Core-modified porphyrins. Part 6: Effects of lipophilicity and core structures on physicochemical and biological properties in vitro

Article abstract of DOI:10.1016/j.bmc.2007.12.023

Thiaporphyrins 2-8 were prepared as analogues of 5,20-diphenyl-10,15-bis[4-(carboxymethyleneoxy)-phenyl]-21,23- dithiaporphyrin (1) to examine the effect of structural modifications: substituent changes in meso aryl groups of dithiaporphyrins with one wat

Full text of DOI:10.1016/j.bmc.2007.12.023

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 3171–3183
Core-modified porphyrins. Part 6: Effects of lipophilicity and
core structures on physicochemical and biological properties in vitro
Ethel J. Ngen,a Thalia S. Daniels,b Rajesh S. Murthy,a
Michael R. Dettyb and Youngjae Youa,*
aDepartment of Chemistry and Biochemistry, South Dakota State University, Brookings, SD 57007, USA
bDepartment of Chemistry, University at Buffalo, The State University of New York, Buffalo, NY 14260, USA
Received 30 October 2007; revised 10 December 2007; accepted 11 December 2007
Available online 27 December 2007
Abstract—Thiaporphyrins 2–8 were prepared as analogues of 5,20-diphenyl-10,15-bis[4-(carboxymethyleneoxy)-phenyl]-21,23- dith-
iaporphyrin (1) to examine the effect of structural modifications: substituent changes in meso aryl groups of dithiaporphyrins with
one water-solubilizing group (2–5), dihydroxylation of a pyrrole double bond and reduction to dihydroxychlorins (6 and 7), and the
removal of two meso aryl groups to give unsubstituted meso positions (8). The impact of these structural modifications was mea-
sured in both physicochemical (UV spectra, generation of singlet oxygen, lipophilicity, and aggregate formation) and biological
properties (dark toxicity and phototoxicity, cellular uptake, and subcellular localization). Mono-functionalized porphyrins had
much higher lipophilicity than di-functionalized porphyrin 1 and, consequently, formed more aggregates in aqueous media. The for-
mation of aggregates might lower the efficiency of lipophilic porphyrins as photosensitizers. Interestingly, dihydroxylation of a core
pyrrole group in the dithiaporphyrin core did not affect either the absorption spectrum or the efficiency for generating singlet oxy-
gen. The phototoxicity of dihydroxydithiachlorins mainly depended on their intracellular uptake. The potent phototoxicity of 6,
IC50 = 0.18 lM, was attributed to the extraordinarily high uptake. The intracellular uptake of 6 was about 7.6 times higher than
1. In contrast, thiaporphyrin 8 with only two meso aryl groups was less effective as a photosensitizer, perhaps due to poorer uptake
and a lower quantum yield for the generation of singlet oxygen.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
problems.3,4 Physically, Photofrinꢁ is a mixture of com-
pounds, which makes pharmacological evaluation diffi-
cult and causes problems in terms of reproducibility in
its preparation. In addition, some patients suffer long-
term skin photosensitization following treatment with
Photofrinꢁ. Furthermore, Photofrinꢁ is only weakly
absorbing at its absorption maximum (ꢀ630 nm), which
limits the photoefficiency of treatment and its absorp-
tion maximum limits the depth of penetration of activat-
ing light.
Photodynamic therapy (PDT) offers a great opportunity
to reduce severe side effects and enhance selectivity in
cancer treatments due to its unique mechanism of ac-
tion.1,2 The expression of biological activity in PDT is
derived from the combined effect of three components:
a photosensitizer, light, and oxygen. Excited-state
photosensitizers generate reactive oxygen species by
the transfer of energy to produce singlet oxygen or by
the transfer of electrons to produce superoxide, hydro-
gen peroxide, or hydroxyl radicals. Therefore, although
photosensitizers are generally administered systemically,
damage by PDT occurs only near/around the areas irra-
diated by the light. Although Photofrinꢁ has been effec-
tive as a first-generation photosensitizer, it has several
Core-modified porphyrins, in particular dithiaporphy-
rins, have several advantages: flexibility in synthesis,
preparation in high purity, absorption at longer wave-
lengths (band I absorption maxima of ꢀ700 nm), and
high photostability.5 In an effort to develop second-gen-
eration, dithiaporphyrin photosensitizers, we synthe-
sized numerous core-modified porphyrins and studied
the relationship between their structure and their photo-
toxicity.5–11 Dithiaporphyrins with two carboxylic acid
groups were found to be more efficient photosensitizers
than dithiaporphyrins with one, three, and four carbox-
Keywords: Core-modified porphyrin; Thiaporphyrin; Photodynamic
therapy; Anticancer therapy; SAR.
*
Corresponding author. Tel.: +1 605 688 6905; fax: +1 605 688
6364; e-mail: youngjae.you@sdstate.edu
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.12.023

3172
E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
X1
X3
N
H
S
OH
S
S
N
N
N
N
N
N
OH
S
S
H
H
O
O
Z
Z
X2
Y
O
O
OH
HO
6: Z = H
7: Z = OCH2COOH
1: X1 = X3 = H, X2 = Y = OCH2COOH
2: X1 = X2 = X3 = H, Y = OCH2COOH
3: X1 = X2 = X3 = F, Y = OCH2COOH
4: X1 = X2 = H, X3 = C(CH3)3, Y = OCH2COOH
5: X1 = X2 = X3 = F, Y = SO3Na
8
Figure 1. Structures of 21-mono- or 21,23-dithiaporphyrins.
ylic acid groups.8 Among dithiaporphyrins with two car-
boxylic acid groups, steric effects at the meso-aryl sub-
stituents were more determinant of phototoxicity than
electronic effects at the meso-aryl groups and a prefer-
ence was noted for smaller substituents and two different
meso substituents.5,9
tate to give 15 followed by saponification of the ethyl
ester. Dithiaporphyrin 4 was prepared using similar
chemistry to that used for the preparation of 3 start-
ing with methoxy porphyrin 16.10 Compound 21 was
obtained in 14% isolated yield through the condensa-
tion reaction of 19 and 2,5-bis[1-(4-fluorophenyl)-1-
hydroxymethyl]-thiophene 20. Sulfonation of the
meso-phenyl group of 21 afforded 5 in 50% isolated
yield (Scheme 1).
In this report, we extend our SAR study to mono-func-
tionalized dithiaporphyrins 2–5, dihydroxydithiachlo-
rins 6 and 7, and monothiaporphyrin 8 bearing only
two meso-aryl groups (Fig. 1). We describe the syntheses
of 2–8 as well as the measurement of their cellular
uptake and phototoxicity. In order to elucidate the
determining factors for phototoxicity, the absorption
spectra, relative yields for the generation of singlet
oxygen, the lipophilicity (logD7.4), the aggregation
tendency, and sites of sub-cellular localization were
determined.
2.1.2. Synthesis of dihydroxydithiachlorins. Dihydroxydi-
thiachlorins, 6 and 7, were synthesized via oxidation of a
pyrrole double bond with OsO4 to give syn-dihydroxyla-
tion.12 The precursor dithiaporphyrins, 22 and 23, were
prepared using previously reported chemistry.6,8 7,8-
Dihydroxydithiaporphyrins 6 and 24 were prepared by
oxidation with OsO4 in a solvent mixture of chloroform
and 1% pyridine, in 34% and 19% yields, respectively.
Saponification of the esters in 24 gave dihydroxy-
dithiaporphyrin 7 with two carboxylic acids in 75% yield
(Scheme 2).
2. Results and discussion
2.1. Chemistry
2.1.3. Synthesis of core-modified porphyrin 8 with two
meso-aryl groups and two unsubstituted meso positions.
Core-modified porphyrin 8 with two unsubstituted
meso-positions was prepared in several steps from dime-
thoxy thiophene 25.13 Initial condensation of 25 with
benzaldehyde and pyrrole in the presence of p-toluene-
sulfonic acid and TCBQ gave 27 in 6% isolated yield.
Demethylation of 27 with BBr3, alkylation of the result-
ing diol 28 with ethyl bromoacetate, and saponification
of the two ester groups of 29 provided carboxylic acid 8
(Scheme 3).
Compounds 1 and 2 were prepared as described in our
earlier work.8 The syntheses of 3–8 used methods that
we developed to construct asymmetric core-modified
porphyrins.
2.1.1. Synthesis of mono-functionalized core-modified
porphyrins. Trifluorinated mono-carboxylic acid dith-
iaporphyrin, 3, was prepared based on previous syn-
thetic methods used for the preparation of
asymmetric dithiaporphyrins.8 Compound 9 was syn-
thesized from thiophene and 4-fluorobenzaldehyde in
46% isolated yield. After protection of the hydroxyl
2.2. Photophysical properties
group of
9
with tert-butyldimethylsilyl chloride
2.2.1. Absorption spectra. Mono-functionalized dithia-
porphyrins 2–5 absorb light at wavelengths similar to
those reported for other dithiaporphyrins6–11 with band
I maxima near 700 nm. Dihydroxydithiachlorins 6 and 7
have band I maxima at 698 nm consistent with reported
data for other dihydroxydithiachlorins (Table 1).12 In
contrast, thiaporphyrin 8 with two unsubstituted meso
positions has a band I absorption maximum at
665 nm. The shorter wavelength is presumably due to
(TBSCl), the dihydroxy compound 11 was prepared
from 9 and p-anisaldehyde in 74% yield after depro-
tection of the TBS group with 1 M TBAF. Condensa-
tion
of
11
with
2,5-bis[1-(4-fluorophenyl)-1-
pyrrolomethyl]thiophene 12 afforded the dithiaporphy-
rin 13 in 8.5% isolated yield. Demethylation of 13
with BBr3 gave 14 in 82% yield. The final product,
3, was obtained after alkylation with ethyl bromoace-

E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
3173
F
F
F
a
b
OHC
S
S
OH
O
Si
c
10
9
F
F
O
F
S
d
N
N
S
12
HO
OH
S
11
F
R
R = OCH3
13
e
g
g
f
R = OCH2CO2H
3
R = OH
R = OCH2CO2CH2CH3
14
15
f
R = OH
R = OCH2CO2H
R = OCH2CO2CH2CH3
17
4
18
S
N
S
R = OCH3
N
16
R
F
F
R
F
S
S
i
h
N
N
R = SO3Na
S
20
5
NH
HN
F
R
R = H
R = H
19
21
Scheme 1. Reagents: (a) i—1 equiv n-BuLi, ii—1 equiv thiophene; (b) TBSCl, DMAP, Et3N; (c) i—1 equiv n-BuLi, ii—1 equiv 4-methoxy-
benzaldehyde, iii—aqueous HCl; (d) 12, TCBQ, TsOHÆH2O, CH2Cl2; (e) BBr3, CH2Cl2; (f) BrCH2CO2Et, K2CO3, acetone; (g) NaOH, aqueous
THF; (h) 20, TCBQ, TsOHÆH2O, CH2Cl2; (i) concd H2SO4.
the incorporation of only one sulfur atom in the core
and the lack of two meso-aryl groups.14
solution of DPBF (90 lM) and core-modified porphyrin
(1 lM) was irradiated with 3 mW/cm2 of the water-fil-
tered halogen source (400–850 nm) for 10 min (1.8 J/
cm2).15 THF was used as a solvent to avoid potential
problems from aggregation. The percentages of oxidized
DPBF by each porphyrin are summarized in Table 2.
Consistent with our previous studies,5,8,9 dithiaporphy-
rins 1–5 were excellent singlet oxygen generators. Inter-
estingly, dihydroxydithiachlorins 6 and 7 generated
2.2.2. Generation of singlet oxygen. To estimate the rela-
tive rates of singlet oxygen in core-modified porphyrins
1–8, we measured the decrease in 1,3-diph-
enylisobenzofuran (DPBF) absorbance as DPBF re-
acted with singlet oxygen generated by irradiation of
the core-modified porphyrin. A tetrahydrofuran (THF)

3174
E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
S
S
OH
OH
a
b
N
N
7
N
N
S
S
R
R
R
R
6, R = H
24, R = OCH2COOCH2CH3
22, R = H
23
, R = OCH2COOCH2CH3
Scheme 2. Reagents: (a) OsO4, CHCl3/pyridine (100:1), H2S, NaOH; (b) NaOH, aqueous THF.
b
R = OH
O
c
O
N
H
28
OH
a
N
HO
N
S
26
R = OCH2CO2CH2CH3
S
H
29
25
H
d
R
R
R = OCH3CO2H
R = OCH3
8
27
Scheme 3. Reagents: (a) 2,5-bis(2-phenyl-1-hydroxymethyl)thiophenepyrrole (26), TCBQ, TsOHÆH2O, CH2Cl2; (b) BBr3, CH2Cl2; (c) BrCH2CO2Et;
(d) NaOH, aqueous THF.
Table 1. UV–vis–near-IR band maxima and molar absorptivities for core-modified porphyrins in tetrahydrofurana
Compound
Soret band
Band IV
Band III
Band II
Band I
1
438 (119)
438 (205)
442 (185)
439 (200)
436 (132)
438 (154)
438 (108)
428 (70.9)
516 (20.8)
516 (25.2)
520 (24.8)
513 (26.3)
512 (20.3)
516 (22.8)
516 (18.7)
505 (21.9)
547 ( 11.1)
547 (9.8)
548 (10.7)
548 (10.3)
544 (5.6)
547 (3.3)
547 (15.0)
561 (3.0)
633 (2.2)
633 (2.2)
635 (2.6)
634 (2.1)
631 (1.5)
633 (10.9)
633 (3.6)
604 (5.2)
698 (6.0)
698 (6.1)
699 (6.1)
698 (5.8)
694 (3.2)
698 (4.9)
698 (8.6)
665 (2.9)
2
3
4
5b
6
7
8
a kmax nm (e · 103 MÀ1 cmÀ1).
b In MeOH.
Table 2. Percentiles of oxidized DPBF by the irradiation with
porphyrins and n-octanol/water partition coefficients in pH 7.4
phosphate buffer
2.2.3. n-Octanol/pH 7.4 buffer partition coefficients. The
lipophilicity of a molecule is important not only in cellu-
lar uptake, but also in the determination of water solu-
bility, which is closely related to the tendency to form
aggregates. This is especially true for porphyrins. Parti-
tion coefficients of porphyrins 1–8 were determined as
logD7.4 using n-octanol and pH 7.4 phosphate buffer.
Experimental logD7.4 values reflected the effects of sub-
stituents at meso-aryl groups except 8. Monocarboxylic
acids 2–4 and monosulfonated porphyrin 5 had higher
values of logD7.4 than dicarboxylic acid porphyrin, 1.
Sulfonated porphyrin 5 had a lower value of logD7.4
due to the higher acidity of the sulfonic acid residue rel-
ative to the carboxylic acid. Monocarboxylic acid por-
phyrin 4 with a tert-butyl group showed a higher value
of logD7.4 than 2. The hydrophilic effect of the two alco-
hol substituents of 6 was much smaller than the effect of
the two carboxylic acid substituents of 1 (logD7.4 of >2
vs logD7.4 of 0.1). However, the addition of two hydro-
Compound
Oxidized DPBF (%)
logD7.4
1
2
3
4
5
6
7
8
68
78
75
75
64
78
62
48
0.1
1.6
1.8
>2
0.8
>2
À0.5
0.8
singlet oxygen, 78% and 62% oxidation, respectively, as
effectively as dithiaporphyrins 1–5. The mono-thia-
porphyrin 8, with two unsubstituted meso-positions,
was less efficient than the others for the generation of
singlet oxygen with only 48% oxidized DPBF.

E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
3175
xyl groups in 1 to give dithiachlorin 7 significantly low-
ered logD7.4 from 0.1 for 1 to À0.5 for 7. Interestingly,
logD7.4 of 8 was higher than that of 1 (0.8 vs 0.1) even
though both have two carboxylic acid residue and 8
lacks two aromatic meso substituents.
rin for 24 h. With the exception of the intracellular con-
centration of 6, concentrations of all porphyrins were
less than 1 fmole per cell (0.31–0.59 fmole/cell). In par-
ticular, the uptake of 7 and 8 was much less than the
others, 0.11 and 0.16 fmole/cell, respectively. On the
other hand, 6 showed much higher uptake, 4.5 fmole/
cell. The uptake of the porphyrins in cells seems to cor-
relate with phototoxicity. The potent activity of 6 is
accompanied by the highest uptake (4.5 fmole/cell)
among core-modified porphyrins 1–8. The lower uptake
of 7 and 8 also correlates with the low phototoxicity of
these two compounds. The relatively low lipophilicity of
7 (logD7.4, À0.5) might make passage through the lipid
bilayer of cell membrane difficult. This observation is
consistent with our previous results where core-modified
porphyrins with three or four carboxylic acids showed
much lower uptake.8 The lower efficiency of singlet oxy-
gen generation in 8 also contributes to the lower potency
of 8 (Table 2). The sulfonated porphyrin 5 showed lower
potency than the carboxylated porphyrins, although the
uptake (0.30 fmole/cell) was close to the uptake ob-
served with 2 (0.31 fmole/cell) and 3 (0.33 fmole/cell)
and singlet oxygen was generated nearly as effectively
(64%) as the others (75–78%).
2.3. Biology
2.3.1. Dark and phototoxicity of core-modified porphyrins
toward cultured R3230AC cells. We tested the photody-
namic activity of the core-modified porphyrins against
R3230AC cells. The porphyrins were added to the cells
24 h before the irradiation with water-filtered halogen
light source (3 mW/cm2 for 1 h, 10.8 J/cm2). Cell viabil-
ity was determined by MTT colorimetric assay.16 The
Hill (sigmoid Emax) equation was used to fit the data
(Fig. 2) and to calculate values of IC50. Values of IC50
ranged from 0.13 to 2.6 lM: a 20-fold difference between
the best and the worst. Dithiachlorin derivative 7 and
thiaporphyrin 8 with two unsubstituted meso positions
showed much lower potency than the others with values
of IC50 > 2 lM. Among the mono-functionalized por-
phyrins, 4 and 5 were less phototoxic than 2 and 3.
Interestingly, dithiachlorin 6 was more phototoxic than
derivatives 2–5, 7, and 8, and was similar to 1 in photo-
toxicity. In dark and light controls, no dark toxicity was
observed in R3230AC cells at porphyrin concentrations
up to 10 lM and no toxicity was observed in R3230AC
cells treated with 10.8 J/cm2 of water-filtered halogen
light (data not shown).
Extraordinarily high uptake of 6 (4.5 fmol/cell) was
unexpected because it was highly lipophilic and formed
aggregates in the media. Compound 4 also showed high
lipophilicity and formed aggregates well, but uptake was
much lower (0.46 fmol/cell). Compound 6 might cross
cell membrane more readily due to two reasons: (1) it
is smaller in size and (2) it does not have flexible hydro-
philic functional group, which could interfere the por-
phyrin’s crossing through plasma membrane as both a
monomer and an aggregate. However, we do not ex-
clude the possibility that compound 6 forms different
2.3.2. Intracellular accumulation of core-modified por-
phyrins in R3230AC cells. The cellular concentration of
the various porphyrins was determined by fluorescence
and is expressed in fmole per cell as shown in Figure
3. Cells were treated with 10 lM core-modified porphy-
Figure 2. Cell viability of cultured R3230AC cells after photosensitization in the presence of thiaporphyrins 1–8. Each data point represents the mean
of at least 3 separate experiments performed in duplicate and error bars are the SEM. Data are expressed as the surviving fraction of viable cells
relative to untreated controls. IC50 (lM): 1 (0.13), 2 (0.26), 3 (0.38), 4 (0.59), 5 (0.77), 6 (0.18), 7 (2.57), and 8 (2.28).

3176
E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
Figure 3. Cellular uptake of thiaporphyrins 1–8 in cultured R3230AC cells. Each bar represents the mean intracellular uptake of each compound
incubated with R3230AC cells for 24 h at 1 · 10À5 M. Data are expressed as femtomole porphyrin/cell and error bars are the SEM.
type of aggregates due to the absence of the flexible
hydrophilic chain.
2.3.4. Sub-cellular localization of core-modified porphyrin
in R3230AC cells. The site of intracellular localization
of the photosensitizer has been suggested to be one
of the main factors in determining the efficiency of
photodynamic activity.19 This statement is supported
by two observations: (1) the site of localization of
the photosensitizers will be the site of cellular damage
since the diffusion distance of singlet oxygen is no fur-
ther than 20 nm in a biological system20 and (2) the
sensitivity of organelles to damage by singlet oxygen
can be quite different. Photodynamic damage to mito-
chondria can lead to the induction of apoptosis,21–23
which has made mitochondria the premier target of
photosensitizers.24,25
2.3.3. Aggregates of core-modified porphyrins in media.
The extent of core-modified porphyrin aggregation in
growth medium was indirectly determined by a compar-
ison of their fluorescence yield in medium relative to
their fluorescence observed in DMSO (Fig. 4). The loss
of fluorescence emission upon aggregation of porphyrins
has been reported for both anionic and cationic porphy-
rins.17,18 The decrease in fluorescence of lipophilic por-
phyrins 2–6 in medium compared to that in DMSO
was quite apparent as shown in Figure 4. Fluorescence
from the more hydrophilic porphyrins 1, 7, and 8
showed little difference between growth medium and
DMSO. These data indicate that the lipophilic porphy-
rins are more prone to form aggregates than hydrophilic
porphyrins in medium.
In order to discern the subcellular localization of the
core-modified porphyrins, R3230AC cells were treated
with core-modified porphyrins 1, 4, or 6, or with rho-
damine-123 (Rh-123), a standard mitochondrial stain-
ing dye (Fig. 5).26 R3230AC cells were incubated
with 20 lM of the core-modified porphyrin for 24 h
or with 13 lM Rh-123 for 30 min and were then
washed three times with media to remove residual
porphyrin or Rh-123. The core-modified porphyrins
did not provide the fine, granular fluorescence associ-
ated with mitochondrial specificity as shown with Rh-
123 in Figure 5. While all three core-modified porphy-
rins gave similar fluorescence images, some slight dif-
ferences were observed. Staining with 4 looked more
specific without more diffuse staining area. The core-
modified porphyrins did not stain nuclei and, while
mitochondria may be stained, the core-modified por-
phyrins were not specific for mitochondria. These
images did not provide conclusive information to ex-
plain the differences of the efficiency between hydro-
philic core-modified porphyrin, 1, and lipophilic
porphyrins, 4, 6.
The tendency to aggregate seemed the most striking
factor responsible for the lower efficiency of highly
lipophilic porphyrins, for example, 4 and 6. Although
the intracellular uptake of 4 (0.46 fmole/cell) was close
to that of 1 (0.59 fmole/cell), 1 was about 4.5 times
more potent than 4 (IC50, 0.13 lM vs 0.59 lM). The
IC50 of 6 was slightly higher than that of 1 (0.18 vs
0.13 lM), but the cellular concentration of 6 was 7.6
times higher than that of 1 (4.5 vs 0.59 fmole/cell).
On a ‘molecule per cell’ basis, both 4 and 6 were
much less efficient as photosensitizers than 1. In gen-
eral, when dyes form aggregates, quantum yields for
photophysical processes such as fluorescence emission
and singlet oxygen generation are decreased. We sug-
gest that the high tendency of 6 and 4 to form aggre-
gates in media may be mirrored in the cytoplasm
causing the lowered phototoxicity of the lipophilic
photosensitizers.

E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
3177
Figure 4. Fluorescence emission from core-modified porphyrins 1–8 in DMSO (upper) and in aqueous media (lower) at three different
concentrations: 0.05, 0.1, and 0.2 lM.
3. Summary and conclusions
absorption spectrum of dihydroxydithiachlorin
6
relative to tetranitrogenic chlorins was reported.12 In
the tetranitrogenic porphyrins, chlorins usually have
longer-wavelength absorption maxima than the corre-
sponding porphyrins. However, chlorins of dithiapor-
phyrins do not show a similar bathochromic shift.
Interestingly, dihydroxydithiachlorins 6 and 7 generated
singlet oxygen as effectively as dithiaporphyrin 1. Lipo-
philicity was affected mainly by the functional groups.
This study was designed to observe structure–activity
relationships of core-modified porphyrins with physico-
chemical and biological properties as a continuation of
our previous studies.5–9 Mono-functionalized dithiapor-
phyrins 2–5, dihydroxydithiachlorins 6 and 7, and
mono-thiaporphyrin 8 with two unsubstituted meso-
positions were prepared and compared with dith-
iaporphyrin 1. Compounds 2–8 were prepared based
on our previous methods with slight modification. In
the preparation of dihydroxydithiachlorins 6 and 7,
OsO4 was used.12 Mono-thiaporphyrin 8 was prepared
as a surrogate for dithiaporphyrin with only two meso
aryl groups.
Although none of compounds 2–8 was more potent than
1, we were able to observe relations between structure
and activity. The phototoxicity of the core-modified
porphyrins was dependent on multiple factors such as
intracellular uptake, lipophilicity, the tendency to form
aggregates, and singlet-oxygen quantum yield. The high-
er potency of 6 is attributed to extraordinarily high up-
take compared to the other core-modified porphyrin
derivatives. The lower efficiency of highly lipophilic por-
phyrins, for example, 4 and 6, might be, in part, due to
the formation of inactive aggregates in the cytoplasm.
Water-soluble derivative 7 had the lowest uptake in
the series and poor phototoxicity. On the other hand,
the poor phototoxicity of 8 might be due to the com-
The photophysical properties were similar within the
series of dithiaporphyrins 1–7. The band I absorption
maxima of 1–7 were between 694 and 698 nm (665 nm
for thiaporphyrin 8). All of the dithiaporphyrins gener-
ated singlet oxygen quite effectively and oxidized 62–
78% of DPBF (48% for thiaporphyrin 8). The differences
observed with 8 might be due to the core system and/or
the absence of two meso aryl groups. The unusual

3178
E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
Figure 5. Fluorescence images of R3230AC cells treated with 20 lM of 1, 4, 6, and 13 lM of Rh-123. Cell culture conditions and experimental details
are described in Section 4.
bined effects of lower uptake and less effective genera-
tion of singlet oxygen.
ducted by the Campus Chemical Instrumentation Cen-
ter of the Ohio State University (Columbus, OH), the
Instrument Center of the Department of Chemistry at
the University at Buffalo, or the Campus Mass Spec-
trometry Facility of at South Dakota State University.
4. Experimental
4.1. General methods
4.2. Synthesis
Solvents and reagents were used as received from
PHARMCO-AAPER, Sigma-Aldrich Chemical Co. or
Thermo Fisher Scientific, Inc. unless otherwise noted.
Chemicals for tissue culture were purchased from Sig-
ma-Aldrich Chemical Co. or Thermo Fisher Scientific,
Inc. NMR spectra were recorded at 23 ꢂC on a Varian
Gemini-300, Inova 400 (Bruker AVANCE 400), or Ino-
va 500 instrument with residual solvent signal as the
internal standard: CDCl3 (d 7.26 for proton, d 77.16
for carbon). UV–visible near-IR spectra were recorded
on a Perkin-Elmer Lambda 12 or CHEM4-UV-FIBER
spectrophotometer (Ocean Optics Inc.). Elemental anal-
yses were conducted by Atlantic Microlabs, Inc. Q-TOF
2 electrospray and ESI mass spectrometry were con-
Compounds 1, 2, 12, 16, 20, 22, 23, and 26 were pre-
pared as previously described in our earlier works.5–8,10
4.2.1. Synthesis of 2-[1-(4-fluorophenyl)-1-hydroxy-
methyl]-thiophene (9). Thiophene (12.0 mL, 150 mmol)
was added to a solution of n-butyllithium (93.8 mL of
a 1.6 M solution in hexanes, 150 mmol) and TMEDA
(22.7 mL, 150 mmol) in 400 mL of hexanes under an
Ar atmosphere. The reaction mixture was heated at re-
flux for 1 h, cooled to ambient temperature to make 2-
lithiophene. This 2-lithiothiophene suspension was
transferred via a cannula to a solution of 4-fluorobenz-
aldehyde (15.0 mL, 143 mmol) in 400 mL of anhydrous
THF cooled to 0 ꢂC, which had been degassed with Ar

E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
3179
for 15 min. After the addition was complete, the mixture
was warmed to ambient temperature, 500 mL of aque-
ous 1M NH4Cl was added, and the organic phase was
separated. The aqueous phase was extracted with ether
(3 · 400 mL). The combined organic extracts were
washed with water (3 · 400 mL) and brine (400 mL),
dried over MgSO4, and concentrated. The crude product
was purified by silica gel column chromatography with
the mixture of EtOAc and hexanes giving 13 g (46%)
of 9 as a light yellow oil. The structure was confirmed
by the comparison with the reference data.27
6.88 (2H, d, J = 6.5 Hz), 7.15 (2H, t, J = 6.9 Hz), 7.29
(2H, m), 7.42 (2H, t, J = 5.2 Hz); HR Q-TOF MS: m/z
344.0882 (calcd for C19H17FO3S, 344.0885).
4.2.4. 5,15,20-Tri(4-fluorophenyl)-10-(4-methoxyphenyl)-
21,23-dithiaporphyrin (13). Diol 11 (4.0 g, 12 mmol), 2,5-
bis[1-(4-fluorophenyl)-1-pyrrolomethyl]thiophene (12,
5.0 g, 12 mmol), and 2,3,5,6-tetrachloro-1,4-benzoqui-
none (TCBQ, 8.6 g, 35 mmol) were dissolved in
500 mL CH2Cl2. p-Toluenesulfonic acid monohydrate
(2.2 g, 12 mmol) was added and the reaction mixture
was stirred for 0.5 h in the dark. The reaction mixture
was concentrated and the residue was redissolved in
minimal CH2Cl2. The crude product was purified via
chromatography on basic alumina eluted with CH2Cl2.
A red band containing dithiaporphyrin 13 was isolated.
The crude product was washed with acetone to give
4.2.2. 2-[1-(4-Fluorophenyl)-1-(tert-butyldimethylsilyloxy)-
methyl]thiophene (10). Compound 10 was prepared fol-
lowing literature procedures.8 Briefly, a solution of com-
pound 9, TBSCl (28.0 g, 186 mmol), DMAP (7.5 g,
61 mmol), and Et3N (11.0 mL, 188 mmol) in 300 mL of
dry CH2Cl2 was stirred at 0 ꢂC for 2 h and was then
warmed to ambient temperature for 24 h. The reaction
mixture was partitioned between 400 mL of ether and
400 mL of saturated aqueous NaHCO3. The organic layer
was washed with water (3 · 400 mL) and brine (400 mL),
dried over MgSO4, and concentrated to give a yellow oil.
Chromatography on SiO2 eluted with 25% EtOAc/hex-
anes gave 14.7 g (74%) of 10 as a colorless oil. 1H NMR
(300 MHz, CDCl3): d 0.04 (3H, s), 0.11 (3H, s), 0.99
(9H, s), 6.01 (1H, s), 7.25 (2H, t, J = 8.7 Hz), 7.44–7.65
(5H, m); 13C NMR (75 MHz, CDCl3): d À4.89, À4.70,
18.42, 25.91, 72.72, 115.10, 115.38, 123.75, 124.86,
126.52, 127.91, 128.02, 140.42, 140.46, 150.33, 160.65,
163.90; HR Q-TOF MS: m/z 345.1105 (calcd for
C17H23FOSSi + Na, 345.1121).
1
0.72 g (9%) of 13 as a purple solid. Mp: >300 ꢂC; H
NMR (400 MHz, CDCl3): d 4.13 (3H, s), 7.40 (2H, d,
J = 8.6 Hz), 7.54 (6H, t, J = 8.6 Hz), 8.22 (8H, m),
8.68 (3H, m), 8.75 (1H, d, J = 4.5 Hz), 9.68 (3H, m)
9.77 (1H, d, J = 5.0 Hz); HR ESI MS: m/z 733.1602
(calcd for C45H27F3N2OS2 + H, 733.1595).
4.2.5. 5,15,20-Tri(4-fluorophenyl)-10-(4-hydroxyphenyl)-
21,23-dithiaporphyrin (14). Dithiaporphyrin 13 (0.47 g,
0.64 mmol) was dissolved in 50 mL of CH2Cl2 and
BBr3 (0.2 mL, 1.9 mmol) was added at 0 ꢂC. The result-
ing solution was stirred overnight at ambient tempera-
ture. The reaction mixture was added to 150 mL of
EtOAc and 150 mL of saturated NaHCO3. The organic
layer was separated and washed three times with 150 mL
of brine, dried over MgSO4, and concentrated. The
crude solid was washed with 25% EtOAc/hexanes sev-
eral times to give 0.38 g (82%) of 14 as a dark blue solid.
4.2.3. 2-[1-(4-Fluorophenyl)-1-hydroxymethyl]-5-[1(4-
methoxyphenyl)-1- hydroxymethyl]thiophene (11). Com-
pound 10 (7.0 g, 22 mmol) was added to a solution of
n-butyllithium (14.9 mL, 1.6 M in hexanes, 24 mmol)
and TMEDA (3.6 mL, 24 mmol) in 50 mL of hexanes
under an Ar atmosphere. The reaction mixture was stir-
red at ambient temperature for 30 min. The suspension
of lithio-10 was transferred dropwise via cannula to a
0 ꢂC solution of 4-methoxybenzaldehyde (2.5 mL,
21 mmol) in anhydrous THF (50 mL), which had been
degassed with Ar for 15 min. After addition was com-
plete, the mixture was allowed to warm to ambient tem-
perature, 100 mL of a 1 M solution of NH4Cl was
added, and the organic phase was separated. The aque-
ous phase was extracted with ether (3 · 150 mL). The
combined organic extracts were washed with water
(3 · 150 mL) and brine (150 mL), dried over MgSO4,
and concentrated to give a yellow oil. The oil was dis-
solved in a 1 M solution of Bu4NF in THF (50 mL,
50 mmol) and stirred at ambient temperature for 1 h at
which point 50 mL of saturated aqueous NH4Cl was
added. The resulting mixture was extracted with ether
(4 · 70 mL). The combined organic extracts were
washed with water (3 · 150 mL) and brine (150 mL),
dried over MgSO4, and concentrated. The crude diol
was purified by column chromatography on SiO2 eluted
with 25% EtOAc/hexanes to give 5.5 g (74%) of 11 as a
1
Mp: >300 ꢂC; H NMR (400 MHz, DMSO-d6): d 6.45
(2H, d, J = 8.5 Hz), 6.89 (6H, m), 7.23 (2H, d,
J = 8.4 Hz), 7.45 (6H, m), 7.78 (3H, m), 7.86 (1H, d,
J = 4.5 Hz), 8.87 (3H, m), 8.97 (1H, d, J = 5.1 Hz),
9.26 (1H, s); HR Q-TOF MS: m/z 719.1433 (calcd for
C44H25F3N2OS2 + H, 719.1442).
4.2.6. 5,15,20-Tri(4-fluorophenyl)-10-[4-(ethoxycarbon-
ylmethyleneoxy)-phenyl]-21,23-dithiaporphyrin (15). Dith-
iaporphyrin 14 (0.33 g, 0.46 mmol), K2CO3 (0.63 g,
4.6 mmol), and ethyl bromoacetate (0.5 mL, 4.6 mmol)
in 200 mL acetone were heated at reflux for 10 h. The
reaction mixture was cooled to ambient temperature
and the K2CO3 was removed by filtration. The filter cake
was washed with acetone until the filtrate became color-
less. The combined filtrates were concentrated. The crude
product was washed with MeOH to give 0.32 g (87%) of
15 as a purple solid. Mp: 168–170 ꢂC; 1H NMR
(400 MHz, CDCl3): d 1.44 (3H, t, J = 7.1 Hz), 4.44 (2H,
q, J = 7.1 Hz), 4.94 (2H, s), 7.40 (2H, d, J = 8.6 Hz),
7.54 (6H, t, J = 8.6 Hz), 8.22 (8H, m), 8.68 (3H, m),
8.73 (1H, d, J = 4.5 Hz), 9.67 (3H, m), 9.74 (1H, d,
J = 5.1 Hz); HR Q-TOF MS: m/z 805.1801 (calcd for
C48H31F3N2O3S2 + H, 805.1812).
1
yellow solid. H NMR (400 MHz, CDCl3): d 3.73 (3H,
s), 5.78 (1H, m), 5.85 (1H, m), 6.02 (1H, d,
J = 3.4 Hz), 6.17 (1H, d, J = 3.4 Hz), 6.65 (2H, m),
4.2.7. 5,15,20-Tri(4-fluorophenyl)-10-[4-(carboxymethyl-
eneoxy)-phenyl]-21,23-dithiaporphyrin (3). Core-modi-
fied porphyrin 15 (0.30 g, 0.37 mmol) was dissolved in

3180
E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
40 mL THF and 40 mL of 1 M aqueous NaOH was
added. The resulting solution was stirred at ambient
temperature for 15 h. The solution was acidified by the
addition of 4.1 mL of 10 N HCl. The reaction mixture
was diluted with 100 mL H2O and the products were ex-
tracted with EtOAc (3 · 100 mL). The combined organ-
ic extracts were dried over MgSO4 and concentrated.
The crude product was washed with several portions
of hexanes/MeOH to give 0.19 g (66%) of 3 as a purple
(8H, m), 7.81 (8H, dd, J = 5.2, 8.2 Hz), 7.35 (2H, dd,
J = 7.6 Hz), 4.94 (2H, s), 1.62 (9H, s); 13C NMR
(300 MHz, CDCl3): d 158.2, 157.1, 156.8, 151.5, 148.5,
148.3, 141.7, 138.7, 136.1, 135.9, 135.5, 135.2, 134.9,
134.9, 134.7, 134.5, 134.4, 133.7, 128.6, 127.9, 125.0,
114.3, 65.7, 35.5, 32.2; HR ESI MS: m/z 779.2421 (calcd
for C50H39O3N2S2 + H, 779.2397).
4.2.12. 2-[1-(4-Fluorophenyl)-1-pyrrolomethyl]-5-[1-phe-
nyl-1-pyrrolomethyl]-thiophene (19). 2-[1-(4-Fluorophenyl)-
1-hydroxymethyl]-5-[1-phenyl-1-hydroxylmethyl]-thiophene5
(6.0 g, 18 mmol) was dissolved in excess pyrrole (25 mL).
Boron trifluoride etherate was added (0.40 mL, 3.6 mmol)
and the resulting mixture was stirred for 1 h at ambient
temperature. The reaction was stopped by the addition
of CH2Cl2 (150 mL), followed by 40% NaOH (80 mL).
The organic layer was separated, washed with water
(3 · 200 mL) and brine (200 mL), dried over MgSO4,
and concentrated. The excess pyrrole was removed at re-
duced pressure at ambient temperature. The residual oil
was purified via chromatography on SiO2 eluted with
25% EtOAc/hexanes to give 5.0 g (68%) of 19 as a yellow
oil. 1H NMR (400 MHz, CDCl3): d 5.61 (1H, s), 5.63 (1H,
s), 5.96 (1H, s), 6.00 (1H, s), 6.22 (2H, s), 6.67 (1H, s), 6.69
(1H, s), 6.76 (2H, s), 7.06 (2H, dd, J = 8.1, 7.9 Hz),
7.23–7.36 (5H, m), 7.38 (2H, d, J = 6.8 Hz), 7.95 (2H, br
s); HR EI MS: m/z 412.1404 (calcd for C26H21FN2S,
412.1404).
1
solid. Mp: >300 ꢂC: H NMR (400 MHz, DMSO-d6): d
4.15 (2H, s), 6.59 (2H, d, J = 8.6 Hz), 6.88 (6H, t,
J = 8.7 Hz), 7.34 (2H, d, J = 8.6 Hz), 7.44 (6H, m),
7.78 (3H, m), 7.82 (1H, d, J = 4.5 Hz), 8.88 (3H, m),
8.93 (1H, d, J = 5.1 Hz); HR Q-TOF MS: m/z
777.1509 (calcd for C46H27F3N2O3S2 + H, 777.1488).
4.2.8.
5-(4-tert-Butylphenyl)-10-(4-methoxyphenyl)-15,
20-bis-phenyl-21,23-dithiaporphyrin (16). Dithiaporphy-
rin 16 was prepared as previously described.10
4.2.9. 5-(4-tert-Butylphenyl)-10-(4-hydroxyphenyl)-15,20-
diphenyl-21,23-dithiaporphyrin (17). Dithiaporphyrin 16
(0.16 g, 0.21 mmol) was treated with BBr3 (0.25 mL,
2.7 mmol) as described for the preparation of 14 to give
0.14 g (92%) of 17 as a metallic purple solid. Mp:
1
>300 ꢂC; H NMR (500 MHz, CDCl3): d 9.71 (4H, dd,
J = 4.9, 8.8 Hz), 8.71 (4H, dd, J = 4.3, 6.1 Hz), 8.25
(4H, dd, J = 1.6 Hz, 5.5 Hz), 8.19 (2H, d, J = 7.9 Hz),
8.11 (2H, d, J = 8.2 Hz), 7.82 (8H, dd, J = 7.9 Hz,
8.2 Hz), 7.23 (2H, d, J = 8.5 Hz), 5.09 (1H, br s),
1.62 (9H, s); 13C NMR (300 MHz, CDCl3): d 157.2,
156.6, 151.5, 148.5, 148.3, 141.8, 138.7, 136.1, 135.8,
135.2, 134.9, 134.9, 134.8, 134.7, 134.7, 134.4,
134.3, 128.5, 127.9, 125.0, 115.0, 35.5, 32.2; HR ESI
MS: m/z 721.2693 (calcd for C48H36ON2S2 + H,
721.2710).
4.2.13. 5,15,20-Tri-(4-fluorophenyl)-10-phenyl-21,23-di-
thiaporphyrin (21). Cyclization of 2,5-bis[(4-fluoro-
phenyl)-hydroxymethyl]-thiophene7 (20, 2.9 g, 7.0 mmol)
and 19 (2.3 g, 6.9 mmol) with TsOHÆH2O (1.3 g,
6.9 mmol) and TCBQ (5.1 g, 21 mmol) was performed
as described for the preparation of 13 to give 0.67 g
1
(14%) of 21 as a purple solid; mp: >300 ꢂC. H NMR
(400 MHz, CDCl3): d7.55 (6H, m), 7.85 (3H, m), 8.23
(8H, m), 8.69 (4H, m), 9.68 (4H, m); HR EI MS: m/z
702.1406 (calcd for C44H25F3N2S2, 702.1407).
4.2.10. 5-(4-tert-Butylphenyl)-10-[4-(ethoxycarbonylm-
ethyleneoxy)-phenyl]-15,20-bis-phenyl-21,23-dithiaporph-
yrin (18). Dithiaporphyrin 17 (0.14 g, 0.19 mmol) was
reacted with ethyl bromoacetate (0.67 mL, 6.0 mmol)
and K2CO3 (0.82 g, 5.8 mmol) in 50 mL of acetone as
described for the preparation of 15 to give 0.15 g
(95%) of 18 as a dark purple solid. Mp: 185–187 ꢂC;
1H NMR (500 MHz, CDCl3): d 9.70 (4H, dd, J = 4.9,
8.8 Hz), 8.70 (4H, dd, J = 4.0, 6.1 Hz), 8.22 (8H, dd,
J = 5.8, 8.2 Hz), 7.82 (8H, dd, J = 7.3, 8.8 Hz), 7.36
(2H, d, J = 7.3 Hz), 4.92 (2H, s), 4.42 (2H, q,
J = 7.0 Hz), 1.61 (9H, s), 1.42 (3H, t, J = 7.2 Hz); 13C
NMR (300 MHz, CDCl3): d 169.5, 158.6, 157.2, 156.9,
151.5, 148.4, 148.4, 141.9, 138.8, 136.2, 135.9, 135.3,
135.0, 134.9, 134.8, 134.5, 134.4, 134.1, 134.0, 128.5,
128.0, 125.0, 114.3, 66.3, 62.1, 35.5, 32.2, 14.8; HR
ESI MS: m/z 807.2693 (calcd for C52H43O3N2S2 + H,
807.2710).
4.2.14. 5,15,20-Tri-(4-fluorophenyl)-10-(4-sulfonatophe-
nyl)-21,23-dithiaporphyrin sodium salt (5). Dithiaporphy-
rin 21 (0.30 g, 0.43 mmol) was dissolved in excess
concentrated H2SO4 (40 mL) and allowed to stir at
100 ꢂC overnight. The acid was slowly neutralized with
concentrated NaOH until the solution was slightly basic.
An equal volume of MeOH was added, and the solid
Na2SO4 was removed by filtration. The filtrate was con-
centrated, and the residue was dissolved in acetone. The
resulting solution was chilled precipitating more
Na2SO4, which was removed via filtration. The acetone
solution was concentrated, and the residue was recrys-
tallized from 10% aqueous MeOH to give 0.17 g (50%)
of 5as a purple solid. Mp: >300 ꢂC; 1H NMR
(400 MHz, CD3OD): d7.78 (6H, t, J = 8.4 Hz), 8.16
(2H, d, J = 7.3 Hz), 8.26 (2H, d, J = 7.4 Hz), 8.35 (6H,
t, J = 5.8 Hz), 8.67–8.73 (4H, s), 9.76–9.83 (4H, s); HR
ESI MS: m/z 783.1052 (calcd for C44H24F3N2O3S3 + H,
783.1069).
4.2.11. 5-(4-tert-Butylphenyl)-10-[4-(carboxymethylene-
oxy)-phenyl]-15,20-bis-phenyl-21,23-dithiaporphyrin (4).
Dithiaporphyrin 18 (0.15 g, 0.18 mmol) was hydrolyzed
by 1 M NaOH (16 mL) as described for the preparation
of 3 to give 0.095 g (68%) of 4 as a purple solid; mp:
4.2.15. 5,20-Bis-phenyl-10,15-bis[4-(ethoxycarbonylmeth-
yleneoxy)-phenyl]-21,23-dithia-7,8-dihydroxychlorin (24).
Dihydroxylation of dithiaporphyrin 23 followed litera-
1
>300 ꢂC. H NMR (500 MHz, CDCl3): d 9.69 (4H, dd,
J = 4.9, 8.8 Hz), 8.67 (4H, dd, J = 4.3, 7.9 Hz), 8.17

E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
3181
1
ture procedures.12 Briefly, dithiaporphyrin 23 (0.60 g
0.70 mmol) was dissolved in a 100:1 mixture of
CHCl3:pyridine (140 mL). To the solution, OsO4
(0.27 g, 1.1 mmol) was added and the flask was closed
with stoppers, covered with aluminum foil, and stirred
at ambient temperature for 3 d. The reaction was then
quenched by purging with H2S for 5 min and the excess
H2S was trapped by 6 M NaOH aqueous solution. The
solution was filtered through a plug of Celite and the fil-
trate was evaporated. The resulting residue was purified
on a silica gel column eluted with a mixture of CH2Cl2
and MeOH to give 0.15 g (24%) of 24 as a purple solid.
1H NMR (400 MHz, CDCl3 + CD3OD): d 1.39 (3H, t,
J = 4.4 Hz), 1.42 (3H, t, J = 4.7 Hz), 3.19 (1H, d,
J = 4.2 Hz), 3.25 (1H, d, J = 4.3 Hz), 4.35–4.44 (4H, s),
4.87 (2H, s), 4.90 (2H, s), 6.38–6.47 (2H, m), 7.19 (4H,
s), 7.28–7.35 (2H, m), 7.70–7.83 (5H, m), 7.80 (1H, d,
J = 7.9 Hz), 7.92–7.98 (1H, m), 8.04–8.24 (5H, m),
8.46–8.54 (2H, m), 9.12 (2H, t, J = 7.1 Hz), 9.45 (1H,
d, J = 5.0 Hz), 9.48 (1H, d, J = 4.9 Hz); HR ESI MS:
m/z 887.2974 (calcd for C52H42N2O8S2 + H, 887.2461).
Mp: 119–121 ꢂC; H NMR (400 MHz, CDCl3): d 1.38
(6H, t, J = 7.1 Hz), 4.44 (4H, q, J = 7.1 Hz), 5.76 (4H,
s), 7.75 (9H, m), 8.18 (4H, d, J = 7.9 Hz), 8.67 (2H, d,
J = 4.4 Hz), 8.91 (1H, d, J = 1.9 Hz), 9.01 (2H, d,
J = 4.4 Hz), 10.71 (1H, s); HR ESI MS: m/z 684.2468
(calcd for C40H33N3O6S + H, 684.2168).
4.2.20.
5,20-Bis-phenyl-12,13-carboxylatomethoxy-21-
thiaporphyrin (8). Mono-thiaporphyrin 30 (0.20 g,
0.29 mmol) was hydrolyzed by 1 M NaOH (4 mL) as de-
scribed for the preparation of 3 to give 0.10 g (55%) of 8
as a purple solid. Mp: 199–201 ꢂC; 1H NMR (400 MHz,
CDCl3): d À1.61 (1H, br s), 7.86 (4H, s), 8.22 (6H, s),
8.22 (4H, s), 8.56 (2H, s), 8.92 (2H, s), 8.92 (2H, s),
10.82 (2H, s), 12.12 (2H, s); HR ESI MS: m/z
628.1807 (calcd for C36H25N3O6S + H, 628.1542).
4.3. Photophysical properties
4.3.1. Relative quantum yields for singlet oxygen gener-
ation. A stock solution of 1,3-diphenylisobenzofuran
(DPBF) (A, 180 lM) was prepared by dissolving
4.9 mg of DPBF in 100 mL of THF. Stock solutions
of the porphyrins (B, 2 mM) were prepared by dissolv-
ing the desired amount of porphyrin in THF. Ten
microliters of (B) was added to 10 mL of THF to get a
2 lM stock solution of the porphyrins (C). The reaction
mixture was prepared by mixing 2 mL of (A) and 2 mL
of (C), so that the final concentration of DPBF is 90 lM
and that of porphyrin is 1 lM.
4.2.16. 5,20-Diphenyl-10,15-bis[4-(carboxymethylene-
oxy)-phenyl]-21,23-dithia-7,8-dihydroxychlorin (7). Dith-
iaporphyrin 24 (0.14 g, 0.16 mmol) was hydrolyzed by
1 M NaOH (40 mL) as described for the preparation
1
of 3 to give 0.10 g (76%) of 7 as a purple solid. H
NMR (400 MHz, CDCl3 + CD3OD): d 4.87 (2H, s),
4.90 (2H, s), 6.36 (2H, s), 7.29 (1H, dd, J = 6.1,
1.8 Hz), 7.30–7.40 (3H, m), 7.59 (2H, s), 7.65–7.93
(7H, m), 8.03–8.24 (5H, m), 8.39 (1H, d, J = 4.4 Hz),
8.42 (1H, d, J = 4.4 Hz), 9.13 (1H, d, J = 5.0 Hz), 9.17
(1H, d, J = 5.0 Hz), 9.47 (1H, d, J = 5.0 Hz), 9.51 (1H,
d, J = 5.0 Hz); HR ESI MS: m/z 831.2313 (calcd for
C48H34N2O8S2 + H, 831.1835).
The UV absorbance of the reaction mixture was ob-
tained before irradiation, using THF as a blank solu-
tion. The irradiation was carried out using a halogen
lamp with a water filter (400–850 nm). The light inten-
sity was 3 mW/cm2 and the reaction mixture was contin-
uously stirred during the irradiation. The progress of
reaction was monitored after every 2 min of irradiation
up to 10 min, using UV absorbance. 300 lL of the sam-
ple was used each time for UV measurement.
4.2.17. 5,20-Bis-phenyl-12,13-dimethoxy-21-thiaporphy-
rin (27). Cyclization of 2,5-bis(hydroxymethyl)-3,4-
dimethoxy-thiophene28 25(0.50 g, 2.5 mmol), pyrrole
(0.52 mL, 7.5 mmol), and benzaldehyde (0.42 mL,
4.2 mmol) with BF3Æetherate (0.052 mL, 0.41 mmol)
and TCBQ (1.7 g, 7.4 mmol) was performed as de-
scribed for the preparation of 13 to give 0.08 g (6%) of
27 as a purple solid. Mp: 159–161 ꢂC; 1H NMR
(500 MHz, CDCl3): d 4.73 (6H, s), 7.60–7.72 (6H, m),
8.06 (4H, d, J = 5.4 Hz), 8.56 (2H, d, J = 3.5 Hz), 8.71
(2H, s), 8.89 (2H, d, J = 3.5 Hz), 10.42 (2H, s); HR
ESI MS: m/z 540.2460 (calcd for C34H25N3O2S + H,
540.1746).
4.3.2. Determination of n-octanol/water partition coeffi-
cients at pH 7.4 phosphate buffer. The n-octanol/water
partition coefficients were determined at pH 7.4 using
the absorbance of the core-modified porphyrins. A
‘shake flask’ direct measurement29 with minor modifica-
tion was used. Individual porphyrins were dissolved in a
mixture of equal volumes of n-octanol and a pH 7.4
phosphate buffer, then placed in an ultrasound bath
for 30 min. The mixture was then left to settle for 4 h
and the partition coefficients determined by measuring
the absorbance of the core-modified porphyrins, using
an Ocean Optics USB4000 UV–vis spectrometer. Re-
sults were reported as logD7.4 values.
4.2.18. 5,20-Bis-phenyl-12,13-dihydroxy-21-thiaporphy-
rin (28). Mono-thiaporphyrin 27 (0.18 g, 0.33 mmol)
was demethylated with BBr3 (0.16 mL, 1.67 mmol) in
CH2Cl2 as described for the preparation of 14 to give
0.15 g (88%). The crude product was used without fur-
ther purification.
4.4. Biology
4.2.19. 5,20-Bis-phenyl-12,13-bis(ethoxycarboxylmethyl-
eneoxy)-21-thiaporphyrin (29). Mono-thiaporphyrin 28
(0.30 g, 0.59 mmol) was treated with ethyl bromoacetate
(0.70 mL, 5.9 mmol) and K2CO3 (0.81 g, 5.9 mmol) in
150 mL of acetone as described for the preparation of
15 to give 0.20 g (51%) of 29 as a dark purple solid.
4.4.1. Cells and culture conditions. Cells cultured from
the rodent mammary adenocarcinoma cell line
(R3230AC) were used. The cells were maintained on
60 mm diameter polystyrene dishes (Becton Dickinson
Labware, Franklin lakes, NJ) in 7 mL minimum essen-
tial medium (a-MEM) supplemented with 10% bovine

3182
E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
growth serum (HyClone, No.: SH3054103), 50 units/mL
penicillin G, 50 lg/mL streptomycin, and 1.0 g/mL fun-
gizone (complete medium). Cells were incubated at
37 ꢂC in 5% CO2 using an incubator (Sanyo MCO-
18AIC-UV). Passage was accomplished by aspirating
the culture medium, then adding a 1.0 mL solution con-
taining 0.25% trypsin and waiting for 4–5 min to remove
the cells from the dish’s surface. New culture dishes were
then seeded with the appropriate number of cells in
7.0 mL of complete medium. Cell counts were done
using a hematocytometer. The cell doubling time was
approximately 20 h.
ker (Lab-line, Barnstead International, IA). The well
plate’s lid was then removed and the wells exposed for
an hour to broadband visible light delivered at
3 mW cmÀ2 from a 60 W halogen light source, through
a 3.5 cm water filter (400–850 nm). Uniform illumina-
tion of the entire well plate was achieved by gently orbit-
ing the well plate on the shaker. After an hour of
irradiation, the clear medium was removed, and
190 lL of complete medium added to the wells. The cul-
tures were incubated at 37 ꢂC in 5% CO2, and in the
dark for 24 h, after which the cytotoxicity was deter-
mined by MTT assay and expressed as a percent of
the controls, cells exposed to light in the absence of
porphyrins.
4.4.2. Incubation of cell cultures with dithiaporphyrins. In
experiments to determine the porphyrin intracellular
accumulation, R3230AC cells were seeded on 96-well
plates at cell densities between 2.0 and 3.0 · 104 cells/
well in the complete medium and incubated at 37 ꢂC in
5% CO2 for 24 h. The porphyrins were dissolved in
DMSO at 2 mM. The stock solutions were diluted to
the appropriate concentrations with complete medium
immediately before the addition to cells. The porphyrin
samples were then added to the wells and incubated for
24 h. After incubation, the medium was removed and
the cell monolayer rinsed twice with a 0.9% NaCl solu-
tion. 190 lL of DMSO was then added to solubilize the
cells and the fluorescence from the porphyrins read
using a fluorescence multi-well plate reader (Molecular
Devices, SpectraMax M2 model) set at the appropriate
excitation and emission wavelengths. The intracellular
porphyrin concentrations were then determined from a
standard fluorescence curve obtained by dissolving por-
phyrin standards in DMSO. Results were expressed in
fmol/cell.
4.4.4. Fluorescence microscopy with thiaporphyrins.
R3230AC cells were seeded at cell densities between
2.0 and 3.0 · 104 cells/well in a 24-well plate containing
12 mm diameter coverslips in 1 mL of complete medium
and then incubated at 37 ꢂC in 5% CO2for 24 h. The
porphyrin samples, dissolved in medium, were then
added to the well plate at 2.0 · 10À5 M and incubated
again for 24 h. The medium was then removed and the
cell monolayer rinsed twice with 3 mL of complete med-
ium. After this the cover slide was immediately removed
and mounted on a slide and the images taken. The
images were captured using a Leica DMI4000B fluores-
cence microscope fitted with a QImaging Fast 1394 cam-
era and Qcapture processing software. The images were
modified for better visualization with Adobe Photoshop
Element 5.0.
4.4.5. Statistical analyses. Statistical analyses were per-
formed using the Student’s t-test for pairwise compari-
sons. A P value of <0.05 was considered significant.
The Hill (sigmoid Emax) equation was fitted to the data
to obtain IC50 values.
In experiments to determine the cytotoxicity of the por-
phyrins in either the dark (dark toxicity) or after light
exposure (phototoxicity), R3230AC cells were seeded on
96-well plates at cell densities between 1.0 and
1.5 · 104 cells/well in the complete medium. Cultures
were then incubated for 24 h at 37 ꢂC in 5% CO2, and then
the porphyrins, dissolved in the complete medium to the
appropriate concentrations, were added to the wells and
again incubated for 24 h. The medium was then removed
and the cell monolayer rinsed twice with 190 lL of a 0.9%
NaCl solution. Clear medium without phenol red and bo-
vine growth serum, was then added to the wells and the
well plates either kept in the dark (dark toxicity) or irradi-
ated (phototoxicity) for an hour. After this the clear med-
ium was removed and 190 lL of complete medium added.
The cultures were then incubated at 37 ꢂC in 5% CO2, for
24 h, after which the cytotoxicity was determined by MTT
assay and expressed as a percent of the controls. The con-
trols used in the dark toxicity tests were cells kept in the
dark and in the absence of porphyrins, while those used
in the phototoxicity tests were cells exposed to light in
the absence of porphyrins.
Acknowledgments
This research was supported by the Department of De-
fense [Breast Cancer Research Program] under award
number (W81XWH-04-1-0500). Views and opinions
of, and endorsements by, the author(s) do not reflect
those of the US Army or the Department of Defense.
References and notes
1. 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.
2. Dolmans, D. E.; Fukumura, D.; Jain, R. K. Nat. Rev.
Cancer 2003, 3, 380.
3. Sternberg, E. D.; Dolphin, D.; Bruckner, C. Tetrahedron
1998, 54, 4151.
4. Detty, M. R.; Gibson, S. L.; Wagner, S. J. J. Med. Chem.
2004, 47, 3897.
4.4.3. Irradiation of cultured cells. After 24 h of incuba-
tion with the porphyrins, the medium was removed,
and the cell monolayer was rinsed twice with 190 lL
of a 0.9% NaCl solution. Clear medium, was then added
to the wells and the well plate placed on an orbital sha-
5. You, Y.; Gibson, S. L.; Detty, M. R. Bioorg. Med. Chem.
2005, 13, 5968.
6. Stilts, C. E.; Nelen, M. I.; Hilmey, D. G.; Davies, S. R.;
Gollnick, S. O.; Oseroff, A. R.; Gibson, S. L.; Hilf, R.;
Detty, M. R. J. Med. Chem. 2000, 43, 2403.

E. J. Ngen et al. / Bioorg. Med. Chem. 16 (2008) 3171–3183
3183
7. Hilmey, D. G.; Abe, M.; Nelen, M. I.; Stilts, C. E.; Baker,
G. A.; Baker, S. N.; Bright, F. V.; Davies, S. R.; Gollnick,
S. O.; Oseroff, A. R.; Gibson, S. L.; Hilf, R.; Detty, M. R.
J. Med. Chem. 2002, 45, 449.
8. You, Y.; Gibson, S. L.; Hilf, R.; Davies, S. R.; Oseroff, A.
R.; Roy, I.; Ohulchanskyy, T. Y.; Bergey, E. J.; Detty, M.
R. J. Med. Chem. 2003, 46, 3734.
9. You, Y.; Gibson, S. L.; Hilf, R.; Ohulchanskyy, T. Y.;
Detty, M. R. Bioorg. Med. Chem. 2005, 13, 2235.
10. You, Y.; Daniels, T. S.; Dominiak, P. M.; Detty, M. R.
J. Porphyrins Phthalocyanines 2007, 11, 1.
11. You, Y.; Gibson, S. L.; Detty, M. R. J. Photochem.
Photobiol., B 2006, 85, 155.
19. Juzeniene, A.; Moan, J. Photodiagn. Photodyn. Ther. 2007,
4, 3.
20. Moan, J. J. Photochem. Photobiol. B 1990, 6, 343.
21. Kamat, J. P.; Devasagayam, T. P. Toxicology 2000, 155,
73.
22. Ichinose, S.; Usuda, J.; Hirata, T.; Inoue, T.; Ohtani, K.;
Maehara, S.; Kubota, M.; Imai, K.; Tsunoda, Y.;
Kuroiwa, Y.; Yamada, K.; Tsutsui, H.; Furukawa, K.;
Okunaka, T.; Oleinick, N. L.; Kato, H. Int. J. Oncol. 2006,
29, 349.
23. Chernyak, B. V.; Izyumov, D. S.; Lyamzaev, K. G.;
Pashkovskaya, A. A.; Pletjushkina, O. Y.; Anton-
enko, Y. N.; Sakharov, D. V.; Wirtz, K. W.;
Skulachev, V. P. Biochim. Biophys. Acta 2006,
1757, 525.
12. Lara, K. K.; Rinaldo, C. R.; Bruckner, C. Tetrahedron
Lett. 2003, 44, 7793.
13. Tong, T.-H.; Chien, L.-C. J. Polym. Sci., A: Polym. Chem.
2000, 38, 1450.
24. Morgan, J.; Oseroff, A. R. Adv. Drug Deliv. Rev. 2001, 49,
71.
14. Latos-Grazynski, L. In The Porphyrin Handbook; Kadish,
L. M., Smith, K. M., Guildard, R., Eds.; Academic Press:
San Diego, 2000; Vol. 2, p 391.
25. Hilf, R. J. Bioenerg. Biomembr. 2007, 39, 85.
26. Johnson, L. V.; Walsh, M. L.; Chen, L. B. Proc. Natl.
Acad. Sci. U.S.A. 1980, 77, 990.
15. Krieg, M. J. Biochem. Biophys. Methods 1993, 27, 143.
16. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
17. Akins, D. L.; Ozcelik, S.; Zhu, H. R.; Guo, C. J. Phys.
Chem. 1996, 100, 14390.
27. Stoner, E. J.; Cothron, D. A.; Balmer, M. K.; Roden, B.
A. Tetrahedron 1995, 51, 11043.
28. Agarwal, N.; Hung, C. H.; Ravikanth, M. Eur. J. Org.
Chem. 2003, 2003, 3730.
18. Kano, K.; Minamizono, H.; Kitae, T.; Negi, S. J. Phys.
Chem. A 1997, 101, 6118.
29. Sangster, J. Octanol–Water Partition Coefficients; Wiley:
New York, 1997.