Physicochemical properties and distinct DNA binding capacity of the repressor of temperate Staphylococcus aureus phage /11 Tridib Ganguly*, Malabika Das*, Amitava Bandhu, Palas K. Chanda, Biswanath Jana, Rajkrishna Mondal and Subrata Sau
Department of Biochemistry, Bose Institute, Calcutta, India
Keywords dimer; major groove; operator; phage /11; repressor (CI)
Correspondence S. Sau, Department of Biochemistry, Bose Institute, P1 ⁄ 12 – CIT Scheme VII M, Calcutta 700 054, India Fax: +91 33 2355 3886 Tel: +91 33 2569 3200 E-mail: subratasau@gmail.com
*These authors contributed equally to this work
(Received 20 November 2008, revised 16 January 2009, accepted 21 January 2009)
doi:10.1111/j.1742-4658.2009.06924.x
The repressor protein and cognate operator DNA of any temperate Staph- ylococcus aureus phage have not been investigated in depth, despite having the potential to enrich the molecular biology of the staphylococcal system. In the present study, using the extremely pure repressor of temperate Staphylococcus aureus phage /11 (CI), we demonstrate that CI is composed of a-helix and b-sheet to a substantial extent at room temperature, pos- sesses two domains, unfolds at temperatures above 39 (cid:2)C and binds to two sites in the /11 cI-cro intergenic region with variable affinity. The above CI binding sites harbor two homologous 15 bp inverted repeats (O1 and O2), which are spaced 18 bp apart. Several guanine bases located in and around O1 and O2 demonstrate interaction with CI, indicating that these 15 bp sites are used as operators for repressor binding. CI interacted with O1 and O2 in a cooperative manner and was found to bind to operator DNA as a homodimer. Interestingly, CI did not show appreciable binding to another homologous 15 bp site (O3) that was located in the same primary immunity region as O1 and O2. Taken together, these results sug- gest that /11 CI and the /11 CI–operator complex resemble significantly those of the lambdoid phages at the structural level. The mode of action of /11 CI, however, may be distinct from that of the repressor proteins of k and related phages.
repeats [10], other repressors bind to asymmetric oper- ators [2,3,11–13] to establish lysogeny. Interestingly, the repressor of Vibrio cholerae phage CTX/ binds to extended operators, stopping lytic growth, as well as ensuring lysogeny of this phage [4]. Although these regulatory elements have enriched both basic and applied molecular biology enormously, they have not been cloned from most temperate phages or character- ized in any depth.
The basic regulatory elements that most temperate phages use for the establishment and maintenance of their lysogeny are the phage-encoded repressor and the cognate operator DNA [1–12]. A temperate phage generally enters into the lysogenic life cycle once its repressor inhibits the transcription of the phage- specific lytic genes from the early promoter by binding to the overlapped operator DNA. Repressors of the temperate phages, although varying greatly in size and in primary sequence level, mostly harbor a DNA bind- ing domain and an oligomerization domain. The size and type of the operator DNAs also vary from phage to phage. Although some repressors bind to operators with dyad symmetry [1,5–9] or operators with direct
The temperate Staphylococcus aureus phage /11 [14] harbors the cI and cro genes in a divergent orientation to that in lambdoid phages [1,8]. The sequence of the immunity region of /11, however, differs significantly from those of the lambdoid phages and other temper-
Abbreviations CI, repressor of temperate Staphylococcus aureus phage /11; CTD, C-terminal domain; DMS, dimethyl sulfate; DTNB, 5,5¢-dithiobis-(2- nitrobenzoic acid); NTD, N-terminal domain.
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1975
T. Ganguly et al.
Characterization of /11 repressor
procedures), analyzed the resulting protein containing elution fractions by 13.5% SDS ⁄ PAGE (Fig. 1A) and found that only fractions F2 and F3 (loaded in lanes 2 and 3) contain intact CI with an estimated purity of almost 98%. The overall yield of CI was approxi- mately 1 mgÆL)1 of induced E. coli culture. Because the above highly-purified CI did not show any degra- dation upon storage on ice for more than 1 month and possessed operator DNA binding activity (described below), it was utilized in all the in vitro experiments performed in the present study.
similar
ate S. aureus phages, such as /PVL, /13, /53, 3A, 77 and /Sa3ms [14–17]. By contrast, the 239 amino acid product of the /11 cI gene shows a moderate homol- ogy over the entire length of the k repressor. Interest- ingly, although the sequences of the C-terminal ends of the above S. aureus phage repressors are identical, the sequences of their N-terminal ends vary considerably [15]. The predicted secondary structures of the repres- sors of S. aureus phages show a notable similarity to that of k repressor, especially at the C-terminal ends. As noted with the C-terminal end of k repressor [1], the repressors of S. aureus the C-terminal ends of phages may be involved in oligomerization. The N-ter- minal half of /11 repressor carries a putative helix- turn-helix DNA binding motif to that of lambdoid phages, indicating that this half of the /11 repressor most likely participates in the binding of operator DNA. An N-terminal histidine-tagged form of the repressor of temperate S. aureus phage /11 (CI) was overexpressed in Escherichia coli and was purified to some extent [15]. An additional (cid:2) 19 kDa protein was always co-purified at a low level along with the intact (cid:2) 31 kDa repressor. This smaller protein, found to comprise the N-terminal end fragment of repressor, was most possibly the result of cleavage of the repres- its alanine–glycine site. The histidine-tagged sor at repressor, however, was shown to form dimers in solu- tion and bind to two sites in the /11 cI-cro intergenic region. Two homologous 15 bp inverted repeats with partial two-fold symmetry, identified in the /11 cI-cro intergenic region, were suggested to act as operator sites because synthetic DNA fragments carrying either repeat showed appreciable binding to CI [15]. Little is its cognate known about the structures of /11 CI, operators and CI–operator complex, the precise bind- ing affinity of CI to the two operators, and the mecha- nism of action of CI. In the present study, we report the purification of /11 CI to near homogeneity and, for the first time, present evidence for the two-domain structure, its thermolability and the binding of CI to two 15 bp operator sites in the cI-cro intergenic region with variable affinity. We also suggest putative tertiary structures for the domains of both the CI and the CI–operator complex.
Results and Discussion
Purification, physicochemical properties and structure of CI
To purify CI to homogeneity, we subjected affinity column chromatography-purified CI [15] to gel filtra- tion chromatography (for details, see Experimental
To map the possible flexible region or domain struc- ture in CI, we performed a partial proteolysis of CI by trypsin and found that protein fragments I and II were the two major products generated from CI at a very early stage of the enzymatic cleavage (Fig. 1B). Both the fragments remained mostly undigested throughout the entire period of digestion. Interestingly, limited proteolysis of CI with chymotrypsin also generated a similar digestion pattern (data not shown). Neither of the above fragments interacted with anti-(his Ig) (data not shown), indicating the loss of the N-terminal histi- dine tag from CI immediately after exposure to the enzyme. The first three N-terminal end amino acid residues of fragment I were determined to be LVS (corresponding to amino acid residues 156–158 of CI), suggesting that it belonged to the C-terminal end of CI. The fragment I most possibly harbors residues 156–276 of CI, with a molecular mass of 13.3 kDa. The fragment II, having a molecular mass of almost 12.14 kDa (as shown by MALDI-TOF analysis), might originate from the N-terminal end of CI because the intensity of fragment I did not decrease with time. The N-terminal end sequencing of one of the chymotryp- sin-digested fragments revealed that the junction region between the C-terminal end of the histidine tag and the N-terminal end of the native /11 repressor (which carries both chymotrypsin and trypsin cleavage sites) is exposed to the surface of the CI. Taken together, this suggests that the histidine-tagged CI carries two flexi- ble regions: one at the N-terminal end and another almost at the middle of the molecule. Tryptic digestion of CI at the above two regions yielded two extremely folded structures or domains [designated N-terminal domain (NTD) and C-terminal domain (CTD)] of CI where the majority of the thirty four trypsin cleavage sites are buried. The two-domain structure of /11 CI monomer therefore approximately resembles that of k CI and related repressor monomers [1,8]. Interestingly, the putative tertiary structure of the CTD of the /11 repressor (Fig. 1C), modeled using amino acid residues 119–238 of the native /11 repressor (equivalent to indeed showed residues
156–275 of
fragment I),
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1976
T. Ganguly et al.
Characterization of /11 repressor
A
D
1 2 3 4 5 6 7 8
a D k
30 25 20 15
B
º
º 0 2 1
º 0 5 1
º 5 1
º 0 3
º 0 9
º 5
º 0 º 0 8 6 1 – + + + + + + + +
Min 0 Try
a D k
I II
30 25 20 15 10
C
E
15
–6
) M n
≈
41ºC
y t i c i t
Tm
–10
10
p
i l l
E
2 2 2 (
–14
5
25
33
41
49
1 – l o m d
Temp (ºC)
0
· 2 m c · g e d
–5
48 ºC
–10
40 ºC
3 – 0 1 ×
] θ [
–15
25 º & 39 ºC
–20
260
200
240 220 Wavelength (nm)
Fig. 1. Purification and properties of /11 CI. (A) The protein-containing elution fractions from different chromatographys were analyzed by 13.5% SDS ⁄ PAGE (for details, see Experimental procedures). Almost 10 lg of protein was loaded in each lane. Lane 1, elution fraction from affinity chromatography; lanes 2–8, elution fractions F2 to F8. Molecular masses (kDa) of the marker protein bands are shown to the right of the gel. (B) Approximately 4 lg of CI was incubated with 16 ng of trypsin (Try) at 25 (cid:2)C in 20 lL of buffer C and aliquots, withdrawn at the indicated time intervals, were analyzed by Tris–Tricine 15% SDS ⁄ PAGE. Molecular masses (kDa) of marker proteins are shown to the right of the gel. For some unknown reason, Try-generated fragments I and II showed a 3–4 kDa higher molecular mass than their actual masses. (C) Schematic tertiary structure of CTD of /11 CI. The ribbons, helices and tubes represent a-helices, b-sheets and loops, respectively. (D) Schematic representation of NTD of /11 CI; notation as in (C). (E) Far-UV CD-spectra of 10 lM repressor in 200 lL of buffer C, were mea- sured at temperatures in the range 25–48 (cid:2)C. Spectra obtained at 25, 39, 40 and 48 (cid:2)C are shown. The inset shows the plot of the molar ellipticity (h) values at 222 nm (obtained from the above CD spectra) versus the incubation temperatures of CI. The melting temperature (Tm) of CI is also indicated.
a-helix
composed
and
remarkable structural resemblance to the LexA CTDs (r.m.s.d. = 0.46 A˚ ) [18] and to k CI CTD (r.m.s.d. = 1.09 A˚ ) [19]. Similarly, the NTD (Fig. 1D) generated with residues 10–69 of the native /11 repressor exhib- ited structural similarity to a putative DNA-binding protein from Bacteriodes fragilis (r.m.s.d. = 0.06 A˚ ) and to the NTD of k CI (r.m.s.d. = 1.43 A˚ ) [20].
The CD spectrum of /11 CI showed a peak of large negative ellipticity at (cid:2) 208 nm and 25 (cid:2)C, indicating the presence of a-helix in CI at room temperature the spectrum by CD neural (Fig. 1E). Analysis of
networks [21] revealed approximately 23.6% a-helix and 18.5% b-sheet in CI at 25 (cid:2)C. The above CD data are as expected because the NTD and CTD of /11 CI are mostly b-sheet of (Fig. 1C,D). The peaks in the CD spectra of CI at 208 nm, however, were reduced substantially once the incubation temperature of CI was raised above 39 (cid:2)C (Fig. 1C). The plot of molar ellipticity at 222 nm ver- sus the incubation temperature (Fig. 1C, inset) shows that the melting temperature of CI is close to 41 (cid:2)C. At this temperature, the concentration ratio of native
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1977
T. Ganguly et al.
Characterization of /11 repressor
concentration percent operator bound versus CI (Fig. 2E), the CI concentrations that gave 50% satura- tion of input O1 and O2 DNAs (i.e. the apparent equi- librium dissociation constants) were calculated to be almost 32 nm and 120 nm, respectively. Thus, CI binds to O1 nearly four-fold more strongly than to O2.
and denatured CI is 1. The data therefore suggest that the a-helical content of CI, which decreases at temper- atures above 39 (cid:2)C, might be responsible for the alter- ation of the conformation of CI, as well as the reduced operator DNA binding affinity of CI [15]. /11 CI, although structurally similar, is more thermosensi- tive than k repressor [22]. The biological significance of this phenomenon is not known with any certainty. However, we found that the alanine and proline con- tents in /11 CI are significantly less than that in k repressor. Several studies have demonstrated that a higher alanine and ⁄ or proline content contributes sig- nificantly to the enhanced thermostability of various proteins, including phage repressor [22–24].
(DTNB)
During the course of the present study, we identified eight additional 15 bp inverted repeats in the /11 gen- ome sequence [including one (designated O3) in the /11 cI-cro intergenic region; Figs 2A and 3B], which showed 60% or more identity with O1. The O3 site is located 31 bp upstream of O2. Surprisingly, CI was found to bind to O3 DNA (Fig. 2F) at concentrations that are required for its binding to S. aureus cspC DNA carrying no operator (Fig. 2G). An additional gel shift assay (Fig. 2H) using labeled O DNA and higher CI concen- trations showed that CI does not bind to O3, even in the presence of O1 and O2. The data therefore indicate that binding of CI to O3 is nonspecific in nature. Interest- ingly, /11 Cro that neither binds to O1 or O2 demon- strates specific binding to O3 DNA [26].
/11 CI carries three cysteine residues at positions 125, 159 and 207 [14]. To obtain clues about the status (bur- ied versus exposed) of these cysteine residues, we deter- mined the free sulfhydryl group content in CI by the 5,5¢-dithiobis-(2-nitrobenzoic acid) test and found that the number of free thiols in CI is almost 1.5, indicating that two cysteine residues are partially exposed to its surface. The putative surface structure of CTD of /11 CI (data not shown) reveals that cysteine 125 and cysteine 207 are approximately 27% and 30% surface exposed, respectively, whereas, cysteine 159 is mostly buried. The former two cysteine residues most likely showed reactivity with DTNB. Interestingly, k CI also harbors three cysteine residues in its CTD, but none of them are exposed to the surface [25].
Cooperative binding of CI to two sites in the /11 cI-cro intergenic region
To determine whether the binding of CI to O1 and O2 is cooperative in nature, we also studied the equilibrium binding of CI to radiolabeled O1O2 DNA by a gel shift assay. It was found that the O1O2 DNA formed two shifted complexes (1 and 2) with increasing CI concen- trations (Fig. 2I). The complex 1 appears at (cid:2) 3 nm, reaching a maximum at (cid:2) 46 nm and starts disappear- ing at higher CI concentrations. By contrast, complex 2 is barely detectable at (cid:2) 10 nm and starts appearing as the predominant form only when the intensity of com- plex 1 declines at more than (cid:2) 50 nm CI. Complex 1 was estimated to contain 36% of the labeled O1O2 DNA at 46 nm CI (Fig. 2J). Under these conditions, the extent of labeled O1O2 DNA that remained in free form or was retained in complex 2 was determined to be approximately 30%. Using the above data, the cooper- ativity parameter was calculated to be approximately 5 (for details, see Experimental procedures), indicating that binding of CI to O1 causes an approximately five- fold increase of the binding affinity of CI to O2, which is 18 bp away from the former operator (Fig. 3B).
Only 15 bp O1 and O2 interact with CI
To identify the precise location of the repressor binding sites in the primary immunity region of /11 (Fig. 2A), we performed a DNase I footprinting experiment using 200 nm CI and radioactively labeled O DNA (Fig. 2B). The footprints of both the top and bottom strands of O DNA reveal that two regions in O DNA became resis- tant to digestion by DNase I in the presence of CI. More precisely, the )21 to )48 and )52 to )87 regions of the top strand and )24 to )53 and )58 to )87 regions of the bottom strand were protected by CI (Fig. 3B). The centers of these two sites harbor the 15 bp O1 and O2, which are the two putative CI binding sites [15].
sulfate
To confirm that the 15 bp O1 and O2 operators inter- act with CI, we performed the guanine base-specific dimethyl (DMS) protection assay in the presence ⁄ absence of saturating amounts of CI and 32P-labeled O DNA (Fig. 3A). Only the guanine chosen base-specific methylation experiment was because both the operators were found to carry more than one guanine base (Fig. 3B). The results revealed
Previously, we reported that the binding affinity of CI to O1 DNA is slightly higher than that to O2 DNA [15]. To determine the relative affinities of the repressor to O1 and O2 sites more accurately, we again performed gel shift assays using a repressor of better quality and smaller O1 and O2 DNA fragments. As expected, both O1 (Fig. 2C) and O2 (Fig. 2D) yielded one shifted com- plex with increasing CI concentrations. From the plot of
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1978
T. Ganguly et al.
Characterization of /11 repressor
A
cl
cro
O3
O2
O1
O
O1O2
B
O3
[CI] –
+
G + A
G
[CI] – +
O2
O1
C
O1
5 0 1
0 3
0 4
0 5
0 6
0 7
0 0 1
5 2 1
0 5 1
[CI] (nM)
O1
O2
O1
O2
0
5 7
0 0 1
5 2 1
0 5 1
5 7 1
0 0 2
0 0 3
0 0 4
D [CI] (nM)
O2
E
O1
O2
d n u o b
r o t a r e p O %
120 100 80 60 40 20 0
0
75 150 225 300 375 450
Bottom
Top
[CI] (nM)
F
I
0 0 3 1
5 2
0 5
5 7
0 0 1
0 0 1
0 5 2
0 0 5
0 5
0 0 0 1
0
[CI] (nM)
CI (nM)
2 1
*
O1O2
O3
G
0 0 5
0 0 1
0 5 2
0 0 5
0 0 0 1
100
J
CI (nM)
75
2
cspC
f
50
2 O 1 O %
H
0 0 5
0 0 1
0 5 2
0 0 5
1
25
CI (nM)
0
0
15 30 45
60 75 90 105
O
[CI] (nM)
Fig. 2. DNA–protein interaction. (A) A schematic representation of the primary immunity region of /11 (not drawn to scale). The coding regions of cI and cro genes (divergent arrows), the 15 bp O1, O2 and O3 operator sites (gray boxes) in the cI-cro intergenic region, and the different DNA fragments of the immunity region (black horizontal bars), which were utilized in the gel shift or footprint assays, are shown. (B) Autoradiograms of DNase I footprints. O DNA labeled (with 32P) at the top (Top) or bottom (Bottom) strand was incubated with (+) ⁄ with- out ()) 200 nM CI, digested with DNase I and the resulting DNA fragments were resolved through urea ⁄ 6% PAGE. The guanine (G) and adenosine + guanine (A + G) markers were generated from labeled O DNA by standard methods. Locations of the 15 bp O1 and O2 sites within the protected regions are indicated by solid bars. (C, D, F–I) Autoradiograms of different gel shift assays. Each autoradiogram repre- sents the gel shift assay with a specific 32P-labeled DNA (noted in the left bottom corner) and the indicated amounts of CI. All gel shift assays were performed three of four times and only representative data are presented. The arrow and asterisk indicate the shifted complex and contaminating band, respectively. (E) Using the scanned data from the autoradigrams (C, D), plots of percent operator bound versus repressor concentration were generated. Curves O1 and O2 denote the equilibrium binding of CI to O1 and O2 DNA respectively. (J). Coop- erative binding: the operator DNA contents in the shifted complexes 1 and 2 and in the unbound labeled O1O2 DNA were determined by scanning the intensities of all the bands shown in the autoradiogram of the gel shift assay (I) and plotted against the respective repressor concentrations. Curves 1, 2 and f denote the status of O1O2 DNA concentrations in complexes 1 and 2 and in the unbound state. The maxi- mum amount of bound operator in complex 1 was estimated from curve 1. The amounts of operator in complex 2 and in the unbound state at the condition of maximum bound operator in complex 1 were determined from curves 2 and f, respectively. All these values were used for calculation of the cooperativity parameter by a standard method (see Experimental procedures). All curves are best-fit curves.
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1979
T. Ganguly et al.
Characterization of /11 repressor
[CI]
A
B
–140
–130
–120
–110
–100
–90
+
–
[CI] –
+
O3 5'CATTTTCTTACCTCCTTAAATTTACCTATAGTATAACCCAATTATTTTTGGTATTCA GTAAAAGAATGGAGGAATTTAAATGGATATCATATTGGGTTAATAAAAACCATAAGT
cl
* O1 *
*
–80
–70
–60
–50
–40
–30
O1
O2
O2
*
ACAAAAAAATACACGAAAAGCAAACTTTTATGTTGACTCAAGTACACGTATCGTGTAT TGTTTTTTTATGTGCTTTTCGTTTGAAAATACAACTGAGTTCATGTGCATAGCACATA
O2
*
*
*
–20
–10
+1 cro
O1
AGTAGGTTTTGTAAGCGGGAGGTGACAACATG TCATCCAAAACATTCGCCCTCCACTGTTGTAC 5'
*
Bottom
Top
Fig. 3. Interaction of CI with 15 bp operator DNA. (A) Autoradiograms of DMS protection footprints. O DNA labeled at the top (Top) or bottom (Bottom) strand was incubated with (+) ⁄ without ()) 0.25 lM CI followed by treatment of the reaction mixture with DMS as described in the Experimental procedures. Solid bars indicate the locations of O1 and O2 sites. Stars and arrowheads indicate the hypermethylation sites and protected guanine bases, respectively. (B) Summary of different footprinting experiments. Angled lines at the top and bottom of the DNA sequence (cI-cro intergenic region) indicate the DNase I-protected regions. The 15 bp O1 and O2 DNA sequences are surrounded by a solid box, whereas O3 is surrounded by a broken box. The protected guanine bases and hyper-methylated bases detected in the DMS protection experiment are denoted by vertical arrowheads and stars, respectively. The start codons of CI and Cro are indicated by angled arrows. The first base of the start codon of Cro was considered as +1 and the whole sequence was numbered with respect to +1.
[8], A2 [27], /g1e [28], HK022 [6] and N15 [29] bear more than two CI binding sites. Lactococcal phage Tuc2009 [30] and S. aureus /Sa3ms [17], however, bear two CI binding sites, similar to that of /11 in the cI-cro intergenic region. Transcription of k cI mRNA from PRM, which overlaps OR2 and OR3, was shown to be positively regulated by k CI [1]. At very high concentrations, k CI binds to OR3, which in turn inhibits the expression of k cI transcripts. The )35 element of promoter of /11 cI was found to partly overlap with the 15 bp O2 site (data not shown). Taken together, this suggests that the transcription of /11 cI is most possibly regulated by O2 alone and O3 is needed merely to stop the transcription of /11 cI by /11 Cro (which favors the lytic development of /11).
Binding stoichiometry
that the intensities of six bottom strand guanine bases and five top strand guanine bases of O DNA are decreased notably in the presence of CI. The guanines protected by CI correspond to )41G, )43G, )63G, )67G, )74G and )76G (bottom strand) and )33G, )35G, )46G, )56G and )68G (top strand) (Fig. 3B). the protected guanine bases except )56G are All located in and around O1 and O2. Interestingly, )35G, )41G and )43G in O1 and )68G, )74G and )76G in O2 are conserved. The )40G in O1 and )73G in O2, although conserved, most likely do not interact with the CI. The data, however, confirm that 15 bp O1 and O2 DNAs are involved in the binding of CI. The intensities of some top ()53G) and bottom ()36G and )49G) strand guanine bases were also increased nota- bly, suggesting that these bases became more exposed as a result of a conformational change of the operator DNA upon CI binding. The N7 group of guanine, which is methylated by DMS, is exposed in the major groove of the DNA helix [1]. Therefore, the data also suggest that the interaction between CI and the opera- tor DNA may occur through the major groove of the operator DNA helix.
To determine the CI binding stoichiometry precisely, we performed glutaraldehyde-mediated crosslinking experiments with CI in the presence ⁄ absence of varying amounts of O1 DNA. As shown in Fig. 4A, dimeric CI is the predominant form formed in the presence of O1 DNA. Although the tetrameric and hexameric forms of CI (formed without O1 DNA) disappeared, a small amount of monomeric CI reappeared in the presence of O1 DNA. The reason for the presence of
The absence of detectable interaction between O3 and /11 CI (as evident from both the gel shift and footprint assays) is quite unexpected because the pri- mary immunity regions of phages k [1], P22 [12], 434
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1980
T. Ganguly et al.
Characterization of /11 repressor
B
kb
0 0.2 0.3 0.4 0.5 0.6 0.7 0.8 1.0 2.0
– 0
+ 0
+ 4
+ 20
+ 40 M
[CI] (µM)
A GCHO O1 (µM)
85 60
O1
200 150 120 100 70 50 40 30
D
C
120
CTD
CTD
100
80
60
NTD
NTD
40
d n u o b A N D 1 O %
20
0
0
2
3
1 [CI]/[O1 DNA]
Fig. 4. Binding stoichiometry. (A) 10% SDS ⁄ PAGE analysis of glutaraldehyde (GCHO) treated CI. 4 lM CI was incubated with the indicated amount of O1 DNA prior to treatment with (+) ⁄ without ()) GCHO. Protein marker bands and their respective molecular masses (kDa) are shown to the right of the gel. (B) Autoradiogram of the gel shift assay shows the binding of varying concentrations (0.2–2 lM) of CI to a 32P-labeled O1 DNA plus (cid:2) 0.4 lM cold O1 DNA). Using the scanned data from the autoradiogram, fixed amount of O1 DNA mix ((cid:2) 0.1 nM a plot of percent O1 bound versus CI concentration was generated (C). (D) The schematic model structure of the CI–O1 DNA complex, developed as based on our present experimental data, reveals that two NTDs (light gray balls) of dimeric CI are pointed towards two adja- cent major grooves of O1 DNA located on the same face of DNA helix. CI monomers in dimeric CI contact each other through their CTDs (dark gray balls). The G bases that interact with the NTDs of dimeric CI are circled.
operator DNA slightly slowing down the migration of dimeric CI remains unclear at present.
To confirm that dimeric repressor binds to a single operator, we carried out gel shift assays under condi- tions (i.e. using very high CI and O1 DNA concentra- tions) that strongly favor the formation of the CI–O1 complex (Fig. 4B). The corresponding plot of CI bind- ing to O1 DNA, as obtained from quantitation of the gel shift data, is also shown (Fig. 4C). It is apparent that the binding stoichiometry is approximately two CI monomers per O1 DNA. Taken together, the data suggest that, similar to k CI and Cro [8] /11, CI binds to 15 bp operator DNA as a homodimer.
major groove of operator DNA helix (Fig. 3A). The average size of each DNase I-protected region of oper- ator DNA was found to be approximately 25–27 bp (Fig. 2B), suggesting the involvement of at least two adjacent (full) turns of DNA helix in the interaction with /11 CI. Thus, two NTDs of dimeric /11 repres- sor may attain a specific conformation in space for easing the interaction of its two HTH motifs to two adjacent major grooves located on the same face of operator DNA helix (Fig. 4D). The )33G, )35G, )41G and )43G bases of O1 possibly contact CI from the front, whereas )46G may contact from the back of helix. The way that NTD contacts with )46G and other bases on the back of the DNA helix remains unclear at present.
Conclusions
The present study provides valuable insights into the basic structures of /11 CI, its cognate operators and the /11 CI–operator complex, and these are found to be quite similar to those in k and related phage systems. Despite structural relatedness, the mechanism of action of /11 CI does not completely resemble that
The CTDs of k CI [1] and LexA [18] (i.e. the struc- tural homologs of /11 CTD) are involved in the homodimerization of these repressors. Sequence align- ment of /11 CI and LexA revealed that several resi- dues involved in the dimerization of LexA CTD were also present in the CTD of /11 CI (data not shown). The CTDs of two /11 CI monomers may therefore be responsible for the formation of a dimeric /11 CI [15]. By contrast, the NTD of /11 CI, which harbors a potential helix-turn-helix DNA binding motif, could participate in the binding of the dimeric /11 CI to the
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1981
T. Ganguly et al.
Characterization of /11 repressor
the repressor proteins of
the lambdoid phages. of Although k CI requires three operators to regulate the expression of genes flanking the k cI-cro intergenic region, /11 CI possibly requires two operators to regu- late the transcription of genes located on the two sides of the /11 cI-cro intergenic region. Furthermore, the information gathered in the present study may prove useful in the construction of S. aureus-based expression inducer vectors that could be induced by a physical such as temperature.
Experimental procedures
Bacterial and phage strains and plasmids
Biochemical and biophysical analysis of /11 repressor
phy, as described previously [15]. To further purify the /11 repressor, we loaded almost 2.8 mg of repressor (derived from the above affinity chromatography) onto a 40 mL Sephadex G-50 column (diameter 1.5 cm) pre-equilibrated with buffer C [10 mm Tris–Cl¢ (pH 8.0), 200 mm NaCl, 1 mm EDTA, 5% glycerol]. Repressors were eluted at a flow rate of 24 mLÆh)1 in buffer C. Twenty 600 lL frac- tions (marked F1 to F20) were collected and protein esti- mation revealed that only fractions F2 to F8 contained protein. Because fractions F2 and F3 contained mainly intact repressor (discussed below), we stored these fractions on ice until use. The concentration of CI was calculated using the molecular mass of monomeric CI.
the content of respectively,
S. aureus RN4220 [31] and E. coli BL21 (DE3) (Novagen, Madison, WI, USA) cells were routinely grown in Trypti- case soy broth [32] and LB [33], respectively. Growth media were supplemented with appropriate antibiotics if required. The temperate phage /11 and its growth conditions have been described previously [32]. The construction of plasmid pSAU1201 and pSAU1220 was also described previously [15]. The 269 bp /11 DNA insert in pSAU1201 carrying the /11 cI-cro intergenic region was designated as O DNA. Plasmid pSAU1220 was utilized for overexpression of /11 CI in E. coli.
Molecular biological techniques
Homology modeling
Glutaraldehyde-mediated crosslinking, partial proteolysis and recording of CD spectrum of the repressor were car- ried out as described previously [13,15]. Using the molar extinction coefficients for 5-thio-2-nitrobenzoic acid at 412 nm and for CI at 280 nm of 14150 m)1Æcm)1 and 18005 m)1Æcm)1, free sulf- hydryl (-SH) groups in CI in buffer C was determined by DTNB according to a standard procedure [35]. MALDI- TOF analysis of protein fragments was carried out using an Autoflex II TOF ⁄ TOF instrument (Bruker Daltonics, Ettlingen, Germany) according to the manufacturer’s protocol.
Amino acid residues 1–118 and 119–239 of native /11 CI were used to develop 3D model structures of the NTD and CTD of this protein by the First Approach Mode of swiss-model (http://ExPasy.org). Although the crystal structure of E. coli LexA CTD (Protein Databank code: 1jhc) was utilized as a template for developing the model structure of the CTD of /11 CI, the X-ray structure of a putative DNA binding protein (Protein Databank code: 3bs3) of Bacteroides fragilis was used as a template for generating the model structure of NTD of /11 CI. Using the coordinates of the resulting model structures, molecu- lar visualization, superimposition of the structures, surface structure determination and drawing of Ramachandran plots were carried out by the swiss-pdb viewer (http:// ExPasy.org).
Gel shift assay
Overexpression and purification of /11 repressor
Plasmid DNA isolation, DNA estimation, digestion of DNA by restriction enzymes, modification of DNA frag- ments by modifying enzymes, PCR, purification of DNA fragments, labeling of DNA fragments with radioactive materials and agarose gel electrophoresis were carried out following standard procedures [33] or according to the protocols provided by respective manufacturer’s the (Qiagen, Hilden, Germany; Fermentas GmbH, St Leon- Rot, Germany; Bangalore Genei P. Ltd., Bangalore, India). Protein estimation, native and SDS ⁄ PAGE, staining of polyacrylamide gel and western blotting were performed as described previously [13,34]. DNA from /11 phage particles was isolated as described previously [32]. Sequencing of all /11 DNA inserts (amplified by PCR) were performed at the DNA sequencing facility at the University of Delhi, South Campus (Delhi, India). Sequencing of the N-terminal ends of all protein fragments was performed using a protein sequencer (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s protocol.
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1982
investigated by the standard gel /11 CI was overexpressed in E. coli BL21 (DE3) (pSAU1220) and purified by Ni-NTA column chromatogra- Equilibrium binding of CI to various 0.1 nm 32P-labeled DNAs (harboring one ⁄ two /11 operators or no operator) shift assay, as was described previously [15]. The 154 bp O1O2 DNA fragment
T. Ganguly et al.
Characterization of /11 repressor
Table 1. Details of the oligonucleotides used.
Name
Purpose
Sequence (5¢- to 3¢)
GGATCCTAAATCTTCTTGAGTAC
pHC1
Synthesis of O and
O1O2 DNAs
GAATTCTTGGTTCTATAGTATCTG
pHC2 PCR11 GACTCAAGTACACGTATCGTGTATA
Synthesis of O DNA Synthesis of O1DNA
GTAGGTTTA
PCR21 AAACCTACTATACACGATACGTGTA
Synthesis of O1 DNA
CTTGAGTCA
IIa
Synthesis of O2 and
O1O2 DNAs
IIb
Synthesis of O2 DNA
ATTCAACAAAAAAATACACGAAAAG CAAACTTTTATGTTGACTCAAGTA TACTTGAGTCAACATAAAAGTTTGC TTTTCGTGTATTTTTTTGTTGAAT GAATTCTCGCTAATTCTTTTTTATC TTTTTTTGTTGAATACCAAAAATAA
PCI51 IIId
Synthesis of O3 DNA Synthesis of O3 DNA
CSP4
TTGGGTTATACTATAG CATGCCATGGATGAATAACGGTACAG Synthesis of S. aureus
cspC DNA
CTCGAGCATTTTAACTACGTTTG
CSP6
Synthesis of S. aureus
cspC DNA
end-labeled with [c-32P] ATP followed by the PCR amplifi- cation of O DNA by Taq polymerase using pSAU1201 DNA or /11 DNA as a template and the oligonucleotides pHC1 and labeled pHC2 as primers. The resulting DNA fragment was purified from an agarose gel.
DMS protection assay
DNase I footprinting was performed according to a stan- dard procedure [5] with some modifications. Briefly, 60 nm labeled DNA fragment ((cid:2) 5000 c.p.m.) was incubated with varying concentrations of CI in 50 lL buffer C for 20 min on ice. Every reaction mixture was made 1 mm with MgCl2 and treated with 0.15 units of DNase I for 4 min at room temperature followed by termination of the reactions by the addition of 90 lL of Stop solution [200 mm NaCl, 80 mm EDTA (pH 8.0), 1% SDS, 0.03% glycogen]. Cleaved DNA fragments, prepared by sequential passage of each reaction mixture through phenol–chloroform (1 : 1) extraction and ethanol precipitation steps, were resuspended in sequencing [98% deionized formamide, 10 mm EDTA gel buffer (pH 8.0), 0.025% bromophenol blue]. Each labeled DNA was treated with DNase I identically in the absence of CI and the recovered DNA fragments were used as controls. Finally, both experimental and control DNA fragments were analyzed by urea ⁄ 6% PAGE along with guanine and ⁄ or adenosine + guanine sequencing ladders generated from the identically labeled DNA fragments by standard procedures [38].
contained higher CI
(Fig. 2A) was synthesized by PCR using pSAU1201 DNA as a template and primers IIa and pHC1 (Table 1). Similarly, 90 bp O3 (a third putative operator in the /11 cI-cro intergenic region) DNA was amplified using primers PCI15 and IIId and pSAU1201 DNA. On the other hand, 214 bp cspC DNA was amplified using primers CSP4 and CSP6 and the chromosomal DNA of S. aureus Newman as a template [36]. All three DNA fragments were purified from agarose gel using the QIAquick Gel Extraction Kit (Qiagen). The 34 bp O1, and 49 bp O2 DNAs (Fig. 2A) were prepared by mixing and annealing primers PCR11 and PCR21 and IIa and IIb, respectively (Table 1). The cooperativity parameter for the binding of CI to O1O2 DNA was determined from the scanned data of the autora- diogram (Fig. 2I) according to Monini et al. [37]. To study CI binding stoichiometry, a gel shift assay was performed using essentially the same method (see above), except that reaction mixtures concentrations (0.2–2 lm) and (cid:2) 0.4 lm cold O1 DNA along with (cid:2) 0.1 nm 32P-labeled O1 DNA. The CI preparation used in the binding stoichiometry experiment was considered to have 100% activity.
DNase I footprinting
Acknowledgements
The DMS protection assay was performed as described pre- viously [39]. Briefly, 0.5 lm repressor was incubated with 60 nm 32P-labeled O DNA ((cid:2) 5000 c.p.m.) in 100 lL of buffer C for 20 min at room temperature followed by the treatment of repressor–operator complexes with 0.2% DMS for 2 min at room temperature. After termination of the reaction with DMS stop solution [1.5 m sodium acetate (pH 7.0), 1 m beta-mercaptoethanol], DNA was recovered by successive passage of the reaction mixture through phe- nol–chloroform (1 : 1) extraction and ethanol precipitation steps in the presence of glycogen. The same labeled O DNA was also treated directly with DMS as above in the absence of CI and the recovered DNA was used as a control. The gunaine-specific ladder DNAs, prepared from both control and experimental DNAs by a standard procedure [38], were analyzed by urea ⁄ 6% PAGE.
This work was supported by the financial assistance from the Department of Atomic Energy (Government of India, Mumbai, India) to S. Sau. The authors thank Drs P. Parrack, R. Chattopadhyaya and N. C. Mandal for critically reading, correcting and modifying the to manuscript. The authors are extremely grateful
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1983
the oligonucleotide pHC2 Different 32P-labeled DNA fragments, utilized in different footprinting assays, were prepared by standard end labeling procedures [33]. Briefly, to label the bottom strand of O DNA with 32P, pSAU1201 was treated successively with EcoRI, Klenow polymerase and [a-32P] dATP, and BamHI. Finally, the bottom strand labeled O DNA was purified from an agarose gel. To label the top strand of O DNA with 32P, (Table 1) was
T. Ganguly et al.
Characterization of /11 repressor
Bacillus subtilis phage phi 105 repressor. J Biol Chem 264, 14784–14791.
12 Susskind MM & Youderian P (1983) Bacteriophage P22 Antirepressor and its control. In Bacteriophage P22 Antirepressor and its control. Lambda II (Hendrix RW, Roberts JW, Stahl FW & Weisberg RA, eds), pp. 93–121. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 13 Ganguly T, Bandhu A, Chattoraj P, Chanda PK, Das
Dr C. Y. Lee (UAMS, Little Rock, AR, USA) for providing plasmids and strains used in the study. The authors would like to thank Mr A. Banerjee, Mr A. Poddar, Mr J. Guin and Mr M. Das for their excellent technical help. Mr Tridib Ganguly, Ms Mal- abika Das and Mr Amitava Bandhu received Senior Research fellowships from the Council of Scientific and Industrial Research (Government of India, New the Delhi). Mr Palas K. Chanda is a recipient of Senior Research fellowship of Bose Institute. Mr Bisw- anath Jana received a Junior Research fellowship from the Department of Biotechnology (Government of India, New Delhi).
References
M, Mandal NC & Sau S (2007) Repressor of temperate mycobacteriophage L1 harbors a stable C-terminal domain and binds to different asymmetric operator DNAs with variable affinity. Virol J 4, 64.
14 Iandolo JJ, Worrell V, Groicher KH, Qian Y, Tian R, Kenton S, Dorman AH, Ji H, Lin S, Loh P et al. (2002) Comparative analysis of the genomes of the temperate bacteriophages phi 11, phi 12 and phi 13 of Staphylococcus aureus 8325. Gene 289, 109–118.
15 Das M, Ganguly T, Chattoraj P, Chanda PK, Bandhu A, Lee CY & Sau S (2007) Purification and character- ization of repressor of temperate S. aureus phage phi11. J Biochem Mol Biol 40, 740–748.
1 Gussin GN, Johnson AD, Pabo CO & Sauer RT (1983) Repressor and Cro Protein: Structure, Function, and Role in Lysogenization. In Repressor and Cro Protein: Structure, Function, and Role in Lysogenization. Lambda II (Hendrix RW, Roberts JW, Stahl FW & Weisberg RA, eds), pp. 93–121. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 2 Hatfull GF (2000) Molecular Genetics of Mycobacteria. ASM Press, Washington, DC. 16 Kaneko J, Kimura T, Narita S, Tomita T & Kamio Y (1998) Complete nucleotide sequence and molecular characterization of the temperate staphylococcal bacte- riophage phiPVL carrying Panton-Valentine leukocidin genes. Gene 215, 57–67.
3 Heinrich J, Velleman M & Schuster H (1995) The tri- partite immunity system of phages P1 and P7. FEMS Microbiol Rev 17, 121–126.
17 Sumby P & Waldor MK (2003) Transcription of the toxin genes present within the Staphylococcal phage phiSa3ms is intimately linked with the phage’s life cycle. J Bacteriol 185, 6841–6851.
4 Kimsey HH & Waldor MK (2004) The CTX phi repres- sor RstR binds DNA cooperatively to form tetrameric repressor-operator complexes. J Biol Chem 279, 2640– 2647. 5 Koudelka AP, Hufnagel LA & Koudelka GB (2004) 18 Luo Y, Pfuetzner RA, Mosimann S, Paetzel M, Frey EA, Cherney M, Kim B, Little JW & Strynadka NC (2001) Crystal structure of LexA: a conformational switch for regulation of self-cleavage. Cell 106, 585–594.
Purification and characterization of the repressor of the shiga toxin-encoding bacteriophage 933W: DNA bind- ing, gene regulation, and autocleavage. J Bacteriol 186, 7659–7669. 6 Carlson NG & Little JW (1993) Highly cooperative 19 Bell CE, Frescura P, Hochschild A & Lewis M (2000) Crystal structure of the lambda repressor C-terminal domain provides a model for cooperative operator binding. Cell 101, 801–811.
DNA binding by the coliphage HK022 repressor. J Mol Biol 230, 1108–1130.
20 Stayrook S, Jaru-Ampornpan P, Ni J, Hochschild A & Lewis M (2008) Crystal structure of the lambda repres- sor and a model for pairwise cooperative operator bind- ing. Nature 452, 1022–1025. 21 Bohm G, Muhr R & Jaenicke R (1992) Quantitative
7 Ogawa T & Ogawa H (1988) Organization of the early region of bacteriophage 80. J Mol Biol 202, 537–550. 8 Ptashne M (ed.) (1992) A Genetic Switch Gene Control and Phage k. Harvard University, Cell Press & Black- well Scientific Publications, Cambridge, MA. analysis of protein far UV circular dichroism spectra by neural networks. Protein Eng 5, 191–195.
9 Dodd IB, Shearwin KE & Egan JB (2005) Revisited gene regulation in bacteriophage Lambda. Curr Opin Genet Dev 15, 145–152. 10 Rousseau P, Betermier M, Chandler M & Alazard R 22 Hecht MH, Sturtevant JM & Sauer RT (1984) Effect of single amino acid replacements on the thermal stability of the NH2-terminal domain of phage lambda repres- sor. Proc Natl Acad Sci USA 81, 5685–5689. 23 Stearman RS, Frankel AD, Freire E, Liu BS & Pabo
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1984
(1996) Interactions between the repressor and the early operator region of bacteriophage Mu. J Biol Chem 271, 9739–9745. CO (1988) Combining thermostable mutations increases the stability of lambda repressor. Biochemistry 27, 7571–7574. 11 van Kaer L, van Montagu M & Dhaese P (1989) Purifi- cation and in vitro DNA-binding specificity of the
T. Ganguly et al.
Characterization of /11 repressor
24 Watanabe K, Chishiro K, Kitamura K & Suzuki Y
31 Novick RP (1990) The Staphylococcus as a Molecular Genetic System. In Molecular Biology of the Staphylo- cocci (Novick RP, ed.), pp 1–37. VCH Publishers, Inc., New York, NY. 32 Lee CY & Iandolo JJ (1988) Structural analysis of (1991) Proline residues responsible for thermostability occur with high frequency in the loop regions of an extremely thermostable oligo-1,6-glucosidase from Bacillus thermoglucosidasius KP1006. J Biol Chem 266, 24287–24294. 25 Chattopadhyaya R & Ghosh KA (2003) A comparative staphylococcal bacteriophage phi 11 attachment sites. J Bacteriol 170, 2409–2411.
33 Sambrook J & Russell DW (2001) Molecular Cloning: A Laboratory Manual. 3rd ed., Cold Spring Harbor Laboratory Press, CSH, New York, NY. three-dimensional model of the carboxy-terminal domain of the lambda repressor and its use to build intact repressor tetramer models bound to adjacent operator sites. J Struct Biol 141, 103–114.
34 Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA & Struhl K (1998) Current Protocols in Molecular Biology. John Wiley & Sons, New York, NY. 35 Saha RP, Basu G & Chakrabarti P (2006) Cloning, 26 Das M, Ganguly T, Bandhu A, Mondal R, Chanda PK, Jana B & Sau S (2009) Thermally stable Cro repressor of phage /11 possesses novel DNA-binding capacity and physicochemical properties. BMB Rep (in press). 27 Ladero V, Garcı´ a P, Alonso JC & Sua´ rez JE (2002)
expression, purification, and characterization of Vibrio cholerae transcriptional activator, HlyU. Protein Expr Purif 48, 118–125. Interaction of the Cro repressor with the lysis ⁄ lysogeny switch of the Lactobacillus casei temperate bacterio- phage A2. J Gen Virol 83, 2891–2895.
28 Kakikawa M, Ohkubo S, Sakate T, Sayama M, Taketo A & Kodaira K (2000) Purification and DNA-binding properties of the cro-type regulatory repressor protein cng encoded by the Lactobacillus plantarum phage phi g1e. Gene 249, 161–169. 29 Lobocka MB, Svarchevsky AN, Rybchin VN & 36 Chanda PK, Mondal R, Sau K & Sau S (2008) Anti- biotics, arsenate and H2O2 induce the promoter of Staphylococcus aureus cspC gene more strongly than cold. J Basic Microbiol, doi: 10.1002/jobm.200800065. 37 Monini P, Grossman SR, Pepinsky B, Androphy EJ & Laimins LA (1991) Cooperative binding of the E2 pro- tein of bovine papillomavirus to adjacent E2-responsive sequences. J Virol 65, 2124–2130. 38 Maxam AM & Gilbert W (1980) Sequencing end-
labeled DNA with base-specific chemical cleavages. Meth Enzymol 65, 499–560. 39 Ramesh V & Nagaraja V (1996) Sequence-specific DNA
FEBS Journal 276 (2009) 1975–1985 ª 2009 The Authors Journal compilation ª 2009 FEBS
1985
Yarmolinsky MB (1996) Characterization of the pri- mary immunity region of the Escherichia coli linear plasmid prophage N15. J Bacteriol 178, 2902–2910. 30 Kenny JG, Leach S, de la Hoz AB, Venema G, Kok J, Fitzgerald GF, Nauta A, Alonso JC & van Sinderen D (2006) Characterization of the lytic-lysogenic switch of the lactococcal bacteriophage Tuc2009. Virol 347, 434–446. binding of the phage Mu C protein: footprinting analysis reveals altered DNA conformation upon protein binding. J Mol Biol 260, 22–33.
