The synthesis of [26,27-11C]dihydroxyvitamin D3, a tracer for positron emission tomography (PET)

Article abstract of DOI:10.1016/S0968-0896(01)00178-X

1α,25-Dihydroxyvitamin D₃, an endogenous ligand with the highest affinity for the vitamin D receptor (VDR), was labeled with ¹¹C for use in biological experiments. The radionuclide was incorporated via the reaction of [¹¹C]methyllithium on a methyl ketone precursor in tetrahydrofuran at -10°C. Deprotection of the labeled intermediate yielded 2.5-3 GBq [26,27-¹¹C]1α,25-dihydroxyvitamin D₃ [¹¹C-1,25(OH)₂ D₃] with specific radioactivity averaging 100 GBq/μmol at the end of synthesis and HPLC purification. The entire process took 48 min from the end of radionuclide production. In vitro binding experiments in rachitic chick purified VDR demonstrated the high affinity binding of this novel tracer. Thus; ¹¹C-1,25(OH)₂ D₃ is available for in vivo distribution studies and may be suitable for the positron emission tomography (PET) determination of VDR levels and occupancy in animals and humans.

Full text of DOI:10.1016/S0968-0896(01)00178-X

Bioorganic & MedicinalChemistry 9 (2001) 3123–3128
The Synthesis of [26,27-11C]Dihydroxyvitamin D3,
a Tracer for Positron Emission Tomography (PET)
Thomas A. Bonasera,a Gunnar Grue-Sørensen,b
Giuseppina Ortu,a Ernst Binderup,b Mats Bergstrom,a
Fredrik Bjorklingb and Bengt Langstroma,*
aUppsala University PET Centre, UAS, SE-751 85 Uppsala, Sweden
bDepartment of Medicinal Chemistry, Leo Pharmaceutical Products,
Industriparken 55, DK-2750 Ballerup, Denmark
Received 13 October 1999; accepted 8 May 2001
Abstract—1a,25-Dihydroxyvitamin D3, an endogenous ligand with the highest affinity for the vitamin D receptor (VDR), was
labeled with 11C for use in biological experiments. The radionuclide was incorporated via the reaction of [11C]methyllithium on a
methylketone precursor in tetrahydrofuran at
À10 ꢀC. Deprotection of the labeled intermediate yielded 2.5–3 GBq
[26,27-11C]1a,25-dihydroxyvitamin D3 [11C-1,25(OH)2 D3] with specific radioactivity averaging 100 GBq/mmolat the end of
synthesis and HPLC purification. The entire process took 48 min from the end of radionuclide production. In vitro binding
experiments in rachitic chick purified VDR demonstrated the high affinity binding of this noveltracer. Thus; 11C-1,25(OH)2 D3 is
available for in vivo distribution studies and may be suitable for the positron emission tomography (PET) determination of VDR
levels and occupancy in animals and humans. # 2001 Elsevier Science Ltd. All rights reserved.
Introduction
tion of a pharmaceuticaldose of a VDR-binding drug.
Through the use of positron emission tomography
(PET) and a tracer based on a VDR ligand, the goal of
in vivo VDR measurements was our aim.
The hormone 1a,25-dihydroxyvitamin D3 [1,25(OH)2
D3], a metabolite of vitamin D3, maintains calcium
homeostasis, inhibits proliferation, stimulates differ-
entiation, and induces apoptosis in a wide range of
normal and malignant cells by binding to specific intra-
cellular receptors,1,2 namely the vitamin D receptor
(VDR), and is the most potent endogenous ligand for
this receptor. Other, nongenomic, activities have been
attributed to the binding of 1,25(OH)2 D3 to a mem-
brane-bound receptor.3 The actions of 1,25(OH)2 D3
and its analogues are thus of interest as pharmaceuticals
in fields such as oncology,4,5 endocrinology, immunol-
ogy, and bone disease.6À8
In this report, the 11C (t1/2=20 min) labeling of
1,25(OH)2 D3 and its potentialfor VDR imaging were
investigated. The choice for the incorporation of the
positron-emitting radionuclide was at the 26,27-car-
bons. 26,27-Positional labeling of 1,25(OH)2 D3 has
been reported incorporating tritium by reacting tritiated
methylmagnesium bromide with 1a-hydroxy-26,27-
dinorvitamin D3-25-carboxylate.9 In the current work,
[11C]methyllithium10 was selected as the carbanion
source, employing as precursor 1a and 3b hydroxyl tert-
butyldimethylsilyl-protected
1a-hydroxy-25-keto-27-
Of potentialinterest to researchers and cilnicians woudl
be the ability to locate and quantify the VDR in tumors
or other tissues in vivo in laboratory animals and
humans. Of potentialinterest to those in VDR-realted
drug development would be the ability to measure the
fraction of unoccupied VDR in vivo after administra-
norvitamin D3. Following rapid 11C incorporation,
protecting group removalwoudl furnish the desired
compound in a time frame compatible with the short
half-life of this radionuclide.
