Prokaryotic Motility Systems: A Written Symposium (Special Issue Journal of Molecular Microbiology and Biotechnology 2004)

Citation preview

Prokaryotic Motility Systems A Written Symposium

Guest Editor

Ken F. Jarrell, Kingston, Ontario, Canada

33 figures, 10 in color, and 9 tables, 2004

Basel • Freiburg • Paris • London • New York • Bangalore • Bangkok • Singapore • Tokyo • Sydney

This symposium is dedicated to Dr. Robert Macnab whose sudden death saddened all researchers in the field of prokaryotic motility and left a void that will be impossible to fill. His enthusiasm for the area of bacterial motility was contagious to all and his research contributions set the standard for the field.

S. Karger Medical and Scientific Publishers Basel • Freiburg • Paris • London New York • Bangalore • Bangkok Singapore • Tokyo • Sydney

Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Drug Dosage The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.

All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher or, in the case of photocopying, direct payment of a specified fee to the Copyright Clearance Center (see ‘General Information’). © Copyright 2004 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel ISBN 3–8055–7772–9

Vol. 7, No. 1–2, 2004

Contents

5 Self-Assembly and Type III Protein Export of the Bacterial Flagellum Minamino, T.; Namba, K. (Osaka) 18 Dual Flagellar Systems Enable Motility under Different Circumstances McCarter, L.L. (Iowa City, Iowa) 30 The Periplasmic Flagellum of Spirochetes Limberger, R.J. (Albany, N.Y.) 41 Recent Advances in the Structure and Assembly of the Archaeal

Flagellum Bardy, S.L.; Ng, S.Y.M.; Jarrell, K.F. (Kingston, Ont.) 52 Pulling Together with Type IV Pili Nudleman, E.; Kaiser, D. (Stanford, Calif.) 63 Cytophaga-Flavobacterium Gliding Motility McBride, M.J. (Milwaukee, Wisc.) 72 The Junctional Pore Complex and the Propulsion of Bacterial Cells Wolgemuth, C.W. (Farmington, Conn.); Oster, G. (Berkeley, Calif.) 78 Shaping and Moving a Spiroplasma Trachtenberg, S. (Jerusalem)

88 Author Index and Subject Index

© 2004 S. Karger AG, Basel Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Access to full text and tables of contents, including tentative ones for forthcoming issues: www.karger.com/mmb_issues

J Mol Microbiol Biotechnol 2004;7:5–17 DOI: 10.1159/000077865

Self-Assembly and Type III Protein Export of the Bacterial Flagellum Tohru Minamino a Keiichi Namba a, b a Dynamic

NanoMachine Project, ICORP, JST, b Graduate School of Frontier Biosciences, Osaka University, Suita, Osaka, Japan

Key Words Flagellum W Motility W Assembly W Capping structure W Type III protein export W ATPase W Chaperone W Substrate specificity switch

Abstract The bacterial flagellum is a supramolecular structure consisting of a basal body, a hook and a filament. Most of the flagellar components are translocated across the cytoplasmic membrane by the flagellar type III protein export apparatus in the vicinity of the flagellar base, diffuse down the narrow channel through the nascent structure and self-assemble at its distal end with the help of a cap structure. Flagellar proteins synthesized in the cytoplasm are targeted to the export apparatus with the help of flagellum-specific chaperones and pushed into the channel by an ATPase, whose activity is controlled by its regulator to enable the energy of ATP hydrolysis to be efficiently coupled to the translocation reaction. The export apparatus switches its substrate specificity by monitoring the state of flagellar assembly in the cell exterior, allowing this huge and complex macromolecular assembly to be built efficiently by a highly ordered and wellregulated assembly process. Copyright © 2004 S. Karger AG, Basel

ABC Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

© 2004 S. Karger AG, Basel 1464–1801/04/0072–0005$21.00/0 Accessible online at: www.karger.com/mmb

Introduction

Bacteria such as Escherichia coli and Salmonella swim in liquid environments and undergo abrupt directional changes by rotating their flagellar filaments. The rotary motor at the base of the filament is powered by the proton motive force. The motor is placed in either of two states: counterclockwise (CCW) or clockwise (CW) rotation. Bacteria can swim smoothly when their flagellar motors rotate CCW, resulting in the formation of a flagellar bundle that produces the thrust. By quick reversal of the motor rotation, the bundle is disrupted so that bacteria can tumble and change their swimming direction [Macnab, 1996]. The bacterial flagellum is a supramolecular architecture consisting of at least three substructures: a basal body, a hook and a filament (fig. 1). The basal body, which works as a motor, is imbedded within the cell envelope, while the hook and filament, which function as a universal joint and a propeller, respectively, extend outwards from the cells. Flagellar assembly begins with the basal body, followed by the hook and finally by the filament [Suzuki et al., 1978; Kubori et al., 1992]. Since the addition of new monomer of flagellar components to the growing flagellar structure occurs at the distal end, where a cap protein complex specific for each substructure is attached to promote the efficient assembly process, the

Keiichi Namba Graduate School of Frontier Biosciences Osaka University, 1–3 Yamadaoka Suita, Osaka 565-0871 (Japan) Tel. +81 6 6879 4625, Fax +81 6 6879 4652, E-Mail [email protected]

physical path for the flagellar protein export is believed to be the central channel of the growing flagellum with a diameter of 2–3 nm [Iino, 1969; Morgan et al., 1993; Namba et al., 1989]. How are these flagellar proteins selectively recognized and transported through the narrow channel, and how do they self-assemble at the distal end? This review describes our current understanding of the mechanism of flagellar assembly and type III protein export in Salmonella.

The Basal Body

The basal body consists of the MS ring, the rod, the P ring and the L ring. The MS, P and L rings which are made of a cytoplasmic MS ring protein FliF, a periplasmic P ring protein FlgI and an outer membrane L ring protein FlgH, respectively, exist within the cytoplasmic membrane, peptidoglycan layer, and outer membrane, respectively, while the rod, which consists of three proximal rod proteins FlgB, FlgC, FlgF and a distal rod protein FlgG, fully traverses the periplasmic space. Each of three ring proteins is present in about 26 subunits, each of the proximal rod proteins is present in about 6 subunits, and the distal rod protein is present in about 26 subunits [Jones et al., 1990]. The L and P rings together form a cylindrical architecture to act as a molecular bushing of the flagellar axial structure, because the L-P ring complex is very stable against a wide variety of chemical treatment and has a very smooth inner surface [Akiba et al., 1991]. The MS ring has a rotational symmetry, while the rod has a helical symmetry. Since there is a symmetry mismatch between these structures, a putative junction zone would exist between them. Since FliE interacts with FlgB, FliE is postulated to form a MS ring/rod junction zone [Minamino et al., 2000b; Müller et al., 1992]. FliE is estimated to be present in about 9 subunits. In the type III secretion pathway for virulence effector proteins, a large homomultimeric annular complex called secretin lies in the outer membrane and allows virulence factors to cross the outer membrane [Hueck, 1998]. The initiation of the hook assembly can occur in mutants with defects in the L-P ring complex, while the hook elongation cannot [Kubori et al., 1992]. In these mutants, rod/hooktype proteins, such as the hook-capping protein FlgD and the hook protein FlgE, cross the cytoplasmic membrane but not the outer membrane, suggesting that the L-P ring complex also acts as a secretin-like annulus at the stage of rod completion and hook assembly initiation [Minamino and Macnab, 1999].

6

J Mol Microbiol Biotechnol 2004;7:5–17

The Stator and Rotor of the Flagellar Motor

Two cytoplasmic membrane proteins MotA and MotB are involved in energy coupling between proton influx and torque generation [Berg, 2000]. They form a complex around the basal body. Since MotB has a peptidoglycanbinding motif, the complex is believed to be the stator. Three switch proteins FliG, FliM and FliN are responsible not only for torque generation but also for switching its direction, enabling the motor to rotate either CCW or CW [Berg, 2000]. They form the C ring on the cytoplasmic side of the MS ring and therefore seem likely to be the rotor of the motor [Francis et al., 1994]. It has been established that electrostatic interactions at the rotor-stator interface are important for torque generation [Zhou et al., 1998].

The Hook

The hook is a short tubular structure connecting the filament and the basal body. It is composed of about 120 subunits of the hook protein FlgE, which are assembled with a helical symmetry. One of the remarkable features of the hook is that its length is fairly tightly controlled at approximately 55 B 6 nm in wild-type cells [Hirano et al., 1994]. A hook length control protein FliK, together with an integral membrane export component, FlhB, switches the substrate specificity of the export apparatus when the hook reaches its mature length (see below) [Kutsukake et al., 1994a; Williams et al., 1996]. Mutations in fliK prevent FlhB from switching the substrate specificity from rod/hook-type proteins to filamen-type proteins, producing an abnormally elongated hook structure called polyhook. In contrast to the substrate specificity switching process, FliK is not directly involved in determining the hook length, because fliK mutants show a peak in the polyhook length distribution at the wild-type value of 55 nm, even though the distribution extends much further out than that of the wild type [Koroyasu et al., 1998]. Mutants with defects in the switch proteins forming the C ring, which is also required for the flagellar assembly/protein export process, produce short hooks of distinct lengths, which lead to a hypothesis that hook length may be determined by the amount of hook subunits measured and exported by the export apparatus [Makishima et al., 2001].

Minamino/Namba

Fig. 2. Overall structure of the flagellar filament. a Solid surface representation of a filament density map. b Contour map of the cen-

Fig. 1. Schematic diagram of the bacterial flagellum.

The Hook-Filament Junction Zone

tral section though the filament axis, showing the inner core structure. The density map was obtained by electron cryomicroscopy and helical image analysis of the R-type straight filament [Mimori et al., 1995]. Contour level in red is twice higher than that of blue in b. The filament is oriented as its proximal end top and the distal end bottom.

The Filament

The hook-filament junction is a very short segment formed by FlgK (HAP1) and FlgL (HAP3) in this order [Homma and Iino, 1985; Ikeda et al., 1997]. Each of the junction proteins is present in about 13 molecules [Ikeda et al., 1997]. A mutation in FlgL results in a change of the helical shape of the filament during swimming, suggesting that the hook-filament junction serves to dump out or seal the structural conversion of the hook and to prevent the polymorphic activity of the hook to function as a universal joint from being transmitted to the filament [Fahner et al., 1994]. Therefore, it seems very likely that the junction zone acts as a buffering structure connecting two filamentous structures with distinct mechanical characteristics. This idea is supported by X-ray crystallographic studies [Imada, K., Matsunami, H., Namba, K., unpubl. results].

The filament is the largest portion of the flagellum, growing to a length of around 15 Ìm by self-assembly of as many as 30,000 subunits of flagellin (FliC). The filament is a tubular structure made by helical assembly of flagellin subunits or of 11 protofilaments, which are axially staggered by about a half subunit in their lateral interactions. The filament structures visualized by electron cryomicroscopy and X-ray fiber diffraction at a resolution of about 10 Å have revealed a concentric double-tubular structure in the filament core, in which ·-helices are axially oriented and bundled together (fig. 2) [Mimori et al., 1995; Morgan et al., 1995; Mimori-Kiyosue et al., 1996, 1997, 1998; Yamashita et al., 1998]. Flagellin in the filament is divided into four domains, D0, D1, D2 and D3. Domains D0 and D1 form the inner and outer tubes of the concentric double-tubular structure, respectively. Terminal segments of flagellin form the inner tube structure (D0). Since the filament can be formed by domain inter-

Bacterial Flagellar Assembly and Type III Protein Export

J Mol Microbiol Biotechnol 2004;7:5–17

7

Fig. 3. Complete atomic model of the Rtype filament in stereo pair. a End on view from the distal end. b Side view from outside. c Side view of the inside. Three of the 11 protofilaments are removed in b and c. Only C· backbone is presented as ribbon diagram and color-coded from blue to cyan, green, yellow and red according the amino acid sequence from the N- terminus to the C-terminus. Domain D0 is therefore blue and red; domain D1, cyan and yellow; domains D2 and D3, green.

actions in the outer tube (D1) alone at high ionic strength, the inner tube structure is not essential, albeit important, for efficient polymerization into the highly stable filament structure [Vonderviszt et al., 1991; Mimori-Kiyosue et al., 1997]. Domains D2 and D3 form the outer part of flagellin in the filament, making relatively minor contributions to the filament structure and stability [Yoshioka et al., 1995; Mimori-Kiyosue et al., 1997], but are probably important to increase the effective filament diameter for

8

J Mol Microbiol Biotechnol 2004;7:5–17

the thrust generation. Domain D3 is highly variable, responsible for changing the serotype. Polymorphic supercoiling is the essential structural feature of the filament to be a helical propeller with a switchable helical handedness between left and right [Asakura, 1970; Calladine, 1975, 1976, 1978]. X-ray fiber diffraction studies revealed that the two distinct protofilament conformations, the R- and L-type, with repeat distance of 51.9 and 52.7 Å, respectively, are responsible for

Minamino/Namba

the gentle curvatures of the supercoiled filaments of wild type as well as many other polymorphic forms produced by mechanical perturbations and mutations [Yamashita et al., 1998]. To understand the switching mechanism, the crystal structure of a major fragment of flagellin, F41, which lacks 52 N- and 44 C-terminal residues, was solved at 2 Å by X-ray crystallography [Samatey et al., 2001]. The structure revealed that domains D2 and D3 consist of mainly ß-strands, whereas domain D1 is made of three long ·-helices paired with a ß-hairpin. The crystal structure also revealed the axial subunit interactions between domains D1 that are responsible for the protofilament formation. The protofilament structure in the crystal was in the R-type conformation, the shorter state of the two that produce the gentle curvatures. Simulated extension of the atomic model of the R-type protofilament revealed that a small but distinct conformational change of the ßhairpin in domain D1 is responsible for the switching protofilament conformations between the L- and R-type [Samatey et al., 2001]. Recently, electron cryomicroscopy and helical image analysis allowed the complete atomic model of the R-type filament structure to be built, visualizing ·-helical coil of domain D0 and its intersubunit hydrophobic interactions to form the inner tube and hydrophilic inner surface of the central channel with a diameter of only 2 nm (fig. 3) [Yonekura et al., 2003]. The hydrophobic interactions between domains D0 make the filament structure mechanically very stable in aqueous environments. The hydrophilic inner surface of the channel is important for rapid and efficient transport of flagellin in a largely unfolded conformation with many hydrophobic residues exposed.

newly exported flagellin molecules [Yonekura et al., 2000]. The anchor domain of the cap is composed of 40 N- and 50 C-terminal segments of FliD, which are unfolded in the isolated cap or in the monomeric state of FliD [Vonderviszt et al., 1998]. This anchor domain appears to extend from the tip of the leg-like domain further down into the cavity to bind to domains D0 of flagellin at the distal end of the inner tube for the stable cap binding [Yonekura et al., 2000, 2001; Maki-Yonekura et al., 2003]. There is a symmetry mismatch between the helical subunit array of the filament with 11 protofilaments forming the tube and the pentameric annular structure of the FliD cap, and this symmetry mismatch is the basis of the assembly promotion mechanism by the cap. The leglike domains of the cap appear to be flexible enough to adjust their conformation for binding to the filament end over the symmetry mismatch. Three-dimensional density map of the filament cap-filament complex has shown that the cap binding to the filament end fills four of the five indentations formed by domains D1 of flagellin and leaves the remaining double-width double-step indentation as an open gap available for flagellin binding [Yonekura et al., 2000]. Upon incorporation of flagellin into the gap, the cap rotates by 6.5° and moves up by 4.7 Å by rearranging the conformation of leg-like domains and/or their binding partners to a symmetrically equivalent position in the closest neighbor to prepare a new open gap as the next flagellin binding site (fig. 4) [Yonekura et al., 2000; Maki-Yonekura et al., 2003].

Self-Assembly Process of the Flagellum The Filament Cap

The filament cap exists at the growing end of the filament and plays an important role in the efficient growth of the filament [Homma et al., 1984; Ikeda et al., 1985]. The cap complex, which is composed of 5 subunits of the filament capping protein FliD (HAP2) [Ikeda et al., 1996; Imada et al., 1998], has a plate domain and axially extended domains that plug into the cavity at the distal end of the filament [Maki et al., 1998]. Structural analysis of the cap complex as well as the cap-filament complex by electron microscopy revealed that the cap is made of a pentagonal plate domain as a lid and five axially extended leg-like domains that bind to the indentations formed by domains D1 of flagellin subunits at the distal end, forming a cavity under the cap plate as a folding chamber for

Bacterial Flagellar Assembly and Type III Protein Export

Self-assembly of the flagellum proceeds in an ordered fashion (fig. 5) [Kubori et al., 1992; Suzuki et al., 1978]. First, FliF self-assembles into the MS ring within the cytoplasmic membrane. Using the MS ring structure as a template, FliG, FliM and FliN self-assemble into the C ring onto the cytoplasmic face of the MS ring. The type III flagellar protein export apparatus, which consists of six integral-membrane proteins and three cytoplasmic proteins, are then presumed to be assembled within the putative central pore of the MS ring and in the cytoplasmic space within the C-ring. At the moment, it is still difficult to reconcile the two apparently conflicting results: while six of the type III component proteins have transmembrane domains, the central channel of the MS ring is not wide open to accommodate these domains [Francis et al., 1994; Suzuki et al.,

J Mol Microbiol Biotechnol 2004;7:5–17

9

Fig. 4. Schematic diagram depicting the cap-filament binding and the rotary cap mechanism promoting the flagellin assembly. Rotation and axial translation of the cap plate and accompanied rearrangement of the legs upon every incorporation of a flagellin subunit (from left to right). Top view in the upper panel and an oblique view in the lower panel. In the upper panel, the cap plate is made transparent to show the different ways of leg domain binding, where black dots indicate the five positions of leg domain attachment to the plate. The plate and the black dots have strict five-fold symmetry, while the leg domains do not. In the lower panel, the outer domain of flagellin is

removed for clarity. Subunits in violet are newly incorporated flagellin molecules. Five open circles in the upper panel indicate the initial positions of the cap plate vertices as a reference for the cap rotation. The flagellin assembly proceeds along the 1-start helix, which is in the CCW direction in the view from the top, approximately at every 65.5 ° (360*2/11). This is also the angle of rotation after which the next binding site appears. However, because the legs of the cap are located every 72 ° (360/5), a 6.5 ° CW rotation with permutation of the leg conformations is sufficient to make the appropriate interactions between the leg and flagellin subunits.

2004]. In any case, once the type III export apparatus is assembled, the flagellar protein export through the central channel of the MS ring into the periplasmic space is ready to start. FliE is the first one to be exported and is likely to assemble at the periplasmic surface of the MS ring at the bottom of its receptacle-like structure, in which most of the proximal rod portion is accommodated [Suzuki et al., 2004]. Then, the rod assembly begins with FlgB, FlgC and FlgF (probably in this order) followed by FlgG with the help of the rod cap made of FlgJ, which has a muramidase activity to locally digest the peptidoglycan layer and permit penetration of the rod growing outwards [Hirano et al., 2001; Nambu et al., 1999]. Upon completion of the rod structure, the rod cap is presumably replaced by the hook cap, which is made of FlgD [Ohnishi et al., 1994]. After the rod is formed, FlgI and FlgH assemble around the rod in the peptidoglycan layer and outer membrane, respectively, to form the L-P ring complex. In contrast to the component proteins of the flagellar axial structure, FlgI and FlgH are synthesized as precursor forms with their secretion signal sequence within the cytoplasm,

translocated across the cytoplasmic membrane by a general secretion system called the Sec pathway, and subjected to signal peptide cleavage [Homma et al., 1987; Jones et al., 1987]. The P ring formation requires a periplasmic chaperone FlgA [Nambu and Kutsukake, 2000]. The L ring formation is somehow required for the FlgD cap formation. The hook protein FlgE self-assembles into the hook at the tip of the distal rod with the aid of the hook cap. When the hook reaches its mature length, the hook cap is replaced by FlgK (HAP1), and then FlgL (HAP3) and FliD (HAP2) are bound at the distal end in this order. FliC (flagellin) then starts self-assembling into the long helical filament onto the hook-filament junction with the help of the filament cap made of FliD.

10

J Mol Microbiol Biotechnol 2004;7:5–17

The Type III Flagellar Protein Export Apparatus

Almost all substructures of the flagellum lie beyond the cytoplasmic membrane (fig. 1). Thus the flagellar components, which are synthesized within the cytoplasm, must be exported to their final destinations where their assem-

Minamino/Namba

bly occurs. With three exceptions, FlgA, FlgI and FlgH, all of the external components cross the cytoplasmic membrane through the central channel of the growing flagellar structure by a flagellum-specific protein export apparatus, which consists of six integral membrane components (FlhA, FlhB, FliO, FliP, FliQ, and FliR) and three soluble components (FliH, FliI, and FliJ) (fig. 6) [Minamino and Macnab, 1999]. In addition to these component proteins, other cytoplasmic proteins (FlgN, FliS, and FliT) act as substrate-specific chaperones that facilitate the export of their substrates.

Fig. 5. The self-assembly process of the bacterial flagellum. The

sequence starts from the top left and proceeds to the bottom right. After the motor structure is partially assembled by FliF in the cytoplasmic membrane and FliG, FliM and FliN in the cytoplasmic surface of the FliF ring, the protein export apparatus is assembled onto its cytoplasmic surface, and it starts exporting flagellar proteins into the central channel selectively by using the energy of ATP hydrolysis. Flagellar proteins that reach the distal end of the growing structure self-assemble onto the template structure formed there, where the distal cap helps the efficient self-assembly of the exported proteins and promotes the flagellar growth.

Fig. 6. Schematic diagram of the bacterial flagellar type III flagellar protein export apparatus. FlhA, FlhB, FliO, FliP, FliQ and FliR are integral membrane proteins and are postulated to be located within a central pore of the MS ring. FlhA and FlhB have substantially large cytoplasmic domains, which project into the cavity within the C ring. FlhA and FlhB are depicted as the mounting place for the ATPase FliI and its regulator FliH. A general chaperon FliJ prevents its substrates from premature aggregation in the cytoplasm. CM = Cytoplasmic membrane. This diagram is based on the structures of the basal body [Francis et al., 1994] and the FliF/ FliG ring complex [Suzuki et al., 2004] obtained by electron cryomicroscopy and other data obtained by biochemical and genetic analyses described in the text, with still putative central pore of the MS ring.

Bacterial Flagellar Assembly and Type III Protein Export

J Mol Microbiol Biotechnol 2004;7:5–17

11

Export of flagellar proteins has characteristics in common with secretion of various virulence factors by pathogenic bacteria, and they now carry the common designation of type III export pathways [Hueck, 1998]. Their components share a high degree of sequence similarity. Interestingly, the main structure of the type III virulence factor secretion apparatus has been identified as a needle complex which is morphologically very similar to the hook-basal body structure of the flagellum [Kubori et al., 1998].

Export Signal Recognized by the Export Apparatus

Since only the flagellar components go through the flagellar protein export pathway, they must contain recognition signals for the export. Multiple candidates of signals, such as amino acid sequence and mRNA, have been proposed for the selective export and its regulation in the Salmonella flagellar protein export system [Karlinsey et al., 2000]. Protein synthesis inhibitors such as spectinomycin and chloramphenicol do not abolish the export of flagellar proteins, indicating that flagellar protein export can occur post-translationally and is not obligatorily linked to translation. Thus, the amino acid sequence of these proteins suffices to mediate their recognition and export [Hirano et al., 2003]. The N-terminal polypeptide chains of flagellar proteins retain the ability to be exported through the export pathway [Iyoda and Kutsukake, 1995; Kornacker and Newton, 1994; Kuwajima et al., 1989; Minamino et al., 1999b], suggesting that their N-terminal regions contain the signals. The amino acid sequences of flagellar axial proteins do show some, albeit relatively weak, sequence similarities at their termini [Homma et al., 1990a, b]. However, these sequence similarities seem to reflect a common structural motif forming ·-helical coils and their hydrophobic intermolecular interactions in the inner core of the axial structure, which makes it mechanically stable in aqueous environments just as shown in the atomic model of the filament [Yonekura et al., 2003], rather than signals for export, because they are required for flagellar assembly but not for export as shown by mutation experiments in the hook protein of Caulobacter [Kornacker and Newton, 1994]. Therefore, the export apparatus seems to recognize a higher-order structural motif in these proteins rather than the primary sequence.

12

J Mol Microbiol Biotechnol 2004;7:5–17

Roles of Integral Membrane Components

Since flagellar components are delivered into the central channel of the growing flagellar structure, all the integral membrane components, FlhA, FlhB, FliO, FliP, FliQ and FliR, have been postulated to be located in a patch of membrane within the putative central pore of the MS ring (fig. 5) [Minamino and Macnab, 1999]. It has been experimentally shown that FliP, FliR and FlhA are physically associated with the MS ring [Fan et al., 1997; Kihara et al., 2001]. Katayama et al. [1996] have shown that a structure called the C rod lies at the center of the cytoplasmic face of the MS-ring/C-ring structure and have proposed that the C rod is part of the export apparatus. FlhA is a 75-kD membrane protein, consisting of two regions: the hydrophobic N-terminal transmembrane region that is predicted to cross the cytoplasmic membrane eight times and the C-terminal cytoplasmic domain termed FlhAC [Minamino et al., 1994]. Since mutations within the transmembrane region of FlhA suppress a specific fliF mutation resulting in an in-frame deletion of Ala 174 and Ser 175 of FliF, the transmembrane region is important for the association of FlhA with the MS ring [Kihara et al., 2001]. Temperature-sensitive flhA mutations, which abolish flagellar protein export at restrictive temperature, lie in FlhAC [Minamino T, unpubl. results], indicating that FlhAC plays a central role in the translocation of its export substrates across the cytoplasmic membrane. In agreement with this, FlhAC binds to cytoplasmic components, FliH, FliI and FliJ, the C-terminal cytoplasmic domain of FlhB and its export substrates in vitro [Minamino and Macnab, 2000c; Zhu et al., 2002]. FlhB is a 42-kD membrane protein, consisting of two domains: the hydrophobic N-terminal domain that is predicted to cross the cytoplasmic membrane four times and the C-terminal cytoplasmic domain termed FlhBC [Minamino et al., 1994]. As mentioned later, FlhBC is directly involved in the substrate-specificity switching of the export apparatus during the flagellar assembly [Fraser et al., 2003; Kutsukake et al., 1994a; Minamino and Macnab, 2000b, c, Williams et al., 1996]. Since FlhBC binds to FliH, FliI, FliJ and the C-terminal cytoplasmic domain of FlhA in vitro, it must also be responsible for the protein translocation mechanism [Minamino and Macnab, 2000c; Zhu et al., 2002]. The extragenic fliH bypass mutations, which substantially improve the motility and flagellar protein export of a fliH null mutant, lie near the N-terminus of FlhA and FlhB close to the beginning of their first membrane spans, suggesting that these relatively short N-terminal cytoplas-

Minamino/Namba

mic regions of FlhA and FlhB are also involved in the interaction with FliH and FliI [Minamino et al., 2003]. Taken together, FlhA and FlhB appear to function as the mounting place for FliH and FliI. FliO (13 kD) has a small N-terminal cytoplasmic domain, a single transmembrane helix and a substantially large periplasmic domain. FliP has a signal sequence at its N terminus and its mature protein exists in two forms, 25and 23-kD fragments, by cleavage of N-terminal peptides. FliP undergoes cleavage of these signal peptides in the process of insertion into the cytoplasmic membrane. FliP also has a substantial periplasmic domain between transmembrane span 3 and 4. FliQ (9.6 kD) crosses the cytoplasmic membrane twice. FliR (29 kD) is almost exclusively hydrophobic and appears to consist of a series of five or six transmembrane helices connected by short loops [Ohnishi et al., 1997]. Although the exact functions of FliO, FliP, FliQ and FliR remain unknown, a plausible hypothesis is that, together with the transmembrane domains of FlhA and FlhB, they form a protein-conducting channel [Minamino and Macnab, 1999].

Roles of Cytoplasmic Components FliI, FliH and FliJ

The ATPase FliI is a 49-kD cytoplasmic protein consisting of at least two regions: the N-terminal flagellum-specific region and the C-terminal ATPase region [Fan and Macnab, 1996; Vogler et al., 1991]. The ATPase region has a significant sequence similarity to the catalytic ß-subunit of the proton-driven F0F1ATP synthase. Since FliI with mutation within the catalytic sites of the ATPase domain cannot support the export of any flagellar proteins, FliI is likely to provide the energy for the translocation of its substrates across the cytoplasmic membrane [Fan and Macnab, 1996]. Mutations within the N-terminal flagellumspecific region strongly affect the ATPase activity, suggesting that the N-terminal region exert a regulatory effect on the catalytic mechanism [Fan and Macnab, 1996]. While purified FliI exists as monomer in solution [Fan and Macnab, 1996; Minamino and Macnab, 2000b], in the presence of ATP or no-hydrolysable ATP analog, FliI is capable of assembling into a hexameric ring structure that has a central cavity with a diameter of about 3 nm. The ATPase activity of FliI displays positive co-operation, suggesting that FliI oligomerization would be physiologically significant for flagellar protein export [Claret et al., 2003; Suzuki, H., Minamino, T., Namba, K., unpubl. result].

Bacterial Flagellar Assembly and Type III Protein Export

FliH is a 26-kD cytoplasmic protein, exists as dimer in solution and forms a stable complex with FliI identified as (FliH)2FliI [Minamino and Macnab, 2000b; Minamino et al., 2001]. The N-terminal flagellum-specific region of FliI and the C-terminal region of both FliH molecules of the FliH dimer are responsible for the formation of this heterotrimeric complex [Auvary et al., 2002; Gonza´lez-Pedrajo et al., 2002; Minamino and Macnab, 2000b, Minamino et al., 2001, 2002]. A central region of FliH that is predicted to form ·-helical coiled-coil structure at a significant probability is responsible for its dimerization [Gonza´lez-Pedrajo et al., 2002]. The ATPase activity of purified (FliH)2FliI complex is 10-fold lower than that of FliI alone, indicating that FliH inhibits the ATPase activity of FliI [Minamino and Macnab, 2000b]. Thus, FliH appears to be an ATPase regulatory protein to prevent FliI from wasting ATP during the flagellar protein export process. Since both the motility and export defects of the fliH null mutant are bypassed by overproduction of FliI or by other extragenic bypass mutations, which are localized within FlhA and FlhB, it seems likely that FliH plays an important role in effective docking of FliI to FlhA and FlhB to enable the protein export function of the export apparatus and to make the energy of ATP hydrolysis be efficiently coupled to flagellar protein export [Minamino et al., 2003]. The N-terminal regions of FliH and FliI are responsible for their docking to FlhA and FlhB. Both FliH and FliI are found to be associated with the membrane even in the absence of any other components of the flagellum, indicating that they behave as peripheral membrane proteins [Auvary et al., 2002]. The ATPase activity of FliI is greatly stimulated by acidic phospholipids. In contrast, phospholipids do not relieve the inhibitory effect of FliH on FliI [Auvary et al., 2002]. Since FliH does not directly affect the catalytic sites of the ATPase domain of FliI, and both FliH and FliI interact with FlhA and FlhB, it is plausible that either one or both subunits of the FliH dimer go through a conformational change in the C-terminal region upon binding to integral membrane components of the export apparatus, altering the association between the FliH dimer and FliI so that FliI can fully exert its ATPase activity. FliJ is a 17-kD cytoplasmic protein with a high probability of ·-helical coiled-coil structure near its N terminus. It has several common features with the type III cytoplasmic chaperone family. Several export substrates form inclusion bodies in the cytoplasm when overproduced, while co-overproduction of these substrates with FliJ suppresses their aggregation in the cytoplasm [Minamino et

J Mol Microbiol Biotechnol 2004;7:5–17

13

al., 2000a]. Thus, FliJ has a chaperone-like activity to prevent export substrates from premature aggregation in the cytoplasm. It has been shown that FliJ binds to FliH, FliI, FlhAC and FlhBC [Gonza´lez-Pedrajo et al., 2002; Minamino and Macnab, 2000c]. However, its exact role in flagellar protein export still remains unclear.

Roles of Substrate-Specific Chaperones, FliS, FlgN and FliT

FliS, FlgN, and FliT are cytoplasmic proteins with a deduced molecular mass of 15, 16.5, and 14 kD, respectively, and have several common features with the type III cytoplasmic chaperone family [Kawagishi et al., 1992; Kutsukake et al., 1994b]. Mutations in these proteins impair the export of FliC, FlgK, FlgL and FliD in the following pairs: FliS – FliC; FlgN – FlgK and FlgL; FliT – FlgD [Bennett et al., 2001; Fraser et al., 1999; Yokoseki et al., 1995]. Purified FliS, FlgN and FliT bind to C-terminal amphipathic ·-helical domains of their cognate substrates to prevent their in vitro aggregation [Auvray et al., 2001; Bennett et al., 2001; Fraser et al., 1999], suggesting that these proteins function as substrate-specific chaperones to facilitate the export of their substrates. In addition to the role as substrate-specific chaperones, it has also been reported that FliS is responsible for negative regulation of the export of FlgM (anti-sigma factor), FlgN for translational regulation of FlgM, and FliT for negative control of the flagellar gene expression [Karlinsey et al., 2000; Kutsukake et al., 1999; Yokoseki et al., 1996]. Thus, it is likely that these chaperones also play important roles in fine-tuning of the flagellar assembly process.

Substrate Specificity Switching of the Export Apparatus

The flagellar type III protein export apparatus mediates an ordered export of flagellar proteins by monitoring the state of flagellar assembly [Hirano et al., 2003; Hughes et al., 1993; Kutsukake, 1994; Kutsukake et al., 1994a, Minamino et al., 1999a; Williams et al, 1996]. The rod-type (FliE, FlgB, FlgC, FlgG and FlgJ) and the hooktype proteins (FlgD, FlgE, FliK) belong to one class, and the filament-type proteins (FlgK, FlgL FliD, FliC and FlgM) belong to the other. The export of rod-/hook-type proteins is followed by the export of filament-type proteins, and a switch in export specificity occurs at a well-

14

J Mol Microbiol Biotechnol 2004;7:5–17

defined point in the export/assembly process. At least two flagellar proteins, FliK, a hook-length control protein, and FlhB, an integral membrane component of the export apparatus, are directly involved in this process. During the rod and hook assembly process, FlhB supports the export of rod/hook-type proteins, but not filament-type proteins. Upon completion of the hook structure to its wild-type length of 55 nm, a signal is somehow transmitted to FliK and/or FlhB, shutting off the rod/hook-type protein export and turning on the filament-type protein export. It is established that the type III virulence factor secretion systems have a similar machinery for the length control and substrate specificity switching of their export apparatus [Edqvist et al., 2003; Kubori et al., 2000]. FliK is a 42-kD protein consisting of three regions: an N-terminal region, a proline-rich central region, and a highly conserved C-terminal region [Kawagishi et al., 1996]. The N- and C-terminal regions of FliK are involved in its export and switching of the substrate specificity, respectively, and the middle region is probably just a linker, although its function is less well understood [Minamino et al., 1999b; Williams et al., 1996]. The export deficient fliK mutants produce the polyhook, sometimes with the filament, indicating that the export of FliK during hook assembly is important for the switching of substrate specificity [Minamino et al., 1999b]. However, it remains unknown whether FliK is associated with the growing flagellar structure to sense the completion of hook assembly. The C-terminal cytoplasmic domain of FlhB (FlhBC), in which all known extragenic suppressor mutations of fliK mutants lie, exists in two substrate specificity modes: rod/hook-type; and filament-type [Kutsukake et al., 1994a; Minamino and Macnab, 2000c, Williams et al., 1996]. Wild-type FlhBC is unstable, being specifically cleaved at the site Asn-269/Pro-270 into two subdomains termed FlhBCN and FlhBCC, but these two subdomains retain the ability to tightly interact with each other after cleavage [Minamino and Macnab, 2000a]. Mutant FlhB proteins that can undergo the switching of substrate specificity even in the absence of FliK are much more resistant to cleavage, indicating that their conformation must be different from the wild type. The N269A mutation of FlhB, which completely prevents the cleavage of FlhBC, is locked in the rod/hook-type of export specificity [Fraser et al., 2003]. Taken together all these observations suggest that FlhBC functions as an export switch, and its conformational change, which is probably a change of interaction mode between FlhBCN and FlhBCC, is responsible for the specificity switching process.

Minamino/Namba

Perspective

As we have described, the bacterial flagellum is a complex macromolecular assembly whose assembly process involves many different proteins, not only as its components of the final structure but also as its assembly scaffold, assembly regulator, selective and switchable component transporter, assembly monitor and energy provider. Specific and cooperative interactions, selective export and conformational switching of these proteins, in a timely manner in every step, achieve highly ordered and wellregulated assembly of the component proteins. The energy required for the self-assembly is gained by ATP hydrolysis in the cell interior and transmitted over a long distance to the distal end of the growing structure in the cell exterior via the transport of component proteins though the narrow central channel. We are to look into these processes in much more detail to fully understand these intri-

cate mechanisms. High-resolution structural information by X-ray crystallography and electron microscopy would be essential to advance our mechanistic understanding of the process. Dynamic observation of individual proteins by single molecule technique would also be important to follow the actual assembly process and protein-protein interactions involved. These works together with biochemical, physicochemical and genetic analyses are now under way towards comprehensive understanding of this highly sophisticated and complex system.

Acknowledgements We dedicate this article to the late Robert M. Macnab for his outstanding contributions and leadership in the studies of flagellar assembly and protein export mechanisms. We greatly appreciate his continuous and inspiring advice and encouragement.

References Akiba, T, Yoshimura, H, Namba, K: Monolayer crystallization of flagellar L-P rings by sequential addition and depletion of lipid. Science 1991;252:1544–1546. Asakura, S: Polymerization of flagellin and polymorphism of flagella. Adv Biophys 1970;1:99– 155. Auvray, F, Ozin, AJ, Claret, L, Hughes, C: Intrinsic membrane targeting of the flagellar export ATPase FliI: Interaction with acidic phospholipids and FliH. J Mol Biol 2002;318:941–950. Auvray, F, Thomas, J, Fraser, GM, Hughes, C: Flagellin polymerization control by a cytosolic export chaperone. J Mol Biol 2001;308:221–229. Bennett, JCQ, Thomas, J, Fraser, GM, Hughes, C: Substrate complex and domain organization of the Salmonella flagellar export chaperones FlgN and FliT. Mol Microbiol 2001;39:781– 791. Berg, HC: Constraints on models for the flagellar motor. Phil Trans Roy Soc B 2000;335:491– 502. Calladine, CR: Construction of bacterial flagella. Nature 1975;225:121–124. Calladine, CR: Design requirements for the construction of bacterial flagella. J Theor Biol 1976;57:469–489. Calladine, CR: Change of waveform in bacterial flagella: The role of mechanics at the molecular level. J Mol Biol 1978;118:457–479. Chilcott, GS, Hughes, KT: The type III secretion determinants of the flagellar anti-transcription factor, FlgM, extended from amino-terminus into the anti-Û28 domain. Mol Microbiol 1998; 30:1029–1040.

Bacterial Flagellar Assembly and Type III Protein Export

Claret, L, Susannah, CR, Higgins, M, Hughes, C: Oligomerization and activation of the FliI ATPase central to bacterial flagellum assembly. Mol Microbiol 2003;48:1349–1355. Edqvist, PJ, Olsson, J, Lavander, M, Sundberg, L, Forsberg, Å, Wolf-Wanz, H, Lloyd, SA: YscP and YscU regulate substrate specificity of the Yersinia type III secretion system. J Bacteriol 2003;185, 2259–2266. Fahner, KA, Block, SM, Krishnaswamy, S, Parkinson, JP, Berg, HC: A mutant hook-associated protein (HAP3) facilitates torsionally induced transformations of the flagellar filament of Escherichia coli. J Mol Biol 1994;238:173– 186. Fan, F, Macnab, RM: Enzymatic characterization of FliI: an ATPase involved in flagellar assembly in Salmonella typhimurium. J Biol Chem 1996;271:31981–31988. Fan, F, Ohnishi, K, Francis, NR, Macnab, RM: The FliP and FliR proteins of Salmonella typhimurium, putative components of the type III flagellar export apparatus, are located in the flagellar basal body. Mol Microbiol 1997;26: 1035–1046. Francis, NR, Sosinsky, GE, Thomas, D, DeRosier, DJ: Isolation, characterization, and structure of bacterial flagellar motors containing the switch complex. J Mol Biol 1994;235:1261– 1270. Fraser, GM, Bennett JCQ, Hughes, C: Substratespecific binding of hook-associated proteins by FlgN and FliT, putative chaperones for flagellum assembly. Mol Microbiol 1999;32:569– 580.

Fraser, GM, Hirano, T, Ferris, HU, Devgan, LL, Kihara, M, Macnab RM: Substrate specificity of type III flagellar protein export in Salmonella is controlled by subdomain interactions in FlhB. Mol Microbiol 2003;48:1043–1057. Gonza´lez-Pedrajo, B, Fraser, GM, Minamino, T, Macnab, RM: Molecular dissection of Salmonella FliH, a regulator of the ATPase FliI and the type III flagellar protein export pathway. Mol Microbiol 2002;45:967–982. Hirano, T, Minamino, T, Macnab RM: The Role in flagellar rod assembly of the N-terminal domain of Salmonella FlgJ, a flagellum-specific muramidase. J Mol Biol 2001;312:359–369. Hirano, T, Minamino, T, Namba, K, Macnab, RM: Substrate specificity classes and the recognition signal for Salmonella type III flagellar export. J Bacteriol 2003;185:2485–2492. Hirano, T, Yamaguchi, S, Oosawa, K, Aizawa S-I: Roles of FliK and FlhB in determination of flagellar hook length in Salmonella typhimurium. J Bacteriol 1994;176:5439–5449. Homma, M, DeRosier, DJ, Macnab, RM: Flagellar hook and hook-associated proteins of Salmonella typhimurium and their relationship to other axial components of the flagellum. J Mol Biol 1990a; 213:819–832. Homma, M, Fujita, H, Yamaguchi, S, Iino, T: Excretion of unassembled flagellin by Salmonella typhimurium mutant deficient hook-associated proteins. J Bacteriol 1984;159:1056– 1059. Homma, M, Iino, T: Location of hook-associated proteins in flagellar structures of Salmonella typhimurium. J Bacteriol 1985;162:183–189.

J Mol Microbiol Biotechnol 2004;7:5–17

15

Homma, M, Komeda, Y, Iino, T, Macnab, RM: The flaFIX gene product of Salmonella typhimurium is a flagellar basal body component with a signal peptide for export. J Bacteriol 1987;169:1493–1498. Homma, M, Kutsukake, K, Hasebe, M, Iino, T, Macnab, RM: FlgB, FlgC, FlgF and FlgG. A family of structurally related proteins in the flagellar basal body of Salmonella typhimurium. J Mol Biol 1990b; 211:465–477. Hueck, CJ: Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol Mol Biol Rev 1998;62:379–433. Hughes, KT, Gillen, KL, Semon, MJ, Karlinsey, JE: Sensing structural intermediates in bacterial flagellar assembly by export of a negative regulator. Science 1993;262:1277–1280. Iino, T: Polarity of flagellar growth in Salmonella. J Gen Microbiol 1969;56:227–239. Ikeda, T, Asakura, S, Kamiya, R: ‘Cap’ on the tip of Salmonella flagella. J Mol Biol 1985;184:735– 737. Ikeda, T, Homma, M, Iino, T, Asakura, S, Kamiya, R: Localization and stoichiometry of hookassociated proteins within Salmonella typhimurium flagella. J Bacteriol 1987;169:1168– 1173. Ikeda, T, Oosawa, K, Hotani, H: Self-assembly of the filament capping protein, FliD, of bacterial flagella into an annular structure. J Mol Biol 1996;259:679–686. Imada, K, Vonderviszt, F, Furukawa, Y, Oosawa, K, Namba, K: Assembly characteristics of flagellar cap protein HAP2 of Salmonella: Decamer and pentamer in the pH-sensitive equilibrium. J Mol Biol 1998;277:883–891. Iyoda, S, Kutsukake, K: Molecular dissection of the flagellum-specific anti-sigma factor, FlgM, of Salmonella typhimurium. Mol Gen Genet 1995;249:417–424. Jones, CJ, Homma, M, Macnab, RM: Identification of proteins of the outer (L and P) rings of the flagellar basal body of Escherichia coli. J Bacteriol 1987;169:1489–1492. Jones, CJ, Macnab, RM, Okino, H, Aizawa, S-I: Stoichiometric analysis of the flagellar hook(basal-body) complex of Salmonella typhimurium. J Mol Biol 1990;212:377–387. Katayama, E, Shiraishi, T, Oosawa, K, Baba, N, Aizawa, S-I: Geometry of the flagellar motor in the cytoplasmic membrane of Salmonella typhimurium as determined by stereo-photogrammetry of quick-freeze deep-etch replica images. J Mol Biol 1996;255:458–475. Karlinsey, JE, Lonner, J, Brown, KL, Hughes, KT: Translation/secretion coupling by type III secretion systems. Cell 2000;102:487–497. Kawagishi, I, Homma, M, Williams, AW, Macnab, RM: Characterization of the flagellar hook length control protein FliK of Salmonella typhimurium and Escherichia coli. J Bacteriol 1996;178:2954–2959. Kawagishi, I, Müller, V, Williams, AW, Macnab, RM: Subdivision of the flagellar region III of Escherichia coli and Salmonella typhimurium chromosome and identification of two additional flagellar genes. J Gen Microbiol 1992; 138:1051–1065.

