INTRODUCTION
Glutamate (Glu) is the most abundant excitatory neurotransmitter with receptors throughout the brain. Abnormal glutamergic neurotransmission is associated with various neurological and psychiatric disorders, such as dementia, epilepsy, and schizophrenia.¹

Dynamic ¹³C MRS with administration of ¹³C-labeled glucose is a powerful method for measuring neurotransmitter cycling and neuro-energetics.¹ To improve sensitivity and localization, proton-observed, carbon-edited (POCE) MRS can be used in combination with ¹H single-voxel localization (SVL) methods, such as semi-localization by adiabatic selective refocusing (sLASER) or stimulated echo acquisition mode (STEAM).¹ Most studies that have been carried out in humans, used ¹H and ¹³C surface coils, thus limited to one location in the brain, most often the occipital lobe.^2-4 Importantly, a surface coil setup also does not allow state-of-the-art MRI as obtained with standard ¹H head coils.

The aim of this study was to develop a setup for POCE MRS and ¹H MRI at 7T. Here we show this setup for sLASER-POCE and STEAM-POCE MRS in the occipital and frontal lobe.

METHODS
Set-up
A 7T MR system (Philips, Best, NL) was used together with a ¹³C birdcage head coil combined with 8 transmit-receive ¹H dipole antennas and a 32-channel ¹H receive array (Nova Medical, Wilmington, USA; Figure 1). To ensure that the local specific absorption rate (SAR) limits were controlled for the ¹H antennas when using either B₁ focusing on the frontal lobe or global B₁ shimming, electromagnetic simulations (Sim4Life, Zurich MedTech, ZMT) were done using the human model Duke from the virtual family.⁵

Sequences
POCE MRS was performed using sLASER or STEAM SVL (Figure 2). In both sequences, the POCE TE was set to 7.8 ms (tuned to the H4-C4 J-coupling constant of Glu and glutamine (Gln), together termed Glx) and a composite ¹³C 180° block pulse was used for ¹³C inversion, which was set to 32.95 ppm (in between the C4 resonances of Glu and Gln) and switched on and off every other acquisition. The sequences were tested in three healthy volunteers (male; 22-30 years) during a primed [U-¹³C]glucose infusion of 99% for 20 min and 60% for 100 min, with aimed plasma glucose levels of 6-10 mmol/l. sLASER-POCE was tested in one volunteer, STEAM-POCE in two, where in one session voxel placement was alternated between the frontal and occipital lobe. Acquisition parameters were as follows: sLASER-POCE: TE/TR=48/5000ms, NSA 2x32; STEAM-POCE: TE/TM/TR=7.8/30/3000ms, NSA 2x48. VAPOR water suppression (300 Hz) was used in both cases. B₁⁺ and B₀ shimming were performed on a manually drawn region of interest (ROI). No outer volume suppression was used.

Data analysis
Coil combination was performed using a generalized-least-squares algorithm, using the unsuppressed water signal to compute the coil sensitivity of each channel. Spectra were frequency and phase aligned and subtracted pairwise. The [4-¹³C]-Glx-H4 signals in the difference spectra were fitted in MATLAB (MathWorks, USA).

RESULTS
Electromagnetic simulations showed a peak local SAR of 1.94 versus 3.9 W/kg for an input power of 8 x 1 W, when driven in global head shim versus B₁ focusing on the frontal lobe, providing a maximum B₁⁺ of 1.18 mT (center) versus 1.26 mT (frontal) at this power level (Figure 3A). When driving the sLASER-POCE and STEAM-POCE to a maximum of 15 and 16 mT, respectively, while incorporating the added 3-5% of SAR by the ¹³C pulses, these scans remained well within the IEC SAR guidelines⁶ of 10 W/kg local and 3.2 W/kg global SAR at a TR of 5 and 3 s, respectively. The B₁⁺ map measured in vivo after B₁⁺ shimming in an ROI in the frontal lobe was in good agreement with the simulation (Figure 3B).

[4-¹³C]-Glx-H4 signals were detected in both sLASER-POCE and STEAM-POCE difference spectra (Figure 4A). Although SNR of the n-acetylaspartate (NAA) signal in the separate on/off spectra was ~50% lower for STEAM-POCE (taking into account differences in voxel size and acquisition times), the [4-¹³C]-Glx-H4 signals in the difference spectra had similar SNR. Time courses of ¹³C labeling of Glx-H4 showed similar kinetics for POCE-STEAM and POCE-sLASER (Figure 4B). [4-¹³C]-Glx-H4 could be detected in both the frontal and the occipital lobe within a single scan session (Figure 5).

DISCUSSION & CONCLUSION
We have shown that it is feasible to detect [4-¹³C]-Glx-H4 during a [U-¹³C]glucose infusion at 7T, using a ¹³C birdcage head coil combined with 8 transmit-receive ¹H dipole antennas and a 32-channel ¹H receive array, in two different locations in the brain, i.e. frontal and occipital, in one session. Moreover, we have shown that while STEAM voxel localization leads to ~50% signal loss of the uncoupled resonances in the unedited spectra compared to sLASER, SNR of the [4-¹³C]-Glx-H4 signals was similar in both the STEAM-POCE and sLASER-POCE difference spectra. This can be explained by the relatively larger signal loss for the strongly coupled Glx spin systems in sLASER due to the longer TE (48 vs 7.8 ms). On the other hand, chemical shift displacement errors are generally assumed to be smaller for sLASER, reducing the likelihood of artefacts from extracranial lipids. While a shoulder of [4-¹³C]-Gln-H4 could be observed, [4-¹³C]-Glu-H4 and [4-¹³C]-Gln-H4 signals were not clearly separated. Glu/Gln signal separation could be achieved by implementing selective POCE.⁷

Acknowledgements

No acknowledgement found.

References

1. Rothman DL, de Graaf RA, Hyder F, et al. In vivo ¹³C and ¹ H-[¹³ C] MRS studies of neuroenergetics and neurotransmitter cycling, applications to neurological and psychiatric disease and brain cancer. NMR Biomed. 2019 Oct;32(10):e4172

2. Chen W, Zhu X-H, Gruetter R, Seaquist ER, Adriany G, Ugurbil K. Study of tricarboxylic acid cycle flux changes in human visual cortex during hemifield visual stimulation using 1H-{13C} MRS and fMRI. Magn Reson Med. 2001;45:349–355

3. Pan JW, Stein DT, Telang F, et al. Spectroscopic imaging of glutamate C4 turnover in human brain. Magn Reson Med. 2000;44:673–679

4. Rothman DL, Novotny EJ, Shulman GI, Howseman AM, Petroff OA, Mason G, Nixon T, Hanstock CC, Prichard JW, Shulman RG. 1H-[13C] NMR measurements of [4-13C]glutamate turnover in human brain. Proc Natl Acad Sci U S A. 1992;89:9603–9606

5. Christ A, Kainz W, Hahn EG, Honegger K, et al. The Virtual Family--development of surface-based anatomical models of two adults and two children for dosimetric simulations. Phys Med Biol. 2010 Jan 21;55(2):N23-38

6. IEC 60601‐2‐33:2010, Medical electrical equipment—Part 2‐33: Particular requirements for the basic safety and essential performance of magnetic resonance equipment for medical diagnosis, International Standard, Edition 3.0

7. De Feyter HM, Herzog RI, Steensma BR, et al. Selective proton-observed, carbon-edited (selPOCE) MRS method for measurement of glutamate and glutamine ¹³C-labeling in the human frontal cortex. Magn Reson Med. 2018 Jul;80(1):11-20