Introduction

Quantification and imaging of cerebral glucose consumption rate (CMR_glc) and tricarboxylic acid cycle flux (V_TCA) is crucial for understanding neuroenergetics. Recently, we developed a novel Deuterium (²H) MRS (DMRS) approach for simultaneously measuring CMR_glc and V_TCA in rat brains at ultrahigh field (1). This new method has the advantages of eliminating the usage of radioactive tracer employed by ¹⁸FDG-PET imaging for CMR_glc measurement; and of allowing more signal averaging resulting in substantial sensitivity gains due to the much shorter T₁ of deuterated glucose (~0.05 s), when comparing with ¹³C MRS for V_TCA quantification (1, 2). Instead of using a clamp protocol for ¹³C MRS studies to keep constant plasma ¹³C-glucose level during hours of measurement, the DMRS approach utilizes a brief i.v. infusion of deuterated glucose as a substance and tracer (1). To further simplify the protocol, in this work, we aimed to establish a completely noninvasive delivery of the glucose isotope into the brain. By using oral uptake, DMRS detection sensitivity was evaluated and the metabolic rates were quantified and compared with those measured through i.v. infusion in the rat brain.

Method

Sprague Dawley rats were anesthetized by 2% isoflurane. The femoral arteries were catheterized for blood sampling and physiological monitoring. For the rat under infusion protocol, femoral veins were also catheterized for deuterated glucose injection. For the rat using oral uptake, commercial standard chow was replaced by 25% sugar water for overnight. In the morning of experimental day, a feeding tube was inserted from the mouth, through the throat and esophagus into the stomach for delivery of tracer before animal entering magnet. All MR scans were conducted at 16.4 T/26 cm scanner (Varian/VNMRJ) using ¹H/²H surface coil. A single-pulse-acquire sequence was applied to obtain dynamic ²H spectra from the rat brain (1). For each rat, 5~10 min baseline spectra were acquired followed by 2 min i.v. infusion of 1.3 g/kg D-Glucose-6,6-d₂ (d66, Sigma-Aldrich) or a bolus injection of 1.9 g/kg d66 through the feeding tube. Metabolites concentrations including glucose (Glc) and glutamate/glutamine (Glx) were quantified as previously described (1). Exponential (for i.v. infusion) or polynomial (for oral uptake) functions were used to describe the changes of blood glucose level, which served as inputs of a kinetic model (1) for determination of CMR_glc and V_TCA.

Result

Figure 1 showed excellent DMR spectral quality with four well-resolved deuteriated signals (water, Glc, Glx and lactate (Lac)) detected in the rat brain after oral uptake of d66. As shown in Fig. 2A, the dynamics of blood glucose level under the oral uptake protocol was significantly different from that of using i.v. infusion. Although the dose of d66 used for oral uptake was higher (~46%), the peak of blood glucose level appeared later (delayed by ~30 min) and lower (~half) when delivering the tracer into the stomach instead of directly into the veins. This was due to the dilution effects of the orally taken glucose by other tissues/organs (such as liver) prior to entering into arteries, which resulted in a constantly lower blood glucose level during the whole experiment when comparing to i.v. infusion. Similarly, because of this slower and down-regulated blood glucose behavior as an input to the brain, time courses of cerebral Glc and Glx concentrations under oral uptake protocol displayed an altered dynamics (Fig. 2B vs. 2C). As shown in Fig. 2B, excellent model fittings provided reliable estimations of metabolic rates: 0.2 μmol/g/min for CMR_glc and 0.6 μmol/g/min for V_TCA, which were consistent with the values obtained from the brain using i.v. infusion at the same anesthesia level (1).

Discussion & Conclusion

The results indicate that in vivo DMRS approach is robust and reliable for simultaneously assessing CMR_glc and V_TCA using different delivery methods for the glucose isotope, i.e., oral uptake or i.v. infusion. While high dose was applied for oral uptake to compensate the dilution effects from other tissues/organs, the efficiency of deuterated glucose utilization by the brain needs to be improved to further increase the DMRS detection sensitivity. For example, employing somatostatin would be a choice, which inhibits insulin and glucagon secretion to reduce the conversion between the labeled glucose and glycogen in liver (3). In summary, this pilot study demonstrated the feasibility of using oral uptake of deuterated glucose for DMRS to quantitatively determine the metabolic rates in vivo in a completely noninvasive way, which makes this technique highly suitable and promising for translation to patients.

Acknowledgements

NIH Grants: R01 NS41262, NS57560, NS70839, MH111447, R24 MH106049, P41 EB015894, P30 NS076408, S10 RR025031 and Keck foundation.

References

(1) Lu, M. et al. (2017) J Creb Blood Flow Metab., PMID: 28503999. (2) Gruetter, R. et al. (1994) J. Neurochem. 63, 1377-1385. (3) Chen, W. et al. (2001) Magn Reson Med. 45, 349-355.