Evaluation of 64Cu-ATSM in vitro and in vivo in a hypoxic tumor model.

Abstract

We have evaluated Cu-diacetyl-bis(N4-methylthiosemicarbazone) (Cu-ATSM), an effective marker for the delineation of hypoxic but viable tissue, in vitro in the EMT6 carcinoma cell line under varying degrees of hypoxia and compared it with the flow tracer 64Cu-pyruvaldehyde-bis(N4-methylthiosemicarbazone) (Cu-PTSM) and the hypoxic tracer 18F-fluoromisonidazole (MISO). We have also compared the uptake of Cu-ATSM and Cu-PTSM in vivo and ex vivo in a murine animal model bearing the EMT6 tumor.
Uptake of 64Cu-ATSM, 64Cu-PTSM and 18F-MISO in vitro into EMT6 cells was investigated at the dissolved oxygen concentrations of 0, 1 x 10(3), 5 x 10(3), 5 x 10(4) and 2 x 10(5) ppm. Biodistribution performed at 1, 5, 10, 20 and 40 min compared 64Cu-ATSM with 64Cu-PTSM in BALB/c mice bearing EMT6 tumors. To determine long-term retention of 64Cu-ATSM, biodistribution was also performed at 1, 2 and 4 h. Ex vivo autoradiography of tumor slices after co-injection of 60Cu-PTSM (60Cu, T1/2 = 23.7 min) and 64Cu-ATSM (64Cu, t1/2 = 12.7 h) into the same animal was performed.
After 1 h, 64Cu-ATSM was taken up by EMT6 cells: 90% at 0 ppm, 77% at 1 x 10(3) ppm, 38% at 5 x 10(3) ppm, 35% at 5 x 10(4) ppm and 31% at 2 x 10(5) ppm. 18F-MISO also showed oxygen concentration dependent uptake, but with lower percentages than 64Cu-ATSM. 64Cu-PTSM showed 83%-85% uptake into the cells after 1 h, independent of oxygen concentration. Biodistribution data of 64Cu-ATSM and 64Cu-PTSM showed optimal tumor uptake after 5 and 10 min, respectively (0.76% injected dose (ID)/organ for 64Cu-ATSM and 1.11%ID/organ for 64Cu-PTSM). Ex vivo imaging experiments showed 60Cu-PTSM uniform throughout the EMT6 tumor, but heterogeneous uptake of 64Cu-ATSM, indicative of selective trapping of 64Cu-ATSM into the hypoxic tumor cells.
Cu-ATSM exhibits selectivity for hypoxic tumor tissue both in vivo and in vitro and may provide a successful diagnostic modality for the detection of tumor ischemia.

Full text

Page 1

Evaluation of 64Cu-ATSM In Vitro and In Vivo
in a Hypoxie Tumor Model
Jason S. Lewis, Deborah W. McCarthy, Timothy J. McCarthy, Yasuhisa Fujibayashi and Michael J. Welch
Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, Missouri; BiomédicalImaging Research
Center, Fukui Medical University, Matsuoka, Fukui, Japan
We have evaluated
zone) (Cu-ATSM), an effective marker for the delineation of
hypoxic but viable tissue, in vitro in the EMT6 carcinoma cell line
under varying degrees of hypoxia and compared it with the flow
tracer «Cu-pyruvaldehyde-bisfW-methylthiosemicarbazone)
(Cu-PTSM) and the hypoxic tracer 18F-fluorom¡sonidazole (MISO).
We have also compared the uptake of Cu-ATSM and Cu-PTSM
in vivo and ex vivo in a murine animal model bearing the EMT6
tumor. Methods: Uptake of «Cu-ATSM,«Cu-PTSM and 18F-
MISO in vitro into EMT6 cells was investigated at the dissolved
oxygen concentrations of 0, 1 x to3, 5 x 103,5 x 10" and 2 x
105 ppm. Biodistribution performed at 1, 5, 10, 20 and 40 min
compared «Cu-ATSMwith «Cu-PTSMin BALB/c mice bearing
EMT6 tumors. To determine long-term retention of «Cu-ATSM,
biodistribution was also performed at 1, 2 and 4 h. Ex vivo
autoradiography of tumor slices after co-injection of 60Cu-PTSM
(«Cu,T1/2= 23.7 min) and «Cu-ATSM(«Cu,t,/2 = 12.7 h) into
the same animal was performed. Results: After 1 h, «Cu-ATSM
was taken up by EMT6 cells: 90% at 0 ppm, 77% at 1 x 1o3ppm,
38% at 5 x 103ppm, 35% at 5 x 10" ppm and 31% at 2 x 10s
ppm. 18F-MISO also showed oxygen concentration dependent
uptake, but with lower percentages than 64Cu-ATSM.64Cu-PTSM
showed 83%-85% uptake into the cells after 1 h, independent of
oxygen concentration. Biodistribution data of «Cu-ATSMand
«Cu-PTSMshowed optimal tumor uptake after 5 and 10 min,
respectively (0.76% injected dose (ID)/organ for «Cu-ATSMand
1.11%ID/organ for «Cu-PTSM). Ex vivo imaging experiments
showed 60Cu-PTSM uniform throughout the EMT6 tumor, but
heterogeneous uptake of 64Cu-ATSM, indicative of selective
trapping of «Cu-ATSMinto the hypoxic tumor cells. Conclusion:
Cu-ATSM exhibits selectivity for hypoxic tumor tissue both invivo
and in vitro and may provide a successful diagnostic modality for
the detection of tumor ischemia.
Key Words: hypoxia;«Cu; «>Cu; ATSM; EMT6
J NucÃ- Med 1999; 40:177-183
Cu-diacetyl-bis(/V-methylthiosemicarba-
IV
seriously affect the successful management
poxic cells within the tumor can account, in part, for this
resistance to radiotherapy(1,2) and chemotherapy
ssistanceof tumors to conventional therapies
of cancer. Hy
can
(3-5). A
Received Mar. 9,1998;
For correspondence
PhD, Mallinckrodt
Medicine, Campus Box 8225,510
revision accepted May 21,1998.
and requests for reprints contact: Michael J. Welch,
Instituteof Radiology, Washington
S. Kingshighway Blvd., St. Louis, MO 63110.
UniversitySchool of
noninvasive, quantitative PET imaging agent could measure
the extent of tumor hypoxia before a treatment
Misonidazole derivatives have been labeled with I8Ffor PET
imaging;clinicalstudiesinvolving
l8F-fluoromisonidazole(MISO),
ences between normal and hypoxic tissues. Misonidazole
analogs and 2-nitroimidazole
labeled with 123I(6-10)for SPECT imaging.
compounds,WmTc-HL91 and BMS-181321,
strated increased uptake in hypoxic and low-flow ischemie
myocardium (8,9) and in tumors (11,12).
The production and use of the positron emitting isotopes
of copper ("'Cu, 6lCu, 62Cu and 64Cu) in nuclear medicine
have been reviewed (13). Cu-diacetyl-bisiA^-methylthiosemi-
carbazone)(ATSM) has been shown to be an effective
marker for delineatinghypoxic,
tively trapped in hypoxic tissue but rapidly washed out from
normoxiccells(14,15). 62Cu-ATSM is currently
investigationin humans at Fukui Medical Center, Japan, for
the detection of ischemia (16). In addition, ^Cu-ATSM
approvedat WashingtonUniversity
Louis, for the clinical diagnosis of tumor hypoxia in lung
cancer.
With the availabilityof high specific activity positron
emittingisotopes of copper from a biomedicai
(17), we have evaluated MCu-ATSM in vitro with the EMT6
cell line under varying pO2 and in vivo in a murine animal
model.
regimen.
the lead compound,
showimageablediffer
functionalitieshave also been
The WmTc
have demon
viable tissue; it is selec
under
is
MedicalSchool,St.
cyclotron
MATERIALS AND METHODS
MCu and ^'Cu were produced on the Cyclotron
CS15 cyclotron at the Washington University Medical School as
previously described (17,18). I8F-MISO was synthesized according
Corporation
to procedures by McCarthy et al. (19). Female BALB/c mice were
purchased from Charles River Laboratories (Wilmington, MA). All
chemicals unless otherwise stated were purchased from Aldrich
ChemicalCompany,Inc. (Milwaukee,
prepared with distilled deionized water (Milli-Q: > 18 Mil resistiv
ity). Thin-layer chromatography
silica gel TLC plates with ethyl acetate as the mobile phase. TLC
plates were analyzed on a BIOSCAN System 200 imaging scanner
(Washington, DC). Radioactive samples were counted on a Beck-
man 8000 gammacounter(Irvine,
prepared in Tissue-Tek Embedding Medium (Miles, Inc., Elkhart,
WI). All solutionswere
(TLC) was performed by using
CA). Tumorsliceswere
MCu-ATSMIN EMT6 HYPOXICTUMORMODELS • Lewis et al.177

Page 2

IN). Electronic autoradiography
ager Electronic Autoradiography
Meriden, CT) (20). The mouse mammary tumor line EMT6 was
obtained from the laboratories of Dr. Ronald S. Pardini at the
University of Nevada, Reno, and maintained by serial passage in
cell culture. Oxygen concentration
monitored by an Oxygen Transmitter
Sensor from Mettler Toledo (Wilmington, MA).
was performed on an Instantlm-
System (Packard Instrument Co.,
during the in vitro studies was
(Model 4300) and Oxygen
Radiochemical Synthesis
MCu-ATSM, 64Cu-pyruvaldehyde-bis(Ar*-methylthiosemicarba-
zone) (MCu-PTSM) and '»Cu-PTSM were produced by methods
identical to literature procedures
produced at 1 X 10~2MBq/ug.
(20,21). The compoundswere
Uptake of Radiolabeled Compounds into EMT6 Cells
The apparatus and procedures for the in vitro experiments
based on methods previously described (22,23). Viability of the
cells and cell numberswas measured
according to trypan blue exclusion procedures.
(1.2-1.4X IO6 cells/mL)were equilibrated
bottomed flask at a maintained temperature
are
with a hemocytometer
The EMT6 cells
in a glass round-
of 37°C.A continual
flow of warmed,
[control]; 5% O2, 5% CO2, 90% N, [hypoxia]; 0.5% O2, 5% CO2,
94.5% N2 [hypoxia]; 0.1% O2, 5% CO2, 94.9% N2 [hypoxia]; 0%
O2, 5% CO2, 95% N2 [anoxia]) was passed over the cells. The pO2
and temperatureweremonitored
equilibration, 7.4 MBq (200 uCi) (0.15 ug) MCu-ATSM (0.15 ug),
MCu-PTSM (0.15 ug) or 18F-MISO were added to the cells. At 1, 5,
15, 30, 45 and 60 min, triplicate aliquots were removed. The cells
were pelleted from the media, and the percentage uptake of the
compound into the cells was calculated. Cell viability was >96%.
Additional experiments examined
higher cell concentrations(5.0 X IO6 cells/mL).
EMT6 cells were gassed with the required gas mixture in cell
culture flasks. The flasks were kept at 37°C and, after 1h, 7.4 MBq
(200 pCi) 18F-MISO were added.
constantlymonitored.After 2 h, the cells were separated
washed to calculate percentage uptake.
humidifiedgas (20% O2, 5% CO2, 75% N2
by an oxygenprobe. After
the uptake of 18F-MISO at
Monolayersof
Oxygen concentrationwas
and
Retention of Radiolabeled Compounds in EMT6 Cells
The washout of radioactivity
obtaining a cell pellet after 1 h incubation with the radiolabeled
compounds and resuspendingit in fresh media. The fully oxygen
ated gas mixture was blown over the cells (20% O2). Samples were
taken at 1,5, 15, 30, 45 and 60 min postresuspension.
from the cells was observed by
Animal Biodistribution Studies
All animal experiments were conducted in compliance with the
Guidelines for the Care and Use of Research Animals established
by Washington University's Animal Studies Committee.
BALB/c mice were implanted subcutaneously
with 2.0 X 10s EMT6 cells from cell culture. At 10 days (tumors
approximately0.4-0.5 g), "Cu-ATSM
MCu-PTSM (0.2 MBq [5 uCi]) was injected into the tail vein and
the animals were euthanized by cervical dislocation at 1,5, 10, 20
and 40 min (ATSM and PTSM) and at 1, 2 and 4 h postinjection
(ATSM only) (n = 4). Selected tissues and organs were harvested
and weighed, and the activity counted on the gamma counter. The
percentage injected dose per gram (%ID/g) and percentage injected
dose per organ (%ID/organ) for each tissue were calculated.
into each flank
(0.2 MBq [5 uCi]) or
Electronic Autoradiography
EMT6cells
subcutaneouslyinto each flank of 5 BALB/c mice. At 9 d (tumors
approximately0.25g), the mice were co-injected
(150 uCi) «Cu-ATSMand 0.2 MBq (5 uCi) ^Cu-PTSM,
(2.5 X 10s) from cell culturewere implanted
with 5.5 MBq
in saline,
into the tail vein. The animals were killed by cervical dislocation
after 10 min and the tumors excised. Slices (1 mm thick) were
mounted and placed in the electronic autoradiography
after the initial scan to localize
visualized distribution of MCu-ATSM.
system 5 h
a second ^Cu-PTSM;scan
RESULTS
In Vitro Studies
The uptake of ^Cu-ATSM
gated as a function of pO2 at fixed temperature
into EMT6 cells was investi
(37°C),pH
and cell concentration.
MCu-ATSM was associated
whole range of pO2. By 1 h, the percentage
^Cu-ATSM into the cells was significantly different depend
ing on pO2 (Fig. 1A). The uptake of ^Cu-ATSM
a function of pO2 (in ppm) at 1h in Figure 2A.
^Cu-PTSMwas more rapidly and efficiently taken up by
EMT6 cells after 1min (76%-82%),
In addition, this uptake was maintained over the experiment
(83%-85% at 1h) (Fig. IB).
18F-MISO uptake was not detected when suspended cells
were used. Therefore, uptake was studied in monolayers of
cells at higher concentrations
At 1 min, between 8% and 14% of
with the EMT6 cells over the
uptakeof
is shown as
independent of the pO2.
(5.0 X IO6cells/mL). After 2
20
30 40
time (minutes)
50
60
B
100 -
so
is 60!
J40 •
203040
time (minutes)
SO
60
FIGURE 1.
PTSM into EMT6 cells over time at varying oxygen concentra
tions: 0% O2(A), 0.1% O2(X), 0.5% O2(A), 5% O2(D) and 20%
O2(•).Errors if not indicated are within symbols.
Percentage uptakeof (A)64Cu-ATSMand (B^Cu-
178THEJOURNAL OFNUCLEAR MEDICINE • Vol. 40 • No. 1 • January 1999

Page 3

10100
Oxygen concentration
(ppm)
100010000
100000 1000000
FIGURE 2.
against oxygen concentration (expressed in ppm). (B) Percent
age uptake of 18F-MISO(•) after 2 h against oxygen concentra
tion (expressed in ppm). Values shown are actual percent
multiplied by 5 (for clarity). Errors if not indicated are within
symbols.
(A) Percentage uptakeofMCu-ATSM (*) after 1 h
h, the percentage uptake of I8F-MISO into the cells was seen
to be pO2 dependent, but only approximately
uptake of «Cu-ATSM (10.9% at 0 ppm, 7.7% at 1 X IO3
ppm, 5.6% at 5 X IO3ppm, 1.0% at 5 X IO4ppm and 0.5% at
2 X IO5ppm) (Fig. 2B).
Washout studies of ^Cu-ATSM
EMT6 cells after l h incubation with the tracer at various
pO2 and resuspension in fresh media were performed at 20%
O2- It is clearly seen that the oxygen concentrations
the initial tracer incubation determine the amount of radioac
tivity washed out from the cells incubated with MCu-ATSM
(Fig. 3A). Assuming 100% radioactivity associated with the
cells at O min, after l h the cells previously maintained
anoxic (0%) oxygen levels lost 27% of the radioactivity. For
cells incubated at 5% O2, this value was 32% and at normal
pO2 (20% O2) 42%. The cells incubated with "Cu-PTSM,
however, retainedradioactivity
pO2 (Fig. 3B). The amount of radioactivity lost after l h was
between 9%-l 1% at all pO2.
10% of the
and ^Cu-PTSMfrom the
during
at
independent of the initial
Animal Biodistribution
Biodistributions
formed with MCu-ATSM and ^Cu-PTSM
tribution of ^Cu-ATSM
(data not shown).
Both MCu-ATSM and MCu-PTSM are rapidly and effi
Studies
at 1, 5, 10, 20 and 40 min were per
(Table 1). Biodis-
was also performed at 1, 2 and 4 h
ciently cleared from the blood (3.65%ID/g and 3.82%ID/g,
respectively) at 1 min. After 40 min, 2.22%ID/g ^Cu-ATSM
and 1.87%ID/g "Cu-PTSM is still seen in the blood pool.
For ^Cu-ATSM over 4 h, this value does not decrease
(2.22%ID/g).
64Cu-PTSMisefficiently
(28.69%ID/g)after 1 min and is retained over the 40-min
period (30.07%ID/g).Conversely,
lower uptakeafter 1 min (12.89%ID/g)
^Cu-PTSMand is retained over a 4-h period (10.84%ID/g).
In the heart tissue after 1 min, 20.50%ID/g
extractedfrom the blood pool and retained (16.75%ID/g)
extractedintothelung
MCu-ATSM exhibits
compared
a
with
^u-FTSMis
over 40 min. ^Cu-ATSM
tionvalues
8.75%ID/gat 1 min, lowering significantly
after 40 min (P < 0.001 ) (77% lower than that observed for
^Cu-PTSM).
The brain presents the most suitable control organ for the
comparisonbetween64Cu-ATSM
^Cu-PTSMrapidly crosses the blood-brain
retained in the braintissue
PTSMwas in the brain after
8.39%ID/gover 40 min. "Cu-ATSM,
extractedefficiently from the blood pool into the brain
(10.46%ID/gat 1 min), is not reduced intracellularly
washed out, giving values of 3.34%ID/g
2.85%ID/g at 4 h.
Both MCu-ATSM and MCu-PTSM are cleared through the
liver and kidney. Fast kidney uptake is observed for both
MCu-ATSM (23.45%ID/gat
(28.32%ID/g),decreasing to similar levels after 40 min
(11.94%ID/gand 14.48%ID/g,
increasesfor both complexes
creases from 7.75%ID/g at 1 min to 29.83%ID/g
and "Cu-PTSMincreases from 7.21%ID/g
25.70%ID/g at 40 min. Liver uptake is followed by uptake in
the intestines before excretion for both compounds.
values stabilize for ^Cu-ATSM over 4 h.
The uptakeof ^Cu-ATSM
EMT6 tumor showed optimal uptake after 10 min (4.78 ±
shows markedly smaller extrac
washout fromandgreatermyocardium,
to 3.81%ID/g
and64Cu-PTSM.
barrier and is
9.38%ID/g
1 min and stabilized
although
(24,25);MCu-
at
initially
and is
at 40 min and
1 min) and^Cu-PTSM
respectively).
over time. ^Cu-ATSM
Liver uptake
in
at 40 min,
at 1 min to
These
and "Cu-PTSM into the
so
60
90-
80
Ì?70-
«0-
50
30
time (minutes)
40
FIGURE 3.
from EMT6 cells after 1 h incubation at 37°Cat varying oxygen
concentrations; 0% O2(*), 5% O2(•), 20% O2(•).Errors if not
indicated are within symbols.
Washout of <*Cu-ATSM (A) and MCu-PTSM (B)
MCu-ATSM INEMT6 HYPOXIC TUMORMODELS• Lewis et al.179

Page 4

TABLE 1
Biodistribution(%ID/g ±SD, n = 4) at 1, 5,10, 20 and 40 min of ^Cu-ATSM
Bearing the EMT6 Mammary Tumor
(top) and ^Cu-PTSM (bottom) in BALB/c Mice
Tissue 1 min5 min 10min20min40 min
BloodLung
LiverSpleen
Kidney
HeartBrainBoneTumorIntestines3.65
±0.1512.89
± 0.55
7.75 ±1.474.80
±1.63
23.45 ±1.60
8.75 ±0.0510.46
± 0.411
.82 ±0.293.50
± 0.735.25
± 0.502.
13 ±0.3711.71
± 0.37
20.84 ±2.506.86
±1.37
18.02 ± 0.85
4.15 ±0.594.78
±0.351
.64 ±0.293.41
± 0.367.87
±1.102.25
±
± 0.99
0.2512.21
22.98 ±1.436.40
± 0.60
17.29 ± 0.36
3.94 ±0.253.76
±0.301.86
±0.384.78
±1.009.66
±0.191
.79 ±0.2611.88
± 0.95
27.82 ±1.995.38
± 0.22
14.32 ± 0.97
3.26 ±0.473.18
±0.161
.33 ±0.273.76
± 0.6110.59
±1.052.22
+
± 1.23
0.1514.86
29.83 ±1.205.24
± 0.19
11.94±1.65
3.81 ± 0.423.34
±0.321.47±
0.274.17
± 1.0313.34
±1.08
Tissue
1 min5 min 10 min20 min 40 min
BloodLungLiverSpleenKidneyHeartBrainBoneTumorIntestines3.8228.697.216.2528.3220.509.381.483.350.294.931.190.771.392.360.900.380.545.77
0.292.9133.1412.324.9521.0918.999.041.423.365.68±±-t--t-±±±±±-+-0.310.721.001.052.535.110.620.330.6
±0.273.191.320.894.890.991.190.031.000.621.5528.3920.843.5414.831 1.211.8730.0725.704.7214.4816.758.391.734.2
±34.02
±16.42
±5.93
±22.21
±19.04
±10.16
±1.44
±4.40
±5.90
2.29±
3.65±0.99±0.21±0.69±
2.16±0.68±0.56±
1.32±
1.31±
1.23±0.28±0.81±0.70
1.00%ID/g
MCu-PTSM).
for "Cu-ATSMand 4.40± 1.00%ID/gfor
Ex Vivo Autoradiography
Ex vivo autoradiography of the EMT6 tumors was under
taken by the co-injectionof «'Cu-PTSM and MCu-ATSM
into the same animal (Fig. 4). The MCu-PTSM displays
homogeneous uniform uptake throughout the tumor, suggest
ing uniform blood flow. After allowing for the decay of the
shorterlived ^Cu, the MCu-ATSM distribution
served. All tumors studied exhibited heterogeneous
of the hypoxic tracer, indicative of the selective trapping of
MCu-ATSM into the hypoxic cells of the tumor. Figure 4 is
representativeof the six studies,
MCu-ATSM observed in 15%-30% of the tumor.
was ob
uptake
with intense uptake of
DISCUSSION
Copper complexes
have been investigated for use within the field of radiophar-
maceutical chemistry (11,13-15,21,24-27).
cation is in the selective detection of viable hypoxic tissue.
