general enquiries on this form should be made...

General enquiries on this form should be made to:Defra, Science Directorate, Management Support and Finance Team,Telephone No. 020 7238 1612E-mail: [email protected]

SID 5 Research Project Final Report

SID 5 (2/05) of 30

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

A SID 5A form must be completed where a project is paid on a monthly basis or against quarterly invoices. No SID 5A is required where payments are made at milestone points. When a SID 5A is required, no SID 5 form will be accepted without the accompanying SID 5A.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code SE0767

2. Project title

Investigations of the potential for virulent strains of Newcastle disease virus to emerge from avirulent progenitors

3. Contractororganisation(s)

Veterinary Laboratories Agency, Woodham Lane, New Haw, ADDLESTONE, Surrey

54. Total Defra project costs £ 311,380.00

5. Project: start date................ 01 April 2002

end date................. 31 March 2005

SID 5 (2/05) of 30

6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They

should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.This proposal addressed a commissioned research call (R1) in the statutory and exotic viruses programme. The proposed research was directly relevant to the policy objective of controlling certain diseases that are considered to be such a threat that their control is the subject of legislation. In Great Britain, the notifiable disease, Newcastle disease [avian paramyxovirus type 1] is subject to control by the Diseases of Poultry Order 1994 and the Racing Pigeons (vaccination) Order 1994, in accordance with EU council directive 92/66/EEC by both prophylactic vaccination and slaughter of affected poultry.

Newcastle disease is a serious disease of poultry of which avian paramyxovirus –1 (APMV-1) is the aetiological agent. The disease can spread extremely rapidly, causing up to 100% mortality in susceptible poultry flocks. Whilst many of the mechanisms of the virulence of this virus are understood, the emergence and origins of virulent viruses are unclear. In this proposal, ‘Investigations of the potential for virulent strains of Newcastle disease virus to emerge from avirulent progenitors’ some of these ‘unclear’ areas have been investigated. The main aim of the study was to investigate the potential of avirulent Newcastle disease viruses (NDVs), which are present in wild birds, to act as progenitor strains that have the potential to emerge as virulent viruses.

It is widely accepted that the principle determinant of virulence in NDV is the fusion protein gene, more specifically the motif at the cleavage site of this protein. The fusion protein is synthesised as an inactive precursor protein, requiring proteolytic cleavage in order to acquire infectivity; it is this cleavage activation step, and the sensitivity to cleavage that is the defining step in virulence determination. Virulent viruses have a multiple basic motif followed by a phenylalanine enabling these viruses to be cleaved by a ubiquitous enzyme and have unlimited tissue distribution; avirulent viruses have single basic amino acids and a leucine, meaning they can only be cleaved by trypsin-like proteases and have their distribution restricted to tissues containing these enzymes. However, whilst this is the principle virulence determinant, The variable nature of RNA genomes is a widely researched area, and one that is a main focus of this work. RNA viruses exist as quasispecies; a term used to describe a single virus that population exists as a swarm of different, but very closely related genomes. RNA viruses exist as these swarms of closely related viruses due to the error-prone nature of the RNA polymerase and it is this increased error rate, in relation to DNA replication, and the population of variable genomes within a single virus sample, that gives rise to the question of whether avirulent virus populations have minor sub-populations of virulent viruses ever present within their quasispecies swarm. If this were the case, it would be likely that the virulent phenotype of any minor subpopulation would not be expressed, which is probably due to competition with the more dominant and better-adapted majority avirulent strain. In this example, virulent sub-populations would be maintained within the swarm of genomes that comprise the avirulent quasispecies, and could

SID 5 (2/05) of 30

only rise to dominance (and change the expressed pathotype) in the infecting virus population if the environment changed, allowing a shift in the equilibrium in the population and therefore allowing the virulent virus to dominate. This was investigated in this work by partial sequence analysis of plaque-purified virus clones; fifty clones were picked, of which only 30 could be reisolated after amplification in embryonated eggs, and all except one (which had a single nucleotide synonymous change) were identical to the original sequence.

Developing these hypotheses further, the theory that virulent viruses arise through spontaneous mutations that are selected for when the environment changes was investigated, more specifically that they do not exist as a ‘silent’ sub-population within an avirulent population, but that in a favourable environment, when mutations occur in the replicating avirulent virus genome that predispose, or take a step closer to a virulent genome, the changes are conserved and through a series of mutations a virulent virus can emerge. An example of a change in environment would be a change in host species, e.g. when an avirulent wild waterfowl isolate infects a domestic poultry host. These mechanisms were investigated using two sources of avirulent viruses, an avirulent vaccine strain (V4) and an avirulent isolate acquired from a wild waterfowl (AV229/03).

In this study we developed a model, with in vivo and in vitro stages, that simulated environmental (host) change. It was proposed that by using this system, the stepwise mutations occurring at different passage levels could have been identified. A number of mutations were identified in the post-passage isolates, although no increase in virulence was recorded, differences in receptor binding were identified.

The question of genetic pathogenicity potential and pathotype has also been investigated in this work. Typically, viruses that express the multiple basic (virulent) motif at the fusion protein cleavage site cause virulent Newcastle disease; those with the mono-basic (avirulent) cleavage site motif lead to an asymptomatic or very mild infection. However, there are a group of variant viruses (AMPV-1 lineage 4b) that do not follow this pattern. The viruses responsible for the 1978-present panzootic of AMPV-1 in pigeons are variant APMV-1 viruses and are both antigenically and genetically distinct from the more classical strains. Some of these variant viruses do not exhibit typical pathogenicity pattern and have conflicting results from in vivo pathogenicity tests, where one result defines it as a mesogenic or lentogenic strain and the other indicates a more velogenic pathotype. The genetic basis for this variation has been investigated, focusing on the two surface glycoproteins (fusion (F) and haemagglutinin-neuraminidase (HN)). In this section of the work, a single isolate was selected that was known to exhibit these unusual pathotype properties. From this starting isolate, using a range of passaging and cloning procedures, it was possible to isolate viruses with intracerebral pathogenicity indexes (ICPI) ranging from 0.025 to 1.3, all of which expressed the virulent cleavage site motif. This work may be the first time that a change outside the cleavage site, and one that has occurred naturally and not as a result of genetic manipulation, has an impact on the virulence of an isolate. Mutations in the F and HN were identified, but further work is required to identify their role. This work has demonstrated that the age of the embryos used for the passaging did not affect the potential to increase the virulence of this group of viruses post-passage.

In the final section of work, the phylogenetic relatedness of isolates representing the 16 so far defined sub lineages of NDV have been investigated according to the partial nucleotide sequence of their matrix protein gene. Three pairs of isolates, selected according to their close genetic relatedness and their widely different pathotypes, were included in the dataset to try to identify if more ‘ancestral’ relationships between isolates could be identified when using a more conserved internal gene. This work confirmed that the genetic groupings of NDV based on the fusion protein gene and identified that for most virulent lineages the most closely related avirulent virus was in a different lineage.

From the data obtained and analysed during the course of this study, it is concluded that virulent viruses probably do emerge from avirulent viruses, but only under selective conditions, which are specific to the individual virus strain. It is proposed that avirulent viruses show a range of potential to acquire virulence, where some strains are more likely to acquire virulence under specific conditions, whereas other strains will remain unchanged. In particular, some of the avirulent viruses (probably comprising a specific genetic group) may exhibit an increased likelihood of acquiring virulence when infecting the domestic fowl.

Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with

details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or

SID 5 (2/05) of 30

Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).

SID 5 (2/05) of 30

Objectives1. Investigate the properties of NDV from wild waterfowl and the potential for emergence of virulent viruses from avirulent progenitors.

2. Assess factors that increase pathogenicity as measured using standard tests (ICPI and IVPI)

3. Investigate the potential for virulent NDV to emerge for live vaccine strains following interspecies transmission

4. Phylogenetic analysis of a selected gene encoding an internal protein of closely related virulent and avirulent isolates

All objectives were completed successfully.

This report has been written in three parts as follows:Part 1: Adaptation of avirulent NDV to new hosts: the potential for the emergence of virulent viruses. This is an amalgamation of the data produced for objective 1 and 3.

Part 2: Investigation of the molecular basis for different pathotypes of PPMV-1 1073/98 248VB. This section presents the data generated for objective 2.

Part 3: Phylogenetic analysis of NDV based on partial matrix protein gene sequence. This part presents the data generated for objective 4.

Part 1: Adaptation of avirulent NDV to new hosts: the potential for the emergence of virulent viruses

IntroductionThe ability of a NDV to induce disease in an infected organism is the result of a complex interplay of a multitude of factors that are determined by the biological, biochemical and genetic characteristics of the virus on one hand, and the reactivity of the host on the other. There is evidence that adaptation of a particular NDV strain to a new host may affect pathogenicity, and it has been postulated that virulent and avirulent NDV strains arose from each other by spontaneous mutation (Rott & Klenk, 1988).

The evolutionary success of RNA viruses is likely to be due to their ability to adapt very rapidly to the varying biological niches encountered during spread in a single or multiple host network (Steinhauer & Holland, 1987), facilitated mostly by their high mutation rate. Nearly all large populations of RNA viruses exist as quasispecies: collections of many differing, but closely related genomes; even cloned virus samples inevitably consist of a complex mixture of different but related genomes, all of which must compete during replication of the clone and its progeny (Holland et al., 1992). It is widely recognised that mixtures of virulent and avirulent viruses often co-exist in a single NDV sample, but only the majority phenotype is expressed (Granoff, 1964). This variation has been found regardless of whether the virus population originated from a single virus particle (cloning) or from many. Most sub-populations in an NDV strain exist at ratios to each other that range from 1:1 to 1:50 (Hanson, 1988). Virulent strains of NDV have been found to be more consistently heterogeneous than strains of lower virulence (McMillan et al., 1986). Variation within individual strains of NDV can be assessed according to the ability of the virus to form plaques on cell monolayers, in particular chick embryo fibroblasts (CEF) where plaques can be differentiated according to their morphology (Daniel & Hanson, 1968; Hanson, 1988).

Despite the great heterogeneity reported within RNA virus populations, high mutation rates do not always mean rapid evolution. Under some conditions these rapidly mutating populations can remain quite stable; conversely, under other conditions, extremely rapid evolution can occur if the population equilibrium is disturbed (Steinhauer & Holland, 1987). This relative stasis can be explained by the selection of fit master-sequences in constant environments, but does not mean that these populations are not quasispecies, nor that their mutation frequencies are lower or that they are incapable of rapid evolution in different environments (Holland et al., 1992). Rarely, events can give rise to a more-fit variant, and some degree of straying from the master sequence can occur. A change in environmental conditions that offers variants

within the population the opportunity to compete favourably with the predominating virus can shift the equilibrium and drive virus evolution. The fastest way to disrupt stable virus equilibrium is to change its adaptive landscape, which happens frequently in nature through host immune response, infection of new host or tissues or interference by defective interfering particles (Holland et al., 1992).

Amplification and passage of a virus sample through specific selective media can increase the likelihood of the emergence of virus populations with altered properties from the parental strain. Specific selection methods have been used to disrupt the equilibrium within a sample and selectively amplify more pathogenic variants in avian influenza and NDV samples and include passage in embryonated eggs (either 9 or 14-day-old), passage through chicks, in vitro passage, plaquing efficiency, chemical and temperature treatments.

Work with avian influenza has utilised passage of low pathogenicity avian influenza (LPAI) viruses in tissue culture, 14-day-old embryonated fowls’ eggs and in chickens to encourage the emergence of highly pathogenic avian influenza virus strains from the low virulence precursors. Ultimately, the success of the different selective approaches was influenced principally by the individual virus strain. Passage in 14-day-old embryos was based on the theory that these may provide a replicative advantage to minority components of a virus population that had increased pathogenicity (Brugh & Beck, 1993). Whilst the basis for this age effect was initially unknown, it was suggested that cellular protease availability may change during the course of embryo development; if the proteases responsible for the activation of the high pathogenicity avian influenza virus variants was at a lower concentration in the younger embryos then this would provide a competitive advantage to other subpopulations not dependant on this protease.

Later work suggested that the selective properties present in 14-day-old embryonated fowls’ eggs might be due to differences in anatomical or protease properties or urate quantity. Haemagglutinin degradation in the allantoic fluid of older eggs was observed, whereas it was not in younger eggs. Based on these results it has been proposed that the virulent viruses may invade a broader range of embryo cells and thereby evade degradation, whereas the avirulent viruses remained in the fluid and were degraded, therefore affording a replicative advantage to the more virulent viruses (Horimoto & Kawaoka, 1995). This concept was confirmed by Rott & Klenk (1988), where they demonstrated that infection of the chorioallantoic membrane in embryonated eggs led to a systemic infection with velogenic viruses, whereas multiplication of the lentogenic strains was restricted to the cell layer that was inoculated.

Avirulent NDV will only replicate in some tissue culture lines if supplied with an exogenous source of trypsin. Therefore, growth of avirulent NDV in medium without trypsin is strongly selective for virulent variants within the parental population. A further method of selection based on this technique is plaque selection, also in the presence and absence of trypsin, where those with the higher plaquing efficiencies have greater virulence. Selection of plaques according to their morphology can assist with the identification of populations with varying biological properties, and possibly virulence potential.

There have been many studies that report the altered phenotype of a parent NDV strain after laboratory manipulation. In a number of examples the aim has been to attenuate the virulence of a strain with a view to potential vaccine development. Conversely, in other studies the aim was to increase disease pathogenicity and investigate virus virulence. Increased virulence has been achieved by laboratory manipulations including: treatment of avirulent vaccine strains with the lipid solvent chloroform (Kaleta et al., 1980); passage in BHK cells (Slosaris et al., 1989); chemical mutagenesis (Pritzer et al., 1990); intracerebral passage in chicks (Islam et al., 1994; Shengqing et al., 2002), passage in chickens (Kommers et al., 2003) and reverse genetics (Peeters et al., 1999; deLeeuw et al., 2003). There are also similar examples where low pathogenicity influenza isolates have increased their pathogenicity by: growth of clones in the absence of trypsin and the passage in chickens (Brugh & Perdue, 1991); intra-cerebral passage in chicks (Ito et al., 2001); passage in the 14-day-old embryonated eggs (Horimoto & Kawaoka, 1995) and through passage in trypsin-free tissue culture (Ouchi et al., 1989). In the last two examples the same low pathogenicity virus was used and the virulent viruses that emerged had acquired their increased virulence by two different mechanisms. The same virus was present immediately preceding an outbreak of highly pathogenic avian influenza (HPAI) in the field and was proposed as the precursor. The method of virulence acquisition was the same as that in the 14-day-old embryonated egg system, which would seem to suggest that the selective pressures involved with the generation of highly pathogenic avian influenza viruses in 14-day-old eggs are more similar to those that exist in nature, than those responsible for the selection of virulent sub populations in trypsin-free tissue culture systems.

There are two field examples where it seems likely that a virulent strain of NDV emerged from an endemic avirulent strain. The first example was in Ireland in 1990, the second was in Australia in 1998. In the first example, two outbreaks of ND occurred in chickens in Ireland in 1990; the viruses isolated in both were highly virulent and apparently identical (Alexander et al., 1992). Genetic and antigenic analysis of these two

samples proved them to be most closely related to isolates of low virulence that generally infect wild waterfowl (Collins et al., 1998). Further, these viruses formed a group that was antigenically and genetically distinct from other, more classical, strains of APMV-1. The virulent virus was found to differ by four nucleotides at the site coding for the F0 cleavage site, when compared to genetically similar viruses of low virulence (Table 1). These cleavage site mutations account for the increased virulence of these isolates. Based on the uniqueness of these isolates it is proposed that the virulent virus emerged by mutation from the avirulent virus, the method of how, and indeed in which host is, however, unclear.

In the second example, phylogenetic analysis of the viruses responsible for the outbreaks of ND in Australia in the late 1990s, proved the virulent strains to be very closely related to a virus of low virulence, also isolated from chickens in the same vicinity. Based on the sequence at the F0 cleavage site, the virulent viruses differed from the avirulent virus by only two nucleotides (Table 1). Retrospective analysis of the events preceding and during the emergence of these virulent NDVs in Australia has led to the suggestion of the presence of an endemic avirulent strain with an unstable genome, possibly represented by the ‘Peat’s Ridge’ strain, which due to its genetic sequence has an increased likelihood of mutating to virulence (Kirkland, 2000). It has been suggested that these highly virulent ND viruses are likely to have evolved from an endemic avirulent strain that had been circulating in Australia for some time; further substantiated by the identification of two transitional or intermediate isolates (Westbury, 2001), and that effective control of NDV will be reliant on surveillance to look for the distribution of this ‘unstable’ (and other similar) viruses, and then consider the possibility of eradicating it.

If, as has been suggested by these two examples, virulent viruses do emerge from avirulent precursor viruses then this could have important implications for the current strategies in place for the control of NDV, particularly in light of the widespread use of live vaccines. The overall aim of the work described in this part of the project was to investigate the genetic variation that may arise in the surface glycoproteins of an avirulent NDV when passaged through a new host medium. In particular with a view to identifying changes that may lead to an altered virus phenotype, specifically the pathotype. Avirulent NDVs are endemic in many countries worldwide, where they are perpetuated and maintained in wild waterfowl. The origins of virulent viruses are unclear but it is widely accepted they emerge through the acquisition of sporadic random mutations (Rott & Klenk, 1988). This idea was investigated in the present study, using in vitro models and genetic analyses of progeny viruses. The work focused on adapting an avirulent isolate to a new host environment and to determine whether a virus with an altered (possibly virulent) phenotype or genotype emerged.

Methods

Acquisition of isolatesThree different clinical specimens of NDV obtained from wild waterfowl (WB isolates) were acquired from Esther Shihmanter (teal: Israel), Poul Jorgensen (mallard: Denmark) and Ron Fouchier (mallard: Netherlands). Vaccine strain (V4) was supplied by the ND reference laboratory at VLA Weybridge.

Passage in embryonated eggsViruses were passaged in 9-to 18-day-old embryonated Pekin duck eggs or 9-to 14-day-old embryonated fowls’ eggs, using a standard protocol.

Rapid tissue culture screen for putative virulence markerThe growth of the original WB and progeny WB viruses (pre- and post-egg passage) in confluent MDCK cells in 96 well plates with and without the addition of 1ug/ml trypsin was used as an indicator of phenotype. Serial ten-fold dilutions (10-1 to 10-10) of virus were prepared; 25µl of each dilution was inoculated per well in a 96 well plate, 8 replicate wells for each dilution. The plate was overlaid with media, half of the wells had media with trypsin, half was without. After 72 hours of incubation the plates were frozen then thawed. Each well was tested for HA activity. The lowest HA positive dilutions in the presence and absence of trypsin were recorded. These two values were used for each isolate to assess the effect of trypsin on the growth of virus.

Standard 50oC PCRThis is the standard protocol for the amplification of the first ~500 nucleotides of the fusion protein gene, which includes the cleavage site, using an MSF1 cDNA template, and primers MSF1 and Taq 2. The protocol is as described in Aldous et al., 2003.

Cleavage site analysis screen for altered virulence The nucleotide sequence of the cleavage site was used as an indicator of changed pathogenicity. Using the standard protocol for determining the nucleotide sequence of the 3’ end of the F gene, single direction

sequence was determined using primer 2 only (Aldous et al., 2003). From this nucleotide sequence the amino acid motif at the cleavage site was deduced, allowing the pathogenicity of the isolate to be estimated.

Whole F and HN gene PCR and nucleotide sequencing systemA PCR system for the amplification of the whole gene sequence of the two surface glycoproteins, the fusion (F) and haemagglutinin-neuraminidase (HN) genes, was designed and optimised. The F gene was amplified using primer MSF1 as the cDNA and forward PCR primer and FCR3 as the reverse PCR primer; the HN was amplified using FCF12 as the cDNA primer and FCF6 and HNCR5 as the forward and reverse PCR primers. All primer details are listed in Table 2. The reaction components and cycling parameters are as the generic PCR protocol; the cycling parameters are the same, except the annealing temperature in the PCR was increased to 57°C. The product sizes for the two PCRs are 1867 nucleotides for the F and 2253 nucleotides for the HN.

Cloning by plaque purificationBiological cloning of viruses was achieved by plaque purification in CEF, DEF or DIV-1 cells. Serial 10-fold dilutions of virus were prepared and 200µl of each dilution was inoculated, in duplicate, onto confluent cell monolayers grown in a 6-well plate and overlaid with media either with or without added trypsin. The plates were incubated at 37°C for 3 to 7 days. The plaques were visualised by adding 1ml of a solution of 0.05% (w/v) neutral red (NR) in sterile water into each well. After this time the cells were stained using neutral red. Plaques were ‘picked’ by choosing well-separated plaques and stabbing each through the agar layer with a sterile Pasteur pipette. This removed an agar plug (with some virus infected cells attached), which was flushed into 100µl of sterile PBS and used directly to inoculate an embryonating fowls’ egg to amplify the virus.

HA tests with non-chicken blood cellsHA tests were done using 1% v/v red blood cell suspensions with mouse, rat, rabbit, guinea pig, turkey, sheep and horse bloods. The whole bloods were received as a 50% v/v suspension in Alsevers solution. The cells were washed, pelleted and resuspended three times with 0.1M PBS. The cells were resuspended in 0.1M PBS as a 1% (v/v) solution.

Intracerebral passageTwo one-day-old chicks were used for each passage. 0.05ml of a 1:10 dilution of infectious allantoic fluid in sterile PBS was inoculated intracerebrally into each chick. Chicks were monitored daily; three days after infection the chicks were euthansed and the brains were removed. The brain tissue was cut up finely and made up to 5ml with virus dilution medium and shaken vigorously. 100µl of this brain homogenate was inoculated into three 9-day-old embryonated fowls’ eggs. The eggs were candled daily and chilled and harvested four days after infection. The allantoic fluid was harvested and HA titre determined. The allantoic fluid was then used for the next intracerebral passage, this was repeated six times.

ResultsGrowth of NDV in duck eggsEmbryonated Pekin duck eggs, 9 to 18 days-old, were determined to be suitable for the amplification of virulent and avirulent strains of NDV, using the same existing inoculation and harvesting procedures as those in place for fowls’ eggs.

Sourcing clinical materialThe clinical sample of NDV from a wild waterfowl was obtained through the International NDV Reference Laboratory at VLA. Of the three samples that were submitted, only the final sample, AV229/03 mallard Netherlands, yielded infectious virus, as determined by PCR and HA tests. The sample was acquired from wild mallard, swabbed as part of an influenza surveillance project in The Netherlands, from this point onwards this will be referred to as isolate WB. A freeze-dried vaccine sample (V4) was supplied by the International NDV reference laboratory at VLA.

Initial amplification and characterisation of AV229/03 swabThe original WB swab was eluted into 1ml VDM and 100µl was inoculated into three 12-day-old embryonated Pekin duck eggs. After four days these were chilled and the allantoic fluid was harvested, confirmed as positive using an HA test, pooled and frozen at –70oC in aliquots of 1ml.

WB was characterised as: monoclonal type GQ, genetic group 1, ICPI 0.00, MDT >96 hours

Surface glycoproteins PCR development

Using alignments of F and HN gene sequence data available on GenBank, potential primer regions were identified for amplification of the F and HN genes. Primers were designed or selected from the existing pool available in this laboratory. Primer combinations were evaluated and PCR optimisation carried out. A range of different conditions and components were evaluated (data not shown), but ultimately, the protocol used in the standard 50°C programme was adopted, with an increased annealing temperature (57°C). Two separate PCR reactions were designed to amplify the two surface glycoprotein genes independently.

Tissue culture screenSeven cell lines were evaluated (MDCK; BHK-21; NPTR; NSK; VERO; MDBK and CEF) for their sensitivity to trypsin and their ability to support the growth of the viruses used in this study. It was established that 0.5µg/ml trypsin was the best concentration for primary cell cultures (CEFs and DEFs) and 1µg/ml for the cell lines (data not shown). The presence or absence of virus in each well was confirmed by the detection of HA activity. Of the cells tested, only MDCKs restricted growth of isolate WB completely at the dose used in the absence of trypsin. The results are presented as a single value, which is the ratio between the total number of virus positive wells in the presence and absence of trypsin (number of HA positive wells with trypsin/ number of HA positive wells without trypsin) (Table 3a & b)). Values of 1, indicate even growth with, or without trypsin; values greater than 1 indicate increased growth with trypsin; values less than 1 indicate more growth in cells without trypsin. Simply, the further the results are from 1, the greater the difference in growth between the two trypsin treatments. Obviously, in cases where no virus is detected in either of the treatments, a value could not be produced.

Cleavage site screeningIn addition to the screening of isolates according to their growth in tissue culture, the nucleotide sequence across the cleavage site was also determined (Table 3 a & b). The sequence was identical for all isolates screened, unchanged from that of the original swab isolate.

Nucleotide sequencing for detection of mixed samplesA series of five synthetic mixtures of virus were made up and tested. Two viruses, one virulent and one avirulent, were amplified by PCR using the standard 50°C PCR F gene protocol and quantified using a DNA gradient. The synthetic mixtures were made up using the diluted purified DNA as detailed in Table 5. The software assigned bases as a virulent cleavage site in mix 1-3 and avirulent in 4 and 5. Manual inspection of the data however showed clear extra peaks, clearly distinguishable from background noise and pull-ups, where there were base changes between the two sequences. Therefore, using the sequencing method described here, backed up by manual editing, it is possible to identify samples that are composed of both avirulent and virulent viruses. In this example, the mixture could only be detected in mix 3, composed of 175fg of virulent and 300fg of avirulent virus.

Determination of relative sensitivity of detection methodsUsing a dilution range of infectious allantoic fluid from 10-1 to 10-10 it was possible to detect virus using the standard PCR method at 10-4 dilution; using the embryonated eggs system, HA activity was detected in eggs at 10-8 concentration.

Virus titreThe EID50 and TCID50 of the viruses were determined according to the method by Karber (1931). Based on this method, 1 EID50 of the p1 duck egg amplified WB material was calculated as 10-8.38 and 1 TCID50 (in MDBK cells + trypsin) as 10-6.5. Vaccine virus was supplied from as a seed stock; the TCID50 of the egg-amplified p1 material was calculated as 10-5.6 in MDBK cells.

Passage schedulesBoth viruses were passaged in two forms: firstly as the original sample, and secondly as plaque-purified clones. Initially, the WB swab was amplified by a single passage in three 12-day-old embryonated Pekin-duck eggs to yield sufficient sample volume of allantoic fluid for all subsequent analysis. The freeze-dried vaccine seed isolate was resuspended in 1ml dH2O and inoculated into three 9-day-old embryonated fowls’ eggs; the allantoic fluid from these eggs was harvested after four days. These materials were the starting point for all further work. The selected passage schedule for this work was to be 10 serial passages in 14-day-old fowls’ eggs. Six WB derived-plaques and four vaccine-derived plaques were selected from the third round of plaquing and were harvested and stored. This was the material used for the 10 serial passages in embryonated eggs. Each isolate was passaged 10 times in two or three embryonated eggs at each pass.

Unplaqued sampleFor the unplaqued WB isolate, several amplification steps were carried out prior to the serial passaging to generate a pool of potential starting materials, from which the isolates for the serial passaging could be selected. The proposed schedule of pre-serial passage amplification steps is detailed in Table 6. It was

anticipated that these amplification steps might allow the introduction of a small amount of variation into the chosen starting samples for the serial passaging in 14 day-old embryonated fowls’ eggs. The final samples used in the serial passaging were selected according to highest HA titres and time to kill eggs. Of the WB passage ten (p10) isolates, one (AB6) was selected, on the basis of the MDCK ratio, to be passaged a further 6 times intracerebrally in day-old chicks.

Plaque purified clonesThree rounds of plaquing in duck embryo fibroblast cells (DEFs) and chick embryo fibroblast cells (CEFs) were completed for the WB and vaccine sample respectively. At each round, a variety of different sized plaques were picked and inoculated directly into embryonated eggs. Post-harvest the HA titre for each egg was determined, and selected isolates were re-plaqued. Isolates were selected for re-plaquing according to the size of the plaques and the HA titre.

Screening of passage ten (p10) virusesAll of the p10 WB viruses were screened using the tissue culture system and cleavage site nucleotide sequencing (Table 3 a & b). Two p10 isolates, from both of the plaqued and unplaqued passage series, were selected for further characterisation. There were no clear examples of isolates exhibiting probable increased virulence based on the results of either of the screens. AB6 and L1 were selected from the unplaqued samples because their MDCK ratios were the two lowest values, and the AB6 ratio had decreased between passage 0 and 10. In the plaqued samples, PQ64 and PQ68 were selected since both had reduced MDCK ratios between p0 and p10, and both had positive HA activity at relatively high dilutions (10-6 and 10-5 respectively).

Haemagglutinating propertiesAny change in the haemagglutinin binding of the p10 WB isolates from the original swab was assessed by HA tests using 1% v/v red blood cell preparations from different host species (chicken, mouse, rat, rabbit, turkey, sheep, horse and guinea pig). Of the red blood cells tested, mouse had non-specific agglutination with all isolates, rabbit had non-specific binding with the negative egg fluids, but showed some variation in binding with the virus samples. Guinea pig red blood cells were agglutinated by all virus samples. Based on binding or not binding as the indicator (since the titre of each virus sample was not measured and likely to be quite variable), the sheep and horse red blood cells showed the most selectivity. The p10 isolates L1, IC6, PQ64, PQ68, A, B, AB3, L9, L14 all gained the ability to agglutinate horse red blood cells whereas PQ64 and PQ68 p0 viruses lost the ability to agglutinate sheep red blood cells, but by p10 had reverted back and regained the binding capacity.

Nucleotide sequencing of the surface glycoproteinsThe F and HN genes of the two selected isolates from the plaqued and unplaqued series of both the WB and vaccine samples were sequenced to allow identification of any regions that have mutated during the course of the passaging. The nucleotide substitutions identified are detailed in table 4 a & b.

Cleavage site analysis of cloned isolatesFifty clones were isolated from plaques formed on monolayers of DEFs infected with the p1 ‘original’ WB material. Of these fifty initial plaques, only thirty could be reisolated after amplification in duck eggs. The partial nucleotide sequence for the F gene for each amplified clone was determined using standard RT_PCR and automated nucleotide sequencing protocols. The sequences were arranged into a single data set and aligned; no variation was identified except in one isolate (WBPQ68) at position 348 in the alignment, which was heterologous C/T. This is a synonymous (non-coding) change.

DiscussionIn this section of work a number of techniques have been developed and evaluated, which have been essential in this piece of work, but will also be useful tools in future research projects. It has been determined that 9- to 18-day-old embryonated duck eggs do support the growth of virulent and avirulent NDV, and can be used for the isolation of NDV from clinical material. The PCR system developed to amplify the surface glycoprotein genes and the primers designed to determine the nucleotide sequence of these products were designed to be suitable for use with a broad range of isolates, as well as the sample used here, for potential use in future projects. The growth and plaquing of NDV in a wide range of cells has been thoroughly evaluated, and working protocols have been written to reflect these findings.

The acquisition of an infective clinical sample from a wild duck proved to be quite a task in itself. Samples from wild birds are generally taken as part of surveys, and in most cases, when virus has been detected and amplified the retained infective material is allantoic fluid. These samples were unsuitable for this study since we needed previously unpassaged clinical material. Furthermore, of the three clinical samples submitted, despite all having been confirmed as virus positive prior to transit, infectious virus could not be detected or

amplified from 2 samples on receipt, thereby demonstrating that the isolation of viable NDV from submissions of clinical material can be problematical, particularly post-transit. This was surprising, since it has been reported that NDV can remain viable in faeces for long periods (Lancaster, 1966; Alexander, 1988b).

MDBK cells have been reported to offer the restrictive environment required for the proposed tissue culture screen to be developed in the present study, specifically that they would not support the amplification of avirulent viruses without the addition of exogenous trypsin. However, with the WB sample, after passage in the absence of trypsin the virus could still be detected. These results were confirmed when the experiments were repeated. It is likely that the virus particles causing HA activity in the trypsin negative wells were the uninfectious progeny virus from this first round of infection, with uncleaved fusion glycoproteins. Quantifying the virus to be screened and adding an infectious dose of considerably less than one multiplicity of infection could have overcome this problem. Since a positive HA result requires about 106 virus particles; infecting with a virus dose lower than this would mean for a positive HA result to be recorded multiple cycles of replication would be required, which would require infectious progeny to be produced. However, since the criteria for this assay were to be simple and rapid, the necessity to quantify each virus before inoculation to ensure a low dose was deemed unfeasible. It was considered more appropriate to select a cell line that would facilitate detection of this particular virus at the existing titre. When more cell types were investigated, using the same dose and titre of virus as that used for the MDBK cells, MDCK cells were found to offer the best restriction of growth, no virus could be detected in the wells infected with WB sample without trypsin after passage one. MDCK cells are a well-characterised and robust cell line that is readily available – therefore fitting with the aim of this screen to be simple. Predictably however, as the virus dose increased (titre increased by serial passaging) virus could be detected by HA assay in the trypsin-negative wells. This was overcome by using the two treatments to generate a simple ratio. The total number of HA positive trypsin positive wells, divided by the number of HA positive trypsin negative wells. Since virulent viruses grow to similar levels in MDCKs in the presence or absence of trypsin, then the closer the value is to 1, the less the effect of trypsin on the growth of the virus.

The agglutination of red blood cells from different mammalian species by NDV has been investigated (Ito et al., 1999), where it was found that isolates obtained from wild waterfowl had the broadest receptor range and could agglutinate red bloods cells from all species tested, whereas with viruses from chickens and other hosts agglutination varied between strains. The use of blood cells from different mammalian species in the present study has been used in a comparative context. The results are not intended to make a statement about the binding properties of the WB isolate, more it was a means to compare directly the receptor binding properties between the viruses generated in the present study.

The F and HN mutations identified in the p10 gene sequences in the isolates passaged in the present study are as detailed table 4 a & b. The proposed functions of the different regions in these two glycoproteins are detailed on Figure 1. The majority of mutations seen in the in vitro adapted WB virus used in the present study were located in the HN, which can be divided into four fairly distinct regions: cytoplasmic tail, transmembrane domain, stalk and globular head. The latter contains the antibody recognition sites, receptor recognition and neuraminidase active site (Wang & Iorio, 1999), whereas the transmembrane and cytoplasmic domains interact with the fusion glycoprotein to enhance fusion.

Examination of changes in the WB sample: the 394F mutation is located in the fusor peptide (amino acid 131), an essential region of the gene and therefore is not surprising that this is a synonymous change. Mutation 711F is non-synonymous (gly>glu) and corresponds to amino acid 237, located between the two heptad repeats, 1 and 3 in the fusion protein. Mutation 250HN a synonymous mutation and corresponds to amino acid 83, in the stalk region of the HN. The remaining 6 nucleotide mutations are all non-synonymous and correspond to amino acids 221, 229, 230, 307, 488, 524 all falling within the globular head region. The specific functions of these mutations have not been determined, however, since they are located in the globular head, it is probable that they have implications in receptor binding. Precise definitions of the effect of these mutations could be resolved through the use of reverse genetics, and analysing all aspects of the recombinant virus phenotype. For the purpose of this study however, it is clear that these mutations have no effect on the virulence of the virus, as determined by ICPI. They may influence the receptor binding, as indicated by variable growth patterns in MDCK cells and agglutination patterns of mammalian red blood cells.

In this section of this project, serial passaging of avirulent NDVs (isolated from clinical material obtained from a wild mallard and vaccine strain V4, and plaque purified clones derived from them,) through 14-day-old embryonated fowls’ eggs and 9- to 18-day-old embryonated duck eggs respectively and by intracerebral passage in chicks (WB only) is reported. No difference in virulence could be detected between the original and the passaged material as detected by ICPI tests. Further, no difference in mAb binding pattern could be observed between the isolates either. Differences could be observed however, in the growth of the isolates in

MDCK cells in the presence and absence of trypsin, agglutination of red blood cells from different host species and in the genetic composition of their surface glycoprotein genes. These results are in clear contrast to other studies where the virulence of isolates appears to have increased post-passage. In the two ‘natural’ examples of where virulent NDV emerged from avirulent precursors, the ‘passaging’ occurred in the field and the virulent virus is proposed to have emerged from a closely related avirulent precursor virus (Alexander et al., 1992; Gould et al., 2001). Clearly, since these examples occurred ‘in the field’ complete passage history of the samples is not available and a number of assumptions have had to be made.

In the first example, the outbreak of virulent ND in Ireland in 1990, the virulent virus was found to be very closely related to an avirulent virus with similar antigenic properties isolated during the same period and at the similar location. The virulent virus had a cleavage site that was 3 nucleotides different from the avirulent strain, generating the virulent cleavage motif (Alexander et al., 1992). No intermediate examples of viruses exhibiting stepwise progression between these two isolates were identified. This is not proof for them not existing however, since it is likely that these ‘intermediate’ viruses would have been present in very low numbers in this population of viruses and therefore their detection would have required a thorough and detailed surveillance of a large number of samples from similar hosts prior to and during the time of the outbreak of virulent virus.

In the second example, in Australia in the late 1990s, the virulent virus responsible for outbreaks of disease was found to be genetically most closely related to an endemic virus of low virulence (Gould et al., 2001). In this case, intermediate viruses were identified (Table 1). The ‘Peat’s Ridge’ (progenitor) virus was isolated in 1998, about 2 weeks prior to the outbreak of virulent (neurological) disease, from poultry presenting with respiratory disease. Wide spread and thorough surveillance was then carried out to determine the extent of spread of this (Peat’s Ridge) strain (Westbury, 2001). In these thousands of surveillance isolates, two were identified as unusual according to their time taken to kill embryos, and the Somersby and PR-32 strains were identified (P. Selleck, personal communication). By sequencing a large number of clones of the PR-32 isolate, it was established that about 25% had the typical virulent motif (RRQRR*F). Phylogenetic analysis revealed close association between these outbreak viruses and two previously isolated avirulent Australian viruses, NSW 12/86 and Qld 1/87 (Table 1). It is suggested that there is a strong likelihood that the Peat’s Ridge strain evolved from these isolates (Gould et al., 2003; Gould et al., 2001). Based on these data, it has been widely accepted that these virulent viruses emerged from the endemic avirulent strain.

Figure 1 HN and fusion protein structure

1. Fusion protein (adapted from Lamb & Kolakofsky, 1996; and (Sergel et al., 2000))

2. HN protein structure and function (adapted from Deng et al, 1995)

cleavage site (117)

NH2 COOH

transmembrane domain (26-48)

Cytoplasmic Tail (0-25)

Stalk (49-143)

Globular head(144-571)

COOH

F2 F1signal

sequence(11-31)

NH2

FusorPeptide

(117-136)

HeptadRepeat 1

HeptadRepeat 2

TransmembraneDomain (501-527)

HeptadRepeat 3(268-289)

F2 F1

However, the factors involved that lead to the emergence of these virulent strains are unclear, as indeed are the precise origins of the avirulent progenitor strain.

Both of these examples seem to support the hypothesis that virulent viruses do emerge from avirulent precursor viruses, possibly as a consequence of the mutation and selection that occurs when a virus enters a new host and adapts. This could well hold true for both of the above examples if the origins of the avirulent viruses were found to be from non-domestic species. Since the reservoir for avirulent NDVs is principally wild waterfowl, this would offer the change of host proposed when these viruses cross the species barrier and infect domestic poultry.

During the course of this work, two other groups have published data on this subject, as already stated, with quite different results from those published in the present study (Shengqing et al., 2002; Berinstein, personal communication, 2004). In the first example, Shengqing et al., (2002) report the generation of virulent NDV from an avirulent waterfowl isolate by passaging in chickens. In this study, the avirulent isolate Goose/Alaska/415/91 was serially passaged 9 times through the air sacs of 3-day-old chicks. The passage 9 material was then passaged intracerebrally five times in day-old chicks, and was found to have an increased ICPI (1.88) and tissue distribution in the host.

In the second example, an avirulent NDV obtained from a wild swan was passaged twenty times in embryonated fowls’ eggs. The MDT of the p20 material was reduced, and produced illness in the inoculated chickens, whereas no disease signs were seen in birds inoculated with the original virus. Cleavage site analysis revealed a change from the original isolate motif in the p5 isolate, with no further changes in the F gene sequence recorded thereafter (Berinstein, 2004 personal communication).

