OUP user menu

Characterization of the chemotaxis fliY and cheA genes in Bacillus cereus

Francesco Celandroni , Emilia Ghelardi , Manuela Pastore , Antonella Lupetti , Anne-Brit Kolstø , Sonia Senesi
DOI: http://dx.doi.org/10.1111/j.1574-6968.2000.tb09294.x 247-253 First published online: 1 September 2000


This paper describes the first identification of chemotaxis genes in Bacillus cereus. We sequenced and studied the genomic organization and the expression of the cheA and fliY genes in two different B. cereus strains, ATCC 14579 and ATCC 10987. While cheA encodes a highly conserved protein acting as the main regulator of the chemotactic response in flagellated eubacteria, fliY, which has been previously described only in B. subtilis, is one of the three genes encoding proteins of the flagellar switch complex. Although the sequences and relative position of cheA and fliY were found to be identical in the two B. cereus strains analyzed, the restriction fragment containing both genes was located differently on the physical maps of B. cereus ATCC 14579 and ATCC 10987. Evidence is shown that the genomic organization and the expression of fliY and cheA in B. cereus differ significantly from that described for B. subtilis, which is considered a model microorganism for chemotaxis in Gram-positive bacteria.

  • Bacillus cereus
  • Motility
  • Chemotaxis
  • Gene organization
  • Physical map
  • Gene expression

1 Introduction

Bacillus cereus is a motile, Gram-positive, rod-shaped and spore-bearing bacterium, which is frequently responsible for food poisoning [1,2] and has recently been recognized as an emerging opportunistic pathogen able to cause local and systemic infections in humans [3]. B. cereus is regarded as the reference species of a group of closely related bacteria encompassing also B. thuringiensis, B. mycoides, and B. anthracis, which all together are in fact related as the B. cereus group [4,5]. While these species have been allocated to the B. subtilis group of bacilli by 16S and 23S rRNA classification [6], increasing evidence is being produced suggesting that the B. cereus genome organization differs widely from that of B. subtilis[7]. Moreover, in contrast to the genetic stability described for B. subtilis, a highly variable gene organization has been reported for B. cereus in certain genomic regions, perhaps more prone to rearrangements [810]. For these reasons, the B. cereus genome is now intensively studied as a model for shedding light onto gene organization within the B. cereus group, and for establishing genetic relatedness with the entirely sequenced B. subtilis genome.

Chemotactic behavior is a relevant physiological trait shared by flagellated eubacteria. Nothing is known about B. cereus chemotaxis while much has been learned about genes and proteins involved in the regulation of B. subtilis chemotactic response. More than 30 genes, organized in a major che/fla operon [11], have been identified in this microorganism and the molecular properties of several proteins regulating B. subtilis chemotaxis have been characterized [11]. Many reports demonstrate notable homologies in most molecular components involved in chemotactic behavior of B. subtilis in comparison with those described for enteric bacteria, in particular for Escherichia coli and Salmonella typhimurium[11]. Among the conserved chemotaxis genes, cheA has been shown to play a key role since it encodes a histidine kinase that acts as the main regulator of the signal transduction mechanism involved in chemotactic response of both B. subtilis and enteric bacteria [12]. Several chemotaxis genes, however, have been described only in B. subtilis[13,14]. One of them, fliY, encodes an essential component of the motor-switch complex (C-ring), which controls the direction of flagella rotation [13]. The C-ring comprises FliG, FliM and FliY in B. subtilis while it is constituted by FliG, FliM and FliN in E. coli and S. typhimurium[13]. Interestingly, although the primary sequence of FliY differs significantly from that of FliN, it was proven that FliY is able to complement the motility defect exhibited by a S. typhimurium FliN mutant [13].

Following the identification of a short B. cereus sequence (EMBL accession number Y08031), characterized by a high degree of homology with a portion of B. subtilis fliY, the present study was addressed to ascertain the presence of such a gene in the B. cereus genome, as well as to analyze the organization of B. cereus chemotaxis genes and compare it to that of B. subtilis. For this purpose, a 5-kb genomic region, comprising fliY, was analyzed in two B. cereus strains. Herein we describe the sequence, organization, genomic localization, and expression of two chemotaxis genes, fliY and cheA, in two B. cereus strains.

2 Materials and methods

2.1 Bacterial strains and growth conditions

The strains ATCC 14579 and ATCC 10987 of B. cereus, obtained from the American Type Culture Collection (Rockville, MD), were grown in Luria–Bertani (LB) medium. For cloning experiments, E. coli strain Xl1-Blue (Stratagene Ltd., UK) was grown in LB medium supplemented with ampicillin (100 μg ml−1) when required.

