Tax impairs DNA replication forks and increases DNA breaks in specific oncogenic genome regions

Abstract

Background: Human T-cell leukemia virus type 1 (HTLV-I) is a human retrovirus associated with adult T-cell leukemia (ATL), an aggressive CD4 T-cell proliferative disease with dismal prognosis. The long latency preceding the development of the disease and the low incidence suggests that the virus itself is not sufficient for transformation and that genetic defects are required to create a permissive environment for leukemia. In fact, ATL cells are characterized by profound genetic modifications including structural and numerical chromosome alterations.

Results: In this study we used molecular combing techniques to study the effect of the oncoprotein Tax on DNA replication. We found that replication forks have difficulties replicating complex DNA, fork progression is slower, and they pause or stall more frequently in the presence of Tax expression. Our results also show that Tax-associated replication defects are partially compensated by an increase in the firing of back-up origins. Consistent with these effects of Tax on DNA replication, an increase in double strand DNA breaks (DDSB) was seen in Tax expressing cells. Tax-mediated increases in DDSBs were associated with the ability of Tax to activate NF-kB and to stimulate intracellular nitric oxide production. We also demonstrated a reduced expression of human translesion synthesis (TLS) DNA polymerases Pol-H and Pol-K in HTLV-I-transformed T cells and ATL cells. This was associated with an increase in DNA breaks induced by Tax at specific genome regions, such as the c-Myc and the Bcl-2 major breakpoints. Consistent with the notion that the non-homologous end joining (NHEJ) pathway is hyperactive in HTLV-I-transformed cells, we found that inhibition of the NHEJ pathway induces significant killing of HTLV-I transformed cells and patient-derived leukemic ATL cells.

Conclusion: Our results suggest that, replication problems increase genetic instability in HTLV-I-transformed cells. As a result, abuse of NHEJ and a defective homologous repair (HR) DNA repair pathway can be targeted as a new therapeutic approach for the treatment of adult T-cell leukemia.

Figure 1

Replication forks are slower in the presence of the HTLV-I Tax oncoprotein. (A) HTLV-I Tax expression was confirmed after induction with Cadmium chloride (CdCl) in Tax inducible Jurkat, JPX9 cells. (B). Induction of Tax expression was associated with an increase in DNA breaks detected by FACS analyses of p-H2AX. (C) Schematic representation of a single molecule analysis of DNA replication by molecular combing. (D) MT4, JPX9 and JPX9 Tax+ cells were pulse-labelled for 15 min with IdU then 15 min with CldU, and fibers were stretched by DNA combing. Blue: DNA. Red: IdU. Green: CldU. An example of a fiber representative for each condition is shown. (E) Replication fork speed analysis. The lengths of CldU tracks (ImageJ) were measured and divided by the length of the pulse (15 min). (F) Box and whisker plots: Box: 25–75 percentile range and Whiskers: 5–95 percentile range. More than 300 values coming from 3 independent experiments were compiled. Statistical analysis was performed using the non-parametric Mann–Whitney rank sum test. ns: not significant. *: p < 0.05; **: p < 0.01; ****: p < 0.0001.

Figure 2

Tax blocks DNA replication fork progression. (A-B) Box and whisker plots. Box: 25–75 percentile range. Whiskers: 5–95 percentile range. More than 300 values coming from 3 independent experiments were compiled. Statistical analysis was performed using the non-parametric Mann–Whitney rank sum test. ns: not significant. *: p < 0.05; **: p < 0.01; ****: p < 0.0001. (A): Measurement of acceleration and deceleration of replication forks, IdU/CldU ratio. (B): Analyses of the center-to-center distance which reflects the distance between two replication origins and is indicative of origin firing frequency.

Figure 3

Tax expression induces DDSBs by stimulating intracellular NO. HR activity (A) is reduced in 289 cells when compared to BRCA1 cells. Tax triggers H2AX phosphorylation in BRCA1-deficient cell. The 289 and BRCA1 cells were infected with GFP or Tax-IRES GFP-expressing virus during 48 hours. 293T cells were transfected with 1.5 μg DR-GFP and 0.5 μg pSCE I expression vector or empty vector. The percentage of GFP+ cells was estimated by FACS. (B) Tax triggers H2AX phosphorylation in BRCA1-deficient cell. The 289 and BRCA1 cells were infected with GFP or Tax-IRES GFP-expressing virus during 48 hours. The cells were analyzed (B) by Flow cytometry and (C) by western blot. (D-F) Tax and M47 Tax mutant stimulate NO production and trigger H2AX phosphorylation. 293T cells were transfected with 2 μg of empty vector (Ø), Tax, or Tax mutants (M47 and G148V). At 72 h, cell culture supernatant was harvested and tested for NO production (D), and cells were further analyzed by FACS for ɣ-H2AX quantification (E) and by western blot for Tax expression (F). (G-I) The NOS inhibitor L-NMMA blocks NO production and DDSBs induced by Tax. 293T cells were transfected with the empty vector (Ø), Tax, or Tax mutants (M47 and G148V) during 24 h and treated with DMSO or 30 μM L-NMMA for 48 h. Cell culture supernatant was harvested and tested for NO production (G). Cells were analyzed by FACS for ɣ-H2AX quantification (H) and western blot to confirm Tax expression (I).

Figure 4

Reduced expression of Pol-H (A) and Pol-K (B) in HTLV-1 cell. Real-time PCR was performed in duplicate and samples were normalized to GAPDH expression. Fold change was calculated by comparing values with T cells (sorted from healthy donor) and/or Jurkat. (C) Viral tax mRNA expression was demonstrated by RT-PCR. Tax protein expression was detected by immunoblots in HTLV-I-transformed cell lines in vitro.

Figure 5

Tax expression increases DNA breaks and rearrangement in a human c-MYC promoter region rich in non-B-DNA forming sequences. (A) Establishment of stable cell lines containing a non-DNA sequence control from the GAPDH gene (pULCtrl), a c-MYC promoter region (pUMyc), and a BCL-2 gene major break region (pUMBR) as described in the Material and Methods. (B) pUCONT, pUMYC and pUMBR cell lines were infected with HRCMV or HRCMVTax virus for 48 h and analyzed by Flow cytometry to quantify ɣ-H2AX. Scatter plots are represented with ɣ-H2AX intensity on the y axis and propidium iodide intensity on the x axis. The cell populations used for the measurement of the ɣ-H2AX-positive cells were enclosed by rectangles. (C) Tax induces a small scale deletion in the c-MYC promoter. After 48 h infection by HRCMV or HRCMVTax, the genomic DNA was isolated from the pUMyc cells and subjected to PCR to amplify c-MYC promoter region. The PCR product was cloned into pGEM-T Easy vector and the inserts were verified by DNA sequencing.

Figure 6

NU7026 inhibits growth and induces apoptosis in HTLV-I cell lines, but not RI-1. (A) MT4 cells were treated with increasing concentrations of NU7026. After 3 days of treatment, cells were stained with XTT and measurements were quantified at OD: 450 nm. The results represent the means of 2 separate experiments + SD. (B) MT4 cells were treated with NU7026 (10 μM). After 3 and 5 days of treatment, cells were stained with Annexin V and PI, and analyzed by flow cytometry. (C) MT4 cells were treated with an increasing concentration of RI-1. After 3 days of treatment, cells were stained with XTT and measurements were quantified at OD: 450 nm. The results represent the means of 2 separate experiments + SD. (D) MT4 cells were treated with RI-1 (20 μM). After 3 and 5 days of treatment, cells were stained with Annexin V and PI, and analyzed by flow cytometry. (E) ATL cell lines were treated with NU7026 (10 μM) for 3 days. Then, cells were stained with XTT and measurements were quantified at OD: 450 nm. The results represent the means of 2 separate experiments + SD.