Annals of Biomedical Engineering (Ó 2016) DOI: 10.1007/s10439-016-1557-y
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure During the Deep Photodynamic Therapy In Vitro WEITAO LI,1 DONG HUANG,1 YAN ZHANG,1 YANGYANG LIU,1 YUEQING GU,2 and ZHIYU QIAN1 1 Department of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao Street, Nanjing 210016, China; and 2Department of Biomedical Engineering, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, China
(Received 31 December 2015; accepted 27 January 2016) Associate Editor James J. Moon oversaw the review of this article.
Abstract—Photodynamic therapy (PDT) is an effective noninvasive method for the tumor treatment. The major challenge in current PDT research is how to quantitatively evaluate therapy effects. To our best knowledge, this is the first time to combine multi-parameter detection methods in PDT. More specifically, we have developed a set of system, including the high-sensitivity measurement of singlet oxygen, oxygen partial pressure and fluorescence image. In this paper, the detection ability of the system was validated by the different concentrations of carbon quantum dots. Moreover, the correlation between singlet oxygen and oxygen partial pressure with laser irradiation was observed. Then, the system could detect the signal up to 0.5 cm tissue depth with 660 nm irradiation and 1 cm tissue depth with 980 nm irradiation by using up-conversion nanoparticles during PDT in vitro. Furthermore, we obtained the relationship among concentration of singlet oxygen, oxygen partial pressure and tumor cell viability under certain conditions. The results indicate that the multi-parameter detection system is a promising asset to evaluate the deep tumor therapy during PDT. Moreover, the system might be potentially used for the further study in biology and molecular imaging. Keywords—Near infrared spectrum, Nanoparticles, Fluorescence imaging, Therapy assessment.
INTRODUCTION Photodynamic therapy (PDT) is an effective cancer treatment method. Comparing with conventional treatments, such as surgery, chemotherapy and radiotherapy, PDT is less or non-invasive. During PDT, the thermal production was based on photochemical reaction, thus the target tissue and cells were destructed. The first generation photosensitizers mainly include Address correspondence to Zhiyu Qian, Department of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao Street, Nanjing 210016, China. Electronic mail: [email protected], [email protected]
hematoporphyrin and its derivatives. However, they have many disadvantages, such as complex composition and poor selectivity. The second generation photosensitizers mainly include Visudyne (Verteporfin Injection), 5-ALA, HMME, hypocrellins, and phthalocyanine. Among them, the most commonly used photosensitizer is ZnPc which belongs to phthalocyanine. Compared with the first generation photosensitizers, the physical and chemical characteristics of the second generation photosensitizers are more stable and the absorption coefficient in the red zone is 10–50 times higher than that of the hematoporphyrin.2,17 The first step of PDT is injecting a certain amount of substance, which can naturally attach to the cells in the patient. After approximate 10 h, the concentration of ZnPc in healthy tissues decreases, while it remains unchanged in tumor cells. When the areas, highly concentrated in ZnPc, are irradiated with 660 nm laser, the photo-chemical reaction between the light and ZnPc occurs. The main consequence of this reaction is the destruction of the cells.7,10 During PDT, it needs the multi-parameter synchronous detection method and the tissue penetration depth of excitation light is low. Several single-parameter detection methods were utilized to investigate the therapy process and assessment.22 These methods verified that the 1 O2 is correlated strongly to the cell survival, but they did not obtain the valid quantitative relation between them.15,16 As we known, the current PDT mainly focus on skin tumor and monitor single parameter per time. Here, we utilized the up-conversion nanoparticles (FASOCUCNP-ZnPc) performing PDT experiment, which could kill tumor cells no less than 10 mm in depth using 980 nm laser.5,6 Moreover, the system could monitor multi-parameters during deep tumor therapy. Our main object is to monitor the multiple parameters during the deep PDT in vitro, and estimate the Ó 2016 Biomedical Engineering Society
LI et al.
viability of the target tumor precisely, which makes it to be an asset for doctors.11,14,20
pled and digitalized by a 16-bit data acquisition card (DAQ, AI USB-6211, National Instrument) and processed by the computer. Thus, we obtain the concentration of singlet oxygen by this indirect method. Furthermore, a commercial NeoFox O2 sensor (Ocean Optics) is used to detect oxygen concentration in the solution. Moreover, the high sensitivity CCD (PIXIS1024, B_eXcelon) is utilized to capture the realtime images of weak fluorescence. The maximum imaging size is 1024 9 1024; the pixel size is 13 9 13 lm2; the lowest operating temperature is 270 °C. The integrated air cooling system can ensure the long time continuous operation in the experiment. The NeoFox O2 sensor and CCD are integrated into the system by connecting with the computer via USB port and are controlled with the lab-made interface software. The whole system is sealed in a black box to avoid external light interferences.
MATERIALS AND METHODS Physical Description System Setup The Minimally Invasive Fiber Multi-parameter Monitoring and Imaging (MIFMI) system is schematically described in Fig. 1. The main parts in the system are a high sensitive CCD (PIXIS1024B_eXcelon, Princeton Instruments), an O2 sensor (NeoFox, Ocean optics), an optical fiber probe (Nanjing Fiberglass Research & Design Institute), and a laser source (MTOlaser). The laser source emits light with a fixed wavelength (660 or 980 nm), which is 9 mm in diameter and has a Gaussian distribution.4 The laser beam passes through the Filter1 (in the light of 625–670 nm the transmission rate is 93%, the remaining transmission band rate is close to 1%, BP650, Giai Photonics Co., Ltd) and a beam expander (GIAI-10.6-C0.57:4.5D1.55-FX,Giai Photonics Co., Ltd). Then a light spot with diameter of 14 mm is delivered to a sample pool. A homemade optical fiber probe with the diameter of 0.8 mm is employed to collect the weak fluorescence produced by 1O2 and Luminol reaction. The delayed luminescence (peak wavelength is 480 nm) is collected by one of the optical fibers in the probe. After passing through the Filter2 (in the light of 350–570 nm the transmission rate is 95%, the remaining transmission band rate close to 1%, SP550, Giai Photonics Co., Ltd), the fluorescence photon is captured by a photoelectric multiplier tube (PMT CR 186, Beijing Hamamatsu Corporation), whose spectral response range is 300–650 nm and the peak wavelength is 420 nm. The current generated by PMT is translated to voltage by using an I/V circuit and multiply channels amplify circuit. Then, the analog voltage is sam-
Interface and Control Software The design of user interface for the control system is another important part of MIFMI. The user interface software integrated several equipment, including CCD image capture part, O2 pressure collection part, and weak fluorescence monitor part. The main functions of the software include hardware connection detection, image acquisition, singlet oxygen concentration (fluorescence intensity) collection, oxygen partial pressure and temperature collection and image processing. MIFMI integrates with multi-parameter detection methods synchronously. The software development was based on the CCD drivers and settings supplied by the SITK of PI enterprise, which contained two important aspects: exposure time and cooling temperature. To reduce the image blurring due to the motion of the object and obtain better signal to noise ratio (SNR), we chose exposure time 0.008 ms. Furthermore, we chose
USB
CCD the optical fiber probe Coupler PMT
different thickness chicken tissues
Filter2
I/V change circuit
Multiply channels amplify circuit
PCI DAQ PC
Filter1 Fiber Interface
Laser Beam expander Sample
FIGURE 1. Schematic of the MIFMI system for deep PDT monitoring.
