A Constitutively Mannose-Sensitive Agglutinating Salmonella enterica subsp. enterica Serovar Typhimurium Strain, Carrying a Transposon in the Fimbrial Usher Gene stbC, Exhibits Multidrug Resistance and Flagellated Phenotypes

Static broth culture favors Salmonella enterica subsp. enterica serovar Typhimurium to produce type 1 fimbriae, while solid agar inhibits its expression. A transposon inserted in stbC, which would encode an usher for Stb fimbriae of a non-flagellar Salmonella enterica subsp. enterica serovar Typhimurium LB5010 strain, conferred it to agglutinate yeast cells on both cultures. RT-PCR revealed that the expression of the fimbrial subunit gene fimA, and fimZ, a regulatory gene of fimA, were both increased in the stbC mutant when grown on LB agar; fimW, a repressor gene of fimA, exhibited lower expression. Flagella were observed in the stbC mutant and this phenotype was correlated with the motile phenotype. Microarray data and RT-PCR indicated that the expression of three genes, motA, motB, and cheM, was enhanced in the stbC mutant. The stbC mutant was resistant to several antibiotics, consistent with the finding that expression of yhcQ and ramA was enhanced. A complementation test revealed that transforming a recombinant plasmid possessing the stbC restored the mannose-sensitive agglutination phenotype to the stbC mutant much as that in the parental Salmonella enterica subsp. enterica serovar Typhimurium LB5010 strain, indicating the possibility of an interplay of different fimbrial systems in coordinating their expression.


Introduction
Salmonella enterica subsp. enterica contains more than 2,300 serovars; one of these, Typhimurium, is an important cause of gastroenteritis [1]. The ability to adhere to the host epithelial cell is considered a prerequisite step for infection. Fimbriae, proteinaceous hair-like appendages on the outer membrane of bacteria, have been implicated in such adherence. Many types of fimbriae have been described for Salmonella enterica subsp. enterica serovar Typhimurium, among which type 1 fimbriae is the most common; therefore, type 1 fimbriae are also referred to as common fimbriae [2]. Type 1 fimbriae adhere to a variety of cells, including erythrocytes, leukocytes, intestinal cells, and even plant root hairs. More than 80% of Salmonella species isolates produce type 1 fimbriae, and this fimbrial type may play some role in the life cycle of bacteria [3].
Old et al. described the phenotypic variation of the expression of type 1 fimbriae in Salmonella enterica subsp. enterica serovar Typhimurium [4,5]. In brief, strongly type 1 fimbriate-phase bacterial cells were obtained following successive passage every 48 hr in static broth culture medium, while nonfimbriated-phase bacteria were found when cultured on solid media [4,5]. Subsequent investigations have 2 The Scientific World Journal indicated that phenotypic expression of type 1 fimbriae in Salmonella enterica subsp. enterica serovar Typhimurium involves the interaction of several genes in the fim gene cluster [6][7][8][9]. To explore whether there is another genetic determinant outside the fim genes that would also participate in regulation of type 1 fimbriae in Salmonella enterica subsp. enterica serovar Typhimurium, Chuang et al. constructed an insertional library to screen for mutant strains that showed different type 1 fimbrial phenotypes than the parental strain [10]. Yeast agglutination test and southern blot analysis were used to screen for those mutants and validated that the transpositional event occurred only once. A group of genes classified as having fimbrial biosynthesis and regulation were involved in the expression of type 1 fimbriae in response to different culture conditions [10]. One mutant stbC strain agglutinated yeast obtained either from static LB broth or on solid LB agar medium, and the agglutination was mannosesensitive [10]. The gene stbC encoded the usher protein for Stb fimbriae in Salmonella enterica subsp. enterica serovar Typhimurium [11]. The usher protein helps anchor the developing fimbrial subunit structure to the outer membrane [12]. Cross-talk between different fimbrial systems in one strain has been documented [13][14][15]. Therefore, the interaction of fim and stb fimbrial systems is an intriguing topic. This paper describes the interesting characteristics of Salmonella enterica subsp. enterica serovar Typhimurium stbC mutant that we discovered.

Bacterial Strains and Culture
Media. The Salmonella enterica subsp. enterica serovar Typhimurium strain used in this study was the LT2 derivative strain LB5010 [16]. The stbC mutant strain was obtained from the Salmonella enterica subsp. enterica serovar Typhimurium transposon library collection in our laboratory [10]. This strain has a transposon inserted in the position of stbC, encoding the usher protein of StbA fimbriae. Salmonella cells were cultured in Luria-Bertani (LB) broth (Difco/Becton Dickinson, Franklin Lakes, NJ) or on LB agar. Mueller-Hinton agar (Difco/Becton Dickinson) was used when performing the antimicrobial susceptibility test, and modified semisolid Rappaport-Vassiliadis (MSRV) (Difco/Becton Dickinson) was used to detect the motility of Salmonella strains.

Yeast Agglutination Test.
The LB agar plates were incubated at 37 • C for 18 h, while the broth preparations were incubated statically at 37 • C for 48 h. Bacterial cells from the solid agar were collected by a sterile loop and resuspended in 100 µL of 1 × phosphate-buffered saline (PBS). Cells in the broth medium were collected by centrifugation, and the pellet was resuspended in 100 µL of 1 × PBS. Subsequently, 30 µL of a 3% (vol/vol) suspension of Candida albicans in PBS and an equal amount of bacterial cells to be tested were mixed together on a glass slide [4]. Visible agglutination after gentle agitation indicated a positive reaction for the presence of type 1 fimbriae. Any bacterial suspension that produced type 1 fimbriae was further mixed with C. albicans, along with 3% (wt/vol) of a D-mannose solution (Sigma Chemical Company, St. Louis, MO). The mannose-sensitive agglutination conferred by type 1 fimbriae was inhibited in the presence of mannose.

Electron Microscopy.
Bacterial strains were grown in static broth or on solid agar and resuspended in 1 × PBS. The bacterial cells were then negatively stained with 2% phosphotungstic acid and observed with a Hitachi H-600 transmission electron microscope (Hitachi Ltd., Tokyo, Japan).

Motility
Test. Salmonella enterica subsp. enterica serovar Typhimurium LB5010 and the stbC mutant strains were grown in static broth for 48 h and on solid agar for 18 h. Cells were resuspended in 1 × PBS and adjusted to the same turbidity. A drop of cell suspension was then spotted on the modified semisolid Rappaport-Vassiliadis (MSRV) agar medium and incubated at 42 • C for 16 h.
2.6. RNA Purification. Bacterial cells were harvested in TRIzole Reagent (Invitrogen, Carlsbad, CA) and disrupted using the MagNA Lyser System (Roche Diagnostics, Mannheim, Germany), with ceramic bead shaking at 5,000 rpm for 15 sec. After phenol-chloroform extraction, the aqueous layer was applied to an RNeasy column (Qiagen, Valencia, CA) for RNA purification according to the protocol provided by the manufacturer. The RNA was quantified by using a ND-1000 spectrophotometer (NanoDrop Technology, Wilmington, DE) and then examined with a Bioanalyzer 2100 (Agilent Technology, Palo Alto, CA) with an RNA 6000 Nano LabChip kit (Agilent). To enhance the sensitivity of microarray signal, the purified total RNA was subjected to a ribosomal RNA removal procedure by MICROB Express Bacterial mRNA Purification Kit (Ambion, Applied Biosystems, Foster City, CA).

Custom Array Design.
Probes were designed by eArray of Agilent Technologies. In the design process, 4,718 Salmonella enterica subsp. enterica serovar Typhimurium LT2 transcripts were uploaded to Agilent eArray and designed by Tm matching methodology. Target sequences that were duplicated were removed. Probes with Tm at around 80 • C, optimal base content, and low cross-hybridization were The Scientific World Journal 3 selected. The resulting 4,646 probes were then generated. The custom microarray was manufactured in 4 × 44 k format by in situ synthesis of oligonucleotide probes. Each array consisted of 4,646 Salmonella enterica subsp. enterica serovar Typhimurium LT2-specific probes and printed in 9 replicates.
2.8. Microarray Experiment. One microgram of enriched mRNA was reverse-transcribed to cDNA with a CyScribe 1st-strand cDNA labeling kit (GE Healthcare, Buckinghamshire, UK) and labeled with Cy3-CTP or Cy5-CTP (CyDye, PerkinElmer, Waltham, MA). Correspondingly labeled cDNA was then pooled and hybridized to microarrays at 60 • C for 17 h. After washing and drying by nitrogen gun blowing, microarrays were scanned with an Agilent microarray scanner at 535 nm for Cy3-CTP and 625 nm for Cy5-CTP. Scanned images were then analyzed by Feature Extraction 9.5.3 software (Agilent). Image analysis and normalization software was used to quantify signal and background intensity for each feature and substantially normalized the data by rank-consistency-filtering LOWESS method.

Reverse Transcription Polymerase Chain Reaction (RT-PCR) Analysis.
The bacteria were first stabilized by adding RNAProtect Bacteria Reagent (Qiagen) and total RNA was isolated by using an RNeasy Mini Kit (Qiagen) and RNasefree DNase I (1 unit/1 µg RNA) (Promega, Madison, WI), according to the protocol provided by the manufacturer. RT-PCR was performed by a SuperScript III One-Step RT-PCR System (Invitrogen). RNA was denatured at 58 • C for 5 min and followed by cDNA synthesis at 42 • C for 30 min. The reaction was stopped by heating the bacteria at 94 • C for 2 min. The following PCR programming was used: 35 cycles of denaturing at 94 • C for 30 sec, annealing at 54 • C for 30 sec, and extension at 72 • C for 30 sec. An additional extension was performed at 72 • C for 10 min. Primers used in the present study are listed in Table 1.

Complementation Test.
The primer set, stbC-F and stbC-R, was used to amplify the stbC coding sequence from the genomic DNA of Salmonella enterica subsp. enterica serovar Typhimurium LB 5010 with the Epicentre FailSafe PCR PreMix Selection kit (Epicentre, Madison, WI). The resulting DNA fragment was cleaved with BamHI and ligated into the pACYC 184 vector. HindIII and SalI were then used to cleave a 2.9 kb DNA fragment possessing the stbC coding sequence from the aforementioned recombinant plasmid and ligated into the pBBR1MCS-5 vector that carries a gentamycin-resistant gene [18]. The resulting recombinant plasmid was transformed into the stbC mutant strain, and transformants were selected on gentamycin-containing (100 µg/mL) LB agar medium. A transformant labeled stbC (pStbC) was selected for further study. For the control, the pBBR1MCS-5 vector was transformed into the stbC mutant, and a resulting transformant labeled stbC (vector) was selected.

Yeast Agglutination Test.
A stbC mutant prepared on both static LB broth and LB agar medium both exhibited agglutination when mixed with yeast cells on a glass slide. The addition of mannose inhibited agglutination, indicating that agglutination was mannose-sensitive. Salmonella enterica subsp. enterica serovar Typhimurium LB5010 from static broth agglutinated yeast cells, while that from agar medium did not. The stbC (pStbC) harboring the plasmid that contains the coding sequence of stbC exhibited the same agglutination pattern as the parental strain, while stbC (vector), containing the vector alone, prepared on both agar and broth medium, agglutinated yeast cells as did the stbC mutant strain. Table 2 compares the yeast agglutination capabilities conferred by these strains.

Electron
Microscopy. Salmonella enterica subsp. enterica serovar Typhimurium LB5010 prepared in static LB broth culture showed fimbrial appendages on the outside of the cell (Figure 1(a)). Besides fimbriae, additional flagellalike structures were present on the stbC mutant strain cultured in static broth (Figure 1(b)). On the contrary, Salmonella enterica subsp. enterica serovar Typhimurium LB5010 grown on agar medium did not produce type1 fimbriae (Figure 2(a)). The stbC mutant prepared from agar medium did not produce fimbriae; either, however, flagellalike structures were still observed around the stbC mutant prepared from agar medium (Figure 2(b)).

Motility Test.
The semisolid characteristic of MSRV allows mobility to be detected as "halos" of growth around the point of inoculation. Salmonella enterica subsp. enterica serovar Typhimurium LB5010 did not exhibit a "halo" effect on the MSRV medium, and the medium remained blue around the inoculated drop (Figure 3(a)). In contrast, a gray-white zone was observed extending from the inoculated drop of the stbC mutant, prepared either from static LB broth or LB agar (Figure 3(b)). The stbC was a motile Salmonella strain. We also used polyvalent Salmonella H antiserum (Difco/Becton Dickinson) to confirm that the flagellar antigen was present on the stbC mutant but not on the Salmonella enterica subsp. enterica serovar Typhimurium LB5010 by slide agglutination test.

Antimicrobial Susceptibility Test.
Since we had difficulty complementing the stbC with a recombinant plasmid containing the coding sequence of stbC, we tested whether stbC mutant exhibited resistance to the antibiotic marker carried in the cloning vector. We found that stbC was resistant to ampicillin and tetracycline carried in TA cloning and pACYC 184 vectors, respectively. Therefore, a battery of antibiotics was tested on the stbC mutant and the parental strain LB5010. Both strains were resistant to penicillin and streptomycin. In addition, the stbC mutant also exhibited resistance to ampicillin, cephalothin, tetracycline, doxycycline, chloramphenicol, florfenicol, and ciprofloxacin.

4
The Scientific World Journal Table 1: Oligonucleotide primers used in the present study.

Primer
Sequence The underlined sequence denotes the BamHI restriction site The underlined sequence denotes the BamHI restriction site

Microarray Analysis.
Since the stbC mutant strain prepared from solid agar produced different results on yeast agglutination testing and exhibited multidrug resistant characteristics, it is tempting to identify the genes that would express differently between the stbC mutant strain and its parental strain Salmonella enterica subsp. enterica serovar Typhimurium LB5010 when grown on solid LB agar. Total RNA was isolated from both strains cultured on solid LB agar and analyzed by hybridization to a Salmonella enterica subsp. enterica serovar Typhimurium LT2 DNA microarray. The median hybridization results of 9 arrays showed that about 50 genes from the stbC mutant grown on agar were upregulated more than 8-fold (Table 3). Analysis of the microarray data revealed that these genes can be classified according to their functions, including fimbrial structure, motility, drug resistance, gene regulation, transportation, outer membrane porin structure, prophage, ribosomal protein, inner membrane protein, periplasmic protein, and enzymes of different functions. Twenty-two genes are enzymes with a variety of functions (44%), and these comprised the major part of the genes upregulated in the stbC mutant strain grown on solid agar. The detail results of the microarray assay in the present study can be accessed at GEO accession number GSE34685.  represent the ratio of stbC/LB5010.) This was correlated with the microarray data. Expression of type 1 fimbriae involved several Fim regulatory proteins and other gene products outside the fim gene cluster. FimZ and FimY are positive regulators for type 1 fimbriae, while FimW is a repressor for fimbrial expression [6][7][8][9]. We also used RT-PCR to investigate the transcription level of the major fimbrial subunit gene fimA and those of three regulatory genes fimZ, fimY, and fimW. Total RNAs from the parental strain Salmonella enterica subsp. enterica serovar Typhimurium LB5010, stbC mutant, stbC (pStbC), and stbC (vector) were prepared and analyzed for fimA, fimZ, fimY, fimW mRNA, and 16S ribosomal (r)RNA expression by RT-PCR. Figure 4 shows the RT-PCR results. Expression levels of fim genes of LB5010 obtained on LB agar were used as the reference. When LB 5010 was cultured in static LB broth, fimA, fimZ, fimY, and fimW had higher levels of expression than when cultured on LB agar. The stbC strain demonstrated the same tendency, except for fimW. The expression of fimA obtained from the stbC mutant strain was The Scientific World Journal approximately 4.6-fold higher than that of the LB5010 strain when both strains were grown on LB agar. The expression of fimZ was also higher than that of LB5010 when grown on LB agar. As a control, 16S rRNA was consistently expressed in all of the strains tested. Transforming the recombinant plasmid harboring the stbC coding sequence to the stbC strain conferred on it the ability to express a level of fimA similar to that of LB5010. Gene expression of fimZ, fimY, and fimW was higher in stbC (StbC) grown in broth than on agar medium. The stbC (vector), possessing the cloning vector alone, exhibited similar gene expression patterns as the stbC strain.

Discussions
Salmonella enterica subsp. enterica serovar Typhimurium LB5010, a LT2 strain derivative, exhibits mannose-sensitive phenotype when grown in static broth culture but not on the solid agar medium culture. This strain is nonflagellated and not multidrug resistant. However, the present study revealed that a transposon inserted in the stbC gene of LB5010 strain conferred it to become a constitutively mannose-sensitive agglutinating, multidrug resistance, and flagellated strain. Whole-genome sequence analysis of Salmonella enterica subsp. enterica serovar Typhimurium LT2 strain has revealed the presence of 13 gene clusters that contain open reading frames (ORFs) to encode putative fimbrial subunit and fimbrial-accessory proteins [11]. Laboratory-grown cultures of Salmonella enterica subsp. enterica serovar Typhimurium commonly produce only type 1 fimbriae and thin aggregative fimbriae (curli fimbriae) [19,20]. Although the laboratory culture condition could not induce bacteria to produce Stb fimbriae, such fimbriae did express in vivo by flow cytometry using bovine ligated ileal loops [21]. The fact that mice infected with Salmonella enterica subsp. enterica serovar Typhimurium seroconvert to StbA, a major fimbrial subunit of Stb fimbriae, also provides evidence for in vivo Std fimbrial expression [21]. The amino acid sequence of the StbC subunit shares similarities to those of the usher proteins of the chaperone/usher assembly pathway. Usher protein is an integral outer membrane protein that interacts with the chaperone/fimbrial subunit complex, facilitating the release of the fimbrial subunits and their secretion through the usher channel [12,22]. This finding leads to an interesting question: Does cross-talk occur between different fimbrial systems in one bacterial strain? For example, expression of pyelonephritis-associated pili [23] represses type 1 fimbrial expression in the same E. coli strain [14]. A transposon inserted in the stbC gene of Salmonella enterica subsp. enterica serovar Typhimurium conferred it to exhibit mannosesensitive agglutination constitutively when cultured in static broth or agar medium. Nonetheless the stbC mutant collected from agar medium exhibited less agglutinating power than those cultured in static broth medium. Titration of the original stbC mutant/agar suspension to 8-fold abolished the agglutination power. One reason could be due to the decreased number of fimbriae present on the stbC mutant.
Previously we have encountered difficulties when attempting to transform the recombinant plasmid containing the coding sequence of stbC to the stbC mutant to perform a complementation test. The results of our study revealed that one major reason for this could be the multidrug-resistant (MDR) phenotype of the stbC strain. The antimicrobial susceptibility test indicated that the stbC mutant resisted a battery of antibiotics. Since these antimicrobial agents have different mechanisms of antibiotic action, it is speculated that the transposon inserted in stbC had changed gene(s) expression that is associated with MDR. Concurrently, the microarray result indicated that ramA and yhcQ were enhanced in the stbC mutant compared to its parental strain LB5010. Our RT-PCR analysis also confirmed this result. RamA is a member of the AraC-XylS family of transcriptional regulators that controls one of the efflux pump genes acrB [24], and the product of yhcQ is a putative MDR pump [25]. Deletion of yhcQ in E. coli reduced the penicillin G resistance [26]. The MDR phenotype of the stbC mutant could be reasonably explained, at least partially, by the increased expression of ramA and yhcQ.
Salmonella enterica subsp. enterica serovar Typhimurium LB5010 was derived from LT2 strain, which is routinely used in laboratories for molecular genetics and as a representative of the wild type of Salmonella enterica subsp. enterica serovar Typhimurium [16]. This strain has defects in its flagellar synthesis gene, making it suitable for observing fimbrial appendages without flagella background [27]. Accordingly, Salmonella enterica subsp. enterica serovar Typhimurium LB5010 did not demonstrate motility on MSRV, a semisolid agar medium for isolating motile Salmonella. There were no flagellar structures on Salmonella enterica subsp. enterica serovar Typhimurium LB5010, prepared either from static broth or on agar medium, while the stbC mutant exhibited flagellar structures under electron microscopy, which was unexpected. When tested in MSRV, the stbC mutant exhibited extended growth from the inoculated center on MSRV, indicating the motile capability of the tested strain. A transposon inserted in stbC could cause a suppression effect that alleviated the original phenotype exhibited by LB5010. However, some flagella of the stbC mutant observed under electron microscopy were not anchored on the bacterial cells. This could be due to the method used to prepare the samples or to the fact that some flagella were actually secreted and without function. Interestingly, microarray data and RT-PCR also indicated that the expression of three genes, motA, motB, and cheM, was enhanced in the stbC mutant compared to the LB5010. MotA and MotB are integral to the cell membrane and are required for motor rotation [28]. The product of the cheM gene is a methyl-accepting chemotaxis protein II, a sensory transducer. All three genes are components of a complicated chemotaxis/flagella mechanism [29]. These results were correlated with the motile characteristic of the stbC mutant.
One dilemma of the stbC mutant strain we observed was the absence of type 1 fimbriae structures under electron microscopy. Agglutination of yeast cells did appear, and it was mannose-sensitive, indicating that the agglutination was mediated by type 1 fimbriae whose receptor contains mannose residue. RT-PCR also indicated that expression of the fimA was higher in the stbC mutant than that in the parental LB5010 when these bacteria were grown on LB agar. This finding was correlated with the fact that stbC had higher fimZ expression and lower fimW expression than LB5010 when grown on agar, which correlated with the previous findings. These results demonstrate that the fimZ gene encodes a positive regulator for fimA, while FimW is a repressor for fimA [6,7]. FimA protein may secrete and attach on the outer membrane but did not assemble into an intact fimbrial appendage. Use of immune-gold electron microscopy would test this hypothesis. Another possibility is that stbC mutant grown on LB agar induced a mannose-sensitive adhesion protein that has not previously been characterized and that is present on the outer membrane. The expression of the fimbrial major subunit gene of thin aggregative fimbriae csgA was increased in stbC mutant grown on LB agar. However, this laboratory condition is not suitable for Salmonella enterica subsp. enterica serovar Typhimurium to produce thin aggregative fimbriae. Collinson et al. demonstrated that static colonization factor antigen (CFA) broth at 30 • C could induce Salmonella enterica subsp. enterica serovar Enteritidis to produce thin aggregative fimbriae [30].
To better understand whether the stbC gene really does account for the mannose-sensitive agglutination, flagellar formation, and MDR characteristics of this interesting strain, a complementation test was performed. When a recombinant plasmid carrying the stbC coding sequence and a gentamycin-resistant cassette were constructed and transformed into the stbC strain, mannose-sensitive agglutination phenotypes and gene expression levels of the type 1 fimbrial subunit gene, fimA, and 3 fimbrial regulatory genes, fimZ, fimY, and fimW, exhibited the same tendency toward the LB5010 parental strain by the stbC strain. This evidence suggests that the stbC gene product, an usher of Stb fimbriae, may play some role in the regulatory network The Scientific World Journal 9 of type 1 fimbrial expression. The interaction of different fimbrial systems in a single strain was reported [13][14][15], but in the present study, we could not identify 1 or more specific fim genes with which StbC directly interacts, nor could we determine if other non-fim genes were involved in connecting these 2 different fimbrial system. However, these interesting topics warrant further investigation. Proteinprotein interaction using yeast two hybrid system is being investigated in our laboratory. Complementation tests did not restore the MDR and flagellar formation characteristics of the LB5010 parental to stbC (pStbC). Whether this was due to polar effect or other reasons was under exploration.