Radiosynthesis of Carbon-11 Labeled Puromycin as a Potential PET Candidate for Imaging Protein Synthesis in Vivo

Article abstract of DOI:10.1021/acsmedchemlett.6b00093

In order to address the limitations associated with the present range of PET radiotracers used for imaging protein synthesis in vivo we have synthesized a candidate PET radiotracer based on Puromycin (3, PURO), a protein synthesis inhibitor. The desmethylPURO 9 precursor for radiolabeling with carbon-11 radioisotope was synthesized in two steps employing EDC/HOBt amide coupling in overall 76% yield. Optimal conditions for radiolabeling were then established via methylation/deprotection sequence. Under these conditions as determined by NMR analysis 9 showed partial stability (ca. 80%) under acidic conditions. Limited evidence of stereochemical stability of 3 was also found. The radiolabeling of intermediate [¹¹C]12 was accomplished with up to 57% conversion from [¹¹C]iodomethane. An automated method was then developed for high radioactivity radiosynthesis to produce [¹¹C]3 ([¹¹C]PURO) in 16 ± 6% (n = 3) decay corrected radiochemical yields.

Full text of DOI:10.1021/acsmedchemlett.6b00093

Letter
pubs.acs.org/acsmedchemlett
Radiosynthesis of Carbon-11 Labeled Puromycin as a Potential PET
Candidate for Imaging Protein Synthesis in Vivo
Selena Milicevic Sephton* and Franklin I. Aigbirhio
Molecular Imaging Chemistry Laboratory, Wolfson Brain Imaging Centre, Department of Clinical Neurosciences, University of
Cambridge, Box 65, Cambridge Biomedical Campus, Hills Road, Cambridge CB2 0QQ, U.K.
S
* Supporting Information
ABSTRACT: In order to address the limitations associated
with the present range of PET radiotracers used for imaging
protein synthesis in vivo we have synthesized a candidate PET
radiotracer based on Puromycin (3, PURO), a protein
synthesis inhibitor. The desmethylPURO 9 precursor for
radiolabeling with carbon-11 radioisotope was synthesized in
two steps employing EDC/HOBt amide coupling in overall
76% yield. Optimal conditions for radiolabeling were then
established via methylation/deprotection sequence. Under
these conditions as determined by NMR analysis 9 showed
partial stability (ca. 80%) under acidic conditions. Limited evidence of stereochemical stability of 3 was also found. The
radiolabeling of intermediate [11C]12 was accomplished with up to 57% conversion from [11C]iodomethane. An automated
method was then developed for high radioactivity radiosynthesis to produce [11C]3 ([11C]PURO) in 16 6% (n = 3) decay
corrected radiochemical yields.
KEYWORDS: Positron emission tomography, puromycin, rate of protein synthesis
Chart 1. (A) Structures of Most Commonly Employed
Amino Acids for PET Imaging of Protein Synthesis and (B)
Structures of Puromycin Derivatives and Aminoacyl-tRNA
he rate of protein synthesis (RPS), which represents one
Tof the hallmarks of cancer, is a target for imaging with
positron emission tomography (PET). To image RPS in vivo,
protein building blocks, amino acids, for example, radiolabeled
with PET radionuclides (e.g., fluorine-18; t1/2 = 109.8 min or
carbon-11; t1/2 = 20.4 min) have been employed extensively.1
To date, the most commonly used amino acid for PET
imaging is [11C]methionine ([11C]1, [11C]MET, Chart 1).1,2
The radiolabeling of [11C]1 is straightforward, and [11C]1 is
successful in crossing the blood−brain barrier (BBB). The
highest accumulation of [11C]1 was found in the organs with
the highest rate of protein synthesis in vivo: brain, liver, and
pancreas.3 Although PET imaging with radiolabeled amino
acids is routinely used, quantification of the PET data is
complicated due to amino acid involvement in other metabolic
processes (e.g., amino acid active transport).1 In 1993, a study
compared three radiolabeled amino acids: [3H]1, [18F]tyrosine
([18F]TYR, structure not shown), and [14C]leucine ([14C]2,
[14C]LEU, Chart 1)4 for their ability to image protein synthesis
in vivo. The authors found that the rate of protein synthesis was
