ArfA recruits release factor 2 to rescue stalled ribosomes by peptidyl-tRNA hydrolysis in Escherichia coli.

The ribosomes stalled at the end of non-stop mRNAs must be rescued for productive cycles of cellular protein synthesis. Escherichia coli possesses at least three independent mechanisms that resolve non-productive translation complexes (NTCs). While tmRNA (SsrA) mediates trans-translation to terminate translation, ArfA (YhdL) and ArfB (YaeJ) induce hydrolysis of ribosome-tethered peptidyl-tRNAs. ArfB is a paralogue of the release factors (RFs) and directly catalyses the peptidyl-tRNA hydrolysis within NTCs. In contrast, the mechanism of the ArfA action had remained obscure beyond its ability to bind to the ribosome. Here, we characterized the ArfA pathway of NTC resolution in vitro and identified RF2 as a factor that cooperates with ArfA to hydrolyse peptidyl-tRNAs located in the P-site of the stalled ribosome. This reaction required the GGQ (Gly-Gly-Gln) hydrolysis motif, but not the SPF (Ser-Pro-Phe) codon-recognition sequence, of RF2 and was stimulated by tRNAs. From these results we suggest that ArfA binds to the vacant A-site of the stalled ribosome with possible aid from association with a tRNA, and then recruits RF2, which hydrolyses peptidyl-tRNA in a GGQ motif-dependent but codon-independent manner. In support of this model, the ArfA-RF2 pathway did not act on the SecM-arrested ribosome, which contains an aminoacyl-tRNA in the A-site.

Functional and expressional analyses of the anti-FlhD4C2 factor gene ydiV in Escherichia coli.

Although Escherichia coli and Salmonella enterica serovar Typhimurium have a similar flagellar regulatory system, the response of flagellar synthesis to nutrient conditions is quite different between the two: that is, in low-nutrient conditions, flagellar synthesis is inhibited in Salmonella and enhanced in E. coli. In Salmonella, this inhibition is mediated by an anti-FlhD(4)C(2) factor, YdiV, which is expressed in low-nutrient conditions and binds to FlhD(4)C(2) to inhibit the expression of the class 2 flagellar genes. The fliZ gene encodes a repressor of the ydiV gene, and thus is required for efficient flagellar gene expression in low-nutrient conditions in Salmonella. In this study, we showed that the E. coli ydiV gene encodes a protein which inhibits motility and flagellar production when expressed from a multicopy plasmid. We showed further that E. coli YdiV binds to FlhD(4)C(2) and inhibits its binding to the class 2 flagellar promoter. These results indicate that E. coli YdiV can also act as an anti-FlhD(4)C(2) factor. However, although the ydiV gene was transcribed efficiently in E. coli cells, the intracellular level of the YdiV protein was extremely low due to its inefficient translation. Consistent with this, E. coli cells did not require FliZ for efficient motility development. This indicates that, unlike in Salmonella, the FliZ-YdiV regulatory system does not work in the nutritional control of flagellar gene expression in E. coli.

The transcript from the σ(28)-dependent promoter is translationally inert in the expression of the σ(28)-encoding gene fliA in the fliAZ operon of Salmonella enterica serovar Typhimurium.

There are three classes of promoters for flagellar operons in Salmonella. Class 2 promoters are transcribed by σ(70) RNA polymerase in the presence of an essential activator, FlhD(4)C(2), and activated by an auxiliary regulator, FliZ. Class 3 promoters are transcribed by σ(28) RNA polymerase and repressed by an anti-σ(28) factor, FlgM. σ(28) (FliA) and FliZ are encoded by the fliA and fliZ genes, respectively, which together constitute an operon transcribed in this order. This operon is transcribed from both class 2 and class 3 promoters, suggesting that it should be activated by its own product, σ(28), even in the absence of FlhD(4)C(2). However, σ(28)-dependent transcription occurs in vivo only in the presence of FlhD(4)C(2), indicating that transcription from the class 2 promoter is a prerequisite to that from the class 3 promoter. In this study, we examined the effects of variously modified versions of the fliA regulatory region on transcription and translation of the fliA gene. We showed that FliA is not significantly translated from the class 3 transcript. In contrast, the 5'-terminal AU-rich sequence found in the class 2 transcript confers efficient fliA translation. Replacement of the Shine-Dalgarno sequence of the fliA gene with a better one improved fliA translation from the class 3 transcript. These results suggest that the 5'-terminal AU-rich sequence of the class 2 transcript may assist ribosome binding. FliZ was shown to be expressed from both the class 2 and class 3 transcripts.

