DUSP1/6 Inhibition Reduces Tumor Cells and Activates Immune Response in Chronic Lymphocytic Leukemia

Chronic lymphocytic leukemia (CLL) is characterized by an expansion of monoclonal, mature B cells that carry auto/polyreactive B cell receptors (BCRs). It is one of the most prevalent B cell malignancies in Western countries. Over the past decade, extensive research has led to significant improvement in understanding its pathogenesis, particularly in regard to the BCR signaling pathway. This has led to the development of novel and promising treatment options, however, none of these approaches are curative and relapses are challenging to treat. Until now allogeneic hematopoietic stem cell transplantation remains the only treatment option for durable remission in poor-risk disease. The therapeutic difficulties result from the heterogeneity, the advanced age of the patients, chemoresistance and relapse of the disease. Here, we test a novel strategy for simultaneously targeting the malignant cells and reverting the CLL-suppressed immune response.

Stimulation of the antigen receptors of immune cells readily induces proliferation, differentiation and functional activation. However, strong binding of antigen can also induce negative selection in lymphocytes. CLL B cells constitutively signal through their BCR and thus we set out to investigate whether they rely on the counterbalancing, negative regulation of specific downstream signaling pathways, in particular the mitogen-activated protein kinase (MAPK) pathway. The dual specificity phosphatases 1 and 6 (DUSP1 and DUSP6) dephosphorylate extracellular-signal regulated kinase 1/2 (ERK1/2) and thereby limit ERK1/2 activation. These molecules are frequently downregulated in solid tumors (Khor et al., Int. J. Med.Sci 2013; Okudela et al., Am. J. Pathol. 2009). We therefore set out to analyze the expression levels of DUSP1 and DUSP6 in CLL and found them readily expressed at various levels, comparable to normal B cells. To determine the functional relevance of DUSP1 and DUSP6, we blocked their phosphatase function using the small molecule inhibitor BCI. By treatment with BCI, we induced hyperactivation of the MAPK signaling cascade followed by cell death of the CLL cells. Interestingly, the induction of cell death is specific for CLL cells and does not occur to the same extent in other malignant B cell lymphoma cells or healthy donor-derived B cells. This deleterious effect of BCI was evident in primary patient-derived CLL cells as well as in the CLL-like cell lines MEC-1 and EHEB and in CLL cells derived from the T cell leukemia/lymphoma 1 (TCL1)-driven mouse model. To further investigate the downstream signaling event upon BCI treatment, we conducted a global phosphoproteome analysis. After treatment of primary CLL cells with BCI, the most significant alterations were within the BCR signaling pathway, including hyperphosphorylation of ERK1/2 and followed by a rapid induction of a DNA damage response. These results were validated by immunoblot analysis of human and murine CLL cells and were not detected in BCI-resistant cell lines.

Beside the direct effects of BCI on the CLL cells we set out to investigate effects on other immune cells, directly by BCI and secondary via CLL cells treated with BCI. Indeed, we observed changes in immune cell compartment: BCI-treated CLL patient-derived peripheral blood mononuclear cells (PBMCs) resulted in selective enrichment of cytotoxic T cells. Furthermore, BCI treatment of CLL cells fed with Ovalbumin, in co-culture with or without BMDCs and OT-I cytotoxic CD8 T cells (specifically recognizing the SINFEKL peptide), resulted in the induction of immunogenic cell death of CLL cells. This was evidenced by enhanced antigen-specific T cell proliferation and release of the high mobility group box 1 protein (HMGB1). To investigate indirect effects of BCI, we treated CLL cells with BCI at sublethal doses, then washed the cells and co-cultured these primary CLL cells or CLL cell lines with healthy donor-derived PBMCs. The frequencies of CLL-induced myeloid-derived suppressor cells (MDSCs) as well as regulatory T cells (Tregs) were reduced after co-cultivation of PBMCs with BCI-pretreated CLL cells.

Taken together, our data indicate that negative feedback inhibition reduces CLL content by induction of immunogenic cell death and activates immune cells to target the CLL-induced dysfunction of the immune system. We therefore propose that inhibition of DUSP1/6 is a promising therapeutic approach for CLL.