- Email: [email protected]
Role of p53, Bax, p21, and DNA-PKcs in radiation sensitivity of HCT-116 cells and xenografts
Role of p53, Bax, p21, and DNA-PKcs in radiation sensitivity of HCT-116 cells and xenografts Sergio Huerta, MD, FACS,a,* Xiaohuan Gao, PhD,a,* Sean Dineen, MD,a Payal Kapur, MD,b Debrabata Saha, PhD,c and Jeffrey Meyer, MD, MS,c Dallas, Texas
Background. Molecular factors that dictate tumor response to ionizing radiation in rectal cancer are not well described. Methods. We investigated the contribution of p53, p21, Bax, and DNA-PKcs in response to ionizing radiation in an isogeneic colorectal cancer system in vitro and in vivo. Results. HCT-116 DNA-PKcs/ cells and xenografts were radiosensitive compared with wild-type (WT) HCT-116 cells. HCT-116 p53/ cells and tumor xenografts displayed a radioresistant phenotype. Separately, p21 or Bax deficiency was associated with a radiosensitive phenotype in vitro and in vivo. In vivo, Bax deficiency led to increased tumor necrosis and decreased microvessel density. In vitro, HCT-116 Bax/ cells had decreased levels of vascular endothelial growth factor. HCT-116 WT cells had a more radioresistant phenotype after pancaspase inhibition, but pancaspase inhibition did not alter radiosensitivity in HCT-116 Bax/ cells subjected to ionizing radiation. There was no difference in cell growth in HCT-116 WT cells subjected to transient apoptosis-inducing factor (AIF) inhibition; however, HCT-116 Bax/ cells treated with AIF siRNA followed by ionizing radiation had a significant survival advantage compared with control-treated cells, implicating AIF in the radiosensitivity of Bax/ cells. Conclusion. These data might be used along with other markers to predict response to radiation in patients with rectal cancer. (Surgery 2013;154:143-51.) From the Departments of Surgery,a Pathology,b and Radiation Oncology,c University of Texas Southwestern Medical Center and the North Texas VA Health Care System, Dallas, Texas
IN PATIENTS DIAGNOSED with stage II or III rectal cancer, preoperative concurrent chemotherapy and radiation (chemoradiation) facilitates operative intervention, decreases the rate of locoregional recurrence, and leads to complete obliteration of the tumor (pathologic complete response [pCR]) in about 20% of patients.1 Patients who achieve a pCR have superior long-term outcomes compared with patients who only experience a partial response to the same preoperative strategies.2 Further, data continue to accumulate suggesting that patients with rectal cancer who achieve a clinical complete response (cCR) after neoadjuvant treatment might be observed without operative intervention.3-6 *Both authors contributed equally. Accepted for publication March 28, 2013. Reprint requests: Sergio Huerta, MD, FACS, Associate Professor, University of Texas Southwestern, 4500 S. Lancaster Road, Surgical Service (112), Dallas, TX 75216. E-mail: Sergio. [email protected]. 0039-6060/$ - see front matter Ó 2013 Mosby, Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2013.03.012
Thus, identifying factors that lead to resistance to ionizing radiation is clinically critical to be able to distinguish patients that might achieve a cCR and might be candidates for nonoperative management. Examination of tumors from patients who received preoperative chemoradiation showed that Ki-67, Bcl-2, Bax, and p53 were associated with pCR.1 A systematic analysis of the literature, however, failed to indicate that Ki-67, p53, and Bcl-2 were likely to predict a response to ionizing radiation.7 There are currently inconsistent and inconclusive results regarding various molecular pathways that lead to a radioresistant phenotype in rectal cancer (reviewed in Huerta et al8). We, therefore, investigated the role of central factors of the cell cycle, apoptosis, and DNA repair in cell death and survival after DNA damage induced by ionizing radiation in a systematic approach both in vitro and in vivo. We hypothesized that a p53/ and Bax/ genotype would lead to radioresistance, whereas cells deficient in DNA-PKcs and p21 would be radiosensitive in vitro and in vivo models of colorectal cancer. We SURGERY 143
144 Huerta et al
tested this hypothesis by an analysis of HCT-116 cells with a stable knockout (KO) genotype for p53 (HCT116 p53/), p21 (HCT-116 p21/), Bax (HCT-116 Bax/), and DNA-PKcs (HCT-116 PKcs/). We exposed these clones to ionizing radiation and compared them with the wild-type (WT) HCT-116 and the radioresistant HT-29 cell lines. These studies were also performed in vivo in immunecompromised mice bearing tumors on both limbs for each cell genotype. Our results corroborate previous findings in similar models with respect to p21-deficient9 and p53-deficient10 cells in vitro and in vivo as well as DNA-PKcs KO studies. We show that Bax-deficient HCT-116 cells have a more radiosensitive phenotype in vitro and in vivo. We explored the mechanism that led to this unexpected response. METHODS Cell culture. All lines were cultured in McCoy’s 5A medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (HyClone, Logan, UT). They were grown and handled as described previously.11 HT-29 (HTB-38) and HCT-116 (CCL-247) colorectal cancer cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD). HT-29 cells have mutations of both alleles of the p53 gene.12 HCT-116 cells are WT for p53. HCT-116 cells are also WT for most of the genes in DNA repair and chromosome stability as well as DNA checkpoints. Further, they are diploid and have a stable karyotype.13 HCT-116 p53/, HCT-116 p21/, and HCT-116 Bax/ cells were obtained from Dr B. Vogelstein (The Johns Hopkins University, Baltimore, MD). HCT-116 PKcs/ cells were kindly provided by Dr E. Hendrickson (University of Minnesota Medical School, Minneapolis, MN).13 Clonogenic survival assays. Clonogenic survival assays were performed as described previously.11 Cells were plated in complete McCoy’s 5A medium (ATCC) into 60-mm culture dishes in triplicate per data point and irradiated using a 137Cs irradiator (J.L. Shepherd and Associates, San Fernando, CA) at room temperature at a dose rate of 3.6 Gy/min. After 14 days, colonies were stained with 0.5% crystal violet solution as described previously.11 Survival fraction was calculated as:
Surgery August 2013 (colonies counted)/ [(cell seeded) 3 (plating efficiency)]. Plating efficiency was defined as: (colonies counted in non-irradiated control cells)/(cells seeded). All experiments were repeated $3 times (each time in triplicate). In vivo studies. All animal experiments were performed in accordance with institutional guidelines from Institutional Animal Care and Use Committee at the University of Texas Southwestern Medical Center for animal welfare. For tumor growth experiments, a total of 1 3 106 cells were injected subcutaneously with a 27-gauge needle into both left and right posterior flanks of 6-week-old, female, athymic nude mice (nu/nu, National Cancer Institute, Bethesda, MD). Once the animals developed palpable tumors (50–100 mm3), each group (n = 9) was assigned randomly to receive ionizing radiation on 1 side. The contralateral side received no radiation and served as control for each mouse. All tumors were subjected to external beam radiation (2 Gy every other day 3 5 [total, 10 Gy]) as described previously.11,14 Specifically, mice were irradiated using an X-RAD 320 irradiator (Precision X-Ray, Inc, North Branford, CT) at 10 Gy/min for 0.2 minute. Lead blocks were used to shield the non-tumor parts of the mice as well as the contralateral tumor that served as control. Tumor response to a given treatment was assessed over 6 weeks with twice-weekly, bidimensional measurements by means of a vernier caliper, utilizing the longest axis of the tumor and its perpendicular, and tumor volumes were estimated by multiplying the length of the tumor by the width squared by an investigator blinded to the treatment, as described previously.11,14 The response to ionizing radiation was evaluated by comparing the non-irradiated with the irradiated side for each group and expressed as percentage of response according to the following formula. For each group (n = 9), the average tumor size was determined at each data point. The average tumor size in the irradiated limb of the mice was divided by the average tumor size of the nonirradiated limb of the mice. This approach provided the percent in tumor change between the irradiated tumor and the non-irradiated limb. To determine the percent of tumor regression this ratio was subtracted from one:
ionizing radiation response ð%Þ ¼ 1:0 Tumor volume mm3 ðionizing radiation-treated xenograftÞ O Tumor volume mm3 ðcontrol xenograftÞ 100%:
Surgery Volume 154, Number 2
Western blot analysis. For each experiment, protein extracts of HT-29 and HCT-116 WT as well as KO cells were prepared from either non-irradiated cells or irradiated cells at 72 hours after exposure to 4 Gy of ionizing radiation. Western blots were performed with an equal amount of 25-mg protein extracts as described previously.11 Mouse anti-p53 and anti-actin (Santa Cruz Biotechnology, Santa Cruz, CA) and rabbit anti-Bax, anti-p21, anti-Bcl-2, anti-survivin, anti–PARP-1, and anti-caspase 3 (Cell Signaling Technology, Beverly, MA) antibodies were used at a dilution of 1:1,000. After overnight exposure to the first antibody, all experiments were incubated with a second antibody at a 1:4,000 dilution (antiimmunoglobulin G; Bio-Rad, Hercules, CA). All Western blots were repeated $3 times. Relative intensity were assessed by densitometric analysis of digitized autographic images, using the public domain NIH Image J Program (available online). Assessment of xenograft necrosis. For HCT-116 WT and HCT-116 Bax KO mice, a complete necropsy was performed. After tissue processing and embedding in paraffin, tissue blocks were cut into 3- to 4-mm slices. Hematoxylin and eosin-stained sections were examined for tumor histomorphology and the amount of tumor necrosis. The amount of tumor necrosis (percentage of tumor necrosis) was calculated by dividing the area of tumor necrosis by the total area of the tumor and multiplying by 100. A board certified pathologist (PK) who was unaware of the treatments the mice had received performed all gross examinations, histomorphologic evaluations, and tumor necrosis estimations. Assessment of microvessel density. For HCT-116 WT and HCT-116 Bax KO tumors, immunohistochemical staining for CD31 was performed to highlight endothelial cells of microvessels using standard immunoperoxidase technique with anti-CD31 antibody (1:30, rabbit polyclonal; Thermo Fisher Scientific, Pittsburgh, PA). Hematoxylin and eosin-stained sections of tumors were used to choose 1 paraffin-embedded tissue block representative of the invasive carcinoma and only one 5-mm-thick section was stained for CD31. Intratumoral microvessel density was assessed by light microscopic analysis for areas of the tumor that contained the most capillaries and small venules (microvascular ‘‘hot spots’’). After the area of greatest neovascularization was identified, and individual microvessel counts were made on a 2003 field (203 objective and 103 ocular, Nikon Eclipse E400 microscope, 0.92 mm2 per field). Results were expressed as the greatest number of microvessels in any single 2003 field.
Huerta et al 145
Enzyme-linked immunosorbent assay: Vascular endothelial growth factor. HCT-116 WT and Bax KO cells were plated in 60-mm dishes at a density of 4 3 105 cells per dish. The next day, cells were irradiated at 0 and 4 Gy of ionizing radiation, after which the medium was replaced with fresh medium. Conditioned media from cultured cells were collected after 24 hours and centrifuged at 1,000g to remove dead cells and debris. An enzyme-linked immunosorbent assay kit (Human VEGF Quantikine kit, R&D system) was used to measure total vascular endothelial growth factor (VEGF) concentrations in the culture media according to the manufacturer-recommended protocol. Pan-caspase inhibition (z-VAD-fmk). HCT-116 WT and Bax KO cells were plated in 60-mm dishes at a density of 1 3 106 cells per dish. The next day, cells were treated with 20 mmol/L of pan-caspase inhibitor z-VAD-fmk (EMD Chemicals, Inc, Darmstadt, Germany) for 3 hours before ionizing radiation. Clonogenic survival assays were performed as described. Apoptosis-inducing factor small interfering RNA. Both control and apoptosis-inducing factor (AIF)-specific small interfering (si)RNAoligos (siGENOME Non-Targeting siRNA Pool and siGENOMESMARTpool AIF siRNA) were purchased from DharmaconRNAi Technologies (Lafayette, CO). siRNA transfections using DharmaFECT transfection reagent were done in 6-well format according to the manufacturer’s instructions. Briefly, the day before transfection, both HCT116 WT and Bax KO cells were plated at a density of 1 3 105 cells per well in antibiotic-free medium. The next day, 80 pmol (4 mL of a 20 mmol/L stock) siRNAs were incubated with 2 mL DharmaFECT 2 in 400-mL serum-free medium for 20 minutes. The complexes of DharmaFECT 2 and siRNAs were then added with 1.6 mL of antibiotic-free complete medium and applied directly to each well. After 72 hours of incubation, both control siRNA and AIF siRNA treated HCT-116 WT and Bax KO cells were subject to clonogenic assays. Statistical analysis. PRISM statistical analysis software (GraphPad Software, Inc, San Diego, CA) was used for statistical analysis. Data are expressed as mean values ± standard error of the mean. Difference in cell survival (clonogenic assays) was analyzed by 1-way analysis of variance (ANOVA). The difference in tumor load in tumor xenografts receiving ionizing radiation compared with control was also evaluated by ANOVA. Differences between groups assessed by densitometry analysis derived from Western blot analysis in
146 Huerta et al
Fig 1. Clonogenic survival assays were performed on single-cell suspension of mid-log growing cells. HT-29 cells were highly radioresistant, whereas HCT-116 DNA-PKcs/ cells were exquisitely sensitive to ionizing radiation. Each data point demonstrates the average of 3 independent experiments ± the standard error of the mean (*P # .05 compared with HCT-116 wt at SF-4).
HCT-116 wt and HCT-116 KO cells were compared by Student’s t test. Student’s t test was also used for comparison between 2 groups for analysis of necrosis and microvessel density as well as assessment of VEGF between radiation-treated compared with untreated cells. Difference in cell survival in pancaspase and siRNA cells was assessed by ANOVA. Bonferroni comparison posttest was used to identify significant differences (P # .05) in treatment effects. RESULTS HCT-116 Bax/ cells have a radiosensitive phenotype. Clonogenic studies demonstrated a radioresistant phenotype in (A) HT-29 cells versus HCT-116 WT cells (surviving fraction at 4 Gy [SF-4]: 0.50 vs 0.17; P # .001) and (B) HCT-116 p53/ cells versus HCT-116 WT cells (SF-4, 0.22 vs 0.17; P # .05). These studies also showed a radiosensitive phenotype in (A) HCT-116 DNAPKcs/ versus WT cells (SF-4, 0.04 vs 0.17; P # .001); (B) HCT-116 p21/ versus WT (SF-4, 0.10 vs 0.17; P # .001); and (C) HCT-116 Bax/ cells versus HCT-116 WT cells (SF-4, 0.12 vs 0.17; P # .001; Fig 1). In vivo study results parallel the response to ionizing radiation in vitro. Our in vivo results were consistent with the in vitro data. HT-29 xenografts were highly resistant to ionizing radiation. At the end of the treatment (day 46), HT-29 xenografts experienced only a 14% decrease in tumor volume. WT HCT-116 xenografts demonstrated a 32% decrease in tumor volume (on day 46).
Surgery August 2013
Fig 2. Assessment of tumor growth in immunocompromised mice. Each group consisted of mice bearing 2 identically implanted xenografts in both limbs (n = 9 per group). One side was treated with ionizing radiation, whereas the contralateral side served as control. The average tumor volume of the ionizing radiation-treated group was divided by the average tumor volume of the control groups and expressed as the percentage of tumor response to treatment. HT-29 tumors demonstrated a poor response to ionizing radiation. Xenografts with isogenic HCT-116 cells were most radiosensitive if deficient in PKcs, Bax, and p21 and relatively radioresistant if deficient in p53 when compared with HCT-116 WT tumors (*P # .05 compared with HCT-116 wt at the end of each treatment).
Similarly, HCT-116 p53-null tumors were relatively radioresistant in vivo compared with their WT counterparts (19% tumor reduction on day 43). HCT-116 p21/ xenografts had 40% decrease in volume at the end of the treatment (on day 50). The most radiosensitive xenografts originated from HCT-116 DNA PKcs/ cells with a response of 60% (on day 46) followed by HCT-116 Bax/ xenografts, which had a decrease to 58% (on day 46). All P values at the end of treatment were #.05 versus HCT-116 WT xenografts (Fig 2). Protein levels of AIF are increased in Bax-deficient cells. Western blot analysis showed that, after treatment with ionizing radiation of HCT-116 Bax KO cells, there was a similar increase in the levels of p53 in these cells compared with HCT-116 WT cells. This response was also followed by a similar augmentation of p21 after treatment with 4 Gy of ionizing radiation in both WT and HCT-116 Bax KO cells. PARP-1 cleavage was more pronounced in HCT-116 Bax-deficient cells compared with WT. The increased levels of PARP-1 were unaccompanied with an increase in the levels of cleaved caspase-3 in Bax KO cells. HCT-116 WT cells demonstrated an increase in cleaved caspase-3 after treatment with ionizing
Surgery Volume 154, Number 2
Fig 3. Western blot analysis of HCT-116 WT and Bax KO cells. Densitometry analysis is shown below each line. There is a similar response in p53 and p21 in both isogenic cells. Bcl-2 was downregulated after 4 Gy ionizing radiation in HCT-116 WT cell and absent in HCT-116 Bax/ cells. PARP-1 was markedly increased in HCT-116 Bax-deficient cells, which was accompanied by corresponding increase in AIF, but not cleaved caspase-3.
radiation. The levels of cleaved caspase-3 did not increase after treatment with ionizing radiation in Bax-null cells. The levels of AIF were increased in HCT-116 Bax deficient cells compared with its WT counterpart (Fig 3). Bax-deficient xenografts demonstrate a decrease in angiogenesis and VEGF. Tumor necrosis was 66 ± 7% vs 55 ± 10% in HCT-116 WT ionizing radiation-treated versus control xenografts (P # .05). There was more necrosis in HCT-116 Bax/ xenografts both in the control group (78 ± 4%) and the ionizing radiation-treated tumors (86 ± 2%) relative to the WT counterparts (both P # .05). Compared with untreated HCT116 xenografts, there was an increase in tumor necrosis in HCT-116 Bax/ xenografts that were subjected to ionizing radiation (P # .05; Fig 4). Microvessel density was decreased in both treated and control HCT-116 Bax/ xenografts compared with HCT-116 WT tumors (22 ± 3% and 23 ± 3% vs 31 ± 2% and 32 ± 2%, respectively; P # .05 for both sets of comparisons; Fig 4). Bax/ tumors had a lesser level of secreted VEGF both in control cells (298 ± 15 pg/mL) and ionizing radiation-treated cells (129 ± 6 pg/mL) compared with HCT-116
Huerta et al 147
Fig 4. Assessment of necrosis and intratumoral microvessel density (iMVD) in Xenografts. HCT-116 WT (n = 9) and HCT-116 Bax KO (n = 7) xenografts were collected for assessment of microscopic necrosis. The Bax KO genotype demonstrated more tissue necrosis after treatment with ionizing radiation compared with untreated Bax-deficient tumors (*P # .001). There was no difference in the wild-type HCT-116 xenografts after treatment with ionizing radiation. There was a decrease in iMVD in both treated and untreated Bax-deficient xenografts compared with HCT-116 WT tumors (*P # .05). No difference in iMVD was observed in either WT or Bax KO tumors after treatment with ionizing radiation.
WT cells (P # .05 for both sets of comparisons; Fig 5). Pan-caspase inhibition influences radiosensitivity in HCT-116 WT cells, but not in Bax-deficient cells. We selected colony-forming assays to determine the role of caspases in HCT-116 Bax KO compared with WT. Our results demonstrated that HCT-116 WT cells treated with the pan-caspase inhibitor (z-VAD-fmk) were more radioresistant compared with control group SF-2 (0.48 ± 0.14 vs 0.43 ± 0.01, respectively; P # .001) and SF-4 (0.16 ± 0.01 vs 0.12 ± 0.01, respectively; P # .0001). In contrast, HCT-116 Bax/cells demonstrated no difference whether treated or untreated with this pan-caspase inhibitor (Fig 6). AIF siRNA treatment shows a difference in survival in HCT-116 Bax KO cells, but not in WT cells. Given the finding that AIF was upregulated in HCT-116 Bax/ cells, we hypothesized that a caspase-independent apoptosis pathway may account for the more radiosensitive phenotype in HCT-116 Bax/ cells. We examined this hypothesis by transient inhibition of AIF expression followed by clonogenic assays to study the survival of these cells. There was no difference in response to
148 Huerta et al
Fig 5. Assessment of vascular endothelial growth factor (VEGF) in vitro. There was a decrease in VEGF in both ionizing radiation-treated and untreated Bax-deficient cells compared with WT cells. In HCT-116 WT cells, there was a 1.7-fold change in ionizing radiation treated versus untreated cells. In HCT-116 deficient cells a 2.3-fold decrease in VEGF was observed in ionizing radiationtreated compared with untreated cells (*P # .05).
ionizing radiation in HCT-116 WT cells treated with this siRNA. In contrast, HCT-116 Bax KO cells were more radioresistant after AIF-siRNA treatment compared with untreated cells (SF-2, 0.38 ± 0.00 vs 0.31 ± 0.01 [P # .0001]; SF-4, 0.07 ± 0.01 vs 0.05 ± 0.01 [P # .001]; Fig 7). DISCUSSION This study was designed to investigate the relationship, within the same experimental protocol, of p21, p53, Bax, and DNA-PKcs in response to ionizing radiation in our in vitro and in vivo models of rectal cancer. Understanding the role of these molecules in response to ionizing radiation is of clinical importance in identifying phenotypic differences that might predict tumor response to neoadjuvant therapy in patients affected by rectal cancer. Three recent, independent studies have corroborated the observations of Habr-Gama et al4 in the safe approach of observing rectal cancer patients who achieve a cCR after neoadjuvant treatment.3,5,6 The approach of nonoperative management in patients who demonstrate a cCR after chemoradiation mandates careful selection of patients who might not respond to this modality. Thus, understanding tumor genotype that determines a response to ionizing radiation is of clinical importance. Because p21, p53, and Bax play a
Surgery August 2013
Fig 6. Pan-caspase inhibition. The pan-caspase inhibitor z-VAD-fdk was used to treat cells for $3 hours before ionizing radiation. After treatment with ionizing radiation, cells were plated for clonogenic survival assays as described in the Methods. Pan-caspase inhibition resulted in a more radioresistant phenotype in HCT-116 WT cells compared with untreated cells. Pan-caspase inhibition had no effect on the survival of HCT-116 Baxdeficient cells. Each data point demonstrates the average of 3 independent experiments ± the standard error of the mean (**P # .01 and ***P # .001 HCT-116 WT cells treated with z-VAD-fmk versus HCT-116 WT cells that did not receive z-VAD-fmk).
central role in radiation-induced cell death and because the current literature demonstrates inconclusive results regarding their clinical role as biomarkers, we undertook a systematic analysis of cells deficient in p21, p53, and Bax to compare the response to ionizing radiation with a well-known radiosensitive cell line (HCT-116 DNA-PKcs/) and a radioresistant cell clone (HT-29). Our results corroborate previous findings with regard to the role of p21, p53, and DNAPKcs in their response to ionizing radiation and demonstrate that a Bax-KO phenotype renders a radiosensitive phenotype in HCT-116 cells and xenografts. Our results also point to a potential mechanism leading to this effect. Consistent with previous observations, HCT-116 DNA-PKcs/ cells were sensitive to ionizing radiation.13 For instance, DNA-PKcs–deficient Chinese hamster ovary cells showed profound cell death after treatment with ionizing radiation compared with the DNA-PKcs complimented V3-YAC cells.15 Our observations in this study suggest that the status of tumors with regard to DNA-PKcs might be used as a marker for radiosensitivity. Similarly, HCT-116 p21/ cells are more radiosensitive to ionizing radiation compared with isogenic WT cells.9,14 We observed irradiated
Surgery Volume 154, Number 2
Fig 7. AIF-siRNA treatment. A, Cells were treated with siRNA specific for AIF for 72 hours. After siRNA treatment, all cells were subjected to ionizing radiation and plated for clonogenic survival assays. Transient siRNA inhibition of AIF led to a more radioresistant phenotype in HCT-116 Bax-deficient cells compared with HCT-116 WT cells. Each data point demonstrates the average of 3 independent experiments ± the standard error of the mean (**P # .01 and ***P # .001 HCT-116 Bax/ cells treated with AIF-siRNA versus HCT-116 WT cells that did not receive AIF-siRNA). B, Knock down of AIF was corroborated by Western blot analysis.
HCT-116 p21/ cells to decrease survival after treatment with ionizing radiation compared with the HCT-116 WT clone in vitro. Mice bearing irradiated p21-deficient xenografts also had decreased tumor growth compared with irradiated HCT-116 WT xenografts at the end of the treatment period in our in vivo study. Thus, rectal tumors bearing mutations of this molecule might demonstrate more radiosensitivity compared with wild-type cancers. Our results showed that compared with HCT-116 WT cells and xenografts, p53 deficiency conferred cell survival and resistance to ionizing radiation. This finding is consistent with previous reports that examined the effects of ionizing radiation in HCT-116 WT and p53/ cells.10 p53 plays an essential role in cellular response to radiation in vitro; that is, WT p53 renders a radiosensitive phenotype in cultured cells. Ex vivo studies have failed to provide predictive or prognostic information as a result of the low number of subjects included in the studies, techniques utilized, inability of p53 antibodies to recognize the mutated versus the WT form of p53, or a combination of these. This finding might explain the conflicting results of the literature in utilizing p53 as a biomarker in tumor response to chemoradiation and underscored the need for further investigation on this subject.8,16
Huerta et al 149
Because Bax is a pro-apoptotic protein, we anticipated that a Bax KO status would render HCT-116 cells radioresistant. Contrary to the pro-apoptotic function of Bax, we found that HCT-116 cells deficient in this gene were highly radiosensitive in vitro and in vivo. Although a few studies demonstrate that Bax-deficient cells are resistant to chemotherapeutic agents,17-19 evidence indicating the response of Bax-deficient colorectal cancer cells to ionizing radiation in preclinical studies is lacking. Limited ex vivo studies have shown that tumors expressing Bax had a positive response to chemoradiation compared with patients treated for rectal cancer who did not show Bax staining in their tumors.7,20 Preclinical studies showed that Bax-deficient HCT-116 cells were slightly resistant to apoptotic cell death after treatment with 5-fluorouracil. In contrast, whereas HCT-116 WT cells demonstrated substantial apoptosis to nonsteroidal anti-inflammatory drugs, HCT-116 Bax KO cells showed no apoptosis when exposed to the same drug.19 This study did not assess the response of HCT-116–deficient cells to ionizing radiation. Our study shows unambiguously that HCT-116, Bax-deficient cells are more sensitive to ionizing radiation in vitro and in vivo compared with the isogeneic counterpart HCT-116 WT clone. Thus, rectal tumors demonstrating lack of immunoreactivity to Bax might prove to be more radiosensitive compared with Bax-positive tumors. Because the results with respect to Bax were unexpected, we determined whether there was a specific pathway leading to this response. To assess the mechanism that led to a more radiosensitive phenotype in HCT-116 Bax-deficient cells, we inspected the tumor xenografts from both WT and Bax KO origin and found an increase in the degree of necrosis and a decrease in angiogenesis both in treated and untreated tumor xenografts that originated from the Bax-deficient cells. We corroborated these findings in vitro by demonstrating a decrease in VEGF in HCT-116 Bax deficient compared with HCT-116 WT cells. Whereas a decrease in VEGF led to a decrease in microvessel count and an increase in necrosis in tumors originating from HCT-116 cells deficient in Bax, this mechanism did not explain the parallel observation that we showed in vitro. We observed an increase in PARP-1 in HCT-116 Bax/ cells after treatment with ionizing radiation compared with HCT-116 WT cells. A parallel increase in cleaved caspase-3, however, did not occur. Thus, our in vitro findings did not demonstrate a clear pathway leading to cell death in Bax-deficient cells
150 Huerta et al
with regard to cell cycle arrest, necrosis, or caspasemediated apoptosis. Because colony formation assays assess all parameters related to cell growth, including apoptosis, cell cycle, and necrosis, these assays are ideal to examine the relative effects of an intervention, over time, in cell survival after treatment with ionizing radiation. We performed clonogenic assays to determine the mechanisms that led to a more radiosensitive phenotype in Bax-deficient cells. Because Western blot analysis suggested that AIF was increased moderately in Bax KO compared with WT cells and because caspase-3–mediated apoptosis was not detected in our studies, we studied the effects of pan-caspase inhibition in HCT-116 Bax-deficient and WT cells. We showed that HCT-116 cells treated with the pancaspase inhibitor (z-VAD-fmk) had a more radioresistant phenotype, although there was no difference in survival in Bax-deficient cells after the same form of treatment, indicating that caspase-mediated apoptosis was not the mechanism of increased radiosensitivity. We then proceeded with transient inhibition of AIF followed by clonogenic studies in cells subjected to various doses of ionizing radiation as well as in untreated cells, which demonstrated a more pronounced radioresistant phenotype in Bax-deficient compared with WT cells. Our results suggest that the mechanism of ionizing radiation treatment in Bax-deficient cells may be mediated, in part, by AIF. Because our results do not show a marked difference in Bax WT and Bax KO cells, AIF cannot be the sole contributor to the radiosensitivity in these observations. These findings point to a multifactorial effect that might include AIF but not caspases. Further studies are required to further shed light into this pathway. In the present study, we revealed that Bax deficiency in HCT-116 cells led to radiosensitivity in vitro and in vivo in colorectal cancer. In vivo, this response was associated with a decrease in angiogenesis. In vitro, we showed that AIF might play a role in cell death compared with caspasemediated mechanisms as RNA inhibition of AIF led to survival advantage in HCT-116 Bax/, but not in HCT-116 WT cell treated under the same conditions. Bax may be a potential biomarker to select patients with rectal cancer who might respond well to neoadjuvant treatment. The authors thank Derrick Chen from the University of Texas Southwestern (UTSW) Medical School for his assistance with assessment of necrosis and microvessel density in xenografts. We thank Monica Lu from UTSW
Surgery August 2013
Medical School for her editorial assistance. This work was supported by funds from the Department of Surgery at UTSW. The authors acknowledge the assistance of the Southwestern Small Animal Imaging Resource, which is supported in part by NCI U24 CA126608, the Harold C. Simmons Cancer Center through an NCI Cancer Center Support Grant, 1P30 CA142543-01, and The Department of Radiology.
REFERENCES 1. Huerta S, Hrom J, Gao X, Saha D, Anthony T, Reinhart H, et al. Tissue microarray constructs to predict a response to chemoradiation in rectal cancer. Dig Liver Dis 2010;42: 679-84. 2. Maas M, Nelemans PJ, Valentini V, Das P, Rodel C, Kuo LJ, et al. Long-term outcome in patients with a pathological complete response after chemoradiation for rectal cancer: a pooled analysis of individual patient data. Lancet Oncol 2010;11:835-44. 3. Dalton RS, Velineni R, Osborne ME, Thomas R, Harries S, Gee AS, et al. A single-centre experience of chemoradiotherapy for rectal cancer: is there potential for nonoperative management? Colorectal Dis 2012;14:567-71. 4. Habr-Gama A, Perez RO, Nadalin W, Sabbaga J, Ribeiro U Jr, Silva e Sousa AH Jr, et al. Operative versus nonoperative treatment for stage 0 distal rectal cancer following chemoradiation therapy: long-term results. Ann Surg 2004;240:711-7. 5. Maas M, Beets-Tan RG, Lambregts DM, Lammering G, Nelemans PJ, Engelen SM, et al. Wait-and-see policy for clinical complete responders after chemoradiation for rectal cancer. J Clin Oncol 2011;29:4633-40. 6. Smith JD, Ruby JA, Goodman KA, Saltz LB, Guillem JG, Weiser MR, et al. Nonoperative management of rectal cancer with complete clinical response after neoadjuvant therapy. Ann Surg 2012;256:965-72. 7. Kuremsky JG, Tepper JE, McLeod HL. Biomarkers for response to neoadjuvant chemoradiation for rectal cancer. Int J Radiat Oncol Biol Phys 2009;74:673-88. 8. Huerta S, Gao X, Saha D. Mechanisms of resistance to ionizing radiation in rectal cancer. Expert Rev Mol Diagn 2009;9:469-80. 9. Tian H, Wittmack EK, Jorgensen TJ. p21WAF1/CIP1 antisense therapy radiosensitizes human colon cancer by converting growth arrest to apoptosis. Cancer Res 2000;60:679-84. 10. Cook T, Wang Z, Alber S, Liu K, Watkins SC, Vodovotz Y, et al. Nitric oxide and ionizing radiation synergistically promote apoptosis and growth inhibition of cancer by activating p53. Cancer Res 2004;64:8015-21. 11. Gao X, Saha D, Kapur P, Anthony T, Livingston EH, Huerta S. Radiosensitization of HT-29 cells and xenografts by the nitric oxide donor DETANONOate. J Surg Oncol 2009; 100:149-58. 12. Shibata D, Peinado MA, Ionov Y, Malkhosyan S, Perucho M. Genomic instability in repeated sequences is an early somatic event in colorectal tumorigenesis that persists after transformation. Nat Genet 1994;6:273-81. 13. Ruis BL, Fattah KR, Hendrickson EA. The catalytic subunit of DNA-dependent protein kinase regulates proliferation, telomere length, and genomic stability in human somatic cells. Mol Cell Biol 2008;28:6182-95. 14. Huerta S, Gao X, Saha D. Murine orthotopic model for the assessment of chemoradiotherapeutic interventions in rectal cancer. Anticancer Drugs 2011;22:371-6.
Surgery Volume 154, Number 2
15. Zhao Y, Thomas HD, Batey MA, Cowell IG, Richardson CJ, Griffin RJ, et al. Preclinical evaluation of a potent novel DNA-dependent protein kinase inhibitor NU7441. Cancer Res 2006;66:5354-62. 16. Huerta S, Goulet EJ, Huerta-Yepez S, Livingston EH. Screening and detection of apoptosis. J Surg Res 2007; 139:143-56. 17. Wagener C, Bargou RC, Daniel PT, Bommert K, Mapara MY, Royer HD, et al. Induction of the death-promoting gene bax-alpha sensitizes cultured breast-cancer cells to drug-induced apoptosis. Int J Cancer 1996;67:138-41.
Huerta et al 151
18. Yamaguchi H, Bhalla K, Wang HG. Bax plays a pivotal role in thapsigargin-induced apoptosis of human colon cancer HCT116 cells by controlling Smac/Diablo and Omi/HtrA2 release from mitochondria. Cancer Res 2003;63:1483-9. 19. Zhang L, Yu J, Park BH, Kinzler KW, Vogelstein B. Role of BAX in the apoptotic response to anticancer agents. Science 2000;290:989-92. 20. Chang HJ, Jung KH, Kim DY, Jeong SY, Choi HS, Kim YH, et al. Bax, a predictive marker for therapeutic response to preoperative chemoradiotherapy in patients with rectal carcinoma. Hum Pathol 2005;36:364-71.
