Human importin alpha and RNA do not compete for binding to influenza A virus nucleoprotein

Virology 409 (2011) 84–90

Contents lists available at ScienceDirect

Virology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y v i r o

Human importin alpha and RNA do not compete for binding to influenza A virus nucleoprotein Sébastien Boulo a,b, Hatice Akarsu b, Vincent Lotteau c,d,e, Christoph W. Müller a, Rob W.H. Ruigrok b, Florence Baudin a,b,⁎ a

European Molecular Biology Laboratory (EMBL), Meyerhofstrasse 1, 69117 Heidelberg, Germany UJF-EMBL-CNRS UMI 3265, Unit of Virus Host-Cell Interactions, 6 rue Jules Horowitz, 38042 Grenoble Cedex 9, France Inserm, U851, Lyon, France d Université de Lyon, IFR128 BioSciences Lyon-Gerland, Lyon, France e Hospices civils de Lyon, Hôpital de la Croix Rousse, Laboratoire de virologie, Lyon, France b c

a r t i c l e

i n f o

Article history: Received 5 July 2010 Returned to author for revision 23 August 2010 Accepted 1 October 2010 Available online 25 October 2010 Keywords: Influenza virus nucleoprotein Importin α NLS Viral RNA

a b s t r a c t Influenza virus has a segmented genome composed of eight negative stranded RNA segments. Each segment is covered with NP forming ribonucleoproteins (vRNPs) and carries a copy of the heterotrimeric polymerase complex. As a rare phenomenon among the RNA viruses, the viral replication occurs in the nucleus and therefore implies interactions between host and viral factors, such as between importin alpha and nucleoprotein. In the present study we report that through binding with the human nuclear receptor importin α5 (Impα5), the viral NP is no longer oligomeric but maintained as a monomer inside the complex. In this regard, Impα5 acts as a chaperone until NP is delivered in the nucleus for viral RNA encapsidation. Moreover, we show that the association of NP with the host transporter does not impair the binding of NP to RNA. The complex human Impα5-NP binds RNA with the same affinity as wt NP alone, whereas engineered monomeric NP through point mutations binds RNA with a strongly reduced affinity. © 2010 Elsevier Inc. All rights reserved.

Introduction Influenza A viruses are important viral pathogens of humans and animals. Despite the seasonal character of the disease, human pandemics also occur. Last year's swine origin H1N1 pandemic has clearly demonstrated that outbreaks can occur suddenly and unexpectedly despite constant worldwide surveillance. The current highly virulent H5N1 avian strain circulating in Asia may as well adapt to man and become a serious threat for the worldwide population. For these reasons, we should maintain our efforts to understand why and how influenza virus can adapt to humans and more precisely which host cell proteins are hijacked by the virus for its own purpose. Influenza A viruses are enveloped viruses of the orthomyxovirus family whose genome comprises 8 negative strand RNA segments. Together with a copy of the heterotrimeric polymerase, the viral nucleoprotein (NP) forms ribonucleoprotein (RNP) complexes with the viral RNA with a stoichiometry of one NP protomer for 24 nucleotides (Compans et al., 1972; Ortega et al., 2000). During infection, the incoming RNPs are released in the cytoplasm and are then actively imported into the nucleus where transcription and replication take place (Whittaker et al., 1996). Viral replication is not ⁎ Corresponding author. Mailing address: EMBL, Meyerhofstrasse 1, D-69117 Heidelberg, Germany. Fax: +49 6221 387 8518. E-mail address: [email protected] (F. Baudin). 0042-6822/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2010.10.001

possible without this nucleoprotein, i.e. the nucleoprotein–vRNA complex is the template for the viral polymerase (Bishop et al., 1971). The influenza A virus nucleoprotein is 498 amino acids long (56 kDa) and is positively charged. As expected from its role in the formation of the RNPs, NP has been shown to possess a high affinity for ssRNA with no sequence specificity (Baudin et al., 1994; Digard et al., 1999; Kingsbury et al., 1987; Scholtissek and Becht, 1971; Yamanaka et al., 1990). In viral RNP or in a reconstituted NP-RNA complex, RNA binds to NP through its phosphate-sugar backbone (Baudin et al., 1994). The interaction between NP and RNA has been shown to be cooperative (Yamanaka et al., 1990). However, the delivery process of NP on the viral RNA template remains unknown. The atomic structure of recombinant NP free of RNA was recently determined (Ye et al., 2006). The structure of the nucleoprotein shows three domains (Fig. 1), a helical Body domain (residues 21 to 149; residues 273 to 396 and residues 453 to 489), a helical Head domain (residues 150 to 272 and residues 438 to 452), and a C-terminal Tail domain (residues 402 to 428). The Tail domain is involved in domain exchange and protein oligomerization. The putative RNA-binding site is located between the Head and Body domains. A large number of basic residues present in this region could constitute the RNA binding platform (Ng et al., 2008). The RNA would be largely exposed on the external surface of nucleoprotein molecules in the RNP; consistent with the fact that influenza virus RNA covered with nucleoprotein is still sensitive to RNases (Baudin et al, 1994; Duesberg, 1969).

