Investigating Drug Resistance In Leukemia Initiating Cell Populations.

Abstract 1570

Outcomes for childhood leukemia have improved significantly in recent years with remission rates of over 98% and reported cure rates of 80% for standard risk cases. However, 20% of patients relapse due to failure to eradicate the disease. Further improvement in outcomes will require a better understanding of the biology of this malignancy and the mechanisms of drug resistance. Evidence that several leukemia subpopulations can initiate and maintain this disease in xenograft models and that some of these subpopulations are resistant to current therapeutic agents suggests that relapse may arise from these cells. Parthenolide (PTL), a sesquiterpene lactone compound, has been shown to cause apoptosis in malignant cells by inducing oxidative stress and by inhibiting NF-κB mediated cell survival. In this investigation we have assessed the effects of PTL on leukemia subpopulations in a cohort of childhood ALL cases from mixed prognostic subgroups. Cells from 15 B-ALL cases were stained with antibodies against CD34 and CD19, while CD34 and CD7 were used for 7 T-ALL cases. Cells were then sorted based on expression or lack of expression of the antibody combinations. Unsorted cells and the 4 sorted subpopulations from each type of leukemia were treated with 7.5 and 10 μM PTL for 18–24 hours. The effect of PTL on viability was studied by flow cytometry using Annexin V and Propidium iodide. In B-ALL cases, the CD34⁺/CD19^- population was the least affected with 89.1±6.9% cells surviving PTL treatment. This was significantly higher than the unsorted cells and the other sorted populations (<53%; P<0.01). Most of the T-ALL cases (6/7) were classed as high risk by MRD analyses at day 28. Despite this, unsorted T-ALL cells were more responsive to PTL with only 29.7±12.8% surviving treatment. The CD34⁺/CD7^- population was the least affected (59.9±13.3% viable cells). The functional capacity of the PTL treated unsorted cells and sorted populations was also assessed in vivo. NOD/SCID IL2Rγ null (NSG) mice were inoculated with untreated or PTL treated cells and the levels of engraftment after 10 weeks were compared. The results to date indicate that PTL treatment prevented engraftment of unsorted ALL cells. Mean engraftment levels of 65±20% CD45⁺ (range 29–99%) were observed using untreated cells while there was no detectable human cell engraftment with the PTL treated cells. This suggests PTL is more effective on unsorted ALL cells than the data from the short term apoptosis assays indicated. Engraftment was achieved using CD34^- cells from 3 cases (73±29%, range 40–96%). However, no engraftment was observed when CD34^- cells were treated with PTL. In contrast, the levels of engraftment observed with PTL treated CD34⁺/CD19^- B-ALL cells were similar to or greater than those observed with the untreated counterparts (95±8% and 64±9% CD45⁺ respectively, P≤0.07). The levels of engraftment observed with CD34⁺/CD7^- T-ALL cells were reduced with PTL treatment from 67±21% to 12±9% CD45+ (P≤0.03) but not eliminated. Subsequently, we investigated the mechanisms for this apparent resistance to PTL in the primitive cell populations. PTL has been associated with induction of oxidative stress, activation of p53 and inhibition of NF-κB in AML and CLL. We used confocal microscopy to investigate whether NF-κB is constitutively expressed in ALL cases and to evaluate the effect of PTL on the phosphorylation of NF-κB. Three B-ALL and 3 T-ALL cases, where the unsorted populations had been affected by PTL while the respective CD34⁺/CD19^- and CD34⁺/CD7^- populations were more resistant, were investigated. Cells were stained with anti-phospho-p65 polyclonal antibody and Alexa fluor 488. NF-κB was constitutively activated in all cases. There was evidence of decreased phosphorylation in unsorted PTL treated cells indicating inhibition of NF-κB. However, in the phenotypically primitive cells there was no difference in the phosphorylation levels compared to untreated cells or phosphorylation was increased. This suggests NF-κB was not inhibited, which could explain the observed resistance of these leukemia populations to PTL. These data demonstrate that some leukemia initiating cell populations in childhood B-ALL and T-ALL are resistant to PTL. A more thorough understanding of these leukemia initiating cell populations and their mechanisms of resistance will be required for the development of more effective therapies.