NSC 19630 inhibitor induces S-phase cell cycle arrest
HTLV-1-derived cell lines and Tax-expressing cells display impaired DNA replication and repair, leading us to hypothesize that these cells may be sensitive to treatment with a small helicase inhibitor. In order to determine if the small inhibitor NSC 19630 affects cellular proliferation, we exposed in vitro HTLV-1-transformed cell lines (MT4, C8166, and C91PL) and patient-derived ATLL cell lines (ED) to 3 μM of NSC 19630 or DMSO control for 48 h. Cells were stained with propidium iodide and DNA content was analyzed by FACS. Consistent with the fact that WRN helicases are required to unwind double-stranded DNA to single-stranded DNA during DNA replication [
43], NSC 19630 treatment showed significant accumulation of cells in the S-phase when compared with DMSO-exposed cells (Fig.
1a,
b). Previous studies demonstrated that cells expressing a WRN-specific shRNA displayed a reduction in cellular growth [
44]. In fact, WRN-depleted human fibroblasts show a marked delay in completing the cell cycle by spending more time in late S- and/or G2-phases of the cell cycle [
45]. Consistent with these observations, perturbation of cell cycle progression was noted in HTLV-1-transformed and ATL-derived cell lines (Fig.
1a,
b). We included Western blot of the Tax viral protein in cellular lysates derived from MT4, C8166, C91PL, and ED (Fig.
1c). As previously reported, our analysis identified ED as Tax-negative and MT4, C8166, and C91PL as Tax-positive cell lines. [
23,
46]. Our analysis shows that NSC 19630 induces perturbation of cell cycle progression in both Tax-negative and Tax-positive cells.
The expression of cell cycle progression regulatory proteins was studied by Western blot in ED cells exposed to 3 μM of WRN inhibitor for 72 h. We compared the protein level of cyclins D1, E, A, and B1 in ED cells treated with NSC 19630 versus DMSO-treated controls (Fig.
1d). Cyclin E has a critical role in the control of the G1- and S-phase transitions and in the initiation of DNA replication [
47]. Cyclin D1 levels vary during the cell cycle, with an elevated level of cyclin D1 maintained through G1-phase and required for the initiation of S-phase, while levels are reduced to allow DNA synthesis in S-phase [
48]. However, increased levels of cyclin D1 are required to exit S-phase [
48]. Similarly, cyclin A is accumulated during S-phase and is degraded before metaphase, while cyclin B1 is accumulated during the G2/M-phase [
49]. Treatment of ATL cells with the WRN inhibitor NSC 19630 was associated with a decrease in the expression of cyclin D1, which may prevent treated cells from S-phase exit and result in accumulation of cells in S-phase (Fig.
1d). While there was no significant change in cyclin A expression in NSC 19630-treated cells, expression of cyclin E and cyclin B1 was significantly reduced (Fig.
1d). We believe that a relative decreased population in G2/M as a result of S-phase arrest accounts for the decrease in cyclin B1.
In addition, we decided to include Western blot of cyclins D1, E, A, and B1 in Tax-expressing cell line lysate C8166 extracted from cells exposed to NSC 19630 compared to DMSO-treated controls. Our analysis shows a reduction of cyclins E, D1, and B1 in Tax-positive cells (Fig.
1d), suggesting that cells arrested during the S-phase of the cell cycle.
Overall, our data revealed a profound alteration of the cyclin expression profile in ATL cells exposed to the WRN helicase inhibitor consistent with S-phase arrest.
NSC 19630 inhibits cellular proliferation and induces apoptosis in HTLV-1-transformed and patient-derived cells
Recent evidence shows that treatment with the WRN inhibitor NSC 19630 significantly affects the cellular growth of different leukemia cell lines [
30]. We next exposed HTLV-1-transformed C91PL to increasing logarithmic doses (0.2, 2, and 20 μM) or DMSO vehicle as a control (Fig.
