A highly effective bifunctional ligand for radioimmunotherapy applications

Article abstract of DOI:10.1039/c0cc05707j

A novel bifunctional ligand (3p-C-NETA) for antibody-targeted radioimmunotherapy (RIT) of β-emitting radioisotopes ⁹⁰Y and ¹⁷⁷Lu was efficiently synthesized via an unexpected regiospecific ring opening of an aziridinium ion. 3p-C-NET

Full text of DOI:10.1039/c0cc05707j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 5584–5586
www.rsc.org/chemcomm
COMMUNICATION
A highly effective bifunctional ligand for radioimmunotherapy applicationsw
Hyun-Soon Chong,* Hyun A. Song, Chi Soo Kang, Thien Le, Xiang Sun, Mamta Dadwal,
Hyunbeom Lee, Xiaoli Lan, Yunwei Chen and Anzhi Dai
Received 21st December 2010, Accepted 13th February 2011
DOI: 10.1039/c0cc05707j
A novel bifunctional ligand (3p-C-NETA) for antibody-targeted
radioimmunotherapy (RIT) of b-emitting radioisotopes 90Y and
177Lu was efficiently synthesized via an unexpected regiospecific
ring opening of an aziridinium ion. 3p-C-NETA instantly formed
a very stable complex with 90Y or 177Lu. 3p-C-NETA is an
excellent bifunctional ligand for RIT.
complexes of C-DTPA produced unfavorable in vitro and
in vivo stability profiles.9 The bifunctional ligands in the
CHX-DTPA (Fig. 1, trans-cyclohexyl-C-DTPA) series contain-
ing a cyclohexyl ring were developed as efforts to improve
stability of C-DTPA radiolabeled with a- or b-emitters.10a
Among the CHX-DTPA-based ligands, enantiomerically pure
CHX-A00-DTPA radiolabeled with 88Y (Fig. 1) was shown to
produce more favorable in vitro and in vivo complex stability
profiles relative to 88Y-1B4M-DTPA.10 CHX-A00-DTPA or
antibody conjugates of CHX-A00-DTPA were reported to
rapidly form a complex with 88Y or 86Y. However, the in vitro
serum stability data indicate that the Y(III) complexes of
CHX-A00-DTPA were less stable than those of C-DOTA.10b–d
A tumor-targeting monoclonal antibody (mAb) has been linked to
a cytotoxic radioisotope for antibody-targeted radiation cancer
therapy (radioimmunotherapy, RIT) for selective delivery of a
cytotoxic radioactive metal to cancer cells while minimizing
radiation exposure to healthy cells.1–3 RIT drugs for various
cancers have been extensively tested in clinical trials.4 Zevalins
is the first FDA-approved RIT drug for treatment of B-cell non-
Hodgkins lymphoma (NHL) and consist of a mAb (anti-CD20), a
90
1
pure beta-emitter Y (t = 64.1 h, Emax = 2.3 MeV), and a
2
bifunctional ligand 1B4M-DTPA(2-(4-isothiocyanatobenzyl)-6-
methyl-diethylene-triamine pentaacetic acid, Fig. 1).5
A successful RIT requires a bifunctional ligand that can
effectively bind a radioisotope with clinically acceptable com-
plexation kinetics and in vivo stability to minimize toxicity due to
dissociation of metal complex or radiolytic damage resulted from
extended exposure of the protein during radiolabeling.6 Despite
great potential of RIT as a cancer therapeutic modality proven
by Zevalins, less progress has been made on improvement of
chelation chemistry. Among the bifunctional ligands available
for RIT, C-DOTA (2-(4-nitrobenzyl)-1,4,7,10-tetraazacyclotetra-
decane-1,4,7,10-tetra-acetic acid, Fig. 1) and C-DTPA
(2-(4-nitrobenzyl)-diethylenetriamine pentaacetic acid, Fig. 1)
have been extensively investigated. The macrocyclic ligand
C-DOTA is known to form a stable complex with many different
radionuclides.7 However, slow complexation kinetics of DOTA
under mild reaction conditions remains an obstacle limiting the
wide practical use of the ligand in RIT of relatively short
half-lived a- and b-emitters such as 90Y, 212Bi (t1/2 = 60.6 m),
213Bi (t1/2 = 45.6 m), or 212Pb (t1/2 = 10.64 h).8 Acyclic C-DTPA
instantly binds to the radionuclides, but the radiolabeled
Fig. 1 Ligands in preclinical or clinical use for RIT.
We reported that NETA ({4-[2-(bis-carboxymethyl-amino)-
ethyl]-7-carboxymethyl-[1,4,7]triazonan-1-yl}-acetic acid, Fig. 1)
is a promising chelator for RIT of a- and b-emitters including
90Y, 177Lu (t1/2 = 6.7 d), 212Bi, 213Bi, and 212Pb.11–13 NETA
possesses both a parent macrocyclic NOTA (1,4,7-triazacyclo-
nonane-N,N0,N0 0-triacetic acid, Fig. 1) backbone and a flexible
acyclic tridentate pendant arm. The idea of designing octadentate
NETA was to integrate the advantage of both the macrocyclic
NOTA and acyclic DTPA frameworks, i.e., both high thermo-
dynamic stability and favorable formation kinetics. Indeed,
NETA was found to bind to the metals with favorable
complexation kinetics, and the corresponding NETA–metal
complexes were stable in vitro and in vivo.11–13
As part of our continued effort to develop an optimized
bifunctional ligand for RIT, a bifunctional version of NETA,
3p-C-NETA ({4-[2-(bis-carboxy-methylamino)-5-(4-nitrophenyl)-
pentyl]-7-carbo-xymethyl-[1,4,7]tri-azonan-1-yl} acetic acid,
Fig. 1) is designed. 3p-C-NETA possesses a longer propyl
chain that is used to connect the NETA backbone and the
functional p-NO2 benzyl group for conjugation to antibody.
The extended alkyl spacer was proposed to reduce steric
Chemistry Division, Biological, Chemical, and Physical Sciences
Department, Illinois Institute of Technology, 3101 S. Dearborn St,
LS 182, Chicago, IL, USA. E-mail: chong@iit.edu;
Fax: 312-567-3494; Tel: 312-567-3235
w Electronic supplementary information (ESI) available: Experimental
section including synthetic procedures and NMR spectra of compounds
10, 11, and 13 and ITLC and SE-HPLC data and spectra for assessment of
radiolabeling and serum stability. See DOI: 10.1039/c0cc05707j
c
5584 Chem. Commun., 2011, 47, 5584–5586
This journal is The Royal Society of Chemistry 2011

View Article Online
hindrance in complexation of the ligand with a metal resulting
in improved complexation kinetics as compared to the stan-
dard methyl chain present in the known bifunctional ligands
(Fig. 1). Herein, we report synthesis of a highly effective
bifunctional ligand 3p-C-NETA based on an unexpected
regiospecific ring opening of an aziridinium ion by nucleo-
philic bisubstituted TACN. 3p-C-NETA was evaluated for
radiolabeling reaction kinetics with 90Y and 177Lu. The corres-
ponding 90Y-3p-C-NETA and 177Lu-3p-C-NETA complexes
were evaluated for in vitro serum stability. The in vitro
complexation kinetics and stability data of 3p-C-NETA were
compared to those of C-DOTA.
We have recently reported that bromination of N,N-
bisubstituted b-amino alcohols provided aziridinium salts that
promptly underwent regiospecific ring opening by the counter-
anion bromide to produce secondary N,N-bisubstituted
b-amino bromides.14 We wanted to apply aziridinium ion
chemistry to the synthesis of 3p-C-NETA. The key reaction
steps in the synthesis of 3p-C-NETA are the formation and the
regiospecific ring opening of 9 by bisubstituted TACN 1213 to
provide 13 (Schemes 1 and 2).
Scheme 2 Synthesis of 3p-C-NETA via regiospecific ring opening of
aziridinium ion 9.
Reaction of 10 with bisubstituted TACN (12) in the presence of
DIPEA as a base provided 13 as the exclusive product in an
excellent isolated yield (82%). It is reasonably proposed that 10
first rearranged to form aziridinium ion 9 wherein the less hindered
methylene carbon was attacked by the bulky nucleophile 12 to
provide the regioisomer 13. We initially attempted the synthesis of
13 by the reaction of 10 with 12 in CH3CN under reflux. However,
the reaction provided 13 in a very poor yield (30%) due to the
formation of an oxomorpholine derivative 14 as the major product
from an intramolecular rearrangement of 9. To avoid the forma-
tion of 14, the reaction was repeated at room temperature and
provided only 13 in a significantly improved isolated yield along
with the unreacted starting materials 10 and 12. The regioisomer
15 that could be formed by nucleophilic attack of 12 at the more
hindered methylene carbon in 9 was not isolated from either of the
reactions. The t-butyl groups in 13 were removed by the treatment
of 13 with 4M HCl/1,4-dioxane, and 3p-C-NETA in the nitro
form was obtained in a quantitative yield (Scheme 2).
Scheme 1 Synthesis of secondary b-amino bromide 10 and aziridindium
ion 11.
To confirm the regiochemistry observed in the nucleophilic
ring opening of the aziridinium ion 9, compound 13 was
separately prepared based on the reaction route starting from
compound 6 (Scheme 3). Swern oxidation of N-BOC protected
amino alcohol 16 that was prepared by reaction of 6 with
BOC-ON provided compound 17. Reductive amination of 17
with 18 using sodium triacetoxyborohydride provided 19
which was subsequently treated with HCl (g) in 1,4-dioxane
to afford 20. A base-promoted reaction of 20 with tert-butyl
bromoacetate produced compound 13 in a very poor isolated
yield (o5%) due to the formation of the partially- or poly-
alkylated byproducts in the reaction mixture and resultant low
purification yield. The isolated yield of 20 may be improved
through optimization of alkylation reaction condition. The
low-yield alkylation reaction of 20 clearly highlights the value
of the efficient synthetic method for polar macrocyclic
bifunctional ligands based on regiospecific ring opening of
labile aziridinium ion (Scheme 2) reported herein. Finally,
regiochemistry in the ring opening of 9 was confirmed by the
identical 1H and 13C NMR data of 13 that was separately
prepared via the two synthetic routes (Schemes 2 and 3).
The new ligand 3p-C-NETA was evaluated for radiolabeling
efficiency with the b-emitting radioisotopes, 90Y and 177Lu.
Radiolabeling reaction kinetics of 3p-C-NETA with 90Y or
177Lu was performed at room temperature and various pH
Secondary b-amino bromide 10 was prepared by a practical,
reproducible, and readily scalable synthetic route as shown in
Scheme 1. The starting material p-nitrophenylpropyl bromide
2 was prepared via modification of a known procedure15 and
reacted with sodium salt of diethyl acetamido malonate to
afford compound 3. Racemic p-nitrophenylpropylalanine 4
that was prepared from decarboxylation and removal of the
acetyl protection group in 3 was converted to amino methyl
ester 5. Reduction of 5 with NaBH4 followed by alkylation of
6 with t-butyl bromoacetate provided 7. Bromination of 7 in
CH2Cl2 using NBS and PPh3 provided the secondary b-amino
bromide 10 in 66% isolated yield. Compound 10 was fully
characterized by 1H and 13C NMR and CHN and HRMS
analysis (supporting informationw). A proposed reaction
mechanism for the transformation of 7 to 10 is shown in
Scheme 1. Aziridinium salt 9 containing counter anion
bromide was initially formed from intramolecular rearrange-
ment of 8 via attack of the nucleophilic nitrogen atom followed
by removal of triphenylphosphine oxide. Regiospecific ring
opening of 9 at the more hindered carbon provided secondary
b-amino bromide 10. Aziridinium ion 11 was isolated from the
reaction of 10 with silver perchlorate and characterized by 1H
and 13C NMR analysis (the Supporting Informationw).
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 5584–5586 5585

View Article Online
the complexes was observed (Supporting informationw).
90Y-C-DOTA and 177Lu-C-DOTA were quite stable in serum
over the time period of measurements by ITLC and HPLC
(Supporting informationw). However, both 90Y-C-DOTA and
177Lu-C-DOTA in serum produced an unbound 90Y or
177Lu-related impurity that appeared as a very small peak in
radio SE-HPLC chromatograms (tR = B14 min).
In summary, the new bifunctional ligand was efficiently
prepared by the new synthetic method centered on the regio-
specific ring opening of the aziridinium ion. Radiolabeling of
3p-C-NETA with 90Y or 177Lu was found to be highly efficient
under mild conditions. 90Y-3p-C-NETA and 177Lu-3p-C-NETA
were stable in serum for at least 14 days. The radiolabeling kinetics
and in vitro serum stability data suggest that 3p-C-NETA is a very
promising bifunctional ligand that can be directly used for RIT
applications of various cancers and may overcome the limitations
associated with the currently available ligands C-DOTA and
C-DTPA for RIT. 3p-C-NETA conjugated to an antibody is being
investigated for RIT applications of a or b-emitting radioisotopes.
We appreciate the financial support from the National
Institutes of Health (R01CA112503). Hyun A Song is the
recipient of 2008 and 2009 IIT Kilpatrick fellowship.
Scheme 3 Structural determination of 13 for confirmation of regio-
chemistry in ring opening of 9.
compared to that of C-DOTA (Table 1 and the Supporting
Informationw). 3p-C-NETA instantly formed a complex with
90Y at pH 5.5 with the excellent radiolabeling efficiency (97.4 ꢀ
0.7%) in 1 min (Table 1). Radiolabeling of 3p-C-NETA with
177Lu was essentially complete in 1 min at all the range of pH
studied. C-DOTA was significantly slower in binding 90Y at pH
5.5 (83.5 ꢀ 8.13%, 1 h) relative to 3p-C-NETA, and radio-
labeling of C-DOTA with 90Y was incomplete even at 1 h time
point. Formation of 177Lu-C-DOTA was relatively faster than
that of 90Y-C-DOTA at all the range of pH. In vitro serum
stability of the radiolabeled complexes was performed to
determine if 3p-C-NETA or C-DOTA radiolabeled with 90Y
or 177Lu remained stable without loss of 90Y or 177Lu in human
serum ITLC and radio size-exclusion HPLC (Supporting
Informationw). 90Y-3p-C-NETA and 177Lu-3p-C-NETA were
readily prepared from the reactions of 3p-C-NETA with 90Y or
177Lu at room temperature and directly used for serum stability
studies. C-DOTA is reported to form less stable intermediate
metal complexes with Lanthanides under mild reaction condi-
tion based on a three step complexation mechanism.16 To
ensure complete radiolabeling and formation of the most stable
complexes of C-DOTA, 90Y-C-DOTA and 177Lu-C-DOTA
were prepared by the reaction of C-DOTA with 90Y or 177Lu
at either room temperature (177Lu) or 37 1C (90Y) over an
extended period of time. Radiolabeling of C-DOTA with 90Y
was determined to be incomplete even at 30 h time point by
SE-HPLC. 90Y-C-DOTA was separated from a small amount
of unbound 90Y present in the reaction mixture through a
Chelex-100 column. 177Lu-3p-C-NETA, 90Y-3p-C-NETA, and
177Lu-C-DOTA were used without further purification. Both
90Y-3p-C-NETA and 177Lu-3p-C-NETA were stable in human
serum for 2 weeks as evidenced by SE-HPLC and ITLC
analysis, and no measurable radioactivity dissociated from
Notes and references
1 S. J. Knox and R. F. Meredith, Semin. Radiat. Oncol., 2000, 10, 73–93.
2 M. R. McDevitt, D. Ma, L. T. Lai, J. Simon, P. Borchardt,
R. K. Frank, K. Wu, V. Pellegrini, M. J. Curcio, M. Miederer,
N. H. Bander and D. A. Scheinberg, Science, 2001, 294, 1537–1540.
3 S. Srivastava and E. Dadachova, Semin. Nucl. Med., 2001, 31, 330–341.
4 D. E. Milenic, E. D. Brady and M. W. Brechbiel, Nat. Rev.
Drug Discovery, 2004, 3, 488–499.
5 (a) S. J. Knox, M. L. Goris, K. Trisler, R. Negrin, T. Davis,
T. M. Liles, A. J. Grillo-Lopez, P. Chinn, C. Varns, S. C. Ning,
S. Fowler, N. Deb, M. Becker, C. Marquez and R. Levy, Clin.
Cancer Res., 1996, 2, 457–470.
6 (a) D. Parker, Chem. Soc. Rev., 1990, 19, 271–291;
(b) M. C. Chakrabarti, N. Le, C. H. Paik, W. G. De Graff and
J. A. Carrasquillo, J. Nucl. Med., 1996, 37, 1384–1388.
7 (a) M. K. Moi, C. F. Meares and S. J. DeNardo, J. Am. Chem.
Soc., 1988, 110, 6266–7; (b) M. Kodama, T. Koike, A. B. Mahatma
and E. Kimura, Inorg. Chem., 1991, 30, 1270–1273.
8 (a) J. B. Stimmel, M. E. Stockstill and F. C. Kull, Jr., Bioconjugate
Chem., 1995, 6, 219–225; (b) L. L. Chappell, D. Ma, D. E. Milenic,
K. Garmasteni, V. Venditto, M. P. Beitzel and M. W. Brechbiel,
Nucl. Med. Biol., 2003, 30, 581–595.
9 S. Hassfjell and M. W. Brechbiel, Chem. Rev., 2001, 101, 2019–36.
10 (a) M. W. Brechbiel and O. A. Gansow, J. Chem. Soc., Perkin
Trans. 1, 1992, 1173–8; (b) L. Camera, S. Kinuya, K. Garmestani,
C. Wu, M. W. Brechbiel, L. H. Pai, T. J. McMurry, O. A. Gansow,
I. Pastan, C. H. Paik and J. A. Carrasquillo, J. Nucl. Med., 1994,
35, 882–889; (c) M. J. Blend, J. J. Stastny, S. M. Swanson and
M. W. Brechbiel, Cancer Biother. Radiopharm., 2003, 18, 355–63;
(d) H. Kobayashi, C. Wu, T. M. Yoo, B. F. Sun, D. Drumm,
I. Pastan, C. H. Paik, O. A. Gansow, J. A. Carrasquillo and
M. W. Brechbiel, J. Nucl. Med., 1998, 39, 829–36.
Table 1 Radiolabling efficiency (%) of 3p-C-NETA or C-DOTA
with 90Y or 177Lu (RT, 0.25M NH4OAC, pH 5.5)a
11 H. S. Chong, K. Garmestani, D. E. Millenic and M. W. Brechbiel,
J. Med. Chem., 2002, 45, 3458–3464.
12 H. S. Chong, D. E. Milenic, K. Garmestani, E. D. Brady, H. Arora,
D. Pfiester and M. W. Brechbiel, Nucl. Med. Biol., 2006, 33, 459.
13 H. S. Chong, H. A. Song, N. Birch, T. Le, S. Lim and X. Ma.
Bioorg, Bioorg. Med. Chem. Lett., 2008, 18, 3436–39.
14 H. S. Chong, H. A. Song, M. Dadwal, X. Sun, I. Sin and Y. Chen,
J. Org. Chem., 2010, 75, 219–21 and 7966.
15 R. L. Blankespoor, T. DeVries, E. Hansen, J. M. Kallemeyn,
A. M. Klooster, J. A. Mulder and R. P. D. A. Vander Griend,
J. Org. Chem., 2002, 67, 2677–2681.
3p-C-NETA
Time (min) 90Y
C-DOTA
177Lu
90Y
177Lu
1
5
10
20
30
60
97.4 ꢀ 0.7 100.0 ꢀ 0.0 77.1 ꢀ 3.7
98.1 ꢀ 1.1 100.0 ꢀ 0.0 78.8 ꢀ 5.1
94.5 ꢀ 3.9
98.8 ꢀ 1.1
98.7 ꢀ 1.6 100.0 ꢀ 0.0 69.4 ꢀ 10.6 99.5 ꢀ 0.5
98.7 ꢀ 2.2 100.0 ꢀ 0.0 71.2 ꢀ 11.2 99.9 ꢀ 0.1
99.4 ꢀ 0.9 100.0 ꢀ 0.0 76.1 ꢀ 9.52 99.9 ꢀ 0.1
99.5 ꢀ 1.0 100.0 ꢀ 0.0 83.5 ꢀ 8.13 100.0 ꢀ 0.0
a
ITLC (CH3CN : H2O = 3 : 2); Radiolabeling efficiency (mean ꢀ
16 J. Moreau, E. Guillion, J.-C. Pierrard, J. Rimbault, M. Port and
M. Aplincourt, Chem.–Eur. J., 2004, 10, 5218–32.
standard deviation %) was measured in triplicate (n = 3).
c
5586 Chem. Commun., 2011, 47, 5584–5586
This journal is The Royal Society of Chemistry 2011