The opposing sides were represented by John Chiorini of the National Institutes of Health fighting for viral delivery, and Leaf Huang of the University of North Carolina who spoke in support of non-viral delivery. An initial vote showed that Chiorini had a lot of work on his hands to convince the audience of his argument, with the majority of his listeners already siding with the opposition. However, he pushed on, laying the foundations of his argument and outlining some of the factors that are key in gene transfer vectors and how viral vectors satisfied these requirements more satisfactorily than non-viral agents. "There's going to be no one ideal vector for every application, and I think you're going to have to tailor it, " he cautioned, "but I think there's going to be some common themes in all the different vectors." Critical features such as tolerability, stability, efficiency, targeting, how regulatable it is, long term expression, the ability to modify and ease of manufacture were all highlighted by Chiorini as requisite in an ideal vector. Viral vectors do well in fulfilling these criteria, according to Chiorini, in particular achieving the crucial requirements of very efficient gene transfer and good tolerability - "some viruses are actually considered non-pathogenic," he argued - as well as long-term gene transfer without eliciting an immune response. Chiorini was keen to stress that viruses are essentially "nature's gene delivery service," having specifically evolved over millions of years to be able to deliver nucleic acid into a cell in a manner that is tolerated by the host, and as such are perfectly designed for use in gene therapies. However, Chiorini had a fight on his hands, and Leaf Huang took to the podium to defend non-viral gene delivery and present his case to the audience. "The biggest advantage of non-viral systems is that they are non-toxic," claimed Huang in his opening statement. "Even the safest viral vectors, like adeno-associated viral vectors and lentiviral vectors, are still immunogenic." To further drive home his argument, Huang made the point that in all gene therapy clinical trials where death has occurred, the viral vector itself was the cause of death. Huang went on to describe a particular method of non-viral gene transfer, hydrodynamic injection of naked DNA, which leads to very high and prolonged transgene expression in the liver. After injection into the tail vein of a mouse, this technique resulted in hundreds of days' of gene expression at an extremely high level, according to Huang. "I challenge any viral vector to do this with one single injection," he said. Huang also went on to describe an interesting alternative method of non-viral delivery, known as mechanical massage. Although eliciting some amused responses from the assembled crowd, the technique involves injecting into a mouse tail vein, then pressing down four times on the stomach of the mouse - this method also resulted in a high level of gene expression in the liver, according to Huang. Huang also argued that the targeting ability and flexibility of non-viral vectors was a key point in the favour of the delivery system, again showing an example from his research in which 70-80 per cent of the injected dose reached the site of the targeted tumour. One point of contention between the two sides focused on the efficiency of the two techniques, specifically in regard to the increase volume of DNA that has to be injected when employing non-viral techniques. While Huang argued that as long as the dose is affordable and non-toxic it is perfectably acceptable to inject a higher volume than common with viral vectors, Chiorini claimed that any peptides that are in non-viral vectors are going to be immunogenic in some way, and that large amounts of material are simply going to be more immunogenic than less material - "less is better in this case," he said. As to the risk of insertional mutagenesis with viral vectors and the cases of the x-Severe Combined Immunodeficiency syndrome (x-SCID) trials in France that were found to cause leukaemia in some children, Chiorini pointed out that only a small number of subjects were affected, and that as soon as the observation was made the trial design was modified and no further occurrences have been reported. Despite the fact that Chirorini noted that there are now trials starting using viral vectors for the treatment of blindness and haemophilia, and viral vectors being the treatment of choice in patients for whom a matching bone marrow transplant cannot be found, the audience appeared reluctant to share his level of confidence in the viral technique. With both opponents admitting that each other's approach had significant points in its favour, and going as far as to say that they both in fact work to some extent using the opposition's vectors, the session ended on a light-hearted note with the debate having generated significant discussion and involvement from the assembled audience. Despite a valiant effort, however, Chiorini was unable to win over the crowd, with non-viral vectors emerging as the favourite in the future of gene delivery.