Download pdf - Cold Spring Harb Perspect Biol-2010-Shaevitz

14, 20102010; doi: 10.1101/cshperspect.a000364 originally published online JulyCold Spring Harb Perspect Biol

Joshua W. Shaevitz and Zemer Gitai

The Structure and Function of Bacterial Actin Homologs

Subject Collection Cell Biology of Bacteria

Electron CryotomographyElitza I. Tocheva, Zhuo Li and Grant J. Jensen

Cyanobacterial Heterocysts

James W. GoldenKrithika Kumar, Rodrigo A. Mella-Herrera and

Protein Subcellular Localization in BacteriaDavid Z. Rudner and Richard Losick Cell Division in Bacteria

Synchronization of Chromosome Dynamics and

Martin Thanbichler

Their Spatial RegulationPoles Apart: Prokaryotic Polar Organelles and

Clare L. Kirkpatrick and Patrick H. ViollierMicroscopyAutomated Quantitative Live Cell Fluorescence

Michael Fero and Kit Pogliano

MorphogenesisMyxobacteria, Polarity, and Multicellular

Dale Kaiser, Mark Robinson and Lee KroosHomologsThe Structure and Function of Bacterial Actin

Joshua W. Shaevitz and Zemer GitaiMembrane-associated DNA Transport Machines

Briana Burton and David DubnauBiofilms

Daniel Lpez, Hera Vlamakis and Roberto KolterThe Bacterial Cell Envelope

WalkerThomas J. Silhavy, Daniel Kahne and Suzanne III Injectisome

Bacterial Nanomachines: The Flagellum and Type

Marc Erhardt, Keiichi Namba and Kelly T. HughesCell Biology of Prokaryotic Organelles

Dorothee Murat, Meghan Byrne and Arash Komeili Live Bacteria CellsSingle-Molecule and Superresolution Imaging in

Julie S. Biteen and W.E. Moerner

SegregationBacterial Chromosome Organization and

Esteban Toro and Lucy Shapiro

http://cshperspectives.cshlp.org/cgi/collection/ For additional articles in this collection, see

Copyright 2010 Cold Spring Harbor Laboratory Press; all rights reserved

on July 22, 2014 - Published by Cold Spring Harbor Laboratory Press http://cshperspectives.cshlp.org/Downloaded from on July 22, 2014 - Published by Cold Spring Harbor Laboratory Press http://cshperspectives.cshlp.org/Downloaded from

The Structure and Function of BacterialActin Homologs

Joshua W. Shaevitz1 and Zemer Gitai2

1Department of Physics and the Lewis-Sigler Institute for Integrative Genomics, Princeton University,Princeton, New Jersey 08544

2Department of Molecular Biology, Princeton University, Princeton, New Jersey 08544

Correspondence: [email protected]

During the past decade, the appreciation and understanding of how bacterial cells canbe organized in both space and time have been revolutionized by the identification andcharacterization of multiple bacterial homologs of the eukaryotic actin cytoskeleton.Some of these bacterial actins, such as the plasmid-borne ParM protein, have highly special-ized functions, whereas other bacterial actins, such as the chromosomally encoded MreBprotein, have been implicated in a wide array of cellular activities. In this review we coverour current understanding of the structure, assembly, function, and regulation of bacterialactins. We focus on ParM as a well-understood reductionist model and on MreB as acentral organizer of multiple aspects of bacterial cell biology. We also discuss the outstand-ing puzzles in the field and possible directions where this fast-developing area may progressin the future.

The discovery of cytoskeletal proteins in bac-teria has fundamentally altered our under-standing of the organization and evolution ofbacteria as cells. Homologs of eukaryotic actinrepresent themostmolecularly and functionallydiverse family of bacterial cytoskeletal elements.Recent phylogenetic studies have identifiedmore than 20 subgroups of bacterial actinhomologs (Derman et al. 2009) (Fig. 1). Manyof these bacterial actins are encoded on extra-chromosomal plasmids, but most bacterial spe-cies with nonspherical morphologies alsoencode chromosomal actin homologs (Danieland Errington 2003). The two earliest proteinsto be characterized as bacterial actins were the

chromosomal protein MreB (Jones et al. 2001)and the plasmidic protein ParM (Jensen andGerdes 1997). MreB and ParM remain the best-characterized of the bacterial actins and we willthus focus on these two proteins formost of thisarticle.

The appreciation that bacteria possess actinhomologs only occurred in the past decade.MreB was first identified as a protein involvedin cell shape regulation in Escherichia coli inthe late1980s (Doi et al. 1988). In theearly1990s,pioneering bioinformatic studies identifiedsimilarities in a group of ATPases that have fiveconserved motifs (Bork et al. 1992), a featuredubbed the actin superfamily fold. Although

Editors: Lucy Shapiro and Richard Losick

Additional Perspectives on Cell Biology of Bacteria available at www.cshperspectives.org

Copyright# 2010 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a000364Cite this article as Cold Spring Harb Perspect Biol 2010;2:a000364

1

on July 22, 2014 - Published by Cold Spring Harbor Laboratory Press http://cshperspectives.cshlp.org/Downloaded from

this group includes actin and MreB, it also con-tains proteins that do not polymerize into fila-ments, such as sugar kinases like hexokinaseand chaperones like Hsp70. A number of bacte-rial proteins are present in the actin superfamily,including the bacterial cell division protein FtsAwhich interacts with the tubulin homolog FtsZand may or may not form filaments in differentcontexts (van den Ent and Lowe 2000). BecauseMreB did not appear significantly more relatedto actin than these nonfilamentous proteins,the weak sequence similarity with actin waslargely ignored for the better part of a decade.This changed in 2001 when two seminal papersshowed that Bacillus subtilisMreB forms cytos-keletal filaments in vivo (Jones et al. 2001) andthatThermotogamaritimaMreB forms cytoske-letal filaments in vitro (van den Ent et al. 2001).Indeed, structural and biochemical studies ofbothMreB and ParM have convincingly showedthat these proteins closely resemble actin andpolymerize into linear filaments in a nucleotide-dependent manner (Fig. 2).

Research following the identification ofbacterial cytoskeletal proteins has focused onunderstanding their assembly, regulation, andfunction. Here, we will summarize our currentunderstanding of these issues and highlightthe outstanding questions. We will begin withParM, whose well-characterized assembly anddynamics represent a model for future studiesof all cytoskeletal proteins. We will then focuson MreB, whose diverse activities appear to becentral to the cell biology of many bacterialspecies.

PARM AS A MODEL FOR STUDYINGBACTERIAL ACTINS

Perhaps the most completely understood bacte-rial actin is ParM, the plasmid segregating pro-tein from the R1 plasmid. An interdisciplinaryeffort by a number of groups over the last dec-ade has illuminated the details of ParM functionfrom its fundamental molecular properties to areconstitution of plasmid segregation in an

MreB

FtsA

MamK

AlfA

Alp7Alp8

Alp6

ParM

Figure 1. The superfamily of bacterial actin homologs. Shown is a phylogenetic tree of the bacterial actinsubfamilies that have been identified to date based on sequence homology. The subfamilies that have beenexperimentally shown to polymerize are labeled and colored. (Courtesy of Joe Pogliano, based on Dermanet al. 2009.)

J.W. Shaevitz and Z. Gitai

2 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a000364


artificial system. This multipronged approachbrings together techniques from biochemistry,molecular biology, imaging, and materials sci-ence to provide an integrated solution to howParM filaments form and how they lead to themovement of DNA. It is likely that this kindof approach will be useful in the study of theother, lesser-understood bacterial actins.

ParM Function and Mechanism

Plasmids are naturally occurringmolecular par-asites that often exploit their host cells machi-nery for replication but provide their ownsegregation machinery. By 1997 it was knownthat the low-copy number E. coli plasmid R1is actively partitioned during cell division sothat each daughter cell retains the drug resist-ance conferred by the plasmid (Jensen andGerdes 1997). This partitioning is achieved bythe active positioning of two R1 sister plasmidsto opposite ends of a cell during cell division(Jensen and Gerdes 1997). The machinery thatdirects the two plasmids to opposite cell polesis grouped in a locus termed par. The presenceof this operon in a plasmid lowers the frequencyof plasmid loss during division by several orders

of magnitude (Gerdes et al. 1985). The Parmachinery consists of three parts: a cis-actingregion of DNA, parC, that acts as a centromere;the ParR protein that binds to 10 repeats withinparC and has kinetichore-like activity; and theactin-homolog ParM, originally labeled as thepartitioning motor, which drives segregationthrough polymerization dynamics (Dam andGerdes 1994).

The details of plasmid segregation are ele-gant both in their simplicity and their efficiency.Throughout the cytoplasm, ParM filamentsactively lengthen and shorten in a processknown as dynamic instability that somewhatmimics the action of spindle microtubules(Garner et al. 2004). However, when the ParMfilaments are bound to a plasmid via theParR/parC complex they are stabilized againstdepolymerization (Garner et al. 2004). Conse-quently, plasmid bound ParM filaments pro-ceed to elongate as much as they can, which ina rod-shaped cell pushes the plasmids to thecell poles at opposite ends of the cell axis(Garner et al. 2007) (Fig. 3).

This mechanism was discovered through aninterdisciplinary approach to the study of plas-mid segregation. Imaging work in vivo first

F-actin MreB ParM:ADP apo ParM ParM filament

Prot

ofila

men

t axis

Figure 2. Structures of F-actin (Holmes et al. 1990),MreB (van den Ent et al. 2001), and ParM (van den Ent et al.2002). (Left) Structures of F-actin filaments (PDB entry 1YAG). (Second from the left) MreB filaments fromT. maritima (PDB entry 1JCE). (Center) ParM:ADP monomer in the closed conformation. (Second fromthe right) apo ParM monomer in the open conformation. (Right) ParM filament. Shown are the positionof the nucleotide within the interdomain cleft, the conservation of fold, and the axis of the protofilamentextension (arrow). Note that the conformational change shown for ParM from the open to closed state ispredicted for all actin homologs. (Adapted, with permission from, Michie and Lowe 2006.)


Cite this article as Cold Spring Harb Perspect Biol 2010;2:a000364 3


defined the segregation pattern of plasmids andshowed that the ParM protein forms long fila-ments in cells (Moller-Jensen et al. 2002;Moller-Jensen et al. 2003). Mutagenesis andother molecular techniques led to the mappingof the par locus (Dam and Gerdes 1994). Invitro work, including structural, biochemical,and imaging experiments confirmed the ini-tial hypotheses from in vivo experiments andallowed for the reconstitution of the entire seg-regation process (Moller-Jensen et al. 2003;Garner et al. 2004; Garner et al. 2007). RichardFeynman once said that What I cannot create, I

do not understand. It is in this spirit that wethink that ParM is well understood. In the para-graphs to follow, we discuss in greater detailwhat is currently known about ParM from themolecular to the cellular scales.

ParM Polymerization and Structure

ParM forms filaments both in vivo and in vitro.Initial immunofluorescence imaging showedthat ParM formed long, curved filaments(Moller-Jensen et al. 2002). These authorsobserved that ParM filaments often spannedthe length of rod-shaped cells and that the fila-ments were gently curved. Interestingly, thelength of the filaments was observed to bedynamic. In vitro, purified ParM was shownto polymerize into long, straight filaments.Polymerization requires ATP, or the nonhydro-lyzable ATP analogs ATPgS and AMPPNP(Moller-Jensen et al. 2002). In the presence ofADP, ParM does not polymerize. This observa-tion, together with the dynamic nature of ParMfilaments in cells, led to a dynamic instabilitymodel for plasmid segregation, similar to thatfor tubulin, where ATP-bound ParMmonomersbind stably to the endof ParMfilamentswhereasposthydrolysis, ADP-bound monomers lead todepolymerization from the filament end.

The three-dimensional crystal structure ofmonomeric ParM shows remarkable similarityto that of actin (Fig. 2). ParM is an asymmetricprotein with a barbed and a pointed end madeup of four domains: IA, IB, IIA, and IIB, whichrespectively correspond to the actin domains 1,2, 3, and 4 (van den Ent et al. 2002). ParM hasbeen successfully crystallized in the nucleotide-free and ADP bound states. These structuresshow that the nucleotide binds in the interdo-main cleft of the barbed end. On nucleotidebinding, a rigid-body rotation of about 258 ofdomains I and II closes the cleft slightly. Thisconformational change is thought to cause thenucleotide-induced depolymerization of ParMfilaments.

The structure of ParM filaments was solvedby docking the monomeric crystal structuresinto three-dimensional electron microscopyimages of filaments (van den Ent et al. 2002).

Stable

Unstable

Stable

Unstable

Unstable

Unstable

Figure 3. The assembly dynamics of ParM drive R1plasmid segregation. Free ParM filaments ends (red)are dynamically unstable and undergo catastrophe.ParR (yellow) associates with the parC loci on theR1 plasmid (plasmids are green). Plasmid-boundParR captures and stabilizes ParM filaments (blue).When both ends of a ParM filament are stabilizedby ParR, a productive spindle is formed. Insertionalpolymerization at the filament ends serves to drivethe two plasmids apart. (Adapted, with permission,from Garner 2008.)




Filaments assemble with the pointed end of onemonomer enclosed by the barbed-end of thenext. Along the filament, adjacent monomersare rotated by about 1668 and translated by 2.5nm from each other. Even though this is aslightly tighter twist and shorter axial spacingthan that of actin, ParM still forms a twisted,two-start helix. One significant differencebetween actin and ParM filaments is that theParM helix is left-handed whereas actin is right-handed. The juxtaposition of a similar but chir-ally opposite geometry suggests that this helicalshape might afford the filaments a high level ofstability that has been arrived at independentlythrough convergent evolution. Similar conclu-sions about the shape of ParM filaments invivo have been made using high-resolutioncryo-EM tomography (Salje et al. 2009). Morerecently, analysis of additional bacterial actinhomologs supports the idea that though theseproteins all polymerize, the details of their fila-ment formation can be quite divergent (Polkaet al. 2009). Although the sequence and 3Dstructure of the ATPase domains of ParM havesignificant homology with actin, the exposedresidues of ParM are very different in theirchemical properties from those of actin. Conse-quently, proteins that bind to actin and to ParMare not expected to share much similarity.

ParM Biochemistry and Biophysics

In vitro experiments have been very successful atverifying and enhancing the early models ofplasmid segregation derived from in vivo ex-periments. Jensen and Gerdes showed that mu-tation of the aspartate at position 170 of ParMabolished both the proteins ATPase activityand its ability to segregate plasmids in vivo(Jensen and Gerdes 1997; Jensen and Gerdes1999). Immunofluorescence imaging showedthat these mutants showed hyperfilamentation,lending support to the idea that ParMdynamicsare required for proper plasmid segregation.

Right-angle light scattering measurementshave been used to measure the kinetics ofParM filament polymerization. Moller-Jensenand others showed that after an initial bout ofpolymerization, ParM slowly hydrolyzes bound

ATP leading to filament depolymerization(Moller-Jensen et al. 2002). Additional ATPadded after this depolymerization phase causesa new round of fast polymerization and slowdepolymerization, indicating that the mono-mers are capable of going through multiplerounds of nucleotide stimulated polymeriza-tion and depolymerization. Most interestingly,filaments are stabilized by the addition of bothParR and parC. When these three componentsare mixed together, filaments do not depoly-merize and the critical concentration for poly-merization is reduced (Garner et al. 2007).These results were confirmed by subsequentFRET measurements that verified the nuclea-tion condensation mechanism of ParM poly-merization. Garner and coworkers found acritical nucleus size of three monomers thatdefines the two-stranded helical filament geom-etry (Garner et al. 2004).

Although light scattering is a useful tool tomeasure the typical size of an ensemble of olig-omers in solution, it cannot resolve the natureof individual filament dynamics. For example,it can be difficult to distinguish whether somemonomers are incapable of forming polymersor if the ensemble of polymers is very dynamic.Garner et al. used total-internal fluorescencemicroscopy to directly visualize single Alexa-dye labeled ParM filaments in vitro (Garneret al. 2004). Using a two-color assay in whichred-labeled monomers were added to green-labeled filament seeds, they were able to distin-guish growth from the two filament ends. Usingthe nonhydrolyzable substrate AMPPNP, theyobserved steady growth from both ends oflong, stable filaments. In the presence of ATP,however, filaments grew symmetrically for awhile but then rapidly depolymerized to com-pletion fromone end. The time scales associatedwith the filament growth and shrinking wereconsistent with a mechanism in which an ATPcap stabilizes the ParM filament. If the terminalmonomers hydrolyze their bound nucleotide,the filament becomes unstable. These studiesthus showed that in contrast to actin, whichpreferentially polymerizes at one filament endat a constant rate, ParM polymerizes symmetri-cally and experiences dynamic instability with




bouts of steady polymerization interrupted byrapid depolymerization.

Putting all of these ideas together, Garnerand colleagues showed that ParM, ParR, andparC are sufficient to reconstitute plasmid seg-regation in vitro (Garner et al. 2007). The au-thors coated small, 350-nm diameter beadswith a piece of DNA that contained the parCsequence. When added to a solution of ParRand fluorescently labeled ParM, the beadsformed aster-like clouds of highly dynamicParM. When two beads were in proximity, theirParM structures stabilized one another to gen-erate long spindle-like structures. EM imagesindicated that single filaments extended all theway from one bead to the other, suggestingthat the long structures resulted from stabiliza-tion of both ends of the ParM filaments. Thesefilaments were able to elongate by insertionalpolymerization between the tips of the fila-ments and the beads, thereby pushing the DNA-coated beads farther and farther apart (Fig. 3).The energetics of this elongation were proposedto be driven by a monomer excess generated bythe dynamic instability of the nonstabilizedParM. Most excitingly, when the parC-coatedbeads, ParM, and ParR were placed in longmicrofabricated channels, the ParM spindleelongated along the long axis of the channel,thereby breaking symmetry to separate pairsof beads to opposite ends of the cylinder.

Although the ability to reconstitute theentire plasmid segregation process is nothingshort of remarkable, it is worth noting that itshows sufficiency for how the system couldwork, but does not prove how it does work invivo. Thus, several exciting questions aboutParM remain to be answered. For example, doesthe rest of the plasmid DNA really just play apassive role in the whole process? Also, why isit that ParM and actin filaments are structurallysimilar yet one polymerizes symmetrically atboth ends while the other polymerizes asym-metrically? Along similar lines, EM studies bothin vitro and in vivo suggest that both ends of thefilament associate with the ParR/parC complex(Garner et al. 2007; Salje et al. 2009), and workby Choi and colleagues used very small goldbeads to show that in a ParM spindle, a single

copy of parC is bound to the end of a singleParM filament (Choi et al. 2008). So how doesthe ParR/parC manage to interact with bothends of an asymmetric polymer? Other unre-solved issues surround the nature of thedynamic instability and the ability of othernucleotides such as GTP to stabilize ParM fila-ments (Popp et al. 2008). Nevertheless, ParM-mediated plasmid segregation remains a proto-type for how combining genetics, cell biology,biochemistry, and biophysics can lead to anemergent understanding of dynamic systems.

THE LOCALIZATION AND FUNCTIONOF MREB

ParM is encoded by a subset of plasmids andcarries out one highly specific function, plasmidsegregation. In contrast, the most widely con-served bacterial actin homolog, MreB, is en-coded in the chromosomes of many differentspecies and can participate in many differentcellular activities (Daniel and Errington 2003).MreB homologs have been implicated in nearlyevery spatially organized cellular process, in-cluding cell growth, morphogenesis, polarity,protein localization, organelle positioning, di-vision, and differentiation, as well as chromo-some segregation, replication, and decatenation(reviewed in Carballido-Lopez 2006). The cur-rent challenge is to understand the mechanismby which MreB impacts these cellular processesand distinguish its primary roles from their sec-ondary consequences. MreB also assembles intoan interesting, often helical, localization patternthat may help it execute its many functions(Jones et al. 2001). Finally, MreB clearly cannotbe doing all of these things on its own; indeed, agrowing number of MreB interactors are beingidentified. In this section we will focus on thebest characterized of these MreB localizations,functions, and interactors.

MreB Proteins are Relatively Diverse

Discussion of MreB is inherently complicatedbecause of the vast diversity of MreB homo-logs in the bacterial and archael kingdoms. Atleast one MreB homolog is found in most




nonspherical bacteria (Daniel and Errington2003). The exceptions to this rule include anumber of plant and animal pathogens thatare rod-shaped but lack a clear MreB homolog,includingMycoplasmas,Mycobacteria, and Rhi-zobiae. There are also several spherical bacteriasuch as Cyanobacteria and Planctomycetes thathave MreB homologs. Many bacteria alsoencode multiple MreB homologs. The best-characterized example is B. subtilis, which hasthree MreB homologs: MreB, Mbl, and MreBH(Jones et al. 2001). The three B. subtilis MreBhomologs each has similar homology to otherMreBs (50% identity to T. maritima MreB),such that even though only one of them bearsthe name MreB, they should each be viewedequivalently. Spiroplasmas have even moreMreB homologs, as many of these species havefiveMreB proteins withmore divergent sequen-ces (Kurner et al. 2005). Yet other species, likeMagnetospirillum magnetotactum, have both arelatively well-conserved MreB homolog and asecond more divergent actin superfamily mem-ber, MamK (Komeili et al. 2006). All MreBhomologs share the same basic actin superfam-ily signature, and all MreB homologs character-ized to date have been found to be able topolymerize. Nevertheless, the diversity in num-ber and sequence ofMreB homologs means thatMreB cannot be treated as a single entity andthat it is important to specify which MreBhomolog one is discussing in a specific context.Most of the structural and biochemical work onMreB has been performed on the T. maritimahomolog, whereas the in vivo properties ofMreB have best been characterized in B. subtilis,Caulobacter crescentus, and E. coli. Our discus-sion here will primarily focus on these systems.

MreB Localization and Dynamics

MreB localization was first characterized inB. subtilis for both the mreB and mbl genes(Jones et al. 2001). These proteins were foundto form right-handed helical structures bydeconvolution microscopy. In some cells a dou-ble helix is observed, whereas others resemble asingle helix. Assessing the exact dimensions andtopologies of these helices remains an active

area of investigation that is complicated bytwo issues. First, the small size of these struc-tures is near the diffraction-limited resolutionof light microscopy. Second, immunofluores-cence (IF) microscopy requires fixation that canalter cell ultrastructure, whereas both amino-and carboxy-terminal fluorescent protein fu-sions to MreB can perturb both the functionand localization of native MreB. Very recently,an internal fusion that placed mCherry in themiddle of the E. coli MreB protein was foundto be largely functional (Bendezu et al. 2009),suggesting that these issues, perhaps in combi-nation with recent advances in subdiffraction-limited microscopy, might soon be resolved.

Nevertheless, in many cases the IF and theGFP-MreB images agree, leading to a consensusview that MreB prefers, at least locally, to forma helix with several turns per cell length. InB. subtilis, the three MreB homologs colocalizewith an approximate pitch of 0.75 mm, sug-gesting that the three MreB isoforms may func-tionally interact or copolymerize into a singlestructure (Carballido-Lopez et al. 2006). As cellselongate, the pitch appears to remain constantwith addition of new helical turns. In Caulo-bacter, both IF and GFP-MreB reporter studiesindicate that the localization of MreB is regu-lated during the cell cycle (Figge et al. 2004;Gitai et al. 2004). Early in the cell cycle, Caulo-bacter MreB forms a patchy, potentially helical,pattern that extends from pole to pole. As thecell cycle progresses, this helix condenses intoa ring at the presumptive division plane. Beforecell division occurs, MreB expands from a ringback into a helical form such that each daughtercell inherits a similar polymer structure. Thetransition from helix to ring depends on theFtsZ tubulin homolog, the central organizerof the division machinery (Figge et al. 2004).Localization experiments in other species suchas E. coli, P. aeruginosa, and Rhodobacter havealso characterized both helical and medialMreB distributions (Slovak et al. 2005; Vatset al. 2009; Cowles and Gitai 2010), thoughthe physiological consequences of this transi-tion remain unclear.

In contrast to B. subtilis cells that colocalizethree closely related MreB homologs, MreB




does not appear to colocalize with more highlydiverged actin homologs. For example, inM. magnetotactum, the divergent actin MamKforms a straight filament structure unlike anyMreB localization pattern that has been charac-terized (Komeili et al. 2006). Similarly, plasmi-dic actin homologs like ParM, AlfA, and the Alpproteins form straight structures that do notcolocalize with MreB (Moller-Jensen et al.2002; Becker et al. 2006; Derman et al. 2009).Some of these proteins, including ParM andAlfA, have highly divergent polymer interactionsurfaces and geometries (van den Ent et al.2001; van den Ent et al. 2002; Polka et al.2009), which may function to actively preventtheir copolymerization with MreB.

While the entire MreB structure is dynami-cally rearranged in Caulobacter, it also appearsthat the subunits of the MreB helix are highlydynamic in all cells examined. These dynam-ics have been assessed by FRAP for Mbl(Carballido-Lopez and Errington 2003), byspeckle tracking for all three MreB proteins inB. subtilis (Defeu Soufo and Graumann 2004),and by single-molecule imaging of MreB inCaulobacter (Kim et al. 2006). The CaulobacterMreB appears to have faster turnover kineticsthan observed in B. subtilis, but it remainsunclear whether this reflects species-specificdifferences or differences in the types of assaysused (FRAP vs. single-molecule imaging).Although the Caulobacter studies suggest thatindividual MreB filaments may be polarlyassembled (Kim et al. 2006), there does notappear to be an overall pole-to-pole polarityin the MreB helix, suggesting that the MreBhelix includes individual filaments with mixedpolarities.

Further evidence for the dynamic natureof MreB assembly comes from experiments withthe small molecule A22 (S-(3,4-dichlorobenzyl)isothiourea). A22 was first found in a chemicalgenetic screen for compounds that increased therate of chromosome loss in E. coli (Iwai et al.2002). A22 was found to cause cells to becomeround, but its cellular target was unknown untilA22-resistant mutations were mapped to themreB gene, first in Caulobacter (Gitai et al.2005), and subsequently in other species such

as E. coli and P. aeruginosa (Kruse et al. 2006;Robertson et al. 2007; Cowles and Gitai 2010).In Caulobacter, A22 rapidly delocalizes MreBfilaments in vivo, and the resistant mutantsall map to the nucleotide-binding pocket (Gitaiet al. 2005). Indeed, biochemical studies(detailed below) show that A22 functions bybinding the MreB nucleotide-binding pocketand mimicking the ADP-bound low polymer-ization affinity monomer state (Bean et al.2009). When A22 is applied to either purifiedMreB in vitro (Bean et al. 2009) or MreB thathas been heterologously expressed in the eukar-yote S. pombe (Srinivasan et al. 2007), A22inhibits new MreB polymerization but doesnot stimulate depolymerization. Consequently,the rapid delocalization of MreB observed inbacterial cells likely results from dynamicMreB depolymerization that can no longer bebalanced by A22-inhibited polymerization.The increased dynamics of MreB assembly invivo also suggests that additional cellular factorsexist to stimulate these dynamics. The identifi-cation and characterization of such factors willbe an important area of future investigation.

