The New Zealand-born Williams was tempted back to this part of the world with the offer of the directorship of the Monash Institute of Medical Research (MIMR), which he took up at the start of 2006.

In Melbourne, he is continuing to pursue his wide research interests while also dealing with the challenge of directing a large research institute and heading its Centre for Cancer Research. This centre is further investigating the study of why siRNAs activate innate immunity through their interaction with the RIG-I helicase.

"We were very interested in determining how siRNAs could activate RIG-I because the size wasn't an issue with RIG-I. There seemed to be something about the structure of the siRNAs.

"We managed to determine that the end structure of the siRNAs was important for RIG-I activation, the 3' end. Where you had a two-nucleotide 3' overhang, these were not very efficient activators of RIG-I.

"But if you had a blunt end molecule - and these are molecules that are substrates of the enzyme Dicer - these got incorporated into the silencing complexes much more efficiently, but it turns out that those ones are very active in activating innate immunity via their ability to turn on RIG-I.

"We discovered that you could avoid it by making siRNAs that were exact replicas of what is made in vivo, under natural circumstances."

Bifunctional siRNAs

Williams and his team are not actively involved in developing or refining siRNA technology, but they are interested in looking at specific targets that the technology could be useful for.

"The obvious targets in cancers are oncogenes, so genes whose transcripts are really driving the tumourogenic phenotype," he says.

"There's also some targets we are looking at that appear to be implicated in metastasis - the spread of cancer is something we all want to work on. And we've identified some targets that might be unique in that area."

While this understanding is helping fine-tune RNAi technologies, Williams is now looking at how to harness the immune response in an intriguing way - by giving cancers the double whammy of silencing an oncogene, for example, and then stimulating an immune response as a second line of defence.

He is interested in developing what he calls 'bifunctional siRNA' - siRNAs that target the disease-causing gene or particular receptors expressed on the surface of tumour cells, but that will also stimulate an immune response.

"We are quite interested in exploiting the findings that we have that siRNAs can activate innate immunity by making siRNAs that have these two functions.

"So we are building two functions into one molecule, if you like. I think it's a neat idea. We have gone some way along that road - we've managed to generate siRNAs that are very good at doing both activities, so they target the disease-causing gene and they stimulate the innate immune system."

In viruses, of course, you don't want to be doing that. Williams and colleagues have been working closely with various biotechnology companies to try to identify all of the aspects of siRNAs that would activate innate immunity and then trying to blunt those without disrupting their silencing activity when it comes to fighting viruses.

"It's a bit of a challenge I have to say, because usually if you blunt the siRNAs' stimulating activity you also blunt their own activity, so it's a bit of a balance."

It is still a surprise to many that despite our long understanding of post-transcriptional gene silencing in plants, it was only in 1995 that RNAi was first tracked down in C. elegans. It is a very short time indeed from that seminal discovery to early phase trials in humans, most notably in macular degeneration and Respiratory Syncytial Virus (RSV), which Williams has also investigated.

There is still a long way to go, however. "There are already some molecules in phase I trials so it hasn't taken long from the discovery right through to being first into patients, but I think it will take a while to develop," he says.

"And the delivery issue is a major one that has to be solved. I think it is likely that there will be some interesting advances in that area, but whether they will get through to being a drug or not will take a number of years to determine.

"Cancer drugs take an awful long time to get from discovery to approval, so it will take a while. I think I'd describe myself as being reasonably optimistic that they will have a role to play as therapeutic agents but it is going to take some time."

In the meantime, the technology is remarkable for what it can do for basic research.

"In terms of the pharmaceutical industry, where this technology has been most interesting is in target validation. So if you want to develop a small molecule drug towards a specific target and you want to determine what the effect is in taking that target out of a particular pathway, you can use an siRNA to turn down the expression of the target and that gives you an idea of what the phenotype is going to be.

"But then of course for all of us in basic research it has been remarkable. We all use siRNA technology on a daily basis now."