This is Part 2 of a two-part series. See also Part 1.
11 December 2008. While some neurodegenerative disease therapies focus on signature proteins—for example, amyloid and tau in the case of Alzheimer’s—a growing number are addressing more fundamentally what sets neurons up for self-destruction. Some of these approaches target histone deacetylases (HDACs), chromatin-compressing enzymes that promote transcriptional repression and have recently come into the spotlight as regulators of DNA repair, too. In the scenario emerging from two new studies, certain deacetylases keep neurons healthy by silencing a diverse set of genes, but will ignore these loci and move to DNA breaks to rally repair crews in times of stress (see ARF companion story). Extra deacetylase would thus help ensure sufficient resources for both essential roles. But one could also imagine the opposite situation, in which having more of a particular deacetylase might be a bad thing if, for instance, that HDAC repressed a gene whose upregulation helps treat or prevent a disease. In those cases, HDAC inhibitors could turn up winners. As reported in recent literature and at the Society for Neuroscience (SfN) annual meeting held 15-19 November in Washington, DC, several such compounds have passed muster in preclinical testing, and a few are moving toward human trials. At first glance, these findings seem to fly in the face of recent work suggesting protective roles for select deacetylases (see Part 1 of this two-part story). Considering the collective data, experts agree it is impossible to label HDACs as generally “good” or “bad” for aging and neurodegeneration, and highlight the importance of identifying their target genes and evaluating HDAC-targeting compounds for specific conditions.
In an SfN poster featuring data to be published in the journal Neuropsychopharmacology, researchers led by Alberto Pérez-Mediavilla and Ana Garcia-Osta of University of Navarra in Pamplona, Spain, showed that the HDAC inhibitor sodium 4-phenylbutyrate rescues cognitive deficits in Tg2576 mice, a widely used AD model that overexpresses mutant human amyloid precursor protein (APP) and develops memory impairment by nine to 10 months. Phenylbutyrate is an orally bioavailable short-chain fatty acid. By inhibiting deacetylases, it activates genes involved in regulating cell development and proliferation. It also appears to act as a molecular chaperone that reverses pathological protein aggregation in human diseases. A related compound, the HDAC inhibitor sodium butyrate, improved learning and long-term memory recovery in inducible p25 transgenic mice, which exhibit tau pathology, cognitive defects, and neurodegeneration (Fischer et al., 2007 and see ARF related news story).
Poster lead author Garcia-Osta and colleagues gave 16-week-old Tg2576 or non-transgenic mice intraperitoneal injections of 4-phenylbutyrate (200 mg/kg) or saline once a day for five weeks. The drug restored spatial memory in Tg2576 animals to wild-type levels in the Morris water maze with invisible platform and in 15- and 60-second probe trials. Treated Tg2576 mice showed no change in Aβ40 or Aβ42 levels or plaque load, but had reduced tau phosphorylation (with steady tau protein levels) and GSK3β activity, compared to wild-type and saline-injected controls. The effects on tau but not Aβ are intriguing, as a study published last month showed that an inhibitor of sirtuins (Class 3 HDACs) had similar tau-selective effects and improved cognition in a different AD mouse model (3xTg) (see ARF related news story). In the SfN study, phenylbutyrate treatment also enhanced expression of synaptic plasticity markers GluR1 and PSD95 in Tg2576 mice.
In experiments to tease out the compound’s specificity, the researchers found that cortical brain extracts from Tg2576 mice have reduced levels of acetylated histone 4 (H4) relative to non-transgenic animals, and that drug treatment partially restores this effect (but does nothing to levels of acetylated H3). In cultured primary neurons, drops in H3 and H4 acetylation in Tg2576 relative to wild-type cells were both reversed by four days of drug treatment. These data leave open the possibility that 4-phenylbutyrate may affect acetylase activity as well, Pérez-Mediavilla noted in an e-mail to ARF. He added that it is unclear at this point which HDACs are targeted by the drug.
The SfN meeting also featured preclinical data on an HDAC inhibitor that appears to enhance memory in normal mice. Developed by EnVivo Pharmaceuticals, a small biotech in Watertown, Massachusetts, the compound EVP-0334 can be taken orally and penetrates the brain well. As detailed on posters by Holger Patzke and Liza Leventhal, of EnVivo, and colleagues, EVP-0334 primarily targets Class 1, 2A, and 4 HDACs, inhibiting deacetylase activity in mouse cortical neurons and human astrocytes with IC50 values ranging from 0.3-1microM. In brain extracts from mice given the drug orally, histones 2A, 3, and 4 show increased acetylation with a minimal effective dose of 10 mg/kg. Wild-type mice receiving a similar dose had better short-term (90-minute) and long-term (24-hour) memory than did vehicle-treated controls in a novel object recognition test. “This is striking to us because the drug is completely gone from the mouse at 24 hours,” EnVivo scientist Michael Ahlijanian said at an SfN press briefing.
In a post-briefing conversation, he told reporters the compound has not been tested in any AD mouse models. He also could not say much about transcriptional profiles affected by the drug. “We can’t point to any single gene or group of genes that is responsible for these memory-enhancing effects,” Ahlijanian said. Nevertheless, the drug is moving toward the clinical pipeline. A Phase 1 trial focusing on pharmacokinetics and safety should start within a year, he told ARF.
HDAC inhibitors may also hold promise for treating Huntington disease, for which transcriptional dysregulation is emerging as a key pathology. Reporting in the 7 October issue of PNAS, Elizabeth Thomas and colleagues at the Scripps Research Institute in La Jolla, California, showed that oral administration of an HDAC3-prefering inhibitor improved motor performance, overall appearance, and body weight in an HD mouse model (R6/2300Q). The drug slowed brain atrophy and partially relieved gene expression changes caused by accumulation of mutant huntingtin protein in the striatum, cortex, and cerebellum. Repligen Corporation in Waltham, Massachusetts, is licensing this and another related compound for Friedreich’s ataxia, an inherited neurodegenerative condition that affects one in every 50,000 people in the U.S. The company “is also committed to pushing our compounds into trials for HD,” Thomas wrote in an e-mail to ARF. “The most optimistic time frame would be one year from now.”
Not to be left out, Parkinson disease may have some HDAC-based drug candidates as well. Last year, scientists reported that inhibitors of SIRT2 (a Class 3 HDAC) can prevent α-synuclein toxicity in fly and cell models of PD (Outeiro et al., 2007 and see ARF related news story). SIRT2 inhibition also appears to stave off neurodegeneration in a fly model of HD (Pallos et al., 2008).
HDAC-targeting compounds are certainly making a splash in CNS drug development. The global transcriptional effects of such drugs can be a two-edged sword, though—powerful if the right genes are changed in the right directions, devastating if not. “It’s very clear that different HDACs have very different functions,” said Li-Huei Tsai of Massachusetts Institute of Technology, whose study in today’s issue of Neuron proposes HDAC1 deregulation as a common culprit for cell cycle reactivation and DNA damage in p25 mice (see Part 1 of this two-part story).
Philipp Oberdoerffer, whose new work identified a deacetylase (SIRT1) as a joint mediator of genome stability and DNA repair (Oberdoerffer et al., 2008) agreed, but noted that specificity concerns need not overshadow a compound’s therapeutic promise.
“These inhibitors have many, many effects, so if you happen to activate some genes that may be protective in a certain setting, that is already an achievement,” he wrote in an e-mail to ARF.
However, given his and Tsai’s recent studies, which suggest that certain HDACs themselves can be protective, Oberdoerffer cautions against using HDAC inhibitors to prevent neurodegeneration in healthy people. “In the long run, messing with transcriptional regulation and the role of HDACs in the DNA damage response may have adverse effects,” he noted.
To optimize the potential of HDAC-targeting compounds, efforts should be made to nail down their specificity, Tsai suggested. “It’s extremely important to know what you are dealing with. I would say that the way to go is to develop isoform-specific inhibitors. They are going to be more selective and therefore more potent and probably a lot safer,” she told ARF.—Esther Landhuis.
This is Part 2 of a two-part series. See also Part 1.