Reflections on a sticky situation: how surface contact pulls the trigger for bacterial adhesion

(1)

Article

Reference

Reflections on a sticky situation: how surface contact pulls the trigger for bacterial adhesion

KIRKPATRICK, Clare, VIOLLIER, Patrick

Abstract

Adhesion of bacterial cells to surfaces can be mediated by a wide variety of extracellular structures, which can either recognize specific molecular motifs or adhere in non-specific ways to multiple types of surfaces. The attachment is thought to be highly regulated, but the underlying sensory mechanism(s) are poorly understood. In the α-proteobacterium Caulobacter crescentus, the formation of adhesive organelles is 'hardwired' into the cell cycle regulatory circuitry. In this issue of Molecular Microbiology, Li et al. (2011) employed this model organism to examine the adhesion process and the transition from temporary to permanent attachment using total internal reflection fluorescence (TIRF) microscopy.

Surprisingly, they observed that adhesin production was not only under developmental control, but was also stimulated by surface contact. Initial reversible contact of the pili with the surface was followed by flagellum rotation arrest and subsequent induction of the holdfast to allow irreversible surface adhesion. These findings demonstrate that Caulobacter produces its holdfast only at the appropriate time for [...]

KIRKPATRICK, Clare, VIOLLIER, Patrick. Reflections on a sticky situation: how surface contact pulls the trigger for bacterial adhesion. Molecular Microbiology , 2012, vol. 83, no. 1, p. 7-9

DOI : 10.1111/j.1365-2958.2011.07913.x PMID : 22092444

Available at:

http://archive-ouverte.unige.ch/unige:18248

Disclaimer: layout of this document may differ from the published version.

1 / 1

(2)

MicroCommentary

Reﬂections on a sticky situation: how surface contact pulls the trigger for bacterial adhesion

mmi_7913 7..9

Clare L. Kirkpatrick and Patrick H. Viollier*

Department of Microbiology and Molecular Medicine, University of Geneva, 1 rue Michel-Servet, 1211 Geneva 4, Switzerland.

Summary

Adhesion of bacterial cells to surfaces can be medi- ated by a wide variety of extracellular structures, which can either recognize specific molecular motifs or adhere in non-specific ways to multiple types of surfaces. The attachment is thought to be highly regulated, but the underlying sensory mechanism(s) are poorly understood. In thea-proteobacteriumCau- lobacter crescentus, the formation of adhesive organelles is ‘hardwired’ into the cell cycle regulatory circuitry. In this issue of Molecular Microbiology, Li et al. (2011) employed this model organism to examine the adhesion process and the transition from temporary to permanent attachment using total inter- nal reflection fluorescence (TIRF) microscopy. Sur- prisingly, they observed that adhesin production was not only under developmental control, but was also stimulated by surface contact. Initial reversible contact of the pili with the surface was followed by flagellum rotation arrest and subsequent induction of the holdfast to allow irreversible surface adhesion.

These findings demonstrate that Caulobacter pro- duces its holdfast only at the appropriate time for surface attachment, preventing premature export of the adhesin, which could then be inactivated by

‘curing’ or be masked by occluding particles. Impor- tantly, their results support the notion that the flagel- lum serves as a mechanosensor for adhesion.

Adhesion of bacteria to surfaces, either organic or inor- ganic, is the first step required for formation of a biofilm (Karatan and Watnick, 2009). The ability to adhere pro- vides a major advantage to both free-living and patho-

genic bacteria in terms of colonization of environmental niches or hosts respectively. Indeed, adhesion of pathogens to surfaces within the human body is often vital for their retention in the host, and poses a substantial problem for treatment of persistent infections, particularly in immunocompromised patients (Klemm et al., 2010).

The molecules which mediate adhesion (‘adhesins’) can be polysaccharide or protein-based, and can either form specific structures such as pili and fimbriae or generate a matrix around the cells as they accumulate. In general, bacterial adhesion is a two-step process. Initial contact with a substrate is mediated through extracellular structures such as flagella and pili, and is reversible. The production of polysaccharides leads to the ensuing irreversible attachment to the surface. However, it is not well known how the transition from reversible to irreversible attachment proceeds and how attachment conditions are sensed.

Adhesins can be produced either in response to environmental stimuli or be ‘hardwired’ into the developmental program. One example of the latter case isCaulobacter crescentus, an aquatica-proteobacterium. This organism reproduces according to an asymmetric division program that gives rise to two morphologically distinct cell types, a swarmer and a stalked cell (Skerker and Laub, 2004;

Curtis and Brun, 2010). The swarmer cell possesses a polar flagellum and adhesive pili at the same pole and is motile but incapable of chromosome replication, while the stalked cell cements itself to surfaces using the holdfast located at the tip of the polar stalk. The stalked cell ini- tiates chromosome replication immediately after cell division, while the swarmer cell must differentiate into a stalked cell before it can do so. The holdfast, an N-acetylglucosamine-containing polysaccharide, appears on swarmer cells before the flagellum is shed (Levi and Jenal, 2006) and before stalk growth starts. Remarkably, the holdfast is capable of binding to surfaces with aston- ishingly high strength (in themN range; Tsanget al., 2006) for a naturally occurring adhesive.

