Spying on parahydrogen-induced polarization transfer using a half-tesla benchtop MRI and hyperpolarized imaging enabled by automation

Nuclear magnetic resonance (NMR) and magnetic resonance imaging (MRI) are powerful techniques limited by low sensitivity. Hyperpolarization methods, like dissolution dynamic nuclear polarization (dDNP), significantly enhance NMR signals, enabling real-time in vivo metabolic imaging. However, dDNP requires expensive equipment. Parahydrogen-induced polarization (PHIP) offers an alternative, but its translation to routine use is hindered by the lack of portable, cost-effective polarizers. This study addresses this limitation by developing and characterizing a portable, automated PHIP polarizer using a readily available, low-field (0.5 T) benchtop MRI system. The use of permanent magnets in the MRI system minimizes cost and maintenance compared to high-field superconducting magnets. Automation simplifies the workflow and enhances reproducibility, crucial steps for wider adoption of PHIP in various fields, including biomedical applications where real-time metabolic monitoring is essential for improving diagnostics and treatment. The authors aim to demonstrate the potential of this compact and automated polarizer to facilitate further translation of hyperpolarized MRI towards in vivo applications.

PHIP, using parahydrogen (pH2) as a source of nuclear spin order, has been successfully applied to various systems, including metal-organic complexes, catalysis, and in vivo angiography. PHIP sidearm hydrogenation (PHIP-SAH), with pyruvate precursors, has advanced biomedical applications. Automation in NMR/MRI experiments, encompassing aspects like magnetic field cycling, laser irradiation, and chemical circulation, improves reproducibility and workflow efficiency. Existing PHIP polarizers range from simple shake-and-run methods to complex, stand-alone devices. Early PHIP polarizers for MRI contrast agents utilized strong spin-spin interactions at low magnetic fields (<0.1 mT) for polarization transfer. Later developments introduced spin order transfer (SOT) pulse sequences and low-field polarizers, achieving increased polarization. However, limitations exist, particularly concerning polarization transfer efficiency due to strong indirect interactions among nuclei. High-field systems using RF-induced SOT have yielded high polarizations but necessitate expensive hardware. The intermediate field (1 mT to 1 T) is theoretically optimal for polarization, which can be conveniently achieved using portable permanent magnets, offering a cost-effective and low-maintenance solution. Previous polarizers used in MRI systems, like SAMBADENA, presented challenges in terms of limited sample volume and material compatibility. This study aims to overcome these limitations by creating a more efficient and accessible method.

The researchers constructed an automated PHIP polarizer around a commercially available 0.55 T benchtop teaching MRI system with a 10 mm bore. The system incorporated a gas-liquid control system managing seven solenoid gas valves, two HPLC valves, a syringe pump, and an RF unit. This setup enabled automated sample loading, hyperpolarization, quantification, and disposal. The system's design allowed for multiple experiments without human intervention, facilitating parameter variations. A custom-built gas-liquid path facilitated automatic sample handling and processing. To exploit the 1H polarization generated by pH2, an out-of-phase spin echo (OPE) sequence was implemented, converting anti-phase PHIP signals into in-phase peaks to enhance the signal-to-noise ratio (SNR). A modified OPE sequence enabled real-time observation of spin evolution during SOT without significant sample consumption. The hydrogenation of vinyl acetate (VA) and vinyl pyruvate (VP) was studied using this approach. For 13C hyperpolarization, ESOTHERIC (efficient spin order transfer to heteronuclei via relayed inept chains) SOT sequences were employed to hyperpolarize 1-13C-ethyl pyruvate-d6 (1-13C-EP-d6), a metabolic tracer. High-resolution 13C images were obtained in situ and using a 3D printed phantom. The system’s performance was evaluated by assessing the hydrogenation kinetics of VA and VP under various pH2 pressures, quantifying hyperpolarization levels (1H and 13C), and measuring T2 relaxation times. The system's magnetic field stability was also analyzed. Chemical shifts, coupling constants, and hydrogenation kinetics were analyzed using appropriate fitting functions.

The automated 0.55 T PHIP polarizer achieved a high degree of automation with a short duty cycle (≤1 min per sample preparation). The system successfully hyperpolarized ethyl acetate-d6 and ethyl pyruvate-d6 to 14.4% and 16.2% 1H polarization, respectively, and 1-13C-ethyl pyruvate-d6 to 7% 13C polarization. The use of the OPE sequence significantly improved the SNR (by a factor of 5–6) by converting anti-phase signals into in-phase peaks. The modified OPE sequence allowed efficient “spying” on the real-time spin evolution during SOT, substantially reducing sample consumption. The 13C hyperpolarization, achieved using ESOTHERIC SOT, proved sufficient for acquiring high-resolution FLASH images both in situ and after transferring the sample to a 3D-printed phantom. The hydrogenation kinetics of VA and VP, monitored using the hyperpolarized signal, showed that VA hydrogenates faster than VP. An increase in pH2 pressure increased the hydrogenation rate. The T2 relaxation time and magnetic field stability were evaluated demonstrating suitability for efficient polarization transfer despite field inhomogeneity. The system’s performance was robust over a temperature range relevant to room temperature operations.

This study successfully demonstrated a compact, automated PHIP polarizer that overcomes several limitations of existing systems. The low-field benchtop MRI system offers a significant cost and maintenance advantage over high-field NMR/MRI systems while providing sufficient performance for hyperpolarization and imaging. The achieved hyperpolarization levels, although not the highest reported in literature, are adequate for various applications. The automated nature of the system, with its rapid duty cycle, significantly increases throughput and allows for efficient parameter optimization studies. The ability to monitor reaction kinetics in real-time is valuable for process optimization. The successful in situ and ex situ imaging showcases the potential of this system for hyperpolarized metabolic MRI studies. While the achieved 13C polarization is lower than that reported by some high-field systems, the advantages in cost, portability, and automation outweigh this difference in many applications. The compact design is especially suitable for preclinical settings.

This work presents a significant advance in PHIP technology by introducing a portable, automated polarizer based on a low-field benchtop MRI. The system demonstrates efficient hyperpolarization and allows real-time monitoring of reaction kinetics and high-resolution imaging. The improvements in automation and the short duty cycle promise to greatly facilitate the use of hyperpolarization techniques in biochemical research and medical applications. Future work could focus on optimizing the hydrogenation reaction to further increase polarization yields, investigating the use of alternative catalysts, and exploring the application of the polarizer in diverse preclinical and clinical settings. Scaling up the reaction chamber volume to accommodate larger sample volumes suitable for human studies is another important avenue for future research.

The main limitation of the current system is the inhomogeneity and instability of the magnetic field inherent to permanent magnet-based systems. This resulted in somewhat lower 13C polarization compared to high-field systems. While the system demonstrated good stability for the SOT experiments within the timescale used, longer experiments might be affected more by field drifts. The reaction volume in the present setup is limited to 300 µL, limiting its applicability to high-throughput studies and studies requiring larger volumes such as those for human applications. Finally, the use of deuterated solvents simplifies the analysis by reducing background signals, but might limit general applicability.