A two-photon antenna for photochemical delivery of nitric oxide from a water-soluble, dye-derivatized iron nitrosyl complex using NIR light

Article abstract of DOI:10.1021/ja057977u

The experiments described here demonstrate the use of two-photon excitation (TPE) to sensitize nitric oxide (NO) release from a dye-derivatized iron/sulfur/nitrosyl cluster Fe₂(μ-RS)₂(NO)₄ (Fluor-RSE, RS = 2-thioethyl ester of fluorescein) with near-infrared (NIR) light in the form of femtosecond pulses from a Ti:sapphire laser. TPE at 800 nm leads both to weak fluorescence from the organic chromophore at λ_max = 532 nm and to NO labilization from the cluster. Since the emission from the reference compound Fluor-Et (the ethyl ester of fluorescein) under identical conditions (50/50 CH₃CN/phosphate buffer (1 mM) at pH 7.4) is considerably more intense, the weaker emission from Fluor-RSE and the NO generation indicate that the fluorescein excited states initially formed by TPE are largely quenched by energy transfer to the cluster core. The two-photon absorption (TPA) cross section of Fluor-RSE at 800 nm was determined to be δ = 63 ± 7 GM via the TPA photoluminescence technique. This can be compared to the TPA cross section of 36 GM reported for fluorescein dye in pH 11 aqueous solution and of 32 ± 3 GM for Fluor-Et measured under conditions comparable to those used for Fluor-RSE. Pulse intensity dependence studies showed that the quantity of NO released from the latter as the result of NIR photoexcitation follows a quadratic relationship to excitation intensity, consistent with the expectation for a TPE process. These studies demonstrate the potential utility of a two-photon antenna for sensitization of the photochemical release of an active agent (in this case, NO) from a photoactive pro-drug.

Full text of DOI:10.1021/ja057977u

Published on Web 02/25/2006
A Two-Photon Antenna for Photochemical Delivery of Nitric
Oxide from a Water-Soluble, Dye-Derivatized Iron Nitrosyl
Complex Using NIR Light
Stephen R. Wecksler, Alexander Mikhailovsky, Dmitry Korystov, and Peter C. Ford*
Contribution from the Department of Chemistry and Biochemistry, UniVersity of California,
Santa Barbara, Santa Barbara, California 93106-9510
Received December 8, 2005; E-mail: ford@chem.ucsb.edu
Abstract: The experiments described here demonstrate the use of two-photon excitation (TPE) to sensitize
nitric oxide (NO) release from a dye-derivatized iron/sulfur/nitrosyl cluster Fe2(µ-RS)2(NO)4 (Fluor-RSE,
RS ) 2-thioethyl ester of fluorescein) with near-infrared (NIR) light in the form of femtosecond pulses from
a Ti:sapphire laser. TPE at 800 nm leads both to weak fluorescence from the organic chromophore at λmax
) 532 nm and to NO labilization from the cluster. Since the emission from the reference compound Fluor-
Et (the ethyl ester of fluorescein) under identical conditions (50/50 CH3CN/phosphate buffer (1 mM) at pH
7.4) is considerably more intense, the weaker emission from Fluor-RSE and the NO generation indicate
that the fluorescein excited states initially formed by TPE are largely quenched by energy transfer to the
cluster core. The two-photon absorption (TPA) cross section of Fluor-RSE at 800 nm was determined to
be δ ) 63 ( 7 GM via the TPA photoluminescence technique. This can be compared to the TPA cross
section of 36 GM reported for fluorescein dye in pH 11 aqueous solution and of 32 ( 3 GM for Fluor-Et
measured under conditions comparable to those used for Fluor-RSE. Pulse intensity dependence studies
showed that the quantity of NO released from the latter as the result of NIR photoexcitation follows a
quadratic relationship to excitation intensity, consistent with the expectation for a TPE process. These
studies demonstrate the potential utility of a two-photon antenna for sensitization of the photochemical
release of an active agent (in this case, NO) from a photoactive pro-drug.
Introduction
thermally stable NO precursors that can be photochemically
triggered upon demand to release nitric oxide.3 In this regard,
The biological messenger nitric oxide (nitrogen monoxide)
is a key player in numerous mammalian functions, including
vasodilation, immune response, and neurotransmission.1 NO has
also been demonstrated to be a γ-radiation sensitizer, and this
property may prove useful in enhancing the therapeutic effects
of radiation treatment of hypoxic tumors.2 As a result, there is
an interest in developing materials for the controlled delivery
of NO to specific biological targets. The key words are
“controlled” and “specific” since systemic release could have
serious physiological effects, most importantly, markedly
reduced blood pressure inducing shock. One strategy to
circumvent such an undesired side effect would be to develop
studies in this laboratory have been concerned with character-
izing the quantitative photochemistry of transition metal com-
pounds that may serve this role, such as metal nitrosyl and nitrito
complexes4,6,7 and the iron/sulfur/nitrosyl clusters known as the
Roussin’s salts and esters.5,8
A desirable characteristic of photochemical drugs would be
activation by excitation wavelengths effective for tissue trans-
(3) (a) Ford, P. C.; Bourassa, J.; Miranda, K. M.; Lee, B.; Lorkovic, I.; Boggs,
S.; Kudo S.; Laverman L. Coord. Chem. ReV. 1998, 171, 185-202. (b)
Pavlos, C. M.; Xu, H.; Toscano, J. P. Curr. Top. Med. Chem. 2005, 5,
635-645.
