First organometallic inhibitors for human thymidine kinase: Synthesis and in vitro evaluation of rhenium(I)- and technetium(I)-tricarbonyl complexes of thymidine

Article abstract of DOI:10.1016/S0022-328X(02)02100-9

Six 5′-carboxamide derivatives of 5′-aminothymidine have been synthesized with alkyl chains of various length and a tridentate, imino diacetic acid based chelating system. The thymidine analogues have been reacted with the precursor fac-[M(H₂O)₃(CO)₃]⁺ (M=^99mTc, Re) in aqueous media to form, water-soluble and stable organometallic complexes in good yields. ¹H-NMR and IR spectroscopic analyses confirmed in all cases the tridentate complexation of the metal-tricarbonyl fragment exclusively via the tridentate chelates, and no unspecific interaction with other functional groups of the pharmacophor. The organometallic rhenium-nucleoside complexes have been tested in vitro for competitive inhibition of human cytosolic thymidine kinase (hTK1) and herpes simplex virus thymidine kinase type 1 (HSV1-TK). In case of hTK1 it could be observed, that the inhibition capacity of the complexes improved with increasing spacer length. On the other hand, all six complexes showed no or only slight inhibition of the HSV1-TK. The corresponding radioactive technetium-99m complexes have been prepared and challenged for stability in physiological phosphate buffer and human serum albumin at 37 °C for 24 h. Only minor decomposition of the complexes could be detected under these conditions proving the high kinetic inertness and/or stability of these complexes.

Full text of DOI:10.1016/S0022-328X(02)02100-9

Journal of Organometallic Chemistry 668 (2003) 67ꢂ74
/
www.elsevier.com/locate/jorganchem
First organometallic inhibitors for human thymidine kinase: synthesis
and in vitro evaluation of rhenium(I)- and technetium(I)-tricarbonyl
complexes of thymidine
R. Schibli a,*, M. Netter a, L. Scapozza b, M. Birringer b, P. Schelling b, C. Dumas a,
J. Schoch a, P.A. Schubiger a
a Center for Radiopharmaceutical Science ETH-PSI-USZ, Paul Scherrer Institute, 5232 Villigen, Switzerland
b Department of Applied Biosciences of the ETH, 8056 Zurich, Switzerland
¨
Received 5 September 2002
Abstract
Six 5?-carboxamide derivatives of 5?-aminothymidine have been synthesized with alkyl chains of various length and a tridentate,
imino diacetic acid based chelating system. The thymidine analogues have been reacted with the precursor fac-[M(H2O)3(CO)3]ꢀ
(Mꢁ99m
spectroscopic analyses confirmed in all cases the tridentate complexation of the metal-tricarbonyl fragment exclusively via the
/
Tc, Re) in aqueous media to form, water-soluble and stable organometallic complexes in good yields. H-NMR and IR
1
tridentate chelates, and no unspecific interaction with other functional groups of the pharmacophor. The organometallic rheniumꢂ
/
nucleoside complexes have been tested in vitro for competitive inhibition of human cytosolic thymidine kinase (hTK1) and herpes
simplex virus thymidine kinase type 1 (HSV1-TK). In case of hTK1 it could be observed, that the inhibition capacity of the
complexes improved with increasing spacer length. On the other hand, all six complexes showed no or only slight inhibition of the
HSV1-TK. The corresponding radioactive technetium-99m complexes have been prepared and challenged for stability in
physiological phosphate buffer and human serum albumin at 37 8C for 24 h. Only minor decomposition of the complexes could
be detected under these conditions proving the high kinetic inertness and/or stability of these complexes.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Rhenium(I); Technetium(I); Thymidine analogues; Human thymidine kinase; Viral thymidine kinase
1. Introduction
of therapeutic gene transfer in vivo. These reporter
genes can be coupled with a therapeutic gene of interest
to indirectly monitor the expression of the therapeutic
gene. In case of HSV1-TK a selective targeting is
facilitated due to the relaxed substrate specificity of
the enzyme as compared with mammalian thymidine
kinase. In both respects (tumor proliferation and gene
therapy monitoring) radioactive labeled probes allow a
non-invasive diagnosis and early assessment of tumor
growth and response to therapeutic approaches. Typi-
cally, either positron-labeled substrates for intracellular
enzymes or positron-labeled ligands for cellular recep-
tors are used as radioactive probes. However, there
would be a great interest to develop Tc-99m based
diagnostic probes because of the favorable decay
Nucleoside kinases play a pivotal role in the use of
nucleosides for cancer and antiviral therapy [1]. In this
context the human cytosolic thymidine kinase (hTK1)
and the herpes simplex virus type 1 thymidine kinase
(HSV1-TK) are important target enzymes. Proliferating
cells reveal a dramatically increased hTK1 activity
compared to quiescent cells. These characteristics form
the base of a selective targeting of growing tumor cells.
On the other hand, introduction of reporter genes, such
as for the HSV1-TK has emerged as a very powerful tool
to monitor the delivery, magnitude, and time variation
* Corresponding author. Tel.:
3102849
E-mail address: roger.schibli@psi.ch (R. Schibli).
ꢀ
/
41-56-3102837; fax:
ꢀ/41-56-
properties (g, 140 keV, t1/2
availability of this radionuclide. In addition, due to the
ꢁ6 h), low costs and easy
/
0022-328X/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 2 - 3 2 8 X ( 0 2 ) 0 2 1 0 0 - 9

68
R. Schibli et al. / Journal of Organometallic Chemistry 668 (2003) 67ꢂ74
/
188W/188Re-generator technique (Re-188: bꢃ, 2.1 MeV,
t1/2 17 h) a therapeutic pendant of technetium is
relative to residual solvent protons as a reference. IR
spectra were recorded on a PerkinꢂElmer FT-IR 16PC
ꢁ
/
/
nowadays also readily available. Particularly organo-
metallic technetium(I) and rhenium(I) modified analo-
gues of thymidine could be attractive candidates for
gene therapy monitoring and diagnosis/therapy of pro-
liferating tumor cells. The high kinetic inertness and
relatively small size of e.g. the technetium(I)- and
rhenium(I)-tricarbonyl core as compared to common
Tc(V)/Re(V) complexes of the Werner type is a decisive
advantage for bio-medical purposes due to increased in
vivo stability and reduced interference with biological
activity.
using KBr pellets. Mass and elemental analyses were
performed at the Swiss Federal Institute of Technology,
ETH, Zurich.
2.1. In vitro HPLC assay
Human cytosolic TK was assayed by using a protocol
published by Pilger et al. [8] with some modifications.
Briefly: hTK1 (2ꢂ5 mg) was incubated at 37 8C for 15
/
min in a mixture (75 ml) containing 50 mM Tris, pH 7.4,
5 mM ATP, 5 mM MgCl2, 1 mM deoxythymidine and
Herein, we report the functionalization, synthesis and
structural characterization of a series of novel 5?-
aminothymidine analogues and their corresponding
technetium-99m(I)/rhenium(I)-tricarbonyl labeled com-
plexes. The biological characterization and evaluation of
the organometallic rhenium compounds were performed
in vitro against hTK1 respectively HSV1 TK. The Tc-
99m analogues have been challenged in vitro in physio-
logical phosphate buffer and human serum albumin.
different concentrations of the rhenium complexes (0.5ꢂ
/
2 mM). The reaction was terminated by adding EDTA
(2.5 mM, 675 ml) and analyzed by HPLC. 0.2 M
NaH2PO4, 25 mM tetrabutylammonium hydrogensul-
fate and 3% MeOH were used as the mobile phase and
the flow rate was 1.1 ml minꢃ1. A Macherey-Nagel C-18
reversed phase column (10 mm, 150ꢄ/4.1 mm) was used.
UV-absorption was measured at 254 nm. Samples
without substrate (thymidine) or without inhibitor
were used as blank reactions. The formation of adeno-
sine diphosphate and thymidine monophosphate was
monitored qualitatively. HSV1-TK was assayed under
the same reaction conditions.
2. Experimental
Solvents for syntheses were purchased from Aldrich
Chemical Co. or Fluka, Buchs, Switzerland and were
dried according to standard methods. The organome-
tallic precursor [NEt4]2[ReBr3(CO)3] and the radioactive
precursor [99mTc(H2O)3(CO)3]ꢀ were prepared as pre-
viously reported [2,3]. 5?-Amino-thymidine [4] and 2-
hydroxy-5-amino-pentanoic acid 13 [5] were synthesized
according to the literature. The g-cyano-3-phenylbuta-
noic acid 10 was synthesized starting from ethyl-a-
carbethoxy-g-cyano-b-phenyl-butyrate [6]. Compounds
2.2. Syntheses of compounds 6ꢂ9
/
The syntheses of compounds 6ꢂ
according to the following, general procedure: Com-
pounds 2ꢂ5 (one equivalent) were dissolved in DMF.
/
9 was performed
/
N,N?-dicyclohexyl carbodiimide (DCC; 1.1 equivalent)
and N-hydroxy succinimide (NHS; 1.1 equivalent) were
added. The reactions were stirred at 60 8C for 2 h. The
formation of the active ester was monitored by TLC. 5?-
Aminothymidine (1 equivalent) was added and the
reaction stirred over night at 40 8C. When the reaction
has finished (monitored by TLC) the solvent was
removed under vacuum, and the crude product was
2ꢂ/5 have been synthesized by a previously published
procedure [7]. Na[99mTcO4] was eluted from
a
99Mo/99mTc generator (Mallinckrodt-Tyco, Petten,
Netherlands) using 0.9% saline. HPLC analyses of the
rhenium and technetium-99m complexes were per-
formed on a Merck-Hitachi L-7000-system equipped
with an L-7400 tunable absorption detector and a
Berthold LB 506 B radiometric detector using a
Macherey-Nagel C-18 reversed phase column (10 mm,
purified by chromatography on silicagel (EtOAcꢂ
/
CH2Cl2ꢂMeOH: 8/1/1). The saponification of the meth-
/
yl ester was performed in 1 M NaOH for 3 h at room
temperature (r.t.) and monitored by HPLC. Analytical
1
data for 6: H-NMR (D2O): d 1.7 (s, 3H), 1.8 (m, 2H),
150ꢄ/4.1 mm). HPLC solvents: Aqueous 0.05 M TEAP
2.2 (m, 2H), 2.2 (m, 2H), 3.2 (m, 2H), 3.3 (d, 2H), 3.8 (q,
1H), 4.0 (s, 2H), 4.2 (m, 1H), 6.1 (t, 1H), 7.3 (s, 1H), 13C-
NMR (Me2SO): d 168.1, 167.2, 160.0, 146.4, 131.9,
105.8, 81.0, 79.7, 67.8, 67.6, 46.5, 36.2, 34.6, 28.5, 21.9,
8.3; HPLC: Rt: 9.3 min.
