Introduction

Signal amplification by reversible exchange (SABRE) is a promising route to hyperpolarising a wide range of biologically relevant contrast agents for magnetic resonance^1,2. SABRE utilises singlet state para-hydrogen (p-H₂) as the source of polarisation³. Spin polarisation transfers to the target agent via a catalytic and reversible exchange (Figure 1) and importantly does not change the chemical identity of the agent/substrate it hyperpolarizes. Additionally the SABRE process takes just a few seconds and does not require expensive hardware to achieve (like other polarisation methods). SABRE polarization transfer between p-H₂ and the agent occurs through concurrent binding to an iridium catalyst in a solvent such as deuterated alcohols. Up to 50% ¹H polarisation for di‐deuterio‐nicotinate using this method have been reported⁴. However such solvents limit the biocompatibility for in-vivo investigation of SABRE hyperpolarised agents. Here we present recent progress in the development of bio-compatible routes to SABRE hyperpolarisation⁵. This method relies on a biphasic solvent system to separate the hyperpolarised agent into a D₂O layer for subsequent safe injection. As such we have now captured ¹H and ¹³C in-vivo preclinical MR images of this novel hyperpolarisation technology; working towards fast and efficient mapping of real time in-vivo biochemistry for application in cardiology, neurology and oncology.

Methods

All aspects of these methods and their development were performed with UK Home Office approval under the Animals (Scientific Procedures) Act 1986. Female Sprague-Dawley rats (n=12) weighing (250–400 g) were kept in a 12 h dark/light cycle, at a constant temperature of 22 °C, with food and water ad libitum. Prior to surgery, animals were anesthetised with urethane (1.25 g/kg i.p). Rectal temperature was maintained at 37 °C throughout surgical and experimental procedures using a homeothermic blanket (Harvard Apparatus Inc, USA). Animals were artificially ventilated and both left/right hand side femoral arteries and veins cannulated for invasive monitoring of arterial BP/blood gases and intravenous infusion respectively. Phenylephrine was infused to increase BP to normal physiological limits with blood gas measurements taken periodically (pO₂ = 105 mm Hg ± 4; pCO₂ = 38 mm Hg ± 5).

Following surgery animals were inserted into the bore of a 7 Tesla pre-clinical MRI scanner (Bruker BioSpec 70/30, 310mm bore with preinstalled BGA-12s imaging gradients). An 86mm ID volume coil (Quadrature ¹H or dual tuned ¹H/¹³C for ¹H or X-nuclei detection) was used for RF TX/RX. A 1.5 mL sub-cutaneous or 0.4 mL intravenous bolus injection of SABRE hyperpolarised agents (nicotinamide, methyl-nicotinate, thienopyridazine and pyruvate) were administered. These agents were chosen for both ease of SABRE hyperpolarisation (fringe field polarisation transfer fields) and their biochemical/medical importance. Nicotinamide is a precursor for NAD+, an important mediator for cellular metabolism. Methyl-nicotinate is a rubefacient with known vaso-dilatory effects. The thienopyridazine motif has anti-cancer properties, inhibiting IKK⁶. The use of hyperpolarised pyruvate has already been proven (with dynamic nuclear polarisation) as good marker for detecting anaerobic metabolism in cancer⁷.

Subsequent MR measurements utilised either standard (RARE, FISP & FLASH) or spectroscopic (CSI & EPSI) imaging pulse sequences for ¹H and ¹³C nuclei. Functional images were superimposed on high resolution structural scans where appropriate. ¹H imaging was also tested with long duration saturation RF pulses and spectra-spatial RF pulses to null contamination effects from the abundant proton water pool.

Results

We showcase in-vivo ¹H SABRE hyperpolarised subcutaneous images of nicotinamide, methyl-nicotinate (figure 2), thienopyridazine, and ¹³C images of pyruvate. We show that EPSI produces robust spectral responses across all ¹H agents, with T₁ times of ~40s (sequence repetition time ~6s) and reduced under-sampling/alias artefacts when compared to low spatial resolution FID based CSI (sequence repetition time ~1.2mins). However, EPSI measurements require substantial post processing to limit water contamination and ghosting artefacts from reversed echoes off resonance. Data are presented which show that significant volumes of hydrogen remain dissolved, even in water based bio-compatible solvents. We therefore add a sonication step before infusion;however polarisation levels decrease substantially (100 fold - figure 2) during this process. Residual heavy metal catalyst is sequestered from the solution (confirmed with inductively coupled plasma mass spectrometry). I.V injection of SABRE hyperpolarised nicotinamide, following these post polarisation procedures, is shown to have no adverse effects on animal physiology. In separate experiments cardiac, brain and tumour based detection was undertaken.

Conclusion & Discussion

This research demonstrates the recent advances in the in-vivo detection of SABRE hyperpolarised agents; detailing developments to help realise its potential as a valuable weapon for in-vivo metabolic monitoring. We have generated hyperpolarised images for biochemical/medically important nicotinamide, methyl-nicotinate, thienopyridazine and pyruvate. Imaging utilises both ¹H and X-nuclei imaging strategies. Targeted biochemical functionalisation of these agents to map in-vivo metabolism through spectral differentiation is on-going.

Acknowledgements

Financial support from the Wellcome Trust (Grants 092506 and 098335) and the MRC (MR/M008991/1) is gratefully acknowledged by the University of York. We thank the technical staff of the biological services facility (H Daniels & C Turnbull) and chemistry department (Dr R John and Dr V Annis) for their assistance.