Syntheses and cellular investigations of 173-, 152-, and 131-amino acid derivatives of chlorin e6

Article abstract of DOI:10.1021/jm2005139

A series of amino acid conjugates of chlorin e₆, containing lysine or aspartic acid residues in positions 17³, 15², or 13¹ of the macrocycle were synthesized and investigated as photosensitizers for photodynamic therapy of tumors. All three regioisomers were synthesized in good yields and in five steps or less from pheophytin a (1). In vitro investigations using human carcinoma HEp2 cells show that the 15 ²-lysyl regioisomers accumulate the most within cells, and the most phototoxic are the 13¹ regioisomers. The main determinant of biological efficacy appears to be the conjugation site, probably because of molecular conformation. Molecular modeling investigations reveal that the 17³-substituted chlorin e₆ conjugates are L-shaped, the 15² and 13¹ regioisomers assume extended conformations, and the 13¹ derivatives are nearly linear. It is hypothesized that the 13¹-aspartylchlorin e₆ conjugate may be a more efficient photosensitizer for PDT than the commercial currently used 15 ² derivative.

Full text of DOI:10.1021/jm2005139

ARTICLE
pubs.acs.org/jmc
3
2
1
Syntheses and Cellular Investigations of 17 -, 15 -, and 13 -Amino
Acid Derivatives of Chlorin e6
R. G. Waruna Jinadasa, Xiaoke Hu, M. Gra c- a H. Vicente, and Kevin M. Smith*
Louisiana State University, Department of Chemistry, Baton Rouge Louisiana 70803, United States
S Supporting Information
b
ABSTRACT: A series of amino acid conjugates of chlorin e ,
6
3
2
containing lysine or aspartic acid residues in positions 17 , 15 , or
1
1
3 of the macrocycle were synthesized and investigated as
photosensitizers for photodynamic therapy of tumors. All three
regioisomers were synthesized in good yields and in five steps or
less from pheophytin a (1). In vitro investigations using human
2
carcinoma HEp2 cells show that the 15 -lysyl regioisomers
accumulate the most within cells, and the most phototoxic are
1
the 13 regioisomers. The main determinant of biological efficacy
appears to be the conjugation site, probably because of molecular
conformation. Molecular modeling investigations reveal that the
3
2
1
1
7 -substituted chlorin e conjugates are L-shaped, the 15 and
6
1 1
1
3 regioisomers assume extended conformations, and the 13 derivatives are nearly linear. It is hypothesized that the 13 -
aspartylchlorin e conjugate may be a more efficient photosensitizer for PDT than the commercial currently used 15 derivative.
2
6
’
INTRODUCTION
(HPPH), lutetium(III) texaphyrin (LuTex), mono-L-aspartyl-
3
ꢀ6
chlorin e , and phthalocyanine 4 (Pc4).
Chlorins (dihydro-
6
Photodynamic therapy (PDT) is a binary cancer therapy that
porphyrins) are particularly promising photosensitizers for PDT
because of their intense absorptions above 640 nm; in particular,
chlorophyll a derivatives are inherently amphiphilic macrocycles
that have been extensively investigated and shown to have low
dark toxicities while being able to effectively generate singlet
relies on the selective uptake of a photosensitizer into cancer cells
followed by irradiation with red light, which produces singlet
oxygen and other reactive oxygen species.
1
ꢀ4
Highly cytotoxic
singlet oxygen readily reacts with electron-rich biomolecules in5
its vicinity such as unsaturated lipids, amino acids, and DNA.
7ꢀ9
oxygen upon light activation. Among these, HPPH and mono-
Because of the known limited diffusion of singlet oxygen through
tissues the PDT effects are largely localized to the photosensi-
tizer-containing cells, thus reducing potential damage to normal
cells in the vicinity of the tumor. The selectivity of the PDT
treatment depends both on the tumor-targeting ability of the
photosensitizer and the light used to activate it. An ideal
photosensitizer should have minimal toxicity in the dark, a high
quantum yield of triplet state formation in the presence of light,
high selectivity for tumor cells over normal cells, rapid clearance
from normal tissues, and a strong absorption peak within the
L-aspartylchlorin e have been investigated and are currently in
6
10ꢀ13
advanced-stage clinical trials for oncologic PDT applications;
both of these chlorophyll a based compounds have shown
superior PDT activity as well as rapid clearance from normal
tissue and decreased patient photosensitivity compared with
commercial hematoporphyrin derivative.
Chlorophyll a derivatives of the chlorin e series possess three
6
carboxylic side chains that can be derivatized in multiple ways to
produce novel amphiphilic photosensitizers for PDT investiga-
tions. For example, their conjugation to peptides, sugars, lipo-
“
therapeutic window” (600ꢀ800 nm) for optimal light pene-
14ꢀ19
proteins, and polyamines have been reported.
In particular,
tration through tissue.
amino acid residues have been found to improve the biological
effects of porphyrin-based compounds, and their nature and
position about the macrocycle can have a significant impact on
Commercial hematoporphyrin derivative is a porphyrin-based
photosensitizer that has been commercially developed and ap-
proved in several countries for the PDT treatment of melanoma,
early and advanced stage cancer of the lung, digestive tract,
5
,20,21
PDT efficacy.
In the present study, we synthesized and
3
2
1
3
ꢀ6
investigated 17 , 15 , and 13 amino acid conjugates of chlorin e
6
genitourinary tract, and Barrett’s esophagus.
However, it has
(
Figure 1) to evaluate the effect of the nature, conjugation site,
only a weak absorption within the therapeutic window and often
induces skin photosensitivity in patients. For these reasons, several
second-generation photosensitizers bearing stronger and red-
shifted absorption within the therapeutic window have been
investigatedinPDT;theseincludetetra(meta-hydroxyphenyl)chlor-
in, (mTHPC), 2-[1-hexyloxyethyl]-2-devinyl-pyropheophorbide a
and position of the amino acid on the PDT efficacy of the
conjugates in vitro, using human carcinoma HEp2 cells. The
Received: April 27, 2011
Published: September 21, 2011
r 2011 American Chemical Society
7464
dx.doi.org/10.1021/jm2005139
|J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
ARTICLE
3
2
1
6
Figure 1. 17 -, 15 -, and 13 -regioisomers of “R”-substituted chlorin e .
a
Scheme 1
a
Reaction conditions: (a) TFA:H
°C, 4 h, 89%; (d) TFA, CH Cl , thioanisole, 0 °C to rt, 12 h, 72%; (e) 5% H
LiI, EtOH, reflux, 24 h, 21%; (h) 5% H SO /MeOH, rt, 12 h, 99%.
2
O 4:1, 0 °C, 1 h, 90%; (b) DCC, DMAP, H-Lys(Boc)-OMe HCl, DIEA, CH
2
CL
2
, rt, 14 h, 52%; (c) NaOMe, THF,
3
0
2
2
2 4
SO /MeOH, rt, 12 h, 96%; (f) NaOMe, THF, 0 °C, 4 h, 98%; (g) 18 equiv
2
4
structure of mono-L-aspartylchlorin e , bearing the aspartyl
conjugation, rather than the presence of Pd(II) and even the
nature of amino acid, is the major determinant of phototoxicity.
6
2
3
residue in position 15 (rather than in 17 as initially proposed)
22
has recently been established. This study led us to design new
3
1
syntheses of the 17 and 13 aspartyl regioisomers of mono-L-
’
RESULTS AND DISCUSSION
aspartylchlorin e as well as to investigate the corresponding
6
3
cationic lysine derivatives, which could potentially show stronger
interactions with negatively charged biological molecules and
1. Syntheses. The synthetic route to the 17 -lysyl conjugate of
chlorin e (3) is shown in Scheme 1. Pheophytin a (1) was
6
23
plasma membranes and consequently enhanced PDT efficacy.
obtained by extraction from Spirulina pacifica alga, an ideal
source for chlorophyll a because of the complete absence of
chlorophyll b. The selective hydrolysis of the phytyl ester group
of pheophytin a (1) using aqueous TFA produced pheophorbide
Furthermore, we investigated the introduction of a spacer group,
ethylene diamine, and of metalation, via insertion of palladium-
(
II). The photosensitizing properties of porphyrin derivatives are
known to be modulated by inner coordinated metal ions and
a in high yield without affecting the β-keto ester of the isocyclic
24ꢀ26
29,30
associated axial ligands.
In particular, Pd(II) complexes
ring.
This ring serves as a natural protecting group for the
1
2
have been shown to be efficient generators of singlet oxygen,
13 - and 15 -positions during the coupling reactions. Selective
hydrolysis of the phytyl ester produced a free 17 -carboxylic acid
3
and the palladium(II)-bacteriopheophorbide is currently being eval-
27,28
uated in clinical trials for the treatment of prostate cancer.
The work described herein indicates that the site of amino acid
group, which was activated with DCC/DMAP and coupled with
H-lysine(OtBu) methyl ester to form the lysine(OtBu) methyl
7
465
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
ARTICLE
a
Scheme 2
a
Reaction conditions: (a) DCC, DMAP, CH
2
Cl
2
, rt, 2 h; H-Asp(Boc)
2
HCl or H-Lys(Boc)-OMe HCl, DIEA, CH
2
Cl
2 2 2 2 2
, rt, 14 h; (b) CH N , CH Cl ,
3
3
3
0 min; (c) TFA, CH
2
Cl
2
, thioanisole, 0 °C to rt, 12 h (7a, 45%; 7b, 77%); (d) Pd(OAc)
2
, THF, 60 °C, 3 h (8a, 98%; 8b, 99%).
a
Scheme 3
(4) by transesterification of the ester group using sulfuric acid in
31
methanol, as previously reported, followed by isocyclic ring-
opening, as shown in Scheme 1. An optimized yield of 98% of
chlorin e trimethyl ester was obtained when 1.1 equiv of freshly
6
prepared sodium methoxide in THF were used in the ring-
opening reaction. Chlorin e (5) was obtained by hydrolysis of
6
the three methyl esters using an excess of LiI in ethyl acetate
33
under reflux conditions for 24 h. Alternative hydrolysis under
basic conditions using NaOH and LiOH gave lower yields of
1
chlorin e . Under acidic conditions the 13 -carboxylic acid of
6
34
chlorin e is deactivated, allowing for the selective methylation
of the 15 -and 17 -carboxylic acids with 5% H SO /MeOH to
give chlorin e dimethyl ester (6) in quantitative yield. Chorin
e6 (5) was used as starting material for the preparation of the 15
6
2
3
2
4
2
2
6
2
amino acid derivatives of chlorin e (as shown in Scheme 2),
6
whereas the chlorin (6) and methyl pheophorbide a (4) were
1
used to prepare the 13 amino acid derivatives of chlorin e (as
6
shown in Schemes 3 and 4). 2
2
2
We have previously shown that the 15 -carboxylic acid of
chlorin e is the most reactive, regardless of the coupling reagent
6
employed. Therefore, upon activation of chlorin e (5) with 1
6
equiv of DCC/DMAP, followed by addition of protected amino
35
acid, reaction with diazomethane and deprotection under
2
acidic conditions, the 15 -aspartyl (7a) and -lysyl (7b) deriva-
tives of chlorin e were obtained in overall 36% and 56% yields,
6
respectively (Scheme 2). The use of excess DCC significantly
3
2
reduced the yields obtained due to the formation of the 17 ,15 -
diamino acid conjugate as a side product.
a
Reaction conditions: (a) L-Asp(OtBu)2 HCl, DIEA, CH Cl , HOBt,
3
2
2
TBTU, DMF, rt, 14 h, 66%; (b) TFA, CH
h, 88%; (c) β-Ala(OtBu) HCl, DIEA, CH
1
Asp(OtBu)2 HCl, DIEA, CH Cl , HOBt, TBTU, DMF, rt, 12 h, 87%;
(
2
Cl
Cl
2 h, 68%; (d) TFA, CH Cl , thioanisole, 0 °C to rt, 12 h, 95%; (e) L-
2
, thioanisole, 0 °C to rt, 12
2
Insertion of palladium into the chlorin ring was attempted
before the deprotection step to simplify the purification process;
however, low yields of the target metallochlorins were obtained.
