Roles of Phosphorylation of the Nucleocapsid Protein of MumpsVirus in Regulating Viral RNA Transcription and Replication

James Zengel, Adrian Pickar, Pei Xu,* Alita Lin,* Biao He

Department of Infectious Diseases, University of Georgia College of Veterinary Medicine, Athens, Georgia

ABSTRACT

Mumps virus (MuV) is a paramyxovirus with a negative-sense nonsegmented RNA genome. The viral RNA genome is encapsi-dated by the nucleocapsid protein (NP) to form the ribonucleoprotein (RNP), which serves as a template for transcription andreplication. In this study, we investigated the roles of phosphorylation sites of NP in MuV RNA synthesis. Using radioactive la-beling, we first demonstrated that NP was phosphorylated in MuV-infected cells. Using both liquid chromatography-mass spec-trometry (LC-MS) and in silico modeling, we identified nine putative phosphorylated residues within NP. We mutated thesenine residues to alanine. Mutation of the serine residue at position 439 to alanine (S439A) was found to reduce the phosphoryla-tion of NP in transfected cells by over 90%. The effects of these mutations on the MuV minigenome system were examined. TheS439A mutant was found to have higher activity, four mutants had lower activity, and four mutants had similar activity com-pared to wild-type NP. MuV containing the S439A mutation had 90% reduced phosphorylation of NP and enhanced viral RNAsynthesis and viral protein expression at early time points after infection, indicating that S439 is the major phosphorylation siteof NP and its phosphorylation plays an important role in downregulating viral RNA synthesis.

IMPORTANCE

Mumps virus (MuV), a paramyxovirus, is an important human pathogen that is reemerging in human populations. Nucleocap-sid protein (NP) of MuV is essential for viral RNA synthesis. We have identified the major phosphorylation site of NP. We havefound that phosphorylation of NP plays a critical role in regulating viral RNA synthesis. The work will lead to a better under-standing of viral RNA synthesis and possible novel targets for antiviral drug development.

Mumps virus (MuV) infects humans, causing acute infectionwith hallmark enlargement of the parotid gland (1). Before

widespread vaccination in the late 1960s, mumps was the leadingcause of aseptic meningitis and caused deafness in children (2).Although vaccination has greatly reduced the number of infec-tions, large outbreaks have occurred recently in vaccinated popu-lations. The largest recent outbreak in the United States originatedat a university in Iowa in 2006, where over 5,000 cases were re-ported, compared to approximately 250 cases per year in the pre-ceding years (3). In 2014, there were over 1,100 cases of mumpsreported, mainly centered around universities (4). At least 90% ofthe individuals infected received the measles, mumps, and rubella(MMR) vaccine, and the majority of people received two doses(3). New strategies to control these outbreaks are needed. Under-standing the roles of each MuV protein in virus replication andpathogenesis will aid development of countermeasures for MuV.

Mumps virus (MuV) is a member of the family Paramyxoviri-dae in the genus Rubulavirus (1). It has a negative-sense, nonseg-mented RNA genome of 15,384 nucleotides. The genome iscomprised of seven transcriptional units that encode nine viralproteins in the order 3=-NP-V/I/P-M-F-SH-HN-L-5= with RNAsynthesis initiating at a single site at the 3= end. The RNA genomeassociates with NP to form the helical ribonucleoprotein (RNP),which protects the genome from degradation and serves as thetemplate for RNA synthesis. The NP also associates with the phos-phoprotein (P) and indirectly with the large protein (L), in whichP and L form the viral RNA-dependent RNA polymerase (vRdRp)(5). The vRdRp uses NP-encapsidated RNA as a template for bothreplication of the vRNA genome and production of mRNA (2).The vRdRp transcribes the NP-encapsidated RNA into 5= capped

and 3= polyadenylated mRNAs in the cytoplasm (6). Although theexact details of mRNA production are not known, the process iscurrently believed to involve termination and reinitiation (stopand start) at each gene junction. The vRdRp also replicates theviral RNA genome (7–10). It is thought that vRdRp transcribesvRNA first and replicates vRNA at a later stage after entry into hostcells. The regulation of the switch from transcription to replica-tion by vRdRp is not clear. It is thought that the phosphorylationstate of the P protein plays a critical role. Interestingly, P interactswith NP in RNP as well as free NP (11–14). Mumps virus NP isalso involved in virus budding. It interacts with the matrix (M)protein, which is critical for virus egress (15).

For all negative-sense nonsegmented RNA viruses, their ge-nomes are encapsidated with nucleoprotein to form RNP. It hasbeen shown that phosphorylation of NP plays a role in transcrip-tion, replication, and genome stability. In measles virus, phos-phorylation of NP has been shown to upregulate transcriptional

Received 12 March 2015 Accepted 2 May 2015

Accepted manuscript posted online 6 May 2015

Citation Zengel J, Pickar A, Xu P, Lin A, He B. 2015. Roles of phosphorylation of thenucleocapsid protein of mumps virus in regulating viral RNA transcription andreplication. J Virol 89:7338 –7347. doi:10.1128/JVI.00686-15.

Editor: D. S. Lyles

Address correspondence to Biao He, bhe@uga.edu.

* Present address: Pei Xu, Microbiology Department, University of Chicago,Chicago, Illinois, USA; Alita Lin, Simon Fraser University, Burnaby, BC, Canada.

Copyright © 2015, American Society for Microbiology. All Rights Reserved.

doi:10.1128/JVI.00686-15

7338 jvi.asm.org July 2015 Volume 89 Number 14Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

activity in a minigenome assay (16). A similar phenotype has alsobeen seen in rabies virus (17) and Nipah virus (18). In measlesvirus, phosphorylation of NP is involved in genome stability andphosphorylation of NP affects genome stability (19). In Marburgvirus, only phosphorylated NP is incorporated into nucleocapsidcomplexes (20). Similarly, in measles virus, phosphorylated NP ispreferentially incorporated into the nucleocapsid (21). Althoughit has been shown that NP is phosphorylated in MuV-infectedchicken cells, the role of phosphorylation is unclear (22).

In this study, we used in silico modeling and mass spectrometry(MS) to determine phosphorylation sites in the NP of MuV. Westudied the function of NP in RNA transcription and replicationthrough the use of a minigenome system (23) and a reverse genet-ics system (24).

MATERIALS AND METHODSPlasmids and cells. All plasmids were constructed using standard molec-ular cloning techniques. Plasmid sequences were based on the sequence ofa mumps virus isolated during an outbreak in Iowa from 2006 (MuV-IA;GenBank JN012242.1). MuV NP, P, and L were previously cloned into thepCAGGS expression vector (24, 25). Firefly luciferase (pFF-Luc) and a

MuV minigenome plasmid expressing Renilla luciferase flanked byMuV-IA trailer and leader sequences and under a T7 promoter (pT7-MG-RLuc) were also previously produced (23). Mutations were introducedinto pCAGGS-NP as previously described for introduction of pCAGGS-Pmutations (23). Plasmids containing the full-length genome of MuV-IApreviously used to rescue virus were mutated as necessary. Plasmids andsequences are available upon request.

HEK293T cells were maintained in Dulbecco’s modified Eagle’s me-dium (DMEM) supplemented with 5% fetal bovine serum (FBS) and 1%penicillin-streptomycin (P/S). BSR-T7 cells were maintained in DMEMsupplemented with 10% FBS, 1% P/S, 10% tryptose phosphate broth(TPB), and 400 �g/ml G418 to maintain T7 RNA polymerase (RNAP)expression. Vero and HeLa cells were maintained in DMEM with 10%FBS and 1% P/S. All cells were cultured at 37°C and 5% CO2. Cells werepassaged the day before to achieve about 85 to 95% confluence for infec-tion and 60 to 80% confluence for transfection.

Minigenome. BSR-T7 cells (1 day, 60 to 80% confluent, 24-well plate)were transfected with pCAGGS-P (80 ng), pCAGGS-L (500 ng), pT7-MG-RLuc (100 ng), pFF-Luc (1 ng), and various amounts ofpCAGGS-NP (wild type [wt] or mutant at 0, 12.5, 25, 50, or 100 ng) usingjetPRIME (Polyplus Transfection, France) according to the manufactur-er’s protocol. Empty pCAGGS vector was used to maintain a constant

FIG 1 Analysis of NP phosphorylation by mass spectrometry. (A) Phosphorylation of NP in infected or transfected cells. Vero cells were infected with MuV-IA,and HEK293T cells were transfected with NP and P. After 24 h, proteins were labeled with [35S]Met or [33P]orthophosphoric acid. Cells were lysed andimmunoprecipitated with anti-NP MAb. The samples were resolved by SDS-PAGE. (B) Immunoprecipitation (IP) of NP by anti-NP and anti-P. Vero cells wereinfected with MuV-IA, and lysate was immunoprecipitated with an anti-P MAb (sample 1). Unbound protein was immunoprecipitated with an anti-NP MAb(sample 2). The samples were resolved by SDS-PAGE followed by visualization using Coomassie blue, and the NP band was excised for analysis by LC-MS/MS.The P1 and P2 bands of MuV P were excised for analysis in another study. (C) Coverage of MS of NP after anti-P immunoprecipitation. (D) Coverage of MS ofNP after anti-NP immunoprecipitation. Phosphorylated positions are in bold, and positions not covered are struck through. Phosphorylation was consideredsignificant with a random probability score of less than 5%.

Phosphorylation of Mumps Virus NP

July 2015 Volume 89 Number 14 jvi.asm.org 7339Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

amount of total plasmid transfected per well. After 48 h, medium wasremoved and 100 �l of passive lysis buffer (Promega, Madison, WI) wasadded to each well, followed by shaking on an orbital shaker for 15 min.Forty microliters of lysate was transferred to a white 96-well plate, and adual-luciferase assay (Promega) was performed according to the manu-facturer’s protocol. Luminescence was detected using a GloMax 96-mi-croplate luminometer (Promega). The ratio of Renilla to firefly lumines-cence was determined for each well, and the average of 3 to 6 biologicalreplicates was calculated. The peak activity for each NP plasmid was de-termined, and each experimental data set was normalized to wt NP. Thedata reported are the combined data for at least 3 experimental replicates.

Virus rescue and sequencing. BSR-T7 cells (1 day, 60 to 80% conflu-ent, 6-well plate) were transfected with pCAGGS-NP (100 ng),pCAGGS-P (160 ng), pCAGGS-L (2,000 ng), and full-length genome(2,500 ng) using jetPRIME. After 48 to 72 h, transfected BSR-T7 cells weretrypsinized and cocultured with Vero cells at a ratio of 1:5 in a 10-cm dish.When cytopathic effect (CPE) was observed (2 to 7 days), the medium,likely containing virus, was collected and a plaque assay was performedusing Vero cells. Single plaques were isolated 6 to 7 days later and culturedin fresh Vero cells in 6-well plates to produce passage 1 (P1). After titerdetermination, P1 was passaged again in T75 or T150 flasks at a multiplic-ity of infection (MOI) of 0.01 to produce P2. After 72 h, virus was col-lected, bovine serum albumin (BSA) was added to a 1% final concentra-tion, and aliquots were stored at �80°C. Titer was determined by plaqueassay. Viral RNA was isolated using the QIAamp viral RNA minikit (Qia-gen, Valencia, CA) followed by synthesis of DNA templates using theSuperScript III one-step reverse transcription-PCR (RT-PCR) systemwith Platinum Taq (Life Technologies, Grand Island, NY), and 5 sets ofprimers were used to amplify the entire genome. Fragments were sent toGenewiz (South Plainfield, NJ) for sequencing using 6 to 10 primers perfragment. Only viruses matching the full-length plasmid sequence wereused for further experiments. Primer sequences are available upon re-quest.

Immunoprecipitation. Cells were lysed with whole-cell extractionbuffer (WCEB) (50 mM Tris-HCl [pH 8], 280 mM NaCl, 0.5% NP-40, 0.2

mM EDTA, 2 mM EGTA, and 10% glycerol) supplemented with proteaseinhibitors (1� protease inhibitor, 0.1 mM phenylmethylsulfonyl fluoride,or 1� protease-phosphatase inhibitor cocktail for radioactive labelingexperiments). Insoluble material was pelleted at 14,000 � g for 2 min, andthe supernatant was transferred to a new tube. Rec-protein G-Sepharose4B beads and anti-P or anti-NP monoclonal antibody (MAb) were addedto each tube and subjected to nutation overnight at 4°C. The next day,tubes were spun at 600 � g for 2 min and supernatant was aspirated. Threewashes were performed with 1 ml of WCEB using the same process. Thebead pellet was resuspended in 50 to 200 �l of 2� Laemmli sample buffer(Bio-Rad, Hercules, CA) plus 5% �-mercaptoethanol followed by heatingat 95°C for 5 min.

Mass spectrometry. Vero cells in a 10-cm plate were infected withMuV-IA at an MOI of 0.5. After 24 h, immunoprecipitation was per-formed as described above with an anti-MuV-P MAb. After overnightincubation, the sample was spun to pellet the beads and the supernatantwas collected. The washes were continued, and loading buffer was addedas described above. This produced the “anti-P” sample. The supernatantcollected prior to the first wash was used for a second immunoprecipita-tion with anti-MuV-NP MAb. This produced the “anti-NP” sample. Bothsamples were resolved on a 10% acrylamide gel by SDS-PAGE. The gel wasstained with Coomassie blue G250 in 10% acetic acid and 45% methanolfor 4 h, followed by destaining with destain buffer (10% acetic acid, 40%methanol, 50% water). Bands were excised from the gel and sent to theMass Spectrometry and Proteomics W. M. Keck Foundation Biotechnol-ogy Resource Laboratory (Yale University, New Haven, CT) for furtherprocessing and MS. Briefly, the protein was digested with trypsin andenriched for phosphoproteins on a TiO2 column (two times). Peptideswere separated on a nanoAcquity column (Waters, Milford, MA) (75 �mby 250 mm; eluted at 300 nl/min; 80-min run) with MS analysis on anOrbitrap Elite mass spectrometer (Thermo Scientific). Both the fractionenriched by the column and the flowthrough were analyzed by liquidchromatography-tandem mass spectrometry (LC-MS/MS), and peak lists

TABLE 1 Summary of phosphorylation site prediction in MuV NPa

Amino acidNetPhos (in silico prediction)value

LC-MS/MSphosphorylation score

Anti-P IP Anti-NP IP

T12 0.689 ND NDS25 0.071 0.0024 29T30 0.950 ND NDT42 0.891 ND NDS67 0.990 ND NDS94 0.996 0.28 0.29T183 0.091 ND 8.4S191 0.749 ND NDS226 0.510 ND NDS298 0.042 ND 0.059T368 0.816 ND NDT387 0.565 0.0004 0.0032T395 0.500 ND 150S439 0.992 0.0074 0.00036S520 0.979 ND NDS542 0.028 0.00015 3.4a Phosphorylation of MuV NP determined by in silico prediction and massspectrometry. Phosphorylation site prediction was performed using the NetPhos 2.0server (http://www.cbs.dtu.dk/services/NetPhos/). Values of �0.5 were considered toshow likely phosphorylation and are shown in bold. Phosphorylation sites found bymass spectrometry (as described in Fig. 1) are shown in the two right columns. Therandom probability score for each site is listed; scores of �0.05 were considered toshow likely phosphorylation and are shown in bold. IP, immunoprecipitation; ND, notdetected.

FIG 2 Phosphorylation of NP mutants in transfected cells. (A) Detection ofNP mutant phosphorylation. Residue S439 was found to be the major phos-phorylation site in NP. HEK293T cells were transfected with plasmids encod-ing either wt or NP with S/T residues mutated to A followed by [35S]Met-Cysor [33P]orthophosphoric acid. Immunoprecipitation was performed with ananti-NP MAb, and samples were resolved by SDS-PAGE. A representative gelis shown along with data from three separate experiments. (B) Summary ofquantified NP phosphorylation. The relative densities of the phosphorylatedand total NPs were calculated for each experiment. All data were normalized towt NP. Statistical analysis was performed by one-way ANOVA with the Holm-Šidák multiple-comparison test (n � 3; *, P � 0.001).

Zengel et al.

7340 jvi.asm.org July 2015 Volume 89 Number 14Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

were combined prior to a Mascot search against the NCBInr database withtaxonomy restricted to viruses. Phosphorylated peptides were consideredsignificant with a random probability score of less than 5%. For peptideswith more than one possible phosphorylation site, the Mascot Delta scoreand PhosphoRS score were used to determine which site was phosphory-lated.

Radioactive labeling for phosphorylation analysis. In order to exam-ine phosphorylation of NP expressed from transfected plasmid, 1 �g ofpCAGGS-NP (wt or mutant) was transfected into HEK293T cells in a6-well plate using jetPRIME in duplicate. After 24 h, cells were starved in1 ml DMEM lacking methionine and cysteine for 30 min followed bylabeling with about 50 �Ci/ml 35S-EasyTag Express 35S protein labelingmix (PerkinElmer, Waltham, MA) for 6 h. Alternatively, the cells werestarved with 1 ml DMEM lacking sodium phosphate followed by labelingwith about 100 �Ci [33P]orthophosphoric acid (PerkinElmer) for 6 h. Thecells were then lysed, and immunoprecipitation was performed withanti-NP MAb as described above. The samples were resolved on a 10%acrylamide gel by SDS-PAGE, and gels were dried. Radioactivity was de-tected by exposing the gel to a BAS-IP storage phosphor screen MS (Fuji)overnight. The screen was read on a Typhoon FLA 7000 laser scanner (GEHealthcare Life Sciences, Pittsburgh, PA), and the densitometry analysiswas performed using ImageQuant TL software (GE Healthcare). The ratioof 33P to 35S was calculated and reported.

In order to determine NP phosphorylation in infected cells, Vero cellsin a 6-well plate were infected with MuV (wt or S439A, S520A, or 542A

mutant) at an MOI of 0.1 for 1 h. Medium was replaced with DMEMcontaining 2% FBS and 1% P/S and incubated for 24 h. After 24 h, the cellswere lysed and labeled, and immunoprecipitation and quantification wereperformed as described above.

Growth curves. Vero or HeLa cells in a 10-cm dish were infected withMuV (wt or S439A, S520A, or 542A mutant) at an MOI of 0.01 or 5 in 5 mlof DMEM-2% FBS-1% P/S for 1 h in triplicate. Cells were washed fourtimes with phosphate-buffered saline (PBS), and 10 ml of DMEM-2%FBS-1% P/S was added to the cells. One sample was taken immediatelyafter the DMEM was added and labeled as 0 hours postinfection (hpi). Foran MOI of 5, samples were collected at 0, 6, 12, 24, 48, and 72 hpi. For anMOI of 0.01, samples were collected at 0, 24, 48, 72, 96, and 120 hpi. Allsamples were supplemented with 1% BSA after collection and stored at�80°C. Virus titers were determined by plaque assay on Vero cells. Resultswere confirmed in a second experiment. Significance was determined bytwo-way analysis of variance (ANOVA) using the Holm-Šidák method tocorrect for multiple comparisons.

Real-time PCR. Vero cells in a 6-well plate were infected with MuV(wt or S439A, S520A, or S542A mutant) at an MOI of 0.1 for 1 h andwashed three times with PBS, and 2 ml of DMEM-2% FBS-1% P/S wasadded to each well. At 0, 6, 12, and 18 hpi, total RNA was collected usingthe RNeasy Plus minikit with QIAshredder homogenization (Qiagen) ac-cording to the manufacturer’s instructions. cDNA was generated usingSuperScript III reverse transcriptase (Life Technologies) using 5 �l ofRNA according to the manufacturer’s directions. Oligo(dT)15 (Promega)

FIG 3 Effects of NP mutants on the MuV minigenome system. (A) Peak minigenome activity of the NP mutants. A MuV minigenome assay was performed usingplasmids encoding NP with possible phosphorylation sites mutated to alanine. The amount of NP plasmid was varied (12.5, 25, 50, and 100 ng/well). The ratioof Renilla luciferase to firefly luciferase activity was normalized to wt for each sample, and the peak titer is reported. Mutating position S439 was found tosignificantly increase minigenome activity (n � 3; ANOVA with Dunnett’s multiple-comparison test; *, P � 0.01; **, P � 0.001). (B) Representative activitycurves for the minigenome assay. The minigenome activities for wt and S439A mutant NP are shown at each concentration tested. (C) The expression of NP wasassessed by Western blotting. All mutant proteins were shown to be expressed at similar levels, as seen by blotting for NP using an NP-specific MAb.

Phosphorylation of Mumps Virus NP

July 2015 Volume 89 Number 14 jvi.asm.org 7341Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

was used for mRNA cDNA synthesis, and a primer specific for the nega-tive-sense genome (TGAACTAGCGAGGCCTATCCCCAAG) was usedfor genomic cDNA synthesis. Five microliters of cDNA was used for real-time PCR using a MuV-F-specific, 6-carboxyfluorescein (FAM)-taggedprobe (Life Technologies) and TaqMan gene expression master mix (LifeTechnologies) according to the manufacturer’s instructions. Real-timePCR was run on a StepOnePlus real-time PCR system (Life Technologies).Biological triplicate samples were run for each sample. Threshold cycle(CT) values were normalized to genomic RNA at 0 hpi. Significance wasdetermined by two-way ANOVA using the Holm-Šidák method to correctfor multiple comparisons.

Protein quantification. Cells were infected at an MOI of 0.1 for 6 h oran MOI of 5 for 24 h. Cells were washed once with PBS and trypsinized.Cells were collected into a 1.5-ml tube and pelleted (all spins at 600 � g),washed twice with DMEM-2% FBS, and fixed and permeabilized usingCytofix/Cytoperm solution (BD Biosciences, San Jose, CA) overnight at4°C. The MAbs were conjugated using Zenon Alexa 488 (A488) or allo-phycocyanin (APC) mouse IgG1 labeling kits (Life Technologies) accord-ing to the manufacturer’s specifications. Cells were then washed twicewith Perm/Wash buffer (BD Biosciences) followed by staining withanti-NP (A488) or anti-P (APC). After staining for 20 min at 4°C, sampleswere washed twice with Perm/Wash buffer and once with PBS-1% BSA.Cells were then resuspended in 500 �l of PBS-1% BSA. Flow cytometrywas performed using the LSRII flow cytometer (BD), and data were col-lected and analyzed using FACSDiva (BD). The mean fluorescence inten-sity was calculated for the stained population.

Total protein was also measured by infecting cells at an MOI of 5 asdescribed above, lysing them with 2� Laemmli sample buffer (Bio-Rad),and heating them at 95°C for 5 min. Samples were then resolved on a 10%acrylamide gel by SDS-PAGE and transferred to Amersham Hybond LFPpolyvinylidene difluoride (PVDF) membranes (GE Healthcare Life Sci-ences). Immunoblotting was performed by incubating the membraneswith anti-NP and anti-P MAb and anti-glyceraldehyde-3-phosphate de-hydrogenase (anti-GAPDH) (GT239) (GeneTex, Irvine, CA) in 5% milk-PBS-0.1% Tween 20 (PBST) overnight at 4°C, followed by three washeswith PBST, followed by incubation with Cy3-conjugated goat anti-mouseIgG diluted 1:2,500 (Jackson ImmunoResearch, West Grove, PA) in 5%milk-PBST for 1 h at room temperature. After the incubation, the mem-brane was washed four times with PBST and dried. The blot was visualizedon the Typhoon FLA 7000 laser scanner (GE Healthcare Life Sciences),and the densitometry analysis was performed using ImageQuant TL soft-ware (GE Healthcare).

RESULTSMuV NP is phosphorylated. Previously, it was shown that MuVNP was phosphorylated when mumps virus was grown in chickenembryo cells. In order to determine if NP is phosphorylated inmammalian cells, Vero cells were infected with the recombinantmumps virus rMuV(Iowa/US/06) (referred to as MuV), at anMOI of 0.1 for 24 h, and labeled with [35S]Met-Cys or[33P]orthophosphoric acid. Immunoprecipitation was performedusing anti-NP MAb (24), and samples were resolved by SDS-PAGE. NP was detected in the 33P labeling, indicating that NP wasphosphorylated in infected cells (Fig. 1A, left panel). To determineif NP phosphorylation was dependent on other viral proteins, cellswere transfected with a plasmid encoding MuV NP, labeled withradioactive reagents, and immunoprecipitated as described above(Fig. 1A, right panel). NP was detected in NP-transfected cellslabeled with 33P, indicating that NP was phosphorylated withoutany other viral proteins present.

Phosphorylation sites in NP were determined by in silicomodeling and mass spectrometry. Potential phosphorylationsites within NP were first identified using NetPhos 2.0 (http://www.cbs.dtu.dk/services/NetPhos/), a sequence-based predic-

tion method (26). Using this program, 12 phosphorylation siteswere predicted above the cutoff value of 0.5. These sites are sum-marized in Table 1.

To identify the phosphorylated residues in MuV NP in infectedcells, Vero cells were infected with MuV at an MOI of 0.1 for 48 h.In order to determine if there were differences in the phosphory-lation states of NP interacting with P, two sequential immunopre-cipitations were performed. Cells were first lysed and immuno-precipitated with a MAb specific for MuV-P, which pulled downall of the P protein in the sample, as well as NP that was associatedwith P (Fig. 1B, sample 1). The unbound protein from the firstimmunoprecipitation was then immunoprecipitated again usinga MAb specific for MuV-NP, which pulled down all of the non-P-associated NP (Fig. 1B, sample 2). These samples were resolved bySDS-PAGE and stained with Coomassie blue. The labeled NPbands were excised from the gel. The samples were subjected totryptic digestion and phosphopeptide enrichment and analyzedby LC-MS/MS. The coverage was between 93 and 94% for eachsample (Fig. 1C and D), with residues T387 and S439 found to bephosphorylated in both samples, and residues S25 and S542 phos-phorylated in only sample 1. The detected sites along with othersites that were below the defined cutoff for confirmed phosphor-ylation are summarized in Table 1.

The S439 residue in MuV NP was found to be the major phos-phorylation site in transfected cells. To assess which serine andthreonine residues contributed to NP phosphorylation, we choseto examine seven residues (S25, S94, T183, S298, T387, S439, andS542) based on identification by mass spectrometry and two res-idues (S67 and S520) based on high in silico prediction scores(0.990 and 0.979). Plasmids encoding NP were made with muta-tions to convert the serine or threonine residues to alanine in theencoded protein. Effects of the mutations on phosphorylationwere determined by transfection of cells with plasmids expressingwt NP or the NP with alanine substitutions in duplicate. After 24

FIG 4 Phosphorylation of NP mutants in infected cells. (A) Detection ofphosphorylation of NP mutants. Vero cells were infected with MuV (wt) ormutant viruses. Radioactive labeling was performed, and lysates were immu-noprecipitated with an anti-NP MAb. Samples were resolved by SDS-PAGE. Arepresentative gel is shown. (B) Summary of NP phosphorylation in infectedcells. MuV-NP-S439A was found to have significantly reduced phosphoryla-tion in infected cells (n � 3, Student t test; *, P � 0.01).

Zengel et al.

7342 jvi.asm.org July 2015 Volume 89 Number 14Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

h, one replicate was labeled with 35S and the other was labeled with33P. After 6 h of labeling, the cells were lysed and immunoprecipi-tation with anti-NP antibody was performed. The samples wereresolved by SDS-PAGE (Fig. 2A), and the ratio of 33P to 35S wascalculated (Fig. 2B). NP-S439A had little or no phosphorylation,while there was no significant difference between wt NP and theother mutants. The addition of P and L in the transfection had noeffect on the phosphorylation of wt NP during transfection (datanot shown).

The role of NP residues was assessed with a MuV minige-nome system. The role of NP in transcription and replication wasstudied using the MuV minigenome system previously developedin our lab (23). The minigenome system consists of plasmids re-quired for transcription and replication of viral RNA (NP, P, andL), as well as a plasmid that encodes a viral negative-sense mini-genome under a T7 promoter. When transfected into T7 RNAP-expressing BSR-T7 cells, the minigenome plasmid produces neg-ative-sense RNA containing the sequence for Renilla luciferaseflanked by the MuV leader and trailer. The MuV replication ma-chinery (NP, P, and L) replicates this RNA through a positive-sense intermediate and produces Renilla luciferase mRNA, whichis translated by host machinery. Changes in Renilla luciferase ac-tivity are due to changes in the replicative and transcriptionalactivity of the MuV replication system. The plasmids encoding NPmutants were tested at four concentrations, and the peak activitywas reported for each (Fig. 3A). An example of a minigenometitration is shown comparing wt NP and the S439A mutant (Fig.3B). Western blot assays were performed for each minigenome setto examine NP expression levels (Fig. 3C). Two mutants and wt

NP are shown, but all mutants had similar protein amounts whenthe same amount of plasmid was transfected, with some slightvariation. The same plasmids showed no difference in proteinlevels when using radioactive labeling. The use of multiple con-centrations of plasmid in the minigenome system also controls forany differences in expression. We found that the S439A mutanthad a higher level of minigenome activity. Four of the substitu-tions (S25A, S94A, T183A, and S298A) had lower activity, andthere was no change with the other four substitutions (S67A,T387A, S520A, and S542A).

S439 was the major phosphorylation site in NP in virus. Wehave constructed plasmids containing full-length MuV genomewith alanine substitutions in NP and produced seven plasmids(S25A, S94A, S183A, T387A, S439A, S520A, and S542A plasmids).The reverse genetics system previously developed in our lab wasused to successfully rescue three viruses (rMuV-NP-S439A,-S520A, and -S542A) (24, 25). Complete genome sequences wereconfirmed as outlined in Materials and Methods. At least threerescue attempts were made to rescue the other viruses, withoutsuccess, while wild-type viruses were consistently rescued. To ex-amine the phosphorylation states of NP in these viruses, Vero cellswere infected with wt MuV and the three mutant viruses. Radio-active labeling of infected cells and immunoprecipitation of celllysates were performed, and phosphorylation was determined asin the previous experiment (Fig. 4A). After calculating the ratio of33P to 35S (Fig. 4B), we found that there was a significant decreasein phosphorylation for MuV-NP-S439A, while there was nochange in the other two mutant viruses, indicating that S439 is the

FIG 5 Growth kinetics of MuV mutants. In each experiment, cells were infected with MuV (wt) or the S439A, S520A, or S542A mutant. Medium was collectedat various time points. The titer of virus in the medium was determined by plaque assay using Vero cells. (A) Vero cells infected at an MOI of 5. (B) Vero cellsinfected at an MOI of 0.01. (C) HeLa cells infected at an MOI of 5. (D) HeLa cells infected at an MOI of 0.01 (for all growth curves, n is 3; ANOVA with Dunnett’smultiple-comparison test; *, P � 0.05).

Phosphorylation of Mumps Virus NP

July 2015 Volume 89 Number 14 jvi.asm.org 7343Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

major phosphorylation site of NP. This is consistent with the re-sults obtained using transfected NP.

rMuV (wt), rMuV-NP-S439A, -S520A, and -S542A had changedgrowth rates in cell culture. To determine the effect of the NP mu-tations on virus growth in cell culture, single-cycle and multicyclegrowth curve assays were performed in Vero and HeLa cells. In asingle-cycle growth curve, cells were infected with an MOI of 5and supernatant was collected at 0, 6, 12, 24, 48, and 72 h postin-fection (hpi). In a multicycle growth curve, cells were infected atan MOI of 0.01 and supernatant was collected every 24 h until 144hpi. During single-cycle replication in Vero cells (Fig. 5A), therewas a lower virus titer for rMuV-NP-S439A at 6 hpi than for rMuV(wt) but an increased titer at 12, 24, 48, and 72 hpi. rMuV-NP-S542A had decreased titers at 12, 24, 48, and 72 hpi, and rMuV-NP-S520A had even lower titers at each of those time points.During multicycle replication in Vero cells (Fig. 5B), rMuV-NP-S439A had increased titers at 48, 96, and 120 hpi, while bothrMuV-NP-S520A and S542A had reduced titers at 72 and 120 hpicompared to rMuV (wt).

In HeLa cells, the growth characteristics of the viruses weresimilar to those in Vero cells for the single-cycle growth, but therewere differences between HeLa and Vero cells for the multicyclegrowth. There still was a lag in single-cycle growth for MuV-NP-S439A (Fig. 5C), but the virus was able to reach a significantlyhigher titer than rMuV (wt) by 48 hpi. During multicycle growthin HeLa cells (Fig. 5D), MuV-NP-S439A had lower titers from 72to 144 hpi. MuV-NP-S542A has higher titers than rMuV(wt) at 48and 96 hpi with slightly lower titers at 144 hpi. MuV-NP-S520A

had decreased titers compared to rMuV (wt) after 72 hpi, similarto growth in Vero cells.

rMuV-NP-S439A had increased protein present at 6 and 24 hpostinfection. To determine if there were differences in proteinproduction for these viruses, the amount of protein producedduring viral infection was examined by Western blotting first(Fig. 6A). rMuV-NP-S439A had increased viral protein in cells(Fig. 6B). rMuV-NP-S542A also had a small increase in viral pro-tein levels by Western blotting, although the increase was not sig-nificant. To determine if this difference was due to protein pro-duction on a per-cell basis, flow cytometry was used to stain forviral protein expression after infection. In order to determineearly protein production, cells were collected 6 h after infection(MOI of 0.1) and stained with antibodies specific to MuV NP or P.The mean fluorescence intensity was determined for each of thestained populations (Fig. 6C). rMuV-NP-S439A produced moreprotein at 6 hpi, as seen by staining for NP, while amounts of Pwere not detectable at this time. Furthermore, rMuV-NP-S439Ahad an increase in the amount of both NP and P produced on aper-cell basis using high-MOI (MOI of 5) infection at 24 hpi (Fig.6D). rMuV-NP-S542A had a trend toward higher protein levels byflow cytometry, but the difference was not significantly differentfrom rMuV (wt) (P � 0.12 to 0.4).

rMuv-NP-S439A had increased genome replication andmRNA production. To understand the difference in protein pro-duction and viral titer, the amounts of genome RNA and mRNAwere measured by quantitative real-time, reverse transcription-PCR (qRT-PCR) in infected Vero cells (MOI of 0.1). The cDNA

FIG 6 Protein production in infected Vero cells. In each experiment, cells were infected with MuV wt or the S439A, S520A, or S542A mutant. (A) Total proteinproduction in Vero cells infected at an MOI of 5 after 24 h. Samples were resolved by SDS-PAGE, and NP was quantified by Western blotting. (B) Summary oftotal protein production in Vero cells infected at an MOI of 5. Average density was calculated over multiple experiments, and the 439A mutant was found to haveincreased protein production. (C) Protein production in Vero cells infected at an MOI of 0.1 after 6 h. Cells were collected and stained using anti-NP (A488) andanti-P (APC) for flow cytometry. The mean fluorescence intensity was calculated for the stained population. (D) Protein production in Vero cells infected at anMOI of 5 after 24 h. Cells were treated as described for panel C. Statistical analysis was performed by one-way ANOVA with the Holm-Šidák multiple-comparisontest (n � 3; *, P � 0.05).

Zengel et al.

7344 jvi.asm.org July 2015 Volume 89 Number 14Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

was generated using a genome-specific primer to quantify genomeRNA and oligo(dT) to quantify mRNA. The probe used for allsamples was specific to MuV F or HN. Data from assays using theMuV F-specific probe are reported. It was found that rMuV-NP-S439A had increased genomic RNA production at 6, 12, and 18hpi, while rMuV-NP-S520A and -S542A had decreased genomicRNA production at 12 and 18 hpi compared to rMuV (wt) (Fig.7A). rMuV-NP-S439A also had increased mRNA production at alltime points, while rMuV-NP-S520A and -S542A had decreasedlevels at 6, 12, and 18 hpi compared to MuV (wt) (Fig. 7B). Whencomparing the ratio of mRNA to genomic RNA, rMuV-NP-S439A had increased relative mRNA levels at 0 and 6 hpi, rMuV-NP-S542A had reduced relative levels at 6 and 12 hpi, and rMuV-NP-S520A had no significant differences compared to MuV (wt)(Fig. 7C).

Mutations in NP affect NP-P interaction during infectionbut not transfection. To investigate the mechanism of the phos-phorylation of NP in regulating viral RNA synthesis, NP and Pinteraction was examined. Cells were transfected with plasmidsencoding NP and P and immunoprecipitation was performed us-ing either anti-NP (Fig. 8A) or anti-P (Fig. 8B) MAbs. After coim-munoprecipitation of NP and P when using plasmids encodingany of the mutant NPs, no difference was detected. To assess dif-ferences in NP and P association in infected cells, Vero cells wereinfected with MuV (wt) or S439A, S520A, and S542A mutants.Coimmunoprecipitation and total protein visualization were per-formed (Fig. 8C). The ratio of NP to P was calculated (Fig. 8D),and MuV-S439A had decreased amounts of NP coimmunopre-

cipitated with P during the anti-P pulldown. While there was adifference in NP-P interaction during infection, there was no dif-ference in the amounts of NP or P in sucrose gradient-purifiedvirus from infected cells (data not shown).

DISCUSSION

In this study, we identified and confirmed multiple phosphory-lated residues in MuV NP by mass spectrometry and directedmutagenesis. We showed that S439 was the major site of phos-phorylation. Mutating this residue to alanine caused an increase inminigenome activity and higher levels of viral RNA and proteinexpression in rMuV-NP-S439A-infected cells than in wild-typevirus-infected cells at early time points after infection. This is incontrast to previous work on measles, rabies, and Nipah viruses,which have decreased activity in their respective minigenome sys-tems when NP phosphorylation is reduced (16–18). To the best ofour knowledge, this is the only case in which decreased phosphor-ylation of the nucleoprotein of a virus resulted in increased activ-ity, indicating that phosphorylation of NP downregulates viralRNA synthesis. We hypothesize that phosphorylation can bothup- and downregulate activity and that these differences may de-pend on the site that is being phosphorylated.

The mechanism of downregulation of viral RNA synthesis byS439 of NP is not clear. We assessed the RNA binding of the S439Amutant along with all other alanine substitution mutants, but nodifferences were found compared to any of the mutants and wt NP(data not shown). One interesting difference between wt NP andNP-S439A is their interaction with P. While we were not able to

FIG 7 Genomic RNA and mRNA levels in infected Vero cells. Vero cells were infected at an MOI of 0.1 with MuV (wt) or the S439A, S520A, or S542Amutant. Total RNA was extracted from biological replicates (n � 3). Real-time PCR was performed on each sample using a MuV-F-specific FAM-taggedprobe. (A) Levels of genomic RNA. Genome replication was calculated after normalization to genomic RNA levels at 0 hpi. (B) Levels of viral mRNA.mRNA production was calculated after normalization to genomic RNA levels at 0 hpi. (C) Quantification of relative levels of mRNA to genomic RNA. Theratio of mRNA to genomic RNA was calculated at each time point. Statistical analysis was performed by multiple t tests with the Holm-Šidák multiple-comparison test (n � 3; *, P � 0.05).

Phosphorylation of Mumps Virus NP

July 2015 Volume 89 Number 14 jvi.asm.org 7345Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

show any differences in interactions between NP and P or NP andM in transfected cells, we found that there was much less NPpulled down in cells infected with rMuV-NP-S439A by an anti-PMAb in coimmunoprecipitation. This result suggests that phos-phorylation at S439 increases NP association with P in infectedcells, although phosphorylation at S439 was found in both P-as-sociated NP and free NP (Fig. 1B; Table 1). This is rather surpris-ing because the domain of NP interacting with P is located at theN-terminal amino acid residue 400 of NP and mutation at residue439 should not have affected binding with P (13, 14). The differ-ence in expression levels of NP, P, and L in transfection and infec-tion may contribute to the difference observed in the NP-Pbinding. It is possible that a previously undetected region of NP(C-terminal tail domain) can interact with P, in the presence ofother viral proteins such as L and M.

The fact that rMuV-NP-S439A caused a slight lag in virionproduction even though there was more RNA and protein at thattime point than for the other viruses suggests that there may besome defect in packaging of RNA for production and release ofprogeny virus. The lag in virion production may be detrimental tovirus growth in vivo, which could explain why this position ishighly conserved among all MuV strains (data not shown). Thelower titer of rMuV-NP-S439A in HeLa cells, a type I interferon(IFN)-producing cell line, than in wt MuV is consistent with thisresidue being critical for MuV growth in vivo. It is possible that therMuV-NP-S439A virus was more sensitive to type I IFN during

HeLa infection, which is not observed in Vero cells, an IFN-defec-tive cell line.

While mutations at other amino acid residues did not producea significant decrease in phosphorylation in transfected cells, theydid play critical roles. There was a small decrease in phosphoryla-tion of the S542A mutant. There was also a small but significantdecrease in both genomic RNA and mRNA, although there was aslight increase in protein produced. It is possible that there may besome defect in budding for the rMuV-NP-S542A virus, whichmight have caused the decrease in RNA and increased protein inthe cells. Less protein may be exported in progeny virions. Inparainfluenza virus type 5 (PIV5), a closely related paramyxovi-rus, it is known that negatively charged residues in the tail of NPare important for NP-M interaction and virus budding (27). Theimpact of the mutation at S542 may also be attributed to the tran-sient nature of phosphorylation at this site. This is consistent withthe mass spectrometry data that show that S542 was significantlyphosphorylated only in the P-associated sample. rMuV-NP-S520A had a lower virus titer than MuV (wt), suggesting that therewas some defect in virus growth, although there were only modestdecreases in the amounts of genomic RNA and mRNA produced.We were also unable to find any phosphorylation at this site bymass spectrometry and saw no decrease in phosphorylation whenthe residue was changed to alanine. It is possible that phosphory-lation at this residue per se does not have an impact on virus lifecycle but that the residue itself is important for the virus life cycle.

FIG 8 Interaction between MuV NP and P. (A) Interaction between NP and P in transfected cells. HEK293T cells were transfected with wt and mutant NPs andP. Proteins were labeled with [35S]Met-Cys, and coimmunoprecipitation was performed using antibodies specific to NP. No difference was detected in theamounts of NP or P pulled down. (B) Interaction between NP and P in transfected cells. Using the same samples as those in panel A, coimmunoprecipitation wasperformed using antibodies specific to P. No difference was detected in the amounts of NP or P pulled down. (C) Interaction between NP and P in infected cells.Vero cells were infected with MuV (wt) or the S439A, S520A, or S542A mutant or mock infected. Total protein was labeled with [35S]Met-Cys, and coimmu-noprecipitation was performed using antibodies specific to NP or P. (D) The mean of the NP-to-P ratio for the anti-P immunoprecipitation is graphed with thestandard error of the mean shown. There was less NP coimmunoprecipitated with P during infection with rMuV-NP-S439A. n � 3; Student t test; *, P � 0.05.

Zengel et al.

7346 jvi.asm.org July 2015 Volume 89 Number 14Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

When the amino acid residue at position S94, T183, or S298 wasreplaced with alanine, there was a large decrease in minigenomeactivity. Viruses containing these mutations were not obtainedafter multiple attempts, likely due to the low level of replicativeactivity seen in the minigenome system, suggesting that these res-idues play important roles in the virus life cycle. It is surprisingthat viruses containing mutations at S67 and T387 were not ob-tained since these mutations did not affect minigenome activity. Itis likely that these residues, not necessarily their phosphorylationstatus, may play a role beyond viral RNA synthesis. Interestingly,mutations at the unstructured C terminus of NP allowed rescue ofinfectious virus, and we were unable to rescue infectious virusescontaining mutations at the N-terminal end of NP, suggesting thatresidues in the more-structured N terminus need to be preserved.

Understanding the roles of phosphorylation of MuV NP willnot only contribute to our knowledge of viral RNA synthesis butalso aid design of novel antivirals and the next generation of vac-cines. Since MuV is not known to encode its own kinase to phos-phorylate NP, the host kinases responsible for NP phosphoryla-tion may be viable drug targets. While host kinase responsible forphosphorylation of NP’s S439 residue is not likely a good target forantiviral drug development, kinases responsible for phosphoryla-tion of the N terminus of NP may be good targets since mutatingthese residues resulted in difficulties in obtaining infectious vi-ruses. Identifying the host kinases responsible for phosphorylat-ing these critical sites of NP may lead to development of small-molecule inhibitors of the kinases as anti-MuV drugs, which donot exist at present. Preventing phosphorylation at NP-S439 wasable to increase viral replication in Vero cells. This mutation canbe incorporated into vaccine viruses, enabling faster growth andhigher-titer viruses, which will reduce the cost of future vaccines.

ACKNOWLEDGMENTS

We appreciate the helpful discussions and technical assistance from allmembers of Biao He’s laboratory.

This work was supported by grants (R01AI097368 and R01AI106307)from the National Institutes of Health.

REFERENCES1. Carbone KM, Wolinsky JS. 2001. Mumps virus, p 1381–1400. In Knipe

DM, Howley PM, Griffin DE, Lamb RA, Martin MA, Roizman B, StrausSE (ed), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadel-phia, PA.

2. Knipe DM, Howley PM, Griffin DE, Lamb RA, Martin MA, Roizman B,Straus SE (ed). 2007. Fields virology, 5th ed. Lippincott Williams &Wilkins, Philadelphia, PA.

3. Marin M, Quinlisk P, Shimabukuro T, Sawhney C, Brown C, LebaronCW. 2008. Mumps vaccination coverage and vaccine effectiveness in alarge outbreak among college students—Iowa, 2006. Vaccine 26:3601–3607. http://dx.doi.org/10.1016/j.vaccine.2008.04.075.

4. National Center for Immunization and Respiratory Diseases. 15 Janu-ary 2015, posting date. Mumps cases and outbreaks. Division of ViralDiseases, National Center for Immunization and Respiratory Diseases,Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/mumps/outbreaks.html.

