- Email: [email protected]
Differential sensitivity to interferon influences the replication and transcription of Urabe AM9 mumps virus variants in nerve cells
Microbes and Infection 9 (2007) 864e872 www.elsevier.com/locate/micinf
Original article
Differential sensitivity to interferon influences the replication and transcription of Urabe AM9 mumps virus variants in nerve cells Nora Rosas-Murrieta a,b, Irma Herrera-Camacho b, Vero´nica Vallejo-Ruiz a, Lourdes Milla´n-Pe´rez-Pe~ na b, Carlos Cruz c, Jose´ Tapia-Ramı´rez c, Gerardo Santos-Lo´pez a, Julio Reyes-Leyva a,* a
Lab. de Virologı´a y Biologı´a Molecular, Centro de Investigacio´n Biome´dica de Oriente, Instituto Mexicano del Seguro Social, Hospital General de Zona No. 5, Km. 4.5 Carretera Atlixco-Metepec, 74360 Metepec, Pue., Mexico b Centro de Quı´mica, Instituto de Ciencias, Beneme´rita Universidad Auto´noma de Puebla, Complejo de Ciencias, Ciudad Universitaria, 72570 Puebla, Pue., Mexico c Dept. de Gene´tica y Biologı´a Molecular, Centro de Investigacio´n y Estudios Avanzados, Instituto Polite´cnico Nacional, Av. Instituto Polite´cnico Nacional 2508, Col. San Pedro Zacatenco, 07360 Me´xico D.F., Mexico Received 19 November 2006; accepted 8 March 2007 Available online 15 March 2007
Abstract Urabe AM9 mumps virus vaccine causes post-vaccination meningitis. Two variants of Urabe AM9 virus differ in their replication efficiency in human nerve cells, HN-A1081 variant being more neurotropic than HN-G1081. The effect of interferon (IFN) on viral replication and transcription was analyzed. Priming of nerve cells with IFN reduced more significantly the replication of HN-G1081 variant (from 102.5 to 101.3 TCID50) than that of HN-A1081 (from 103.5 to 102.6 TCID50). IFN-priming also reduced the transcription of HN-G1081 genes, but not of HN-A1081. The effect of viral infection on the transcription of cellular IFN responsive genes was analyzed. HN-A1081 virus reduced the transcription of STAT1, STAT2, p48 and MxA in both unprimed and IFN-primed cells; whereas HN-G1081 virus just reduced MxA transcription. Since rubulavirus V protein inhibits IFN signaling, the V mRNA was cloned and sequenced, finding that HN-G1081 but not HN-A1081 presented three extra G in the P/V edition site, producing the insertion of Gly156 in the V protein. Our results suggest that the replication efficiency of Urabe AM9 mumps virus variants is influenced by their sensitivity to interferon and their capacity to reduce the antiviral response. Ó 2007 Elsevier Masson SAS. All rights reserved. Keywords: Mumps vaccine; Interferon inhibition; Transcription; Urabe AM9; Neurovirulence
1. Introduction Mumps virus (MuV) causes an acute febrile infection that frequently affects the central nervous system, producing aseptic meningitis and sudden deafness [1]. Massive vaccination programs have decreased the incidence of MuV infection worldwide. However, reports indicate that w1/1000e10,000 recipients of Urabe AM9 and other vaccine strains developed aseptic meningitis [2,3]. The unacceptably high rate of vaccine
* Corresponding author. Tel./fax: þ52 24 44 44 01 22. E-mail addresses: [email protected], [email protected] (J. Reyes-Leyva). 1286-4579/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.micinf.2007.03.005
associated meningitis and parotitis cases has resulted in vaccine withdrawal and public resistance to mumps vaccination [2]. In consequence, mumps epidemics have re-emerged, and the incidence is rising in several countries [4]. Therefore, it is important to identify the basis of MuV neurovirulence in order to improve the safety and efficiency of mumps virus vaccines. MuV belongs to Rubulavirus genus of the Paramyxoviridae family; it has a single-stranded negative-sense RNA genome of 15,384 nucleotides that encodes seven structural proteins: nucleoprotein (NP), phosphoprotein (P), matrix (M), fusion (F), small hydrophobic (SH), hemagglutinin-neuraminidase (HN) and RNA polymerase, the so-called large (L) protein. In addition, the P gene is co-transcriptionally edited giving
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
origin to P, I and V proteins; the last regulates viral replication and inhibits antiviral response [1]. Previous reports suggested that a nucleotide change (G/A) at position 1081 in the HN gene of Urabe AM9 mumps virus was causally associated with presentation of post-vaccination meningitis cases, because HN-A1081 viruses were isolated during aseptic meningitis at higher frequency than HN-G1081 viruses [5]. Partial characterization of the HN gene amongst several vaccine and wild-type mumps virus strains led to the identification of an association of HN-A1081 genotype with neurovirulence [5,6]. We have shown that two variants of Urabe AM9 virus that were initially characterized by their difference in the HN gene nt 1081 also differ in their replication efficiency in nerve cells. Indeed, HN-A1081 variant was efficiently replicated in both human neuroblastoma cells and newborn rat brain (105 and 104 PFU, respectively), whereas HN-G1081 variant was replicated at low titers (102 PFU in both cases) [7]. The neurotropism of these variants was supported in part by their differences in cell receptor-binding affinity; i.e., HN-A1081 virus showed highest affinity towards a2-6 linked sialic acids that are highly expressed in human nerve cells; whereas HN-G1081 variant preferred a2-3 linked sialic acids that are scarcely presented in nerve cells. However, HN-G1081 variant also recognized a2-6 linked sialic acids but with lesser affinity than HN-A1081 virus [8]. The role of HN-A1081 mutation in the neurovirulence of mumps virus has been questioned because complete genome sequencing has shown that Urabe AM9 based vaccines are composed of more than two viral quasispecies presenting genetic heterogeneity at several sites [9]. In addition, adaptation of the non-neurotropic Jeryl Lynn virus strain to nerve cells increased its neurovirulence in the newborn rat model and involved changes in NP, M and L genes [10]. Moreover, it was recently found that the loss of genetic heterogeneity had a profound participation in the attenuation of Urabe AM9 virus; this included changes in coding nucleotides in NP, F, HN and L genes as well as in non-coding regions [11]. These results indicate that other viral and cellular factors are involved in determining the permissibility for mumps virus replication in nerve cells. Viral infection is controlled at cellular level by type I interferon (IFN a/b), which are produced in response to the presence of viral double stranded RNA (dsRNA). Recognition of IFN a/b by their specific cell receptors activates the signaling pathway JAK-STAT, which includes the signal transducers and activators of transcription STAT1 and STAT2. Upon phosphorylation, STAT1 and STA2 form heterodimers which translocate to the nucleus, where they associate with the DNAbinding protein p48 (IRF9) to induce the activation of several IFN responsive genes, including: dsRNA-dependent protein kinase R (PKR), 20 ,50 oligoadenylate synthetase (OAS) and MxA, leading together to inhibition of viral replication, transcription and protein expression [12]. In the present paper we analyzed whether cell priming with IFN modifies the cell permissibility for mumps virus variants; and inversely whether infection with either HN-A1081 or HNG1081 virus affects the transcription of IFN responsive genes.
865
2. Material and methods 2.1. Viruses and cells HN-A1081 and HN-G1081 clones of Urabe AM9 mumps virus vaccine were purified by plaque assays as reported previously [7]. Viruses were replicated in human neuroblastoma SH-SY5Y cells and uterine cervix carcinoma HeLa cells maintained in Dulbecco’s minimum essential medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 mg/ml streptomycin and 1% nonessential amino acids. 2.2. Fluorescence in situ hybridization SH-SY5Y and HeLa cells (104/well) cultured on chamber slides (Corning; Acton, MA) were infected with each virus variants (0.02 PFU/cell). At 48 h post infection (p.i.), cells were washed and fixed in 4% paraformaldehyde in PBS. Cells were consecutively treated with permeation solution (10 mM TriseHCl pH 7.4, 0.1% Triton X-100) for 10 min, then with pre-hybridization solution (2 SSC, 10 Denhardt’s solution, 0.1% SDS, 250 mg/ml salmon sperm DNA) 1 h, and with hybridization solution (50% deionized formamide, 2SSC, 10 Denhardt’s solution, 5% dextran sulfate, 0.2% BSA, 0.1% SDS) containing 50 ng/ml of an M gene anti-sense probe conjugated to psoralen-biotin (Ambion Inc., Austin, TX) during 18 h. After that, cells were washed with 1% SSC, incubated with FITC-conjugated streptavidin (SigmaeAldrich Chemicals, St. Louis, MO), further washed and examined under an epifluorescence microscope (Nikon Inc., Mel, NY). Images were processed and analyzed with MetaMorph Software (Universal Imaging Co., Downingtown, PA). 2.3. Mumps virus infection on interferon-primed cells SH-SY5Y and HeLa cells were seeded in 35 mm culture plates with DMEM 10% FBS and incubated at 37 C during 24 h. Cells were then treated with 2000 IU/ml human recombinant IFN-a2b (Probiomed, Mexico City) in DMEM without serum during 5 h. After that, unprimed and IFN-primed cells were inoculated with HN-A1081 and HN-G1081 virus (0.02 PFU/cell). Non-infected cells, both primed and unprimed, were used as controls. 2.4. Virus titration The supernatants of infected HeLa and SH-SY5Y cells were collected at 48 h p.i. and used to determine viral yields on Vero cells cultured on 24-wells microassay plates. Viral titers were expressed as 50% tissue culture infective doses (TCID50), which were calculated according to conventional methods. 2.5. Transcription of viral and cellular genes Mock and IFN primed HeLa and SH-SY5Y cells seeded on 35 mm plates were infected with each virus variant. Total
866
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
RNA was extracted at 48 h p.i. using Trizol reagent (Invitrogen, Carlsbad, CA). mRNA was purified with Oligotex kit (Qiagen, Germany) and treated with DNase (SigmaeAldrich), cDNA was synthesized and amplified by RTePCR using SuperScriptÔII One-Step RTePCR System with Platinum Taq (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. Design of Urabe AM9 mumps virus specific primers was based on published genome sequence (GenBank accession no. AF314559) using the Primer3 software Release 1.0.1 (http://www.sk.embnet.org/cgi-bin/primer3_www.cgi). Primer sequences for viral genes are available under request. Primers for cellular genes were obtained from previous reports [13]. The house-keeping gene cyclophilin (CP) was used as an internal control. Amplified products were analyzed by electrophoresis on a 2% agarose gel. Images were captured in a Kodak EDAS 290 LE system (Eastman Kodak Co., Rochester, NY) and analyzed with Quantity-One software (Bio-Rad, Hercules, CA). Results were reported as optical density gene/CP ratios. 2.6. Cloning and sequencing of V gene The V mRNA of both HN-A1081 and HN-G1081 virus variants was amplified by RTePCR as mentioned above using the primers MuV-VP1 (50 -GTTCTACGAGCATCGACACA-30 ) and MuV-VP2 (50 -AGCACCTTGTCCACCTTTTG-30 ). The RTePCR products were cloned into the pCR2.1TOPO vector according with the manufacturer instructions (Invitrogen). Cloned products were sequenced in both sense using M13þ and M13 primers (included in the pCR2.1-TOPO kit), MuV-VP1 and MuV-VP2 primers and the BigDye Terminator Cycle Sequencing Kit v2.0 in the ABI Prism 310 gene
analyzer (Applied Biosystems, Foster City, CA). Sequences were assembled with DNASIS Max for Windows (Hitachi Software Engineering Co. Ltd.) and compared with published V gene sequence (GenBank accession no. AF314559). 3. Results 3.1. Mumps virus replication in cell cultures Differences in viral replication in human SH-SY5Y nerve cells were confirmed by means of a fluorescent in situ hybridization (FISH) assay using a 680-bp cDNA probe towards M gene in the anti-genome sense; this probe identifies RNA particles produced during viral replication. HN-A1081 virus-infected cells showed broader fluorescence distribution and intensity than HN-G1081 virus-infected nerve cells (Fig. 1B,C). Because HeLa cells are highly susceptible to MuV infection, they were used as a model to compare the viral replication level in cells of not nervous origin. Performing of the FISH assays showed that HN-A1081 and HN-G1081 viruses replicate at similar levels in HeLa cells (Fig. 1E,F). These results mean that HN-G1081 was better replicated in HeLa cells than in nerve cells. 3.2. Mumps virus replication in interferon-primed cells To test their sensitivity to interferon, virus variants were inoculated in SH-SY5Y and HeLa cells, either unprimed or previously primed with IFN-a2b. HN-A1081 virus was replicated at 1 log higher than HN-G1081 (103.5 vs. 102.5 TCID50, respectively) in untreated nerve cells (Fig. 2A). Pretreatment of nerve cells with IFN reduced the replication of both virus
Fig. 1. Detection of viral replication by fluorescence in situ hybridization. SH-SY5Y nerve cells (AeC) and HeLa cervix cells (DeF) were infected with 0.02 PFU/ cell of HN-A1081 (B, E) and HN-G1081 (C, F) variants of Urabe AM9 mumps virus strain. Viral genomes were detected at 48 h p.i., using a biotin-conjugated anti-M gene probe followed by FITC-conjugated streptavidin. Note cytoplasmic fluorescence in infected cells. Uninfected HeLa cells showed nuclear unspecific fluorescence (D). Magnification 400.
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
867
Fig. 2. Effect of IFN-priming on virus replication. HN-A1081 and HN-G1081 variants of Urabe AM9 mumps virus were inoculated in SH-SY5Y (A) and HeLa cells (B) unprimed or primed with 2000 IU/ml of IFN-a2b 5 h before infection. Viral titers were determined at 48 h p.i.
variants in nerve cells, although HN-A1081 was less affected than HN-G1081 variant; indeed, HN-A1081 virus was still replicated 1.3 logs higher than HN-G1081 (102.6 vs. 101.3 TCID50, respectively). In unprimed HeLa cells, HN-A1081 virus showed a replication level similar to that in nerve cells, but it was just 0.75 log higher than HN-G1081 (103.5 versus 102.75, respectively); this confirmed that HN-G1081 was better replicated in HeLa cells than in nerve cells (Fig. 2B). IFN-a treatment reduced the replication of both virus variants in HeLa cells, but the 0.7 log difference in replication was maintained between HN-A1081 and HN-G1081 variants (103 and 102.3, respectively). Thus, IFN priming of HeLa cells just reduced 0.5 and 0.45 log the replication of HN-A1081 and HN-G1081 variants, respectively. 3.3. Transcription of viral genes In order to know the effect of IFN-a priming on viral gene transcription, cells were infected with HN-A1081 and HNG1081 virus and their transcriptional level was analyzed by semi-quantitative RTePCR. Although slightly higher transcription of HN-A1081 than HN-G1081 virus was observed, their differences did not reach more than 20% from each other for NP, P, V, M, F and SH genes in untreated nerve cells, but higher transcription of HN and L genes was found in HNA1081 variant (Fig. 3A). Priming of nerve cells with IFN-a reduced the transcription of all the genes of HN-G1081 virus, but the reduction was more notable (30e45%) in M, L, SH, and P genes (Fig. 3B). In contrast, IFN-a priming just reduced NP transcription and increased (40%) the V transcript of HNA1081 variant. Thus, HN-G1081 virus was more sensitive than HN-A1081 virus to IFN-a priming in nerve cells. HN-A1081 and HN-G1081 viruses showed almost similar transcription patterns for all the genes in untreated HeLa cells (Fig. 3C). Priming of HeLa cells with IFN-a reduced the F gene transcript and increased (41%) the V transcript of HNA1081 variant (Fig. 3C and D), with no important effect on the other genes. Contrarily to that which occurred in nerve cells, IFN priming of HeLa cells just reduced the SH transcript, but did not affect any other gene of HN-G1081 variant
(Fig. 3D). This confirmed that HN-G1081 virus was less sensitive to the effect of IFN-a priming in HeLa cells. 3.4. Transcription of IFN responsive genes The transcription of some IFN responsive genes was analyzed to identify if viral infection modifies the antiviral response. Fig. 4A shows that unprimed nerve cells presented basal levels of STAT1, STAT2, p48, PKR and OAS transcripts. Priming of nerve cells with IFN-a2b increased STAT1, p48, PKR and OAS and induced MxA transcription. Infection with HN-A1081 variant reduced the transcription of STAT1, STAT2, p48 and MxA genes. However, HN-G1081 virus did not significantly modify the transcription of IFN responsive genes except for the reduction of MxA in IFN-primed cells (Fig. 4B). HeLa cells presented basal levels of STAT1, STAT2, PKR and OAS transcripts, but not p48 and MxA (Fig. 4C). IFNpriming increased STAT1 and highly induced p48 and MxA transcription (Fig. 4D). Infection with both variants of mumps virus reduced the transcription of all the IFN responsive genes under study: however, their highest effect was on STAT2, p48 and MxA (Fig. 4C and D). HN-A1081 also considerably affects STAT1 transcription in HeLa cells. 3.5. Variants of Urabe AM9 mumps virus differ in the P/V gene edition site The complete open reading frame that codes for the V protein (nt 1906e2704 in genome sense) were cloned and sequenced. The nucleotide sequence of HN-A1081 virus V gene (Fig. 5A) was identical to that reported for Urabe AM9 mumps virus strain; while the V sequence from HN-G1081 virus presented three extra G inserted at the P/V edition site, nt 2468e2473 (Fig. 5B). These changes at nucleotide level theoretically imply the insertion of a glycine at position 156 in the V protein of HN-G1081 virus, while preserving the rest of the amino acid sequence (Fig. 5C). Therefore the predicted amino acid sequences of V proteins of both viral clones were identical to each other except for the inserted G156 in HN-G1081 virus.
868
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
Fig. 3. Viral gene transcription. HN-A1081 and HN-G1081 variants of Urabe AM9 mumps virus were inoculated into unprimed or IFN-primed SH-SY5Y (A, B) and HeLa cells (C, D). mRNA was purified at 48 h p.i., specific segments of NP (352 bp), P (355 bp), V (220 bp), M (680 bp), F (480 bp), SH (168 bp), HN (431 bp) and L (450 bp) viral genes were reverse transcribed and amplified by RTePCR and their products were analyzed by means of agarose gel electrophoresis and densitometry. Data represent the mean of three experiments. Representative electrophoretic gels are shown.
4. Discussion The basis of Urabe AM9 neurovirulence is an issue of intense research. We have found that two variants of Urabe
AM9 mumps virus differ in their replication efficiency in nerve cells [7]. It has been recently shown that genetic heterogeneities at coding and regulatory non-translated regions have profound implications in the neurovirulence of Urabe AM9
Fig. 4. Cellular gene transcription. HN-A1081 and HN-G1081 variants of Urabe AM9 mumps virus were inoculated in unprimed or IFN-primed SH-SY5Y (A, B) and HeLa cells (C, D).The mRNA was purified at 48 h p.i., treated with DNasa and fragments of the IFN responsive genes: STAT1 (713 bp), STAT2 (498 bp), p48 (529 bp), PKR (809 bp), OAS (265 bp), and MxA (729 bp) were amplified by RTePCR. Products were analyzed by means of agarose gel electrophoresis and densitometry. Data represent the mean of three experiments. Representative electrophoretic gels are shown.
870
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
Fig. 5. Sequencing of the V mRNA. The complete V open reading frame of both virus variants was cloned and sequenced. Partial electropherograms show the insertion of three G (arrows) at the P/V edition site of HN-G1081 variant (B) but not in HN-A1081 virus (A). An alignment of the V protein complete amino acid sequence is shown in C. The V protein of HN-A1081 virus was identical to that reported for Urabe AM9 mumps virus strain, but the V protein of HN-G1081 virus presented a Gly inserted at position 156 (asterisk).
virus [11]; this suggests that the replication efficiency of Urabe AM9 variants might be determined by mechanisms affecting the regulation of viral replication and transcription. Since the products of several IFN responsive genes are involved in controlling the permissibility for viral infection [12], we tested if mumps virus variants differ in their sensitivity to IFN. Pretreatment of nerve cells with IFN reduced the replication of both virus variants, but HN-G1081 was more affected; in addition, all the transcripts of HN-G1081 virus were reduced. Although observed reductions do not indicate a complete suppression of virus transcription, it is noteworthy that P and L genes were the most affected. Since P protein is the cofactor of the viral RNA-dependent RNA polymerase, L protein, its reduction might imply a down-regulation of viral replication [1]. In contrast, P and L transcripts were not affected in HNA1081 virus due to IFN treatment, but the transcript of V increased. Viral replication activates the transcription of IFN-ab genes, but induction of an efficient antiviral response depends on the presence of a functional IFN signaling pathway [12,13].
Several viruses including paramyxoviruses can attenuate the antiviral response in order to avoid their elimination [13,14]. The capacity of mumps virus to circumvent the IFN response mainly relies on the function of V protein, which promotes STAT1 degradation and disruption of the IFN signaling; this has been amply studied [12e19]. In our work, the increment of V transcript correlated with low transcription of STAT1, STAT2, p48 and MxA genes in cells infected with HN-A1081 virus. This suggests that HN-A1081 virus possesses either higher capacity to block the IFN mediated response [15,16] or lower capacity to activate IFN genes [17]. These results contribute to understanding in part the neurovirulence of Urabe AM9, because neurons are very responsive to IFN with a great STAT1-dependent transcriptional response and an innate potential to defend against viral infection. Indeed, STAT1, PKR, OAS and MxA are involved in host protection against injury or loss of neurons [20,21]. The viral inhibitory effect on the transcription of IFN responsive genes might be a consequence of an interaction of V protein with ISG promoters; as occurs with mda-5, an
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
IFN-inducible RNA helicase, which is bound by the V protein inhibiting the activation of the IFN-b promoter and thereby blocking the antiviral response [22]. We have found that the replication efficiency of mumps virus variants depends on the cell type. In HeLa cells, HN-G1081 virus was replicated at almost similar levels than HN-A1081 virus, contrarily to what occurred in nerve cells. In addition, IFN priming of HeLa cells did not notably affect HN-G1081 virus replication or transcription. This was a probable consequence of IFN defects that have been observed in HeLa cells; indeed, lack of expression of MxA protein, which participates in the control of paramyxovirus transcription [23,24]. HN-G1081 virus has been shown to efficiently replicate in cells of non-nervous origin such as Vero cells [5,7,25], which also possess defects in their IFN b response [24]. This suggests that HN-G1081 presented higher viral yields in cells that have an altered IFN response. In respect of this, it is surprising that transcription of STAT1 was notably reduced in HeLa cells infected with HNA1081 virus, but not in cells infected with HN-G1081; however, both virus variants replicate at similar levels as was confirmed in the hybridization assay (Fig. 1). We thought that the lack of p48 transcription in infected HeLa cells contributes to produce similar viral yields, since p48 is required to establish the transcriptional activation complex that induces expression of IFN responsive genes required to control viral replication [12]. In order to identify differences at the nucleotide level, the V mRNA of both virus variants was sequenced. HN-G1081 virus differs from HN-A1081 because the first possesses nine G in the P/V gene edition site while the latter presented just six G. It is broadly known that MuV P mRNA is co-transcriptionally modified by the addition of two G nucleotides between positions 2468 and 2473; while I protein surges by addition of four G nucleotides in the same region. However, the MuV V protein is codified by a non-edited mRNA [1,18]. In the present work we obtained an edited V mRNA that includes three extra G in the edition site, which produce the insertion of Gly156 in the HN-G1081V protein. To our knowledge, this is the first time that an edited V mRNA has been found in rubulaviruses. It is difficult to propose a hypothesis about the putative effect of the inserted Gly156 on the V protein function. The rubulavirus V proteins possess a cysteine-rich motif located in the carboxyl end that binds two atoms of Zn2þ, adopting the typical zinc finger conformation that is involved in proteineprotein interactions [25e28]. The site of specific interaction with STAT1 in the HuPIV-2 V protein has been mapped downstream of the zinc finger motif; this is a tryptophan-rich motif (Trp-aa3Trp-aa9-Trp) [18], which corresponds to residues 170e224 in the MuV V sequence. Therefore Gly156 does not constitute part of any of these functional motifs. Published sequences of P/V gene do not show any difference between Urabe AM9 mumps virus isolates [9], because they sequenced the viral genome directly obtained from vaccine samples; in contrast we purified, cloned and sequenced viral mRNA. Thus our results suggest that observed nucleotide insertions originated during the co-transcriptional edition process. Thus, the effect of this amino acid insertion in either the function of V protein or in the infectious process should be determined experimentally.
871
5. Conclusions Our results showed that the replication efficiency of Urabe AM9 mumps virus variants in human nerve cells is influenced by their capacity to circumvent the antiviral defense. Acknowledgments This work was supported by grants from CONACYT (Salud 2003-C01-085) and FOFOI, IMSS (IMSS-2003/028; IMSS-2004/091) Mexico. References [1] K.M. Carbone, J.S. Wolinsky, Mumps virus, in: D. Knipe, P. Howley (Eds.), Fields Virology, Lippincott, Williams and Wilkins, Philadelphia, 2001, pp. 1381e1400. [2] S.A.M. Galazka, S.E. Robertson, A. Kraigher, Mumps and mumps vaccine: a global review, Bull. WHO 77 (1999) 3e14. [3] J. Dourado, S. Cunha, M.G. Teixeira, C.P. Farrington, A. Melo, R. Lucena, M.L. Barreto, Outbreak of aseptic meningitis associated with mass vaccination with a Urabe-containing measles-mumps-rubella vaccine: implications for immunization programs, Am. J. Epidemiol. 151 (2000) 524e530. [4] R. Dobson, Mumps cases on the rise in England and Wales, BMJ 330 (2005) 324. [5] E.G. Brown, K. Dimock, K.E. Wright, The Urabe AM9 mumps vaccine is a mixture of viruses differing at amino acid 335 of the hemagglutininneuraminidase gene with one form associated with disease, J. Infect. Dis. 174 (1996) 619e622. [6] M.G. Cusi, L. Santini, S. Bianchi, M. Valassina, P.E. Valensin, Nucleotide sequence at position 1081 of the hemagglutinin-neuraminidase gene in wild-type strains of mumps virus is the most relevant marker of virulence, J. Clin. Microbiol. 36 (1998) 3743e3744. [7] G. Santos-Lo´pez, C. Cruz, N. Pazos, V. Vallejo, J. Reyes-Leyva, J. TapiaRamirez, Two clones obtained from Urabe AM9 mumps virus vaccine differ in their replicative efficiency in neuroblastoma cells, Microbes Infect. 332 (2006) 332e339. [8] J. Reyes-Leyva, R. Ba~nos, M. Borraz-Argu¨ello, G. Santos-Lopez, N. Rosas, G. Alvarado, I. Herrera, V. Vallejo, J. Tapia-Ramirez, Amino acid change 335 E to K affects the sialic-acid-binding and neuraminidase activities of Urabe AM9 mumps virus hemagglutinin-neuraminidase glycoprotein, Microb. Infect. 9 (2007) 234e240. [9] G. Amexis, N. Fineschi, K. Chumakov, Correlation of genetic variability with safety of mumps vaccine Urabe AM9 strain, Virology 287 (2001) 234e241. [10] S.A. Rubin, G. Amexis, M. Pletnikov, Z. Li, J. Vanderzanden, J. Mauldin, C. Sauder, T. Malik, K. Chumakov, K.M. Carbone, Changes in mumps virus gene sequence associated with variability in neurovirulent phenotype, J. Virol. 77 (2003) 11616e11624. [11] C.J. Sauder, K.M. Vandenburgh, R.C. Iskow, T. Malik, K.M. Carbone, S.A. Rubin, Changes in mumps virus neurovirulence phenotype associated with quasispecies heterogeneity, Virology 350 (2006) 48e57. [12] S. Goodbourn, L. Didcock, R.E. Randall, Interferons: cell signaling, immune modulation, antiviral response and virus countermeasures, J. Gen. Virol. 81 (2000) 2341e2364. [13] N. Fujii, N. Yokosawa, S. Shirakawa, Suppression of interferon response gene expression in cells persistently infected with mumps virus, and restoration from its suppression by treatment with ribavirin, Virus Res. 65 (1999) 175e185. [14] B. Gotoh, T. Komatsu, K. Takeuchi, J. Yokoo, Paramyxovirus strategies for evading the interferon response, Rev. Med. Virol. 2 (2002) 337e357. [15] T. Kubota, N. Yokosawa, S. Yokota, N. Fujii, C-terminal cys-rich region of mumps virus structural V protein correlates with block of interferon
872
[16]
[17]
[18]
[19]
[20]
[21]
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872 alpha and gamma signal transduction pathway through decrease STAT1alpha, Biochem Biophys Res Commun. 283 (2001) 255e259. T. Kubota, N. Yokosawa, S. Yokota, N. Fujii, M. Tashiro, A. Kato, Mumps virus V protein antagonizes interferon without the complete degradation of STAT1, J. Virol. 79 (2005) 4451e4459. L. Marie, J.E. Durbin, D.E. Levy, Differential viral induction of distinct interferon-alpha genes by positive feedback through interferon regulatory factor-7, EMBO J. 17 (1998) 6660e6669. M. Nishio, D. Garcin, V. Simonet, D. Kolakofsky, The carboxyl segment of the mumps virus V protein associates with STAT proteins in vitro via a tryptophan-rich motif, Virology 300 (2002) 92e99. N. Yokosawa, S. Yokota, T. Kubota, N. Fujii, C-terminal region of STAT1a is not necessary for its ubiquitination and degradation caused by mumps virus v protein, J. Virol. 76 (2002) 12683e12690. J. Wang, R.D. Schreiber, I.L. Campbell, STAT1 deficiency unexpectedly and markedly exacerbates the pathophysiological actions of IFN-a in the central nervous system, Proc. Natl. Acad. Sci. 99 (2002) 16209e16214. J. Wang, I.L. Campbell, Innate STAT1-Dependent genomic response of neurons to the antiviral cytokine alpha interferon, J. Virol. 79 (2005) 8295e8302.
[22] J. Andrejeva, K.S. Childs, D.F. Young, T.S. Carlos, N. Stock, S. Goodbourn, R.E. Randall, The V proteins of paramyxoviruses bind the IFN-inducible RNA helicase, mda-5, and inhibit its activation of the IFN-b promoter, PNAS 101 (2004) 4917264e4917269. [23] P. Staeheli, Interferon-induced proteins and the antiviral state, Adv. Virus Res. 38 (1990) 147e200. [24] J.D. Mosca, P.M. Pitha, Transcriptional and posttranscriptional regulation of exogenous human beta interferon gene in simian cells defective in interferon synthesis, Mol. Cell. Biol. 6 (1986) 2279e2283. [25] K.E. Wright, K. Dimock, E.G. Brown, Biological characteristics of genetic variants of Urabe AM9 mumps vaccine virus, Virus Res. 67 (2000) 49e57. [26] C.M. Horvath, Weapons of STAT destruction, Eur. J. Biochem. 271 (2004) 4621e4628. [27] A. Kato, K. Kiyotani, Y. Sakai, T. Yoshida, T. Shioda, Y. Nagai, Importance of the cysteine-rich carboxyl-terminal half of V protein for Sendai virus pathogenesis, J. Virol. 71 (1997) 7266e7272. [28] C. Huang, K. Kiyotani, Y. Fujii, N. Fukuhara, A. Kato, Y. Nagai, T. Yoshida, T. Sakaguchi, Involvement of the zinc-binding capacity of Sendai virus V protein in viral pathogenesis, J. Virol. 74 (2000) 7834e7841.
Original article
Differential sensitivity to interferon influences the replication and transcription of Urabe AM9 mumps virus variants in nerve cells Nora Rosas-Murrieta a,b, Irma Herrera-Camacho b, Vero´nica Vallejo-Ruiz a, Lourdes Milla´n-Pe´rez-Pe~ na b, Carlos Cruz c, Jose´ Tapia-Ramı´rez c, Gerardo Santos-Lo´pez a, Julio Reyes-Leyva a,* a
Lab. de Virologı´a y Biologı´a Molecular, Centro de Investigacio´n Biome´dica de Oriente, Instituto Mexicano del Seguro Social, Hospital General de Zona No. 5, Km. 4.5 Carretera Atlixco-Metepec, 74360 Metepec, Pue., Mexico b Centro de Quı´mica, Instituto de Ciencias, Beneme´rita Universidad Auto´noma de Puebla, Complejo de Ciencias, Ciudad Universitaria, 72570 Puebla, Pue., Mexico c Dept. de Gene´tica y Biologı´a Molecular, Centro de Investigacio´n y Estudios Avanzados, Instituto Polite´cnico Nacional, Av. Instituto Polite´cnico Nacional 2508, Col. San Pedro Zacatenco, 07360 Me´xico D.F., Mexico Received 19 November 2006; accepted 8 March 2007 Available online 15 March 2007
Abstract Urabe AM9 mumps virus vaccine causes post-vaccination meningitis. Two variants of Urabe AM9 virus differ in their replication efficiency in human nerve cells, HN-A1081 variant being more neurotropic than HN-G1081. The effect of interferon (IFN) on viral replication and transcription was analyzed. Priming of nerve cells with IFN reduced more significantly the replication of HN-G1081 variant (from 102.5 to 101.3 TCID50) than that of HN-A1081 (from 103.5 to 102.6 TCID50). IFN-priming also reduced the transcription of HN-G1081 genes, but not of HN-A1081. The effect of viral infection on the transcription of cellular IFN responsive genes was analyzed. HN-A1081 virus reduced the transcription of STAT1, STAT2, p48 and MxA in both unprimed and IFN-primed cells; whereas HN-G1081 virus just reduced MxA transcription. Since rubulavirus V protein inhibits IFN signaling, the V mRNA was cloned and sequenced, finding that HN-G1081 but not HN-A1081 presented three extra G in the P/V edition site, producing the insertion of Gly156 in the V protein. Our results suggest that the replication efficiency of Urabe AM9 mumps virus variants is influenced by their sensitivity to interferon and their capacity to reduce the antiviral response. Ó 2007 Elsevier Masson SAS. All rights reserved. Keywords: Mumps vaccine; Interferon inhibition; Transcription; Urabe AM9; Neurovirulence
1. Introduction Mumps virus (MuV) causes an acute febrile infection that frequently affects the central nervous system, producing aseptic meningitis and sudden deafness [1]. Massive vaccination programs have decreased the incidence of MuV infection worldwide. However, reports indicate that w1/1000e10,000 recipients of Urabe AM9 and other vaccine strains developed aseptic meningitis [2,3]. The unacceptably high rate of vaccine
* Corresponding author. Tel./fax: þ52 24 44 44 01 22. E-mail addresses: [email protected], [email protected] (J. Reyes-Leyva). 1286-4579/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.micinf.2007.03.005
associated meningitis and parotitis cases has resulted in vaccine withdrawal and public resistance to mumps vaccination [2]. In consequence, mumps epidemics have re-emerged, and the incidence is rising in several countries [4]. Therefore, it is important to identify the basis of MuV neurovirulence in order to improve the safety and efficiency of mumps virus vaccines. MuV belongs to Rubulavirus genus of the Paramyxoviridae family; it has a single-stranded negative-sense RNA genome of 15,384 nucleotides that encodes seven structural proteins: nucleoprotein (NP), phosphoprotein (P), matrix (M), fusion (F), small hydrophobic (SH), hemagglutinin-neuraminidase (HN) and RNA polymerase, the so-called large (L) protein. In addition, the P gene is co-transcriptionally edited giving
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
origin to P, I and V proteins; the last regulates viral replication and inhibits antiviral response [1]. Previous reports suggested that a nucleotide change (G/A) at position 1081 in the HN gene of Urabe AM9 mumps virus was causally associated with presentation of post-vaccination meningitis cases, because HN-A1081 viruses were isolated during aseptic meningitis at higher frequency than HN-G1081 viruses [5]. Partial characterization of the HN gene amongst several vaccine and wild-type mumps virus strains led to the identification of an association of HN-A1081 genotype with neurovirulence [5,6]. We have shown that two variants of Urabe AM9 virus that were initially characterized by their difference in the HN gene nt 1081 also differ in their replication efficiency in nerve cells. Indeed, HN-A1081 variant was efficiently replicated in both human neuroblastoma cells and newborn rat brain (105 and 104 PFU, respectively), whereas HN-G1081 variant was replicated at low titers (102 PFU in both cases) [7]. The neurotropism of these variants was supported in part by their differences in cell receptor-binding affinity; i.e., HN-A1081 virus showed highest affinity towards a2-6 linked sialic acids that are highly expressed in human nerve cells; whereas HN-G1081 variant preferred a2-3 linked sialic acids that are scarcely presented in nerve cells. However, HN-G1081 variant also recognized a2-6 linked sialic acids but with lesser affinity than HN-A1081 virus [8]. The role of HN-A1081 mutation in the neurovirulence of mumps virus has been questioned because complete genome sequencing has shown that Urabe AM9 based vaccines are composed of more than two viral quasispecies presenting genetic heterogeneity at several sites [9]. In addition, adaptation of the non-neurotropic Jeryl Lynn virus strain to nerve cells increased its neurovirulence in the newborn rat model and involved changes in NP, M and L genes [10]. Moreover, it was recently found that the loss of genetic heterogeneity had a profound participation in the attenuation of Urabe AM9 virus; this included changes in coding nucleotides in NP, F, HN and L genes as well as in non-coding regions [11]. These results indicate that other viral and cellular factors are involved in determining the permissibility for mumps virus replication in nerve cells. Viral infection is controlled at cellular level by type I interferon (IFN a/b), which are produced in response to the presence of viral double stranded RNA (dsRNA). Recognition of IFN a/b by their specific cell receptors activates the signaling pathway JAK-STAT, which includes the signal transducers and activators of transcription STAT1 and STAT2. Upon phosphorylation, STAT1 and STA2 form heterodimers which translocate to the nucleus, where they associate with the DNAbinding protein p48 (IRF9) to induce the activation of several IFN responsive genes, including: dsRNA-dependent protein kinase R (PKR), 20 ,50 oligoadenylate synthetase (OAS) and MxA, leading together to inhibition of viral replication, transcription and protein expression [12]. In the present paper we analyzed whether cell priming with IFN modifies the cell permissibility for mumps virus variants; and inversely whether infection with either HN-A1081 or HNG1081 virus affects the transcription of IFN responsive genes.
865
2. Material and methods 2.1. Viruses and cells HN-A1081 and HN-G1081 clones of Urabe AM9 mumps virus vaccine were purified by plaque assays as reported previously [7]. Viruses were replicated in human neuroblastoma SH-SY5Y cells and uterine cervix carcinoma HeLa cells maintained in Dulbecco’s minimum essential medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 mg/ml streptomycin and 1% nonessential amino acids. 2.2. Fluorescence in situ hybridization SH-SY5Y and HeLa cells (104/well) cultured on chamber slides (Corning; Acton, MA) were infected with each virus variants (0.02 PFU/cell). At 48 h post infection (p.i.), cells were washed and fixed in 4% paraformaldehyde in PBS. Cells were consecutively treated with permeation solution (10 mM TriseHCl pH 7.4, 0.1% Triton X-100) for 10 min, then with pre-hybridization solution (2 SSC, 10 Denhardt’s solution, 0.1% SDS, 250 mg/ml salmon sperm DNA) 1 h, and with hybridization solution (50% deionized formamide, 2SSC, 10 Denhardt’s solution, 5% dextran sulfate, 0.2% BSA, 0.1% SDS) containing 50 ng/ml of an M gene anti-sense probe conjugated to psoralen-biotin (Ambion Inc., Austin, TX) during 18 h. After that, cells were washed with 1% SSC, incubated with FITC-conjugated streptavidin (SigmaeAldrich Chemicals, St. Louis, MO), further washed and examined under an epifluorescence microscope (Nikon Inc., Mel, NY). Images were processed and analyzed with MetaMorph Software (Universal Imaging Co., Downingtown, PA). 2.3. Mumps virus infection on interferon-primed cells SH-SY5Y and HeLa cells were seeded in 35 mm culture plates with DMEM 10% FBS and incubated at 37 C during 24 h. Cells were then treated with 2000 IU/ml human recombinant IFN-a2b (Probiomed, Mexico City) in DMEM without serum during 5 h. After that, unprimed and IFN-primed cells were inoculated with HN-A1081 and HN-G1081 virus (0.02 PFU/cell). Non-infected cells, both primed and unprimed, were used as controls. 2.4. Virus titration The supernatants of infected HeLa and SH-SY5Y cells were collected at 48 h p.i. and used to determine viral yields on Vero cells cultured on 24-wells microassay plates. Viral titers were expressed as 50% tissue culture infective doses (TCID50), which were calculated according to conventional methods. 2.5. Transcription of viral and cellular genes Mock and IFN primed HeLa and SH-SY5Y cells seeded on 35 mm plates were infected with each virus variant. Total
866
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
RNA was extracted at 48 h p.i. using Trizol reagent (Invitrogen, Carlsbad, CA). mRNA was purified with Oligotex kit (Qiagen, Germany) and treated with DNase (SigmaeAldrich), cDNA was synthesized and amplified by RTePCR using SuperScriptÔII One-Step RTePCR System with Platinum Taq (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. Design of Urabe AM9 mumps virus specific primers was based on published genome sequence (GenBank accession no. AF314559) using the Primer3 software Release 1.0.1 (http://www.sk.embnet.org/cgi-bin/primer3_www.cgi). Primer sequences for viral genes are available under request. Primers for cellular genes were obtained from previous reports [13]. The house-keeping gene cyclophilin (CP) was used as an internal control. Amplified products were analyzed by electrophoresis on a 2% agarose gel. Images were captured in a Kodak EDAS 290 LE system (Eastman Kodak Co., Rochester, NY) and analyzed with Quantity-One software (Bio-Rad, Hercules, CA). Results were reported as optical density gene/CP ratios. 2.6. Cloning and sequencing of V gene The V mRNA of both HN-A1081 and HN-G1081 virus variants was amplified by RTePCR as mentioned above using the primers MuV-VP1 (50 -GTTCTACGAGCATCGACACA-30 ) and MuV-VP2 (50 -AGCACCTTGTCCACCTTTTG-30 ). The RTePCR products were cloned into the pCR2.1TOPO vector according with the manufacturer instructions (Invitrogen). Cloned products were sequenced in both sense using M13þ and M13 primers (included in the pCR2.1-TOPO kit), MuV-VP1 and MuV-VP2 primers and the BigDye Terminator Cycle Sequencing Kit v2.0 in the ABI Prism 310 gene
analyzer (Applied Biosystems, Foster City, CA). Sequences were assembled with DNASIS Max for Windows (Hitachi Software Engineering Co. Ltd.) and compared with published V gene sequence (GenBank accession no. AF314559). 3. Results 3.1. Mumps virus replication in cell cultures Differences in viral replication in human SH-SY5Y nerve cells were confirmed by means of a fluorescent in situ hybridization (FISH) assay using a 680-bp cDNA probe towards M gene in the anti-genome sense; this probe identifies RNA particles produced during viral replication. HN-A1081 virus-infected cells showed broader fluorescence distribution and intensity than HN-G1081 virus-infected nerve cells (Fig. 1B,C). Because HeLa cells are highly susceptible to MuV infection, they were used as a model to compare the viral replication level in cells of not nervous origin. Performing of the FISH assays showed that HN-A1081 and HN-G1081 viruses replicate at similar levels in HeLa cells (Fig. 1E,F). These results mean that HN-G1081 was better replicated in HeLa cells than in nerve cells. 3.2. Mumps virus replication in interferon-primed cells To test their sensitivity to interferon, virus variants were inoculated in SH-SY5Y and HeLa cells, either unprimed or previously primed with IFN-a2b. HN-A1081 virus was replicated at 1 log higher than HN-G1081 (103.5 vs. 102.5 TCID50, respectively) in untreated nerve cells (Fig. 2A). Pretreatment of nerve cells with IFN reduced the replication of both virus
Fig. 1. Detection of viral replication by fluorescence in situ hybridization. SH-SY5Y nerve cells (AeC) and HeLa cervix cells (DeF) were infected with 0.02 PFU/ cell of HN-A1081 (B, E) and HN-G1081 (C, F) variants of Urabe AM9 mumps virus strain. Viral genomes were detected at 48 h p.i., using a biotin-conjugated anti-M gene probe followed by FITC-conjugated streptavidin. Note cytoplasmic fluorescence in infected cells. Uninfected HeLa cells showed nuclear unspecific fluorescence (D). Magnification 400.
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
867
Fig. 2. Effect of IFN-priming on virus replication. HN-A1081 and HN-G1081 variants of Urabe AM9 mumps virus were inoculated in SH-SY5Y (A) and HeLa cells (B) unprimed or primed with 2000 IU/ml of IFN-a2b 5 h before infection. Viral titers were determined at 48 h p.i.
variants in nerve cells, although HN-A1081 was less affected than HN-G1081 variant; indeed, HN-A1081 virus was still replicated 1.3 logs higher than HN-G1081 (102.6 vs. 101.3 TCID50, respectively). In unprimed HeLa cells, HN-A1081 virus showed a replication level similar to that in nerve cells, but it was just 0.75 log higher than HN-G1081 (103.5 versus 102.75, respectively); this confirmed that HN-G1081 was better replicated in HeLa cells than in nerve cells (Fig. 2B). IFN-a treatment reduced the replication of both virus variants in HeLa cells, but the 0.7 log difference in replication was maintained between HN-A1081 and HN-G1081 variants (103 and 102.3, respectively). Thus, IFN priming of HeLa cells just reduced 0.5 and 0.45 log the replication of HN-A1081 and HN-G1081 variants, respectively. 3.3. Transcription of viral genes In order to know the effect of IFN-a priming on viral gene transcription, cells were infected with HN-A1081 and HNG1081 virus and their transcriptional level was analyzed by semi-quantitative RTePCR. Although slightly higher transcription of HN-A1081 than HN-G1081 virus was observed, their differences did not reach more than 20% from each other for NP, P, V, M, F and SH genes in untreated nerve cells, but higher transcription of HN and L genes was found in HNA1081 variant (Fig. 3A). Priming of nerve cells with IFN-a reduced the transcription of all the genes of HN-G1081 virus, but the reduction was more notable (30e45%) in M, L, SH, and P genes (Fig. 3B). In contrast, IFN-a priming just reduced NP transcription and increased (40%) the V transcript of HNA1081 variant. Thus, HN-G1081 virus was more sensitive than HN-A1081 virus to IFN-a priming in nerve cells. HN-A1081 and HN-G1081 viruses showed almost similar transcription patterns for all the genes in untreated HeLa cells (Fig. 3C). Priming of HeLa cells with IFN-a reduced the F gene transcript and increased (41%) the V transcript of HNA1081 variant (Fig. 3C and D), with no important effect on the other genes. Contrarily to that which occurred in nerve cells, IFN priming of HeLa cells just reduced the SH transcript, but did not affect any other gene of HN-G1081 variant
(Fig. 3D). This confirmed that HN-G1081 virus was less sensitive to the effect of IFN-a priming in HeLa cells. 3.4. Transcription of IFN responsive genes The transcription of some IFN responsive genes was analyzed to identify if viral infection modifies the antiviral response. Fig. 4A shows that unprimed nerve cells presented basal levels of STAT1, STAT2, p48, PKR and OAS transcripts. Priming of nerve cells with IFN-a2b increased STAT1, p48, PKR and OAS and induced MxA transcription. Infection with HN-A1081 variant reduced the transcription of STAT1, STAT2, p48 and MxA genes. However, HN-G1081 virus did not significantly modify the transcription of IFN responsive genes except for the reduction of MxA in IFN-primed cells (Fig. 4B). HeLa cells presented basal levels of STAT1, STAT2, PKR and OAS transcripts, but not p48 and MxA (Fig. 4C). IFNpriming increased STAT1 and highly induced p48 and MxA transcription (Fig. 4D). Infection with both variants of mumps virus reduced the transcription of all the IFN responsive genes under study: however, their highest effect was on STAT2, p48 and MxA (Fig. 4C and D). HN-A1081 also considerably affects STAT1 transcription in HeLa cells. 3.5. Variants of Urabe AM9 mumps virus differ in the P/V gene edition site The complete open reading frame that codes for the V protein (nt 1906e2704 in genome sense) were cloned and sequenced. The nucleotide sequence of HN-A1081 virus V gene (Fig. 5A) was identical to that reported for Urabe AM9 mumps virus strain; while the V sequence from HN-G1081 virus presented three extra G inserted at the P/V edition site, nt 2468e2473 (Fig. 5B). These changes at nucleotide level theoretically imply the insertion of a glycine at position 156 in the V protein of HN-G1081 virus, while preserving the rest of the amino acid sequence (Fig. 5C). Therefore the predicted amino acid sequences of V proteins of both viral clones were identical to each other except for the inserted G156 in HN-G1081 virus.
868
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
Fig. 3. Viral gene transcription. HN-A1081 and HN-G1081 variants of Urabe AM9 mumps virus were inoculated into unprimed or IFN-primed SH-SY5Y (A, B) and HeLa cells (C, D). mRNA was purified at 48 h p.i., specific segments of NP (352 bp), P (355 bp), V (220 bp), M (680 bp), F (480 bp), SH (168 bp), HN (431 bp) and L (450 bp) viral genes were reverse transcribed and amplified by RTePCR and their products were analyzed by means of agarose gel electrophoresis and densitometry. Data represent the mean of three experiments. Representative electrophoretic gels are shown.
4. Discussion The basis of Urabe AM9 neurovirulence is an issue of intense research. We have found that two variants of Urabe
AM9 mumps virus differ in their replication efficiency in nerve cells [7]. It has been recently shown that genetic heterogeneities at coding and regulatory non-translated regions have profound implications in the neurovirulence of Urabe AM9
Fig. 4. Cellular gene transcription. HN-A1081 and HN-G1081 variants of Urabe AM9 mumps virus were inoculated in unprimed or IFN-primed SH-SY5Y (A, B) and HeLa cells (C, D).The mRNA was purified at 48 h p.i., treated with DNasa and fragments of the IFN responsive genes: STAT1 (713 bp), STAT2 (498 bp), p48 (529 bp), PKR (809 bp), OAS (265 bp), and MxA (729 bp) were amplified by RTePCR. Products were analyzed by means of agarose gel electrophoresis and densitometry. Data represent the mean of three experiments. Representative electrophoretic gels are shown.
870
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
Fig. 5. Sequencing of the V mRNA. The complete V open reading frame of both virus variants was cloned and sequenced. Partial electropherograms show the insertion of three G (arrows) at the P/V edition site of HN-G1081 variant (B) but not in HN-A1081 virus (A). An alignment of the V protein complete amino acid sequence is shown in C. The V protein of HN-A1081 virus was identical to that reported for Urabe AM9 mumps virus strain, but the V protein of HN-G1081 virus presented a Gly inserted at position 156 (asterisk).
virus [11]; this suggests that the replication efficiency of Urabe AM9 variants might be determined by mechanisms affecting the regulation of viral replication and transcription. Since the products of several IFN responsive genes are involved in controlling the permissibility for viral infection [12], we tested if mumps virus variants differ in their sensitivity to IFN. Pretreatment of nerve cells with IFN reduced the replication of both virus variants, but HN-G1081 was more affected; in addition, all the transcripts of HN-G1081 virus were reduced. Although observed reductions do not indicate a complete suppression of virus transcription, it is noteworthy that P and L genes were the most affected. Since P protein is the cofactor of the viral RNA-dependent RNA polymerase, L protein, its reduction might imply a down-regulation of viral replication [1]. In contrast, P and L transcripts were not affected in HNA1081 virus due to IFN treatment, but the transcript of V increased. Viral replication activates the transcription of IFN-ab genes, but induction of an efficient antiviral response depends on the presence of a functional IFN signaling pathway [12,13].
Several viruses including paramyxoviruses can attenuate the antiviral response in order to avoid their elimination [13,14]. The capacity of mumps virus to circumvent the IFN response mainly relies on the function of V protein, which promotes STAT1 degradation and disruption of the IFN signaling; this has been amply studied [12e19]. In our work, the increment of V transcript correlated with low transcription of STAT1, STAT2, p48 and MxA genes in cells infected with HN-A1081 virus. This suggests that HN-A1081 virus possesses either higher capacity to block the IFN mediated response [15,16] or lower capacity to activate IFN genes [17]. These results contribute to understanding in part the neurovirulence of Urabe AM9, because neurons are very responsive to IFN with a great STAT1-dependent transcriptional response and an innate potential to defend against viral infection. Indeed, STAT1, PKR, OAS and MxA are involved in host protection against injury or loss of neurons [20,21]. The viral inhibitory effect on the transcription of IFN responsive genes might be a consequence of an interaction of V protein with ISG promoters; as occurs with mda-5, an
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872
IFN-inducible RNA helicase, which is bound by the V protein inhibiting the activation of the IFN-b promoter and thereby blocking the antiviral response [22]. We have found that the replication efficiency of mumps virus variants depends on the cell type. In HeLa cells, HN-G1081 virus was replicated at almost similar levels than HN-A1081 virus, contrarily to what occurred in nerve cells. In addition, IFN priming of HeLa cells did not notably affect HN-G1081 virus replication or transcription. This was a probable consequence of IFN defects that have been observed in HeLa cells; indeed, lack of expression of MxA protein, which participates in the control of paramyxovirus transcription [23,24]. HN-G1081 virus has been shown to efficiently replicate in cells of non-nervous origin such as Vero cells [5,7,25], which also possess defects in their IFN b response [24]. This suggests that HN-G1081 presented higher viral yields in cells that have an altered IFN response. In respect of this, it is surprising that transcription of STAT1 was notably reduced in HeLa cells infected with HNA1081 virus, but not in cells infected with HN-G1081; however, both virus variants replicate at similar levels as was confirmed in the hybridization assay (Fig. 1). We thought that the lack of p48 transcription in infected HeLa cells contributes to produce similar viral yields, since p48 is required to establish the transcriptional activation complex that induces expression of IFN responsive genes required to control viral replication [12]. In order to identify differences at the nucleotide level, the V mRNA of both virus variants was sequenced. HN-G1081 virus differs from HN-A1081 because the first possesses nine G in the P/V gene edition site while the latter presented just six G. It is broadly known that MuV P mRNA is co-transcriptionally modified by the addition of two G nucleotides between positions 2468 and 2473; while I protein surges by addition of four G nucleotides in the same region. However, the MuV V protein is codified by a non-edited mRNA [1,18]. In the present work we obtained an edited V mRNA that includes three extra G in the edition site, which produce the insertion of Gly156 in the HN-G1081V protein. To our knowledge, this is the first time that an edited V mRNA has been found in rubulaviruses. It is difficult to propose a hypothesis about the putative effect of the inserted Gly156 on the V protein function. The rubulavirus V proteins possess a cysteine-rich motif located in the carboxyl end that binds two atoms of Zn2þ, adopting the typical zinc finger conformation that is involved in proteineprotein interactions [25e28]. The site of specific interaction with STAT1 in the HuPIV-2 V protein has been mapped downstream of the zinc finger motif; this is a tryptophan-rich motif (Trp-aa3Trp-aa9-Trp) [18], which corresponds to residues 170e224 in the MuV V sequence. Therefore Gly156 does not constitute part of any of these functional motifs. Published sequences of P/V gene do not show any difference between Urabe AM9 mumps virus isolates [9], because they sequenced the viral genome directly obtained from vaccine samples; in contrast we purified, cloned and sequenced viral mRNA. Thus our results suggest that observed nucleotide insertions originated during the co-transcriptional edition process. Thus, the effect of this amino acid insertion in either the function of V protein or in the infectious process should be determined experimentally.
871
5. Conclusions Our results showed that the replication efficiency of Urabe AM9 mumps virus variants in human nerve cells is influenced by their capacity to circumvent the antiviral defense. Acknowledgments This work was supported by grants from CONACYT (Salud 2003-C01-085) and FOFOI, IMSS (IMSS-2003/028; IMSS-2004/091) Mexico. References [1] K.M. Carbone, J.S. Wolinsky, Mumps virus, in: D. Knipe, P. Howley (Eds.), Fields Virology, Lippincott, Williams and Wilkins, Philadelphia, 2001, pp. 1381e1400. [2] S.A.M. Galazka, S.E. Robertson, A. Kraigher, Mumps and mumps vaccine: a global review, Bull. WHO 77 (1999) 3e14. [3] J. Dourado, S. Cunha, M.G. Teixeira, C.P. Farrington, A. Melo, R. Lucena, M.L. Barreto, Outbreak of aseptic meningitis associated with mass vaccination with a Urabe-containing measles-mumps-rubella vaccine: implications for immunization programs, Am. J. Epidemiol. 151 (2000) 524e530. [4] R. Dobson, Mumps cases on the rise in England and Wales, BMJ 330 (2005) 324. [5] E.G. Brown, K. Dimock, K.E. Wright, The Urabe AM9 mumps vaccine is a mixture of viruses differing at amino acid 335 of the hemagglutininneuraminidase gene with one form associated with disease, J. Infect. Dis. 174 (1996) 619e622. [6] M.G. Cusi, L. Santini, S. Bianchi, M. Valassina, P.E. Valensin, Nucleotide sequence at position 1081 of the hemagglutinin-neuraminidase gene in wild-type strains of mumps virus is the most relevant marker of virulence, J. Clin. Microbiol. 36 (1998) 3743e3744. [7] G. Santos-Lo´pez, C. Cruz, N. Pazos, V. Vallejo, J. Reyes-Leyva, J. TapiaRamirez, Two clones obtained from Urabe AM9 mumps virus vaccine differ in their replicative efficiency in neuroblastoma cells, Microbes Infect. 332 (2006) 332e339. [8] J. Reyes-Leyva, R. Ba~nos, M. Borraz-Argu¨ello, G. Santos-Lopez, N. Rosas, G. Alvarado, I. Herrera, V. Vallejo, J. Tapia-Ramirez, Amino acid change 335 E to K affects the sialic-acid-binding and neuraminidase activities of Urabe AM9 mumps virus hemagglutinin-neuraminidase glycoprotein, Microb. Infect. 9 (2007) 234e240. [9] G. Amexis, N. Fineschi, K. Chumakov, Correlation of genetic variability with safety of mumps vaccine Urabe AM9 strain, Virology 287 (2001) 234e241. [10] S.A. Rubin, G. Amexis, M. Pletnikov, Z. Li, J. Vanderzanden, J. Mauldin, C. Sauder, T. Malik, K. Chumakov, K.M. Carbone, Changes in mumps virus gene sequence associated with variability in neurovirulent phenotype, J. Virol. 77 (2003) 11616e11624. [11] C.J. Sauder, K.M. Vandenburgh, R.C. Iskow, T. Malik, K.M. Carbone, S.A. Rubin, Changes in mumps virus neurovirulence phenotype associated with quasispecies heterogeneity, Virology 350 (2006) 48e57. [12] S. Goodbourn, L. Didcock, R.E. Randall, Interferons: cell signaling, immune modulation, antiviral response and virus countermeasures, J. Gen. Virol. 81 (2000) 2341e2364. [13] N. Fujii, N. Yokosawa, S. Shirakawa, Suppression of interferon response gene expression in cells persistently infected with mumps virus, and restoration from its suppression by treatment with ribavirin, Virus Res. 65 (1999) 175e185. [14] B. Gotoh, T. Komatsu, K. Takeuchi, J. Yokoo, Paramyxovirus strategies for evading the interferon response, Rev. Med. Virol. 2 (2002) 337e357. [15] T. Kubota, N. Yokosawa, S. Yokota, N. Fujii, C-terminal cys-rich region of mumps virus structural V protein correlates with block of interferon
872
[16]
[17]
[18]
[19]
[20]
[21]
N. Rosas-Murrieta et al. / Microbes and Infection 9 (2007) 864e872 alpha and gamma signal transduction pathway through decrease STAT1alpha, Biochem Biophys Res Commun. 283 (2001) 255e259. T. Kubota, N. Yokosawa, S. Yokota, N. Fujii, M. Tashiro, A. Kato, Mumps virus V protein antagonizes interferon without the complete degradation of STAT1, J. Virol. 79 (2005) 4451e4459. L. Marie, J.E. Durbin, D.E. Levy, Differential viral induction of distinct interferon-alpha genes by positive feedback through interferon regulatory factor-7, EMBO J. 17 (1998) 6660e6669. M. Nishio, D. Garcin, V. Simonet, D. Kolakofsky, The carboxyl segment of the mumps virus V protein associates with STAT proteins in vitro via a tryptophan-rich motif, Virology 300 (2002) 92e99. N. Yokosawa, S. Yokota, T. Kubota, N. Fujii, C-terminal region of STAT1a is not necessary for its ubiquitination and degradation caused by mumps virus v protein, J. Virol. 76 (2002) 12683e12690. J. Wang, R.D. Schreiber, I.L. Campbell, STAT1 deficiency unexpectedly and markedly exacerbates the pathophysiological actions of IFN-a in the central nervous system, Proc. Natl. Acad. Sci. 99 (2002) 16209e16214. J. Wang, I.L. Campbell, Innate STAT1-Dependent genomic response of neurons to the antiviral cytokine alpha interferon, J. Virol. 79 (2005) 8295e8302.
[22] J. Andrejeva, K.S. Childs, D.F. Young, T.S. Carlos, N. Stock, S. Goodbourn, R.E. Randall, The V proteins of paramyxoviruses bind the IFN-inducible RNA helicase, mda-5, and inhibit its activation of the IFN-b promoter, PNAS 101 (2004) 4917264e4917269. [23] P. Staeheli, Interferon-induced proteins and the antiviral state, Adv. Virus Res. 38 (1990) 147e200. [24] J.D. Mosca, P.M. Pitha, Transcriptional and posttranscriptional regulation of exogenous human beta interferon gene in simian cells defective in interferon synthesis, Mol. Cell. Biol. 6 (1986) 2279e2283. [25] K.E. Wright, K. Dimock, E.G. Brown, Biological characteristics of genetic variants of Urabe AM9 mumps vaccine virus, Virus Res. 67 (2000) 49e57. [26] C.M. Horvath, Weapons of STAT destruction, Eur. J. Biochem. 271 (2004) 4621e4628. [27] A. Kato, K. Kiyotani, Y. Sakai, T. Yoshida, T. Shioda, Y. Nagai, Importance of the cysteine-rich carboxyl-terminal half of V protein for Sendai virus pathogenesis, J. Virol. 71 (1997) 7266e7272. [28] C. Huang, K. Kiyotani, Y. Fujii, N. Fukuhara, A. Kato, Y. Nagai, T. Yoshida, T. Sakaguchi, Involvement of the zinc-binding capacity of Sendai virus V protein in viral pathogenesis, J. Virol. 74 (2000) 7834e7841.