In aim 1, we showed that CD20 is one of the key proteins involved in BCR-mediated CLL cell activation. This also has important therapeutic implications since CD20 is used as a therapeutic target for 20 years, but its regulation and function remain largely unknown. We have described for the first time that T-cell interactions determine CD20 levels on malignant B cells, and such levels impact IL4 signaling in a feed-forward regulatory loop. We have also revealed that drugs targeting PI3K kinase will lead to reduced CD20 levels and, therefore, are less suitable for combination with anti-CD20 antibodies. In aim 1, we also described a mechanism that allows malignant B lymphocytes to survive therapy with so-called BCR inhibitors (ibrutinib/idelalisib). We have described two completely novel mechanisms. The first one involves the induction of GAB1 protein and subsequent increase in tonic pro-survival signaling mediated by AKT kinase. The second one involved the induction of the mTORC2 assembly protein Rictor via the transcription factor FoxO1, which activates Akt irrespective of cell surface receptors/BCR-associated kinases. We have further shown that GAB1 or FoxO1 inhibitors can kill leukemic cells and could be used alone or in combination with BTK or PI3K inhibitors.
In aim 2, we described the first example of the role of short non-coding RNAs, called microRNAs , in regulating CLL-T cell interactions. The described mechanism provides a mechanistic way for B cell to synchronously activate both BCR and CD40, which is required to enter B cells into the cell cycle and multiplication. We have proved the regulation of CD40 signaling by miR-29 in CLL and also in a related disease, follicular lymphoma. We also showed that the BCR inhibitors ibrutinib or idelalisib affect this axis, and the propensity of CD40 signaling also changes during the transformation of lymphomas into more aggressive disease. These data also explain why BCR inhibitors have such an unexpected and dramatic effect on the proliferation of malignant B cells.
In aim 3, we integrated our data on microenvironmental signaling and developed a novel CLL co-culture model that allows us to mimic CLL-T cell interactions and triggers robust CLL cell proliferation. We utilized the analyses of microenvironmental interactions to define what signals are missing in in vitro and in immunodeficient mice, and preclude the proliferation/engraftment of leukemic CLL cells. We genetically engineered supportive cells to express three T-cell factors. According to our data, this model is the most robust model for studying CLL cell proliferation, which is otherwise impossible since CLL cells in vitro do not spontaneously proliferate. Moreover, in line with our plan, we managed to transfer this model on 3D scaffold to a mouse model for patient-derived xenograft (PDX). This allows stable engraftment of primary CLL cells and studies of CLL biology. The co-culture model allowed us to also reveal for the first time that pan-RAF inhibitors are able to block CLL cells proliferation.
