Journal of Virological Methods 130 (2005) 140–144

Short communication

Visualization of infectious pancreatic necrosis virus (IPNV)particles labeled with fluorescent probes

Manuel Osorio, Juan Carlos Espinoza, Juan Kuznar∗

Laboratorio de Bioquımica y Virologıa, Instituto de Ciencias Biologicas y Quımicas, Facultad de Ciencias,Universidad de Valparaıso, Casilla 50 30, Valparaıso, Chile

Received 2 May 2005; received in revised form 9 June 2005; accepted 14 June 2005Available online 19 July 2005

Abstract

Infectious pancreatic necrosis virus (IPNV) particles were labeled with SYBR Green I or a monoclonal antibody and FITC-conjugatedsecondary antibody and examined in a fluorescence microscope. Labeled viral particles were visualized in a narrow range of pixels. ComparingIPNV particles with fluorescent phage T4 virions, the former, as expected, were seen smaller in size. The method allows the rapid and accuratecounting of viral particles both on filters and bound to the cell surface. In addition, IPNV particles can be specifically enumerated in the

allynec-sicalisu-

y of

tio-rates-ical

rderesisthodsey a

ndedplica-iralh p.i.

s inand

presence of other virions and the ratio between physical particles and virus infectivity can be easily calculated as well.© 2005 Published by Elsevier B.V.

1. Introduction

The visualization of single virus particles until recentlyhad been restricted to the use of electron microscopy tech-niques. It is known that optical microscopes have a resolutionlimited to observation of objects far bigger than viruses. How-ever, nowadays the improvement of fluorescence microscopyallows detection of continuously expanding number of viri-ons as well as their interaction with host cells.

Different strategies have been used to label virions withfluorophores. Adenovirus and adeno-associated virus havebeen chemically labeled in order to examine the earlystages of virus infection (Persson et al., 1983; Bartlett andSamulski, 1998; Leopold et al., 1998). Fluorescent chimerasof viral proteins have been incorporated into virions in orderto examine virus assembly in infected cells (Andrawiss etal., 2003) and virions have also been labeled by indirectimmunostaining (Pizzato et al., 2001).

Visualizing single viral particles provides important qual-itative and quantitative information about virus-cell interac-tions, virus concentration in suspensions and the calculation

of physical particles/PFU ratio. Moreover, to biochemicfollow the interactions between viruses and cells, it isessary to reduce as much as possible the ratio of phyparticles/PFU in order to have representative results. Valization of single labeled virions increases the reliabilitperforming such controls.

Infectious pancreatic necrosis virus (IPNV) is the elogical agent of a disease, which causes high mortalityin young salmonid fish (Dorson, 1982). Due to the expansion and diversification of aquaculture and the ecologimportance of IPNV, much effort has been done in oto gain knowledge about its biology and the pathogenof the disease and to develop advanced diagnostic me(Rodriguez Saint-Jean et al., 2003). IPNV belongs to thBirnaviridae family whose members are characterized bgenome which consists of two segments of double straRNA and a naked icosahedral single-shelled capsid. Retion of IPNV takes place in the cytoplasm and progeny vproteins can be detected by immunofluorescence from 4(Espinoza and Kuznar, 2002; Dobos, 1995). The replicativecycle of IPNV has been studied in its major aspectinfected cells as well as the timing of protein synthesis

∗ Corresponding author. Tel.: +56 32 508072; fax: +56 32 508072.E-mail address: Juan.Kuznar@uv.cl (J. Kuznar).

RNA transcription and replication (Dobos, 1995). However,new tools are required to further analyze virus-cell inter-

0166-0934/$ – see front matter © 2005 Published by Elsevier B.V.d

oi:10.1016/j.jviromet.2005.06.007

M. Osorio et al. / Journal of Virological Methods 130 (2005) 140–144 141

actions and to enumerate single viral particles in general.Many features of IPNV biology could be addressed if singleIPNV particles could be seen as fluorescent dots: virus pen-etration, uncoating, defective interfering particles and viralpersistence.

In this study it is shown that labeled IPNV particles canbe specifically enumerated from captured digitalized images.They also seem to be the smallest viral particles seen with afluorescent microscope.

2. Materials and methods

2.1. Cells and virus

Chinook salmon embryo cells (CHSE-214) were grownas monolayers at 20◦C in Eagle’s minimal essential medium(EMEM) with Earle’s salts and supplemented with 2 mMl-glutamine, 0.1 mM non-essential amino acids (GIBCO),Grand Island, NY, sodium bicarbonate (1 mg/ml), gentamicin(50�g/ml) and 10% of fetal bovine serum (FBS, GIBCO).Cells were grown on 24 well plastic plates containing 12 mmcircular glass coverslips or 35 mm diameter plastic dishes.The virus was quantitated using the method of fluorescentfoci (FF) (Espinoza and Kuznar, 2002).

The IPNV strain used in this study was isolated inC ta byi of0 ffect( lls. Tor awedt ge d byc

2

ughd sw ,W lteru NVp ize-s ) wasa thev on.T ce-d dos iod,tw lidew m,V h ac

The stained samples were examined in an OlympusBX60 epifluorescence microscope and the images were cap-tured, digitalized and processed with Scion Image software(http://www.scioncorp.com). To enumerate virions with thissoftware, the size in pixels of the fluorescent microsphereswas used as a reference.

2.3. Immunostaining of IPNV particles

The Anodisc filter with virions was placed in a plastic dishand incubated with a monoclonal antibody (mAb) againstVP3 (1�g/ml) in PBS during 1 h at room temperature (RT).After the staining period, the filter was mounted in a filter-ing tower and washed under vacuum with 20 ml of filtered(0.02�m) Milli Q quality water. After placing again the filterin a plastic dish, a secondary antibody conjugated with FITC(Sigma, 1/100 in PBS) was added to the filter and maintainedat RT during 30 min. After washing, the filters were mountedon slides as described before.

In order to stain the virus particles bound to cells, 12 mmround coverslips containing monolayers of CHSE-214 cellswere incubated with partially purified IPNV (2× 108 FF/ml)during 1 h at 4◦C. After the incubation period, the cells werewashed with cold PBS and fixed with 3.7% paraformaldehydein PBS during 5 min at RT and stained with the antibodies inthe same conditions as described for the virions retained ont

3

sesa encem thenf darkb le tos acityo tchedw nmd stan-d BRG inseto ge T4i rn,a nm.P res of1 fluo-r

fields se thei ea ec thefi s ared een

hile and identified as a VR-299 serotype (Espinoza el., 1985). The virus was propagated in plastic flasks

nfecting susceptible cells at a multiplicity of infection.001 FF/cell. After 3 days, when a massive cytophatic eCPE) occurred, the flasks were shaken to detach the ceelease the virus, the infected cells were frozen and thwice and stored in aliquots at−20◦C. Before the labelinxperiments the thawed viral suspensions were clarifieentrifuging at 20 000 g during 15 min.

.2. Staining of viral particles with SYBR Green I

Suspensions containing IPNV were filtered first throisposable filters of 0.2�m pore diameter. 1ml fractionere further filtered through a 25 mm, 0.02�m pore-sizehatman Anodisc (Fisher Scientific, Inc.) membrane fi

sing a vacuum pump. In order to calibrate the size of IParticles, partially purified phage T4 and/or fluorescent standardized spheres of 100 nm (Molecular Probes, Inc.dded in some experiments. The dry filters containingiral particles were removed with the vacuum pump stillhe virions were stained in the filters following the proure described byNoble (2001). Briefly, the filter is loadever a drop of SYBR Green I solution (0.2�g/ml), sampleide up, during 15 min in the dark. After the staining perhe filter is washed with filtered (0.02�m) Milli Q qualityater in a filtering tower, then mounted on a glass sith an antifading agent (Vecta shield R Mounting Mediuector Laboratories, Burlingham, CA) and covered witoverslip.

he filters.

. Results and discussion

In order to determine if small double stranded RNA virus IPNV can be visualized by means of epifluorescicroscopy, partially purified virus was stained withucleic acid stain SYBR Green I. As shown inFig. 1, uni-

ormly sized fluorescent particles can be seen against aackground. Slight differences in size can be attributablight differences in focus as well as the stain binding capf the particle. To test whether the fluorescent spots maith the sizes expected for IPNV particles, we used 100iameter fluorescent microspheres and phage T4 as sizeards. The latter, stained in its DNA with the probe SYreen I, is seen brighter and bigger than IPNV particles (f Fig. 1). These differences are expected because pha

s 200 nm long and 80–100 nm wide; IPNV particles in ture spherical particles with an average diameter of 60hage T4 particles and standard fluorescent microsphe00 nm (FluoSpheres, Molecular Probes) appeared asescent spots of the same size (not shown).

