• Aucun résultat trouvé

Minimal features of efficient incorporation of the hemagglutinin-neuraminidase protein into sendai virus particles

N/A
N/A
Protected

Academic year: 2022

Partager "Minimal features of efficient incorporation of the hemagglutinin-neuraminidase protein into sendai virus particles"

Copied!
13
0
0

Texte intégral

(1)

Article

Reference

Minimal features of efficient incorporation of the

hemagglutinin-neuraminidase protein into sendai virus particles

ESSAIDI, Manel, SHEVTSOVA, Anastasia, ROUX, Laurent

Abstract

Two transmembrane glycoproteins form spikes on the surface of Sendai virus, a member of the Respirovirus genus of the Paramyxovirinae subfamily of the Paramyxoviridae family: the hemagglutinin-neuraminidase (HN) and the fusion (F) proteins. HN, in contrast to F, is dispensable for viral particle production, as normal amounts of particles can be produced with highly reduced levels of HN. This HN reduction can result from mutation of an SYWST motif in its cytoplasmic tail to AFYKD. HNAFYKD accumulates at the infected cell surface but does not get incorporated into particles. In this work, we derived experimental tools to rescue HNAFYKD incorporation. We found that coexpression of a truncated HN harboring the wild-type cytoplasmic tail, the transmembrane domain, and at most 80 amino acids of the ectodomain was sufficient to complement defective HNAFYKD incorporation into particles.

This relied on formation of disulfide-bound heterodimers carried out by the two cysteines present in the HN 80-amino-acid (aa) ectodomain. Finally, the replacement of the measles virus H cytoplasmic and transmembrane domains with the corresponding [...]

ESSAIDI, Manel, SHEVTSOVA, Anastasia, ROUX, Laurent. Minimal features of efficient

incorporation of the hemagglutinin-neuraminidase protein into sendai virus particles. Journal of Virology , 2014, vol. 88, no. 1, p. 303-13

DOI : 10.1128/JVI.02041-13 PMID : 24155372

Available at:

http://archive-ouverte.unige.ch/unige:34120

Disclaimer: layout of this document may differ from the published version.

1 / 1

(2)

Published Ahead of Print 23 October 2013.

2014, 88(1):303. DOI: 10.1128/JVI.02041-13.

J. Virol.

Roux

Manel Essaidi-Laziosi, Anastasia S. Shevtsova and Laurent

Protein into Sendai Virus Particles of the Hemagglutinin-Neuraminidase

http://jvi.asm.org/content/88/1/303

Updated information and services can be found at:

These include:

REFERENCES

http://jvi.asm.org/content/88/1/303#ref-list-1 at:

This article cites 25 articles, 16 of which can be accessed free

CONTENT ALERTS

more»

articles cite this article),

Receive: RSS Feeds, eTOCs, free email alerts (when new

http://journals.asm.org/site/misc/reprints.xhtml Information about commercial reprint orders:

http://journals.asm.org/site/subscriptions/

To subscribe to to another ASM Journal go to:

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(3)

Minimal Features of Efficient Incorporation of the Hemagglutinin- Neuraminidase Protein into Sendai Virus Particles

Manel Essaidi-Laziosi, Anastasia S. Shevtsova, Laurent Roux

‹Department of Microbiology and Molecular Medicine, University of Geneva, Faculty of Medicine, CMU, Geneva, Switzerland

Two transmembrane glycoproteins form spikes on the surface of Sendai virus, a member of the

Respirovirus

genus of the

Paramyxovirinae

subfamily of the

Paramyxoviridae

family: the hemagglutinin-neuraminidase (HN) and the fusion (F) proteins.

HN, in contrast to F, is dispensable for viral particle production, as normal amounts of particles can be produced with highly reduced levels of HN. This HN reduction can result from mutation of an SYWST motif in its cytoplasmic tail to AFYKD.

HN

AFYKD

accumulates at the infected cell surface but does not get incorporated into particles. In this work, we derived experi- mental tools to rescue HN

AFYKD

incorporation. We found that coexpression of a truncated HN harboring the wild-type cytoplas- mic tail, the transmembrane domain, and at most 80 amino acids of the ectodomain was sufficient to complement defective HN

AFYKD

incorporation into particles. This relied on formation of disulfide-bound heterodimers carried out by the two cysteines present in the HN 80-amino-acid (aa) ectodomain. Finally, the replacement of the measles virus H cytoplasmic and transmem- brane domains with the corresponding HN domains promoted measles virus H incorporation in Sendai virus particles.

P aramyxoviruses are enveloped viruses containing two integral envelope glycoproteins, the hemagglutinin-neuraminidase protein (HN), which is responsible for cell receptor binding/cleav- age, and the fusion protein (F), which is responsible for fusion of the viral envelope with the cellular membrane. The inner side of the viral envelope is carpeted by a layer of the matrix M protein that bridges the envelope to the nucleocapsid, the inner core of the particle. The nucleocapsid is composed of a single-stranded RNA of negative polarity, tightly wrapped by nucleocapsid proteins (N) in a structure of helicoidal symmetry, and is associated with the viral RNA-dependent RNA polymerase made of the two proteins P and L (for recent reviews about

Paramyxoviridae, see reference 1). In Sendai virus (SeV), a member of theParamyxovirinae

sub- family, genus

Respirovirus, HN is a type II glycoprotein of 576

amino acids (aa) having at its N terminus a cytoplasmic tail of 35 aa followed by a transmembrane domain of 25 aa (2). HN has two functions. On one hand, it binds to the sialic acid-containing sug- ars on cellular glycoproteins, allowing the specific attachment of the virus particle on the cell surface. This in turn triggers the fu- sion of the viral envelope with the cell membrane effectuated by the F protein. On the other hand, HN contains a neuraminidase (NA) activity, which, by analogy with the influenza virus NA, is likely to allow detachment of the viral particles at the end of the life cycle. HN protrudes from the viral envelope (and is expressed at the cell surface) as a homotetramer composed of two homodimers in which the monomers are covalently linked by disulfide bonds (1,

3,4). SeV HN protein shares these structural features with the

other members of the family. The C-terminal portion of HN forms a globular head that is the site of its biological activities (receptor binding and cleavage), and the N-terminal tail contains the signals for HN incorporation into cellular and viral mem- branes. Between the membrane anchor and the globular head is a stalk region of 80 to 90 aa containing the cysteines that covalently link the dimers (5,

6).

The cytoplasmic tail (c) of HN is thought to be responsible for its incorporation into viral particles (7–11). For SeV HN, an SYWST motif situated at aa 10 to 14 from the N terminus was found to be essential for this incorporation. Its deletion or muta-

tion resulted in viral particles depleted of HN (7,

12). This surpris-

ing result also highlighted the dispensability of HN for viral par- ticle production. This was known for some time due to studies of a thermosensitive mutant (ts271) (13,

14) in which HN was de-

graded at nonpermissive temperatures before reaching the cell membrane (15,

16). The finding was then confirmed with the

replacement of the SYWST motif by AFYKD, a motif present at the same position of the measles virus (MeV) H protein (7). This finding was revisited more recently using small interfering RNA (siRNA) suppression of HN expression (8). In that study, in an effort to better describe the mechanism of HN incorporation in viral particles, a chimeric protein consisting of the SeV HN cyto- plasmic tail and transmembrane domains (hn-ct) fused to the MeV H ectodomain was ectopically expressed. Surprisingly, when SeV HN

AFYKD

, which is normally excluded from SeV particles, was coexpressed with the chimeric MeV hn

ct

H protein, HN

AFYKD

was recovered in virus particles (8).

The present study was devoted to understanding the mecha- nism of this transcomplementation. To this effect, hn-ct, plus in- creasing extensions of the ectodomain (hn-cte), was fused to green fluorescent protein (GFP) and expressed from a supplemental transcription unit of the SeV genome, which in addition expressed HN

AFYKD

. The analysis of the rescued recombinant viruses (rSeV) showed that as soon as the ectodomain portion in the fused hn- cte-GFP protein extended past 80 amino acids, the GFP open reading frame (ORF) was systematically interrupted, leaving ex- pression of an hn-cte capable of bringing the HN

AFYKD

into virus particles. This was achieved by formation of hetero-oligomers (hn-cte-HN

AFYKD

) dependent on cysteine residues at positions

Received22 July 2013 Accepted15 October 2013 Published ahead of print23 October 2013

Address correspondence to Laurent Roux, laurent.roux@unige.ch.

Copyright © 2014, American Society for Microbiology. All Rights Reserved.

doi:10.1128/JVI.02041-13

January 2014 Volume 88 Number 1 Journal of Virology p. 303–313 jvi.asm.org 303

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(4)

129 and 138. In this case, then, hn-cte was found in virus particles as well, along with HN

AFYKD

.

MATERIALS AND METHODS

Cells and virus infection.LLC-MK2 cells were grown at 37°C under a 5%

CO2 atmosphere in Dulbecco’s modified Eagle medium (DMEM) (Gibco-Invitrogen) supplemented with penicillin, streptomycin, and 5%

fetal calf serum (FCS) (Brunshwing). BSRT7 cells were provided by K. K.

Conzelmann (17) and grown under the same conditions except for the medium (BHK-21 Glascow minimum essential medium [GMEM];

Gibco-Invitrogen). They were treated with 0.5 ␮g/ml of Geneticin (Gibco) once every two passages. During infection, cells were incubated for 1 h at 33°C with virus appropriately diluted in basic salt solution (BSS).

After adsorption, virus-containing BSS was removed; cells were rinsed and incubated for 24 to 48 h at 33°C in 2% FCS–DMEM. Cells were lysed in lysing buffer II (150 mM NaCl, 1% sodium deoxycholate, 1% Triton X-100, 0.1% SDS, and 50 mM Tris-HCl, pH 8.3). Viral particles in the cell supernatants were pelleted through a 25% glycerol–Tris–NaCl–EDTA cushion at 13,000 rpm at 4°C for 60 min and resuspended in appropriate buffers (see the figure legends).

Constructions of full-length cDNA plasmids.SeV full-length cDNA genomes were generated by in-house standard molecular cloning strate- gies using an Fl5 backbone as previously described (18). Mutations in the HN gene (SYWST motif or cysteine at positions 69 and 78) were intro- duced by site-directed mutagenesis of the insert flanked by two unique restriction sites: ClaI (in the F gene) and ApaI (in the L gene). Supplemen- tal genes were inserted in the unique MluI restriction site flanking the gene start and end sequences of an additional transcription unit inserted be- tween the M and F genes. Schematic representations of the rSeV genomes prepared in this study are shown inFig. 1A,3A,6A, and7A. All endoge- nous HN genes encoded a protein carrying a hemagglutinin (HA) tag at the C terminus, as has been described previously (7) (Fig. 1B). All pre- pared rSeV cDNAs were sequenced, and they all obey the rule of six (19).

Rescue and stock preparation of recombinant viruses.Mutated and chimera rSeVs were rescued as previously described (18). Briefly, BSRT7 cells were transfected simultaneously with full-length genome cDNA con- structions and three pTM1-based plasmids expressing the N, P/Cstop, and L viral proteins. At 48 h posttransfection, cells were rinsed, treated with acetylated trypsin (1.5␮g/ml, 15 min, 33°C), and further incubated for 24 h at 33°C. Finally, about 2 million cells were injected into the allantoic cavities of 9-day-old embryonated chicken eggs. After 72 h of incubation at 33°C, allantoic fluids (AF) were collected and served as viral stocks. Viral particles were pelleted from 1 ml AF and analyzed by SDS- PAGE and Coomassie blue staining to monitor the presence of virus par- ticles. Additional egg-to-egg passages were performed when needed.

Titers of infectious virus were obtained as described by Sugita et al. (20).

The numbers of rescue attempts and of egg serial passages required and virus titers are summarized inTable 1.

Antibodies.In Western blotting, HN was probed with (i) a rabbit serum targeting a peptide in the cytoplasmic domain of HN and reacting equally with wild-type (wt) and SYWST/AFYKD-mutated HN (␣-HNc) (Fig. 1B), (ii) a rabbit serum raised against SDS-denatured HN (␣- HNSDS), and (iii) a mouse monoclonal antibody against the HA tag (␣- HA) (clone 16B12; Covance).␣-HA was also used in HN immunoprecipi- tations (IP). The M, N, and F proteins were probed in Western blots with anti-MSDS, -NSDS, and -FSDStargeting the SDS-denatured form of the proteins (18,21,22). Mouse monoclonal␣-N (clone M6) was obtained from Allen Portner (St. Jude Children’s Research Hospital, Memphis, TN) and used in immunoprecipitations. Mouse monoclonal antiactin (clone 4; Roche) and anti-MeV H (catalog number MAB8905; Chemicon) antibodies were used in Western blotting and immunoprecipitations, re- spectively. Mouse monoclonal anti-GFP purchased from Sigma-Aldrich (catalog number G1546, clone GSN149) was used for both techniques.

Cell surface and total IP.The same infected LLC-MK2 samples were used for cell surface and total immunoprecipitations (IP). At 24 h postin-

fection, intact cells were washed with phosphate-buffered saline (PBS) and incubatedin situwith PBS-diluted antibodies. After 2 h of incubation at 4°C, cells were washed 5 times with PBS and lysed in lysing buffer II. Cell lysates were then split into two equal parts. The first part (surface IP) was directly incubated for 2 h at 4°C with protein A-Sepharose (Roche). The second part (total IP) was further incubated (2 h at 4°C) with the same antibody before protein A-Sepharose addition. Surface and total IP sam- ples were washed twice with NET buffer (NaCl, 150 mM; EDTA, 5 mM;

Tris-HCl [pH 7.8], 50 mM; NP-40, 2%) and once with washing buffer (LiCl, 500 mM; Tris-HCl [pH 7.8], 100 mM;␤-mercaptoethanol, 1%) or three times with NET buffer only (for analysis under reducing [R] or nonreducing [NR] conditions, respectively). Finally, the samples were eluted in sample buffer containing (R conditions) or not containing (NR conditions) 1%␤-mercaptoethanol.

Western blot analysis.Cellular extracts and virus particle and immu- noprecipitation samples were analyzed by 10% or 17.5% SDS-PAGE. Sep- arated proteins were transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore) using a Trans-Blot SD transfer cell (Bio-Rad).

After incubation with appropriate antibody(ies) and the corresponding anti-mouse or anti-rabbit horseradish peroxidase (HRP)-coupled sec- ondary antibodies (Bio-Rad), proteins were detected using ECL substrate (PNR 2106 ECL Western blotting detection system [Amersham] or hy- persensitive ECL [Pierce]) and a Fujifilm LAS-4000 development system.

