An intrinsically disordered linker controlling the formation and the stability of the bacterial flagellar hook.

BACKGROUND: In a macro-molecular complex, any minor change may prove detrimental. For a supra-molecular nano-machine like the bacterial flagellum, which consists of several distinct parts with specific characteristics, stability is important. During the rotation of the bacterial flagellar motor, which is located in the membrane, the flagella rotate at speeds between 200 and 2000 rpm, depending on the bacterial species. The hook substructure of the bacterial flagellum acts as a universal joint connecting the motor to the flagellar filament. We investigated the formation of the bacterial flagellar hook and its overall stability between the FlgE subunits that make up the hook and attempted to understand how this stability differs between bacteria. RESULTS: An intrinsically disordered segment plays an important role for overall hook stability and for its structural cohesion during motor rotation. The length of this linker segment depends on the species of bacteria; for Salmonella enterica and Campylobacter jejuni it is approximately 37 and 54 residues, respectively. Few residues of the linker are conserved and mutating the conserved residues of the linker yields non-flagellated cells. In the case of Campylobacter, which rotates its flagella at a speed much higher than that of Salmonella, shortening the linker leads to a rupture of the hook at its base, decreasing cell motility. Our experiments show that this segment is required for polymerization and stability of the hook, demonstrating a surprising role for a disordered region in one of the most finely tuned and closely studied macromolecular machines. CONCLUSIONS: This study reveals a detailed functional characteristic of an intrinsically disordered segment in the hook protein. This segment evolved to fulfill a specific role in the formation of the hook, and it is at the core of the stability and flexibility of the hook. Its length is important in the case of bacteria with high-speed rotating flagella. Finding a way of disrupting this linker in Campylobacter might help in preventing infections.

Function of the conserved FHIPEP domain of the flagellar type III export apparatus, protein FlhA.

The Type III flagellar protein export apparatus of bacteria consists of five or six membrane proteins, notably FlhA, which controls the export of other proteins and is homologous to the large family of FHIPEP export proteins. FHIPEP proteins contain a highly-conserved cytoplasmic domain. We mutagenized the cloned Salmonella flhA gene for the 692 amino acid FlhA, changing a single, conserved amino acid in the 68-amino acid FHIPEP region. Fifty-two mutations at 30 positions mostly led to loss of motility and total disappearance of microscopically visible flagella, also Western blot protein/protein hybridization showed no detectable export of hook protein and flagellin. There were two exceptions: a D199A mutant strain, which produced short-stubby flagella; and a V151L mutant strain, which did not produce flagella and excreted mainly un-polymerized hook protein. The V151L mutant strain also exported a reduced amount of hook-cap protein FlgD, but when grown with exogenous FlgD it produced polyhooks and polyhook-filaments. A suppressor mutant in the cytoplasmic domain of the export apparatus membrane protein FlhB rescued export of hook-length control protein FliK and facilitated growth of full-length flagella. These results suggested that the FHIPEP region is part of the gate regulating substrate entry into the export apparatus pore.

Assembling flagella in Salmonella mutant strains producing a type III export apparatus without FliO.

The type III export apparatus of the Salmonella flagellum consists of six transmembrane proteins (FlhA, FlhB, FliO, FliP, FliQ, and FliR) and three soluble proteins (FliH, FliI, and FliJ). Deletion of the fliO gene creates a mutant strain that is poorly motile; however, suppressor mutations in the fliP gene can partially rescue motility. To further understand the mechanism of suppression of a fliO deletion mutation, we isolated new suppressor mutant strains with partially rescued motility. Whole-genome sequence analysis of these strains found a missense mutation that localized to the clpP gene [clpP(V20F)], which encodes the ClpP subunit of the ClpXP protease, and a synonymous mutation that localized to the fliA gene [fliA(+36T→C)], which encodes the flagellar sigma factor, σ(28). Combining these suppressor mutations with mutations in the fliP gene additively rescued motility and biosynthesis of the flagella in fliO deletion mutant strains. Motility was also rescued by an flgM deletion mutation or by plasmids carrying either the flhDC or fliA gene. The fliA(+36T→C) mutation increased mRNA translation of a fliA'-lacZ gene fusion, and immunoblot analysis revealed that the mutation increased levels of σ(28). Quantitative real-time reverse transcriptase PCR showed that either the clpP(V20F) or fliA(+36T→C) mutation rescued expression of class 3 flagellar and chemotaxis genes; still, the suppressor mutations in the fliP gene had a greater effect on bypassing the loss of fliO function. This suggests that the function of FliO is closely associated with regulation of FliP during assembly of the flagellum.