5. Takeuchi K, Hishiyama M, Yamada A, Sugiura A. 1988. Molecularcloning and sequence analysis of the mumps virus gene encoding the Pprotein: mumps virus P gene is monocistronic. J Gen Virol 69:2043–2049.http://dx.doi.org/10.1099/0022-1317-69-8-2043.

6. Emerson SU, Yu Y-H. 1975. Both NS and L proteins are required for invitro RNA synthesis by vesicular stomatitis virus. J Virol 15:1348 –1356.

7. Abraham G, Banerjee AK. 1976. Sequential transcription of the genes ofvesicular stomatitis virus. Proc Natl Acad Sci U S A 73:1504 –1508. http://dx.doi.org/10.1073/pnas.73.5.1504.

8. Ball LA, White CN. 1976. Order of transcription of genes of vesicularstomatitis virus. Proc Natl Acad Sci U S A 73:442– 446. http://dx.doi.org/10.1073/pnas.73.2.442.

9. Emerson SU. 1982. Reconstitution studies detect a single polymeraseentry site on the vesicular stomatitis virus genome. Cell 31:635– 642. http://dx.doi.org/10.1016/0092-8674(82)90319-1.

10. Iverson LE, Rose JK. 1982. Sequential synthesis of 5=-proximal vesicularstomatitis virus mRNA sequences. J Virol 44:356 –365.

11. Cox R, Pickar A, Qiu S, Tsao J, Rodenburg C, Dokland T, Elson A, HeB, Luo M. 2014. Structural studies on the authentic mumps virus nucleo-capsid showing uncoiling by the phosphoprotein. Proc Natl Acad SciU S A 111:15208 –15213. http://dx.doi.org/10.1073/pnas.1413268111.

12. Cox R, Green TJ, Purushotham S, Deivanayagam C, Bedwell GJ, Pre-velige PE, Luo M. 2013. Structural and functional characterization of themumps virus phosphoprotein. J Virol 87:7558 –7568. http://dx.doi.org/10.1128/JVI.00653-13.

13. Kingston RL, Gay LS, Baase WS, Matthews BW. 2008. Structure of thenucleocapsid-binding domain from the mumps virus polymerase; an ex-ample of protein folding induced by crystallization. J Mol Biol 379:719 –731. http://dx.doi.org/10.1016/j.jmb.2007.12.080.

14. Kingston RL, Baase WA, Gay LS. 2004. Characterization of nucleocapsidbinding by the measles virus and mumps virus phosphoproteins. J Virol78:8630 – 8640. http://dx.doi.org/10.1128/JVI.78.16.8630-8640.2004.

15. Li M, Schmitt PT, Li Z, McCrory TS, He B, Schmitt AP. 2009. Mumpsvirus matrix, fusion, and nucleocapsid proteins cooperate for efficientproduction of virus-like particles. J Virol 83:7261–7272. http://dx.doi.org/10.1128/JVI.00421-09.

16. Hagiwara K, Sato H, Inoue Y, Watanabe A, Yoneda M, Ikeda F, FujitaK, Fukuda H, Takamura C, Kozuka-Hata H, Oyama M, Sugano S,Ohmi S, Kai C. 2008. Phosphorylation of measles virus nucleoproteinupregulates the transcriptional activity of minigenomic RNA. Proteomics8:1871–1879. http://dx.doi.org/10.1002/pmic.200701051.

17. Yang J, Koprowski H, Dietzschold B, Fu ZF. 1999. Phosphorylation ofrabies virus nucleoprotein regulates viral RNA transcription and replica-tion by modulating leader RNA encapsidation. J Virol 73:1661–1664.

18. Huang M, Sato H, Hagiwara K, Watanabe A, Sugai A, Ikeda F, Kozuka-Hata H, Oyama M, Yoneda M, Kai C. 2011. Determination of a phos-phorylation site in Nipah virus nucleoprotein and its involvement in virustranscription. J Gen Virol 92:2133–2141. http://dx.doi.org/10.1099/vir.0.032342-0.

19. Sugai A, Sato H, Yoneda M, Kai C. 2013. Phosphorylation of measlesvirus nucleoprotein affects viral growth by changing gene expression andgenomic RNA stability. J Virol 87:11684 –11692. http://dx.doi.org/10.1128/JVI.01201-13.

20. Becker S, Huppertz S, Klenk H-D, Feldmann H. 1994. The nucleopro-tein of Marburg virus is phosphorylated. J Gen Virol 75:809 – 818. http://dx.doi.org/10.1099/0022-1317-75-4-809.

21. Gombart AF, Hirano A, Wong TC. 1995. Nucleoprotein phosphorylatedon both serine and threonine is preferentially assembled into the nucleo-capsids of measles virus. Virus Res 37:63–73. http://dx.doi.org/10.1016/0168-1702(95)00020-Q.

22. Naruse H, Nagai Y, Yoshida T, Hamaguchi M, Matsumoto T, IsomuraS, Suzuki S. 1981. The polypeptides of mumps virus and their synthesis ininfected chick embryo cells. Virology 112:119 –130. http://dx.doi.org/10.1016/0042-6822(81)90618-8.

23. Pickar A, Xu P, Elson A, Li Z, Zengel J, He B. 2014. Roles of serine andthreonine residues of mumps virus P protein in viral transcription and repli-cation. J Virol 88:4414–4422. http://dx.doi.org/10.1128/JVI.03673-13.

24. Xu P, Li Z, Sun D, Lin Y, Wu J, Rota PA, He B. 2011. Rescue ofwild-type mumps virus from a strain associated with recent outbreakshelps to define the role of the SH ORF in the pathogenesis of mumps virus.Virology 417:126 –136. http://dx.doi.org/10.1016/j.virol.2011.05.003.

25. Xu P, Luthra P, Li Z, Fuentes S, D’Andrea JA, Wu J, Rubin S, Rota PA,He B. 2012. The V protein of mumps virus plays a critical role in patho-genesis. J Virol 86:1768 –1776. http://dx.doi.org/10.1128/JVI.06019-11.

26. Blom N, Gammeltoft S, Brunak S. 1999. Sequence and structure-basedprediction of eukaryotic protein phosphorylation sites1. J Mol Biol 294:1351–1362. http://dx.doi.org/10.1006/jmbi.1999.3310.

27. Schmitt PT, Ray G, Schmitt AP. 2010. The C-terminal end of parainflu-enza virus 5 NP protein is important for virus-like particle production andM-NP protein interaction. J Virol 84:12810 –12823. http://dx.doi.org/10.1128/JVI.01885-10.

Phosphorylation of Mumps Virus NP

July 2015 Volume 89 Number 14 jvi.asm.org 7347Journal of Virology

on February 4, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from