ausblenden:
Schlagwörter:
-
Zusammenfassung:
Trimeric autotranporter adhesins (TAAs) represent a group of non- fimbrial, non-pilus adhesins that are widespread in -, -, and g- proteobacteria. They include a number of prominent pathogenicity factors including Yersinia YadA, Neisseria NadA and Bartonella BadA that are involved in pathogen adhesion as well as in the defence against host responses. TAAs are targeted by the type Vc secretion pathway through the outer membrane into the extracellular space. Their architecture follows a general head-stalk-anchor assembly from the N- to the C-terminus. TAAs are highly modular multidomain proteins with a variable number of head and stalk domains that are linked by several types of connector domains. The highly conserved C-terminal membrane anchor harbours the autotransporter function and defines the protein family [1]. In order to explore the domain diversity of trimeric autotransporter adhesins, we set out to produce a dictionary approach (daTAA, available at http://toolkit.tuebingen.mpg.de/dataa) which allows the detailed and automated annotation of TAAs from sequence data [2]. daTAA provides information on the sequence, structure and function of so far 25 different domain types as well as the rules by which these are combined to form the observed long fibers on the cell surface. As complete TAA fibers are not amenable for X-ray crystallography, we turned to solve the structures of single domains in order to assemble them into the full fiber in silico in a later step. The Salmonella adhesin SadA served as a perfect model as it is a highly complex adhesin composed of different types of domains. Closest SadA homologues are found in almost all enterobacteria, such as UpaG, an adhesin involved in the infection process of uropathogenic E. coli. Exploiting the observation that almost all domain types of TAAs begin and end in coiled-coil segments, we produced a pASK IBA - based expression vector system that fuses the extremely stable trimeric pII variant of the GCN4 leucine zipper in register to the N- and C-terminal ends of the domain constructs [3, 4]. We solved the structure of all exemplars of domain types of SadA by molecular replacement and assembled them together with homology models of isolated domains into a complete structural model of the full SadA fiber. Our work successfully approved the applicability of the dictionary approach to understand the structural organization and to perform the annotation of this complex class of proteins.
