Researchers at Massachusetts General Hospital last week began the final year of a four-year NIH-funded study to develop a model of drug-induced cardiac repolarization abnormalities in larval zebrafish.
Working with a $414,000 grant from the National Institute of General Medical Sciences, the investigators are now developing a panel of reporter fish modeling repolarization reserve, and several human proarrythmic disease states, a Mass General official told CBA News this week.
They will use these reporter fish to generate a profile of repolarization responses for cardiotoxic or non-cardiotoxic drugs in order to identify features that could help predict proarrythmia in human patients, said Calum MacRae, an assistant professor of medicine and the principal investigator on the grant.
“The grant that we have is to try and build an empiric algorithm using our initial low-resolution/high-throughput assays, as well as sequential assays with much higher resolution, which we have also developed using zebrafish,” he said.
The researchers will combine those assays with disease models and environmental variables so that they can develop a particular profile of activities in their model system that can predict clinical events, including toxicity.
“If we do that, we can develop a more precise risk analysis of a given drug that allows the matching of the drug’s efficacy with the appropriate risk,” said MacRae.
MacRae and his research group have been developing quantitative metrics to predict risk based on compound profiling in vivo. “Our endpoint initially is heart rate with [ atrioventricular] block, but in our second-round assays, it would be action potential duration at 90 percent repolarization, as well as calcium transient duration.”
Predicting Heart Risk
Drugs that cause the particular kind of cardiotoxicity that the researchers are studying are designed to slow the heart and prolong the action potential duration, which results in atrioventricular, or AV, block.
MacRae said his team has developed an automated in vivo assay that allows them to study how drugs cause toxicity. He said the assay can screen hundreds of fish per hour in a 96- or 384-well plate format.
Scientists have found that researchers can’t readily predict how the majority of drugs that cause AV block will affect an intact organism, although these drugs share a specific activity against a particular potassium channel in the heart called iKR.
“As people have better understood the mechanism of action of some of these cardiotoxic effects, we have seen that it requires a more and more complex assay system to recapitulate each of these mechanisms. We find that the zebrafish allow us to do that.”
MacRae said that he and his team have very high-throughput assays that can test multiple parameters, and for any given compound, “we can test not only what it does in wild-type organisms, we can understand what it does in disease models, particularly those that are relevant to the clinical situation where the toxicity occurs,” MacRae explained.
The assays can also help the investigators better understand how a patient’s polypharmacy, diet, or the environment may affect the metabolism of the drug being tested.
“Because we have an assay that is very high throughput, we can test all of those in very rapid succession, and so we can begin to build a profile of the things that predict eventual clinical toxicity,” MacRae said. “As people have better understood the mechanism of action of some of these cardiotoxic effects, we have seen that it requires a more and more complex assay system to recapitulate each of these mechanisms. We find that the zebrafish allow us to do that.”
MacRae's work has been “instrumental in establishing the zebrafish as a biologically relevant model system for the study of QT prolongation and cardiac arrhythmias,” Peter Eimon, director of research for Zygogen, told CBA News in an e-mail. He added that MacRae’s lab has done “some great work using electrocardiograms to show that drugs associated with QT prolongation in humans have the same physiological effect on zebrafish.”
According to Eimon, Zygogen has “received scientific input and advice” from MacRae, though it was not immediately clear if it was performed for a fee. Also, he said his firm has “worked [with MacRae] in the past, specifically when we were in the process of developing, automating, and validating our own zebrafish cardiotoxicity assay.” Eimon said that Zygogen is “not currently collaborating” with MacRae.
Performing cardiotoxicity profiling in zebrafish offers some advantages over current cell-based assays, said Eimon. Specifically, since fish embryos are small enough and cheap enough to be screened in multi-well plates, they allow drugs to be rapidly evaluated in an intact organism with all the associated biological complexity.
“This means that effects of both the parent compound and its metabolites will be picked up,” he said. “It also means that the interplay of drug candidates on multiple molecular pathways can be assessed simultaneously and drug-drug interactions can be readily evaluated.”
“Given the large amount of background work that's already been done on the zebrafish cardiovascular system by MacRae and other labs, the time seems ripe for developing these kindd of finely-tuned zebrafish models,” said Eimon. “I know from personal experience that the use of transgenic reporters in particular can significantly increase both the throughput and reproducibility of zebrafish assays.”
MacRae and his team “are going into greater depth and trying to dissect in greater detail the events leading to QT prolongation,” Stephane Berghmans, director of drug discovery for Znomics, told CBA News this week. He added that zebrafish have a lot to contribute to safety pharmacology, “and MacRae’s assay demonstrates that.”
The next step would be proving that “our assay is predictive in a blinded fashion, and that is what we are actively doing at the moment as part of the NIH-funded work,” MacRae said. Another goal is to identify a toxicity that could not be identified in any other way.
“If we did that, it is conceivable that this assay could be part of the screening of new molecular entities during the preclinical phase, before they get further down the drug development pipeline,” he added.
Ultimately, the real goal would be the ability to do toxicity screening in parallel with drug discovery. “That would allow you to match the risk of the disease and the risk of toxicity with magnitude of the therapeutic effect in real time,” said MacRae. That is probably the most optimized use of the assays, and MacRae said that it would take a change in the way the zebrafish is viewed as well as robust data linking the zebrafish phenotypes and responses with human phenotypes and responses.
“And I think that we need to work pretty hard to be able to do that,” he said.