Fig 1.
The choline-acetylcholine cycle.
The enzyme choline acetyltransferase builds acetylcholine (ACH) from choline (CHO) and acetyl-CoA. After ACH is released in the synaptic cleft, the enzyme acetylcholinesterase converts ACH into the inactive metabolites CHO and acetate. After re-uptake into the pre-synaptic terminal, free CHO is phosphorylated into phosphocholine (PHC), a reaction catalysed by choline kinase. PHC is available to mobilise CHO for further ACH production via phospholipase C. CHO is also bound in the cell membrane as phosphatidylcholine (PYC). PHC can be then converted to CDP-choline by CTP:phosphocholinecytidyltransferase. The enzyme CDP-choline:1,2-diacylglycerol cholinephosphotransferase then converts the CDP-choline into phosphatidylcholine (PYC) [20]. PYC can also be broken down (via phospholipase C) into glycerophosphocholine (GPC), phosphocholine (PHC), and finally CHO (and side products), or (via phospholipase D) directly into free CHO [19]. In the chemical spectra acquired with magnetic resonance spectroscopy (MRS) CHO, GPC and PHC are the only “visible” metabolites of the CHO cycle.
Fig 2.
Overview of trial and MRS acquisition timing.
Trial timing (top): Each trial had a total length of 5s and begun with a short onset jitter of 0–0.29 s, followed by a cue (0.5s) and a jittered inter stimulus interval (ISI) of 0.8 to 1.3s. The stimuli were presented for 0.5s followed by a maximum response interval of 2.2s and an inter trial interval (ITI) of 0.2 to 0.5s. MRS acquisition (bottom): We collected two MRS acquisitions for each trial (2.5s)—an early epoch, covering the cue, the attention shift phase and part of the stimuli, and a late epoch covering the rest of the trial.
Fig 3.
Sum of MRS voxels over all subjects.
For better visualization the MRS voxel masks were transformed to MNI space. All subjects contributed to coordinates with the highest overlap, including the average center coordinate (MNI -18, -72, 42) (yellow color).
Fig 4.
Choline-containing metabolite spectrum.
Spectral patterns for the metabolites CHO, PHC and GPC simulated using VeSPA (https://scion.duhs.duke.edu/vespa/project). Metabolites were simulated at a field strength of 3T (main field 123.25MHz) using a PRESS pulse sequence (TE1 = 20ms, TE2 = 10ms).
Fig 5.
Changes in the concentration of choline between the early and late epochs (early-late) in the different experimental conditions (Ipsi: MRS acquisition ipsilateral to the shift; Neutral: no attention shift (free choice) trial; Contra: MRS acquisition contralateral to the attention shift).Choline increased only after contralateral visuospatial attention shifts, and there was a statistically significant interaction between condition and epoch. Error bars indicate the standard error.
Fig 6.
Reference metabolites and functional control.
There were no effects of condition (ipsilateral, neutral or contralateral) or epoch (attention or baseline), and no condition by epoch interaction on reference and control metabolites: CRE (left), NAA (middle) and GLU (right).
Fig 7.
Choline-containing metabolite amplitude changes in synthetic data.
Amplitudes of choline (CHO, red) and the combined phosphocholine/glycerophosphocholine peak (PHC/GPC, blue) estimated using the MOD_CHOsep (top) and MOD_CHOglobal (bottom) models in synthetic data. Estimated levels of CHO (red) and PHC/GPC (blue) show an increase in line with the simulated levels of CHO and PHC. Also shown are the levels of NAA and CRE and GPC (right panel). Note that the bottom panel shows combined amplitudes for choline-containing compounds as the MOD_CHOglobal model does not differentiate between CHO and other choline-containing metabolites.
Fig 8.
Sensitivity of amplitude change detection.
Results of the significance tests between neighboring CHO (red) and PHC (blue) simulated levels over 20 iterations. Bright colors represent differences between the neighboring levels detected as statistically significant after correcting for multiple comparison using FDR (q<0.05). Dark colors represent uncorrected significance and black denotes no significant result.