Fluorous ligand capture (FLC): A chemoselective solution-phase strategy for isolating 99mTc-labelled compounds in high effective specific activity

Article abstract of DOI:10.1039/c1cc11079a

A new approach for preparing ^99mTc-labelled compounds in high effective specific activity was developed by utilizing a novel fluorous ligand capture (FLC) agent and a chemoselective filtration strategy. This paradigm eliminates the need to use HPLC to obtain technetium(i) based molecular imaging probes free from residual precursor.

Full text of DOI:10.1039/c1cc11079a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 7518–7520
www.rsc.org/chemcomm
COMMUNICATION
Fluorous ligand capture (FLC): a chemoselective solution-phase strategy
for isolating 99mTc-labelled compounds in high effective specific activityw
Justin W. Hicks, Laura E. Harrington and John F. Valliant*
Received 23rd February 2011, Accepted 4th May 2011
DOI: 10.1039/c1cc11079a
A new approach for preparing 99mTc-labelled compounds in high
effective specific activity was developed by utilizing a novel
fluorous ligand capture (FLC) agent and a chemoselective
filtration strategy. This paradigm eliminates the need to use
HPLC to obtain technetium(I) based molecular imaging probes
free from residual precursor.
and dispensing of PET and SPECT radiopharmaceuticals has
been an issue of concern and attempts are made to minimize
sample handling or maximize shielding.6,7
The move toward solid-phase extraction and automated
synthesis and purification systems will serve to reduce operator
exposure by minimizing sample handling and accelerating
preparation and purification times as well as facilitating the
process of meeting GMP guidelines; the current focus on
processes that can be easily automated is a general trend in
the field.8 An alternative strategy that has been developed to
address the issue of removal of excess unlabeled precursor
without HPLC are solid-phase labelling methods9,10 which are
effective in removing the excess ligand but the labelling yields
are not ideal.9,10 An additional disadvantage is that not all
radiopharmaceuticals can be linked to solid-supports such that
the desired product is released upon labelling.
Molecular radioimaging using positron emission tomography
(PET) and single photon emission computed tomography
(SPECT)1 make it possible to visualize biochemical changes
in vivo thereby providing the opportunity for earlier detection
and molecular characterization of diseases and injuries. These
methods depend upon targeted molecular imaging probes;
radiolabeled compounds which can localize medical isotopes
to sites in direct proportion to the concentration of a specific
target. When developing molecular radioimaging probes for
specific receptors, a Bmax/KD of 10 or greater is desirable to
achieve high target to non-target ratios.2,3 For systems of
modest expression levels it is essential that the probe have
high affinity for the target and that the concentration of
competing ligands be minimized to prevent non-specific bind-
ing which has a detrimental impact on target to non-target
ratios and the general utility of a probe.4
An alternative approach is to employ ligand capture following
labelling; both solid and liquid-based systems are possible.
Reported solid-phase capture reactions, such as those used
for mass spectrometric identification of phosphoproteins11
and those for capture and removal of bacterial pathogens12
are generally not suitable for working with radioactive
compounds. Moreover, solid-phase labelling methods, which
are favoured in high throughput screening of combinatorial
libraries,13 have often been shown to result in high non-specific
binding of the radionuclide to solid supports leading to
decreased radiochemical yields.10 An alternative approach
would be to employ a solution-phase ligand capture system
based on fluorous chemistry.14 Ley and coworkers have used a
comparable ‘‘phase-switch’’ method to isolate reagents and
scavengers used in synthetic transformations.15
The traditional process of preparing SPECT radiopharma-
ceuticals using 99mTc, the most widely used radionuclide in
diagnostic medicine, involves the use of instant kits which
contain a reductant, buffers and a large excess of the ligand to
be labelled. Upon the addition of technetium, the product
mixture contains the desired product and unlabelled vector
which is problematic in the case of molecular imaging probes
because the excess unlabelled precursor often maintains simi-
lar binding affinity as the radiotracer for the target.3,5 For
research, preparative HPLC is a common method for removing
the excess unlabelled ligands; however, this approach is not
optimal for routine clinical use, primarily because of prolonged
operator exposure to radiation and the general length of time
required for purification. Staff exposure during the preparation
Fluorous radiolabelling methods offer a convenient means
to purify radiolabelled compounds. The general approach to
date involved systems whereby the fluorous group is cleaved
upon reaction with the radionuclide. The labelled compound
can then be isolated by passing the reaction mixture through a
fluorous solid-phase extraction cartridge where the precursor
is selectively retained.16,17 Our approach is to incorporate the
fluorous synthon into the scavenging agent which will capture
the unreacted ligand after the radiolabelling reaction. This
route avoids the step of preparing the fluorous precursor of the
desired radiolabelled product.
Departments of Chemistry and Chemical Biology & Medical Physics
and Applied Radiation Sciences, McMaster University,
Hamilton, ON, Canada. E-mail: valliant@mcmaster.ca;
Fax: +905-522-7776; Tel: +905-525-9140 ext. 20182
w Electronic supplementary information (ESI) available: Details of
experimental procedures and characterization of novel compounds.
See DOI: 10.1039/c1cc11079a
A fluorous copper chelate complex that can selectively bind
unlabelled ligand, rendering it fluorous, without degrading the
c
7518 Chem. Commun., 2011, 47, 7518–7520
This journal is The Royal Society of Chemistry 2011

View Article Online
Scheme 1 (i) BrCH2CO2Bn, DIPEA (ii) H2, Pd/C, CH3OH (iii) CuBr2,
EtOH.
labelled compound was developed as the FLC reagent. The
soluble ligand capture agent was prepared by combining
3-(perfluorooctyl)propyl amine (1) with 2-bromobenzylacetate
to yield 2 in 98% yield (Scheme 1). Use of unprotected
bromoacetic acid led to a large mixture of side products
which could not be easily separated from the desired chelate.
Following hydrogenation, the copper complex was prepared
using a saturated CuBr2 in EtOH solution which produced the
product as a blue precipitate in 91% yield.
Fig. 1 HPLC chromatograms associated with the FLC of 5/6a. Top:
UV trace of the crude reaction mixture showing 5 prior to FLC.
Second from top (note scale change): UV trace following FLC. Third:
g trace following FLC showing only 6a. Note that the UV and
radioactivity detectors are connected in series.
To test the extraction efficiency, 4 was combined with a
simple Tc(I) chelate system, bispyridyl valeric acid (5) prior to
working with a peptide-targeted agent (Scheme 2). 5 was
chosen because it contains a highly effective chelator which
forms inert complexes with the [99mTc(CO)3]+ core.18 In
addition, the ligand has a strong UV absorption making it
possible to determine the extraction efficiency by HPLC.
Compound 5 (3.3 mmol) was combined with [99mTc(CO)3(H2O)3]+
in a microwave reactor for 2 min at 150 1C (Scheme 2). The
g-HPLC chromatogram displayed the formation of the desired
BPV-Tc complex (tR = 12.9 min.), with a retention time
matching the Re standard (tR = 12.6 min.)19 while unlabelled
BPV ligand could be seen in the corresponding UV trace
(tR = 10.7 min.). Compound 4 (10 mg) was dissolved in
0.5 mL DMF and then added to reaction mixture. After
stirring for 5 min, the mixture was loaded onto a F-SPE
cartridge and a fluorophobic wash of 20% water in MeOH
was used to elute the BPV-Tc complex. Analytical HPLC
(Fig. 1) showed that 6a was obtained as the only radioactive
product with greater than 99% of the ligand removed and an
average of 8.5 Æ 1.6% (n = 4) of the activity remained on the
cartridge (Table 1). The compound could then potentially be
dried by rapid evaporation on the Biotage V10 and reconsti-
tuted as an injectible formulation. Ethanol can be substituted
for methanol to give a formulation suitable for injection
(following dilution with an isotonic aqueous solution).
Table 1 Comparison of ligand removal methods
Method
Ligand removed Non-specific binding
Solid-phase (20–75 mm)
Solid-phase (75–150 mm)
86 Æ 2%
90 Æ 3%
27 Æ 3%
22 Æ 1%
21 Æ 1%
8 Æ 2%
Solid-phase (150–300 mm) 91 Æ 5%
FLC (5/6a)
FLC (7/8a)
>99%
95 Æ 1%
99 Æ 1%
95 Æ 1%
12 Æ 1%
2 Æ 1%
Preloaded FLC (5/6a)
Preloaded FLC (7/8a)
33 Æ 3%
an acidic saline solution (pH = 4). The mixture containing 5
and 6a was dripped onto the resin slowly, where approxi-
mately 1/3 of the activity was eluted during the loading process
and another 40–50% was eluted with a 3 mL saline wash.
Further washes with saline and then MeOH did not elute the
significant amounts (>20%) of activity that remained on the
cartridge.
The solid-phase extraction procedure, which also required a
C-18 SPE to eliminate residual copper, was able to remove
between 86 and 91% of the excess ligand which is reasonable
but less effective than the fluorous system. In addition, the
high non-specific binding to the solid-support greatly favours
the FLC method where the inherently poor interaction
between fluorous compounds and polar organic/inorganic
molecules was particularly advantageous.
An alternative strategy, a hybrid of the FLC and solid phase
methods, was investigated whereby fluorous silica preloaded
with 4 was used to purify the model reaction mixture. As the
radiolabelling mixture passed through the silica, the excess
ligand was captured by 4 and remained behind while the
product was eluted. This approach is more amenable to an
instant kit type strategy. As a test, compound 4 was preloaded
onto a 2g F-SPE cartridge using a 1 : 1 mixture of MeCN and
DMF containing 5 drops of 1N HCl and the reaction mixture
containing 5 and 6a added subsequently. Following a gradient
elution, 6a was selectively eluted with 99% of the ligand
removed and 90% of the activity recovered. This represents
an improvement over the solid phase extraction procedure
however it requires more manipulation.
For comparison, the solid-phase analogue of 4 was prepared
following the method of Ley et al.15,20 and evaluated as a
ligand capture reagent. Three Amberlite IRC-784 resins of
different particle sizes were used (20–75, 75–150, and 150–300 mm).
These resins, which are functionalized with iminodiacetic acid
moieties, were loaded with copper by shaking a suspension of
the resins in saturated solutions of CuSO4 for 24 h. After
thorough washing, the resins were loaded into standard solid-
phase extraction tubes (100 mg) and activated prior to use with
With the successful extraction of the free dipyridyl amine
chelate using the FLC method, the focus shifted to peptide
conjugates. Peptides are attractive vectors for targeting radio-
metals to specific proteins. For this work, a dipyridyl amine
Scheme 2 (i) [M(CO)3(OH2)3]+, microwave heating (ii) 4 then FSPE.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 7518–7520 7519

View Article Online
conjugate of a LTVSPWY peptide that is known to target
erbB2 receptors was prepared.21 ErbB2 is a human epidermal
growth factor whose over expression is present in approxi-
mately 30% of breast cancer cases. In such cases, the
prognosis is poor as this receptor is normally a sign of a
metastasizing tumor.22 To incorporate the dipyridyl ligand in
this peptide, an analogue derived from Fmoc-protected lysine
as opposed to the valeric acid derivative was used (see ESI for
structurew). The peptide (7) and the Re analogue (8b) were
prepared using a CEM Liberty synthesizer according to a
literature procedure.22
Notes and references
1 G. D. Luker and D. Piwnica-Worms, Acad. Radiol., 1999, 8, 4.
2 W. C. Eckelman, Nucl. Med. Biol., 1998, 25, 169.
3 W. C. Eckelman, M. Bonardi and W. A. Volkert, Nucl. Med. Biol.,
2008, 35, 523.
