Low-power laser irradiation (LPLI) can modulate various biological processes.1 For instance, LPLI can induce cell proliferation and differentiation.2, 3 It has been used to treat diseases with regeneration limitations4 and to promote wound healing.5, 6 LPLI, in combination with a photosensitizer, can also induce apoptosis via photodynamic actions.7, 8 High-fluence LPLI interferes with cell cycles and inhibits cell proliferation; thus, it can be used to control certain types of hyperplasia.9, 10, 11 Some studies suggest that LPLI has a stimulatory effect on cell proliferation at relatively low fluences but an inhibitory effect at high fluences.12, 13 Furthermore, the biological effects of LPLI also depend on laser wavelength, power density, irradiation procedure, and cell type.14, 15 Our previous studies showed that low-fluence LPLI induced proliferation of human lung adenocarcinoma cells (ASTC-a-1),16 while high fluence LPLI induced caspase-3 activation and apoptosis.17
There have been many reports on the mechanism of low-fluence LPLI-induced cell proliferation. For example, it can induce expression or activation of certain signal proteins, such as PKC (protein kinase Cs),16 ERK (extracellular signal-regulated kinase),18 Akt,19 PDGF (platelet-derived growth factor),20 VEGF (vascular endothelial growth factor),21 NGF (nerve growth factor),22 CyclinD1,22 IL-8 (interleukin-8),23 and (interleukin-1 alpha).23 Low-fluence LPLI can also induce changes in reactive oxygen species (ROS) production,24 concentration,25 ATP quantity,26 mitochondrial membrane potential ,26 and intracellular pH.27 However, so far there are no definitive conclusions on the mechanism of high-fluence LPLI-induced cell death.
Apoptosis is an important cellular event that plays a key role in pathology and treatment of many diseases.28, 29 The mechanisms of initiation and regulation of apoptosis are complex and diverse. Some studies suggest that ROS generation, decrease, and caspase activation are all involved in photodynamic therapy (PDT)–induced cell apoptosis.30, 31 There are two main pathways to initiate and regulate cell apoptosis. One is the death receptor pathway (receptor/caspase-8/mitochondria/caspase-3), and the other is the mitochondria pathway (mitochondria/caspase-3).32
Fluorescence resonance energy transfer (FRET) has been widely used to study protein-protein interactions in living cells.33, 34, 35 It can be effectively used to study cellular activities, such as caspase activation during cell apoptosis, with specifically designed genetic reporters.36, 37
In this study, both ASTC-a-1 cells and COS-7 cells were irradiated by a He-Ne laser under a fluence of and , respectively, for a duration of in the absence of a photosensitizer to induce apoptosis. In order to study the mechanism of the induced apoptosis, we measured and analyzed the dynamics of ROS generation, change of , and activity of caspase-3 before and after the laser irradiation. In order to determine the pathway of the induced cells apoptosis, we also observed in vivo caspase-8 activity.
Materials and Methods
Human lung adenocarcinoma cells (ASTC-a-1) and African green monkey SV40-transformed kidney fibroblast cells (COS-7) were grown on culture glasses, in Dulbecco’s modified Eagle’s medium (DMEM, Life Technologies, Inc.), supplemented with 15% fetal bovine serum, penicillin, and streptomycin in 5% , 95% air at in a humidified incubator. In all experiments, 70 to 85% confluent cultures were used.
Chemicals and Plasmids
The following fluorophore probes were used: dichlorodihydrofluorescein diacetate ( , , Molecular Probes, Inc.) to detect the generation of ROS; MitoTracker Deep Red 633 ( , Molecular Probes, Inc.) to stain mitochondria; and Rhodamine 123 ( , Alexis Biochemicals, Inc.) to monitor . The cells were incubated in DMEM with different fluorophore probes. The optimal concentrations and incubation times for each of the probes were determined experimentally.
In addition, we used Lipofectin reagent (Invitrogen life Technologies, Inc.) to transfect plasmid DNA into ASTC-a-1 cells and G418 antibiotic ( , Sino-American Biotec, Inc., China) to screen the cells stably expressing reporter [SCAT3]. The pSCAT3 was a gift from Dr. Masayuki Miura, and the pBid-CFP was a gift from Dr. Kazunari Taira.
Laser Irradiation of Cells
For irradiation of the ASTC-a-1 cells and COS-7 cells, a He-Ne laser inside a confocal laser scanning microscope (LSM510-ConfoCor2, Zeiss, Jena, Germany) was used. Laser irradiation was performed through the objective lens of the microscope. In this setup, only the cells under observation were irradiated by the laser. The cells in the selected area were irradiated for with fluences of and . The power dose densities of the laser on the cellular level were and . The laser irradiation sizes were and . A minitype culture chamber with supply (Tempcontrol 37-2 Digital, Zeiss, Germany) was used in order to keep cells under normal culture conditions ( , 5% ) during irradiation.
A genetic FRET reporter, SCAT3, was used to monitor caspase-3 activity. The working mechanism of SCAT3 is shown in Fig. 1 . This reporter consists of a donor cyan fluorescent protein (CFP) and an acceptor Venus (a mutant of yellow fluorescent protein). The donor and the acceptor are linked with a caspase-3 recognition and cleavage sequence (DEVD). Activation of caspase-3 leads to the cleavage of the linker, thus effectively reducing the FRET. Therefore, a decrease in the FRET represents activation of the caspase-3.36 Another genetic reporter, Bid-CFP, was used to measure the caspase-8 activity. The working mechanism of this specific reporter is shown in Fig. 2 . In the plasmid Bid-CFP, CFP is fused to the C terminus of Bid. The activation of caspase-8 leads to the cleavage of the Bid protein and yields two fragments (p15 and p7). An exposed glycine residue at the C terminus of Bid (tBid) undergoes N-myristoylation, and the resulting CFP-tBid is translocated to the mitochondrial outer membrane. The aggregation and localization of the CFP-tBid in mitochondria represent activation of caspase-8.38 Using a genetic reporter technique based on SCAT3 and Bid-CFP, the spatiotemporal dynamics of caspase-3 and caspase-8 activity in living cells was monitored in real time during laser-irradiation–induced apoptosis.
Imaging Analysis of Living Cells
In order to image the activities of a single cell, a confocal laser scanning microscope (LSM) system was used. The system is equipped with a krypton-argon air-cooled laser and a He-Ne laser for excitation illumination. The illumination power was reduced to 3 to 0.3% of the maximum power of the excitation lasers to avoid fluorescence saturation and bleaching. Imaging was performed before and after laser irradiation, and the images were acquired with a Plan-Neofluar 40X/NA1.3 oil-immersed objective lens. Cells were maintained at , 5% during imaging with the minitype culture chamber. For the measurement of ROS and , cells stained with and Rhodamine 123 were excited using the krypton-argon laser. The excitation wavelength was , the beamsplitter was NFT , and the emission detection filter was BP . For the measurement of the in vivo FRET effect, cells transfected with SCAT3 were excited using the krypton-argon laser. The excitation wavelength was , the beamsplitter was NFT , and the emission detection filters were BP and LP for emissions of CFP and Venus, respectively. For the measurement of caspase-8 activity and the mitochondria staining, cells transfected with Bid-CFP and stained with MitoTracker Deep Red were excited by the krypton-argon laser and the He-Ne laser. The wavelengths used were for the excitation of CFP and for the excitation of MitoTracker Deep Red. The beamsplitter was NFT , and the emission detection filters were BP and LP for the emissions of CFP and MitoTracker Deep Red, respectively. For intracellular measurements, the desired measurement area was chosen in the LSM image. To quantify the results, the fluorescence emission intensities were obtained with Zeiss Rel3.2 image processing software (Zeiss, Jena, Germany). All imaging experiments were repeated in different cells. For the analysis of Bid-CFP translocation, images were analyzed with MATLAB 6.5 by drawing regions around individual cells and then computing the SD of the intensity of the pixels (punctuate/diffuse) and integrated brightness (total brightness). All imaging experiments were repeated in different cells. As a rough indication of the range of variability among cells within an experiment, error bars show the SD.
