Heterogeneous nuclear ribonucleoproteins (hnRNPs) represent a large family of RNA-binding proteins. Heterogeneous nuclear ribonucleoprotein D-like (HNRPDL) is a member of this family, which is able to bind RNA thus regulating multiple steps of mRNA metabolism. Chronic myeloid leukemia (CML) originates from hematopoietic stem cells acquiring BCR-ABL fusion gene. The specific inhibitor against BCR-ABL (e.g. imatinib mesylate, IM) has improved CML management though single agent is not a cure yet. Thus the delineation of growth/survival mechanisms of CML stem/progenitor cells is still needed. We have reported the aberrant expression of HNRPDL in CML CD34+ cells; however, the function of HNRPD and its underlying mechanism remain unclear yet.

In the present study, we showed that HNRPDL silencing significantly inhibited the growth and the colony-forming cell (CFC) production of K562, MEG-01 and BV173 cells. Moreover, the CFC production of CML CD34+ cells was suppressed by HNRPDL silencing (ranging from 36% to 73%, P=0.014, n=6). Then BaF3/BCR-ABL and HNRPDL silenced BaF3/BCR-ABL cells were injected into lethally irradiated mice; though most mice developed leukemia (8/8 vs. 7/8, 1×105 cells/mouse), HNRPDL silencing attenuated BCR-ABL induced leukemia analyzed with Kaplan-Meier method (P<0.0001). In addition, less leukemic cells were detected in both bone marrow (24±2% vs. 6±2%, P=0.002) and spleen (12±1% vs. 3±2%, P=0.012) of the leukemic mice in HNRPDL silenced group than those in control group. Conversely, HNRPDL overexpression conferred BaF3 cells hyper-responsiveness upon murine IL-3 treatment and induced lethal leukemia in mice (8/8, 1×106cells/mouse). HNRPDL and BCR-ABL cooperatively promoted the growth of BaF3 cells as well. Lastly, we found that HNRPDL silencing sensitized K562 cells upon IM treatment as more apoptotic cells were induced; in contrast, HNRPDL overexpression protected K562 cells from IM treatment.

To investigate how HNRPDL acts, microarray data comparing HNRPDL silenced K562 cells with their control have been generated. Q-RT-PCR was performed to validate that the expression of pre-B-cell leukemia homeobox 1 (PBX1) was significantly decreased in HNRPDL silenced K562 and CML CD34+ cells (n=3) than their control cells. We found that the PBX1 expression decreased more rapidly in HNRPDL silenced cell than control cells upon Actinomycin D treatment. Sequence analysis showed that PBX1 mRNA contained a HNRPDL consensus binding motif (5'- ACUAGC-3') in its 3'-untranslated region (3'-UTR). A piece of 3'-UTR only containing this consensus motif was sub-cloned in the 3'-end of a luciferase reporter vector, which conferred elevated reporter activity than control vector (3-fold, P=0.001, n=4). The elevated reported activity was entirely abolished when a mutant motif (5'-UAACAA-3') introduced in the reporter vector. In addition, the enhanced reporter activity was decreased in HNRPDL silenced cells than their control. Importantly, the PBX1 transcript expression was higher in CML CD34+ cells (n=10) compared with that in normal control cells (n=7) significantly (4-fold, P=0.019). PBX1 silencing inhibited the growth of K562 cells and sensitized them to IM treatment; while PBX1 overexpression "rescued" the suppressed growth and sensitized IM response caused by HNRPDL silencing. Taken together, our data has shown that HNRPDL plays a pivotal role in the growth and IM response of CML cells partially through its post-transcriptional regulation of PBX1.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution