.JPG?ad=18507&adview=true)
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Research Articles: Therapeutics, Targets, and Development
Photodynamic therapy mediates the oxygen-independent activation of hypoxia-inducible factor 1
Departments of 1 Imaging Sciences and 2 Microbiology and Immunology, University of Rochester Medical Center, and 3 Department of Physics and Astronomy, University of Rochester, Rochester, New York
Requests for reprints: Thomas H. Foster, Department of Imaging Sciences, University of Rochester, 601 Elmwood Avenue, Box 648, Rochester, NY 14642. Phone: 585-275-1347. E-mail: thomas.foster{at}rochester.edu
Abstract
Photodynamic therapy (PDT) induces the expression of the hypoxia-inducible factor 1
(HIF-1) subunit of the HIF-1 transcription factor and its target genes in vitro and in vivo. PDT also induces the expression of the enzyme cyclooxygenase-2 and its metabolite, prostaglandin E2 (PGE2). PGE2 and hypoxia act independently and synergistically to increase HIF-1 accumulation and nuclear translocation. To examine the expression of HIF-1 target genes in response to PDT-mediated oxidative stress and PGE2 under normoxic conditions, we established EMT6 cells transfected with a plasmid consisting of a hypoxia response element promoter and a downstream gene encoding for green fluorescent protein (GFP). To examine the temporal kinetics of HIF-1 nuclear translocation in response to PDT, we transfected a second line of EMT6 cells with a GFP-tagged HIF-1 fusion vector. Cell monolayers were incubated with 1 µg mL–1 Photofrin for 24 h and irradiated with fluences of 1, 2.5, and 5 J cm–2. Direct measurement of oxygen concentration during irradiation confirmed that cells remained well oxygenated. Cells were also exposed to 1 and 10 µmol/L PGE2 for 3 h. In normoxic conditions, Photofrin, PDT, and PGE2 treatment activated HIF-1 and induced its nuclear translocation. Maximal Photofrin-PDT–mediated HIF-1 activation was intermediate in magnitude between that induced by PGE2 and that by the hypoxia mimic cobalt chloride. This work establishes that PDT induces significant activation of the HIF-1 pathway in the absence of hypoxia and supports the interpretation that the induction of HIF-1 target genes by PDT may be mediated, at least in part, by the prostaglandin pathway. [Mol Cancer Ther 2006;5(12):3268–74]
Introduction
Photodynamic therapy (PDT) is a promising treatment for cancer that is receiving increasing acceptance in the clinic. Health agency approvals in the United States, Europe, and elsewhere have been granted for four different photosensitizing agents, and ambitious clinical trials are under way in a number of important oncologic and nononcologic conditions. A recent review by Brown et al. (1) summarizes the current international status of health agency approvals. PDT involves the localization of a photosensitive drug in the target tissue before illumination using an appropriate wavelength of visible or near-IR light (2). On illumination, a cascade of photochemical events results in the generation of cytotoxic singlet oxygen, which causes tumor destruction directly through tumor cell damage, through microvascular damage and the induction of vascular stasis (3–5), or through a combination of these mechanisms. Various immune responses to PDT have also been described (2, 6–8). The therapy-induced decrease in blood flow leads to strong and persistent tumor tissue hypoxia (4). Further, because the PDT photochemistry consumes oxygen, oxygen consumption during the therapy can also create rapid and severe transient tissue hypoxia (9–11).
