Histone Deacetylase: An Epigenetic Target for Cancer Therapy
ACVIM 2008
Douglas H. Thamm, VMD, DACVIM (Oncology)
Fort Collins, CO, USA

Epigenetics--What Does It Mean?

The new science of epigenetics is leading to more innovative methods of treating cancer. Epigenetic changes are heritable traits that are mediated by changes in DNA other than nucleotide sequences, and play an important role in regulating gene expression. Aberrant gene transcription resulting from epigenetic changes is a frequent event in the molecular pathogenesis of cancer, and transcriptional silencing of key non-mutated genes such as tumor suppressor or DNA repair genes is common in the pathway toward malignancy.1 The phenotype of most cancers is likely the result of a combination of some mutated genes and some genes functionally modified by these epigenetic changes. In contrast to permanent genetic defects, the reversible nature of epigenetic aberrations constitutes an attractive therapeutic target.

The epigenetic changes associated with altered gene transcription can be divided into 3 main groups: histone modifications, DNA methylation, and genomic imprinting. Methylation involves the addition of a methyl group to cytosines, often in the promoter regions of genes, and typically when they are part of a CpG dinucleotide repeat, which results in heritable transcriptional silencing. Genomic imprinting refers to the selective expression of alleles in a parent-of-origin specific manner, i.e., a single allele from one parent but not the other, is selectively expressed as a result of chemical modifications to DNA. This occurs as a result of a combination of differential methylation and histone modifications.1 Histone modifications are discussed below.

The Histone Deacetylase Enzymes

It is easy to forget that, in the cell, DNA is not a linear molecule; rather, it adapts secondary and tertiary structure and assembles itself into complexes with a variety of structural proteins. The core unit of structure of DNA is the histone, a phylogenetically conserved octamer of protein subunits around which the DNA strand associates. Histones undergo a variety of posttranslational modifications, including methylation, acetylation, ubiquitination, and phosphorylation among others. It is this combination of modifications that is thought to comprise the "histone code", which governs diverse functions such as transcriptional regulation, DNA repair, and chromosome condensation in mitosis.2

The differential acetylation of histones, controlled by various histone acetyltransferases and histone deacetylases (HDACs) is important in governing the structure of chromatin and thereby plays a role in modulating expression of genes associated with cellular proliferation, differentiation and survival.2 The acetylation of histones by HDACs results in formation of heterochromatin (compact chromatin) and transcriptional repression. This transcriptional repression can negatively regulate the expression of a variety of developmental and tumor suppressor genes.2

To date, 18 mammalian HDAC enzymes have been reported. They are generally divided into 2 classes, based on sequence homology to yeast counterparts. Class I HDACs are expressed in most cell types, and are found almost exclusively in the nucleus. Class II HDACs have a more restricted expression pattern, and are capable of both nuclear and cytoplasmic localization in response to different cellular signals.3

The precise mechanisms of action of all HDAC enzymes are not yet well defined. The primary function of HDACs is to affect repression of transcription by altering chromatin condensation, selectively altering gene expression.2 However, HDACs also are capable of differentially acetylating a variety of non-histone protein substrates, which can further regulate gene transcription, as well as regulating protein stability and protein-protein interactions. Well-characterized non-histone targets of HDACs include p53, estrogen and androgen receptors, the E2F1 and HIF1-alpha transcription factors, c-MYC, NF-kappaB, and the HSP90 chaperone protein.4 It is likely that there are many other non-histone targets that have yet to be elucidated.

The pharmacologic inhibition of HDAC is capable of antitumor effects in a wide variety of human cancer models.Importantly, normal cells are more resistant to the cell-death inducing properties of HDAC inhibitors than transformed cells.5 Growth inhibition is commonly observed, at least in part as a result of induction of expression of the cell cycle checkpoint regulator p21 (CDKN1A). The pathways identified in mediating HDAC inhibitor-induced transformed cell death include apoptosis by the intrinsic and extrinsic pathways, mitotic catastrophe/cell death, autophagic cell death, senescence, and reactive oxygen species facilitated cell death.2

In addition to direct effects on tumor cell proliferation and survival, inhibitors of HDAC can mediate a variety of other putative antineoplastic effects, which are listed in Table 1.

Histone Deacetylase Inhibitors

A number of compounds that specialize in inhibiting HDACs have progressed rapidly through clinical trials.2 For example, suberoylanilide hydroxamic acid (SAHA, also known as Vorinostat and marketed as Zolinza®; Merck & Co., Inc., NJ, USA) was approved last year for treatment of cutaneous T cell lymphoma. Approval was granted as part of the FDA orphan drug program, following a priority review. A variety of other natural, semi-synthetic, and synthetic inhibitors of HDAC are in various stages of clinical development (Table 2). There has been encouraging antitumor activity in a variety of tumor types in these early studies, including hematopoietic and solid tumors.

Table 1. Putative antineoplastic mechanisms associated with HDAC inhibition.



Angiogenesis inhibition

Reduction in activity of Hif-1/VHL/VEGF axis: Decreased angiopoietin-2 synthesis: Decreased endothelial commitment of progenitor cells: Decreased endothelial NOS: Decreased bFGF synthesis: Inhibition of in vitro endothelial proliferation and tube formation: Induction of endothelial cell apoptosis and growth inhibition

Decreased bone loss

Induction of osteoclast apoptosis: Promotion of osteoblast maturation: Enhanced bony differentiation of mesenchymal stem cells


Upregulation of MHC class I and class II antigen expression: Upregulation of costimulatory molecule expression (B7.1, B7.2 CD40, ICAM-1): Re-expression of repressed tumor antigens

Telomerase inhibition

Decreases in TERT expression, telomerase activity, and telomerase length

Anti-inflammatory effects

COX-2 inhibition: Decreased Toll-like receptor mediated proinflammatory gene transcription: Decreased IL-6 and IL-1beta production

Table 2. HDAC Inhibitors in human clinical development.

Phase of


Company / sponsor





MethylGene / Pharmion




LBH589, LAQ824




FK228 / Depsipeptide / Romidepsin




Pivaloxymethyl butyrate / AN-9 / Pivanex


PDX101 / Belinostat

CuraGen / TopoTarget

Baceca® (topical valproate)


Savicol® (timed-release valproate)



SAHA / vorinostat / Zolinza®
(approved for cutaneous T cell lymphoma)


Phenylbutyrate / Buphenyl®
(approved for urea cycle disorders)


Sodium valproate / Depakote®
(approved for epilepsy)


There is tremendous structural heterogeneity among the HDAC subclasses and even among the individual isoforms. Increasingly, this heterogeneity is being shown to give rise to distinct functions.3 Despite these emerging differences, many of the known HDAC inhibitors are class indiscriminate with respect to the HDACs they inhibit. For example, SAHA is a non-specific agent, inhibiting the activity of class I and II HDAC enzymes equally. Therefore, much of the activity of HDAC inhibitors has been broadly attributed to all HDAC subclasses. However, with respect to cancer therapy it is emerging that class I HDAC enzymes may be more clinically relevant targets.6 This is still very controversial and not absolute, as illustrated by findings that inhibition of HDAC6, a class II HDAC, leads to acetylation and disruption of the chaperone function of heat-shock 90 protein (Hsp90) in leukemic cells.7 Nevertheless, the current evidence is swaying towards class I HDACs as the important cancer targets and this may call the therapeutic potential of broad-spectrum inhibitors such as SAHA into question.

HDAC Inhibitors in Veterinary Oncology

To date, there is little to no information available regarding the use of HDAC inhibitors in domestic animals. A single case report described the postoperative administration of SAHA to a dog with hemangiosarcoma with a favorable outcome;8 however, the dose and schedule were empirically chosen, and no pharmacokinetic or pharmacodynamic evaluation was performed to demonstrate biologic activity. A recent abstract reported on the administration of SAHA to a slightly larger cohort of dogs with hemangiosarcoma.9 Gastrointestinal toxicity was dose-limiting, and the study was terminated prematurely as a result of toxicity. Again, pharmacodynamic data were not reported. The astronomical cost of SAHA will undoubtedly limit its broad use in veterinary patients for the next several years.

The HDAC inhibitors phenylbutyrate and sodium valproate (VPA) are FDA approved for other indications, and are comparatively inexpensive versus SAHA. VPA is marketed as an anticonvulsant and is available in both oral and injectable forms, and preliminary pharmacokinetic information in dogs is available from studies performed in the 1980's. VPA is rarely used as an anticonvulsant in dogs due to its very short half-life. Sustained-release oral formulations of VPA are now available; however, published pharmacokinetic information in dogs is not available. Human clinical trials have evaluated VPA as a single-agent, as well as in combination with demethylating agents, retinoids, or conventional cytotoxic agents. It appears that VPA concentrations capable of modulating HDAC activity are achievable in vivo in humans. It remains to be seen if this will be the case in dogs.

HDAC Inhibitors and Traditional Cancer Therapy

As with most novel, targeted therapies, they may find their greatest utility when combined with standard forms of therapy such as chemotherapy, radiation therapy and surgery. Preclinical data support the concept that HDAC inhibition can potentiate the efficacy of both chemotherapy and radiation therapy.

Following pretreatment with HDAC inhibitors, enhanced sensitivity to many classes of chemotherapeutic agents has been demonstrated. These include topoisomerase inhibitors such as doxorubicin, epirubicin and etoposide, antimetabolites such as 5-fluorouracil, gemcitabine, fludarabine and cytosine arabinoside, antimicrotubule agents such as paclitaxel and docetaxel, and platinum drugs, as well as many other novel/targeted agents. In many cases, the mechanisms of chemosensitization are not completely understood. One mechanism responsible for enhancement of sensitivity to DNA-interacting agents by HDAC inhibitors may be through induction of a conformational change in chromatin structure, allowing better access of the drug to the DNA molecule.10 Hyperacetylation of tubulin may alter the sensitivity of cells to microtubule-interacting agents such as vinca alkaloids and taxanes.11 Double-strand DNA breaks may be repaired inefficiently as a result Ku70 hyperacetylation, leading to decreased DNA binding affinity.12 Additionally, alterations in the expression of pro- and anti-apoptotic proteins may enhance apoptosis in response to diverse chemotherapeutic agents. In many, but not all cases, these synergistic effects are sequence-specific and most profound when cells are pretreated with the HDAC inhibitor. Interestingly, at least one study has suggested that HDAC inhibitors may protect against the cardiotoxic effects of doxorubicin as well.13 However, there are potential adverse effects associated with HDAC inhibition and chemotherapy: some studies have suggested that continuous exposure to HDAC inhibitors may induce the expression of P-glycoprotein, an ATP-dependent efflux pump which may confer resistance to a variety of antitumor natural products.14 Importantly, these data suggest that short-term HDAC inhibitor treatment around the time of chemotherapy administration may be effective, and even preferable to continuous treatment. This schedule of treatment may reduce toxicity and cost.

Similarly, enhancement of radiation sensitivity has been observed with HDAC inhibitor treatment. Interestingly, the topical application of HDAC inhibitors was shown in one study to mitigate acute cutaneous effects from radiation therapy and to accelerate radiation-induced cutaneous wound healing.15

HDAC Inhibitors--Current and Future Veterinary Studies

Our laboratory has been evaluating combinations of VPA and doxorubicin in canine and human osteosarcoma in vitro, and in a mouse model. We have demonstrated that pretreatment of osteosarcoma cells with clinically relevant concentrations of VPA enhances the antiproliferative and pro-apoptotic effects of doxorubicin, and significantly inhibits tumor growth in a murine subcutaneous xenograft model of canine osteosarcoma. This appears to be due, at least in part, to increased nuclear accumulation of doxorubicin by the tumor cells following VPA exposure. We have recently completed accrual to a phase-I clinical trial in tumor-bearing dogs of escalating VPA followed by a standard dose of doxorubicin; analysis of pharmacokinetic and pharmacodynamic endpoints is ongoing. Future studies may include a randomized, placebo-controlled trial of doxorubicin +/- VPA in canine osteosarcoma, as well as evaluation of other, novel HDAC inhibitors.


1.  Jones PA, et al. Cell 2007;128(4):683.

2.  Xu WS, et al. Oncogene 2007;26:5541.

3.  de Ruijter AJM, et al. Biochem J 2003;370:737.

4.  Glozak MA, et al. Gene 2005;363:15.

5.  Kelly WK, Marks PA. Nat Pract Clin Oncol 2005;2:150.

6.  Karagiannis TC, El-Osta A. Leukemia 2005;21:61.

7.  Bali P, et al. J Biol Chem 2005;280:26729.

8.  Cohen LA, et al. Vet Comp Oncol 2004;2:243.

9.  Overley B, et al. Proc Vet Cancer Soc 2007:84.

10. Marchion DC, et al. J Cell Biochem 2004;92:223.

11. Catalano MG, et al. Endocr Relat Cancer 2007;14(3):839.

12. Chen C-S, et al. Cancer Res 2007;67(11):5318.

13. Daosukho C, et al. Free Radical Biol Med 2007;42:1818.

14. Tabe Y, et al. Blood 2006;107(4):1546.

15. Chung Y-L, et al. Mol Cancer Ther 2004;3(3):317.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Douglas Thamm, VMD, DACVIM (Oncology)
Colorado State University
Ft. Collins, CO