Development of marker vaccines for rinderpest virus using reverse genetics technology

Full text: Rinderpest or 'cattle plague' is an economically devastating disease which is still endemic in areas of Eastern Africa, its aetiological agent being a morbillivirus (RPV) closely related to human Measles virus. A global rinderpest eradication programme (GREP) is in place and its goal is the eradication of the disease by the year 2010. In the final stages of the eradication campaign, when mass vaccination must be discontinued, it would be desirable to use a vaccine that enables vaccinated animals to be distinguished from animals which have recovered from natural infection, so called 'marker vaccines'. These are now being produced using reverse genetics, the process whereby the genomes of RNA viruses can be genetically manipulated through a DNA copy and live virus rescued from the altered DNA. This very powerful new technology, in addition to its practical usefulness in allowing the development of marker vaccines, also enables us to investigate, on a rational scientific basis, the molecular determinants of virulence and attenuation in this virus group. This will be an important consideration when assessing the safety of such vaccines for general use. Reverse genetics for RPV was established in the Pirbright laboratory some six years ago and several candidate marker vaccines have been produced which are now ready for field testing. The first approach we have taken was to place genes encoding 'positive' marker proteins into the RPV genome. The markers chosen were the green fluorescent protein (GFP) and a modified form of the influenza haemagglutinin (fluHA) protein. A strong antibody response was generated to the fluHA marker protein in all animals vaccinated with the RPV-fluHA recombinant. The GFP gene was modified so that the protein, in addition to cytoplasmic expression, would be either secreted from the infected cells or expressed as a membrane anchored protein of the cell surface. The antibody response to the GFP protein elicited in the vaccinated animals depended on its mode of expression. The form of GFP that was only expressed in the cytoplasm failed to elicit an immune response in any of the vaccinated animals, despite very high expression levels of the protein during in vitro experiments. Animals vaccinated with a virus expressing a secreted form of GFP varied in their responses to the marker protein with only 50% showing strong anti-GFP antibody levels. Only the membrane-anchored form of the protein gave a strong antibody response in all vaccinated animals and thus was a suitable marker protein. In addition to positive markers, a 'negatively' marked vaccine, i.e. one lacking a RPV-specific antigenic component, was also produced. To achieve this a chimeric virus in which the nucleocapsid protein (N) gene of RPV was replaced by that from the related morbillivirus Peste des petits ruminants virus (PPRV), was produced (rRPV-PPRN) and tested in cattle. Use of this recombinant virus vaccine enables vaccinated animals to be distinguished from those that have been naturally infected. This distinction can be made using two currently available ELISAs that detect antibodies specific to either the N or haemagglutinin (H) proteins of these two viruses. Vaccinated animals become positive in the PPRV N-specific ELISA and negative in the RPV N-specific assay. The opposite is true for the H proteinspecific ELISAs. Vaccinated animals which subsequently become infected with wild type virus will become double positive for the N protein antibodies of both viruses. In a further development we have introduced a positive marker gene into this virus genome, namely that encoding the membrane-anchored GFP protein. This vaccine (rRPV-PPRN-ancGFP), in addition to being distinguishable using the above mentioned tests, produces antibodies against the marker protein in vaccinated animals which are absent in naturally infected and recovered animals. Unfortunately this 'positive-negative' marker vaccine, unlike the other marker vaccines produced thus far, does not grow to high titre in tissue culture and we are currently trying to improve its growth by introducing a chimeric RPV-PPRV N protein gene to replace the complete PPRV N gene in this virus. One major consideration when developing genetically modified virus vaccines is safety. The possibility that the marker proteins could be incorporated into the virus envelope and thus alter the tropism and possibly the pathogenicity, of the vaccine produced, must be considered. Generally viruses have mechanisms to exclude the incorporation of foreign proteins into their envelopes and immunoprecipitation studies using antibodies to the marker proteins failed to precipitate the recombinant vaccine viruses, indicating that the marker is excluded from the virion envelopes. As an additional safety precaution, the gene for the HA protein was modified to remove its receptor-binding capacity to ensure that a novel tissue tropism cannot be induced in the recombinant virus. Immunoelectron microscopy also showed that virion envelopes were free of the marker fluHA protein. These findings have implications for the design of new vaccines based on the rescue of negative strand viruses where either marker proteins or additional immunogens from other pathogens are incorporated into their genomes to produce dual vaccines. (author)