The fluorescent material observed in the microscopichould be of a size range between 20 and 200 nm becaunfected cell lysate was filtered through a 0.2�m membrannd the virions were collected in a 0.02�m filter. Under thesonditions, IPNV particles can be easily enumerated fromlter using the appropriate software. Fluorescent particleistributed in a narrow range of sizes, most of them betw

142 M. Osorio et al. / Journal of Virological Methods 130 (2005) 140–144

Fig. 1. Visualization of IPNV particles stained with SYBR Green I. Clarifiedsuspensions of IPNV were filtered through filters of 0.2�m pore diameterand further retained in a 25 mm, 0.02�m pore-size filter. The particles werestained with SYBR Green I and observed under a fluorescence microscope.Inset; two T4 phage particles also stained with SYBR Green I, are shownfor comparative purposes.

3 and 5 pixels, indicating that the counting method is suitedto the measurement of physical particles in the range sizeexpected for IPNV particles (Fig. 2). It has been proposedthat 50 nm sized particles represents the lower limit allowingthe direct visualization of fluorescent viruses (Pizzato et al.,2001). With a mean size of 60 nm, IPNV seems to be the

Fig. 2. Size distribution of IPNV particles stained with SYBR Green I. Aftercapturing the fluorescent dots by digital imaging, randomly selected areas ofthe filters were further processed using the Scion Image program in order tor xels.

Table 1Enumeration of IPNV particles in the presence of phage T4

Sample Physical particles

IPNVa T4b

1 8 (S.D. 0.5,n = 3)× 109 4.2× 109

2 (1:2) 4.7 (S.D. 0.6,n = 3)× 109 10.4× 109

3 (1:5) 1.7 (S.D. 0.3,n = 3)× 109 20.8× 109

a Immunofluorescent labeled IPNV particles were enumerated from threefilters prepared from each sample.

b SYBR Green I stained T4 particles were enumerated from a stock solu-tion. The figures in the table were calculated from the amount of the solutionadded to each sample containing IPNV.

smallest virus particle which has been visualized by fluores-cence staining.

IPNV particles can be specifically visualized in mix-tures containing other viruses or similar sized particles, pro-vided that fluorescent labeling is performed with fluorescentantibodies instead of SYBR Green I or other general pur-poses dyes. Uniform size fluorescent IPNV particles canbe easily distinguished against a dark background whenpartially purified viral preparations are filtered and stainedwith a monoclonal antibody (Fig. 3A–C). Negative controls,omitting monoclonal antibody or using mock-infected cellswere prepared and the obtained results were as expected(not shown).

In order to verify that IPNV quantitation was not affectedby the presence of another virus, a suspension of IPNV wasdiluted 1:2 and 1:5 in PBS. To each of the suspensions differ-ent amounts of phage T4 were added. As seen inFig. 3andTable 1, the fluorescent dots of stained IPN virions decreasedin number at the expected proportion irrespective of the factthat phage T4 particles were added to the containing IPNVsuspensions.

Interestingly, the method can be applied to the study ofearly interactions between IPN virions and their host cells.As shown inFig. 3D, immunostained IPNV particles can beeasily visualized and enumerated at the cell surface. Conse-quently, adsorption kinetics can be followed. The penetrationo ellm cols.

nicm on-t tot untedp repa-r gingt arec thev w-b g ofv uratem rdert m thee do d

ecord the distribution of virions having different sizes expressed in pi

f the virus can be followed as well if the integrity of the cembrane is maintained using appropriate fixation protoCounting viral particles by means of electro

icroscopy is time consuming and requires carefully crolled conditions to maximize the fixation of the virionshe supporting membranes, otherwise, the number of coarticles can be underestimated. Moreover, the virus pation should be clean from contaminating material ranhe size of the virions. Alternatively, if physical particlesalculated from protein and nucleic acid composition ofirus, highly purified virions are mandatory. All these draacks are avoided using the immunofluorescent labelinirus particles which allows, among other uses, the acceasurement of the physical particles/PFU ratio. In o

o give an example, the same samples used to perforxperiment shown inFig. 3, were titrated by the methof fluorescent foci (FF) described recently (Espinoza an

M. Osorio et al. / Journal of Virological Methods 130 (2005) 140–144 143

Fig. 3. Specific visualization of IPNV particles stained by indirect immunofluorescence in filters (A–C) and in cell surface (D). Virus particles retained in0.02�m pore diameter Anodisc filters were stained with monoclonal antibody against VP3 and a secondary antibody conjugated to FITC. Dilutions (1:2:5) ofan IPNV suspension were mixed with various amounts of phage T4 virions. Mixture of viruses containing; (A) 8× 109 ml−1 IPNV particles and 4.2× 109 ml−1

of phage T4 virions. (B) 4.7× 109 ml−1 IPNV particles and 1× 1010 T4 virions/ml and (C) 1.7× 109 ml−1 IPNV particles and 2.8× 1010 T4 virions/ml. (D)Virus was adsorbed at 4◦C to monolayers of CHSE-214 cells grown in glass coverslips. After 1 h, non-adsorbed virus was washed out and the cells fixed andstained as indicated in the text.

Kuznar, 2002). With these figures a ratio of 760 (S.D. = 52,n = 3) physical particles/FF is obtained.

IPNV particles stained with SYBR Green I are brighterthan those labeled with antibodies. This indicates that SYBRGreen I could be used to enumerate physical particles inrather purified preparations. In the case of non-purified prepa-rations, i.e., lysates and contaminated hatchery water, themethod of choice is the antibody mediated fluorescent label-ing of virions. Opposite results were reported when vacciniavirus particles were visualized by conventional fluorescentmicroscopy; DAPI stained virions were seen as faint dots con-trasting with the brighter and bigger fluorescent dots obtainedusing indirect immunofluorescent staining (Vanderplasschenand Smith, 1997). Clearly, the amount of bound fluorophoreby each stained particle can vary depending on its structuralcomplexity, size, whether enveloped or not, the amount ofnucleic acid per virion, accessibility of the dye or the numberof bound antibodies per virion in the case of immunostaining.

These results show that single particles of IPNV can bevisualized by means of epifluorescence microscopy. Depend-ing on the aims sought, the virions can be stained specificallywith antibodies or by a simpler methodology using non-specific stains.

Acknowledgement

The author is grateful to Dr. James Robeson for providingthe samples of T4 and grant DIPUV-Tec 2002.

References

Andrawiss, M., Takeuchi, Y., Hewlett, L., Collins, M., 2003. Murineleukemia virus particle assembly quantitated by fluorescencemicroscopy: role of gag–gag interactions and membrane association.J. Virol. 77 (21), 11651–11660.

Bartlett, J.S., Samulski, R.J., 1998. Fluorescent viral vectors: a new tech-nique for the pharmacological analysis of gene therapy. Nat. Med. 4,635–637.

Dobos, P., 1995. The molecular biology of infectious pancreatic necrosisvirus (IPNV). Ann. Rev. Fish Dis. 5, 25–54.

Dorson, M., 1982. Infectious pancreatic necrosis of salmonids: presentstatus of knowledge concerning the viruses and the possibilities ofcontrolling the disease. Bull. Franc. Piscicult. 305, 61–66.

Espinoza, E., Farias, G., Soler, M., Kuznar, J., 1985. Identity betweeninfectious pancreatic necrosis virus VR-299 and a Chilean isolate.Intervirology 24, 58–60.

Espinoza, J.C., Kuznar, J., 2002. Rapid and simultaneous detection andquantitation of infectious pancreatic necrosis virus. J. Virol. Meth.105, 81–85.

144 M. Osorio et al. / Journal of Virological Methods 130 (2005) 140–144

Leopold, P.L., Ferris, B., Grinberg, I., Worgall, S., Hackett, N.R., Crystal,R.G., 1998. Fluorescent virions: dynamic tracking of the pathway ofadenoviral gene transfer vectors in living cells. Hum. Gene Ther. 9,367–378.

Noble, R.T., 2001. Enumeration of viruses. In: Marine Microbiology, vol.30, John H. Paul (Ed.), University of South Florida, Department ofMarine Science, St. Petersburg, FL, pp. 43–51.

Persson, R., Svensson, U., Everitt, E., 1983. Virus receptor interaction inthe adenovirus system. II. Capping and cooperative binding of virionson HeLa cells. J. Virol. 46, 956–963.

Pizzato, M., Blair, E.D., Fling, M., Kopf, J., Tomassetti, A., Weiss, R.A.,Takeuchi, Y., 2001. Evidence for nonspecific adsorption of targetedretrovirus vector particles to cells. Gene Ther. 8 (14), 1088–1096.

Rodriguez Saint-Jean, S., Borrego, J., Perez-Prieto, S., 2003. Infectiouspancreatic necrosis virus: biology, pathogenesis and diagnostic meth-ods. Adv. Vir. Res. 62, 113–165.

Vanderplasschen, A., Smith, G.L., 1997. A novel binding assay usingconfocal microscopy: demonstration that the intracellular and extra-cellular vaccinia virions bind to different cellular receptors. J. Virol.71, 4032–4041.