Effects of the antifungal agent ciclopirox in HPV‐positive cancer cells: Repression of viral E6/E7 oncogene expression and induction of senescence and apoptosis

The malignant growth of human papillomavirus (HPV)‐positive cancer cells is dependent on the continuous expression of the viral E6/E7 oncogenes. Here, we examined the effects of iron deprivation on the phenotype of HPV‐positive cervical cancer cells. We found that iron chelators, such as the topical antifungal agent ciclopirox (CPX), strongly repress HPV E6/E7 oncogene expression, both at the transcript and protein level. CPX efficiently blocks the proliferation of HPV‐positive cancer cells by inducing cellular senescence. Although active mTOR signaling is considered to be critical for the cellular senescence response towards a variety of prosenescent agents, CPX‐induced senescence occurs under conditions of severely impaired mTOR signaling. Prolonged CPX treatment leads to p53‐independent Caspase‐3/7 activation and induction of apoptosis. CPX also eliminates HPV‐positive cancer cells under hypoxic conditions through induction of apoptosis. Taken together, these results show that iron deprivation exerts profound antiviral and antiproliferative effects in HPV‐positive cancer cells and suggest that iron chelators, such as CPX, possess therapeutic potential as HPV‐inhibitory, prosenescent and proapoptotic agents in both normoxic and hypoxic environments.


Introduction
About 5% of the total human cancer incidence is attributable to infections with oncogenic types of human papillomaviruses (HPVs) 1 which includes prevalent cancers in the anogenital region and of the oropharynx. 2 The viral E6 and E7 oncoproteins play a central role in the oncogenicity of HPVs. Both are pleiotropic factors that modulate the function of a broad variety of cellular proteins and pathways. 3,4 The inactivation of the p53 and pRb tumor suppressor proteins by E6 and E7, respectively, plays a central role for HPV-linked transformation. 3,4 Under many experimental conditions, the combined inhibition of E6/E7 expression leads to a rapid induction of cellular senescence, [5][6][7] an irreversible growth arrest. This indicates that HPV-positive cancer cells are "oncogene addicted" in that they require continuous E6/E7 expression for their proliferation. Thus, it will be important to identify regulatory mechanisms that can block E6/E7 expression and thereby possess the potential to be developed into new treatment options for HPV-positive (pre)neoplastic lesions. 4 Cancer cells typically exhibit metabolic alterations, which contribute to their malignant growth and to their resistance towards anticancer therapies. 8 A classic example of metabolic reprogramming is the increased rate of aerobic glycolysis in cancer cells (Warburg effect) which provides building blocks necessary for cell proliferation, such as nucleotides, amino acids and lipids. 9 More recently, changes in iron metabolism have also been linked to carcinogenesis. In specific, tumor cells typically reprogram various cellular processes that ultimately lead to enhanced iron influx and reduced iron efflux. The resulting increase in the intracellular iron pool could exert protumorigenic effects, for example, by supporting DNA synthesis and cell proliferation or by affecting cancer-associated signal transduction pathways, including Wnt (Wingless integrated) or hypoxia-inducible factors (HIF) signaling. 10,11 Although, on the one hand, it is now well recognized that oncogenic HPVs are major human carcinogens, and, on the other hand, metabolic alterations are a hallmark of cancer cells, studies on the potential crosstalk between the viral oncogenes and the host cell metabolism in HPV-transformed cells are surprisingly sparse. In the present study, we show that iron chelators, such as the synthetic antifungal agent ciclopirox (CPX), 12,13 strongly inhibit HPV E6/E7 oncogene expression. CPX leads to growth inhibition of HPV-positive cancer cells through the induction of cellular senescence or apoptosis, depending on treatment conditions. Notably, CPX-induced senescence can occur under conditions of impaired mTOR (mechanistic target of rapamycin) signaling and appears to be p53-independent. In addition, CPX also can target hypoxic HPV-positive cancer cells, which are typically more resistant towards radiotherapy and chemotherapy, 14 leading to induction of apoptosis. Collectively, these findings show that HPV oncogene expression and the growth of HPVpositive cells are highly sensitive to iron deprivation and indicate that CPX may possess therapeutic potential as an antiviral, prosenescent and/or proapoptotic agent in both normoxic and hypoxic HPV-positive cancer cells.
applied for visualization through the Fusion SL Detection System (Vilber Lourmat, Germany).

