Introduction: Histone lysine methylation, a reversible event dynamically and reciprocally regulated by lysine methyltransferases (KMTs) and demethylases (KDMs), represents one of the major epigenetic mechanisms for regulation of chromatin remodeling and gene expression re-programming. The KDM4 family, which belongs to the Jumonji C (JmjC)-domain-containing proteins (JMJDs), consists of five members, including KDM4A-E that demethylate H3K9me2/3 and/or H3K36me2/3 in a Fe2+- and α-ketoglutarate-dependent manner. KDM4 proteins are involved in various cellular processes such as gene transcription and translation, DNA replication, DNA repair, apoptosis, and stem cell renewal. Notably, increasing evidence implicates KDM4 dysregulation in promoting genomic instabilities and oncogenesis, thereby which is considered as a potential target for emerging cancer epigenetic therapy. Although KDM4A, a member of the KDM4 family, has been widely studied in many solid tumors including breast, prostate, bladder cancer, its role in hematopoietic malignancies, including multiple myeloma (MM), remains unknown.

Materials and Methods: Human MM cell lines (U266, RPMI8226, H929, OPM-2) were employed. After exposed to hypoxia (or the chemical hypoxia mimetic lactic acid) and anti-MM agents (e.g., bortezomib/Btz), cells were analyzed by flow cytometry, qPCR, Western blot to monitor apoptosis, cell cycle, proliferation (Ki67), DNA double-strand break/DSB (γH2A.X), expression of 1q21 and anti-apoptotic genes, as well as activation of the NF-κB and HIF pathways. The shRNA approach was used to knock down KDM4A for functional evaluation. The findings from in vitro experiments involving cell lines were then validated in primary MM samples to link KDM4A expression to disease progression and therapeutic response.

Results: Analysis of the MM genome-wide GEP databases revealed that KDM4A mRNA was significantly up-regulated in MGUS and MM, but not SMM, compared to normal control, as well as in relapsed MM, compared to newly-diagnosed MM. To our surprise, KDM4A expression rather favored overall survival of MM patients, including those carrying 1q21 gain in whom KDM4A expression was indeed lower than those who did not have this high risk cytogenetic abnormality. Moreover, KDM4A expression correlated adversely with expression of 1q21 genes (e.g., CKS1B, MCL1, PSMD4, ARNT). Whereas basal KDM4A protein level was moderately but clearly higher in MM cell lines carrying 1q21 gain or acquired drug resistance than their counterparts, exposure to hypoxia or lactic acid (but not cobalt chloride) resulted in marked KDM4A up-regulation, accompanied by NF-κB and HIF pathway activation. However, while NF-κB inhibition and to a lesser extent ARNT/HIF-1β knockdown led to a robust increase in hypoxia-induced KDM4A expression, shRNA knockdown or pharmacological inhibition of KDM4A triggered NF-κB activation and HIF expression, as well as up-regulated anti-apoptotic proteins (e.g., Mcl-1, TNFAIP3/A20, CKS1B), in association with increased H3K36me3 rather than H3K9me3. Furthermore, KDM4A knockdown or inhibition sharply diminished Btz lethality and overrode hypoxia-mediated cytoprotection. Interestingly, KDM4A knockdown also increased MM cell proliferation, promoted S phase entry, and attenuated Btz-induced DSB. Last, IHC of sequential bone marrow biopsies revealed that while KDM4A protein was relatively low at diagnosis, its level was markedly increased when patients achieved CR and then fell to the baseline low level at relapse.

Conclusion: KDM4A/JMJD2A, a lysine demethylase that has been recognized as an pro-oncogenic protein via its epigenetic and/or non-epigenetic properties, is identified for the first time as a potential tumor suppressor in MM, particularly in a high risk subtype carrying 1q21 gain. Whereas KDM4A is expressed in MM and can be further induced by hypoxia that naturally exists in bone marrow niche, it seems to play multiple inhibitory roles in cell growth, cell cycle, DNA repair, and drug resistance by suppressing expression of oncogenic and anti-apoptotic genes (especially 1q21 genes), likely via H3K36me3 demethylation, and antagonizing NF-κB and HIF activation. These findings suggest that in contrast to its pro-oncogenic role in certain solid tumors, KDM4A might instead act as a tumor suppressor in MM.

This work was supported by NNSFC (81471165, 81670189, and 81670190).

Disclosures

Kumar:AbbVie: Membership on an entity's Board of Directors or advisory committees, Research Funding; KITE: Membership on an entity's Board of Directors or advisory committees, Research Funding; Janssen: Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution