Hyperpolarization of amino acid precursors to neurotransmitters with parahydrogen induced polarization

Article abstract of DOI:10.1039/c3cc40426a

Several important neurotransmitter precursors were hyperpolarized via homogeneous hydrogenation with parahydrogen. Polarization enhancement was achieved for ¹H and ¹³C spins by several orders of magnitude compared to thermal spectra.

Full text of DOI:10.1039/c3cc40426a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Hyperpolarization of amino acid precursors to
neurotransmitters with parahydrogen induced
polarization†
Cite this: Chem. Commun., 2013,
49, 5304
Received 17th January 2013,
Accepted 21st April 2013
Pei Che Soon,z Xiang Xu,z Boyang Zhang,z Francesca Gruppi, James W. Canary*
and Alexej Jerschow*
DOI: 10.1039/c3cc40426a
www.rsc.org/chemcomm
Several important neurotransmitter precursors were hyperpolarized acid products are capable of deacylation (e.g., by aminoacylase I6,7
)
via homogeneous hydrogenation with parahydrogen. Polarization and could be readily transformed into neurotransmitters.8 In
enhancement was achieved for 1H and 13C spins by several orders of addition, such amino acid analogs have been radiolabeled as
magnitude compared to thermal spectra. Such large signal enhance- tracers for tumor imaging using positron emission tomography
ments of these molecules could facilitate neurotransmitter studies. (PET).9,10 These agents therefore could also have the potential to
become dual-modality probes for simultaneous PET-MRI.11
Neurotransmitters are chemical messengers that transmit signals
The hyperpolarization of some amino acids, such as phenyl-
across synapses in the nervous system. These chemicals play an alanine and histidine, has been achieved at moderate enhancement
important role in cognitive processing (in particular dopamine, factors using the SABRE method and detected at low magnetic
glutamate and GABA). It is believed that hypo- or hyperproduction fields.12,13 A preliminary report indicated that g-aminobutyric acid
of certain neurotransmitters is associated with disorders such as (GABA) was hyperpolarized by PHIP,13 which typically provides large
Alzheimer’s and Parkinson’s diseases,1 and schizophrenia,2 enhancements. This protocol is compatible with hydrogenation at
respectively. Limited imaging techniques exist to follow the high magnetic fields (PASADENA approach14). Therefore, we pre-
formation of neurotransmitters from their biological precursors sent here synthetic N-acetyl dehydro-amino acids, which upon
to detect neurotransmitter production.3
hydrogenation, form amino acid neurotransmitter precursors. We
Hyperpolarization techniques such as dynamic nuclear polariza- investigated the polarization enhancement of the hyperpolarized
tion (DNP), optical pumping, and parahydrogen induced polariza- precursors and transferred the hyperpolarization to 13C to explore
tion (PHIP) have gained great interest in the past decade because the possibility of exploiting the longer signal lifetimes.
they can provide sensitivity boosts by several orders of magnitude.
Fig. 1 shows the nine N-acetyl dehydro-amino acids prepared
PHIP is an economical and potentially high-throughput technique by the Erlenmeyer synthesis.15 Parahydrogen was generated using
to yield hyperpolarized substrates, and it can be implemented either the procedure described previously.16 Hydrogenation of N-acetyl
through a hydrogenation reaction using parahydrogen (p-H2) or dehydro-amino acids (Scheme 1) was initiated by bubbling p-H2
through reversible exchange (signal amplification by reversible
exchange, SABRE).4,5 Both approaches can provide hyperpolarized
species that may serve as promising tools for studies of enzymatic
transformations and associated pathology.
Here, we report the hyperpolarization of a collection of neuro-
transmitter precursors using PHIP. We chose to focus on amino
acid biosynthetic precursors of the neurotransmitters dopamine
(phenylalanine, tyrosine, and DOPA), serotonin (tryptophan) and
histamine (histidine). The requisite N-acetyl dehydro-amino acids
are easily prepared and their structures are well suited for
obtaining both 1H and 13C PHIP signals. The hydrogenated amino
Chemistry Department, New York University, New York, NY 10003, USA.
E-mail: alexej.jerschow@nyu.edu, canary@nyu.edu; Tel: +1-212-998-8451
† Electronic supplementary information (ESI) available: General methods, synth-
esis and characterization data of all compounds, and additional PHIP 1H NMR
spectra of 2a–6a and 8a. See DOI: 10.1039/c3cc40426a
Fig. 1 Synthesized N-acetyl dehydro-amino acids.
