Deregulated EVI1 Expression Leads to Genomic Instability and G1 Cell Cycle Arrest

Abstract 2431

Insertional activation of the zinc finger transcription factor ecotropic viral integration site 1 (EVI1) by promoter and enhancer elements within the gamma-retroviral vector LTR has led to clonal dominance and malignant transformation in clinical gene therapy of chronic granulomatous disease (cGD). EVI1 has been shown to inhibit stress induced cell death or TGFβ signaling and is essential for embryonic development. Cytogenetic rearrangements leading to the activation of the gene locus are a marker for poor prognosis in myeloid malignancies. Very little is known about its larger splice variant MDS1/EVI1 in normal or malignant hematopoiesis. We aim to systematically analyze the role of deregulated EVI1 and MDS1/EVI1 expression in hematopoiesis to further elucidate the stepwise progression of clonal selection up to malignant transformation.

Lentiviral vector particles encoding for EVI1 (E) or MDS1/EVI1 (ME) and eGFP as marker protein were produced to stably overexpress the transgenes. Transgene expression was verified in myeloid HL60 cells by western blotting and immunofluorescence. Analysis of growth kinetics revealed a 1.5 – 3 fold lower proliferation of ME and E expressing cells as compared to eGFP control vector transduced cells. Additional analysis on cell cycle distribution revealed that in both, E and ME transduced cells, a higher percentage of cells could be detected in the G1/G0 phase of the cell cycle (52.5 ± 1.6% in ME cells and 55.2 ± 1.8% in E cells) compared to untransduced (48.9 ± 1.6%) and control vector cells (47.7 ± 0.3%) (p<0.01). Consequently, a 1.3 – 1.4 fold lower proportion of E and ME cells was observed in the S/G2/M phases compared to control cells. For further analyzing the transgene effect on cell cycle activity, 3 populations with different intensity of transgene expression (negative, intermediate and high eGFP⁺ cells) were isolated. With raising ME expression a 5 fold decrease of cells in sub-G1 phase but a 1.3 fold increase of cells in G1/G0 phase was detected. In the highly EVI1 expressing fraction 91.4% arrested in G1/G0 phase of the cell cycle (50.4% in G1/G0 phase in eGFP⁻ E cells). In line with this, we observed a decrease of eGFP⁺ E and ME transduced human hematopoietic CD34⁺ cells from 8% at day 3 after transduction to 0.5 – 2.5% at day 14, respectively. In contrast, the proportion of eGFP⁺ human primary cells remained stable for the time period analyzed after transduction with the control vector. To investigate E and ME overexpression in vivo, hematopoietic stem and progenitor cells were isolated from murine bone marrow, transduced with the lentiviral vectors (LV) and transplanted into lethally irradiated recipient mice. Although all recipients showed donor cell engraftment, eGFP expression dropped from 47.7 – 49% at 4 to 7 weeks to 0.9 – 4.4% at 40 weeks after transplantation (BMT). Interestingly, EVI1 transplanted mice showed a significantly lower thrombocyte recovery (2.2 fold; p<0.05) within the first 13 weeks after BMT in comparison to untransduced, control vector or ME transplanted mice. We then asked if the cell cycle arrest in G1 is associated with genetic instability, as patients with insertional activation of EVI1 developed a myelodysplastic syndrome with monosomy 7. E and ME transduced human fibroblasts showed a significant higher percentage of cells with abnormal centrosome numbers as compared to controls (p<0.05). In addition, staining of γ-H2AX, an indirect marker for double strand breaks (DSB), in E and ME transduced HL60 cells revealed that EVI1⁺ γ-H2AX⁺ cells were 2-fold enriched in G1 as compared to control vector transduced cells.

In summary, our data show that EVI1 overexpression causes G1 cell cycle arrest of hematopoietic cells that may be caused by genetic instability or inefficient DSB repair. No sign for clonal selection could be detected, neither in vitro nor in vivo, but EVI1 transplanted mice showed a delay in thrombocyte recovery. Systematic investigation of EVI1 and MDS1/EVI1 overexpression in human hematopoietic cells will help us to gain insights into regulatory processes of hematopoietic stem cells and mechanisms leading to dominant clones in gene modified hematopoiesis.