The creation of stable cell lines expressing Ebola virus glycoproteins and the matrix protein VP40 and generating Ebola virus-like particles utilizing an ecdysone inducible mammalian expression system

Available online at www.sciencedirect.com

Journal of Virological Methods 148 (2008) 237–243

The creation of stable cell lines expressing Ebola virus glycoproteins and the matrix protein VP40 and generating Ebola virus-like particles utilizing an ecdysone inducible mammalian expression system P.L. Melito a,∗,1 , X. Qiu a,1 , L.M. Fernando a , S.L. deVarennes b , D.R. Beniac b , T.F. Booth b , S.M. Jones a,c,d a

Special Pathogens Program, National Microbiology Laboratory, Public Health Agency of Canada, 1015 Arlington Street, Winnipeg Manitoba, R3E 3R2 Canada b Viral Diseases Division, National Microbiology Laboratory, Public Health Agency of Canada, 1015 Arlington Street, Winnipeg Manitoba, R3E 3R2 Canada c Department of Medical Microbiology, University of Manitoba, Room 603 Basic Medical Science Building, 730 William Avenue, Winnipeg Manitoba, R3E 0W3 Canada d Department of Immunology, University of Manitoba, Room 603 Basic Medical Science Building, 730 William Avenue, Winnipeg Manitoba, R3E 0W3 Canada Received 29 August 2007; received in revised form 7 December 2007; accepted 13 December 2007 Available online 1 February 2008

Abstract Ebolavirus is a filovirus that causes hemorrhagic fever in humans and is associated with case fatality rates of up to 90%. The lack of therapeutic interventions in combination with the threat of weaponizing this organism has enhanced research investigations. The expression of key viral proteins and the production of virus-like particles in mammalian systems are often pursued for characterization and functional studies. Common practice is to express these proteins through transient transfection of mammalian cells. Unfortunately the transfection reagents are expensive and the process is time consuming and labour intensive. This work describes utilizing an ecdysone inducible mammalian expression system to create stable cell lines that express the Ebolavirus transmembrane glycoprotein (GP), the soluble glycoprotein (sGP) and the matrix protein (VP40) individually as well as GP and VP40 simultaneously (for the production of virus like particles). These products were the same as those expressed by the transient system, by Western blot analysis and electron microscopy. The inducible system proved to be an improvement of the current technology by enhancing the cost effectiveness and simplifying the process. Crown Copyright © 2007 Published by Elsevier B.V. All rights reserved. Keywords: Ebola; Inducible system; Proteins; Virus-like particles

1. Introduction Ebola virus (EBOV) is a member of the family Filoviridae of which there are two genera, Ebolavirus and Marburgvirus. The genus Ebolavirus is divided further into species, named by the locations from which they were first isolated (Zaire, Sudan, Reston and Cote d’Ivoire) (Peters and LeDuc, 1999). EBOV causes hemorrhagic fever and of all the viruses that induce this type of pathology it stimulates the most severe form of the

∗ 1

Corresponding author. Tel.: +1 204 789 5097; fax: +1 204 789 2140. E-mail address: pasquale [email protected] (P.L. Melito). Equal contribution to this work.

pathology and features the highest case fatality rate (Feldmann et al., 2003). The emergence of evidence outlining efforts to weaponize filoviruses combined with the heightened sensitivity to terrorist activities has encouraged studies into the pathogenesis of the organism and into its components both of which can be applied for diagnostic and preventative measures (Bray, 2003). The 19 kb genome of EBOV codes for at least eight proteins; nucleoprotein (NP), virion proteins (VP35, VP40, VP30 and VP24), polymerase protein (L), the transmembrane glycoprotein (GP) and a soluble glycoprotein (sGP) (Feldmann et al., 2003; Geisbert and Hensley, 2004). The GP gene codes for both of the glycoproteins, the dominant product is the sGP and GP is the result of RNA editing (Sanchez et al., 1996; Voclhkov et al., 1995). The NP, VP30, VP35 and L proteins combine with the

0166-0934/$ – see front matter. Crown Copyright © 2007 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jviromet.2007.12.004

238

P.L. Melito et al. / Journal of Virological Methods 148 (2008) 237–243

viral genomic RNA to form a ribonucleoprotein complex while VP40, VP24 and GP are associated with the viral membrane (Feldmann et al., 2003; Geisbert and Hensley, 2004). The membrane associated proteins as well as sGP have served as formidable targets for the study of EBOV pathogenesis (Barrientos et al., 2004; Bosio et al., 2004; Martinez et al., 2007; Panchal et al., 2003; Sui and Marasco, 2002). They may also serve as reagents for diagnostic tests or in the case of VLPs, a vaccine alternative (Ksiazek et al., 1999; Warfield et al., 2005). Current technology to study these proteins has mainly relied on the transient transfection of mammalian cell lines with plasmids coding for the specific proteins (Noda et al., 2002; Wahl-Jensen et al., 2005a,b; Warfield et al., 2003, 2005). Production of the proteins of interest by transient transfection has been effective for small-scale studies; however the transfection reagents and preparation of the large amounts of plasmid DNA required is prohibitively expensive and limits a scale up progression. This is of particular importance where large quantities and consistent preparations are required such as for structural or vaccination studies. Employing any of these proteins as diagnostic reagents would also require large-scale production. To fulfill these characteristics the current technology would require multiple batches to be prepared and this may result in batch variability. A recent attempt to use a baculovirus expression system has been described and would seem promising to accommodate scale up and cost (Ye et al., 2006). The baculovirus expression system, however, relies on the expression of the proteins in a non-mammalian insect based cell line, giving rise to concerns of post-translational modifications such as glycosylation that may be different. Glycosylation of surface proteins contributes to both antigenic and in some cases functional properties (Paulson, 1989). Thus, it seems prudent to maintain a mammalian glycosylation pattern in order to ensure the correct function of the proteins for downstream applications and studies. The creation of stable mammalian cell lines expressing EBOV proteins and producing EBOV VLPs would address the limitations of current technology and uphold an expression system necessary to study appropriately and use these proteins.

Table 1 Primers for cloning Ebola Zaire VP40, sGPand GP1, 2 into pIND(SP1) Primer

Sequence

ZebovVP40 EcoRIf

GAC GAA TTC ATG AGG CGG GTT ATA TTG CCT AC GAC CTC GAG TTA CTT CTC AAT CAC AGC TGG GAC GAA TTC ATG GGC GTT ACA GGA ATA TTG GAC CTC GAG CTA AAA GAC AAA TTT GC GAC GAA TTC ATG GGC GTT ACA GGA ATA TTG GAC CTC GAG TTA CTA GCG CCG GAC TCT GAC C

ZebovVP40 XhoIr ZebovGP1, 2 EcoRIf ZebovGP1, 2 XhoIr ZEBOVsGP1, 2 EcoRIf ZEBOsGP XhoIr

2.2. Plasmids and cloning strategy cDNAs were amplified from existing plasmids (Wahl-Jensen et al., 2005a) utilizing primers that incorporated the required restriction sites for down stream cloning (Tables 1 and 2). Primers were synthesized at the DNA core facility of the National Microbiology Laboratory, Winnipeg, MB. cDNAs encoding EBOV GP, VP40, sGP were cloned separately into the inducible expression vector pIND (SP1) (Invitrogen Cat #V70020) at the EcoR 1 and Xho 1 restriction sites in the MCS. To accommodate expression of both VP40 and GP in the inducible expression vector the cDNAs were first cloned on either side of the internal ribosome entry site (IRES) segment of the pIRES plasmid (BD Biosciences, Cat #631605) at the Nhe 1 and EcoR 1 of MCS A and Xba 1 and Not 1 of MCS B, respectively. The resulting VP40-IRES-GP segment was then subcloned into pIND(SP1) at the Nhe 1 and Not 1 site of the MCS. Transformed bacterial clones were confirmed by restriction enzyme digestion and sequencing. The plasmid pVgRXR (Invitrogen Cat #V730-20) was used in combination with the above constructs to complete the inducible system and creation of stable cell lines. 2.3. Creation of stable cell lines

2. Materials and methods 2.1. Cells and cell lines AD-293 human embryonic kidney cells (Stratagene, Cat #240085) were maintained in Dulbecco’s modified Eagle Medium supplemented with 10% fetal bovine serum, lglutamine (l-Glut) and penicillin–streptomycin (P/S) and grown at 37 ◦ C under 5% CO2 . Medium for the stable cell lines were supplemented with an approximate 50% lethal dose (LD50) for native 293 cells to the selective reagents ZeocinTM (Invitrogen) and Geneticin® (Gibco), 300 ␮g/ml and 400 ␮g/ml, respectively. During induction the cell lines were grown in serum reduced medium OPTI-MEM® (Gibco Cat #31985) with an optimized concentration of 10 ␮M Ponesterone A (Invitrogen) and no selective reagents for 72 h (hours). Cloning and subcloning utilized the invitrogen One Shot® Top 10 chemically competent E. coli (Invitrogen) and S.O.C (Invitrogen Cat #15544-034).

Death curves of ZeocinTM (Invitrogen Cat #R250-01) and Geneticin® (Invitrogen Cat #10131) were established for the 293 cells. Lethal doses were estimated through the observation of cellular morphology (ZeocinTM ) and cell death (Geneticin® ). The lethal dose for ZeocinTM was approximately at 0.6 mg/ml Table 2 Primers for cloning Ebola Zaire VP40 and GP1, 2 into pIRES Primer

Sequence

ZebovVP40NheIf

GAC GCT AGC ATG AGG CGG GTT ATA TTG CCT AC GAC GAA TTC TTA CTT CTC AAT CAC AGC TGG GAC TCT AGA ATG GGC GTT ACA GGA ATA TTG GAC GCG GCC GCC TAA AAG ACA AAT TTG C

ZebovVP40EcoRIr ZebovGP1, 2 XbaIf ZebovGP1, 2 NotIr

P.L. Melito et al. / Journal of Virological Methods 148 (2008) 237–243

and the lethal dose for Geneticin® was approximately at 0.8 mg/ml. Primary clones were first established harbouring the pVgRXR plasmid using the methods outlined in the Ecdysone-Inducible Mammalian Expression System manual (Version H, September 2002, 25-0243). Briefly, approximately 1.0 × 106 cells were transfected with 1 ␮g of pVgRXR using 6 ␮l of Fugene 6 transfection reagent (Roche Cat #11 814 443 001) in OPTI-MEM® sera reduced media (Gibco Cat #31985), at 24 h the media was changed to DMEM 10% FBS, l-Glut, P/S and the cells were incubated for an additional 24 h, cells were then split to achieve an approximate 25% confluency in DMEM 10% FBS, l-Glut, P/S with 0.3 mg/ml of ZeocinTM and allowed to grow. Media was changed every 3 days and the formation of islets or foci was observed over a period of 10–15 days. Foci were either picked utilizing cloning cylinders (Sigma Cat #Z37,078-9), or representative plates showing distinct foci formation were trypsinized and subjected to a limited dilution in 96 well plates containing DMEM 10% FBS, l-Glut, P/S with 0.3 mg/ml of ZeocinTM . Primary clones were tested for the presence of the pVgRXR by subjecting them to a transient transfection with pIND (SP1) EBOV GP. After 24 h a 70% confluent culture of the transfected clones in 6 well tissue culture plates were induced with 10 ␮M Ponesterone A for 72 h. Supernatant was then run on 10% SDS PAGE gel and subjected to analysis by Western blot using inhouse mouse monoclonal antibodies to EBOV GP. Detection was accomplished with goat anti-mouse horse radish peroxidase (HRP) conjugated antibody (BioRad Cat #170-658) and the ECL Western blot Detection System (Amersham Biosciences Cat #RPN 2132). The pVgRXR clone that showed the strongest presence of the EBOV GP in the Western blot was chosen for further manipulation. Secondary clones were established in the same manner as the primary clones except the selected primary 293-pVgRXR clone was transfected with the pIND (SP1) plasmids containing the EBOV open reading frames of VP40, GP, sGP and VP40-IRESGP. Media used for the growth of the clones contained 0.4 mg/ml Geneticin® as well as 0.3 mg/ml of ZeocinTM . Secondary clones were tested by induction with 10 ␮M Ponesternone A in OPTIMEM® for 72 h. The culture supernatant was analyzed in the same way as that of the primary clone test except using an inhouse mouse monoclonal antibodies directed to EBOV VP40 and sGP as well as GP. Clones showing the highest levels of expression of the proteins of interest were chosen, scaled-up, and subjected to further analysis. 2.4. VLP and glycoprotein virosome purification Cell culture supernatant was harvested from a T-150 tissue culture flask after 72 h induction or transfection and centrifuged at 500 × g for 10 min at 4 ◦ C to remove any cell debris. The clarified supernatant was then layered over an 8 ml 20% sucrose cushion in 25 mm × 89 mm Ultra Clear centrifuge tubes (Beckman Cat #344058). The sucrose cushion was centrifuged in a Beckman Optima L-70K Ultracentrifuge using a SW28 rotor for 2 h at 28,000 rpm. The supernatant and sucrose cushion was

239

then poured off and the pellet resuspended in one tube volume of 20 mM Tris, 0.1 M NaCl, 0.1 mM EDTA, pH 7.4. The preparation was then centrifuged under identical conditions for 30 min. The final pellet was resuspended in 0.5 ml of 20 mM Tris, 0.1 M NaCl, 0.1 mM EDTA, pH 7.4. 2.5. sGP purification Cell culture supernatant was harvested from a T-150 tissue culture flask after 72 h induction and centrifuged at 500 × g to remove any cell debris. The sample was then processed by fast protein liquid chromatography (FPLC) with desalting and anion exchange columns (Amersham Cat #17-408-01 and Cat #17-5053-01, respectively). The binding buffer for the anion exchange was 20 mM Tris pH 7.4 with no NaCl and the elution buffer was 20 mM Tris, 1 M NaCl pH 7.4. All peaks were analyzed by Western blot using mouse anti-EBOV sGP antibody. The sGP eluted at approximately 30% elution buffer. Fractions were collected and concentrated with an amicon centricon 4 molecular weight cut off of 30 kDa prior to analysis. 2.6. Electron microscopy The EBOV VLPs and glycoprotein virosome samples were adsorbed to a carbon coated formvar film on a 400 mesh copper grid for 1 min, washed in PBS three times for 1 min, followed by fixation for 2 min (1% paraformaldehyde, 2% glutaraldehyde in PBS). Grids were then washed in deionised water and negatively contrasted with 2% methylamine tungstate (Nano-W, Nanoprobes, and Yaphank, NY). Specimens were imaged in a FEI Tecnai 20 transmission electron microscope (TEM) operating at 200 kV, and at nominal instrument magnifications of 14,500× and 50,000×. Digital images of the specimens were acquired by an AMT Advantage XR-12 CCD camera (AMT, Danvers, MA). 2.6.1. Immuno-electron microscopy Three microliters of EBOV VLP or virosome suspension was applied to a formvar-carbon coated 400-mesh nickel grid, and incubated for 1 min. All incubations were carried out at 20 ◦ C. Grids were then washed in PBS six times for 1 min, followed by a 10-min block in PBS-G-BSA (PBS pH7.2, 0.2% glycine, 2% BSA). Grids were then washed in PBS-G (PBS pH7.2, 0.2% glycine) six times for 1 min, followed by a 1h incubation on a 30 ␮l drop of mouse monoclonal antibody directed to EBOV GP (diluted 1:3000 in PBS-G). Grids were then washed in PBS-G (PBS pH7.2, 0.2% glycine) six times for 1 min, followed by a 30-min incubation with goat antimouse IgG (conjugated to 10 nm gold, Sigma; diluted 1:10 in PBS-G). Grids were then washed in PBS-G (PBS pH7.2, 0.2% glycine) three times for 1 min, followed by fixation for 2 min (1% paraformaldehyde, 2% glutaraldehyde in PBS). Grids were then washed in deionised water, negatively stained with 2% methylamine tungstate (Nanoprobes), and observed in the Tecnai 20 G2 TEM operated under the same conditions as described in the previous section.

