doi:10.1016/j.jmb.2007.03.079 J. Mol. Biol. (2007) 369, 1244–1257
Dynamic Structural Rearrangements BetweenDNA Binding Modes of E. coli SSB Protein
Rahul Roy1, Alexander G. Kozlov2, Timothy M. Lohman2⁎and Taekjip Ha1,3,4⁎
1Center for Biophysics andComputational Biology,University of Illinois,Urbana-Champaign,IL 61801, USA2Department of Biochemistryand Molecular Biophysics,Washington University Schoolof Medicine, St. Louis,MO 63110, USA3Department of Physics,University of Illinois,Urbana-Champaign,IL 61801, USA4Howard Hughes MedicalInstitute, Urbana,IL 61801, USA
Abbreviations used: SSB, single-stssDNA, single-stranded DNA; FRETresonance energy transfer; PEG, polE-mail addresses of the correspon
Escherichia coli single-stranded (ss)DNA binding (SSB) protein binds ssDNAin multiple binding modes and regulates many DNA processes via protein-protein interactions. Here, we present direct evidence for fluctuationsbetween the twomajor modes of SSB binding, (SSB)35 and (SSB)65 formed on(dT)70, with rates of interconversion on time scales that vary as much as 200-fold for a mere fourfold change in NaCl concentration. Such remarkableelectrostatic effects allow only one of the two modes to be significantlypopulated outside a narrow range of salt concentration, providing a contextfor precise control of SSB function in cellular processes via SSB expressionlevels and interactions with other proteins. Deletion of the acidic C terminusof SSB, the site of binding of several proteins involved in DNA metabolism,does not affect the strong salt dependence, but shifts the equilibriumtowards the highly cooperative (SSB)35 mode, suggesting that interactionsof proteins with the C terminus may regulate the binding mode transitionand vice versa. Single molecule analysis further revealed a novel lowabundance binding configuration and provides a direct demonstration thatthe SSB-ssDNA complex is a finely tuned assembly in dynamic equilibriumamong several well-defined structural and functional states.
The presence of extensive amounts of single-stranded DNA (ssDNA) in the cell is often a signof trouble, as ssDNA is an obligate intermediate inrecombination-mediated DNA repair. Single-stranded DNA binding (SSB) proteins are essentialin all organisms and bind selectively to ssDNAindependent of sequence.1,2 They also modulate thefunctions of many DNA processing enzymes eithervia protein–protein interactions or by controllingaccessibility to ssDNA.1–5 While it is now wellestablished that SSB proteins can display a multi-plicity of modes of interaction with DNA,5,6 thedynamic nature of the transitions among thesemodes is yet to be explored. We have therefore ex-
randed DNA binding;, fluorescencey-ethyleneglycol.ding authors:iuc.edu
lsevier Ltd. All rights reserve
amined the dynamic interconversion among theseSSB/DNA complexes at the single molecule level.The Escherichia coli SSB protein forms a stable
homotetramer, with each of its 19 kDa subunitspossessing an oligonucleotide/oligosaccharide bind-ing (OB) fold,5,7–9 hence the tetramer has four po-tential ssDNA binding sites. The SSB tetramer canbind long ssDNA in a variety of binding modesdepending on solution conditions, especially saltconcentration and type.5,10–14 At low monovalentsalt concentrations (<10 mM NaCl) and highprotein to DNA ratios, an SSB tetramer binds tossDNAwith high inter-tetramer cooperativity usingonly two out of four subunits on average, and oc-cluding ∼35 nucleotides (nt) ((SSB)35 mode).15
However, at higher salt concentrations (>0.2 MNaCl), an SSB tetramer binds to ssDNA using allfour subunits with approximately 65 nt wrappedaround the tetramer to form the low cooperativity(SSB)65 mode.5,8
A model for SSB bound to ssDNA in its (SSB)65binding mode has been suggested based on theX-ray crystallographic structure of a tetramer of
SSBc (a C-terminal truncation of 42 residues) boundto two molecules of (dC)35.
8 In this model, ∼65 nt ofssDNA interact with all four SSB subunits with atopology approximating the stitching around abaseball (Figure 1(a)), hence the entry and exit sitesof a ∼65 nt ssDNA are in close proximity. A morespeculative structural model for the (SSB)35 bindingmode was also proposed as shown in Figure 1(b).8The transition between these binding modes isreversible13 but little information is currently avail-able about the dynamics of interconversion betweenthese binding modes.Here, we use single molecule fluorescence reso-
nance energy transfer (smFRET) techniques16,17 toexamine directly the dynamics of the spontaneous
Figure 1. E. coli SSB binding modes. (a) Model of the (SSBchymotrypsin truncated SSB tetramer (missing 42 C-terminal(70 nucleotides; white tube)wraps around the SSB tetramer (subto all four subunits. (b) Model of the (SSB)35 binding mode. ∼3subunits on average.8 (a) and (b) were generated using PyMoexperiment for SSB. (c) Partial duplex DNA, (dT)70 immobillow FRET. (d) Binding of a single SSB tetramer in (SSB)65 biSSB (in (SSB)35 binding mode) will result in an intermediate
structural changes between the two SSB-DNAbinding modes using ssDNA that accommodatesonly one tetramer bound in the (SSB)65 mode or twotetramers in the (SSB)35 mode. The results agree wellwith stopped-flow ensemble studies and furtherreveal a new binding configuration that we term(SSB)35b that can be formed from the (SSB)35 modevia a major rearrangement of the protein-DNAcomplex without protein dissociation. Deletion of42 amino acid residues from the C termini of SSBshifts the equilibrium to favor the (SSB)35 mode,suggesting that interaction of SSB with other rep-lication/recombination proteins via its C terminuscan potentially be regulated by these binding modesand vice versa.
)65 binding mode derived from the crystal structure of aresidues) bound to two (dC)35 oligonucleotides.
8 ssDNAunits colored individually; C terminus not shown) binding5 nucleotides (white tube) are bound to approximately twol.54 (c)–(e) Schematic design for the single molecule FRETized on “biocompatible” PEG coated quartz surface hasnding mode results in high FRET. (e) Binding of anotherFRET state.
1246 DNA Binding Mode Transitions of SSB Protein
Results
DNA constructs for monitoring SSB bindingmode transitions
For single molecule experiments, we used asurface-tethered partial duplex DNA labeled witha donor (Cy3) and an acceptor (Cy5) so that FRETbetween them reports on the conformations of theintervening 70 nt of ssDNA (Figure 1(c), referred tohereafter as (dT)70). While the DNA alone has lowFRET due to a large time-averaged distance betweenthe two fluorophores,18 ssDNAwrapping around aSSB tetramer in the (SSB)65 mode should resultin very high FRET (Figure 1(d)).19 Upon bindingtwo tetramers in the (SSB)35 mode, an intermediateFRET signal is expected due to a lower degree ofcompaction (Figure 1(e)). To demonstrate this inbulk solution, we performed an equilibrium titrationof Cy5-(T)68-Cy3-T-3′ (referred to as (dT)68) with SSBat 80 mM NaCl. Figure 2 shows the results of atitration of (dT)68 with SSB plotted as the normalizedCy5 fluorescence as a function of the ratio of the total[SSB] to total [(dT)68]. Upon addition of SSB, the Cy5fluorescence increases linearly up to a ratio of oneSSB tetramer per (dT)68, indicating stoichiometricformation of the fully wrapped (SSB)65 complex (1:1molar complex) characterized by high FRET indicat-ing that Cy3 and Cy5 are in close proximity. Upon
Figure 2. SSB concentration dependent changes inbinding modes observed in bulk solution. Results of anequilibrium titration of Cy5-(T)68-Cy3-T-3′ (0.1 μM) withSSB (buffer T, 80 mM NaCl, 25 °C) plotted as normalizedCy5 fluorescence (Fn=(F–F0)/F0) versus ratio of totalprotein to DNA concentrations (where F0 is fluorescenceintensity of DNA alone and F is the fluorescence measuredat each point in the titration). The biphasic character of thebinding isotherm indicates that two types of complexescan form, the first having one and the second having twotetramers bound and characterized by high and inter-mediate FRET values ((SSB)65 and (SSB)35, respectively),respectively. The continuous line represents the best fit tothe data with k1=5×10
10 M−1 (minimum estimate) andk2=(3.3±0.6)×10
7 M−1 and two additional parametersF1=7.1±0.1 and F2=2.7±0.1, reflecting the maximum Cy5fluorescence observed for one and two tetramers bound,respectively.
further addition of SSB, there is a more gradualdecrease in Cy5 fluorescence reflecting formation ofthe (SSB)35 complex with two SSB tetramers boundand characterized by an intermediate FRET value.Binding of the first SSB tetramer occurs with veryhigh affinity (stoichiometrically), whereas the appar-ent affinity of the second SSB tetramer is substan-tially lower. Although the second SSB tetramerbinds with high cooperativity in the (SSB)35 mode,this apparent negative cooperativity is due to thevery high affinity of a single tetramer in the (SSB)65mode.15 The binding isotherm in Figure 2 was fitwell by a model in which the DNA can bind two SSBtetramers (see Supplementary Data) characterizedby the two step-wise binding constants, k1=5×10
10
M−1 and k2= (3.3±0.6)×107 M−1, reflecting binding
of the first and the second SSB tetramers to DNA.The data in Figure 2 show that the DNA binds twoSSB tetramers at saturation, with each SSB bound inthe (SSB)35 binding mode.
Single molecule studies show a salt-dependentdistribution of binding modes
We first tested whether the two major SSB bindingmodes can be detected as two distinct FRET states atthe single molecule level (Figure 3(a) and (b)). In thepresence of 10 nM SSB tetramer, at low NaClconcentrations (<50 mM), we observed a singlepopulation with low apparent FRET efficiency(Eapp∼0.2) calculated as the acceptor intensitydivided by the total intensity. This low FRET speciesat 10 nM SSB is easily distinguishable from DNAalone (Eapp∼0.1) and from the donor only popula-tion (Eapp∼0) (FRET histograms in Figure 3(a)).Upon increasing the [NaCl], the low FRET speciesdisappeared concomitantly with the increase of ahigh FRET species (Eapp∼0.75). Since higher saltconcentrations favor the (SSB)65 mode,13 we tenta-tively assigned the high and low FRET states to the(SSB)65 and (SSB)35 modes, respectively.We evaluated whether the high FRET values that
we observe are consistent with the expectationsbased on the model of the (SSB)65 complex inferredfrom the SSB-DNA crystal structure.8 The apparentFRET efficiency, Eapp, should provide a goodmeasure of the true FRET efficiency because thecorrection factor, γ, which accounts for the differ-ences in quantum yield and detection efficiencybetween the donor and the acceptor is 1.03±0.02(details in Materials and Methods) for NaCl con-centrations ranging from 100 mM to 500 mM.20
FRET values increased from 0.73 at 100 mMNaCl to0.80 at 500 mM NaCl (Figure 3(a)), probablyreflecting some compaction of the extra bases notin contact with the protein at the higher saltconcentrations.18 Based on a R0 value of 60 Å forthe Cy3-Cy5 FRET pair,18 the calculated inter-dyedistance for the high FRET state is 47–51 Å. From thestructure-based model of a (dT)70 bound to an SSBctetramer, we estimate an average distance betweenthe first and seventieth base to be 33 Å. Thisagreement is reasonably good considering that the
1247DNA Binding Mode Transitions of SSB Protein
dyes on the (dT)70 in the single molecule experi-ments are placed on two different strands of thepartial duplex via flexible linkers, and suggests thatDNAwrapping in the (SSB)65 mode indeed inducesthe DNA ends to come together as inferred from thecrystal structure.The midpoint of the NaCl-induced transition be-
tween the low and high FRET states ranged from
Figure 3 (legend
80–130 mM NaCl for the SSB concentrations (10–200 nM) used here. We also observed a transitionfrom the low to high FRET state induced by in-creasing [MgCl2], with a midpoint of ∼2 mM MgCl2at 10 nM SSB (Supplementary Data Figure S1).Previous ensemble studies of SSB binding to poly(dT) reported midpoints of ∼20 mM NaCl (or0.06 mM MgCl2) and ∼160 mM NaCl (or 64 mM
on next page)
1248 DNA Binding Mode Transitions of SSB Protein
MgCl2) for the (SSB)35→(SSB)56 and (SSB)56→(SSB)65transitions, respectively.5,10,11 Direct comparison ofthese midpoints is difficult because the (SSB)56 modeis presumably only formed on long ssDNA, such aspoly(dT), and appears to be absent in our experi-ments with (dT)70. However, the salt concentrationranges over which the binding mode transitionsoccur are in general agreement, suggesting that thesmFRET measurements of DNA tethered on asurface recapitulate the major features of SSB-DNAcomplexes in bulk solution.Figure 3(b) shows representative time traces of the
donor and acceptor intensities and Eapp from asingle DNA-SSB complex which exhibits clearfluctuations between discrete states with, Eapp∼0.2and 0.75 (100 mM NaCl, 10 nM SSB). Similar two-state transitions were observed for a range of SSBand NaCl concentrations indicating the clear exis-tence of two distinct SSB-(dT)70 binding modes,which interconvert on a time scale ranging fromsub-seconds to a minute as we discuss below.
Effects of SSB concentration
Previous ensemble studies15 have shown a SSBconcentration-dependent transition from a statewith one SSB tetramer bound to (dA)70 ((SSB)65mode) to a state with two tetramers bound to (dA)70((SSB)35 mode). An ensemble titration of (dT)68 withSSB at 80 mM NaCl also shows this transition(Figure 2). Therefore, to further test our FRET stateassignments in the single molecule experimentswith (dT)70, we examined the effect of SSB concen-tration at three NaCl concentrations. At 500 mMNaCl, a single FRET population in the high FRETstate (Eapp∼0.75) was observed at all proteinconcentrations (10 pM–100 nM) (data not shown).SSB binding was very tight such that even at 100 pMSSB, we did not observe unliganded ssDNA (thatcould be detected as molecules with Eapp∼0.1). At100 mM NaCl, the high FRET state (Eapp∼0.73) of(dT)70 was first formed at low SSB concentrations,followed by a population shift to the low FRET state(∼0.2) at higher SSB concentrations (Figure 3(c)),indicating that the low FRET state of (dT)70 has moreSSB bound. This is consistent with our assignmentthat the high and low FRET states have one and twotetramers bound, respectively.
Figure 3. Salt and SSB-dependent changes in SSB bindingaverage FRET values for single molecules of (dT)70. The FRETprotein bound state (E∼0.2) at low salt corresponding to the (Sconversion to a high FRET state representing predominantly threveals the presence of another minor binding configuration, te(b) Donor and acceptor intensity and FRET time record of a sindistinct transitions between low and high FRET states, corespectively. (c) FRET histograms as a function of SSB concentrafavored at low SSB concentrations, but is replaced by (SSB)35 bihistograms as a function of SSB concentration at 10 mM NaCalthough with a somewhat lower FRET efficiency, Eapp∼0.6 (DNA from the SSB), although lower SSB concentrations (>30mode (low FRET) with two tetramers bound to DNA.
Even at the lowest NaCl concentration examined(10 mM), which favors the (SSB)35 mode, we firstobserve binding of a single tetramer (at 100 pM SSB),followed by binding of a second SSB tetramerresulting in formation of a low FRET state as theSSB concentration is increased (Figure 3(d)). Thesingle tetramer-(dT)70 complex formed at 10 mMNaCl displays Eapp∼0.6, which is significantlylower than the values observed at higher salts (100and 500 mM NaCl). We suggest that this reductionin FRET at 10 mM NaCl may reflect the possibilitythat a fully wrapped (SSB)65 complex is in dynamicequilibrium with partially unwrapped forms due tothe significant negative cooperativity for ssDNAbinding to the third and fourth SSB subunits thatoccurs at low [NaCl].21–23 Our preliminary investi-gations indicate that similar complexes formed atlow [NaCl] are indeed undergoing fast fluctuations(DNA unwrapping and rewrapping) that are fullyaveraged during the FRET histogram binning time(300 ms) used here (R. R., unpublished data).
Single molecule transition rates
Two-state time trajectories (Figure 3(b), 30 ms timeresolution) were analyzed (details in Materials andMethods) to yield transition rates for a range ofsolution conditions (50–200 mM NaCl; 10–200 nMSSB). The integrated dwell time histograms were fitto single or bi-exponential functions. For the highFRET state, single exponential fits were adequate athigh salt ([NaCl]>100 mM), but at lower salt (50–100 mM NaCl) the data were better fit by a bi-exponential function (Figure 4(a)), suggesting thattwo different species share the same high FRETvalue. The fast component (with lifetimes<180 msindependent of SSB concentration) is detectable onlyat low salt (its amplitude drops sharply with respectto the amplitude of the slow component withincreasing [NaCl]; Figure 4(b)) and is due to ashort-lived species we term (SSB)35b as will bediscussed later. The slow component has a lifetimethat is inversely proportional to [SSB] and thereforeits inverse represents the observed rate constant forthe (SSB)65→(SSB)35 transition, k65→35,obs,sm, whichinvolves binding of an additional SSB tetramer. Theobserved rate constants at higher salt concentrations(NaCl>100 mM) were determined from single
modes. (a) Histograms representing the distributions ofpeak for DNA only (E∼0.1) can be distinguished from a
SB)35 bindingmode. Increasing salt concentration results ine (SSB)65 binding mode. Kinetic analysis of the transitionsrmed (SSB)35b, within the high FRET population (see text).gle DNA molecule (100 mM NaCl and 10 nM SSB) showsrresponding to the (SSB)35 and (SSB)65 binding modes,tion at 100mMNaCl. (SSB)65 bindingmode (high FRET) isnding mode (E∼0.2) at higher SSB concentration. (d) FRETl. Formation of a 1:1 complex is still favored at low [SSB],due to suggested rapid partial unwrapping of the labeled0 pM) are required to induce the transition to the (SSB)35
Figure 4. Dependence of transition kinetics between different binding modes on NaCl and SSB concentrations (fromsingle molecule experiments). (a) Integrated dwell time distribution in high FRET state for (50 mM NaCl and 10 nM SSB)fit to single (red continuous line) and double (blue continuous line) exponentials. Residuals of the corresponding fits areshown in the lower panel. (b) Ratio of amplitudes of fast to slow component (a) as a function of [NaCl]. Amplitude offaster component from the double-exponential fits decreases dramatically with increasing salt concentration such thatonly single exponential fits are needed above 100 mM NaCl. (c) Rates of transition from the (SSB)65 to the (SSB)35 mode,k65→35,obs,sm, as a function of [NaCl] at different [SSB] (black squares, 10 nM SSB; red circles, 40 nM SSB; green triangles,120 nM SSB; blue inverted triangles, 200 nM SSB). (d) Dependence of observed rates, k65→35,obs,sm, on [SSB] at various[NaCl] (white, black, red, green and blue represent 50, 80, 100, 150, 200 mM NaCl, respectively). The slopes of the linearleast squares fits (shown in grey, black, red, green and blue) yield the bi-molecular rate constant for the transition atspecified [NaCl] (tabulated in Table 1). (e) Rate of transition from (SSB)35→(SSB)65 binding mode, k35→65,obs,sm increaseswith increasing [NaCl], but is independent of [SSB] (symbols and colors are similar to those in (c)).
1249DNA Binding Mode Transitions of SSB Protein
exponential fits. The values of k65→35,obs,sm increasedsharply with decreasing [NaCl] (Figure 4(c) and (d)),consistent with low salt favoring the (SSB)35 mode.13
The slope of a plot of k65→35,obs,sm versus [SSB](Figure 4(d)) yields the apparent bimolecular rateconstant, k65→35,sm, which ranges from 106 to 2×107
M−1s−1 depending on the [NaCl] (Table 1).Dwell times for the low FRET state were well
described by single exponential fits and yielded the
Table 1. Kinetic parameters for transitions between (SSB)35 a
a Values of k35→65,sf are calculated as an average of ksf,obs, at differethese [NaCl] (see Figure 5(b)).
observed rate klow→high. We obtained the actualrate constant for the (SSB)35→(SSB)65 transition(k35→65,obs,sm) by correcting for the contributionsfrom the alternative path of depopulating (SSB)35to form (SSB)35b (see Materials and Methods).Remarkably, k35→65,obs,sm increased 200-fold for amere fourfold increase in [NaCl] (Figure 4(e)). Thevalue of k35→65,obs,sm does not depend on SSBconcentration as expected for a dissociation reaction
nd (SSB)65 binding modes for wild-type SSB
lecule)k65→35,sf (M
−1s−1)(slope, stopped-flow)
k65→35,sm (M−1s−1)(slope, single molecule)
(6.96±0.34)×106 –– (2.0±0.02)×107
(4.73±0.17)×106 –(3.42±0.16)×106 (1.04±0.06)×107
(2.53±0.17)×106 (3.96±0.4)×106
(0.77±0.32)×106 (2.33±0.17)×106
– (1.29±0.09)×106
– –– –– –– –
nt protein–DNA ratios, since there is no dependence on [SSB] at
1250 DNA Binding Mode Transitions of SSB Protein
and the average value of k35→65,sm, is shown in Table1. This represents the first determination of thekinetic parameters for this SSB binding modetransition.
