Arginine within a specific motif near the N-terminal of FimY is critical for the maximal production of type 1 fimbriae in Salmonella enterica serovar Typhimurium.

An important Salmonella serovar for both human and animals Salmonella Typhimurium possesses 13 gene clusters that have the potential to produce fimbrial structure, among which the type 1 fimbriae with the binding specificity to mannose residue is the most commonly found type. Six structural genes and five regulatory genes comprise the fim gene cluster that is responsible for the production of type 1 fimbriae in S. Typhimurium. The fimY gene encodes a positive regulator for type 1 fimbrial expression since a deletion in fimY abolished the production of fimbriae. The N-terminal portion of FimY contains amino acid residues that exhibit some similarity as those found in the proteins possessing the PilZ domain, which is engaged in cyclic di-GMP binding. A fimY allele that had a change from arginine to alanine at position 7 (R7A) or 7 and 11 (R7/11A) generated by site-directed mutagenesis in a 6 RRERH11 R motif near N-terminal, when cloned in pACYC184 and transformed into a fimY-deleted strain, decreased the expression of fimA and fimZ. The number of type 1 fimbriae in these two transformants was also less than those of the other transformants that contained different fimY alleles in pACYC184 when observed in electron microscopy. However, changing from arginine to alanine at position 11 (R11A) remained the same as the wild-type fimY allele. It is likely that the arginine at the 7th position of FimY is critical for its maximal activating activity upon fimZ. Another motif 83 DI85 SLWIEK91 G motif did not affect the function of FimY. Although FimY has the two aforementioned motifs, which contain some amino acids that are present within those of the PilZ domain proteins, secondary structure prediction analysis did not reveal that FimY has a conformation commonly observed in PilZ-like proteins. Therefore, FimY and PilZ domain proteins are not homologs. Further investigation for a detailed analysis of FimY is thus warranted.

INTRODUCTION

Salmonella belongs to the member of the family Enterobacteriaceae and is an important zoonotic agent of public health concern. Nontyphoid Salmonella accounted for over 59,000 deaths and they were with the highest rank for disability adjusted life years among the foodborne disease hazards in 2010 (Kirk et al., 2015). Salmonella enterica contains more than 2,579 serovars while S. Typhimurium is one of the major serovars of Salmonella that account for infections in human and animals (CDC, 2011). Adhesion to the host cell is the first step to establish infection for many bacteria including S. Typhimurium. Adhesion molecules on the surface of bacteria or conjugated on the shaft of a structure called fimbria have been implicated in engaging such an adherent event. In addition, signal transduction in bacteria can also be mediated by adhesion to the host cell (Moorthy, Keklak, & Klein, 2016). Fimbriae are hair‐like appendages present on the outer membrane of many bacteria. S. Typhimurium has been documented to possess 13 different fimbrial gene clusters that may have the potential to produce fimbriae (McClelland et al., 2001). Among these fimbriae types, type 1 fimbriae with the binding specificity to mannosylated residue is frequently found and is also referred to as the common fimbriae (Duguid, Anderson, & Campbell, 1966). The fim gene cluster in the genome of S. Typhimurium is responsible for the phenotypic expression of type 1 fimbriae. The fim gene cluster consists of genes for structure, biosynthesis, and those for regulation. FimA, FimI, FimF, and FimH are fimbrial structural subunits that incorporate each other to form a fimbrial shaft. Fimbrial subunits are assembled and anchored in the outer membrane of bacteria by chaperone protein FimC and usher protein FimD, respectively (Clegg & Swenson, 1994). The genes fimZ, fimY, fimW, stm0551, and fimU have been shown to regulate type 1 fimbrial production through a complicated circuit involving both the transcriptional and translational levels and protein–protein interaction as well (Tinker & Clegg, 2001; Tinker, Hancox, & Clegg, 2001; Yeh, Hancox, & Clegg, 1995). FimY is a positive regulator for type 1 fimbrial expression in S. Typhimurium. The C‐terminal part of FimY possesses a helix‐turn‐helix motif and FimY was shown to bind to the promoter region of fimZ (Wang, Hsu, Huang, Lin, & Yeh, 2014); however, some aspects of its mechanism in terms of fimbrial regulation remains to be elucidated. For example, the members of the PilZ family regulator, such as YcgR of Escherichia coli and MrkH of Klebsiella pneumoniae (Ryjenkov, Simm, Romling, & Gomelsky, 2006; Wilksch et al., 2011), often contain two conserved domains RRxxxR and DxSxxG (x designates any amino acid), which involve in binding cyclic di‐GMP (c‐di‐GMP), a second messenger that controls many physiological processes of bacteria (Jenal, Reinders, & Lori, 2017). According to the protein sequence alignment, the N‐terminal of FimY also possesses some amino acids that are present in these motifs (Figure 1). FimY harbors a 6RRERH11R and a 83DI85SLWIEK91G motif, with the second motif having extra three amino acids as compared with the DxSxxG motif; yet, its role as a c‐di‐GMP–binding protein has not been reported. Interestingly, a gene, stm0551, located just upstream of fimY was proven to encode a phosphodiesterase with an EAL domain whose function is to degrade c‐di‐GMP (Wang, Hsu, Huang, & Yeh, 2012). Nevertheless, there is no protein resembling a diguanylate cyclase to synthesize c‐di‐GMP within the fim gene cluster.

Figure 1

Conservation of PilZ‐containing domain proteins. Residues that show strict identity are written in white characters and highlighted in red. Similarity is indicated with black characters and highlighted in yellow. Residues that are crucial for c‐di‐GMP binding in the PilZ domain are marked with an asterisk. Protein names and microorganisms are as follows: FimY‐LT2, FimY of Salmonella Typhimurium LT2; MrkH‐KP, MrkH of Klebsiella pneumoniae AJ218; YcgR‐K12, YcgR of Escherichia coli K‐12

The present study reported here attempts to uncover the potential roles of 6RRERH11R and 83DI85SLWIEK91G of FimY and investigate how it may affect type 1 fimbrial expression in S. Typhimurium. We found that the 7th arginine at the 6RRERH11R motif of FimY may affect the production of type 1 fimbriae by activating fimZ and subsequently resulting in the fimbrial major subunit gene fimA expression.

MATERIALS AND METHODS

Bacterial strains, plasmids, and culture conditions

Table 1 lists the bacterial strains and plasmids used in this study. All strains were grown in Luria–Bertani (LB) broth or on an LB agar at 37°C. For transformant selection, media may be required to be supplemented with antibiotics such as ampicillin (100 μg/ml), kanamycin (25 μg/ml), or chloramphenicol (25μg/ml). Escherichia coli strains were carried out for plasmid construction, plasmid purification, and cloning using standard techniques. For the type 1 fimbriae inducing condition, all S. Typhimurium strains were cultured in static broth at 37°C for 48 hr.

Table 1

Bacterial strains and plasmids used in this study

Strain or plasmid	Description	Abbreviation	Source or reference
Strains
Salmonella Typhimurium
LB5010	a S. Typhimurium LT2 derivative, wild type, fimbriate with the complete fim gene cluster	Wild type
LB5010ΔfimY	a fimY‐deleted strain	ΔfimY	This study
LB5010ΔfimY (fimY)	a fimY‐deleted strain transformed with pfimY	ΔfimY (fimY)	This study
LB5010ΔfimY (pACYC184)	a fimY‐deleted strain transformed with pACYC184	ΔfimY (pACYC184)	This study
LB5010ΔfimY (fimY R7A)	a fimY‐deleted strain transformed with pfimY R7A	R7A	This study
LB5010ΔfimY (fimY R11A)	a fimY‐deleted strain transformed with pfimY R11A	R11A	This study
LB5010ΔfimY (fimY R7/11A)	a fimY‐deleted strain transformed with pfimY R7/11A	R7/11A	This study
LB5010ΔfimY (fimY DSG)	a fimY‐deleted strain transformed with pfimY DSG	DSG	This study
LB5010ΔfimY (fimY MUT)	a fimY‐deleted strain transformed with pfimY MUT	MUT	This study
Escherichia coli
One Shot® TOP10	a chemically competent cell strain for molecular cloning		Invitrogen
Plasmids
pKD13	A template plasmid for gene inactivation, Kanr		Datsenko & Wanner, 2000
pKD46	The plasmid expressing λ Red recombinase; Ampr		Datsenko & Wanner, 2000
pACYC184	Cloning vector; Tetr; and Cmr		ATCC
pSTBLUE‐1	Cloning vector; Kanr; and Ampr		Novagen
pfimY	A complete fimY coding sequence cloned into pACYC184; Cmr		This study
pfimY R7A	A fimY coding sequence with R7A cloned into pACYC184; Cmr		This study
pfimY R11A	A fimY coding sequence with R11A cloned into pACYC184; Cmr		This study
pfimY R7/11A	A fimY coding sequence with R7/11A cloned into pACYC184; Cmr		This study
pfimY DSG	A fimY mutant sequence with D83A, S85A, and G91A cloned into pACYC184; Cmr		This study
pfimY MUT	A fimY mutant sequence with R7/11A, D83A, S85A, and G91A cloned into pACYC184; Cmr		This study

Construction of a S. Typhimurium  fimY‐deleted strain

The S. Typhimurium  fimY‐deleted strain was constructed by a one‐step gene inactivation method (Datsenko & Wanner, 2000). Kanamycin resistance gene from pKD13 tagged with a flanking sequence of the fimY gene was generated by a polymerase chain reaction (PCR) technique using fimY‐P1 and fimY‐P4 and transformed by electroporation into S. Typhimurium LB5010 that has previously been transformed with pKD46, the plasmid that encodes a λ red recombinase. The primers and PCR conditions are listed in Table 2. All transformants were grown on LB agar supplemented with kanamycin. The constructed mutants were verified by PCR with primers located in the flanking sequence of the fimY gene.

Table 2

Primers used in this study

Construction of the recombinant plasmids possessing the fimY PilZ‐like domain mutant alleles

The mutant alleles of fimY were generated by site‐directed mutagenesis using an overlapping extension PCR with LB5010 strain genomic DNA (Ho, Hunt, Horton, Pullen, & Pease, 1989). The primers are listed in Table 2. Briefly, fimYSalI‐F and R7A‐R were used to amplify the first DNA fragment, while the second DNA fragment was amplified using fimYBamHI‐R and R7A‐F primers. Ligation of these two DNA fragments with two overlapping ends was achieved with fimYSalI‐F and fimYBamHI‐R primers. The PCR conditions consisted of initial denaturation at 95°C for 30 s, followed by 35 cycles at 95°C for 30 s, 63°C for 30 s, and 72°C for 45 s using KOD DNA polymerase (Novagen, Madison, WI). The ligated fragments were cloned into the pSTBLUE‐1 blunt vector (Novagen) and sequenced to determine whether the codon encoding the specific amino acid had been replaced with alanine. The other fimY alleles were constructed with the same methods, except that all five amino acids of fimPilZ‐like domain mutant were accomplished by primers DSG‐F and DSG‐R using pfimY R7/11A as the template. Table 3 lists the fimY mutant alleles by site‐directed mutagenesis. For the complementation test, the ligated fragment was cloned into the pACYC184 vector and transformed into LB5010ΔfimY strain by electroporation, and the transformants were tested for the ability to produce type 1 fimbriae as described below. Confirmation that only appropriate sequences had been mutated was performed by nucleotide sequence analysis (Mission Biotech, Taipei, Taiwan).

Table 3

Site‐directed mutagenesis of fimY PilZ‐like domain

Plasmid	FimY amino acid sequence
pfimY	7RERH11R 83DI85SLWIEK91G
pfimY R7A	7AERH11R 83DI85SLWIEK91G
pfimY R11A	7RERH11A 83DI85SLWIEK91G
pfimY R7/11A	7AERH11A 83DI85SLWIEK91G
pfimY DSG	7RERH11R 83AI85ALWIEK91A
pfimY MUT	7AERH11A 83AI85ALWIEK91A

Electron microscopy

Salmonella strains were prepared for electron microscopy using the negative staining procedure. Briefly, the bacterial culture was resuspended in ddH2O and mixed with an equal volume of sodium phosphotungstate (2%). A drop of the suspension was placed on a glow‐discharged carbon‐coated copper grid, and surplus fluid was removed with filter papers. The grids were air‐dried and observed with a JEOL JEM‐1400 transmission electron microscope (JEOL Ltd, Tokyo, Japan) operated at 60 or 80 kV.

Detection of type 1 fimbriae by microplate yeast agglutination test

To analyze the expression ability of type 1 fimbriae of Salmonella strains, a microplate agglutination test (MAT) using yeast cells was performed in this study. All the Salmonella strains were cultured in 10 ml LB broth at 37°C for 48 hr statically, followed by centrifugation at 3,000 rpm for 10 min. MAT was performed as follows: First, the bacterial suspensions were adjusted to OD600 1 ± 0.05, followed by twofold serial dilution in phosphate‐buffered saline (PBS), and 100 µl of each dilution preparation was added in the well of a 96‐well round‐bottom microplate. Then, an equal volume (100 µl) of 0.5% Saccharomyces cerevisiae (Sigma‐Aldrich, Darmstadt, Germany) was added to each well and mixed. The plates were incubated at 37°C for 30 min. The highest dilution of the suspensions showing agglutination was defined as the titer end point.

Quantitative RT‐PCR analysis

Total RNA was extracted and purified from bacterial strains cultured in static broth with the ZR Fungal/Bacterial RNA Miniprep™ kit (Zymo Research, Irvine, CA) according to the manufacturer's instructions. To remove the residual DNA, all the extractions were treated with RNase‐free DNase I (ThemoFisher Scientific, Waltham, MS). The purified RNA (0.1 μg) was converted to cDNA using iScript™ cDNA Synthesis kit (Bio‐Rad Laboratories, Hercules, CA), and transcribed into mRNA using iTaq™ universal SYBR® Green supermix (Bio‐Rad Laboratories). Transcription of fimA, fimY, fimZ, and 16s rRNA as an internal control were detected by quantitative RT‐PCR. Cycling conditions were performed using CFX Connect™ Real‐Time System (Bio‐Rad Laboratories) as follows: 95°C for 3 min followed by 49 cycles of 95°C for 10 s and 60°C for 30 s. Melting curves and nontemplate controls were included to detect any primer dimerization or other artifacts. The mRNA transcript levels were analyzed by the software Bio‐Rad CFX Manager 3.1 (Bio‐Rad Laboratories).

Secondary structure analysis

A simple secondary structure prediction program, Profile network from Heidelberg (PHD) in NPS@: Network Protein Sequence Analysis [available at https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSA/npsa_phd.html] (Combet, Blanchet, Geourjon, & Deleage, 2000; Rost & Sander, 1993) was used to predict the secondary structures of FimY, MrkH, DgrA, and VCA0042. To predict the distributions of the secondary structures (e.g., α‐helix, random coil, or β‐sheet), 100 amino acids starting from 6RRERH11R of FimY, 106RRRDP111R of MrkH, 11RRAHP16R of DgrA, and 134QLRKEP140R of VCA0042 were analyzed using PHD.

RESULTS

The electron microscopy

All Salmonella strains were observed for the phenotypic expression of type 1 fimbriae by electron microscopy. S. Typhimurium LB5010 wild type and the ΔfimY (fimY) strains produced type 1 fimbrial appendages, whereas no fimbrial structures were detected in the ΔfimY strain and its transformant containing pACYC184 cloning vector (Figure 2). The ΔfimY strain transformants which received recombinant plasmids possessing different fimY alleles demonstrated different levels of type 1 fimbriae. It was shown that the R7A, R7/11A, and MUT strains revealed relatively fewer numbers of type 1 fimbriae than those that possessed other fimY alleles (Figure 3).

Figure 2

Observation of type 1 fimbriae by electron microscopy. (a) S. Typhimurium LB5010 wild‐type strain produced type 1 fimbriae on the outer membrane of the cell. (b) ΔfimY did not produce type 1 fimbriae. (c) There are no type 1 fimbriae present in the ΔfimY (pACYC184). (d) ΔfimY (fimY) resumed the ability to express type 1 fimbriae. Bacterial cells were negatively stained with 2% of phosphotungstic acid (60,000 x–80,000 x). Fimbriae are indicated by arrow

Figure 3

Observation of type 1 fimbrial expression in the Δ fimY strains that harbored plasmids expressing different fimY allels by electron microscopy. (a) R7A strain (b). R11A strain (c). R7/11A strain (d). DSG strain (e). MUT strain. All the strains produced type 1 fimbriae, whereas R7A and R7/11A strains produced comparatively less fimbriae than the others. Bacterial cells were negatively stained with 2% of phosphotungstic acid (60,000 x‐80,000 x)

Microplate yeast agglutination test was used to quantify the type 1 fimbriae expression in all Salmonella strains. Figure 4 demonstrated the base two logarithms of a geometric average titer for each sample. The ΔfimY and ΔfimY (pACYC184) strains did not produce type 1 fimbriae and no titer was detected. The titer of the ΔfimY (fimY) was 4.64, which was higher than that of the wild type (2.76). The titer of other ΔfimY transformants were as follows: R11A and DSG were the same as ΔfimY (fimY), R7/11A and MUT were nearly 4, and the lowest was R7A (3.17). Compared with the ΔfimY (fimY), whose titer was considered as a baseline, the decrease in the titer of R7A was over 50%, while R7/11A and MUT were close to 36%.

Figure 4

Detection of type 1 fimbriae by microplate yeast agglutination test. The numbers indicate the geometric average of the titer end point (N = 3) of different Salmonella strains

The DSG strain exhibited about 0.8‐fold of the fimZ expression as compared to the control strain ΔfimY (fimY), while the other strains R7A, R11A, R7/11A, and MUT produced less fimZ expression. Similar profile was observed on fimA expression, the DSG strain exhibited relatively similar fimA expression as the ΔfimY (fimY) strain and the others exhibited the fimA level significantly lower than that of the ΔfimY (fimY) strain (Figure 5).

Figure 5

Quantitative real‐time polymerase chain reaction analysis of fimA and fimZ of different S. Typhimurium strains. The expression of (a) fimZ and (b) fimA was calculated using the ΔΔCt method and designed as the fold change compared with the geometric mean expression level of the ΔfimY (fimY) strain in triplicate. A p value < 0.05 was considered to represent a significant difference. *p < 0.05; **p < 0.01; ***p < 0.001

If FimY serves as a PilZ domain‐like protein in S. Typhimurium, the secondary structure prediction program would reveal similarities between the configurations of FimY and other well‐known PilZ domain proteins, such as MrkH of K. pneumoniae, DgrA of Caulobacter crescentus, and VCA0042 of Vibrio cholerae (Heidelberg et al., 2000; Ryjenkov et al., 2006; Wilksch et al., 2011). Figure 6 shows that the secondary structure pattern of FimY did not match that of the other well‐documented PilZ domain proteins—the RRxxxR at the N‐terminal followed by 6–7 β‐strands forming a β‐barrel and a C‐terminal α‐helix (Amikam & Galperin, 2006). However, the N‐terminal of FimY only contained an α‐helix.

Figure 6

Secondary structure prediction. Protein size of 100 amino acids beginning from 6RRERH11R of FimY in Salmonella Typhimurium (accession No.: AAL19504), 106RRRDP111R of MrkH in Klebsiella pneumoniae (accession No.: AEO27488), 11RRAHP16R of DgrA in Caulobacter crescentus (accession No.: AE005673.1), and 134QLRKEP140R of VCA0042 in Vibrio cholerae (accession No.: NP_232443) were analyzed using PHD to predict their secondary structures; red: β‐strand; pink: random coli; blue: α‐helix.

DISCUSSION

S. Typhimurium possesses three well‐defined processes which are regulated by c‐di‐GMP, they are flagellar‐based motility, curli fimbriae formation, and cellulose production (Kader, Simm, Gerstel, Morr, & Romling, 2006; Ryjenkov et al., 2006; Solano et al., 2002; Zogaj, Nimtz, Rohde, Bokranz, & Romling, 2001). Previous studies have demonstrated that c‐di‐GMP binding depends on residues in the RRxxxR and DxSxxG sequence motif conserved in the PilZ domains (Christen et al., 2007; Merighi, Lee, Hyodo, Hayakawa, & Lory, 2007). Like most of the Enterobacteriaceae members, S. typhimurium has two PilZ domain proteins, YcgR and BcsA, which control motility and cellulose production, respectively (Pultz et al., 2012). Within the elements of the fim gene cluster of S. Typhimurium only stm0551 had been described to involve in c‐di‐GMP metabolism, whose product is a phosphodiesterase to degrade c‐di‐GMP (Wang et al., 2012). No PilZ domain protein had been reported previously. In this study, we found that the amino acid of FimY does possess a RRxxxR motif and another DxSxxxxxG sequence motif; however, the function of these motifs has not been explored. Therefore, it is interesting to investigate the function of these motifs in terms of type 1 fimbrial production.

Since FimY is a positive regulator for type 1 fimbrial production, a fimY‐deleted strain ΔfimY did not possess fimbrial appendages. Transforming an intact fimY allele cloned in pACYC184 into the ΔfimY restored the fimbriate positive phenotype. However, transformation of the plasmids possessing the fimY alleles with amino acid changes in different positions enabled ΔfimY to produce different levels of fimbrial structures. Comparatively, the amount of fimbriae found in R7A, R7/11A, and MUT strains was less than those of the other transformants. Since the R11A strain did not decrease the amount of fimbriae, it is highly possible that the change from arginine to alanine at position 7, but not 11, in FimY had an impact on type 1 fimbrial expression. In another study, DgrA of Caulobacter crescentus controls flagellar motor function and it functions as a c‐di‐GMP–binding protein; mutation of R11A/R12A in the 11RRxxxR16 motif of DgrA abolished the binding ability of c‐di‐GMP (Christen et al., 2007). It was postulated that the side chains of arginine participate in hydrogen bond or in electrostatic interactions with c‐di‐GMP, as the similar to that in the allosteric binding site of the diguanylate cyclases such as PleD and DgcA (Chan et al., 2004; Christen et al., 2006). In addition, the positive charged head groups of arginine are required for transient binding to the phosphate groups of c‐di‐GMP (Shoemaker, Portman, & Wolynes, 2000). The R7A and R7/11A strains decreased the production of fimbriae and R7A exhibited a decreased microplate yeast agglutination titer; these findings did somehow underscore the important role of R7 in FimY. As to the yeast agglutination test, FimH protein incorporated on the fimbrial shaft mediates the agglutination of yeast cell. The titer end point of the microplate agglutination test should correlate with the number of type 1 fimbriae on S. typhimurium. The wild‐type strain showed an agglutination titer of 2.75 while ΔfimY and ΔfimY (pACYC184) strains were both not fimbriate, hence no agglutination was detected. The reason that ΔfimY (fimY) strain exhibited a higher agglutination titer than that of the wild‐type strain (4.64 vs. 2.75) may be due to the copy number effect of the pACYC184. In the regulatory network of type 1 fimbriae in S. typhimurium, fimY may act upstream of fimZ; that is to say, fimY activates fimZ, and in turn fimZ activates fimA, resulting in the production of fimbriae (Wang et al., 2014; Zeiner, 2012). Many fimY transcripts in ΔfimY (fimY) strain should be responsible for the mass production of fimA through the fimY→fimZ→fimA pathway, leading to a high agglutination titer.

The DSG strain contains an intact 6RRERHR11 motif but with aspartic acid, serine, and glycine at 83, 85, and 91 position, respectively, substituted with alanine at the 83DI85SLWIEK91G motif. A previous study indicated that the aspartic acid and glycine residues of DxSxxG motif are required for c‐di‐GMP binding (Christen et al., 2007). The R11A and DSG strains both had the same agglutination titer as the ΔfimY (fimY) strain and these two strains also produced similar number of fimbriae as the ΔfimY (fimY) strain. Besides, the fimZ and fimA expression of DSG strain were close to that of the ΔfimY (fimY) strain. It is possible that the 83DI85SLWIEK91G motif did not involve the overall function of FimY to activate fimZ, and consequently fimA. Since there are three extra amino acid residues in the 83DI85SLWIEK91G motif of FimY than DxSxxG motif of other PilZ domains, we are not sure if these extra residues cause this motif a nonfunctional vestige. Therefore, it seems that the intact 6RRERHR11 motif may play a crucial role for the ultimate function of FimY.

There was a discrepancy between the number of fimbriae, yeast agglutination titer, and gene expression of fimZ and fimA in the R11A strain. This strain exhibited low fimZ and fimA expression in real‐time RT‐PCR, however, this did not correlate with the number of fimbriae and yeast agglutination titer, which were similar to the ΔfimY (fimY) strain. So far we have no reason to explain this phenomenon.

PHD revealed that N‐terminal of FimY possessed only an α‐helix, which is not likely to form a PilZ domain; however, the same analysis for MrkH, DgrA, and VCA0042 predicted six to seven β‐strands and a C‐terminal helix, consistent with a PilZ domain fold (Benach et al., 2007). Although FimY possesses 6RRxxx11R and 83Dx85Sxxxxx91G motifs, evidence to designate FimY as a PilZ domain protein was unavailable. FimY and PilZ domain proteins are not homologs. Nevertheless, the arginine at position 7 of the 6RRxxx11R motif has a crucial role in activating fimZ and consequently fimA, resulting in an expression of type 1 fimbriae in S. typhimurium.

CONFLICT OF INTERESTS

The authors declare that they have no conflict of interest.

AUTHOR CONTRIBUTIONS

Nan‐Ling Kuan drafted the manuscript and participated all the experiments. Kuang‐Sheng Yeh conceived this study and helped to draft the manuscript.

ETHICS STATEMENT

Not required.