November 1, 2008 (Vol. 28, No. 19)
Vicki Glaser Writer GEN
Numerous Targets for Small Molecule Drug Development Are Being Explored
Therapeutic drug discovery targeting Alzheimer’s disease (AD) is a wide open field, as this progressive neurodegenerative disorder presents various structural and biochemical abnormalities that could serve as drug targets.
The etiologic origin for the amyloid-beta plaques and neurofibrillary tangles characteristic of AD remains unclear, and at present, therapeutic interventions aimed at slowing disease progression and the resulting cognitive impairment and dementia have been unremarkable.
Numerous targets for small molecule drug development are being explored, and the range of hypotheses and signaling pathways of interest were evident at the recent “International Conference on Alzheimer’s Disease Drug Discovery,” sponsored by The Alzheimer’s Drug Discovery Foundation (ADDF) in New York City.
Many of these emerging therapeutic strategies are originating in academic laboratories as a result of basic research on how the biological and biochemical properties of neuronal development and physiology go astray in the early stages of neurodegenerative disease.
These novel drug targets represent attractive investments for the biotech and pharma industries, as there is a large unmet medical need in this disease area, and the growing population of older adults will contribute to an increasing disease burden.
Recent news has included findings from the University of Pennsylvania School of Medicine that detailed how a small molecule—4,5-dianilinophthalimide (DAPH)—can selectively dismantle abnormally folded protein fibers including the amyloid fibers associated with neurodegenerative disease.
DAPH and Abeta 42 Toxicity
Building on studies initiated at MIT, researchers have shown that DAPH blocks the growth of these fibers by wedging itself into the crevices between fiber subunits. By remodeling fiber architecture, DAPH can reduce amyloid-beta (Abeta) toxicity.
Columbia University Medical Center recently published findings from a 20-year study of elderly individuals who were either at risk of developing or already had AD. The results demonstrated an association between elevated plasma levels of Abeta 42 peptide and increased risk for AD, while the blood levels of Abeta 42 declined in people with established AD.
The level of Abeta 42 appears to increase before the onset of disease and to decrease shortly after the onset of dementia—perhaps because the peptide becomes trapped in the developing Abeta plaques in the brain. This peptide may, therefore, have a role as a marker of early disease, allowing emerging therapeutic drugs to be administered before neurodegenerative changes cause significant damage.
The September issue of NeuroInvestment featured a review of Alzheimer’s therapeutics that included more than 170 programs at 110 companies. In addition to established players in the field, the review highlighted several young companies that have active discovery programs for AD. These included Allon Therapeutics, TauRx, Accera, EnVivo Pharmaceuticals, NeuroNascent, and Sanomune.
Repair and Restoration
Allon Therapeutics, one of the companies presenting at the ADDF-organized conference, announced results in August, of a pharmacokinetic study showing that its drug, AL-108, now entering a Phase IIb trial in patients with AD, is able to penetrate the blood brain barrier in quantities sufficient to have a therapeutic effect.
AL-108 is derived from activity-dependent neurotrophic factor, a naturally occurring protein secreted by the brain in response to injury. It interacts with neuronal tubulin to repair the damage to microtubular networks caused by neurodegenerative disease. This may help re-establish the structural integrity of the neurons, aid in restoring axonal transport within neurons, and enhance chemical transmission between nerve cells. AL-108 also promotes neurite growth.
Earlier this year, Allon released results of a Phase IIa trial in patients with amnestic mild cognitive impairment, which is a precursor to AD. In addition to being safe and well tolerated, the drug was associated with statistically significant, dose-dependent, and durable improvement in memory.
A different approach to AD therapeutics development began in the laboratory of Anthony Koleske, Ph.D., associate professor of molecular biophysics and biochemistry at Yale University. Dr. Koleske’s group studies how changes in the extracellular environment cause cells to alter their shape and the way they move, and how these responses are disrupted in cancer and neurodegenerative disease. His group has identified a biochemical pathway sensitive to changes in a cell’s adhesive environment that inhibits the Rho GTPase. Hyperactivity of Rho has a negative effect on neurons.
Dr. Koleske has shown that knocking out this pathway in mice causes dendrites to become unstable and to disappear. His work has focused on two proteins in this pathway, the Abl and Arg tyrosine kinases, which can interact with both Abeta and tau.
One aspect of this group’s research involves determining whether Rho signaling or some other pathway component has a role in the development of AD pathology. “There is some speculation that extracellular Abeta deposits may be interacting with adhesion receptors or extracellular deposits,” said Dr. Koleske, but that has not yet been tested.
“We believe that Rho signaling is hyper-activated in neurodegenerative conditions that lead to synapse loss and dendritic regression,” he added. Dr. Koleske is currently studying the effects of a small molecule that inhibits Rho kinase 2 (Rock2), which is an approved drug in Japan.
In September, Pfizer acquired the rights to Dimebon, an experimental drug for AD developed by Medivation. The deal, which includes collaboration on ongoing Phase III trials in AD and Huntington’s disease, was valued at up to $725 million. Dimebon is an orally administered small molecule that appears to improve mitochondrial function and stimulate neurite growth.
Drug discovery efforts in AD at P2D Biosciences focus on a disease-modifying therapeutic approach that targets the neuroinflammatory response to elevated levels of cytokines such as TNF-a and IL-1, believed to underlie neuronal loss in Alzheimer’s disease. S. Prasad Gabbita, Ph.D., vp R&D at P2D Biosciences, gave a talk entitled, “Small Molecular TNF-Alpha Inhibitors for Alzheimer’s Disease.”
Postmortem evidence has shown that TNF-a levels are elevated in the brain, cerebrospinal fluid, and serum of patients with AD. Preclinical study results support a role for TNF-a in neurodegeneration.
Dr. Gabbita pointed to findings from three separate studies completed during the past 12 months: one showed TNF-a to be a critical signaling mediator in the development of synaptic deficit and cognitive impairment in an Abeta mouse model; another study demonstrated improved function of brain vasculature resulting in decreased hippocampal neuronal loss following administration of thalidomide, a TNF-a inhibitor.
The third study, in an Abeta-induced impairment model, produced evidence that thalidomide could enhance learning ability and memory in mice.
“We are developing compounds that are up to 70-fold stronger inhibitors than thalidomide,” explained Dr. Gabbita. “We have identified four lead compounds and are now focused on modifying these small molecules to improve their ability to cross the blood brain barrier and to optimize their toxicity profile.”
The company plans to test these compounds in two AD models, one created by direct delivery of Abeta into the mouse brain and the other a transgenic mouse model.
“When the first signs of probable AD appear, a significant number of neurons in the temporal cortex and hippocampal regions of the brain have already died, and this is irreversible,” stated Ajay Gupta, Ph.D., chairman and CEO of Osta Biotechnologies. Osta is developing a novel class of drugs based on the inhibition heme-oxygenase-1 (HO-1) protein, a stress protein that catabolize heme to biliverdin, free iron, and carbon monoxide. The HO-1 gene is “exquisitely sensitive” to oxidative stress and is induced in brain and other tissues in disease and trauma, according to Dr. Gupta.
Hyman Schipper, M.D., Ph.D., professor of neurology and medicine at McGill University has demonstrated that HO-1 protein is significantly overexpressed in AD-affected temporal cortex and hippocampus compared to control brain tissue. Furthermore, HO-1 upregulation by transient transfection of the human HO-1 gene or by stimulation of endogenous HO-1 expression promotes intracellular oxidative stress, increased mitochondrial permeability, and iron deposition in the mitochondria.
Dr. Schipper’s work has also shown that gial iron sequestration makes cocultured neurons increasingly sensitive to oxidative injury.
Dr. Schipper and Dr. Gupta believe that the induction of the astroglial HO-1 gene may constitute a “common pathway leading to pathological brain iron deposition, intracellular oxidative damage, and bioenergetic failure in AD and other human CNS disorders.” Targeting HO-1, however, could provide an early therapeutic intervention for the treatment of AD.
In a talk entitled, “Suppression of Glial HO-1 Activity as a Potential Neurotherapeutic Intervention in AD,” Dr. Schipper described the efforts under way at Osta to develop a potent, brain-permeable small molecule inhibitor of HO-1 that does not affect HO-2, a noninducible form of HO involved in cellular housekeeping activities and is important in neuroprotection.
Osta has demonstrated that overexpression of HO-1 in cultured astrocytes can damage nearby neurons, resulting in cell death. Osta has also shown that its small molecule HO-1 inhibitors are selective and potent and can attenuate oxidative damage to the glia over expressing HO-1.
At therapeutic doses of up to 100 mg/kg/day, the company’s lead HO-1 inhibitor did not cause any significant toxicity in mice. Osta, in collaboration with Walter Szarek, Ph.D., at Queen’s University (Kingston, Ontario), has designed and synthesized about 200 analogs of these HO-1 inhibitors and is now studying their pharmacokinetic and pharmacodynamic properties.
Targeting Abeta and Tau
Luciano D’Adamio, M.D., Ph.D., professor at Albert Einstein College of Medicine and cofounder and acting CSO of RemeGenix, described how the company’s early work to identify endogenous proteins that have a role in regulating Abeta processing led to its focus on the BRI2 protein.
BRI2 is a membrane protein that binds Abeta precursor protein (APP). It is mutated in other forms of AD-like familial dementia and, if overexpressed in tumor cell lines inhibits Abeta formation. It tightly binds APP in the region where the secretase anzymes act to cleave the precursor protein into its constituent parts.
Dr. D’Adamio’s group has demonstrated in both cell lines and mice that over-expression of BRI2 can block access to the APP cleavage site and prevent APP processing. Conversely, knocking out BRI2 expression increases APP processing. In a mouse model of AD, overexpression of BRI2 leads to a reduction in plaque formation.
The researchers have shown that BRI2 does not directly affect the activity of the secretase enzymes and acts only on APP to block secretase binding. This observation has fueled hope that molecules capable of mimicking BRI2 activity would not interfere with other secretase functions. RemeGenix has synthesized a small peptide capable of blocking the APP cleavage site without inhibiting secretase. This peptide blocks the docking of both ß and g secretase.
RemeGenix has produced data showing that the peptide binds a region of APP that affects the toxicity and polymerization of Abeta. Ongoing efforts are proceeding in three distinct areas: studies to determine the structure of the peptide bound to APP as a basis for developing improved small peptides and designing a small molecule therapeutic; characterization of the chemistry of the peptide and modifications aimed at improving its bioavailability, stability, and other pharmacokinetic properties; and behavioral studies in animal models to assess therapeutic efficacy.
Research on the role of molecular chaperones in cancer led to the development of heat shock protein 90 (Hsp90) inhibitors as potential therapeutic agents for AD.
Gabriela Chiosis, Ph.D., Frederick R. Adler chair, molecular pharmacology and chemistry, at Memorial Sloan-Kettering Cancer Center, and Paul Greengard, Ph.D., Vincent Astor professor, laboratory of molecular and cellular neuroscience, at The Rockefeller University, recently presented evidence that a brain-permeable Hsp90 inhibitor blocks the production of mutant tau and reduces the levels of phosphorylated and aggregated tau in a mouse model of dementia.
This approach arose from the recognition that cancer and AD are both heterogeneous diseases that result from the accumulation of aberrant molecular events, and both cancer cells and altered neurons appear to co-opt molecular chaperones such as Hsp90 to stabilize mutant proteins, enabling them to continue to function and be retained by the cell.
A synthetic small molecule inhibitor of Hsp90 developed by Dr. Chiosis and colleagues is currently in clinical testing in patients with cancer, with a second generation compound in the late-IND stage of development. Dr. Chiosis believes that Hsp90 is important in the early stages of AD pathogenesis, and that affected neurons may become increasingly dependent on Hsp90 as the disease progresses.
Challenges of CNS Disorders
One of the challenges with CNS disorders is designing inhibitors capable of crossing the blood brain barrier. “We have compounds with good brain permeability, and we are looking at time-and dose-dependent changes in pharmacodynamic markers in animals,” Dr. Chiosis says.
She notes that Hsp90 also has housekeeping functions essential for normal cell activity. The small molecule inhibitors that her group is developing, are designed to have higher affinity for the form of Hsp90 that binds aberrant proteins rather than the form associated with housekeeping functions.
What differentiates the various forms is not well understood yet but may depend on post-translational modifications to the Hsp90 protein or the types of co-chaperone molecules which the protein associates.