The formation of electronically excited species in the human multiple myeloma cell suspension

In this study, evidence is provided on the formation of electronically excited species in human multiple myeloma cells U266 in the growth medium exposed to hydrogen peroxide (H2O2). Two-dimensional imaging of ultra-weak photon emission using highly sensitive charge coupled device camera revealed that the addition of H2O2 to cell suspension caused the formation of triplet excited carbonyls 3(R = O)*. The kinetics of 3(R = O)* formation in the real time, as measured by one-dimensional ultra-weak photon emission using low-noise photomultiplier, showed immediate enhancement followed by a slow decay. In parallel to the formation of 3(R = O)*, the formation of singlet oxygen (1O2) in U266 cells caused by the addition of H2O2 was visualized by the imaging of 1O2 using the green fluorescence of singlet oxygen sensor green detected by confocal laser scanning microscopy. Additionally, the formation of 1O2 after the addition of H2O2 to cell suspension was detected by electron paramagnetic resonance spin-trapping spectroscopy using 2,2,6,6-tetramethyl-4-piperidone. Presented results indicate that the addition of H2O2 to cell suspension results in the formation of 3(R = O)* and 1O2 in U266 cell suspension. The contribution of the cell-free medium to the formation of electronically excited species was discussed.

In this study, evidence is provided on the formation of electronically excited species in human multiple myeloma cells U266 in the growth medium exposed to hydrogen peroxide (H 2 O 2 ). Two-dimensional imaging of ultra-weak photon emission using highly sensitive charge coupled device camera revealed that the addition of H 2 O 2 to cell suspension caused the formation of triplet excited carbonyls 3 (R 5 O)*. The kinetics of 3 (R 5 O)* formation in the real time, as measured by one-dimensional ultra-weak photon emission using low-noise photomultiplier, showed immediate enhancement followed by a slow decay. In parallel to the formation of 3 (R 5 O)*, the formation of singlet oxygen ( 1 O 2 ) in U266 cells caused by the addition of H 2 O 2 was visualized by the imaging of 1 O 2 using the green fluorescence of singlet oxygen sensor green detected by confocal laser scanning microscopy. Additionally, the formation of 1 O 2 after the addition of H 2 O 2 to cell suspension was detected by electron paramagnetic resonance spin-trapping spectroscopy using 2,2,6,6-tetramethyl-4-piperidone. Presented results indicate that the addition of H 2 O 2 to cell suspension results in the formation of 3 (R 5 O)* and 1 O 2 in U266 cell suspension. The contribution of the cell-free medium to the formation of electronically excited species was discussed. R eactive oxygen species (ROS) are produced as a byproduct of either metabolic processes such as cellular respiration in mitochondria and photosynthesis in chloroplast [1][2][3] or oxidative burst in phagocytic cells known to play role in the defense against infection 4 . The one-electron reduction of molecular oxygen produces superoxide anion radical (O 2 N2 ) known to dismutate to hydrogen peroxide (H 2 O 2 ) and subsequently to hydroxyl radical (HO N ). When ROS are not sufficiently eliminated by non-enzymatic (low molecular weight scavengers) and enzymatic (the superoxide dismutase and peroxidase families of enzymes) defense systems, ROS oxidize biomolecules comprising lipid, protein and nucleic acid [5][6][7][8] . The oxidation of lipid and protein by abstraction of weakly bonded hydrogen atom by HO N results in the formation of lipid and protein alkyl radicals (R N ), respectively. Subsequently, lipid and protein peroxyl radicals (ROO N ) are formed by the interaction of R N with molecular oxygen. Peroxyl radicals abstract hydrogen from lipid and protein, while lipid and protein hydroperoxides (ROOH) are formed. Under reducing conditions (reduced transition metals such as Fe 21 , Mn 21 , Zn 21 or Cu 1 bound in metalloproteins, transport proteins, and storage proteins), ROOH is reduced to lipid and protein alkoxyl radicals (RO N ), whereas under oxidizing conditions (oxidized transition metals, cytochrome c, peroxynitrite, chloroperoxide, and hypochlorous acid), ROOH is oxidized back to lipid and protein ROO N9, 10 .
The oxidation of biomolecules by ROS was proposed to be associated with the formation of electronically excited species such as triplet excited carbonyl 3 (R 5 O)* and singlet oxygen ( 1 O 2 ) [11][12][13][14] . Based on chemical system experiments, 3 (R 5 O)* was proposed to be produced by the decomposition of cyclic (1,2-dioxetane, ROOR) and linear (tetroxide, ROOOOR) high-energy intermediates. 1,2-dioxetane is formed by the cycloaddition of 1 O 2 to lipid and protein, by the cyclization of lipid and protein ROO N15 , and by the enzymatic reactions [16][17][18] . Tetroxide is formed by the recombination of two lipid and protein ROO N via Russell mechanism 15,19 . Once 3 (R 5 O)* is formed, the excitation energy of 3 (R 5 O)* can be transferred either to molecular oxygen resulting in the formation of 1 O 2 20 or to chromophore (C) forming singlet excited chromophore ( 1 C*) 14 . Beside the reactions related to the transfer of excitation energy, 3 (R 5 O)* is known to undergo a variety of other reactions comprising of isomerization, cleavage, hydrogen abstraction, and cycloaddition 21,22 . In addition to 3 (R 5 O)* pathway, the direct decomposition of ROOOOR to 1 O 2 by Russell mechanisms was evidenced in the chemical system 23,24 . More recently, several lines of evidence were provided on the formation of 1 O 2 by the decomposition of lipid and protein ROOH by metal ions, cytochrome c, peroxynitrite, chloroperoxide, and hypochlorous acid [25][26][27][28] . As the yield of 3 (R 5 O)* formation during the decomposition of ROOOOR was found to be 10 3 -10 4 lower compared to the yield of 1 O 2 formation, it was proposed that the Russell mechanism leads predominantly to the formation of 1 O 2 , whereas the formation of 3 (R 5 O)* is rather negligible 29 .
The excitation energy of electronically excited species formed under oxidative processes is emitted as ultra-weak photon emission 14,[30][31][32] . The excitation energy of 3 (R 5 O)* is emitted as near UVA and blue-green photons in the spectral range of 350-550 nm 14,19 . The photon emission of 1 C* is in the green-red region of the spectrum (550-750 nm) 14 , whereas dimol and monomol photon emission of 1 O 2 is in the red (634, 703 nm) and infra-red (1270 nm) region of the spectrum 33,34 , respectively. Recent development in one-dimensional detection of ultra-weak photon emission using low-noise photomultiplier tubes (PMT) and two-dimensional detection of ultra-weak photon emission using highly sensitive charge coupled device (CCD) cameras allows temporal and spatial visualization of the formation of electronically excited species.
In spite of the fact that several lines of evidence were provided on the formation of 3 (R 5 O)* and 1 O 2 in chemical system, a limited number of studies have focused on the production of electronically excited species in cells. In this study, experimental evidence on the formation of 3 (R 5 O)* and 1 O 2 in human multiple myeloma cell line U266 suspended in the growth medium containing free amino acids, fetal bovine serum and traces of transition metal ions is provided. The detection of the electronically excited species was performed using ultra-weak photon emission, confocal laser scanning microscopy and electron paramagnetic (EPR) spin-trapping spectroscopy. It is demonstrated here that both 3 (R 5 O)* and 1 O 2 are formed in the cell suspension exposed to oxidative environment.

Results
The effect of various H 2 O 2 concentrations on the formation of the electronically excited species and cell viability. To select the most suitable concentration of H 2 O 2 for the formation of 3 (R 5 O)* and 1 O 2 , the effect of various H 2 O 2 concentrations on the formation of 3 (R 5 O)* and 1 O 2 was tested using one-dimensional ultra-weak photon emission and EPR spin-trapping spectroscopy, respectively. Figure 1A shows the one-dimensional ultra-weak photon emission measured from cell suspension after the addition of H 2 O 2 at different concentrations. The addition of H 2 O 2 at different concentrations to the cell suspension resulted in the immediate significant increase of one-dimensional ultraweak photon emission as compared to the untreated cell suspension. In further, the formation of TEMPONE EPR signal in cell suspension was measured after the addition of H 2 O 2 at various concentrations. Figure 1B shows that TEMPONE EPR signal was enhanced nonlinearly with the increasing concentration of H 2 O 2 . These observations reveal that the addition of 5 mM H 2 O 2 to the cell suspension results in sufficient response allowing the detection of electronically excited species. To test whether 5 mM H 2 O 2 affects the viability of cells, the number of live cells was determined after the addition of 5 mM H 2 O 2 . Figure 1C shows that the cell viability remained unaffected by the addition of 5 mM H 2 O 2 to the cell suspension. Based on the results from one-dimensional ultra-weak photon emission, EPR spin-trapping spectroscopy, and cell viability, the concentration of 5 mM H 2 O 2 was chosen for further experiments.
The detection of triplet excited carbonyls by two-dimensional ultraweak photon emission. To visualize the formation of 3 (R 5 O)* in cell suspension treated with H 2 O 2 , the two-dimensional ultra-weak photon emission was examined using a highly sensitive CCD camera. Figure 2 shows the two-dimensional ultra-weak photon emission measured from cell suspension after the addition of H 2 O 2 . The treatment of cell suspension with H 2 O 2 resulted in the pronounced increase in two-dimensional ultra-weak photon emission compared to control. To quantify the differences in two-dimensional ultra-weak photon emission from cell suspension treated with H 2 O 2 , the spatial profile of photon emission in the middle strip of the image was used. The number of counts of two-dimensional ultra-weak photon  Detection of triplet excited carbonyls by one-dimensional ultraweak photon emission. To study the kinetics of ultra-weak photon emission, one-dimensional ultra-weak photon emission was measured using low-noise PMT. Spontaneous one-dimensional ultra-weak photon emission from cell suspension measured in the absence of H 2 O 2 shows a steady-state level, whereas the addition of H 2 O 2 after 30 min induced pronounced enhancement of one-dimensional ultraweak photon emission followed by slow decay (Fig. 3A). To confirm that one-dimensional ultra-weak photon emission is in the region of the spectrum assigned to 3 (R 5 O)*, the one-dimensional ultra-weak photon emission was measured using long-pass edge filter (600 nm). Figure 3B shows that one-dimensional ultra-weak photon emission  was significantly suppressed in the presence of the long-pass edge filter. To determine the participation of 1 O 2 in one-dimensional ultra-weak photon emission originated from red region of the spectrum, the effect of histidine, 1 O 2 scavenger, on one-dimensional ultra-weak photon was measured. Figure 3C demonstrates that the one-dimensional ultra-weak photon emission in the red region was almost fully suppressed in the presence of histidine. These results prove that most of the one-dimensional ultra-weak photon emission originates from the blue-green region of the spectrum known to be associated with the photon emission of 3 (R 5 O)*. Moreover, the measurement of one-dimensional ultra-weak photon emission reveals that the formation of 3 (R 5 O)* starts immediately after the addition of H 2 O 2 to the cell suspension.
Formation of singlet oxygen detected by confocal laser scanning microscopy. To visualize the formation of 1 O 2 in U266 cells caused by the addition of H 2 O 2 , singlet oxygen sensor green (SOSG) fluorescence was detected by confocal laser scanning microscopy. In spite of its high selectivity to 1 O 2 without any undesirable interaction with O 2 N2 and HO N35 , the SOSG has been reported to suffer from selfphotosensitization 35 . To prevent the formation of 1 O 2 by SOSG itself, the dye was strictly protected from light exposition during the whole experiment. Figure 4A shows the combination of channels for SOSG fluorescence (in green) and Nomarski DIC (greyscale) of U266 cells treated with H 2 O 2 for 0, 30, 60 and 90 min. Negligible SOSG fluorescence detected from untreated U266 cells was due to the weak SOSG fluorescence upon excitation by light at 480 nm prior to the reaction with 1 O 2 . The formation of SOSG endoperoxide by the cycloaddition of 1 O 2 precludes the intramolecular electron transport thus resulting in the enhancement of the fluorescence. The intensity of SOSG fluorescence was pronouncedly higher in the H 2 O 2 -treated U266 cells compared to untreated U266 cells. Figure 4B represents the integral distribution of SOSG fluorescence intensity within the corresponding upper image. Although SOSG was originally devised for extracellular applications, several studies reported its penetration inside of cells 36,37 . To test whether SOSG penetrates into U266 cells, SOSG fluorescence was measured in the multiple layers of sample (Fig. 5). The presence of SOSG fluorescence in the multiple layers of sample reveals the formation of 1 O 2 inside of U266 cells after the addition of H 2 O 2 .
Formation of singlet oxygen detected by EPR spin-trapping spectroscopy. To study the kinetics of 1 O 2 formation in the cell suspension, EPR spin-trapping spectroscopy was used. The spintrapping was accomplished by utilizing the oxidation of hydrophilic diamagnetic 2,2,6,6-tetramethyl-4-piperidone (TMPD) by 1 O 2 known to yield paramagnetic 2, 2, 6, 6-tetramethyl-4piperidone-1-oxyl (TEMPONE). The observation that no TEMPONE EPR signal was detected in untreated cell suspension indicates that no 1 O 2 was formed in the untreated cell suspension (Fig. 6). The addition of H 2 O 2 to the cell suspension caused the formation of TEMPONE EPR signal. To confirm that EPR signal is attributed to TEMPONE, the EPR spectrum of pure TEMPONE was measured. The simulation of TEMPONE EPR spectra using one spectral component with the hyperfine coupling constant a N 5 16 G provided agreement with the hyperfine coupling constant described   Quantitative analysis of triplet excited carbonyls and singlet oxygen formation. In order to quantify 3 (R 5 O)* and 1 O 2 , two-dimensional ultra-weak photon emission, one-dimensional ultra-weak photon emission, SOSG fluorescence and TEMPONE EPR signal were plotted as a function of H 2 O 2 treatment period. To quantify twodimensional images of ultra-weak photon emission from cell suspension treated with H 2 O 2 , the aera below the curve of spatial profile of photon emission at different strips of the image was counted. The intensity of the two-dimensional ultra-weak photon emission decreased by 50% and 62% after the addition of H 2 O 2 for 60 and 90 min compared to 30 min (Fig. 7A). To evaluate onedimensional ultra-weak photon emission from cell suspension treated with H 2 O 2 , the area below curve was counted over the 30 min of the treatment (Fig. 7B). Two-dimensional ultra-weak photon emission measured from the cell suspension decreased by 55% and 72% after the addition of H 2 O 2 for 60 and 90 min compared to 30 min. To evaluate the SOSG fluorescence obtained using confocal laser scanning microscopy, the intensity of SOSG fluorescence within confocal images was analyzed by image analysis (Fig. 7C). Due to the high background of SOSG fluorescence, the intensity of SOSG fluorescence observed in untreated U266 cells was subtracted from the SOSG fluorescence observed in treated U266 cells prior to the evaluation. The SOSG fluorescence was decreased by 50% and 80% after the addition of H 2 O 2 for 60 and 90 min compared to 30 min. The plotting of the height of the center peak of TEMPONE EPR spectrum against the time showed that TEMPONE EPR signal decreased by 50% and 75% after the addition of H 2 O 2 for 60 and 90 min compared to 30 min (Fig. 7D). These results show that the formation of both 3 (R 5 O)* and 1 O 2 decays rapidly shortly after the addition of H 2 O 2 to cell suspension followed by slow decay over the whole time of experiment.
The involvement of the medium in the formation of electronically excited species. In order to evaluate the effect of the medium on the formation of electronically excited species, the cell suspension was gently centrifuged and the supernatant was taken for further experiments. The addition of H 2 O 2 to both cell suspension and cellfree medium resulted in significant increase in both the one-dimensional ultra-weak photon emission and TEMPONE EPR signal. The comparison of results revealed that the one-dimensional ultra-weak photon emission originated from the cell-free medium was lower by   17% compared to the cell suspension (Fig. 8A). On the other hand, the TEMPONE EPR signal detected from cell-free medium treated with H 2 O 2 was higher by 44% compared to the TEMPONE EPR signal detected from cell suspension treated with H 2 O 2 (Fig. 8B). These results suggest that the contribution of the cell-free medium to the formation of electronically excited species is not negligible.

Discussion
Two-dimensional ultra-weak photon emission confirms the formation of 3 (R 5 O)* upon the addition of H 2 O 2 to cell suspension. Based on the results obtained by one-dimensional ultra-weak photon emission, it is concluded here that the formation of 3 (R 5 O)* occurs immediately after the addition of H 2 O 2 to cell suspension. Spectral analysis of the ultra-weak photon emission from different samples such as rat perfused lung, rat brain and liver homogenate, spinach mitochondria, cotyledons of etiolated seedlings of Cicer arietinum L., DNA, hemodialysis plasma cells, esophageal carcinoma cell line, porcine ex-vivo skin model induced by different treatment, showed that ultra-weak photon emission originates mostly from 3 (R 5 O)* 14,[43][44][45][46][47][48] . Several lines of evidence have been provided that 3 (R 5 O)* is formed by decomposition of ROOR or ROOOOR 11,13 . The decomposition of ROOR involves the cleavage of oxygen-oxygen and carbon-carbon bonds by two different mechanisms. The concerted mechanism involves the simultaneous cleavage of oxygen-oxygen and carbon-carbon bonds. The diradical mechanism is a two-step reaction involving the cleavage of oxygen bound prior to the cleavage of carbon-carbon bond 15 . Tetroxide is formed by the recombination of two ROO N . It has been demonstrated that t-butyl hydroperoxide does not participate in the ROOOOR formation due to the fact that the presence of a-hydrogen is required in order to undergo the Russell pathway 24,49 . Based on this observation, it was established that solely the primary and secondary ROO N are involved in the ROOOOR formation, whereas the tertiary ROO N undergoes propaga-tion of lipid peroxidation and protein oxidation. The conditions required for the Russell mechanism makes it rather unlikely, but not impossible, to occur in biological system.
Based on the data obtained from EPR spin-trapping spectroscopy itis concluded here that the formation of 1 O 2 occurs after the addition of H 2 O 2 to cell suspension. The visualization of 1 O 2 formation in the multiple layers of sample confirmed that 1 O 2 was produced inside the U266 cells. These results clearly show that U266 cells are oxidized upon the exposure of cell suspension to H 2 O 2 resulting in 1 O 2 formation. It has been previously demonstrated in the chemical systems that 1 O 2 is formed by the decomposition of organic hydroperoxides such as linoleic acid hydroperoxides, thymidine hydroperoxides, tbutyl hydroperoxides 9,50-52 . The decomposition of linoleic acid hydroperoxide formed by the photooxidation using methylene blue has been shown to result in the formation of 1 O 2 as confirmed by chemical trapping and monomol photon emission in near infra-red region 27 . The decomposition of ROOH results in the formation of ROO N which leads to the formation of unstable ROOOOR known to decompose to 1 O 2 , R 5 O, and ROH via the Russell mechanism 23 . The experimental evidence indicates that 1 O 2 is generated at a yield close to 10% by the Russell mechanism 52 . Furthermore, 3  In conclusion, the addition of H 2 O 2 to the cell suspension results in the formation of electronically excited species by three simultaneous reactions: the oxidation of cellular components [39][40][41][42] , the oxidation of growth medium components, especially free amino acids and fetal bovine serum, which might further cause the oxidation of the cellular components, and the oxidation of the medium components.  In this study, we mainly focused on the detection of electronically excited species and possible explanation of their formation rather than the place of their origin. The detail characterization of the site of origin of the electronically excited species will be the subject of the forthcoming study.