16

Kihara, M, Minamino, T, Yamaguchi, S, Macnab, RM: Intergenic suppression between the flagellar MS ring protein FliF of Salmonella and FlhA, a membrane component of its export apparatus. J Bacteriol 2001;183:1655–1662. Kornacker, MG, Newton, A: Information essential for cell-cycle-dependent secretion of the 591residue Caulobacter hook protein is confined to a 21-amino-acid sequence near the N-terminus. Mol Microbiol 1994;14:73–85. Koroyasu, S, Yamazato, M, Hirano, T, Aizawa, S-I: Kinetic analysis of the growth rate of the flagellar hook in Salmonella typhimurium by the population balance method. Biophys J 1998; 74:436–443. Kubori, T, Matsushima, Y, Nakamura, D, Uralil, J, Lara-Tejero, M, Sukhan, A, Gala´n, JE, Aizawa, S-I: Supramolecular structure of the Salmonella typhimurium type III protein secretion system. Science 1998;280:602–605. Kubori. T, Shimamoto, N, Yamaguchi, S, Namba, K, Aizawa, S-I: Morphological pathway of flagellar assembly in Salmonella typhimurium. J Mol Biol 1992;226:433–446. Kubori, T, Sukhan, A, Aizawa, S-I, Gala´n, JE: Molecular characterization and assembly of the needle complex of the Salmonella typhimurium type III protein secretion system. Proc Natl Acad Sci USA 2000;97:10225–10230. Kutsukake, K: Excretion of the anti-sigma factor through a flagellar structure couples flagellar gene expression with flagellar assembly in Salmonella typhimurium. Mol Gen Genet 1994; 243:605–612. Kutsukake, K, Ikebe, T, Yamamoto, S: Two novel regulatory genes, fliT and fliZ, in the flagellar regulon of Salmonella. Genes Genet Syst 1999; 74:287–292. Kutsukake, K, Minamino, T, Yokoseki, T: Isolation and characterization of FliK-independent flagellation mutants from Salmonella typhimurium. J Bacteriol 1994a; 176:7625–7629. Kutsukake, K, Okada, T, Yokoseki, T, Iino, T: Sequence analysis of the flgA gene and its adjacent region in Salmonella typhimurium, and identification of another flagellar gene, flgN. Gene 1994b; 143:49–54. Kuwajima, G, Kawagishi, I, Homma, M, Asaka, J, Kondo, E, Macnab, RM: Export of an N-terminal fragment of Escherichia coli flagellin by a flagellum-specific pathway. Proc Natl Acad Sci USA 1989;86:4953–4957. Macnab, RM: 1996. Flagella and motility; in Neidhardt, FC, Curtiss III, R, Ingraham, JL, Lin, ECC, Low, KB, Magasanik, B, Reznikoff, WS, Riley, M, Schaechter, M, Umbarger, H (eds): Escherichia coli and Salmonella: Cellular and Molecular Biology, ed 2. Washington, ASM Press, pp 123–145. Maki, S, Vonderviszt, F, Furukawa, Y, Imada, K, Namba, K: Plugging interactions of HAP2 pentamer into the distal end of flagellar filament revealed by electron microscopy. J Mol Biol 1998;277:771–777. Maki-Yonekura, S, Yonekura, K, Namba, K: Domain movements of HAP2 in the cap-filament complex formation and growth process of the bacterial flagellum. Proc Natl Acad Sci USA 2003;100:15528–15533.

J Mol Microbiol Biotechnol 2004;7:5–17

Makishima, S, Komoriya, K, Yamaguchi, S, Aizawa, S-I: Length of the flagellar hook and the capacity of the type III export apparatus. Science 2001;291:2411–2413. Mimori, Y, Yamashita, I, Murata, K, Fujiyoshi, Y, Yonekura, K, Toyoshima, C, Namba, K: The structure of the R-type straight flagellar filament of Salmonella at 9 Å resolution by electron cryomicroscopy. J Mol Biol 1995;249:69– 87. Mimori-Kiyosue, Y, Vonderviszt, F, Namba, K: Locations of terminal segments of flagellin in the filament structure and their roles in polymerization and polymorphism. J Mol Biol 1997;270:222–237. Mimori-Kiyosue, Y, Vonderviszt, F, Yamashita, I, Fujiyoshi, Y, Namba, K: Direct interactions of flagellin termini essential for polymorphic ability of flagellar filament. Proc Natl Acad Sci USA 1996, 93:15108–15113. Mimori-Kiyosue, Y, Yamashita, I, Fujiyoshi, Y, Yamaguchi, S, Namba, K: Role of the outermost subdomain of Salmonella flagellin in the filament structure revealed by electron cryomicroscopy. J Mol Biol 1998;284:521–530. Minamino, T, Chu, R, Yamaguchi, S, Macnab, R.M: Role of FliJ in flagellar protein export in Salmonella. J Bacteriol 2000a; 182:4207– 4215. Minamino, T, Doi, H, Kutsukake, K: Substrate specificity switching of the flagellum-specific export apparatus during flagellar morphogenesis in Salmonella typhimurium. Biosci Biotechnol Biochem 1999a, 63:1301–1303. Minamino, T, Gonza´lez-Pedrajo, B, Kihara, M, Namba, K, Macnab, RM: The ATPase FliI can interact with the type III flagellar protein export apparatus in the absence of its regulator, FliH. J Bacteriol 2003;185:3983–3988. Minamino, T, Gonza´lez-Pedrajo, B, Oosawa, K, Namba, K, Macnab, RM: Structural properties of FliH, an ATPase regulatory component of the Salmonella type III flagellar export apparatus. J Mol Biol 2003;322:281–290. Minamino, T, Gonza´lez-Pedrajo, B, Yamaguchi, K, Aizawa, S, Macnab, R.M: FliK, the protein responsible for flagellar hook length control in Salmonella, is exported during hook assembly. Mol Microbiol 1999b, 34:295–304. Minamino, T, Iino, T, Kutsukake, K: Molecular characterization of the Salmonella typhimurium flhB operon and its protein products. J Bacteriol 1994;176:7630–7637. Minamino, T, Macnab, RM: Components of the Salmonella flagellar export apparatus and classification of export substrates. J Bacteriol 1999; 181:1388–1394. Minamino, T, Macnab, RM: Domain structure of Salmonella FlhB, a flagellar export component responsible for substrate specificity switching. J Bacteriol 2000a; 182:4906–4919. Minamino, T, Macnab, RM: FliH, a soluble component of the type III flagellar export apparatus of Salmonella, forms a complex with FliI and inhibits its ATPase activity. Mol Microbiol 2000b;37:1494–1503.

Minamino/Namba

Minamino, T, Macnab, RM: Interactions among components of the Salmonella flagellar export apparatus and its substrates. Mol Microbiol 2000c; 35:1052–1064. Minamino, T, Tame, JRH, Namba, K, Macnab, RM: Proteolytic analysis of the FliH/FliI complex, the ATPase component of the type III flagellar export apparatus of Salmonella. J Mol Biol 2001;312:1027–1036. Minamino, T, Yamaguchi, S, Macnab, RM: Interaction between FliE and FlgB, a proximal rod component of the flagellar basal body of Salmonella. J Bacteriol 2000b;182:3029–3036. Morgan, DG, Macnab, RM, Francis, NR, DeRosier, DJ: Domain organization of the subunit of the Salmonella typhimurium flagellar hook. J Mol Biol 1993;229:79–84. Morgan, DG, Owen, C, Melanson, L A, DeRosier, DJ: Structure of bacterial flagellar filaments at 11 Å resolution; packing of the ·-helices. J Mol Biol 1995:249:88–110. Müller, V, Jones, CJ, Kawagishi, I, Aizawa, S-I, Macnab RM: Characterization of the fliE gene of Escherichia coli and Salmonella typhimurium and identification of the FliE protein as a component of the flagellar hook-basal body complex. J Bacteriol 1992;174:2298–2304. Namba, K, Yamashita, I, Vonderviszt, F: Structure of the core and central channel of bacterial flagella. Nature 1989;342:648–654. Nambu, T, Kutsukake, K: The Salmonella FlgA protein, a putative periplasmic chaperone essential for flagellar P ring formation. Microbiology 2000;146:1171–1178. Nambu, T, Minamino, T, Macnab, RM, Kutsukake, K: Peptidoglycan-hydrolyzing activity of the FlgJ protein, essential for flagellar rod formation in Salmonella typhimurium. J Bacteriol 1999;181:1555–1561.

Bacterial Flagellar Assembly and Type III Protein Export

Ohnishi, K, Fan, F, Schoenhals, GJ, Kihara, M, Macnab, RM: The FliO, FliP, FliQ, and FliR proteins of Salmonella typhimurium: Putative components for flagellar assembly. J Bacteriol 1997;179:6092–6099. Ohnishi, K, Ohto, Y, Aizawa, S-I, Macnab, RM, Iino, T: FlgD is a scaffolding protein needed for flagellar hook assembly in Salmonella typhimurium. J Bacteriol 1994;176:2272–2281. Samatey, FA, Imada, K, Nagashima, S, Vonderviszt, F, Kumasaka, T, Yamamoto, M, Namba, K: Structure of the bacterial flagellar protofilament and implications for a switch for supercoiling. Nature 2001;410:331–337. Suzuki, H, Yonekura, K, Namba, K: Structure of the rotor of the bacterial flagellar motor revealed by electron cryomicroscopy and singleparticle image analysis. J Mol Biol 2004;337: 105–113. Suzuki, T, Iino, T, Horiguchi, T, Yamaguchi, S: Incomplete flagellar structures in nonflagellate mutants of Salmonella typhimurium. J Bacteriol 1978;133:904–915. Vogler, AP, Homma, M, Irikura, VM, Macnab, RM: Salmonella typhimurium mutants defective in flagellar filament regrowth and sequence similarity of FliI to F0F1, vacuolar, and archaebacterial ATPase subunits. J Bacteriol 1991; 173:3564–3572. Vonderviszt, F, Aizawa, S-I, Namba, K: Role of the disordered terminal regions of flagellin in filament formation and stability. J Mol Biol 1991; 221:1461–1474. Vonderviszt, F, Imada, K, Furukawa, Y, Uedaira, H, Taniguchi, H, Namba, K: Mechanism of self-association and filament capping by flagellar HAP2. J Mol Biol 1998;284:1399–1416. Williams, AW, Yamaguchi, S, Togashi, F, Aizawa, S-I, Kawagishi, I, Macnab, RM: Mutations in fliK and flhB affecting flagellar hook and filament assembly in Salmonella typhimurium. J Bacteriol 1996;178:2960–2970.

Yamashita, I, Hasegawa, K, Suzuki, H, Vonderviszt, F, Mimori-Kiyosue, Y, Namba, K: Structure and switching of bacterial flagellar filament studied by X-ray fiber diffraction. Nature Struct Biol 1998;5:125–132. Yokoseki, T, Kutsukake, K, Ohnishi, K, Iino, T: Functional analysis of the flagellar genes in the fliD operon of Salmonella typhimurium. Microbiology 1995;141:1715–1722. Yokoseki, T, Iino, T, Kutsukake, K: Negative regulation by FliD, FliS, and FliT of the export of the flagellum-specific anti-sigma factor, FlgM, in Salmonella typhimurium. J Bacteriol 1996; 178:899–901. Yonekura, K, Maki, S, Morgan, DG, DeRosier, DJ, Vonderviszt, F, Imada, K, Namba, K: The bacterial flagellar cap as the rotary promotor of flagellin self-assembly. Science 2000;290:2148– 2152. Yonekura, K, Maki-Yonekura, S, Namba, K: Structure analysis of the flagellar cap-filament complex by electron cryomicroscopy and single particle image analysis. J Struct Biol 2001;133: 246–253. Yonekura, K, Maki-Yonekura, S, Namba, K: Complete atomic model of the bacterial flagellar filament by electron cryomicroscopy. Nature 2003;424:643–650. Yoshioka, K, Aizawa, S-I, Yamaguchi, S: Flagellar filament structure and cell motility of Salmonella typhimurium mutants lacking part of the outer domain of flagellin. J Bacteriol 1995;177: 1090–1093. Zhou, J, Lloyd, SA, and Blair, DF: Electrostatic interactions between rotor and stator in the bacterial flagellar motor. Proc Natl Acad Sci USA 1998;95:6436–6441. Zhu, K, Gonza´lez-Pedrajo, B, Macnab, RM: Interactions among membrane and soluble components of the flagellar export apparatus of Salmonella. Biochemistry 2002;41:9516–9524.

J Mol Microbiol Biotechnol 2004;7:5–17

17

J Mol Microbiol Biotechnol 2004;7:18–29 DOI: 10.1159/000077866

Dual Flagellar Systems Enable Motility under Different Circumstances Linda L. McCarter Department of Microbiology, The University of Iowa, Iowa City, Iowa, USA

Key Words Flagella W Motility W Swimming W Swarming W Vibrio parahaemolyticus

Abstract Flagella are extremely effective organelles of locomotion used by a variety of bacteria and archaea. Some bacteria, including Aeromonas, Azospirillum, Rhodospirillum, and Vibrio species, possess dual flagellar systems that are suited for movement under different circumstances. Swimming in liquid is promoted by a single polar flagellum. Swarming over surfaces or in viscous environments is enabled by the production of numerous peritrichous, or lateral, flagella. The polar flagellum is produced continuously, while the lateral flagella are produced under conditions that disable polar flagellar function. Thus at times, two types of flagellar organelles are assembled simultaneously. This review focuses on the polar and lateral flagellar systems of Vibrio parahaemolyticus. Approximately 50 polar and 40 lateral flagellar genes have been identified encoding distinct structural, motor, export/assembly, and regulatory elements. The sodium motive force drives polar flagellar rotation, and the proton motive force powers lateral translocation. Polar genes are found exclusively on the large chromo-

ABC

© 2004 S. Karger AG, Basel

Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/mmb

some, and lateral genes reside entirely on the small chromosome of the organism. The timing of gene expression corresponds to the temporal demand for components during assembly of the organelle: RpoN and lateral- and polar-specific Û54-dependent transcription factors control early/intermediate gene transcription; lateral- and polarspecific Û28 factors direct late flagellar gene expression. Although a different gene set encodes each flagellar system, the constituents of a central navigation system (i.e., chemotaxis signal transduction) are shared. Copyright © 2004 S. Karger AG, Basel

Introduction

Propulsion by flagella is an extremely effective mode of motility, and these organelles are responsible for movement in many bacteria and some archaea. Flagella act as helical propellers that are F20 nm in diameter and F10– 20 Ìm in length [reviewed in Doetsch and Sjoblad, 1980; Namba and Vonderviszt, 1997]. Rotary motors embedded in the membrane power these propellers [reviewed in Berg, 2003]. The motors are energized by the transmembrane potential of specific ions, most commonly the proton motive force or the sodium motive force [Imae and Atsumi, 1989; Manson et al., 1977]. Passage of ions

Linda L. McCarter Department of Microbiology The University of Iowa Iowa City, Iowa 52246 (USA) Tel. +1 319 335 9721, Fax +1 319 335 7679, E-Mail [email protected]

gellins (e.g., Caulobacter or Vibrio species). In electron micrographs, some flagella appear smooth (e.g., Caulobacter crescentus) whereas others possess knobby protrusions (e.g., S. typhimurium) [Trachtenberg and DeRosier, 1987, 1988; Trachtenberg et al., 1987, 1998]. The flagella of Sinorhizobium meliloti and Sinorhizobium lupini appear to have grooved surfaces, which have been postulated to improve motility in viscous environments [Cohen-Krausz and Trachtenberg, 2003; Götz et al., 1982; Trachtenberg et al., 1987].

through the motor is coupled to rotation of the flagellum. The motor can spin at speeds of F100–1,500 Hz [Kudo et al., 1990; Lowe et al., 1987; Magariyama et al., 1994], translating into speeds of many cell body lengths per second (commonly 20–60 Ìm/s) [Macnab and Aizawa, 1984; Wilson and Beveridge, 1993]. The direction of rotation of the motor can be reversible, and modulation of the direction of rotation in response to signals in the environment results in taxis, allowing movement towards more favorable conditions [reviewed in Blair, 1995]. Although the mechanism of flagellar function is conserved, the numbers and arrangement of flagella are amazingly diverse. Some bacteria elaborate a single polar flagellum, which can be sheathed, e.g., Vibrio cholerae and Bdellovibrio bacteriovorus [Hranitzky et al., 1980; Thomashow and Rittenberg, 1985a, b] or unsheathed, e.g., Pseudomonas aeruginosa, Caulobacter crescentus, and Campylobacter species [Hernandez and Monge-Najera, 1994; McCoy et al., 1975; Trachtenberg and DeRosier, 1988; Wilson and Beveridge, 1993]. Other bacteria display tufts of multiple polar flagella (usually 2–7 in number, which also can be sheathed, e.g., Vibrio fischeri [Baumann and Baumann, 1981] and Helicobacter pylori [Geis et al., 1993], or unsheathed, e.g., Pseudomonas putida [Harwood et al., 1989]. The nature and function of the sheath are somewhat mysterious [Fuerst, 1980; Sjoblad et al., 1983]. In electron micrographs it appears continuous with the cell outer membrane; however, compositional analyses suggest that it may be a distinct domain both with respect to protein and membrane content [Fuerst and Perry, 1988; Furuno et al., 2000; Hranitzky et al., 1980; Sjoblad et al., 1983; Thomashow and Rittenberg, 1985a]. Peritrichously flagellated cells (which are characteristic of the enteric bacteria) are arranged around the cell body and are generally unsheathed. E. coli and S. typhimurium possess F4–8 peritrichous, or lateral, flagella when grown in liquid medium [Macnab, 1996]. Most unusual are the periplasmic, or endoflagella, of the spirochetes. These flagella are attached subterminally to the cell ends and are found internal to the cell outer membrane [reviewed in Charon and Goldstein, 2002; Li et al., 2000]. Unfolded flagellin monomers pass through a central channel in the filament (F3 nm in diameter) and polymerize at the distal end [Namba and Vonderviszt, 1997; Yonekura et al., 2002, 2003]. Composition of the selfassembling flagellar filament also varies [Joys, 1988; Wilson and Beveridge, 1993]. Flagella can be polymerized from single flagellin subunits (e.g., F20,000 subunits polymerize to form the filament of E. coli) or multiple fla-

Some bacteria produce two kinds of flagella. Usually, a single polar flagellum (Fla) is present under all growth conditions and lateral flagella (Laf) are produced when cells are grown on solid media. The polar flagellum is used for locomotion in liquid (swimming) and the lateral flagella are used for movement over surfaces or through viscous environments (swarming). In some of these organisms, surface sensing, i.e., induction of the lateral flagella in response to growth on surfaces, has been linked with performance of the polar organelle. The best studied examples are described below. Certain members of the ·-proteobacterial genus Azospirillum, including A. brasilense, A. lipoferum, and A. irakense [Hall and Krieg, 1984b; Khammas et al., 1989; Tarrand et al., 1978], display mixed flagellation and swarming over the surface can be observed on 0.4–0.8% agar [Hall and Krieg, 1984b; Moens et al., 1995]. The lateral flagella appear thinner and have a shorter waveform than the polar flagellum [Hall and Krieg, 1984a]. These Azospirillum species are nitrogen-fixing soil organisms that colonize the rhizosphere of cereals and grasses. Motility and chemotaxis have been demonstrated to be important for plant colonization. The polar flagellum plays a key role in adsorption to roots, whereas the lateral flagella do not appear important for initial stages of colonization [Moens et al., 1995]. Hindrance of polar flagellar rotation induces expression of the structural gene encoding lateral flagellin [Moens et al., 1996]. Some mutants lack both kinds of flagella, suggesting there may be common structural or assembly elements [Moens et al., 1996]. Other mutants, lacking a polar flagellum, constitutively express lateral flagella in liquid [Alexandre et al., 1999]. Some evidence suggests that a polar flagellum is not absolutely required for surface sensing: increasing viscosity of the environment further increases lateral flagellar gene expression in

Polar and Lateral Flagella

J Mol Microbiol Biotechnol 2004;7:18–29

Survey of Bacteria with Mixed Patterns of Flagellation

19

Fig. 1. The swimmer cell possesses a single, sheathed polar flagellum. Electron micrograph of V. parahaemolyticus grown in liquid medium and stained with phosphotungstic acid. Bar is F1 Ìm.

liquid even in a mutant lacking a polar flagellum [Alexandre et al., 1999]. In liquid medium, the ·-purple photosynthetic bacterium Rhodospirillum centenum possesses a single sheathed polar flagellum. When it is grown on solid medium, numerous unsheathed peritrichous flagella are synthesized [McClain et al., 2002; Ragatz et al., 1995]. The lateral flagella enable R. centenum to move over surfaces (e.g., 0.8% agar) in response to light [Ragatz et al., 1994]. Although lateral rotation occurs in the absence of a functional polar flagellum, and separate polar and lateral motor and switch elements have been identified, surface movement requires functional polar and lateral flagella [Jiang et al., 1998; McClain et al., 2002]. A third example of mixed flagellation is found in some Aeromonas strains [Shimada et al., 1985]. The aeromonads are aquatic bacteria that are significant pathogens of amphibians, fish, reptiles and humans. The single unsheathed polar flagellum, which is expressed constitutively promotes swimming in liquid and has been shown to be important for colonization of fish [Merino et al., 1997] and human cells [Rabaan et al., 2001]. Numerous lateral flagella propel the bacteria over surfaces (e.g., 0.5% Eiken agar) [Kirov et al., 2002]. Mutants with defects in the lateral flagellar system also show decreased adherence to human epithelial cells and biofilm formation [Gavin et al., 2002]. Moreover, the introduction of lateral flagellar genes into lateral-flagella-negative strains increases adhesion and invasion of HEp-2 cells and the capacity for biofilm formation [Gavin et al., 2003]. Thus, Kirov proposes that lateral flagella are virulence factors due to their contributions to host cell invasion and biofilm formation [Kirov, 2003]. Like R. centenum, some mutations that abolish polar flagellar function also eliminate lateral flagella [Altarriba et al., 2003; Kirov et al., 2002]; however, unlike R. centenum, genetic evidence suggests that the polar and

20

J Mol Microbiol Biotechnol 2004;7:18–29

lateral flagellar motors are shared in Aeromonas species [Gavin et al., 2002]. Many marine Vibrio species possess sheathed polar and unsheathed lateral flagellar systems, including the majority of V. campbellii, V. neptuna, V. aginolyticus and V. parahaemolyticus strains [Allen and Baumann, 1971] and some strains of V. harveyi and V. splendida [Baumann and Baumann, 1981]. Surface-associated movement is quite vigorous, and robust swarming occurs on 1.5% agar. The best studied dual flagellar systems of the marine Vibrio are those of V. parahaemolyticus, and these systems are described in more depth below.

The Dual Flagellar Systems of V. parahaemolyticus

V. parahaemolyticus is a ubiquitous gram-negative marine bacterium and human pathogen [Joseph et al., 1982; McCarter, 1999]. It can be isolated as a free-living bacterium and from a variety of animate and inanimate surfaces [Kaneko and Colwell, 1975]. In liquid medium the bacterium swims using a sheathed polar flagellum (fig. 1) [Allen and Baumann, 1971; Shinoda and Okamoto, 1977]. This is an extremely effective propulsive organelle. Swimming speeds average 60 Ìm per second [Kawagishi et al., 1996]. The sodium motive force powers the polar flagellum [Atsumi et al., 1992; Yorimitsu and Homma, 2001]. Rotation has been measured (in the closely related bacterium V. alginolyticus) by using laser darkfield microscopy at rates as fast as 100,000 rpm compared with F15,000 rpm for the proton-driven peritrichous flagella of E. coli. [Lowe et al., 1987; Magariyama et al., 1994, 1995]. On contact with surfaces or viscous environments, however, the polar flagellum is not an effective propulsive organelle. Such conditions impede rotation of the polar flagellum. As the polar flagellum slows down, swarmer cell differentiation occurs [Belas et al., 1986; Kawagishi et al., 1996; McCarter et al., 1988]. Cell division is suppressed, resulting in elongated cells, and the second flagellar system is elaborated (fig. 2) [McCarter and Silverman, 1990; Shinoda and Okamoto, 1977]. Production of the polar flagellum is continuous, so at times the bacterium produces two distinct types of organelles, i.e., the single-sheathed polar and numerous lateral flagella (fig. 3). Lateral flagella, which are driven by the proton motive force [Atsumi et al., 1992], enable rapid colonization of surfaces. An example of a swarming colony, which has been inoculated in the center of a petri plate with 1.5% agar, is shown in figure 4.

McCarter

Flagellar Components and Genes in V. parahaemolyticus

Flagellar structure and assembly have been studied in many organisms and most intensively in E. coli and S. typhimurium [reviewed in Macnab, 1986, 1996, 1999, 2003]. Approximately 35 genes are required to encode components of the flagellum and its export structure. In V. parahaemolyticus, these components are entirely distinct for each flagellar system. Both of the polar and lateral flagellar gene sets encode homologs for most of the components found in E. coli. Mutants with defects in swimming motility show no defect in swarming, and swarmingdefective mutants are swim competent [Kim and McCarter, 2000; McCarter et al., 1988; McCarter, 1995; Stewart and McCarter, 2003]. Genes for the polar and lateral systems are listed in tables 1 and 2, respectively. They are named with the common designations used in other flagellar systems when identifiable counterparts exist; the subscripts P and L are used to distinguish polar and lateral genes, respectively. The genome sequence of V. parahaemolyticus has recently been completed (http://genome.gen-info.osakau.ac.jp/bacteria/vibrio/); the organism has two circular chromosomes with a total of 5.2 Mb of DNA [Makino et al., 2003]. Polar flagellar genes are found exclusively on the large chromosome (which is F3.3 Mb), and all of the lateral flagellar genes are on the small chromosome (which is F1.9 Mb). There is no general bias in the percent GC content, which is F46% for lateral as well as polar genes (compared to 45% for the entire genome). The products of many of the flagellar genes can be described much like those of an engine [Aizawa, 1996; Berg, 2002; Macnab, 2003]. The stator contains MotA and MotB, which participate in torque generation. There is a rotor, containing FliF, FliG, and the C-ring components FliM and FliN. The drive shaft, or flagellar rod, contains FlgB, FlgC, FlgF, and FlgG. FlgH and FlgI form bushings, called the L and P rings, that are found in the outer membrane and peptidoglycan layer, respectively. The bushings surround the drive shaft and keep it in place. There is a flexible universal joint called the hook (FlgE) that joins the drive shaft to the propeller via two adaptor proteins or hook-associated proteins, HAP1 (FlgK) and HAP2 (FlgL). The propeller, or flagellar filament, is polymerized from multiple flagellin subunit monomers called flagellins. At the tip of the filament is a cap, called FliD or HAP3. The flagellum is assembled sequentially, initiating with the membrane-imbedded basal body (rotor, drive shaft,

Polar and Lateral Flagella

Fig. 2. Elongated swarmer cell possessing numerous unsheathed lateral flagella in addition to the sheathed polar. Electron micrograph of V. parahaemolyticus grown on solid medium and stained with phosphotungstic acid. Bar is F1 Ìm. Lateral flagella tend to bundle, which gives the appearance of variable flagellar thickness in this photograph.

Fig. 3. Zoom image of polar and lateral flagella. Electron micrograph

of V. parahaemolyticus grown on solid medium and stained with phosphotungstic acid. Large arrow points to sheathed polar flagellum, and smaller arrow shows lateral filament. The diameter of the sheathed polar flagellum is F 30 nm, and the unsheathed lateral is F15 nm.

Fig. 4. Swarming over surfaces. V. parahaemolyticus (carrying a

green fluorescence reporter) was inoculated in the center of a 1.5% agar plate (Heart Infusion medium) and incubated at 30 ° C. Photograph was taken 10 h after inoculation by using a Fuji Luminescence Imager.

J Mol Microbiol Biotechnol 2004;7:18–29

21

Table 1. Chromosome 1: V. parahaemolyticus polar flagellar and chemotaxis genes

Genea

VP locus tagb

Predicted function

Region 1 flgN flgM flgA cheV cheR flgB flgC flgD flgE flgF flgG flgH flgI flgJ flgK flgL flaC flaD flaE

VP0770 VP0771 VP0772 VP0773 VP0774 VP0775 VP0776 VP0777 VP0778 VP0780 VP0781 VP0782 VP0783 VP0784 VP0785 VP0786 VP0788 VP0790 VP0791

Chaperone for FlgK and FlgL Anti-Û28 factor P-ring chaperone/assembly CheW/CheY hybrid Chemotaxis methyl transferase Proximal rod Proximal rod Hook assembly Hook Proximal rod Distal rod L ring P ring peptidoglycan hydrolase HAP1 (hook-filament junction) HAP3 (hook-filament junction) flagellin flagellin flagellin

Region 2 flaF flaB flaA flaG fliD flaI fliS flaK flaL flaM fliE

VP2261 VP2259 VP2258 VP2257 VP2256 VP2255 VP2254 VP2253 VP2252 VP2251 VP2250

fliF fliG fliH fliI fliF fliJ fliK

VP2249 VP2248 VP2247 VP2246 VP2249 VP2245 VP2244

flagellin flagellin flagellin unknown HAP2; filament cap (a.k.a. flaH ) unknown flagellin chaperone (a.k.a. flaJ ) Û54-dependent regulator Two-component sensor kinase Two-component response regulator Hook-basal body component; MS ring-rod junction MS ring Rotor/Switch component Fla export; negative regulator of FliI Fla export; ATPase MS ring Fla export Hook-length control

and bushings), followed by addition of the hook and finally the filament. There are proteins that act as specific chaperones for different classes of exported flagellar substrates. A dedicated type III secretion apparatus, containing FlhA, FlhB, FliO, FliP, FliQ and FliR, FliH, FliI, and FliJ, directs morphogenesis. There are some differences between E. coli/S. typhimurium and V. parahaemolyticus flagellar gene systems. The lateral, but not the polar, system lacks a fliO homolog.

22

J Mol Microbiol Biotechnol 2004;7:18–29

Genea

VP locus tagb

Predicted function

fliL fliM fliN fliO fliP fliQ fliR flhB flhA flhF

VP2243 VP2242 VP2241 VP2240 VP2239 VP2238 VP2237 VP2236 VP2235 VP2234

flhG

VP2233

fliA cheY

VP2232 VP2231

cheZ cheA cheB ORF 1

VP2230 VP2229 VP2228 VP2227

ORF 2 cheW ORF 3

VP2226 VP2225 VP2224

unknown C ring/Switch component C ring/Switch component Fla export Fla export Fla export Fla export Fla export Fla export Flagellar GTP-binding protein; homologous to FtsY Conserved domain COG0455 (ATPase involved in cell division & chromosome binding); flagellar synthesis regulator (a.k.a. fleN ) RNA polymerase Û28 factor Response regulator; causes change in direction of flagellar rotation CheY phosphatase Histidine autokinase Chemotaxis methyl esterase Unknown; Soj-like and other chromosome-partitioning ATPase proteins Unknown (370 aa) Chemotaxis coupling protein Unknown (163 aa)

Motor genes motA VP0689 motB VP0690 motX VP2811 motY VP2111

Na+ motor stator/force generator Na+ motor stator/force generator Na+ motor component Na+ motor component

a

Genes were assigned designations when a clear homolog was identified or a mutation in the coding region caused altered motility. The subscripts P or L are used to distinguish polar and lateral genes when necessary. b VP tag number is plain or boldfaced corresponding to a counterclockwise or clockwise (respectively) direction of gene transcription on the genome sequence map.

Four genes are required for the polar, sodium-type motor function: motA, motB, motX, and motY [Boles and McCarter, 2000]. Three genes are required for the lateral, proton-type motor function: motA, motB, and motY [Stewart and McCarter, 2003]. In contrast, only two motor genes, motA and motB, encode the proton-driven E. coli motors. There are six polar flagellin genes: flaA, B, C, D, E, and F [Kim and McCarter, 2000]. Similar to E. coli, a single gene (lafA) encodes subunits for the lateral filament

McCarter

[McCarter and Wright, 1993]. The polar flagellar system possesses two genes not found in E. coli, flhF and flhG, but which can be found in other bacteria, e.g., Pseudomonas, Bacillus, Campylobacter, Borrelia and other Vibrio species. FlhF shows homology to FtsY, which is a GTP-binding protein, that is part of the signal recognition particletargeting pathway. In Bacillus subtilis, FlhF is required for motility [Carpenter et al., 1992]. In P. putida it has been implicated as playing a role in polar flagellar placement [Pandza et al., 2000]. Vibrio FlhF homologs contain an insertion of F170 amino acids not found in other FlhF sequences derived from organisms possessing protondriven flagella. FlhG, which shows homology to MinD and some other ATPases, has been best studied in Pseudomonas species where it is named FleN and believed to act as an anti-activator of a flagellar regulatory protein [Dasgupta et al., 2000; Dasgupta and Ramphal, 2001].

Chemotaxis Genes in V. parahaemolyticus

The navigation system, i.e., chemotaxis, is one point of overlap between the dual flagellar systems. Potential chemotaxis genes are listed in table 3. Transposon mutants with defects in chemotaxis have been isolated and the insertions map in two regions on chromosome 1 near polar flagellar gene clusters. By using mutant strains that are proficient in only one of the two motility systems (i.e., Fla+Laf– and Fla–Laf+) [Sar et al., 1990], effects of chemotaxis defects could be assessed separately for swimming and swarming. Insertions in the potential operon encoding cheYZABW or cheVR affect both polar and lateral motility [L.M., unpubl.]. The genome sequence reveals that there is only one set of the central cytoplasmic chemotaxis proteins. This configuration is surprisingly different from V. cholerae, P. aeruginosa, and several other organisms, which appear (by sequence analysis) to have multiple central chemotaxis genes, and is more like E. coli and S. typhimurium. Multiple homologs (four) exist only for CheV in V. parahaemolyticus. One homolog (VP0773) maps within a polar flagellar cluster and is contiguous with cheR. CheV looks like a CheW/CheY hybrid. Although CheV does not exist in E. coli, genetic studies in B. subtilis suggest that function of CheV and CheW overlap [Rosario et al., 1994]. The genome sequence annotation suggests there is a second CheY (VP1376) on chromosome 1. Close inspection suggests that it may be a transcriptional type response regulator with its own cognate sensor histidine kinase rather than a second chemotaxis response regulator; moreover,

Polar and Lateral Flagella

Table 2. Chromosome 2: V. parahaemolyticus lateral flagellar genes

Genea

VP locus tagb

Predicted function

Region 1 flgN flgM flgA flgB flgC flgD flgE flgF flgG flgH flgI flgJ flgK flgL flgU

VPA0061 VPA0062 VPA0063 VPA0264 VPA0265 VPA0266 VPA0267 VPA0268 VPA0269 VPA0270 VPA0271 VPA0272 VPA0273 VPA0274 VPA0275

Chaperone for FlgK and FlgL Anti-Û28 factor P-ring chaperone/assembly Proximal rod Proximal rod Hook assembly Hook (a.k.a. lfgE and lafX ) Proximal rod Distal rod L ring P ring peptidoglycan hydrolase HAP1; hook-filament junction HAP3; hook-filament junction Unknown but required

Region 2 fliJ fliI fliH fliG fliF fliE

VPA1532 VPA1533 VPA1534 VPA1535 VPA1536 VPA1537

lafK motY fliM fliN fliP fliQ fliR flhB flhA ORF lafA ORF fliD fliS fliT fliK fliL fliA motA

VPA1538 VPA1539 VPA1540 VPA1541 VPA1542 VPA1543 VPA1544 VPA1545 VPA1546 VPA1547 VPA1548 VPA1549 VPA1550 VPA1551 VPA1552 VPA1553 VPA1554 VPA1555 VPA1556

motB

VPA1557

Laf export Laf export; ATPase Laf export; negative regulator of FliI Rotor/Switch component MS ring Hook-basal body component; MS ring-rod junction Û54-dependent regulator H+ motor component C ring/Switch component C ring/Switch component Laf export Laf export Laf export Laf export Laf export unknown flagellin unkown HAP2 (a.k.a. lafB ) (a.k.a. lafC ) (a.k.a. lafD ) Hook-length control (a.k.a. lafK ) Unknown (a.k.a. lafL ) RNA polymerase Û28 factor (a.k.a. lafS ) H+ motor stator/force generator (a.k.a. lafT ) H+ motor stator/force generator (a.k.a. lafU )

a

Genes were assigned designations when a clear homolog was identified or a mutation in the coding region caused altered motility. The subscripts P or L are used to distinguish polar and lateral genes when necessary. b VP tag number is plain or boldfaced corresponding to a counterclockwise or clockwise (respectively) direction of gene transcription on the genome sequence map.

J Mol Microbiol Biotechnol 2004;7:18–29

23

Table 3. Potential V. parahaemolyticus chemotaxis genes

Chromosome 1 genea

VP locus tag

Chromosome 2 predicted function/ conserved domainsb

gene

Predicted cytoplasmic products cheV VPO773 CheW/CheY hybrid VPO2037 CheW/CheY hybrid cheR VPO774 Methyl transferase cheY VP2231 Response regulator tumble effector cheW VP2225 Chemotaxis adapter protein cheB VP2228 Methyl esterase cheA VP2229 Sensor kinase cheZ VP2230 CheY phosphatase Predicted sensors involved in signal reception and transduction (potential chemoreceptors or methyl-accepting taxis proteins)c A, Tar VP0183 B, Tar VP0421 A, Chase3, Tar, VP0963 A, Tar VP1088 A, Tar VP1185 C, Tar VP1486 B, Tar VP1628 A, Tar VP1892 A, Cache, Tar VP1904 A, Cache, Tar VP1981 A, Cache, Tar VP2159 A, Tar VP2629 A, Cache, Tar VP2827

VP locus tag

predicted function/ conserved domainsb

VPA0431 VPA0746

CheW/CheY hybrid CheW/CheY hybrid

VPA0024 VPA0491 VPA0511 VPA0554 VPA0562 VPA0596 VPA0612 VPA0693 VPA0842 VPA1000 VPA1182 VPA1189 VPA1449 VPA1462 VPA1492 VPA1651

A, Tar A, Tar A, Tar D, Tar A, Tar B, Pas, Tar D, Pas/Pac, Tar A, Tar A, Tar A, Tar C, Tar A, Tar A, Tar A, Tar A, Tar A, Tar

a

Mutations that map in or near boldfaced genes cause defects in chemotaxis [Sar et al., 1990; LM unpubl.]. Roles of other potential genes are unknown and therefore gene names are not assigned. b Searches for homology were performed at http://www.ncbi.nlm.nih.gov/BLAST and conserved domains are noted. Tar (COG0840) is described as a methyl-accepting chemotaxis domain; Chase3 (pfam05227.1) is an extracellular sensory domain found in various transmembrane receptors that are parts of signal transduction pathways [Zhulin et al., 2003]; Cache is a signaling domain common to a class of prokaryotic chemotaxis receptors [Anantharaman and Aravind, 2000]; Pas/Pac domains (AtoS; COG2202.1) bind ligands and act as sensors involved in signal transduction; the Pac domain (cd00130) is recognized as part of the ß-scaffold usually found with the PAS core [Taylor and Zhulin, 1999]. c Potential chemotaxis receptor proteins are given letter designations to specify subfamily type according to Falke and Kim [2000]: A, two transmembrane domains with intervening periplasmic domain, B, two transmembrane domains with few intervening aa, C, one transmembrane domain, and D, no transmembrane domains. Gene designations have not yet been assigned. Transmembrane domain predictions were performed using the DAS transmembrane prediction server http://www.sbc.su.se/Fmiklos/DAS/maindas.html [Cserzo et al., 1997].

this gene has been disrupted with no effect on motility [L.M., unpubl.]. Like V. cholerae and P. aeruginosa but in contrast to E. coli and S. typhimurium, a plethora of potential chemoreceptors (MCPs) is found within the genome; 13 on

24

J Mol Microbiol Biotechnol 2004;7:18–29

chromosome 1 and 16 on chromosome 2 (listed in table 3). It suggests a great capacity for sensing and responding to signals in the environment. None of the genes encoding the potential methyl-accepting taxis proteins are linked to flagellar genes. MCP localization has

McCarter

Table 4. Regulatory and morphogenetic pathway for V. parahaemolyticus polar flagellar biogenesisa

Early

Middle

Late

FlaK (Û54-interacting regulator) FlaL & M (two-component sensor response regulator pair)

Switch, export and assembly apparatus Û28 CheZABW chemotaxis proteins Hook and basal body parts Junction proteins (HAP1 and HAP3) MotY Na+-motor protein FlhF and G

MotA and B Na+-motor proteins MotX Na+-motor protein HAP2 (filament cap) Chaperones Flagellins (6) CheV and R chemotaxis proteins FlgM and N (anti-Û28 and chaperone)

a A cascade of gene expression has been elucidated [Kim and McCarter, 2001], and the gene products are divided into categories that correspond to early, middle, and late gene expression. RpoN (Û54) is also required for polar motility [Stewart and McCarter, 2003].

Table 5. Regulatory and morphogenetic

pathway for V. parahaemolyticus lateral flagellar biogenesisa

Early

Middle

LafK (Û54-interacting regulator) ?

MotY H+-motor protein Switch, export and assembly apparatus Hook and basal body parts Û28

Late Û28 Flagellin (1) Chaperone Junction proteins (HAP1, 2 and 3) MotA and B H+-motor proteins FlgM and N (anti-Û28 and chaperone)

a

A cascade of gene expression has been elucidated [Stewart and McCarter, 2003], and the gene products are divided into categories that correspond to early, middle and late gene expression. RpoN (Û54) is required for lateral motility. Additional regulators most probably exist at the top of the hierarchy.

been performed in V. parahaemolyticus using antibodies directed against the E. coli chemoreceptor Trg [Gestwicki et al., 2000]. In swimmer cells, MCPs localize to both poles, and in swarmer cells MCPs are found at the poles and at intervals along the cell body.

Flagellar systems require a sizable commitment in terms of numbers of genes, amount of protein synthesis and energy expended for rotation. As a result, all flagellar systems that have been examined are stringently regulated. Regulation of flagellar gene expression and protein production is temporally coupled to assembly of the flagellar organelle [reviewed in Aldridge and Hughes, 2002]. Multiple checkpoints exist gating transcriptional and post

transcriptional events to coordinate assembly of the flagellum. Since the pathway of organelle assembly is generally conserved, temporal regulation of flagellar gene expression is also conserved. Differences occur primarily in the particular solutions, such as transcription factors, employed to coordinate gene expression and component production. Transcriptional regulation of both polar and lateral flagellar genes has been explored. Tables 4 and 5 depict the simplest models of the inferred morphogenetic pathway for flagellar assembly based on the temporal hierarchy of gene expression. Regulation of polar flagellar genes appears similar in many Vibrio and Pseudomonas species [Arora et al., 1997; Klose et al., 1998; McCarter, 2001; Prouty et al., 2001; Ritchings et al., 1995]. Û54-dependent transcription factors regulate intermediate gene expression, and a polar flagellar Û28 directs late polar gene expression. Although the lateral genes encode peritri-

Polar and Lateral Flagella

J Mol Microbiol Biotechnol 2004;7:18–29

Regulation of V. parahaemolyticus Flagellar Gene Expression

25

Why Two Flagellar Systems?

Fig. 5. Lateral flagella production in liquid by a strain lacking a polar

flagellum. Electron micrograph of V. parahaemolyticus Fla– mutant strain grown in liquid medium and stained with phosphotungstic acid. Bar is F1 Ìm. Lateral flagellar bundling is apparent.

chous, proton-driven flagella employed for swarming, regulation is more similar to the polar system than it is to the regulatory hierarchy of peritrichous, proton-driven flagella employed for swimming and swarming by E. coli, S. typhimurium or P. mirabilis. Specifically, intermediate lateral gene expression is not dependent on FlhDC homologs. It requires Û54 and a Û54-dependent transcription factor, LafK, which is similar to the polar regulator FlaK [Stewart and McCarter, 2003]. Late lateral gene expression requires a lateral flagellar-specific Û28. Details of each of these regulatory cascades remain to be elucidated, and most certainly additional levels of regulation exist. For example, lateral flagellar production is also known to be regulated by the availability of iron [McCarter and Silverman, 1989], OpaR, a V. harveyi LuxR homolog [McCarter, 1998], and a GGDEF-EAL motif-containing protein ScrC [Boles and McCarter, 2002]. The two flagellar systems interact. All mutations that have been isolated affecting polar flagellar performance cause expression of lateral flagellar genes in liquid. These include mutations that affect assembly (Fla–) (fig. 5) and rotation (Mot–) [Boles and McCarter, 2000; McCarter et al., 1988; McCarter, 2001]. Mutations in the central chemotaxis genes (Che–) do not perturb lateral flagellar gene expression [L.M. unpubl. obs.]. The genetic linkage between performance of the polar system and induction of the lateral flagellar system is not known.

