Valproic acid: an old drug newly discovered as inhibitor of histone deacetylases.

Fusion proteins encoded by several types of chromosomal translocations in promyelocytic leukemia can serve as aberrant transcriptional repressors relying on recruitment of histonedeacetylases (HDACs) into DNA-associated multi-protein complexes. Thus, inappropriate modulation of chromatin structure by HDACs and subsequently repression of gene expression that is critical for myeloid differentiation appear to be major factors in the development of the disease. They identify inhibitors of HDACs as prime candidates for novel anti leukemic drugs. Over the last years several candidate compounds have been introduced into clinical trials and have successfully been used in compassionate use protocols. Amongst them phenylbutyrate served as the first example to establish proof of principle. Novel drugs such as suberoylanilide hydroxamic acid (SAHA) are developed for example by modifications of the microbial HDAC inhibitory compound trichostatin A with a hydroxamic acid as the key structural element. The branched chain carboxylic acid valproic acid (VPA) that is in use as antiepileptic drug over decades was also discovered to inhibit HDACs and preferentially class I HDACs. HDAC inhibition is likely to mediate the teratogenic side effects of VPA but not the antiepileptic activity. In contrast to other HDAC inhibitors VPA also induces proteasomal degradation of HDAC2. None of the currently available compounds may be the optimum HDAC inhibitory drug but each of them may serve to answer urgent questions concerning the concept of HDAC inhibition in the treatment of malignant diseases. Prominent questions are i) whether and by which mechanisms HDAC inhibition can be expected to affect a malignant disease not only in the early stage but also at later stages that have acquired additional genetic defects, ii) which forms of cancer in addition to myelocytic leukemia respond to HDAC inhibition, iii) by which markers those susceptible forms could be identified and iv) which individual HDACs are the most critical isoenzymes to address in treatment of malignant diseases.