Using a short interfering RNA library to screen roughly 700 protein kinases, a research team led by scientists at Imperial College London has identified the protein LMTK3 as a potential drug response biomarker in breast cancer patients and a possible therapeutic target for the disease.
The protein appears to be a key regulator of estrogen receptor α, which plays a key role in patient response to endocrine therapies like tamoxifen, aromatase inhibitors, and fulvestrant, which is marketed by AstraZeneca under the brand name brand name Faslodex.
In the study, which was published this week in the online edition of Nature Medicine, the researchers demonstrated that LMTK3 regulates ERα by phosphorylating it to protect it from degradation and showed via immunohistochemistry done in 613 breast cancer samples that high LMTK3 expression levels are associated with significantly shorter disease-free survival and overall survival.
Protein kinases are key targets in drug development, and protein phosphorylation signatures are under investigation as prognostic and diagnostic tools for a variety of cancers including breast cancer (PM 11/12/2010). The Imperial College team was led to focus on protein kinases based on previous research showing that phosphorylation plays a key role in ERα activity, said Georgios Giamas, a postdoctoral researcher at the college and an author on the paper.
"We decided to focus on kinases because estrogen receptor had been shown to be a target for phosphorylation, and it appears that phosphorylation affects many of [its] functions," he told ProteoMonitor. The researchers chose to use siRNAs for the kinome screening, Giamas said, due to the lack of specific inhibitors for the large number of kinases they wanted to look at.
A significant challenge to the study, Giamas noted, was devising a readout for ERα activity. ERα RNA and protein expression levels don't always correspond to activation levels, and doing phosphorylation reporter assays to measure changes to phospho-levels at various ERα phosphor-sites in response to kinase inhibition would require "a huge amount of work" given the roughly 700 proteins being screened, he said.
Given these constraints, the researchers chose to use expression of the estrogen-responsive TFF1 gene as a readout for the altered ERα activity, detecting levels via real-time PCR – a technique, Giamas said, not typically used for this sort of screening assay given the time and expense of running the large number of reactions required.
"We looked at almost 700 kinases, doing [the initial discovery round] in triplicate" and then following that up with two additional smaller rounds," he said. "At the end of the day, we ended up doing [roughly] 2,500 real-time PCR reactions."
This, plus the siRNA kinase inhibition library, cost around £20,000 ($33,000), Giamas estimated. The screen "was very time consuming and very expensive," he said. "It was a risk, but this is research, and at the end of the day you have to risk at some point. So we thought it was worth doing."
Ultimately, he said, the screen "worked very well," providing the researchers with positive confirmation of several kinases identified in the past as regulating ERα activity, as well as identifying a number of kinases – like LMTK3 – not previously known to be involved in ERα regulation.
While they plan to investigate some of these other proteins in the future, the researchers chose to focus more immediately on LMTK3 based on its novelty and evolutionary history, Giamas said.
"We wanted something that was new, something that hadn't been identified so far, and it turns out that nothing is known about this kinase at all," he said. And, indeed, a PubMed search reveals only two other papers mentioning LMTK3 – in both cases as a kinase other siRNA screens identified as worthy of further investigation with regard to colon cancer and leukemia.
Evolutionarily, the researchers were interested in investigating a kinase that had been positively selected for between humans and chimpanzees, under the theory that this selection might have altered the substrate binding characteristics of the protein in humans and led to increased human susceptibility to breast cancer.
In work done since the submission of the Nature Medicine paper, the researchers have confirmed the results from the 613-sample study linking LMTK3 expression to breast cancer patient survival rates in another blind patient cohort, Giamas said, and have submitted this new data for publication.
The researchers have also patented the protein as a potential drug target for breast cancer patients, and are now working to better characterize it in hopes of developing an inhibitor for it in the future.
"From a clinical point of view, from the studies we've done so far, it's pretty clear that [LMTK3] is involved in and important for breast cancer progression and overall survival of patients," Giamas said. "The main steps now are trying to characterize this kinase and make an inhibitor against it – first to understand how [LMTK3] is working and then to be able to use it in the future [as a therapeutic]."
As of yet the team hasn't contacted any pharmaceutical or biotech firms about collaborating on this work, he said, but they "are in contact and in discussions with" several other Imperial College researchers who have expertise in inhibitor development.
"This is not something you do [quickly]," Giamas added. Developing such a molecule "could take five years, it could take 10 years, depending on how lucky we are with the crystallography data [characterizing the protein] and how lucky we are with the chemistry of developing a specific inhibitor."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.