YouTube Facebook LinkedIn Google+ Twitter Xingrss  

Fly Fishing on the Brain

By John Brokars

June 12, 2002 | One of the largest concerns in today's drug discovery market is the increasing time and expense associated with the R&D that precedes FDA approval of a new drug. After a pharmaceutical company has spent millions on research, many promising compounds are brought to pre-clinical and clinical trials, only to be found ineffective or worse — toxic and dangerous to humans.

EnVivo Pharmaceuticals Inc., a Redwood City, Calif.-based biotech startup, has developed an innovative approach to this problem that blends high-tech automation with fruit

Neurodegeneration in a Nutshell 
Neurodegenerative and psychiatric disorders account for more hospitalizations and long-term care than nearly all other disorders combined.

Read More 
fly genetics in the hopes of streamlining several steps in drug discovery: molecule identification, target binding, and efficacy and safety assessment. Principals at EnVivo (, which was founded in 2001 by researchers from Baylor College of Medicine and the European Molecular Biology Laboratory (EMBL) in Heidelberg, Germany, say its technology has the potential to shave as much as three years from the front end of the average 12-year drug discovery and development process.

Working with one of the most reliable genetic models, Drosophila melanogaster, EnVivo researchers — with the help of proprietary Baylor technology — insert known human brain disease genes into the fruit fly. This living model faithfully mimics many human neurodegenerative diseases, including Parkinson's, Huntington's, and Alzheimer's diseases. Another California-based company, Exelixis Inc., already uses the Drosophila model of central nervous system (CNS) disease, as well as genetic disease models in worms, zebrafish, and mice, to identify drug targets.

Using Drosophila allows researchers to test novel compounds directly in a living system, and can reveal not only if a molecule fits a target, but also how well a compound permeates the brain, the extent to which it relieves the illness, and whether it causes toxicity. What's new is EnVivo's use of proprietary monitoring technology and data analysis software from EMBL to conduct high-speed, high-throughput screening on living organisms.

For EnVivo, the payoff could be tremendous. CNS drugs represent $50 billion of the $400 billion spent on central nervous system disorders annually, and the market is growing 20 percent each year, according to the company.

The Fruitful Drosophila 
Proof that the fruit fly is indeed a useful model for human CNS diseases has accumulated in the last few years. In March 2000, for example, a report in Nature by Mel B. Feany and colleagues demonstrated that single-gene mutation models of human neurodegenerative diseases such as Parkinson's could be studied using fruit flies. Since then, complete genome maps of both Drosophila and humans have revealed significant interrelatedness and conservation between fruit fly and human genes and neural pathways. The most dramatic example came a few years ago, when Walter Gehring, a professor at the University of Basel in Switzerland, demonstrated that the human PAX6 gene was perfectly capable of triggering eye development in Drosophila lacking the fly counterpart. Huda Zoghbi, one of EnVivo's scientific founders and a Howard Hughes Medical Institute investigator at Baylor, adds that a fly gene involved in hearing and its "very similar" analog in humans perform identical functions.

The fly model is certainly not perfect. Many of the cellular pathways in the human CNS are not present in the fly, and obviously, humans and flies do not share many essential anatomical traits. This may be why a fly model of Lou Gehrig's disease (amyotrophic lateral sclerosis, or ALS) has not been achieved.

Even so, flies carrying a single-gene mutation responsible for a rare form of Parkinson's disease do show the age-dependent, progressive degeneration of certain dopamine neurons, accumulation of plaques rich in the protein alpha synuclein, rigidity, and tremors characteristic of Parkinson's-like disorders (see "Neurodegeneration in a Nutshell" below). Flies that mimic Alzheimer's show similar age-dependent, progressive neurodegeneration, as well as the specific neuronal loss, beta amyloid plaque accumulation, and premature death commonly associated with this disease.

Lewis Sudarsky, director of the movement disorders division of neurology at Brigham and Women's Hospital and Harvard Medical School, calls these transgenic models of Parkinson's a "terrific vehicle" and a "powerful tool." He says that Drosophila in particular is a "great accelerator for understanding" because its lifespan is only a few weeks, as opposed to the much longer generation times of humans and mice. Drosophila is also attractive for its low cost, smaller size, and CNS complexity.

One major stumbling block, says Sudarsky, is that "Parkinson's in humans may turn out to be polygenetic and multifactorial." In fact, single-gene models are unlikely to provide comprehensive insight into the pathogenesis of such common diseases. This would challenge companies, such as EnVivo, seeking to reduce the discovery time for new pharmaceuticals aimed at CNS disorders. Even if that's true, however, Zoghbi says single-gene models can still be useful in finding therapeutics, because accumulation of alpha synuclein plaques is common to most Parkinson's cases and therefore provides a common focus in the effort to identify means of relief.

