64Cu-labeled triphenylphosphonium and triphenylarsonium cations as highly tumor-selective imaging agents

Article abstract of DOI:10.1021/jm0704088

This report presents synthesis and evaluation of the ⁶⁴Cu- labeled triphenylphosphonium (TPP) cations as new radiotracers for imaging tumors by positron emission tomography. Biodistribution properties of ⁶⁴Cu-L1, ⁶⁴Cu-L2,

Full text of DOI:10.1021/jm0704088

J. Med. Chem. 2007, 50, 5057-5069
5057
Articles
64Cu-Labeled Triphenylphosphonium and Triphenylarsonium Cations as Highly
Tumor-Selective Imaging Agents
Jianjun Wang,† Chang-Tong Yang,† Young-Seung Kim,† Subramanya G. Sreerama,† Qizhen Cao,‡ Zi-Bo Li,‡ Zhengjie He,§
Xiaoyuan Chen,‡ and Shuang Liu*,†
School of Health Sciences, Purdue UniVersity, West Lafayette, Indiana, Molecular Imaging Program at Stanford, Department of Radiology &
Bio-X, Stanford UniVersity, Stanford, California, and State Key Laboratory of Elemento-Organic Chemistry, Nankai UniVersity,
Tianjin 300071, P. R. China
ReceiVed April 5, 2007
This report presents synthesis and evaluation of the 64Cu-labeled triphenylphosphonium (TPP) cations as
new radiotracers for imaging tumors by positron emission tomography. Biodistribution properties of 64Cu-
L1, 64Cu-L2, 64Cu-L3, and 99mTc-Sestamibi were evaluated in athymic nude mice bearing U87MG human
glioma xenografts. The most striking difference is that 64Cu-L1, 64Cu-L2, and 64Cu-L3 have much lower
heart uptake (<0.6% ID/g) than 99mTc-Sestamibi (∼18% ID/g) at >30 min p.i. Their tumor/heart ratios
increase steadily from ∼1 at 5 min p.i. to ∼5 at 120 min p.i. The tumor/heart ratio of 64Cu-L3 is ∼40 times
better than that of 99mTc-Sestamibi at 120 min postinjection. Results from in vitro assays show that 64Cu-L1
is able to localize in tumor mitochondria. The tumor is clearly visualized in the tumor-bearing mice
administered with 64Cu-L1 as 30 min postinjection. The 64Cu-labeled TPP/TPA cations are very selective
radiotracers that are able to provide the information of mitochondrial bioenergetic function in tumors by
monitoring mitochondrial potential in a noninvasive fashion.
Introduction
cells is sufficient to account for a 10-fold greater accumulation
of cationic compounds in carcinoma cells than in normal
cells.1-4,11 In addition, the plasma transmembrane potential (30-
90 mV) can also preconcentrate the cationic species, thus
affecting their cytoplasmic concentration and the availability
for their mitochondrial uptake.1-4
Cationic radiotracers, such as 99mTc-Sestamibi and 99mTc-
Tetrofosmin (Figure 1), originally developed as radiotracers for
myocardial perfusion imaging by single photon emission
computed tomography (SPECT), are also able to localize in
tumor cells because of the increased mitochondrial transmem-
brane potential. Both 99mTc-Sestamibi and 99mTc-Tetrofosmin
have been successfully used for imaging cancers and the
transport function of multidrug resistance P-glycoproteins
(particularly MDR1).17-25 However, their cancer diagnostic
values are often limited because of their insufficient tumor
localization and high uptake in the heart and liver, which makes
it very difficult to detect small lesions in the chest and abdominal
regions.
More than 20 years ago, radiolabeled cations of quaternary
ammonium, phosphonium and arsonium were studied as ra-
diotracers for heart imaging.26-28 Recently, several groups
proposed to use the radiolabeled triphenylphosphonium (TPP)
cations, such as 4-(18F-benzyl)triphenylphosphonium (Figure
1: 18F-BzTPP), as positron emission tomography (PET) radio-
tracers for both tumor and myocardial perfusion imaging.29-37
3H-Tetraphenylphosphonium (Figure 1: 3H-TPP) was reported
to have a better tumor uptake than 99mTc-Sestamibi, but its tumor
selectivity is very poor with the tumor/heart ratio being
,1.0.31,35 In addition, 3H-TPP is not suitable for imaging
purposes. The high uptake of 18F-BzTPP in the heart and liver
may also impose a significant challenge for its routine clinical
applications in diagnosis of cancer in the chest and abdominal
Alteration in the mitochondrial transmembrane potential
(∆Ψm) is an important characteristic of cancer and is caused
directly by mitochondrial dysfunction, such as DNA mutation
and oxidative stress.1-4 It has been demonstrated that the
mitochondrial transmembrane potential in carcinoma cells is
significantly higher than that in normal epithelial cells.5-9 For
example, the difference in ∆Ψm between the colon carcinoma
cell line CX-1 and the control green monkey kidney epithelial
cell line CV-1 was approximately 60 mV (163 mV in tumor
cells versus 104 mV in normal cells). The observation that the
enhanced mitochondrial transmembrane potential is prevalent
in the tumor cell phenotype provides the conceptual basis for
development of mitochondrion-targeting pharmaceuticals and
imaging probes.1-3,10-13
Measurement of mitochondrial potential provides the most
comprehensive reflection of mitochondrial bioenergetic function
because it directly depends on the proper integration of diverse
metabolic pathways that converge at mitochondria. Since plasma
and mitochondrial transmembrane potentials are negative,
delocalized cationic molecules with appropriate structural
features can be driven electrophoretically through these mem-
branes and accumulate inside the energized mitochondria.1-4,11
3
Lipophilic organic cations, such as rhodamine-123 and H-
tetraphenylphosphonium (3H-TPP), have been widely used to
measure mitochondrial potentials in tumor cells.1,14-16 According
to the Nernst equation, the 60 mV difference in mitochondrial
transmembrane potential between the carcinoma and epithelial
* To whom correspondence should be addressed. Phone:
765-494-0236; fax 765-496-1377; e-mail: lius@pharmacy.purdue.edu.
† Purdue University.
‡ Stanford University.
§ Nankai University.
10.1021/jm0704088 CCC: $37.00 © 2007 American Chemical Society
Published on Web 09/15/2007

5058 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Wang et al.
Figure 1. Cationic radiotracers useful for tumor imaging. 99mTc-Sestamibi and 99mTc-Tetrofosmin have been widely used to image tumors of
different origin by SPECT. 64Cu-L1, 64Cu-L2, and 64Cu-L3 are evaluated in this study.
regions. Thus, there is an urgent need for new radiotracers that
have high tumor-selectivity and are able to provide the informa-
tion of mitochondrial bioenergetic function in tumors by
monitoring mitochondrial potential in the noninvasive fashion.
Copper has several radionuclides, such as 60Cu, 61Cu, 62Cu,
64Cu, and 67Cu. Among them, 62Cu and 64Cu are particularly
useful for PET imaging. 62Cu has a half-life of 9.7 min and
decays by positron emission (194% abundance). The short half-
life of 62Cu allows the repeated dose without imposing a
significant radiation burden to the patient. 64Cu has a half-life
of 12.7 h and a low â+ emission rate (18%) with maximum â+
energy of 0.655 MeV. The long half-life of 64Cu makes it more
feasible to prepare small biomolecule radiotracers. The rich
coordination chemistry of copper in combination with diverse
nuclear properties of its radionuclides offers many opportunities
for development of diagnostic (60Cu, 61Cu, 62Cu, and 64Cu) and
therapeutic (64Cu and 67Cu) radiotracers. Copper chemistry and
radiochemistry, along with medical applications of 62Cu- and
64Cu-labeled biomolecules, have been reviewed extensively.38-41
Recently, we prepared three DO3A (1,4,7,10-tetraazacyclo-
dodecane-4,7,10-triacetic acid)-conjugated TPP and triphen-
ylarsonium (TPA) cations and their 64Cu complexes, 64Cu-L1,
64Cu-L2, and 64Cu-L3 (Figure 1). TPP or TPA acts as the
“mitochondrion-targeting biomolecule” to carry 64Cu into tumor
cells that have a much higher mitochondrial transmembrane
potential than normal cells. DO3A is of particular interest as a
bifunctional chelator because it is able to form a stable 64Cu-
DO3A chelate. DO3A also forms highly stable radiometal
chelates with many other radionuclides, such as 68Ga for PET,
111In for SPECT, and 90Y and 177Lu for radiotherapy.
In this report, we present the synthesis and evaluation of
64Cu-L1 (H3L1a ) triphenyl(4-((4,7,10-tris(carboxymethyl)-
1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)phosphon-
ium), 64Cu-L2 (H3L2 ) triphenyl(4-((4,7,10-tris(carboxymeth-
yl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)arson-
ium), and 64Cu-L3 (H3L3 ) tris(4-methoxyphenyl)(4-((4,7,10-
tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)-
benzyl)phosphonium) as new PET radiotracers for tumor
imaging in athymic nude mice bearing subcutaneous U87MG
human glioma xenografts. The U87MG glioma cell line was
chosen because it has very limited multidrug resistance gene
(particularly MDR1) expression.42,43 We are particularly inter-
ested in the impact of heteroatoms (Figure 1: P versus As) and
methoxy groups of the TPP/TPA cations on biodistribution
characteristics of the corresponding 64Cu radiotracers. The main
objective of this study is to demonstrate the intrinsic capability
of the 64Cu-labeled TPP/TPA cations to localize in tumors. This
study represents the first to use the radiometal-labeled TPP/
TPA cations for tumor imaging.
a Abbreviation: H3L1, triphenyl(4-((4,7,10-tris(carboxymethyl)-1,4,7,-
10-tetraazacyclododecan-1-yl )methyl)benzyl)phosphonium; H3L2, triphen-
yl(4-((4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)-
methyl)benzyl)arsonium; H3L3, tris(4-methoxyphenyl)(4-((4,7,10-tris(car-
boxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)phospho-
nium.

Tumor-SelectiVe Imaging Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5059
Chart 1. Synthesis of DO3A-Conjugated TPP and TPA Cations
The white solid was filtered, washed with toluene (5 mL) and
diethyl ether (20 mL), and then dried under vacuum to give the
intermediate product: (4-bromomethylbenzyl)triphenylphosphonium
bromide. The yield was 0.93 g (88.4%). 1H NMR (CDCl3, chemical
shift δ in ppm relative to TMS): 4.35 (s, 2H, CH2Br), 5.48 (d, 2H,
PCH2, JPH ) 15 Hz), 7.09 (m, 4H, C6H4), 7.74-7.55 (m, 15H
C6H5). ESI-MS: m/z ) 446.3 for [M + H]+ (calcd 446 for [C26H23-
PBr]+).
Experimental Section
Materials and Instruments. Chemicals were purchased from
Sigma/Aldrich (St. Louis, MO) without further purification. DO3A-
(OBu-t)3 (1,4,7,10-tetraazacyclododecane-4,7,10-tris(tert-butyl-
acetate)) was purchased from Macrocyclics Inc. (Dallas, TX). The
NMR (1H and 13C) data were obtained using a Bruker DRX 300
MHz FT NMR spectrometer. The chemical shifts as δ are reported
in ppm relative to TMS. Infrared (IR) spectra were recorded on a
Perkin-Elmer FT-IR spectrometer. Mass spectral data were collected
using positive mode on a Finnigan LCQ classic mass spectrometer,
School of Pharmacy, Purdue University. Elemental analysis was
performed by Dr. H. Daniel Lee using a Perkin-Elmer Series III
analyzer, Department of Chemistry, Purdue University. Cardiolite
vials were obtained as a gift from Bristol Myers Squibb Medical
Imaging (North Billerica, MA) and were reconstituted according
to the manufacturer’s insert. 64Cu was produced using a CS-15
biomedical cyclotron at Washington University School of Medicine
by the 64Ni(p,n)64Cu nuclear reaction.
HPLC Methods. Method 1 used a LabAlliance semiprep HPLC
system equipped with a UV/vis detector (λ ) 254 nm) and Zorbax
C18 semiprep column (9.4 mm × 250 mm, 100 Å pore size) for
peptide purification. The flow rate was 2.5 mL/min. The mobile
phase was isocratic with 90% solvent A (0.1% acetic acid in water)
and 10% solvent B (0.1% acetic acid in acetonitrile) at 0-5 min,
followed by a gradient mobile phase going from 90% solvent A
and 10% solvent B at 5 min to 60% solvent A and 40% solvent B
at 20 min. Method 2 used the LabAlliance HPLC system equipped
with a UV/vis detector (λ ) 254 nm), a â-ram IN-US detector,
and Vydac protein and peptide C18 column (4.6 mm × 250 mm,
300 Å pore size). The flow rate was 1 mL/min with the mobile
phase being isocratic with 90% solvent A (10 mM ammonium
acetate) and solvent B (acetonitrile) at 0-20 min, followed by a
gradient mobile phase going from 40% B at 20 min.
(4-(Bromomethyl)benzyl)triphenylarsonium Bromide. Tri-
phenylarsine (0.90 g, 3.0 mmol) and R,R′-dibromomethyl-p-xylene
(0.87 g, 3.3 mmol) were dissolved in nitromethane (8 mL). The
resulting solution was refluxed for 6 h. The mixture was cooled to
room temperature and kept still overnight to afford a white solid.
After filtration, the white solid was washed with a small amount
of acetone and dried under vacuum overnight. The yield was 0.41
1
g (24%). H NMR (CDCl3): 7.69-7.53 (m, 19H), 7.07 (s, 2H),
5.48 (s, 2H). ESI-MS: m/z ) 490.3 for [M + H]+ (calcd 490 for
[C26H23AsBr]+).
(4-(Bromomethyl)benzyl)tris(4-methoxyphenyl)phosphoni-
um Bromide. Tris(4-methoxyphenyl)phosphine (0.74 g, 2.0 mmol)
was dissolved in toluene (6 mL). The resulting solution was added
dropwise to a toluene solution (5 mL) containing R,R′-dibromo-
p-xylene (0.53 g, 2.0 mmol) at 100 °C. After the reaction mixture
was refluxed at 100 °C for 18 h, the white solid was filtered, washed
with toluene (5 mL) and diethyl ether (20 mL), and then dried
under vacuum to give the product: (4-(bromomethyl)benzyl)tris-
(4-methoxyphenyl)phosphonium bromide. The yield was 1.02 g
1
(84.2%). H NMR (CDCl3, chemical shift δ in ppm relative to
TMS): 3.86 (s, 9H, 4-OCH3); 4.36 (s, 2H, CH2Br); 5.08 (d, 2H,
PCH2, JPH ) 15 Hz); 6.86-7.63 (m, 16H aromatic). ESI-MS: m/z
) 536.4 for [M+ + H] (calcd 536 for [C29H29O3PBr]+). The
intermediate product was used for the next step reaction without
further purification.
Triphenyl(4-((4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacy-
clododecan-1 -yl)methyl)benzyl)phosphonium (H3L1). Triethyl-
amine (0.5 mmol) was added to a solution of (4-bromomethyl-
benzyl)triphenylphosphonium bromide (52.6 mg, 0.1 mmol) and
DO3A(OBu-t)3 (52.0 mg, 0.1 mmol) in 3 mL of anhydrous DMF.
The mixture was stirred overnight at room temperature. Upon
(4-(Bromomethyl)benzyl)triphenylphosphonium Bromide.
Triphenylphosphine (0.54 g, 2 mmol) was dissolved in toluene (6
mL), and the resulting solution was added dropwise to a toluene
solution (5 mL) containing R,R′-dibromomethyl-p-xylene (0.53 g,
2 mmol) at 100 °C. The mixture was refluxed at 100 °C for 18 h.

5060 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Wang et al.
Table 1. HPLC Retention Time and log P Values for 64Cu-L1,
64Cu-L2, 64Cu-L3, and 99mTc-Sestamibi
retention
compound
RCP (%)
time (min)
log P value
99mTc-Sestamibi
64Cu-L1
>98
>95
>95
>90
16.5
17.5
17.7
20.9
1.09 ( 0.15
-2.67 ( 0.21
-2.65 ( 0.02
-2.02 ( 0.01
64Cu-L2
64Cu-L3
Table 2. Solution Stability Data for the HPLC Purified 64Cu-L1 in
Saline and in the Presence of EDTA mg/mL in 25 mM Phosphate
Buffer, pH ) 7.5)
time after HPLC
purification, h
RCP
(in saline)
RCP
(in EDTA solution)
0.5
1.0
2.0
4.0
6.0
98
98
98
98
98
98
98
98
98
98
subjected to HPLC purification. Fractions at ∼16.5 min were
collected and lyophilized to give a white powder. The yield was
14 mg (52%). The HPLC retention time was ∼16.3 min (Method
1) with the purity >98%. 1H NMR (D2O, chemical shift δ in ppm
relative to TMS): 2.80-3.45 (m, 22H), 3.67 (s, 2H, CH2), 4.63 (s,
2H, AsCH2), 6.94 (d, 2H, C6H4), 7.23 (d, 2H, C6H4), 7.37-7.73
(m, 15H, C6H5). ESI-MS: m/z ) 755.8 for [M + H]+ (calcd 756
for [C40H48N4O6As]+). Anal. Calcd for C42H51AsN4O8‚2.5H2O: C,
58.67; H, 6.56; N, 6.52. Found: C, 58.75; H, 6.31; N, 6.68.
Tris(4-methoxyphenyl)(4-((4,7,10-tris(carboxymethyl)-
1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)phosphon-
ium (H3L3). To a solution of (4-(bromomethyl)benzyl)tris(4-
methoxyphenyl)phosphonium bromide (61.9 mg, 0.1 mmol) and
DO3A(OBu-t)3 (52.0 mg, 0.1 mmol) in 3 mL of anhydrous DMF
was added triethylamine (0.07 mL, 0.5 mmol). The reaction mixture
was stirred overnight at room temperature. Upon complete removal
of DMF and excess triethylamine, the residue was dissolved in 2.2
mL of concentrated HCl (37%). After the mixture was stirred at
room temperature for 10-20 min, excess HCl was removed under
the reduced pressure. The residue was dissolved in 5 mL of 50%
DMF/water mixture, and the product was subjected to semi-prep
HPLC purification. Fractions at ∼19 min were collected. The
collected fractions were combined and lyophilized to give a white
powder. The yield was 39 mg (48.7%). The HPLC retention time
was ∼19.2 min (Method 1) with the purity >98%. 1H NMR (D2O,
chemical shift δ in ppm relative to TMS): 2.64-3.38 (m, 22H);
3.63 (s, 2H, CH2); 3.73 (s, 9H, CH3); 4.35 (d, 2H, PCH2, JPH ) 13
Hz), 6.67-7.37 (m, 16H, C6H4). ESI-MS: m/z ) 801.4 for
[M + H]+ (calcd 801 for [C43H54N4O9P]+). Anal. Calcd for
C45H57N4O11P: C, 62.78; H, 6.67; N, 6.51. Found: C, 62.68; H,
7.05; N, 6.60.
Cu-L1. To the solution containing H3L1 (40 mg, 0.56 mmol)
and Cu(OAc)2·2H2O (12.0 mg, 0.60 mmol) in H2O (2 mL) was
added 0.2 mL of NH4OAc buffer (0.5 M, pH ) 6.0). The resulting
solution was heated at 100 °C for 45 min. After filtration, diethyl
ether (15 mL) was added to the filtrate above. The precipitate was
separated and dried under vacuum overnight before being submitted
for elemental analysis. The yield was 26.5 mg (∼61.2%). A sample
was analyzed by HPLC. The HPLC retention time was 17.2 min
with the purity >95%. IR (cm-1, KBr pellet): 1633.5 (s, υCdO),
1721.3 (s, υCdO) and 3432.1 (bs, υO-H). ESI-MS (positive mode):
m/z ) 773.3 for [M + H]+ (calcd 773 for [C40H46CuN4O6P]+).
Anal. Calcd for C40H46CuN4O6‚2H2O: C, 59.33; H, 6.18; N, 6.92.
Found: C, 59.11; H, 6.13; N, 6.63.
Figure 2. Representative HPLC chromatograms of 64Cu-L1, Cu-L1,
and Mn-L1 under identical chromatographic conditions. Their almost
identical HPLC retention times suggest that the same metal complex
was prepared at the tracer (64Cu) and macroscopic (Cu) levels, and
that Cu-L1 most likely exists as its zwitterion form as observed for
Mn-L1.
removal of DMF and excess triethylamine under reduced pressure,
the residue was dissolved in 2 mL of concentrated HCl. After being
stirred at room temperature for 10-20 min, volatiles were removed
under the reduced pressure. The residue was dissolved in 4 mL of
50% DMF/water mixture, and the product was separated by HPLC.
Fractions at ∼16.5 min were collected and lyophilized to give a
white powder. The yield was 45 mg (63%). The HPLC retention
time was ∼16.5 min (Method 1) with purity >98%. 1H NMR (D2O,
chemical shift δ in ppm relative to TMS): 2.80-3.7 (m, 22H),
4.63 (s, 2H, CH2), 4.68 (d, 2H, PCH2, JPH ) 13 Hz), 6.86-7.75
(m, 19H, aromatic). ES-MS: m/z ) 711.3 for [M + H]+ (calcd
711 for [C40H48N4O6P]+). Anal. Calcd for C42H51N4O8P‚4H2O: C,
59.85; H, 7.06; N, 6.65. Found: C, 59.53; H, 6.84; N, 6.79.
Triphenyl(4-((4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacy-
clododecan-1 -yl)methyl)benzyl)arsonium (H3L2). To a solution
of DO3A(OBu-t)3 (15.6 mg, 0.03 mmol) and (4-bromomethylben-
zyl)triphenylarsonium bromide (52.6 mg, 0.1 mmol) in 3 mL of
anhydrous DMF was added triethylamine (0.02 mL, 0.15 mmol).
The mixture was stirred overnight at room temperature. Upon
removal of volatiles under reduced pressure, the residue was
dissolved in 2 mL of concentrated HCl to remove the tert-butyl
ester groups. After being stirred at room temperature for 10-20
min, volatiles removed under reduced pressure. The residue was
dissolved in 4 mL of 50% DMF/water mixture, and product was
64Cu-Labeling. Caution! 64Cu has a half-life of 12.7 h with a
â+ emission (Emax ) 0.655 MeV, 19.3%), a â- emission (Emax
) 0.573 MeV, 39.6%) and an Auger-electron emission (41%). All
manipulations must be performed in a laboratory approVed for
handling of radioisotopes. Radiation safety procedures must be
followed at all times to preVent contamination. To a 5 mL vial

Tumor-SelectiVe Imaging Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5061
Figure 3. Cellular and mitochondrial uptake (left) and efflux (right) kinetics of 64Cu-L1 in both U87MG human glioma cells ([) and tumor cell
mitochondria (9). The radiotracer uptake is expressed as the percentage of total added radioactivity (%ID). The radiotracer efflux is represented by
the percentage of the intracellular radioactivity at t ) 0 before being recultured.
were added 0.5 mL of 0.1 M NaOAc buffer (pH ) 6.9) containing
50 µg of the DO3A conjugate and 0.12 mL of 64CuCl2 solution
(1.0-2.0 mCi) in 0.05 N HCl. The final pH in the reaction mixture
was 5.0-5.5. The mixture was then heated at 100 °C for 30 min.
After being cooled to room temperature, a sample of resulting
solution was analyzed by radio-HPLC (Method 2). The radiochemi-
cal purity (RCP) was >90% for all three radiotracers before
purification. The specific activity was 30-50 Ci/mmole for all three
64Cu radiotracers. The HPLC retention time was ∼17.5 min for
64Cu-L1 and 64Cu-L2, and 20.4 min for 64Cu-L3. The HPLC
concordance experiment was performed by sequential injection of
HPLC-purified 64Cu-L1 and Cu-L1 under identical chromatographic
conditions.
of 10 µL of Mitochondria Isolation Reagent B, the mixture was
vortexed at the maximum speed for 5 s and was then incubated on
ice for another 5 min. Upon addition of Mitochondria Isolation
Reagent C (800 µL), the mixture was centrifuged at 700g for 10
min at 4 °C. The supernatant was transferred to a new tube and
was then centrifuged at 12 000g for 15 min at 4 °C. The supernatant
was discarded. To the pellet was added 500 µL of Mitochondria
Isolation Reagent C. After centrifugation at 12 000g for 5 min
at 4 °C, the mitochondrial pellet was put on ice for the uptake and
efflux experiments.
In Vitro Cellular and Mitochondrial Uptake. U87MG cells
were seeded into 12-well plates (8 × 105 cells/well) overnight.
Adherent cells were washed with PBS buffer, followed by addition
of 64Cu-L1 (10 µCi in 1 mL of DMEM medium per well). After
incubation at 37 °C for the specified time (15, 30, 60, 120, and
180 min), cell monolayers were washed with PBS (3×) and then
lysed with NaOH-SDS (0.2 N NaOH, 1% SDS). The lysate was
counted by a γ-counter (Perkin-Elmer Wizard - 1480). The same
procedure was used to carry out the mitochondrial uptake experi-
ment with the mitochondria isolated from the same number of
glioma cells.
Time Dependent Efflux from Cells and Mitochondria. U87MG
cells were seeded into 12-well plates (8 × 105 cells/well) overnight.
Adherent cells were washed with PBS, followed by addition of
64Cu-L1 (10 µCi in 1 mL of DMEM medium per well). After
incubation at 37 °C for 180 min, the medium was removed. The
glioma cells were washed with PBS (3×), isolated, and recultured
with the radioactivity free DMEM medium for the specified time
(0, 15, 30, 60, 120, and 180 min). Cell monolayers were washed
with PBS (3×) and were then lysed with NaOH-SDS (0.2 N
NaOH, 1% SDS). The lysate was counted by a γ-counter (Perkin-
Elmer Wizard - 1480). For the mitochondrial efflux experiment,
the mitochondria isolated from the same number of glioma cells
were incubated with 64Cu-L1 in the DMEM medium for 180 min
and were then incubated in nonradioactive DMEM medium for the
specified time (0, 15, 30, 60, 120, and 180 min). The mitochondrion-
bound activity was then isolated from centrifugation after PBS wash
(3×) and counted.
Dose Preparation. All three 64Cu radiotracers were purified by
HPLC before being used for biodistribution studies. Volatiles in
the HPLC mobile phase were completely removed. The residue
was dissolved in saline to give a concentration of ∼10 µCi/mL.
The resulting solution was filtered with a 0.20 µm filter unit to
eliminate any particles. Each animal was administered with ∼0.10
mL of the dose solution. For the imaging study, 64Cu-L1 was
prepared and the resulting mixture was used without purification.
The reaction mixture was diluted to ∼5 mCi/mL with saline. The
injected dose for each tumor-bearing mouse was about 250 µCi of
64Cu-L1.
Solution Stability. For solution stability in saline, 64Cu-L1 was
prepared and purified by HPLC. Volatiles in the HPLC mobile
phase were removed under vacuum. The residue was dissolved in
saline to ∼1 mCi/mL. Samples were analyzed by HPLC (Method
2) at 0, 1, 2, 4, and 8 h post purification. In EDTA challenge
experiment, the radiotracer was dissolved in 25 mM phosphate
buffer (pH ) 7.4) containing EDTA (1 mg/mL) to give a
concentration of 1 mCi/mL. Samples were analyzed by radio-HPLC
(Method 2) at 0, 1, 2, 4, and 8 h post purification.
Partition Coefficient. All three radiotracers were purified by
HPLC. After complete removal of volatiles in the mobile phase,
the residue was dissolved in a mixture of 3 mL of saline and 3 mL
of n-octanol in a round-bottom flask. The mixture was vigorously
stirred for 20 min at room temperature and was then transferred to
an Eppendorf microcentrifuge tube. The tube was centrifuged at
12 500 rpm for 5 min. Samples in triplets from n-octanol and
aqueous layers were obtained and were counted in a gamma-counter
(Perkin-Elmer Wizard - 1480). For comparison purposes, the log
P of 99mTc-Sestamibi was also determined using the same procedure.
The log P value was reported as an average of the data obtained in
three independent measurements.
Animal Model. Biodistribution studies were performed using
the athymic nude mice bearing U87MG human glioma xenografts
in compliance the NIH animal experiment guidelines (Principles
of Laboratory Animal Care, NIH Publication No. 86-23, revised
1985). The animal protocol has been approved by Purdue University
Animal Care and Use Committee (PACUC). Female athymic nu/
nu mice were purchased from Harlan (Charles River, MA) at 4-5
weeks of age. The mice were orthotopically implanted with 5 ×
106 the U87MG human glioma cells into the mammary fat pad.
Tumor cells were grown at 37 °C in Minimal Essential Medium
(Alpha Modification) containing 3.7 g of sodium bicarbonate/L,
10% fetal bovine serum v/v, in a humidified atmosphere of 5%
carbon dioxide. Four weeks after inoculation, the tumor size was
Mitochondria Isolation. U87MG cells were collected and
mitochondria were isolated by Mitochondria Isolation Kit for
Cultured Cells (Pierce Biotechnology, Rockford, IL). Briefly, 800
µL of Mitochondria Isolation Reagent A was added to a pellet of
2 × 107 U87MG cells. The mixture was vortexed at the medium
speed for 5 s and was then incubated on ice for 2 min. After addition

5062 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Wang et al.
Table 3. Biodistribution Data of 64Cu-L1 in Athymic Nude Mice Bearing the U87MG Human Glioma Xenograftsa
uptake (% ID/g)
organ
5 min
30 min
60 min
120 min
blood
brain
heart
intestine
kidney
liver
lungs
muscle
spleen
4.54 ( 0.34
0.18 ( 0.06
2.04 ( 0.35
13.05 ( 1.52
21.32 ( 5.44
31.40 ( 4.52
4.28 ( 0.58
1.42 ( 0.39
1.54 ( 0.34
3.22 ( 0.12
0.71 ( 0.04
21.17 ( 10.69
0.79 ( 0.08
2.03 ( 0.18
0.91 ( 0.18
0.13 ( 0.17
0.59 ( 0.16
8.83 ( 9.72
3.19 ( 0.35
18.73 ( 3.51
1.72 ( 0.48
0.10 ( 0.08
1.06 ( 0.48
1.61 ( 0.55
0.71 ( 0.63
34.66 ( 24.58
0.97 ( 0.42
7.36 ( 10.10
1.37 ( 0.67
0.04 ( 0.03
0.84 ( 0.22
5.54 ( 1.12
3.29 ( 0.73
15.48 ( 3.53
1.99 ( 0.64
0.37( 0.37
0.81 ( 0.19
2.09 ( 0.82
2.08 ( 1.86
119 ( 45
0.55 ( 0.09
0.05 ( 0.02
0.59 ( 0.02
2.60 ( 0.24
2.20 ( 0.42
11.09 ( 2.61
1.56 ( 0.05
0.01 ( 0.00
0.55 ( 0.17
2.51 ( 0.38
3.72 ( 1.47
39.72 ( 17.99
1.66 ( 0.25
226 ( 50
tumor
tumor/blood ratio
tumor/brain ratio
tumor/lung
tumor/muscle ratio
1.58 ( 0.14
137 ( 183
a Each data point represents an average of biodistribution data in four animals.
Table 4. Biodistribution Data of 64Cu-L2 in Athymic Nude Mice Bearing the U87MG Human Glioma Xenograftsa
uptake (% ID/g)
organ
5 min
30 min
60 min
120 min
blood
brain
heart
intestine
kidney
liver
lungs
muscle
spleen
4.00 ( 0.43
0.19 ( 0.14
1.87 ( 0.66
13.80 ( 7.28
14.17 ( 0.77
26.29 ( 0.45
3.56 ( 1.05
1.47 ( 0.82
1.30 ( 0.14
2.52 ( 0.30
0.63 ( 0.01
18.47 ( 9.84
0.75 ( 0.23
2.31 ( 1.62
0.67 ( 0.08
0.04 ( 0.00
0.28 ( 0.13
7.56 ( 3.61
2.66 ( 0.32
14.83 ( 2.22
0.96 ( 0.07
0.03 ( 0.02
0.46 ( 0.17
1.25 ( 0.22
1.87 ( 0.35
29.98 ( 7.02
1.30 ( 0.14
27.30 ( 3.09
0.66 ( 0.27
0.02 ( 0.01
0.37 ( 0.13
5.17 ( 3.82
2.48 ( 1.08
12.58 ( 1.55
1.01 ( 0.53
0.07 ( 0.11
0.50 ( 0.36
1.66 ( 0.64
2.56 ( 0.58
114 ( 52.49
1.72 ( 0.29
251 ( 16.57
0.57 ( 0.13
0.05 ( 0.03
0.51 ( 0.22
2.12 ( 0.88
2.20 ( 0.49
10.85 ( 2.93
1.48 ( 0.25
0.01 ( 0.00
0.46 ( 0.13
2.30 ( 1.03
4.16 ( 2.05
69.01 ( 54.86
1.62 ( 0.36
170 ( 15.70
tumor
tumor/blood ratio
tumor/brain ratio
tumor/lung
tumor/muscle ratio
a Each data point represents an average of biodistribution data in four animals.
Table 5. Biodistribution Data of 64Cu-L3 in Athymic Nude Mice Bearing the U87MG Human Glioma Xenograftsa
uptake (% ID/g)
organ
5 min
30 min
60 min
120 min
blood
brain
heart
intestine
kidney
liver
lungs
muscle
spleen
4.76 ( 1.54
0.13 ( 0.06
2.25 ( 0.75
7.29 ( 2.15
18.83 ( 7.53
12.61 ( 1.34
5.04 ( 0.42
1.48 ( 0.25
1.83 ( 0.25
2.79 ( 0.66
0.60 ( 0.09
25.50 ( 17.92
0.55 ( 0.12
1.88 ( 0.37
0.48 ( 0.20
0.01 ( 0.00
0.42 ( 0.04
5.34 ( 3.55
3.60 ( 0.71
9.69 ( 1.00
1.02 ( 0.30
0.11 ( 0.15
0.36 ( 0.11
1.47 ( 0.81
3.03 ( 0.83
147 ( 81.32
1.39 ( 0.37
46.74 ( 43.14
0.41 ( 0.21
0.01 ( 0.00
0.32 ( 0.20
1.97 ( 0.97
2.63 ( 0.51
6.21 ( 1.45
0.79 ( 0.12
0.10 ( 0.16
0.52 ( 0.09
1.16 ( 0.20
3.56 ( 2.11
110 ( 10.18
1.49 ( 0.24
71.35 ( 57.57
0.24 ( 0.08
0.01 ( 0.00
0.34 ( 0.10
1.50 ( 0.15
1.99 ( 0.23
4.71 ( 0.06
1.02 ( 0.21
0.01 ( 0.00
0.25 ( 0.19
1.85 ( 0.33
8.27 ( 2.84
185 ( 32.94
1.82 ( 0.21
184 ( 32.94
tumor
tumor/blood ratio
tumor/brain ratio
tumor/lungs
tumor/muscle ratio
a Each data point represents an average of biodistribution data in four animals.
in the range of 0.3-0.6 g, and animals were used for biodistribution
and imaging studies.
was also performed on 99mTc-Sestamibi, a well-known radiophar-
maceutical approved for both myocardial perfusion and tumor
imaging, using the same protocol. The biodistribution data and
target-to-background ratios are reported as an average plus the
standard variation. Comparison between two different radiotracers
was also made using the one-way ANOVA test. The level of
significance was set at p < 0.05.
Biodistribution Protocol. Sixteen tumor-bearing mice (20-25
g) were randomly divided into four groups, each of which had four
animals. The HPLC-purified 64Cu radiotracer (∼1 µCi dissolved
in 0.1 mL of saline) was administered each animal via tail vein.
Four animals were sacrificed by exsanguinations and opening of
thoracic cavity at 5, 30, 60, and 120 min postinjection (p.i.). Organs
were excised, washed with saline, dried with absorbent tissue,
weighed, and counted on a γ-counter (Perkin-Elmer Wizard -
1480). Organs of interest included tumor, blood, brain, spleen, lungs,
liver, kidneys, muscle, and intestine. The organ uptake was
calculated as a percentage of the injected dose per gram of wet
tissue (%ID/g). For comparison purpose, the biodistribution study
Metabolism. The metabolic stability of 64Cu-L1 was evaluated
in normal athymic nude mice. Two animals were anesthetized with
IP injection of ketamine (40-100 mg/kg) and xylazine (2-5 mg/
kg). Each mouse was administered with 64Cu-L1 at a dose of 100
µCi dissolved in 0.2 mL of saline via tail vein. Urine samples were
collected at 30 and 120 min p.i. by manual void and were mixed
with an equal volume of acetonitrile. The mixture was centrifuged

Tumor-SelectiVe Imaging Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5063
Table 6. Biodistribution Data of 99mTc-Sestamibi in Athymic Nude Mice Bearing the U87MG Human Glioma Xenograftsa
uptake (% ID/g)
organ
blood
brain
heart
intestine
kidney
liver
lungs
muscle
spleen
tumor
tumor/blood
tumor/brain
tumor/lungs
tumor/muscle
5 min
30 min
60 min
120 min
0.87 ( 0.48
0.22 ( 0.05
19.22 ( 7.62
9.80 ( 3.01
37.45 ( 13.15
14.37 ( 1.01
6.20 ( 2.59
4.84 ( 1.22
3.21 ( 1.39
3.06 ( 1.15
6.28 ( 4.44
16.69 ( 8.72
0.69 ( 0.07
0.63 ( 0.36
0.23 ( 0.25
0.14 ( 0.06
17.24 ( 3.84
10.26 ( 2.70
28.49 ( 6.00
8.59 ( 1.17
2.57 ( 0.70
4.72 ( 1.22
1.94 ( 0.63
2.11 ( 0.47
12.88 ( 2.49
16.87 ( 7.95
0.95 ( 0.34
0.46 ( 0.05
0.07 ( 0.02
0.11 ( 0.02
17.82 ( 3.42
13.31 ( 3.89
24.28 ( 4.91
7.29 ( 1.27
1.81 ( 0.34
4.83 ( 1.46
1.51 ( 0.23
1.47 ( 0.54
18.55 ( 5.72
12.53 ( 5.02
0.89 ( 0.15
0.26 ( 0.10
0.07 ( 0.03
0.07 ( 0.01
19.19 ( 5.32
6.19 ( 2.94
13.50 ( 4.48
4.70 ( 2.31
1.55 ( 0.54
5.45 ( 1.24
1.77 ( 0.90
1.55 ( 0.36
23.50 ( 9.83
22.62 ( 10.34
1.22 ( 0.32
0.24 ( 0.06
a Each data point represents an average of biodistribution data in three animals.
of the 64Cu radiotracers. DO3A is the bifunctional chelator of
our choice because it is able to form stable radiometal chelates
with many radionuclides, including 64Cu and 68Ga for PET,
111In for SPECT, and 90Y and 177Lu for radiotherapy. H3L1,
H3L2, and H3L3 were prepared by reacting DO3A(OBu-t)3 with
the corresponding bromide (Chart 1). Hydrolysis of the tert-
butyl ester with concentrated HCl gave the expected product,
which was then subjected to HPLC purification. All three DO3A
conjugates were isolated as the monoacetate salt and have been
characterized by NMR, ESI-MS, and elemental analysis. The
purity of the HPLC-purified DO3A conjugate must be >95%
before being used for radiolabeling.
Synthesis and Characterization of Cu-L1. Cu-L1 is used
as a model compound for structural characterization of 64Cu-
labeled TPP/TPA cations. Cu-L1 was prepared by reacting H3-
L1 with 1 equiv of Cu(II) acetate in 0.5 M ammonium acetate
buffer (pH ) 6), was isolated as its Zwitterion salt, and has
been characterized by IR, ESI-MS, and elemental analysis. The
IR spectrum of Cu-L1 shows a broad band at 3432.1 cm-1 from
the crystallization water, a strong band at 1721.3 cm-1 from
the uncoordinated carboxylate group, and a strong band at
1637.1 cm-1 from two coordinated carboxylate groups. The ESI-
MS spectrum of Cu-L1 displays two molecular ions: m/z )
773.3 for [M + H]+ and m/z ) 795.2 for [M + Na]+. Many
attempts were made to grow crystals for structural determination
of Cu-L1 by X-ray crystallography, but they were unsuccessful.
On the basis of its elemental analysis and IR spectroscopic data,
we believe that Cu-L1 might exist as its zwitterion form in the
solid state as observed for Mn-L1, which has been structurally
characterized by X-ray crystallography in our previous study.44
Radiochemistry. All three 64Cu radiotracers were prepared
by reacting 64CuCl2 with the DO3A conjugate (H3L1, H3L2, or
H3L3) in 0.1 M NaOAc buffer (pH ) 5.0-5.5) at 100 °C for
30 min. Their radiochemical purity (RCP) was >90% with the
specific activity in the range of 30-50 Ci/mmol. Their log P
values (Table 1) were determined to be -2.67 ( 0.21, -2.65
( 0.02, and -2.02 ( 0.01, respectively, in the mixture of
n-octanol and 25 mM phosphate buffer (pH 7.4). 64Cu-L1 is
stable for >6 h after purification and remains intact for >6 h
without any decomposition (Table 2) in the presence of EDTA
(1 mg/mL in 25 mM phosphate buffer, pH ) 7.4), which is
completely consistent with the metabolic stability of 64Cu-L1
in urine samples of normal mice under anesthesia. We performed
the HPLC concordance experiment using HPLC-purified 64Cu-
L1 and Cu-L1. Figure 2 shows radio-HPLC chromatograms of
64Cu-L1 and Cu-L1 under the same chromatographic conditions.
Since 64Cu-L1 and Cu-L1 have almost identical HPLC retention
at 8000 rpm. The supernatant was collected and filtered through a
0.20 µm Millex-LG syringe-driven filter unit to remove the
precipitate and large proteins. The filtrate was analyzed by radio-
HPLC. Feces samples were collected at ∼120 min p.i. and were
suspended in a mixture of 50% acetonitrile aqueous solution. The
mixture was vortexed for 5-10 min. After centrifuging at 8000
rpm for 5 min, the supernatant was collected and passed through a
0.20 µm Millex-LG syringe driven filter unit. The filtrate was then
analyzed by radio-HPLC (Method 2).
Calibration of microPET. Scanner activity calibration was
performed to map between microPET image units and units of
activity concentration. A preweighed 50-mL centrifuge tube filled
with solution containing 64CuCl2 (∼9.3 MBq as determined by the
dose calibrator) was used to simulate the whole body of the mouse.
This tube was weighed, centered in the scanner aperture, and imaged
for a 30 min static image. From the sample weight and assuming
a density of 1 g/mL, the activity concentration in the bottle was
calculated in units of µCi/mL. Eight planes were acquired in the
coronal section. A rectangular region of interest (ROI) (counts/
pixel/ s) was drawn on the middle of eight coronal planes. Using
these data, a calibration factor was obtained by dividing the known
radioactivity in the cylinder (µCi/mL) by the image ROI. This
calibration factor was determined periodically and did not vary
significantly with time.
MicroPET Imaging Studies. PET imaging of tumor-bearing
mice was performed using a microPET R4 rodent model scanner
(Concorde Microsystems, Knoxville, TN). The U87MG tumor-
bearing mice (n ) 3) were imaged in the prone position in the
microPET scanner. The tumor-bearing mice were injected with
∼250 µCi of 64Cu-L1 via the tail vein and then anesthetized with
2% isoflurane and placed near the center of the FOV where the
highest resolution and sensitivity are obtained. Multiple static scans
were obtained at 0.05, 0.2, 0.6, 2, and 19 h p.i. The images were
reconstructed by a two-dimensional ordered subsets expectation
maximum (OSEM) algorithm. No correction was necessary for
attenuation or scatter. At each microPET scan, ROIs were drawn
over each tumor and major organs on decay-corrected whole-body
coronal images. The average radioactivity concentration (accumula-
tion) within the tumor or an organ was obtained from mean pixel
values within the multiple ROI volume, which were converted to
counts/mL/ min by using the calibration constant C. Assuming a
tissue density of 1 g/mL, the ROIs were then converted to counts/
g/min and then divided by the total administered activity to obtain
an imaging ROI-derived percentage administered activity per gram
of tissue (%ID/g).
Results
Synthesis of DO3A Conjugates. We are particularly inter-
ested H3L1, H3L2, and H3L3 since they allow us to assess the
possible impact of heteroatoms (P versus As) and methoxy
groups of the TPP/TPA cations on biodistribution characteristics

5064 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Wang et al.
Figure 4. Direct comparison of organ uptake (tumor, blood, heart, liver, lungs, and muscle) between 64Cu-L1, 64Cu-L2, 64Cu-L3, and 99mTc-
Sestamibi in athymic nude mice bearing U87MG human glioma xenografts. The most striking difference between 64Cu radiotracers and 99mTc-
Sestamibi is uptake in the heart and muscle.
times, it is reasonable to conclude that the same metal complex
was prepared at both tracer (64Cu) and macroscopic (Cu) levels.
For comparison purposes, we also included the HPLC chro-
matogram of Mn-L1 under the same chromatographic condi-
tions.44 It is quite clear that Cu-L1 and Mn-L1 have almost
identical HPLC retention times (∼17.5 min).
designed to demonstrate that 64Cu-labeled TPP cations are
indeed able to across plasma and mitochondrial membranes and
localize in mitochondria of tumor cells. Figure 3 illustrates the
cellular and mitochondrial uptake (left) and efflux (right) kinetics
of 64Cu-L1. The radiotracer uptake is expressed as the percentage
of total added radioactivity (%ID). The radiotracer efflux is
expressed by the percentage of intracellular radioactivity at t )
0 before being recultured. The uptake of 64Cu-L1 in the tumor
cell mitochondria (1.39 ( 0.43 and 0.97 ( 0.27 %ID/million
In vitro Cellular and Mitochondrial Uptake Kinetics. The
cellular and mitochondrial uptake kinetics of 64Cu-L1 was
examined by using U87MG glioma cells. These studies were

Tumor-SelectiVe Imaging Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5065
64Cu-L2, and 64Cu-L3 increased steadily from ∼1 at 5 min p.i.
to ∼5 at 120 min p.i. The tumor/heart ratio of 64Cu-L3 was
about 40-fold better than that of 99mTc-Sestamibi at 120 min
p.i. (Figure 5). The muscle uptake for 64Cu-L1, 64Cu-L2, and
64Cu-L3 was almost undetectable at >30 min p.i. In contrast,
99mTc-Sestamibi has a very high muscle uptake (4.84 ( 1.22 at
5 min p.i. and 5.45 ( 1.24 %ID/g at 120 min p.i.). In addition,
the lung uptake of 99mTc-Sestamibi is significantly higher (p <
0.05) than that of 64Cu-L1, 64Cu-L2, and 64Cu-L3 at 5-60 min
p.i., but this difference disappeared at 2 h p.i.
PET Imaging. We performed a MicroPET imaging study
on 64Cu-L1 using athymic nude mice (n ) 3) bearing U87MG
human glioma xenografts. Figure 6 illustrates the coronal
microPET images (A) and the activity accumulation quantifica-
tion (B) in several organs of the tumor-bearing mice adminis-
tered with ∼250 mCi of 64Cu-L1. The glioma tumor was clearly
visualized as early as 30 min p.i. with very high tumor-to-
background contrast. No significant radioactivity accumulation
was detected in the brain. The radioactivity accumulation in
the heart and muscle was also very low, which is consistent
with the results from ex vivo biodistribution studies. After
normalization, the tumor uptake of 64Cu-L1 is 1.94 ( 0.82 %ID/
g, 2.31 ( 0.90 %ID/g, 2.57 ( 0.41 %ID/g, 3.51 ( 1.20 %ID/
g, and 3.11 ( 1.44 %ID/g at 3 min, 12 min, 35 min, 2 and 19
h p.i., respectively. The steady increase in tumor uptake between
0.5 and 2 h p.i. is completely consistent with that observed in
the biodistribution study (Figure 4), and is supported by the
slow “diffusion kinetics” of 64Cu-L1 in both glioma cells and
mitochondria of glioma cells (Figure 3).
Metabolic Properties. We performed metabolism studies on
64Cu-L1 using normal athymic nude mice. Each mouse was
administered with ∼100 µCi of 64Cu-L1. Since 64Cu-L1 is
excreted from both renal and hepatobiliary routes, we tried to
collect urine and feces samples from the mice administered with
64Cu-L1. The radio-HPLC was used for sample analysis to
determine if 64Cu-L1 is able to retain its chemical integrity at
2 h p.i. Attempts to collect the urine sample at 120 min p.i.
were not successful. Figure 7 illustrates the representative radio-
HPLC chromatograms of 64Cu-L1 (HPLC purified) in saline
before injection (A), in the urine sample at 30 min p.i. (B), and
in the feces sample at 120 min p.i. (C). There is no significant
metabolite detectable in the urine sample at 30 min p.i. while
only about 15% of 64Cu-L1 remains intact in the feces sample
at 120 min p.i. The identity of metabolite from the feces sample
is not known. Since the species at ∼4 min is very hydrophilic,
it is reasonable to believe that the TPP moiety of 64Cu-L1 is
probably cleaved in the metabolite.
Figure 5. Direct comparison of the tumor/heart ratios between 64Cu-
L1, 64Cu-L2, 64Cu-L3, and 99mTc-Sestamibi in athymic nude mice
bearing U87MG human glioma xenografts. The high tumor selectivity
is observed for all three 64Cu radiotracers as demonstrated by their
higher tumor/heart and tumor-lung ratios than those of 99mTc-Sestamibi.
cells) is fast and is significantly higher than that in glioma cells
(0.31 ( 0.03 and 0.53 ( 0.05 %ID/million cells) between 15
and 120 min postincubation (Figure 3: left). This radiotracer
uptake difference disappeared at 180 min postincubation. No
peak uptake was achieved in the glioma cells after >3 h
incubation at 37 °C. In contrast, the efflux kinetics (Figure 3:
right) of 64Cu-L1 from glioma cells was significantly slower
than that from tumor cell mitochondria.
Biodistribution Characteristics. Biodistribution character-
istics of 64Cu-L1, 64Cu-L2, and 64Cu-L3 were evaluated using
the athymic nude mice bearing U87MG human glioma xe-
nografts. 99mTc-Sestamibi was evaluated in the same model for
comparison purposes. Biodistribution data of 64Cu-L1, 64Cu-
L2, 64Cu-L3, and 99mTc-Sestamibi are summarized in Tables
3-6. Figure 4 shows the comparison of organ uptake between
64Cu-L1, 64Cu-L2, 64Cu-L3, and 99mTc-Sestamibi in tumor,
blood, heart, liver, lungs, and muscle. Figure 5 illustrates the
comparison of tumor/heart ratios between 64Cu-L1, 64Cu-L2,
64Cu-L3, and 99mTc-Sestamibi.
In general, biodistribution properties of 64Cu-L1 and 64Cu-
L2 are almost identical within the experimental error. Their
initial blood activity level was high (4.54 ( 0.34 and 4.00 (
0.43 %ID/g, respectively, at 5 min p.i.), but both are able to
clear rapidly from blood (0.91 ( 0.18 and 0.67 ( 0.08 %ID/g,
respectively, at 30 min p.i.) via renal and hepatobiliary routes.
The high initial blood activity level might contribute to the high
uptake of 64Cu-L1 and 64Cu-L2 in tumor (3.22 ( 0.12 and 2.52
( 0.30 %ID/g), heart (2.04 ( 0.35 and 1.87 ( 0.66 %ID/g),
kidneys (21.32 ( 5.44 and 14.17 ( 0.77 %ID/g), lungs (4.28
( 0.58 and 3.56 ( 1.05 %ID/g), and muscle (1.42 ( 0.39 and
1.47 ( 0.82 %ID/g) at 5 min p.i. The biodistribution pattern of
64Cu-L3 is similar to that of 64Cu-L1, but its blood clearance is
faster. The liver uptake of 64Cu-L3 is significantly lower (p <
0.01) than that of 64Cu-L1 and 64Cu-L2 at all four time points.
Introduction of three methoxy groups seems to reduce liver
uptake and increase renal excretion (Tables 3-6). The heart
uptake for all three 64Cu radiotracers was very low (<0.6% ID/
g) at >30 p.i.
Discussion
In this study, we reported the synthesis and evaluation of
three 64Cu-labeled TPP cations as PET radiotracers for imaging
tumors with much higher mitochondrial potentials than normal
cells. Cu-L1 was prepared as a model compound for structural
characterization of 64Cu-labeled TPP/TPA cations. On the basis
of its spectroscopic and HPLC data, we believe that Cu-L1 most
likely exists in its zwitterion form as observed for Mn-L1.44
The exact structure of Cu-L1 remains unknown. Since the ionic
radius of Cu(II) (0.67 Å)45 is very close to that of Ga(III) (0.62
Å),45 the coordinated DO3A in Cu-L1 might be the same as
that in [Ga-L1]+,44 with one acetate group remaining deproto-
nated to balance to cationic charge of the TPP moiety.
We carried out the biodistribution on 99mTc-Sestamibi because
it has been widely used to image tumors by SPECT. The most
striking difference between 64Cu-L1, 64Cu-L2, 64Cu-L3, and
99mTc-Sestamibi is their uptake in the heart and muscle (Figure
4). For example, the heart uptake was <0.6% ID/g at >30 min
p.i. for 64Cu-L1, 64Cu-L2, and 64Cu-L3 while the heart uptake
of 99mTc-Sestamibi was 19.22 ( 7.62 at 5 min p.i. and 19.19 (
5.32 %ID/g at 120 min p.i. The tumor/heart ratios of 64Cu-L1,
99mTc-Sestamibi and 99mTc-Tetrofosmin have been widely
used for both myocardial perfusion imaging and diagnosis of
tumors in clinics.46-52 However, their cancer diagnostic value

5066 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Wang et al.
Figure 6. The coronal microPET images (top) and the radioactivity accumulation quantification (bottom) in selected organs of the tumor-bearing
mice administered with ∼250 mCi of 64Cu-L1. Arrows indicate the presence of glioma tumors. All microPET images were decay-corrected.
is often limited because of its insufficient tumor localization
and high liver uptake, which makes it very difficult to detect
small lesions in the chest and abdominal regions. Radiolabeled
The difference in mitochondrial potential between carcinoma
cells and normal epithelial cells is ∼60 mV.1-4,15 Thus, the
enhanced mitochondrial transmembrane potential in tumor cells
is most likely responsible for the tumor localization of 64Cu-
L1, 64Cu-L2, 64Cu-L3, and 99mTc-Sestamibi.
3
lipophilic cations, such as H-TPP and 18F-BzTPP, have also
been evaluated as radiotracers for imaging heart and tumors.29-37
3H-TPP has better tumor uptake than 99mTc-Sestamibi, but its
tumor selectivity is poor with the tumor/heart ratio <0.2. In
addition, there is always a high muscle uptake for both 3H-TPP
and 99mTc-Sestamibi.31,35
The difference between myocardium and normal tissue is their
mitochondrial population. Myocardium has the highest popula-
tion of mitochondria, which occupy 20-25% of the total vol-
ume of myocytes.31 Other mitochondrion-rich organs also
include salivary gland, liver and kidneys. Mechanism studies
on 99mTc-Sestamibi indicate that both the lipophilicity and
cationic charge play an important role in the accumulation and
retention of cationic 99mTc radiotracers in tumor cells and
myocytes.53-55 While the mitochondrial potential provides
electrochemical driving force for the radiolabeled lipophilic
cations to enter mitochondria of tumor cells, the lipophilicity
modulates their penetration kinetics into plasma and mitochon-
drial membranes. For more lipophilic cationic radiotracers, such
as 99mTc-Sestamibi (log P ) 1.09 ( 0.15), their diffusion
kinetics are so fast that they tend to localize in the mitochon-
drion-rich organs, such as heart, liver, kidneys, and salivary
gland.56-59 For more hydrophilic cations, such as 64Cu-L1 (log
P ) -2.67 ( 0.21), their membrane diffusion kinetics is very
slow (Figure 3: left), which makes it difficult for the more
hydrophilic cations across plasma and mitochondrial membranes,
thereby forcing them to localize in tissues where the mitochon-
drial transmembrane potential is elevated. Therefore, it is not
surprising that all three 64Cu radiotracers (log P ) -2.03 to
-2.67) have low uptake in the heart and muscle with very high
tumor/heart selectivity while lipophilic cations, such as 99mTc-
Sestamibi and 3H-TPP, are excellent myocardial perfusion
radiotracers because of the high heart and muscle uptake and
long myocardial retention. In this respect, myocardial perfusion
radiotracers with very high lipophilicity (log P ≈ 0.5-1.3) are
particularly useful for imaging “mitochondrial population” while
radiotracers with low lipophilicity (log P ) 0 to -3.0) are useful
for imaging “the enhanced mitochondrial potential”. This
conclusion is consistent with the high tumor selectivity of
rhodamine-123 (log P ) -0.62) and MKT-077 (1-ethyl-2-{-
[3-ethyl-5-(3-methylbenzothiazolin-2-yliden)]-4-oxothiazolin-2-
ylidenemethyl}pyridinium chloride), an anticancer drug with the
In this study, we discovered a new class of radiotracers with
very high tumor selectivity. All three 64Cu-labeled TPP cations
show the tumor uptake comparable to or better than that of
99mTc-Sestamibi (Figure 4), but their heart uptake is much lower
and their tumor/heart ratios are significantly better than those
of 99mTc-Sestamibi at all four time points (Figure 5). For
example, the tumor/heart ratio of 64Cu-L1 is 1.57 ( 0.18 at 5
min p.i. and increases to 4.25 ( 0.23 at 120 min p.i. while the
tumor/heart ratio of 99mTc-Sestamibi is <0.2 during the 2 h study
period. The heart uptake for 64Cu-L1, 64Cu-L2, and 64Cu-L3 is
also significantly lower than that of 3H-TPP in the same animal
species,35,36 most likely because of high hydrophilicity of the
64Cu-DO3A chelate. The tumor selectivity of 64Cu-L1, 64Cu-
L2, and 64Cu-L3 is unprecedented among all cationic radiotrac-
ers reported in the literature,35,36 including 99mTc-Sestamibi and
3H-TPP. Results from the in vitro cellular and mitochondrial
uptake experiments (Figure 3) clearly demonstrate that 64Cu-
labeled TPP cations are able to cross the plasma and mitochon-
drial membranes and localize in mitochondria of tumor cells.
There are several factors affecting the tumor uptake and tumor
selectivity of a cationic radiotracer. These include mitochondrial
population, mitochondrial potential, and lipophilicity of the
radiotracer. Various tumor cell lines exhibit differences in the
number, size, and shape of their mitochondria relative to normal
cells.3 The mitochondria of rapidly growing tumors tend to be
fewer in number and have fewer cristae than mitochondria from
slowly growing tumors; the latter are larger and have charac-
teristics more closely resembling those of normal cells.3 Thus,
the high tumor uptake (2-3 %ID/g) of 64Cu-L1, 64Cu-L2,
64Cu-L3, and 99mTc-Sestamibi is not likely related to mitochon-
drial population in tumor cells. In contrast, the enhanced
mitochondrial potential is prevalent in tumor cell phenotype.

Tumor-SelectiVe Imaging Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5067
epithelial carcinoma KB-3-1 cells and HT29par colon carcinoma
cells. The slow cellular and mitochondrial uptake kinetics is
consistent with the observation of a steady increase in the tumor
uptake of all three radiotracers (64Cu-L1, 64Cu-L2, and 64Cu-
L3) between 30 and 120 min p.i. in the biodistribution study.
The fact that the uptake of 64Cu-L1 (0.31 ( 0.03 to 0.53 (
0.05 %ID/million cells) in glioma cells between 15-120 min
postincubation is significantly (p < 0.01) lower than that in
mitochondria (1.39 ( 0.43 to 0.97 ( 0.27 %ID/million cells)
isolated from glioma cells strongly suggests that the diffusion
process across the plasma membrane is a key step for the
radiotracer to enter mitochondria of tumor cells. This conclusion
is also supported by the slower efflux kinetics of 64Cu-L1 from
glioma cells (Figure 3: right). It must be emphasized that these
in vitro data serve only as complimentary evidence, and should
not be viewed as “absolute” proof for the in vivo behavior of
the radiotracer.
Lipophilicity of a cationic radiotracer has a significant impact
on its diffusion kinetics to penetrate both plasma and mito-
chondrial membranes. For more lipophilic radiotracers, such as
99mTc-Sestamibi (log P ) 1.09 ( 0.15), their membrane
diffusion kinetics are fast and they can localize in tumor quickly.
For more hydrophilic cations, such as 64Cu-L1 (log P ) -2.67
( 0.21), their membrane penetration kinetics is very slow and
it takes them several hours to reach the maximal uptake in
glioma tumor cells. This explanation is consistent with the slow
cellular uptake kinetics of 64Cu-L1 (Figure 3: left) and is also
supported by the fact that 64Cu-L1 reaches its peak tumor uptake
at ∼6 h p.i. as observed in PET images of tumor-bearing mice
(Figure 5). Results from studies of cationic organic dyes also
showed that very hydrophilic and water-soluble organic cations
(log P < -1.5) will not be reach the equilibrium after 24 h
incubation with tumor cells.63,65
MDR gene (particularly MDR1) expression is another im-
portant factor that influences the tumor uptake of cationic
radiotracers. In this study, we choose the U87MG glioma cell
line since it has very limited MDR1 expression.42,43 High MDR1
expression is expected to result in a rapid efflux of cationic
radiotracer from tumor cells and significant reduction of its
tumor uptake.22-25 Thus, the tumor uptake data reported herein
reflects the intrinsic capability of 64Cu-L1, 64Cu-L2, and 64Cu-
L3 to localize in tumors due to the enhanced mitochondrial
potential. It is well-documented that cationic 99mTc radiotracers,
such as 99mTc-Sestamibi and 99mTc-Tetrofosmin originally
approved as myocardial perfusion radiopharmaceuticals, are
clinically useful for noninvasive imaging of the MDR1 expres-
sion in tumors.17-25 From this point of view, 64Cu-L1, 64Cu-
L2, and 64Cu-L3 might be useful for noninvasive imaging of
the MDR function in different tumors. Biodistribution and
imaging studies in tumor-bearing mice with and without MDR1
expression are still in progress, and the results will be reported
as a separate account.
In addition to diagnostic applications, both 64Cu and 67Cu
are useful for radiotherapy.39-41 64Cu has a low-energy â-emis-
sion (Emax ) 0.573 MeV, 39.6%) and an Auger-electron
emission (41%). 67Cu has three low-energy â-emissions (Emax
) 0.577 MeV, 20%; 0.484 MeV, 35%; and 0.395 MeV, 45%).
DO3A is useful for chelation of various therapeutic radionu-
clides, such as 111In(Auger-electron emission), 90Y (pure
â-emission, Emax ) 2.28 MeV), and 177Lu (low-energy â-emis-
sion Emax ) 0.497 MeV). The fact that 64Cu-labeled TPP/TPA
cations are able to localize in the energized mitochondria of
tumor cells suggest that the TPP cations chelated with a low-
energy â-particle or an Auger-electron radionuclide might be
Figure 7. Radio-HPLC chromatograms of the HPLC purified 64Cu-
L1 in saline before injection (A), in the urine sample at 30 min
postinjection (B), and in the feces sample at 120 min p.i. (C). Attempts
to collect the urine sample at 120 min p.i. were not successful. Each
mouse was administered with ∼100 µCi of 64Cu-L1.
log P value of -1.6.60,61 It is also supported by the results
obtained from studies on cationic triarylmethane dyes.62-65
It is interesting to note that all three 64Cu radiotracers
(64Cu-L1, 64Cu-L2, and 64Cu-L3) show a sharp decrease in their
tumor uptake between 5 and 30 min p.i., and a steady increase
in their tumor uptake between 30 and 120 min p.i. while
99mTc-Sestamibi has a slow washout during the 2 h study period
(Figure 4). We believe that the high initial tumor uptake of these
radiotracers is likely caused by their high blood radioactivity
level. The steady increase in tumor uptake suggests that it may
take time for radiotracer to form the equilibrium inside/outside
tumor cells. The fact that all three 64Cu radiotracers have a
steady increase in their tumor uptake suggests that they are
indeed able to localize inside mitochondria, instead of being
the “blood-flow” radiotracers. If they were simple blood flow
radiotracers, there would have been a significant washout over
time due to their rapid blood clearance.
In order to understand the biodistribution properties of
64Cu-L1, 64Cu-L2, and 64Cu-L3, we performed cellular and
mitochondrial uptake experiments using 64Cu-L1 in U87MG
glioma cells. It was found that there is a steady increase in
cellular uptake of 64Cu-L1 over 3 h (Figure 3: left). No peak
uptake is observed after >3 h incubation at 37 °C. In contrast,
99mTc-Sestamibi shows the cellular uptake plateau in <60 min
in several MDR-negative tumor cell lines,66-68 including human

5068 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21
Wang et al.
(9) Dairkee, S. H.; Hackett, A. J. Differential retention of rhodamine
123 by breast carcinoma and normal human mammary tissue. Breast
Cancer Res. Treat. 1991, 18, 57-61.
(10) Gottlieb, E.; Thompson, C. B. Targeting the mitochondria to enhance
tumor suppression. Methods Mol. Biol. 2003, 223, 543-554.
(11) Mannella, C. A. The relevance of mitochondrial membrane topology
to mitochondrial function. Biochim. Biophys. Acta 2006, 1762, 140-
147.
(12) Ross, M. F.; Kelso, G. F.; Blaikie, F. H.; James, A. M.; Cocheme,
H. M.; Filipovska, A.; Da Ros, T. D.; Hurd, T. R.; Smith, R. A. J.;
Murphy, M. P. Lipophilic triphenylphosphonium cations as tools in
mitochondrial bioenergetics and free radical biology. Biochemistry
(Moscow) 2005, 70, 222-230.
(13) Jakobs, S. High resolution imaging of live mitochondria. Biochim.
Biophys. Acta 2006, 1763, 561-575.
(14) Lichtshtein, D.; Kaback, H. R.; Blume, A. J. Use of a lipophilic cation
for determination of membrane potential in neuroblastomas-glioma
hybrid cell suspensions. Proc. Natl. Acad. Sci. U.S.A. 1979, 76, 650-
654.
(15) Hockings, P. D.; Rogers, P. J. The measurement of transmembrane
electric potential with lipophilic cations. Biochim. Biophys. Acta 1996,
1282, 101-106.
(16) Huang, S. G. Development of a high throughput screening assay for
mitochondrial membrane potential in living cells. J. Biomol. Screen-
ing 2002, 7, 383-389.
(17) Herman, L. W.; Sharma, V.; Kronauge, J. F.; Barbarics, E.; Herman,
L. A.; Piwnica-Worms, D. Novel hexakis(areneisonitrile)technetium-
(I) complexes as radioligands targeted to the multidrug resistance
P-glycoprotein. J. Med. Chem. 1995, 38, 2955-2963.
(18) Agrawal, M.; Abraham, J.; Balis, F. M.; Edgerly, M.; Stein, W. D.;
Bates, S.; Fojo, T.; Chen, C. C. Increased 99mTc-Sestamibi accumula-
tion in normal liver and drug resistant-tumors after the administration
of the glycoprotein inhibitor, XR9576. Clin. Cancer Res. 2003, 9,
650-656.
(19) Luker, G. D.; Francisco, P. M.; Dobkin, J.; Piwbica-Worms, D.
Modulation of the multidrug resistance P-glycoprotein: Detection
with technetium-99m sestamibi in vivo. J. Nucl. Med. 1997, 38, 369-
372.
very attractive radiotracers for tumor radiotherapy because they
will have more opportunities to interact intracellular biomol-
ecules, such as DNA molecules, than the extracellular receptor-
based radiotracers. They may also produce more “intracellular”
free radicals, which are very reactive toward many biomolecules
inside mitochondria of tumor cells. However, in the absence of
experimental data, these assumptions remain largely speculation.
Conclusion
In this study, we discovered a new class of PET radiotracers
with very high tumor selectivity. Results from the in vitro
cellular and mitochondrial uptake kinetics clearly demonstrate
that the 64Cu-labeled TPP cations are able to across the plasma
and mitochondrial membranes and localize in mitochondria of
tumor cells. The most striking difference between 64Cu-L1,
64Cu-L2, 64Cu-L3 and 99mTc-Sestamibi is their tumor/heart and
heart/muscle ratios. The heart uptake for 64Cu-L1, 64Cu-L2, and
64Cu-L3 is much lower than that of 99mTc-Sestamibi in the same
animal model. The tumor selectivity of 64Cu-L1, 64Cu-L2, and
64Cu-L3 is unprecedented among all the cationic radiotracers
reported in literature. While their high tumor uptake is due to
the enhanced negative mitochondrial transmembrane potentials
in glioma cells, their unprecedented high tumor selectivity is
attributed to their low lipophilicity (log P ) -2.03 to -2.67).
The results from this study strongly suggest that the 64Cu-labeled
TPP/TPA cations are very selective tumor-imaging radiotracers,
and can provide the information of mitochondrial bioenergetic
function by monitoring mitochondrial potential in tumors.
Acknowledgment. Authors would like to thank Dr. Sulma
I. Muhammed, the Director of Purdue Cancer Center Drug
Discovery Shared Resource, Purdue University, for her as-
sistance with the tumor-bearing animal model. This work is
supported, in part, by research grants: R01 CA115883 A2 (S.L.)
from National Cancer Institute (NCI), BCTR0503947 (S.L.)
from Susan G. Komen Breast Cancer Foundation, AHA0555659Z
(S.L.) from the Greater Midwest Affiliate of American Heart
Association, R21 EB003419-02 (S.L.) from National Institute
of Biomedical Imaging and Bioengineering (NIBIB), and R21
HL083961-01 from National Heart, Lung, and Blood Institute
(NHLBI).
(20) Liu, Z. L.; Stevenson, G. D.; Barrett, H. H.; Kastis, G. A.; Bettan,
M.; Furenlid, L. R.; Wilson, D. W.; Woolfenden, J. M. Imaging
recognition of multidrug resistance in human breast tumors using
99mTc-labeled monocationic agents and a high-resolution stationary
SPECT system. Nucl. Med. Biol. 2004, 31, 53-65.
(21) Liu, Z. L.; Stevenson, G. D.; Barrett, H. H.; Furenlid, L. R.; Wilson,
D. W.; Kastis, G. A.; Bettan, M.; Woolfenden, J. M. Imaging
recognition of inhibition of multidrug resistance in human breast
cancer xenografts using 99mTc-labeled sestamibi and tetrofosmin.
Nucl. Med. Biol. 2005, 32, 573-583.
(22) Sharma, V.; Piwnica-Worms, D. Metal complexes for therapy and
diagnosis of drug resistance. Chem. ReV. 1999, 99, 2545-2560.
(23) Muzzammil, T.; Ballinger, J. R.; Moore, M. J. 99mTc-sestamibi
imaging of inhibition of the multidrug resistance transporter in a
mouse xenograft model of human breast cancer. Nucl. Med. Commun.
1999, 20, 115-122.
(24) Sharma, V. Radiopharmaceuticals for assessment of multidrug
resistance P-glycoprotein-mediated drug transport activity. Biocon-
jugate Chem. 2004, 15, 1464-1474.
(25) Vaidyanathan, G.; Zalutsky, M. R. Imaging drug resistance with
radiolabeled molecules. Curr. Pharm. Des. 2004, 10, 2965-2979.
(26) Woo, D. V. Process of radioimaging the myocardium of mammals
utilizing radiolabeled lipophilic cations. US Patent 4,446,123, 1984.
(27) Srivastava, P. C.; Knapp, F. K. [(E)-1-[123I]Iodo-penten-5-yl]triph-
enylphosphonium iodide: convenient preparation of a potentially
useful myocardial perfusion agent. J. Med. Chem. 1984, 27, 978-
981.
References
(1) Kroemer, G.; Dallaporta, B.; Resche-Rigon, M. The mitochondrial
death/ life regulator in apoptosis and necrosis. Annu. ReV. Physiol.
1998, 60, 619-642.
(2) Modica-Napolitano, J. S.; Aprille, J. R. Delocalized lipophilic cations
selectively target the mitochondria of carcinoma cells. AdV. Drug
DeliVery ReV. 2001, 49, 63-70.
(3) Duchen, M. R. Mitochondria in health and disease: perspectives on
a new mitochondrial biology. Mol. Aspects Med. 2004, 25, 365-
451.
(28) Srivastava, P. C.; Hay, H. G.; Knapp, F. K. Effects of alkyl and aryl
substitution on the myocardial specificity of radioiodinated phos-
phonium, arsinium, and ammonium cations. J. Med. Chem. 1985,
28, 901-904.
(29) Krause, B. J.; Szabo, Z.; Becker, L. C.; Dannals, R. F.; Scheffel, U.;
Seki, C.; Ravert, H. T.; Dipaola, A. F., Jr.; Wagner, H. N., Jr.
Myocardial perfusion with [11C] triphenyl phosphonium: measure-
ments of the extraction fraction and myocardial uptake. J. Nucl. Biol.
Med. 1994, 38, 521-526.
(30) Madar, I.; Anderson, J. H.; Szabo, Z.; Scheffel, U.; Kao, P. F.; Ravert,
H. T.; Dannals, R. F. Enhanced uptake of [11C]TPMP in canine brain
tumor: a PET study. J. Nucl. Med. 1999, 40, 1180-1185.
(31) Madar, I.; Weiss, L.; Izbicki, G. Preferential accumulation of 3H-
tetraphenylphosphonium in non-small cell lung carcinoma in mice:
comparison with 99mTc-MIBI. J. Nucl. Med. 2002, 43, 234-238.
(32) Ravert, H. T.; Madar, I.; Dannals, R. F. Radiosynthesis 3-[18F]-
fluoropropyl and 4-[18F]fluorobenzyl triarylphosphonium ions. J.
Labelled Compd. Radiopharm. 2004, 47, 469-476.
(4) Modica-Napolitano, J. S.; Singh, K. K. Mitochondria as targets for
detection and treatment of cancer. Expert ReViews in Molecular
Medicine. http://www.ermm.cbcu.cam.ac.uk, 2002, (02)00445-3a
(short code: txt001ksb).
(5) Johnson, L. V.; Walsh, M. L.; Chen, L. B. Localization of
mitochondria in living cells with rhodamine 123. Proc. Natl. Acad.
Sci. U.S.A. 1980, 77, 990-994.
(6) Summerhayes, I. C.; Lampidis, T. J.; Bernal, S. D.; Nadakavukaren,
J. J.; Nadakavukaren, K. K.; Shepard, E. L.; Chen, L. B. Unusual
retention of rhodamine 123 by mitochondria in muscle and carcinoma
cells. Proc. Natl. Acad. Sci. U.S.A. 1982, 79, 5292-5296.
(7) Modica-Napolitano, J. S.; Aprille, J. R. Basis for the selective
cytotoxicity of rhodamine 123. Cancer Res. 1987, 47, 4361-4365.
(8) Davis, S.; Weiss, M. J.; Wong, J. R.; Lampidis, T. J.; Chen, L. B.
Mitochondrial and plasma membrane potentials cause unusual
accumulation and retention of rhodamine 123 by human breast
adenocarcinoma-derived MCF-7 cells. J. Biol. Chem. 1985, 260,
3844-3850.

Tumor-SelectiVe Imaging Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 21 5069
(33) Madar, I.; Ravert, H. T.; Du, Y.; Hilton, J.; Volokh, L.; Dannals, R.
F.; Frost, J. J.; Hare, J. M. Characterization of uptake of the new
PET imaging compound [18F]fluorobenzyl triphenylphosphonium in
dog myocardium. J. Nucl. Med. 2006, 47, 1359-1366.
(34) Cheng, Z.; Subbarayan, M.; Chen, X.; Gambhir, S. S. Synthesis of
(4-[18F]fluorophenyl)triphenylphosphonium as a potential imaging
agent for mitochondrial dysfunction. J. Labelled Compd. Radiopharm.
2005, 48, 131-137.
(54) Carvalho, P. A.; Chiu, M. L.; Kronauge, J. F.; Kawamura, M.; Jones,
A. G.; Holman, B. L.; Piwnica-Worms, D. Subcellular distribution
and analysis of technetium-99m-MIBI in isolated perfused rat hearts.
J. Nucl. Med. 1992, 33, 1516-1522.
(55) Crane, P.; Laliberte, R.; Heminway, S.; Thoolen, M.; Orlandi, C.
Effect of mitochondrial viability and metabolism on technetium-99m-
sestamibi myocardial retention. Eur. J. Nucl. Med. 1993, 20, 20-
25.
(56) Liu, S.; He, Z. J.; Hsieh, W. Y.; Kim, Y. S. Evaluation of novel
cationic 99mTc-nitrido complexes as radiopharmaceuticals for heart
imaging: Improving liver clearance with crown ether groups. Nucl.
Med. Biol. 2006, 33, 419-432.
(57) He, Z. J.; Hsieh, W. Y.; Kim, Y. S.; Liu, S. Evaluation of novel
cationic 99mTc(I)-tricarbonyl complexes as potential radiotracers for
myocardial perfusion imaging. Nucl. Med. Biol. 2006, 33, 1045-
1053.
(35) Min, J. J.; Biswal, S.; Deroose, C.; Gambhir, S. S. Tetraphenylphos-
phonium as a novel molecular probe for imaging tumors. J. Nucl.
Med. 2004, 45, 636-643.
(36) Steichen, J. D.; Weiss, M. J.; Elmaleh, D. R.; Martuza, R. L.
Enhanced in vitro uptake and retention of 3H-tetraphenylphosphonium
by nervous system tumor cells. J. Neurosurg. 1991, 74, 116-122.
(37) Cheng, Z.; Winant, R. C.; Gambhir, S. S. A new strategy to screen
molecular imaging probe uptake in cell culture without radiolabeling
using matrix-assisted laser desorption/ionization time-of-flight mass
spectrometry. J. Nucl. Med. 2005, 46, 878-886.
(38) Blower, P. J.; Lewis, J. S.; Zweit, J. Copper radionuclides and
radiopharmaceuticals in nuclear medicine. Nucl. Med. Biol. 1996,
23, 957-980.
(39) Reichert, D. E.; Lewis, J. S.; Anderson, C. J. Metal complexes as
diagnostic tools. Coord. Chem. ReV. 1999, 184, 3-66.
(40) Liu, S.; Edwards, D. S. Bifunctional chelators for therapeutic
lanthanide radiopharmaceuticals. Bioconjugate Chem. 2001, 12,
7-34.
(41) Liu, S. The role of coordination chemistry in development of target-
specific radiopharmaceuticals. Chem. Soc. ReV. 2004, 33, 445-461.
(42) Ba¨hr, O.; Wick, W.; Weller, M. Modulation of MDR/MRP by wild-
type and mutant p53. J. Clin. InVest. 2001, 107, 643-645.
(43) Ba¨hr, O.; Rieger, J.; Duffner, F.; Meyermann, R.; Weller, M.; Wick,
W. P-glycoprotein and multidrug resistance-associated protein medi-
ate specific patterns of multidrug resistance in malignant glioma cell
lines, but not in primary glioma cells. Brain Pathol. 2003, 13, 482-
494.
(44) Yang, C. T.; Li, Y. X.; Liu, S. Synthesis and structural characteriza-
tion of complexes of a DO3A-conjugated triphenylphosphonium
cation with diagnostically important metal ions. Inorg. Chem., in
press.
(45) Shannon, R. D. Revised effective ionic radii and systematic studies
of interatomic distances in halides and chalcogenides. Acta Crystal-
logr. Sect. A. 1976, 32, 751-767.
(46) Acampa, W.; Di, Benedetto, C.; Cuocolo, A. An overview of
radiotracers in nuclear cardiology. J. Nucl. Cardiol. 2000, 7, 701-
707.
(47) Dilsizian, V. The role of myocardial perfusion imaging in vascular
endothelial dysfunction. J. Nucl. Cardiol. 2000, 7, 180-184.
(48) Schwaiger, M.; Melin, J. Cardiological applications of nuclear
medicine. Lancet 1999, 354, 661-666.
(49) Beller, G. A.; Zaret, B. L. Contributions of nuclear cardiology to
diagnosis and prognosis of patients with coronary artery disease.
Circulation 2000, 101, 1468-1478.
(50) Parker, J. A. Cardiac nuclear medicine in monitoring patients with
coronary heart disease. Semin. Nucl. Med. 2001, 31, 223-237.
(51) Saha, G. B.; Go, R. T.; Macintyre, W. J. Radiopharmaceuticals for
cardiovascular imaging. Nucl. Med. Biol. 1992, 19, 1-20.
(52) Jain, D. Technetium-99m labeled myocardial perfusion imaging
agents. Semin. Nucl. Med. 1999, 29, 221-236.
(53) Piwnica-Worms, D.; Kronauge, J. F.; Chiu, M. L. Uptake and
retention of hexakis(2-methoxyisobutylisonitrile) technetium(I) in
cultured chick myocardial cells: mitochondrial and plasma membrane
potential dependence. Circulation 1990, 82, 1826-1838.
(58) Kim, Y. S.; He, Z. J.; Hsieh, W. Y.; Liu, S. Impact of bidentate
chelators on lipophilicity, stability and biodistribution characteristics
of cationic 99mTc-nitrido complexes. Bioconjugate Chem. 2007, 18,
929-936.
(59) Liu, S. Ether and crown ether-containing cationic 99mTc complexes
useful as radiopharmaceuticals for heart imaging. Dalton Trans. 2007,
1183-1193.
(60) Fantin, V. R.; Berardi, M. J.; Scorrano, L.; Korsmeyer, S. J.; Leder,
P. A novel mitochondriotoxic small molecule that selectively inhibits
tumor cell growth. Cancer Cell 2002, 2, 29-42.
(61) Britten, C. D.; Rowinsky, E. K.; Baker, S. D.; Weiss, G. R.; Smith,
L.; Stephenson, J.; Rothenberg, M.; Smetzer, L.; Cramer, J.; Collins,
W.; Von, Hoff, D. D.; Eckhardt, S. G. A phase I and pharmacokinetic
study of the mitochondrial-specific rhodacyanine dye analog MKT-
077. Clin. Cancer Res. 2000, 6, 42-49.
(62) Kandela, I. K.; Bartlett, J. A.; Indig, G. L. Effect of molecular
structure on the selective phototoxicity of triarylmethane dyes towards
tumor cells. Photochem. Photobiol. Sci. 2002, 1, 309-314.
(63) Kandela, I. K.; Lee, W.; Indig, G. L. Effect of the lipophilic/
hydrophilic character of cationic triarylmethane dyes on the selective
phototoxicity toward tumor cells. Biotech. Histochem. 2003, 78, 157-
169.
(64) Trapp, S.; Horobin, R. W. A predictive model for the selective
accumulation of chemicals in tumor cells. Eur. Biophys. J. Biophys.
Lett. 2005, 34, 959-966.
(65) Horobin, R. W. Uptake, distribution, and accumulation of dyes and
fluorescent probes with living cells: a structure-activity modeling
approach. AdV. Colour Sci. Technol. 2001, 4, 101-107.
(66) Ma´ria´n, T.; Balkay, L.; Szabo´, G.; Krasznai, Z. T.; Hera´di, Z.;
Galuska, L.; Szabo´-Pe´li, J.; EÄ sik, O.; Tro´n, L.; Krasznai, Z. Biphasic
accumulation kinetics of [99mTc]-hexakis-2-methoxyisobutyl isonitrile
in tumour cells and its modulation by lipophilic P-glycoprotein
ligands. Eur. J. Pharm. Sci. 2005, 25, 201-209.
(67) Ma´ria´n, T.; Szabo´, G.; Nagy, H.; Szincsa´k, N.; Juha´sz, I.; Galuska,
L.; Balkay, L.; Mikecz, P.; Tro´n, L.; Krasznai, Z. In vivo and in
vitro multitracer analysis of P-glycoprotein expression-related mul-
tidrug resistance. Eur. J. Nucl. Med. Mol. Imaging 2003, 30, 1147-
1154.
(68) Lorke, D. E.; Kru¨ger, M.; Buchert, R.; Bohuslavizki, K. H.; Clausen,
M.; Schumacker, U. In vitro and in vivo tracer characteristics of an
established multidrug-resistant human colon cancer cell line. J. Nucl.
Med. 2001, 42, 646-654.
JM0704088