3703

Cost-efficient hyperpolarization of long-lived nuclear spin states on carbon-13 spin pairs

Thomas Theis¹, Zijian Zhou¹, Jin Yu¹, Johannes Colell¹, and Warren Warren^1,2

¹Department of Chemistry, Duke University, Durham, NC, United States, ²Departments of Physics, Radiology and BME, Duke University, Durham, NC, United States

Synopsis

Current hyperpolarization methods used in preclinical research are limited by high cost (>$2M) and short hyperpolarization lifetimes (<1 min). Here we demonstrate hyperpolarization of long-lived states with inexpensive equipment. Specifically, we use parahydrogen which is simple to produce (<$10k in equipment costs) and transfer its singlet hyperpolarization by non-reactive polarization transfer to long-lived singlet states on 13C spin pairs in molecular substrates. We detail polarization transfer mechanisms and obtain hyperpolarization lifetimes in excess of two minutes. Moving forward, we expect to achieve hyperpolarization lifetimes of hours as has been demonstrated on thermally polarized 13C spin pairs.

INTRODUCTION

Hyperpolarized MRI is a powerful tool to study metabolic conversions in animals and patients.¹ However, current technology is limited by high cost (>$2M) and short signal lifetimes (< 1 min). Therefore, we develop approaches that are cost-efficient and generate long-lasting hyperpolarization. This can be done by using parahydrogen, which can be produced with <$10k in equipment cost, and transfer its hyperpolarization to substrates that support long-lived singlet states.² For this purpose, we recently used SABRE (Signal Amplification by Reversible Exchange)³ and hyperpolarized ¹⁵N₂ spin pairs supporting hyperpolarization with lifetimes in excess of 20 min.⁴ Others recently used SABRE to hyperpolarize long-lived singlet states in ¹H pairs⁵ and observed lifetimes extensions from 10 s to 50 s. Here we show that long-lived singlet states can also be hyperpolarized by SABRE in ¹³C₂ pairs. This is significant because current preclinical and clinical hyperpolarized MRI exclusively uses hyperpolarized carbon-13.

METHODS

We produce parahydrogen by cooling normal hydrogen gas to about 37K in the presence of iron oxide and obtain ~90% enriched parahydrogen. This parahydrogen is bubbled through a solution at room temperature where polarization transfer is performed with a SABRE catalyst as depicted in Fig. 1. Parahydrogen and ¹³C labeled substrate reversibly bind to an iridium center. This establishes the hyperpolarization transfer complex. This complex has an average lifetime of about 50 ms. During this time, hyperpolarization is transferred from parahydrogen to substrate. Many repeated catalytic cycles build up hyperpolarization on the substrate (within ~ 5 min). Most critical to the hyperpolarization transfer is nuclear spin energy mixing of parahydrogen and substrate states. This mixing is established by going to very low magnetic fields where the frequency difference between ¹H and ¹³C becomes comparable to the strength of J-coupling interactions in the system. Therefore, we perform the hyperpolarization process in a magnetically shielded environment established by mu-metal shields providing microTesla fields, about 200 times smaller than the Earth’s field. We introduced this approach as SABRE-SHEATH (SABRE in SHield Enables Alignment Transfer to Heteronuclei), initially focused on ¹⁵N hyperpolarization⁶, and expand it to carbon-13 in the present contribution. After hyperpolarization in the magnetic shield, the samples are transferred to an 8.5 T NMR magnet for detection. To measure decay time constants after hyperpolarization, the samples are kept at an intermediate field for variable time periods and subsequently transferred to the NMR spectrometer for detection.

RESULTS

Figure 2 shows ¹³C SABRE-SHEATH for two compounds, symmetric bipyridyl acetylene and asymmetric pyridyl benzene acetylene. Both are pyridine derivatives that easily bind to the iridium center. We obtain between 40 to 200 fold enhancements over thermal measurements at 8.5 T. ¹³C-polarization is enhanced in samples with naturally abundant ¹³C (1.1%) or in ¹³C enriched samples. For the enriched samples, we specifically chose the acetylene group because those carbons display large mutual J-couplings (of ~180 Hz), ideal to support long-lived singlet states. On the examined compounds two types of hyperpolarization can be induced: magnetization and singlet order. This is illustrated in Fig. 3. We can easily distinguish between these two types of hyperpolarization. Magnetization is associated with in-phase signals as displayed in Fig. 3 A and C, whereas singlet order is associated with anti-phase signals as seen in part B. Also, we can select the hyperpolarization type by setting the magnetic field as indicated in Fig. 3 D-F. At very low fields of about 0.4 microT we obtain magnetization, whereas at slightly higher fields of ~6 microT we obtain primarily singlet order. Importantly, these two types of hyperpolarization are associated with different relaxation rates as displayed in Fig. 4. Magnetization decays with a T₁ time of about 0.2 min (12 s) whereas singlet order can extend the hyperpolarization lifetime up to two minutes.

DISCUSSION

Most importantly, the data shows that SABRE-SHEATH works for hyperpolarizing carbon-13. We can hyperpolarize either magnetization or singlet order by adjusting the magnetic field in the magnetically shielded environment. Singlet order can be highly advantageous because it displays slower relaxation such that the hyperpolarization can be observed on longer timescales; ideal for tracking and imaging of slower processes as is the case for many metabolic transformations.

CONCLUSION

In future research hyperpolarization levels need to be increased and biologically relevant ¹³C labeled substrates need to be developed. Then, the cost-efficient SABRE-SHEATH modality will give researchers on modest budgets access to hyperpolarized probes for studying metabolism by non-invasive molecular imaging. Further down the road, we envision biomolecular imaging using SABRE-SHEATH at the cost of an X-Ray.

Acknowledgements

We would like to thank the NSF (CHE-1363008) for financial support of this research.

References

1. S. J. Nelson, J. Kurhanewicz, D. B. Vigneron, et al. Metabolic Imaging of Patients with Prostate Cancer Using Hyperpolarized 1-C-13 Pyruvate. Sci. Transl. Med. 5, 198ra108 (2013).

2. G. Stevanato, J. T. Hill-Cousins, P. Håkansson, et al. A Nuclear Singlet Lifetime of More than One Hour in Room-Temperature Solution. Angew. Chem. Int. Ed. 54, 3740-3743 (2015).

3. R. W. Adams, J. A. Aguilar, K. D. Atkinson, et al., Reversible interactions with para-hydrogen enhance NMR sensitivity by polarization transfer. Science 323, 1708-1711 (2009).

4. T. Theis, G. X. Ortiz , A. W. J. Logan, et al. Direct and cost-efficient hyperpolarization of long-lived nuclear spin states on universal ¹⁵N₂-diazirine molecular tags. Sci. Adv. 2, e1501438 (2016).

5. S. S. Roy, P. J. Rayner, P. Norcott, et al. Long-lived states to sustain SABRE hyperpolarised magnetisation. Phys Chem Chem Phys 18, 24905-24911 (2016).

6. T. Theis, M. L. Truong, A. M. Coffey et al. Microtesla SABRE Enables 10% Nitrogen-15 Nuclear Spin Polarization. J. Am. Chem. Soc. 137, 1404-1407 (2015).

Figures

Fig. 1. SABRE polarization transfer catalyst. Parahydrogen and substrate reversibly bind to the iridium center. During the lifetime of the polarization transfer catalyst, polarization can be transferred from the parahydrogen derived hydrides to the substrate molecule. Repeated catalytic cycles build up hyperpolarization on the free substrate in solution. In the depicted case carbon-13 spins in the substrate are hyperpolarized.

Fig. 2. ¹³C NMR spectra of SABRE-SHEATH hyperpolarized compounds. A and B show spectra of pyridyl-phenyl-acetylene and bi-pyridyl-acetylene with ¹³C at natural abundance. Only ¹³C sites without attached protons are observed. Likely because ¹³C sites with attached protons relax too fast. C and D show spectra of the same compounds where the acetylenic carbons are enriched with¹³C. For the asymmetric compound detected in C the carbons have a chemical shift difference. Therefore, two doublets are detected, split by the C-C J-coupling. In the symmetric compound the carbons are chemically equivalent and only a single line is observed.

Fig. 3. Singlet and magnetization are selectively hyperpolarized. A) hyperpolarized magnetization on the asymmetric pyridyl-phenyl-acetylene. B) hyperpolarized singlet order on the same molecule. C) magnetization on the symmetric. Because bi-pyridyl-acetylene is symmetric the singlet order remains undetectable with the current experimental protocol. D-F) The corresponding dependence of magnetization and singlet order on the magnetic field in the mu-metal shield. Again, for the symmetric compound singlet order has not been detected.

Fig. 4. Hyperpolarization decay time constant measurements. The decay time constants are measured for magnetization (T₁) and for singlet order (T_S) for the asymmetric pyridyl-phenyl-acetylene. The measurements are performed at low magnetic fields where the frequency difference between the two carbons is small such that the singlet state remains a relaxation protected state. The data is fit with single exponentials to extract the characteristic decay constants.

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)

3703