Randomly selected suppressor mutations in genes for NADH : quinone oxidoreductase-1, which rescue motility of a Salmonella ubiquinone-biosynthesis mutant strain.

The primary mobile electron-carrier in the aerobic respiratory chain of Salmonella is ubiquinone. Demethylmenaquinone and menaquinone are alternative electron-carriers involved in anaerobic respiration. Ubiquinone biosynthesis was disrupted in strains bearing deletions of the ubiA or ubiE genes. In soft tryptone agar both mutant strains swam poorly. However, the ubiA deletion mutant strain produced suppressor mutant strains with somewhat rescued motility and growth. Six independent suppressor mutants were purified and comparative genome sequence analysis revealed that they each bore a single new missense mutation, which localized to genes for subunits of NADH : quinone oxidoreductase-1. Four mutants bore an identical nuoG(Q297K) mutation, one mutant bore a nuoM(A254S) mutation and one mutant bore a nuoN(A444E) mutation. The NuoG subunit is part of the hydrophilic domain of NADH : quinone oxidoreductase-1 and the NuoM and NuoN subunits are part of the hydrophobic membrane-embedded domain. Respiration was rescued and the suppressed mutant strains grew better in Luria-Bertani broth medium and could use l-malate as a sole carbon source. The quinone pool of the cytoplasmic membrane was characterized by reversed-phase HPLC. Wild-type cells made ubiquinone and menaquinone. Strains with a ubiA deletion mutation made demethylmenaquinone and menaquinone and the ubiE deletion mutant strain made demethylmenaquinone and 2-octaprenyl-6-methoxy-1,4-benzoquinone; the total quinone pool was reduced. Immunoblotting found increased NADH : quinone oxidoreductase-1 levels for ubiquinone-biosynthesis mutant strains and enzyme assays measured electron transfer from NADH to demethylmenaquinone or menaquinone. Under certain growth conditions the suppressor mutations improved electron flow activity of NADH : quinone oxidoreductase-1 for cells bearing a ubiA deletion mutation.

FliO regulation of FliP in the formation of the Salmonella enterica flagellum.

The type III secretion system of the Salmonella flagellum consists of 6 integral membrane proteins: FlhA, FlhB, FliO, FliP, FliQ, and FliR. However, in some other type III secretion systems, a homologue of FliO is apparently absent, suggesting it has a specialized role. Deleting the fliO gene from the chromosome of a motile strain of Salmonella resulted in a drastic decrease of motility. Incubation of the ΔfliO mutant strain in motility agar, gave rise to pseudorevertants containing extragenic bypass mutations in FliP at positions R143H or F190L. Using membrane topology prediction programs, and alkaline phosphatase or GFPuv chimeric protein fusions into the FliO protein, we demonstrated that FliO is bitopic with its N-terminus in the periplasm and C-terminus in the cytoplasm. Truncation analysis of FliO demonstrated that overexpression of FliO43-125 or FliO1-95 was able to rescue motility of the ΔfliO mutant. Further, residue leucine 91 in the cytoplasmic domain was identified to be important for function. Based on secondary structure prediction, the cytoplasmic domain, FliO43-125, should contain beta-structure and alpha-helices. FliO43-125-Ala was purified and studied using circular dichroism spectroscopy; however, this domain was disordered, and its structure was a mixture of beta-sheet and random coil. Coexpression of full-length FliO with FliP increased expression levels of FliP, but coexpression with the cytoplasmic domain of FliO did not enhance FliP expression levels. Overexpression of the cytoplasmic domain of FliO further rescued motility of strains deleted for the fliO gene expressing bypass mutations in FliP. These results suggest FliO maintains FliP stability through transmembrane domain interaction. The results also demonstrate that the cytoplasmic domain of FliO has functionality, and it presumably becomes structured while interacting with its binding partners.