Putative cancer stem cells (CSCs), which have enhanced DNA damage-repair mechanisms to overcome the DNA-damaging effects of radiograph radiation,1 are implicated in tumor cell resistance to radiotherapy or chemotherapy.2–7 Treatment strategies targeting CSCs may help to overcome this resistance in cancer patients.8 Charged particles with high linear-energy transfer, such as neutrons, may be capable of overcoming CSC DNA repair, but their severe effect on normal tissue renders treatment with these particles problematic for most radiotherapeutic applications. Like any charged particle, protons possess a radiation dose deposition characteristic called the Bragg peak.9 It is because of this property that proton therapy is able to spare normal tissues surrounding the tumor target to a much greater degree than can photon therapy. This characteristic may allow dose intensification for cancers such as non–small-cell lung cancer (NSCLC) and minimize dose-related toxicities to normal tissue.10–12
Because of its promising safety and efficacy profile, proton therapy is operational in 39 cancer treatment centers around the world, and many more centers are under construction or being planned. However, the biological efficacy of proton versus photon (radiograph) therapy is still poorly understood. The ability of protons to exert cytotoxic damage to cells has been considered to be 10% higher than that of photons such that the relative biological effectiveness (RBE) of protons is 1.1, regardless of the cell or tissue type.13,14 This RBE value is used routinely to adjust the clinical radiotherapy dose with protons as compared with photons. However, this adjustment is a gross oversimplification. Because of their charge and mass, protons produce locally higher ionization density regions along their tracks than photons do, producing a diffuse field of ionization through secondary electrons. Furthermore, neutrons produced by high-energy protons have a relatively high RBE. These factors may be responsible for observed differences in the sublethal damage:cell kill ratio between protons and photons and suggest that radiation-resistant cells may be more likely to be killed by protons than photons.15,16
We recently reported results from our phase II clinical trials of early-stage and locally advanced NSCLC treated with dose-escalated proton therapy.11 Our data showed that compared with photon therapy, irradiation with protons seems to reduce side effects and have high local control. The latter regimen might have had better clinical outcomes partially by having greater biologic effectiveness than what an RBE of 1.1 suggests. We hypothesize that the differences in the effectiveness were because of how CSC-like cells respond to proton therapy and photon therapy. We tested this hypothesis by using a previously established model of CSC-like cells from paclitaxel-resistant (chemotherapy resistant) NSCLC cell lines.2 We found that protons kill more CSC-like cells than photons do at the same radiation dose.
MATERIALS AND METHODS
Cell Lines and Reagents
Human NSCLC cell lines A549 and H460 were obtained from the American Type Culture Collection and routinely maintained in Roswell Park Memorial Institute (RPMI)-1640 medium (Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine serum, 10,000 U/ml of penicillin–streptomycin, and 2 mmol/liter of glutamine. Normal human bronchial epithelial (NHBE) cells were purchased from Clonetics Corporation (Walkersville, MD) and cultured as recommended by the manufacturer. The identities of these cell lines were validated by short tandem repeat profiling (performed by the Characterized Cell Line Core of The University of Texas MD Anderson Cancer Center) using the AmpFlSTR Identifiler polymerase chain reaction amplification kit (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. The short tandem repeat profiles for these cell lines matched their known American Type Culture Collection fingerprints.
Generation of Treatment-Resistant Cell Lines
We established two CR- and CSC-enriched NSCLC cell lines, H460/CR and A549/CR, by repeatedly treating parental H460 and A549 cells with paclitaxel to induce chemoresistance, as described previously.2 The cells were initially treated with 5 nM paclitaxel, and the surviving cells were then treated with increasing doses of up to 100 nM.
Stem cells can efflux Hoechst 33342 (Life Technologies) dye by means of verapamil-sensitive adenosine triphosphate (ATP)-binding cassette transporters. This classic stem cell characteristic is referred to as having a side population (SP). For each treatment-resistant cell line, SP cells (H460/CR/SP and A549/CR/SP cells) were obtained and sorted serially. We refer to these cells as CSC-like cells. Reanalysis of SP cells from the cultured SP cells demonstrated enrichment of the SP cells and production of non-SP (NSP) cells (H460/CR/NSP and A549/CR/NSP cells).2
Cell irradiation with proton beams was done at the Proton Therapy Center at the MD Anderson Cancer Center. Cells were irradiated at the center of the spread-out Bragg peak modulated to the depth of the cell layer. The irradiation system and biophysical characteristics of proton beams have been detailed elsewhere.11,12 For irradiation with photon beams, γ-rays from the cobalt-60 machine at MD Anderson Cancer Center were used with the same cobalt gray equivalent (CGE) on the same day.
