Todralazine hepatotoxicity: A sting in the histone tail
Article Outline
Considering the amount of DNA that must be packaged into every cell of our bodies it is not surprising to find that a highly complex system of compaction exists in every nucleus. DNA is wound around protein complexes known as nucleosomes, with 146 base pairs of DNA being wrapped around each nucleosome in 1.75 super helical turns [1], [2]. Each nucleosome is an octamer typically made up of four canonical histones: H2A, H2B, H3 and H4 and is formed during the S phase of the cell cycle. In situations where specialised chromatin regions are required variants of H2A (such as H2A.Z and H2A.X) and H3 (such as H3.3 and C-ENPA) are inserted [3]. Unusually, this process is independent of DNA replication. Nucleosomes are then packaged into higher order chromatin structures such as chromonema fibres that eventually form chromosomes [2].
The nucleosomes in themselves are relatively stable structures due to the multitude of histone–DNA contacts. Yet chromatin is also surprisingly dynamic and a series of complicated nucleosomal remodelling and modifying systems have emerged to allow regulation of regions of the genome that need to be read constantly and regions that can be switched off when the genes are not required; during terminal differentiation for example.
Although histone modifications were discovered more than 40 years ago, [4] the last 10 years have seen big advances in our understanding of how these events are regulated [5]. Key to this is the appreciation of how modifications to lysine, arginine and serine amino acids of the N terminal tails of individual histones in a nucleosome occur and how these can affect the structure and compaction of chromatin as well as influencing which factors bind to the individual histones and adjacent DNA. The modifications discovered to date include phosphorylation, methylation, acetylation, sumoylation, ribosylation and ubiquitination. Additional complexity is provided by the fact that the various amino acids can be modified in more than one way and that modifications on one amino acid can affect modifications on an adjacent amino acid. In addition, modifications in the cores of histones are increasingly recognised to play an important part in regulation of chromatin structure [6]. The huge number of possible chromatin modifications has been proposed to form a ‘histone code’ elements of which are heritable giving rise to the increasingly prominent field of epigenetics [7].
One of the most common modifications studied to date is histone acetylation of lysine residues in histone tails; in particular H3 and H4. Put simply, histone acetylation generally results in an open chromatin structure enabling genes in that region to be read. This acetylation is mediated by enzymes known as histone acetyltransferases (HATs). Deacetylation is mediated by histone deactylases (HDACs) and leads to closed repressed chromatin (see Fig. 1). This, however, is a highly simplified version of events and increasing levels of complexity are being continuously discovered (see below) [5].
Fig. 1.
Illustration of changes that occur during acetylation and deacetylation of histone tails. Acetylation by histone acetyl transferases (HATs) allows unwinding of the compacted chromatin structure and access by the RNA polymerase and other components of the basic transcriptional machinery allowing gene transcription. The reverse is true of histone deacetylation.
Most work in the epigenetic field has concentrated on the study of changes in histones that accompany malignant transformation [8]. For example, recent work published in Nature has demonstrated that prostate cancers can be prognostically stratified according to various histone acetylation patterns [9]. However, few studies have been undertaken outside the field of oncology and even less in Hepatology circles. A study back in 2001 started to address the role that histone acetylation may play in hepatocyte regeneration using an in vitro model. The authors showed that H4 hyperacetylation of the MAT2A gene occurs in response to hepatocyte growth factor stimulation of hepatocytes in culture [10]. In addition, a recent paper in Cell, Death and Differentiation highlighted the role of histone modifications in the transdifferentiation of hepatic myofibroblasts in an in vitro model of liver fibrosis [11]. Little if any work has focused on the potential role of histone modifications in various forms of liver disease in vivo.
This is now no longer the case. In this issue of the Journal, Murata et al. [12] provide data to suggest a new mechanism of drug induced hepatotoxicity associated with a defect in histone acetylation. The study started with a clinical observation. Having previously reported seven cases of todralazine (a hydralazine derivative used in the treatment of hypertension) induced hepatic failure [13], they report here a further case with fulminant hepatic failure requiring liver transplantation 20 days after admission. Histology of the explant revealed submassive hepatocyte necrosis with little inflammation and little hepatocyte proliferation. Given that todralazine and other hydralazine derivatives seem to induce failure of tissue regeneration (they cause aplastic anaemia for example) and that histone acetylation plays such a fundamental role in activation of gene transcription, the authors hypothesised that the two phenomena were linked. This seems especially likely as histone acetyltransferases (HATs) detoxify hydralazine derivatives by acetylation. They, therefore, proposed that todralazine caused a reduction of histone acetylation with subsequent failure in liver regeneration.
The authors started off by looking at acetylated histone H4 by immunohistochemistry in patients with todralazine induced liver failure compared to those with acute hepatitis A or thiopronine induced liver injury. They showed that whilst the latter two groups had high levels of acetylated H4, the former had very low levels. Staining with proliferating cell nuclear antigen (PCNA; a marker of dividing cells) mirrored the findings with acetylated H4. These findings were not observed in bile ducts suggesting that the defect in acetylation was localised to hepatocytes. In addition, acetylated H3 and methylated H4 levels were similar between all the groups.
Murata et al. then went on to study the effect of hydralazine derivatives on histone acetylation in vitro. They showed that todralazine, isoniazid and hydralazine all dose dependently inhibited core histone acetylation whereas diltiazem (a hydralazine derivative that cannot be acetylated by HATs) did not.
