Reverse genetics for peste-des-petits-ruminants virus (PPRV): Promoter and protein specificities

Virus Research 126 (2007) 250–255

Short communication

Reverse genetics for peste-des-petits-ruminants virus (PPRV): Promoter and protein specificities Dalan Bailey ∗ , Louisa S. Chard, Pradyot Dash, Tom Barrett, Ashley C. Banyard Institute for Animal Health, Pirbright Laboratory, Ash Road, Woking, Surrey GU24 ONF, United Kingdom Received 10 November 2006; received in revised form 16 January 2007; accepted 21 January 2007 Available online 9 March 2007

Abstract Peste-des-petits-ruminants virus (PPRV) (family Paramyxoviridae, genus Morbillivirus) causes an acute febrile illness in sheep and goats resulting in significant morbidity and mortality in infected herds. The paramyxoviruses all have negative sense, non-segmented RNA genomes and their host range and pathogenic determinants have been extensively studied using reverse genetics. This technology also enables a more rational approach to be taken with respect to vaccine design. In order to initiate this type of work for PPRV we constructed a PPRV minigenome and studied its expression in transfected cells. As for other morbilliviruses, the minimum requirements for minigenome rescue were shown to be the cis-acting elements of the genome (GP) and antigenome (AGP) promoters as well as the three trans-acting helper proteins N (nucleocapsid), P (phosphoprotein) and L (large polymerase). Homologous PPRV helper proteins were compared to their heterologous analogues from the closely related rinderpest virus (RPV) and heterologous minigenome rescue was found to be a much less efficient process. By engineering two GP/AGP chimeric minigenomes we also identified differences between the two viruses in the specific interactions between the promoters and the transcriptase/replicase complexes. The PPRV minigenome was also shown not to strictly comply with the “rule of six” in vitro. © 2007 Elsevier B.V. All rights reserved. Keywords: Peste-des-petits-ruminants virus (PPRV); Minigenome; Reverse genetics; Homologous/heterologous rescue; Chimeric minigenomes; PPRV proteins

Peste-des-petits-ruminants virus (PPRV) causes high mortality in its host species, sheep and goats, and as a result is responsible for serious socio-economic problems in some of the poorest developing countries (Perry et al., 2001). PPRV has a non-segmented negative sense RNA genome and is classified within the Morbillivirus genus of the Paramyxoviridae. Other members of this genus include measles virus (MV), a serious human pathogen, rinderpest virus (RPV), the cause of cattle plague, canine distemper virus (CDV) and phocine, porpoise and dolphin distemper viruses that affect marine species. Morbillivirus genomes encode six transcription units flanked by untranslated regions (UTRs) that control viral transcription and replication. These are referred to as the genome (GP) and antigenome (AGP) promoter, and are found at the 3 and 5 end of the negative sense RNA genome, respectively. Reverse genetics systems have been developed for many paramyxoviruses, including MV, RPV and CDV (Baron and

∗

Corresponding author. Tel.: +44 1483 231066; fax: +44 1483 236448. E-mail address: [email protected] (D. Bailey).

0168-1702/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.virusres.2007.01.015

Barrett, 1997; Gassen et al., 2000; Radecke et al., 1995). These have allowed extensive analysis of the molecular pathways involved in virus replication and the determination of the molecular basis of pathogenicity and host range specificity for these viruses. A similar system for PPRV, however, is not available and is required if we are to improve our understanding of the molecular biology of the virus and this, in combination with other morbillivirus reverse genetics systems, will help to define host range restrictions in these viruses. The first step in the development of a reverse genetics system for PPRV involved the production of a minigenome where the six transcription units were replaced by a single reporter gene, in this case the bacterial chloramphenicol acetyl transferase (CAT) gene. The minimal replicative unit of morbilliviruses has been defined as the negative sense RNA in addition to three proteins; the nucleocapsid protein (N) that is required for encapsidation of the genome, the phosphoprotein (P) that is required to mediate an interaction between N and the polymerase, and the large polymerase (L) that is required for genome transcription and replication (Baron and Barrett, 1997). It has also been established for many paramyxoviruses that effective rescue of both full length and minigenome

D. Bailey et al. / Virus Research 126 (2007) 250–255

251

constructs requires the length of the primary transcript to be divisible by six (i.e. comply to the “rule of six”) as each N protein is believed to bind and encapsidate six nucleotides (Calain and Roux, 1993; Kolakofsky et al., 1998). We therefore constructed a PPRV-based minigenome and studied its expression with homologous and heterologous protein and promoter combinations and its compliance with the “rule of six” in vitro. 1. Construction of the PPRV minigenome and helper plasmids The PPRV GP and AGP promoters were amplified from the RNA of cells infected with the PPRV Turkey 2000 (Tu00) strain using rapid amplification of cDNA ends (RACE). These were amplified using Pfu polymerase (Promega) to incorporate a ClaI restriction site at the end of the GP and a SacII restriction site at the end of the AGP. These elements were sequentially cloned into the previously described pMDB1 plasmid (Baron and Barrett, 1997) to generate the plasmid shown in Fig. 1A. The minigenome was sequenced to confirm the size of the primary minigenome transcript using a Beckmann CEQ8000 automated sequencer. Helper plasmids expressing N, P or L were also produced from RNA extracted from PPRV Tu00-infected cells using Superscript-II (Invitrogen) reverse transcription (RT)-PCR followed by amplification by KOD polymerase (Novagen). These were cloned into the pGEM-3Z vector (Promega), in which they are under transcriptional control of a T7 promoter, to create PPRV Tu00 helper plasmids. The construction of the RPV-based minigenome and helper plasmids have been described previously (Baron and Barrett, 1997; Mioulet et al., 2001). All minigenome assays were carried out in FP-T7-infected (MOI of 0.18) Vero cells by transfecting 1 ␮g of the minigenome DNA with 1 ␮g of the plasmids expressing N and P and 0.05 ␮g of the plasmid expressing L (Das et al., 2000). The pathways involved in CAT production from the minigenome are detailed in Fig. 1B. CAT production was assessed at 72 h post-transfection using a commercially available ELISA system (Roche). Controls in which L was omitted from the transfection showed no CAT production from the transfected minigenome (Fig. 2, panel A) presumably because the absence of polymerase proteins prevents viral transcription and replication of the encapsidated minigenome. Three separate transfections were carried out for each experiment out and a mean value and standard deviation calculated. In agreement with previous findings for other morbilliviruses (Baron and Barrett, 1997; Radecke et al., 1995) the minimal trans-acting elements required for PPRV minigenome rescue were the N, P and L proteins (Fig. 2A). Small amounts of CAT protein were observed in transfections containing the PPRV minigenome without N, P or L, presumably due to the translation of the primary positive sense T7-derived transcript. This was prevented by the inclusion of N and P, and therefore encapsidation of the RNA transcript (data not shown). This description of a functional minigenome rescue system for PPRV follows the recent development of minigenome rescue systems for related viruses such as Borna disease virus (Rosario et al., 2005), Newcastle disease virus (of goose origin) (Zhang et al., 2005) and an RNA

Fig. 1. Schematic of the PPRV and RPV minigenomes and their replication and transcription pathways. (A) Schematic of the RPV and PPRV minigenomes, showing the length of the GP and AGP regions, the restriction sites used for construction of each minigenome and the unique EcoRI restriction site present in the CAT ORF that was used (in conjunction with a unique AflIII site present in the pMDB1 backbone) for construction of chimeric minigenomes. Extra nucleotides were inserted as shown into the SacII site at the beginning of the PPRV AGP to produce minigenome mutants with 1–5 extra nucleotides in order to assess the compliance of PPRV with the paramyxovirus “rule of six”. (B) The functional elements of the PPRV minigenome are as follows: the T7 promoter element (T7p ), the PPRV genome promoter (GP), the antisense (−) chlorampenicol acetyl transferase ORF (CAT), the PPRV antigenome promoter (AGP), the hepatitis delta ribozyme sequence (HD␦) and the T7 terminator sequence (T7t ). Step (i): (a) is transcribed by T7 polymerase (provided by the fowlpox recombinant virus FP-T7) to generate a positive sense (+) primary transcript (b). Step (ii): Co-transfection of the minigenome with the viral N, P and L helper plasmids enables encapsidation and replication of (b) to produce an antisense (−) minigenome that contains the full length GP and AGP sequences (c). Step (iii): The negative sense minigenome is then replicated to produce positive sense full length minigenomes (iiib) or shorter sense viral transcripts (d) lacking the complete GP and AGP (iiia) that act as mRNAs for translation of CAT.

252

D. Bailey et al. / Virus Research 126 (2007) 250–255

Fig. 2. PPRV minigenome rescue and investigation of the rule of six mutants. (A) The mean CAT production and standard deviation (from triplicate transfections assayed by ELISA at 72 h post-transfection) recorded when the pPPRV minigenome of correct hexamer length (6n ± 0) was co-transfected into FP-T7 infected Vero cells with the PPRV helper plasmids expressing N and P or N, P and L. (B) Five non-hexameric minigenome mutants (6n + 1 to +5), containing additional nucleotides at a specific site between the AGP and the CAT ORF, were co-transfected into FP-T7 infected Vero cells. The CAT production (from the +NPL transfections) was assayed at 72 h post-transfection and expressed as a percentage of the CAT recorded for the wild type hexamer length minigenome (pPPRV 6n ± 0). The mean percentage and standard deviation from triplicate transfections is provided. CAT production was not observed for any minigenome co-transfected with only N and P (data not shown).

polymerase I-driven minigenome system for Ebola (Groseth et al., 2005).

2. Compliance with the “rule of six” The PPRV genome is a multiple of six in length (Bailey et al., 2005), and therefore appears to obey the rule of six established for paramyxoviruses that co-transcriptionally edit their P gene transcript. To establish the relevance and degree of flexibility, if any, of this rule for PPRV, mutant minigenomes were constructed with one to five extra nucleotides inserted between the CAT ORF and AGP resulting in non-hexameric transcripts (6n ± 1 to 5) (Fig. 1A). CAT production was again assayed at 72 h posttransfection. The greatest amount of CAT produced was in the wild type pCAT +NPL transfection, where the minigenome transcript is a multiple of six (6n + 0) in length (Fig. 2A). However, there was significant production (greater than 50% of wild type) from the 6n + 1(−5), 6n + 2(−4) and 6n + 5(−1) minigenomes. A small amount of CAT was seen from the 6n + 3(−3) construct and nothing was observed for pCAT 6n + 4(−2) (Fig. 2B). This data deviates from what was expected, and implies PPRV does not strictly obey the rule of six. Equivalent experiments on both hPIV3 and Nipah virus have shown that the rule of six is strictly obeyed even at the minigenome level (Durbin et al., 1997; Halpin et al., 2004). The investigation of the rule of six for PPRV indicates that this virus has a degree of flexibility between its replicase/transcriptase complex and non-hexameric encapsidated minigenome RNA. It is presumed that non-hexameric transcripts are inefficiently encapsidated at the terminus disrupt-

ing both the hexamer phasing of the promoters and the RNA–N protein positioning in the polymerase docking site (Bhella et al., 2004; Lamb and Kolakofsky, 2001). We believe the mechanism of transcription and replication of PPRV favours hexamer length, but will actually tolerate small deviations (+1, +2 and −1 nt). The reason why PPRV does not appear to strictly obey the rule of six is the subject of continued work in our laboratory.

3. Heterologous helper plasmids and minigenome rescue It has been reported for other members of the order Mononegavirales, including the rhabdoviruses and filoviruses, that heterologous helper plasmids (Hoffmann et al., 2003; Muhlberger et al., 1999; Yunus et al., 1999) or helper viruses (Biacchesi et al., 2000; Le Mercier et al., 2002) vary in their ability to encapsidate and transcribe foreign minigenomes. To determine the ability of homologous and heterologous helper plasmids to rescue PPRV, the minigenome rescue was compared to an analogous system for RPV established previously in the laboratory. In initial experiments, FP-T7-infected Vero cells were transfected with either the homologous or heterologous helper plasmid set (RPV or PPRV N, P and L) as well as the RPV minigenome (pRPV) or PPRV minigenome (pPPRV). Although in all instances heterologous rescue was successful, the amount of CAT produced was greatly reduced compared to that in the homologous system. An 80% reduction was seen for pPPRV (Fig. 3A, column 4, blue and green bars) and an 84% reduction for pRPV (Fig. 3A, column 1, blue and green bars) when heterologous helper plasmids were used. This indi-

D. Bailey et al. / Virus Research 126 (2007) 250–255

253

Fig. 3. PPRV and RPV minigenome comparisons. (A) The two wild type minigenomes (pPPRV and pRPV) were compared to each other and to GP/AGP chimeras (pRPVGP–pPPRVAGP and pPPRVGP–pRPVAGP) under the trans-mediated support of either RPV or PPRV N, P and L helper plasmids. FP-T7 infected Vero cells were transfected with these varying combinations of plasmids, and the CAT production from triplicate transfections assayed by ELISA 72 h post-transfection. For each set of helper plasmids, rescued CAT is expressed as a percentage of homologous rescue (i.e. pPPRV and PPRV N, P and L or pRPV and RPV N, P and L). (B) pPPRV and pRPV were assayed with all possible combinations of both PPRV and RPV N, P and L. CAT production was assayed from triplicate transfections at 72 h post-transfection. For each heterologous combination of helper plasmids the CAT produced was expressed as a percentage of homologous minigenome rescue (i.e. pPPRV and PPRV N, P and L or pRPV and RPV N, P and L). The average percentage of homologous rescue is plotted along with the recorded standard deviation for both A and B. CAT production was not observed for any minigenome co-transfected with only N and P (data not shown).

254

D. Bailey et al. / Virus Research 126 (2007) 250–255

cates, unsurprisingly, that homologous helper proteins are more efficient mediators of transcription and replication than their heterologous counterparts. The heterologous restriction may lie in the inability of these helper proteins to form optimal and functional ribonucleoprotein replication complexes with the foreign RNA minigenome (Hoffmann et al., 2003). Interestingly, in previous experiments where RPV was compared to MV and CDV using a similar assay no heterologous restriction was seen. In fact some heterologous helper plasmid sets generated a higher level of CAT then their homologous counterparts (Brown et al., 2005). 4. Roles of the GP and AGP in heterologous rescue To further investigate the reduction in CAT expression observed upon use of heterologous helper plasmid sets, pPPRV and pRPV GP/AGP chimeras were constructed. These chimeras were produced using EcoRI (present in the CAT ORF) and AflII (present in the pMDB1 backbone) to swap the AGP regions between the PPRV and RPV-based plasmids, while maintaining the fidelity of the CAT ORF and the “rule of six”. Chimeric minigenomes were used to assess the role of the promoters in heterologous restriction. A PPRV GP had little effect on the ability of RPV helper plasmids to rescue the chimeric pPPRVGP–pRPVAGP minigenome (Fig. 3A, compare columns 1 and 3 blue bars), whereas a PPRVAGP in the pRPVGP–pPPRVAGP led to a three-fold decrease in CAT production (Fig. 3A, compare columns 1 and 2 blue bars). This suggests that the restriction for RPV lies mostly with the AGP, and therefore the limiting step is either an encapsidation or antigenome to genome replication problem. It is possible that a poor interaction of RPV N with the encapsidation signals in the PPRV AGP increases the amount of soluble N in the cytoplasm. Optimal levels of cytoplasmic N are required for efficient replication and excessive amounts may be deleterious (Chandrika et al., 1995; Muhlberger et al., 1999). The ability of the RPV helper proteins to efficiently recognise a foreign GP could be due to a less specific protein–RNA interaction, interaction of the RPV N protein with a cis-acting signal conserved between PPRV and RPV and within the GP, or indirectly as a result of a more processive polymerase. In contrast, PPRV plasmid rescues were particularly sensitive to a RPV GP (Fig. 3A, compare columns 2 and 4, green bars) and were restricted by a heterologous AGP (Fig. 3A, compare columns 3 and 4, green bars). This non-reciprocity made it difficult to identify the exact cause of the restriction observed using the PPRV helpers but it is clear that the RNA sequence plays a role. 5. Combinations of PPRV and RPV helper proteins vary in their ability to rescue minigenomes The chimeric genome work was extended by carrying out helper protein substitution experiments, in which the dynamics of mixed homologous and heterologous N, P and L-mediated transcription and replication were examined. The PPRV and RPV minigenomes were assayed in conjunction with all combinations of PPRV and RPV N, P and L. None of the combinations

of RPV/PPRV helper plasmids was more efficient at rescuing pRPV or pPPRV than the homologous set (Fig. 3B, columns 1 and 8); however, these experiments did highlight major differences between the N, P and L of RPV and PPRV. Only two combinations, RPV N with PPRV P and L, and RPV N and P with PPRV L (Fig. 3B, columns 2 and 5), rescued both pRPV and pPPRV minigenomes efficiently, representing some complementation within this system. The only other reported example of this type of complementation for morbilliviruses was the rescue of a MV minigenome by MV N and P with RPV L although the level of CAT produced was much lower (Brown et al., 2005). Interestingly, these two functional combinations contain the PPRV polymerase (L) suggesting that this protein does not represent a significant restricting factor during heterologous rescue. An equivalent combination with RPV L showed severe restriction implying the binding-specificity of the PPRV L protein is broader than RPV L (Fig. 3B, columns 2 and 6). The P protein was also identified as a key restricting factor in heterologous rescue. A heterologous P in combination with homologous N and L did not result in significant CAT production (Fig. 3B, columns 3 and 6) confirming the important role of this protein in the transcription/replication complex. In accordance with the observations for L, the PPRV P was shown to have a broader scope of interactions than RPV P (Fig. 3B, columns 2 and 3). While the N protein of RPV could substitute for the N protein of PPRV efficiently (Fig. 3B, columns 2) the opposite was not the case (Fig. 3B, column 7), most probably because of a less effective N–P or N–RNA interaction. This contrasts with the efficient rescue of a chimeric RPV virus containing a substituted N protein from PPRV (Barrett et al., 2003). Perhaps mutations introduced into the PPRV N of this chimeric virus during passaging of the rescued virus removed this restriction and increased the replication efficiency. Interestingly, a heterologous combination of N and P (Fig. 3B, column 2) exhibited variable rescue efficiencies depending on the origin of the RNA (RPV of PPRV). Although closely related viruses were used in the homologous/heterologous comparisons, the RPV helper proteins and minigenome were derived from the vaccine strain RBOK virus whilst the PPRV-based plasmids were derived from a virulent field isolate. Postulated vaccine related changes at positions 5 and 26 in the RBOK GP may account for the inability of PPRV helper proteins to recognise or interact efficiently with the RPV GP (Mioulet et al., 2001). Nucleotides 1–11 have previously been identified in RSV as key elements of the promoter region and intrinsically involved in polymerase recruitment (Cowton and Fearns, 2005). The nucleotide change at position 5 may therefore represent a restriction to wild type polymerases. It is also attractive to postulate that, given the broader specificity of PPRV L, the observed complementation is in part due to variation in the polymerases of vaccine and wild type strains of morbilliviruses. It has been established that, phenotypically, MV polymerases from vaccine strains have greater transcriptional processivity than their wild type equivalents (Bankamp et al., 2002), a fact that might contribute to an overall attenuated phenotype. Interestingly the complete PPRV helper plasmid set can support the rescue of live virus from a full length RPV RBOK

D. Bailey et al. / Virus Research 126 (2007) 250–255

clone (unpublished data) and so a sub-optimal heterologous N, P and L complex can still function in this respect. In conclusion, the efficiency of minigenome rescue is dependant on both protein–protein interactions and RNA–protein interactions. Although the two ruminant morbilliviruses PPRV and RPV can heterogeneously support varying combinations of promoters and proteins, this complementation is restricted and is not reciprocal. References Bailey, D., Banyard, A., Dash, P., Ozkul, A., Barrett, T., 2005. Full genome sequence of peste des petits ruminants virus, a member of the Morbillivirus genus. Virus Res. 110 (1–2), 119–124. Bankamp, B., Kearney, S.P., Liu, X., Bellini, W.J., Rota, P.A., 2002. Activity of polymerase proteins of vaccine and wild-type measles virus strains in a minigenome replication assay. J. Virol. 76 (14), 7073–7081. Baron, M.D., Barrett, T., 1997. Rescue of rinderpest virus from cloned cDNA. J. Virol. 71 (2), 1265–1271. Barrett, T., Parida, S., Mahapatra, M., Walsh, P., Das, S., Baron, M.D., 2003. Development of new generation rinderpest vaccines. In: Brown, F., Roth, J. (Eds.), Vaccines for OIE List A and Emerging Animal Diseases, vol. 114. Dev. Biol., Basel, Karger, pp. 89–97. Bhella, D., Ralph, A., Yeo, R.P., 2004. Conformational flexibility in recombinant measles virus nucleocapsids visualised by cryo-negative stain electron microscopy and real-space helical reconstruction. J. Mol. Biol. 340 (2), 319–331. Biacchesi, S., Yu, Y.X., Bearzotti, M., Tafalla, C., Fernandez-Alonso, M., Bremont, M., 2000. Rescue of synthetic salmonid rhabdovirus minigenomes. J. Gen. Virol. 81 (Pt. 8), 1941–1945. Brown, D.D., Collins, F.M., Duprex, W.P., Baron, M.D., Barrett, T., Rima, B.K., 2005. ‘Rescue’ of mini-genomic constructs and viruses by combinations of morbillivirus N, P and L proteins. J. Gen. Virol. 86 (Pt. 4), 1077–1081. Calain, P., Roux, L., 1993. The rule of six, a basic feature for efficient replication of Sendai virus defective interfering RNA. J. Virol. 67, 4822–4830. Chandrika, R., Horikami, S.M., Smallwood, S., Moyer, S.A., 1995. Mutations in conserved domain I of the Sendai virus L polymerase protein uncouple transcription and replication. Virology 213 (2), 352–363. Cowton, V.M., Fearns, R., 2005. Evidence that the respiratory syncytial virus polymerase is recruited to nucleotides 1–11 at the 3 end of the nucleocapsid and can scan to access internal signals. J. Virol. 79 (17), 11311–11322. Das, S.C., Baron, M.D., Skinner, M., Barrett, T., 2000. Improved technique for transient expression and negative strand virus rescue using fowlpox T7 recombinant virus in mammalian cells. J. Virol. Meth. 89, 119–127. Durbin, A.P., Siew, J.W., Murphy, B.R., Collins, P.L., 1997. Minimum protein requirements for transcription and RNA replication of a minigenome of

255

human parainfluenza virus type 3 and evaluation of the rule of six. Virology 234 (1), 74–83. Gassen, U., Collins, F.M., Duprex, W.P., Rima, B.K., 2000. Establishment of a rescue system for canine distemper virus. J. Virol. 74 (22), 10737– 10744. Groseth, A., Feldmann, H., Theriault, S., Mehmetoglu, G., Flick, R., 2005. RNA polymerase I-driven minigenome system for Ebola viruses. J. Virol. 79 (7), 4425–4433. Halpin, K., Bankamp, B., Harcourt, B.H., Bellini, W.J., Rota, P.A., 2004. Nipah virus conforms to the rule of six in a minigenome replication assay. J. Gen. Virol. 85 (Pt. 3), 701–707. Hoffmann, B., Schutze, H., Mettenleiter, T.C., 2003. Recognition of cis-acting elements of infectious haematopoietic necrosis virus and viral hemorrhagic septicemia virus by homologous and heterologous helper proteins. Virus Res. 93 (1), 79–89. Kolakofsky, D., Pelet, T., Garcin, D., Hausmann, S., Curran, J., Roux, L., 1998. Paramyxovirus RNA synthesis and the requirement for hexamer genome length: the rule of six revisited. J. Virol. 72, 891–899. Lamb, R.A., Kolakofsky, D., 2001. Paramyxoviridae: the viruses and their replication. In: Knipe, D.M., Howley, P.M. (Eds.), Fields Virology, 3rd ed. Lippincott Williams and Wilkins, Philadelphia, pp. 1305–1340. Le Mercier, P., Jacob, Y., Tanner, K., Tordo, N., 2002. A novel expression cassette of lyssavirus shows that the distantly related Mokola virus can rescue a defective rabies virus genome. J. Virol. 76 (4), 2024–2027. Mioulet, V., Barrett, T., Baron, M.D., 2001. Scanning mutagenesis identifies critical residues in the rinderpest virus genome promoter. J. Gen. Virol. 82 (Pt. 12), 2905–2911. Muhlberger, E., Weik, M., Volchkov, V.E., Klenk, H.D., Becker, S., 1999. Comparison of the transcription and replication strategies of Marburg virus and Ebola virus by using artificial replication systems. J. Virol. 73 (3), 2333–2342. Perry, B., McDermott, J., Randolph, T., 2001. Can epidemiology and economics make a meaningful contribution to national animal-disease control? Prev. Vet. Med. 48 (4), 231–260. Radecke, F., Spielhofer, P., Schneider, H., Kaelin, K., Huber, M., Dotsch, C., Christiansen, G., Billeter, M.A., 1995. Rescue of measles viruses from cloned DNA. EMBO J. 14 (23), 5773–5784. Rosario, D., Perez, M., de la Torre, J.C., 2005. Functional characterization of the genomic promoter of borna disease virus (BDV): implications of 3 -terminal sequence heterogeneity for BDV persistence. J. Virol. 79 (10), 6544–6550. Yunus, A.S., Krishnamurthy, S., Pastey, M.K., Huang, Z., Khattar, S.K., Collins, P.L., Samal, S.K., 1999. Rescue of a bovine respiratory syncytial virus genomic RNA analog by bovine, human and ovine respiratory syncytial viruses confirms the “functional integrity” and “cross-recognition” of BRSV cis-acting elements by HRSV and ORSV. Arch. Virol. 144 (10), 1977–1990. Zhang, Y.M., Liu, Y.L., Huang, Y., Jia, L.J., Liu, X.F., 2005. Construction of minigenome of Newcastle disease virus of goose origin and its preliminary application. Wei Sheng Wu Xue Bao 45 (1), 72–76.