2.2 Construction of B. cereus genomic libraries

Genomic DNA was extracted from B. cereus and purified as described by Wagner et al. [15] with some minor modifications. Cells were harvested from liquid cultures at the late exponential growth phase, washed with TES buffer (5 mM EDTA, 50 mM NaCl, 30 mM Tris–HCl, pH 8.0), and incubated at 37°C for 40 min in 12 ml of TES containing 20 mg lysozyme and 5 mg RNase. After the addition of 8% Triton X-100 (2.1 ml) and of 10 mg ml−1 proteinase K (0.6 ml), samples were incubated at 37°C for 1 h, 2.5 ml CTAB/NaCl solution (10% CTAB; 0.7 M NaCl) was added, and the samples were incubated for an additional 10 min at 65°C. DNA was purified with chloroform–isoamyl alcohol (24:1), extracted by phenol, and precipitated with isopropanol. Purified DNA was digested with HindIII. The resulting DNA fragments were ligated into pUC19 (New England Biolabs Inc., USA). E. coli was transformed with the ligation mixture and cells were plated onto nylon membranes lying on LB-ampicillin agar plates. After overnight incubation at 37°C, Ampr colonies were transferred onto nylon membranes by replica plating in duplicate [16] and the filters were used for subsequent colony hybridization.

2.3 Screening of B. cereus libraries with a fliY-specific probe

Genomic DNA of B. cereus ATCC 14579 was used as a template for PCR amplification of a 567-bp fragment (p-567) to produce a fliY-specific probe. PCR mixtures contained 1 ng of strain ATCC 14579 genomic DNA, 1 μmol of each primer, 2.5 U of DynaZyme II DNA polymerase (Finnzymes Oy, Finland), in the supplied buffer, to a final volume of 50 μl. Primers D-4 (5′-GCTCCTCATGTAGAGTTAGTGG-3′) and F-2 (5′-CTTCTACAATCTCGGCTTCC-3′) were deduced from a DNA sequence of B. cereus NCIB 8122 (EMBL accession number Y08031). Reactions were carried out for 30 cycles at 94°C for 1 min, 55°C for 1 min and 72°C for 1 min. Aliquots containing 100 ng of the amplified p-567 were end-labeled using 2 μl [γ-32P]dATP (5000 Ci (18.5×1013 Bq) mmol−1) and 10 U T4 polynucleotide kinase (New England Biolabs Inc., USA) in the supplied buffer at 37°C for 30 min. The labeled p-567 was used as a probe for screening B. cereus libraries in colony hybridization experiments performed according to standard methods [16].

2.4 DNA techniques and analysis

Plasmid DNA was extracted from E. coli using Qiagen plasmid kits (Qiagen Inc., Valencia, CA, USA). Restriction enzymes, T4 DNA ligase, and calf intestinal phosphatase were used as recommended by the manufacturers. DNA restriction fragments were purified from agarose gels using the Qiaex II gel extraction kit (Qiagen Inc.). Nucleotide sequencing was performed by the dideoxy-chain termination method on double-stranded DNA, with the ALFexpress AutoRead sequencing kit and the ALFexpress automatic sequencer (Pharmacia, Uppsala, Sweden). Oligonucleotide primers were synthesized by Genset (Genset, Paris, France) and at the DNA Synthesis Laboratory, Biotechnology Centre of Oslo (Oslo, Norway). DNA sequences were analyzed using the GCG sequence analysis programs [17]. Open reading frames (ORFs) and restriction sites were identified using MAP and translated using TRANSLATE. Translated ORFs were used to perform queries in protein data banks by the use of FASTA and BLAST algorithms. Theoretical molecular masses and isoelectric points of predicted proteins were calculated with COMPUTE pI/Mw. Putative RNA secondary structures having different ΔG values were calculated with MFOLD.

2.5 Pulsed field gel electrophoresis

Intact B. cereus ATCC 14579 and B. cereus ATCC 10987 genomic DNA was prepared in agarose blocks as described [8]. DNA was digested with the enzymes NotI, SfiI and AscI prior to undergoing pulsed field gel electrophoresis (PFGE) by a Bio-Rad CHEF Mapper XA Pulsed Field Electrophoresis System (Bio-Rad, USA). DNA digestions were carried out by incubating each agarose block for 12 h at 37°C (50°C for SfiI) in a volume of 200 μl containing 30 U of each restriction enzyme in the appropriate buffer and bovine serum albumin when required. When double digestions (NotI/SfiI, NotI/AscI and SfiI/AscI) were performed, the agarose blocks were washed three times in 10 mM Tris–NaCl buffer (pH 7.2) for a total of 8 h and treated with a second endonuclease. The agarose blocks were then incubated in TE buffer for 40 min and for an additional 3 h in 1 mM aminoethyl-benzenesulfonyl fluoride. Two different sets of PFGE runs were used to resolve fragments ranging from 10 up to 40 and from 40 up to 400 kb, according to the algorithms calculated by the Bio-Rad CHEF Mapper. Separation of 10–40-kb fragments was achieved by the use of 1% agarose gel in 0.5×TBE buffer, at 14°C for 25.58 h and 9.0 V cm−1. The forward and reverse switch times had an initial value of 0.30 s and a final value of 0.68 s. For the separation of the 40–400-kb fragments, samples were electrophoresed in 0.5×TBE buffer, at 14°C and 6.0 V cm−1 for 27 h, using an initial and a final switch time of 2.91 s and 35.38 s, respectively. A linear ramping factor was adopted for both runs. DNA fragments were transferred to positively charged nylon membranes (Boehringer Mannheim Italia, Italy) by capillary blotting and hybridized with the labeled p-567 probe.

2.6 RNA isolation, RT-cDNA synthesis and PCR

Total RNA was purified from mid-exponential B. cereus cultures grown in LB broth. Briefly, cells were washed with diethylpyrocarbonate-treated water and 1×108 cells were suspended in 450 μl RLT buffer (RNeasy Mini Kit, Qiagen Inc.). After the addition of one volume of glass beads (diameter 0.1 mm), suspensions were vortexed for 15 min to rupture the cells. Samples were centrifuged for 2 min at 10 000×g and each supernatant, supplemented with 250 μl absolute ethanol, was applied to an RNeasy mini spin column (Qiagen Inc.). After being digested with 40 Kunitz units of RNase-free DNase (Qiagen Inc.) for 20 h, the total RNA was eluted from the column following the instructions of the manufacturer. An aliquot of the RNA was electrophoresed on agarose gel to ensure its integrity and absence of DNA contamination. For cDNA synthesis, up to 2 μg RNA was annealed with 1 μg random hexamer primers at 70°C for 5 min, chilled in a ice-bath for 5 min, and supplemented with RT buffer (50 mM Tris (pH 8.3), 75 mM KCl, 3 mM MgCl2), 13 mM dithiothreitol, 0.5 mM deoxynucleoside triphosphate, and 200 U Superscript II reverse transcriptase (Gibco BRL, Life Technologies). Reactions were carried out at 42°C for 1 h and stopped by incubation at 72°C for 15 min.

For the detection of cheA and fliY expression, PCR amplifications were carried out using cDNA as a template. The forward and reverse primers used in the different reactions were: YF2 (5′-ATATCATTTGGTTCAGCTTCG-3′) and YR1 (5′-ACATTTGGCGGCGTCATG-3′) for the amplification of fliY cDNA; AF1 (5′-ATGCAAACAGATCTATTAAAT-3′) and CS2CA (5′-GACTATTTCTCTTTGACCGA-3′) for the amplification of cheA cDNA; AF1 and YR1 for the amplification of a cDNA fragment containing both cheA and fliY. Reactions were carried out for 30 cycles at 94°C for 1 min, 60°C for 1 min and 72°C for 1.5 min.