S. Boulo et al. / Virology 409 (2011) 84–90

85

Results Recombinant wt and monomeric NP

Fig. 1. Structure of influenza A virus nucleoprotein and location of R416 and E339 (PDB accession number 2IQH). The Head domain is coloured in blue, the Body domain is coloured in green and the Tail domain is coloured in orange. The two residues R416 and E339 are indicated in black.

The viral polymerase and nucleoprotein play active roles in host adaptation and pathogenesis when an avian virus changes host to a mammalian species (Chen et al., 2006; Gabriel et al., 2005, 2008; Naffakh et al., 2008; Nagata et al., 2008). Indeed, influenza transcription, translation, and protein trafficking are dependent on the host machinery. A multitude of cellular factors interacting with influenza virus polymerase and nucleoprotein are being identified (Watanabe et al., 2010). Subsequent to viral protein translation in the cytoplasm, NP and polymerase need to be actively transported into the nucleus. This transport is achieved through importin α binding, which recognizes the NLS motif present at the surface of the protein to be imported. Then this complex binds in turn to importin β that directs the complex through the nucleopore (reviewed by Cook et al., 2007). Different studies have already shown that isolated influenza nucleoprotein interacts in vivo and in vitro with several human importins α such as α1, α3 and α5 (O'Neill et al., 1995; Wang et al., 1997; Melen et al., 2003). NP was shown to possess a non-conventional NLS present in its N-terminal region (amino acids 1-13) (Wang et al., 1997; Neumann et al., 1997) and/or a classical bipartite NLS between amino acids 198 and 216 (Weber et al., 1998), although some evidence favours the use of the N-terminal signal only (Cros et al., 2005). Moreover, the crystal structure of NP (Ye et al., 2006; Ng et al., 2008) shows a distance of 15 Å between the two basic clusters 198-KR-199 and 213-RKTR-216. This length is clearly shorter than the minimal required distance of 28 Å necessary for the efficient binding of importin alpha (Fontes et al., 2000). Unfortunately, the first twenty amino acids of NP are not visible in the crystal structure of NP. In the present work, we studied the interactions of influenza virus NP with human importin alpha 5 (Impα5) as well as the binding of RNA to NP and to NP inside the NP–Impα5 complex. We show through biochemical assays that Impα5 maintains NP as a monomer and that the RNA binding activity inside the NP–Impα5 complex is preserved. Mutated monomeric NP binds with a 20 times lower affinity to RNA than wt recombinant NP. However, NP associated with importin alpha binds to viral RNA with the same affinity as wt NP, suggesting that monomerisation is not responsible for the decreased affinity. The mutations engineered on NP (R416A and E339A) are responsible for the observed low affinity to RNA which could be the consequence of a conformational change of NP leading to a change in the RNA binding platform.

Recombinant influenza virus NP (recNP) with an N-terminal histidine-tag (His-Tag) was expressed in Escherichia coli and purified according to a previously published protocol (Baudin et al., 1994). Monomeric influenza virus NP (monoNP) was made by mutating R416 or E339 to alanines. The crystal structure of NP (Ye et al., 2006) shows a Tail domain (Fig. 1) that is involved in the oligomerization of the NP protomers. The interaction between NP protomers is stabilized by a salt-bridge between arginine R416 of one protomer and glutamate E339 from a second NP protomer. The mutation of R416 and E339 to alanine destroys the salt-bridge and leads to monomeric NP. The importance of amino acid R416 in the oligomerization process was first described by Elton et al. (1999a). The polymerisation state of these two nucleoprotein preparations was first visualised using negative stain electron microscopy (Fig. 2). The NP samples were first incubated at 37 °C for 10 min, then stained with a freshly prepared 2% uranyl-acetate solution. Images were taken with a Morgagni FEI electron microscope at 100 kV with a magnification of 71,000 times. RecNP forms small oligomers with many monomers, dimers, trimers and tetramers in the background as pointed out in white circles (Fig. 2 recNP). As expected, the R416A and the E339 mutant NP only produced monomers (Fig. 2 monoNP, only the result for the R416 mutant is shown). Binding of influenza NP to the host factor importin α5 (Impα5) Importin α5 was produced and purified as described before (Tarendeau et al., 2007, see Materials and methods). Since both wild type recNP and recombinant Impα5 were individually expressed and purified through an N-terminal His-tag as a first step of purification, we removed one of the tags in order to purify the complex. Since NP polymerises easily, we used a three times excess of Impα5 over recNP (w/w) in order to shift the NP oligomerization towards NP–Impα5 hetero-dimer formation. Therefore, the His-tag was kept on NP in order to get rid of the excess of Impα5 protein. The two purified proteins were mixed and purified over a Co2+ charged affinity column. After complex elution, an exclusion size chromatography was performed (Fig. 3a). The first peak elutes at 9.1 ml and corresponds to the void volume of the column. The second peak elutes at 11.2 ml and most likely contains small aggregates such as dimers or trimers of NP as well as dimers of importin α5 as seen on an SDS gel (Fig. 3b,

Fig. 2. Electron micrographs of negatively stained influenza virus nucleoprotein recNP; recombinant wild type NP produced in E. coli. monoNP; recombinant NP mutant R416A. Negative staining was done with 2% uranyl acetate.

86

S. Boulo et al. / Virology 409 (2011) 84–90

size of roughly 100 kDa, in agreement with the expected molecular weight of the complex; i.e. 58 kDa for the recombinant NP (including His-tag) plus 50 kDa for importin α5. We verified the size of the complex in fraction 8 by analytical ultra centrifugation (Fig. 3c) and found that it sedimented with a size of 106 kDa. Therefore, the two proteins are present in equimolar amounts in the complex. This result suggests that recNP is no longer oligomeric upon association with Impα5, implying that importin α5 binding impairs the binding of free recNP to the NP–Impα5 dimer. We also characterized the interaction between monomeric R416A NP mutant and importin α in vitro. The monoNP alone was first injected on a size exclusion chromatography (gel filtration S200, Pharmacia) and the curve shows a single peak eluting at 214 ml (Fig. 4a and b, black curve). Importin α5 alone (green curve) was then injected on the same column and shows a peak eluting at 200 ml (Fig. 4a and b). Injection of a mixture of monoNP and impα5 (red curve) shows 3 peaks on the size exclusion chromatography (Fig. 4a). The first one corresponds to the void volume of the column (110 ml). The second peak elutes at 174 ml and the third peak elutes at 214 ml. Aliquots from fractions of the elution curve were applied to an SDS gel and shows the presence of monoNP and Impα5 in peak 2 and the presence of monoNP in excess in peak 3 (Fig. 4a and b). Here again, we estimate that the two bands from peak 2 on the SDS gel in Fig. 4b present a similar intensity. Therefore, the two proteins appear to be present in equimolar amounts in the complex. We again performed analytical ultracentrifugation experiments to determine precisely the molecular weight of the complex. Fig. 4c shows the sedimentation profile of R416A NP alone with a molecular weight of 56 kDa, confirming the electron microscopy result. Importin α5 sediments with a molecular weight of 50 kDa and the ultracentrifugation experiment performed on the R416A NP–Impα5 complex reveals the same molecular weight as that for the wt recNP–Impα5 complex. We then determined the binding affinity of influenza NP to importin α5 using an ITC experiment (isothermal calorimetry) as shown in Fig. 5. Such experiment cannot be performed with wt recNP because the heat effects of NP-NP interactions interfere with those of the NP–Impα5 interaction. MonoNP binds to importin α5 with an affinity of 26 nM. This value is very similar to the affinities measured for other nuclear import substrates for their receptors (Yang et al., 2010). RNA binding to NP and NP–Impα5 complex

Fig. 3. Interaction of recNP with importin α5. (a) Elution profile from a Superdex 200 column (GE Healthcare) of the recNP–Impα5 complex. The molecular weights of the proteins that were used for the calibration of the column (protein standards from Amersham Biosciences: ferritin (440 kDa), catalase (232 kDa), aldolase (158 kDa), albumin(67 kDa), ovalbumin (43 kDa), chymotrypsin (25 kDa); vv= void volume) are indicated above the elution profile. (b) SDS gel of fractions 5 and 7–9 of the elution profile shown in panel 3a. (c) Analytical ultracentrifugation experiment on the recNP–Impα5 complex, fraction 8 from the elution shown in panel a. The plot shows the sedimentation coefficient distribution derived from sedimentation velocity profile using SEDFIT.

fraction 5). The main peak elutes at 14.3 ml and contains two bands corresponding to recNP and Impα5 as seen on the SDS gel (Fig. 3b, fractions 7–9) and we estimate that the two bands on the SDS gel in Fig. 3b have a similar intensity. According to the calibration of the S200 column, indicated above the elution profile in Fig. 3a, an elution volume of 14.3 ml corresponds to that of a spherical protein with a

The affinity of both recNP and monoNP for panhandle RNA was measured by a filter binding assay. Panhandle RNA is an 81 nt RNA molecule comprising the conserved 3' and 5' ends of influenza A vRNA connected by a sequence derived from the NP gene (Baudin et al., 1994). In filter binding experiments, a very small amount of radioactively labelled RNA is mixed with increasing concentrations (from 10−9 M to 10−4 M) of NP. The mixture is then filtered through presoaked nitrocellulose membrane and only RNA that is bound to NP is retained on the filter and can be counted by Cerenkov counting. Since the amount of RNA used was extremely small (0.03 pmol), the NP concentration at 50% retention corresponds to the apparent dissociation constant Kd. The results are shown in Fig. 6a. Wild type recNP binds to panhandle RNA with an affinity of 3.8 × 10−7 M. However, the monomeric mutant R416A binds RNA with a more than tenfold lower affinity (Kd 2.6 × 10−6 M). Similarly, the E339A mutant binds to panhandle RNA with an affinity of 1.6 × 10−6 M. We further analyzed if RNA was able to bind to importin α5 or to the NP–Impα5 complex, see Fig. 6b. The importin alone did not bind RNA but the complex bound with an affinity identical to that of wt NP alone;

Fig. 4. Characterization of the monoNP–Impα5 complex. a) Elution profiles on a preparative S200 column of the influenza NP R416A mutant (monoNP) (black curve), Impα5 alone (green curve) and the incubation mixture between monoNP and Impα5 (red curve). b) SDS gels of the black (monoNP), green (Impα5) and red (monoNP–Impα5) elution profiles shown in a. c) analytical ultracentrifugation experiments on monoNP, Impα5 alone and Impα5 in complex with monoNP (second peak in the profile in a. The plot shows the sedimentation coefficient distribution derived from sedimentation velocity profile using SEDFIT. The protein concentration for the 3 different species was 0.8 mg/ml.

S. Boulo et al. / Virology 409 (2011) 84–90

87

88

S. Boulo et al. / Virology 409 (2011) 84–90

a

Time (min) -10

0

10 20 30 40 50 60 70 80 90 100

recNP

:Kd= 3.8 10-7 M

R416A NP

:Kd= 2.6 10-6 M

0.8 E339A NP

:Kd= 1.6 10-6 M

1.0

Fraction RNA bound

0.0

µcal/sec

-0.2 -0.4 -0.6

0.6

0.4

0.2

2 0 -2 -4 -6 -8 -10 -12 -14 -16 -18 -20 -22

0.0 10-9

10-8

10-7

10-6

10-5

[NP] in nM

b 1.0

0.0

0.5

1.0

1.5

2.0

2.5

:Kd= 3.8 10-7 M

recNP

recNP/Impα5 :Kd= 3.5 10-7 M

Kd = 26 +/-3 nM

3.0

Molar Ratio Fig. 5. Affinities measurement of monoNP to human importin α5. ITC determination of the binding thermodynamics of monoNP to human Impα5. Titration of 126 μM monoNP (R416A) into a solution of 11 μM Impα5. The experiments were performed in 40 mM Tris–HCl buffer, pH 7.5, 200 mM NaCl and 1 mM β-mercaptoethanol at 25 °C.

Fraction RNA bound

kcal/mole of injectant

-0.8

0.8

Impα5

:Kd= -

0.6

0.4

0.2

0.0 10-9

10-8

10-7

10-6

10-5

[NP] in nM 3.5 × 10−7 M for the complex compared to 3.8 × 10−7 M for NP alone. The shapes of the two curves are very similar and these results suggest that importin α binding to influenza NP leaves the RNA binding platform on NP accessible. Discussion The gene for the recombinant NP used in this work was isolated from the A/PR/8/34 strain whereas the nucleoprotein used by Ye et al. (2006) was from the A/WSN/33 strain. Although strain specific differences may exist in the polymerization state of isolated NP (Ye et al., 2006), our results are in general agreement with those of Ye and coworkers that most of the protein exists as small oligomers going from monomers to trimers and tetramers with only a few larger aggregates. These results are also more or less in agreement with those on NP isolated from A/PR/ 8/34 virus and prepared for EM under the same conditions (incubation at 37 °C) (see Fig. 2b in Ruigrok and Baudin, 1995). We also confirmed the results by Elton et al. (1999a) and Ye et al. (2006) on the monomeric state of the R416A mutant. R416 forms a salt bridge with E339 in the crystal structures of the NP trimer (Ye et al., 2006; Ng et al., 2008) and we show here that the E339A mutant is also monomeric, confirming the crucial role for this salt bridge in trimer formation. Wild type recombinant NP binds to RNA with an affinity 10 times lower than NP isolated from A/PR/8/34 virus (Kd 3.8 × 10−7 M for recombinant NP vs. 3.7 × 10−8 M for viral NP, both measured with the same 81 nt panhandle RNA (Baudin et al., 1994). The experiment with viral NP was repeated with the same result (not shown). For recNP the presence or absence of an N-terminal His-tag did not influence the affinity for RNA. We do not have a clear cut explanation for the difference in RNA affinity between NP isolated from virus and recNP. We did find, however, that viral NP is phosphorylated to some extend

Fig. 6. RNA-binding affinity of wt and mutant influenza NP and the NP–Impα5 complex. (a) Titration curves from filter binding experiments of panhandle vRNA with a) nucleoprotein recNP (―), NP R416A mutant (– ∙ ∙ –) and NP E339A mutant (∙∙∙∙∙∙). The mid-points of the curves represent the Kd values. The bars represent the standard deviation of the results from 4 independent experiments. (b) Titration curves from filter binding assays of panhandle vRNA with recNP (―), the recNP– Impα5 complex (-∙– –) and Impα5 alone (­­­­). The mid-point of the curve represents the Kd value.

whereas recombinant NP is not (not shown). The two monomeric NP mutants bind RNA with an even lower affinity; 2.6 and 1.6 × 10−6 M for R416A and E339A, respectively. This could be due to the fact that these proteins cannot oligomerize onto the RNA (Chan et al., 2010), or to the fact that these mutations lead to a conformational change of NP or to the fact that the RNA binding platform is affected by these mutations. Elton et al. (1999b) performed UV cross-linking experiments on the binding properties of NP for RNA. Using this technique, they could not detect RNA binding to R416A at 37 °C although binding was observed when the protein was grown and assayed at 30 °C. The incubation for the filter binding experiments shown here was done at 4 °C. Importin α5 binds to NP with an affinity similar to other importin– cargo complexes (Yang et al., 2010). Two importin binding sites have been proposed on NP, a monopartite N-terminal region (residues 1–13) and a bipartite motif located between amino acids 198-KR-199 and 213-RKTR-216. Because the bipartite signal overlaps with the putative RNA binding platform (Ng et al., 2008) and because importin binding has no effect on RNA binding of NP, it is most likely that the N-terminal NLS is used. Recombinant NP in the NP-importin complex is monomeric, like the R416A and E339A mutants but has the same affinity for RNA as free recNP. This implies that it is not the fact that the two mutants are monomeric that lowers the affinity for RNA by 20

S. Boulo et al. / Virology 409 (2011) 84–90

times compared to wt NP. Therefore, it is probable that the two mutations have an influence on the structure of the RNA binding platform or that they do not allow possible conformational changes in NP upon RNA binding. Finally, the non-segmented RNA viruses code for a phosphoprotein that, as one of its functions, keeps the viral nucleoprotein soluble and prevents it from binding to non-viral RNA before it is delivered onto the newly produced viral or complementary RNA (Albertini et al., 2008). We had hypothesized that importin α could fulfil the same function but our RNA binding experiments using the NP–importin complex show that this is not the case. However, importin binding to NP may prevent oligomerization of NP in the nucleus and, as such, perform a chaperone function during the influenza virus infection process. Materials and methods Recombinant NP proteins The nucleoprotein gene from influenza virus A/PR/8/34 (recNP for recombinant NP), which codes for a 498-residue protein, was expressed using a recombinant plasmid pET-16b (Novagen) containing the full length gene with a His-Tag at its N-terminal end. To remove the His-Tag, we changed the cleavage site Factor X to a TEV site. The mutations R416A and E339A of NP were constructed by, respectively, replacing the codons R416 and E339 to GCA coding for alanine, using an inverse PCR method (Quickchange, Stratagene). E. coli BL21 (DE3) pLysS (Stratagene) was used for protein expression. Expression of the recombinant proteins was induced with 0.5 mM isopropyl-1-thio-β-D-galactopyranoside (IPTG) for 5 h at 22 °C. Cells were harvested by centrifugation and then resuspended in lysis buffer containing 20 mM Tris–HCl, pH 7.9, 500 mM NaCl, 10 mM imidazole, 10% glycerol and 1 mM β-SH. After opening the cells by sonication, the cell debris was removed by centrifugation for 1 h at 20,000g at 4 °C. The supernatant was filtered through a 0.2 μm membrane and loaded onto a Co2+ charged affinity column (Sigma). The column was washed with 20 mM Tris–HCl, pH 7.9, 1 M NaCl, 10 mM imidazole, 10% glycerol, 1 mM β-SH. NP was eluted with 500 mM imidazole and dialysed against 20 mM Tris–HCl, pH 7.9, 200 mM NaCl, 10% glycerol and 1 mM β-SH. When needed, the His-Tag was removed with TEV protease (ratio 1/100) overnight at room temperature. The cut protein was further purified by gel filtration on a HiLoad 16/60 Superdex 200 column (GE Healthcare) equilibrated with 40 mM Tris–HCl, pH 7.5, 200 mM NaCl, 1 mM β-mercaptoethanol buffer. Purified protein was at least 95% pure according to Coomassie SDS-polyacrylamide gel electrophoresis.

