A novel tetrabranched neurotensin(8-13) cyclam derivative: Synthesis, 64Cu-labeling and biological evaluation

Article abstract of DOI:10.1016/j.jinorgbio.2011.02.011

New macrocyclic 1,4,8,11-tetraazacyclotetradecane (cyclam) derivatives with 1, 2 and 4 neurotensin(8-13) units 4, 5 and 7 have been synthesized. Compounds 4 and 5 were prepared by the reaction of non-stabilized neurotensin(8-13) and cyclamtetrapropionic acid 2 using 1-ethyl-3-(3-dimethylaminocarbonyl) carbodiimide-hydrochloride and N-hydroxysulfosuccinimide. The tetrameric compound 7 was synthesized by Michael addition of neurotensin(8-13) acrylamide 6 and cyclam 1. The copper(II) complexation behavior of 4, 5 and 7 was investigated by UV/visible spectrophotometry and shows that the metal center resides inside the N4 chromophore with additional apical interactions established with pendant arms. The novel tetrabranched NT(8-13) cyclam 7 with nanomolar neurotensin receptor 1 binding affinity was efficiently radiolabeled with ⁶⁴Cu under mild conditions. ⁶⁴Cu ? 7 showed slow transchelation in the presence of a large amount of cyclam as competing ligand, while it completely remains intact in the presence of EDTA. The in vivo behavior of ⁶⁴Cu ? 7 was studied in rats and mice. The metabolic stability in rodent models was high with a half-life of intact ⁶⁴Cu ? 7 in plasma of 34 min in rats and 60 min in the mice, respectively. The binding affinity was high enough to demonstrate in vivo binding of ⁶⁴Cu ? 7 to NTR1 overexpressing HT-29 tumor xenotransplants in nude mice. Regarding elimination, ⁶⁴Cu ? 7 showed a substantial renal and reticuloendothelial accumulation. On the other hand, metabolization of the compound in vivo with a resulting metabolite-postulated to be the ⁶⁴Cu-cyclam-tetraarginine complex-also showed long retention in the circulating blood, preventing a better contrast of tumor imaging.

Full text of DOI:10.1016/j.jinorgbio.2011.02.011

Journal of Inorganic Biochemistry 105 (2011) 821–832
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
A novel tetrabranched neurotensin(8–13) cyclam derivative: Synthesis,
64Cu-labeling and biological evaluation
Anika Röhrich, Ralf Bergmann, Anne Kretzschmann, Steffi Noll, Jörg Steinbach,
⁎
Jens Pietzsch, Holger Stephan
Institute of Radiopharmacy, Helmholtz-Zentrum Dresden-Rossendorf, PF 510119, D-01314 Dresden, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
New macrocyclic 1,4,8,11-tetraazacyclotetradecane (cyclam) derivatives with 1, 2 and 4 neurotensin(8–13)
units 4, 5 and 7 have been synthesized. Compounds 4 and 5 were prepared by the reaction of non-stabilized
neurotensin(8–13) and cyclamtetrapropionic acid 2 using 1-ethyl-3-(3-dimethylaminocarbonyl)carbodii-
mide-hydrochloride and N-hydroxysulfosuccinimide. The tetrameric compound 7 was synthesized by
Michael addition of neurotensin(8–13) acrylamide 6 and cyclam 1. The copper(II) complexation behavior of 4,
5 and 7 was investigated by UV/visible spectrophotometry and shows that the metal center resides inside the
N4 chromophore with additional apical interactions established with pendant arms. The novel tetrabranched NT(8–
13) cyclam 7 with nanomolar neurotensin receptor 1 binding affinity was efficiently radiolabeled with 64Cu under
mild conditions. 64Cu⊂7 showed slow transchelation in the presence of a large amount of cyclam as competing
ligand, while it completely remains intact in the presence of EDTA. The in vivo behavior of 64Cu⊂7 was studied in
rats and mice. The metabolic stability in rodent models was high with a half-life of intact 64Cu⊂7 in plasma of
34 min in rats and 60 min in the mice, respectively. The binding affinity was high enough to demonstrate in vivo
binding of 64Cu⊂7 to NTR1 overexpressing HT-29 tumor xenotransplants in nude mice. Regarding elimination,
64Cu⊂7 showed a substantial renal and reticuloendothelial accumulation. On the other hand, metabolization of the
compound in vivo with a resulting metabolite—postulated to be the 64Cu-cyclam-tetraarginine complex—also
showed long retention in the circulating blood, preventing a better contrast of tumor imaging.
Received 19 November 2010
Received in revised form 21 January 2011
Accepted 21 February 2011
Available online 22 March 2011
Keywords:
Neurotensin receptor
Cyclam derivatives
Tetramer
Copper-64
Colorectal adenocarcinoma
Small animal positron emission tomography
© 2011 Elsevier Inc. All rights reserved.
1. Introduction
NTR2, and to sortilin/NTR3. It is also known that neurotensin is
involved in developing neoplastic cells. There is mounting evidence
In the mid-eighties, it was discovered that certain tumors over-
express receptors for neuropeptides [1,2]. Since then, neuropeptides
have aroused increasing interest for the diagnosis and radiotherapy of
tumors which overexpress neuropeptide receptors. Compared to
macromolecules, such as monoclonal antibodies, the use of small
peptides (fewer than 30 amino acids) as tumor-localizing radiophar-
maceuticals is advantageous due to their low molecular weights,
which empirically improves the penetration through the tumor tissue.
Moreover, a faster clearance from healthy tissue can be expected
resulting in an improved tumor/non-tumor activity ratio. Peptides are
usually non-antigenic and feature high affinity and specificity to a
number of neuropeptide receptors [3–10].
Radiolabeled peptides based on neurotensin derivatives represent
very interesting targeting vector molecules for certain cancer entities.
Neurotensin is a tridecapeptide (pGlu1-Leu2-Tyr3-Gly4-Asp5-Lys6-
Pro7-Arg8-Arg9-Pro10-Tyr11-Ile12-Leu13) and was first isolated from
bovine hypothalamus [11]. This neuropeptide shows high selectivity
and affinity to the G-protein-coupled neurotensin receptors NTR1 and
that NTR1 is overexpressed in a number of human tumors, such as
breast, prostate, lung, colorectal and pancreatic cancer [12–20]. Most
radiopharmaceutical investigations have been performed with the
hexapeptide (Arg8-Arg9-Pro10-Tyr11-Ile12-Leu13) NT(8–13)—the
shortest neurotensin binding motif for efficient and selective
ligand–receptor interaction—and their metabolic stabilized deriva-
tives [21]. These monomeric neurotensin fragments have been mainly
radiolabeled with 99mTc, but also to a lesser extent with 111In, 125I,
177Lu, 188Re, 201Tl and 18F [22–40]. Hitherto, radiolabeling of
neurotensin with 64Cu has not been reported.
Due to the emerging possibilities of metabolic stabilization as well
as to induce multivalency effects in receptor–ligand interaction, mul-
timeric neurotensin derivatives (cf. Chart 1) gain specific interest.
Bracci et al. prepared tetrabranched peptide dendrons I and reported
excitingly high inhibition of [3H]neurotensin binding to HT-29 cells
using the tetramers coupled via the C-terminus of NT(8–13). These
so-called “magic forks” show remarkably high resistance of peptide
cleavage in human plasma and serum. The synthetic pathway applied
allows the introduction of appropriate molecules, such as tar-
geting units, fluorescent labels and cytotoxic agents, at the focal
point [41–45]. Artificial neurotensin tetramers II became available by
⁎
Corresponding author. Tel.: +49 351 260 3091; fax: +49 351 260 3232.
E-mail address: h.stephan@hzdr.de (H. Stephan).
0162-0134/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jinorgbio.2011.02.011

822
A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
Chart 1. Multimeric neurotensin derivatives.
applying solid phase peptide synthesis, coupling NT(8–13) fragments
with Fmoc-Glu-Glu-Glu-OH [46]. After deprotection, the remaining
amino group of the glutamic acid trimer can be utilized for the intro-
duction of 18F-containing agents. Attempts to label aminooxy-func-
tionalized tetramers II using 4-[18F]fluorobenzaldehyde ([18F]FBA)
and 2-[18F]fluoro-2-deoxy-D-glucose ([18F]FDG) led to low radio-
chemical yields. The replacement of the aminooxy moiety by
sulfhydryl groups facilitates the 18F-labeling of II with an appropriate
maleimide, and in doing so, the labeling efficacy can be improved, but
the specific activity is rather low [47,48].
150 min were used. The complete separation of 64Cu and 61Co was
confirmed by gamma-ray spectroscopy. Nickel targets were prepared
by electrodeposition of enriched 64Ni (99.6%) on gold disks at
amounts of 95–120 mg. The plated diameter was 7 mm, matching
more than 80% intensity of the cyclotron proton beam, as measured by
autoradiography of a natNi disk irradiated with 15 MeV protons to
induce the natNi(p,x)57Ni reaction. The yields of the nuclear reaction
64Ni(p,n)64Cu constituted between 3.6 and 5.2 GBq (EOB) with
specific activities of 150–250 GBq/μmol Cu.
Cyclam (1,4,8,11-tetraazacyclotetradecane) forms very stable com-
plexes with Cu(II) and it is therefore one of the most important ligands
for radiopharmaceutical applications [49,50]. Recently, a concept was
reported for dendritic encapsulation of copper(II), employing cyclam as
central-arranged chelating agent. PEGylation and carbohydration of a
tetraamino-functionalized branched cyclam derivative provide ligands
with interesting solubility and targeting properties [51,52]. Drawing on
a similar ensemble of building blocks, we prepared the tetrabranched
NT(8–13) star-shaped multimer III comprising cyclam as core element.
This synthetic approach aims at both the metabolic stabilization of
peptide residues and improved resistance of radiocopper complexes
formed to demetalation in vivo.
The present work reports the synthesis of branched cyclam deri-
vatives with one, two, and four NT(8–13) fragments and subsequent
evaluation of in vitro binding characteristics in NTR1 overexpressing
human colorectal adenocarcinoma (HT-29) cells and xenografts. UV/
vis spectroscopy was applied to study the complexation behavior of
the peptide–cyclam conjugates with copper(II). The neurotensin(8–
13) tetramer 7 was radiolabeled with 64Cu, and its radiopharmaco-
logical characterization includes in vitro and in vivo metabolism and
biodistribution studies in normal rats, as well as small animal PET
studies in NMRI nu/nu mice, bearing the human HT-29 tumor.
2.2. Methods and instrumentation
1H and 13C NMR spectra were recorded on a 400 MHz Varian Inova
spectrometer. The chemical shifts (δ, ppm) are relative to TMS or the
solvent. The abbreviations for the peak multiplicities are as follows:
d (doublet), t (triplet) and m (multiplet). Elemental analyses were
performed on a Leco Elemental Analyzer CHNS-932. A Specord S10 UV/vis
spectrometer (Zeiss) was used to record electronic absorption spectra.
Electrospray ionization mass spectrometry (ESI-MS) was carried out using
a Micromass Tandem Quadrupole Mass Spectrometer (Quattro LC). HPLC
was performed using different instruments and methods. HPLC (Äkta
basic from GE Healthcare, UV-900, software Unicorn 5.0) was used for the
purification of 4, 5, 6 and 7. Method 1: Phenomenex Jupiter 4 μm Proteo
C12 90 Å, 10×250 mm, 3 mL/min, A: water+0.1% TFA, B: acetonitrile+
0.1% TFA, 0% B to 60% B in 40 min. Method 2: Eurospher 100 C18, 10 μm,
20×250 mm, 10 mL/min, A: acetonitrile+0.1% TFA, B: water+0.1% TFA,
70% B to 40% B in 25 min. Method 3: Eurospher 100 C18, 10 μm,
20×250 mm, 10 mL/min, A: acetonitrile+0.1% TFA, B: water+0.1% TFA,
70% B to 40% B in 25 min. Method 4: The characterization of the radioactive
conjugate 64Cu⊂7 was made by a Knauer system consisting of a pump
Wellchrom HPLC K1001, UV-detector K-2501, software Chromgate 2.8,
activity detector Raytest Ramona Star, Eurospher 100 C18, 4.6×200 mm,
1 mL/min, A: water+0.1% TFA, B: acetonitrile+0.1% TFA, 2 min 10% B, in
13 min from 10% to 90% B. Method 5: HPLC Hewlett Packard Series 1100,
activity detector Raytest Ramona Star, Zorbax (Agilent) 300SB-C18, 300 Å,
5 μm, 250×9.4 mm; A: water+0.1% TFA, B: acetonitrile+0.1% TFA, 5 min
95% A, 10 min from 95% A to 95% B, 5 min 95% B, 5 min from 95% B to 95%
A, 3 mL/min.
2. Experimental section
2.1. Materials
Chemical reagents and solvents were purchased from commercial
sources (Fisher Scientific, Fluka, Merck, Sigma-Aldrich) and used
without further purification. Cyclam 1 was purchased from Fluka, and
NT(8–13) 3 from Bachem. The production of 64Cu was performed at a
PET cyclotron, similar to those recently described for the 64Cu–61Co-
production process [53–55]. For the 64Ni(p,n)64Cu nuclear reaction,
15 MeV protons of a Cyclone® 18/9 with a beam current of 12 μA for
2.3. Synthesis
2.3.1. 1,4,8,11-Tetraazacyclotetradecane-1,4,8,11-tetrapropionic acid (2)
The tetramethylester-functionalized cyclam derivative was syn-
thesized following the previously reported procedure. Analytical data

A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
823
were in agreement with that described in the literature [51]. For
further reaction, 328 mg (0.6 mmol) of this precursor was suspended
in 2.4 mL 1 M aqueous sodium hydroxide solution. After stirring for
45 min at room temperature, a clear solution was obtained. After
filtration, the solution was filled in a small column with 2.5 mL cation
exchanger Dowex50WX8 in H2O, and the free acid 2 was eluted with
60 mL of water. After removal of the solvent under reduced pressure
and lyophilization, 236 mg (80%) of a colorless solid was obtained. 1H
NMR (D2O): δ 3.12−2.95 (m, 16 H, 8×CH2), 2.83−2.75 (t, 8 H,
4×CH2), 2.67−2.20 (t, 8 H, 4×CH2), 1.96−1.87 (t, 8 H, 4×CH2), 1.80−
1.72 (m, 4 H, 2×CH2). ESI-MS (m/z): 487.4 (calcd. 487.6) [M−H]−.
Elemental analysis: calculated for C22H40N4O8·0.5H2O: C 53.10%, H
11.26%, N 8.31%, found: C 52.64%, H 11.14%, N 8.22%.
colorless solid after lyophilization as tetra trifluoro acetate. tR(HPLC):
10.5 min, 1H NMR (D2O): δ 7.15−7.07 (d, 2 H, 2×CHarom., 42, 46),
6.84−6.72 (d, 2 H, 2×CHarom., 43, 45), 6.38−6.17 (m, 2 H, 1×CH2, 1),
5.82−5.75 (m, 1 H, 1×CH, 2), 4.62−4.50 (m, 2 H, 2×CH, 11, 14),
4.47−4.08 (m, 4 H, 4×CH, 5, 8, 17, 20), 3.81−3.57 (m, 2 H, 1×CH2,
37), 3.22−3.13 (m, 4 H, 2×CH2, 25, 32), 3.00−2.93 (m, 2 H, 1×CH2,
40), 2.27−2.16 (m, 1 H, 1×CH, 48), 2.02−1.92 (m, 2 H, 1×CH2, 31),
1.85−1.57 (m, 13 H, 1×CH, 6×CH2, 23, 30, 38, 39, 49, 52, 53), 1.44−
1.05 (m, 2 H, 1×CH2, 24), 0.97−0.76 (m, 12 H, 4×CH3, 50, 51, 54, 55),
ESI-MS (m/z): 871.5 (calcd. 872.1) [M+H]+. Elemental analysis:
calculated for C41H66N12O9⋅4TFA: C 44.35%, H 5.86%, N 12.66%, found:
C 44.07%, H 6.08%, N 12.85%.
2.3.6. N,N′,N″,N‴-Tetrakis((N-(arginyl-arginyl-prolyl-tyrosyl-isoleucyl-
leucine)-aminocarbonyl)-ethyl)-1,4,8,11-tetraazacyclotetradecane (7)
76 mg (0.057 mmol) 6 was suspended in 400 μL deionized water
and stirred at 40 °C for 90 min. After addition of 14.5 μL (0.07 mmol)
5 M aqueous NaOH solution a white precipitate was temporarily
observed. After 15 min, 1.14 mg (0.0057 mmol) cyclam 1 in 33 μL of
water/methanol (1:1) was added. The reaction was stirred for two
weeks at 40 °C. Evaporated solvent was replaced by water. After one
week, pH was adjusted to 10 by the addition of 25 μL MeOH, 50 μL
water and 75 μL 0.1 M aqueous NaOH solution. The crude product was
concentrated under reduced pressure, followed by the addition of
3 mL water at 40 °C. pH was adjusted to 4 with 10 μL TFA. After
concentration and addition of 1.5 mL of a mixture of 30% acetonitrile,
70% water and 0.1% TFA, the crude product was filtered. Purification
by HPLC using method 3 gave 10.8 mg (52%) of 7 as colorless solid
after lyophilization. tR(HPLC): 15.8 min, 1H NMR (D2O): δ 7.27−7.12 (d,
8 H, 8×CHarom., 42′, 46′), 6.96−6.80 (d, 8 H, 8×CHarom., 43′, 45′), 4.73−
4.10 (m, 24 H, 24×CH, 5′, 8′, 11′, 14′, 17′, 20′), 3.92−3.66 (m, 8 H, 4×CH2,
37′), 3.36−2.70 (m, 56 H, 28×CH2, 1′, 2, 2′, 3, 5, 7, 9, 10, 12, 14, 25′, 32′,
40′), 2.36−2.24 (m, 4 H, 4×CH, 48′), 2.02−1.92 (m, 8 H, 4×CH2, 31′),
2.00−1.63 (m, 52 H, 4×CH, 24×CH2, 23′, 30′, 38′, 39′, 49′, 52′, 53′),
1.57−1.12 (m, 12 H, 6×CH2, 6, 13, 24′), 1.10−0.88 (m, 48 H, 16×CH3,
2.3.2. Procedure A for the preparation of cyclam ligands 4 and 5
17.6 mg (0.09 mmol) 1-ethyl-3-(3-dimethylaminocarbonyl)car-
bodiimide-hydrochloride (EDC) was dissolved in 1 mL deionized
water, and a solution of 5 mg (0.01 mmol) cyclamtetrapropionic
acid 2 in 1 mL deionized water was added. 15.3 mg (0.07 mmol)
N-hydroxysulfosuccinimide in 1 mL deionized water was dropped
into the solution and stirred at ambient temperature. After 35 min
33.3 mg (0.04 mmol) H-NT(8–13)-OH in 1 mL deionized water
was added and pH adjusted to 8.5 with 0.1 M aqueous disodium
phosphate solution. The reaction mixture was stirred for 24 h and
concentrated under reduced pressure. The crude products were
purified by HPLC using method 1.
2.3.3. N-((N-(arginyl-arginyl-prolyl-tyrosyl-isoleucyl-leucine)
aminocarbonyl)-ethyl)-N′,N″,N‴-tris(2-(carboxy)ethyl)-1,4,8,
11-tetraazacyclotetradecane (4)
Procedure A afforded 6.1 mg (47%) of 4 as a white solid after
lyophilization. tR(HPLC): 22 min, 1H NMR (D2O): δ 6.96−6.89 (d, 2
H, 2×CHarom., 42′, 46′), 6.68−6.62 (d, 2 H, 2×CHarom., 43′, 45′),
4.48−4.30 (m, 2 H, 2×CH, 11′, 14′), 4.30−4.09 (m, 2 H, 2×CH, 8′, 17′),
4.09−3.72 (m, 2 H, 2×CH, 5′, 20′), 3.72−3.20 (m, 10 H, 5×CH2, 1′,
37′), 3.18−2.48 (m, 30 H, 15×CH2, 2, 2′, 3, 5, 7, 9, 10, 12, 14, 25′, 32′,
40′), 2.11−2.01 (m, 1 H, CH, 48′), 1.86−1.36 (m, 19 H, 9×CH2, 1×CH,
6, 13, 23′, 30′, 31′, 38′, 39′, 49′, 52′, 53′), 1.30−0.87 (m, 2 H, CH2, 24′),
0.79−0.62 (m, 12 H, 4×CH3, 50′, 51′, 54′, 55′), ESI-MS (m/z): 1288.2
(calcd. 1288.6) [M+H]+.
50′, 51′, 54′, 55′), ESI-MS (m/z): 1841.3 (calcd. 1842.1) [M+2H]2+
.
2.4. Radiochemistry
2.4.1. Radiolabeling experiments
Labeling of 7 was carried out by addition of 10 to 100 μg (2.7–
27 nmol) 7 to 200 μL [64Cu]CuCl2 (20 MBq) solution in 0.05 M 2-[N-
morpholino]ethanesulfonic acid (MES)-NaOH buffer (pH=5.5).
Specific activity AS of 64Cu was determined to be 200 55 GBq/μmol
(n=4) using the titration method with TETA [53,55]. Formation
kinetics and labeling yields of 64Cu⊂7 were investigated utilizing same
conditions as function of temperature and ligand concentration, and
were assayed using radio-HPLC (method 4, 64Cu⊂7, tR=12.5 min;
64CuCl2, tR=2.6 min).
Challenge studies with both EDTA and cyclam were performed
using radiotracer experiments. 100 μg 7 were added to 200 μL [64Cu]
CuCl2 (20 MBq) in 0.05 M MES-NaOH buffer (pH=5.5) and the
solution was incubated for 1 h at 50 °C. Full complexation of the ligand
was checked by radio-HPLC (method 4). Then 1 mg EDTA (125-fold
excess of EDTA referring to 7) and either 110 or 550 μg cyclam (20- and
100-fold excess of cyclam referring to 7) was added to the complex
solution and studied for 24 h. At several time points, the degree of
transchelation was determined by radio-HPLC (method 4, 64Cu⊂7,
tR =12.5 min; 64Cu⊂EDTA, tR =2.5 min; 64Cu⊂1, tR =4.5 min).
2.3.4. N,N′-Di((N-(arginyl-arginyl-prolyl-tyrosyl-isoleucyl-leucine)
aminocarbonyl)-ethyl)-N″,N‴-di(2-(carboxy)ethyl)-1,4,8,
11-tetraazacyclotetradecane (5)
Procedure A afforded 2.6 mg (12%) of 5 as a white solid after
lyophilization. tR(HPLC): 26 min, 1H NMR (D2O): δ 6.94−6.92 (d, 4 H,
4×CHarom., 42′, 46′), 6.65−6.58 (d, 4 H, 4×CHarom., 43′, 45′), 4.47−
4.35 (m, 4 H, 4×CH, 11′, 14′), 4.21−4.13 (m, 4 H, 4×CH, 8′, 17′), 3.96−
3.88 (m, 4 H, 4×CH, 5′, 20′), 3.68−3.23 (m, 4 H, 2×CH2, 37′), 3.18−
2.48 (m, 44 H, 22×CH2, 1′, 2, 2′, 3, 5, 7, 9, 10, 12, 14, 25′, 32′,
40′), 2.15−1.95 (m, 2 H, 2×CH, 48′), 1.86−1.36 (m, 28 H,
14×CH2, 2 H, 2×CH, 4 H, 2×CH2, 6, 13, 23′, 30′, 31′, 38′, 39′, 49′, 52′,
53′), 1.30−0.80 (m, 4 H, 2 × CH2, 24′), 0.80−0.55 (m, 24 H, 8×CH3,
50′, 51′, 54′, 55′), ESI-MS: m/z=2087.3 (calcd. 2087.6) [M+H]+
.
2.3.5. N-Acrylamido-(arginyl-arginyl-prolyl-tyrosyl-isoleucyl-leucine)-
tetra trifluoro acetate (6)
44.9 mg (0.055 mmol) H-NT(8–13)-OH 3 was dissolved in 375 μL
1 M aqueous NaOH solution. After addition of 20 μL phenolphthalein,
the reaction mixture was stirred at 0 °C. Within 12 min 3×10 μL
(3×0.124 mmol) acrylic acid chloride and 3×25 μL 1 M aqueous
NaOH solution were alternately added. The reaction was stirred at
0 °C and at room temperature for 50 min, respectively. After solvent
removal, the crude product was dissolved in 1 mL of a mixture of 30%
acetonitrile, 70% water and 0.1% TFA, and acidified with 24 μL TFA.
Purification by HPLC using method 2 gave 22.9 mg (31%) of 6 as
2.4.2. Distribution in water/1-octanol
Information about the lipophilicity of 64Cu complexes of cyclam
ligands 1, 2, 4, 5 and 6 were obtained by distribution experiments in
water/1-octanol systems. The experiments were performed with
100 μM buffered solution of the ligands (0.05 M 4-(2-hydroxyethyl)-
1-piperazine ethanesulfonic acid (HEPES)-NaOH, pH=7.2, 7.4, 7.6).

824
A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
The aqueous solution contains Cu(NO3)2 (10 μM) and was spiked with
64Cu]CuCl2 (500 kBq). In the case of ligands 4, 5 and 6, the solution
Wistar rats (Harlan Winkelmann GmbH, Borchen, Germany) were
housed under standard conditions with free access to standard food and
tap water. Animals aged between 7 and 9 weeks of age (body weight
171 14 g). The NMRI nu/nu mice (35 2 g) were housed under
standard conditions except the temperature (27 1 °C) in controlled-
airflow cabinets. Animals for each time point were intravenously
injected into a lateral tail vein with approximately 0.3 MBq 64Cu⊂7
conjugate in 0.5 mL electrolyte solution E-153 (Serumwerk Bernburg,
Germany) with 5% Tween 80. Animals were sacrificed at 5 and 60 min
post-injection. Blood and the major organs were collected, weighed, and
counted in a Wallac WIZARD Automatic Gamma Counter (PerkinElmer,
Germany). The activity of the tissue samples was decay-corrected and
calibrated by comparing the counts in tissue with the counts in aliquots
of the injected tracer that had been measured in the gamma counter at
the same time. The activity in the selected tissues and organs was
expressed as percent injected dose per organ (% ID) or as standardized
uptake value (SUV). Values are expressed as means standard de-
viation (mean SD; n=4) for each group.
[
was pre-equilibrated for 1 h at 50 °C. Full complexation was checked
by radio-HPLC (method 4) to give no evidence of free copper(II) in the
aqueous phase. The distribution experiments in 1-octanol/buffer
systems were performed at 25 1 °C in microcentrifuge tubes
(2 mL) by means of mechanical shaking for 30 min. The phase ratio
V(1-octanol):V(aq) was 1:1 (0.5 mL each). All samples were centrifuged,
the phases separated and the copper concentration was determined in
both phases radiometrically using γ-radiation [64Cu, NaI(Tl) scintilla-
tion counter automatic gamma counter 1480, Wizard 3″, PerkinElmer].
The results are the average values of three independent experiments.
2.5. In vitro and in vivo studies
2.5.1. Receptor binding in vitro
Human colorectal adenocarcinoma cell line HT-29 (ATCC Ref. No.
HTB-38; DSMZ, Braunschweig, Germany) was routinely cultivated in
T-25/75 flasks (Cellstar®, Greiner bio-one) as specified by ATCC in
McCoy's 5A medium containing Glutamax-I (Gibco-BRL), 10% fetal
calf serum (Sigma), and penicillin/streptomycin (10,000 U/mL, Gibco)
in humidified 5% CO2/95% air at 37 °C. For the experiments, cells were
detached by trypsin/EDTA (0.02–0.05%). The experiments were
conducted when cells were confluent as published by us elsewhere
[46]. For competitive binding inhibition experiments, a 1 mM stock
solution of the NT derivatives dissolved in DMSO was prepared. Final
concentrations of 10 μM, 6 μM, 1 μM, 600 nM, 200 nM, 100 nM,
60 nM, 10 nM, 6 nM, 2 nM, and 1 nM were obtained by dilution
with Tris buffer (50 mM Tris–HCl, 1 mM EDTA, 0.1% BSA, and 0.5 mM
o-phenanthroline, pH 7.4) containing 10% DMSO. The cells in the
flasks were washed three times with Tris buffer, removed from the
flask and resuspended in 8 mL of Tris buffer. The cell suspension was
then homogenized in a glass homogenizer. 200 μL of the cell
homogenate was mixed with 100 μL of the peptide solution and
500 μL of Tris buffer. 200 μL of [3H]NT was added and the mixture was
allowed to incubate for 30 min at room temperature. Non-specific
binding was assessed in the presence of 100 mM NT(8–13).
Separation of bound and free ligand was performed by rapid filtration
through Whatman GF/B glass fiber filters using a Brandel cell
harvester (Gaithersburg, MD, USA). Filters were washed with three
volumes of Tris buffer and were placed into vials with 4 mL
scintillation cocktail (ULTIMA Gold®, Packard Instruments, USA).
The [3H]NT remaining on the filters was quantified by liquid
scintillation spectrometry (TriCarb®, Packard Instruments, USA).
2.5.4. Small animal positron emission tomography
General anesthesia of rats was induced with inhalation of
desflurane 9% (v/v) (Suprane, Baxter, Germany) in 40% oxygen/air
(gas flow 1 L/min), and was maintained with desflurane 6% (v/v).
General anesthesia of NMRI nu/nu mice was induced with inhalation
of desflurane 12% (v/v) (Suprane, Baxter, Germany) in 40% oxygen/air
(gas flow 1 L/min), and was maintained with desflurane 8% (v/v). In
the PET experiments, approximately 18.5 MBq of the 64Cu⊂7 con-
jugate in 0.5 mL was administered intravenously over 1 min into a tail
vein. PET imaging of 64Cu⊂7 in mice was performed over 60 min
using a microPET P4 scanner (Siemens CTI Molecular Imaging Inc.,
Knoxville). Data acquisition was performed in 3D list mode. For
estimation of photon attenuation in the subjects a transmission scan
was carried out prior to the injection of 64Cu⊂7 using a 57Co point
source. Emission data were collected continuously for 60 min after
injection of 64Cu⊂7. The list mode data were sorted into sinograms
using a framing scheme of 12×10 s, 6×30 s, 5×300 s, and 3×600 s.
The data were attenuation corrected, and the frames were recon-
structed by Ordered Subset Expectation Maximization applied to 3D
sinograms (OSEM3D) with 14 subsets, 6 OSEM3D iterations, 2 max-
imum a posteriori (MAP) iterations, and 0.05 beta-value for smooth-
ing. The pixel size was 0.8 by 0.8 by 1.2 mm3, and the resolution in the
center of field of view was 1.85 mm. No correction for recovery and
partial volume effects was applied. The image files were processed
using the ASIPRO software (Siemens, CTI Molecular Imaging Inc.
Knoxville) and the ROVER software (ABX GmbH, Radeberg, Germany).
Summed frames from 30 to 90 s post-injection (p.i.) and 30 to 60 min
p.i. were used to define the regions of interest (ROI). The data were
normalized to the injected activity by using 64Cu standards from the
injection solution measured in a γ-well counter (Isomed 2000,
Dresden, Germany) cross-calibrated to the PET scanner and expressed
in SUV.
2.5.2. In vivo stability
The in vivo stability of 64Cu⊂7 was analyzed using rat arterial
blood samples at various time points after intravenous injection of
approximately 12 MBq of the radiotracer. For rat arterial blood
sampling, a catheter prefilled with heparinized saline was inserted
into the right carotid artery, and 100 μL blood samples were taken at
1, 3, 5, 10, 20, and 60 min after intravenous tracer administration. The
samples were immediately centrifuged to precipitate the blood cells.
A solution consisting of 50 μL CH3CN, 45 μL H2O, and 5 μL TFA was
added to 200 μL of the supernatant, and the mixture was then cooled
to 20 °C. After centrifugation, the supernatant was analyzed by radio-
HPLC using method 5. The chromatogram of 64Cu⊂7 (plasma, 0 min)
was obtained by addition of 64Cu⊂7 to rat blood and immediately
treating the mixture as described for the in vivo samples. Urine,
kidney and liver homogenates (10% in PBS) were treated analogously
to the arterial blood samples.
2.5.5. Autoradiography
The animals were sacrificed by CO2 inhalation, quickly frozen by
immersion in isopentane/dry ice solution at −75 °C. The frozen
animals were dried and stored at −18 °C until embedded in chilled
(4 °C), low viscosity, 3% carboxymethyl cellulose. Specimens were
sectioned coronally (40 μm) at −18 °C using a heavy-duty microtome
housed in a Jung Cryomacrocut (Leica Instruments GmbH, Nussloch,
Germany). The sections were stored at −18 °C and then dried for 1 h
in a lyophilizer (Sublimator 400, Martin Christ Gefriertrocknungsan-
lagen, Osterode, Germany) and exposed to imaging plates (BAS-SR
Imaging Plate, FUJI, Japan) together with a tissue activity standard
overnight. The imaging plates were scanned in the bio-imaging ana-
lyzer BAS 5000 (FUJI). The data were recorded as photo-stimulated
luminescence values (PSL), which are proportional to the activity of
2.5.3. Animals, feeding, husbandry and biodistribution studies in rats
All animal experiments were carried out according to the guide-
lines of the German Regulations for Animal Welfare. The protocol was
approved by the local Ethical Committee for Animal Experiment. The

A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
825
the exposed sample. The data analysis was carried out with the
3. Results and discussion
software program AIDA 4.20 (Raytest, Germany).
3.1. Synthetic approach
2.5.6. Statistical analysis
Data are expressed as means SD. One-way analysis of variance
(ANOVA) was used for statistical evaluation.
Cyclam derivatives 4 and 5 with one and two NT(8–13) fragments
were built-up by treating of H-NT(8–13)-OH 3 with cyclamtetrapropionic
Scheme 1. Synthetic procedure for the preparation of cyclam derivatives with one and two NT(8–13) fragments 4 and 5 (a) (i) excess methyl acrylate, MeOH, 6 days, r.t.; (ii) 1 M
NaOH, neutralization with DOWEX 50WX8, 80%, (b) EDC, sulfo-NHS, 0.1 M Na2HPO4, pH 8.5, 24 h, r.t., RP-HPLC, 4: (47%), 5: (12%).

826
A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
acid 2 using standard peptide coupling reagents EDC and sulfo-NHS
in aqueous media. Several synthetic routes for the preparation of 2
have been described in the literature. Condensation of cyclam with
ω-bromopropylethylesters followed by acid hydrolysis gave only a
modest yield of 2 [56]. The tetra hydrochloride of 2 can be achieved
by reaction of cyclam with methyl acrylate followed by acidification
with aqueous HCl solution in 70% yield [57]. In the same way, Michael
addition of acrylic acid with cyclam was proven high yielding [58,59].
We found that Michael addition of methyl acrylate with cyclam 1
followed by basic hydrolysis with aqueous NaOH solution is far
superior particularly with respect to purity of 2 to the previous
synthetic protocols applied. Neutralization of the sodium salt of
Scheme 2. Synthetic procedure for the preparation of the cyclam derivative with four NT(8–13) fragments 7 (c) acrylic acid chloride, 1 M NaOH, 50 min at 0 °C, 50 min at r.t., RP-
HPLC, 31%, (d) 6/1=10/1, H2O (CH3OH), 0.06 M NaOH, 40 °C, 2 weeks, RP-HPLC, 52%.

