DYRK1A Is a Therapeutic Target in B-ALL in Children with Down Syndrome

Children with Down syndrome have a spectrum of associated disorders including a 20-fold increased incidence of B-cell acute lymphoblastic leukemia (DS-ALL). Although a number of genetic alterations have been found in this ALL subtype, such as activating mutations in JAK2 and overexpression of CRLF2, the mechanisms by which trisomy 21 promotes the leukemia are largely unknown. Previous studies have implicated chromosome 21 genes HMGN1 and DYRK1A in both malignant and normal lymphopoiesis.

DYRK1A is a member of the dual-specificity tyrosine phosphorylation-regulated kinase family that has been well studied in non-hematopoietic tissues. Its targets include proteins that regulate multiple pathways including cell signaling, cell cycle, and brain development. We have previously shown that DYRK1A is a megakaryoblastic leukemia-promoting gene through its negative regulation of NFAT transcription factors. Furthermore, in studies with a conditional Dyrk1a knock-out mouse, we found that the kinase is required for lymphoid, but not myeloid cell development. In developing lymphocytes, Dyrk1a regulates the cell cycle by destabilizing cyclin D3. Consequently, loss of Dyrk1a resulted in the failure of these cells to switch from a proliferative to quiescent phase for subsequent maturation (Thompson et al. J. Exp. Med. 2015 212:953-70). Despite this deficiency in exiting the cell cycle, Dyrk1a-deficient lymphocytes also exhibit impaired proliferation before undergoing apoptosis. These data reveal a critical role for DYRK1A in lymphopoiesis and suggest that it may be a target for therapeutic intervention.

We assayed the activity of the highly selective and potent DYRK1 inhibitor, EHT 1610, in multiple ALL cell lines. EHT 1610 inhibited the growth of Jurkat and MHH-CALL-4 cells with EC50s of 0.83mM and 0.49mM, respectively. Next, we treated primary human ALL blasts with EHT 1610 and the less selective DYRK1A inhibitor harmine. Growth of 16 out of 30 specimens, which included DS-ALL, pre-B ALL, and T-ALL, was sensitive to DYRK1A inhibition at doses between 0.5 and 10mM. Of note, growth of 9 of the 11 of the DS-ALL samples was inhibited by EHT 1610. This result indicates that the increased dosage of DYRK1A in DS samples sensitizes the cells to DYRK1A inhibition.

To further study the contributions of DYRK1A to normal and malignant lymphopoiesis, we performed phosphoproteomic analysis on primary murine pre-B cells treated with EHT 1610. After 2 hours of EHT 1610 treatment, the cells were collected and analyzed for changes in the phosphoproteome. Phosphorylation of 36 proteins was significantly altered. Bioinformatics analysis led to the identification of a number of notable pathways that appear to be regulated by DYRK1A including cell cycle, cell division and mitosis, RNA metabolism, and JAK-STAT signaling. Differentially phosphorylated proteins included geminin, which is important in cell division and whose loss enhances megakaryopoiesis, and POLR2M, which is intriguing because DYRK1A phosphorylates the CTD of RNA Pol II and binds chromatin at specific sites in glioblastoma cells. Another interesting target is STAT3, which is phosphorylated by DYRK1A on Ser727, a residue whose phosphorylation is required for maximal STAT3 activation. Treatment of murine pre-B cells with EHT 1610 significantly reduced the level of phosphorylation of Ser727 and Tyr705, suggesting that DYRK1A may provide a priming event for STAT3 activation similar to its priming effect on GSK3b phosphorylation. Consistent with a role for JAK/STAT signaling and STAT3 activity, B-ALL cells were highly sensitive to ruxolitinib therapy. Taken together, our study suggests that DYRK1A is a therapeutic target in DS-ALL and likely functions in part by enhancing JAK/STAT signaling.