Despite the uncertainty regarding Parkinson's disease's genetic underpinning, Zach Hall, EnVivo's president and CEO, says that his company has the technological potential to meet the market need for many CNS therapeutics. Hall is the former director of the National Institute of Neurological Diseases and Stroke (NINDS) and, before joining EnVivo, was most recently executive vice chancellor at the University of California at San Francisco.

Flies on Film 
Much of EnVivo's work depends on genetic intellectual property regarding human CNS disease that has been licensed from Baylor. Indeed, Juan Botas and Zoghbi, two Baylor pioneers of genetic in vivo drug discovery for CNS diseases, have played pivotal roles in EnVivo's research.

However, it's EnVivo's processing technology that allows hundreds of thousands of potentially therapeutic novel compounds to be tested quickly in a so-called "living test tube." The general process works like this: Researchers maintain an inventory of normal flies and transgenic flies (containing inserted brain disease genes). Unlike the normal fly's lifespan of roughly 70 days, diseased flies typically become sick after 10 days, with deteriorating motor performance until death occurs after about 20 days. Potentially therapeutic molecules are introduced in the flies' food, which is available to them ad libitum.

High-speed cameras and proprietary software monitor fruit flies to detect the restoration of normal function. The actual "bioassay" is per-formed once daily and takes only 10 to 20 seconds. (EnVivo is developing a conveyor belt system to streamline the process even further.) Software integrates the information from the 50 to 100 individual frames captured per assay by each camera and then performs pattern recognition analysis to uncover any trends.

Cayetano González, an EnVivo founder, developed much of the high-speed screening technology at EMBL, where he is a group leader. González' EMBL colleague, Louis Serrano, coordinator of the EMBL Structure and Biocomputing program, helped develop EnVivo's proprietary software, known as PhenoScreen. This software can currently track 12 to 15 flies in each test tube and analyze more than 20 variables reflecting motor activity (speed, frequency of turning, direction of movement, etc.) to determine a disease-specific electronic "signature." It also monitors the phenotype of the diseased flies for improvements coincident with the introduction of new molecules.

Once the system has identified what EnVivo calls "bioactive lead substances" that both reach the disease target and successfully reverse the disease symptoms in a living animal, then EnVivo can go back and find the targets reached. Promising targets are then tested in vertebrates.

The EnVivo system is run using a simple 1Ghz Pentium PC, and analysis is performed on a dual-Pentium server. As the Oracle database grows, EnVivo says it is considering migrating to a Sun Microsystems workstation. The company is building a comprehensive IT infrastructure that will include a laboratory information management system to track compounds, flies, and all processes involving PhenoScreen assays.

Cutting Clinical Trial Length 
Validating new molecules quickly could shorten the process of getting a molecule to clinical trial from an average of 50 months to somewhere between 12 and 18 months, according to EnVivo, which would not only reduce costs but also expedite relief for patients.

EnVivo asserts that it has further verified its process by testing several FDA-approved Parkinson's drugs on the flies, resulting in documented improvements. This development has investors interested, and EnVivo expects collaboration funding from larger pharmaceutical companies within the year. EnVivo officials would not say which approved CNS drugs had been tested.

Still, significant hurdles remain. For example, a molecule's ability to mollify neural symptoms depends largely on the dose in which it is given. At high-throughput rates, with hundreds of thousands of compounds to test, determining the correct dose could prove especially difficult. For example, given at too high a dose, a compound could be seen as toxic, whereas at too low a dose, it could appear ineffective; either way, the result could be that a potentially useful compound is discarded prematurely.

This dosing problem is not unique to EnVivo, however, and Dan O'Connell, general partner of venture capital firm Neuroventures Partners — an EnVivo backer — argues that "the underappreciated aspect of this technology is its ability to better qualify lead candidates as being both bio-available and not lethally toxic." EnVivo has the capability to show a therapeutic index of which doses of a lead compound show efficacy and which show toxicity. Knowing the capability for toxicity issues in a molecule before clinical trials can be particularly helpful.

"Some validation needs to occur, but we are investigating some collaborations with [major] pharmaceutical companies to share resources," says O'Connell. Adds Thorsten Melcher, EnVivo's chief scientific officer, "The [big pharmas] have the infrastructure and resources to run clinical trials, and a partnership can help us bring a drug to market."

Only time will tell if EnVivo's attempt to turn the classic fruit fly into a modern film star will produce a commercial hit.

John Brokars is a writer based in Chestnut Hill, Mass. He can be reached at 617-630-0124. 


For reprints and/or copyright permission, please contact  Jay Mulhern, (781) 972-1359,