The synthesis of [26,27-11C]1a,25-dihydroxyvitamin D3
(11C-1,25(OH)2 D3) and the precursor required to make
it, along with validation of radiotracer binding in vitro,
are described herein.
*Corresponding author. Tel.: +46-18-471-5381; fax: +46-18-471-
5390; e-mail: bengt.langstrom@pet.uu.se
0968-0896/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00178-X

3124
T. A. Bonasera et al. / Bioorg. Med. Chem. 9 (2001) 3123–3128
Results and Discussion
1,25(OH)2 D3 was radiolabeled with 11C. The synthetic
approach was one employing formation of a single car-
bon–carbon bond on an otherwise intact molecule,
allowing completion of the synthesis and HPLC puri-
fication within the constraints of the 20 min half-life. In
addition to standard chromatographic analysis for
chemicaland radiochemicalpurities, in vitro binding
studies were performed using a commercially available
solubilized rachitic chick purified VDR preparation to
verify that the quality of the radiotracer was sufficient
for further in vivo work.
Chemistry and radiochemistry
Starting from the iodo compound 1, the precursor for
11C labeling 3 was synthesized in two steps11 (Scheme
1). Synthesis of 4, an HPLC standard for a 11C-labeled
intermediate, was achieved in one step11 from
1,25(OH)2 D3 (Scheme 2).
11C-1,25(OH)2 D3 was synthesized rapidly (48 min
average from the end of cyclotron irradiation) using a
reliable method (more than 14 syntheses without fail-
ure) as outlined in Scheme 3. The first step, reaction of
[11C]methyllithium10 with methylketone 3 was per-
formed using a published method in which the reaction
vialis charged with the precursor prior to trapping
12
[11C]methyliodide.
The temperatures used in this
work during trapping of methyliodide ( À70 ꢀC) and
reaction of [11C]methyllithium (À10 ꢀC) were sig-
nificantly lower than those reported (À10 ꢀC and room
temperature, respectively). We chose the lower tem-
perature because in our experience13 higher yields of
both [11C]methyliodide trapping and [ 11C]methylithium
reaction are achieved with fewer side reactions. The 11C
incorporation yield was estimated to be as high as 50%,
although this value was difficult to measure due to loss
of radioactivity as [11C]methane and the very lipophilic
nature of the labelled intermediate, [11C]4. The second
step, cleavage of two tert-butyldimethylsilyl (TBDMS)
groups with tetrabutylammonium fluoride, was esti-
mated to be quantitative.
Scheme 1. (a) Prepare methylvinylketone, NiCl 2Â6H2O, Zn dust,
pyridine, 65 ꢀC, 30 min; (b) then 1, pyridine, THF, rt, 2 h, dark; (c) 9-
acetylanthracene, TEA, toluene, hn, 10 ꢀC, 15 min. Overall yield 38%.
The quantity of 11C-1,25(OH)2 D3 produced, 2.5–3 GBq
at end of synthesis, was more than adequate for both in
vitro and in vivo studies (typically 0.4–0.8 GBq 11C is
used in a human PET examination). The specific radio-
activity of 11C-1,25(OH)2 D3 at the end of synthesis
averaged 100Æ39 GBq/mmol( n=14), which is in the
range of other tracers made from [11C]methyliodide
synthesized at this laboratory.
