- Email: [email protected]
Rapid identification of myxoma virus variants by long-range PCR and restriction fragment length polymorphism analysis
Journal of Virological Methods 161 (2009) 284–288
Contents lists available at ScienceDirect
Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet
Rapid identification of myxoma virus variants by long-range PCR and restriction fragment length polymorphism analysis Kevin P. Dalton ∗ , Franziska Ringleb, Jose Manuel Martín Alonso, Francisco Parra Instituto Universitario de Biotecnología de Asturias, Departamento de Bioquímica y Biología Molecular, Edificio Santiago Gascón, Campus El Cristo, Universidad de Oviedo, 33006 Oviedo, Spain
a b s t r a c t Article history: Received 25 March 2009 Received in revised form 24 June 2009 Accepted 29 June 2009 Available online 8 July 2009 Keywords: Myxoma virus Long-range PCR RFLP Strain differentiation
A long-range PCR method directed at the Myxoma virus (MV) left hand and right hand terminal inverted repeats (TIRs) for rapid amplification of genomic DNA and MV isolate differentiation by restriction fragment length polymorphism (RFLP) analysis is described. The efficacy of this method was tested by comparing the results from full genome RFLPs with those from TIRs amplified separately using reference strain Lausanne (Lu) and a field MV strain characterised previously for its virulence in rabbits. The usefulness of this method was also demonstrated by amplifying MV DNA directly from the eyelid tissue of an infected rabbit and comparative RFLP analysis with respect to Lu. The results proved the long-range PCR technique to be a simple highly efficient method for identifying mutations between MV genomes by RFLP analyses of the amplified TIRs and may be used in future studies to identify variable regions for phylogenetic studies. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Myxoma virus (MV) is a member of the Leporipoxvirus genus within the Poxviridae family (ICTVdB, 2002). MV has a large DNA genome (161 kb) containing 171 open reading frames and terminal inverted repeats (TIR) of 11.5 kb (Cameron et al., 1999). The virus is highly contagious and myxomatosis is a major problem for rabbit farmers throughout Europe (Farsang et al., 2003; Kritas et al., 2008) causing major economic losses. The disease is endemic in many countries, with strains of varying virulence levels circulating. Although vaccines are available, the existence of a reservoir of infected rabbits in the wild population and the evolution of new strains makes control difficult (Barcena et al., 2000a). Epidemiological studies would aid greatly in the identification of circulating strains and in tracking the spread of virus. Two methods are used in order to characterise and differentiate MV strains. The first relies on virulence studies which require experimental infection of rabbits: different strains or isolates are identified by measuring the mean survival time of animals. This method differentiates strains into five groups based on virulence (Rosell et al., 2000; Saint et al., 2001). One such study was used to characterise MV isolates circulating in wild rabbits in Spain between the years 1992 and 1995 (Barcena et al., 2000b). From the isolates examined the highly attenuated 6918 strain has been sequenced completely and over 80 mutations were identified. A total of 77.7% of the non-
∗ Corresponding author. Tel.: +34 985 104215; fax: +34 985 103157. E-mail address: [email protected] (K.P. Dalton). 0166-0934/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jviromet.2009.06.026
conservative amino acid mutations and 41% of the total nucleotide changes occurred in the terminal (37.2 kb) regions (Morales et al., 2009). The second method of strain differentiation requires virus isolation and propagation in cell culture, genome purification and RFLP analysis (Kerr et al., 2003; Saint et al., 2001). Many MV isolates in Australia have been characterised by extensive RFLP and sequence analysis (Saint et al., 2001). Such studies are essential for a better understanding of the epidemiology of MV. However, they are laborious, time consuming and difficult to accomplish logistically, because they require the amplification of large quantities of virus. In their study, Saint et al. (2001) analysed 37 Australian MV isolates with 21 restriction endonucleases and identified 15 RFLPs. Of these 15 RFLPs, 9 occurred in the terminal 13.5 kb regions, either in the TIR or in the immediate flanking unique region. The large size of the MV genome complicates the analysis and the interpretation of RFLP results. In order to increase the likelihood of identifying mutations, it is necessary to choose enzymes that cut the genome more frequently, leading to the generation of complex electrophoretic profiles which can give ambiguous results. Kerr et al. (2003) used PCR and sequencing of four small (204–334 nt) regions of MV genomes to monitor the spread of an MV isolate characterised previously and released. In European studies a PCR protocol has been described for the confirmation of the presence of MV (Farsang et al., 2003), but this analysis has not been shown to be effective in epidemiological studies or for the identification of new strains, due to the small target size of amplified product (492 nt) compared to the large size of the viral genome (161 kb).