NeoFox
USB
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure
270 °C as cooling temperature to obtain higher quality images. We utilized DAQmx9.3 (National Instrument) to develop the data collection program. The curves of some parameters (such as fluorescence intensity, and concentration of O2) could be displayed and the related data could be stored in real time. Moreover, we utilized SIKT to implement the real-time imaging collection programs. Considering the image format compatibility between IMAQ and SITK, we employed a raw image format (TIFF, 16-bit unit). The resolution is selectable in 1024 9 1024 pixels, 512 9 512 pixels or any pixels defined by user. The ROI (Region of Interest) analysis tool facilitated the on-line image processing. Then, the histogram, maximum value, minimum value, average value, standard deviation of the ROI could be calculated automatically. Moreover, a pseudo colored mask was overlapped on the gray image to highlight the target region. The sequence diagram is shown in Fig. 2a. MIFMI detects the PO2 and 1O2 synchronously. During the whole process, the CCD captures fluorescence images at a certain frequency. Figure 2b shows the software flow chart. Here, we present the general procedures to operate the system. The mainly operations include device connection and CCD parameters setting. After
that, we could acquire the oxygen pressure, fluorescence intensity and CCD image simultaneously. At last, we could use ROI analysis to process the raw data. Figure 2c shows the designed user interface based on LabVIEW 8.5 for multiply parameters detection. The detection frame possessed ‘Physical Channel’, ‘Clock Source’, ‘Max Value’, ‘Min Value’, ‘Sample Model’, ‘Sample Rate’, ‘Channel Samples’, ‘Cutoff Frequency’ and ‘Input Configure’ toggles, which were used to select configuration information (information or parameters) during oxygen pressure and fluorescence intensity monitoring. Furthermore, the results could be saved and displayed in three graph frames. Several experiments have been designed to evaluate the performance of MIFMI.16 Experimental Methods To validate the MIFMI system, a carbon quantum dots (CQs) solution has been explored. We configured 0.1 mol/L carbon quantum dots (Janus New-Materials Co., Ltd) solution. Then the solution was diluted 25 times (group d), 125 times (group c) and 1250 times (group b), respectively. Group a was a pure water sample. The excitation wavelength of the laser was 365 nm and the center wavelength of fluorescence was
FIGURE 2. Design of user interface for deep PDT system. (a) Time flow; (b) Software flow chart; (c) The photo of graphic user interface (GUI).
LI et al.
430 nm. To minimize the interference of the excitation light to the fluorescence signal, we adjusted the laser and samples in the appropriate position. The spectra of different solutions (Table 1) in 660 nm laser irradiation were investigated. Based on these spectra, we can select the optimized excitation wavelength in MIFMI system. The different concentration of Luminol (N-(4-aminobutyl)-N-ethylisoluminol, Sigma-Aldrich) and ZnPc (Q103715, D&B) solutions were prepared. PBS (Phosphate Buffered Salin) is a kind of organic solvent. We chose 1 ml PBS as the control group. The solutions with different concentration of Luminol and ZnPc is shown in Table 1. Then the UV–vis-NIR spectra of five solutions were obtained using Fluorescence Spectrometer (LS 55, PerkinElmer). Then we detected the oxygen partial pressure in PDT solution independently. Here, we chose air (standard atmospheric pressure), pure water, Luminol (50 lM/L) and ZnPc (20 lM/L) as samples. The volume of the sample solutions were 1 ml, except for the air sample. We detected the oxygen partial pressure in the air (duration time is 10 s), in the pure water (duration time is 20 s), in the Luminol solution (duration time is 36 s), in the ZnPc solution (duration time is 30 s), respectively. Then, we added ZnPc into Luminol
TABLE 1. The concentration of sample solutions. Groups
Sample solutions
A B C
PBS (1 ml) Luminol (100 lM,1 ml) Luminol (100 lM,0.5 ml) and ZnPc (100 lM,0.5 ml) Luminol (500 lM,0.5 ml) and ZnPc (100 lM,0.5 ml) Luminol (500 lM,0.5 ml) and ZnPc (200 lM,0.5 ml)
D E
(a)
solution, the changes of oxygen partial pressure during the whole process (duration time is 100 s) were recorded.18,21 Moreover, we explored the therapy depth by using FASOC-UCNP-ZnPc5 and ZnPc. The tissues with different thicknesses were designed. Our study focused on the detection of PO2 and 1O2 by using the present MIFMI system. We prepared two groups of solutions of PDT. One was mixed 50 lM/L Luminol (0.5 ml) with 20 lM/L ZnPc (0.5 ml). The other was mingled 50 lM/L Luminol (0.5 ml) with 20 lM/L FASOCUCNP-ZnPc (0.5 ml). Moreover, the different thickness of chicken breast was provided. The thickness was 0.0, 2.5, 5, 7.5, and 10 mm, respectively. The position of these tissues was shown in Fig. 1. First, we adjusted the fluorescence signal to zero using the 1 mL PBS. Second, we utilized 1 mL ZnPc and Luminol in different depth models using 660 nm irradiation. Third, we used 1 mL FASOC-UCNP-ZnPc and Luminal with 980 nm irradiation. Here, we aimed to obtain the maximum thickness under 660 nm and 980 nm irradiation with PMT at 1000 V. Furthermore, the paper explored the relationship among oxygen partial pressure, concentration of singlet oxygen and tumor cell viability under 980 nm irradiation.1,23 In the experiment, we prepared DMSO (dimethylsulfoxide), 5, 10, 15, 20 lM/L FASOC-UCNP-ZnPc, 50 lM/L Luminol and U87 tumor (ATCC). To study tumor cell viability, we cultured six groups (each group included six holes) of U87 for 24 h in 96 hole cell culture plate by using culture medium and different concentrations FASOCUCNP-ZnPc solution. After that, we detected tumor cell viability in the condition without and with laser irradiation for 5 min by MTT test, respectively. Then, we obtained the optimal photosensitizer concentration, at which the minimum tumor cell survival rate occured. In addition, we cultured U87 and performed experiment using the optimal photosensitizer concentration.
(b) group a
Voltage (V)
3 2.5
group b
2 1.5
group c
1 0.5
group a
group b
group c
group d
group d
FIGURE 3. Performance testing of the MIFMI system: (a) Intensity of fluorescence in different solutions; (b) Fluorescence images.
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure
of Luminol solutions. Comparing with group D and group E, the fluorescence peak rises as the concentration of ZnPc increases. Moreover, the detected peak wavelength is 480 nm, which is in the expected dtection wavelength range from 420 to 540 nm. The detected peak wavelength is very helpful to choose the properties of the filter in the MIFMI. In Fig. 5, the baseline of the oxygen partial pressure is 20.7% (arrow a), which was measured in the dark environment. The oxygen partial pressures are 30.6, 26.3, and 17.7% when the probe was in the pure water (arrow b), Luminol solution (arrow c), and ZnPc solution (arrow d), respectively. Then, the oxygen partial pressure as indicated by arrow e is 26.3% in Luminol with several fluctuations. After that, 0.5 mL ZnPc was dissolved in the solution with amount 1 mL totally and the oxygen partial pressure increases sharply to 54% (arrow f). With the photodynamic chemical reacting, the oxygen partial pressure in PDT solution decreases and reaches 30.9% in the end. Under the radiation at 660 nm, the fluorescence intensity curves of the models with different depth from 0 to 10 mm are shown in Fig. 6a. Considering the higher detection depth under the radiation at 900 nm, the fluorescence intensity curves shown in Fig. 6b are at the depth of 7.5 and 10 mm. In summary, all the data is shown in Table 2. We chose 10 mm depth model in the following synchronous experiment. Figure 7a shows that the tumor cell viability is at a high level without laser irradiation. Figure 7b shows that the tumor cell viability decreases with the increasing of FASOC-UCNP-ZnPc concentration after 5 min laser irradiation. Thus, we chose 20 lM/L as an optimal concentration and conduct 5 group experiments. After 5 min laser irradiation, the tumor cell viability results are shown in Fig. 7c. Then, the singlet oxygen and oxygen partial pressure were detected by using the
RESULTS The intensity of the fluorescence was acquired by the optical probe and converted to voltage signal shown in Fig. 3a. The intensity values were significantly different according to the different concentrations of CQs in the solution. Moreover, the related fluorescence images captured by CCD are shown in Fig. 3b, which include gray, pseudo, and the overlap images (from the left column to right column). The images show fluorescence of CQs in the solutions. Thus, the results prove that the present system has the ability to detect the weak fluorescence in the solution. The spectra of the five groups (Table 1) are shown in Fig. 4. The group A (PBS) and group B (Luminol) solutions do not have any peaks in the spectra. By analyzing group C and group D, we found that there was no signal difference in the different concentrations 1.6
E
1.4
Intensity (Cd)
1.2
C 1.0 0.8
D
0.6
A 0.4
B
0.2
400
440
480
520
560
600
Wavelength (nm)
FIGURE 4. Spectra of different PDT solutions in Table 1. A: PBS(1 ml); B: Luminol(100 lM,1 ml); C: Luminol(100 lM, 0.5 ml) and ZnPc(100 lM,0.5 ml); D: Luminol(500 lM,0.5 ml) and ZnPc(100 lM,0.5 ml); E: Luminol(500 lM,0.5 ml) and Zn Pc(200 lM,0.5 ml).
55
f 45
35
b e
c 25
15
a 10
d 20
36
30
100
Duration time (s) FIGURE 5. Changes of oxygen partial pressure in different PDT solutions. Each process is indicated by arrow a to arrow f.
LI et al.
FIGURE 6. The deep chicken chest tumor cell experiment. (a) Under 660 nm laser radiation; (b) Under 980 nm radiation.
TABLE 2. Different deep under different conditions.
Tissue depth (mm) 0 2.5 5.0 7.5 10
660 nm + PBS(V)
660 nm + ZnPc(V)
980 nm + FASOC-UCNP-ZnPc (V)
0.000 0.000 0.000 0.000 0.000
0.310 0.230 0.110 0.045 0.020
NA NA NA 0.410 0.150
FIGURE 7. Different tumor cell viability and parameters synchronous monitor in deep PDT. (a) Without laser irradiation; (b) After 5 min laser irradiation; (c) After 5 min laser irradiation of the same concentration (20 lM/L) of FASOC-UCNP-ZnPc; (d) Relationship curve between PO2 (arrow a) and 1O2 (arrow b) during deep PDT.
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure TABLE 3. Oxygen partial pressure, singlet oxygen and tumor cell viability in deep PDT (15 Groups).
Group 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Oxygen partial pressure (%)
Singlet oxygen (V)
Tumor cell viability in deep PDT (%)
1.624 1.527 1.773 2.434 1.727 1.549 1.892 2.507 2.566 2.447 2.420 1.876 2.552 1.455 1.675
0.241 0.242 0.244 0.239 0.246 0.243 0.243 0.237 0.238 0.241 0.239 0.245 0.242 0.241 0.238
30.1 25.6 25.5 22.7 28.6 26.5 25.3 26.6 29.9 25.1 23.6 33.1 24.9 25.4 22.8
MIFMI system at the same time. Figure 7d shows the changes of both singlet oxygen (arrow b) and oxygen partial pressure (arrow a) simultaneously. The values of singlet oxygen, oxygen partial pressure and tumor cell viability are shown in Table 3. According to the original data in Table 3, the tumor cell viability, f(x,y), can be defined by the oxygen partial pressure (x) and the singlet oxygen (y) as follows. fðx; yÞ ¼ 3:4x2 þ 98181:8y2 833:6xy þ 186:6x 45240:9y þ 5239:9
ð1Þ
Here are several default parameters to make Eq. (1) valid. Briefly, the distance between laser and the samples is 10 cm; the power of laser is 250 mw; the concentration of FASOC-UCNP-ZnPc is 20 lM/L; the concentration of Luminol is 50 lM/L. Under the above conditions, Eq. (1) can provide a theoretical fundamental relationship among these three variables.
DISCUSSION This is the first time to introduce an integrated system to detect multiple parameters during deep PDT experiment. We designed an optical probe and a series of signal process circuits to obtain the weak fluorescence, which was produced by the reaction of Luminol and the singlet oxygen released by the photosensitizers (ZnPc and FASOC-UCNP-ZnPc) under laser irradiation. The fluorescence intensity is proportional to the concentration of Luminol and singlet oxygen. If the concentration of Luminol is constant, the concentration of singlet oxygen will be obtained indirectly.3 To acquire the oxygen partial pressure, we integrated a commercial detection device into the present system. The device was controlled by the synchronizing signal
from the user interface software. In brief, the MIFMI system can measure the most important two parameters in PDT, including the concentration of singlet oxygen and the oxygen partial pressure, at the same time. Comparing with the previously reported monitor systems, which can only detect single parameter, the present MIFMI system proposes an improved multiparameter monitor and analysis ability. In general, the main mechanism cancer treatment using PDT is the reaction between oxygen and photosensitizers under the light irradiation, which generates cytotoxic Reactive Oxygen Species (ROS) that directly destroys the tumor cells.24 It has been demonstrated that PDT has two ways to kill cells, defined as Type I and Type II mechanisms.12 The Type I mechanism involves hydrogen-atom abstraction or electron-transfer reactions between the photosensitizer and a substrate to yield free radicals and radical ions, which can interact with molecular oxygen to either generate ROS (such as superoxide anions and hydroxyl radicals) or can cause irreparable biological damage. The Type II mechanism results from an energy transfer between the photosensitizers and the molecular oxygen to produce singlet oxygen. In the mechanisms, the molecular oxygen and singlet oxygen have similar relationship. Due to a short lifetime (0.04 ls) of singlet oxygen, a small radiation area (0.02 lm), and the weak intensity of emission spectroscopy (1270 nm), the direct detection of singlet oxygen is very difficult. In this paper, we utilized Luminol as fluorescence probe to detect singlet oxygen indirectly.9,25 We employed the advantages of the chemical reaction between Luminol and singlet oxygen, which produced a delayed fluorescence signal. Comparing to the visible light, PDT triggered by NIR light enables the feasibility of deep-tissue PDT treatment. In carbon quantum dots concentration gradient experiment, we aimed to investigate a valid real-time monitor system during PDT for deep-seated tumors treatment. The amplitude of the voltage is obviously distinguished due to different concentration. Even if the concentration of solution is very low (0.08 lM/L), we could also obtain the signal at 0.18 V. Then, we acquired the spectral curves of several PDT solutions (Table 1) under 660 nm laser irradiation, which could guide us to optimize the wavelength (from 420 to 540 nm). Furthermore, we conducted oxygen partial pressure change curves during PDT, the results show that the oxygen partial pressure in Luminol and ZnPc solutions are 26.3 and 17.7%, respectively. When putting the ZnPc into Luminol, the maximum oxygen partial pressure increase is 54%, and the minimum increase is 30.9%.8,13,19 In order to evaluate the deep PDT treatment using FASOC-UCNP-ZnPc and MIFMI system, we
LI et al.
explored another experiment. Comparing with the ZnPc, the effective depth of FASOC-UCNP-ZnPc can be up to 10 mm, which is consistent with the result in former study. Here, we conducted the multi-parameter synchronous detection experiment using 980 nm laser and FASOC-UCNP-ZnPc under 10 mm chicken tissue. Finally, we derived the therapy evaluation model among the oxygen partial pressure, the concentration of singlet oxygen and the tumor cell viability. Moreover, the concentration of singlet oxygen, oxygen partial pressure, as well as the CCD images were displayed in the computer. These multiple parameters are more accurate to remind doctor of the current expectation of the survival rate of each dose of light irradiation. For example, the tumor cell viability curve will be calculated in real time using the relation equation (Eq. 1), which is the most important factor to evaluate the effects of PDT. In the following studies, the MIFMI system can be explored in the PDT animal experiments in vivo. In summary, we have successfully developed and verified a MIFMI system, which can measure oxygen partial pressure, singlet oxygen and capture fluorescence images simultaneously in vitro. Furthermore, the MIFMI system has been used to explore the relationship among the oxygen partial pressure, the concentration of singlet oxygen and the tumor cell viability in deep PDT experiments. It all boils down to relation equation and the specific conditions have been presented. Our studies demonstrate that the MIFMI system has the potential for the evaluation of deep PDT using multiple parameters.