trans-translation-mediated tight regulation of the expression of the alternative ribosome-rescue factor ArfA in Escherichia coli.

Ribosomes translating mRNA without an in-frame stop codon (non-stop mRNA) stall at its 3' end. In eubacteria, such ribosomes are rescued by SsrA-mediated trans-translation. Recently, we have shown that Escherichia coli ArfA (formerly YhdL) also rescues stalled ribosomes by a mechanism distinct from that of trans-translation. Synthetic lethality phenotype of ssrA arfA double mutants suggests that accumulation of stalled ribosomes is deleterious to E. coli cells. In this report, we show that the expression of ArfA is tightly regulated by the system involving trans-translation. Both premature transcription termination and specific cleavage by RNase III were programmed at the specific sites within the arfA open reading frame (ORF) and produced arfA non-stop mRNA. C-terminally truncated ArfA protein synthesized from arfA non-stop mRNA was tagged through SsrA-mediated trans-translation and degraded in wild type cell. In the absence of SsrA, however, C-terminally truncated ArfA escaped from degradation and had a function to rescue stalled ribosomes. Full-length ArfA produced only when arfA mRNA escapes from both premature transcription termination and RNase III cleavage was unstable. From these results, we illustrate a regulatory model in which ArfA is expressed only when it is needed, namely, when the ribosome rescue activity of trans-translation system is insufficient to support cell viability. This sophisticated regulatory mechanism suggests that the ArfA-mediated ribosome rescue is a backup system for trans-translation.

FliZ acts as a repressor of the ydiV gene, which encodes an anti-FlhD4C2 factor of the flagellar regulon in Salmonella enterica serovar typhimurium.

YdiV acts as an anti-FlhD4C2 factor, which negatively regulates the class 2 flagellar operons in poor medium in Salmonella enterica serovar Typhimurium. On the other hand, one of the class 2 flagellar genes, fliZ, encodes a positive regulator of the class 2 operons. In this study, we found that the FliZ-dependent activation of class 2 operon expression was more profound in poor medium than in rich medium and not observed in the ydiV mutant background. Transcription of the ydiV gene was shown to increase in the fliZ mutant. Purified FliZ protein was shown in vitro to bind to the promoter region of the nlpC gene, which is located just upstream of the ydiV gene, and to repress its transcription. These results indicate that FliZ is a repressor of the nlpC-ydiV operon and activates the class 2 operons by repressing ydiV expression. Therefore, the fliZ and ydiV genes form a regulatory loop.

Escherichia coli YaeJ protein mediates a novel ribosome-rescue pathway distinct from SsrA- and ArfA-mediated pathways.

Accumulation of stalled ribosomes at the 3' end of mRNA without a stop codon (non-stop mRNA) is supposed to be toxic to bacterial cells. Escherichia coli has at least two distinct systems to rescue such stalled ribosomes: SsrA-dependent trans-translation and ArfA-dependent ribosome rescue. Combination of the ssrA and arfA mutations is synthetically lethal, suggesting the significance of ribosome rescue. In this study, we identified the E. coli yaeJ gene, encoding a peptide-release factor homologue with GGQ motif, as a multicopy suppressor of the lethal phenotype of ssrA arfA double mutant. The YaeJ protein was shown to bind to ribosomes. Both in vivo and in vitro, YaeJ showed the ribosome-rescue activity and promoted the hydrolysis of peptidyl-tRNA residing in the stalled ribosome. Missense mutation in the GGQ motif or deletion of the C-terminal unstructured tail abolished both the suppressor activity for ssrA arfA synthetic lethality and the ribosome-rescue activity, suggesting the importance of these structural features. On the basis of these observations, we propose that YaeJ acts as a stop codon-independent peptidyl-tRNA hydrolysing factor through binding to ribosomes stalled at the 3' end of non-stop mRNAs. It was also suggested that ArfA and YaeJ rescue the stalled ribosomes by distinct mechanisms.

EAL domain protein YdiV acts as an anti-FlhD4C2 factor responsible for nutritional control of the flagellar regulon in Salmonella enterica Serovar Typhimurium.

Flagellar operons are divided into three classes with respect to their transcriptional hierarchy in Salmonella enterica serovar Typhimurium. The class 1 gene products FlhD and FlhC act together in an FlhD(4)C(2) heterohexamer, which binds upstream of the class 2 promoters to facilitate binding of RNA polymerase. In this study, we showed that flagellar expression was much reduced in the cells grown in poor medium compared to those grown in rich medium. This nutritional control was shown to be executed at a step after class 1 transcription. We isolated five Tn5 insertion mutants in which the class 2 expression was derepressed in poor medium. These insertions were located in the ydiV (cdgR) gene or a gene just upstream of ydiV. The ydiV gene is known to encode an EAL domain protein and to act as a negative regulator of flagellar expression. Gene disruption and complementation analyses revealed that the ydiV gene is responsible for nutritional control. Expression analysis of the ydiV gene showed that its translation, but not transcription, was enhanced by growth in poor medium. The ydiV mutation did not have a significant effect on either the steady-state level of flhDC mRNA or that of FlhC protein. Purified YdiV protein was shown in vitro to bind to FlhD(4)C(2) through interaction with FlhD subunit and to inhibit its binding to the class 2 promoter, resulting in inhibition of FlhD(4)C(2)-dependent transcription. Taking these data together, we conclude that YdiV is a novel anti-FlhD(4)C(2) factor responsible for nutritional control of the flagellar regulon.

Ribosome rescue by Escherichia coli ArfA (YhdL) in the absence of trans-translation system.

Although SsrA(tmRNA)-mediated trans-translation is thought to maintain the translation capacity of bacterial cells by rescuing ribosomes stalled on messenger RNA lacking an in-frame stop codon, single disruption of ssrA does not crucially hamper growth of Escherichia coli. Here, we identified YhdL (renamed ArfA for alternative ribosome-rescue factor) as a factor essential for the viability of E. coli in the absence of SsrA. The ssrA-arfA synthetic lethality was alleviated by SsrA(DD) , an SsrA variant that adds a proteolysis-refractory tag through trans-translation, indicating that ArfA-deficient cells require continued translation, rather than subsequent proteolysis of the truncated polypeptide. In accordance with this notion, depletion of SsrA in the ΔarfA background led to reduced translation of a model protein without affecting transcription, and puromycin, a codon-independent mimic of aminoacyl-tRNA, rescued the bacterial growth under such conditions. That ArfA takes over the role of SsrA was suggested by the observation that its overexpression enabled detection of the polypeptide encoded by a model non-stop mRNA, which was otherwise SsrA-tagged and degraded. In vitro, purified ArfA acted on a ribosome-nascent chain complex to resolve the peptidyl-tRNA. These results indicate that ArfA rescues the ribosome stalled at the 3' end of a non-stop mRNA without involving trans-translation.

Measurement of sialic acid content is insufficient to assess bioactivity of recombinant human erythropoietin.

Assessment of biological potency and its comparison with clinical effects are important in the quality control of therapeutic glycoproteins. Animal models are usually used for evaluating bioactivity of these compounds. However, alternative methods are required to simplify the bioassay and avoid ethical issues associated with animal studies. Negatively charged sialic acid residues are known to be critical for in vivo bioactivity of recombinant human erythropoietin (rhEPO). In this study, we used capillary zone electrophoresis, a charge-based separation method, to estimate the sialic acid content for predicting in vivo bioactivity of rhEPO. In vivo bioactivities of rhEPO subfractions were measured and compared with sialylation levels. The results obtained indicated that in vivo bioactivity of rhEPO is not simply correlated with the sialylation level, which suggests that it is difficult to predict biological potency from the sialic acid content alone. N-Glycan moieties as well as sialic acid residues may have a significant impact on in vivo bioactivity of rhEPO.

Suppression by enhanced RpoE activity of the temperature-sensitive phenotype of a degP ssrA double mutant in Escherichia coli.

SsrA is a small RNA playing a crucial role in trans-translation, which leads to rescue of stalled ribosomes on or at the end of mRNA and addition of the degradation tag to a growing polypeptide. The lack of SsrA has been shown to enhance the temperature-sensitive (ts) phenotype of an E. coli strain defective in the degP gene, which encodes one of the periplasmic proteases. This severe ts phenotype was relieved only partially by an SsrADD variant, which can lead to ribosome rescue but adds a protease-resistant tag instead of the degradation tag, suggesting that accumulation of polypeptides programmed by truncated mRNAs is responsible for growth defect of the ssrA degP mutant. Expression of an S210A-mutant DegP protein, which lacks the protease activity but retains the chaperone activity, could relieve the ts phenotype of the double mutant, suggesting that the chaperone activity but not the protease activity of DegP is required for growth of the ssrA-deficient cells at high temperature. Overexpression of the rpoE gene, which encodes sigmaE responsible for the expression of factors involved in extracellular stress response, also suppressed the ts phenotype of the ssrA degP mutant. This suggests that the stress-responsing pathway(s) may be involved in the enhancement of ts phenotype of degP mutant in the absence of SsrA.

Production of novel anti-recombinant human erythropoietin monoclonal antibodies and development of a sensitive enzyme-linked immunosorbent assay for detection of bioactive human erythropoietin.

Erythropoietin (EPO) is a growth factor, regulating the proliferation and differentiation of erythroid progenitor cells. In this study, we generated five monoclonal antibodies (mAbs) that reacted specifically with recombinant human EPO (rhEPO). Epitope exclusion and other experiments showed that the mAbs obtained were divided into two groups, differing in recognition sites for rhEPO: group 1 mAbs recognize the N-terminal region of rhEPO, whereas group 2 mAbs seem to recognize a conformation-dependent epitope. Although most of the previously reported anti-EPO antibodies that recognized the N-terminal region of EPO lacked the EPO-neutralizing activity, the group 1 mAbs obtained here had the rhEPO-neutralizing activity. Therefore, the group 1 mAbs may be useful for future study on structure-function relationship of EPO. One of the group 2 mAbs, 5D11A, showed the highest affinity for rhEPO with K(D) value 0.52 nM and had the highest rhEPO-neutralizing activity. Using this mAb, we developed a reproducible and sensitive enzyme-linked immunosorbent assay for the quantification of bioactive rhEPO.

Expressed and cryptic flagellin genes in the H44 and H55 type strains of Escherichia coli.

Bacterial H antigens are specified by flagellin molecules, which constitute the flagellar filament. Escherichia coli 781-55 and E2987-73 are the type strains for H44 and H55 antigens, respectively. Unlike E. coli K-12, they possess two flagellin genes, fliC and fllA, on their chromosomes. However, they are monophasic, expressing exclusively the fllA genes, which specify the type antigens. In this study, the flagellin genes were cloned from these strains and their structure and expression were analyzed. It was found that the fliC genes encode apparently intact flagellin subunits but possess inefficient sigma28-dependent promoters, which may result in these genes being silent. The chromosomal locations of the fllA genes are approximately, but not exactly, identical with that of the phase-2 flagellin gene, fljB, of diphasic Salmonella strains. However, unlike the Salmonella fljB gene, the invertible H segment and the fljA gene responsible for the control of flagellar phase variation are both absent from the fllA loci. The fllA genes are highly homologous to the E. coli fliC gene but distantly related to the Salmonella fljB gene. These results suggest a hypothesis that the fllA genes may have emerged by an intra-species lateral transfer of the fliC gene. This hypothesis is further supported by the observation that the fllA genes are flanked by several IS elements and located within cryptic prophage elements.

FliT acts as an anti-FlhD2C2 factor in the transcriptional control of the flagellar regulon in Salmonella enterica serovar typhimurium.

Flagellar operons are divided into three classes with respect to their transcriptional hierarchy in Salmonella enterica serovar Typhimurium. The class 1 gene products FlhD and FlhC act together in an FlhD(2)C(2) heterotetramer, which binds upstream of the class 2 promoters to facilitate binding of RNA polymerase. Class 2 expression is known to be enhanced by a disruption mutation in a flagellar gene, fliT. In this study, we purified FliT protein in a His-tagged form and showed that the protein prevented binding of FlhD(2)C(2) to the class 2 promoter and inhibited FlhD(2)C(2)-dependent transcription. Pull-down and far-Western blotting analyses revealed that the FliT protein was capable of binding to FlhD(2)C(2) and FlhC and not to FlhD alone. We conclude that FliT acts as an anti-FlhD(2)C(2) factor, which binds to FlhD(2)C(2) through interaction with the FlhC subunit and inhibits its binding to the class 2 promoter.

Two DNA invertases contribute to flagellar phase variation in Salmonella enterica serovar Typhimurium strain LT2.

Salmonella enterica serovar Typhimurium strain LT2 possesses two nonallelic structural genes, fliC and fljB, for flagellin, the component protein of flagellar filaments. Flagellar phase variation occurs by alternative expression of these two genes. This is controlled by the inversion of a DNA segment, called the H segment, containing the fljB promoter. H inversion occurs by site-specific recombination between inverted repetitious sequences flanking the H segment. This recombination has been shown in vivo and in vitro to be mediated by a DNA invertase, Hin, whose gene is located within the H segment. However, a search of the complete genomic sequence revealed that LT2 possesses another DNA invertase gene that is located adjacent to another invertible DNA segment within a resident prophage, Fels-2. Here, we named this gene fin. We constructed hin and fin disruption mutants from LT2 and examined their phase variation abilities. The hin disruption mutant could still undergo flagellar phase variation, indicating that Hin is not the sole DNA invertase responsible for phase variation. Although the fin disruption mutant could undergo phase variation, fin hin double mutants could not. These results clearly indicate that both Hin and Fin contribute to flagellar phase variation in LT2. We further showed that a phase-stable serovar, serovar Abortusequi, which is known to possess a naturally occurring hin mutation, lacks Fels-2, which ensures the phase stability in this serovar.

FljA-mediated posttranscriptional control of phase 1 flagellin expression in flagellar phase variation of Salmonella enterica serovar Typhimurium.

Flagellar phase variation of Salmonella is a phenomenon where two flagellin genes, fliC (phase 1) and fljB (phase 2), are expressed alternately. This is controlled by the inversion of a DNA segment containing the promoter for the fljB gene. The fljB gene constitutes an operon with the fljA gene, which encodes a negative regulator for fliC expression. Previous biochemical analysis suggested that phase variation might depend on alternative synthesis of phase-specific flagellin mRNA (H. Suzuki and T. Iino, J. Mol. Biol. 81:57-70, 1973). However, recently reported results suggested that FljA-dependent inhibition might be mediated by a posttranscriptional control mechanism (H. R. Bonifield and K. T. Hughes, J. Bacteriol. 185:3567-3574, 2003). In this study, we reexamined the mechanism of FljA-mediated inhibition of fliC expression more carefully. Northern blotting analysis revealed that no fliC mRNA was detected in phase 2 cells. However, only a moderate decrease in beta-galactosidase activity was observed from the fliC-lacZ transcriptional fusion gene in phase 2 cells compared with that in phase 1 cells. In contrast, the expression of the fliC-lacZ translational fusion gene was severely impaired in phase 2 cells. The half-life of fliC mRNA was shown to be much shorter in phase 2 cells than in phase 1 cells. Purified His-tagged FljA protein was shown to bind specifically to fliC mRNA and inhibit the translation from fliC mRNA in vitro. On the basis of these results, we propose that in phase 2 cells, FljA binds to fliC mRNA and inhibits its translation, which in turn facilitates its degradation.

