Contractile Injection Systems

We studied such a divergent T6SS in the bacterial symbiont Amoebophilus asiaticus. The in situ architecture revealed three modules, including a contractile sheath-tube, a baseplate, and an anchor. We found that similar to the sheath-tube, the baseplate module also undergoes a conformational change upon firing. Lateral baseplate interactions coordinated T6SSs in hexagonal arrays. The system mediated interactions with host membranes and may participate in phagosome escape. Sequence analyses suggested that T6SSs are widespread and lilkely evolved multiple times independently. external page See publication

Microbial communities are shaped by cell-cell interactions. Although archaea are often found in associations with other microorganisms, the mechanisms structuring these communities are poorly understood.

We reported on the structure and function of haloarchaeal CISs. Using a combination of functional assays and time-lapse imaging, we showed that Halogeometricum borinquense exhibits antagonism toward Haloferax volcanii by inducing cell lysis and inhibiting proliferation. This antagonism is contact-dependent and requires a functional CIS, which is encoded by a gene cluster that is associated with toxin-immunity pairs. Cryo-focused ion beam milling and imaging by cryoET revealed that these CISs are bound to the cytoplasmic membrane, resembling the bacterial T6SS. We showed that related T6SS gene clusters are conserved and expressed in other haloarchaeal strains, which exhibit antagonistic behavior. Our data provide a mechanistic framework for understanding how archaea may shape microbial communities and affect the food webs they inhabit. external page See publication

Extracellular contractile injection systems (eCISs)

Extracellular contractile injection systems (eCIS) are assembled inside the bacterial cell but then released into the medium through cell lysis. Via their tail fibers, they can specifically bind to the cell surface of their target, followed by firing and effector translocation. external page In an initial collaborative study including Nick Shikuma, Grant J. Jensen and Dianne K. Newman, we studied an eCIS and showed that it triggered metamorphosis of marine tubeworm larvae. These eCIS furthermore formed a unique sea mine-like superstructure composed of ~100 individual eCIS. In a external page follow-up study, we also identified the metaorphosis-inducing effector that was translocated by these eCIS. 

In the genome of the marine bacterium Algoriphagus machipongonensis, we discovered a novel putative CIS gene cluster. Using an integrative approach, we showed that the system is compatible with an eCIS mode of action. Our cryoEM structure revealed several features that differ from those seen in other CISs: a 'cap adaptor' located at the distal end, a 'plug' exposed to the tube lumen, and a 'cage' formed by massive extensions of the baseplate. These elements are conserved in other CISs, and our genetic tools identified that they are required for assembly, cargo loading and function. Furthermore, our atomic model highlights specific evolutionary hotspots and will serve as a framework for understanding and re-engineering eCISs. external page See publication

Also multicellular cyanobacterial frequently harbor CIS gene clusters. We characterized a CIS in Anabaena, which showed features that were distinct from eCISs and T6SSs. CryoET of focused ion beam-milled cells revealed that CISs were anchored in thylakoid membrane stacks, facing the cell periphery. Single particle cryoEM showed that this unique in situ localization was mediated by extensions of tail fibre and baseplate components. Upon stress, cyanobacteria induced the formation of ghost cells, presenting thylakoid-anchored CISs to the environment. Functional assays suggest that these CISs may mediate ghost cell formation and/or interactions of ghost cells with other organisms. Collectively, these data provide a framework for understanding the evolutionary re-engineering of CISs and potential roles of these CISs in cyanobacterial programmed cell death. external page See publication

While CIS gene clusters are highly abundant across diverse bacterial phyla, representatives from Gram-positive organisms remain poorly studied. We characterized a CIS in the Gram-positive multicellular model organism Streptomyces coelicolor and showed that, in contrast to most other CIS, S. coelicolor CIS (CIS^Sc) mediate cell death in response to stress and impact cellular development. CISSc were expressed in the cytoplasm of vegetative hyphae and are not released into the medium. Our cryoEM structure enabled the engineering of non-contractile and fluorescently tagged CIS^Sc assemblies. CryoET imaging showed that CIS^Sc contraction was linked to reduced cellular integrity. Fluorescence light microscopy furthermore revealed that functional CIS^Sc mediate cell death upon encountering different types of stress. The absence of functional CIS^Sc had an impact on hyphal differentiation and secondary metabolite production. Finally, we identified three putative effector proteins, which when absent, phenocopied other CIS^Sc mutants. Our results provide new functional insights into CISs in Gram-positive organisms and a framework for studying novel intracellular roles, including regulated cell death and life-cycle progression in multicellular bacteria. external page See publication

In a external page follow-up study, we identified the membrane protein CisA that is required for firing and function and may anchor the apparatus to the cell envelope.