Annual influenza epidemics and pandemics cause a significant disease burden and death rates in human beings, and have a high economical impact on the EU. The major drawbacks of influenza vaccines are their limited efficacy and production methods that are hard to adapt to requirements of continuously changing influenza viruses. We aim to improve the efficacy and production methods of epidemic and pandemic influenza vaccines through:
Annually occurring influenza virus epidemics, associated with antigenic 'drift' of the surface glycoproteins haemagglutinin and neuraminidase (HA and NA) of the influenza A and B viruses, cause a significant disease burden and high death rates in human beings which exceed deaths due, for example, to road accidents in Europe. Less frequent pandemic outbreaks of influenza (three in the last century) associated with the introduction of a new influenza A virus subtype from an animal reservoir - antigenic 'shift' - are even more serious in this respect, causing millions of deaths in a short period of time.
Although recently a new generation of antiviral compounds (neuraminidase inhibitors) with promising anti-viral effects have been developed, the most cost-effective and efficient way of controlling influenza remains preventive vaccination. The major drawbacks of the currently used multi-component influenza vaccines, produced in embryonated chicken eggs, are their limited efficacy and classical production methods which are hard to adapt to requirements related to 'drift' and 'shift' phenomena.
The aim of the project was the development of more effective strategies for the vaccination of human beings against epidemic and pandemic influenza. Divided over the respective work packages four objectives were identified. These were:
- Development of optimal strategies for vaccine selection: Computer algorithms were developed which are used for the interpretation of complex serological data and allow the easy assessment of antigenic relatedness of influenza viruses. These tools have been implemented and fully integrated in the bi-annual WHO vaccine strain selection process. In addition, these newly developed tools can be used for the antigenic characterisation of the potentially pandemic H5N1 strains. Thus in the case of a pandemic outbreak, the prior knowledge of the antigenic properties of these viruses can aid in the selection of the best vaccine strains to provide the broadest protection possible. This scenario was exercised with the avian H7N7 viruses that caused an outbreak of fowl plague in the Netherlands in 2003 (see c). Extensive surveillance of avian influenza viruses amongst wild birds resulted in a library of virtually all subtypes of HA and NA, which were used for vaccine preparation and the preparation of reference reagents to aid in the adequate diagnosis and identification of avian influenza virus infections in man and animals.
- Development of alternative approaches for vaccine production with the recent registration of MDCK-cells as an influenza vaccine production platform and the advent of reverse genetics technology which allows the genetic manipulation of viruses. It was the objective to engineer influenza viruses and develop reverse genetics technology that would allow the rapid generation of reassortant strains that give high yields of the relevant viral proteins, the HA and the NA.
In Work Packages 2 and 3 the development of these reverse genetic systems was addressed. For the production of viruses in embryonated chicken eggs or chicken embryo fibroblasts, the avian POL1 promoter was cloned which can drive the transcription of viral RNA in avian cells. It was not possible to optimise the extremities of the HA gene segments for enhanced vaccine production yield since with a reporter gene expression system it was demonstrated that these non-coding regions are already compatible with the RNA complexes of various viruses including that of the vaccine back-bone strain A/PR/8/34. Furthermore, a universal reverse genetics system was developed based on the T7 polymerase and the T7 promotor system. This would allow the transcription of RNA in any cell type independent of the species from which the cells originated (e.g. MDCK cells).
A patent application was filed on the T7 reverse genetics system. The availability of an alternative reverse genetic system constitutes a very valuable alternative that would prevent the use of existing systems and for which the intellectual property rights lie outside the EU. In addition the use of the T7 system is better tailored for the use in MDCK cells to which EU companies have access.
- Development of novel vaccine candidates: This was addressed in three different ways. First, recombinant MVA vectors were developed and evaluated as a novel delivery system for influenza viral proteins. MVA vectors were constructed that express the HA of two different strains of H5N1 virus. These recombinant H5N1 poxvirus vector vaccines were evaluated in a mouse model and it was demonstrated that with these vaccine preparations protective immunity could be induced against the lethal challenge of H5N1 virus. Based on these data a patent application was filed. Furthermore, MVA were constructed expressing the nucleoprotein (NP) gene of various influenza viruses, including the H5N1 viruses. It is anticipated that recombinant MVA vaccine candidates, of which the production is independent of existing influenza vaccine production capacity, can contribute to overcoming the envisaged vaccine shortage in the face of pandemic outbreaks of influenza.
The second approach that was evaluated was the use of immune stimulating complexes (ISCOMS) as an alternative antigen delivery system. Using a combination of reverse genetics technology, avian influenza surveillance and novel vaccine strain selection procedures an H7N7 vaccine strain was prepared for the production of the viral antigens HA and NA. This preparation was adjuvanted with ISCOMS and used for the immunisation of mice, which developed virus specific antibodies and were protected against infection with a lethal H7N7 virus.
The third approach for the development of novel candidate vaccines is the use of defective influenza virus particles. Again, reverse genetics technology was exploited for the production of defective virus particles. Viruses were produced that lack one functional gene that is essential for its replication. It proved possible to produce defective particles by trans-complementation and a patent application was filed on this procedure. A vaccine based on defective virus particles is considered promising, since it is expected that it will not only induce antibody responses but also cell-mediated immunity.
- Definition of correlates of protection: For the determination of correlates of protection a lot of emphasis was put on virus specific cytotoxic T lymphocytes (CTL). A detailed analysis was made of the interaction between influenza viruses and the human CTL response. First, it was observed that epitopes recognised by human CTL are under selective pressure, which is indirect evidence that CTL are important immune correlates in the control of influenza virus infections. The impact of variation in individual epitopes was studied on the virus specific human CTL response and it was found that a single amino acid change in an epitope affected the CTL response significantly, indicating that this variation is advantageous to the virus.
In one case a number of co-mutations was observed with a mutation in an epitope and it proved impossible to rescue virus by reverse genetics technology with a single substitution in the epitope. It was found that the co-mutations were necessary to functionally compensate for the detrimental effect of the amino acid substitution in the epitope. Also, the extent of variation of CTL epitopes was assessed and some more examples of point mutations associated with escape from CTL were identified.
Although the variation of CTL epitopes could complicate the development of vaccines that aim at the induction of virus specific CTL response, it was also found that certain conserved epitopes are under functional constraints. In a mutational analysis it proved impossible to introduce amino acid substitutions without the loss of viral fitness. This may indicate that the induction of T cell immunity against these conserved epitopes may be a feasible approach for the induction of broad-spectrum immunity, also against pandemic strains of influenza.
Results obtained during the project were presented at national and international conferences on 91 occasions. A total of 75 publications in international peer-reviewed journals emanated from the project and three patent applications were filed.
- The computer algorithm for the antigenic characterisation of influenza viruses is already applied in the annual influenza vaccine strain selection process and could be used for pandemic vaccines strain selection.
- A library of HA/NA subtypes could be used for the preparation of prototype vaccines.
- Novel reverse genetics technology can be used for the preparation of vaccine strains in MDCK cells and CEF cells.
- Novel antigen delivery systems (MVA, defective virus particles and use of ISCOMS) can be used for the production of effective vaccines and may aid in overcoming the envisaged shortage of vaccines in the face of a pandemic.
- The induction of virus-specific CTL may be a way to prepare broadly protective vaccines.
Prof. Dr A D M E Osterhaus
Dept of Virology
Dr Molewaterplein 50
Tel: +31 10 40 88 firstname.lastname@example.org
Dr D J Smith
University of Cambridge
Cambridge CB2 3EJ
Tel: +44 12 23 33 44 email@example.com
Solvay Pharmaceutical Bv
C J van Houtenlaan 36
Tel: +31 29 44 77 firstname.lastname@example.org
Prof. Dr S van der Werf
28 Rue du Docteur Roux
Tel: +33 14 56 88 email@example.com
Prof. Dr G Sutter
Tel: +49 61 03 77 21 firstname.lastname@example.org
Prof. Dr J Oxford
327 Mile End Road
London E1 4NS
Tel: +44 20 78 82 79 email@example.com