K.P. Dalton et al. / Journal of Virological Methods 161 (2009) 284–288
Data from the experimental infection of rabbits and the molecular characterisation of MV isolates show that, as is the case for other poxviruses, the terminal regions are more variable while the central regions more conserved. Therefore it was hypothesised that PCR could be used to amplify the terminal regions and the products used directly for RFLP analysis and MV strain differentiation. This report describes the use of a long-range PCR to amplify specifically the entire left or right TIR regions of myxoma virus DNA and unique genes immediately flanking these TIRs. PCR amplification was achieved of 13.5 kb and 23.4 kb from genomic DNA extracted from semipurified virions and of 13.5 kb from DNA extracted from infected rabbit eyelid tissue for use in strain identification and differentiation. The long-range PCR described proved sufficiently sensitive to not require the cultivation of the virus in tissue culture or purification of virions while allowing the differentiation of a field strain by RFLP. 2. Materials and methods 2.1. Cells and virus Myxoma virus (MV)–Lausanne (Lu) strain was grown and titered in RK13 cells. RK13 cells were maintained in Dulbecco’s-modified Eagle medium (DMEM (Gibco)) supplemented with 10% foetal calf serum (FCS-PAA Laboratories, UK) and 80 mg/L gentamicin (Gibco). 2.2. Concentration of MV virions and purification of viral genomic DNA Infected cells were washed with PBS, rinsed briefly with TE9 (10 mM Tris, 1 mM EDTA, pH 9) buffer. Cells were incubated for 15 min at RT in 5 mL TE9 buffer (per T75 cm2 flask) in the cell culture flask. Cells were broken by repeated pipetting (10 times) and cell breakage monitored using light microscopy. This mixture was transferred to a sterile 15 mL tube (BD Biosciences San Jose, CA, USA) and the cell debris was removed by centrifugation at 750 × g for 5 min. The supernatant was transferred to a fresh 15 mL tube and used for the concentration of virus. In order to precipitate the virions one volume of PEG solution (40% PEG and 41 mM NaCl solution (pH 7.2)) was added to 4 volumes of virus containing supernatant, mixed by inversion and incubated at 4 ◦ C (without further agitation) for a minimum of 1 h and then centrifuged at 3000 × g for 1 h. The supernatant was removed and the resultant pellet was resuspended in 150 L sterile H2 O. After treatment with DNaseI (15 units, 1 h, 37 ◦ C; Fermentas, Ontario, Canada) DNA was extracted from precipitated virions using a commercial kit (MasterpureTM Complete DNA and RNA purification kit, Epicentre® Biotechnologies, Madison, WI, USA), following the total DNA purification protocol.
285
1 × DNase buffer; Fermentas) and DNA was extracted using the commercial kit mentioned above (MasterpureTM Complete DNA and RNA purification kit, Epicentre® Biotechnologies). The final DNA pellet was resuspended in 50 L nuclease free water (Promega, Madison, WI, USA). 2.5. Long-range PCR The left hand MV genomic sequence (positions 331–13,225) was amplified using the primers MV331 5 CTAGACGGTACCTATCCTATC3 (genome position 331–351, 161,423–161,443) and MV13225 5 GACAATGACTCCATCTAGGAG-3 (genome position 13,205– 13,225) and the right hand MV genomic sequence (137,061– 161,443) was amplified using the primers MV331 and MV137061 5 GAGAGACGATGCGTGTGTTAAG-3 (genome position 137,061– 137,082). Each primer was used at a final concentration of 0.2 M. Cycle conditions were the same for both reactions and were as follows: 94 ◦ C 2 min, then 35 cycles of 94 ◦ C 30 s, 65 ◦ C 30 s and 68 ◦ C 15 min, with a final extension of 30 min at 68 ◦ C. Reactions were carried out using TaKaRa LA Taq (Takara) in 1× reaction buffer, 2.5 mM MgCl2 and 2.5 mM of each dNTP and 2.5 units of enzyme. Additional oligonucleotides used to amplify regions suspected of containing mutations (as determined by the RFLP analysis) were Mlu04 5 GGAATCTAGATAAGGAACATTG-3 (genome position 4271–4292, 157,482–157,503) with Mlu04C 5 CGTCTTCCCGTAGAAGTC3 (genome position 5001–5018, 156,756–156,773); Mlu02 5 GAATTCCACGCTGATGTAG-3 (genome position 2121–2139, 159,635–159,653) with Mlu02C 5 CGAGTACGAACCACGCTC-3 (genome position 2417–2434, 159,340–159,357); and Seq9a 5 CGAACGTATCATTAGACAATG-3 (genome position 13,219–13,239) with Seq9b 5 CGCAGGTCCACGTATAAACC-3 (genome position 11,482–11,501). Additionally Seq9c 5 GTTCGAAAATGTCCAGATCG3 (genome position 12,568–12,587) was used in sequencing reactions. The cycle conditions used to amplify these regions were 30 cycles of 94 ◦ C 30 s, 55 ◦ C 30 s and 68 ◦ C 30 s, with a final extension of 5 min at 68 ◦ C, using TaKaRa LA Taq and the reaction conditions as described above. 2.6. Gel extraction and RFLP analysis PCR products were analysed by agarose gel electrophoresis. Full size PCR products were cut from gels and purified using the Wizard SV gel and PCR clean-up system (Promega, Madison, WI, USA). Purified PCR products or MV genomic DNA were subjected to digestions using a variety of restriction enzymes (as per manufacturer’s instructions) and analysed by agarose gel electrophoresis in order to look for RFLPs. 2.7. DNA analysis, gels and software
2.3. Virus isolation from tissue samples Veterinary surgeons provided eyelid tissue from rabbits which died from suspected MV infections. Fifty to 100 mg of tissue was diced into small cubes and incubated in 500 L TE9 buffer for 15 min at room temperature. This material was filtered through 0.22 m syringe filters and mixed with 50 L 10× DMEM and used to infect subconfluent monolayers of RK13 cells which were incubated for 4–7 days and monitored for cytopathic effects.
Agarose (SeaKem Gold Agarose (Lonza, Basel, Switzerland)) gels (0.75% in 0.04 M Tris, 1.14% acetic acid and 0.002 M EDTA (TAE)) were run at room temperature and stained with ethidium bromide or with 1× SYBR green stain (Invitrogen, Carlsbad, CA), photographed using the Gel logic 200I system, and images were analysed with a Kodak Molecular Imaging software (Version 4), Kodak (NY, USA). 3. Results
2.4. DNA extraction from tissue samples infected with myxoma virus Fifty to 100 mg of tissue was diced into small cubes and incubated in 500 L TE9 buffer for 15 min at room temperature. Solid material was pelleted by centrifugation (1000 × g, 5 min). The supernatant was digested with DNaseI (15 units, 1 h, 37 ◦ C in
Oligo pairs MV331/MV13225 and MV331/MV137061 were used to amplify the left and right MV (Lu) genomic sequences, respectively. The expected sizes of the products were 13.5 kb for the left TIR and 23.4 kb for the right MV TIR. The position of the oligonucleotides in relation to the EcoRI restriction map (Russell and Robbins, 1989; Saint et al., 2001) and the location of the viral genes amplified are
286
K.P. Dalton et al. / Journal of Virological Methods 161 (2009) 284–288
Fig. 1. Diagram showing EcoRI RFLP fragments corresponding to the MV Lu TIRs and the flanking regions identified by a letter as described in Saint et al. (2001). (A) Also shown is the relative position of the oligonucleotides (black arrows) used in the long-range PCRs. Checkered arrows represent unique genes and solid grey arrows show genes which are repeated in both TIRs. (B) Agarose gel electrophoresis of long-range PCR products corresponding to the left hand (lane L) and right hand (lane R) MV TIRs. Lane M—DNA size markers (Generuler 1 kb plus DNA ladder-Fermentas).
shown in Fig. 1A. The left hand PCR amplified the entire left TIR and the unique M009L gene sequence, while the right hand PCR amplified the entire right TIR and the unique gene sequences M141R to M156R. Fig. 1B shows the agarose gel electrophoresis of the amplified products obtained. Bands of the expected sizes for each PCR were cut from the gel, purified and analysed by RFLP analysis using a variety of restriction enzymes to confirm that the bands purified corresponded to the MV Lu sequence (data not shown). Results from this analysis proved that the bands were DNA amplified from the left hand terminal 13.5 kb and the right hand terminal 23.4 kb regions of MV Lu genomic DNA. Both long-range PCR amplifications gave a minor nonspecific PCR product which was removed by the gel purification process. RFLP analysis of purified MV genomic DNA was carried out on a highly virulent MV strain isolated in Spain and characterised for virulence by experimental infections in rabbits (Barcena et al., 2000b). Strain 87 was used, as a test to determine if conventional RFLP approaches, using purified full-length MV genomes, could be used to differentiate field and reference MV strains and if the methods described above were more simple and efficient while giving rise to results which can be interpreted easily. Viral genomic DNA was purified from virus infected RK13 cells and analysed by RFLP using various restriction enzymes (data not shown). Using this type of analysis one RFLP mutation was identified using Acc65I, which was characterised by sequencing as a mutation (G-T mutation nt 68,719 in genome) in the M072L gene. With the remaining enzymes no RFLPs were detected. To increase the possibility of discovering further RFLPs restriction enzymes were chosen that cut more frequently. However this led to very complex restriction patterns and to difficulties in the analysis and interpretation of results. An RFLP was identified by comparing MluI
digested Lu and strain 87 genomic DNAs (Fig. 2A, lanes Lu and 87, respectively). A band shift from approximately 2–2.7 kb was observed. Even though the presence of this RFLP could be determined the exact nature of the mutation could not be identified easily without cloning a large number of fragments since a total of 19 bands were expected to have migrated between 2 and 3 kb and 6 bands were predicted to have a size of 2 kb. In view of these difficulties, RFLP analysis of the left TIR longrange PCR was investigated to provide a workable alternative for differentiating better these two MV strains. In the MluI analysis of the long-range PCR amplified strain 87 DNA left TIR a RFLP can be observed easily (Fig. 2B, lane 87) resulting from a shift in size (to approximately 2.8 kb) of one of the comigrating bands of 2098 or 2069 (genome coordinates 4605–6674 and 1897–3995) represented by the thicker band observed (black asterisk) in the MluI RFLP analysis of the amplified Lu left TIR (Fig. 2B, lane Lu). A detailed analysis of the MV restriction map (shown in Table 1) identified two potential sites for mutations which could cause a band shift from 2 kb to approximately 2.8 kb. Bands C plus F indicating a mutation in the MluI site at position 4935 or bands B plus F indicating a mutation at position of 4325. Other mutations or combinations of bands were excluded since the predicted sizes of fused bands did not correspond to the observed size. The regions containing the potential mutations were amplified by PCR using the appropriate oligonucleotide pairs (Mlu02/Mlu02C or Mlu04/Mlu04C), sequenced and the mutation identified as (C to T mutation at nt position 4937 in the MV genome). The main benefit of using PCR for the differentiation of MV strains is the sensitivity of the technique and the possibility of working with field samples without previous virus propagation and purification. The long-range PCR of the left hand TIR was
K.P. Dalton et al. / Journal of Virological Methods 161 (2009) 284–288
287
Fig. 2. Genomic RFLP analysis, long-range PCR amplification and RFLP analysis for the differentiation of MV variants. (A) MluI RFLP analysis of purified genomic MV DNA. (B) MluI RFLP analysis of gel purified left hand TIR regions amplified by long-range PCR. (C) Electrophoretic analysis of gel purified left hand TIR long-range PCR amplification products. (D) MluI RFLP analysis of gel purified left hand TIR region amplified by long-range PCR. Lane Lu, Lausanne reference strain. Lane 87, field strain 87 (Barcena et al., 2000b), lanes FS template DNA extracted from 50 to 100 mg of infected rabbit eyelid tissue from an uncharacterised field strain. Lane M corresponded to DNA size markers (Generuler 1 kb plus DNA ladder-Fermentas). The identified RFLPs are indicated with asterisks.
attempted on DNA extracted from a sample of rabbit eyelid which was confirmed previously to contain MV by virus cultivation, immunofluorescence and sequencing (data not shown). Using DNA extracted from 50 to 100 mg of eyelid tissue a band of DNA that corresponded in size to the left TIR was amplified from DNA extracted from laboratory reference strain Lu purified virions (Fig. 2C, compare lanes Lu and FS). This DNA was gel purified and used in RFLP analysis using four restriction enzymes. One of the enzymes revealed an RFLP profile (Fig. 2D) which could be used to differenti-
ate this virus from both Lu and 87 strains. This RFLP was a shift in band size from approximately 0.8 kb to approximately 1.2 kb. Analysis of the predicted restriction map (Table 1) showed that the possible sites containing mutations were located at positions 11,531 or 12,371. The mutation was identified as a deletion of 21 nt, which removed the last nucleotide of the MluI site (genomic position 12,350–12,370), using PCR amplification with the oligonucleotides seq9a and seq9b and sequencing using these oligonucleotides in addition to seq9c.
288
K.P. Dalton et al. / Journal of Virological Methods 161 (2009) 284–288
Table 1 MluI restriction fragment map of MV Lausanne 5 LTR and M009L region. Fragment name
Fragment size
Base positiona
A B C D E F G H I J K L M
3645 2098 2069 1466 840 610 592 422 382 263 252 178 78
7004–10,649 2227–4325 4935–7004 761–2227 11,531–12,371 4325–4935 12,634–end (13,225) 10,649–11,071 11,149–11,531 12,371–12,634 509–761 Start (331)–509 11,071–11,149
a
Base position in MV genome.
4. Discussion Due to the unique nature of its history, MV is a model for the coevolution of host-pathogen ecosystems. Since its release in Europe, where it is the scourge of rabbit farmers and its introduction as a biological control agent in Australia, there has been much interest in the identification of strain variants and the search for mutations with biological significance, both in the search for vaccines and the search and identification of more virulent strains. Epidemiological studies of MV also require the differentiation of strains. Such studies in Europe are lacking but would be useful for identifying circulating strains, the identification of predominant strains, new strains and for tracking their spread. The traditional methods for the analysis and differentiation of MV DNAs are laborious and require large quantities of infected cells for the purification of sufficient quantities of DNA for analysis by RFLP using a wide range of restriction enzymes. This leads to logistical problems when large numbers of strains are analysed. The long-range PCR described above simplifies greatly analysis, provides a sensitive way for testing large numbers of samples without the need for producing large quantities of virus and viral genomic DNA. Using both long-range PCRs a total of 23% of the genome was amplified, and the regions where most RFLPs have been identified in previous, extensive, studies (Morales et al., 2009; Saint et al., 2001) were included. Analysing these “hotspots” for mutations has permitted the identification of mutations which allow the differentiation of a viral strain without the need for cultivating virus in the laboratory. Such analysis proved sufficient to differentiate between closely related strains and may help in tracking virus evolution both in the field in wild rabbit
populations and also in outbreaks on rabbit farms. The mutations identified in this study allow the differentiation of these strains from one another and from the reference strain Lu. However, additional studies will be required to determine if the sites of these mutations can provide useful data for phylogenetic studies. In summary, a long-range PCR is described which when coupled with RFLP analysis can be used to differentiate rapidly between MV strains without the requirement for cultivation of virus and will provide data on variable regions in the MV genome which could be used for future phylogenetic studies. Acknowledgements This work was supported in part by Ministerio de Medio Ambiente y Medio Rural y Marino, INTERCUN and by the INIA grant FAU2006-00214-CO2-01 cofinanced by FEDER. References Barcena, J., Morales, M., Vazquez, B., Boga, J.A., Parra, F., Lucientes, J., Pages-Mante, A., Sanchez-Vizcaino, J.M., Blasco, R., Torres, J.M., 2000a. Horizontal transmissible protection against myxomatosis and rabbit hemorrhagic disease by using a recombinant myxoma virus. J. Virol. 74, 1114–1123. Barcena, J., Pages-Mante, A., March, R., Morales, M., Ramirez, M.A., Sanchez-Vizcaino, J.M., Torres, J.M., 2000b. Isolation of an attenuated myxoma virus field strain that can confer protection against myxomatosis on contacts of vaccinates. Arch. Virol. 145, 759–771. Cameron, C., Hota-Mitchell, S., Chen, L., Barrett, J., Cao, J.X., Macaulay, C., Willer, D., Evans, D., McFadden, G., 1999. The complete DNA sequence of myxoma virus. Virology 264, 298–318. Farsang, A., Makranszki, L., Dobos-Kovacs, M., Virag, G., Fabian, K., Barna, T., Kulcsar, G., Kucsera, L., Vetesi, F., 2003. Occurrence of atypical myxomatosis in Central Europe: clinical and virological examinations. Acta Vet. Hung. 51, 493–501. ICTVdB, 2002. The Universal Virus Database of the International Committee on Taxonomy of Viruses, http://www.ncbi.nlm.nih.gov/ICTVdb/Ictv/index.htm. Kerr, P.J., Merchant, J.C., Silvers, L., Hood, G.M., Robinson, A.J., 2003. Monitoring the spread of myxoma virus in rabbit Oryctolagus cuniculus populations on the southern tablelands of New South Wales, Australia. II. Selection of a strain of virus for release. Epidemiol. Infect. 130, 123–133. Kritas, S.K., Dovas, C., Fortomaris, P., Petridou, E., Farsang, A., Koptopoulos, G., 2008. A pathogenic myxoma virus in vaccinated and non-vaccinated commercial rabbits. Res. Vet. Sci. 85, 622–624. Morales, M., Ramirez, M.A., Cano, M.J., Parraga, M., Castilla, J., Perez-Ordoyo, L.I., Torres, J.M., Barcena, J., 2009. Genome comparison of a nonpathogenic myxoma virus field strain with its ancestor, the virulent Lausanne strain. J. Virol. 83, 2397–2403. Rosell, J.M., Argüello, J.L., Badiola, J.I., Cuervo, L., Vanderkerchhove, D., 2000. Enfermedades víricas, 302–353. Russell, R.J., Robbins, S.J., 1989. Cloning and molecular characterization of the myxoma virus genome. Virology 170, 147–159. Saint, K.M., French, N., Kerr, P., 2001. Genetic variation in Australian isolates of myxoma virus: an evolutionary and epidemiological study. Arch. Virol. 146, 1105–1123.
Contents lists available at ScienceDirect
Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet
Rapid identification of myxoma virus variants by long-range PCR and restriction fragment length polymorphism analysis Kevin P. Dalton ∗ , Franziska Ringleb, Jose Manuel Martín Alonso, Francisco Parra Instituto Universitario de Biotecnología de Asturias, Departamento de Bioquímica y Biología Molecular, Edificio Santiago Gascón, Campus El Cristo, Universidad de Oviedo, 33006 Oviedo, Spain
a b s t r a c t Article history: Received 25 March 2009 Received in revised form 24 June 2009 Accepted 29 June 2009 Available online 8 July 2009 Keywords: Myxoma virus Long-range PCR RFLP Strain differentiation
A long-range PCR method directed at the Myxoma virus (MV) left hand and right hand terminal inverted repeats (TIRs) for rapid amplification of genomic DNA and MV isolate differentiation by restriction fragment length polymorphism (RFLP) analysis is described. The efficacy of this method was tested by comparing the results from full genome RFLPs with those from TIRs amplified separately using reference strain Lausanne (Lu) and a field MV strain characterised previously for its virulence in rabbits. The usefulness of this method was also demonstrated by amplifying MV DNA directly from the eyelid tissue of an infected rabbit and comparative RFLP analysis with respect to Lu. The results proved the long-range PCR technique to be a simple highly efficient method for identifying mutations between MV genomes by RFLP analyses of the amplified TIRs and may be used in future studies to identify variable regions for phylogenetic studies. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Myxoma virus (MV) is a member of the Leporipoxvirus genus within the Poxviridae family (ICTVdB, 2002). MV has a large DNA genome (161 kb) containing 171 open reading frames and terminal inverted repeats (TIR) of 11.5 kb (Cameron et al., 1999). The virus is highly contagious and myxomatosis is a major problem for rabbit farmers throughout Europe (Farsang et al., 2003; Kritas et al., 2008) causing major economic losses. The disease is endemic in many countries, with strains of varying virulence levels circulating. Although vaccines are available, the existence of a reservoir of infected rabbits in the wild population and the evolution of new strains makes control difficult (Barcena et al., 2000a). Epidemiological studies would aid greatly in the identification of circulating strains and in tracking the spread of virus. Two methods are used in order to characterise and differentiate MV strains. The first relies on virulence studies which require experimental infection of rabbits: different strains or isolates are identified by measuring the mean survival time of animals. This method differentiates strains into five groups based on virulence (Rosell et al., 2000; Saint et al., 2001). One such study was used to characterise MV isolates circulating in wild rabbits in Spain between the years 1992 and 1995 (Barcena et al., 2000b). From the isolates examined the highly attenuated 6918 strain has been sequenced completely and over 80 mutations were identified. A total of 77.7% of the non-
∗ Corresponding author. Tel.: +34 985 104215; fax: +34 985 103157. E-mail address: [email protected] (K.P. Dalton). 0166-0934/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jviromet.2009.06.026
conservative amino acid mutations and 41% of the total nucleotide changes occurred in the terminal (37.2 kb) regions (Morales et al., 2009). The second method of strain differentiation requires virus isolation and propagation in cell culture, genome purification and RFLP analysis (Kerr et al., 2003; Saint et al., 2001). Many MV isolates in Australia have been characterised by extensive RFLP and sequence analysis (Saint et al., 2001). Such studies are essential for a better understanding of the epidemiology of MV. However, they are laborious, time consuming and difficult to accomplish logistically, because they require the amplification of large quantities of virus. In their study, Saint et al. (2001) analysed 37 Australian MV isolates with 21 restriction endonucleases and identified 15 RFLPs. Of these 15 RFLPs, 9 occurred in the terminal 13.5 kb regions, either in the TIR or in the immediate flanking unique region. The large size of the MV genome complicates the analysis and the interpretation of RFLP results. In order to increase the likelihood of identifying mutations, it is necessary to choose enzymes that cut the genome more frequently, leading to the generation of complex electrophoretic profiles which can give ambiguous results. Kerr et al. (2003) used PCR and sequencing of four small (204–334 nt) regions of MV genomes to monitor the spread of an MV isolate characterised previously and released. In European studies a PCR protocol has been described for the confirmation of the presence of MV (Farsang et al., 2003), but this analysis has not been shown to be effective in epidemiological studies or for the identification of new strains, due to the small target size of amplified product (492 nt) compared to the large size of the viral genome (161 kb).
K.P. Dalton et al. / Journal of Virological Methods 161 (2009) 284–288
Data from the experimental infection of rabbits and the molecular characterisation of MV isolates show that, as is the case for other poxviruses, the terminal regions are more variable while the central regions more conserved. Therefore it was hypothesised that PCR could be used to amplify the terminal regions and the products used directly for RFLP analysis and MV strain differentiation. This report describes the use of a long-range PCR to amplify specifically the entire left or right TIR regions of myxoma virus DNA and unique genes immediately flanking these TIRs. PCR amplification was achieved of 13.5 kb and 23.4 kb from genomic DNA extracted from semipurified virions and of 13.5 kb from DNA extracted from infected rabbit eyelid tissue for use in strain identification and differentiation. The long-range PCR described proved sufficiently sensitive to not require the cultivation of the virus in tissue culture or purification of virions while allowing the differentiation of a field strain by RFLP. 2. Materials and methods 2.1. Cells and virus Myxoma virus (MV)–Lausanne (Lu) strain was grown and titered in RK13 cells. RK13 cells were maintained in Dulbecco’s-modified Eagle medium (DMEM (Gibco)) supplemented with 10% foetal calf serum (FCS-PAA Laboratories, UK) and 80 mg/L gentamicin (Gibco). 2.2. Concentration of MV virions and purification of viral genomic DNA Infected cells were washed with PBS, rinsed briefly with TE9 (10 mM Tris, 1 mM EDTA, pH 9) buffer. Cells were incubated for 15 min at RT in 5 mL TE9 buffer (per T75 cm2 flask) in the cell culture flask. Cells were broken by repeated pipetting (10 times) and cell breakage monitored using light microscopy. This mixture was transferred to a sterile 15 mL tube (BD Biosciences San Jose, CA, USA) and the cell debris was removed by centrifugation at 750 × g for 5 min. The supernatant was transferred to a fresh 15 mL tube and used for the concentration of virus. In order to precipitate the virions one volume of PEG solution (40% PEG and 41 mM NaCl solution (pH 7.2)) was added to 4 volumes of virus containing supernatant, mixed by inversion and incubated at 4 ◦ C (without further agitation) for a minimum of 1 h and then centrifuged at 3000 × g for 1 h. The supernatant was removed and the resultant pellet was resuspended in 150 L sterile H2 O. After treatment with DNaseI (15 units, 1 h, 37 ◦ C; Fermentas, Ontario, Canada) DNA was extracted from precipitated virions using a commercial kit (MasterpureTM Complete DNA and RNA purification kit, Epicentre® Biotechnologies, Madison, WI, USA), following the total DNA purification protocol.
285
1 × DNase buffer; Fermentas) and DNA was extracted using the commercial kit mentioned above (MasterpureTM Complete DNA and RNA purification kit, Epicentre® Biotechnologies). The final DNA pellet was resuspended in 50 L nuclease free water (Promega, Madison, WI, USA). 2.5. Long-range PCR The left hand MV genomic sequence (positions 331–13,225) was amplified using the primers MV331 5 CTAGACGGTACCTATCCTATC3 (genome position 331–351, 161,423–161,443) and MV13225 5 GACAATGACTCCATCTAGGAG-3 (genome position 13,205– 13,225) and the right hand MV genomic sequence (137,061– 161,443) was amplified using the primers MV331 and MV137061 5 GAGAGACGATGCGTGTGTTAAG-3 (genome position 137,061– 137,082). Each primer was used at a final concentration of 0.2 M. Cycle conditions were the same for both reactions and were as follows: 94 ◦ C 2 min, then 35 cycles of 94 ◦ C 30 s, 65 ◦ C 30 s and 68 ◦ C 15 min, with a final extension of 30 min at 68 ◦ C. Reactions were carried out using TaKaRa LA Taq (Takara) in 1× reaction buffer, 2.5 mM MgCl2 and 2.5 mM of each dNTP and 2.5 units of enzyme. Additional oligonucleotides used to amplify regions suspected of containing mutations (as determined by the RFLP analysis) were Mlu04 5 GGAATCTAGATAAGGAACATTG-3 (genome position 4271–4292, 157,482–157,503) with Mlu04C 5 CGTCTTCCCGTAGAAGTC3 (genome position 5001–5018, 156,756–156,773); Mlu02 5 GAATTCCACGCTGATGTAG-3 (genome position 2121–2139, 159,635–159,653) with Mlu02C 5 CGAGTACGAACCACGCTC-3 (genome position 2417–2434, 159,340–159,357); and Seq9a 5 CGAACGTATCATTAGACAATG-3 (genome position 13,219–13,239) with Seq9b 5 CGCAGGTCCACGTATAAACC-3 (genome position 11,482–11,501). Additionally Seq9c 5 GTTCGAAAATGTCCAGATCG3 (genome position 12,568–12,587) was used in sequencing reactions. The cycle conditions used to amplify these regions were 30 cycles of 94 ◦ C 30 s, 55 ◦ C 30 s and 68 ◦ C 30 s, with a final extension of 5 min at 68 ◦ C, using TaKaRa LA Taq and the reaction conditions as described above. 2.6. Gel extraction and RFLP analysis PCR products were analysed by agarose gel electrophoresis. Full size PCR products were cut from gels and purified using the Wizard SV gel and PCR clean-up system (Promega, Madison, WI, USA). Purified PCR products or MV genomic DNA were subjected to digestions using a variety of restriction enzymes (as per manufacturer’s instructions) and analysed by agarose gel electrophoresis in order to look for RFLPs. 2.7. DNA analysis, gels and software
2.3. Virus isolation from tissue samples Veterinary surgeons provided eyelid tissue from rabbits which died from suspected MV infections. Fifty to 100 mg of tissue was diced into small cubes and incubated in 500 L TE9 buffer for 15 min at room temperature. This material was filtered through 0.22 m syringe filters and mixed with 50 L 10× DMEM and used to infect subconfluent monolayers of RK13 cells which were incubated for 4–7 days and monitored for cytopathic effects.
Agarose (SeaKem Gold Agarose (Lonza, Basel, Switzerland)) gels (0.75% in 0.04 M Tris, 1.14% acetic acid and 0.002 M EDTA (TAE)) were run at room temperature and stained with ethidium bromide or with 1× SYBR green stain (Invitrogen, Carlsbad, CA), photographed using the Gel logic 200I system, and images were analysed with a Kodak Molecular Imaging software (Version 4), Kodak (NY, USA). 3. Results
2.4. DNA extraction from tissue samples infected with myxoma virus Fifty to 100 mg of tissue was diced into small cubes and incubated in 500 L TE9 buffer for 15 min at room temperature. Solid material was pelleted by centrifugation (1000 × g, 5 min). The supernatant was digested with DNaseI (15 units, 1 h, 37 ◦ C in
Oligo pairs MV331/MV13225 and MV331/MV137061 were used to amplify the left and right MV (Lu) genomic sequences, respectively. The expected sizes of the products were 13.5 kb for the left TIR and 23.4 kb for the right MV TIR. The position of the oligonucleotides in relation to the EcoRI restriction map (Russell and Robbins, 1989; Saint et al., 2001) and the location of the viral genes amplified are
286
K.P. Dalton et al. / Journal of Virological Methods 161 (2009) 284–288
Fig. 1. Diagram showing EcoRI RFLP fragments corresponding to the MV Lu TIRs and the flanking regions identified by a letter as described in Saint et al. (2001). (A) Also shown is the relative position of the oligonucleotides (black arrows) used in the long-range PCRs. Checkered arrows represent unique genes and solid grey arrows show genes which are repeated in both TIRs. (B) Agarose gel electrophoresis of long-range PCR products corresponding to the left hand (lane L) and right hand (lane R) MV TIRs. Lane M—DNA size markers (Generuler 1 kb plus DNA ladder-Fermentas).
shown in Fig. 1A. The left hand PCR amplified the entire left TIR and the unique M009L gene sequence, while the right hand PCR amplified the entire right TIR and the unique gene sequences M141R to M156R. Fig. 1B shows the agarose gel electrophoresis of the amplified products obtained. Bands of the expected sizes for each PCR were cut from the gel, purified and analysed by RFLP analysis using a variety of restriction enzymes to confirm that the bands purified corresponded to the MV Lu sequence (data not shown). Results from this analysis proved that the bands were DNA amplified from the left hand terminal 13.5 kb and the right hand terminal 23.4 kb regions of MV Lu genomic DNA. Both long-range PCR amplifications gave a minor nonspecific PCR product which was removed by the gel purification process. RFLP analysis of purified MV genomic DNA was carried out on a highly virulent MV strain isolated in Spain and characterised for virulence by experimental infections in rabbits (Barcena et al., 2000b). Strain 87 was used, as a test to determine if conventional RFLP approaches, using purified full-length MV genomes, could be used to differentiate field and reference MV strains and if the methods described above were more simple and efficient while giving rise to results which can be interpreted easily. Viral genomic DNA was purified from virus infected RK13 cells and analysed by RFLP using various restriction enzymes (data not shown). Using this type of analysis one RFLP mutation was identified using Acc65I, which was characterised by sequencing as a mutation (G-T mutation nt 68,719 in genome) in the M072L gene. With the remaining enzymes no RFLPs were detected. To increase the possibility of discovering further RFLPs restriction enzymes were chosen that cut more frequently. However this led to very complex restriction patterns and to difficulties in the analysis and interpretation of results. An RFLP was identified by comparing MluI
digested Lu and strain 87 genomic DNAs (Fig. 2A, lanes Lu and 87, respectively). A band shift from approximately 2–2.7 kb was observed. Even though the presence of this RFLP could be determined the exact nature of the mutation could not be identified easily without cloning a large number of fragments since a total of 19 bands were expected to have migrated between 2 and 3 kb and 6 bands were predicted to have a size of 2 kb. In view of these difficulties, RFLP analysis of the left TIR longrange PCR was investigated to provide a workable alternative for differentiating better these two MV strains. In the MluI analysis of the long-range PCR amplified strain 87 DNA left TIR a RFLP can be observed easily (Fig. 2B, lane 87) resulting from a shift in size (to approximately 2.8 kb) of one of the comigrating bands of 2098 or 2069 (genome coordinates 4605–6674 and 1897–3995) represented by the thicker band observed (black asterisk) in the MluI RFLP analysis of the amplified Lu left TIR (Fig. 2B, lane Lu). A detailed analysis of the MV restriction map (shown in Table 1) identified two potential sites for mutations which could cause a band shift from 2 kb to approximately 2.8 kb. Bands C plus F indicating a mutation in the MluI site at position 4935 or bands B plus F indicating a mutation at position of 4325. Other mutations or combinations of bands were excluded since the predicted sizes of fused bands did not correspond to the observed size. The regions containing the potential mutations were amplified by PCR using the appropriate oligonucleotide pairs (Mlu02/Mlu02C or Mlu04/Mlu04C), sequenced and the mutation identified as (C to T mutation at nt position 4937 in the MV genome). The main benefit of using PCR for the differentiation of MV strains is the sensitivity of the technique and the possibility of working with field samples without previous virus propagation and purification. The long-range PCR of the left hand TIR was
K.P. Dalton et al. / Journal of Virological Methods 161 (2009) 284–288
287
Fig. 2. Genomic RFLP analysis, long-range PCR amplification and RFLP analysis for the differentiation of MV variants. (A) MluI RFLP analysis of purified genomic MV DNA. (B) MluI RFLP analysis of gel purified left hand TIR regions amplified by long-range PCR. (C) Electrophoretic analysis of gel purified left hand TIR long-range PCR amplification products. (D) MluI RFLP analysis of gel purified left hand TIR region amplified by long-range PCR. Lane Lu, Lausanne reference strain. Lane 87, field strain 87 (Barcena et al., 2000b), lanes FS template DNA extracted from 50 to 100 mg of infected rabbit eyelid tissue from an uncharacterised field strain. Lane M corresponded to DNA size markers (Generuler 1 kb plus DNA ladder-Fermentas). The identified RFLPs are indicated with asterisks.
attempted on DNA extracted from a sample of rabbit eyelid which was confirmed previously to contain MV by virus cultivation, immunofluorescence and sequencing (data not shown). Using DNA extracted from 50 to 100 mg of eyelid tissue a band of DNA that corresponded in size to the left TIR was amplified from DNA extracted from laboratory reference strain Lu purified virions (Fig. 2C, compare lanes Lu and FS). This DNA was gel purified and used in RFLP analysis using four restriction enzymes. One of the enzymes revealed an RFLP profile (Fig. 2D) which could be used to differenti-
ate this virus from both Lu and 87 strains. This RFLP was a shift in band size from approximately 0.8 kb to approximately 1.2 kb. Analysis of the predicted restriction map (Table 1) showed that the possible sites containing mutations were located at positions 11,531 or 12,371. The mutation was identified as a deletion of 21 nt, which removed the last nucleotide of the MluI site (genomic position 12,350–12,370), using PCR amplification with the oligonucleotides seq9a and seq9b and sequencing using these oligonucleotides in addition to seq9c.
