High affinity anchoring of the decoration protein pb10 onto the bacteriophage T5 capsid

Bacteriophage capsids constitute icosahedral shells of exceptional stability that protect the viral genome. Many capsids display on their surface decoration proteins whose structure and function remain largely unknown. The decoration protein pb10 of phage T5 binds at the centre of the 120 hexamers formed by the major capsid protein. Here we determined the 3D structure of pb10 and investigated its capsid-binding properties using NMR, SAXS, cryoEM and SPR. Pb10 consists of an α-helical capsid-binding domain and an Ig-like domain exposed to the solvent. It binds to the T5 capsid with a remarkably high affinity and its binding kinetics is characterized by a very slow dissociation rate. We propose that the conformational exchange events observed in the capsid-binding domain enable rearrangements upon binding that contribute to the quasi-irreversibility of the pb10-capsid interaction. Moreover we show that pb10 binding is a highly cooperative process, which favours immediate rebinding of newly dissociated pb10 to the 120 hexamers of the capsid protein. In extreme conditions, pb10 protects the phage from releasing its genome. We conclude that pb10 may function to reinforce the capsid thus favouring phage survival in harsh environments.


Supplementary
: Structure conservation of the isolated domains compared to full-length pb10. (a) Chemical shift differences between full-length pb10 and its isolated domains. The threshold was calculated based on the resolution of the NMR spectra. (b) Structures of the isolated NTD and CTD with residues coloured according to the chemical shift differences between fulllength pb10 and its isolated domains: absence of data in grey, no significant difference (lower than the threshold) in green, a significant difference (higher than the threshold) in orange. Figure S4: Comparison between the experimental (red line) and calculated SAXS curves. The calculated curves correspond to the 15 best structures of pb10 determined using a subset of the NMR NOE and dihedral angle restraints obtained for the isolated NTD and CTD together with SAXS data obtained on the full length pb10. In the upper panel, the residual is calculated according to R(q)= (I calc( q) -I obs (q))/I err (q) where I calc , I obs and I err correspond to the calculated SAXS intensity, the experimental intensity and the associated error, respectively. Expanded capsids (32 µg/ml), incubated with limited amounts of pb10-His6 (1 and 10 nM, respectively 6.8x10 -3 and 68x10 -3 pb10/hexamer), were captured on the surface of a nickel activated NTA (Ni 2+ -NTA) sensorchip. They were then saturated with untagged pb10 (100 nM). Many available pb10 binding sites are observed in the case where capsids were incubated with pb10-His6 1nM (response of 35 RU for 750 RU of capsid), while very few pb10 sites remain available when capsids were incubated with pb10-His6 10nM (4 RU for 750 RU of captured capsid). Figure S7: (a) Sequence alignment of the major head proteins from 17 T5-like (closely-related and more distant) phages. The A-loop region is highlighted in the black box. The alignment figure was generated using ESPript 3.0 2 . (b) Structural model of the major head protein pb8 of phage T5 coloured from blue (N-terminus) to red (C-terminus). The Aloop is located within the black oval. (c) CD spectra of wild-type pb10 in black and pb10 mutants in other colours as indicated. The spectra are presented on 3 separated panels for clarity. (d) Secondary structure contents of wild-type pb10 and pb10 mutants as predicted by the SELCON3, CONTIN and CDSSTR algorithms available on the DichroWeb server 6 and the BeStSel algorithm 7 .

Ramachandran plot statistics (%) b
Residue in most favoured regions 89 79.7 Residue in additional allowed regions 10.9 15.6 Residue in generously allowed regions 0.1 1.9 Residue in disallowed regions 0 2.8 Rms diff. to the mean structure (Å) c 0.43 ± 0.08 0.44 ± 0.08 a Electrostatic energy was calculated with CHARMM 8 using Charmm22 parameters and a distant dependent dielectric constant. b The percentage corresponds to residues 4-68 and 79-160 for N77 and C72 respectively. They were determined by PROCHECK 9,10 . c The rms values correspond to residues 4-68 and 79-160 for N77 and C72 respectively. They were calculated on the backbone atoms C, N and Cα. 2.0 a After relaxation in the Charmm22 force field

Supplementary Methods
Protein production and purification. E coli BL21(DE3) cells harbouring each of the different pb10 expression vectors were grown at 37°C in 2-YT medium supplemented with 50 µg/mL kanamycin. For NMR studies, 15 N or 15 N/ 13 C-labeled proteins were produced by growing E coli BL21(DE3) cells in M9 minimum medium containing ( 15 NH 4 ) 2 SO 4 as the only nitrogen source and/or 13 C-glucose as the only carbon source (Cambridge Isotope Laboratories, Inc.). At mid-exponential growth phase (OD 600 =0.6-0.8), protein expression was induced by addition of 0.4 mM isopropyl-β-D-thiogalactopyranoside (IPTG) and the growth continued for 3h at 37 °C. Bacterial cells harvested by centrifugation were suspended in 20 mL of lysis buffer (50mM Tris-HCl pH 7.2, 150 mM NaCl, 1 mM EDTA, 0.6% Nonidet P-40, 0.1 g/L lysozyme) and incubated at 37°C for 10 min. DNAse (Universal Nuclease, Pierce) and 1 mM MgCl 2 were then added for a 10 min additional incubation and the suspension was centrifuged at 20,000g for 30min at 4°C. The supernatant was loaded onto a 5 mL HisTrap TM FF column (GE Healthcare) pre-equilibrated in either 25 mM Tris-HCl pH 8.0 or 50 mM Hepes pH 7.0 containing 150 mM NaCl that was connected to an ÄKTA purifying system. The column was extensively washed with the loading buffer supplemented with 1M NaCl and pb10 or derivative proteins were eluted with a 0-1 M imidazole gradient. The eluted fractions were collected and proteins further purified by cation or anion exchange chromatography, depending on their calculated isoelectric point. Fulllength pb10 (pI=7.89) and pb10-NTD (pI=9.75) were loaded onto a 5 mL HiTrap SP column (GE Healthcare) preequilibrated in 50 mM Hepes buffer pH 7.0. Pb10-CTD (pI=5.76) mL was loaded onto a 5 mL HiTrap Q HP column (GE Healthcare) pre-equilibrated in 25 mM Tris-HCl pH 8.0. The proteins were eluted with a 0-1 M NaCl gradient. Purified proteins were dialyzed against 25 mM sodium phosphate pH 7.2 and concentrated if needed on a centrifugal filter (Amicon Ultra-4, 10 kD, Millipore) to a final concentration of at least 200 µM. Untagged pb10 was produced from a pET28 expression vector including a sequence coding the TEV protease recognition site, ENLYFQS, upstream of the C-terminal His-tag of pb10. After a first purification step on an HisTrap TM FF column as described above, the pb10-TEV-His protein was dialyzed in 25 mM Tris Buffer pH 8.0 containing 0.5 mM EDTA, 1mM DTT and 150 mM NaCl. The His-tag was then cleaved by incubating the protein with purified His-tagged TEV protease (prepared in our laboratory) at 4°C overnight with a ratio of 1 OD 280nm of TEV protease for 10 OD 280nm pb10-TEV-His protein. Purification of cleaved pb10 and removal of His-TEV protease was achieved in one step by loading the reaction mixture onto a 5 mLHisTrap TM FF column and harvesting of untagged pb10 in the flow-through.

SEC-MALLS (Size-exclusion-chromatography-coupled with multi-angle laser light scattering).
Chromatographic analysis of pb10 was carried out by SEC-MALLS using a GPCMax-TDA system (Viscotek, Malvern, France). A pb10 sample (200 µL at ≈ 4 mg/mL) was injected at a flow rate of 0.5 ml/min onto a Superdex 75 HR10/30 column (GE Healthcare) equilibrated with 25 mM sodium phosphate pH 7.2 containing 150 mM NaCl. Elution was monitored on-line by a UV-visible spectrophotometer, a differential refractometer, a 7° low angle light scattering detector, a 90° right angle light scattering detector and a differential pressure viscometer. The instrument was calibrated using bovine serum (Sigma-Aldrich). The OmniSEC program (Malvern) was used for the acquisition and analysis of the data.
Details of SAXS data processing. Purified pb10 protein was injected in a Superdex 75 column (GE Healthcare) that was pre-equilibrated with preparative buffer comprising 25 mM sodium phosphate buffer pH 7.0. Flow rate was 0.2ml/min, frame duration was 1.0 s and the dead time between frames was 0.5 s. For each frame, the protein concentration (about 3.5 mg/ml at the top of elution peak) was estimated from UV absorption at 280 and 295 nm using a spectrometer located immediately upstream of the SAXS measuring cell. Selected identical frames corresponding to the elution peak were averaged. A large number of frames were collected before the void volume and averaged to account for buffer scattering. SAXS data were normalized to the intensity of the incident beam and background (i.e. the elution buffer) subtracted using the programs FoxTrot (courtesy of SWING beamline) and Primus 11 . The scattered intensities were displayed on an absolute scale using the scattering by water. The conformation of pb10 in solution was determined ab initio from the scattering curve using the program GASBOR that generates a volume filled with a compact chain of dummy residues. Models obtained from a hundred calculations are superimposed and compared using the program suite DAMAVER 12 . Their similarity is quantified using a normalized spatial discrepancy (NSD). The value found for NSD (0.96) indicates a high similarity between them.
Pb10 structure and dynamics analysis. The 3D structure of the full-length pb10 was determined using a simulated annealing protocol with the XPLOR-NIH 2.36 software1.2. All calculations were achieved on a cluster equipped with bi-pro Intel(R) XEON Nehalem X86 processors (I2BC, GIPSI, CEA). NOE-derived inter-proton distances, chemical shift-derived dihedral angles and small angle X-ray diffusion (SAXS) data were used as restraints during the simulated annealing protocol. A "synthetic" NOE distance set was built from those obtained for structure determination of the isolated N and C-terminal domains. The NOE-derived distances composing this "synthetic" set were the NOE distance restraints obtained for the isolated domains with the exception of those involving residues with altered 15 N or 1 Hn chemical shifts in pb10 as compared to the isolated domains. This criterion was also applied to select a subset of φ, ψ dihedral restraints determined for the isolated domains using the TALOS software3. This led to a set of 2,490 NOE distance restraints and 214 dihedral angle restraints. The structure determination was achieved in two steps. A high-temperature simulated annealing stage was run to obtain a good starting structure that was further used in the so-called refinement step to generate a large number of structures compatible with the NMR and SAXS data. The annealing step started with randomization of the torsion angles of an initial extended conformation of the protein. This structure was submitted to 1,000 steps of high-temperature (3,500 K) molecular dynamics. Temperature was controlled using the coupling to an external bath. During the high temperature molecular dynamics steps, radii were significantly reduced for all atoms except the CA and the experimental NOE distance restraints were taken into account via a soft potential and a scale factor of 2. The chemical shift derived-dihedral angle restraints were taken into account via a harmonic term with a scale factor of 10. Then the system was cooled to 25 K in steps of 12.5 K. At each temperature, 100 molecular dynamics steps were calculated. During cooling, the scale factor of the NOE restraints was gradually increased from 2 to 30 and the scale factor of the torsion angle was set to 200. The van der Waals radii of all atoms were increased during cooling via the rcon parameter that was decreased from its high temperature value 0.004 to 4 at the end of the cooling step. The atom-atom repulsive strength was also gradually decreased through the repel parameter from 0.9 to 0.8. During annealing, 150 structures were generated and the 30 best structures were kept for refinement. Refinement of the selected pb10 structures was achieved using a simulated annealing protocol essentially similar to the one used for the annealing step excepted that the starting structure was not a random one but the starting coordinates were those of the structures obtained via the annealing protocol. In both annealing and refinement protocols, SAXS data were taken into account via a specific potential term. This term allowed minimizing the root mean square difference between the experimental intensities of the experimental SAXS spectrum and the intensities calculated from the Cartesian coordinates of a 3D structure of the protein. During all calculations, the X-ray scale factor was set to 400. 50 points were used for the calculation of each spectrum. The calculated intensities were corrected for higher accuracy by fitting of the experimental spectrum every 10th temperature. This included the evaluation of the solvent contribution. Circular dichroism. The protein samples at 0.5 mg/mL were analysed on a Jasco J-180 spectropolarimeter at 20°C in 20 mM phosphate buffer at pH 7.0. Scans were performed between 185 and 260 nm with intervals of 1 nm at 50 nm/min. 3 to 8 accumulations were recorded for each protein. After buffer subtraction and conversion to international units (Δε) the spectra were smoothed to reduce noise.

Construction of T5∆dec and T5D18am-∆dec mutants.
A deleted version of the dec gene encoding pb10 was constructed by PCR-based oligonucleotide-directed mutagenesis. The deletion encompassed 444 bp corresponding to codons 2-149 of the dec gene and the final PCR product included 484 bp and 558 bp flanking the 5'-and 3'-sides of the dec gene respectively. This PCR fragment was cloned into pUC19 EcoRI/SalI.and E. coli cells transformed with the resulting pUC19/T5∆dec plasmid were infected with T5wt. The phage progeny were titrated to individual plaques, which were transferred to Hybond-N membranes (Amersham) and hybridized with a 32 P-labeled oligonucleotide covering the opposite sides of the deleted region (5'-GTAGCTCCTGAAGCATCATTCTTAATCTCCCT-3'). Deleted mutants were screened by hybridization overnight at room temperature in 2xSSC, 0.1% SDS, 0.1 mg/mL calf thymus DNA, followed by incubation for 30 min in 2xSSC, 0.1% SDS at 65°C and autography with PhosphorImager (Cyclone, Packard) and OptiQuant software. Positive plaques were selected as desired T5∆dec mutant phages and checked by sequencing. In order to produce non-decorated T5 heads, we constructed the double mutant T5D18am-∆dec by cross-infection of the E. coli CR63 suppressive strain with T5∆dec and T5D18amH5 mutant defective in tail assembly 13 . Phage progeny were plated on suppressive CR63 and non-suppressive F strains. Individual plaques exhibiting the amber phenotype were screened for the presence or absence of pb10 production by dot blots using anti-pb10 antibodies.
Capsid concentration determination. The empty capsids molar concentrations C mol were determined based on the OD 280 and OD 260 as follows: where F = 1.47 (correction factor based on the amino acids quantitative analysis), R th = 1.58 (theoretical OD 260 /OD 280 ratio based on the absorbance of the major head protein aromatic amino acids), R exp is the measured OD 260 /OD 280 ratio, R = 280 4 /260 4 , MW C = 26 MDa (capsid molecular weight), E = ε pb8 /MW pb8 = 0.878 (ε pb8 is the major head protein extinction coefficient and MW pb8 is the major head protein molecular weight).