Radiochemicalpurity was in alcases greater than 99%.
Chemicalpurity, based on HPLC UV detection at
l=265 nm, averaged 79%, and was generally higher in
the later syntheses (91% average for the last four
experiments) due to improved preparative HPLC
separation and taking of a center cut fraction of the 11C-
1,25(OH)2 D3 eluent. The major mass impurity, thought
to be fully tert-butyldimethylsilyl-deprotected starting
materialbased on LC–MS (data not presented) and its
Scheme 2. (a) TBDMSCl, imidazole, DMF, rt, 1 h. Yield 80%.

T. A. Bonasera et al. / Bioorg. Med. Chem. 9 (2001) 3123–3128
3125
lmax of 265 nm, eluted prior to 11C-1,25(OH)2 D3 on
both preparative and analytical HPLC.14 The suspected
impurity, 1(S),3(R)-bis(hydroxy)-25-keto-9,10-seco-27-
addition adduct was not observed. We propose that the
low reaction temperatures employed (À70 to À10 ꢀC)
and short reaction time (10 min) allowed only the kin-
etically favored methyl anion to react with the ketone
prior to the TBAF quench.
norcholesta-5(Z),7(E),10(19)-triene
keto-27-norvitamin D3), has been shown to exhibit
vitamin like activity, although, compared to
(1a-hydroxy-25-
D
1,25(OH)2 D3, doses of this compound at least 40 and
2000 times higher were required, respectively, for similar
effect on intestinal calcium transport and bone calcium
mobilization.15 Thus while the impurity can potentially
bind to VDR, its influence on the effective specific
radioactivity of the tracer would appear negligible due
to its low relative potency in addition to its presence
averaging about 9% (in the last four experiments) of the
The position of the 11C label was confirmed in a 11/13C-
1,25(OH)2 D3 co-labeling experiment, followed by 13C
NMR analysis of fully radio-decayed product. The
resonances found, 29.26 and 29.11 ppm, correspond
within experimentalerror to the resonances assigned to
the 26 and 27 carbons, respectively, 29.31 and
29.18 ppm.16
totalVDR-active mass [the impurity puls 1,25(OH)
2
In vitro VDR binding
D3]. Minimizing the VDR-binding mass is important
due to its potentialto bind the high-affinity, ilmited
capacity receptor system, potentially blocking 11C-
1,25(OH)2 uptake. In addition, prior to use of the tracer
in humans, chemicalimpurities must be minimzed.
The VDR kit allowed determination of an equilibrium
dissociation constant (Kd) of 61 pM and the maximum
number of binding sites (Bmax) of 22.3 fmolreceptor per
experiment as shown in Figure 1. These values were
calculated from a non-linear regression curve fit of the
specific binding data (Fig. 1, top) and the numbers were
confirmed in a conventionalScatchard polt (Fig. 1,
bottom). The Kd value measured in these studies is
within the range of others reported for 1,25(OH)2 D3
binding to the VDR.9 The calculated relative amount of
Curiously, considering the facts that butyllithium is in
great excess over the mass associated with [11C]methyl-
lithium and the starting ketone is present at the end of
[11C]methyllithium reaction, formation of the butyl
Figure 1. Top: Plot of specific 11C-1,25(OH)2 D3 binding in rachitic
chick purified VDR versus 1,25(OH)2 D3 concentration. The curve
shown is a fit as described in the experimentalsection from which Kd
and Bmax were derived. Error bars represent the standard error of the
mean (n=3). Bottom: Scatchard plot of the ratio of specifically bound
radiotracer to free radiotracer versus specifically bound radiotracer.
The line shown is a linear regression fit. The values À1/slope and
X-intercept are shown for comparison, respectively, to Kd and Bmax
obtained using non-linear regression.
Scheme 3. (a)
[
11C]methyl iodide, butyllithium, À70 to À10 ꢀC,
10 min; (b) TBAF, THF, 105 ꢀC, 3–5 min.

3126
T. A. Bonasera et al. / Bioorg. Med. Chem. 9 (2001) 3123–3128
specific binding17 to totalbinding at Kd and twice Kd,
18
were 29 and 21%, respectively.
seco-27-norcholesta-5(E),7(E),10(19)-triene 111 (576 mg,
0.84 mmol) in pyridine–THF 1:2 (v/v) (6 mL) was added
and the mixture was stirred in the dark for 2 h. The
mixture was poured into ethylacetate (75 mL) and fi-l
tered through Hyflo supercel. The Hyflo supercel was
washed with ethylacetate (2 Â75 mL). The combined
organic filtrates were washed with 1 M hydrochloric
acid (4Â75 mL), a solution of EDTA disodium salt
dihydrate (12 g) and NaHCO3 (12 g) in water (150 mL)
and brine (2Â75 mL). The crude product obtained from
drying (MgSO4) and concentration in vacuo was chro-
matographed on Si gel(20 g) with 10% ether in hexane
as eluent. Yield of 2: 251 mg (0.40 mmol, 53%). 1H NMR
as described.11 13C NMR d 209.0, 153.4, 143.1, 135.1,
121.5, 116.2, 106.4, 70.0, 67.0, 56.2, 56.1, 45.7, 44.0, 43.8,
40.3, 36.4, 35.8, 35.2, 29.6, 28.7, 27.4, 25.7, 25.6, 23.3, 22.0,
20.2, 18.5, 18.0, 17.9, 11.8, À5.0, À5.1, À5.1.
Conclusion
A
reliable method for the production of
[26,27-11C]1a,25-dihydroxyvitamin D3, a positron-emit-
ting radiopharmaceuticalfor the vitamin D receptor, in
adequate quantity, purity and specific radioactivity for
PET studies has been developed. In vitro experiments
demonstrated the high affinity binding of this novelPET
tracer. However, further evaluation of this tracer in
vitro in, for example, human tumor samples is war-
ranted. 11C-1,25(OH)2 D3 is thus available as a potential
toolfor in vivo assessment of the VDR in animasl and
humans using PET. Suggested applications are drug
development, and cancer- and bone disease-related
imaging.
1(S),3(R)-Bis[(tert-butyldimethylsilyl)oxy]-25-keto-9,10-
seco - 27 - norcholesta - 5(Z),7(E),10(19) - triene (3). See
Scheme 1. The method employed was adapted from a
published procedure.11
A mixture of 2 (251 mg,
Experimental
General
0.40 mmol), 9-acetylanthracene (18.5 mg, 0.08 mmol)
and triethylamine (290 mL) in toluene (32 mL) was irra-
diated under an argon atmosphere with a 500 W Hanau
Z-2 UV lamp at 10 ꢀC for 15 min. The reaction mixture
was concentrated in vacuo and the residue was chro-
matographed on Si gel (80 g) with toluene as eluent to
[11C]Carbon dioxide production was performed by the
14N(p,a)11C nuclear reaction using a Scanditronix MC-
17 cyclotron. A nitrogen (AGA, Nitrogen 6.0) gas target
containing 0.05% oxygen (AGA, Oxygen 4.8) was used.
Synthia, a robotic system developed in the laboratory,19
was used for [11C]methyliodide production, HPLC
injection and fraction collection. HPLC was performed
on binary pump systems (126 Solvent Module, Beck-
man) using either, for semi-preparative, UV (265 nm,
166 Detector Module, Beckman) and radioactivity
detectors, or, for analytical, diode array (190–300 nm,
168 Detector Module, Beckman) and radioactivity
detectors. The HPLC columns used were an Ultrasphere
ODS (5 mm, 10Â250 mm, Beckman) for semi-pre-
parative work and an Ultrasphere ODS (5 mm,
4.6Â250 mm plus 4.6Â45 mm guard column, Beckman)
for analytical. NMR spectra were recorded using either
a Varian XL 300 (Uppsala) or a Bruker AM300 (Bal-
lerup) spectrometer and all chemical shifts are reported
in ppm downfield from tetramethylsilane (d scale). Mass
spectra were recorded on a Micromass Autospec spec-
trometer. Tetrahydrofuran (THF) was distilled from
Na-benzophenone ketyl. Lithium aluminum hydride,
0.05 M, was diluted with THF from Aldrich 1 M in
THF.
1
give 3 (178 mg, 0.28 mmol, 71%). H NMR d 6.22 (d,
1H), 5.99 (d, 1H), 5.16 (m, 1H), 4.84 (m, 1H), 4.35 (m,
1H), 4.17 (m, 1H), 2.80 (m, 1H), 2.11 (s, 3H), 0.92 (d,
3H), 0.86 (s, 18H), 2.49–0.79 (m, 22H), 0.51 (s, 3H),
0.04 (m, 12H); 13C NMR d 209.1, 148.1, 140.8, 134.8,
122.9, 117.7, 111.0, 71.8, 67.3, 56.1, 56.0, 45.8, 45.5,
44.6, 44.0, 40.4, 35.8, 35.2, 29.6, 28.6, 27.5, 25.6, 25.6,
23.3, 21.9, 20.2, 18.5, 18.0, 17.9, 11.7, À4.9, À5.0,
À5.3.MS (EI+) calcd for C38H68O3Si2: 628.4707,
found: 628.4694. Elemental analysis of 3 was not possi-
ble due to residual solvent. Compound 3 was stored at
À25 ꢀC in ethylacetate (18 mg/mL) for more than 6
months with no apparent decomposition.
1,3-Bis[(tert-butyldimethylsilyl)oxy]-1ꢀ,25-dihydroxyvi-
tamin D3 (4). See Scheme 2. 1,25(OH)2 D3 (50 mg,
0.12 mmol), tert-butyldimethylsilyl chloride (54 mg,
0.36 mmol) and imidazole (49 mg, 0.72 mmol) were stir-
red in dimethylformamide (1.2 mL) at room temperature
for 1 h. Calcium chloride (3 M, 3 mL) was added and the
mixture was extracted with hexane (3Â3 mL). The
organic extracts were washed with water (3 mL) and
brine (3 mL), dried with MgSO4, concentrated in vacuo
and chromatographed on Si gel(15 g) with pentane–di-
chloromethane–ethyl acetate 50:50:4 as eluent to give 4
Chemistry and radiochemistry
1(S),3(R)-Bis[(tert-butyldimethylsilyl)oxy]-25-keto-9,10-
seco - 27 - norcholesta - 5(E),7(E),10(19) - triene (2). See
Scheme 1. The method employed was adapted from a
published procedure.11 A suspension of methylvinyl
ketone (130 mg, 1.85 mmol), NiCl2Â6H2O (180 mg,
0.76 mmol), and Zn dust (98 mg, 1.5 mmol) was stirred
in pyridine (10 mL) under argon and slowly heated to
65 ꢀC. After 30 min of heating at 65 ꢀC, the mixture was
cooled to room temperature. A solution of 1(S),3(R)-
bis[(tert-butyldimethylsilyl)oxy]-20(S)-(iodomethyl)-9,10-
(62 mg, 0.096 mmol, 80%). 1H NMR as described.11 13
C
NMR d 148.1, 140.9, 134.7, 123.0, 117.7, 111.0, 71.9,
70.9, 67.3, 56.3, 56.1, 45.8, 45.6, 44.6, 44.2, 40.4,
36.2,35.9,29.2, 29.0, 28.7, 27.5, 25.7, 25.6,23.3, 21.9, 20.7,
18.6, 18.0, 18.0, 11.8, À4.9, À5.0, À5.3. Compound 4
was stored at À25 ꢀC in ethylacetate (17 mg/mL) for
more than 6 months with no apparent decomposition.
[26,27-11C]1ꢀ,25-Dihydroxyvitamin D3 (11C-1,25(OH)2
D3). 11C-1,25(OH)2 D3 was synthesized according to

T. A. Bonasera et al. / Bioorg. Med. Chem. 9 (2001) 3123–3128
3127
Scheme 3. Stock 3 in ethylacetate (100 mL solution;
1.8 mg, 2.6 mmol) was added to an oven-dried 3 mL glass
vial. The solvent was removed under a stream of nitro-
gen gas and the vialwas pal ced under high vacuum for
5–10 min. The vialwas seaeld with a septum (PTFE-
lined)-equipped screw cap and thoroughly flushed with
nitrogen gas. THF was added, and 3 was dissolved via
vigorous vortexing. [11C]Methyliodide (estimated 0.03–
0.1 mmol, made from [11C]CO2, lithium aluminum
hydride, and HI as described13) was distilled into the
vial through a PTFE line and stainless steel needle over
3 min. The vialwas cooeld in a À70 ꢀC bath (dry ice,
ethanol) at the beginning of and throughout the
[11C]methyl iodide distillation.
