ETNK1 Mutations in Atypical Chronic Myeloid Leukemia Induce a Mutator Phenotype That Can be Reverted with Phosphoethanolamine

ETNK1 kinase is responsible for the phosphorylation of ethanolamine to phosphoethanolamine (P-Et) (Kennedy, 1956, J Biol Chem). Recurrent somatic mutations occurring on ETNK1 were identified in about 13% of patients affected by atypical chronic myeloid leukemia (aCML), in 3-14% of chronic myelomonocytic leukemia (CMML), and in 20% of systemic mastocytosis (SM) patients with eosinophilia (Gambacorti-Passerini, 2015, Blood; Lasho, 2015, Blood Cancer J). ETNK1 mutations, encoding for H243Y, N244S/T/K, and G245V/A amino acid substitutions, cluster in a very narrow region of the ETNK1 catalytic domain and cause an impairment of ETNK1 enzymatic activity leading to a significant decrease in the intracellular concentration of P-Et (Gambacorti-Passerini, 2015, Blood). Despite this evidence, however, their oncogenic role remained largely unexplained. Here, we investigated the specific role of these mutations by using cellular CRISPR/Cas9 and ETNK1 overexpression models as well as aCML patients' samples. We showed that mutated ETNK1 causes a significant increase in mitochondrial activity (1.87 fold increase compared to WT; p=0.0002) and in ROS production (2.05 fold increase compared to WT; p<0.0001). Since ROS are responsible for DNA oxidative damage, we firstly generated ChIP-Seq data for ETNK1 mutated cells using an antibody raised against the oxoguanine (oxoG) and we compared oxoG signal against the wild-type cell line, to assess whether ETNK1 mutations could cause accumulation of DNA lesions. This analysis revealed a significant increase in oxoG in mutated cells, compared to WT (p=0.018). Then, we investigated if these lesions were driving the onset of a mutator phenotype by applying the 6-thioguanine (6-TG) resistance assays to our cell models, showing that in the mutated cells there was a 5.4 fold increase in colony number compared to the WT line (p<0.0001). Moreover, we investigated if the ROS-mediated genotoxic insult operating in ETNK1-mutated lines could be also associated with an increase in DNA double-strand breaks. Comparison of ETNK1-N244S and ETNK1-WT lines revealed a sharp increase in the number of γH2AX foci (2.52 fold increase; p=0.0002) in the former. At this point, we hypothesized that the decreased P-Et concentration in ETNK1-mutated cells could be responsible for the increased mitochondrial activity. ETNK1-N244S cells treated with P-Et showed a complete restoration of the normal mitochondrial membrane potential and generation of ROS. Moreover, the mutator phenotype was reverted by P-Et treatment, supporting the hypothesis of a direct involvement of P-Et in the induction of DNA damage. To dissect the mechanism by which P-Et intracellular levels were able to control mitochondria activity, we isolated the mitochondrial oxidative phosphorylation complexes I to IV and we measured the activity of each complex in absence/presence of increasing P-Et concentrations. This approach revealed a profound, dose-dependent decrease in redox activity for mitochondrial complex II (P-Et 10μM: 1.80 fold decrease; p=0.0012; P-Et 20μM: 7.40 fold decrease; p<0.0001; P-Et 50μM: 28.85 fold decrease; p<0.0001) and virtually no effect for the other three complexes, indicating that P-Et controls mitochondria potential through direct inhibition of complex II. To gain insight into the specific mechanism by which P-Et could repress complex II, we analyzed its activity in competition assays in presence of both P-Et and increasing concentration of succinate, the endogenous substrate of succinate dehydrogenase (SDH), showing that succinate supplementation was able to restore the normal SDH activity starting from 50µM. Taken globally, these data suggest that P-Et acts as a competitive inhibitor of succinate for SDH activity. In line with these data, automatic docking of P-Et inside the SDH catalytic domain confirmed that P-Et can occupy the succinate binding site in an energetically favorable conformation, mimicking succinate. In conclusion, the reduced activity of mutant ETNK1 leads to the accumulation of new mutations through the reduced competition of P-Et with succinate, increased mitochondrial activity and ROS production. This mechanism can be blocked, at least in vitro, by P-Et supplementation, suppressing the accumulation of new mutations mediated by the ETNK1-dependent mutator phenotype. In vivo studies will address the therapeutic potential of P-Et.