Cu-ATSM has been investigated
ischemia (75). The successful determination
fraction within a tumor is important
treatment regimen with radiotherapy
Cu-ATSM is a complex of low molecular weight, high mem
brane permeabilityand low redox potential
retention within hypoxic tissue. Therefore, the use of Cu-ATSM
in the detection of tumor hypoxia is of significant clinical
interest. We report here our findings on the selectivity
Cu-ATSM for hypoxic tissue in comparison with 64Cu-PTSM,
based on thiosemicarbazoneligands
One such appli
for the detection of heart
of the hypoxic
when considering
or chemotherapy
a
(3).
for selective
of
Max
Min
w'Cu(PTSM)"CiKATSM)
Max
Min
«•Cu(PTSM) MCu(ATSM)
FIGURE
«"Cu-ATSM (right) and «Cu-PTSM
graph of EMT6 tumor with «Cu-ATSM (right) and "Cu-PTSM
4.(A)Typical autoradiograph of EMT6 tumor
autoradio
with
(left). (B) Typical
(left).
180THEJOURNAL OFNUCLEAR MEDICINE • Vol.40 • No. 1 • January 1999

Page 5

a known blood-flow and tumor-flow tracer (24,25), and
I8F-MISO,a known tracer for the detection of hypoxic tissue
(28-30).
The EMT6 carcinoma cell line is an in vitro cultured cell
line available in monolayer, spheroidal and invivo form (3Ì,
32). This cell line has been used extensively to investigate
hypoxic agents, because the cells cultured as spheroids or as
solid tumors in vivo in BALB/c mice contain a significant
hypoxic fraction dependent on size and age (33,34).
The apparatus used in the in vitro experiments was
designed to provide the cells with a controlled environment.
The use of 5% CO2inall the certified gas mixtures preserved
the pH of the media, and the continual agitation of the cells
avoided clumping and possible formation of spheroidal
microenvironments. Cell concentration, although not di
rectly investigated as a determining factor in compound
uptake in this investigation, was maintained at constant
levels. The maintenance of temperature and cell density/
concentration ensures that these factors do not contribute to
differences in uptake between the compounds investigated,
an observation noted in an investigation on the technetium-
labeled nitroimidazole derivative-BMS181321 (11). The
gases were warmed and humidified by passing through a
glass bubbler that contained water at 37°C. The glass vessels
and Tygon R3603 tubing (Norton, Akron, Ohio) were used
in the apparatus because of their relative impermeability to
oxygen. Therefore, the uptake of radiolabeled compound is
entirely based on the dissolved pO2in the cell media.
In the invitro study,the uptake of MCu-ATSMisrelated in
a sigmoidal fashion to the pO2of the media, where retention
was markedly increased under hypoxic and anoxic condi
tions (Fig. 2A). The uptake of I8F-MISOin the EMT6 cells
in monolayers follows a similar pattern (Fig. 2B). This same
binding profile is observed for the uptake of [3H]F-MISO
into EMT6 and V79 cells (35,36). With I8F-MISO,binding
to the EMT6 cells initiates at higher oxygen concentrations
than with "Cu-ATSM. The percentage uptake of 18F-MISO
is also much lower than that of "Cu-ATSM after a longer
incubation time. It was necessary to use monolayers of cells
at a higher concentration to show the oxygen-dependent
binding of 18F-MISOto EMT6 cells. The significance of the
much lower cellular uptake of 18F-MISOand the uptake at
different pO2to in vivo situations has yet to be determined.
"Cu-PTSM showed uptake completely independent of oxy
gen concentration. From Figure la, itisseen that"Cu-ATSM
is selectively trapped at levels depending on the pO2. The
significant (P < 0.01) and equal uptake of the flow tracer,
"Cu-PTSM, at all oxygen concentrations, acts as a suitable
control to "Cu-ATSM.
The difference in retention between Cu-ATSM and
Cu-PTSM is due to the different redox potentials of the
complexes. Minkel et al. (37) reported the redox potential of
Cu-PTSM (-208 mV), and Fujibayashi et al. (15) published
the redox potential of Cu-ATSM (-297 mV). Taniuchi et al.
(27) have reportedthatCu-PTSM is reduced atthe Complex
I site of the electron transport chain using NADH as the
electron donor. NADH has a redox potential of —315mV
(38), similar to that of Cu-ATSM. Cu-ATSM, with a redox
potential 89 mV lower than that of Cu-PTSM, is reduced less
efficiently at the Complex I site under the same conditions
(75). Under hypoxic conditions, depletion of oxygen causes
hyper-reduction of Complex I and an increase in NADH
concentration (39). This accounts for the difference in
uptake and retention between "Cu-ATSM and "Cu-PTSM.
This is observed in the EMT6 cell uptake studies, since
under hypoxic conditions, the cells would contain Complex
I with particularly high electron and/or NADH concentra
tions.
Others researchers have suggested that ubiquitous sulfhy-
dryl (SH) groups, such as glutathione (GSH), are responsible
for the retention of Cu-PTSM in normal cells (24,25,40). It
would be reasonable to assume that under hypoxic condi
tions, the retention of Cu-ATSM is due to enzymatic
reduction, most probably at the Complex I site as discussed
previously.
The washout studies further support this hypothesis.
"Cu-PTSM is reduced and trapped in all cells, and therefore
little washout from the cells is observed when the media is
changed (<11%) (Fig. 3B). With "Cu-ATSM, over 42% of
the complex is washed out from the normoxic cells after 1h.
In hypoxic cells, this value is only 27%, suggesting the
selective retention of the complex in a more reducing
cellular environment (Fig. 3A). Fujibayashi et al. (75)
confirmed this observation, in that hypoxic retention of
Cu-ATSM is a reversible phenomenon dependent only on
pO2and not on irreversible cellular damage such as mem
brane disruption.
I8F-MISO, although used for the detection of hypoxic
tissue, has two disadvantages: low cellular uptake and slow
washout from normoxic tissue (29). This is demonstrated by
the in vitro uptake studies in which, over a 1-hperiod, less
than 2% of I8F-MISOwas taken up by the EMT6 at any pO2.
After 2 h, binding of the tracer was seen to be dependent on
oxygen concentration in EMT6cells at higher cell concentra
tions (Fig. 2B). However, the more efficient uptake and
washout kinetics of "Cu-ATSM in hypoxic and normoxic
cells offers the possibility of a fast and efficient means of
detecting tumor hypoxia by PET imaging. Currently,clinical
studies involving I8F-MISOproduce images showing differ
ences between normal and hypoxic tissues but only hours
postinjection as a result of slow blood clearance and low
tumor-to-muscle ratios (29,30,41).
Both "Cu-ATSM and "Cu-PTSM
lipophilic, square planar compounds that can easily trans
verse the blood-brain barrier (14,24). Both "Cu-ATSM and
"Cu-PTSM were extracted rapidly from the blood compart
ment. The major organs such as the lung, brain and kidney
showed efficient extraction and retention of the flow tracer
"Cu-PTSM as already described (24,25). The extraction of
MCu-ATSMat l min was either equal or less efficient than
"Cu-PTSM. The heart and brain extraction of "Cu-PTSM
was followed by retention of the complex over the 40-min
are small neutral,
"Cu-ATSM INEMT6 HYPOXIC TUMORMODELS• Lewis et al.181