Sean Scott, the late president of the ALS Therapy Development Institute (ALS TDI) in Cambridge, Mass., used to compare ALS to a building fire in which firemen show up focused on one thing: putting out the fire as quickly as possible.
Scott believed that same sense of urgency and focus should be directed at stopping ALS, recalled ALS TDI Director of Communications Rob Goldstein during an hour-long public webinar on June 29.
To further the analogy, just as firemen need pickaxes and fire hoses, so ALS researchers need a set of tools to stop ALS; one of the most important of these is an animal model of the disease that can be used in research.
During the webinar, ALS TDI’s director of in vivo validation, Fernando Vieira, informed viewers of the current state of animal models in ALS research.
About animal models
Animal models have long been used in testing new drugs, enabling researchers to evaluate how the body's cells and tissues interact with a drug, or to determine whether a drug has an effect against one or more processes that cause disease, Vieira noted.
Researchers also study animal models to better understand how a disease starts and progresses. Some of the more common animal models include invertebrates such as worms and flies; rodents including mice, rats, Guinea pigs and rabbits; large animals such as dogs, horses, pigs and birds; and, finally, nonhuman primates such as monkeys and chimps.
Vieira said he doesn't believe there's a "standard" animal model for ALS. Researchers typically select a model based on cost and its applicability to the problem they're trying to solve, he explained, assessing such trade-offs as size and handling requirements, risk and resemblance to human physiology. All models can be useful depending on the application, Vieira noted.
Animal models of ALS
In 1994, researchers associated the SOD1 gene with some forms of familial (inherited) ALS. When that gene is flawed it can lead to production of mutated SOD1 protein and — following a cascade of events not yet fully understood — motor neurons die.
A number of ALS animal models are based around the SOD1 gene, including some zebra fish, fruit flies and worms.
Vieira explained that when scientists put flawed SOD1 genes into these animals, they look for development of a disease that resembles ALS in humans. Although they often see neuronal dysfunction typical of ALS — fruit flies exhibit lethargy and zebrafish lack appropriate axonal development, for example — none of these models exhibit the same complexity that exists in a mammalian system, with lots of cell-to-cell interactions between support cells, motor neurons and the muscle itself.
Typically, said Vieira, "you're modeling parts, but not all, of ALS." Still, such models have great value.
For example, Vieira noted, to test a potential therapeutic designed to affect axonal growth, a researcher may use a fish model if he only wants to look at axons. "But if you want your drug to impact immune cells and their effect on axons, you can't use a fish. You'd have to look at something like a mouse."
Mice have complex systems with many similarities to humans, including an immune system, vascular flow and neurons that are surrounded by immune-specific support cells called astrocytes and microglia.
But, so long as rigorous standards are applied, "you can find the right place and the right use for a fly model, or a worm model, or even different types of mouse models," he said.
Studying ALS in the SOD1 mouse
ALS TDI uses a particular variant of the SOD1 mouse model known as SOD1-G93A, which contains 23 to 25 copies of a flawed human SOD1 gene. This mouse exhibits a disease that looks like an aggressive form of ALS, dying at 135 days of age (as opposed to the normal 2 years) following muscle weakening and atrophy, and motor neuron loss.
A number of symptoms exhibited in the SOD1 mouse model of ALS are evident in human disease as well. These include functional muscle loss, respiratory distress, upper and lower motor neuron involvement, an immune-system component, blood-brain barrier disruption, inclusion bodies and protein aggregation (problems found in cells), and neuromuscular junction disruption.
Although the SOD1 mouse provides a huge advantage over other models, the fact remains that it's not a human and important differences remain. Still, "what we're waiting for, and what we want and are looking forward to," Vieira said, "is a drug that works in the SOD1 mouse model at ALS TDI, that also works in humans."
Is the SOD1 mouse relevant to sporadic ALS?
The SOD1 mouse has the familial, or inherited, form of ALS, which accounts for approximately 5 percent to 10 percent of ALS cases in humans. Although about 90 percent to 95 percent of human ALS cases are sporadic (non-inherited), “it's important to remember that familial and sporadic ALS look clinically identical," Vieira said.
A physician can't tell the difference between familial and sporadic ALS without a family history and a genetic test. Both forms present with the same physical and biological symptoms.
"In the end," Vieira said, "the SOD1 mouse model is a model in which motor neurons die. And ALS is a disease where motor neurons die. The model is also one where muscles waste away and where there's central nervous system inflammation – again, all of these are things that we see in both familial and sporadic ALS."
The hope then is to find a drug that targets these downstream effects in both the mouse model and some, if not all, ALS patients with either familial or sporadic ALS.
In mouse model, details matter
In 2007, ALS TDI research scientists conducted testing called "whole genome expression profiling" in the SOD1 mouse. They looked at all 30,000 genes in the mouse genome and studied how those genes are affected in SOD1 mice compared to healthy mice. They then compared treated SOD1 mice with untreated SOD1 mice.
Starting with approximately 8,000 mice in 2007, the Institute has built its SOD1 mouse model database to 42,000 mice, and in the process learned that "noise" in the model can confound study results. In other words, details matter.
Failure to control the system carefully can lead to false positive or false negative results. This means, for example, that test and control groups should contain equal numbers of male and female mice, as well as equal numbers from the same litter.
ALS TDI used its optimized study design to test a number of drugs that had showed positive results in mouse studies in other labs, but which failed to show benefit in human clinical trials. ALS TDI was unable to reproduce or confirm the previous positive results in mice, and in fact, the results mirrored those seen in human testing.
Probably the most well-known example of this discrepancy is minocycline. Several studies had shown that minocycline was efficacious against ALS in the mouse model, which led to a large clinical trial in humans. But in the human trial, not only did the drug fail to have beneficial effects, in some cases it even hastened the course of the disease. In later testing using its own rigorous standards (balancing for gender, littermates and other considerations), ALS TDI found minocycline had either no effect or a negative effect in mice.
“We've used this type of testing to really understand what's happening in ALS in these mice, and then over time build a database that helps us home in on the best possible treatment strategies,” Vieira said.
TDP43: New mouse on the block
Mutant TDP43 protein produced from instructions carried by flawed TDP43 genes has recently been associated with both sporadic and some familial forms of ALS. Now, several models of a new mouse engineered to have a flawed TDP43 gene have been developed (see New ALS Mouse Added to Research Toolkit).
ALS TDI currently is working to build a colony of the new TDP43 mice, which TDI researchers plan to fully characterize in the same way they did the SOD1 mouse, to determine how best to use it in studies.
TDI researchers plan to do the same gene expression profiling that they did with the SOD1 mouse. They'll compare the 30,000 genes of the TDP43 mouse to those of the SOD1 mouse and to healthy mice, seeking to determine what the mice have in common and what traits are specific to each individual strain.
Vieira said the institute is "anxiously awaiting news" from research groups working on mouse models of FUS, another gene that recently was implicated in both familial and sporadic ALS.
"As these models come out, and as we evaluate their relevance to ALS, we'll be bringing them in-house," Vieira said. "We want to use every tool we possibly can as long as our resources allow."
Coming up from ALS TDI
ALS TDI's third-quarter 2010 Research Update Webcast and 6th Annual Leadership Summit Webcast is scheduled for 8 a.m. Eastern Time, Oct. 5.
The Institute has scheduled introductory webinars for July 27, Aug. 17, Sept. 21, Oct. 19, Nov. 16 and Dec. 21 (times vary). These contain similar basic information about ALS and allow participants to ask questions about research progress and new trends in therapy development. Registration is required.
To learn more about ALS TDI’s archived and upcoming webcasts, webinars and conference calls, visit the Institute's website at www.als.net and click on "Get Involved."
(MDA and the nonprofit biotech ALS Therapy Development Institute (ALS TDI) of Cambridge, Mass., forged a historic partnership in January 2007 when they launched the largest ALS drug discovery project to date, a three-year, $36 million collaboration to identify biochemical targets in ALS and find drugs that hit them. In January 2010 MDA renewed its partnership with ALS TDI with a grant of $2.5 million (see Muscular Dystrophy Association Renews Partnership).)