Gene transcription within the tumor cell and its microenvironment can be affected by epigenetic modulation in histones and inhibition of histone deacetylases is presently a promising therapeutic option in multiple myeloma (MM). MicroRNAs (miRNAs) are non-coding RNAs that control gene expression; a subclass of miRNAs, named epi-miRNAs, exerts anti-tumor activity by targeting effectors of the epigenetic machinery. We recently demonstrated a key role of tumor suppressor miR-29b as epi-miRNA able to target DNA-methyltransferases and to reduce aberrant hypermethylation in MM. In silico search of miR-29b targets further clarifying its epi-miRNA function, unveiled the histone deacetylase HDAC4. Here, we aimed at characterizing HDAC4 expression, function and its regulation by miR-29b in MM. HDAC4 mRNA, protein levels and activity were found increased in 11 out of 11 MM cell lines as compared to 3 healthy controls. The analysis of our microarray dataset consisting of plasmacells from 4 healthy donors, 55 MM and 29 plasma cell leukemia patients indicated HDAC4 overexpression in cancer samples, suggesting that high HDAC4 expression might be involved in MM pathogenesis. In the same dataset, HDAC4 mRNA expression inversely correlated with miR-29b levels.

To investigate the mechanism by which miR-29b could affect HDAC4 expression, we transfected SKMM1 cells with miR-29b mimics, which resulted in down-modulation of HDAC4 at mRNA and protein level, along with increased acetylation at lysine 8 of histone H4 as well as of tubulin acetylation. This effect was isoform-specific, since levels of other HDACs (namely HDAC1, 2, 6) remained unchanged after miR-29b overexpression. Moreover, miR-29b mimics inhibited the 3’UTR of HDAC4 cloned in a luciferase vector whereas failed to regulate a mutant devoid of a predicted miR-29b target site.

Through loss of function experiments, we assessed the functional significance of HDAC4 in MM cells. Stable shRNA-mediated silencing of HDAC4 phenocopied our previously reported miR-29b-triggered biological effects, since it led to growth inhibition, caspase 3/7-dependent apoptosis, inhibition of cell migration and induction of autophagy in KMS11 and NCI-H929 MM cells. HDAC4 depletion by shRNAs potentiated the in vitro anti-MM activity of dexamethasone, bortezomib and vorinostat. Notably, silencing of HDAC4 led to miR-29b upregulation, likely as consequence of histone H4 hyperacetylation at miR-29b promoter, as assessed by Chip experiments; conversely, HDAC4 overexpression strongly reduced miR-29b levels. These results underscore the presence of a negative feedback loop between miR-29b and its target. Adhesion to BM stromal cells reduced miR-29b levels while induced up-regulation of HDAC4 mRNA levels in NCI-H929 cells, thus suggesting that the huBMM might increase HDAC activity in MM cells via miR-29b down-modulation.

Importantly, the pan-HDAC inhibitor vorinostat also triggered apoptosis, autophagy and anti-migratory effects in MM cells, along with the induction of miR-29b and the down-regulation of HDAC4 and of other miR-29b targets like Sp1, CDK6 and MCL-1; these phenotypical changes were potentiated by miR-29b mimics, whereas were strongly reduced by antagomiR-29b transduction, thus demonstrating a functional involvement of miR-29b in the anti-MM activity of vorinostat.

Finally, in vivo studies using a KMS11 xenograft murine model of MM demonstrated a significant potentiation of vorinostat-induced tumor growth inhibition by subcutaneously delivered neutral lipid emulsion-formulated miR-29b synthetic mimics.

Collectively, our results demonstrate that HDAC4 is aberrantly expressed and represents a novel therapeutic target in MM; moreover, our investigation provides the preclinical rationale for using miR-29b mimics to potentiate HDAC inhibitors activity in MM.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.