reduced in brain and pancreas only with [14C]2 under blocking
conditions using the protein synthesis inhibitor cyclohexamide
in excess amounts. Subsequently, [11C]2 emerged as a superior
PET radiotracer for imaging protein synthesis in vivo;5−7
however, it is not without limitations. Amino acid [11C]2 is
radiolabeled with the short-lived carbon-11, and the radiosyn-
thesis is lengthy and challenging. This is primarily because it
requires carbon-11 HCN, which is not readily available from
the cyclotron.8,9 In addition, the radiosynthesis requires chiral
prep-HPLC separation to obtain the L-form from the crude
mixture. Furthermore, quantitative PET analysis requires
arterial blood sampling coupled with metabolite analysis,
Received: March 2, 2016
Accepted: April 12, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acsmedchemlett.6b00093
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
which involves trapping and measuring of gaseous [11C]CO2
radiometabolite.10
corresponding DSC ester 11. However, with the more reactive
free phenol functionality in the molecule, only a mixture of
inseparable products was obtained. Faced with the complica-
tions with the DSC ester coupling method, we revisited the
synthesis of the 3 (PURO) manifold17,18 and developed a two-
step synthesis of 9 as depicted in Scheme 1. The two hydroxyl
groups in 6 were protected with TBS to facilitate the
subsequent amide coupling by avoiding side reactions observed
when unprotected 6 was employed. Only a modest 30% yield of
9 was reached when unprotected 6 was employed for the
coupling directly. Similarly, protected precursor 9 is less likely
to undergo side reactions during radiolabeling. In the second
step, TBS-protected aminonucleoside 7 was coupled to
commercially available Boc-protected tyrosine 8 using HOBt/
EDC coupling conditions19 to afford 9 in 86% yield.
In order to overcome difficulties associated with the PET
analysis using amino acid based radiotracers, our group sought
to explore a potential PET radiotracer based on the Puromycin
(3, PURO, Chart 1) scaffold.11 Amide 3 is an aminonucleoside
antibiotic from Streptomyces alboniger and a protein synthesis
inhibitor. As such, 3 exhibits inhibitory activity via the covalent
attachment to the nascent protein chains.12−14 Eukaryotic
ribosome can then insert 3 (PURO) in place of aminoacyl-
tRNA (5, Chart 1) causing premature termination of
translation, resulting in truncated proteins. The concept of
imaging protein synthesis in vivo using protein synthesis
inhibitor has been employed previously.15 In 2012, the Salic
group16 employed 4, (O-PropagylPURO, Chart 1) and
fluorescence imaging was accomplished via the [2 + 3]
cycloaddition between 4 and fluorophore tagged azide. The
authors were successful in establishing that protein synthesis
was indeed inhibited as quantified by fluorescence measure-
ment in cultured cells as well as ex vivo.
With 9 in hand, we next explored conditions for methylation
of the phenolic functionality in 9 (Table 1), which could then
Table 1. Conditions for the Nonradioactive methylation of
DesmethylPURO 9
To assess this concept for imaging RPS in vivo by PET we
selected [11C]3 (O-carbon-11 labeled 3 (PURO), Chart 1) as a
potential radiotracer candidate. In analogy to fluorescence
imaging with 4,16 the concept of a PET PURO radiotracer
would be based on accumulation of radioactivity from [11C]3
within the ribosome, which can then be visualized and
quantified by PET. However, several challenges were
considered in the design and synthesis of [11C]3 arising from
the chemical scaffold of 3 having the following features: five
stereocenters, three reactive exchangeable functional groups
(two hydroxyl and one amine), and one labile amide bond.
Herein, we report the synthesis of a respective precursor for the
radiosynthesis of [11C]3, NMR experimental evidence for its
chemical and stereochemical stability as well as carbon-11
modular radiosynthesis of [11C]3.