Stopped-flow studies of the (SSB)35-(SSB)65binding mode transition initiated by salt-jumps
To compare with the single molecule studies, wealso performed stopped-flow kinetic studies of thetransition between the (SSB)35 and (SSB)65 modes inbulk solution (buffer T, pH 8.1, 25 °C). The (SSB)35complex was pre-formed at low salt concentration(no added NaCl with SSB in molar excess over(dT)68) and the transition to the (SSB)65 mode wasinitiated by the addition of buffer containing a seriesof higher [NaCl]. Example time-courses (20 nM Cy5(T)68Cy3-T and 40 nM SSB) are shown in Figure 5(a)and are all well described by a single exponentialincrease in Cy5 fluorescence yielding observed rateconstants, ksf,obs. Figure 5(b) shows a plot of logksf,obsversus log [NaCl] determined for a variety of SSB tossDNA concentration ratios indicating that ksf,obs isvery sensitive to [NaCl]. For example, for thestarting concentrations used for the time-coursesshown in Figure 5(a), ksf,obs increases by almost5000-fold (light green line in Figure 5(b)) as the[NaCl] is increased from below 20 mM to 1 M.The value of ksf,obs is independent of SSB
concentration at high salt (above 0.15 M NaCl) butdisplays a linear dependence on [SSB] at lowersalt concentrations (Figure 5(c)). The slopes of theplots in Figure 5(c) decrease with increasing[NaCl], whereas the intercepts increase (also tabu-lated in Table 1). A simple one-step kinetic mecha-nism, as shown in the inset to Figure 5(a), canexplain the data, where k35→65,sf and k65→35,sf arethe rate constants for the (SSB)35→(SSB)65 and(SSB)65→(SSB)35 transitions, respectively. Based on
Figure 5. Stopped-flow kinetics of the (SSB)35 to (SSB)65 bintime-courses (green) obtained upon addition of buffer T wicomplex (20 nM: 40 nM final ratio) in the stopped-flow in bincrease in FRET signal reflecting formation of (SSB)65 modeshown in red. (b) Observed rates (ksf,obs) determined from thfunction of [NaCl] for different initial ratios of [Cy5-T68-Cy3-T20:160 nM and (5) 20:200 nM. (c) Dependences of ksf,obs on [SSBand intercept of the linear fits yield the bimolecular rate con(SSB)35→(SSB)65 (k35→65,sf), respectively (see Table 1). For [NaC
this mechanism, the observed rate, ksf,obs, is givenby equation (1):
ksf;obs ¼ k35Y65;sf þ k65Y35;sf½SSB� ð1ÞTherefore, the individual rate constants, k35→65,sfand k65→35,sf can be obtained from the intercept andslope of the linear dependences of ksf,obs on [SSB]shown in Figure 5(c) (Table 1). It is clear fromFigure 5(b) and (c) that the first term in equation (1)dominates at [NaCl]>0.15 M where the (SSB)65mode is favored. As the [NaCl] concentration isdecreased, k35→65,sf also decreases and the secondterm in equation (1) becomes significant.The rates for the (SSB)35→(SSB)65 transition
determined from the single molecule and stopped-flow measurements agree very well as shown inTable 1 (compare k35→65,sm and k35→65,sf). Thebimolecular rate constants determined for the(SSB)65→(SSB)35 transition are also comparable. Forexample, k65→35,sm, which varies by a factor of tenupon increasing the [NaCl] from 50 mM to 200 mMwas only 1.5–3-fold higher than k65→35,sf. This smalldifference may be due to the fact that the ssDNAused in the single molecule studies is longer, by twonucleotides, than that used in the stopped-flowstudies, which should facilitate the binding of thesecond tetramer (each with an occluded site size of∼35 nt). A similar enhancement (∼twofold) in rateis observed for SSB binding to (dT)68 versus (dT)66 instopped-flow measurements (A. G. K., unpublishedresults).
C-terminal truncated SSB and electrostaticeffects
The C terminus of E. coli SSB protein is highlyacidic,2,24 a feature shared by many other SSB pro-teins, such as phage T4 gene 32 protein (gp32)25,26and phage T7 gene 2.5 protein (gp2.5).1,27,28 The
ding mode transition initiated by salt-jumps. (a) Cy5 FRETth increasing [NaCl] to a premixed (Cy5-T68-Cy3-T)-SSBuffer T with no salt at 25 °C. Each time-course shows an. The fits of the traces to single exponential functions aree single exponential fits each time-course are plotted as a]tot/ [SSB]tot : (1) 20:40 nM, (2) 20:80 nM, (3) 20:120 nM, (4)] obtained from the data in (b) for [NaCl]<0.2 M. The slopestant of (SSB)65→(SSB)35 (k65→35,sf) and rate constant forl]≥0.2 M ksf,obs is independent of [SSB] (see Table 1).
1251DNA Binding Mode Transitions of SSB Protein
C-terminal domain of T4 gp32 has previously beenshown to modulate DNA binding via a salt-dependent structural rearrangement.25,26,29,30 Bothfor E. coli SSB2,5,31–35 and the T7 gp2.5,28,36,37 theC-terminal domain is the site for interactions withother proteins involved in DNA processing. Sincethe DNA wrapping topology (Figure 1(a)) wasderived from a crystal structure of ssDNA boundto a 42 amino acid residue C-terminal truncation ofSSB (termed SSBc)8 and the DNA was very poorlyresolved in a crystal structure of the full lengthSSB,38 we examined the binding mode transitionsfor ssDNA binding to SSBc, to determine whetherdeletion of the C termini influences the salt-dependent dynamics of these complexes. Impor-tantly, deletion of the 42 amino acid residues fromeach SSB subunit does not affect either the stabilityof the SSB tetramer or its high salt occluded site size(A. G. K., unpublished experiments).Upon performing single molecule FRET experi-
ments with the SSBc tetramer, we still observedtransitions between two FRET states with identicalFRET values as observed for the full length SSBtetramer except that the equilibrium is shifted infavor of the (SSBc)35 mode (data not shown). Only
Figure 6. Kinetics of binding mode transition for SSBc (Cfrom (SSBc)65→(SSBc)35 mode, k65→35,obs,sm decreases with inred circles, green triangles and blue inverted triangles represek65→35,obs,sm plotted as a function of SSBc concentration yieldsfor wild-type SSB for all NaCl concentrations (see Tables 1 andtransition, k35→65,obs,sm increases with [NaCl] albeit twofoconcentrations. There is no change in k35→65,obs,sm with [SSBc] (equilibrium constant, K65→35 (calculated as k65→35/k35→65 fro(circles) and SSBc (squares). The slopes, −6.3±0.2 and −6.0±0∼six ions.
the lowest salt concentration tested (100 mM NaCl)displayed any bi-exponential behavior in the highFRET state dwell time distribution, and for thatcase the slower component was used to estimatethe rate constant for the (SSBc)65→(SSBc)35 transi-tion. Both the forward and reverse rate constantshad similar dependences on NaCl concentration aswas observed for full length SSB (Figure 6(a)–(c)).However, there is a twofold decrease in the rate forthe (SSBc)35→(SSBc)65 transition and a ∼twofoldincrease in the bimolecular rate constant for the(SSBc)65→(SSBc)35 transition compared to full lengthSSB (Tables 1 and 2). The net result is a fourfoldchange in the equilibrium constant for the transition,K65→35≡k65→35,sm/k35→65,sm (Figure 6(d)), indicatingthat deletion of the C termini favors the (SSB)35mode.From the slope of a plot of log(K65→35) versus log
[NaCl], one can estimate the net number of ions(Na++Cl−) released or absorbed in the transitionfrom the (SSB)65 to the (SSB)35 modes.11 Such plotsare shown in Figure 6(d) for both the SSB and SSBcbinding mode transitions. In each case, the log-logslopes are negative, with slopes of −6.3±0.2 and−6.0±0.5 for SSB and SSBc, indicating a net release
-terminal truncated SSB). (a) Observed rate of transitioncreasing [NaCl] but increases with [SSBc]. Black squares,nt 10, 40, 120 and 200 nM SSBc, respectively. (b) Slope ofa bimolecular rate constant which is ∼twofold higher than2). (c) Observed rate for (SSBc)35→(SSBc)65 binding modeld slower than for wild-type SSB for the same salt[SSBc] color coding same as in (a)). (d) Dependences of them single molecule data) on [NaCl] (log-log plot) for SSB.5 for SSB and SSBc, respectively, indicate a net release of
Table 2. Kinetic parameters for transitions between(SSBc)35 and (SSBc)65 binding modes for truncated SSB(SSBc) from single molecule experiments
of ∼six ions upon binding the second SSB tetramerto form the (SSB)35 mode. Overall, the resultssuggest that the DNA wrapping topology deducedfrom the ssDNA-SSBc structures is likely to be thesame for SSBc and the full length protein. However,the C terminus does exert a significant influence onthe equilibrium between the two binding modes.Since the C terminus of SSB is the region involved ininteractions with other accessory proteins, it ispossible that the binding of these proteins to the Cterminus could alter either the equilibrium and/ordynamics of the transition between the SSB bindingmodes.