Isotopic radiolabeling of cells.Pulse-chase radiolabeling assays were performed at 24 h postinfection. Infected cells were deprived of methio- nine, serine, and FCS for 30 min at 37°C and then pulse-labeled for 10 or 20 min with 300␮Ci/ml of35S-labeled methionine and cysteine (Pro- mix-[35S]; Amersham Biosciences). Cells were then either collected or chased for 90 min in DMEM supplemented with 10 mM cold methionine or cysteine.35S-labeled proteins in cellular extracts or viral particles were immunoprecipitated using appropriate antibodies or directly analyzed.

Prolonged [35S]methionine-cysteine radiolabeling assays were performed from 16 to 24 h postinfection. Infected cells were incubated in FCS-free medium containing 1/10 the amount of cold methionine and cysteine and 30␮Ci/ml of [35S]methionine and [35S]cysteine at 33°C. Finally, the35S- labeled samples were resuspended in sample buffer (under reducing or nonreducing conditions) and analyzed by SDS-PAGE. The proteins were detected in the dried gel using a Typhoon FLA 7000 phosphorimager (GE Healthcare).

Immunofluorescence staining and confocal microscopy.Infected LLC-MK2 cells were seeded on sterilized coverslips coated with polylysine (Sigma). Twenty-four hours later, cells were buffered with 20 mM HEPES (pH 7.5) in DMEM and fixed for 15 min at room temperature with 4%

paraformaldehyde in H2O, pH 7.3 (PFA). The nuclei were stained with DAPI (4=,6=-diamidino-2-phenylindole) (Boehringer Mannheim GmbH). Cells were mounted in Fluoromount-G (Southern Biotech) and analyzed with an LSM510 (Carl Zeiss) confocal microscope via a 63⫻/1.4 oil immersion objective. Acquisition, analysis, and treatment imaging were performed using the Zeiss LSM Image Browser.

RESULTS

The cytoplasmic and transmembrane domains of HN are not sufficient to promote HN

AFYKD

incorporation into viral parti- cles. As Sendai virus (SeV) HN protein is dispensable for virion production, its incorporation into particles appears to be inde- pendent from the steps needed to form these particles. In attempts to clarify the mechanism underlying this incorporation, we have previously shown that ectopic expression of the cytoplasmic and transmembrane domains of HN

wt

was apparently sufficient for the uptake into viral particles of a HN

AFYKD

that was otherwise excluded (see the introduction) (8). In an attempt to clarify the mechanism underlying this transcomplementation, we generated various rSeVs which expressed green fluorescent protein (GFP) fused to HN cytoplasmic and transmembrane domains (hn-ct). In

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(5)

some cases, this included increasing lengths of amino acids from the HN ectodomain proximal to the transmembrane domain (Fig. 1, hn-cte’s).

Figure 2

presents our first attempts to rescue HN

AFYKD

incor- poration by coexpression in

trans

of hn-ct or hn-cte fused to GFP.

Three ectopic HN domains were tested: (i) the cytoplasmic tail in its wt configuration plus the transmembrane domain (Fig. 1, hn- ct-GFP), the same two domains with the AFYKD mutation (hn- c

m

t-GFP), and the same two domains in the wt configuration plus 26 aa of the ectodomain proximal to the membrane (hn-cte

26

- GFP). No differences were noted in the levels of the viral proteins detected in cells infected with the various viruses indicated (Fig. 2A,

IC); that is, HN was uniformly expressed. The analysis of the viral particles produced shows that HN

AFYKD

levels were highly reduced compared to those of HN

wt

(Fig. 2A, VP, lane 2 and lanes 3 to 6) in all cases.

Figure 2A

(IC, middle panel, lanes 4 to 6) shows the level of hn-ct expression detected with antibodies raised against a peptide from the HN cytoplasmic tail (Fig. 1B,

␣-HNc). It appears that addi-

tion of the 26 aa of the ectodomain favors the expression of the chi- meric GFP (Fig. 2A, IC, middle panel, lane 5); this is detected as well with

-GFP (Fig. 2A, lower panel, lane 5). When cell surface immu- noprecipitations were performed (Fig. 2B,

S

lanes), no sign of cell surface expression of the chimeric GFP proteins was obtained (

- GFP lanes). This contrasts with the efficient cell surface expression of

FIG 1Schematic representation of rSeV genomes and of some HN protein features relevant to this study. (A) Outline of the parental SeV genome with its 6 genes flanked by the genomic promoter (GP) and the complement of the antigenomic promoter (AGPc). 1, Blow-up of the HN gene as it stands in rSeV-HNwtvirus.

The HN open reading frame (ORF) is flanked with gene start and end signals (gray bent arrows and lines) and with 3=and 5=untranslated sequences (UTR) (light gray thin tubes). In its ORF (wide colored tubes), the domains are indicated: cytoplasmic (c, in blue) carrying the SYWST motif, transmembrane (t, in red), ectodomain (e, in gray), and HA tag (HA, in yellow). 2, Outline of rSeV-GFP/HNwtvirus, as in construct 1, with the rSeV genome containing an additional transcriptional unit between the M and F genes, encoding the green fluorescent protein (GFP) (in green). 3, Outline of rSeV-GFP/HNAFYKDvirus, as in construct 2 except that the SYWST motif is replaced by AFYKD. 4, Outline of rSeV-hn-ct-GFP/HNAFYKDvirus, as in construct 3 but with the GFP ORF fused at its N terminus to the hn cytoplasmic and transmembrane domains. 5, Outline of rSeV-hn-cmt-GFP/HNAFYKDvirus, as in construct 4 but with the motif SYWST replaced by AFYKD in the hn cytoplasmic domain. 6, Outline of rSeV-hn-cte26-GFP/HNAFYKDvirus. As in construct 4 but with the hn-ct domains extended with 26 amino acids of the hn ectodomain. Only highlighted parts (white on black) of the virus denominations were used for the annotation of the viruses in the figures and in Results. (B) Outline of the HN protein pointing to features important for this study (color code is as in panel A). HN is 593 amino acids long (576 aa⫹ HA tag). Its wild-type (wt) cytoplasmic domain contains the10SYWST14motif, replaced by AFYKD as in HNAFYKD.␣-HNc, outline of the cytoplasmic peptide sequence used to raise the␣-HNc rabbit serum reacting against HN (wild type and mutated HNAFYKD). hn-cte26, hn-cte80, and hn-cte130, sequence additions with increasing extensions of the ectodomain fused to GFP, as in recombinant viruses analyzed inFig. 2,3,4, and6. The numbers in the denominations refer to the length of the ectodomain only, while the numbers in the figure refer to the total length form the HN N terminus. The cysteine residues (in red in the ectodmain, positions 129 and 138) refer to the cysteines changed to serines in experiments presented inFig. 6and7.

SeV HN Incorporation in Virus Particles

January 2014 Volume 88 Number 1 jvi.asm.org 305

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(6)

HN

wt

and HN

AFYKD

(

-HA lanes), confirming previous results (7,

8).

GFP expressed alone (Fig. 2B, GFP/HN

AFYKD

and

-GFP lanes) is found exclusively in total cell immunoprecipitates, and this validates the technique of total cell versus cell surface immunoprecipitation.

Confocal microscopy images of the hn-ct-GFP constructs failed as well to show the characteristic ring of cell surface-expressed proteins (Fig. 2D); the GFP staining appears to be more consistent with a perinuclear endoplasmic reticulum (ER) localization. Pulse-chase

35

S labeling (Fig. 2C) shows that the three chimeric GFP proteins are unstable compared to unmodified GFP and confirms the more effi- cient synthesis of hn-cte

26

-GFP. In conclusion, ectopic expression of hn-ct’s fused to a virally irrelevant protein (GFP) could not restore the HN

AFYKD

uptake in viral particles.

Approximately 80 amino acids of the HN ectodomain are needed for HN

AFYKD

incorporation into virus particles. Since the addition of a portion of the ectodomain (e

26

) was beneficial for the level of expression of the chimeric protein, as well as for a possible endoplasmic reticulum (ER) localization (Fig. 2D), new constructs were prepared in which the ectodomain was extended to 80 and 130 aa (Fig. 3A, hn-cte

80

-GFP and hn-cte

130

-GFP).

Analysis of the cellular extracts and viral particles produced after infection with these new rSeVs showed that, in contrast to the finding for hn-cte

26

-GFP (repeated here), coexpression of hn-

TABLE 1Rescue attempts, egg serial passages, and virus titers

rSeVa

Rescue frequencyb

Rescue at egg passagec:

Titer in AF (⫻109 PFU/ml)d

rSeV-GFP/HNwt 1/1 1 2.2

rSeV-GFP/HNAFYKD 1/1 4 0.8

rSeV-hn-ct-GFP/HNAFYKD 1/1 4

rSeV-hn-cte26-GFP/HNAFYKD 1/2 4

rSeV-hn-cmt-GFP/HNAFYKD 1/1 4

rSeV-hn-cmte26-GFP/HNAFYKD 0/2 4

rSeV-hn-cte80-GFP/HNAFYKD 1/2* 5 0.1

rSeV-hn-cte130-GFP/HNAFYKD 1/2* 6 0.3

rSeV-hn-cte80/HNAFYKD 1/1 1 2

rSeV-hn-cmte80/HNAFYKD 1/1 2 2

rSeV-hn-cte80-S2/HNAFYKD 1/1 2 0.5

rSeV-hn-cmte80-S2/HNAFYKD 1/1 2 0.2

rSeV-GFP/HN-S2 1/1 3 0.3

rSeV-MeV-H/HNAFYKD 1/1 4 0.35

rSeV-MeV-hnctH/HNAFYKD 1/1 4 0.2

aSchematic descriptions of these viruses can be found inFig. 1A,3A,5B,6A,7A, and8A.

bSuccessful rescues/total independent rescue attempts. *, virus obtained with no GFP expression due to the insertion of stop codons in ORF.

cNumber of serial egg-to-egg passages to obtain a high-titer virus stock.

dPFU/ml of the allantoic fluid stock.

FIG 2First attempt to rescue incorporation of HNAFYKDin SeV particles. LLC-MK2 cells were infected with the indicated viruses described inFig. 1A. (A) Western blot analysis of the cellular extracts (IC) and viral particles in the cell supernatants (VP) probed with␣-HNSDS,-MSDS, -NSDS, and -FSDS(upper panel),

␣-HNc (middle panel), and␣-GFP (lower panel) antibodies. GFP (black arrowhead) and hn-ct-, hn-cmt-, or hn-cte26-GFP proteins (*) are indicated. (B) Infected cells were radiolabeled with [35S]methionine and cysteine from 16 to 24 h postinfection. Surface (S) and total (T) immunoprecipitations (IP) were performed using␣-GFP or␣-HA as indicated. Lane V,35S-SeV labeled viral proteins loaded as protein markers. Immunopreciptated GFP (green arrow) and hn-ct- or hn-cte-GFP proteins (*) are indicated. (C) Eighteen hours postinfection, LLC-MK2 cells infected with the indicated viruses were radio-pulse-labeled for 10 or 20 min to control synthesis. After the 20-min pulse, the cells were further chased for 90 min. The GFP-containing proteins were then immunoprecipi- tated using␣-GFP and directly analyzed by PAGE and detected by autoradiography. (D) LLC-MK2 cells grown on coverslips were infected with the rSeVs expressing the hn-ct’s and hn-cte26fused to GFP as described for panel A. At 24 hours postinfection, the cells were fixed, stained with DAPI, and observed by confocal microscopy. GFP proteins, green; nuclei, blue.

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(7)

ct

80

-GFP and hn-cte

130

-GFP was associated with increased levels of HN

AFYKD

in virus particles (Fig. 3B, VP, upper panel, lane 4 and lanes 5 and 6). These levels could reach about 60% of normal (Fig.

3C). Surprisingly, the detailed analysis of the construct expression

failed to reveal the presence of the new chimeric proteins. For instance, in

Fig. 3B

(upper left panel), hn-cte

26

-GFP is clearly vis- ible above the M protein at its expected molecular mass (lane 4, *), but in lanes 5 and 6, no visible bands are detected at positions where hn-cte

80

- and hn-cte

130

-GFP should migrate (above the band in lane 4). In the middle panel, however, two bands were found at about 15 to 20 kDa, a much smaller molecular mass than expected (lanes 5 and 6,

). When probed with

-GFP (Fig. 3B, lower panel), these two bands were not detected, compared to free GFP or hn-cte

26

-GFP (lanes 2, 3, and 4 [*] compared to lanes 5 and 6). Finally, we note that two smaller bands are detected in virus particles when probed with

-HNc (Fig. 3B, VP, middle panel, lanes 5 and 6,

●). In conclusion, uptake of HNAFYKD

in viral particles can be partially restored by coexpression of chimeric GFP containing an HN ectodomain longer than 26 aa.

Normally, HN is found at the cell surface in the form of a homotetramer composed of two dimers in which the monomers

are linked by disulfide bonds, which are visible when analyzed under nonreducing (NR) conditions.

Figure 4A

shows the results of such an analysis using immunoprecipitates of the samples pre- sented in

Fig. 3B. The expected forms (HN monomer, HN2

, and HN

2

-HN

2

) were readily detected, either in total cell or in cell surface immunoprecipitates, for HN

wt

(lanes GFP/HN

wt

), for HN

AFYKD

(lanes-GFP/HN

AFYKD

), or for hn-cte

26

-GFP. However, the profile obtained from cells infected with hn-cte

80

- or hn- cte

130

-GFP clearly differed by exhibiting a more complex pattern, namely, by a distinct band between the HN monomer (HN) and dimer (HN

2

) and by multiple bands between the dimer and the tetramer (HN

2

-HN

2

) (Fig. 4A, ?). Similar results were obtained when a direct analysis of HN (detected by

␣-HNc) present in the

viral particle was made (Fig. 4B, NR). HN dimers and tetramers were readily detected for HN

wt

(in this case, HN monomer is not present), and other forms were detected in lanes hn-cte

80

- and hn-cte

130

-GFP (marked with a question mark). This pattern was better resolved with the use of 17.5% PAGE and a more sensitive ECL (Fig. 4C, NR). In particular, we note two forms migrating faster than the N protein (

), which mirror the two small bands detected under reducing conditions (Fig. 4C,

R,●) and inFig. 3B.

FIG 3Second attempt to rescue incorporation of HNAFYKDin SeV particles. (A) Schematic representation of rSeV viruses encoding hn-cte26-GFP, hn-cte80- GFP, and hn-cte130-GFP. Note that the hn-cte-GFPs are all in the wt (SYWST) configuration and that the endogenous HN is in its mutated (AFYKD) one. The same color code as forFig. 1Ais used. (B) LLC-MK2 cells were infected with the indicated viruses, and Western blot analysis of the cellular extracts (IC) and the viral particles in the cell supernatant, probed with the indicated antibody preparations, are presented. *, hn-cte26-GFP protein detected with␣-HNSDS,␣-HNc, and␣-GFP antibodies at the expected molecular mass (all panels).␣-HNc recognizes both full HN and GFP HN tails. ?, potential hn-cte80-GFP and hn-cte130- GFP proteins detected using␣-HNc antibody but with an unexpected molecular mass (middle panel). (C) HNAFYKDincorporation in VP after infections with the indicated viruses, quantified (using ImageJ software) in relation to N and normalized to incorporation of wild-type HN (lane 2 of panel B) taken as 100%.

Data were obtained from three separate experiments.

SeV HN Incorporation in Virus Particles

January 2014 Volume 88 Number 1 jvi.asm.org 307

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(8)

The complexity of the HN oligomer pattern observed upon infection by rSeV-hn-cte

80

-GFP and hn-cte

130

-GFP prompted us to verify the sequences of these viral transgenes (Fig. 5A). Remark- ably, in both cases a stop codon that prevented translation of the GFP protein had been introduced. For the hn-cte

80

-GFP gene, this

stop codon created an hn-cte protein carrying an ectodomain of 80 aa with an additional 4 aa of GFP. In the case of the hn-cte

130

- GFP, the stop codon arrested the ectodomain translation at aa 98 (Fig. 5B). These findings are consistent with the sizes of the pro- teins detected in

Fig. 3

and

4

and with the lack of reactivity with

-GFP antibody. In the end, rescue of HN

AFYKD

incorporation into viral particles was achieved with coexpression in

trans

of wt- hn-cte domains where the ectodomain was at the least 80 aa long.

These hn-cte’s were likely incorporated in viral particles (Fig. 3B and

4B) and were likely responsible for the complex pattern of HN

isoforms found under nonreducing conditions, either at the cell surface (Fig. 4A) or in viral particles (Fig. 4B), as they form het- erodimers with HN

AFYKD

. This would explain the complexity of the HN pattern observed in

Fig. 4. For example, the band above

HN in

Fig. 4A

could represent an oligomer (HN-hn-cte). Simi- larly, the two bands below N in

Fig. 4C

(NR,

), could represent homodimers of hn-cte’s as seen in

Fig. 4C

(R,

●), suggesting that

these hn-cte’s are indeed incorporated in viral particles.

Importance of Cys129 and Cys138 for rescue of HN

AFYKD

in- corporation into virus particles. Remarkably, the counterselec- tion of the GFP protein expression in rSeV-hn-cte

80

-GFP and -hn-cte

130

-GFP resulted in retaining an HN ectodomain longer than 80 amino acids in both cases. At positions 129 and 138 of the ectodomain (Fig. 1B), two cysteines are found. We therefore ex- amined the possibility that formation of disulfide-bound het- erodimers constituted the driving force for retaining this ectodo- main length and that the ability to form HN oligomers had

FIG 4Study of endogenous HN oligomerization in the presence of hn-cte26-

GFP, hn-cte80-GFP, and hn-cte130-GFP. LLC-MK2 cells were infected with the indicated viruses as forFig. 3. Cellular extracts were collected at 24 h postin- fection and subjected to surface (S) or total (T) immunoprecipitations using

␣-HA antibody. Viral particles were purified form the cell supernatants. (A) Western blot analysis of the cellular immunoprecipitates performed under nonreducing (NR) conditions and probed with␣-HNc (upper panel). N was recovered from the same samples with␣-N antibody as a control for surface IP (lower panel). (B) Western blot analysis of the purified viral particles per- formed under reducing (R) and nonreducing (NR) conditions probed with

␣-HNc antibody. (C) Exactly as in panel B except for 17.5% SDS-PAGE, as opposed to 10% in panel B, and for use of a higher-sensitivity ECL (Pierce ECL Plus).

FIG 5Sequence comparison of hn-cte-GFPs in plasmids and rescued viruses.

(A) Viral RNA was recovered from recombinant virus particles, and the genes encoding hn-cte80-GFP and hn-cte130-GFP were reverse transcribed (RT) us- ing a primer of positive polarity positioned in the 5=extremity of the M gene.

The RT products were then amplified by PCR using primers flanking the additional transcription units carrying the hn-cte-GFP genes situated between the M and F genes. RT-PCR products were sequenced and compared to the sequence of the original cDNA used for virus rescue. Highlighted nucleotides in gray indicate the G-to-T and the G-to-A substitutions leading to the inser- tion of stop codons in the hn-cte80-GFP and hn-cte130-GFP ORFs, respec- tively. (B) Schematic representation of hn-cte80-GFP and hn-cte130-GFP pro- teins theoretically expressed by the full-length cDNA plasmid used for virus rescue (plasmid) and the proteins actually expressed by the recovered viruses (virus).

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(9)

selected for the closing of the chimeric GFP ORFs. New rSeVs were therefore generated (Fig. 6A), which simply expressed hn-cte’s with an 80-aa-long ectodomain, both in the wt (hn-cte

80

) and AFYKD (hn-c

m

te

80

) configurations, as transgenes. Moreover, cys- teines 129 and 138 were alternatively replaced by serines. The vi- ruses were tested with their appropriate controls expressing HN

wt

or HN

AFYKD

. In the cellular extracts, the viral proteins, namely, HN, were equally expressed (Fig. 6B, IC). For virus particles (VP), lanes 1 and 2 serve as controls, showing normal and restricted HN incorporation (Fig. 6D). We found that HN was recovered to a significant level only when hn-cte was in its wt configuration (Fig.

6B, lane 3, andD). Cysteine-to-serine substitutions had a negative

effect on this rescue (Fig. 6B, lane 4). As for hn-c

m

te, where AFYKD replaces SYWST, HN uptake in viral particles was mini- mal if not undetectable (Fig. 6B, lanes 5 and 6 and,

D). We note the

presence in

Fig. 6B

(VP, lanes 3 and 4) of a small band in VP samples corresponding to hn-cte

80

. Analysis performed under nonreducing conditions showed evidence of hn-cte involvement in heterodimer (HN-hn-cte) formation only when the 2 cysteines were present (Fig. 6C, IC, lanes 3 and 5, and NR, *). In addition, hn-cte was present in immunoprecipitates obtained with

␣-HA

(Fig. 6C, IC, lanes 3 and 5, R,

), and evidence of mixed oligomer (HN-hn-cte) correlated with HN

AFYKD

incorporation in viral par- ticles only when hn-cte was wild type (Fig. 6C, VP, lane 3, *).

hn-cte dimer (hn-cte2) was formed only with the cysteine-con-

taining hn-cte (Fig. 6C, VP, compare lanes 3 and 4, lower panel, *).

In summary, these data show that one or both of the two cysteines are engaged in disulfide-bound oligomer formation (likely hn- cte-HN

AFYKD

) resulting in incorporation of HN

AFYKD

in viral par- ticles. Replacement of cysteines with serines reduced HN

AFYKD

uptake, possibly by weakening the hn-cte-HN

AFYKD

interaction (Fig. 6B and

C, VP, lane 4).

Cysteines 129 and 138 are not essential for HN

wt

incorpora- tion in SeV particles. We next examined the role of the two cys- teines, namely, the ability to form covalently linked dimers, in the normal situation where integral HN

wt

s are incorporated into virus particles. A new rSeV was produced where the endogenous HN

wt

carried the two serine substitutions in place of the two cysteines (S2) (Fig. 7A). Clearly, S2 substitutions had no detrimental effect on HN incorporation in virus particles (Fig. 7B, VP, lanes 1 and 3, and C). Remarkably, this normal incorporation takes place in the apparent absence of homodimers and homotetramers at the in- fected cell surface and in virus particles, detectable by PAGE under nonreducing conditions (Fig. 7B, NR, IC, and VP). The two cys- teines (or one of the two) are thus required for formation of de- tectable HN dimers and tetramers. However, they are not essential for HN

wt

incorporation in virus particles.

Incorporation of MeV H in SeV particles. We previously re- ported that MeV H expressed in the context of rSeV-HA- HN

AFYKD

infection was not incorporated into virus particles, even

FIG 6Importance of disulfide bond formation for HNAFYKDrescue in viral particles. (A) Schematic representation of the two hn-cte80proteins harboring SYWST or AFYKD (hn-cte80 and hn-cmte80), each presented with the normal cysteine composition at positions 69 and 78 of the ectodomain or with the replacement of the 2 cysteines (C/C) by 2 serines (S/S). The 4 rSeVs express HA-HNAFYKD, as endogenous HN. The color code is as inFig. 1A. (B and C) LLC-MK2 cells were infected with the viruses described in panel A andFig. 1A. (B) Western blot analysis of the cellular extracts (IC) and viral particles present in the cell supernatants (VP) probed with␣-HNSDS,-MSDS, -NSDS, and -FSDS(upper panel) and␣-HA (lower panel). C and S indicate the presence of cysteine or serine, respectively, at position 69 or 78 of the ectodomain of hn-cte80and hncmte80. (C) Left panels, Western blot analysis of immunoprecipitations (IP) performed with

␣-HA antibody under nonreducing (NR) (upper panel) and reducing (R) (lower panel) conditions and probed with␣-HNc. Right panels, similar Western blot analysis of the corresponding viral particles (VP) resuspended in a nonreducing sample buffer. (D). Western blots for the experiment presented in panel B (VP) plus two similar independent experiments were quantified using ImageJ software for estimation of HN incorporation in viral particles. In each lane HN was normalized to N, and the result of HNwtincorporation was taken as 100%.

SeV HN Incorporation in Virus Particles

January 2014 Volume 88 Number 1 jvi.asm.org 309

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(10)

when MeV H carried hn-ct in place of its own domains (8). In those experiments, MeV H was expressed from minigenomes am- plified in mixed virus stocks, a procedure that cannot ensure the same level of expression as that from the nondefective genome.

Based on the present results, however, hn

ct

-H would be expected to be incorporated, especially since formation of disulfide-linked dimers appears not to be essential. We therefore prepared rSeVs carrying as supplemental genes MeV H or MeV hn

ct

-H (Fig. 8A) so that MeV H expression now corresponded to that of the other SeV proteins. Infected cells were

35

S labeled, and MeV H was re- covered by immunoprecipitations from cellular extracts and from purified viral particles (Fig. 8B, lower panel).

Figure 8B

also shows expression of intracellular HN

wt

by Western blotting (IC, middle panel, lane 1) or HN

AFYKD

(IC, middle panel, lanes 2 to 4). In the right part of

Fig. 8B

(VP), at the position of the missing HN

AFYKD

, a faint protein band is visible upon H

wt

expression, and a strong protein band is detected in the case of hn

ct

-H expression (upper panel, lanes 3 and 4, respectively). Probing the viral particle pro- teins with an

␣-HA antibody showed that these proteins were not

HN

AFYKD

(middle panel, lanes 2 to 4), and specific

-H immuno- precipitations confirmed that these proteins were MeV H (lower panel, lanes 3 and 4). Thus, quantitative incorporation of MeV H into SeV particles takes place when the MeV Hct domains are replaced by those of SeV-HN. Note that although hn

ct

-H is incor- porated into SeV particles, this does not promote HN

AFYKD

up- take. Analysis of the two H proteins under nonreducing condi- tions shows that they can be found mainly in the forms of dimers and tetramers (Fig. 8C). The fact that hn

ct

-H does not promote HN

AFYKD

uptake in particles suggests that no hetero-oligomers (H-HN) can be formed, further suggesting that the HN ectodo- main (stalk) and not HN-ct is involved in HN oligomerization. In conclusion, HN-ct appears to be required and sufficient for incor- poration into SeV particles.

DISCUSSION

The

10

SYWST

14

amino acid motif in the SeV HN cytoplasmic tail was recognized as essential for HN incorporation in SeV particles.

Replacement of SYWST with the corresponding sequence AFYKD present in MeV H made HN highly defective for incorporation into virus particle (20- to 50-fold reduction). Remarkably, enough residual incorporation of HN occurred so that virus expressing HN

AFYKD

could be rescued (albeit after 4 serial passages in embry- onated eggs). rSeV-HN

AFYKD

cDNA was constructed in three dif- ferent full-length cDNA clone backbones by two different re- searchers, and this led to the recovery of three virus preparations exhibiting the same phenotype. Despite the four serial passages consistently required for virus rescue, deep sequencing of the virus obtained and used in this study showed no sequence differences relative to the wt reference sequence other than those introduced (unpublished data). The availability of rSeV-HN

AFYKD

then made possible the study of the mechanism of HN incorporation in viral particles by defining conditions that complement the HN

AFYKD

incorporation deficiency. These were, in a way, “naturally se- lected” when chimeric hn-cte (80/130)-GFP proteins were coex- pressed with HN

AFYKD

and led to the selection of hn-cte domains devoid of the GFP sequence. Moreover, the required length of the ectodomain (e) was found in the same experiments to be at least 80 amino acids. This selection presumably required 5 or 6 serial passages to recover hn-cte’s lacking the GFP ORF (Fig. 5;

Table 1),

since the recovery of virus expressing hn-cte

80

alone was achieved after a single egg passage (i.e., similar to the case for wt virus).

Thus, hn-cte

80

provided an SYWST-containing cytoplasmic tail and two cysteines at the distal end of the ectodomain (position 129 and 138), allowing disulfide linkage between hn-cte’s and HN

AFYKD

. In the end, one SYWST-containing cytoplasmic tail appears to be sufficient to recruit an otherwise deficient HN for

FIG 7Cysteines 69 and 78 are not required for HNwtincorporation in viral particles. (A) Schematic representation of rSeV-GFP/HN-C2S (construct 3), in which HN harbors replacements of cysteines 69 and 78 by serine. (B) Western blots of cellular extracts (IC, left panels) and supernatant (VP, right panels) from noninfected (M) LLC-MK2 or cells infected with the indicated viruses analyzed under reducing conditions (R) (upper panel) using

␣-HNSDS,-MSDS, -NSDS, and -FSDSand under nonreducing (NR) (lower panel) conditions using␣-HNc. (C) Quantification of HN incorporation into VP.