The required desmethylPURO radiolabeling precursor
phenol 9 (Scheme 1) also has an advantage of being a
Scheme 1. Concise Synthesis of Common Intermediate/
Precursor 9 for Derivatization of 3 and Radiosynthesis of
[11C]3
a
b
Very low mass recovery (ca. 50%). Reaction done in MeCN as
solvent.
be translated to the radiosynthesis. Several bases were
investigated (entries 1−8), with NaOH (entries 1 and 2) and
K2CO3 (entries 3−6) showing good conversions to product.
However, with NaOH at 70 and 90 °C, low mass recovery
(50%) was found or the conversion to product was lower,
respectively (entries 1 vs 2). To avoid the possibility of
racemization with NaOH, further optimization of conditions
was conducted with K2CO3 and showed higher yield with
higher reaction temperature but shorter reaction time (entry 3
vs 4 and 5). The yield was significantly reduced with MeCN as
solvent (entry 6) and interestingly an aqueous solution of
K2CO3 yielded 82% of 12 (entry 5). A homogeneous reaction
mixture was particularly well suited for automation of
radiolabeling procedure.
In order to establish optimal reaction conditions for removal
of TBS and Boc protecting groups in one step, acidic hydrolysis
of 12 was investigated. Complete decomposition, particularly
seen through cleavage of the amide bond, was observed when
12 was treated with 2 M aq. HCl for 1 h at either 100 or 80 °C.
Similarly, 12 decomposed upon treatment with 1.25 M
methanolic HCl at 60 °C over 1 h as well as at ambient
common intermediate for further derivatization of 3.17,18 In
their synthesis of 4, Salic and co-workers used commercially
available puromycin aminonucleoside (6, Scheme 1) and
utilized a disuccinimidyl (DSC) ester coupling (not shown)
to form the critical amide bond in the framework of 3.16 In the
synthesis of 9, tyrosine 8 was initially used to form the
B
DOI: 10.1021/acsmedchemlett.6b00093
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
temperature over 30 min. Initial failed attempts to achieve
complete deprotection of 12 prompted us to investigate acidic
hydrolysis in an NMR experiment using a DCl/D2O/MeOD-d4
mixture with 12. Successful monitoring of the reaction progress
was dependent on the solubility of 12, which dissolved only
partly in DCl/D2O for which reason MeOD-d4 was added to
ensure solubility was complete. The mixture was heated to 70
°C and NMR analysis performed at intervals indicated in Table
2. An unidentified byproduct was observed and accounted for
Table 3. Comparison of Commercially Available 3 and
Synthesized Radiolabelling Intermediate 12 in One-Step
Hydrolysis (NMR Monitored)
Table 2. Time-Dependent Experiments for the One-Step
Deprotection to Obtain 3 (NMR Monitored)
formed 3 could not be unambiguously determined by NMR
despite good chemical shift agreement between synthesized 3
and that commercially sourced. However, a doubling of peaks
would be expected should epimerization occur, and this was not
observed.
Next, we explored methylation/deprotection sequence under
radiochemical (i.e., hot) conditions. For this purpose, aq.
K2CO3 was used, [11C]iodomethane produced from cyclotron-
generated [11C]CO2 was trapped, in DMF and this solution
was added to radiolabeling precursor 9. To reduce the reaction
time, reaction temperature was increased to 70, 75, or 80 °C
(Table 4). Aliquots were analyzed by HPLC after 5 and 10 min
in molar ratios with starting material (SM), product, and
byproduct. SM was consumed within 30 min (entries 1−3) and
the amount of byproduct increased with time (entries 1−4).
After 10 min (entry 1) there was 37% of product and 9% of
byproduct, and this ratio changed to 42% product and 35%
byproduct after 20 min (entry 2). With time more SM was
consumed, but the relative amount of desired product did not
improve. For this reason 10 min was set as the optimal reaction
time. Although the NMR data showed no change in chemical
shifts for the desired product once formed (see Supporting
Information), actual ratios of SM/product/byproduct sug-
gested partial instability of 3 (entry 3 vs 4).
Table 4. Radiochemistry of 9 (DesmethylPURO) To Obtain
[11C]3
Both TBS and Boc protecting groups were removed in a
single step with aqueous hydrochloric acid; however, further
NMR experiment was required to confirm the stereochemical
identity and determine stability of formed product 3.
Both commercially available 3 and fully protected inter-
mediate 12 were treated with 2 M aq. DCl in DMF-d7 at 80 °C
for 10 min and then neutralized with aq. NaOD and subjected
to analysis by NMR (Table 3). These reaction conditions were
selected considering two requirements for the choice of solvent.
Methylation step was performed in DMF (9 to 12, Table 1),
and the deprotection under radiochemical conditions was
foreseen without previous removal of DMF from the reaction
mixture. Also, suggested instability of 3 (Table 2) could likely
be due to the presence of protic solvent (MeOD-d4). Analysis
of NMR spectral data revealed 82% intact sample after the
treatment of 3 with acid. Gratifyingly, under the same
hydrolysis conditions 12 afforded 77% desired product identical
to commercially available 3 when comparing corresponding
chemical shifts. The NMR data also indicated presence of
byproduct in 18% and 23% from 3 and 12, respectively (Table
3). This strongly suggests that although the removal of TBS
and Boc protecting groups yielded 3, under acidic conditions 3
was partly chemically unstable. Due to minimally different
chemical shifts among epimers of 318 stereochemical identity of
and the results revealed no significant difference among three
temperatures at 5 min (entries 1−3). Longer reaction time was
proven beneficial (entries 1 vs 4, 2 vs 5), unless temperature
was 80 °C in which case decomposition of both radiolabeled
precursor and formed product was predominant (entries 3 vs
6). The deprotection step was performed under acidic
conditions previously established with nonradioactive 12 (i.e.,
cold) to afford [11C]3 as confirmed by HPLC coinjection with
commercially available 3.
For initial proof of principle, development method based on
[11C]CH3I trapping was sufficient to demonstrate that [11C]3
can be prepared; however, automated radiosynthesis was
C
DOI: 10.1021/acsmedchemlett.6b00093
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
required to enable production of [11C]3 on larger scale. We
selected to use iPHASE module20 with an attached semiprep
HPLC. The iPHASE module contains a small volume (1 mL)
reactor for which reason the radiosynthesis was modified as
follows; i) the volumes of acid and base were reduced while
increasing their concentration (4 M vs previously used 2 M
HCl and NaOH); ii) the reaction temperature was kept at 75
°C for both methylation and deprotection; and iii) the reaction
time was reduced to 5 min for each step. Before injection onto
semiprep HPLC the crude reaction mixture was neutralized
with aq. NaOH to ensure pH was >7. Within 70 min from the
end-of-bombardment (EOB) ca. 350 MBq of [11C]3 was
isolated with 66 GBq/μmol specific activity in 16 6% (n = 3)
decay corrected radiochemical yield and >99% radiochemical
purity at end-of-synthesis (EOS).
ABBREVIATIONS
■
[11C]MET, carbon-11 methionine; PET, positron emission
tomography; [3H]MET, tritiated methionine; [18F]TYR,
fluorine-18 tyrosine; [14C]LEU, carbon-14 leucine; [11C]LEU,
carbon-11 leucine; PURO, puromycin; tRNA, transcription
ribonucleic acid; NMR, nuclear magnetic resonance; DSC,
disuccinimydil; TBS, tert-butyldimethylsilyl; DMF, N,N′-
dimethylformamide; Boc, tert-butyloxycarbonyl; HOBt, hydrox-
ybenzotriazole; EDC, 1-ethyl-3-(3-(dimethylamino)propyl)-
carbodiimide; HPLC, high pressure liquid chromatography;
HCN, hydrogen cyanide; EOB, end-of-bombardment; EOS,
end-of-synthesis; RPS, rate of protein synthesis
REFERENCES
■
In conclusion, we have prepared 9 (desmethylPURO
precursor) in two steps from commercially available materials
in 76% overall yield improving the published synthesis of
similar PURO derivatives. NMR analysis of 3 and fully
protected radiolabeling intermediate 12 showed partial (ca.
80%) chemical stability of 3 and limited evidence provided
support for the lack of epimerization. Optimal conditions were
found for nonradioactive [12C]- and radioactive [11C]-
methylation/deprotection sequence to produce [11C]3 using
[11C]iodomethane. An automated method was then developed
(1) Jager, P. L.; Vaalburg, W.; Pruim, J.; de Vries, E. G.; Langen, K. J.;
Piers, D. a. Radiolabeled Amino Acids: Basic Aspects and Clinical
Applications in Oncology. J. Nucl. Med. 2001, 42 (3), 432−445.
(2) Klasner, B. D.; Krause, B. J.; Beer, A. J.; Drzezga, A. PET Imaging
̈
of Gliomas Using Novel Tracers: A Sleeping Beauty Waiting to Be
Kissed. Expert Rev. Anticancer Ther. 2010, 10 (5), 609−613.
(3) Shahbazian, F. M.; Jacobs, M.; Lajtha, A. Rates of Protein
Synthesis in Brain and Other Organs. Int. J. Dev. Neurosci. 1987, 5 (1),
39−42.
(4) Ishiwata, K.; Kubota, K.; Murakami, M.; Kubota, R.; Sasaki, T.;
Ishii, S.; Senda, M. Re-Evaluation of Amino Acid PET Studies: Can the
Protein Synthesis Rates in Brain and Tumor Tissues Be Measured in
Vivo? J. Nucl. Med. 1993, 34 (11), 1936−1943.
(5) Hawkins, R. a; Huang, S. C.; Barrio, J. R.; Keen, R. E.; Feng, D.;
Mazziotta, J. C.; Phelps, M. E. Estimation of Local Cerebral Protein
Synthesis Rates with L-[11C]leucine and PET: Methods, Model, and
Results in Animals and Humans. J. Cereb. Blood Flow Metab. 1989, 9,
446−460.
(6) Tomasi, G.; Bertoldo, A.; Bishu, S.; Unterman, A.; Smith, C. B.;
Schmidt, K. C. Voxel-Based Estimation of Kinetic Model Parameters
of the L-[1-11C]leucine PET Method for Determination of Regional
Rates of Cerebral Protein Synthesis: Validation and Comparison with
Region-of-Interest-Based Methods. J. Cereb. Blood Flow Metab. 2009,
29 (7), 1317−1331.
(7) Veronese, M.; Schmidt, K. C.; Smith, C. B.; Bertoldo, A. Use of
Spectral Analysis with Iterative Filter for Voxelwise Determination of
Regional Rates of Cerebral Protein Synthesis with L-[1-11C]leucine
PET. J. Cereb. Blood Flow Metab. 2012, 32 (6), 1073−1085.
(8) Barrio, J. R.; Keen, R. E.; Ropchan, J. R.; Macdonald, N. S.;
Baumgartner, F. J.; Padgett, H. C.; Phelps, M. E. L-[11C] Leucine:
Routine Synthesis by Enzymatic Resolution L-[11C] Leucine: Routine
Synthesis by Enzymatic Resolution. J. Nucl. Med. 1983, 24 (6), 515−
521.
to enable preparation of [11C]3 in 16
6% (n = 3) decay
corrected radiochemical yield with high radiochemical purity
(>99%) and specific activity of 66 GBq/μmol at EOS. The
radiosynthesis of [11C]3 included a two-step radiolabeling
followed by semipreparative HPLC purification and C18 Light
Sep-Pak trapping, and [11C]3 was produced within 70 min from
EOB. In vitro and in vivo studies are ongoing in our group to
evaluate [11C]3 as a potential PET radiotracer for imaging
protein synthesis and will be reported in due course.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acsmedchem-
lett.6b00093.
Full experimental procedures and characterization of all
compounds, results of NMR experiments, and radio and
UV traces for [11C]3 (PDF)
(9) Mu, F.; Mangner, T. J.; Chugani, H. T. Facile Synthesis of L-
[11C] Leucine as a PET Radiotracer for the Measurement of Cerebral
Protein Synthesis. J. Label. Compd. Radiopharm. 2005, 48, S189.