Radiometric determination of radioligand binding in a
solubilized rachitic chick purified VDR system. Three
sequentialbinding experiments were performed using
components from, and following the instructions in, the
Amersham ‘1a,25-dihydroxyvitamin D [3H]assay reagents
system’ (Amersham Pharmacia Biotech, Uppsala, Swe-
den, product code TRK 870), substituting 11C-
1,25(OH)2 D3 for the tritiated tracer supplied. Briefly,
the purified VDR supplied by the manufacturer was
incubated (30 min, 21 ꢀC, with shaking) with different
concentrations (9.6–308.7 pM) of 11C-1,25(OH)2 D3 to
give total binding. Parallel experiments with the same
concentration of radiotracer plus 100 nM cold 1,25
(OH)2 D3 were performed to give non-specific binding.
Dextran-coated charcoalwas added at the end of incu-
bation to adsorb free radioligand. Centrifugation pel-
leted the dextran-coated charcoal, and radioactivity in
the supernatant [containing receptor-bound 11C-
1,25(OH)2 D3] was determined using conventional
gamma counting. The receptor supplied underwent a
single additional freeze-thaw cycle between the first and
second experiments. Kd (equilibrium dissociation con-
stant) and Bmax (maximum number of binding sites)
were calculated using the software GraphPad Prism
(GraphPad Software, Inc., San Diego, USA) via non-
linear-regression analysis, fit to a single site binding
model(hyperboal ) on the specific binding data.
At the end of distillation, the transfer and vent needles
were removed, leaving a sealed vial, cooled to À70 ꢀC,
containing unreacted 3 and [11C]methyliodide in THF.
To the vial was added butyllithium (Lancaster, 1.6 M
in hexanes, 30 mL, 48 mmol). After 1 min the vial was
moved to a À10 ꢀC bath (ice, ethanol) for 10 min
(contents turned yellow), providing 11C-labeled inter-
mediate [11C]4 in about 30–50% yield [calculated from a
few early experiments, decay-corrected, from trap-
ped [11C]methyliodide and based on anyatlical
HPLC, acetonitrile–dichloromethane 1:1, 1 mL/min,
tR (4 standard)=5.28 min, lmax=267 nm, product not
isolated].
The [11C]methylation reaction was quenched by adding
tetrabutylammonium fluoride (Lancaster, 1 M in THF,
thought to contain about 5 wt% water, 200 mL,
200 mmol), and the vial was moved from À10 to 105 ꢀC
for protecting group cleavage. After 3–5 min, the vial
was cooled to À10 ꢀC and the reaction was quenched by
adding 200 mL saturated aqueous sodium bicarbonate
plus 2 mL ethanol–water 1:1. Semi-preparative HPLC
[water–99.5% ethanol9:13, 4 mL/min, tR (1,25(OH)2 D3
standard)=13.2 min] afforded 2.5–3 GBq 11C-labeled
1,25(OH)2 D3.
Acknowledgements
Support was provided by the Swedish Cancer Fund
(TAB). NMR analyses were, in part, performed at the
Department of Organic Chemistry, Uppsala University.
NMR and MS analyses were performed at Leo,
Department of Spectroscopy. The technicalassistance
of Ms. Nadja Bøggild and Ms. Elise Blomsterberg is
greatly acknowledged.
The HPLC eluent fraction containing 11C-1,25(OH)2 D3
was used directly in the biological experiments. Analysis
of the fraction by HPLC [water–acetonitrile 1:9, 1 mL/
min, tR (1,25(OH)2 D3 standard)=5.8 min, lmax
=265 nm] was performed to determine radiochemical
and chemicalpurity, and the concentration of
1,25(OH)2 D3 mass in the fraction for calculation of
specific radioactivity.
References and Notes
1. Walters, M. R. Endocr. Rev. 1992, 13, 719.
2. Whitfield, G. K.; Hsieh, J. C.; Jurutka, P. W.; Selznick,
S. H.; Haussler, C. A.; MacDonald, P. N.; Haussler, M. R. J.
Nutr. 1995, 125, 1690S.
3. Norman, A. W.; Bishop, J. E.; Collins, E. D.; Seo, E. G.;
Satchell, D. P.; Dormanen, M. C.; Zanello, S. B.; Farach-
Carson, M. C.; Bouillon, R.; Okamura, W. H. J. Steroid Bio-
chem. Mol. Biol. 1996, 56, 13.
4. El Abdaimi, K.; Dion, N.; Papavasiliou, V.; Cardinal, P. E.;
Binderup, L.; Goltzman, D.; Ste-Marie, L. G.; Kremer, R.
Cancer Res. 2000, 60, 4412.
5. Hansen, C. M.; Hamberg, K. J.; Binderup, E.; Binderup, L.
Curr. Pharm. Des. 2000, 6, 803.
6. Bikle, D. D. Endocr. Rev. 1992, 13, 765.
7. Binderup, L.; Binderup, E.; Godtfredsen, W. O.; In Vitamin
D; Feldman, D., Glorieux, F. H., Pike, W., Eds.; Academic:
San Diego, 1997; pp 1027–1043.
8. ; Holick, M. F. AM. J. Clin. Nutr 1994, 60, 619.
9. Napoli, J. L.; Mellon, W. S.; Fivizzani, M. A.; Schnoes,
H. K.; DeLuca, H. F. Biochemistry 1980, 19, 2515.
10. Reiffers, S.; Vaalburg, W.; Wiegman, T.; Wynberg, H.;
Woldring, M. G. Appl. Radiat. Isot. 1980, 31, 535.
[26,27-11/13C]1ꢀ,25-Dihydroxyvitamin D3
(
11/13C-1,25
(OH)2 D3). The process for the 11/13C co-labeling
synthesis was a slight modification of that described
above for 11C-1,25(OH)2 D3. At the end of [11C]methyl
iodide distillation, (13C)methyliodide (Larodan, Swe-
den, 20% v/v in octane, 10 mL, 30 mmol) was added to
the cooled reaction vessel, proceeding as described
above. The HPLC fraction containing 11/13C-1,25(OH)2
D3 was allowed to fully radio-decay at À25 ꢀC, after
which the solvent was removed in vacuo. The resulting
residue was dissolved in CD3OD (0.5 mL) and a 13C
NMR spectrum was recorded (ref CD3OD=49.00 ppm).
13C NMR d 29.26, 29.11.

3128
T. A. Bonasera et al. / Bioorg. Med. Chem. 9 (2001) 3123–3128
11. Manchand, P. S.; Yiannikouros, G. P.; Belica, P. S.;
Madan, P. J. Org. Chem. 1995, 60, 6574.
Fivizzani, M. A. World/US Patents WO8002502/US4264513,
1981.
12. Dence, C. S.; Napolitano, E.; Katzenellenbogen, J. A.;
Welch, M. J. Nucl. Med. Biol. 1996, 23, 491.
13. Neu, H.; Bonasera, T. A.; Langstrom, B. J. Label. Compd.
Radiopharm. 1997, 41, 227.
14. Synthesis and spectralcharacterization of the suspected
impurity was performed. HPLC analysis in comparison with
1,25(OH)2 D3 indicated the same relative retention observed in
the radiolabeling experiments.
16. Leo PharmaceuticalProducts, NMR spectra archives.
17. From the non-linear regression curve fit, Figure 1.
18. The sum of calculated specific binding and non-specific
binding. The latter was obtained from a linear regression fit of
the non-specific binding data (not shown).
19. Bjurling, P.; Reineck, R.; Westerberg, G.; Schultz, J.; Gee,
A.; Sutcliffe, J.; Langstrom, B. Proceedings of the Sixth
Workshop on Targetry and Target Chemistry, Vancouver,
Canada, Aug 17–19, 1995; p 282–287.
15. DeLuca, H. F.; Schnoes, H. K.; Joseph L. Napoli, J.;