Plasticity of the domain structure in FlgJ, a bacterial protein involved in flagellar rod formation.

Bacterial flagellar rod structure is built across the peptidoglycan (PG) layer. A Salmonella enterica flagellar protein FlgJ is believed to consist of two functional domains, the N-terminal half acting as a scaffold or cap essential for rod assembly and the C-terminal half acting as a PG hydrolase (PGase) that makes a hole in the PG layer to facilitate rod penetration. In this study, molecular data analyses were conducted on FlgJ data sets sampled from a variety of bacterial species, and three types of FlgJ homologs were identified: (i) "canonical dual-domain" type found in beta- and gamma-proteobacteria that has a domain for one of the PGases, acetylmuramidase (Acm), at the C terminus, (ii) "non-canonical dual-domain" type found in the genus Desulfovibrio (delta-proteobacteria) that bears a domain for another PGase, M23/M37-family peptidase (Pep), at the C terminus and (iii) "single-domain" type found in phylogenetically diverged lineages that lacks the Acm or Pep domain. FlgJ phylogeny, together with the domain architecture, suggested that the single-domain type was the original form of FlgJ and the canonical dual-domain type had evolved from the single-domain type by fusion of the Acm domain to its C terminus in the common ancestor of beta- and gamma-proteobacteria. The non-canonical dual-domain type may have been formed by fusion of the Pep domain to the single-domain type in the ancestor of Desulfovibrio. In some lineages of gamma-proteobacteria, the Acm domain appeared to be lost secondarily from the dual-domain type FlgJ to yield again a single-domain type one. To rationalize the underlying mechanism that gave rise to the two different types of dual-domain FlgJ homologs, we propose a model assuming the lineage-specific co-option of flagellum-specific PGase from diverged housekeeping PGases in bacteria.

Interactions between bacterial flagellar axial proteins in their monomeric state in solution.

The axial structure of the bacterial flagellum is composed of many different proteins, such as hook protein and flagellin, and each protein forms a short or long axial segment one after another in a well-defined order along the axis. Under physiological conditions, most of these proteins are stable in the monomeric state in solution, and spontaneous polymerization appears to be suppressed, as demonstrated clearly for flagellin, probably to avoid undesirable self-assembly in the cytoplasmic space. However, no systematic studies of the possible associations between monomeric axial proteins in solution have been carried out. We therefore studied self and cross-association between hook protein, flagellin and three hook-associated proteins, HAP1, HAP2 and HAP3, in all possible pairs, by gel-filtration and analytical centrifugation, and found interactions in the following two cases only. Flagellin facilitated HAP3 aggregation into beta-amyloid-like filaments, but without stable binding between the two. Addition of HAP3 to HAP2 resulted in disassembly of preformed HAP2 decamers and formation of stable HAP2-HAP3 heterodimers. HAP2 missing either of its disordered terminal regions did not form the heterodimer, whereas HAP3 missing either of its disordered terminal regions showed stable heterodimer formation. This polarity in the heterodimer interactions suggests that the interactions between HAP2 and HAP3 in solution are basically the same as those in the flagellar axial structure. We discuss these results in relation to the assembly mechanism of the flagellum.

The ClpXP ATP-dependent protease regulates flagellum synthesis in Salmonella enterica serovar typhimurium.

The ClpXP protease is a member of the ATP-dependent protease family and plays a dynamic role in the control of availability of regulatory proteins and the breakdown of abnormal and misfolded proteins. The proteolytic activity is rendered by the ClpP component, while the substrate specificity is determined by the ClpX component that has ATPase activity. We describe here a new role of the ClpXP protease in Salmonella enterica serovar Typhimurium in which ClpXP is involved in the regulation of flagellum synthesis. Cells deleted for ClpXP show "hyperflagellate phenotype," exhibit overproduction of the flagellar protein, and show a fourfold increase in the rate of transcription of the fliC encoding flagellar filament. The assay for promoter activity of the genes responsible for expression of the fliC showed that the depletion of ClpXP results in dramatic enhancement of the expression of the fliA encoding sigma factor final sigma(28), leaving the expression level of the flhD master operon lying at the top of the transcription hierarchy of flagellar regulon almost normal. These results suggest that the ClpXP may be responsible for repressing the expression of flagellar regulon through the control of the FlhD/FlhC master regulators at the posttranscriptional and/or posttranslational levels. Proteome analysis of proteins secreted from the mutant cells deficient for flhDC and clpXP genes demonstrated that the DeltaflhD mutation abolished the enhanced effect by DeltaclpXP mutation on the production of flagellar proteins, suggesting that the ClpXP possibly defines a regulatory pathway affecting the expression of flagellar regulon that is dependent on FlhD/FlhC master regulators.

The Salmonella FlgA protein, a putativeve periplasmic chaperone essential for flagellar P ring formation.

P ring is a periplasmic substructure of the flagellar basal body and is believed to connect with the peptidoglycan layer in Salmonella. Two flagellar genes, flgA and flgI, are known to be indispensable for P ring formation. The flgI gene encodes the component protein of the P ring. However, the role of the flgA gene product in P ring assembly remained unknown. Here, evidence is presented that FlgA is synthesized as a precursor form and exported via the Sec secretory pathway into the periplasmic space where P ring formation takes place. Overproduction of the FlgI protein led flgA mutants to form flagella with a P ring, suggesting that FlgA plays an auxiliary role in P ring assembly. Far-Western blot analysis revealed that FlgA binds in vitro to both FlgI and FlgA itself. Though a direct FlgI-FlgI interaction in the absence of FlgA could not be demonstrated, an indirect or direct interaction between the FlgI proteins was observed in the presence of FlgA. FlgA alone was very unstable in vivo, but co-expression with FlgI could stabilize FlgA. This suggests the presence of FlgA-FlgI interaction in vivo. On the basis of these results, a hypothesis is proposed that FlgA acts as a periplasmic chaperone, which assists a polymerization reaction of FlgI into the P ring through FlgA-FlgI interaction.

Structure and expression of the fliA operon of Salmonella typhimurium.

The fliA gene encodes the flagellum-specific sigma factor sigma28 In Salmonella typhimurium. The transcription in vivo and in vitro of this gene was analysed and it was found that there are two promoters for the expression of this gene. One is a class 2 promoter which is recognized by sigma70-RNA polymerase in the presence of the FlhD and FlhC activator proteins. The other is a class 3 promoter which is recognized by sigma28-RNA polymerase. Therefore, the fliA operon is under dual positive control from FlhD/FlhC and from FliA itself. The nucleotide sequence downstream of the fliA gene was determined. The sequence contains two ORFs following the fliA gene. On the basis of their sequence homology, it is concluded that these two correspond to the fliZ and fliY genes of Escherichia coil. Northern blot analysis revealed that the fliZ gene is transcribed from the fliA promoters, whereas the fliY gene is transcribed from both the fliA promoters and its own FlhD/FlhC-independent promoter. A fliZ-disruption mutant was constructed by inserting a kanamycin-resistance gene cassette into the fliZ gene on the chromosome. The mutant showed poor motility, and introduction of a fliZ+ plasmid into this mutant restored the wildtype level of motility. These results suggest that the fliZ gene may be required for expression of maximal motility.