ACKNOWLEDGMENTS This work is supported by ‘‘the Fundamental Research Funds for the Central Universities’’, NO. NS2015032. We thanked Dr. Xinzeng Wang for close reading of the manuscript and made corresponding revisions in the whole paper.
REFERENCES 1
Bown, S. G. Photodynamic therapy for cancers of solid organs. Lasers Electro Opt. Soc. 1:253–254, 2002. 2 Celli, J. P., B. Q. Spring, I. Rizvi, C. L. Evans, K. S. Samkoe, S. Verma, B. W. Pogue, and T. Hasan. Imaging and photodynamic therapy: mechanisms, monitoring, and optimization. Chem. Rev. 110:2795–2838, 2010. 3 Chen, T. C., L. Huang, C. C. Liu, P. J. Chao, and F. H. Lin. Luminol as the light source for in situ photodynamic therapy. Process Biochem. 47:1903–1908, 2012.
4
Crimaldi, J. P. Planar laser induced fluorescence in aqueous flows. Exp. Fluids. 44:851–863, 2008. 5 Cui, S., H. Chen, H. Zhu, J. Tian, X. chi, Z. Qian, S. Achilefu, and Y. Gu. Amphiphilic chitosan modified upconversion nanoparticles for in vivo photodynamic therapy induced by near-infrared light. J. Mater. Chem. 22:4861–4873, 2012. 6 Cui, S., D. Yin, Y. Chen, Y. Di, H. Chen, Y. Ma, S. Achilefu, and Y. Gu. In vivo targeted deep-tissue photodynamic therapy based on near-infrared light triggered upconversion nanoconstruct. ACS Nano. 7:676–688, 2013. 7 Garcez, A. S., E. R. Freqnani, H. M. Rodriquez, S. C. Nunez, C. P. Sabino, H. Suzuki, and M. S. Ribeiro. The use of optical fiber in endodontic photodynamic therapy. Is it really relevant? Lasers Med. Sci. 28:79–85, 2013. 8 Gehmert, S., S. Geis, P. E. Lamby, C. Roll, U. Braumandl, M. Hidayat, M. Sultan, B. Fuchtmeier, E. M. Jung, and L. Prantl. Evaluation of hyperbaric oxygen therapy for free flaps using planar optical oxygen sensors. Preliminary results. Clin. Hemorheol. Microcirc. 48:75–79, 2011. 9 Hackbarth, S., J. Schlothauer, A. Preuss, and B. Roder. New insights to primary photodynamic effects-singlet oxygen kinetics in living cells. J. Photochem. Photobiol. B 98:173–179, 2010. 10 Hewett, J., V. Nadeau, J. Ferquson, H. Moseley, S. Ibbotson, J. W. Allen, W. Sibbett, and M. Padgett. The application of a compact multispectral imaging system with integrated excitation source to in vivo monitoring of fluorescence during topical photodynamic therapy of superficial skin cancers. Photochem. Photobiol. 73:278– 282, 2001. 11 Jerjes, W., T. Upile, C. Alexander Mosse, Z. Hamdoon, M. Morcos, S. Morley, and C. Hopper. Prospective evaluation of 110 patients following ultrasound-guided photodynamic therapy for deep seated pathologies. Photodiagnosis Photodyn Ther. 8:297–306, 2011. 12 Jessica, T., C. Sandra, and C. Alison. Validation of a non-invasive fluorescence imaging system to monitor dermatological PDT. Photodiagnosis Photodyn. Ther. 7:86–97, 2010. 13 Jonas, R., T. Schaal, M. Krimmel, D. Gu¨licher, S. Reinert, and J. Hoffmann. Monitoring in microvascular tissue transfer by measurement of oxygen partial pressure: four years experience with 125 microsurgical transplants. J. Craniomaxillofac Surg. 41:303–309, 2013. 14 Jung, J., A. J. Nichols, O. J. Klein, E. Roussakis, and C. L. Evans. Label-free, longitudinal visualization of PDT response in vitro with optical coherence tomography. Isr. J. Chem. 52:728–744, 2012. 15 Kasˇ cˇa´kova´, S., A. Giuliani, S. Lacerda, A. Pallier, P. Mercere, E. Toth, and M. Refregiers. X-ray induced radiophotodynamic therapy (RPDT) using lanthanide micelles: beyond depth limitations. Nano Res. 8:2373–2379, 2015. 16 Niedre, M. J., A. J. Secord, M. S. Patterson, and B. C. Wilson. In vitro tests of the validity of singlet oxygen luminescence measurements as a dose metric in photodynamic therapy. Cancer Res. 63:7086–7094, 2003. 17 Pacheco, L., S. S. Stolik, J. De la Rosa, and A. De la Cadena. Laser diode system for interstitial photodynamic therapy. IEEE Lat. Am. Trans. 12:574–579, 2014. 18 Papkovsky, D. B., and R. I. Dmitriev. Biological detection by optical oxygen sensing. Chem. Soc. Rev. 42:8700–8732, 2013.
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure 19
Piffaretti, F., A. M. Novello, R. S. Kumar, E. Forte, C. Paulou, P. Nowak-Sliwinska, H. van den Bergh, and G. Wagnie`res. Real-time, in vivo measurement of tissular pO2 through the delayed fluorescence of endogenous protoporphyrin IX during photodynamic therapy. J. Biomed. Opt. 17:115007, 2012. 20 Sabino, C. P., A. S. Garcez, S. C. Nu´n˜ez, M. S. Ribeiro, and M. R. Hamblin. Real-time evaluation of two light delivery systems for photodynamic disinfection of Candida albicans biofilm in curved root canals. Lasers Med. Sci. 30:1657–1665, 2015. 21 Sargent, B. J., and D. A. Gough. Design and validation of the transparent oxygen sensor array. IEEE Trans. Biomed. Eng. 38:476–482, 1991.
22
Sunar, U. Monitoring photodynamic therapy of head and neck malignancies with optical spectroscopies. World J. Clin. Cases. 1:96–105, 2013. 23 Tokatli-Apollon, A., H. Lee, M. El-Rifai, and A. Punjabi. Upconversion nanoparticles for photodynamic therapy. Northeast Bioeng. Conf. 40:1–2, 2014. 24 Trushina, A. P., V. G. Goldort, S. A. Kochubei, and A. V. Baklanov. UV-photoexcitation of encounter complexes of oxygen O2–O2 as a source of singlet oxygen O2(1Dg) in gas phase. Chem. Phys. Lett. 485:11–15, 2010. 25 Wang, J., D. Xing, Y. He, and X. Hu. Experimental study on photodynamic diagnosis of cancer mediated by chemiluminescence probe. FEBS Lett. 523:128–132, 2002.
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure During the Deep Photodynamic Therapy In Vitro WEITAO LI,1 DONG HUANG,1 YAN ZHANG,1 YANGYANG LIU,1 YUEQING GU,2 and ZHIYU QIAN1 1 Department of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao Street, Nanjing 210016, China; and 2Department of Biomedical Engineering, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, China
(Received 31 December 2015; accepted 27 January 2016) Associate Editor James J. Moon oversaw the review of this article.
Abstract—Photodynamic therapy (PDT) is an effective noninvasive method for the tumor treatment. The major challenge in current PDT research is how to quantitatively evaluate therapy effects. To our best knowledge, this is the first time to combine multi-parameter detection methods in PDT. More specifically, we have developed a set of system, including the high-sensitivity measurement of singlet oxygen, oxygen partial pressure and fluorescence image. In this paper, the detection ability of the system was validated by the different concentrations of carbon quantum dots. Moreover, the correlation between singlet oxygen and oxygen partial pressure with laser irradiation was observed. Then, the system could detect the signal up to 0.5 cm tissue depth with 660 nm irradiation and 1 cm tissue depth with 980 nm irradiation by using up-conversion nanoparticles during PDT in vitro. Furthermore, we obtained the relationship among concentration of singlet oxygen, oxygen partial pressure and tumor cell viability under certain conditions. The results indicate that the multi-parameter detection system is a promising asset to evaluate the deep tumor therapy during PDT. Moreover, the system might be potentially used for the further study in biology and molecular imaging. Keywords—Near infrared spectrum, Nanoparticles, Fluorescence imaging, Therapy assessment.
INTRODUCTION Photodynamic therapy (PDT) is an effective cancer treatment method. Comparing with conventional treatments, such as surgery, chemotherapy and radiotherapy, PDT is less or non-invasive. During PDT, the thermal production was based on photochemical reaction, thus the target tissue and cells were destructed. The first generation photosensitizers mainly include Address correspondence to Zhiyu Qian, Department of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao Street, Nanjing 210016, China. Electronic mail: [email protected], [email protected]
hematoporphyrin and its derivatives. However, they have many disadvantages, such as complex composition and poor selectivity. The second generation photosensitizers mainly include Visudyne (Verteporfin Injection), 5-ALA, HMME, hypocrellins, and phthalocyanine. Among them, the most commonly used photosensitizer is ZnPc which belongs to phthalocyanine. Compared with the first generation photosensitizers, the physical and chemical characteristics of the second generation photosensitizers are more stable and the absorption coefficient in the red zone is 10–50 times higher than that of the hematoporphyrin.2,17 The first step of PDT is injecting a certain amount of substance, which can naturally attach to the cells in the patient. After approximate 10 h, the concentration of ZnPc in healthy tissues decreases, while it remains unchanged in tumor cells. When the areas, highly concentrated in ZnPc, are irradiated with 660 nm laser, the photo-chemical reaction between the light and ZnPc occurs. The main consequence of this reaction is the destruction of the cells.7,10 During PDT, it needs the multi-parameter synchronous detection method and the tissue penetration depth of excitation light is low. Several single-parameter detection methods were utilized to investigate the therapy process and assessment.22 These methods verified that the 1 O2 is correlated strongly to the cell survival, but they did not obtain the valid quantitative relation between them.15,16 As we known, the current PDT mainly focus on skin tumor and monitor single parameter per time. Here, we utilized the up-conversion nanoparticles (FASOCUCNP-ZnPc) performing PDT experiment, which could kill tumor cells no less than 10 mm in depth using 980 nm laser.5,6 Moreover, the system could monitor multi-parameters during deep tumor therapy. Our main object is to monitor the multiple parameters during the deep PDT in vitro, and estimate the Ó 2016 Biomedical Engineering Society
LI et al.
viability of the target tumor precisely, which makes it to be an asset for doctors.11,14,20
pled and digitalized by a 16-bit data acquisition card (DAQ, AI USB-6211, National Instrument) and processed by the computer. Thus, we obtain the concentration of singlet oxygen by this indirect method. Furthermore, a commercial NeoFox O2 sensor (Ocean Optics) is used to detect oxygen concentration in the solution. Moreover, the high sensitivity CCD (PIXIS1024, B_eXcelon) is utilized to capture the realtime images of weak fluorescence. The maximum imaging size is 1024 9 1024; the pixel size is 13 9 13 lm2; the lowest operating temperature is 270 °C. The integrated air cooling system can ensure the long time continuous operation in the experiment. The NeoFox O2 sensor and CCD are integrated into the system by connecting with the computer via USB port and are controlled with the lab-made interface software. The whole system is sealed in a black box to avoid external light interferences.
MATERIALS AND METHODS Physical Description System Setup The Minimally Invasive Fiber Multi-parameter Monitoring and Imaging (MIFMI) system is schematically described in Fig. 1. The main parts in the system are a high sensitive CCD (PIXIS1024B_eXcelon, Princeton Instruments), an O2 sensor (NeoFox, Ocean optics), an optical fiber probe (Nanjing Fiberglass Research & Design Institute), and a laser source (MTOlaser). The laser source emits light with a fixed wavelength (660 or 980 nm), which is 9 mm in diameter and has a Gaussian distribution.4 The laser beam passes through the Filter1 (in the light of 625–670 nm the transmission rate is 93%, the remaining transmission band rate is close to 1%, BP650, Giai Photonics Co., Ltd) and a beam expander (GIAI-10.6-C0.57:4.5D1.55-FX,Giai Photonics Co., Ltd). Then a light spot with diameter of 14 mm is delivered to a sample pool. A homemade optical fiber probe with the diameter of 0.8 mm is employed to collect the weak fluorescence produced by 1O2 and Luminol reaction. The delayed luminescence (peak wavelength is 480 nm) is collected by one of the optical fibers in the probe. After passing through the Filter2 (in the light of 350–570 nm the transmission rate is 95%, the remaining transmission band rate close to 1%, SP550, Giai Photonics Co., Ltd), the fluorescence photon is captured by a photoelectric multiplier tube (PMT CR 186, Beijing Hamamatsu Corporation), whose spectral response range is 300–650 nm and the peak wavelength is 420 nm. The current generated by PMT is translated to voltage by using an I/V circuit and multiply channels amplify circuit. Then, the analog voltage is sam-
Interface and Control Software The design of user interface for the control system is another important part of MIFMI. The user interface software integrated several equipment, including CCD image capture part, O2 pressure collection part, and weak fluorescence monitor part. The main functions of the software include hardware connection detection, image acquisition, singlet oxygen concentration (fluorescence intensity) collection, oxygen partial pressure and temperature collection and image processing. MIFMI integrates with multi-parameter detection methods synchronously. The software development was based on the CCD drivers and settings supplied by the SITK of PI enterprise, which contained two important aspects: exposure time and cooling temperature. To reduce the image blurring due to the motion of the object and obtain better signal to noise ratio (SNR), we chose exposure time 0.008 ms. Furthermore, we chose
USB
CCD the optical fiber probe Coupler PMT
different thickness chicken tissues
Filter2
I/V change circuit
Multiply channels amplify circuit
PCI DAQ PC
Filter1 Fiber Interface
Laser Beam expander Sample
FIGURE 1. Schematic of the MIFMI system for deep PDT monitoring.
NeoFox
USB
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure
270 °C as cooling temperature to obtain higher quality images. We utilized DAQmx9.3 (National Instrument) to develop the data collection program. The curves of some parameters (such as fluorescence intensity, and concentration of O2) could be displayed and the related data could be stored in real time. Moreover, we utilized SIKT to implement the real-time imaging collection programs. Considering the image format compatibility between IMAQ and SITK, we employed a raw image format (TIFF, 16-bit unit). The resolution is selectable in 1024 9 1024 pixels, 512 9 512 pixels or any pixels defined by user. The ROI (Region of Interest) analysis tool facilitated the on-line image processing. Then, the histogram, maximum value, minimum value, average value, standard deviation of the ROI could be calculated automatically. Moreover, a pseudo colored mask was overlapped on the gray image to highlight the target region. The sequence diagram is shown in Fig. 2a. MIFMI detects the PO2 and 1O2 synchronously. During the whole process, the CCD captures fluorescence images at a certain frequency. Figure 2b shows the software flow chart. Here, we present the general procedures to operate the system. The mainly operations include device connection and CCD parameters setting. After
that, we could acquire the oxygen pressure, fluorescence intensity and CCD image simultaneously. At last, we could use ROI analysis to process the raw data. Figure 2c shows the designed user interface based on LabVIEW 8.5 for multiply parameters detection. The detection frame possessed ‘Physical Channel’, ‘Clock Source’, ‘Max Value’, ‘Min Value’, ‘Sample Model’, ‘Sample Rate’, ‘Channel Samples’, ‘Cutoff Frequency’ and ‘Input Configure’ toggles, which were used to select configuration information (information or parameters) during oxygen pressure and fluorescence intensity monitoring. Furthermore, the results could be saved and displayed in three graph frames. Several experiments have been designed to evaluate the performance of MIFMI.16 Experimental Methods To validate the MIFMI system, a carbon quantum dots (CQs) solution has been explored. We configured 0.1 mol/L carbon quantum dots (Janus New-Materials Co., Ltd) solution. Then the solution was diluted 25 times (group d), 125 times (group c) and 1250 times (group b), respectively. Group a was a pure water sample. The excitation wavelength of the laser was 365 nm and the center wavelength of fluorescence was
FIGURE 2. Design of user interface for deep PDT system. (a) Time flow; (b) Software flow chart; (c) The photo of graphic user interface (GUI).
LI et al.
430 nm. To minimize the interference of the excitation light to the fluorescence signal, we adjusted the laser and samples in the appropriate position. The spectra of different solutions (Table 1) in 660 nm laser irradiation were investigated. Based on these spectra, we can select the optimized excitation wavelength in MIFMI system. The different concentration of Luminol (N-(4-aminobutyl)-N-ethylisoluminol, Sigma-Aldrich) and ZnPc (Q103715, D&B) solutions were prepared. PBS (Phosphate Buffered Salin) is a kind of organic solvent. We chose 1 ml PBS as the control group. The solutions with different concentration of Luminol and ZnPc is shown in Table 1. Then the UV–vis-NIR spectra of five solutions were obtained using Fluorescence Spectrometer (LS 55, PerkinElmer). Then we detected the oxygen partial pressure in PDT solution independently. Here, we chose air (standard atmospheric pressure), pure water, Luminol (50 lM/L) and ZnPc (20 lM/L) as samples. The volume of the sample solutions were 1 ml, except for the air sample. We detected the oxygen partial pressure in the air (duration time is 10 s), in the pure water (duration time is 20 s), in the Luminol solution (duration time is 36 s), in the ZnPc solution (duration time is 30 s), respectively. Then, we added ZnPc into Luminol
TABLE 1. The concentration of sample solutions. Groups
Sample solutions
A B C
PBS (1 ml) Luminol (100 lM,1 ml) Luminol (100 lM,0.5 ml) and ZnPc (100 lM,0.5 ml) Luminol (500 lM,0.5 ml) and ZnPc (100 lM,0.5 ml) Luminol (500 lM,0.5 ml) and ZnPc (200 lM,0.5 ml)
D E
(a)
solution, the changes of oxygen partial pressure during the whole process (duration time is 100 s) were recorded.18,21 Moreover, we explored the therapy depth by using FASOC-UCNP-ZnPc5 and ZnPc. The tissues with different thicknesses were designed. Our study focused on the detection of PO2 and 1O2 by using the present MIFMI system. We prepared two groups of solutions of PDT. One was mixed 50 lM/L Luminol (0.5 ml) with 20 lM/L ZnPc (0.5 ml). The other was mingled 50 lM/L Luminol (0.5 ml) with 20 lM/L FASOCUCNP-ZnPc (0.5 ml). Moreover, the different thickness of chicken breast was provided. The thickness was 0.0, 2.5, 5, 7.5, and 10 mm, respectively. The position of these tissues was shown in Fig. 1. First, we adjusted the fluorescence signal to zero using the 1 mL PBS. Second, we utilized 1 mL ZnPc and Luminol in different depth models using 660 nm irradiation. Third, we used 1 mL FASOC-UCNP-ZnPc and Luminal with 980 nm irradiation. Here, we aimed to obtain the maximum thickness under 660 nm and 980 nm irradiation with PMT at 1000 V. Furthermore, the paper explored the relationship among oxygen partial pressure, concentration of singlet oxygen and tumor cell viability under 980 nm irradiation.1,23 In the experiment, we prepared DMSO (dimethylsulfoxide), 5, 10, 15, 20 lM/L FASOC-UCNP-ZnPc, 50 lM/L Luminol and U87 tumor (ATCC). To study tumor cell viability, we cultured six groups (each group included six holes) of U87 for 24 h in 96 hole cell culture plate by using culture medium and different concentrations FASOCUCNP-ZnPc solution. After that, we detected tumor cell viability in the condition without and with laser irradiation for 5 min by MTT test, respectively. Then, we obtained the optimal photosensitizer concentration, at which the minimum tumor cell survival rate occured. In addition, we cultured U87 and performed experiment using the optimal photosensitizer concentration.
(b) group a
Voltage (V)
3 2.5
group b
2 1.5
group c
1 0.5
group a
group b
group c
group d
group d
FIGURE 3. Performance testing of the MIFMI system: (a) Intensity of fluorescence in different solutions; (b) Fluorescence images.
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure
of Luminol solutions. Comparing with group D and group E, the fluorescence peak rises as the concentration of ZnPc increases. Moreover, the detected peak wavelength is 480 nm, which is in the expected dtection wavelength range from 420 to 540 nm. The detected peak wavelength is very helpful to choose the properties of the filter in the MIFMI. In Fig. 5, the baseline of the oxygen partial pressure is 20.7% (arrow a), which was measured in the dark environment. The oxygen partial pressures are 30.6, 26.3, and 17.7% when the probe was in the pure water (arrow b), Luminol solution (arrow c), and ZnPc solution (arrow d), respectively. Then, the oxygen partial pressure as indicated by arrow e is 26.3% in Luminol with several fluctuations. After that, 0.5 mL ZnPc was dissolved in the solution with amount 1 mL totally and the oxygen partial pressure increases sharply to 54% (arrow f). With the photodynamic chemical reacting, the oxygen partial pressure in PDT solution decreases and reaches 30.9% in the end. Under the radiation at 660 nm, the fluorescence intensity curves of the models with different depth from 0 to 10 mm are shown in Fig. 6a. Considering the higher detection depth under the radiation at 900 nm, the fluorescence intensity curves shown in Fig. 6b are at the depth of 7.5 and 10 mm. In summary, all the data is shown in Table 2. We chose 10 mm depth model in the following synchronous experiment. Figure 7a shows that the tumor cell viability is at a high level without laser irradiation. Figure 7b shows that the tumor cell viability decreases with the increasing of FASOC-UCNP-ZnPc concentration after 5 min laser irradiation. Thus, we chose 20 lM/L as an optimal concentration and conduct 5 group experiments. After 5 min laser irradiation, the tumor cell viability results are shown in Fig. 7c. Then, the singlet oxygen and oxygen partial pressure were detected by using the
RESULTS The intensity of the fluorescence was acquired by the optical probe and converted to voltage signal shown in Fig. 3a. The intensity values were significantly different according to the different concentrations of CQs in the solution. Moreover, the related fluorescence images captured by CCD are shown in Fig. 3b, which include gray, pseudo, and the overlap images (from the left column to right column). The images show fluorescence of CQs in the solutions. Thus, the results prove that the present system has the ability to detect the weak fluorescence in the solution. The spectra of the five groups (Table 1) are shown in Fig. 4. The group A (PBS) and group B (Luminol) solutions do not have any peaks in the spectra. By analyzing group C and group D, we found that there was no signal difference in the different concentrations 1.6
E
1.4
Intensity (Cd)
1.2
C 1.0 0.8
D
0.6
A 0.4
B
0.2
400
440
480
520
560
600
Wavelength (nm)
FIGURE 4. Spectra of different PDT solutions in Table 1. A: PBS(1 ml); B: Luminol(100 lM,1 ml); C: Luminol(100 lM, 0.5 ml) and ZnPc(100 lM,0.5 ml); D: Luminol(500 lM,0.5 ml) and ZnPc(100 lM,0.5 ml); E: Luminol(500 lM,0.5 ml) and Zn Pc(200 lM,0.5 ml).
55
f 45
35
b e
c 25
15
a 10
d 20
36
30
100
Duration time (s) FIGURE 5. Changes of oxygen partial pressure in different PDT solutions. Each process is indicated by arrow a to arrow f.
LI et al.
FIGURE 6. The deep chicken chest tumor cell experiment. (a) Under 660 nm laser radiation; (b) Under 980 nm radiation.
TABLE 2. Different deep under different conditions.
Tissue depth (mm) 0 2.5 5.0 7.5 10
660 nm + PBS(V)
660 nm + ZnPc(V)
980 nm + FASOC-UCNP-ZnPc (V)
0.000 0.000 0.000 0.000 0.000
0.310 0.230 0.110 0.045 0.020
NA NA NA 0.410 0.150
FIGURE 7. Different tumor cell viability and parameters synchronous monitor in deep PDT. (a) Without laser irradiation; (b) After 5 min laser irradiation; (c) After 5 min laser irradiation of the same concentration (20 lM/L) of FASOC-UCNP-ZnPc; (d) Relationship curve between PO2 (arrow a) and 1O2 (arrow b) during deep PDT.
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure TABLE 3. Oxygen partial pressure, singlet oxygen and tumor cell viability in deep PDT (15 Groups).
Group 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Oxygen partial pressure (%)
Singlet oxygen (V)
Tumor cell viability in deep PDT (%)
1.624 1.527 1.773 2.434 1.727 1.549 1.892 2.507 2.566 2.447 2.420 1.876 2.552 1.455 1.675
0.241 0.242 0.244 0.239 0.246 0.243 0.243 0.237 0.238 0.241 0.239 0.245 0.242 0.241 0.238
30.1 25.6 25.5 22.7 28.6 26.5 25.3 26.6 29.9 25.1 23.6 33.1 24.9 25.4 22.8
MIFMI system at the same time. Figure 7d shows the changes of both singlet oxygen (arrow b) and oxygen partial pressure (arrow a) simultaneously. The values of singlet oxygen, oxygen partial pressure and tumor cell viability are shown in Table 3. According to the original data in Table 3, the tumor cell viability, f(x,y), can be defined by the oxygen partial pressure (x) and the singlet oxygen (y) as follows. fðx; yÞ ¼ 3:4x2 þ 98181:8y2 833:6xy þ 186:6x 45240:9y þ 5239:9
ð1Þ
Here are several default parameters to make Eq. (1) valid. Briefly, the distance between laser and the samples is 10 cm; the power of laser is 250 mw; the concentration of FASOC-UCNP-ZnPc is 20 lM/L; the concentration of Luminol is 50 lM/L. Under the above conditions, Eq. (1) can provide a theoretical fundamental relationship among these three variables.
DISCUSSION This is the first time to introduce an integrated system to detect multiple parameters during deep PDT experiment. We designed an optical probe and a series of signal process circuits to obtain the weak fluorescence, which was produced by the reaction of Luminol and the singlet oxygen released by the photosensitizers (ZnPc and FASOC-UCNP-ZnPc) under laser irradiation. The fluorescence intensity is proportional to the concentration of Luminol and singlet oxygen. If the concentration of Luminol is constant, the concentration of singlet oxygen will be obtained indirectly.3 To acquire the oxygen partial pressure, we integrated a commercial detection device into the present system. The device was controlled by the synchronizing signal
from the user interface software. In brief, the MIFMI system can measure the most important two parameters in PDT, including the concentration of singlet oxygen and the oxygen partial pressure, at the same time. Comparing with the previously reported monitor systems, which can only detect single parameter, the present MIFMI system proposes an improved multiparameter monitor and analysis ability. In general, the main mechanism cancer treatment using PDT is the reaction between oxygen and photosensitizers under the light irradiation, which generates cytotoxic Reactive Oxygen Species (ROS) that directly destroys the tumor cells.24 It has been demonstrated that PDT has two ways to kill cells, defined as Type I and Type II mechanisms.12 The Type I mechanism involves hydrogen-atom abstraction or electron-transfer reactions between the photosensitizer and a substrate to yield free radicals and radical ions, which can interact with molecular oxygen to either generate ROS (such as superoxide anions and hydroxyl radicals) or can cause irreparable biological damage. The Type II mechanism results from an energy transfer between the photosensitizers and the molecular oxygen to produce singlet oxygen. In the mechanisms, the molecular oxygen and singlet oxygen have similar relationship. Due to a short lifetime (0.04 ls) of singlet oxygen, a small radiation area (0.02 lm), and the weak intensity of emission spectroscopy (1270 nm), the direct detection of singlet oxygen is very difficult. In this paper, we utilized Luminol as fluorescence probe to detect singlet oxygen indirectly.9,25 We employed the advantages of the chemical reaction between Luminol and singlet oxygen, which produced a delayed fluorescence signal. Comparing to the visible light, PDT triggered by NIR light enables the feasibility of deep-tissue PDT treatment. In carbon quantum dots concentration gradient experiment, we aimed to investigate a valid real-time monitor system during PDT for deep-seated tumors treatment. The amplitude of the voltage is obviously distinguished due to different concentration. Even if the concentration of solution is very low (0.08 lM/L), we could also obtain the signal at 0.18 V. Then, we acquired the spectral curves of several PDT solutions (Table 1) under 660 nm laser irradiation, which could guide us to optimize the wavelength (from 420 to 540 nm). Furthermore, we conducted oxygen partial pressure change curves during PDT, the results show that the oxygen partial pressure in Luminol and ZnPc solutions are 26.3 and 17.7%, respectively. When putting the ZnPc into Luminol, the maximum oxygen partial pressure increase is 54%, and the minimum increase is 30.9%.8,13,19 In order to evaluate the deep PDT treatment using FASOC-UCNP-ZnPc and MIFMI system, we
LI et al.
explored another experiment. Comparing with the ZnPc, the effective depth of FASOC-UCNP-ZnPc can be up to 10 mm, which is consistent with the result in former study. Here, we conducted the multi-parameter synchronous detection experiment using 980 nm laser and FASOC-UCNP-ZnPc under 10 mm chicken tissue. Finally, we derived the therapy evaluation model among the oxygen partial pressure, the concentration of singlet oxygen and the tumor cell viability. Moreover, the concentration of singlet oxygen, oxygen partial pressure, as well as the CCD images were displayed in the computer. These multiple parameters are more accurate to remind doctor of the current expectation of the survival rate of each dose of light irradiation. For example, the tumor cell viability curve will be calculated in real time using the relation equation (Eq. 1), which is the most important factor to evaluate the effects of PDT. In the following studies, the MIFMI system can be explored in the PDT animal experiments in vivo. In summary, we have successfully developed and verified a MIFMI system, which can measure oxygen partial pressure, singlet oxygen and capture fluorescence images simultaneously in vitro. Furthermore, the MIFMI system has been used to explore the relationship among the oxygen partial pressure, the concentration of singlet oxygen and the tumor cell viability in deep PDT experiments. It all boils down to relation equation and the specific conditions have been presented. Our studies demonstrate that the MIFMI system has the potential for the evaluation of deep PDT using multiple parameters.
