Accelerating pandemic vaccine manufacture

April 1, 2010

Pharmaceutical Technology Europe

Florian Krammer explains how a novel technology using insect cells can accelerate the manufacture of pandemic influenza vaccines.

Pharmaceutical Technology Europe speaks to Florian Krammer, a researcher from the baculovirus group at the Vienna Institute of BioTechnology (VIBT) about a novel technology that could provide pandemic influenza vaccines faster than conventional egg-based methods.

Can you provide more details about the insect cell-based vaccine manufacture technique?
The technique itself is very simple. The hemagglutinin gene as well as the matrix protein gene of influenza virus is cloned into a recombinant baculovirus under control of strong baculoviral promotors. This recombinant baculovirus is than used to infect insect cells in a bioreactor. The baculovirus drives the insect cell to express hemagglutinin and matrix protein, hemagglutinin locates at the membrane, matrix protein co-associates and the virus-like particles are released into the supernatant by budding. After 72 hours post infection the cell supernatant, containing the VLPs, is harvested and purification takes place.

Why is it so much faster than traditional methods?
Time from isolation of a new virus strain to the first batch of vaccine is very short. A VLP vaccine against a new strain (based on a mock-up vaccine) could be on the market within three months from the first isolation of a new influenza strain.

As we have seen, this takes about six months for traditional produced vaccines.

For traditional egg-based vaccines it is necessary to make re-assortants of the new strain with “high-growth” strains which give high yields in eggs. This is a very time consuming technique which needs a lot of screening.

Cell-culture produced inactivated vaccines on the other hand can be a problem because new isolates don’t necessarily grow well/give high yields. Additionally, new isolates dedicated for cell-culture based production have to undergo strict purification procedures to rule out a carry over of unknown pathogens (plaque purifications) and mutations of the HA caused by host adaption have to be ruled out (these vaccines are produced on Vero or MDCK cells, Vero is african green monkey derived, MDCK is derived from a cocker spaniel, so HA from human isolates may adapt to these new hosts). These procedures are very time coming as well.

These procedures can entirely be circumvented by the VLP technique. The RNA from a new virus strain is extracted, amplified by PCR, sequenced and cloned into a baculovirus genome. This can be done within 4 to 6 days. The baculovirus genome is than used to generate recombinant baculovirus which takes another 5 days. The recombinant baculovirus is than amplified an can be directly used for vaccine production (the first batch of VLPs for immunological testing can be ready within 6 weeks).

Influenza VLP vaccines bypass most disadvantages of traditional egg produced inactivated influenza vaccines like problems with yield, the necessity of high-growth re-assortants, high risk of contamination, limited specific pathogen free egg supply in case of a pandemic, process upscale, purification and therefore problems with vaccinees with allergies against egg proteins and unwanted selection of unwanted antigenic variants. They also circumvent many disadvantages of cell-culture based production of inactivated influenza vaccines like slow growth of isolates and therefore unpredictable yields, mutation of HA through host adaption and the need for biosafety level 3 facilities which are obligatory in cell-culture based production systems for pandemic vaccines. VLP vaccines furthermore offer predictable yields with various types of hemagglutinins (H1, H3, H5, H9). Another major advantage of VLPs is the absence of any genetic information; therefore VLPs are completely replication deficient and non-infectious.

This makes their application in the human population very safe, norecombination with wild type virus or complications in immune suppressed individuals as observed with attenuated vaccines can occur. Influenza VLPs are structurally very similar to their correspondent pathogenic virus and therefore, they are readily uptaken into antigenpresenting cells and therefore able to induce CD4+ cell proliferation which results in B-cell as well as in cytotoxic T-cell immune responses.

What were the main challenges associated with using insect cells?/ Are there any drawbacks or problems that still need to be resolved regarding the use of the technology?
As influenza VLPs are very difficult to separate from baculovirus (due to similar densities) this is the main challenge of the technology. The original method used Sf9 cells, derived from the fall armyworm Spodoptera frugiperda. With our technology it is possible to reduce baculovirus background 100fold. This was achieved by establisment of the production system in another expression cell line, namely BTI-TN5B1-4 (High Five), derived from the American cabbage looper Trichoplusia ni. This change went along with increased yield (a publication dealing with these details has been accepted by Molecular Biotechnology and will be published within the coming months). However, purification remains the main challenge.

How important do you think virus-like particles will be to the future of vaccine manufacture?
There are currently two VLP vaccines for human use on the market: recombinant hepatitis B vaccines (produced in yeast) and human papilloma vaccines (produced in insect cells or yeast). VLP vaccines have an excellent safety profile and are very efficient. There are VLP vaccine candidates for many viruses, including influenza virus, in the pipeline. Additionally, a lot of approaches for displaying foreign proteins and epitopes on the surface of various VLPs have been developed (we are currently working on displaying tuberculosis antigens on their surface). I think that VLP vaccines will play a very important role in the future of vaccine manufacturing.

The paper can be read at: