Effect of oxygen concentration on singlet oxygen
luminescence detection
Longchao Chen a,Lisheng Lin a,Yirong Li a,Huiyun Lin a,Zhihai Qiu a,
Ying Gu b,Buhong Li a,n
a MOE Key Laboratory of OptoElectronic Science and Technology for Medicine,Fujian Provincial Key Laboratory for Photonics Technology,
Fujian Normal University,Fuzhou350007,PR China
b Department of Laser Medicine,Chinese PLA General Hospital,Beijing100853,PR China
a r t i c l e i n f o
Available online20October2013
Keywords:
Singlet oxygen
Fluorescence
Phosphorescence
Luminescence
Oxygen concentration
a b s t r a c t
Singlet oxygen(1O2)is a major phototoxic component in photodynamic therapy(PDT)and its generation
is dependent on the availability of tissue oxygen.To examine the effect of oxygen concentration on1O2
detection,two hydrophilic photosensitizer(PS),rose bengal(RB)and meso-metra(N-methyl-4-pyridyl)
porphine tetra tosylate(TMPyP)were used as model PS.Irradiation was carried out using523nm under
hypoxic(2%,13%),normoxic(21%)and hyperoxic(65%)conditions.The spectral and spatial resolved1O2
luminescence was measured by near-infrared(NIR)photomultiplier tube(PMT)and camera,respectively.
Upon the irradiation,the emission signal mainly consisted of background scattering light,PSfluorescence
and phosphorescence,and1O2luminescence.The PS phosphorescence was evidently dependent on the
oxygen concentration and PS type,which resulted in the change of emission profile of1O2luminescence.
This change was further demonstrated on1O2luminescence image.The present study suggests that the
low oxygen concentration could affect1O2luminescence detection.
&2013Elsevier B.V.All rights reserved.
1.Introduction
Singlet oxygen(1O2)is a highly reactive oxygen species(ROS)
that plays a key role in a number of chemical and photochemical
reactions in different biological systems and particularly in photo-
dynamic therapy(PDT)[1–3].This has driven intense interests to
develop novel sensitive techniques for direct1O2detection,eluci-
date the kinetic mechanisms of1O2generation,and establish
reliable protocols for the quantification of1O2generation[4–6].
Photosensitizer(PS),light and oxygen are three major compo-
nents in PDT.Many studies have been previously performed
to investigate the influence of PS type[7–13],oxygen concentration
[14–19]and light dose[2,13]on1O2generation during photosensi-
tization.Among these,the time-resolved detection of near-infrared
(NIR)luminescence at around1270nm is the most common method
for the study of kinetic mechanisms of1O2generation[3,5].In
particular,the effects of oxygen concentration on the kinetic
mechanisms of1O2generation,such as the lifetimes of PS phosphor-
escence,triplet-state and1O2luminescence have been investigated
under different oxygen concentrations[14,15,18].The detection of
1O
2
luminescence can also be used as a sensor for monitoring oxygen
concentration.However,there is limited knowledge about the effect
of oxygen concentration on the emission intensity and gray value of
1O
2
luminescence image,which is important for quantifying1O2
generation.
In this study,two hydrophilic PS,rose bengal(RB)and meso-
tetra(N-methyl-4-pyridyl)porphine tetra tosylate(TMPyP)were
used as model PS.RB and TMPyP mediated1O2luminescence were
detected under hypoxic(2%,13%),normoxic(21%)and hyperoxic
(65%)conditions using spectral resolved1O2luminescence and NIR
imaging system.To the best of our knowledge,this is thefirst
study that dedicates to study the effect of oxygen concentration on
1O
2
generation during photosensitization by using spectral and
spatial resolved1O2luminescence measurement.
2.Materials and methods
2.1.Chemicals
Stock solutions(100μM)of RB(Sigma-Aldrich,St.Louis MO,
USA)and TMPyP(Frontier Scientific,Logan UT,USA)were pre-
pared in air-saturated phosphate buffered saline(PBS,pH7.5)and
stored at41C in the dark.Aqueous working solutions were
prepared from the stock solutions before use.The RB and TMPyP
concentrations in the working solution were controlled at the
absorbance of0.2at the corresponding excitation wavelength
(523nm).The absorbance was measured with a UV/Vis/NIR
spectrophotometer(Lambda950,Perkin Elmer,Waltham MA,
Contents lists available at ScienceDirect
journal homepage:www.elsevier/locate/jlumin
Journal of Luminescence
0022-2313/$-see front matter&2013Elsevier B.V.All rights reserved.
/10.1016/j.jlumin.2013.10.034
n Corresponding author.Tel.:þ8659188037959;fax:þ8659183465373.
E-mail address:bhli@fjnu.edu(B.Li).
Journal of Luminescence152(2014)98–102
USA).The concentrations of RB and TMPyP in the working solution were5.0and19.0μM,respectively.
2.2.Control of dissolved oxygen
Hypoxic(2%or13%),normoxic(21%)and hyperoxic(65%) solutions were prepared by bubbling the air-saturated samples with pure nitrogen or oxygen,respectively.The concentration of dissolved oxygen in solutions was determined by an oxygen sensor (NeoFox,Ocean Optics,Dunedin FL).The sensor uses ruthenium (II)complexes suspended in a support matrix and attached to the tip of thefiber optic cable.When excited at475nm,the ruthenium complex generates afluorescence emission at620nm.The sensor assessed thefluorescence lifetime by modulation of light excita-tion frequency where the oxygen dependent shift of the emission is detected[19].
2.3.Measurement offluorescence,phosphorescence and1O2 luminescence
The system developed for the measurement of PSfluorescence, phosphorescence and1O2luminescence is shown in Fig.1.Laser of 523nm from a diode-pumped,Q-switched and frequen
cy-doubled Nd:YLF laser(QG-523-500;Crystalaser Inc.,Reno NV,USA)was used as excitation light source.The excitation pulses of15μJ per pulse and12kHz repetition were used for all experiments.The solutions were placed in a standard quartz cuvette mounted on a hotplate-stirrer unit(Variomag mini;HþP Labortechnik GmbH, Oberschleissheim,Germany).The source–sample–detector geo-metry was maintained at the same position during measurements. To minimize oxygen diffusion from the atmosphere,the cuvette was tightly sealed with aflexible PVC plastic tubing.
Forfluorescence measurement,a550nm long-passfilter(Omega Optical,Brattleboro VT,USA)was used to block out unwanted scatt-ering excitation light.A photomultiplier tube(PMT,R928P;Hama-matsu Corp.,Hamamatsu,Japan)with supply voltage ofÀ1300V and PCS900PC card(Edinburgh Instruments Ltd.,Livingston,UK)were used to record thefluorescence spectra.In order to monitor the PS photobleaching as a function of irradiation time,an opticalfiber connected to a miniature spectrometer(USB4000,Ocean Optics) through a550nm long passfilter(Omega Optical)was used to record PSfluorescence spectroscopy.The measured sample will be replaced with a fresh sample once the PSfluorescence decreased to95%of its initial value after photobleaching.
For NIR signal measurement,a900nm long-passfilter(Omega Optical)was used to block out excitation light andfluorescence from the sample.After excitation,the1O2luminescence reached the PMT(R5509;
Hamamatsu Corp.)photocathode through a monochromator(TMS300,Bentham Instruments Ltd.,UK).The operating voltage of the PMT was set toÀ1650V.The PMT output was amplified,converted to a voltage pulse using a high-speed wide band pre-amplifier(HFAC-26;Becker&Hickl GmbH,Berlin, Germany),and recorded by a fast multiscaler(MSA-300;Becker& Hickl GmbH).The1O2luminescence was recorded from1220to 1340nm at10nm intervals.The emission slits forfluorescence and NIR signal measurement were set at0.10and20.0nm, respectively.
2.4.1O2luminescence imaging
An NIR-camera based imaging system was developed for1O2 luminescence detection in which the spectral imaging discrimination of the NIR luminescence was achieved using three narrow-band filters centered at1215,1270and1315nm(OD5,Omega Optical), respectively.Thefilter was mounted on a slider in front of the cooled NIR-camera.Thesefilters were used to spectrally isolate the1O2 luminescence near1270nm from the long wavelength spectral background signal,such as PSfluorescence,phosphorescence and ambient background,as previously reported by Lee et al.[20].The images recorded at1215and1315nm were acquired and averaged to generate a single spectral image of background signal.This formed a first order estimation of the signal level of the background signal that contributed to the1270nm image.This averaged background image was subseq
uently subtracted on a pixel by pixel basis from the image obtained with the1270nmfilter.The1O2luminescence image with a resolution of30μm and with a maximal area of9.60Â7.68mm2 could be acquired using this system.
2.5.Data analysis
Each experiment was repeated in triplicate at room tempera-ture and all values were presented as mean7standard deviation (S.D.).The data were processed and graphed using Origin8.0soft-ware(Origin Lab Corporation,Northampton MA,USA).
3.Results and discussion
3.1.Effect of oxygen concentration onfluorescence and phosphorescence spectra
The typicalfluorescence and phosphorescence spectra of RB and TMPyP under the same PS concentration but different oxygen concentrations are shown in Fig.2.Under the excitation of 523nm,thefluorescence emission peaks were observed at570 and625nm for RB and at625,725and935nm for TMPyP.For both RB and TMPyP,the emission peaks of phosphorescence were not evident under normoxic and hyperoxic conditions.However, when the oxygen concentration w
as reduced to2%,the phosphor-escence peaks at730and1055nm were observed for RB and TMPyP,respectively.Thesefindings agree well with those observed by Zhang et al.and Dědic et al.[21,22].As shown in Fig.2,the oxygen concentration had a significant influence on the emission offluorescence and phosphorescence,which resulted in the change of emission profile of1O2luminescence.In particular, the intensity of phosphorescence increased with a decrease of dissolved oxygen
concentration.
Fig.1.Schematic diagram of the detection system.
L.Chen et al./Journal of Luminescence152(2014)98–10299
3.2.Effect of oxygen concentration on time and spectral resolved 1O 2luminescence spectra
The representative time resolved 1
O 2luminescence spectra of RB and TMPyP under different oxygen concentrations are shown in Fig.3a and b,respectively.The insert of Fig.3b indicates the 1O 2luminescence spectra of TMPyP after removing the initial spike.The time-integrated spectra were generated by summing the luminescence for each individual wavelength from 0to 70μs,respectively.As expected,a strong spectral peak was observed at 1270nm as shown in Fig.3c and d.When the oxygen concentra-tion reduced to 2%,the 1O 2luminescence at 1270nm decreased while the background signal outside the 1O 2emission band markedly increased.
3.3.Effect of oxygen concentration on triplet-state and 1O 2lifetime Based on the previously established method [4],the triplet state and 1O 2lifetimes of RB and TMPyP were derived through fittin
g the 1
O 2luminescence spectra.As shown in Table 1,the triplet-state lifetimes correlated well with the corresponding phosphorescence lifetimes,which were independently determined by fitting the time-resolved phosphorescence spectra at 730nm for RB and 1055nm for TMPyP,respectively.For TMPyP,the phosphorescence lifetime of 105.50713.40μs obtained under the hypoxic concentration (2%)is comparable to the reported value of 102.471.3μs [22].The triplet-state and 1O 2lifetimes of 2.3470.03and 3.1470.07μs for normoxic condition (21%)are in good agreement with the reported values of 2.0and 3.5μs [23],respectively.Most recently,Scholz et al.
clearly
Fig.2.Fluorescence and phosphorescence spectra of RB (a)and TMPyP (b)under different oxygen
concentrations.
Fig.3.Time resolved 1O 2luminescence of RB (a)and TMPyP (b),and spectral resolved 1O 2luminescence of RB (c)and TMPyP (d)in PBS under different oxygen concentrations.
L.Chen et al./Journal of Luminescence 152(2014)98–102
100
showed that the decay fluorescence of some kinds of photosensiti-zers may be extended to the NIR region,which is crucial for accurately determining the real phosphorescence [24].In this regard,the decay fluorescence should be carefully considered for the phosphorescence measurement of photosensitizers.3.4.
1
O 2luminescence image
The spatially resolved images of RB and TMPyP (Fig.4)showed that the average gray values (AGV)of images recorded from the 1215,1270and 1315nm filter for TMPyP were higher than that of RB.The ima
ge obtained with the 1215nm filter was gradually enhanced with the decrease of oxygen concentration,which was strongly in fluenced by the tail of NIR phosphorescence.These observations were consistent with the spectral-resolved 1O 2lumi-nescence seen in Fig.3c and d.In order to obtain the real image of
the 1O 2luminescence,the pixel by pixel averaged values of the 1215and 1315nm images were subtracted from the image recorded at 1270nm.In contrast,no signi ficant 1O 2signals were observed from the control samples without PS (data not shown)under four oxygen concentrations,which suggested that the background signal could be ignored.
3.5.Effect of oxygen concentration on 1O 2detection
As shown in Fig.3c and d,the luminescence intensities recorded at 1220and 1320nm,which lie outside the 1O 2emission band,were used to determine the luminescence background in order to obtain the real 1O 2luminescence intensity.As illustrated in Fig.5,the real 1O 2luminescence intensities were consistent with the AGV of real 1O 2luminescence images for all four oxygen concentrations.No signi ficant difference was found for the 1O 2
Table 1
The triplet-state (τ1270T ),1O 2(τ1270D )and phosphorescence (τ730T /τ1055
T )lifetimes of RB and TMPyP in PBS.Oxygen concentration (%)
RB TMPyP τ1270D
(μs)τ1270T
(μs)τ730T
(μs)τ1270D
(μs)τ1270T
(μs)τ1055T
(μs)2.00
0.4770.0764.0074.2462.8772.040.5870.05105.8773.96105.50713.4013.00  2.3670.23  6.3170.47  5.0870.17  3.4070.08  5.5670.17  5.2770.3421.00  3.9470.09  2.1770.05  2.5570.02  3.1470.07  2.3470.03  2.0570.3165.10
3.5070.06
1.0070.03
0.9370.08
3.4370.01
0.9570.04
1.117
0.42
Fig.4.1O 2luminescence images of RB (a)and TMPyP (b)under different oxygen
concentrations.
Fig.5.Effect of oxygen concentration on the intensity of 1O 2luminescence and AGV of 1O 2luminescence image of RB (a)and TMPyP (b),respectively.
L.Chen et al./Journal of Luminescence 152(2014)98–102101
generation between the oxygen concentrations greater than13% but a significant difference was found between2%and13%.These findings suggest that the effect of oxygen concentration on1O2 luminescence detection is particularly important when the oxygen concentration is extremely low,which could happen in hypoxic tumor and during PDT[25].Therefore,further in vivo studies should be performed in order to quantify the oxygen effect on1O2 generation under hypoxic conditions.
4.Conclusions
Both spectral and spatial resolved1O2luminescence measure-ments suggest that the oxygen concentration might affect1O2 luminescence detection,particularly under hypoxic condition.The monitoring of PSfluorescence and phosphorescence is crucial to accurately analyze1O2luminescence kinetics and quantify the1O2 generation during photosensitization.In order to quantify the1O2 generation,background signal from the referencefilters should be appropriately subtracted from the signal recorded at1270nm. Acknowledgments
The authors wish to thank Dr.Brian C.Wilson at University of Toronto for his insightful discussion.This work was supported by the National Natural Science Foundation of China(60978070, 61036014,and61275216),the program for New Century Excellent Talents in University of China(NCET-
10-0012),the Fujian Provin-cial Natural Science Foundation(2011J06022)and the Program for Changjiang Scholars and Innovative Research Team in University (IRT1115).
References
[1]А.Krasnovsky Jr.,J.Photochem.Photobiol.A196(2008)210.
[2]M.T.Jarvi,M.S.Patterson,B.C.Wilson,Biophys.J.102(2012)661.
[3]B.Li,H.Lin,D.Chen,B.C.Wilson,Y.Gu,J.Innov.Opt.Health Sci.6(2013)
1330002.
[4]H.Lin,D.Chen,M.Wang,J.Lin,B.Li,S.Xie,J.Opt.13(2011)125301.
[5]N.R.Gemmell,A.McCarthy,B.Liu,M.G.Tanner,S.N.Dorenbos,V.Zwiller,
M.S.Patterson,G.S.Buller,B.C.Wilson,R.H.Hadfield,Opt.Express21(2013) 5005.
[6]H.Lin,Y.Shen,D.Chen,L.Lin,B.C.Wilson,B.Li,S.Xie,J.Fluoresc.23(2013)41.
[7]P.Bilski,R.Dabestani,C.Chignell,J.Phys.Chem.B95(1991)5784.
[8]J.Baier,T.Fuss,  C.Pöllmann,  C.Wiesmann,K.Pindl,R.Engl,  D.Baumer,
M.Maier,M.Landthaler,W.Bäumler,J.Photochem.Photobiol.B87(2007) 163.
reactive materials studies[9]J.W.Snyder,J.D.Lambert,P.R.Ogilby,Photochem.Photobiol.82(2006)177.
[10]R.Dědic,V.Vyklický,A.Svoboda,J.Hála,J.Lumin.131(2011)442.
[11]J.C.Schlothauer,S.Hackbarth,L.Jäger,K.Drobniewski,H.Patel,S.M.Gorun,
B.Röder,J.Biomed.Opt.17(2012)115005.
[12]A.Felgenträger,F.P.Gonzales,T.Maisch,W.Bäumler,J.Biomed.Opt.18(2013)
045002.
[13]E.F.da Silva,  B.W.Pedersen,T.Breitenbach,R.Toftegaard,M.K.Kuimova,
L.G.Arnaut,P.R.Ogilby,J.Phys.Chem.B116(2011)445.
[14]T.Maisch,J.Baier,  B.Franz,M.Maier,M.Landthaler,R.-M.Szeimies,
W.Bäumler,Proc.Natl.Acad.Sci.USA104(2007)7223.
[15]J.Baier,T.Maisch,J.Regensburger,M.Loibl,R.Vasold,W.Bäumler,J.Biomed.
Opt.12(2007)064008.
[16]S.Hatz,L Poulsen,P.R.Ogilby,Photochem.Photobiol.84(2008)1284.
[17]M.Scholz,R.Dědic,J.Hála,S.Nonell,J.Mol.Struct.1044(2013)303.
[18]A.Gollmer,J.Regensburger,T.Maisch,W.Bäumler,Phys.Chem.Chem.Phys.
15(2013)11386.
[19]S.Boehme,  B.Duenges,K.U.Klein,V.Hartwich,  B.Mayr,J.Consiglio,
J.E.Baumgardner,K.Markstaller,R.Basciani,  A.Vogt,PLoS One8(2013) e60591.
[20]S.Lee,M.E.Isabelle,K.L.Gabally-Kinney,B.W.Pogue,S.J.Davis,Biomed.Opt.
Express2(2011)1233.
[21]Y.Zhang,K.Aslan,M.J.R.Previte,S.N.Malyn,C.D.Geddes,J.Phys.Chem.B110
(2006)25108.
[22]R.Dědic,V.Vyklický,A.Svoboda,J.Hála,J.Mol.Struct.924(2009)153.
[23]X.Ragàs,M.Agut,S.Nonell,Free.Radic.Biol.Med.49(2010)770.
[24]M.Scholz,R.Dědic,T.Breitenbach,J.Hála,Photochem.Photobiol.Sci.12(2013)
1873.
[25]Z.Huang,Q.Chen,A.Shakil,H.Chen,J.Beckers,H.Shapiro,F.W.Hetzel,
Photochem.Photobiol.78(2003)496.
L.Chen et al./Journal of Luminescence152(2014)98–102 102

版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。