A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
827
Table 1
Table 3
UV/vis data of Cu(II) complexes with cyclam ligands 4, 5 and 7 and formation kinetics in
Partition coefficients (log Do/w) of the 64Cu complexes of 1, 2, 4, 5 and 7 at different pHs.
aqueous solution (2 mM ligand, 2 mM Cu(NO3)2).
pH
log Do/w
64Cu⊂1
log Do/w
64Cu⊂2
log Do/w
64Cu⊂4
log Do/w
64Cu⊂5
log Do/w
64Cu⊂7
Complex
λmax [nm] εmax [M−1 cm−1
]
kobs [M−1 s−1] (tCuL(50%) [min])
[Cu⊂4](NO3)2 706
[Cu⊂5](NO3)2 711
[Cu⊂7](NO3)2 710
105
160
186
7.92 (0.525)
2.64 (1.58)
25 °C: 0.014 (300)
40 °C: 0.107 (39)
7.2
7.4
7.6
−3.74
−3.76
−3.72
−3.85
−3.81
−3.80
−2.41
−2.45
−2.50
−2.28
−2.32
−2.45
−2.18
−2.23
−2.37
assigned to ligand-to-metal charge transfer (LMCT) transition in-
volving cyclam amine donor atoms. The broad band at about 710 nm
is caused by transitions from dxy, dxz and dyz-type orbitals of the
copper(II) center. LMCT and dd transitions were recorded as function
of time. A relatively slow attainment of complexation equilibrium was
obtained at room temperature, in particular for the peptide tetramer
7. This finding can be explained on the way that the Cu(II) ion is
bound by the peptide residue in the initial complexation step. Overall,
the complexation kinetics of macrocyclic amines with copper(II) in
aqueous media are quite complex [61–63]. Under the experimental
conditions chosen, a second-order formation of the non-radioactive
Cu(II) complexes with ligands 4, 5 and 7 was observed. The second-
order rate constants are summarized in Table 1, showing that an
increasing substitution degree of the cyclam core by NT(8–13)
fragments led to a significant deceleration in complexation kinetics.
While, 50% of Cu(II) was complexed within 0.5 and 1.5 min by ligands
4 and 5, five hours were needed in the case of the tetramer 7 at room
temperature. However, the formation kinetics can be significantly
accelerated at higher temperature. An increase of temperature from
25 to 40 °C gives an about eight-fold higher rate constant.
As already found for PEGylated and carbohydrated cyclam
derivatives, a red-shift of dd transition and an increase of molar
absorptivity were observed for Cu(II) complexes with NT(8–13)-
containing derivatives in comparison to cyclam [51,52]. The dd bands
of the copper(II) complexes with 4, 5 and 7 have the maximum at
about 710 nm which is indicative for apical interactions of pendant
arms of the copper(II)-N4 chromophore [64–67]. Molar absorptivity
rises as the number of N-substituents is increased. That reflects the
rising tetragonal distortion of the coordination geometry with higher
substitution of the cyclam donor atoms [68].
cyclamtetrapropionic acid with HCl gave 2 in 92% yield. The salt-free
compound can be achieved by treatment of the sodium salt of 2 with
cation exchanger instead of neutralization with HCl in 80% yield
(Scheme 1).
Coupling reactions in organic media were restricted by the low
solubility of 2 and H-NT(8–13)-OH 3 in organic solvents. Nonetheless,
numerous reactions have been performed in suspension (TBTU/
DIPEA, HATU/DIPEA, HATU/HOAt/NEt3, DSC/NEt3, DCMME, isobutyl
chloroformate/NEt3 in N-methylpyrrolidone, DSC/NEt3, DCC/HOBt in
dimethylformamide, HATU/HOAt/NEt3, TBTU, NMM in DMSO or
dimethylformamide) to give no evidence of cyclam derivatives
containing peptide residues. The desired coupling reaction did also
not occur in the presence of EDC either in water or DMSO. The only
successful way to couple peptide residues with cyclamtetrapropionic
acid 2 was obtained by activation of 2 with both EDC and sulfo-NHS
followed by the addition of the peptide 3 in aqueous solution.
However, only cyclam derivatives with one and two NT(8–13)
fragments 4 and 5 could be isolated in modest yield. Applying this
coupling reaction, the formation of products containing three or four
NT(8–13)-units was not observed even upon addition of a high excess
of peptides.
Taking into account that Michael acceptors react smoothly with
azamacrocycles, the synthetic protocol was changed to achieve the
desired derivative with four peptide residues. In a first step, H-NT(8-
13)-OH 3 was converted into the corresponding acrylamide 6 by
reaction of 3 with acrylic acid chloride in aqueous NaOH solution. The
tetrabranched NT(8–13) star-shaped multimer 7 comprising cyclam
as core element was synthesized by using Michael addition of acry-
lamides 6 to 1 according to the published synthetic route, grafting
alkylglucosylacrylamides to cyclam [60]. After a reaction time of two
weeks, the target compound 7 with four covalently bound NT(8–13)-
OH-units was isolated in 52% yield following purification by RP-HPLC.
It is worth mentioning that the surplus of peptide-acrylamide 6
remained intact and can be recycled (Scheme 2).
Dendritic compounds I (cf. Chart 1) with four appending NT(8–13)
fragments show improved biostability and interesting tumor-target-
ing properties [41,42]. For this reason, the NT(8–13) tetramer 7 has
been selected for radiolabeling with 64Cu, aiming at studies of
metabolic stability as well as biodistribution in tumor-bearing mice.
The first tests were performed with 0.1 M NH4OAc that is successfully
applied for 64Cu-labeling of bispidines, TACN and also cyclam de-
rivatives [69–71]. However, the rate of copper(II) complexation with
3.2. Complexation properties and radiochemistry
UV/vis experiments have been performed in order to achieve
information about the kinetics of complex formation of Cu(II) with 4,
5 and 7 in aqueous solution (Table 1). Upon addition of 1 equivalent
Cu(NO3)2 to 4, 5 and 7 in water, the absorption spectra of the free
ligands were changed and two new broad bands at 315 and 706–
711 nm appeared. The absorption band with a maximum at 315 nm
shows a high molar absorptivity (εN4000 M−1 cm−1) and can be
Table 2
Radiochemical yield (RCY) and tCuL(50%) value as a function of ligand concentration for
64Cu-labeling of 7 at 25 and 50 °C.
Concentration of 7
[μg/200 μL]
Temperature
[°C]
Duration
[h]
RCY
[%]
tCuL(50%)
[min]
20
20
10
20
30
100
25
25
50
50
50
50
3
5
1
1
1
1
91
95
80
N95
N95
100
54
25
12
8
Fig. 1. Competitive displacement of [3H]NT(8–13) by 4, 5, 7, Cu⊂7, neurotensin (NT)
and NT(8–13) on HT-29 cell membranes (mean SD).
5

828
A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
7 was found to be very slow. After 14 h, the 64Cu-labeling yield was
lower than 5% in ammonium acetate buffer solution at ambient
temperature. Such slow complexation kinetics were also obtained for
the labeling of cyclam-antibody conjugates under comparable
conditions [72]. Replacement of acetate by the non-coordinating 2-
[N-morpholino]ethanesulfonate anion causes a significant accelera-
tion. Thus, more than 90% radiochemical yield was achieved at very
low ligand concentration (c=20 μg/200 μL) even at room tempera-
ture after 3 h (Table 2). Increasing the temperature to 50 °C led to a
suitable labeling efficiency with yields higher than 95% within 1 h.
The radiochemical yield as a function of ligand concentration and
the time at which 50% of 64Cu is complexed are summarized in
Table 2. The labeling conditions as absolute activity, high As, and
ligand concentration result in a ratio of radioactive plus non-radio-
active copper(II) ions (63Cu+64Cu+65Cu; 20 MBq, As =200 GBq/μg
Cu) to 7 in the range of 1:30 (10 μg 7) to 1:300 (100 μg 7). Typical
pseudo first order n.c.a. radiochemical labeling reactions have a
precursor excess of 100 to 1000 and elucidate fast reaction courses of
such bimolecular reactions. As expected, kinetics of 64Cu-labeling
with an large excess of 7 over Cu(II) are faster in comparison to UV/vis
experiments, using equimolar concentrations of non-radioactive
copper(II) and ligands. Thus, 50% of 64Cu-7 is formed within 54 min
at room temperature under labeling conditions (cf. Table 2). This
corresponds to a five-fold acceleration regarding to UV/vis experi-
ments. By increasing of temperature and ligand concentration, the
radiochemical yield can be enhanced. Altogether, the NT(8–13)
tetramer 7 can be efficiently labeled with 64Cu in a 0.05 M MES/
NaOH buffer solution. For biodistribution and small animal PET
experiments, 20 to 30 μg of 7 dissolved in 200 μL 0.05 M MES-NaOH
buffer (pH=5.5) was radiolabeled with [64Cu]CuCl2 (200 MBq in
0.01 M HCl, specific activity N200 GBq/μmol). Incubation at 50 °C for
1 h yielded ≥99% radiochemical purity (radio-HPLC, method 4). In
order to get information about the chemical stability of 64Cu⊂7,
challenge studies in the presence of EDTA and cyclam were performed
using radio-HPLC. This allowed to clearly distinguish between the
64Cu complexes formed (64Cu⊂7, tR =12.5 min; 64Cu⊂EDTA,
tR =2.5 min; 64Cu⊂1, tR =4.5 min). In the presence of 1 mg EDTA,
64Cu⊂7 completely remains intact at least for 24 h, pointing to a high
stability that is comparable to copper(II) radionuclide complexes of N,
N′,N″-tris(2-pyridylmethyl)-cis,cis-1,3,5-triaminocyclohexane (tach-
pyr) ligands predicted as potential radiopharmaceuticals [73].
Challenge studies with a 20-fold and 100-fold excess of cyclam
showed a slow transchelation from 64Cu⊂7 to 64Cu⊂1. Radio-HPLC
results indicated that more than 90% of 64Cu⊂7 is unchanged in the
presence of cyclam up to 1 h. About 50% of 64Cu⊂7 could be detected
at a 20-fold cyclam excess after 1 day. Nearly 80% of the radiolabeled
tetramer was converted into 64Cu⊂1 after 24 h when a 100-fold
cyclam excess has been applied.
showed rapid blood clearance and excretion via the renal pathway
[66,67].
3.3. In vitro binding to the neurotensin receptor 1 (NTR1) in HT-29 cells
The binding of 7 and Cu⊂7 towards the neurotensin receptor 1
(NTR1) was determined in a competitive in vitro binding inhibition
assay using the human colon adenocarcinoma cell (HT-29) membrane
preparation. HT-29 cells were reported to express NTR1 on their
membranes, and they have been used as a suitable test system to
evaluate a variety of different neurotensin analogs [19]. The binding
inhibition assay was performed with [3H]neurotensin as reference
compound and radiotracer. The novel compounds inhibited the
binding of [3H]neurotensin in a dose dependent manner, showing
the typical sigmoid curves (Fig. 1).
In comparison to the determined IC50 value of 0.3 nM for the high
affinity ligand NT(8–13), the calculated IC50 values of 7.0 nM (4.6 to
10.6 nM, 95% confidence interval) for Cu⊂7 and 10.9 nM (7.1 to
16.9 nM) for 7 indicate a lower but still appropriate IC50 for in vivo
imaging of NTR1. The binding affinity of the apo-complex and Cu-
bound holo-complex7 was not different. This observation was not
surprising because the receptor–ligand interaction is suggested to be
determined by the NT(8–13) side chains. In this regard, an increasing
number of NT(8–13) side chains coupled to the cyclam core clearly
results in an enhancement of affinity. However, the affinity of the
tetravalent Cu⊂7 is lower than that of both the single NT(8–13) and
of neurotensin (IC50 1.4 nM). This could be caused by steric hindrance
of the peptide side chains and, on the other hand, the rigid cyclam
core. Hence, attaching the NT(8–13) in very close proximity of the
cyclam core reduces affinity in an extent that is not compensated by
multivalency. This finding is in agreement with results recently
described for NT(8–13) multimers II. The decrease of affinity can be
circumvented by introduction of a PEG spacer between core matrix
and NT(8–13) moieties [46]. Nevertheless, the affinity of Cu⊂7 to the
NTR1 was assumed to be high enough for in vivo binding.
3.4. In vivo studies
3.4.1. Stability in rodent models
Metabolic stability of compound 64Cu⊂7 in vivo was studied in
Wistar rats by radio-HPLC analysis of deproteinized plasma samples at
different time points after injection. The results indicate a substantial
in vivo degradation. The metabolic fractions of 64Cu⊂7 were 0.13 at
5 min p.i., 0.32 at 20 min p.i., and 0.80 at 60 min p.i. The main
metabolite showed an identical retention time (at 11.4 min) as the
64Cu-cyclam-tetraarginine complex which fits to one of the most
important peptidolytic cleavage sites of Arg8-Arg9, Pro10-Tyr11 and
Tyr11-Ile12 within NT(8–13) [77]. Consequently, the hydrolytic clea-
vage of the most sensitive peptide bonds at Arg8-Arg9 of the NT(8–13)
molecule is supposed to be the main metabolic reaction. This is
consistent with former data published by us [26,37]. In rat plasma low
abundance hydrophilic metabolites (eluting with retention times of
3.7 and 4.5 min, respectively) were observed, but not identified. The
metabolite also occurred in rat urine samples (Fig. 2).
Lipophilicity is an important parameter for the prediction of
absorption and distribution of radiotracers in biosystems. A compre-
hensive study on cyclen derivatives showed that the biodistribution
of their 64Cu complexes is remarkably influenced by the hydrophi-
licity/hydrophobicity of the derivatives. Increasing lipophilicity of the
complexes resulted in a higher liver accumulation compared to more
hydrophilic counterparts [74]. Copper(II) complexes of non-
substituted cyclen and cyclam ligands were expected to be very
hydrophilic [75]. However, quantitative lipophilicity data has not
previously been published. The radiotracer technique allows the
determination of precise log D values [76]. As summarized in Table 3,
log D values (1-octanol/buffer system) showed that 64Cu complexes
of cyclam 1 and cyclamtetrapropionic acid 2 almost completely
remain in the aqueous phase (N99.9%). The introduction of NT(8–13)
fragments into the ligand architecture leads to a slight increase of the
lipophilicity, yielding distribution ratios for 64Cu⊂4, 64Cu⊂5 and
64Cu⊂7 in the same range as obtained for copper(II) complexes of
bombesin conjugates with bispidine and TACN derivatives which
In order to investigate differences in rodent species, plasma and
urine samples obtained from mice at 60 min p.i. were analyzed.
Mouse plasma showed only the original compound (retention time
13 min) and the metabolite eluting with a retention time of 11.4 min.
The metabolite also occurred in mouse urine. The more hydrophilic
metabolites obtained in rat experiments could not be observed for
mice. Of note, in mouse plasma compound 64Cu⊂7 with a metabolic
fraction of about 0.5 at 60 min p.i. was more stable compared to rats
(data not shown in detail). Species-related differences revealed in
vivo stabilities of less than 1 min for non-stabilized neurotensin de-
rivatives [37,39]. Accordingly, the tetrabranched NT(8–13) cyclam
derivative shows a quite high in vivo stability (half-life: 34 min in rats,

A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
829
Fig. 2. Examples of radiochromatograms of deproteinized rat plasma (supernatant) (A, 1 min p.i.; B, 30 min p.i.; C, 60 min p.i.), and (D) urine at 60 min p.i. after application of
64Cu⊂7.
60 min in mice) which is in the range reported for doubly and triply
stabilized derivatives (99mTc-NT-XII: 45 min, 99mTc-NT-XIX: 84 min,
111In-NT-XI: 47 min in mice) [31,39].
3.4.2. Biodistribution studies
The biodistribution of 64Cu⊂7 was studied in rats with body
weight of 171.5 14.2 g (5 min) and 174.5 15.2 g (60 min) (Fig. 3),
and mice (body weight 35.4 1.4 g) after dynamic PET studies at
60 min p.i. (Fig. 4).
In rats the highest activity concentration after 5 min was observed
in liver, lung, spleen, kidneys, and adrenals. Beginning with the
highest activity concentration, this order was changed to kidney, liver,
spleen, and lung after 1 h. All other tissues contained a 64Cu activity
concentration lower than 1 SUV. In mice the organs with peak activity
concentration were the liver, kidney, and spleen at 60 min p.i. The
SUV data of rats and mice after 60 min p.i. match very well, except for
the liver (rat 5.0 0.1 SUV vs. mouse 13.5 3.0 SUV), and kidneys (rat
6.69 0.74 SUV vs. mouse 1.98 1.34 SUV) (Fig. 5).
However, the overall pattern of organ-specific biodistribution of
64Cu activity at 60 min p.i. did not substantially differ between mice
and rats. The different quantitative metabolism in the liver could
reflect the different amount of injected radiotracer per body weight.
The body weight of the rats is about 4.9 times higher than that of the
mice which corresponds with the relevant concentration relation. The
Fig. 3. Biodistribution of 64Cu⊂7 in rats at 5 and 60 min after single intravenous
application. The data represent the distribution of the 64Cu activity concentration (SUV)
in selected organs and tissues (mean SD of 4 animals each time point).

830
A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
Fig. 6. Representative serial transversal (1), coronal (2), and sagittal (3) images
obtained from dynamic small animal PET scans showing distribution of 64Cu activity
amount at 0.5 (A), 5 (B), and 45 min (C) after single intravenous injection of 64Cu⊂7 in
a HT-29 tumor-bearing nude mouse. Selected planes illustrate 64Cu activity entry into
the liver and the tumor (tumor is marked by arrows).
Fig. 4. Biodistribution of 64Cu⊂7 in different organs of mice 60 min after single
intravenous application. The data represent the distribution of the 64Cu activity amount
(A, %ID, mean SD of 6 animals) and 64Cu activity concentration (B, SUV, mean SD of
6 animals) in selected organs and tissues.
overall pattern of organ-specific biodistribution indicates that
compound 64Cu⊂7 became substantially entrapped by the reticulo-
endothelial system (liver and spleen). In this regard, the contribution
of, in particular, Kupffer cells (liver), marginal zone macrophages
Fig. 7. Ex vivo autoradiograms of 64Cu⊂7 of a HT-29 tumor-bearing nude mouse (A,
whole body coronal section), and representative organ sections of liver (B), kidney (C),
and tumor (D). The arrow marks the tumor.
Fig. 5. Comparison of biodistribution data (SUV, mean SD) between rats and mice.

A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
831
activity in the tumor was increasing over the time; the activity in the
muscle did not change. As result, the tumor to muscle ratio increased
up to 3.1 at 60 min p.i. (Fig. 8).
4. Conclusion
The modification of NT(8–13) to the acrylamide derivative 6
followed by Michael addition with cyclam was shown to be an efficient
route for the synthesis of tetrabranched compound 7. This reaction
offers promising opportunities for the creation of tetrabranched cyclam
derivatives with differently stabilized neurotensin units or other tumor-
targeting peptides. The star-shaped multimer 7 was efficiently
radiolabeled with 64Cu in MES/NaOH buffer which also can be expected
to be the case for other tetrabranched cyclam compounds. High binding
affinity of 64Cu⊂7 towards NTR1 receptor could be demonstrated in
vitro. Furthermore, the observed accumulation of 64Cu⊂7 in tumors
indicates substantial binding affinity of 64Cu⊂7 towards NTR1 receptor
in vivo. In this regard, the contribution of NTR2 and NTR3 to bind the
ligands also has to be considered. Altogether, substantial metabolic
degradation in vivo prevented a better contrast for sensitive imaging
tumors in small animal positron emission tomography studies. It is
noteworthy that the non-stabilized NT(8–13) part of the tetrabranched
64Cu⊂7 showed a significantly improved stability (half-life: 34 min in
rats, 60 min in mice) compared to other radiolabeled non-stabilized NT
(8–13) derivatives (half-life: b1 min). Application of stabilized neuro-
tensin derivatives to minimize or avoid metabolic degradation should
improve the in vivo targeting characteristics. In summary, the approach
offers promisingopportunities fordesign and synthesis of various highly
selective and potent tetrabranched peptide cyclam derivatives targeted
towards tumor-specific peptide receptors.
Abbreviations
BSA
bovine serum albumin
DCMME dichloromethylmethylether
Fig. 8. Time activity curves were derived from ROIs over the vena cava (circulating
blood pool), tumor, and muscle, expressed as SUVmean (A), and the tumor to muscle
ratio (B). The values are means SD of 8 animals.
DIPEA
DCC
N,N-Diisopropylethylamine
dicyclohexylcarbodiimide
DSC
disuccinimidylglutarate
EDC
1-ethyl-3-(3-dimethylaminocarbonyl)carbodiimide-
hydrochloride
end of bombardment
(spleen), but also mesangial cells (kidneys) to the observed organ-
specific accumulation cannot be excluded.
EOB
The overall high 64Cu accumulation in the liver accounted for the
original compound as determined by radio-HPLC measurement in
liver homogenates (data not shown). However, it could not be
excluded that some transchelation occurred after cellular uptake and
metabolic degradation. The different activity accumulation in the
kidneys between the both species in part can be ascribed to decreased
perfusion in mice kidneys during the PET study under anesthesia. The
main metabolite of 64Cu⊂7 was supposed to be the 64Cu-cyclam-
tetraarginine complex.
ESI-MS electrospray ionization mass spectrometry
HATU
O-(7-aza-benzotriazole-1-yl)-tetramethyl-uroniumhexa-
fluorophosphate
4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid
1-hydroxy-7-aza-benzotriazole
2-[N-morpholino]ethanesulfonic acid
N-methylmorpholine
HEPES
HOAt
MES
NMM
NT(8–13)neurotensin(8–13)
NTR
PET
Sulfo-NHS N-hydroxysulfosuccinimide
SUV
neurotensin receptor
positron emission tomography
3.4.3. PET
The biokinetics of 64Cu⊂7 in mice were studied using small
animal PET. Summarized images are presented in Fig. 6.
The tumor clearly appeared only in the late images, when the
tumor to muscle ratio reached 3.1. This was caused by the low
absolute uptake and the high blood background. The predominant
organ, appearing in the PET images is the liver. This is also shown
in the autoradiograms (Fig. 7); the liver has the highest activity
concentration.
standardized uptake value
O-(benzotriazol-1-yl)-N,N,N′,N′-tetramethyluroniumtetra-
fluoroborate
TBTU
TFA
trifluoroacetic acid
tCuL(50%) time at which 50% of Cu(II) is complexed
TMS
tR
Trimethylsilyl group
retention time
UV/visible
UV/vis
In the kidney, the activity is mainly located in the cortex and the
tumor shows a heterogeneous distribution. The PET kinetic analysis
unveils a fast distribution of 64Cu⊂7 in the body and blood followed
by an elimination phase with a half-life of 64Cu activity in mouse blood
of 117.8 s with a 95% confidence interval from 76.8 s to 252.5 s
(n=6). This was calculated using a monoexponential elimination
model from the time activity curves of ROIs over the vena cava. The
Acknowledgments
The authors would like to thank Brigitte Große, Regina Herrlich,
Karin Landrock, Sonja Lehnert and Andrea Suhr for the excellent
technical assistance. For providing 64Cu, Philipp Große-Gehling,

832
A. Röhrich et al. / Journal of Inorganic Biochemistry 105 (2011) 821–832
[37] R. Bergmann, M. Scheunemann, C. Heichert, P. Mäding, H. Wittrisch, M.
Kretzschmar, H. Rodig, D. Tourwe, K. Iterbeke, K. Chavatte, D. Zips, J.C. Reubi, B.
Johannsen, Nucl. Med. Biol. 29 (2002) 61–72.
[38] T. Ramenda, R. Bergmann, F. Wuest, Lett. Drug Des. Discovery 4 (2007) 279–285.
[39] F. Alshoukr, C. Rosant, V. Maes, J. Abdelhak, O. Raguin, S. Burg, L. Sarda, J. Barbet, D.
Tourwe, D. Pelaprat, A. Gruaz-Guyon, Bioconjug. Chem. 20 (2009) 1602–1610.
[40] V. Lungu, D. Chiper, L. Matei, C. Barna, I. Gruia, C. Tuta, C. Postolache, C. Cimpeanu,
J. Labelled Compd. Radiopharm. 53 (2010) 466–468.
[41] L. Bracci, C. Falciani, B. Lelli, L. Lozzi, Y. Runci, A. Pini, M.G. De Montis, A.
Tagliamonte, P. Neri, J. Biol. Chem. 278 (2003) 46590–46595.
[42] C. Falciani, M. Fabbrini, A. Pini, L. Lozzi, B. Lelli, S. Pileri, J. Brunetti, S. Bindi, S. Scali,
L. Bracci, Mol. Cancer Ther. 6 (2007) 2441–2448.
[43] C. Falciani, L. Lozzi, A. Pini, F. Corti, M. Fabbrini, A. Bernini, B. Lelli, N. Niccolai, L.
Bracci, Chem. Biol. Drug Des. 69 (2007) 216–222.
[44] A. Pini, C. Falciani, L. Bracci, Curr. Protein Pept. Sci. 9 (2008) 468–477.
[45] C. Falciani, A. Pini, L. Bracci, Expert Opin. Biol. Ther. 9 (2009) 171–178.
[46] C. Hultsch, B. Pawelke, R. Bergmann, F. Wuest, Bioorg. Med. Chem. 14 (2006)
5913–5920.
[47] C. Hultsch, M. Berndt, R. Bergmann, F. Wuest, Appl. Radiat. Isot. 65 (2007)
818–826.
[48] F. Wuest, C. Hultsch, M. Berndt, R. Bergmann, Bioorg. Med. Chem. Lett. 19 (2009)
5426–5428.
[49] D. Parker, Chem. Soc. Rev. 19 (1990) 271–291.
Stephan Preusche and Martin Walther are gratefully acknowledged.
This work was partially supported by the BMBF grant (contract no 03
NUK 006A).
References
[1] J.C. Reubi, A.M. Landolt, J. Clin. Endocrinol. Metab. 59 (1984) 1148–1151.
[2] J.C. Reubi, Endocr. Rev. 24 (2003) 389–427.
[3] M. de Visser, S.M. Verwijnen, M. de Jong, Cancer Biother. Radiopharm. 23 (2008)
137–157.
[4] S.M. Okarvi, Cancer Treat. Rev. 34 (2008) 13–26.
[5] I.U. Khan, A.G. Beck-Sickinger, Anticancer Agents Med. Chem. 8 (2008) 186–199.
[6] D. Zwanziger, A.G. Beck-Sickinger, Curr. Pharm. Des. 14 (2008) 2385–2400.
[7] L. Bodei, D. Ferone, C.M. Grana, M. Cremonesi, A. Signore, R.A. Dierckx, G. Paganelli,
J. Endocrinol. Invest. 32 (2009) 360–369.
[8] C. Ansquer, F. Kraeber-Bodéré, J.F. Chatal, Curr. Pharm. Des. 15 (2009) 2453–2462.
[9] M. de Jong, W. Breeman, D.J. Kwekkeboom, R. Valkema, E.P. Krenning, Acc. Chem.
Res. 42 (2009) 873–880.
[10] M.F. Tweedle, Acc. Chem. Res. 42 (2009) 958–968.
[11] R. Carraway, S.E. Leeman, The isolation of
a new hypotensive peptide,
neurotensin, from bovine hypothalami, J. Biol. Chem. 248 (1973) 6854–6861.
[12] C.M. Perou, T. Sørlie, M.B. Eisen, M. van de Rijn, S.S. Jeffrey, C.A. Rees, J.R. Pollack,
D.T. Ross, H. Johnsen, L.A. Akslen, O. Fluge, A. Pergamenschikov, C. Williams, S.X.
Zhu, P.E. Lønning, A.-L. Børresen-Dale, P.O. Brown, D. Botstein, Nature 406 (2000)
747–752.
[50] P.V. Bernhardt, P.C. Sharpe, Inorg. Chem. 39 (2000) 4123–4129.
[51] H. Stephan, G. Geipel, D. Appelhans, G. Bernhard, D. Tabuani, H. Komber, B. Voit,
Tetrahedron Lett. 46 (2005) 3209–3212.
[52] H. Stephan, A. Röhrich, S. Noll, J. Steinbach, R. Kirchner, J. Seidel, Tetrahedron Lett.
48 (2007) 8834–8838.
[13] I. Sehal, S. Powers, B. Huntley, G. Powis, M. Pittelkow, N.J. Maihle, Proc. Natl. Acad.
Sci. U. S. A. 91 (1994) 4673–4677.
[53] M.A. Avila-Rodriguez, J.A. Nye, R.J. Nickles, Appl. Radiat. Isot. 65 (2007)
1115–1120.
[14] M. Ocejo-Garcia, S.I. Ahmed, J.M. Coulson, P.J. Woll, Lung Cancer 33 (2001) 1–9.
[15] J.J. Maoret, D. Pospai, C. Rouyer-Fessard, A. Couvineau, C. Laboisse, T. Voisin, M.
Laburthe, Biochem. Biophys. Res. Commun. 203 (1994) 465–471.
[16] J.C. Reubi, B. Waser, H. Friess, M. Buchler, J. Laissue, Gut 42 (1998) 546–550.
[17] R.E. Carraway, A.M. Plona, Peptides 27 (2006) 2445–2460.
[18] B.M. Evers, Peptides 27 (2006) 2424–2433.
[19] C. Haase, R. Bergmann, J. Oswald, D. Zips, J. Pietzsch, Anticancer Res. 26 (2006)
3527–3533.
[20] R.M. Myers, J.W. Shearman, M.O. Kitching, A. Ramos-Montoya, D.E. Neal, S.V. Ley,
ACS Chem. Biol. 4 (2009) 503–525.
[21] C. Granier, J. van Rietschoten, P. Kitabgi, C. Poustis, P. Freychet, Eur. J. Biochem.
124 (1982) 117–124.
[22] E. Garćia-Garayoa, L. Allemann-Tannahill, P. Bläuenstein, M. Willmann, N.
Carrel-Remy, D. Tourwe, K. Iterbeke, P. Conrath, P.A. Schubiger, Nucl. Med. Biol.
28 (2001) 75–84.
[23] M. Bruehlmeier, E. Garćia-Garayoa, A. Blanc, B. Holzer, S. Gergely, D. Tourwe, P.A.
Schubiger, P. Bläuenstein, Nucl. Med. Biol. 29 (2002) 312–327.
[24] E. Garćia-Garayoa, P. Bläuenstein, M. Bruehlmeier, A. Blanc, K. Iterbeke, P. Conrath,
D. Tourwe, P.A. Schubiger, J. Nucl. Med. 43 (2002) 374–383.
[25] F. Buchegger, F. Bonvin, M. Kosinski, A.O. Schaffland, J. Prior, P. Bläuenstein, D.
Tourwe, E. Garćia-Garayoa, A. Bischof Delaloye, J. Nucl. Med. 44 (2003) 1649–1654.
[26] S. Seifert, J.-U. Künstler, E. Schiller, H.-J. Pietzsch, B. Pawelke, R. Bergmann, H.
Spies, Bioconjug. Chem. 15 (2004) 856–863.
[27] E. Garćia-Garayoa, V. Maes, P. Bläuenstein, A. Blanc, A. Hohn, D. Tourwe, P.A.
Schubiger, Nucl. Med. Biol. 33 (2006) 495–503.
[28] K.J. Zhang, R. An, Z.R. Gao, Y.X. Zhang, M.R. Aruva, Nucl. Med. Biol. 33 (2006)
505–512.
[29] V. Maes, E. Garćia-Garayoa, P. Bläuenstein, D. Tourwe, J. Med. Chem. 49 (2006)
1833–1836.
[30] T. Maina, A. Nikolopoulou, E. Stathopoulou, A.S. Galanis, P. Cordopatis, B. Nock,
Eur. J. Nucl. Med. Mol. Imaging 34 (2007) 1804–1814.
[31] E. Garćia-Garayoa, P. Bläuenstein, A. Blanc, V. Maes, D. Tourwe, P.A. Schubiger, Eur.
J. Nucl. Med. Mol. Imaging 36 (2009) 37–47.
[32] K. Chavatte, D. Terriere, L. Jeannin, K. Iterbeke, M. Briejer, J. Schuurkes, J.J.R.
Mertens, E. Bruyneel, D. Tourwe, J.E. Leysen, A. Bossuyt, J. Labelled Compd.
Radiopharm. 42 (1999) 423–435.
[33] M.H. de Boisferon, O. Raguin, C. Thiercelin, M. Dussaillant, W. Rostene, J. Barbet, A.
Pelegrin, A. Gruaz-Guyon, Bioconjug. Chem. 13 (2002) 654–662.
[34] M. de Visser, P.J.J.M. Janssen, A. Srinivasan, J.C. Reubi, B. Waser, J.L. Erion, M.A.
Schmidt, E.P. Krenning, M. de Jong, Eur. J. Nucl. Med. Mol. Imaging 30 (2003)
1134–1139.
[54] F. Szelecsényi, G. Blessing, S.M. Qaim, Appl. Radiat. Isot. 44 (1993) 575–580.
[55] D.W. McCarthy, R.E. Shefer, R.E. Klinkowstein, L.A. Bass, W.H. Margeneau, C.S.
Cutler, C.J. Anderson, M.J. Welch, Nucl. Med. Biol. 24 (1997) 35–43.
[56] C. Trabaud, J. Dessolin, M. Camplo, J.L. Kraus, Synth. Commun. 28 (1998) 311–317.
[57] V. Bulach, D. Mandon, J. Fischer, R. Weiss, Inorg. Chim. Acta 210 (1993) 7–9.
[58] V. Dahaoui-Gindrey, C. Lemote, H. Chollet, A.K. Mishra, C. Mehadji, R. Guilard,
New J. Chem. 19 (1995) 839–850.
[59] C. Lecomte, V. Dahaoui-Gindrey, H. Chollet, C. Gros, A.K. Mishra, F. Barbette, P.
Pullumbi, R. Guilard, Inorg. Chem. 36 (1997) 3827–3838.
[60] C. Larpent, A. Laplace, T. Zemb, Angew. Chem. 116 (2004) 3225–3229.
[61] C.-T. Lin, D.B. Rorabacher, G.R. Cayley, D.W. Margerum, Inorg. Chem. 14 (1975)
919–925.
[62] G.W. Liesegang, E.M. Eyring, in: R.M. Izatt, J. Christensen (Eds.), Synthetic
Multidentate Macrocyclic Compounds: Kinetic Studies of Synthetic Multidentate
Macrocyclic Compounds, Academic Press, New York-San Francisco-London, 1978,
pp. 245–286.
[63] H. Elias, Coord. Chem. Rev. 187 (1999) 37–73.
[64] M. Hediger, T.A. Kaden, Helv. Chim. Acta 66 (1983) 861–870.
[65] C.F.G.C. Geraldes, M.P.M. Marques, B. de Castro, E. Pereira, Eur. J. Inorg. Chem.
(2000) 559–565.
[66] Y. Dong, G.A. Lawrance, L.F. Lindoy, P. Turner, Dalton Trans. (2003) 1567–1576.
[67] A. de Castries, A. Escande, H. Fensterbank, E. Magnier, J. Marrot, C. Larpent,
Tetrahedron 63 (2007) 10330–10336.
[68] P. Comba, T.W. Hambley, M.A. Hitchman, H. Stratemeier, Inorg. Chem. 34 (1995)
3903–3911.
[69] S. Juran, M. Walther, H. Stephan, R. Bergmann, J. Steinbach, W. Kraus, F.
Emmerling, P. Comba, Bioconjug. Chem. 20 (2009) 347–359.
[70] G. Gasser, L. Tjioe, B. Graham, M.J. Belousoff, S. Juran, M. Walther, J.-U. Künstler, R.
Bergmann, H. Stephan, L. Spiccia, Bioconjug. Chem. 19 (2008) 719–730.
[71] C.J. Anderson, T.S. Pajeau, W.B. Edwards, E.L. Sherman, B.E. Rogers, M.J. Welch,
J. Nucl. Med. 36 (1995) 2315–2325.
[72] J.R. Morphy, D. Parker, R. Kataky, M.A.W. Eaton, A.T. Millican, R. Alexander, A.
Harrison, C. Walker, J. Chem. Soc. Perkin Trans. 2 (1990) 573–585.
[73] D. Ma, F. Lu, T. Overstreet, D.E. Milenic, M.W. Brechbiel, Nucl. Med. Biol. 29 (2002)
91–105.
[74] J. Yoo, D.E. Reichert, M.J. Welch, J. Med. Chem. 47 (2004) 6625–6637.
[75] T.M. Jones-Wilson, K.A. Deal, C.J. Anderson, D.W. McCarthy, Z. Kovacs, R.J.
Motekaitis, A.D. Sherry, A.E. Martell, M.J. Welch, Nucl. Med. Biol. 25 (1998)
523–530.
[76] H. Stephan, S. Juran, B. Antonioli, Ke. Gloe, Ka. Gloe, in: C.A. Schalley (Ed.),
Analytical Methods in Supramolecular Chemistry: Extraction Methods, Wiley-
VCH, Weinheim, 2007, pp. 79–103.
[35] M. Melis, E.P. Krenning, B.F. Bernard, M. de Visser, E. Rolleman, M. de Jong, Nucl.
Med. Biol. 34 (2007) 633–641.
[36] K. Chavatte, J.J.R. Mertens, P. Van Den Winkel, J. Labelled Compd. Radiopharm.
43 (2000) 1227–1234.
[77] P. Kitabgi, F. Denadai, C. Rovere, J. Bidard, Ann. NY Acad. Sci. 668 (1992) 30–42.