Molecular composition and physiological effects of R-bodies and description of new strains of Caedibacter

The R-body, or refractile body, is a proteinaceous structure comprised of a ribbon coiled inside bacterial cells of genus Caedibacter, endosymbionts of the ciliate Paramecium. In Caedibacter, R-bodies are associated with the phenomenon called “killer trait”. The killer trait is the ability of killer paramecia, those harbouring Caedibacter as symbionts, to kill other paramecia (Caedibacter-free) that are consequently defined sensitives. Released Caedibacter cells carrying an R-body can be ingested by sensitive paramecia. Once inside the digestive vacuoles of a sensitive Paramecium, the R-bodies extrude in a telescopic fashion causing the rupture of the vacuoles and the release of the bacterial cell contents into the Paramecium cytoplasm and finally causing its death. The genetic determinants of R-body production are encoded on extrachromosomal elements like plasmids or bacteriophages. In C. taeniospiralis the involved genes are rebA, rebB, rebC and rebD;. Also in C. caryophilus some genes putatively involved in the production of R-body have been recently characterized. The aims of my thesis were: i) to describe how the structural R-body components from C. taeniospiralis (RebA, RebB, RebC) interact to constitute a R-body by using biochemical techniques as SDS-Page and Western Blot; ii) to test the potential toxicity of recombinant R-bodies expressed in Escherichia coli on a sensitive strain of Paramecium through in vivo tests; iii) to molecularly characterize new strains of Caedibacter species in ciliates applying the full cycle rRNA approach. SDS-Page and Western blot analyses revealed as major components of recombinant R-bodies RebB and RebA, while RebC is present only in a minor proportion; probably RebC carries out functions that are not strictly structural. In vivo analyses permitted to test the potential toxic effects of recombinant R-bodies on the sensitive strain Paramecium tetraurelia 51S. Different experimental conditions were tested. All obtained results confirmed the absence of toxicity from the R-body itself and therefore support the hypothesis that the R-body is acting as releasing vehicle for an unidentified toxin which finally brings the sensitive Paramecium to death. Two novel strains of Caedibacter caryophilus (> 99% similarity with characterized C. caryophilus) were described, one in the cytoplasm of Paramecium primaurelia HC+ and the other in the macronucleus of Spirostomum sp. 72. Closely related bacteria reacting unspecifically with the C. caryophilus FISH probe have been also characterized from the macronucleus of P. caudatum SH42 and the cytoplasm of P. primaurelia CS1. These bacteria were morphologically identified as Holospora caryophila; they cluster phylogentically with Holospora obtusa, but share a sequence similarity of less than 87%. Therefore they should be described as members of a new genus. Additionally, two Rickettsia-like symbionts (>99% similarity with Rickettsia-like bacteria of Diophrys appendiculata) were characterized as bacteria co-infecting the macronucleus of P. caudatum SH42 and the macronucleus of Spirostomum sp. 72. The results of this thesis contribute to a better understanding of the biochemistry of the R-body and its phenotypical effects in the killer trait; moreover, newly described strains of Caedibacter (and of Rickettsia and Holospora caryophila) extend our knowledge about endosymbiont distribution, occurrence and host specificity