(10) Bishu, S.; Schmidt, K. C.; Burlin, T.; Channing, M.; Conant, S.;
Huang, T.; Liu, Z.; Qin, M.; Unterman, A.; Xia, Z.; Zametkin, A.;
Herscovitch, P.; Smith, C. B. Regional Rates of Cerebral Protein
Synthesis Measured with L-[1-11C]leucine and PET in Conscious,
Young Adult Men: Normal Values, Variability, and Reproducibility. J.
Cereb. Blood Flow Metab. 2008, 28 (8), 1502−1513.
(11) Azzam, M. E.; Algranati, I. D. Mechanism of Puromycin Action:
Fate of Ribosomes after Release of Nascent Protein Chains from
Polysomes. Proc. Natl. Acad. Sci. U. S. A. 1973, 70 (12), 3866−3869.
(12) Starck, S. R.; Roberts, R. W. Puromycin Oligonucleotides Reveal
Steric Restrictions for Ribosome Entry and Multiple Modes of
Translation Inhibition. RNA 2002, 8, 890−903.
AUTHOR INFORMATION
Corresponding Author
*E-mail: sms96@wbic.cam.ac.uk.
Author Contributions
All authors have given approval to the final version of the
manuscript.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors acknowledge Dr. Lindsay McMurray (WBIC) and
Mr. Robert Bielik (CRUK) for technical assistance with the
production of [11C]CH3I and working with iPHASE as well as
many useful discussions. Dr. Mark A. Sephton (Discovery from
Charles River, UK) is acknowledged for proofreading the
manuscript. NMSF (EPSRC) at the University of Swansea is
acknowledged for running HRMS data.
(13) Miyamoto-Sato, E.; Nemoto, N.; Kobayashi, K.; Yanagawa, H.
Specific Bonding of Puromycin to Full-Length Protein at the C-
Terminus. Nucleic Acids Res. 2000, 28 (5), 1176−1182.
(14) Eckermann, D. J.; Greenwell, P.; Symons, R. H. Peptide-Bond
Formation on the Ribosome. A Comparison of the Acceptor-Substrate
D
DOI: 10.1021/acsmedchemlett.6b00093
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Specificity of Peptidyl Transferase in Bacterial and Mammalian
Ribosomes Using Puromycin Analogues. Eur. J. Biochem. 1974, 41
(3), 547−554.
(15) Eigner, S.; Vera, D. R. B.; Fellner, M.; Loktionova, N. S.; Piel,
M.; Lebeda, O.; Rosch, F.; Roß, T. L.; Henke, K. E. Imaging of Protein
̈
Synthesis: In Vitro and in Vivo Evaluation of 44Sc-DOTA-Puromycin.
Mol. Imaging Biol. 2013, 15 (1), 79−86.
(16) Liu, J.; Xu, Y.; Stoleru, D.; Salic, A. Imaging Protein Synthesis in
Cells and Tissues with an Alkyne Analog of Puromycin. Proc. Natl.
Acad. Sci. U. S. A. 2012, 109 (2), 413−418.
(17) Lee, H.; Fong, K. L.; Vince, R. Puromycin Analogues. Effect of
Aryl-Substituted Puromycin Analogues on the Ribosomal Peptidyl-
transferase Reaction. J. Med. Chem. 1981, 24 (3), 304−308.
(18) Starck, S. R.; Qi, X.; Olsen, B. N.; Roberts, R. W. The
Puromycin Route to Assess Stereo- and Regiochemical Constraints on
Peptide Bond Formation in Eukaryotic Ribosomes. J. Am. Chem. Soc.
2003, 125 (27), 8090−8091.
(19) Valeur, E.; Bradley, M. Amide Bond Formation: Beyond the
Myth of Coupling Reagents. Chem. Soc. Rev. 2009, 38 (2), 606−631.
(20) Ackermann, U.; Plougastel, L.; Wichmann, C.; Goh, Y. W.;
Yeoh, S. D.; Poniger, S. S.; Tochon-Danguy, H. J.; Scott, A. M. Fully
Automated Synthesis and Coupling of [18F]FBEM to Glutathione
Using the iPHASE FlexLab Module. J. Labelled Compd. Radiopharm.
2014, 57 (2), 115−120.
E
DOI: 10.1021/acsmedchemlett.6b00093
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX