Characterisation of the dissimilatory reduction of Fe(III)- oxyhydroxide at the microbe – mineral interface: the application of STXM–XMCD V. S. COKER, 1 J. M. BYRNE, 1 N. D. TELLING, 2 G. VAN DER LAAN, 3 J. R. LLOYD, 1 A. P. HITCHCOCK, 4 J. WANG 5 AND R. A. D. PATTRICK 1 1 School of Earth, Atmospheric & Environmental Sciences and Williamson Research Centre for Molecular Environmental Science, University of Manchester, Manchester, UK 2 Institute for Science and Technology in Medicine, Keele University, Stoke-on-Trent, UK 3 Diamond Light Source Ltd, Didcot, Oxfordshire, UK 4 Brockhouse Institute for Materials Research, McMaster University, Hamilton, ON, Canada 5 Canadian Light Source Inc., University of Saskatchewan, Saskatoon, SK, Canada ABSTRACT A combination of scanning transmission X-ray microscopy and X-ray magnetic circular dichroism was used to spatially resolve the distribution of different carbon and iron species associated with Shewanella oneidensis MR- 1 cells. S. oneidensis MR-1 couples the reduction of Fe(III)-oxyhydroxides to the oxidation of organic matter in order to conserve energy for growth. Several potential mechanisms may be used by S. oneidensis MR-1 to facili- tate Fe(III)-reduction. These include direct contact between the cell and mineral surface, secretion of either exog- enous electron shuttles or Fe-chelating agents and the production of conductive ‘nanowires’. In this study, the protein ⁄ lipid signature of the bacterial cells was associated with areas of magnetite (Fe 3 O 4 ), the product of dis- similatory Fe(III) reduction, which was oversaturated with Fe(II) (compared to stoichiometric magnetite). How- ever, areas of the sample rich in polysaccharides, most likely associated with extracellular polymeric matrix and not in direct contact with the cell surface, were undersaturated with Fe(II), forming maghemite-like (c-Fe 2 O 3 ) phases compared to stoichiometric magnetite. The reduced form of magnetite will be much more effective in environmental remediation such as the immobilisation of toxic metals. These findings suggest a dominant role for surface contact-mediated electron transfer in this study and also the inhomogeneity of magnetite species on the submicron scale present in microbial reactions. This study also illustrates the applicability of this new synchro- tron-based technique for high-resolution characterisation of the microbe–mineral interface, which is pivotal in controlling the chemistry of the Earth’s critical zone. Received 6 September 2011; accepted 24 February 2012 Corresponding author: V. S. Coker. Tel.: +44(0)161 275 3803; fax: +44(0)161 306 9361; e-mail: vicky.coker@ manchester.ac.uk INTRODUCTION Microbial Fe(III) reduction is widespread in the subsurface and has significant environmental consequences, as it has been shown to control the mobility of radionuclides, toxic metals and organic molecules in many different environments (Lov- ley et al., 2004). Therefore, it is of great interest to identify the precise mechanisms that dissimilatory Fe(III)-reducing bacteria use to reduce Fe(III)-bearing minerals. The Gram- negative, facultative anaerobe Shewanella oneidensis MR-1 is able to reduce solid-phase Fe(III) oxyhydroxides by coupling the reaction to the oxidation of organic matter (Lovley et al., 1989; Nealson & Saffarini, 1994). This electron-transfer pro- cess can result in the formation of a range of Fe(II)-rich phases, including magnetite, green rusts, siderite or vivianite (Fredrickson et al., 1998; Ona-Nguema et al., 2002; Lloyd, 2003). As well as direct contact between cell surface and mineral, transferring electrons via outer membrane cyto- chromes (Myers & Myers, 1992; Shi et al., 2007), there are a number of alternative methods that this bacterium can use to facilitate the electron-transfer process, particularly to reduce less bioavailable solid-phase minerals. These include the secre- tion of electron-shuttling compounds (Newman & Kolter, 2000; Marsili et al., 2008; von Canstein et al., 2008), the use Ó 2012 Blackwell Publishing Ltd 347 Geobiology (2012), 10, 347–354 DOI: 10.1111/j.1472-4669.2012.00329.x