(SSB)35b, a novel (SSB)35 complex
The results presented above indicate evidence fortwo distinct SSB-DNA species that co-exist withinthe high FRET state, but that have markedlydifferent lifetimes. We assigned the longer-livedspecies to the (SSB)65 mode and the shorter-livedspecies to a new binding configuration termed(SSB)35b, since it exhibits the same stoichiometry asin the (SSB)35 mode. The rate constant for thetransition from the (SSB)35b to the (SSB)35 mode,k35b→35, was observed to be independent of [SSB](open squares in Figure 7(a)) and only slightlydependent on [NaCl]. The rate constant for thereverse step, k35→35b, calculated as described inMaterials and Methods (open circles in Figure 7(a)),is independent of [NaCl] and nearly 100-fold smallerin magnitude than k35b→35.We hypothesize that (SSB)35b represents a new
configuration of SSB binding to ssDNA consisting oftwo tetramers bound per (dT)70 as in (SSB)35, butwhich differs in that the two ends of (dT)70 are inmuch closer proximity. If this is the case, the (SSB)35bcomplex should be able to be formed directly fromthe (SSB)35 mode without any change in the extent ofSSB binding to the ssDNA (Figure 7(b)). Indeed,when we form the (SSB)35 mode at low [NaCl] andremove free SSB from solution, an isolated complexin the (SSB)35 mode displays transient excursions tothe high FRET state (Figure 7(c)).Next, we compared the kinetic rate constants for
the transition between the (SSB)35b and (SSB)35modes determined from the two sets of experimentsdescribed above. The values of k35b→35 obtainedfrom analysis of the time-dependent fluctuations inthe isolated (SSB)35 complexes display only amoderate salt dependence over the range from1 mM to 25 mM NaCl (Figure 7(a), filled blacksquares). These values approach the values of
k35b→35 determined from the fast phase of the bi-exponential decay in the presence of free SSB in therange from 50–100 mM NaCl (Figure 7(a), opensquares). Furthermore, the values of k35→35b, calcu-lated from analysis of the isolated (SSB)35 complexes(filled circles in Figure 7(a)) were similar to thevalues of k35→35b estimated at higher salt concentra-tions in the presence of free SSB (open circles inFigure 7(a)).Using these rate constants, we estimated the
equilibrium population of (SSB)35b for the range ofsolution conditions in this study (details in Materialsand Methods). Figure 7(d) shows the relativepopulation of (SSB)35b within the high FRET statethat increases with increasing protein concentrationand decreasing [NaCl]. However, the fraction of(SSB)35b in the total population of SSB-DNA com-plexes at equilibrium is less than 2% under allsolution conditions examined (Figure 7(d) inset).On the basis of the arguments presented above,
we use the kinetic scheme depicted in Figure 7(b) todescribe the transitions among these three confor-mations. The (SSB)35 binding mode with two SSBtetramers bound per (dT)70 can undergo transitionseither to the (SSB)65 mode upon dissociation of onetetramer or to the (SSB)35b configuration via a con-formational rearrangement without SSB dissocia-tion. Although the kinetic scheme in Figure 7(b)indicates a direct pathway for interconversion be-tween the two high FRET binding modes, (SSB)65and (SSB)35b (broken arrows in Figure 7(b)), the rateconstants indicate that the preferred pathway forthis interconversion would proceed through the(SSB)35 binding mode.
Discussion
In this study, we have used single molecule andstopped-flow FRET methods to determine, for thefirst time, the dynamics of the spontaneous transi-tions between the major SSB/DNA binding modes.Importantly, our single molecule data show in amost direct way that SSB binds to ssDNA in well-defined structural states that reflect these differentbinding modes that may regulate the functionalactivity of SSB in vivo.5 The experiments describedhere lay the foundation for further single moleculeinvestigations of the central role of this key proteinin DNA replication and recombination processes.Kinetic analysis of single molecule FRET data
revealed evidence that three distinct complexes canform on a 70 nt ssDNA: (SSB)35 with two SSBtetramers, (SSB)65 with one tetramer, and (SSB)35b, anewly uncovered minority species with two tetra-mers bound. Furthermore, we have established thereaction pathways connecting these species (Figure7(b)). The rate constants of the transitions betweenthe (SSB)35 and (SSB)65 modes are extremelysensitive to salt concentration. For example, k65→35decreases 10–16-fold with a fourfold increase in[NaCl] (from 50 mM to 200 mM). Even moreremarkably, k35→65 varies by 200-fold over the
Figure 7. Kinetics of the (SSB)35b↔(SSB)35 transition. (a) Rate constants for the (SSB)35↔(SSB)35b transition k35→35b(∼0.3 s−1; filled red circles) and k35b→35 (20–12 s−1 filled black squares) determined for isolated (SSB)35 complexes ([SSB]:[dT]70=2:1, no SSB in solution) at low [NaCl] (1–25 mM) from the analysis of dwell times of the traces as shown in (c).These rates are close to those estimated from experiments performed in the presence of free SSB in solution at intermediate[NaCl] (50–100 mM), k35→35b (∼0.1–0.2 s−1; open red circles) and k35b→35 (∼15–6 s−1, open squares; different colorsrepresent different SSB concentrations: 10 nM (black), 40 nM (red), 120 nM SSB (green) and 200 nM (blue)). (b) Kineticscheme showing interconversion among the SSB binding modes on (dT)70 (see text for details). (c) Fluorescence intensityand FRET time record of single DNA complexed with two SSB tetramers in 10 mM NaCl in the absence of free SSB showbrief excursions to a high FRET state indicating that the (SSB)35 binding mode can undergo a transition to a high FRETstate without dissociation of SSB from the ssDNA. (d) Fraction of (SSB)35b in the high FRET state as a function of NaCl andSSB concentrations (10 nM (black ), 40 nM (red), 120 nM SSB (green) and 200 nM (blue)). This is a minor species in thepopulation except at low salt (50 mM) and high protein concentrations (>40 nM SSB). Inset: Fraction of (SSB)35b in the totalequilibrium population is always <2%. Colors for SSB concentrations are same as in main panel.
1253DNA Binding Mode Transitions of SSB Protein
same range of salt concentration. As a consequence,outside of a narrow range of salt concentrations,only one or the other of the two modes is populatedsignificantly. We have not performed a kineticanalysis as a function of Mg2+, but equilibriumdata (Figure S1) suggest that a similar dichotomyholds for Mg2+ around its physiologically relevantconcentrations (which most likely lies close to thehigher end of the reported range of 6 μM to10 mM39–41). In fact, the 35 to 56 transition canalso be induced at micromolar concentrations of thepolyamine, spermidine.42 Therefore, charged spe-cies present in the cell or a change in the expressionlevel of SSB can easily shift the delicate balance
between the two modes, providing the possibilityfor significant regulatory consequences.We observe that the (SSB)35 complex can also
undergo a conformational transformation to form apreviously undetected, low abundance ternarycomplex (2SSB: DNA), which we term (SSB)35b.This minor complex was revealed only throughkinetic analysis of the single molecule data and isnot populated sufficiently to be detected in ensem-ble experiments. Extremely fast dynamics of DNAwrapping/unwrapping occurring in the microse-cond time scale have been observed within the(SSB)65 complex.43 In the (SSB)65 complex, SSBcontacts ssDNA via all four subunits and therefore
1254 DNA Binding Mode Transitions of SSB Protein
should be capable of remaining bound to DNA evenif a long segment of the DNA is unraveled. Here, weprovide evidence that the (SSB)35 binding conforma-tion is also dynamically capable of samplingalternate topologies without protein dissociation,even though the SSB is bound to ∼35 nt using onlytwo subunits. These results suggest that small scaleDNA wrapping/unwrapping dynamics mightoccur in this mode with implications for how SSBtetramers might undergo direct transfer betweensingle strands of DNA.44Deletion of the negatively charged C terminus of
SSB shifts the equilibrium constant for the (SSB)65 to(SSB)35 transition by a factor of four in favor of the(SSB)35 mode. This raises the possibility that thebinding of SSB to replication/recombination pro-teins, such as the χ subunit of DNA polymeraseIII,34,45 exonuclease I,31 PriA33 and RecO,46 whichoccurs primarily via the C terminus of SSB may alsoinfluence the equilibrium between SSB-ssDNAbinding modes. The (SSB)35 mode has been pro-posed to function during DNA replication by virtueof its high cooperativity,5,14 and our observationwith SSBc suggests the interesting possibility thatbinding of the χ subunit of DNA polymerase IIIholoenzyme to the C terminus of SSB35 may stabilizethe (SSB)35 mode and induce the formation of longSSB clusters. More generally, other proteins maymodulate SSB function by controlling its DNAbinding mode. Future studies will test this modelof reciprocity where the DNA binding mode of SSBis influenced by its interaction with other proteinsvia the C terminus and vice versa.The existence of DNA binding proteins that can
bind in multiple, salt-dependent binding modes ismore prevalent than generally thought. In fact, anyDNA binding protein possessing multiple DNAbinding sites has the potential for binding DNA inmultiple binding modes. Such salt-dependent bind-ing mode transitions have been observed for ssDNAbinding of yeast RPA (replication protein A, theeukaryotic SSB protein, which possesses four–fiveOB-folds),6 as well as for hnRNP (heterogeneousnuclear ribonucleoprotein) binding to pre-mRNA.47
The strong dependence of the stabilities and rates ofinterconversion between the SSB-ssDNA bindingmodes on salt and SSB concentration emphasizes theneed to consider explicitly the different properties ofits different binding modes5,11,14 when interpretingexperiments that include SSB as a component.
Materials and Methods
Oligodeoxyribonucleotides and SSB proteins
For single molecule experiments, a partial duplex with a71 base long ssDNA 3′ tail was generated by annealingstrands 5′-Cy5-GCCTCGCTGCCGTCGCCA-Biotin-3′and 5′-TGGCGACGGCAGCGAGGC(T)70-Cy3-T-3′ sothat the FRET (Cy3-Cy5) pairs are 70 bases apart((dT)70). Stopped-flow studies and equilibrium titrationswere done with 5′-Cy5-(T)68-Cy3-T-3′ with concentration
determined as described earlier.19 The oligodeoxynucleo-tides were synthesized using standard β-cyanoethylphosphoramidite chemistry using an ABI model 391automated DNA synthesizer (Applied Biosystems, FosterCity, CA). Biotin was incorporated using BiotinTEG CPG(all modification reagents were from Glen Research,Sterling, VA). E. coli SSB and SSBc proteins (>99% homo-geneity) were purified and concentrations determinedspectrophotometrically as described.48
Single-molecule TIR spectroscopy and analysis
All single molecule measurements were performed at23(±1) °C in 10 mM Tris (pH 8.0), 0.1 mM Na3EDTA,0.1 mg/ml BSA (NEB, Ipswich, MA), oxygen scavengingsystem49,50 (0.5% (w/v) glucose, 1.5 mM Trolox,51 165units/ml glucose oxidase and 2170 units/ml catalase(Roche, Indianapolis, IN)) and indicated amounts of NaCland SSB. Trolox concentration was determined spectro-photometrically using an extinction coefficient ε290=2350(±100) M−1cm−1.A wide-field prism-type total internal reflection (TIR)
microscope (IX 70, Olympus) was used to image singleimmobilized (<50 pM) DNA on NeutrAvidin (Pierce,Rockford, IL) treated biotin-PEG (PEG MW 5000, NektarTherapeutics, Huntsville, AL) quartz surface.50 Donor(Cy3) was excited with an Nd:YAG laser (532 nm, 50 mW;Reno, NV). Donor and acceptor (Cy5) fluorescence wascollected using water immersion objective (60X, 1.2numerical aperture, Olympus). Laser scatter was rejectedusing a long pass filter at 550 nm (Chroma, Rockingham,VT). Cy3 and Cy5 fluorescence was split using a dichroicmirror at 630 nm (Chroma) and imaged side by side on anelectron multiplying charge coupled device (CCD) camera(iXon DV 887-BI, Andor Technology, CT). Mappingcalibration between the donor and acceptor channels isachieved with the help of immobilized fluorescent beads(0.2 μm, 540/560 nm, Molecular Probes). Fluorescencedata as successive frames of images were acquired withcustomVisual C++ (Microsoft) routines and IDL (ResearchSystems Inc.) programs were used to obtain donor andacceptor intensity time traces with 30 ms integrationtime.After correcting for the leakage and background in both
channels, apparent FRET efficiency was calculated asEapp= IA/(IA+ ID), where IA and ID represent acceptor anddonor intensity, respectively. Single-molecule FRET histo-grams were generated by averaging for 300 ms. Gammacorrection factor, γ was calculated by estimating the ratioof change in the acceptor intensity, ΔIA to change inthe donor intensity, ΔID after acceptor photobleaching(γ=ΔIA/ΔID).
20 Gamma factor calculated (>50 molecules)was binned and fitted to a gaussian to yield the rep-resentative gamma factor for each solution condition(data not shown). Dwell times in each of the FRET stateswere estimated from time traces (having at least tenturnovers; 60–120 s long) using custom MATLAB (Math-works) routines using a “thresholding” criterion.52 Histo-grams of dwell times in each of the FRET states wereintegrated and fitted to single or bi-exponential decayfunctions to obtain the average lifetimes and respectiveamplitudes. Rate constants were estimated as the inverseof the lifetimes. For rate constants≤0.1 s−1, time tracesfrom multiple molecules were concatenated for dwelltime analysis to prevent underestimation of the dwelltimes.In the presence of two high FRET states ((SSB)65 and
(SSB)35b), the observed rate for the transition from the low
1255DNA Binding Mode Transitions of SSB Protein
to the high FRET state (klow→high, obtained as the inverse oflow FRET lifetime) reflects theweighted sum of k35→65,obs,sm(the observed rate constant for (SSB)35→(SSB)65) andk35→35b,obs,sm (the rate constant for (SSB)35→(SSB)35b).Hence, k35→65,obs,sm and k35→35b,obs,sm can be approximatedusing equation (2):
whereAmpslow andAmpfast represent the amplitudes of thefast and slow components determined from the bi-exponential fits. The fraction of (SSB)35b state in the highFRET state and the fraction of (SSB)35b in the totalpopulation are related to the rate constants k35→35b,k35b→35, k35→65, k65→35 and [SSB] according to the expres-sions in equations (3a) and (3b), respectively:
Stopped-flow kinetics and equilibrium fluorescencetitrations
Stopped-flow data were acquired using an SX.18MVstopped-flow instrument (Applied Photophysics Ltd,Leatherhead, UK) as described.19 Cy3 was excited at515 nm and sensitized Cy5 emission was collected using a665 nm long pass filter. Reactant solutions were preparedon ice and pre-incubated at 25 °C for 5 min prior tomixing.All kinetic time-courses used here represent an average ofsix to ten individual traces that were fitted to singleexponentials. Fluorescence titrations were performedusing a QM-4 spectrofluorometer (Photon TechnologyInternational, Lawrenceville, NJ) exciting Cy3 donor(515 nm) and monitoring sensitized emission from Cy5acceptor at 664 nm applying all corrections as described.53
All measurements were carried out in buffer T (10mMTris(pH 8.1), 0.1 mM Na3EDTA) at 25(±0.1) °C. All chemicalswere reagent grade and from Sigma-Aldrich (St. Louis,MO), if not stated otherwise.
Acknowledgements
R.R. thanks S. McKinney for data acquisitionprograms; C. Joo, I. Rasnik, S. Hohng, M. Nahas andS. Myong for experimental help and discussion. Wealso thank Dr G. Waksman for providing the PDBstructures for Figure 1 and T. Ho for DNA synthesis.Funding for the work was provided by NIH grants
to T.H. (GM065367) and T.M.L. (GM030498). T.H. isan investigator with the Howard Hughes MedicalInstitute.
Supplementary Data
Supplementary data associated with this articlecan be found, in the online version, at doi:10.1016/j.jmb.2007.03.079
References
1. Chase, J. W. & Williams, K. R. (1986). Single-strandedDNA binding proteins required for DNA replication.Annu. Rev. Biochem. 55, 103–136.
2. Meyer, R. R. & Laine, P. S. (1990). The single-strandedDNA-binding protein of Escherichia coli.Microbiol. Rev.54, 342–380.
3. Yuzhakov, A., Kelman, Z. & O'Donnell, M. (1999).Trading places on DNA–a three-point switch under-lies primer handoff from primase to the replicativeDNA polymerase. Cell, 96, 153–163.
4. Sun, W. & Godson, G. N. (1998). Structure of theEscherichia coli primase/single-strand DNA-bindingprotein/phage G4oric complex required for primerRNA synthesis. J. Mol. Biol. 276, 689–703.
5. Lohman, T. M. & Ferrari, M. E. (1994). Escherichia colisingle-stranded DNA-binding protein: multiple DNA-binding modes and cooperativities. Annu. Rev. Bio-chem. 63, 527–570.
6. Kumaran, S., Kozlov, A. G. & Lohman, T. M. (2006).Saccharomyces cerevisiae replication protein A binds tosingle-stranded DNA in multiple salt-dependentmodes. Biochemistry, 45, 11958–11973.
7. Raghunathan, S., Ricard, C. S., Lohman, T. M. &Waksman, G. (1997). Crystal structure of the homo-tetrameric DNA binding domain of Escherichia colisingle-stranded DNA-binding protein determined bymultiwavelength X-ray diffraction on the seleno-methionyl protein at 2.9-A resolution. Proc. NatlAcad. Sci. USA, 94, 6652–6657.
8. Raghunathan, S., Kozlov, A. G., Lohman, T. M. &Waksman, G. (2000). Structure of the DNA bindingdomain of E. coli SSB bound to ssDNA. Nature Struct.Biol. 7, 648–652.
9. Theobald, D. L., Mitton-Fry, R. M. & Wuttke, D. S.(2003). Nucleic acid recognition by OB-fold proteins.Annu. Rev. Biophys. Biomol. Struct. 32, 115–133.
10. Bujalowski, W. & Lohman, T. M. (1986). Escherichia colisingle-strand binding protein forms multiple, distinctcomplexes with single-stranded DNA. Biochemistry,25, 7799–7802.
11. Bujalowski, W., Overman, L. B. & Lohman, T. M.(1988). Binding mode transitions of Escherichia colisingle strand binding protein-single-stranded DNAcomplexes. Cation, anion, pH, and binding densityeffects. J. Biol. Chem. 263, 4629–4640.
12. Griffith, J. D., Harris, L. D. & Register, J., 3rd (1984).Visualization of SSB-ssDNA complexes active in theassembly of stable RecA-DNA filaments. Cold SpringHarbor Symp. Quant. Biol. 49, 553–559.
13. Lohman, T. M. & Overman, L. B. (1985). Two bindingmodes in Escherichia coli single strand bindingprotein-single stranded DNA complexes. Modulationby NaCl concentration. J. Biol. Chem. 260, 3594–3603.
14. Lohman, T. M., Bujalowski, W. & Overman, L. B.(1988). E. coli single strand binding protein: a new look
1256 DNA Binding Mode Transitions of SSB Protein
at helix-destabilizing proteins. Trends Biochem. Sci. 13,250–255.
15. Ferrari, M. E., Bujalowski, W. & Lohman, T. M. (1994).Co-operative binding of Escherichia coli SSB tetramersto single-stranded DNA in the (SSB)35 binding mode.J. Mol. Biol. 236, 106–123.
16. Ha, T. (2001). Single-molecule fluorescence resonanceenergy transfer. Methods, 25, 78–86.
17. Ha, T. (2001). Single-molecule fluorescence methodsfor the study of nucleic acids. Curr. Opin. Struct. Biol.11, 287–292.
18. Murphy, M. C., Rasnik, I., Cheng, W., Lohman, T. M.& Ha, T. (2004). Probing single-stranded DNAconformational flexibility using fluorescence spectro-scopy. Biophys. J. 86, 2530–2537.
19. Kozlov, A. G. & Lohman, T. M. (2002). Stopped-flowstudies of the kinetics of single-stranded DNAbinding and wrapping around the Escherichia coliSSB tetramer. Biochemistry, 41, 6032–6044.
20. Ha, T., Ting, A. Y., Liang, J., Caldwell, W. B., Deniz,A. A., Chemla, D. S. et al. (1999). Single-moleculefluorescence spectroscopy of enzyme conformationaldynamics and cleavage mechanism. Proc. Natl Acad.Sci. USA, 96, 893–898.
21. Bujalowski, W. & Lohman, T. M. (1989). Negative co-operativity in Escherichia coli single strand bindingprotein-oligonucleotide interactions. I. Evidence and aquantitative model. J. Mol. Biol. 207, 249–268.
22. Bujalowski, W. & Lohman, T. M. (1989). Negative co-operativity in Escherichia coli single strand bindingprotein-oligonucleotide interactions. II. Salt, tempera-ture and oligonucleotide length effects. J. Mol. Biol.207, 269–288.
23. Lohman, T. M. & Bujalowski, W. (1988). Negativecooperativity within individual tetramers of Escher-ichia coli single strand binding protein is responsiblefor the transition between the (SSB)35 and (SSB)56DNA binding modes. Biochemistry, 27, 2260–2265.
24. Chase, J. W., L'Italien, J. J., Murphy, J. B., Spicer, E. K.& Williams, K. R. (1984). Characterization of theEscherichia coli SSB-113 mutant single-stranded DNA-binding protein. Cloning of the gene, DNA andprotein sequence analysis, high pressure liquidchromatography peptide mapping, and DNA-bindingstudies. J. Biol. Chem. 259, 805–814.
25. Lonberg, N., Kowalczykowski, S. C., Paul, L. S. & vonHippel, P. H. (1981). Interactions of bacteriophage T4-coded gene 32 protein with nucleic acids. III. Bindingproperties of two specific proteolytic digestion pro-ducts of the protein (G32P*I and G32P*III). J. Mol. Biol.145, 123–138.
