Rapid and sensitive LC-MS approach to quantify non-radioactive transition metal impurities in metal radionuclides

Article abstract of DOI:10.1039/c3cc39071c

A rapid and sensitive LC-MS approach for the quantification of non-radioactive metal contaminants present in metal radionuclide formulations was developed. Traditional ¹²C/¹H and heavy stable isotope ¹³C/²H-labeled chelator-amino acid conjugates are used as chelating agents to quantify contaminating transition metals, allowing for determination of effective specific activity of radio-metals. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3cc39071c

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Rapid and sensitive LC-MS approach to quantify
non-radioactive transition metal impurities in
metal radionuclides†
Cite this: Chem. Commun., 2013,
49, 2697
Received 19th December 2012,
Accepted 13th February 2013
Dexing Zeng and Carolyn J. Anderson*
DOI: 10.1039/c3cc39071c
www.rsc.org/chemcomm
A rapid and sensitive LC-MS approach for the quantification of can be achieved, but also the ESA of radio-metals for a parti-
non-radioactive metal contaminants present in metal radionuclide cular chelator can be determined. Furthermore, this LC-MS
formulations was developed. Traditional 12C/1H and heavy stable method allows for optimization of the radiolabelling condi-
isotope 13C/2H-labeled chelator–amino acid conjugates are used as tions to achieve high specific activity by using conditions
chelating agents to quantify contaminating transition metals, allowing favouring the chelation of a particular radio-metal.
for determination of effective specific activity of radio-metals.
Stable isotope labelling agents, such as ICAT, CILAT, iTRAQ,
and SILAC, have been developed and widely used for quantita-
Metal-based radiopharmaceuticals have been widely used for tive proteomics with high precision.11–18 The use of stable
radiotherapy and diagnostic imaging, and in recent decades, isotope-labelled tags involves the addition of a chemically
their development has been a rapidly growing area for both identical form of the analyte(s) containing a stable heavy
clinical and pre-clinical studies.1–3 Radio-metals typically con- isotope (e.g., 2H, 13C, 15N, etc.) to a solution containing an
tain significant amounts of cold metal impurities that impede internal standard. Due to the ionization variability for different
effective radiolabelling at high specific activities.4,5 Achieving compounds, the best internal standard for quantification is
high specific activity (the amount of radioactivity per unit of typically the same compound labelled with a stable isotope(s).
mass) is extremely important for imaging agents that target The internal standard and analyte differ only by the incorpora-
proteins, such as tumour receptors, present in very low (nM or tion of either the heavy stable isotopes, or the native stable
less) concentrations in vivo.6,7 The effective specific activity isotopes. In theory, all compounds in the combined sample
(ESA) of a radio-metal is typically obtained by radiolabelling exist as analyte pairs of identical sequence but differing
titrations with DOTA (1,4,7,10-tetraazacyclododecane-1,4,7,10- masses. The compounds have the same properties and behave
tetraacetic acid) and/or TETA (1,4,8,11-tetraazacyclotetra- with identical characteristics under any isolation or separation
decane-1,4,8,11-tetraacetic acid).8,9 However, the titration step. Thus, the ratio between intensities of the pair of com-
method will not provide information on the identity of the pounds provides an accurate measurement of relative abun-
contaminating metals or the relative amounts present, and dance, and therefore the abundance of protein containing this
additionally suffers from inconsistent results that are depen- pair of compounds can also be measured.
dent on the chelator and/or the labelling conditions used.
In this study, we used a similar approach to quantify
Although ion chromatography (IC) has also been used to contaminating transition metal complexes, where 13C/2H
analyse metal impurities,10 it is limited by low sensitivity, chelator–amino acid conjugates were added to a radio-metal
and requires high concentrations of radio-metals that can be solution. The amounts of contaminating transition metal com-
hazardous from a radiation safety standpoint. To overcome plexes were quantified by LC-MS analysis using the corre-
these problems, a rapid and sensitive LC-MS approach using sponding light metal complexes containing only native 12C/1H
traditional 12C/1H and stable isotope 13C/2H-labeled agents was as internal standards. Based on the intensities of the pairs of
developed. Through this methodology, not only quantification metal complexes, non-radioactive metal contaminants present
of non-radioactive metal contaminants in metal radionuclides in metal radionuclide formulations can be readily quantified
for determination of the ESA for the specific chelator.