2a,
b). Induction of cell death was measured by using annexin V/PI staining. Apoptotic cells were scored as annexin V+, necrotic dead cells as PI+, versus live cells, which were annexin V-/PI-. We calculated the IC50 by using logarithmic transformation and we compared it to normal PBMCs isolated from healthy donors. Resting PBMCs were treated with the same concentrations of NSC 19630, 0.2, 2, and 20 μM, and induction of apoptosis by using annexin V/PI staining was measured (Fig.
2b). The IC50 in normal cells is higher compared to the HTLV-1-transformed cell line, C91PL (9.28 ± 0.23 and 2.76 ± 0.29, respectively), showing that resting PBMCs isolated from a healthy donor are less sensitive to the drug.
We expanded our analysis by testing the HTLV-1-transformed (MT-4, C8166, C91PL, 1186.94) and ATL-derived (ED, TL, ATL-25, ATL-43T, ATL-55T, LMY1, KK1, SO4, and KOB) cell lines with increasing doses of NSC 19630, 0.2, 2, and 20 μM. Inhibition of cellular growth was measured by using cell count and reported in Fig.
2c. We calculated the IC50 for every cell line by using logarithmic transformation, and the values are indicated in Table
1. Interestingly, LMY1 and ATL-55T displayed a limited reduction of cellular growth when cells were treated with 2 μM of WRN inhibitor (Fig.
3c). Consistently, the IC50 was found to be higher in those lines compared to the other cell lines tested in our study, suggesting that LMY1 and ATL-55T are less sensitive to the drug (Table
1). We included normal PBMCs isolated from healthy donors in our analysis as a negative control. As expected, limited inhibition of proliferation was noted in resting PBMCs treated with WRN inhibitor (Fig.
2c) even at high concentrations.
Table 1
Estimated IC50 of NSC 19630 in HTLV-1-transformed and patient-derived cells
HTLV-1-transformed cell lines | IC50 (μM) |
MT-4 | 1.99 ± 0.065 |
C8166 | 2.84 ± 0.19 |
C91PL | 2.76 ± 0.28 |
1186.94 | 2.23 ± 0.3 |
ATL-derived cell lines | IC50 (μM) |
ED | 0.75 ± 0.053 |
TL | 1.73 ± 0.29 |
ATL-25 | 1.79 ± 0.22 |
ATL-43T | 1.69 ± 0.23 |
ATL-55T | 6.1 ± 0.15 |
LMY1 | 4.35 ± 0.21 |
KK1 | 1.64 ± 0.038 |
SO4 | 1.45 ± 0.12 |
KOB | 1.73 ± 0.086 |
Our analysis shows a dose-dependent inhibition of cellular proliferation, suggesting that targeting WRN activity represents a promising strategy to kill ATL cells. These results were further confirmed by clonogenic assay. Two patient-derived ATL cell lines with adherent characteristics were exposed to 3 μM of NSC 19630 or DMSO for 72 h, washed, and then stained with crystal violet. A significant reduction in the number of cells was noted in ATL-25 (Fig.
2d); however, no significant changes were observed in LMY1, suggesting that these cells are resistant to the WRN inhibitor (Fig.
2e). To confirm these results, the anti-proliferative effect of NSC 19630 was quantified by measuring cleavage of XTT to an orange formazan dye using an ELISA reader at 450 nm and confirmed by microscopic cell count. Consistently, our analysis identified LMY1 as less sensitive to NSC 19630 (Fig.
2e). In contrast, ATL-25 was sensitive to the anti-proliferative effect of the WRN inhibitor (Fig.
2d).
The lack of effective therapeutic treatment for ATL patients led us to investigate the cytotoxicity of NSC 19630. WRN inhibitor is reported to induce cell death through accumulation of DNA double-strand breaks (DDSB) [
30]. A previous article published from our laboratory shows that Tax affects the DNA repair machinery, more specifically by inhibiting Homologous Recombination (HR) repair [
21]. That evidence leads us to speculate that NSC 19630 might induce cell death more prominently in Tax-expressing cells. To verify our hypothesis, we tested NSC 19630’s apoptotic effect on a Tax-expressing cell line (MT-4) versus a Tax-negative cell line (ED). Evident induction of cell death was noted in both Tax-positive and Tax-negative cells (Fig.
