Synthesis and preliminary biological evaluation of a 99mTc- labeled hypericin derivative as a necrosis avid imaging agent

Article abstract of DOI:10.1002/jlcr.1468

Mono-[¹²³I]iodohypericin and mono-[¹²³I]iodohypericin monocarboxylic acid are iodine-123-labeled hypericin derivatives which have shown great promise in preclinical studies as necrosis avid imaging agents in animal models of infarction. In view of the more attractive properties of a ^99mTc-labeled hypericin derivative, we have synthesized a conjugate of protohypericin monocarboxylic acid with S-benzoylmercaptoacetyldiglycyl- diaminopentane in an overall yield of 15%. The conjugate was labeled with technetium-99m by exchange labeling at pH 10 in a labeling yield of 95% followed by photocyclization to yield ^99mTc-mercaptoacetyldiglycyl-1,5- diaminopentylene hypericincarboxamide (^99mTc-13). The negatively charged ^99mTc-13 complex was purified by reversed phase high-pressure liquid chromatography and the logP_7.4 was determined to be 2.36. In normal NMRI mice, the complex showed slow hepatobiliary clearance while plasma clearance was rapid. The tracer was evaluated in rats with reperfused hepatic infarction by ex vivo autoradiography, gamma counting and histochemical techniques. Unlike the radioiodinated hypericin derivatives, the new tracer agent did not show preferential uptake in necrotic tissue on autoradiography and gamma counting techniques. Conjugation of hypericin with a ^99mTc- chelate, resulting in a change in size, charge and lipophilicity, had a profound effect on the necrosis avidity of the tracer agent. The results show that ^99mTc-13 is not suitable for imaging necrosis. Copyright

Full text of DOI:10.1002/jlcr.1468

Research Article
Received 28 June 2007,
Revised 25 August 2007,
Accepted 17 September 2007
Published online 2 January 2008 in Wiley Interscience
(www.interscience.wiley.com) DOI: 10.1002/jlcr.1468
Synthesis and preliminary biological
evaluation of a Tc-labeled hypericin
99m
derivative as a necrosis avid imaging agent
Humphrey Fonge,a Lixin Jin,a Huaijun Wang,b Guy Bormans,a Yicheng Ni,b
Ã
and Alfons Verbruggena
Mono-[123I]iodohypericin and mono-[123I]iodohypericin monocarboxylic acid are iodine-123-labeled hypericin derivatives
which have shown great promise in preclinical studies as necrosis avid imaging agents in animal models of infarction. In
view of the more attractive properties of a 99mTc-labeled hypericin derivative, we have synthesized a conjugate of
protohypericin monocarboxylic acid with S-benzoylmercaptoacetyldiglycyl-diaminopentane in an overall yield of 15%. The
conjugate was labeled with technetium-99m by exchange labeling at pH 10 in a labeling yield of 95% followed by
photocyclization to yield 99mTc-mercaptoacetyldiglycyl-1,5-diaminopentylene hypericincarboxamide
(
99mTc-13). The
negatively charged 99mTc-13 complex was purified by reversed phase high-pressure liquid chromatography and the
log P7.4 was determined to be 2.36. In normal NMRI mice, the complex showed slow hepatobiliary clearance while plasma
clearance was rapid. The tracer was evaluated in rats with reperfused hepatic infarction by ex vivo autoradiography, gamma
counting and histochemical techniques. Unlike the radioiodinated hypericin derivatives, the new tracer agent did not show
preferential uptake in necrotic tissue on autoradiography and gamma counting techniques. Conjugation of hypericin with a
99mTc-chelate, resulting in a change in size, charge and lipophilicity, had a profound effect on the necrosis avidity of the
tracer agent. The results show that 99mTc-13 is not suitable for imaging necrosis.
Keywords: necrosis avidity; hypericin; 99mTc labeling
has more ideal imaging and availability characteristics than
iodine-123. Since technetium is a transition metal, labeling
Introduction
requires initial conjugation of hypericin to an appropriate
bifunctional chelating agent (BFC). The choice of the BFC is
critical in the design of the technetium-labeled tracer agent,
because of the alterations in charge, size and lipophilicity
introduced by coupling a bioactive compound with some 99mTc-
chelates. This study reports the synthesis of a novel technetium-
99m-labeled mercaptoacetyldiglycyl-1,5-diaminopentylene hy-
pericincarboxamide and its evaluation in normal mice and in rats
with reperfused hepatic infarction. Tracer uptake in reperfused
hepatic infarct was evaluated by ex vivo biodistribution studies,
autoradiography, triphenyltetrazolium chloride (TTC) and hema-
toxylin and eosin (H&E) staining techniques for histological
correlations.
Cell death occurs by two distinct processes, i.e. programmed cell
death, called apoptosis and necrosis.1,2 Necrosis and apoptosis
play a central role in the pathogenesis of various disorders and
results from different types of insults. Their non-invasive
identification and localization are therefore very important in
the diagnosis, prognosis and assessment of response to
revascularization therapies. This is particularly the case in acute
myocardial infarction (AMI) whereby early diagnosis and knowl-
edge of infarct size should permit a timely intervention to
salvage the organ as well as to assess response to treatment,
which is important for survival and quality of life of the
patients.3,4 99mTc-pyrophosphate, 111In/99mTc-antimyosin Fab
antibody and 99mTc-glucarate have all to some extent been
used in nuclear medicine to locate and quantify infarct size5–7
but none has optimal imaging characteristics. These can be
summarized as rapid localization to the infarct, high avidity and
specificity for necrotic tissue and reasonable duration of scan
positivity.
aLaboratory of Radiopharmacy, Faculty of Pharmaceutical Sciences, K.U. Leuven,
BE-3000 Leuven, Belgium
Our group has been focusing on infarct avid agents for
contrast enhancement in magnetic resonance imaging (MRI)
and single photon-emission-computed tomography (SPECT).8–10
We have shown that mono-[123I]iodohypericin ([123I]MIH, Figure
1(A))9 and mono-[123I]iodohypericin monocarboxylic acid
([123I]MIHA, Figure 1(B))10 are very avid for necrotic tissue in
different animal models of necrosis. However, technetium-99m
bDepartment of Radiology, Faculty of Medicine, K.U. Leuven, BE-3000 Leuven,
Belgium
*Correspondence to: Alfons Verbruggen, Herestraat 49, Onderwijs & Navorsing 2
Box 821, BE-3000 Leuven, Belgium.
E-mail: alfons.verbruggen@pharm.kuleuven.be
J. Label Compd. Radiopharm 2008, 51 33–40
Copyright r 2008 John Wiley & Sons, Ltd.

H. Fonge et al.
include: S-benzylmercaptoacetyl diaminoethane, N,N0 bis-BOC
triethylene triamine, ethyl iminodiacetate derivatives (ethyl or
methyl esters), S-benzylmercaptoacetyldiglycyllysine and S-
benzylmercaptoacetylcysteinyldiglycine. Using S-benzylmercap-
toacetyl diaminoethane (initially chosen because of enhanced
stability of the S-benzyl-precursor over the S-benzoyl for
example) as BFC, the resulting hypericin conjugate, S-benzyl-
mercaptoacetyldiglycyl-1,2-diaminoethylene hypericincarboxa-
mide, could not be labeled because of instability under
different conditions used to deprotect the thiol group. For this
reason, the pentyl linker with an S-benzoyl-protected mercap-
toacetyldiglycine derivative was chosen as the BFC. The MAG3-
like system is known to form a stable complex with technetium-
99m. S-benzoylmercaptoacetyldiglycine (S-Bz-MAG2) was
synthesized as reported by Brandau et al.,11 coupled to N-BOC
substituted 1,5-diaminopentane and after removal of the -BOC
protective group the resulting intermediate was conjugated
with protohypericin carboxylic acid (Scheme 1, Scheme 2 and
Scheme 3). The overall synthetic yield of the novel hypericin
derivative (compound 12) was 15%. This allowed a one-pot
deprotection and labeling with a radiochemical yield of 95%. As
shown in Figure 2, a reversed phase high-pressure liquid
chromatography (RP-HPLC) analysis of the labeling reaction
mixture showed the presence of mainly two peaks, probably
corresponding to 99mTc-mercaptoacetyldiglycyl-1,5-diamino-
Figure 1. Chemical structures of iodine-123-labeled hypericin derivatives with
avidity for necrotic tissue. (A) R 5 ꢀCH3 5 mono-[123I]iodohypericin ([123I]MIH) and
(B) R 5 ꢀCOOH 5 mono[123I]iodohypericin monocarboxylic acid ([123I]MIHA).
Results and discussions
Chemistry and radiolabeling
Scheme 1 shows the synthetic procedure for the selected BFC, S-
benzoylmercaptoacetyldiglycyl-1,5-diaminopentane. Synthesis
of the precursor protohypericin monocarboxylic acid (9) was
carried out starting from emodin as shown in Scheme 2.10
Attempts to derivatise hypericin via an acidic phenol (to obtain
pentylene
protohypericincarboxamide
(
99mTc-proto13,
ether derivatives) at the ‘bay region’ (pk ¼ 1:7) resulted in ether
a
tR 5 32.3 min) and 99mTc-mercaptoacetyldiglycyl-1,5-diamino-
pentylene hypericincarboxamide (99mTc-13, tR 5 32.5 min), re-
spectively. Photocyclization necessary to convert 99mTc-proto13
to 99mTc-13 was done post labeling by irradiating with a 400-W
halogen lamp for 30 min. Monitoring of the cyclization reaction
using RP-HPLC showed conversion of the mixture of peaks to a
conjugates that were unstable under different deprotection
conditions and/or during the labeling step. As a result amide
derivatives were preferred because of enhanced stability. This
was done starting from protohypericin monocarboxylic acid
(Scheme 3). Examples of BFCs that were used to derivatize
hypericin for labeling with a [Tc(V)O]31 or a [Tc(CO)3]11 core
Scheme 1
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 33–40

H. Fonge et al.
Scheme 2
single peak with a slight shift in retention time for the originally significantly bound to plasma lipoproteins.
most abundant peak (Figure 2(B)), consistent with the conver-
sion of the radiolabeled protohypericin derivative to the cyclized had a much faster hepatobiliary clearance. Urinary clearance of
[
123I]MIH and
[
123I]MIHA also showed similar initial high liver uptake, but
intended hypericin derivative.
the new tracer agent was very low.
Partition coefficient
Necrosis avidity
The partition coefficient (log P7.4) of 99mTc-13 was found to be
2.36 (70.12). For comparison, the log P7.4 of [123I]MIH and
Necrosis avidity of the tracer was evaluated in rats with
reperfused hepatic infarction by ex vivo autoradiography and
gamma counting in correlation with TTC and H&E staining
techniques. The model of reperfused hepatic infarction is
preferred in screening experiments of novel compounds to a
model of occlusive/reperfused AMI because it is easier to make
and because of the fact that very reproducibly an infarct is
[
123I]MIHA was previously determined to be 3.0812 and 1.4710,
respectively.
Biodistribution in normal mice
Biodistribution was evaluated in normal NMRI mice and the obtained with a low mortality rate of the animals. Owing to the
results are shown in Table 1. The tracer was slowly cleared via limited potential clinical applications of the model it therefore
the hepatobiliary pathway resulting in high amounts of tracer in serves mainly as a reliable first screening tool.9,10 Typically, the
the intestines, amounting to 33 and 63% of injected dose (%ID) necrosis/viable liver tissue activity ratio for an avid tracer is low
at 4 and 24 h post injection (p.i.), respectively. Blood clearance early after injection (o4 h p.i.) but increases to a maximum value
was rapid with only 0.5% ID per gram blood (%ID/g) at 30 min at later time points (47 h p.i.) due to specific uptake in necrotic
p.i. In previous studies, it was found that [123I]MIH shows high tissue and clearance from viable liver tissue.9 Necrosis was
affinity for plasma lipoproteins and has
a slow plasma confirmed by TTC staining (TTC negative areas). The images of
clearance.13 The faster plasma clearance of 99mTc-13 as TTC staining did not fully match with those of autoradiography
compared with that of [123I]MIH suggests that 99mTc-13 is not and H&E staining due to disruptions that occurred during
J. Label Compd. Radiopharm 2008, 51 33–40
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

H. Fonge et al.
Scheme 3
Figure 2. RP-HPLC chromatograms of labeling reaction mixture. (A) 99mTc-protohypericin derivative (99mTc-proto13, before ring closure) and (B) 99mTc-hypericin derivative
99mTc-13, after ring closure by irradiation with a 400-W halogen lamp). (See section on instruments for HPLC conditions.)
(
freezing and microtome sectioning. Region of interest delineated by the tracer at 9 h p.i.15 123I]MIHA showed even
[
analysis on autoradiograms shows no preferential uptake of more favorable characteristics (high avidity and rapid clearance
tracer in necrotic tissue (Figure 3) at 4 and 24 h p.i. Low from non-target organs) than [123I]MIH as a necrosis avid
uptake in necrotic tissue was also confirmed by gamma imaging agent.10 The results of the current study indicate that
counting (necrotic to viable tissue activity ratios were 0.5 the presence of the 99mTc-chelate in the new tracer agent has a
and 0.8 at 4 and 24 h p.i., respectively). In rats with reperfused negative impact on the necrosis avidity, probably as a result of
hepatic infarction
[
123I]MIH showed specific uptake in the charge and /or size alterations associated with the
necrotic tissue with a necrotic to viable tissue activity ratio of conjugation of hypericin with a negatively charged rather bulky
3.1 at 24 h p.i.14 In reperfused AMI necrotic to viable tissue chelate. Probably, also a reasonable plasma half-life is a
uptake ratio of [123I]MIH was as high as 81 by autoradiography requirement for sufficient accumulation of tracer agents in
at 24 h p.i. and myocardial infarction was unequivocally necrotic tissue.
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 33–40

H. Fonge et al.
Table 1. Biodistribution of 99mTc-mercaptoacetyldiglycyl-1,5-diaminopentylene hypericincarboxamide (99mTc-13) in normal
mice (nX3 per time point)
% Injected dose (7SD)
% Injected dose/g (7SD)
Organ
30 min
4 h
24 h
30 min
4 h
24 h
Urine
Kidneys
Liver
Spleen1pancreas
Lungs
Heart
Intestines1feces
Stomach
Brain
0.1 (70.0)
1.5 (70.7)
75.5 (73.6)
4.4 (73.0)
1.7 (70.2)
1.1 (70.2)
5.4 (73.2)
1.9 (71.4)
0.0 (70.0)
0.9 (70.3)
0.7 (70.9)
0.8 (70.0)
57.2 (72.6)
3.0 (70.7)
0.4 (70.2)
0.0 (70.0)
33.3 (71.1)
1.5 (71.8)
0.0 (70.0)
0.6 (70.2)
2.1 (71.2)
0.6 (70.1)
30.6 (72.6)
1.9 (70.4)
0.2 (70.1)
0.0 (70.0)
62.9 (72.3)
0.8 (70.5)
0.0 (70.0)
0.1 (70.1)
ND
ND
1.8 (70.0)
ND
1.4 (70.2)
2.2 (71.2)
36.3 (73.9)
13.4 (78.4)
5.3 (70.6)
0.0 (70.0)
ND
ND
0.0 (70.0)
0.5 (70.1)
34.9 (71.8)
14.3 (74.7)
2.4 (70.9)
0.0 (70.0)
ND
ND
0.0 (70.0)
0.3 (70.1)
19.4 (73.1)
9.6 (7 3.9)
0.9 (7 0.2)
0.0 (70.0)
ND
ND
0.0 (70.0)
0.1 (70.1)
Blood
Note: ND, not determined; SD, standard deviation.
Figure 3. Ex vivo analysis of 99mTc-13 uptake in rat with reperfused hepatic infarction. (A) Autoradiogram of necrotic liver lobe with low tracer uptake in the necrotic
regions (arrow) while adjacent viable tissue shows normal tracer distribution and (B) TTC-stained section of the necrotic liver lobe with arrow pointing to regions of severe
necrosis (TTC negative). This figure is available in color online at www.interscience.wiley.com/journal/jlcr.
thick silica gel 60 plates with UV fluorescence (lot 610/284,
Macherey-Nagel, Du¨ren, Germany) and spots were visualized
Experimental procedure
Reagents
with UV light (Spectroline, Westbury, NY). TLC plates were
developed using the following mobile phase systems: system A:
toluene/EtOAc/formic acid 50/40/10 V/V or system B: CHCl3/
methanol/acetic acid 60/40/1 or system C: CH2Cl2/acetone/
acetic acid 70/30/1. RP-HPLC analysis was performed on a
LaChrom Elite system (Merck-Hitachi, Darmstadt, Germany)
using an XTerra RP-C18 column (10 mm, 10 mm ꢁ 250 mm;
Waters, Milford, USA). The compounds were eluted from the
column using mixtures of 0.05 M ammonium acetate buffer (pH
6.8) (solvent A) and ethanol (solvent B): 0–10 min: linear gradient
from 100% A to 40% A; 10–25 min: 40% A; 25–35 min: linear
gradient to 100% B, at a flow rate of 2.2 mL/min over 35 min. The
column effluent was monitored using a UV-detector set at
254 nm and the output signal was acquired on a Laura Lite
system (Lablogic, Sheffield, UK). For the analysis of radiolabeled
compounds, the HPLC eluate was led over a 3-in NaI(Tl)
scintillation detector connected to a single-channel analyzer.
Thioglycolic acid, benzoyl chloride, N-Boc-1,5-diaminopentane,
diglycine, N,N0-ethyldiisopropylamine, N,N0-dicyclohexylcarbodii-
mide and N-hydroxysuccinimide were obtained from Sigma
Aldrich (Steinheim, Germany). All solvents were obtained from
Acros Organics (Geel, Belgium) and were used as purchased.
Emodin was a kind gift from Prof. Peter de Witte (Pharmaceutical
Biology, K.U. Leuven). Generator eluate containing Na99mTcO4
was obtained from an Ultratechnekow generator (Tyco Health-
care, Petten, The Netherlands).
Instruments
1H NMR spectra were acquired with a Gemini 300 MHz spectro-
meter (Varian, Palo Alto, CA). Chemical shifts are reported in
parts per million (ppm) relative to TMS (d 5 0). Thin-layer
chromatography (TLC) was performed on normal phase 0.2-mm
J. Label Compd. Radiopharm 2008, 51 33–40
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

H. Fonge et al.
Radioactivity counting for biodistribution studies was done pale brown oil. Compound 6 was precipitated from the oil after
using an automated system with a gamma counter (3-in NaI(Tl) addition of acetonitrile. TLC: Rf 5 0.4 (system B). Accurate mass
well crystal) coupled to a multi-channel analyzer in a sample [C18H27N4O4S1H]1, theoretical 395.1747 Da, found 395.1716 Da.
changer (Wallac, 1480 Wizards 30in, Turku, Finland). The results
Tri-acetylemodin10 (7):
were corrected for background radiation and physical decay The reaction product (compound 7) consisted of a mixture of di-
during counting. Accurate mass measurement was performed and tri-acetylated emodin. Yield 99%. Accurate mass di-
on a time-of-flight mass spectrometer (LCT, Micromass, Man- acetylated emodin [C19H14O7-H]ꢀ theoretical 352.0739 Da, found
chester, UK) equipped with an electrospray ionization (ESI) 352.0694 Da.
interface, operated in positive (ES1) or negative (ESꢀ) mode.
Samples were infused in acetonitrile/water using a Harvard 22 Yield 98%. Accurate mass of tri-acetylated emodic acid
Tri-acetyl Emodic Acid10 (8):
syringe pump (Harvard instruments, Massachusetts, USA). [C21H14O10-H]ꢀ : theoretical 425.0514 Da, found 425.0547 Da.
Accurate mass determination was performed by co-infusion
Protohypericin Monocarboxylic Acid (9) and Hypericin Mono-
with a 10 mg/mL solution of a reference as internal calibration carboxylic Acid (10):
standard. Acquisition and processing of data were done using Emodin (0.1 g, 0.37 mmol), compound 8 (0.32 g, 0.75 mmol) and
Masslynx software (Micromass, version 3.5).
hydroquinone (0.162 g, 1.5 mmol) were dissolved in 5 mL of
All animal experiments were conducted with the approval of 0.8 M KOH protected from light in a capped vial placed in a
the institutional ethical committee for conduct of experiments heavy steel reaction vial holder. The vial was placed in an
on animals.
oven at 1551C and allowed to react for 5 days during which it
was agitated every 24 h to allow mixing. After cooling, the
reaction mixture was acidified to pH 4–4.5 with 1 M HCl and
the dark colored precipitate was separated by centrifugation,
after which the filtrate was extracted three times with 2%
NaHCO3 and EtOAc. The aqueous layer was evaporated in vacuo
and the residue was dissolved in 25 mL methanol and
applied on a silica gel column. Purification was afforded
by gradient elution starting from 100% CH2Cl2 to CH2Cl2/
MeOH (80/20 V/V). The first major fraction was collected,
evaporated in vacuo and repurified on silica gel. Evaporation
of solvents yielded the deep purple solid 9. The residue was
dissolved in dry methanol and the solution was irradiated with a
400-W halogen lamp for 30 min to give compound 10. TLC:
Rf 5 0.7 (system A, compounds 9 and 10). Accurate mass
of 9 [C30H14O10-H]ꢀ theoretical 533.0510 Da, found 533.0478 Da,
accurate mass of 10 [C30H16O10-H]ꢀ theoretical 535.0703 Da,
found 535.0671 Da. 1H NMR 9 (DMSO) d 7.96 (s, CH-11),
7.67 (s, CH-9), 7.28 (s, CH-14), 7.18 (s, CH-2, CH-5) 6.76 (s, CH-
12), 2.50 (s, CH3).
N-Succinimidyl protohypericin monocarboxylic acid (11):
Compound 9 (0.1 g, 0.187 mmol) and NHS (25 mg, 0.22 mmol)
were dissolved in THF (20 mL). The solution was cooled to 01C
and DCC (45 mg, 0.22 mmol) dissolved in THF (10 mL) was added
dropwise over a 5 min period with stirring. The mixture was
further stirred at 01C for 1 h and then at RT overnight.
Dicyclohexyl urea was filtered off and the filtrate evaporated
to yield a dark purple residue 11. Yield 70%. TLC: Rf 5 0.6
(system A). Accurate mass [C34H18NO12-H]ꢀ theoretical
682.0834 Da, found 682.0843 Da.
Chemistry
N-Hydroxysuccinimidyl-S-benzoylthioglycolate11 (2):
Yield 20.7 g (70.6%). TLC: Rf 5 0.85 (system C), Rf 5 0.7 (system B).
Accurate mass [C13H10NO5S1Na]1, theoretical 316.0250 Da,
found 316.0258 Da.
S-Benzoylmercaptoacetylglycylglycine11 (3):
Yield 6.75 g (72.2%). TLC: Rf 5 0.1 (system C). Accurate mass
[C13H13N2O5S-H]ꢀ, theoretical 309.0551 Da, found 309.0563 Da.
1H NMR (DMSO-d6) d 8.5 (t, 1H, CONH), 8.2 (t, 1H, CONH), 7.9–7.5
(m, 5H, ArH), 3.9 (s, 2H, SCH2) 3.7 (d, 4H, 2 ꢁ CH2).
N-Succinimidyl-S-benzoylmecrcaptoacetylglycylglycine (4):
Compound 3 (3.10 g, 0.01 mol) and NHS (1.15 g, 0.01 mol) were
dissolved in CH2Cl2/DMF 75/25 V/V. The solution was cooled to
01C and DCC (2.48 g, 0.012 mol) dissolved in 25 mL CH2Cl2 was
added dropwise while stirring. It was further stirred at 01C for 1 h
and at room temperature (RT) for 16 h. The resulting precipitate
was filtered off and washed with CH2Cl2/DMF 75/25 V/V. The
filtrate was evaporated under reduced pressure and compound
4 was recrystallized from ethyl acetate as a white powder.
Yield 3.42 g (84%). TLC: Rf 5 0.4 (system C). Accurate mass
[C17H16N3O7S1Na]1 theoretical 430.0679Da, found 430.0612Da.
1H NMR (DMSO-d6) d 8.6 (t, 2H, 2 ꢁ CONH), 7.9–7.5 (m, 5H, ArH),
4.3 (d, 2H, CH2), 3.9 (s, 2H, SCH2), 3.8 (d, 2H, CH2), 2.8 (s, 4H,
2 ꢁ COCH2).
S-Benzoylmercaptoacetyldiglycyl-1-amino-5-N-BOC-aminopen-
tane (5):
To a solution of compound 4 (1.06 g, 2.6 mmol) in 25 mL THF/
DMF 4/1 V/V, a mixture of N-Boc-1,5-diaminopentane (0.578 g,
2.86 mmol) and triethylamine (0.41 mL, 3.12 mmol) in 5 mL THF
was added with stirring at RT. After stirring overnight, the
solvents were evaporated and a white precipitate was obtained
after addition of 25 mL water. Yield 1.2 g (93.3%). TLC: Rf 5 0.9
(system B). Accurate mass [C23H33N4O6S1Na]1 theoretical
517.2091 Da, found 517.2123 .Da 1H NMR (DMSO-d6) d 8.5–8.2
(m, 4H, 4 ꢁ CONH), 7.9–7.4 (m, 5H, ArH), 3.8 (s, 2H, SCH2), 3.7 (d,
2H, CH2), 3.6 (d, 2H, CH2), 3.0–2.8 (m, 4H, 2 ꢁ CH2), 1.7 (d, 4H,
2 ꢁ CH2), 1.4 (s, 9H, 3 ꢁ CH3), 1.2 (m, 2H, CH2).
S-Benzoylmercaptoacetyldiglycyl-1,5-diaminopentane (6):
To a reaction flask under N2, containing compound 5 (100 mg)
at ꢀ51C was added 2 mL TFA. The solution was stirred at ꢀ51C
for 20 min and then at RT for 40 min. TFA was evaporated
azeotropically with hexane under reduced pressure to give a
S-Benzoylmercaptoacetyldiglycyl-1,5-diaminopentylene protohy-
pericin-carboxamide (12):
To a solution of compound 6 (0.158 g, 0.4 mmol) in 5 mL DMF a
solution of compound 11 (0.12 g, 0.187 mmol) and diisopropy-
lethylamine (70 mL, 0.41 mmol) in THF (20 mL) was added
dropwise. The reaction mixture was further stirred at RT and
monitored on TLC (system A). The reaction was complete after
15 h. The solvents were evaporated under reduced pressure and
compound 11 was obtained after two-fold purification on RP-
HPLC using an isocratic mobile phase consisting of 0.05 M
NH4OAc/acetonitrile 40/60 V/V, employing a semi-preparative
column (XTerras prep RP18 10 mm, 10 mm ꢁ 250 mm) with a
flow rate of 2.5 mL/min. Yield 60%. TLC: Rf 5 0.1 (system A).
Accurate mass [C48H38N4O13S-H]ꢀ, theoretical 911.2240 Da,
found 911.2162 Da.
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 33–40

H. Fonge et al.
Radiolabeling with technetium-99m
Evaluation in animal model of hepatic infarction
99mTc-mercaptoacetyldiglycyl-1,5-diaminopentylene hypericincar- Rat model of reperfused hepatic infarction: Four adult Wistar rats
boxamide (99mTc-13):
weighing 350–450 g were anesthetized with i.p. injection of
s
´
Deprotection (removal of the S-benzoyl group) and labeling pentobarbital (Nembutal ; Sanofi Sante Animale, Brussels,
with 99mTc were achieved in a one-pot reaction by a classical Belgium) at a dose of 40 mg/kg. Under laparotomy, reperfused
exchange labeling method at pH 10. This was done by hepatic infarction was induced by clamping the hilum of
consecutively adding 20 mg NaK tartrate in 0.3 mL H2O, 100 mg the right liver lobe for 3 h. After reperfusion by declamping
stannous chloride dihydrate in 25 mL 0.05 N HCl and 400–600 hepatic inflow, the abdominal cavity was closed with two-
MBq Na99mTcO4 in 0.3–0.5 mL saline to a solution of 0.5 mg of S- layered sutures and the animals were allowed to recover for at
benzoylmercaptoacetyldiglycyl-1,5-diaminopentylene protohy- least 8 h. An activity of 18.5–74 MBq of the 99mTc-labeled
pericin-carboxamide in 0.3 mL EtOH, buffered with 0.2 mL of hypericin derivative formulated in 0.5 mL water/PEG 400 80/
0.5 M phosphate buffer at pH 10. The reaction mixture was 20 V/V was injected via a tail vein and the rats were sacrificed
heated at 1001C for 20 min followed by irradiation with a 400-W under anesthesia at 4 h (n 5 2) or 24 h (n 5 2) p.i. followed by ex
halogen lamp for 30 min in order to cyclize the compound. The vivo studies.
labeled compound was purified by RP-HPLC after which the
Ex vivo studies: For ex vivo studies, the necrotic tissues
solvents were evaporated by a gently N2 flush at 401C prior to (liver and muscle) and viable control tissues were thoroughly
formulation in water/polyethylene glycol 400 (PEG 400) 80/20 V/ washed with saline (41C) to remove blood activity. The
V for biodistribution studies. Reanalysis of the HPLC purified tissues were then stained separately in TTC solution (1%
tracer was done by RP-HPLC at different time points post solution in normal saline) at 371C for 15 min. Guided by
labeling to verify its stability.
TTC, the necrotic and viable tissues were identified and
immediately weighed and activity concentration in the
tissues was counted in a 3-in NaI(TI) scintillation well detector
mounted in a sample changer and tracer concentration was
expressed as % ID/g tissue or counts per minute (CPM)/g tissue.
Digital photographs were taken prior to and after staining.
The tissues were then quickly frozen in isopentane cooled in
liquid nitrogen and separately cut immediately with a cryotome
(Microm HM 550, Walldorf, Germany) into 5–50 mm thick
serial sections which then were mounted on slides. Autoradio-
grams were obtained from these sections by exposing the
slides for 2 days to a high-performance phosphor screen
(superresolution screen; Canberra-Packard, Meriden, CT).
The images were analyzed with Optiquant software (Canberra-
Packard) and the activity concentration in the autoradiograms
was expressed in digital light units (DLU)/mm2. The
relative tracer concentration in the necrotic and viable sections
of the tissues was estimated by regions of interest (ROIs) analysis
of the necrotic and the viable regions of 30 and 50-mm
thick sections of the autoradiograms of all slices. After exposure
for autoradiography, the samples were stained with H&E
Octanol/phosphate buffer partition coefficient
As a measure of the lipophilicity of the RP-HPLC-purified 99mTc-
labeled hypericin derivative, its octanol/phosphate buffer
partition coefficient was determined using a modification of
the method of Yamauchi et al.16 In a test vial, 50 mL of the HPLC
purified radiolabeled compound was mixed with 2 mL of 1-
octanol and 2 mL of 0.025 M phosphate buffer pH 7.4. The test
vial was vortexed at RT for 2 min and then centrifuged at 1700g
for 10 min. Approximately 50 mL of the 1-octanol phase and
0.5 mL of the phosphate buffer phase were pipetted into
separate tared test tubes with adequate care to avoid cross
contamination between the phases. The volume of fluid
pipetted was calculated by dividing the net weight of the fluid
by its density. The radioactivity of the test tubes was counted
using a 3-in NaI(Tl) scintillation detector mounted in a sample
changer (Wallac). Corrections were made for background
radiation and physical decay during counting. The octanol/
buffer partition coefficient, P, was calculated using the following
equation:
using
a conventional procedure. The H&E stained slices
were used to examine microscopic evidence of myocardial
necrosis.
cpm=mL in octanol
P ¼
cpm=mL in buffer
Conclusion
where cpm 5 counts per min.
The results show that the conjugation of a 99mTc-chelate to
hypericin alters the necrosis avidity of the resulting derivative,
presumably because of the changes in lipophilicity, size and
charge of the molecule, factors which on their turn may affect
residence time in plasma and protein binding. The novel 99mTc-
mercapto-acetyldiglycyl-1,5-diaminopentylene protohypericin-
carboxamide is therefore not suitable for visualization of
necrotic cell death in vivo by SPECT.
Biodistribution in normal mice
Twelve normal NMRI mice (n 5 4 per time point) weighing
20–32 g were anesthetized by intraperitoneal injection of 0.1 mL
of a mixture containing 2.25 mg/mL xylazine hydrochloride and
30 mg/mL ketamine hydrochloride. The mice were injected via a
tail vein with 400 kBq of the 99mTc-hypericin derivative in 0.1 mL
water/PEG 400 80/20 V/V and sacrificed by decapitation under
anesthesia 30 min, 4 or 24 h p.i. The organs were dissected and
weighed in tared tubes and radioactivity in all organs and blood
was counted in a gamma counter (Wallac). Corrections were
Acknowledgement
made for background radiation and physical decay during This work was supported by a grant awarded by Geconcerteerde
counting. Activity in the organs was expressed as %ID/organ Onderzoeksactie (GOA) of the Flemish Government, FWO-
and %ID/g of organ. Activity in blood was calculated on the Vlaanderen (grant G. 0257.05) and in part by the EC-FP6-project
assumption that blood constitutes 7% of the total body weight. DiMI, LSHB-CT-2005-512146.
J. Label Compd. Radiopharm 2008, 51 33–40
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

H. Fonge et al.
References
[11] W. Brandau, B. Bubeck, M. Eisenhut, D. M. Taylor. Int. J. Rad. Appl.
Instrum. [A] 1988, 39, 121–129.
[12] G. Bormans, D. Huyghe, A. Christiaen, K. Verbeke, T. de Groot, H.
Vanbilloen, P. de Witte, A. Verbruggen, J. Label. Compd. Radio-
pharm. 2004, 47, 191–198.
[13] H. Fonge, M. Van de Putte, D. Huyghe, G. Bormans, Y. Ni, P. de
Witte, A. Verbruggen, Contrast Media Mol. Imaging 2007, 2,
113–119.
[14] H. A. Fonge, G. Bormans, Y. Ni, F. Chen, K. Verbeke, G. Marchal, L.
Mortelmans, A. Verbruggen, Eur. J. Nucl. Med. Mol. Imaging 2004,
31, S381.
[15] H. Fonge, K. Vunckx, Y. Ni, J. Nuyts, L. Mortelmans, H. Wang, T. de
Groot, P. Vermaelen, G. Bormans, A. Verbruggen, Abstract of
papers. Fifty-fourth Annual Meeting of the Society of Nuclear
Medicine, Washington, DC. The Society of Nuclear Medicine,
Reston, VA, 2007, Abstract 610.
[16] H. Yamauchi, J. Takahashi, S. Seri, H. Kawashima, H. Koike, M.
Kato-Azuma. In vitro and in vivo characterisation of a new
series of 99mTc-complexes with N2S2 ligands, in Technetium
and Rhenium in Chemistry and Nuclear Medicine (Eds.:
M. Nicolini, G. Bandoli, U. Mazzi), Cortina International, Padova,
Italy, 1989, pp. 475–502.
[1] I. Bohm, H. Schild, Mol. Imaging Biol. 2003, 5, 2–14.
[2] S. Nagata, P. Golstein, Science 1995, 267, 1449–1456.
[3] B. A. Khaw, Semin. Nucl. Med. 1999, 29, 259–270.
[4] A. Flotats, I. Carrio, Eur. J. Nucl. Med. Mol. Imaging 2003, 30,
615–630.
[5] B. A. Khaw, H. W. Strauss, R. Moore, J. T. Fallon, T. Yasuda, H. K.
Gold, E. Haber, J. Nucl. Med. 1987, 28, 76–82.
[6] B. A. Khaw, T. Yasuda, H. K. Gold, R. C. Leinbach, J. A. Johns, M.
Kanke, M. Barlai-Kovach, H. W. Strauss, E. Haber, J. Nucl. Med. 1987,
28, 1671–1678.
[7] G. Mariani, G. Villa, P. F. Rossettin, P. Spallarossa, G. P. Bezante, C.
Brunelli, K. Y. Pak, B. A. Khaw, H. W. Strauss, J. Nucl. Med. 1999, 40,
1832–1839.
[8] Y. Ni, G. Bormans, F. Chen, A. Verbruggen, G. Marchal, Invest.
Radiol. 2005, 40, 526–535.
[9] Y. Ni, D. Huyghe, K. Verbeke, P. A. de Witte, J. Nuyts, L. Mortelmans,
F. Chen, G. Marchal, A. M. Verbruggen, G. M. Bormans. Eur. J. Nucl.
Med. Mol. Imaging 2006, 33, 595–601.
[10] H. Fonge, L. Jin, H. Wang, Y. Ni, G. Bormans, A. Verbruggen, Bioorg.
Med. Chem. Lett. 2007, 17, 4001–4005.
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 33–40