Western blots were quantified using ImageJ software, and HN was normalized to N content. Incorporation of HNwtwas taken as 100%. Data were obtained from three independent experiments.

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(11)

incorporation, providing that a covalently linked heterodimer can form. In some way, GFP must interfere with the proper function of the dimer formed, or its transit through the exocytosis pathway is counterselected. This counterselection does not take place in hn-cte

26

-GFP, presumably, because this protein does not have the possibility to form a heterodimer with HN

AFYKD

.

In the end, the present study shows transcomplementation by hn-cte of HN

AFYKD

uptake in viral particles. This complementa- tion takes place by hn-cte dragging HN

AFYKD

so that both partners end up in the particles. This result differs from that reported be- fore, where transcomplementation was proposed to take place without detection of the hn-cte-carrying protein (in that case, MeV hn

ct

H) in virus particles (8). In view of the selection of proper hn-cte from hn-cte-GFP described in this study, we pro- pose that a similar hn-ct emerged from hn

ct

H and went unde- tected. Although this did not occur in the present study (see

Fig.

8), the experimental approaches are sufficiently different to sup-

port the two different issues. In the 2010 study, the MeV hn

ct

H was encoded by an SeV minigenome, and this makes possible the pro- duction of the complementing hn-ct by only a fraction of mini- genomes, while the remaining still produced MeV hn

ct

H, albeit in a reduced amount. This could explain the transcomplementation and lack of detection of hn

ct

H in viral particles. Alternatively, the data obtained previously could have resulted from selection of AFYKD motif revertants that have later been obtained (unpub- lished data). These putative explanations have been excluded in the present study by systematic sequencing of the obtained vi-

ruses. In addition, hn

ct

H was well represented in viral particles and showed no ability to restore HN

AFYKD

incorporation (see

Fig. 8).

The requirement for the covalently linked dimers appears not to apply when each monomer harbors the SYWST recruiting sig- nal, as HN

wt

carrying the S2 substitutions (Fig. 7A) was equally incorporated in viral particles. It is interesting that the disulfide bonds are not required for HN insertion in the plasma membrane, suggesting that the traffic of HN through the exocytic pathway does not require formation of HN oligomers. We find it more surprising that the absence of covalently linked HN dimers does not have a larger overall effect on the virus life cycle but is re- stricted to the appearance of a small-plaque phenotype, accompa- nied by only a minimal loss of titer (Fig. 9 and

Table 1). In this

case, it is not clear whether oligomerization is totally dispensable or whether an oligomer forms that is not stable enough to be detected.

This work also provides information about the need for HN during the virus life cycle. We confirmed in this study that rSeV stocks reaching titers reduced only to a third of those of wt viruses (Table 1) (7) can have 20- to 50-fold less HN on the particles. We also show that the price to pay is a decreased infectivity, as revealed by the small-plaque phenotype of HN

AFYKD

viruses in titration assays (Fig. 9 and

Table 1). We have evidence that the different

multiplication steps develop at the same rate (including produc- tion of virus particles). It would therefore appear that virus de- pleted of HN is penalized during the attachment step, where re- ceptor binding is less efficient. Alternatively, HN depletion on the

FIG 8Incorporation of MeV H into SeV particles. (A) Schematic representation of the rSeVs expressing MeV H. As shown in construct 4, hnctH refers to MeV H in which the cytoplasmic and transmembrane regions have been replaced with the corresponding domains of SeV HN. (B) LLC-MK2 cells were infected with the indicated viruses. The infected cells were radiolabeled with [35S]methionine-cysteine at 16 h postinfection and were collected 24 h later. Upper panels, SDS-PAGE autoradiogaphy of cellular extracts (IC) and virus particles (VP). Middle panel, Western blot analysis of the SDS-PAGE probed with␣-HA. Lower panel, SDS-PAGE autoradiography of immunoprecipitates performed with␣-MeV H on total cellular extracts and VP prepared as for panel A. Note that MeV H and HN exhibit the same migration profile. (C) LLC-MK2 cells were mock infected (lanes M) or infected with rSeV-GFP/HNAFYKD(lanes 2), rSeV-MeV H/HNAFYKD(lanes 3), or rSeV-MeV hnctH/HNAFYKD(lanes 4) viruses as described for panel A. Sixteen hours postinfection, cells were [35S]methionine-cysteine radiolabeled for 24 h. Cellular extracts were directly analyzed under reducing conditions (IC/R) or immunoprecipitated using an␣-MeV H antibody (IP) and analyzed under reducing (R) or nonreducing (NR) conditions. Lower panels, Western blot analysis of the cellular extracts (IC) and of the immunoprecipitates probed with␣-HA.

SeV HN Incorporation in Virus Particles

January 2014 Volume 88 Number 1 jvi.asm.org 311

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(12)

viral envelope may limit F/HN interactions required to regulate fusion activation (23).

This work has added to our understanding of HN uptake in virus particles. Since HN

AFYKD

is transported normally to the cell surface, where it accumulates in amounts generally higher than that of HN

wt

, it is likely that it is at the cell surface that HN is recruited to the assembly complex that participates in virus budding. This raises the question (not addressed in this work) of the mechanism of this re- cruitment. The simplest explanation would involve an interaction of the HN cytoplasmic tail (via the SYWST motif) with another viral protein. The protein of choice should be M, as numerous studies have shown indirect evidence of M-HN association (reviewed in refer- ences24,

25, and11). This mechanism has not yet been examined and

will be part of another study.

ACKNOWLEDGMENTS

We thank Geneviève Mottet-Osman for invaluable technical help and Dominique Garcin and Caroline Tapparel for critical and constructive comments. We are thankful to Daniel Kolakofsky for critical reading of the manuscript.

This work was carried out with the financial support of the Swiss National Foundation for Scientific Research.

The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

REFERENCES

1.Lamb RA, Parks GD.2007. Paramyxoviridae: the viruses and their rep- lication, p 1449 –1496.InKnipe DM, Howley PM (ed), Fields virology, 5th ed, vol 1. Lippincott Williams & Wilkins, Philadelphia, PA.

2.Blumberg B, Giorgi C, Roux L, Raju R, Dowling P, Chollet A, Kola- kofsky D.1985. Sequence determination of the Sendai virus HN gene and

its comparison to the influenza virus glycoproteins. Cell41:269 –278.http:

//dx.doi.org/10.1016/0092-8674(85)90080-7.

3. Markwell MA, Fox CF. 1980. Protein-protein interactions within paramyxoviruses identified by native disulfide bonding or reversible chemical cross-linking. J. Virol.33:152–166.

4.Vidal S, Mottet G, Kolakofsky D, Roux L. 1989. Addition of high- mannose sugars must precede disulfide bond formation for proper folding of Sendai virus glycoproteins. J. Virol.63:892–900.

5.Bose S, Welch BD, Kors CA, Yuan P, Jardetzky TS, Lamb RA.2011.

Structure and mutagenesis of the parainfluenza virus 5 hemagglutinin- neuraminidase stalk domain reveals a four-helix bundle and the role of the stalk in fusion promotion. J. Virol.85:12855–12866.http://dx.doi.org/10 .1128/JVI.06350-11.

6.Yuan P, Swanson KA, Leser GP, Paterson RG, Lamb RA, Jardetzky TS.

2011. Structure of the Newcastle disease virus hemagglutinin- neuraminidase (HN) ectodomain reveals a four-helix bundle stalk. Proc.

Natl. Acad. Sci. U. S. A.108:14920 –14925.http://dx.doi.org/10.1073/pnas .1111691108.

7.Fouillot-Coriou N, Roux L. 2000. Structure-function analysis of the Sendai virus F and HN cytoplasmic domain: different role for the two proteins in the production of virus particle. Virology270:464 – 475.http:

//dx.doi.org/10.1006/viro.2000.0291.

8.Gosselin-Grenet AS, Mottet-Osman G, Roux L.2010. Sendai virus par- ticle production: basic requirements and role of the SYWST motif present in HN cytoplasmic tail. Virology405:439 – 447.http://dx.doi.org/10.1016 /j.virol.2010.06.030.

9.Kim SH, Yan Y, Samal SK.2009. Role of the cytoplasmic tail amino acid sequences of Newcastle disease virus hemagglutinin-neuraminidase pro- tein in virion incorporation, cell fusion, and pathogenicity. J. Virol.83:

10250 –10255.http://dx.doi.org/10.1128/JVI.01038-09.

10. Schmitt AP, Leser GP, Waning DL, Lamb RA.2002. Requirements for budding of paramyxovirus simian virus 5 virus-like particles. J. Virol.

76:3952–3964.http://dx.doi.org/10.1128/JVI.76.8.3952-3964.2002.

11. Takimoto T, Portner A.2004. Molecular mechanism of paramyxovirus budding. Virus Res. 106:133–145. http://dx.doi.org/10.1016/j.virusres .2004.08.010.

12. Takimoto T, Bousse T, Coronel EC, Scroggs RA, Portner A.1998.

Cytoplasmic domain of sendai virus HN protein contains a specific sequence required for its incorporation into virions. J. Virol.72:9747–

9754.

13. Portner A, Marx PA, Kingsbury DW.1974. Isolation and character- ization of Sendai virus temperature-sensitive mutants. J. Virol. 13:

298 –304.

14. Portner A, Scroggs RA, Marx PA, Kingsbury DW.1975. A temperature- sensitive mutant of Sendai virus with an altered hemagglutinin- neuraminidase polypeptide: consequences for virus assembly and cytopathol- ogy. Virology67:179 –187.http://dx.doi.org/10.1016/0042-6822(75)90415-8.

15. Stricker R, Roux L.1991. The major glycoprotein of Sendai virus is dispensable for efficient virus particle budding. J. Gen. Virol.72:1703–

1707.http://dx.doi.org/10.1099/0022-1317-72-7-1703.

16. Tuffereau C, Portner A, Roux L. 1985. The role of haemagglutinin- neuraminidase glycoprotein cell surface expression in the survival of Sen- dai virus-infected BHK-21 cells. J. Gen. Virol.66:2313–2318.http://dx.doi .org/10.1099/0022-1317-66-11-2313.

17. Buchholz UJ, Finke S, Conzelmann KK. 1999. Generation of bovine respiratory syncytial virus (BRSV) from cDNA: BRSV NS2 is not essential for virus replication in tissue culture, and the human RSV leader region acts as a functional BRSV genome promoter. J. Virol.73:251–259.

18. Mottet-Osman G, Iseni F, Pelet T, Wiznerowicz M, Garcin D, Roux L.

2007. Suppression of the Sendai virus M protein through a novel short interfering RNA approach inhibits viral particle production but does not affect viral RNA synthesis. J. Virol.81:2861–2868.http://dx.doi.org/10 .1128/JVI.02291-06.

19. Calain P, Roux L.1993. The rule of six, a basic feature for efficient replication of Sendai virus defective interfering RNA. J. Virol.67:4822– 4830.

20. Sugita K, Maru M, Sato K.1974. A sensitive plaque assay for Sendai virus in established line of monkey kidney cells. Jpn. J. Microbiol.

18:262–264.

21. Mottet G, Muhlemann A, Tapparel C, Hoffmann F, Roux L.1996. A Sendai virus vector leading to the efficient expression of mutant M pro- teins interfering with virus particle budding. Virology221:159 –171.http:

//dx.doi.org/10.1006/viro.1996.0362.

22. Tuffereau C, Roux L.1988. Direct adverse effects of Sendai virus DI FIG 9Infectivity tests of recombinant rSeV viruses. LLC-MK2 cells grown in

six-well petri dishes were infected with the indicated viruses (for their descrip- tion, seeFig. 3and6) at appropriate dilutions. After 1 h of adsorption, infected cells were incubated for 5 days at 33°C under conditions of infectious virus titration. Cells were then fixed with trichloroacetic acid (5% in H2O) for 10 min at room temperature and stained with crystal violet. (A) Images of stained wells showing the plaques. (B) Pictures of the well were taken to allow mea- surement of the plaque size using ImageJ software. In each well, the plaque size was normalized to the size of the well. n, number of plaques measured for each recombinant virus.

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

(13)

particles on virus budding and on M protein fate and stability. Virology 162:417– 426.http://dx.doi.org/10.1016/0042-6822(88)90482-5.

23. Bose S, Zokarkar A, Welch BD, Leser GP, Jardetzky TS, Lamb RA.2012.

Fusion activation by a headless parainfluenza virus 5 hemagglutinin- neuraminidase stalk suggests a modular mechanism for triggering. Proc.

Natl. Acad. Sci. U. S. A.109:E2625–E2634.http://dx.doi.org/10.1073/pnas .1213813109.

24. Coronel EC, Murti KG, Takimoto T, Portner A.1999. Human parain- fluenza virus type 1 matrix and nucleoprotein genes transiently expressed in mammalian cells induce the release of virus-like particles containing nucleocapsid-like structures. J. Virol.73:7035–7038.

25. Takimoto T, Murti KG, Bousse T, Scroggs RA, Portner A.2001. Role of matrix and fusion proteins in budding of Sendai virus. J. Virol.75:11384 – 11391.http://dx.doi.org/10.1128/JVI.75.23.11384-11391.2001.

SeV HN Incorporation in Virus Particles

January 2014 Volume 88 Number 1 jvi.asm.org 313

on December 19, 2013 by UNIVERSITE DE GENEVE http://jvi.asm.org/ Downloaded from

Références

Documents relatifs

(A) The 293T cells were transfected with a non-targeting si-CTL RNA or with si-ITCH for 72 h and then transfected with a HA-tagged ubiquitin and/or with plasmids encoding

Indeed for cone singularities of angles less than 2π, the induced metric on the boundary of the convex core might also have cone singularities of angle between π and 2π, and for

Pathogenic missense mutations decrease the proportion of cells showing one fluorescent spot (Spot Formation assay) and increase the delocalization of BRCA1 nuclear

At about the same time strong solutions were obtained by Lions and Perthame in [15] with a different method (see also [7] for a slightly simpler proof and [17] for an application to

Involvement of an Arginine Triplet in M1 Matrix Protein Interaction with Membranes and in M1 Recruitment into Virus-Like Particles of the Influenza A(H1N1)pdm09 Virus... Involvement

The indiscerned (or discerned) quantum particles converge to indiscerned (or discerned) classical particles and we deduce that the de Broglie-Bohm pilot wave is the

In such a physical situation, we keep a kinetic description, obtaining when passing to the limit a Vlasov-Fokker-Planck equation which involves a diffusion matrix defined from

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des