89

and importin α5 protein was eluted with 500 mM imidazole. The HisTag was removed with TEV protease (1/100) overnight and then purified by a gel filtration chromatography on a HiLoad 16/60 Superdex 200 column (GE Healthcare) equilibrated with 40 mM Tris–HCl, pH 7.5, 200 mM NaCl, 1 mM β-mercaptoethanol buffer. The flow rate was 2.5 ml/min and the A280nm was measured. The elution volume of selected protein standards was used to construct a calibration curve (data not shown). Purified importin α5 was at least 95% pure according to Coomassie SDS-polyacrylamide gel electrophoresis and had an A260nm/A280nm ratio of at least 0.6. Complex formation between recNP and importin α5 After His-Tag removal, importin α5 and NP were mixed together in a ratio : 3 Impα5/1 recNP (mole/mole). The mixture was loaded on a Superdex 200 column in the buffer Tris–HCl (pH 7.5), 30 mM, NaCl 150 mM and β-SH 1 mM. The complex was eluted at a retention volume of 14.6 ml. Samples were further visualized on a Coomassie SDS PAGE stained gel. Filter binding experiments The method was described in Baudin et al. (1994). Briefly, the radiolabelled RNA was first denatured in water at 95 °C for 1 min, and then properly folded in the following buffer: 50 mM Tris–HCl (pH 7.5), 100 mM NaCl, 10 mM MgCl2 and 2.5 mM DTT. Protein samples were serially diluted in the same buffer from 10−4 M to 10−9 M before addition of fixed amount of RNA. After an incubation of 1 h at 4 °C, the mixture was filtrated through a nitrocellulose filter. These filters were counted in a Cerenkov scintillation counter. Analytical ultracentrifugation Sedimentation velocity experiments were done at 20 °C using 2-channel charcoal centerpices at 46,000 rpm in a Beckman Optima XL-A centrifuge (Beckman-Coulter) fitted with a four-hole AN-60Ti rotor. Sedimentation velocity profiles were collected by monitoring the absorbance signal at 280 nm in 40 mM Tris–HCl, pH 7.5, 200 mM NaCl, 1 mM β-mercaptoethanol buffer. Sedimentation coefficients and molecular weight distributions were analysed by the C(s) method implemented in the Sedfit software package (Schuck, 2000). Buffer density and viscosity corrections were made according to the data published by Laue et al. (1992). Isothermal titration calorimetry (ITC)

Recombinant human importin α5 Human importin α5 (473-residues), was expressed using a recombinant plasmid pProEX HTb (Invitrogen). The first sixty-six amino acid residues were deleted as well as the last twenty. Expressed importin α5 contains an N-terminal His-Tag provided by the vector and a TEV cleavage site. The sequence of importin α5 gene was verified by standard dideoxy sequencing. The plasmid was transformed into E. coli BL21 competent cells (Stratagene). The cells were grown in Luria–Bertoni medium at 37 °C until OD 600 nm reached 0.8. Expression of the recombinant protein was then induced with 0.5 mM isopropyl-1-thio-β-D-galactopyranoside (IPTG) for 5 h at 22 °C. Cells were harvested by centrifugation and then suspended in lysis buffer containing 20 mM Tris–HCl, pH 8.45, 100 mM NaCl, 10 mM imidazole, 10% glycerol and 1 mM β-mercaptoethanol. The cells were disrupted by sonication on ice. The cell debris was removed by centrifugation for 1 h at 20,000g, 4 °C. The supernatant was filtered over a 0.2 μm membrane and loaded onto a Co2+ charged affinity column (Sigma). The column was washed with 20 mM Tris–HCl, pH 8.45, 500 mM NaCl, 10 mM imidazole, 10% glycerol, 1 mM β-mercaptoethanol buffer

ITC was performed using a VP-ITC Microcal calorimeter (Microcal, Northhampton, MA, USA). The protein was dialysed extensively against ITC buffer (40 mM Tris–HCl, pH 7.5, 200 mM NaCl, 1 mM β-mercaptoethanol) prior to all titrations. Experiments were performed at 25 °C. A typical titration consisted of injecting 12 μL aliquots of 50–70 μM of protein with a syringe into 5–7 μM protein solutions in the cell at time intervals of 5 min to ensure that the titration peak returned to the baseline. ITC data were corrected for the heat of dilution by subtracting of the mixing enthalpies for titrant solution injections into protein-free buffer. The ITC data were analyzed using program Origin version 5, provided by the manufacturer. Acknowledgments We thank Dr Vladimir Rybin at the Protein Expression and Purification Core Facility, EMBL Heidelberg for AUC and ITC experiments and Claire Blachier-Batisse, EMBL Heidelberg, for electron microscopy. This work was funded by the EU FLUPOL contract (SP5BCT-2007-044263) and the French ANR FLU INTERPOL contract (ANR-

90

S. Boulo et al. / Virology 409 (2011) 84–90

06-MIME-014-02). S.B. was supported by Conseil Régional Rhône Alpes (cluster 10 infectiologie). References Albertini, A.A.V., Schoehn, G., Weissenhorn, W., Ruigrok, R.W.H., 2008. Structural aspects of rabies virus replication. Cell. Mol. Life Sci. 65, 282–294. Baudin, F., Bach, C., Cusack, S., Ruigrok, R.W.H., 1994. Structure of influenza virus RNP. I. Influenza virus nucleoprotein melts secondary structure in panhandle RNA and exposes the bases to the solvent. EMBO J. 13, 3158–3165. Bishop, D.H., Obijeski, J.F., Simpson, R.W., 1971. Transcription of the influenza ribonucleic acid genome by a virion polymerase. I. Optimal conditions for in vitro activity of the ribonucleic acid-dependent ribonucleic acid polymerase. J. Virol. 8, 66–73. Chan, W.H., Ng, A.K., Robb, N.C., Lam, M.K., Chan, P.K., Au, S.W., Wang, J.H., Fodor, E., Shaw, P.C., 2010. Functional analysis of the influenza virus H5N1 nucleoprotein tail loop reveals amino acids that are crucial for oligomerization and ribonucleoprotein activities. J. Virol. 84, 7337–7345. Chen, G.W., Chang, S.C., Mok, C.K., Lo, Y.L., Kung, Y.N., Huang, J.H., Shih, Y.H., Wang, J.Y., Chiang, C., Chen, C.J., Shih, S.R., 2006. Genomic signatures of human versus avian influenza A viruses. Emerg. Infect. Dis. 12, 1353–1360. Compans, R.W., Content, J., Duesberg, P.H., 1972. Structure of the ribonucleoprotein of influenza virus. J. Virol. 10, 795–800. Cook, A., Bono, F., Jinek, M., Conti, E., 2007. Structural biology of nucleocytoplasmic transport. Annu. Rev. Biochem. 76, 647–671. Cros, J.F., Garcia-Sastre, A., Palese, P., 2005. An unconventional NLS is critical for the nuclear import of the influenza A virus nucleoprotein and ribonucleoprotein. Traffic 6 (3), 205–213. Digard, P., Elton, D., Bishop, K., Medcalf, E., Weeds, A., Pope, B., 1999. Modulation of nuclear localization of the influenza virus nucleoprotein through interaction with actin filaments. J. Virol. 73, 2222–2231. Duesberg, P.H., 1969. Distinct subunits of the ribonucleoprotein of influenza virus. J. Mol. Biol. 42, 485–499. Elton, D., Medcalf, E., Bishop, K., Digard, P., 1999a. Oligomerization of the influenza virus nucleoprotein: identification of positive and negative sequence elements. Virology 260 (1), 190–200. Elton, D., Medcalf, L., Bishop, K., Harrison, D., Digard, P., 1999b. Identification of amino acid residues of influenza virus nucleoprotein essential for RNA binding. J. Virol. 73 (9), 7357–7367. Fontes, M.R., Teh, T., Kobe, B., 2000. Structural basis of recognition of monopartite and bipartite nuclear localization sequences by mammalian importin-alpha. J. Mol. Biol. 297 (5), 1183–1194. Gabriel, G., Dauber, B., Wolff, T., Planz, O., Klenk, H.D., Stech, J., 2005. The viral polymerase mediates adaptation of an avian influenza virus to a mammalian host. Proc. Natl Acad. Sci. USA 102, 18590–18595. Gabriel, G., Herwig, A., Klenk, H.D., 2008. Interaction of polymerase subunit PB2 and NP with importin α1 is a determinant of host range of influenza A virus. PLoS Pathog. 4, e11. Kingsbury, D.W., Jones, I.M., Murti, K.G., 1987. Assembly of influenza ribonucleoprotein in vitro using recombinant nucleoprotein. Virology 156, 396–403. Laue, T.M., Shah, B.D., Ridgeway, T.M., Pelletier, S.L., 1992. Analytical Ultracentrifugation in Biochemistry and Polymer Science. In: Harding, S.E., Rowe, A.J., Horton, J.C. (Eds.), The Royal Society of Chemistry, Cambridge, pp. 90–125.

