MLL-Rearranged Infant ALL, but Not Standard-Risk ALL, Induces Inflammasome Signaling and Immune Activation

T cell-based therapy for infants with MLL-rearranged B-ALL (MLL-ALL) represents a promising approach in this high-risk, poor-prognosis subset. Factors in the immune microenvironment that influence T-cell function, however, are poorly understood in MLL-ALL. We have observed complete inhibition of healthy donor T-cell proliferation by ALL cells from MLL-ALL, but not standard risk ALL (SR-ALL) patients. T-cell function is restored in the presence of bone marrow macrophages. In this study, we hypothesized that MLL-ALL cells selectively induce the inflammasome- a cytokine signaling complex formed by macrophages in response to damage-associated molecular patterns (DAMPs) released by leukemic cells. In the canonical inflammasome pathway, DAMPs engage macrophage Toll-like receptors, which activate caspase-1 and drive production of T-cell activating cytokines such as IL1-β.

First, we determined whether ALL cells obtained from MLL-ALL vs. SR-ALL patients could produce high mobility group box 1 (HGMB1), a prototypical DAMP associated with tumor cell stress and macrophage PRR activation, particularly after anthracycline therapy. CD19+ cells were isolated from the diagnostic bone marrow specimens of patients with MLL-ALL (n=1), SR-ALL (n=4), and healthy donor PBMC (n=2) and treated with either control media or 2uM doxorubicin. Prior to treatment both MLL-ALL and SR-ALL, but not healthy donor cells, expressed HMGB1 protein as measured by Western immunoblot (29 kDa, GADPH internal control). Following doxorubicin treatment, HMGB1 protein was increased in MLL-ALL but not SR-ALL. Soluble HMGB1, measured by ELISA, was significantly increased in MLL-ALL (12.6 + 3.7 ng/ml untreated vs. 51.4 + 1.9 treated) compared to SR-ALL cells (5.0 + 0.8 ng/ml untreated, 4.3 + 1.5 treated, p<0.05). Soluble HMGB1 was not detectable in healthy donor cells before or after treatment.

Next, we determined whether the increased HMGB1 secretion observed in MLL-ALL vs. SR-ALL cells was associated with macrophage caspase-1 activation. Control media and 2 uM doxorubicin treated-and-washed MLL-ALL, SR-ALL and healthy control CD19+ cells were cultured alone or 1:1 with allogeneic primary human macrophages for 24 hours. The active isoform of caspase-1 (20 kD tetramer) was detectable by immunoblot only in macrophages exposed to doxorubicin-treated MLL-ALL. Mean levels of IL1-β, measured by ELISA, were significantly increased in cocultures of doxorubicin-treated MLL-ALL + macrophages (1255 pg/ml + 322) compared to untreated MLL-ALL (820 pg/ml + 183, p=<0.05) and doxorubicin-treated SR-ALL + macrophages (494 pg/ml + 85 p<0.01). Macrophages cultured alone did not produce significant levels of IL1-β before or after doxorubicin treatment.

Finally, we wished to determine whether expression of Toll-like receptor 4, a surface receptor for HMGB1 known to signal caspase-1, was increased on macrophages exposed to doxorubicin-treated MLL-ALL vs. SR-ALL. Folowing coculture with untreated and doxorubicin treated MLL-ALL and SR-ALL cells, macrophage TLR4 mRNA and surface expression was quantified by qPCR and flow cytometry, respectively. Both mRNA and surface expression of TLR4 was high at baseline, and did not change significantly after coculture. Unexpectedly, however, the MLL-ALL cells themselves displayed a significant increase in TLR4 mRNA (20-fold compared to endogenous beta-glucuronidase, p<0.0001; 12-fold compared to untreated MLL-ALL, p<0.05 by ΔΔCt) and surface expression (median fluorescence intensity 1005 + 89 vs. 367 + 32 respectively, p<0.05) following doxorubicin treatment. This phenomenon was not observed in SR-ALL or healthy CD19+ cells and is consistent with other tumor models of autocrine immune activation by HMGB1.

Taken together, these data indicate that primary MLL-ALL cells, but not SR-ALL cells, are capable of inducing a caspase-1 associated inflammasome that enhances both innate and T-cell immune activation following anthracycline therapy. Studies are currently underway with additional primary MLL-ALL samples to determine how strategic augmentation of this pathway may impact the anti-leukemia killing capacity of therapeutic T-cells.