(triethylammonium phosphate) buffer, pH 2.25 (solvent
A), MeOH (solvent B). The HPLC system started with
100% of A from 0 to 3 min. The eluent switched at 3 min
to 75% A and 25% B and at 9 min to 66% A and 34% B
followed by a linear gradient 66% A/34% B to 100% B
from 9 to 20 min. The gradient remained at 100% B for 2
min before switching back to 100% A. The flow rate was
1 ml minꢃ1. Nuclear magnetic resonance spectra were
recorded on a 300 MHz Varian Gemini 2000 spectro-
meter. The 1H and 13C chemical shifts are reported
Analytical data for 7: 1H-NMR (D2O): d 1.1 (m,
12H), 1.4ꢂ
/
1.6 (m, 4H), 1.7 (s, 3H), 2.1ꢂ
/2.2 (m, 4H),
2.9ꢂ
/
3.0 (m, 10H), 3.1 (s, 4H), 3.3 (m, 2H), 3.8 (m, 1H),
4.2 (m, 1H), 6.0 (t, 1H), 7.3 (s, 1H); 13C-NMR
(CD3OD): d 174.3, 169.1, 165.1, 151.1, 137.2, 110.5,
85.9, 85.1, 71.8, 71.5, 57.4, 41.1, 38.6, 34.8, 24.9, 22.3,

R. Schibli et al. / Journal of Organometallic Chemistry 668 (2003) 67ꢂ
/
74
69
1
10.0 min. Analytical data for 8: H-
11.1; HPLC: Rtꢁ
/
6.2 (t, 1H), 7.5 (s, 1H); 13C-NMR (D2O): d 176.6, 169.3,
166.9, 152.1, 139.1, 136.9, 112.0, 85.8, 85.0, 71.7, 70.9,
NMR (CD3OD): d 1.4 (m, 6H), 1.6ꢂ
/
1.8 (m, 4H), 2.0 (s,
3H), 2.3 (m, 4H), 2.8 (m, 2H), 3.5 (d, 2H), 3.8 (s, 4H),
4.0 (m, 1H), 4.3 (m, 1H), 6.3 (t, 1H), 7.6 (s, 1H); 13C-
NMR (CD3OD): d 175.5, 172.2, 165.1, 151.2, 138.2,
136.1, 110.7, 85.9, 84.5, 72.6, 71.5, 54.6, 54.1, 41.3, 39.0,
57.7, 55.7, 38.2, 31.8, 20.1, 12.9; HPLC: Rtꢁ
/
9.9 min.
2.5. Syntheses of the complexes 16aꢂ21a
/
36.0, 29.2ꢂ
data for 9: 1H-NMR (D2O): d 1.2 (m, 12H), 1.5 (m, 2H),
1.7 (m, 2H), 1.9 (s, 3H), 2.2 (m, 2H), 2.3ꢂ2.4 (m, 4H),
3.4 (d, 2H), 3.7 (s, 4H), 4.1 (m, 1H), 4.4 (m, 1H), 6.3 (t,
1H), 7.5 (s, 1H); 13C-NMR (CD3OD): d 177.6, 170.7,
166.5, 151.7, 137.5, 111.6, 85.4, 84.4, 71.6, 57.1, 56.3,
/
25.3, 12.2; HPLC: Rtꢁ
/
10.6 min. Analytical
The complexes 16aꢂ
/
21a were prepared according to
the following general procedure: 900 ml of a solution of
/
fac-[99mTc(OH2)3(CO)3]ꢀ (1.8ꢂ
8 GBq) and 100 ml of a
/
10ꢃ3 M solution of the corresponding ligand in PBS
buffer (0.1 M NaCl/0.05 M sodium phosphate buffered,
pH 7.4) were placed in a 10-ml glass vial under nitrogen.
The vial was sealed and the reaction heated to 75 8C for
60 min and cooled on an ice bath. The complex
formation was checked by HPLC.
40.9, 38.1, 35.9, 32.7, 28.1ꢂ
/
29.2, 25.5, 11.7; HPLC: Rtꢁ
/
16.9 min.
2.3. Synthesis of compound 11
2.6. Syntheses of complexes 16bꢂ21b
/
The g-cyano-3-phenylbutanoic acid 10 (1.7 mmol)
was solved in DMF (10 ml) and DCC (1.8 mmol) and
NHS (1.8 mmol) were added. The solution was stirred
for 90 min at 50 8C. 5?-Aminothymidine (1.6 mmol) was
added in portions to the reaction solution. The DMF
was removed in vacuum and the product was purified by
The Re(CO)3ꢂ
according to the following general procedure: (NE-
t4)2[ReBr3(CO)3] and the ligands 6ꢂ9, 12 and 15 in
equimolar amounts were solved in a H2OꢂMeOH
mixture (1/1) (5 ml) and stirred at 60 8C till the
disappearance of the starting material (monitored by
means of HPLC). The solution was evaporated to
dryness and resolved in a small amount of water. The
resulting complexes were purified by chromatography
on SepPak†-cartridge (Waters, C18-cartridges) using a
/
thymidine-complexes were synthesized
/
/
chromatography on silica gel (EtOAcꢂMeOH: 9/1) and
/
checked via 1H-NMR spectroscopy. Yield: 284 mg
1
(52%). H-NMR (CD3OD): d 2.0 (d, 3H), 2.3 (m, 2H),
2.8 (m, 4H), 3.5 (m, 3H), 3.9 (m, 1H), 4.3 (m, 1H), 6.2
(m, 1H), 7.4ꢂ7.5 (m, 5H), 7.6 (s, 1H). The cyano group
/
was reduced in dry MeOH with HCl conc. (100 ml) and
Pd/C 10%. H2 was bubbled through the solution over
night. The solution was filtered over celite and the
product (301 mg, 98%) was used without further
purification. 1H-NMR (CD3OD): d 2.0 (d, 3H), 2.1
(m, 2H), 2.3 (m, 2H), 2.7 (m, 4H), 3.5 (m, 3H), 3.9 (m,
waterꢂMeOH gradient. Analytical data for complex
/
1
16b: Yield: 64 mg (76%); H-NMR (D2O): d 1.12 (t,
4H), 1.78 (s, 3H), 1.8 (m, 2H), 2.2 (m, 2H), 2.2 (m, 2H),
3.0 (q, 3H), 3.2 (m, 2H), 3.4 (d, 2H), 3.4ꢂ3.6 (AB, 4H),
/
3.8 (q, 1H), 4.2 (m, 1H), 6.1 (t, 1H), 7.4 (s, 1H), 13C-
NMR (D2O): d 192.4 (CO), 191.8 (CO), 178.4, 170.7,
162.1, 146.9, 133.1, 106.8, 81.0, 79.7, 66.8, 63.6, 57.5,
47.2, 36.6, 33.5, 27.9, 16.1, 6.3, 1.9; Anal. Calc. for
1H), 4.3 (m, 1H), 6.2 (m, 1H), 7.4ꢂ/7.5 (m, 5H), 7.6 (s,
1H).
C21H24N4O12Re(C8H20N)0.3Na3.2Br2.5
27.23; H, 3.03; N, 5.83. Found: C, 27.51; H, 3.98; N,
5.72%; IR (KBr): 3750 br, 2014 s, 1890 vs, 1684 s, cmꢃ1
×
/
0.5H2O:
C,
2.4. Syntheses of compounds 12 and 15
;
To a solution of 11 or 14 (0.6 mmol) in MeOH, NEt3
(1.9 mmol) and methylbromoacetate (1.3 mmol) was
added drop wise under an atmosphere of nitrogen. The
reaction mixture was refluxed over night, the solvent
removed and the product was purified by chromato-
HPLC: Rt: 18.5 min; MS (ESI): m/z (%) 709 (60), 711
(100) [Mꢃ].
Analytical data for complex 17b: Yield: 57.8 mg
1
(61%); H-NMR (CD3OD): d 1.3ꢂ
/
1.4 (m, 12H), 1.75ꢂ
2.4 (m, 4H), 3.3ꢂ3.4 (m,
3.8 (AB, 4H), 4.0 (q, 1H), 4.4 (m,
/
1.85 (m, 4H), 2.0 (s, 3H), 2.3ꢂ
/
/
graphy on silica gel (EtOAcꢂ/MeOH: 9/1). The saponi-
10H), 3.5 (d, 2H), 3.5ꢂ
/
fication of the ester groups was performed in 1 N NaOH
1
for 3 h at r.t. (Scheme 2). Analytical data for 12: H-
1H), 6.3 (t, 1H), 7.6 (s, 1H); 13C-NMR (CD3OD): d
197.8 (CO), 196.9 (CO), 181.8, 174.7, 165.3, 151.2,
137.0, 110.6, 86.0, 85.4, 71.7, 63.0, 52.0, 52.5, 50.8,
41.4, 38.0, 35.3, 24.4, 22.9, 11.7; Anal. Calc. for
NMR (CD3OD): d 2.0 (s, 3H), 2.2 (m, 2H), 2.6 (m, 2H),
2.6 (m, 4H), 3.2 (m, 1H), 3.4 (d, 2H), 3.5 (s, 4H), 3.9 (m,
1H), 4.2 (m, 1H), 6.2 (t, 1H), 7.2ꢂ
1H); 13C-NMR (CD3OD): d 172.8, 169.8, 165.3, 151.2,
142.2, 137.1, 128.8ꢂ127.1, 110.3, 85.7, 84.9, 71.7, 71.6,
57.5, 41.0, 40.6, 38.6, 30.2, 10.9; HPLC: Rtꢁ15.4 min.
Analytical data for 15: H-NMR (CD3OD): d 1.8ꢂ1.9
/
7.4 (m, 5H), 7.6 (s,
C22H26N4O12Re(C8H20N)1.5Na1.5Br1.3Cl0.7
×
/
H2O:
37.08; H, 5.31; N, 6.99. Found: C, 37.37; H, 5.38; N,
C,
/
6.16%; IR (KBr): 3418 br, 2014 s, 1998 vs, 1874 w, 1684
s, 1654 m cmꢃ1; HPLC: Rtꢁ
16.1 min; MS (ESI): m/z
/
/
1
/
(%) 725.2 (100), 723.2 (60) [Mꢃ].
Analytical data for complex 18b: Yield: 76 mg (86%);
(m, 4H), 1.9 (s, 3H), 2.4 (m, 2H), 3.4 (m, 2H), 3.6 (d,
2H), 4.1 (m, 1H), 4.1 (s, 4H), 4.2 (m, 1H), 4.4 (m, 1H),
1H-NMR (CD3OD): d 1.3ꢂ
/
1.4 (m, 18H) 1.7ꢂ1.8 (m,
/

70
R. Schibli et al. / Journal of Organometallic Chemistry 668 (2003) 67ꢂ74
/
Scheme 1. (i) DCC, NHS in DMF, 60 8C, 2 h; (ii) 1 N NaOH, r.t., 3 h.
4H), 2.0 (s, 3H), 2.3 (m, 4H), 3.3ꢂ
/
3.4 (m, 10H), 3.5 (d,
HPLC: Rtꢁ20.5 min; MS (ESI): m/z (%) 809.0 (100),
/
2H), 3.6ꢂ
/
3.8 (AB, 4H), 4.0 (m, 1H), 4.3 (m, 1H), 6.3 (t,
807.0 (60) [Mꢃ].
Analytical data for complex 20b: Yield: 87.8 mg
1H), 7.6 (s, 1H); 13C-NMR (CD3OD): d 199.0 (CO),
198.0 (CO), 183.0, 176.8, 166.0, 151.2, 139.6, 137.4,
112.0, 88.0, 85.5, 74.0, 71.0, 64.4, 52.0, 42.7, 40.2, 37.3,
1
(75%); H-NMR (CD3OD): d 1.3ꢂ
/
1.4 (m, 5H), 2.0 (s,
2.3 (m, 4H), 2.6 (m, 2H), 3.0ꢂ3.25 (m, 3H),
3.5 (m, 3H), 3.5 (d, 2H), 3.6ꢂ3.8 (AB, 4H), 3.84 (m,
1H), 4.23 (m, 1H), 6.2 (t, 1H), 7.3ꢂ7.4 (m, 5H), 7.5 (s,
1H); 13C-NMR (CD3OD): d 197.3 (CO), 196.9 (CO),
181.3, 173.0, 165.3, 151.2, 142.5, 137.1, 128.6ꢂ126.8,
3H), 2.15ꢂ
/
/
3.3ꢂ
/
/
30.0ꢂ
(C8H20N)0.6Na0.4
Found: C, 40.19; H, 5.45; N, 6.88%; IR (KBr): 3384 br,
2020 s, 1880 vs, 1700 w, 1684 s, 1652 s, 1646 m, cmꢃ1
HPLC: Rtꢁ20.9 min; MS (ESI): m/z (%) 766.8 (100),
764.8 (60) [Mꢃ].
/26.0, 13.4; Anal. Calc. for C25H32N4O12Re-
/
×
/
1.5H2O: C, 40.62; H, 5.38; N, 7.31.
;
/
/
110.4, 85.9, 85.0, 71.7, 67.6, 62.9, 51.9, 43.0, 40.9, 38.8,
30.9, 11.1, 6.3; Anal. Calc. for (C28H30N4O12)1.17R-
e(C8H20N)0.4Na0.6: C, 44.47; H, 4.47; N, 7.32. Found: C,
44.51; H, 5.17; N, 7.15%; IR (KBr): 3739 br, 2014 s,
Analytical data for complex 19b: Yield: 74.2 mg
(79%); 1H-NMR (CD3OD): d 1.4 (m, 18H), 1.7 (m,
2H) 1.8 (m, 2H), 2.0 (s, 3H), 2.3ꢂ
/
2.4 (m, 4H), 3.4 (m,
1892 vs, 1703 s, 1671 s, cmꢃ1; HPLC: Rtꢁ
20.2 min;
/
4H), 3.5 (d, 2H), 3.6ꢂ
/
3.8 (AB, 4H), 4.0 (m, 1H), 4.3 (m,
MS (ESI): m/z (%) 801.2 (100), 799.2 (60) [Mꢃ].
Analytical data for complex 21b: Yield: 45 mg (92%);
1H), 6.3 (t, 1H), 7.6 (s, 1H); 13C-NMR (CD3OD): d
197.6 (CO), 197.1 (CO), 181.7, 175.3, 165.0, 151.2,
136.9, 110.5, 85.6, 85.2, 71.8, 69.5, 62.8, 51.9, 46.8,
1H-NMR (CD3OD): d 1.3ꢂ
4H), 2.0 (s, 3H), 2.3 (m, 2H), 3.3ꢂ
3.8 (AB, 4H), 4.0 (m, 1H), 4.2 (m, 1H), 4.39
/
1.4 (m, 6H), 1.78ꢂ/1.9 (m,
/
3.4 (m, 6H), 3.6 (d,
46.9, 41.1, 38.8, 35.8, 29.2ꢂ
Calc. for C28H38N4O12Re(C8H20N)0.3Na0.7
40.57; H, 5.38; N, 6.69. Found: C, 40.00; H, 5.41; N,
6.63%. IR (KBr): 3550 br, 2015 s, 1888 vs, 1672 s, cmꢃ1
/
28.8, 26.6ꢂ
/24.9, 11.1; Anal.
2H), 3.6ꢂ
/
×
/
2H2O: C,
(m, 1H), 6.28 (t, 1H), 7.6 (s, 1H); 13C-NMR (CD3OD):
d 197.8 (CO), 196.9 (CO), 181.8, 176.0, 165.3, 151.2,
138.5, 136.2, 110.6, 86.0, 85.4, 72.3, 71.5, 63.0, 52.0,
;
Scheme 2. (i) DCC, NHS in DMF, 60 8C, 2 h; (ii) methanol, conc. HCl, 10% Pd/C, H2, r.t., 12 h; (iii) BrCH2CO2CH3, methanol, NEt3, reflux, 12 h;
(iv) 1 N NaOH, r.t., 3 h.

R. Schibli et al. / Journal of Organometallic Chemistry 668 (2003) 67ꢂ
/
74
71
41.4, 38.6, 31.8, 20.1, 12.0; Anal. Calc. for
C22H26N4O13Re(C8H20N)0.7Na0.3: C, 39.52; H, 4.81;
N, 7.85. Found: C, 38.71; H, 5.29; N, 8.07%; IR
(KBr): 3408 br, 2020 s, 1878 vs, 1646 s, cmꢃ1; HPLC:
Rt: 14.9 min; MS (ESI): m/z (%) 741.1 (100), 739.1 (60)
[Mꢃ].
biologically active technetium and rhenium based thy-
midine analogues in the literature. In order to elucidate
the potential of organometallic technetium and rhe-
nium-tricarbonyl complexes of thymidine (or uridine)
for radiopharmaceutical application, we are currently
focusing on the functionalization and labeling of
thymidine at the C-5?-position. From our previous
studies we have recognized, that poly amino poly
carboxylic acid based ligand systems react readily with
the precursor 1 to form stable and hydrophilic com-
3. Results and discussion
plexes [14ꢂ16]. Thus, we synthesized a series of 5?-
/
3.1. Syntheses
carboxamide derivatives of 5? amino thymidine with an
imino diacetic acid chelating moiety and alkyl spacers of
variable chain lengths and with various side chains. The
Martin et al. recently presented a series of carbox-
amide derivatives of 5?-amino-2?,5?-dideoxy-5-ethyluri-
dine and could show, that they are highly potent
inhibitors of HSV1-TK and HSV2-TK [9]. The IC50
values for these enzymes were found to range from 1 mM
to 1 nM which is orders of magnitude more potent than
thymidine derivatives 6ꢂ
through coupling with the corresponding N,N-dicar-
boxy methyl amino carboxylic acids 2ꢂ5 in the presence
/9, 12 and 15 were obtained
/
of 1,3-dicyclohexyl carbodiimide (DDC) and N-hydro-
xysuccinimid (NHS) in DMF as shown in Scheme 1. The
e.g. the antiviral drug Acyclovir (IC50ꢁ28 mM). The
/
compounds were purified over silica gel (ethylacetateꢂ
/
design of these HSV TK inhibitors was based on
isosteric and isoelectronic analogues of thymidine
monophosphate (Fig. 1). Based on the X-ray structure
of HSV-1 TK and molecular modeling calculation, there
is evidence that the aromatic residues of these com-
pounds are located outside the binding site of the
enzyme [10,11]. Other X-ray structure analyses of the
viral enzyme predicted, that functionalization of thymi-
dine/uridine at positions other than C-5? causes serious
reduction of binding to the enzyme [12]. Thus, it would
be interesting and reasonable to develop synthetic
strategies for high affinity, C5?-functionalized and
99mTc-labeled thymidine/uridine analogues for potential
use in radiopharmacy.
methanol: 8/1). Yields varied between 64 and 79% for
the coupling reaction. For the derivatives 12 and 15 with
additional side chains the chelating moiety was build up
after coupling of the spacers to 5?-amino-thymidine. N-
Boc-5-amino-2(S)-hydroxy-pentanoic acid [17] or 4-
cyano-3-phenyl butanoic acid 10 were amidically linked
to 5?-amino-thymidine (Scheme 2). After reduction of
the cyano group or cleavage of the Boc group, the
pendent primary amino function of compounds 11 and
14 were double N-alkylated with methyl bromo acetate
in presence of NEt3 to form the nucleoside derivatives 12
and 15 in yields of 38ꢂ
purification on silica gel (ethylacetateꢂ
ethylacetateꢂmethanol: 9/1, respectively). All nucleoside
/59% after chromatographic
/methanol: 8/2,
/
Discovered for radiopharmaceutical application, the
water-soluble, organometallic complex [M(H2O)3-
derivatives have been unambiguously characterized by
1
means of H-, 13C-NMR and mass spectroscopy.
The reactions with the organometallic precursor 1b
(CO)3]ꢀ (Mꢁ
Tc, Re), 1, has gained considerable
/
attention in the past years. The high kinetic inertness,
its small size and the versatility of appropriate chelating
systems make this organometallic core particularly
suitable for in vivo stable radiolabeling of small
biomolecules. Mundwiler et al. have recently presented
C-5 functionalized organo-metallic thymidine complexes
of Tc-99m and Re [13]. However, to our knowledge,
there exists no other systematic in vitro investigation of
have been performed in a waterꢂ
at 60 8C (Scheme 3). The crude products were purified
over SepPak† reversed phase columns using a waterꢂ
methanol gradient. The complexes 16bꢂ21b were ob-
tained in good yields (61ꢂ84%) as a mixture of the
/methanol mixture (1/1)
/
/
/
corresponding Naꢀ and NEt4ꢀ salts as evident from the
elemental analyses. The IR spectra of all complexes
revealed the typical fac-M(CO)3 pattern of a symme-
trically coordinated complex with significantly blue-
shifted CO stretch frequencies compared to the starting
material [NEt4]2[ReBr3(CO)3] (2000 and 1868 cmꢃ1). In
1
the H-NMR spectra of complexes 16bꢂ
/21b, the four
protons of the coordinated IDA moiety showed the
typical pattern of an AB spin system (Fig. 2B). The
singlet of the previously identical NCH2CO2 protons in
6ꢂ9, 12 and 15 (Fig. 2A) became non-equivalent upon
/
rigid coordination to the metal center by virtue of their
axial and equatorial orientation. The same spectroscopic
features have been previously reported for rhenium-
Fig. 1. Examples of a new generation of potent inhibitors of herpes
simplex virus thymidine kinase [9].

72
R. Schibli et al. / Journal of Organometallic Chemistry 668 (2003) 67ꢂ74
/
Scheme 3. (i) Mꢁ99m
/
Tc: PBS buffer, pH 7.4, 75 8C, 1 h; Mꢁ
/
Re: methanolꢂH2O, 60 8C, 5 h.
/
The corresponding radioactive technetium-99m com-
plexes 16aꢂ21a have been almost quantitatively pre-
pared in physiological phosphate buffer (PBS; pH 7.4)
at concentration of 10ꢃ4 M of compounds 6ꢂ
9, 12 and
/
/
15 after 60 min at 75 8C (Scheme 3). Specific activities
between 18 and 180 MBq mmolꢃ1 could be achieved
under these labeling conditions. Single species have been
produced with all nucleoside derivatives. The character-
ization of complexes 16aꢂ21a was accomplished by
/
comparison of the retention times of the g-traces with
the UV-traces (254 nm) of the corresponding rhenium
complexes (Fig. 3). The organometallic 99mTc-complexes
were challenged in vitro in PBS buffer and human serum
albumin for 24 h at 37 8C. Decomposition or dissocia-
ꢃ
tion of the complexes to either TcO4 or 1a was less
than 8% of the initial activity. This stabilities and
specific activities would be sufficient for a potential
radiopharmaceutical application of such compounds.
3.2. In vitro evaluation
For an initial evaluation of the biological activity, the
Fig. 2. Section of the 1H-NMR spectra between 3.0 and 4.5 ppm of the
thymidine analogue 9 (A) and of the corresponding complex [NE-
rhenium complexes 16bꢂ21b have been screened and
/
tested in vitro for inhibition of HSV-1 TK as well as
human TK 1 according to the procedure described by
Pilger et al. [8]. The inhibition of the enzymatic reaction
t4][Re(9)(CO)3], 19b (B) (Rꢁ5?-amidodecyl thymidine). Only NCH2
protons of the IDA-moiety and of the counter ion NEt4 (in case of
complex 19b) are labeled.
/
ꢀ
TK
ATPꢀdT 0 ADPꢀdTMP
tricarbonyl complexes of IDA-functionalized desoxyglu-
cose and glucose derivatives [16]. This proved the
desired symmetric, tridentate coordination of the me-
tal-tricarbonyl core via the IDA chelate. There was no
significant chemical shift observed in the 1H-NMR
was monitored. The results of these in vitro experiments
are depicted in Fig. 4. The formation of ADP or dTMP
in presence of hTK 1 at various concentrations of
complexes 16bꢂ/21b was measured. The organometallic
spectra of complexes 16bꢂ21b for protons, which are
/
rheniumꢂthymidine complexes revealed competitive in-
/
not in close vicinity of the chelate. Therefore, unspecific
coordination or interaction of the metal center via other
functional groups of thymidine can largely be excluded.
All complexes revealed a good solubility in water and
polar organic solvents, which was a prerequisite for
enzymatic studies in vitro.
hibition of hTK 1. One could observe a correlation
between length of the spacer and the inhibitorial effect
of the rhenium complexes (Fig. 4A). Inhibition was
found to be increased for complexes with longer alkyl
chains. On the other hand, the introduction of an
additional hydrophobic (complex 20b) or hydrophilic

R. Schibli et al. / Journal of Organometallic Chemistry 668 (2003) 67ꢂ
/
74
73
ality [1]. In case of the organometallic complexes 16ꢂ
/
21
the selectivity for hTK 1 is presumably related to the
introduction of the organometallic complex moiety at
the C-5? position of 5?-amino-thymidine. To the best of
our knowledge, such an effect has never been observed
and there are no other compounds reported in the
literature, which reveal a comparable TK selectivity. Till
now, we do not have a concluding explanation for these
results. It can be speculated that steric interference of the
rhenium-tricarbonyl complex moiety with the viral
enzyme prohibits a binding of the complexes to the
active site. Computer-aided docking studies based on
the X-ray structure of HSV1-TK are in progress and
might help to understand the absence of inhibition of
the viral enzyme.
Fig. 3. (A) HPLC trace of complex 19b (254 nm); (B) g-trace of the
radioactive complex 19a. Peak at 4.2 min represent unreacted
[
99mTc(H2O)3(CO)3]ꢀ (B
5% of total activity).
/
4. Conclusion
In conclusion, the first series of organometallic
technetium and rheniumꢂthymidine complexes have
/
been synthesized and fully characterized. The biological
affinity of the rhenium-tricarbonyl compounds was
tested in vitro with hTK 1 and HSV1-TK. The selective
and exclusive inhibition of hTK 1 prohibits the use of
the corresponding radioactive technetium-tricarbonyl
complexes for gene therapy monitoring. On the other
hand, with the current functionalization and labeling
strategy at position C-5? thymidine derivatives could be
produced, which might have the potential to serve as
proliferation markers for tumor diagnosis.
Acknowledgements
Fig. 4. Enzymatic formation of ADP at various concentrations of
We thank Regula Johner for assistance and Judith
Stahel for preparing the Tc-99m complexes. This work
was supported by Mallinckrodt-Tyco Inc.
complexes 16bꢂ19b (A) and of complexes 20b and 21b (B) in presence
/
of hTK 1. Complex [Re(IDA)(CO)3]ꢃ was used as a negative control.
(complex 21b) group does not significantly change the
inhibitory activity (Fig. 4B). The complex [NEt4]-
[Re(IDA)(CO)3] without the pharmacophoric group
was used as a negative control (Fig. 4B). It is reasonable
to assume, that a better spacial separation of the
pharmacophor (thymidine) and the complex moiety
reduce potential steric and/or electrostatic interference
with the protein. Using the viral enzyme, only insignif-
icant inhibition of dTMP/ADP formation could be
detected in presence of the complexes. These findings
are surprising since the thymidine moiety should be
recognized by both human and viral TK. In case of the
inhibitors shown in Fig. 1 the selectivity for HSV1-TK,
respectively, is not related to the organic residues
introduced at position C-5? but by substitution of the
methyl group at the C-5 position by an ethyl function-
References
[1] D. Baker, C. Chu, Nucleosides and Nucleotides as Antitumor and
Antiviral Agents, Kluwer Academic/Plenum, New York, 1993.
[2] R. Alberto, K. Ortner, N. Wheatley, R. Schibli, A.P. Schubiger, J.
Am. Chem. Soc. 123 (2001) 3135.
[3] R. Alberto, R. Schibli, A. Egli, P.A. Schubiger, W.A. Herrmann,
G. Artus, U. Abram, T.A. Kaden, J. Organomet. Chem. 493
(1995) 119.
[4] W. Bannwarth, Helv. Chim. Acta 71 (1988) 1517.
[5] D.S. Karanewsky, M.C. Badia, D.W. Cushman, J.M. Deforrest,
T. Dejneka, V.G. Lee, M.J. Loots, E.W. Petrillo, J. Med. Chem.
33 (1990) 1459.
[6] A.P. Krapcho, Synthesis (1982) 805.
[7] R. Schibli, R. Schwarzbach, R. Alberto, K. Ortner, H. Schmalle,
C. Dumas, A. Egli, A.P. Schubiger, Bioconjugate Chem. (2002) in
press.

74
R. Schibli et al. / Journal of Organometallic Chemistry 668 (2003) 67ꢂ74
/
[8] B.D. Pilger, R. Perozzo, F. Alber, C. Wurth, G. Folkers, L.
[13] S. Mundwiler, R. Alberto, J. Labelled Compd. Radiopharm. 44
(2001) S494.
Scapozza, J. Biol. Chem. 274 (1999) 31967.
[9] J.A. Martin, R.W. Lambert, J.H. Merrett, K.E.B. Parkes, G.J.
Thomas, S.J. Baker, D.J. Bushnell, J.E. Cansfield, S.J. Dunsdon,
A.C. Freeman, R.A. Hopkins, I.R. Johns, E. Keech, H. Simmo-
nite, A. Walmsley, P.W. Kai-In, M. Holland, Bioorg. Med. Chem.
Lett. 11 (2001) 1655.
[14] R. Schibli, R. La Bella, R. Alberto, E. Garcia-Garayoa, K.
Ortner, U. Abram, P.A. Schubiger, Bioconjugate Chem. 11 (2000)
345ꢂ351.
/
[15] R. Waibel, I. Novak-Hofer, R. Schibli, P. Blauenstein, E. Garcia-
Garayoa, R. Schwarzbach, K. Zimmermann, R. Pellikka, O.
Gasser, A. Blanc, M. Bruhlmeier, P.A. Schubiger, Chimia 54
(2000) 683.
[10] L. Scapozza, personal communication.
[11] J. Vogt, R. Perozzo, A. Pautsch, A. Prota, P. Schelling, P. Pilger,
G. Folkers, L. Scapozza, G.E. Schulz, Proteins 41 (2000)
545.
[16] J. Petrig, R. Schibli, C. Dumas, R. Alberto, P.A. Schubiger,
Chem.-Eur. J. 7 (2001) 1868.
[12] K. Wild, T. Bohner, G. Folkers, G.E. Schulz, Protein Sci. 6 (1997)
2097.
[17] I. Shin, M.R. Lee, J. Lee, M. Jung, W. Lee, J. Yoon, J. Org.
Chem. 65 (2000) 7667.