Impaired apoptosis of megakaryocytes and bone marrow mononuclear cells in essential thrombocythemia: correlation with JAK2V617F mutational status and cytoreductive therapy

The impaired balance between the ability for cells to proliferate and undergo apoptosis may be crucial for the development of malignant hematological disorders including ET [1]. To the best of our knowledge, this is the first report showing inhibited apoptosis of ET MKCs in bone marrow as assessed by flow cytometry. We observed a markedly lower percentage of MKCs positive for both Ann-V staining and active Cas-3 in patients with ET naïve to therapy as compared to healthy volunteers. This is contrary to the results presented by Florena et al. [8] who found a similar percentage of Cas-3-expressing MKCs in ET versus controls, however, this study was based just on an immunohistochemical evaluation of bone marrow samples. This is an important distinction, because the flow cytometry method is a much more sensitive method of detecting a relatively small population of characteristic cells. Interestingly, cases with a low apoptotic fraction of Cas-3 (<5%) were associated with the occurrence of thrombotic events, which are still the main cause of morbidity and mortality in ET [8].

The actual fraction of bone marrow mononuclear cells committed to apoptosis was assessed by the Ann-V staining, with estimation of effector Cas-3 activation. Our study showed a markedly lower percentage of apoptotic MKCs and BMMCs in ET patients than in controls. Moreover, the Bax/Bcl-2 index was much lower in ET patients, comparing to the control.

Activation of caspases and other events during the process of apoptosis is regulated on multiple levels by Bcl-2 family members acting on the intrinsic pathway of apoptosis [9‐11]. The Bcl-2 family consists of several proteins, with the most important members being the pro-apoptotic Bax and anti-apoptotic Bcl-2 proteins. The inhibited Bax/Bcl-2 ratio is one of the most sensitive markers of the anti-apoptotic cell profile. Moreover, it has been previously demonstrated that transcription of the anti-apoptotic Bcl-2 protein is regulated by the STAT3/5 signaling pathway and that Bcl-2 is overexpressed in erythroid cells from patients with polycythemia vera [12]. It has also been discovered that MKCs from patients with ET as opposed to primary myelofibrosis are characterized by elevated expression of Bcl-XL and markedly lower expression of pro-apoptotic Bax (8). Additionally, Zhang et al. [13] observed strong expression of Bcl-XL in immature MKCs and decreased expression in mature MKCs. They concluded that increased Bcl-XL expression may be essential to MKC maturation. In the present study, we found downregulation of Bax expression in bone marrow BMMC in patients with ET as compared to controls. This, together with the higher expression of Bcl-2 protein and markedly lower Bax/Bcl-2 index, confirms the anti-apoptotic profile of BMMCs in ET patients.

In 2005, a somatic activation mutation in Janus Kinase 2, occurring in approximately 50% of ET patients, was identified [14, 15]. The JAK2V617F gene mutation results in constitutive signaling through the JAK2 tyrosine kinase, leading to increased cellular proliferation. On the other hand, it has been speculated that the constitutive activation of JAK2 could protect cells from death receptor-induced apoptosis by different means, in this number by upregulation of the anti-apoptotic gene c-FLIP [16‐19]. In our study, the cohort of JAK2V617F positive patients demonstrated a markedly higher activation of Cas-3, as well as higher Bax expression, and also a tendency toward increased apoptosis measured by Ann-V method as compared to JAK2 negative cases. Also, the Bax/Bcl-2 ratio was shifted toward more pro-apoptotic in JAK2+ patients. Our observation can partially explain the phenomenon of increased efficacy of cytoreductive drugs in patients positive for JAK2 mutation [20].

It has been stated repeatedly that not all patients with ET require therapy, unless certain risk factors (age >60 years, prior thrombotic or hemorrhagic event, cardiovascular diseases, possibly JAK2V617F mutation and elevated leukocytosis) are present [21‐25]. HU and ANA are most widely used as first-line cytoreductive therapy in high-risk ET patients. HU has emerged as the treatment of choice in two randomized clinical trials [24, 25]. HU is an antimetabolite, mediating its anti-myeloproliferative effects via the inhibition of ribonucleoside diphosphate reductase needed for synthesis and repair of DNA [26]. Beyond its broad myelosuppressive activity, HU can also exert an antithrombotic activity by affecting polymorphonuclear leukocyte function and their interactions with platelets [27‐29]. Recent studies have shown that HU can induce apoptosis in many types of cells: for example, endothelial cells, human mesenchymal stem cells and mouse embryonic stem cells [30, 31]. HU appears to promote cell death by regulating the expression levels of Bcl-2 and the tumor suppressor p53 protein [32, 33]. In contrast, ANA is a non-cytostatic imidazoquinazoline derivative, which reversibly disrupts MKC maturation (e.g., ploidy, size, cytoplasmic maturation) in a dose-dependent fashion by influencing the post-mitotic phases of MKC development [34]. There is no clear-cut evidence to suggest a role is played by ANA in reducing the total number of MKCs or stimulating apoptosis [35, 36]. In the present study, different markers of apoptosis were examined in 21 patients with ET treated with HU or ANA who were in complete or partial remission according to European Leukemia Net criteria [37]. Our analysis showed that, in relation to patients naïve to therapy, both ANA and HU downregulated concentration of anti-apoptotic Bcl-2 while HU, surprisingly, downregulated the concentration of pro-apoptotic Bax. However, it has to be stressed that the overall Bax/Bcl-2 ratio in patients on ANA or HU therapy was more pro-apoptotic than in patients at the time of diagnosis. There were no statistical differences in other studied parameters although there was a tendency toward a higher percentage of Ann-V+ MKc and Cas-3+ BMMCs in the HU group.

In conclusion, we found evidence for caspase-dependent MKC and BMMC apoptosis inhibition with upregulation of Bcl-2 protein and concurrent downregulation of Bax in ET patients in relation to healthy volunteers. Patients positive for the JAK2V617F mutation had a tendency toward increased apoptosis of MKCs and BMMCs in bone marrow in relation to JAK2V617F negative cases, which could contribute to the better therapy outcome observed in this group of patients. Patients treated with HU showed a slightly higher pro-apoptotic Bax/Bcl-2 index than patients on ANA therapy, although no significant differences in apoptosis markers were observed between these treatment modalities.