526 | JULY 2004 | VOLUME 2 www.nature.com/reviews/micro

Many bacteria show polar characteristics, suchas the positioning of a flagellum at one end ofthe cell. However, the mechanisms involved inestablishing this polarity are poorly understood.A recent study from Lucy Shapiro and col-

leagues has made important progress in under-standing these processes, identifying a masterregulator of polarity in Caulobacter crescentus.

The C. crescentus life cycle involves two dif-ferent cell types, both with specialized structureslocated at one pole of the cell. Several proteinsthat are involved in the development of thesestructures and that have corresponding polardistributions have been identified, providinguseful markers of polarity. Shapiro and col-leagues made use of these features to investigatewhether the actin-like protein MreB is requiredfor polarization in C. crescentus.

MreB has a distinctive localization pattern,forming a spiral structure that extends along the length of C. crescentus cells. By analogy toeukaryotic actin, MreB molecules might havean intrinsic polarity, so the spirals they formcould be used for the asymmetric localization ofmolecules required for the development ofpolar structures. To test this, the authorsanalysed the effect of MreB depletion on thedistribution of four signalling proteins — PleC,DivJ, CckA and DivK — that are required forpolar development in C. crescentus. Depletion ofMreB abolished the polar foci that are usuallyformed by all four proteins at certain points inthe cell cycle, consistent with a role for MreB asa global regulator of polarity.

Importantly, MreB seems to be activelyrequired for specifying polarity, rather than hav-ing a passive role in protein localization. UnlikeCckA and DivK, which form foci at both polesof C. crescentus cells, PleC and DivJ are asym-metrically distributed at certain points in thecell cycle, localizing to only one pole. WhenMreB was depleted and then re-expressed,although polar foci of PleC and DivJ wererestored, these were located at the wrong pole in~50% of cells. This indicates that when MreB isdepleted, cells lose all memory of their initialpolarity, so that when MreB is re-expressedpolarity becomes randomized. MreB musttherefore be required for the initial decision thatdetermines which pole of the cell is which.

This study shows intriguing similaritiesbetween the establishment of polarity in C. cres-centus and the corresponding processes ineukaryotic cells, in which actin has a centralrole. It will be interesting to see whether similarmechanisms operate in other bacteria.

Louisa Flintoft

References and linksORIGINAL RESEARCH PAPERGitai, Z., Dye, N. & Shapiro, L. An actin-like gene can determinecell polarity in bacteria. Proc. Natl Acad. Sci. USA 101,8643–8648 (2004)WEB SITELucy Shapiro’s laboratory: http://caulo.stanford.edu/shaplab/

RNA interference (RNAi) experiments haveprovided the long-awaited experimental proofthat the cell-wall component α-(1,3)-glucan is avirulence factor in the dimorphic fungalpathogen Histoplasma capsulatum.

The yeast form of H. capsulatum is anintracellular pathogen and is found withinmacrophages, where it survives andproliferates. Techniques for the direct geneticanalysis of H. capsulatum have been developedand can be used in conjunction with genome-sequence data to identify virulence factors.However, these techniques are slow, oftentaking weeks or months to obtain results. Now,Rappleye et al. have developed an efficientRNAi system for use in H. capsulatum.

The system is plasmid-based and was designedand optimized using an easily quantifiabletarget — green fluorescent protein (GFP). Testvectors were constructed to determine themost efficient plasmid design for gene silencingin H. capsulatum. Transformation of an

H. capsulatum strain that stably expresses GFP with a construct in which a gfp hairpin isformed resulted in an average sixfold reductionin fluorescence. Serial passage of the H.capsulatum strain expressing this plasmidshowed that the vector and the RNAi effect werestably maintained. Rappleye et al. went on toshow that the RNAi effect was still observedonce the transformed H. capsulatum cells hadcolonized murine macrophages.

α-(1,3)-glucan has long been suspected to bea virulence factor in H. capsulatum —spontaneous avirulent mutants can be isolatedthat are referred to as ‘smooth’ colony variantsas they lack α-(1,3)-glucan. However, untilnow, there has been no direct experimentalevidence of a causal link between α-(1,3)-glucan and virulence. Rappleye et al. followedup their proof-of-concept work by targetingthe H. capsulatum α-(1,3)-glucan synthasegene, AGS1, which encodes an essential enzymefor α-(1,3)-glucan biosynthesis.Transformation of H. capsulatum with anRNAi vector targeting AGS1 generated‘smooth’ colonies, and immunofluorescencestudies confirmed that AGS1 had beeneffectively silenced in most transformants. Therelative virulence of H. capsulatum yeast cellsexpressing the AGS1–RNAi construct wasanalysed in a macrophage monolayer and in a

murine infection model and it was shown thatthe virulence of these cells was substantiallyreduced compared with controls, thus providingfunctional evidence that α-(1,3)-glucan is avirulence factor in H. capsulatum.

Sheilagh Clarkson

References and linksORIGINAL RESEARCH PAPER Rappleye, C. A., Engle, J. T.& Goldman, W. E. RNA interference in Histoplasma capsulatumdemonstrates a role for α-(1,3)-glucan in virulence. Mol. Microbiol. (21 May 2004) doi: 10.1111/j.1365-2958.2004.04131.x

Interferencecomes good

F U N G A L PAT H O G E N I C I T Y

Polar exploration

C E L L U L A R M I C R O B I O LO G Y

R E S E A R C H H I G H L I G H T S

Histoplasma capsulatum proliferating within macrophages.The red fluorescence corresponds to immunoreactivity to α-(1,3)-glucan. Image kindly provided by Christian Coleman,Chad Rappleye and William Goldman and used withpermission of Blackwell Publishing © (2004).