To demonstrate this new technology, we synthesized 12C/1H-
Department of Radiology, University of Pittsburgh, 100 Technology Drive,
and 13C/2H-labelled agents of EdF-DOTA (Ethylenediamine-
Suite 452F, Pittsburgh, PA 15219, USA. E-mail: andersoncj@upmc.edu;
F(Phe)-DOTA, Fig. 1), and used them to quantify non-radioactive
Fax: +1 412-624-2598; Tel: +1 412-624-6887
metal impurities in solutions of the PET radionuclide, Cu-64
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c3cc39071c
(T1/2 = 12.7 h, b+: 0.655 MeV, 17.8%; bÀ: 0.573 MeV, 41%).
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 2697--2699 2697

View Article Online
Communication
ChemComm
Fig. 1 Structure of EdF-DOTA consisting of a MS tag and DOTA chelator.
Fig. 2 MS spectrum of Fe(III)-DOTA-EdF complexes.
The use of the 13C/2H MS tag was inspired by a recent report on the
use of deuterium-labelled reagents for protein quantification,19
as the cost of a 13C/2H-encoded agent is significantly less
expensive than a fully encoded 13C-labelled compound. The
two reagents are a set of structurally identical molecules con-
sisting of a MS tag (moieties that can be easily identified by MS)
and a metal chelation group. The ‘‘heavy’’ MS tag is encoded
with 13C2 (50 atom% 13C in phenylalanine) and 2H4 (98 atom%
D in ethylenediamine), and the ‘‘light’’ MS tag includes only
native 12C/1H. DOTA was chosen as the metal chelation group
since it is widely used in metal-based radiopharmaceuticals
and chelates with the transition metals Fe3+, Ni2+, Cu2+, Co2+,
and Zn2+ that are the major impurities in cyclotron-produced
Cu-64.20 In order to simplify the quantification process, a
complete separation between MS peaks from the heavy and
light complexes (at least 5 Da) is required. For example, the
MS of a Ni2+ complex of light EdF-DOTA is distributed from
650 to 654 due to naturally occurring nickel (Ni), which is
primarily composed of four stable isotopes, 58Ni, 60Ni, 61Ni and
62Ni. In order to minimize the overlap between the light and
heavy Ni complex, the MS of a Ni(II) complex with the heavy
reagent should have a MW of at least 655, which is a 5 Da
(or more) difference between the heavy and light reagents
(Fig. S1, ESI†).
was observed. To further demonstrate the capability to quantify
metal chelates by LC-MS, a heavy Fe3+ complex solution (Fe(III)-
DOTA-EdF; 0.5 nmol) was mixed with the light Fe3+ complex
solution (0.1 nmol), and diluted with NH4OAc buffer for LC-MS
analysis (Fig. 2). By comparing the total ion counts between
heavy and light Fe(III)-DOTA-EdF complexes, the ratio of heavy
(Fe(III)) to light (Fe(III)) complexes was 401/82 = 4.89, which is
very close to the predetermined ratio of 5.0. The LC-MS
approach was also used to measure the ratios of heavy and
light Cu(II) complexes that were predetermined from 0.1 to 6.0,
and a good linear correlation (R = 0.9934) between the pre-
determined ratio and LC-MS ratio was demonstrated over a
broad range (Fig. S2, ESI†).
Following the above validation of the LC-MS method, the
light and heavy EdF-DOTA agents were used to quantify the
metal impurities in a solution of cyclotron produced Cu-64
from one of the two vendors in the U.S. The Cu-64 solution was
neutralized and buffered with 0.1 M NH4OAc solution (pH =
6.80), and when it was completely decayed (1 week), the excess
heavy EdF-DOTA agent was added and incubated at 22 1C for 10
or 30 min, or at 60 1C for 30 min. After the incubation, the light
complex solutions (internal standards) were added and mixed
thoroughly. The mixture was then loaded into the LC-MS for
analysis. Under the optimized LC-MS conditions, the five com-
plexes were completely separated (Fig. 3). The UV trace was not
shown here because the low molar absorptivity of the complex
solutions led to very low signals. The mole amounts of the
heavy metal complexes were calculated by comparing the MS
The heavy EdF-DOTA agent was prepared by conventional
solid-phase synthesis (Scheme S1, ESI†). Firstly, 2H enriched
ethylenediamine was loaded onto the trityl chloride resin via its
primary amine group. Next, Fmoc-(13C2)Phe-OH that was pre-
pared from the 13C2-enriched phenylalanine was attached to
the resin using HBTU/HOBt as the coupling reagents. After
the Fmoc group was removed, a primary amine group was
generated for the subsequent DOTA–NHS coupling. Finally, the
EdF-DOTA agent with the heavy tag was cleaved from the resin
by acid treatment and purified by HPLC. The light-tagged agent
was synthesized in a similar manner using native ethylene-
diamine and phenylalanine. After obtaining the purified light
EdF-DOTA reagent, internal standards of transition metal com-
plexes (10 mM) were prepared by incubating excess light EdF-
DOTA with a known amount of the transition metals for 30 min
at 60 1C. The complexes are stored at À80 1C.
To prove that the above five transition metals do not bind
with the MS tag, the MS tag without DOTA was prepared and
incubated with those transition metals under the radiolabelling
conditions. Based on HPLC and MS analysis, there was only
the MS tag in the incubation mixture and no metal complex Fig. 3 TIC trace of LC-MS chromatography of the light–heavy complex mixtures.
c
2698 Chem. Commun., 2013, 49, 2697--2699
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
Table 1 Heavy complexes and calculated ESA at 60 1C for 30 min
Metal complex
Fe(III)
Ni(II)
Cu(II)
Co(II)
Zn(II)
Standard intensity
Sample intensity
Sample (pmol)
726 Æ 13
1943 Æ 134
133
2987 Æ 203
2299 Æ 131
39
2405 Æ 219
1440 Æ 211
30
2148 Æ 210
345 Æ 9
8
734 Æ 10
2537 Æ 101
172
Effective SA (GBq mmolÀ1
)
19.4 = 7.4 MBq/(133 + 39 + 30 + 8 + 172) pmol
complexed with Cu2+ in the presence of other contaminating
metals, such as Fe3+, Zn2+.
In summary, a novel LC-MS approach has been developed
to quantify non-radioactive metal impurities and determine
effective specific activity (ESA) of radio-metals. 13C/2H- and
12C/1H-labelled EdF-DOTA agents were synthesized and were
used for rapid and sensitive quantification of metal impurities
Table 2 Heavy complexes and calculated ESA under different conditions
Complex (pmol) Fe(III) Ni(II) Cu(II) Co(II) Zn(II) ESA (GBq mmolÀ1
)
22 1C, 10 min
22 1C, 30 min
60 1C, 30 min
37
48
133
30
32
39
23
25
30
11
7.5 152 28.0
172 19.4
120 33.5
8
intensities between the heavy and light complexes to the known found in cyclotron produced Cu-64. Besides the quantification
amount of the light complex (50 pmol). The effective specific of radio-metals for ESA, this LC-MS method can also be used
activities were then calculated based on the Cu-64 activity as a valuable tool to optimize radiolabelling conditions and
(7.4 MBq (200 mCi), decay corrected to the time of receipt) evaluate the specificity of a chelator for a particular radio-
and the total mole amount of five heavy complex contaminants metal. Achieving high ESA for radio-metals is critical for PET
(382 pmol).
imaging or targeted radiotherapy of low capacity receptor
The LC-MS measurements were performed in triplicate and systems. This novel methodology can be applied for a variety
the results obtained at 60 1C with 30 min incubation are of radio-metals and chelators in the development of PET
summarized in Table 1. Only 7.4 MBq of Cu-64 was needed to imaging radiopharmaceuticals and metal-based targeted radio-
obtain adequate TIC chromatograms, which demonstrates the therapy agents for many diseases, including cancer, cardio-
high sensitivity of this LC-MS approach. It should be noted that vascular disease and pulmonary disorders.
although we allowed the Cu-64 to decay prior to analysis,
Notes and references
radioactive samples are expected to give the same results. The
major metal impurities in the test solutions were confirmed to
be Fe3+, Ni2+, Cu2+, Co2+ and Zn2+; there were no other MS peaks
in the TIC chromatogram. The metal-complex contaminants
were quantified, and the results indicate that the Zn2+ and Fe3+
dramatically affected the radiolabelling of EdF-DOTA with the
cyclotron produced Cu-64. Therefore, purification of Fe3+ and
Zn2+ should be a focus for secondary purification of Cu-64.
Alternatively, developing a chelator that specifically chelates
Cu2+ with minimal Fe3+ and Zn2+ chelation is highly desired for
improving ESA of Cu-64 radiopharmaceuticals.
1 D. E. Reichert, J. S. Lewis and C. J. Anderson, Coord. Chem. Rev.,
1999, 184, 3–66.
2 K. Ogawa and H. Saji, Int. J. Mol. Imaging, 2011, 2011, 537687.
3 S. Liu, Adv. Drug Delivery Rev., 2008, 60, 1347–1370.
4 T. J. Wadas, E. H. Wong, G. R. Weisman and C. J. Anderson, Chem.
Rev., 2010, 110, 2858–2902.
5 M. A. Avila-Rodriguez, J. A. Nye and R. J. Nickles, Appl. Radiat. Isot.,
2007, 65, 1115–1120.
6 D. A. Mankoff, J. M. Link, H. M. Linden, L. Sundararajan and
K. A. Krohn, J. Nucl. Med., 2008, 49(suppl. 2), 149S–163S.
7 I. Velikyan, G. J. Beyer, E. Bergstrom-Pettermann, P. Johansen,
M. Bergstrom and B. Langstrom, Nucl. Med. Biol., 2008, 35, 529–536.
8 D. W. McCarthy, R. E. Shefer, R. E. Klinkowstein, L. A. Bass,
W. H. Margeneau, C. S. Cutler, C. J. Anderson and M. J. Welch,
Nucl. Med. Biol., 1997, 24, 35–43.
9 L. P. Szajek, W. Meyer, P. Plascjak and W. C. Eckelman, Radiochim.
Acta, 2005, 93, 239–244.
10 J. van Doornmalen, J. T. van Elteren and J. J. de Goeij, Anal. Chem.,
2000, 72, 3043–3049.
11 S. P. Gygi, B. Rist, S. A. Gerber, F. Turecek, M. H. Gelb and
R. Aebersold, Nat. Biotechnol., 1999, 17, 994–999.
12 S. Li and D. Zeng, Chem. Commun., 2007, 2181–2183.
13 S. E. Ong and M. Mann, Nat. Protocols, 2006, 1, 2650–2660.
14 D. Zeng and S. Li, Bioorg. Med. Chem. Lett., 2009, 19, 2059–2061.
This LC-MS method also allows for optimization of the
labelling conditions to achieve high specific activity by using
conditions favouring the chelation of a specific radio-metal.
From the data shown in the Table 2, it was observed that the
ESA varied with different incubation times and temperatures.
Extending the incubation time or increasing the incubation
temperature resulted in decreased ESA. At elevated tempera-
tures, significantly more Fe3+ and Zn2+ chelated EdF-DOTA,
whereas only slightly more Cu2+ complexed the EdF-DOTA 15 P. L. Ross, Y. N. Huang, J. N. Marchese, B. Williamson, K. Parker,
reagent, resulting in the ESA decreasing from 28.0 GBq mmolÀ1
S. Hattan, N. Khainovski, S. Pillai, S. Dey, S. Daniels, S. Purkayastha,
P. Juhasz, S. Martin, M. Bartlet-Jones, F. He, A. Jacobson and
D. J. Pappin, Mol. Cell. Proteomics, 2004, 3, 1154–1169.
at 22 1C to 19.4 GBq mmolÀ1 at 60 1C, and after decay correction,
it was close to the titration result from vendor. A similar result 16 K. Gevaert, F. Impens, B. Ghesquiere, P. Van Damme, A. Lambrechts
and J. Vandekerckhove, Proteomics, 2008, 8, 4873–4885.
17 L. V. Schneider and M. P. Hall, Drug Discovery Today, 2005, 10, 353–363.
was obtained with extended incubation time (10 to 30 min), and
the ESA decreased from 33.5 GBq mmolÀ1 to 28.0 GBq mmolÀ1
.
18 A. Treumann and B. Thiede, Expert Rev. Proteomics, 2010, 7,
647–653.
19 D. Zeng and S. Li, Chem. Commun., 2009, 3369–3371.
20 D. Zeng, N. S. Lee, Y. Liu, D. Zhou, C. S. Dence, K. L. Wooley,
J. A. Katzenellenbogen and M. J. Welch, ACS Nano, 2012, 6,
5209–5219.
These data demonstrate that this LC-MS approach can be used
to optimize radiolabelling conditions to decrease chelation of
contaminating metals for achieving higher ESA, since it can be
challenging to separate DOTA-conjugated peptides that are
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 2697--2699 2699