Eötvös Lóránd University Biology Doctorate School Classical and molecular genetics program Project leader: Dr. László Orosz, corresponding member of HAS PHYLOGENY AND EVOLUTION OF NEWCASTLE DISEASE VIRUS GENOTYPES Doctoral thesis Alíz Czeglédi Supervisor: Dr. Béla Lomniczi, D.Sc. Veterinary Medical Research Institute of the Hungarian Academy of Sciences Budapest 2007
2 Introduction Newcastle disease virus (NDV), an avian paramyxovirus, is the causative agent of Newcastle disease (ND), a highly contagious ailment that elicits 100% morbidity and mortality in susceptible chickens. ND was one of the most serious problems of the poultry industry in the 20. th century but vaccination introduced in the 1950s, allowed poultry industry to remain profitable even during an unprecedented growth of the number of chickens from 1 to 20 billion in the past 50 years. However, in the developing countries vaccination was not sufficient to eliminate the infection by virulent NDV-strains therefore the disease has become endemic in 80 countries while further 30 suffered one or several introduction of the virus even in the past decade. Genetic analyses, performed in our laboratory, of NDV strains isolated in the past 80 years have revealed the existence of at least 9 genotypes (and further subtypes) that showed not only region specific and host species associations but their temporal distribution was also apparent (Lomniczi és Czeglédi, 2005). It was shown that early genotypes [II.-IV. and Herts 33(W)] prevalent before the 1960s were replaced by recent genetic groups (V.-VIII.) following the introduction of vaccination. Recently sublineages of the Far East genotype VII have spread to other geographic areas, e.g. to Europe. Replacement of genotypes appears to be an evolutionary process rather than random epidemiological event in the distribution of NDV strains. The emergence of novel virulent genotypes seems to be inconsistent with the application of vaccination but experimental infections shed light on the process whereby immunized chicken population became the reservoir of the novel virulent viruses. As to the ecology two major reservoirs of NDV strains exist in nature. The primordial reservoir consists of wild water-bird species that harbour primitive (apatogenic) viruses but, surprisingly, only two genetic lineages are known in the wild: class I and genotype I (belonging to class II). By contrast, the remainder (genotypes II.-VIII.) comprises virulent strains and are maintained in the secondary (artificial) reservoirs of chickens. It is hypothesized that the chicken populations were seeded with apathogenic viruses and pathogenic strains emerged in the chicken host. Prior to the immunization period at least two independent colonisations could take place (with genotype I and II) whereas a novel strategy of generating virulent genotypes must have emerged recently.
3 Aims of the study 1) Phylogenetic analysis of NDV strains to reveal epidemiological relationships and infer evolutionary changes during epizootics: 1a) to reveal the temporal occurrence and replacement dynamics epidemiological types (genotypes and subtypes) in epizootics using virus strains derived from Bulgaria and encompassing four decades; 1b) to estimate of the rate of change of the virus strains under field condition derived from genetically separated endemic lineages in order to assess the approximate time when ND was introduced to Europe; 1c) genetic analysis of vaccine strains to verify the authenticity and origin of these strains based on early publication. 2) Reconstruction of the phylogenetic relationship of the NDV genotypes and subtypes based on the analysis of all genes of NDV strains: 2a) building a database comprising sequences of the 6 genes of 60 representative NDV strains; 2b) to compare the different gene-trees in order to confirm the topology and relationships of the genotypes and subtypes that was established on the basis of a region of the fusion (F) protein gene; 2c) to compare the relative divergence of genes in order to see if surface protein genes are more variables than those coding for internal proteins; 2c) to reveal recombination events based on incongruencies of the different gene trees. 3) Genealogy of NDV genotypes to assess the overrepresentations of virulent groups: 3a) analysis of molecular structural features to use in the grouping above genotypic level; 3b) relationship between genome size classes and the history of virus-host relationships.
4 Materials and Methods Growth of viruses, preparation of viral RNA and reverse transcription These procedures were performed without modification as described previously (Czeglédi et al., 2002). Polymerase chain reaction Partial sequencing of the 6 genes (3 -NP-P-M-F-HN-L-5 ) from 60 NDV strains, and the 5 non-coding region of the NP gene of 23 NDV strains, representing the 9 genotypes and further subtypes was performed. Primers were selected by the OLIGO 5.0 computer program. Complete genome amplification of the PHY-LMV42 virus strain by RT-PCR The genome was amplified in five overlapping portions. Three specific primers were used for the RT and three primer pairs for the amplification of the inner regions encompassing 94% of the genome. The sequences of the 3 - and 5 termini of the viral genome were amplified by 3 - and 5 -RACE. All these procedures were performed as described previously (Czeglédi et al., 2006). Cloning of the amplified products PCR products containing the 5 non-coding regions of 23 NDV strains, and the complete genome of the PHY-LMV42/66 strain were cloned into sequencing plasmids as described previously (Czeglédi et al., 2006). Sequencing and sequence analysis DNA sequences were determined at Genotype GmbH, Hirschhorn, Medigenomix GmbH, Martinsried/München, Germany and at the Agricultural Biotechnology Centre, Gödöllő, Hungary. In case of the complete genome sequencing, primer-walking technology was used. The sequence data of the 60 NDV strains were aligned by the MegAlign program using the CLUSTAL W multiple alignment algorithm. Distance matrix based phylogenetic analysis was performed using the TREECON for Windows 1.3b software, that created a distance tree by the neighbour-joining method using the Kimura two-parameter model and 100 bootstrap values to assign confidence values to topology. MEGA 2.1 was also utilized to estimate the overall averages of the characterized 6 genome regions. Character analysis of the 60 NDV strain were also performed. The best fitting nucleotide substitution model was chosen by the program Modeltest v3.06 program, then phylogenetic trees were reconstructed in PAUP*v.4.0b10 using maximum likelihood approach. Tree topologies were evaluated using a heuristic search approach (Czeglédi et al., 2006).
5 Results and Discussion 1) Partial sequence analysis of the fusion (F) protein gene of NDV isolates deriving from a collection made in Bulgaria and encompassing 4 decades was used to reveal epidemiological relationships and infer evolutionary change. It was estimated that the rate of change of the variable region of the F gene under field condition was 1%/decade. A number of endemic lineages composed of old European strains belonging to genotype IV and fully separated by the 1960s were identified in Bulgaria and other countries in the region. Using the above value to estimate the age of these lineages it was concluded that the most recent common ancestor (or the founder) of these viruses must have been present in Europe not later than the turn of century. This correlates with early documentation of the disease in Europe. A major division of genotype VI was shown which resulted in an Asian and African lineage by the 1970s. 2) Genetic analysis of an authentic sample of the first European isolate, Herts 33(W), revealed that it represented a highly diverged novel early lineage. Contrarily to a 1940 publication from England in which the derivation of strain H, one of the most successful early vaccines, from Herts 33(W) by egg passage was reported, genetic analysis precluded relationships between them. On the other hand, strain Mukteswar claimed to be an independent vaccine, was found to be identical with strain H. 3) The sequence analyses of the 6 genes of 60 representative NDV strains reconstructed basically the same phylogenetic relationships of genotypes and subtypes. All trees were congruent and no signs of recombination were seen at the sequenced regions. A database was established for facilitating genotyping on the bases of any of the genes. 4) Phylogenetic analysis revealed that two major separations resulting in three genome size categories occurred during the history of NDV. An ancient division in the primordial reservoir (wild waterbird species) led to two basal sister clades, class I and II, with genom sizes 15 198 (due to a 12 nucleotide insert in the phosphoprotein gene) and 15 186 nucleotides, respectively. Ancestors of only class II viruses colonized chicken populations and subsequently converted to virulent forms. A second division occurred in the 20th century in the secondary (chicken) host. This gave rise to the branching-off of a clade with the concomitant insertion of 6 nucleotides into the 5 non-coding region of the nucleoprotein gene thereby increasing the genome size to 15 192 nucleotides. In cladistic
6 terms the 6 nucleotide insertion constitutes a synapomorphic character for the recent genotypes while the lack of it the plesiomorphic state. In class I the 12 nucleotide insertion corresponds to an autapomorphic character. 5) Based on the above results two distinct evolutionary mechanisms are proposed for the emergence of lineages comprising virulent strains. The first involves two major steps: independent colonisations of chicken populations with distinct lineages of primitive (avirulent) viruses from the primordial reservoir, which is followed by separate avirulent virulent conversion in the chicken host. The second is less elaborate: surviving a bottle neck effect due to environmental pressure (e.g. immunisation of the host) the virulent ancestor diversifies to further virulent lineages (adaptive radiation). Old genotypes (I.-IV.) appeared to follow the first scenario while recent ones (genotypes V.- VIII.) the second.
7 References 1) Czeglédi, A., Herczeg, J., Hadjiev, G., Doumanova, L., Wehmann, E., Lomniczi, B., 2002. The occurrence of five major Newcastle disease virus genotypes (II, IV, V, VI and VIIb) in Bulgaria between 1959 and 1996. Epidemiology and Infection 129, 679-688. 2) Wehmann, E., Czeglédi, A., Werner, O., Kaleta, E.F., Lomniczi, B., 2003. Occurrence of genotypes IV, V, VI and VIIa in Newcastle disease outbreaks in Germany between 1939 and 1995. Avian Pathology 32, 157-163. 3) Czeglédi, A., Wehmann, E., Lomniczi, B., 2003. On the origins and relationships of Newcastle disease virus vaccine strains Hertfordshire and Mukteswar, and virulent strain Herts 33. Avian Pathology 32, 271-276. 4) Lomniczi, B., Czeglédi, A., 2005. History of Newcastle disease 1. Molecular epidemiology and evolution of Newcastle disease virus. (in Hungarian) Magyar Állatorvosok Lapja 127, 707-719. 5) Czeglédi, A., Ujvári, D., Somogyi, E., Wehmann, E., Werner, O., Lomniczi, B., 2006. Third genome size category of avian paramyxovirus serotype 1 (Newcastle disease virus) and evolutionary implications. Virus Research 120, 36-48.