Expression of DLK/PREF1 Results in Inhibition of Human Myeloid Cell Differentiation and Proliferation through Distinct Molecular Mechanisms.

The molecular pathogenesis of myelodysplastic syndromes (MDS) is poorly understood. Microarray based analyses have shown that the Delta-Like/Preadipocyte-Factor 1 (DLK/PREF1) gene is overexpressed in CD34+ cells from patients with MDS. DLK encodes an EGF-like transmembrane protein subject to proteolytic cleavage and release. Stromal expression of DLK appears to play a role in maintenance of murine hematopoietic stem cells and in supporting B- and T-lymphopoiesis. However, the functional consequences of DLK overexpression in human hematopoietic cells are not known. We have studied the effects of ectopic expression of DLK on the differentiation and growth of HL-60 promyelocytic cells, and found that ectopic expression of DLK inhibits induced differentiation in response to ATRA, PMA and DMSO exposure. DLK mutants were generated to explore mechanisms underlying its effects on hematopoietic cells. These included DLK-EC, expressing only the extracellular portion of the molecule; DLK-dIC, a transmembrane form lacking the intracellular domain; and DLK-ALT, an alternatively spliced form lacking the protease sensitive site. We observed that proteolytic cleavage and release of the extracellular domain was not required for inhibition of hematopoietic cell differentiation, whereas intracellular domain interactions were critical to this DLK function. We have also tested the effect of DLK expression on myeloid cell proliferation. DLK expressing cells demonstrated significantly reduced proliferation compared with controls (91-fold expansion of control cells after 7 days culture vs. 66-fold expansion for DLK expressing cells, p<0.0001). Studies using SNARF-1 labeling to track cell division confirmed the inhibitory effect of DLK expression on proliferation (proliferation index of 6.39±1.0 for control vs. 4.99±0.3 for DLK expressing cells, p=0.04, n=3). Cell cycle analysis revealed that DLK expressing cells exhibited slower progress through G0/G1 phase into S-phase than controls following release from G2/M phase synchronization with Nocodazole treatment (45±5% of DLK expressing cells vs. 66±10% of control cells were in S-phase after 14 hours, p=0.027, n=3). We also evaluated the effects of mutations in the DLK molecule on its anti-proliferative effect. DLK-dIC inhibited cell proliferation to a similar extent as wild type DLK (p=0.42), whereas the DLK-ALT resulted in further reduction in proliferation (p=0.0014). In contrast, the DLK-EC resulted in significantly enhanced cell proliferation compared with wild type DLK (p<0.0001). These results indicate that the inhibitory effects of DLK on proliferation require its expression as a transmembrane molecule but do not require intracellular domain mediated interactions. On the other hand proteolytic cleavage and release of the extracellular domain appears to have a stimulatory effect on cell proliferation. In summary, DLK overexpression in myeloid cells has important functional consequences including inhibition of proliferation and differentiation. The inhibitory effects of DLK are independent of proteolytic cleavage of the molecule and require its expression as a transmembrane protein. The intracellular portion of the molecule is critical to inhibition of differentiation but not to proliferation inhibition. In contrast, proteolytic release of the extracellular portion of the molecule may stimulate cell proliferation but does not affect differentiation. Further investigation of the role of DLK overexpression in abnormal hematopoiesis in MDS is warranted.