Previous studies have examined the effect of PDT-mediated microvascular damage and the resulting hypoxia in initiating a variety of molecular and physiologic responses, including gene activation (12, 13). A primary step in the molecular response to hypoxia is the formation of the hypoxia-inducible factor 1 (HIF-1) complex, a key transcriptional regulator. HIF-1 is a heterodimeric transcription factor consisting of and ß subunits. Regulation of HIF-1 occurs principally through the subunits (HIF-1), which are degraded in the presence of oxygen following their binding to the von Hippel-Landau protein (14–16). In the absence of oxygen, HIF-1 is stabilized and translocates to the nucleus, where it forms a dimer with HIF-1ß, also called the aryl hydrocarbon receptor nuclear translocator. The resulting HIF-1 complex binds to specific sequences within the hypoxia responsive element (HRE) present in oxygen-regulated target genes, ultimately activating transcription of these genes, which encode for erythropoietin, vascular endothelial growth factor (VEGF), and a variety of other proteins (16). However, recent studies by Liu et al. (17) and Fukuda et al. (18) have shown that, independent of hypoxia, HIF-1 protein levels are elevated and undergo increased nuclear accumulation in the presence of prostaglandin E2 (PGE2), which is a metabolite of cyclooxygenase-2 (COX-2). Further, these authors showed that the combination of PGE2 and hypoxia resulted in greater nuclear accumulation of HIF-1 protein levels, therefore implying that PGE2 and hypoxia act both independently and synergistically to increase HIF-1 accumulation and nuclear translocation.
There is evidence in the literature that, under treatment conditions not expected to induce hypoxia, PDT with the photosensitizer Photofrin induced increased expression of VEGF in BA carcinoma tumor cells in vitro (12). Further, under identical treatment conditions in the radiation-induced fibrosarcoma (RIF) tumor cell line, PDT induced increased expression of COX-2 and PGE2 (13). It is therefore feasible that, in the absence of hypoxia, PDT-generated oxidative stress may induce enhanced activation of HIF-1 and its nuclear translocation through the prostaglandin pathway and/or through other signaling pathways initiated by reacting oxygen species (19).
In the current study, we examine the activation of an HRE promoter in response to PDT-mediated oxidative stress and PGE2 exposure using fluorescence imaging of murine tumor cells transfected with a plasmid consisting of the HRE promoter and a downstream green fluorescent protein (GFP) gene. In addition, we investigate whether these same treatment conditions induce the translocation of the HIF-1 from the cytosol to the nucleus. Nuclear translocation of HIF-1 is considered to be an important regulatory step in the transcriptional activation of its target genes. To visualize this process, we establish cells stably transfected with a GFP-tagged HIF-1 fusion vector. Therefore, using fluorescent reporter genes and two transfected cell lines, we explore the role of PDT in regulating HIF-1 translocation/activation and the underlying molecular pathways. We show that in the absence of hypoxia, PDT-mediated oxidative stress activates HIF-1 and induces its nuclear translocation. Consistent with previous reports, we show that, under normoxic conditions, PGE2 exposure also induces similar effects. These findings in conjunction with the reports of PGE2 induction by PDT (13), therefore, are consistent with the hypothesis that the increased expression of the HIF-1 target genes in response to PDT may be mediated, at least in part, by the prostaglandin pathway.
Materials and Methods
Chemicals and Photosensitizer
The photosensitizer Photofrin (Axcan Pharma, Inc., Mont-Saint-Hilaire, Quebec, Canada) was received in powder form as a gift from Steven Hahn (University of Pennsylvania, Philadelphia, PA). PGE2 was obtained from Cayman Chemical (Ann Arbor, MI). FCS was purchased from Atlanta Biologicals (Atlanta, GA). HBSS, Lipofectin reagent, and G-418 were purchased from Invitrogen (Carlsbad, CA). Unless otherwise stated, all other chemicals and reagents were obtained from Sigma Chemical Co. (St. Louis, MO).
Cell Lines
We established a mouse mammary cancer EMT6 cell line stably transfected with a plasmid containing the gene for destabilized EGFP (Clontech, Palo Alto, CA) placed under the control of a promoter region consisting of five copies of a 35-bp fragment from the HRE of the human VEGF gene and a human cytomegalovirus (CMV) minimal promoter. The destabilized EGFP has a half-life of 
2 h. The plasmid was kindly provided by J. Martin Brown (Stanford University, Palo Alto, CA). The characterization of the plasmid and the methods for its construction have been described in detail by Shibata et al. (20) and Vordermark et al. (21). This reporter plasmid was transfected into EMT6 cells using Lipofectin reagent. To select stably transfected clones, cells were grown in Basal Eagle Medium supplemented with 10% FCS containing the antibiotic G-418 at a concentration of 400 µg mL–1. G-418-resistant clones were then expanded and screened for GFP expression by exposure to cobalt chloride (CoCl2)–simulated hypoxia as a positive inducible treatment. A single clone exhibiting high CoCl2-induced expression and low basal levels of GFP fluorescence was selected for subsequent studies, and the cells derived from this clone are referred to as 5HRE-GFP/EMT6.
We also established an EMT6 cell line stably transfected with a GFP-tagged HIF-1 fusion protein vector to visualize the nuclear accumulation of HIF-1 in response to PGE2 exposure and PDT-induced oxidative stress. The fusion protein vector was a gift from Alice Levine (Mount Sinai Medical School, New York, NY) and was prepared by fusing HIF-1 cDNA to the pEGFP (Clontech; ref. 17). The transfection and clone selection procedures were similar to those followed for the 5HRE-GFP/EMT6 cells. Screening of the G-418-resistant clones was done by exposure to CoCl2 and imaging of the HIF-1 nuclear accumulation. The cells of this selected clone are designated as GFP-HIF1/EMT6.
PDT Treatment, PGE2, and CoCl2 Exposure
The transfected 5HRE-GFP/EMT6 and GFP-HIF1/EMT6 cells were grown on microscope grade coverslips placed in Petri dishes containing Basal Eagle Medium supplemented with 10% FCS and G-418. For PDT experiments, the cells were incubated with 1 µg mL–1 Photofrin for 24 h. After incubation, the medium containing the photosensitizer was removed and replaced with fresh medium. PDT irradiation was done on cell monolayers using 514-nm light from an argon ion laser (Innova 90, Coherent, Santa Clara, CA). An irradiance of 5 mW cm–2 was used to deliver treatment fluences ranging from 1 to 5 J cm–2. These treatment conditions were chosen to avoid treatment-induced hypoxia. To investigate the effects of PGE2 exposure in inducing HIF-1 target genes under normoxia, cells were exposed to 1 and 10 µmol/L PGE2 for 3 h. For both cell lines, the positive control was obtained by exposure to 100 µmol/L CoCl2 for 4 h (18). CoCl2 is a transition metal compound that is widely used as a hypoxia mimic in cell culture. Therefore, exposing cells to CoCl2-simulated hypoxia provides an estimate of the highest possible enhancement of HIF-1-induced GFP fluorescence in 5HRE-GFP/EMT6 cells and maximum nuclear translocation of GFP-tagged HIF-1 fusion protein in GFP-HIF1/EMT6 cells. Immediately after PDT treatment or exposure to PGE2 or CoCl2, the coverslips were returned to the incubator to allow for GFP synthesis and nuclear translocation of the fusion protein in 5HRE-GFP/EMT6 and GFP-HIF1/EMT6 cells, respectively. Cells that were exposed only to drug sensitization or that were used as controls were treated identically (including the medium change), with the exception of irradiation.
Oxygen Measurement
Photofrin-sensitized cell monolayers grown on coverslips were placed in Petri dishes and immersed in HBSS, and a Clark-style microelectrode (Diamond Micro Sensor, Ann Arbor, MI) with a tip diameter of 10 µm and a temporal response of <1 s was placed in close proximity to the coverslip glass surface. The oxygen-sensitive electrode recorded changes in oxygen concentration that occurred during PDT at an irradiance of 5 mW cm–2 of 514-nm light.
Fluorescence Imaging
Due to the short half-life of the destabilized EGFP in the 5HRE-GFP/EMT6 cells, these cells were imaged for GFP expression 3 h following PDT irradiation or exposure to PGE2 or CoCl2. The control, Photofrin only–, PGE2-, CoCl2-, and PDT-treated cell monolayers on coverslips were taken out of the incubator and washed with HBSS. The coverslips were then placed in a Leiden microscope coverslip dish and positioned on the stage of an inverted Nikon Diaphot fluorescence microscope. Selective excitation of GFP and its fluorescence collection were done using a GFP band-pass filter cube (Endow GFP Bandpass, Chroma, Rockingham, VT). Cell monolayer images were obtained with a 20x air immersion objective (0.75 numerical aperture) and acquired by a 12-bit monochrome cooled charge-coupled device camera (RT KE monochrome; Diagnostic Instruments, Sterling Heights, MI). To account for day-to-day variations in background GFP expression in control cells, coverslips containing control cells were processed in parallel with every experimental group. Throughout, fluorescence from treated cells was normalized to these parallel control samples.
The subcellular localization of the GFP-HIF1 fusion protein in the control, PGE2-treated, Photofrin-sensitized, and PDT-treated GFP-HIF1/EMT6 cells was examined by imaging on the microscope as described above. To observe the temporal kinetics of nuclear translocation of the fusion protein, the cells were imaged at 3 to 24 h following PDT irradiation.
Image Analysis
The analysis procedure was similar to that described in Mitra et al. (22). Briefly, the digitized images were analyzed using ImageJ (NIH)4 and MATLAB (The MathWorks, Inc., Natick, MA). Quantifying the extent of GFP expression in these samples required various image processing and analysis steps. To calculate the area occupied by cells, the contrast in the bright-field images was enhanced using the background subtraction tool in ImageJ. This enhancement allowed the ImageJ thresholding tool to successfully select and calculate all the pixels corresponding to cells in the field of view. The fluorescence images were first corrected for background signal, and then the pixels with GFP signal were easily selected using the thresholding tool. To calculate the average GFP intensity, the signals from all the GFP-positive pixels in a fluorescence image were summed and then divided by the total number of pixels containing cells, as calculated from the corresponding bright-field image.
Results
Figure 1
shows the levels of HIF-1 activation in the 5HRE-GFP/EMT6 cells as reported by GFP fluorescence for various treatment conditions. As illustrated in Fig. 1A, there was a low basal level of HRE-promoter-driven GFP expression in control cells, consistent with low constitutive activation of HIF-1. In comparison, the levels were maximally induced following exposure to the hypoxia mimic CoCl2 (Fig. 1B). In Fig. 1C, we observe that GFP levels were increased relative to controls with 10 µmol/L PGE2 incubation. Interestingly, incubation of cells with 1 µg mL–1 Photofrin for 24 h induced a statistically significant increase in GFP signal compared with the control (Fig. 1D). Following PDT treatment with fluences of 2.5 and 5 J cm–2 (Fig. 1E and F), the cells exhibited a significant increase in GFP expression compared with the drug-only control, with levels for 5 J cm–2 slightly but not significantly decreased relative to 2.5 J cm–2–treated cells.
|
|
3-fold higher than control levels and was obtained following exposure to CoCl2. Incubation of cells with 1 µg mL–1 Photofrin resulted in an 50% increase in GFP fluorescence relative to control. Photofrin-PDT with fluences of 1, 2.5, and 5 J cm–2 resulted in significantly higher GFP expression compared with completely untreated controls with a maximum increase of 2.2-fold observed with a fluence of 2.5 J cm–2. Relative to Photofrin-only controls, the irradiated cells exhibited a modest but statistically significant (P < 0.05) increase with an 38% higher GFP fluorescence obtained in the case of 2.5 J cm–2–treated cells. The results also showed a trend of initial increase in GFP expression with increasing treatment fluences followed by an apparent decrease. However, the differences in GFP fluorescence levels among the three different fluences were not statistically significant. In Fig. 4, we report the levels of GFP signal induced with the exposure of cells to two different concentrations of PGE2. Cells exposed to 1 µmol/L PGE2 produced 62% enhanced GFP fluorescence compared with untreated control levels. Increasing the PGE2 incubation concentration to 10 µmol/L did not increase the GFP signal, suggesting that the HIF-1 activation through the prostaglandin pathway was maximally induced with the 1 µmol/L PGE2 exposure. The findings summarized in Figs. 3 and 4 show that both Photofrin-PDT and PGE2 exposure induce appreciable levels of GFP accumulation, and therefore are consistent with the hypothesis that the increased expression of HIF-1 target genes in response to PDT is initiated, at least in part, by the prostaglandin pathway. Statistical analysis of the GFP signals also shows, however, that there is a significant difference (P < 0.05) between the PDT and PGE2 treatment groups, with treatment of Photofrin-sensitized cells with a fluence of 2.5 J cm–2 inducing an 35% higher effect than exposure to 10 µmol/L PGE2. Our attempts to define the role of PGE2 in mediating the activation of HIF-1 in Photofrin-PDT–treated cells were frustrated by the fact that the cells did not tolerate PDT at the doses used here in the presence of the COX-2-specific inhibitor NS398. In the presence of 100 µmol/L NS398, the transfected EMT6 cells reproducibly rounded and lifted off of the coverslip following doses of PDT that were clearly nontoxic in the absence of the inhibitor.
|
|
/EMT6 cells transfected with the fusion protein vector CMV-GFP-HIF1 displayed HIF-1 translocation from the cytosol (Fig. 5A
) to the nucleus in response to 10 µmol/L PGE2 exposure for 3 h (Fig. 5B). Exposure of the cells to 100 µmol/L CoCl2 also induced a large fraction of the cells to undergo HIF-1 nuclear translocation (data not shown). In response to 1 µg mL–1 Photofrin incubation for 24 h, we observed some evidence of cells in the imaged field exhibiting nuclear translocation of the fusion protein (Fig. 5C). This result is consistent with the summary data of Fig. 3, which shows that Photofrin sensitization induces a small but significant increase in the expression levels of GFP driven by HIF-1 activation. Three hours following PDT irradiation of Photofrin-sensitized cell monolayers with a fluence of 2.5 J cm–2, a large proportion of the cells showed nuclear translocation (Fig. 5D), whereas at 7 h post-PDT with the same fluence, almost all of the cells in the treated field displayed bright GFP fluorescence from the nucleus (Fig. 5E). Identically treated cells imaged 24 h postirradiation also displayed nuclear accumulation of HIF-1 (Fig. 5F), indicating that PDT induces strong and prolonged activation of the HIF-1 pathway. Similar patterns of nuclear translocation were also observed in cells that were irradiated with a fluence of 5 J cm–2 (data not shown).
|
Numerous reports have shown that PDT induces microvascular damage within treated tumors (3–5) and that this leads to tumor hypoxia (4, 11). Because hypoxia induces increased expression and stabilization of HIF-1 and activates the HIF-1 transcription factor, recent molecular studies have thus far investigated the effects of PDT-induced hypoxia on expression of proangiogenic factors and their relevance to treatment efficacy. For example, a study by Ferrario et al. (12) was the first to show that Photofrin-mediated PDT induces a modest increase in VEGF levels in BA carcinoma cells in vitro. They also found that PDT done on tumors in vivo grown from the same cells resulted in increased expression of HIF-1 and VEGF. Further, they reported that significantly enhanced tumor response is obtained when PDT was combined with antiangiogenic treatments that attenuated the therapy-induced increase in VEGF levels. In a subsequent study, Ferrario et al. (13) examined the activation of COX-2 after Photofrin-PDT and NPe6-PDT, and found that both photosensitizers were able to elicit increased PDT-mediated COX-2 expression and PGE2 synthesis in mouse sarcoma and carcinoma cells in vitro. They also observed that Photofrin-PDT induced elevated levels of COX-2, PGE2, and VEGF in RIF tumors in vivo. These results were consistent with earlier work of Henderson and Donovan (23), who reported that Photofrin-PDT stimulates release of PGE2 from macrophages and RIF tumor cells in culture and that the extent of release was dose dependent. Thus, these studies have shown that PDT-mediated tumor damage can result in the up-regulation of proangiogenic factors.
Based on this previous literature, the primary focus of our current study was to examine the PDT-induced molecular pathways that govern the oxygen-independent activation/stabilization of HIF-1, its nuclear translocation, and the consequent expression of HIF-1 target genes through optical imaging of fluorescent reporter proteins in cell monolayers. We therefore restricted our experiments to cells in vitro, where culture and treatment conditions could be carefully controlled to avoid hypoxia. To examine these responses to PDT and PGE2, we established two stably transfected cell lines. The 5HRE-GFP/EMT6 cells were used to quantify the levels of HIF-1 activation through the expression of the reporter protein GFP. The GFP-HIF1/EMT6 cells were used to directly visualize the nuclear translocation and accumulation of the HIF-1 protein following PDT-induced oxidative stress or PGE2 exposure. A distinct advantage of using these transfected cell lines and optical imaging is that experiments with fluorescent reporters permit the study of inducible expression or subcellular localization of proteins on a per cell basis, whereas biochemical assays provide complementary information but are limited to measurements obtained from whole populations of cells.
As shown in the summary plot of Fig. 3, measurements in 5HRE-GFP/EMT6 cell monolayers showed that Photofrin sensitization alone resulted in increased levels of GFP. These results are consistent with the observation that cells exposed to Photofrin incubation exhibited nuclear translocation of GFP-tagged HIF-1 (Fig. 5C). Identifying the molecular process through which Photofrin activates this promoter is an interesting subject but beyond the scope of our study. However, based on ample evidence in the literature (24) that transition metal ions, iron chelators, pyruvate, lactate, and other molecules stimulate HIF-1 by impairing the degradation of HIF-1 under normoxic conditions, we can speculate that the photosensitizer Photofrin results in increased GFP levels in these cells via a similar molecular mechanism. It is interesting to note that Ferrario et al. (12) reported no significant increase in VEGF levels with Photofrin incubation compared with untreated controls. It is possible that the Photofrin-induced GFP expression is more easily detectable in our studies for two reasons: (a) quantifying gene expression through reporter protein fluorescence may be a more sensitive measurement than the ELISA assay used by Ferrario et al., and (b) the reporter protein construct in the 5HRE-GFP/EMT6 cells contains five copies of HRE, which is designed to make the target gene expression more responsive (20).
With PDT irradiation of these cells, we observed a trend of initial increase in GFP levels in response to fluences of 1 and 2.5 J cm–2 and then a slight decrease with an increased fluence of 5 J cm–2. However, the difference in GFP expression among the three fluences was not significant. These results are in qualitative agreement with the study of Ferrario et al. (12), where the authors reported that VEGF levels showed a similar pattern with increasing PDT doses. The 12% decrease in GFP intensity at 5 J cm–2 relative to 2.5 J cm–2 is most likely due to damage to cells that resulted in compromised GFP protein synthesis without reduced cell viability. In a recently reported study (22), we measured GFP expression driven by the heat shock 70 promoter in transfected EMT6 cells that were subjected to mTHPC-PDT, and observed that at treatment fluences that resulted in significant loss of cell viability, the cells expressed decreased levels of GFP.
CoCl2 activates the hypoxia-mediated signaling pathway by causing the stabilization of HIF-1 under normoxia and mimics hypoxia, at least in part, by occupying the von Hippel-Landau binding domain of HIF-1, thereby preventing its degradation even in the presence of oxygen (25). Liu et al. (26) showed that the CoCl2-simulated hypoxia led to increased expression of VEGF mRNA and protein levels in three prostate cancer cell lines and that this increase was mediated by COX-2. Thus, for experiments with both of the transfected cell lines, CoCl2 served as a useful positive control. We observed that the extent of HIF-1 activation due to either PDT-generated oxidative stress or direct PGE2 exposure was significantly less than that induced by CoCl2-simulated hypoxia. However, we note that Photofrin-PDT with 2.5 J cm–2 induced 70% of the maximum CoCl2-induced response. Although the physiologic relevance of this substantial PDT-mediated activation is not yet established, this result supports the use of antiangiogenic treatments in combination with PDT, as has been suggested by Ferrario et al. (12) and Zhou et al. (27). Before leaving the discussion of CoCl2, we note that reactive oxygen species are generated by hypoxia and by CoCl2 and these have been implicated recently in HIF-1-mediated cellular responses. Thus, the situation is complex, and it is possible, if not likely, that there is crosstalk between mechanisms that have, until recently, been considered separate (19, 28).
As illustrated in Fig. 5B, exposure of GFP-HIF1/EMT6 cells to 10 µmol/L PGE2 induced nuclear translocation of the GFP-tagged HIF-1 protein. This is in excellent agreement with the findings of Liu et al. (17), who showed that PGE2 exposure resulted in significant nuclear localization of HIF-1 in PC-3ML human prostate cancer cells under normoxic conditions. We find that Photofrin-PDT also strongly induced the translocation of the fusion protein from the cytosol to the nucleus. This nuclear localization can be visualized as early as 3 h postirradiation and exists for at least 24 h following PDT, indicating that the translocation process is not transient. The rapid translocation of the HIF-1 to the nucleus at 3 h post-PDT correlates well with the enhanced GFP expression observed in 5HRE-GFP/EMT6 cells at this time point. Nuclear accumulation of HIF-1 takes place only after it is stabilized in the cytosol. Stabilization occurs before HIF-1 dimerizes with aryl hydrocarbon receptor nuclear translocator, after which the HIF-1 complex binds to the HRE promoter regions to initiate transcription of target genes (29). Therefore, these images provide evidence that, in the absence of hypoxia, PDT-mediated oxidative insult has the ability to stabilize HIF-1 and induce its nuclear translocation. Thus, integrating the results of the prior study of Ferrario et al. (13), which showed that PDT induced increased levels of PGE2, the recent reports of HIF-1 activation by reactive oxygen species as reviewed by Kietzmann and Görlach (19), and our current observations lends support to the hypothesis that Photofrin-PDT initiates HIF-1 stabilization, its nuclear translocation, and subsequent transcription of HIF-1 target genes through the prostaglandin pathway and/or other signaling mechanisms triggered by reactive oxygen species such as singlet oxygen.
In summary, the findings from this study confirm that Photofrin-PDT is an efficient inducer of all of the biologically relevant aspects of the oxygen-independent activation of HIF-1. PDT-mediated oxidative stress results in damage to cells, so the observation of nuclear translocation of GFP-tagged HIF-1 and increased GFP expression driven by the HRE promoter in these cells indicates that the HIF-1 pathway is intact following the treatment under these conditions. Not all PDT treatments induce persistent hypoxia, and those that do tend to target microvascular damage. A question not addressed by this study is whether or not transient PDT-induced hypoxia, of which the origin is photochemical oxygen depletion (9–11), is capable of activating HIF-1. In the absence of microvascular occlusion, this photochemical oxygen depletion lasts only as long as the PDT irradiation, which clinically is 30 min. On the basis of the very long (12–14 h) exposures to severe hypoxia used typically in HIF-1 activation studies in vitro (17, 18), it seems unlikely that the duration of transient hypoxia would be sufficient to bring about activation. Although this issue remains open, our findings offer evidence that, even in the absence of hypoxia, sublethal Photofrin-PDT–mediated HIF-1 activation is intermediate in magnitude between PGE2 and the hypoxia mimic CoCl2 and could therefore be clinically relevant.
Acknowledgments
We thank Dr. Steve Hahn for the gift of Photofrin and Drs. J. Martin Brown and Alice Levine for kindly providing us with the plasmid constructs.
Footnotes
Grant support: U.S. NIH grant CA68409 awarded by the National Cancer Institute.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
4 http://rsb.info.nih.gov/ij/. 
Received 7/18/06; revised 10/ 6/06; accepted 10/27/06.
References
stabilization and nuclear localization in a human prostate cancer cell line. J Biol Chem 2002;277:50081–6. and von Hippel-Landau protein by direct binding to hypoxia-inducible factor-. J Biol Chem 2003;278:15911–6.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
0.05), and fluorescence in all of the three PDT treatment groups was significantly higher than in the Photofrin-only group. No statistical difference was obtained among the three PDT groups. Columns, mean of at least 20 separate fields of view; bars, SE.
