JOURNAL OF CLINICAL MICROBIOLOGY,0095-1137/99/$04.0010

Dec. 1999, p. 4139–4141 Vol. 37, No. 12

Copyright © 1999, American Society for Microbiology. All Rights Reserved.

Multiplex Reverse Transcriptase PCR Assay for SimultaneousDetection of Three Fish Viruses†

K. WILLIAMS,‡ S. BLAKE, A. SWEENEY,§ J. T. SINGER, AND B. L. NICHOLSON*

Department of Biochemistry, Microbiology and Molecular Biology, University of Maine, Orono, Maine 04469

Received 8 April 1999/Returned for modification 9 July 1999/Accepted 31 August 1999

A multiplex reverse transcriptase (RT)-PCR assay was developed for the simultaneous detection of threedifferent fish viruses: infectious pancreatic necrosis virus (IPNV), infectious hematopoietic necrosis virus(IHNV), and viral hemorrhagic septicemia virus (VHSV). The sensitivity levels of the multiplex RT-PCR assaywere 100, 1, and 32 50% tissue culture infective doses/ml for IPNV, IHNV, and VHSV, respectively.

The aquatic birnaviruses are the largest and most diversegroup of viruses within the family Birnaviridae and include avariety of viruses from numerous species of fish and marineinvertebrates (12, 24). Many of these viruses, such as infectiouspancreatic necrosis virus (IPNV), have been proven or impli-cated to be the etiological agents of diseases in a variety ofspecies used in fish farming and aquaculture worldwide.Aquatic birnaviruses are characterized by a bisegmented, dou-ble-stranded RNA genome within a nonenveloped, icosahedralcapsid approximately 60 nm in diameter (7). The smaller ge-nome segment (B) encodes a single protein (VP1, 90 to 110kDa), the virion-associated transcriptase. The larger genomesegment (A) (approximately 3,000 bp) contains a large openreading frame that encodes a precursor polyprotein (100 kDa)which is subsequently cleaved to form three viral proteins(pVP2, 63 kDa; NS, 29 kDa; and VP3, 29 to 31 kDa) byprotease activity associated with the NS protein (16). The pro-teins are encoded in the following order: pVP2, NS, and VP3.The pVP2 protein is further processed to yield the majorcapsid protein VP2 (50 to 55 kDa). The vast majority ofaquatic birnaviruses, regardless of host species or geographicorigin, are closely related antigenically and form a major se-rogroup (serogroup A) comprising nine serotypes (5, 6, 19).

Infectious hematopoietic necrosis virus (IHNV) is a memberof the family Rhabdoviridae. It causes a lethal disease in severalsalmonid species (23). Epizootics of the disease generally oc-cur in juvenile fish through waterborne horizontal transmis-sion. Mortality levels from IHNV in the most susceptible spe-cies can reach 100%. In chronically infected populations,animals are typically debilitated and secondary bacterial infec-tions are common. Viral hemorrhagic septicemia virus(VHSV) is another rhabdovirus which also causes a seriousdisease in several salmonid species (23). The disease has beenfound in all age groups of susceptible fish, with mortality ratesreaching 80 to 100%, particularly in fry and fingerlings. It haspresented serious problems to rainbow trout culture in Europefor over 50 years. Indeed, VHSV is arguably the most impor-

tant cause of economic loss in the European trout farmingindustry. Moreover, cases of VHSV have been increasinglyreported in other species, including brown trout, grayling, pike,and whitefish.

IHNV is a bullet-shaped virion 150 to 190 nm in length by 65to 75 nm in diameter (11, 23). The enveloped, helical nucleo-capsid contains a negative-sense, single-stranded RNA ge-nome (14). The genome organization for the six separatelytranslated genes is 39-N, M1, M2, G, NV, L-59 (18). The pro-teins encoded by these viral genes are the nucleoprotein (N),matrix proteins (M1 and M2), the envelope glycoprotein (G),a nonvirion protein (NV), and the RNA polymerase (L). Themorphology of VHSV is similar to those of IHNV and otherrhabdoviruses. The virion is approximately 120 to 180 nm longby 60 to 90 nm wide. The viral RNA genome organization andgene products are similar to those of IHNV.

The aquatic birnavirus infectious pancreatic necrosis virus(IPNV) and the rhabdoviruses IHNV and VHSV are signifi-cant pathogens that cause high levels of mortality among arti-ficially propagated populations of fish. Currently, the mosteffective method of controlling diseases caused by these virusesis by prevention of exposure of fish to the virus. Consequently,a number of countries, including the United States, have man-dated fish health management programs that include inspec-tions of artificially propagated fish for the presence of variousfish pathogens, including these three viruses. The currentmethod for detection of fish viruses requires isolation of virusby inoculation of cell cultures with homogenates of tissue sam-ples collected from a statistically significant portion of thepopulation (1). When a virus is isolated, identification of thevirus is required. This usually is accomplished by neutralizationtests with specific polyclonal antisera, which can take 1 to 4weeks, or by enzyme immunoassays with monoclonal antibod-ies or polyclonal antisera (1, 5). However, individual PCRassays for detecting and identifying these fish viruses have beendeveloped (2–4, 6, 8, 9, 15, 21–23). Increasingly, PCR assaysare becoming incorporated into fish health management pro-grams. However, the need to perform separate, individualPCR assays for each virus complicates the diagnostic processand increases the costs. McAllister et al. (17) demonstrated thepotential of using a single PCR assay for the simultaneousidentification of several fish viruses in a single PCR assay.Subsequently, however, it was shown that the IPNV-specificprimers used in their investigation failed to identify all aquaticbirnavirus serotypes (3).

In this investigation, we developed a multiplex PCR assayfor the simultaneous detection of three fish viruses in a singleassay: all serotypes of aquatic birnavirus serogroup A, IHNV,

* Corresponding author. Mailing address: Department of Biochem-istry, Microbiology and Molecular Biology, University of Maine,Orono, ME 04469. Phone: (207) 581-2800. Fax: (207) 581-2801. E-mail: brucen@maine.maine.edu.

† Maine Agricultural and Forest Experiment Station publication no.2369.

‡ Present address: DNA Sequencing Facility, University of Maine,Orono, ME 04469-5735.

§ Present address: Department of Dermatology, University of NorthCarolina, Chapel Hill, NC 27599.

4139

on August 19, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

and VHSV. Routine testing for all three viruses is required inmany fish health management programs worldwide.

Viruses and cell cultures. IPNV (West Buxton strain) andIHNV (unknown electropherotype) were propagated in Chi-nook salmon embryo 214 (13) cell cultures, and VHSV (Spainstrain; provided by Paul Reno, Oregon State University, New-port) was propagated in epithelioma papulosum cyprini (10)cell cultures at 15 to 16°C in Eagle’s minimum essential me-dium with L-glutamine and Earle’s balanced salts solution (Sig-ma Chemical Co., St. Louis, Mo.) supplemented with 0.2%sodium bicarbonate (Sigma Chemical Co.) and 10% fetal bo-vine serum (Atlanta Biologicals, Norcross, Ga.).

RNA extraction and reverse transcription. Total RNA wasextracted from 90 to 300 ml of infected cell culture supernatantwith TriReagent LS (Sigma Chemical Co.) by the protocolsupplied by the manufacturer. Viral cDNA was obtained byreverse transcription by incubating 4 ml of viral RNA prepa-ration with 1 ml of random hexamer primers (1.25 mM randomprimer; Promega) and 2.2 ml of nuclease-free water (Promega)at 80°C for 5 min, followed by cooling to 37°C for 5 min. Thesample was brought to a final volume of 20 ml with a reversetranscription mixture consisting of 4 ml of 53 reverse transcrip-tase (RT) buffer (250 mM Tris-HCl, 375 mM KCl, 15 mMMgCl2, 50 mM dithiothreitol [pH 8.3]; Promega), 200 U ofMoloney murine leukemia virus RT (Promega), 20 U ofRNasin (Promega), and 200 mM each dNTP. The cDNA wassynthesized at 37°C for 1 h, and the reaction was terminated byheating to 95°C for 5 min.

Primers. The aquatic birnavirus-specific primers WB1 andWB2 (WB1, CCGCAACTTACTTGAGATCCATTATGC;WB2, CGTCTGGTTCAGATTCCACCTGTAGTG) and theIHNV-specific primers IHN3 and IHN4 (IHN3, GTTCAACTTCAACGCCAACAGG; IHN4, TGAAGTACCCCACCCCGAGCATCC) had been developed previously in our laboratory(22). Primers WB1 and WB2 (which recognize a 206-bp cDNAfragment within the VP2 gene of aquatic birnaviruses) previ-ously had been shown to identify representative isolates of allnine serotypes of aquatic birnavirus serogroup A (23). PrimersIHN3 and IHN4 recognize a 371-bp cDNA fragment withinthe N gene of IHNV.

Several sets of VHSV-specific primers were designed andevaluated with the aim of developing primers to specificallyidentify VHSV and not produce any amplification productsfrom the other rhabdovirus, IHNV. Bruchhof et al. (4) deter-mined that VHSV and IHNV could not be differentiated basedon genomic sequences of the N gene. Considerable variation inthe matrix protein genes (M1 and M2) has been demonstratedamong the VHSV serotypes, and the NV genes of the NorthAmerican and European strains of VHSV are markedly dis-similar. Little genomic sequence information is available forthe L gene. However, considerable sequence information isavailable for the G gene of VHSV. Therefore, the G gene ofVHSV was chosen as the target for RT-PCR amplification inthis investigation. G gene sequences, including conserved re-gions, were available for the four serotypes from the followingstrains: Makah (representing the North American isolates, se-rotype 1), He-70 (Denmark, serotype 2), 23/75 (France, sero-type 3), and 02-84 (France, serotype 4). These sequences werealigned and used to design primers that would identify com-mon, conserved sequence fragments with the program Oligo(National Biosciences, Inc., Plymouth, Minn.). Ultimately,primers VHS3 and VHS4 (VHS3, CGGCCAGCTCAACTCAGGTGTCC; VHS4, CCAGGTCGGTCCTGATCCATTCTGTC), which amplify a 625-bp cDNA fragment within the Ggene, were selected for use in the multiplex RT-PCR assay.

Multiplex PCR. For multiplex RT-PCR, IPNV, IHNV, andVHSV viral RNA templates were reverse transcribed simulta-neously, and PCR was performed with different combinationsof multiple primer pairs for each virus. Various thermal cyclerprograms were evaluated to optimize the reaction conditions.In the optimal procedure, PCR was performed with 10 ml ofreverse transcribed cDNA in a 50-ml reaction mixture consist-ing of 4 ml of 103 PCR buffer (500 mM KCl, 100 mM Tris-HCl[pH 9.0], 1.0% Triton X-100; Promega), 1.5 mM MgCl2, 0.5 to0.6 mM each primer, and 5 U of AmpliTaq DNA polymerase(Perkin-Elmer Co.). Amplification (40 cycles) was performedin an MJ Research (PTC-100) programmable thermal cycler bythe following protocol: initial denaturation at 94°C for 4 min,denaturation at 94°C for 30 s, annealing at 60°C for 30 s,extension at 72°C for 1 min 30 s, and final extension at 72°C for10 min. Agarose gel electrophoresis was used to separate thePCR products. Agarose gels (2% agarose; SeaKem LE; FMC,Rockland, Maine) containing 0.5 mg of ethidium bromide perml in 13 TAE electrophoresis buffer (40 mM Tris, 20 mMacetate, 2 mM EDTA) were loaded with 12 ml of PCR sampleand electrophoresed at 65 V for 2 h or 100 V for 1 h. EitherPCR Markers (50 to 1,000 bp), 100-bp DNA ladder (100 to2,072 bp), or an HaeIII-digested fX174 DNA (118 to 1,353 bp)molecular weight standard was also electrophoresed on eachgel. A UV transilluminator was used to visualize the bands, andresults were recorded by photography.

Typical ethidium bromide-stained agarose gels of the ampli-fication products of the multiplex RT-PCR assay are shown inFig. 1. Figure 1A shows amplification by the multiplex RT-PCR assay from infected cell culture supernatants of each viraltemplate separately and of all three target templates from amixture of the three viruses. Also, the multiplex RT-PCR assaywas tested for ability to amplify the three viral templates fromuninfected kidney and spleen tissue homogenates from Atlan-tic salmon which had been mock infected with all three viruses.Figure 1B shows amplification of all three viral templates fromboth infected cell culture supernatants and mock-infected tis-sue homogenates.

FIG. 1. Typical agarose gels showing simultaneous RT-PCR identification ofIPNV (West Buxton strain), IHNV, and VHSV. (A) Lane 1, IPNV alone; lane2, IHNV alone; lane 3, VHSV alone; lane 4, multiplex RT-PCR (IPNV plusIHNV plus VHSV); lane 5, HaeIII-digested fX174 DNA marker ladder (118 to1,353 bp). (B) Lane 1, negative cell culture control; lane 2, multiplex RT-PCR(IPNV plus IHNV plus VHSV from infected cell culture supernatants); lane 3,HaeIII-digested fX174 DNA marker ladder (118 to 1,353 bp); lane 4, multiplexRT-PCR (IPNV plus IHNV plus VHSV from mock-infected fish tissue [kidneyand spleen] homogenates).

4140 NOTES J. CLIN. MICROBIOL.

on August 19, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

Sensitivity of multiplex PCR assay. The sensitivity of thePCR assay was determined with 10-fold serial dilutions of eachvirus. RNA was extracted from 90 ml of each sample. RT-PCRwas performed as previously described. The sensitivity wasdetermined from the highest dilution of the sample exhibitinga positive PCR result. The sensitivity levels of the multiplexRT-PCR assay were 100, 1, and 32 50% tissue culture infectivedoses [TCID50]/ml by the method of Reed and Muench [20])for IPNV, IHNV, and VHSV, respectively (Fig. 2). Theselevels of sensitivity are comparable to those of direct isolationof virus in cell culture with dilutions of fish tissue homogenates.

In summary, the multiplex RT-PCR assay, which can becompleted in 48 h or less, provides significant savings in time,cost, and materials in comparison to the antibody neutraliza-tion tests or the separate, individual PCR assays presently usedto identify viruses isolated in cell culture from clinical materi-als. Furthermore, the data presented here and from otherpreliminary investigations (data not shown) in our laboratoryindicate that the multiplex RT-PCR assay compares in sensi-tivity to isolation of virus in cell cultures for the direct detec-tion of these viruses in clinical specimens. Confirmation of thesensitivity of the multiplex RT-PCR assay in field trials usingtissue samples from naturally infected fish will provide a diag-nostic assay with significant advantages over the currentmethod of virus isolation in cell culture, which requires 3 to 4weeks.

This work was supported by grants from the NOAA Sea Grant(R/FMD-217 and R/FMD-235), the NSF EPSCoR Program (EHR91-08766), and the Maine Agricultural and Forest Experiment Station.

REFERENCES

1. Amos, K. H., K. A. Hopper, and L. LeVander. 1989. Absence of infectioushematopoietic necrosis virus in adult sockeye salmon. J. Aquat. Anim.Health 1:281–283.

2. Arakawa, C. K., R. E. Deering, K. H. Higman, K. H. Oshima, P. J. O’Hara,and J. R. Winton. 1990. Polymerase chain reaction (PCR) amplification of anucleoprotein gene sequence of infectious hematopoietic necrosis virus. Dis.Aquat. Org. 8:165–170.

3. Blake, S. L., W. B. Schill, P. E. McAllister, M. K. Lee, J. T. Singer, and B. L.Nicholson. 1995. Detection and identification of aquatic birnaviruses by PCRassay. J. Clin. Microbiol. 33:835–839.

4. Bruchhof, B., O. Marquardt, and P.-J. Enzmann. 1995. Differential diagno-sis of fish pathogenic rhabdoviruses by reverse transcriptase-dependent poly-merase chain reaction. J. Virol. Methods 55:111–119.

5. Caswell-Reno, P., V. Lipipun, P. W. Reno, and B. L. Nicholson. 1989. Use ofa group-reactive and other monoclonal antibodies in an enzyme immunodotassay for identification and presumptive serotyping of aquatic birnaviruses.J. Clin. Microbiol. 27:1924–1929.

6. Chiou, P. P., B. S. Drolet, and J. C. Leong. 1995. Polymerase chain ampli-fication of infectious hematopoietic necrosis virus RNA extracted from fixedand embedded fish tissue. J. Aquat. Anim. Health 7:9–15.

7. Dobos, P., B. J. Hill, R. Hallett, D. T. C. Kells, H. Becht, and D. Teninges.1979. Biophysical and biochemical characterization of five animal viruseswith bisegmented double-stranded RNA genomes. J. Virol. 32:593–605.

8. Einer-Jensen, K., N. J. Olesen, N. Lorenzen, and P. E. V. Jorgensen. 1995.Use of the polymerase chain reaction (PCR) to differentiate serologicallysimilar viral haemorrhagic septicemia (VHS) virus isolates from Europe andAmerica. Vet. Res. 26:464–469.

9. Estepa, A., C. De Blas, F. Ponz, and J. M. Coll. 1995. Detection of trouthaemorrhagic septicaemia rhabdovirus by capture with monoclonal antibod-ies and amplification with PCR. Vet. Res. 26:530–532.

10. Fijan, N., D. Sulimanovic, M. Bearzotti, D. Muzinic, L. O. Zwillenberg, S.Chilmonoczyk, J. F. Vautherot, and P. De Kinkelin. 1983. Some propertiesof Epithelioma papulosum cyprini (EPC) cell line from carp Cyprinus carpio.Ann. Virol. (Paris) 134:207–220.

11. Hill, B. J. 1975. Physico-chemical and serological characterization of fiverhabdoviruses infecting fish. J. Gen. Virol. 27:369–378.

12. Hill, B. J., and K. Way. 1983. Serological classification of fish and shellfishbirnaviruses, p. 10. Abstracts of the First International Conference of theEuropean Association for Fish Pathology. European Association of FishPathologists, Plymouth, England.

13. Lannan, C. N., J. R. Winton, and J. L. Fryer. 1984. Fish cell lines: establish-ment and characterization of nine cell lines from salmonids. In Vitro 20:107–144.

14. Levy, J. A., H. Fraenkel-Conrat, and R. A. Owens. 1994. Virology, p. 77–84.Prentice Hall, Inc., Englewood Cliffs, N.J.

15. Lopez-Lastra, M., M. Gonzalez, M. Jashes, and A. M. Sandino. 1994. Adetection method for infectious pancreatic necrosis virus (IPNV) based onreverse transcription (RT)-polymerase chain reaction (PCR). J. Fish Dis.17:269–282.

16. Manning, D. S., and J. C. Leong. 1990. Expression in Escherichia coli of thelarge genomic segment of infectious pancreatic necrosis virus. Virology 179:16–25.

17. McAllister, P. E., W. B. Schill, W. J. Owens, and D. L. Hodge. 1991. Infec-tious pancreatic necrosis virus: a comparison of methods used to detect andidentify virus in fluids and tissues of fish, p. 191–201. In J. L. Fryer (ed.),Proceedings of the Second International Symposium of Viruses of LowerVertebrates. Oregon State University, Corvallis, Oreg.

18. Morzunov, S. P., J. R. Winton, and S. T. Nichol. 1995. The complete genomestructure and phylogenetic relationship of infectious hematopoietic necrosisvirus. Virus Res. 38:175–192.

19. Nicholson, B. L. 1993. Use of monoclonal antibodies in fish disease research.Annu. Rev. Fish Dis. 1993:241–257.

20. Reed, L. J., and H. Muench. 1938. A simple method of estimating fiftypercent end points. Am. J. Hyg. 27:493–497.

21. Rodriguez, S., M. P. Vilas, M. Alonso, and S. L. Perez. 1995. Study of aviral-dual infection in rainbow trout (Oncorhynchus mykiss) by seroneutral-ization, Western blot, and polymerase chain reaction assays. Microbiol. Se-min. 11:461–470.

22. Sweeney, A. 1996. Multiplex PCR assay for the simultaneous detection ofIPNV and IHNV. M.S. thesis. University of Maine, Orono, Maine.

23. Sweeney, A., S. Blake, J. T. Singer, and B. L. Nicholson. 1997. Detection andidentification of infectious pancreatic necrosis virus (IPNV) and infectioushematopoietic necrosis virus (IHNV) by polymerase chain reaction (PCR),p. 42–47. In Y. Inui and J. Winton (ed.), New approaches to viral disease ofaquatic animals. National Research Institute of Aquaculture, Nansei, Japan.

24. Wolf, K. 1988. Fish viruses and fish viral diseases, p. 83–158, 217–249.Cornell University Press, Ithaca, N.Y.

FIG. 2. Agarose gel showing sensitivity of multiplex RT-PCR assay for de-tection of virus in infected cell culture supernatants. Lanes 1 to 7, IPNV (107.0,106.0, 105.0, 104.0, 103.0, 102.0, and 101.0 TCID50/ml, respectively), IHNV (106.0,105.0, 104.0, 103.0, 102.0, 101.0 and 100 TCID50/ml, respectively), and VHSV (106.5,105.5, 104.5, 103.5, 102.5, and 101.5 and 100 TCID50/ml, respectively); lane 8,HaeIII-digested fX174 DNA marker ladder (118 to 1,353 bp).

VOL. 37, 1999 NOTES 4141

on August 19, 2020 by guest

http://jcm.asm

.org/D

ownloaded from