Scientists are working out whether inhibitors of particular enzymes in one histone deacetylase (HDAC) family may one day restore normal transcription of genes in certain neurodegenerative disorders. Until now, though, they have been unable to determine just where HDACs are active in the healthy living human brain, and how that changes with disease and drug therapy. Enter a new positron emission tomography tracer called [11C]Martinostat. Scientists led by Jacob Hooker, Massachusetts General Hospital, Boston, report in the August 10 Science Translational Medicine that this ligand binds a specific class of HDACs in people, and reveals a distinct regional pattern of enzymatic activity in healthy controls. 

“This is the first quantitative map of class I HDAC in the human brain,” Hooker told Alzforum. “Our goal is to characterize what the signal would look like in healthy individuals so we can have a basis for comparison across different diseases.”

Healthy Signal. A [11C]Martinostat PET scan of a young, healthy person reveals the typical pattern of HDAC expression throughout the brain, with varying levels in gray- and white-matter regions. [Courtesy of Sci Transl Med/AAAS.]

In order to be transcribed, genes are unwound and loosened from their histones by histone acetyl transferases, enzymes that attach an acetyl group to a histone lysine residue. Countering that process are histone deacetylases (HDACs), which pop that acetyl group off and lock the DNA back up. Class I HDACs—HDAC1, 2, and 3 specifically—silence the subset of genes important for learning, memory, and synaptic plasticity (Guan et al., 2009; Haettig et al., 2011). This class of HDACs has been found to be overactive in several neurological disorders—including Alzheimer’s disease (AD) frontotemporal dementia (FTD). HDAC inhibitors have been proposed as treatments (Gräff et al., 2012), and one such drug has been moved into Phase 2 to evaluate its ability to boost expression of the progranulin gene in patients with FTD FRM-0334

To measure levels of these enzymes in the living brain, Hooker collaborated with other groups to develop [11C]Martinostat. This tracer visualizes the density of class I HDACs across the brain at nanomolar doses (Wang et al., 2014). The researchers had tested the tracer in rodents and monkeys, but had yet to try it in humans.

First authors Hsiao-Ying Wey and Tonya Gilbert recruited eight healthy volunteers—four men and four women, average age 29 years—to undergo scans. Uptake peaked 30 minutes after injection and remained stable during the 90-minute scan. Nearly twice as much bound in the gray matter as white. Levels in the gray matter varied by region, with high uptake in the putamen and cerebellum and low uptake in the hippocampus and amygdala.

To the authors’ surprise, this pattern was consistent from one subject to another, implying that HDAC expression is tightly regulated and is a major contributor to local levels of gene transcription. “That tells you that it’s tapping into something that is neurobiologically real,” said Mike Yassa, University of California, Irvine. “The initial set of data is as promising as you would like to see for a tracer of this sort.” Yassa was not involved in the current work, but plans to collaborate with the Hooker lab on research with this tracer in the near future.

To better characterize which HDAC isoforms [11C]Martinostat detected, Wey and colleagues performed thermal shift assays on human brain homogenates. These tests measure the increased thermal stability of a target protein bound to an inhibitor. The compound stabilized HDAC1, HDAC2, and HDAC3, but not other HDACs. To double check that [11C]Martinostat targeted HDACs that affected downstream genes governing neuroplasticity, the authors treated human stem cell–derived neural progenitor cells with micromolar doses of Martinostat, high enough to influence gene expression. At these concentrations, Martinostat itself acts as an HDAC inhibitor. Treatment boosted the expression of several genes relevant for learning, memory, and neurological disease, including BDNF, EGR1, CDK5, SYT1, SYP, and GRN, which encodes progranulin.  

Taken together, these data suggest [11C]Martinostat detects a subset of HDACs that regulate the expression of genes important for neuroplasticity. The tracer could help determine changing epigenetic patterns over the lifespan, as well as in any disease that involves epigenetic modifications mediated by these three subclasses of HDAC including neuropsychiatric disorders such as schizophrenia, depression, and psychosis, said Yassa.

Since submitting this paper, Hooker’s lab has tested people as old as 81, as well as patients with schizophrenia and Huntington’s disease. He declined to share preliminary results about whether those people exhibited any differences in signal. Hooker next plans to collaborate with Yassa and other scientists to pilot these scans in a small sample of AD patients. If they detect changes in HDAC patterns, they plan to conduct scans on about 120 more patients to determine whether [11C]Martinostat can help track disease or predict decline in early disease. The compound might eventually help drug companies decide which patients to include in HDAC inhibitor trials, and reveal whether and how those therapies are changing the epigenetic landscape.

“It's an exciting new area that could have great potential should it continue to develop beyond this initial step,” said William Klunk, University of Pittsburgh. “While this new approach is exciting, it remains to be seen how useful this class I HDAC tracer will be for any disease and whether other similar gene-expression tracers are forthcoming.”

Li-Huei Tsai at the Massachusetts Institute of Technology in Cambridge was excited to see this development, but considered it more as a proof of concept. “The authors show that it’s possible to develop a probe and an imaging approach to look at the spatial distribution of HDACs in the brain.” However, she said that since [11C]Martinostat binds three different isoforms of HDACs, it may not be specific enough. Therapies will need to be as selective as possible to avoid toxicity and safety concerns. A change in one isoform could be masked by expression of the others. Further, if an inhibitor modifies activity but not expression of an HDAC, this compound may not detect any difference. Hooker agreed that more research would be needed to find out how expression changed with disease and therapy.

Further down the road, Hooker plans to seek approval for an F18-labeled derivative of Martinostat his group has developed (Strebl et al., 2015). He also wants to partner with pharmaceutical companies to develop Martinostat or a related compound as a potential therapeutic HDAC inhibitor.—Gwyneth Dickey Zakaib


  1. The paper by Wey et al. on PET imaging of HDACs appears to provide an important new proof of principle, showing that key epigenetic regulators can be imaged in human brain. There is growing evidence that class I HDACs play important roles in a highly diverse range of brain functions, including, as shown primarily by Tsai and colleagues, learning and memory. Class I HDACs are also increasingly suspected to be significant players in Alzheimer’s disease and other major disorders.

    However, in part because of their pleiotropic functions and ubiquitous involvement in chromatin modulation, studies of specific targets and actions of HDACs have been hindered. Most research to date has used drugs or genetic manipulations, the effects of which can be difficult to interpret in a region-, cell-, or gene-specific manner. Therefore, there is a clear need for technological advances that will enable more specific analyses of the targets and mechanisms of these key epigenetic regulators.

    The development of this new selective HDAC ligand for PET imaging appears to be an important step in this direction that could seemingly support analyses of concurrent HDAC expression in multiple brain regions during cognitive activity or in disease. Clearly, the availability of this ligand for use in humans has the important advantage of not having to rely on animal models that may not fully model a human disease or function. Nonetheless, it seems likely that a substantial amount of additional work in animals will be required to assess how well the ligand informs about HDAC function during dynamic interventions and pathological conditions.

Make a Comment

To make a comment you must login or register.


Therapeutics Citations

  1. FRM-0334

Paper Citations

  1. . HDAC2 negatively regulates memory formation and synaptic plasticity. Nature. 2009 May 7;459(7243):55-60. PubMed.
  2. . HDAC inhibition modulates hippocampus-dependent long-term memory for object location in a CBP-dependent manner. Learn Mem. 2011 Feb;18(2):71-9. PubMed.
  3. . An epigenetic blockade of cognitive functions in the neurodegenerating brain. Nature. 2012 Mar 8;483(7388):222-6. PubMed.
  4. . In vivo imaging of histone deacetylases (HDACs) in the central nervous system and major peripheral organs. J Med Chem. 2014 Oct 9;57(19):7999-8009. Epub 2014 Sep 18 PubMed.
  5. . Development of a Fluorinated Class-I HDAC Radiotracer Reveals Key Chemical Determinants of Brain Penetrance. ACS Chem Neurosci. 2015 Dec 21; PubMed.

Further Reading


  1. . Histone deacetylases in memory and cognition. Sci Signal. 2014 Dec 9;7(355):re12. PubMed.
  2. . Pharmacological Selectivity Within Class I Histone Deacetylases Predicts Effects on Synaptic Function and Memory Rescue. Neuropsychopharmacology. 2015 Sep;40(10):2307-16. Epub 2015 Apr 3 PubMed.
  3. . PET Imaging of Epigenetic Influences on Alzheimer's Disease. Int J Alzheimers Dis. 2015;2015:575078. Epub 2015 Oct 22 PubMed.

Primary Papers

  1. . Insights into neuroepigenetics through human histone deacetylase PET imaging. Sci Transl Med. 2016 Aug 10;8(351):351ra106. PubMed.