288
K.P. Dalton et al. / Journal of Virological Methods 161 (2009) 284–288
Table 1 MluI restriction fragment map of MV Lausanne 5 LTR and M009L region. Fragment name
Fragment size
Base positiona
A B C D E F G H I J K L M
3645 2098 2069 1466 840 610 592 422 382 263 252 178 78
7004–10,649 2227–4325 4935–7004 761–2227 11,531–12,371 4325–4935 12,634–end (13,225) 10,649–11,071 11,149–11,531 12,371–12,634 509–761 Start (331)–509 11,071–11,149
a
Base position in MV genome.
4. Discussion Due to the unique nature of its history, MV is a model for the coevolution of host-pathogen ecosystems. Since its release in Europe, where it is the scourge of rabbit farmers and its introduction as a biological control agent in Australia, there has been much interest in the identification of strain variants and the search for mutations with biological significance, both in the search for vaccines and the search and identification of more virulent strains. Epidemiological studies of MV also require the differentiation of strains. Such studies in Europe are lacking but would be useful for identifying circulating strains, the identification of predominant strains, new strains and for tracking their spread. The traditional methods for the analysis and differentiation of MV DNAs are laborious and require large quantities of infected cells for the purification of sufficient quantities of DNA for analysis by RFLP using a wide range of restriction enzymes. This leads to logistical problems when large numbers of strains are analysed. The long-range PCR described above simplifies greatly analysis, provides a sensitive way for testing large numbers of samples without the need for producing large quantities of virus and viral genomic DNA. Using both long-range PCRs a total of 23% of the genome was amplified, and the regions where most RFLPs have been identified in previous, extensive, studies (Morales et al., 2009; Saint et al., 2001) were included. Analysing these “hotspots” for mutations has permitted the identification of mutations which allow the differentiation of a viral strain without the need for cultivating virus in the laboratory. Such analysis proved sufficient to differentiate between closely related strains and may help in tracking virus evolution both in the field in wild rabbit
populations and also in outbreaks on rabbit farms. The mutations identified in this study allow the differentiation of these strains from one another and from the reference strain Lu. However, additional studies will be required to determine if the sites of these mutations can provide useful data for phylogenetic studies. In summary, a long-range PCR is described which when coupled with RFLP analysis can be used to differentiate rapidly between MV strains without the requirement for cultivation of virus and will provide data on variable regions in the MV genome which could be used for future phylogenetic studies. Acknowledgements This work was supported in part by Ministerio de Medio Ambiente y Medio Rural y Marino, INTERCUN and by the INIA grant FAU2006-00214-CO2-01 cofinanced by FEDER. References Barcena, J., Morales, M., Vazquez, B., Boga, J.A., Parra, F., Lucientes, J., Pages-Mante, A., Sanchez-Vizcaino, J.M., Blasco, R., Torres, J.M., 2000a. Horizontal transmissible protection against myxomatosis and rabbit hemorrhagic disease by using a recombinant myxoma virus. J. Virol. 74, 1114–1123. Barcena, J., Pages-Mante, A., March, R., Morales, M., Ramirez, M.A., Sanchez-Vizcaino, J.M., Torres, J.M., 2000b. Isolation of an attenuated myxoma virus field strain that can confer protection against myxomatosis on contacts of vaccinates. Arch. Virol. 145, 759–771. Cameron, C., Hota-Mitchell, S., Chen, L., Barrett, J., Cao, J.X., Macaulay, C., Willer, D., Evans, D., McFadden, G., 1999. The complete DNA sequence of myxoma virus. Virology 264, 298–318. Farsang, A., Makranszki, L., Dobos-Kovacs, M., Virag, G., Fabian, K., Barna, T., Kulcsar, G., Kucsera, L., Vetesi, F., 2003. Occurrence of atypical myxomatosis in Central Europe: clinical and virological examinations. Acta Vet. Hung. 51, 493–501. ICTVdB, 2002. The Universal Virus Database of the International Committee on Taxonomy of Viruses, http://www.ncbi.nlm.nih.gov/ICTVdb/Ictv/index.htm. Kerr, P.J., Merchant, J.C., Silvers, L., Hood, G.M., Robinson, A.J., 2003. Monitoring the spread of myxoma virus in rabbit Oryctolagus cuniculus populations on the southern tablelands of New South Wales, Australia. II. Selection of a strain of virus for release. Epidemiol. Infect. 130, 123–133. Kritas, S.K., Dovas, C., Fortomaris, P., Petridou, E., Farsang, A., Koptopoulos, G., 2008. A pathogenic myxoma virus in vaccinated and non-vaccinated commercial rabbits. Res. Vet. Sci. 85, 622–624. Morales, M., Ramirez, M.A., Cano, M.J., Parraga, M., Castilla, J., Perez-Ordoyo, L.I., Torres, J.M., Barcena, J., 2009. Genome comparison of a nonpathogenic myxoma virus field strain with its ancestor, the virulent Lausanne strain. J. Virol. 83, 2397–2403. Rosell, J.M., Argüello, J.L., Badiola, J.I., Cuervo, L., Vanderkerchhove, D., 2000. Enfermedades víricas, 302–353. Russell, R.J., Robbins, S.J., 1989. Cloning and molecular characterization of the myxoma virus genome. Virology 170, 147–159. Saint, K.M., French, N., Kerr, P., 2001. Genetic variation in Australian isolates of myxoma virus: an evolutionary and epidemiological study. Arch. Virol. 146, 1105–1123.