Histone deacetylase (HDAC) inhibitors are being developed clinically for the treatment of leukemia. Because HDACs are composed of a large number of different proteins, and substrate specificity may differ among different HDAC inhibitors, it is important to understand if human leukemias are characterized by specific HDAC expression patterns. To study this, we have analyzed using real-time PCR and Western blots, all major class I and II HDACs in normal marrow controls (N=13, including 10 CD19+ B cell specimens), leukemia cell lines (N=25), samples from patients with AML (N=6), MDS (N=12), CLL (N=10) and human samples (N=6) obtained pretreatment and sequentially from patients with leukemia treated with two different phase I clinical trials of HDAC inhibitors: MGCD0103 and vorinostat. In general, normal controls were characterized by low levels of HDACs 1 to 10, although normal CD19+ B cells exhibited a significant increased expression of HDAC1 and 5. In leukemia cell lines, HDAC 1, 2 and 3 were expressed at higher levels than 4 to 10 but there were no differences between leukemia cell lines and normal controls or B cells. HDAC mRNA expression was not modified by cell proliferation or treatment with HDAC inhibitors. No specific HDAC expression profiles were detected in primary human AML or MDS samples. In contrast, CLL primary samples were characterized by an overexpression of HDAC 1,3,5 and 10, although this pattern was not significantly different than that of normal CD19+ B cells. Sequential analysis of human samples obtained from patients treated with two different HDAC inhibitors, vorinostat or MGCD0103 on two different clinical trials, did not affect expression profiles in patients with MDS or AML. Overall, mRNA expression results correlated with protein levels. In summary, our results indicate that AML or MDS are not characterized by a leukemia specific HDAC expression profile but that B cells and B cell leukemia are characterized by a significant overexpression of HDAC 1. This could explain the activity observed with HDAC inhibitors in B cell malignancies and serve as the bases for clinical studies of HDAC inhibitors in CLL.

Author notes

Disclosure: No relevant conflicts of interest to declare.