Synergistic anticancer activity of combined ATR and ribonucleotide reductase inhibition in Ewing's sarcoma cells

Cell culture

WE-68 (RRID: CVCL_9717) cells were kindly provided by Dr. F. van Valen (Münster, Germany), SK-ES-1 (RRID: CVCL_0627) cells were purchased from the DSMZ (Braunschweig, Germany), and A673 (RRID: CVCL_0080) cells were purchased from Sigma Aldrich (Deisenhofen, Germany). WE-68 and SK-ES-1 cells were cultured in RPMI 1640 medium, and A673 cells were cultured in DMEM (Capricorn Scientific, Ebsdorfergrund, Germany). Media were supplemented with 10% foetal calf serum (Capricorn Scientific), 100 units/ml penicillin G sodium and 100 µg/ml streptomycin sulphate (Lonza, Basel, Switzerland). Cells were cultured throughout in rat-tail collagen-coated (5 µg/cm²; Merck, Darmstadt, Germany) cell culture vessels. Cells were maintained in an incubator at 37 °C and 5% CO₂ and regularly passaged at a confluence of $\sim$ 90%. Mycoplasma contamination was tested with the qPCR Mycoplasma Test Kit from Applichem (Darmstadt, Germany).

Treatment of cells

WE-68 and SK-ES-1 cells were cultured in 12-well tissue culture plates, and A673 cells were cultured in 6-well tissue culture plates. For flow-cytometric analyses, cells were seeded at a density of 150,000 cells/well, and for caspase 3/7 assay and PCR analyses, cells were seeded at a density of 200,000 cells/well. For immunoblotting, cells were seeded in 25 cm³ tissue culture flasks at a density of 1.5 × 10⁶ cells/flask. Cells were treated with the ATRi VE821 (1–5 µM; Medchem Express, Monmouth Junction, NJ, USA) 1 h before treatment with the RNRi triapine (0.125–1 µM; Medchem Express) or didox (25–100 µM; Cayman Chemical, Ann Arbor, MI, USA) for either 48 h (flow-cytometric analyses) or 24 h (caspase 3/7 activity, immunoblotting, PCR). In the respective experiments, cells were pretreated with the broad-spectrum caspase inhibitor z-VAD-fmk (20 µM; Enzo Life Sciences, Lörrach, Germany) 1 h before treatment with VE821.

Flow-cytometric analysis of cell death and mitochondrial transmembrane potential (Δψ_m) dissipation

Cell death was analysed by assessing the integrity of the cell membrane by flow-cytometric analysis of propidium iodide (PI; Sigma Aldrich) uptake. After harvesting, cells were incubated in 2 µg/ml PI in PBS for 5 min at 4 °C in the dark. The loss of Δψ_m was analysed by determining 3,3′-dihexyloxacarbocyanine iodide (DiOC₆(3); Thermo Fisher Scientific, Dreieich, Germany) staining of mitochondria. Before harvesting, cells were incubated with 50 nM DiOC₆(3) for 45 min at 37 °C in the dark. In both read-outs, 10,000 cells per sample were analysed on a BD FACS Canto II (Heidelberg, Germany) using BD FACSDiva software; events were gated to exclude debris and doublets. The results of the cell death determinations were analysed for synergism/antagonism by the combination index (CI) method according to Chou and Talalay using Calcusyn software from Biosoft (Cambridge, UK); CI values of < 1, = 1 and > 1 indicate synergism, additivism and antagonism, respectively (Chou 2006).

Cell cycle analysis

For cell cycle analysis, ethanol-fixed cells were flow-cytometrically assessed for PI incorporation into DNA. After harvesting, cells were washed twice with PBS and fixed in 70% ethanol at − 20 °C over night. After washing, cells were resuspended in PBS containing 1% glucose, 2.5 µl/ml RNase (Roche, Mannheim, Germany) and 50 µg/ml PI. After 45 min of incubation in the dark, 20,000 cells/sample were analysed on a BD FACS Canto II using BD FACSDiva software; events were gated to exclude debris and doublets.

Caspase 3/7 activity

Caspase 3/7 activity was determined by measuring the fluorescence of the caspase 3/7 substrate Ac-DEVD-AMC (Bachem, Weil am Rhein, Germany). After harvesting, cells were lysed in 10 mM Tris–HCl, 10 mM NaH₂PO₄/NaHPO₄ (pH 7.5), 130 mM NaCl, 1% Triton X-100 and 10 mM Na₄P₂O₇ for 15 min at 4 °C in the dark. Samples were mixed with 20 mM Hepes (pH 7.5), 10% glycerol, 2 mM DTT and 25 µg/ml Ac-DEVD-AMC. The release of AMC was measured at an excitation of 355 nm and an emission of 460 nm on a Tecan Infinite M200 Pro (Crailsheim, Germany) plate reader. Relative caspase 3/7 activities were calculated as the ratio of the emission of treated to untreated cells.

Immunoblotting

Cells were centrifuged at 250×g for 5 min and lysed in 200 µl RIPA buffer (Abcam, Cambridge, UK) supplemented with 20 µl/ml protease and phosphatase (Serva Electrophoresis, Heidelberg, Germany) inhibitor cocktails. Sample volumes equivalent to 50 μg of protein were prepared in Laemmli SDS sample buffer (Thermo Fisher Scientific) and incubated at 85 °C for 2 min. 20 µl sample per lane were separated by standard SDS-PAGE on 4–12% precast gels (Serva) and electrophoretically transferred to PVDF membranes (Thermo Scientific). After 1 h blocking in TBS (pH 7.6) containing 5% BSA and 0.1% Tween-20, the membranes were incubated overnight at 4 °C with mouse anti-p53 (Santa Cruz Biotechnology, Heidelberg, #sc-126, RRID: AB_628082; 1:100) antibody. Equal loading of protein was verified by using rabbit anti-GAPDH (Cell Signaling, Danvers, MA, USA, #2118, RRID: AB_561053; 1:3000). HRP-conjugated anti-mouse IgG (Cell Signaling, #7076, RRID: AB_330924; 1:3000) and HRP-conjugated anti-rabbit IgG (Cell Signaling, #7074, RRID: AB_2099233; 1:20,000) were used as secondary antibodies followed by detection of specific signals using Immobilon Forte Western HRP Substrate (Sigma Aldrich). Imaging was done on an MF ChemiBis 3.2 imaging system (DNR Bio Imaging Systems, Neve Yamin, Israel).

Real-time RT-PCR

All procedures were conducted as per the manufacturers’ instructions. In brief, total RNA was isolated using the Peqgold Total RNA Kit including DNase digestion (Peqlab, Erlangen, Germany). RNA was transcribed into cDNA using the Omniscript RT Kit (Qiagen, Hilden, Germany). Real-time PCR was carried out on a Thermo Fisher Scientific Applied Biosystems 7900HT Real-Time PCR system. Reactions were done in duplicate using Applied Biosystems Gene Expression Assays and Universal PCR Master Mix. CDKN1A (#Hs00355782_m1) and BBC3 (#Hs00248075_m1) expression levels were normalized to B2M (#Hs00187842_m1) expression levels, and the relative gene expressions were calculated by the 2(^–ΔΔCt) method.

Statistical analysis

Statistical significance of differences between experimental groups was evaluated by the paired two-tailed Student’s t test using Microsoft Excel 2016 (*p < 0.05, **p < 0.01, ***p < 0.001).

ATRi and RNRi synergise to induce cell death in ES cells

This study aimed at exploring the potential anticancer cooperativity between the ATRi VE821 and the RNRi triapine (3-aminopyridine-2-carboxaldehyde thiosemicarbazone (3-AP)) in cultured ES cells. VE821 was developed as a selective ATRi (Reaper et al. 2011), and its clinical formulation berzosertib (also known as VE822, VX970 and M6620) is being tested in several phase I and II clinical trials (Groelly et al. 2023). Triapine inhibits the M2 subunit of RNR (RRM2; also termed β2) (Greene et al. 2020). It was evaluated in a number of phase I and II studies (Mannargudi and Deb 2017) and is currently being examined in a phase III study of advanced-stage cervical and vaginal cancers (ClinicalTrials.gov Identifier: NCT02466971). We used three ES cell lines with different TP53 status to test VE821-triapine combination treatment: p53 wild-type WE-68 cells, p53 missense mutant (C176F) SK-ES-1 cells (Sonnemann et al. 2015) and p53 null A673 cells (Ottaviano et al. 2010), to consider a potential impact of p53 on the response of ES cells to the treatment.

We initially monitored cell death to examine a possible favourable cytotoxic interaction between VE821 and triapine. One hour after treatment with VE821, cells were exposed to triapine for another 48 h. As shown in Fig. 1a, WE-68 and A673 cells were marginally sensitive to triapine alone under these conditions. Yet, when cells were cotreated with VE821, triapine evoked cell death in a concentration-dependent manner. For example, in combination with VE821 at a concentration of 1 µM—which by itself did not cause any cell death in the three cell lines (see Fig. 1a and b)—triapine-induced cell killing reached 66.5% in WE-68 cells and 61.1% in A673 cells. SK-ES-1 cells were moderately responsive to triapine, viz., treatment with triapine alone resulted in up to 35.9% cell death. In conjunction with VE821, however, triapine-triggered cell death amounted up to 83.3%. We tested the combination of VE821 with triapine for synergy by the CI method (Chou 2006). In WE-68 cells, synergy was seen after all treatment combinations except for the combinations of 1 µM VE821 with 0.125 µM or 0.25 µM triapine (Table 1). In SK-ES-1 cells, all combinations with the two higher concentrations (0.5 µM and 1 µM) of triapine produced synergistic effects (Table 2). In A673 cells, a synergistic interaction was observed in all cases except for 1 µM VE821 combined with 0.125 µM or 0.25 µM triapine, and for 2 µM VE821 combined with 0.125 µM triapine (Table 3).

To confirm the usefulness of RNRi as combination partners for ATRi, we in addition assessed the combination of VE821 with another RNRi, didox. Didox is an inhibitor of the RRM2 subunit, too, and was evaluated in a few phase I and II trials (Mannargudi and Deb 2017). Our assessments of VE821-didox-induced cell death revealed a cooperative effect also of this ATRi-RNRi combination (Fig. 1b). As observed for triapine (compare Fig. 1a), WE-68 and A673 cells were again only modestly responsive to RNR inhibition alone, while SK-ES-1 cells displayed clear responsiveness to didox. At any rate, however, coexposure to VE821 enhanced didox-provoked cell death. The combinatorial effect of VE821 and didox was especially pronounced in A673 cells: when administered alone, didox killed up 19.8% of cells, but when combined with VE821, didox-induced cell death reached 64.7%. The CI analyses validated the synergistic interaction between VE821 and didox at all concentrations except for a few combinations with 25 µM didox (Tables 4, 5 and 6).

The synergistic activity of ATRi and RNRi involves apoptosis

Although cell death can proceed via various pathways, anticancer drugs typically engage the apoptotic one (Bhola and Letai 2016). Thus, we asked whether the synergistic anticancer action of ATRi-RNRi combination treatment also involved apoptosis. To this end, we examined Δψ_m decay, caspase 3/7 activity as well as the effect of the broad-spectrum caspase inhibitor z-VAD-fmk on VE821-triapine-elicited cell death, Δψ_m decay and DNA fragmentation. In accord with our cell death findings, triapine alone had a modest effect on Δψ_m in WE-68 and A673 cells (Fig. 2a). When the same experiment was carried out in cells coexposed to VE821, however, triapine evoked Δψ_m dissipation in up to 97.3% of WE-68 cells and in up to 81.5% of A673 cells. Noncotreated SK-ES-1 cells showed some responsiveness to triapine (i.e., 36.9% loss of Δψ_m after 1 µM triapine), whereas VE821-cotreated SK-ES-1 cells displayed a very strong one (i.e., up to 91.7% loss of Δψ_m after 1 µM triapine). Very similar effect patterns were found in the caspase 3/7 activity measurements: VE821 or triapine applied individually activated caspase 3/7 only weakly, and their combination produced potentiated effects (Fig. 2b). In line with these results, z-VAD-fmk significantly reduced VE821-triapine-triggered cell death in the three cell lines, albeit to different extents: the pan-caspase inhibitor had a substantial effect in WE-68 and SK-ES-1 cells, yet an only weak one in A673 cells (Fig. 2c). Likewise, z-VAD-fmk markedly alleviated VE821-triapine-mediated Δψ_m dissipation in WE-68 and SK-ES-1 cells, but did much less so in A673 cells (Fig. 2d).

To further substantiate that the combination of ATRi with RNRi elicited apoptosis, we determined the sub-G1 fraction of cells with fragmented DNA < 2n by cell cycle analysis. As illustrated in Fig. 2e, VE821-triapine induced DNA fragmentation in the three cell lines. Yet again, z-VAD-fmk demonstrated different effectiveness: it prevented DNA fragmentation in WE-68 and SK-ES-1 cells, while it was only moderately effective in A673 cells. The cell cycle analyses also served to assess the impact of VE821 and triapine on cell cycle phase distribution. VE821 alone unsurprisingly caused an accumulation of cells in the G1 phase, while triapine alone increased the G2/M phase fraction of cells (Fig. S1). Remarkably, VE821-triapine combination treatment blocked cell cycle progression to the G2/M phase. It should be noted, however, that triapine treatment resulted in cell cycle profiles without a clear separation of G1, S and G2/M populations (Fig. S2), rendering the quantification of cell cycle phases challenging in these cases.

We likewise tested the combination of VE821 with didox for the induction of apoptosis. As demonstrated in Fig. 3a, didox triggered Δψ_m decay, and VE821 amplified this effect. Like in the analyses of VE821-didox-elicited cell death (compare Fig. 1b), strongest amplification was seen in A673 cells. z-VAD-fmk had a considerable impact on VE821-didox-mediated cell death in WE-68 and SK-ES-1 cells, but a weak one in A673 cells (Fig. 3b), in keeping with its differential impact on VE821-triapine-evoked cell death (compare Fig. 2c). With regard to VE821-didox-induced Δψ_m dissipation, a substantial effect of z-VAD-fmk was observed only in WE-68 cells (Fig. 3c).

VE821-triapine combination treatment activates p53

About 90% of ES tumours harbour wild-type TP53 (Thoenen et al. 2019). Hence, and in view of the key role of p53 in human cancers (Levine 2020), we evaluated if VE821 and triapine had an effect on p53 in p53 wild-type ES cells. Immunoblot detection revealed a moderate rise of p53 abundance after exposure to VE821 alone and a very strong one after exposure to VE821-triapine in WE-68 cells (Fig. 4a). Consistent with the high expression of mutant p53 found in a wide array of cancers (Bartek et al. 1991), p53 missense mutant SK-ES-1 cells exhibited a pronounced constitutive expression of p53 that, however, was not further increased by the treatment. To corroborate the p53-activating effect of VE821-triapine in WE-68 cells, we determined the mRNA expression of two major p53 target genes, CDKN1A (coding for the cell cycle-inhibitory protein p21^WAF1/CIP1) and BBC3 (coding for the proapoptotic BCL2 family protein PUMA) (Fischer 2017). As shown in Fig. 4b, VE821-triapine boosted CDKN1A expression most notably in the p53 wild-type cells (i.e., 73.6-fold), but also to some degree in the p53 mutant ones (i.e., 4.4-fold in SK-ES-1 cells and 13.4-fold in A673 cells). Similarly, BBC3 expression was massively induced by VE821-triapine in WE-68 cells (i.e., 105.7-fold), yet only slightly in SK-ES-1 and A673 cells (i.e., 2.4- and 2.5-fold, respectively). In agreement with its moderate effect on p53 abundance, VE821 alone produced also a moderate increase in CDKN1A and BBC3 expression (i.e., both 2.9-fold) in WE-68 cells.

Below is the link to the electronic supplementary material.

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

Not applicable.

Consent to participate consent for publication

Not applicable.