26

J Mol Microbiol Biotechnol 2004;7:18–29

In summary, V. parahaemolyticus possesses two distinct motility systems adapted for life in different circumstances. The polar flagellar system propels the bacterium in liquid and the lateral flagella move the bacteria through viscous environments and over surfaces. The polar filament, composed of multiple flagellin subunits, is sheathed by membrane and can rotate extraordinarily fast using energy derived from the sodium membrane potential. Lateral filaments, composed of a single flagellin, can also rotate, albeit not a quickly as the polar, using energy derived from the proton motive force. Polar flagella are produced continuously whereas lateral flagellar are synthesized only under conditions that do not favor polar flagellar rotation, i.e., growth on surfaces or viscous environments. Lateral flagella perform well under such conditions and allow movement over and colonization of surfaces. Why two flagellar systems? A second flagellar system does not seem to be requisite for swarming. Robust swarming requires hyperflagellation. Many bacteria swarm simply by upregulating existing peritrichous flagella, e.g., E. coli, S. typhimurium, and Serratia and Proteus species [Fraser and Hughes, 1999; Harshey, 1994; Harshey and Matsuyama, 1994]. In particular, swarming of P. mirabilis is tremendously vigorous. So, overproduction of the same flagella used for swimming can be an effective means of propulsion on surfaces. However, in liquid environments, polar flagella generally outperform peritrichous flagella [Macnab, 1976; Wilson and Beveridge, 1993]. For example, swimming speeds (measured by laser dark-field microscopy at 30 ° C) of polarly propelled V. alginolyticus and peritrichously flagellated Salmonella typhimurium were 100 and 43 Ìm/s, respectively [Magariyama et al., 2001]. Thus, a polar flagellum seems to provide maximal swimming proficiency, whereas numerous lateral flagella support the most vigorous swarming. Motility may have important consequences with respect to survival and colonization of specific niches. For example, studies using V. alginolyticus and other bacteria have shown that adhesion to glass surfaces is directly proportional to swimming speed [Kogure et al., 1998; Morisaki et al., 1999]. It seems that some bacteria choose a modest lifestyle, economizing with a single flagellar system that accommodates moderate swimming and swarming; whereas other bacteria, opting for maximal speed and performance, choose to maintain dual capacities for locomotion. Of course, there are variations to such different forms of resource management: some bacteria (such as

McCarter

R. centenum, Azospirillum, and Aeromonas species) appear to maintain dual flagellar systems that can share parts or regulation. The implications for competitiveness and survival in different environments will be very interesting to explore.

Acknowledgements Work on the V. parahaemolyticus lateral flagellar system was supported by the National Science Foundation (MCB0077327) and the polar flagellar system was supported by National Institutes of Health (USPH grant GM43196).

References Aizawa, S.I. 1996. Flagellar assembly in Salmonella typhimurium. Mol. Microbiol. 19:1–5. Aldridge, P., Hughes, K.T. 2002. Regulation of flagellar assembly. Curr. Opin. Microbiol. 5:160– 165. Alexandre, G., Rohr, R., Bally, R. 1999. A phase variant of Azospirillum lipoferum lacks a polar flagellum and constitutively expresses mechanosensing lateral flagella. Appl. Environ. Microbiol. 65:4701–4704. Allen, R.D., Baumann, P. 1971. Structure and arrangement of flagella in species of the genus Beneckea and Photobacterium fischeri. J. Bacteriol. 107:295–302. Altarriba, M., Merino, S., Gavin, R., Canals, R., Rabaan, A., Shaw, J.G., Tomas, J.M. 2003. A polar flagella operon (flg) of Aeromonas hydrophila contains genes required for lateral flagella expression. Microb. Pathog. 34:249–259. Anantharaman, V., Aravind, L. 2000. Cache – a signaling domain common to animal Ca(2+)channel subunits and a class of prokaryotic chemotaxis receptors. Trends Biochem. Sci. 25: 535–537. Arora, S.W., Ritchings, B.W., Almira, E.C., Lory, S., Ramphal, R. 1997. A transcriptional activator FleQ, regulates mucin adhesion and flagellar gene expression in Pseudomonas aeruginosa in a cascade manner. J. Bacteriol. 179:5574– 5581. Atsumi, T., McCarter, L., Imae, Y. 1992. Polar and lateral flagellar motors of marine Vibrio are driven by different ion-motive forces. Nature 355:182–184. Baumann, P., Baumann, L. 1981. The marine Gram-negative eubacteria: Genera Photobacterium, Beneckea, Alteromonas, and Alcaligenes, In M. Starr, H. Stolp, H. Truper, A. Balows, and H. Schlegel (ed.), The Prokaryotes, Springer-Verlag, New York, pp. 1302–1331. Belas, R., Simon, M., Silverman, M. 1986. Regulation of lateral flagella gene transcription in Vibrio parahaemolyticus. J. Bacteriol. 167: 210–218. Berg, H.C. 2003. The rotary motor of bacterial flagella. Annu. Rev. Biochem. 12:19–54. Blair, D.F. 1995. How bacteria sense and swim. Annu. Rev. Microbiol. 49:489–522. Boles, B.R., McCarter, L.L. 2000. Insertional inactivation of genes encoding components of the sodium-type flagellar motor and switch of Vibrio parahaemolyticus. J. Bacteriol. 182:1035– 1045.

Polar and Lateral Flagella

Boles, B.R., McCarter, L.L. 2002. Vibrio parahaemolyticus scrABC, a novel operon affecting swarming and capsular polysaccharide regulation. J. Bacteriol. 184:5946–5954. Carpenter, P.B., Hanlon, D.W., Ordal, G.W. 1992. flhF, a Bacillus subtilis flagellar gene that encodes a putative GTP-binding protein. Mol. Microbiol. 6:2705–2713. Charon, N.W., Goldstein, S.F. 2002. Genetics of motility and chemotaxis of a fascinating group of bacteria: The spirochetes. Annu. Rev. Genet. 36:47–73. Cohen-Krausz, S., Trachtenberg, S. 2003. The structure of the helically perturbed flagellar filament of Pseudomonas rhodos: Implications for the absence of the outer domain in other complex flagellins and for the flexibility of the radial spokes. Mol. Microbiol. 48:1305–1316. Cserzo, M., Wallin, E., Simon, I., von Heijne, G., Elofsson, A. 1997. Prediction of transmembrane alpha-helices in procaryotic membrane proteins: The Dense Alignment Surface method. Prot. Eng. 10:673–676. Dasgupta, N., Arora, S.K., Ramphal, R. 2000. fleN, a gene that regulates flagellar number in Pseudomonas aeruginosa. J. Bacteriol. 182:357– 364. Dasgupta, N., Ramphal, R. 2001. Interaction of the antiactivator FleN with the transcriptional activator FleQ regulates flagellar number in Pseudomonas aeruginosa. J. Bacteriol. 183:6636– 6644. Doetsch, R.N., Sjoblad, R.D. 1980. Flagellar structure and function in eubacteria. Annu. Rev. Microbiol. 34:69–108. Falke, J.J., Kim, S.H. 2000. Structure of a conserved receptor domain that regulates kinase activity: The cytoplasmic domain of bacterial taxis receptors. Curr. Opin. Struct. Biol. 10: 462–469. Fraser, G.M., Hughes, C. 1999. Swarming motility. Curr. Opin. Microbiol. 2:630–635. Fuerst, J.A. 1980. Bacterial sheathed flagella and the rotary motor model for the mechanism of bacterial motility. J. Theor. Biol. 4:761–774. Fuerst, J.A., Perry, J.W. 1988. Demonstration of lipopolysaccharide on sheathed flagella of Vibrio cholerae O:1 by protein A-gold immunoelectron microscopy. J. Bacteriol. 170:1488– 1494. Furuno, M., Sato, K., Kawagishi, I., Homma, M. 2000. Characterization of a flagellar sheath component, PF60, and its structural gene in marine Vibrio. J. Biochem. (Tokyo). 127:29– 36.

Gavin, R., Merino, S., Altarriba, M., Canals, R., Shaw, J.G., Tomas, J.M. 2003. Lateral flagella are required for increased cell adherence, invasion and biofilm formation by Aeromonas spp. FEMS Microbiol. Lett. 224:77–83. Gavin, R., Rabaan, A.A., Merino, S., Tomas, J.M., Gryllos, I., Shaw, J.G. 2002. Lateral flagella of Aeromonas species are essential for epithelial cell adherence and biofilm formation. Mol. Microbiol. 43:383–397. Geis, G., Suerbaum, S., Forsthoff, B., Leying, H., Opferkuch, W. 1993. Ultrastructure and biochemical studies of the flagellar sheath of Helicobacter pylori. J. Med. Microbiol. 38:371– 377. Gestwicki, J.E., Lamanna, A.C., Harshey, R.M., McCarter, L.L., Kiessling, L.L., Adler, J. 2000. Evolutionary conservation of methyl-accepting chemotaxis protein location in Bacteria and Archaea. J. Bacteriol. 182:6499–6502. Götz, R., Limmer, N., K. Ober, K., Schmitt, R. 1982. Motility and chemotaxis in two strains of Rhizobium with complex flagella. J. Gen. Microbiol. 128:789–798. Hall, P.G., Krieg, N.R. 1984a. Application of the indirect immunoperoxidase staining technique to the flagella of Azospirillum brasilense. Appl. Environ. Microbiol. 47:433–435. Hall, P.G., Krieg, N.R. 1984b. Swarming of Azospirillum brasilense on solid media. Can. J. Microbiol. 29:1592–1594. Harshey, R.M. 1994. Bees aren’t the only ones: Swarming in Gram-negative bacteria. Mol. Microbiol. 13:389–394. Harshey, R.M., Matsuyama, T. 1994. Dimorphic transition in Escherichia coli and Salmonella typhimurium: Surface-induced differentiation into hyperflagellate swarmer cells. Proc. Natl. Acad. Sci. USA 91:8631–8635. Harwood, C.S., Fosnaugh, K., Dispensa, M. 1989. Flagellation of Pseudomonas putida and analysis of its motile behavior. J. Bacteriol. 171: 4063–4066. Hernandez, F., Monge-Najera, J. 1994. Ultrastructure of the bacteria Campylobacter and Helicobacter: Implications for the phylogeny of mammal gastric bacteria. Rev. Biol. Trop. 42:85– 92. Hranitzky, K.W., Mulholland, A., Larson, A.D., Eubanks, E.R., Hart, L.T. 1980. Characterization of a flagellar sheath protein of Vibrio cholerae. Infect. Immun. 27:597–603. Imae, Y., Atsumi, T. 1989. Na+-driven bacterial flagellar motors. J. Bioenerg. Biomemb. 21: 705–716.

J Mol Microbiol Biotechnol 2004;7:18–29

27

Jiang, Z.Y., Rushing, B.G., Bai, Y., Gest, H., Bauer, C.E. 1998. Isolation of Rhodospirillum centenum mutants defective in phototactic colony motility by transposon mutagenesis. J. Bacteriol. 180:1248–1255. Joseph, S.W., Colwell, R.R., Kaper, J.B. 1982. Vibrio parahaemolyticus and related halophilic Vibrios. Crit. Rev. Microbiol. 10:77–124. Joys, T. 1988. The flagellar filament protein. Can J. Microbiol. 34:452–458. Kaneko, T., Colwell, R.R. 1975. Adsorption of V. parahaemolyticus onto chitin and copepods. Appl. Microbiol. 29:269–274. Kawagishi, I., Imagawa, M., Imae, Y., McCarter, L., Homma, M. 1996. The sodium-driven polar flagellar motor of marine Vibrio as the mechanosensor that regulates lateral flagellar expression. Mol. Microbiol. 20:693–699. Khammas, K.M., Ageron, E., Grimont, P.A., Kaiser, P. 1989. Azospirillum irakense sp. nov., a nitrogen-fixing bacterium associated with rice roots and rhizosphere soil. Res. Microbiol. 140: 679–693. Kim, Y.K., McCarter, L.L. 2000. Analysis of the polar flagellar gene system of Vibrio parahaemolyticus. J. Bacteriol. 182:3693–3704. Kirov, S.M. 2003. Bacteria that express lateral flagella enable dissection of the multifunctional roles of flagella in pathogenesis. FEMS Microbiol. Lett. 224:151–159. Kirov, S.M., Tassell, B.C., Semmler, A.B., O’Donovan, L.A., Rabaan, A.A., Shaw, J.G. 2002. Lateral flagella and swarming motility in Aeromonas species. J. Bacteriol. 184:547–555. Klose, K.E., Novik, V., Mekalanos, J.J. 1998. Identification of multiple sigma54-dependent transcriptional activators in Vibrio cholerae. J. Bacteriol. 180:5256–5259. Kogure, K., Ikemoto, E., Morisaki, H. 1998. Attachment of Vibrio alginolyticus to glass surfaces is dependent on swimming speed. J. Bacteriol. 180:932–937. Kudo, S., Magariyama, Y., Aizawa, S. 1990. Abrupt changes in flagellar rotation observed by laser dark-field microscopy. Nature 346: 677–680. Li, C., Motaleb, A., Sal, M., Goldstein, S.F., Charon, N.W. 2000. Spirochete periplasmic flagella and motility. J. Mol. Microbiol. Biotechnol. 2:345–354. Lowe, G., Meister, M., Berg, H.C. 1987. Rapid rotation of flagellar bundles in swimming bacteria. Nature 325:637–640. McCarter, L.L. 1995. Genetic and molecular characterization of the polar flagellum of Vibrio parahaemolyticus. J. Bacteriol. 177:1595–1609. McCarter, L.L. 1998. OpaR, a homolog of Vibrio harveyi LuxR, controls opacity of Vibrio parahaemolyticus. J. Bacteriol. 180:3166–3173. McCarter, L. 1999. The multiple identities of Vibrio parahaemolyticus. J. Mol. Microbiol. Biotechnol. 1:51–57. McCarter, L.L. 2001. Polar flagellar motility of the Vibrionaceae. Microbiol. Mol. Biol. Rev. 65: 445–462. McCarter, L., Hilmen, M., Silverman, M. 1988. Flagellar dynamometer controls swarmer cell differentiation of V. parahaemolyticus. Cell. 54:345–351.

28

McCarter, L., Silverman, M. 1989. Iron regulation of swarmer cell differentiation of Vibrio parahaemolyticus. J. Bacteriol. 171:731–736. McCarter, L., Silverman, M. 1990. Surface-induced swarmer cell differentiation of Vibrio parahaemolyticus. Mol. Microbiol. 4:1057– 1062. McCarter, L.L., Wright, M.E. 1993. Identification of genes encoding components of the swarmer cell flagellar motor and propeller and a sigma factor controlling differentiation of Vibrio parahaemolyticus. J. Bacteriol. 175:3361–3371. McClain, J., Rollo, D.R., Rushing, B.G., Bauer, C.E. 2002. Rhodospirillum centenum utilizes separate motor and switch components to control lateral and polar flagellum rotation. J. Bacteriol. 184:2429–2438. McCoy, E.C., Doyle, D., Wiltberger, H., Burda, K., Wintberger, A.J. 1975. Flagellar ultrastructure and flagella-associated antigens of Campylobacter fetus. J. Bacteriol. 122:307–315. Macnab, R.M. 1976. Examination of bacterial flagellation by dark-field microscopy. J. Clin. Microbiol. :258–265. Macnab, R.M. 1986. Proton-driven bacterial flagellar motor. Methods Enzymol. 125:563–579. Macnab, R.M. 1996. Flagella, motility, In R.C. F.C. Neidhardt, C.A. Gross, J.L. Ingraham, E.C.C. Lin, K.B. Low, Jr., B. Magasanik, W. Reznikoff, M. Riley, M. Schaechter,, H. E. Umbarger (ed.), Escherichia coli and Salmonella cellular and molecular biology, 2nd ed, ASM Press, Washington, D.C., pp. 123–146. Macnab, R.M. 1999. The bacterial flagellum: Reversible rotary propellor and type III export apparatus. J. Bacteriol. 181:7149–7153. Macnab, R.M. 2003. How bacteria assemble flagella. Annu. Rev. Microbiol. 57:77–100. Macnab, R.M., Aizawa, S.-I. 1984. Bacterial motility and the bacterial flagellar motor. Annu. Rev. Biophys. Bioeng. 13:51–83. Magariyama, Y., Sugiyama, S., Kudo, S. 2001. Bacterial swimming speed and rotation rate of bundled flagella. FEMS Microbiol. Lett. 199: 125–129. Magariyama, Y., Sugiyama, S., Muramoto, K., Kawagishi, I., Imae, Y., Kudo, S. 1995. Simultaneous measurement of bacterial flagellar rotation rate and swimming speed. Biophys. J. 69: 2154–2162. Magariyama, Y., Sugiyama, S., Muramoto, K., Maekawa, Y., Kawagishi, I., Imae, Y., Kudo, S. 1994. Very fast flagellar rotation. Nature 371: 752. Makino, K., Oshima, K., Kurokawa, K., Yokoyama, K., Uda, T., Tagomori, K., Iijima, Y., Najima, M., Nakano, M., Yamashita, A., Kubota, Y., Kimura, S., Yasunaga, T., Honda, T., Shinagawa, H., Hattori, M., Iida, T. 2003. Genome sequence of Vibrio parahaemolyticus: A pathogenic mechanism distinct from that of V. cholerae. Lancet 361:743–749. Manson, M.D., Tedesco, P., Berg, H.C., Harold, F.M., Van der Drift, C. 1977. A proton motive force drives bacterial flagella. Proc. Natl. Acad. Sci. USA 74:3060–3064.

J Mol Microbiol Biotechnol 2004;7:18–29

Merino, S., Rubires, X., Aguilar, A., Tomas, J.M. 1997. The role of flagella and motility in the adherence and invasion to fish cell lines by Aeromonas hydrophila serogroup O: 34 strains. FEMS Microbiol. Lett. 151:213–217. Moens, S., Michiels, K., Keijers, V., Van Leuven, F., Vanderleyden, J. 1995. Cloning, sequencing, and phenotypic analysis of laf1, encoding the flagellin of the lateral flagella of Azospirillum brasilense Sp7. J. Bacteriol. 177:5419– 5426. Moens, S., Schloter, M., Vanderleyden, J. 1996. Expression of the structural gene, laf1, encoding the flagellin of the lateral flagella in Azospirillum brasilense Sp7. J. Bacteriol. 178:5017– 5019. Morisaki, H., Nagai, S., Ohshima, H., Ikemoto, E., Kogure, K. 1999. The effect of motility and cell-surface polymers on bacterial attachment. Microbiology. 145:2797–2802. Namba, K., Vonderviszt, F. 1997. Molecular architecture of bacterial flagellum. Q. Rev. Biophys. 30:1–65. Pandza, S., Baetens, M., Park, C.H., Au, T., Keyhan, M., Matin, A. 2000. The G-protein FlhF as a role in polar flagellar placement and general stress response induction in Pseudomonas putida. Mol. Microbiol. 36:414–423. Prouty, M., Correa, N., Klose, K. 2001. The novel s54- and s28-dependent flagellar gene transcription hierarchy of Vibrio cholerae. Mol. Microbiol. 39:1595–1609. Rabaan, A.A., Gryllos, I., Tomas, J.M., Shaw, J.G. 2001. Motility and the polar flagellum are required for Aeromonas caviae adherence to HEp-2 cells. Infect. Immun. 69:4257–4267. Ragatz, L., Jiang, Z.Y., Bauer, C.E., Gest, H. 1995. Macroscopic phototactic behavior of the purple photosynthetic bacterium Rhodospirillum centenum. Arch. Microbiol. 163:1–6. Ragatz, L., Jiang, S.-Y., Gest, H., Bauer, C. 1994. Phototactic purple bacteria. Nature 370:104. Ritchings, B.W., Almira, E.C., Lory, S., Ramphal, R. 1995. Cloning and phenotypic characterization of fleS and fleR, new response regulators of Pseudomonas aeruginosa which regulate motility and adhesion to mucin. Infect. Immun. 63: 4868–4876. Rosario, M.M.L., Fredrick, K.L., Ordal, G., Helmann, J. 1994. Chemotaxis in Bacillus subtilis requires either of two functionally redundant CheW homologs. J. Bacteriol. 176:2736–2739. Sar, N., McCarter, L., Simon, M., Silverman, M. 1990. Chemotactic control of the two flagellar systems of Vibrio parahaemolyticus. J. Bacteriol. 172:334–341. Shimada, T., Sakazaki, R., Suzuki, K. 1985. Peritrichous flagella in mesophilic strains of Aeromonas. Jpn. J. Med. Sci. Biol. 38:141–145. Shinoda, S., Okamoto, K. 1977. Formation and function of Vibrio parahaemolyticus lateral flagella. J. Bacteriol. 129:1266–1271. Sjoblad, R.D., Emala, C.W., Doetsch, R.N. 1983. Bacterial sheaths: Structures in search of function. Cell Motility. 3:93–103. Stewart, B.J., McCarter, L.L. 2003. Lateral flagellar gene system of Vibrio parahaemolyticus. J. Bacteriol. 185:4508–4518.

McCarter

Tarrand, J.J., Krieg, N.R., Dobereiner, J. 1978. A taxonomic study of the Spirillum lipoferum group, with descriptions of a new genus, Azospirillum gen. nov. and two species, Azospirillum lipoferum (Beijerinck) comb. nov. and Azospirillum brasilense sp. nov. Can. J. Microbiol. 24:967–980. Taylor, B.L., Zhulin, I.B. 1999. PAS domains: Internal sensors of oxygen, redox potential, and light. Microbiol. Mol. Biol. Rev. 63:479–506. Thomashow, L.S., Rittenberg, S.C. 1985a. Isolation and composition of sheathed flagella from Bdellovibrio bacteriovorus 109J. J. Bacteriol. 163:1047–1054. Thomashow, L.S., Rittenberg, S.C. 1985b. Waveform analysis and structure of flagella and basal complexes from Bdellovibrio bacteriovorus 109J. J. Bacteriol. 163:1038–1046.

Polar and Lateral Flagella

Trachtenberg, S., DeRosier, D.J. 1987. Threedimensional structure of the frozen-hydrated flagellar filament. The left-handed filament of Salmonella typhimurium. J. Mol. Biol. 195: 581–601. Trachtenberg, S., DeRosier, D.J. 1988. Threedimensional reconstruction of the flagellar filament of Caulobacter crescentus. A flagellin lacking the outer domain and its amino acid sequence lacking an internal segment. J. Mol. Biol. 202:787–808. Trachtenberg, S., DeRosier, D.J., Macnab, R.M. 1987. Three-dimensional structure of the complex flagellar filament of Rhizobium lupini and its relation to the structure of the plain filament. J. Mol. Biol. 195:603–620. Trachtenberg, S., DeRosier, D.J., Zemlin, F., Beckmann, E. 1998. Non-helical perturbations of the flagellar filament: Salmonella typhimurium SJW117 at 9.6 A resolution. J. Mol. Biol. 276: 759–773.

Wilson, D.R., Beveridge, T.J. 1993. Bacterial flagellar filaments and their component flagellins. Can. J. Microbiol. 39:451–472. Yonekura, K., Maki-Yonekura, S., Namba, K. 2002. Growth mechanism of the bacterial flagellar filament. Res Microbiol. 153:191–197. Yonekura, K., Maki-Yonekura, S., Namba, K. 2003. Complete atomic model of the bacterial flagellar filament by electron cryomicroscopy. Nature. 424:643–650. Yorimitsu, T., Homma, M. 2001. Na+-driven flagellar motor of Vibrio. Biochem. Biophys. Acta. 1505:82–93. Zhulin, I.B., Nikolskaya, A.N., Galperin, M.Y. 2003. Common extracellular sensory domains in transmembrane receptors for diverse signal transduction pathways in bacteria and archaea. J. Bacteriol. 185:285–294.

J Mol Microbiol Biotechnol 2004;7:18–29

29

J Mol Microbiol Biotechnol 2004;7:30–40 DOI: 10.1159/000077867

The Periplasmic Flagellum of Spirochetes Ronald J. Limberger Wadsworth Center-Axelrod Institute, New York State Department of Health, Albany, N.Y., USA

Key Words Motility W Flagellum W Spirochete W Structure W Periplasm

Abstract Although spirochete periplasmic flagella have many features similar to typical bacterial flagella, they are unique in their structure and internal periplasmic location. This location provides advantages for pathogenic spirochetes to enter and to adapt in the appropriate host, and to penetrate through matrices that inhibit the motility of most other bacteria. These flagella are complex, and they dynamically interact with the spirochete cell cylinder in novel ways. Electron microscopy, tomography and three-dimensional reconstructions have provided new insights into flagellar structure and its relationship to the spirochetal cell cylinder. Recent advances in genetic methods have begun to shed light on the composition of the spirochete flagellum, and on the regulation of its synthesis. Because spirochetes have a high length to width ratio, their cells provide an opportunity to study two important features. These include the polarity or distribution of flagellar synthesis as well as the mechanisms required for coordination of the movement of the cell ends, to enable it to move in the forward or reverse direction. Copyright © 2004 S. Karger AG, Basel

ABC

© 2004 S. Karger AG, Basel

Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/mmb

Introduction

Spirochetes are a diverse group of pathogenic and nonpathogenic bacteria that possess a unique helical or wavelike morphology [Holt et al., 1994]. Notable diseases caused by spirochetes include syphilis (Treponema pallidum), Lyme disease (Borrelia burgdorferi), swine dysentery (Brachyspira hyodysenteriae), and leptospirosis (Leptospira interrogans). Certain oral spirochetes including Treponema denticola have been associated with periodontal disease [Moter et al., 1998; Paster et al., 2001; Qiu et al., 1994; Simonson et al., 1988]. In addition to their distinctive shape, a unique feature that differentiates spirochetes from other bacteria is the presence of flagellar filaments that lie within the periplasm and therefore do not contact the external environment. Evidence indicates that the rotation of the spirochete flagellar filaments results in specific movements of the cell body, which in turn enable the locomotion of the cell [Berg, 1976, Canale-Parola, 1984, Charon et al., 1992a, b, Li et al., 2000b]. This periplasmic location of the flagellum imparts spirochetes the ability to propel themselves through viscous media that would inhibit the rotation of external flagellar filaments [Greenberg and Canale-Parola, 1977; Kimsey and Spielman, 1990; Klitorinos et al., 1993; Ruby and Charon, 1998]. Depending on the spirochete species, this characteristic may enable movement in low- or high-viscosity matrices (mud, periodontal scrapings, water), as well as penetration through tissues [Lux et al., 2001; Sadziene et al., 1991; Thomas et al., 1988].

Ronald J. Limberger Wadsworth Center-Axelrod Institute, New York State Department of Health PO Box 22002 Albany, NY 12201 (USA) Tel. +1 518 474 4177, Fax +1 518 486 7971, E-Mail [email protected]

Fig. 1. Diagrammatic representation of a T. denticola cell, showing the periplasmic location of the flagellar filaments.

Also shown is an enlarged region as well as a cross-sectional view. Although the size of T. denticola varies, a typical cell is 0.2 Ìm in diameter by 11 Ìm in length. PF = Periplasmic flagella; CF = cytoplasmic filaments; BB = basal body; IM = inner membrane; OM = outer membrane.

Because of the great diversity in spirochete morphology, there is no prototypical spirochete flagellum. Each spirochete species displays a unique structure, composition and number of flagella. For example, the number of flagella per cell end can range from one in the Leptospiraceae to hundreds in the genus Cristispira. The periplasmic location of the flagellum is the primary commonality in flagellar structure among the spirochetes. For the purposes of this review, the term flagellum refers to the entire flagellar structure, including the basal body, hook, and filament. The spirochete flagellum (primarily referring to the filamentous portion) has been previously designated as the axial filament, periplasmic fibril, periplasmic flagellum, endoflagellum and many other terms. Because the spirochete flagellum is responsible for motility, it is properly designated as a flagellum. However, I will often utilize the term periplasmic flagellar filaments to emphasize that the filaments are found in the periplasm in contrast to other bacteria that have external filaments. The taxonomy of the spirochetes is evolving as more is learned about the genetics of these organisms. Notably, Leptospira illini is now designated as Leptonema illini. The recently renamed Brachyspira hyodysenteriae was formerly assigned to the genera Serpulina and Trepone-

ma. These organisms will be referred to by their most recent taxonomic designation in the present review. For additional information on spirochete motility, the reader is referred to other reviews [Canale-Parola, 1984; Charon et al., 1992b; Charon and Goldstein, 2002; Goldstein and Charon, 1988; Li et al., 2000b; Lux et al., 2000].

The Periplasmic Flagellum of Spirochetes

J Mol Microbiol Biotechnol 2004;7:30–40

Structure of the Spirochete Flagellum

The flagellar structure of many spirochete species has been analyzed by transmission electron microscopy. Spirochete flagella are inserted subterminally into the cell cylinder near each pole of the cell, but are localized within the periplasmic space. The filaments wrap around the cell cylinder or are along the cell axis (fig. 1) [Goldstein et al., 1996; Holt et al., 1994; Holt, 1978]. The distal end of the filament does not appear to be anchored to the cell wall or cell membrane. The number of flagellar filaments varies depending on the species. The length of the flagellar filament varies as well. Some species have long filaments that overlap with the filaments that originate at the other end of the cell (B. burgdorferi), whereas others have short filaments that terminate prior to reaching the cell center and do not lengthwise overlap (Treponema phagedenis).

31

Fig. 2. Diagrammatic representation of the flagellar basal body region of spirochetes. Although the genes encoding

these structures have been noted in many spirochetes, the actual location of the encoded gene product has not always been directly demonstrated in spirochetes.

Fig. 3. Electron micrograph of a negatively stained flagellar basal body, including the rod and hook from T. phagedenis. The hook structure has an average length of 69 nm and a diameter of 23 nm.

32

J Mol Microbiol Biotechnol 2004;7:30–40

The basic overall structure of the spirochetal flagellum is similar to that of flagella in other bacteria such as Salmonella, in that it consists of a basal body, rod, hook, and filament (fig. 2, 3) [Holt, 1978; Macnab, 2003, 1992]. However, the complexity of the flagellar filament is somewhat greater than that of the enteric bacterial flagellum. The spirochete flagellar filament is often composed of multiple polypeptides that form both core and sheath structures. The diameter of the spirochete flagellar filament is usually larger than that of other bacterial flagella due to the usual presence of an outer protein sheath (23 nm diameter in T. denticola vs. 20 nm in Salmonella). In B. hyodysenteriae the flagellar filament is 24–25 nm in diameter whereas mutants lacking the sheath are 19.6 nm in diameter [Li et al., 2000a]. The purpose and function of this sheath are unclear. Multiple protein species that comprise the core and sheath of the flagellar filament are typical, but not universal, in spirochetes [Norris et al., 1988]. Additional details on the composition of the spirochete periplasmic flagellum are included later in this review.

Limberger

Table 1. Open reading frames of

Treponema pallidum associated with flagellar structure and function organized by sequential location on the chromosome

Gene product

ORF number2

Function3

Promoter4

1

FliG-1

026

Unknown

Unknown

2

FlaA1

249

Filament sheath

Sigma 70

3

FlgB FlgC FliE FliF FliG FliH FliI FliJ

396 397 398 399 400 401 402 403

Rod Rod MS ring/rod export MS ring Rotor/switch Neg. regulator of FliI ATPase Chaperone

Sigma 28

4

FlgL FlgK

659 660

Hook/filament junction 3 Hook/filament junction 1

Unknown

5

FlaA2

664

Unknown

Unknown

6

FlhF

713

GTP-binding

Unknown

7

FlhA FlhB FliR FliQ FliP FliN/Y FliM FliL MotB MotA FlbD/orf4 FlgE FlgD FliK/Tap1

714 715 716 717 718 720 721 722 724 725 726 727 728 729

Export component Export component Export component Export component Export component C-ring;rotor/switch C-ring;rotor/switch Assoc. with basal body Stator Stator Unknown Hook Hook capping Hook length control

Sigma 28

8

FlaB2

792

Filament core

Sigma 28

9

FlaB1 FlaB3 FliD

868 870 872

Filament core Filament core HAP2 filament cap

Sigma 28

10

FliS/FlaJ

943

Chaperone flagellin

Unknown

11

FlgG-1 FlgG-2

960 961

Distal rod Distal rod

Unknown

Group number1

f

d

f

1

Group number refers to the cluster or operon of genes in each region. ORF number is from the T. pallidum genome sequencing project [Fraser et al., 1998]. 3 Function in Salmonella as designated by [Macnab, 2003] or as determined in spirochetes. 4 Promoter assignment is based either on experimental data or from sequence similarity. Arrow indicates the direction of transcription. 2

The Periplasmic Flagellum of Spirochetes

J Mol Microbiol Biotechnol 2004;7:30–40

33

The structure of periplasmic flagellar filaments isolated from numerous spirochetal species has been analyzed by electron microscopy. The isolated flagellar filaments retain their typically helical shape in vitro. Spirochete flagellar filaments have varying wavelengths and pitches depending on the species examined, and are usually left-handed. One notable exception is the Leptospiraceae which have coiled flagellar filaments when these are isolated in vitro [Birch-Andersen et al., 1973; Bromley and Charon, 1979]. Nonmotile mutants of L. illini have been described as having noncoiled filaments. The spirochete hook structure likely plays the same role as do hooks in other bacteria, and appears by electron microscopy to be similar in structure to other bacterial hooks. The average length of a T. phagedenis hook is 69 nm, with a diameter of 23 nm, dimensions which are within the range found for other bacteria [Limberger et al., 1994]. A slightly smaller hook has been described for Spirochaeta aurantia [Brahamsha and Greenberg, 1988]. The rod structure has not been extensively analyzed in spirochetes, but it appears by electron microscopy to be similar in structure to that of other bacteria and likely has the same function. The basal body of the spirochetes has been difficult to elucidate by electron microscopy. In the few studies that have examined this structure, there have been notable differences from other bacteria. In some Treponema species there appears to be a basal disk (also known as basal plate or terminal knob) consisting of, or surrounded by, one or possibly two rings [Hovind-Hougen and Birch-Andersen, 1971; Hovind-Hougen, 1972, 1974]. In Leptospira, multiple rings (4–5) are found in the basal body regions [Hovind-Hougen, 1976; Nauman et al., 1969]. In S. aurantia, the basal body appears to consist of two closely associated rings that assume several different configurations upon electron microscopic analysis [Brahamsha and Greenberg, 1988]. Although the actual visualization of the components of the spirochete basal body region is often unclear, whole genome sequencing of T. pallidum, T. denticola, and B. burgdorferi has enabled the prediction of basal body structures [Fraser et al., 1997, 1998] (www.tigr.org, www.hgsc.bcm.tmc.edu). Many of the known basal bodyassociated structures are presumed to be present in T. pallidum, on the basis of the identification of genes encoding polypeptides related to FliG, FliM, FliN (switch) FlgB, FlgC, FlgF FlgG (rod), MotA, MotB (motor), FliF (MSring), FlgD, FlgE, FliK (hook assembly) and FlgK, FlgL (hook-filament junction) (see table 1). In T. pallidum and T. denticola, there is no structural or genetic evidence for

34

J Mol Microbiol Biotechnol 2004;7:30–40

an L-ring (FlgH): such a structure is not needed since the spirochete flagellum does not penetrate the outer cell membrane as it does in Salmonella. Interestingly, genetic evidence for a P-ring (FlgI) is lacking in T. pallidum. Nevertheless, it is possible (though unlikely based on the absence of structural evidence), that T. pallidum does possess a P-ring that has a unique amino acid sequence, not homologous to that of other known P-ring proteins. Paradoxically, the Leptospira genome reveals sequences homologous to both FlgH and FlgI [Ren et al., 2003]. Further work is needed to elucidate the functions of these recently identified genes. One interesting aspect of this structural characterization is the evidence demonstrating that flagellar filaments of T. phagedenis, T. denticola, and B. burgdorferi can protrude through the outer membrane under certain conditions [Charon et al., 1992a; Goldstein et al., 1994]. Although these protruding filaments may appear to be external flagellar filaments, they are in fact still covered, in part, by a membranous material. These protruding filaments show evidence of rotating externally, and the cells are still viable. The biological basis for the protruding filaments is unclear.

Three-Dimensional Analysis of Flagellar and Cytoplasmic Filament Structures

The use of high-voltage electron microscopy together with three-dimensional reconstruction has elucidated the spatial location and structure of the periplasmic flagellar filament in L. illini and in B. burgdorferi [Goldstein et al., 1996]. In Leptonema, the flagellar filament lies in the central channel, along the axis of a right-handed cell cylinder. In B. burgdorferi, the flagellar filaments are found in a ridge that spans the length of the cell. The flagellar filaments wrap around the cell cylinder in a right-handed sense, but the filaments are left-handed in space. The wavelength of the cell body and the helix pitch of the flagellar filament are identical. In addition to the periplasmic flagella filaments, Treponema spp. possess cytoplasmic filaments, encoded by cfpA [Eipert and Black, 1979; Hovind-Hougen and BirchAndersen, 1971; Izard et al., 1999; You et al., 1996]. Such cytoplasmic filaments have not been detected, either genetically or by structural analysis, in Borrelia but they have been noted by electron microscopy in Leptonema [Hovind-Hougen, 1976]. These filaments form a ribbon within the treponemal cell cytoplasm and lie just underneath the periplasmic flagella [Eipert and Black, 1979;

Limberger

The spirochete flagellar filament is typically a complex association of multiple protein species. Flagellar filaments of several treponemal species have been analyzed by oneand two-dimensional SDS-PAGE and are composed of 1– 3 core proteins and 1–2 sheath proteins [Norris et al., 1988]. In T. denticola, the core of the filament is composed of three FlaB core polypeptides (one of 34 kD, and two of 35 kD), encoded by separate genes (flaB1, flaB2, flaB3), that have a high amino acid sequence identity with one another [Cockayne et al., 1989; Li et al., 2000a; Ruby et al., 1997]. These core polypeptides, which have moderate amino acid identity with flagellin of other bacteria, are transcribed from promoters that have a sequence similar to that of the sigma 28 promoter of E. coli [www.tigr.org, www.hgsc.bcm.tmc.edu, R. Limberger unpubl.]. flaB1 and flaB3 are adjacent to one another in the T. denticola genome and appear to be part of the same operon, whereas flaB2 is located elsewhere on the chro-

mosome and is transcribed as a single gene. T. denticola also possesses a sheath that is composed of a 38-kD FlaA polypeptide [Cockayne et al., 1989; Ruby et al., 1997]. The FlaA polypeptide is transcribed from a sigma 70-like promoter. This protein composition and gene arrangement are very similar to what has been described for the structure and genetics of T. pallidum [Champion et al., 1990; Fraser et al., 1998; Isaacs et al., 1989]. Immunolocalization, immuneprecipitation, and/or gene knockout experiments in Treponema, Leptospira, Brachyspira, and Spirochaeta have shown that the FlaA protein comprises the sheath and possesses external epitopes available for binding to antibodies [Ge et al., 1998; Li et al., 2000a; Limberger and Charon, 1986b; Radolf et al., 1986; Trueba et al., 1992]. The sheath polypeptide FlaA has a signal sequence and is likely exported through the general secretory system into the periplasm, where it is added to the flagellar filament via an unknown mechanism [Charon et al., 1992b]. The treponemal flagellar filament, hook, and rod are likely exported via a type III secretion system as in enteric bacteria. Whole-genome sequencing has shown that some spirochetes, such as T. pallidum, possess two genes encoding modestly related FlaA polypeptides; however, the expression and function of the second gene copy are unknown [Fraser et al., 1998]. What is the structural arrangement of the flagellar core polypeptides on those spirochetes that possess periplasmic flagella composed of multiple core polypeptides? One could envision a cell having each individual flagellar filament composed of only one of the three core polypeptides. Alternatively, each filament could be composed of all three core polypeptides, or various combinations of each. If each flagellar filament is comprised of multiple polypeptides, the spatial arrangement of the core polypeptides would need to be resolved: are they arranged in sequential order from the proximal to the distal ends or are they randomly associated in equal molar amounts? Analysis of B. hyodysenteriae revealed that each flagellin has flagellar filament proteins present in the following stoichiometry: FlaA (1): FlaB1 (0.42): FlaB2 (0.22): FlaB3 (0.45) [Li et al., 2000a]. Analysis of double mutants suggests that the core filament polypeptides are distributed along the entire length of the flagellar filament [Charon and Goldstein, 2002]. B. burgdorferi shows a very different composition of its flagellar filament. In B. burgdorferi, only one flagellar filament protein, FlaB, was originally described [Barbour et al., 1986, Coleman and Benach, 1989]. However, modification of the method of isolation resulted in the detection

The Periplasmic Flagellum of Spirochetes

J Mol Microbiol Biotechnol 2004;7:30–40

Hovind-Hougen and Birch-Andersen, 1971]. The cytoplasmic filaments appear to originate near the end of the cell close to the flagellar basal body. These results were recently confirmed through the use of both traditional transmission electron microscopy and three-dimensional reconstruction of tomographic images of T. phagedenis cells [Izard et al., 1999, 2003]. This close structural relationship might suggest that cytoplasmic filaments interact with the periplasmic flagellar filaments to influence cell shape or motility. A knockout mutant was constructed in T. denticola to determine whether a lack of cytoplasmic filaments would affect the structure or function of the periplasmic flagellar filaments [Izard et al., 2001]. The CfpAminus mutant that lacked the cytoplasmic filaments had an unusually long cell cylinder, showed an impairment in DNA segregation, and had altered motility on agar plates. However, the mutant cells were shown by electron microscopy to have normal flagellar structures. Moreover, the rarely observed single cell appeared to have motility functions similar to those of the wild type when observed by dark-field microscopy [Izard et al., 2001]. This analysis was unable to demonstrate any defined structural connection between the cytoplasmic and periplasmic filaments. It appears that these cytoplasmic filaments, despite their close spatial association with periplasmic flagella, are instead involved in a broad range of functions associated with cell division.

Biochemical Composition of the Spirochete Flagellum

35

of both FlaA and FlaB associated with the purified flagellar filaments [Ge et al., 1998]. Unlike other spirochetes that synthesize high levels of FlaA, the FlaA levels of the B. burgdorferi flagellum appear to reach only 10% of the FlaB levels. Although the structural location of FlaA of B. burgdorferi is uncertain, there is some evidence that it is associated with the flagellar filament [Charon and Goldstein, 2002]. The biochemical composition of the hook, rod, and basal body region of spirochete species has not been extensively studied. Enriched hook/basal bodies of S. aurantia consist of multiple polypeptides, in the range of 62–66 kD each, which presumably comprise the hook structure [Brahamsha and Greenberg, 1988]. The hook polypeptide of T. phagedenis has been shown to consist of one major polypeptide, FlgE, that has modest amino acid sequence identity with the hook polypeptides of other bacteria [Limberger et al., 1994]. Interestingly, flagellar hooks of T. phagedenis run as a polypeptide ladder on SDS-PAGE, suggesting a cross-linking among monomers. This ladder is also noted for T. denticola but not for B. burgdorferi for which the hook polypeptide migrates as a single band on SDS-PAGE [Jwang et al., 1995]. In T. denticola, interruption of flgE using an antibiotic resistance cassette rendered the cell nonmotile and deficient in flagellar hook structures [Li et al., 1996, Ruby et al., 1997]. This mutant could be restored to near wild-type motility by complementation with FlgE encoded on a plasmid; this also restored the polypeptide ladder [Chi et al., 2002]. T. denticola FliK, also known as Tap1, assists in controlling the length of the flagellar hook, and was first identified by mutational analysis [Limberger et al., 1999]. T. denticola cells that were deficient in FliK due to insertional inactivation had a polyhook phenotype similar to that of Salmonella FliK mutants [Hirano et al., 1994; Kawagishi et al., 1996; Minamino et al., 1999; Muramoto et al., 1998; Williams et al., 1996]. The T. denticola FliK mutants lacked flagellar filaments but did possess hooks of varying lengths; some were as long as a flagellar filament. Interestingly, the amino acid sequence identity between Salmonella and T. denticola FliK polypeptides is relatively weak and is restricted to a small region of the polypeptide. Moreover, T. pallidum and T. denticola which are two closely related treponemes, share only 21% amino acid sequence identity among their FliK polypeptides, and most of this identity is located in the carboxyterminal one-third of the polypeptide. Thus, the use of genetic methods to inactivate FliK was essential for deciphering the function of this gene.

36

J Mol Microbiol Biotechnol 2004;7:30–40

Because the spirochete flagella are located within the periplasm, they likely have unique polypeptides involved in the interaction of the flagellum with cellular components. Thus, it is probable that hypothetical proteins found in the known genomes of spirochetes, but not found in the genomes of other bacteria, are involved in motility. Moreover, some spirochete genomes possess additional genes encoding polypeptides of unknown function, yet showing relatedness to known motility genes. For example, T. denticola, T. pallidum, and B. burgdorferi all possess a theoretical polypeptide that is modestly related (about 28% amino acid sequence identity) to their respective FliG polypeptides and has been designated as FliG-1 [Fraser et al., 1997, 1998]. Although the role of the FliG-1 polypeptide is currently unclear, we do know that it cannot compensate for the loss of FliG [R. Limberger, L. Slivienski-Gebhardt, J. Izard, unpubl.]. It should be noted that FliG is designated as FliG-2 in the published papers related to the genome sequencing of B. burgdorferi and T. pallidum. Interestingly, the genome of Leptospira possesses four copies of FliG, each of which is related to the others to varying degrees [Ren et al., 2003; R. Limberger, unpubl.]. Spirochete genome sequencing has also revealed open reading frames of unknown function interspersed among genes of known function related to motility. For example, orf4 of T. denticola is a small 65-amino acid polypeptide that is found in the fla operon and consists of genes involved in flagellar structure. Deletion of orf4 renders the T. denticola cell nonmotile, but the actual function of orf4 is unknown [R. Limberger and L. SlivienskiGebhardt, unpubl.]. The development of transposon mutagenesis systems for spirochetes, together with ongoing improvements in directed mutagenesis, will greatly assist in identification of spirochete-specific polypeptides involved in flagellar structure and motility.