4 R. J. Mairs, M. N. Gaze, D. G. Watson, G. G. Skellern,
P. Constable, K. McKeller, J. Owens, G. Valdyarathan and
M. R. Zalutsky, Nucl. Med. Commun., 1994, 15, 268.
5 W. C. Eckelman, Nucl. Med. Biol., 2002, 29, 777.
6 (a) C. Tsopelas, P. J. Collins and C. Blefari, J. Nucl. Med. Tech.,
2003, 31, 37; (b) T. H. Wu and L. H. Shen, Nucl. Sci. J., 1997, 34,
140; (c) J. Trujillo, S. Krinsky, B. Wilson, G. Young, E. Teague
and K. Kamer, J. Nucl. Med. Tech., 1982, 10, 197.
Compound 7 was combined with [99mTc(CO)3(OH2)3]+ in a
microwave reactor for 10 min at 60 1C. The g-HPLC
chromatogram showed a peak representing the 99mTc-peptide
complex (8a) (tR = 15.6 min.) which matched the retention
time of the Re standard (tR = 15.2 min.) while the unlabelled
peptide could be seen in the UV chromatogram (tR = 13.0 min.).
To remove the unlabelled ligand, 4 was added to the reaction
mixture (20 mg). After 10 min of sonication the pale blue
solution was loaded onto a FSPE cartridge and 8a was eluted
using a fluorophobic wash of 20% water in CH3CN.
A Biotage V10 rapid evaporation system was used to remove
the solvent and the residue was redissolved into 1 mL of 25%
CH3CN in water for HPLC analysis. The UV chromatogram
showed that an average of 95 Æ 1% of the ligand had been
removed demonstrating that the FLC method can be
applied to larger ligands as well as small molecule chelates.
Unfortunately, but not unexpectedly, the loss of product from
non-specific binding increased (average of 12 Æ 1%, n = 3).
The value is still superior to solid-phase capture. Purification
of the labelled peptides using the FSPE cartridge precoated
with 4 was also attempted. The approach resulted in 95% of
the ligand being removed but higher loss of product (33%).
The reason for this is likely interaction with uncapped silica
groups on the SPE cartridge. Evaluating different solid-phase
materials is a focus of current research.
7 D. D. Dischino, C. J. Bernard and J. J. Mongillo, U.S. Pat. Appl.
Publ. US 20070031492 A1 20070208, 2007.
8 (a) C. Lemaire, A. Plenevaux, J. Aerts, G. Del Fiore, C. Brihaye,
D. Le Bars, D. Comar and A. Luxen, J. Labelled Compd. Radio-
pharm., 1999, 42, 63; (b) D. Ellison, J. Kaufman and S. J. Mather,
Nucl. Med. Commun., 2010, 31, 173; (c) G. Quincoces, L. Lopez-
Sanchez, M. Sanchez-Martınez, M. Rodrıguez-Fraile and
I. Penuelas, Appl. Radiat. Isot., 2010, 68, 2298.
9 R. W. Riddoch, P. Schaffer and J. F. Valliant, Bioconjugate Chem.,
2006, 17, 226.
10 (a) G. N. Ranadive and W. D. Bloomer, Nucl. Med. Biol., 1995, 22,
607; (b) D. H. Hunter and X. Zhu, J. Labelled Compd. Radio-
pharm., 1999, 42, 653; (c) G. Vaidyanathan, D. J. Affleck,
K. L. Alston, X.-G. Zhao, M. Hens, D. H. Hunter, J. Babich
and M. R. Zalutsky, Bioorg. Med. Chem., 2007, 15, 3430;
(d) A. Pollak, D. G. Roe, C. M. Pollock, L. F. L. Lu and
J. R. Thornback, J. Am. Chem. Soc., 1999, 121, 11593;
(e) S. Mundwiler, L. Candreia, P. Haefliger, K. Ortner and
R.
Alberto,
Bioconjugate
Chem.,
2004,
15,
195;
(f) R. J. Flanagan, U.S. Pat 4,874,601, 1989; (g) K. Kawai,
H. Ohta, M. A. Channing, A. Kubodera and W. C. Eckelman,
Appl. Radiat. Isot., 1996, 47, 37.
11 L. Zhao, R. Wu, G. Han, H. Zhou, L. Ren, R. Tian and H. Zou,
J. Am. Soc. Mass Spectrom., 2008, 19, 1176.
12 Y.-F. Huang, Y. Fanwang and X. Pingyan, Environ. Sci. Technol.,
2010, 44, 7908.
13 M. K. J. Gagnon, S. H. Hausner, J. Marik, C. K. Abbey,
J. F. Marshall and J. L. Sutcliffe, Proc. Natl. Acad. Sci. U. S. A.,
2009, 106, 17904.
14 (a) D. P. Curran, S. Hadida and M. He, J. Org. Chem., 1997, 62,
6714; (b) D. P. Curran and Z. Y. Luo, J. Am. Chem. Soc., 1999,
121, 9069; (c) D. P. Curran, S. Hadida, S. Y. Kim and Z. Y. Luo,
J. Am. Chem. Soc., 1999, 121, 6607.
The results described demonstrate that fluorous ligand
capture is a feasible alternative to HPLC for the purification
of new Tc(I)-based molecular imaging probes. This system can
be readily adapted for use with an automated synthesis plat-
form which is routinely used in manufacturing and purifying
radiopharmaceuticals. The general utility of this method for
other Tc chelates needs to be assessed including a more
detailed measurement of the optimal and relative stability
constants. Nonetheless, a change in how and where Tc
radiopharmaceuticals are prepared (i.e. in large central
radiopharmacies equipped with automated synthesis units as
opposed to small hospital pharmacies) in combination with
the development of new labelling and purification strategies
should allow for more effective agents to be developed and
formulated.23
15 J. Siu, I. R. Baxendale, R. A. Lewthwaite and S. V. Ley, Org.
Biomol. Chem., 2005, 3, 3140.
16 A. C. Donovan and J. F. Valliant, Nucl. Med. Biol., 2008, 35, 741.
17 R. Bejot, T. Fowler, L. Carroll, S. Boldon, J. E. Moore, J. Declerck
and V. Gouverneur, Angew. Chem., Int. Ed., 2009, 48, 586.
18 M. Bartholomae, J. Valliant, K. Maresca, J. Babich and J. Zubieta,
Chem. Commun., 2009, 493.
19 C. Sundararajan, T. Besanger, R. Labiris, K. J. Guenther,
T. Strack, R. Garafalo, T. T. Kawabata, D. Finco-Kent,
J. Zubieta, J. W. Babich and J. F. Valliant, J. Med. Chem., 2010,
53, 2612.
20 (a) S. V. Ley, I. R. Baxendale, D. A. Longbottom and
R. M. Myers, Drug Discovery and Development, 2007, 2, 51;
(b) S. V. Ley, A. Massi, F. Rodrıguez, D. C. Horwell,
´
R. A. Lewthwaite, M. C. Pritchard and A. M. Reid, Angew.
Chem., Int. Ed., 2001, 40, 1053.
21 M. Shadidi and M. Sioud, FASEB J., 2003, 17, 256.
22 X.-F. Wang, M. Birringer, L.-F. Dong, P. Veprek, P. Low,
E. Swettenham, M. Stantic, L.-H. Yuan, R. Zobalova, K. Wu,
M. Ledvina, S. J. Ralph and J. Neuzil, Cancer Res., 2007, 67, 3337.
23 W. C. Eckelman, P. A. Erba, M. Schwaiger, H. N. Wagner, Jr.,
R. Alberto and U. Mazzi, Nucl. Med. Biol., 2007, 34, 1.
We thank the Canadian Institute of Health Research
(CIHR) and the Ontario Institute for Cancer Research
(OICR) for funding.
c
7520 Chem. Commun., 2011, 47, 7518–7520
This journal is The Royal Society of Chemistry 2011