Synthesis, properties, and photodynamic properties in vitro of heavy-chalcogen analogues of tetramethylrosamine

Article abstract of DOI:10.1016/j.bmc.2004.03.029

Thio and seleno analogues of tetramethylrosamine were prepared by the directed-metalation/cyclization of the corresponding N,N-diethyl 2-(3-dimethylaminophenylchalcogeno)-4-dimethylaminobenzamide to the 2,7-bis-(N,N-dimethylamino)-9H-chalcogenoxanthen-9-one followed by the addition of phenylmagnesium bromide, dehydration, and ion exchange to the chloride salt. The thio and seleno tetramethylrosamines had longer wavelengths of absorption and higher quantum yields for the generation of singlet oxygen than tetramethylrosamine. Both the thio and selenoanalogues of tetramethylrosamine were efficient photosensitizers against R3230AC rat mammary adenocarcinoma cells in vitro.

Full text of DOI:10.1016/j.bmc.2004.03.029

Bioorganic & Medicinal Chemistry 12 (2004) 2537–2544
Synthesis, properties, and photodynamic properties in vitro
of heavy-chalcogen analogues of tetramethylrosamine
Michael R. Detty,a,* Paras N. Prasad,a David J. Donnelly,a Tymish Ohulchanskyy,a
Scott L. Gibsonb and Russell Hilf b
aInstitute for Lasers, Photonics, and Biophotonics, Department of Chemistry, University at Buffalo,
The State University of New York, Buffalo, NY 14260, USA
bDepartment of Biochemistry and Biophysics, University of Rochester Medical Center, 601 Elmwood Avenue, PO Box 607,
Rochester, NY 14642, USA
Received 26 January 2004; revised 1 March 2004; accepted 14 March 2004
Available online 13 April 2004
Abstract—Thio and seleno analogues of tetramethylrosamine were prepared by the directed-metalation/cyclization of the corre-
sponding N,N-diethyl 2-(3-dimethylaminophenylchalcogeno)-4-dimethylaminobenzamide to the 2,7-bis-(N,N-dimethylamino)-9H-
chalcogenoxanthen-9-one followed by the addition of phenylmagnesium bromide, dehydration, and ion exchange to the chloride
salt. The thio and seleno tetramethylrosamines had longer wavelengths of absorption and higher quantum yields for the generation
of singlet oxygen than tetramethylrosamine. Both the thio and selenoanalogues of tetramethylrosamine were efficient photosensi-
tizers against R3230AC rat mammary adenocarcinoma cells in vitro.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
combination of a tumor-specific photosensitizer, light,
and molecular oxygen to induce cellular toxicity, pre-
sumably via the generation of singlet oxygen.12 While
rhodamine and rosamine dyes exhibit selective uptake in
cancer cells, they are poor producers of excited-state
triplets14 and, consequently, of singlet oxygen. The poor
triplet production limits the use of rhodamine and ros-
amine dyes as photosensitizers PDT. Furthermore, the
rhodamines and rosamines absorb light of wavelengths
too short for effective penetration in tissue.12
The xanthylium nucleus is the heart of many important
chromophores in chemistry and biology. The rhodamine
and rosamine dyes are representative of the xanthylium
class and have been used as laser dyes, fluorescent labels,
and fluorescence emission standards where their high
fluorescence quantum yields and photostabilities are
exploited.1–5 The rhodamine dyes are also useful fluo-
rescence probes in cell biology studies, showing specific
fluorescence staining of mitochondria and other cell
organelles.6;7 The rhodamines have been found to
accumulate selectively in carcinoma cells8–10 and to be
toxic to cancer cells both in vitro10 and in vivo.11
Heavy-atom effects in brominated rhodamine dyes give
increased triplet yields and quantum yields for the gen-
eration of singlet oxygen [/(1O2)]15–17 relative to
unmodified rhodamines.18 However, wavelengths of
absorption are little changed relative to their light-atom
counterparts. The brominated analogues still target the
mitochondria and have increased phototoxicity toward
cancer cells.17 We describe the synthesis of thio and
seleno analogues of tetramethylrosamine (TMR-S and
TMR-Se, respectively) where the heavy atom is part of
the xanthylium chromophore. Not only does substitu-
tion of sulfur or selenium for oxygen give dramatic
One area where the xanthylium dyes have been mini-
mally utilized is photodynamic therapy (PDT), where
their tumor specificity might truly be exploited.12;13 PDT
is a treatment for various cancers that utilizes the
Keywords: Photodynamic therapy; Anticancer drugs; Photosensitizers;
Tetramethylrosamines; Thioxanthylium; Selenoxanthylium.
* Corresponding author. Tel.: +1-716-645-6800; fax: +1-716-645-6963;
e-mail: mdetty@buffalo.edu
1
increases in the production of O2, but this substitution
also gives longer wavelengths of absorption and
0968-0896/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.03.029

2538
M. R. Detty et al. / Bioorg. Med. Chem. 12 (2004) 2537–2544
emission and provides highly effective photosensitizers
for PDT against R3230AC rat mammary adenocarci-
noma cells in vitro.
and 71% isolated yields, respectively. Anion exchange
with Amberlite IRA-400 chloride exchange resin gave
the corresponding chloride salts TMR-Se, and TMR-S
in 78% and 75% isolated yields, respectively.
2. Results and discussion
2.1. Synthesis
2.2. Photophysical properties
Various photophysical properties of TMR-S and TMR-
Se are compared to those of commercially available
tetramethylrosamine (TMR-O, Molecular Probes, Inc.)
in Table 1. In MeOH, a 20-nm bathochromic shift for
TMR-S and a 30-nm bathochromic shift for TMR-Se in
the wavelength of the absorption maximum, kmax, is
observed as the chalcogen atom becomes larger and the
oscillator strength of the band decreases as indicated by
a decrease in the molar extinction coefficient, e. Other
series of chalcogenopyrylium dyes show similar trends.20
We have described the synthesis of 2,7-bis-N,N-di-
methylamino-9H-selenoxanthen-4-one (1) from acyclic
precursors via directed metalation reactions as shown in
Scheme 1.19 The same approach was used for the syn-
thesis of 2,7-bis-N,N-dimethylamino-9H-thioxanthen-4-
one (2). Lithiation of N,N-diethyl 4-dimethylamino-
benzamide with tert-BuLi in the presence of N,N,N0,N0-
tetramethylethylenediamine (TMEDA) followed by the
addition of bis(3-dimethylaminophenyl)diselenide (3) or
disulfide (4) gave N,N-diethyl-2-(30-dimethylamino-
phenylseleno)-4-dimethylaminobenzamide (5) or 2-N,N-
diethyl-2-(30-dimethylaminophenylthio)-4-dimethylamino-
benzamide (6) in 57% and 51% isolated yields, respectively.
Cyclization of 5 and 6 with 2 equiv of lithium diiso-
propylamide (LDA) gave selenoxanthenone 1 in 23%
yield and thioxanthenone 2 in 13% yield, respectively,
with 70% recovered starting material in each case. The
starting material could be recycled to give more 1 and 2.
Use of larger excesses of LDA gave up to 30% yields of 1
and 2, but no starting material was recovered under
these conditions.
A 25–28 nm Stokes shift is observed in the fluorescence
emission maximum, kF, which is typical of other cationic
dyes.20 The radiative lifetime, srad, decreases from
2.1 ꢀ 0.1 ns in TMR-O to 1.5 ꢀ 0.1 ns for TMR-S to
approximately 50 ps for TMR-Se, which is consistent
with heavy-atom effects in other cationic dyes.20
The rigid nature of the xanthylium core minimizes
internal conversion back to the ground state, which
leads to dramatic changes in /F and /(1O2) as the
chalcogen atom becomes larger. Fluorescence in TMR-
O is highly efficient with /U of 0.84. Intersystem crossing
to the triplet, which is necessary for the production of
singlet oxygen, is relatively slow with /(1O2) of 0.08 for
TMR-O. As the chalcogen atom becomes larger, the
spin-orbit (heavy-atom) effects promote triplet forma-
tion relative to fluorescence and /(1O2) decreases to 0.44
and 0.009 for TMR-S and TMR-Se, respectively, while
The addition of phenylmagnesium bromide to chalco-
genoxanthen-9-ones 1 and 2 followed by dehydration
with HPF6 gave 2,7-bis-N,N-dimethylamino-9-phen-
ylselenoxanthylium and 2,7-bis-N,N-dimethylamino-9-
phenylthioxanthylium hexafluorophosphates in 98%
Scheme 1. Preparation of TMR-Se and TMR-S.
Table 1. Absorption (kmax) and fluorescence (kF) maxima in MeOH, quantum yields for fluorescence (/F) and the generation of singlet oxygen
[/(1O2)], reduction potentials (E°), and n-octanol/water partition co-efficients (log P) for TMR-O, TMR-S, and TMR-Se
Compound
kmax (nm) (log e)
kF (nm)a
/
F ꢀ sdb
/(1O2) ꢀ sdc
E° (V)d
Log Pe
TMR-O
TMR-S
TMR-Se
552 (4.92)
571 (4.70)
582 (4.84)
575
599
608
0.84 ꢀ 0.01
0.08 ꢀ 0.01
0.21 ꢀ 0.02
0.87 ꢀ 0.01
ꢁ0.94
ꢁ0.79
ꢁ0.77
1.5
1.2
1.1
0.44 ꢀ 0.01
0.009 ꢀ 0.001
a Excitation at 532 nm.
b Quantum efficiencies using rhodamine 6 G as a standard.
c Direct detection of singlet-oxygen luminescence using rose bengal as a standard.
d In CH2Cl2 with 0.2 N Bu4NBF4 as supporting electrolyte. V versus the ferrocene/ferrocinium couple (E° ¼ þ0:40 V).
e pH 6 Phosphate buffer as the aqueous phase.

M. R. Detty et al. / Bioorg. Med. Chem. 12 (2004) 2537–2544
2539
values of /(1O2) increase to 0.21 and 0.87 for TMR-S
and TMR-Se, respectively.
2.4. Biological studies. Dark and phototoxicity toward
R3230AC cells
The tetramethylrosamine series was evaluated for dark
and phototoxicity against R3230AC rat mammary
adenocarcinoma cells. Cell cultures were incubated for
3 h in the dark with various concentrations of dye and
were then washed prior to treatment with filtered 360–
2.3. Chemical properties
Mitochondrial reduction of a cationic photosensitizer
can bleach the photosensitizer limiting the useful lifetime
of the photosensitizer. Furthermore, the cationic photo-
sensitizer’s role as an electron acceptor might impact the
redox cascade in mitochondrial respiration. The reduc-
tion potentials of TMR-O, TMR-S, and TMR-Se were
determined by cyclic voltametry and values of E° for the
cation/neutral radical couple [vs the ferrocene/ferro-
cinium couple (Fc/Fcþ) at þ0.40 V] are compiled in
Table 1. An anodic (positive) shift in E° is observed as
the chalcogen atom becomes heavier. The reduction of
the tetramethylrosamines is more cathodic (more diffi-
cult to reduce) than the thiopyrylium dye AA1 [E° of
ꢁ0.71 V (vs Fc/Fcþ)], which shows antitumor activity by
targeting the mitochondria of carcinoma cells21 and
inhibiting mitochondrial ATPase activity.21
800 nm light from
1.4 mW cmꢁ2 for a total light dose up to 5.0 J cmꢁ2
a tungsten–halogen source at
.
Light-treated cells and dark controls were incubated for
24 h and cell survival was determined. At concentrations
61.0 lM, none of the tetramethylrosamine series
showed any significant dark toxicity toward R3230AC
cells relative to untreated dark controls (P < 0:05) and
values of LD50, the concentration necessary to give 50%
cell kill in the dark with 3 h incubation, were >5lM for
each dye.
TMR-O displayed no added phototoxicity that was
statistically significant upon irradiation, which is con-
sistent with results with other rhodamines. However,
both TMR-S and TMR-Se were efficient photosensitiz-
ers toward R3230AC cells with phototoxicity increasing
with dye concentration and constant light dose (Fig. 1a)
or with constant dye concentration and increasing light
dose (Fig. 1b). With 0.1 lM solutions of either TMR-S
or TMR-Se and irradiation with 5.0 J cmꢁ2 of 360–
800 nm light, surviving fractions were less than 0.50
(Fig. 1a). Furthermore, statistically significant
(P < 0:05) phototoxicity was also observed with
0.05 lM TMR-S and TMR-Se (Fig. 1a). For both
TMR-S and TMR-Se, the therapeutic ratio LD50
divided by the effective concentration to kill 50% of the
cells (EC50, 60.1 lM for each) was >50.
Values of the n-octanol/water partition coefficient
(log P) were determined for the tetramethylrosamine
series using a pH 6 phosphate buffer22 as the aqueous
phase and are compiled in Table 1. Values of log P are
somewhat smaller for the tetramethylrosamine series
(log P of 1.1–1.5) relative to AA1 (log P of 1.9), which
might have some impact with respect to uptake and
localization within the cell.
2.4.1. Intracellular accumulation of TMR-O, TMR-S,
and TMR-Se. The intracellular accumulation of TMR-O
and TMR-S was determined in R3230AC cells incu-
bated with 10 lM dye for 3 h. Longer incubation periods
had little impact on final intracellular concentrations.
Intracellular dye concentration was determined by
Figure 1. Effect of dyes TMR-O, TMR-S, and TMR-Se on the viability of R3230AC cells in culture in the dark or 24 h after exposure to (a) various
concentrations of dye and 5.0 J/cm2 of filtered 360–800 nm light or (b) 0.1 lM dye and variable light doses. Each point represents three separate
experiments performed in duplicate for the viability of cultured R3230AC cells maintained in the dark or 24 h after exposure of cells to light. Data are
expressed as the fraction of control cell viability, exposed to neither dyes nor light, error bars are the SEM.

2540
M. R. Detty et al. / Bioorg. Med. Chem. 12 (2004) 2537–2544
ated with TMR-S is the result of improved photophys-
ical properties. While we could not quantify the cellular
uptake of TMR-Se, confocal microscopy studies con-
firmed the uptake of TMR-Se by the R3230AC cells.
Continuing studies will focus on the subcellular locali-
zation of TMR-S and TMR-Se as well as the perfor-
mance of TMR-S, and TMR-Se in vivo.
4. Experimental
4.1. General methods
Figure 2. CSLM image of R3230AC cells treated with 2 lM TMR-Se
for 1.5 h with excitation at 532 nm.
Solvents and reagents were used as received from
Sigma–Aldrich Chemical Co. (St. Louis, MO) unless
otherwise noted. Tetramethylrosamine (TMR-O) was
purchased from Molecular Probes, Inc. Cell culture
media and antibiotics were obtained from Grand Island
Biological (Grand Island, NY). Fetal bovine serum
(FBS) was purchased from Atlanta Biologicals (Atlanta,
fluorescence in cell lysates and was 16 ꢀ 3 fmol of dye
per cell for TMR-O and 14 ꢀ 2 fmol of dye per cell for
TMR-S.
€
GA). Concentration in vacuo was performed on a Buchi
The fluorescence from cell lysates of TMR-Se treated
cells was too weak to determine the intracellular dye
concentration accurately. However, the cellular uptake
of TMR-Se could be observed by confocal scanning
laser microscopy (CSLM). R3230AC cells were incu-
bated for 1.5 h with 2 lM TMR-Se in culture media and
were then washed and covered with fresh media. The
CSLM fluorescence micrographs of these cells show a
diffuse fluorescence from TMR-Se within the cell with
minimal staining of the nucleus (Fig. 2).
rotary evaporator. Chalcogenoxanthones 1 and 2 were
prepared according to Ref. 19. NMR spectra were
recorded on a Varian Inova 500 instrument with resid-
ual solvent signal as internal standard. UV–vis-near-IR
spectra were recorded on a Perkin–Elmer Lambda 12
spectrophotometer equipped with a circulating con-
stant-temperature bath for the sample chambers. Ele-
mental analyzes were conducted by Atlantic Microlabs,
Inc.