3a). Then we investigated if the compound affects Tax expression in three Tax-positive cell lines (C8166, MT-4, and C91PL); an ED-negative cell line was included in our analysis. Tax expression was found to be unchanged in NSC-treated cells compared to a DMSO control (Fig.
3b), suggesting that Tax expression is not a marker of drug sensitivity. A possible explanation of our finding is that Tax inhibits Homologous Recombination repair in an NF-κB-dependent manner [
21]; however, constitutive activation is described in ATL cells that do not express detectable Tax [
50].
Our preliminary data show that a WRN helicase inhibitor induces apoptosis in vitro (Fig.
3a), so we decided to include additional cell lines (MT-4, C8166, C91PL, and 1186.94) and nine ATL patient-derived cell lines (Fig.
3c). Previous studies suggest that IL-2-dependent ATL cell lines represent a model of smoldering and chronic forms of ATL, while IL-2-independent lines may better relate to the acute form of ATL. To investigate the potential use of WRN in various stages of ATL disease, we selected ATL-IL-2-independent (ED, TL, and ATL-25) and ATL-IL-2-dependent (ATL-43T, ATL-55T, LMY1, KK1, SO4, and KOB) cell lines for study. These cells were exposed for 72 h to the WRN helicase inhibitor and analyzed by annexin V/PI staining to measure the percentage of apoptosis induced by the compound (Fig.
3c). Significant levels of apoptosis were detected in IL-2-dependent and IL-2-independent ATL cell lines (Fig.
3c). These results suggest that HTLV-1-transformed ATL cells are highly sensitive to the WRN inhibitor NSC 19630 and that patients with the chronic, smoldering, or acute form of ATL are potential candidates for this therapeutic agent.
To gain some insights into the molecular mechanisms involved in NSC 19630’s effects on ATL cells, we next investigated disruption of the mitochondrial transmembrane potential (ΔΨm) (Fig.
3f). Activation of mitochondrial pathway cell death leads to the opening of the mitochondrial permeability transition (MPT) pore. The major consequences of this event are the disruption of ΔΨm and the release of pro-apoptotic proteins. Our analyses show that treatment with the NSC 19630 inhibitor resulted in the collapse of mitochondrial transmembrane potential in ED cells (Fig.
3f). Since the loss of mitochondrial transmembrane potential is associated with the activation of the caspase pathway [
51], we investigated the activation of caspase-3, an essential mediator of apoptosis activated by proteolytic cleavage. Our data indicate that cleaved caspase-3 products were readily detected in ED-treated cells when compared to DMSO-treated control cells (Fig.
3d). We then analyzed the expression of B-cell lymphoma 2 (Bcl-2) (Fig.
3d), a protein that prevents apoptosis either by sequestering caspases or by preventing the release of mitochondrial apoptogenic factors, such as cytochrome c, and an apoptosis-inducing factor, AIF, into the cytoplasm [
52]. Consistent with results from annexin V/PI, the treatment with the WRN helicase inhibitor led to decreased Bcl-2 expression in ED-treated cells (Fig.
3d). Previous studies showed that an increased expression of Mcl-1 significantly inhibits progression through the S-phase of the cell cycle [
53]. Consistent with our cell cycle results demonstrating that exposure to the WRN helicase inhibitor results in S-phase arrest (Fig.
1a), increased levels of Mcl-1 were detected in ED-treated cells (Fig.
3d). Moreover, we performed Western blot of caspase-3, Bcl-2, and Mcl-1 on protein lysates of ATL-55T and LMY1 treated with DMSO or NSC 19630. As expected, no significant change of expression was noted (Fig.
3d), confirming that these lines are less sensitive to the drug.
The WRN helicase stabilizes and maintains the replication fork during DNA replication. Failure to stabilize the fork induces DNA double-strand breaks (DDSB); in fact, treatment with replication inhibitors induces fork collapse, leading to serious DNA damage and cell death [
54]. Consistent with this concept, M. Aggarwal et al. demonstrated that a WRN inhibitor, NSC 19630, induces DDSB and accumulation of PCNA foci, which is associated with stalled replication forks [
30]. In order to investigate if apoptosis is WRN-dependent in an HTLV-1 context, we dual-stained ɣ-H2AX (a specific marker of DDSB) and PCNA in cells exposed to NSC 19630 compared to DMSO-treated cells. Our analysis shows accumulation of PCNA and ɣ-H2AX foci, suggesting that the treatment induces DNA replication issues and, consequentially, DNA damage (Fig.
3e).
NSC 19630-resistant ATL-55T and LMY1 cell lines are sensitive to NSC 617145
Data presented above suggest that NSC 19630 is a promising agent for the treatment of ATL. However, our analyses show that the apoptotic effect of the WRN helicase inhibitor was very limited in two ATL lines, namely LMY1 and ATL-55T (Fig.
3c), suggesting potential resistance mechanisms that warrant further investigations. We studied the endogenous expression of WRN helicases in HTLV-1-transformed and ATL-derived cell lines. Consistent with previously published studies, our analyses show no direct correlation between levels of WRN protein expression and sensitivity to the WRN inhibitor (data not shown).
We then investigated NSC 617145, a WRN inhibitor identified as a close structural analog of NSC 19630 but with more potent inhibition of WRN helicase activity. We next exposed HTLV-1-transformed, ATL-derived cell lines (MT-4, C8166, C91PL, 1186.94, ED, TL, ATL-25, ATL-43T, ATL-55T, LMY1, KK1, SO4, KOB) and resting PBMCs isolated from healthy donors to increasing doses of NSC 617145 (0.02, 0.2, 2, and 20 μM), and inhibition of cellular growth was measured by cell count (Fig.
4a). We calculated the IC50 for every cell line by using logarithmic transformation and the values are reported in Table
2. Our analysis clearly shows that NSC 617145 was found to be more potent in inhibiting cellular growth compared to NSC 19630. In fact, ATL-55T cells were sensitive when exposed to increasing concentrations of NSC 617145, which inhibited cellular growth, as shown by cell count and XTT proliferation assay (Fig.
4b,
c). Next, we exposed LMY1 cells to 0.25 μM for 4 days then stained with crystal violet. Consistently, again, a significant reduction in the number of cells was observed in LMY1, which was confirmed by cell count and XTT assay (Fig.
4d). Finally, we performed apoptosis assay and found that low concentrations of NSC 617145 induced high levels of cell death in both ATL-55T and LMY1 cell lines (Fig.
4e). Interestingly, limited inhibition of proliferation was noted in normal PBMCs (Fig.
4a). To confirm the limited effect of WRN inhibitor on normal cells, we estimated the IC50 in an HTLV-1-transformed cell line, C91PL, and in resting PBMCs based on induction of apoptosis. C91PL and normal resting PBMCs were exposed to increasing logarithmic doses (0.02, 0.2, and 2 μM) or DMSO vehicle as a control. Induction of cell death was measured by using annexin V/PI staining (data not show). Our analysis shows that the IC50 in normal cells is higher compared to the HTLV-1-transformed cell line C91PL (0.32 ± 0.013 and 0.13 ± 0.047, respectively), suggesting that NSC 617145 might be suitable for treating ATL patients.
Table 2
Estimated IC50 of NSC 617145 in HTLV-1-transformed and patient-derived cells
HTLV-1-transformed cell lines | IC50 (μM) |
MT-4 | 0.19 ± 0.042 |
C8166 | 0.22 ± 0.023 |
C91PL | 0.21 ± 0.0091 |
1186.94 | 0.22 ± 0.063 |
ATL-derived cell lines | IC50 (μM) |
ED | 0.15 ± 0.033 |
TL | 0.17 ± 0.039 |
ATL-25 | 0.16 ± 0.007 |
ATL-43T | 0.099 ± 0.013 |
ATL-55T | 0.22 ± 0.0012 |
LMY-1 | 0.29 ± 0.0087 |
KK1 | 0.28 ± 0.032 |
SO4 | 0.14 ± 0.061 |
KOB | 0.17 ± 0.031 |
Overall, our data suggest that HTLV-1-transformed ATL cells are very sensitive to the anti-proliferative and apoptotic effects of WRN helicase inhibitors.