3 Results and discussion

3.1 Identification and characterization of B. cereus fliY

The organization of flagella and chemotaxis genes has never been investigated for B. cereus. The only available B. cereus sequence exhibiting significant homology with a chemotaxis gene is represented by a 567-bp short gene fragment (p-567) (EMBL accession number Y08031) highly homologous to a part of B. subtilis fliY. Specific primers were designed to amplify the p-567 fragment from B. cereus ATCC 14579. This DNA probe was used to screen genomic libraries prepared from both ATCC 14579 and ATCC 10987 B. cereus strains. Among the E. coli colonies (about 5×103) screened with the p-567 probe, few positive clones contained recombinant plasmids carrying an insert of about 5 kb. The plasmids derived from the ATCC 14579 and ATCC 10987 libraries were named pKI183 and pKI75, respectively. Sequencing of pKI183 and pKI75 led to identification of two almost identical 1131-bp ORFs both having typical ribosome binding sites (AGGAGG) at positions −9 and −8 from the ATG start codon (EMBL accession numbers AJ272330 and AJ272332). Deduced amino acid sequences showed a theoretical pI of 4.4 for both strains and a predicted molecular mass of 41.6 kDa for the ATCC 14579 strain and of 41.3 kDa for the ATCC 10987 strain. The amino acid sequences from the two B. cereus strains were almost identical (98% aa identity) and were homologous to B. subtilis FliY, with 43% homology in 196 aa (52% of B. cereus FliY) for the ATCC 14579 strain and 40% homology in 205 aa (54% of B. cereus FliY) for the ATCC 10987 strain. The highest degree of homology between B. cereus and B. subtilis FliY was detected in the middle of both sequences while the C-terminal sequence was less conserved (Fig. 1). Although the specific function of FliY in B. cereus has not yet been determined, the structural similarity between B. cereus and B. subtilis FliY strongly suggests their similar role in the chemotactic response. Interestingly, in contrast to what has been observed for B. subtilis FliY [13], no significant homology arose by comparing B. cereus FliY with E. coli FliN and FliM, and only weak homology was found between B. cereus FliY and Borrelia burgdorferi FliM. These observations, indicating that B. cereus fliY is less genetically related to fliN/fliM of enteric bacteria than B. subtilis fliY, may support the idea that a different phylogenetic evolution of this gene has occurred in the two Bacillus species.

Figure 1

Alignment between FliY of B. cereus (B.c.) ATCC 14579 (EMBL accession number AJ272330) and ATCC 10987 (EMBL accession number AJ272332) and B. subtilis (B.s.). Asterisks represent non-conserved residues between the sequences of the two B. cereus strains.

3.2 Identification and characterization of B. cereus cheA

From the sequence analysis of pKI183 and pKI75 inserts, two 2016-bp ORFs (EMBL accession numbers AJ272333 and AJ272331) with a putative ribosome binding site (AGGTAGG) at position −9 from the ATG start codon were identified. The deduced amino acid sequences showed a theoretical pI of 4.9 and predicted molecular masses of 74.8 and 75.1 kDa for strains ATCC 14579 and ATCC 10987, respectively. By BLAST searches, the sequences were identified as CheA. The highest homologies were seen with Listeria monocytogenes CheA (55% overall identity) followed by B. subtilis CheA (42% identity), and homology ranging from 40 to 31% was observed by comparing the same protein in several other microorganisms. Alignments of representative CheA proteins are given in Fig. 2.

Figure 2

Alignment between CheA of B. cereus (B.c.) ATCC 14579 (EMBL accession number AJ272333) and ATCC 10987 (EMBL accession number AJ272331) and CheA of other microorganisms (Listeria monocytogenes, L.m.; Bacillus subtilis, B.s.; Escherichia coli, E.c.; Salmonella typhimurium, S.t.). The autophosphorylation domain containing the histidine 45 (♦) and the ATP binding motifs (*) are shown as gray areas.

CheA was very well conserved in B. cereus, showing more than 98% identity in the two B. cereus strains analyzed. B. cereus CheA contains the active sites characteristic of this protein, which are a NH2-terminal autophosphorylated histidine (position 45) included in the conserved motif IFRSAHTFKG, and two ATP binding motifs [12] (Fig. 2). These findings strongly suggest that B. cereus may be included in the growing list of microorganisms for which CheA has been recognized as a highly conserved protein. The low sequence variability of CheA even within phylogenetically distant genera is consistent with its key role as the main regulator of the chemotactic response in both Gram-negative and Gram-positive bacteria [18]. It is tempting, therefore, to hypothesize that B. cereus CheA plays a central role in the regulation of chemotactic response also in this microorganism.

3.3 Localization of fliY and cheA on the B. cereus chromosome

The genomic region containing fliY and cheA was very well conserved in both the ATCC 14579 and ATCC 10987 B. cereus strains analyzed. The gene organization appeared identical as cheA was located about 170 bp upstream of fliY in both strains, with the common 170-bp non-coding region containing a palindromic sequence followed by a run of T residues. This stem-loop sequence putatively acts as a ρ-independent terminator of transcription (ΔG=−21 kcal mol−1). The relative position of fliY to cheA in B. cereus differs remarkably from that described for B. subtilis, in that B. subtilis fliY is upstream of cheA and 10 chemotaxis and motility genes are inserted between them [19]. In addition, while B. subtilis fliY is translationally coupled to fliM, which is located upstream of fliY with an 11-nucleotide overlap [13], no significant homology with B. subtilis fliM was detected upstream of B. cereus fliY (Fig. 3).

Figure 3

Comparison between the organization of fliY and cheA in B. subtilis (a) and B. cereus ATCC 10987 (b) and ATCC 14579 (c). The lower part of the figure (d) shows the stem-loop structure of the deduced mRNA of an intergenic region that may act as a ρ-independent transcription terminator.

In order to localize the position of fliY and cheA on the B. cereus chromosome, genomic DNA fragments resolved by PFGE were hybridized with the fliY-specific probe (p-567). Since no restriction site for NotI, AscI, SfiI is present within the fliYcheA region of both ATCC 14579 and ATCC 10987 strains, the hybridization with p-567 allowed the identification of DNA fragments containing both genes. The position of fliY and cheA on the B. cereus chromosome was identified by comparing the restriction fragments we obtained with those previously reported for the two strains [810] (Fig. 4). In particular, in B. cereus ATCC 10987, the two genes were detected on a genomic fragment corresponding to the 250-kb AscI region (A5), which is a part of the larger NotI N3 region [7]. The hybridization of B. cereus ATCC 14579 DNA fragments with p-567 led to the localization of fliY and cheA onto the N1 (NotI larger fragment), A1 (AscI larger fragment), and S1 (SfiI larger fragment) overlapping genome region (Fig. 4). Interestingly, since it was previously described that the A5 fragment of B. cereus ATCC 10987 contains the flagellin gene [7], these results could indicate such a DNA fragment as a candidate for containing at least a part of a putative che/fla operon in B. cereus.

Figure 4

Physical maps of B. cereus ATCC 14579 and ATCC 10987. Black regions represent the genomic fragments containing fliY and cheA. Concentric circles represent physical maps obtained with a specific restriction endonuclease: A, AscI; N, NotI; S, SfiI. Three different enzymes were used to construct the physical map of strain ATCC 10987 and two enzymes for strain ATCC 10987 of B. cereus. The outermost ring represents the physical map obtained by combining the digestions performed with the different restriction endonucleases.

3.4 Expression of B. cereus cheA and fliY

The expression of B. cereus cheA and fliY was investigated to evaluate whether these genes are components of a unique transcriptional unit as reported for B. subtilis[11]. To this end, total cDNAs extracted from strains ATCC 14579 and ATCC 10987 of B. cereus were amplified by the use of primers that are specific for each single gene (YF2/YR1 for fliY; AF1/CS2CA for cheA) or for a cDNA fragment comprising both genes (AF1/YR1). Separate amplifications performed with the primer pair YF2/YR1 and AF1/CS2CA produced DNA bands corresponding to fliY and cheA, respectively, as they were obtained when the genomic DNA was used as the template (Fig. 5). These data demonstrate that both genes are expressed in the B. cereus strains analyzed. However, the lack of any cDNA amplification with the primers AF1 and YR1 (Fig. 5) clearly shows that B. cereus cheA and fliY are not part of the same transcriptional unit. This result is in contrast to what has been reported for B. subtilis, in which fliY and cheA constitute a single transcriptional unit being included in the same major fla/che operon [11]. The finding that two separate mRNAs are derived from fliY and cheA is consistent with the presence of the stem-loop structure observed within the 170-bp region located between the two genes (Fig. 4); most likely, this structure acts as a transcriptional terminator downstream of cheA.

Figure 5

Amplification of cDNA with primers specific for fliY (lane 1), cheA (lane 2), and cheAfliY (lane 3). Lane 4 represents the direct amplification of RNA with primers specific for fliY (negative control). Lanes 5–7 represent the amplification of fliY, cheA, and cheAfliY respectively from genomic DNA (positive controls).

The substantial differences observed in the expression of two B. cereus chemotaxis genes in comparison with the expression of those of B. subtilis underline the necessity to further investigate chemotaxis mechanisms in B. cereus and related species to address whether B. subtilis may be really regarded as a model for studying chemotaxis in Gram-positive bacteria.

In conclusion, the novelties of this paper consist in: (i) the characterization of the first two chemotaxis genes, fliY and cheA, in B. cereus; (ii) the demonstration that no differences in sequence and gene organization exist within B. cereus strains; (iii) the assessment of a markedly different organization of the two genes in comparison with those described for B. subtilis; (iv) the finding that, differently from B. subtilis, B. cereus fliY and cheA are part of two distinct transcriptional units.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
View Abstract