Dual inhibition of DNA-PKcs and mTOR by CC-115 potently inhibits human renal cell carcinoma cell growth

CC-115 inhibits human RCC cell survival and proliferation

To begin to test the efficacy of the DNA-PKcs/mTOR dual inhibitor CC-115 as a treatment for RCC, the established human RCC cell line, 786-O, was treated with gradually-increasing concentrations of CC-115 [11, 21]. Using the CCK-8 assay to test cell viability, we showed that CC-115 inhibited 786-O cell viability in a dose-dependent manner (Figure 1A). CC-115’s significant anti-survival activity was observed after 48-72h (Figure 1A). The IC-50 of CC-115, or the concentration resulting in 50% reduction of viability, was between 1-5 μM (48h and 72h treatment) (Figure 1A). Performing a soft agar colony formation assay, results confirmed that CC-115, at 1-10 μM, significantly decreased the number of viable 786-O cell colonies (Figure 1B). Furthermore, BrdU incorporation was suppressed after CC-115 (1-10 μM) treatment (Figure 1C). These results indicated a significant anti-proliferative activity by CC-115.

Figure 1. CC-115 inhibits human RCC cell survival and proliferation. Established human RCC cell lines (786-O and A498), the primary human RCC cells (derived from two patients, “RCC1/2”), immortalized HK-2 tubular epithelial cells as well as the primary human renal epithelial cells (“Pri-Epi”) were treated with indicated concentration of CC-115 for the applied time periods, cell viability was tested by CCK-8 assay (A, D); Cell proliferation was tested by soft agar colony formation assay (B) and the BrdU ELISA assay (C, E). Data were expressed as mean ± standard deviation (SD, same for all Figures). * P < 0.05 vs. untreated control group (“Ctrl”). All in vitro experiments were repeated 3-4 times, and similar results were obtained.

The potential activity of CC-115 on other RCC cells was studied. In both established (A489 cell line) and primary human RCC cells (derived from two patients, “RCC1/RCC2”), treatment with CC-115 (5 μM) for 48/72h significantly inhibited cell viability (CCK-8 optical density/OD, Figure 1D) and proliferation (BrdU ELISA OD, Figure 1E). Importantly, in the immortalized HK-2 human proximal tubule epithelial cells and primary human renal epithelial cells (“Pri-Epi”), CC-115 treatment (5 μM, 48/72h) was non-cytotoxic (Figure 1D) nor anti-proliferative (Figure 1E). These results indicated a cancer cell specific effect by the compound. Collectively, CC-115 potently inhibited RCC cell survival and proliferation.

CC-115 induces apoptosis activation in human RCC cells

Apoptosis activation is an important mechanism for reduced viability and proliferation of cancer cells. By analyzing caspase activity, we show that the caspase-3 and the caspase-9 activities were significantly increased in CC-115 (1-10 μM)-treated 786-O cells (Figure 2A). Furthermore, CC-115 resulted in single strand DNA (ssDNA) accumulation in 786-O cells, further confirming apoptosis activation (Figure 2B). The caspase-3 specific inhibitor z-DEVD-fmk and the pan caspase inhibitor z-VAD-fmk largely attenuated CC-115 (5 μM)-induced viability reduction in 786-O cells (Figure 2C). These results demonstrate that CC-115 induced apoptotic death in 786-O RCC cells.

Figure 2. CC-115 induces apoptosis activation in human RCC cells. Established human RCC cell lines (786-O and A498), the primary human RCC cells (“RCC1/2”), HK-2 tubular epithelial cells and the primary human renal epithelial cells (“Pri-Epi”) were treated with indicated concentration of CC-115 for the applied time periods, cell apoptosis was tested by the assays mentioned in the text (A, B, D); For (C), 786-O cells were pretreated for 30 min with 50 μM of z-VAD-fmk (“zvadfmk”) or z-DEVD-fmk (“zdevdfmk”) before CC-115 treatment, and cell viability analyzed by CCK-8 assay. *P < 0.05 vs. untreated control group (“Ctrl”). ^#P < 0.05 vs. “dmso” (0.5% DMSO) (C).

In established A489 cells and primary RCC cells (“RCC1/RCC2”), treatment with CC-115 (5 μM) induced a significant increase of ssDNA content (Figure 2D), indicating apoptosis activation. On the contrary, CC-115 failed to induce significant apoptosis activation in HK-2 cells and primary human renal epithelial cells (“Pri-Epi”) (Figure 2D).

CC-115 blocks DNA-PKcs and mTOR activation in RCC cells

To demonstrate that CC-115 acts as a DNA-PKcs and mTOR dual inhibitor, we tested its inhibitory effect on signaling pathways in RCC cells. When analyzing DNA-PKcs activity using a previously-described assay [22], we found that CC-115 treatment (5 μM, 2h) significantly inhibited DNA-PKcs activity in 786-O and primary RCC cells (“RCC1/2”) (Figure 3A). DNA-PKcs protein level was unchanged with CC-115 treatment (Figure 3B). The dual inhibitor blocked phosphorylation of S6K1 (Ser-389, the indicator of mTORC1 activation [3]) and AKT (Ser-473, the indicator of mTORC2 activation) in 786-O cells (Figure 3B). Furthermore, one mTORC2 target protein, HIF-2α [15, 23], was downregulated (Figure 3B).

Figure 3. CC-115 blocks DNA-PKcs and mTOR activation in RCC cells. 786-O RCC cells, the primary human RCC cells (“RCC1/2”), or the primary human renal epithelial cells (“Pri-Epi”), were treated with CC-115 (“CC”, 5 μM) for 2h, DNA-PKcs activity (A) and expression of listed proteins in total cell lysates (B, C-”Inputs”, D, E) were shown. mTOR-Sin1-DNA-PKcs association was tested by co-immunoprecipitation assay (C, “IP”). Expression of the listed proteins in control 786-O cells (“C”) and DNA-PKcs- and mTOR-double knockout (DKO) 786-O cells were shown (F), cells were also treated with/without CC-115 (5 μM) for 48/72h, cell viability and proliferation were tested by CCK-8 assay (G) and BrdU ELISA assay (H), respectively. The exact same number of “C” and DKO cells were plated initially (“0h”) for the functional assays (G, H). *P < 0.05 vs. untreated control group (“Ctrl”) (A). *P < 0.05 vs. “C” cells (G, H).

Our previous study has shown that DNA-PKcs forms a complex with mTOR and Sin1, required for mTORC2 activation in RCC cells [15]. Performing a co-IP assay from 786-O cells, results demonstrated that CC-115 disrupted mTOR-Sin1-DNA-PKcs association (Figure 3C, “IP”), but did not inhibit their expression (Figure 3C, “Input”). In the primary human RCC cells (“RCC1/2”), CC-115 blocked AKT-S6K1 phosphorylation and downregulated HIF-2α (Figure 3D). Thus, CC-115 blocked DNA-PK and mTOR signaling in RCC cells.

In line with our previous findings [15], basal AKT-S6K1 phosphorylation as well as expression of DNA-PKcs and HIF-2α were extremely low in primary human renal epithelial cells (“Pri-Epi”) (Figure 3E). This could explain the ineffectiveness of the compound in normal epithelial cells (Figures 1, 2).

If DNA-PKcs-mTOR inhibition is the primary mechanism of CC-115-induced cytotoxicity in RCC cells, the compound should be ineffective in DNA-PKcs plus mTOR-depleted cells. To test this hypothesis, the lenti-CRISPR/Cas9 mTOR-KO construct and the lenti-CRISPR/Cas9 DNA-PKcs-KO construct were used to deplete expression in 786-O cells. Stable cells were established via FACS-mediated GFP sorting and puromycin selection. The immunoblotting confirmed DNA-PKcs and mTOR double knockout (DKO) (Figure 3F). When compared to the control cells (“C”), DKO cells presented with decreased cell viability (CCK-8 OD, Figure 3G) and proliferation (BrdU OD, Figure 3H). Importantly, adding CC-115 to the DKO 786-O cells failed to further inhibit cell viability and proliferation (Figure 3G, 3H). These results indicated that DNA-PKcs-mTOR dual inhibition is the reason of CC-115-induced RCC cell death and apoptosis.

In 786-O cells mTOR single KO, by CRISPR/Cas9-mediated gene editing (Supplementary Figure 1A), inhibited cell viability (Supplementary Figure 1B) and proliferation (BrdU incorporation, Supplementary Figure 1B). Importantly, in mTOR single KO cells adding CC-115 was able to induce further anti-786-O cell activity (Supplementary Figure 1B, 1C). Similarly, CRISPR/Cas9-induced DNA-PKcs single KO (Supplementary Figure 1D) only moderately inhibited 786-O cell viability (Supplementary Figure 1E) and proliferation (Supplementary Figure 1F), both were however augmented with CC-115 treatments (Supplementary Figure 1E, 1F). These results again confirmed that CC-115-induced anti-RCC cell activity was due to DNA-PKcs-mTOR dual inhibition.

Autophagy inhibition sensitizes CC-115 in RCC cells

Our previous studies have shown that mTOR inhibition by AZD-2014 induced feedback autophagy activation, which counteracted RCC cell death/apoptosis. Whereas inhibition of autophagy sensitized AZD-2014-induced RCC cell activity [12]. In cancer cells, inhibition of DNA-PKcs could also induce autophagy activation as a negative feedback pro-survival factor [22]. In 786-O cells CC-115 induced autophagy activation, evidenced by light chain 3B (LC3B)-II induction, Beclin-1/ATG-5/ATG-7 upregulation and p62 degradation (Figure 4A). Pharmacologic autophagy inhibition by two well-established autophagy inhibitors, 3-MA and Cq [24, 25], significantly potentiated the CC-115-induced viability reduction (Figure 4B) and apoptosis (Figure 4C) in 786-O cells.

Figure 4. Autophagy inhibition sensitizes CC-115 in RCC cells. 786-O cells were treated with CC-115 (5 μM) for 24h, expression of listed proteins in total cell lysates were shown (A); 786-O cells were pre-treated with 3-methyladenine (3-MA, 1 mM), chloroquine (Cq, 5 μM) or 0.5% DMSO vehicle control (“dmso”) for 1h, followed by CC-115 (5 μM) treatment for indicated time periods, cell viability and apoptosis were tested by CCK-8 assay (B) and ssDNA ELISA assay (C), respectively. Stable 786-O cells with Beclin-1 shRNA (“sh-Beclin1”), LC3 shRNA (“sh-LC3”) or scramble control shRNA (“sh-C”), were treated with CC-115 (1/5 μM) for indicated time periods, expression of listed proteins (D), cell viability (E) and apoptosis (F) were tested. Vector control or the “Beclin-1-OE” 786-O cells were treated with/without CC-115 (5 μM) for applied time; Listed proteins were tested (G); Cell viability (H) and apoptosis (I) were tested similarly. Primary human RCC cells (“RCC-1”), transfected with Beclin-1 siRNA-a (“siBeclin1-a”, Santa Cruz Biotech), Beclin-1 siRNA-b (“siBeclin1-b”, Cellular Signaling Tech) or scramble control siRNA (“siCtrl”) (200 nM each), were treated with/without CC-115 (5 μM) for applied time; Listed proteins were tested by Western blot assay (J); Cell viability (K) and apoptosis (L) were tested. “C” stands for parental control cells (H, I). Expression of listed proteins were quantified and normalized to the loading control (A, D, G, J). *P < 0.05 vs. untreated control group (“Ctrl”). ^#P < 0.05 vs. “dmso” cells (B, C). ^#P < 0.05 vs. “sh-C”/”si-C” cells (E, F, K, L). ^#P < 0.05 vs. Vector control cells (H, I).

To exclude possible off-target effects of the applied autophagy inhibitors, the shRNA strategy was used to silence key autophagy proteins. As described, 786-O cells were transduced with lentivirus encoding Beclin-1 shRNA or LC3 shRNA, and stable cells established. Western blotting confirmed the silencing of targeted proteins by the specific lentiviral shRNA in cells with CC-115 treatment (Figure 4D). As shown CC-115-induced viability reduction (Figure 4E) and apoptosis activation (Figure 4F) were enhanced in Beclin-1-silenced or LC3-silenced 786-O cells. Therefore, autophagy inhibition sensitized 786-O cells to CC-115-induced cell death.

Beclin-1 is required for autophagosome formation [26, 27]. Ectopic overexpression of Beclin-1 will initiate autophagy [26]. A Beclin-1-overexpressing stable 786-O cell line (“Beclin-1-OE”) was established (Figure 4G). When compared to the vector control cells, CC-115-induced viability reduction (Figure 4H) and apoptosis (Figure 4I) were dramatically inhibited in Beclin-1-OE cells. In primary human RCC cells (“RCC-1”), transfection of “Beclin-1 siRNA-a” or “Beclin-1 siRNA-b” (see Methods) efficiently downregulated Beclin-1 (Figure 4J). Beclin-1-silenced cells were sensitized to CC-115-induced cytotoxicity (Figure 4K) and apoptosis (Figure 4L). These results confirmed the pro-survival function of autophagy activation in CC-115-treated RCC cells.

CC-115 oral administration inhibits 786-O xenograft growth in vivo

The in vivo activity of CC-115 was tested. As described [11, 12], 786-O cells were s.c. injected into the flanks of the nude mice. Xenografts were established within three weeks (tumor volumes close to 100 mm³, labeled as “Day-0”). The tumor-bearing mice were randomly assigned into three groups, receiving CC-115 or vehicle control. Results demonstrated that oral administration of CC-115, at 2 mg/kg or 5 mg/kg, significantly inhibited subcutaneous 786-O xenograft growth in nude mice (Figure 5A). The estimated daily 786-O tumor growth, calculated by (tumor volume at Day-30 subtracting tumor volume at Day-30)/30, was dramatically inhibited in CC-115-treated mice (Figure 5B). Tumors of all three groups were isolated at Day-30 and weighted (Figure 5C). Tumors with CC-115 treatment were much lighter than the vehicle-treated tumors (Figure 5C). The mouse body weights were not significantly different between the three groups (Figure 5D). Collectively, oral administration of CC-115 inhibited subcutaneous 786-O xenograft growth in mice.

Figure 5. CC-115 oral administration inhibits 786-O xenograft growth in vivo. 786-O cells were inoculated via s.c. injection to the flanks of nude mice (five million cells per mouse). When tumors were close to 100 mm³ (“Day-0”), mice were randomly assigned into three groups, with ten mice per group. One group was treated with vehicle control (“Vehicle”) and the other two with CC-115 (2 and 5 mg/kg, daily, for 20 consecutive days). Tumor volumes (A) and estimated daily tumor growth (B) were shown. At Day-30, tumors of all three groups were isolated and weighted individually (C). Mice body weights were recorded (D). * P < 0.05 vs. “Vehicle” group. The mice experiments were performed once.

Chemicals and antibodies

CC-115 was provided by MCE China (Shanghai, China). Puromycin, the caspase inhibitors z-DEVD-fmk and z-VAD-fmk as well as the autophagy inhibitors chloroquine (Cq) and 3-methyaldenine (3-MA) were obtained from Sigma-Aldrich (St. Louis, MO). The antibodies of this study were provided by Cell Signaling Tech (Shanghai, China). Cell culture reagents were purchase from Gibco (Suzhou, China).

Cell culture

786-O and A498 RCC cells and HK-2 human proximal tubule epithelial cells were cultured as previously reported [12, 15]. The primary human RCC cells, derived from two different patients (“RCC1 and RCC-2”) as well as the primary human renal epithelial cells (“Pri-Epi”), were established previously [15, 30]. Primary human cells of passage 3-6 were utilized for in vitro experiments. The protocols were approved by the Ethical Committee of Nantong University (Nantong, China).

Cell Counting Kit-8 (CCK-8) assay

Cells were seeded in 96-well tissue culture plates (at 3×10³ cells per well). Following treatments, CCK-8 (Dojindo, Kumamoto, Japan) assay was performed to quantify cell viability. CCK-8 optical densities (ODs) at 450 nm were recorded.

Soft agar colony formation assay

Briefly, 786-O cells were plated onto 60 mm plates with 5×10⁴ cells per plate. After incubation with CC-115 for 10 days (CC-115 medium was renewed every two days), 786-O cell colonies were stained with 0.05% crystal violet, and manually counted.

BrdU (5-bromo-2-deoxyuridine) ELISA assay

Following the applied treatments, the BrdU ELISA kit (Cell Signaling Tech) was employed to test cell growth in vitro. BrdU ELISA ODs at 450 nm were recorded.

Single-stranded DNA (ssDNA) assay

Following the treatments, a ssDNA Apoptosis ELISA Kit (Chemicon International, Temecula, CA) was employed to quantify cell apoptosis via the attached protocol. ELISA ODs at 405 nm were recorded.

Caspase activity assay

Caspase activity assay was performed as described [12, 30] The release of AFC was quantified by a Fluoroskan system with an excitation value of 355 nm and emission value of 525 nm.

Western blotting

The detailed protocol of Western blotting assay was described as previously reported [12, 31]. Band intensity was quantified via the ImageJ software (NIH).

Co-Immunoprecipitation (Co-IP)

As reported [12], from each treatment, aliquots of 1000 μg of protein (per treatment) were pre-cleared by adding protein A/G Sepharose (“Beads”, Sigma). The pre-cleared lysates were incubated with the specific anti-mTOR antibody overnight. Afterwards, protein A/G Sepharose were added to the lysates. Beads were washed and boiled, detected by Western blotting.

shRNA and stable cell selection

For shRNA experiments, lentivirus were produced by constructing a lentiviral GV248 construct (Genechem, Shanghai, China) containing a puromycin resistance gene and either scramble control shRNA, or shRNA to Beclin-1 (“sh-Beclin-1”) or light chain 3 (LC3, “sh-LC3”). RCC cells were grown in six-well tissue culture plates (50×10³ cells per plate). Lentiviral-shRNA was added to the cells. Stable cells were selected by puromycin (5.0 μg/mL) for 10 days [32]. Expression of targeted protein in the stable cells was tested by Western blotting assay.

DNA-PKcs and mTOR double knockout

The small guide RNAs (sgRNAs) for human DNA-PKcs (targeted DNA sequence, 5’-GATCAGTTGATGACCGGCCA) and mTOR (targeted DNA sequence, 5’-CAATACTGTTGAGTGACTTC) were synthesized by Genechem (Shanghai, China). Afterwards, the sgRNA was individually inserted into the lenti-CRISPR-GFP plasmid (Addgene, Beijing, China). mTOR and DNA-PKcs constructs were transfected to 786-O cells. Afterwards, GFP-positive single cells were FACS-sorted. Stable 786-O cells were achieved by further puromycin (5.0 μg/mL) selection for another week. In the stable cells, DNA-PKcs and mTOR double knockout (“DKO”) or single knockout (SKO) was confirmed by Western blotting.

Beclin-1 siRNA

Two different Beclin-1 siRNAs were provided by Santa Cruz Biotech (sc-29797, labeled as “siBeclin-1-a”) and Cell Signaling Tech (#6222, labeled as “siBeclin-1-b”) [33]. The scramble control siRNA (labeled as “siCtrl”) was provided by Santa Cruz Biotech as well. Primary human RCC cells were initially seeded onto the 6-well tissue culture plates. Lipofectamine 2000 was utilized for siRNA (200 nM) transfection with the standard procedure. Beclin-1 knockdown was verified by Western blotting assay.

Ectopic Beclin-1 overexpression

Beclin-1 cDNA inserted into the hU6-MCS-Ubiquitin-EGFP-IRES-puromycin vector (GV428 [34]) to generate Beclin-1 construct. The latter and the lentivirus plasmids (Genepharm) were co-transfected to HEK-293 cells to generate viral particles. After filtration and enrichment (10⁸ TU/mL), virus were added to 786-O cells. Cells were cultured in puromycin (5 μg/mL) medium with 10-12 days. Beclin-1 overexpression was verified by Western blotting assay. Control cells were infected with virus with empty vector [34].

Xenograft model

As reported [12], eight-week-old female nude/beige mice (4-5 week old, 17-18 g) were provided by Nantong University Animal Laboratories. For each mouse, five million 786-O cells (in 200 μL DMEM/Matrigel) were subcutaneously (s.c.) injected to the right flank. Xenograft tumors were established within three weeks when volume of each tumor was close to 100 mm³ (“Day-0”). Mice were then randomly assigned into three groups (n=10 of each group). Mice body weights and bi-dimensional tumor volumes were recorded every six days [12]. All experimental protocols were approved by the IACUC of Nantong University.

Statistical analyses

Data were expressed as mean ± standard deviation (SD). Statistical analyses were performed as described [11, 30]. P values of <0.05 were considered statistically significant.