Melen, K., Fagerlund, R., Franke, J., Kohler, M., Kinnunen, L., Julkunen, I., 2003. Importin alpha nuclear localization signal binding sites for STAT1, STAT2, and influenza A virus nucleoprotein. J. Biol. Chem. 278 (30), 28193–28200. Naffakh, N., Tomoiu, A., Rameix-Welti, M.A., van der Werf, S., 2008. Host restriction of avian influenza viruses at the level of the ribonucleoproteins. Annu. Rev. Microbiol. 62, 403–424. Nagata, K., Kawaguchi, A., Naito, T., 2008. Host factors for replication and transcription of the influenza virus genome. Rev. Med. Virol. 18, 247–260. Neumann, G., Castrucci, M.R., Kawaoka, Y., 1997. Nuclear import and export of influenza virus nucleoprotein. J. Virol. 71 (12), 9690–9700. Ng, A.K., Zhang, H., Tan, K., Li, Z., Liu, J.H., Chan, P.K., Li, S.M., Chan, W.Y., Au, S.W., Joachimiak, A., Walz, T., Wang, J.H., Shaw, P.C., 2008. Structure of the influenza virus A H5N1 nucleoprotein: implications for RNA binding, oligomerization, and vaccine design. FASEB J. 22, 3638–3647. O'Neill, R.E., Jaskunas, R., Blobel, G., Palese, P., Moroianu, J., 1995. Nuclear import of influenza virus RNA can be mediated by viral nucleoprotein and transport factors required for protein import. J. Biol. Chem. 270 (39), 22701–22704. Ortega, J., Martin-Benito, J., Zurcher, T., Valpuesta, J.M., Carrascosa, J.L., Ortin, J., 2000. Ultrastructural and functional analyses of recombinant influenza virus ribonucleoproteins suggest dimerization of nucleoprotein during virus amplification. J. Virol. 74 (1), 156–163. Ruigrok, R.W., Baudin, F., 1995. Structure of influenza virus ribonucleoprotein particles. II. Purified RNA-free influenza virus ribonucleoprotein forms structures that are indistinguishable from the intact influenza virus ribonucleoprotein particles. J. Gen. Virol. 76 (Pt 4), 1009–1014. Scholtissek, C., Becht, H., 1971. Binding of ribonucleic acids to the RNP-antigen protein of influenza viruses. J. Gen. Virol. 10, 11–16. Schuck, P., 2000. Size-distribution analysis of macromolecules by sedimentation velocity ultracentrifugation and lamm equation modeling. Biophys. J. 78, 1606–1619. Tarendeau, F., Boudet, J., Guilligay, D., Mas, P.J., Bougault, C.M., Boulo, S., Baudin, F., Ruigrok, R.W., Daigle, N., Ellenberg, J., Cusack, S., Simorre, J.P., Hart, D.J., 2007. Structure and nuclear import function of the C-terminal domain of influenza virus polymerase PB2 subunit. Nat. Struct. Mol. Biol. 14, 229–233. Wang, P., Palese, P., O'Neill, R.E., 1997. The NPI-1/NPI-3 (karyopherin alpha) binding site on the influenza a virus nucleoprotein NP is a nonconventional nuclear localization signal. J. Virol. 71 (3), 1850–1856. Watanabe, T., Watanabe, S., Kawaoka, Y., 2010. Cellular networks involved in the influenza virus life cycle. Cell Host Microbe 7 (6), 427–439. Weber, F., Kochs, G., Gruber, S., Haller, O., 1998. A classical bipartite nuclear localization signal on Thogoto and influenza A virus nucleoproteins. Virology 250 (1), 9–18. Whittaker, G., Bui, M., Helenius, A., 1996. Nuclear trafficking of influenza virus ribonuleoproteins in heterokaryons. J. Virol. 70 (5), 2743–2756. Yamanaka, K., Ishihama, A., Nagata, K., 1990. Reconstitution of influenza virus RNA– nucleoprotein complexes structurally resembling native viral ribonucleoprotein cores. J. Biol. Chem. 265, 11151–11155. Yang, S.N.Y., Takeda, A.A.S., Fontes, M.R.M., Harris, J.M., Jans, D.A., Kobe, B., 2010. Probing the specificity of binding to the major nuclear localization sequencebinding site of importin-α using oriented peptide library screening. J. Biol. Chem. 285, 19935–19946. Ye, Q., Krug, R.M., Tao, Y.J., 2006. The mechanism by which influenza A virus nucleoprotein forms oligomers and binds RNA. Nature 444, 1078–1082.