Cell Shape and the Flagellar Filament

The shape of the periplasmic flagellar filament has been shown to influence overall cell shape in several spirochetes. For example, the tightly coiled filaments of Leptonema and Leptospira appear to be responsible for imparting a hook-shaped end to the cell during periods of directional swimming activity [Berg et al., 1978, Bromley and Charon, 1979, Nauman et al., 1969]. Leptonema and Leptospira can transform their cell shape from hookshaped ends to spiral ends, and back, for various swimming needs. A nonmotile L. illini mutant obtained through chemical mutagenesis possessed flagellar struc-

Limberger

flexible than others. An open reading frame with amino acid sequence identity to mreB has been identified in T. pallidum. This gene has been shown to be involved in cell shape of some bacteria but the role of MreB, if any, in the maintenance of spiral shape in this spirochete is unclear [Young, 2003].

tures that did not have the coiled structure that is found in the wild-type cell [Bromley and Charon, 1979]. Moreover, the ends of the mutant cell were unable to form the hook shape necessary for Leptospira to swim. Although the nature of the mutation was unknown, revertants were isolated that had regained the ability to swim, and to form hook-shaped ends, and that possessed coiled periplasmic flagella. Direct experimental evidence was obtained by targeted interruption of FlaB of Leptospira biflexa which rendered the cells nonmotile and unable to form the hookshaped ends [Picardeau et al., 2001]. There are several other examples of analysis of mutants which revealed the alteration of spirochete cell shape by the flagellar filaments. In T. phagedenis, the periplasmic flagella cause the ends of the cell to have a bent shape while the central region of the cell is a regular helix [Charon et al., 1991]. Nonmotile, flagellum-deficient mutants of T. phagedenis retain their helical cell shape [Limberger and Charon, 1986a]. However, unlike the wild type, these mutants do not possess the characteristic bent-end shape. Similarly, cells of T. denticola are irregular-shaped, consisting of right-handed helices as well as planar regions. Nonmotile mutants that lack flagellar filaments have a regular helical cell shape. Thus, the irregular cell shape, in part, is due to the flagellar filaments [Ruby et al., 1997]. In the above examples, the flagellar filaments modestly alter cell shape, probably by distorting the cell cylinder to varying degrees. An example in which the periplasmic flagellar filaments strongly impart shape to a spirochetal cell is B. burgdorferi. Nonmotile mutants that lack flagellar filaments possess straight or rod-shaped cell cylinders, as opposed to the wave-shaped cell cylinder of the wild type [Motaleb et al., 2000; Sadziene et al., 1991]. Thus B. burgdorferi flagellar filaments are thought to possess cytoskeletal as well as motility-related functions. The evidence revealing the effect of flagellar structure on cell shape was primarily obtained from analysis of nonflagellated mutants. Recently, the FlaB knockout mutant of B. burgdorferi has been complemented to the wild-type morphology [Sartakova et al., 2001]. Interestingly, complementation was only achieved after integration of the complementing gene and associated plasmid into the chromosome. The reason for this integration requirement remains unknown. While it is clear that the flagellum affects the shape of the cell, the effect differs among the various genera of spirochetes. The breadth of the range of effects on the cell shape may stem from the variable composition of the spirochete cell cylinders, some of which are likely to be more

The complex spirochete flagellar filament structure has been directly analyzed by genetic methods. In B. hyodysenteriae, flagellar filaments are composed of two FlaA and three FlaB polypeptides [Koopman et al., 1992]. Single knockouts of FlaA1 or FlaB1, resulted in mutant strains that exhibited altered motility [Kennedy et al., 1997; Rosey et al., 1995]. However, the structure of the flagellum from disrupted FlaA1 or FlaB1 mutant cells appeared morphologically similar to wild-type upon analysis by electron microscopy. Despite the loss of a major flagellar sheath protein, the apparent diameter of the filament measured in disrupted FlaA1 cells was similar to that of the wild type [Rosey et al., 1995]. However, more extensive analysis of purified flagellar filaments from FlaA-deficient mutants revealed that they were thinner than in the wild type, indicating that FlaA does indeed correspond to the flagellar sheath protein [Li et al., 2000a]. Additional studies have shown that mutants inactivated for either FlaB2 or FlaB3 also exhibit altered motility [Li et al., 2000a]. Interestingly, the cells of B. hyodysenteriae do not appear to compensate for the loss of one FlaB polypeptide by upregulating the synthesis of the other FlaB polypeptides. This is consistent with the finding of similar sigma 28 promoter sequences found upstream of the flaB genes in B. hyodysenteriae and Treponema spp. Although the morphology of the B. hyodysenteriae FlaB mutant flagellar filaments was similar to that of the wild type, the FlaA mutants had flagellar filaments with a smaller helix pitch and helix diameter as compared to the wild-type strain; thus FlaA impacts flagellar helical morphology [Li et al., 2000a]. Cellular processes that define and direct the number of periplasmic flagellar filaments that are inserted at each end of the spirochete cell have not been elucidated. Although the numbers of filaments found at each cell end varies, overall the filament numbers per cell end are rather tightly programmed in most species. Similarly, the distribution of the filaments at each cell end is invariant. During spirochete cell division, new filaments are seen

The Periplasmic Flagellum of Spirochetes

J Mol Microbiol Biotechnol 2004;7:30–40

Additional Genetic Analysis of Flagellar Filament Composition and Distribution

37

growing near the site of septation, and these short nascent filaments are one of the earliest indicators of the cell division process [Izard et al., 1999]. The factors that actually determine the number and location of flagellar filaments are unknown. Recently, we constructed three mutants of T. denticola that had FliG deleted or truncated by a varying number of amino acids. These mutants were all severely defective, but not totally deficient, in flagellar synthesis. Overall, truncation and deletion of T. denticola FliG markedly decreased flagellar filament numbers, length, and also altered their distribution [R. Limberger, L. Slivienski-Gebhardt, J. Izard, unpubl.]. These mutants demonstrate that FliG is important for an intact flagellar assembly process in spirochetes. The control of flagellar filament synthesis and polarity in spirochetes is an area ripe for exploration, especially since the long length of the spirochetal cell provides an opportunity to observe the processes involved in molecular distribution.

Regulation of Flagellar Filament Expression

As indicated earlier, the spirochetes have an organization of their flagellar genes that is clearly different from that in other bacteria. Moreover, the complex treponemal flagellar filament gene regulatory system apparently involves at least two promoters: the sigma 70-like promoter for the sheath protein, and the sigma 28-like promoters for the core filament proteins and other flagellar structures. The necessity of this regulatory system to the cell is unclear. It appears that spirochetes in general do not have an aflagellate stage during growth in cell culture. Changes in growth media designed to inhibit flagellar synthesis have failed to reduce flagellar synthesis in wild-type cells of T. denticola [R. Limberger, unpubl.]. Perhaps the inability to observe flagella-less wild-type spirochetes is only due to the limited likelihood of detecting this stage. B. burgdorferi motility genes, on the other hand, possess only sigma 70 promoters, and their regulation is distinct from that in Treponema as well as other bacteria [Ge et al., 1997]. One consistent paradox has emerged from study of flagellar filament-deficient mutants of spirochetes. Specific mutations that knock out flagellum-associated structures often have no deleterious effect on the transcription of flagellar-filament mRNA. This phenomenon has been noted in both Treponema [Limberger et al., 1999] and Borrelia [Charon and Goldstein, 2002]. For example, we have made insertional mutations in T. denticola FliK that also abolished the flagellar filament structures. Moreover, no

38

J Mol Microbiol Biotechnol 2004;7:30–40

FlaA or FlaB filament polypeptides were detectable in the cells or growth medium by immunoblotting [Limberger et al., 1999]. However, flaB mRNA was easily detectable by Northern analysis, and it appeared to be present in similar quantity as in the wild type. Sequencing of one flaB gene, and flanking DNA revealed that the DNA sequence was identical to wild type [R. Limberger, unpubl.]. Knockouts of flgE, fliG, and orf4 in T. denticola all have a deficiency in flagellar filament synthesis, yet quantitative RT-PCR has confirmed that levels of flaB mRNA are comparable to the levels in the wild type [R. Limberger, L. SlivienskiGebhardt, unpubl.]. It is unclear whether post-translational control mechanisms are the primary explanation for this phenomenon.

Coordination between the Flagella of a Spirochete Cell

The length of the spirochete cell provides an advantage in the study of motility and chemotaxis of bacteria. For spirochetes, chemotactic ability is dependent on coordination of the flagellar rotation between each cell end [Li et al., 2002; Shi et al., 1998]. Cells that are not coordinated may be able to move their ends, but the cells will be unable to propel themselves toward or away from substrates. The mechanisms for sensing substrates and subsequent coordination of the movement of the cell ends are unknown, but such spatial sensing functions are within the capabilities of the bacterial cell [Shapiro and Losick, 2000; Thar and Kuhl, 2003]. The recent review on spirochete motility expands on these theories [Charon and Goldstein, 2002], but future work will be aimed at resolving the coordination and communication between the two widely separated flagellar insertion regions of spirochetes.

Acknowledgments The author would like to thank all present and former members of his laboratory for their efforts, Stuart Lehrman for the illustrations, Bill Samsonoff for the electron microscopy, and Jacques Izard and Linda Slivienski-Gebhardt for their extensive assistance and helpful discussions. A special thanks to Nyles Charon for helpful comments and for introducing me to the world of spirochetes. The work in the author’s laboratory has been funded through the National Institutes of Health grant AI34354.

Limberger

References Barbour, A.G., Hayes, S.F., Heiland, R.A., Schrumpf, M.E., and Tessier, S.L. 1986. A Borrelia-specific monoclonal antibody binds to a flagellar epitope. Infect. Immun. 52:549–554. Berg, H.C. 1976. How spirochetes may swim. J. Theor. Biol. 56:269–273. Berg, H.C., Bromley, D.B., and Charon, N.W. 1978. Leptospira motility. In S. R. Y., H. J. Rogers & J. B. Ward (ed.), Relations between structure and function in the prokayotic cell: 28th Symposium for the Society for General Microbiology. Cambridge University Press, Cambridge. Birch-Andersen, A., Hovind Hougen, K., and BorgPetersen, C. 1973. Electron microscopy of Leptospira. 1. Leptospira strain Pomona. Acta Pathol. Microbiol. Scand. [B] Microbiol Immunol. 81:665–676. Brahamsha, B., and Greenberg, E.P. 1988. A biochemical and cytological analysis of the complex periplasmic flagella from Spirochaeta aurantia. J. Bacteriol. 170:4023–4042. Bromley, D.B., and Charon, N.W. 1979. Axial filament involvement in the motility of Leptospira interrogans. J. Bacteriol. 137:1406–1412. Canale-Parola, E. 1984. Motility and chemotaxis of spirochetes. Annu. Rev. Microbiol. 32:69–99. Champion, C.I., Miller, J.N., Lovett, M.A., and Blanco, D.R. 1990. Cloning, sequencing, and expression of two class B endoflagellar genes of Treponema pallidum subsp. pallidum encoding the 34.5 and 31.0-kilodalton proteins. Infect. Immun. 58:1697–1704. Charon, N.W., and Goldstein, S.F. 2002. Genetics of motility and chemotaxis of a fascinating group of bacteria: the spirochetes. Annu. Rev. Genet. 36:47–73. Charon, N.W., Goldstein, S.F., Block, S.M., Kurci, K., Ruby, J.D., Kreiling, J.A., and Limberger, R.J. 1992a. Morphology and dynamics of protruding spirochete periplasmic flagella. J. Bacteriol. 174:832–840. Charon, N.C., Goldstein, S.F., Curci, K., and Limberger, R.J. 1991. The bent-end morphology of Treponema phagedenis is associated with short, left-handed, periplasmic flagella. J. Bacteriol. 173:4820–4826. Charon, N.W., Greenberg, E.P., Koopman, M.B.H., and Limberger, R.J. 1992b. Spirochete chemotaxis, motility, and the structure of the spirochetal periplasmic flagella. Res. Microbiol. 143:597–603. Chi, B., Limberger, R.J., and Kuramitsu, H.K. 2002. Complementation of a Treponema denticola flgE mutant with a novel coumermycin A1-resistant T. denticola shuttle vector system. Infect. Immun. 70:2233–2237. Cockayne, A., Sanger, R., Ivic, A., Strugnell, R.A., MacDougall, J.H., Russell, R.R.B., and Penn, C.W. 1989. Antigenic and structural analysis of Treponema denticola. J. Gen. Microbiol. 135: 3209–3218. Coleman, J.L., and Benach, J.L. 1989. Identification and characterization of an endoflagellar antigen of Borrelia burgdorferi. J. Clin. Invest. 84:322–330.

The Periplasmic Flagellum of Spirochetes

Eipert, S.R., and Black, S.H. 1979. Characterization of the cytoplasmic fibrils of Treponema refringens (Nichols). Arch. Microbiol. 120: 205–214. Fraser, C.M., Casjens, S., Huang, W.M., Sutton, G.G., Clayton, R., Lathigra, R., White, O., Ketchum, K.A., Dodson, R., Hickey, E.K., Gwinn, M., Dougherty, B., Tomb, J.F., Fleischmann, R.D., Richardson, D., Peterson, J., Kerlavage, A.R., Quackenbush, J., Salzberg, S., Hanson, M., van Vugt, R., Palmer, N., Adams, M.D., Gocayne, J., Weidman, J., Utterback, T., Watthey, L., McDonald, L., Artiach, P., Bowman, C., Garland, S., Fujii, C., Cotton, M., Horst, K., Roberts, K., Hatch, B., Smith, H.O., and Venter, J.C. 1997. Genomic sequence of a Lyme disease spirochaete, Borrelia burgdorferi. Nature 390:580–586. Fraser, C.M., Norris, S.J., Weinstock, G.M., White, O., Sutton, G.G., Dodson, R., Gwinn, M., Hickey, E.K., Clayton, R., Ketchum, K.A., Sodergren, E., Hardham, J.M., McLeod, M.P., Salzberg, S., Peterson, J., Khalak, H., Richardson, D., Howell, J.K., Chidambaram, M., Utterback, T., McDonald, L., Artiach, P., Bowman, C., Cotton, M.D., Fujii, C., Garland, S., Hatch, B., Horst, K., Roberts, K., Sandusky, M., Weidman, J., Smith, H.O., and Venter, J.C. 1998. Complete genome sequence of Treponema pallidum, the syphilis spirochete. Science 281:375–388. Ge, Y., Li, C., Corum, L., Slaughter, C.A., and Charon, N.W. 1998. Structure and expression of the FlaA periplasmic flagellar protein of Borrelia burgdorferi. J. Bacteriol. 180:2418–2425. Ge, Y., Old, I.G., Girons, I.S., and Charon, N.W. 1997. The flgK motility operon of Borrelia burgdorferi is initiated by a sigma 70-like promoter. Microbiology. 143:1681–1690. Goldstein, S.F., Buttle, K.F., and Charon, N.W. 1996. Structural analysis of the Leptospiraceae and Borrelia burgdorferi by high-voltage electron microscopy. J. Bacteriol. 178:6539–6545. Goldstein, S.F., and Charon, N.W. 1988. Motility of the spirochete Leptospira. Cell. Motil. Cytoskeleton. 9:101–118. Goldstein, S.F., Charon, N.W., and Kreiling, J.A. 1994. Borrelia burgdorferi swims with a planar waveform similar to that of eukaryotic flagella. Proc. Natl. Acad. Sci. U S A. 91:3433–3437. Greenberg, E.P., and Canale-Parola, E. 1977. Relationship between cell coiling and motility of spirochetes in viscous environments. J. Bacteriol. 131:960–969. Hirano, T., Yamaguchi, S., Oosawa, K., and Aizawa, S. 1994. Roles of FliK and FlhB in determination of flagellar hook length in Salmonella typhimurium. J. Bacteriol. 176:5439–5449. Holt, J.G., Krieg, N.R., Sneath, P.H.A., Staley, J.T., and Williams, S.T. 1994. The Spirochetes, pp. 27–21, Bergey’s Manual of Determinative Bacteriology, 9th ed. Williams & Wilkins, Baltimore.

Holt, S.C. 1978. Anatomy and chemistry of spirochetes. Microbiol. Rev. 42:114–160. Hovind-Hougen, K. 1972. Further observations on the ultrastructure of Treponema pallidum Nichols. Acta Pathol. Scand. Sect. B 80:297–304. Hovind-Hougen, K. 1974. The ultrastructure of cultivable treponemes. I. Treponema phagedenis, Treponema vincentii, and Treponema refringens. Acta Pathol. Scand. Sect. B 82:329– 344. Hovind-Hougen, K. 1976. Determination by means of electron microscopy of morphological criteria of value for classification of some spirochetes, in particular treponemes. Acta Pathol. Scand. Sect. B 255:1–41. Hovind-Hougen, K., and Birch-Andersen, A. 1971. Electron microscopy of endoflagella and microtubules in Treponema reiter. Acta Pathol. Microbiol. Scand. 79:37–50. Isaacs, R.D., Hanke, J.H., Guzman-Verduzco, L.M., Newport, G., Agabian, N., Norgard, M.V., Lukehart, S.A., and Radolf, J.D. 1989. Molecular cloning and DNA sequence analysis of the 37-kilodalton endoflagellar sheath protein gene of Treponema pallidum. Infect. Immun. 56:3403–3411. Izard, J., McEwen, B.F., Barnard, R.M., Portuese, T., Samsonoff, W.A., and Limberger, R.J. 2003. Tomographic reconstruction of treponemal cytoplasmic filaments reveals novel bridging and anchoring components. Mol. Microbiol., in press. Izard, J., Samsonoff, W.A., Kinoshita, M.B., and Limberger, R.J. 1999. Genetic and structural analysis of the cytoplasmic filaments of wildtype and flagellar filament mutant of Treponema phagedenis. J. Bacteriol. 181:6739–6746. Izard, J., Samsonoff, W.A., and Limberger, R.J. 2001. Cytoplasmic filament-deficient mutant of Treponema denticola has pleiotropic defects. J. Bacteriol. 183:1078–1084. Jwang, B., Dewing, P., Fikrig, E., and Flavell, R.A. 1995. The hook protein of Borrelia burgdorferi, encoded by the flgE gene, is serologically recognized in Lyme disease. Clin. Diagn. Lab. Immunol. 2:609–615. Kawagishi, I., Homma, M., Williams, A.W., and Macnab, R.M. 1996. Characterization of the flagellar hook length control protein fliK of Salmonella typhimurium and Escherichia coli. J. Bacteriol. 178:2954–2959. Kennedy, M.J., Rosey, E.L., and Yancey, R.J., Jr. 1997. Characterization of flaA – and flaB – mutants of Serpulina hyodysenteriae: Both flagellin subunits, FlaA and FlaB, are necessary for full motility and intestinal colonization. FEMS Microbiol. Lett. 153:119–128. Kimsey, R.B., and Spielman, A. 1990. Motility of Lyme disease spirochetes in fluids as viscous as the extracellular matrix. J. Infect. Dis. 162: 1205–1208. Klitorinos, A., Noble, P., Siboo, R., and Chan, E.C. 1993. Viscosity-dependent locomotion of oral spirochetes. Oral Microbiol. Immunol. 8:242– 244.

J Mol Microbiol Biotechnol 2004;7:30–40

39

Koopman, M.B., Baats, E., van Vorstenbosch, C.J., van der Zeijst, B.A., and Kusters, J.G. 1992. The periplasmic flagella of Serpulina (Treponema) hyodysenteriae are composed of two sheath proteins and three core proteins. J. Gen. Microbiol. 138 ( Pt 12):2697–2706. Li, C., Bakker, R.G., Motaleb, M.A., Sartakova, M.L., Cabello, F.C., and Charon, N.W. 2002. Asymmetrical flagellar rotation in Borrelia burgdorferi nonchemotactic mutants. Proc. Natl. Acad. Sci. USA 99:6169–6174. Li, C., Corum, L., Morgan, D., Rosey, E.L., Stanton, T.B., and Charon, N.W. 2000a. The spirochete FlaA periplasmic flagellar sheath protein impacts flagellar helicity. J. Bacteriol. 182: 6698–6706. Li, C., Motaleb, A., Sal, M., Goldstein, S.F., and Charon, N.W. 2000b. Spirochete periplasmic flagella and motility. J. Mol. Microbiol. Biotechnol. 2:345–354. Li, H., Ruby, J., Charon, N., and Kuramitsu, H. 1996. Gene inactivation in the oral spirochete Treponema denticola: Construction of an flgE mutant. J. Bacteriol. 178:3664–3667. Limberger, R.J., and Charon, N.W. 1986a. Treponema phagedenis has at least two proteins residing together on its periplasmic flagella. J. Bacteriol. 166:105–112. Limberger, R.J., and Charon, N.W. 1986b. Antiserum to the 33,000-dalton periplasmic-flagellum protein of ‘Treponema phagedenis’ reacts with other treponemes and Spirochaeta aurantia. J. Bacteriol. 168:1030–1032. Limberger, R.J., Slivienski, L.L., Izard, J., and Samsonoff, W.A. 1999. Insertional inactivation of Treponema denticola tap1 results in a nonmotile mutant with elongated flagellar hooks. J. Bacteriol. 181:3743–3750. Limberger, R.J., Slivienski, L.L., and Samsonoff, W.A. 1994. Genetic and biochemical analysis of the flagellar hook of Treponema phagedenis. J. Bacteriol. 176:3631–3637. Lux, R., Miller, J.N., Park, N.H., and Shi, W. 2001. Motility and chemotaxis in tissue penetration of oral epithelial cell layers by Treponema denticola. Infect. Immun. 69:6276–6283. Lux, R., Moter, A., and Shi, W. 2000. Chemotaxis in pathogenic spirochetes: Directed movement toward targeting tissues? J. Mol. Microbiol. Biotechnol. 2:355–364. Macnab, R.M. 1992. Genetics and biogenesis of bacterial flagella. Annu. Rev. Genet. 26:131– 158. Macnab, R.M. 2003. How bacteria assemble flagella. Annu. Rev. Microbiol. 57:77–100. Minamino, T., Gonzalez-Pedrajo, B., Yamaguchi, K., Aizawa, S.I., and Macnab, R.M. 1999. FliK, the protein responsible for flagellar hook length control in Salmonella, is exported during hook assembly. Mol. Microbiol. 34:295– 304.

40

Motaleb, M.A., Corum, L., Bono, J.L., Elias, A.F., Rosa, P., Samuels, D.S., and Charon, N.W. 2000. Borrelia burgdorferi periplasmic flagella have both skeletal and motility functions. Proc. Natl. Acad. Sci. U S A. 97:10899–10904. Moter, A., Hoenig, C., Choi, B.K., Riep, B., and Göbel, U.B. 1998. Molecular epidemiology of oral treponemes associated with periodontal disease. J. Clin. Microbiol. 36:1399–1403. Muramoto, K., Makishima, S., Aizawa, S.I., and Macnab, R.M. 1998. Effect of cellular level of FliK on flagellar hook and filament assembly in Salmonella typhimurium. J. Mol. Biol. 277: 871–882. Nauman, R.K., Holt, S.C., and Cox, C.D. 1969. Purification, ultrastructure, and composition of axial filaments from Leptospira. J. Bacteriol. 98:264–280. Norris, S.J., Charon, N.W., Cook, R.G., Fuentes, M.S., and Limberger, R.J. 1988. Antigenic relatedness and N-terminal sequence homology define two classes of periplasmic flagellar proteins of Treponema pallidum subsp. pallidum and Treponema phagedenis. J. Bacteriol. 170: 4072–4082. Paster, B.J., Boches, S.K., Galvin, J.L., Ericson, R.E., Lau, C.N., Levanos, V.A., Sahasrabudhe, A., and Dewhirst, F.E. 2001. Bacterial diversity in human subgingival plaque. J. Bacteriol. 183: 3770–3783. Picardeau, M., Brenot, A., and Saint Girons, I. 2001. First evidence for gene replacement in Leptospira spp. Inactivation of L. biflexa flaB results in nonmotile mutants deficient in endoflagella. Mol. Microbiol. 40:189–199. Qiu, Y.S., Klitorinos, A., Rahal, M.D., Siboo, R., and Chan, E.C. 1994. Enumeration of viable oral spirochetes from periodontal pockets. Oral Microbiol. Immunol. 9:301–304. Radolf, J.D., Blanco, D.R., Miller, J.N., and Lovett, M.A. 1986. Antigenic interrelationship between endoflagella of Treponema phagedenis biotype Reiter and Treponema pallidum (Nichols). II. Molecular characterization of endoflagellar proteins. Infect. Immun. 54:626–634. Ren, S.X., Fu, G., Jiang, X.G., Zeng, R., Miao, Y.G., Xu, H., Zhang, Y.X., Xiong, H., Lu, G., Lu, L.F., Jiang, H.Q., Jia, J., Tu, Y.F., Jiang, J.X., Gu, W.Y., Zhang, Y.Q., Cai, Z., Sheng, H.H., Yin, H.F., Zhang, Y., Zhu, G.F., Wan, M., Huang, H.L., Qian, Z., Wang, S.Y., Ma, W., Yao, Z.J., Shen, Y., Qiang, B.Q., Xia, Q.C., Guo, X.K., Danchin, A., Saint Girons, I., Somerville, R.L., Wen, Y.M., Shi, M.H., Chen, Z., Xu, J.G., and Zhao, G.P. 2003. Unique physiological and pathogenic features of Leptospira interrogans revealed by whole-genome sequencing. Nature 422:888–893.

J Mol Microbiol Biotechnol 2004;7:30–40

Rosey, E.L., Kennedy, M.J., Petrella, D.K., Ulrich, R.G., and Yancey, R.J., Jr. 1995. Inactivation of Serpulina hyodysenteriae flaA1 and flaB1 periplasmic flagellar genes by electroporationmediated allelic exchange. J. Bacteriol. 177: 5959–5970. Ruby, J.D., and Charon, N.W. 1998. Effect of temperature and viscosity on the motility of the spirochete Treponema denticola. FEMS Microbiol. Lett. 169:251–254. Ruby, J.D., Li, H., Kuramitsu, H., Norris, S.J., Goldstein, S.F., Buttle, K.F., and Charon, N.W. 1997. Relationship of Treponema denticola periplasmic flagella to irregular cell morphology. J. Bacteriol. 179:1628–1635. Sadziene, A., Thomas, D.D., Bundoc, V.G., Holt, S.C., and Barbour, A.G. 1991. A flagella-less mutant of Borrelia burgdorferi: structural, molecular and in vitro functional characterization. J. Clin. Invest. 88:82–92. Sartakova, M.L., Dobrikova, E.Y., Motaleb, M.A., Godfrey, H.P., Charon, N.W., and Cabello, F.C. 2001. Complementation of a nonmotile flaB Mutant of Borrelia burgdorferi by chromosomal Integration of a plasmid containing a wild-type flaB Allele. J. Bacteriol. 183:6558– 6564. Shapiro, L., and Losick, R. 2000. Dynamic spatial regulation in the bacterial cell. Cell. 100:89– 98. Shi, W., Yang, Z., Geng, Y., Wolinsky, L.E., and Lovett, M.A. 1998. Chemotaxis in Borrelia burgdorferi. J. Bacteriol. 180:231–235. Simonson, L.G., Goodman, C.H., Bial, J.J., and Morton, H.E. 1988. Quantitative relationship of Treponema denticola to severity of periodontal disease. Infect. Immun. 56:726–728. Thar, R., and Kuhl, M. 2003. Bacteria are not too small for spatial sensing of chemical gradients: an experimental evidence. Proc. Natl. Acad. Sci. USA. 100:5748–5753. Thomas, D.D., Navab, M., Haake, D.A., Fogelman, A.M., Miller, J.N., and Lovett, M.A. 1988. Treponema pallidum invades intracellular junctions of endothelial cell monolayers. Proc. Natl. Acad. Sci. USA. 8:3608–3612. Trueba, G.A., Bolin, C.A., and Zuerner, R.L. 1992. Characterization of the periplasmic flagellum proteins of Leptospira interrogans. J. Bacteriol. 174:4761–4768. Williams, A.W., Yamaguchi, S., Togashi, F., Aizawa, S.I., Kawagishi, I., and Macnab, R.M. 1996. Mutations in fliK and flhB affecting flagellar hook and filament assembly in Salmonella typhimurium. J. Bacteriol. 178:2960–2970. You, Y., Elmore, S., Colton, L.L., Mackenzie, C., Stoops, J.K., Weinstock, G.M., and Norris, S.J. 1996. Characterization of the cytoplasmic filament protein gene (cfpA) of Treponema pallidum subsp. pallidum. J. Bacteriol. 178:3177– 3187. Young, K.D. 2003. Bacterial shape. Mol. Microbiol. 49:571–580.

Limberger

J Mol Microbiol Biotechnol 2004;7:41–51 DOI: 10.1159/000077868

Recent Advances in the Structure and Assembly of the Archaeal Flagellum Sonia L. Bardy Sandy Y.M. Ng Ken F. Jarrell Department of Microbiology and Immunology, Queen’s University, Kingston, Ont., Canada

Key Words Methanococcus voltae W Preflagellin peptidase W Type IV pili W Glycosylation W Motility

Abstract Archaeal motility occurs through the rotation of flagella that are distinct from the flagella found on bacteria. The differences between the two structures include the multiflagellin nature of the archaeal filament, the widespread posttranslational modification of the flagellins and the presence of a short signal peptide on each flagellin that is cleaved by a specific signal peptidase prior to the incorporation of the mature flagellin into the flagellar filament. Research has revealed similarities between the archaeal flagellum and the type IV pilus, including the presence of similar unusual signal peptides on the flagellins and pilins, similarities in the amino acid sequences of the major structural proteins themselves, as well as similarities between potential assembly and processing components. The recent suggestion that type IV pili are part of a family of cell surface complexes, coupled with the similarities between type IV pili and archaeal flagella, raise questions about the evolution of these systems and possible inclusion of archaeal flagella into this surface complex family.

Introduction

Motility is a common trait in prokaryotes, and Archaea possess a structure for swimming that appears to be unique among the prokaryotes. While Bacteria also use swimming as one of their forms of motility, it has become apparent that organisms from these two prokaryotic Domains of life [Woese and Fox, 1977] achieve this similar type of motility through very different mechanisms. Archaeal genomes lack any homologues to genes involved in bacterial flagellation [Faguy and Jarrell, 1999], and examination of the physical flagellar structure has revealed a myriad of differences between bacterial and archaeal flagella (table 1). These differences include, but are not limited to, the multiflagellin nature of the archaeal filament, novel enzymes involved in the biosynthesis of archaeal flagella, and likely the mechanism of assembly. In fact, the more the archaeal flagella are studied, the more they are found to be similar to another bacterial motility system, type IV pili. Comparisons of archaeal flagella with type IV pili have revealed the presence of homologous proteins in these two systems, N-terminal sequence similarities between flagellins and pilins, as well as the presence of signal peptides on both archaeal flagellins and type IV pilins that are processed by homologous specialized signal peptidases.

Copyright © 2004 S. Karger AG, Basel

ABC Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

© 2004 S. Karger AG, Basel 1464–1801/04/0072–0041$21.00/0 Accessible online at: www.karger.com/mmb

Ken F. Jarrell Department of Microbiology and Immunology Queen’s University Kingston, Ontario, K7L 3N6 (Canada) Tel. +1 613 533 2456, Fax +1 613 533 6796, E-Mail [email protected]

Table 1. Comparison of bacterial flagella, archaeal flagella and type IV pili

Bacterial flagella

Archaeal flagella

Bacterial type IV pili

Single flagellin with some exceptions No sequence similarities between bacterial and archaeal flagellins

multiple flagellins

single major pilin with minor pilin-like proteins N-terminal sequence similarities with archaeal flagellins

Flagellins rarely posttranslationally modified

flagellins commonly posttranslationally modified, usually through glycosylation

pilins may be glycosylated or phosphorylated depending on the species

Flagellins do not have N-terminal leader peptides

flagellins have N-terminal leader peptides cleaved by FlaK (homolog of PilD)

prepilins have N-terminal leader peptides cleaved by PilD (homolog of FlaK) FlaI is homologous to PilB, PilT and TadA, NTPases involved in the assembly/ disassembly process of the type IV pili FlaJ has sequence similarity with TadB, an integral membrane protein involved in type IV pili biogenesis

No archaeal homologues of genes involved in bacterial flagellation (flagellins, rod, hook, hook-associated, ring, switch, mot)

Basal body with rings and rod

anchoring structure not observed

anchoring structure not reported

Hook region

no hook

Well-defined hook region length of about 55 nm

curved hook-like region likely composed of FlaB3 (in M. voltae) located close to the anchoring structure great variability in length of the curved region (72–265 nm)

F20 nm diameter 2 nm channel

F10–15 nm diameter channel not detected

F 5–7 nm diameter no channel

Swimming motility

swimming motility

surface-mediated twitching motility

Rotating filament swim/tumble motion

rotating filament push/pull motion in Halobacterium

movement by filament retraction/ elongation filament assembly mediated by PilB/filament disassembly by PilT

Type IV pili are members of a family of cell surface complexes which are widely distributed throughout both gram-negative and gram-positive bacteria, and which are involved in macromolecular transport. These systems all contain multiple sets of homologies between structural proteins (pilins and pseudopilins with atypical signal peptides) as well as assembly and processing proteins (ATPases and specific signal peptidases). Besides type IV pili, these protein families are involved in type II secretion in a mechanism which appears to involve a pseudopilus structure (the type II secretion piston) [Vignon et al., 2003] and in DNA uptake in certain gram-positive bacteria. While all these systems have evolved different functions, they appear to be linked by an overall similar core structure [Mattick, 2002]. In the archaea, the presence of

42

J Mol Microbiol Biotechnol 2004;7:41–51

pilin-like proteins (the archaeal flagellins), the conserved ATPase and unique signal peptidase, as well as a similar core structure [Cohen-Krausz and Trachtenberg, 2002], suggest that in this Domain these components have been adapted to form the archaeal flagellum. A thorough review of archaeal flagella has recently been completed [Thomas et al., 2001a], and this review will focus on those areas where progress has since been made. It is possible that further understanding of archaeal flagellation, assembly and function will provide insights into many unique archaeal processes, including protein secretion.

Bardy/Ng/Jarrell

Fig. 1. Flagella gene families of selected archaea. Similar colours represent homologues shared among families. The

white, unlabeled box in Archaeoglobus fulgidus has no sequence similarity to genes in universal databases. The genes are transcribed in the direction of the respective arrows. The flaF gene designated in H. salinarum is not a homologue of the flaF genes in M. voltae and T. volcanium. The B flagellin genes of H. salinarum are adjacent to the accessory genes, while the A flagellin genes are located elsewhere on the chromosome.

Distribution

Flagellation is widespread throughout the Domain Archaea. It is a feature that encompasses many genera in both archaeal Kingdoms (Crenarchaeota and Euryarchaeota) as well as all the major subgroupings (methanogens, extreme halophiles, haloalkaliphiles, hyperthermophiles and even Thermoplasma species which lack a cell wall). Three particularly intriguing cases are Methanosarcina, Methanopyrus and Pyrobaculum. In Methanosarcina, apparently intact flagellin genes are found in both M. acetivorans and M. mazei, although the cells of both species have never been reported as motile or flagellated [Galagan et al., 2002; Deppenmeier et al., 2002]. Perhaps there are special growth conditions that result in the production of flagella in this genus. In the case of both Methanopyrus kandleri and Pyrobaculum aerophilum, electron micrographs show apparently flagellated organisms [Hu-

Archaeal Flagella

ber and Stetter, 2001; Kurr et al., 1991], yet the completely sequenced genomes lack obvious homologues to the archaeal flagellins [Fitz-Gibbon et al., 2002]. Obviously these structures need to be isolated and the genes identified through reverse genetics in order to confirm identification of these structures as flagella.

Archaeal Flagella Gene Families

In all the flagellated archaeal species with available sequence data, multiple flagellin genes arranged in tandem are always present (with the rare exceptions of Sulfolobus spp. which have genomes annotated with single flagellin genes). The multiflagellin nature of the archaeal flagellum is intriguing, and available evidence suggests that individual flagellins have specialized roles and are not interchangeable entities (see below).

J Mol Microbiol Biotechnol 2004;7:41–51

43

While the genomic arrangement and composition of genes involved in flagellation can be quite idiosyncratic depending on the archaeon being studied (fig. 1), flaHIJ are present in every flagellated archaeon with available sequence data. This conservation, taken together with their homology to bacterial type IV pili genes [Bayley and Jarrell, 1998], led to the postulation that this Archaea-specific gene cluster is vital in flagellum biosynthesis and/or function. Early studies involving an insertional mutation in flaH revealed that this mutant strain is nonmotile and nonflagellated [Thomas et al., 2001b]. Immunoblotting experiments using antibodies against flagellins ascertained that the mutant makes all the flagellins and at least some putative flagella accessory proteins upstream of flaH; hence the defect is likely in the assembly process. However, since the flaHIJ cluster is co-transcribed [Thomas et al., 2002], this finding only demonstrated that at least one of the genes, flaHIJ, was critical for flagella assembly. A subsequent study utilized a novel mutagenic vector harboring a methanococcal promoter (hmvA) that was able to restart transcription downstream of the insertion to individually characterize the functions of flaI and flaJ [Thomas et al., 2002]. It was found that both mutants were similarly nonmotile and nonflagellated. The requirement of flaJ in flagellation is clear, since flaJ is the last gene in the transcriptional unit and the phenotype observed is not due to a transcriptional polar effect or improper protein expression. Due to the lack of FlaJ antibodies, FlaJ expression in the flaI knockout could not be proven. Despite this, RTPCR demonstrated that the flaI mutant possessed an flaJ mRNA transcript, leading to the belief that flaI is most likely essential as well. Similar results were found in H. salinarium, in which deletion mutants of flaI were found to be nonmotile and nonflagellated [Patenge et al., 2001]. The flaI deletion in this case was complemented in trans, and the phenotype of the deletion strain was found to be directly attributable to the lack of functional FlaI and not from polar effects on downstream genes. mRNA analysis revealed no difference in flagellin gene mRNA levels between wild type and the mutant strain, indicating that flaI is not involved in transcriptional regulation of flagellin gene expression. Indeed, all available archaeal flaI sequences contain a Walker box A nucleotide-binding site motif. FlaI is an archaeal member of the type IV family of predicted secretion NTPases which include PilB/T and appears to be specifically related to the TadA subfamily [Bayley and Jarrell, 1998; Kachlany et al., 2001]. Taken together, it is believed that the entire flaHIJ cluster is required for proper flagella translocation. Consistent

44

J Mol Microbiol Biotechnol 2004;7:41–51

with this theory, FlaH and FlaI have been demonstrated to localize to the cell membrane [Thomas and Jarrell, 2001]. In addition, FlaJ is predicted by dense alignment surface analysis to be a protein with 7–9 transmembrane domains [Cserzo et al., 1997], and is phylogenetically related to the TadB protein required for piliation of Actinobacillus actinomycetemcomitans [Kachlany et al., 2000]. Thomas and Jarrell [2001] proposed that FlaH, FlaI and FlaJ may form a transmembrane secretory complex, which allows for the secretion of flagellins for the filament across the membrane. The most recently identified gene involved in archaeal flagellation is the gene encoding the preflagellin peptidase (flaK). In archaeal flagella, the flagellin subunits are synthesized as preproteins, with signal peptides that are removed prior to the incorporation of the mature protein into the flagellar filament [Bardy and Jarrell, 2003; Correia and Jarrell, 2000]. In M. jannaschii, flaK is located immediately downstream of the flagellin and flagella associated genes. However, the organization in at least two other methanococci is very different. In both M. maripaludis and M. voltae, flaK is located separate from the flagella gene family. A gene following flaJ in H. salinarum was identified as flaK, but does not show similarity to the preflagellin peptidases [Patenge et al., 2001]. Instead, it has been cited as having similarity to fleN, which is responsible for regulating the number of bacterial flagella [Dasgupta et al., 2000]. The presence of other flagella accessory genes varies depending on the species, suggesting that some gene products likely have specific roles and are not required in all archaea. For instance, flaF is present in the flagella gene families of methanogens and Thermoplasma spp. The gene designated flaF in H. salinarum is unique and unrelated to the aforementioned flaF. FlaG protein shares sequence similarity to archaeal flagellins at the N-terminus. It has been speculated that FlaF and FlaG might be the equivalents of PilE and PilV in the type IV pilus system [Thomas et al., 2001a]. PilE and PilV have sequence similarity to pilins and may be minor structural proteins of the pilus [Mattick and Alm, 1995].

Ultrastructure

Flagellar Filament In bacterial flagella the filament is, in most cases, composed of thousands of copies of a single flagellin protein [Aizawa, 2000]. However, in the Archaea, multiple flagellins are an almost absolute feature of archaeal fla-

Bardy/Ng/Jarrell

gella. The only exceptions thus far are the species of Sulfolobus that have genomes annotated with single flagellin genes. This appears to be a specific feature of Sulfolobus and not of the Crenarchaeotes in general as the Aeropyrum pernix genome has been annotated as including two flagellin genes. In addition, the archaeal flagellins are often, and perhaps even universally posttranslationally modified by glycosylation. This is extremely rare in the case of bacterial flagellins [Doig et al., 1996; Moens et al., 1995]. The almost universal presence of multiple flagellins on such varied organisms suggests a requirement for each of the individual flagellins, and perhaps a specialized role for each. Transcriptional data support this theory, with different flagellins being transcribed at different levels [Kalmokoff et al., 1988]. Research in M. voltae, Halobacterium salinarum, and Natrialba magadii provides some insight into the need for multiple flagellins. Initial studies of the M. voltae flagellar filament revealed that the two major structural proteins were FlaB1 and FlaB2, while the other flagellins were not detected within the filament by SDS-PAGE [Kalmokoff et al., 1988]. Electron microscopy of purified flagellar filaments revealed the presence of a curved, hook-like region proximal to the cell surface. Shearing of the cells prior to isolation of the flagellar filament enriched for cell proximal proteins, and allowed for the detection of FlaB3 within the flagellar filament. Immunoblotting and immuno-electron microscopy revealed that FlaB3 localizes proximal to the cell surface, and is likely the major component of this curved hook-like region [Bardy et al., 2002]. Additional work using polyclonal antisera to the remaining flagellin, FlaA, indicated that this flagellin was likely found throughout the flagellar filament [Bardy et al., 2002]. That each of these flagellins may have a specialized role was further supported through mutant studies in M. voltae. Insertional inactivation of FlaA resulted in M. voltae possessing flagella that appeared similar to wild type, yet these cells were significantly less motile than wild type [Jarrell et al., 1996a]. These results demonstrate that while not all of the flagellins are required for assembly of the helical filament, the flagellins are not interchangeable, and all are required for full function. In H. salinarum, the flagellin localization studies were done entirely through insertional mutation studies [Tarasov et al., 2000]. In this organism, there are five flagellins composing the filament. The genes flgA1 and flgA2 are co-transcribed and located separate from the three remaining flagellin genes (flgB1, flgB2, and flgB3), which are also co-transcribed [Gerl and Sumper, 1988]. The fla-

gellar accessory genes are located on a separate transcript from both sets of flagellin genes [Patenge et al., 2001], which is distinct from the situation seen in M. voltae. Inactivation of both flgA genes resulted in short, curved flagella located at the sides and poles of the cells, instead of at the normal polar location. These truncated flagella were composed entirely of FlgB flagellins. Transcriptionally, there was a reduced amount of the flgB message, suggesting that the flgB genes are positively regulated by the FlgA flagellins. Specific inactivation of flgA2 resulted in the straight flagella located mainly at the poles. Transcriptional data revealed that these flagella were entirely composed of FlgA1, as there was no flgB transcript. Finally, the inactivation of the flgB genes resulted in spiral flagella located at the poles, similar in length to wild-type flagella, which were composed of both FlgA flagellins. At the cell proximal end of the flagella were outgrowths that were speculated to be basal-body-filled membrane sacs [Tarasov et al., 2000]. All cells resulting from these mutational studies were unable to make fully functional flagella. From these studies, the multiflagellin nature of H. salinarum flagella was explained by the requirement of both FlgA1 and FlgA2 for the formation of helical filaments, and the proposed role of the three B flagellins as cell proximal proteins (such as hook-associated proteins, terminators or anchors) that are essential for full function. The additional proposal that the FlgA flagellins are likely inserted into the filament prior to the insertion of the cell proximal proteins (FlgB flagellins; based on the proposed transcriptional regulation) is in direct contradiction with the established mechanism of assembly for bacterial flagella, and consistent with the current model of type IV pilus assembly [Mattick, 2002], and the proposed model for archaeal flagella [Jarrell et al., 1996b]. While the mechanism of assembly for archaeal flagella has not been directly demonstrated, evidence is accumulating which supports a unique mechanism distinct from the bacterial flagella (see below). In N. magadii, there are four flagellins (FlaB1-B4) comprising the flagellar filament. It has been found that when the salt concentration is lowered to less than 8%, the flagellar filaments tend to dissociate into protofilaments. These protofilaments show a disordered surface, have a thinner diameter (3–5 nm compared to 10–12 nm), and protein aggregates begin to appear. This change in morphology continues with decreasing NaCl concentrations down to zero, where only aggregates remained [Fedorov et al., 1994]. Polyclonal antibodies were raised against the individual flagellins to study the flagellar filaments and the protofilaments [Pyatibratov et al., 2002]. Due to the

Archaeal Flagella

J Mol Microbiol Biotechnol 2004;7:41–51

45

conserved nature of the flagellins, antibodies raised against any individual flagellin tended to be cross-reactive with other flagellins. Fortunately, the antibodies raised against FlaB4 were only cross-reactive with FlaB2; this antiserum was used in immunolabeling of both flagellar filaments and protofilaments. By immuno-electron microscopy all segments of the flagellar filaments were labeled, suggesting that FlaB2 and FlaB4 are evenly distributed throughout the filament. However, upon examining the protofilaments, it was apparent that some were homogenously labeled while others were not labeled at all. There was no difference in the thickness or morphology of the labeled and unlabeled protofilaments. This corresponds with the previous theory that the archaeal flagellar filament is formed by several rows (protofilaments) with each composed of one flagellin type [Fedorov et al., 1994]. It is currently unclear how this theory corresponds to the results seen in both H. salinarum and M. voltae, which show some flagellins having distinct roles and localization (i.e. cell proximal or throughout the filament). However, the authors also proposed that the helicity of archaeal flagella requires multiple flagellins (in this case FlaB4(2) and FlaB1), while FlaB3 likely has an auxiliary role as it is present in lesser amounts [Pyatibratov et al., 2002]. Beyond the requirement for at least two flagellins for filament helicity, it is currently not possible to make a universal statement about the requirement for multiple flagellins within the archaeal flagellum, and roles for each, due to the variety in number of flagellins present in each archaeon.

Anchoring Structures

The anchoring structure of the archaeal flagellum is unknown. Current isolation techniques generally reveal only the presence of a knob-like structure that lacks distinct features [Bardy et al., 2002; Kupper et al., 1994]. Examination of completely sequenced archaeal genomes demonstrated that archaea do not possess homologues to any genes encoding components of the bacterial hook basal-body complex [Faguy and Jarrell, 1999]. This indicates that the anchoring system of the archaeal flagellum is composed of completely different and archaeal specific proteins, which may or may not fulfill the same function. Antibodies have been generated to some of the flagellar associated proteins in M. voltae. While most of the flagellar-associated proteins localize to the cell membrane, immunoblotting of isolated flagella reveal that these proteins are not part of the flagellar structure, or if they are, the

46

J Mol Microbiol Biotechnol 2004;7:41–51

structure does not remain intact through the currently used isolation procedure [Thomas and Jarrell, 2001].

Archaeal Flagellins

Archaeal flagellins are synthesized as preproteins, with a signal peptide that is removed prior to the incorporation of the mature protein into the flagellar filament. This is one of the distinguishing features between archaeal and bacterial flagella; bacterial flagellins are not synthesized as preproteins, and are exported through the hollow flagellar filament by a type III secretion system, and are incorporated at the tip of the filament. The synthesis of archaeal flagellins initially as preproteins is similar to the situation in type IV pili, where the pilin subunits are also made as preproteins with a signal peptide that is removed prior to the incorporation of the mature pilins into the pilus structure [Strom et al., 1994]. An additional difference between archaeal flagella and bacterial flagella is the noticeable lack of sequence similarity between the two types of flagellins, whereas flagellins from distantly related bacteria share N- and C-terminal similarities [Wilson and Beveridge, 1993; Winstanley et al., 1994]. On the other hand, comparison of archaeal flagellins with type IV pilins does reveal readily detected amino acid sequence similarities at the N-termini [Faguy et al., 1994a]. Study of the flagellins has revealed sequence conservation in the portion of the signal peptide immediately preceding the cleavage site, and extending over the first approximately 50 amino acids of the mature protein [Thomas et al., 2001a]. Within the N-terminus of the mature protein, especially conserved are the amino acids for positions +8 to +19 (relative to the cleavage site). All 57 archaeal flagellins, for which sequence data are available, have alanine at position +12; all but the Sulfolobus flagellins have VAA at positions +16 to +18. All flagellins have glycine at +3, and all but one have isoleucine at +8 and valine at +23 (fig. 2). Interestingly, study of the +1 position revealed that while a variety of amino acids are found at this position, there appears to be absolute conservation within a genus. With the sole exception of FlaB1 of M. vannielii, all Methanococcus (17) and Pyrococcus (9; not including one apparent translational start error in P. furiosus) flagellins have an alanine at +1. All Thermoplasma flagellins (4) have a +1 glutamic acid, and all Methanosarcina flagellins have a +1 phenylalanine, as do the two flagellins from Archaeoglobus. Both Sulfolobus flagellins (from two species) have leucine at +1, and the flagellins from Aeropyrum have a +1 isoleucine. Finally, all

Bardy/Ng/Jarrell

Fig. 2. A sequence LOGO of archaeal fla-

gellins aligned at their cleavage site (without gaps) using the default settings of the WebLogo site (http://www.bio.cam.ac.uk/ cgi-bin/seqlogo/logo.cgi). The final amino acid in the signal peptide is shown at position 23.

halophilic flagellins from both Halobacterium and Natrialba have glutamine at the +1 position. The signal peptide of archaeal flagellins shares similarities with the signal peptide of type IV pilins [Faguy et al., 1994a]. The type IV pilin typically has a short signal peptide that ends in KG, with an alanine at –1 also permitting partial processing of the signal peptide [Strom and Lory, 1991]. The signal peptides of archaeal flagellins end with a basic amino acid followed by glycine, as well. The signature +1 phenylalanine and +5 glutamic acid of type IV pilins is also found in certain archaeal flagellins, specifically those of Methanosarcina spp., A. fulgidus and Methanospirillum hungatei [Faguy et al., 1994b]. Within the archaeal signal peptide itself, the –3 to –1 positions are extremely well conserved. The –1 position is almost always a glycine (51/57), with alanine found 5 times and a serine reported in one T. acidophilum flagellin. Site directed mutagenesis studies on M. voltae flagellin FlaB2 revealed that changing the –1 glycine to alanine permitted some processing of the signal peptide to occur [Thomas et al., 2001c]. A serine substitution was not tested. The –2 position in the signal peptide is either a lysine (32/57) or an arginine (25/57). Substitutions of other amino acids in this position did not allow for processing. Although the –3 position is less highly conserved, this place is often held by a charged amino acid, with both basic and acid residues occurring (19 lysine, 17 arginine, 10 glutamic acid, 6 aspartic acid, and 2 asparagine). Serine has been reported in this position in two T. acidophilum flagellins, and glycine in one N. magadii flagellin. The site-directed mutagenesis studies in M. voltae have revealed that changes in this position do allow for processing of the signal peptide [Thomas et al., 2001c].

The average length of the signal peptide of archaeal preflagellins is not entirely clear. In most cases, the amino acid sequences are derived from annotated complete genomes, and several translation start sites may be in error. From the TIGR website (www.tigr.org), there are cases of different annotations assigned by TIGR compared to the primary sequencing agency. This is particularly seen in the case of Halobacterium sp. NRC-1. In other cases, translational start sites begin at what is likely an interior site within the protein. Some predicted translational start sites in the annotated sequences result in extra long signal peptides that contain in frame methionines that would result in signal peptides of a more typical length (approximately 12 amino acids). A good example of this is the annotated start for FlaB3 of M. jannaschii, which results in a signal peptide 18 amino acids long (Genbank accession number NP_247888). However, this protein could have a signal peptide 11 amino acids in length if the in frame methonine contained in this signal peptide is the correct translational start. Still other flagellins have extremely short predicted signal peptides (4 amino acids), but to date none of these extremely short signal peptides have been biochemically proven to be the correct length. Given that truncating the signal peptide of M. voltae from 12 amino acids to 6 amino acids prevented processing [Thomas et al., 2001c], and that all biochemically proven signal peptide lengths have between 9 and 13 amino acids [Albers et al., 2003; Correia and Jarrell, 2000], it may be best to regard these exceptionally short signal peptides with a degree of skepticism until biochemical evidence demonstrates otherwise.

Archaeal Flagella

J Mol Microbiol Biotechnol 2004;7:41–51

47

Preflagellin Peptidase

It has long been recognized that the signal peptide of type IV pilins is removed by the prepilin peptidase (PilD and homologues). The prepilin peptidase is a membranebound bifunctional enzyme responsible for both the cleavage of the peptide bond and methylation of the resulting N-terminal amino acid of the mature protein. Knowing that the signal peptide of archaeal preflagellins is also removed prior to incorporation into the filament, an in vitro assay was developed to test for the removal of the signal peptide. This assay revealed that the required enzymatic activity was found in membranes of M. voltae [Correia and Jarrell, 2000]. Recently, the enzymes responsible for this processing were discovered in M. maripaludis [Bardy and Jarrell, 2002] and M. voltae [Bardy and Jarrell, 2003]. Overexpression of either methanococcal FlaK enzyme in Escherichia coli allowed for the processing of the M. voltae preflagellin in an in vitro assay [Bardy and Jarrell, 2002]. Insertional inactivation of flaK in M. voltae resulted in cells that were nonmotile and nonflagellated, and the flagellins of the flaK – cells migrated slower in SDSPAGE than the flagellins of wild-type cells, as expected if they retained their 11–12 amino acid signal peptide. The glycosylation of the M. voltae flagellins still occurs despite the inactivation of FlaK, indicating that processing of the signal peptide is not required for post-translational modification to occur [Bardy and Jarrell, 2003]. Additionally, membranes of the flaK – mutants were unable to process the signal peptide of M. voltae preflagellin in the in vitro assay. These data confirm that FlaK is indeed the preflagellin peptidase, and that this enzymatic activity is essential for flagellation within the archaea. Alignments of M. maripaludis FlaK with PppA of Esherichia coli revealed two conserved aspartic acid residues [Bardy and Jarrell, 2002]. These residues are part of the active site in type IV prepilin peptidases, which are members of a novel family of aspartic acid proteases [LaPointe and Taylor, 2000]. Additionally, alignments of multiple methanococcal FlaK enzymes revealed conservation of three additional aspartic acid residues [Bardy and Jarrell, 2003]. Site-directed mutagenesis was performed on all five conserved aspartic acid residues of M. voltae FlaK. The numbering of amino acids is based on the M. voltae sequence. Independent alteration of two aspartic acid residues (D18 and D79) to either alanine or asparagine residues eliminated the ability of overexpressed FlaK to process the signal peptide. In the case of D79, mutagenesis to glutamate resulted in the retention of activity (FlaK containing the D79E mutation could not be

48

J Mol Microbiol Biotechnol 2004;7:41–51

successfully expressed in E. coli). These two aspartic acid residues correspond to those of the type IV prepilin peptidases required for activity. A BLAST search of completed archaeal genomes for homologues of M. voltae FlaK revealed homologues in P. abyssi, A. fulgidus, M. barkeri, M. mazei, and M. maripaludis. Interestingly, two homologues were identified in M. jannaschii and three in M. acetivorans. An alignment of these putative FlaK homologues (ClustalW) revealed that while all possess the first aspartic acid residue (D18), thought to be involved in signal peptide processing, the second homologue from M. jannaschii (Mj1282.1) lacks the second aspartic acid (D79). Most of the FlaK homologues have not yet been tested to determine if they are biochemically active, or if the multiple homologues in M. jannaschii or M. acetivorans are able to substitute for one another. A FlaK homologue (PibD) has been identified in Sulfolobus solfataricus, and found to have additional substrates, aside from preflagellins. PibD is also responsible for processing the signal peptide of sugar-binding proteins prior to their insertion into the cell membrane. Additional substrates have been proposed but have not yet been biochemically confirmed [Albers et al., 2003]. While the preflagellin peptidases do have multiple substrates, it is likely that in autotrophs such as certain methanogens, the only substrate will be the preflagellins.

Model

The mechanism of assembly of archaeal flagella has not yet been demonstrated. However, with the body of evidence that is accumulating, it is becoming more apparent that the assembly of archaeal flagella is a different process than the assembly of bacterial flagella. It is entirely possible that a completely novel mechanism of assembly will be revealed, but at this point it appears to be similar to the mechanism of assembly used by type IV pili. The number of genes involved in both bacterial flagellation and the formation of type IV pili (in Pseudomonas aeruginosa) is much greater than the maximum of 13 genes (in M. voltae) that have been identified as likely playing a role in archaeal flagellation [Macnab, 1999; McBride, 2001]. The low number of identified genes, along with the identification of genes involved in flagellation that are not clustered with the flagellin genes (such as flaK in M. voltae) opens the possibility that there may be many more genes in archaeal flagellation that have not yet been discovered and cannot be identified by homology searches to the bacterial counterparts.

Bardy/Ng/Jarrell

In our proposed model of archaeal flagellation, the preflagellins are bound by chaperones in the cytoplasm to prevent N-terminal aggregation [Polosina et al., 1998]. Upon insertion of the preflagellin into the cytoplasmic membrane, at least three events occur. The signal peptide of the preflagellin is removed by the preflagellin peptidase, post-translational glycosylation occurs, probably by enzymes on the outer surface of the cytoplasmic membrane [Sumper, 1987], and the mature protein is incorporated at the base of the flagellar filament. The exact order of these events is unknown, but mutational studies have revealed that the post-translational modification of the flagellins is not dependent on the removal of the signal peptide [Bardy and Jarrell, 2003], as would be indicated by a reduction in the molecular mass of the flagellins by approximately 8–10 kD. Additionally, neither the removal of the signal peptide nor the post-translational modification of the flagellins is dependent on the presence of FlaHIJ, which are thought to be essential for the assembly of the flagellar filament. The role that the additional flagellum-associated proteins have in flagellation is unclear. It is possible that they are part of the structure that currently cannot be isolated, or perhaps they are only temporarily involved in flagella biosynthesis, and act for example, as scaffolding proteins [Thomas and Jarrell, 2001]. Current research suggests that the flagellins that compose the majority of the archaeal flagellar filament are assembled prior to the addition of the cell proximal flagellins, which contradicts the order of assembly of the bacterial flagella. This is supported by flagellin mutant studies and recent work examining the structure of the archaeal flagellar filament, both of which were done in H. salinarum. Examination of the flagellar filament revealed that a central channel, or even a central region of lower mass density, could not be detected [Cohen-Krausz and Trachtenberg, 2002]. The lack of a central channel would preclude any possibility that archaeal flagella are assembled in the same manner as bacterial flagella, one which is absolutely dependent on the presence of a hollow flagellar structure to allow passage of subunits to their incorporation site at the distal tip of the structure. While this model is incomplete, it allows for speculation as to additional requirements for successful flagellation and motility that will likely be discovered as research progresses. This includes the identification of an anchoring structure and with that, a reversible rotary motor. It has been shown that while archaeal flagella do rotate in both directions, they do not undergo the bacterial style swim/tumble. Instead, it has been found in H. salinarum

that the reversing of rotational direction results in a push/ pull effect [Alam and Oesterhelt, 1984]. With a rotational motor comes the need for both a rotor and a stator. In bacterial flagella, the stator is bound to the peptidoglycan in the cell wall. It is unclear which part of the archaeal cell wall the stator would be bound to, but given the varied nature of the archaeal cell wall [Kandler and Konig, 1985] this does raise some questions about the consistency of this structure. It has been suggested that the flagella may extend through the cytoplasmic membrane into a cytoplasmic structure, such as a polar cap, which would serve to stabilize the rotor [Alam and Oesterhelt, 1984; Kupper et al., 1994]. The bacterial chemotaxis system appears to be a unifying component between bacterial flagella, type IV pili and archaeal flagella [Bardy et al., 2003]. It is known that CheY interacts with the switch protein (FliM) to control flagellar rotation in response to attractants and repellents [Bourret et al., 2002]. Homologous chemotaxis systems have been found in P. aeruginosa and Myxococcus xanthus that are involved in the regulation of pili production and twitching motility [Darzins and Russell, 1997; Mattick, 2002]. While the exact mechanism is unknown, it is speculated that the CheY homologue interacts with either PilT or PilU of the type IV pilus system, as these proteins are responsible for providing the required energy for retraction [Wall and Kaiser, 1999]. Intriguingly, a homologous chemotaxis system has also been found in some archaea [Faguy and Jarrell, 1999; Rudolph and Oesterhelt, 1996]. It is currently unknown what protein component the archaeal CheY homologue interacts with to control flagellar rotation. Archaea lack a switch protein homologue, but this interaction could be a target for future research.

Archaeal Flagella

J Mol Microbiol Biotechnol 2004;7:41–51

Future Directions

Thus far only the genes surrounding the major structural proteins (flagellins) have been identified as being involved in flagellation. These genes are often co-transcribed with the flagellins. The functions of the flagellumassociated genes are unknown, and only flaI and flaJ have known homologues in the databases. The sole exception is flaK encoding the preflagellin peptidase that is required for the processing of the signal peptide from the preflagellins. While found as part of the fla gene cluster in M. jannaschii, it is located separately from the fla genes in M. voltae and M. maripaludis.

49

Since genes involved in flagellation cannot be identified by homology searches, different approaches must be used to identify so far undetected genes. One approach is to perform a random mutagenesis and look for nonmotile transformants. Since transposon mutagenesis is not yet a routine methodology in archaea, the approach taken in M. voltae is to clone random fragments of the genome into a mutagenic nonreplicating vector. These random fragments are generated using the restriction enzyme Tsp509I which recognizes the 4-bp sequence AATT leaving the ends compatible with the EcoRI restriction site in the vector. Since the %G+C of the M. voltae is approximately 30%, the ATTA sequence is found on average every 50 bp. Partial digests followed by a size screening for 400to 600-bp fragments leads to a library of plasmids containing random fragments of the genome. This has been used to transform wild-type cells and led to the isolation of several nonmotile, puromoycin-resistant transformants that are currently under study. Another approach to the identification of additional genes involved in flagellation is to improve the isolation techniques so that more of the anchoring structure is preserved. Reverse genetics can then be used to identify the genes of the extra proteins associated with the more complete structure. The isolation procedure is followed by EM to assess the appearance of the structure. Furthermore, since most archaeal flagellins are modified, likely by glycosylation, there must be numerous genes involved in making the glycosyl units and their attachment to the flagellin. Recent work on the M. voltae flagellins has indicated up to about 8–10 kD of glycosyla-

tion may be present as N-linked material [S. Voisin, S. Logan, J. Kelly, S. Bardy and K. Jarrell, unpubl. data]. This accounts for most of the discrepancy initially reported between the predicted flagellin molecular weights obtained by SDS-PAGE, and those predicted from the flagellin gene sequence. Once the structure of the glycosylation modification is known, it should be possible to identify from the complete genome the enzymes needed to make the units. Mutations in such genes should confirm their presumed role in the flagellin process, assuming they are not essential to the cell for other purposes (such as vital to S-layer modification and assembly).

Concluding Thoughts

While much work remains to be done on the study of archaeal flagella, both at the structural and genetic level, recent developments have been made in the identification and understanding of the preflagellin peptidase, and the multi-flagellin nature of the flagellar filament. While a complete model of the archaeal flagellum and the nature of its assembly remain elusive, it will be interesting to follow the research on this motility system. Given the similarities between archaeal flagella and type IV pili, it remains to be seen whether the archaeal flagellum is solely a unique archaeal motility system. It may develop into a model system for protein export in archaea or perhaps even in certain archaea such as Sulfolobus a model for sugar uptake.

References Aizawa SI: Flagella; in Lederberg J (ed): Encyclopedia of Microbiology. New York, Academic Press, 2000, pp 380–389. Alam M, Oesterhelt D: Morphology, function and isolation of halobacterial flagella. J Mol Biol 1984;176:459–475. Albers S-V, Szabo Z, Driessen AJM: Archaeal homolog of bacterial type IV prepilin signal peptidases with broad substrate specificity. J Bacteriol 2003;185:3918–3925. Bardy SL, Jarrell KF: FlaK of the archaeon Methanococcus maripaludis possesses preflagellin peptidase activity. FEMS Microbiol Lett 2002; 208:53–59. Bardy SL, Jarrell KF: Cleavage of preflagellins by an aspartic acid signal peptidase is essential for flagellation in the archaeon Methanococcus voltae. Mol Microbiol 2003;50:1339–1347.

50

Bardy SL, Mori T, Komoriya K, Aizawa S-I, Jarrell KF: Identification and localization of flagellins FlaA and FlaB3 within the flagella of Methanococcus voltae. J Bacteriol 2002;184:5223– 5233. Bardy SL, Ng SYM, Jarrell KF: Prokaryotic motility structures. Microbiology 2003;149:295– 304. Bayley DP, Jarrell KF: Further evidence to suggest that archaeal flagella are related to bacterial type IV pili. J Mol Evol 1998;46:370–373. Bourret RB, Charon NW, Stock AM, West AH: Bright lights, abundant operons – Fluorescence and genomic technologies advance studies of bacterial locomotion and signal transduction: Review of the BLAST meeting, Cuernavaca, Mexico, January 2001. J Bacteriol 2002;184:1– 17.

J Mol Microbiol Biotechnol 2004;7:41–51

Cohen-Krausz S, Trachtenberg S: The structure of the archaebacterial flagellar filament of the extreme halophile Halobacterium salinarum R1M1 and its relation to eubacterial flagellar filaments and type IV pili. J Mol Biol 2002; 321:383–395. Correia JD, Jarrell KF: Posttranslational processing of Methanococcus voltae preflagellin by preflagellin peptidases of M. voltae and other methanogens. J Bacteriol 2000;182:855–858. Cserzo M, Wallin E, Simon I, Von Heijne G, Elofsson A: Prediction of transmembrane alphahelices in prokaryotic membrane proteins: The dense alignment surface method. Protein Eng 1997;10:673–676. Darzins A, Russell MA: Molecular genetic analysis of type-4 pilus biogenesis and twitching motility using Pseudomonas aeruginosa as a model system – a review. Gene 1997;192:109–115.

Bardy/Ng/Jarrell

Dasgupta N, Arora SK, Ramphal R: fleN, a gene that regulates flagellar number in Pseudomonas aeruginosa. J Bacteriol 2000;182:357–364. Deppenmeier U, Johann A, Hartsch T, Merkl R, Schmitz R, Martinez-Arias R, Henne A, Wiezer A, Bäumer S, Jacobi C, Brüggemann H, Lienard T, Christmann A, Bömeke M, Steckel S, Bhattacharyya A, Lykidis A, Overbeek R, Klenk H, Gunsalus R, Fritz H, Gottschalk G: The genome of Methanosarcina mazei: Evidence for lateral gene transfer between bacteria and archaea. J Mol Microbiol Biotechnol 2002; 4:453–461. Doig P, Kinsella N, Guerry P, Trust TJ: Characterization of a posttranslational modification of Campylobacter flagellin: Identification of a sero-specific glycosyl moiety. Mol Microbiol 1996;19:379–387. Faguy DM, Jarrell KF: A twisted tale: The origin and evolution of motility and chemotaxis in prokaryotes. Microbiology 1999;145:279–281. Faguy DM, Jarrell KF, Kuzio J, Kalmokoff ML: Molecular analysis of archaeal flagellins: Similarity to the type IV pilin-transport superfamily widespread in bacteria. Can J Microbiol 1994a;40:67–71. Faguy DM, Koval SF, Jarrell KF: Physical characterization of the flagella and flagellins from Methanospirillum hungatei. J Bacteriol 1994b;176:7491–7498. Fedorov OV, Pyatibratov MG, Kostyukova AS, Osina NK, Tarasov VY: Protofilament as a structural element of flagella of haloalkalophilic archaebacteria. Can J Microbiol 1994; 40:45–53. Fitz-Gibbon ST, Ladner H, Kim UJ, Stetter KO, Simon MI, Miller JH: Genome sequence of the hyperthermophilic crenarchaeon Pyrobaculum aerophilum. Proc Natl Acad Sci USA 2002;99: 984–989. Galagan JE, Nusbaum C, Roy A, Endrizzi MG, Macdonald P, FitzHugh W, Calvo S, Engels R, Smirnov S, Atnoor D, Brown A, Allen N, Naylor J, Stange-Thomann N, DeArellano K, Johnson R, Linton L, McEwan P, McKernan K, Talamas J, Tirrell A, Ye W, Zimmer A, Barber RD, Cann I, Graham DE, Grahame DA, Guss AM, Hedderich R, Ingram-Smith C, Kuettner HC, Krzycki JA, Leigh JA, Li W, Liu J, Mukhopadhyay B, Reeve JN, Smith K, Springer TA, Umayam LA, White O, White RH, Conway de Macario E, Ferry JG, Jarrell KF, Jing H, Macario AJ, Paulsen I, Pritchett M, Sowers KR, Swanson RV, Zinder SH, Lander E, Metcalf WW, Birren B: The genome of M. acetivorans reveals extensive metabolic and physiological diversity. Genome Res 2002;12:532–542. Gerl L, Sumper M: Halobacterial flagellins are encoded by a multigene family. J Biol Chem 1988;263:13246–13251. Huber R, Stetter KO: Genus III: Pyrobaculum; in Garrity M (ed): Bergey’s Manual of Systematic Bacteriology. The Archaea and the Deeply Branching Phototrophic Bacteria. New York, Springer-Verlag, 2001, vol 1, pp 175–177.

Archaeal Flagella

Jarrell KF, Bayley DP, Florian V, Klein A: Isolation and characterization of insertional mutations in flagellin genes in the archaeon Methanococcus voltae. Mol Microbiol 1996a;20:657– 666. Jarrell KF, Bayley DP, Kostyukova AS: The archaeal flagellum: A unique motility structure. J Bacteriol 1996b;178:5057–5064. Kachlany S, Planet P, Bhattacharjee M, DeSalle R, Fine D, Figurski D: Nonspecific adherence by Actinobacillus actinomycetemcomitans requires genes widespread in bacteria and archaea. J Bacteriol 2000;182:6169–6176. Kachlany S, Planet P, DeSalle R, Fine D, Figurski D: Genes for tight adherence of Actinobacillus actinomycetemcomitans: From plaque to plague to pond scum. Trends Microbiol 2001; 9:429–437. Kalmokoff ML, Jarrell KF, Koval SF: Isolation of flagella from the archaebacterium Methanococcus voltae by phase separation with Triton X114. J Bacteriol 1988;170:1752–1758. Kandler O, Konig H: Cell envelopes of Archaebacteria; in Woese CR, Wolfe RS (ed): Archaebacteria. New York, Academic Press, 1985, vol 8, pp 413–457. Kupper J, Marwan W, Typke D, Grunberg H, Uwer U, Gluch M, Oesterhelt D: The flagellar bundle of Halobacterium salinarium is inserted into a distinct polar cap structure. J Bacteriol 1994;176:5184–5187. Kurr M, Huber R, König H, Jannasch HW, Fricke H, Trincone A, Kristjansson JK, Stetter KO: Methanopyrus kandleri, gen. and sp. nov. represents a novel group of hyperthermophilic methanogens, growing at 110 ° C. Arch Microbiol 1991;156:239–247. LaPointe CF, Taylor RK: The type 4 prepilin peptidases comprise a novel family of aspartic acid proteases. J Biol Chem 2000;275:1502–1510. McBride MJ: Bacterial gliding motility: Multiple mechanisms for cell movement over surfaces. Annu Rev Microbiol 2001;55:49–75. Macnab RM: The bacterial flagellum: Reversible rotary propellor and type III export apparatus. J Bacteriol 1999;181:7149–7153. Mattick JS: Type IV pili and twitching motility. Annu Rev Microbiol 2002;56:289–314. Mattick JS, Alm RA: Common architecture of type 4 fimbriae and complexes involved in macromolecular traffic. Trends Microbiol 1995;3: 411–413. Moens S, Michiels K, van der Leyden J: Glycosylation of the flagellin of the polar flagellum of Azospirillum brasilense, a Gram-negative nitrogen fixing bacterium. Microbiology 1995;141: 2651–2657. Patenge N, Berendes A, Englehardt H, Schuster SC, Oesterhelt D: The fla gene cluster is involved in the biogenesis of flagella of Halobacterium salinarum. Mol Microbiol 2001;41:653–663.

Polosina YY, Jarrell KF, Fedorov OV, Kostyukova AS: Nucleoside diphosphate kinase from haloalkaliphilic archaeon Natronobacterium magadii. Extremophiles 1998;2:333–338. Pyatibratov MG, Leonard K, Tarasov VY, Fedorov OV: Two immunologically distinct types of protofilaments can be identified in Natrialba magadii flagella. FEMS Microbiol Lett 2002; 212:23–27. Rudolph J, Oesterhelt D: Deletion analysis of the che operon in the archaeon Halobacterium salinarium. J Mol Biol 1996;258:548–554. Strom MS, Nunn DN, Lory S: Posttranslational processing of type IV prepilin and homologs by PilD of Pseudomonas aeruginosa. Methods Enzymol 1994;235:527–540. Strom MS, Lory S: Amino acid substitutions in pilin of Pseudomonas aeruginosa. J Biol Chem 1991;266:1656–1664. Sumper M: Halobacterial glycoprotein biosynthesis. Biochim Biophys Acta 1987;906:69–79. Tarasov VY, Pyatibratov MG, Tang S, DyallSmith M, Fedorov OV: Role of flagellins from A and B loci in flagella formation of Halobacterium salinarum. Mol Microbiol 2000;35:69– 78. Thomas NA, Bardy SL, Jarrell KF: The archaeal flagellum: A different kind of prokaryotic motility structure. FEMS Microbiol Rev 2001a;25:147–174. Thomas NA, Chao ED, Jarrell KF: Identification of amino acids in the leader peptide of Methanococcus voltae preflagellin that are important in posttranslational processing. Arch Microbiol 2001c;175:263–269. Thomas NA, Jarrell KF: Characterization of flagellum gene families of methanogenic archaea and localization of novel flagellum accessory proteins. J Bacteriol 2001;183:7154–7164. Thomas NA, Mueller S, Klein A, Jarrell KF: Mutants in flaI and flaJ of the archaeon Methanococcus voltae are deficient in flagellum assembly. Mol Microbiol 2002;46:879–887. Thomas NA, Pawson CT, Jarrell KF: Insertional inactivation of the flaH gene of the Archaeon Methanococcus voltae results in nonflagellated cells. Mol Gen Genomics 2001b;265:596–603. Vignon G, Köhler R, Larquet E, Giroux S, Prévost M, Roux P, Pugsley A: Type IV pili formed by the type II secreton: Specificity, composition, bundling, polar localization, and surface presentation of peptides. J Bacteriol 2003;185: 3416–3428. Wall D, Kaiser D: Type IV pili and cell motility. Mol Microbiol 1999;32:1–10. Wilson DR, Beveridge TJ: Bacterial flagellar filaments and their component flagellins. Can J Microbiol 1993;39:451–472. Winstanley C, Morgan JA, Pickup RP, Saunders JR: Molecular cloning of two Pseudomonas flagellin genes and basal body structural genes. Microbiology 1994;140:2019–2031. Woese CR, Fox GE: Phylogenetic structure of the procaryotic domain: Primary kingdoms. Proc Natl Acad Sci USA 1977;74:5088–5090.

J Mol Microbiol Biotechnol 2004;7:41–51

51

J Mol Microbiol Biotechnol 2004;7:52–62 DOI: 10.1159/000077869

Pulling Together with Type IV Pili Eric Nudleman Dale Kaiser Stanford University, Departments of Biochemistry and of Developmental Biology, Stanford, Calif., USA

Key Words Social motility W Twitching motility W Biofilms W PilT W PilQ W Secretins W Fibrils

Abstract Type IV pili are an efficient and versatile device for bacterial surface motility. They are widespread among the ß-, Á-, and ‰-proteobacteria and the cyanobacteria. Within that diversity, there is a core of conserved proteins that includes the pilin (PilA), the motors PilB and PilT, and various components of pilus biogenesis and assembly, PilC, PilD, PilM, PilN, PilO, PilP, and PilQ. Progress has been made in understanding the motor and the secretory functions. PilT is a motor protein that catalyzes pilus retraction; PilB may play a similar role in pilus extension. Type IV pili are multifunctional complexes that can act as bacterial virulence factors because pilus-based motility is used to spread pathogens over the surface of a tissue, or to build multicellular structures such as biofilms and fruiting bodies. Copyright © 2004 S. Karger AG, Basel

ABC Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

© 2004 S. Karger AG, Basel 1464–1801/04/0072–0052$21.00/0 Accessible online at: www.karger.com/mmb

Where Are Type IV Pili Found?

Henrichsen [1972] was the first to observe the association between surface-dependent motility and type IV pili (tfp). Henrichsen collected and studied the movement of bacteria capable of spreading over surfaces, typically the surface of moist agar. He observed that some of the organisms had flagella and could swim in suspension, but could also swarm over the surface. Other bacteria, lacking flagella, were nevertheless capable of rapid swarm-spreading by motility that was described as ‘gliding’ or ‘twitching’. These organisms formed spreading zones at the colony edge that frequently were one, or at most a few layers of cells – so thin as to be barely visible to the naked eye. Henrichsen also observed the correlation between twitching among Moraxella strains and Acinetobacter strains and the presence of polar pili, or fimbriae recognized by electron microscopy [Henrichsen et al., 1972; Henrichsen and Blom, 1975]. Pseudomonas aeruginosa was shown to have tfp by Bradley [1973, 1974], Darzins [1994] and Mattick et al. [1996]. MacRae and McCurdy [1976] discovered a correlation between pili and gliding motility among myxobacteria in the 1970s. These and later investigations have enlarged Henrichsen’s list of strains possessing tfp to those shown in figure 1. The list is based on a

Dale Kaiser Department of Biochemistry and Developmental Biology Stanford University School of Medicine Stanford, CA 94305 (USA) Tel. +1 650 723 6165, Fax +1 650 725 7739, E-Mail [email protected]

Aeromonas hydrophila Vibrio cholerae Shewanella putrefaciens Pasteurella multocida Legionella pneumophila Escherichia coli Suttonella indologenes gamma proteobacteria

Dichelobacter nodosus Pseudomonas syringae Pseudomonas stutzeri Pseudomonas putida Pseudomonas aeruginosa Moraxella nonliquefaciens Moraxella lacunata Moraxella catarrhalis Azoarcus spp. Ralstonia solanacearum

Proteobacteria beta proteobacteria

Neisseria meningitidis Neisseria gonorrhoeae Kingella kingae Kingella denitrificans Eikenella corrodens Bacteroides ureolyticus

Bacteria

Fig. 1. Phylogeny of tfp. Organisms that

have the conserved core of tfp genes are shown classified by their taxonomic lineage. The tree is simplified to show the subdivisions of the Proteobacteria that are common to these organisms, but no further branches. The tree was generated using the NCBI Taxonomic classifications and the TreeView Software [Page, 1996].

Myxococcus xanthus delta/epsilon proteobacteria

Myxococcus virescens Myxococcus fulvus Wolinella spp.

alpha proteobacteria Cyanobacteria

No species with tfp Synechocystis sp. PCC 6803

twitching motility phenotype, the presence of polar pili, or the presence of genes encoding tfp in the sequenced genome, as suggested by Mattick [2002]. A wide distribution that extends across the ß-, Á-, and ‰-proteobacteria and the cyanobacteria might be taken to imply that the common ancestor of these groups had tfp (fig. 1). If so, tfp would have been expected among the ·proteobacteria, but they have cpa pili instead [Skerker and Shapiro, 2000]. Rather than postulating common ancestory, the high degree of structural similarity across organisms may instead indicate that groups of tfp genes have spread across these diverse genera by lateral gene transfer. Lateral transfer might also account for the finding that groups of tfp genes are often clustered, or reside

on plasmids [Giron et al., 1991; Gophna et al., 2003; Stone et al., 1996; Wall and Kaiser, 1999]. Significantly, many of the piliated strains are important pathogens of animals and plants. Type IV pili endow bacteria with social motility, and cooperating cells might have an advantage in overcoming the barriers to infection erected by a potential host, or be able to move into better environmental conditions.

Pulling Together with Type IV Pili

J Mol Microbiol Biotechnol 2004;7:52–62

Fluid Mechanics and Tiny Organisms The movement of tiny organisms requires different physical mechanisms from the swimming of large organisms. For instance, a fish can swim by imparting rearward momentum to the water around it [Vogel, 1994]. But bac-

53

teria are too small to use momentum, rather they must deal with viscosity, drag, and the reversibility of fluid flow at low Reynolds numbers [Purcell, 1977]. Pulling the narrow, streamlined front end of a cylindrical bacterium forward is a good way to translate force into movement across a surface that is covered with a viscous fluid.

Retraction

David Bradley obtained the first evidence that tfp were engines of movement. He observed that the P. aeruginosa bacteriophage PP7 attached initially to the distal end of a pilus, and that the average pilus length was significantly reduced after incubation with phage [Bradley, 1972]. From these and related data, Bradley hypothesized that the pili retract, pulling the phage to the cell surface. Bradley also reported that both the nonpiliated and the hyperpiliated mutants lacked twitching motility, and he concluded that retractile pili were the mechanical basis for twitching motility [Bradley, 1980]. Mattick [2002] extensively reviews Bradley’s historic experiments. Direct evidence that tfp forcefully retract has been reported recently. Using an optical trap on Neisseria gonorrhoeae tfp, Merz et al. [2000] measured the velocity, timing, and force of retraction. Single diplococci were immobilized on a bead attached to a coverslip. The tfp were pulled by manipulating a bead coated with anti-pilin monoclonal antibody that was held in a laser trap. When the cell retracted tfp that were bound to the bead, the bead was pulled from the laser trap with a force greater than 80 pN. These tfp did not retract continuously; spates of retraction were separated by intervals of 1–20 s. A retraction velocity of 1.17 B 0.49 Ìm s –1 was recorded, which was independent of the length of a given pilus. This value correlates well with the rate at which cells crawled on coverslips and also pulled out of the laser trap towards microcolonies (F1 Ìm s –1), suggesting that the measured force of retraction is responsible for the cell movement observed. An objection was raised to the measurements of Merz et al. [2000] in that several tfp from the same cell may have been attached to a single bead. If more than one pilus had retracted, the force would not apply to a single tfp. To address this uncertainty, Maier et al. [2002] used an inducible promoter to express low levels of pilin in order that 80% of the cells would have no tfp, and the rest would either have a single pilus or several well-separated tfp. Force measurements under these new conditions agreed with the previously reported values; the stall forces aver-

54

J Mol Microbiol Biotechnol 2004;7:52–62

aged 110 B 30 pN. The initial velocity of pilus retraction was also confirmed as 1.2 B 0.2 Ìm s –1. The diameter of a single tfp being much less than the wavelength of visible light, they are very difficult to see, except by electron microscopy. However, cells labeled with a fluorescent dye that binds amino groups rendered those tfp in live P. aeruginosa cells visible with total internal reflection microscopy [Skerker and Berg, 2001]. As had been observed in N. gonorrhoea a tfp of P. aeruginosa would retract, pause, then retract further. When the distal tip of the pilus had adhered to the substratum, the cell body was pulled forward by the retraction. The speed of pilus retraction (F0.5 Ìm s –1) agreed with the velocity of cell movement (0.31 B 0.21 Ìm s –1). Following retraction, some tfp re-extended, now with labeled pilin, suggesting that fluorescently labeled monomers released by retraction had been stored in the membrane, then reassembled into a new pilus. Extension was not associated with cell movement, suggesting that, although tfp can pull, they cannot push. Observations consistent with tfp retraction have been made in Myxococcus xanthus. Piliated cells were seen to adhere end-on to a polystyrene surface, then to ‘jiggle’, which was taken to indicate pilus retraction [Sun et al., 2000]. Wild-type cells from which the tfp had been sheared, and mutants that lacked tfp were unable to attach. A pilT mutant adhered, but failed to jiggle. Cells that had jiggled were then occasionally observed to lie flat on the surface and to glide short distances [Sun et al., 2000].

The tfp Apparatus

Among the tfp proteins common to M. xanthus, P. aeruginosa, N. gonorrhoeae, and Synechocystis PCC6803, the most highly conserved are PilA, PilB, PilC, PilD, PilM, PilN, PilO, PilP, PilQ, and PilT, using the gene designation terminology of M. xanthus and P. aeruginosa (table 1). Protein localization is also conserved. In M. xanthus, this entire gene set is in a single cluster [Wall and Kaiser, 1999]. PilA, the Pilin Type IV pilins, which are encoded by the pilA gene, share a conserved amino terminal region of about 60 amino acids [reviewed in Strom and Lory, 1993; Wu and Kaiser, 1995]. All pilins are synthesized as prepilins, and the prepilins are processed by PilD, a peptidase. Satisfying the substrate specificity of PilD would partly account

Nudleman/Kaiser

Table 1. Core tfp genes

Protein

Function

Cellular localization

Reference

PilA PilB PilT PilC PilD PilM PilN PilO PilP PilQ

Pilin: monomer of the tfp filament Pilus extension Pilus retraction Unknown PilA leader peptidase ATPase, unknown Unknown Unknown PilQ stability in N. gonorrhoeae Secretin

Inner membrane and pilus fiber Periplasm/inner membrane Cytoplasm/inner membrane Inner membrane Inner membrane Inner membrane Periplasm Periplasm Anchored in outer membrane Outer membrane

see text see text see text Nunn et al. [1990], Wu et al. [1997] see text Martin et al. [1995] Martin et al. [1995] Martin et al. [1995] Drake et al. [1997], Martin et al. [1995] see text

for the conservation of the pilin sequence around the cleavage site. A second constraint on the amino acid sequence of the pilin amino terminus is that it must form an ·-helix that is capable of staggered coiling with several copies of itself. Structure of the tfp Fiber The structure of a pilus fiber is important for its mechanical strength and flexibility, for pilus assembly, and for pilus retraction. A tfp is a fibrous repeating polymer made of many thousands of copies of the processed pilin, encoded by pilA. The fiber is a layered structure of ·-helices surrounded by ß-strands illustrated in figure 2. The N-terminal amino acids of adjacent monomers (after processing of the prepilin), shown in blue in figure 2A, form an ·-helical coiled coil [Parge et al., 1995]. The parallel, staggered ·-helices coil around and make hydrophobic bonds with one another. In N. gonorrhoeae, this solid inner layer of the fiber is then covered with the Cterminal regions of adjacent monomers that form a scaffold of ß-strands; individual ß-strands are shown as green ribbons in figure 2A. A 2.6-angström resolution X-ray crystal structure of N. gonorrhoeae pilin dimers was the first to be obtained, and the dimer interactions were used to infer the structure of the fiber [Parge et al., 1995]. The fiber model was confirmed with anti-peptide antibodies that distinguished between regions of pilin that were buried and regions that were exposed in the assembled fiber [Forest et al., 1996; Forest and Tainer, 1997]. The fiber of N. gonorrhoeae was predicted to have 5 pilin monomers per helical turn, a rise of about 4 nm per monomer, and an outer diameter of about 6 nm. A space-filling model of the assembled structure is shown in figure 2B. Helix parameters and helix diameter of N. gonorrhoeae also agree with

Pulling Together with Type IV Pili

fiber and crystal diffraction of P. aeruginosa PAK tfp [Craig et al., 2003; Folkhard et al., 1981; Hazes et al., 2000]. Toxin-coregulated pili of Vibrio cholerae (TCP), and the bundle forming pili of enteropathogenic Escherichia coli (EHEC) are both type IVb. They have a different structure that defines a subtype different from the type IVa pili of N. gonorrhoeae and P. aeruginosa. Nevertheless, TCP and EHEC tfp share with the other tfp an assembly pathway, a core gene set, a pilin C-terminal disulfide bond, and sequence similarity in the N-terminal 60 residues of pilin, but have little or no other similarity in the C-terminal domains of pilin [Giron, 1997; Strom and Lory, 1993]. X-ray crystallography and cryo-electron microscopy of TCP show 5 ß-strands (one more than the type IVa pilin) and a new overall protein fold. Nevertheless, as shown by comparing figure 2C with 2A and 2B, TCP have a structural scaffold that consists of ·-helices and ß-sheets [Craig et al., 2003]. The difference from the type IVa pili lies in the folded arrangement of the ßsheets. Perhaps the hydrophobic bonding and the flexibility of ·-helices in all tfp allow them to bend, twist, or bundle with other pili. In all tfp, the ß-sheets of one pilin monomer interact with the ß-strands of the next, as shown in figure 2C, to give mechanical strength to the fiber. Despite its narrow diameter of 6 nm, the fiber can withstand tension stresses of more than 100 pN. The absence of a channel in the center of that narrow fiber implies that the pilus cannot be assembled from the tip like a flagellum, but rather must be assembled from its base. Also, the base is inferred to be the locus of disassembly, because of the location of PilT in the inner membrane (table 1).

J Mol Microbiol Biotechnol 2004;7:52–62

55

Fig. 2. Structure of pilin in the tfp fiber. A The secondary structure

of the pilin monomer from N. gonorrhoeae, type IVa. Reprinted from Parge et al. [1995], with permission. The highly conserved N-terminus (blue) is part of the coiled-coil core that is tucked inside the pilus fiber. Also shown are the four major ß-strands (green), the conserved disulphide region (yellow), and the sugar loop (red) of each monomer. B Space-filling model of five, helically staggered monomers, as they are arranged in the pilus fiber. Each monomer has the structure shown in A. [From Parge et al., 1995]. Successive monomers are shown in green, purple, red, yellow, and blue. While the N-terminal

PilD PilD is a leader peptidase that recognizes an N-terminal pre-sequence of PilA, which is distinct from the sequences recognized by signal peptidases I and II [Nunn and Lory, 1991]. PilD is identical with XcpA, the leader peptidase used in the main terminal branch of the type II general secretory pathway [reviewed by Pugsley et al., 1997]. PilD proteolytically removes the leader sequence from several other proteins of the general secretory pathway, which for that reason are called pseudopilins. PilD is a bifunctional enzyme that also methylates the newly created N-terminal amino acid of the pilin or pseudopilin. In vitro and mutational studies in P. aeruginosa have shown that the two activities of this integral membrane protein are at two adjacent active sites in the enzyme protein [Pepe and Lory, 1998].

56

J Mol Microbiol Biotechnol 2004;7:52–62

·-helices of the 5 monomers of one helical turn wrap around each other, the surface of the fiber is made up entirely of residues from the carboxyl domains of the monomers. C Structure-based model of TCP filament from V. cholerae, typeIVb [Craig et al., 2003]. The model illustrates the packing of the N-terminal ·-helices (red/orange) that form the core. The model also shows the interactions between the ß-sheets of adjacent monomers in the filament (blue). The image was generated from PDB code 1OR9 using VRML 2.0 software [Suhnel, 1996].

