Hyperthermia Selectively Targets Human Papillomavirus in Cervical Tumors via p53-Dependent Apoptosis

Human papillomavirus (HPV)–associated cancers are responsible for approximately 5% of all tumors worldwide (1). Over 100 different HPV types have been identified, although only a few are oncogenic. These high-risk HPVs, such as types 16 and 18, can develop precancerous lesions. If these lesions remain untreated, they can process into cancer. HPV is associated with several types of cancers, such as carcinoma of the head and neck, anus, vagina, vulva, and penis (2–4). HPV is particularly notorious for causing cervical cancer, which is the third most common cancer in women (www.cancer.gov) and the second most frequent cause of cancer-related death in women (5). The high-risk HPV types 16 and 18 are found in over 70% of cervical cancers (6–8).

The standard treatment for cervical carcinoma is radical surgery and/or concurrent cisplatin chemoradiotherapy. Hyperthermia is clinically applied as concomitant therapy with either radiotherapy or chemotherapy, particularly as an alternative treatment if cisplatin-based chemotherapy is contraindicated or for recurrent cancers in a previously irradiated area (9). Hyperthermia is given simultaneously with cisplatin-based chemotherapy and shortly before or after radiotherapy. Several phase III studies on cervical cancer have shown a significantly better overall survival after radiotherapy combined with hyperthermia compared with radiotherapy alone (10–13). Because a high percentage of cervical tumors are HPV infected, we examined the response on HPV-positive cells to hyperthermia.

The high-risk HPV types 16 and 18 both produce early protein 6 (E6), which binds to the tumor suppressor protein p53 (14–18). E6 mediates ubiquitination of p53 and as a result targets p53 as well as itself for proteasomal degradation protein complex (14, 19). As a consequence, both p53-induced cell-cycle arrest and apoptosis are abrogated (20). Therefore, E6 might be an interesting target in cancer therapies (21, 22). In this study, we describe that hyperthermia interferes in this degradation process and allows p53 to escape proteasomal degradation, reactivating its tumor-suppressive capacities.

The cervical carcinoma cell lines: SiHa, HeLa, C33A, Caski, C4I, and HT3; the human prostate carcinoma cells lines: Du145, LNCaP, and PC3 were all obtained from the American Type Culture Collection. Colon carcinoma cell lines RKO, RC10.1, and RC10.2 were kindly provided by Dr. Kathleen Cho, University of Michigan (Ann Arbor, MI), and HCT116 p53wt and p53null were obtained from Horizon Discovery, Cambridge, UK. All cell lines were passed for fewer than 6 months after receipt, except for RKO, RC10.1, and RC10.2. Their p53 status has been tested regularly using Western blots. Human cervical carcinoma cells SiHa (HPV 16-infected), HeLa (HPV 18-infected), and C33A cells (HPV-negative) were grown in Media Modified Eagle's Medium (BioWhittaker/Lonza). HPV 16–infected cervical carcinoma Caski cells were grown in RPMI-1640 (Gibco-Brl Life Technologies). HPV-negative human cervical carcinoma cell line C4I was grown in Waymouth's medium (Gibco-Brl Life Technologies). HPV-negative human cervical carcinoma cells line HT3, colon carcinoma cell lines HCT116 p53wt and p53null, RKO, RC10.1, and RC10.2 were grown in McCoy's 5a (Gibco-Brl Life Technologies). All media contained 25 mmol/L Hepes (Gibco-Brl Life Technologies) supplemented with 10% heat-inactivated FBS and 2 mmol/L glutamine. Cells were maintained at 37°C in an incubator with humidified air supplemented with 5% CO₂. The doubling time of these cells during exponential growth is approximately 24 to 60 hours.

Human SiHa cells were injected into the hind leg of athymic mice. When tumors had grown to a volume of about 100 mm³, mice were treated with or without hyperthermia. The animals were cooled, which prevented an increase of the body core temperature. During all treatments, the animals were anesthetized with a mixture of 2.5% isoflurane in oxygen. Directly after treatment mice were sacrificed and the tumors were taken out, both untreated and treated tumors were prepared for either Western blotting or immunofluorescence.

Animal experiments were approved by the animal welfare committee of the Academic Medical Center as required by Dutch law LEX102767.

Cervical carcinoma (AMC/MEC 03/137) biopsies of patients were divided into two parts, one half was left untreated, whereas the other half was treated ex vivo with hyperthermia. Four biopsies were sliced and sonificated into cell suspensions to perform Western blotting. Three biopsies were submerged in paraformaldehyde, to be used for paraffin coupes.

Hyperthermia of the cells was performed by submerging the culture dishes in a thermostatically controlled water bath (Lauda aqualine AL12, Beun de Ronde) for 1 hour at 40, 41, or 42°C and 30 minutes at 42°C. Temperature was checked in parallel dishes, and the desired temperature (±0.1°C) was reached in approximately 5 minutes. The atmosphere was adjustable by a connection with air and CO₂ supplies. All cell cultures were heated in a 5% CO₂/95% air atmosphere and air inflow of 2 L/min.

Hyperthermia of mouse tumors was performed by submerging the tumor-bearing hind leg in a thermostatically controlled water bath for 1 hours at 42.7°C, resulting in a tumor temperature of 42°C, as confirmed by thermocouple measurements at the tumor surface. Hyperthermia (1 hour at 42°C) of patient biopsies was performed in culture dishes in the same thermostatically controlled water bath as described for the in vitro cultures.

Cells were irradiated with or without prior hyperthermia treatment. In order to allow observation of an additional effect of hyperthermia, all radiation treatments were performed with single dose (4 Gy) of gamma rays from a ¹³⁷Cs source at a dose rate of about 0.5 Gy/min. For survival curves, cells were irradiated with single doses up to 8 Gy.

Proteasomal degradation was tested using an inhibitor (MG132, 10 μmol/L for 1 hour; Sigma). For lysosomal degradation, a lysosomal inhibitor, chloroquine (100 μmol/L for 16 hours; Sigma), was used and a second compound inhibiting fusion between autophagosomes and lysosomes, so called bafilomycin A1 (200 nmol/L for 4 hours; Sigma). To check the effect of hyperthermia on the p53–E6 interaction, an E6-siRNA (Santa Cruz Biotechnology) transfection was performed. Furthermore, cells were transfected with p53-siRNA and scrambled p53-siRNA (Cell Signaling Technology) prior to hyperthermia in order to study the p53-dependent apoptosis. Cells were also incubated with MG132 in order to investigate the localization of E6 after radiation. To test whether the accumulated p53 was produced de novo, cycloheximide (1 μmol/L for 1 hour; Sigma-Aldrich) was used, which prevents translation from taking place.

Clonogenic assays were conducted to investigate the radiosensitization of hyperthermia. Experiments were performed in HPV 16 and 18–positive cervical cancer cell lines. Cells were plated before treatment into six-well culture plates (Costar). Dishes were placed in an incubator with 5% CO₂ at 37°C until sufficiently large colonies were formed. Afterwards, the medium was removed and cells were washed with PBS. A mixture of 6.0% glutaraldehyde and 0.5% crystal violet was added for at least 30 minutes at room temperature (20°C). After removing the mixture of glutaraldehyde and crystal violet, plates were washed with water and eventually dishes were dried in normal air at 20°C. Colonies were counted under a light microscope. Survival fractions were calculated by dividing the plating efficiency of treated cells by that of control cells ± SEM (23, 24).

Surviving fractions after dose D, [S(D)/S(0)], were corrected for the cytotoxicity of hyperthermia alone, and survival curves were analyzed to calculate values of the linear and quadratic parameters α and β, using SPSS 14.0 statistical software by means of a fit of the data by weighted linear regression, according to the linear-quadratic formula: Ln(S(D)/S(0)) = −(αD + βD²) (25–28). Data on clonogenic assays and apoptosis were analyzed using a t test.

To understand the additional effect of hyperthermia to radiation on HPV-positive tumors, Western blots were carried out to study p53 levels. Controls and treated cells were harvested 4 hours after treatment. Pellets were lysed in ice-cold RIPA buffer (20 mmol/L Tris-HCl, 150 mmol/L NaCl, 1 mmol/L Na₂EDTA, 1% NP-40, 1% sodium deoxycholate, 2.5 mmol/L sodium pyrophosphate, 1 mmol/L β-glycerophosphate, 1 mmol/L Na₃VO₄, 1 μg/mL leupeptin) for 30 minutes on ice with protein inhibitors (29). Laemmli buffer with 2-mercaptoethanol (355 mmol/L) was added to the supernatant (1:1) and heated in boiling water for 2 to 5 minutes. Finally, samples were sonificated (Sonics & Materials Inc.). One microgram of protein was resolved by 10% SDS-PAGE precast gels (BioRad) and transferred to PVDF membranes. Equal protein loading was checked by Ponceau S staining (29). Immunodetection was performed for p53 mAb (Dako) in combination with a horseradish peroxidase–conjugated secondary anti-mouse IgG (Southern Biotech). Housekeeping protein ERK2 was detected using mAb (Bethyl Laboratories) and a secondary anti-rabbit mAb (Invitrogen Life Technologies). All samples were enhanced using chemoluminescence (Amersham Pharmacia Biotech). Finally, blots were analyzed using LAS4000 (GE, Healthcare Life Sciences).

The Nicoletti assay (30) is another way to study p53 functionality. Apoptosis was studied in all cell lines using this assay. At 48 hours after treatment (p53-siRNA with or without hyperthermia), cells were collected and pellets were resuspended in Nicoletti buffer (0.1% w/v Sodium citrate, 0.1% v/v Triton X-100 in demi water, pH 7.4). We have also performed additional apoptotic assays (cleaved caspase-3 and Annexin V) on SiHa and HeLa cells. Analyses were made using flow cytometry (FACS Canto; BD Biosciences). Statistical analysis was performed using a t test.

SiHa and HeLa cells were grown on coverslips. Inhibitors chloroquine, bafilomycin A1, and MG132 were added to study the breakdown of p53 and E6, in normal conditions and after hyperthermia. Also, a transfection with HPV 16/18-E6-siRNA was performed to investigate whether the same results were found after hyperthermia and a specific E6-siRNA transfection. Afterwards, cells were fixed with 4% paraformaldehyde for 20 minutes on ice. Before blocking cells (PBS containing 0.1% triton x-100 and 5% normal goat serum), cells were washed with PBS. Cells were incubated overnight at 4°C in HPV 16/18-E6 protein (Santa Cruz Biotechnology) or p53 mAb (Transduction Labs). The next day, samples were washed and incubated for 1 hour at room temperature with a secondary antibody Alexa Fluor 488 (Invitrogen Life Technologies), washed and incubated for 10 min in PBS containing 1 μg/mL DAPI (Sigma-Aldrich). After washing, coverslips were stuck to slides using ProLong Gold anti-fade reagent (Invitrogen Life Technologies).

Paraffin-embedded coupes were made from xenograft tumors and patient biopsies. They were deparaffinized and rehydrated. Afterwards, a heat-induced antigen retrieval at pH 9.0 for 20 minutes was performed, followed by a 30-minute cooling period. Next, a 15-minute PO block including H₂O₂ was performed. Then coupes were incubated overnight at 4°C with HPV 16/18-E6 protein or p53 mAb (Transduction Labs). Next, tissue was embedded in Alexa Fluor 488 (Invitrogen Life Technologies), after washing in PBS. DAPI was used to stain the nuclei blue before covering tissue with ProLong Gold antifade reagent (Invitrogen Life Technologies) and a coverslip.

In order to study the cleaved caspase-3 in immunohistochemistry stainings, the heat-induced antigen retrieval was performed at pH 6.0. Primary antibody cleaved caspase-3 antibody (Cell Signaling Technology) was diluted in Primary antibody dilution (KliniPath) and incubated overnight at 4°C. Afterwards, tissue was embedded in Powervision Poly-HRP-GAM/R/R IgG (Immunologic, Immunovision Technologies). For counterstaining Eosin-hematoxyline (Fluka) was used, before covering tissue in mounting solution (Pertex) and a coverslip.

To investigate changes in p53 levels after radiation and after hyperthermia, Western blots and immunocytochemistry stainings were performed. Highly elevated levels of p53 were found in SiHa and HeLa cells after 1 hour of hyperthermia at 42°C and after a combinational treatment of hyperthermia and radiation, but not after a treatment of radiation alone. A temperature of 42°C is required to achieve a significant increase of p53 and either 40°C or 41°C was not sufficient to cause induction of p53 (Fig. 1A). The effectiveness of radiation, hyperthermia, and their combination on p53 levels is tested on HPV-positive (Supplementary Fig. S1A) and HPV-negative cells (Supplementary Fig. S1B). In HPV-positive cells, a treatment including hyperthermia was required to induce p53, whereas radiation alone was sufficient in upregulation p53 levels in HPV-negative p53wt cells. The HPV-negative p53mut or p53null cells did not show any upregulation of p53. Accumulation of p53 was observed immediately after hyperthermia (at 42°C for 1 hour) and lasted until 4 hours after treatment (Fig. 1B).

In immunocytochemistry staining, p53 was only seen after hyperthermia or after the combinational treatment of radiation and hyperthermia (Fig. 1C), p53 was neither observed in controls nor after radiation treatment only. Moreover, in vivo–treated tumors of SiHa cells grown in athymic mice and ex vivo–treated patient biopsies show a p53 induction after a 60-minute in vivo treatment with 42°C (Fig. 1D–G). Results of more patients are shown in Supplementary Fig. S1C and S1D. Quantification of immunofluorescence staining is presented in Supplementary Fig. S1E.

Protein E6, produced by human papillomavirus 16 and 18, is present in the untreated SiHa and HeLa, as shown with immunocytochemistry (Fig. 2A and B), whereas E6 has disappeared immediately after hyperthermia (42°C for 1 hour). In order to test the role of E6 in HPV-infected cells after hyperthermia, an E6-siRNA transfection was used. After transfection with E6-siRNA, neither E6 nor p53 is observed in immunocytochemistry staining. Next, cells were transfected with E6-siRNA and irradiated (4 Gy). Again, there was no E6 measurable, but now an accumulation of p53 was observed, which suggests that hyperthermia also causes stress to the cells that induce p53. Similar results were obtained with Western blots examination as shown in Supplementary Fig. S2A.

In SiHa cells grown in athymic mice, E6 was present in the cells of untreated tumors, but no E6 was detected after treatment with hyperthermia (Fig. 2C). Furthermore, after ex vivo treatment of patient biopsies, E6 disappeared after hyperthermia treatment, which was similar to the outcomes of the cell and xenograft experiments (Fig. 2D and Supplementary Fig. S2B and S2C).

Cells were also treated for 30 minutes at 42°C to understand the time needed before hyperthermia is fully effective on both E6 and p53 (Supplementary Fig. S2D). In the untreated situation, all cells were E6 positive, whereas none of the cells express p53. After a 30-minute treatment, only a few cells were completely positive for p53, and E6 expression is lower compared with the control. This shows that a 30-minute treatment is not sufficient and 60-minute treatment at 42°C is needed to induce p53 in all cells and to degrade E6 completely. Quantification of immunofluorescence staining is presented in Supplementary Fig. S2E.

To examine whether p53 accumulation is newly produced or due to interference with the degradation, cells were treated with 1 μmol/L cycloheximide for 1 hour. This did not result in a p53 accumulation, whereas treatment with 10 μmol/L MG132 for 1 hour did show p53 accumulation (Fig. 3A). It was concluded that p53 is produced de novo after a hyperthermia treatment at 42°C for 60 minutes.

To gain insight into degradation of E6 and p53 after hyperthermia, cells were treated after incubation with a proteasomal inhibitor (MG132) or a lysosomal inhibitor (chloroquine). In untreated conditions, no p53 is detected in HPV-positive cells. After blocking the lysosome using chloroquine, no difference was observed using Western blot, but after blocking the proteasome using MG132, a substantial increase of p53 was observed. Compared with blocking of the proteasome alone, combined blocking of both lysosome and proteasome showed a similar increase in p53. After a 60-minute treatment at 42°C, the lysosomal inhibition with chloroquine resulted in a slight increase in the levels of p53. However, this effect was much more pronounced when the activity of the proteasomal pathway was blocked, indicating that the degradation of p53 is to a large extent dependent on the proteasome (Fig. 3B).

In order to study the localization of p53 within the cells, the proteasome was blocked using an inhibitor (MG132) to prevent degradation. After hyperthermia p53 is only present in the nucleus, whereas after radiotherapy p53 is also present in the cytoplasm (Fig. 3C and D, top).

After studying localization of p53, the localization of E6 has been investigated as well. As we cannot detect E6 after hyperthermia, a proteasomal inhibitor (MG132) was used before treating the cells with hyperthermia or radiation. After both radiation and hyperthermia treatment, in the presence of MG132, E6 is localized in the cytoplasm and nucleus (Fig. 3C and D, bottom).

To understand the interaction of E6 and p53 after hyperthermia, immunocytochemistry stainings were performed using both lysosomal and proteasomal inhibitors. In untreated SiHa cells, E6 was present in control, and also after blocking the lysosome (bafilomycin) and after blocking the proteasome (MG132). However, p53 was absent both in control and after blocking the lysosome, whereas p53 was detected after blocking the proteasome, showing p53 is only degraded via the proteasome after complex formation with E6. Both E6 and p53 were present after treating cells with hyperthermia before blocking either the lysosome or proteasome (Fig. 3E and Supplementary Fig. S3A), suggesting that after hyperthermia E6 is degraded via both the lysosomal and proteasomal pathways. Besides functional p53, it is observed that some p53 is still degraded via the proteasomal pathway.

Moreover, conditions containing components of the standard treatment (radiation with platin-based chemotherapy) were tested. Cisplatin (cDDP) or standard chemoradiation (cDDP+RT) did not result in an accumulation of p53 compared with untreated controls (Supplementary Fig. S3B). A treatment with hyperthermia was necessary to induce p53. Quantification of immunocytochemistry staining is presented in Supplementary Fig. S3C.

To test whether the accumulated p53 is functional, apoptosis was tested by the Nicoletti assay. About 6% of the untreated SiHa and HeLa cells were apoptotic (Fig. 4A). No difference was observed after radiation or after transfection of p53-siRNA prior to hyperthermia. Both after hyperthermia and after hyperthermia combined with radiation, a significant increase in apoptosis was observed. The effects of more treatments and treatment combinations on apoptosis in SiHa and HeLa cells are summarized in Supplementary Fig. S4A. Additional HPV-positive cell lines (Supplementary Fig. S4B) show similar results as were found in SiHa and HeLa cells: induction of apoptosis occurs only if hyperthermia is part of the treatment. In the p53wt HPV-negative cell lines, radiation alone already caused a significant increase in apoptosis (Supplementary Fig. S4C), whereas there was no apoptosis observed in the p53mut or p53null cell lines. Cleaved caspase-3 immunohistochemistry staining has been performed on paraffin-embedded coupes to test the apoptotic induction in xenograft and patient biopsies. After hyperthermia, there is a clear upregulation of the cleaved caspase-3, indicating that apoptosis is induced (Fig. 4B).

Untreated SiHa cells showed a cell percentile distribution of 62.6 ± 3.1 in G₀–G₁, 15.8 ± 1.1 in S, and 26.0 ± 0.5 in G₂. After hyperthermia, a clear G₂-phase arrest is observed as the percentages change to 44.2 ± 1.2 in G₀–G₁, 11.0 ± 0.6 in S, and 49.1 ± 0.7 in G₂–M. Using p53-siRNA prior to hyperthermia, the G₂ arrest is abrogated again as the distribution of cell phases almost returned to control level, showing 65.6 ± 1.0 in G₀–G₁, 13.5 ± 0.6 in S, and 25.9 ± 1.2 in G₂–M. Similar results were found in HeLa cells: in untreated conditions, 65.2 ± 1.6 in G₀–G₁, 12.1 ± 1.2 in S, and 22.7 ± 0.4 in G₂–M; after hyperthermia, 44.0 ± 1.2 in G₀–G₁, 14.6 ± 0.2 in S, and 41.4 ± 1.3 in G₂–M. After transfection with p53-siRNA and hyperthermia, cell distribution almost returned to control levels 66.4 ± 0.5 in G₀–G₁, 11.5 ± 0.6 in S, and 22.2 ± 1.2 in G₂–M. These data in combination with more HPV-positive cell lines are shown in Supplementary Table SI. The effectiveness of hyperthermia or hyperthermia and radiation on HPV-negative cell lines is presented in Supplementary Table SII. Histograms of cell cycle show a clear G₂ arrest and induction of apoptosis after hyperthermia (Fig. 4C). Additional apoptotic assays performed on SiHa and HeLa (cleaved caspase-3 and Annexin V staining) confirm results found by the Nicoletti assay, only apoptosis induction after a treatment including hyperthermia (Supplementary Fig. S4D and S4E). In these figures, the functionality of p53-siRNA was demonstrated as well. Transfection with scrambled p53-siRNA prior to hyperthermia caused apoptosis, whereas the normal p53-siRNA prior to hyperthermia was able to prevent the induction of p53. This has been confirmed using Western blot (Supplementary Fig. S4F).

Cell survival of HPV-positive cells was tested using a clonogenic assay (24). The survival curves are shown in Fig. 4D. All cell lines show a consistently significant lower survival after radiotherapy combined with hyperthermia compared with radiation alone. This hyperthermic sensitization is shown for clinically relevant radiation doses. The corresponding α- and β values of the linear-quadratic analysis can be found in Supplementary Table SIII, showing an impressive enhancement of the linear parameter, α.

In this study, we aimed to elucidate the cellular mechanisms that may explain why hyperthermia is particularly clinically effective in HPV-positive cancers. In HPV-positive carcinoma cells, E6 binds and ubiquitinates p53, after which this complex is degraded via the proteasomal pathway (14, 15, 17, 18). As a consequence, neither p53-dependent G₂ arrest nor apoptosis can be induced. We found that hyperthermia at 42°C for a duration of 60 minutes abrogates the interaction of p53 with E6, resulting in a major accumulation of p53. This was demonstrated in multiple HPV-positive cell lines, both in vitro and in xenografts and in patient cervical carcinoma biopsies. We found that hyperthermia restored p53 function inducing p53-dependent apoptosis and G₂ arrest in HPV-positive cells, which could be inhibited by p53-siRNA (Fig. 4A and C). Our data highlight the difference between HPV-negative and -positive cells, because the HPV-negative cells induce p53 and causes apoptosis after radiation alone, whereas HPV-positive cells require hyperthermia to accomplish these effects. G₂-phase arrest and apoptosis induction after hyperthermia at 42°C for 1 hour in HPV-E6–transfected cells were observed earlier (31). This important finding may explain why patients with HPV-positive cervical cancers respond particularly well to the combination of standard radiotherapy or chemotherapy with mild hyperthermia and may boost further clinical studies with hyperthermia in other HPV-associated cancers. Activation of the p53-dependent apoptotic pathway by hyperthermic degradation of E6 is specific for HPV-positive tumor cells, which may provide a therapeutic benefit to cancer patients by minimizing normal tissue damage compared with classical cisplatin-based treatments.

Hyperthermia induces de novo p53 (Fig. 3A). This induction of p53 is observed after 1-hour treatment at 42°C; treatment at lower temperatures did not show any accumulation of p53 (Fig. 1A). Furthermore, this p53 accumulation occurs immediately after hyperthermia and remains elevated up to 4 hours later (Fig. 1B). Also in immunofluorescence staining, p53 upregulation is observed directly after hyperthermia, in both HPV 16 and 18–positive tumor models (Fig. 1C). Previous studies support the hypothesis that blocking E6 might lead to reactivation of p53 (32). In these studies, blocking of E6 alone was not effective, and it was postulated that an additional stimulus would be necessary to transport and activate p53. Our results confirm these earlier findings by showing that p53 is not elevated in E6-siRNA–transfected cells, but that p53 levels in these cells are highly increased after adding radiation. This p53 accumulation appears to be about similar in quantity to the p53 induction found after hyperthermia (Fig. 2).

By blocking the proteasome or lysosome it can be observed that E6 is degraded via the proteasomal and lysosomal pathways, and p53 is degraded exclusively via the proteasomal pathway (Fig. 3). After hyperthermia, no significant differences in p53 levels were detected, suggesting that p53 degradation remains via the proteasomal pathway. Figure 5 presents a schematic overview summarizing these findings.

Reactivating or restoration of p53 function in tumor cells has long been studied as a rational cancer treatment (33, 34). Also, targeting E6 resulted into effective killing of HPV-positive cancer cells (35). Targeting of E6, resulting in reactivation of p53, has also been studied in HPV-positive head and neck tumors (36). The main difference between the present study and previous investigations is our use of hyperthermia; although this clinical treatment has been used for decades, the sensitization mechanism in tumor cells is only now starting to unfold (37).

Our study presents a possible biologic explanation for the tumor-specific radiosensitization by hyperthermia for HPV-positive tumor cells. These findings suggest that patients with HPV-positive tumors benefit from mild hyperthermia due to abrogation of the p53–E6 complex causing activation of the p53-dependent apoptotic pathway.

No potential conflicts of interest were disclosed.

Conception and design: A.L. Oei, L.J.A. Stalpers, J.P. Medema, N.A.P. Franken

Development of methodology: A.L. Oei, R. ten Cate, L.J.A. Stalpers, N.A.P. Franken

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A.L. Oei, R. ten Cate, H.M. Rodermond, M.R. Buist, L.J.A. Stalpers, N.A.P. Franken

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): A.L. Oei, C.M. van Leeuwen, J. Crezee, H.P. Kok, N.A.P. Franken

Writing, review, and/or revision of the manuscript: A.L. Oei, C.M. van Leeuwen, R. ten Cate, M.R. Buist, L.J.A. Stalpers, J. Crezee, H.P. Kok, J.P. Medema, N.A.P. Franken

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): H.M. Rodermond, N.A.P. Franken

Study supervision: L.J.A. Stalpers, N.A.P. Franken

The authors thank Jan Sijbrands for the technical support in building the water bath for the in vivo experiments.

This work was supported by the Dutch Cancer Society (UVA 2012-5540).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.