[Journal logo]

Volume 69 
Part 1 
Pages 53-56  
January 2013  

Received 19 September 2012
Accepted 24 November 2012
Online 20 December 2012


[Cited in]
[Download citation]

© International Union of Crystallography 2013

Preliminary crystallographic analysis of a polyadenylate synthase from Megavirus

Audrey Lartigue,a Sandra Jeudy,a Lionel Bertauxa and Chantal Abergela*

aInformation Génomique et Structurale, IGS UMR7256, CNRS, Aix-Marseille Université, IMM, FR3479, 163 Avenue de Luminy - case 934, 13288 Marseille CEDEX 09, France
Correspondence e-mail: chantal.abergel@igs.cnrs-mrs.fr

Megavirus chilensis, a close relative of the Mimivirus giant virus, is also the most complex virus sequenced to date, with a 1.26 Mb double-stranded DNA genome encoding 1120 genes. The two viruses share common regulatory elements such as a peculiar palindrome governing the termination/polyadenylation of viral transcripts. They also share a predicted polyadenylate synthase that presents a higher than average percentage of residue conservation. The Megavirus enzyme Mg561 was overexpressed in Escherichia coli, purified and crystallized. A 2.24 Å resolution MAD data set was recorded from a single crystal on the ID29 beamline at the ESRF.

1. Introduction

Megavirus chilensis was recently isolated from an environmental sample collected on the Pacific Chilean coast near the Tunqúen river outfall (Arslan et al., 2011[Arslan, D., Legendre, M., Seltzer, V., Abergel, C. & Claverie, J.-M. (2011). Proc. Natl Acad. Sci. USA, 108, 17486-17491.]). Its natural host being unknown, we successfully isolated Megavirus by co-cultivation with Acanthamoeba castellanii. Megavirus is an icosahedral virus of roughly 750 nm in diameter that encloses a 1.259 Mb double-stranded DNA genome encoding 1120 genes. It is clearly related to Mimivirus, with which it shares ~600 genes with an average of 50% sequence identity at the protein level (Arslan et al., 2011[Arslan, D., Legendre, M., Seltzer, V., Abergel, C. & Claverie, J.-M. (2011). Proc. Natl Acad. Sci. USA, 108, 17486-17491.]; Legendre et al., 2012[Legendre, M., Arslan, D., Abergel, C. & Claverie, J.-M. (2012). Commun. Integr. Biol. 5, 102-106.]).

As in Mimivirus, the replicative cycle of Megavirus takes place entirely in the host cytoplasm in a virion factory initiated by a single viral particle. Transcription and replication of the viral DNA take place in this transitory organelle (Claverie et al., 2009[Claverie, J.-M., Abergel, C. & Ogata, H. (2009). Curr. Top. Microbiol. Immunol. 328, 89-121.]; Mutsafi et al., 2010[Mutsafi, Y., Zauberman, N., Sabanay, I. & Minsky, A. (2010). Proc. Natl Acad. Sci. USA, 107, 5978-5982.]).

The regulatory elements governing Mimivirus gene expression, such as a highly conserved early and a less stringent late promoter (Suhre et al., 2005[Suhre, K., Audic, S. & Claverie, J.-M. (2005). Proc. Natl Acad. Sci. USA, 102, 14689-14693.]; Legendre et al., 2010[Legendre, M., Audic, S., Poirot, O., Hingamp, P., Seltzer, V., Byrne, D., Lartigue, A., Lescot, M., Bernadac, A., Poulain, J., Abergel, C. & Claverie, J.-M. (2010). Genome Res. 20, 664-674.]), are also conserved in Megavirus. Another example is the `hairpin rule' governing Mimivirus transcript polyadenylation: all Mimivirus mRNAs are polyadenylated at a site coinciding exactly with unrelated, but strongly palindromic, genomic sequences (Byrne et al., 2009[Byrne, D., Grzela, R., Lartigue, A., Audic, S., Chenivesse, S., Encinas, S., Claverie, J.-M. & Abergel, C. (2009). Genome Res. 19, 1233-1242.]; Legendre et al., 2010[Legendre, M., Audic, S., Poirot, O., Hingamp, P., Seltzer, V., Byrne, D., Lartigue, A., Lescot, M., Bernadac, A., Poulain, J., Abergel, C. & Claverie, J.-M. (2010). Genome Res. 20, 664-674.]). We verified using bioinformatic and experimental approaches that this `hairpin rule' also applies to Megavirus (Arslan et al., 2011[Arslan, D., Legendre, M., Seltzer, V., Abergel, C. & Claverie, J.-M. (2011). Proc. Natl Acad. Sci. USA, 108, 17486-17491.]; Legendre et al., 2012[Legendre, M., Arslan, D., Abergel, C. & Claverie, J.-M. (2012). Commun. Integr. Biol. 5, 102-106.]). The two viruses must thus share the machinery responsible for recognition of the gene termini, transcription arrest and transcript polyadenylation.

In both genomes we identified a well conserved gene (62% protein sequence identity) predicted to encode a polyadenylate synthase which could be involved in the above processes. The first viral polyadenylate synthase was characterized in Vaccinia virus (Moure et al., 2006[Moure, C. M., Bowman, B. R., Gershon, P. D. & Quiocho, F. A. (2006). Mol. Cell, 22, 339-349.]). However, the sequence similarity between this enzyme and the Megaviridae homologue is too low (<20% sequence identity) to use the available three-dimensional structure in molecular replacement. Moreover, the `hairpin rule' does not apply in poxviruses, suggesting that this high level of divergence corresponds to significant differences in their polyadenylation processes. We thus decided to conduct an ab initio structural and functional study of the Megavirus polyadenylate synthase in order to obtain insights into the polyadenylation machinery in Megaviridae.

2. Materials and methods

2.1. Expression of the Mg561 gene product

The gene encoding the Megavirus Mg561 protein was amplified from Megavirus chilensis genomic DNA and cloned by restriction/ligation.

For PCR amplification, we used a forward primer (MG561_F; CCCGGATCCTCCGAAAATAGAAATAGAAAATTATC) containing the BamHI site followed by specific nucleotides of the 5' Mg561 gene (bold sequence) and a reverse primer (Mg561_R; TTATGCGGCCGCTTAATTAGTATTAATATCTGTAGTAGTATT) specific for the 3' sequence including the STOP codon of the Mg561 gene (bold sequence) followed by an NotI site.

The PCR product was inserted by restriction/ligation into an in-house-modified pETDuet expression vector (Novagen) allowing the expression of proteins fused with a His6 tag to facilitate purification. A human rhinovirus 3C protease cleavage site allows tag removal to facilitate the recovery of highly purified proteins free from classical Escherichia coli contaminants.

Expression was performed in E. coli C41 (Avidis) transfected cells grown at 310 K in Superior Broth (SB; Athena Enzyme Systems). Induction was performed at 294 K using 0.5 mM IPTG when the A600 reached ~1. The pellet from 1 l overnight culture was resuspended in 40 ml 50 mM Tris-HCl, 300 mM NaCl buffer pH 9 (buffer A) containing 0.1% Triton X-100, 5% glycerol, 0.01%(w/v) DNAse, 0.01%(w/v) lysozyme and three antiprotease tablets (Roche Diagnostics), and the total proteins were extracted by sonication.

Selenomethionine-substituted protein was produced using the appropriate protocol to inhibit methionine synthesis in the presence of selenomethionine and M9 minimal medium (Doublié, 1997[Doublié, S. (1997). Methods Enzymol. 276, 523-530.]).

2.2. Purification

The cleared lysate was applied onto a 5 ml HisTrap FF Column (GE Healthcare) charged with Ni2+ and equilibrated with buffer A on an ÄKTAexplorer 10S FPLC system (GE Healthcare) at a flow rate of 1.5 ml min-1. The column was washed with nine column volumes of buffer A, nine column volumes of buffer A containing 25 mM imidazole and nine column volumes of buffer A containing 50 mM imidazole at a flow rate of 5 ml min-1. Elution was performed with a linear gradient from 50 to 500 mM imidazole over 20 column volumes.

The in-house-produced His-tagged human rhinovirus 3C protease was added to the pooled elution fractions and dialysed overnight at 277 K against 25 mM Tris-HCl buffer pH 9, 150 mM NaCl, 1 mM DTT, 0.1 mM EDTA.

A second run of affinity purification was performed to separate the cleaved Mg561 protein from the uncleaved Mg561 protein and the His-tagged protease, which were retained on the HisTrap FF column. The 534-amino-acid Mg561 recombinant protein corresponds to the native enzyme in which the initial methionine is replaced by Gly-Pro-Gly-Ser resulting from the protease cleavage site. After recovery from the flowthrough, Mg561 was concentrated using a centrifugal filter device (Vivaspin 30K, Vivascience) and desalted on a desalting column (HiPrep 26/10 Desalting, GE Healthcare) in 10 mM Tris-HCl pH 9, 100 mM NaCl.

Gel filtration (performed on a Pharmacia Superdex S200 column) indicated that the Mg561 protein was monomeric in solution.