26. Waidner, L. A., Flynn, E. K., Wu, M., Li, X. & Karpel,R. L. (2001). Domain effects on the DNA-interactiveproperties of bacteriophage T4 gene 32 protein. J. Biol.Chem. 276, 2509–2516.
27. Marintcheva, B., Hamdan, S. M., Lee, S. J. & Richard-son, C. C. (2006). Essential residues in theC terminus ofthe bacteriophage T7 gene 2.5 single-stranded DNA-binding protein. J. Biol. Chem. 281, 25831–25840.
28. Kong, D. C. & Richardson, C. C. (1998). Role of theacidic carboxyl-terminal domain of the single-stranded DNA-binding protein of bacteriophage T7in specific protein-protein interactions. J. Biol. Chem.273, 6556–6564.
29. Hosoda, J. & Moise, H. (1978). Purification andphysicochemical properties of limited proteolysis pro-ducts of T4 helix destabilizing protein (gene 32 pro-tein). J. Biol. Chem. 253, 7547–7558.
30. Pant, K., Karpel, R. L., Rouzina, I. & Williams, M. C.
(2005). Salt dependent binding of T4 gene 32 proteinto single and double-stranded DNA: single moleculeforce spectroscopy measurements. J. Mol. Biol. 349,317–330.
31. Genschel, J., Curth, U. & Urbanke, C. (2000). Interac-tion of E. coli single-stranded DNA binding protein(SSB) with exonuclease I. The carboxy-terminus of SSBis the recognition site for the nuclease. Biol. Chem. 381,183–192.
32. Curth, U., Genschel, J., Urbanke, C. & Greipel, J.(1996). In vitro and in vivo function of the C-terminusof Escherichia coli single-stranded DNA bindingprotein. Nucl. Acids Res. 24, 2706–2711.
33. Cadman, C. J. & McGlynn, P. (2004). PriA helicase andSSB interact physically and functionally. Nucl. AcidsRes. 32, 6378–6387.
34. Kelman, Z., Yuzhakov, A., Andjelkovic, J. & O'Don-nell, M. (1998). Devoted to the lagging strand-thesubunit of DNA polymerase III holoenzyme contactsSSB to promote processive elongation and slidingclamp assembly. EMBO J. 17, 2436–2449.
35. Witte, G., Urbanke, C. & Curth, U. (2003). DNApolymerase III chi subunit ties single-stranded DNAbinding protein to the bacterial replication machinery.Nucl. Acids Res. 31, 4434–4440.
36. He, Z. G., Rezende, L. F., Willcox, S., Griffith, J. D. &Richardson, C. C. (2003). The carboxyl-terminaldomain of bacteriophage T7 single-stranded DNA-binding protein modulates DNA binding and inter-action with T7 DNA polymerase. J. Biol. Chem. 278,29538–29545.
37. He, Z. G. & Richardson, C. C. (2004). Effect of single-stranded DNA-binding proteins on the helicase andprimase activities of the bacteriophage T7 gene 4protein. J. Biol. Chem. 279, 22190–22197.
38. Savvides, S. N., Raghunathan, S., Futterer, K., Kozlov,A. G., Lohman, T. M. & Waksman, G. (2004). TheC-terminal domain of full-length E. coli SSB isdisordered even when bound to DNA. Protein Sci. 13,1942–1947.
39. Kung, F. C., Raymond, J. & Glaser, D. A. (1976). Metalion content ofEscherichia coli versus cell age. J. Bacteriol.126, 1089–1095.
40. Lusk, J. E., Williams, R. J. & Kennedy, E. P. (1968).Magnesium and the growth of Escherichia coli. J. Biol.Chem. 243, 2618–2624.
41. Hurwitz, C. & Rosano, C. L. (1967). The intracellularconcentration of bound and unbound magnesiumions in Escherichia coli. J. Biol. Chem. 242, 3719–3722.
42. Wei, T. F., Bujalowski, W. & Lohman, T. M. (1992).Cooperative binding of polyamines induces theEscherichia coli single-strand binding protein-DNAbinding mode transitions. Biochemistry, 31, 6166–6174.
43. Kuznetsov, S. V., Kozlov, A. G., Lohman, T. M. &Ansari, A. (2006). Microsecond dynamics of protein-DNA interactions: direct observation of the wrap-ping/unwrapping kinetics of single-stranded DNAaround the E. coli SSB tetramer. J. Mol. Biol. 359, 55–65.
44. Kozlov, A. G. & Lohman, T. M. (2002). Kineticmechanism of direct transfer of Escherichia coli SSBtetramers between single-stranded DNA molecules.Biochemistry, 41, 11611–11627.
45. Witte, G., Urbanke, C. & Curth, U. (2003). DNApolymerase III chi subunit ties single-stranded DNAbinding protein to the bacterial replication machinery.Nucl. Acids Res. 31, 4434–4440.
46. Hobbs, M. D., Sakai, A. & Cox, M. M. (2007). SSBprotein limits RecOR binding onto single-strandedDNA. J. Biol. Chem. 282, 11058–11067.
1257DNA Binding Mode Transitions of SSB Protein
47. Nadler, S. G., Merrill, B. M., Roberts, W. J., Keating,K. M., Lisbin, M. J., Barnett, S. F. et al. (1991). Inter-actions of the A1 heterogeneous nuclear ribonucleo-protein and its proteolytic derivative, UP1, with RNAand DNA: evidence for multiple RNA bindingdomains and salt-dependent binding mode transi-tions. Biochemistry, 30, 2968–2976.
48. Lohman, T. M., Green, J. M. & Beyer, R. S. (1986). Large-scale overproduction and rapid purification of theEscherichia coli ssb gene product. Expression of the ssbgene under lambda PL control. Biochemistry, 25, 21–25.
49. Benesch, R. E. & Benesch, R. (1953). Enzymaticremoval of oxygen for polarography and relatedmethods. Science, 118, 447–448.
50. Rasnik, I., Myong, S., Cheng, W., Lohman, T. M. & Ha,T. (2004). DNA-binding orientation and domain
conformation of the E. coli rep helicase monomerbound to a partial duplex junction: single-moleculestudies of fluorescently labeled enzymes. J. Mol. Biol.336, 395–408.
51. Rasnik, I., McKinney, S. A. & Ha, T. (2006). Nonblink-ing and long-lasting single-molecule fluorescenceimaging. Nature Methods, 3, 891–893.
52. McKinney, S. A., Declais, A. C., Lilley, D. M. & Ha, T.(2003). Structural dynamics of individual Hollidayjunctions. Nature Struct. Biol. 10, 93–97.
53. Lohman, T. M. & Mascotti, D. P. (1992). Nonspecificligand-DNA equilibrium binding parameters deter-mined by fluorescence methods. Methods Enzymol.212, 424–458.
54. DeLano, W. L. (2002). The PyMOL Molecular GraphicsSystem. DeLano Scientific, San Carlos, CA, USA.
Edited by J. O. Thomas
(Received 22 December 2006; received in revised form 29 March 2007; accepted 29 March 2007)Available online 5 April 2007
![35 [2,3]-sigmatropic rearrangements](https://static.documents.pub/doc/80x56/55504042b4c905b2788b48e9/35-23-sigmatropic-rearrangements.jpg)
![[3,3]-Sigmatropic rearrangements - Massey gjrowlan/stereo2/ Organic Chemistry Claisen rearrangements • One of the most useful sigmatropic rearrangements is the Claisen rearrangement](https://static.documents.pub/doc/80x56/5adcada77f8b9a213e8bd8b0/33-sigmatropic-rearrangements-massey-gjrowlanstereo2lecture11pdf123702.jpg)