4.1.1. Preparation of bis(3-dimethylaminophenyl)disulfide
(4). 3-Bromo-N,N-dimethylaniline (1.0 g, 5.0 mmol) was
added to a stirred mixture of Mg turnings (0.243 g,
10.0 mmol) in 10 mL of anhydrous THF. The resulting
mixture was heated reflux for 2 h and then cooled to
ambient temperature. Sulfur powder (0.39 g, 5.0 mmol)
was added and the resulting mixture was heated at reflux
for 2 h. The reaction mixture was cooled to ambient
temperature and then poured over 6 g of ice. HCl
(10 mL, 5%) was added and the resulting mixture was
filtered through a pad of Celite. The crude thiol was
oxidized via the addition of 0.074 g (0.24 mmol) of
dihexyltelluride and 10 mL of 30% H2O2.23 After stirring
for 1 h, the reaction mixture was poured into 50 mL of
water and the crude product was extracted with ether
(3 ꢂ 25 mL). The combined organic extracts were
washed with brine, dried over MgSO4, and concen-
trated. The crude oil was purified via chromatography
on SiO2 eluted with 10% cyclohexane/CH2Cl2 to give
0.79 g (52%) of the disulfide 4 as a white powder, mp 91–
3. Conclusions
The rhodamines and related structures have interesting
biological properties that included selective localization
in the mitochondria of carcinoma cells.8–10 At high
concentrations, these materials can show antitumor
activity.10;11 Unfortunately, the rhodamines have had
limited use in PDT since these molecules absorb only
wavelengths of light that do not penetrate tissues effec-
tively and show no added phototoxicity upon irradia-
tion unless heavy atoms are incorporated into their
structure. While brominated rhodamines function as
photosensitizers, their wavelengths of absorption are
unchanged by this substitution.16;17 Highly efficient
photosensitizers for photodynamic therapy can be ob-
tained by substituting the heavier chalcogen atoms sul-
fur and selenium for the oxygen atom in these molecules.
Both TMR-S and TMR-Se absorb longer wavelengths
of light and generate singlet oxygen much more effi-
ciently upon irradiation than TMR-O.
1
92 °C: H NMR (CD2Cl2) d 7.81 (d, 2H, J ¼ 6:7 Hz),
7.21 (d ꢂ d, 1H, J ¼ 2:0, 6.7 Hz), 3.35 (s, 12 H); HRMS
(EI), m=z 304.1065 (calcd for C16H20N2S2: 304.1068).
Anal. Calcd for C16H20N2S2: C, 63.12; H, 6.62; N, 9.20.
Found: C, 63.08; H, 6.62; N, 9.23.
In R3230AC rat mammary adenocarcinoma cells, values
of EC50 for both TMR-S and TMR-Se are <0.1 lM with
5.0 J cmꢁ2 of filtered 360–800 nm light while the LD50 of
dark controls is >5 lM. These in vitro studies have a
therapeutic ratio of >50. Fluorescence studies indicated
that the uptake of TMR-S and TMR-O was nearly
identical, which suggests that the phototoxicity associ-
4.1.2. Preparation of N,N-diethyl-2-(30-dimethylamino-
phenylthio)-4-dimethylaminobenzamide (6). sec-Butyl-

M. R. Detty et al. / Bioorg. Med. Chem. 12 (2004) 2537–2544
2541
lithium (1.3 M in cyclohexane, 2.7 mL, 3.6 mmol) was
added dropwise to a stirred solution of N,N-diethyl 4-
N,N-dimethylamino benzamide (0.79 g, 3.6 mmol) and
N,N,N,N-tetramethylethylenediamine (TMEDA, 0.42 g,
3.6 mmol) in 25 mL of anhydrous THF at ꢁ78 °C. After
1 h at ꢁ78 °C, disulfide 4 (1.43 g, 3.6 mmol) in 5 mL of
THF was added dropwise. After 0.5 h at ꢁ78 °C, the
reaction mixture was warmed to ambient temperature.
Saturated NH4Cl (10 mL) was added and the products
were extracted with CH2Cl2 (3 ꢂ 30 mL). The combined
organic extracts were washed with brine, dried over
MgSO4, and concentrated. The crude product was
purified via chromatography on SiO2 eluted with 20%
ether/CH2Cl2 to give 0.25 g (51%) of 6 as a pale yellow
oil: 1H NMR (CD2Cl2) d 7.17 (t, 1H, J ¼ 8 Hz), 7.08 (d,
1H, J ¼ 8:5 Hz), 6.81 (t, 2H, J ¼ 2 Hz), 6.69 (d, 1H,
J ¼ 8 Hz), 6.63 (dd, 1H, J ¼ 2, 8.5 Hz), 6.60 (d, 1H,
J ¼ 2 Hz), 6.58 (d ꢂ d, 1H, J ¼ 2, 8.5 Hz); 3.50 (ex-
change br s, 4 H), 3.18 (s, 6 H), 2.86 (exchange br s, 6H);
13C NMR (CD2Cl2) d 169.5, 151.4, 151.0, 135.7, 133.8,
129.7, 127.5, 127.2, 119.7, 115.6, 115.0, 111.6, 111.0,
43.2 (br), 40.5, 40.2, 14.2 (br); IR (film, NaCl) 1621,
1594 cmꢁ1; HRMS (ES) m=z 372.2109 (calcd for
C21H30N3OS þ H: 372.2104).
product was recrystallized by dissolving the solid in
2 mL of hot CH3CN, cooling to ambient temperature,
slowly adding 2 mL of ether, and chilling. The recrys-
tallized product was collected by filtration, washed with
ether (2 ꢂ 5 mL), and dried to give 0.109 g (98%) of the
hexafluorophosphate salt as a dark green solid: mp
1
>260 °C; H NMR (500 MHz, CD2Cl2) d 7.62 (m, 3H),
7.44 (d, 2H, J ¼ 9:8 Hz), 7.30 (m, 2H), 7.27 (d, 2H,
J ¼ 2:5 Hz), 6.83 (d ꢂ d, 2H, J ¼ 2:5, 9.8 Hz), 3.25 (s,
12H); 13C NMR (500 MHz, CD2Cl2) d 161.3, 152.6,
146.1, 138.1, 136.9, 128.9, 128.8, 128.3, 119.6, 114.4,
109.2, 40.4; kmax (CH2Cl2) 582 nm (e ¼ 6:9 ꢂ 104
M
ꢁ1 cmꢁ1); HRMS (ES) m=z 407.1038 (calcd for
C23H23N820Se:
407.1026).
Anal.
Calcd
for
C23H23F6N2OPSe: C, 50.10; H, 4.20; N, 5.08. Found: C,
50.20; H, 4.32; N, 4.96.
The hexafluorophosphate salt (0.109 g, 0.068 mmol) was
dissolved in 20 mL of acetonitrile and 0.500 g of Am-
berlite IRA-400 chloride ion exchange resin was added.
The resulting mixture was stirred 0.5 h, the exchange
resin was removed via filtration, and the process was
repeated with two additional 0.500 g aliquots of the ion
exchange resin. Following the final ion exchange, the
filtrate was concentrated and the solid residue was
recrystallized from acetonitrile, and a small amount of
diethyl ether to give 0.081 g (75%) of 2,7-bis-N,N-di-
methylamino-9-phenylselenoxanthylium chloride (TMR-
Se) as a dark purple solid: mp >260 °C; 1H NMR
(500 MHz, CD2Cl2) d 7.63 (m, 3H), 7.44 (d, 2H,
J ¼ 6:1), 7.30 (d ꢂ d, 2H, J ¼ 1:8, 7.3 Hz), 7.28 (d ꢂ d,
2H, J ¼ 2:1, 7.3 Hz), 6.83 (d ꢂ d, 2H, J ¼ 2:1, 9.1 Hz),
3.25 (s, 12H); kmax (H2O) 580 nm (e ¼ 5:9 ꢂ 104 Mꢁ1
cmꢁ1); HRMS (ES) m=z 407.1038 (calcd for
C23H23N820Se: 407.1026). Anal. Calcd for C23H23-
ClN2OSe: C, 62.52; H, 5.25; N, 6.34. Found: C, 62.46;
H, 5.08; N, 6.28.
4.1.3. Preparation of 2,7-bis-N,N-(dimethylamino)thio-
xanthen-9-one (2). To a solution of 6 (0.52 g, 1.5 mmol)
in 10 mL of THF at 25 °C was added LDA (1.0 M in
hexanes, 3.6 mL, 3.6 mmol). The resulting mixture was
stirred at ambient temperature for 15 h and was quen-
ched by the addition of 20 mL of saturated NH4Cl. The
products were extracted with CH2Cl2 (3 ꢂ 10 mL) and
the combined organic extracts were washed with brine,
dried over Na2SO4, and concentrated. The products
were purified via chromatography on SiO2 eluted with
10% ether/CH2Cl2 to give 0.37 g (70%) of recovered 6
and 0.52 g (13%) of 2 as a yellow powder, mp 260–
261 °C: 1H NMR (500 MHz, CD2Cl2) d 8.33 (d, 2H,
J ¼ 9:2 Hz), 6.80 (d ꢂ d, 2H, J ¼ 2:1, 9.2 Hz), 6.77 (d,
2H, J ¼ 2:1 Hz), 3.07 (s, 12H); 13C NMR (CD2Cl2)
177.2; 151.7, 138.6, 130.2, 118.5, 110.9, 104.8, 39.6; IR
(KBr) 1592 cmꢁ1; HRMS (ES) m=z 299.1217 (calcd for
C17H18ON2 S þ H: 299.1213).
The incorporation of selenium into the xanthylium core
is clearly shown in Figure 3 where the parent ion cluster
for TMR-Se is consistent with the theoretical spectrum
for C23H23N2Se.
4.1.5. Preparation of 2,7-bis-N,N-dimethylamino-9-phen-
ylthioxanthylium chloride (TMR-S). A solution of 2,7-
4.1.4. Preparation of 2,7-bis-N,N-dimethylamino-9-phen-
ylselenoxanthylium chloride (TMR-Se). A solution of
2,7-bis-N,N-dimethylamino-9H-selenoxanthen-9-one (1,
0.070 g, 0.20 mmol) in THF (3 mL) was added dropwise
to a solution of phenylmagnesium bromide (1.0 mL of a
1.0 M solution in THF, 1.0 mmol) in THF (2 mL) heated
to reflux. After addition was complete, the resulting
solution was heated at reflux for an additional 0.5 h. The
reaction mixture was then cooled to ambient tempera-
ture and poured into acetic acid (3.0 mL). Hexafluoro-
phosphoric acid (60% weight solution in water) was
added dropwise until the initial deep blue solution
turned reddish yellow. Water (50 mL) was added and the
resulting mixture was cooled at ꢁ10 °C precipitating the
selenoxanthylium hexafluorophosphate salt. The dye
was collected by filtration and the solid was washed with
water (2 ꢂ 5 mL) and diethyl ether (2 ꢂ 5 mL). The crude
bis-N,N-dimethylamino-9H-thioxanthen-9-one
(2,
0.050 g, 0.17 mmol) in THF (3 mL) was treated with
phenylmagnesium bromide (1.0 mL of a 1.0 M solution
in THF, 1.0 mmol) in THF (2 mL) as described for
TMR-Se. Workup and recrystallization gave 0.078 g
(71%) of 2,7-bis-N,N-dimethylamino-9-phenylthioxan-
1
thylium hexafluorophosphate: mp >260 °C; H NMR
(500 MHz, CD2Cl2) d 7.68 (m, 3H), 7.47 (d, 2H,
J ¼ 9:8 Hz), 7.37 (m, 2H), 7.14 (d, 2H, J ¼ 2:4 Hz), 6.95
(d ꢂ d, 2H, J ¼ 2:4, 9.8 Hz), 3.31 (s, 12H); 13C NMR
(500 MHz, CD2Cl2) d 153.5, 144.4, 136.5, 135.4, 129.5,
129.2, 128.7, 119.2, 115.1, 114.2, 105.4, 40.5; kmax
(CH2Cl2) 571 nm (e ¼ 5:0 ꢂ 104 Mꢁ1 cmꢁ1); HRMS
(ESI) m=z 359.1580 (calcd for C23H23N2S: 359.1582).
Anal. Calcd for C23H23F6N2OPS: C, 54.76; H, 4.60; N,
5.55. Found: C, 54.75; H, 4.75; N, 5.24.

2542
M. R. Detty et al. / Bioorg. Med. Chem. 12 (2004) 2537–2544
(Millenia X, Spectra-Physics) at 532 nm was the excita-
tion source. The sample solution in a quartz cuvette was
placed directly in front of the entrance slit of the spec-
trometer and the emission signal was collected at 90°
relative to the exciting laser beam. An additional long-
pass filter (850LP) was used to attenuate the excitation
laser and the fluorescence from the photosensitizer.
4.3. Fluorescence quantum yields and radiative lifetimes
Fluorescence quantum yields (/F) and radiative life-
times (srad) were determined using techniques and
equipment that we have previously described.25
4.4. Electrochemical procedures
A BAS 100 potentiostat/galvanostat and programmer
were used for the electrochemical measurements. The
working electrode for cyclic voltammetry was a plati-
num disk electrode (1 mm diameter) obtained from
Princeton Applied Research. The auxiliary and refer-
ence electrodes were silver wires. The reference for cyclic
voltammetry was the ferrocene/ferrrocinium couple at
þ0.40 V at a scan rate of 0.1 V sꢁ1. All samples were run
in J. T. Baker HPLC-grade dichloromethane that had
ꢀ
been stored over 3 A molecular sieves and freshly dis-
tilled prior to use. Tetrabutylammonium fluoroborate
(Sigma–Aldrich) was recrystallized from EtOAc–ether
and then dried overnight at 80 °C under vacuum before
it was used as supporting electrolyte at 0.2 M. Nitrogen
was used for sample deaeration.
Figure 3. A comparison of the experimental mass spectrum of TMR-
Se (top) with the theoretical mass spectrum of a molecule with the
formula C23H23N2Se (bottom).
4.5. Determination of partition co-efficients
The octanol/water partition co-efficients were all mea-
sured at pH 6 (phosphate-buffered) using UV–vis spec-
trophotometry. The measurements were done using a
‘shake flask’ direct measurement.26 Mixing for 3–5 min
was followed by 1 h of settling time. Equilibration and
measurements were made at 23 °C using a Perkin–Elmer
Lambda 12 spectrophotometer. HPLC grade 1-octanol
was obtained from Sigma–Aldrich.
The hexafluorophosphate salt (0.045 g, 0.068 mmol) was
treated with Amberlite IRA-400 chloride ion exchange
resin as described for the preparation of TMR-Se to give
0.026 g (78%) of 2,7-bis-N,N-dimethylamino-9-phenyl-
thioxanthylium chloride (TMR-S) as a dark green,
crystalline solid: mp >260 °C; 1H NMR (500 MHz,
CD2Cl2) d 7.67 (m, 3H), 7.46 (d, 2H, J ¼ 9:1 Hz), 7.36
(m, 2H), 7.14 (d, 2H, J ¼ 2:5 Hz), 6.95 (d ꢂ d, 2H,
J ¼ 2:5, 9.1 Hz), 3.31 (s, 12H); kmax (H2O) 570 nm
(e ¼ 4:0 ꢂ 104 Mꢁ1 cmꢁ1); HRMS (ESI), m=z 359.1579
(calcd for C23H23N2S: 359.1582). Anal. Calcd for
C23H23ClN2OS: C, 69.94; H, 5.87; N, 7.07. Found: C,
70.03; H, 5.75; N, 7.05.
4.6. Cells and culture conditions
R3230AC, a rat mammary adenocarcinoma line,27;28
was maintained in a growth medium of minimum
essential media (a-MEM) supplemented with 10%
bovine serum albumin (FBS, Atlanta Biologicals,
Atlanta, GA), 50 units/mL of penicillin G, 50 mg/mL of
streptomycin, and 1.0 mg/mL of Fungizone.
4.2. Quantum yield determinations for the generation of
singlet oxygen
The quantum yields for singlet-oxygen generation with
chalcogenoxanthylium dyes TMR-O, TMR-S, TMR-Se,
and 6-Se were measured by direct methods in MeOH.24
A SPEX 270M spectrometer (Jobin Yvon) equipped
with InGaAs photodetector (Electro-Optical Systems
Inc., USA) was used for recording singlet-oxygen
emission spectra. A diode pumped solid-state laser
4.7. Irradiation of cultured R3230AC cells
After seeding on 12-well plates, R3230AC cells were
incubated for 24 h to allow cells to attach to the surface.
Stock solutions of TMR-O, TMR-S, or TMR-Se, pre-
pared at 1 ꢂ 10ꢁ3 M, were then added directly to the cell

M. R. Detty et al. / Bioorg. Med. Chem. 12 (2004) 2537–2544
2543
culture medium to give a final dye concentration of
0.01–10 lM. Cell monolayers were then incubated for
3 h at 37 °C in the dark in a humidified 5% CO2 atmo-
sphere. The media was then removed and 1.0 mL of
a-MEM minus FBS and phenol red (clear medium) was
added to each well. One plate, with the lid removed, was
then exposed to 360–800 nm light delivered at
1.4 mW cmꢁ2 for 1 h (5.0 J cmꢁ2) from a filtered tungsten
source for dye concentrations 60.1 lM. The remaining
plate was kept in the dark during the irradiation period.
Immediately following irradiation the clear medium was
replaced with complete medium and the monolayers
were incubated for an additional 24 h period as above.
Subsequently, cells were trypsinized and counted to
determine cell viability.
optical fiber of core diameter 1 mm, and was delivered to
a spectrometer (Holospec from Kaiser Optical Systems,
Inc.) equipped with a cooled charge coupled device
(CCD) camera (Princeton Instruments) as detector. A
comparison of the fluorescence spectra from cells and
the fluorescence spectra of the photosensitizers con-
firmed that the origin of the fluorescence observed in the
image channel was from the photosensitizer. Control
cells showed no fluorescence under the same imaging
conditions in the absence of photosensitizer.
4.10. Statistical analysis
Pairwise inter-comparisons among the experimental
groups and comparisons with the controls were per-
formed using the students’ t test. A value of P < 0:05 is
considered significant.
4.8. Measurement of dye uptake into cell monolayers
Twenty-four hours after seeding R3230AC cells on 12-
well plates, TMR-Se, TMR-S, or TMR-O were added at
10 lM in a-MEM plus FBS and phenol red. Cells were
incubated for 3 h in the presence of each dye in a
humidified atmosphere in the dark as described above.
The medium was then removed and the monolayers
washed once with 1.0 mL of 0.9% NaCl and 1.0 mL of
50% EtOH/1% glacial acetic acid was added. Cells
detached from the surface within 5 min and the dye
emission in the resulting cell lysates was determined.
Intracellular dye concentration was calculated by com-
paring emission values obtained from cell lysates with
standard curves generated from known concentrations
of dyes dissolved in a mixture of 50% EtOH/1% glacial
acetic acid. Intracellular dye concentration is expressed
in femtomole dye per cell.
Acknowledgements
This work was done under research grants to M.R.D.
from the NIH (Grant CA69155) and to P.N.P. from the
Directorate of Chemistry and Life Sciences of the Air
Force Office of Scientific Research through Defense
University Research Initiatives on Nanotechnology
(Grant F496200110358).
References and notes
1. Peterson, O. G.; Tuccio, S. A.; Snavely, B. B. Appl. Phys.
Lett. 1970, 17, 245.
2. Ali, M.; Moghaddasi, J.; Ahmed, S. A. J. Appl. Opt. 1990,
29, 3945.
3. Karstens, T.; Kobs, K. J. Phys. Chem. 1980, 84, 1871.
4. Preininger, C.; Mohr, G. J.; Klimant, I.; Wolfbeis, O. S.
Anal. Chim. Acta 1996, 334, 113.
5. Liu, J.; Diwu, Z.; Leung, W.-Y.; Lu, Y.; Patch, B.;
Haugland, R. P. Tetrahedron Lett. 2003, 44, 43555.
6. Johnson, L. V.; Walsh, M. L.; Bockus, B. J.; Chen, L. B.
J. Cell Biol. 1981, 88, 526.
7. Shea, C. R.; Sherwood, M. E.; Flotte, T. J.; Chen, N.;
Scholz, M.; Hasan, T. Cancer Res. 1990, 50, 4167.
8. Summerhayes, I. C.; Lampidis, T. J.; Bernal, S. D.;
Nadakavukaren, J. J.; Nadakavukaren, K. K.; Shepard,
E. L.; Chen, L. B. Proc. Natl. Acad. Sci. U.S.A. 1982, 79,
5292.
4.9. Confocal laser scanning microscopy studies
R3230AC cells in 60-mm culture plates were treated
with 5 mL of MEM alpha medium (without phenol red)
containing 2 lM TMR-Se. The plates were returned to
the incubator (37 °C, 5% CO2) for a duration of 2 h.
After incubation, the plates were rinsed with sterile PBS,
and fresh media (without phenol red) was replaced at a
volume of 5 mL/plate. The plates were then observed
directly under the confocal microscope.
Confocal laser scanning microscopy images were
obtained using a commercial confocal laser scanning
microscope system (CLSM), model MRC-1024 (Bio-
Rad, Richmond, CA), which was attached onto an
upright microscope (Nikon, model Eclipse E800). A
water immersion objective lens (Nikon, Fluor-60X,
NA ¼ 1.0) was used for cell imaging. A solid-state diode-
pumped laser (Verdi, Coherent) was used as a source of
excitation (532 nm). An additional long-pass filter
(585LP) was introduced in front of the photomultiplier
to minimize leakage of the excitation light. To confirm
that the observed fluorescence was from photosensitizer
permeated to cells, we used localized spectrofluorome-
try. The fluorescence signal was collected from the upper
port of the confocal microscope using a multi-mode
9. Davis, S.; Weiss, M. J.; Wong, J. R.; Lampidis, T. J.;
Chen, L. B. J. Biol. Chem. 1985, 260, 13844.
10. Lampidis, T. J.; Bernal, S. D.; Summerhayes, I. C.; Chen,
L. B. Ann. N.Y. Acad. Sci. 1982, 395, 299.
11. Bernal, S. D.; Lampidis, T. J.; McIsaac, R. M.; Chen, L.
B. Science 1986, 222, 169.
12. Dougherty, T. J.; Gomer, C. J.; Henderson, B. W.; Jori,
G.; Kessel, D.; Korbelik, M.; Moan, J.; Peng, Q. J. Natl.
Cancer Inst. 1998, 90, 889.
13. Shea, C. R.; Chen, N.; Wimberly, J.; Hasan, T. Cancer
Res. 1989, 49, 3961.
14. Menzel, R.; Thiel, E. Chem. Phys. Lett. 1998, 291, 237.
15. Kessel, D.; Woodburn, K. Br. J. Cancer 1995, 71, 306–
310.
16. Pal, P.; Zeng, H.; Durocher, G.; Girard, D.; Li, T. C.;
Gupta, A. K.; Giasson, R.; Blanchard, L.; Gaboury, L.;

2544
M. R. Detty et al. / Bioorg. Med. Chem. 12 (2004) 2537–2544
Balassy, A.; Turmel, C.; Laperriere, A.; Villeneuve, L.
Photochem. Photobiol. 1996, 63, 161.
23. You, Y.; Ahsan, K.; Detty, M. R. J. Amer. Chem. Soc.
2003, 125, 4918.
17. Villeneuve, L.; Pal, P.; Durocher, G.; Migneault, D.;
Girard, D.; Giasson, R.; Balassy, A.; Blanchard, L.;
Gaboury, L. J. Fluoresc. 1996, 6, 209.
18. Stracke, F.; Heupel, M.; Thiel, E. J. Photochem. Photobiol.
A 1999, 126, 51.
24. Pandey, R. K.; Sumlin, A. B.; Constantine, S.; Aoudia,
M.; Potter, W. R.; Bellnier, D. A.; Henderson, B. W.;
Rodgers, M. A.; Smith, K. M.; Dougherty, T. J. Photo-
chem. Photobiol. 1996, 64, 194.
25. Baker, G. A.; Bright, F. V.; Detty, M. R.; Pandey, S.;
Stilts, C. E.; Yao, H. R. J. Porphyrins Phthalocyanines
2000, 4, 669.
19. Brennan, N. K.; Donnelly, D. J.; Detty, M. R. J. Org.
Chem. 2003, 68, 3344.
20. Detty, M. R.; Merkel, P. B. J. Amer. Chem. Soc. 1990,
112, 3845.
21. Sun, X.; Wong, J. R.; Song, K.; Hu, J.; Garlid, K. D.;
Chen, L. B. Cancer Res. 1994, 54, 1465.
22. Brennan, N. K.; Hall, J. P.; Davies, S. R.; Gollnick, S. O.;
Oseroff, A. R.; Gibson, S. L.; Hilf, R.; Detty, M. R.
J. Med. Chem. 2002, 45, 5123.
26. Sangster, J. In Octanol–Water Partition Co-efficients:
Fundamentals and Physical Chemistry; Fogg, P. G. T.,
Ed.; John Wiley and Sons: New York, 1997.
27. Hilf, R.; Michel, I.; Bell, C.; Freeman, J. J.; Borman, A.
Cancer Res. 1965, 25, 286.
28. Hissin, P. J.; Hilf, R. Cancer Res. 1978, 38,
3646.