Loss of p19 Arf Increases the Self-Renewal Capacity of Immature Thymocytes That Over-Express LMO2

Abstract 4198 LMO2 is a lim-only domain transcription factor that mediates protein-protein interactions in a multimeric complex containing LDB1, E47, Tal1 and either GATA1 or GATA2. It was discovered as a result of chromosomal translocations in T-cell leukemia; where it becomes abnormally regulated by T-cell receptor elements and is involved in approximately 9% of cases. LMO2 transgenic mice develop T-cell leukemia with a latency of between 10–12 months indicating that other co-operating genetic events are required. In recent X-SCID gene therapy trials 4 patients developed T-cell leukemia due to vector–induced activation of LMO2 gene expression. Two out of the 4 patients had loss of the CDKN2A locus with another having an activating insertion in BMI1, a known repressor of the CDKN2A locus. The CDKN2A locus encodes both p19 Arf and p16 INK4A , and p19 Arf suppression is known to increase self-renewal in a variety of different stem cell systems. Because LMO2 has been shown to promote thymocyte self-renewal (Curtis et al, Science 2010); we investigated whether Arf loss could collaborate with LMO2 in regulating self-renewal of thymocyte progenitors. We utilized OP9-DL1 stromal cells that express the Notch ligand; Delta-ligand 1 which enables differentiation of immature bone marrow cells and thymocytes into mature T-cells. When cultured on OP9-DL1 stromal cells LMO2+ Arf +/+ and LMO2+ Arf -/- populations had a block in progression to mature T-cells at the DN2 stage of thymocyte differentiation. These DN2 thymocytes were sorted and replated onto fresh OP9-DL1 stromal cells and analysis 7 days later showed that both populations had 99% DN2 stage thymocytes, indicating that LMO2 was inducing a complete block in thymocyte differentiation. LMO2+ Arf -/- DN2 thymocytes had 3 fold more thymocytes than the Arf +/+ population 11 days after the thymocytes had been sorted. This indicates that Arf loss may be increasing the self-renewal of the DN2 population. We next transplanted either 2×10 5 LMO2+ Arf -/- or 2×10 5 LMO2 Arf +/+ DN2 thymocytes together with 2×10 5 WT bone marrow cells into 2 groups of 10 WT lethally irradiated syngenic recipients. After only three weeks LMO2+ Arf -/- (n=7) cells were detected in the thymi of recipient mice. Marked cells came to represent about 18% of total thymocytes between weeks 6 to 15 post-transplant, these LMO2+ Arf -/- thymocytes were able to differentiate into mature T cell lineages in the thymus. By week 18 the LMO2+ Arf -/- cells dramatically increased to 97% of total thymocytes indicating that they maybe a preleukemic pool. In contrast, no thymic repopulation with LMO2 Arf +/+ (N=7) cells was noted. Serial transplant experiments were performed by isolating mCherry+, LMO2+ thymocytes 12 weeks post transplant. Then 2×10 5 LMO2+ thymocytes were transplanted into secondary Rag2 -/- γ c -/- recipients. After 6 weeks the thymi of 2 secondary mice were analyzed and were 100% mCherry+, LMO2+ and contained mature T-cells. These LMO2+ Arf -/- thymocytes had the capacity to be serially transplanted demonstrating that Arf loss was contributing to the self-renewal potential of these LMO2+DN2 thymocytes. Comparing our data to the published result for LMO2+ Arf +/+ , a 100 fold less thymocytes were transplanted in our assay potentially explaining the lack of engraftment with our LMO2+ Arf +/+ cells. Furthermore we observed activation of the Arf locus when these thymocytes were transplanted in vivo, indicating that somehow these LMO2+thymocytes were inducing Arf activation at later times after transplant. p19 Arf expression in thymocytes has been reported to induce p53 dependent apoptosis (Miyazaki et al, Immunity 2008). This supports the concept that Arf loss contributes to the self-renewal potential of thymocytes in and above the context of LMO2 over-expression. The combination of an increase in self-renewal induced by LMO2 and loss of the gatekeeper function of self-renewal through p19 Arf loss leads to acceleration in the pathogenesis of LMO2 induced T-cell malignancies. Disclosures: No relevant conflicts of interest to declare.