Type 1 phototherapeutic agents, part I: Preparation and cancer cell viability studies of novel photolabile sulfenamides

Article abstract of DOI:10.1021/ml2001483

Novel type 1 phototherapeutic agents based on compounds containing S"N bonds (sulfenamides) were synthesized, assessed for free radical generation, and evaluated in vitro for cell death efficacy in four cancer cell lines (U937, HTC11, KB, and HT29). All o

Full text of DOI:10.1021/ml2001483

LETTER
pubs.acs.org/acsmedchemlett
Type 1 Phototherapeutic Agents, Part I: Preparation and Cancer Cell
Viability Studies of Novel Photolabile Sulfenamides
Amolkumar S. Karwa,† Amruta R. Poreddy,† Bethel Asmelash,† Tien-Sung Lin,‡ Richard B. Dorshow,†
and Raghavan Rajagopalan*,†
†Covidien Pharmaceuticals, 675 McDonnell Boulevard, Hazelwood, Missouri 63042, United States
‡Department of Chemistry, Washington University, 1 Brookings Drive, St. Louis, Missouri 63130, United States
S Supporting Information
b
ABSTRACT: Novel type 1 phototherapeutic agents based on compounds containing
SÀN bonds (sulfenamides) were synthesized, assessed for free radical generation, and
evaluated in vitro for cell death efficacy in four cancer cell lines (U937, HTC11, KB,
and HT29). All of the compounds were found to produce copious free radicals upon
photoexcitation with UV-A and/or UV-B light, as determined by electron spin
resonance spectroscopy. Among the sulfenamides, the most potent compounds were
derived from dibenzazepine 7b and dihydroacridine 8b as determined in all of the four cancer cell lines.
KEYWORDS: Sulfenamides, ROS, type 1, type 2, photosensitizers, radicals, ESR
hototherapeutic agents are classified according to the reac-
tion pathway involved in the cell death process. In general,
nor is it restricted by wavelength of light. Hence, type 1
photosensitizers could be useful for ablation of lesions under
hypoxic conditions. In the type 1 process, the photosensitzer
undergoes irreversible change, and the delivery of an efficaci-
ous amount of type 1 agents to a lesion could be a limiting factor.
However, the advent of newly developed drug delivery systems,
especially the many varieties of nanoparticles, should allow efficaci-
ous delivery of these agents with increased safety margins.15 Thus,
we have implemented a research effort on developing type 1
phototherapeutic agents, and in this letter, we present a prelimin-
ary study on the synthesis and in vitro evaluation of type 1 com-
pounds containing photolabile SÀN bonds16 1bÀ14b (Chart 1).
The sulfenamides 1b, 2b, and 6bÀ10b were prepared in one
step by the reaction of 4-nitrobenzenesulfenyl chloride with
corresponding commercially available arylamines 1a, 2a, and
6aÀ10a (Scheme 1). Compounds 3bÀ5b and 11bÀ14b were
prepared similarly from the corresponding known tricyclic
amines 3aÀ5a and 11aÀ14a.17,18 The reaction of sulfenyl chlo-
rides with diarylamines is known to generate diarlyl sulfide B
(Scheme 1) resulting from C-sulfenylation of the phenyl ring,
but the electronic factors controlling the product distribution of
A and B has not been fully understood.19 For example, the
diarylamines 1aÀ5a, 7a, and 14a gave N-sulfenylated products A
exclusively, whereas 6a and 12a gave the C-sulfenylated isomers
B exclusively. All other amines gave variable mixtures of A and
B. It may also be noted that the preparation of sulfenamides
without the nitro group or with an electron-donating group on
the thiophenyl ring was not successful, which is consistent with
the observation reported earlier by others for the preparation of
sulfenates.20
P
photosensitizers operate via two main mechanisms, type 1 and
type 2,1À3 and type 2 photosensitzers have been used extensively
for the treatment of various lesions.4À9 For example, commercial
photosensitizers such as Visudyne used for the treatment of wet
macular degeneration and Photofrin used for the treatment of
certain types of cancers, are porphyrin-based type 2 agents. The
vast majority of current research is focused on improving the
efficacy of such porphyrin-based type 2 agents. Surprisingly, very
little attention has been devoted to type 1 agents despite the fact
that there are numerous classes of compounds suitable for the
development of such photosensitizers.
In type 2 process (referred to as photodynamic therapy
or “PDT”), the energy of the excited photosensitizer is trans-
ferred to nearby molecular oxygen, creating singlet oxygen and
subsequent reactive oxygen species (ROS) such as hydroxyl
radical, superoxide radical anion, etc. that are responsible for the
cell death.10À12 This process may be repeated continuously
because a substantial proportion of the photosensitizer returns
to the ground state without photodegradation, but the depletion
of oxygen level (local hypoxia) induces inflammatory response
that causes vasodilatation and enables the transport of cancer
cells to other regions resulting in tumor metastases.13 However,
the risk of metastases can be mitigated by using combination
therapy, where an anti-VEGF (vascular endothelial growth
factor) agent, such as Avastin, is coadministered with the
photosenstizer.14
On the other hand, in type 1 process, the absorption of light
causes the photosensitizer itself to undergo bond fragmentation
to generate reactive intermediates such as free radicals, which
induces the cell death. Whereas the PDT is mediated by singlet
oxygen and requires red light for optimal generation of singlet
oxygen, the type 1 process does not require oxygen for activity,
Received: June 20, 2011
Accepted: September 13, 2011
Published: September 13, 2011
r
2011 American Chemical Society
828
dx.doi.org/10.1021/ml2001483 ACS Med. Chem. Lett. 2011, 2, 828–833
|

ACS Medicinal Chemistry Letters
LETTER
Chart 1. Structures of Sulfenamides
Figure 1. ESR spectra of sulfenamides 1b and 7b in benzene with
DMPO spin trapping agent: (a) compound 1b during light exposure,
(b) compound 1b after irradiation in light-off period, (c) compound 7b
during light exposure, and (d) compound 7b during light-off period. The
fast-decaying spin adducts DMPO/ArS• become much weaker, and the
weak signals of spectra b and d are assigned to the nitrogen-centered
DMPO/Ar2N• spin adducts.
Scheme 1
The decay half-life (t1/2) of DMPO/ArS• spin adducts from 1b and
7b is about 10 s. On the other hand, the nitrogen-centered DMPO
spin adducts (DMPO/Ar2N•) (Figure 1b,d) are substantially differ-
ent because the aromatic portion of the N-centered spin adducts are
different and may account for the differences in the cell death
properties. The typical parameters for DMPO/Ar2N• spin adducts
are g= 2.0050; aNα =1.77mT;aHβ =1.77mT;andaNβ =0.26mTfor
compound 1b. Similar parameters are obtained for compounds 7b
except that more proton splittings (aHγ) are observed due to
extended conjugation of the π system.
The values for aNβ and aHγ parameters varied slightly among
compounds depending on the delocalization of the unpaired
electron in the aromatic π system. Note that these parameters
refer to the spin adducts and not the original nascent radical upon
irradiation. Some of these parameters are similar to those of
reported values.22À25 We further observed that the N-centered
radicals form molecular complexes with oxygen in the ESR
spectra of nondegassed samples (unpublished results). The
oxygen complex of nitrogen-centered radical may further under-
go electron transfer to form superoxide anion radicals (•O2À),
which is detrimental to cell survival.
All of the sulfenamides generated copious free radicals upon
photofragmentation of SÀN bond as evidenced from the elec-
tron spin resonance (ESR) measurements. ESR spectroscopy is
a powerful and widely used technique for the detection and
characterization of chemical species with unpaired electrons
(e.g., free radicals). Representative ESR spectra corresponding
to the photolyzed sulfenamides 1b and 7b are shown in Figure 1.
Because most of the free radicals are short-lived, a spin trapping
agent, 5,5-dimethyl-1-pyrroline-1-oxide (DMPO), was used to
trap radical species in the ESR studies. The DMPO spin adducts
are relatively stable nitroxides with unique ESR spin parameters
and spectral patterns depending on the type of free radicals added
to the carbon atom at the 2-position (i.e., the β-carbon) in
DMPO.21 All of the compounds generated the same fast-decay-
ing sulfur-centered DMPO spin adducts (DMPO/ArS•) upon
photolysis as evidenced by the same paramagnetic parameters and
decay characteristics (Figure 1a,c). The paramagnetic parameters of
all DMPO/ArS• spin adducts are g = 2.0052; aN = aHβ = 1.32 mT.
829
dx.doi.org/10.1021/ml2001483 |ACS Med. Chem. Lett. 2011, 2, 828–833

ACS Medicinal Chemistry Letters
LETTER
The viability of cancer cells exposed to the sulfenamides in the
absence and presence of light was assessed by the standard WST-
1 assay26 using U937 leukemia cell lines. For the two most active
sulfenamides 7b and 8b, three additional cell lines (HTC11, KB,
and HT29) were also used. All of the compounds were dissolved
in DMSO at an initial concentration of about 8 mM and were
diluted with cell culture media such that amount of DMSO
exposed to the cells were below 0.5% (64 mM). The cells were
incubated with various concentrations of the photosensitizer for
2 h prior to light exposure. The corresponding control conditions
were (a) DMSO only (no photosensitizer), no light; (b) DMSO
only (no photosensitizer) light; and (c) photosensitizer, no light
(dark toxicity). The cells were irradiated for 20 min with 100 W
UV lamp (model B-100A, UVP, Upland, CA) with peak output at
365 nm and a bandwidth of about 50 nm. The light source was
not optimized for each compound with respect to power and
wavelength. The total power near the surface of the microtiter
plate is about 8 mW. The temperature at the surface of the
microtiter plates was kept below 37 °C with external air cooling.
The viability of cells was assessed after 24 h following light
exposure.
The λmax and IC50 values (defined as the concentration at
which 50% decrease in cell viability is observed when the cells are
exposed to light and the photosensitizer for 20 min) are given in
Table 1. The absorption maxima for all of the sulfenamides fall in
the range of 305À420 nm. The sulfenamides can be modified by
fusing additional aromatic rings into their structures for extended
conjugation to enable them to be activated with visible light. All
of the sulfenamides except 6b displayed dose-dependent and
light exposure time-dependent decrease in cell viability. It should
be noted that DMSO itself exhibited cytotoxicity only at high
concentration and at long exposure time to light. In fact, the cell
viability of DMSO was reduced by only about 15% at 100 mM
concentration when irradiated for 20 min. The concentration
of DMSO in the viability studies is less than 64 mM. Exemplary
cell viability doseÀresponse graphs for the sulfenamides 7b and
8b in four cancer cell lines (HTC116, HT29, KB, and U937) are
shown in Figures 2 and 3.
Table 1. Absorption and Cell Viability (IC50) Data
photosensitizer
λmax (nm)
IC50 (μM)a
none (DMSO)
>100000
6.8 ( 1.3
>20
1b
307
325
325
341
325
350
321
321
327
345
335
420
2b
3b
7.3 ( 1.1
6.0 ( 1.2
5.7 ( 1.2
1.2 ( 1.1
1.3 ( 1.2
12.4 ( 1.2
>20
4b
5b
7b
8b
The parent sulfenamide 1b exhibits moderate activity with an
IC50 of about 9 μM. Connecting the two phenyl groups with a
single bond essentially abrogates the activity. Insertion of a
methylene or an ethylene bridge between the nitrogen and the
phenyl ring (compounds 3b and 4b, respectively) or insertion of
one methylene between the nitrogen and the phenyl ring and the
other between the two phenyl rings (compound 5b) restores the
9b
10b
11b
13b
14b
17.2 ( 1.1
2.0 ( 1.1
>20
a Average of at least three independent runs.
Figure 2. Phototoxicity of 7b. Blue, no light exposure (dark toxicity); red, 20 min of light exposure (phototoxicity).
830
dx.doi.org/10.1021/ml2001483 |ACS Med. Chem. Lett. 2011, 2, 828–833

ACS Medicinal Chemistry Letters
LETTER
Figure 3. Phototoxicity of 8b. Blue, no light exposure (dark toxicity); red, 20 min of light exposure (phototoxicity).
activity. The activity of compound 6a, which contains an ethylene
bridge between the two phenyl groups, could not be determined
due to the exclusive formation of the C-sulfenylated thioether
product. In contrast, inserting a vinylene bridge between the two
phenyl rings in 1 (compound 7b) or transposing one of the
phenyl groups in 2 from an angular to a linear tricyclic arrange-
ment (compound 8b) greatly increases the activity. Indeed, the
dibenzazepine 7b and dihydroacridine 8b are the most potent
among all of the compounds with IC50 of about 1.3 and 1.2 μM,
respectively. Replacing the methylene bridge in 8b with oxygen
or sulfur atom substantially reduces or even obviates the activity
(compounds 9b and 10b). Likewise, replacing a methylene unit
in 5b or 6b (compounds 11b and 12b) also eliminated the
activity. Introduction of imino bridge between the two phenyl
groups in 1 resulted in a potent compound 13b, but surprisingly,
introduction of an azo bridge (compound 14b) completely
abrogated the activity.
The difference in activity between the sulfenamides 2b and
7b could be rationalized on the basis of transition states leading
to the formation of reactive species as illustrated in Scheme 2. If
the transition state is dipolar in nature or if homolytic cleavage
of SÀN bond followed by electron transfer from nitrogen to
sulfur leads to ionic species such as 15 and 16, then the
transition state for the photofragmentation of compound 2b
should resemble cyclopentadienyl cation 15, which is antiaro-
matic, whereas for 7b the transition state should resemble
azatropylium cation27 16, which is aromatic. The high resolu-
tion mass spectrum of 7b clearly indicates the formation of
two fragments: the azatropylium cation (exact mass, 193.0886)
and the p-nitrophenylsulfide anion (exact mass, 153.9967).
Furthermore, cyclic voltammetry studies with 7a indicates
that the formation of cation 16 from the radical 17 is a highly
favored process with E1/2 of 0.88 V.28 Thus, it could be
expected that the formation of reactive intermediate from 2b
would be considerably suppressed as compared to 7b under
the same photolytic conditions. Although the contribution, if
any, of p-nitrophenylsulfide anion on cytotoxicity is not deter-
mined, it should be the same for all of the sulfenamides
and should not affect the relative cell viabilities among these
photosensitzers.
The high activities of 7b and 8b could be explained on the
basis of two different pathways as illustrated in Scheme 2. In the
case of 7b, the N-centered radical 17 could also react with
molecular oxygen to produce, via electron transfer, an ion pair
18, where superoxide anion radical is known to be detrimental
to cell viability. In the photolysis of 8b, the dihydroacridine
radical 19 could itself cause cell death directly or react with
oxygen to form a stable nitroxide radical 20,29 which could
cause cell death. Alternatively, the radicals 19 or 20 could
undergo oxidation to the nitrone 21, which may form an ion
pair with superoxide anion radical (•O2À). Indeed, the ESR
spectrum of 8b as well as previous ESR studies with phenox-
azines and phenothiazines24,25 supports the formation of such
nitroxide radicals.
In this preliminary communication, we have demonstrated
that incorporation of SÀN moiety into the photosensitizer
structure is a viable approach for the design of suitable type 1
phototherapeutic agents. The results clearly support the hypoth-
esis that the cell death occurs through initial formation of
nitrogen-centered radicals. Whether these radicals cause cell
death directly or through secondary ROS as well as the precise
mechanism of action can only be ascertained with additional
ESR and ultrafast spectroscopic studies and rigorous theoretical
831
dx.doi.org/10.1021/ml2001483 |ACS Med. Chem. Lett. 2011, 2, 828–833

ACS Medicinal Chemistry Letters
LETTER
Scheme 2
’ REFERENCES
(1) Grossweiner, J. B.; Rogers, G. B. H. The Science of Phototherapy:
An Introduction; Jones, L. R., Ed.; Springer: Norwell, 2010; pp 93À195.
(2) Moore, A. C. E.; Ortel, B.; Hasan, T. Mechanisms of Photo-
dynamic Therapy. In Photodynamic Therapy; Patrice, T., Ed.; RSC:
Cambridge, 2003; pp 19À57.
(3) O'Connor, A. E.; Gallagher, W. M.; Byrne, A. T. Porphyrin and
Non-porphyrin Photosensitizes in Oncology: Preclinical and Clinical
Advances in Photodynamic Therapy. Photochem. Photobiol. 2009, 85,
1053–1074.
(4) Huang, A. A Review of progress in Clinical Photodynamic
Therapy. Technol. Cancer Res. Treat. 2005, 4, 283–293.
(5) Bessler, N. M. VAM Study Writing Committee. Verteporphin
Therapy in Age-related Macular Degeneration (VAM). Retina 2004,
24, 512–520.
(6) van Lier, J. E.; Tian, H.; Ali, H.; Cauchon, N.; Hassessian, H. M.
Trisulfonated porphyrazines: New Photosensitizers for the Treatment
of Retinal and Subretinal Edema. J. Med. Chem. 2009, 52, 4107–4110.
(7) Mitton, D.; Ackroyd, R. Photodynamic therapy of Barrett’s
oesophagus and oesophageal carcinoma—How I do it. Photodiag.
Photodyn. Ther. 2006, 3, 9698.
(8) Li, L.; Luo, R.; Liao, W.; Zhang, M.; Luo, Y.; Miao, J.
Clinical study of photofrin photodynamic therapy for the treatment of
relapse nasopharyngeal carcinoma. Photodiagn. Photodyn. Ther. 2006, 3,
266–271.
(9) Hahn, S. M.; Putt, M. E.; Metz, J.; Shin, D. B.; Rickter, E.; Menon,
C.; Smith, D.; Glatstein, E.; Fraker, D. L.; Busch, T. M. Photofrin uptake
in the tumor and normal tissues of patients receiving intraperitoneal
photodynamic therapy. Clin. Cancer Res. 2006, 12, 5464–5470.
(10) Kehrer, J. P. Free Radicals as Mediators of Tissue Injury and
Disease. Crit. Rev. Toxicol. 1993, 23, 21.
(11) Dugan, L. L.; Lin, T.-S.; He, Y. Y.; Hsu, C. Y.; Choi, D.
Detection of Free Radicals by Microdialysis /Spin Trapping EPR
Following Focal Cerebral Ischemia Reperfusion and a Cautionary Note
on the Stability of 5,5-Dimethyl-1-pyrroline N-Oxide (DMPO). Free
Radical Res. 1995, 23, 27–32.
(12) Wardman, P.; Dennis, M. F.; Everett, S.A.; Patel, K. B.;
Stratford, M. R. L.; Tracy, M. Radicals from One-electron Reduction
of Nitro Compounds, Aromatic N-Oxides and Quinones: The kinetic
Basis for Hypoxiaselective, Bioreductive Drugs. Biochem. Soc. Symp.
1995, 6 (1), 171–194.
(13) Momma, T.; Hamblin, M. R.; Wu, H. C.; Hasan, T. Photo-
dynamic Therapy of Orthotopic Prostate Cancer with Benzoporphyrin
Derivative: Local Control and Distal Metastasis. Cancer Res. 1998,
58, 5425–5431.
calculations. These are currently being investigated and will be
reported subsequently.
(14) Bhuvaneswari, R.; Yuen, G. Y.; Chee, S. K.; Olivio, M. Hyper-
icin-mediated Photodynamic Therapy in Combination with Avastin
(Bevacizumab) Improves Tumor Response by Downregulating Angio-
genic Proteins. Photochem. Photobiol. Sci. 2007, 6, 1275–1283.
(15) Gyenge, E. B.; Hiestand, S.; Graefe, S.; Walt, H.; Maake, C.
Cellular and Molecular Effects of the Liposomal mTHPC Deivative
Foslipos in Prostate Carcinoma Cells In Vitro. Photodiagn. Photodyn.
Ther. 2011, 8, 86–96.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures for the
b
synthesis of compounds 1bÀ14b and cell viability assays. This material
is available free of charge via the Internet at http://pubs.acs.org.
(16) Davis, F. A. Adventures in Sulfur-Nitrogen Chemistry. J. Org.
Chem. 2006, 71, 8993–2003.
’ AUTHOR INFORMATION
(17) Hunziker, von F.; Kunzle, F.; Schindler, O.; Schmutz, J.
Dibenzo-Azepine, -Diazepine, -Oxazepine, and -Thiazepine. Helv. Chim.
Acta 1964, 47, 1163–1172.
(18) Yang, M.; Wu, L.; She, D.; Hui, H.; Zhao, Q.; Chen, M.; Huang,
G.; Liang, Y. Synthesis of Dibenzo-fused Seven-membered Heterocycles
via Copper-catalyzed Cyclization of 2-Haloaniline Compounds. Synth-
esis 2008, 448–452.
(19) Craine, L.; Raban, M. The Chemistry of Sulfenamides. Chem.
Rev. 1989, 89, 689–712.
(20) Guo, G.; Jenks, W. S. Photochemistry and Photophysics of
Aromatic Sulfoxides: α-Cleavage and the Role of the Sulfenic Ester.
J. Org. Chem. 1995, 60, 5480–5486.
Corresponding Author
*Tel: 314-654-3800. Fax: 314-654-8900. E-mail: Raghavan.
rajagopalan@covidien.com.
’ ACKNOWLEDGMENT
We thank Drs. James Kostelc, Jigyue Yang, and Bich Vu for the
mass spectral analysis of some of the compounds. We also thank
Drs. Gary Cantrell, J. J. Shieh, David Pipes, Mary Dyszlewski, and
Prof. Richard Loomis for insightful discussions.
832
dx.doi.org/10.1021/ml2001483 |ACS Med. Chem. Lett. 2011, 2, 828–833

ACS Medicinal Chemistry Letters
LETTER
(21) Janzen, E. G. Radical Addition Reactions of 5,5-Dimethyl-l-
pyrroline-l-oxide. ESR Spin Trapping with a Cyclic Nitrone. J. Magn.
Reson. 1973, 9, 510–512.
(22) Migita, C. T.; Migita, K. Spin Trapping of the Nitrogen Radicals.
Chem. Lett. 2003, 32, 466–467.
(23) Mile, B.; Rowlands, C. C.; Sillmana, P.D.; Fildes, M. The EPR
Spectra of Thiyl Radical Spin Adducts Produced by Photolysis of
Disulfides in the Presence of 2,4,6-Tri-tert-butylnitrosobenzene and
5,5-Dimethyl-1-pyrroline N-Oxide. J. Chem. Soc., Perkin Trans. 1992,
2, 1431–1437.
(24) Lin, T.-S.; Retsky, J. ESR Studies of Photochemical Reactions of
Diphenylamines, Phenothiazines and Phenoxazines. J. Phys. Chem. B
1986, 90, 2687–2689.
(25) Gilbert, B. C.; Chiu, M. F.; Hanson, P. Electron Spin Resonance
of Some Heterocyclic Radicals Containing Elements of Group VI.
J. Chem. Soc [Sec. B] Phys. Org. 1970, 9, 1700–1708.
(26) Ishiyama, M.; Miyazuno, Y.; Sasamoto, K.; Ohkura, Y.; Ueno,
K. A Highly Water-Soluble Tetrazolium Salt as a Chromogenic Indicator
for NADH as well as Cell Viability. Talanta 1997, 44, 1299–1305.
(27) Cann, M. C. Formation of Acridine from the reaction of
Dibenz[b,f]azepine with Silver(I): Formation of an Aromatic Nitrenium
Ion? J. Org. Chem. 1988, 53, 1112–1113.
(28) Bausch, M. J.; Gostowski, R. Homolytic and Heterolytic Bond
Strengths. J. Org. Chem. 1991, 56, 6260–6262.
(29) Xu, H. J.; Lin, S. Q.; Shen, S. Y.; Chen, D. W.; Xu, G. Z. Radical
Mechanism in the Photoinduced Oxygen Transfer from Acridine
N-Oxide to Cyclohexene. J. Photochem. Photobiol., A 1989, 48, 53–59.
833
dx.doi.org/10.1021/ml2001483 |ACS Med. Chem. Lett. 2011, 2, 828–833