During the course of a year there are some 40 billion chickens in the world (with 40 billion new chickens in the succeeding year), some six times the human population. Virtually all of them will catch IB. At its simplest IBV causes in chickens what we, in humans, would call the common cold (though IBV is very different genetically from human coronaviruses and does not cause disease in humans). At best infection with IB slows down the growth of meat-type chickens, making them uneconomic. Unfortunately the birds also get secondary bacterial infections, which can be fatal to the young ones. Infection of egg-layers results in a precipitous loss of production, which usually does not get back to normal.

There is a lot to be gained if we can make new vaccines quicker, vaccines that will not revert to virulence in the field, vaccines that are easy to distinguish from the their virulent counterparts. We and others have shown that it is the large surface spike protein that induces protective immune responses. Moreover, the virus avoids destruction by the immune system by virtue of great diversity exhibited by the spike protein. If we could replace the spike protein gene of a vaccine strain with that from a current field strain, that might be sufficient to make the old vaccine effective against the new field strain. That is one of the notions that we are testing with our infectious clone system.

In recent years we have diverted some of our attention towards turkeys and pheasants, both of which are infected with coronaviruses that are genetically very similar to IBV. We would like to know the molecular basis for the host restriction of the avian coronaviruses.

  What first interested you in a biological career?

As a child of 11 or so I was into doing experiments at home. I recall a chemistry set and getting books describing experiments from the local library. My concentration on biology was one of those twists of fate. At the age of 14 I opted for art as an optional subject. After one lesson in the new academic year I switched to biology, a decision I’ve never regretted. I chose to read microbiology at university (Reading), following a hunch that I would enjoy it. Reading University was strong on virology, no doubt a factor in my choosing to specialize in that area. I was then fortunate to get a graduate research student position at what was then called the Animal Virus Research Institute, at Pirbright. I owe a lot to my main mentors there, Dave Sangar, Dave Rowlands, and Fred Brown.

  How did you come to work on coronaviruses?

After Pirbright I dipped my toe briefly into the world of immunology. No one immunologist seemed able to agree with another, so I returned to the apparently more rational field of virology. After working on tissue-tropism aspects of human influenza virus (and managing to remain unbitten by our ferrets) at the University of Birmingham, I had a desire to analyse viruses at greater depth. The first time that I read the job opportunities pages of New Scientist I saw a position at the Houghton Poultry Research Station (now part of the Institute for Animal Health), near Cambridge, to work on IBV. I applied, got the job and—here I am. (We transferred from the Houghton Laboratory to the Compton Laboratory in 1992).

  Has the current concern about SARS affected the course of your research?

In the very short term, yes. Information about the etiology of the disease was coming thick and fast on the web; reading and discussing that distracted me from my established tasks. I was initially apprehensive talking to journalists but became impressed with the quality of their questions. The emergence of the SARS coronavirus has made the public and the authorities more aware of what virologists have always suspected; that there are lots of unknown viruses out there, and that occasionally they can jump species, sometimes with nasty consequences. I believe that we should be doing more to discover what is out there; this is a feature of current discussions with research commissioners.

  How important are these viruses in human disease?

Before SARS there were two human coronaviruses known to us. These are genetically very different from each other, although they cause similar upper respiratory tract disease. They cause about 25% of common colds. Given that these are rarely life-threatening, coronaviruses have not been, until recently, at the forefront of peoples’ minds. Less commonly known is that human coronaviruses are associated with enteric infections. This should come as no surprise as most of the known animal coronaviruses are associated with respiratory or enteric infections—or both, though any one strain of a virus tends to be more damaging in one site or the other. "My" virus, IBV, although most obviously causing upper respiratory infection, also replicates in the oviduct and kidney. Some strains of IBV cause severe kidney disease, killing the chicken. IBV also grows in many gut tissues, though without causing pathology. In contrast, the turkey coronavirus that we know of causes enteritis.

  Where do you see this research going in 10 years from now?

The turn of the century ushered in a new era for coronavirus research; systems were developed whereby virus could be recovered from full-length DNA copies of several coronavirus RNA genomes, including ours for IBV. This had been a long time coming, as coronaviruses have the largest RNA genomes, 28,000-plus nucleotides. Coronavirologists can now modify any gene—to study how the virus replicates, causes disease, induces/avoids immunity—in a much more precise manner than heretofore. This technological development may also lead to a new generation of vaccines—rationally designed vaccines, in which highly specific modifications are made to the virus genome, based on the results of research. In the case of IBV this would include swapping spike protein genes, something that we have done recently, to combat new serotypes (Casais et al., 2003). The accessory genes—those encoding proteins that do not become part of virus particles—are being found to be non-essential for replication per se. Deletion of these from murine coronavirus has already been found to attenuate the pathogenicity of the virus, with obvious potential for vaccine design with other coronaviruses. It will also be possible to add "signatures" to a vaccinial strain, for ease of differentiation from similar field strains.