Role of N and C-terminal Tails in DNA Binding and Assembly in Dps: Structural Studies of Mycobacterium smegmatis Dps Deletion Mutants

doi:10.1016/j.jmb.2007.05.004

J. Mol. Biol. (2007) 370, 752–767

Role of N and C-terminal Tails in DNA Binding and Assembly in Dps: Structural Studies of Mycobacterium smegmatis Dps Deletion Mutants Siddhartha Roy 1 , Ramachandran Saraswathi 1 , Surbhi Gupta 1 K. Sekar 2 , Dipankar Chatterji 1 and M. Vijayan 1 ⁎ 1

Molecular Biophysics Unit, Indian Institute of Science, Bangalore 560012, India 2

Bioinformatics Centre, Indian Institute of Science, Bangalore 560012, India

Mycobacterium smegmatis Dps degrades spontaneously into a species in which 16 C-terminal residues are cleaved away. A second species, in which all 26 residues constituting the tail were deleted, was cloned, expressed and purified. The first did not bind DNA but formed dodecamers like the native protein, while the second did not bind to DNA and failed to assemble into dodecamers, indicating a role in assembly also for the tail. In the crystal structure of the species without the entire C-terminal tail the molecule has an unusual open decameric structure resulting from the removal of two adjacent subunits from the original dodecameric structure of the native form. A Dps dodecamer could assemble with a dimer or one of two trimers (trimer-A and trimer-B) as intermediate. Trimer-A is the intermediate species in the M. smegmatis protein. Estimation of the surface area buried on trimerization indicates that association within trimer-B is weak. It weakens further when the Cterminal tail is removed, leading to the disruption of the dodecameric structure. Thus, the C-terminal tail has a dual role, one in DNA binding and the other in the assembly of the dodecamer. M. smegmatis Dps also has a short N-terminal tail. A species with nine N-terminal residues deleted formed trimers but not dodecamers in solution, unlike wild-type M. smegmatis Dps, under the same conditions. Unlike in solution, the Nterminal mutant forms dodecamers in the crystal. In native Dps, the Nterminal stretch of one subunit and the C-terminal stretch of a neighboring subunit lock each other into ordered positions. The deletion of one stretch results in the disorder of the other. This disorder appears to result in the formation of a trimeric species of the N-terminal deletion mutant contrary to the indication provided by the native structure. The ferroxidation site is intact in the mutants. © 2007 Elsevier Ltd. All rights reserved.

*Corresponding author

Keywords: DNA-binding protein; N and C-terminal tails; ferroxidation; quaternary assembly

Introduction Oxidative stress is a universal phenomenon experienced by both aerobic and anaerobic organ-

Abbreviation used: Dps, DNA-binding protein from stationary phase cells. E-mail address of the corresponding author: [email protected]

isms. Reactive oxygen species are generated during the stress, which can damage most cellular components, including proteins, lipids and DNA.1,2 Naturally, organisms have evolved defense mechanisms to prevent oxidative damage. In prokaryotic systems, DNA-binding protein from stationary phase cells (Dps) forms an important component of the mechanisms.3 Free iron catalyzes the formation of free radicals via the Fenton reaction.4 The excess free iron is converted from the ferrous form to the ferric form by Dps and

0022-2836/$ - see front matter © 2007 Elsevier Ltd. All rights reserved.

753

Assembly and DNA Binding in Dps Mutants

sequestered, and it binds non-specifically to DNA and condenses it into a compact structure, thereby protecting the DNA from the onslaught of free radicals.5,6 This dual action provides maximum protection to DNA. Dps is known to be produced maximally during the stationary phase of bacterial growth.7 Dps is now known to exist in different types of bacteria.8 The three-dimensional structures of many of them have been determined.9–22 The subunit in all of them is similar in structure to that of ferritin and is an elaboration of a four-helix bundle motif.9 However, unlike ferritin, which is composed of 24 subunits,23,24 Dps is a dodecamer with 12 subunits located at the corners of a distorted icosahedron with 23 symmetry.10 The hollow core of the dodecamer, with net negative charge on the inner lining, is the location of iron mineralization.9 The ferroxidation center is well characterized and conserved in Dps from different species, and is located at the interface between two subunits.10 Not all of them bind DNA but, when they do, the binding takes place on the outer surface of the molecule as in single-stranded DNA-binding proteins, 25–27 and nucleosomes. 28 The protein lacks any of the known conventional DNA-binding motifs. In Dps molecules, such as that from Escherichia coli (EcDps), which are known to bind to DNA, a lysine-rich, N-terminal tail located on the surface of the dodecamer is believed to be responsible for DNA binding. We reported earlier the structure of Dps from Mycobacterium smegmatis (MsDps) in three crystal forms.10 Subsequently, structural studies on the protein have been reported by another laboratory.29 MsDps has a short N-terminal tail and a long C-terminal tail that has been shown to be involved in DNA binding. The latter is substantially ordered in the crystal structure and located on the surface of the molecule.10 On the basis of the crystal structure, a plausible mode of DNA sequestering was suggested.10 Biochemical experiments had suggested more than one mode of assembly of the dodecamer for the protein from different species. On the basis of a comparative study involving Dps molecules of known three-dimensional structure, it was possible to elucidate the structural basis of the different modes of assembly. Here again, the Cterminal stretch appeared to have a significant role. As in the case of the C-terminal tail, the short Nterminal tail in MsDps does not occur in Dps from many other sources, and appears to have a role in assembly. Therefore, it was important to investigate the roles of N and C-terminal tails in DNA binding, as well as the assembly of MsDps. Here, we report an attempt in this direction using two C-terminal deletion mutants of the protein, one involving the deletion of 16 residues (MsDpsΔC16) and the other involving the deletion of 26 residues (MsDpsΔC26), and an N-terminal deletion mutant involving nine residues (MsDpsΔN9), particularly with the help of the crystal structures of MsDpsΔC26 and MsDpsΔN9.

Results and Discussion Biochemical characterization Native MsDps is trimeric on purification, but converts to a DNA-binding dodecameric form when incubated overnight at 37 °C.30 Both forms exhibit ferroxidation ability. Incubation of the protein at 4 °C for an extended period of time resulted in a species with a subunit molecular mass of 18,542 Da instead of the original molecular mass of 21,658 Da. The degraded species failed to show any signal upon Western blot analysis with anti-His antibody, unlike the full-length protein, and it did not bind to a Ni-NTA column whereas the full-length protein did. These data and the comparison of molecular mass indicated that the degraded protein was produced by the removal of the 16 C-terminal residues through cleavage of the peptide bond between Gly167 and Gln168, concurrently with the formation of dodecamers. Peptide bonds involving Asn and Gln residues adjacent to Gly are known to be more susceptible to cleavage because of a spontaneous deamination reaction through the formation of a transient succinimidyl derivative.31,32 The protein with 16 terminal residues removed (MsDpsΔC16) did not show DNA binding activity, but retained the dodecameric structure. The DNA retardation assay for this mutant as well as for the native protein is illustrated by Figure 1. Comparison with other bacterial species (Figure 2) indicates that the C-terminal tail of MsDps begins at residues 158 soon after the fourth helix in the subunit

Figure 1. DNA retardation assay illustrating the abrogation of DNA binding ability on truncation of the C-terminal tail. Free pUC19 DNA (lane 1), incubated with MsDpsΔC16 (lane 2) and the full-length Ms-Dps (lane 3) at 37 °C for 30 min in 50 mM Tris–HCl (pH 7.9), 150 mM NaCl at a DNA to protein molar ratio of 1:10.3

754

Assembly and DNA Binding in Dps Mutants

Figure 2. Multiple sequence alignment of MsDps with other members of known structure in the Dps family. Ec, Escherichia coli (PDB code 1DPS); Hs, Halobacterium salinarum (1MOJ); At, Agrobacterium tumefaciens (1O9R); Ss, Streptococcus suis (2BW1); Tp, Treponema pallidum (2FJC); Te, Thermosynechococcus elongatus (2C41); Lm, Listeria monocytogenes (2IY4); Li, Listeria innocua (1QGH); Bb, Bacillus bravis (1N1Q); Sa, Staphylococcus aureus (2D5K); Ba, Bacillus anthracis (1JI5, 1JIG); Hp, Helicobacter pylori (1JI4); Dr, Deinococcus radiodurans (2C2F). Ferroxidation (♦) and cleavage (∇) sites are shown.

Assembly and DNA Binding in Dps Mutants

(Figure 3). A species in which 26 C-terminal residues starting with 158 deleted (MsDpsΔC26) was therefore cloned and expressed with a His 6 tag. MsDpsΔC26 did not bind DNA, but exhibited ferroxidation activity. Gel-filtration chromatography indicated clearly that this deletion mutant exists as a trimer before incubation (Figure 4). The eluted protein gave a single band on native PAGE, confirming that it is a single species. The mutant did not oligomerize further under conditions in which MsDps formed dodecamers. However, after an extensive search, different conditions for oligo-

755 merization were identified. However, even under these conditions, MsDpsΔC26 adopts a series of aggregation states instead of a unique dodecamer as native MsDps does (Figure 5). A nine residue tail precedes the first helix in the subunit (Figure 3). This tail is not a common feature of Dps from all species, although it is unusually long and is involved in metal binding in the protein from Deinococcus radiodurans.16,18 A mutant with nine N-terminal residues deleted (MsDpsΔN9) was cloned and expressed. The mutant does not bind DNA and exists in solution only as a trimer.

Figure 3. Structure of a subunit of MsDps (ribbon diagram). The five different helices, α1(10-41), α2(46-76), α3(102128), α4(131-155) and α5(82-90) are shown in different colors. The N-terminal and C-terminal loops, and loops connecting helices are colored green.

756

Assembly and DNA Binding in Dps Mutants

Figure 4. Gel-permeation chromatography of (a) full-length protein and (b) MsDpsΔC26 mutant on a Pharmacia Superdex 2009HR(10/ 30) column in 50 mM Tris–HCl (pH 7.9), 150 mM NaCl at a flow-rate of 0.3 ml/min.The column was calibrated as described in Materials and Methods. The inset in (a) shows the calibration curve.

Dodecamerization was observed in its crystals. The molecules in a solution prepared by dissolving the crystals failed to reconstitute into dodecamers and remained as trimers (discussed later). An extensive search for conditions for further aggregation was unsuccessful. The essential properties of MsDpsΔN9 in solution and those of native MsDps, MsDpsΔC16 and MsDpsΔC26 are given in Table 1. Crystal structures of the deletion mutants: overall features Biochemical experiments indicated that, although MsDpsΔC16, as expected, does not bind DNA, it assembles in the same way as the native protein does. MsDpsΔC26 is incapable of binding DNA, and is unable to assemble like the full-length protein. Therefore, the crystal structure of MsDpsΔC26 was determined at a resolution of 3.3 Å, which is not as high as in the case of the cubic MsDps structure, but

is good enough to answer the specific questions on DNA binding and assembly. Solution studies indicated that MsDpsΔN9 forms trimers, contrary to indications that it should be a dimer obtained from the structure of the native protein (Table 2). It exhibits ferroxidase activity, but does not assemble into dodecamers under conditions in which the native protein does. It does not bind DNA. Thus, biochemical studies on MsDpsΔN9 pose interesting questions. The structure of this protein was determined at 2.53 Å resolution. Each subunit of the deletion mutant in both structures has nearly the same geometry as that of the subunits in the native structure. The rms deviation in Cα positions between pairs of MsDps ΔC26 subunits, when they are superposed, varies between 0.5 Å and 0.8 Å. The corresponding range in MsDpsΔN9 is 0.4–0.5 Å. A similar range of deviations was observed when the subunits of these structures were superposed on a subunit of native

757

Assembly and DNA Binding in Dps Mutants

Table 2. Surface area (Å2) buried on oligomerisation in native Dps and the deletion mutants

MsDps MsDps without 16 C-terminal residues MsDps without 26 C-terminal residues MsDps without 9 N-terminal residues MsDpsΔC26 MsDpsΔN9

Dimer

Trimer-A

Trimer-B

1350 (801)

1601 (1082)

967 (652)

1350 (801)

1601 (1082)

967 (652)

1350 (801)

1315 (879)

787 (476)

1331 (790) 1120 (740) 1277 (760)

869 (547) 835 (405) 838 (410)

966 (510) – 812 (402)

The non-polar component is given in parentheses.

Figure 5. Native 10% (w/v) polyacrylamide gel electrophoresis of MsDpsΔC26 at 4 C (lane 1) and MsDpsΔC26 after incubation with 50 mM cadmium chloride at 42 °C (lane 2).

MsDps. Naturally, the 26 C-terminal residues do not exist in the structure of MsDpsΔC26. However, surprisingly to start with, the ten N-terminal residues are not defined in the structure, whereas they are in native MsDps. Equally surprisingly, the electron density for the N-terminal nine residues and that for the 26 C-terminal residues are absent from the crystals of MsDpsΔN9. Unexpected quaternary association The MsDpsΔC26 oligomer has an unusual decameric structure that resembles the native dodecameric structure only partly . Three nearly orthogonal views of the decamer are shown in Figure 6. It has an open structure resulting from the removal of a dimeric unit from the dodecameric native structure, which is a distorted icosahedron.10 This removal does not leave the rest of the structure unaffected. As shown in Figure 7, substantial movements occur in regions in the neighborhood of the original location of the removed dimer. The native dodecamer may be thought of as consisting of a top hexamer and a bottom hexamer. The decamer in MsDpsΔC26 results when a dimer from, say the top hexamer, is removed. The bottom hexamer in native Dps and Table 1. Biochemical characterization of Dps and its mutants Oligomeric state of the protein in solution Protein

Native state

Msdps Trimer MsDpsΔC16 Trimer MsDpsΔC26 Trimer MsDpsΔN9

Trimer

After incubation Dodecamer Dodecamer Higher heterogeneous oligomers Trimer

DNA- Ferroxidation binding activity Yes No No

Yes Yes Yes

No

Yes

MsDpsΔC26 have identical structures. The four remaining subunits in MsDpsΔC26 move away from each other relative to their original positions in the top hexamer. As mentioned with reference to native MsDps, the subunits could form 2-fold symmetric dimers, and the dimers could then assemble into dodecamers with 23 symmetry. It is possible also that 3-fold symmetric trimers are formed first and then assemble into dodecamers. Two types of trimers are possible, referred to for convenience here as trimer-A and trimer-B. In a 23 symmetric dodecamer, the two types of trimers occur, as they do in native MsDps, at the opposite ends of a 3-fold axis. In an assembled state, the dimeric and the two trimeric arrangements co-exist and the course of assembly is not immediately obvious from the structure. It is interesting to see how the three possible arrangements survive in the decameric MsDpsΔC26. As can be seen from Figure 8, the dimeric arrangement remains intact, and so does trimer-A, but trimer-B opens up and is no longer a stable entity. Native MsDps first assemble into trimers and the trimers then assemble into a dodecamer. There are Dps molecules in other species that first assemble into dimers and then into dodecamers.10 On the basis of inter-subunit interactions, particularly surface area buried on association (Table 2), in the crystal structure of MsDps, it was demonstrated that trimer-A is the most stable of the three species in the M. smegmatis protein. The surface area buried still favors trimer-A when 16 C-terminal residues are deleted from the observed structure. Surface area buried on dimerisation remains unaffected by the removal of 26 C-terminal residues from the experimentally determined MsDps structure. The surface area per subunit buried on the formation of the dimer and trimer-A are nearly the same. That on the formation of trimer-B becomes substantially low and it is not surprising that this trimeric arrangement breaks up in the crystal structure of MsDpsΔC26. Rather unexpectedly, however, the surface area buried on the formation of trimer-A is substantially lower than that on the formation of the dimer in the crystal structure of MsDpsΔC26. As indicated later, this anomaly was addressed successfully using modeling.

758

Assembly and DNA Binding in Dps Mutants

Figure 6. Crystal structure of MsDpsΔC26: (a) top view; (b) and (c) side views. Subunits are colored differently.

Assembly and DNA Binding in Dps Mutants

759

Figure 6 (legend on previous page)

Figure 7. Superposition of ten subunits of MsDpsΔC26 (blue) on corresponding subunits of native MsDps (red). Arrows indicate deviation in structure.

760

Assembly and DNA Binding in Dps Mutants

Figure 8 (legend on next page)

Assembly and DNA Binding in Dps Mutants

761

Figure 8. CPK representation of the decameric MsDpsΔC26 molecule showing the arrangement of subunits in (a) a dimer, (b) trimer-A, and (c) trimer-B in MsDpsΔC26.

Unlike the case of MsDpsΔC26, MsDpsΔN9 forms dodecamers, very similar to those of native MsDps, in its crystal structure. In fact, the symmetry of the molecule is made use of in the crystal with only three subunits in the asymmetric unit. It is not entirely clear why these dodecamers are not seen in solution. Despite extensive searches, the conditions for dodecamer formation in solution are yet to be found. Alternatively, a high concentration, as in the crystal, facilitates dodecamerization. As expected, the surface area buried on dimerization was substantially higher than that buried on trimerization. Thus, the crystal structure does not provide an immediate answer as to how trimers instead of dimers are formed in solution. This problem was again addressed through modeling. Disorder in terminal stretches and its implications for assembly In the structure of native MsDps, the C-terminal tail of one subunit and the N-terminal stretch of another subunit in trimer-A interact, as illustrated in Figure 9. The interactions between the two subunits primarily involve residues 1–4 of one and residues 157,158,161,163, and 166 of the other and they are mainly van der Waals in nature. When

the C-terminal tail is deleted, as in MsDpsΔC26, or the N-terminal stretch is removed, as in MsDpsΔN9, this can no longer happen and the two peptide segments lose the ability to anchor in a single position. That appears to be the reason for the disorder in the N-terminal segment in MsDpsΔC26 and that in the C-terminal segment in MsDpsΔN9. The disorder in the two segments and its implications for assembly were explored using modeling. The Met-Thr-Ser-Phe-Thr-Iso-Pro-Gly-Leu sequence of the N-terminal stretch indicated that a possible source of disorder is the conformational flexibility about Gly8. The N-terminal heptapeptide was rotated using all the allowed values of Φ and Ψ at Gly8 at 10° interval. Each of the resulting structures was then energy minimized. In each case, the total energy of the system containing the two subunits and the N-terminal segment, as well as the interaction energy of the segment with the neighboring subunit and the surface area buried on the association of the two subunits, were calculated (Table 3A). The disorder of the Cterminal stretch was a more complex problem. The stretch is 28 residues long and 18 of them are ordered in native MsDps. This 18 residue stretch has a convoluted structure, which need not necessarily be retained when the stretch is allowed

762

Assembly and DNA Binding in Dps Mutants

Figure 9. Interaction of the N-terminal stretch (blue) of one subunit with the C-terminal stretch (hot pink) of another subunit related by 3-fold symmetry in native MsDps.

to move. However, in order to approach the problem at a proof of concept level, this convoluted structure was retained in the calculations. Furthermore, it was decided to explore the conformations resulting from rotations at Gly160, which is followed by another glycine in the sequence. The calculations were similar to those performed in the case of the N-terminal stretch. The results of the calculations are given in Table 3B. For purposes of comparison, the two subunits in MsDps with ordered terminal segments were also energy minimized. The movements of the segments in them were marginal. Also minimized was the same system without the N-terminal segment and that without the C-terminal segment. Again the movements were small. The results of these calculations are given in Table 3C. Energy values obtained from calculations need to be treated with caution. However, the values presented in Table 3 provide some useful insights. In the C-terminal deletion mutant, in some of the possible minimum energy conformations of the N-terminal stretch indicated in Table 3A, the surface area buried in each subunit is comparable to that in the native structure. This can be seen

also in the distribution of minimum energy conformations illustrated in Figure 10(a). That provides a plausible explanation for the occurrence of the trimeric species of MsDpsΔC26, as other interactions such as hydrogen bonds are comparable in the interfaces of the dimer and trimer-A. The results in relation to the N-terminal deletion mutant are even more interesting. In the absence of the N-terminal stretch, the position of the C-terminal segment in the native structure appears to become energetically unstable. Therefore, it appears to move to other low-energy conformations of the type illustrated in Figure 10(b) and described in Table 3B. When the C-terminal segment is in the original position, the energy of its interaction with the neighboring subunit in trimer-A is zero, and the surface area buried in each subunit is less than 1000 Å2 (869 Å2). When it moves to energetically more favored positions consequent to the deletion of the N-terminal segment, the interaction energy with the neighboring subunit and the surface area buried per subunit increase substantially. That appears to explain the occurrence of trimers of N-terminal mutants.

763

Assembly and DNA Binding in Dps Mutants Table 3. Parameters pertaining to minimum energy conformations Conformation at Gly8

Phi

Psi

Interaction energy of the stretch Surface area Total with the buried in each interaction neighboring subunit in energy subunit trimer-A (kcal/mol) (kcal/mol) interface(Å2)

A. When the position of the N-terminal stretch is varied 99 67 −1397 −13 64 −101 −1423 −51 72 −92 −1415 −31 72 −100 −1423 −50 73 −133 −1374 −7 83 −129 −1399 −29 85 −125 −1389 −14 89 −119 −1381 −12 94 −127 −1422 −43 113 −120 −1411 −35 114 −121 −1409 −18 162 −114 −1419 −29 176 −165 −1394 −13 −81 93 −1410.9 −28 −1413.9 −26 −141 61 −69 −177 −1396.8 −18 −70 −49 −1396.4 −13 −107 −88 −1404.6 −22 −120 −173 −1391.6 −13 −137 −135 −1426.8 −28

1260 1749 1314 1708 1040 1491 1154 1198 1642 1431 1208 1339 1156 1423 1256 1506 1228 1314 1320 1294

Conformation at Gly160 B. When the C-terminal stretch is varied 67 32 −1500.4 77 33 −1490.5 77 112 −1473.6 78 121 −1580.0 80 35 −1502.0 94 24 −1497.1 66 −115 −1475.5 78 −111 −1484.2 82 −117 −1473.6 −1477.7 85 −120 103 −16 −1503 115 −126 −1470.4 118 −101 −1473.4 121 −127 −1471.6 −62 −74 −1472.1 −67 −65 −1466.1 −73 −87 −1466.2 −76 −76 −1465.7 −83 −103 −1463.5 −85 −56 −1466.7 −85 −113 −1466.4 −85 −123 −1462.0 −1461.4 −100 −69 −103 −134 −1463.2 −139 −109 −1466.0 −165 −135 −1461.3

−36.749 −30.383 −19.999 −26.928 −37.507 −37.680 −19.825 −28.529 −21.838 −21.502 −37.840 −20.768 −7.715 −20.326 −20.972 −13.058 −12.997 −11.571 −8.915 −10.084 −13.901 −10.069 −7.955 −11.056 −12.381 −6.964

1451 1539 1396 1516 1461 1483 1392 1480 1357 1332 1418 1301 1102 1264 1322 1195 1181 1194 1057 1112 1184 1105 1033 1142 1180 1027

C. When the native structure and its C and N-terminal deletion mutants are minimized separately 1650 Native Dps −1509 −22, −49a C-terminal −1498 −25 1373 deletion 838 N- terminal −1415 0.0 deletion a

The appropriate stretch is at the C terminus in an N-terminal deletion and vice versa.

Ferroxidation center The ferroxidation center in MsDps lie at the intersubunit interface of the dimer and is composed of His39, Asp66, and Glu70.10 The dimeric arrangement is intact and the residues concerned occupy the original positions in MsDpsΔC26 and MsDpsΔN9 deletion mutants. However, the position of iron in the native structure is vacant in the structures of the mutants, which are presumably unable to retain iron. Conclusion Crystallographic and biochemical studies on the C-terminal deletion mutant of MsDps demonstrated conclusively the role of C-terminal tail in DNA binding. The tail plays a major role also in the assembly of the dodecameric molecule. The removal of the last 16 residues results only in loss of DNA binding activity. However, the deletion of the remaining ten residues of the tail, in addition, seriously disrupts the assembly of the molecule and leads to an open decamer instead of a closed dodecamer. However, the deletion does not affect the ability of the subunits to form trimers to start with. The open decamer retains ferroxidase activity but, understandably, loses the ability to mineralize. The short N-terminal tail also is important in assembly. It holds the long Cterminal stretch in position. It is unclear if DNAbinding dodecamers can be formed in its absence. Its deletion leads to disorder of the C-terminal stretch, which results in a trimeric species of the N-terminal deletion mutant.

Materials and Methods Purification and characterization The wild-type Ms-Dps, MsDpsΔC26 and MsDpsΔN9 were purified as described.30 In brief, E. coli strain BL21 DE3 was transformed with respective vectors pETmsdps, pRSET-msdpsΔC26 and pET-msdpsΔN9. Single-step purification was performed using the Qiagen Ni-NTA affinity matrix according to the manufacturer's instructions. After checking the purity of the proteins by SDS-PAGE (12% (w/v) polyacrylamide gel), proteins were dialyzed against 50 mM Tris–HCl(pH 7.9), 150 mM NaCl. All mass spectra were recorded in positive ion mode on an HP 100 series, electron spray ionization mass spectrometer, equipped with a conventional single quadruple. Optimal nebulization/desolvation of ions was achieved using a nebulizer pressure of 15 bars (1 bar = 105 Pa). Ions were scanned from 700–2300 m/z at a scan rate of 1.8 cycles/s. A minimum of five ions, well above the base-line, were taken from deconvolution with Chemstation software (Hewlett Packard). Deconvolution of the mass spectrum of the native protein showed a single prominent peak. That of MsDpsΔC16 gave a major peak at the expected position, in addition to a few minor peaks probably due to heterogeneity in degradation over time. Gel-permeation chromatography

764

Assembly and DNA Binding in Dps Mutants

Figure 10. Ensemble of low energy conformers of (a) the N-terminal stretch obtained by variation of Φ and Ψ at Gly8 and (b) the C-terminal stretch obtained by variation of the angles at Gly160. Adjacent subunits are colored differently. The modeled positions of the disordered loops are in magenta. See the text for details.

765

Assembly and DNA Binding in Dps Mutants

of Ms-Dps and the mutant MsDpsΔC26 was done on Pharmacia Superdex 200 HR (10/30) column. The buffer used was 50 mM Tris–HCl (pH 7.9), 150 mM NaCl. Standard markers of known molecular mass, horse spleen ferritin (450 kDa), rabbit muscle aldolase (158 kDa), BSA (63 kDa) and myoglobin(17.6 kDa) were run independently under identical buffer conditions. Apparent molecular mass was determined from a plot of log molecular mass versus Kav: Kav ¼ ðVe  V0 Þ=ðVt  V0 Þ where Ve is the elution volume, V0 is the void volume, and Vt is the total bed volume of the column. Native PAGE, gel-retardation assay for DNA binding and spectroscopic analysis of ferroxidation (iron incorporation) were done as they were in the case of MsDps.30 In the conditions under which MsDpsΔC26 oligomerized, a solution of 45 μM MsDpsΔC26 in 10 mM Hepes (pH 7.9), 150 mM NaCl was exposed to the transition temperature (54 °C) for 5 min followed by cooling to room temperature in the presence of 50 mM cadmium chloride. The incubation was continued at 42 °C for 8 h, resulting in the conversion of the trimeric form to heterogeneous oligomeric forms. These were subsequently analyzed by native PAGE (10% polyacrylamide gel) using a discontinuous buffer system at a constant current of 16 mA.33x Crystallization MsDpsΔC26 crystals were grown by the hangingdrop, vapor-diffusion method at 293 K. Each droplet was made by mixing equal volumes of the protein and the reservoir solutions. The concentration of the protein solution was 10 mg/ml in 30 mM Tris–HCl (pH 7.9). The reservoir solution was 0.1 M Tris–HCl (pH 8.5),30% (w/v) PEG-1140, 100 mM CaCl2. Plate-like crystals appeared within two weeks and reached maximum dimensions of 0.4 mm × 0.3 mm × 0.1 mm. Crystals of MsDpsΔN9 were obtained by the microbatch method. A 1:1 (v/v) ratio of paraffin oil and silicone oil was used. A 2 μl sample of 10 mg/ml protein in 50 mM Tris–HCl (pH 8.0) was mixed with 2 μl of reservoir solution (0.1 M Hepes-Na (pH-7.5), 0.2 M CaCl2, 28% v/v polyethylene glycol 400). Crystals suitable for X-ray diffraction experiment were obtained after three days. Data collection and processing Data were collected using an MAR 345 imaging plate mounted on a Rigaku Ultrax-18 rotating-anode X-ray generator. In the case of MsDpsΔC26, data were collected at room temperature (293 K), whereas data from MsDpsΔN9 were collected at 100 K. In both cases, the diffraction pattern was indexed using program DENZO and the intensities were scaled and merged using SCALEPACK from the HKL program package.34 Intensities were converted into structure-factor amplitudes using the program TRUNCATE from the CCP4 suite.35 Data statistics are presented in Table 4.

Table 4. Crystal data, and data collection, refinement and model statistics

A. Crystal data Space group Unit cell parameters a b c (Å) β (deg.) VM (Å3 Da−1) Solvent content (%, v/v) Number of subunits in the a.u. Resolution range (Å) No. measurements No. unique reflections Completeness of data (%) Rmergea (%) B. Refinement and model statistics No. reflections used R (%) Rfree (%) No. protein atoms RMS deviation from ideal Bond lengths (Å) Bond angles (deg.) Dihedral angles (deg.) Improper angles (deg.) Ramachandran plot statistics Residues in the Core regions (%) Allowed regions (%) Generously allowed regions (%) Disallowed regions (%)

MsDpsΔC26

MsDpsΔN9

P21

P6222

96.5 83.9 111.9 105.9 2.5 49.8 10

87.4 87.4 212.5

30.0–3.3 (3.42–3.30) 102,443 25,976 (2467) 99.2 (95.1) 10.8 (33.1)

30.0–2.53 (2.62–2.53) 2,491,497 15,936 (1411) 94.7 (86.5) 9.8 (49.7)

25,975 22.4 25.9 11,607

15,897 22.9 24.6 3543

0.012 1.25 20.4 0.9

0.007 1.20 19.0 0.8

84.9 13.8 1.2

91.6 7.4 1.0

0.1

0.0

2.0 37.5 3

Values for the highest resolution shell are given in parentheses. a.u., asymmetric unit. a Rmerge = ∑|I(k) − bIN|∑|I(k)|, where I(k) is the kth intensity measurement of a reflection, bIN is the average intensity value of that reflection, and the summation is over all measurements.

subunit of MsDps (PDB code 1VEI) containing 157 amino acid residues (without the C-terminal stretch) as the search model did not lead to solution. A model consisting of only the four-helix bundle, accounting for 73% of the polypeptide chain, resulted in a solution consisting of ten subunits in the asymmetric unit. The structure of MsDpsΔN9 was solved uneventfully by molecular replacement using AMoRe37 with a subunit of native Dps as the search probe. Both the structures were subjected to rigid body, positional, torsion angle dynamics and group B-factor refinement using CNS 1.1.38 Density for the N-terminal stretch in MsDpsΔC26 and the 26 C-terminal stretch in MsDpsΔN9 were not seen in the electron density maps. Due to poor resolution in the case of the MsDpsΔC26 deletion mutant, only group B-factor refinement was performed. Individual B-factors were refined in the case of MsDpsΔN9. The final refinement statistics are given in Table 4. COOT was used for model building,39 and the final refinement was done using REFMAC-5.40

Structure solution and refinement Modeling The structure of MsDpsΔC26 was solved by molecular replacement using maximum-likelihood methods implemented in program PHASER v.1.2.36 The use of a

The various conformations of the terminal tails about Gly8 and Gly160 were generated by rotating Φ and Ψ at

766 10° interval using INSIGHTII (Biosym Technologies, 1992). There are 36 possibilities for each Φ and Ψ, and 1996 combinations were generated. All conformations resulting in a short contact of less than 1.8 Å or no contact of the tail (either N terminus or C terminus) with a neighboring subunit were excluded from further calculation. The remaining models were subjected to energy minimization using X-plor 3.85.41 A distance-dependent dielectric constant was used throughout. Structural analysis and comparisons The model was validated using program PROCHECK.42 Structural superposition was done with the program ALIGN43 and accessible buried surface area was calculated using NACCESS† employing all default parameters. Interatomic contacts were calculated using program CONTACT from the CCP4 program suite. The Figures were prepared using PYMOL‡. Protein Data Bank accession codes The atomic coordinates and structure factors of the N and C-terminal deletion mutants of M. smegmatis Dps have been deposited in the RCSB Protein Data Bank with accession codes 2YW6 and 2YW7.

Acknowledgements The data sets used in the present work were collected at the X-ray facility for the Structural Biology at the Institute, supported by the Department of Science & Technology (DST). Computations were performed at the Super Computer Education and Research Centre of the Institute and the Bioinformatics Centre and Graphics facility, both supported by the Department of Biotechnology (DBT). We acknowledge the help of Ms Rakhi P. Chowdhury in performing the ferroxidation assay. M.V. is supported by a Distinguished Biotechnologist Award from the DBT. R.S is a CSIR fellow.

References 1. Farr, S. B. & Kogoma, T. (1991). Oxidative stress responses in Escherichia coli and Salmonella typhimurium. Microbiol. Rev. 55, 561–585. 2. Storz, G., Tartaglia, L. A., Farr, S. B. & Ames, B. N. (1990). Bacterial defenses against oxidative stress. Trends Genet. 6, 363–368. 3. Almiron, M., Link, A. J., Furlong, D. & Kolter, R. (1992). A novel DNA-binding protein with regulatory and protective roles in starved Escherichia coli. Genes Dev. 6, 2646–2654.

† http://wolf.bms.umist.ac.uk/naccess/ ‡ http://pymol.sourceforge.net/

Assembly and DNA Binding in Dps Mutants

4. Haber, F. & Weiss, J. (1934). The catalytic decomposition of hydrogen peroxide by iron salts. Proc. Roy. Soc. ser A, 147, 332–351. 5. Martinez, A. & Kolter, R. (1997). Protection of DNA during oxidative stress by the nonspecific DNAbinding protein Dps. J. Bacteriol. 179, 5188–5194. 6. Ferguson, G. P., Creighton, R. I., Nikolaev, Y. & Booth, I. R. (1998). Importance of RpoS and Dps in survival of exposure of both exponential- and stationary-phase Escherichia coli cells to the electrophile N-ethylmaleimide. J. Bacteriol. 180, 1030–1036. 7. Azam, A. T., Iwata, A., Nishimura, A., Ueda, S. & Ishihama, A. (1999). Growth phase-dependent variation in protein composition of the Escherichia coli nucleoid. J. Bacteriol. 181, 6361–6370. 8. Ohniwa, L. R., Morikawa, K., Kim, J., Ohta, T., Ishihama, A., Wada, C. & Takeyasu, K. (2006). Dynamic state of DNA topology is essential for genome condensation in bacteria. EMBO J. 25, 5591–5602. 9. Grant, R. A., Filman, D. J., Finkel, S. E., Kolter, R. & Hogle, J. M. (1998). The crystal structure of Dps, a feritin homolog that binds and protects DNA. Nature Struct. Biol. 5, 294–303. 10. Roy, S., Gupta, S., Das, S., Sekar, K., Chatterji, D. & Vijayan, M. (2004). X-ray analysis of Mycobacterium smegmatis Dps and a comparative study involving other Dps-like molecules. J. Mol. Biol. 18, 1103–1113. 11. Ilari, A., Stefanini, S., Chiancone, E. & Tsernoglou, D. (2000). The dodecameric ferritin from Listeria innocua contains a novel intersubunit iron-binding site. Nature Struct. Biol. 7, 38–43. 12. Papinutto, E., Dundon, W. G., Pitulis, N., Battistutta, R., Montecucco, C. & Zanotti, G. (2002). Structure of two iron binding proteins from Bacillus anthracis. J. Biol. Chem. 277, 15093–15098. 13. Zanotti, G., Papinutto, E., Dundon, W. G., Battistutta, R., Seveso, M., DelGiudice, G. et al. (2002). Structure of the Neutrophil-activating protein from Helicobacter pylori. J. Mol. Biol. 323, 125–130. 14. Ren, B., Tibbelin, G., Kajino, T., Asami, O. & Ladenstein, R. (2003). The multi-layered structure of Dps with a novel di-nuclear ferroxidase centre. J. Mol. Biol. 6, 467–477. 15. Ceci, P., Ilari, A., Falvo, E. & Chiancone, E. (2003). The Dps protein of Agrobacterium tumefaciens does not bind to DNA but protects it toward oxidative cleavage: X-ray crystal structure, iron binding, and hydroxyl-radical scavenging properties. J. Biol. Chem. 278, 20319–20326. 16. Kim, S. G., Bhattacharyya, G., Grove, A. & Lee, Y. H. (2006). Crystal structure of Dps-1, a functionally distinct Dps protein from Deinococcus radiodurans. J Mol Biol. 361, 105–114. 17. Zeth, K., Offermann, S., Essen, L. O. & Oesterhelt, D. (2004). Iron-oxo clusters biomineralizing on protein surfaces: structural analysis of Halobacterium salinarum DpsA in its low- and high-iron states. Proc. Natl Acad. Sci. USA, 101, 13780–13785. 18. Romão, C. V., Mitchell, E. P. & McSweeney, S. (2006). The crystal structure of Deinococcus radiodurans Dps protein (DR2263) reveals the presence of a novel metal centre in the N terminus. J. Biol. Inorg. Chem. 11, 891–902. 19. Franceschini, S., Ceci, P., Alaleona, F., Chiancone, E. & Ilari, A. (2006). Antioxidant Dps protein from the thermophilic cyanobacterium Thermosynechococcus elongatus. FEBS J, 273, 4913–4928.

Assembly and DNA Binding in Dps Mutants

20. Gauss, G. H., Benas, P., Wiedenheft, B., Young, M., Douglas, T. & Lawrence, C. M. (2006). Structure of the DPS-like protein from Sulfolobus solfataricus reveals a bacterioferritin-like dimetal binding site within a DPS-like dodecameric assembly. Biochemistry, 45, 10815–10827. 21. Kauko, A., Haataja, S., Pulliainen, A. T., Finne, J. & Papageorgiou, A. C. (2004). Crystal structure of Streptococcus suis Dps-like peroxide resistance protein Dpr: implications for iron incorporation. J. Mol. Biol. 338, 547–558. 22. Stillman, T. J., Upadhyay, M., Norte, V. A., Sedelnikova, S. E., Carradus, M. et al. (2005). The crystal structures of Lactococcus lactis MG1363 Dps proteins reveal the presence of an N-terminal helix that is required for DNA binding. Mol Microbiol. 57, 1101–1112. 23. Carrondo, M. A. (2003). Ferritins, iron uptake and storage from the bacterioferritin viewpoint. EMBO J. 22, 1959–1968. 24. Chasteen, N. D. & Harrison, P. M. (1999). Mineralization in ferritin: an efficient means of iron storage. J. Struct. Biol. 126, 182–194. 25. Raghunathan, S., Kozlov, A. G., Lohman, T. M. & Waksman, G. (2000). Structure of the DNA binding domain of E.coli SSB bound to SSDNA. Nature Struct. Biol. 7, 648–652. 26. Matsumoto, T., Morimoto, Y., Shibata, N., Kinebuchi, T., Shimamoto, N., Tsukihara, T. & Yasuoka, N. (2000). Roles of functional loops and the C-terminal segment of a single-stranded DNA binding protein elucidated by X-ray structure analysis. J. Biochem. 127, 329–335. 27. Saikrishnan, K., Jeyakanthan, J., Venkatesh, J., Acharya, N., Sekar, K., Varshney, U. & Vijayan, M. (2003). Structure of Mycobacterium tuberculosis singlestranded DNA-binding protein. Variability in quaternary structure and its implications. J. Mol. Biol. 331, 385–393. 28. Luger, K., Mader, A. W., Richmond, R. K., Sargent, D. F. & Richmond, T. J. (1997). Crystal structure of the nucleosome core particle at 2.8 Å resolution. Nature, 389, 251–260. 29. Ceci, P., Ilari, A., Falvo, E., Giangiacomo, L. & Chiancone, E. (2005). Reassessment of protein stability, DNA binding, and protection of Mycobacterium smegmatis Dps. J. Biol. Chem. 280, 34776–34785.

767 30. Gupta, S. & Chatterji, D. (2003). Bimodal protection of DNA by Mycobacterium smegmatis DNA-binding protein from stationary phase cells. J. Biol. Chem. 278, 5235–5241. 31. Creighton, T. E. (1993). Proteins: Structures and Molecular Properties. W.H. Freeman and Company, New York. 32. Wearne, S. J. & Creighton, T. E. (1989). Effect of protein conformation on rate of deamidation: ribonuclease A. Proteins: Struct. Funct. Genet. 5, 8–12. 33. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685. 34. Otwinowski, Z. & Minor, W. (1997). Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326. 35. Collaborative Computational Project Number 4 (1994). The CCP4 suite: programs for protein crystallography. Acta Crystallog. sect. D, 50, 760–763. 36. Storoni, L. C., McCoy, A. J. & Read, R. J. (2004). Likelihood enhanced fast rotation functions. Acta Crystallog. sect. D, 60, 432–438. 37. Navaza, J. (1994). AMoRe: an automated package for molecular replacement. Acta Crystallog. sect. A, 50, 445–449. 38. Brunger, A. T., Adams, P. D., Clore, G. M., Delano, W. L., Gross, P., Grosse-Kunstleve, R. W. et al. (1998). Crystallography and NMR system: a new software suite for macromolecular structure determination. Acta Crystallog. sect. D, 54, 905–921. 39. Emsley, P. & Cowtan, K. (2004). Coot: model-building tools for molecular graphics. Acta Crystallog. sect. D, 60, 2126–2132. 40. Murshudov, G. N., Vagin, A. A. & Dodson, E. J. (1997). Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallog. sect. D, 53, 240–255. 41. Brünger, A. (1992). XPLOR 3.1: a system for X-ray crystallography and NMR. Yale University Press, New Haven. 42. Laskowski, R. A., Moss, D. S. & Thornton, J. M. (1993). Main-chain bond lengths and bond angles in protein structures. J. Mol. Biol. 231, 1049–1067. 43. Cohen, G. E. (1997). ALIGN: a program to superimpose protein coordinates, accounting for insertions and deletions. J. Appl. Crystallog. 30, 1160–1161.

Edited by R. Huber (Received 27 January 2007; received in revised form 2 May 2007; accepted 3 May 2007) Available online 10 May 2007