Methods
Cell Culture. Human MM cell line U266 was used in this study 53 (Fig. 9). U266 cells were grown in RPMI-1640 supplemented with 2 mM L-glutamine, 10% FBS, antibiotics at 37uC in humidified 5% CO 2 atmosphere. Viability of the cells was measured by Trypan Blue viability test. Subsequently, 10 ml of cell suspension was mixed with 10 ml of 0.4% Trypan Blue. Cells were counted by TC20 automated cell counter (Bio-Rad Laboratories, California, USA).
Two-dimensional ultra-weak photon emission imaging. Two-dimensional ultraweak photon emission was detected by CCD camera installed in a temperature controlled black box placed in a black painted inner darkroom. The measurement systems inside the inner darkroom were controlled and data were recorded with the computer located in the outer darkroom. The highly sensitive CCD camera VersArray 1300B (Princeton instruments, Trenton, NJ, USA) with spectral sensitivity in the range 350 to 1000 nm and close to 90% quantum efficiency in the visible range of the spectra was used to record the 2-D spectra. Objective lens of 50 mm focal distance (F mount Nikkor 50-mm, f: 1,2, Nikon) was used to enhance the light collecting efficiency. The CCD element was cooled down to 2110uC in order to reduce the dark count. The following parameters were used during the measurement: scan rate 100 kHz; gain 2; image format 1340 3 1300 pixels; binning mode 4; distance between detector and the reflecting mirror 40 cm; and accumulation time 30 min.
One-dimensional ultra-weak photon emission. One-dimensional ultra-weak photon emission was detected by PMT system installed in a black painted inner darkroom. The measurement systems inside the inner darkroom were controlled and data were recorded with the computer located in the outer darkroom. A low-noice PMT R7518P, sensitive in the spectral range 185 to 730 nm, and a photon counting unit C9744 (Hamamatsu Photonics K.K., Iwata City, Japan) were employed to measure one-dimensional photon emission. To reduce the dark noise, PMT was cooled down to 230uC using thermoelectric cooler C9143 (Hamamatsu Photonics, K.K., Iwata City, Japan). All the measurements were recorded at 2960 mV. The PMT was placed 5 cm above the sample during the measurement. In order to cut off the blue-green region of the spectra long-pass edge interference filter (600 nm) was used.
Detection of singlet oxygen by electron paramagnetic resonance spin-trapping spectroscopy. EPR spin-trapping spectroscopy was used to monitor the formation of 1 O 2 in cell suspension. Hydrophilic compound TMPD was used in order to detect 1 O 2 . To eliminate the impurity of TMPD EPR signal, the TMPD was purified twice by vacuum distillation. Cell suspension was treated with 5 mM H 2 O 2 in the presence of 50 mM TMPD and culture medium. Cell suspension previously treated with H 2 O 2 were put into a glass capillary tube (Blaubrand intraMARK, Brand, Germany) and EPR spectra were recorded using an EPR spectrometer MiniScope MS400 (Magnetechh GmbH, Berlin, Germany). EPR conditions were as follows: microwave power, 10 mW; modulation amplitude, 1 G; modulation frequency, 100 kHz, sweep width, 100 G, scan rate, 1.62 Gs 21 , gain 500.
Singlet oxygen imaging by confocal laser scanning microscopy and image analysis. SOSG (Molecular Probes Inc. Eugene, OR, USA) was applied to the U266 cells in order to visualize 1 O 2 production. Cell suspension treated with H 2 O 2 was incubated with 50 mM SOSG for 30 min in dark. Following incubation, U266 cells were gently washed with 20 mM K-buffer, and consequently the 1 O 2 imaging was performed by a confocal laser scanning microscope, Fluorview 1000 (Olympus Czech Group, Prague, Czech Republic). The transition images were obtained by transmitted light detection module with 405 nm excitation with a near-ultraviolet (UV) laser diode and Nomarski DIC filters (Olympus). Simultaneously visualized fluorescence channel resulted from excitation by a 488 nm line of argon laser, representing the signals for SOSG fluorescence detected by 505-525 nm emission filter set. The proper intensity of lasers was set according to unstained samples at the beginning of each experiment 54 . The integral distribution of signal intensity (ranging from 0 to 4095 levels of brightness) within the images was evaluated using software FV10-ASW 3.0 Viewer (Olympus).