Molecular Dissection of the Homotrimeric Sliding Clamp of T4

Sliding clamp proteins are circular dimers or trimers that encircle DNA and serve as processivity factors during DNA replication. Their presence in al...
2 downloads 10 Views 2MB Size
Download PDF
Article pubs.acs.org/biochemistry

Molecular Dissection of the Homotrimeric Sliding Clamp of T4 Phage: Two Domains of a Subunit Display Asymmetric Characteristics Manika Indrajit Singh and Vikas Jain* Microbiology and Molecular Biology Laboratory, Department of Biological Sciences, Indian Institute of Science Education and Research (IISER), Bhopal 462023, India S Supporting Information *

ABSTRACT: Sliding clamp proteins are circular dimers or trimers that encircle DNA and serve as processivity factors during DNA replication. Their presence in all the three domains of life and in bacteriophages clearly indicates their high level of significance. T4 gp45, besides functioning as the DNA polymerase processivity factor, also moonlights as the late promoter transcription determinant. Here we report a detailed biophysical analysis of gp45. The chemical denaturation of gp45 probed by circular dichroism spectroscopy, tryptophan fluorescence anisotropy, and blue-native polyacrylamide gel electrophoresis suggests that the protein follows a three-state denaturation profile and displays an intermediate molten globule-like state. The three-state transition was found to be the result of the sequential unfolding of the two domains, the N-terminal domain (NTD) and the C-terminal domain (CTD), of gp45. The experiments involving Trp fluorescence quenching by acrylamide demonstrate that the CTD undergoes substantial changes in conformation during formation of the intermediate state. Further biophysical dissection of the individual domain reveals contrasting properties of the two domains. The NTD unfolds at low urea concentrations and is also susceptible to protease cleavage, whereas the CTD resists urea-mediated denaturation and is not amenable to protease digestion even at higher urea concentrations. These experiments allow us to conclude that the two domains of gp45 differ in their dynamics. While the CTD shows stability and rigidity, we find that the NTD is unstable and flexible. We believe that the asymmetric characteristics of the two domains and the interface they form hold significance in gp45 structure and function. similarity.5,12,13 The crystal structure of gp45 shows that the protein is a homotrimer.13 Each subunit of gp45 has two domains connected through a 20-residue proline-rich interdomain linker. The subunits interact with each other in a head-to-tail fashion, with the N-terminal region of one subunit interacting with the Cterminal region of the other through hydrogen bonds (H-bonds) at the interface. Although it has been shown that gp45 stays as an open trimer in solution, with one open and two closed interfaces,14,15 it is not clear how an open trimer state is maintained by the protein. In this paper, we report a detailed biophysical examination of gp45. The chemical denaturation of gp45 as probed by fluorescence anisotropy, fluorescence quenching by acrylamide, circular dichroism spectroscopy, blue-native polyacrylamide gel electrophoresis, and pulse proteolysis shows a three-state transition, which is the result of sequential unfolding of the two domains of gp45. Furthermore, these domains differ in their dynamics. While the C-terminal domain (CTD) shows stability and rigidity, we find that the N-terminal domain (NTD) is

DNA replication is an inherent property of all living organisms. The presence of this process even in bacteriophages makes the virus a connecting link between living and nonliving systems. The machinery for DNA replication is conserved in all domains of life and necessarily includes a DNA polymerase (DNAP). The ability of DNAP to conduct chain elongation without being dissociated from the template is affected by the sliding clamp, a ring-shaped protein molecule.1 The clamp encircles the DNA and interacts with DNAP, preventing the enzyme from falling off the DNA and, thereby, acting as a processivity-promoting factor.2−4 Clamp proteins have been discovered in all three domains of life (bacteria, archaea, and eukarya) and bacteriophages.5 Some of the well-studied clamps include β clamps of bacterial origin, such as those from Escherichia coli, Mycobacterium tuberculosis, Streptococcus pyogenes, and Deinococcus radiodurans, proliferating cell nuclear antigen (PCNA) of human, yeast, and archaea, and gp45 of T4 and RB69 bacteriophages.3,6−11 Whereas bacterial clamps are dimers, those from bacteriophage, eukaryotes, and archaea are trimeric.3,9 It is intriguing that during evolution, while the bacteria developed a dimeric clamp, eukaryotes retained the phage-like trimeric molecules. Nevertheless, all of the identified clamps have a ring-shaped geometry, despite having a low degree of sequence © 2016 American Chemical Society

Received: November 6, 2015 Revised: December 31, 2015 Published: January 6, 2016 588

DOI: 10.1021/acs.biochem.5b01204 Biochemistry 2016, 55, 588−596

Article

Biochemistry unstable and flexible. We believe that the asymmetric characteristics of the two domains and the interface they form hold significance in gp45 structure and function and allow the protein to maintain an open trimer structure in solution.

CD values were obtained at the same wavelengths. The tryptophan fluorescence anisotropy measurements at varying temperatures were performed to monitor thermal denaturation; 8 μM (monomer) protein in gp45 buffer was heated from 5 to 95 °C at a ramp rate of 1 °C/min. The excitation and emission wavelengths were kept as 295 and 344 nm, respectively. The slit width was set at 5 nm for both excitation and emission, with an integration time of 5 s. The reversibility of the reaction was assessed by following a method similar to that used for CD spectroscopy. To assess the contribution of heat to the anisotropy values, unfolded protein (in 8 M urea) was taken as a control. Experiments were performed using a Fluoromax 4c spectrofluorometer (Horiba Jobin-Yvon). Chemical Denaturation of gp45. For protein unfolding studies, 2 mg/mL protein in a buffer containing 50 mM sodium phosphate (pH 7.4) and 50 mM NaCl was mixed with freshly prepared and deionized 10 M urea in the same buffer to achieve specified concentrations of urea while the protein was maintained at a final concentration of 0.2 mg/mL (8 μM monomer). The protein was allowed to equilibrate for 24 h before further analysis. The far-UV CD profile at different urea concentrations was recorded using a JASCO J-815 spectropolarimeter. A total of three accumulations per sample were averaged and blank-subtracted before the molar ellipticity (ME) values were calculated.17 The ME at 222 nm was used to calculate fraction unfolded (f u) as