High-Fluence LPLI-Induced Mitochondrial ROS Generation
The generation of ROS in ASTC-a-1 cells by high-fluence LPLI was monitored by measuring changes in fluorescence resulting from the oxidation of intracellular (nonfluorescent) to DCF (fluorescent). Cells were incubated with for in DMEM culture liquid in an incubator. Fluorescence progressively increased after the cells were irradiated, indicating the generation of ROS. The typical time-course images of cells loaded with after laser irradiation are shown in Fig. 3a . The average fluorescence intensity of DCF after high-fluence LPLI from five different cells is shown in Fig. 3b. The DCF fluorescence was observed as soon as the laser irradiation was completed, while there was a low level of detectable DCF fluorescence before irradiation corresponding to the ROS level in rest cells (data not shown). This result indicated an immediate generation of ROS following the laser irradiation. This was followed by a gradual increase of DCF fluorescence intensity. About after laser irradiation, the DCF fluorescence intensity reached a plateau [Figs. 3a and 3b].In order to determine whether this phenomenon happened in other apoptosis doses, we used LPLI to irradiate ASTC-a-1 cells; the result then showed that at after laser irradiation, the DCF fluorescence intensity reached a plateau [Figs. 3a and 3b]. Similar results were also obtained in COS-7 cells [Fig. 3c].
To confirm that the increased DCF fluorescence after laser irradiation originated from mitochondria, ASTC-a-1 cells were coloaded with MitoTracker Deep Red 633, a mitochondria-specific dye. The perinuclear fluorescence of DCF from the cells colocalized with MitoTracker Deep Red is shown in Fig. 3d.The overlaid images of the DCF fluorescence and MitoTracker Deep Red fluorescence indicate that ROS were mostly formed in mitochondria. Similar results were also obtained in COS-7 cells (data not shown).
High-Fluence LPLI-Induced Mitochondrial Membrane Potential Decrease
ASTC-a-1 cells were loaded with Rhodamine 123 dyes in order to monitor in single cells by confocal laser scanning microscopy. Polarized mitochondria would appear as bright fluorescent spheres; thus, a reduction in fluorescence would indicate the decrease of . The typical time-course images of cells loaded with Rhodamine 123 after laser irradiation are shown in Fig. 4a . The average fluorescence intensities of Rhodamine 123 at different times after the high-fluence LPLI from five different cells are shown in Fig. 4b. About after laser irradiation, the fluorescence intensity started to decrease [Fig. 4b], which indicated the decrease of . About after laser irradiation, the fluorescence intensity reached the lowest level. Without laser treatment, was nearly unchanged, as shown by the control data in Fig. 4b. Comparing Fig. 3b with Fig. 4b, it is apparent that the generation of mitochondrial ROS (immediately following laser irradiation) precedes the decrease of ( after irradiation). The results suggest that the ROS production under high-fluence LPLI might have caused the decrease. Similar results were also obtained in COS-7 cells [Fig. 4c].
High-Fluence LPLI-Activated Caspase-3
FRET imaging was used to determine the dynamic changes of caspase-3 in ASTC-a-1 cells transfected by SCAT3 reporters after high-fluence LPLI . The typical time-course Venus/CFP ratio images of cells stably expressing SCAT3 reporters after LPLI show an intensity decrease in Fig. 5a (top). The average emission ratio of Venus/CFP from five different cells is plotted as function of time during laser irradiation [Fig. 5b]. The fluorescence intensities of Venus/CFP first decreased slightly during the first after LPLI then decreased significantly. About after laser irradiation, the fluorescence intensity from the cells reached a low, stable level. These results indicated that the caspase-3 activity increased gradually due to laser irradiation, noticeably after irradiation. Comparing Fig. 4b with Fig. 5b, it is clear that the onset of decrease ( after irradiation) precedes that of caspase-3 activation ( after irradiation). The results suggest that the decrease might affect the activity of caspase-3. The similar results were also obtained in COS-7 cells [Figs. 5a and 5b].
Effect of High-Fluence LPLI on Caspase-8
In order to determine the pathway of cell apoptosis induced by high-fluence LPLI , caspase-8 activity was observed using the fluorescence of the Bid-CFP reporter, because aggregation and localization of Bid-CFP in mitochondria would represent the activation of caspase-8.38 The typical time-course images of ASTC-a-1 cells transfected with pBid-CFP reporters after laser irradiation are shown in Fig. 6a (top). The Bid-CFP punctuate/diffuse index at different times after the high-fluence LPLI from five different cells is shown in Fig. 6c. The Bid-CFP fluorescence was distributed evenly in the cytoplasm, and no aggregation occurred, even after apoptosis. To determine the location of Bid-CFP, the cells transfected with pBid-CFP were also loaded with MitoTracker Deep Red. The time-course fluorescence images of MitoTracker Deep Red overlaid with images of pBid-CFP are shown in Fig. 6a (bottom). There was no colocalization. These results demonstrate that caspase-8 was not activated by the high-fluence LPLI, suggesting that high-fluence LPLI-induced apoptosis was initiated directly from the mitochondria pathway, independent of the activity of caspase-8. Similar results were also obtained in COS-7 cells [Figs. 6b and 6c].
The mechanism of low-fluence LPLI-induced cell proliferation has been reported in various studies, but the mechanism of high-fluence LPLI-induced cell death remains unclear. This study is an attempt to understand the mechanism of cell apoptosis induced by high-fluence LPLI, by investigating a cascade of phenomena induced by laser irradiation of ASTC-1-a cells. Cellular functions such as generation of ROS and change of were studied using fluorescent-vital probes. Caspase-3 and caspase-8 activities were monitored in real time in single living cells using genetic reporters.
In our studies, we chose two cell lines, ASTC-a-1 and COS-7, to study the mechanism of high-fluence LPLI induced cell apoptosis. The former were human lung adenocarcinoma cells; hence, our results could be applied to cancer therapy. The latter, although transformed with SV40, were not cancer cells and did not have multiple oncogene activation; hence, our results should be more applicable to the normal cells that were in tissues treated with LPLI.
Our previous work shows that lower-fluence LPLI can promote ASTC-a-1 proliferation at a dose of (Ref. 16), and high fluence LPLI can induce ASTC-a-1 apoptosis at the dose . In this study, we choose as the major stimulation dose and as the parallel dose. We found that both doses of LPLI cause similar biological effects, and the difference is major in the initiate time and end time of these cellular events, mentioned earlier.17
In PDT-induced apoptosis, there is evidence that ROS plays an important role during the progression of apoptosis.30 One of the objectives of this study was to detect the generation of ROS as a result of high-fluence LPLI. Our results show an immediate mitochondrial ROS generation induced by the laser irradiation, followed by a gradual increase of ROS, as indicated by the increase of DCF fluorescence intensity (see Fig. 3) from irradiated cells containing . This observation agrees with the suggestions that the initial ROS, such as , , and , are produced as a result of light absorption either by endogenous porphyrins or cytochrome in mitochondria. We suggest that endogenous porphyrins may play a role based on the increase of cellular ROS levels after higher-fluence LPLI (Fig. 3). Following are three examples:
3. Low-fluence LPLI can also induce the mitochondrial ROS generation, but the difference is the relative low level compared with the high level induced by high-fluence LPLI.27
The decrease of the mitochondrial membrane potential, , may promote inner membrane permeabilization, disrupt the mitochondria structure, and cause the release of proapoptotic factors. In our study, after laser irradiation, the decrease in began, as indicated by Rhodamine fluorescence intensity in irradiated cells [Fig. 4a].
The release of proapoptotic factors, such as cytochrome C, can activate caspase-9, and subsequently, caspase-3. Utilizing the FRET, we found that the caspase-3 activity in ASTC-a-1 cells increased gradually after high-fluence LPLI, as indicated by the decrease of the fluorescence intensity ratio of Venus/CFP from the cells transfected by SCAT3, as shown in Fig. 5.
Caspase-8 activity was studied in order to determine whether high-fluence LPLI-induced cell apoptosis was initiated directly from the mitochondria pathway or mediated by upstream caspase-8. Figure 6 shows that the fluorescence from Bid-CFP in irradiated cells was uniformly distributed in the cytoplasm without aggregation in the mitochondria. Therefore, caspase-8 was not activated by high-fluence LPLI, which indicates that high-fluence LPLI-induced cell apoptosis was initiated directly from the mitochondria pathway, independent of caspase-8. This observation explains why high-fluence LPLI-induced caspase-3 activation is a relatively earlier event [about after laser irradiation, Fig. 5b], compared with the TNF- -induced apoptosis (about after TNF- treatment), which was mediated by the caspase-8.45
1. Immediate generation of mitochondrial ROS following laser irradiation, reaching a maximum level after irradiation.
2. Onset of decrease after laser irradiation, reaching a minimum level after irradiation.
3. Activation of caspase-3 between and after laser irradiation.
Therefore, we propose that the cellular responses to laser irradiation follow a temporal sequence shown as follows: Without evidence of caspase-8 activation in our experiments, we conclude that light is absorbed directly by endogenous porphyrins in mitochondria, resulting in the initial ROS generation. And we found that both ASTC-a-1 cells and COS-7 cells can be induced by the increase of mitochondrial ROS. Therefore, the porphyin may exist not only in ASTC-a-1 cells but also in many cell lines.27 Interestingly, we found that at a fluence of , the mitochondrial ROS level in COS-7 cells reaches the plateau much more quickly than in ASTC-a-1 cells . This result indicates that COS-7 cells are more sensitive to light than ASTC-a-1 cells, although the distinction is insignificant [Figs. 3b and 3c]. After light absorption, oxidative stress induces the redox activity of respiratory chain acceleration, resulting in massive production of ROS. Subsequently, the generation of ROS induces the decrease of and inner membrane permeabilization, thereby promoting mitochondrial structure destruction and causing the release of proapoptotic factor. The release of proapoptotic factors, such as cytochrome C, activates the caspase-9, and subsequently, caspase-3, which finally results in apoptosis.
In conclusion, our experimental results show that the mitochondrial ROS generation, the decrease of , and the activation of caspase-3, in that order, play a critical role in the high-fluence LPLI-induced ASTC-a-1 cell apoptosis. To our best knowledge, this is the first time that the temporal sequence of cellular activities during high-fluence LPLI-induced apoptosis has been unequivocally determined using fluorescence imaging techniques. Our study contributes to the understanding of the mechanism of LPLI-induced biological responses. In addition, this research would be useful due to the fact that high fluences producing cell death by apoptosis may explain some of the negative findings in LPLI studies where too high a fluence may have been inadvertently used.
The authors thank Dr. Masayuki Miura for providing pSCAT3 and Dr. Kazunari Taira for providing pBid-CFP. This research is supported in part by the National Natural Science Foundation of China (Contract Grant Nos. 60378043, 30470494), by the Natural Science Foundation of Guangdong Province (Contract Grant Nos. 015012, 04010394), and by a grant from the U.S. National Institute of Health (Grant No. P20 RR016478 from the INBRE Program of the National Center for Research Resources).