Background. Molecular factors that dictate tumor response to ionizing radiation in rectal cancer are not well described. Methods. We investigated the contribution of p53, p21, Bax, and DNA-PKcs in response to ionizing radiation in an isogeneic colorectal cancer system in vitro and in vivo. Results. HCT-116 DNA-PKcs/ cells and xenografts were radiosensitive compared with wild-type (WT) HCT-116 cells. HCT-116 p53/ cells and tumor xenografts displayed a radioresistant phenotype. Separately, p21 or Bax deficiency was associated with a radiosensitive phenotype in vitro and in vivo. In vivo, Bax deficiency led to increased tumor necrosis and decreased microvessel density. In vitro, HCT-116 Bax/ cells had decreased levels of vascular endothelial growth factor. HCT-116 WT cells had a more radioresistant phenotype after pancaspase inhibition, but pancaspase inhibition did not alter radiosensitivity in HCT-116 Bax/ cells subjected to ionizing radiation. There was no difference in cell growth in HCT-116 WT cells subjected to transient apoptosis-inducing factor (AIF) inhibition; however, HCT-116 Bax/ cells treated with AIF siRNA followed by ionizing radiation had a significant survival advantage compared with control-treated cells, implicating AIF in the radiosensitivity of Bax/ cells. Conclusion. These data might be used along with other markers to predict response to radiation in patients with rectal cancer. (Surgery 2013;154:143-51.) From the Departments of Surgery,a Pathology,b and Radiation Oncology,c University of Texas Southwestern Medical Center and the North Texas VA Health Care System, Dallas, Texas
IN PATIENTS DIAGNOSED with stage II or III rectal cancer, preoperative concurrent chemotherapy and radiation (chemoradiation) facilitates operative intervention, decreases the rate of locoregional recurrence, and leads to complete obliteration of the tumor (pathologic complete response [pCR]) in about 20% of patients.1 Patients who achieve a pCR have superior long-term outcomes compared with patients who only experience a partial response to the same preoperative strategies.2 Further, data continue to accumulate suggesting that patients with rectal cancer who achieve a clinical complete response (cCR) after neoadjuvant treatment might be observed without operative intervention.3-6 *Both authors contributed equally. Accepted for publication March 28, 2013. Reprint requests: Sergio Huerta, MD, FACS, Associate Professor, University of Texas Southwestern, 4500 S. Lancaster Road, Surgical Service (112), Dallas, TX 75216. E-mail: Sergio. [email protected]. 0039-6060/$ - see front matter Ó 2013 Mosby, Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2013.03.012
Thus, identifying factors that lead to resistance to ionizing radiation is clinically critical to be able to distinguish patients that might achieve a cCR and might be candidates for nonoperative management. Examination of tumors from patients who received preoperative chemoradiation showed that Ki-67, Bcl-2, Bax, and p53 were associated with pCR.1 A systematic analysis of the literature, however, failed to indicate that Ki-67, p53, and Bcl-2 were likely to predict a response to ionizing radiation.7 There are currently inconsistent and inconclusive results regarding various molecular pathways that lead to a radioresistant phenotype in rectal cancer (reviewed in Huerta et al8). We, therefore, investigated the role of central factors of the cell cycle, apoptosis, and DNA repair in cell death and survival after DNA damage induced by ionizing radiation in a systematic approach both in vitro and in vivo. We hypothesized that a p53/ and Bax/ genotype would lead to radioresistance, whereas cells deficient in DNA-PKcs and p21 would be radiosensitive in vitro and in vivo models of colorectal cancer. We SURGERY 143
144 Huerta et al
tested this hypothesis by an analysis of HCT-116 cells with a stable knockout (KO) genotype for p53 (HCT116 p53/), p21 (HCT-116 p21/), Bax (HCT-116 Bax/), and DNA-PKcs (HCT-116 PKcs/). We exposed these clones to ionizing radiation and compared them with the wild-type (WT) HCT-116 and the radioresistant HT-29 cell lines. These studies were also performed in vivo in immunecompromised mice bearing tumors on both limbs for each cell genotype. Our results corroborate previous findings in similar models with respect to p21-deficient9 and p53-deficient10 cells in vitro and in vivo as well as DNA-PKcs KO studies. We show that Bax-deficient HCT-116 cells have a more radiosensitive phenotype in vitro and in vivo. We explored the mechanism that led to this unexpected response. METHODS Cell culture. All lines were cultured in McCoy’s 5A medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (HyClone, Logan, UT). They were grown and handled as described previously.11 HT-29 (HTB-38) and HCT-116 (CCL-247) colorectal cancer cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD). HT-29 cells have mutations of both alleles of the p53 gene.12 HCT-116 cells are WT for p53. HCT-116 cells are also WT for most of the genes in DNA repair and chromosome stability as well as DNA checkpoints. Further, they are diploid and have a stable karyotype.13 HCT-116 p53/, HCT-116 p21/, and HCT-116 Bax/ cells were obtained from Dr B. Vogelstein (The Johns Hopkins University, Baltimore, MD). HCT-116 PKcs/ cells were kindly provided by Dr E. Hendrickson (University of Minnesota Medical School, Minneapolis, MN).13 Clonogenic survival assays. Clonogenic survival assays were performed as described previously.11 Cells were plated in complete McCoy’s 5A medium (ATCC) into 60-mm culture dishes in triplicate per data point and irradiated using a 137Cs irradiator (J.L. Shepherd and Associates, San Fernando, CA) at room temperature at a dose rate of 3.6 Gy/min. After 14 days, colonies were stained with 0.5% crystal violet solution as described previously.11 Survival fraction was calculated as:
Surgery August 2013 (colonies counted)/ [(cell seeded) 3 (plating efficiency)]. Plating efficiency was defined as: (colonies counted in non-irradiated control cells)/(cells seeded). All experiments were repeated $3 times (each time in triplicate). In vivo studies. All animal experiments were performed in accordance with institutional guidelines from Institutional Animal Care and Use Committee at the University of Texas Southwestern Medical Center for animal welfare. For tumor growth experiments, a total of 1 3 106 cells were injected subcutaneously with a 27-gauge needle into both left and right posterior flanks of 6-week-old, female, athymic nude mice (nu/nu, National Cancer Institute, Bethesda, MD). Once the animals developed palpable tumors (50–100 mm3), each group (n = 9) was assigned randomly to receive ionizing radiation on 1 side. The contralateral side received no radiation and served as control for each mouse. All tumors were subjected to external beam radiation (2 Gy every other day 3 5 [total, 10 Gy]) as described previously.11,14 Specifically, mice were irradiated using an X-RAD 320 irradiator (Precision X-Ray, Inc, North Branford, CT) at 10 Gy/min for 0.2 minute. Lead blocks were used to shield the non-tumor parts of the mice as well as the contralateral tumor that served as control. Tumor response to a given treatment was assessed over 6 weeks with twice-weekly, bidimensional measurements by means of a vernier caliper, utilizing the longest axis of the tumor and its perpendicular, and tumor volumes were estimated by multiplying the length of the tumor by the width squared by an investigator blinded to the treatment, as described previously.11,14 The response to ionizing radiation was evaluated by comparing the non-irradiated with the irradiated side for each group and expressed as percentage of response according to the following formula. For each group (n = 9), the average tumor size was determined at each data point. The average tumor size in the irradiated limb of the mice was divided by the average tumor size of the nonirradiated limb of the mice. This approach provided the percent in tumor change between the irradiated tumor and the non-irradiated limb. To determine the percent of tumor regression this ratio was subtracted from one:
ionizing radiation response ð%Þ ¼ 1:0 Tumor volume mm3 ðionizing radiation-treated xenograftÞ O Tumor volume mm3 ðcontrol xenograftÞ 100%:
Surgery Volume 154, Number 2
Western blot analysis. For each experiment, protein extracts of HT-29 and HCT-116 WT as well as KO cells were prepared from either non-irradiated cells or irradiated cells at 72 hours after exposure to 4 Gy of ionizing radiation. Western blots were performed with an equal amount of 25-mg protein extracts as described previously.11 Mouse anti-p53 and anti-actin (Santa Cruz Biotechnology, Santa Cruz, CA) and rabbit anti-Bax, anti-p21, anti-Bcl-2, anti-survivin, anti–PARP-1, and anti-caspase 3 (Cell Signaling Technology, Beverly, MA) antibodies were used at a dilution of 1:1,000. After overnight exposure to the first antibody, all experiments were incubated with a second antibody at a 1:4,000 dilution (antiimmunoglobulin G; Bio-Rad, Hercules, CA). All Western blots were repeated $3 times. Relative intensity were assessed by densitometric analysis of digitized autographic images, using the public domain NIH Image J Program (available online). Assessment of xenograft necrosis. For HCT-116 WT and HCT-116 Bax KO mice, a complete necropsy was performed. After tissue processing and embedding in paraffin, tissue blocks were cut into 3- to 4-mm slices. Hematoxylin and eosin-stained sections were examined for tumor histomorphology and the amount of tumor necrosis. The amount of tumor necrosis (percentage of tumor necrosis) was calculated by dividing the area of tumor necrosis by the total area of the tumor and multiplying by 100. A board certified pathologist (PK) who was unaware of the treatments the mice had received performed all gross examinations, histomorphologic evaluations, and tumor necrosis estimations. Assessment of microvessel density. For HCT-116 WT and HCT-116 Bax KO tumors, immunohistochemical staining for CD31 was performed to highlight endothelial cells of microvessels using standard immunoperoxidase technique with anti-CD31 antibody (1:30, rabbit polyclonal; Thermo Fisher Scientific, Pittsburgh, PA). Hematoxylin and eosin-stained sections of tumors were used to choose 1 paraffin-embedded tissue block representative of the invasive carcinoma and only one 5-mm-thick section was stained for CD31. Intratumoral microvessel density was assessed by light microscopic analysis for areas of the tumor that contained the most capillaries and small venules (microvascular ‘‘hot spots’’). After the area of greatest neovascularization was identified, and individual microvessel counts were made on a 2003 field (203 objective and 103 ocular, Nikon Eclipse E400 microscope, 0.92 mm2 per field). Results were expressed as the greatest number of microvessels in any single 2003 field.
Huerta et al 145
Enzyme-linked immunosorbent assay: Vascular endothelial growth factor. HCT-116 WT and Bax KO cells were plated in 60-mm dishes at a density of 4 3 105 cells per dish. The next day, cells were irradiated at 0 and 4 Gy of ionizing radiation, after which the medium was replaced with fresh medium. Conditioned media from cultured cells were collected after 24 hours and centrifuged at 1,000g to remove dead cells and debris. An enzyme-linked immunosorbent assay kit (Human VEGF Quantikine kit, R&D system) was used to measure total vascular endothelial growth factor (VEGF) concentrations in the culture media according to the manufacturer-recommended protocol. Pan-caspase inhibition (z-VAD-fmk). HCT-116 WT and Bax KO cells were plated in 60-mm dishes at a density of 1 3 106 cells per dish. The next day, cells were treated with 20 mmol/L of pan-caspase inhibitor z-VAD-fmk (EMD Chemicals, Inc, Darmstadt, Germany) for 3 hours before ionizing radiation. Clonogenic survival assays were performed as described. Apoptosis-inducing factor small interfering RNA. Both control and apoptosis-inducing factor (AIF)-specific small interfering (si)RNAoligos (siGENOME Non-Targeting siRNA Pool and siGENOMESMARTpool AIF siRNA) were purchased from DharmaconRNAi Technologies (Lafayette, CO). siRNA transfections using DharmaFECT transfection reagent were done in 6-well format according to the manufacturer’s instructions. Briefly, the day before transfection, both HCT116 WT and Bax KO cells were plated at a density of 1 3 105 cells per well in antibiotic-free medium. The next day, 80 pmol (4 mL of a 20 mmol/L stock) siRNAs were incubated with 2 mL DharmaFECT 2 in 400-mL serum-free medium for 20 minutes. The complexes of DharmaFECT 2 and siRNAs were then added with 1.6 mL of antibiotic-free complete medium and applied directly to each well. After 72 hours of incubation, both control siRNA and AIF siRNA treated HCT-116 WT and Bax KO cells were subject to clonogenic assays. Statistical analysis. PRISM statistical analysis software (GraphPad Software, Inc, San Diego, CA) was used for statistical analysis. Data are expressed as mean values ± standard error of the mean. Difference in cell survival (clonogenic assays) was analyzed by 1-way analysis of variance (ANOVA). The difference in tumor load in tumor xenografts receiving ionizing radiation compared with control was also evaluated by ANOVA. Differences between groups assessed by densitometry analysis derived from Western blot analysis in
146 Huerta et al
Fig 1. Clonogenic survival assays were performed on single-cell suspension of mid-log growing cells. HT-29 cells were highly radioresistant, whereas HCT-116 DNA-PKcs/ cells were exquisitely sensitive to ionizing radiation. Each data point demonstrates the average of 3 independent experiments ± the standard error of the mean (*P # .05 compared with HCT-116 wt at SF-4).
HCT-116 wt and HCT-116 KO cells were compared by Student’s t test. Student’s t test was also used for comparison between 2 groups for analysis of necrosis and microvessel density as well as assessment of VEGF between radiation-treated compared with untreated cells. Difference in cell survival in pancaspase and siRNA cells was assessed by ANOVA. Bonferroni comparison posttest was used to identify significant differences (P # .05) in treatment effects. RESULTS HCT-116 Bax/ cells have a radiosensitive phenotype. Clonogenic studies demonstrated a radioresistant phenotype in (A) HT-29 cells versus HCT-116 WT cells (surviving fraction at 4 Gy [SF-4]: 0.50 vs 0.17; P # .001) and (B) HCT-116 p53/ cells versus HCT-116 WT cells (SF-4, 0.22 vs 0.17; P # .05). These studies also showed a radiosensitive phenotype in (A) HCT-116 DNAPKcs/ versus WT cells (SF-4, 0.04 vs 0.17; P # .001); (B) HCT-116 p21/ versus WT (SF-4, 0.10 vs 0.17; P # .001); and (C) HCT-116 Bax/ cells versus HCT-116 WT cells (SF-4, 0.12 vs 0.17; P # .001; Fig 1). In vivo study results parallel the response to ionizing radiation in vitro. Our in vivo results were consistent with the in vitro data. HT-29 xenografts were highly resistant to ionizing radiation. At the end of the treatment (day 46), HT-29 xenografts experienced only a 14% decrease in tumor volume. WT HCT-116 xenografts demonstrated a 32% decrease in tumor volume (on day 46).
Surgery August 2013
Fig 2. Assessment of tumor growth in immunocompromised mice. Each group consisted of mice bearing 2 identically implanted xenografts in both limbs (n = 9 per group). One side was treated with ionizing radiation, whereas the contralateral side served as control. The average tumor volume of the ionizing radiation-treated group was divided by the average tumor volume of the control groups and expressed as the percentage of tumor response to treatment. HT-29 tumors demonstrated a poor response to ionizing radiation. Xenografts with isogenic HCT-116 cells were most radiosensitive if deficient in PKcs, Bax, and p21 and relatively radioresistant if deficient in p53 when compared with HCT-116 WT tumors (*P # .05 compared with HCT-116 wt at the end of each treatment).
Similarly, HCT-116 p53-null tumors were relatively radioresistant in vivo compared with their WT counterparts (19% tumor reduction on day 43). HCT-116 p21/ xenografts had 40% decrease in volume at the end of the treatment (on day 50). The most radiosensitive xenografts originated from HCT-116 DNA PKcs/ cells with a response of 60% (on day 46) followed by HCT-116 Bax/ xenografts, which had a decrease to 58% (on day 46). All P values at the end of treatment were #.05 versus HCT-116 WT xenografts (Fig 2). Protein levels of AIF are increased in Bax-deficient cells. Western blot analysis showed that, after treatment with ionizing radiation of HCT-116 Bax KO cells, there was a similar increase in the levels of p53 in these cells compared with HCT-116 WT cells. This response was also followed by a similar augmentation of p21 after treatment with 4 Gy of ionizing radiation in both WT and HCT-116 Bax KO cells. PARP-1 cleavage was more pronounced in HCT-116 Bax-deficient cells compared with WT. The increased levels of PARP-1 were unaccompanied with an increase in the levels of cleaved caspase-3 in Bax KO cells. HCT-116 WT cells demonstrated an increase in cleaved caspase-3 after treatment with ionizing
Surgery Volume 154, Number 2
Fig 3. Western blot analysis of HCT-116 WT and Bax KO cells. Densitometry analysis is shown below each line. There is a similar response in p53 and p21 in both isogenic cells. Bcl-2 was downregulated after 4 Gy ionizing radiation in HCT-116 WT cell and absent in HCT-116 Bax/ cells. PARP-1 was markedly increased in HCT-116 Bax-deficient cells, which was accompanied by corresponding increase in AIF, but not cleaved caspase-3.
radiation. The levels of cleaved caspase-3 did not increase after treatment with ionizing radiation in Bax-null cells. The levels of AIF were increased in HCT-116 Bax deficient cells compared with its WT counterpart (Fig 3). Bax-deficient xenografts demonstrate a decrease in angiogenesis and VEGF. Tumor necrosis was 66 ± 7% vs 55 ± 10% in HCT-116 WT ionizing radiation-treated versus control xenografts (P # .05). There was more necrosis in HCT-116 Bax/ xenografts both in the control group (78 ± 4%) and the ionizing radiation-treated tumors (86 ± 2%) relative to the WT counterparts (both P # .05). Compared with untreated HCT116 xenografts, there was an increase in tumor necrosis in HCT-116 Bax/ xenografts that were subjected to ionizing radiation (P # .05; Fig 4). Microvessel density was decreased in both treated and control HCT-116 Bax/ xenografts compared with HCT-116 WT tumors (22 ± 3% and 23 ± 3% vs 31 ± 2% and 32 ± 2%, respectively; P # .05 for both sets of comparisons; Fig 4). Bax/ tumors had a lesser level of secreted VEGF both in control cells (298 ± 15 pg/mL) and ionizing radiation-treated cells (129 ± 6 pg/mL) compared with HCT-116
Huerta et al 147
Fig 4. Assessment of necrosis and intratumoral microvessel density (iMVD) in Xenografts. HCT-116 WT (n = 9) and HCT-116 Bax KO (n = 7) xenografts were collected for assessment of microscopic necrosis. The Bax KO genotype demonstrated more tissue necrosis after treatment with ionizing radiation compared with untreated Bax-deficient tumors (*P # .001). There was no difference in the wild-type HCT-116 xenografts after treatment with ionizing radiation. There was a decrease in iMVD in both treated and untreated Bax-deficient xenografts compared with HCT-116 WT tumors (*P # .05). No difference in iMVD was observed in either WT or Bax KO tumors after treatment with ionizing radiation.
WT cells (P # .05 for both sets of comparisons; Fig 5). Pan-caspase inhibition influences radiosensitivity in HCT-116 WT cells, but not in Bax-deficient cells. We selected colony-forming assays to determine the role of caspases in HCT-116 Bax KO compared with WT. Our results demonstrated that HCT-116 WT cells treated with the pan-caspase inhibitor (z-VAD-fmk) were more radioresistant compared with control group SF-2 (0.48 ± 0.14 vs 0.43 ± 0.01, respectively; P # .001) and SF-4 (0.16 ± 0.01 vs 0.12 ± 0.01, respectively; P # .0001). In contrast, HCT-116 Bax/cells demonstrated no difference whether treated or untreated with this pan-caspase inhibitor (Fig 6). AIF siRNA treatment shows a difference in survival in HCT-116 Bax KO cells, but not in WT cells. Given the finding that AIF was upregulated in HCT-116 Bax/ cells, we hypothesized that a caspase-independent apoptosis pathway may account for the more radiosensitive phenotype in HCT-116 Bax/ cells. We examined this hypothesis by transient inhibition of AIF expression followed by clonogenic assays to study the survival of these cells. There was no difference in response to
148 Huerta et al
Fig 5. Assessment of vascular endothelial growth factor (VEGF) in vitro. There was a decrease in VEGF in both ionizing radiation-treated and untreated Bax-deficient cells compared with WT cells. In HCT-116 WT cells, there was a 1.7-fold change in ionizing radiation treated versus untreated cells. In HCT-116 deficient cells a 2.3-fold decrease in VEGF was observed in ionizing radiationtreated compared with untreated cells (*P # .05).
ionizing radiation in HCT-116 WT cells treated with this siRNA. In contrast, HCT-116 Bax KO cells were more radioresistant after AIF-siRNA treatment compared with untreated cells (SF-2, 0.38 ± 0.00 vs 0.31 ± 0.01 [P # .0001]; SF-4, 0.07 ± 0.01 vs 0.05 ± 0.01 [P # .001]; Fig 7). DISCUSSION This study was designed to investigate the relationship, within the same experimental protocol, of p21, p53, Bax, and DNA-PKcs in response to ionizing radiation in our in vitro and in vivo models of rectal cancer. Understanding the role of these molecules in response to ionizing radiation is of clinical importance in identifying phenotypic differences that might predict tumor response to neoadjuvant therapy in patients affected by rectal cancer. Three recent, independent studies have corroborated the observations of Habr-Gama et al4 in the safe approach of observing rectal cancer patients who achieve a cCR after neoadjuvant treatment.3,5,6 The approach of nonoperative management in patients who demonstrate a cCR after chemoradiation mandates careful selection of patients who might not respond to this modality. Thus, understanding tumor genotype that determines a response to ionizing radiation is of clinical importance. Because p21, p53, and Bax play a
Surgery August 2013
Fig 6. Pan-caspase inhibition. The pan-caspase inhibitor z-VAD-fdk was used to treat cells for $3 hours before ionizing radiation. After treatment with ionizing radiation, cells were plated for clonogenic survival assays as described in the Methods. Pan-caspase inhibition resulted in a more radioresistant phenotype in HCT-116 WT cells compared with untreated cells. Pan-caspase inhibition had no effect on the survival of HCT-116 Baxdeficient cells. Each data point demonstrates the average of 3 independent experiments ± the standard error of the mean (**P # .01 and ***P # .001 HCT-116 WT cells treated with z-VAD-fmk versus HCT-116 WT cells that did not receive z-VAD-fmk).
central role in radiation-induced cell death and because the current literature demonstrates inconclusive results regarding their clinical role as biomarkers, we undertook a systematic analysis of cells deficient in p21, p53, and Bax to compare the response to ionizing radiation with a well-known radiosensitive cell line (HCT-116 DNA-PKcs/) and a radioresistant cell clone (HT-29). Our results corroborate previous findings with regard to the role of p21, p53, and DNAPKcs in their response to ionizing radiation and demonstrate that a Bax-KO phenotype renders a radiosensitive phenotype in HCT-116 cells and xenografts. Our results also point to a potential mechanism leading to this effect. Consistent with previous observations, HCT-116 DNA-PKcs/ cells were sensitive to ionizing radiation.13 For instance, DNA-PKcs–deficient Chinese hamster ovary cells showed profound cell death after treatment with ionizing radiation compared with the DNA-PKcs complimented V3-YAC cells.15 Our observations in this study suggest that the status of tumors with regard to DNA-PKcs might be used as a marker for radiosensitivity. Similarly, HCT-116 p21/ cells are more radiosensitive to ionizing radiation compared with isogenic WT cells.9,14 We observed irradiated
Surgery Volume 154, Number 2
Fig 7. AIF-siRNA treatment. A, Cells were treated with siRNA specific for AIF for 72 hours. After siRNA treatment, all cells were subjected to ionizing radiation and plated for clonogenic survival assays. Transient siRNA inhibition of AIF led to a more radioresistant phenotype in HCT-116 Bax-deficient cells compared with HCT-116 WT cells. Each data point demonstrates the average of 3 independent experiments ± the standard error of the mean (**P # .01 and ***P # .001 HCT-116 Bax/ cells treated with AIF-siRNA versus HCT-116 WT cells that did not receive AIF-siRNA). B, Knock down of AIF was corroborated by Western blot analysis.
HCT-116 p21/ cells to decrease survival after treatment with ionizing radiation compared with the HCT-116 WT clone in vitro. Mice bearing irradiated p21-deficient xenografts also had decreased tumor growth compared with irradiated HCT-116 WT xenografts at the end of the treatment period in our in vivo study. Thus, rectal tumors bearing mutations of this molecule might demonstrate more radiosensitivity compared with wild-type cancers. Our results showed that compared with HCT-116 WT cells and xenografts, p53 deficiency conferred cell survival and resistance to ionizing radiation. This finding is consistent with previous reports that examined the effects of ionizing radiation in HCT-116 WT and p53/ cells.10 p53 plays an essential role in cellular response to radiation in vitro; that is, WT p53 renders a radiosensitive phenotype in cultured cells. Ex vivo studies have failed to provide predictive or prognostic information as a result of the low number of subjects included in the studies, techniques utilized, inability of p53 antibodies to recognize the mutated versus the WT form of p53, or a combination of these. This finding might explain the conflicting results of the literature in utilizing p53 as a biomarker in tumor response to chemoradiation and underscored the need for further investigation on this subject.8,16
Huerta et al 149
Because Bax is a pro-apoptotic protein, we anticipated that a Bax KO status would render HCT-116 cells radioresistant. Contrary to the pro-apoptotic function of Bax, we found that HCT-116 cells deficient in this gene were highly radiosensitive in vitro and in vivo. Although a few studies demonstrate that Bax-deficient cells are resistant to chemotherapeutic agents,17-19 evidence indicating the response of Bax-deficient colorectal cancer cells to ionizing radiation in preclinical studies is lacking. Limited ex vivo studies have shown that tumors expressing Bax had a positive response to chemoradiation compared with patients treated for rectal cancer who did not show Bax staining in their tumors.7,20 Preclinical studies showed that Bax-deficient HCT-116 cells were slightly resistant to apoptotic cell death after treatment with 5-fluorouracil. In contrast, whereas HCT-116 WT cells demonstrated substantial apoptosis to nonsteroidal anti-inflammatory drugs, HCT-116 Bax KO cells showed no apoptosis when exposed to the same drug.19 This study did not assess the response of HCT-116–deficient cells to ionizing radiation. Our study shows unambiguously that HCT-116, Bax-deficient cells are more sensitive to ionizing radiation in vitro and in vivo compared with the isogeneic counterpart HCT-116 WT clone. Thus, rectal tumors demonstrating lack of immunoreactivity to Bax might prove to be more radiosensitive compared with Bax-positive tumors. Because the results with respect to Bax were unexpected, we determined whether there was a specific pathway leading to this response. To assess the mechanism that led to a more radiosensitive phenotype in HCT-116 Bax-deficient cells, we inspected the tumor xenografts from both WT and Bax KO origin and found an increase in the degree of necrosis and a decrease in angiogenesis both in treated and untreated tumor xenografts that originated from the Bax-deficient cells. We corroborated these findings in vitro by demonstrating a decrease in VEGF in HCT-116 Bax deficient compared with HCT-116 WT cells. Whereas a decrease in VEGF led to a decrease in microvessel count and an increase in necrosis in tumors originating from HCT-116 cells deficient in Bax, this mechanism did not explain the parallel observation that we showed in vitro. We observed an increase in PARP-1 in HCT-116 Bax/ cells after treatment with ionizing radiation compared with HCT-116 WT cells. A parallel increase in cleaved caspase-3, however, did not occur. Thus, our in vitro findings did not demonstrate a clear pathway leading to cell death in Bax-deficient cells
150 Huerta et al
with regard to cell cycle arrest, necrosis, or caspasemediated apoptosis. Because colony formation assays assess all parameters related to cell growth, including apoptosis, cell cycle, and necrosis, these assays are ideal to examine the relative effects of an intervention, over time, in cell survival after treatment with ionizing radiation. We performed clonogenic assays to determine the mechanisms that led to a more radiosensitive phenotype in Bax-deficient cells. Because Western blot analysis suggested that AIF was increased moderately in Bax KO compared with WT cells and because caspase-3–mediated apoptosis was not detected in our studies, we studied the effects of pan-caspase inhibition in HCT-116 Bax-deficient and WT cells. We showed that HCT-116 cells treated with the pancaspase inhibitor (z-VAD-fmk) had a more radioresistant phenotype, although there was no difference in survival in Bax-deficient cells after the same form of treatment, indicating that caspase-mediated apoptosis was not the mechanism of increased radiosensitivity. We then proceeded with transient inhibition of AIF followed by clonogenic studies in cells subjected to various doses of ionizing radiation as well as in untreated cells, which demonstrated a more pronounced radioresistant phenotype in Bax-deficient compared with WT cells. Our results suggest that the mechanism of ionizing radiation treatment in Bax-deficient cells may be mediated, in part, by AIF. Because our results do not show a marked difference in Bax WT and Bax KO cells, AIF cannot be the sole contributor to the radiosensitivity in these observations. These findings point to a multifactorial effect that might include AIF but not caspases. Further studies are required to further shed light into this pathway. In the present study, we revealed that Bax deficiency in HCT-116 cells led to radiosensitivity in vitro and in vivo in colorectal cancer. In vivo, this response was associated with a decrease in angiogenesis. In vitro, we showed that AIF might play a role in cell death compared with caspasemediated mechanisms as RNA inhibition of AIF led to survival advantage in HCT-116 Bax/, but not in HCT-116 WT cell treated under the same conditions. Bax may be a potential biomarker to select patients with rectal cancer who might respond well to neoadjuvant treatment. The authors thank Derrick Chen from the University of Texas Southwestern (UTSW) Medical School for his assistance with assessment of necrosis and microvessel density in xenografts. We thank Monica Lu from UTSW
Surgery August 2013
Medical School for her editorial assistance. This work was supported by funds from the Department of Surgery at UTSW. The authors acknowledge the assistance of the Southwestern Small Animal Imaging Resource, which is supported in part by NCI U24 CA126608, the Harold C. Simmons Cancer Center through an NCI Cancer Center Support Grant, 1P30 CA142543-01, and The Department of Radiology.
REFERENCES 1. Huerta S, Hrom J, Gao X, Saha D, Anthony T, Reinhart H, et al. Tissue microarray constructs to predict a response to chemoradiation in rectal cancer. Dig Liver Dis 2010;42: 679-84. 2. Maas M, Nelemans PJ, Valentini V, Das P, Rodel C, Kuo LJ, et al. Long-term outcome in patients with a pathological complete response after chemoradiation for rectal cancer: a pooled analysis of individual patient data. Lancet Oncol 2010;11:835-44. 3. Dalton RS, Velineni R, Osborne ME, Thomas R, Harries S, Gee AS, et al. A single-centre experience of chemoradiotherapy for rectal cancer: is there potential for nonoperative management? Colorectal Dis 2012;14:567-71. 4. Habr-Gama A, Perez RO, Nadalin W, Sabbaga J, Ribeiro U Jr, Silva e Sousa AH Jr, et al. Operative versus nonoperative treatment for stage 0 distal rectal cancer following chemoradiation therapy: long-term results. Ann Surg 2004;240:711-7. 5. Maas M, Beets-Tan RG, Lambregts DM, Lammering G, Nelemans PJ, Engelen SM, et al. Wait-and-see policy for clinical complete responders after chemoradiation for rectal cancer. J Clin Oncol 2011;29:4633-40. 6. Smith JD, Ruby JA, Goodman KA, Saltz LB, Guillem JG, Weiser MR, et al. Nonoperative management of rectal cancer with complete clinical response after neoadjuvant therapy. Ann Surg 2012;256:965-72. 7. Kuremsky JG, Tepper JE, McLeod HL. Biomarkers for response to neoadjuvant chemoradiation for rectal cancer. Int J Radiat Oncol Biol Phys 2009;74:673-88. 8. Huerta S, Gao X, Saha D. Mechanisms of resistance to ionizing radiation in rectal cancer. Expert Rev Mol Diagn 2009;9:469-80. 9. Tian H, Wittmack EK, Jorgensen TJ. p21WAF1/CIP1 antisense therapy radiosensitizes human colon cancer by converting growth arrest to apoptosis. Cancer Res 2000;60:679-84. 10. Cook T, Wang Z, Alber S, Liu K, Watkins SC, Vodovotz Y, et al. Nitric oxide and ionizing radiation synergistically promote apoptosis and growth inhibition of cancer by activating p53. Cancer Res 2004;64:8015-21. 11. Gao X, Saha D, Kapur P, Anthony T, Livingston EH, Huerta S. Radiosensitization of HT-29 cells and xenografts by the nitric oxide donor DETANONOate. J Surg Oncol 2009; 100:149-58. 12. Shibata D, Peinado MA, Ionov Y, Malkhosyan S, Perucho M. Genomic instability in repeated sequences is an early somatic event in colorectal tumorigenesis that persists after transformation. Nat Genet 1994;6:273-81. 13. Ruis BL, Fattah KR, Hendrickson EA. The catalytic subunit of DNA-dependent protein kinase regulates proliferation, telomere length, and genomic stability in human somatic cells. Mol Cell Biol 2008;28:6182-95. 14. Huerta S, Gao X, Saha D. Murine orthotopic model for the assessment of chemoradiotherapeutic interventions in rectal cancer. Anticancer Drugs 2011;22:371-6.
Surgery Volume 154, Number 2
15. Zhao Y, Thomas HD, Batey MA, Cowell IG, Richardson CJ, Griffin RJ, et al. Preclinical evaluation of a potent novel DNA-dependent protein kinase inhibitor NU7441. Cancer Res 2006;66:5354-62. 16. Huerta S, Goulet EJ, Huerta-Yepez S, Livingston EH. Screening and detection of apoptosis. J Surg Res 2007; 139:143-56. 17. Wagener C, Bargou RC, Daniel PT, Bommert K, Mapara MY, Royer HD, et al. Induction of the death-promoting gene bax-alpha sensitizes cultured breast-cancer cells to drug-induced apoptosis. Int J Cancer 1996;67:138-41.
Huerta et al 151
18. Yamaguchi H, Bhalla K, Wang HG. Bax plays a pivotal role in thapsigargin-induced apoptosis of human colon cancer HCT116 cells by controlling Smac/Diablo and Omi/HtrA2 release from mitochondria. Cancer Res 2003;63:1483-9. 19. Zhang L, Yu J, Park BH, Kinzler KW, Vogelstein B. Role of BAX in the apoptotic response to anticancer agents. Science 2000;290:989-92. 20. Chang HJ, Jung KH, Kim DY, Jeong SY, Choi HS, Kim YH, et al. Bax, a predictive marker for therapeutic response to preoperative chemoradiotherapy in patients with rectal carcinoma. Hum Pathol 2005;36:364-71.