Finally to study the effects of todralazine in an experimental model the authors fed mice with the drug for 4 months. They showed that the drug per se induced mild hepatitis with small rises in ALT (67 compared with 27 in the control group). At baseline, acetylated H4 levels and PCNA positivity were reduced in the drug treated mice. In addition, levels of cyclin D1 (another marker of cell cycling) and HAT activity were also reduced. Feeding the mice with todralazine for 1 month only was insufficient to induce these changes consistent with a cumulative effect on saturating natural acetylating enzyme activity in hepatocytes.
When the drug treated mice were then treated with Fas ligand to induce acute liver injury (a well-recognised experimental model) all of them died between 16 and 24
h compared to none of the controls. This was associated with wide ranging hepatic necrosis, reduction in acetylated H4, reduction in PCNA staining and a reduction in liver weights suggesting a failure of hepatocyte regeneration in the drug treated group.
This study provides the first evidence of drug induced hepatotoxicity potentially mediated through inhibition of a histone modification. However, several important and critical caveats remain. The authors used a strain of mice (A/J strain) that were slow acetylators in their in vivo model. These mice would be expected to accumulate much higher levels of todralazine in their livers to those normally seen in most patients. In addition, the authors failed to detect any defects in acetylating enzymes in their genotyping of the seven patients they had reported with todralazine induced FHF. All the patients that they did admit with FHF secondary to todralazine had clearly suffered a “second hit” inducing the acute hepatitis from which they could not recover due to their inability to induce hepatocyte regeneration. The nature of this “second hit” is not at all clear but is crucial for the prediction and prevention of future cases.
But perhaps most importantly of all, although there is circumstantial evidence for the link between the defect in acetylation and the failure of hepatocyte regeneration on the basis of their data, no direct causality was demonstrated. Moreover, the authors provide no data to suggest which genes in particular are suppressed by the failure of H4 acetylation. These data could be easily generated using chromatin immunoprecipitation assays of genes known to be involved in hepatocyte regeneration [14].
The complexities of histone acetylation have also been largely overlooked in this study. It is important to state that HATs and HDACs do not exclusively act on histones. In fact, they are essentially protein acetylators and deacetylators, respectively. As a result the toxicity of todralazine could be related to a defect in acetylation of potentially one of hundreds of proteins. In addition, HATs also function as transcriptional regulators and the toxicity associated with todralazine could be as a result of a defect in transcription unrelated to histone modifications [15].
There is emerging evidence that acetylation on different lysine residues on the tails of H4 is associated with different effects on gene transcription. Acetylation events on lysines 5, 8 and 12 are known to be permissive whereas acetylation on lysine 16 is associated with specific cell signalling cascades [16], [17]. Moreover, loss of H4 lysine 16 acetylation is a common feature of transformed cells [8]. The exact lysine residues affected by the treatment with todralazine were not addressed in this study.
Furthermore, it is increasingly recognised that histone acetylation is an extremely dynamic process with different pools of acetyl groups being added and removed from histone tails at different speeds [5]. Taking a snapshot of total acetylation as done in this study will underestimate some of the signalling complexities occurring at the gene promoter level.
These caveats aside, Murata et al. have demonstrated a potentially novel mechanism of drug induced liver toxicity; something which should spur other researchers in the field to investigate similar mechanisms in the pathogenicity of other hepatotoxic drugs. The authors would do well to continue to investigate the mechanism of todralazine toxicity using less blunt tools than the ones used in the current study. And finally, what this study also highlights is the relatively embryonic stage of epigenetic research in liver disease; something that needs addressing urgently in the next few years.
References
- . Chromatin remodelling: the industrial revolution of DNA around histones. Nat Rev Mol Cell Biol. 2006;7:437–447
- . Gene activation and deactivation related changes in the three-dimensional structure of chromatin. Chromosoma. 2005;114:331–337
- In and out: histone variant exchange in chromatin. Trends Biochem Sci. 2005;30:680–687
- . The acid extraction of histones from calf thymus deoxyribonucleoprotein. J Mol Biol. 1966;15:409–419
- . Enhanced histone acetylation and transcription: a dynamic perspective. Mol Cell. 2006;23:289–296
- . The tale beyond the tail: histone core domain modifications and the regulation of chromatin structure. Nucleic Acids Res. 2006;34:2653–2662
- . Reading signals on the nucleosome with a new nomenclature for modified histones. Nat Struct Mol Biol. 2005;12:110–112
- . Towards the human cancer epigenome: a first draft of histone modifications. Cell Cycle. 2005;4:1377–1381
- Global histone modification patterns predict risk of prostate cancer recurrence. Nature. 2005;435:1262–1266
- Hepatocyte growth factor induces MAT2A expression and histone acetylation in rat hepatocytes: role in liver regeneration. FASEB J. 2001;15:1248–1250
- Mann J, Oakley F, Akiboye F, Elsharkawy A, Thorne AW, Mann DA. Regulation of myofibroblast transdifferentiation by DNA methylation and MeCP2: implications for wound healing and fibrogenesis. Cell Death Differ 2006.
- . A novel mechanism for drug-induced liver failure: Inhibition of histone acetylation by hydralazine derivatives. J Hepatol. 2007;46:322–329
- . Fulminant hepatic failure caused by ecarazine hydrochloride (a hydralazine derivative). Hepatology. 1996;23:465–470
- . Liver regeneration. Hepatology. 2006;43:S45–S53
- . HATs off to PIC assembly. Mol Cell. 2006;23:776–777
- . Genomic characterization reveals a simple histone H4 acetylation code. Proc Natl Acad Sci USA. 2005;102:5501–5506
- . Mapping global histone acetylation patterns to gene expression. Cell. 2004;117:721–733
PII: S0168-8278(06)00633-7
doi:10.1016/j.jhep.2006.11.007
© 2006 European Association for the Study of the Liver. Published by Elsevier Inc. All rights reserved.
