The molecular mechanism of the type IVa pilus motors

Abstract

Type IVa pili are protein filaments essential for virulence in many bacterial pathogens; they extend and retract from the surface of bacterial cells to pull the bacteria forward. The motor ATPase PilB powers pilus assembly. Here we report the structures of the core ATPase domains of Geobacter metallireducens PilB bound to ADP and the non-hydrolysable ATP analogue, AMP-PNP, at 3.4 and 2.3 Å resolution, respectively. These structures reveal important differences in nucleotide binding between chains. Analysis of these differences reveals the sequential turnover of nucleotide, and the corresponding domain movements. Our data suggest a clockwise rotation of the central sub-pores of PilB, which through interactions with PilC, would support the assembly of a right-handed helical pilus. Our analysis also suggests a counterclockwise rotation of the C2 symmetric PilT that would enable righthanded pilus disassembly. The proposed model provides insight into how this family of ATPases can power pilus extension and retraction.

Structural insights into the mechanism of Type IVa pilus extension and retraction ATPase motors

“…Crystals appeared after about 5 months of setting up of the trials suggesting N-terminal region got proteolyzed in situ. A similar situation was observed in previously characterized PilB Gme structures as well [28]. While in PilB Tth , the N1D domain was removed to aid crystallization [27].…”

“…The data suggest that there is negative co-operativity in binding ATP as when ATP occupies two sites, then none of the other sites in a hexamer can bind ATP. Available literature suggest that ATP-bound PilB chains adopt open-state conformation [27,28]. In light of the existing literature, ITC experiments and structural data presented here probably suggests that two open chains in a PilB Gsu hexamer bind ATP with high affinity.…”

“…A. The central orifice resembles the dumbbell shape as reported earlier [28] with two subpores having a diameter of~23…”

“…3). All the structures of PilB solved so far have ligands present in their active site, but PilB Gsu does not contain ATP/ADP [27,28]. Even in the absence of ATP/ADP, PilB Gsu forms a distorted hexamer with twofold symmetry as observed in the previously solved structures [25,26,32].…”

“…4). Based on the existing literature [27,28] and the fact that we observed all three chains having distinct domain motions, we hypothesized that PilB Gsu could adopt three different conformations, [25,26,28].…”