Lorne Protein report: Tagging proteases

A US researcher has described to the Lorne Protein Conference a powerful new technique for 'tagging' protease enzymes in living cells, illuminating potential new drug targets for parasitic diseases like malaria, and metastatic cancers.

Dr Matt Bogyo, of the Department of Pathology at Stanford University School of Medicine, said the small, fluorescent chemical probes technique could also be used to track the cellular processes involved in the progression of infection or metastasis, and to visualise the effects of selective 'chemical knockouts' of proteases in lysed or living cells, or even in live animals.

Proteases -- protein-digesting enzymes -- are ubiquitous, diverse, and functionally indispensable in eukaryotic and prokaryotic cells. The Human Genome Program has catalogued more than 500 protease genes, while the malaria parasite, Plasmodium falciparum, has at least 90. All bacteria and viruses employ proteases to infect their host species.

Bogyo's team has synthesised a diverse suite of 'plug drugs', each designed to resemble the natural substrate of its target enzyme. The molecules are equipped with reactive 'warheads' that cause them to bind irreversibly to the target protease's active site, and have a fluorophore 'tail' that glows at a particular wavelength under laser light.

Within minutes, the fluorescence provides quantitative readout of the amount of the target protease in the cell, and illuminates where it is active -- whether it is secreted, or active at the cell membrane, or within the cytosol.

"The probes only bind to an enzyme if it is active, so they provide a readout of enzyme activity," Bogyo said.

Bogyo's team has used radio-labelled chemical probes to study the role of proteases in the life cycle of the malaria parasite. "The parasites go through their life cycle in erythrocytes within 25 hours, and we take extracts from each stage to see which proteases are turned on or off," he said.

"The probes typically aren't great drugs, but they can be useful tools for screening for selective inhibitors. When we selectively turned off the activity of one protease, falcipain, the parasite was unable to invade red blood cells after completing its life cycle. So it's an interesting tool for drug discovery."

In a murine model of metastatic cancer of the pancreas, Bogyo's team was able to observe the multi-stage process of tumorigenesis in insulin-secreting beta cells in islet tissues, which is normally accompanied by rapid growth of new blood vessels.

They monitored the expression profiles of three different cathepsin proteases -- A, B and C -- in the islet tissues, which comprise a mixed population of cancerous beta cells, fibroblasts, endothelial cells, and invading inflammatory cells from the immune system.

As the beta cells began to multiply and form tumours, the researchers observed a huge increase in protease activation, particularly in the lysosomal compartments of the inflammatory cells. Bogyo said there was strong protease activity in immune-system cells at the leading margins of the tumours, adjacent to healthy tissues, suggesting that protease activity in inflammatory cells may increase the invasiveness of tumours.

Bogyo said his team administered an inhibitory probe for cathepsin to living mice with pancreatric tumours at age 10 weeks. Normally, the tumours are lethal at 15 weeks, but in the treated mice, the tumours regressed dramatically in volume, and the mice remained healthy at least to 16 weeks, when the experiment was terminated.

There appeared to be no toxicity associated with the treatment, which reduced protease activity by 90 per cent -- in fact, fluorescent probes administered to them indicated no protease activity in their cancerous cells.

Bogyo's team suspects blocking cathepsin activity inhibits the development of new blood vessels that fuel tumour growth.

"Identifying the substrates for proteases is extremely difficult, but our probes could be very good diagnostic markers of disease," he said. "We could perform non-invasive imaging, using using positron emission tomography or near infra-red fluorescence, to identify of patterns protease activity associated with particular diseases."

Glove drug

In the same session at Lorne, Prof David Fairlie, of the Institute for Molecular Biology (IMB) at the University of Queensland, described his team's work to identify structural themes in the active sites of proteases, as potential targets for inhibitory 'plug drugs' to prevent infectious and parasitic diseases.

Fairlie said his team had shown that all proteases, irrespective of their selectivity for particular amino acid residues -- serine, cysteine, metallo -- recognise essentially the same basic structures, in the same conformation in their substrate proteins: beta sheets and beta barrels.

Fairlie said the world's second deadliest parasitic disease after malaria, schistosomiaisis, infected at least 250 million people around the world, mainly in Africa and Asia, and was conservatively responsible for some 250,000 deaths a year.

The worm-transmitted disease, which has aquatic snails as an intermediate host, feeds on haemoglobin, digesting it with a protease enzyme, schistosomal cathepsin D.

If an inhibitor of cathepsin D could be designed, that was selective for schistosomal cathepsin D, but not for human cathepsin D, it might be possible to starve the parasites by disrupting their ability to digest haemoglobin.

Fairlie's research team has identified a structural element of the active site of the schistosomal cathepsin D that could be exploited selectively. The researchers have developed some experimental inhibitors, and were able to enhance the activity of the first one they made by a factor of 600, by strategically adding small-molecle 'foliage' to the backbone of the drug, so that it bound to complementary residues around the enzyme's active site.

Such inhibitory molecules would act more like a flexible 'glove' with a broad grip on the active site, than a high-affinity 'plug drug' designed to anchor itself to a particular residue in the active site.

By providing more points of contact between the drug, Fairlie believes it should be possible to minimise the risk of a parasite, bacterium or virus evolving resistance by randomly mutating the 'anchor' residue.

Fairlie said that despite intensive research into proteases as drug targets, only two types of human protease-inhibitor drugs are currently in clinical use -- one for the AIDS virus, and ACE inhibitors, which have been used for more than 30 years to treat high blood pressure.

He said scores of protease inhibitor molecules were currently in clinical trials, but some promising candidates had failed because of side-effects brought on by a lack of selectivity. With advances in genomics, it would be possible to design protease inhibitors that were more potent, more selective, and more bio-available, he said.

His team's 'glove drug' strategy might be a fruitful approach to developing new drugs, particularly for infectious viruses and the parasites that cause malaria and schiztosomiasis, to prevent them evolving resistance.

Graeme O'Neill will file daily reports on highlights from the Lorne Protein Conference. A special wrap-up of the Lorne Protein, Cancer and Genome conferences will be published in the March issue of Australian Life Scientist