Therefore, palladium insertion was performed after deprotec-
tion, using either palladium(II) acetate or palladium(II) chloride.
Quantitative yields for the metal insertion reaction were obtained
using 1.2 equiv of palladium(II) acetate in THF at 40 °C to
produce the corresponding Pd(II) complexes (8). As expected,
3
2
, HOBt, TBTU, DMF, rt,
2
2
3
2
2
2 2
f) TFA, CH Cl , thioanisole, 0 °C to rt, 12 h, 97%.
32 1
ester pheophorbide a (2) in 52% yield. H NMR spectroscopy
demonstrated the existence of the isocyclic ring and new peaks
for the lysine residue. Subsequent isocyclic ring-opening with
sodium methoxide in THF produced the 17 -lysyl(Boc) chlorin
e6 TME in high yield (89%); deprotection of the amine group of
the lysine side chain with TFA in CH Cl followed and yielded
1
significant blue-shifts in the UVꢀvis spectrum along with H
3
1
3
NMR and mass spectrometry confirmed the insertion of palla-
dium. Heavy metal atomsinthe coreofthetetrapyrroles have been
shown to facilitate intersystem crossing and increase the produc-
2
2
3
36,37
the 17 -lysylchlorin e (3) in 30% overall yield from pheophytin
a (1).
tion of singlet oxygen due to the inner heavy atom effect.
6
1
Two routes were used to synthesize the 13 -amino acid con-
jugates: (1) Selective esterification of 17 - and 15 -carboxylic acids
2
1
3
2
To synthesize the 15 and 13 amino acid derivatives of chlorin
e6, pheophytin a (1) was converted into methyl pheophorbide a
of chlorin e to produce the monocarboxylic acid (6) followed by
6
7
466
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
ARTICLE
a
Scheme 4
a
Reaction conditions: (a) NH C H NH , CHCl , rt, 24 h, 91%; (b) (Boc)Lys(Boc)-OH, DIEA, CH Cl , HOBt, TBTU, DMF, rt, 12 h, 78%; (c) TFA,
2
2
4
2
3
2
2
CH Cl , thioanisole, 0 °C to rt, 12 h, 81%.
2
2
3
4
coupling (Scheme 3) and (2) isocyclic ring-opening of methyl
1
pheoporbide a (4) with a nucleophile (Scheme 4). The 13 -
carboxylic acid of chlorin (6) was activated using HOBt/TBTU
38
under basic conditions and coupled with L-aspartic di(tert)butyl
ester. With DCC/DMAP, a lower yield was obtained for the
coupling reaction. Subsequent deprotection of both carboxylic
acids in the aspartic chain provided (9) in 57% overall yield from
1
chlorin e (Scheme 3). Coupling of the 13 -carboxylic acid of
6
chlorin (6) with β-alanine tert-butyl ester followed by deprotection
1
using TFA gave the 13 -β-alanylchlorin e dimethyl ester (10) in
6
64% overall yield. Aspartic di(tert)butyl ester was then coupled
under similar conditions, followed by removal of the tert-butyl
protecting group to give (11) in 54% yield from (10). β-Alanine
provides a 4-atom link between the aspartic acid and chlorin e6
dimethyl ester and allows for direct comparison with the lysine
derivative 13 (see Scheme 4). The use of a linker significantly
increased the yield of the coupling reaction (from 66 to 87%), and it
20,39
can also affect the biological properties of the photosensitizer.
Figure 2. Energy minimized conformations in gas phase for chlorin e
6
The nucleophilic ring-opening of the isocyclic ring in pheo-
1
+
derivatives (a) 3, (b) 7b, (c) 13, and (d) [13 LysCe
6
TME] . Optimiza-
phorbide a with ethylenediamine and ethanolamine has already
25,40
tion by energy was carried out at HF/6-31G level.
been previously reported.
We took advantage of this high
yielding isocyclic ring-opening reaction to develop a potentially
2
high yielding synthetic route to novel chlorin e photosensitizers
to the macrocyclic plane (Figure 2a), while in the 15 -lysyl
derivative 7b it forms approximately a 120° angle (Figure 2b).
On the other hand, the lysine residue extends away from the
6
directly from pheophytin a (1) in four steps. Nucleophilic ring-
opening of the isocyclic ring with ethylenediamine produced
molecule (12) in 91% yield. Subsequent coupling with protected
lysine followed by deprotection gave the desired product (13) in
four steps and in 55% overall yield from pheophytin a (1). No
absorption quenching of the chlorin macrocycle was observed
upon conjugation with the amino acid lysine in comparison with
the aspartic acid derivative (see Figure S1 of the Supporting
Information).
1
macrocycle in the 13 -lysyl derivatives, with and without the
short spacer (Figures 2c,d). Consequently, in the L-shape
derivative, the amino
acid shelters one face of the chlorin ring, while in the case of the
15 - and 13 -lysyl derivatives, it extends away from the macro-
cycle, resulting in a nearly linear conformation for the 13
3
conformation of the 17 -lysylchlorin e
6
2
1
1
derivatives. The use of a short linker and the presence of two
(rather than one) positive charges, as a result from conjugation to
the C-terminus rather than the N-terminus of the amino acid, do
not appear to have a significant effect upon the preferred
conformation; the main determinant of molecule conformation
is the site of substitution. While the minimum energy conforma-
tions found in water were similar to those in the gas phase, it is
possible that in water intermolecular forces (such as πꢀπ
3
2
2
. Molecular Modeling. Conformation analyses for 17 -, 15 -,
1
and 13 -lysylchlorin e derivatives 3, 7b, and 13 were performed
6
in both the gas phase and in water at the HF/6-31G level. These
calculations were based on the atom coordinates from the X-ray
2
22
structure of 15 -aspartylchlorin e6 tetramethyl ester. The
minimum energy conformations found for compounds 3, 7b,
and 13 in the gas phase, shown in Figure 2aꢀc, respectively, were
very similar to those found in water phase (Figure S2 of the
stacking) predominate over the intramolecular interactions, in
1
Supporting Information). In addition, the conformation of a
particular for the linear 13 -lysylchlorin e
6
derivatives.
1
monocationic 13 -lysylchlorin e derivative without a linker was
3. Cell Culture Studies. 3.1. Time-Dependent Cellular Uptake.
The results obtained for the time-dependent uptake of chlorin e6
and its derivatives at a concentration of 10 μM in human HEp2
6
also investigated in the gas phase (Figure 2d). As expected, the
3
lysine residue in the 17 -lysyl derivative 3 is nearly perpendicular
7
467
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
ARTICLE
Figure 5. Phototoxicity of chlorin e (5, black line) and its derivatives
6
3
2
1
7 -LysChlorin e TME (3, brown line), 15 -AspChlorin e DME (7a,
6 6
2 2
Figure 3. Time-dependent uptake of chlorin e (5, black line) and its
6
3
2
light-blue line), 15 -LysChlorin e TME (7b, red line), 15 -AspPdChlor-
6
derivatives 17 -LysChlorin e TME (3, brown line), 15 -AspChlorin
6
2
2
2
in e
6
DME (8a, green line), 15 -LysPdChlorin e
6
TME (8b, blue line),
DME (9, maroon line), 13 -βAlaAspChlorin e DME
DME (13, pink line) toward
e
6
DME (7a, light-blue line), 15 -LysChlorin e
6
TME (7b, red line), 15 -
1
1
2
13 -AspChlorin e
6
6
AspPdChlorin e
6
DME (8a, green line), 15 -LysPdChlorin e TME (8b,
blue line), 13 -AspChlorin e DME (9, maroon line), 13 -βAlaAsp-
Chlorin e DME (11, purple line), and 13 -EDLysChlorin e DME (13,
6 6
6
1
1
1
(11, purple line), and 13 -EDLysChlorin e
6
6
2
1
HEp2 cells using 1 J/cm light dose and the Cell Titer Blue assay.
pink line) at 10 μM by HEp2 cells.
Table 1. Cytotoxicity (HEp2 cells) for Chlorin e and Its
6
Derivatives Using the Cell Titer Blue Assay
dark toxicity phototoxicity
compd
(5)
(IC50, μM)
(IC50, μM)
ratio
chlorin e
6
>400
>400
373.1
>400
324.8
>400
284.6
383.9
268.4
20.8
26.2
4.0
>19.2
>15.3
93.4
3
1
1
1
1
1
1
1
7 -LysChlorin e
6
TME (3)
2
5 -AspChlorin e
6
DME (7a)
2
5 -LysChlorin e
6
TME (7b)
28.8
16.7
3.3
>13.9
19.4
2
5 -AspPdChlorin e
6
DME (8a)
2
5 -LysPdChlorin e
6
TME (8b)
>121.2
466.6
468.2
200.3
1
3 -AspChlorin e
6
DME (9)
0.61
0.82
1.34
1
3 -βAlaAspChlorin e
6
DME (11)
Figure 4. Dark toxicity of chlorin e
6
(5, black line) and its derivatives
1
3
2
13 -EDLysChlorin e DME (13)
6
1
7 -LysChlorin e
6
TME (3, brown line), 15 -AspChlorin e
6
DME (7a,
2
2
light blue line), 15 -LysChlorin e
in e DME (8a, green line), 15 -LysPdChlorin e TME (8b, blue line),
1
6
TME (7b, red line), 15 -AspPdChlor-
2
2
extended conformation of the 15 -lysylchlorin e derivative 7b
6
6
6
1
1
3 -AspChlorin e
6
DME (9, maroon line), 13 -βAlaAspChlorin e
6
DME
(Figure 2b) might be the most favored for penetration across the
1
(
11, purple line), and 13 -EDLysChlorin e
6
DME (13, pink line) toward
plasma membrane, compared with the L-shape and linear conforma-
2
3
1
HEp2 cells using 1 J/cm light dose and the Cell Titer Blue assay.
tions of the 17 - and 13 -lysyl derivatives (Figures 2a,c), respectively.
The presence of a chelated palladium ion, as well as the more
linear structure of the 13 -lysyl substituent might lead to
1
cells are shown in Figure 3 (also see Figure S3 of the Supporting
Information). All amino acid conjugates of chlorin e were
enhanced πꢀπ stacking of the macrocycles, thereby decreasing
6
readily taken up by cells and showed uptake kinetics similar to
cellular uptake.
2
those of unconjugated chlorin e . Interestingly, the 15 -lysylchlorin
3.2. Cytotoxicity. The dark cytotoxicity and phototoxicity of
6
e6 derivatives 7b and its palladium(II) complex 8b accumulated
chlorin e and its derivatives was evaluated in HEp2 cells exposed
6
to a much higher extent than all other compounds at all time
to increasing concentrations of each compound up to 400 μM;
points studied. In comparison with chlorin e , the lysyl deriva-
the results are shown in Figures 4 and 5, respectively, and are
6
tives 7b and 8b showed 18-fold and 4-fold higher cellular uptake,
summarized in Table 1. Chlorin e and its lysyl derivatives 3, 7b,
6
respectively, after 24 h. The observed high uptake for derivatives
and 8b were found to be nontoxic in the dark up to the highest
concentration (400 μM) investigated. All other amino acid
2
7
b and 8b is probably due to the lysine residue in position 15
because the corresponding aspartyl derivatives 7a and 8a accu-
mulated to a significantly lower extent within cells. Presumably,
the stronger interactions between the positively charged lysine
derivatives (compared with the corresponding aspartyl de-
rivatives) with the negatively charged plasma membrane leads
derivatives showed very low dark cytotoxicities, with IC
>
50
1
320 μM except for the 13 -chlorin e derivatives 9 and 13, which
6
showed IC50 of 285 and 268 μM, respectively. However, upon
2
exposure to a low light dose (1 J/cm ), all chlorin e derivatives
6
were found to be highly toxic to HEp2 cells (Figure 3). The most
20,23
1
to enhanced cellular uptake.
residue in position 17 , as in derivative 3, showed a dramatic
On the other hand, a lysine
phototoxic were the 13 -chlorin e derivatives 9, 11, and 13, with
6
3
estimated IC50 values of 0.61, 0.82, and 1.34 μM, respectively.
Among these, the most promising aspartyl derivatives for PDT
applications are compounds 9 and 11 because they have
the highest dark cytotoxicity/phototoxicity ratio > 466:1. The
decrease in cellular uptake compared with the same residue at
2
position 15 , suggesting that the molecule conformation plays an
important role on the mechanism of cellular uptake. Indeed the
7
468
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
ARTICLE
Table 2. Major (++) and Minor (+) Subcellular Sites of
Localization for Chlorin e and Its Derivatives in HEp2 Cells
6
compd
lysosomes ER Golgi mitochondria
chlorin e
6
(5)
+
+
++
++
++
++
+
ꢀ
+
ꢀ
ꢀ
+
3
1
1
1
1
1
1
1
1
7 -LysChlorin e
6
TME (3)
2
5 -AspChlorin e
6
DME (7a)
TME (7b)
++
++
++
+
ꢀ
+
2
5 -LysChlorin e
6
ꢀ
ꢀ
ꢀ
++
ꢀ
ꢀ
2
5 -AspPdChlorin e
6
DME (8a)
++
ꢀ
2
5 -LysPdChlorin e
6
TME (8b)
++
1
3 -AspChlorin e
6
DME (9)
+
++ ++
++ ++
++ ++
1
3 -βAlaAspChlorin e
6
DME (11)
++
++
1
3 -EDLysChlorin e
6
DME (13)
1
presence of the β-alanine spacer between the 13 carbonyl group
and the aspartic acid residue seems to have only a small effect,
slightly decreasing compound cytotoxicity. On the other hand,
2
the 15 -aspartylchlorin e derivative 7a was less phototoxic than
6
1
its 13 regioisomer 9 by approximately 7-fold, and the introduc-
tion of palladium(II) further reduced its phototoxicity. The
3
2
positively charged 17 - and 15 -lysylchlorin e derivatives 3
6
and 7b were the least phototoxic, and the introduction of
2
palladium increased the phototoxicity of the 15 derivative by
about 10-fold. These results show for the first time that the
phototoxicity of amphiphilic conjugates of chlorin e depends
6
mainly on the site of conjugation, probably as a result of its effect
on molecular conformation; the nature of the amino acid, the
molecule overall charge, and the presence of palladium(II) also
affect phototoxicity, but apparently to a smaller extent. Our
results suggest that the most extended, nearly linear conforma-
1
tion of the 13 regioisomers probably facilitates binding to
specific biological substrates, enhancing their toxic effect.
3
.3. Intracellular Localization. The preferential sites of sub-
cellular localization of this series of chlorin e derivatives were
6
evaluated by fluorescence microscopy upon exposure of HEp2
cells to 10 μM compound concentrations for 18 h. Figure S4 of
the Supporting Information shows the fluorescence pattern
observed for all compounds, and Table 2 summarizes their main
sites of subcellular localization. Overlay experiments using the
organelle specific fluorescence probes BODIPY Ceramide
(
Golgi), LysoSensor Green (lysosomes), MitoTracker Green
mitochondria), and ER Tracker Blue/White (ER) were con-
1
Figure 6. Subcellular localization of conjugate 13 AspCe
6
DME (9) in
(
HEp2 cells at 10 μM for 18 h (a) phase contrast, (b) overlay of conjugate
9 compound and phase contrast, (c) ER Tracker Blue/White fluores-
cence, (e) MitoTracker Green fluorescence, (g) BoDIPY Ceramide,
i) LysoSensor Green fluorescence, and (d,f,h,j) overlays of organelle
tracers with compound fluorescence. Scale bar: 10 μm.
ducted to evaluate the preferential sites of compound localization,
1
as seen in Figures 6, 7, and 8 for the most phototoxic 13 -
regioisomers, respectively, and Figures S5ꢀS10 of the Support-
(
ing Information. All chlorin e derivatives 3, 7a, 7b, 8a, 8b, 9, 11,
6
and 13 were found to localize in the lysosomes and the ER
(
1
Table 2). This is not surprising because the structurally related
5 -aspartylchlorin e is a known lysosomal localizer, and HPPH
6
2
’ CONCLUSIONS
(
2-[1-hexyloxyethyl]-2-devinyl-pyropheophorbide a) localizes
A series of eight chlorin e derivatives, conjugated with either
aspartic acid or lysine residues at positions 17 , 15 , or 13 of the
chlorin macrocycle, have been synthesized in good yields from
6
41
3
2
1
preferentially in the ER. Photodamage to the ER and/or
lysosomes has been shown to lead to activation of apoptotic
pathways.
4
1ꢀ43
1
Furthermore, all the 13 -chlorin e6 regioi-
pheophytin a (1). In comparison with chlorin e (5), all amino acid
6
somers 9, 11, and 13 were also found in Golgi, and in addition
the most phototoxic derivative 9 also localized in mitochondria;
presumably, the photodamage effect to multiple organelles
derivatives readily accumulated within human HEp2 cells, and in
particular the 15 -lysyl derivatives were taken up by cells to a
2
significantly higher extent than were all other regioisomers. On the
1
2
caused by the 13 derivatives can trigger various apoptotic
other hand, the metal-free 15 -lysylchlorin e derivative (7b)
6
3
pathways, leading to more effective cell destruction. The multi-
showed the least phototoxicity, followed by the 17 -lysyl regioisomer
(3); insertion of palladium into 15 -lysychlorin e (7b) increased
1
2
ple sites of intracellular localization observed for the 13 -chlorin
6
e6 derivatives might again be due to their linear conformations
that facilitate their binding to various intracellular sites.
phototoxicity by approximately 9-fold. However, the most photo-
toxic compounds were found to be the 13 -regioisomers, bearing
1
7
469
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
ARTICLE
1
1
Figure 8. Subcellular localization of conjugate 13 ED-LysCe
(
6
DME
Figure 7. Subcellular localization of conjugate 13 β-Ala-AspCe
(
6
DME
13) in HEp-2 cells at 10 μM for 18 h (a) phase contrast, (b) overlay of
conjugate 11 and phase contrast, (c) ER Tracker Blue/White fluores-
cence (e) MitoTracker Green fluorescence, (g) BoDIPY Ceramide,
i) LysoSensor Green fluorescence, and (d,f,h,j) overlays of organelle
11) in HEp-2 cells at 10 μM for 18 h (a) phase contrast, (b) overlay of
conjugate 11 and phase contrast, (c) ER Tracker Blue/White fluores-
cence (e) MitoTracker Green fluorescence, (g) BoDIPY Ceramide,
i) LysoSensor Green fluorescence, and (d,f,h,j) overlays of organelle
tracers with compound fluorescence. Scale bar: 10 μm.
(
(
tracers with compound fluorescence. Scale bar: 10 μm.
’
EXPERIMENTAL SECTION
either an aspartic acid or a lysine residue directly conjugated to
position-13 of the chlorin macrocycle or connected via a β-alanine
or ethylene diamine spacer. The most phototoxic compound of
1
. Chemistry. All air and moisture sensitive reactions were per-
formed in dried and distilled solvents under an argon atmosphere. All
solvents and reagents were purchased from commercial sources, unless
otherwise stated. Silica gel 60 (230 ꢁ 400 mesh, Sorbent Technologies)
was used for column chromatography. Analytical thin-layer chromatog-
raphy (TLC) was carried out using polyester backed TLC plates 254
1
this series, and the most promising for PDT applications, is 13 -
aspartylchlorin e (9). Molecular modeling calculations show that
6
1
the 13 -regioisomersassume extended, nearly linear conformations
that might facilitate their binding to multiple intracellular compo-
nents and subsequent photodamage to multiple cellular sites.
Therefore, the site of amino acid conjugation at the chlorin e6
macrocycle is a major determinant of compound phototoxicity, and
(precoated, 200 μm) from Sorbent Technologies. NMR spectra were
1
recorded on an AV-400 spectrometer (400 MHz for H, 100 MHz for
1
3
C). The chemical shifts are reported in δ ppm using the following
partially deuterated solvents as internal references: CD Cl 5.32 ppm -
1
we hypothesize that the 13 -regioisomer of mono-L-aspartylchlorin
2
2
1
13
1
13
e (9) might be a more efficient photosensitizer for PDT than is the
( H), 54 ppm ( C); DMSO-d 2.49 ppm ( H), 39.5 ppm ( C);
6
2
1
13
1
commercial regioisomer (15 -mono-L-aspartylchlorin e ).
3 3
CH OH-d 4.78 ppm ( H), 49.0 ppm ( C); CDCl 7.26 ppm ( H),
6
7
470
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
ARTICLE
1
3
1
13
7
7.16 ppm ( C); (CH
3
)
2
CO 2.50 ppm ( H), 29.84 ppm ( C).
and 1 mL of TFA were added, and the reaction mixture was allowed to
stir overnight. The reaction mixture was evaporated several times with
diethyl ether to remove TFA. The resulting precipitate was washed
several times with diethyl ether to remove thioanisole. The precipitate
Electronic absorption spectra were measured on an Agilent 8453 UV/
vis spectrophotometer. Mass spectra were obtained on a Bruker Omni-
flex MALDI time-of-flight mass spectrometer. Spirulina pacifica alga was
purchased as a spray-dried powder from Cyanotech, Hawaii. Pheophytin
a (1) was extracted from Spirulina pacifica alga as previously published,
and its spectroscopic characterization agreed with the published data.
All compounds synthesized were purified and isolated in g95% purity,
as evidenced by analytical TLC in at least two solvent systems and
confirmed by the absence of extraneous tetrapyrrole resonances in H
was dissolved in CH
with 10% NaHCO to remove last traces of TFA. The organic layer was
dried over anhydrous Na SO . Solvent was evaporated, and 43 mg
(0.056 mmol), 72% yield, of 17 -lysylchlorin e
Cl and washed three times with H O and once
2 2
2
3
2
9
2
4
3
TME (C43
H
N
O
)
6
54
6
7
ꢀ
1
ꢀ1
was obtained. UVꢀvis (acetone): λmax (ε/M cm ) 664 (65000), 608
1
1
3 2
(4871), 528 (4461), 500 (16100), 400 (213000). H NMR [(CD ) CO,
13
and C NMR spectra. Energy minimization of all compounds was
performed in the framework of HartreeꢀFock. The restricted Hartreeꢀ
Fock functional was used at the 6-31G level. All structures were optimized
without any symmetry constrains. The solvent effects were accounted for
using the Polarizable Continuum Model (PCM). All calculations were
400 MHz]: δ 9.79 (s, 1H), 9.63 (s, 1H), 9.03 (s, 1H), 8.17 (dd, J = 17.87,
11.58 Hz, 1H), 7.59 (d, J = 7.6 Hz, 1H), 6.38 (d, J = 17.87, 1.33 Hz, 1H),
6.12 (d, J = 11.61, 1.36 Hz, 1H), 5.37 (m, 2H), 4.63 (q, J = 7.22 Hz, 1H),
4.55 (dd, J = 10.34, 1.95 Hz, 1H), 4.44 (m, 1H), 4.25 (s, 3H), 3.76 (m,
2H) 3.75 (s, 3H), 3.69 (s, 3H), 3.55 (s, 3H) 3.48 (s, 3H), 3.25 (3H, s),
3.09 (m, 2H), 2.67 (m, 1H), 2.30 (m, 1H), 2.20 (m, 1H), 1.85ꢀ1.62 (br,
3H), 1.78 (d, J = 7.24, 3H), 1.68 (t, J = 7.58, 3H), 1.51 (m, 2H), 1.39 (m,
performed using the Gaussian 09 program package.
3
1
7 Lysyl(Boc)OMe-pheophorbide a (2). Pheophytin a (1) (250 mg,
3
0
.29 mmol) was selectively hydrolyzed to the 17 carboxylic acid without
2H), ꢀ1.33 (s, 1H), ꢀ1.53 (s, 1H). HRMS (MALDI-TOF) m/z
1
31
+
affecting the 13 carbomethoxy group, as previously reported. After
767.442 [M + H] , calcd for C43
H
55
N
6
O
7
767.413.
evaporation of solvent, 155 mg (0.26 mmol), 90% of pheophorbide a
Chlorin e (5). Chlorin e TME was prepared as previously published,
6 6
(
C H N O ) was obtained. The spectroscopic data obtained for the
and its spectroscopic characterization agreed fully with the published
data. Chlorin e TME (50 mg, 0.078 mmol) was dissolved in anhy-
6
drous ethyl acetate (10 mL) under argon. Lithium iodide (124 mg, 0.94
mmol) was added. The reaction mixture was refluxed for 48 h under
argon. The reaction mixture was diluted with water and adjusted to pH 3
3
5 36 4 5
2
2
compound agreed with published data. Pheophorbide a (100 mg; 0.169
mmol) was dissolved in 10 mL of CH Cl . Then DCC (42 mg) and
DMAP (10 mg) were added and left for 15 min. H-Lys(Boc)OMe HCl
2
2
3
(
2 2
60 mg) and DIEA (0.035 mL) were dissolved in 5 mL of CH Cl ,
added to the reaction mixture, and stirred for 4 h. [Note: this reaction is
favored by dilute conditions; concentrated conditions favor the forma-
tion of the anhydride (bispheophorbide) and will slow the reaction].
The reaction mixture was washed with 10% citric acid, water, and brine,
2 2
with aqueous citric acid and then washed with CH Cl . The solution was
evaporated, redissolved in acetone, and evaporated several times. The
solid was washed with water and then dried under vacuum. The residue
was dissolved in methanol and purified on a Sephadex LH-20 column to
dried over anhydrous Na
2
SO
4
, and purified on a silica gel column (10%
6 36 4 6
yield 10 mg, 0.017 mmol, 21% yield of chlorin e (C34H N O ).
ꢀ1
ꢀ1
acetone in CH Cl ). After the major brown band was eluted from the
UVꢀvis (MeOH): λ (ε/M cm ) 666 (45271), 610 (8706), 558
2
2
max
column, the solvent was evaporated and the solid was dissolved in ethyl
acetate and filtered (DCC precipitated in ethyl acetate.) Evaporation of
(7835), 530 (9721), 502 (15525), 402 (145100). MS MALDI: m/z 597
+
1
3 2
(M + H) . H NMR [(CD ) CO, 400 MHz]: δ 9.73 (s, 1H), 9.51 (s,
the filtered ethyl acetate gave 74 mg, 0.089 mmol, 52% yield of lysyl-
1H), 9.04(s, 1H), 8.05 (dd, J = 17.86, 11.61 Hz, 1H), 6.28 (dd, J = 17.85,
1.26 Hz, 1H), 6.02 (dd, J = 11.57, 1.37 Hz, 1H), 5.60 (d, J = 18.94 Hz,
1H), 5.40 (d, J = 18.94 Hz, 1H), 4.65 (m, 1H), 4.55 (m, 1H), 3.64 (m,
2H), 3.63 (s, 3H), 3.42 (s, 3H), 3.15 (s, 3H), 2.72 (m, 1H), 2.35 (m,
1H), 2.26 (m, 1H), 2.06 (m, 1H), 1.75 (d, J = 7.5 Hz, 3H), 1.64 (t, J = 7.5
Hz, 3H), ꢀ1.6 (s, 2H)
1
(
Boc)OMe-pheophorbide a (C H N O ). H NMR [(CD ) CO, 400
47
58
6
8
3 2
MHz]: δ 9.12 (s, 1H), 8.71 (s, 1H), 8.70 (s, 1H), 7.64 (dd, J = 17.8, 11.63
Hz, 1H), 7.20 (d, J = 7.2 Hz, 1H), 6.33 (s, 1H), 6.02 (d, J = 17.85 Hz, 1H),
5
(
7
(
.93 (d, J = 11.61 Hz, 1H), 5.77 (s, 1H), 4.65 (m, 1H), 4.41 (m, 1H), 4.21
m, 1H), 3.91 (s, 3H), 3.59 (s, 3H), 3.47 (s, 3H), 3.19 (s, 3H) 3.09 (q, J =
.5 Hz, 2H), 2.92 (3H, s), 2.83 (m, 2H), 2.66 (m, 1H), 2.48 (m, 1H), 2.25
m, 1H), 1.85ꢀ1.62 (br, 4H), 1.86 (d, J = 7.24, 3H), 1.37 (t, J = 7.58, 3H),
.43 (m, 2H), 1.26 (s, 9H), ꢀ1.95 (s, 1H), ꢀ2.23 (s, 1H).
6 6
Chlorin e DME (6). Chlorin e DME was prepared as previously
published, and its spectroscopic characterization agreed with the pub-
2
2
1
lished data.
3
2
1
7 Lysyl-chlorin e
dissolved in dry 2:1 THF/MeOH (10 mL) and stirred under argon for
0 min. Sodium methoxide (0.17 mL of a 0.5 M solution) was added,
6
TME (3). Compound 2 (74 mg, 0.089 mmol) was
15 Aspartylchlorin e
was dissolved in dry CH
added and allowed to stir until completely dissolved. After 3 h, H-Asp-
(OtBu)2 HCl (47.2 mg) and DIEA (0.029 mL) were mixed in CH Cl2
6
DME (7a). Chlorin e
6
(100 mg, 0.168 mmol)
2
Cl . DCC (35 mg) and DMAP (9 mg) were
2
1
and the reaction was allowed to stir at 0 °C for 1 h. The reaction was
followed using spectrophotometry. The solution turned from brown to
green as the isocyclic ring opens. The reaction mixture was then poured
3
2
and added to the reaction mixture. The reaction was allowed to stir
overnight. The mixture was diluted with CH Cl and then washed with
2
2
into water. The mixture was extracted with CH
layer was washed with water and 5% citric acid, dried over anhydrous
2
Cl
2
, and the organic
5% aqueous citric acid, followed by washing with brine and water. It was
dried over anhydrous Na SO and then evaporated. The residue was
2
4
Na
CH
2
SO
Cl
4
, and then evaporated. The residue was dissolved in 2% MeOH/
dissolved in CH
residue was dissolved in 2% MeOH/CH
column chromatography using the same mobile phase to afford 65 mg,
2
Cl
2
and treated with ethereal diazomethane. The
2
2
and purified on a silica gel plug with the same mobile phase. The
2
2
Cl and purified via silica gel
3
solvent was evaporated, and 68 mg (0.078 mmol), 89% yield, of 17 -
1
2
lysyl(Boc)chlorin e
(CD CO, 400 MHz]: δ 9.80 (s, 1H), 9.65 (s, 1H), 9.03 (s, 1H),
.19 (dd, J = 17.87, 11.58 Hz, 1H), 7.44 (d, J = 7.6 Hz, 1H), 6.40 (d, J =
7.87, 1.33 Hz, 1H), 6.13 (d, J = 11.61, 1.36 Hz, 1H), 5.89 (s, 1H), 5.38
6
TME (C48
H
62
N
6
O
9
) was obtained. H NMR
0.076 mmol, 45% yield of 15 -monoaspartylchlorin e
dimethyl ester (C48 ). MS (MALDI) m/z 853 (M + H) . H
NMR [(CD CO, 400 MHz]: δ 9.64 (s, 1H), 9.32 (s, 1H), 9.02
6
di(tert)butyl
+
1
[
8
1
3
)
2
61 5 9
H N O
3 2
)
(s, 1H), 7.81 (dd, J = 11.58, 17.85 Hz, 1H), 7.00 (s, 1H), 6.09 (dd, J =
17.87, 1.26 Hz, 1H), 5.85 (dd, J = 11.60, 1.29 Hz, 1H), 5.38 (s, 2H), 4.6
(br, 3H), 4.32 (s, 3H), 3.62 (s, 3H), 3.54 (s, 3H), 3.52 (q, J = 7.3 Hz,
2H), 3.30 (s, 3H), 3.02 (s, 3H), 2.78 (m, 3H), 2.44 (m, 2H), 1.87 (m,
(m, 2H), 4.63 (q, J = 7.22 Hz, 1H), 4.55 (dd, J = 10.34, 1.95 Hz, 1H),
4
3
2
1
1
2
.44 (m, 1H), 4.25 (s, 3H), 3.78 (q, J =7.67, 2H) 3.75 (s, 3H), 3.69 (s,
H), 3.56 (s, 3H) 3.49 (s, 3H), 3.27 (3H, s), 3.00 (m, 2H), 2.66 (m, 1H),
.33 (m, 1H), 2.20 (m, 1H), 1.85ꢀ1.62 (br, 4H), 1.78 (d, J = 7.24, 3H),
.69 (t, J = 7.58, 3H), 1.43 (m, 2H), 1.32 (s, 9H), ꢀ1.32 (s, 1H), ꢀ1.52 (s,
1H), 1.65 (d, J = 7.3 Hz, 3H), 1.58 (t, J = 7.7 Hz, 3H), 1.26 (s, 9H), 1.16
2
(s, 9H), ꢀ1.42 (s, 1H), ꢀ1.53 (s, 1H). The 15 -Aspartylchlorin e
6
H). Lysyl(Boc)chlorin e
6
TME (68 mg, 0.078 mmol) was dissolved in
di(tert)butyl dimethyl ester (65 mg, 0.076 mmol) was dissolved in 2 mL
mL of dry CH Cl in an ice bath under argon. Thioanisole (0.057 mmol)
of dry CH Cl in an ice bath under argon. Thioanisole (0.006 mL) and
2
2
2
2
7
471
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
mL of TFA were added, and the reaction mixture was allowed to stir
ARTICLE
6
Palladium(II) 15 -Aspartylchlorin e DME (8a). 15 -Aspartylchlorin
2
2
2
overnight. The reaction mixture was evaporated several times with
diethyl ether to remove TFA. The resulting precipitate was washed
several times with diethyl ether to remove thioanisole and then dissolved
e6 DME (44 mg, 0.06 mmol) was dissolved in 2 mL of dry THF.
Palladium(II) acetate (14.2 mg, 0.063 mmol) was dissolved in THF and
added to the reaction vessel and allowed to stir at 40 °C for 3 h. The
reaction was followed by spectrophotometry. The solution turned
from green to bluishꢀgreen as the complex formed. After evaporation
of solvent, the residue was dissolved in methanol and purified via
in CH
NaHCO
acid until all precipitate redissolved in the organic layer. The organic
layer was dried over anhydrous Na SO . Solvent was evaporated, and
4 mg, 0.06 mmol, 79% yield of 15 -aspartylchlorin e
2
Cl
2
and washed three times with H
2
O and once with 10%
3
to remove TFA. The organic layer was washed with citric
Sephadex LH-20 chromatography using the same mobile phase to
2
4
2
2
4
6
DME was
afford 50 mg, 0.059 mmol, 98% yield of palladium 15 -lysylchlorin e
6
ꢀ
1
ꢀ1
ꢀ1
ꢀ1
obtained (C40
H
45
N
5
O
9
). UVꢀvis (acetone) λmax (ε/M cm ) 664
TME (C40
H
43
N
5
O
9
Pd). UVꢀvis (acetone): λmax (ε/M cm ) 619
(48400), 609 (6000), 560 (2000), 529 (6280), 500 (14160), 402
(31240), 579 (6210), 489 (4035), 393 (44160). MS (MALDI) m/z
+
1
+ 1
(151800). MS (MALDI) m/z 841 (M + H) . H NMR [(CD ) CO,
845 (M + H) . H NMR [(CD CO, 400 MHz]: δ 9.53 (s, 1H), 9.54
3 2
)
3
2
4
1
1
3
3
2
00 MHz]: δ 9.76 (s, 1H), 9.58 (s, 1H), 9.04 (s, 1H), 8.10 (dd, J = 17.85,
1.59 Hz, 1H), 6.33 (dd, J = 17.87, 0.98 Hz, 1H), 6.07 (dd, J = 11.61,
.09 Hz, 1H), 5.32 (m, 2H), 4.84 (m, 1H), 4.60 (m, 2H), 4.26 (s, 3H),
.72 (q, J = 7.3 Hz, 2H), 3.57 (s, 3H), 3.54 (s, 3H), 3.45 (s, 3H), 3.21 (m,
H), 2.95 (dd, J = 16.85, 5.53 Hz, 1H), 2.88 (dd, J = 16.85, 5.09 Hz, 1H)
.68 (m, 1H) 2.32 (m, 2H), 1.78 (m, 1H), 1.74 (d, J = 7.3 Hz, 3H), 1.66
(s, 1H), 8.93 (s, 1H), 8.03 (dd, J = 17.82, 11.54 Hz, 1H), 6.17 (d, J =
17.79 Hz, 1H), 6.00 (dd, J = 11.61, 1H), 5.19 (d, J = 18.77 Hz, 1H),
5.04 (d, J = 18.59 Hz, 1H), 4.92 (m, 1H), 4.63 (m, 2H), 4.20 (s, 3H),
3.62 (s, 3H), 3.57 (m, 2H), 3.57 (s, 3H), 3.37 (s, 3H), 3.07 (m, 3H),
2.68 (m, 2H) 2.38 (m, 1H) 2.12 (m, 1H), 1.78 (m, 2H), 1.74 (d, J =
7.3 Hz, 3H), 1.50 (t, J = 7.7 Hz, 3H). HRMS (MALDI-TOF) m/z
+
(
t, J = 7.7 Hz, 3H), ꢀ1.39 (s, 1H), ꢀ1.58 (s, 1H). HRMS (MALDI-
843.109 [M] , calcd for C40
H
43
N
5
O
9
Pd 843.210.
+
2
2
6 6
TOF) m/z 740.376 [M + H] , calcd for C40
H
46
N
5
O
9
740.330
(75 mg, 0.12 mmol) was
dissolved in dry CH Cl (10 mL). DCC (31.3 mg) and DMAP (11 mg)
Palladium(II) 15 -Lysylchlorin e TME (8b). 15 -Lysylchlorin e TME
2
1
5 Lysylchlorin e
6
TME (7b). Chlorin e
6
(56 mg) was dissolved in 5 mL of dry THF. Palladium(II) acetate
(53 mg) was dissolved in THF and added to the reaction vessel and
stirred at 60 °C for 3 h. The reaction was followed by spectrophoto-
metry. The solution turned from green to bluishꢀgreen as the complex
formed. After evaporation of solvent, the residue was dissolved in
methanol and purified via Sephadex LH-20 chromatography using the
2
2
were added and allowed to stir until completely dissolved. After 3 h,
H-lysyl(Boc)OMe HCl (45 mg) and DIEA (0.022 mL) were mixed in
CH Cl and added to the reaction mixture. The reaction was allowed to
2 2
stir overnight. It was then diluted with CH Cl and washed with 5%
aqueous citric acid, followed by a wash with brine and water. It was dried
over anhydrous Na SO and then evaporated to dryness. The residue
was dissolved in CH
The residue was dissolved in 12% acetone/CH
silica gel column chromatography with the same mobile phase to afford
3
2
2
same mobile phase to afford 61 mg, 0.07 mmol, 99% yield of palladium-
2
(
λ
(
II) 15 -lysylchlorin e TME (C H N O Pd). UVꢀvis (acetone):
6
43 53 6 7
2
4
ꢀ
1
ꢀ1
max (ε/M cm ) 620 (97520), 581 (14500), 490 (7600), 394
140900). H NMR [(CD
2
Cl
2
and treated with excess ethereal diazomethane.
1
3 2
) CO, 400 MHz] all the peaks were broad
2
2
Cl and purified via
due to partial paramagnetic nature of the compound. HRMS (MALDI-
+
+
2
TOF) m/z 872.337 [M + 2H] , 767.450 [M + H ꢀ Pd] calcd for
8
H
4 mg, 0.097 mmol, 77% yield of 15 -lysyl(Boc)chlorin e
6
TME (C48
-
+
1
C H N O Pd 871.301, C H N O 767.413.
4
3
54
6
7
43 55 6 7
62
N
6
O
9
). MS (MALDI) m/z 868 (M + H) . H NMR [(CD CO,
3 2
)
1
1
6 6
3 Aspartylchlorin e DME (9). Chlorin e dimethyl ester (55 mg,
4
1
00 MHz]: δ 9.77 (s, 1H), 9.55 (s, 1H), 9.04 (s, 1H), 8.08 (dd, J = 17.87,
1.58 Hz, 1H), 7.07 (m, 1H), 6.31 (dd, J = 17.87, 1.33 Hz, 1H), 6.06 (dd,
0
.088 mmol) was dissolved in dry CH Cl . A mixture of HOBt (12 mg),
2 2
TBTU (29 mg), and DIEA (0.017 mL) in DMF was added, and the
mixture was allowed to stir for 30 min. H-Asp(OtBu) HCl (60 mg) and
J = 11.61, 1.36 Hz, 1H), 5.87 (s, 1H), 5.32 (s, 2H), 4.65 (q, J = 7.22 Hz,
H), 4.60 (dd, J = 10.34, 1.95 Hz, 1H), 4.49 (m, 1H), 4.26 (s, 3H), 3.71
q, J =7.67, 2H) 3.57 (s, 3H), 3.54 (s, 3H), 3.47 (s, 3H) 3.44 (s, 3H),
2
1
(
3
2 2
DIEA (0.033 mL) were mixed in CH Cl and added to this reaction
mixture. The mixture was stirred overnight. It was diluted with CH Cl2
2
3
1
.19 (3H, s), 2.97 (m, 2H), 2.71 (m, 1H), 2.37 (m, 2H), 1.80 (m, 3H),
.77 (d, J = 7.24, 3H), 1.66 (t, J = 7.58, 3H), 1.39 (s, 9H), 1.27 (m, 2H),
and then washed with 5% aqueous citric acid, followed by a wash with
2
brine and water. It was dried over anhydrous Na
2 4
SO and then
ꢀ
1.35 (s, 1H), ꢀ1.56 (s, 1H). The 15 -lysyl(Boc)chlorin e TME (84
6
evaporated. The residue was dissolved in 5% acetone/CH Cl and
2
2
2 2
mg 0.097 mmol) was dissolved in 3 mL of dry CH Cl in an ice bath
purified via silica gel column chromatography using the same mobile
under argon. Thioanisole (0.01 mL) and 1 mL of TFA were added, and
the reaction mixture was allowed to stir overnight. The reaction mixture
was evaporated several times with diethyl ether to remove residual
TFA. The resulting precipitate was washed several times with diethyl
ether to remove thioanisole. Then the precipitate was dissolved in
CH
NaHCO
acid until the precipitate was redissolved in the organic layer. The
organic layer was dried over anhydrous Na SO and then was
evaporated to give 56 mg, 0.071 mmol, 73% yield of 15 -lysylchlorin
1
phase to afford 50 mg, 0.058 mmol, 66% yield of 13 -aspartylchlorin e
di(tert)butyl dimethyl ester (C48 ). MS (MALDI) m/z 852
M + H) . H NMR [(CD ) CO, 400 MHz]: δ 9.79 (s, 1H), 9.70 (s,
6
61 5 9
H N O
+
1
(
3
2
1H), 9.10 (s, 1H), 8.39 (d, J = 7.93 Hz, 1H), 8.19 (dd, J = 11.58, 17.85
Hz, 1H), 6.37 (dd, J = 17.85, 1.23 Hz, 1H), 6.10 (dd, J = 11.87, 1.26 Hz,
2
Cl
2
and washed three times with H
2
O and once with 10%
1
4
3
0
1
H), 5.71 (d, J = 18.95 Hz, 1H), 5.30 (m, 2H), 4.67 (q, J = 7.22 Hz, 1H),
.49 (dd, J = 10.34, 1.95 Hz, 1H), 3.76 (q, J = 7.3 Hz, 2H), 3.74 (s, 3H),
.64 (s, 3H), 3.61 (s, 3H), 3.50 (s, 3H), 3.27 (s, 3H), 3.16 (dd, J = 5.8,
.78 Hz, 2H), 2.73 (m, 1H), 2.32 (m, 2H), 1.79 (m, 1H), 1.69 (m, 6H),
3
to remove TFA. The organic layer was washed with citric
2
4
2
1
.64 (s, 9H), 1.53 (s, 9H), ꢀ1.57 (s, 1H), ꢀ1.85 (s, 1H). The 13 -
ꢀ
1
ꢀ1
e TME (C H N O ). UVꢀvis (acetone) λ
(ε/M cm ) 664
6
43 54
6
7
max
6
Aspartylchlorin e di(tert)butyl dimethyl ester (50 mg, 0.058 mmol) was
1
(
57800), 608 (4600), 529 (4900), 500 (14800), 400 (187200). H
NMR [(CD CO, 400 MHz]: δ 9.68 (s, 1H), 9.42 (s, 1H), 9.02
s, 1H), 7.95 (dd, J = 17.87, 11.58 Hz, 1H), 7.12 (d, J = 6.8 Hz, 1H),
dissolved in 2 mL of dry CH
2 2
Cl in a ice bath under argon. Thioanisole
3 2
)
(0.005 mL) and 2 mL of TFA were added, and the reaction mixture was
(
stirred overnight. The reaction mixture was evaporated several times
with diethyl ether to remove residual TFA. The resulting precipitate was
washed several times with diethyl ether to remove thioanisole. Then the
precipitate was dissolved in CH Cl and washed three times with H O
6
5
1
.20 (dd, J = 17.87, 1.33 Hz, 1H), 5.95 (dd, J = 11.61, 1.36 Hz, 1H),
.35 (s, 2H), 4.65 (q, J = 7.22 Hz, 1H), 4.61 (dd, J = 10.34, 1.95 Hz,
H), 4.53 (m, 1H), 4.27 (s, 3H), 3.59 (m, 2H) 3.55 (s, 6H), 3.48
2
2
2
(
(
1
s, 3H) 3.37 (s, 3H), 3.09 (3H, s), 2.99 (m, 3H), 2.73 (m, 1H), 2.38
m, 2H), 1.80 (m, 3H), 1.78 (d, J = 7.24, 3H), 1.61 (t, J = 7.58, 3H),
and once with 10% NaHCO to remove TFA. The organic layer was
3
dried over anhydrous Na SO and then evaporated to give 38 mg, 0.051
2
4
1
.43 (m, 2H), ꢀ1.40 (s, 1H), ꢀ1.57 (s, 1H). HRMS (MALDI-TOF)
6 40 45 5 9
mmol, 88% yield of 13 -aspartylchlorin e DME (C H N O ).
+
ꢀ1
ꢀ1
m/z 767.399 [M + H] , calcd for C H N O 767.413
UVꢀvis (acetone): λmax (ε/M cm ) 663 (126800), 607 (9000),
4
3
55
6
7
7
472
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
ARTICLE
1
5
28 (7450), 500 (31540), 399 (385800). H NMR [(CD
MHz]: δ 9.79 (s, 1H), 9.70 (s, 1H), 9.12 (s, 1H), 8.50 (d, J = 7.93 Hz,
H), 8.19 (dd, 11.58, 17.85 Hz, 1H), 6.37 (d,
3
)
2
CO, 400
2.57 Hz, 1H), 2.79 (d, J = 3.06 Hz, 1H), 2.31 (m, 2H), 1.79 (m, 1H), 1.71
(d, J = 7.24, 3H), 1.68 (t, J = 7.58, 3H), 1.43 (s, 9H), 1.42 (s, 9H), ꢀ1.61
1
1
J
=
(s, 1H), ꢀ1.91 (s, 1H). The 13 -β-alanyl-aspartylchlorin e di(tert)butyl
6
J = 18.85 Hz, 1H), 6.10 (d, J = 11.87 Hz, 1H), 5.71 (d, J = 18.95
Hz, 1H), 5.45 (dd, J = 13.76, 5.80 Hz, 1H), 5.30 (d, J = 18.93 Hz, 1H),
dimethyl ester (45 mg, 0.048 mmol) was dissolved in 2 mL of dry
2 2
CH Cl in a ice bath under argon. Thioanisole (0.004 mL) and 2 mL of
4
7
.67 (q, J = 7.22 Hz, 1H), 4.50 (dd, J = 10.34, 1.95 Hz, 1H), 3.76 (q, J =
.3 Hz, 2H), 3.71 (s, 3H), 3.63 (s, 3H), 3.60 (s, 3H), 3.51 (s, 3H), 3.31
TFA were added, and the reaction mixture was stirred overnight before
being evaporated several times with diethyl ether to remove residual
TFA. The resulting precipitate was washed several times with diethyl
ether to remove more residual thioanisole. The precipitate was dissolved
in CH Cl and washed three times with H O and once with 10%
(dd, J = 5.76, 3.47 Hz, 2H), 3.26 (s, 3H), 2.70 (m, 2H), 2.31 (m, 2H),
1
1
.79 (m, 1H), 1.71 (d, J = 7.24, 3H), 1.67 (t, J = 7.58, 3H), ꢀ1.57 (br s,
+
H), ꢀ1.87 (s, 1H). HRMS (MALDI-TOF) m/z 740.344 [M + H] ,
2
2
2
3
NaHCO to remove TFA. The organic layer was dried over anhydrous
46 5 9
calcd for C40H N O 740.330
1
Na
2 4
SO , and the solvent was evaporated to give 38 mg, 0.046 mmol, 97%
1
3 β-Alanylchlorin e DME (10). Chlorin e DME (55 mg, 0.088
6
6
1
yield of 13 -aspartylchlorin e DME (C H N O ). UVꢀvis (acetone):
mmol) was dissolved in dry CH
TBTU (29 mg), and DIEA (0.017 mL) in DMF was added, and the
mixture was stirred for 30 min. β-Alanyl(OtBu) HCl (18 mg) and DIEA
2
Cl
2
. A mixture of HOBt (12 mg),
6
43 50 6 10
ꢀ
1
ꢀ1
λmax (ε/M cm ) 663 (77125), 607 (3621), 528 (1831), 500 (17200),
1
3 2
399 (237100). H NMR [(CD ) CO, 400 MHz]: δ 9.72 (s, 1H), 9.70
3
(0.017 mL) were mixed in CH Cl and added to the reaction mixture.
(s, 1H), 9.10 (s, 1H), 8.20 (dd, J = 11.58, 17.85 Hz, 1H), 8.11 (br t, J =
5.66 Hz, 1H), 7.64 (br d, J = 8.11 Hz, 1H), 6.32 (d, J = 19.18 Hz, 1H),
6.10 (d, J = 12.94 Hz, 1H), 5.63 (d, J = 19.03 Hz, 1H), 5.37 (d, J = 19.09
Hz, 1H), 4.92 (dt, J = 8.14, 5.74 Hz, 1H), 4.66 (q, J = 7.22 Hz, 1H), 4.51
(dd, J = 10.34, 1.95 Hz, 1H), 4.07 (m, 1H), 3.95 (m, 1H), 3.74 (s, 3H),
3.75 (q, J = 7.3 Hz, 2H), 3.60 (s, 3H), 3.51 (s, 3H), 3.50 (s, 3H), 3.25 (s,
3H), 2.96 (d, J = 5.27 Hz, 1H), 2.69 (d, J = 3.06 Hz, 1H), 2.31 (m, 2H),
1.79 (m, 1H), 1.71 (d, J = 7.24, 3H), 1.66 (t, J = 7.58, 3H), ꢀ1.61 (s, 1H),
2
2
The mixture was then allowed to stir for overnight before being diluted
with CH Cl and then washed with 5% aqueous citric acid, followed by a
wash with brine and water. It was dried over anhydrous Na SO and then
evaporated to dryness. The residue was dissolved in 5% acetone/
CH Cl and purified via silica gel column chromatography using the
2
2
2
4
2
2
1
same mobile phase to afford 45 mg, 0.06 mmol, 68% yield of 13 -β-
alanylchlorin e (tert)butyl dimethyl ester (C H N O ). H NMR
1
6
43 53 5 7
+
[(CD
3
)
2
CO, 400 MHz]: δ 9.70 (s, 1H), 9.63 (s, 1H), 9.01 (s, 1H), 8.17
ꢀ1.91 (s, 1H). HRMS (MALDI-TOF) m/z 811.381 [M + H] , calcd for
(t, J = 5.71 Hz, 1H), 8.11 (dd, J = 11.58, 17.85 Hz, 1H), 6.31 (d, J = 18.85
C H N O 811.367.
4
3 51 6 10
1
Hz, 1H), 6.04 (d, J = 11.87 Hz, 1H), 5.65 (d, J = 18.95 Hz, 1H), 5.38 (d,
J = 19.06 Hz, 1H), 4.66 (q, J = 7.22 Hz, 1H), 4.52 (dd, J = 10.34, 1.95 Hz,
6
13 Ethylenediaminylchlorin e DME (12). Methyl pheophorbide a,
(100 mg, 0.164 mmol) was dissolved in dry CHCl and stirred under
3
1
3
H), 4.02 (m, 1H), 3.09 (m, 1H), 3.77 (s, 3H), 3.68 (q, J = 7.3 Hz, 2H),
.61 (s, 3H), 3.51 (s, 3H), 3.46 (s, 3H), 3.21 (s, 3H), 2.71 (m, 2H), 2.31
argon for 10 min. Then 0.2 mL of ethylenediamine was added to the
solution and the mixture stirred for 24 h. The reaction was monitored by
spectrophotometry. The reaction mixture was evaporated, and the
residue was dissolved in 2.5% MeOH/CH Cl and then chromato-
(
(
(
m, 2H), 1.79 (m, 1H), 1.72 (d, J = 7.24, 3H), 1.65 (t, J = 7.58, 3H), 1.54
1
s, 9H), ꢀ1.62 (s, 1H), ꢀ1.91 (s, 1H). The 13 -β-alanylchlorin e
tert)butyl dimethyl ester (45 mg, 0.059 mmol) was dissolved in 1.5 mL
Cl in a ice bath under argon. Thioanisole (0.005 mL) and
.5 mL of TFA were added, and the reaction mixture was stirred
overnight before being diluted with CH Cl and washed three times
with H O and once with 10% NaHCO . The organic layer was dried
over anhydrous Na SO , and the solvent was evaporated to give 39 mg,
6
2
2
graphed on a short silica gel column using the same mobile phase to
remove byproduct, and then the product was eluted using 50% MeOH/
of dry CH
1
2
2
1
CH Cl to afford 100 mg, 0.150 mmol, 91% yield of 13 -ethylenedia-
2
2
1
38 46 6 5 3 2
2
2
minylchlorin e6 DME (C H N O ). H NMR [(CD ) CO, 400
2
3
MHz]: δ 9.69 (s, 1H), 9.63 (s, 1H), 9.09 (s, 1H), 8.09 (dd, J = 17.78,
11.58 Hz, 1H), 8.07 (br s, 1H), 6.29 (d, J = 18.85 Hz, 1H), 6.02 (d, J =
11.87 Hz, 1H), 5.67 (d, J = 19.08 Hz, 1H), 5.39 (d, J = 19.11 Hz, 1H),
4.66 (q, J = 7.22 Hz, 1H), 4.51 (dd, J = 10.34, 1.95 Hz, 1H), 4.01 (m,
1H), 3.87 (m, 1H), 3.75 (s, 3H), 3.66 (q, J = 7.3 Hz, 2H), 3.61 (s, 3H),
3.50 (s, 3H), 3.45 (s, 3H), 3.20 (s, 3H), 2.71 (m, 1H), 2.31 (m, 2H), 1.95
(br m, 1H), 1.80 (m, 1H), 1.72 (d, J = 7.24, 3H), 1.64 (t, J = 7.58, 3H),
2
4
1
0
.056 mmol, 95% yield of 13 -β-alanylchlorin e DME (C H N O ).
6 39 45 5 7
1
H NMR [(CD
H), 8.22 (dd, J = 11.58, 17.85 Hz, 1H), 8.15 (t, J = 5.71 Hz, 1H), 6.39
d, J = 18.85 Hz, 1H), 6.12 (d, J = 11.87 Hz, 1H), 5.65 (d, J = 18.95 Hz,
3 2
) CO, 400 MHz]: δ 9.74 (s, 1H), 9.72 (s, 1H), 9.10 (s,
1
(
1H), 5.37 (d, J = 19.06 Hz, 1H), 4.66 (q, J = 7.22 Hz, 1H), 4.52 (dd, J =
1
3
1
0.34, 1.95 Hz, 1H), 4.07 (m, 1H), 3.95 (m, 1H), 3.77 (s, 3H), 3.68
ꢀ1.64 (s, 1H), ꢀ1.93 (s, 1H). C NMR [(CD ) CO, 100 MHz] δ
3
2
(
2
(
q, J = 7.3 Hz, 2H), 3.61 (s, 3H), 3.53 (s, 3H), 3.52 (s, 3H), 3.28 (s, 3H),
.69 (m, 2H), 2.31 (m, 2H), 1.79 (m, 1H), 1.72 (d, J = 7.24, 3H), 1.65
t, J = 7.58, 3H), ꢀ1.62 (s, 1H), ꢀ1.91 (s, 1H).
173.1, 173.0, 169.2, 168.3, 167.7, 153.6, 149.1, 144.4, 138.1, 136.0, 135.3,
134.7, 133.9, 130.1, 129.8, 129.3, 120.9, 103.2, 100.7, 98.4, 93.9, 53.1,
51.4, 50.8, 50.2, 48.8, 41.2, 37.2, 30.7, 29.6, 22.6, 19.0, 17.2, 11.3, 11.0,
1
+
1
3 β-Alanyl-aspartylchlorin e
6
DME (11). Chlorin e
6
DME (39 mg,
10.3. HRMS (MALDI-TOF) m/z 667.395 [M + H] , calcd for
0
.056 mmol) was dissolved in dry CH Cl . A mixture of HOBt (8 mg),
C H N O 667.361.
13 Ethylenediaminyl-lysylchlorin e DME (13). (Boc)Lysine(Boc)OH
(55 mg, 0.082 mmol) was dissolved in dry CH Cl (10 mL) A mixture of
HOBt (15 mg), TBTU (33 mg), andDIEA (0.025 mL) in DMF was added
2
2
38 47 6 5
1
TBTU (18 mg), and DIEA (0.017 mL) in DMF was added and stirred
for 30 min. H-Asp(OtBu) HCl (40 mg) and DIEA (0.025 mL) were
mixed in CH
6
2
3
2
2
2
Cl
2
and added to the reaction mixture. The mixture was
1
stirred overnight before being diluted with CH Cl and then washed
and then stirred for 1 h. 13 -Ethylenediaminylchlorin e DME (50 mg,
2
2
6
with 5% aqueous citric acid, followed by a wash with brine and water. It
was dried over anhydrous Na SO and then evaporated. The residue was
dissolved in 10% acetone/CH Cl and purified via silica gel column
chromatography using the same mobile phase to afford 45 mg, 0.048
0.075 mmol) was added to the reaction mixture, which was stirred
overnight. The mixture was diluted with CH Cl and then washed with
2
4
2
2
2
2
5% aqueous citric acid, followed by a wash with brine and water. It was dried
over anhydrous Na SO and then evaporated. The residue was dissolved in
2
4
1
mmol, 87% yield of 13 -β-alanyl-aspartylchlorin e
6
di(tert)butyl dimeth-
CO, 400 MHz]: δ 9.74 (s, 1H), 9.70 (s, 1H),
.10 (s, 1H), 8.19 (dd, J = 11.58, 17.85 Hz, 1H), 8.11 (t, J = 5.66 Hz, 1H),
.64 (d, J = 8.11 Hz, 1H), 6.37 (d, J = 19.18 Hz, 1H), 6.10 (d, J = 12.94
2.5% MeOH/CH Cl and purified via silica gel column chromatography
2 2
1
1
yl ester. H NMR [(CD
3
)
2
using the same mobile phase to afford 65 mg, 0.065 mmol, 78% yield of 13 -
ethylenediaminyl(Boc)lysyl(Boc)chlorin e6 DME (C H N O ). H
54 74 8 10
1
9
7
NMR [(CD ) CO, 400 MHz]: δ 9.54 (s, 1H), 9.53 (s, 1H), 9.07 (s,
3
2
Hz, 1H), 5.66 (d, J = 19.03 Hz, 1H), 5.38 (d, J = 19.09 Hz, 1H), 4.77 (dt,
J = 8.14, 5.74 Hz, 1H), 4.66 (q, J = 7.22 Hz, 1H), 4.52 (dd, J = 10.34, 1.95
Hz, 1H), 4.07 (m, 1H), 3.95 (m, 1H), 3.75 (s, 3H), 3.76 (q, J = 7.3 Hz,
1H), 8.14 (br s, 1H) 7.94 (dd, J= 17.78, 11.58 Hz, 1H), 7.78 (br s, 1H), 6.18
(d, J = 18.85 Hz, 1H), 6.16 (br s, 1H), 5.93 (d, J = 11.87 Hz, 1H), 5.89 (br s,
1H), 5.63 (d, J = 19.08 Hz, 1H), 5.39 (d, J = 19.11 Hz, 1H), 4.67 (q, J = 7.22
Hz, 1H), 4.52 (dd, J = 10.34, 1.95 Hz, 1H), 4.17 (m, 1H), 3.85 (m, 1H),
2H), 3.60 (s, 3H), 3.53 (s, 3H), 3.50 (s, 3H), 3.27 (s, 3H), 2.80 (d, J =
7
473
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
.75 (s, 3H), 3.66 (m, 3H), 3.61 (s, 3H), 3.52 (br q, J = 7.14 Hz, 2H), 3.41
ARTICLE
3
dose. The cells were kept cool by filtering the IR radiation through
10 mm of water and placing the culture in an iceꢀwater bath. After
exposure to light for 20 min, the plate was incubated in the 37 °C, 5%
2
incubator overnight. Cell viability was then measured as des-
cribed above using CellTiter blue Viability Assay Kit, as previously re-
(
(
(
ꢀ
s, 3H), 3.39 (s, 3H), 3.11 (s, 3H), 3.00 (m, 2H), 2.71 (m, 1H), 2.32
m, 2H), 1.95 (br m, 1H), 1.84 (m, 2H), 1.72 (d, J = 7.24, 3H), 1.67
m, 1H), 1.59 (t, J = 7.58, 3H), 1.41 (m, 3H), 1.37 (s, 9H), 1.34 (s, 9H),
CO
1
1.66 (s, 1H), ꢀ1.94 (s, 1H). The 13 -ethylenediaminyl(Boc)lysyl(Boc)-
2
0,21,24
chlorin e DME (65 mg, 0.065 mmol) was dissolved in 3 mL of dry CH Cl
2
ported.
6
2
in a ice bath under argon. Thioanisole (0.003 mL) and 2 mL of TFA were
added, and the reaction mixture was stirred overnight. The reaction mixture
was evaporated several times with diethyl ether to remove TFA. The
resulting precipitate was washed several times with diethyl ether to remove
2.4. Microscopy. The cells were incubated in a glass bottom 6-well
plate (MatTek) and allowed to grow for 48 h. The cells were then
exposed to 10 μM of each compound for 18 h. Organelle tracers were
obtained from Invitrogen and used at the following concentrations:
LysoSensor Green 50 nM, MitoTracker Green 250 nM, ER Tracker
Blue/white 100 nM, and BODIPY FL C5 ceramide 1 mM. After 30 min
thioanisole. Then the precipitate was dissolved in CH
times with H O and once with 10% NaHCO to remove TFA. The organic
layer was dried over anhydrous Na SO , and the solvent was evaporated to
give 42 mg, 0.052 mmol, 81% yield of 13 -ethylenediaminyl-lysylchlorin e
2 2
Cl and washed three
2
3
incubation in the 37 °C, 5% CO incubator, both the media and the
2
4
2
1
organelle tracers were removed and washed with PBS buffer for 3 times.
6
ꢀ1
ꢀ1
DME (C44
76420), 607 (1810), 527 (657), 500 (14,560), 399 (242,100).
NMR [(CD CO, 400 MHz]: δ 9.76 (s, 1H), 9.69 (s, 1H), 9.10 (s,
H), 8.29 (br s, 1H) 8.17 (dd, J= 17.78, 11.58 Hz, 1H), 7.86 (br s, 1H), 6.35
d, J= 18.85 Hz, 1H), 6.07 (d, J= 11.87 Hz, 1H), 5.64 (d, J= 19.08 Hz, 1H),
H
58
N
8
O
6
). UVꢀvis (acetone): λmax (ε/M cm ) 663
Images were acquired using a Leica DMRXA microscope with 40ꢁ NA
1
0.8 dip objective lens and DAPI, GFP, and Texas Red filter cubes
(
H
(Chroma Technologies).
3 2
)
1
(
’ ASSOCIATED CONTENT
Supporting Information. Normalized absorption spec-
5.40 (d, J = 19.11 Hz, 1H), 4.66 (q, J = 7.22 Hz, 1H), 4.52 (dd, J = 10.34,
1.95 Hz, 1H), 3.92 (m, 2H), 3.75 (br m, 5H), 3.74 (s, 3H), 3.60 (s, 3H),
3.53 (s, 3H), 3.49 (s, 3H), 3.26 (s, 3H), 2.96 (br t, 2H), 2.70 (m, 1H), 2.29
S
b
tra, energy minimized conformations in water, time-dependent
uptake, and subcellular localization microscopy figures. Proton
NMR spectra for compounds 2ꢀ13 and other synthetic inter-
mediates, carbon-13 NMR shifts, carbon-13 NMR spectra. This
material is available free of charge via the Internet at http://pubs.
acs.org.
(m, 2H), 1.95 (br m, 1H), 1.84 (m, 2H), 1.72 (d, J = 7.24, 3H), 1.68 (t, J =
7
.58, 3H), 1.45 (m, 2H), 1.31 (m, 2H) ꢀ1.60 (s, 1H), ꢀ1.89 (s, 1H).
+
59 8 6
HRMS (MALDI-TOF) m/z 795.492 [M + H] , calcd for C44H N O
7
95.456.
. Cell Studies. Human carcinoma HEp2 cells were maintained in a
0:50 mixture of DMEM:AMEM (Invitrogen) supplemented with 5%
2
5
FBS (Invitrogen), 1% Primocin antibiotic (Invitrogen) in a humidified,
% CO incubator at 37 °C. The cells were subcultured twice weekly to
’ AUTHOR INFORMATION
5
2
Corresponding Author
*
maintain subconfluent stocks. The fourth to fifteenth passage cells were
used for all experiments.
Phone: 225-578-7442. Fax: 225-578-3458. E-mail: kmsmith@
lsu.edu.
2
.1. Time-Dependent Cellular Uptake. The HEp2 cells were plated at
7500 cells per well in a Costar 96-well plate and allowed to grow for 48 h.
Compound stock solutions were prepared at 32 mM in DMSO and
Cremophor (10% of Cremophor in DMSO). Further dilution into the
cells of the 96-well plate gave a final concentration of 10 μM with
maximum DMSO concentration of 1% and Cremophor concentration
of 0.1%. For the uptake test, the compounds were diluted to 20 μM (2ꢁ
stock solution) and added to the 96-well to give a final concentration of
’
ACKNOWLEDGMENT
This work was supported by the National Institutes of Health,
grant numbers CA132861 (K.M.S.) and CA139385 (M.G.H.V.).
’
ABBREVIATIONS LIST
10 μM at 0, 1, 2, 4, 8, 12, and 24 h interval. The uptake was terminated by
PS, photosensitizer; PDT, photodynamic therapy; DMF, di-
methylformamide; DMSO, dimethylsulfoxide; THF, tetrahydro-
furan; DME, dimethyl ester; TME, trimethyl ester; DBU, 1,
removing the loading medium and washing the wells with 200 μL of PBS
buffer. Cells and compounds were then solubilized using 100 μL of
0.25% Triton X-100 in PBS buffer. The compound concentration was
8
-diazabicyclo[5.4.0]undec-7-ene; DCC, 1,3-dicyclohexylcarbo-
measured using intrinsic fluorescence as measured with a BMG FLUOs-
tar plate reader equipped with a 355 nm excitation and a 650 nm
emission filter. The cells were measured using a CyQuant Cell prolifer-
ation assay (Invitrogen) as per manufacturer’s instructions, as previously
diimide; TFA, trifluoroactic acid; DIEA, N,N-diisopropylethyla-
mine; TBTU, 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluro-
nium tetrafluoroborate; DMAP, 4-dimethylaminopyridine; HOBt,
1-hydroxybenzotriazole; PBS, phosphate buffered saline; FBS, fetal
bovine serum; MEM, modified eagle medium; ER, endoplasmic
reticulum
20,21,24
reported.
.2. Dark Cytotoxicity. The HEp2 cells were plated as described above
2
for the uptake experiment. The compounds were diluted into media to
give 400 μM solution concentrations. Two-fold serial dilutions were
then prepared from 400 to 50 μM, and the cells were incubated for 18 h.
The loading medium was removed, and the cells were fed new medium,
followed by 20 μL of CellTiter blue per 100 μL of medium and
incubated for 4 h. Cell toxicity was measured using Promega’s CellTiter
Blue Viability Assay Kit as per manufacturer’s instructions; untreated
cells were considered 100% viable and cells treated with 0.2% saponin as
’
REFERENCES
(
1) Dougherty, T. J.; Gomer, C. J.; Henderson, B. W.; Jori, G.;
Kessel, D.; Korbelik, M.; Moan, J.; Peng, Q. Photodynamic therapy.
J. Natl. Cancer Inst. 1998, 90, 889–905.
(
therapy as a major modality in cancer treatment. Rev. Contemp.
Pharmacother. 1999, 10, 69–74.
(
J. Porphyrins Phthalocyanines 2000, 4, 368–373.
(4) Brown, S. B.; Brown, E. A.; Walker, I. The present and future role
of photodynamic therapy in cancer treatment. Lancet Oncol. 2004,
5, 497–508.
2) Hahn, S. M.; Glatstein, E. The emergence of photodynamic
20,21,24
0% viable, as previously reported.
2.3. Phototoxicity. The cells were prepared as described above with
3) Pandey, R. K. Recent advances in photodynamic therapy.
compound concentration range from 6.25 to 100 μM. After loading
overnight, the medium was replaced with medium containing 50 mM
HEPES pH 7.2. The cells were exposed to a 100 W halogen lamp filtered
ꢀ
2
through a 610 nm long pass filter to provide approximately 1 J cm light
7
474
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
5) Vicente, M. G. H. Porphyrin-based Sensitizers in the Detection
and Treatment of Cancer: Recent Progress. Current Med. Chem., Anti-
Cancer Agents 2001, 1, 175–194.
ARTICLE
(
distribution on cellular uptake, distribution and phototoxicity of cationic
porphyrins in HEp2 cells. J. Photochem. Photobiol., B 2010, 100,
100–111.
(
6) Huang, Z. A review of progress in clinical photodynamic therapy.
Technol. Cancer Res. Treat. 2005, 4, 283–293.
7) Kostenicha, G. A.; Zhuravkina, I. N.; Zhavrid, E. A. Experimental
(24) Josefsen, L. B.; Boyle, R. W. Photodynamic therapy and
development of metal-based photosensitizers. Metal-Based Drugs
2008, 276109, 1–24.
(
grounds for using chlorin e6 in the photodynamic therapy of malignant
tumors. J. Photochem. Photobiol., B 1994, 22, 211–217.
(25) Ol’shevskaya, V. A.; Savchenko, A. N.; Zaitsev, A. V.; Konono-
va, E. G.; Petrovskii, P. V.; Ramonova, A. A., Jr.; Uvarov, V. V. T.;
Moisenovich, O. V.; Kalinin, M. M.; Shtil, V. N. A. A. Novel metal
(8) Allison, R. R.; Downie, G. H.; Cuenca, R.; Hu, X.-H.; Sibata,
C. H.; Childs, C. J. Photosensitizers in clinical PDT. Photodiagn.
Photodyn. Ther. 2004, 1, 27–42.
6
complexes of boronated chlorin e for photodynamic therapy. J.
Organomet. Chem. 2009, 694, 1632–1637.
(
9) Kessel, D. Determinants of photosensitization by mono-L-aspartyl
chlorin e . Photochem. Photobiol. 2008, 49, 447–452.
10) Sunar, U.; Rohrbach, D.; Rigual, N.; Tracy, E.; Keymel, K.;
(26) Obata, M.; Hirohara, S.; Tanaka, R.; Kinoshita, I.; Ohkubo, K.;
Fukuzumi, S.; Tanihara, M.; Yano, S. In vitro heavy-atom effect
of palladium(II) and platinum(II) complexes of pyrrolidine-fused
chlorin in photodynamic therapy. J. Med. Chem. 2009, 52, 2747–
2753.
(27) Tremblay, T.; Leroy, S.; Freitag, L.; Copin, M. C.; Brun,
P. H.; Marquette, C.-H. Endobronchial phototoxicity of WST 09
(Tookad), a new fast-acting photosensitizer for photodynamic
therapy: preclinical study in the pig. Photochem. Photobiol. 2003, 78,
124–130.
(28) Trachtenberg, J.; Bogaards, A.; Weersink, R. A.; Haider, M. A.;
Evans, A.; McCluskey, S. A.; Scherz, A.; Gertner, M. R.; Yue, C.; Appu,
S.; Aprikian, A.; Savard, J.; Wilson, B. C.; Elhilali., M. Vascular targeted
photodynamic therapy with palladium-bacteriopheophorbide photosen-
sitizer for recurrent prostate cancer following definitive radiation
therapy: assessment of safety and treatment response. J. Urol. 2007,
178, 1974–1979.
(29) Smith, K. M.; Goff, D. A.; Simpson, D. J. Meso substitution of
chlorophyll derivatives: direct route for transformation of bacteriopheo-
phorbides d into bacteriopheophorbides c. J. Am. Chem. Soc. 1985,
107, 4946–4954.
6
(
Cooper, M. T.; Baumann, H.; Henderson, B. H. Monitoring photo-
bleaching and hemodynamic responses to HPPH-mediated photody-
namic therapy of head and neck cancer: a case report. Opt. Express 2010,
1
8, 14969–14978.
(
11) (a) Usuda, J.; Ichinose, S.; Ishizumi, T.; Hayashi, H.; Ohtani, K.;
Maehara, S.; Ono, S.; Kajiwara, N.; Uchida, O.; Tsutsui, H. Management
of multiple primary lung cancer in patients with centrally located early
cancer lesions. J. Thorac. Oncol. 2010, 5, 62–68. (b) Usuda, J.; Ichinose, S.;
Ishizumi, T.; Hayashi, H.; Ohtani, K.; Maehara, S.; Ono, S.; Honda, H.;
Kajiwara, N.; Uchida, O. Outcome of Photodynamic Therapy Using
NPe6 for Bronchogenic Carcinomas in Central Airways >1.0 cm in
Diameter. Clin. Cancer Res. 2010, 16, 2198–2204.
(
12) Wang, S.; Bromley, E.; Xu, L.; Chen, J. C.; Keltner, L.
Talaporfin sodium. Expert Opin. Pharmacother. 2010, 11, 133–140.
13) Bromley, E.; Briggs, B.; Keltner, L.; Wang, S.-S. Characteriza-
(
tion of cutaneous photosensitivity in healthy volunteers receiving
talaporfin sodium. Photodermatol., Photoimmunol. Photomed. 2011,
2
7, 85–89.
14) Akhlynina, T. V.; Jans, D. A.; Rosenkranz, A. A.; Statsyuk, N. V.;
Balashova, I. Y.; Toth, G.; Pavo, I.; Rubin, A. B.; Sobolev, A. S. Nuclear
targeting of chlorin e enhances its photosensitizing activity. J. Biol.
Chem. 1997, 272, 20328–20331.
15) Soukos, N. S.; Hamblin, M. R.; Hasan, T. The effect of charge
(
(30) Kenner, G. W.; McCombie, S. W.; Smith., K. M. Separation and
oxidative degradation of chlorophyll derivatives. J. Chem. Soc., Perkin
Trans. 1 1973, 2517–2523.
(31) Wasielewsi, M. R.; Svec, W. A. Synthesis of covalently linked
dimeric derivatives of chlorophyll a, chlorophyll b and bacteriochlor-
ophyll a. J. Org. Chem. 1980, 45, 1969–1674.
6
(
6
on cellular uptake and phototoxicity of polylysine chlorin e conjugates.
Photochem. Photobiol. 1997, 65, 723–729.
(32) Han, S.-Y.; Kim, Y.-A. Recent development of peptide coupling
reagents in organic synthesis. Tetrahedron 2004, 60, 2447–2467.
(33) Fisher, J. W.; Trinkle, K. L. Iodide dealkylation of benzyl, PMB,
PNB, and t-Butyl N-acyl amino acid esters via lithium ion coordination.
Tetrahedron Lett. 1994, 35, 2505–2508.
(16) Schmidt-Erfurth, U.; Diddens, H.; Birngruber, R.; Hasan, T.
Photodynamic targeting of human retinoblastoma cells using covalent
low-density lipoprotein conjugates. Br. J. Cancer 1997, 75, 54–61.
(17) Bisland, S. K.; Singh, D.; Gari ꢀe py, J. Potentiation of Chlorin e
6
Photodynamic Activity in Vitro with Peptide-Based Intracellular Vehi-
cles. Bioconjugate Chem. 1999, 10, 982–992.
(34) Smith, K. M.; Lewis, M. W. Partial synthesis of chlorophyll a
from rhodochlorin. Tetrahedron Lett. 1981, 37, 399–403.
(35) Hudlicky, M. An improved apparatus for the laboratory pre-
paration of diazomethane. J. Org. Chem. 1980, 45, 5377–5378.
(36) Wiehe, A.; Stollberg, H.; Runge, S.; Paul, A.; Senge, M. O.;
Roder, B. PDT-related photophysical propertices of conformationally
distorted palladium(II) porphyrins. J. Porphyrins Phthalocyanines 2001,
5, 853–860.
(37) Romanova, Z. S.; Deshayes, K.; Piotrowiak, P. Remote inter-
molecular “heavy-atom effect”: spin-orbit coupling across the wall of a
hemicarcerand. J. Am. Chem. Soc. 2001, 123, 2444–2445.
(38) Meseguer, B.; Alonso-Díaz, D.; Griebenow, N.; Priv.-Doz.,
T. H.; Waldmann, H. Solid-phase synthesis and biological evaluation
of a teleocidin library—discovery of a selective PKCδ down regulator.
Chem.—Eur. J. 2000, 6, 3943–3957.
(18) Uzdensky, A. B.; Dergacheva, O. Y.; Zhavoronkova, A. A.;
Reshetnikov, A. V.; Ponomarev, G. V. Photodynamic effect of novel
chlorin e derivatives on a single nerve cell. Life Sci. 2004, 74, 2185–
6
2197.
(
19) Zheng, X.; Morgan, J.; Pandey, S. K.; Chen, Y.; Tracy, E.;
Baumann, H.; Missert, J. R.; Batt, C.; Jackson, J.; Bellnier, D. A.;
0
Henderson, B. W.; Pandey, R. K. Conjugation of 2-(1 -hexyloxyethyl)-
2-devinylpyropheophorbide-a (HPPH) to carbohydrates changes its
subcellular distribution and enhances photodynamic activity in Vivo.
J. Med. Chem. 2009, 52, 4306–4318.
(20) Sibrian-Vazquez, M.; Jensen, T. J.; Fronczek, F. R.; Hammer,
R. P.; Vicente, M. G. H. Synthesis and characterization of positively
charged porphyrinꢀpeptide conjugates. Bioconjugate Chem. 2005, 16,
852–863.
(39) Shiah, J. -G.; Sun, Y.; Peterson, C. M.; Straight, R. C.; Kopec, J.
Antitumor activity of N-(2-hydroxypropyl)methacrylamide copoly-
(21) Sibrian-Vazquez, M.; Jensen, T. J.; Vicente, M. G. H. Influence
of the number and distribution of NLS peptides on the photosensitizing
merꢀmesochlorin e
6
and adriamycin conjugates in combination treat-
ments. Clin. Cancer Res. 2000, 6, 1008–1015.
activity of multimeric porphyrinꢀNLS. Org. Biomol. Chem. 2010, 8,
1160–1172.
(40) Ol’shevskaya, A. V.; Nikitina, R. G.; Savchenko, A. N.;
Malshakova, M. V.; Vinogradov, A. M.; Golovina, G. V.; Belykh,
D. V.; Kutchin, A. V.; Kaplan, M. A.; Kalinin, V. N.; Kuzmin, V. A.;
Shtil, A. A. Novel boronated chlorin e based photosensitizer: synthesis,
binding to albumin and antitumor efficacy. Bioorg. Med. Chem. 2009,
17, 1297–1306.
(
22) Hargus, J. A.; Fronczek, F. R.; Vicente, M. G. H.; Smith, K. M.
Mono-(L)-aspartylchlorin-e . Photochem. Photobiol. 2007, 83, 1006–
015.
23) Jensen, T. J.; Vicente, M. G. H.; Luguya, R.; Norton, J.;
Fronczek, F. R.; Smith, K. M. Effect of overall charge and charge
6
1
6
(
7
475
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476

Journal of Medicinal Chemistry
ARTICLE
(
41) Kessel, D. Correlation between subcellular localization and
photodynamic efficacy. J. Porphyrins Phthalocyanines 2004, 8, 1009–
014.
42) Wan, Q.; Liu, L.; Xing, D.; Chen, Q. Bid is required in NPe6-
PDT-induced apoptosis. Photochem. Photobiol. 2008, 84, 250–257.
43) Reiners, J. J., Jr.; Caruso, J. A.; Mathieu, P.; Chelladurai, B.; Yin,
X. -M; Kessel, D. Release of cytochrome c and activation of pro-caspase-
following lysosomal photodamage involves bid cleavage. Cell Death
Differ. 2002, 9, 934–944.
1
(
(
9
7
476
dx.doi.org/10.1021/jm2005139 |J. Med. Chem. 2011, 54, 7464–7476