Synthesis and in vitro properties of trimethylamine- and phosphonate-substituted carboranylporphyrins for application in BNCT

Article abstract of DOI:10.1016/j.bmc.2007.12.020

A series of carboranylporphyrins containing either amine or phosphonic acid functionalities and two to six closo-carborane clusters have been synthesized via a [2 + 2] condensation of a dimethylamino- or diethylphosphonate-substituted dipyrromethane with

Full text of DOI:10.1016/j.bmc.2007.12.020

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 3191–3208
Synthesis and in vitro properties of trimethylamine-
and phosphonate-substituted carboranylporphyrins
for application in BNCT
Michael W. Easson, Frank R. Fronczek,* Timothy J. Jensen and M. Grac¸a H. Vicente*
Department of Chemistry, Louisiana State University, Baton Rouge, LA 70803-1804, USA
Received 22 October 2007; revised 5 December 2007; accepted 11 December 2007
Available online 21 February 2008
Abstract—A series of carboranylporphyrins containing either amine or phosphonic acid functionalities and two to six closo-carbo-
rane clusters have been synthesized via a [2 + 2] condensation of a dimethylamino- or diethylphosphonate-substituted dipyrrome-
thane with a dicarboranylmethyl-benzaldehyde. The X-ray structures of four key reaction intermediates (1, 2, 3, and 4a) and of two
target porphyrins, the diphosphonate ester- and the diamino-tetracarboranylporphyrins 5b and 6a, are presented and discussed. In
vitro studies using human carcinoma HEp2 and human glioblastoma T98G cells show that these porphyrins are non-toxic in the
dark up to 100 lM concentrations, and that a tetracarboranylporphyrin bearing two quaternary ammonium groups is the most effi-
ciently taken up by cells at short times (up to 8 h), followed by a dicarboranylporphyrin bearing three phosphonic acid substituents.
All carboranylporphyrins delivered therapeutic amounts of boron to T98G cells and localized mainly within the cell lysosomes.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
active area of research. Several boron-containing mole-
cules have been synthesized and investigated as boron
delivery agents for BNCT, including nucleosides, amino
acids, phospholipids, monoclonal antibodies, dendri-
mers, and porphyrins.10,11 Among these, porphyrins
are very promising boron delivery vehicles because of
their known tumor selectivity, low dark toxicity, and
long persistence within tumors.12,13 A promising BNCT
delivery porphyrin agent should exhibit the following
characteristics: (1) low dark toxicity, (2) high tumor cell
selectivity, that is, high tumor/normal tissue and tumor/
blood boron concentration ratios, (3) deliver therapeutic
concentrations of 10B to tumors, that is, >20 lg 10B/g tu-
mor,4 (4) localize within tumor cells, preferentially with-
in or near sensitive organelles such as the nuclei,14,15 (5)
have long retention times within tumors and rapid clear-
ance from normal tissues and blood, and (6) be easily
monitored by fluorescence techniques in order to facili-
tate radiation dosimetry. The major challenge in the
development of BNCT delivery agents has been the
selective delivery of therapeutic amounts of boron to
targeted tumors with minimal normal tissue toxicity.4
Boron neutron capture therapy (BNCT) is a binary ther-
apy that involves the activation of a tumor-localized
10B-containing sensitizer with low energy neutrons,
resulting in the release of high linear energy transfer
(high-LET) particles (4He2+ and 7Li3+) containing
2.4 MeV of kinetic energy.1–3 The short-range high-
LET particles are highly cytotoxic and thus can
selectively destroy 10B-containing malignant cells in the
presence of boron-free normal tissues. BNCT is particu-
larly attractive for the treatment of high-grade gliomas,
malignant melanomas, meningiomas, head and neck
tumors, and oral cancer.4 Two boron-containing
compounds, mercapto-closo-dodecaborate (BSH) and
4-dihydroxyborylphenylalanine (BPA), are currently
being used in BNCT clinical trials, in the USA, Europe,
and Japan.4–9 Since in the clinical studies performed to
date there has been evidence of a therapeutic response
to BNCT, the discovery of BNCT agents with higher
biological efficacy than BSH and BPA has been an
Herein we report the syntheses of new amphiphilic closo-
carboranylporphyrins (5d, 6c,d, 7c,d, and 8d) bearing
either positively charged quaternary ammonium or neg-
atively charged phosphonate water-solubilizing groups,
Keywords: BNCT; Carborane; Porphyrin; Cytotoxicity; Cellular
uptake.
*
Corresponding authors. Tel.: +1 225 578 7405; fax: +1 225 578 3458
(M.G.H.V.); e-mail: vicente@lsu.edu
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.12.020

3192
M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
and two to six closo-carborane cages.16 Only a few
water-soluble closo-carboranylporphyrins have been
previously reported in the literature, bearing carboxyl-
ates or a PEG as water-solubilizing groups.12 Our new
closo-carboranylporphyrins contain either phosphonate
groups that can form multiple hydrogen bonds with bio-
logical molecules, or quaternary ammonium salts that
can potentially target highly vulnerable intracellular
sites, such as the nuclei, and induce effective DNA and
RNA damage.17–20 The X-ray molecular structures of
four key intermediates (1, 2, 3, and 4a) and of two target
porphyrins (5b and 6c) are discussed. We investigated
and compared the in vitro toxicity, cellular uptake,
and subcellular distribution properties of this series of
compounds using both human carcinoma HEp2 and hu-
man glioblastoma T98G cells.
2. Results
2.1. Porphyrin syntheses
Porphyrins 5–8 containing two to six carborane cages
(18–38% boron by weight) were synthesized via a
[2 + 2] condensation of dicarboranyl-benzaldehyde 3
with an amine- or phosphonate-substituted dipyrrome-
thane 4, as shown in Scheme 1. Due to acid-catalyzed
scrambling in these reactions porphyrins 5–8 were ob-
CN
CN
CHO
i
ii
Me
Me
Me
R
Me
Br
Br
2
3
1
CHO
a: R = NMe2
= BH
= C
b: R = PO(OEt)2
c: R = NMe3 I
d: R = PO(OH)2
iii
iv, v
R
NH HN
4a,b
R
R
Me
Me
Me
NH
N
NH
N
N
R
N
HN
HN
Me
Me
Me
Me
Me
R
5b,d
6a-d
R
R
Me
Me
Me
NH
N
NH
N
R
N
HN
N
HN
Me
Me
Me
Me
Me
R
8b,d
7a-d
Scheme 1. Reagents and conditions: (i) 1-Lithium-2-methyl-o-carborane, THF, LiI, rt, 20 h (51%); (ii) DIBAL, toluene followed by 10% H2SO4
(62%); (iii) pyrrole, TFA, CH2Cl2, rt (70–86%); (iv) BF3OEt2, CH2Cl2, rt, 4 min–1 h; (v) DDQ, rt for 30 min (overall 17–22%).

M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
3193
tained in a single-pot reaction and isolated in 2–8%
yields, along with 2–9% of the corresponding octa-car-
boranylporphyrin (H2OCP),21 and <1% of other minor
porphyrin products. The benzonitrile 1, prepared
according to the procedure of Bodwell et al.,22 reacted
with in situ prepared 1-lithium-2-methyl-o-carborane
producing the dicarboranyl-benzonitrile 2 in 51% yield.
The cyano group of compound 2 was reduced to formyl
using DIBAL-H, giving benzaldehyde 3 in 62% yield.
We have previously reported a related synthesis of
dicarboranyl-benzaldehyde 323 from 3,5-dimethyl-
bromobenzene in 15% overall yield. Our new synthesis
of 3 relies on the readily reduction of the cyano group
of 2, as previously reported for other benzonitriles,24
rather than on a lithium/bromide exchange followed
by DMF addition and acidic hydrolysis,23 thus involv-
ing less steps. Dipyrromethanes 4 were prepared from
the reaction of the corresponding benzaldehyde with
an excess of pyrrole using TFA as the catalyst, in 70–
86% yields, as previously reported.25–27 The condensa-
tion of compounds 3 and 4a via a [2 + 2] methodol-
ogy,27–29 using BF3OEt2 for 1 h at room temperature,
followed by oxidation with 2,3-dichloro-5,6-dicyano-
1,4-benzoquinone (DDQ) gave porphyrins 6a, 7a and
H2OCP in 2.1%, 6.3% and 8.5% yields, respectively, as
major products. On the other hand, the condensation
of 3 and 4b under similar conditions but using BF3OEt2
for only 4 min afforded porphyrins 5b, 6b, 7b, 8b and
H2OCP in 3.6%, 2.6%, 8.0%, 4.5% and 2.1% yields,
respectively. Other porphyrins were also obtained as
minor products (<1% yield) as a result of the scrambling
but were not isolated in sufficient amounts for character-
ization. In order to achieve water-solubility, the phos-
phonate esters were cleaved to the corresponding
phosphonic acids using bromotrimethylsilane followed
by hydrolysis30 in 78–99% yields, and the dimethyl-
amino groups were quaternized with excess iodometh-
ane in dichloromethane,19 in 83–92% yields. While the
porphyrins containing either the diester phosphonate
or dimethylamino functionalities are highly soluble in
dichloromethane, the phosphonic acid-containing por-
phyrins are only soluble in DMSO and partially soluble
in water, and the trimethylammonium-porphyrins are
soluble in polar solvents (such as acetone and methanol)
and also partially water-soluble. Porphyrin 5d exhibited
particularly low solubility in all solvents, including
DMSO, and therefore was not evaluatedin in vitro studies.
Figure 1. The molecular structure of 1 as determined by X-ray
crystallography, with ellipsoids at the 50% level.
try-required to be rigorously coplanar. The entire por-
phyrin ring deviates little from coplanarity, exhibiting
˚
a mean deviation of 0.029 A for the 24-atom core and
˚
maximum deviation of 0.067(4) A. The phenyl rings car-
rying the phosphonates form dihedral angles of 70.9(1)ꢁ
with the porphyrin plane, and the phenyl rings carrying
the carboranes form dihedral angles of 67.0(1)ꢁ with it.
In porphyrin 6a, the four porphyrin N atoms in Figure
˚
6 are coplanar within 0.02(2) A, but the 24-atom por-
phyrin core deviates somewhat from planarity, its atoms
˚
exhibiting a mean deviation of 0.12 A from their best
˚
plane, with maximum deviation of 0.26(3) A. The distor-
tion is mostly a twist about an axis bisecting the pyrrole
between the two Me2NPh groups and the pyrrole be-
tween the two carborane-substituted phenyl groups.
This twist places the two NMe2 nitrogen atoms 1.54(4)
˚
and ꢀ0.70(3) A above and below the porphyrin best
plane, as illustrated in Figure 6a, a view down the twist
axis. The phenyl rings form dihedral angles which fall in
the range 71.3(6)–82.3(6)ꢁ with the porphyrin best plane.
2.3. Dark cytotoxicity
The dark toxicity of porphyrins 6c, 6d, 7c, 7d, and 8d to-
ward human carcinoma HEp2 cells was evaluated at
concentrations up to 100 lM and the results obtained
are shown in the Supplementary data. All compounds
were found to be non-toxic to HEp2 cells in the dark un-
der the conditions tested, up to 100 lM concentrations.
2.2. Molecular structures
Crystals of intermediate dibromobenzonitrile 1 were
grown from ethyl acetate/hexane, and those of carbora-
nyl-benzonitrile 2, carboranyl-benzaldehyde 3, and
dipyrromethane 4a were grown from CHCl3/hexane.
Their molecular structures are shown in Figures 1–4,
respectively, confirming their identities. Crystals of the
toluene solvate of porphyrin 5b were grown from
CHCl3/toluene, and those of the chloroform solvate of
6a were obtained by slow evaporation of a chloroform
solution over a period of several days. The X-ray struc-
tures of porphyrins 5b and 6a are shown in Figures 5
and 6, respectively. In the crystal, porphyrin 5b lies on
an inversion center, so the four N atoms are symme-
2.4. Cellular uptake
The time-dependent uptake of porphyrins 6c, 6d, 7c, 7d,
and 8d was evaluated in human glioblastoma T98G cells
at a concentration of 10 lM over a period of 24 h and
the results are shown in Figure 8. All porphyrins accu-
mulated rapidly in the first 2 h, after which a plateau
was reached for 6c, 6d, 7c, and 7d, while 8d continued
to steadily accumulate over time up to the 24 h period
investigated. Porphyrin 6c bearing two quaternary

3194
M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
Figure 2. The molecular structure of 2 with 50% ellipsoids. H atoms are not shown.
Figure 3. The molecular structure of 3 with 50% ellipsoids. H atoms are not shown.
ammonium groups on adjacent p-phenyl positions accu-
mulated the most of this series of compounds at short
time points (<4 h), while porphyrin 8d with three phos-
phonic acid groups accumulated the most at longer time
points (>8 h). Porphyrin 7c containing only one quater-
nary ammonium group was the least taken-up by the
T98G cells at all time points investigated. Similar uptake
results were also obtained using human carcinoma
HEp2 cells (results not shown).
rins 6c, 6d, 7c, 7d, and 8d, using both HEp2 and T98G
cells. The cells were exposed overnight to 10 lM of each
porphyrin and then examined for intracellular fluores-
cence. Similar results were obtained with both cell lines;
Figures 9b–13b show the fluorescence patterns observed
in HEp2 cells. For the co-localization experiments the
cells were incubated concurrently with porphyrin and
one of the following organelle tracers for 30 min: Mito-
Tracker Green 250 nM, LysoSensor Green 50 nM,
BODIPY FL C5-ceramide at 1 mM and ER Tracker
Green FM 2 lg/mL. The slides were washed three times
with growth medium and new medium containing
50 mM HEPES, pH 7.4, was added. The images were ac-
quired using a Zeiss Axiovert 200 M inverted fluores-
2.5. Intracellular localization
Fluorescence microscopy was used to investigate the
preferential sites of intracellular localization of porphy-

M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
3195
Figure 4. The molecular structure of 4a with 50% ellipsoids.
cence microscope fitted with standard FITC and Texas
Red filter sets. The preferential site of intracellular local-
ization for porphyrins 6c, 6d, 7c, 7d, and 8d was found
to be the lysosomes (Figs. 9f–13f).
porphyrins;16 in fact, no carborane-containing phospho-
nate-substituted porphyrins were obtained using this
methodology. The acid-catalyzed scrambling was more
significant in the [2 + 2] condensation of 4a with 3 prob-
ably as a result of the strong electron-donating effect of
the dimethylamino group and the longer reaction time
necessary to drive the reaction to completion; in this
case the trans-porphyrin 5a was not among the major
products of the reaction and H2OCP was the main prod-
uct obtained in 8.5% yield. Under the above optimal
conditions for the condensation of 4b and 3, porphyrin
7b bearing a single phosphonate ester was the major
product obtained, followed by 8b and 6b; in this case
H2OCP was only obtained in 2.1% yield. Although
acid-catalyzed scrambling reactions are generally
avoided in order to minimize challenging chromato-
graphic separations, this methodology allowed us to iso-
late reasonable amounts of several interesting porphyrin
targets for structure/in vitro biological activity
relationships.
3. Discussion
The new series of closo-carboranylporphyrins 5–8 was
prepared in a single reaction using a MacDonald-type
[2 + 2] condensation31 of either dimethylamino- or
diethylphosphonate–dipyrromethane with dicarboranyl-
benzaldehyde 3 (Scheme 1). We investigated various
reaction conditions that led to the highest yields of the
target porphyrins 5–8, bearing one to three water-solu-
bilizing groups and two to six closo-carborane cages,
including different acid catalysts, reagent concentra-
tions, solvents, reaction times, and temperatures.27,28
The optimal conditions involved the use of 10 mM
dipyrromethane and benzaldehyde concentrations in
dichloromethane and 2 equiv of BF3OEt2 for either 1 h
(in the case of 4a + 3) or 4 min (for 4b + 3). The use of
dipyrromethanes 4 in a [2 + 2] condensation rather than
a mixed aldehyde condensation afforded higher yields of
the target porphyrins containing at least one dipyrro-
methane motif, by minimizing the formation of
H2OCP, which was always a side product in these reac-
tions. When 5-(4-acetamidophenyl)dipyrromethane was
used in place of 4a, lower yields of the targeted porphy-
rins were obtained due to the additional step required to
cleave the acetamido group (ethanol/HCl at reflux for
24 h) and the isolation of a tetracarboranylcorrole as
one of the main products. The condensation of 5-
[bis(3,5-(2-methyl-o-carboran-1-yl)methyl)phenyl]dipyr-
romethane with phosphonate- or amino-substituted
benzaldehydes also gave lower yields of the targeted
All new compounds were characterized by spectroscopic
techniques (MS, H NMR, UV–Vis) and in addition,
1
benzonitriles 1 and 2, benzaldehyde 3, dipyrromethane
4a, and porphyrins 5b and 6a were also characterized
by X-ray crystallography. It is interesting to note that
while the porphyrin ring in porphyrin 5b, bearing two
phosphonic esters on opposite p-phenyl positions is
essentially flat, that of porphyrin 6a with two adjacent
p-dimethylaminophenyl groups is moderately distorted
from planarity, showing a twisted conformation. Such
porphyrin distortions from planarity have been previ-
ously observed32 and seem to be associated with the elec-
tronic interactions between the water-solubilizing
groups rather than with the 3,5-dicarboranylmethylphe-
nyl groups.21,23 Upon quaternization of the dimethyl-

3196
M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
Figure 5. The molecular structure of porphyrin 5b with 20% ellipsoids. Solvent molecules and H atoms are not shown.
amino groups of 6a, that is, in 6c, a more pronounced
distortion of the porphyrin is expected, as we have pre-
viously observed for a 5,10-di(4-trimethylamino)phenyl-
15,20-diphenylporphyrin.19
result of the formation of large aggregates and their sub-
sequent precipitation in aqueous media.35 Porphyrins
5d, 6c, 6d, 7c, 7d, and 8d showed similar absorption
and emission spectra in DMSO, typical for meso-tetra-
phenylporphyrins, as shown in Figure 7. However, in
HEPES buffer (20 mM, pH 7.4) containing 1% DMSO
all porphyrins showed significant broadening of their
emission bands and fluorescence quenching indicating
aggregation (see Fig. 7 and Supplementary data), as
we have previously observed.36 Interestingly, porphyrins
5d, 6c, 7c, and 7d displayed red-shifted (1–4 nm) emis-
sion bands, while porphyrins 6d and 8d showed blue-
shifted (3–5 nm) bands, suggesting the formation of dif-
ferent types of aggregates.
As a consequence of the highly hydrophobic character
of the closo-carborane cages, porphyrins 5b, 6a,b, 7a,b,
and 8b were completely insoluble in water. Two strate-
gies are usually employed for conferring partial
water-solubility to carboranylporphyrins,12 (1) the intro-
duction of water-solubilizing groups (such as amino, hy-
droxyl or carboxylate), or (2) the deboronation of
o-carborane cages to the corresponding anionic nido-
carboranes. In the present study we synthesized, charac-
terized, and compared the in vitro biological properties
of a new series of closo-carboranylporphyrins bearing
either positively charged quaternary ammonium or neg-
atively charged phosphonate water-solubilizing groups;
these compounds could potentially interact with
DNA33,34 and the phosphonate-substituted porphyrins
could form multiple hydrogen bonds with biological
substrates. Interestingly, porphyrin 5d bearing two
phosphonic acid substituents on opposite p-phenyl posi-
tions showed very limited water-solubility, probably as a
Porphyrins 6c, 6d, 7c, 7d, and 8d were all found to be
non-toxic to HEp2 cells in the dark up to 100 lM con-
centrations (see Supplementary data). Higher concentra-
tions were not investigated because the compounds
precipitated. Our results are in agreement with previous
studies reporting very low dark toxicities for both closo-
and nido-carboranylporphyrins.21,23,34,36–42 The cellular
uptake of this series of closo-carboranylporphyrins did
not increase with their hydrophobic character. We deter-

M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
3197
behavior of these polyprotic compounds, as we have
previously investigated.35 At physiologic pH the phos-
phonic acid groups mainly exist in their mono-anionic
form (i.e., as PO3Hꢀ) and the resulting anionic porphy-
rins can aggregate as a result of both p-p interactions
and hydrogen bond formation. It has been recently
shown that negatively charged cobaltacarborane-por-
phyrins containing up to eight negative charges form
large aggregates in aqueous solutions.36,43 On the other
hand, porphyrin 6c, bearing two quaternary ammonium
groups, accumulated within T98G cells about 2.5 times
more than 7c with only one positive charge; in fact por-
phyrin 7c was the least taken up by cells of all the por-
phyrins studied and delivered the least amount of boron
to T98G cells. The higher amphiphilicity of porphyrin 6c
compared with 7c as a result of its more favorable lipo-
philic/hydrophilic balance as a consequence of its charge
distribution might be responsible for the observed re-
sults. The uptake kinetics for this series of carboranyl-
porphyrins is similar to that previously observed for
both closo- and nido-carboranylporphyrins.21,36,37,39,44,45
The preferential sites of intracellular localization for
porphyrins 6c, 6d, 7c, 7d, and 8d were found to be the
cell lysosomes, as evidenced by the punctate pattern ob-
served by fluorescence microscopy and its overlay with
the lysosome-specific probe LysoSensor Green. These
results are in agreement with previous studies showing
that carboranylporphyrins localize within cell lyso-
somes,21,36,39,45–48 maybe as a result of an endocytic
mechanism of uptake.49,50
4. Conclusions
The syntheses of six new amphiphilic closo-carboranyl-
porphyrins bearing either positively charged quaternary
ammonium (6c,7c) or negatively charged phosphonate
(5d, 6d, 7d, and 8d) groups are reported. The X-ray
molecular structures of four intermediates (1, 2, 3, and
4a) and of two target porphyrins (5b and 6c) are dis-
cussed. All porphyrins were found to be non-toxic in
the dark up to 100 lM concentrations and the extent
of their cellular uptake did not increase with their
hydrophobic character. The tetracarboranylporphyrin
6c, bearing two quaternary ammonium groups, was
the most efficiently taken up by human glioma T98G
cells at short times (up to 8 h), followed by di-
carboranylporphyrin 8d bearing three phosphonic acid
substituents. The hexacarboranylporphyrin 7d was
found to deliver the highest amount of boron to T98G
cells after 24 h. The preferential sites of subcellular
localization of this series of porphyrins were found to
be the cell lysosomes.
Figure 6. The molecular structure of porphyrin 6a with 20% spheres.
(a) top view; (b) side view. Solvent molecules and H atoms are not
shown.
mined the following order of increasing uptake at times
<8 h: 6c > 8d > 7d > 6d > 7c (Fig. 8). At longer time
periods porphyrin 8d, bearing three phosphonic acid
groups, was the most accumulated within cells, and at
24 h twice the amount of this porphyrin was found with-
in cells than 6d, containing two phosphonic acid groups
on adjacent phenyl rings. Porphyrin 7d with only one
phosphonic acid substituent and the highest percentage
of boron by weight of all porphyrins showed slightly
higher uptake than 6d at all time points investigated
and delivered the highest amount of boron to T98G cells
after 24 h (see Table 1 in Supplementary data). After
24 h the calculated amount of intracellular boron (in
lg per billion cells or per gram of wet tissue) delivered
by the porphyrins to T98G cells was between
92.6 16.2 (for 7c) and 220.0 35.0 (for 7d). Therefore,
all porphyrins are able to deliver therapeutic amounts of
boron to human glioblastoma T98G cells. The differ-
ences in uptake observed for the phosphonate-contain-
ing porphyrins might be due to the unique aggregation
5. Experimental
5.1. Chemistry
All reactions were carried out under an argon atmo-
sphere in dried solvents. Commercially available starting
compounds were purchased from Sigma-Aldrich or

3198
M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
a
b 400
400
350
300
250
200
150
100
50
350
300
250
200
150
100
50
0
0
600
650
700
750
600
650
700
750
Wavelength (nm)
Wavelength (nm)
d 400
400
350
300
250
200
150
100
50
c
350
300
250
200
150
100
50
0
0
600
650
700
750
600
650
700
750
Wavelength (nm)
Wavelength (nm)
400
f
400
350
300
250
200
150
100
50
e
350
300
250
200
150
100
50
0
0
600
650
700
750
600
650
700
750
Wavelength (nm)
Wavelength (nm)
Figure 7. Fluorescence emission spectra of porphyrins 5d (a), 6c (b), 6d (c), 7c (d), 7d (e), and 8d (f), at 1 · 10ꢀ6 M in DMSO (solid line), and in
freshly prepared HEPES buffer (20 mM, pH 7.4) containing 1% DMSO (dashed line). Excitation at 420 nm.
from Katchem, Ltd. and used directly without further
purification. Silica gel 60 (70–230 mesh, Merck) was
used for column chromatography. Analytical thin-layer
chromatography (TLC) was performed using Merck 60
F254 silica gel (precoated sheets, 0.2 mm thick). 1H
NMR spectra were obtained using a Bruker DPX
250 MHz spectrometer; chemical shifts are expressed
in ppm relative to TMS (0 ppm). Electronic absorption
spectra were measured on a Perkin Elmer Lambda 35
UV–Vis spectrophotometer. Low resolution mass spec-

M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
3199
Figure 8. Time-dependent uptake of porphyrins 6c (purple), 6d (black), 7c (blue), 7d (red), and 8d (green) at 10 lM by T98G cells.
tra (MS) were obtained on a Bruker Proflex III MAL-
DI-TOF mass spectrometer. High resolution mass spec-
tra (HRMS) were obtained using ESI-TOF under
negative mode on an Agilent Technologies Time-of-
Flight LC/MS 6210; the isotope peaks were matched
with calculated isotope patterns (see Supplementary
data) and only the most abundant peaks are listed below
for each porphyrin. Melting points were measured on a
MELT-TEMP apparatus.
The crude residue was dissolved in ethyl acetate and
washed once with water, once with brine, dried over
Na2SO4, and the solvent removed under vacuum. The
yellow residue was dried onto silica and purified by silica
gel column chromatography using ethyl acetate/hexane
1:4 for elution. The pure title compound was obtained
as an off-white solid (2.70 g) in 51% yield (40% conver-
sion). Recovered 1.05 g of compound 1. Mp = 235–
237 ꢁC; MS (ESI) m/z 443.24; 1H NMR (CDCl3) d
ppm 1.4–3.0 (br s, 20H, BH), 2.22 (s, 6H, CH3), 3.58
(s, 4H, CH2), 7.36 (s, 1H, p-ArH), 7.56 (s, 2H, o-ArH).
5.1.1. Bis(3,5-bromomethyl)benzonitrile (1).22 To a
round-bottomed flask containing dichloromethane
(460 mL) were added 3,5-dimethylbenzonitrile (5.00 g,
38 mmol), NBS (15.00 g, 84 mmol), and benzoyl perox-
ide (0.048 g, 0.20 mmol). The reaction mixture was
heated to reflux while stirring under argon. Upon reach-
ing reflux a white floodlight was positioned within four
inches of the reaction vessel. After 1 h the reaction mix-
ture was cooled to room temperature and 50 mL of dis-
tilled water was added to quench the reaction. The
organic phase was washed once with water, once with
brine, and then dried over Na2SO4, and the solvent re-
moved under vacuum to yield a yellow/orange residue.
The crude residue was dried onto silica and purified by
silica gel column chromatography using ethyl acetate/
hexane 1:4 for elution. The pure product was collected
affording 4.47 g of the title compound as a white solid
in 40% yield. Mp = 114–115 ꢁC; MS (GC) m/z 289.8
5.1.3. Bis[3,5-(2-methyl-o-carboran-1-yl)methyl]benzalde-
A solution of compound 2 (11.14 g,
hyde (3).
25.14 mmol) in dry toluene (500 mL) was flushed with
argon for 10 min. The reaction vessel was then cooled
to 0 ꢁC and DIBAL-H (1.5 M in toluene) (36.87 mL,
55.3 mmol) was slowly added dropwise while stirring
under argon. The reaction was heated to 60–65 ꢁC for
39 h, then cooled to 0 ꢁC and a cold 10% H2SO4 aque-
ous solution (228 mL) was slowly added to avoid foam-
ing. Toluene (800 mL) was added and the reaction
mixture was stirred for an additional 62 h under argon.
The organic phase was washed once with Na2HCO3,
once with water, once with brine, dried over Na2SO4,
and the solvent removed under vacuum to yield a crude
white solid. This solid was dissolved in dichlorometh-
ane, dried onto silica, and purified by silica gel column
chromatography using 1:1 hexane/dichloromethane for
elution. Pure title compound was obtained as a white so-
lid (7.07 g) in 62% yield. The spectroscopic data for the
title compound are in full agreement with those reported
in the literature.21,23
1
[M+1]; H NMR (CDCl3) d ppm: 4.50 (s, 4H, CH2),
7.65 (s, 2H, o-ArH), 7.68 (s, 1H, p-ArH).
5.1.2. Bis[3,5-(2-methyl-o-carboran-1-yl)methyl]benzoni-
trile (2). n-Butyllithium (23.63 mL, 1.6 M in hexane)
was added dropwise to a solution of 1-methyl-o-carbo-
rane (4.76 g, 30 mmol) in dry THF (450 mL) at a tem-
perature between ꢀ5 and 0 ꢁC, under argon. The
mixture was stirred at this temperature for 1.5 h, and
5.1.4. 5-(4-Dimethylamino)phenyldipyrromethane (4a).
The title compound was prepared as described in the lit-
erature and its spectroscopic data are in full agreement
with those previously reported.27
then compound
1 (4.37 g, 15.10 mmol) and LiI
(0.628 g, 4.69 mmol) in THF (50 mL) were added. The
final reaction mixture was allowed to warm to room
temperature and after 20 h the reaction was quenched
with water and the solvent removed under vacuum.
5.1.5. 5-(4-Phosphonate diethylester)phenyldipyrrome-
thane (4b). A solution of 4-(phosphonate diethyl ester)
benzaldehyde (0.047 g, 0.20 mmol) and freshly distilled

3200
M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
Figure 9. Intracellular localization of porphyrin 6c in HEp2 cells at 10 lM for 24 h. (a) phase contrast; (b) porphyrin fluorescence; (c) BODIPY FL-
C5-Ceramide fluorescence (Golgi label); (d) overlay with porphyrin; (e) LysoSensor fluorescence (lysosome label); (f) overlay; (g) MitoTracker
fluorescence (mitochondria label); (h) overlay; (i) ERTracker Green FM fluorescence (ER label); (j) overlay. Scale bar: 10 lm.
pyrrole (0.80 mL, 11.5 mmol) was purged with nitro-
gen for 5 min. TFA (2 lL, 0.023 mmol) was added
and the final solution was stirred at room temperature
in a sealed reaction vessel for 20 min. The light green
crude material was dried onto silica and purified by
silica gel column chromatography using methanol/
dichloromethane 1:33. The title compound was ob-
tained (65 mg) in 86% yield, as a yellow residue. 1H

M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
3201
Figure 10. Intracellular localization of porphyrin 6d in HEp2 cells at 10 lM for 24 h. (a) phase contrast; (b) porphyrin fluorescence; (c) BODIPY FL-
C5-Ceramide fluorescence (Golgi label); (d) overlay with porphyrin; (e) LysoSensor fluorescence (lysosome label); (f) overlay; (g) MitoTracker
fluorescence (mitochondria label); (h) overlay; (i) ERTracker Green FM fluorescence (ER label); (j) overlay. Scale bar: 10 lm.
NMR (CDCl3) d ppm: 1.36 (t, 6H, CH3, J = 7.0 Hz),
4.10 (m, 4H, CH2), 5.60 (s, 1H, CH), 6.01 (s, 2H, o-
ArH), 6.18 (s, 2H, m-ArH), 6.78 (s, 2H, a-H), 7.35
(m, 2H, b-H), 7.62 (m, 2H, b-H), 9.56 (br s, 2H, NH).
5.2. General procedure for porphyrin synthesis
A solution of dipyrromethane 4 (1.12 mmol) and
bis[3,5-(2-methyl-o-carboran-1-yl)methyl]benzaldehyde 3

3202
M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
Figure 11. Intracellular localization of porphyrin 7c in HEp2 cells at 10 lM for 24 h. (a) phase contrast; (b) porphyrin fluorescence; (c) BODIPY FL-
C5-Ceramide fluorescence (Golgi label); (d) overlay with porphyrin; (e) LysoSensor fluorescence (lysosome label); (f) overlay; (g) MitoTracker
fluorescence (mitochondria label); (h) overlay; (i) ERTracker Green FM fluorescence (ER label); (j) overlay. Scale bar: 10 lm.
(0.498 g, 1.12 mmol) in 112 mL of dichloromethane was
purged with argon for 15 min. BF3 Æ OEt2 (0.283 mL,
2.23 mmol) was added and the reaction mixture was stir-
red in a sealed reaction vessel. After 1 h (for 4a) or 4 min
(for 4b) DDQ (0.336 g, 1.48 mmol) and Et3N (0.31 mL,
2.23 mmol) were added and the reaction mixture stirred

M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
3203
Figure 12. Intracellular localization of porphyrin 7d in HEp2 cells at 10 lM for 24 h. (a) phase contrast; (b) porphyrin fluorescence; (c) BODIPY FL-
C5-Ceramide fluorescence (Golgi label); (d) overlay with porphyrin; (e) LysoSensor fluorescence (lysosome label); (f) overlay; (g) MitoTracker
fluorescence (mitochondria label); (h) overlay; (i) ERTracker Green FM fluorescence (ER label); (j) overlay. Scale bar: 10 lm.
for 30 min. The mixture was then dried onto silica and
filtered through a plug of silica gel using dichlorometh-
ane/hexane 2:1 for elution. The crude mixture was fur-
ther purified using preparative silica TLC plates and
dichloromethane/hexane 3:1 for elution to yield the
porphyrins.
5.2.1. Deprotection. The diethyl ester porphyrins were
dissolved in dichloromethane and purged with argon
for 5 min. The reaction vessel was then sealed and
bromotrimethylsilane was added by microliter syringe
(16 equiv/phosphonate diester) while stirring at room
temperature. The final mixture was stirred for 24 h.

3204
M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
Figure 13. Intracellular localization of porphyrin 8d in HEp2 cells at 10 lM for 24 h. (a) phase contrast; (b) porphyrin fluorescence; (c) BODIPY FL-
C5-Ceramide fluorescence (Golgi label); (d) overlay with porphyrin; (e) LysoSensor fluorescence (lysosome label); (f) overlay; (g) MitoTracker
fluorescence (mitochondria label); (h) overlay; (i) ERTracker Green FM fluorescence (ER label); (j) overlay. Scale bar: 10 lm.
All solvents were removed under vacuum and the
crude green residue was dissolved in a 1:4 H2O/THF
solution and allowed to stir for 2 h. The organic sol-
vent was removed under vacuum and the deprotected
porphyrins were filtered and washed repeatedly with
distilled water.

M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
3205
5.2.2. Quaternization. The dimethylamino porphyrins
were dissolved in a 2:5 solution of CH2Cl2/CH3I and
the resulting mixture allowed to stand without stirring
in a sealed reaction vessel for 72 h at room temperature.
The quaternized products precipitated out of solution
and were filtered and washed repeatedly with dichloro-
methane. Porphyrins 6c and 7c were partially demethy-
lated upon standing over a period of one week (for 6c)
or one month (for 7c), thus reforming the starting por-
phyrins 6a and 7a, respectively.
(m, 8H, CH2), 7.58 (s, 2H, ArH), 8.13 (s, 4H, ArH),
8.25 (m, 4H, ArH), 8.41 (m, 4H, ArH), 8.97 (m, 8H,
b-H). UV–Vis (CH2Cl2) kmax: 419 nm (e 352,900), 514
(18,000), 549 (6800), 589 (5500), 643 (3400). Deprotec-
tion afforded the corresponding phosphonic acid por-
phyrin (6d) in 78% yield as a green solid. Mp >300 ꢁC;
HRMS (ESI) m/z 1457.0260 [M+H]+, calculated for
C60H88B40N4O6P2 1456.0195; 1H NMR (d-DMSO) d
ppm: ꢀ2.92 (br s, 2H, NH), 1.3–3.4 (br s, 40H, BH),
2.27 (s, 12H, CH3), 3.97 (s, 8H, CH2), 7.72 (s, 2H,
ArH), 8.08–8.16 (br s, 8H, ArH), 8.32 (br s, 4H, ArH),
8.84 (m, 8H, b-H). UV–Vis (DMSO) kmax: 419 nm (e
395,100), 514 (17,200), 549 (8300), 589 (5900), 645
(4900).
5.2.3. 5,15-Bis-(4-phosphonate diethyl ester)-10,20-
bis[3,5-(2-methyl-o-carboran-1-yl)methylphenyl]porphyrin
(5b). This porphyrin was obtained (31 mg, 3.6% yield) as
a purple/red solid. Mp: >300 ꢁC; MS (MALDI) m/z
1
1552.78 [MꢀCH3]+; H NMR (CD2Cl2) d ppm: ꢀ2.80
5.2.6.
5-(4-Dimethylaminophenyl)-10,15,20-tris[3,5-(2-
(br s, 2H, NH), 1.60 (t, 12H, CH3, J = 7.0 Hz), 1.3–3.4
(br s, 40H, BH), 2.27 (s, 12H, CH3), 3.85 (s, 8H,
CH2), 4.40 (m, 8H, CH2), 7.58 (s, 2H, ArH), 8.12 (s,
4H, ArH), 8.26 (m, 4H, ArH), 8.40 (m, 4H, ArH),
8.95 (m, 8H, b-H). UV–Vis (CH2Cl2) kmax: 419 nm (e
331,000), 514 (15,400), 548 (6000), 588 (4800), 643
(2800). Deprotection afforded the corresponding phos-
phonic acid porphyrin (5d) in 83% yield as a green solid.
Mp: >300 ꢁC; HRMS (ESI) m/z 1457.0262 [M+H]+, cal-
culated for C60H88B40N4O6P2 1456.0195; 1H NMR
(DMSO-d6) d ppm: ꢀ2.92 (br s, 2H, NH), 2.21 (s,
12H, CH3), 1.3–3.4 (br s, 40H, BH), 3.92 (s, 8H,
CH2), 7.63 (s, 2H, ArH), 8.15 (br s, 8H, ArH), 8.29
(br s, 4H, ArH), 8.85 (br s, 8H, b-H). UV–Vis (DMSO)
methyl-o-carboran-1-yl)methylphenyl] porphyrin (7a).
This porphyrin was obtained in 6.3% yield as a red/
brown solid. Mp >300 ꢁC; MS (MALDI) m/z 1679.10
1
[M]+; H NMR (CD2Cl2) d ppm: ꢀ2.73 (s, 2H, NH),
1.25–3.1 (br s, 60H, BH), 2.27 (s, 18H, CH3), 3.29 (s,
6H, CH3), 3.85 (s, 12H, CH2), 7.20 (br s, 2H, ArH),
7.57 (s, 3H, ArH), 8.12–8.16 (m, 8H, ArH), 8.90 (br s,
6H, b-H), 9.10 (br s, 2H, b-H). UV–Vis (DMSO) kmax
:
418 nm (e 253,900), 518 (14,300), 562 (11,000), 592
(8100), 653 (6300). Quaternization afforded the corre-
sponding quaternary ammonium iodide salt (7c) in
92% yield as purple red solid. Mp 186.3 ꢁC; HRMS
(ESI) m/z 1695.7502 [M+H–I]+, calculated for
C71H122B60N5 1694.5730; 1H NMR (acetone-d6) d
ppm: ꢀ2.78 (s, 2H, NH), 1.25–3.1 (br s, 60H, BH),
2.38 (s, 18H, CH3), 4.09 (s, 12H, CH2), 4.28 (s, 9H,
CH3), 7.85 (s, 3H, ArH), 8.23 (s, 6H, ArH), 8.60 (s,
k
max: 420 nm (e 379,800), 515 (18,300), 550 (9100), 589
(6300), 645 (5300).
5.2.4. 5,10-Bis-(4-dimethylaminophenyl)-15,20-bis[3,5-(2-
methyl-o-carboran-1-yl)methylphenyl] porphyrin (6a).
This porphyrin was obtained in 2.1% yield as a black so-
4H, ArH), 8.92 (m, 8H, b-H); UV–Vis (DMSO) kmax:
420 nm (e 389,800), 514 (14,900), 548 (7700), 588
(5800), 644 (4400).
1
lid. Mp: >300 ꢁC, MS (MALDI) m/z 1380.90 [M+]; H
NMR (CD2Cl2) d ppm: ꢀ2.69 (s, 2H, NH), 1.25–3.1
(br s, 40H, BH), 2.24 (s, 12H, CH3), 3.25 (s, 12H,
CH3), 3.79 (s, 8H, CH2), 7.17 (br s, 4H, ArH), 7.51 (s,
2H, ArH), 8.07–8.12 (m, 8H, ArH), 8.85 (br s, 4H, b-
H), 8.99 (s, 2H, b-H), 9.03 (br s, 2H, b-H). UV–Vis
(DMSO) kmax: 423 nm (e 171,000), 523 (13,500), 568
(14,200), 595 (8500), 657 (7300). Quaternization of this
porphyrin afforded the corresponding quaternary
ammonium iodide salt (6c) in 83% yield as a purple/
red solid. Mp: >300 ꢁC, HRMS (ESI) m/z 706.1089
[Mꢀ2I]2+, calculated for C66H102B40N6 706.1090; 1H
NMR (acetone-d6) d ppm: ꢀ2.79 (s, 2H, NH), 1.25–
3.1 (br s, 40H, BH), 2.37 (s, 12H, CH3), 4.07 (s, 8H,
CH2), 4.29 (s, 18H, CH3), 7.86 (s, 2H, p-ArH), 8.22 (s,
4H, ArH), 8.55 (br s, 4H, ArH), 8.65 (br s, 4H, ArH),
5.2.7. 5-(4-Phosphonic acid diethyl ester)-10,15,20-
tris[3,5-(2-methyl-o-carboran-1-yl)methyl phenyl]porphy-
rin (7b). This porphyrin was obtained in 8.0% yield as a
purple solid. Mp 294.7 ꢁC; MS (MALDI) m/z 1756.10
[MꢀCH3]+; 1H NMR (CD2Cl2) d ppm: ꢀ2.80 (br s,
2H, NH), 1.36 (t, 6H, CH3, J = 7.0 Hz), 1.3–3.4 (br s,
60H, BH), 2.28 (s, 18H, CH3), 3.86 (s, 12H, CH3),
4.38 (m, 4H, CH2), 7.59 (s, 3H, ArH), 8.12 (s, 6H,
ArH), 8.26 (m, 2H, ArH), 8.41 (m, 2H, ArH), 8.97 (m,
8H, b-H). UV–Vis (CH2Cl2) kmax: 419 nm (e 359,100),
514 (21,900), 548 (11,000), 588 (7400), 643 (3500).
Deprotection afforded the corresponding phosphonic
acid porphyrin (7d) in 94% yield as a black solid. Mp
>300 ꢁC; HRMS (ESI) m/z 1717.4781 [M+H]+, calcu-
lated for C68H115B60N4O3P 1716.4735; 1H NMR
(DMSO-d6) d ppm: ꢀ2.91 (s, 2H, NH), 2.24 (s, 18H,
CH3), 1.3–3.4 (br s, 60H, BH), 3.92 (s, 12H, CH2),
7.68 (br s, 3H, ArH), 8.12 (br s, 8H, ArH), 8.26 (br s,
8.88–8.99 (m, 8H, b-H). UV–Vis (DMSO) kmax
:
419 nm (e 406,100), 514 (11,800), 548 (6300), 588
(4500), 644 (3200).
2H, ArH), 8.86 (br s, 8H, b-H). UV–Vis (DMSO) kmax
420 nm (e 418,100), 515 (16,900), 549 (8000), 589 (5700),
645 (4700).
:
5.2.5. 5,10-Bis-(4-phosphonic acid diethyl ester)-15,20-
bis[3,5-(2-methyl-o-carboran-1-yl)methylphenyl]porphyrin
(6b). This porphyrin was obtained in 2.6% yield as a
purple solid. Mp 282.8 ꢁC; MS (MALDI) m/z 1552.78
[MꢀCH3]+; 1H NMR (CD2Cl2) d ppm: ꢀ2.83 (br s,
2H, NH), 1.55 (t, 12H, CH3, J = 7.0 Hz), 1.3–3.4 (br s,
40H, BH), 2.27 (s, 12H, CH3), 3.85 (s, 8H, CH2), 4.39
5.2.8. 5,10,15-Tris-(4-phosphonic acid diethyl ester)-20-
bis[3,5-(2-methyl-o-carboran-1-yl)methylphenyl]porphyrin
(8b). This porphyrin was obtained in 4.5% yield as a
purple solid. Mp 267 ꢁC; MS (MALDI) m/z 1362.40

3206
M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
3
[M]+; 1H NMR (CD2Cl2) d ppm: ꢀ2.83 (br s, 2H, NH),
1.55 (t, 18H, CH3, J = 7.0 Hz), 1.3–3.4 (br s, 20H, BH),
2.27 (s, 6H, CH3), 3.84 (s, 4H, CH2), 4.40 (m, 12H,
CH3), 7.58 (s, 1H, ArH), 8.08 (s, 2H, ArH), 8.26 (m,
6H, ArH), 8.42 (m, 6H, ArH), 8.97 (m, 8H, b-H).
c = 20.075(4) A, b = 96.093(12)ꢁ, V = 1410.7(5) A ,
˚
˚
Z = 4,
q
calcd = 1.249 g cmꢀ3
,
l = 0.075 mmꢀ1
,
T =
110 K, 22814 data collected, R = 0.045 (F2 > 2r),
Rw = 0.121 (all F2) for 5080 unique data having
h < 32.5ꢁ and 189 refined parameters.
UV–Vis (CH2Cl2) kmax
: 419 nm (e 385,000), 514
(16,300), 549 (6700), 589 (5000), 643 (3000). Deprotec-
tion afforded the corresponding phosphonic acid por-
phyrin (8d) in quantitative yield as a blue-green solid.
Mp: >300 ꢁC; HRMS (ESI) m/z 1196.5739 [M+H]+, cal-
culated for C52H61B20N4O9P3 1195.5655; 1H NMR
(DMSO) d ppm: ꢀ2.94 (br s, 2H, NH), 1.3–3.4 (br s,
20H, BH), 2.27 (s, 6H, CH3), 3.96 (br s, 4H, CH2),
7.72 (s, 1H, ArH), 8.08–8.16 (m, 8H, ArH), 8.31–8.34
(m, 6H, ArH), 8.84 (br s, 8H, b-H). UV–Vis (DMSO)
Crystal data for 5b: brown, C68H104B40N4O6P2Æ3C7H8,
Mr = 1844.3,
monoclinic
space
group
C2/c,
˚
a = 25.249(7), b = 22.017(6), c = 19.894(5) A, b =
105.52(2)ꢁ, V = 10656(5) A , Z = 4, qcalcd = 1.150 g cmꢀ3
,
3
˚
l = 0.092 mmꢀ1
, T = 110 K, 48441 data collected,
R = 0.112 (F2 > 2r), Rw = 0.343 (all F2) for 7630 unique
data having h < 23.3ꢁ and 569 refined parameters. All
toluene solvent molecules are disordered, and were trea-
ted in half-populated sites with isotropic C atoms and a
common Uiso for each molecule. Solvent H atoms were
not included.
k
max: 420 nm (e 331,700), 515 (14,600), 549 (7100), 589
(5100), 645 (4200).
5.3. Molecular structures
Crystal data for 6a: Black, C64H96B40N6Æ4CHCl3,
Mr = 1859.3, triclinic space group P1, a = 11.3640(12),
ꢀ
˚
Crystal structures were determined using data collected
at low temperature on a Nonius KappaCCD diffractom-
eter equipped with Mo Ka radiation (k = 0.71073 A)
b = 18.753(2), c = 23.905(4) A, a = 85.692(5), b =
3
˚
86.915(5), c = 73.912(6)ꢁ, V = 4878.1(11) A , Z = 2,
calcd = 1.266 g cmꢀ3
,
l = 0.384 mmꢀ1
,
T = 106 K,
˚
q
and a graphite monochromator. Refinement was by
full-matrix least squares, and all nonhydrogen atoms
were treated anisotropically with H atoms included, ex-
cept as described below. Crystallographic data (exclud-
ing structure factors) for the structures in this paper
have been deposited with the Cambridge Crystallo-
graphic Data Centre as supplementary publication
nos. CCDC 663420–663423, CCDC 663925, and CCDC
644633. Copies of the data can be obtained, free of
charge, on application to CCDC, 12 Union Road, Cam-
bridge CB2 1EZ, UK (Fax: +44 (0)1223 336033 or
e-mail: deposit@ccdc.cam.ac.uk).
11914 data collected, R = 0.171 (F2 > 2r), Rw = 0.457
(all F2) for 7736 unique data having h < 20.0ꢁ and 573
refined parameters. Due to the very weak scattering
power of the crystals even at low temperature, it was
not possible to do full anisotropic refinement. Only the
Cl atoms were refined anisotropically, while the lighter
atoms were treated as isotropic.
5.4. Absorption and emission spectra
The absorption spectra were measured on a Perkin El-
mer Lambda 35 UV–Vis spectrometer with 10 mm path
length quartz cuvettes equipped with a photodiode
detector, in the 300–700 nm wavelength range with
0.1 nm accuracy. Emission spectra were obtained on a
Perkin Elmer LS55 luminescence spectrophotometer
equipped with a Xenon lamp, in the 500–700 nm wave-
length range with 1 nm accuracy. Stock solutions of all
porphyrins in DMSO at 1 · 10ꢀ4 M were prepared.
The aqueous solutions were prepared by adding
100 lL of each porphyrin DMSO stock solution into
10 mL of HEPES (N-2-hydroxyethylpiperazine-N0-2-
ethanesulfonic acid) buffer, 20 mM, pH 7.4. All mea-
surements were performed within 2 h of solution
preparation.
Crystal data for 1: colorless, C9H7Br2N, Mr = 289.0,
monoclinic space group P21/c, a = 7.574(2), b =
˚
4.4039(10), c = 28.315(8) A, b = 96.475(8)ꢁ, V =
3
938.4(4) A , Z = 4, qcalcd = 2.045 g cmꢀ3, l = 8.579 mmꢀ1
,
˚
T = 90 K, 8408 data collected, R = 0.021 (F2 > 2r),
Rw = 0.050 (all F2) for 2279 unique data having
h < 28.2ꢁ and 110 refined parameters. Absorption cor-
rections by multi-scan method.
Crystal data for 2: colorless, C15H33B20N, Mr = 443.6,
monoclinic space group P21/n, a = 17.526(2),
˚
b = 7.8927(10), c = 20.262(3) A, b = 109.743(6)ꢁ, V =
3
2638.0(6) A ,
˚
Z = 4,
q
calcd = 1.117 g cmꢀ3
T = 150 K, 21359 data collected,
,
l =
0.052 mmꢀ1
,
5.5. Cell culture
R = 0.071 (F2 > 2r), Rw = 0.202 (all F2) for 5089 unique
data having h < 27.1ꢁ and 328 refined parameters.
HEp2 and T98G cells were acquired from the ATCC
and maintained in a 50:50 mix of Advanced MEM:-
DMEM supplemented with 5% FBS (all from Invitro-
gen). The cultures were passaged twice weekly to
maintain subconfluent stocks. Each porphyrin was pre-
pared as a 10 mM stock in DMSO and diluted to a
working concentration of 10 lM in two stages. First,
equal volumes of 10 mM stock and Cremophor EL
(Fluka) were mixed in a microfuge tube and sonicated
using a bath sonicator for 10 min. This mixture was then
diluted with DMSO to a final porphyrin concentration
of 1 mM. This was then diluted by dropwise addition
Crystal data for 3: colorless, C15H34B20O, Mr = 446.6,
orthorhombic space group P212121, a = 7.8923(10),
3
˚
˚
b = 12.211(2), c = 27.548(5) A, V = 2654.9(7) A , Z =
4, qcalcd = 1.117 g cmꢀ3, l = 0.054 mmꢀ1, T = 110 K,
22125 data collected, R = 0.050 (F2 > 2r), Rw = 0.143
(all F2) for 4549 unique data having h < 30.5ꢁ and 327
refined parameters.
Crystal data for 4a: tan, C17H19N3, Mr = 265.4, mono-
clinic space group P21/n, a = 8.0907(10), b = 8.735(2),

M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
3207
of medium to give a final concentration of 10 lM. The
medium-containing porphyrin was then sonicated for
30 min.
with this article can be found, in the online version, at
doi:10.1016/j.bmc.2007.12.020.
5.5.1. Dark cytotoxicity. HEp2 cells were plated as de-
scribed above and allowed 36–48 h to attach. The cells
were then exposed to various concentrations of porphy-
rin up to 100 lM and incubated overnight. The loading
medium was supplemented with CellTitre Blue (Prome-
ga) as per manufacturer’s instructions. Cell viability was
measured by reading the fluorescence at 520/584 nm
using a BMG FLUOstar plate reader. The signal was
normalized to 100% viable (untreated cells) and 0% via-
ble (cells treated with 0.2% saponin (Sigma)).
References and notes
1. Barth, R. F.; Soloway, A. H.; Goodman, J. H.; Gahbauer,
R. A.; Gupta, N.; Blue, T. E.; Yang, W.; Tjarks, W.
Neurosurgery 1999, 44, 433–451.
2. Barth, R. F. J. Neuro-Oncol. 2003, 62, 1–5.
3. Barth, R. F.; Coderre, J. A.; Vicente, M. G. H.; Blue, T. E.
Clin. Cancer Res. 2005, 11, 3987–4002.
4. Barth, R. F.; Joensuu, H. Radiother. Oncol. 2007, 82, 119–
122.
5. Nakagawa, Y.; Pooh, K.; Kobayashi, T.; Kageji, T.;
Uyama, S.; Matsumura, A.; Kumada, H. J. Neuro-Oncol.
2003, 62, 87–99.
6. Diaz, A. Z. J. Neuro-Oncol. 2003, 62, 101–109.
7. Busse, P. M.; Harling, O. K.; Palmer, M. R.; Kiger, W. S.,
III; Kaplan, J.; Kaplan, I.; Chuang, C. F.; Goorley, J. T.;
Riley, K. J.; Newton, T. H.; Santa Cruz, G. A.; Lu, X.-Q.;
Zamenhof, R. G. J. Neuro-Oncol. 2003, 62, 111–121.
8. Joensuu, H.; Kankaanranta, L.; Seppa¨la¨, T.; Auterinen, I.;
Kallio, M.; Kulvik, M.; Laakso, J.; Va¨ha¨talo, J.;
5.5.2. Time-dependent cellular uptake. T98G cells were
plated at 20000 per well in a Costar 96-well plate and al-
lowed to grow for 36 h. Porphyrin stocks were prepared
as above to 2· stocks. The porphyrin was then diluted in
medium to final working concentrations. The cells were
exposed to 10 lM porphyrin for 0, 1, 2, 4, 8, and 24 h.
The uptake was stopped by removing the loading med-
ium and by washing the cells with PBS. The cells were
solubilized by the addition of 100 ll of 0.25% Triton
X-100 (Calbiochem) in PBS. To determine compound
concentration, fluorescence emission was read at 410/
650 nm (excitation/emission) using a BMG FLUOstar
plate reader. The cell numbers were quantified using Cy-
Quant reagent (Molecular Probes).
´
Kortesniemi, M.; Kotiluoto, P.; Seren, T.; Karila, J.;
Brander, A.; Ja¨rviluoma, E.; Ryyna¨nen, P.; Paetau, A.;
Ruokonen, I.; Minn, H.; Tenhunen, M.; Ja¨a¨skela¨inen, J.;
Fa¨rkkila¨, M.; Savolainen, S. J. Neuro-Oncol. 2003, 62,
123–134.
9. Capala, J.; Stenstam, B. H.; Sko¨ld, K.; af Rosenscho¨ld, P.
M.; Giusti, V.; Persson, C.; Wallin, E.; Brun, A.; Franzen,
L.; Carlsson, J.; Salford, L.; Ceberg, C.; Persson, B.;
Pellettieri, L.; Henriksson, R. J. Neuro-Oncol. 2003, 62,
135–144.
10. Soloway, A. H.; Tjarks, W.; Barnum, B. A.; Rong, F.-G.;
Barth, R. F.; Codogni, I. M.; Wilson, J. G. Chem. Rev.
1998, 98, 1515–1562.
11. Bregadze, V. I.; Sivaev, I. B.; Glazun, S. A. Curr. Med.
Chem., Anti-Cancer Agents 2006, 6, 75–109.
12. Renner, M. W.; Miura, M.; Easson, M. W.; Vicente, M.
G. H. Curr. Med. Chem., Anti-Cancer Agents 2006, 6, 145–
157.
13. Ratajski, M.; Osterloh, J.; Gabel, D. Curr. Med. Chem.,
Anti-Cancer Agents 2006, 6, 159–166.
14. Fairchild, R. G.; Bond, V. P. Int. J. Rad. Oncol. Biol.
Phys. 1985, 11, 831–840.
15. Gabel, D.; Foster, S.; Fairchild, R. G. Radiat. Res. 1987,
111, 14–25.
5.5.3. Intracellular localization. HEp2 cells were plated
on LabTek II two chamber coverslips and incubated
for 36 h. The cells were then exposed to 10 lM of each
porphyrin for 24 h. For co-localization experiments
using organelle tracers, cells were incubated concur-
rently with compound the following morning for
30 min with the following organelle tracers: Mitochon-
dria: MitoTracker Green (Molecular Probes) 250 nM;
Lysosomes: LysoSensor Green (Molecular Probes)
50 nM; Endoplasmic reticulum: ERTracker Green FM
(Molecular Probes) 2 lg/ml (for 1 h); Golgi: BODIPY
FL- C5-Ceramide (Molecular Probes) 1 mM. The slides
were washed three times with growth medium and fed
medium containing 50 mM HEPES, pH 7.4. Fluores-
cence microscopy was performed using a Zeiss Labovert
200 M inverted fluorescence microscope fitted with a
standard FITC filter set (Ex/Em 470 nm/540 nm) for
organelle tracers and standard Texas Red filter set
(560 nm/600 nm LP) for compound detection (Chroma).
The images were acquired with a Zeiss Axiocam MRM
CCD camera fitted to the microscope.
16. This work was partially presented at the 12th Interna-
tional Congress on Neutron Capture Therapy for Cancer,
Takamatzu (Japan), October 2006: Vicente, M. G. H.;
Easson, M. W. In Proc. 12th ICNCT for Cancer, 2006,
pp 231–233.
17. Mettath, S.; Munson, B. R.; Pandey, R. K. Bioconj. Chem.
1999, 10, 94–102.
18. Bustamante, C.; Gurrieri, S.; Pasternack, R. F.; Purrello,
R.; Rizzarelli, E. Biopolymers 1994, 34, 1099–1104.
19. Kessel, D.; Luguya, R.; Vicente, M. G. H. Photochem.
Photobiol. 2003, 78, 431–435.
Acknowledgment
20. Sibrian-Vazquez, M.; Nesterova, I. V.; Jensen, T. J.;
Vicente, M. G. H. Bioconj. Chem., accepted for publication.
21. Gottumukkala, V.; Luguya, R.; Fronczek, F. R.; Vicente,
M. G. H. Bioorg. Med. Chem. 2005, 13, 1633–1640.
22. Bodwell, G. J.; Bridson, J. N.; Houghton, T. J.; Yarlag-
adda, B. Tetrahedron Lett. 1997, 38, 7475–7478.
23. Vicente, M. G. H.; Wickramasinghe, A.; Nurco, D. J.;
Wang, H. J. H.; Nawrocky, M. M.; Makar, M. S.; Miura,
M. Bioorg. Med. Chem. 2003, 11, 3101–3108.
The research described was supported by the National
Institutes of Health, Grant No. R01 CA098902.
Supplementary data
Dark cytotoxicity, fluorescence, and HRMS data for
porphyrins are available. Supplementary data associated

3208
M. W. Easson et al. / Bioorg. Med. Chem. 16 (2008) 3191–3208
´
24. McIlroy, S. P.; Clo, E.; Nikolajsev, L.; Frederiksen, P. K.;
Nielsen, C. B.; Mikkelsen, K. V.; Gothelf, K. V.; Ogilby,
P. R. J. Org. Chem. 2005, 70, 1134–1146.
39. Vicente, M. G. H.; Edwards, B. F.; Shetty, S. J.; Hou,
Y.; Boggan, J. E. Bioorg. Med. Chem. 2002, 10, 481–
492.
25. Lee, C.-H.; Lindsey, J. S. Tetrahedron 1994, 50, 11427–
11440.
26. Littler, B. J.; Miller, M. A.; Hung, C.-H.; Wagner, R. W.;
O’Shea, D. F.; Boyle, P. D.; Lindsey, J. S. J. Org. Chem.
1999, 64, 1391–1396.
27. Durantini, E. N. Molecules 2001, 6, 533–539.
28. Littler, B. J.; Ciringh, Y.; Lindsey, J. S. J. Org. Chem.
1999, 64, 2864–2872.
40. Miura, M.; Micca, P. L.; Fisher, C. D.; Gordon, C. R.;
Heinrichs, J. C.; Slatkin, D. N. Br. J. Radiol. 1998, 71,
773–781.
41. Ol’shevskaya, V. A.; Zaitsev, A. V.; Luzgina, V. N.;
Kondratieva, T. T.; Ivanov, O. G.; Kononova, E. G.;
Petrovskii, P. V.; Mironov, A. F.; Kalinin, V. N.;
Hofmann, J.; Shtil, A. A. Bioorg. Med. Chem. 2006, 14,
109–120.
29. Smith, K. M.; Vicente, M. G. H.. In Houben-Weyl Science
of Synthesis; Weinreb, S. M., Ed.; Georg Thieme: Stutt-
gart, 2004; Vol. 17, pp 1081–1235.
30. Muthukumaran, K.; Loewe, R. S.; Ambroise, A.; Tamaru,
S.; Li, Q.; Mathur, G.; Bocian, D. F.; Misra, V.; Lindsey,
J. S. J. Org. Chem. 2004, 69, 1444–1452.
42. Ol’shevskaya, V. A.; Nikitina, R. G.; Zaitsev, A. V.;
Luzgina, V. N.; Kononova, E. G.; Morozova, T. G.;
Drozhzhina, V. V.; Ivanov, O. G.; Kaplan, M. A.;
Kalinin, V. N.; Shtil, A. A. Org. Biomol. Chem. 2006, 4,
3815–3821.
43. Kubat, P.; Lang, K.; Cigler, P.; Kozisek, M.; Matejicek,
P.; Janda, P.; Zelinger, Z.; Prochazka, K.; Kral, V. J.
Phys. Chem. B 2007, 111, 4539–4546.
31. Arsenault, G. P.; Bullock, E.; MacDonald, S. F. J. Am.
Chem. Soc. 1960, 82, 4384–4389.
32. Senge, M. O.. In The Porphyrin Handbook; Kadish, K. M.,
Smith, K. M., Guilard, R., Eds.; Academic Press: San
Diego, 2000; Vol. 1, pp 239–347.
33. Lauceri, R.; Purrello, R.; Shetty, S. J.; Vicente, M. G. H.
J. Am. Chem. Soc. 2001, 123, 5835–5836.
34. Vicente, M. G. H.; Nurco, D. J.; Shetty, S. J.; Osterloh, J.;
Ventre, E.; Hegde, V.; Deutsch, W. A. J. Photochem.
Photobiol. B: Biol. 2002, 68, 123–132.
35. De Napoli, M.; Nardis, S.; Paolesse, R.; Vicente, M. G.
H.; Lauceri, R.; Purrello, R. J. Am. Chem. Soc. 2004, 126,
5934–5935.
36. Hao, E.; Sibrian-Vazquez, M.; Serem, W.; Garno, J. C.;
Fronczek, F. R.; Vicente, M. G. H. Chem. Eur. J. 2007,
13, 9035–9042.
37. Woodburn, K.; Phadke, A. S.; Morgan, A. R. Bioorg.
Med. Chem. Lett. 1993, 3, 2017–2022.
38. Hill, J. S.; Kahl, S. B.; Stylli, S. S.; Nakamura, Y.; Koo,
M.-S.; Kaye, A. H. Proc. Natl. Acad. Sci. U.S.A. 1995, 92,
12126–12130.
44. Matsumura, A.; Shibata, Y.; Yamamoto, T.; Yoshida, F.;
Isobe, T.; Nakai, K.; Hayakawa, Y.; Kiriya, M.; Shimojo,
N.; Ono, K.; Sakata, I.; Nakajima, S.; Okumura, M.;
Nose, T. Cancer Lett. 1999, 141, 203–209.
45. Nguyen, T.; Brownell, G. L.; Holden, S. A.; Teicher, B. A.
Biochem. Pharma. 1993, 45, 147–155.
46. Nguyen, T.; Brownell, G. L.; Holden, S. A.; Kahl, S.;
Miura, M.; Teicher, B. A. Radiat. Res. 1993, 133,
33–40.
47. Callahan, D. E.; Forte, T. M.; Afzal, S. M. J.; Deen,
D. F.; Kahl, S. B.; Bjornstad, K. A.; Bauer, W. F.;
Blakely, E. A. Int. J. Rad. Oncol. Biol. Phys. 1999, 45,
761–771.
48. Hao, E.; Jensen, T. J.; Courtney, B. H.; Vicente, M. G. H.
Bioconj. Chem. 2005, 16, 1495–1502.
49. Osterloh, J.; Vicente, M. G. H. J. Porphyr. Phthalocya.
2002, 6, 305–324.
50. Novick, S.; Laster, B.; Quastel, M. R. Int. J. Biochem. Cell
Biol. 2006, 38, 1374–1381.