Molecular and behavioral analysis of magneto-aerotaxis in Magnetospirillum gryphiswaldense

Published: Oct. 22, 2015, 11 a.m.

Magnetotactic bacteria (MTB) contain nanometer-sized crystals of a magnetic iron mineral enabling directed swimming along geomagnetic field lines. However, although this unique behavior was discovered already 40 years ago, it still has remained poorly understood at the cellular level and the molecular mechanisms responsible for sensing environmental stimuli and transducing signals to the flagellar motors have been unknown. Therefore, the major goal of this thesis was to investigate the swimming behavior of Magnetospirillum gryphiswaldense both at the behavioral and molecular level. Individual motors of tethered M. gryphiswaldense cells were found to rotate both clockwise and counterclockwise with equal speed. Cells swam at speeds of up to 60 µm s-1 and commonly displayed runs of several hundred µm in length. In striking contrast to E. coli, which reorients the cell body between run intervals at random angles, motor switching events caused swimming reversals with reorientation angles close to 180°. The sensory repertoire of M. gryphiswaldense was analyzed by classical macroscopic chemotaxis assays, and aerotaxis was found to be the dominant behavior. In addition to the strong microaerophilic response in oxygen gradients, I observed tactic bands also under anoxic conditions within gradients of the alternative electron acceptor nitrate, suggesting that aerotaxis is part of a general redox or energy taxis mechanism. The aerotactic response of M. gryphiswaldense was furthermore analyzed by recording and tracking single cells under controlled atmospheric conditions in a gas perfusion chamber. Compared to other well-studied bacteria, M. gryphiswaldense displayed unusually low swimming reversal rates (<0.1 s-1) under equilibrium conditions. Abruptly shifting oxygen levels from 2% to 0% only slightly increased reversal rates, whereas a reverse shift from 0% to 2% caused a transient threefold increase in reversal rates that was directly followed by an extraordinarily sustained smooth-swimming phase without return to pre-stimulus levels. Apart from 56 putative genes encoding chemoreceptors that might be involved in magnetotaxis, four putative chemotaxis operons (cheOp1-4) were identified in the genome of M. gryphiswaldense, containing genes commonly involved in signal transduction from chemoreceptors to the flagellar motors. Single or combined deletions of cheOp2-4 did not have any pronounced effect on motility or aerotaxis. In striking contrast, deletion of cheOp1, which comprises only the canonical set of chemotaxis genes (cheAWYBR), caused individual cells to swim straight without reversing, resulting in a complete loss of aerotaxis. When analyzed under oxic conditions, most MTB possess a clear directional preference corresponding to downward movement in their natural habitat, referred to as “polar magneto-aerotaxis”. Although cultivated strains of magnetotactic spirilla were previously assumed to lack any directional preference, in this work polar swimming behavior could be restored in M. gryphiswaldense through repeated cultivation of cells in magnetic fields superimposed on oxygen gradients. Individual cells displayed a gradual bias of swimming runs with one of the cell poles leading that depended on ambient oxygen levels. In anoxic microdroplets, addition of 2% oxygen rapidly reversed the overall swimming direction of the entire population. However, in the absence of CheOp1 swimming polarity could be no longer selected and no reversal of swimming bias was observed. These findings for the first time show that there is a direct molecular link between aerotactic sensing and the determination of magnetotactic polarity, through the sensory pathway CheOp1. In a joint project in the last part of this thesis, I demonstrated how magnetotactic behavior can be manipulated through artificial recruitment of polarly localized CheW1-GFP fusion proteins to midcell anchors. GFP-labelled proteins were trapped by expressing GFP-binding nanobodies on the magnetosome membrane surface (referred to as “nanotrap”). By varying the expression level of the nanobody, a gradual knockdown of magneto-aerotaxis was achieved.