Enhanced neurochemical profile of the rat brain using in vivo ¹H NMR spectroscopy at 16.4 T

Abstract

Single voxel magnetic resonance spectroscopy with ultrashort echo time was implemented at 16.4 T to enhance the neurochemical profile of the rat brain in vivo. A TE of 1.7 msec was achieved by sequence optimization and by using short-duration asymmetric pulses. Macromolecular signal components were parameterized individually and included in the quantitative analysis, replacing the use of a metabolite-nulled spectrum. Because of the high spectral dispersion, several signals close to the water line could be detected, and adjacent peaks could be resolved. All 20 metabolites detected in this study were quantified with Cramé r-Rao lower bounds below 20%, implying reliable quantification accuracy. The signal of acetate was detected for the first time in rat brain in vivo with Cramé rRao lower bounds of 16% and a concentration of 0.52 mmol/g. Localized in vivo 1 H NMR spectroscopy provides a unique opportunity for measuring brain metabolite concentrations noninvasively, thus providing neurochemical information of in vivo processes (1). This capability has been shown to benefit from increasing magnetic field strength because of gains in signal-to-noise ratio (SNR) and chemical shift resolution. These advantages were demonstrated by detecting and quantifying 18 metabolites in the rat brain in vivo with an ultrashort echo time stimulated echo acquisition mode (STEAM) sequence at 9.4 T (2). Subsequently, the detection of ascorbate (Asc) was reported with both ultrashort TE STEAM and a J-difference editing technique (3), also at 9.4 T. Recently, several signals adjacent to the water peak, including N-acetylaspartate (NAA) at 4.38 ppm, glycerophosphorylcholine (GPC) at 4.31 ppm, and phosphorylcholine (PCho) at 4.27 ppm, were resolved with narrow RF bandwidths for water suppression, taking advantage of the increased spectral dispersion at 14.1 T (4).The availability of 16.4 T provides the potential of quantifying additional metabolites not detectable at lower field strength. Acetate (Ace), the anion of a shortchain organic acid, is widely recognized to be a substrate for glial cells (5). With in vivo 1 H NMR spectroscopy, the detection of the Ace methyl resonance at 1.9 ppm has not yet been possible due to its low concentration and virtually complete overlap with the methylene resonances of both g-aminobutyric acid (GABA) at 1.889 ppm and N-acetylaspartylglutamate (NAAG) at 1.9 ppm. Detection of Ace has, therefore, been limited to 1 H-[ 13 C]-NMR spectroscopy (6) and high resolution 1 H-NMR on perchloric acid (PCA) extracts of the rat brain (7).Although ultrashort TE makes it possible to obtain valuable metabolite information, it also leads to larger contributions from signals of macromolecules (MM); assessing this component is a critical factor in accurate metabolite quantification. Three different methods have been used to take account of the MM components in short TE in vivo 1 H NMR spectra. The first approach is to acquire an MM spectrum with an inversion-recovery metabolite-nulled measurement and then use this as a basis...

Metabolite and macromolecule T1 and T2 relaxation times in the rat brain in vivo at 17.2T

“…This difference could be originated from the different handling of the macromolecule baseline among studies. Contrary to the study of Hong et al , we were not able to detect acetate when its model spectrum was added to the LCModel basis‐set. However, we were able to detect other low‐concentration metabolites such as alanine (CRLB = 14%), glycine (CRLB = 16%), and serine (CRLB = 20%).…”

Neuropharmacological and Neurobiological Relevance of In vivo 1H-MRS of GABA and Glutamate for Preclinical Drug Discovery in Mental Disorders

“…Water line widths were at the propitious lower end of the 8-15 Hz range reported previously (Mlynarik et al, 2008;Tkac et al, 2003). The normalized SNR (normalized to volume and number of transients acquired but uncorrected for field strength) in the striatum and the cortex reached favorable values of 32 (cm À 3 Â transient À 1/2 ) and 43 (cm À 3 Â transient À 1/2 ), thus approaching the SNR of B42 (cm À 3 Â transient À 1/2 ) reported for 1 H-MRS in rat brain at the higher field strength of 16.4 T (Hong et al, 2011b). CRLBs for GABA, glutamate, and glutamine quantification were at the favorable very low end of the few values disclosed in previous preclinical studies (Cudalbu et al, 2009;Hong et al, 2011a;Pfeuffer et al, 1999;Tkac et al, 2004).…”

Rat strain‐dependent variations in brain metabolites detected by in vivo1H NMR spectroscopy at 16.4T

“…Quantitative analysis was performed with LCModel , using the basis set described in detail previously . Briefly, 20 numerically calculated metabolite spectra were generated by a density matrix computation program, containing the following metabolites: acetate (Ace), Ala, aspartate (Asp), Asc, creatine (Cr), γ‐aminobutyric acid (GABA), Glc, glutamine (Gln), glutamate (Glu), Glycine (Gly), glutathione (GSH), glycerophosphorylcholine (GPC), phosphorylcholine (PCh), myo‐inositol (mI), lactate (Lac), NAA, NAAG, phosphocreatine (PCr), phosphorylethanolamine (PE) and taurine (Tau).…”

“…A comparative study between 4 T and 7 T reported an SNR increase by a factor of two and a 14% increase in spectral resolution, demonstrating comparable quantification results at 7 T with those at 4 T for a 16‐fold decreased scan time . With a field strength of 9.4 T, it was possible to accurately quantify 18 metabolites in rat brain , whereas recent experiments at 16.4 T obtained quantitative values for 20 metabolites, including alanine (Ala), ascorbate (Asc), glucose (Glc) and N ‐acetylaspartylglutamate (NAAG) with Cramér–Rao lower bounds (CRLB) below 20%, indicating good agreement between fitting results and the data .…”