Mixed bis(thiosemicarbazone) ligands for the preparation of copper radiopharmaceuticals: Synthesis and evaluation of tetradentate ligands containing two dissimilar thiosemicarbazone functions

Article abstract of DOI:10.1021/jm9605703

A series of four 'mixed' bis(thiosemicarbazone) keto aldehyde derivatives containing dissimilar thiosemicarbazone functions were synthesized and evaluated as ligands for preparation of radiocopper-labeled radiopharmaceuticals. The pyruvaldehyde-based mixed bis(thiosemicarbazone) ligands CH₃C[=NNHC(S)NH₂]CH[=NNHC(S)NHMe] (4a), CH₃C[=NNHC(S)NHMe]- CH[=NNHC(S)NH₂] (4b), CH₃C[=NNHC(S)NH₂]CH[=NNHC(S)NMe₂] (4c), and CH₃C[=NNHC-(S)NHMe]CH[=NNHC(S)NMe₂] (4d) were obtained by reaction of thiosemicarbazide, N⁴-methylthiosemicarbazide, or N⁴,N⁴- dimethylthiosemicarbazide with pyruvaldehyde 2-thiosemicarbazones that had been generated by oxidative cleavage of the appropriate pyruvic aldehyde dimethyl acetal 2-thiosemicarbazone. The ⁶⁷Cu-labeled complexes of ligands 4a-d were prepared and screened in a rat model to assess the potential of each chelate as a ⁶²Cu radiopharmaceutical for imaging with positron emission tomography. In the rat model the ⁶⁷Cu complexes of ligands 4a-d exhibit significant uptake into the brain and heart after intravenous injection, following trends similar to those previously reported for the related bis(thiosemicarbazone) complexes, Cu-PTS, Cu-PTSM, and Cu-PTSM₂ (derived from pyruvaldehyde bis(thiosemicarbazone), pyruvaldehyde bis(N⁴- methylthiosemicarbazone), and pyruvaldehyde bis(N⁴,N⁴- dimethylthiosemicarbazone), respectively). Ultrafiltration studies using solutions of dog and human serum albumin reveal that the ⁶⁷Cu complexes of ligands 4a-d, like the Cu(II) complex of pyruvaldehyde bis(N⁴- methylthiosemicarbazone), interact more strongly with human albumin than dog albumin.

Full text of DOI:10.1021/jm9605703

1
32
J . Med. Chem. 1997, 40, 132-136
Notes
Mixed Bis(th iosem ica r ba zon e) Liga n d s for th e P r ep a r a tion of Cop p er
Ra d iop h a r m a ceu tica ls: Syn th esis a n d Eva lu a tion of Tetr a d en ta te Liga n d s
Con ta in in g Tw o Dissim ila r Th iosem ica r ba zon e F u n ction s
J ohn K. Lim, Carla J . Mathias, and Mark A. Green*
Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University, West Lafayette, Indiana 47907-1333
X
Received August 5, 1996
A series of four “mixed” bis(thiosemicarbazone) keto aldehyde derivatives containing dissimilar
thiosemicarbazone functions were synthesized and evaluated as ligands for preparation of
radiocopper-labeled radiopharmaceuticals. The pyruvaldehyde-based mixed bis(thiosemicar-
bazone) ligands CH
CH[dNNHC(S)NH
S)NHMe]CH[dNNHC(S)NMe
3
C[dNNHC(S)NH
] (4b), CH C[dNNHC(S)NH
] (4d ) were obtained by reaction of thiosemicarbazide, N -meth-
2
]CH[dNNHC(S)NHMe] (4a ), CH
3
C[dNNHC(S)NHMe]-
2
3
2
]CH[dNNHC(S)NMe ] (4c), and CH
2
3
C[dNNHC-
4
(
2
4
4
ylthiosemicarbazide, or N ,N -dimethylthiosemicarbazide with pyruvaldehyde 2-thiosemicar-
bazones that had been generated by oxidative cleavage of the appropriate pyruvic aldehyde
6
7
dimethyl acetal 2-thiosemicarbazone. The Cu-labeled complexes of ligands 4a -d were pre-
6
2
pared and screened in a rat model to assess the potential of each chelate as a Cu radiopharm-
6
7
aceutical for imaging with positron emission tomography. In the rat model the Cu complexes
of ligands 4a -d exhibit significant uptake into the brain and heart after intravenous injection,
following trends similar to those previously reported for the related bis(thiosemicarbazone)
2
complexes, Cu-PTS, Cu-PTSM, and Cu-PTSM (derived from pyruvaldehyde bis(thiosemi-
4
4
4
carbazone), pyruvaldehyde bis(N -methylthiosemicarbazone), and pyruvaldehyde bis(N ,N -
dimethylthiosemicarbazone), respectively). Ultrafiltration studies using solutions of dog and
6
7
human serum albumin reveal that the Cu complexes of ligands 4a -d , like the Cu(II) complex
4
of pyruvaldehyde bis(N -methylthiosemicarbazone), interact more strongly with human
albumin than dog albumin.
In tr od u ction
4
Copper-62-labeled pyruvaldehyde bis(N -methylthio-
semicarbazonato)copper(II), Cu-PTSM (Figure 1, R1 )
R2 ) NHCH3), has been widely investigated as a
generator-based radiopharmaceutical for evaluation of
regional tissue perfusion with positron emission tomog-
raphy (PET).1
-10
The Cu-PTSM radiopharmaceutical
62
is attractive as a marker of both cerebral and myocar-
dial perfusion, since following intravenous administra-
tion it affords reasonably high first-pass extraction
accompanied by prolonged microsphere-like tissue re-
tention of the copper radiolabel. Thus, Cu-PTSM
allows the 9.3 min 62Cu half-life to be exploited with
relatively long image acquisition periods designed to
afford good counting statistics, while the physical decay
of the radiolabel remains sufficiently rapid to permit
repeat imaging without excessive delay between scans.
Unfortunately, contrary to experience with the dog
F igu r e 1. General structural formula of (A) the pyruvalde-
hyde bis(thiosemicarbazone) ligands and (B) the corresponding
bis(thiosemicarbazonato)copper(II) complexes.
complexes derived from â-keto aldehydes have been
13
studied.
Two related agents, Cu-ETS and Cu-n-
PrTS, have been identified that do not exhibit the
problem of interspecies variability in albumin binding
12
that has been problematic with Cu-PTSM. This has
led to interest in exploration of copper(II) radiophar-
maceutical chemistry with bis(thiosemicarbazone) ligands
exhibiting greater structural diversity.
3
model, in humans it is found that myocardial perfusion
is underestimated with Cu-PTSM at high rates of
4
-8
flow.
The properties of the copper(II) bis(thiosemicarba-
zone) radiopharmaceuticals are known to be quite
The interspecies variability in the performance of
Cu-PTSM as a perfusion tracer appears to result from
6
2
sensitive to substitution patterns at the R1 and R2 sites
species-dependent binding of the radiopharmaceutical
to serum albumin, with the radiotracer’s interaction
13-15
(Figure 1, where R1, R2 ) NH2, NHCH3, or N(CH3)2).
Unfortunately, conventional synthetic methods (involv-
ing condensation of a â-ketoaldehyde with two equiva-
lents of a thiosemicarbazide derivative) generate only
ligands in which R1 and R2 cannot be independently
varied (i.e., R1 ) R2). To gain access to ligands in which
1
1,12
being especially strong with human serum albumin.
An extensive number of related bis(thiosemicarbazone)
X
Abstract published in Advance ACS Abstracts, December 15, 1996.
S0022-2623(96)00570-5 CCC: $14.00 © 1997 American Chemical Society

Notes
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1 133
F igu r e 2. Synthetic route successfully employed for preparation of the mixed bis(thiosemicarbazone) ligands and copper(II)
chelates.
the properties of the final Cu complex might be more
finely-tuned to optimize radiopharmaceutical biodistri-
bution and pharmacokinetics, it is desirable to develop
methods for preparation of “mixed” bis(thiosemicarba-
zone) ligands containing dissimilar thiosemicarbazone
functions (i.e., R1 * R2). We report here the develop-
ment of synthetic methods that provide access to such
mixed bis(thiosemicarbazone) radiopharmaceuticals.
acetal 2-thiosemicarbazone, 2. Our attempts to hydro-
lyze the acetal in protic acid media resulted in the
concomitant hydrolysis of the imine as well, ultimately
yielding a mixture of bis(thiosemicarbazones) analogous
to that obtained by the thiosemicarbazone 1-oxime
route. This problem was resolved through the oxidative
cleavage of the acetal with a mild Lewis acid, lithium
tetrafluoroborate, in 2% aqeous acetonitrile at roomtem-
perature over 30 min to cleanly and efficiently generate
the pyruvaldehyde 2-thiosemicarbazones 3a and 3b.
The final step in the synthesis involved a second
condensation reaction of 3 with a dissimilar thiosemi-
carbazide to yield the desired mixed bis(thio-
semicarbazone) ligands 4a -d .
Resu lts a n d Discu ssion
Cold Ch em istr y. Attempts to generate pure samples
of the mixed bis(thiosemicarbazone) ligands by methods
described in the literature16 produced unsatisfactory
results. Our attempts to selectively synthesize mixed
bis(thiosemicarbazones) by stepwise reaction of dis-
similar thiosemicarbazides with pyruvic aldehyde failed
because we were unable to cleanly isolate the â-keto
imine intermediate; reaction of pyruvaldehyde with 1
Ra d ioch em istr y. The no-carrier-added 67Cu com-
plexes of the tetradentate mixed bis(thiosemicarbazone)
ligands, 4a -d , were prepared by reaction of the 67Cu
ion with the thiosemicarbazone ligand in an ethanolic
solution of acetate buffer. The radiochemical purity of
the radiotracers was always found to exceed 98% by thin
layer chromatography. As expected, the lipophilicity of
these compounds increases with increasing methyl
substitution at the terminal amine groups of the ligand
backbone. Measured octanol/water partition coeffi-
2+
4
equiv of thiosemicarbazide or N -methylthiosemicar-
bazide resulted in condensation at both the aldehyde
and ketone sites to generate a mixture of both possible
monothiosemicarbazones and the “unmixed” bis(thio-
semicarbazone). We were also unsuccessful with the
second reported synthetic route in which a thiosemi-
carbazide is first condensed onto pyruvic aldehyde
6
7
cients, P, for the Cu complexes of mixed bis(thiosemi-
1
7
1
-oxime to generate the thiosemicarbazone 1-oxime.
carbazone) ligands 4a -d
fall between the values
Selective hydrolysis of the oxime moiety, followed by
reaction with a dissimilar thiosemicarbazide, report-
edly16 provides the desired mixed bis(thiosemicarba-
zone) ligands; however, in our hands this method
invariably resulted in the formation of a mixture of both
unmixed and mixed bis(thiosemicarbazones). NMR
studies of the reaction products showed four imine C-H
proton resonances in approximate 1:1:1:1 ratios, indi-
cating nearly random mixed and unmixed product
formation (assignments subsequently confirmed by
comparison to the authentic products). Therefore, an
alternative strategy was devised for the synthesis of the
desired mixed bis(thiosemicarbazone) ligands.
previously reported for the corresponding unmixed
complexes, Cu-PTS, Cu-PTSM, and Cu-PTSM2.13
Complex lipophilicity was found to depend somewhat
on the position of the N-alkyl group relative to the
pyruvaldehyde methyl; for example, the copper complex
of ligand 4b is slightly more lipophilic than the isomeric
complex of ligand 4a (∆ log P ) 0.16).
An im a l Stu d ies. The biodistribution of each 67Cu
complex was determined in Sprague-Dawley rats at 1,
5, and 120 min following femoral vein injection of the
radiotracer in order to screen for possible utility in
imaging the brain and/or heart. The potential clinical
6
2
utility of the Cu-bis(thiosemicarbazone) radiophar-
maceuticals as perfusion tracers derives from two
characteristic factors in their biodistribution and phar-
macokinetics. Specifically, the most promising of the
previously studied compounds provide high first-pass
tissue extraction of tracer (thereby assuring that the
The route successfully employed for selective synthe-
sis of the desired mixed bis(thiosemicarbazone) ligands
is outlined in Figure 2. One equivalent of a thiosemi-
carbazide was condensed with pyruvic aldehyde di-
methyl acetal, 1, to form the pyruvic aldehyde dimethyl

1
34 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1
Notes
Ta ble 1. Myocardial Uptake of
Cu(II)-Bis(thiosemicarbazone) Complexes in the Rat
following Intravenous Injection
Ta ble 3. Binding of Copper(II)-Bis(thiosemicarbazone)
6
7
Complexes to Serum Albumina
free (unbound) 67Cu-L determined by
ultrafiltration (%)
percentage of the injected dose in the hearta
ligand
PTSb
1 min
5 min
120 min
dog albumin
(35 mg/mL saline)
human albumin
(35 mg/mL saline)
complex (Cu-L)
2.8 ( 0.4
2.3 ( 0.1
6
7
Cu-PTSMb
Cu-4a
4
4
a
b
2.73 ( 0.15
2.84 ( 0.16
2.7 ( 0.3
2.44 ( 0.14
2.86 ( 0.01
3.4 ( 1.2
2.03 ( 0.23
2.59 ( 0.22
3.3 ( 0.9
1.05 ( 0.09
0.89 ( 0.13
40.2 ( 6.7
57.4 ( 1.8
46.5 ( 1.5
57.6 ( 2.9
25.7 ( 1.7
44.3 ( 6.4
30.3 ( 2.4
5.1 ( 1.2
10.4 ( 0.6
8.6 ( 0.5
17.2 ( 0.6
5.4 ( 0.3
41.5 ( 4.0
37.0 ( 2.8
6
7
b
67
PTSM
Cu-4b
6
6
7
7
4
4
c
d
1.59 ( 0.27
1.43 ( 0.05
0.97 ( 0.15
1.23 ( 0.17
1.27 ( 0.19
0.43 ( 0.14
Cu-4c
Cu-4d
b
67
Cu-ETSb
Cu-n-PrTSb
PTSM2
6
7
a
Values represent the mean ( standard deviation of data
b
a
obtained from three rats. From ref 13.
Amicon Centrifree (30 000 Da) ultrafiltration devices centri-
fuged at 1000g for 20 min in a 45° fixed angle rotor at 20 °C.
Values shown represent the mean ( standard deviation of 6-12
measurements. All values are corrected for nonspecific binding
of tracer to the ultrafiltration device using binding data indepen-
Ta ble 2. Brain Uptake of 67Cu(II)-Bis(thiosemicarbazone)
Complexes in the Rat Following Intravenous Injection
percentage of the injected dose in the braina
b
dently measured for tracer in protein-free saline solution. From
ligand
PTSb
1 min
5 min
120 min
ref 12.
0.46 ( 0.15
1.40 ( 0.06
1.72 ( 0.17
3.0 ( 0.7
2.79 ( 0.62
2.76 ( 0.17
3.1 ( 0.4
0.29 ( 0.05
1.37 ( 0.26
1.81 ( 0.09
3.0 ( 0.6
1.97 ( 0.12
1.80 ( 0.48
2.2 ( 0.3
4
4
a
b
1.10 ( 0.24
1.67 ( 0.17
3.2 ( 0.4
1.98 ( 0.33
1.63 ( 0.20
bis(thiosemicarbazone) radiotracers are fully consistent
67
with substituent effects reported for the Cu-PTS,
b
PTSM
67
67
13,14
Cu-PTSM, and Cu-PTSM2 series.
For both the heart and brain, levels of the 67Cu
4
4
c
d
b
PTSM2
radiolabel are essentially unchanged between 1 and 5
min postinjection using ligands 4a and 4b, while the
complexes of 4c and 4d exhibit clearance of tissue
radiotracer over the same time frame (Tables 1 and 2).
As was the case in analyzing structural effects on initial
cerebral and myocardial uptake of radiotracer, the
observed variations in the efficiency with which the
tracer is trapped by tissue correlate with whether the
NMe2 substituent is present in the ligand backbone. The
present observations that the NMe2 substituent inhibits
tracer trapping by tissue is consistent with findings from
previous biodistribution studies with radiolabeled Cu-
a
Values represent the mean ( standard deviation of data
b
obtained from three rats. From ref 13.
regional tissue distribution of radionuclide initially
maps the pattern of capillary blood flow), while also
exhibiting prolonged tissue retention of the extracted
radiolabel. The high first-pass tissue extraction of these
lipophilic 62Cu agents results from high diffusibility of
3
,14,18
the blood-borne radiotracer.
The subsequent tissue
trapping of the radiolabel results from liberation of ionic
copper by reductive intracellular decomposition of the
uncharged, square-planar Cu(II)-bis(thiosemicarba-
1
3,14
1
3,15,19,20
(II)-bis(thiosemicarbazone) complexes.
This struc-
zone) complex.
The 67Cu complexes of 4a , 4b, 4c, and 4d are all found
tural effect can be explained by the known influences
of the electron-releasing NMe2 substituent on chelate
redox potential and susceptibility to intracellular reduc-
to clear from blood rather rapidly following intravenous
administration, with only 5.7-9.0% of the injected
1
3-15
6
7
tive decomposition.
Cu radioactivity remaining in circulation at 1 min
Ultr afiltr ation Stu dies. Ultrafiltration studies were
postinjection. Tables 1 and 2 show the heart and brain
6
7
uptake of the 67Cu complexes of 4a , 4b, 4c, and 4d ,
performed with the Cu complexes of ligands 4a -d in
order to assess the amout of free (i.e., non-albumin-
associated) radiotracer present in solutions of dog or
human serum albumin that approximate the albumin
concentration of plasma. These ultrafiltration studies
expressed in each case as a percentage of the injected
radioactivity per organ. For the purposes of this study
the tissue uptake at 1 min postinjection can be assumed
to largely reflect the unidirectional first-pass tissue
extraction of tracer from blood.13 The Cu complexes
of 4a , 4b, 4c, and 4d all show significant myocardial
uptake at this 1 min time point (ranging from 1.4% to
67
67
demonstrate that the four new Cu-labeled mixed bis-
(thiosemicarbazone) complexes, like Cu-PTSM, bind
more strongly to human serum albumin than to dog
serum albumin (Table 3). In the dog albumin solutions,
albumin-bound tracer was in the range of 40-75% for
2
.8% of the injected dose), as well as the ability to
penetrate the blood-brain barrier (brain uptake also
ranging from 1.4% to 2.8% of the injected dose at 1 min
postinjection).
6
7
the Cu complexes of 4a -d ; while in human albumin
solutions of these agents, 83-95% of the radiotracer was
albumin-associated. Thus, copper radiopharmaceuticals
derived from ligands 4a -d are unlikely to overcome the
interspecies variability in radiotracer biodistribution
that somewhat hinders clinical utilization of 62Cu-
PTSM as a PET radiopharmaceutical for myocardial
perfusion imaging. Nevertheless, this derivatization
strategy may still be useful as a means of improving
the properties of the related copper(II) radiopharma-
ceuticals, Cu-ETS and Cu-n-PrTS, which are based
on the ethylgyloxal and n-propylglyoxal bis(thiosemi-
carbazone) ligand backbones and which do not exhibit
the strong affinity for human serum albumin that has
In this new series of 67Cu complexes the brain uptake
at 1 min is highest for the more lipophilic complexes in
which the ligand contains at least one NMe2 group. For
reasons that are unclear, in contrast to the brain results
the presence of the NMe2 group is associated with the
lowest myocardial uptake of radiotracer. We find this
somewhat surprising, since one might expect the ef-
ficiency of first-pass myocardial extraction to be in-
creased by the same physicochemical properties that
favor high first-pass cerebral extraction. However, the
disparate effects of the NMe2 substituent on the initial
cerebral and myocardial uptake of these mixed Cu(II)-

Notes
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1 135
a 67Cu2+/HCl solution to dryness with heating under a stream
of nitrogen and then reconstituting in 0.25 N acetate buffer
characterized the pyruvaldehyde-based copper bis(thio-
semicarbazone) complexes.
1
2
(pH 5.5). Typically, 100 µL of ethanol and ca. 0.1 mg of the
Exp er im en ta l Section
bis(thiosemicarbazone) ligand dissolved in 1-2 µL of DMSO
was added to the aqueous 67Cu-acetate (30-100 µL). The
resulting solution was diluted with saline to ca. 5% ethanol
and e0.5% DMSO (final 67Cu concentration of ∼2 µCi/µL) and
then filtered through a 0.2 µm PTFE membrane before use.
The radiochemical purity of the 67Cu-bis(thiosemicarbazone)
complexes was determined by thin-layer chromatography on
silica gel plates eluted with ethanol and was always found to
exceed 98%.
Gen er a l. Unless otherwise noted, materials obtained from
commercial suppliers were used without further purification.
Pyruvic aldehyde dimethyl acetal, lithium tetrafluoroborate,
4
4
4
thiosemicarbazide, N -methylthiosemicarbazide, and N ,N -
dimethylthiosemicarbazide were purchased from the Aldrich
Chemical Co. Whatman KSF silica gel 60A plates (without
Florisil indicator) were used for TLC analysis. Microanalysis
were performed at the Purdue Microanalysis Laboratory, and
all values were within (0.4% of the calculated compositions.
Copper-67 copper(II) chloride was obtained from Brookhaven
National Laboratory and Los Alamos National Laboratory.
Human serum albumin (essentially globulin and fatty acid
free) and dog serum albumin (essentially fatty acid free) were
purchased from Sigma Chemical Co. (St. Louis, MO). Melting
points were measured on a Gallenkamp MFB-595 capillary
Partition coefficients of the 67Cu radiotracers were measured
following 1 min of vigorous vortex mixing of 1 mL of 1-octanol
2
4
and 1 mL of isotonic TRIS-buffered saline (pH ) 7) with
approximately 0.1 µCi of the radiolabeled copper complex.
Following centrifugation at >1200g for 5 min, the octanol and
aqueous phases were sampled and counted in an automatic
well counter. A 500 µL sample of the octanol phase from this
partitioning was repartitioned two to three times with fresh
buffer to ensure that trace hydrophilic 67Cu impurities did not
alter the calculated P values. The reported log P values are
the average of the second and third extractions from three to
1
melting point apparatus and are uncorrected. H NMR spectra
were measured on either a Bruker ARX 300 spectrometer (300
MHz) or a Varian XL-500 (500 MHz) spectrometer. FAB mass
spectra were obtained from a Kratos MS50 mass spectrometer.
All experiments involving animals were carried out following
protocols approved by the Purdue Animal Care and Use
Committee. Rat biodistribution studies were performed as
1
7
four independent measurements, so that the reported log P
values represent the mean ((standard deviation) of six to eight
measurements.
Ultr a filtr a tion Stu d ies w ith 67Cu Com p lexes. Serum
albumin binding of the 67Cu-bis(thiosemicarbazone) complexes
was quantitatively evaluated by ultrafiltration as described
previously.12 Solutions of serum albumin were prepared fresh
for each experiment at 35 mg/mL normal saline and refriger-
ated when not in use. Each Amicon (Beverly, MA) Centrifree
ultrafiltration device (30 000 Da NMWL) was loaded with
1
1,13
described previously
following injection of tracer (∼2 µCi
in 0.2 mL) into the femoral vein of ether-anesthetized animals.
P r epar ation of P yr u valdeh yde Dim eth yl Acetal 2-Th io-
sem ica r ba zon es. To a stirred solution of pyruvic aldehyde
dimethyl acetal (5.00 g, 42.3 mmol) in DMF (75 mL) was added
freshly activated molecular sieves (4 Å). The appropriate
thiosemicarbazide (0.95 equiv) was then added, resulting in a
clear, pale yellow to deep amber solution. The reaction was
allowed to stir at ambient temperature for a total of 3 days.
The molecular sieves were filtered, and the DMF was removed
in vacuo. The crude material was dissolved in DMF (1-2 mL),
which was loaded onto a silica gel column (22 cm × 60 cm)
packed in diethyl ether. The column was eluted with diethyl
ether (flow rate 2 mL/min), and fractions corresponding to the
product (TLC) were pooled and evaporated to dryness under
reduced pressure. The product was then recrystallized from
diethyl ether. Isolated yields of the purified material obtained
3
00-600 µL of either an albumin solution (35 mg/mL saline)
or normal saline (control) within 2 min of mixing with the
6
7
Cu complex. Typically, 1 mL of protein solution was mixed
with 1-2 µL of 67Cu-bis(thiosemicarbazone) solution. The
Centrifree devices were loaded and immediately (<1 min)
centrifuged in a Sorvall RC2-B refrigerated centrifuge (20
°
C) with a SS-34 45° fixed angle rotor at 1000g for 20 min.
Cu-bis(thiosemicarbazone) complex concentrations (cpm/mL)
6
7
in the unfiltered protein solutions and their ultrafiltrates were
determined by counting measured aliquots in a Packard
Autogamma 5530 automatic γ counter. The percentage of free
2
1
were between 60 and 80%.
(
unbound) 67Cu-bis(thiosemicarbazone) complex was calcu-
P r ep a r a tion of P yr u va ld eh yd e 2-Th iosem ica r ba zon es.
Pyruvaldehyde dimethyl acetal 2-thiosemicarbazone and lithium
tetrafluoroborate (2 equiv) were combined, and approximately
lated as
6
7
1
5 mL of a 2% aqueous solution of acetonitrile was added at
(
Cu-L concentration in protein ultrafiltrate)
ambient temperature to form a clear, pale yellow solution.
Approximately 30 min into the reaction, the formation of a
white precipitate was observed. Completion of the reaction
[(67
]
Cu-L concentration in unfiltered protein solution)
× 100
6
7
(
Cu-L concentration in saline ultrafiltrate)
was confirmed by thin-layer chromatography (Et
2
O, silica gel).
[
(
Cu-L concentration in unfiltered saline solution)]
6
7
A saturated solution of sodium carbonate (30 mL) was added
to the suspension, resulting in an intense, bright yellow
homogeneous solution which was extracted with diethyl ether
Ack n ow led gm en t. This work was supported by a
grant from the National Heart, Lung, and Blood Insti-
tute of the Public Health Service (R41-HL53067).
(
3 × 50 mL). The ether extracts were combined, dried over
magnesium sulfate, filtered, and solvent removed in vacuo.
The solid material was washed with diethyl ether and pen-
tanes to leave behind white to pale yellow flaky, needle-like
crystals. Isolated yields of the purified material were generally
Su p p or tin g In for m a tion Ava ila ble: Complete tables of
rat biodistribution data for each Cu compound, calculated
as both a percentage of the injected dose per gram tissue (wet
weight) and as a percentage of the injected dose per organ (4
pages). Ordering information is given on any current mast-
head page.