ACKNOWLEDGMENTS This work is supported by ‘‘the Fundamental Research Funds for the Central Universities’’, NO. NS2015032. We thanked Dr. Xinzeng Wang for close reading of the manuscript and made corresponding revisions in the whole paper.
REFERENCES 1
Bown, S. G. Photodynamic therapy for cancers of solid organs. Lasers Electro Opt. Soc. 1:253–254, 2002. 2 Celli, J. P., B. Q. Spring, I. Rizvi, C. L. Evans, K. S. Samkoe, S. Verma, B. W. Pogue, and T. Hasan. Imaging and photodynamic therapy: mechanisms, monitoring, and optimization. Chem. Rev. 110:2795–2838, 2010. 3 Chen, T. C., L. Huang, C. C. Liu, P. J. Chao, and F. H. Lin. Luminol as the light source for in situ photodynamic therapy. Process Biochem. 47:1903–1908, 2012.
4
Crimaldi, J. P. Planar laser induced fluorescence in aqueous flows. Exp. Fluids. 44:851–863, 2008. 5 Cui, S., H. Chen, H. Zhu, J. Tian, X. chi, Z. Qian, S. Achilefu, and Y. Gu. Amphiphilic chitosan modified upconversion nanoparticles for in vivo photodynamic therapy induced by near-infrared light. J. Mater. Chem. 22:4861–4873, 2012. 6 Cui, S., D. Yin, Y. Chen, Y. Di, H. Chen, Y. Ma, S. Achilefu, and Y. Gu. In vivo targeted deep-tissue photodynamic therapy based on near-infrared light triggered upconversion nanoconstruct. ACS Nano. 7:676–688, 2013. 7 Garcez, A. S., E. R. Freqnani, H. M. Rodriquez, S. C. Nunez, C. P. Sabino, H. Suzuki, and M. S. Ribeiro. The use of optical fiber in endodontic photodynamic therapy. Is it really relevant? Lasers Med. Sci. 28:79–85, 2013. 8 Gehmert, S., S. Geis, P. E. Lamby, C. Roll, U. Braumandl, M. Hidayat, M. Sultan, B. Fuchtmeier, E. M. Jung, and L. Prantl. Evaluation of hyperbaric oxygen therapy for free flaps using planar optical oxygen sensors. Preliminary results. Clin. Hemorheol. Microcirc. 48:75–79, 2011. 9 Hackbarth, S., J. Schlothauer, A. Preuss, and B. Roder. New insights to primary photodynamic effects-singlet oxygen kinetics in living cells. J. Photochem. Photobiol. B 98:173–179, 2010. 10 Hewett, J., V. Nadeau, J. Ferquson, H. Moseley, S. Ibbotson, J. W. Allen, W. Sibbett, and M. Padgett. The application of a compact multispectral imaging system with integrated excitation source to in vivo monitoring of fluorescence during topical photodynamic therapy of superficial skin cancers. Photochem. Photobiol. 73:278– 282, 2001. 11 Jerjes, W., T. Upile, C. Alexander Mosse, Z. Hamdoon, M. Morcos, S. Morley, and C. Hopper. Prospective evaluation of 110 patients following ultrasound-guided photodynamic therapy for deep seated pathologies. Photodiagnosis Photodyn Ther. 8:297–306, 2011. 12 Jessica, T., C. Sandra, and C. Alison. Validation of a non-invasive fluorescence imaging system to monitor dermatological PDT. Photodiagnosis Photodyn. Ther. 7:86–97, 2010. 13 Jonas, R., T. Schaal, M. Krimmel, D. Gu¨licher, S. Reinert, and J. Hoffmann. Monitoring in microvascular tissue transfer by measurement of oxygen partial pressure: four years experience with 125 microsurgical transplants. J. Craniomaxillofac Surg. 41:303–309, 2013. 14 Jung, J., A. J. Nichols, O. J. Klein, E. Roussakis, and C. L. Evans. Label-free, longitudinal visualization of PDT response in vitro with optical coherence tomography. Isr. J. Chem. 52:728–744, 2012. 15 Kasˇ cˇa´kova´, S., A. Giuliani, S. Lacerda, A. Pallier, P. Mercere, E. Toth, and M. Refregiers. X-ray induced radiophotodynamic therapy (RPDT) using lanthanide micelles: beyond depth limitations. Nano Res. 8:2373–2379, 2015. 16 Niedre, M. J., A. J. Secord, M. S. Patterson, and B. C. Wilson. In vitro tests of the validity of singlet oxygen luminescence measurements as a dose metric in photodynamic therapy. Cancer Res. 63:7086–7094, 2003. 17 Pacheco, L., S. S. Stolik, J. De la Rosa, and A. De la Cadena. Laser diode system for interstitial photodynamic therapy. IEEE Lat. Am. Trans. 12:574–579, 2014. 18 Papkovsky, D. B., and R. I. Dmitriev. Biological detection by optical oxygen sensing. Chem. Soc. Rev. 42:8700–8732, 2013.
Real-Time Monitoring of Singlet Oxygen and Oxygen Partial Pressure 19
Piffaretti, F., A. M. Novello, R. S. Kumar, E. Forte, C. Paulou, P. Nowak-Sliwinska, H. van den Bergh, and G. Wagnie`res. Real-time, in vivo measurement of tissular pO2 through the delayed fluorescence of endogenous protoporphyrin IX during photodynamic therapy. J. Biomed. Opt. 17:115007, 2012. 20 Sabino, C. P., A. S. Garcez, S. C. Nu´n˜ez, M. S. Ribeiro, and M. R. Hamblin. Real-time evaluation of two light delivery systems for photodynamic disinfection of Candida albicans biofilm in curved root canals. Lasers Med. Sci. 30:1657–1665, 2015. 21 Sargent, B. J., and D. A. Gough. Design and validation of the transparent oxygen sensor array. IEEE Trans. Biomed. Eng. 38:476–482, 1991.
22
Sunar, U. Monitoring photodynamic therapy of head and neck malignancies with optical spectroscopies. World J. Clin. Cases. 1:96–105, 2013. 23 Tokatli-Apollon, A., H. Lee, M. El-Rifai, and A. Punjabi. Upconversion nanoparticles for photodynamic therapy. Northeast Bioeng. Conf. 40:1–2, 2014. 24 Trushina, A. P., V. G. Goldort, S. A. Kochubei, and A. V. Baklanov. UV-photoexcitation of encounter complexes of oxygen O2–O2 as a source of singlet oxygen O2(1Dg) in gas phase. Chem. Phys. Lett. 485:11–15, 2010. 25 Wang, J., D. Xing, Y. He, and X. Hu. Experimental study on photodynamic diagnosis of cancer mediated by chemiluminescence probe. FEBS Lett. 523:128–132, 2002.