2.3. Crystallization

The recombinant Mg561 protein was concentrated to 5.9 mg ml-1 in the desalting buffer using a centrifugal filter device (Ultracel 30K, Millipore). The recombinant protein crystallization assays were performed at 293 K against 588 different conditions corresponding to commercially available solution sets (Crystal Screen from Hampton Research and Wizards from Emerald BioStructures) and conditions designed in-house using the SAmBA software (Audic et al., 1997[Audic, S., Lopez, F., Claverie, J.-M., Poirot, O. & Abergel, C. (1997). Proteins, 29, 252-257.]). The screening for crystallization conditions was performed on 3 × 96-well crystallization plates (Greiner) loaded using an eight-needle dispensing robot (a Tecan WS 100/8 workstation modified for our needs), mixing 0.5 µl protein solution with 0.5 µl reservoir solution in a sitting drop (Abergel et al., 2003[Abergel, C. et al. (2003). J. Struct. Funct. Genomics, 4, 141-157.]).

Crystals appeared in 0.1 M Tris-HCl buffer, 50%(v/v) PEG 200 pH 6.8. We refined these conditions at 293 K by hanging-drop vapour diffusion using 24-well culture plates (Greiner). Each hanging drop was prepared by mixing 1 µl 5.9 mg ml-1 Mg561 solution with 1 µl reservoir solution. The hanging drop on the cover glass was vapour-equilibrated against 1 ml reservoir solution in each well of the tissue-culture plate. We used increasing PEG concentrations from 10% to 50% PEG 200. The best crystals appeared after 1 d for the highest PEG concentration and reached typical dimensions of 0.2 × 0.2 × 0.1 mm (Fig. 1[link]a). Data were collected from a crystal grown by mixing 1 µl selenomethionine-substituted Mg561 protein at 2.8 mg ml-1 with 0.5 µl of a reservoir solution consisting of 0.1 M Tris-HCl buffer, 50%(v/v) PEG 200 pH 6.8.

[Figure 1]
Figure 1
Photograph (a) and diffraction image (b) of an Mg561 crystal. In the inset, the edge of the diffraction circle corresponds to 2.24 Å resolution.

2.4. Data collection and processing

This crystal was collected in a Hampton Research 0.2 × 0.2 mm loop, flash-cooled to 105 K in a cold nitrogen-gas stream and subjected to X-ray diffraction; the crystal diffracted to a resolution of 2.24 Å (Fig. 1[link]b). To use the MAD method (Hendrickson et al., 1990[Hendrickson, W. A., Horton, J. R. & LeMaster, D. M. (1990). EMBO J. 9, 1665-1672.]), a three-wavelength data set was collected on the ID29 beamline at the European Synchrotron Radiation Facility (ESRF; Grenoble, France). A fluorescence scan was performed to determine the peak and inflection energies; the exact values were 12.6603 and 12.6584 keV for the peak and inflection, respectively. Data collection was performed with an oscillation angle of 0.5° and a crystal-to-detector distance of 374.27 mm, and 240 images were collected for each wavelength on a PILATUS detector. The crystal belonged to the orthorhombic space group P212121, with unit-cell parameters a = 86.7, b = 96.1, c = 153.9 Å. Superposition of the two monomers in the asymmetric unit shows that they are related by a twofold axis located at polar angles [omega] = 111° and [varphi] = -85°. Weak peaks are detected at these positions in the self-rotation function (Fig. 2[link]). The packing density for two monomers of Mg561 (62743.8 Da) in the asymmetric unit of the crystal is 2.55 Å3 Da-1, indicating an approximate solvent content of 51.85% (Matthews, 1968[Matthews, B. W. (1968). J. Mol. Biol. 33, 491-497.]). Data statistics are presented in Table 1[link].

