Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M607785200 on October 17, 2006

J. Biol. Chem., Vol. 281, Issue 51, 39380-39387, December 22, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/51/39380    most recent
M607785200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Qian, X.
Right arrow Articles by Luo, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Qian, X.
Right arrow Articles by Luo, Y.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Calcium Stiffens Archaeal Rad51 Recombinase from Methanococcus voltae for Homologous Recombination*

Xinguo Qian{ddagger}, Yujiong He{ddagger}, Xinfeng Ma{ddagger}, Michel N. Fodje§, Pawel Grochulski{ddagger}, and Yu Luo, Recipient of CIHR new investigator salary award{ddagger}1

From the {ddagger}Department of Biochemistry, University of Saskatchewan, Saskatoon, Saskatchewan S7N 5E5, Canada and the §Canadian Light Source Inc., University of Saskatchewan, Saskatoon, Saskatchewan S7N 0X4, Canada

Received for publication, August 15, 2006 , and in revised form, September 21, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Archaeal RadA or Rad51 recombinases are close homologues of eukaryal Rad51 and DMC1. These and bacterial RecA orthologues play a key role in DNA repair by forming helical nucleoprotein filaments in which a hallmark strand exchange reaction between homologous DNA substrates occurs. Recent studies have discovered the stimulatory role by calcium on human and yeast recombinases. Here we report that the strand exchange activity but not the ATPase activity of an archaeal RadA/Rad51 recombinase from Methanococcus voltae (MvRadA) is also subject to calcium stimulation. Crystallized MvRadA filaments in the presence of CaCl2 resemble that of the recently reported ATPase active form in the presence of an activating dose of KCl. At the ATPase center, one Ca2+ ion takes the place of two K+ ions in the K+-bound form. The terminal phosphate of the nonhydrolyzable ATP analogue is in a staggered conformation in the Ca2+-bound form. In comparison, an eclipsed conformation was seen in the K+-bound form. Despite the changes in the ATPase center, both forms harbor largely ordered L2 regions in essentially identical conformations. These data suggest a unified stimulation mechanism by potassium and calcium because of the existence of a conserved ATPase center promiscuous in binding cations.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
In homologous recombination, a hallmark DNA strand exchange reaction is promoted by RecA-like recombinases between a single-stranded DNA (ssDNA) and a homologous double-stranded DNA (dsDNA). Such a reaction plays critical role in the repair of double-stranded DNA breaks and restart of stalled replication forks (1-5). This recombinase superfamily (6) is composed of bacterial RecA (7), archaeal RadA or Rad51 (8), and eukaryal Rad51 (9), and meiosis-specific DMC1 (10). Large differences in their primary structures exist between bacterial RecAs and nonbacterial orthologues. Their tertiary and quaternary structures, however, are clearly conserved (11-13). Such filamentous assemblies are classic allosteric systems equipped with at least two functional sites, one located at the subunit interface for binding and hydrolyzing ATP and the other located near the filament axis for binding DNA and promoting strand exchange. In vitro studies have clearly shown that optimal reaction conditions differ for the well characterized Escherichia coli RecA (EcRecA) and members of the archaeal/eukaryal group (14-16). EcRecA favors low salt conditions with little or no monovalent cation. Human and yeast Rad51 and DMC1, on the other hand, favor the presence of a salt, typically KCl or (NH4)2SO4. Fishel and coworkers further found that human Rad51 is stimulated by K+ or bigger monovalent cations (17). Using MvRadA2 as a prototype, we have structurally rationalized that monovalent cations stimulate RadA/Rad51/DMC1 by bridging and stabilizing a critical L2 region with the ATP cofactor (18, 19). However, the reported MvRadA structures could not explain the recently discovered stimulatory role by divalent Ca2+ ions on human Rad51 (20) and DMC1 (21). Though initially suspected of being relevant only to higher eukaryotes in sensing the second messenger Ca2+, similar effects have later been found for yeast DMC1 (22). To test whether a similar effect by Ca2+ exists in archaeal orthologues, we carried out assays on MvRadA in the presence of Ca2+. The strand exchange activity but not the ATPase activity of MvRadA is stimulated by Ca2+, suggesting that this archaeal orthologue also possesses a structural element for sensing Ca2+. We further determined the crystal structure of MvRadA in the presence of Ca2+. A largely ordered L2 is anchored by Ca2+ with the ATP cofactor in an essentially identical conformation to that previously elucidated in the K+-activated forms of MvRadA. Analysis of sequence and structure suggests that archaeal/eukaryal recombinases may share a negatively charged cavity lined by ATP, the catalytic Glu, an Asp in the ATP cap, and the C terminus of a short helix in the L2 region. We interpret these results as suggesting a unified stimulation mechanism by various cations because of the existence of such a conserved and promiscuous cation-binding pocket in the ATPase center of closely related RadA/Rad51/DMC1 recombinases.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cloning, Protein Preparation, and Crystallization—RadA from Methanococcus voltae was subcloned into pET28a (Novagen). The recombinant proteins were overexpressed in BL21-CodonPlus (DE3)-RIPL cells (Stratagene) and purified as reported (13, 19). In brief, the purification procedure involved steps of polymin P (Sigma) precipitation, high salt extraction, and three chromatography steps using heparin (Amersham Biosciences), hydroxyapatite (Bio-Rad), and DE52 (Whatman) columns. The RadA protein pooled between 0.3 and 0.4 M NaCl from the DE52 column was concentrated to ~30 mg/ml by ultra-filtration.

