Two HlyIIR dimers bind to a long perfect inverted repeat in the operator of the hemolysin II gene from Bacillus cereus

FEBS Letters 581 (2007) 1190–1196

Two HlyIIR dimers bind to a long perfect inverted repeat in the operator of the hemolysin II gene from Bacillus cereus Ekaterina A. Rodikovaa, Oleg V. Kovalevskiya, Sergey G. Mayorova, Zhanna I. Budarinaa, Victor V. Marchenkovc, Bogdan S. Melnikc, Andrew P. Leechb, Dmitri V. Nikitina, Michael G. Shlyapnikova, Alexander S. Solonina,* a

Institute of Biochemistry and Physiology of Microorganisms, Russian Academy of Sciences, Prospekt Nauki, 5, Pushchino, Moscow Region 142290, Russia b Department of Biology, University of York, Heslington, York YO10 5YW, UK c Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region 142290, Russia Received 7 January 2007; revised 12 February 2007; accepted 14 February 2007 Available online 28 February 2007 Edited by Lev Kisselev

Abstract HlyIIR is a negative transcriptional regulator of hemolysin II gene from B. cereus. It binds to a long DNA perfect inverted repeat (44 bp) located upstream the hlyII gene. Here we show that HlyIIR is dimeric in solution and in bacterial cells. No protein–protein interactions between dimers and no significant modification of target DNA conformation upon complex formation were observed. Two HlyIIR dimers were found to bind to native operator independently with Kd level in the nanomolar range. The minimal HlyIIR binding site was identified as a half of the long DNA perfect inverted repeat.  2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: TetR family; DNA-binding; Transcriptional regulator

1. Introduction The Bacillus cereus group includes Gram-positive, sporeforming, aerobic rod-shaped bacteria [1]. Well-known members of the group are B. anthracis, the causative agent of anthrax, insecticidal bacteria B. thuringiensis [2,3], and B. cereus, an opportunistic human pathogen that causes food poisoning and other infection diseases [4]. In the previous studies, we have described the hemolysin II gene, encoding one of several cytolytic toxins produced by B. cereus [5–7]. HlyII is a cytolysin capable to lyse different kinds of eukaryotic cells, a member of the b-barrel pore-forming toxin (PFT) family of secreted microbial proteins [8,9]. Many b-barrel PFTs are important pathogenic factors adapting bacteria to a hostile environment during their transient stay in warm-blooded organisms [1,10]. Their expression is tightly regulated in bacterial cells. Recently, three transcriptional regulators that control the expression of virulence genes in B. cereus have been identified: (i) PlcR, a global transcription regulator for most B. cereus cytotoxins. Genome analysis data suggest that PlcR directly controls more than one hundred genes and operons including * Corresponding author. Fax: +7 495 956 33 70. E-mail address: [email protected] (A.S. Solonin).

genes encoding numerous pore-forming and lipolytic toxins, proteases, and other pathogenic factors [11]; (ii) Fur, a ferric uptake repressor. The studies using an insect infection model have demonstrated that a fur null-mutant strain of B.cereus is significantly attenuated [12]; and (iii) HlyIIR, a specific transcriptional regulator for the hlyII gene [13]. Other factors that regulate pathogenic factors expression in B. cereus are unknown. The hlyIIR gene located immediately downstream the hlyII gene encodes a 201 amino acid long protein [13]. In vivo experiments have demonstrated that in the presence of hlyIIR gene the level of hlyII expression is decreased. The HlyIIR protein specifically binds to the operator region (Fig. 1) of the hlyII gene and interacts with RNA polymerase in vitro [13]. The N-terminal part of HlyIIR displays sequence similarity to a number of bacterial transcriptional regulators belonging to the TetR repressors family, while the C-terminal part is more divergent. Overall sequence identity of HlyIIR to the characterized proteins of TetR family is as low as 21% (for QacR from Staphylococcus aureus) or less. Members of TetR family are implicated in the regulation of apparently unrelated pathways and share several common features such as: (i) a highly conserved helix-turn-helix (HTH) motif implicated in DNA binding; (ii) dependence on cofactors that regulate the factor’s activity; (iii) involvement in the adaptation to a changing environment; and (iv) acting as homodimers [14]. Usually, the proteins of TetR family bind to DNA inverted repeats of 15 bp [15], but HlyIIR was shown to interact with a 50 bp region located within the hlyII operator containing an unusually long (44 bp in size) perfect inverted repeat [13], which includes short subrepeats. The details of HlyIIR interaction with this complicated operator as well as the reasons for complicity of operator arrangement were obscure. To address these issues, we have determined oligomerization state of HlyIIR protein and characterized HlyIIR interactions with the complete hlyII operator and its parts.

2. Materials and methods 2.1. HlyIIR protein His6-tagged HlyIIR protein was purified as reported previously [13] with slight modification. The ammonium sulfate precipitation step (40% saturation) was added before metal-affinity chromatography.

0014-5793/$32.00  2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2007.02.035

E.A. Rodikova et al. / FEBS Letters 581 (2007) 1190–1196

1191

Fig. 1. Structure of promoter-operator region of hlyII gene The 44bp palindrome, the part of the HlyIIR protected in DNase I footprinting experiments sequence [13] is indicated by two long arrows. Two subrepeats within inverted repeat are shown below by four short arrows.

Pellet was collected by centrifugation, dissolved in buffer A [13], and loaded onto Ni-NTA agarose column. In this study, we have used the His6–tagged HlyIIR protein named as HlyIIR. 2.2. Analytical gel filtration The molecular weight of purified HlyIIR protein in solution was determined on a Superose – 12 HR 10/30 column (Amersham Pharmacia Biotech) equilibrated with 50 mM Na-phosphate, pH 7.5, 200 mM NaCl. 200 lg of purified HlyIIR protein, at a concentration of 5 mg/ ml, was loaded onto the column and eluted at a flow rate of 0.4 ml/ min. The molecular mass of HlyIIR was determined by interpolation, using a calibration curve of proteins of the known molecular masses: bovine serum albumin (67 kDa), ovalbumin (45 kDa), and chymotripsinogen A (20 kDa). 2.3. Chemical cross-linking Chemical cross-linking of HlyIIR protein was carried out as follows. Purified 6His-HlyIIR protein samples (20 lg) were preincubated in the presence or absence of appropriate amount of 132 bp DNA fragment in a total reaction volume of 150 ll containing 20 mM Na-phosphate (pH 7.6) and 100 mM NaCl for 20 min at 20 C. After addition of glutaraldehyde to a final concentration of 0.1% and incubation for 30 min at 20 C, the reaction was quenched by addition of NaBH4 (freshly prepared 2 M in 0.1 M NaOH) to a final concentration of 100 mM and incubation for 20 min at 20 C. Protein was precipitated by TCA, pellet was washed with acetone, dissolved in 1· loading buffer, and then heated for 5 min at 95 C prior to the analysis on SDS PAGE. 2.4. In vivo oligomerization assay The Escherichia coli strains and plasmids for the in vivo oligomerization assay in a k cI fusion hybrid system [16] were kindly provided by Dr. J. Hu. To construct the pHCF1 plasmid, the hlyIIR gene was amplified with 5 0 GAGGTGTGGTCGACGGGGAAGTCTCGTGA 3 0 (sense) and 3 0 AAACACTGGGATCCTCATATATTAGGCTT 5 0 (antisense) primers carrying SalI and BamHI sites (underlined), respectively. The pUJ1 plasmid [5] was used as a template. The PCR-amplified product was treated with SalI and BamHI restriction enzymes and cloned into the pJH391 vector. The resulting plasmid pHCF1 carried an in-frame fusion construct containing the coding sequence for the k CI N-terminal domain and full-sized HlyIIR. To test protein oligomerization in the hybrid system, the b-galactosidase activity in reporter E. coli strains (JH372, JH607 and XZ980) harboring pHCF1 plasmid was measured as described by Miller [17]. All measures were made in triplicate. 2.5. Analytical ultracentrifugation Sedimentation equilibrium centrifugation was done in an AN-50Ti rotor using Beckman 12 mm path length six channel charcoal-filled Epon centrepieces and quartz windows, in Beckman Optima XL/I analytical ultracentrifuge, at 20 C using speeds between 8000 and 22 000 rpm. Approximately 120 ll of reference buffer and a slightly lower volume (115 ll) of protein sample were loaded into the cells. Absorbance scans were taken at 3000 rpm to check loading concentrations and that the cell contents were uniformly distributed. The speed was increased and absorbance scans (260 or 280 nm) had been taken at

approximately 3-h intervals until sedimentation equilibrium had been achieved (judged by the absence of change in subtractions of successive scans). Sedimentation velocity experiments were performed using Beckman cells with 12 mm path length double sector charcoal-filled Epon centerpieces and sapphire windows, in an AN-60Ti rotor. Approximately 420 ll of reference buffer and a slightly lower volume (416 ll) of sample were loaded into the cells. Absorbance scans were taken at 3000 rpm to check loading concentrations and that the cell contents were uniformly distributed. The speed was increased to 45 000 rpm and absorbance scans (260 or 280 nm) had been taken at 3-min intervals until sedimentation was complete. Partial specific volumes, buffer densities and viscosities were estimated using the program SEDNTERP [18]. The data were analyzed using SEDFIT program [19] and the c(s) model (continuous distribution of sedimentation coefficients). Sedimentation coefficient distribution graphs were not corrected for the buffer density or viscosity. Sedimentation equilibrium data were analyzed using the Origin software (Beckman). 2.6. Oligonucleotides All oligonucleotides used in this study were synthesized on a DNA synthesizer by the solid-phase phosphoramidite method and purified by 15% PAGE. Two synthetic oligonucleotides: 5 0 -CCGTTTAAACAAGAATTTTAAATATGCC-3 0 and 5 0 -GGCATATTTAAAATTCTTGTTTAAACGG-3 0 were annealed to form a double stranded (ds) DNA fragment containing a half of the 44 bp operator region of the hlyII gene enhanced by 6 additional nucleotides (O_half) (Fig. 1). The [c-32P] ATP-labeled O_half dsDNA fragments were purified by PAGE. A 132 bp DNA fragment containing the native operator region

Fig. 2. Determination of the molecular weight of HlyIIR by gelfiltration.

1192 of the hlyII gene was amplified from plasmid pUJ1 [5] and uniformly labeled [a-32P] ATP by PCR using Taq DNA polymerase (‘‘Fermentas’’). The primers used were 5 0 -CGCCAGGGTTTTCCCAGTCACGAC-3 0 (sense) and 5 0 -TTAAAACACAGAATAACGATTTA-3 0 (antisense). 5 0 -F-CGCCAGGGTTTTCCCAGTCACGAC-3 0 (‘‘Sintol’’ Ltd., Moscow) was used to generate a PCR product of the same length carrying fluorescein (F) at one 5 0 -end. The PCR products were purified by 7% PAGE and used as probes for EMSA, flourescence stoichiometric titration, and CD spectroscopy. 2.7. Electrophoretic mobility shift assay (EMSA) The purified 5 0 [c-32P] ATP-labeled oligonucleotide O_half (0.75 pM) and [a-32P] ATP – labeled 132 bp DNA fragment (0.24 pM) were incubated with various amounts of purified HlyIIR for 20 min at 37 C in a buffer containing 20 mM Tris–HCl, pH 8.0, and 100 mM NaCl, in a

E.A. Rodikova et al. / FEBS Letters 581 (2007) 1190–1196 total volume of 10 ll. The 0.2 lg of BSA and 1lg of chicken erythrocyte DNA as non-specific competitor DNA were added to the reaction. After incubation, the samples were immediately loaded onto a 12% or 7% PAGE and electrophoresed in Tris borate-EDTA buffer under non-denaturing conditions. The radioactive unbound DNA and HlyIIR–DNA complexes were visualized by X-ray film. To determine the Kd values, different amounts of the purified labeled 132 bp DNA fragment (0.044, 0.066, 0.1, 0.15, 0.225, 0.3375, 0.5 pM) and 5 0 [c-32P] ATP-labeled oligonucleotide O_half (0.33, 0.5, 0.75, 1.13, 1.67, 2.5, 3.75 pM) were incubated with the constant amounts of HlyIIR: 0.66 pM and 1.31 pM, respectively. Gel segments containing individual electrophoretic species were excised from the gel using the film as a guide and counted in a scintillation counter [20]. The Kd values were estimated from a Scatchard plot [21]. This is a plot of [S]bound/[S]free vs. [S]bound. Kd is the negative reciprocal of the slope.

Fig. 3. Analytical ultracentrifugation of HlyIIR. (a) Sedimentation velocity of HlyIIR protein analyzed as c(s) distribution. (b) c(M) plot from sedimentation velocity experiment on HlyIIR. (c) Fit of equilibrium data for 0.2 mg/ml HlyIIR at 16 000 rpm, 20 C. The best fit from the data for a single species gives an estimated MW of 46 743 ± 700 Da. (d) Sedimentation velocity data for HlyIIR–O_half DNA complex (solid line) and 24 bp DNA duplex (dashed line), at 45 000 rpm 20 C.

E.A. Rodikova et al. / FEBS Letters 581 (2007) 1190–1196

1193

2.8. Anisotropy assay Fluorescence anisotropy measurements were made with a Shimadzu RF-5301PC (Japan) fluorescence spectrometer using an L-format setup. The samples were excited at 490 nm using the 5 nm slit. The vertical and horizontal emission was monitored at 520 nm with the slit width of 10 nm. Ten measurements were taken and averaged for each anisotropy value. The integration time was 4 s. Stoichiometric titration was performed in a temperature-controlled cuvette at 37 C. Protein diluted in the binding buffer (20 mM Tris–HCl, pH 8.0, 100 mM NaCl, 0.1 mg/ml BSA) was added stepwise to the 1 ml of 2.5 nM fluoresceinlabeled 132-bp PCR fragment in the same buffer. DNA dilution was taken into account during the data analysis. Experimental data were analyzed accordance to theoretical justification of protein–DNA interaction [22], using final Eq. (1) r ¼ ½A2 B=½BðrAB  rB Þ þ rB

ð1Þ

were r is the observed anisotropy and rB and rAB are the anisotropy of the free and bound DNA. [A2] and [B] represent the concentrations of total protein dimer and total DNA. 2.9. Circular dichroism (CD) spectroscopy CD measurements were performed in 20 mM Tris–HCl, pH 8.0, 100 mM NaCl with DNA (132 bp PCR fragment) concentration of 1.5 · 106 M and different HlyIIR concentrations. The investigations were carried out on a Jasko-600 spectropolarimeter (Japan Spectroscopic Co., Tokyo, Japan) using a 1.0 mm – pathlength cell equilibrated at 37 C.

3. Results and discussion 3.1. HlyIIR is a dimer in solution The oligomerization state of the purified HlyIIR protein was analyzed by gel filtration and analytical ultracentrifugation experiments. When subjected to gel filtration, HlyIIR eluted as a single symmetric peak at a position corresponding to a globular protein of approximately 45 kDa (Fig. 2). Since the molecular mass of HlyIIR monomer is 23.5 kDa, it suggests that the protein to form stable dimers in solution. The results of sedimentation velocity analysis show that HlyIIR sediments as a single species at 3.24S (3.41S if corrected for buffer density) (Fig. 3a). This value is consistent with the dimeric HlyIIR form with the frictional ratio 1.3; conversion of the data to a c(M) distribution gives a single peak at 48 143 Da, about 3% higher than expected for dimer (Fig. 3b). The results of sedimentation equilibrium analysis provide molecular weight (MW) estimates of 46.7–51.5 kDa, which is again consistent with the expected MW of the dimer, 47 kDa. An example of sedimentation plot is shown in Fig. 3c. The plots of estimated MW against concentration do not indicate any significant association/dissociation behavior at accessible concentrations (data not shown). 3.2. HlyIIR binds to DNA as a dimer The ability of HlyIIR protein to form other oligomeric structures during the DNA complex formation was examined by chemical cross-linking, sedimentation velocity and hybrid fusion system in vivo. Glutaraldehyde cross-linking of HlyIIR in the presence of the 132 bp fragment operator DNA seemed not to change the cross-linking pattern observed for the free protein, with a major band corresponding to the cross-linked HlyIIR dimers (Fig. 4) and few minor bands corresponding to higher molecular weight complexes. It seems likely that these bands, some of which may correspond to tetramers of HlyIR, resulted from non-specific cross-linking. Thus, binding

Fig. 4. Glutaraldehyde cross-linking. Lane 1 – markers, 2 – HlyIIR, 3 – cross-linked HlyIIR, 4 – HlyIIR cross-linked in the presence of its operator DNA.

to the specific operator region does not alter HlyIIR oligomerization state. The hybrid k cI fusion system was also used to analyze HlyIIR oligomerization in vivo [16,23]. The C-terminal domain of the k cI repressor is responsible for dimerization, whereas the N-terminal domain binds to DNA. The N-terminal domain alone cannot function as an effective repressor. The fusion of k cI DNA-binding domain with any protein capable of forming oligomers can restore the repressor activity of this fusion protein. Thus, the protein oligomerization state can be tested in vivo in this hybrid system by comparing the repressor activity of the fusion protein with the wild-type k cI using reporter strain that carries the lacZ gene under the control of k PROR promoter/operator (JH372). Two additional reporter strains, each containing synthetic operator regions, were used. The first strain carried combination of strong k operator Os1 and weak k operator Os2 overlapping promoter of the reporter gene (JH607). The second reporter strain carried only weak Os2 operator (XZ980). Such system allowed discrimination between (i) dimerization and (ii) formation of higher oligomeric forms or establishment of additional protein–protein interactions by tested protein. Full-sized hlyIIR gene was cloned in-frame with the N-terminal domain of the k cI gene to create pHCF1 plasmid (see Section 2). The analysis of b-galactosidase activity in the reporter E. coli strains clearly shows that the fusion protein and hence HlyIIR formed dimers in bacterial cells and were not able to form higher oligomers (Tables 1 and 2). Similar level of reporter activity in strains JH607 and XZ980 also indicates the absence of protein–protein interaction within HlyIIR dimers in vivo.

Table 1 b-Galactosidase activity in the E. coli reporter strain JH372 Plasmid

Description

Miller units

pZ150 pKH101 pUJ1 pFG157 pHCF1

Vector plasmid cI repressor N-terminal domain HlyIIR Wild-type cI repressor cI-HlyIIR fusion protein

2200 1200 2200 130 150

1194

E.A. Rodikova et al. / FEBS Letters 581 (2007) 1190–1196

Table 2 b-Galactosidase activity in the E. coli reporter strains JH607 and XZ980 Miller units pKH101 pJH370 pJH622 pHCF1

cI repressor N-terminal domain Dimer controla Tetramer controla cI-HlyIIR fusion protein

XZ980/JH607 ratio

JH607

XZ980

1700

1600

0.94

880 225 800

800 620 850

0.9 2.8 1.1

a

Control plasmids were provided as part of the hybrid k cI fusion system [16].

3.3. Two HlyIIR dimers bind specifically to the operator region of the hlyII gene The most intriguing thing about HlyIIR is that it interacts with an unusually long (44 bp in size) perfect inverted repeat located in the operator of the hemolysin II gene, while usually the members of TetR family interact with DNA inverted repeats of 15 bp [15]. Two distinct bands corresponding to different DNA–protein complexes appeared simultaneously at a low protein–DNA molar ratio when EMSA was carried out with a 132 bp PCR fragment containing native operator (Fig. 5a). The complete transition of all DNA substrates to protein– DNA complexes was observed at a molar ratio of 6.5:1, and then further shift to the higher molecular weight complex was observed until complete disappearance of the lower weight complexes. Further changes in the mobility of the complexes were apparently due to non-specific HlyIIR–DNA complex formation. The appearance of two different DNA–protein complexes suggests that several HlyIIR dimers bind to the native operator region.

Fig. 5. EMSA. Binding of HlyIIR to the 132 bp PCR fragment (a) and to the O_half ds DNA (b). Line 1 – without HlyIIR protein, lines 2–9: 83.4 pM, 20.8 pM, 5.2 pM, 1.3 pM, 0.3 pM, 0.16 pM, 0.08 pM, 0.04 pM of HlyIIR (a); lines 2-8: 10.4 pM, 5.2 pM, 2.6 pM, 1.3 pM, 0.7 pM, 0.3 pM, 0.16 pM of HlyIIR (b) . Molar ratio of HlyIIR/DNA are indicated in the bottom cell of each panel. Left arrows indicate DNA marker positions.

Fig. 6. Fluorescence anisotropy titration of 2.5 nM fluorescein-labeled native operator DNA with HlyIIR.

The fluorescence anisotropy titration assay was used to determine the exact molar ratio of HlyIIR and DNA in a saturated complex. The constant DNA concentration of 132 bp DNA fragment carrying a fluorescein chromophore at one end was mixed with different amounts of HlyIIR protein, and fluorescence anisotropy was measured. Protein had been added until all binding sites were saturated and the value of anisotropy reached a plateau (Fig. 6). The intersection of two linear parts of the binding curve yields a stoichiometric equivalence at 5.2 nM protein dimer concentration, while DNA concentration in the reaction mixture was 2.5 nM. Thus, HlyIIR dimers bind to hlyII operator DNA with a stoichiometry of 2:1, suggesting that two HlyIIR binding sites are located within operator’s perfect inverted repeat. However, chemical cross-linking of HlyIIR in the presence of operator DNA did not reveal any additional protein forms. We could interpret this result as two HlyIIR dimers bound to the operator DNA located at a distance exceeding the bond length of cross-linking reagent. The shape of the fluorescence anisotropy titration curve suggests that HlyIIR–DNA interactions are non-cooperative in character. Our previous structural data suppose that the distance between the two DNA-recognition helices of the HlyIIR dimer ˚ [24], suggesting that the two helices fitting into is about 35 A a major groove are separated by one full turn of the DNA. In such a complex, the protein dimer covers about 20 bp of the DNA. As HlyIIR specifically protects 50 bp of the hlyII operator DNA from DNase I digestion [13], at least two dimers of HlyIIR should bind to cover the hlyII operator region. We have checked if HlyIIR is able to introduce structural changes in the bound DNA, since it has been shown for other DNA-binding proteins binding more than one site within the operator region, e.g., TetR protein [25]. The conformational changes in the target DNA upon complex formation resulted in increase in the CD signal (De) at wavelengths above 250 nm [26]. However, no changes in CD signal were observed, when increasing amounts of HlyIIR were added to the 132 bp DNA fragment (Fig. 7). Thus, HlyIIR protein does not induce significant changes in the structure of the hlyII operator DNA. Perfect inverted repeat located in the operator region of the hlyII gene contains two subrepeats (Fig. 1). We supposed that these subrepeats could represent two consequent (proximal

E.A. Rodikova et al. / FEBS Letters 581 (2007) 1190–1196

1195

Fig. 7. CD spectra of the operator DNA, HlyIIR (a) and HlyIIR-operator DNA complex (b). The solid line shows the CD of the 132 bp DNA; the dotted line indicates the CD of the free HlyIIR (a). Black, light grey, dashed and grey lines show the CD of the repressor-operator complex with 0.4, 1.6, 3.8 and 14.7 protein/DNA molar ratios, respectively (b).

and distal) HlyIIR binding sites and examined HlyIIR interaction with the half of the operator region. The results of sedimentation velocity experiments showed that two species were present in the mixture of HlyIIR and O_half DNA (Fig. 3d, solid line) with sedimentation coefficients 2.56S and 4.63S (17% and 81% of the total signal, respectively). A 24bp DNA duplex was run as a control (Fig. 3d, dashed line), behaving as a single species with sedimentation coefficient 2.43S. It is likely that the slower species in the mixture is unbound DNA and the faster is a protein–DNA complex. The expected sedimentation velocity of the HlyIIR dimer–O_half complex is close to the obtained experimental value. EMSA showed that HlyIIR interacts with the half of the inverted repeat and forms only one DNA–protein complex at any molar ratio (Fig. 5b), suggesting it is the minimal HlyIIR binding site. The total transition of the O_half nucleotide substrate to the DNA–protein complex was observed at a protein to DNA molar ratio as 3.5:1. The apparent dissociation constants for complexes with both DNA substrates were determined by the Scatchard plot. Kd was estimated as 4.4 ± 1.7 · 109 M for HlyIIR – native operator and as 8.3 ± 3.0 · 109 M for HlyIIR–O_half oligonucleotide complex. Additional EMSA experiments indicated that wide range variations in ionic strength (50–500 mM NaCl) and pH (from 6.8 to 8.5) had almost no effect on the binding of HlyIIR to the operator DNA. Moreover, the DNA–HlyIIR complex is very stable, as it has been observed in solutions containing 2 M urea (data not shown). Operator region of the hlyII gene is represented by a long perfect inverted repeat centered 48 bp upstream of the hlyII promoter transcription initiation point. The data presented here indicate that this region includes two consequent subrepeats located in distal and proximal parts and centered 61 and 36 bp, respectively (Fig. 1). HlyIIR dimers bind both subrepeats with high affinity but independently, as suggested by the data obtained from k cI fusion system and by chemical cross-linking. Taking together with the results of fluorescence anisotropy titration experiment these data indicate non-cooperative interactions. It is still not clear why hlyII operator has such a complicated structure, with two HlyIIR binding sites joined together to form a long perfect inverted repeat. However, this structure seems to be evolutionary stable and it is present in all strains

of B. cereus group, for which genome sequences are available and which contain the hlyII gene (B. cereus, B. thuringiensis, and B. anthracis) [6]. There are a few nucleotide substitutions located within the inverted repeat in different strains (Fig. 1). It is notable that all these substitutions are symmetrical, preserving perfect symmetry of the inverted repeat. This arrangement of the operator region points to an exceptional biological importance of such a complicated structure. However, we have found additional nucleotide sequences similar to a half the hlyII gene operator region in complete genomes of microorganisms of B. cereus group published earlier. These sequences carry single nucleotide substitutions within the operator region. Preliminary EMSA results for the two of them suggest that HlyIIR is able to take part in regulation of additional genes from B. cereus. Finally, we would like to emphasize that the study of HlyIIR oligomerization and detailed clarification of the HlyIIR–DNA interactions are crucial for understanding the molecular mechanisms of regulation of B. cereus hemolysin II expression. Further in vivo experiments are required to elucidate the role of this unusual organization of the cytotoxin gene operator. This will extend our knowledge about the pathogenesis processes and/or microbial environmental adaptation. Acknowledgements: This work was supported by the Russian Foundation for Basic Research, Project 03-04-48623 and 04-04-49693 (A.S.), and by a sub grant with Rutgers University, Office of Research and Sponsored Programs, under Sponsor Award No. 0853 from the Burroughs Welcome. We are grateful to Dr. J. Hu for providing the Escherichia coli strains and plasmids for the in vivo oligomerization assay in a k cI fusion hybrid system and to Dr. O. Denisenko and Dr. M. Sinev for critically reading the manuscript.

References [1] Jensen, G.B., Hansen, B.M., Eilenberg, J. and Mahillon, J. (2003) The hidden lifestyles of Bacillus cereus and relatives. Environ. Microbiol., 631–640. [2] Helgason, E., Caugant, D.A., Lecadet, M.M., Chen, Y., Mahillon, J., Lo¨vgren, A., Hegna, I., Kvaløy, K. and Kolstø, A.B. (1998) Genetic diversity of Bacillus cereus/B. thuringiensis isolates from natural sources. Curr. Microbiol. 37, 80–87. [3] Helgason, E., Økstad, O.A., Caugant, D.A., Johansen, H.A., Fouet, A., Mock, M., Hegna, I. and Kolstø, A.B. (2000) Bacillus anthracis, Bacillus cereus, and Bacillus thuringiensis – one species

1196

[4] [5]

[6]

[7]

[8]

[9]

[10] [11]

[12] [13]

[14]

E.A. Rodikova et al. / FEBS Letters 581 (2007) 1190–1196 on the basis of genetic evidence. Appl. Environ. Microbiol. 66, 2627–2630. Granum, P.E. and Lund, T.A. (1997) Bacillus cereus and its food poisoning toxins. FEMS Microbiol. Lett. 157, 223–228, Review. Sinev, M.A., Budarina, Zh.I., Gavrilenko, I.V., Tomashevskii, A.Iu. and Kuz’min, N.P. (1993) Evidence of the existence of hemolysin II from Bacillus cereus: cloning the genetic determinant of hemolysin II. Mol. Biol. (Mosk.) 27, 1218–1229, (in Russian). Budarina, Z.I., Sinev, M.A., Mayorov, S.G., Tomashevski, A.Y., Shmelev, I.V. and Kuzmin, N.P. (1994) Hemolysin II is more characteristic of Bacillus thuringiensis than Bacillus cereus. Arch. Microbiol. 161, 252–257. Baida, G., Budarina, Z.I., Kuzmin, N.P. and Solonin, A.S. (1999) Complete nucleotide sequence and molecular characterization of hemolysin II gene from Bacillus cereus. FEMS Microbiol. Lett. 180, 7–14. Andreeva, Z.I., Nesterenko, V.F., Fomkina, M.G., Ternovsky, V.I., Suzina, N.E., Bakulina, A.Y., Solonin, A.S. and Sineva, E.V. (2007) The properties of Bacillus cereus hemolysin II pores depend on environmental conditions. Biochim. Biophys. Acta 1768 (2), 253–263. Andreeva, Z.I., Nesterenko, V.F., Yurkov, I.S., Budarina, Z.I., Sineva, E.V. and Solonin, A.S. (2006) Purification and cytotoxic properties of Bacillus cereus hemolysin II. Protein Expr. Purif. 47, 186–193. Schmitt, C.K., Meysick, K.C. and O’Brien, A.D. (1999) Bacterial toxins: friends or foes? Emerg. Infect. Dis. 5 (2), 224–234, Review. Ivanova, N., Sorokin, A., Anderson, I., Galleron, N., Candelon, B., Kapatral, V., Bhattacharyya, A., Reznik, G., Mikhailova, N., Lapidus, A., Chu, L., Mazur, M., Goltsman, E., Larsen, N., D’Souza, M., Walunas, T., Grechkin, Y., Pusch, G., Haselkorn, R., Fonstein, M., Ehrlich, S.D., Overbeek, R. and Kyrpides, N. (2003) Genome sequence of Bacillus cereus and comparative analysis with Bacillus anthracis. Nature 423, 87–91. Harvie, D.R., Vilchez, S., Steggles, J.R. and Ellar, D.J. (2005) Bacillus cereus Fur regulates iron metabolism and is required for full virulence. Microbiology 151, 569–577. Budarina, Z.I., Nikitin, D.V., Zenkin, N., Zakharova, M., Semenova, E., Shlyapnikov, M.G., Rodikova, E.A., Masyukova, S., Ogarkov, O., Baida, G.E., Solonin, A.S. and Severinov, K. (2004) A new Bacillus cereus DNA-binding protein, HlyIIR, negatively regulates expression of B. cereus haemolysin II. Microbiology 150, 3691–3701. Ramos, J.L., Martinez-Bueno, M., Molina-Henares, A.J., Teran, W., Watanabe, K., Zhang, X., Gallegos, M.T.,

[15]

[16]

[17] [18]

[19] [20] [21] [22] [23]

[24]

[25] [26]

Brennan, R. and Tobes, R. (2005) The TetR family of transcriptional repressors. Microbiol. Mol. Biol. Rev. 69, 326–356, Review. Engohang-Ndong, J., Baillat, D., Aumercier, M., Bellefontaine, F., Besra, G.S., Locht, C. and Baulard, A.R. (2004) EthR, a repressor of the TetR/CamR family implicated in ethionamide resistance in mycobacteria, octamerizes cooperatively on its operator. Mol. Microbiol. 51, 175–188. Hu, J.C., O’Shea, E.K., Kim, P.S. and Sauer, R.T. (1990) Sequence requirements for coiled-coils: analysis with lambda repressor-GCN4 leucine zipper fusions. Science 250, 1400–1403. Miller, J.H. (1972) Experiments in Molecular Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, pp. 352–355. Laue, T.M., Shah, D.D., Ridgeway, T.M. and Pelletier, S.L. (1992) Computer-aided interpretation of analytical sedimentation data for proteins in: Analytical Ultracentrifugation in Biochemistry and Polymer Science (Harding, S.E., Rowe, A.J. and Horton, J.C., Eds.), pp. 90–125, Royal Society of Chemistry, Cambridge, UK. Schuck, P. (2000) Size-distribution analysis of macromolecules by sedimentation velocity ultra centrifugation and Lamm equation modeling. Biophys. J. 78, 1606–1619. Fried, M.G. (1989) Measurement of protein–DNA interaction parameters by electrophoresis mobility shift assay. Electrophoresis 10, 366–376. Scatchard, G. (1949) The attractions of proteins for small molecules and ions. Ann. NY Acad. Sci., 660–672. Lundblad, J.R., Laurance, M. and Goodman, R.H. (1996) Fluorescence polarisation analysis of protein–DNA and protein–protein interactions. Mol. Endocrinol. 10, 607–612. Di Lallo, G., Ghelardini, P. and Paolozzi, L. (1999) Two-hybrid assay: construction of an Escherichia coli system to quantify homodimerization ability in vivo. Microbiology 145, 1485–1490. Kovalevskiy, O.V., Lebedev, A.A., Surin, A.K., Solonin, A.S. and Antson, A.A. (2007) Crystal structure of Bacillus cereus HlyIIR, a transcriptional regulator of the gene for pore-forming toxin hemolysin II. J. Mol. Biol. 365, 825–834. Altschmied, L. and Hillen, W. (1984) TET repressor.tet operator complex formation induces conformational changes in the tet operator DNA. Nucleic Acids Res. 12, 2171–2180. Culard, F. and Maurizot, J.C. (1982) Binding of lac repressor induces different conformational changes on operator and nonoperator DNAs. FEBS Lett. 146, 153–156.