Table 1
X-ray data-collection statistics

Values in parentheses are for the highest resolution shell.
Beamline ID29, ESRF
Method MAD
Space group P212121
Unit-cell parameters at 100 K (Å) a = 86.7, b = 96.1, c = 153.9
Packing density# (Å3 Da-1) 2.55
Wavelength (Å) 0.9793 0.9795 0.9770
Resolution range (Å) 51.3-2.24 (2.36-2.24) 51.3-2.24 (2.36-2.24) 51.3-2.24 (2.36-2.24)
Observations 210289 (32884) 219559 (34075) 209823 (32763)
Unique reflections 53432 (7970) 54650 (8145) 53198 (7983)
Multiplicity 3.9 (4.1) 4.0 (4.2) 3.9 (4.1)
Completeness (%) 99.0 (99.0) 99.2 (99.2) 99.0 (99.0)
<I/[sigma](I)>+ 6.4 (1.1) 5.0 (0.9) 6.7 (1.5)
Rmerge§ (%) 8.1 (62) 8.8 (69) 6.9 (46)
#For two molecules in the asymmetric unit.
+<I/[sigma](I)> is the mean signal-to-noise ratio, where I is the integrated intensity of a measured reflection and [sigma](I) is the estimated error in the measurement.
§Rmerge = [\textstyle \sum_{hkl}\sum_{i}|I_{i}(hkl)- \langle I(hkl)\rangle|/][\textstyle \sum_{hkl}\sum_{i}I_{i}(hkl)], where Ii(hkl) is the integrated intensity of reflection hkl having i observations and <I(hkl)> is the mean recorded intensity of reflection hkl over multiple recordings.
[Figure 2]
Figure 2
Self-rotation Patterson plot. The self-rotation function was calculated using the Self Rotation Function Server (http://services.mbi.ucla.edu/selfrot/ ) with default parameters. These sections have [omega] = 0 or 180° at the centre, [omega] = 90° around the edge and [varphi] as marked around the periphery.

MOSFLM (Leslie, 1992[Leslie, A. G. W. (1992). Jnt CCP4/ESF-EACBM Newsl. Protein Crystallogr. 26.]) and SCALA (Evans, 1997[Evans, P. R. (1997). Proceedings of the CCP4 Study Weekend. Recent Advances in Phasing, edited by K. S. Wilson, G. Davies, A. W. Ashton & S. Bailey, pp. 97-102. Warrington: Daresbury Laboratory.], 2006[Evans, P. (2006). Acta Cryst. D62, 72-82.]) from the CCP4 package (Winn et al., 2011[Winn, M. D. et al. (2011). Acta Cryst. D67, 235-242.]) were used for processing, scaling and reduction of the data sets.

A self-rotation function analysis of the data in the resolution range 9-4 Å with an integration radius of 25 Å was carried out using the Self Rotation Function Server (http://services.mbi.ucla.edu/selfrot/ ; Winn et al., 2011[Winn, M. D. et al. (2011). Acta Cryst. D67, 235-242.]).

3. Results and discussion

Phase determination was performed with autoSHARP (de La Fortelle & Bricogne, 1997[La Fortelle, E. de & Bricogne, G. (1997). Methods Enzymol. 276, 472-494.]; Bricogne et al., 2003[Bricogne, G., Vonrhein, C., Flensburg, C., Schiltz, M. & Paciorek, W. (2003). Acta Cryst. D59, 2023-2030.]). Phases were calculated at three wavelengths in the resolution range 51.3-2.24 Å, and a single solution was found with 34 Se atoms (for 36 methionines). The mean figure of merit for this solution is 0.39. The electron-density map is currently being used to construct the main chain of the Mg561 protein structure (Fig. 3[link]).

[Figure 3]
Figure 3
Projection of the solvent-flattened electron-density map produced by autoSHARP.

The transcriptome study of A. castellanii cells infected by Mimivirus through the entire viral cycle revealed that Mimivirus genes start to be transcribed immediately after infection (Legendre et al., 2010[Legendre, M., Audic, S., Poirot, O., Hingamp, P., Seltzer, V., Byrne, D., Lartigue, A., Lescot, M., Bernadac, A., Poulain, J., Abergel, C. & Claverie, J.-M. (2010). Genome Res. 20, 664-674.]). The expression of the Mimivirus and Megavirus polyadenylate synthase genes is controlled by the highly conserved early promoter and the Mimivirus enzyme exhibits continuous expression from the beginning to the end of the infection cycle. Interestingly, the Mimivirus polyadenylate synthase protein is also loaded into the virion capsids (Renesto et al., 2006[Renesto, P., Abergel, C., Decloquement, P., Moinier, D., Azza, S., Ogata, H., Fourquet, P., Gorvel, J.-P. & Claverie, J.-M. (2006). J. Virol. 80, 11678-11685.]; Claverie et al., 2009[Claverie, J.-M., Abergel, C. & Ogata, H. (2009). Curr. Top. Microbiol. Immunol. 328, 89-121.]), suggesting that this function is required for the processing of the very first viral transcripts, including its own. The three-dimensional structure of Mg561 will be compared with its cellular homologues (bacterial, archaeal and eukaryal polyadenylate synthases) as well as with the structure of the Vaccinia virus enzyme. This comparison should provide some clues to the origin of the Megaviridae enzymes. Molecular studies will also be performed to address the affinity of the enzyme for folded nucleic acids (RNA hairpins and DNA hairpins) and its ability to polyadenylate transcripts in vitro.

Acknowledgements

We thank the BM30, ID14, ID23, ID29 (ESRF) and PROXIMA (SOLEIL) beamline teams for expert assistance. We also thank Dr Alain Roussel for collecting the MAD data set and Professor Nobert Sträter for helpful suggestions. This work was partially funded by CNRS, ANR grant No. ANR-08-BLAN-0089, IBISA and the Provence-Alpes-Côte d'Azur region.

References

Abergel, C. et al. (2003). J. Struct. Funct. Genomics, 4, 141-157.  [CrossRef] [PubMed] [ChemPort]
Arslan, D., Legendre, M., Seltzer, V., Abergel, C. & Claverie, J.-M. (2011). Proc. Natl Acad. Sci. USA, 108, 17486-17491.  [CrossRef] [ChemPort] [PubMed]
Audic, S., Lopez, F., Claverie, J.-M., Poirot, O. & Abergel, C. (1997). Proteins, 29, 252-257.  [CrossRef] [ChemPort] [PubMed]
Bricogne, G., Vonrhein, C., Flensburg, C., Schiltz, M. & Paciorek, W. (2003). Acta Cryst. D59, 2023-2030.  [ISI] [CrossRef] [ChemPort] [details]
Byrne, D., Grzela, R., Lartigue, A., Audic, S., Chenivesse, S., Encinas, S., Claverie, J.-M. & Abergel, C. (2009). Genome Res. 19, 1233-1242.  [ISI] [CrossRef] [PubMed] [ChemPort]
Claverie, J.-M., Abergel, C. & Ogata, H. (2009). Curr. Top. Microbiol. Immunol. 328, 89-121.  [CrossRef] [PubMed] [ChemPort]
Doublié, S. (1997). Methods Enzymol. 276, 523-530.  [CrossRef] [ChemPort] [PubMed] [ISI]
Evans, P. R. (1997). Proceedings of the CCP4 Study Weekend. Recent Advances in Phasing, edited by K. S. Wilson, G. Davies, A. W. Ashton & S. Bailey, pp. 97-102. Warrington: Daresbury Laboratory.
Evans, P. (2006). Acta Cryst. D62, 72-82.  [ISI] [CrossRef] [details]
Hendrickson, W. A., Horton, J. R. & LeMaster, D. M. (1990). EMBO J. 9, 1665-1672.  [ChemPort] [PubMed]
La Fortelle, E. de & Bricogne, G. (1997). Methods Enzymol. 276, 472-494.
Legendre, M., Arslan, D., Abergel, C. & Claverie, J.-M. (2012). Commun. Integr. Biol. 5, 102-106.  [CrossRef] [ChemPort] [PubMed]
Legendre, M., Audic, S., Poirot, O., Hingamp, P., Seltzer, V., Byrne, D., Lartigue, A., Lescot, M., Bernadac, A., Poulain, J., Abergel, C. & Claverie, J.-M. (2010). Genome Res. 20, 664-674.  [ISI] [CrossRef] [ChemPort] [PubMed]
Leslie, A. G. W. (1992). Jnt CCP4/ESF-EACBM Newsl. Protein Crystallogr. 26.
Matthews, B. W. (1968). J. Mol. Biol. 33, 491-497.  [CrossRef] [ChemPort] [PubMed] [ISI]
Moure, C. M., Bowman, B. R., Gershon, P. D. & Quiocho, F. A. (2006). Mol. Cell, 22, 339-349.  [ISI] [CrossRef] [PubMed] [ChemPort]
Mutsafi, Y., Zauberman, N., Sabanay, I. & Minsky, A. (2010). Proc. Natl Acad. Sci. USA, 107, 5978-5982.  [CrossRef] [ChemPort] [PubMed]
Renesto, P., Abergel, C., Decloquement, P., Moinier, D., Azza, S., Ogata, H., Fourquet, P., Gorvel, J.-P. & Claverie, J.-M. (2006). J. Virol. 80, 11678-11685.  [CrossRef] [PubMed] [ChemPort]
Suhre, K., Audic, S. & Claverie, J.-M. (2005). Proc. Natl Acad. Sci. USA, 102, 14689-14693.  [CrossRef] [PubMed] [ChemPort]
Winn, M. D. et al. (2011). Acta Cryst. D67, 235-242.  [ISI] [CrossRef] [details]

Acta Cryst (2013). F69, 53-56   [ doi:10.1107/S1744309112048257 ]