Tolerance of Whole Genome Duplication in Megakaryocytes, and Consequences for Their Genome Integrity

Introduction: Megakaryocytes (MKs) are unique cells that acquire multiple copies of their nuclear genome via endomitosis - an unusual variant of the cell cycle in which failure of late cytokinesis leads to successive rounds of whole genome duplication (WGD). WGD+ cells are genetically unstable. To protect against this, in most human cells, centrosome accumulation triggers activation of a cell apparatus called the PIDDosome. This degrades MDM2, causing p53-mediated cell cycle arrest and apoptosis. How MKs bypass this checkpoint, and the consequences of WGD for MK genome integrity are poorly understood. Of note, MK ploidy varies over human ontogeny and in disease states. While adult MKs are typically 16N, fetal MKs are 2-4N, and low ploidy MKs occur in blood cancers eg myelofibrosis and certain inherited thrombocytopenias. Crucially, WGD is also a common feature of cancer genomes, where it is a catalytic event for cancer evolution. 1/3 cases occur with an intact p53 pathway, similar to MKs. Therefore, understanding the tolerance mechanisms and consequences of WGD in MKs has broad implications for human health.

Methods: Cell cycle analysis, ploidy (Hoechst), single cell ATAC & RNAseq (SmartSeq-3), and DNA damage was functionally assessed in ex vivo isolated and in vitro differentiated MKs from adult bone marrow (BM), WT and p53 KO murine BM and healthy human cord blood (CB).

Results: Significantly more γH2AX foci (indicative of double stranded DNA breaks) were found in unperturbed, healthy adult BM polyploid MKs than diploid MKs (% MKs with > 5 γH2AX foci, Diploid = 0.85%, Polyploid = 14.6%, p = 0.03, n = >150 cells). Polyploid MKs also showed impaired resolution of γH2AX foci after low-dose irradiation. These data confirm that WGD is detrimental for MK genome integrity.

To interrogate this, we analysed >1400 highly purified human MKs ranging 2-16N (n=5 healthy adult donors). Single cell analysis is vital to accurately distinguish MKs from non-MK cells and enable two important corrections – removing doublets by genetic demultiplexing and predicting cell cycle phase to correct for G2M cells ‘masquerading’ as a higher ploidy class. We confirmed that, in contrast to most other human cells, MKs continually progress through cell cycle, even after WGD. While genes encoding essential nuclear proteins increased with ploidy, DNA repair pathway genes were insufficiently scaled with nuclear size, potentially explaining the impaired γH2AX clearance observed.

We predicted that p53 activation may restrain MK ploidy. Surprisingly, BM MKs from p53KO mice showed no difference in ploidy vs WT MKs, but mature WT MKs showed highly efficient p53-mediated cell cycle arrest upon irradiation. Together, this indicated that p53 is proficient in MKs, but does not respond to WGD.

We next explored how MKs might uncouple WGD sensing from p53 activation. ANKRD26 activates the PIDDosome upon centrosome amplification, and its expression is downregulated during MK maturation. To test whether this explains the lack of p53 activation in response to WGD, we differentiated MKs from CD34+ HSPCs of patients with ANKRD26-Related Thrombocytopenia (ANKRD26-THC), in whom genetic variants prevent RUNX1-mediated transcriptional repression of ANKRD26. ANKRD26-THC MKs were low ploidy, comparable to CB MKs (% 8N+ Cells, Adult: 10.95%, ANKRD26-THC: 5.62%, CB: 2.43%, n=3-5), and showed strikingly increased p53 activation by imaging. Ongoing interrogation of the molecular regulation of ploidy in healthy adult, CB and ANKRD26-THC MKs will delineate the role of the centrosome-p53 axis across ontogeny/disease states.

Conclusions: To our knowledge, this study presents the first detailed examination of MK nuclear dynamics and genome integrity. We demonstrate a previously unappreciated accumulation of DNA damage with ploidy, and show that transcriptional repression of ANKRD26 uncouples p53 responses to WGD from DNA damage response, while enforced ANKRD26 expression has profound effects on MK ploidy. Differential PIDDosome activity may also explain the physiological variation in MK ploidy across human ontogeny and in blood disorders. More broadly, hijacking these mechanisms could also provide a mechanism by which p53-intact cancer genomes tolerate WGD, indicating a potential therapeutic vulnerability. This study highlights how interrogating unusual aspects of MK biology is broadly relevant to understanding human physiology and disease states.