PilT and PilB, Complementary Motor Proteins The hyperpiliated mutants that Bradley found to be phage resistant and to lack twitching motility were found to have internal deletions in the pilT gene [Whitchurch et al., 1990a]. Mutations in the pilT homologues of P. aeruginosa, M. xanthus and Synechocystis PCC6803 were also found to be hyperpiliated, nonmotile, and unable to retract their tfp [Bhaya et al., 2000; Skerker and Berg, 2001; Sun et al., 2000; Whitchurch et al., 1990b; Wu et al., 1997]. The pilT gene is the most highly conserved of those required for tfp [Wall and Kaiser, 1999]. PilT has an NTP-binding cassette, or ‘Walker box’, that is necessary for retraction [Herdendorf et al., 2002; Wu et al., 1997]. PilT in M. xanthus has significant sequence homology to PilB (32% identity, 50% similarity). Mutations in pilB destroy the cells’ capacity to assemble tfp. The two proteins appear to have opposing roles:

Nudleman/Kaiser

PilQ, a Secretin Tfp cross the outer membrane through a large oligomeric channel made of a single protein [Bitter et al., 1998; Collins et al., 2001; Drake and Koomey, 1995; Liu et al.,

2001; Martin et al., 1995; Schmidt et al., 2001; Wall et al., 1999; Yoshihara et al., 2001]. The PilQ of Neisseria spp., Pseudomonas spp., M. xanthus and Synechocystis PCC6803 are members of the large secretin family, proteins that form multimeric pores in the outer membranes of gram-negative bacteria [Genin and Boucher, 1994]. Secretins facilitate the passage of folded proteins, filamentous phage particles, DNA, and other macromolecules across the outer membrane [Dubnau, 1999; Genin and Boucher, 1994; Linderoth et al., 1997]. PilQ is essential for tfp biogenesis [Drake and Koomey, 1995; Liu et al., 2001; Wall et al., 1999; Yoshihara et al., 2001]. Secretins were first identified as proteins required for secretion that form highly stable complexes, resistant to boiling in detergent [Chen et al., 1996; Hardie et al., 1996a; Kazmierczak et al., 1994; Newhall et al., 1980]. Electron microscopy of purified PilQ multimers from P. aeruginosa, N. meningitidis, and E. coli (EPEC) have revealed ring-shaped structures with 12-fold symmetry [Bitter et al., 1998; Collins et al., 2001; Schmidt et al., 2001]. By electron microscopy, the PilQ pore has an internal diameter of 5–7 nm, which matches the 6-nm diameter of the pilus fiber. Secretins are most highly conserved at their C-termini. This region, embedded in the outer membrane, is predicted to form a ß-barrel composed of ß-strands from adjacent monomers in the complex [Brok et al., 1999; Daefler et al., 1997; Guilvout et al., 1999; Wall et al., 1999]. The C-terminal domain also interacts with specific cognate lipoproteins that are necessary for secretin multimerization [Daefler et al., 1997; Daefler and Russel, 1998]. Secretins form electrochemically gated channels, and purified protein embedded in planar lipid bilayers form voltage-gated, ion-conducting channels [Brok et al., 1999; Nouwen et al., 1999]. As the applied membrane potential is ramped up, the secretin channel conductance increases nonlinearly, suggesting that the conformation of the channel proteins can change with the voltage applied. A recent 2.5-nm resolution structure of the PilQ complex of N. meningitidis revealed a funnel-shaped structure that constricts to a closed point that presumably lies in the periplasm [Collins et al., 2003]. Thus, the channel conductance may reflect the conformational flexibility of the closed tip of the funnel. Structural comparisons of the open and closed conformations might give insight into the gating mechanism, and how the pilus might trigger the channel to open. Small lipoproteins are often required to aid assembly and localization of the secretin to the outer membrane

Pulling Together with Type IV Pili

J Mol Microbiol Biotechnol 2004;7:52–62

PilT is required for pilus retraction, PilB for pilus extension. Synechocystis PCC6803 has 2 pilT genes (pilT1, pilT2) and a pilB gene. Mutants in pilT1 are hyperpiliated; interestingly, mutants in pilT2 exhibit negative phototaxis (wildtype cells are positively phototactic) [Bhaya et al., 2000]. The PilT protein is a (distant) member of the AAA family of motor ATPases that generally form hexameric complexes, and that mediate the unidirectional disassembly of macromolecular complexes [Vale, 2000]. PilT purified from the gliding thermophile Aquifex aeolicus has an ATPase activity of 15.7 U/mg of protein in the presence of Mg2+. The protein formed stable oligomers of 5–6 subunits [Herdendorf et al., 2002]. Purified PilT from Synechocystis PCC6803 had a similar specific ATPase activity of 18.4 U/mg [Okamoto and Ohmori, 2002]. By regulating the level of the PilT protein in cells, Maier et al. have obtained evidence that the PilT motor undergoes multiple cycles of ATP hydrolysis as it advances along the pilus filament [Maier et al., 2002]. Reducing the level of PilT in cells reduced the frequency of retraction events, but the length of the pauses between spurts of retraction was similar in cells with a low level of PilT as in those with normal levels of PilT. Neither the stall force nor the force-velocity dependence was affected by reducing the level of PilT. This suggests that once a (hexameric) complex has formed at the base of a pilus, it can catalyze retraction at the maximum rate. The PilT sequence has no evident transmembrane domains [Herdendorf et al., 2002; Wu et al., 1997], and the protein is thought to localize to the cytoplasmic face of the inner membrane where it would be in a position to encircle the base of the pilus fiber. G. Oster has suggested that the PilT oligomer might deliver its power stroke by a mechanism similar to that of the F1-ATPase [reported in Kaiser, 2000]. Similarity of catalytic sites and a hexameric structure align the F1-ATPase with a predicted PilT/ PilB motor. The PilT oligomer would break the proteinprotein interaction between monomers in the fiber in an ATP-dependent manner, and dissolve the pilus into a monomer pool in the membrane, from which they could be recycled [Herdendorf et al., 2002; Merz et al., 2000; Skerker and Berg, 2001]. Retraction is rapid: an estimated 1,500 pilin monomers are disassembled per second [Merz et al., 2000].

57

[Crago and Koronakis, 1998; Daefler et al., 1997; Daefler and Russel, 1998; Hardie et al., 1996a, b; Koster et al., 1997; Nouwen et al., 1999; Shevchik and Condemine, 1998]. In N. gonorrhoeae, mutations in the pilP lipoprotein reduce assembly of PilQ multimers [Drake et al., 1997]. In E. coli (EPEC), the secretin (BfpB) is itself a lipoprotein [Ramer et al., 1996]. In this case, the assembly of BfpB into a multimer requires the small (14 kD) protein, BfpG [Schmidt et al., 2001]. In M. xanthus, the Tgl lipoprotein is required for assembly of the PilQ secretin [Nudleman et al., unpubl.].

Binding Targets of tfp

Tfp mediated motility usually involves their contact with another cell. Although tfp have been reported to attach to inert surfaces, those reports offer little information about targets. Attachment of tfp to cells has been shown to lead to retraction. However, cells separated by more than a pilus length (several micrometers) rarely move [Kaiser, 1979; Merz et al., 2000; Semmler et al., 1999]. As a result, the formation of Neisseria microcolonies depends on functional tfp [Merz et al., 2000]. Cells with mutations in pilT can tether to other cells, or to inert surfaces, but they are not able to form microcolonies in N. gonorrhoeae or wide rafts of cells in M. xanthus [Merz et al., 2000; Skerker and Berg, 2001; Sun et al., 2000; Whitchurch et al., 1990b; Wu et al., 1997]. The surfaces of human epithelia are usually covered with harmless biofilms that may offer a measure of protection. Harmful biofilms are associated with persistent multicellular infections. Tfp-dependent cell-to-cell adhesions are required for biofilm formation in many bacteria pathogenic for humans [Costerton et al., 1999]. Such biofilms are found on surfaces of the middle ear, urinary tract, bone, and heart valves; biofilms also grow on the abiotic surfaces of implanted medical devices. Investigating the role of tfp in biofilm formation, O’Toole and Kolter [1998] found several different pil– mutants of P. aeruginosa in a screen for biofilm defects. Compared to the wild type, these pil– mutants attached poorly to the polyvinylchloride plastic surface. However, by 8 h, the mutant cells had scattered about on the plastic, but were unable to move or to aggregate, while the wild-type did both [O’Toole and Kolter, 1998]. Heydorn et al. [2002] compared the behavior of biofilms in a flow chamber formed by wild type with a pil– mutant. Their pil– mutants formed discrete, dense microcolonies after 98 h, whereas the wild-type cells formed a confluent, featureless flat

58

J Mol Microbiol Biotechnol 2004;7:52–62

sheet. Chiang and Burrows reported that under static (i.e., no flow) conditions, pilT mutants formed more dense biofilms after 24 h than wild type, while the pil– (pilA) mutants formed less dense biofilms [Chiang and Burrows, 2003]. Under flow conditions, the pilA mutants were unable to adhere, while the pilT mutant adhered and formed a dense mat. However, the pilT biofilm mushroom structures were less dense than the wild type. Twitching motility is thus implicated in forming the dense mushroom-shaped structure of mature P. aeruginosa biofilms, particularly in the development of their mushroom caps [Klausen et al., 2003]. Tfp-based motility of M. xanthus is important for building its multicellular fruiting body that has a species-specific shape [Kaiser, 2003]. Induced to sporulate by starvation, fruiting body cells of M. xanthus must continue to move within the structure in order to signal each other and to reach the signaling threshold adequate for sporulation [Kim and Kaiser, 1990; Sager and Kaiser, 1993a, b]. In sum, tfpbased movement helps to give shape to multicellular structures in biofilms and in fruiting bodies. Adhesion by tfp is often critical for pathogenesis. That tfp bind to tissues to initiate an infection is indicated by the reduced adhesion of bacteria that have lost their tfp [Chi et al., 1991; Collyn et al., 2002; Farinha et al., 1994; Rothbard et al., 1985; Ruehl et al., 1993; Strom and Lory, 1993; Zhang et al., 2000]. In addition to adhesion, several observations suggest that tfp are cytotoxic by virtue of retraction. E. coli (EPEC) cells with mutations in bfpF (the pilT homolog) are avirulent, even though they have tfp and adhere to tissue monolayers [Bieber et al., 1998]. Similarly, pilT mutants of P. aeruginosa and N. meningitidis lack cytotoxicity [Comolli et al., 1999; Pujol et al., 1999]. Retraction might also help the bacteria spread over the surface of a host cell, as proposed by Henrichsen [1983]. Spreading would also facilitate the establishment of a biofilm on the surface of a tissue [Costerton et al., 1999]. Retraction of tfp may bring about an intimate contact between a bacterial pathogen and its host cell, which might suffice to allow type II or type III secretion of toxins into the host. Type II secretion is reviewed by Sandkvist [2001]; type III by Cornelis and Van Gijsegem [2000]. It has been reported by Kirn et al. [2003] that a soluble toxin requires the TCP of V. cholerae to be translocated across the bacterial outer membrane. However, it has not been demonstrated that retraction is required. tfp prefer to adhere by their tip when they bind the surface of quartz [Skerker and Berg, 2001]. The distal tip of the pilus filament is unique. Each pilin monomer at the

Nudleman/Kaiser

tip exposes a region of its surface that otherwise is part of the monomer-monomer interface [Forest and Tainer, 1997]. Carbohydrates have been identified as the cellular targets of pilus adhesion. In P. aeruginosa, the C-terminal disulfide bonded region of 12–17 residues of the monomer is exposed at the tip [Hazes et al., 2000] and is required for binding to the carbohydrate moiety of the glycosphingolipids asialo-GM1 and asialo-GM2 on epithelial cells [Lee et al., 1994; Sheth et al., 1994]. It has been reported that the PilC protein in N. gonorrhoeae, and its PilY1 homolog in P. aeruginosa are required for binding to host cell tissues [Scheuerpflug et al., 1999; Wolfgang et al., 2000]. PilC and PilY1 are found associated with the pilus filament and within the cell membrane. These proteins are required for tfp biogenesis, but pili can be restored to mutants defective in either gene by a mutation in pilT [Wolfgang et al., 1998]. Even though piliation is restored, the suppressed mutants cannot bind host tissues. Wolfgang et al. suggest that PilC may be required to cap or to stabilize the tfp filament. In their absence, the pilus would be retracted. Myxobacterial tfp Prefer to Bind Fibrils The fibrils of M. xanthus make up a linked network of amorphous strands, about 30 nm in diameter, that are often seen to join neighboring cells into a cluster [Behmlander and Dworkin, 1994a; Dworkin, 1999]. Fibrils consist of almost equal amounts of protein and of polysaccharide that contains galactose, glucosamine, glucose, rhamnose, and xylose. Several different fibrillar proteins can be distinguished from each other by their antigens as well as by their electrophoretic mobility [Behmlander and Dworkin, 1994b]. Most of the mutants of M. xanthus that lack social motility, lack tfp; social motility depends on tfp [Kaiser, 1979]. However, three groups of social motility mutants, the pilT mutants, the dsp (dispersed growth) mutants, and certain lipopolysaccharide-defective mutants retain tfp. Unlike the pilT mutants, which grow in clumps, the dsp mutants are less cohesive and grow dispersed in liquid culture [D. Morandi, unpubl. results]. Arnold and Shimkets [1988a, b] and Shimkets [1986] discovered that the dsp mutants lack fibrils, although dsp appears not to encode constituents of the fibrils. A group of chemosensory mutants, called dif, also lack fibrils [Yang et al., 2000]. Bowden and Kaplan [1998] have reported that mutants defective in lipopolysaccharide (LPS) O-antigen biosynthesis lack social motility, yet have tfp. In fact, the LPS mutants are hyperpiliated, like pilT mutants. In a separate study, it had been shown that M. xanthus dif mutants that

Pulling Together with Type IV Pili

lack fibrils also have more tfp than wild-type cells because their tfp do not retract. It was shown that the addition of crude fibril material brought the dif mutants down to normal levels of tfp. However, a pilT mutant was not brought down by addition of isolated fibrils or by a fibril donor strain [Li et al., 2003]. Using the same assay, Li et al. [2003] showed that protease-treated fibrils, or chitin, a ß1,4-linked polymer of N-acetyl glucosamine brought the dif mutants down to normal levels of tfp. Unlike the dsp and dif mutants, the LPS mutants have fibrils and can even donate fibrils to a dif mutant [Li et al., 2003]. They also have the pilus-based cohesivity of wild-type cells. Taking these data together, the observation that LPS is necessary for S-motility suggests that the O-antigen mutants are deficient in pilus retraction. Because chains of the O-antigen completely cover the cell surface, a retracting pilus sliding through this covering would be expected to interact repeatedly with the O-antigen chains, and the absence of O-antigen might therefore disable retraction. Since social motility is not observed among cells that are more than a pilus length apart [Kaiser and Crosby, 1983], the observations together imply that a pilus extends ahead of an M. xanthus cell, adheres to fibrils on other cells, then retracts, pulling the leading end of the piliated cell forward toward the other cells. In this series of actions, M. xanthus offers a model for social gliding and twitching motility.

J Mol Microbiol Biotechnol 2004;7:52–62

59

References Arnold JW, Shimkets L: Inhibition of cell-cell interactions in Myxococcus xanthus by Congo red. J Bacteriol 1988a;170:5765–5770. Arnold JW, Shimkets LJ: Cell surface properties correlated with cohesion in Myxococcus xanthus. J Bacteriol 1988b;170:5771–5777. Behmlander RM, Dworkin M: Biochemical and structural analyses of the extracellular matrix fibrils of Myxococcus xanthus. J Bacteriol 1994a; 176:6295–6303. Behmlander RM, Dworkin M: Integral proteins of the extracellular matrix fibrils of Myxococcus xanthus. J Bacteriol 1994b;176:6304–6311. Bhaya D, Bianco NR, Bryant D, Grossman A: Type IV pilus biogenesis and motility in the cyanobacterium Synechocystis sp. PCC6803. Mol Microbiol 2000;37:941–951. Bieber D, Ramer SW, Wu CY, Murray WJ, Tobe T, Fernandez R, Schoolnik GK: Type IV pili, transient bacterial aggregates, and virulence of enteropathogenic Escherichia coli. Science 1998;280:2114–2118. Bitter W, Koster M, Latijnhouwers M, de Cock H, Tommassen J: Formation of oligomeric rings by XcpQ and PilQ, which are involved in protein transport across the outer membrane of Pseudomonas aeruginosa. Mol Microbiol 1998; 27:209–219. Bowden MG, Kaplan HB: The Myxococcus xanthus lipopolysaccharide O-antigen is required for social motility and multicellular development. Mol Microbiol 1998;30:275–284. Bradley DE: Shortening of Pseudomonas pili after RNA-phage adsorption. J Gen Microbiol 1972; 72:303–319. Bradley DE: Basic characterization of a Pseudomonas aeruginosa pilus-dependent bacteriophage with a long noncontractile tail. Virology 1973;12:1139–1148. Bradley DE: The adsorption of Pseudomonas aeruginosa pilus-dependent bacteriophages to a host mutant with nonretractile pili. Virology 1974;58:149–163. Bradley DE: A function of Pseudomonas aeruginosa PAO polar pili: Twitching motility. Can J Microbiol 1980;26:146–154. Brok R, Van Gelder P, Winterhalter M, Ziese U, Koster AJ, de Cock H, Koster M, Tommassen J, Bitter W: The C-terminal domain of the Pseudomonas secretin XcpQ forms oligomeric rings with pore activity. J Mol Biol 1999;294: 1169–1179. Chen LY, Chen DY, Miaw J, and Hu NT: XpsD, an outer membrane protein required for protein secretion by Xanthomonas campestris pv. campestris, forms a multimer. J Biol Chem 1996;271:2703–2708. Chi E, Mehl T, Nunn D, Lory S: Interaction of Pseudomonas aeruginosa with A549 pneumocyte cells. Infect Immun 1991;59:822–828. Collins RF, Davidsen L, Derrick JP, Ford RC, Tonjum T: Analysis of the PilQ secretin from Neisseria meningitidis by transmission electron microscopy reveals a dodecameric quaternary structure. J Bacteriol 2001;183:3825–3832.

60

Collins RF, Ford RC, Kitmitto A, Olsen RO, Tonjum T, Derrick JP: Three-dimensional structure of the Neisseria meningitidis secretin PilQ determined from negative-stain transmission electron microscopy. J Bacteriol 2003;185: 2611–2617. Collyn F, Lety MA, Nair S, Escuyer V, Ben Younes A, Simonet M, Marceau M: Yersinia pseudotuberculosis harbors a type IV pilus gene cluster that contributes to pathogenicity. Infect Immun 2002;70:6196–6205. Comolli JC, Hauser AR, Waite L, Whitchurch CB, Mattick JS, Engel JN: Pseudomonas aeruginosa gene products PilT and PilU are required for cytotoxicity in vitro and virulence in a mouse model of acute pneumonia. Infect Immun 1999;67:3625–3630. Cornelis G, Van Gijsegem F: Assembly and function of type III secretory systems. Annu Rev Microbiol 2000;54:735–774. Costerton JW, Stewart PS, Greenberg EP: Bacterial biofilms: A common cause of persistent infections. Science 1999;284:1318–1322. Crago AM, Koronakis V: Salmonella InvG forms a ring-like multimer that requires the InvH lipoprotein for outer membrane localization. Mol Microbiol 1998;30:47–56. Craig L, Taylor R, Pique M, Adair B, Arvai A, Singh M, Lloyd S, Shin D, Getzoff E, Yeager M, Forest K, and Tainer J: Type IV pilin structure and assembly: X-ray and EM analyses of Vibrio cholerae toxin-coregulated pilus and Pseudomonas aeruginosa PAK pilin. Molecular Cell 2003;11:1139–1150. Daefler S, Guilvout I, Hardie KR, Pugsley AP, Russel M: The C-terminal domain of the secretin PulD contains the binding site for its cognate chaperone, PulS, and confers PulS dependence on pIVf1 function. Mol Microbiol 1997; 24:465–475. Daefler S, Russel M: The Salmonella typhimurium InvH protein is an outer membrane lipoprotein required for the proper localization of InvG. Mol Microbiol 1998;28:1367–1380. Darzins A: Characterization of a Pseudomonas aeruginosa gene cluster involved in pilus biogenesis and twitching motility: Sequence similarity to the chemotaxis genes of the gliding bacterium Myxococcus xanthus. Mol Microbiol 1994; 11:137–153. Drake SL, Koomey M: The product of the pilQ gene is essential for the biogenesis of type IV pili in Neisseria gonorrhoeae. Mol Microbiol 1995;18:975–986. Drake SL, Sandstedt SA, Koomey M: PilP, a pilus biogenesis lipoprotein in Neisseria gonorrhoeae, affects expression of PilQ as a highmolecular-mass multimer. Mol Microbiol 1997;23:657–668. Dubnau D: DNA uptake in bacteria. Annu Rev Microbiol 1999;53:217–244.

J Mol Microbiol Biotechnol 2004;7:52–62

Dworkin M: Fibrils as extracellular appendages of bacteria: Their role in contact-mediated cellcell interactions in Myxococcus xanthus. BioEssays 1999;21:590–595. Farinha MA, Conway BD, Glasier LM, Ellert NW, Irvin RT, Sherburne R, Paranchych W: Alteration of the pilin adhesin of Pseudomonas aeruginosa PAO results in normal pilus biogenesis but a loss of adherence to human pneumocyte cells and decreased virulence in mice. Infect Immun 1994;62:4118–4123. Folkhard W, Marvin DA, Watts TH, Paranchych W: Structure of polar pili from Pseudomonas aeruginosa strains K and O. J Mol Biol 1981; 149:79–93. Forest KT, Bernstein S, Getzoff ED, So M, Tribbick G, Geysen HM, Deal CD, Tainer JA: Assembly and antigenicity of the Neisseria gonorrhoeae pilus mapped with antibodies. Infect Immun 1996;64:644–652. Forest KT, Tainer JA: Type-4 pilus-structure: Outside to inside and top to bottom – a minireview. Gene 1997;192:165–169. Genin S, Boucher CA: A superfamily of proteins involved in different secretion pathways in gram-negative bacteria: Modular structure and specificity of the N-terminal domain. Mol Gen Genet 1994;243:112–118. Giron JA: Longus pilus on enterotoxigenic Escherichia coli and its relatedness to other type-4 pili – a minireview. Gene 1997;192:39–43. Giron JA, Ho A, Schoolnik GK: An inducible bundle-forming pilus of enteropathogenic Escherichia coli. Science 1991;254:710–713. Gophna U, Parket A, Hacker J, Ron EZ: A novel ColV plasmid encoding type IV pili. Microbiology 2003;149:177–184. Guilvout I, Hardie KR, Sauvonnet N, Pugsley AP: Genetic dissection of the outer membrane secretin PulD: Are there distinct domains for multimerization and secretion specificity? J Bacteriol 1999;181:7212–7220. Hardie KR, Lory S, Pugsley AP: Insertion of an outer membrane protein in Escherichia coli requires a chaperone-like protein. EMBO J 1996a;15:978–988. Hardie KR, Seydel A, Guilvout I, Pugsley AP: The secretin-specific, chaperone-like protein of the general secretory pathway: Separation of proteolytic protection and piloting functions. Mol Microbiol 1996b; 22:967–976. Hazes B, Sastry P, Hayakawa K, Read R, Irvin R: Crystal structure of Pseudomonas aeruginosa PAK pilin suggests a main-chain-dominated mode of receptor binding. J Mol Biol 2000;299: 1005–1017. Henrichsen J: Bacterial surface translocation: A survey and a classification. Bacteriol Rev 1972; 36:478–503. Henrichsen J: Twitching motility. Annu Rev Microbiol 1983;37:81–93. Henrichsen J, Blom J: Correlation between twitching motility and possession of polar fimbriae in Acinetobacter calcoaceticus. Acta Pathol Microbiol Scand B 1975;83:103–115.

Nudleman/Kaiser

Henrichsen J, Froholm LO, Bovre K: Studies on bacterial surface translocation. 2. Correlation of twitching motility and fimbriation in colony variants of Moraxella nonliquifaciens, M. bovis, and M. kingi. Acta Pathol Microbiol Scand B 1972;80:445–452. Herdendorf TJ, McCaslin DR, Forest KT: Aquifex aeolicus PilT, homologue of a surface motility protein, is a thermostable oligomeric NTPase. J Bacteriol 2002;184:6465–6471. Heydorn A, Ersboll B, Kato J, Hentzer M, Parsek MR, Tolker-Nielsen T, Givskov M, Molin S: Statistical analysis of Pseudomonas aeruginosa biofilm development: Impact of mutations in genes involved in twitching motility, cell-to-cell signaling, and stationary-phase sigma factor expression. Appl Environ Microbiol 2002;68: 2008–2017. Kaiser D: Social gliding is correlated with the presence of pili in Myxococcus xanthus. Proc Natl Acad Sci USA 1979;76:5952–5956. Kaiser D: How do pili pull? Curr Biol 2000;10: R777–R780. Kaiser D: Coupling cell movement and multicellular development in Myxobacteria. Nature Rev Microbiol 2003;1:45–54. Kaiser D, Crosby C: Cell movement and its coordination in swarms of Myxococcus xanthus. Cell Motil 1983;3:227–245. Kazmierczak BI, Mielke DL, Russel M, Model P: pIV, a filamentous phage protein that mediates phage export across the bacterial cell envelope, forms a multimer. J Mol Biol 1994;238:187– 198. Kim SK, Kaiser D: Cell motility is required for the transmission of C-factor, an intercellular signal that coordinates fruiting body morphogenesis of Myxococcus xanthus. Genes Dev 1990;4: 896–905. Kirn TJ, Bose N, Taylor RK: Secretion of a soluble colonization factor by the TCP type 4 pilus biogenesis pathway in Vibrio cholerae. Mol Microbiol 2003;49:81–92. Klausen M, Heydorn A, Ragas P, Lambertsen L, Aaes-Jorgensen A, Molin S, Tolker-Nielsen T: Biofilm formation by Pseudomonas aeruginosa wild type, flagella and type IV pili mutants. Mol Microbiol 2003;48:1511–1524. Koster M, Bitter W, de Cock H, Allaoui A, Cornelis GR, Tommassen J: The outer membrane component, YscC, of the Yop secretion machinery of Yersinia enterocolitica forms a ring-shaped multimeric complex. Mol Microbiol 1997;26: 789–797. Lee KK, Sheth HB, Wong WY, Sherburne R, Paranchych W, Hodges RS, Lingwood CA, Krivan H, and Irvin RT: The binding of Pseudomonas aeruginosa pili to glycosphingolipids is a tip-associated event involving the C-terminal region of the structural pilin subunit. Mol Microbiol 1994;11:705–713. Li Y, Sun H, Ma X, Lu A, Lux R, Zusman D, Shi W: Extracellular polysaccharides mediate pilus retraction during social motility of Myxococcus xanthus. Proc Natl Acad Sci USA 2003;100: 5443–5448.

Pulling Together with Type IV Pili

Linderoth NA, Simon MN, Russel M: The filamentous phage pIV multimer visualized by scanning transmission electron microscopy. Science 1997;278:1635–1638. Liu H, Kang Y, Genin S, Schell MA, Denny TP: Twitching motility of Ralstonia solanacearum requires a type IV pilus system. Microbiology 2001;147:3215–3229. MacRae TH, McCurdy HD: Evidence for motilityrelated fimbriae in the gliding microorganism Myxococcus xanthus. Can J Microbiol 1976; 22:1589–1593. Maier B, Potter L, So M, Seifert HS, Sheetz MP: Single pilus motor forces exceed 100 pN. Proc Natl Acad Sci 2002;99:16012–16017. Martin PR, Watson AA, McCaul TF, Mattick JS: Characterization of a five-gene cluster required for the biogenesis of type 4 fimbriae in Pseudomonas aeruginosa. Mol Microbiol 1995;16: 497–508. Mattick JS: Type IV pili and twitching motility. Annu Rev Microbiol. 2002;56:289–314. Mattick JS, Whitchurch CB, Alm RA: The molecular genetics of type-4 fimbriae in Pseudomonas aeruginosa – a review. Gene 1996;179:147– 155. Merz AJ, So M, Sheetz MP: Pilus retraction powers bacterial twitching motility. Nature 2000;407: 98–102. Newhall WJ, Wilde CE, Sawyer WD, Haak RA: High-molecular-weight antigenic protein complex in the outer membrane of Neisseria gonorrhoeae. Infect Immun 1980;27:475–482. Nouwen N, Ranson N, Saibil H, Wolpensinger B, Engel A, Ghazi A, Pugsley AP: Secretin PulD: Association with pilot PulS, structure, and ionconducting channel formation. Proc Natl Acad Sci USA 1999;96:8173–8177. Nunn D, Bergman S, Lory S: Products of three accessory genes, pilB, pilC, and pilD, are required for biogenesis of Pseudomonas aeruginosa pili. J Bacteriol 1990;172:2911–2919. Nunn DN, Lory S: Product of the Pseudomonas aeruginosa gene pilD is a prepilin leader peptidase. Proc Natl Acad Sci USA 1991;88:3281– 3285. O’Toole GA, Kolter R: Flagellar and twitching motility are necessary for Pseudomonas aeruginosa biofilm development. Mol Microbiol 1998; 30:295–304. Okamoto S, Ohmori M: The cyanobacterial PilT protein responsible for cell motility and transformation hydrolyzes ATP. Plant Cell Physiol 2002;43:1127–1136. Page RDM: TREEVIEW: An application to display phylogenetic trees on personal computers. Comput Appl Biosci 1996;12:357–358. Parge HE, Forest KT, Hickey MJ, Christensen DA, Getzoff ED, Tainer JA: Structure of the fibreforming protein pilin at 2.6 Å resolution. Nature 1995;378:32–38. Pepe JC, Lory S: Amino acid substitutions in PilD, a bifunctional enzyme of Pseudomonas aeruginosa. J Biol Chem 1998;273:19120–19129. Pugsley AP, Francetic O, Possot OM, Sauvonnet N, Hardie KR: Recent progress and future directions in studies of the main terminal branch of the general secretory pathway in gram-negative bacteria – a review. Gene 1997;192:13–19.

Pujol C, Eugene E, Marceau M, Nassif X: The meningococcal PilT protein is required for induction of intimate attachment to epithelial cells following pilus-mediated adhesion. Proc Natl Acad Sci USA 1999;96:4017–4022. Purcell EM: Life at low Reynolds number. Am J Physiol 1977;45:3–11. Ramer SW, Bieber D, Schoolnik GK: BfpB, an outer membrane lipoprotein required for the biogenesis of bundle-forming pili in enteropathogenic Escherichia coli. J Bacteriol 1996;178: 6555–6563. Rothbard JB, Fernandez R, Wang L, Teng NN, Schoolnik GK: Antibodies to peptides corresponding to a conserved sequence of gonococcal pilins block bacterial adhesion. Proc Natl Acad Sci USA 1985;82:915–919. Ruehl WW, Marrs C, Beard MK, Shokooki V, Hinojoza JR, Banks S, Bieber D, Mattick JS: Q pili enhance the attachment of Moraxella bovis to bovine corneas in vitro. Mol Microbiol 1993;7:285–288. Sager B, Kaiser D: Spatial restriction of cellular differentiation. Genes Dev 1993a; 7:1645–1653. Sager B, Kaiser D: Two cell-density domains within the Myxococcus xanthus fruiting body. Proc Natl Acad Sci USA 1993b;90:3690–3694. Sandkvist M: Type II secretion and pathogenesis. Infect Immunity 2001;69:3523–3535. Scheuerpflug I, Rudel T, Ryll R, Pandit J, Meyer TF: Roles of PilC and PilE proteins in pilusmediated adherence of Neisseria gonorrhoeae and Neisseria meningitidis to human erythrocytes and endothelial and epithelial cells. Infect Immun 1999;67:834–843. Schmidt SA, Bieber D, Ramer SW, Hwang J, Wu CY, Schoolnik G: Structure-function analysis of BfpB, a secretin-like protein encoded by the bundle-forming-pilus operon of enteropathogenic Escherichia coli. J Bacteriol 2001;183: 4848–4859. Semmler ABT, Whitchurch CB, Mattick JS: A reexamination of twitching motility in Pseudomonas aeruginosa. Microbiology 1999;145: 2863–2873. Sheth HB, Lee KK, Wong WY, Srivastava G, Hindsgaul O, Hodges RS, Paranchych W, Irvin RT: The pili of Pseudomonas aeruginosa strains PAK and PAO bind specifically to the carbohydrate sequence beta GalNAc(1–4)beta Gal found in glycosphingolipids asialo-GM1 and asialo-GM2. Mol Microbiol 1994;11:715– 723. Shevchik VE, Condemine G: Functional characterization of the Erwinia chrysanthemi OutS protein, an element of a type II secretion system. Microbiology 1998;144:3219–3228. Shimkets L: Correlation of energy-dependent cell cohesion with social motility in Myxococcus xanthus. J Bacteriol 1986;166:837–841. Skerker JM, Berg HC: Direct observation of extension and retraction of type IV pili. Proc Natl Acad Sci USA 2001;98:6901–6904. Skerker J, Shapiro L: Identification and cell cycle control of a novel pilus system in Caulobacter crescentus. EMBO J 2000;19:3223–3234.

J Mol Microbiol Biotechnol 2004;7:52–62

61

Stone KD, Zhang HZ, Carlson LK, Donnenberg MS: A cluster of fourteen genes from enteropathogenic Escherichia coli is sufficient for the biogenesis of a type IV pilus. Mol Microbiol 1996;20:325–337. Strom MS, Lory S: Structure-function and biogenesis of the type IV pili. Annu Rev Microbiol 1993;47:565–596. Suhnel J: Image library of biological macromolecules. Comput Appl Biosci 1996;12:227–229. Sun H, Zusman DR, Shi W: Type IV pilus of Myxococcus xanthus is a motility apparatus controlled by the frz chemosensory system. Curr Biol 2000;10:1143–1146. Vale RD: AAA proteins: Lords of the ring. J Cell Biol 2000;150:F13–F19. Vogel S: Life in moving fluids. Princeton, Princeton University Press, 1994. Wall D, Kaiser D: Type IV pili and cell motility. Mol Microbiol 1999;32:1–10.

62

Wall D, Kolenbrander PE, Kaiser D: The Myxococcus xanthus pilQ (sglA) gene encodes a secretin homolog required for type IV pilus biogenesis, social motility, and development. J Bacteriol 1999;181:24–33. Whitchurch CB, Hobbs M, Livingston SP, Krishnapillai V, Mattick JS: Characterisation of a Pseudomonas aeruginosa twitching motility gene and evidence for a specialised protein export system widespread in eubacteria. Gene 1990;101:33–44. Wolfgang M, Park HS, Hayes SF, van Putten JP, Koomey M: Suppression of an absolute defect in type IV pilus biogenesis by loss-of-function mutations in pilT, a twitching motility gene in Neisseria gonorrhoeae. Proc Natl Acad Sci USA 1998;95:14973–14978. Wolfgang M, van Putten JPM, Hayes SF, Dorward D, Koomey M: Components and dynamics of fiber formation define a ubiquitous biogenesis pathway for bacterial pili. EMBO J 2000;19: 6408–6418. Wu SS, Kaiser D: Genetic and functional evidence that type IV pili are required for social gliding motility in Myxococcus xanthus. Mol Microbiol 1995;18:547–558.

J Mol Microbiol Biotechnol 2004;7:52–62

Wu SS, Wu J, Kaiser D: The Myxococcus xanthus pilT locus is required for social gliding motility although pili are still produced. Mol Microbiol 1997;23:109–121. Yang Z, Ma X, Tong L, Kaplan HB, Shimkets LJ, Shi W: The Myxococcus xanthus dif genes are required for the biogenesis of cell surface fibrils essential for social gliding motility. J Bacteriol 2000;182:5793–5798. Yoshihara S, Geng X, Okamoto S, Yura K, Murata T, Go M, Ohmori M, Ikeuchi M: Mutational analysis of genes involved in pilus structure, motility and transformation competency in the unicellular motile cyanobacterium Synechocystis sp. PCC 6803. Plant Cell Physiol 2001;42: 63–73. Zhang XL, Tsui IS, Yip CM, Fung AW, Wong DK, Dai X, Yang Y, Hackett, J Morris C: Salmonella enterica serovar typhi uses type IVB pili to enter human intestinal epithelial cells. Infect Immun 2000;68:3067–3073.

Nudleman/Kaiser

J Mol Microbiol Biotechnol 2004;7:63–71 DOI: 10.1159/000077870

Cytophaga-Flavobacterium Gliding Motility Mark J. McBride Department of Biological Sciences, University of Wisconsin-Milwaukee, Milwaukee, Wisc., USA

Key Words Gliding motility W Flavobacterium W Cytophaga W Chitin W Transporter

Abstract Flavobacterium johnsoniae, like many other members of the Cytophaga-Flavobacterium-Bacteroides group, displays rapid gliding motility. Cells of F. johnsoniae glide over surfaces at rates of up to 10 Ìm/s. Latex spheres added to F. johnsoniae bind to and are rapidly propelled along cells, suggesting that adhesive molecules move laterally along the cell surface during gliding. Genetic analyses have identified a number of gld genes that are required for gliding. Three Gld proteins are thought to be components of an ATP-binding-cassette transporter. Five other Gld proteins are lipoproteins that localize to the cytoplasmic membrane or outer membrane. Disruption of gld genes results not only in loss of motility, but also in resistance to bacteriophages that infect wild-type cells, and loss of the ability to digest the insoluble polysaccharide chitin. Two models that attempt to incorporate the available data to explain the mechanism of F. johnsoniae gliding are presented. Copyright © 2004 S. Karger AG, Basel

ABC Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

© 2004 S. Karger AG, Basel 1464–1801/04/0072–0063$21.00/0 Accessible online at: www.karger.com/mmb

Introduction

Surfaces are important features of many environments. Bacteria often preferentially colonize surfaces and have evolved a number of strategies for moving over them. Some bacteria use numerous flagella to spread over moist surfaces by a process known as swarming motility [Harshey, 1994]. Others use type IV pili to move by twitching motility [Henrichsen, 1983; Wall and Kaiser, 1999]. Finally, phylogenetically diverse bacteria such as Flavobacterium johnsoniae, Myxococcus xanthus, Phormidium uncinatum and many others crawl over surfaces by a process known as gliding motility [Hoiczyk, 2000; McBride, 2001; Spormann, 1999; Youderian, 1998]. Bacterial gliding motility is defined as smooth translocation of cells over a surface by an active process that does not involve flagella and requires the expenditure of energy. The cells move in the direction of their long axes and produce colonies that have thin spreading edges (fig. 1A). A number of mechanisms have been proposed to explain gliding [Burchard, 1984; Hoiczyk, 2000; Lapidus and Berg, 1982; McBride, 2001; McBride et al., 2003; Pate, 1988; Spormann, 1999; Youderian, 1998]. Recent results suggest that there are probably several different types of gliding ‘motors’ that have evolved independently. Myxobacterial ‘social gliding motility’ and gliding of unicellular cyanobacteria such as Synechocystis PCC6803 are

Mark J. McBride Department of Biological Sciences, University of Wisconsin-Milwaukee 3209 N. Maryland Ave/181 Lapham Hall Milwaukee, WI 53211 (USA) Tel. +1 414 229 5844, Fax +1 414 229 3926, E-Mail [email protected]

ments. Swimming motility is rare within this group, but gliding motility over surfaces is common. The gliding behavior of F. johnsoniae is characteristic of members of this group. Cells of F. johnsoniae glide at rates of 2– 10 Ìm/s over wet glass surfaces. Cells occasionally reverse their direction of movement, and also flip end over end. Finally, cells that attach to a surface by a single pole often rotate in place at frequencies of about 2 revolutions/s.

Ultrastructural Analyses of F. johnsoniae and Related Gliding Bacteria

Electron microscopic studies have been performed on F. johnsoniae and related bacteria, but they have not conclusively revealed the gliding motility machinery. Pili have not been detected on F. johnsoniae or related gliding bacteria [Henrichsen and Blom, 1975; Reichenbach, 1992]. Ring-like structures have been observed in the cell envelope of F. johnsoniae [Pate and Chang, 1979], and goblet-shaped structures have been seen in the cell wall of Flexibacter polymorphus [Ridgway and Lewin, 1973]. Helical bands have also been observed in the cell envelopes of several members of the Cytophaga-Flavobacterium group [Lunsdorf and Schairer, 2001]. The involvement of any of these structures in cell movement remains uncertain.

Fig. 1. Colonies and cells of F. johnsoniae. A Spreading colony of wild-type cells. Bar indicates 1 mm. B Nonspreading colonies of a gldA mutant. Bar indicates 1 mm. C Wild-type cells moving over an

agar surface. Bar indicates 20 Ìm. Reprinted with permission from the Annu Rev Microbiol, vol. 55, 2001. Annual Reviews, www. annualreviews.org.

similar to bacterial twitching motility and rely on type IV pilus extension and retraction [Wall and Kaiser, 1999]. In contrast, myxobacterial ‘adventurous gliding motility’ and gliding of filamentous cyanobacteria may rely on polysaccharide secretion [Hoiczyk and Baumeister, 1998; Wolgemuth et al., 2002], whereas mycoplasma gliding appears to involve the cytoskeleton [Korolev et al., 1994; Miyata et al., 2000]. Many members of the Cytophaga-FlavobacteriumBacteroides (CFB) branch of the bacterial phylogenetic tree exhibit a form of rapid gliding motility that appears to be distinct from each of those described above [McBride, 2001; Pate, 1988]. The members of the CFB group are diverse and are abundant in many environ-

64

J Mol Microbiol Biotechnol 2004;7:63–71

Energy Source for Gliding

The source of energy for gliding of F. johnsoniae and for other members of the CFB group has been investigated by determining the effects of respiratory poisons, uncouplers, or ATPase inhibitors [Duxbury et al., 1980; Dzink-Fox et al., 1997; Pate and Chang, 1979; Ridgway, 1977]. Proton motive force (PMF) appears to be required for gliding. Uncouplers or poisons that diminish PMF block gliding motility. In contrast inhibitors such as arsenate, which result in a decrease in ATP levels, have little effect on gliding. The ATP pools were not completely depleted in these studies, so it remains possible that both proton motive force and ATP are required for cell movement.

Observations of Cell Surface Movements

The motility apparatus can be observed in action by adding latex spheres to cells [Lapidus and Berg, 1982; Pate and Chang, 1979; Ridgway and Lewin, 1988].

McBride

Spheres drifting near a cell do not exhibit any directed motion until they contact the cell. Once contact has been made, the spheres are propelled along the cell surface (fig. 2). Spheres often travel along the length of a cell and may reverse direction abruptly. A single sphere may migrate the length of a cell, proceed around the pole, and continue down the opposite side. Spheres attached to a cell pole may also rotate in place. When multiple spheres bind to a cell they may follow the same or different paths and two spheres may even pass close by one another in opposite directions with neither sphere altering its course. Sphere movement and cell movement appear to be mediated by the same machinery since spheres move at approximately the same speed as gliding cells, and metabolic poisons that inhibit motility also block sphere movement [Pate and Chang, 1979]. Furthermore, nonmotile mutants of F. johnsoniae fail to move spheres, and complementation results in restoration of both gliding motility and sphere movement [Agarwal et al., 1997; Hunnicutt and McBride, 2000, 2001; Hunnicutt et al., 2002; Kempf and McBride, 2000; McBride et al., 2003]. Analyses of latex sphere movements suggest that gliding involves movement of adhesive molecules along the cell surface. Other proposed mechanisms, such as inchworm-like contraction and expansion of cytoskeletal components, or propulsion by polymer extrusion appear unlikely. Spheres are not propelled away from cells as would be predicted by a polymer extrusion model and it is difficult to imagine how polymer extrusion or cytoskeletal contractions could result in spheres moving past each other in opposite directions.

Genetic Analysis of F. johnsoniae Gliding: Identification of Genes and Proteins Required for Gliding

Tools and techniques that allow the genetic manipulation of F. johnsoniae have been developed to aid the identification of components of the gliding motility machinery [McBride and Kempf, 1996]. These include shuttle plasmids and cosmids and methods for gene transfer, transposon mutagenesis, and site-directed gene disruption. Many nonmotile mutants have been isolated as spontaneous mutants or by transposon or chemical mutagenesis [Agarwal et al., 1997; Chang et al., 1984; Godchaux et al., 1990; Gorski et al., 1992; Hunnicutt and McBride, 2000, 2001, 2002; Kempf and McBride, 2000; McBride et al., 2003; Wolkin and Pate, 1985]. Analysis of these mutants has resulted in the identification of genes and proteins

Cytophaga-Flavobacterium Gliding Motility

Fig. 2. Movement of latex spheres by F. johnsoniae cells. Latex spheres bind to cells and are propelled along their surfaces. Two spheres attached to the same cell may move in the same or different directions, and may pass each other moving in opposite directions. Spheres attached to a cell pole may also rotate in place.

that are required for gliding motility. Colonies of nonmotile mutants fail to spread (fig. 1B) and individual cells fail to glide over glass or agar surfaces. Cells of these mutants also lack the rotary movements seen with wild-type cells, and fail to propel latex spheres over their surfaces. Table 1 lists proteins that are required for F. johnsoniae gliding motility. gldA encodes a protein that exhibits sequence similarity to components of ATP-binding cassette (ABC) transporters [Agarwal et al., 1997]. ABC transporters are found in most if not all organisms. They use the energy of ATP hydrolysis to transport molecules across one or more membranes into or out of cells. Molecules as diverse as inorganic ions, amino acids, lipids, polysaccharides and proteins are transported by ABC transporters. Many bacterial ABC transporters are composed of multiple subunits. GldA is similar to the ATPhydrolyzing components of ABC transporters. GldA has no obvious transmembrane domains, but cell fractionation followed by Western blot analyses demonstrated its presence in membrane fractions [Hunnicutt et al., 2002]. gldF and gldG, which were identified by transposon mutagenesis, are required for gliding and for proper membrane localization of GldA [Hunnicutt et al., 2002]. gldF and gldG constitute an operon that is not closely linked to gldA. GldF and GldG are similar to membrane components of ABC transporters and probably interact with GldA to form the gld ABC transporter. GldF and GldG exhibit limited similarity to transporter components of known function, but analysis of GldA homologs is more informative. GldA is similar to proteins such as Bacillus subtilis SpaF (40% identity over 220 amino acids), Bacil-

J Mol Microbiol Biotechnol 2004;7:63–71

65

Table 1. Proteins involved in F. johnsoniae gliding motility

Protein

Sizea kD

Localizationb

Homologs of known function

Possible function in Reference gliding motility

GldA GldB GldD GldEc GldF GldG GldH GldI GldJ

33.6 36.3 19.3 48.6 26.9 63.7 16.1 22.3 60.8

CM/C CM, lipoprotein CM, lipoprotein CMd CMd CM CM, lipoprotein OMd, lipoprotein OM, lipoprotein

Transport ? ? Transport Transport Transport ? Protein folding Transport?

Agarwal et al. [1997] Hunnicutt and McBride [2000] Hunnicutt and McBride [2001] Hunnicutt and McBride [2001] Hunnicutt et al. [2002] Hunnicutt et al. [2002] McBride et al. [2003] Genbank accession No. AF527792 Genbank accession No. AF527793

SecDF FtsX

108 32.9

ABC transporter components None None CorC (involved in magnesium export) ABC transporter components ABC transporter components None Peptidyl-prolyl isomerases CarF (carbapenam resistance), XiaF (extracellular xylanase activity) SecD, SecF FtsX (cell division)

Protein secretion Protein secretion?

[Nelson and McBride, unpubl. data] Kempf and McBride [2000]

CMd CMd

a

Molecular mass calculated from amino acid sequences. Localization determined experimentally except as indicated. Abbreviations: cytoplasm (C), cytoplasmic membrane (CM), outer membrane (OM). c Overexpression of GldE partially suppresses a gldB point mutation. Mutations in gldE have not been isolated and it is not known whether GldE is required for motility. d Localization predicted from sequence analyses. b

lus licheniformis BcrA (33% identity over 295 amino acids), Sinorhizobium meliloti NodI (33% identity over 299 amino acids), S. meliloti NosF (38% identity over 216 amino acids), and M. xanthus PilH (38% identity over 213 amino acids). SpaF [Klein and Entian, 1994] and BcrA [Podlesek et al., 1995] are involved in resistance to the peptide antibiotics subtilin and bacitracin, respectively, presumably by exporting them. NodI is thought to function in lipochitin oligosaccharide export [Spaink et al., 1995], and NosF interacts with NosD and NosY (a GldF homolog) to form a transporter that is involved in maturation of the periplasmic nitrous oxide reductase, NosZ [Holloway et al., 1996]. M. xanthus PilH is involved in pilus biogenesis and social gliding motility [Wu et al., 1998]. It is unlikely that pili are responsible for F. johnsoniae motility since pili have not been observed on cells of F. johnsoniae and genes similar to those required for pilus assembly and function (other than pilH) have not been identified. However, GldA and M. xanthus PilH may perform analogous roles in the assembly, modification or functioning of their respective motility machineries. The GldA homologs have diverse functions, but most are thought to be involved in export. Several cell surface molecules have been implicated in F. johnsoniae gliding. Sulfonolipids are found in the outer membrane of F. john-

66

J Mol Microbiol Biotechnol 2004;7:63–71

soniae, and several mutants that fail to make sulfonolipids exhibit motility defects [Abbanat et al., 1986; Godchaux and Leadbetter, 1988; Godchaux et al., 1990]. Sulfonolipids may be directly involved in gliding or may be needed for proper assembly and functioning of the motility apparatus. Cells with mutations in gldA, gldF or gldG have normal levels of sulfonolipids [Leadbetter, unpubl. data], but it is possible that they are not localized properly. Cell surface polysaccharides and glycoproteins have also been implicated in gliding [Godchaux et al., 1990, 1991; Pate, 1988; Pate and De Jong, 1990] and GldA, GldF, and GldG may be involved in export or modification of these molecules. The gld ABC transporter might also be involved in export of protein components of the gliding machinery to the periplasm or outer membrane. Given the many functions of ABC transporters, other roles are also possible. GldB, GldD, and GldH are each required for motility but they do not exhibit sequence similarity to proteins of known function [Hunnicutt and McBride, 2000, 2001; McBride et al., 2003]. Analysis of the predicted amino acid sequences of GldB, GldD, and GldH suggested that they might be lipoproteins since each has a hydrophobic signal peptide followed by a cysteine residue. Incubation of cells with 3H-palmitate resulted in labeling of each protein, confirming this prediction [McBride et al., 2003].

McBride

Western blot analyses of cell fractions separated by gradient centrifugation suggest that GldB, GldD, and GldH are tethered to the cytoplasmic membrane by their lipid tails [Braun and McBride, unpubl. data]. The exact roles of these proteins in cell movement remain to be determined. gldE lies immediately upstream of gldD on the F. johnsoniae genome. Tenfold overproduction of GldE partially suppresses a gldB point mutation, suggesting that it may be involved in gliding [Hunnicutt and McBride, 2001]. Unlike the other Gld proteins, it is not known whether GldE is required for cell movement since cells with mutations in gldE have not been constructed. GldE is similar in sequence to Salmonella enterica serovar Typhimurium CorC which is thought to be involved in magnesium export [Gibson et al., 1991], and to many proteins of unknown function. GldI is similar to peptidyl-prolyl cis/trans-isomerases that are involved in protein folding [McBride and Braun, 2004]. Sequence analysis and labeling studies indicate that GldI, like GldB, GldD and GldH, is a lipoprotein. Sequence analysis suggests that GldI is anchored to the outer membrane by its lipid tail. GldI may play a role in folding components of the motility machinery. Cells with mutations in gldI grow at the same rate as wild-type cells so GldI is not essential for viability and may have a more specialized role in motility. GldJ is yet another lipoprotein that is required for gliding motility [Braun and McBride, unpubl. data]. GldJ localizes to the outer membrane fraction of cells. GldJ exhibits similarity to CarF, which confers carbapenem resistance on Erwinia carotovora [McGowan et al., 1997], and to Bacillus stearothermophilus XiaF, which is involved in extracellular xylanase activity [Cho and Choi, 1998]. CarF and XiaF may function in export processes, but other roles are also possible. Several pieces of evidence indicate that GldJ may interact with the other Gld proteins to form a complex. First, mutations in any of the other gld genes result in decreased levels of GldJ protein, suggesting that GldJ is unstable in the absence of its partners [Braun and McBride, unpubl. data]. More direct evidence for an interaction between GldJ and GldB comes from pull-down experiments using C-terminal His-tagged versions of these proteins [Braun and McBride, unpubl. data]. secDF is also involved in gliding since cells with a transposon insertion in secDF are severely crippled and form nonspreading colonies [Nelson and McBride, unpubl. data]. F. johnsoniae SecDF is a hybrid protein with domains similar to E. coli SecD and SecF. Although the

exact roles of SecD and SecF in E. coli are not known, they are thought to function in protein secretion [Murphy and Beckwith, 1996]. E. coli secD and secF mutants exhibit severe growth defects, suggesting that secretion of essential proteins is impaired. In contrast, F. johnsoniae secDF mutants grow as rapidly as wild-type cells. Apparently, SecDF is not required for secretion of essential proteins in F. johnsoniae. It may be needed to transport some components of the motility machinery to their site of assembly, or it may play a more direct role in cell movement. Most of the genes described above were identified by Tn4351 mutagenesis. Nonmotile mutants have also been independently isolated as spontaneous mutants or by chemical mutagenesis [Chang et al., 1984; Wolkin and Pate, 1985]. Introduction of gldA, gldB, gldD, gldF, gldG, gldH, gldI and gldJ individually into 50 of these mutants restored motility to 33 of them. Apparently, this limited collection constitutes a significant fraction of the genes required for gliding. Inspection of table 1 reveals that many of the homologs of Gld proteins are thought to be involved in transport, suggesting that transporter proteins may constitute an important part of the motility machinery. Additionally, at least five of the Gld proteins are lipoproteins. The Gld proteins may assemble to form a multiprotein complex in the cell envelope and anchoring of the Gld lipoproteins may facilitate formation of this complex.

Cytophaga-Flavobacterium Gliding Motility

J Mol Microbiol Biotechnol 2004;7:63–71

Filamentous Nongliding Mutants and Motile-Nonspreading Mutants

The gld mutants described above are nonmotile and exhibit normal cell morphology. Two other classes of mutants that fail to form spreading colonies have been isolated: ‘filamentous-nonmotile’ and ‘motile-nonspreading’ mutants. Filamentous-nonmotile mutants have a defect in cell division that results in the formation of abnormally long cells. Mutations in F. johnsoniae ftsX, for example, disrupt cell division at a late step and also result in loss of motility [Kempf and McBride, 2000]. The exact function of FtsX in E. coli is not known, but it may play a role in translocation of proteins involved in potassium transport and in cell division into the cytoplasmic membrane [de Leeuw et al., 1999; Ukai et al., 1998]. Motile-nonspreading mutants form nonspreading colonies that are indistinguishable from those of nonmotile mutants. However, individual cells of motile-nonspreading mutants glide in wet mounts. Cells of some motile-nonspreading mutants glide over wet glass surfaces as well as wild-type cells,

67

Fig. 3. F. johnsoniae gld mutants fail to digest chitin. Approximately 4 ! 107 cells of wild-type F. johnsoniae (WT), of the gldI mutant UW102–41, and of UW102–41 complemented with gldI on a plasmid were spotted on media containing the insoluble polysaccharide chitin and incubated for 6 days at 25 ° C. Similar results were obtained with cells with mutations in gldA, gldB, gldD, gldF, gldG, and gldH [McBride et al., 2003].

whereas others (such as secDF mutants) are severely crippled. We are only beginning to characterize the genes that are involved in colony spreading and a clear picture regarding the reason for the lack of colony spreading in motile-nonspreading mutants has not yet emerged. Sugars and polysaccharides have previously been implicated in colony spreading by F. johnsoniae. Mutants with reduced levels of high-molecular-weight cell surface polysaccharide maintained some ability to glide, but formed colonies that spread much less than wild-type colonies [Godchaux et al., 1991]. Addition of some sugars to wild-type cells severely inhibited colony spreading on agar, but did not effect the ability of cells to glide on glass microscope slides [Gorski et al., 1993; Wolkin and Pate, 1984].

Motility, Bacteriophage Resistance and Chitin Digestion

Cells of nonmotile mutants form nonspreading colonies, fail to glide on glass or agar surfaces, and fail to propel latex spheres. They also display two additional phenotypes that distinguish them from wild-type cells. They are completely resistant to infection by bacteriophages that infect wild-type cells [Agarwal et al., 1997; Hunnicutt and McBride, 2000, 2001; Hunnicutt et al., 2002; Kempf and McBride, 2000; McBride et al., 2003; Wolkin and Pate,

68

J Mol Microbiol Biotechnol 2004;7:63–71

1986] and they are unable to digest the insoluble polysaccharide chitin [Chang et al., 1984; McBride et al., 2003] (fig. 3). The connection between cell movement, phage resistance, and ability to digest chitin is not understood. Bacteriophage infection and chitin utilization each require transport across the cell envelope. Bacteriophage infection requires the export of receptor molecules to the cell surface, and the entry of phage nucleic acid into the cell. Chitinase and chitin-binding proteins must be exported to their sites of action for chitin utilization, and chitin oligomers may be transported across the outer membrane during chitin digestion. Interestingly, the GldA homolog NodI has been implicated in export of the lipochitin oligosaccharide Nod factor [Spaink et al., 1995]. Many members of the CFB group are proficient at utilizing polysaccharides and other macromolecules [Kirchman, 2002; Reichenbach, 1992; Salyers et al., 1996; Xu et al., 2003]. Bacteroides thetaiotaomicron utilizes starch by a mechanism that may be common to other members of the CFB group [Salyers et al., 1996; Shipman et al., 2000]. B. thetaiotaomicron binds the polysaccharide on the cell surface. Hydrolysis into smaller units occurs after binding to the cell surface, and much of the digestion is thought to occur after transit across the outer membrane. F. johnsoniae gliding motility may have evolved by modifying such a polysaccharide utilization system. In this model export and import of macromolecules, such as polysaccharides or proteins, at different sites form ‘conveyor belts’ along the cell surface which propel cells (fig. 4A). The motility apparatus may consist of two types of transporters as indicated in figure 4A, or a single type of transporter that can function as both an exporter and an importer. GldA, GldF, and GldG could constitute part of an ATP-driven ABC transporter that functions in cell movement. It is unlikely that ATP hydrolysis by GldA is the sole driving force for cell movement, however, because PMF is required for gliding of F. johnsoniae [Dzink-Fox et al., 1997; Pate and Chang, 1979]. An alternative model (fig. 4B) that is consistent with the observations of cell movement involves outer membrane components that are driven along tracks by periplasmic and cytoplasmic membrane proteins that obtain energy from the PMF [Lapidus and Berg, 1982]. The gld ABC transporter may function in assembly or modification of the motility machinery in this model. In both models macromolecules are propelled along the cell surface. Temporary adhesive interaction of these molecules with the substratum is necessary for cell movement. Both models are necessarily speculative and leave much unex-

McBride

PMF? ATP?

PMF? ATP?

Substratum

CM

Fig. 4. Speculative models to explain F. johnsoniae gliding motility. A Polymer

transport model. Gliding involves coordinated export and import of polymeric material (‘conveyor belt’). Import of polymer temporarily attached to substratum propels cells. B Lateral movement of outer membrane adhesion molecules. Proteins in the cytoplasmic membrane (CM) harvest the proton motive force and propel outer membrane (OM) proteins along tracks (dotted line) that are anchored to the peptidoglycan (PG). Temporary attachment of the OM proteins to the substratum results in cell movement. Modified with permission from the Annu Rev Microbiol, vol 55, 2001. Annual Reviews, www.annualreviews.org.

PG OM

A

Direction of cell movement

Exporter

Importer

H+

Substratum

CM PG OM

B

Direction of cell movement

plained. For example, each postulates the existence of numerous motor elements per cell, but it is not known how these would be coordinated to allow directed movement. Clearly, the mechanism of F. johnsoniae gliding remains somewhat mysterious. An open mind and a healthy dose of skepticism are useful when considering any proposed model to explain F. johnsoniae gliding.

obvious GldD, GldF and GldG homologs. The functions of the Gld protein homologs in B. thetaiotaomicron are not known. They may function in polysaccharide utilization, since the F. johnsoniae Gld proteins are needed for chitin utilization (but not for utilization of other polysaccharides) in addition to gliding motility.

Summary and Future Prospects Genomic Analysis of gld Genes and Gliding Motility

Nearly complete genome sequence information is available for Cytophaga hutchinsonii, a cellulose-degrading gliding bacterium that belongs to the CFB group and is distantly related to F. johnsoniae. C. hutchinsonii displays the same gliding behaviors as F. johnsoniae and has homologs to each of the F. johnsoniae Gld proteins. In contrast, preliminary analysis of the nearly complete genome sequence of M. xanthus reveals few genes that are predicted to encode proteins with significant similarity to F. johnsoniae Gld proteins. Homologs to gldB, gldD, gldG, gldH, and gldI have not been detected by analysis of the M. xanthus genome, suggesting that F. johnsoniae gliding and M. xanthus gliding are probably not genetically closely related. B. thetaiotaomicron, a nonmotile member of the CFB group for which the complete genome sequence is available [Xu et al., 2003] has apparent homologs to GldA, GldB, GldH, GldI, and GldJ but lacks

Cytophaga-Flavobacterium Gliding Motility

Significant progress has been made in the last few years toward identifying genes and proteins required for F. johnsoniae gliding. Several lines of evidence suggest that transporters may play important roles in gliding, but the exact mechanism responsible for cell movement is still uncertain. Identification of the remaining Gld proteins, analysis of their functions, and ultrastructural and behavioral studies of wild-type and mutant cells should help determine the mechanism of Cytophaga-Flavobacterium gliding motility. A number of other questions regarding F. johnsoniae motility also remain unanswered. Motilenonspreading mutants exhibit cell movement but form nonspreading colonies. What, in addition to cell movement, is required for the formation of a spreading colony? Cells of F. johnsoniae display directed movement [Liu and Fridovich, 1996], but the underlying mechanism is not understood. Much is known regarding the mechanism of chemotaxis of swimming bacteria such as E. coli. The gliding bacterium M. xanthus utilizes a chemotactic sig-

J Mol Microbiol Biotechnol 2004;7:63–71

69

nal transduction system whose components are related to those of E. coli [McBride et al., 1989; Ward and Zusman, 1999]. Is the mechanism of F. johnsoniae chemotactic signal transduction also similar to the enteric paradigm? Finally, although we have not encountered much similarity between F. johnsoniae motility genes and those of gliding bacteria outside of the CFB group, it is possible that the underlying mechanisms for cell movement exhibit greater similarity than the proteins that comprise the machineries. Transporters have been linked to gliding in F. johnsoniae [Agarwal et al., 1997; Hunnicutt et al., 2002], M. xanthus [Wolgemuth et al., 2002; Youderian et al., 2003], and P. uncinatum [Hoiczyk and Baumeister, 1998]. Are there other similarities between the gliding

machineries of these organisms? Clearly much remains to be discovered, and many surprises await those who probe these common but poorly understood bacteria.

Acknowledgements This work was supported by a grant from the National Science Foundation (MCB-0130967) and by a Shaw Scientist Award from The Milwaukee Foundation. Preliminary sequence data for C. hutchinsonii were obtained from the DOE Joint Genome Institute at http://jgi.doe.gov. Preliminary genomic sequence data for M. xanthus were obtained from The Institute for Genomic Research website at http://www.tigr.org. I thank E. Leadbetter, T. Braun and S. Nelson for providing unpublished data, and D. Saffarini for careful reading of the manuscript.

References Abbanat DR, Leadbetter ER, Godchaux W III, Escher A: Sulphonolipids are molecular determinants of gliding motility. Nature 1986;324: 367–369. Agarwal S, Hunnicutt DW, McBride MJ: Cloning and characterization of the Flavobacterium johnsoniae (Cytophaga johnsonae) gliding motility gene, gldA. Proc Natl Acad Sci USA 1997; 94:12139–12144. Burchard RP: Gliding motility and taxes; in Rosenberg E (ed): Myxobacteria, development and cell interactions. New York, Springer, 1984, pp 301. Chang LYE, Pate JL, Betzig RJ: Isolation and characterization of nonspreading mutants of the gliding bacterium Cytophaga johnsonae. J Bacteriol 1984;159:26–35. Cho SG, Choi YJ: Characterization of the xaiF gene encoding a novel xylanase-activity-increasing factor, XaiF. J Microbiol Biotechnol 1998;8:378–387. Duxbury T, Humphrey BA, Marshall KC: Continuous observations of bacterial gliding motility in a dialysis microchamber: The effects of inhibitors. Arch Microbiol 1980;124:169–175. Dzink-Fox JL, Leadbetter ER, Godchaux W III: Acetate acts as a protonophore and differentially affects bead movement and cell migration of the gliding bacterium Cytophaga johnsonae (Flavobacterium johnsoniae). Microbiology 1997;143:3693–3701. Gibson MM, Bagga DA, Miller CG, Maguire ME: Magnesium transport in Salmonella typhimurium: The influence of new mutations conferring Co2+ resistance on the CorA Mg2+ transport system. Mol Microbiol 1991;5:2753– 2762. Godchaux W III, Gorski L, Leadbetter ER: Outer membrane polysaccharide deficiency in two nongliding mutants of Cytophaga johnsonae. J Bacteriol 1990;172:1250–1255.

70

Godchaux W III, Leadbetter ER: Sulfonolipids are localized in the outer membrane of the gliding bacterium Cytophaga johnsonae. Arch Microbiol 1988;150:42–47. Godchaux W III, Lynes MA, Leadbetter ER: Defects in gliding motility in mutants of Cytophaga johnsonae lacking a high-molecular-weight cell surface polysaccharide. J Bacteriol 1991; 173:7607–7614. Gorski L, Godchaux W III, Leadbetter ER: Structural specificity of sugars that inhibit gliding motility of Cytophaga johnsonae. Arch Microbiol 1993;160:121–125. Gorski L, Godchaux W III, Leadbetter ER, Wagner RR: Diversity in surface features of Cytophaga johnsonae motility mutants. J Gen Microbiol 1992;138:1767–1772. Harshey RM: Bees aren’t the only ones: Swarming in gram-negative bacteria. Mol Microbiol 1994;13:389–394. Henrichsen J: Twitching motility. Annu Rev Microbiol 1983;37:81–93. Henrichsen J, Blom J: Examination of fimbriation of some gram-negative rods with and without twitching and gliding motility. Acta Pathol Microbiol Scand B1975;83:161–170. Hoiczyk E: Gliding motility in cyanobacteria: Observations and possible explanations. Arch Microbiol 2000;174:11–17. Hoiczyk E, Baumeister W: The junctional pore complex, a prokaryotic secretion organelle, is the molecular motor underlying gliding motility in cyanobacteria. Curr Biol 1998;8:1161– 1168. Holloway P, McCormick W, Watson RJ, Chan YK: Identification and analysis of the dissimilatory nitrous oxide reduction genes, nosRZDFY, of Rhizobium meliloti. J Bacteriol 1996;178: 1505–1514.

J Mol Microbiol Biotechnol 2004;7:63–71

Hunnicutt DW, Kempf MJ, McBride MJ: Mutations in Flavobacterium johnsoniae gldF and gldG disrupt gliding motility and interfere with membrane localization of GldA. J Bacteriol 2002;184:2370–2378. Hunnicutt DW, McBride MJ: Cloning and characterization of the Flavobacterium johnsoniae gliding motility genes, gldB and gldC. J Bacteriol 2000;182:911–918. Hunnicutt DW, McBride MJ: Cloning and characterization of the Flavobacterium johnsoniae gliding motility genes gldD and gldE. J Bacteriol 2001;183:4167–4175. Kempf MJ, McBride MJ: Transposon insertions in the Flavobacterium johnsoniae ftsX gene disrupt gliding motility and cell division. J Bacteriol 2000;182:1671–1679. Kirchman DL: The ecology of Cytophaga-Flavobacteria in aquatic environments. FEMS Microbiol Ecol 2002;39:91–100. Klein C, Entian KD: Genes involved in self-protection against the lantibiotic subtilin produced by Bacillus subtilis ATCC 6633. Appl Environ Microbiol 1994;60:2793–2801. Korolev EV, Nikonov AV, Brudnaya MS, Snigirevskaya ES, Sabinin GV, Komissarchik YY, et al: Tubular structures of Mycoplasma gallisepticum and their possible participation in cell motility. Microbiology 1994;140:671–681. Lapidus IR, Berg HC: Gliding motility of Cytophaga sp. strain U67. J Bacteriol 1982;151:384– 398. Leeuw E de, Graham B, Phillips GJ, ten HagenJohnman CM, Oudega B, Luirink J: Molecular characterization of Escherichia coli FtsE and FtsX. Mol Microbiol 1999;31:983–993. Liu ZX, Fridovich I: Negative chemotaxis in Cytophaga johnsonae. Can J Microbiol 1996;42: 515–518. Lunsdorf H, Schairer HU: Frozen motion of gliding bacteria outlines inherent features of the motility apparatus. Microbiology 2001;147: 939–947.

McBride

McBride MJ: Bacterial gliding motility: Multiple mechanisms for cell movement over surfaces. Annu Rev Microbiol 2001;55:49–75. McBride MJ, Braun TF: GldI is a lipoprotein that is required for Flavobacterium johnsoniae gliding motility and chitin utilization. J Bacteriol 2004;186:2295–2302. McBride MJ, Braun TF, Brust JL: Flavobacterium johnsoniae GldH is a lipoprotein that is required for gliding motility and chitin utilization. J Bacteriol 2003;185:6648–6657. McBride MJ, Kempf MJ: Development of techniques for the genetic manipulation of the gliding bacterium Cytophaga johnsonae. J Bacteriol 1996;178:583–590. McBride MJ, Weinberg RA, Zusman DR: ‘Frizzy’ aggregation genes of the gliding bacterium Myxococcus xanthus show sequence similarities to the chemotaxis genes of enteric bacteria. Proc Natl Acad Sci USA 1989;86:424–428. McGowan SJ, Sebaihia M, O’Leary S, Hardie KR, Williams P, Stewart GS, et al.: Analysis of the carbapenem gene cluster of Erwinia carotovora: Definition of the antibiotic biosynthetic genes and evidence for a novel beta-lactam resistance mechanism. Mol Microbiol 1997;26:545–556. Miyata M, Yamamoto H, Shimizu T, Uenoyama A, Citti C, Rosengarten R: Gliding mutants of Mycoplasma mobile: Relationships between motility and cell morphology, cell adhesion and microcolony formation. Microbiology 2000; 146:1311–1320. Murphy CK, Beckwith J: Export of proteins to the cell envelope in Escherichia coli; in Neidhardt FC (ed): Escherichia coli and Salmonella. Washington, ASM Press, 1996, vol 2, pp 967– 978. Pate JL: Gliding motility. Can J Microbiol 1988; 34:459–465. Pate JL, Chang L-YE: Evidence that gliding motility in prokaryotic cells is driven by rotary assemblies in the cell envelopes. Curr Microbiol 1979;2:59–64.

Cytophaga-Flavobacterium Gliding Motility

Pate JL, De Jong DM: Use of nonmotile mutants to identify a set of membrane proteins related to gliding motility in Cytophaga johnsonae. J Bacteriol 1990;172:3117–3124. Podlesek Z, Comino A, Herzog-Velikonja B, ZgurBertok D, Komel R, Grabnar M: Bacillus licheniformis bacitracin-resistance ABC transporter: Relationship to mammalian multidrug resistance. Mol Microbiol 1995;16:969–976. Reichenbach H: The Order Cytophagales; in Balows A, Truper HG, Dworkin M, Harder W, Schleifer KM (eds): The Prokaryotes. New York, Springer, 1992, pp 3631–3675. Ridgway HF: Source of energy for gliding motility in Flexibacter polymorphus: Effects of metabolic and respiratory inhibitors on gliding movement. J Bacteriol 1977;131:544–556. Ridgway HF, Lewin RA: Goblet-shaped subunits from the wall of a marine gliding microbe. J Gen Microbiol 1973;79:119–128. Ridgway HF, Lewin RA: Characterization of gliding motility in Flexibacter polymorphus. Cell Motility Cytoskeleton 1988;11:46–63. Salyers AA, Reeves A, D’Elia J: Solving the problem of how to eat something as big as yourself: Diverse bacterial strategies for degrading polysaccharides. J Indust Microbiol 1996;17:470– 476. Shipman JA, Berleman JE, Salyers AA: Characterization of four outer membrane proteins involved in binding starch to the cell surface of Bacteroides thetaiotaomicron. J Bacteriol 2000; 182:5365–5372. Spaink HP, Wijfjes AHM, Lugtenberg BJJ: Rhizobium NodI and NodJ proteins play a role in the efficiency of secretion of lipochitin oligosaccharides. J Bacteriol 1995;177:6276–6281. Spormann AM: Gliding motility in bacteria: Insights from studies of Myxococcus xanthus. Microbiol Mol Biol Rev 1999;63:621–641. Ukai H, Matsuzawa H, Ito K, Yamada M, Nishimura A: ftsE(Ts) affects translocation of K+pump proteins into the cytoplasmic membrane of Escherichia coli. J Bacteriol 1998;180:3663– 3670.

Wall D, Kaiser D: Type IV pili and cell motility. Mol Microbiol 1999;32:1–10. Ward MJ, Zusman DR: Motility in Myxococcus xanthus and its role in developmental aggregation. Curr Opin Microbiol 1999;2:624–629. Wolgemuth C, Hoiczyk E, Kaiser D, Oster G: How myxobacteria glide. Curr Biol 2002;12:369– 377. Wolkin RH, Pate JL: Translocation of motile cells of the gliding bacterium Cytophaga johnsonae depends on a surface component that may be modified by sugars. J Gen Microbiol 1984;130: 2651–2669. Wolkin RH, Pate JL: Selection for nonadherent or nonhydrophobic mutants co-selects for nonspreading mutants of Cytophaga johnsonae and other gliding bacteria. J Gen Microbiol 1985;131:737–750. Wolkin RH, Pate JL: Phage adsorption and cell adherence are motility-dependent characteristics of the gliding bacterium Cytophaga johnsonae. J Gen Microbiol 1986;132:355–367. Wu SS, Wu J, Cheng YL, Kaiser D: The pilH gene encodes an ABC transporter homologue required for type IV pilus biogenesis and social gliding motility in Myxococcus xanthus. Mol Microbiol 1998;29:1249–1261. Xu J, Bjursell MK, Himrod J, Deng S, Carmichael LK, Chiang HC, et al.: A genomic view of the human-Bacteroides thetaiotaomicron symbiosis. Science 2003;299:2074–2076. Youderian P: Bacterial motility: Secretory secrets of gliding bacteria. Curr Biol 1998;8:R408– R411. Youderian P, Burke N, White DJ, Hartzell PL: Identification of genes required for adventurous gliding motility in Myxococcus xanthus with the transposable element mariner. Mol Microbiol 2003;49:555–570.

J Mol Microbiol Biotechnol 2004;7:63–71

71

J Mol Microbiol Biotechnol 2004;7:72–77 DOI: 10.1159/000077871

The Junctional Pore Complex and the Propulsion of Bacterial Cells Charles W. Wolgemuth a George Oster b a University of Connecticut Health Center, Department of Physiology, Farmington, Conn., b University of California, Departments of Molecular & Cellular Biology and ESPM, Berkeley, Calif., USA

Key Words Gliding motility W Cyanobacteria W Myxobacteria W Junctional pore complex

Abstract Gliding motility is defined as translocation in the direction of the long axis of the bacterium while in contact with a surface. This definition leaves unspecified any mechanism and, indeed, it appears that there is more than one physiological system underlying the same type of motion. Currently, two distinct mechanisms have been discovered in myxobacteria. One requires the extension, attachment, and retraction of type IV pili to pull the cell forwards. Recent experimental evidence suggests that a second mechanism for gliding motility involves the extrusion of slime from an organelle called the ‘junctional pore complex’. This review discusses the role of slime extrusion and the junctional pore complex in the gliding motility of both cyanobacteria and myxobacteria. Copyright © 2004 S. Karger AG, Basel

Introduction

Prokaryote propulsion is bewilderingly diverse. Many bacteria swim through fluids by rotating a long helical flagellum driven by a rotary motor [Namba and Vonderviszt, 1997; Berg, 2003]. Other bacteria, such as Synechococcus [Waterbury et al., 1985] and Spiroplasma [Gilad et

ABC

© 2004 S. Karger AG, Basel

Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/mmb

al., 2003], are themselves helical and swim using mechanochemical filaments [Trachtenberg et al., 2003; Wolgemuth et al., 2003]. Cells that move on solid surfaces use other mechanisms. One such motion, generically called ‘gliding motility’, refers to translocation in the absence of any visible propulsive organelle. This led Reichenbach [1981] to propose the deliberately vague definition as ‘translocation in the direction of the long axis of the bacterium while in contact with a surface’. Many different bacterial species glide, including cyanobacteria, myxobacteria, flexibacteria, mycoplasmas, Chloroflexacae, and Beggiatoaceae [Burchard, 1981]. Gliding is slow compared to swimming, with average speeds ranging from micrometers per second in Mycoplasma mobile, down to micrometers per minute in Myxococcus xanthus [Spormann and Kaiser, 1995; Hoiczyk, 2000]. Gliding motility exhibits a number of different characteristics [see recent reviews: McBride, 2000, 2001]. For example, some species rotate during translocation [Ladipus and Berg, 1982], others do not; myxobacteria leave telltale slime ‘tracks’ in an agar substratum on which they glide, but mycoplasmas neither secrete slime nor leave any detectable tracks. M. xanthus has been shown to have two genetically distinct mechanisms for gliding. Social (S)-motility requires the extension and retraction of type IV pili [Wall and Kaiser, 1999; Kaiser, 2000]. This type of motility is similar to so-called ‘twitching motility’ observed in many other species. The second type of locomotion, so-called Adventurous (A)-motility, depends on the secretion of slime. Recent experimental evidence strongly suggests that, in both cyanobacteria and myxobacteria, slime extrusion from a sur-

Charles Wolgemuth University of Connecticut Health Center 263 Farmington Avenue Farmington, CT 06030-3305 (USA) Tel. +1 860 679 1655, Fax +1 860 679 1267, E-Mail [email protected]

Fig. 1. Slime extrusion and nozzles in cyanobacteria. a–c Show secreted slime stained with India ink in P. uncinatum. a Immobilized filament showing helically wrapped slime bands. b, c Slime bands shear off the surface when fluid flow is passed over the filaments. Slime emanates from points near septa and elongates at a rate comparable to the gliding rate. a–c The scale bar represents 10 Ìm. d Electron micrograph of a cross-section of P. uncinatum. JPC (arrowheads) are visible near the septum. The scale bar represents 200 nm. e Averages of the side- and top-view projections of the pore complex of P. uncinatum. Reprinted from Hoiczyk and Baumeister [1998], with permission from Elsevier.

face organelle drives gliding locomotion. In the first section, we will discuss the experimental evidence related to slime secretion and the secreting organelle that supports this notion. In the second section, we will describe two recent models that describe this type of motility.

Experimental Findings

Cyanobacteria Cyanobacteria are a morphologically diverse group of phototrophic, gram-negative bacteria. The cell walls of these bacteria are much thicker than the walls of other gram-negative bacteria, ranging from 30 to 700 nm [Hoiczyk and Baumeister, 1995]. Many members of this group of bacteria can glide at speeds of up to 10 Ìm/s. The most efficient gliders among the cyanobacteria are found in the filamentous forms, such as Oscillatoria, Spirulina, Phormidium, and Anabena [Reichenbach, 1981]. Some species, such as P. uncinatum and Oscillatoria, rotate about their long axis while gliding; others, such as A. variabilis, translate laterally [Hoiczyk and Baumeister, 1998]. Other cyanobacteria, such as Synechocystis, move by ‘twitching’, analogous to S-motility in myxobacteria [Bhaya et

Junctional Pore Complex and Propulsion of Bacterial Cells

al., 1999] which is also driven by type IV pili [Wall and Kaiser, 1999]. Cyanobacterial gliders secrete mucilage, or slime, while moving. Hoiczyk and Baumeister [1998] discovered that diluted India ink particles stuck to the secreted mucus in P. uncinatum and A. variabilis filaments; this enabled them to visualize the secretion process using light microscopy (fig. 1a–c). In P. uncinatum, they observed that the cells were either covered in a tight-fitting slime tube or had a number of thin bands of slime that wrapped in a helical fashion over the surface (fig. 1a). Flowing fluid past the bacteria, sheared the slime bands off the surface of the cells except at fixed points located near the septa between cells from which the slime emerged (fig. 1b, c). In A. variabilis, slime was extruded in bands perpendicular to the long axis of the filaments, which also originated near the septa between cells. The slime bands elongated at F 3 Ìm/s, comparable to the speed at which P. uncinatum glides. This strongly suggests that slime extrusion plays an active, rather than passive, role in cyanobacterial gliding motility. Electron microscopic studies on acid-treated, isolated cell walls from over two dozen species of cyanobacteria showed rows of fine pores, 14–16 nm in diameter, clustered circumferentially around the septa (fig. 1d) [Gugliel-

J Mol Microbiol Biotechnol 2004;7:72–77

73

Fig. 2. Slime extrusion and nozzles in myxobacteria. a Fluorescent light micrograph of gliding wild-type M. xanthus DK1622 cell where the slime has been stained with acridine orange. b Electron micrograph of the cell pole of a gliding M. xanthus DK1622 cell showing several slime bands emanating from the cell pole. c Negatively stained electron micrograph of an isolated M. xanthus DK1622 cell envelope showing multiple ring-like structures located predominantly at the poles of the cell. d Gallery of electron micrographs of negatively stained isolated nozzles from M. xanthus DK1622. Reprinted from Wolgemuth et al. [2001], with permission from Elsevier.

mi and Cohen-Bazire, 1982]. However, these pores did not penetrate the entire cyanobacterial cell wall and so were initially disregarded as playing a role in gliding motility [Castenholtz, 1982]. Hoiczyk and Baumeister [1998] later discovered that these pores were actually part of a larger structure 70–80 nm in diameter, about 32 nm long, sufficient to span the entire cell wall. The entire structure is called the junctional pore complex (JPC). It consists of an outer membrane pore complex attached to a channel F13 nm in diameter that spans the peptidoglycan layer. The pore is both mirror and cylindrically symmetric with a bulge in the center. The ends are roughly 8 and the center bulge 14 nm in diameter (fig. 1e). Ring-like structures encompass the pore near the central bulge. In both P. uncinatum and A. variabilis, the JPCs encircle the cell and are located near the septa. The channels formed by the JPCs are inclined at an angle of 30–40° relative to the cell axis, and are oppositely directed on either side of the septum. This angle provides directionality to the exuded slime that propels the cells forward. In P. uncinatum, the pores are aligned in a single row. In A. variabilis, several rows of pores

74

J Mol Microbiol Biotechnol 2004;7:72–77

line both sides of the septum. In some circumstances, prolonged culture leads to the formation of filaments that are non-motile. These filaments do not secrete slime and the JPC organelles disappear leaving behind only the transpeptidoglycan channels [Hoiczyk and Baumeister, 1995]. The outer surface of four different gliding bacteria from the species Oscillatoriaceae consist of parallel, helically arranged protein fibrils [Hoiczyk and Baumeister, 1995]. In Phormidium, the protein that comprises this layer is oscillin, a Ca2+ binding protein that is required for motility [Hoiczyk and Baumeister, 1997]. Hoiczyk and Baumeister [1995, 1998] hypothesized that the surface striations formed by these proteins act as channels for the extruded slime to flow along. Therefore, if the filaments are helically arranged, the cell will rotate as it glides with the handedness of the rotation corresponding to the handedness of the wrapping. If the filaments are aligned radially, the cell will not rotate. In all species studied to date, this correlation is consistent, and provides a structural explanation for why some species rotate as they glide while others do not [Hoiczyk, 1998].

Wolgemuth/Oster

Myxobacteria Myxobacteria are common, rod-shaped, gram-negative bacteria whose natural habitat is soil [Shimkets, 1990; Koch and White, 1998]. As mentioned above, myxobacteria have two propulsive motors. Retraction of type IV pili provides the force that propels S-motility [Wall and Kaiser, 1999; Kaiser, 2000; Merz et al., 2000; Sun et al., 2000; Skerker and Berg, 2001], and A-motility is associated with slime secretion. Gliding of A-motility mutants is quite different from that of S-motility mutants. For example, the gliding direction of wild-type (S+A+) M. xanthus can be altered by deforming the agar substratum, a property called ‘elasticotaxis’ [Stanier, 1942; Fontes and Kaiser, 1999]. However, A-motility mutants (S+A–) do not show such a response. A-motility and Smotility mutants also respond differently to the concentration of agar in the substrate. A-motility shows higher speeds on surfaces that are drier and firmer, whereas Smotility favors softer and wetter surfaces [Shi and Zusman, 1993]. Another marked difference between the two mechanisms is observed in filamentous mutants. A-motility gliding speed is not affected by cell elongation, but Smotility gliding speed is greatly reduced as filamentous cells lengthen [Sun et al., 1999]. If slime secretion from the JPC drives the motility of cyanobacteria, it is likely that the same mechanism provides thrust for A-motility in M. xanthus. Indeed, electron micrographs of whole cells and isolated cell envelopes from M. xanthus show ring-like structures in the cell wall nearly identical to the JPC in cyanobacteria [Wolgemuth et al., 2001] (fig. 2a). The M. xanthus rings have a central core roughly 6.5 nm in diameter with a less dense peripheral zone 12–14 nm in diameter. The rings are located predominantly at the ends of the cells with up to 250 nozzles at each end (fig. 2b). S-motility (S–A+) mutants still possess nozzles, as well as cells carrying a mutation in the mglA gene, the only known gene required for both Amotility and S-motility. Side-view images of the nozzle have not yet been obtained so it is not possible to make a complete morphological comparison between the nozzles in M. xanthus and cyanobacteria. Since S-motility is driven by the retraction of pili, might slime trails deposited by A– mutants be different from those left by S– mutants? Indeed, comparison of slime deposition by wild-type cells with that of A+S– and A–S+ motility mutants shows that both wild-type and the A+S– mutant cells leave slime trails, but an A–S+ mutant (¢cglB) leaves ‘puddles’ of slime around each cell [Wolgemuth et al., 2001]. This suggests that slime is still secreted in this A-motility mutant, but without directionality. To

Junctional Pore Complex and Propulsion of Bacterial Cells

test whether slime secretion occurred at the nozzle sites, acridine orange was used to stain the secreted and deposited slime. As expected, slime trails originated at the rear of cells. M. xanthus cells were also induced to crawl over electron micrograph grids. High magnification visualization showed that slime trails originated from the rear of the cell in numerous small streams. These observations, combined with the location of the ring-like structures in M. xanthus, strongly implicate slime secretion from JPC organelles as the mechanism for A-motility in myxobacteria. This mechanism is also consistent with the observation that gliding speed in A+S– filamentous cells does not depend on the length of the filament. Since the number of cells increases in proportion to the length in filamentous cells, the number of nozzles, and therefore the propulsive force, will also increase in proportion with the length.

Models for A-Motility

The Slime Gun Slime extrusion was proposed as a mechanism for thrust generation in gliding bacteria over 75 years ago [Jahn, 1924; Kuhlwein, 1953]. This model was largely discounted because it was unclear how slime excretion could produce sufficient thrust and directionality to propel the bacteria. Also, estimates of the energetic cost for slime production seemed excessively large. However, slime is a polyelectrolyte gel; that is, a charged, crosslinked polymer mesh embedded in a fluid solvent. Such gels can swell to many hundreds of times their dry volume, and generate enormous swelling forces. Recently, Wolgemuth et al. [2001] proposed a model for force generation based on the hydration of slime (fig. 3a). They proposed that the slime is introduced into the nozzle cavity in a deswelled state, perhaps (as in mucin granules) [Verdugo, 1991] by divalent cations. Fluid perfuses into the nozzle from outside of the cell and hydrates the slime, causing it to swell and extrude from the nozzle. As the gel stream leaves the nozzle it adheres to the substrate. This provides the swelling gel with a footing allowing it to push the cell forward. The mathematical model permitted calculating the force produced within a single nozzle as the slime hydrates and expands. By multiplying by the number of observed nozzles at a cell pole, and comparing this force with an estimate of the drag on a cell, Wolgemuth et al. [2001] were able to show that slime extrusion could produce a propulsive force sufficient to account for the observed gliding speed of both myxobacteria and cyanobacteria, and to predict how the velocity of the bacterium depends on resisting forces.

J Mol Microbiol Biotechnol 2004;7:72–77

75

Fig. 3. a The slime gun. Top: Illustration of

the JPC nozzle with deswelled slime being introduced near the inner membrane. The slime then swells and flows out of the pore. Bottom: Placement of the nozzles in M. xanthus near the end of the cell. Slime flows out of the nozzle, adheres to the surface and pushes the cell forward. b The slime ratchet. Cartoon of a possible mechanism for Amotility in myxobacteria. Helically wrapped filaments are able to contract and expand periodically changing the length of the bacterial cell. Slime is extruded from pores in the bacterial cell wall and adheres to the surface preventing the cell from moving backwards. The cell then expands, pushing itself forward against the adhered slime. Relaxation of the filaments returns the cell to its original size.

This slime propulsion mechanism is consistent with a number of the observed phenomena mentioned previously. Slime Trail Following [Burchard, 1981]. If extruded slime adheres more strongly to itself than to the substratum, a cell that encounters a trail left by another cell will be pivoted into alignment with the existing slime trail and will continue moving along the direction of the trail. Elasticotaxis [Burchard, 1981; Fontes and Kaiser, 1999]. If the slime adheres to the solid part of the agar substrate, alignment of the slime polymer with those of the strain-aligned agar provides more adhesive sites for traction, which favors propulsion along strain lines. Velocity Dependence on the Volume Fraction of the Agar Substrate [Shi and Zusman, 1993]. That there is an optimal volume fraction for the agar substrate suggests two things. First, at low volume fractions, the agar gel is not sufficiently stiff to provide the necessary resistive force. Therefore, the slime pushes against a soft wall that yields, reducing the thrust and hence the cell velocity. At high agar volume fractions, the gliding velocity in M. xanthus also decreases [Shi and Zusman, 1993]. This can be understood since both the agar and the slime are gels, and when in contact, they will tend towards an osmotic equilibrium. If the agar is less hydrated than the slime, it will draw water out of the slime, causing the slime to dehydrate and so generate less propulsive thrust. Etching of the Agar [Burchard, 1981]. Gliding cells ‘etch’ the agar substrate on which they move. This occurs because the osmotic pressure of the gel is greater than that of the agar, so water is drawn out of the agar gel, producing a local depression [Burchard, 1981].

76

J Mol Microbiol Biotechnol 2004;7:72–77

The Slime Ratchet Electron micrographs of Myxococcus cells reveal helical filamentous structures wrapping the periplasmic space [Burchard et al., 1977; Lunsdorf and Reichenbach, 1989; Freese et al., 1997; Lunsdorf and Schairer, 2001]. These filaments have been implicated in gliding motility, and recent experiments suggest that they are responsible for morphological changes, such as twisted forms, in actively gliding myxobacteria; these twisted morphologies are not observed in non-motile mutants [Lunsdorf and Schairer, 2001]. Dynamic oscillatory morphology changes have been shown to drive motility in wall-less mollicute bacteria [Miyata and Uenoyama, 2002; Gilad et al., 2003; Trachtenberg et al., 2003]. A similar mechanism could also drive motility in myxobacteria through a slime ratchet-type mechanism [Wolgemuth et al., 2003]. This model hypothesizes that the mechanochemical cycle of conformational changes in the helical filament subunits, coupled with the cell turgor pressure, cyclically changes the overall cell length. Slime secreted from the rear of the cell adheres to the substrate and holds the cell in place during lengthening of the cell and prevents the cell from moving backward during the contraction phase. Repetition of this process slowly ratchets the cell forward, as illustrated in figure 3b. This ‘slime ratchet’ mechanism can be experimentally distinguished from direct propulsion by slime extrusion by observing the correlations between the front and rear of gliding cells. Perfect correlation implies the cell is moving as a rigid body, and supports the slime nozzle model. Conversely, if the front and rear are cyclically correlated out of phase, then the slime ratchet model is more likely.