Another important yet unresolved issue isthe mechanism by which MreB adopts its local-ization pattern. MreB filaments are not obvi-ously helical either in vitro or on heterologousexpression in eukaryotes (van den Ent et al.2001; Srinivasan et al. 2007). It is possible thatMreB helices are the mechanical consequenceof forcing a linear polymer into a cylindricalcontainer. Alternatively, accessory factors maychange the preferred conformation of MreB fil-aments in bacteria. The dynamic redistributionof MreB during the Caulobacter cell cycle sup-ports the idea that MreB conformation andlocalization can be regulated. One candidatefor such regulation is RodZ, which was recentlycharacterized as a protein that binds MreB andpotentially links it to the inner membrane(Shiomi et al. 2008; Alyahya et al. 2009; Bendezuet al. 2009; van den Ent et al. 2010). RodZ andMreB have an interdependent genetic relation-ship and similar loss-of-function phenotypes,making it difficult to dissect a linear localizationhierarchy between these proteins. Yet otherMreB-interacting proteins such as Ef-Tu (Defeu




Soufo et al. 2010), Pbp2 (Figge et al. 2004), orother as-yet-unidentified proteins could alsoinfluence MreB localization.

MreB is a Key Regulator of Cell ShapeDetermination

MreBwas first characterized as an E. colimutantthat caused normally rod-shaped cells to be-come spherical (Doi et al. 1988). Indeed, lossof proper cell shape has emerged as the mostcommon defect associated with MreB proteinsacross most species. The central importance ofMreB for achieving rodlike elongation is sup-ported by the phylogenetic observation thatMreB proteins are primarily found in non-spherical cells (Daniel and Errington 2003).However, there are exceptions to this rule. Forexample, Helicobacter pylori MreB has beenreported to influence chromosome dynamicsand virulence factor secretion without affectingcell shape (Waidner et al. 2009), and in Strepto-myces coelicolor MreB affects sporulation butdoes not affect the shape of vegetatively growingcells (Mazza et al. 2006). To more generallysurvey how rodlike bacteria grow, the Erringtonlab chemically labeled nascent cell wall synthe-sis and characterized two distinct cylindricalgrowth modes in different species (Daniel andErrington 2003). One mode is more commonand uses MreB to direct the insertion of newcell wall material along the length of the cylin-der. These species generally have inert cell poles.The second mode is MreB-independent andinvolves insertion of new cell wall material atthe cell poles. A third, FtsZ-dependent elonga-tion mechanism has also been more recentlyproposed (Aaron et al. 2007).

MreB regulates cell shape in three well-characterized model systems: B. subtilis, E. coli,and Caulobacter. In B. subtilis, mutants in thethreeMreB homologs have different morpholo-gies (Jones et al. 2001). It is possible that theseMreB proteins have specialized for distinctfunctions. Alternatively, a study demonstrat-ing partial redundancy between these mutantssuggests that they may all influence the sameprocess to different extents, perhaps becauseof differential expression levels (Kawai et al.

2009). In E. coli grown under normal condi-tions, cells lacking MreB become spheres andeventually lyse (Bendezu and de Boer 2008).However, the lethality of loss of MreB can besuppressed either byoverexpressing cell divisionproteins or reducing the rate of cell growth(Bendezu and de Boer 2008). In these condi-tions mreB mutants are viable but accumulateintracellular vesicles, suggesting that membraneproduction is disregulated (Bendezu and deBoer 2008). The lethality of a mutation in oneof the B. subtilis mreB genes can also be sup-pressed by growth in increased levels of Mg

(Formstone and Errington 2005).Studies from both B. subtilis and Caulo-

bacter suggest that at least one way in whichMreB influences the cell wall is by directingthe insertion of new cell wallmaterial in a helicalpattern. The bacterial cell wall is composed ofstiff peptidoglycan strands that are polymerizedby transglycosylases and crosslinked by trans-peptidases to generate a meshlike superstruc-ture (Holtje and Heidrich 2001). The cell wallhas been thought to be the pressure-bearing cellshape determinant because cell wall lysis causescells to round up and isolated cell walls retaintheir generalmorphology (Young 2006). Becauseboth MreB and new cell wall insertion followhelical patterns, the basic model for how MreBdirects cell shape determination is that MreBhelically localizes proteins that in turn lead to hel-ical peptidoglycan assembly (Daniel and Erring-ton 2003). It is also possible that as in eukaryotes,the MreB cytoskeleton plays a mechanical role indirecting proper morphogenesis.

Consistent with the cell wall patterningmodel, Caulobacter MreB biochemically asso-ciates with the cytoplasmic tail of a cell wallassembly enzyme, the Pbp2 peptidoglycantranspeptidase, and directs its localization(Figge et al. 2004; Dye et al. 2005). CaulobacterMreB has also been proposed to direct the local-ization of cytoplasmic proteins that direct pep-tidoglycan subunit synthesis (White et al. 2010)and inhibition of MreB with the small moleculeA22 leads to shortened cell wall glycan strands(Takacs et al. 2010), indicating that MreB mayhave a general influence on peptidoglycanassembly. In B. subtilis, the Pbp proteins adopt




a helical localization pattern (Scheffers et al.2004). However, single mreB mutants do notdisrupt this localization (Scheffers et al. 2004),possibly because of partial redundancy betweenthe different MreB. Unlike E. coli and Caulo-bacter,which are gram-negative bacteria, B. sub-tilis are gram-positive and their cell wall is alsocomposed of teichoic acids. The proteins thatassemble teichoic acids are also helically local-ized, though it remains unclear if this helicaldistribution is related to that of the MreBproteins (Formstone et al. 2008). In addition,MreBH interacts with the cell wall hydrolaseLytE (Carballido-Lopez et al. 2006), and theo-retical studies suggest that cell shape could bedictated by patterning either peptidoglycan in-sertion or cleavage (Huang et al. 2008).

In many species, mreB is found in the sameoperon as two other genes, mreC and mreD.MreB, MreC, and MreD, along with the cellwall assembly proteins Pbp2 and RodA havebeen proposed to form a complex that collabo-rates to coordinate cell elongation (Kruse et al.2005). This complex of predominantly trans-membrane proteins may explain how the cyto-plasmic MreB structure can have such animpact on the assembly of the cell wall, whichoccurs outside of the inner membrane. In Cau-lobacter, MreC is cleaved and released into theperiplasm where it forms a helical structurewhose assembly is independent of MreB (Diva-karuni et al. 2005). The crystal structure ofMreC suggests that it may form polymers andcould thus act as a periplasmic cytoskeleton(van den Ent et al. 2006). Further genetic andcolocalization studies in Caulobacter suggestthat MreC acts to localize peptidoglycan assem-bly in the periplasm (Divakaruni et al. 2005;Dye et al. 2005), whereas MreB acts to localizepeptidoglycan precursor synthesis in the cyto-plasm (Figge et al. 2004; White et al. 2010).However, this model must either be oversimpli-fied or species-specific because some speciesthat have MreB lack MreC or vice-versa, andin yet other species, MreB localization dependson MreC (Kruse et al. 2005).

The growing body of evidence that MreBcan influence nearly every aspect of cell wall as-sembly suggests either that all of these processes

are highly interdependent or that MreB hasmultiple independent roles in regulating cellshape. Dissecting the specific roles of MreB inthis process has been complicated by the factthat we do not have very good tools for studyingpeptidoglycan structure and dynamics. Fluores-cent derivatives of antibiotics that target nascentpeptidoglycan structures have proved useful(Daniel and Errington 2003; Tiyanont et al.2006), but can also perturb cell wall assemblyandoccasionally produceconflicting results.Re-cent studies have used high-resolution atomicforce microscopy to study the exposed cell wallsof gram positives or isolated gram-negative cellwalls (Yao et al. 1999; Touhami et al. 2004), andan elegant electron cryotomography study gavethe first glimpses into the structure of Gram-negative cell walls (Gan et al. 2008). By combin-ing these new approaches for studying peptido-glycan, the next few years promise to reveal agreat deal about the assembly and regulationof bacterial cell walls.

MreB Regulates Protein Localization

MreB has also emerged as a key regulator of thesubcellular organization of bacterial proteins.The role of MreB in polar protein localizationwas first characterized in Caulobacter, whereMreBwas found to be necessary for the localiza-tion of multiple polar protein markers (Gitaiet al. 2004). Subsequently,MreB has been impli-cated in a wide range of protein localizationprocesses in many experimental systems. Inaddition to the proteins involved in cell wallassembly discussed in the previous section oncell shape, these include chemotaxis receptorsin E. coli (Shih et al. 2005), glidingmotility pro-teins in Myxococcus xanthus (Mauriello et al.2010), and pilus-associated proteins in Pseudo-monas aeruginosa (Cowles and Gitai 2010).There is also evidence that bacterial actins areimportant for localizing larger protein com-plexes or organelles. For example, MamK isessential for magnetosome localization in M.magnetotactum (Komeili et al. 2006). Similarly,MreB influences stalk assembly and localizationin Caulobacter (Wagner et al. 2005; Divakaruniet al. 2007), pilus assembly in P. aeruginosa and




M. Xanthus (Cowles and Gitai 2010; Maurielloet al. 2010), inclusion body localization inE. coli (Rokney et al. 2009), and carboxysomelocalization in Synechoccus elongates (Savageet al. 2010). These larger structures may partic-ularly benefit from cytoskeletal-mediated local-ization because of their inherently reduced ratesof diffusion.

One issue with studying MreBs functions isthat MreB is pleiotropic, such that it can be dif-ficult to untangle the direct effects of MreB onsubcellular localization from indirect effectscaused by decreased growth or disrupted mor-phology. One advance that has been of immensehelp in this area has been the discovery of smallmolecules, such as A22, that rapidly perturbMreB. The rapid action of these compoundshelps temporally uncouple primary and secon-dary consequences of MreB disruption. Forexample, although MreB can be delocalized inas little as 30 s with these agents, the cell shapedefects induced on MreB disruption requirehours of new growth to manifest, producing atime window during which MreB is disruptedbut cell shape is still unaffected (Gitai et al.2005). One concern with using any small mole-cule inhibitor is the degree of specificity for thetarget protein of interest. Although A22 appearsto primarily affect MreB, genetic studies fromboth E. coli and P. aeruginosa suggest that itmay have additional off-target effects (Cowlesand Gitai 2010; Takacs et al. 2010). Recently,two additional MreB antagonist compounds,CBR-4830 and MP265 have been characterized(Robertson et al. 2007; Takacs et al. 2010). Bycombining comparisons of multiple distinctMreB inhibitors and the proper use of drug-resistant control strains, small molecule antago-nists remain the premier method for dissectingMreB function.

AlthoughMreB is clearly important for pro-tein localization, and this role may explain howMreB can affect so many cellular processes, themechanism by which MreB directs proteinlocalization remains mysterious. Increasing evi-dence suggests that MreB may often be involvedin the initiation, but not the maintanence, ofprotein localization. For example, in Caulo-bacter, the polar marker PopZ localizes to one

cell pole early in the cell cycle and later redistrib-utes to the other cell pole (Bowman et al. 2008).Although MreB is not required for maintainingthe localization of PopZ at the initial pole, it isrequired for relocalizing PopZ to the secondpole (Bowman et al. 2008). By treating cellswith A22 before or after the induction of a flu-orescent fusion to a protein of interest, one candirectly distinguish whetherMreB is involved inmaintaining the localization of old protein(which would be delocalized regardless ofwhether A22 is administered before or afterinduction), or whether MreB is involved in ini-tiating new protein localization (which wouldonly be delocalized when A22 is administeredbefore induction). Consistent with the PopZexample, MreB is also required for the initiationof Pbp2 localization in Caulobacter (Dye et al.2005). These functions can also be regulated,as seen in P. aeruginosa, where MreB is requiredfor both the initiation andmaintenance of polarPilT localization when cells are grown in liquidmedia but is only required for initiation of PilTlocalization when cells are grown on solid sur-faces (Cowles and Gitai 2010). Thus, althoughseveral studies have taken a proteins persistentlocalization in the presence of A22 as indicationof MreB-independent localization, such experi-ments only address the importance of MreBfor the maintenance of protein localization. Abetter understanding of both the proteins thatdirectly interact with MreB and the single-moleculemotions of these target proteins shouldlead to a better understanding of the molecularmechanism by which MreB directs the dynam-ics of initiating protein localization.

MreB and Chromosome Dynamics

Bacterial actins have been implicated in regulat-ing the organization of bacterial DNA. Thisfunction is clearest for the plasmidic actins suchas ParM and AlfA, as discussed earlier. MreBproteins have also been implicated in chromo-some dynamics, though their specific functionsand mechanisms remain unclear. In E. coli, adominant-negative MreB mutant was shown toperturb chromsome segregation (Kruse et al.2003), and A22 treatment was found to both




affect chromosome segregation and increase therate of chromosome loss (Iwai et al. 2002; Kruseet al. 2006). A mechanism for this effect wassuggested by the interaction of MreB withRNA polymerase, which could serve as a motorto push the two chromosomes apart (Kruseet al. 2006). In contrast, recent studies haveshown that E. coli can survive and presumablysegregate its chromosomes in the absence ofMreB when cell growth is slowed (Bendezuand de Boer 2008), and can segregate its chro-mosomes when MreB localization is aberrant(Karczmarek et al. 2007). Similarly, MreB hasbeen implicated in chromosome segregationin B. subtilis in some cases but not others (Soufoand Graumann 2003; Formstone and Errington2005), though a careful analysis in the absenceof all three MreB homologs has yet to be com-pleted. Experiments in Vibrio cholerae andHelicobacter pylori support the role of MreB inchromosome segregation (Srivastava et al.2007; Waidner et al. 2009), whereas the nones-sential nature of MreB in other species such asS. coelicolor suggest that MreB is not necessaryfor segregation in these organisms (Mazzaet al. 2006; Hu et al. 2007). Some insight intothis apparent paradox might be gleaned fromstudies in Caulobacter. Under certain environ-mental conditions such as growth on agarosepads, the segregation of the region of the chro-mosome near the origin of replication is eitherblocked or significantly delayed by treatmentwith A22 (Gitai et al. 2005). However, underother conditions, such as growth in liquidmedia,A22 treatment only results in a mild delay in theonset of segregation, which can be largely attrib-uted to a delay in the onset of DNA replication(Shebelut et al. 2009). Genetic studies suggestthat there are multiple pathways that contributeto the process of segregation (Shebelut et al.2009). It is possible that these pathways mayact redundantly in some conditions, but thateachmay become essential under other, perhapsstressful, conditions. Although it remains un-clear whether MreB directly or indirectly affectschromosome segregation, viewing segregationas a sequence of distinct processes may helpdefine the specific mechanisms that collaborateto govern segregation in different contexts.

MreB has also been implicated in aspects ofchromosome dynamics other than segregation.In V. cholerae, MreB perturbations affect chro-mosome condensation (Srivastava et al. 2007).Meanwhile, in E. coli, MreB was found to regu-late chromosome decatenation (Madabhushiand Marians 2009). This effect appears to bedirect, as in vitro studies showed that the activ-ity of Topoisomerase IV was stimulated by puri-fied MreB polymers and inhibited by purifiedMreB monomers. Finally, MreB may play arole in DNA replication, as A22 treatment slowsreplication in Caulobacter (Shebelut et al. 2009)and MreB can affect the positioning of DNAreplication proteins in B. subtilis (Defeu Soufoand Graumann 2005). MreB may also berequired for the replication of foreign DNA inbacteria. MreB is required for the replicationof a number of phages that infect different bac-terial species, and MreB specifically mediatesthe localization of phage replication proteinsfor the B. subtilis phage 29 (Munoz-Espin et al.2009).

MreB May be Important for BacterialPathogenesis

The studyof bacterial cell biology in general andbacterial actins in particular is rapidly advanc-ing our understanding of the fundamentals ofcellular organization and dynamics. Mean-while, these studies also promise to yield excit-ing advances in our ability to combat in-fectious diseases. MreB is essential for the rapidgrowth of many bacteria, such that small mole-cule inhibitors of MreB could represent power-ful broad-spectrum antibiotics. Recent studieshave begun to suggest that MreB may alsohave been co-opted by pathogenic bacteria toorganize their virulence mechanisms. In P. aer-uginosa, MreB regulates the polar assembly oftype IV pili, which are important for virulence(Cowles and Gitai 2010). InH. pylori, MreB reg-ulates the secretion of virulence factors (Waid-ner et al. 2009), and Vibrio parahaemalyticusdifferentially regulates MreB expression on in-teractionwith its host (Chiu et al. 2008). Bdello-vibrio bacteriovorus is a bacterium that infectsother bacteria, and perturbing B. bacteriovorus




MreB perturbs this ability (Fenton et al. 2010).Consequently, understanding bacterial actinsmay help understand and ultimately developboth broad and narrow spectrum approachesto combating bacterial pathogenesis.

MREB STRUCTURE AND ASSEMBLY

Like ParM, MreB shares significant sequencehomology with actin. An early search of knownbacterial sequences indicated that MreB was alikely member of the actin superfamily basedon its catalytic core (Bork et al. 1992). Bysequence, MreB is the most closely related toactin of all the actin family proteins althoughsmall differences between MreB and actin existin both the nucleotide binding site and the res-idues that form themonomermonomer inter-face within a protofilament. The differences atthe monomermonomer interface are interest-ing because they must have evolved concomi-tantly to retain the polymerization. In vitropolymerization of MreB and subsequent solu-tion of the polymer crystal structure was per-formed in 2001 by van den Ent, Amos, andLowe (van den Ent et al. 2001). Using purifiedMreB1 from T. maritima, they found that poly-mers formed in a variety of conditions andrequired ATP or GTP.

MreB was the first actin for which the poly-meric crystal structure was solved (van den Entet al. 2001), as opposed to conventional actinfor which only EM-based polymer structuresexist (Oda et al. 2009). This breakthroughrevealed a two-domain, V-shaped configurationwith the nucleotide bound in the interdomaincleft, in good agreement with models of actinfilaments. The lateral spacing of monomersalong an MreB protofilament is 5.1 nm and ahigh concentration of hydrophobic residueslies at the monomermonomer interface, pro-ducing a strong binding interaction. Unlikeactin and ParM, however, MreB appears toform straight protofilaments and does notassemble into two-filament twisted helices.The ability to express, purify, polymerize, andcrystallize recombinant MreB polymers pro-vides the potential to study structural propertiesof filamentous proteins in a way that has not

been possible with actin or tubulin and shouldresult in some major breakthroughs in thenear future.

MreB Filament Structure and Biophysics

Although MreB appears to form a helical struc-ture in cells, in vitro MreB usually forms verystraight and stiff filaments, and sometimestight, ringlike spirals depending on the poly-merization conditions (van den Ent et al.2001; Esue et al. 2005; Esue et al. 2006). TheMreB protofilament is straight with a longitudi-nal spacing of 5.1 nm. These protofilamentsassociate laterally to build thick, crystallinebundles. Bean and Amann used fluorescent im-aging to examine Alexa-fluor-labeled Cys332-MreB and observed thick bundles 3 micronsin length that appeared very rigid because oftheir lack of thermally driven bending fluctua-tions (Bean and Amann 2008). Similar featureswere observed when a GFP-labeled MreB wasexpressed in fission yeast (Srinivasan et al.2007). Esue et al. (2006)measured the bulk rhe-ological properties of MreB gels (Esue et al.2006). At physiological concentrations, theMreB formed a very stiff gel (10 dyn/cm2)in about 2 min. However, these types of bulkmeasurements are difficult to interpret and therelationship between gel stiffness and MreB fil-ament function remains to be explored. Veryrecent measurements have shown that MreBcontributes significantly to the overall stiffnessof E. coli cells (Wang et al. 2010).

MreB Biochemistry

A consensus view of the in vitro kinetics ofMreB polymerization has beenmore difficult tomeasure than for ParM because different exper-imental groups have produced varying resultsusing different MreB constructs. To date, thebiochemical mechanisms of MreB polymeri-zation and dynamics remain an active area ofresearch.

Esue and coworkers published the firstmeasurements of MreB polymerization kineticsin 2005 (Esue et al. 2005). Using a His-taggedform of T. Maritima MreB1, the authors foundthat polymerization was strongly dependant on




temperature and the concentrations of differentions in solution. They measured a critical con-centration of 3 nM, a hundred times smallerthan that for actin. Further experiments showedthat MreB can use ATP or GTP as a substrateequally well, unlike other proteins such as actin(Esue et al. 2006).

Work in 2008 from Bean and Amann fo-cused on the same protein but measured a dif-ferent behavior (Bean and Amann 2008).These authors purified the native form ofT. Maritima MreB1 without a His-tag throughion exchange and gel filtration chromatographyand differential centrifugation. They found twophases of polymerization, one that uses divalentcations and one that does not. These phaseshave been hypothesized to correspond withthe nucleation and elongation phases of poly-mer growth. Their more native protein wasalso less temperature sensitive than thosereported previously. ATP hydrolysis of theenzyme was fast, implying that in vivo mostpolymerized MreB is in an ADP-bound state.In addition, the authors measured the criticalconcentration for polymerization in the pres-ence of ATP and ADP and found these two val-ues to be close to each otherone requirementfor treadmilling.

Very recent work from Mayer and Amman(2009) successfully purified and polymerizedMreB from B. subtilis (Mayer and Amann2009). This protein is 56% identical and 76%similar to MreB1 from T. maritima. The kineticbehavior of this enzymewas drastically differentfrom that ofT.maritimaMreB1. Polymerizationof B. subtilisMreB required millimolar concen-trations of divalent cations, was favored by lowpH, and was inhibited by monovalent saltsand low temperatures. The authors found thatB. subtilis MreB binds and hydrolyzes ATP andGTP, but surprisingly, does not require nu-cleotide to polymerize. Indeed, the criticalconcentration for polymerization was 900nM regardless of the presence or absence ofnucleotide.

Bean and Amann also produced the firstfluorescently labeled MreB in vitro by bindinga dye to an engineered cysteine residue at posi-tion 332 (Bean and Amann 2008). Using a

fluorescence-resonance energy transfer meas-urement between adjacent monomers in afilament, the authors found similar bulk poly-merization kinetics to that measured with lightscattering. Unlike the measurements fromParM, no one has yet measured the kinetics ofMreB polymer elongation using fluorescencemicroscopy. These measurements are likely tobe very important because of the combinationof elongation and filament bundling seen inMreB.

Because of MreBs propensity to self-as-semble laterally into bundles, many questionsremain about translating in vitro kinetic datato an in vivo context. In vitro bundles can bevery thick and quite crystalline. Is MreB bun-dled inside a cell, and if so, how thick are thesebundles? Fast growing E. coli cells contain40,000 MreB monomers which, if fully poly-merized into a single helical bundle that spansthe cell, suggests a bundle thickness of greaterthan 50 protofilaments (Kruse et al. 2003).Assuming that these filaments are 5 nm inwidth, a tightly packed bundle would have awidth of at least 50 nm. The mechanisms bywhich cells control MreB filament size, geome-try, and conformation remains to be discovered.Future experiments that take advantage ofrecent improvements in electron-microscopyor super-resolution fluorescence imaging willlikely be required to address these issues.

CONCLUSIONS AND OUTLOOK

The ParM-based system of R1 plasmid segrega-tion suggests that at least some bacterial cytos-keletons have relatively simple and specificfunctions. Here, regulated polymerization ofParM mechanically drives two plasmids toopposite cell poles. In contrast, the shear num-ber of functions that have been associated withMreB suggests that the MreB situation may befar more complicated than that observed withother cytoskeletal proteins. The primary out-standing challenge for the field is to understandthe mechanistic basis for these many functionsand distinguish which effects are directly orindirectly mediated by MreB. It is possible thatMreB carries out a small set of relatively simple




functions that are used over and over for differ-ent purposes. For example, MreBs entire func-tion could be to define a helical path along thelong axis of rodlike cells, thereby directly orindirectly organizing the rest of the cellularcomponents. Multiple helical structures havebeen identified in bacteria in the past decade.It will be important to determine whether thesehelices are independent such that the helix isperhaps simply an energetically favorable con-formation for many structures, or whether allof these helices are ultimately patterned interde-pendently, either by MreB or another structure.Another outstanding question concerns thefunctional significance of the MreB helix. Thefact that some cells transition from having heli-cal to medial MreB suggests that differentorganizations of MreB filaments could play dif-ferent functions, but how and why this mayoccur remains mysterious.

By analogy to eukaryotic actin, it has beenlargely assumed that MreB assembly is relatedto its function. However, it is possible that theeukaryotic analogy is stifling our perspectiveon MreB, which could be playing yet unconsid-ered functions. The properties of insertionalpolymerization, treadmilling, and structuralmechanics could be combined and controlledby different proteins for different cellular pur-poses. Alternatively, assembly dynamics mayonly be important for forming the helical struc-ture, and amultitude of MreB-binding proteinsmay have been adapted to tailor MreBs func-tion for each individual activity. Indeed, eukar-yotes usemany regulators to control each step ofactin assembly and use divergent myosins totraffic cargoes along actin. No such assemblyregulators or motor proteins have thus farbeen identified for MreB, and it will be of in-terest to see if they exist and whether they arephylogenetically conserved. Advancing ourunderstanding of MreB will thus require bring-ing to MreB the level of mechanistic detail thatwe currently enjoy for ParM. The analysis ofParM was largely driven by a bottom-up ap-proach of reconstructing the system in vitrofrom its individual parts. It remains unclearwhether this approach will work for morecomplex systems with many components and

regulators. In light of the fact that a decadeago bacteria were not even thought to possessactin proteins, we now know a great deal aboutthe bacterial actin superfamily. The prospect ofcombining newmolecular, genetic, biophysical,and imaging approaches promises to reward thebacterial actin field with a far deeper under-standing in the decade to come.

REFERENCES

Aaron M, Charbon G, Lam H, Schwarz H, Vollmer W,Jacobs-Wagner C. 2007. The tubulin homologue FtsZcontributes to cell elongation by guiding cell wall precur-sor synthesis in Caulobacter crescentus.Mol Microbiol 64:938952.

Alyahya SA, Alexander R, Costa T, Henriques AO, Emonet T,Jacobs-Wagner C. 2009. RodZ, a component of the bacte-rial core morphogenic apparatus. Proc Natl Acad Sci U SA 106: 12391244.

Bean GJ, Amann KJ. 2008. Polymerization properties of theThermotoga maritima actin MreB: roles of temperature,nucleotides, and ions. Biochemistry 47: 826835.

Bean GJ, Flickinger ST, Westler WM, McCully ME, Sept D,Weibel DB, Amann KJ. 2009. A22 disrupts the bacterialactin cytoskeleton by directly binding and inducinga low-affinity state in MreB. Biochemistry 48: 48524857.

Becker E, Herrera NC, Gunderson FQ, Derman AI, DanceAL, Sims J, Larsen RA, Pogliano J. 2006. DNA segregationby the bacterial actin AlfA during Bacillus subtilis growthand development. Embo J 25: 59195931.

Bendezu FO, de Boer PA. 2008. Conditional lethality, divi-sion defects, membrane involution, and endocytosis inmre andmrd shapemutants of Escherichia coli. J Bacteriol190: 17921811.

Bendezu FO, Hale CA, Bernhardt TG, de Boer PA. 2009.RodZ (YfgA) is required for proper assembly of theMreB actin cytoskeleton and cell shape in E. coli.EMBO J 28: 193204.

Bork P, SanderC, Valencia A. 1992. AnATPase domain com-mon to prokaryotic cell cycle proteins, sugar kinases,actin, and hsp70 heat shock proteins. Proc Natl Acad SciU S A 89: 72907294.

Bowman GR, Comolli LR, Zhu J, Eckart M, Koenig M,Downing KH, Moerner WE, Earnest T, Shapiro L.2008. A polymeric protein anchors the chromosomalorigin/ParB complex at a bacterial cell pole. Cell 134:945955.

Carballido-Lopez R. 2006. The bacterial actin-like cytoske-leton. Microbiol Mol Biol Rev 70: 888909.

Carballido-Lopez R, Errington J. 2003. The bacterial cytos-keleton: in vivo dynamics of the actin-like proteinMbl ofBacillus subtilis. Dev Cell 4: 1928.

Carballido-Lopez R, Formstone A, Li Y, Ehrlich SD, NoirotP, Errington J. 2006. Actin homolog MreBH governs cellmorphogenesis by localization of the cell wall hydrolaseLytE. Dev Cell 11: 399409.




Chiu SW, Chen SY, Wong HC. 2008. Localization andexpression ofMreB inVibrio parahaemolyticus under dif-ferent stresses. Appl Environ Microbiol 74: 70167022.

Choi CL, Claridge SA, Garner EC, Alivisatos AP, MullinsRD. 2008. Protein-nanocrystal conjugates support a sin-gle filament polymerization model in R1 plasmid segre-gation. J Biol Chem 283: 2808128086.

Cowles KN, Gitai Z. 2010. Surface association and theMreBcytoskeleton regulate pilus production, localization andfunction in Pseudomonas aeruginosa. Mol Microbiol.

Dam M, Gerdes K. 1994. Partitioning of plasmid R1. Tendirect repeats flanking the parA promoter constitute acentromere-like partition site parC, that expressesincompatibility. J Mol Biol 236: 12891298.

Daniel RA, Errington J. 2003. Control of cellmorphogenesisin bacteria: Two distinct ways to make a rod-shaped cell.Cell 113: 767776.

Defeu Soufo HJ, Graumann PL. 2004. Dynamic movementof actin-like proteins within bacterial cells. EMBO Rep 5:789794.

Defeu Soufo HJ, Graumann PL. 2005. Bacillus subtilis actin-like protein MreB influences the positioning of the repli-cation machinery and requires membrane proteinsMreC/D and other actin-like proteins for proper local-ization. BMC Cell Biol 6: 10.

Defeu Soufo HJ, Reimold C, Linne U, Knust T, Gescher J,Graumann PL. 2010. Bacterial translation elongation fac-tor EF-Tu interacts and colocalizes with actin-like MreBprotein. Proc Natl Acad Sci U S A 107: 31633168.

Derman AI, Becker EC, Truong BD, Fujioka A, Tucey TM,Erb ML, Patterson PC, Pogliano J. 2009. Phylogeneticanalysis identifies many uncharacterized actin-like pro-teins (Alps) in bacteria: Regulated polymerization,dynamic instability and treadmilling in Alp7A. MolMicrobiol 73: 534552.

Divakaruni AV, Baida C,White CL, Gober JW. 2007. The cellshape proteins MreB and MreC control cell morphogen-esis by positioning cell wall synthetic complexes. MolMicrobiol 66: 174188.

Divakaruni AV, Loo RR, Xie Y, Loo JA, Gober JW. 2005. Thecell-shape protein MreC interacts with extracytoplasmicproteins including cell wall assembly complexes inCaulobacter crescentus. Proc Natl Acad Sci 102:1860218607.

Doi M, Wachi M, Ishino F, Tomioka S, Ito M, Sakagami Y,Suzuki A, Matsuhashi M. 1988. Determinations oftheDNA sequence of themreB gene and of the gene prod-ucts of the mre region that function in formation of therod shape of Escherichia coli cells. J Bacteriol 170:46194624.

Dye NA, Pincus Z, Theriot JA, Shapiro L, Gitai Z. 2005. Twoindependent spiral structures control cell shape inCaulo-bacter. Proc Natl Acad Sci 102: 1860818613.

Esue O, Wirtz D, Tseng Y. 2006. GTPase activity, structure,and mechanical properties of filaments assembled frombacterial cytoskeleton protein MreB. J Bacteriol 188:968976.

Esue O, CorderoM,Wirtz D, Tseng Y. 2005. The assembly ofMreB, a prokaryotic homolog of actin. J Biol Chem 280:26282635.

Fenton AK, Lambert C, Wagstaff PC, Sockett RE. 2010.Manipulating each MreB of Bdellovibrio bacteriovorusgives diverse morphological and predatory phenotypes.J Bacteriol 192: 12991311.

Figge RM, Divakaruni AV, Gober JW. 2004. MreB, the cellshape-determining bacterial actin homologue, co-ordinates cell wall morphogenesis in Caulobacter cres-centus. Mol Microbiol 51: 13211332.

Formstone A, Errington J. 2005. A magnesium-dependentmreB null mutant: implications for the role of mreB inBacillus subtilis. Mol Microbiol 55: 16461657.

Formstone A, Carballido-Lopez R, Noirot P, Errington J,Scheffers DJ. 2008. Localization and interactions of tei-choic acid synthetic enzymes in Bacillus subtilis. J Bacter-iol 190: 18121821.

Gan L, Chen S, Jensen GJ. 2008. Molecular organization ofGram-negative peptidoglycan. Proc Natl Acad Sci 105:1895318957.

Garner EC. 2008. GE Prize essay. Understanding a minimalDNA-segregating machine. Science 322: 14861487.

Garner EC, Campbell CS,Mullins RD. 2004.Dynamic insta-bility in a DNA-segregating prokaryotic actin homolog.Science 306: 10211025.

Garner EC, Campbell CS, Weibel DB, Mullins RD. 2007.Reconstitution of DNA segregation driven by assemblyof a prokaryotic actin homolog. Science 315: 12701274.

Gerdes K, Larsen JE, Molin S. 1985. Stable inheritance ofplasmid R1 requires two different loci. J Bacteriol 161:292298.

Gitai Z, Dye N, Shapiro L. 2004. An actin-like gene candetermine cell polarity in bacteria. Proc Natl Acad Sci101: 86438648.

Gitai Z, Dye NA, Reisenauer A, Wachi M, Shapiro L. 2005.MreB actin-mediated segregation of a specific region ofa bacterial chromosome. Cell 120: 329341.

Holmes KC, Popp D, Gebhard W, Kabsch W. 1990. Atomicmodel of the actin filament. Nature 347: 4449.

Holtje JV, Heidrich C. 2001. Enzymology of elongation andconstriction of the murein sacculus of Escherichia coli.Biochimie 83: 103108.

Hu B, Yang G, Zhao W, Zhang Y, Zhao J. 2007. MreB isimportant for cell shape but not for chromosome segre-gation of the filamentous cyanobacterium Anabaena sp.PCC 7120. Mol Microbiol 63: 16401652.

HuangKC,Mukhopadhyay R,WenB,Gitai Z,WingreenNS.2008. Cell shape and cell-wall organization in Gram-negative bacteria. Proc Natl Acad Sci 105: 1928219287.

Iwai N, Nagai K, Wachi M. 2002. Novel S-benzylisothioureacompound that induces spherical cells in Escherichia coliprobably by acting on a rod-shape-determining pro-tein(s) other than penicillin-binding protein 2. BiosciBiotechnol Biochem 66: 26582662.

Jensen RB, Gerdes K. 1997. Partitioning of plasmid R1. TheParM protein exhibits ATPase activity and interacts withthe centromere-like ParR-parC complex. J Mol Biol 269:505513.

Jensen RB, Gerdes K. 1999. Mechanism of DNA segregationin prokaryotes: ParM partitioning protein of plasmid R1co-localizes with its replicon during the cell cycle. EMBOJ 18: 40764084.




Jones LJ, Carballido-Lopez R, Errington J. 2001. Control ofcell shape in bacteria: helical, actin-like filaments inBacil-lus subtilis. Cell 104: 913922.

Karczmarek A,Martinez-Arteaga R, Alexeeva S, Hansen FG,VicenteM,NanningaN, den Blaauwen T. 2007. DNA andorigin region segregation are not affected by the transi-tion from rod to sphere after inhibition of Escherichiacoli MreB by A22. Mol Microbiol 65: 5163.

Kawai Y, Asai K, Errington J. 2009. Partial functional redun-dancy of MreB isoforms, MreB, Mbl and MreBH, in cellmorphogenesis of Bacillus subtilis. Mol Microbiol 73:719731.

Kim SY, Gitai Z, Kinkhabwala A, Shapiro L, Moerner WE.2006. Single molecules of the bacterial actin MreBundergo directed treadmilling motion in Caulobactercrescentus. Proc Natl Acad Sci U S A 103: 1092910934.

Komeili A, Li Z, Newman DK, Jensen GJ. 2006. Magneto-somes are cell membrane invaginations organized bythe actin-like protein MamK. Science 311: 242245.

Kruse T, Bork-Jensen J, Gerdes K. 2005. ThemorphogeneticMreBCD proteins of Escherichia coli form an essentialmembrane-bound complex. Mol Microbiol 55: 7889.

Kruse T, Blagoev B, Lobner-Olesen A, Wachi M, Sasaki K,Iwai N, Mann M, Gerdes K. 2006. Actin homolog MreBand RNA polymerase interact and are both required forchromosome segregation in Escherichia coli. Genes Dev20: 113124.

Kruse T, Moller-Jensen J, Lobner-Olesen A, Gerdes K. 2003.DysfunctionalMreB inhibits chromosome segregation inEscherichia coli. EMBO J 22: 52835292.

Kurner J, Frangakis AS, Baumeister W. 2005. Cryo-electrontomography reveals the cytoskeletal structure of Spiro-plasma melliferum. Science 307: 436438.

Madabhushi R, Marians KJ. 2009. Actin homolog MreBaffects chromosome segregation by regulating topoiso-merase IV in Escherichia coli. Mol Cell 33: 171180.

Mauriello EM, Mouhamar F, Nan B, Ducret A, Dai D, Zus-man DR, Mignot T. 2010. Bacterial motility complexesrequire the actin-like protein, MreB and the Ras homo-logue, MglA. EMBO J 29: 315326.

Mayer JA, Amann KJ. 2009. Assembly properties of theBacillus subtilis actin, MreB. Cell Motil Cytoskeleton 66:109118.

Mazza P, Noens EE, Schirner K, Grantcharova N,MommaasAM, Koerten HK, Muth G, Flardh K, van Wezel GP,Wohlleben W. 2006. MreB of Streptomyces coelicolor isnot essential for vegetative growth but is required forthe integrity of aerial hyphae and spores. Mol Microbiol60: 838852.

Michie KA, Lowe J. 2006. Dynamic filaments of the bacterialcytoskeleton. Annu Rev Biochem 75: 467492.

Moller-Jensen J, Borch J, Dam M, Jensen RB, Roepstorff P,Gerdes K. 2003. Bacterial mitosis: ParM of plasmid R1moves plasmid DNA by an actin-like insertional poly-merization mechanism. Mol Cell 12: 14771487.

Moller-Jensen J, Jensen RB, Lowe J, Gerdes K. 2002. Pro-karyotic DNA segregation by an actin-like filament.EMBO J 21: 31193127.

Munoz-Espin D, Daniel R, Kawai Y, Carballido-Lopez R,Castilla-Llorente V, Errington J, Meijer WJ, Salas M.2009. The actin-like MreB cytoskeleton organizes viral

DNA replication in bacteria. Proc Natl Acad Sci 106:1334713352.

Oda T, Iwasa M, Aihara T, Maeda Y, Narita A. 2009. Thenature of the globular- to fibrous-actin transition.Nature457: 441445.

Polka JK, Kollman JM, Agard DA, Mullins RD. 2009. Thestructure and assembly dynamics of plasmid actin AlfAimply a novel mechanism of DNA segregation. J Bacteriol191: 62196230.

Popp D, Narita A, Oda T, Fujisawa T, Matsuo H, Nitanai Y,Iwasa M, Maeda K, Onishi H, Maeda Y. 2008. Molecularstructure of the ParM polymer and the mechanism lead-ing to its nucleotide-driven dynamic instability. EMBO J27: 570579.

Robertson GT, Doyle TB, Du Q, Duncan L, Mdluli KE,Lynch AS. 2007. A novel indole compound that inhibitsPseudomonas aeruginosa growth by targeting MreB is asubstrate for MexAB-OprM. J Bacteriol 189: 68706881.

Rokney A, Shagan M, Kessel M, Smith Y, Rosenshine I,Oppenheim AB. 2009. E. coli transports aggregated pro-teins to the poles by a specific and energy-dependentprocess. J Mol Biol 392: 589601.

Salje J, Zuber B, Lowe J. 2009. Electron cryomicroscopy ofE. coli reveals filament bundles involved in plasmidDNA segregation. Science 323: 509512.

Savage DF, Afonso B, Chen AH, Silver PA. 2010. Spatiallyordered dynamics of the bacterial carbon fixationmachi-nery. Science 327: 12581261.

Scheffers DJ, Jones LJ, Errington J. 2004. Several distinctlocalization patterns for penicillin-binding proteins inBacillus subtilis. Mol Microbiol 51: 749764.

Shebelut CW, Jensen RB, Gitai Z. 2009. Growth conditionsregulate the requirements for Caulobacter chromosomesegregation. J Bacteriol 191: 10971100.

Shih YL, Kawagishi I, Rothfield L. 2005. The MreB and Mincytoskeletal-like systems play independent roles in pro-karyotic polar differentiation. Mol Microbiol 58:917928.

Shiomi D, Sakai M, Niki H. 2008. Determination of bacte-rial rod shape by a novel cytoskeletal membrane protein.EMBO J 27: 30813091.

Slovak PM, Wadhams GH, Armitage JP. 2005. Localizationof MreB in Rhodobacter sphaeroides under conditionscausing changes in cell shape and membrane structure.J Bacteriol 187: 5464.

Soufo HJ, Graumann PL. 2003. Actin-like proteins MreBand Mbl from Bacillus subtilis are required for bipolarpositioning of replication origins. Curr Biol 13:19161920.

Srinivasan R, Mishra M, Murata-Hori M, BalasubramanianMK. 2007. Filament formation of the Escherichia coliactin-related protein, MreB, in fission yeast. Curr Biol17: 266272.

Srivastava P, Demarre G, Karpova TS, McNally J, ChattorajDK. 2007. Changes in nucleoid morphology and originlocalization upon inhibition or alteration of the actin-homolog, MreB, of Vibrio cholerae. J Bacteriol.

Takacs CN, Poggio S, Charbon G, Pucheault M, Vollmer W,Jacobs-Wagner C. 2010. MreB drives de novo rod mor-phogenesis in Caulobacter crescentus via remodeling ofthe cell wall. J Bacteriol 192: 16711684.




Tiyanont K, Doan T, Lazarus MB, Fang X, Rudner DZ,Walker S. 2006. Imaging peptidoglycan biosynthesis inBacillus subtilis with fluorescent antibiotics. Proc NatlAcad Sci U S A 103: 1103311038.

Touhami A, Jericho MH, Beveridge TJ. 2004. Atomic forcemicroscopy of cell growth and division in Staphylococcusaureus. J Bacteriol 186: 32863295.

van den Ent F, Amos LA, Lowe J. 2001. Prokaryotic origin ofthe actin cytoskeleton. Nature 413: 3944.

van den Ent F, Johnson CM, Persons L, de Boer P, Lowe J.2010. Bacterial actin MreB assembles in complex withcell shape protein RodZ. EMBO J 29: 10811090.

van den Ent F, Lowe J. 2000. Crystal structure of the cell divi-sion protein FtsA from Thermotoga maritima. EMBO J19: 53005307.

van den Ent F, Leaver M, Bendezu F, Errington J, de Boer P,Lowe J. 2006. Dimeric structure of the cell shape proteinMreC and its functional implications. Mol Microbiol 62:16311642.

van den Ent F, Moller-Jensen J, Amos LA, Gerdes K, Lowe J.2002. F-actin-like filaments formed by plasmid segrega-tion protein ParM. EMBO J 21: 69356943.

Vats P, Shih YL, Rothfield L. 2009. Assembly of theMreB-associated cytoskeletal ring of Escherichia coli.Mol Microbiol 72: 170182.

Wagner JK, Galvani CD, Brun YV. 2005. Caulobacter cres-centus requires RodA and MreB for stalk synthesis andprevention of ectopic pole formation. J Bacteriol 187:544553.

Waidner B, SpechtM,Dempwolff F, Haeberer K, Schaetzle S,Speth V, Kist M, Graumann PL. 2009. A novel system ofcytoskeletal elements in the human pathogen Helico-bacter pylori. PLoS Pathog 5: e1000669.

Wang S, Arellano-Santoyo H, Combs PA, Shaevitz JW. 2010.An actin-like cytoskeleton contributes to cell mechanicsin bacteria. Proc Natl Acad Sci 107: 91829185.

White CL, Kitich A, Gober JW. 2010. Positioning cell wallsynthetic complexes by the bacterial morphogenetic pro-teins MreB and MreD. Mol Microbiol 76: 616633.

Yao X, Jericho M, Pink D, Beveridge T. 1999. Thickness andelasticity of gram-negative murein sacculi measured byatomic force microscopy. J Bacteriol 181: 68656875.

Young KD. 2006. The selective value of bacterial shape.Microbiol Mol Biol Rev 70: 660703.


