July 2010

Virology:

A Bacterial Influenza Vaccine Factory

At the start of the influenza A H1N1 "swine flu" pandemic last year, no one knew for sure whether this would be a virulent variant of the influenza virus. The pandemic occurred only a few years after the SARS (Severe Acute Respiratory Syndrome) outbreak of 2002-2003, of which one variant was extremely deadly, and the influenza A H5N1 "bird flu" outbreak of the past few years, which kills over half of the people it infects.

It's natural that people and governments were wary of the new influenza strain in their midst, considering how contagious it was. Fortunately, it turned out to be more contagious, but far less deadly, than typical seasonal flu.

On the other hand, the United States government spent tons of money to build up an enormous stockpile of H1N1 vaccines. There's certainly plenty of room for debate as to whether this stockpile was necessary.

Nevertheless, it's still noteworthy that nearly half of this vaccine stockpile has expired (or will expire), and will ultimately be destroyed. It's possible that this unused excess stockpile will cost up to several hundred million dollars (in the United States alone).

I personally won't weigh in on the necessity or wisdom of this H1N1 vaccine stockpile, but bringing down the cost of producing influenza vaccines in a future emergency should be a priority. The research of Mario Alvarez (Tecnologico de Monterrey, Mexico) and coworkers should be of help in this respect.

They have developed a method of quickly producing incomplete influenza viruses in bacteria. These incomplete viruses can nevertheless serve as effective vaccines, avoiding the need for more complex production currently available on a commercial level.

Why haven't bacteria been used for vaccine production so far?

Current commercial influenza vaccine production starts with infecting early-stage chicken embryos, one per vaccine dose (that's hundreds of millions of eggs), followed by extensive purification. If bacteria could be used for vaccine production instead, production could be much faster.

On the other hand, scientists often think that glycosylation of the viruses (attaching sugar units to them), which is important for generating a fully functional virus, is important for an effective vaccine as well. This would seem to rule out the classes of bacteria that cannot perform glycosylation.

Alvarez and coworkers have countered this common assumption. Their bacterial factories produce viral subunits that are incomplete yet are nevertheless effective vaccines, judged by their efficacy in ferrets against H1N1 infection.

Synthesizing and isolating viral proteins.

The scientists genetically engineered their bacteria, Escherichia coli, to synthesize part of the hemagglutinin protein of the influenza virus. A hemagglutinin protein causes red blood cells to clump together, functioning in the immune response.

The viral protein segment synthesized by the bacteria is highly conserved. This means that almost all H1N1 viruses possess it.

Why is this important? Viruses tend to rapidly mutate; consequently, you want your immune system to recognize the H1N1 virus whether or not it has mutated somewhat.

The scientists deliberately left off certain water-repelling segments of their viral protein fragment, to facilitate successful expression in bacteria. Through extensive trial and error, they observed a protein production rate of 3.4 grams per liter, at a scale of 5L, in only 12 hours.

The vaccine protects ferrets from H1N1 infection.

The scientists found that over 90% of the viral protein fragment was recognized in the blood serum of human patients who had previously contracted H1N1, and had therefore developed antibodies to the virus. In these lab tests, the vaccine was quite effective.

They next turned to ferrets, 16 of whom were immunized with the vaccine and given a booster (additional dose) at day 15. All of these ferrets exhibited an immune response less than a week after the booster.

Five ferrets were not vaccinated. All 21 of them were intranasally infected with H1N1 (the exposure was 45 days after immunization).

The scientists developed an "overall sickness index" to quantitate protection against viral infection, including four symptoms (change in weight, temperature, sneezing, and mucus). By this classification, 3 of the 16 vaccinated ferrets had moderate symptoms, and only 1 had severe symptoms.

In contrast, all of the unvaccinated ferrets had severe symptoms, and were humanely killed. Although immunization was clearly less than perfect, it did provide a great deal of protection.

Overall evaluation.

The scientists have not yet optimzed their vaccine dosages. It's presently approximately 10 times greater than current commercial influenza vaccines.

Nevertheless, their bacteria can produce over a thousand times the amount of vaccine, per unit of time and volume, than current commercial technology. Vaccine production could conceivably could be scaled up to roughly 500 million vaccine doses per month in a mid-size, appropriately-equipped pharmaceutical factory.

Their process is clearly more efficient than the "hundreds of million of eggs approach" to vaccine production available today. Hopefully future work will rigorously test vaccine efficacy in humans.

Also note that, with additional research, the concept reported herein could be extended to producing a wide range of antiviral vaccines, not simply those that protect against H1N1. The research reported by Alverez and coworkers should be very valuable for quickly and relatively cheaply synthesizing large quantities of antiviral vaccines, as long as the antigenic (immune system-activating) protein is known.

ResearchBlogging.org for more information:
Aguilar-Yáñez, J., Portillo-Lara, R., Mendoza-Ochoa, G., García-Echauri, S., López-Pacheco, F., Bulnes-Abundis, D., Salgado-Gallegos, J., Lara-Mayorga, I., Webb-Vargas, Y., León-Angel, F., Rivero-Aranda, R., Oropeza-Almazán, Y., Ruiz-Palacios, G., Zertuche-Guerra, M., DuBois, R., White, S., Schultz-Cherry, S., Russell, C., & Alvarez, M. (2010). An Influenza A/H1N1/2009 Hemagglutinin Vaccine Produced in Escherichia coli PLoS ONE, 5 (7) DOI: 10.1371/journal.pone.0011694