Proteomics at the hub of prostate cancer research

Whether it is mastering a new suite of expensive protein analysis equipment or the basics of 4WD survival in the outback, Associate Professor Terry Walsh applies the same calm, intuitive logic and intellect that inspires and reassures those who come to know him.

As one of those lucky ones, this reporter was delighted with her recent assignment to find out how Walsh and the Queensland University of Technology's new Institute for Health and Biomedical Innovation (IHBI) are using proteomics to develop better diagnostic and therapeutic targets for prostate cancer.

The newly housed institute integrates health, science and biomedical engineering research. Based at QUT's Kelvin Grove campus in Brisbane and affiliated with nationally recognised hospitals and health care service providers, IHBI seeks to build on QUT's track record in successful commercial partnerships and biotechnology applications.

The research at IHBI encompasses several areas ranging from human health and wellbeing, injury prevention and rehabilitation to tropical crops and biocommodities to cells and tissue, which is where Terry Walsh's proteomics work fits in.

The research that Walsh oversees is predominantly in the hormone-dependent cancer program of IHBI, which was established and is headed by Professor Judith Clements; several other programs in the cells and tissue domain of the institute are now also moving into the proteomics area.

The cancer program comprises eight or nine groups with the common aim of better understanding the molecular and cellular basis of prostate cancer, one of several hormone-dependent cancers and one of the most common cancers globally.

Relatively little is known about these cancers' epidemiology or underlying cellular and molecular biology. Walsh's research is centred on the role of proteases in the biological activation and regulation of prostate cancer, particularly from a structural point of view, and high-throughput proteomics analysis is now a big part of that thrust.

Hormone-dependant cancers

The current workhorse of Walsh's proteomics approach is the ProteomeLab PF2D protein fractionation system from Beckman Coulter.

The main application of the system thus far has been in prostate cancer research, looking at either samples from disease cohorts - men with prostate cancer confirmed by biopsy and those that have tested negative by this method.

"We have a large bank of samples (serum and ejaculates) available through excellent clinical collaborations at the Princess Alexandra and Royal Brisbane Hospitals," Walsh says.

"Professor Clements also leads the Australian Prostate Cancer Bioresource initiative, which has the potential to extend the number and scope of samples."

They also use prostate cancer cell lines generated from primary tumours or normal cells transfected with a molecule known to be associated with the disease.

"There are several identified differences between prostate cancer cells and normal cells, so we can take the genes that encode for these differences and put them into normal cell lines to analyse differences in protein profiles between transfected and non-transfected cells."

The work has concentrated in particular on the kallikrein family of proteolytic enzymes and its role in prostate cancer, a long-standing focus of the hormone-dependent cancers program.

Judith Clement's group originally characterised the human kallikrein locus and identified 11 of the newer kallikreins. There are 15 members of this family in total and a number of them (eg. 4 and 14) have been linked to the development of prostate and other hormone-dependent cancers such as breast and ovarian cancer.

Kallikrein 3 is the marker used clinically in blood tests used to screen for prostate cancer, called the prostate-specific antigen or PSA test.

However, PSA is not perfect due to issues such as sensitivity, and a better marker is clearly needed. In fact, despite an increase in awareness, routine PSA testing and the range of treatment options available, there have been essentially no major advances in prostate cancer management in the past decade, Walsh says.

There is a particular need to detect early signs of the disease and predict aggressiveness of the cancer and the information would greatly improve treatment options.

For example, many men with a high PSA reading have a form of the cancer that will never progress to a life-threatening stage, not warranting radical treatment; however, this cannot be determined until much further down the track.

"We are primarily hunting for diagnostic targets but quite often when you identify a new diagnostic marker it ends up also being important therapeutically. Also, sometimes the difference can be not just diagnostic but also prognostic."

Future studies will continue to examine the key relationships between the kallikrein proteases and prostate cancer cells. In addition, bone cells will be investigated to try and determine why prostate cancer has a predilection for bone metastases.

The proteomics approach, together with a host of other studies conducted in the prostate cancer group, is ultimately geared towards finding a markers or suite of markers that are unique to the cancer, present in different fluids and easy to devise simple tests for such as an antibody-based ELISA or strip test.

"We are still probably five to 10 years down the track from finding that good clinical marker but so far the results look promising," he says.

Protein fractionation

The Beckman Coulter machine that Walsh uses was originally 'acquired' via a postdoctoral fellow and a competition in 2004. The prize was placement of the system in Walsh's lab at QUT for a one-year period with access available to the company as a demonstration model.

During this time, Walsh's team worked out all the appropriate protocols for the cancer samples under investigation and generated a significant amount of useful data. The program decided that the PF2D was worth having permanently and it was subsequently purchased following a successful application for funding from the Australian Research Council.

The PF2D system provides an automated, two-dimensional fractionation system expressly designed for high-resolution analysis of complex protein mixtures including cell lysates. Walsh confirmed with his usual understated acuity that "the system works pretty well" - it can fractionate multiple complex samples and compare protein profiles accurately, quickly and with reasonable throughput.

The proteomics system basically comprises two (and potentially three) chromatography components used sequentially.

"The machine comes supplied with an isoelectric focusing unit followed by a reverse phase setup, but you can put anything at the front end that you want," Walsh says.

Some start with simple ion exchange chromatography to separate proteins from the crude fluid sample based on charge and then analyse the collected fractions by reverse-phase chromatography, which further separates the proteins based on overall size and hydrophobicity.

"The end result is a pattern that the computer software turns into something that looks like a two-dimensional gel with a series of bands," he says.

"The computer software then compares different profiles and highlights 'bands' that are present or elevated in one profile and down or absent in another, for example in samples from patients with and without proven prostate cancer.

"The fractions of interest are analysed further by mass spectrometry to identify the proteins contained within that area of the profile.

"The beauty of it is that it is information intensive - if you have a high throughput mass spec at the end of it, you can mine this information, and analyse a decent number of samples in one run to generate useful results very quickly, which would take a long time by conventional methods."