Colony formation, senescence and apoptosis assays
For colony formation assays (CFAs), cells were treated, then replated and cultivated in drug-free medium for the time periods indicated in the text, fixed and stained with formaldehyde-crystal violet as described before. 7 For senescence detection, cells were treated, then replated and cultivated in drug-free medium, for the time periods specified in the text. Cells were stained for SA-β-gal activity as described 19 and visualized by the EVOSxl Core Cell Imaging System (Thermo Fisher) with 20× magnification.
For the detection of apoptosis, TUNEL (Terminal deoxynucleotidyltransferase-mediated UTP end labeling) assays were performed as detailed before 21 using the in situ cell death detection kit (Roche Molecular Biochemicals, Indianapolis, IN). Total DNA was stained with 4 0 ,6-diamidino-2-phenylindole (DAPI, Roche Molecular Biochemicals). Apoptotic DNA strand breaks and total DNA were visualized by transmission epifluorescence microscopy. To quantify TUNEL assays, five images per coverslips were taken and the percentage of TUNEL positive cells was determined using an ImageJ Macro (written by Damir Krunic, Light Microscopy Core Facility, DKFZ). For Annexin V and PI staining, the Annexin V Apoptosis Detection Kit I for flow cytometry (BD Pharmingen, San Jose, CA) was used, following the protocol of the supplier. Cells were investigated by using the BD LSRFortessa™ cell analyzer (BD, Germany) and the BD FACS Diva Software version v8.0.1. Analyses and image generation were done with FlowJo version 10. Caspase-3/7 activation was detected by Live Cell Imaging (see below). Cleaved PARP was analyzed by immunoblot. CFAs, senescence and apoptosis assays were independently performed at least thrice with consistent results.

Cell cycle analyses
For cell cycle analyses, cells were harvested and processed essentially as previously described. 19 Cell cycle distribution was analyzed by flow cytometry using the BD LSRFortessa™ cell analyzer (BD, Germany) and the BD FACS Diva Software version v8.0.1. Image generation was done with FlowJo version 10. The Dean-Jett-Fox model was used for quantification of the cells in the different cell cycle phases. 22

Live cell imaging
For real-time analysis of cellular proliferation in 2D cell culture, 3,000 cells per well were seeded in 96-well plates and analyzed with the IncuCyte ® S3 live-cell imaging system (Essen BioScience, Hertfordshire, UK). mCherry H2B-labeled HeLa and SiHa cells were treated with different concentrations of CPX, as specified in the text, and labeled nuclei were counted. Four images per well were acquired every 4 hr at a magnification of 10× over a time period of 5 days. Analysis was performed with the IncuCyte ® S3 2018A software. For the analysis of cell proliferation in 3D cell culture, HeLa and SiHa spheroids were treated with different concentrations of CPX, as specified in the text, for up to 7 days. The size of the spheroids was continuously determined with the IncuCyte ® S3 spheroid module, measuring the spheroid area every 4 hr.
For detection of Caspase-3/7 activation, cells were seeded and incubated in a 96-well plate for 24 hr (HeLa and HeLa "p53 null") or for 48 hr (HCT116 and HCT116 p53 −/− ) and then cultivated in fresh medium containing 5 μM IncuCyte ® Caspase-3/7 Green Reagent for Apoptosis (Essen BioScience, Ann Arbor, MI). This reagent couples a recognition motif for Caspase-3/7 to a DNA intercalating dye. When the nonfluorescent substrate is cleaved by activated Caspase-3 or Caspase-7, it releases the dye which then binds to DNA and leads to the emission of green fluorescence by apoptotic cells. Images were taken every 3 hr with the IncuCyte ® S3 live-cell imaging system. The IncuCyte ® S3 2018A software was used to analyze the number of green objects normalized to cell confluence.

Statistical analyses
SigmaPlot version 12.5 (Systat Software Inc., San Jose, CA) was used for statistical tests. For comparison of relative mRNA levels upon CPX or DFO treatment, a one-sample t-test was performed with the test mean set to zero. Shapiro-Wilk was used for normality statistic and the alpha value was set to 0.05. For the comparison of EdU incorporation in EtOH-and CPX-treated cells a t-test was performed with the same settings. Values of p ≤ 0.05 (*), p ≤ 0.01 (**) and p ≤ 0.001 (***) were considered statistically significant.

Data availability
The data and other items supporting the results of the study will be made available upon reasonable request.

Iron chelators repress HPV E6/E7 oncogene expression
To analyze the effects of iron chelators on HPV-positive cells, expression of the viral E6 and E7 oncogenes were determined on mRNA and protein level upon treatment with the iron chelator deferoxamine (DFO; Figs. 1a and 1b). As indirect measure of intracellular iron levels, expression of the H-subunit of the Ferritin protein (Fig. 1a) and the transferrin receptor 1 (TFR1) mRNA (Fig. 1b) were assessed, which are expected to decrease or increase, respectively, upon iron depletion. 23 As shown in Fig. 1a, DFO strongly decreases HPV E6 and E7 protein expression in a time-dependent manner in both HPV18-positive HeLa and HPV16-positive SiHa cells. Efficient E6/E7 repression by DFO is also detectable at the mRNA level as shown by quantitative realtime (qRT)-PCR analyses (Fig. 1b).
Concordant results were obtained upon treatment of HPVpositive cancer cells with the structurally unrelated iron chelator CPX, a drug used in the clinic as a topical antifungal agent. 12,13,24 CPX also efficiently represses HPV16 and HPV18 E6 and E7 protein expression after 24-48 hr treatment in a dose-dependent manner (Fig. 1c). This effect is linked to a strong downregulation of E6/E7 transcript levels after 48 hr of CPX treatment (Fig. 1d). HeLa cells reproducibly showed a transient increase of E6/E7 transcript levels at 24 hr (Fig. 1d) which, however, was not linked to a detectable increase in E6/E7 protein levels (Fig. 1c). The increase of TFR1 mRNA levels and decrease of Ferritin protein levels, as indicators for iron deprivation, precede E6/E7 repression (Figs. 1c and 1d). The downregulation of the HPV oncogenes is counteracted by saturating 25 CPX with a twofold molar excess of two different iron donors (FeSO 4 , and ferric ammonium citrate or FAC) but not by providing a two-fold molar excess of zinc (ZnSO 4 , ZnCl 2 ; Fig. 1e). Taken together, these findings show that endogenous HPV16 and HPV18 E6/E7 expression is highly sensitive to iron deprivation as it is efficiently inhibited by iron chelators.

CPX inhibits the proliferation of HPV-positive Cancer cells and induces senescence
Next, we analyzed the phenotypic consequences of CPX treatment in HPV-positive cancer cells. CPX induces an efficient, dose-dependent growth inhibition of HeLa and SiHa cells in 2D cell culture, as assessed by live-cell imaging (Fig. 2a). Moreover, CPX also blocks the growth of HeLa and SiHa cells in 3D (spheroid) cell culture in a dose-dependent manner (Fig. 2b). Flow cytometry analyses revealed that CPX leads to an arrest in the G1 and S phases of the cell cycle, as indicated by increased G1 populations (Fig. 2c) and inhibition of incorporation of the thymidine analog EdU into the replicating DNA during S phase (Fig. 2d).
Moreover, both HeLa and SiHa cells undergo senescence after treatment with CPX for 2 days, as indicated by positive staining for the senescence marker senescence-associated ß-galactosidase (SA-β-gal) as well as by morphological alterations typical for senescence (enlargement and flattening of the cells, cytoplasmic extensions) [5][6][7]19 (Fig. 2e). Colony formation assays (CFAs) further support this notion since substantially fewer colonies grow out after release from CPX treatment, in line with the induction of senescence (an irreversible growth arrest) by CPX in these cells (Fig. 2f ). The induction of senescence upon CPX treatment is efficiently counteracted by saturating CPX with a twofold molar excess of iron (FeSO 4 or FAC) but not with a twofold molar excess of zinc (ZnSO 4 or ZnCl 2 ; Supporting Information Figs. S1a and S1b), indicating that-alike the inhibition of viral E6/E7 expression (Fig. 1e)-the senescence response of HPVpositive cancer cells is the result of iron deprivation by CPX. In line, treatment of HeLa cells for 2 days with a different iron chelator, DFO, also results in senescence (Supporting Information Fig. S1c, S1d).

CPX represses E6/E7 and induces senescence independently of p53
Under many experimental conditions, the repression of E6/E7 leads to an increase of total p53 levels by counteracting E6-mediated p53 degradation. 3,5,21 However, despite efficient E6/ E7 repression, 2 days of CPX treatment of HeLa cells results in a reduction of total p53 protein levels (Fig. 3a) which was linked to a decrease of p53 transcript levels (Supporting Information Fig. S2). In contrast, phospho-p53 (Ser 15) protein amounts increase under CPX treatment (Fig. 3a). This phosphorylated form of p53 plays an important role in the transactivation of p53 target genes, such as CDKN1A encoding the p21 cell cycle regulator. 26 Yet, following an initial increase after 24 hr, p21 protein levels decrease under conditions where phospho-p53 (Ser 15) amounts remain increased (Fig. 3a), suggesting uncoupling of this regulation under prolonged CPX treatment.
Since phospho-p53 (Ser 15) amounts can increase in response to DNA damage 27 and CPX has recently been linked to the induction of DNA damage in rhabdomyosarcoma and breast cancer cells, 28 we tested whether CPX acts as a DNA damaging agent in cervical cancer cells under our experimental conditions. Both HeLa (Supporting Information Fig. S3a) and SiHa cells (Supporting Information Fig. S3b) showed a clear increase in γ-H2AX staining, which is a widely used biomarker for DNA double-strand breaks, 29 in response to CPX treatment. This finding indicates that the increase of phospho-p53 (Ser 15) levels could be a result of CPX-induced genotoxicity.
p53 is known to play a major role in senescence induction in response to several prosenescent stimuli. 30 We thus tested whether p53 is important for CPX-induced senescence in HPV-positive cancer cells. To this end, we studied the effects of CPX on HeLa "p53 null" cells in which endogenous p53 expression is efficiently silenced by stable expression of a TP53-targeting sh (short hairpin) RNA. 15,19 Both E6 and E7 are strongly downregulated by CPX in HeLa "p53 null" cells ( Fig. 3b, upper panel) and both total p53 and phospho-p53 (Ser 15) remain undetectable (Fig. 3b, lower panel). Yet, HeLa "p53 null" cells also undergo an arrest in the G1 and S cell cycle phases (Figs. 3c and 3d) as well as induce senescence in response to CPX treatment, as indicated by positive staining for SA-β-gal (Fig. 3e) and the strong reduction of colony formation capacity upon release from CPX treatment (Fig. 3f ). These findings show that both the HPV E6/E7 oncogene repression and the induction of senescence upon CPX treatment are not linked to a detectable activation of p53.

CPX induces senescence in HPV-positive Cancer cells despite impaired mTORC1 signaling
Active mTORC1 signaling is considered to play a critical role in the induction of senescence and can be counteracted by the mTORC1 inhibitor rapamycin. 31 This is also the case in HPV-positive cancer cells exposed to prosenescent stimuli, such as E6/E7 repression or chemotherapy. 7 Alike other iron chelators, 32,33 CPX has been reported to possess the potential of inhibiting mTOR signaling. 34,35 Thus, the question arises whether CPX may be able to induce senescence under conditions of mTOR impairment.
Phospho-S6K and phospho-4E-BP1 protein amounts are two well-defined parameters for detecting active mTOR signaling. 36 As expected, the mTORC1 inhibitor rapamycin inhibits S6K phosphorylation but not 4E-BP1 phosphorylation (Fig. 4a). 37 CPX also efficiently blocks mTOR signaling in HPV-positive cancer cells, as indicated by the strong reduction of both phospho-S6K and phospho-4E-BP1 levels (Fig. 4a). Notably, CPX induces senescence in both HeLa and SiHa cervical cancer cells, even when pretreated with rapamycin, as indicated by positive SA-ß-gal staining (Fig. 4b), and substantially reduced colony formation capacity upon release from CPX (Fig. 4c). Collectively, these data show that CPX can induce cellular senescence under conditions of severely impaired mTOR signaling.

Prolonged CPX treatment induces apoptosis in normoxic HPV-positive Cancer cells
Notably, when treatment of HPV-positive cells with CPX was prolonged from 2 to 5 days, virtually no colonies emerged after subsequent release of the cells from CPX treatment (please compare Fig. 5a with Supporting Information Fig. S1b). Terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) assays indicate that a growing number of cells undergo apoptosis, starting after approximately 3 days of CPX treatment (Fig. 5b) and occurring in a dose-dependent manner (Supporting Information Fig. S4). Annexin V and PI (propidium iodide) staining after 3 days CPX treatment (Fig. 5c) reveal an increase from 8.7% to 45.3% in the fraction of late apoptotic cells (Annexin V and PI positive). Apoptosis induction is also associated with the accumulation of the cleaved form of the Caspase substrate PARP (poly(ADP-ribose) polymerase) (Fig. 5d) and a steep increase in Caspase-3/7 activity, starting at approximately 3 days of CPX treatment (Fig. 5e). Collectively, these findings show that prolonged CPX treatment induces apoptosis in HPV-positive cancer cells.
Since the proapoptotic activity of the iron chelator DFO has been linked to the induction of the proapoptotic Bcl-2 protein family member Bim, 38 we tested the effect of CPX on Bim expression. We observed that Bim expression is increased by CPX in both HeLa and SiHa cells (Supporting Information Fig. S5).  To assess a possible role for p53 during CPX-induced apoptosis, we comparatively analyzed the response of HeLa "p53 null" cells. Live-cell imaging indicates that CPX treatment ultimately leads to a similar extent of Caspase-3/7 activation in HeLa and HeLa "p53 null" cells ( Fig. 5e), indicating that the proapoptotic effect of CPX is not dependent on p53. Yet, p53 appears to be important for setting the time course of CPXinduced apoptosis, since Caspase-3/7 activation is accelerated in the presence of p53 (Fig. 5e). To test if this observation is a peculiarity of HeLa cells, we compared the response toward CPX of p53 wild-type HCT116 colon cancer cells and HCT116 p53 −/− cells, the latter bearing deletions in both TP53 alleles. 16 Both cell lines exhibited a similar extent of Caspase-3/7 activation under CPX treatment, but this process was again accelerated in the presence of functional p53 (Supporting Information Fig. S6).
Finally, we compared the apoptotic response of NOK (normal oral keratinocytes) cells which were either transduced with an HPV16 E6/E7-expressing lentiviral vector (NOK HPV16 E6/E7) or with the corresponding empty control vector (NOK pWPI). 17 CPX induced apoptosis in both cell populations, but more efficiently in the HPV16 E6/E7-expressing cells (Supporting Information Fig. S7). These results suggest that keratinocytes expressing the HPV oncogenes are sensitized towards CPX treatment.

CPX treatment induces apoptosis in hypoxic HPV-positive cancer cells
The observation that CPX can induce mTOR-independent senescence raises the question whether CPX can also act prosenescent in HPV-positive cancer cells under hypoxic conditions, where mTORC1 signaling is impaired. 7 As previously reported, 7 hypoxia leads to a strong repression of viral E6/E7 expression (Fig. 6a). CPX treatment of hypoxic HPV-positive cancer cells further reduces the already low residual E7 levels, whereas E6 expression is undetectable (Fig. 6a). Notably, and in contrast to the induction of senescence after 2 days of CPX treatment under normoxia (Fig. 2e, Supporting Information Fig. S1a), both HeLa and SiHa cells do not exhibit signs of increased senescence when treated with CPX for 2 days under hypoxia (Supporting Information Fig. S8). This is evidenced by comparable SA-ß-gal activation rates as untreated control cells (Supporting Information Fig. S8a) as well as by similar colony formation capacities, upon release from 2 days CPX exposure and subsequent cultivation under normoxia (Supporting Information Fig. S8b).
However, when HPV-positive cancer cells were treated with CPX for 3 days (instead of 2 days) under hypoxia and subsequently released from CPX and recultivated under normoxia, colony formation was clearly diminished (please compare the CFAs after CPX treatment at 1% O 2 in Fig. 6b with those in Supporting Information Fig. S8b). These findings reveal an antiproliferative effect of CPX after prolonged treatment of hypoxic HPV-positive cancer cells. Whereas hypoxic HPV-positive cancer cells exposed to 3 days CPX treatment show no increase in the number of SA-β-gal-positive cells (Fig. 6c), they exhibit enhanced apoptosis over time, as revealed by TUNEL staining (Fig. 6d). Collectively, these data indicate that hypoxic HPVpositive cancer cells are sensitive to prolonged treatment with CPX, resulting in their elimination by induction of apoptosis.

Discussion
The present work uncovers that iron chelators, such as CPX or DFO, can efficiently block HPV E6/E7 oncogene expression, both at the RNA and protein level. Detailed analyses of the effects of CPX on the phenotype of HPV positive cancer cells show a strong repression of cellular proliferation in both 2D and 3D cell culture and induction of cellular senescence. Interestingly, CPX-induced senescence of HPV-positive cancer cells appears to be p53-independent and, furthermore, occurs under conditions of severely impaired mTOR signaling. Moreover, prolonged exposure of HPV-positive cancer cells to CPX results in induction of apoptosis in both normoxic and hypoxic HPVpositive cancer cells. Collectively, these findings indicate that HPV oncogene expression as well as the tumorigenic phenotype of HPV-positive cancer cells are highly sensitive to iron deprivation. They also uncover the potential of CPX in efficiently activating tumorsuppressive pathways, that is, senescence and/or apoptosis, in both normoxic and hypoxic HPV-positive cancer cells.
In addition to the strong downregulation of the HPV oncogenes, the antiproliferative, prosenescent and proapoptotic effects of CPX in HPV-positive cancer cells are further supported by the reported potential of CPX to interfere with cellular pathways linked to tumorigenesis. For example, CPX has been reported to block growth-promoting signaling cascades, such as Wnt 39 or mTOR 35,40 signaling, and to inhibit specific iron-dependent enzymes, including ribonucleotide reductase which is required for DNA synthesis. 24 Moreover, CPX has been shown to inhibit the iron-dependent activity of deoxyhypusine hydroxylase (DOHH), an enzyme which is required for the synthesis of the eukaryotic translation initiation factor 5A-1 (eIF5A1). 41,42 Expression of eIF5A1 is important for the proliferation of eukaryotic cells and, interestingly, has been recently reported to be activated by the HPV16 E6 protein. 43 The antiproliferative potential of CPX has also been linked to its ability to induce the proteolytic degradation of the pro-proliferative Cdc25A protein. 44 Further, CPX can inhibit histone methylases, thereby epigenetically affecting gene expression. 45 Most of these CPXinduced effects have been attributed to its function as an iron chelator. Our results indicate that also the repression of the HPV oncogenes and induction of senescence by CPX occurs through iron deprivation, since both responses can be blocked by excess iron and are also induced by treatment with a structurally unrelated iron chelator, DFO.
Functional investigation of senescence induction by CPX revealed two interesting mechanistic features. First, CPX can induce senescence in HeLa "p53 null" cells, suggesting that p53 is not essential for this process. This is surprising, given the key role for p53 in the induction of cellular senescence under many conditions, 30 including in HPV-positive cancer cells in response to E6/E7 repression. 5,6 However, there is also evidence for regulatory pathways able to induce senescence independently of p53. For example, different TAp63 isoforms, members of the p53 protein family, can induce senescence in a p53-null background. 46 Notably, the TAp63β isoform can be degraded by the HPV E6 oncoprotein. 47 It will thus be interesting to investigate in future studies the effects of CPX on the regulation and function of p63 proteins in HPV-positive cancer cells. Second, CPX induces senescence under conditions of severely impaired mTOR signaling and, by itself, is a strong inhibitor of mTOR signaling in HPV-positive cancer cells. This observation is interesting in the light of the concept that-under many circumstances-active mTOR signaling is a key requirement for senescence induction by converting a reversible growth arrest into irreversible senescence (geroconversion). 31 Furthermore, studies indicate that active mTOR signaling is responsible for the secretion of protumorigenic components by the SASP (senescence-associated secretory phenotype) of senescent stromal fibroblasts, which is considered to be a major obstacle for the efficacy of prosenescent anticancer therapies, including prosenescent chemotherapy and radiotherapy (CT, RT). 48,49 Our data indicates that CPX utilizes an mTOR-independent pathway for senescence induction, and it will be interesting to determine whether the SASP of cells after CPX-induced senescence may be less tumorigenic than the SASP induced by other prosenescent agents, including CT and RT.
Whereas exposure of HPV-positive cancer cells for 2 days to CPX triggers subsequent induction of senescence, prolonged treatment over 3 days or longer programs the cells to undergo apoptosis. The potential of CPX to induce apoptosis is supported by a previous study indicating that CPX can diminish expression of antiapoptotic proteins, such as Survivin and Bcl-xL, and can induce cleavage of the antiapoptotic Bcl-2 protein. 34 Moreover, we found that CPX treatment induces expression of the proapoptotic Bim protein, further supporting the notion that iron chelators induce Bim expression 38 and raising the possibility that Bim activation may be involved in CPX-induced apoptosis.
Corroborating the results of the present investigation, CPXinduced apoptosis was reported to be p53-independent, 34 although we observed an influence of p53 on the time course of Caspase-3/7 activation, which was accelerated in the presence of p53. Interestingly, CPX also induced apoptosis in hypoxic HPVpositive cancer cells which are typically more resistant to conventional anticancer therapies in the clinic, including CT and RT. 14 Due to its antitumorigenic potential, CPX increasingly gains attention for being repurposed as a novel anticancer agent. 13 A clinical study applying CPX olamine in an oral formulation in patients with hematological malignancies was discontinued due to the low bioavailability upon oral administration and dose-limiting gastrointestinal toxicity. Recently, a clinical trial in bladder cancer patients was initiated, administrating a phosphoryl-oxymethyl ester of CPX (fosciclopirox) by intravenous injection. 50,51 The systemic application of iron chelators has been associated with severe side effects. 52 Notably, however, and in contrast to most other clinically used iron chelators, CPX can be applied topically, and is used since decades for the treatment of fungal infections of the skin and mucosa with an excellent pharmacological safety profile. 12,13 This may make CPX particularly interesting for accessing HPV-induced (pre)neoplasias, which typically affect skin and mucosa. Furthermore, the topical application route may yield high local drug concentrations and avoid side effects associated with the systemic administration of iron chelators. 52 The here observed profound antitumorigenic effects of CPX on the phenotype of HPV-positive cancer cells, which include the repression of the HPV oncogenes, the induction of senescence under conditions of impaired mTOR and p53 signaling, and the induction of apoptosis upon prolonged treatment of both normoxic and hypoxic cells, indicate that further preclinical and clinical exploration of the therapeutic potential of CPX against HPV-positive (pre)neoplastic lesions is warranted.