Wolgemuth/Oster

Open Questions

Experiments provide strong support for the idea that slime secretion from the JPC organelle is the force-driving motility in many cyanobacteria and A-type motility in myxobacteria. However, the details of the propulsive mechanism are still vague. Here we have suggested two models for slime-associated motility consistent with the available evidence. Experiments that can distinguish between these models are proposed, but remain undone [Wolgemuth et al., 2001; 2003]. A number of other questions also remain. The hydration model suggests specific material qualities for the slime; does the chemical composition of the slime conform to these values? How is slime

imported into the nozzles? How do myxobacteria alternate the active cell end during their periodic reversals? Are the nozzles of M. xanthus morphologically similar in side view to those observed in cyanobacteria? While the fog of mystery surrounding gliding motility is slowly lifting, much remains to be learned before we can declare the mystery solved.

Acknowledgments C.W. was supported by NSF Grant MCB-0327716. G.O. was supported by NIH Grant GM59875-02. The authors would like to thank Dale Kaiser for his comments and insights.

References Berg HC: The rotary motor of bacterial flagella. Annu Rev Biochem 2003;79:19–54. Bhaya D, Watanabe N, Ogawa T, Grossman AR: The role of an alternative sigma factor in motility and pilus formation in the cyanobacteria Synechocystis sp. strain PCC6803. Proc Natl Acad Sci USA 1999;96:3188–3193. Burchard AC, Burchard RP, Kloetzel JA: Intracellular, periodic structures in the gliding bacterium Myxococcus xanthus. J Bacteriol 1977;132: 666–672. Burchard RP: Gliding motility of prokaryotes: Ultrastructure, physiology, and genetics. Annu Rev Microbiol 1981;35:497–529. Castenholtz RW: The movements of cyanobacteria. in Carr NG, Whitton BA (eds): The Biology of Cyanobacteria; London, Blackwell, 1982, pp 413–439. Fontes M, Kaiser D: Myxococcus cells respond to elastic forces in their substrate. Proc Natl Acad Sci USA 1999;96:8052–8057. Freese A, Reichenbach H, Lunsdorf H: Further characterization and in situ localization of chain-like aggregates of the gliding bacteria Myxococcus fulvus and Myxococcus xanthus. J Bacteriol 1997;179:1246–1252. Gilad R, Porat P, Trachtenberg S: Motility modes of Spiroplasma melliferum BC3: A helical, wall-less bacterium driven by a linear motor. Mol Microbiol 2003;47:657–669. Guglielmi G, Cohen-Bazire G: Structure et distribution des pores et des perforations de l’enveloppe de peptidoglycane chez quelques cyanobactéries. Protistologica 1982;18:151–165. Hoiczyk E: Structural and biochemical analysis of the sheath of Phormidium uncinatum. J Bacteriol 1998;180:3923–3932. Hoiczyk E: Gliding motility in cyanobacteria: Observations and possible explanations. Arch Microbiol 2000;174:11–17. Hoiczyk E, Baumeister W: Envelope structure of four gliding filamentous cyanobacteria. J Bacteriol 1995;177:2387–2395.

Junctional Pore Complex and Propulsion of Bacterial Cells

Hoiczyk E, Baumeister W: Oscillin, an extracellular, Ca2+-binding glycoprotein essential for the gliding motility of cyanobacteria. Mol Microbiol 1997;26:699–708. Hoiczyk E, Baumeister W: The junctional pore complex, a prokaryotic secretion organelle, is the molecular motor underlying gliding motility in cyanobacteria. Curr Biol 1998;8:1161– 1168. Hoiczyk E, Hansel A: Cyanobacterial cell walls: News from an unusual prokaryotic envelope. J Bacteriol 2000;182:1191–1199. Jahn E: Beiträge zur botanischen Protistologie. I. Die Polyangiden. Leipzig, Gebrüder Bornträger, 1924. Kaiser D: Bacterial motility: How do pili pull? Curr Biol 2000;10:R777–R780. Koch A, White D: The social lifestyle of myxobacteria. Bioessays 1998;20:1030–1038. Kuhlwein H: Weitere Untersuchungen an Myxobacterien. Arch Mikrobiol 1953;19:365–371. Lapidus IR, Berg HC: Gliding motility of Cytophaga sp. strain U67. J Bacteriol 1982;151:384– 398. Lunsdorf H, Reichenbach H: Ultrastructural details of the apparatus of gliding motility of Myxococcus fulvus (Myxobacterales). J Gen Microbiol 1989;135:1633–1641. Lunsdorf H, Schairer H: Frozen motion of gliding bacteria outlines inherent features of the motility apparatus. Microbiology-UK 2001;147: 939–947. McBride M: Bacterial gliding motility: Mechanisms and mysteries. Am Soc Microbiol News 2000;66:203–210. McBride M: Bacterial gliding motility: Multiple mechanisms for cell movement. Annu Rev Microbiol 2001;55:49–75. Merz A, Sheetz M, So M: Pilus retraction powers bacterial twitching motility. Nature 2000;407: 98–102. Miyata M, Uenoyama A: Movement on the cell surface of the gliding bacterium, Mycoplasma mobile, is limited to its head-like structure. FEMS Microbiol Lett 2002;215:285–289.

Namba K, Vonderviszt F: Molecular architecture of bacterial flagellum. Q Rev Biophys 1997;30: 1–65. Reichenbach H: The taxonomy of the gliding bacteria. Annu Rev Microbiol 1981;35:339–364. Shi W, Zusman DR: The two motility systems of Myxococcus xanthus show different selective advantages on various surfaces. Proc Natl Acad Sci USA 1993;90:3378–3382. Shimkets LJ: Social and developmental biology of the myxobacteria. Microbiol Rev 1990;54: 473–501. Skerker JM, Berg HC: Direct observation of extension and retraction of type IV pili. Proc Natl Acad Sci USA 2001;98:6901–6904. Spormann A, Kaiser D: Gliding Movements in Myxococcus xanthus. J Bacteriol 1995;177: 5846–5852. Stanier R: Elasticotaxis in Myxobacteria. J Bacteriol 1942;44:405–412. Sun H, Yang Z, Shi W: Effect of cellular filamentation on adventurous and social gliding motility of Myxococcus xanthus. Proc Natl Acad Sci USA 1999;96:15178–15183. Sun H, Zusman D, Shi W: Type IV pilus of Myxococcus xanthus is a motility apparatus controlled by the frz chemosensory system. Curr Biol 2000;10:1143–1146. Trachtenberg S, Gilad R, Geffen N: The bacterial linear motor of Spiroplasma melliferum BC3: From single molecules to swimming cells. Mol Microbiol 2003;47:671–697. Verdugo P: Mucin exocytosis. Am Rev Respir Dis 1991;144:533–537. Wall D, Kaiser D: Type IV pili and cell motility. Mol Microbiol 1999;32:1–10. Waterbury JB, Willey JM, Franks DG, Valois FW, Watson SW: A cyanobacterium capable of swimming motility. Science 1985;230:74–76. Wolgemuth C, Hoiczyk E, Kaiser D, Oster G: How myxobacteria glide. Curr Biol 2001;12:369– 377. Wolgemuth CW, Igoshin O, Oster G: The motility of Mollicutes. Biophys J 2003;85:1–15.

J Mol Microbiol Biotechnol 2004;7:72–77

77

J Mol Microbiol Biotechnol 2004;7:78–87 DOI: 10.1159/000077872

Shaping and Moving a Spiroplasma Shlomo Trachtenberg Department of Membrane and Ultrastructure Research, The Hebrew University of Jerusalem-Hadassah Medical School, Jerusalem, Israel

Key Words Bacterial motility W Electron microscopy W Linear motors W Mollicutes W Motor proteins W Spiroplasma

Abstract The Mollicutes (Spiroplasma, Mycoplasma and Acholeplasma) are the most minimal cells known to exist, being the smallest and simplest free-living and self-replicating forms of life. Phylogenetically, the Mollicutes are related to gram-positive bacteria and have evolved, by regressive evolution and genome reduction, from Clostridia. The smallest genome in this group (Mycoplasma genitalium – 5.77 ! 105 bp) is only twice that of a large virus (e.g., Entomopox viruses). The largest Mollicute genome (Spiroplasma LB12 – 2.2 ! 106 bp) is only about half that of, e.g., Escherichia coli. Structurally, the Mollicutes lack cell walls and flagella, but have internal cytoskeletons and are motile and chemotactic. Only a cholesterol-containing unit membrane envelops the cells. No analogs to the bacterial chemotactic and motility (che, mot, fla) genes, genes for a two-component signal transduction system, genes associated with gliding, or genomic homologs for the eukaryotic cytoskeleton and motor proteins were found in the Mollicutes. The Spiroplasmas are unique amongst the Mollicutes in having a well-defined basic helical cell geometry. In this respect, the Spiroplasma cell can, essentially, be viewed as a helical dynamic membranal tube (diameter F0.2 Ìm; equivalent to that

ABC Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

© 2004 S. Karger AG, Basel 1464–1801/04/0072–0078$21.00/0 Accessible online at: www.karger.com/mmb

of one eukaryotic flagellar axoneme or to a bacterial flagellar bundle). A flat cytoskeletal ribbon of parallel fibrils is attached to the inside of the cellular tube. Both tube and cytoskeleton are mutually coiled into a dynamic helix driven by differential length changes of the fibrils, which function as linear motors. The cytoskeletal ribbon follows the shortest (inner) helical line on the inner surface of the cellular tube. Being helical allows for further analytical reduction and consequent structural quantification of Spiroplasma. Of particular importance is the ability to correlate light and electron microscopy data and to calculate the fibril lengths (and corresponding molecular dimensions) in the helical and nonhelical dynamic states. The structural unit of the contractile cytoskeleton is a F50-Ångstrom-wide filament comprised of pairs of the 59-kD fib gene product. The monomers are arranged in pairs with opposite polarities allowing for a F100-Ångstrom-long axial repeat. The functional unit of the contractile cytoskeletal ribbon is a fibril comprised of an aligned pair of filaments. Neighboring repeats form a tetrameric ring with a lateral repeat of F100 Å. The axial length of the rings may shorten by F40%, driving the changes in the fibril lengths and, consequently, helical dynamics. Local length changes result in helical symmetry breaking and nonreciprocating cell movements allowing for net directional displacement. Flexing allows for changes in swimming direction. Copyright © 2004 S. Karger AG, Basel

Shlomo Trachtenberg Department of Membrane and Ultrastructure Research The Hebrew University of Jerusalem-Hadassah Medical School PO Box 12272, Jerusalem 91120 (Israel) Tel. +972 2 6758166, Fax +972 2 6784010, E-Mail [email protected]

Bacterial Motility – Rotary Motors

Cells are commonly driven by either linear (eukaryotes) or rotary (prokaryotes) motors. Most nongliding prokaryotes are flagellated. They swim by means of rigid helical propellers driven by rotary motors. The motor is embedded in the cell envelope and is powered by a proton or a sodium motive force generated across the cell membrane [for reviews, see Berg, 2003; DeRosier, 1998]. The rotation of the motor is converted into linear thrust by the corkscrew-shaped filament (propeller). The propeller is connected to the motor’s drive-shaft via a flexible, slightly curved hook (universal joint) enabling the transmission of force in off-axial directions and the formation of filament bundles [for a review, see Namba and Vonderviszt, 1997]. In some bacteria (e.g., Spirochetes), the flagellum is inserted internally in such a way that the rigid propeller rotates within the periplasmic space, causing the helical bacterial body to roll about its axis and move in the axial direction [Berg, 1976; Goldstein and Charon, 1988; Charon et al., 1991]. Other helical bacteria (e.g., Spirillum) may roll about their axis by flagellar bundles rotating at their poles [Ramia, 1991]. The number of flagella per cell and their distribution vary. Individual flagellar filaments having the same helical sense (hand) and rotating in the same direction form coordinated bundles. The concerted filament bundles may fly apart upon the reversal of the direction of motor rotation. These two alternating states – causing ‘smooth running’ and ‘tumbling’, respectively – result in a random walk in space. The frequency of ‘runs’ is higher in the presence of attractants. The frequency of tumbles is higher in the presence of repellents, thus biasing the random walk and enabling directional tactic responses. The alternations between running and tumbling are possible due to the reversibility of the motor’s direction of rotation, the flexibility of the hook and the dynamic polymorphism of the filament [Asakura, 1970; Calladine, 1983]. Filament polymorphism (i.e., the ability to change reversibly the helical parameters) stems from the conformationally bistable nature of the molecule flagellin from which it self-assembles. The filament is constructed from 11 helical protofilaments, which can switch cooperatively between two conformational states. The switch introduces minute length changes [Trachtenberg and DeRosier, 1991] in the protofilaments. To relieve the stress introduced by mixing protofilaments of different lengths and stabilize the structure, the straight, 11-stranded helical filament bends and twists and deforms into a supercoiled propeller. Progressive length changes of the 11 pro-

Shaping and Moving a Spiroplasma

tofilaments allow for the formation of 12 discrete superhelical polymorphs. Interestingly, spiroplasmas – the subject of this review – exhibit a similar cellular and molecular phenomenology yet they are driven by a completely different mechanism as described below.

Bacterial Motility – Gliding

Nonflagellar motility in bacteria is, usually, a form of ‘gliding’ or ‘twitching’ which involves interactions between the bacterial cell surface and a solid or semisolid substrate. At present, clearly defined structural motor elements found in gliding bacteria are (a) the slime secretion pores in Cyanobacteria [Hoiczyk and Baumeister, 1998; Wolgermuth et al., 2002] in which the excreted slime polymers push the cells linearly forward and (b) the retracting type IV pili which pull gliding and twitching bacteria linearly [for a recent review, see Kaiser, 2000].

Linear Motors in Eukaryotes

In eukaryotes, a variety of linear motor complexes constructed from cytoskeletal elements (actin filaments and microtubules) and motor proteins (mechano-enzymes such as myosin, kinesin, dynein and ncd) drive organelles, cells and whole organisms through ATP or GTP hydrolysis. These complexes function in several ways: (a) by the cytoskeletal elements sliding linearly past each other due to molecular switching in the motor domains of the mechano-enzymes and consequent force generation; (b) the motor proteins themselves – utilizing their switchable domains – can move directionally along the cytoskeletal elements (which function as tracks) carrying, e.g., membranal vesicles; (c) linear force generation and movement can also be accomplished by polarity-controlled polymerization of G-actin and tubulin and depolymerization of actin filaments and microtubules [for a review, see Bray, 2001]. No rotary motors involved directly in cell motility are known in eukaryotes.

Bacterial Cytoskeletons

A characteristic of bacteria was believed to be the lack of either a unique or prokaryote-like internal cytoskeleton. Recently, analogs of actin, such as MreB [van den Ent et al., 2001; Jones et al., 2001], ParM [van den Ent et al.,

J Mol Microbiol Biotechnol 2004;7:78–87

79

2002] tubulin (FtsZ) [Lowe and Amos, 1998; Ben-Yehuda and Losick, 2002] and intermediate filaments [Ausmees et al., 2003] were found in bacterial cells. A network of presumably EF-TU filaments was recently described in a gliding Mycoplasma [Mayer, 2003]. However, none of these cytoskeleton-like assemblies is involved in cell motility and swimming except in the Mollicute Spiroplasma, which swims by means of a contractile cytoskeleton functioning as a linear motor, as described below.

The Mollicutes

The Mollicutes (Spiroplasma, Mycoplasma and Acholeplasma) are the most minimal cells known to exist being the smallest and simplest free-living and self-replicating forms of life. Phylogenetically, the Mollicutes are related to gram-positive bacteria and have evolved, by regressive evolution and genome reduction, from Clostridia. The smallest genome in this group (Mycoplasma genitalium – 5.77 ! 105 bp) is only twice that of a large virus (e.g., Entomopox viruses). The largest Mollicute genome (Spiroplasma LB12 – 2.2 ! 106 bp) is only about half that of, e.g., Escherichia coli [for reviews, see Bové, 1993; Razin et al., 1998]. Structurally, the Mollicutes are unique bacteria: they lack cell walls and flagella but have internal cytoskeletons and are motile and chemotactic. Only a cholesterol-containing unit membrane envelops the cells. Cholesterol, uncommon in bacteria, increases the rigidity of the membrane. The Mollicute cells are well defined in shape and exhibit a rich repertoire of cell movements and dynamic morphologies [for reviews, see Kirchhoff, 1992; Trachtenberg, 1998, 2003]. The Mycoplasmas glide [Miyata et al., 2002; Wolgemuth et al., 2003] and the Spiroplasmas are active polymorphic and chemotactic swimmers [Gilad et al., 2002; Berg, 2002; Wolgemuth et al., 2003]. No analogs to the bacterial chemotactic and motility (che, mot, fla) genes and genes associated with gliding were found in the Mollicute genome. Similarly, no genomic homologs for the eukaryotic cytoskeleton and motor proteins were found by whole genome sequencing of Mollicutes [Fraser et al., 1998; Himmelreich et al., 1997; Glass et al., 2000; Chambaud et al., 2001; Sasaki et al., 2002]. The sequenced Spiroplasma cytoskeletal and associated genes [fib – Williamson et al., 1991; scm1 – Jacob et al., 1997] are completely unrelated to those of eukaryotes and have no homologs in other prokaryotes. Historically, Spiroplasma was first described by Davis et al. [1972], Davis and Worley [1973] and Saglio et al.

80

J Mol Microbiol Biotechnol 2004;7:78–87

[1973]. Williamson [1974] reported the isolation of cytoskeletons upon cell lysis and sodium deoxycholate extraction of Spiroplasma isolated from Drosophila. Similar cytoskeletal structures were detergent extracted from Spiroplasma citri and Spiroplasma kunkelii [Williamson and Whitcomb, 1974]. Cytoskeletons were also released from Spiroplasma cells by repeated freezing and thawing [Townsend et al., 1980]. Townsend et al. [1980], Davis et al., [1981]; Charbonneau and Ghiorse, [1984]; Schmitt et al. [1984] and Williamson et al. [1984] applied standard electron microscopy techniques to both whole and lysed cells as well as to purified cytoskeletons of Spiroplasma melliferum BC3 and Spiroplasma floricola OMBG.

The Underlying Helical Geometry of Spiroplasma

Spiroplasmas are, most likely, the simplest free-living and self-replicating known helical cells. Having a greatly reduced genome minimizes the number of functional components in the cell. Being helical allows for further analytical reduction and consequent structural quantification. Of particular importance is the ability to correlate light and electron microscopy data and to calculate the generator lengths of the contractile cytoskeleton (and corresponding molecular dimensions) in the helical and nonhelical states, as described below. The Spiroplasmas are unique amongst the Mollicutes in having a well-defined basic helical cell geometry (fig. 1A, B). In this respect, the Spiroplasma cell can, essentially, be viewed as a helical dynamic membranal tube (diameter F0.2 Ìm; equivalent to that of one eukaryotic flagellar axoneme or to a bacterial flagellar bundle). A flat cytoskeletal ribbon (fig. 1B, C) of parallel fibrils (fig. 1C, D) is attached to the cellular tube. Both tube and cytoskeleton are mutually coiled into a dynamic helix (fig. 2). The cytoskeletal ribbon follows the shortest (inner) helical line on the inner surface of the cellular tube (fig. 1B). One repeat of a dynamic helical tube of diameter d is shown in figure 2. The tube represents the actual average helical repeat (P = 0.87 Ìm), outer coil diameter (D + d = 0.57 Ìm), cell diameter (d = 0.2 Ìm), centerline diameter (D = 0.37 Ìm) and length (L = 1.6 Ìm) of a Spiroplasma population. An average cell has 4–5 repeats. In the figure, a straight cylinder of length L (left) is gradually coiled into a tight coil (right). The coiling is reversible.

Trachtenberg

Fig. 1. A A dark-field video image (with reversed contrast) of a Spi-

roplasma melliferum BC3 cell. The image is a single frame from a video movie. The cell is straight, entirely in the focal plane and normal to the optical axis of the microscope. The helical parameters are labeled in figure 2. B A thin, longitudinal section of a S. melliferum BC3 cell. Cell pellets were impact-frozen and freeze-substituted in acetone containing 2% uranyl acetate. The cytoplasm and the membranes are extracted under these conditions leaving behind only ribosomes and the cytoskeleton. C A cytoskeleton of S. melliferum BC3

obtained by on-the-grid lysis and detergent extraction of whole cells. The ribbon is flat, monolayered and composed of parallel fibrils with ring-like substructures (see D). D A single, isolated, negatively stained fibril from a flat, seven-membered ribbon. The axial length of subunits along the fiber varies from nearly circular to nearly flattened (see fig. 7). The inserts are of class averages of these extreme states. See Trachtenberg [1998] and Trachtenberg and Gilad [2001] for further details.

Fig. 2. The range of helical states and reciprocating (‘breathing’) motion of a helix, one repeat long, between a straight

cylinder of length L and a fully contracted helix (P = d ) depicted in three dimensions. Fifteen intermediate steps are sampled between a fully expanded (D = d ) and fully contracted helix. P = pitch; d = cell diameter; D = centerline diameter and D + d = outer diameter of the coiled cell are marked. To swim, the cell has to deviate from helical symmetry and reciprocating movements. This is achieved through local dynamic deformations [see Gilad et al., 2003].

Shaping and Moving a Spiroplasma

J Mol Microbiol Biotechnol 2004;7:78–87

81

Fig. 3. The relation between the helical repeat, P, and diameter, D, for a fixed centerline of length L (L2 = P2 + (D)2). This relation underlies the reciprocating movement shown in figure 2. For a full account see Trachtenberg et al. [2003b].

In a helical line of length L, P and D are related (fig. 3) by helical symmetry: L2 = P2 + (D)2. The geometrical limits of a fully expanded and fully contracted tube of diameter d (constructed about a centerline D) in which helical symmetry is maintained are D = d and P = d, respectively. Stretching beyond D = d and straightening requires deviation from helical symmetry. Contracting beyond P = d is not possible due to spatial constraints. For a full analytical account see Trachtenberg et al. [2003b].

The Phenomenology of Spiroplasma Swimming and Chemotaxis

Pure, dynamic helical symmetry, as described in the previous section (e.g., linear extension and contraction; fig. 1A, 2), would not allow for motility. Reciprocal shape changes in a low Reynolds number environment cannot generate net cell displacement [Purcell, 1977, 1997]. Therefore, for a generation of directional motility and, equally important, direction changes (the combination of which underlies the chemotactic response), mechanisms of breaking helical symmetry have to be employed [Gilad et al., 2003; Trachtenberg and Geffen, unpubl. data]. Besides regular shape changes, complying to a good approximation with helical symmetry, Spiroplasma were observed to deviate from helical symmetry in the following regular ways: (a) Propagating helical waves – a bulge formed on one of the helical turns propagates towards the cell end and causes the cell to roll about its axis. (b) Propa-

82

J Mol Microbiol Biotechnol 2004;7:78–87

Fig. 4. The underlying principle of coiling a straight cylindrical tube into a helix by means of differentially contractile generators. A A straight tube with two straight, identical generators (L). B The generators have isometrically shortened (S) and, being identical, caused the tube to bend in the plane, towards the viewer. The tube would have bent symmetrically away from the viewer if the generators, positioned where they are, had lengthened. If the length of only one generator changes relative to the other (C), both bending and twisting forces are introduced and the flexible tube relieves the stress by coiling in the direction (hand) of the shorter generator. To maintain regularity of shape and symmetry in complex systems, the active generators have to be clustered and their length has to change in a coordinated manner. L = Long; S = short.

gating a local change of helical sense – the relatively straight segment moving towards one of the cell’s ends is flanked by helical cell segments of opposite hand. (c) Flexing – sharp, random bending about an arbitrary point along the cell’s body. Flexing causes the cell to change its swimming direction. Thus, by temporary deviations from helical symmetry (fig. 2), the cell is capable of ‘running’ and ‘tumbling’, allowing for chemotactic and viscotactic responses [Gilad et al., 2003; Trachtenberg, 2003]. The frequency of flexing is well correlated with the presence and gradient of repellents. In the presence of attractants, the cells tend to straighten, rotate about their axis and align with the gradient [Daniels et al., 1980; Daniels and Longland, 1984]. All of the local deviations from helical symmetry comply with the allowed spectrum of generator lengths and molecular dimensions, as described below [Trachtenberg and Geffen, unpubl. results].

Coiling a Spiroplasma – Differential Length Changes of Contractile Generators

How can such a dynamic helix be generated from a straight membranal tube? Figure 4 depicts schematically the principle of introducing bending and twisting, due to differential length

Trachtenberg

changes of surface lines (generators), underlying this phenomenon. If the generators are of equal length (L, fig. 4A), the membranal cylinder is straight. If both lines shorten isometrically (S, fig. 4B), the membranal tube bends in the plane towards the viewer. If only one of the lines shortens (S, fig. 4C), the membranal tube both bends and twists in a coordinated manner to form a coil. The relative position of the short generator determines the handedness of the coil and the location of its shortest helical line. From a biological point of view, what may determine the number of generators is apparently the dimensional range of a unit motor contraction and the force it can generate. The abstract generators described above are, in fact, contractile linear polymers assembled into a flat, stable membrane-associated ribbon, as described below.

A Contractile Linear Motor – The Cellular and Molecular Machinery

fluid extracted the membrane and cytoplasm, leaving behind only the cytoskeleton and ribosomes. Note that the cytoskeleton is a flat, monolayered helical ribbon comprised of parallel fibrils. The cytoskeleton can be isolated nearly intact from single cells, as seen in figure 1C. In the negatively stained ribbon, clear parallel fibrils with regular repeats can be identified. Seven fibrils are usually observed in such specimens, particularly when the ribbon is folded. An isolated fibril, the functional building block of the cytoskeletal ribbon, purified from the bulk, is shown in figure 1D. Note the ring-like structure of the repeating units of the fibril and variability in their axial length (see also fig. 7, 8). The insert shows class averages (obtained by correspondence analysis) of a circular ring (top) and flattened ring (bottom).

Generating a Dynamic Helical Tube – Fibril Length and Molecular Dimensions

The largely reduced genome and simplicity of the cell leave only one prominent interpretable feature (besides the membrane and ribosomal particles and clusters in the cytoplasm) to be correlated with cell shape determination and motility – the cytoskeleton. The Spiroplasma cytoskeleton amounts to F1% of the total cellular proteins [Townsend et al., 1980]. The building block of the cytoskeletal fibrils in S. citri is a 59-kD protein composed of 515 amino acids – a product of the fib gene [Williamson et al., 1991]. The Spiroplasma cytoskeleton is, apparently, both unique and largely conserved within the group: antibodies raised against S. melliferum BC3 cytoskeletons react with fibrils from 13 Spiroplasmas [Townsend and Archer, 1983] but do not react with Mycoplasma and Acholeplasma cytoskeletons [Williamson et al., 1984]. The helical geometry, which is well maintained for brief instances during cell motility, allows quantifying and correlating video light microscopy of live cells and electron microscopy of subcellular and molecular components of the cytoskeleton [Trachtenberg and Gilad, 2001; Trachtenberg et al., 2003a, b]. This is demonstrated by a progressive reduction of a cell to its molecular components in figure 1. A live, video image of a straight, entirely in-focus helical cell is shown in figure 1A (the corresponding helical parameters are labeled in fig. 2). A longitudinal section through a cryofixed and freezesubstituted cell is shown in figure 1B. The substitution

The population of helical Spiroplasma cells can be divided into size groups based on the volume of one helical repeat, assuming that the cytoplasm is an incompressible viscid fluid. We found in stationary populations F14 size groups. The helical geometry of the cells allows to calculate from direct measurements of P and D + d (light microscopy) and ! d 1 (electron microscopy), the length of each fibril in the cytoskeletal ribbon (fig. 1) for each size group/helical state. Figure 5 depicts the actual lengths (Lg) for each of the seven fibrils in each size group at each of the helical states between D = d and P = d. The helical states are determined arbitrarily by D (the corresponding P, for a given L, can be determined from fig. 3). The shortest generator coincides with the shortest helical line. Note that the actual differential length changes of the generators required to generate the entire spectrum of helical states are very small. Propagating local, rather than uniform length changes generates nonhelical dynamics. How may the differential length changes of the fibrils, required to drive helical dynamics, actually be achieved? The number of tetramer-like rings (fig. 1D) per unit length of fibril is fixed. A fully expanded tetramer (in a cylindrical tube) is F100 Å in diameter (fig. 8B) allowing for F100 tetramers/Ìm. For each size group, one can calculate the length of the shortest fibril in the most contracted helical state (P = d; fig. 5, square) and the longest fibril at the most extended helical state (D = d; fig. 5, triangle). The number of fully extended tetramers per repeat, L, can be estimated from

Shaping and Moving a Spiroplasma

J Mol Microbiol Biotechnol 2004;7:78–87

83

Lg 2.15 2.10 2.05 2.00 1.95 1.90 1.85 1.80 1.75 1.70 1.65 1.60 1.55 1.50 1.45 1.40 1.35 1.30

6

1.25 1.20 1.15 1.10 1.05 1.00 0.95 0.90 0.85 0.80 0.75 0.70 0.65 0.60

Lcyl

0.55 0.50

Lshortest

0.45

Llongest 0.70

0.65

0.60

0.55

0.50

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.40

D

5

the most extended (cylindrical; fig. 5, circle) state. These data are marked, for the median size/age group, in figure 5 and provide the allowed range of dimensional changes. One can now estimate the minor axis lengths of each tetramer along the fibrils required to generate the corre-

84

J Mol Microbiol Biotechnol 2004;7:78–87

Fig. 5. The relation between helical state (L, P and D combinations; see fig. 3) and generators’ length. The length, Lg (Ìm), per one helical repeat (see fig. 2) of the seven generators (see fig. 1) is shown for each sampled helical state between P = d and D = d (as a function of D; see fig. 3). Each cluster represents one of the 14 helical size/age families (based on constant cell volume assuming that the cytoplasm is an incompressible viscid fluid) found in the cell populations analyzed. Note the small differences in length. The corresponding repeat, P, for each D sampled can be found from a curve, such as in figure 3, for the corresponding L. The extreme values, at Lcyl (straight cylinder), Lshortest (the shortest generator at P = d ) and Llongest (longest generator at D = d ) are marked for size/age group 8 (the median group). The events in the gap on the left-hand side of the figure (straightening and hand switching) are nonhelical [see Trachtenberg et al., 2003b]. Fig. 6. The spectrum of generator lengths (Ìm) in the cell population studied. Straight cylinder (Lcyl), Llongest – longest fibril at the fully extended helical state (D = d ) and Lshortest – shortest fibril at the fully contracted helical state (P = d ). Llongest – Lshortest (x) and Lcyl – Lshortest (+) are shown as well. This spectrum reflects the switching capacity of the fib subunits. See also figure 7 for the corresponding tetramer axial dimensions.

sponding fibril length changes by isometric contraction. These data, regarding the extreme fibril lengths and the range of required changes in fibril lengths (x and +), are depicted in figure 6 for each of the 14 size groups. The spectrum of actual minor axis lengths of tetramers (and their frequencies), measured directly from isolated

Trachtenberg

200

15 14 13 12 11 10 9 8

100 7

Size group

Measurement number

150

6 5 4

50

3 2 1 0

0 40

50

60

70 Length

80

Fig. 7. The range of axial tetrameric ring dimensions and their relation to the range of fibril lengths. Filled circles presented in ascending order depict accumulated bulk data, obtained by direct measurements on electron micrographs of isolated fibrils. Fibril extreme length data, calculated from helical parameters obtained by darkfield light microscopy, are shown for the upper (Llongest; triangles) and lower (Lshortest; squares) limits within the helical range of each of the 14 size/age groups (see fig. 5). The dashed horizontal lines cover the axial length range from a straight cylinder (100 Å; Lcyl) (see fig. 5, 6) through an extended helix (triangle) to a tightly coiled (square) one (see fig. 2, 3).

fibrils (fig. 1D), is depicted as filled circles in figure 7. The corresponding extreme tetrameric axial lengths, calculated for each size group from the data in figures 5 and 6, are shown for the helical states as squares and triangles. The cylindrical state is presented by the right-hand axis. Note that the calculated and directly measured spectra are congruent, allowing to postulate that fibril length changes – and the corresponding helical geometries observed – are driven by molecular switching of the 59-kD tetrameric subunits.

Shaping and Moving a Spiroplasma

90

100

Fig. 8. A A model Spiroplasma with helical parameters of an average cell (see fig. 1A, 5) at a nearly stretched (D = d ) state. The flat cytoskeletal ribbon is nearly straight and follows the shortest helical line inside the coiled membranal tube (see fig. 1B). B A model of a functional fibril. The fibril constitutes a pair of filaments comprised of fib monomers (gray circles) F50 Å in diameter arranged in pairs of opposite polarities (inner dots). Neighboring pairs from each filament form tetrameric rings (large circles) F100 Å in diameter as seen in figure 1C and D. For a full account, see Trachtenberg et al. [2003b].

A Functional Model

An average Spiroplasma cell is modeled in figure 8A. The cell has 5 helical repeats and is nearly fully stretched. The nearly straight cytoskeletal ribbon is depicted in black following the innermost (shortest) helical line [Trachtenberg and Gilad, 2001; Trachtenberg et al., 2003a]. The ribbon is a monomolecular layer comprised of 7 fibrils and attached to the inner surface of the cell membrane. A schematic fibril is depicted in figure 8B based on diffraction and mass distribution analysis [Trachtenberg et al., 2003b]. The structural unit is a F50-angstrom-wide

J Mol Microbiol Biotechnol 2004;7:78–87

85

filament comprised of pairs of the 59-kD fib gene product. The monomers (gray circles) are arranged in pairs with opposite polarities (gray dots within each gray circle) allowing for a F100-angstrom axial repeat. The functional unit of the contractile cytoskeletal ribbon is a fibril comprised of an aligned pair of filaments. Neighboring repeats form a tetrameric ring (large circle) with a lateral repeat of F100 Å. The flatness on the monomolecular thick ribbon maximizes the number of attachment sites with the membrane and allows for simple and reversible hand switching. The position of the ribbon along the shortest helical line in the cellular tube maximizes the efficiency of the motor, as the smallest changes in length in this circumferential location have the largest effect on cell shape. The cytoskeletal ribbon is a stable structure highly resistant to depolymerization. The fibrils seem to be tight-

ly and strongly connected to each other. Even at extreme states of curvature, the fibrils are well aligned and in register. This suggests that whatever the discrete mechanism of molecular switching might be, the axial size range of the tetrameric units in the ribbon is practically continuous. One may view the cytoskeletal ribbon as a functional, membrane-associated analog of an unrolled flagellar filament. In practice, the whole cell functions as a helical propeller.

Acknowledgment This project is supported by funds from the Israel Science Foundation.

References Asakura S: Polymerization of flagellin and polymorphism of flagella. Adv Biophys 1970;1:99– 155. Ausmees N, Kuhn JR, Jacobs-Wagner C: The bacterial cytoskeleton: An intermediate filamerlike function in cell shape. Cell 2003;115:706– 713. Ben-Yehuda S, Losick R: Asymmetric cell division in B. subtilis involves a spiral-like intermediate of the cytokinetic protein FtsZ. Cell 2002;109: 257–266. Berg HC: How spirochetes may swim. J Theor Biol 1976;56:269–273. Berg HC: How Spiroplasma might swim. J Bacteriol 2002;184:2063–2064. Berg HC: The rotary motor of bacterial flagella. Annu Rev Biochem 2003;72:19–54. Bové JM: Molecular features of mollicutes. Clin Infect Dis 1993;17:S10–S31. Bray D: Cell Movements: From Molecules to Motility, ed 2. New York, Garland, 2001. Calladine CR: Construction and operation of bacterial flagella. Sci Prog 1983;68:365–385. Chambaud I, Heilig R, Ferris S, Barbe V, Samson D, Galisson F, Moszer I, Dybvig K, Wroblewski H, Viari A, Rocha EPC, Blanchard A: The complete genome sequence of the murine respiratory pathogen Mycoplasma pulmonis. Nucleic Acid Res 2001;29:2145–2153. Charbonneau DL, Ghiorse WC: Ultrastructure and location of cytoplasmic fibrils in Spiroplasma floricola OBMG. Curr Microbiol 1984;10:65– 72. Charon NW, Goldstein SF, Curci K, Limberger RJ: The bent-end morphology of Treponema phagedenis is associated with short, left-handed, periplasmic flagella. J Bacteriol 1991;173: 4820–4826.

86

Daniels MJ, Longland JM: Chemotactic behavior of spiroplasmas. Curr Microbiol 1984;10:191– 194. Daniels MJ, Longland JM, Gilbart J: Aspects of motility and chemotaxis in spiroplasmas. J Gen Microbiol 1980;118:429–436. Davis RE, Lee, IM, Worley JF: Spiroplasma floricola, a new species isolated from surfaces of flowers of the tulip tree, Liriodendron tulipifera. Int J Sys Bacteriol 1981;31:456–464. Davis RE, Worley, JF: Spiroplasma: Motile, helical microorganisms associated with corn stunt disease. Phytopatholgy 1973;63:403–408. Davis, RE, Worley JF, Whitcomb RF, Ishijima T, Steere RL: Helical filaments produced by a mycoplasma-like organism associated with corn stunt disease. Science 1972;176:521–523. DeRosier DJ: The turn of the screw: The bacterial flagellar motor. Cell 1998;93:17–20. Fraser CM, Gocayne JD, White O, Adams MD, Clayton RA, Fleischmann RD, Bult CJ, Kerlavage AR, Sutton G, Kelly JM, Fritchman JL, Weidman JF, Small KV, Sandusky M, Fuhrmann J, Nguyen D, Utterback TR, Saudek DM, Phillips CA, Merrick JM, Tomb J-F, Daugherty BA, Bott KF, Hu P-C, Lucier TS, Petterson SN, Smith HO, Huchison CA III, Venter JC: The minimal gene complement of Mycoplasma genitalium, Science 1998;270: 397–403. Gilad R, Porat A, Trachtenberg S: Motility modes of Spiroplasma melliferum BC3: A helical, wall-less bacterium driven by a linear motor. Mol Microbiol 2003;47:657–669. Glass JI, Lefkowitz EJ, Glass JS, Heimer CR, Chen EY, Cassel GH: The complete sequence of the mucosal pathogen Ureaplasma urealyticum. Nature 2000;407:757–762.

J Mol Microbiol Biotechnol 2004;7:78–87

Goldstein SF, Charon NW: Motility of the spirochete Leptospira. Cell Motil Cytoskeleton 1988;9:101–110. Himmelreich R, Plagens H, Hilbert H, Reiner B, Herrmann R: Comparative analysis of the genomes of the bacteria Mycoplasma pneumoniae and Mycoplasma genitalium, Nucleic Acids Res 1997;25:701–712. Hoiczyk E, Baumeister W: The junctional pore complex, a prokaryotic secretion organelle, is the molecular motor underlying gliding motility in cyanobacteria. Curr Biol 1998;8:1161– 1168. Jacob C, Nouzieres F, Duret S, Bové J-M, Renaudin J: Isolation, characterization and complementation of a motility mutant of Spiroplasma citri. J Bacteriol 1997;179:4802–4810. Jones LJF, Carballido-Lopez R, Errington J: Control of cell shape in bacteria: Helical, actin-like filaments in Bacillus subtilis. Cell 2001;104: 913–922. Kaiser, D: How do pili pull? Curr Biol 2000;10: R777–R780. Kirchhoff H: Motility; in Maniloff J, McElhaney RN, Finch LR, Baseman JB (eds): Mycoplasmas: Molecular Biology and Pathogenesis. Washington, American Society for Microbiology. 1992, pp 289–306. Loewe J, Amos LA: Crystal structure of the bacterial cell-division protein FtsZ. Nature 1998; 391:203–206. Mayer F: Cytoskeletons in prokaryotes. Cell Biol Int 2003;27:429–438. Miyata M, Ryu W, Berg HC: Force and velocity of Mycoplasma mobile. J Bacteriol 2002;184: 1827–1832. Namba K, Vonderviszt F: Molecular architecture of bacterial flagellum. Q Rev Biophys 1997;30: 1–65.

Trachtenberg

Purcell EM: Life at low Reynolds number. Am J Physics 1977;45:3–11. Purcell EM: The efficiency of propulsion by a rotating flagellum. Proc Natl Acad Sci USA 1997; 94:11307–11311. Ramia M: Numerical models for the locomotion of Spirilla. Biophys J 1991;60:1057–1078. Razin S, Yogev D, Naot J: Molecular biology and pathogenicity of mycoplasmas. Microbiol Mol Biol Rev 1998;624:1094–1156. Saglio P, L’Hospital M, Laflèche D, Dupont G, Bové JM, Tully JG, Freundt EA: Spiroplasma citri: A mycoplasma-like organism associated with ‘stubborn’ disease of citrus. Int J Syst Bacteriol 1973;23:191–204. Sasaki Y, Ishikawa J, Yamashita A, Oshima K, Kenri T, Furuya K, Yoshino C, Horino A, Shiba T, Sasaki and T, Hattori M: The complete genomic sequence of Mycoplasma penetrans, an intracellular bacterial pathogen in humans. Nucleic Acids Res 2002;30:5293–5300. Schmitt U, Petzold H, Marwitz R: An in situ freeze-fracture study of Spiroplasma citri and the corn stunt Spiroplasma. J Phytopathol 1984;111:297–304. Townsend R, Archer DB: A fibril protein antigen specific to Spiroplasma. J Gen Microbiol 1983; 129:199–206.

Shaping and Moving a Spiroplasma

Townsend R, Archer DB, Plaskitt KA: Purification and preliminary characterization of Spiroplasma fibrils. J Bacteriol 1980;142:694–700. Trachtenberg, S: Mollicutes – wall-less bacteria with internal cytoskeletons. J Struct Biol 1998; 124:244–256. Trachtenberg, S: Spiroplasma motility and chemotaxis: Molecular aspects of cell behavior; in Browning G, Blanchard A (eds); Mycoplasmas: Pathogenesis, Molecular Biology, and Emerging Strategies for Control. Norfolk Horizon Scientific Press, in press. Trachtenberg S, Andrews SB, Leapman RD: Mass distribution and spatial organization of the linear bacterial motor of Spiroplasma citri R2A8. J Bacteriol 2003a;185:1987–1994. Trachtenberg S, DeRosier DJ: A molecular switch: Subunit rotations involved in the right-handed to left-handed transitions of Salmonella typhimurium flagellar filaments. J Mol Biol 1991; 220:67–77. Trachtenberg S, Gilad R: A bacterial linear motor: Cellular and molecular organization of the contractile cytoskeleton of the helical bacterium Spiroplasma melliferum. Mol Microbiol 2001; 41:827–848. Trachtenberg S, Gilad R, Geffen N: The bacterial linear motor of Spiroplasma melliferum BC3: From single molecules to swimming cells. Mol Microbiol 2003b;47:671–697.

Van den Ent F, Amos LA, Loewe J: Prokaryotic origin of the actin cytoskeleton. Nature 2001;413: 39–44. Van den Ent F, Moelle-Jensen J, Amos LA, Gerdes K, Loewe J: F-actin-like filaments formed by plasmid segregation protein ParM. EMBO J 2002;21:6935–6943. Williamson DL: Unusual fibrils from the spirochete-like sex ratio organism. J Bacteriol 1974; 117:904–906. Williamson DL, Brink PR, Zieve GW: Spiroplasma fibrils. Isr J Med Sci 1984;20:830–835. Williamson DL, Renaudin J, Bové JM: Nucleotide sequence of the Spiroplasma citri fibril protein gene. J Bacteriol 1991;173:4353–4362. Williamson DL, Whitcomb RF: Helical, wall-free prokaryotes in Drosophila, leafhoppers and plants; in Bové JM, Duplan JF (eds): Les mycoplasmas de l’homme, des animaux, des végétaux et des insectes. Paris, INSERM, 1974, pp 283–290. Wolgemuth C, Hoiczyk E, Kaiser, D, Oster G: How myxobacteria glide. Curr Biol 2002;12:369– 377. Wolgemuth C, Igoshin O, Oster G: The motility of mollicutes. Biophys J 2003;85:828–842.

J Mol Microbiol Biotechnol 2004;7:78–87

87

Author Index Vol. 7, No. 1–2, 2004

Bardy, S.L. 41 Jarrell, K.F. 41 Kaiser, D. 52 Limberger, R.J. 30 McBridge, M.J. 63 McCarter, L.L. 18 Minamino, T. 5 Namba, K. 5 Ng, S.Y.M. 41 Nudleman, E. 52 Oster, G. 72 Trachtenberg, S. 78 Wolgemuth, C.W. 72

Subject Index Vol. 7, No. 1–2, 2004

Assembly 5 ATPase 5 Bacterial motility 78 Biofilms 52 Capping structure 5 Chaperone 5 Chitin 63 Cyanobacteria 72 Cytophaga 63 Electron microscopy 78 Fibrils 52 Flagellum 5, 18, 30 Flavobacterium 63 Glycosylation 41 Junctional pore complex 72 Linear motors 78 Methanococcus voltae 41 Mollicutes 78

ABC

© 2004 S. Karger AG, Basel

Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/mmb

Motility 5, 18, 30, 41, 52, 63, 72 Motor proteins 78 Myxobacteria 72 Periplasm 30 PilT, PilQ 52 Preflagellin peptidase 41 Secretins 52 Spirochete 50 Spiroplasma 78 Structure 30 Substrate specificity switch 5 Swarming 18 Swimming 18 Transporter 63 Type III protein export 5 Type IV pili 41 Vibrio parahaemolyticus 18