Taking RNAi from in vitro to in vivo

RNA interference (RNAi) already enjoys routine status in many research laboratories despite the process being described less than a decade ago. Synthetic 'interfering' RNA molecules are used in vitro to investigate gene function across many different applications.

Manipulating this fundamental cellular mechanism for in vivo therapy, however, remains largely a brave but promising new world.

One of the few groups exploring this approach in Australia is led by Dr Nigel McMillan, senior research fellow at the University of Queensland Diamantina Institute for Cancer, Immunology and Metabolic Medicine, which is based at the Princess Alexandra Hospital in Brisbane.

Since 2002, shortly after siRNAs were shown to work in mammalian cells, McMillan's group began investigating cervical cancer therapies based on RNAi gene silencing of cancer-addicted oncogenes.

McMillan established his current laboratory in Professor Ian Frazer's Centre for Immunology and Cancer Research (CICR) in 1995 following a postdoctoral period at the Hospital for Sick Children in Toronto and the Cleveland Clinic.

The CICR recently merged with the Centre for Diabetes and Endocrinology Research to form the new Diamantina Institute, officially opened on the first of January this year. His group has grown from two to 14 in that time, thanks to continuous funding from the NHMRC.

As a self-confessed virologist, McMillan's early interests lay in the diverse means used by viruses to evade the innate immune system. Discovery of the RNAi mechanism provided a clue as to how viruses do this.

By that stage, McMillan was less into viruses and more into cancer, so he decided to focus on the therapeutic use of RNA interference in cervical cancer, which is caused by the human papillomavirus (HPV). "We were interested in both developing RNAi therapies for cervical cancer as a means unto itself, and using this cancer as a model system to develop RNAi as a therapeutic tool," McMillan says.

The process of RNA interference was initially described in plants, but it was a series of elegant worm experiments published in Nature in 1998 by Americans, Andrew Fire and Craig Mello that established this previously unrecognised cellular mechanism for regulating gene expression as an entire new field of biology.

Fire and Mello introduced double-stranded RNA (dsRNA) complementary to the coding sequence of a muscle protein into C. elegans, and instead of the enhancement effect they were expecting, gene expression was turned off and the protein not produced.

It seemed that the dsRNA activated cellular machinery to catalytically degrade the target mRNA, thereby post-transcriptionally regulating gene expression via protein synthesis.

Until then it was easy to add gene expression to a cell but almost impossible to take it away. RNAi is now used widely as an experimental tool to investigate a protein or pathway of choice or for conducting screens of whole cells or organisms by shutting down one gene at a time, and asking what happens to a particular function or process.

We now know that gene silencing via RNAi happens in all plants, animals and lower organisms, and that RNAi comes in a number of different forms - short-interfering RNA (siRNA), microRNA (miRNA) and short-hairpin RNA (shRNA).

RNAi has also emerged as a powerful new technology to treat virtually any disease with a genetic basis using normal cellular machinery and synthetic molecules that are easy and cheap to produce.

RNAi and cancer

McMillan's primary focus is cervical cancer, and of course he is surrounded by plenty of expertise in this area. He and his colleagues quickly realised the enormous potential of RNAi-based therapies for their particular cancer because the viral oncogenes that cause cervical cancer, E6 and E7, are expressed only in tumour cells, and the cells are addicted to these genes.

For instance HeLa cells (originated from cervical cancer tissue) have been cultured now for over 50 years, but if you knock out these two genes, they die. "Even after all this time in tissue culture being abused by students and postdocs, those oncogenes do such a good job that they have survived," he says.

The approach showed promise immediately - RNAi moieties designed to knock out either E6 or E7 were put into HeLa cells in culture, which in all cases stopped growing and died.

McMillan then took this work one step further to show that combining RNAi with cisplatin (a chemotherapeutic agent used for the treatment of advanced cervical cancer) enhanced the sensitivity of the cancer cells to chemotherapy by almost four-fold. This means four times less nasty drug being used, which is immediately better for the patient.

A major issue became obvious right from these earliest experiments - delivery of the dsRNA. RNA is notoriously unstable in biological systems, especially serum, with a half-life of about six seconds.

The group initially developed a means of delivering shRNA molecules inside a lentivirus particle. It works well in the tissue culture dish and more recently in a mouse model of cervical cancer.

"At low doses of virus the tumours stopped growing and at high doses no tumours formed," McMillan says.

These findings were exciting and provided proof of principle that the shRNA was delivered to the cancer site, got rid of the tumour cells and that the effect was dose dependent.

The RNAi effect also worked to clear tumour cells in a lung metastasis model simulating cervical cancer secondaries.

"So, this worked really well but still does not solve the delivery problem because what we were doing is essentially gene therapy, which is not an acceptable way to cure cancer in terms of regulatory authorities and clinical acceptance."

Delivery systems

Current work in McMillan's group is therefore centred on better ways to deliver this technology, proven by them to work for cervical cancer. In one approach, they are developing delivery systems involving nanoparticles and liposomes together with Nigel Davies in the School of Pharmacy at UQ.

"One of our lipid-rich vehicles for the systemic delivery of shRNA has a few little trick," McMillan says. "These particles do not get stuck in the lungs, which is usually a problem with liposomes, and they can be targeted to the tumour cells."

McMillan is also working with a company in America to test using some interesting chemical stabilisation methods for siRNAs. Half-lives can now reach up to 48 hours, which makes delivery of naked molecules clinically feasible.

Despite the many delivery problems, the feeling at the recent international Keystone meeting on the potential of RNAi in therapy was positive. McMillan says.

"I think we will see small breakthroughs for specific cancers like cervical cancer and definitely for viral diseases in the near future, although for other cancers involving self-genes, the issue is far more complex."

His enthusiasm for this new field, however, is evident. "What you realised at this meeting was how far we have come - RNAi in mammals was only described six years ago and we already have a Phase I trial completed late last year for the treatment of macular degeneration and many other in the pipeline. We really are looking at the next revolution in therapy."