SINGLE-STRAND BINDING PROTEIN

Abstract

The present invention relates to the use of a single-strand DNA binding protein (SSB) which exhibits at least 50% of its maximum ssDNA binding capability in the presence of 500 mM of sodium ions, to dehybridize a DNA molecule or to prevent hybridisation of a complementary ssDNA, wherein the SSB comprises the amino acid sequence of SEQ ID NO:1 or an amino acid sequence which is at least 75% identical to SEQ ID NO:1, or a functional fragment thereof, and wherein the DNA molecule or ssDNA is present in or exposed to a solution containing one or more of the following (i) at least 350 mM of sodium ions; (ii) at least 50 mM of potassium ions; (iii) at least 150 mM of magnesium ions; or (iv) at least 200 mM of calcium ions.

Claims

1. Use of a single-strand DNA binding protein (SSB) which exhibits at least 50% of its maximum ssDNA binding capability in the presence of 500 mM of sodium ions, to dehybridize a DNA molecule, wherein the SSB comprises the amino acid sequence of SEQ ID NO:1 or an amino acid sequence which is at least 75% identical to SEQ ID NO:1, or a functional fragment thereof, wherein the amino acid at position 17 and/or 71 has been substituted with a replacement amino acid that lacks a negative charge on its side chain, and wherein the DNA molecule is present in or exposed to a solution containing one or more of the following: (i) at least 350 mM of sodium ions, (ii) at least 50 mM of potassium ions, (iii) at least 150 mM of magnesium ions, or (iv) at least 200 mM of calcium ions.

2. The use of claim 1, wherein the SSB comprises the amino acid sequence of SEQ ID NO:1 or an amino acid sequence which is at least 90% identical thereto, wherein the amino acid at position 17 and/or 71 has been substituted with a replacement amino acid that lacks a negative charge on its side chain.

3. The use of claim 1, wherein the replacement amino acid carries a positive charge on its side chain at pH 7.0.

4. The use of claim 3, wherein the replacement amino acid is selected from the group consisting of lysine, histidine, arginine, tyrosine, asparagine and glutamine.

5. The use of claim 4, wherein the replacement amino is lysine or arginine

6. The use of claim 5, wherein the replacement amino acid is lysine.

7. The use of claim 1, wherein the SSB comprises the amino acid sequence of SEQ ID NO:2.

8. The use of claim 1, wherein said use is performed during a method of: i) nucleic acid amplification, purification or sequencing, preferably nanopore sequencing; ii) site directed mutagenesis; iii) examining nucleic acid structures in a sample using a microscope; iv) restriction enzyme digestion; v) reverse transcription; vi) enhancing the activity of T4 polymerase; or vii) protecting ssDNA or RNA from nuclease digestion.

9. Use of a single-strand DNA binding protein (SSB) which exhibits at least 50% of its maximum ssDNA binding capability in the presence of 500 mM of sodium ions to prevent hybridisation of complementary ssDNA, wherein the SSB comprises the amino acid sequence of SEQ ID NO:1 or an amino acid sequence which is at least 75% identical to SEQ ID NO:1, or a functional fragment thereof, wherein the amino acid at position 17 and/or 71 has been substituted with a replacement amino acid that lacks a negative charge on its side chain, and wherein the ssDNA is present in or exposed to a solution containing one or more of the following: (i) at least 350 mM of sodium ions, (ii) at least 50 mM of potassium ions, (iii) at least 150 mM of magnesium ions, or (iv) at least 200 mM of calcium ions.

10. The use of claim 9, wherein the SSB comprises the amino acid sequence of SEQ ID NO:1 or an amino acid sequence which is at least 90% identical thereto, wherein the amino acid at position 17 and/or 71 has been substituted with a replacement amino acid that lacks a negative charge on its side chain.

11. The use of claim 10, wherein the replacement amino acid carries a positive charge on its side chain at pH 7.0.

12. The use of claim 11, wherein the replacement amino acid is selected from the group consisting of lysine, histidine, arginine, tyrosine, asparagine and glutamine.

13. The use of claim 12, wherein the replacement amino is lysine or arginine

14. The use of claim 13, wherein the replacement amino acid is lysine.

15. The use of claim 9, wherein the SSB comprises the amino acid sequence of SEQ ID NO:2.

16. The use of claim 9, wherein said use is performed during a method of: i) nucleic acid amplification, purification or sequencing, preferably nanopore sequencing; ii) site directed mutagenesis; iii) examining nucleic acid structures in a sample using a microscope; iv) restriction enzyme digestion; v) reverse transcription; vi) enhancing the activity of T4 polymerase; or vii) protecting ssDNA or RNA from nuclease digestion.

Description

[0130] The invention is further described in the following non-limiting examples and with reference to:

[0131] FIGS. 1A), B), C) and D) which show binding of S. ruber SSB (wild-type and mutant forms) and E. coli SSB to DNA in the presence of monovalent and divalent salts.

[0132] FIG. 2A) which shows SrSSB D17KD71K mutant in the presence of a mixture of salts (350 mM NaCl, 600 mM KCl and 150 mM Mg(OAc).sub.2) without ssDNA or incubated with M13 phage ssDNA for 5 or 120′, visualized with electron microscopy; B) which shows binding of S. ruberSSB (wild-type and mutant forms) and E. coli SSB to DNA in the presence of a mixture of salts (350 nM NaCl, 600 nM KCl and 150 mM Mg(OAc).sub.2).

[0133] FIG. 3 which shows the apparent oligomer mass of S. ruber wildtype SSB determined using size exclusion chromatography at 0.15M, 2 M and 4 M NaCl, and the S. ruber SSB cdel 111 terminal deletion mutant, determined using size exclusion chromatography at 0.15M NaCl.

[0134] FIG. 4 which shows the crystal structure of S. ruberSSB in high salt conditions. The structure shows filament-like architecture consisting of polymerized tetramers. (A) Polymerization interface is formed by two adjacent dimers belonging to separate tetramers and repeated along the filament. (B) Crystal packing of tetramers in the filament. Short DNA fragments of bound DNA are soaked-in (dT)8 oligonucleotides.

EXAMPLES

[0135] Identification and Cloning of wt S. ruber SSB and D17K D71K Mutant

[0136] The single SSB open reading frame was identified in S. ruber strain DSM 13855 genome (GenBank: CP000160.1) by using blastp algorithm with E. coli SSB sequence as query (NCBI: WP_042201710.1) (J. Mol. Biol [1990] 215, 403-410). Two point mutations that substitute aspartic acid to lysine in positions 17 and 71 were introduced, yielding the D17K D71K mutant.

[0137] Genes encoding wild-type SSB and D17K D71 K mutant were synthesized with codon optimization for expression in E. coli (using DNA 2.0). In the codon-optimized genes 26% of the original nucleotides were replaced and the GC-content was changed from the original 63% to 47%, which made the resulting genes better suited for expression in E. coli.

TABLE-US-00003 TABLE 1 gene sequences SEQ ID No. NAME SEQUENCE 3 wt SrSSB 5′- Coding ATGGCACGCGGAGTCAACAAGGTCATTCTCATCGGCAAC Sequence CTCGGCGACGATCC (CDS) CGAACTGCGGTACACCGGCAGCGGGACGGCTGTCTGCAA CATGTCGCTCGCGACCAACGAAACCTACACCGATAGCGA CGGCAATGAGGTGCAAAACACCGAGTGGCACGACGTCGT GGCGTGGGGGCGGCTCGGAGAGATCTGCAACGAGTACC TTGACAAGGGCTCCCAGGTCTACTTCGAGGGCAAGCTCC AAACCCGCTCTTGGGAGGACCGCGACAACAACACGCGCT ACTCGACGGAGGTGAAGGCCCAAGAGATGATGTTCCTCG ACAGCAATCGCCAGGGCGGGGCGGACATGGACGGCTTC GACCAGACCCGTGGGGACGAATCCCTGGACCAAACCCGC CAGGAGCAGCCCGCCGGCTCTTCCGGTCCGCAGCCTGG GCAGCAGGCGTCCTCCGGGGGCGAGGACGAGGACACAT TCGAGCCGGACGATGATCTTCCGTTCTAG-3′ 4 wt SrSSB 5′- codon ATGGCACGTGGTGTGAATAAAGTGATTCTGATTGGTAATC optimized, TGGGTGATGATCCG synthetic GAACTGCGTTATACCGGTAGCGGCACCGCAGTTTGTAATA gene TGAGCCTGGCAACCAATGAAACCTATACCGATAGTGATGG TAATGAAGTGCAGAATACCGAATGGCATGATGTTGTTGCA TGGGGTCGTCTGGGTGAAATTTGTAATGAATATCTGGATA AAGGCAGCCAGGTGTATTTTGAAGGTAAACTGCAGACCCG TAGCTGGGAAGATCGTGATAATAACACCCGTTATAGCACC GAAGTTAAAGCCCAAGAAATGATGTTTCTGGATAGCAATC GTCAGGGTGGTGCAGATATGGATGGTTTTGATCAGACCC GTGGTGATGAAAGCCTGGATCAGACACGTCAAGAACAGC CTGCAGGTAGCAGCGGTCCGCAGCCTGGTCAGCAGGCAA GCAGCGGTGGTGAAGATGAAGATACCTTTGAACCGGATG ATGATCTGCCGTTT-3′ 5 SrSSB 5′- D17K D71K ATGGCACGTGGTGTGAATAAAGTGATTCTGATTGGTAATC TGGGTGATAAACCG GAACTGCGTTATACCGGTAGCGGCACCGCAGTTTGTAATA TGAGCCTGGCAACCAATGAAACCTATACCGATAGTGATGG TAATGAAGTGCAGAATACCGAATGGCATGATGTTGTTGCA TGGGGTCGTCTGGGTGAAATTTGTAATGAGTACTTGAAAA AAGGCAGCCAGGTGTATTTTGAAGGTAAACTGCAGACCCG TAGCTGGGAAGATCGTGATAATAACACCCGTTATAGCACC GAAGTTAAAGCCCAAGAAATGATGTTTCTGGATAGCAATC GTCAGGGTGGTGCAGATATGGATGGTTTTGATCAGACCC GTGGTGATGAAAGCCTGGATCAGACACGTCAAGAACAGC CTGCAGGTAGCAGCGGTCCGCAGCCTGGTCAGCAGGCAA GCAGCGGTGGTGAAGATGAAGATACCTTTGAACCGGATG ATGATCTGCCGTTT-3′

[0138] Constructs for expression of SrSSB protein variants in E. coli were cloned with a Fast Cloning method (BMC Biotechnol [2011] 11, p 92). Insert fragments were generated by PCR reaction with SrSSB i FW (forward) and SrSSB i RW (reverse) oligonucleotides with wt SrSSB or SrSSB D17K D71 K synthetic genes as templates. The vector fragment was generated by PCR reaction with SrSSB v RW and SrSSB v FW oligonucleotides with pCOLD II expression vector (Takara) as template. For the removal of template DNA Dpnl restriction enzyme was added and fragments were mixed and incubated for 3 h at 37° C. The mixture was than transformed NovaBlue competent cells (Novagene), yielding constructs for expression of both genes in E. coli cells.

[0139] For generation of both inserts and the vector fragment Phusion High-Fidelity DNA Polymerase (Thermo) was used. Oligonucleotides used for gene cloning were synthesized by Integrated DNA Technologies. Cloned constructs were verified by DNA sequencing.

TABLE-US-00004 TABLE 2 oligonucleotides SEQ ID No. NAME SEQUENCE  6 SrSSB i FW 5′-CTTTACTTCCAGGGGGCCATGGCACGTGGTGTGAAT-3′  7 SrSSB v 5′-ATTCACACCACGTGCCATGGCCCCCTGGAAGTAAAG-3′ RW  8 SrSSB V 5′-GATGATCTGCCGTTTTAATGAGGATCCGAATTCAAG-3′ FW  9 SrSSB i RW 5′-CTTGAATTCGGATCCTCATTAAAACGGCAGATCATC-3′ 10 MB 5′- GGCCCG[S1]AGGAGGAAAGGACATCTTCTAGCA[S2]ACGGG CCGTCAAGTTCATGGCCAGTCAAGTCGTCAGAAATTTCGC ACCAC-3′ [S1] = [dt-DABCYL] [S2] = [dt-FAM] FAM = carboxyfluorescein DABCYL = 4-(4′-dimethylaminophenylazo) benzoic acid Underlined complementary bases form a 8 base pair, double- stranded stem 11 SPR 150 nt 5′-(BTN)- AAAGGGTATTGACGGACCAGATGTAGCGTGGCAGAAAAG GGTATTGAC GGACCAGATGTAGCGTGGCAGAGACTGAAAGGGTATTGA CGGACCAGATGTAGCGTGGCAGAAAAGGGTATTGACGGA CCAGATGTAGCGTGGCAGAGACTG-3′ BTN = biotin

Protein Expression and Purification

[0140] E. coli cells strain B121 transformed with wt SrSSB expression plasmid were grown overnight, than diluted 1:50 with fresh Terrific Broth medium supplemented with ampicillin (100 mg/ml). Cells were cultivated until OD600 reached 0.5 then cultures were cooled down to 18° C. Recombinant protein expression was induced by addition of 0.5 mM IPTG and cultivation was continued at 18° C. overnight. The cells were than harvested by centrifugation (8000×g, 4° C., 10 min), and resuspended in buffer A (Tris-HCl 25 mM pH 7.4, NaCl 0.5 M, imidazole 10 mM, 8-mercaptoethanol 5 mM, Tween 20 0.05%, glycerol 10%). After disruption of cells in a French press (27 kpsi applied) the crude extract was clarified by centrifugation (65,000×g, 4° C., 30 min.) and supernatant was used for protein purification by Immobilized Metal Affinity Chromatography (IMAC) with a 1 mL HisTrap HP column (GE Healthcare). Prior to loading the column was equilibrated with 5 column volumes (CV) of buffer A. After the clarified lysate was applied to the column, unbound material was removed by a 5 CV wash with buffer A. Protein was than eluted in a linear gradient from 10 mM to 500 mM imidazole in 10 CV of buffer A. Fractions containing SSB were detected by SDS-PAGE, after which they were combined and concentrated using Amicon Ultra-15 Filter Device with molecular weight cut-off of 100 kDa (Millipore, USA). Final dialysis was performed against buffer B (Tris-HCl 25 mM pH 7.4, NaCl 0.15 M, β-mercaptoethanol 5 mM, glycerol 10%). The purification of D17K D71K mutant was performed according to the same protocol. The purity of purified SSB proteins was estimated using SDS-PAGE (at over 95%), concentration was measured fluorometrically.

Single-Strand DNA Binding Activity Measurements

Fluorescence Measurements

[0141] Wild-type S. ruber SSB and the D17K D71K mutant were tested for their ability to bind ssDNA in reaction buffer supplemented with NaCl, KCl, MgCl.sub.2 and CaCl.sub.2). E. coli SSB was used as a well characterized, reference protein.

[0142] In the assay a change of fluorescence of a fluorophore-labelled oligonucleotide (Table 2, SEQ ID NO:10) is being detected. The oligonucleotide MB is designed to form a partial hairpin with a short, eight base pair (bp) double-strand stem and a 24 base single-strand loop. The oligo is labeled with a fluorescent dye—FAM (carboxyfluorescein and DABCYL (4-(4′-dimethylaminophenylazo) benzoic acid) which quenches FAM fluorescence. When the oligonucleotide is in its native, hairpin state both 5′ and 3′ ends of the stem are in proximity and FAM fluorescence is quenched by DABCYL. Binding of SSB protein to the 24 nucleotide loop disrupts the hydrogen bonds in the double-strand stem and forces dehybridization of the oligonucleotide. Spatial separation of FAM and DABCYL allows measurement of dye fluorescence (Tang et al. Chemical Communications [2011] 47 (19), p 5485-5487)].

[0143] Both the wild-type and the D17K D71 K mutant proteins are able to efficiently bind and dehybridize the fluorescent probe in high salt concentrations of both mono- and divalent salts. The nucleoprotein complex formation with ssDNA probe in the presence of sodium chloride (NaCl), potassium chloride (KCl) and magnesium chloride (MgCl.sub.2) in ranges from 10 mM to 1 M, 0M to 1M and 0M to 250 mM respectively, is shown in FIG. 1A-C.

[0144] Reactions contained 0.125 μM of molecular beacon and 0.25 μM of E. coli SSB tetramer (single binding unit), 4 μM of wt SrSSB and 2 μM of D17K D71 K. Reaction mixtures were set up in triplicates in 50 μl of reaction buffer MB (10 mM Tris-HCl pH 7, 10 mM NaCl) containing the fluorescent probe and supplemented with an indicated concentration of NaCl, KCl or MgCl.sub.2 in 96-well plate (Corning). The binding reaction was initiated by addition of the indicated SSB protein, next reactions were incubated at 25° C. for 15′ and the fluorescence was measured with excitation at 495 nm and emission at 520 nm on the Gemini plate reader (Molecular Devices). All graphs were normalized to 100% of initial response, measured in fluorescence units.

[0145] With the increasing concentration of sodium chloride, potassium chloride and magnesium chloride both SrSSB variants maintain their ability to bind and dehybridize ssDNA. The D17K D71 K mutant shows levels of binding very similar to wt protein but at only half of the concentration, which indicates its increased affinity to ssDNA.

[0146] When increasing salt concentrations from 200-400 mM (for NaCl), binding affinity of previously described E. coli SSB protein drops drastically (Biochemistry [1988] 27, 456-471). This drop is due both to disruption of electrostatic interactions between the protein and the nucleic acid and the increasing stability of segments of double-strand DNA formed by complementary ssDNA. Consistent with previous observations, the affinity of reference SSB from E. coli to ssDNA decreased significantly with the increased concentration of all tested salts.

[0147] This experiment was also performed to test the susceptibility of E. coli, wild type and mutant S. ruber SSB to increased concentrations of CaCl.sub.2). The results are shown in FIG. 1D.

Surface Plasmon Resonance

[0148] Higher binding affinity of the D17K D71K mutant was also confirmed by probing the protein-ssDNA interaction strength directly with the use of Surface Plasmon Resonance on a T200 instrument (BIACORE). This biosensor-based technique enables real-time measurement of binding parameters that enables calculation of equilibrium dissociation constant (KD) of different complexes, including nucleoprotein complexes. To ensure that the measurement is not biased by nucleotide sequence, a randomly generated, 150 nucleotide, 5′-biotynylated ssDNA oligonucleotide was used (Table 2, SEQ ID NO:11). Oligonucleotides were attached to the surface of the SA S-type chip (BIACORE) via biotin-streptavidin interaction. Next, both proteins were run in dilutions: 10 nM, 20 nM, 40 nM, 100 nM, 150 nM, 200 nM, 300 nM over the chip surface in MB buffer (10 mM Tris-HCl pH 7, 10 mM NaCl). The binding event was detected by mass change on the chip surface, which enabled real time measurements of the kinetic parameters of protein-ssDNA interaction. The kinetic parameters are given in the table in Table 3.

TABLE-US-00005 TABLE 3 S. ruber SSB version KD Chi.sup.2 Wt 3.43 × 10.sup.−8 M 0.529 D17K D71K 1.91 × 10.sup.−8 M 0.624

[0149] Table 3 gives dissociation constants (KD) of binding to the 150 nt single-strand DNA and quality of the fitting to the binding model of wild-type protein and D17D71K S. ruber SSB. Chi.sup.2 values below 1 indicate a very good fit of the experimental data to the model used for KD calculation. KD of the D17KD71K mutant is over 1.8 fold lower, which indicates increased binding affinity of the mutated protein.

[0150] The decrease of KD value confirms that the introduction of D17K D71K mutations significantly increases the initial, already high, affinity of the wild-type SSB protein to the single-strand DNA.

[0151] This experiment was necessarily performed at a low salt concentration but is nonetheless indicative of the superior binding affinity of the mutant protein.

Binding and Dehybridization Fluorescent Assay in a Mixture of Salts

[0152] Wild-type S. ruberSSB, the D17K D71K mutant, and an S. ruber SSB cdel 111 C-terminal deletion mutant lacking the C-terminal 111 residues of the wildtype SSB, were each tested for their ability to bind ssDNA in reaction buffer containing a mixture of NaCl, KCl and Mg(OAc).sub.2. E. coli SSB was used as a well characterized reference protein. Reactions contained 0.125 μM of molecular beacon, 0.25 μM of E. coli SSB tetramer (single binding unit), 4 μM of wt SrSSB, D17KD71K mutant octamer or cdel 111 mutant. Reactions were set up in triplicates in 100 μl of reaction buffer MB (10 mM Tris-HCl pH 7) containing 350 mM NaCl, 600 mM KCl and 150 mM Mg(OAc).sub.2 in a 96-well plate (Corning) and incubated at 25° C. for 5 h. Fluorescence was measured with excitation at 495 nm and emission at 520 nm every 5 min. The baseline fluorescence of the reaction containing molecular beacon was subtracted. Measurements were performed on the Gemini microplate plate reader (Molecular Devices).

[0153] In the course of the experiment, we observed an increase of fluorescence for both wt and mutant S. ruber SSB, indicating binding of the protein to the DNA probe. The D17KD71K mutant had a significantly increased affinity towards ssDNA in the highly concentrated salt mixture. E. coli SSB and cdel 111 had no measurable binding activity under experimental conditions. The results are shown in FIG. 2B.

[0154] Unlike E. coli SSB, which tetramers lose stability in relatively low concentrations of divalent salts, e.g., magnesium chloride, SrSSB is able to maintain its conformation and bind ssDNA in a mixture of highly concentrated mono- and divalent salts (FIG. 2).

Electron Microscopy of SrSSB-ssDNA Complexes

[0155] The SrSSB D17KD71K mutant protein was incubated in the same reaction conditions as under “Binding and dehybridization fluorescent assay in the salts mixture” alone or with ssDNA (M13mp18 Single-stranded DNA, NEB) in 200× excess. Protein concentration was 17 μM (octamer), incubation times varied from 5 to 120 minutes. To prepare the samples for Negative Stain Electron Microscopy, a 3 microliter drop of the analyte was place into a carbon coated grid for 1 min. then blotted away. The grid was washed 5 times with Mili-Q water, blotted off and the stained with 2% uranyl acetate for 1 min. The grid than was blotted and dried in vacuum for observation in a JEOL 1200 EXII in a low dose mode, operated at an acceleration voltage of 100 kV.

[0156] The EM visualizations are shown in FIG. 2A. When the protein alone or in a reaction mixture with M13 phage ssDNA was incubated for 5′, a limited amount of highly ordered, ring-like and sickle-like structures could be observed. When the incubation was prolonged to 120′ a high amount of large, ordered structures could be observed. The observed extensive structures are nucleoprotein complexes of SrSSB-ssDNA, which form extensively when the protein is incubated with ssDNA. It is therefore confirmed that the protein is able to form large, ordered, filament-like nucleoprotein complexes and bind and dehybridize ssDNA a highly concentrated salt solutions.

Oligomeric Structure

[0157] Sequence alignment of Salinibacter ruber (SrSSB) to the other known SSB suggested that the protein forms tetramers like the majority of bacterial SSBs, however during the initial characterization of the SrSSB, it was discovered that the protein exists in an unexpected, higher oligomeric state.

[0158] It has been shown that for E. coli SSB, high salt concentrations prevent SSB function, or, if ssDNA-SSB nucleoprotein complex had already formed, cause hyper-condensation of the ssDNA-protein complex. It is known that in salt concentration equal or lower than 20 mM of NaCl, E. coli SSB binds ssDNA in high-cooperativity mode, forming elongated, filament-like nucleoprotein complex, and in salt concentration between 20 and 200 mM of NaCl E. coli SSB tetramers create a “beads-on-string” structure in which tetramers bind with limited cooperativity.

[0159] To determine the assembly state of SSB from Salinibacter ruber in high and low salt conditions, a recombinantly produced SrSSB was analyzed by size exclusion chromatography and differential scanning calorimetry. In order to investigate the effect of salt concentration on SSB oligomeric state, SSB oligomer molecular mass estimation was made via size exclusion chromatography on a Superdex 200 10/300 GL column (GE healthcare). The analysis was performed in buffer C, containing 25 mM Tris pH 7.5 and 0.15M (buffer C1), 2M (buffer C2), 4M (buffer C3) NaCl. The elution pattern of wildtype S. ruber SSB protein was then compared with those of standard proteins: ferritine (440 kDa), aldolase (158 kDa), concalbumin (75 kDa), ovalbumin (43 kDa).

[0160] The results are shown in FIG. 3. Analysis of purified SrSSB by size exclusion revealed a single peak corresponding to a molecular mass of approximately 225 kDa, which is about 10.3 times the molecular mass of a monomer. This result indicates that the protein forms larger oligomers than previously described SSBs and that the SrSSB tetramers most probably assemble to octamers (dimers of tetramers), dodecamers (trimers of tetramers) and higher oligomeric states as multiples of four. The observed mass of 10.3 times of a monomer, therefore, represents a mixture of approximately 42% octamer and 58% dodecamer in equilibrium. With an increase in salt concentration, the size of the observed oligomer decreased to approximately 7.6 (˜90% octamer, 10% tetramer) and 4.9 (˜25% octamer, 75% tetramer) times the size of the monomer at 2M and 4M of NaCl respectively. Analysis of the oligomeric state was also performed in the presence of 500 mM MgCl.sub.2, where the protein oligomer size was approximately 8 times the size of the monomer (data not shown).

[0161] Thus, the S. ruber SSB displays marked quaternary structure stability in the presence of high salt conditions even in the absence of ssDNA. Such quaternary structure is required for binding to ssDNA. A C-terminal deletion mutant of S. ruber SSB was shown not to bind ssDNA in the mixture of highly concentrated salts (FIG. 2B), and was shown to have an estimated size of 3.8×the size of the monomer (indicative of only tetrameric quaternary structure) at 150 mM NaCl (FIG. 3).

[0162] Surprisingly, unlike the wt SSB, the C-terminal deletion mutant CM 11 was not able to form octamers, forming tetramers instead (data not shown). As discussed above, CM 11 was also not able to bind ssDNA in high salt. Thus in the case of SrSSB, the C-terminal domain is crucial to both filament formation and ssDNA binding activity in high-salt.

Crystallisation and Structure Determination

[0163] Small, prism-shaped crystals of wild-type S. ruberSSB were obtained in sitting drops consisting of 1 μl of protein solution (5 mg/ml) containing 20 μM of d(T).sub.75 oligonucleotide and 1 μl of reservoir solution (27% w/v PEG 3350, 100 mM Tris-HCl pH 8.0, 0.25M MgCl.sub.2, 1 M ammonium sulphate). Soaking experiments were performed by adding d(T).sub.8 oligonucleotide to the drops containing formed crystals, to a final concentration of 0.25 μM.

[0164] FIG. 4 shows the crystal structure SrSSB (PDB ID 5ODN) has an octamer as an asymmetric unit build up from a dimer of tetramers. Each OB-fold has bound one short fragment of ssDNA resulting in eight protein and eight ssDNA fragments in the asymmetric unit. The octamerisation interface is shown in FIG. 4A. FIG. 4B shows that the SSB protein forms filaments of tightly packed tetramers (FIG. 3B), even under the high-salt crystallisation conditions used (0.25M MgCl.sub.2, 1 M ammonium sulphate).