4320
Flow attenuation and saturation-recovery to discriminate between intracellular and inflowing species in renal hyperpolarized 13C MRS1Laboratory for Functional and Metabolic Imaging (LIFMET), École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland, 2Department of Radiology, University of Geneva (UNIGE), Geneva, Switzerland, 3Department of Radiology, University of Lausanne (UNIL), Lausanne, Switzerland
Synopsis
Selective attenuation of signal from flowing hyperpolarized 13C spins can be used to enhance the fraction of signal from locally-produced metabolites. Here, we look at the effect of flow attenuation in the rat kidney on pyruvate metabolite ratios using an interleaved pulse-acquire/spin-echo+bipolar gradient acquisition scheme. With flow-sensitizing gradients applied, lactate- and alanine-to-pyruvate ratios increase as anticipated. Unexpectedly, the ratio of the hydrate and keto forms of pyruvate increases as well, suggesting a greater fraction of flowing keto-pyruvate. The increasing lactate signal fraction after saturation indicates that labelled circulating lactate becomes predominant over the course of the acquisition.
Introduction
Assigning a hyperpolarized 13C metabolite or probe signal to a particular compartment provides information on its distribution. This is particularly useful when characterizing novel probes in a complex organ such as the kidney, and can aid in understanding the isotope exchange kinetics between metabolites. Multiple techniques have been developed to attenuate the signal originating from intravascular flowing spins. Among them, the use of bipolar gradients has been shown to effectively suppress the large macrovascular signal pool1,2,3. Similarly, spin-echo (SE) sequences with gradient lobes around the refocusing pulse have been used to attenuate the flowing signal4. The present study aims to investigate the effect of macro/microvascular flow attenuation in the rat kidney using bipolar gradients and [1-13C]pyruvate as a well-established test substrate.Methods
Samples of [1-13C]pyruvic acid doped with 21 mM of OX063 radical were polarized in a custom-built 7 T polarizer at 1 K by shining microwave radiation at 196.59 GHz, 50 mW in power. The polarization buildup was monitored for approximately 2 hours, after which the samples were rapidly dissolved in 5.5 mL of a hot buffer solution (pH = 7.4) and transferred (3 s) to a separator infusion pump located in a 9.4 T/ 31 cm animal scanner. 1.4 mL of the hyperpolarized pyruvic acid solution (c ≈ 2.5 mM) was injected through the femoral vein of Sprague Dawley rats (n = 3) over 9 s. The acquisition was performed by placing a single loop 1H / quadrature 13C surface coil over the left kidney of the animal and was respiration and cardiac gated. An interleaved pulse-acquire (FA = 30°)/spin-echo (FA = 90°, TE = 24 ms) sequence (Fig. 1) was developed to detect the 13C signal without and with flow attenuation respectively. Partial flow suppression was achieved by adding two pairs of bipolar gradients (G = 360 mT/m, δ = 3.47 ms) after each rf-pulse in the SE acquisitions. The repetition time between each scan was ≈ 2.6 s. Blood gases, pH and physiological parameters were monitored shortly after each infusion. For each infusion, the acquired FIDs were fitted with Bayes (Washington University, St. Louis) to determine the spectral peak amplitudes and the ratios between metabolites.Results and discussion
As expected, the application of bipolar gradients predominantly affects the pyruvate signal level, which decreases relative to the metabolites signals. Consequently, at every spin-echo acquisition, both alanine- and lactate-to-pyruvate ratios increase (Fig. 2). Additionally, recovery of the signals after saturation by the spin-echo acquisition provides information on their origin. The alanine is almost completely saturated after the spin-echo scan, as seen in the low alanine-to-pyruvate ratio in the acquisition following every spin-echo (Fig. 3), whereas a lactate signal remains and becomes proportionally larger after each spin-echo. These observations are in line with the assumption that most of the alanine signal is produced in an intracellular compartment and that a significant portion of inflowing spins contributes to the observed lactate signal. The steady increase in lactate-to-pyruvate ratio in between spin-echo acquisitions (Fig. 4) can be ascribed to both pyruvate-to-lactate conversion in the tissue, as well as the increasing enrichment of the circulating lactate pool by the hyperpolarized label5,6. This circulating labelled lactate may also be partly converted to alanine, with the ratio of alanine to the sum of lactate and pyruvate remaining low and varying less through the experiment while the alanine-to-pyruvate ratio progressively increases (Fig. 3). Finally, an increase in the pyruvate hydrate-to-keto ratio is seen in the spin-echo/bipolar gradient scans (Fig. 5). This unanticipated result implies that the two forms are not in equilibrium in all compartments and that the hydrate form is disproportionally represented in the slower-moving or stationary compartments.Conclusion
An increase in the lactate- and alanine-to-pyruvate ratios is consistently observed when bipolar gradients are applied, revealing a significant suppression of the moving spins. The similar effect on hydrated vs keto pyruvate raises further questions of how a disequilibrium of the two forms may arise. Recovery of substrate and metabolite signals after saturation by the spin-echo/bipolar gradient acquisition suggests an increasing metabolic role of circulating labelled lactate as the experiment progresses. Although the gradient parameters used do not completely distinguish signal originating from different kidney compartments (e.g. intracellular vs tubular/microvascular space), they nonetheless provide additional evidence for the main compartmental locations of pyruvate and its metabolites and how they may evolve during the first minute after infusion.Acknowledgements
This work was supported by the Swiss State Secretariat for Education, Research and Innovation (SERI) within the Marie Curie Initial Training Network EUROPOL project (n° SERI: 15.0164) and by the Centre d’Imagerie BioMédicale (CIBM) of the UNIL, UNIGE, HUG, CHUV, EPFL and the Leenards and Jeantet Foundations.References
1. Gordon J. W., Niles D. J., Adamson E. B. et al. Application of flow sensitive gradients for improved measures of metabolism using hyperpolarized 13C MRI. Magn. Reson. Med. 2016;75(3):1242-8
2. Lee H., Lee J., Joe E. et al. Flow-suppressed hyperpolarized 13C chemical shift imaging using velocity-optimized bipolar gradient in mouse liver tumors at 9.4 T. Magn. Reson. Med. 2017;78(5):1674-1682
3. Lau A. Z.,2, Miller J. J., Robson M. D. et al. Cardiac perfusion imaging using hyperpolarized 13C urea using flow sensitizing gradients. Magn. Reson. Med. 2016;75(4):1474-83
4. Kettunen M. I., Kennedy B. W., Hu D. E., Brindle K. M. Spin echo measurements of the extravasation and tumor cell uptake of hyperpolarized [1-13C]lactate and [1-13C]pyruvate. Magn. Reson. Med. 2013;70(5):1200-9
5. Chen A. P., Leung K., Lam W. Design of spectral-spatial outer volume suppression RF pulses for tissue specific metabolic characterization with hyperpolarized 13C pyruvate. J. Magn. Reson. 2009;200(2):344-8
6. Wespi P., Steinhauser J., Kwiatkowski G., Kozerke S. Overestimation of cardiac lactate production caused by liver metabolism of hyperpolarized [1-13C]pyruvate. Magn. Reson. Med. 2018;80(5):1882-1890
Figures
