Synthesis and biological analysis of prostate-specific membrane antigen-targeted anticancer prodrugs

Article abstract of DOI:10.1021/jm100729b

Ligand-targeted therapeutics have increased in prominence because of their potential for improved potency and reduced toxicity. However, with the advent of personalized medicine, a need for greater versatility in ligand-targeted drug design has emerged, w

Full text of DOI:10.1021/jm100729b

J. Med. Chem. 2010, 53, 7767–7777 7767
DOI: 10.1021/jm100729b
Synthesis and Biological Analysis of Prostate-Specific Membrane
Antigen-Targeted Anticancer Prodrugs
†
Sumith A. Kularatne, Chelvam Venkatesh, Hari-Krishna R. Santhapuram, Kevin Wang,
†
‡
‡
†
†
Balasubramanian Vaitilingam, Walter A. Henne, and Philip S. Low*
,†
†
Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, Indiana 47907, United States, and
Endocyte, Inc., 3000 Kent Avenue, West Lafayette, Indiana 47906, United States
‡
Received June 16, 2010
Ligand-targeted therapeutics have increased in prominence because of their potential for improved
potency and reduced toxicity. However, with the advent of personalized medicine, a need for greater
versatility in ligand-targeted drug design has emerged, where each tumor-targeting ligand should be
capable of delivering a variety of therapeutic agents to the same tumor, each therapeutic agent being
selected for its activity on a specific patient’s cancer. In this report, we describe the use of a prostate-specific
membrane antigen (PSMA)-targeting ligand to deliver multiple unrelated cytotoxic drugs to human
prostate cancer (LNCaP) cells. We demonstrate that the PSMA-specific ligand, 2-[3-(1, 3-dicarboxy
propyl)ureido] pentanedioic acid, is capable of mediating the targeted killing of LNCaP cells with many
different therapeutic warheads. These results suggest that flexibility can be designed into ligand-targeted
therapeutics, enabling adaptation of a single targeting ligand for the treatment of patients with different
sensitivities to different chemotherapies.
Introduction
formed by chromosomal translocation only in chronic mye-
a
7
logenous leukemia (CML) cells. Another example of a
functionality-targeted therapeutic agent is bevacizumab, which
suppresses growth of new blood vessels required for tumor
proliferation, but has little impact on the maintenance or
survival of the vasculature healthy tissues.
A second major class of targeted therapies involves the use
Most cancer therapies today involve treatment with cytotoxic
drugs that distribute indiscriminately into virtually all cells of
the body and cause damage to malignant and healthy cells
alike. Because such conventional chemotherapies are primarily
designed to kill rapidly dividing cells, they also destroy pro-
liferating healthy cells, leading to off-target toxicities that can
include myelosuppression, mucositis, alopecia, nausea/vomiting,
8-11
of tumor-specific ligands to deliver an attached nonselective
12-15
1-6
cancer drug specifically to malignant cells.
When such
anemia, peripheral neuropathy, fatigue, etc.
Clearly, cyto-
a homing ligand exhibits strong specificity for a cancer cell,
the derived ligand-targeted therapies can yield high tumor
response rates with low toxicity to normal tissues. Moreover,
if the chemotherapeutic agent is released only after uptake
and trafficking deep into the target cell, efflux of the drug by
toxic therapies that can be targeted selectively to pathologic cells,
avoiding collateral damage to healthy cells, would constitute a
significant advance in the treatment of cancer.
Two major classes of targeted therapies are currently under
clinical development. First, chemotherapeutic agents are being
designed to target novel functionalities or processes that emerge
primarily during pathogenesis and are critical for proliferation
and/or survival ofthe pathologic cellbutare not importantfor
the function or survival of normal cells. A prominent example
of this class of targeted therapeutics is imatinib mesylate,
a tyrosine kinase inhibitor that is designed to block bcr-abl
16
multidrug resistance pumps can be greatly reduced. Ligands
thathave been exploited for chemotherapeutic agent targeting
17-20
include both monoclonal antibodies
and low molecular
21,22
weight receptor-binding ligands such as peptide hormones,
23
24
receptor antagonists and agonists, aptamers, oligosac-
25,26
charides,
27
oligopeptides, and vitamins.
15,28
In an effort to expand the arsenal of tumor-targeting ligands,
we have synthesized a molecule termed 2-[3-(1,3-dicarboxy
(breakpoint cluster region-abelson), a proto-oncogene product
2
9
propyl)ureido] pentanedioic acid (DUPA) that selectively
binds to a cell surface glycoprotein termed prostate-specific
membrane antigen (PSMA) and enters PSMA-expressing
*To whom correspondence should be addressed. Tel: 765-494-5273.
Fax: 765-494-5272. E-mail: plow@purdue.edu.
a
Abbreviations: PSMA, prostate-specific membrane antigen; LNCaP,
29-32
lymph node prostate cancer; PCa, prostate cancer; CML, chronic myelog-
enous leukemia; DUPA, 2-[3-(1,3-dicarboxy propyl)ureido] pentanedioic
acid; PMPA, 2-(phosphonomethyl)pentanedioic acid; GSH, glutathione;
DTT, dithiothreitol; HOBt, N-hydroxybenzotriazole; HBTU, O-benzotri-
cells by PSMA endocytosis.
Because PSMA is expressed
on both prostate cancer (PCa) cells and the neovasculature
ofother solid tumorsbutis largelyabsent from healthy tissues,
it constitutes an ideal cell surface protein for tumor-specific
0
0
azole-N,N,N ,N -tetramethyl-uronium-hexafluoro-phosphate; SPPS, solid-
phase peptide synthesis; RP-HPLC, reverse-phase high-performance liquid
chromatography; UPLC, ultra high-pressure liquid chromatography; DidB,
didemnin B; Cpt, camptothecin; VrcA, verrucarin A; PTax, paclitaxel;
DAVB, desacetylvinblastine; DAVBH, desacetylvinblastine hydrazide;
Tub, tubulysin B; TubH, tubulysin B hydrazide; Dox, doxorubicin.
3
0,33
targeting.
DUPA conjugates for prostate cancers, we have examined
In a recent effort to evaluate the specificity of
99m
the uptake of a DUPA-targeted Tc radioimaging agent
by human lymph node prostate cancer (LNCaP) tumors in
r
2
010 American Chemical Society
Published on Web 10/11/2010
pubs.acs.org/jmc

7
768 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Kularatne et al.
athymic nude mice and found retention of the agent limited to
Synthesis of 2-(Pyridine-2-yl-disulfanyl) Ethanol (2). To a solu-
tion of methoxycarbonylsulfenyl chloride (3.1 mL, 34.25 mmol) in
2
the PCa xenograft and the kidneys. Because human kidneys,
9
3
unlike murine kidneys, express only low levels of PSMA,
4
30
CH
in CH
3
CN (30.0 mL) at 0 °C, 2-mercaptoethanol (2.4 mL, 34.25 mmol)
CN (1.0 mL) was added dropwise, and the reaction mixture
3
we hypothesized that DUPA might constitute an ideal ligand
for a PCa-targeted therapy. In this paper, we describe the
synthesis, structure validation, and biological properties of a
series of PSMA-targeted disulfide-bridged self-immolative
chemotherapeutic agents and identify several targeted pro-
drugs that warrant consideration for clinical development.
was allowed to stir at -10 °C for 30 min. A solution of 2-thiopyr-
idine (3.46 g, 31.13 mmol) in CH CN (18.0 mL) was added to the
reaction mixture and refluxed for 2 h. The reaction mixture was
3
then stirred at 10 °C for 1 h and filtered. The solid was washed with
CH CN and dried under vacuum at room temperature to yield the
3
1
pure product as a colorless solid (4.0 g, 57.5%). H NMR (DMSO-
d
6
): δ 3.06 (t, J=6.0 Hz, 2H), 4.18 (t, J=6.1 Hz, 2H), 7.25 (t, J=
Experimental Section
6.0 Hz, 1H), 7.78 (d, J=7.9 Hz, 1H), 7.83 (t, J=7.5, 1H), 8.45 (d,
þ
J=4.0 Hz, 1H). ESI/MS (m/z): (M þ H) calcd for C H NOS ,
Materials. Amino acids were purchased from Chem-Impex
Int (Chicago, IL). N-Hydroxybenzotriazole (HOBt) and O-benzo-
7
10
2
188.3; found, 188.0.
0
0
Synthesis of 2-[Benzotriazole-1-yl-(oxycarbonyloxy)ethyl-
disulfanyl]pyridine (3). To a solution of 2-(pyridine-2-yl-disulfanyl)
ethanol (2.0, 8.94 mmol), HOBT (1.58 g, 11.71 mmol), and TEA
(1.25 mL, 8.94 mmol) in CH
triazole-N,N,N ,N -tetramethyl-uronium-hexafluoro-phosphate
HBTU) were obtained from Peptide Int. (Louisville, KY).
-(Phosphonomethyl)pentanedioic acid (PMPA) was purchased
(
2
3
CN (35 mL) at 0 °C, diphosgene
from Axxora Platform (San Diego, CA). Tubulysin B (Tub) and
desacetylvinblastine (DAVB) were provided by Endocyte Inc.
(0.72 mL, 5.98 mmol) was added. The reaction mixture was heated
to 50 °C and stirred for 24 h. The solid was filtered, washed with
(
West Lafayette, IN). Camptothecin (Cpt) was obtained from
the NCI (Rockville, MD). Verrucarin A (VrcA) and didemnin B
DidB) were purchased from Sigma-Aldrich (St. Louis, MO).
CH CN, and dried under vacuum at room temperature to yield the
3
1
pure product as a pale white solid (3.48 g, 95.7%). H NMR
(
(
DMSO-d
6
):δ3.39 (t, J=6.0 Hz, 2H), 4.76 (t, J=6.0 Hz, 2H), 7.22
All other chemicals, cell culture materials, and animal feed were
obtained from major suppliers.
(
m, 1H), 7.66 (t, J=7.8 Hz, 1H), 7.78 (m, 2H), 7.92 (t, J=7.8, 1H),
8.05 (d, J=8.4 Hz, 1H), 8.19 (d, J=8.4 Hz, 1H), 8.43 (d, J=4.7
General Methods. Moisture- and oxygen-sensitive reactions
were carried out under an argon atmosphere. Solid-phase pep-
tide synthesis (SPPS) was performed using a standard peptide
synthesis apparatus (Chemglass, Vineland, NJ). Flash chroma-
tography was conducted with silica gel as the solid phase, and
thin-layer chromatography (TLC) was performed on silica gel
TLC plates (60 f254, 5 cm ꢀ 10 cm) and visualized under UV
light. All peptides and peptide conjugates were purified by pre-
parative reverse-phase high-performance liquid chromatogra-
phy (RP-HPLC; Waters, xTerra C18 10 μm; 19 mmꢀ250 mm)
and analyzed by analytical RP-HPLC (Waters, X-Bridge C
þ
Hz, 1H). ESI/MS (m/z): (M þ H) calcd for C14
13 4 3 2
H N O S , 349.4;
found, 349.2.
Synthesis of Hydrazinecarboxylic Acid 2-(Pyridine-2-yl-disulfanyl)-
ethyl ester (4). To a solution of 2-[benzotriazole-1-yl-(oxycarbon-
yloxy)ethyldisulfanyl]pyridine hydrochloride (685 mg, 1.62
mmol) and DIPEA (600 μL, 3.4 mmol) in CH Cl (5.0 mL) at
2 2
0
stir at 0 °C for 15 and 30 min at room temperature. After the
precipitate was filtered, the crude compound was purified using
flash column chromatography (silica gel, 2% MeOH in CH Cl )
°C, hydrazine was added. The reaction mixture was allowed to
2
2
1
8
1
to yield the product as colorless oil (371 mg, 93.4%). H NMR
DMSO-d ): 3.06 (t, J=6.0 Hz, 2H, S-CH ), 4.18 (t, J=6.0 Hz,
2H), 7.26 (t, J=7.8 Hz, 1H), 7.79 (d, J=8.1 Hz, 1H), 7.83 (t, J=
5
μm; 3.0 mm ꢀ 50 mm) or ultra high-pressure liquid chroma-
(
6
2
tography (UPLC) (Acquity, BEH C18 1.7 μm; 2.1 mmꢀ50 mm).
1
H spectra were acquired with a Bruker 400 or 500 MHz NMR
spectrometer equipped with a TXI cryoprobe. Samples were run
þ
7
.8, 1H), 8.47 (d, J=4.7 Hz, 1H). LC/MS (m/z): (M þ H) calcd
8 12 3 2 2
for C H N O S , 246.0; found, 246.1.
3 6 6 2
in 5 mm NMR tubes using CDCl , DMSO-d , or DMSO-d /D O.
Synthesis of Disulfide-Activated TubH (6). To a solution of
Tub (30 mg, 0.036 mmol) in ethylacetate (600 μL) under argon
at -15 °C, isobutyl chloroformate (4.7 μL, 0.054 mmol) and
diisopropylethylamine (13.2 μL, 0.076 mmol) were added. After
the solution was stirred at -15 °C for another 45 min under
argon, a solution of compound 4 (13.4 mg, 0.054 mmol) in ethyl
acetate (500 μL) was added. The reaction mixture was stirred
at -15 °C for 15 min and then at room temperature for 45 min.
The crude product was purified using a short column (2-8%
Presaturation was used to reduce the intensity of the residual
H O peak. All H signals are recorded in ppm with reference to
residual CHCl (7.27 ppm) or DMSO (2.50 ppm), and data are
3
1
2
reported as s=singlet, d=doublet, t=triplet, q=quartet, m=
multiplet or unresolved, and b=broad, with coupling constants
in Hz. Liquid chromatography/mass spectrometry (LC/MS) ana-
lyses were obtained using a Waters micromass ZQ 4000 mass
spectrometer coupled with a UV diode array detector. High-
resolution mass spectrometric results were obtained by matrix-
assisted laser desorption ionization (MALDI) mass using an
Applied Biosystems (Framingham, MA) Voyager DE PRO mass
spectrometer. This instrument utilizes a nitrogen laser (337 nm
UV laser) for ionization with a time-of-flight mass analyzer. The
matrix used for these samples was R-cyano-4-hydroxy cinnamic
acid, and the peptide LHRH was used as an internal standard.
Cell Culture. LNCaP cells were obtained from the American
Type Culture Collection (Rockville, MD) and grown as a mono-
layer in 1640 RPMI medium containing 10% heat-inactivated
fetal bovine serum, 1% sodium pyruvate (100 mM), and 1% peni-
cillin streptomycin in a 5% carbon dioxide:95% air-humidified
atmosphere at 37 °C.
methanol in dichloromethane) to obtain the activated TubH (6)
1
as a colorless oil (34.4 mg, 90.5%). H NMR (DMSO-d
6
/D
2
O):
δ 0.65 (d, J=6.1 Hz, 3H), 0.78 (m, 6H), 0.94 (d, J=6.1 Hz, 3H),
1.02 (d, J=7.0 Hz, 3H), 1.08 (m, 1H), 1.16 (m, 1H), 1.23 (s, 4H),
1.24 (s, 4H), 1.32 (q, J=7.2 Hz, 2H), 1.43 (m, 2H), 1.49 (m, 4H),
1.59 (m, 1H), 1.67 (m, 1H), 1.79 (m, 1H), 1.82 (s, 1H), 1.89 (m,
1H), 2.09 (s, 6H), 2.11 (m, 1H), 2.21 (m, 2H), 2.35 (m, 1H), 2.69
(m, 1H), 2.80 (bd, 1H), 2.97 (bs, 1H), 3.08 (m, 3H), 4.11 (m, 2H),
4.20 (m, 2H), 4.40 (d, J=9.3 Hz, 1H), 5.25 (d, J=11.9 Hz, 1H),
5.71 (d, J=11.5 Hz, 1H), 6.16 (b, 1H), 6.59 (d, J=8.1 Hz, 2H),
6.97 (d, J=6.6 Hz, 2H), 7.22 (m, 1H), 7.65-7.89 (m, 2H), 8.17
þ
(s, 1H), 8.43 (bs, 1H). ESI/MS (m/z): (M þ H) calcd for C50
73
H -
Synthesis and Characterization. DUPA, the ligand used to
N
8
O
11
S
3
, 1057.4; found, 1057.0.
Characterization of DAVB Hydrazide (DAVBH, 8). H NMR
(DMSO-d /D O): δ 0.61 (m, 1H), 0.74 (t, J=7.3 Hz, 3H), 0.79
3
5
1
target PSMA, was synthesized as previously described. Peptide
spacers used to attach DUPA to its therapeutic warhead were
synthesized by standard fluorenylmethyloxycarbonyl (Fmoc) solid-
phase peptide chemistry starting with Cys(4-methoxytrityl)-Wang
resin (Scheme 2). Disulfide-activated prodrugs [tubulysin B hydra-
6
2
(t, J=7.5 Hz, 3H), 1.03 (d, J=6.1 Hz, 1H), 1.16 (m, 2H), 1.29
(m, 2H), 1.58 (m, 2H), 1.94 (m, 1H), 2.30 (s, 3H), 2.35 (m, 1H),
2.65 (m, 1H), 2.69 (s, 3H), 2.89 (dd, J=14.2, 5.2 Hz, 1H), 3.10
(m, 1H), 3.26 (d, J=13.8 Hz, 1H), 3.54 (s, 3H), 3.72 (s, 3H), 3.82
(d, J=5.9 Hz, 1H), 3.95 (d, J =6.1 Hz, 1H), 3.97 (s, 1H), 4.05
(t, J=14.4 Hz, 1H), 4.24 (sb, 1H), 5.57 (d, J=10.1 Hz, 1H), 5.71
36
37
38
zide (TubH), DAVB hydrazide, and Cpt ] were synthesized
from the commercially available drugs according to literature
procedures.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7769
(
7
dd, J=11.0, 5.1 Hz, 1H), 6.20 (s, 1H), 6.44 (s, 1H), 6.92 (t, J=
.5 Hz, 1H), 7.00 (t, J=7.5 Hz, 1H), 7.15 (m, 1H), 7.25 (t, J=8.4
Hz, 1H), 7.36 (d, J=7.9 Hz, 1H), 8.33 (s, 1H), 8.82 (s, 1H), 9.32
Characterization of DUPA-Linker II (13b). The synthesis of
linker II was performed similarly to DUPA-linker I using SPPS.
Analytical RP-HPLC: R
and B = CH CN; solvent gradient: 10-80% B in 12 min]. H
NMR (DMSO-d ): δ 1.01 (m, 2H), 1.14 (m, 2H), 1.30 (m, 3H),
t 4
=2.2 min [A=10 mM NH OAc (pH 7.0),
þ
1
(
found, 769.5.
s, 1H). LC/MS (m/z): (M þ H) calcd for C43
H
57
N
6
O
7
, 769.9;
3
6
1
Characterization of Disulfide-Activated DAVBH (9). H NMR
1.70 (m, 3H), 1.89 (m, 2H), 1.95 (m, 2H), 2.08 (m, 2H), 2.23 (m,
2H), 2.65 (m, 2H), 2.79 - 3.00 (m, 9H), 3.06 (dd, J = 14.1, 4.8
Hz, 1H), 4.04 (m, 1H, R-H), 4.08 (m, 1H, R-H), 4.43 (m, 1H,
R-H), 4.47 (m, 1H, R-H), 4.58 (m, 1H, R-H), 7.20 (m, 10H). LC/
(
DMSO-d
J=7.5 Hz, 3H), 1.03 (d, J=6.1 Hz, 1H), 1.12 (s, 2H), 1.19 (m, 2H),
.33 (m, 2H), 1.60 (m, 2), 1.78 (m, 1H), 2.38 (m, 2H), 2. 56 (bs, 1H),
.67 (m, 1H), 2.78 (s, 3H), 2.90 (dd, J=14.2, 5.2 Hz, 1H), 3.09 (s,
H), 3.12 (m, 2H), 3.16 (bs, 1H), 3.21 (dd, J= 15.5, 4.1 Hz, 1H),
.28 (d, J=14.2 Hz, 1H), 3.42 (s, 1H), 3.56 (s, 3H), 3.74 (s, 3H), 3.82
d, J=6.5 Hz, 1H), 3.99(s, 1H), 4.09(m, 2H), 4.23(m, 2H), 5.61 (d,
J=10.1 Hz, 1H), 5.71 (dd, J=10.7, 5.8 Hz, 1H), 6.22 (s, 1H), 6.48
s, 1H), 6.94 (t, J=7.5 Hz, 1H), 7.02 (t, J=7.5 Hz, 1H), 7.28 (d, J=
.9 Hz, 2H), 7.39 (d, J=8.0 Hz, 1H), 7.86 (m, 2H), 8.49 (bs, 1H),
.66 (s, 1H), 9.20 (s, 1H), 9.32 (s, 1H), 9.45 (s, 1H). LRMS-LC/MS
6 2
/D O): δ 0.64 (m, 1H), 0.76 (t, J=7.3 Hz, 3H), 0.81 (t,
1
2
3
3
þ
MS (m/z): (M þ H) calcd for C H N O S, 859.95; found,
4
0
55
6
13
859.67. UV/vis: λmax = 257 nm.
Characterization of DUPA-Linker III (13c). The synthesis of
linker III was performed as described for DUPA-linker I using
(
SPPS. Analytical RP-HPLC: white solid, R
TFA (pH 2), and B=CH CN; solvent gradient: 1-50% B in 10
/D O): δ 1.70 (m, 3H), 1.90 (m, 3H),
t
=8.2 min [A=0.1
(
7
8
3
1
min]. H NMR (DMSO-d
6
2
2.10 (m, 2H), 2.17 (m, 2H), 2.23 (m, 2H), 2.36 (q, J = 8.3 Hz,
1H), 2.59 (dd, J=11.0, 5.7 Hz, 1H), 2.79 (m, 3H), 3.04 (dd, J=
9.8, 4.0 Hz, 1H), 4.07 (m, 2H), 4.13 (m, 1H), 4.37 (m, 1H), 4.47
(m, 2H), 7.19 (m, 5H), 7.87 (d, J=8.1 Hz, 1H), 8.20 (d, J=7.8
þ
(
m/z): (M) calcd for C51
Characterization of Disulfide-Activated Cpt (10). H NMR
DMSO-d ): δ 0.92 (t, J=7.4 Hz, 3H), 2.18 (m, 2H), 3.14 (t, J=
.1 Hz, 2H), 4.32 (t, J=6.0 Hz, 2H), 5.30 (s, 2H), 5.52 (d, J=3.5
Hz, 2H), 7.09 (s, 1H), 7.15 (m, 1H), 7.70 (m, 2H), 7.77 (td, J =
.75, 1.8 Hz, 1H), 7.85 (td, J=8.0, 1.0 Hz, 1H), 8.13 (t, J=8.1
64 7 9 2
H N O S , 982.2; found, 982.5.
1
(
6
6
þ
Hz, 1H). LC/MS (m/z): (M þ H) calcd for C H N O S,
3
2
43
6
17
815.7; found, 815.3. UV/vis: λmax = 257 nm.
7
Synthesis of DUPA-TubH I (14). Into a solution of saturated
sodium bicarbonate (2 mL) and HPLC grade water, argon was
bubbled for 10 min. With continuous bubbling of argon, DUPA-
linker I (9.2 mg, 0.0113 mmol) was dissolved in argon-purged
HPLC grade water (2.0 mL), and the pH of the reaction mixture
was increased to ca. 7 using argon-purged bicarbonate. A solu-
tion of disulfide activated-TubH 6 (12.0 mg, 0.0113 mmol) in
THF (2.0 mL) was then added to the reaction mixture, and the
solution was stirred for 10 min. After THF was removed under
reduced pressure, DUPA-TubH I (DUPA-TubH I) was purified
by preparative RP-HPLC [A=2 mM sodium phosphate buffer
Hz, 2H), 8.39 (dt, J = 4.8, 0.7 Hz, 1H), 8.69 (s, 1H), ESI/MS
(
Characterization of Disulfide-Activated VrcA (11). Synthesis
was performed similarly to the DUPA-didemin B conjugate (see
þ
m/z): (M þ H) calcd for C H N O S , 562.6; found, 562.3.
2
8
24
3
6 2
1
above). H NMR (DMSO-d ): δ 0.73 (s, 3H), 0.88 (d, J=6.8 Hz,
6
3
1
1
1
4
5
6
8
H), 1.23 (s, 1H), 1.57 (t, J=12.7 Hz, 1H), 1.62 (s, 3H), 1.72 (s,
H), 1.92 (m, 2H), 2.78 (d, J=3.9 Hz, 1H), 3.03 (d, J=3.9 Hz,
H), 3.16 (t, J=6.1 Hz, 1H), 3.70 (d, J=5.1 Hz, 1H), 3.94 (t, J=
1.7 Hz, 1H), 4.20 (d, J=12.1 Hz, 1H), 4.35 (t, J=6.0 Hz, 1H),
.39 (d, J=12.1 Hz, 1H), 4.84 (s, 1H), 5.31 (d, J=4.7 Hz, 1H),
.82 (m, 1H), 6.24 (d, J=15.5 Hz, 1H), 6.32 (d, J=11.1 Hz, 1H),
.88 (t, J=11.4 Hz, 1H), 7.25 (t, J=6.0 Hz, 1H), 7.82 (m, 2H),
(pH 7.1), and B=CH CN; solvent gradient: 5-80% B in 25 min],
3
yielding the desired product (12.2 mg, yield = 61.3%). Analyt-
ical RP-HPLC: R = 1.8 min [A = 10 mM NH OAc (pH 7.0),
þ
.07 (s, 1H), 8.46 (d, J=4.2 Hz, 1H). LC/MS (m/z): (M þ H)
t
4
1
calcd for C35
H
42NO11
S
2
, 716.8; found, 716.5.
and B = CH CN; solvent gradient: 10-100% B in 4 min]. H
3
Synthesis of Disulfide-Activated DidB (12). To a solution of
DidB (15.0 mg, 0.013 mmol) in CH Cl (0.33 mL) at 0 °C,
NMR (DMSO-d
6 2
/D O): δ 0.67 (d, J=5.5 Hz, 3H), 0.81 (q, J=
2
2
7.8 Hz, 2H), 0.86 (d, J=7.2 Hz, 3H), 0.96 (d, J=6.2 Hz, 3H),
2
(
-[benzotriazole-1-yl-(oxycarbonyloxy)ethyldisulfanyl]pyridine
18.0 mg, 0.047 mmol) and DMAP (8.0 mg, 0.067 mmol) were
1
2
.03 (m, 3H), 1.10 (m, 1H), 1.27 (bs, 6H), 1.47 (q, J = 7.3 Hz,
H), 1.51 (m, 1H), 1.61 (m, 2H), 1.68 (m, 1H), 1.82 (m, 2H), 1.92
added. The reaction mixture was stirred for 2 h at room tem-
perature, and the crude product was purified by column chro-
matography (0-4% methanol in dichloromethane) to afford
(
m, 1H), 2.11 (s, 6H), 2.23 (m, 2H), 2.37 (m, 2H), 2.42 (t, J=8.1
Hz, 1H), 2.71 (m, 1H), 3.00 (bs, 1H), 3.10 (m, 3H), 3.58 (bt, 2H),
.13 (m, 2H), 4.22 (m, 2H, R-H), 4.26 (t, J=7.0 Hz, 1H), 4.42 (t,
4
1
the activated DidB, as a white solid (10.0 mg, 56%). H NMR
J=8.8 Hz, 1H), 5.28 (d, J=12.1 Hz, 1H), 5.73 (d, J=11.4 Hz,
1H), 6.18 (bs, 1H), 6.60 (d, J=7.9 Hz, 2H), 6.97 (d, J=7.9 Hz,
2H), 7.24 (m, 2H), 7.68 (m, 1H), 7.70 (m, 1H), 7.80 (m, 2H), 7.85
(
CDCl
J=6.9 Hz, 3H), 1.38 (m, 4H), 1.51 (d, J=6.7 Hz, 3H), 1.56-1.79
m, 7H), 1.98-2.17 (m, 6H), 2.26 (m, 1H), 2.34 (m, 1H), 2.54 (s,
H), 2.63 (m, 1H), 3.02 (t, J = 6.8 Hz, 2H), 3.14 (s, 3H), 3.17-
.21 (m, 1H), 3.24 (s, 1H), 3.41(dd, J = 13.8, 4.2 Hz, 1H), 3.59
3
): δ 0.84-0.96 (m, 23H), 1.15-1.28 (m, 6H), 1.32 (d,
(
3
3
(m, 2H), 8.21 (s, 1H), 8.46 (s, 1H). LRMS (LC/MS) (m/z): 2007.0
(M þ H) . UV/vis: λ =254 nm. HRMS (MALDI) (m/z) calcd
max
þ
þ
for C H N O S , 2005.8637 (M þ H) ; found, 2005.8624.
9
2
133 16 28 3
(
(
d, J=6.8 Hz, 2H), 3.70 (m, 2H), 3.79 (s, 3H), 4.06 (m, 2H), 4.22
q, J=6.9 Hz, 1H), 4.37 (m, 2H), 4.55 (dd, J=5.8, 2.0 Hz, 1H),
Characterization of DUPA-DAVBH (15). The synthesis of
DUPA-DAVB hydrazide (DUPA-DAVBH) was performed as
4
Hz, 1H), 5.08 (q, J = 6.7 Hz, 1H), 5.18 (d, J = 3.4 Hz, 1H),
.60 (t, J=7.6 Hz, 1H), 4.73 (t, J=7.3 Hz, 1H), 4.79 (t, J=9.7
described for DUPA-TubH. Analytical RP-HPLC: R =6.1 min
t
[A=10 mM NH OAc (pH 7.0), and B = CH CN; solvent
4
3
1
5
8
1
.21-5.23 (m, 1H), 5.36 (dd, J=11.2, 4.2 Hz, 1H), 6.84 (d, J=
.6 Hz, 2H), 7.07 (d, J=8.6 Hz, 1H), 7.11 (dt, J=5.7, 2.1 Hz,
gradient: 10-100% B in 14 min]. H NMR (DMSO-d /D O):
6
2
δ 0.62 (m, 1H), 0.73 (t, J=7.3 Hz, 3H), 0.79 (t, J=7.3 Hz, 3H),
0.86 (d, J = 7.2 Hz, 3H), 1.04 (m, 2H), 1.16 (m, 6H), 1.31 (m,
3H), 1.57-2.16 (ms, 6H), 2.25 (m, 2H), 2.36 (m, 2H), 2.50-2.70
(ms, 4H), 2.77 (bs, 3H), 2.80-3.30 (ms, 10H), 3.53 (s, 3H), 3.71
(m, 3H), 3.98 (m, 1H), 4.10 (m, 1H), 4.18 (m, 2H), 4.42 (m, 2H),
5.60 (d, J=10.1 Hz, 1H), 5.68 (d, J=11.0 Hz, 1H), 6.20 (s, 1H),
6.22 (s, 1H), 6.43 (s, 1H), 6.91 (t, J=7.7 Hz, 2H), 7.00 (t, J=7.3
Hz, 1H), 7.03 (d, J=7.5 Hz, 1H), 7.05-7.29 (ms, 6H), 7.36 (d,
J = 7.8 Hz, 1H), 7.72 (bs, 2H), 7.86 (bs, 2H), 7.93 (s, 1H), 8.02
(s, 1H), 9.16 (s, 1H), 9.34 (s, 1H). LRMS (LC/MS) (m/z): 1932.2
H), 7.66-7.69 (m, 2H), 8.47 (d, J=5.0 Hz, 1H). LC/MS (m/z):
þ
(
M) calcd for C65
H
96
N
8
O
17
S
Characterization of DUPA-Linker I (13a). The synthesis was
2
, 1325.6; found, 1325.8.
3
3
t
performed as described elsewhere. Analytical RP-HPLC: R =
7
.8 min [A=0.1 TFA (pH 2), and B=CH CN; solvent gradient:
3
1
5
-80% B in 10 min]. H NMR (DMSO-d
6
): δ 0.93 (m, 2H), 1.08
(
m, 5H), 1.27 (m, 5H), 1.69 (m, 2H), 1.90 (m, 2H), 1.94 (m, 2H),
.10 (m, 2H), 2.24 (m, 2H), 2.62 (m, 2H), 2.78 (m, 4H), 2.88
m, 1H), 2.96 (t, J=6.8 Hz, 2H), 3.01 (m, 1H), 3.31 (dd, J=13.9,
2
(
þ
5
1
.9 Hz, 1H), 3.62 (dd, J=14.0, 5.9 Hz, 1H), 3.80 (q, J=6.1 Hz,
H), 4.07 (m, 1H, R-H), 4.37 (m, 1H, R-H), 4.42 (m, 2H,
(M þ H) . UV/vis: λmax=254 nm. HRMS (MALDI) (m/z): calcd
þ
for C93
H
124
N
Synthesis and Characterization DUPA-Cpt (16). The synthesis
15
O
26
S
2
, 1930.8283 (M þ H) ; found, 1930.8273.
þ
R-H), 4.66 (m, 1H, R-H), 7.18 (m, 10H). LC/MS (m/z): (M þ H)
calcd for C47
H
66
N
9
O
17S, 1060.4; found, 1060.2. UV/vis:
of DUPA-Cpt was performed as described for DUPA-TubH.
λmax =257 nm.
t 4
Analytical RP-HPLC: R =5.3 min [A=10 mM NH OAc (pH 7.0),

7
770 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Kularatne et al.
1
and B=CH
3
CN; solvent gradient: 10-100% B in 14 min]. H
1.93 (m, 2H), 1.98 (s, 3H), 2.11 (m, 2H), 2.18 (m, 2H), 2.20 (m,
2H), 2.34 (m, 3H), 2.62 (m, 1H), 2.75 (m, 3H), 2.89 (m, 2H), 3.06
(d, J=10.9 Hz, 1H), 3.12 (d, J=11.1 Hz, 1H), 3.76 (t, J=6.2
Hz, 1H), 3.81 (t, J=6.2 Hz, 2H), 3.90 (t, J=6.4 Hz, 1H), 4.04 (b,
1H), 4.21 (b, 2H), 4.34 (m, 1H), 4.37 (m, 1H), 4.49 (m, 2H), 5.20
(d, J=11.5 Hz, 1H), 5.65 (d, J=11.4 Hz, 1H), 6.15 (b, 1H), 6.57
(d, J=8.0 Hz, 2H), 6.96 (d, J=7.4 Hz, 2H), 7.13 (d, J=6.7 Hz,
NMR (DMSO-d /D O): δ 0.89 (t, J=7.5 Hz, 3H), 1.00 (m, 1H),
.11 (m, 2H), 1.28 (m, 1H), 1.67 (m, 1H), 1.78 (m, 1H), 1.85 (m,
H), 1.98 (m, 1H), 2.07 (m, 1H), 2.14 (m, 1H), 2.23 (m, 1H), 2.84
6
2
1
1
(
1
1
1
bt, J=11.9 Hz, 1H), 2.94 (m, 1H), 3.31-2.99 (ms, 3H), 3.96 (m,
H), 4.06 (q, J=6.5 Hz, 1H), 4.17 (m, 1H), 4.27 (t, J=5.8 Hz,
H), 4.34 (t, J=6.1 Hz, 1H), 4.42 (m, 1H), 4.65 (m, 1H), 5.30 (bs,
H), 5.50 (bs, 1H), 7.10 (m, 10H), 7.71 (t, J=7.5 Hz, 1H), 7.81 (t,
3H), 7.18 (d, J=6.5 Hz, 2H), 8.04 (s, 1H). LC/MS (m/z): 1758.7
(M - H) . UV/vis: λ =254 nm. HRMS (MALDI) (m/z): calcd
-
J=6.6 Hz, 1H), 7.86 (t, J=7.7 Hz, 1H), 8.12 (d, J=8.1 Hz, 1H),
max
-
8
7
.19 (d, J=8.5 Hz, 1H), 8.59 (bs, 1H), 8.69 (s, 1H), 8.98 (d, J=
.5 Hz, 1H), 9.47 (bt, 1H). LRMS (LC/MS) (m/z): 1511.1 (M þ
for C H N O S , 1758.6594 (M - H) ; found, 1758.7033.
7
7
110 13 28 3
General Procedure for Relative Affinity Studies. LNCaP cells
were seeded in 24-well (100000 cells/well) Falcon plates and
allowed to form monolayers over a period of 48 h. The spent
þ
H) . UV/vis: λmax=254 nm. HRMS (MALDI) (m/z): calcd for
þ
C H N O S , 1510.5183 (M þ H) ; found, 1510.5176.
7
0
84 11 23 2
9
9m
Synthesis and Characterization of DUPA-VrcA (17). The
synthesis of DUPA-VrcA disulfide was performed as described
medium in each well was replaced with 100 nM DUPA- Tc in
the presence of increasing concentration (0.1 nM to 1 μM) of
DUPA-prodrug in fresh medium (0.5 mL). After they were
incubated for 1 h at 37 °C, cells were rinsed with culture medium
(2ꢀ1.0 mL) and tris buffer (1ꢀ1.0 mL) to remove any unbound
radioactivity. Cells were then resuspended in tris buffer (0.5 mL),
and the cell-bound radioactivity was counted using a γ-counter.
The relative binding affinities were calculated using a plot of %
cell bound radioactivity versus the concentration of the test
article using GraphPad Prism 4.
for DUPA-Tub. Analytical RP-HPLC: R
t
=1.4 min [A=10 mM
NH OAc (pH 7.0), and B=CH CN; solvent gradient: 10-100%
4
3
1
B in 4 min]. H NMR (DMSO-d
6
/D
2
O): δ 0.73 (s, 3H), 0.86 (d, J=
6
1
2
3
1
.8 Hz, 3H), 1.05 (m, 1H), 1.15 (m, 1H), 1.23 (s, 1H), 1.33 (m,
H), 1.55 (m, 1H), 1.63 (s, 3H), 1.72 (s, 2H), 1.88-2.08 (m, 2H),
.78 (bd, J=3.5 Hz, 1H), 2.97 (m, 1H), 3.01 (d, J=2.5 Hz, 1H),
.69 (bd, J = 4.7 Hz, 1H), 3.91 (m, 1H), 3.98 (m, 1H), 4.05 (m,
H), 4.17 (m, 1H), 4.31 (m, 1H), 4.39 (m, 1H), 4.82 (s, 1H), 5.30
Analysis of Efficiency of Drug Release Following Disulfide
Reduction. A solution of test article in phosphate buffer (1 mM,
pH 7.4) was treated with a 10-fold molar excess of either gluta-
thione (GSH) or dithiothreitol (DTT) and incubated for differ-
ent lengths of time at 37 °C. The efficiency of active drug release
from the intact DUPA prodrugs was monitored at λ =254 nm
by analytical RP-HPLC and LC/MS as a function of time.
General Procedure for Dose Dependence Studies. LNCaP cells
were seeded in 24-well (50000 cells/well) Falcon plates and allowed
to form monolayers over a period of 48 h. The old medium was then
replaced with fresh medium (0.5 mL) containing increasing con-
centrations of drug (either targeted or nontargeted) in the presence
or absence of excess PMPA (to block all accessible PSMA sites),
and cells were incubated for an additional 2 h at 37 °C. Cells were
washed 3ꢀ with fresh medium and incubated in fresh medium
(
(
7
bd, J=4.7 Hz, 1H), 5.82 (m, 1H), 6.22 (d, J=15.9 Hz, 1H), 6.30
d, J=11.5 Hz, 1H), 6.88 (t, J=11.3 Hz, 2H), 7.21 (ms, 11H),
.84 (t, J = 14.3 Hz, 1H). LRMS (LC/MS) (m/z): 1664.9 (M þ
þ
H) . UV/vis: λmax=254 nm. HRMS (MALDI) (m/z): calcd for
C
þ
77
H
102
N
9
O
Synthesis and Characterization of DUPA-DidB (18). The
28
S
2
, 1664.6276 (M þ H) ; found, 1664.6265.
synthesis of DUPA-DidB was performed as described for DUPA-
TubH. White solid, analytical RP-HPLC: R =1.9 min [A=10 mM
NH
t
4
OAc (pH 7.00, and B=CH
3
CN; solvent gradient: 10-100%
/D O): δ 0.73-0.95 (m, 9H),
.99 (d, J=5.6 Hz, 1H), 1.05 (m, 1H), 1.15 (d, J=6.8 Hz, 3H),
1
B in 4 min]. H NMR (DMSO-d
6
2
0
1
1
.19-1.29 (ms, 4H), 1.34 (d, J = 6.6 Hz, 3H), 1.47-2.26 (ms,
9H), 2.82 (s, 2H), 2.85-2.99 (m, 3H), 3.02 (s, 3H), 3.04-3.25
(
1
(
ms, 3H), 3.35 (s, 3H), 3.47 (dd, J=13.8, 4.2 Hz, 1H), 3.61 (m,
H), 3.72 (s, 3H), 3.79 (m, 1H), 3.92-4.50 (ms, 15H), 4.60-5.00
ms, 12H), 5.16 (d, J= 7.5 Hz, 1H), 5.22 (dd, J = 11.1, 4.0 Hz,
(0.5 mL) for another 66 h at 37 °C. The spent medium in each well
3
was replaced with fresh medium (0.5 mL) containing [ H]-thymi-
dine (1 mCi/mL), and the cells were incubated for additional 4 h at
1
H), 6.85 (d, J=8.5 Hz, 4H), 7.08-7.29 (ms, 10H). LRMS (LC/
þ2
MS) (m/z): [(M þ 2)/2] 1138.2. UV/vis: λmax=254 nm. HRMS
3
3
7 °C to allow [ H]-thymidine incorporation. The cells were then
(
(
MALDI) (m/z): calcd for C107H157N16O34S2, 2274.0490
M þ H) ; found, 2274.0657.
þ
rinsed with medium (2 ꢀ0.5 mL) and treated with 5% trichloro-
acetic acid (0.5 mL) for 10 min at room temperature. After the
trichloroacetic acid was replaced with 0.25 N NaOH (0.5 mL),
cells were transferred to individual scintillation vials containing
Ecolume scintillation cocktail (3.0 mL) and counted in a liquid
Synthesis and Characterization of DUPA-TubH II (19). The
synthesis of DUPA-TubH (linker II) was performed as descri-
bed for DUPA-Tub. Analytical RP-HPLC: R = 2.2 min [A =
t
1
1
6
1
1
2
2
2
0 mM NH
0-100% B in 4 min]. H NMR (DMSO-d
4
OAc (pH 7.0), and B= CH
3
CN; solvent gradient:
/D O): δ 0.66 (d, J=
1
scintillation analyzer. IC values were calculated by plotting
50
6
2
3
[ H]-thymidine incorporation versus log concentration of
%
DUPA-drug conjugate using in GraphPad Prism 4.
.1 Hz, 3H), 0.80 (q, J=7.5 Hz, 5H), 0.96 (d, J=6.3 Hz, 3H),
.00 (d, J = 6.2 Hz, 3H), 1.14 (m, 3H), 1.27 (m, 1H), 1.32 (m,
H), 1.47 (m, 2H), 1.54 (m, 1H), 1.61-1.68 (ms, 2H), 1.81 (m,
H), 1.92 (m, 3H), 2.04 (s, 3H), 2.10 (s, 3H), 2.17-2.39 (ms, 8H),
.66-2.99 (ms, 6H), 3.06 (dd, J = 13.3, 4.4 Hz, 1H), 3.34 (m,
H), 3.87 (m, 1H), 4.00 (m, 1H), 4.06 (m, 1H), 4.20 (m, 1H), 4.24
Results
Screening of Base Drugs. Because PSMA is expressed at
6
only ∼10 copies/PCa cell, it was reasoned that only highly
(
t, J=7.0 Hz, 1H), 4.40 (t, J=8.8 Hz, 1H), 4.44 (m, 1H), 5.25 (d,
potent chemotherapeutic agents could be targeted in sufficient
quantities to kill PCa cells in vivo. In an effort to identify
chemotherapeutic agents with the required potency, human
PCa (LNCaP cell line) cells were pulsed for 2 h with increas-
ing concentrations of nontargeted cytotoxic agent (see the
structures in the Supporting Information, Figure 1), washed
to remove unbound drug, and chased for 66 h in fresh medium
to allow each cytotoxin time to exert its effect. As shown in
Table 1 and Figure 1, cytotoxic drugs such as Tub, DAVB,
and their semisynthetic hydrazide analogues, as well as protein
synthesis inhibitors such as VrcA and DidB, were all highly
J=12.1 Hz, 1H), 5.74 (d, J=11.4 Hz, 1H), 6.22 (bs, 1H), 6.62 (d,
J=7.9 Hz, 2H), 6.99 (d, J=7.9 Hz, 2H), 7.13-7.27 (m, 10H),
7
9
λ
.92 (m, 2H), 8.24 (s, 1H), 8.46 (s, 1H), 9.16 (s, 1H), 9.21 (s, 1H),
þ
.70 (s, 1H). LRMS (LC/MS) (m/z): 1805.4 (M þ H) . UV/vis:
=254 nm. HRMS (MALDI) (m/z): calcd for C H N -
85 122 13
max
24
þ
O
S
3
, 1804.7888 (M þ H) ; found, 1804.7879.
Synthesis and Characterization of DUPA-TubH III (20). The
synthesis of DUPA-TubH (linker III) was performed as de-
scribed for DUPA-Tub. Analytical RP-HPLC: R =7.3 min [A=
t
1
1
5
(
(
(
0 mM NH
-50% B in 10 min]. H NMR (DMSO-d
.5 Hz, 3H), 0.65 (t, J=7.3 Hz, 3H), 0.72 (t, J=7.3 Hz, 3H), 0.76
4
OAc (pH 7.0), and B= CH
3
CN; solvent gradient:
1
6
/D O): δ 0.59 (d, J=
2
active in blocking LNCaP cell proliferation (IC <13 nM).
50
d, J=6.4 Hz, 3H), 0.86 (m, 1H), 0.90 (d, J=6.0 Hz, 3H), 1.00
d, J=6.7 Hz, 3H), 1.08 (m, 2H), 1.15 (b, 3H), 1.34 (m, 7H), 1.41
m, 3H), 1.49 (b, 1H), 1.58 (m, 5H), 1.68 (m, 2H), 1.81 (m, 7H),
On the basis of past experience with folate-targeted drugs,
-
8
an IC50 of <10 M was considered sufficient to ensure efficacy

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7771
Table 1. Effect of Various Nontargeted Cytotoxic Drugs on the Survival
of LNCaP Cells in Culture
a
drug
PTax
class
IC50 (nM) clog P
taxanes
22.3
9.5
4.40
2.89
DAVB
DAVBH
Tub
vinca alkaloids
vinca alkaloids
vinca alkaloids
vinca alkaloids
12.2
2.5
0.73
7.20
TubH
Cpt
Dox
3.7
6.05
0.89
b
topo-isomerase I
topo-isomerase II
none
77.6
3.3
0.32
-1.93
6.34
VrcA
DidB
cis-platin
protein synthesis inhibitors
protein synthesis inhibitors
DNA alkylating agent
2.8
b
b
b
b
b
b
b
b
none
none
none
none
none
none
none
none
-2.83
1.42
4.14
prednisolone anti-inflammatory agent
RWJ67657
SB203580
wortmannin PI3 kinase inhibitor
P38 MAP kinase inhibitor
P38 MAP kinase inhibitor
2.99
0.21
þ
þ
nigericin
paraquat
atractyloside inhibit oxidative phosphorylation
antibiotic/antiporter of H and K
electron acceptor
2.45
-8.65
1.38
a
b
clog P = calculated log (partition coefficient). None, not active within
the 100 pM to 1 μM range.
4
0
for most ligand-targeted therapeutic agents, and consequently,
the above drugs were examined further for their killing
potencies as DUPA-targeted conjugates.
Figure 1. Effect of the drug concentration on the survival of a human
PCa cell line. Free drugs dissolved in DMSO were added at the
indicated concentrations to LNCaP cells in RPMI culture media and
allowed to incubate for 2 h at 37 °C. Media were then removed and
replaced with fresh media (drug-free), and incubation was continued
Synthesis of PSMA-Targeted Disulfide-Bridged Self-
Immolative DUPA Prodrugs. Syntheses of activated mixed
disulfide forms of TubH (6) and desacetylvinblastine hydrazide
DAVBH) (9) are shown in Scheme 1. β-Mercaptoethanol (1)
was first reacted with methoxycarbonylsulfenyl chloride and
-thiopyridine to generate 2-(pyridine-2-yl-disulfanyl) ethanol
2) in quantitative yield. Compound 2 was then treated with
3
for an additional 66 h. Cell survival was then quantitated using a H-
thymidine assay, as described in the General Methods. PTax, paclitaxel;
Dox, doxorubicin; cis-Pt, cis-platin; DAVBH, desacetylvinblastine
hydrazide. Error bars represent SD (n = 3).
(
2
(
neutral pH under argon to produce the desired PSMA-
targeted disulfide-bridged self-immolative prodrug, DUPA-
TubH I 14 (Scheme 3). Compounds 15-18 (Figure 3) were syn-
thesized by analogous methods. Three closely related
DUPA-TubH conjugates 14, 19, and 20 (Figure 3) were also
prepared by varying the length and chemistry of the spacer to
evaluate the contribution of the spacer to the solubility and
specificity of the final drug conjugate.
diphosgene to form a chloroformate intermediate, followed
by HOBT to furnish compound 3. This HOBT-activated
analogue 3 was transformed into the heterobifunctional
linker 4 by reacting with hydrazine in dichloromethane.
The carboxylic acid moiety in Tub was then activated with
isobutylchloroformate and reacted with the heterobifunctional
linker 4 to generate an activated mixed disulfide derivative
of TubH (6). In the case of the activated mixed disulfide of
DAVBH (8), vinblastine was first reacted with hydrazine in
methanol followed by transformation into the activated mixed
disulfide derivative of DAVBH (9) using HOBT analogue 3,
as shown in Scheme 1. Along similar lines, activated mixed
disulfide derivatives of DidB (12), Cpt (10), and VrcA (11)
were synthesized using their HOBT analogues 3 (Figure 2).
Design of the optimal spacer linking the aforementioned
drugs to DUPA required selection of substituents that would
not only improve binding affinity but also enhance overall
water solubility and avoid nonspecific uptake by scavenger
receptors in the liver. Because earlier studies with DUPA-
targeted radioimaging agents revealed that optimal binding
occurred when the spacer contained two phenylalanine residues
at appropriate distances from DUPA, a similar design was
incorporated into construction of each of the therapeutic conju-
gates. To improve water solubility, several acidic groups
were added toward the C terminus of the spacer and com-
pared for optimal binding. Synthesis of these optimized
DUPA-peptide spacers (13a-c) bearing a COOH-terminal
cysteine was accomplished by standard fluorenylmethyloxy-
carbonyl (Fmoc) SPPS starting from the Fmoc-Cys(Trityl)-
Wang resin, as shown in Scheme 2.
Analysis of the Affinity of DUPA-Prodrug Conjugates for
PSMA and Evaluation of Base Drug Release following Disulfide
Reduction. The affinity of each DUPA-prodrug conjugate
for PSMA was estimated by measuring the concentration of
99m
DUPA-prodrug required to block 50% binding of DUPA- Tc
to LNCaP cells. For this purpose, LNCaP cells were incu-
9
9m
bated in the presence of 100 nM DUPA- Tc plus increas-
ing concentrations (0.1 nM to 1 μM) of DUPA-prodrug (e.g.,
DUPA-TubH). Following three washes in saline to remove
unbound DUPA conjugate, cells were counted for bound radio-
99m
activity. On the basis of an absolute affinity of DUPA- Tc
2
for PSMA of K = 14 nM, the dissociation constant of
9
d
DUPA-TubH I (14) for PSMA was calculated at 26 nM
(Figure 4). The measured affinities of the other DUPA-
prodrug conjugates range from 7 to 70 nM.
3
5
The efficiency of base drug release from its respective DUPA-
prodrug conjugate was next assessed by treating each con-
jugate with either GSH or DTT to reduce the mixed disulfide,
followed by analysis of products by analytical RP-HPLC
and LC/MS. HPLC chromatograms (e.g., see Figure 5a)
indicated that over 98% of disulfide bonds were reduced within
2-4 h upon addition of 10-fold excess GSH. To establish the
mechanism of active drug release, selected samples were also
analyzed by LC/MS at different time intervals (Figure 5b). In
the case of DUPA-TubH III, as shown in Scheme 4, reduction
The activated mixed disulfide-bridged derivative of TubH
was exchanged with the DUPA-spacer-Cys I (13) in water at

7
772 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Kularatne et al.
a
Scheme 1
a
0
Reagents and conditions: (a) (i) MeOC(O)SCl/CH
CN, 0 °C; (ii) Δ/50 °C, 24 h. (c) NH NH , DIPEA/CH
5 min; (ii) 4/ EtOAc, -15 °C to room temperature, 45 min. (e) NH
Py = pyridyl.
3
CN, 0 °C, 0.5 h; (ii) 2,2 -dipyridyl-disulfide/CH
3
CN, 0 °C, 2 h. (b) (i) Diphosgene, HOBt, TEA/
CH
3
2
2
2
Cl
2
, 0 °C to room temperature, 45 min. (d) (i) Isobutylchlorofomate, DIPEA/EtOAc, -15 °C,
NH Cl , room temperature, 1 h.
/MeOH, 60 °C, 12 h. (f) Compound 3, DIPEA/CH
4
2
2
2
2
Figure 2. Structures of activated mixed disulfide-bridged cytotoxic agents. TubH (6), TubH; DAVBH (9), DAVBH; Cpt (10), Cpt; VrcA (11),
VrcA; and DidB (12), DidB (R = p-methoxyphenyl).
of the disulfide bond by DTT generated a thiol intermediate
that would eventually be expected to release unmodified
tubulysin hydrazide via either path A or path B (Scheme 4).
On the basis of LC/MS analysis, over 95% of the disulfide
bonds were found to form free TubH via the carbonate
intermediate 21 (Figure 5b), suggesting that efficient drug
release occurs primarily via path A. Efficient release of unmodi-
fied drug upon exposure to GSH was encouraging, since it
suggested that liberation of unmodified TubH should pro-
ceed spontaneously following disulfide bond reduction within
the endosomes of PCa cells. Importantly, previously pub-
lished in vivo studies using the DUPA-tubulysin III (20)
conjugate have demonstrated that the conjugate is suffi-
ciently stable in circulation to reach its receptor on LNCaP
tumors in its intact nonreduced form.
Cytotoxicity of DUPA-Prodrug Conjugates. Analysis of
the cytotoxicity of each PSMA-targeted prodrug was then
conducted by exposing LNCaP cells to the DUPA-prodrug
conjugate for 2 h, followed by restoration of the cells to
drug-free media for 66 h and then analysis of cell viability by
quantitation of H-thymidine incorporation. As shown in
3
Figure 6a, H-thymidine incorporation dramatically de-
creased with increasing concentrations of DUPA-TubH
I-III and DUPA-DAVBH. Indeed, all DUPA-prodrug
conjugates with the exception of DUPA-VrcA (17) showed
high cytotoxicity for LNCaP cells, with IC50 values in
the low nanomolar range. Most importantly, the high
efficacy of all PSMA-targeted prodrugs was quantitatively
inhibited by competition with 100-fold excess PMPA
(Figure 6b), suggesting that cell killing required the pre-
sence of an empty PSMA receptor and did not occur by
extracellular release of active drug followed by passive diffu-
sion into the cell. This result is significant, since PCa
cells constitute the primary site where PSMA is exposed
3
2
9

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7773
and accessible in humans to parenterally administered PSMA-
targeted drugs.
targeted drug must be high, since most receptor-mediated
endocytosis pathways deliver only limited quantities of drug.
The spacer that links the drug to its targeting ligand must
also be relatively stable to ensure its arrival at the tumor mass
intact, yet sufficiently labile to enable rapid release of the
unmodified therapeutic agent following internalization into
the malignant cell. Indeed, many ligand-drug conjugates have
failed because free drug has been released before the targeting
ligand could mediate its uptake by the target cell (personal
observations). Now, as observed in this paper, the innate stabil-
ity of the therapeutic drug in its ligand-targeted conjugate
would also appear to impact performance of the drug con-
jugate. Although Cpt was relatively inactive against LNCaP
cells in its nontargeted form (Table 1), the drug was still found
to be highly potent as a DUPA conjugate (Table 2). We
hypothesize that attachment of the DUPA-spacer to Cpt at its
Discussion
We have shown here that the PSMA-targeted ligand, DUPA,
can beexploited to deliver a varietyof cytotoxic drugs that can
kill cells that express elevated levels of PSMA. Because future
advances in personalized medicine will almost certainly require
different therapeutic warheads for different PCa patients, this
versatility in drug delivery could prove beneficial when
attempting to fit any desired therapy to any individual patient.
Expression of PSMA on the neovasculature of most, if not all,
solid tumors will only add to the need for an adaptable
targeting ligand for multiple therapeutic needs.
Although ligand targeting shows considerable promise for
improving the potency and toxicity of multiple cancer ther-
apeutics, simple attachment of a cytotoxic drug to a tumor-
specific ligand may not ensure achievement of the above
2
as reported when other moieties have been conjugated to this
0-OHmayhave stabilizedthe lactonering againsthydrolysis,
3
8-40
1
5
same site.
Such stabilization would have then rendered
properties. As noted previously, the potency of any ligand-
the conjugate capable of surviving the 2 h incubation in culture
medium without hydrolyzing to its inactive form. Conversely,
failure of the DUPA-VrcA conjugate to kill LNCaP cells in
culture despite its high potency in free form may have derived
from the different routes of cellular entry of VrcA in its targeted
and nontargeted forms. Thus, hydrolases in endosomal com-
partments could have cleaved either the epoxide or the lactone
ring of VrcA during intracellular PSMA trafficking, whereas
the free drug may have found direct access to its site of action
within the cell. Taken together, it is becoming more apparent
that multiple variables may require optimization before a
potent cytotoxic drug can be converted into a successfully
targeted therapeutic agent.
a
Scheme 2
As efforts to develop ligand-targeted therapies increase, it is
important to consider both their advantages and disadvantages
as compared with molecular-targeted therapies. As noted above,
a major advantage of ligand targeted therapies can stem from
their remarkable therapeutic flexibility, since almost any potent
drug can be targeted to a diseased tissue if it can be linked in
a reversible manner to an appropriate targeting ligand. How-
ever, because only a few chemical functionalities allow for
facile release of a covalently attached drug following uptake
by the targeted cell, drugs must be selected that contain one
or more of a limited number of chemical moieties (-SH,
a
Reagents and conditions: (a) (i) 20% piperidine/DMF, room tempera-
t
ture, 10 min; (ii) Fmoc-Asp(O Bu)-OH, HBTU, HOBt, DMF/DIPEA, 2 h.
(
b) (i) 20% piperidine/DMF, room temperature, 10 min; (ii) Fmoc-diamino-
propionic (DAP) acid, HBTU, HOBt, DMF/DIPEA, 2 h. (c) (i) 20%
piperidine/DMF, room temperature, 10 min; (ii) Fmoc-Phe-OH, HBTU,
HOBt, DMF/DIPEA, 2 h. (d) (i) 20% piperidine/DMF, room temperature,
1
0 min; (ii) Fmoc-Phe-OH, HBTU, HOBt, DMF/DIPEA, 2 h. (e) (i) 20%
piperidine/DMF, room temperature, 10 min; (ii) Fmoc-8-amino-octanoic
EAO) acid, HBTU, HOBt, DMF/DIPEA, 2 h. (f) (i) 20% piperidine/
(
t
DMF, room temperature, 10 min; (O Bu)
DIPEA, 2 h. (g) TFA/H O/TIPS/EDT (92.5:2.5:2.5:2.5), 1 h.
3
-DUPA-OH, HBTU, HOBt,
-COOH, -OH, or -NH ) that can be adapted for intra-
2
15,29,36-38
2
cellular release.
A second advantage of ligand-targeted
a
Scheme 3
a
2 3
Reagents and conditions: (a) (i) H O/NaHCO (pH 7.0-7.2), argon, room temperature; (ii) 6/THF, argon, room temperature, 15 min.

7
774 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Kularatne et al.
Figure 3. Structures of PSMA-targeted disulfide-bridged self-immolative prodrugs. The base drug nomenclature is as in Figure 2.
therapies derives from the ability to design a cognate imaging
1
generation of an imaging agent targeted to the same receptor
renders compliance with this FDA condition relatively easy.
Third, whenever membrane impermeable drugs must be
delivered, ligand-targeted therapies are generally preferred,
since a goodtargeting ligand willoften carry its attached cargo
into the target cell by receptor-mediated endocytosis. Thus,
5,29,35,41-43
agentwiththe sametargetingligand.
Becausesuch
cognate imaging agents can be used to select patients for the
ligand-targeted therapy and because the FDA is increasingly
desirous that a diagnostic agent be employed to identify
potential “responders” to any new drug, the straightforward

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7775
when proteins, dendrimers, liposomes, genes, nanoparticles,
siRNA, peptides, miRNA, etc. constitute the therapeutic cargo,
the use of an internalizing targeting ligand can render an
4
3,44
otherwise inactive drug more efficacious.
Finally, because
overexpression of a receptor on a pathologic cell may be more
common than expression of a novel/upregulated enzyme or
pathway ina pathologic cell, opportunities fordevelopment of
ligand-targeted therapies may arise more frequently than
opportunities for novel functionality-targeted therapies.
Ligand-targeted therapies also exhibit disadvantages rela-
tivetofunctionality-targeted andnontargeted therapies. First,
because most endocytic pathways transport relatively few
molecules into a cell, the ligand-targeted drug must be effec-
tive at very low concentrations. For some diseases, a large
selection of highly potent therapeutic agents is readily avail-
able, but for other pathologies, the repertoire of potent drugs
is small. Second, as noted above, an efficient drug release
Figure 4. Relative binding affinities of DUPA-prodrugs with
respect to DUPA-
9
9 m
Tc (a PSMA-targeted radioimaging agent).
LNCaP cells were incubated for 1 h at 37 °C in the presence of
99 m
1
00 nM DUPA-
Tc with increasing concentrations of DUPA-
prodrugs. Cell-bound radioactivity was assayed as described in the
General Methods.
Figure 5. Release of TubH (21) from DUPA-TubH self-immolative prodrug (14 or 20) by disulfide reduction. DUPA-TubH was analyzed
using (a) analytical RP-HPLC [A = 10 mM NH OAc (pH 7.0), and B = CH CN; λ = 257 nm: 10-100% B in 4 min] and (b) LC/MS [A = 10 mM
NH OAc (pH 7.0), and B = CH
4
3
4
3
CN; λ = 257 nm: 5-80% B in 8 min] in the absence (left chromatogram) and presence (right chromatogram)
of a excess of (a) GSH or (b) DTT. See Scheme 4 for identities of products of the disulfide reduction reaction.
Scheme 4. Possible Mechanisms for Disulfide-Mediated Release of TubH from PSMA-Targeted Self-Immolative Prodrugs, Such as
DUPA-TubH III, in the Presence of Disulfide Reducing Agents

7
776 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Kularatne et al.
compounds, and HPLC profiles for targeted drug conjugates.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(
1) Van der Hul, R. L.; Seynaeve, C.; van Gell, B. N.; Verweij, J. Low
dose of methotrexate and vinblastine, given weekly to patients with
Desmoid tumors, is associated with major toxicity. Sarcoma 2003,
7, 153–157.
(
2) Onyenadum, A.; Gogas, H.; Markopoulos, C.; Bafaloukos, D.;
Aravantinos, G.; Mantzourani, M.; Koutras, A.; Tzorakoelefterakis,
E.; Xiros, N.; Makatsoris, T.; Fountzilas, G.; Kalofonos, H. P.
Mitoxantrone plus vinorelbine in pretreated patients with meta-
static breast cancer. J. Chemother. 2007, 19, 582–589.
(
(
3) Rahmani, R.; Zhou, X. J. Pharmacokinetics and metabolism of
vinca alkaloids. Cancer Surv. 1993, 17, 269–281.
4) Rowinsky, E. K.; Chaudhry, V.; Cornblath, D. R.; Donehower,
R. C. Neurotoxicity of Taxol. J. Natl. Cancer Inst. Monogr. 1993,
1
5, 107–115.
(
5) Rowinsky, E. K.; Eisenhauer, E. A.; Chaudhry, V.; Arbuck, S. G.;
Donehower, R. C. Clinical toxicities encountered with paclitaxel
(
Taxol). Semin. Oncol. 1993, 20, 1–15.
(
6) Ivachtchenko, A. V.; Kiselyov, A. S.; Tkachenko, S. E.; Ivanenkov,
Y. A.; Balakin, K. V. Novel mitotic targets and their small-
molecule inhibitors. Curr. Cancer Drug Targets 2007, 7, 766–784.
7) Druker, B. J. Translation of the Philadelphia chromosome into
therapy for CML. Blood 2008, 112, 4808–4817.
(
(
(
8) Rabatti, D. The discovery of antiangiogenic molecules: a historical
Figure 6. Effect of DUPA-targeted drug concentration on the survival
of human PCa cells in the absence (a) and presence (b) of excess
competitive inhibitor of PSMA (PMPA). DUPA-drug conjugates,
dissolved in saline, were added at the indicated concentrations to
LNCaP cells in RPMI culture media and allowed to incubate for 2 h
at 37 °C. In competition experiments, 100-fold excess PMPA was added
prior to DUPA conjugate addition to block all available PSMA sites on
the LNCaP cells. Following 2 h of incubation, media were removed
and replaced with fresh media (drug-free), and incubation was con-
tinued for an additional 66 h. Cell survival was then quantitated using
review. Curr. Pharm. Des. 2009, 15, 345–352.
9) Li, W. W.; Hutnik, M.; Gehr, G. Antiangiogenesis in haemato-
logical malignancies. Br. J. Hamaetol. 2008, 143, 622–631.
(
10) Mauro, M. J. Appropriate sequencing of tyrosine kinase inhibitors
in chronic myelogenous leukemia: When to change? A perspective
in 2009. Curr. Opin. Hematol. 2009, 16, 135–139.
(11) Breccia, M.; Frustaci, A. M.; Cannella, L.; Soverini, S.; Stefanizzi,
C.; Federico, V.; Grammatico, S.; Santopietro, M.; Alimena, G.
Sequential development of mutant clones in an imatinib resistant
chronic myeloid leukaemia patient following sequential treatment
with multiple tyrosine kinase inhibitors: An emerging problem?
Cancer Chemother. Pharmacol. 2009, 64, 195–197.
3
a H-thymidine assay, as described in the General Methods. Error bars
represent SD (n=3).
(
12) Ojima, I. Guided molecular missiles for tumor-targeting chemotherapy-
case studies using the second-generation taxoids as warheads. Acc.
Chem. Res. 2008, 41, 108–119.
Table 2. Effect of Various DUPA-Targeted Cytotoxic Drugs on the
Survival of LNCaP Cells in Culture
(13) Balhorn, R.; Hok, S.; Burke, P. A.; Lightstone, F. C.; Cosman, M.;
Zemla, A.; Mirick, G.; Perkins, J.; Natarajan, A.; Corzett, M.;
DeNardo, S. J.; Albrecht, H.; Gregg, J. P.; DeNardo, G. L. Selec-
tive high-affinity ligand antibody mimics for cancer diagnosis and
therapy: Initial application to lymphoma/leukemia. Clin. Cancer
Res. 2007, 15, 5621s–5628s.
a
b
targeted drug
RF
IC50 (nM)
clog P
1
1
1
1
1
1
2
4
5
6
7
8
9
0
26
25
7
6 ( 2
31 ( 3
115
3.38
-2.26
-2.99
-4.77
4.19
(
14) Allen, T. M. Ligand-targeted therapeutics in anticancer therapy.
c
34
70
none
Nat. Rev. Cancer 2002, 2, 750–763.
107
24
(15) Low, P. S.; Kularatne, S. A. Folate-targeted therapeutic and
imaging agents for cancer. Curr. Opin. Chem. Biol. 2009, 13, 256–
262.
d
NA
NA
7.70
3.60
d
5 ( 2
(
16) Larsen, A. K.; Escargueil, A. K.; Skladanowski, A. Resistance
mechanisms associated with altered intracellular distribution of
anticancer agents. Pharmacol. Ther. 2000, 85, 217–229.
a
b
RF, relative affinity. clog P = calculated log (partition coefficient).
See the legend to Figure 2 for structures of DUPA-targeted drugs.
None, not active within the 100 pM to 1 μM range; SD (n = 2). NA,
not applicable.
c
d
(17) Isaken, K.; Jonsson, R.; Omdal, R. Anti-CD20 treatment in
primary Sj o€ gren’s syndrome. Scand. J. Immunol. 2008, 68, 554–564.
(
(
(
18) Gisselbrecht, C. Use of rituximab in diffuse large B-cell lymphoma
in the salvage setting. Br. J. Hamaetol. 2008, 143, 607–621.
19) Hall, P. C.; Cameron, D. A. Current perspective-trastuzumab. Eur.
J. Cancer 2009, 45, 12–18.
20) Jean, G. W.; Shah, S. R. Epiderm growth factor receptor mono-
clonal antibodies for the treatment of metastatic colorectal cancer.
Pharmacotherapy 2008, 28, 742–754.
mechanism must be designed that is inert during transit to the
pathologic lesion but rapidly activated following target cell
binding/internalization, leading to drug release at the site of
1
2,37,38,41,43
disease.
Finally, for membrane impermeabledrugs,
an endosome escape strategy must also be designed that will
ensure access of the released drug to the cytosol following
uptake by the diseased cell. While much progress has been
made in the development of these escape mechanisms, signif-
icant improvements will be necessary before the full potential
of macromolecular drugs such as siRNAs, proteins, and nano-
(
21) Zwanziger, D.; Beck-Sickinger, A. G. Radiometal targeted tumor
diagnosis and therapy with peptide hormones. Curr. Pharm. Des.
008, 14, 2385–2400.
2
(22) Schally, A. V. New approaches to the therapy of various tumors
based on peptide analogues. Horm. Metab. Res. 2008, 40, 315–322.
(
23) Nishimura, Y.; Bereczky, B.; Ono, M. The EGFR inhibitor
gefitinib suppresses ligand-stimulated endocytosis of EGFR via
the early/late endocytic pathway in non-small cell lung cancer cell
lines. Histochem. Cell Biol. 2007, 127, 541–553.
24) Ye, X.; Yang, D. Recent advances in biological strategies for
targeted drug delivery. Cardiovasc. Hematol. Disord. Drug Targets
4
5-48
particles can be realized.
Acknowledgment. This work was supported by a grant
from Endocyte, Inc.
(
(
2
25) Yan, H.; Tram, K. Glycotargeting to improve cellular delivery
009, 9, 206–221.
Supporting Information Available: Structures of nontargeted
cytotoxic drugs, copies of H NMR spectra of all newly synthesized
1
efficiency of nucleic acids. Glycoconjugate J. 2007, 24, 107–123.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7777
(
(
(
26) Irache, J. M.; Salman, H. H.; Gamazo, C.; Espuelas, S. Mannose-
targeted systems for the delivery of therapeutics. Expert Opin. Drug
Delivery 2008, 5, 703–724.
chemotherapeutics. Part II: Folic acid conjugates of tubulysins
and their hydrazides. Bioorg. Med. Chem. Lett. 2008, 18, 4558–4561.
(37) Vlahov, I. R.; Santhapuram, H. K.; Kleindl, P. J.; Howard, S. J.;
Stanford, K. M.; Leamon, C. P. Design and regioselective synthesis
of a new generation of targeted chemotherapeutics. Part 1: EC145,
a folic acid conjugate of desacetylvinblastine monohydrazide.
Bioorg. Med. Chem. Lett. 2006, 16, 5093–5096.
(38) Henne, W. A.; Doorneweerd, D. D.; Hilgenbrink, A. R.; Kularatne,
S. A.; Low, P. S. Synthesis and activity of a folate peptide campto-
thecin prodrug. Bioorg. Med. Chem. Lett. 2006, 16, 5350–5355.
(39) Pessah, N.; Reznik, M.; Shamis, M.; Yantiri, F.; Xin, H.; Bowdish,
K.; Shomron, N.; Ast, G.; Shabat, D. Bioactivation of carbamate-
based 20(S)-camptothecin prodrugs. Bioorg. Med. Chem. 2004, 12,
1859–1866.
27) Corti, A.; Curnis, F.; Arap, W.; Pasqualini, R. The neovasculature
homing motif NGR: More than meets the eye. Blood 2008, 112,
2628–2635.
28) Gupta, Y.; Kohli, D. V.; Jain, S. K. Vitamin B12-mediated trans-
port: A potential tool for tumor targeting of antineoplastic drugs
and imaging agents. Crit. Rev. Ther. Drug Carrier Syst. 2008, 25,
347–379.
(
29) Kularatne, S. A.; Wang, K.; Santhapuram, H. K.; Low, P. S.
Prostate-specific membrane antigen targeted imaging and therapy
of prostate cancer using a PSMA inhibitor as a homing ligand. Mol.
Pharmaceutics 2009, 6, 780–789.
(
30) Ghosh, A.; Heston, W. D. Tumor target prostate specific mem-
brane antigen (PSMA) and its regulation in prostate cancer. J. Cell
Biochem. 2004, 91, 528–539.
(40) de Groot, F. M.; Busscher, G. F.; Aben, R. W.; Scheeren, H. W.
Novel 20-carbonate linked prodrugs of camptothecin and 9-amino-
camptothecin designed for activation by tumour-associated plasmin.
Bioorg. Med. Chem. Lett. 2002, 12, 2371–2376.
(41) Kratz, F.; Muller, I. A.; Ryppa, C.; Warnecke, A. Prodrug
strategies in anticancer chemotherapy. Chem. Med. Chem. 2008,
3, 20–53.
(
31) Chang, S. S.; O’Keefe, D. S.; Bacich, D. J.; Reuter, V. E.; Heston,
W. D.; Gaudin, P. B. Prostate-specific membrane antigen is
produced in tumor-associated neovasculature. Clin. Cancer Res.
1999, 5, 2674–2681.
(
(
(
32) Liu, H.; Rajasekaran, A. K.; Moy, P. Constitutive and antibody-
induced internalization of prostate-specific membrane antigen.
Cancer Res. 1998, 58, 4055–4060.
33) Silver, D. A.; Pellicer, I.; Fair, W. R.; Heston, W. D.; Cordon-
Cardo, C. Prostate-specific membrane antigen expression in normal
and malignant human tissues. Clin. Cancer Res. 1997, 3, 81–85.
34) Slusher, B. S.; Tsai, G.; Yoo, G.; Coyle, J. T. Immunocytochemical
localization of the N-acetyl-aspartyl-glutamate (NAAG) hydrolyzing
enzyme N-acetylated R-linked acidic dipeptidase (NAALADase).
J. Comp. Neurol. 1992, 315, 217–229.
(42) Sega, E. I.; Low, P. S. Tumor detection using folate receptor-
targeted imaging agents. Cancer Metastasis Rev. 2008, 27, 655–664.
(43) Zhao, X.; Li, H.; Lee, R. J. Targeted drug delivery via folate
receptor. Expert Opin. Drug Delivery 2008, 5, 309–319.
(44) Tarrag oꢀ -trani, M. T.; Storrie, B. Alternative routes for drug
delivery to the cell interior: Pathways to the Golgi apparatus and
endoplasmic reticulum. Adv. Drug Delivery Rev. 2007, 59, 782–797.
(45) Whitehead, K. A.; Langer, R.; Anderson, D. G. Knocking down
barriers: advances in siRNA delivery. Nat. Rev. Drug Discovery
2009, 8, 129–138.
(35) Kularatne, S. A.; Zhou, Z.; Yang, J.; Post, C. B.; Low, P. S. Design,
(46) Aigner, A. Cellular delivery in vivo of siRNA-based therapeutics.
synthesis, and preclinical evaluation of prostate-specific membrane
Curr. Pharm. Des. 2008, 14, 3603–3619.
99m
antigen targeted
009, 6, 790–800.
36) Vlahov, I. R.; Wang, Y.; Kleindl, P. J.; Leamon, C. P. Design
Tc-radioimaging agents. Mol. Pharmaceutics
(47) Oh, Y. K.; Park, T. G. siRNA delivery systems for cancer treat-
ment. Adv. Drug Delivery Rev. 2009, 61, 850–862.
(48) Walchli, S.; Sioud, M. Vector-based delivery of siRNAs: in vitro
2
(
and regioselective synthesis of a new generation of targeted
and in vivo challenges. Front. Biosci. 2008, 13, 3488–3493.