The heterogeneity of virulence acquisition is clearly emphasised in these examples, where three different laboratory passaging protocols have led to the emergence of two virulent viruses. The role that the brain passage plays in the emergence of virulent viruses is unclear. It has been suggested that the brain provides a highly selective environment for virulent viruses (Islam et al., 1994; Ito et al., 2001; deLeeuw et al., 2003). Its role as one of the principal factors in the selection of virulent viruses seems reasonable, however, from the results discussed so far it is clearly not the only factor. Looking at previous studies, when a mesogenic NDV was passaged intranasally and intracerebrally in parallel, only the brain passaged isolates developed velogenic properties, while the intranasally passaged virus remained unchanged (Islam et al., 1994). In a further example where reverse genetics was used to construct ‘intermediate’ cleavage site motifs in a previously virulent virus, a single intracerebral passage led to all mutants reverting to virulence. However, in the two previous passages in eggs the mutant viruses replicated to similar titres as the wild type, and maintained their mutant sequences. It was proposed that the revertants were present after the first passage in eggs, but only passage in the brain provided the selective environment necessary for them to emerge and become the dominant phenotype (deLeeuw et al., 2003). Conversely, as mentioned above, Berenstein, (2004) completed just five passages in embryonated fowls’eggs and a virulent virus emerged. In the example of the present study, six intracerebral passages after 10 passages in 14-day-old embryonated fowls’ eggs did not lead to the emergence of a virulent virus.

It is widely accepted that many host, virus and environmental factors influence virulence. Whilst is has been demonstrated that the principal determinate of virulence in NDV is the cleavage site motif, and its sensitivity to cleavage by proteases, there are many other factors intrinsically involved that also affect it. Using reverse genetics, Peeters et al, (1999) introduced three nucleotide changes to a La Sota (lentogenic) genome and changed the cleavage site to a typical virulent motif. The pathogenicity of the mutant strain increased from 0 to 1.28, which although considered virulent, could still be more virulent since the scale goes from 0-2, and there are isolates that give a result of 2 or very close e.g. Herts/33 (Alexander, 2003), leaving the question of what factors within the genome would need to change to realise the full virulence. Therefore, in the present study, despite the passage 10 isolates not having a virulent cleavage site, it is possible that other (as yet unseen) adaptations and changes may have occurred. The host and virus factors that may influence the development in virulence could be many.

The use of 14-day-old embryos for the selection of viruses with higher pathogenicity has been used successfully for avian influenza (Brugh & Beck, 1993; Horimoto & Kawaoka, 1995). Brugh and Beck (1993) show that older embryos provide an environment where viruses with increased plaquing efficiency (a marker closely linked with virulence) can evolve to predominance. Horimoto & Kawaoka (1995) demonstrated that low pathogenicity viruses passaged through 14-day-old eggs developed increased virulence through the acquisition of an arginine in the cleavage site, or loss of a glycosylation site nearby. The properties of the 14-day-old eggs that provide this environment are unclear, though it has been proposed that virulent viruses may be able to invade more cells and thereby evade degradation, whereas the avirulent viruses cannot. However, despite this method being appropriate for the selection of influenza viruses with increased

pathogenicity, it was not suitable for the selection of virulent strains of NDV from the avirulent virus used in the present study.

For a new (virulent) strain to emerge from an existing (avirulent) strain, something in the environment of the virus needs to change sufficiently to favour the replication of any potential (virulent) sub-population existing within the quasispecies swarm of related genomes that comprises the original virus population or arising by mutation subsequent to a change in the environment. Through these favourable (selective) conditions, the previously minor sub-population could evolve and replicate to become the dominant strain and change the phenotype. Clearly in this instance the 14-day-old embryonated eggs did not provide this (virulent biased) selective environment, and consequently the virus remained avirulent; equally neither did the intracerebral passage, since the virus isolated post passage was also avirulent. It is possible that the egg amplification step between each intracerebral passage was sufficient to maintain the avirulent proportion of the population, and the avirulent phenotype. Repeating this work and omitting the egg amplification step between each of the intra-cerebral passages could clarify this result. There is a risk however, that if these viruses could not replicate in the brain they may be lost at passage 2.

Whilst the aim of this piece of work was to investigate the potential methods by which virulent virus may emerge, it is recognised that the schedules available and chosen probably do not closely mimic what would happen naturally. This model system was designed to allow the replication of virus in a system with different selection pressures. It is important to consider that disease pathogenicity is only one sign that the virus is changing, or adapting to its new niche or environment. From the point of view of poultry health and welfare, it is probably considered the most important, and logistically its one of the easiest to measure, but there will be many other changes, with more subtle effects that are likely to be difficult or impossible to record.

The emergence of virulent viruses is likely to be a highly complex process, for the cleavage site motif of the WB isolate to mutate and acquire a virulent motif a minimum of three mutations of the F0 cleavage site would have been required. These would either need to be simultaneous, or consecutive and persist through sequential replication cycles as isolates with ‘intermediate’ cleavage site sequences. It is possible that these intermediate viruses may not replicate as well as the progenitor virus, and therefore would be selected against. In both examples, the acquisition of a virulent cleavage site motif by random nucleotide mutations is likely to be an uncommon event. This is illustrated in previous studies completed by this laboratory (SEO762, TDP 3003), where the majority of outbreaks have been caused by reappearances of previous virulent viruses that have persisted and then been re-introduced into the domestic poultry hosts; the number of clear examples where an avirulent virus has led directly to the emergence of a virulent virus are few.

This work, in concert with the other similar studies discussed, illustrates the variability of NDV and the factors that affect their virulence. From the data obtained and analysed during the course of this study, it is concluded that virulent viruses probably do emerge from avirulent viruses, but only under selective conditions, which are specific to the individual virus strain. It is proposed that avirulent viruses show a range of potential to acquire virulence, where some strains are more likely to acquire virulence under specific conditions, whereas other strains will remain unchanged. In particular, some of the avirulent viruses (probably comprising a specific genetic group) may exhibit an increased likelihood of acquiring virulence when infecting the domestic fowl.

Future workIt is unfortunate that the genetic lineage information is not available for the isolates used in both of the two studies where a virulent virus did emerge from an avirulent precursor virus. It would be interesting acquire the original virus in both these examples to sequence them, firstly to gain information on their genetic lineages, but also to identify any other changes that may have occurred in the glycoproteins, with a view to identifying their role by reverse genetics. Following on from this, having determined the genetic lineage it would be interesting to repeat the work using other, very closely related viruses in order to determine whether the potential to acquire virulence is linked to their genetic composition and conserved within lineages, or whether it is a more random process, determined by individual factors associated with each specific virus/host interaction.

Development of reverse genetic techniques is the subject of on-going and future work at VLA. Once developed, the mutations identified in this study, and others identified in the current ROAME looking at virulence markers, will be analysed according to their function and whether they have a role in the emergence of virulent viruses.

Part 2: Investigation of the molecular basis for different pathotypes of PPMV-1 1073/98 248VB

IntroductionCurrently, the pathotyping of Newcastle disease virus isolates is carried out by the intra-cerebral inoculation of day-old chicks. An intracerebral pathogenicity index or ICPI is calculated based on the clinical signs (and the speed of their onset) and ranges from 0.00 to 2.0. An index of >0.7 is indicative of virulent isolates whose control is the subject of legislation; compulsory slaughter can mean compensation claims that run into millions of pounds.

The nucleotide sequencing of many hundreds of NDV isolates has demonstrated that there is a correlation between the amino acid motif at the cleavage site of the fusion protein and the ICPI. Virulent isolates are characterised by one or two pairs of basic amino acids in this region and a phenylalanine residue and have an ICPI value of greater than 0.7; avirulent isolates have single basic amino acids and leucine and an ICPI value of less than 0.7. These observations have led to the motif at the cleavage site being included in the OIE definition of virulent NDV, a definition that once accepted by the European Union should mean a reduction in level of in vivo tests to determine pathogenicity.

The basis for this research is that in recent years some isolates belonging to a variant group of APMV-1 viruses, which are genetically and antigenically distinct from other strains, known as pigeon paramyxovirus -1 (PPMV1) have displayed unusual pathogenic properties. These isolates, despite having a cleavage site motif characteristic of virulent isolates, exhibit ICPIs that are typical of avirulent or the lentogenic strains. Moreover, it is possible that these isolates of low ICPI (<0.7) may become virulent after different passage procedures. This affect has been observed in low pathogenicity avian influenza isolates (with a virulent cleavage site motif) on passage in 14-day-old embryonated fowls’ eggs and PPMV-1 isolates on passage in chickens. Such isolates, if cloned, are excellent candidates for molecular studies aiming to determine the molecular basis for the observed changes in pathogenicity.

In this report we describe the passage procedures used to obtain isolates of varying virulence from an initial sample with a virulent cleavage and low ICPI, the RT-PCR procedure adopted to obtain a 4000bp fragment encompassing the HN and F genes and the sequencing of these products. Nucleotide sequences were compared and an analysis done with respect to molecular change and its possible impact on the observed change in pathotype.

Materials and MethodsOriginal test virus: AV1073/98 248VB submitted to the reference laboratory by Meulemanns, has an ICPI of 0.32 and was obtained from a pigeon in Belgium. The different passage procedures (i), (ii), (iii) and (iv) used to obtain four cloned pathotypes are summarized in Table 7. Briefly, they are:

(i) 1073/98 248VB was passaged three times in 9-day-old embryonated fowls’ eggs at limiting dilution. Pathotype was confirmed by a further passage at limiting dilution in 9-day-old embryonated fowls’ eggs.(ii) Allantoic/amniotic fluid from one egg of the third passage that tested positive in procedure (i) was passaged five times by limiting dilution in 14-day-old embryonated fowls’ eggs. Pathotype was confirmed by a further passage at limiting dilution in 9-day-old embryonated fowls’ eggs. (iii) 1073/98 248VB passaged five times in 14-day-old embryonated fowls’ eggs at low dilution, followed by single passage at low dilution in 9-day-old embryonated fowls’ eggs then passaged three times at limiting dilution in 9-day-old embryonated fowls’ eggs.(iv) 1073/98 248VB passaged five times at low dilution and then four times at limiting dilution in 9-day-old embryonated fowls’ eggs.

The ICPI of virus derived from each procedure was determined using a standard protocol.

RNA extraction, RT-PCR and sequencing Following extraction of viral RNA from infective allantoic fluid using the reagent TRIzol®LS, RT-PCR and sequencing was carried out using the primers shown in Table 2. First strand synthesis was carried out using the SuperscriptTM II Reverse Transcriptase kit (Invitrogen) with the primer 3’UTR1 and extended PCR fragments obtained with the Elongase® Amplification kit (Invitrogen) and primers MSF1 and L14. For both procedures the kits were used by following recommendations provided by the manufacturer.

For nucleotide sequencing, PCR fragments of the predicted size were separated by electrophoresis in a 1% agarose gel containing Tris-acetate and ethidium bromide, excised and purified using the QiaQuick (Qiagen) gel extraction kit.

Nucleotide sequencing reactions were carried out using the ‘BigDye’ DNA sequencing kit V1.1 according to the manufacturers protocol and analysed on a PE Applied Biosystems 310 genetic analyzer. Sequence analysis and alignment was carried out using Lasergene DNASTAR software version 5. The sequences were aligned using the Clustal V method in Megalign.

ResultsEffect of different passage procedures on pathotypeThe passage procedures used and the resultant pathotypes produced are shown in Table 7. Procedure (i) yielded an isolate with an ICPI of 0.025; this low pathotype was confirmed by a further passage, the ICPI of which was 0.125. Procedure (ii) resulted in an increase in the ICPI from 0.025 to 0.55. Procedure (iii) yielded an isolate with an increased ICPI rising from 0.32 to 1.125, which increased still further to 1.3 on passage three times at limiting dilution in 9-day-old embryonated fowls’ eggs. Following procedure (iv) the ICPI again increased significantly from 0.32 to 1.0125.

RT-PCRFollowing RT-PCR a fragment of more than 4000 base pairs encompassing the F and HN genes was obtained for each of the three pathotypes.

Nucleotide sequencesComparison of the nucleotide sequence of the HN and F genes of the three pathotypes showed that there were no nucleotide differences between the pathotypes 0.025 and 0.55. However, there was one nucleotide difference between these viruses and the pathotype 1.3 – a change of C→T at position 1403 of the fusion protein gene or proline → serine at amino acid position 453 of the fusion protein.

DiscussionThe different passage procedures each yielded isolates of a different pathotype with ICPIs of 0.025, 0.55, 1.3 and 1.0125. All of these viruses, despite the range in ICPI values, had identical cleavage sites.

Nucleotide sequencing analysis demonstrated pathotypes 0.025 and 0.55 had identical F and HN sequences. These two isolates differed from the 1.3 pathotype by a single nucleotide: a change of C→T at position 1403 (residue 453) of the fusion protein gene resulted in the amino acid proline being substituted by serine in the virulent pathotype.

As the only identified change between these isolates (in the F and HN genes), it is possible that the presence of proline instead of serine at position 453 may be the reason for the reduced pathogenicity of isolates 0.025 and 0.55. Further work would be needed to confirm this. This amino acid can affect protein structure, which could cause a change in the protein’s biological function. In this case, the proteins ability to promote fusion between a virion and a host cell is an essential requirement for the efficient spread of virus in the host, a change in which could alter virulence. Fusion activation of NDV post cleavage involves the assembly of a complete HR-A coiled coil with the fusion peptides and transmembrane anchors being brought into close proximity (Chen et al., 2001). It does appear possible that the presence of proline and not serine at position 453 of the fusion protein may affect the fusion process of NDV. In studies of HIV mutants where proline has been substituted in the transmembrane protein gp41, it has been shown that this affects oligomerization of a coiled coil by inducing a kink in a long helix which may account for the deficiency in the biological functions of the proline mutants and the inhibitory action of proline substituted peptides (Chang et al., 1999).

ConclusionsThe original isolate (ICPI 0.32) appears to be a mixed population of particles of low, but varying, pathotype hence passage at limiting dilution in 9-day-old embryonated eggs has presumably meant the selection of a clone that represents the majority in the population with a lower ICPI (0.025). In addition, a small number particles with serine at position 453 of their fusion protein may be present but expression of the virulent (if serine at this position is responsible for virulence) phenotype may be suppressed by the inhibitory action of virus with proline at this position. Other PPMV-1 sequences in GenBank were analysed, but were found to not have a proline at position 453.

Passage at limiting dilution in 14-day-old embryos has raised the ICPI of clone 0.025 to 0.55. Although the latter is still avirulent in terms of the EU definition of virulent NDV (i.e. >0.7 represents notifiable NDV) the increase does appear significant. Furthermore, there were no nucleotide/amino acid sequence differences observed between these pathotypes indicating that the reason for this difference in pathotype lies in another gene. The role of embryo age in increasing the virulence of these isolates appears to be of little significance, since low dilution passage in both 9- and 14-day-old embryos both lead to a similar increase in virulence. This work may be the first time that a change outside the cleavage site, and one which has occurred naturally and not as a result of genetic manipulation, has an impact on the virulence of an isolate.

Further workAn indirect immunoperoxidase study would allow the determination of the capacity of each pathotype to fuse; i.e. measurement of cell-to-cell spread would indicate whether the fusion process has been affected by the substitution of serine by proline. Further, investigating the three dimensional structure of the fusion protein of NDV and the location of amino acid 453 in relation to the functional heptad repeats and leucine zipper motifs.

These viruses are ideal candidates for the identification of genetic virulence markers outside the F and HN gene. Determination of the nucleotide sequence for the entire genome and comparison of these data will allow their identification; the biological function of these mutations would need to be investigated using reverse genetics techniques. This is the subject of ROAME SE0774

Part 3: Phylogenetic analysis of NDV based on partial matrix protein gene sequenceIntroductionThe aim of phylogenetic analysis is to arrive at the best possible estimate of the true evolutionary history of the entities under study, i.e. their phylogeny. Phylogenetic trees are a method in which the mutation pathways that link genes are reconstructed and a way to illustrate the genetic relationships between them, and are typically based on nucleic or amino acid sequence data.

The rapid evolutionary rate of RNA viruses means that, commonly, sufficient variation exists to allow differentiation between even extremely closely related strains using molecular epidemiological techniques; techniques that exploit the fact that viral genomes vary at the nucleotide level whilst maintaining their essential characteristics and functions at the protein and virus level (Clewley, 1998a; Crandall et al., 2002). The application of these techniques yields precise and rapid results, and potentially can be used in outbreaks to find the source and identification of the routes of dissemination.

The region of genome selected for phylogenetic analysis will depend on the nature of the samples being analysed. Typically, if the samples included in the study are all closely related, then a gene with a higher rate of evolution would be better to ensure that sufficient variation exists to allow resolution between them. If samples are more distantly related, or the origins are being investigated, then a more conserved gene would be needed for analysis (Hall, 2001). Generally, genes encoding structural proteins that are targets of the host’s immune response will show more variation, whereas those encoding proteins involved with replication tend to be more conserved. A slower evolving gene may be preferable in some studies on the grounds that it should have less reversion mutation (multiple substitutions at a single site), where a fast evolving gene may have mutated into saturation, making inferences and alignments difficult and inaccurate (Hungnes et al., 2000; Leitner, 2002). The length of sequences used in phylogenetic studies varies; commonly, analysis is carried out on fragments rather than whole genes. Lomniczi et al (1998) report that dataset of sequences 250 nucleotides in length gave meaningful phylogenetic analysis.

There are three main families of methods for inferring phylogeny: parsimony and compatibility methods, distance methods and maximum likelihood methods (Felsenstein, 1988). Parsimony and maximum likelihood (ML) are discrete character based analyses where the particular nucleotides or amino acids in the alignment are examined more directly, and the optimal tree is searched for through a large number of tree topologies (Baldauf, 2003). Maximum likelihood is a statistically based method, where the optimal tree is the one that gives the highest probability of observing the actual sequences, given a particular model of evolution (Felsenstein, 1981). Maximum likelihood method was designed for use with nucleotide data, and many consider it to be the method of choice for tree construction and has been shown to give the most accurate estimates of phylogeny (Clewley, 1998c; Hungnes et al., 2000; Leitner, 2002).

Previous studies carried out in this laboratory (TDP3003, SE0748, SE0762, SE0763) have generated genetic sequence data for many viruses, mostly focusing on the fusion protein gene. As one of the two surface glycoproteins of this virus, this protein is one of the main targets for the host immune system and probably as a consequence to this, this gene exhibits considerable genetic heterogeneity. This heterogeneity is essential for analysis closely related viruses; not only the relationships across the whole NDV population (where the maximum sequence divergence is about 50%), but also within specific lineages and even to variation between isolates comprising a single outbreak of disease (where sequence divergence can be >5%). One disadvantage of this considerable variation is that it may be possible to ‘loose site’ of some more ancestral relationships between viruses due to reversion mutations, where a mutation randomly mutates back to the original sequence.

In order to investigate more ancestral relationships between isolates, phylogenetic studies were carried out using the matrix protein gene. As an internal gene, this is a more highly conserved gene and therefore

eliminates the problems of studying more ancestral relationships in less conserved genes, as in the example of the fusion gene. In this section of the work the partial gene sequence of the matrix gene was determined and analysed using maximum likelihood.

MethodsAlignmentAll published sequence data available for the matrix protein of NDV available on gene bank was downloaded and aligned. This alignment was used for selection of conserved regions, and primers were designed based on this data. Data available on request, but not included in this report.

The PCR was based on the existing protocol used in this laboratory, and annealing temperatures adjusted to match the melting temperatures of the primer combinations (Aldous et al., 2003). The nucleotide sequencing was carried out using Applied Biosystem’s ‘Big-dye’ protocol; samples were analysed using a 310 genetic analyser. SOPs are available on request for both of these procedures.

ResultsThe PCR/sequencing system was optimized to enable amplification of isolates from all sixteen defined lineages of NDV (Aldous et al., 2003). The primer combination M220 and 4 generated a PCR product of 750bp for all isolates, but was unreliable for isolates from lineage 6. Reliable amplification of lineage 6 isolates required a nested PCR, with a first round M220 and 4 primer combination, followed by a 3 and 4 combination. All three primers were used for sequencing.

