Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hirano, T. Articles by Hughes, K. T. Search for Related Content PubMed PubMed Citation Articles by Hirano, T. Articles by Hughes, K. T.

 Previous Article  |  Next Article 

Journal of Bacteriology, June 2009, p. 3938-3949, Vol. 191, No. 12
0021-9193/09/$08.00+0     doi:10.1128/JB.01811-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Mutations in Flk, FlgG, FlhA, and FlhE That Affect the Flagellar Type III Secretion Specificity Switch in Salmonella enterica

Takanori Hirano,¹ Shino Mizuno,² Shin-Ichi Aizawa,² and Kelly T. Hughes^1*

Department of Biology, University of Utah, Salt Lake City, Utah 84112,¹ CREST Soft Nano-machine Project, JST, Innovation Plaza Hiroshima, 3-10-23 Kagamiyama, Higashi-Hiroshima 739-0046, Japan²

Received 23 December 2008/ Accepted 1 April 2009

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Upon completion of the flagellar hook-basal body (HBB) structure, the flagellar type III secretion system switches from secreting rod/hook-type to filament-type substrates. The secretion specificity switch has been reported to occur prematurely (prior to HBB completion) in flk-null mutants (P. Aldridge, J. E. Karlinsey, E. Becker, F. F. Chevance, and K. T. Hughes, Mol. Microbiol. 60:630-643, 2006) and in distal rod gene gain-of-function mutants (flgG* mutants) that produce filamentous rod structures (F. F. Chevance, N. Takahashi, J. E. Karlinsey, J. Gnerer, T. Hirano, R. Samudrala, S. Aizawa, and K. T. Hughes, Genes Dev. 21:2326-2335, 2007). A fusion of β-lactamase (Bla) to the C terminus of the filament-type secretion substrate FlgM was used to select for mutants that would secrete FlgM-Bla into the periplasmic space and show ampicillin resistance (Ap^r). Ap^r resulted from null mutations in the flhE gene, C-terminal truncation mutations in the flhA gene, null and dominant mutations in the flk gene, and flgG* mutations. All mutant classes required the hook length control protein (FliK) and the rod cap protein (FlgJ) for the secretion specificity switch to occur. However, neither the hook (FlgE) nor the hook cap (FlgD) protein was required for premature FlgM-Bla secretion in the flgG* and flk mutant strains, but it was in the flhE mutants. Unexpectedly, when deletions of either flgE or flgD were introduced into flgG* mutant strains, filaments were able to grow directly on the filamentous rod structures.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The bacterial flagellum is a biological nanomachine that includes an ion-powered rotary motor and functions as a locomotive organelle (25, 38). The motor or basal body component of the flagellar structure is embedded in the cell wall and membranes and is connected to the external rigid filament by a flexible coupler called the hook. The flexible hook acts as a universal joint between rigid rod and rigid filament structures but still propagates rotational motion generated inside the cell.

In general, inner structural components of the flagellum are assembled before the outer ones. Basal structure assembly initiates when a single protein (FliF) polymerizes into a ring structure, the MS ring, in the cytoplasmic membrane (26). The MS ring serves as a structural nucleus, providing a mounting surface for a cytoplasmic ring structure, the C ring, on its inner (cytoplasmic) face, which forms the rotor of the flagellar motor (27). The membrane-associated part of the flagellar type III secretion (T3S) system forms within the MS ring (32, 33, 37, 42). The proton motive force is used as the energy source for secretion (41, 49). The rod structure acts as a drive shaft and extends from the periplasmic face of the MS ring through the peptidoglycan layer to the outer membrane. The rod and other axial components are secreted from the cytoplasm via flagellar T3S into a narrow central channel within the growing structure and polymerize onto the distal end (10, 19). At this point in assembly, the FlgH and FlgI proteins, which are secreted into the periplasm by the cell's general secretory (Sec) pathway, polymerize around the distal rod to form a bushing-like structure (the PL ring) (20) and function as an outer membrane pore (6). The hook structure then assembles at the distal end of the rod and extends from the cell surface for about 55 nm (17). The secretion specificity switch is determined by the FlhB component of the flagellar T3S system and follows hook completion. Following hook-basal body (HBB) completion, the hook length ruler protein FliK interacts with the C-terminal cytoplasmic domain of FlhB (FlhB_CC) within the flagellar T3S system. This interaction is thought to result in a conformational change in FlhB_CC to switch secretion from the rod/hook-type substrates to filament-type substrates (34). This secretion specificity switch is critical for determining the length of the hook. After HBB completion and the secretion specificity switch, the hook-filament junction proteins and the filament cap (FliD) are added. Finally, the filament polymerizes under the FliD cap, extending to lengths of up to 10 µm. Multiple peritrichous flagella form on Salmonella that bundle behind the cell so that the cell can swim in liquid conditions.

HBB completion and the secretion specificity switch are coupled to the regulation of flagellar gene expression (4, 7). The flagellar regulon is organized into a transcriptional hierarchy of three promoter classes. At the top of the flagellar transcriptional hierarchy is the class 1 flhDC operon. When FlhD and FlhC are produced, they form a heteromultimeric complex that activates class 2 HBB gene expression (54). In addition, the regulatory genes flgM and fliA are expressed with the HBB genes. The fliA gene encodes a flagellum-specific transcription factor, ²⁸, which directs the transcription of class 3 genes (48). The class 3 gene products are needed after HBB completion and include the filament gene fliC and the genes of the chemosensory system. ²⁸-dependent flagellar gene expression is inhibited by the anti-²⁸ factor FlgM prior to hook completion (2). Upon hook completion, FlgM is secreted through the flagellar structure into the culture supernatant (18, 28). Free ²⁸ can then direct transcription of flagellar class 3 genes.

In the last decade, the protein export function of the flagellum has been investigated. The export system consists of six integral membrane proteins (39). Two proteins, FlhA and FlhB, have been extensively characterized for structure and function (12, 13, 40, 50, 51), whereas four proteins, FliO, FliP, FliQ, and FliR, are not yet understood (11, 46). FlhA and FlhB each possess an N-terminal transmembrane domain and a large C-terminal cytoplasmic domain (36).

FlhA and FlhB are encoded in a single operon together with FlhE. Since FlhA and FlhB play essential roles in flagellar protein secretion (21, 39), it seemed plausible that FlhE might play a role in flagellar secretion. FlhE contains an N-terminal signal sequence for export into the periplasm via the Sec pathway. A null mutation in the flhE gene resulted in a swarming defect but had no apparent effect on flagellar assembly or swimming behavior (36, 53). Thus, a specific function for FlhE remains to be determined.

It has been known for a long time that any mutant defective in HBB formation will not undergo the secretion specificity switch. In HBB assembly-defective mutants, FlgM is retained in the cytoplasm and inhibits ²⁸-dependent class 3 transcription. Selection for class 3 transcription in HBB assembly mutants has yielded many classes of mutants, depending on the particular HBB assembly defect use in the selection. In all cases, the mutants have reduced FlgM levels, including null alleles of flgM, or enhanced ²⁸ function, including ²⁸ gain-of-function mutants that have a reduced affinity for FlgM. Two unusual mutant classes that have reduced FlgM levels specific to HBB assembly mutants that are defective in PL-ring assembly (flgA, flgH, or flgI mutants) are flk locus loss-of-function mutants and distal rod gene flgG gain-of-function mutants (flgG*). In both cases, the mutants result in the secretion specificity switch and secretion of FlgM into the periplasm, where it is degraded.

In this study, we sought to capitalize on our recent finding that a fusion of β-lactamase (Bla) to FlgM (FlgM-Bla) could be utilized to monitor substrate specificity switching. In the PL-ring-defective mutants, flk-null allele or flgG* allele FlgM-Bla is secreted into the periplasm but the C-terminal Bla fusion prevents FlgM degradation and confers ampicillin resistance (Ap^r) on the cell. Thus, we can use the Bla fusion to select for mutants that allow FlgM-Bla secretion into the periplasm under conditions where this does not normally occur. Thus, this high background of flgM-null alleles and FlgM bypass mutants in ²⁸ would not complicate the isolation of rare and potentially novel mutant types that affect the secretion specificity switch using Ap^r selection with FlgM-Bla.

Here we report the use of FlgM-Bla secretion to isolate two new classes of mutants that affect the flagellar secretion specificity switch. The FlgM-Bla fusion contains the filament-type secretion signal and confers Ap^r when exported into the periplasm (1). We isolated Ap^r Fla⁺ and PL-ring mutant strains that secreted FlgM-Bla into the periplasm. In addition to previously isolated mutations in the flk and flgG genes, null alleles of the flhE gene and flhA nonsense mutations resulting in C-terminally truncated FlhA were obtained.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains, media, and standard genetic techniques. The bacterial strains used in this study are listed in Table 1 L broth (LB) and motility agar plates were prepared as described before. A 100-µg/ml concentration of Ap was used for LB agar plates, whereas 30 µg/ml was used in MacConkey (Mac)-lactose (lac) agar medium. The generalized transducing phage P22 HT105/1 int-201 was used for all transductional crosses (9).

View this table: [in this window] [in a new window]

TABLE 1. Bacterial strains used in this study

Strain construction. -Red-based recombination was used for targeted chromosomal mutagenesis as described previously (8, 22).

Screening for Ap^r mutants. To isolate mutants that allow FlgM-Bla secretion into the periplasm in a flgHI background, 100-µl volumes of 86 independent TH10406 overnight cultures were plated onto an LB-Ap (100 µg/ml) agar plate and incubated at 37°C overnight. Two hundred fifty-two Ap^r colonies were picked and streaked onto the LB-Ap (100 µg/ml) agar plate containing 1% arabinose to induce a second wild-type copy of the flk gene placed under the control of the promoter of the arabinose-encoding locus on the genome. One hundred twenty-one Ap^s isolates in the presence of 1% arabinose were discarded as recessive flk mutants. The other 131 Ap^r mutants were analyzed further.

Ap MIC assays. Ap MIC assays were performed as described previously (31).

Flagellar immunostaining. Cells were inoculated into LB medium by 1/100 dilution from overnight cultures. The cells were incubated at 30°C for 3 h. The cover glass and glass slide were spaced by two layers of double-sided tape to avoid shear force during flagellar staining. A 60-µl volume of poly-L-lysine was placed in the space between the glass slide and the cover glass just before 60 µl of the cells was added to the space. A final concentration of 5% formaldehyde was used to fix the cells. A petri dish was used as a humid chamber by including a wet Kimwipe to avoid sample drying. The glass slide was placed upside down in the petri dish and incubated for 5 min at room temperature to ensure cell attachment to the surface of the cover glass. Further incubation steps in the staining process were done in the petri dish. Then, 200 µl of phosphate-buffered saline (PBS) was used to wash excess unattached cells away. Forty microliters of diluted anti-FliC polyclonal antibody was added, and the glass slide was incubated for 1 h at room temperature. After washing with 200 µl of PBS, 40 µl of secondary goat anti-rabbit antibody conjugated with fluorescein isothiocyanate (FITC; 4 µg/ml; Molecular Probes) was applied and the sample was incubated for 30 min. Samples were washed with 200 µl of PBS and then incubated for 5 min with 10 µl of 4',6'-diamidino-2-phenylindole (DAPI; 1 µg/ml; Molecular Probes) for DNA staining. Two hundred microliters of PBS was used as a wash, and 2 µl of FM4-64 (0.01 µg/ml; Molecular Probes) was used for membrane staining and incubation for 5 min. After a 200-µl PBS wash, 40 µl of poly-L-lysine was applied and the glass slide was sealed with clear nail polish. Samples were observed using a DeltaVision deconvolution fluorescence microscope (Applied Precision). Images were taken and deconvolved by softWoRx v3.4.2 software (Applied Precision).

Electron microscopy. For visualization by transmission electron microscopy (TEM), isolated flagellar structures were stained with 1% phosphotungstic acid (PTA; pH 7 or 5) and observed with a JEOL 1200Ex electron microscope at 80 kV.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Rationale behind selection for premature flipping of the secretion specificity switch. When HBB assembly is completed, the flagellar T3S system changes specificity from the rod/hook type of secretion substrate to the filament type (15, 39). This is catalyzed by the FliK protein, which measures rod-hook length and switches secretion specificity through interaction of the C terminus of FliK with the FlhB component of the flagellar T3S system (43).

FliK has been shown to be secreted during rod-hook assembly but only switches secretion specificity when the hook reaches its final length (35). It is believed that FliK makes a pause during secretion through an interaction of its N terminus with the assembled hook cap. If the C terminus of FliK is located near the FlhB component of the flagellar T3S system during this pause in secretion, the secretion specificity switch could flip (34, 44). In strains defective in PL-ring formation, flk locus mutants were isolated that allowed the secretion specificity to switch (1, 23). Our working model is that Flk forms a cork at the base of the flagellum to prevent access of both cytoplasmic FliK and the C terminus of secreted FliK to FlhB and prevent flipping of the switch prior to hook completion (see Discussion). It is when the hook reaches its final length that the C terminus of FliK passes the Flk cork, allowing interaction with FlhB.

In an attempt to locate components of the flagellar basal structure that interact with Flk, we decided to isolate flk bypass mutants that allowed the secretion of an FlgM-Bla fusion into the periplasm in either Fla⁺ or PL-ring mutant strains. We surmised that such mutants would allow access of FliK to FlhB prior to HBB completion and result in premature flipping of the flagellar secretion specificity switch. Perhaps an affinity site within the C ring exists for recruitment of Flk, which, if mutated, would give the same phenotype as a flk-null mutation. The C terminus of FlhA could include such an affinity site (see results below).

Loss of FlhE results in FlgM-Bla secretion into the periplasm in the PL-ring mutant background. In the PL-ring^– (flgHI) background, HBB assembly stops prior to outer membrane penetration and hook elongation outside the cell (26, 47). The FliK ruler does not flip the FlhB secretion specificity switch, and FlgM remains in the cytoplasm and inhibits ²⁸-dependent transcription. Selection for ²⁸-dependent transcription in the PL-ring mutant background has yielded flk-null mutants and filamentous rod mutants (flgG*) that result in the secretion of FlgM into the periplasm, where it is degraded (6, 23). We expected that selection for FlgM secretion should yield a higher frequency of flk loss-of-function (null) mutants than of the altered-function Flk bypass mutant class. To circumvent this, we constructed a second copy of the flk⁺ gene transcribed from an arabinose-inducible promoter. In the presence of arabinose, the second flk⁺ gene would presumably complement recessive flk mutations.

In this study, Ap^r colonies were selected for that secreted an FlgM-Bla fusion into the periplasm. Two hundred fifty-two Ap^r colonies were picked from 86 independent cultures on LB-Ap (100 µg/ml) agar plates. In the presence of 1% arabinose, 121 isolates became Ap^s and were presumed to have recessive mutations in flk and were discarded. The remaining 131 Ap^r mutants were screened for linkage to the chromosomal flk gene. Mutations in these 131 strains were mapped by P22 phage transduction, followed by DNA sequencing, and 15 were linked to pyrC3052::FCF (where FCF is FRT-chloramphenicol-FRT) and 31 were linked to pdxB651::Tn10dTc (where Tc is tetracycline). These were presumed to be dominant alleles since they conferred an Flk^– phenotype in the presence of wild-type Flk protein, defining a new class of flk mutants that had not been isolated before. DNA sequence analysis of some of the dominant flk alleles identified single amino acid substitutions in the N terminus of Flk: H2P, L61P, and A77D. Flk is a 333-amino-acid protein that is predominantly in the cytoplasm with a C-terminal membrane anchor (1). Fine mapping and DNA sequence analysis revealed that the remaining 100 Ap^r mutants were altered in either the flgG or the flhE gene. The flgG mutants included previously isolated flgG* alleles shown to produce filamentous rod structures (6) and included previously isolated single-amino-acid substitution mutations in FlgG: G53C, G65E, G65R, G65V, D117Y, G132R, G183W, and S197L. Surprisingly, 85 Ap^r mutants (with linkage to flhE7233::FCF) resulted from null mutations in the flhE gene (Table 2). Null mutations in flhE were reported to be defective in swarming (53), but the reason for the swarming defect was not clear. These results suggest that FlhE may play some role in the secretion process. When moved into an otherwise Fla⁺ background, the flhE-null alleles did not allow FlgM-Bla secretion. Since FlhE is probably a periplasmic protein, it is not clear how it can affect the secretion specificity switch at the cytoplasmic domain of FlhB and only in the PL-ring mutant background. One possibility is that FlhE can interact with periplasmic-facing regions of FlhB to affect the secretion specificity switch.

View this table: [in this window] [in a new window]

TABLE 2. Results of DNA sequence analysis of isolated flhE mutants

Loss of Flk results in FlgM-Bla secretion into the periplasm in an Fla⁺ strain. The Ap^r selection for secretion of FlgM-Bla into the periplasm was repeated in strains with no mutations in flagellar assembly (Fla⁺). Prior to this experiment, we discovered that the FlgM-Bla strain that is otherwise Fla⁺ exhibits a leaky resistance to Ap on LB medium but a tight Ap^s phenotype on Mac medium, suggesting that FlgM-Bla confers Mac-Ap^r only when it is secreted into the periplasm. For this reason, all further selections for FlgM-Bla secretion into the periplasm were done with Mac-Ap medium.

Initially, 10 independent cultures of a Fla⁺ strain expressing chromosomal flgM-bla (TH10239) were plated on Mac-Ap plates and incubated overnight at 37°C, and one Mac-Ap^r colony from each plate was characterized by genetic mapping. Three of the mutations mapped to the flk gene, and seven mapped to the flg region. Fine mapping and DNA sequence analysis revealed that the flg-linked mutants were mutated in the flgG gene and included flgG* alleles resulting from single-amino-acid substitutions that were previously isolated in filamentous rod mutants: G65E(x2), G65V(x2), G133V, E189K, and S197L (6). To determine if Mac-Ap^r results from loss-of-function mutations in flk, null alleles including Tn10dTc and Tn10dCm insertions in flk and a deletion of flk replaced with a chloramphenicol resistance cassette (flk::Cm) were introduced by transduction and the resulting transductants were found to be Mac-Ap^r. This result demonstrates that loss of Flk, even in Fla⁺ strains, will cause enough premature secretion specificity switching to allow FlgM-Bla secretion into the periplasm. The three flk dominant-negative alleles isolated as described above produced a phenotype identical to that produced by an allele with the entire flk coding sequence deleted (flk-7755) when compared for FlgM-Bla secretion and effects on motility in Fla⁺ and PL-ring mutant backgrounds. All four alleles allowed FlgM-Bla secretion in either a flgHI958 or a Fla⁺ background, and none restored motility in the flgHI958 background. The mechanism by which Flk loss-of-function mutants could allow both premature FlgM-Bla secretion and the formation of normal flagella in an Fla⁺ background remains to be elucidated.

C-terminal FlhA truncations result in FlgM-Bla secretion into the periplasm. The selection for FlgM-Bla secretion into the periplasm was repeated with a strain isogenic to the Fla⁺ background used earlier, except that it carried a lac transcriptional reporter fusion to the ²⁸-dependent motA gene (TH10505 = TH10239 motA::MudJ). The selection was performed at 30°C and was followed by screening for temperature-sensitive (ts) alleles. Such a selection and screening can yield unusual rare mutant classes (5). Ten independent cultures were plated on Mac-Ap plates and incubated overnight at 30°C. The Mac-Ap^r colonies were replica printed to two Mac-Ap plates; one plate was incubated at 30°C, and the other was incubated at 42°C. While mutants that secrete FlgM-Bla into the periplasm are expected to express the motA::MudJ reporter fusion (Lac⁺), mutants that show only a slight increase in FlgM-Bla secretion might be Ap^r but remain Lac^– if enough FlgM-Bla is retained to inhibit ²⁸-dependent transcription. The majority of colonies were Ap^r and Lac⁺ at both 30 and 42°C, but three plates yielded mutants that were Ap^r Lac^– at 30°C and Ap^s at 42°C. Additionally, colonies with different Lac phenotypes (Lac⁺, Lac^+/–, and Lac^–) at 30 and 42°C were picked as a means to isolate different types of mutants from individual experiments. The ts mutants were screened for linkage to the flk and flg regions of the chromosome. Mac-Ap^r colonies that were linked to the flk gene were discarded, and mutants linked to the flg region were sequenced for the flgG gene. DNA sequence analysis identified 11 single-amino-acid substitutions in the flgG gene and a deletion of three residues that conferred an Ap^r Lac⁺ phenotype on Mac-lac-Ap plates. The FlgG deletion and substitutions included P52L, (59-61), S64P, G65E, G65V, L66P, G118V, G183R, G183W, E189K, N190K, and S197L. The majority of these alleles had already been shown to result in filamentous rods (6). The three alleles that were Ap^r Lac^– at 30°C and Ap^s at 42°C were not linked to flk or the flg region and mapped to the flhBAE operon. DNA sequence analysis revealed that all three mutants resulted from C-terminal truncation mutations in flhA. The three FlhA truncation mutations were Q588-Stop and two independent Q589-Stop mutations. The fact that the flhA mutants exhibited a Mac-Ap^r phenotype at 30°C suggested that secretion of FlgM-Bla into the periplasm did occur. The Lac^– phenotype suggests a general HBB assembly defect resulting in FlgM-Bla accumulation in the cytoplasm, which inhibited motA-lac transcription. The flhA7299(Q588-Stop) and flhA7300(I600-A644) alleles were moved into an fla⁺ strain, and the cells were nonmotile. These also allowed FlgM-Bla secretion when moved into a flgHI958 (PL-ring mutant) strain. It is known that an flhA-null allele mutant accumulates FlgM since it is defective in secretion; however, the secretion of FlgM-Bla into the periplasm demonstrates that mutants with FlhA truncations retain some flagellar secretion capability and allow early flipping of the secretion specificity switch. Finally, mutants with two deletions that fused the FlgA signal sequence for Sec-dependent secretion to the FlgM-Bla fusion were also isolated as Mac-Ap^r and exhibited strong Lac⁺ phenotypes at both 30 and 42°C.

The secretion specificity switch does not require either the hook or the hook cap protein. During flagellar assembly, the secretion specificity switch is believed to occur when the C terminus of secreted FliK interacts with the FlhB component of the flagellar T3S system (44). This interaction occurs when the hook reaches its terminal length of 55 nm but can occur at shorter hook lengths by overexpression of FliK and at longer lengths by hook protein (FlgE) overexpression (44, 45). The N terminus of FliK has been shown to interact strongly with the hook cap protein FlgD and weakly with FlgE (44). These results led to the model in which the interaction of FliK with assembled FlgD results in a pause in FliK secretion. This pause would allow the FliK C terminus, if it is in the vicinity of FlhB, to interact with FlhB and premature secretion specificity switch flipping to occur. As a result, the final assembled rod-hook length would correspond to the length of the extended unfolded conformation of the N-terminal part of FliK.

Our results have shown that the timing of the secretion specificity switch during flagellar assembly, which allows secretion of FlgM-Bla into the periplasm, can be achieved by loss of Flk, loss of FlhE, C-terminal truncations in FlhA, and alterations in the FlgG distal rod protein that produce filamentous rod structures. Furthermore, a filamentous rod mutation, flgG*, produces a rod corresponding in length to a rod-hook in a normal HBB when FliK is intact (6). Loss of FliK in the flgG* background produces rods of uncontrolled length. If a pause in FliK secretion, via interaction between FliK_N and assembled FlgD/FlgE, is required to switch secretion substrate specificity in normal HBB assembly, then it seems likely that there is also a pausing mechanism in the FliK-dependent length control of the filamentous rod structures. According to this model, FliK-dependent switching to allow FlgM or FlgM-Bla secretion into the periplasm in flgG* mutants would require some hook or hook cap protein to be present at the tip of the filamentous rods. Alternatively, FliK could interact with the assembled rod cap (FlgJ), or perhaps no interaction is required at all. It may be that just getting the N terminus of FliK through the "hole" at the tip of the rod or hook for secretion to occur results in the required pause. Thus, we decided to test the requirements for FliK, FlgD, FlgE, and FlgJ in the mutant strains that allow premature secretion of FlgM-Bla into the periplasm in the PL-ring mutant background.

Secretion of FlgM-Bla into the periplasm in a PL-ring^– background was tested in strains that also carried either a flgD point mutation allele [flgD157(L40P)] or a nonpolar deletion allele of the flgD, flgE, flgJ, or fliK gene. These strains were screened for growth on Mac-Ap plates (Fig. 1) and for ²⁸-dependent transcription as determined by a Lac⁺ phenotype on tetrazolium-lactose (TTC-lac) indicator plates without added Ap (Table 3) Surprisingly, an flgD- or flgE-null allele did not abolish the Mac-Ap^r phenotype in the flgG* mutant strains, while loss of the rod cap protein FlgJ resulted in Ap^s. This suggests that FlgM-Bla is not secreted in the absence of the rod structure but is secreted in the absence of hook (FlgE) or hook cap (FlgD) protein. Deletion of fliK also resulted in a Mac-Ap^s phenotype, confirming the requirement for FliK to flip the secretion specificity switch. The Ap^s phenotypes also corresponded to Lac^– phenotypes, suggesting that FlgM-Bla was retained in the cytoplasm to inhibit ²⁸-dependent transcription. Mutants with either a recessive (flk) or a dominant (L61P) flk allele were Lac^– on TTC-lac indicator plates if combined with a flgD-, flgE-, flgJ-, or fliK-null allele. This suggests a reduction in FlgM-Bla secretion, but not enough to be Ap^s except in the fliK background. These results were confirmed by assaying the MIC of Ap for these strains (Table 4). The parent strain, TH10406, showed a MIC of 62.5 µg/ml, and this value increased to 250 µg/ml when the flk(L61P) mutation was introduced. When flk(L61P) was combined with the point mutation flgD157 or an flgD-, flgE-, or flgJ-null mutation, the MICs were decreased to 125, 125, 150, and 100 µg/ml, respectively. Introduction of fliK brought it back to the parent strain MIC of 62.5 µg/ml. The two flhE alleles, one which has a mutation in the predicted signal sequence (A16V) and the other which has a point mutation after the signal sequence (G110R), were also analyzed. Both flhE mutants required all four of the flagellar components examined (FlgD, FlgE, FlgJ, and FliK) to show both the Ap^r and Lac⁺ phenotypes.

View larger version (72K): [in this window] [in a new window]

FIG. 1. Effects of the rod cap (FlgJ), hook (FlgE), hook cap (FlgD), and ruler (FliK) proteins on the secretion of FlgM-Bla into the periplasm in the different flk bypass mutant classes. Various mutations in flhE, flk, and flgG (flgG* filamentous rod alleles) which allowed FlgM-Bla secretion into the periplasm in either the Fla+ or the PL-ring (flgHI) mutant background were tested alone or in combination with one of the following alleles: flgD157(L40P) (a secretion-competent, polymerization-defective flgD allele [44]), flgD (nonpolar), flgE (nonpolar), flgJ (nonpolar), or fliK. The cells were streaked onto Mac-Ap plates and incubated overnight at 37°C. The flk bypass mutant strains used, with mutations in flgD, flgE, flgJ, or fliK, were TH10406 = flk+ flgG+ flhE+, TH13139 = TH10406 [flgD(L40P)], TH13140 = TH10406 (flgD), TH13188 = TH10406 (flgE), TH12569 = TH10406 (flgJ), and TH13628 = TH10406 (fliK); TH12542 = flk-7420, TH13194 = TH12542 (flgD[L40P]), TH13195 = TH12542 (flgD), TH13199 = TH12542 (flgE), TH12577 = TH12542 (flgJ), and TH13631 = TH12542 (fliK); TH12541 = flk-7419(L61P), TH13148 = TH12541 [flgD(L40P)], TH13149 = TH12541 (flgD), TH13193 = TH12541 (flgE), TH12576 = TH12541 (flgJ), and TH13633 = TH12541 (fliK); TH10545 = flgG6705(G65R), TH1764 = TH10545 (flgD[L40P]), TH13135 = TH10545 (flgD), TH13189 = TH10545 (flgE), TH12570 = TH10545 (flgJ), and TH13151 = TH10545 (fliK); TH10547 = flgG6707(S197L), TH12765 = TH10547 [flgD(L40P)], TH13137 = TH10547 (flgD), TH13190 = TH10547 (flgE), TH12571 = TH10547 (flgJ), and TH13152 = TH10547 (fliK); TH12538 = flhE7416(A16V), TH13142 = TH12538 [flgD(L40P)], TH13143 = TH12538 (flgD), TH13191 = TH12538 (flgE), TH12573 = TH12538 (flgJ), and TH13629 = TH12538 (fliK); and TH12539 = flhE7417(G110R), TH13145 = TH12539 [flgD(L40P)], TH13146 = TH12539 (flgD), TH13192 = TH12539 (flgE), TH12574 = TH12539 (flgJ), and TH13630 = TH12539 (fliK).

View this table: [in this window] [in a new window]

TABLE 3. Effects of flk, flgG*, and flhE alleles on 28-dependent transcription in various HBB mutant backgrounds using an fliC-lac transcriptional fusion reporter and TTC-lac indicator platesa

View this table: [in this window] [in a new window]

TABLE 4. Ap MIC assay for Flk(L61P) in various mutant backgrounds

Flagellar motility in flgG* mutants. It has been reported that mutations in either flk or flhE have no effects on swimming motility (36, 53), suggesting that these mutants still can assemble flagellar structures with a hook of the proper length. In fact, we found that substrate specificity switching without the PL rings in either the flk-null or the flhE-null background still required the structural components of the rod and hook and the FliK ruler. Since the HBB structure cannot be fully assembled in the periplasmic space of the PL-ring^– strain, the FliK ruler seems to be able to tolerate a shorter rod length when making the substrate specificity switch. It is important to note that HBB assembly is completed 15 min after transcription of the HBB genes is initiated (24). Thus, FliK normally flips the switch within a narrow window of time. However, in assays for switching using FlgM-Bla secretion, colonies are scored as Ap^r or Lac⁺ after overnight incubation. FliK-dependent switching in the flk- or flhE-null background need not be as efficient as that which normally occurs in HBB assembly to give an Ap^r or Lac⁺ phenotype. Taken together, the results show that substrate specificity switching and termination of hook assembly in the flk and flhE mutants need not be coupled. The effects of the flgG* filamentous rod mutations on flagellar motility were determined in an otherwise wild-type background (Fig. 2A) and in combination with an flgD-, flgE-, or flgJ-null mutation (Fig. 2B). The flgG* mutants have reduced motility compared to that of the wild-type strain (Fig. 2A), indicating the presence of functional flagella. When combined with the flgD, flgE, or flgJ allele, the cells exhibited a slight degree of motility when flgG filamentous rod mutations were combined with an flgE deletion compared to the flgD and flgJ mutant strains (Fig. 2B). Although it is reproducible, it is not clear if this slight difference is significant.

View larger version (39K): [in this window] [in a new window]

FIG. 2. Effects of the rod cap (FlgJ), hook (FlgE), and hook cap (FlgD) proteins on the motility of flgG* mutants. (A) The motility of flgG* mutants in a wild-type background was tested by using 0.3% agar motility plates. The plates were incubated at 37°C for 7 h. (B) The effect of a point mutation of flgD157(L40P) (hook cap) or of a nonpolar deletions of flgD, flgE (hook), or flgJ (rod cap) on motility was tested in 0.3% agar motility plates. Strains were poked into the plate and incubated at 37°C for 24 h. The plates were incubated at 37°C for 7 h. The strains used were the wild type (TH437) and G65R (TH10354), D117Y (TH10929), G132R (TH10355), G183W (TH10930), E189K (TH11250), and S197L (TH10356) mutants. (B) The effects of the flgD157(L40P) (hook cap) point mutation or a nonpolar deletion in flgD, flgE (hook), or flgJ (rod cap) on motility was tested on 0.3% agar motility plates. Strains were poked onto the plate and incubated a 37°C for 24 h. The strains used were flgG+ TH13469 = flgD157, TH9932 = flgD6540, TH9933 = flgD6541, TH13470 = flgE, and TH13471 = flgJ; TH13453 = flgG*6705(G65R) flgD157, TH13454 = flgG*6705 flgD6540, TH13455 = flgG*6705 flgD6541, TH13456 = flgG*6705 flgE, and TH13457 = flgG*6705flgJ; and TH13458 = flgG*6707(S197L) flgD157, TH13459 = flgG*6707 flgD6540, TH134560 = flgG*6707 flgD6541, TH13461 = flgG*6707 flgE, and TH13462 = flgG*6707 flgJ.

Rod-filament structures form in the absence of hook protein. The secretion specificity switch occurs in flgG* mutants in the absence of the hook (FlgE) and hook cap (FlgD) proteins. In addition, the flgG* mutants exhibited poor motility in the absence of hook (FlgE) protein. We examined these strains by fluorescence microscopy and TEM to determine if filaments could actually form on the filamentous rod structures without an intervening hook structure. The flgG* mutants, which produce the filamentous rod structures, grow flagella in the periplasmic space (6). However, upon examination by fluorescence microscopy using FITC-conjugated anti-FliC antibodies to visualize the flagella, about 20% of the bacteria had a single external filament structure that was, on average, shorter than wild-type filaments (Fig. 3A). These external filaments were also observed on the flgG* cells lacking the hook (FlgE) or hook cap (FlgD) protein but not in the flgG* strain lacking FlgJ. [The flgG(G65R) flgJ mutant strain in Fig. 3A was overstained in an aggressive attempt to find flagella, and some background stain is visible in that panel.] These results strongly suggest that filaments can form on rod structures without intervening hook protein.

View larger version (48K): [in this window] [in a new window]

FIG. 3. Effects of flgD, flgE, and flgJ deletions on the flagellation phenotype of the flgG* mutant. (A) The effect of a deletion mutation of flgD (hook cap) (TH13454), flgE (hook) (TH13456), or flgJ (rod cap) (TH13457) on flagellum formation in combination with the flgG*6705(G65R) allele was tested by immunofluorescence microscopy compared to that of flgG*6705 parent strain TH10354. The flagellar filaments (green) were first probed with an anti-FliC antibody and then with an anti-rabbit secondary antibody conjugated with a fluorescent dye, FITC. The cellular membrane (red) and genomic DNA (blue) were stained with FM4-64 and DAPI, respectively. (B) The flagellar structures were isolated from flgG* strains combined with the deletion flgD (TH13690), flgE (TH13692), or flgDE (TH13723). The isolated flagella were stained with 2% PTA and observed by TEM.

To confirm that filaments form directly on filamentous rods, rod-filament structures were isolated and examined by TEM. The purified flagellar structure of the flgG* strain with the flgD, flgE, or flgDE deletion was stained with 1% PTA. By TEM, flagellar filaments were assembled directly onto the filamentous rod structure perpendicular to the top surface of the MS ring (Fig. 3B). Filament assembly onto the filamentous rod required FlgJ, FlgK, and FlgL (data not shown) but neither FlgD nor FlgE.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
This study was designed to isolate Flk bypass mutants as a means to elucidate the mechanism by which the loss of Flk allowed the flagellar T3S system to change secretion substrate specificity prior to HBB completion. The goal was to understand how wild-type Flk protein prevents premature switching. Flk was first identified in the original isolation of flgM-null mutants (14). It had been known for decades that flagellar mutants that are defective in HBB assembly also fail to transcribe ²⁸-dependent class 3 promoters. Later, it was shown that the FlgM protein is an anti-²⁸ factor that sequesters ²⁸ (48), and will even strip ²⁸ from core RNA polymerase (3), to prevent flagellar class 3 promoter transcription, which is dependent on the ²⁸ RNA polymerase holoenzyme. It was serendipitous that the original selection resulting in the isolation of flgM-null mutants was done by selecting for ²⁸-dependent transcription in an flgI mutant strain defective in PL-ring formation. Transposon mutagenesis yielded insertions in the flgM and flk genes that allowed a flagellar class 3 promoter fused to the lac operon to be transcribed in the flgI mutant background (14). It was found that loss of FlgM allowed ²⁸-dependent transcription in all mutants defective in HBB formation, whereas loss of Flk only allowed wild-type levels of ²⁸-dependent transcription in mutants defective in PL-ring formation: flgA, flgH, and flgI mutants. There was also some increase in ²⁸-dependent transcription in the absence of Flk in HBB assembly mutants that were defective in rod-hook assembly, but not in mutants defective in early HBB assembly, including those defective in MS- and C-ring assembly and in genes of the flagellar T3S system (23).

Later, it was discovered that loss of Flk in mutants defective in PL-ring formation resulted in the flagellar secretion specificity switch, which normally occurs after hook completion, and resulted in secretion of FlgM into the periplasm, where it was degraded (1). The net result was that FlgM was removed from the cytoplasm and ²⁸ was free to transcribe flagellar class 3 genes. Given that loss of Flk showed some increase in class 3 transcription in rod-hook assembly mutants, but not in mutants leading up to rod assembly, we surmise that some low-level secretion specificity switching occurs in rod-hook assembly mutants in the absence of Flk, resulting in FlgM secretion into the periplasm. Any mechanism to explain the effect of Flk on the flagellar secretion specificity switch must also account for this phenotype.

An assay was developed that allowed easy detection of FlgM secretion into the periplasm (1). When wild-type FlgM is secreted into the periplasm, it is rapidly degraded and not detected. However, it was discovered that fusion of β-lactamase to the C terminus of FlgM prevented the degradation of periplasmic FlgM. Furthermore, secretion of FlgM-Bla conferred Ap^r on the cell and thus provided a positive selection for FlgM secretion into the periplasm. During the course of these studies, it was observed that the Ap^s phenotype of flagellar mutants that were unable to secrete FlgM-Bla into the periplasm was leaky on L medium. They exhibited some growth on L-Ap plates, while the parent strain lacking the Bla fusion did not. However, it was then discovered that on a Mac-based medium, mutants unable to secrete FlgM-Bla into the periplasm did not show a leaky Ap^s phenotype. Only mutants able to secrete FlgM-Bla into the periplasm were able to grow on Mac-Ap plates, providing a strong and reliable selection for FlgM secretion into the periplasm.

We sought to use selection of FlgM-Bla into the periplasm as a means to isolate mutants, similar in phenotype to those with flk-null alleles, with a premature switch in the flagellar secretion specificity switch. The same strategy was employed to discover FlgM bypass mutants with changes in the ²⁸ structural gene fliA (2). The ²⁸ mutants unable to bind FlgM (FlgM bypass) have the same phenotype as flgM-null mutants and were used to characterize the mechanism by which FlgM inhibits flagellar class 3 transcription in HBB assembly mutants. For the isolation of Flk bypass mutants, selection for Mac-Ap^r using the FlgM-Bla reporter requires that FlgM-Bla be secreted into the periplasm and thus the selection would be specific to the secretion specificity switch and not yield mutants affecting flagellar gene transcription.

Selection for FlgM-Bla secretion in either Fla⁺ or PL-ring mutant strains yielded both new and expected results. Distal rod gene mutants that produced filamentous rods, flgG* mutants, were obtained. These mutants were isolated in a previous attempt to isolate flk bypass mutants. Such mutants would be predicted to have an Flk-null phenotype (flagellar class 3 transcription in the PL-ring mutant strains) and possibly result from alterations in a protein that interacts with Flk. The previous selection demanded flagellar class 3 transcription in a PL-ring mutant strain that was unlinked to the flk, flgM, and fliA loci (6). The flgG* mutants answered this selection because they allowed the rod to continue growing to the length of the rod-hook structure, at which point the FliK-dependent secretion specificity switch occurred and FlgM was secreted into the periplasm. Remarkably, when the flgG* alleles were moved into an Fla⁺ strain, they produced flagella that grew in the periplasmic space, indicating that rod termination was coupled to outer membrane penetration and growth of the flagellum outside the cell (6). In this study, we further characterized the flgG* alleles by asking what the requirements are for the secretion specificity switch to occur in flgG* mutant strains. As expected, the rod cap structural gene, flgJ, was required, since it is needed for rod polymerization. The fliK gene was also required, since it catalyzes the secretion specificity switch. However, we were surprised to find that the flgD (hook cap) and flgE (hook) genes were not required. Furthermore, we did not expect to find that filaments would grow directly from the filamentous rod structures. However, given the high degree of conservation between the distal rod protein, FlgG, and the hook protein, FlgE, it could have been expected that the hook-filament junction proteins (FlgK and FlgL) would polymerize at the end of the rod once the secretion specificity switch occurred to allow secretion of these flagellar late secretion substrates.

We were not surprised to find some rod-filament structures present on the surface of cells in strains lacking the PL-ring outer membrane pore. It was shown that overexpression of hook protein could suppress the loss of the PL-ring pore (47). This suggests that the pore facilitates outer membrane penetration but is not essential for that process.

Another unexpected finding was that flhE-null mutants allowed the secretion specificity switch to occur in the PL-ring-defective background but not in the Fla⁺ strains. flhE-null alleles were reported to produce a swarming defect on agar surfaces, but by an unknown mechanism. The flhE gene is one of the last flagellar assembly genes whose function remains to be determined. Its presence in an operon with flhA and flhB implies a possible role in the secretion process. However, the known components of the flagellar secretion apparatus are either membrane-embedded (FlhA, FlhB, FliO, FliP, FliQ, and FliR) or cytoplasmic (FliH, FliI, and FliJ) proteins, whereas FlhE is a periplasmic protein and, like those encoded by the genes required for PL-ring assembly (flgA, flgH, and flgI), is secreted by the Sec secretion system. The mechanism that allows the secretion specificity switch in PL-ring mutant strains that are also missing FlhE remains to be elucidated.

Finally, we come to FlhA. It seems likely to us that the flhA mutations could represent an affinity site within the C ring for recruitment of Flk. Thus, the C-terminal truncation mutations in flhA produce the same phenotype as flk-null alleles. The flhE mutations do not produce the same phenotype as flk-null alleles because they allow FlgM-Bla secretion only in the absence of the PL ring, whereas either flk or flhA mutations allow periplasmic secretion of FlgM-Bla in Fla⁺ strains.

There are two requirements for the flagellar T3S system to change specificity from rod-hook substrates to late-secretion substrates: a cleavage of the C-terminal cytoplasmic domain of FlhB and interaction of FlhB with the C terminus of FliK (FliK_C). The FlhB cleavage event has been shown to occur spontaneously with a 5-min half-life (autocleavage) (12), although FliK interaction may catalyze the cleavage as well. The hook grows to a length of 55 ± 6 nm (17). If FliK interaction did not catalyze FlhB autocleavage, then the distribution of hook lengths would depend only on FlhB autocleavage and longer hook lengths would be observed, which is not the case. FliK-dependent hook length control is completely dependent on FliK secretion through the filament. It had been reported that FliK could measure hook length independently of secretion as an internal ruler because FliK mutants that fail to be secreted showed hook length control (52). However, we have repeated these experiments and found that the FliK variants previously reported to control hook length independently of secretion are, in fact, secreted (T. Hirano and K. T. Hughes, unpublished data), ruling out the internal-ruler model. FliK was demonstrated to flip the secretion specificity switch independently of FliK secretion. However, the nonsecreted form of FliK had to be expressed from the trc promoter from a high-copy-number plasmid vector and hook length control was lost (16).

Since a nonsecreted form of FliK must be expressed to very high levels to access FlhB, there must exist a mechanism to prevent the interaction of cytoplasmic FliK with the FlhB component of the basal body for the FliK secretion-dependent hook length control mechanism to work. We propose that the Flk and FlhA proteins provide a barrier within the cytoplasmic face of the basal C-ring structure to prevent cytoplasmic FliK from interacting with FlhB (Fig. 4). This barrier could be in the form of a corking device that blocks cytoplasmic FliK_C from access to FlhB. The secretion of FliK would be required to get the FliK_C past the Flk-FlhA cork and allow interaction with FlhB. Alternatively, Flk-FlhA could interact directly with FliK_C to prevent interaction with FlhB. Furthermore, we suggest that a pause in FliK secretion is required for a productive interaction between FliK_C and FlhB_CC. It is known that FliK is continuously secreted through the basal body of flgE (hook)-null mutants, and therefore past FlhB, but never flips the secretion specificity switch, suggesting that some pause in secretion is required to allow FlhB-FliK interaction to occur. Loss of Flk or C-terminal truncations in flhA would effectively remove the cork or no longer bind FliK_C and allow cytoplasmic FliK_C access to FlhB. Since FlhA is also part of the flagellar T3S system, there would be a limit to how much of the C terminus could be removed to allow FliK access without inhibiting the secretion process.

View larger version (41K): [in this window] [in a new window]

FIG. 4. The Flk-FlhA cork model. (A) During the secretion of FliK through the flagellar basal structure, FliK makes a pause as its N terminus reaches the tip of the growing rod or rod-hook. At this point, Flk (Fluke) and FlhA prevent the C terminus of FliK (FliKC) from interacting with FlhB to flip the flagellar T3S specificity switch prior to hook completion. The interaction of cytoplasmic FliKC with FlhB is prevented by either a physical blockage depicted with Flk sequestered within the basal body (A, top) or an interaction of membrane-anchored Flk in the vicinity of the basal body with FliKC within the cytoplasm (A, bottom). (B) When either the hook (B, top) or the filamentous rod (B, bottom) has reached its terminal length, secretion of FliK through the basal structure and a resulting pause as the N terminus of FliK reaches the tip of the hook occur and FliKC is in close proximity to the FlhB component of the flagellar T3S. Interaction of FliKC with the cleaved C-terminal domain of FlhB (FlhBCC) catalyzes the secretion specificity switch from rod-hook substrate specificity to late substrate or filament-type specificity. (C) In the absence of Flk, when FliK makes a pause during secretion, FliKC can interact with FlhBCC and catalyze the secretion specificity switch prior to HBB completion.

Loss of Flk allows a limited amount of FlgM secretion in rod-hook assembly mutants and wild-type levels of FlgM secretion in PL-ring mutants. This suggests that FliK_C has much greater access to FlhB in the PL-ring mutants than in the rod-hook mutant strains. The N terminus of FliK (FliK_N) interacts with the hook (FlgE) and hook cap (FlgD) proteins (44). This interaction may facilitate FliK secretion in the PL-ring mutants if the hook cap and some hooks are assembled. A working model that illustrates the Flk-FlhA corking mechanism is presented in Fig. 4. Although there are no published data to demonstrate the interaction of Flk with any flagellar protein, the model predicts that Flk probably interacts with FlhA_C, FlhB_C, FliK, or a C-ring component (FliG, FliM, or FliN). The model predicts that Flk is incorporated into the basal structure. The flk gene is not expressed with the flagellar regulon, and the Flk protein is present throughout the cytoplasmic membrane (1, 23). Thus, Flk likely plays another role in cellular metabolism that is independent of its role in the flagellar secretion substrate specificity switch. Our working model predicts that either some Flk protein anchored in the cytoplasmic membrane is sequestered into the flagellar basal structure to prevent interaction of FliK_C with FlhB_CC (Fig. 4A, top), or Flk, which is present throughout the cytoplasmic membrane, is able to interact with the C terminus of FliK to prevent interaction with FlhB_CC without being incorporated within the basal body (Fig. 4A, bottom). FliK_N must be secreted far enough to pass the Flk block, in either a normal HBB (Fig. 4B, top) or a filamentous rod mutant structure (Fig. 4B, bottom), to flip the secretion switch at FlhB. A flk-encoded mutant protein with its C-terminal membrane anchor deleted does not complement an flk-null mutant protein to prevent the secretion specificity switch in a PL-ring mutant strain (1), suggesting that the interaction with the flagellar basal structure, if any, is weak. Regardless, secretion of FliK would still be required to separate FliK_C from Flk and allow interaction with FlhB_CC. These are working models and, in the absence of direct binding data, remain speculative. We realize that any refined models to explain the role of Flk in flagellar T3S will require further studies of protein-protein interactions through biochemical analyses to be better substantiated.

Motile revertants of fliK-null strains result from mutations in flhB that switch without FliK (17, 30). The FliK bypass flhB (flhB^FliK-BP) mutants can make the secretion specificity switch in the absence of FliK, but the flagella have hooks of uncontrolled lengths and are only poorly motile. Later, it was found that the flhB^FliK-BP mutant would not secrete FlgM in a fliK⁺ background if the hook was absent (29). flk loss-of-function mutants (termed rflH mutants) showed FlgM secretion from the cell in the flhB^FliK-BP flgE (fliK⁺) background (29). These were isolated by selecting for Lac⁺ in a flhB^FliK-BP flgE (fliK⁺) background with a ²⁸-dependent motA-lac reporter fusion. The flhB^FliK-BP mutant switches poorly without FliK and switches normally in the presence of FliK. Thus, in the flhB^FliK-BP flgE (fliK⁺) background, the switch to late secretion is not efficient and FlgM accumulates to inhibit ²⁸ because in the absence of a hook, the central channel is not long enough for FliK_C to be within the vicinity of FlhB during secretion. Only when Flk is also absent can FliK_C in the process of secretion interact with FlhB to flip the switch (Fig. 4C).

Continuous secretion of FliK in mutants lacking hooks fails to flip the secretion specificity switch. If a pause in FliK secretion is required to allow a productive interaction between FliK_C and FlhB_CC, our data suggest that this pause is specific to interactions between FliK_N and FlgD/FlgE. Otherwise, we would not have expected the secretion switch to occur in the absence of FlgD/FlgE in the flgG* filamentous rod mutant strains. Thus, it possible that FliK_N can also interact with rod and rod cap proteins (FlgG/FlgJ), which has never been examined. Thus, it is likely that the pause occurs when the very N terminus of FliK reaches the tip of the growing structure, whether capped by FlgJ or by FliD. Furthermore, we also propose that the interaction between FliK_N and FlgD/FlgE would create an affinity site and facilitate FliK secretion during hook growth, which is the time during flagellum assembly when FliK secretion is needed.

 
We acknowledge the Hughes laboratory members for critical reading of the manuscript and for helpful discussions. We thank Noriko Takahashi for her technical help in the preparation of flagellar basal structures.

This work was supported by Public Health Service grant GM056141 from the National Institutes of Health to K.T.H.

 
* Corresponding author. Mailing address: Department of Biology, University of Utah, Salt Lake City, UT 84112. Phone: (801) 581-6517. Fax: (801) 581-4668. E-mail: hughes{at}biology.utah.edu

Published ahead of print on 17 April 2009.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Aldridge, P., J. E. Karlinsey, E. Becker, F. F. Chevance, and K. T. Hughes. 2006. Flk prevents premature secretion of the anti-sigma factor FlgM into the periplasm. Mol. Microbiol. 60:630-643.[CrossRef][Medline]
2
2. Chadsey, M. S., and K. T. Hughes. 2001. A multipartite interaction between Salmonella transcription factor ²⁸ and its anti-sigma factor FlgM: implications for ²⁸ holoenzyme destabilization through stepwise binding. J. Mol. Biol. 306:915-929.[CrossRef][Medline]
3
3. Chadsey, M. S., J. E. Karlinsey, and K. T. Hughes. 1998. The flagellar anti-sigma factor FlgM actively dissociates Salmonella typhimurium ²⁸ RNA polymerase holoenzyme. Genes Dev. 12:3123-3136.[Abstract/Free Full Text]
4
4. Chevance, F. F., and K. T. Hughes. 2008. Coordinating assembly of a bacterial macromolecular machine. Nat. Rev. Microbiol. 6:455-465.[CrossRef][Medline]
5
5. Chevance, F. F., J. E. Karlinsey, C. E. Wozniak, and K. T. Hughes. 2006. A little gene with big effects: a serT mutant is defective in flgM gene translation. J. Bacteriol. 188:297-304.[Abstract/Free Full Text]
6
6. Chevance, F. F., N. Takahashi, J. E. Karlinsey, J. Gnerer, T. Hirano, R. Samudrala, S. Aizawa, and K. T. Hughes. 2007. The mechanism of outer membrane penetration by the eubacterial flagellum and implications for spirochete evolution. Genes Dev. 21:2326-2335.[Abstract/Free Full Text]
7
7. Chilcott, G. S., and K. T. Hughes. 2000. Coupling of flagellar gene expression to flagellar assembly in Salmonella enterica serovar Typhimurium and Escherichia coli. Microbiol. Mol. Biol. Rev. 64:694-708.[Abstract/Free Full Text]
8
8. Datsenko, K. A., and B. L. Wanner. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97:6640-6645.[Abstract/Free Full Text]
9
9. Davis, R. W., D. Botstein, and J. R. Roth. 1980. Advanced bacterial genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
10
10. Emerson, S. U., K. Tokuyasu, and M. I. Simon. 1970. Bacterial flagella: polarity of elongation. Science 169:190-192.[Abstract/Free Full Text]
11
11. Fan, F., K. Ohnishi, N. R. Francis, and R. M. Macnab. 1997. The FliP and FliR proteins of Salmonella typhimurium, putative components of the type III flagellar export apparatus, are located in the flagellar basal body. Mol. Microbiol. 26:1035-1046.[CrossRef][Medline]
12
12. Ferris, H. U., Y. Furukawa, T. Minamino, M. B. Kroetz, M. Kihara, K. Namba, and R. M. Macnab. 2005. FlhB regulates ordered export of flagellar components via autocleavage mechanism. J. Biol. Chem. 280:41236-41242.[Abstract/Free Full Text]
13
13. Fraser, G. M., T. Hirano, H. U. Ferris, L. L. Devgan, M. Kihara, and R. M. Macnab. 2003. Substrate specificity of type III flagellar protein export in Salmonella is controlled by subdomain interactions in FlhB. Mol. Microbiol. 48:1043-1057.[CrossRef][Medline]
14
14. Gillen, K. L., and K. T. Hughes. 1991. Negative regulatory loci coupling flagellin synthesis to flagellar assembly in Salmonella typhimurium. J. Bacteriol. 173:2301-2310.[Abstract/Free Full Text]
15
15. Hirano, T., T. Minamino, K. Namba, and R. M. Macnab. 2003. Substrate specificity classes and the recognition signal for Salmonella type III flagellar export. J. Bacteriol. 185:2485-2492.[Abstract/Free Full Text]
16
16. Hirano, T., S. Shibata, K. Ohnishi, T. Tani, and S. Aizawa. 2005. N-terminal signal region of FliK is dispensable for length control of the flagellar hook. Mol. Microbiol. 56:346-360.[CrossRef][Medline]
17
17. Hirano, T., S. Yamaguchi, K. Oosawa, and S. Aizawa. 1994. Roles of FliK and FlhB in determination of flagellar hook length in Salmonella typhimurium. J. Bacteriol. 176:5439-5449.[Abstract/Free Full Text]
18
18. Hughes, K. T., K. L. Gillen, M. J. Semon, and J. E. Karlinsey. 1993. Sensing structural intermediates in bacterial flagellar assembly by export of a negative regulator. Science 262:1277-1280.[Abstract/Free Full Text]
19
19. Iino, T. 1969. Polarity of flagellar growth in Salmonella. J. Gen. Microbiol. 56:227-239.[Abstract/Free Full Text]
20
20. Jones, C. J., M. Homma, and R. M. Macnab. 1989. L-, P-, and M-ring proteins of the flagellar basal body of Salmonella typhimurium: gene sequences and deduced protein sequences. J. Bacteriol. 171:3890-3900.[Abstract/Free Full Text]
21
21. Jones, C. J., and R. M. Macnab. 1990. Flagellar assembly in Salmonella typhimurium: analysis with temperature-sensitive mutants. J. Bacteriol. 172:1327-1339.[Abstract/Free Full Text]
22
22. Karlinsey, J. E. 2007. -Red genetic engineering in Salmonella enterica serovar Typhimurium. Methods Enzymol. 421:199-209.[Medline]
23
23. Karlinsey, J. E., A. J. Pease, M. E. Winkler, J. L. Bailey, and K. T. Hughes. 1997. The flk gene of Salmonella typhimurium couples flagellar P- and L-ring assembly to flagellar morphogenesis. J. Bacteriol. 179:2389-2400.[Abstract/Free Full Text]
24
24. Karlinsey, J. E., S. Tanaka, V. Bettenworth, S. Yamaguchi, W. Boos, S. I. Aizawa, and K. T. Hughes. 2000. Completion of the hook-basal body complex of the Salmonella typhimurium flagellum is coupled to FlgM secretion and fliC transcription. Mol. Microbiol. 37:1220-1231.[CrossRef][Medline]
25
25. Kojima, S., and D. F. Blair. 2004. The bacterial flagellar motor: structure and function of a complex molecular machine. Int. Rev. Cytol. 233:93-134.[Medline]
26
26. Kubori, T., N. Shimamoto, S. Yamaguchi, K. Namba, and S. Aizawa. 1992. Morphological pathway of flagellar assembly in Salmonella typhimurium. J. Mol. Biol. 226:433-446.[CrossRef][Medline]
27
27. Kubori, T., S. Yamaguchi, and S. Aizawa. 1997. Assembly of the switch complex onto the MS ring complex of Salmonella typhimurium does not require any other flagellar proteins. J. Bacteriol. 179:813-817.[Abstract/Free Full Text]
28
28. Kutsukake, K. 1994. Excretion of the anti-sigma factor through a flagellar substructure couples flagellar gene expression with flagellar assembly in Salmonella typhimurium. Mol. Gen. Genet. 243:605-612.[Medline]
29
29. Kutsukake, K. 1997. Hook length control of the export-switching machinery involves a double-locked gate in Salmonella typhimurium flagellar morphogenesis. J. Bacteriol. 179:1268-1273.[Abstract/Free Full Text]
30
30. Kutsukake, K., T. Minamino, and T. Yokoseki. 1994. Isolation and characterization of FliK-independent flagellation mutants from Salmonella typhimurium. J. Bacteriol. 176:7625-7629.[Abstract/Free Full Text]
31
31. Lee, H. J., and K. T. Hughes. 2006. Posttranscriptional control of the Salmonella enterica flagellar hook protein FlgE. J. Bacteriol. 188:3308-3316.[Abstract/Free Full Text]
32
32. Macnab, R. M. 2003. How bacteria assemble flagella. Annu. Rev. Microbiol. 57:77-100.[CrossRef][Medline]
33
33. Macnab, R. M. 2004. Type III flagellar protein export and flagellar assembly. Biochim. Biophys. Acta 1694:207-217.[Medline]
34
34. Minamino, T., H. U. Ferris, N. Moriya, M. Kihara, and K. Namba. 2006. Two parts of the T3S4 domain of the hook length control protein FliK are essential for the substrate specificity switching of the flagellar type III export apparatus. J. Mol. Biol. 362:1148-1158.[CrossRef][Medline]
35
35. Minamino, T., B. Gonzalez-Pedrajo, K. Yamaguchi, S. I. Aizawa, and R. M. Macnab. 1999. FliK, the protein responsible for flagellar hook length control in Salmonella, is exported during hook assembly. Mol. Microbiol. 34:295-304.[CrossRef][Medline]
36
36. Minamino, T., T. Iino, and K. Kutuskake. 1994. Molecular characterization of the Salmonella typhimurium flhB operon and its protein products. J. Bacteriol. 176:7630-7637.[Abstract/Free Full Text]
37
37. Minamino, T., K. Imada, and K. Namba. 2008. Mechanisms of type III protein export for bacterial flagellar assembly. Mol. Biosyst. 4:1105-1115.[CrossRef][Medline]
38
38. Minamino, T., K. Imada, and K. Namba. 2008. Molecular motors of the bacterial flagella. Curr. Opin. Struct. Biol. 18:693-701.[CrossRef][Medline]
39
39. Minamino, T., and R. M. Macnab. 1999. Components of the Salmonella flagellar export apparatus and classification of export substrates. J. Bacteriol. 181:1388-1394.[Abstract/Free Full Text]
40
40. Minamino, T., and R. M. Macnab. 2000. Domain structure of Salmonella FlhB, a flagellar export component responsible for substrate specificity switching. J. Bacteriol. 182:4906-4914.[Abstract/Free Full Text]
41
41. Minamino, T., and K. Namba. 2008. Distinct roles of the ATPase and proton motive force in bacterial flagellar protein export. Nature 451:485-488.[CrossRef][Medline]
42
42. Minamino, T., and K. Namba. 2004. Self-assembly and type III protein export of the bacterial flagellum. J. Mol. Microbiol. Biotechnol. 7:5-17.[CrossRef][Medline]
43
43. Minamino, T., Y. Saijo-Hamano, Y. Furukawa, B. Gonzalez-Pedrajo, R. M. Macnab, and K. Namba. 2004. Domain organization and function of Salmonella FliK, a flagellar hook length control protein. J. Mol. Biol. 341:491-502.[CrossRef][Medline]
44
44. Moriya, N., T. Minamino, K. T. Hughes, R. M. Macnab, and K. Namba. 2006. The type III flagellar export specificity switch is dependent on FliK ruler and a molecular clock. J. Mol. Biol. 359:466-477.[CrossRef][Medline]
45
45. Muramoto, K., S. Makishima, S. I. Aizawa, and R. M. Macnab. 1998. Effect of cellular level of FliK on flagellar hook and filament assembly in Salmonella typhimurium. J. Mol. Biol. 277:871-882.[CrossRef][Medline]
46
46. Ohnishi, K., F. Fan, G. J. Schoenhals, M. Kihara, and R. M. Macnab. 1997. The FliO, FliP, FliQ, and FliR proteins of Salmonella typhimurium: putative components for flagellar assembly. J. Bacteriol. 179:6092-6099.[Abstract/Free Full Text]
47
47. Ohnishi, K., M. Homma, K. Kutsukake, and T. Iino. 1987. Formation of flagella lacking outer rings by flaM, flaU, and flaY mutants of Escherichia coli. J. Bacteriol. 169:1485-1488.[Abstract/Free Full Text]
48
48. Ohnishi, K., K. Kutsukake, H. Suzuki, and T. Lino. 1992. A novel transcriptional regulation mechanism in the flagellar regulon of Salmonella typhimurium: an anti-sigma factor inhibits the activity of the flagellum-specific sigma factor, sigma F. Mol. Microbiol. 6:3149-3157.[Medline]
49
49. Paul, K., M. Erhardt, T. Hirano, D. F. Blair, and K. T. Hughes. 2008. Energy source of flagellar type III secretion. Nature 451:489-492.[CrossRef][Medline]
50
50. Saijo-Hamano, Y., K. Imada, T. Minamino, M. Kihara, R. M. Macnab, and K. Namba. 2005. Crystallization and preliminary X-ray analysis of the C-terminal cytoplasmic domain of FlhA, a membrane-protein subunit of the bacterial flagellar type III protein-export apparatus. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 61:599-602.[CrossRef][Medline]
51
51. Saijo-Hamano, Y., T. Minamino, R. M. Macnab, and K. Namba. 2004. Structural and functional analysis of the C-terminal cytoplasmic domain of FlhA, an integral membrane component of the type III flagellar protein export apparatus in Salmonella. J. Mol. Biol. 343:457-466.[CrossRef][Medline]
52
52. Shibata, S., N. Takahashi, F. F. Chevance, J. E. Karlinsey, K. T. Hughes, and S. Aizawa. 2007. FliK regulates flagellar hook length as an internal ruler. Mol. Microbiol. 64:1404-1415.[CrossRef][Medline]
53
53. Stafford, G. P., and C. Hughes. 2007. Salmonella typhimurium flhE, a conserved flagellar regulon gene required for swarming. Microbiology 153:541-547.[Abstract/Free Full Text]
54
54. Wang, S., R. T. Fleming, E. M. Westbrook, P. Matsumura, and D. B. McKay. 2006. Structure of the Escherichia coli FlhDC complex, a prokaryotic heteromeric regulator of transcription. J. Mol. Biol. 355:798-808.[CrossRef][Medline]

---

Journal of Bacteriology, June 2009, p. 3938-3949, Vol. 191, No. 12
0021-9193/09/$08.00+0     doi:10.1128/JB.01811-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hirano, T. Articles by Hughes, K. T. Search for Related Content PubMed PubMed Citation Articles by Hirano, T. Articles by Hughes, K. T.