Cell Viability Assay
The viability of parental H460 and A549 cell lines, treatment-resistant H460 and A549 cell lines, and NHBE cells was determined using the sulforhodamine B colorimetric assay 4 days after irradiation. Each experiment was done in quadruplicate and repeated at least twice.
For the clonogenic assay, parental H460 and A549 cell lines and treatment-resistant H460 and A549 cell lines (the latter with or without SP cells) were trypsinized, passed through a 40-μm sieve, and immediately irradiated at room temperature to generate a dose curve of 0, 2, 4, and 8 Gy. Corresponding control cells were sham irradiated. Colony-forming assays were begun immediately after irradiation by replating the cells into triplicate 100-mm culture dishes using the same method described previously.17,18 Colony formation was assessed 14 days after irradiation. To determine survival fractions, counts were normalized using the plating efficiency of the unirradiated corresponding control cells. Each experiment was done in triplicate and repeated at least twice.
Terminal Deoxynucleotidyl Transferase–Mediated Deoxyuridine Triphosphate-Biotin Nick End-Labeling Assay
Apoptosis of parental or treatment-resistant H460 cells (the latter with or without SP cells) was evaluated 48 hours after irradiation by flow cytometry with a terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate-biotin nick end-labeling (TUNEL) kit (Roche Applied Science, Indianapolis, IN) by dual color, as previously described.2,17,19 Three independent experiments were conducted.
Matrigel Cell Migration and Invasion Assay
To determine the invasiveness potential of parental and treatment-resistant H460/CR cells, we used the CytoSelect 24-well cell chemotaxis assay, flourometric format (Cell Biolabs, Inc., San Diego, CA) 48 hours after irradiation. The assay was performed exactly as described by the manufacturer. In brief, cells were starved of serum for 24 hours and subsequently seeded into the upper chamber onto a rehydrated basement membrane covering a Matrigel (BD Bioscineces, San Jose, CA) preparation with a diameter of 8 μm. Cells were allowed to invade toward 10% fetal calf serum for 24 hours. Cells that invaded to the bottom of the membrane were stained and quantified as described in the assay protocol. This assay was repeated three times.
Reactive Oxygen Species Analysis
We used flow cytometry with 2’-7’-dichlorofluorescein diacetate (H2DCF-DA) staining to determine the level of intracellular reactive oxygen species (ROS). Briefly, the day before the assay was performed, H460, H460/CR, H460/CR/SP, H460/CR/NSP, and NHBE cells were seeded at a density of 2 × 105 per well onto 6-well plates overnight. The cells were then irradiated with protons or photons at 0 or 4 Gy at different time points. H2DCF-DA was dissolved in dimethyl sulfoxide and diluted with prewarmed phosphate-buffered saline (PBS) to a final concentration of 5 µM. After treatment with radiation at different time points (from 6 hours to 9 days), the growth medium was replaced with PBS containing H2DCF-DA. After incubation for 40 minutes at 37°C, the cells were returned to a prewarmed growth medium and incubated at 37°C for 10 minutes. Cells were trypsinized and washed with prewarmed PBS once. The samples were then subjected to a flow cytometric assay using a Caliber fluorescence-activated cell sorter (BD Biosciences, San Jose, CA). These were performed at the Flow Cytometry and Cellular Imaging Core Facility at MD Anderson Cancer Center. All experiments were performed at least twice.
Tumor Sphere Formation Assay
H460/CR/SP cells were suspended in 0.8% methylcellulose-based serum-free medium supplemented with 20 ng/ml epidermal growth factor, 20 ng/ml basic fibroblast growth factor, and 4 µg/ml insulin and plated at 2000 cells/ml in ultralow-attachment 24-well plates (Corning, Tewksbury, MA). The medium was replaced or supplemented with fresh growth factors twice a week. To assess the self-renewing potential of the cells, first-generation spheres were collected by gentle centrifugation, dissociated into single-cell suspensions, and filtered and cultured under the stem cell–selective conditions. The same procedures were repeated with the second- and third-generation spheres. Third-generation spheres were used in our experiments. A total of 1000 cells per ml was plated in ultralow-attachment 96-well plates (Corning, Tewksbury, MA) in 200 µl of growth medium cultured under the stem cell–selective conditions. The cells were then irradiated with protons or photons at 0 or 4 Gy. After 2 weeks of culture, tumor spheres were visualized with a phase-contrast microscope and counted in each well as described previously.2
Flow Cytometric Assay of Coxsackievirus and Adenovirus Receptor Expression, β-Catenin Expression, and SP Cells
Ninety-six hours (4 days) after irradiation of H460/CR/SP cells at 4 Gy, we used flow cytometry to determine coxsackievirus and adenovirus receptor (CAR) and β-catenin expression levels as well as the percentage of SP cells.2,17,19
Data are expressed as means and 95% confidence intervals, some are expressed as means ± SD. Analysis of variance and Student’s t tests (two-tailed) were performed to test for significant differences. A difference was considered statistically significant when the p value was 0.05 or less.
Proton Therapy Decreases Cell Viability More So Than Photon Therapy Does
To detect differences in the effects of the proton and photon modalities, we determined cell viability by sulforhodamine B colorimetric assay in NSCLC cell lines H460, H460/CR, A549, A549/CR and in NHBE cells on the fourth day after treatment with radiation doses up to 8 Gy at the same CGE (RBE = 1.1). We observed significantly weaker cell-killing effects of photons versus protons in the treatment-resistant cell lines (H460/CR and A549/CR) but not in the parental nonresistant cell lines or the NHBE cells (Fig. 1).
Proton Therapy Induces Lower Clonogenic Survival Than Photon Therapy Does
To determine whether proton therapy affects clonogenic survival in cancer cells differently than photon therapy does, we performed clonogenic formation assays with parental and treatment-resistant cell lines as well as with the putative stem cell fraction of the resistant cells identified as SP by Hoeschst dye exclusion. On the 14th day after irradiation with a dose of up to 8 Gy, colony formation of CSC-like cells (H460/CR/SP and A549/CR/SP cells) was significantly less with protons than photons (Fig. 2).
We also found that compared with photon radiation, clonogenic survival was significantly lower in treatment-resistant cells irradiated with protons at 4 Gy (p < 0.05 for both lines). However, no significant differences were seen for parental nonresistant cells or for resistant cells with NSPs. Taken together, these results indicate that protons better target CSC-like cells than photons do.
Proton Therapy Induces Higher Levels of Apoptosis Compared with Photon Therapy
To investigate whether the enhanced cell killing by proton radiation occurred through an apoptotic mechanism, we quantified apoptotic cells using the TUNEL assay in H460, H460/CR, H460/CR/SP, and H460/CR/NSP cells at 48 hours after radiation. Both modalities induced marked increases in apoptosis in a dose-dependent manner in all four cell lines (Fig. 3A). However, protons induced significantly greater levels of apoptosis in H460/CR cells and in CSC-like cells (H460/CR/SP cells) than photons did.
Proton Therapy Reduces Migration and Invasiness of CSC-Like Cells More Than Photon Therapy Does
To assess whether the differential effects of proton and photon irradiation on apoptosis in CSC-like cells we observed were reflected by cell motility, we examined the migration of malignant cells 48 hours after irradiation using the chemotaxis assay in H460, H460/CR, H460/CR/SP, H460/CR/NSP cells. In general, both proton and photon radiation suppressed migration in a dose-dependent manner, although we noticed a trend toward an increase in migration at lower doses (0.5 and 1 Gy) of photon irradiation (Fig. 3B). Compared with photons, protons at comparable doses significantly reduced migration in the CSC-like H460/CR/SP cells.
We next focused on changes in the invasive capability of cancer cells after irradiation using the Matrigel invasion assay the same four cell lines. The results mirrored our observations for cell migration (Fig. 3C). Thus, compared with photons, protons suppressed both the migration and invasiveness potential of the cancer cells even at lower radiation doses.
Proton Therapy Induces Higher ROS Levels Than Photon Therapy Does
Radiation is cytotoxic because it induces double-stranded DNA breaks from ROS produced from secondary ionization of molecules.20 To determine whether the differential cytotoxic effects of protons and photons in CSC-like cells we observed were because of differences in ROS production, we measured the intracellular concentrations of pro-oxidants by using H2DCF-DA staining by flow cytometry. Baseline measurements showed lower ROS levels in parental H460 cells than in normal NHBE cells (Fig. 4A, left). We also found lower baseline ROS levels in CSC-like H460/CR/SP cells than in H460/CR or H460/CR/NSP cells (Fig. 4A, right).
Given these observations, we were interested to see whether there were differences in the production of ROS after proton or photon radiation. We measured ROS levels in H460, H460/CR, H460/CR/SP, and NHBE cell lines at 48 hours after irradiation with 0, 2, 4, 8 Gy (RBE = 1.1). Both modalities induced ROS levels increases in a dose-dependent manner in all four cell lines. Protons with 4 Gy induced significantly higher ROS levels than photons did in H460/CR cells (3.8-fold versus 2.6-fold) and H460/CR/SP cells (3.4-fold versus 2.1-fold), but not in H460 (3.6-fold versus 3.0-fold) and NHBE cells (1.2-fold versus 1.1-fold). We further measured ROS levels in H460 and H460/CR/SP cells for up to 9 days (216 hours) after irradiation 4 Gy. Protons induced significantly higher ROS levels than photons did in H460/CR/SP cells at almost all time points, but in the parental H460 cells at 7 days (168 hours) only (Fig. 4D). These results suggest that protons induce greater ROS levels in CSCs and thus have great cytotoxicity than photons do.
Proton Therapy Results in Less Tumor Sphere Formation Compared with Photon Therapy
We tested the ability of proton therapy and photon therapy to affect the tumor sphere formation of CSC-like H460/CR/SP cells grown in suspension cultures using serum-free medium under stem cell–selective conditions.2 Both types of radiation significantly inhibited tumor sphere formation, but more so with proton radiation (17 ± 2 versus 32 ± 5; Fig. 5A, B).
Proton Therapy Results in Lower CAR and β-Catenin Expression Levels and Lower Percentage of SP Cells Compared with Photon Therapy in CSC-Like Cells
Finally, we used flow cytometry to measure the expression levels of the CSC markers CAR, β-catenin, and SP 96 hours after irradiation at 4 CGE in H460/CR/SP cells. Only protons significantly reduced the expression of CAR (80.6% versus 92.1%) and β-catenin (20.3% versus 32.6%; Fig. 5C) as well as the percentage of SP cells (39.5% versus 50.1%; Fig. 5D) in CSC-like cells.
In this article, we investigated the biologic differences between proton beam therapy and photon beam therapy and determined whether protons (delivered with an RBE of 1.1) and photons have differential cytotoxic effects on CSC-enriched CSC-like cells as well. What we found was consistent with our previous data that protons kill more CSC-like cells than photons do at the same radiation dose. Regardless of the assay we used (for cell viability, clonogenic survival, apoptosis, cell migration or invasiveness, or tumor sphere formation), our current results indicate that protons have a greater biologic effect than what an RBE of 1.1 seems to indicate. Why CSC-like cells are more likely to be killed by protons is not entirely clear. Our data suggested that ROS generation was significantly higher in CSC-like cells with proton therapy than with photon therapy. Whether this is difference is related to the high ionization density of protons in the cells is also unclear.
ROS regulates a broad array of signal transduction pathways in various biological processes, including gene expression, cell growth, differentiation, and apoptosis. ROS caused by high glucose, angiotensin, tumor necrosis factor-α, irradiation, or other stressors enhances the apoptosis of various tumor cells. Effective management of ROS in normal tissue stem cells may contribute to their fitness and hence the life span of the organism.21,22 But what happens when the machinery for ROS management that is used by normal stem cells is hijacked by CSCs to maintain cellular fitness? The work of Phillips et al.23 showed that cancer-initiation cells are more resistant to radiation than are cells grown as monolayer cultures and that ROS levels are lower in cancer-initiation cells than in monolayer cultures after irradiation. The most recent findings in this area support and extend this concept. Diehn et al.1 demonstrated that human and mouse breast CSCs, similar to their normal tissue counterparts, maintain low levels of ROS that afford radioprotection, providing a possible explanation for tumor recurrence with photon therapy. Consistent with the idea of ROS being critical mediators of ionizing radiation–induced cell killing, CSCs were found to develop less DNA damage than nontumorigenic cells did and were more likely to be spared after photon irradiation. Lower ROS levels in CSCs are associated with increased expression of free radical scavenging systems.1 Our results are consistent with these reports. Further studies are needed to examine gene expression profiles of CSCs and non-CSCs in treatment-resistant NSCLC to determine whether the former have stronger antioxidant defense systems.
Pietro et al.24 demonstrated that compared with photon radiation, proton radiation significantly increased the intracellular level of ROS in CSCs and that it altered cell structures such as membranes, caused DNA double-stranded breaks, and significantly increased apoptosis and intracellular levels of ROS in the MCF7, PC3, and Ca301D cancer cell lines. Finnberg et al.15 further demonstrated that proton therapy triggers DNA damage and apoptosis in vivo more efficiently than photon therapy does and that the apoptotic responses varied greatly between proton and photon therapy in a tissue- and dose-dependent manner. Moreover, the work by Ogata et al.16 suggest that at lower doses, proton irradiation suppresses metastatic potential whereas photon irradiation promotes cell migration and invasive capabilities. The authors also provided preclinical evidence that proton therapy is superior to conventional photon therapy in preventing the metastasis of irradiated malignant tumor cells. Our own data on cell migration and invasion support these observations.
There are several limitations to our study. First, only CSC-like treatment-resistant cell lines generated from two parental cell lines (H460 and A549) were tested. In addition, only H460 cell line series were used for the test of apoptosis, cell migration/invasion, ROS levels, sphere formation, and CSC marker expression. It is possible that different cell lines may have varied sensitivities to proton beam because of different genetic backgrounds. Second, limited radiation doses and therapy designs were conducted because of limited machine time and biological safety issues in the patient treatment facility. We originally planned to use more cell lines and add in vivo study to validate our results. Thus, this report focuses on hypothesis generation; further studies are needed to validate these observations.
We have shown for the first time that treatment-resistant CSC-like cell lines are more sensitive to protons than photons at the same RBE. The greater effects of proton therapy on CSC-like cells could be because of the higher generation of ROS by protons than photons. Assuming that treatment-resistant CSCs have a role in the tumor recurrence, protons may be more effective than photons in eliminating recurrent or persistent NSCLC. Further laboratory studies, particularly in vivo experiments and clinical evaluation of ongoing and planned randomized trials at various disease sites, are needed to confirm this hypothesis.
Financial support was provided by Radiology Society of North America Research Scholar Award (to JYC), Career Development Award from The University of Texas Lung Cancer Specialized Programs of Research Excellence grant from the National Cancer Institute (to JYC), National Cancer Institute grants RO1 CA 092487 and RO1 CA 098582 (to BF), and National Institutes of Health Cancer Center Core Grant CA 16672. This research was supported in part by P01-CA021239 from the National Cancer Institute. MD Anderson Cancer Center is supported by the National Institutes of Health through grant CA16672.
1. Diehn M, Cho RW, Lobo NA, et al. Association of reactive oxygen species levels and radioresistance in cancer stem cells. Nature. 2009; 458:780–783
2. Zhang X, Fang B, Mohan R, Chang JY. Coxsackie-adenovirus receptor as a novel marker of stem cells in treatment-resistant non-small cell lung cancer. Radiother Oncol. 2012; 105:250–257
3. Baumann M, Krause M, Hill R. Exploring the role of cancer stem cells in radioresistance. Nat Rev Cancer. 2008; 8:545–554
4. Brunner TB, Kunz-Schughart LA, Grosse-Gehling P, Baumann M. Cancer stem cells as a predictive factor in radiotherapy. Semin Radiat Oncol. 2012; 22:151–174
5. Dean M, Fojo T, Bates S. Tumour stem cells and drug resistance. Nat Rev Cancer. 2005; 5:275–284
6. Krause M, Yaromina A, Eicheler W, Koch U, Baumann M. Cancer stem cells: targets and potential biomarkers for radiotherapy. Clin Cancer Res. 2011; 17:7224–7229
7. Mihatsch J, Toulany M, Bareiss PM, et al. Selection of radioresistant tumor cells and presence of ALDH1 activity in vitro. Radiother Oncol. 2011; 99:300–306
8. Nguyen GH, Murph MM, Chang JY. Cancer stem cell radioresistance and enrichment: where frontline radiation therapy may fail in lung and esophageal cancers. Cancers (Basel). 2011; 3:1232–1252
9. Chang JY, Zhang X, Wang X, et al. Significant reduction of normal tissue dose by proton radiotherapy compared with three-dimensional conformal or intensity-modulated radiation therapy in Stage I or Stage III non-small-cell lung cancer. Int J Radiat Oncol Biol Phys. 2006; 65:1087–1096
10. Chang JY, Komaki R, Wen HY, et al. Toxicity and patterns of failure of adaptive/ablative proton therapy for early-stage, medically inoperable non-small cell lung cancer. Int J Radiat Oncol Biol Phys. 2011; 80:1350–1357
11. Chang JY, Komaki R, Lu C, et al. Phase 2 study of high-dose proton therapy with concurrent chemotherapy for unresectable stage III nonsmall cell lung cancer. Cancer. 2011; 117:4707–4713
12. Schulz-Ertner D, Tsujii H. Particle radiation therapy using proton and heavier ion beams. J Clin Oncol. 2007; 25:953–964
13. Wambersie A. RBE, reference RBE, and clinical RBE: applications of these concepts in hadron therapy. Strahlenther Onkol. 1999; 175:39–43
14. Paganetti H, Niemierko A, Ancukiewicz M, et al. Relative biological effectiveness (RBE) values for proton beam therapy. Int J Radiat Oncol Biol Phys. 2002; 53:407–421
15. Finnberg N, Wambi C, Ware JH, Kennedy AR, El-Deiry WS. Gamma-radiation (GR) triggers a unique gene expression profile associated with cell death compared to proton radiation (PR) in mice in vivo. Cancer Biol Ther. 2008; 7:2023–2033
16. Ogata T, Teshima T, Kagawa K, et al. Particle irradiation suppresses metastatic potential of cancer cells. Cancer Res. 2005; 65:113–120
17. Zhang X, Komaki R, Wang L, Fang B, Chang JY. Treatment of radioresistant stem-like esophageal cancer cells by an apoptotic gene-armed, telomerase-specific oncolytic adenovirus. Clin Cancer Res. 2008; 14:2813–2823
18. Zhang X, Cheung RM, Komaki R, Fang B, Chang JY. Radiotherapy sensitization by tumor-specific TRAIL gene targeting improves survival of mice bearing human non-small cell lung cancer. Clin Cancer Res. 2005; 11:6657–6668
19. Chang JY, Zhang X, Komaki R, Cheung R, Fang B. Tumor-specific apoptotic gene targeting overcomes radiation resistance in esophageal cancer. Int J Radiat Oncol Biol Phys. 2006; 64:1482–94
20. Riley PA. Free radicals in biology: oxidative stress and the effects of ionizing radiation. Int J Radiat Biol. 1994; 65:27–33
21. Tothova Z, Kollipara R, Huntly BJ, et al. FoxOs are critical mediators of hematopoietic stem cell resistance to physiologic oxidative stress. Cell. 2007; 128:325–339
22. Tothova Z, Gilliland DG. A radical bailout strategy for cancer stem cells. Cell Stem Cell. 2009; 4:196–197
23. Phillips TM, McBride WH, Pajonk F. The response of CD24(-/low)/CD44+ breast cancer-initiating cells to radiation. J Natl Cancer Inst. 2006; 98:1777–1785
24. Pietro CD, Tabbi G, Ragusa M, et al. Cellular and molecular effects of protons: apoptosis induction and potential implications for cancer therapy. Apoptosis. 2006; 11:57–66