240

P.L. Melito et al. / Journal of Virological Methods 148 (2008) 237–243

2.6.2. Image processing Image processing was carried out using the EMAN and SPIDER/WEB image processing program packages (Frank et al., 1996; Ludtke et al., 1999) on SGI Tezro workstations running IRIX 6.5. Single particle images were selected using EMAN, picking images containing one Ebola GP spike per image. The three data sets contained the following number of images: GP only n = 1275; GP and VP40, inducible expression n = 1142; and GP and VP40, transient expression n = 998. Each data set was initially processed using a reference free-alignment in EMAN to generate class averages. Several characteristic class averages showing individual GP spikes were then selected and aligned to each other. These aligned averages were then used as references for a multireference alignment using the SPIDER software package. The multireference alignment procedure was then repeated three times, to generate the final class averages. 2.7. Proteinase K protection assay EBOV VLPs from the inducible system were incubated with proteinase K alone and with both proteinase K (Worthington, Cat #4222) and Triton X-100 (Fisher Scientific Cat #BP151-100) at the final concentration of 5 ␮g/ml and 0.01%, respectively for 1 h. Samples were heat inactivated for 1 min at 90 ◦ C then prepared for SDS-PAGE and Western blot analysis. If proteins are on the surface of the particle they will be vulnerable to the proteinase K and be digested, whereas proteins inside the membrane vesicle will remain undigested. Incubation with Triton X-100 disrupts the membrane and allows the digestion of both surface and membrane associated proteins. In-house monoclonal mouse anti-EBOV GP and VP40 antibodies were used along with goat anti-mouse HRP conjugate secondary antibody to visualize the results. 3. Results While testing the primary clones for the presence of the pVgRXR plasmid a variable expression of EBOV GP was observed, indicating variability in the efficiency of the system. One clone demonstrating the strongest expression was moved to the second stage of the cell transformation. Variability in expres-

sion was also apparent by testing the secondary clones. One clone from each; GP, VP40, sGP and VP40-IRES-GP cell line was chosen for further analysis. An optimal concentration of Ponesterone A was established by subjecting one of the cell lines to 5, 10 and 15 ␮M of Ponesterone A. An increase in expression from 5 to 10 ␮M was noticed while no significant increase was observed from 10 to 15 ␮M (data not shown). A comparison of protein expression was performed between the inducible cell lines and a transient transfection from the approximately same number of cells. Overall the transient system showed a greater amount of protein expression than that of the inducible clones (Fig. 1). A noticeable difference was also noted in the cell morphology in the two systems, particularly in the GP, VLP and VP40 cell lines. With the transient system there appeared to be a significant amount of cytotoxicty at 72 h, where cells were rounding and detaching. The inducible system showed very little cytotoxicty if any. Alazard-Dany et al. (2006) reported that EBOV GP and VP40 have less of a cytotoxic effect when expressed in low to moderate levels than high levels. This appears to be the case for the inducible system and allows for the extension of expression or induction to accommodate the decrease in product yield. Electron microscopy was performed on purified preparations (20% sucrose cushion) from the EBOV GP, VP40 and VP40-IRES-GP cell lines. The EBOV GP presents itself on a virosome that appears in a non-descript globular manner; the VP40 expressing cell line and the VP40-IRES-GP cell line produced more authentic EBOV like structures (Fig. 2). For all three data sets the EBOV GP spike has an average diameter of ∼7 nm, and height of ∼10 nm. This is consistent with what has been previously reported and comparable to the transient transfection (Noda et al., 2002). Immunogold labelling with anti-EBOV GP monoclonal antibodies and subsequent electron microscopy allowed us to visualize the incorporation of EBOV GP in both the globular structures from the EBOV GP cell line and the purified EBOV VLPs (Fig. 3). The absence of immunogold labelling of the VP40 preparation was as expected (data not shown). The proteinase K protection assay showed that VP40 was protected from digestion by the proteinase whereas GP was not indicating that GP was located on the surface of the particles and

Fig. 1. Comparative analysis of EBOV proteins expressed through transient transfection and through the ecdysone inducible mammalian system. Lane 1: Transient Transfection VLP; Lane 2: Inducible System VLP; Lane 3: Transient Transfection GP; Lane 4: Inducible System GP; Lane 5: Transient Transfection VP40; Lane 6: Inducible System VP40; Lane 7: Transient Transfection sGP; Lane 8: Inducible System sGP, Molecular weight Marker; MagicMarkTM XP (Invitrogen Cat #LC5602).

P.L. Melito et al. / Journal of Virological Methods 148 (2008) 237–243

241

Fig. 2. Electron micrographs, negatively stained with methylamine tungstate. The panels show virosomes from the inducible expression of GP only (A), VLPs from the inducible expression VP40 only (B), VLPs from the inducible expression of both GP and VP40 (C), and VLPs from the transient expression of both GP and VP40 (D). The inset at the top right corner of each panel provides a higher magnification image. Panels (A, C, D) include a gallery of representative class averages of the GP that were calculated by image analysis. Surface glycoproteins are indicated with white arrows. (White scale bars = 100 nm, black scale bars = 5 nm).

Fig. 3. Immuno-electron microscopy of particles. The panels show virosomes from the inducible expression of GP only (A), and VLPs from the inducible expression of both GP and VP40 (B). Both panels show the classical black dots indicating that they are positively labeled with immunogold. (Scale bars = 100 nm).

242

P.L. Melito et al. / Journal of Virological Methods 148 (2008) 237–243

Fig. 4. Proteinase K assay. Lane 1: VLP with both proteinase K and Triton X100; Lane 2: VLP with proteinase K only; Lane 3: control with no proteinase K or Triton X-100. Molecular weight Marker: MagicMarkTM XP (Invitrogen Cat #LC5602).

VP40 at the inner site of the membrane (Fig. 4). The addition of Triton X-100 disrupted the cell membrane completely resulting in full digestion of both proteins by proteinase K. 4. Discussion The findings demonstrate that the expressed recombinant proteins are similar to the authentic viral proteins as well as to the products of transient transfection. Although transient transfection in mammalian cell lines produces higher levels of protein than the created stable cell lines the use of these cell lines do offer advantages. The initial development of the cell lines is time consuming; however, the resulting system can easily be scaled up using less reagents and DNA, thus lowering the cost. The cells from the inducible system also show less cytotoxicity. Overall, in the long-term, greater amounts proteins can be produced at a lower cost. EBOV VLPs have been shown to illicit immune responses and protect mice from lethal challenges of EBOV virus suggesting that these entities could function as a therapeutic by inducing the production of neutralizing antibodies (Takada et al., 2007; Warfield et al., 2003). Further studies are necessary to confirm this application. The ability to scale up the production of EBOV VLPs at a lower cost using the stable cell lines would further progress their assessment as a vaccine candidate. EBOV VLPs or glycoprotein virosomes may also be applied as diagnostic reagents such as capture antigens in enzyme linked immunosorbant assays (ELISA), which can be useful for the assessment of anti-EBOV antibodies in sera or in a competitive assay for epitope mapping of monoclonal antibodies to the EBOV GP. Consistency of the reagents would be paramount. Using the stable cell lines would allow for the production of a large amount of reagent from one batch, minimizing the variability that would be the result of making an equivalent amount of protein from several batches using the transient system. More importantly the stable cell lines developed from this work have a mammalian protein expression system. Ye et al.

(2006) describe an insect cell line based system for the production of VLPs, which is an economical improvement over the transient system. Some activity was maintained and consistent with previous reports of VLP activity, however the expression of the proteins was under a non-mammalian system. Posttranslational modifications such as glycosylation contribute to both structural and functional properties of proteins and may be different in the two systems (Paulson, 1989). EBOV GP is a heavily glycosylated protein, particularly at the mucin domain, which has recently been linked to the stimulation of human dendritic cells (Martinez et al., 2007). Any compromise of the pattern or the extent of the glycosylation on the EBOV GP would certainly influence functional studies and not be authentic, downstream application of the products of a baculovirus expression system as a vaccine or diagnostic reagent would be dubious. 5. Conclusion In conclusion, the use of the ecdysone inducible mammalian expression system to develop stable cell lines that produce EBOV proteins and VLPs was successful and an advancement on the current technology. The resulting products from this system can be produced in large quantities at a lower cost, in a consistent manner and maintain an expression system that substantiates downstream applications. The cell lines created in this work will prove to be useful laboratory tools for further investigations into the structure and function of EBOV membrane proteins and quite possible contribute to the development of a vaccine. Acknowledgements The authors would like to thank Judie Alimonti and Heinz Feldmann for help in preparation of the manuscript. This study was supported by a grant from the Chemical, Biological, Radiological or Nuclear Research and Technology Initiative (CRTI-01-0087RD) awarded to Steven Jones. References Alazard-Dany, N., Volchkova, V., Reynard, O., Carbonnelle, C., Dolnik, O., Ottmann, M., Khromykh, A., Volchkov, V., 2006. Ebola virus glycoprotein GP is not cytotoxic when expressed constitutively at moderate levels. J. Gen. Virol. 87, 1247–1257. Barrientos, L.G., Martin, A.M., Rollin, P.E., Sanchez, A., 2004. Disulfide bond assignment of the Ebola virus secreted glycoprotein SGP. Biochem. Biophys. Res. Commun. 323, 696–702. Bosio, C.M., Noore, B.D., Warfield, K.L., Ruthel, G., Mohamadzadeh, M., Aman, J., Bavari, S., 2004. Ebola and Marburg virus like particles activate human myeloid dendritic cells. Virology 326 (2004), 280–287. Bray, M., 2003. Defence against filoviruses used as biological weapons. Antiviral Res. 57, 53–60. Feldmann, H., Jones, S., Klenk, H.D., Schnittler, H.J., 2003. Ebola virus: from discovery to vaccine. Nat. Rev. Immunol. 3, 677–685. Frank, J., Radermacher, M., Penczek, P., Zhu, J., Li, Y., Ladjadj, M., Leith, A., 1996. SPIDER and WEB: processing and visualization of images in 3D electron microscopy and related fields. J. Struct. Biol. 116, 190–199. Geisbert, T.W., Hensley, L.E., 2004. Ebola virus: new insights into disease aetiopathology and possible therapeutic interventions. Expert Rev. Mol. Med. 6, 1–24.

P.L. Melito et al. / Journal of Virological Methods 148 (2008) 237–243 Ksiazek, T.G., West, C.P., Rollin, P.E., Jahrling, P.B., Peters, C.J., 1999. ELISA for the detection of antibodies to Ebola viruses. J. Infect. Dis. 179 (Suppl. 1), S192–S198. Ludtke, S.J., Baldwin, P.R., Chiu, W., 1999. EMAN: semi automated software for high-resolution single-particle reconstructions. J. Struct. Biol. 128, 82–97. Martinez, O., Valmas, C., Basler, C.F., 2007. Ebola virus-like particle-induced activation of NF-kappaB and Erk signalling in human dendritic cells requires the glycoprotein mucin domain. Virology 364, 342–354. Noda, T., Sagara, H., Suzuki, E., Takada, A., Kida, H., Kawaoka, Y., 2002. Ebola virus VP40 drives the formation of virus-like filamentous particles along with GP. J. Virol. 76, 4855–4865. Panchal, R.G., Ruthel, G., Kenny, T.A., Kallstrom, G.H., Lane, D., Badie, S.S., Li, L., Bavari, S., Aman, M.J., 2003. In vivo oligomerization and raft localization of Ebola virus protein VP40 during vesicular budding. PNAS 100 (26), 15936–15941. Paulson, J.C., 1989. Glycoproteins: what are the sugar chains for? Trends Biochem. Sci. 14, 272–276. Peters, C.J., LeDuc, J.W., 1999. World Health Organization, Centers for Disease Control and Prevention, Institut de m´edecine tropicale Prince L´eopold, National Institutes of Health, and European Commission. In: Ebola: The Virus and The Disease. University of Chicago Press, Chicago. Sanchez, A., Trappier, S.G., Mahy, B.W., Peters, C.J., Nichol, S.T., 1996. The virion glycoproteins of Ebola viruses are encoded in two reading frames and are expressed through transcriptional editing. Proc. Natl. Acad. Sci. U.S.A. 93, 3602–3607.

243

Sui, J., Marasco, W.A., 2002. Evidence against Ebola virus binding to human neutrophils by a specific receptor. Virology 303, 9–14. Takada, A., Ebihara, H., Jones, S., Feldmann, H., Kawaoka, Y., 2007. Protective efficacy of neutralizing antibodies against Ebola virus infection. Vaccine 25, 993–999. Voclhkov, V.E., Becker, S., Volchkova, V.A., Ternovoj, V.A., Kotov, A.N., Netesov, S.V., Klenk, H.D., 1995. GP mRNA of Ebola virus is edited by the Ebola virus polymerase and by T7 and vaccinia virus polymerase. Virology 214 (2), 421–430. Wahl-Jensen, V., Kurz, S.K., Hazelton, P.R., Schnittler, H.J., Stroher, U., Burton, D.R., Feldmann, H., 2005a. Role of Ebola virus secreted glycoproteins and virus-like particles in activation of human macrophages. J. Virol. 79, 2413–2419. Wahl-Jensen, V.M., Afanasieva, T.A., Seebach, J., Stroher, U., Feldmann, H., Schnittler, H.J., 2005b. Effects of Ebola virus glycoproteins on endothelial cell activation and barrier function. J. Virol. 79, 10442–10450. Warfield, K.L., Bosio, C.M., Welcher, B.C., Deal, E.M., Mohamadzadeh, M., Schmaljohn, A., Aman, M.J., Bavari, S., 2003. Ebola virus like particles protect from lethal Ebola virus infection. PNAS 100, 15889–15894. Warfield, K.L., Swenson, D.L., Demmin, G., Bavari, S., 2005. Filovirus-like particles as vaccines and discovery tools. Expert Rev. Vac. 4, 429–440. Ye, L., Lin, J., Sun, Y., Bennouna, S., Lo, M., Wu, Q., Bu, Z., Pulendran, B., Compans, R.W., Yang, C., 2006. Ebola virus-like particles produced in insect cells exhibit dendritic cell stimulating activity and induce neutralizing antibodies. Virology 351, 260–270.