For any graduate student who has ever struggled to collect enough darn data to finally start writing the darn thesis, it adds insult to injury. But Robert Hughes' group has the opposite dilemma.
"One of the problems with high-throughput data," he told BioWorld Today, "is that there's so much data that you have to come up with creative ways to eliminate it."
In the May 2007 issue of PLoS Genetics, senior author Hughes and his colleagues reported the end result of their screening-plus-winnowing approach: 27 proteins, out of an initial pool of more than 200 identified by the group, that interact with mutant huntingtin protein, the culprit at the root of Huntington's disease, and affect the severity of Huntington's symptoms in an animal model.
Hughes, now an assistant professor at the Buck Institute for Aging Research in Novato, Calif., and colleagues at Salt Lake City-based biotechnology firm Prolexys Pharmaceuticals Inc., the Baylor College of Medicine in Houston and the Fred Hutchinson Cancer Research Center in Seattle came by their embarrassment of riches using two separate screening methods to cast a wide net, followed by applying elimination criteria to select a group of proteins for testing in the fruit fly.
The researchers used yeast-two-hybrid screening and a combination of pull down and mass spectroscopy, respectively, to identify 234 proteins that bind to mutant huntingtin. Roughly a quarter of those proteins were chosen for further in vivo study in fruit flies.
"Many of the proteins that interact with Huntington's are modulators of its toxicity," explained Juan Botas of Baylor, whose lab carried out the fruit fly studies. The purpose of the validation was to identify which of the 60 proteins actually affected toxicity; roughly half of them changed the toxicity of mutant huntingtin's when their levels were manipulated in flies.
Hughes noted that the more labor-intensive and low-throughput pull-down method is sometimes seen as more likely to find biologically relevant proteins, because it can detect complexes of more than two proteins interacting with each other. However, in the data published in PLoS, the two approaches had similar validation rates in vivo. "We see the approaches as complementary," Hughes said. "Both produced high-quality data."
The findings allow new insights into the basic biology of Huntington's, as well as providing potential drug targets.
On the basic science side, Hughes and his group identified a number of proteins in the so-called SNARE protein family, which are involved in neurotransmitter release. Though it has long been clear that transmitter release is affected in Huntington's, it has been less clear whether that is an important cause of problems, or just one of many consequences of the toxicity of mutant huntingtin.
Hughes said that though this finding is not proof, it does allow the inference that "the defect in neurotransmission is a very primary event" in Huntington's, not just a consequence of general neuronal damage.
From a more practical standpoint, 17 of the validated proteins were so-called loss-of-function suppressors, meaning that knocking them out improved the Huntington flies' phenotype. Hughes said that a loss-of-function suppressor suggests that pharmacological inhibition of the pathway in question has the potential to improve symptoms of Huntington's disease. Since inhibition often is easier to achieve than activation pharmacologically, the authors wrote that the pathways in question "can be considered potential therapeutic targets."
Those targets, though, will have to be developed by someone else. Prolexys, where much of the work was done (and where Hughes used to be director of therapeutic biology), is itself focused on the development of cancer therapeutics. The Huntington's work "was done in the early days of the company when we were setting up the technology," Prolexys Chief Scientific Officer Sudhir Sahasrabudhe told BioWorld Today. But "we certainly do proteomics-based drug discovery." Prolexys has used the same high-throughput proteomics methodology to identify targets in the beta-catenin pathway.
The Huntington's work, though, was funded partly by grants from the Cure Huntington's Disease Initiative, the Hereditary Disease Foundation and the HighQ Foundation, and Sahasrabudhe said that one of the conditions of funding had been that the results be put in the public domain.