(4) (a) Hoshino, M.; Ozawa, K.; Seki, H.; Ford, P. C. J. Am. Chem. Soc. 1993,
115, 9568-9575. (b) Lorkovic, I. M.; Miranda, K. M.; Lee, B.; Bernhard,
S.; Schoonover, J. R.; Ford, P. C. J. Am. Chem. Soc. 1998, 120, 11674-
11683. (c) Laverman, L.; Ford, P. C. J. Am. Chem. Soc. 2001, 123, 11614-
11622. (d) Joseph, C.; Ford, P. C. J. Am. Chem. Soc. 2005, 127, 6737-
6743.
(5) (a) Bourassa, J.; DeGraff, W.; Kudo, S.; Wink, D. A.; Mitchell, J. B.; Ford,
P. C. J. Am. Chem. Soc. 1997, 119, 2853-2860. (b) Bourassa, J.; Lee, B.;
Bernard, S.; Schoonover, J.; Ford, P. C. Inorg. Chem. 1999, 38, 2947-
2952. (c) Bourassa, J.; Ford, P. C. Coord. Chem. ReV. 2000, 200-202,
887-900.
(6) (a) De Leo, M.; Ford, P. C. J. Am. Chem. Soc. 1999, 121, 1980-1981. (b)
DeLeo, M.; Ford, P. C. Coord. Chem. ReV. 2000, 208, 47-59. (c) DeRosa,
F.; Bu, X.; Ford, P. C. Inorg. Chem. 2005, 44, 4157-4165. (d) DeRosa,
F.; Bu, X.; Pohaku, K.; Ford, P. C. Inorg. Chem. 2005, 44, 4166-4174.
(7) (a) Works, C. F.; Ford, P. C. J. Am. Chem. Soc. 2000, 122, 7592-7593.
(b) Works, C. F.; Jocher, C. J.; Bart, G. D.; Bu, X.; Ford, P. C. Inorg.
Chem. 2002, 41, 3728-3739.
(8) (a) Conrado, C.; Bourassa, J.; Egler, C.; Wecklser, S.; Ford, P. C. Inorg.
Chem. 2003, 42, 2288-2293. (b) Conrado, C. L.; Wecksler, S.; Egler, C.;
Magde, D.; Ford, P. C. Inorg. Chem. 2004, 43, 5543-5549.
(1) (a) Ignarro, L. J.; Buga, G. M.; Wood, K. S.; Byrns, R. E.; Chaudhuri, G.
Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 9265-9269. (b) Palmer, R. M. J.;
Ferrige, A. G.; Moncada, S. Nature 1987, 327, 524-526. (c) Palmer, R.
M. J.; Ashton, D. S.; Moncada, S. Nature 1988, 333, 664-666. (d)
Garthwaite, J.; Charles, S. L.; Chess-Williams, R. Nature 1988, 336, 385-
388. (e) Garthwaite, J.; Garthwaite, G.; Palmer, R. M. J.; Moncada, S. Eur.
J. Pharmacol. 1989, 172, 413-416. (f) Azuma, H.; Ishikawa, M.; Sekizaki,
S. Br. J. Pharmacol. 1986, 88, 411-415. (g) Radomski, M. W.; Palmer,
R. M. J.; Moncada, S. Br. J. Pharmacol. 1987, 92, 181-187. (h) Steuhr,
D. J.; Marletta, M. A. J. Immunol. 1987, 139, 518-525. (i) Iyengar, R.;
Stuehr, D. J.; Marletta, M. A. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 6369-
6373. (j) Nitric Oxide: Biology and Pathobiology; Ignarro, L. J., Ed.;
Academic Press: San Diego, CA, 2000.
(2) (a) Howard-Flanders, P. Nature 1957, 180, 1191-1192. (b) Powers, W.
E.; Tolmach, L. J. Nature 1963, 197, 710-711. (c) Mitchell, J. B.; Wink,
D. A.; DeGraff, W.; Gamson, J.; Keefer, L. K.; Krishna, M. C. Cancer
Res. 1993, 53, 5845-5848. (d) Thomlinson, R. H.; Gray, L. H. Br. J.
Cancer 1995, 9, 539-549.
9
10.1021/ja057977u CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 3831-3837
3831

A R T I C L E S
Wecksler et al.
Figure 1. Formulas for Fluor-RSE, Fluor-Et, and PPIX-RSE.
mission. For this reason, our current focus is on designing
compounds with antenna chromophores that have high absorp-
tion cross sections at longer wavelengths and that sensitize NO
photochemical labilization from the linked metal center. We
have demonstrated this for several supramolecular complexes
whereby pendant chromophores led to markedly improved light
gathering efficiency in the near-ultraviolet and visible wave-
length ranges and consequently more effective photochemistry
at the metal center serving as the nitric oxide precursor.6d,8b
However, for mammalian tissue, light penetration is most
favorable at a 800-1100 nm window in the near-infrared
(NIR).9 Since the excited states responsible for the desired NO
release largely lie at higher energy, we have adopted a modified
strategy to address this issue, namely, the use of an antenna
with a sufficiently large cross section for two-photon excitation
(TPE) at NIR wavelengths to serve as the sensitizer of the
desired photochemistry. A second and perhaps equally important
advantage of such multiple photon excitation is the much greater
spatial resolution of TPE techniques, which is already a feature
utilized with imaging via confocal microscopy and related
methods.10,11
investigation of the two-photon photochemistry and photophys-
ics of the supramolecular complex Fluor-RSE, which has two
fluorescein moieties with much larger TPA cross sections (δ )
38 GM at 782 for fluorescein in pH 11 aqueous solution)14,15
as pendant groups on the RRS cluster (Figure 1).
Like PPIX-RSE, Fluor-RSE is a derivative of the iron/sulfur/
nitrosyl compounds known as the Roussin’s salts.16 The
Roussin’s red salt anion (RRS, Fe2(µ-S)2(NO)42-) has been
shown to be photoactive toward NO release with a moderately
high quantum yield.5 Furthermore, cell culture experiments (with
V79 Chinese hamster lung cells) demonstrated that white light
photolysis of RRS-treated cells simultaneous with γ-radiation
exposure led to marked decreases in cell survival relative to
controls also submitted to γ-radiation.5a This effect was at-
tributed to sensitization of the radiation damage by concurrent
photochemical NO release in the cells.2 However, a limitation
(11) (a) For similar reasons, TPE has drawn attention in approaches to
photodynamic therapy (PDT) involving the generation of singlet oxygen
as a toxic agent via photosensitization by various organic dyes,11b-h and
other recent reports include applications with a phosphorescent oxygen
sensor11i for photochemical uncaging of bioactive substances11j-n and
photoacid generation for 3D microfabrication.11o (b) Dougherty, T. J.;
Grindley, G.; Flel, R. J. Natl. Cancer Inst. 1974, 55, 115. (c) Dougherty,
T. J.; Kaufman, J. E.; Goldbarb, A.; Weishaupt, K. R.; Boyle, D.; Mittleman,
A. Cancer Res. 1978, 38, 2628. (d) Bodaness, R. S.; Heller, D. F.; Krasinski,
J.; King, D. S. J. Biol. Chem. 1986, 261, 12098-12101. (e) Fischer, W.
G.; Partridge, W. P., Jr.; Dees, C.; Wachter, E. A. Photochem. Photobiol.
1997, 66, 141-155. (f) Bhawalkar, J. D.; Kumar, N. D.; Zhao, C.-F.; Prasad,
P. N. J. Clin. Lasers Med. Surg. 1997, 15, 201-204. (g) Frederiksen, P.
K.; Jørgensen, M.; Ogilby, P. R. J. Am. Chem. Soc. 2001, 123, 1215-
1221. (h) Dichtel, W. R.; Serin, J. M.; Edder, C.; Fre´chet, J. M. J.;
Matuszewski, M.; Tan, L.-S.; Ohulchanskyy, T. Y.; Prasad, P. N. J. Am.
Chem. Soc. 2004, 126, 5380-5381. (i) Brinas, R. P.; Troxler, T.;
Hochstrasser, R. M.; Vinogradov, S. A. J. Am. Chem. Soc. 2005, 127,
11851-11862. (j) Nikolenko, V.; Yuste, R.; Zayat, L.; Baraldo, L. M.;
Etchenique, R. Chem. Commun. 2005, 1752-1754. (k) Furuta, T.; Wang,
S. S.; Dantzker, J. L.; Dore, T. M.; Bybee, W. J.; Callaway, E. M.; Denk,
W.; Tsien, R. Y. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 1193-1200. (l)
Jaconi, M.; Pyle, J.; Bortolon, R.; Ou, J.; Clapham, D. Curr. Biol. 1997, 7,
599-602. (m) Lipp, P.; Niggli, E. J. Physiol. 1998, 508, 801-809. (n)
Brown, E. B.; Webb, W. W. Methods Enzymol. 1998, 291, 356-380. (o)
Zhou, W.; Kuebler, S. M.; Braun, K. L.; Yu, T.; Cammack, J. K.; Ober, C.
K.; Perry, J. W.; Marder, S. R. Science 2002, 296, 1106-1109.
(12) Wecksler, S.; Mikhailovsky, A.; Ford, P. C. J. Am. Chem. Soc. 2004, 126,
13566-13567.
In this context, we have recently reported12 the qualitative
observation that TPE of the bichromophoric complex PPIX-
RSE (Figure 1) with femtosecond pulses of 810 nm light leads
both to weak emission from the protoporphyrin IX center with
a λmax of 632 nm and to photochemical generation of NO. This
observation was interpreted in terms of the porphyrin chro-
mophore acting as a two-photon antenna to sensitize the
photoreactivity of the iron/sulfur/nitrosyl core. However, the
solubility properties of PPIX-RSE and recognition that por-
phyrins are generally poor two-photon absorption (TPA) chro-
mophores (for PPIX, δ ) 2 GM at 790 nm, where 1 GM )
10-50 cm4 s photon-1 13
) led us to examine antennae with larger
TPA cross sections. Described in the present article is an
(9) (a) Dougherty, T. J.; Marcus, S. L. Eur. J. Cancer 1992, 28A, 1734-1742.
(b) Master, B. R.; So, P. T.; Gratton, E. Biophys. J. 1997, 72, 2405-2412.
2
(10) (a) Since the probability of TPE depends on I0 (I0 ) the intensity of the
(13) Goyan, R. L.; Cramb, D. T. Photochem. Photobiol. 2000, 72, 821-827.
(14) Webb, W. W.; Xu, C. J. Opt. Soc. Am. B 1997, 13, 481-491.
(15) Fischer, A.; Cremer, C.; Stelzer, E. H. K. Appl. Opt. 1995, 34, 1989-
2003.
(16) (a) Roussin, F. Z. Ann. Chim. Phys. 1858, 52, 285. (b) Butler, A. R.;
Glidewell, C.; Li, M.-H. AdV. Inorg. Chem. 1988, 32, 335-393. (c) Butler,
A. R.; Megson, I. L. Chem. ReV. 2002, 102, 1155-1165.
incident light), the spatial selectivity is due to the highest probability of
TPA being at the focal point of the excitation source. (b) Denk, W.;
Strickler, J. H.; Webb, W. W. Science 1990, 248, 73-76. (c) Ko¨nig, K. J.
Microscopy 2000, 200, 83-104. (d) So, P. T. C.; Dong, C. Y.; Masters, B.
R.; Berland, K. M. Annu. ReV. Biomed. Eng. 2000, 2, 399-429. (d) Rubart,
M. Circulation Res. 2004, 95, 1154-1166.
9
3832 J. AM. CHEM. SOC. VOL. 128, NO. 11, 2006

Photochemical Delivery of Nitric Oxide Using NIR Light
A R T I C L E S
50 v/v) used for spectroscopic studies were prepared by slowly diluting
the buffered solutions with distilled acetonitrile.
for Roussin’s red salt in this regard is that the RRS chromophore
does not absorb light appreciably at longer wavelengths. For
this reason, related compounds known as the Roussin’s red salt
esters17 (RSE, Fe2(µ-SR)2(NO)4) were prepared via reactions
such as eq 1, and their photochemistry was investigated.8 The
premise of these studies is that the choice of the R substituent
may allow control of optical properties and biological specificity
of such materials.
All gases were purchased from Praxair Inc. and used as received
unless further specified. Argon used for inert atmosphere preparation
was passed through a Drierite/molecular sieve (4 Å) column to remove
any remaining water. Solutions were deaerated by sparging with N2 or
Ar for approximately 1 min/mL or by three freeze-pump-thaw (f-
p-t) cycles on a vacuum line. Inert atmosphere work was performed
on a Schlenk line or in an argon-filled glovebox (VAC atmospheric
company, Nexus model).
Instruments. UV-vis absorption spectra were measured using a
HP8572 diode array spectrophotometer or Shimadzu dual beam UV-
2401 PC spectrophotometer. All spectra were recorded in quartz UV-
vis cells with 1.00 cm path lengths. Real time detection of nitric oxide
in solution was accomplished using an amiNO-700 nitric oxide electrode
from Innovative Instruments, Inc. The amiNO-700 is capable of
quantitatively detecting nanomolar concentrations of NO in aqueous
solution.
Two-Photon Excitation Fluorescent Measurements. TPE fluo-
rescent measurements were carried out with the UCSB Optical
Characterization Facility. Samples were excited with a tightly collimated
(diameter ∼120 µm) high-intensity laser beam, and the corresponding
up-converted fluorescence was detected at 90°. TPE was achieved using
a mode-locked Ti:sapphire laser (Spectra Physics Tsunami) with
excitation pulses of ∼100 fs and energy of ∼6 nJ operating with a
repetition rate of 80 MHz. The emitted light from the sample was
collected unto a high numerical aperture lens and focused onto the
spectrometer. The radiation dispersed from the spectrometer was then
detected by a thermoelectrically cooled CCD camera (Roper Scientific
Spec 10:100B/TE). Solutions were irradiated for approximately 30 s
with a central excitation wavelength (λex) of 800 nm, and the
corresponding TPE-induced fluorescence was monitored over the λmon
range of 400-750 nm.
Two-Photon Absorption Cross Section Measurements. Although
there are several experimental methods for determining molecular two-
photon absorption (TPA) cross sections (δ),19 all the measurements
described in this work utilized the TPA photoluminescence technique
described by Xu and Webb.14 With this method, a sample was excited
with a tightly collimated high-intensity laser beam (d0 ∼ 120 µm), and
the emission was collected at 90° and detected (400-750 nm) using
the spectrometer/CCD combination described in the previous paragraph.
The integrated fluorescence intensity (480-700 nm) was utilized to
determine the TPA cross section according to the equation
Photochemical measurements involving direct excitation of
the absorption bands characteristic of the iron/sulfur/nitrosyl
clusters of the “simpler” RSE complexes (where R is an alkyl
or aryl functionality) showed that these decomposed with
moderate quantum yields (ΦRSE ) 0.02-0.13) when irradiated
in the near UV (366 nm).8a Furthermore, nearly 4 equiv of NO
is released per mole of cluster undergoing photodecomposition.
However, the spectra of these simple RSE are similar to that
of RRS and accordingly display little absorptivity at longer
visible wavelengths. To address the latter issue, RSE complexes
were prepared with dye chromophores incorporated into the
pendant R group, including protoporphyrin IX in the case of
PPIX-RSE and fluorescein in the case of Fluor-RSE (Figure
1). Elsewhere, we have described the photochemical/physical
properties of PPIX-RSE8b and Fluor-RSE18 resulting from
continuous photolysis with visible range light. The low intensity
of the continuous photolysis excitation source defines these
experiments as involving single-photon excitation (SPE). For
both compounds, such excitation into the dye chromophore is
followed by energy transfer to the iron/sulfur/nitrosyl core and
sensitized reaction at that center, including NO release (see
below). For the TPE experiments described here, the larger TPA
cross section and the greater solubility in protic solvents make
fluorescein a good candidate as a light-harvesting TPA antennae
for the Roussin’s clusters, and these properties are reported here
for Fluor-RSE. The ethyl ester of fluorescein (Fluor-Et) was
used as a metal-free model compound for comparisons.
ΦrefδrefcrefP2refI
δ )
K
(2)
ΦcP2Iref
where the subscript ref refers to the experimental measurements for
the reference material under the same conditions, Φ is the photolumi-
nescence quantum yield, c is the concentration, P is the average
excitation power (P and Pref were intentionally as similar as possible),
K is a correction factor accounting for the difference in refractive indices
for the respective reference and sample solutions, and I is the integrated
fluorescence intensity. The derivation of this equation is described in
greater detail in the Supporting Information. The reference utilized was
fluorescein dye (10 µM in aqueous solution at pH 11), which has a δ
) 38 ( 9.7 GM at 782 nm under these conditions.14
Experimental Section
Materials. The synthesis and characterization of Fluor-RSE and
Fluor-Et are described elsewhere.18 Nanopure water was obtained for
spectroscopic studies via a Millipore water purification system.
Acetonitrile was distilled under dinitrogen from calcium hydride (CaH2).
Phosphate buffer solutions (1-10 mM) were prepared in Nanopure
water, and their pH was adjusted to 7.4 using an Orion Research (Model
701 A) digital ionalyzer. Acetonitrile/phosphate buffer mixtures (50/
TPE with the NO Electrode. Nitric oxide concentrations photo-
chemically generated were determined with a NO specific electrode
(amiNO-700 from Innovative Instruments) that had been calibrated from
0 to 400 nM using injections of nitrite standards into acidified, aerated
(17) (a) Seyferth, D.; Gallagher, M. K.; Cowie, M. Organometallics 1986, 5,
539-548. (b) Glidewell, C.; Lambert, R. J.; Harman, M. E.; Hursthouse,
M. B. J. Chem. Soc., Dalton Trans. 1990, 2685-2690.
(18) Wecksler, S. R.; Hutchinson, J.; Ford, P. C. Inorg. Chem. 2006, 45, 1192-
1200.
(19) (a) Shen, Y. R. The Principles of Non-linear Optics; J. Wiley: New York,
1984. (b) Lin, S. H.; Fujimura, Y. F.; Neusser, H. J.; Schlag, E. W.
Multiphoton Spectroscopy of Molecules; Academic Press: Orlando, FL,
1984. (c) Prasad, P. N.; Williams, D. J. Introduction to Nonlinear Optical
Effects in Molecules and Polymers; J. Wiley: New York, 1991.
9
J. AM. CHEM. SOC. VOL. 128, NO. 11, 2006 3833

A R T I C L E S
Wecksler et al.
solutions containing the reducing agent sodium iodide. In these
experiments, samples were prepared in the dark in aerated22 solutions
of 50/50 CH3CN/phosphate buffer (10 mM) at pH 7.4, and then were
irradiated at 800 nm with the Ti:sapphire laser for 30 s intervals under
conditions analogous to those of the TPE photoluminescence measure-
ments. The mixed solvent system was used owing to limited solubility
of Flour-RSE in water and is the same medium used for previous SPE
experiments.18 Samples were placed into the optical train in a high-
quality quartz UV-vis cell equipped with a mini stir flea. The nitric
oxide electrode was then immersed into the solution and carefully
positioned in order to avoid direct illumination by the excitation source.
The solution was stirred using a micro stirrer. The quantity of NO
photochemically produced via TPE was calculated by taking the
difference between the minimum and maximum peak amplitudes before
and after a photolysis interval and comparing this to the calibration
curves. A neutral density filter wheel was utilized to attenuate the energy
of the laser pulses to the desired level, and the energy of the pulses
was determined using a thermopile optical power meter (Newport-
815C).
Figure 2. Steady-state luminescence spectra of Fluor-Et (1.80 µM) and
Fluor-RSE (1.88 µM) in deoxygenated 50/50 CH3CN/phosphate buffer (10
mM) solutions at pH 7.4 under argon, prepared with matched absorbances
(∼0.1) at the excitation wavelength (480 nm).
Results and Discussion
TPE Fluorescence Measurements. Luminescence spectra
of Fluor-RSE and the reference compound Fluor-Et have been
recorded in this laboratory18 using a continuous xenon lamp
excitation source and monochromator set to 480 nm, under
which conditions the excitation should be entirely in the SPE
mode. Deoxygenated solutions of Fluor-RSE (1.88 µM) and
Fluor-Et (1.80 µM) in 50/50 CH3CN/phosphate buffer (10 mM)
at pH 7.4 with matched absorbances (∼0.1) at λexc were used,
and the emission spectra were integrated over the range of 485-
700 nm. Although both samples displayed fluorescence (Figure
2), the integrated emission intensities for Fluor-RSE proved
to be 15% that for Fluor-Et. Concomitant photochemistry of
the iron/sulfur/nitrosyl cluster core (NO release) was also
observed, thus in the former case, the principal decay pathway
of the fluorescent state is apparently energy transfer from the
light-gathering antennae to the iron/sulfur/nitrosyl cluster.
To test their potential TPE properties, similar deaerated
solutions of Fluor-RSE and Fluor-Et were subjected to
excitation using 800 nm light using the output from the Ti:
sapphire laser (6 nJ pulses with half-widths of ∼100 fs at 80
MHz pulse frequency). The respective solution concentrations
(3.81 × 10-6 and 1.92 × 10-5 M) were chosen to give similar
Figure 3. TPE PL spectra of Fluor-RSE (3.81 × 10-6 M) and Fluor-Et
(1.92 × 10-5 M) in 50/50 CH3CN/phosphate buffer (1 mM) at pH 7.4.
absorbances (∼0.06) at the wavelength (400 nm) corresponding
to twice the frequency of the excitation source. Notably, the
recorded spectrum of neither solution displays any absorption
at 800 nm; however, both displayed measurable fluorescence
at energy greater than the excitation energy, indicating the
operation of a TPE mechanism. The data shown in Figure 3
are the emission (450-750 nm) resulting from such 800 nm
excitation of these solutions. The up-converted fluorescence was
recorded for 30 s at a 90° angle from the excitation source.
Notably, while both samples displayed emission as the result
of two-photon excitation under these conditions, the integrated
photoluminescence intensity for Fluor-RSE was 6.7% that of
Fluor-Et. Since the two measurements were made under
identical conditions, the ratio of the integrated TPE photolu-
minescence intensities can be calculated from eq 3.
(20) (a) Meyers, F.; Marder, S. R.; Pierce, B. M.; Bre´das, J.-L. J. Am. Chem.
Soc. 1994, 116, 10703-10714. (b) Baur, J. W.; Alexander, M. D., Jr.;
Banach, M.; Denny, L. R.; Reinhardt, B. A.; Vaia, R. A. Chem. Mater.
1999, 11, 2899-2906. (c) Marder, S. R.; Gorman, C. B.; Meyers, F.; Perry,
J. W.; Bourhill, G.; Bre´das, J.-L.; Pierce, B. M. Science 1994, 265, 632-
635. (d) He, G. S.; Lin, T.-C.; Dai, J.; Prasad, P. N.; Kannan, R.;
Dombroskie, A. G.; Vaia, R. A.; Tan, L.-S J. Chem. Phys. 2004, 120, 5275-
5284.
ΦFluor-RSE
δ
Fluor-RSEcFluor-RSE
IFluor-RSE/IFluor-Et
)
(3)
(21) Albota, M.l et al. Science 1988, 281, 1653-1656.
ΦFluor-Et
δFluor-EtcFluor-Et
(22) As we have earlier shown,5,8 both Roussin’s red salt and the red salt esters
show much higher net photodecomposition when irradiated in the presence
of dioxygen (usually air) than in deaerated solution. This effect can be
attributed to oxygen trapping of the Fe2S2(NO)3 clusters formed by NO
photodissociation in competition to the fast back reaction with NO to
regenerate the starting complex. Flash photolysis studies have confirmed
this analysis. It is for this reason that luminescence measurements for the
Roussin’s salt esters with pendant chromophores are normally carried out
in deaerated solutions in order to reduce the complications resulting from
net photoreactions. However, when the goal is to evaluate the photoreaction
chemistry itself, the photolyses are carried out in aerated solutions. Control
experiments have demonstrated that the luminescence behavior of aerated
and deaerated solutions of Fluor-RSE are initially equivalent; that is, there
is no evidence of the luminescence being directly quenched or enhanced
by the presence of dioxygen. The only effects in these cases are the results
of different long-term stabilities of these solutions.
Substituting the respective Fluor-RSE and Fluor-Et values for
Φ (0.11 and 0.77),18 δ (63 and 32 GM), and c (3.81 × 10-6
and 1.92 × 10-5 M) gives the predicted ratio of 5.6 ( 1.0%, in
reasonable agreement with that determined experimentally.
TPA Cross Sections. The δ values were measured for Fluor-
RSE, Fluor-Et, and fluorescein using the photoluminescence
intensity ratio method described in the Experimental Section,
with fluorescein in pH 11 aqueous solution (δ ) 38 GM at 782
nm)14 used as the reference. For these measurements, solutions
of Fluor-RSE and Fluor-Et were prepared in deaerated 50/50
CH3CN/phosphate buffer (1 mM) at pH 7.4. The samples were
(23) Baker, E. S.; Bushnell, J. E.; Wecksler, S. R.; Lim, M. D.; Manard, M. E.;
Dupuis, N. F.; Ford, P. C.; Bowers, M. T. J. Am. Chem. Soc. 2005, 127,
18222-18228.
9
3834 J. AM. CHEM. SOC. VOL. 128, NO. 11, 2006

Photochemical Delivery of Nitric Oxide Using NIR Light
A R T I C L E S
Figure 4. TPA cross section of Fluor-RSE, Fluor-Et, and fluorescein.
Figure 6. NO electrode response due to TPE of Fluor-RSE (9.1 µM) in
50/50 CH3CN/phosphate buffer (10 mM) at pH 7.4 with ∼100 fs pulses
operating at a repetition rate of 80 MHz, a central wavelength at 800 nm,
and an average power of 450 mW.
nm light (∼100 fs pulses operating at 80 MHz with an average
power of 450 mW) for a total of ∼10 min of irradiation. Under
these conditions, the changes in the optical spectra (principally
strong absorption increases across the spectrum) closely parallel
those seen for SPE studies carried out using continuous
photolysis of analogous solutions with 436 nm light.18 Clearly
TPE with 800 nm light leads to sensitized photochemical
reactions. As discussed with regard to the SPE photochemistry
of Fluor-RSE,18 we have no ready explanation for the absorp-
tion increases observed upon photodecomposition, other than
to suggest that these may be due to differences in the pKa’s of
the fluorescein chromophores between Fluor-RSE and the
fluorescein ethyl thiolate photoproduct. Ion mobility mass
spectrometry and computation experiments23 are underway with
the goal of gaining insight into three-dimensional structural
features that might influence photophysical properties.
Photochemical Production of NO from TPE of Fluor-RSE
with NIR Irradiation. The goal of these experiments was to
establish quantitatively the NO production via reaction of the
RRS cluster resulting from photosensitization by TPE of the
pendant chromophores. A nitric oxide specific electrode and
the procedures described in the Experimental Section were used
to evaluate the NO concentration generated by 800 nm excitation
of an aerated Fluor-RSE solution (9.1 µM) in 50/50 CH3CN/
phosphate buffer (10 mM) at pH 7.4. Immediately upon
exposing the solution to the laser, the electrode displayed
substantial current increases, indicating the photochemical
production of NO, which leveled off once excitation was
terminated (Figure 6). There is a lag time in the electrode
response that we attribute to diffusion of NO from the excitation
point to the electrode. This was also seen with SPE experi-
ments.18 Control experiments in which solutions were irradiated
without the Fluor-RSE present showed no change in the current
due to exposure to the laser. The concentrations of nitric oxide
photochemically generated from TPE were obtained by taking
Figure 5. Optical absorption changes of Fluor-RSE (9.1 µM) in 50/50
CH3CN/phosphate buffer (10 mM) at pH 7.4 before and after TPE at 800
nm for ∼10 min of irradiation.
prepared in the dark, blanketed with argon, and kept from
exposure to extraneous light before and during the measurement.
Previous photochemical experiments of Fluor-RSE demon-
strated that the complex undergoes photodecomposition to a
more luminactive fluorophore.18 Therefore, to avoid sample
degradation, data collection times were limited, and each sample
was only used for two data points in the δ versus λex spectrum.
Control experiments were performed under identical conditions
to ensure the integrity of the samples during collection times,
and in these experiments, the fluorescence intensity of Fluor-
RSE changed less than 5% after TPE at 800 nm for a period of
approximately 5 min.
Figure 4 shows the results of these TPA cross section
measurements over the wavelength range of 730-890 nm. The
δ values for fluorescein and Fluor-Et are similar, with δ ) 38
and 32 ( 3 GM, respectively, at 800 nm. Thus, the synthetic
modification of fluorescein by forming the ethyl ester of the
carboxylic acid functionality resulted in little change to the two-
photon photophysics of this complex. Although the TPA cross
section measurement for the Fluor-RSE showed a similar TPA
spectral profile, the δ value proved to be somewhat higher at
800 nm and at shorter λex than for the free chromophores. The
reason for the enhanced δ observed for Fluor-RSE is unknown,
but it may simply be the result of having two independent dye
chromophores. On the other hand, δ is believed to correlate
with the quadrupole moment of the molecule,19-21 and given
the molecular formula illustrated (Figure 1), it seems likely that
Fluor-RSE with two pendant chromophores about an iron/
sulfur/nitrosyl core would have a larger quadrupole moment than
does Fluor-Et or fluorescein.
(24) Continuous photolysis investigations of Roussin’s red salt and the simple
esters demonstrated that net photodecomposition quantum yields (Φd) are
dependent on the photon flux of excitation (Ia).5,8 The Φd decreases
somewhat upon increasing Ia in a manner consistent with reversible NO
labilization to give cluster intermediates that are trapped by solution O2
competitive with back reaction with NO to regenerate the starting complex.
The extent of back reaction is dependent on the steady-state concentrations
of NO and the cluster intermediates that are in turn dependent on Ia. Such
considerations might have some modest impact upon Figure 5, thus
explaining the slope slightly less than 2.0. However, the [NO] generated
under TPE conditions is low, so that impact is likely to be minor.
Does TPA by Fluor-RSE Lead to Photoreaction? Figure
5 illustrates the changes in the absorption spectrum of an
aerated22 solution of Fluor-RSE before and after TPE with 800
9
J. AM. CHEM. SOC. VOL. 128, NO. 11, 2006 3835

A R T I C L E S
Wecksler et al.
Scheme 1. Schematic Illustrating the Consequences of TPA at 800 nm by Fluor-RSE
the difference between the minimum and maximum peak heights
before and after an illumination period and calculated from the
calibrations performed before and after photolysis.
Figure 6 illustrates such a NO electrode measurement for
repeated TPE excitation for a single solution. In this experiment,
the concentration of NO photochemically produced during the
first 30 s of illumination was ∼4.7 nM, and if one assumes
that the red salt esters photochemically generate ∼4 equiv of
NO for each mole of RSE photodecomposition (as in SPE),
then this would correspond to ∼0.015% depletion of the initial
Fluor-RSE within this time period. Experiments of this type
generally gave reproducible data within an experimental un-
certainty of (15%.
Figure 7. Log [NO]/s versus log P plot, where [NO]/s is the concentration
of nitric oxide released in 30 s under 800 nm excitation, and P is the average
power of the pulsed laser (100 fs pulses at 80 MHz) which is proportional
to the pulse intensity I. Slope ) 1.8 ( 0.15.
The observation of NO production via electrochemical
detection and the changes in the optical absorption spectrum
upon using 800 nm laser pulse to excite solutions of Fluor-
RSE both point to the operation of a photochemical mechanism
initiated by simultaneous two-photon absorption (Scheme 1).
However, since it is possible that these observations might be
the result of thermal excitation or some other linear process, it
would be desirable to provide alternative evidence for a TPE
process. According to the “formal intensity law”,19a a photo-
chemical or photophysical process can be described in terms
of an intensity dependence. For example, the transition prob-
ability W(n) of an n photon absorption (nonresonant case) is a
linear function of In, the specific case of TPA being proportional
to I2. Thus a plot of log W(n) versus log I should take the form
log W(n) ) n log I + C, where C is a constant independent of
I. For a TPA process, the parameter being measured (photolu-
minescence intensity, NO release, etc.) must be a function of
the intensity squared; that is, the slope of the log-log plot of
the measured quantity versus intensity should equal 2.
In this context, experiments were undertaken in order to
measure the quantitative NO release as a function of irradiation
intensity. The experimental setup was that described previously
for the TPE of Fluor-RSE in conjunction with the NO electrode,
the only difference being that the irradiation intensity was varied
by the use of neutral density filters. The irradiation intensities
were measured using a thermopile optical power meter and were
found to be 211, 375, and 450 mW for the lowest, medium,
and highest intensities, respectively.24 (Several measurements
at lower irradiation intensity were attempted, but these resulted
in minimal changes in the NO electrode current.) The samples
were irradiated for 30 s while the [NO] was measured using
the NO specific electrode. Three data points were collected for
each sample, and the average [NO] for the photochemically
released nitric oxide was determined using the calibration curves.
The respective concentrations of NO detected in these measure-
ments were 1.3 ( 0.2, 3.8 ( 0.5, and 4.7 ( 0.7 nM. The plot
of log [NO]/s versus log P (Figure 7) displays a slope of 1.8 (
0.15 consistent with the quadratic dependence of NO release
on excitation intensity. In other words, this slope is consistent
with a mechanism for nitric oxide generation resulting from
two-photon excitation of the pendant chromophore(s) of Fluor-
RSE with NIR light, followed by energy transfer to the cluster
core resulting in sensitization of the photochemical production
of NO.
If it is assumed that the products and quantum yields of
Fluor-RSE photoreactions are the same for TPE as for SPE,
then one should be able to use the parameters of the experiments
determined here to estimate the quantity of NO that should be
photochemically generated by TPE under these conditions. From
the rate equation for TPA by a molecule,25 the intensity of light
that is absorbed by this mechanism changes upon passing
through a sample according to
2
γ ‚ l ‚ I0
∆I )
(4)
γ ‚ l ‚ I0 + 1
where ∆I ) I0 - I, I0 is the input intensity for each laser pulse
in W/cm2, I is the transmitted light, and l is the path length (in
cm). The term γ is defined as δn/2hω, where δ is the TPA
cross section (in cm4 s/photon), n is the concentration of the
solution in molecules‚cm-3, h is Planck’s constant (1.05 × 10-34
J‚s), and ω is 2πc/λ. I0 is calculated from
(25) Freidrich, D. M. J. Chem. Educ. 1982, 59, 472-481.
9
3836 J. AM. CHEM. SOC. VOL. 128, NO. 11, 2006

Photochemical Delivery of Nitric Oxide Using NIR Light
A R T I C L E S
of Fluor-RSE at 800 nm would be generated, leading a
concentration of NO of approximately 6.3 nM after 30 s.
Experimentally, the amount of NO generated from TPA of
Fluor-RSE for 30 s was found to be ∼4.7 nM. Given a number
of uncertainties inherent to the experiment and to the estimate
(one being the assumption that the laser pulse is a uniform
cylinder of light rather than having a Gaussian shape in time
and in radial distribution), the estimated and experimental values
are remarkably (more likely, fortuitously) close. Nonetheless,
this calculation provides further support for the operation of a
two-photon excitation mechanism.
4P
I0 )
(5)
2
π ‚ f ‚ τ ‚ d0
where P is the (average) laser power (in W), f is the repetition
rate, τ is the pulse duration, and d0 is the diameter of the laser
beam (assuming that the laser is collimated). If each pulse
contains N0 photons before entering the sample, and ∆N is the
number of photons reduced by TPA, then the number of excited
states generated by TPA from a single pulse will be ∆N/2.
Substituting into eq 4 gives
γ ‚ l ‚ I0
I0 γ ‚ l ‚ I0 + 1
∆N ∆I
)
)
(6)
N0
Summary
Therefore
These studies have successfully demonstrated the viability
of utilizing two-photon excitation with near-infrared light for
the photochemical activation of a model pro-drug, the fluores-
cein dye-derivatized iron/sulfur/nitrosyl complex Fluor-RSE.
The potential advantages of using NIR TPE for photochemical
drug delivery are 2-fold: deeper light penetration through tissue
and enhanced spatial selectivity for drug activation. TPE
photoluminescence measurements of Fluor-RSE show visible
range emission from the fluorescein chromophore(s), indicating
frequency up-conversion of the excitation light at 800 nm but
at a marked decrease in PL intensity relative to that of the
reference compound Fluor-Et. The decreased intensity is an
indicator of efficient energy transfer from the fluorescein
chromophores to the iron/sulfur cluster as seen previously with
single-photon excitation with near UV wavelengths. The TPA
cross section of Fluor-RSE was determined to be 63 ( 5 GM
at 800 nm using the TPE PL technique, and this is a somewhat
larger δ than that found for Fluor-Et or for fluorescein itself.
Last, NIR TPE of Fluor-RSE leads to the release of NO into
the solution as determined quantitatively using a NO specific
electrode. The quadratic dependence of the NO generated on
the excitation pulse intensities shows unequivocally that NO is
being produced via a two-photon process.
N0 γ ‚ l ‚ I0
2 γ ‚ l ‚ I0 + 1
ES(pulse-1) )
‚
(7)
where ES(pulse-1) is the number of excited states formed by
TPE in each pulse. Since N0 ) P/(fhω), one can calculate the
excited states formed per second from
P
B
ES(s-1) )
‚
(8)
2hω B + 1
where B ) γ l I0. Since B , 1, eq 7 simplifies to
BP
ES(s-1) )
2hω
(9)
For the average laser power (P) of 500 mW (0.5 W), a pulse
duration (τ) of 100 fs, operating at a repetition rate (f) of 80
MHz, λexc ) 800 nm, and a beam diameter of 120 µ, I0 is found
to be 5.7 × 108 W/cm2. The experimentally determined cross
section (δ) for Fluor-RSE is 63 GM, or 6.3 × 10-49 cm4
s/photon, at 800 nm. For these measurements, a quartz cuvette
with a 1.0 cm path length, a sample volume of 3.5 mL, and a
concentration of Fluor-RSE approximately 1.0 × 10-5 M were
used. For these conditions, γ ) 7.56 × 10-14 cm/W, entering
these values into eq 5 estimates the rate of excited state
formation under these experimental conditions to be ES ) 1.51
× 1013 cm-3 s-1 or equivalently 21 nM/s.
Acknowledgment. These studies were supported by the
National Science Foundation (CHE-0352650).
We have determined previously from SPE experiments that
the quantum yield for photodecomposition of Fluor-RSE at λirr
) 436 nm is Φ ) 0.0036 ( 0.005, and that the compound
releases 3.2 equiv of NO per mole of cluster decomposed.18
Thus the quantum yield for NO release from this system is
approximately 0.01 or 1.0% at 436 nm. According to this
estimate, approximately 0.21 nM of NO per second upon TPE
Supporting Information Available: Complete ref 21 and
more detailed derivation of eq 2. This material is available free
of charge via the Internet at http://pubs.acs.org.
JA057977U
9
J. AM. CHEM. SOC. VOL. 128, NO. 11, 2006 3837