Acute lymphoblastic leukemia (ALL) is one of the most frequently occurring cancers in infants and young children. For patients suffering from CD20-positive B-cell ALL (B-ALL) and Ph-positive ALL, overall survival rates have been greatly improved after clinical introduction of rituximab and imatinib, respectively. However, T-cell ALL (T-ALL) patients still exhibit poor prognosis, since there is no such an efficient molecular-targeted drug.

It is known that LIM-only transcriptional co-factor LMO2 and its target gene HHEX are essential for self-renewal of T cell precursors and onset of T-ALL. LMO2 directly associates with LDB1 in a large DNA-containing nuclear complex and controls the transcription of T-ALL-related downstream genes. Recently, we reported that overexpression of LIM-homeodomain transcription factor Lhx2 results in liberation of Lmo2 protein from the Lmo2-Ldb1 complex followed by degradation by ubiquitin-proteasome system. We found that proliferation of 5 different human T-ALL-derived cell lines including CCRF-CEM was significantly suppressed by retroviral overexpression of Lhx2. In contrast, enforced overexpression of Lhx2 did not reduce the growth rate of B lymphoma-derived cell line Raji, oral cancer-derived cell line HSC-3, and osteosarcoma-derived cell line Saos-2. Majority of the Lhx2-transduced CCRF-CEM cells was arrested in G0 phase and subsequently underwent apoptosis. Expression of LMO2 protein and HHEX mRNA was repressed by the Lhx2 transduction. Importantly, the Lhx2-mediated growth inhibition was partially rescued by the simultaneous overexpression of Lmo2. However, both C-terminal LIM-domain and homeodomain of Lhx2 were required for the growth-suppressive activity. These data indicated that Lhx2 is capable to blocking proliferation of T-ALL-derived cells in LMO2-dependent and independent fashions. Lhx2 would be a useful molecular tool for designing a new type of anti-T-ALL drug.

In order to develop a new drug that mimics the aforementioned activity of Lhx2, we performed large-scale screening of natural compound libraries to find out a compound that suppresses the proliferation of T-ALL cell line CCRF-CEM, but does not inhibit the growth of B lymphoma cell line Raji. Among 150,000 different compounds, we successfully identified 3 low-molecular-weight compounds (44D-L008, 31D-F005, 21D-D016) that fulfilled the above criteria. 44D-L008 and 31D-F005 possessed a common chemical structure. In the presence of 5μM of 44D-L008 and 31D-F005, proliferation of 5 human T-ALL cell lines including CCRF-CEM was severely blocked. On the other hand, Raji, HSC-3 and Saos-2 were completely resistant to both compounds in the same experimental settings. Intriguingly, 44D-L008 decreased viability of human skin fibroblasts in culture, whereas 31D-F005 displayed no such negative effects on skin fibroblasts, peripheral blood mononuclear cells, and peripheral blood T cells of human origin. These results indicated that small differences in the chemical structure between 44D-L008 and 31D-F005 are responsible for the side effect on normal cells. Finally, we collaborated with a hospital and found that 0.5 μM of 31D-F005 efficiently suppressed the in vitro growth of primary cancer cells of a T-ALL patient. Taken together, theses data demonstrated that 31D-F005 is a promising lead compound for a new anti-T-ALL drug.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.