Sequence analysis was carried out using Lasergene’s DNAstar suite of programmes. Maximum likelihood phylogenetic analysis was carried out using the PHYLIP suite of programmes.

Discussion and conclusionsPreviously, phylogenetic analysis of NDV has focused on the F protein gene; as one of the two surface glycoproteins this is the target of the hosts immune response and as such exhibits considerable genetic variation. In this study, the phylogenetic relationships within the APMV-1 population were assessed based on the matrix gene; this codes for an internal protein and is therefore likely to be more conserved. The purpose of this work was to confirm the lineages as determined by fusion gene phylogeny, but also to identify more ancestrally linked isolates. As an internal protein, and therefore less of antigenic target, the variation within the gene is likely less than that of the fusion gene. Based on this, it was proposed that it may be possible to identify more distantly related isolates, particularly with a view to identifying avirulent precursor viruses from which virulent viruses have emerged. Eighteen isolates of NDV have been analysed, representing all of the lineages identified within the APMV-1 populations. Partial matrix gene nucleotide sequence was generated for all samples, using a single pair of primers; the dataset was analysed using maximum likelihood method. Within the group the divergence (calculated in Lasergenes Megalign) for the fusion gene sequences was greater, with a maximum of 54%; within the matrix gene analysis there is a maximum of 34% divergence, illustrating the more conserved nature of this region of the matrix gene. The analysis confirmed the lineages as determined by F-gene analysis, but also identified that the virulent isolates of lineage 3 are the most closely related virulent samples (excluding Australia 97 samples) to avirulent isolates in lineage 1.

ReferencesAldous, E. W., Mynn, J. K., Banks, J. & Alexander, D. J. (2003). A molecular epidemiological study of avian paramyxovirus type 1 (Newcastle disease virus) isolates by phylogenetic analysis of a partial nucleotide sequence of the fusion protein gene. Avian Pathology 32, 239-257.Alexander, D. J. 2003 Newcastle Disease, other Avian Paramyxovirus and Pneumovirus infections. Y. M. Saif. Diseases of Poultry. 64-87. USA, Iowa State University Press. Alexander, D. J., Campbell, G., Manvell, R. J., Collins, M. S., Parsons, G. & McNulty, M. S.(1992). Characterisation of an antigenically unusual virus responsible for two outbreaks of Newcastle disease in the Republic of Ireland in 1990. Veterinary Record, 130, 65-68.Brugh, M. &Beck, J. R. 1993 Recovery of minority subpopulations of highly pathogenic avian influenza virus. Easterday, B. C. Proceedings of the Third International Symposium on Avian Influenza. 166-174. Richmond, Virginia, United States Animal Health Association. Brugh, M. & Perdue, M. L.(1991). Emergence of highly pathogenic virus during selective chicken passage of the prototype mildly pathogenic chicken/Pennsylvania/83 (H5N2) influenza virus. Avian Diseases , 35, 824-833.Collins, M.S., Franklin, S., Strong, I., Meulemans, G., & Alexander, D.J. (1998). Antigenic and phylogenetic studies on a variant Newcastle disease virus using anti-fusion protein monoclonal antibodies and partial sequencing of the fusion protein gene. Avian Pathology, 27, 90-96.

Daniel, M.D. & Hanson, R.P. (1968). Differentiation of representative Newcastle disease virus strains by their plaque forming ability on monolayers of chick embryo fibroblasts. Avian Diseases, 12, 424-433.deLeeuw, O. S., Hartog, L., Koch, G. & Peeters, B. P. H.(2003). Effect of fusion protein cleavage site mutations on virulence of Newcastle disease virus: non-virulent cleavage site mutants revert to virulence after one passage in chicken brain. Journal of General Virology, 84, 475-484.Gould, A. R., Hansson, E., Selleck, K., Kattenbelt, J. A., Mackenzie, M. & DellaPorta, A. J.(2003). Newcastle disease virus fusion and haemagglutinin-neuraminidase gene motifs as markers for viral lineage. Avian Pathology, 32, 361-373.Gould, A. R., Kattenbelt, J. A., Selleck, P., Hansson, E., Della Porta, A. & Westbury, H. A.(2001). Virulent Newcastle disease in Australia: Molecular epidemiological analysis of viruses isolated prior to and during the outbreaks of 1998-2000. Virus Research, 77, 51-60.Granoff, A. (1964). Nature of Newcastle disease virus populations. InR.P. Hanson (Ed.), Newcastle disease: An evolving pathogen Wisconsin, USA: University if Wisconsin Press.Hanson, R. P. (1988) Heterogeneity within strains of Newcastle disease virus: key to survival. Alexander, D. J. Newcastle Disease. 113-129. Massachusetts, USA, Kluwer Academic Press. Holland, J.J., De la Torre, J.C., & Steinhauer, D.A. (1992). RNA virus populations as quasispecies. Current topics in Microbiology and Immunology, 176, 1-20.Horimoto, T. & Kawaoka, Y. (1995). Molecular changes in virulent mutants arising from avirulent avian influenza viruses during replication in 14-day-old embryonated eggs. Virology, 206, 755-759.Islam, MA, Ito, T., Takakuwa, H., Takada, A., Itakura, C. & Kida, H.(1994). Acquisition of pathogenicity of a Newcastle disease virus isolated from a Japanese quail by intracerebral passage in chickens. Japanese Journal Veterinary Research, 42, 147-56.Ito, T., Goto, H., Yamamoto, E., Tanaka, H., Takeuchi, M., Kuwayama, M., Kawaoka, Y., & Otsuki, K. (2001). Generation of a highly pathogenic avian influenzaA virus from an avirulent field isolate by passaging in chickens . Journal of Virology, 75, 4439-4443.Ito, T., Kawaoka, Y., Kameda, C., Yasuda, J., Kida, H. & Otsuki, K.(1999). Differences in receptor specificity between Newcastle disease viruses originating from chickens and waterfowl. Journal of Veterinary Medical Science, 61, 951-953.Kaleta, E.F., Seigmann, O., Jank-Ladwig, R., & Glunder, G. (1980). Isolation and biological properties of virulent sub-populations from lentogenic Newcastle disease virus strains. Comparative Immunology and Microbiology of Infectious Diseases, 2, 485-496.Karber, G. V.(1931). Beitrag zur kolletiven Behandlung pharmakologischer Reihenversuche. Archiv fur Experimentelle Pathologie und Pharmakologie, 162, 480-483.Kirkland, P. D.(2000). Virulent Newcastle disease virus in Australia: in through the 'back door'. Australian Veterinary Journal, 78, 331-333.Kommers, G. D., King, D. J., Seal, B. S. & Brown, C. C.(2003). Virulence of six heterogeneous-origin Newcastle disease virus isolates before and after sequential passages in domestic chickens. Avian Pathology, 32, 81-93.McMillan, BC, Fellenz, SC & Hanson, RP(1986). The effects of the multiplicity of infection on viral subpopulations during passage of Newcastle disease viruses. Avian Diseases, 30, 122-5.Millar, N. &Emmerson, P. 1988 Molecular cloning and nucleotide sequencing of Newcastle disease virus. Alexander, D. J. Newcastle Disease. 79-97. Massachusetts, USA, Kluwer Academic Press. Peeters, B. P. H., deLeeuw, O. S., Koch, G. & Gielkens, A. L. J.(1999). Rescue of Newcastle disease virus from cloned cDNA: Evidence that cleavability of the fusion protein is a major determinant for virulence . Journal of Virology, 73, 5001-5009.Pritzer, E., Kuroda, K., Garten, W., Nagai, Y. & Klenk, H. D.(1990). A host range mutant of Newcastle disease virus with an altered cleavage site for proteolytic activation of the F protein. Virus Research, 15, 237-42.Rott, R. &Klenk, H. D. 1988 Molecular basis of infectivity and pathogenicity of Newcastle disease virus. Alexander, D. J. Newcastle Disease. 98-112. Massachusetts, USA, Kluwer Academic Press. Sergel, T. A., McGinnes, L. W. & Morrison, T. G. (2000). A single amino acid change in the Newcastle disease virus fusion protein alters the requirement for HN protein in fusion. Journal of Virology, 74, 5101-5107.Shengqing, Y., Kishida, N., Ito, H., Kida, H., Otsuki, K., Kawaoka, Y. & Ito, T.(2002). Generation of velogenic Newcastle disease viruses from a nonpathogenic waterfowl isolate by passaging in chickens. Virology, 301, 206-211.Slosaris, M., Levy, B., Katz, E., Levy, R. & Zakay, Rones Z.(1989). Elevated virulence of Newcastle disease virus strains following serial passages in kidney cells in vitro. Avian Diseases, 33, 248-253.Steinhauer, D.A. & Holland, J. (1987). Rapid evolution of RNA viruses. Annual Review of Microbiology, 41, 409-33.Wang, Z. Y. & Iorio, R. M.(1999). Amino acid substitutions in a conserved region in the stalk of the Newcastle disease virus HN glycoprotein spike impair its neuraminidase activity in the globular domain. Journal of General Virology, 80, 749-753.

Westbury, H.(2001). Newcastle disease virus: an evolving pathogen? Avian Pathology, 30, 5-11.

Table 1 Cleavage site nucleotide sequences and amino acid motifs Virus Country Virulence Nucleotide/ amino acid sequence at the

F0 cleavage siteMC110 Ireland Low GAA CGG CAG GAG CGT CTG

112ERQER*L117

34/90 Ireland High AAA CGG CAG AAG CGT TTT112KRQKR*F117

NSW 12/86 Australia Low GGG AAA CAG GGA CGT CTT111GKQGR*L117

Qld 1/87 Australia Low GGG AAA CAG GGA CGT CTT111GKQGR*L117

1154/98 Australia Low GGA AGG AGA CAG GGG CGT CTT‘Peat’s Ridge’ 111GRRQGR*L117

Somersby/99* Australia Moderate* -112RRQRR*L117

PR-32* Australia High -112RRQGR*F117

1236/98 Australia High GGA AGG AGA CAG AGG CGT TTT111GRRQRR*F117

1249/98 Australia High GGA AGG AGA CAG AGG CGT TTT111GRRQRR*F117

* data provided by P. Selleck, personal communication, reports that birds in this group show signs of disease, but many recover.

Table 2 Primer Details

Name Purpose Sequence 5' - 3' Direction TM Gene 5' positionMSF1 cDNA/ PCR F GACCGCTGACCACGAGGTTA F 73 1025 (M gene)FCF12 cDNA HN CAACTCGATAAGTAATGCTCTGAA F 61 1459 (F gene) FCR3 PCR F AGRGTATTATTCCCAAGCCATA R 59 1676 (F gene) 4 PCR F (WB) TACTGCTGTCGCTACACCTAA R 62 400 (F gene)FCF6 PCR HN AGACCTTAYTATGGCTTGGGAA F 62 1645 (F gene)HNCR5 PCR HN GACCGGAGCTCGCCATGTCC R 74 8 (L gene) MSF1 seq f GACCGCTGACCACGAGGTTA F 73 1025 (M gene)7 seq f TTAGAAAAAACACGGGTAGAA F 56 0 (F gene) 3 seq f (WB) ATGCCCAAAGACAAAGAGCAA F 60 100 (F gene)23 seq f AACTGACTACAGTGTTTGGGCC F 64 693 (F gene)2 seq f AGTCGGAGGATGTTGGCAGC R 64 507 (F gene)16 seq f TCCAAGTAGGTGGCACGCATA R 67 957 (F gene)11FOV seq f CTGCTGCATCTTCCCAACTG R 62 598 (F gene)17 seq f TGTTGACATTCCAAGCTCAG R 61 1460 (F gene)6FOV seq f CGGAATATCAAGCGCCATGTA R 67 168 (HN gene)FCF6 seq hn AGACCTTAYTATGGCTTGGGAA F 62 1646 (F gene)FCF8 seq hn GTGKGGGGCRCCTRTYCATGA F 63 2279 (F gene)15HNOV seq hn AACAGCCACTCTTCATAGTCC F 62 1393 (HN gene)6FOV seq hn & f CGGAATATCAAGCGCCATGTA R 67 168 (HN gene)HNCR9 seq hn TGTAGTRGGYGCCGGGATRAARTT R 67 593 (HN gene)HNCR5 seq hn GACCGGAGCTCGCCATGTCC R 74 2076 (HN gene)HNCR13 seq hn GGACTATGAAGAGTGGCTGTT R 60 1413 (HN gene)2HNOV seq hn CCGTCGAACCCTAACCTCC R 62 927 (HN gene)12HNOV seq hn GGTCTTCGCCTAAGGATGTTG R 64 1247 (HN gene)4HNOV seq hn CCTCGCAAGGTGTGGTTTCTA R 64 1548 (HN gene)3HNOV seq hn GTCTTGCAGTGTGAGTGCAAC F 64 799 (HN gene)L114 seq hn TGTGRCTCTGGTAGGATAATCTG R 68 35 (L gene)

Tables 3a Comparison of p0 and p10 screening data (unplaqued WB)Lowest HA positive dilutionIsolate Passage

numberICPI HA titre MDCK +/- trypsin

ratio+ trypsin - trypsin Cleavage site sequence HA properties1

Original duck 0.0 2.14 -4 -2 GGA GGG AGA CAG GGA CGC CTT

R, Rb, T, S, -, G

1. Unplaqued AV229/03A 0 128 1.57 -3 -2 GGA GGG AGA CAG GGA CGC

CTT10 512 1.75 -4 -2 GGA GGG AGA CAG GGA CGC CTT

R, --, T, S, H, GB 0 128 1.72 -4 -2 GGA GGG AGA CAG GGA CGC


R, Rb, T, S, H, GAB3 0 64 1.72 -4 -2 GGA GGG AGA CAG GGA CGC

CTT10 1024 2 -5 -2 GGA GGG AGA CAG GGA CGC CTT

R, --, T, S, H, GAB4 0 256 1.38 -4 -2 GGA GGG AGA CAG GGA CGC


R, Rb, T, S, H, GAB6 0 128 1.86 -5 -2 GGA GGG AGA CAG GGA CGC

CTTR, Rb, T, S, -, G

10 0.0 512 1.5 -3 -2 GGA GGG AGA CAG GGA CGC CTT

R, Rb, T, S, -, GL1 0 256 1.33 -3 -3 GGA GGG AGA CAG GGA CGC



R, Rb, T, S, H, GL9 0 256 1.63 -4 -2 GGA GGG AGA CAG GGA CGC


R, Rb, T, S, H, GL14 0 256 1.83 -3 -2 GGA GGG AGA CAG GGA CGC

CTT10 256 4 -1 -1 GGA GGG AGA CAG GGA CGC CTT

R, Rb, T, S, H, GT 0 128 ND -2 0 GGA GGG AGA CAG GGA CGC


R, Rb, T, S, -, GU 0 256 ND -2 0 GGA GGG AGA CAG GGA CGC


R, Rb, T, S, -, G

1 Types of blood agglutinated, where H: horse, T: turkey, M: mouse, R: rat, Rb: rabbit; S: sheep and G: guinea pig.

Tables 3b Comparison of p0 and p10 screening data (plaqued WB)Lowest HA positive dilutionIsolate Passage

numberICPI HA titre MDCK +/- trypsin

ratio+ trypsin - trypsin Cleavage site sequence HA properties1

Original duck 0.0 2.14 -4 -2 GGA GGG AGA CAG GGA CGC CTT

R, Rb, T, S, -, G

2. Plaqued AV 229/03WB 61 0 128 1.8 -3 -2 GGA GGG AGA CAG GGA CGC

CTTND

10 256 1.66 -3 -3 GGA GGG AGA CAG GGA CGC CTT

R, Rb, T, S, -, GWB 64 0 256 1.83 -4 -2 GGA GGG AGA CAG GGA CGC

CTTR, Rb, T, -, -, G


R, Rb, T, S, H, GWB 65 0 256 1.4 -2 -2 GGA GGG AGA CAG GGA CGC

CTTND


R, Rb, T, S, -, GWB 68 0 256 1.43 -3 -3 GGA GGG AGA CAG GGA CGC

CTTR, Rb, T, -, -, G


R, Rb, T, S, H, GWB 69 0 256 1.75 -3 -2 GGA GGG AGA CAG GGA CGC

CTTND


R, Rb, T, S, -, GWB 70 0 128 2 -1 -1 GGA GGG AGA CAG GGA CGC

CTTND


R, Rb, T, S, -, G

3. Intra-cerebral passageIC6 0 512 1.5 -3 -2 GGA GGG AGA CAG GGA CGC


6 0.0 64 2 -4 -2 GGA GGG AGA CAG GGA CGC CTT

R, Rb, T, S, H, G

1 Types of blood agglutinated, where H: horse, T: turkey, M: mouse, R: rat, Rb: rabbit; S: sheep and G: guinea pig.

Table 4a Nucleotide substitutions in F and HN genes: WB sample

Position1 394 F2 711 F 250 HN 665 HN 687 HN 690 HN 922 HN 1464 HN 1574 HNcodon AA codon AA codon AA codon AA codon AA codon AA codon AA codon AA codon AA

Original duck CGC Arg GGG Gly CCT Pro GTA Val CAC His TCG Ser TTC Phe TCA Ser GAA Glu

L1 p10 3 TTG Leu GWA Glu/ValAB6 p10IC6 CGC Arg AAA LysPQ64 p0 ATA Ile TTA LeuPQ64 p10 GAG Glu CCG Pro ATA Ile TTA Leu GTA ValPQ68 p0 CGY Arg CCG Pro ATA Ile TTA Leu GWA Glu/ValPQ68 p10 GAG Glu CCG Pro ATA Ile TTA Leu GTA Val

Table 4b Nucleotide substitutions in F and HN genes: V4 vaccine sample

Position177 F 305 Fcodon AA codon AA

Original V4 AAC Asn CCC Pro

V4 uncloned p10V4 clone 66 p10 AAW AsnV4 clone 74 p10 CCT Pro

1 Base 1 is considered to be the first base of the mRNA start sequence (Millar & Emmerson, 1988). Numbering restarts for each gene. 2 Where F stands for fusion, HN for haemagglutinin-neuraminidase3 Blank space means this is the same as the sequence recorded for original sample

Table 5 Mixed sample analysis by nucleotide sequencingMix 1 Mix 2 Mix 3 Mix 4 Mix 5

Virulent 10-3 10-4 10-5 10-6 10-7

DNA in 2.5l 17.5pg 1.75pg 175fg 17.5fg 1.75fgAvirulent 10-7 10-6 10-5 10-4 10-3

DNA in 2.5l 3fg 30fg 300fg 3pg 30pgSequencing result virulent virulent mix (virulent) avirulent avirulent

Table 6 Preliminary passage schedule for unplaqued WB virusStep Passage purpose No

passagesNo. eggs

1 Initial adaptation to 14 day old ck eggs 2 3 x 14doCK2 Growth of step 1 harvest in MDBK +/- trypsin 2 -3 Serial dilution (SD) to end point using step 1 material 1 5/dil 14doCK4 Single passage of first SD eggs to die (A & B pool) 1 40 x 14doCK5 Single passage of lowest SD positive egg (L) 1 40 x 14doCK6 Serial passage of selected isolates 10 3 x 14doCK7 Serial intra-cerebral passage in chicks 6 -

Table 7: Changes in pathogenicity of 1073/98 248VB following different passage procedures in embryonated fowls’ eggs Pass. ID Passage procedure Egg age in days (no passages) ICPI AA @ residue 453 i LDa 9 (3) (4) 0.025, 0.125 Prolineii LDa of 0.025 14 (5) 9 (1) 0.55 Prolineiii Low dilutionb / LDa 14 (5) 9 (1) / 9 (3) 1.125 / 1.3 Serineiv Low dilution/ LD 9 (5) / 9 (4) 1.0125 Serine

a LD = limiting dilutionb Low dilution = 1/10 and 1/100 dilutions

Figure 2. Phylogenetic tree of APMV-1 population based partial nucleotide sequence of the matrix protein gene

Maximum likelihood analysis of 251bp of the NDV matrix protein gene, with group 6 isolates identified as the outgroup.

References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.Aldous E. W. (2004) Molecular investigations into the pathogenicity and epidemiology of Newcastle disease virus. Ph.D. thesis, Reading University.

general enquiries on this form should be made...

Documents