67
2
2
>
90%.
P r ep a r a tion of th e Mixed Bis(th iosem ica r ba zon es)
fr om P yr u va ld eh yd e 2-Th iosem ica r ba zon es. Freshly ac-
tivated molecular sieves (4 Å) were added to a DMF solution
(
10 mL) of pyruvaldehyde 2-thiosemicarbazone. A clear yellow
Refer en ces
solution resulted when the appropriate dissimilar thiosemi-
carbazide was added. The reaction mixture was heated (60
°
proximately 30 min prior to filtering of the solution to remove
the molecular sieves. DMF was removed from the clear yellow
filtrate in vacuo. The resulting solid product was recrystal-
lized from DMF and washed with ethanol and pentanes.
Isolated yields of the mixed bis(thiosemicarbazones) were
(
1) Green, M. A. 62Cu-PTSM: A Generator-Based PET Radiophar-
maceutical for Myocardial Perfusion Imaging. In Cardiac Positron
Emission Tomography; Schwaiger, M., Ed.; Kluwer Academic
Publishers: Boston, 1996; pp 161-172.
C) for 6 h and then cooled at room temperature for ap-
(2) Green, M. A.; Mathias, C. J .; Welch, M. J .; et al. [62Cu]-Labeled
4
Pyruvaldehyde Bis(N -methylthiosemicarbazonato) Copper-
(II): Synthesis And Evaluation As A Positron Emission Tomog-
raphy Tracer For Cerebral And Myocardial Perfusion. J . Nucl.
Med. 1990, 31, 1989-1996.
3) (a) Shelton, M. E.; Green, M. A.; Mathias, C. J .; Welch, M. J .;
Bergmann, S. R. Assessment of Regional Myocardial and Renal
Blood Flow Using Copper-PTSM and Positron Emission Tomog-
2
3
between 50-60%.
(
Ra d iotr a cer Syn th esis. The 67Cu-labeled copper(II) bis-
thiosemicarbazone) complexes were prepared by evaporating
(

1
36 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1
Notes
raphy. Circulation 1990, 82, 990-997. (b) Herrero, P.; Markham,
(17) Measured log P values were 1.35 ( 0.01, 1.51 ( 0.01, 1.97 (
0.01, and 2.62 ( 0.04 for the 67Cu complexes of 4a , 4b, 4c, and
J .; Weinheimer, C. J .; et al. Quantification of Regional Myocar-
6
2
dial Perfusion with Generator-Produced Cu-PTSM and Positron
4
d , respectively.
Emission Tomography. Circulation 1993, 87, 173-183.
(
18) Shelton, M. E.; Green, M. A.; Mathias, C. J .; Welch, M. J .;
Bergmann, S. R. Kinetics of Copper-PTSM in Isolated Hearts:
A Novel Tracer for Measuring Blood Flow with Positron Emis-
sion Tomography. J . Nucl. Med. 1989, 30, 1843-1847.
(
4) (a) Bergmann, S. R.; Herrero, P.; Hartman, J . J .; Anderson, C.
J .; Welch, M. J .; Green, M. A. Use and Limitations of Copper-
6
2 Pyruvaldehyde Bis(methylthiosemicarbazone) and Positron
Emission Tomography for Measuring Regional Myocardial Per-
fusion in Human Subjects. Circulation 1993, 88, 171 (abstract).
(19) (a) Fujibayashi, Y.; Taniuchi, H.; Wada, K.; Yonekura, Y.;
Konishi, J .; Yokoyama, A. Differential Mechanism of Retention
(
b) Herrero, P.; Hartman, J . J .; Green, M. A.; Anderson, C. J .;
4
Welch, M. J .; Markham, J .; Bergmann, S. R. Assessment of
Regional Myocardial Perfusion with Generator-Produced Cu-
PTSM and PET in Human Subjects. J . Nucl. Med. 1996, 37,
of Cu-Pyruvaldehyde-Bis(N -methylthiosemicarbazone) (Cu-
6
2
PTSM) by Brain and Tumor: A Novel Radiopharmaceutical for
Positron Emission Tomography Imaging. Ann. Nucl. Med. 1995,
1
294-1300.
9
, 1-5. (b) Taniuchi, H.; Fujibayashi, Y.; Okazawa, H.; Yoneku-
(
(
(
5) Beanlands, R. S.; Muzik, O.; Mintun, M. A.; Mangner, T.; Lee,
K.; Petry, N. ; Hutchins, G. D.; Schwaiger, M. The Kinetics of
Copper-62-PTSM in the Normal Human Heart. J . Nucl. Med.
4
ra, Y.; Konishi, J .; Yokoyama, A. Cu-Pyruvaldehyde-Bis(N
-
methylthiosemicarbazone) (Cu-PTSM), a Metal Complex with
Selective NADH-Dependent Reduction by Complex I in Brain
Mitochondria: A Potential Radiopharmaceutical for Mitochon-
dria-Function Imaging with Positron Emission Tomography
(PET). Biol. Pharm. Bull. 1995, 18, 1126-1129.
1
992, 33, 684-690.
6) Melon, P. G.; Brihaye, C.; Degueldre, C.; Guillaume, M.; Czi-
chosz, R.; Rigo, P.; Kulbertus, H.; Comar, D. Myocardial Kinetics
of Potassium-38 in Humans and Comparison with Copper-62-
PTSM. J . Nucl. Med. 1994, 35, 1116-1122.
(
20) Baerga, I.; Maickel, R. P.; Green, M. A. Subcellular Distribution
of Tissue Radiocopper Following Intravenous Administration of
7) Beanlands, R. S. B.; Muzik, O.; Hutchins, G. D.; Woulfe, E. R.;
Schwaiger, M. Heterogeneity of Regional Nitrogen-13-Labeled
Ammonia Tracer Distribution in the Normal Human Heart:
Comparison with Rubidium-82 and Copper-62-Labeled PTSM.
J . Nucl. Cardiol. 1994, 1, 225-235.
8) Tadamura, E.; Tamaki, N.; Okazawa, H.; Fujibayashi, Y.; Kudoh,
T.; Yonekura, Y.; Magata, Y.; Nohara, R.; Sasayama, S.; Konishi,
J . Generator-Produced Copper-62-PTSM as a Myocardial PET
Perfusion Tracer Compared with Nitrogen-13-Ammonia. J . Nucl.
Med. 1996, 37, 729-735.
6
7
Cu-Labeled Cu-PTSM. Nucl. Med. Biol. 1992, 19, 697-701.
21) P yr u va ld eh yd e d im eth yl a ceta l 2-th iosem ica r ba zon e (2a )-
(
1
:
d
mp 125-126 °C; TLC R
) δ 10.22 (s, 1H, NH), 8.24 (s, 1H, NH
)), 3.27 (s, 6H, CH(OCH
CH ). P yr u va ld eh yd e d im et h yl a cet a l 2-N
f
0.30 (Et
2
O, silica gel); H NMR (DMSO-
), 7.72 (s, 1H, NH ),
)), 1.86 (s, 3H,
-m et h ylt h io-
0.51 (Et O, silica
) δ 10.21 (s, 1H, NH), 8.35 (m, 1H,
), 4.52 (s, 1H, CH(OCH )), 3.28 (s, 6H, CH(OCH )), 2.96
d, 3H, NHCH ), 1.81 (s, 3H, CH ). P yr u va ld eh yd e d im eth yl
a ceta l 2-N -d im eth ylth iosem ica r ba zon e (2c): mp 84-85 °C;
(
6
2
2
4.56 (s, 1H, CH(OCH
3
3
4
3
sem ica r ba zon e (2b): mp 87-88 °C; TLC R
gel); H NMR (DMSO-d
NHCH
f
2
1
6
(
9) (a) Okazawa, H.; Yonekura, Y.; Fujibayashi, Y.; Nishizawa, S.;
Magata, Y.; Ishizu, K.; Tanaka, F.; Tsuchida, T.; Tamaki, N.;
Konishi, J . Clinical Application and Quantitative Evaluation of
Generator-Produced Copper-62-PTSM as a Brain Perfusion
Tracer for PET. J . Nucl. Med. 1994, 35, 1910-1915. (b) Oka-
zawa, H.; Yonekura, Y.; Fujibayashi, Y.; Mukai, T.; Nishizawa,
S.; Magata, Y.; Ishizu, K.; Tamaki, N.; Konishi, J . Measurement
of Regional Cerebral Blood Flow with Copper-62-PTSM and a
Three-Compartment Model. J . Nucl. Med. 1996, 37, 1089-1093.
3
3
3
(
3
3
4
1
TLC R
NH), 4.51 (s, 1H, CH(OCH
6H, N(CH ) ), 1.89 (s, 3H, CH ).
f
0.68 (Et
2
O, silica gel); H NMR (DMSO-d
6
) δ 9.32 (s, 1H,
)), 3.20 (s,
3
)), 3.28 (s, 6H, CH(OCH
3
3
2
3
22) P yr u va ld eh yd e 2-t h iosem ica r b a zon e (3a ): 1H NMR
(
(
(
1
DMSO-d
s, 1H, NH), 9.33 (s, 1H, CHO), 8.78 (br s, 1H, NH
H, NH ), 1.92 (s, 3H, CH ). P yr u va ld eh yd e 2-N -m eth yl-
th iosem ica r ba zon e (3b): mp 173 °C; TLC R 0.62 (Et O, silica
) δ 11.18 (s, 1H, NH), 9.35 (s, 1H, CHO),
), 3.01 (d, 3H, NHCH ), 1.93 (s, 3H, CH ).
(23) 2-Oxopr opan al 1-N -m eth ylth iosem icar bazon e 2-th iosem i-
ca r ba zon e (4a ): mp 208-209 °C; H NMR (DMSO-d ) δ 11.70
s, 1H, NH), 10.34 (s, 1H, NH), 8.41 (m, 1H, NHCH
H, NH ), 7.88 (s, 1H, NH ), 7.64 (s, 1H, NdCH), 2.98 (d, 3H,
NCH , J ) 5.0 Hz), 2.15 (s, 3H, CH ); FAB mass spectrum, 233
M + H). Anal. (C ‚0.30H O) C, H, N. 2-Oxop r op a n a l
-N -m e t h ylt h iose m ica r b a zon e -1-t h iose m ica r b a zon e
6
): mp 171 °C; TLC R
f
0.58 (Et
2
O, silica gel) δ 11.12
(
10) Young, H.; Carnochan, P.; Zweit, J .; Babich, J .; Cherry, S.; Ott,
2
), 8.38 (br s,
4
R. Evaluation of Copper(II)-Pyruvaldehyde Bis(N -methylthio-
4
2
3
semicarbazone) for Tissue Blood Flow Measurements Using a
Trapped Tracer Model. Eur. J . Nucl. Med. 1994, 21, 336-341.
11) Mathias, C. J .; Bergmann, S. R.; Green, M. A. Development and
Validation of a Solvent Extraction Technique for Determination
of Cu-PTSM in Blood. Nucl. Med. Biol. 1993, 20, 343-349.
12) Mathias, C. J .; Bergmann, S. R.; Green, M. A. Species-Dependent
Binding of Copper(II) Bis(thiosemicarbazone) Radiopharmaceu-
ticals to Serum Albumin. J . Nucl. Med. 1995, 36, 1451-1455.
13) J ohn, E.; Green, M. A. Structure-Activity Relationships for
Metal-Labeled Blood Flow Tracers: Comparison of Ketoaldehyde
Bis(thiosemicarbazonato)copper(II) Derivatives. J . Med. Chem.
f
2
1
(
(
(
gel); H NMR (DMSO-d
9.01 (m, 1H, NHCH
6
3
4
3
3
1
6
(
1
3
), 8.37 (s,
2
2
3
3
(
2
6
H
12
N
6
S
2
2
4
1
990, 33, 1764-1770.
1
(
14) Mathias, C. J .; Welch, M. J .; Raichle, M. E.; et al. Evaluation of
a Potential Generator-Produced PET Tracer for Cerebral Per-
fusion Imaging: Single-Pass Cerebral Extraction Measurements
and Imaging with Radiolabeled Cu-PTSM. J . Nucl. Med. 1990,
(4b): mp 229-230 °C; H NMR (DMSO-d
10.38 (s, 1H, NH), 8.53 (m, 1H, NHCH ), 8.31 (s, 1H, NH
6
) δ 11.69 (s, 1H, NH),
), 7.90
3
2
(s, 1H, NH ), 7.65 (s, 1H, NdCH), 2.98 (d, 3H, NHCH , J ) 4.0
2
3
Hz), 2.12 (s, 3H, CH
3
); FAB mass spectrum, 233 (M + H). Anal.
3
1, 351-359.
4
(C
6
H
12
N
6
S
2
‚0.30H
2
O) C, H, N. 2-Oxop r op a n a l 1-N -d im eth -
(
15) (a) Minkel, D. T.; Saryan, L. A.; Petering, D. H. Structure-
Function Correlations in the Reaction of Bis(thiosemicarbazone)
Copper(II) Complexes with Ehrlich Acites Tumor Cells. Cancer
Res. 1978, 38, 124-129. (b) Winkelmann, D. A.; Bermke, Y.;
Petering, D. H. Comparative Properties of the Antineoplastic
Agent 3-ethoxy-2-oxobutyraldehyde Bis(thiosemicarbazone) Cop-
per(II) and Related Chelates: Linear Free Energy Relationships.
Bioinorg. Chem. 1974, 3, 261-277.
ylth iosem ica r ba zon e 2-th iosem ica r ba zon e (4c): mp 192 °C;
1
H NMR (DMSO-d
s, 1H, NH ), 7.81 (s, 1H, NH
N(CH ), 1.94 (s, 3H, CH ); FAB mass spectrum, 247 (M + H).
Anal. (C ‚1.10 H
d im eth ylth iosem ica r ba zon e 2-(N -m eth ylth iosem ica r ba -
6
) δ 11.17 (s, 1H, NH), 10.40 (s, 1H, NH), 8.37
(
2
2
), 7.76 (s, 1H, NdCH), 3.24 (s, 6H,
3
)
2
3
4
7
H
14
N
6
S
2
2
O) C, H, N. 2-Oxop r op a n a l 1-N -
4
1
zon e (4d ): mp 211-212 °C; H NMR (DMSO-d
NH), 10.42 (s, 1H, NH), 8.47 (m, 1H, NHCH
NdCH), 3.25 (s, 6H, N(CH ), 2.98 (d, 3H, NHCH
.08 (s, 3H, CH ); FAB mass spectrum, 261 (M + H). Anal.
‚0.60H O) C, H, N.
6
) δ 11.21 (s, 1H,
), 7.75 (s, 1H,
, J ) 4.3 Hz),
(
16) (a) Barry, C.; Conalty, M. L.; O’Sullivan, J . F. Anticancer Activity
3
of Bisthiosemicarbazones of Methylglyoxal. Cancer Res. 1966,
3
)
2
3
2
6, 2165-2168. (b) Barry, V. C.; Byrne, J .; Conalty, M. L.;
2
3
O’Sullivan, J . F. Anticancer agentssII Synthesis and Antimy-
cobacterial Activity of Some Novel Methylglyoxal Derivatives.
Proc. R. Ir. Acad. 1967, 65, 269-284. (c) Barry, V. C.; Conalty,
M. L.; O'Callaghan, C. N.; Twomey, D. Anticancer AgentssIII
Synthesis and Anticancer Activity of Some Bis(thiosemicarba-
zones) and Thiosemicarbazides. Proc. R. Ir. Acad. 1967, 65, 309-
(C
8
H
16
N
6
S
2
2
(
24) Bates, R. G.; Vega, C. A.; White, D. R. Standards for pH Meas-
urements in Isotonic Saline Media of Ionic Strength I)0.16.
Anal. Chem. 1978, 50, 1295-1300.
3
24.
J M9605703