LMO2 and TAL1 Cooperate in a Model of Human T−Cell Leukemogenesis.

T−cell acute lymphoblastic leukemia (T−ALL) is associated with the aberrant expression of a limited number of genes, including the basic helix−loop−helix transcription factor TAL1 (SCL) and the LIM−only domain gene LMO2, in the T−cell lineage. These proteins are thought to mediate their leukemogenic effects by interfering with the transcriptional programs that regulate differentiation during normal thymocyte development. The recent X−linked SCID gene therapy trial has highlighted a role for LMO2 overexpression as an early event in T−lineage leukemogenesis, as retroviral integration into the LMO2 locus was detected in multiple patients that went on to develop T−ALL. However, our understanding of the effects of aberrant LMO2 expression upon human T−lymphopoiesis is currently limited. In order to address this area, lineage−depleted human umbilical cord blood cells were transduced with a lentivirus encoding LMO2 or a control virus, then seeded upon OP9−DL1 stroma. As expected, control cells underwent a normal stage−specific program of T cell development concluding with the emergence of a population of CD4+CD8+CD3hi TCRαβ+ cells. Interestingly, LMO2−expressing cells exhibited a differentiation block at the double negative (DN: CD8−CD4−CD7+) stage of T−cell development. These LMO2−expressing DN cells had a growth advantage compared to control cells (23 population doublings over 75 days for LMO2 vs. 15 p.d. over 60 days for control) but were not immortalized as they stopped expanding after 75 days of co−culture.

In the context of T−cell leukemogenesis, these findings suggest that as an initial hit, LMO2 overexpression can induce a blockage in differentiation, resulting in the generation of a proliferative pre−leukemic pool of DN cells. These cells could subsequently accumulate additional mutations leading to the eventual development of an overt leukemia. Given that TAL1 has been shown to accelerate the development of leukemia in LMO2 transgenic mice, and that these two genes are simultaneously overexpressed in a significant percentage of T−ALL cases, this oncogene was an ideal candidate for a second genetic hit. Thus, a retrovirus encoding TAL1 was utilized to infect the LMO2+ DN T−cell population. The expression of TAL1 in these cells significantly increased their proliferative capacity and greatly extended their lifespan, as greater than 60 population doublings occurred over 220 days of culture on stroma. Of note, TAL−1 overexpression appeared to release the LMO2−induced differentiation block at the DN stage, resulting in the emergence of a population of CD4+CD8+CD3− lymphoblasts.

Taken together, these findings describe the first experimental model that studies the early stages of human T−cell leukemogenesis by starting with the physiologically relevant population of primitive primary human hematopoietic cells and analyzing the impact of sequential genetic hits upon T−lymphopoiesis. These data indicate that the aberrant expression of LMO2 contributes to leukemogenesis as an early event by generating a pre−leukemic pool of DN cells and that TAL−1 overexpression in this population acts a cooperating event that leads to the emergence of a highly proliferative, immortalized clone. Given that an experimentally induced leukemia model requires the demonstration of in vivo disease, studies assessing the leukemic potential of human cells co−expressing TAL1 and LMO2 are underway in a novel NOD/SCID system that supports human T cell development.