Since the holdfast is producedde novoby the swarmer cells in every cell cycle,Caulobacteris an excellent model for examining the bacterial adhesion process from Accepted 28 October, 2011. *For correspondence. E-mail patrick.

[email protected]; Tel. (+41) 22 379 41 75; Fax (+41) 22 379 55 02.

Molecular Microbiology(2012)83(1), 7–9 䊏 doi:10.1111/j.1365-2958.2011.07913.x

First published online 18 November 2011

(3)

beginning to end. In this issue ofMolecular Microbiology, the effect of surface availability on holdfast production by Caulobacter swarmer cells was investigated (Li et al., 2011). Holdfast production in synchronized swarmer cells was monitored in real-time in the presence or absence of a surface. Surprisingly, although holdfast production had been thought to be purely developmentally regulated, holdfast appearance was stimulated by surface contact.

In the presence of a surface, holdfasts appeared 1–2 min after surface presentation, independent of the age of the swarmer cell. By contrast, production of the holdfast occurred much later when cells were grown in large shake flasks to minimize surface contact, as the developmental program took precedence over the absence of surface availability. The timing of holdfast appearance was not affected by the addition of protein synthesis inhibitors, indicating that surface contact signals holdfast formation post-translationally.

Since surface contact acted as a stimulatory signal for holdfast production and adhesion, it follows that the holdfast could not be responsible for mediating the initial reversible phase of attachment. Li et al. (2011) investi- gated this aspect using two mutant strains ofCaulobacter;

DhfsA which is incapable of holdfast production (Smith et al., 2003), andDpilAwhich lacks the structural subunit of Caulobacter pili (Skerker and Shapiro, 2000). In the DhfsAmutant, initial attachment was as efficient as in the wild-type, but never progressed to permanent adhesion.

However, in the DpilA mutant, initial attachment was reduced, and almost no attachment was seen until the developmental stage at which the holdfast is synthesized regardless of surface availability, at which point adhesion, if it occurred, was permanent. From these data it can be concluded that pili promote the initial phase of adhesion, and the pili-mediated reversible attachment stimulates the holdfast production for permanent adhesion.

During the swarmer stage of theCaulobacterlife cycle, the cell swims by rotation of its polar flagellum, which is colocalized with the pili and still present (and moving) at the time of reversible attachment. Therefore, at this stage there is an antagonism between the pili trying to connect the cell to the surface, and the flagellum attempting to propel the cell through the medium and potentially away from the surface. Indeed, in theDpilAmutant, occasional and transient surface attachment through the flagellum was observed, but flagellar rotation continued in these cells until they detached and swam away. However, in the wild-type attachment through pili was quickly followed by stopping of flagellar rotation, suggesting that the flagellum is jammed as the pili pull the cell into close contact with the surface. Li et al. (2011) investigated whether flagellar motor arrest was required for permanent attachment by exposure of the cells in shake flasks to crowding polymers (specifically, Ficoll and Dextran). These crowding agents

reduce the solvent volume available to cells for mixing, thus confining the bacteria into small spaces in which flagellar movement is obstructed. Under these conditions cells aggregated within 1 min of crowding polymer addition, and aggregation was quickly (within 3–4 min) followed by flagellar rotation arrest. Examination of the aggregates showed that holdfasts were present on a large majority of cells, while they were not present (in the same timeframe) on untreated cells, or in cells treated with polymers added to the same viscosity but which did not exert the crowding effect. The same effect was observed with DpilA cells, indicating that alternative methods of blocking flagellar rotation are sufficient for stimulation of holdfast production, presumably by bypassing the need for pili-mediated initial attachment and jamming of the flagellum.

To examine whether surface-induced adhesion is unique toCaulobacter, the authors examined adhesin production in two other flagellated a-proteobacteria, Asticcacaulis biprosthecum and the plant pathogen Agrobacterium tumefaciens.A. biprosthecumcells have a polar holdfast and two lateral stalks (Pateet al., 1973), whileA. tumefa- ciensis stalkless and produces a unipolar polysaccharide (UPP) when attached to a substrate or to other bacteria (Tomlinson and Fuqua, 2009). Strikingly, in both these species, holdfast or UPP production was stimulated by surface contact, and with the same timescale (<5 min) as observed inCaulobacter. In addition to testing attachment to glass surfaces, the authors also examined adhesion ofA. tumefaciensto plant roots, which is likely to be a critical early step in infection of plants by this pathogen.

Again, UPP production was enhanced by contact between the bacteria and the roots, after the initial UPP- independent attachment. This demonstrates that not only is this mechanism conserved in other species of bacteria, but it may also be involved in pathogenesis since it occurs on contact between pathogenic bacteria and the surface of a host cell.

This dissection of the adhesion process from initial attachment through flagellar arrest and development of permanent attachment (Fig. 1) constitutes an important step towards understanding how bacterial biofilms form.

Detailed future studies will be required to elucidate the exact signalling mechanisms that induce holdfast production and whether it is the polysaccharide production or export machinery which is activated on surface contact.

The fact that flagellar arrest is required inCaulobacterfor holdfast appearance, and that inA. biprosthecumthe flagellum is also localized at the pole with the holdfast, suggests that the surface contact signal might be transmit- ted, directly or indirectly, through the components of the flagellar motor when the filament can no longer rotate. It would therefore be of interest to investigate whether there is any physical or genetic interaction between flagellar and holdfast biogenesis machines (Smithet al., 2003) to trans- 8 C. L. Kirkpatrick and P. H. Viollier _䊏

(4)

mit this signal. FliL, an inner membrane component of the flagellar basal body, has been proposed to act as a sensor of flagellar torque or proton flux inProteus mirabilis(Ander- sonet al., 2009) and the flagella formed by aCaulobacter DfliLmutant cannot rotate (Jenalet al., 1994). Even though direct comparison of Caulobacter and P. mirabilis FliL reveals little resemblance at the primary structure level, as putative ancillary proteins of the basal body, it might prove interesting to explore if surface-induction of holdfasts is altered in motile suppressors of the Caulobacter DfliL mutant. Alternatively, as suggested in adhesion models of Vibrio choleraecells (Van Dellenet al., 2008), a transient change in membrane potential might signal holdfast production in Caulobacter through an indirect mechanism.

When the flagellum is jammed, the flux of energizing protons (or other ions) across the membrane–embedded flagellar (MotAB stator) complex is interrupted. This might result in a hyperpolarized membrane that could engage proton flux sensors in the flagellar basal body or in other membrane-embedded complexes, such as a hypothetical component in the holdfast or UPP biosynthetic machinery.

The stimulation of UPP production byA. tumefacienson contact with plant roots strongly suggests that this mechanism is important for initiation of biofilm formation on organic surfaces, which may also be the case for other pathogens. If so, then jamming the signal from flagellum to adhesin export could potentially serve as an anti-biofilm strategy, with clear medical implications.

Acknowledgements

Funding support is from the Swiss National Science Founda- tion (31003A_127287), the Human Frontiers Science

Program (RGP0051/2010), the Société Académique de Genève (2011/70), the Fondation Leenaards and the Univer- sity of Geneva.

References

Anderson, J.K., Smith, T.G., and Hoover, T.R. (2009) Sense and sensibility: flagellum-mediated gene regulation.Trends Microbiol18:30–37.

Curtis, P.D., and Brun, Y.V. (2010) Getting in the loop: regulation of development inCaulobacter crescentus.Microbiol Mol Biol Rev74:13–41.

Jenal, U., White, J., and Shapiro, L. (1994) Caulobacter flagellar function, but not assembly, requires FliL, a non- polarly localized membrane protein present in all cell types.

J Mol Biol243:227–244.

Karatan, E., and Watnick, P. (2009) Signals, regulatory net- works, and materials that build and break bacterial biofilms.

Microbiol Mol Biol Rev73:310–347.

Klemm, P., Vejborg, R.M., and Hancock, V. (2010) Prevention of bacterial adhesion.Appl Microbiol Biotechnol88:451–

459.

Levi, A., and Jenal, U. (2006) Holdfast formation in motile swarmer cells optimizes surface attachment duringCaulo- bacter crescentus development. J Bacteriol 188: 5315–

5318.

Li, G., Brown, P.J., Tang, J.X., Xu, J., Quardokus, E.M., Fuqua, C., and Brun, Y.V. (2011) Surface contact stimulates the just-in-time deployment of bacterial adhesins.Mol Microbiol83:41–51.

Pate, J.L., Porter, J.S., and Jordan, T.L. (1973) Asticcacaulis biprosthecum sp.nov. Life cycle, morphology and cultural characteristics. Antonie Van Leeuwenhoek 39: 569–

583.

Skerker, J.M., and Laub, M.T. (2004) Cell-cycle progres- sion and the generation of asymmetry in Caulobacter crescentus.Nat Rev Microbiol2:325–337.

Skerker, J.M., and Shapiro, L. (2000) Identification and cell cycle control of a novel pilus system in Caulobacter crescentus.EMBO J19:3223–3234.

Smith, C.S., Hinz, A., Bodenmiller, D., Larson, D.E., and Brun, Y.V. (2003) Identification of genes required for synthesis of the adhesive holdfast inCaulobacter crescentus.

J Bacteriol185:1432–1442.

Tomlinson, A.D., and Fuqua, C. (2009) Mechanisms and regulation of polar surface attachment in Agrobacterium tumefaciens.Curr Opin Microbiol12:708–714.

Tsang, P.H., Li, G., Brun, Y.V., Freund, L.B., and Tang, J.X.

(2006) Adhesion of single bacterial cells in the micronew- ton range.Proc Natl Acad Sci USA103:5764–5768.

Van Dellen, K.L., Houot, L., and Watnick, P.I. (2008) Genetic analysis of Vibrio cholerae monolayer formation reveals a key role for DeltaPsi in the transition to permanent attachment.J Bacteriol190:8185–8196.

Fig. 1. Stages in surface adhesion ofCaulobacter crescentus.

Surface stimulation of adhesion 9