‡ These authors contributed equally.
c
5304 Chem. Commun., 2013, 49, 5304--5306
This journal is The Royal Society of Chemistry 2013

View Article Online
Communication
ChemComm
Scheme 1 Hydrogenation of N-acetyl dehydro-phenylalanine.
(enriched to 50%) for 15 s into the NMR tube containing the
reaction solution, while catalyzed by [Rh(cod)(+)(diop)]BF4 in
deuterated methanol, except for 6. Compound 6 was hydroge-
nated in deuterated water with sodium dodecylsulfate (SDS)16
because of limited solubility in methanol.
The reaction clearly started at low field, but continued at high
field, as shown using Only Parahydrogen SpectroscopY (OPSY)
experiments17 (Fig. S7, ESI†): the new signal was produced con-
tinuously for more than 60 s at high field. Therefore, the PASA-
DENA mechanism can be assumed for the interpretation of the
observed spectra. The enhancement factors (EFs), calculated by
comparing the hyperpolarized 1H NMR signals to those at thermal
equilibrium, are summarized in Table S1 (ESI†). Compounds 1a–3a
yielded significantly larger EFs compared to 4a–6a. It was observed
that 1–3 hydrogenated faster than 4 and 5 by a factor of ten. This
rate difference may derive from the faster reaction of electron-poor
alkenes.18 A faster reaction rate may translate into a higher EF,
since more hyperpolarized hydrogenation products are produced at
a given time.5 Product 6a has a smaller EF because hydrogenation
is slower in the deuterated water–SDS system19 compared to a
homogeneous system in deuterated methanol.20
Fig. 3 (a) Single-scan 1H NMR spectrum recorded immediately after hydroge-
nation of 8; (b) reference spectrum of a mixture of 8 and 8a after hyperpolariza-
tion decay. ‘‘S’’ indicates the solvent peaks.
The ability to generate large enhancement factors and to use the
polarization in further experiments depends critically on the lifetime of
the polarized states. Deuterating the substrates is a strategy to prolong
signal lifetimes.24,25 In order to minimize the relaxation caused by spin
diffusion, and thereby boosting the enhancement of the hyper-
polarized signals, deuterated substrates 7–9 were prepared. Compound
7 was obtained after hydrolyzing the azlactone that was condensed
from d6-benzaldehyde and d3-N-acetyl glycine in d6-acetic anhydride.
Interestingly, when non-deuterated acetic anhydride was used, the CD3
group of the N-acetyl moiety exchanges with the CH3 group of acetic
anhydride to yield 8. The hydrogenation of 7–9 is stereospecific (Fig. S8,
1
ESI†). The hyperpolarized H NMR signals are shown in Fig. 3a. No
1
The H spectrum of the product, 1a, obtained from hydrogena-
signal is seen at 2.95 ppm, which is consistent with the fact that this
site is deuterated. EFs of 1400 and 2000 (at 11.7 T) were obtained for 7a
and 8a, respectively. These are much higher than the EF of 1a because
deuteration reduces dipolar coupling neighbors. We also observed
longer relaxation times for the protons in the deuterated products: in
8a, T1 of H1* and H2* are 3.3 s and 3.4 s, respectively, while in 1a, T1 of
H1* and H2* are 2.4 s and 1.1 s respectively.
tion of 1 with p-H2 is shown in Fig. 2a. A substantial sensitivity
improvement (EF of 755 at 11.7 T) was achieved for H2* at 3.25 ppm.
The hyperpolarized signal of H1* at 4.85 ppm showed a similar
enhancement. Interestingly, another strong hyperpolarized signal
was observed at 2.95 ppm, arising from H3, the geminal hydrogen on
the b-carbon atom. Evidently, this signal is a result of polarization
transfer from the newly added proton, H2*. The more distant
aromatic protons at around 7.2 ppm were also hyperpolarized, but
at weaker levels. These polarization transfers most likely result from
transfers via the J-coupling network or the nuclear Overhauser effect
(NOE) during or after hydrogenation.21–23
For hyperpolarized contrast agents, it is desirable to transfer the
polarization to carbon nuclei since they typically have much longer
relaxation times than protons (for 9a, T1 of C1 is B12.6 s while T1 of
H1* and H2* are B3 s at 11.7 T). The transfer was initially attempted
via field cycling.26 This approach failed, likely as a result of the weak
1
J-couplings between H and 13C in question (2JCH = 6.6 Hz). This
value would make it necessary to perform adiabatic transport from
zero field through 0.2 mT at a very slow pace, which would
necessitate a high-precision field-cycling apparatus. Furthermore,
the efficiency would be low because of relaxation during the slow
passage, and the transfer would not be compatible with reactions at
high field. Therefore, we chose the PH-INEPT+ pulse sequence27 to
effect the heteronuclear hyperpolarization transfer. The transfer to
13C did not work for non-deuterated compounds, likely as a result of
faster relaxation and/or dilution of polarization over several spins
due to spin diffusion effects. For the deuterated substrates 7–9, by
contrast, large EFs were obtained. When 7 and 8 were hydrogenated
with p-H2, the hyperpolarization was transferred to the adjacent
carbonyl carbon spin, C1, as well as the carbonyl carbon spin of the
N-acetyl moiety, C2, and those of aromatic carbons in the phenyl
Fig. 2 (a) Single-scan 1H NMR spectrum recorded immediately after hydrogena-
tion of 1; (b) reference spectrum of a mixture of 1 and 1a after the hyperpolariza-
tion decay. ‘‘*’’ indicates starting material peaks, ‘‘S’’ indicates solvent peaks.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 5304--5306 5305

View Article Online
ChemComm
Communication
20 mmol for a coil of 25 mm diameter and 30 mm length, hence an
enhancement factor of 100 would be needed to detect this amount
in spectroscopy. Starting from an EF of 1200, after 15 s we would
have an EF of 365. There would likely be further loss in sensitivity
due to shortened T1 in tissue. Therefore, spectroscopy would be
feasible, but imaging would remain relatively challenging.
We are grateful to Dr Jae-Seung Lee for providing 13C-glycine
and Dr Yu-Shin Ding for helpful discussions. This work was
supported by NSF grant CHE-0848234 (to J.W.C.) and NSF grant
CHE-0957586 (to A.J.). Xiang Xu is grateful for the Margaret
and Herman Sokol fellowship. The NMR spectrometer was
purchased with funds from NSF grant CHE-0116222.
Notes and references
Fig. 4 13C NMR spectra of 8: (a) a single-scan with the PH-INEPT+ sequence
recorded immediately after hydrogenation. (b) The thermally polarized reference
spectrum after hydrogenation (8888 scans). ‘‘*’’ indicates starting material peaks,
‘‘S’’ indicates the solvent peak.
1 S.-C. Li and U. Lindenberger, in Cognitive neuroscience of memory, ed.
L. G. N. H. J. Markowitsch, Hogrefe & Huber Publishers, Ashland,
OH, US, 1999, pp. 103–146.
2 J. A. Lieberman, T. S. Stroup and D. O. Perkins, The American
Psychiatric Publishing textbook of schizophrenia, American Psychiatric
Pub., Washington, DC, 2006.
3 S. Guhlke, A. M. Verbruggen and S. Vallabhajosula, in Clin. Nucl.
Med., ed. H. J. Biersack and L. M. Freeman, Springer, Berlin; New
York, 2007, pp. 34–73.
4 R. W. Adams, J. A. Aguilar, K. D. Atkinson, M. J. Cowley, P. I. P.
Elliott, S. B. Duckett, G. G. R. Green, I. G. Khazal, J. Lopez-Serrano
and D. C. Williamson, Science, 2009, 323, 1708–1711.
5 S. B. Duckett and R. E. Mewis, Acc. Chem. Res., 2012, 45, 1247–1257.
6 Y. Endo, Biochim. Biophys. Acta, Enzymol., 1978, 523, 207–214.
7 S. M. Birnbaum, L. Levintow, R. B. Kingsley and J. P. Greenstein,
J. Biol. Chem., 1952, 194, 455–470.
8 C. K. Mathews and K. E. Van Holde, Biochemistry, Benjamin/
Cummings Pub. Co., Redwood City, Calif., 1990.
9 D. Hellwig, M. Menges, G. Schneider, M.-O. Moellers, B. F. Romeike, M. D.
Menger, C.-M. Kirsch and S. Samnick, Nucl. Med. Biol., 2005, 32, 137–145.
10 P. L. Jager, W. Vaalburg, J. Pruim, E. G. E. de Vries, K.-J. Langen and
D. A. Piers, J. Nucl. Med., 2001, 42, 432–445.
11 C. Catana, D. Procissi, Y. Wu, M. S. Judenhofer, J. Qi, B. J. Pichler,
R. E. Jacobs and S. R. Cherry, Proc. Natl. Acad. Sci. U. S. A., 2008, 105,
3705–3710.
12 S. Gloggler, R. Muller, J. Colell, M. Emondts, M. Dabrowski, B. Blumich
and S. Appelt, Phys. Chem. Chem. Phys., 2011, 13, 13759–13764.
13 J. Bargon, A. Koch, R. Rizi and J. Schmiedeskamp, Proc. Intl. Soc.
Mag. Reson. Med., Berlin, Germany, 2007.
Fig. 5 13C NMR spectra of 9: (a) a single-scan PH-INEPT+ sequence recorded
immediately after hydrogenation. The enhancement factor is 1237 ꢀ 212 at
11.7 T. (b) The thermally polarized reference spectrum after hydrogenation
(16 scans). ‘‘*’’ indicates the starting material peak.
14 C. R. Bowers and D. P. Weitekamp, J. Am. Chem. Soc., 1987, 109, 5541–5542.
15 H. N. C. Wong, Z. L. Xu, H. M. Chang and C. M. Lee, Synthesis, 1992,
793–797.
16 J. A. Tang, F. Gruppi, R. Fleysher, D. K. Sodickson, J. W. Canary and
A. Jerschow, Chem. Commun., 2011, 47, 958–960.
17 J. A. Aguilar, P. I. P. Elliott, J. Lopez-Serrano, R. W. Adams and
S. B. Duckett, Chem. Commun., 2007, 1183–1185.
ring (Fig. 4). The signal-to-noise ratio of the hyperpolarized spec-
trum was much larger than that of a thermal spectrum obtained
with 8888 scans (Fig. 4b). To maximize the signal for spectroscopic
or imaging applications, the carbonyl carbon was enriched with 13
C
(substrate 9) and an average EF of 1237 ꢀ 212 was achieved. This
corresponds to 13C polarization of 1.3% (at 11.7 T, 298 K). The
hyperpolarized signals are shown in Fig. 5.
18 J. P. Candlin and A. R. Oldham, Discuss. Faraday Soc., 1968, 46, 60–71.
19 G. Oehme, E. Paetzold and R. Selke, J. Mol. Catal., 1992, 71, L1–L5.
20 R. Selke, C. Facklam, H. Foken and D. Heller, Tetrahedron: Asymmetry,
1993, 4, 369–382.
In summary, we demonstrated the hyperpolarization of several
amino acids, known to be biosynthetic precursors to neurotrans-
mitters, via PHIP and transferred the polarization to 13C in deuter-
ated and 13C-enriched precursors (EF of 1200). With these large
signal enhancement factors, this approach could be used to study
in vitro enzymatic transformations involved in neurotransmitter
biosynthesis. We are looking into adapting the production of
hyperpolarized molecules in biocompatible media. Although signal
lifetimes continue to be a limitation for many potential applications,
it has been reported, for example, that L-DOPA is taken up in the
brain and converted to dopamine within 15 s of injection.28 The
amount of dopamine that can be infused into the mouse brain is
about 200 nmol,29 and the single-scan detection limit of 13C is about
21 L. Kuhn and J. Bargon, in In situ NMR Methods in Catalysis, ed.
J. Bargon and L. Kuhn, Springer, Berlin, Heidelberg, 2007, pp. 25–68.
22 S. Aime, D. Canet, W. Dastru`\, R. Gobetto, F. Reineri and A. Viale,
J. Phys. Chem. A, 2001, 105, 6305–6310.
23 F. Gruppi, X. Xu, B. Zhang, J. A. Tang, A. Jerschow and J. W. Canary,
Angew. Chem., Int. Ed., 2012, 51, 11787–11790.
24 H. Allouche-Arnon, M. H. Lerche, M. Karlsson, R. E. Lenkinski and
R. Katz-Brull, Contrast Media Mol. Imaging, 2011, 6, 499–506.
25 H. Allouche-Arnon, T. Wade, L. F. Waldner, V. N. Miller, J. M. Gomori,
R. Katz-Brull and C. A. McKenzie, Contrast Media Mol. Imaging, 2013,
8, 72–82.
´
26 H. Johannesson, O. Axelsson and M. Karlsson, C. R. Phys., 2004, 5,
315–324.
27 M. Haake, J. Natterer and J. Bargon, J. Am. Chem. Soc., 1996, 118,
8688–8691.
28 L. Wade and R. Katzman, Am. J. Physiol., 1975, 228, 352–359.
29 K. Golman and J. S. Petersson, Acad. Radiol., 2006, 13, 932–942.
c
5306 Chem. Commun., 2013, 49, 5304--5306
This journal is The Royal Society of Chemistry 2013