■

MATERIALS AND METHODS Materials. All the reagents used in this study were procured from Sigma and were of the highest quality available. Restriction endonucleases were purchased from New England Biolabs. Oligonucleotides were procured from Macrogen (Geumcheon District, South Korea). Preparation of Proteins. Wild-type gp45 (gp45WT) was subcloned into vector pET21b (Novagen) from pET-MO (a kind gift from E. Peter Geiduschek, University of California at San Diego, San Diego, CA) with a C-terminal six-His tag. E. coli BL21(DE3) cells were transformed with plasmids for protein preparation. Post-induction, cells were incubated at 37 °C for 3 h for gp45WT and the Trp mutants (gp45W92 and gp45W199, where the number represents the Trp that is intact) and at 22 °C for 12 h for the truncated NTD and CTD. Purification of proteins was conducted as described previously,16 using Ni-NTA affinity purification. For all biophysical studies, after Ni-NTA pull-down, the proteins were dialyzed against gp45 buffer [50 mM sodium phosphate (pH 7.4), 50 mM NaCl, 1 mM EDTA, and 5% glycerol] and concentrated using Millipore Amicon filters. The protein samples were extensively centrifuged at high speed before any experiment was performed. The proteins were quantified by measuring their absorbance at 280 nm and using molar extinction coefficients of 19940 M−1 cm−1 for the wild type and 14440 M−1 cm−1 for single-Trp mutants. The gp45WT concentration is 8 μM monomer in all experiments, unless otherwise stated. Size Exclusion Chromatography. The native oligomeric state of gp45WT was analyzed by performing size exclusion chromatography on a Superdex-75 10/300 GL column (GE Healthcare); 200 μg of gp45 in SEC buffer [50 mM sodium phosphate (pH 7.4) and 150 mM NaCl] was passed through the column, and the elution profile was monitored at 280 nm. The peaks were analyzed and compared against calibration proteins [conalbumin (75 kDa), ovalbumin (43 kDa), ribonuclease A (13 kDa), and aprotinin (6.5 kDa) (GE Healthcare)]. Thermal Denaturation of gp45. The thermal denaturation of gp45WT and its mutants was analyzed using differential scanning calorimetery (DSC), circular dichroism (CD) spectroscopy, and fluorescence anisotropy. For DSC measurements, proteins (20 μM, monomer concentration) were equilibrated at the start temperature for 10−15 min before thermograms were recorded by heating the samples from 20 to 80 °C at a ramp rate of 1 °C/min on a Microcal VP-DSC calorimeter (GE Healthcare). The data were blank-subtracted and concentration-normalized (using the gp45 monomer molecular weight). After fitting using Microcal-VPDSC analysis software, Tm, ΔH, and ΔHvH values were derived. The ΔHvH/ΔH fraction was used to measure the oligomeric state of the protein. The effect of temperature on the secondary structure content of the proteins was determined via CD spectroscopy studies on a JASCO J-815 spectropolarimeter (JASCO Inc., Tokyo, Japan). The thermal melting of 8 μM (monomer) protein was performed in gp45 buffer by heating the protein from 4 to 95 °C at a ramp rate of 1 °C/min. The data were recorded at 208, 215, and 222 nm. The CD values at 222 nm were converted to molar ellipticity (θME) and plotted.17 To assess the reversibility of the reaction, the sample was cooled from 95 to 4 °C at a rate of 1 °C/min and the

fu =

yo − yn yu − yn

where y0 is the ME at any urea concentration and yn and yu are ME values for fully folded (at 0 M urea) and unfolded (at 8 M urea) protein, respectively. These samples were also subjected to the tryptophan fluorescence anisotropy recording using a Fluoromax 4c spectrofluorometer (Horiba Jobin-Yvon). The excitation and emission wavelengths used to monitor fluorescence anisotropy were 295 and 344 nm, respectively. For refolding studies, concentrated protein in 8 M urea was diluted in 1× buffer (containing additional urea to achieve the desired urea concentration) to a final concentration of 8 μM monomer. The free energy of unfolding (ΔG°U) at 0 M urea is derived by fitting the fraction unfolded values, determined from equilibrium unfolding studies conducted using CD spectroscopy, to either two-state or three-state equations.18,19 The fraction unfolded values for gp45NTD and gp45CTD were fitted to a two-state equation, whereas the data for gp45WT were fitted to a three-state equation19 to derive thermodynamic parameters. Other thermodynamic parameters (m value and Cm, representing the cooperativity of unfolding and the chemical denaturation midpoint, respectively) were derived from the fraction unfolded data as described previously.19 Formaldehyde Cross-Linking. The protein (20 μM monomer) sample in 50 mM sodium phosphate (pH 7.4) and 50 mM NaCl was treated with 1% formaldehyde, and the reaction mixture was incubated at 30 °C. The cross-linking was quenched after 20 min using 5× sodium dodecyl sulfate (SDS) loading dye [250 mM Tris-HCl (pH 6.8), 10% SDS, 0.5 M βmercaptoethanol, 0.5% bromophenol blue, and 25% glycerol] to final concentration of 1×. The proteins were separated via SDS− polyacrylamide gel electrophoresis (PAGE), and densitometric analysis of cross-linked bands was performed using ImageJ.20 The fraction cross-linked was calculated as the ratio of the 589

DOI: 10.1021/acs.biochem.5b01204 Biochemistry 2016, 55, 588−596

Article

Biochemistry intensities of the cross-linked species with the sum of the intensities of all species. Blue-Native Polyacrylamide Gel Electrophoresis. Bluenative PAGE was performed to analyze the oligomeric nature of the proteins in urea. The protein samples at different urea concentrations were mixed with native loading dye [50 mM TrisHCl (pH 6.8), 0.5% bromophenol blue, and 5% glycerol] and loaded onto a 4 to 15% gradient nondenaturing gel devoid of SDS. The cathode buffer [15 mM Tris-HCl (pH 7.0) and 50 mM Tricine] contained 0.01% Coomassie Brilliant Blue G-250, and the anode buffer consisted of 12.5 mM Tris-HCl (pH 8.3) and 96 mM glycine. The gel was run at 4 °C and destained using 10% acetic acid in water. Fluorescence Quenching by Acrylamide. The 10× concentrated protein was incubated with different urea concentrations for 24 h to achieve equilibrium before the fluorescence quenching studies were performed. A 6 M stock of acrylamide in 50 mM sodium phosphate (pH 7.4) and 50 mM NaCl was used to achieve various concentrations of acrylamide ranging from 0.05 to 0.5 M to monitor tryptophan fluorescence quenching. The protein samples (final concentrations are ∼8 μM for gp45WT and ∼13 μM for Trp mutants) were mixed with acrylamide and equilibrated for 30 min at 25 °C. The data were recorded using a Fluoromax 4c spectrofluorometer with a λex of 295 nm and emission from 310 to 400 nm. The λem‑max at each urea concentration was then used to prepare a Stern−Volmer plot, after taking into account the correction for the inner filter effect.21,22 The slope, the Stern−Volmer constant, Ksv, of the plot for each urea concentration was calculated and plotted. Pulse Proteolysis. Six micrograms of gp45WT and 2 μg of gp45CTD in varying urea concentrations were treated with 15 ng of proteinase K (NEB) in a 10 μL reaction volume and incubated for 1 min at 37 °C. The reaction was performed in buffer containing 20 mM Tris-HCl (pH 8), 1 mM CaCl2, and 5% glycerol. Quenching of the reaction was achieved by adding 5× SDS loading dye containing 10 mM phenylmethanesulfonyl fluoride (PMSF). The digested protein was run on 15% SDS− PAGE and stained with Coomassie dye. The protein bands were subjected to peptide mass fingerprinting by matrix-assisted laser desorption ionization (MALDI) time of flight (ToF)/ToF after in-gel trypsin digestion as described elsewhere.23

Figure 1. gp45WT remains a trimer in solution. (A) Coomassie-stained SDS−PAGE gel showing purified gp45WT after incubation without (−) and with (+) a formaldehyde cross-linker (CL). The bands are labeled as T (trimer), D (dimer), and M (monomer). The protein molecular weight marker (Ladder) is shown with a few of the bands labeled. A large amount of the trimeric form of protein is obtained after treatment with formaldehyde. (B) Size exclusion chromatography profile of gp45WT along with standard calibrants. gp45WT elutes as a trimer (∼85 kDa), and no other species is detected.

scanning microcalorimetry (DSC) measurements. The far-UV CD spectrum shows that the protein possesses an α+β conformation (Figure 2A) that largely matches the reported

Figure 2. Thermal stability of gp45WT trimer. (A) Circular dichroism profile of gp45WT in molar ellipticity [θ]ME before (green) and after (red) the experiment. The protein was heated to 95 °C, and pre- and post-heating scans were recorded at 4 °C. The native (before denaturation) protein CD profile depicts an α+β conformation, which changes to classical β-sheet (negative maxima at 215 nm) after denaturation. The inset shows a comparison of the secondary structure content of gp45 estimated from the crystal structure (Protein Data Bank entry 1CZD) and CD spectroscopic estimation for the purified protein. gp45WT structural content estimation was done using Reed’s estimation, available with the Spectra analysis package of Spectra manager software (Jasco Analytical Instruments). (B) DSC profile of gp45WT. Tm (apparent Tm), ΔH, and ΔHvH values (with standard error) obtained from fitting the peak using Microcal VP-DSC analysis are given. The ΔHvH/ΔH ratio of ∼3 suggests that the protein is a trimer in solution.

■

RESULTS gp45 Remains Largely a Trimer in Solution. To the folding and assembly of understand wild-type gp45 (gp45WT) into a trimer, we first examined the protein in detail. We addressed the oligomerization state of gp45WT by subjecting the purified protein to formaldehyde-mediated cross-linking followed by denaturing PAGE analysis. We observe all the three species of gp45WT, viz., monomer, dimer, and trimer, with varying intensities (Figure 1A), with the trimer exhibiting the largest population (80.7%) when compared with those of dimer and monomer (3.5 and 15.7%, respectively). Additionally, size exclusion chromatography (SEC) demonstrates that gp45WT is predominantly a trimer in solution (Figure 1B); we could not detect any peak belonging to either monomeric or dimeric species in this experiment. This observation further supports a previous finding that gp45WT trimer has a very low dissociation constant (∼250 nM).11 Thermal Denaturation of gp45WT Is a Two-State Event. After establishing that gp45WT is largely trimeric, we studied the stability of the protein by thermal denaturation using circular dichroism (CD), Trp fluorescence anisotropy, and differential

crystal structure.13 Unfolding of gp45WT is noted to be a highly cooperative two-state event (Figure S1A) that is followed by protein aggregation (data not shown); upon cooling, the sample yields a profile that largely demonstrates the presence of β-sheet (Figure 2A). Thus, the formation of soluble aggregates after heating and cooling suggests that gp45WT unfolding caused by heat is irreversible. Denaturation of a multimeric protein usually involves the breakdown of the multimer into monomeric subunits followed by the denaturation of monomer. We envisaged capturing the breakdown of trimeric gp45WT into individual subunits and the denaturation of monomer by heat using the fluorescence anisotropy of the two tryptophans (W92 590

DOI: 10.1021/acs.biochem.5b01204 Biochemistry 2016, 55, 588−596

Article

Biochemistry and W199) present in gp45WT. Surprisingly, however, the fluorescence anisotropy recorded while heating gp45 WT demonstrates a single major event (Figure S1B). The experiment, nevertheless, further confirms that the thermal unfolding of the protein is irreversible. The measurement of the energetic costs for trimer dissociation and gp45WT unfolding, assessed using microcalorimetry, provides us with only one distinct peak with an apparent Tm, the midpoint of the transition from the native to denatured state when the thermal denaturation is irreversible,24 corresponding to 52.3 °C (Figure 2B). The peak profile suggests cooperative denaturation of gp45WT; although we expected to capture a distinct peak corresponding to dissociation of trimer into monomer, this could not be obtained. The recorded peak profile thus suggests that both events, i.e., dissociation of trimer into monomer and monomer denaturation, occur over a very narrow temperature range, which is in good agreement with the fluorescence anisotropy data. The ΔHvH/ΔH ratio (ratio of van’t Hoff enthalpy ΔHvH and calorimetry enthalpy ΔH), which represents the oligomeric state of the protein, was found to be ∼3, confirming that we are monitoring the energetics of unfolding of trimeric gp45WT. Thus, our thermal denaturation measurements, conducted via fluorescence anisotropy and DSC, allow us to conclude that gp45WT trimer to monomer conversion and denaturation of protein by heat are coupled events. Chemical Denaturation of gp45WT Follows a ThreeState Unfolding. To capture the unfolding transitions of gp45WT, we proceeded with chemical denaturation and followed the local and global changes using fluorescence and CD, respectively. We subjected gp45WT trimer to urea-mediated chemical denaturation and probed the unfolding process using Trp fluorescence anisotropy. Interestingly, unlike thermal denaturation, chemical unfolding of gp45WT is found to be three-state (Figure 3A), with the two distinct states corresponding to the folded trimeric gp45WT (in 0 M urea) and unfolded monomeric gp45WT (in 8 M urea) separated by a distinct intermediate state (I-state). Additionally, guanidine hydrochloride-mediated denaturation also follows the three-state unfolding (Figure S2), confirming that intermediate formation is indeed an intrinsic property of the protein. A simple explanation for this observation could be the formation of monomeric gp45WT, by disruption of the interactions at each interface of trimeric gp45WT by urea during the first transition, thus giving rise to the I-state. The second transition could then be a result of monomer denaturation. To test our hypothesis, we monitored the urea-mediated loss of secondary structure in gp45WT by CD. Surprisingly, we observe a three-state transition in our CD measurements that resembles the anisotropy data (Figure 3B). Thus, the CD and fluorescence anisotropy data together indicate that the first transition occurring during the chemical denaturation of gp45WT may include both trimer to monomer conversion (lowering of anisotropy) and partial (∼50%) loss of gp45WT secondary structure (as demonstrated by CD data). It is worth noting that unlike thermal denaturation, the chemical unfolding of gp45WT is completely reversible and shows no hysteresis (Figure 3A,B). We further estimated the free energy of unfolding of gp45WT at 0 M urea (ΔG°U) for both transitions using the fraction unfolded CD data. While the ΔG°U value for the first transition (ΔG°U1) is found to be 9.24 ± 0.58 kcal mol−1, for the second transition the free energy of unfolding (ΔG°U2) is estimated to be 10.18 ± 0.62 kcal mol−1. Moreover, the cooperativity of the unfolding as measured in terms of m value is estimated to be −3.531 ± 0.11

Figure 3. Chemical denaturation of gp45WT. (A) Three-state reversible unfolding for gp45WT, monitored using fluorescence anisotropy. The Cm1 and Cm2 values, obtained from fitting the data (fits shown as solid lines), are given in the plot (unfolding data colored green, refolding data colored red). Shown here are representative data sets, for the sake of clarity; the average Cm values derived from three independent experiments are as follows: Cm1 = 2.7 ± 0.2 M, and Cm2 = 5.7 ± 0.2 M. (B) Three-state reversible unfolding for gp45WT was also monitored using far-UV CD at 222 nm. Fraction unfolded data calculated from the CD values are plotted, and the Cm1 and Cm2 values obtained from fitting the data (solid lines) are mentioned. A representative data set is shown for the sake of clarity, and fits are displayed with solid lines. The average Cm values derived from three independent experiments are as follows: Cm1 = 2.6 ± 0.2 M, and Cm2 = 5.7 ± 0.3 M.

and −1.77 ± 0.07 kcal mol−1 M−1 for the first and second transitions, respectively. The data thus suggest that the first transition is more cooperative and results in a significant loss of secondary structure as compared to the second transition. The midpoint of each transition, represented by the Cm, is calculated from fraction unfolded data and is found to be 2.6 ± 0.2 M (first transition, Cm1) and 5.7 ± 0.3 M (second transition, Cm2). The Cm values allow us to conclude that although the bulk protein is denatured during first transition, the remaining structure forms a “stable core” that requires a higher concentration of urea. The Two Domains of gp45WT Display Asymmetric Characteristics. The crystal structure of gp45WT shows that the protein is composed of two domains, the NTD and the CTD, that are connected by a long proline-rich linker.13 We asked if the two transitions observed in the chemical denaturation profile of gp45WT represent the unfolding of the two domains in a stepwise manner. In other words, one of the two domains unfolds first followed by the other domain as the concentration of urea is increased. Because gp45WT has two tryptophans, W92, present in the NTD, and W199, present in the CTD, we hypothesized that of the two tryptophans, W92, which is sandwiched at the NTD− CTD interface of gp45WT (Figure 4A), would be rendered conformationally labile in the monomeric state, which would result in a lowering of anisotropy during denaturation. Indeed, W92 yielded a solvent accessible surface area (SASA, calculated using gp45 structure by Discovery Studio version 4.0) of 19.02 ± 5.31 Å2 when it is present in the trimeric form of the protein, 591

DOI: 10.1021/acs.biochem.5b01204 Biochemistry 2016, 55, 588−596

Article

Biochemistry

anisotropy during denaturation. To explore this further and to understand the nature of the stable core that denatures at high urea concentrations as proposed earlier, we generated single-Trp mutants gp45W92 (carrying a W199F mutation) and gp45W199 (carrying a W92F mutation). We affirmed that the Trp mutation did not adversely affect gp45 behavior by cross-linking (Figure S3A) and Tm estimates (Figure S3B). The anisotropy recorded for the single-Trp mutants provides interesting insights. The chemical denaturation of gp45W92 suggests that NTD unfolds at low urea concentrations (Figure 4B). Furthermore, a sharp decline in the anisotropy is also indicative of trimer dissociating into monomer (Figure 4B). While monitoring gp45W199 denaturation, which will primarily demonstrate the changes occurring in the CTD, we notice two important aspects. The anisotropy decrease at a high urea concentration (Cm ∼ 5.8 M) indicates domain unfolding, and the minor decrease in anisotropy is also observed at ∼1.9 M urea that overlaps well with gp45W92, suggesting trimer dissociation (Figure 4B). Nevertheless, the protein unfolding profile recorded for both mutants using CD remains three-state (Figure 4C). Taken together, our data strongly support the possibility that both trimer to monomer dissociation and NTD unfolding occur at low urea concentrations (compare a Cm of ∼2.2 M for gp45W92 with a Cm1 of ∼2.5 M for gp45WT, derived from anisotropy data) and an intermediate state is formed. Increasing the urea concentration further leads to the unfolding of the CTD and results in complete denaturation of the protein. We next asked which of the two events, i.e., NTD unfolding and trimer dissociation, occurs first. To address this, we monitored the quenching of the W92 fluorescence by acrylamide, at different urea concentrations. Fluorescence quenching is a well-studied phenomenon that suggests how exposed the fluorophore is in a protein molecule.21 Because , in gp45WT, W92 is buried at the interface (SASA calculated to be 19.02 ± 5.31 Å2), a trimer dissociation event would cause W92 to be exposed to solvent (SASA calculated to be 42.67 ± 2.78 Å2), thereby making it accessible to the quencher acrylamide. The Stern−Volmer quenching constant (Ksv) measured for W92 in the gp45W92 protein shows that in the absence of urea (i.e., at 0 M), W92 fluorescence is not quenched by acrylamide, clearly indicating that the tryptophan is inaccessible (Figure 5). Addition of urea leads to an increase in Ksv, suggesting that the tryptophan is exposed, which is possible only when the trimer dissociates. This event is comparable to the anisotropy data (Figure 4B) that confirm that the first transition during chemical denaturation leads to trimer dissociation. It is, however, worth

Figure 4. gp45WT trimer dissociation and NTD unfolding are coupled. (A) Crystal structure of gp45 (Protein Data Bank entry 1CZD) showing the two tryptophan residues W92 (blue) and W199 (red). The dashed line marks the three subunit−subunit interfaces in gp45. The W92 (located in the NTD) is seen buried in the interface, whereas W199 (located in the CTD) is present at the surface. (B) Fluorescence anisotropy values obtained for single-Trp mutants gp45W92 (carrying intact Trp at position 92, blue) and gp45W199 (harboring intact Trp199, red) measured in the presence of increasing urea concentrations are plotted with solid lines representing the fits. The Cm values derived from these fits are mentioned. gp45 W92 shows a single transition corresponding to trimer dissociation (and NTD unfolding). gp45W199, on the other hand, shows a three-state unfolding with two transitions, corresponding to trimer dissociation and CTD unfolding, respectively. (C) Fraction unfolded profiles derived from far-UV CD data recorded at 222 nm are shown for both (gp45W92 and gp45W199). The insets show the CD spectra of the proteins in molar ellipticity (ME) with increasing urea concentrations for selective urea concentrations. Plots in panel B and C represent the average data from three independent experiments.

which is significantly lower than the average SASA for tryptophan on a protein surface.25 Furthermore, the SASA for W92 in a monomeric gp45 is estimated to be 42.67 ± 2.78 Å2, which is close to what is obtained for W199 in both gp45 trimer and monomer (44.32 ± 9.68 Å2). It is conceivable therefore that W92 present in trimeric gp45 would demonstrate a lowering of

Figure 5. Tryptophan fluorescence quenching of gp45WT and the single-Trp mutants by acrylamide. Fluorescence quenching of tryptophan in gp45WT, gp45W92, and gp45W199 by acrylamide at different urea concentrations was monitored to understand tryptophan accessibility. The Ksv values derived at each urea concentration are represented as gray bars, and the corresponding anisotropy values (colored red) of the tryptophan at the same urea concentration are replotted for comparison. The data presented here are an average of two independent experiments. 592

DOI: 10.1021/acs.biochem.5b01204 Biochemistry 2016, 55, 588−596

Article

Biochemistry

However, interestingly, at ∼2.5 M urea, an intermediate mobility band (at position “I”) is also observed. The appearance of this “I” band is in good agreement with the I-state observed in our fluorescence anisotropy and CD data obtained with the chemical denaturation of gp45WT protein (Figure 3A,B). Our fluorescence anisotropy and fluorescence quenching data obtained for gp45WT (Figures 3A and 5) and gp45W92 (Figures 4B and 5) together suggest that at ∼2.5 M urea, the gp45 trimer dissociates into monomer. Very interestingly, the corresponding urea concentration treatment of gp45WT gives rise to the “I” band via BNPAGE (Figure 6A). It is conceivable, therefore, that the “I” band observed via BN-PAGE is formed by monomeric gp45WT that displays retarded mobility. We believe that the I-state observed during gp45 denaturation likely adopts a “swollen” molten globule form. Our results allow us to conclude that the gp45WT trimer dissociates beyond a “threshold” urea concentration (Cm1) with simultaneous denaturation of the NTD (Figure 6B). An intermediate I-state is observed during denaturation, which is formed by monomeric gp45WT having a denatured NTD and a differentially folded CTD. A further increase in urea concentration leads to complete denaturation of the protein, giving rise to a three-state unfolding process. The gp45WT CTD Alone Is Sufficient To Form the I-State and Displays Molten Globule-like Characteristics. To gain deeper insight into the properties of the two domains of gp45WT and to understand the properties of the molten globule state obtained during protein denaturation, we separately cloned, expressed, and purified the two domains of gp45, NTD (gp45NTD, residues 1−107) and CTD (gp45CTD, residues 108−228). The far-UV CD spectra suggest that both proteins are well-folded (Figure S4A,B). However, upon chemical denaturation, contrasting features are observed, further confirming our observations of the asymmetric characteristics of the two domains in the full-length protein. gp45NTD undergoes a nearcomplete denaturation (∼80%) between 2 and 4 M urea as judged by the CD data (Figure 7A). However, this is accompanied by a largely noncooperative decrease in fluorescence anisotropy (Figure 7B), which is in stark contrast with the fluorescence anisotropy data obtained for gp45W92 (Figure 4B). The free energy of unfolding of gp45NTD, calculated from fraction unfolded data, is estimated to be 1.54 ± 0.22 kcal mol−1, which is substantially less than what is calculated for the first transition in gp45WT (ΔG°U1 = 9.24 ± 0.58 kcal mol−1). In addition, the cooperativity of denaturation, measured in terms of m value, is also found to be −0.65 ± 0.26 kcal mol−1 M−1, indicating that the NTD lacks a well-packed hydrophobic core but is, instead, an open, solvated structure. Nevertheless, the Cm of gp45NTD denaturation (2.5 ± 0.7 M) matches well with that of gp45WT. It is thus plausible that either the NTD does not form a stable structure when it is constructed separately or additional interactions are established between the NTD and the remaining full-length gp45WT that stabilize the domain; we speculate the latter to be the reason for NTD stability because the CD data of purified gp45NTD (Figure S4A) suggest that the protein is well folded and does not unfold in up to ∼2 M urea. Very interestingly, gp45CTD resists denaturation by urea to a great extent (up to ∼4 M). However, the unfolding free energy for this protein (3.03 ± 0.82 kcal mol−1) is found to be lower than that of the second transition of gp45WT. Further addition of urea leads to a cooperative unfolding of gp45CTD (Figure 7A), with an m value corresponding to −0.52 ± 0.17 kcal mol−1 M−1, which is higher than what is observed for the second transition in the

noting that the urea concentration at which this occurs largely matches with the concentration observed during the first transition, thus indicating that the two events, NTD denaturation and trimer dissociation, occur somewhat together. Additionally, we also studied the accessibility of Trp in gp45WT and gp45W199 during protein unfolding (Figure 5). gp45W199 (containing Trp199 in the CTD) shows a remarkable property. The W in this molecule is exposed (SASA calculated to be 44.32 ± 9.68 Å2) in the native protein molecule and, expectedly, shows a higher Ksv in the absence of urea (Figure 5). However, surprisingly, the Ksv value decreases with an increasing urea concentration up to 3.0 M urea, indicating that the region harboring W199 undergoes a conformational change that causes “immersion” of W199 (Figure 5); the Trp thus becomes inaccessible to the acrylamide quencher. Very interestingly, this state of the protein overlaps well with the I-state observed in the fluorescence anisotropy and CD data of gp45WT (Figure 3A,B) and gp45W199 (Figure 4B,C), during protein denaturation. A further increase in urea concentration leads to the exposure of W199 (Figure 5), suggesting a complete denaturation of the protein as confirmed by the CD data (Figure 4C). Not surprisingly, the fluorescence quenching profile (Ksv) for gp45WT protein (bearing both the tryptophans) displays an average of the values obtained for both gp45W92 and gp45W199 (Figure 5). Taken together, our data suggest that during chemical denaturation of gp45, the intermediate state is formed because of a significant change in the CTD conformation; we are tempted to postulate that such a conformational change in the CTD of gp45 leads to the formation of a “molten globule”. To further explore the properties of the intermediate state, we performed blue-native PAGE (BN-PAGE) analysis of gp45WT after treating the protein with different concentrations of urea (Figure 6A). The protein shows distinct bands at 0 and 8 M urea (Figure 6A).

Figure 6. Unfolding of gp45WT as captured via BN-PAGE. (A) BNPAGE profile of urea-induced unfolding of gp45WT. The urea concentration is indicated above each lane. The protein migrates as a trimer (T) at lower and as a monomer (M) at higher urea concentrations; at ∼2.5−3.0 M urea, i.e., Cm1, the monomer exists as the I-state (I) and shows retarded gel mobility until ∼5 M (Cm2). A further increase in the urea concentration results in monomeric (M) denatured protein as the predominant species. (B) Cartoon representation of urea-induced unfolding of gp45WT. C denotes the CTD and N the NTD. Step 1 is the first transition followed by I-state formation; step 2 is the second transition followed by complete denaturation. The indol rings refer to W92 (blue) and W199 (red). The NTD (blue) unfolds during the first transition coupled with trimer dissociation, resulting in the exposure of W92, while the CTD (red) takes a molten globule-like conformation leading to the immersion of W199. Beyond the second transition, the protein unfolds completely. 593

DOI: 10.1021/acs.biochem.5b01204 Biochemistry 2016, 55, 588−596

Article

Biochemistry

W199 fluorescence anisotropy measured after subjecting gp45CTD to increasing Gdn-HCl concentrations (Figure S4D) shows that protein denaturation by Gdn-HCl not only follows a profile similar to that of urea but also shows a rather clear and “long-lasting” I-state that survives to up to 2.5 M Gdn-HCl. We further confirmed the existence of the molten globule state by comparing the mobility retardation of gp45WT (Figure 6A) with that of gp45CTD (Figure 8A) via BN-PAGE. Here, gp45CTD

Figure 7. Differential stability of the NTD and CTD of gp45WT. (A) Fraction unfolded profiles derived from far-UV CD data recorded at 222 nm for both gp45NTD and gp45CTD proteins. Solid lines represent the fits; Cm values were derived from these fits and are given in the plot. gp45NTD shows a noncooperative unfolding, whereas gp45CTD undergoes cooperative unfolding. The plots represent the average of two independent sets of experimental data. (B) Fluorescence anisotropy values derived for the tryptophan present in gp45NTD and gp45CTD measured in the presence of increasing urea concentrations plotted with solid lines representing the fits. Cm values derived from these fits are given. Note that gp45NTD shows a noncooperative denaturation profile, which is remarkably different from that seen in gp45W92 (see Figure 4B). Also, note an initial increase in the anisotropy in gp45CTD (see the text for details). These plots represent the average of three independent sets of experimental data.

Figure 8. CTD alone shows I-state formation. (A) BN-PAGE profiles of gp45CTD and gp45NTD demonstrate that the CTD is responsible for the retarded mobility band as seen in the gp45WT protein (Figure 6A). Such a state is not observed in gp45NTD. The denatured proteins and the intermediate state are marked as 1 and 2, respectively. (B) Coomassiestained SDS−PAGE gel demonstrating the pulse proteolysis examination of gp45WT and gp45CTD. Each lane shows a digestion profile of the protein at a specific urea concentration (marked above the lane) after treatment with proteinase K for 1 min. WT refers to the full-length gp45WT protein; CTD refers to the digested out C-terminal domain, and CTD* refers to the C-terminal domain lacking 27 residues from the Cterminal end. Residues 209−228 as per the crystal structure and seven residues containing a vector specific hexahistidine tag are digested out, leaving the remaining CTD protease protected. Lane C refers to the control protein that was not treated with either urea or proteinase K.

gp45WT protein (−1.77 ± 0.07 kcal mol−1 M−1), suggesting that the denaturation of gp45CTD is less cooperative than that of the C-terminal domain in gp45WT. Nevertheless, gp45CTD unfolds with a Cm of 5.8 ± 0.3 M, which is similar to the Cm2 of gp45WT (i.e., 5.7 M). This concurs with our previous observations and suggests that the second transition is due to CTD unfolding. The fluorescence anisotropy recorded for gp45CTD displays a very interesting property (Figure 7B). Addition of urea beyond ∼4 M causes an unexpected increase in W199 anisotropy, suggesting that the protein adopts an I-state with non-native contacts of the indole (Figure 7B). Furthermore, the anisotropy values for gp45CTD, beyond ∼4 M, are in good agreement with those of gp45W199 (Figure S4C). This immediately suggests that isolated gp45CTD attains its native structure. Moreover, data reveal that the I-state of gp45WT arises almost exclusively from the CTD, because the NTD is largely unstructured at these urea concentrations We wish to add here that gp45CTD, while experiencing structural loss (Figure 7A), shows an increase in anisotropy during the formation of the molten globule-like state (Figure 7B) at similar urea concentrations. This is unlike CTD unfolding in gp45WT, where it is initiated only after the intermediate state has been formed (Figure 3B). Such differential unfolding of the C-terminal domain in gp45WT and in gp45CTD proteins results in nonoverlapping fraction unfolded curves (data not shown; compare the profiles obtained at urea concentrations of up to ∼4.5−5.0 M in Figures 3B and 7A) and thus makes the equilibrium unfolding parameters (m value) different. However, because the decrease in anisotropy in gp45CTD and gp45WT occurs after the intermediate state has been achieved, the anisotropy curves overlap very well (Figure S4C). The change in

shows a significant retardation of the protein band mobility when subjected to higher urea concentrations; gp45NTD, on the other hand, does not display such behavior (Figure 8A). Thus, our domain truncation studies clearly establish that gp45WT attains an intermediate molten globule state during unfolding that is primarily achieved by structural changes occurring in the CTD. We next asked if CTD of gp45 is indeed a well-packed structure that not only resists urea-mediated denaturation but also gives rise to an altered conformation molecule. It is wellknown that an unstructured protein is highly susceptible to protease-mediated cleavage, whereas a compact well-folded protein resists such digestion. Because gp45CTD resists ureamediated denaturation, we subjected gp45 CTD to pulse proteolysis and compared the digestion profile of the protein to that of gp45WT (Figure 8B). Upon digestion of gp45WT by proteinase K in the presence of increasing urea concentrations, two prominent lower-molecular weight bands are obtained 594

DOI: 10.1021/acs.biochem.5b01204 Biochemistry 2016, 55, 588−596

Article

Biochemistry

The observation that the NTD of gp45 is more dynamic than the CTD is similar to what has been reported for E. coli β clamp domain I that is relatively more dynamic than domains II and III.26 It has been shown that gp45 can load itself onto the DNA without clamp loader complex (CLC) assistance,27 which likely happens because of the highly dynamic nature of the NTD. We hypothesize that for gp45 to open and close without being completely disassembled into monomer, the interface should be made of two components: the flexible component, which can be tweaked by the clamp loader proteins, and the rigid component, which allows the realignment and interface closing. We further suggest that the NTD provides flexibility to gp45, allowing the ring to open at the subunit interface, with the help of CLC, and facilitates the loading of gp45 onto the DNA. Subsequently, the rigidity provided by the CTD may allow the protein to re-form a closed trimer and maintain its state. It is thus likely that a combination of these two components acts as the key player behind the primary (replication) and auxiliary (transcription) functions of gp45 within the cell. The T4 phage DNA polymerase (gp43), for example, has been shown to interact with gp45 at its interface.28 Furthermore, the interacting motif present at the C-terminus of gp43 is also found at the C-terminus of T4 late promoter transcription factors gp55 and gp33. The T4 gp45 clamp is also interesting from an evolutionary point of view. The trimeric phage clamp resembles the archaeal and eukaryotic trimeric PCNAs but differs from the dimeric β clamp of bacteria. It is thus puzzling that during evolution the bacterial clamp was designed to be a dimer whereas the archaea and the eukarya retained the trimeric clamp. It is plausible that the archaeal and eukaryotic PCNAs perform certain additional functions, besides helping in DNA replication, for which a trimeric molecule is required. Therefore, a detailed study of the biology of these clamps will most certainly help in understanding why two anatomically different clamps (dimer and trimer) are required to conduct essentially the same function.

(Figure 8B). These bands were identified as the CTD of gp45 by subjecting them to in-gel trypsin digestion and MALDI mass spectrometry (data not shown); here, the CTD* band corresponds to the His tag-less version of CTD, which is generated after losing the 27-residue C-terminal peptide (data not shown). We did not observe any band corresponding to the NTD in this experiment. The data strongly indicate that the Cterminal domain of gp45WT is a well-folded structure that resists digestion by protease. Not surprisingly, therefore, gp45CTD upon digestion by proteinase K in the presence of increasing urea concentrations gives rise to only the CTD* band similar to that observed in the case of gp45WT (Figure 8B). These experiments thus clearly demonstrate that the N-terminal domain of gp45 is fragile and is susceptible to digestion by protease, whereas the Cterminal domain is very rigid and resists such digestion.

■

DISCUSSION

The sliding clamp proteins contribute to the processivity of the DNA polymerase and are found in all the domains of life. This work is an attempt to understand the biophysical properties of a bacteriophage sliding clamp that structurally resembles the bacterial, archaeal, and eukaryotic clamps.5 Additionally, T4 sliding clamp gp45, besides supporting DNA replication, is also required for the phage late promoter transcription. It is thus an interesting molecule that should be explored in greater detail to examine how two contrasting activities exist in a single protein molecule. Here, we report several important aspects about T4 gp45 after performing a detailed biophysical examination of the protein using several methods such as CD, fluorescence, BNPAGE, and pulse proteolysis. We also demonstrate for the first time the existence of a molten globule-like state of the T4 gp45 molecule during denaturation. The heat denaturation of gp45 displayed a two-state unfolding transition followed by aggregation. On the other hand, the urea-mediated denaturation of the protein revealed a three-state transition with a prominent intermediate state that has lost the tertiary structure but retains some secondary structure with an altered conformation. The thermodynamic parameters also suggest that a significant amount of protein denaturation occurs during the first transition, yet the intermediate state requires a higher urea concentration to achieve complete denaturation. The dissection of gp45 into its domains yields an important understanding of the behavior of the domains. The data obtained for the N-terminal domain from both the full-length and isolated NTD protein suggest that the domain is largely unstable as compared to the CTD. A point worth noting here is that the fluorescence anisotropy profiles of W92 in both gp45W92 and gp45NTD do not match when the proteins are subjected to increasing urea concentrations. Our data allow us to suggest that the largely unstable NTD is stabilized by its interactions with the CTD of another subunit in the gp45 trimer. Our suggestion may further explain why the first transition observed during ureamediated denaturation of gp45 involves both trimer dissociation and NTD unfolding; because the NTD is stabilized by the interactions with the CTD, perturbation at the interface causes the trimer to dissociate and unfolding of NTD. However, the possibility of the NTD unfolding prior to trimer dissociation cannot be ruled out. Here, the argument that the NTD in its isolated form (gp45NTD) will behave in a manner completely different from that of the gp45 trimer because of its (mis)folding appears to be invalid because CD data of gp45NTD demonstrate a well-folded protein.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.5b01204. Thermal stability of gp45WT trimer as analyzed by CD and fluorescence (Figure S1), Gdn-HCl-induced denaturation of gp45WT (Figure S2), single-tryptophan mutants of gp45 that are stable in solution (Figure S3), and properties of the NTD and CTD of gp45 (Figure S4) (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Microbiology and Molecular Biology Laboratory, Indian Institute of Science Education and Research (IISER), Transit campus, ITI (Gas Rahat) Building, Govindpura, Bhopal 462023, Madhya Pradesh, India. Telephone: +91-755-4092318. Fax: +91755-4092392. E-mail: [email protected]. Funding

This work is supported by a grant (SR/FT/LS-06/2010) from Science and Engineering Research Board, Government of India, to V.J. M.I.S. is supported by a senior research fellowship from the Council of Scientific & Industrial Research (CSIR), Government of India. Notes

The authors declare no competing financial interest. 595

DOI: 10.1021/acs.biochem.5b01204 Biochemistry 2016, 55, 588−596

Article

Biochemistry

■

(16) Singh, M. I., and Jain, V. (2013) Tagging the expressed protein with 6 histidines: rapid cloning of an amplicon with three options. PLoS One 8, e63922. (17) Maurya, S. R., Chaturvedi, D., and Mahalakshmi, R. (2013) Modulating lipid dynamics and membrane fluidity to drive rapid folding of a transmembrane barrel. Sci. Rep. 3, 1989. (18) Moon, C. P., and Fleming, K. G. (2011) Using tryptophan fluorescence to measure the stability of membrane proteins folded in liposomes. Methods Enzymol. 492, 189−211. (19) Moon, C. P., and Fleming, K. G. (2011) Side-chain hydrophobicity scale derived from transmembrane protein folding into lipid bilayers. Proc. Natl. Acad. Sci. U. S. A. 108, 10174−10177. (20) Schneider, C. A., Rasband, W. S., and Eliceiri, K. W. (2012) NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671−675. (21) Lakowicz, J. R. (2009) Principles of Fluorescence Spectroscopy, 3rd ed., Springer Science and Business Media, Berlin. (22) Maurya, S. R., and Mahalakshmi, R. (2013) Modulation of human mitochondrial voltage-dependent anion channel 2 (hVDAC-2) structural stability by cysteine-assisted barrel-lipid interactions. J. Biol. Chem. 288, 25584−25592. (23) Syal, K., Joshi, H., Chatterji, D., and Jain, V. (2015) Novel pppGpp binding site at the C-terminal region of the Rel enzyme from Mycobacterium smegmatis. FEBS J. 282, 3773−3785. (24) Bullock, A. N., Henckel, J., DeDecker, B. S., Johnson, C. M., Nikolova, P. V., Proctor, M. R., Lane, D. P., and Fersht, A. R. (1997) Thermodynamic stability of wild-type and mutant p53 core domain. Proc. Natl. Acad. Sci. U. S. A. 94, 14338−14342. (25) Pagano, B., Jama, A., Martinez, P., Akanho, E., Bui, T. T., Drake, A. F., Fraternali, F., and Nikolova, P. V. (2013) Structure and stability insights into tumour suppressor p53 evolutionary related proteins. PLoS One 8, e76014. (26) Fang, J., Engen, J. R., and Beuning, P. J. (2011) Escherichia coli processivity clamp beta from DNA polymerase III is dynamic in solution. Biochemistry 50, 5958−5968. (27) Reddy, M. K., Weitzel, S. E., and von Hippel, P. H. (1993) Assembly of a functional replication complex without ATP hydrolysis: a direct interaction of bacteriophage T4 gp45 with T4 DNA polymerase. Proc. Natl. Acad. Sci. U. S. A. 90, 3211−3215. (28) Alley, S. C., Jones, A. D., Soumillion, P., and Benkovic, S. J. (1999) The carboxyl terminus of the bacteriophage T4 DNA polymerase contacts its sliding clamp at the subunit interface. J. Biol. Chem. 274, 24485−24489.

ACKNOWLEDGMENTS We thank Prof. E. Peter Geiduschek (University of California at San Diego) for providing us with the T4 gp45 construct and for his critical comments about the manuscript. We also thank Dr. R. Mahalakshmi [Indian Institute of Science Education and Research (IISER)] for the constructive inputs during manuscript preparation.

■

ABBREVIATIONS Gdn-HCl, guanidine hydrochloride; NTD, N-terminal domain; CTD, C-terminal domain; DSC, differential scanning calorimetry; BN-PAGE, blue-native polyacrylamide gel electrophoresis; DTT, dithiothreitol; EDTA, ethylenediaminetetraacetic acid; SEC, size exclusion chromatography; CD, circular dichroism.

■

REFERENCES

(1) Huang, C. C., Hearst, J. E., and Alberts, B. M. (1981) Two types of replication proteins increase the rate at which T4 DNA polymerase traverses the helical regions in a single-stranded DNA template. J. Biol. Chem. 256, 4087−4094. (2) Kelman, Z., and O’Donnell, M. (1995) DNA polymerase III holoenzyme: structure and function of a chromosomal replicating machine. Annu. Rev. Biochem. 64, 171−200. (3) Kong, X. P., Onrust, R., O’Donnell, M., and Kuriyan, J. (1992) Three-dimensional structure of the beta subunit of E. coli DNA polymerase III holoenzyme: a sliding DNA clamp. Cell 69, 425−437. (4) O’Donnell, M., Kuriyan, J., Kong, X. P., Stukenberg, P. T., and Onrust, R. (1992) The sliding clamp of DNA polymerase III holoenzyme encircles DNA. Mol. Biol. Cell 3, 953−957. (5) Kelman, Z., and O’Donnell, M. (1995) Structural and functional similarities of prokaryotic and eukaryotic DNA polymerase sliding clamps. Nucleic Acids Res. 23, 3613−3620. (6) Argiriadi, M. A., Goedken, E. R., Bruck, I., O’Donnell, M., and Kuriyan, J. (2006) Crystal structure of a DNA polymerase sliding clamp from a Gram-positive bacterium. BMC Struct. Biol. 6, 2. (7) Gui, W. J., Lin, S. Q., Chen, Y. Y., Zhang, X. E., Bi, L. J., and Jiang, T. (2011) Crystal structure of DNA polymerase III beta sliding clamp from Mycobacterium tuberculosis. Biochem. Biophys. Res. Commun. 405, 272− 277. (8) Indiani, C., and O’Donnell, M. (2006) The replication clamploading machine at work in the three domains of life. Nat. Rev. Mol. Cell Biol. 7, 751−761. (9) Krishna, T. S., Kong, X. P., Gary, S., Burgers, P. M., and Kuriyan, J. (1994) Crystal structure of the eukaryotic DNA polymerase processivity factor PCNA. Cell 79, 1233−1243. (10) Niiranen, L., Lian, K., Johnson, K. A., and Moe, E. (2015) Crystal structure of the DNA polymerase III beta subunit (beta-clamp) from the extremophile Deinococcus radiodurans. BMC Struct. Biol. 15, 5. (11) Yao, N., Turner, J., Kelman, Z., Stukenberg, P. T., Dean, F., Shechter, D., Pan, Z. Q., Hurwitz, J., and O’Donnell, M. (1996) Clamp loading, unloading and intrinsic stability of the PCNA, beta and gp45 sliding clamps of human, E. coli and T4 replicases. Genes Cells 1, 101− 113. (12) Fang, J., Nevin, P., Kairys, V., Venclovas, C., Engen, J. R., and Beuning, P. J. (2014) Conformational analysis of processivity clamps in solution demonstrates that tertiary structure does not correlate with protein dynamics. Structure 22, 572−581. (13) Moarefi, I., Jeruzalmi, D., Turner, J., O’Donnell, M., and Kuriyan, J. (2000) Crystal structure of the DNA polymerase processivity factor of T4 bacteriophage. J. Mol. Biol. 296, 1215−1223. (14) Millar, D., Trakselis, M. A., and Benkovic, S. J. (2004) On the solution structure of the T4 sliding clamp (gp45). Biochemistry 43, 12723−12727. (15) Trakselis, M. A., Alley, S. C., Abel-Santos, E., and Benkovic, S. J. (2001) Creating a dynamic picture of the sliding clamp during T4 DNA polymerase holoenzyme assembly by using fluorescence resonance energy transfer. Proc. Natl. Acad. Sci. U. S. A. 98, 8368−8375. 596

DOI: 10.1021/acs.biochem.5b01204 Biochemistry 2016, 55, 588−596