Strand Exchange Assay Using Virion DNA—Circular single-stranded {phi}X174 virion DNA (5386 nucleotides) and its homologous double-stranded {phi}X174 replication form I DNA (5386 base pairs) were purchased from New England Biolabs. The double-stranded replication form I DNA was linearized by PstI (VWR) digestion before being used as the dsDNA substrate. The reaction solutions were composed of 50 mM Tris-Hepes buffer at pH 7.4, specified amount of CaCl2 or MgCl2, 5 mM ATP or AMP-PNP. The reaction solution with ATP was supplemented with an ATP regenerating system composed of 4 mM creatine phosphate and 0.01 unit/µl creatine kinase (VWR). The circular single-stranded {phi}X174 virion DNA (5 ng/µl or 15 µM in nucleotides) and MvRadA (5 µM) were first incubated in the reaction buffer at 37 °C for 3 min. Then single-stranded DNA-binding protein from E. coli (VWR) was added to a concentration of 1 µM. After a second incubation at 37 °C for 3 min, the PstI-linearized dsDNA was added to a concentration of 10 ng/µl (15 µM in base pairs). After a third incubation at 37 °C for 90 min, SDS (a final concentration of 1%) and proteinase K (a final concentration of 1 µg/µl, VWR) were added to the reaction solution. A further incubation for 10 min was required to fully degrade MvRadA. A 10-µl sample was mixed with 5 µl of loading buffer composed of 30% glycerol and 0.1% bromphenol blue and loaded onto a 0.6% agarose horizontal gel. The DNA gel electrophoresis was developed for 4 h in a constant electric field of 4 V/cm. The ethidium bromide-stained gel was recorded with a Kodak Gel Logic 200 system. Strand exchange yields were quantified using the Kodak MI software. Strand exchange yields were derived by dividing the integrated intensity of the product band by the average intensity of the dsDNA bands in inactive lanes.

Strand Exchange Assay Using Synthetic Oligonucleotides—A 63-nucleotide oligonucleotide and a homologous 36-bp double-stranded DNA were used as the strand exchange substrates as described previously (19). The solution for strand exchange reaction was composed of 5 mM ATP or an analogous nucleotide, 6 mM MgCl2 or CaCl2, 0.1 M NaCl, 50 mM Hepes-Tris buffer at pH 7.4, 21 µM MvRadA, and 1 µM oligonucleotides. The 63-nucleotide ssDNA substrate was pre-incubated at 37 °C with MvRadA for 1 min before adding the 36-bp dsDNA substrate. The reaction was stopped at 30 min by adding EDTA to a concentration of 20 mM and trypsin to a concentration of 1 µg/µl. After 10 min of trypsin digestion at 37 °C, a 10-µl sample was mixed with 5 µl of loading buffer composed of 30% glycerol and 0.1% bromphenol blue and then loaded onto a 17% acrylamide gel. The SDS-PAGE was developed, stained with ethidium bromide, and digitized with the Kodak Gel Logic 200 system.

Crystallization and Structure Determination—The concentrated MvRadA protein (~30 mg/ml) was crystallized using the hanging drop crystallization method at a room temperature of 21 °C. The optimal well solution for crystallization contained 2 mM AMP-PNP, 0.005 M MgCl2, 0.025 M CaCl2, 0.5 M NaCl, 6% polyethylene glycol 3350, and 0.05 M Tris-HCl buffer at pH 7.2-7.8. Crystals grew to a typical size of 0.15 mm wide and 0.4 mm long in 4 days. AMP-PNP and polyethylene glycol 3350 were purchased from Sigma. The other chemicals used in crystallization were from VWR. Crystals were gradually transferred to soaking solution composed of the crystallization reservoir solution supplemented with 10, 15, 20, and 25% glycerol, soaked for ~5 min, then flash-cooled to 100 K in a nitrogen stream generated by an Oxford Cryosystems device. The 0.4° oscillation images were acquired and processed using a Bruker Proteum R system as described (13). Another crystal was cooled by dipping into liquid nitrogen and data collected at the 08ID-1 beamline at the Canadian Light Source. A total of 360 consecutive images of 1° oscillation and 1 s exposure were acquired. This synchrotron data set was processed using XDS (23). The previously solved MvRadA model (Protein Data Bank code 1XU4) was used as the starting model for rigid body refinement. The model was iteratively rebuilt using XtalView (24) and refined using crystallography NMR software (25). Using the data collected with the in-house Bruker system, a 2.8 Å-resolution anomalous difference map was generated using model-derived phases retarded by 90°. An ordered calcium site near the ATP analogue was located at a major peak (~7 {sigma}) in the anomalous difference map, where most minor peaks (3-4 {sigma}) corresponded to sulfur atoms in Met and Cys residues. Using the synchrotron data set, the calcium site corresponded to a higher density peak (~18 {sigma}) in the anomalous difference map. Some sulfur and phosphorous atoms also corresponded to higher anomalous density peaks up to 6 {sigma}. Statistics of the diffraction data, refinement, and geometry are given in Table 1. The molecular figures were generated using Molscript (26) and rendered using Raster3D (27). The coordinates and structure factors of the synchrotron data set have been deposited in the Protein Data Bank (code 2I1Q).


View this table:
[in this window]
[in a new window]

  		TABLE 1
X-ray crystallographic data and structure refinement statistics

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Ca2+ Stimulates Strand Exchange Activity of MvRadA—We have recently rationalized the stimulating roles on MvRadA played by potassium ions (18, 19). Two K+ ions stabilize an ATPase and strand exchange-active conformation by filling in a negatively charged cavity at the ATPase center. However, potassium and other monovalent ions are not the only recombinase-activating cations known to date. To find out whether a similar effect by divalent Ca2+ ions exists in archaeal orthologues as reported for human and yeast Rad51 and DMC1 recombinases (20-22), we carried out ATPase and strand exchange assays on MvRadA in the presence of divalent cations Ca2+ and Mg2+. Despite screening a number of monovalent cations (18) and divalent cations (Mg2+ and Ca2+, data not shown), K+ remains the only cation that stimulates the ATPase activity of MvRadA. In comparison, other tested monovalent and divalent cations registered less than 5% of the maximal ATPase activity recorded in the presence of 100 mM KCl. The stimulation on the strand exchange activity of MvRadA, on the other hand, appeared to require the presence of ions such as but not limited to K+. Similar to its roles on human and yeast homologues, Ca2+ activates the strand exchange activity of MvRadA (Fig. 1) without drastically boosting ATPase activity (data not shown). In the presence of 5 mM ATP, more of 2 mM CaCl2 was needed for yielding visible quantities of the plasmidsize strand-exchange product (Fig. 1A). A maximal yield of ~8% was observed in the presence of 4 mM CaCl2. It is worth noting that the assay was performed in the absence of potassium, the only previously known stimulant for this protein. We also resorted to shorter DNA substrates, the strand exchange between which does not require the presence of single-stranded DNA-binding protein. Stimulation by Ca2+ was observed in the presence of ATP or AMP-PNP (Fig. 1B). Therefore, it appears that stimulation by Ca2+ on DNA strand exchange is a universal phenomenon shared by RadA/Rad51/DMC1 recombinases.


Figure 1
View larger version (64K):
[in this window]
[in a new window]

  		FIGURE 1.
DNA strand exchange in the presence of Ca2+. A, strand exchange between viron DNA substrates. A simplified strand reaction scheme is shown in the top panel. The circular ssDNA and the linear dsDNA first form a larger joint molecule intermediate (JM, not drawn in the scheme), then form the product of a nicked circular DNA species (NC). Strand exchange reactions were assayed in solutions containing 5 mM ATP or AMP-PNP (shown as ANP), specified concentrations of CaCl2 (in mM) and MgCl2 (in mM, optional), 5 µM MvRadA, 1 µM E. coli single-stranded DNA-binding protein, 5 ng/µl circular ssDNA, and 10 ng/µl linear dsDNA. The agarose gels were stained by ethidium bromide. Average yields of at least three experiments are shown along with S.E. B, strand exchange between synthetic oligonucleotides. A reaction scheme is shown on the left. An ethidium bromide-stained acrylamide gel is shown on the right. ATP{gamma}S is shown as {gamma}-S. Wild-type MvRadA appears stimulated by Ca2+ in the presence of either ATP or AMP-PNP. HD, heteroduplex.

 
Structural Basis for Calcium Stimulation—To pursue a structural elucidation of the mechanism of calcium stimulation, we carried out strand exchange assay in the presence of AMP-PNP, the only ATP analogue amenable to MvRadA crystallization. In the presence of 5 mM AMP-PNP and 2-6 mM CaCl2, MvRadA did promote strand exchange between viron DNAs (Fig. 1A). The yields were similar to those in the presence of ATP. We then attempted crystallizing MvRadA in the presence of AMP-PNP and CaCl2. MgCl2 was found to be an essential ingredient in the production of diffractive crystals, though crystallization could be achieved in its absence. The strand exchange activity in the presence of AMP-PNP and mixtures of MgCl2 and CaCl2 was confirmed (Fig. 1A) before proceeding to final optimization of crystallization conditions and structural determination. The calcium-bound structure was first refined to a resolution of 2.1 Å using an in-house data set (Table 1). Because calcium is the major anomalous scattering element in the crystal, the resulting anomalous difference map was used to locate a Ca2+ species near the {gamma}-phosphate of AMP-PNP (Fig. 2A). A synchrotron data set of 1.9 Å-resolution was acquired later at the new 08ID-1 beamline at the Canadian Light Source, which provided improved clarity in electron density maps (Fig. 2B) and agreed with the in-house data-based assignment of a Ca2+ species. All crystals in this and previous studies of MvRadA belong to space group P61 with similar helical pitches and essentially identical packing schemes. The filament pitch coinciding with the crystallographic c axis is 104.2 Å, 1-2 Å shorter than previously reported forms using the same cooling and in-house diffracting procedures. The Ca2+-bound structure is unexpectedly and strikingly similar to the previously determined K+-bound form of MvRadA (Fig. 3B) (18, 28). The Ca2+-bound form also has an octahedral Mg2+ ion in a position seen in many P-loop-containing GTPases and ATPases. The lack of anomalous signal and the distances (1.9-2.2 Å) from this species to its six-ligand oxygen atoms enabled the Mg2+ assignment. As seen in other crystal forms, AMP-PNP, the nonhydrolyzable ATP analogue is buried between MvRadA promoters (Fig. 3). One subunit (Fig. 3, yellow subunit) contacts the ATP analogue and the octahedral Mg2+ largely through its conserved P-loop (residues Gly-105 to Thr-112) (29) and the base-stacking Arg-158. The adjacent subunit (Fig. 3, gray subunit) contributes the ATP cap (residues Asp-302 to Asp-308) and the C-terminal portion of the L2 region in trans. The putative ssDNA-binding L2 region (residues Asn-256 to Arg-285) has an 8-residue helix (residues Gly-275 to Ala-282) analogous to helix G of EcRecA (30). The side chain of His-280 therein forms a direct hydrogen bond with the {gamma}-phosphate of the ATP analogue. One Ca2+ ion (Fig. 3A, large salmon spheres) located by anomalous signals forms a bridge between the {gamma}-phosphate and three backbone carbonyl moieties at the C terminus of the 8-residue helix. Unlike the Mg2+, seven oxygen ligands of this Ca2+ form a pentagonal bipyramid. The carbonyl O atom of Ala-282 and one O{gamma} atom of the ATP analogue constitute the poles. Three water ligands and two carbonyl O atoms in Gly-279 and His-280 constitute the equatorial pentagon. The Ca2+ ion makes indirect contact with Glu-151 or Asp-302 through two of its three water ligands. A candidate for the hydrolyzing water (Fig. 3A, large green spheres) is hydrogen-bonded with the side chains of Glu-151 and Gln-257 from the subunit contributing the P-loop. The analogous residues of EcRecA (Glu-96 and Gln-194) have been proposed as the catalytic residues based on the EcRecA crystal structure (30) and analogy to GTPases (31), respectively. As expected for a P-loop-containing ATPase, the {gamma}-phosphate contacts the {epsilon}-amino group of Lys-111 and the Mg2+ ion. With the added electron-withdrawing effects by His-280 and the Ca2+ ion, the {gamma}-phosphate is likely further polarized for the nucleophilic attack by the hydrolysis water positioned and activated by Glu-151 and Gln-257. However, the terminal phosphate is in a staggered conformation (Fig. 3A), which may lack the maximal phosphate-phosphate repulsion in the eclipsed conformation seen in the K+-bound ATPase active form (Protein Data Bank code 2FPM, Fig. 3B) (28). This may partially explain why calcium does not stimulate optimal ATPase activity. The L2 region, which is more important for strand exchange activity, is in an essentially identical conformation as seen in the K+-bound form. Both forms lacked interpretable electron densities for a 7-residue segment from Pro-262 to Met-268. In comparison, ~19 residues (residues 260-278) in the L2 region are disordered in the recurrent inactive form of MvRadA crystallized in the presence of ADP instead of AMP-PNP or in the absence of an adequate amount of a stimulatory cation. Because both potassium and calcium ions are elucidated to stabilize a largely ordered L2 conformation, which is correlated with the stimulating roles of the ions in DNA strand exchange, we view these findings as further supporting evidence for the notion that a properly dispositioned L2 region is critical for promoting DNA strand exchange as reviewed by Bell (32).

Mg2+ Stimulates Strand Exchange Activity of MvRadA—The structural comparison between K+- and Ca2+-bound structures implies a promiscuous cavity capable of incorporating varied numbers and sizes of cations. Indeed, the location of the Ca2+ ion in the new crystal form lies between the two K+ ions in the previously reported ATPase-active form (Fig. 3). We further tested the possibility that this cavity could house the physiologically abundant magnesium ion. Although Mg2+ appeared a stronger activator than Ca2+, more than 5 mM MgCl2 was required to promote DNA strand exchange to 50% of the maximal yield (Fig. 4A). In comparison, a lower (2 mM) CaCl2 concentration was required to exhibit ~50% of its maximal stimulatory effect (Fig. 1A). It is generally believed that the P-loop-containing ATPases bind M(II)-ATP chelates. We argue that the binding of K+,Ca2+, or other ions in the structurally characterized cavity occurs after the incorporation of M(II)-ATP. As such, the saturation of the ion-binding cavity would be affected by the concentration of free ions. In the presence of 5 mM ATP, an efficient Mg2+ and Ca2+ chelator, free cation concentrations were estimated by using the MAXCHELATOR software (33). Approximately 0.5-1.2 mM free Mg2+ is present when a total of 5-6 mM MgCl2 is added to half-activate MvRadA. A much lower free Ca2+ concentration of ~0.08 mM corresponds to the half-activating dose of 2 mM CaCl2. We speculate that one or two Mg2+ ions could be bound in the MvRadA cavity but with lower affinity than Ca2+. In the presence of AMP-PNP, Mg2+ did not appear to stimulate MvRadA in promoting strand exchange between viron DNA (Fig. 4A) or oligonucleotides (Fig. 4B). As such, it is not surprising that previously reported MvRadA structures crystallized in the presence of Mg2+ did not reveal a second magnesium ion in contact with the {gamma}-phosphate of AMP-PNP.


Figure 2
View larger version (55K):
[in this window]
[in a new window]

  		FIGURE 2.
Electron density maps in stereo. AMP-PNP (cyan) and residues 275-283 in the L2 region (gray) are shown in stick models. Mg2+ (blue), Ca2+ (salmon), and selected water molecules (green) are shown in spheres. The omit Fo - Fc difference maps (yellow) were contoured at 3 {sigma} after simulated annealing starting at 1000 K. The anomalous difference maps (magenta) were generated using the model-derived phases retarded by 90°. A, electron density maps derived from the in-house data set. The 2.8 Å-resolution anomalous difference map was contoured at 4 {sigma}. B, electron density maps derived from the synchrotron data set. The 2.8 Å-resolution anomalous difference map was contoured at 8 {sigma}. One Ca2+ is located near the terminal phosphate of AMP-PNP.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The archaeal RadA protein from M. voltae is a bona fide RecA orthologue that promotes the hallmark strand exchange reaction between homologous DNAs. Consistent with its sequence similarity with eukaryal Rad51/DMC1 recombinases, the archaeal RadA from M. voltae is similarly activated by K+ and Ca2+ ions. Therefore, MvRadA can serve as a structural model for rationalizing the activation mechanism of its human homologues. An active recombinase DNA filament is in a strand exchange-competent state, which harbors an extended and under-wound form of DNA as compared with B-DNA (34). The assembly of an active filament also requires the binding of Mg2+ and ATP cofactors at the ATPase center of each subunit, which dooms the complex because of the intrinsic ATPase activity unless ATP hydrolysis is blocked or ATP is replenished by ADP/ATP exchange. Based on the crystal structure of E. coli RecA, two axially located loops L1 and L2 have been proposed as the structural elements for binding DNA (30). This structure-based hypothesis has been supported by numerous experiments as reviewed (35). In the crystal structures of MvRadA, two loops equivalent to L1 and L2 of EcRecA are closer to the filament axis than seen in the EcRecA structure (13). Their axial location would be suitable for binding DNA. Besides, RecA-like recombinases are classical allosteric proteins with a signature interplay between its ATPase center and DNA-binding loops. The stimulatory roles of potassium ions on the ATPase and strand exchange activities of MvRadA (18, 19) led to the first revelation that concomitant conformational changes in its ATPase center and L2 region (which encompass the L2 loop and its elbows). Structures determined in the presence of strand exchange stimulant ions K+ and Ca2+ harbor largely ordered L2 regions in a recurrent conformation as compared with a more disordered one in structures determined in the absence of adequate stimulating ions or in the presence of ADP (28). These findings are not only consistent with the notion that L2 is critical for promoting DNA strand exchange but also provide a satisfying explanation to why ATP and Mg2+ cofactors and ionic stimulants are needed (Fig. 5). A cation-bridged interaction between the ATP cofactor and the C-terminal short helix in the L2 region appears essential in the proper disposition of L2 as well as in further polarizing ATP for hydrolysis. The negatively charged nature of the cation-binding pocket is further strengthened by two conserved carboxylates, one from the invariable Asp in the ATP cap and the other from the catalytic Glu. As the short helix points its C terminus toward this pocket, electrostatic repulsion would destabilize such a L2 disposition without cationic bridging. Therefore, the K+- and Ca2+-bound structures offer a plausible explanation for the cationic activation on archaeal and eukaryal recombinases. Yet it astonishes that MvRadA and perhaps its eukaryal homologues as well is capable of incorporating both monovalent and divalent cations of various radii. The ~1 Å-shortening of each helical 6-subunit rise of the Ca2+-bound MvRadA filament compared with the K+-bound one amounts to a ~0.2 Å smaller gap between adjacent subunits where the stimulating ions are bound. This would partially offset the ~ 0.3 Å smaller ionic radius of Ca2+. The still wider-than-ideal gap is compensated by seven, one more than usual, oxygen ligands of the Ca2+ to further satisfy the valence rule (36). The ATP in contact with the ions also switches from the eclipsed conformation in the K+-bound form to the staggered conformation in the Ca2+-bound form. Though direct Ca2+-carboxylate contacts with both Glu-151 and Asp-302 are infeasible, the stabilizing effects by these two conserved residues are retained in forming hydrogen bonds with the hydration shell of Ca2+. This conformational variability of the ATPase center allows for promiscuous binding of cations while stabilizing a recurrent L2 disposition for promoting proper interactions with DNA substrates and hence strand-exchange. In the absence of a stimulatory cation or in the aftermath of ATP hydrolysis, L2 becomes disordered (28). This scenario of order-disorder transition appears recurrent in similar ATPase complexes such as the motor in the bacterial Sec protein translocation machinery (37). Also because of the subtle yet noticeable changes in the ATPase center, the similarly polarized ATP, either by one Ca2+ or two K+, may sense enough disparity in its surroundings for efficient hydrolysis. Though unexpected, crystal structures of MvRadA appear to suggest a unified activation mechanism by monovalent and divalent cations on this archaeal recombinase. In light of the findings that such archaeal and eukaryal recombinases share a high degree of similarity in ionic activation consistent with their resemblance in amino acid sequence, it would not be surprising that they share this conserved structural element for cation sensing. In particular, as pointed out by Bugreev and Mazin (20), sensing Ca2+ as a second messenger could play a regulatory role on recombinase activity in higher eukaryotes.


Figure 3
View larger version (54K):
[in this window]
[in a new window]

  		FIGURE 3.
ATPase center of MvRadA in stereo. The view is rotated from that in Fig. 2 by ~90°. Two MvRadA subunits are in yellow and gray, respectively. Ca2+ and K+ ions are in salmon and red, respectively. Mg2+ ions and water molecules are blue and green, respectively. The putative hydrolysis water is highlighted in a larger sphere. A, Ca2+-bound form. B, previously reported K+-bound form (Protein Data Bank entry code 2FPM). The L2 regions are in essentially identical conformations in both forms. Ca2+ and K+ are bound in a conserved cavity lined by the tip of the nucleotide cofactor, the catalytic Glu-151, Asp-302 in the ATP cap, and the C terminus of a short helix in the L2 region. The terminal phosphate is in a staggered conformation in the Ca2+-bound form but in an eclipsed conformation in the K+-bound form.

 
Though we speculate that the primary Ca2+, which anchors the L2 elbow, is likely to play a major role in triggering the large scale conformational change, it is prudent to notice two additional calcium ions located by anomalous signals. Neither is in contact with the flexible L2 loop. The second Ca2+ is partially enclosed by the 2'- and 3'-hydroxyls of the bound AMP-PNP, the carboxylate of Asp-308, and the carbonyl of Tyr-301. In a previously reported inactive filament in which L2 is largely disordered, this cavity is occupied (Protein Data Bank entry code 2FPL) by a potassium ion. Intake of a cation at this site does not appear to be correlated with the ordering in the L2 region. Nevertheless, cations such as potassium and calcium may play a role in the recognition of the ribose moiety of the nucleotide cofactor. The third Ca2+ replaces an Mg2+/Mn2+ seen in all previously reported MvRadA structures. This cation species is liganded by Asp-246. We made a D246A mutant MvRadA,3 the crystal structure of which no longer harbors a cation near Ala-246. However, this D246A protein was observed to be proficient in promoting DNA strand exchange. Therefore, it is unlikely calcium binding near Asp-246 would play a major role in activating MvRadA by triggering the observed conformational change in the L2 region.


Figure 4
View larger version (39K):
[in this window]
[in a new window]

  		FIGURE 4.
DNA strand exchange in the presence of Mg2+. A, strand exchange between viron DNA substrates. Strand exchange reactions were assayed in solutions containing 5 mM ATP or AMP-PNP (shown as ANP), specified concentrations of MgCl2 (in mM), 5 µM MvRadA, 1 µM E. coli single-stranded DNA-binding protein, 5 ng/µl circular ssDNA, and 10 ng/µl linear dsDNA. Average yields of at least three experiments in the presence of ATP are shown along with S.E. B, strand exchange between synthetic oligonucleotides. The wild-type MvRadA appears activated by Mg2+ in the presence of ATP but not in the presence of AMP-PNP.

 
In light of recent development in eukaryal Rad51/DMC1 proteins, activation of DNA strand exchange appears correlated with K+ or Ca2+-triggered transition from inactive ringshaped oligomers to active nucleoprotein filaments (16, 22, 38). On the other hand, two recurrent conformations have been observed in the crystal structures of MvRadA. In the presence of ADP or in the absence of adequate cationic activator, the L2 region is disordered, suggesting that this form resembles an inactive post-hydrolysis intermediate (28). The disordered feature of this form implies that its L2 region is not fully engaged in interacting with DNA. As such, direct conversion of ringshaped oligomers into inactive nucleoprotein filaments is unlikely. The crystallized filaments in the presence of canonical magnesium and nucleotide co-factors as well as calcium or high concentration of potassium harbor a largely ordered L2 region and may resemble the active filament previously seen by imaging techniques. The ordered L2 and L1 regions form a continuous axial groove (Fig. 5) in this seemingly active form implies that their stabilizing interactions with DNA are optimal for the ring-to-filament transition to occur. Archaeal RadA and eukaryal Rad51/DMC1 proteins share a conserved anionic cage highlighted by the invariable Glu-151/Asp-302. By filling this cavity, counter ions such as K+ or Ca2+ would stabilize this conformation and hence triggering increased affinity for DNA and assembly of active nucleoprotein filaments.


Figure 5
View larger version (21K):
[in this window]
[in a new window]

  		FIGURE 5.
Ca2+-triggered conformational change in the L2 region in stereo. The views cover 12 consecutive subunits of MvRadA or two helical turns of the crystallized filaments. The filament axes are shown as vertical lines. The ordered residues in the L1 regions (residues 218-230) and L2 regions (residues 256-285) are shown in green and gold, respectively. The ATP analogues, Mg2+ ions and Ca2+ ions are shown in cyan, blue, and magenta, respectively. A, structure in the absence of Ca2+. The structure of Protein Data Bank entry 1T4G is shown (13). B,Ca2+-bound form. Binding of a Ca2+ ion in every ATPase center triggers a disorder-order transformation in the DNA-interacting L2 region. The largely ordered structure in the presence of stimulatory K+ and Ca2+ ions appears correlated with strand exchange activity of MvRadA.

 

   FOOTNOTES
 
The atomic coordinates and structure factors (code 2I1Q) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

* This work was supported in part by the Natural Sciences and Engineering Research Council of Canada (NSERC), the Saskatchewan Health Research Foundation (SHRF), and the Canadian Institutes of Health Research (CIHR) operating grants (to Y. L.). Some of the research described in this paper was performed at the Canadian Light Source, which is supported by NSERC, National Research Council (NRC), CIHR, and the University of Saskatchewan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 To whom correspondence should be addressed: Dept. of Biochemistry, University of Saskatchewan, A3 Health Sciences Bldg., 107 Wiggins Rd., Saskatoon, Saskatchewan S7N 5E5, Canada. Tel.: 306-966-4379; Fax: 306-966-4390; E-mail: Yu.Luo{at}usask.ca.

2 The abbreviations used are: MvRadA, RadA recombinase from M. voltae; EcRecA, RecA recombinase from E. coli; ATP{gamma}S, adenosine 5'-O-(thiotriphosphate); AMP-PNP, adenosine 5'-(beta,{gamma}-imino)triphosphate; ssDNA, single-stranded DNA; dsDNA, double-stranded DNA. Back

3 X. Qian, Y. He, and Y. Luo, unpublished data. Back


   ACKNOWLEDGMENTS
 
We thank Drs. Gabriele Schatte and Wilson Quail for assistance with the x-ray facility at the Saskatchewan Structural Sciences Centre. We also thank beam line staff of the Canadian Light Source.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Cox, M. M. (1998) Genes Cells 3, 65-78[Abstract]
  2. Cox, M. M., Goodman, M. F., Kreuzer, K. N., Sherratt, D. J., Sandler, S. J., and Marians, K. J. (2000) Nature 404, 37-41[CrossRef][Medline] [Order article via Infotrieve]
  3. Courcelle, J., Ganesan, A. K., and Hanawalt, P. C. (2001) BioEssays 23, 463-470[CrossRef][Medline] [Order article via Infotrieve]
  4. Lusetti, S. L., and Cox, M. M. (2002) Annu. Rev. Biochem. 71, 71-100[CrossRef][Medline] [Order article via Infotrieve]
  5. Kowalczykowski, S. C. (2000) Trends Biochem. Sci. 25, 156-165[CrossRef][Medline] [Order article via Infotrieve]
  6. Seitz, E. M., and Kowalczykowski, S. C. (2000) Mol. Microbiol. 37, 555-560[CrossRef][Medline] [Order article via Infotrieve]
  7. Clark, A. J., and Margulies, A. D. (1965) Proc. Natl. Acad. Sci. U. S. A. 53, 451-459[Free Full Text]
  8. Sandler, S. J., Satin, L. H., Samra, H. S., and Clark, A. J. (1996) Nucleic Acids Res. 24, 2125-2132[Abstract/Free Full Text]
  9. Shinohara, A., Ogawa, H., and Ogawa, T. (1992) Cell 69, 457-470[CrossRef][Medline] [Order article via Infotrieve]
  10. Bishop, D. K., Park, D., Xu, L., and Kleckner, N. (1992) Cell 69, 439-456[CrossRef][Medline] [Order article via Infotrieve]
  11. VanLoock, M. S., Yu, X., Yang, S., Lai, A. L., Low, C., Campbell, M. J., and Egelman, E. H. (2003) Structure (Camb.) 11, 187-196
  12. Conway, A. B., Lynch, T. W., Zhang, Y., Fortin, G. S., Fung, C. W., Symington, L. S., and Rice, P. A. (2004) Nat. Struct. Mol. Biol. 11, 791-796[CrossRef][Medline] [Order article via Infotrieve]
  13. Wu, Y., He, Y., Moya, I. A., Qian, X., and Luo, Y. (2004) Mol. Cell 15, 423-435[CrossRef][Medline] [Order article via Infotrieve]
  14. Rice, K. P., Eggler, A. L., Sung, P., and Cox, M. M. (2001) J. Biol. Chem. 276, 38570-38581[Abstract/Free Full Text]
  15. Liu, Y., Stasiak, A. Z., Masson, J. Y., McIlwraith, M. J., Stasiak, A., and West, S. C. (2004) J. Mol. Biol. 337, 817-827[CrossRef][Medline] [Order article via Infotrieve]
  16. Sehorn, M. G., Sigurdsson, S., Bussen, W., Unger, V. M., and Sung, P. (2004) Nature 429, 433-437[CrossRef][Medline] [Order article via Infotrieve]
  17. Shim, K. S., Schmutte, C., Yoder, K., and Fishel, R. (2006) DNA Repair 5, 718-730[Medline] [Order article via Infotrieve]
  18. Wu, Y., Qian, X., He, Y., Moya, I. A., and Luo, Y. (2005) J. Biol. Chem. 280, 722-728[Abstract/Free Full Text]
  19. Qian, X., He, Y., Wu, Y., and Luo, Y. (2006) J. Mol. Biol. 360, 537-547[CrossRef][Medline] [Order article via Infotrieve]
  20. Bugreev, D. V., and Mazin, A. V. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 9988-9993[Abstract/Free Full Text]
  21. Bugreev, D. V., Golub, E. I., Stasiak, A. Z., Stasiak, A., and Mazin, A. V. (2005) J. Biol. Chem. 280, 26886-26895[Abstract/Free Full Text]
  22. Lee, M. H., Chang, Y. C., Hong, E. L., Grubb, J., Chang, C. S., Bishop, D. K., and Wang, T. F. (2005) J. Biol. Chem. 49, 40980-40984
  23. Kabsch, W. (1993) J. Appl. Crystallogr. 26, 795-800[CrossRef]
  24. McRee, D. E. (1999) J. Struct. Biol. 125, 156-165[CrossRef][Medline] [Order article via Infotrieve]
  25. Brunger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., Grosse-Kunstleve, R. W., Jiang, J. S., Kuszewski, J., Nilges, M., Pannu, N. S., Read, R. J., Rice, L. M., Simonson, T., and Warren, G. L. (1998) Acta Crystallogr. Sect. D Biol. Crystallogr. 54, 905-921[CrossRef][Medline] [Order article via Infotrieve]
  26. Kraulis, P. (1991) J. Appl. Crystallogr. 24, 946-950[CrossRef]
  27. Bacon, D. J., and Anderson, W. F. (1988) J. Mol. Graph. Model. 6, 219-220
  28. Qian, X., Wu, Y., He, Y., and Luo, Y. (2005) Biochemistry 44, 13753-13761[CrossRef][Medline] [Order article via Infotrieve]
  29. Saraste, M., Sibbald, P. R., and Wittinghofer, A. (1990) Trends Biochem. Sci. 15, 430-434[CrossRef][Medline] [Order article via Infotrieve]
  30. Story, R. M., Weber, I. T., and Steitz, T. A. (1992) Nature 355, 318-325[CrossRef][Medline] [Order article via Infotrieve]
  31. Kelley, J. A., and Knight, K. L. (1997) J. Biol. Chem. 272, 25778-25782[Abstract/Free Full Text]
  32. Bell, C. E. (2005) Mol. Microbiol. 58, 358-366[CrossRef][Medline] [Order article via Infotrieve]
  33. Patton, C., Thompson, S., and Epel, D. (2004) Cell Calcium 35, 427-431[CrossRef][Medline] [Order article via Infotrieve]
  34. Egelman, E. H. (2001) J. Mol. Biol. 309, 539-542[CrossRef][Medline] [Order article via Infotrieve]
  35. McGrew, D. A., and Knight, K. L. (2003) Crit. Rev. Biochem. Mol. Biol. 38, 385-432[CrossRef][Medline] [Order article via Infotrieve]
  36. Nayal, M., and Di Cera, E. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 817-821[Abstract/Free Full Text]
  37. Keramisanou, D., Biris, N., Gelis, I., Sianidis, G., Karamanou, S., Economou, A., and Kalodimos, C. G. (2006) Nat. Struct. Mol. Biol. 13, 594-602[CrossRef][Medline] [Order article via Infotrieve]
  38. Ristic, D., Modesti, M., van der Heijden, T., van Noort, J., Dekker, C., Kanaar, R., and Wyman, C. (2005) Nucleic Acids Res. 33, 3292-3302[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/51/39380    most recent
M607785200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Qian, X.
Right arrow Articles by Luo, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Qian, X.
Right arrow Articles by Luo, Y.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH