Importance of Macromolecules for Quantification of Full Neurochemical Profile & GABA Editing
Lijing Xin1
1Centre d'Imagerie BioMédicale (CIBM) Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland

Synopsis

Mobile macromolecules (MM) present as broad resonances underlying sharp metabolite resonances in 1H MR spectra at short to moderate TEs. The accurate estimation of MM is an important prerequisite for reliable quantification of metabolites. This lecture covers up-to-date knowledge about MM, how to handle MM for MRS quantification, and also some open questions.

Background

Performing localized 1H MRS measurement at short TE allows us to attain high SNR with minimal signal loss from T2 relaxation and J-modulation of coupled metabolites. We can therefore achieve neurochemical profiling including a large number of metabolites. In short to moderate TE 1H MR spectra, broad resonances are observed underlying sharp metabolite resonances, which arise from mobile macromolecules (MM). These broad MM resonances demonstrate short T1 and T2 relaxation properties, and low apparent diffusion coefficients. They are likely composed of overlapping spectra (J-coupled multiplets) from various amino acids within different cytosolic proteins(1). Since an inaccurate estimation of MM spectrum may lead to potential systematic errors in quantification of neurochemical profile, the contribution of this broad MM spectrum should be carefully handled.

Impact of TE, B0, brain tissue, age and disease on MM

MM spectrum is constituted of signals of amino acids that are J-coupled spin systems, of which spectral pattern evolves with pulse sequence parameters, such as TE and TM(STEAM). In this case, MM spectrum measured for short TE spectra would not be the correct prior-knowledge for spectra acquired at other TEs. Thus, reliable MRS quantification requires to take into account a TE/TM specific MM spectrum.

The spectral linewidth of individual MM resonances is determined by multiple factors, such as its T2, macroscopic and microscopic B0 inhomogeneities, multiplicity pattern of amino acids, relative shifts of chemical shifts of amino acids within different proteins. Like other J-coupled metabolites, with the increase of B0, spectral resolution could be improved for multiplicity of amino acids, which has been observed for spectral regions of 1-1.8ppm when increasing B0 from 1.5 to 7T(2).

One may pose the question that whether MM differs within various brain regions and tissues (white matter and grey matter). Several studies in human brain attempted to investigate this question. In (3-5), authors found that some MM components differ between white matter and grey matter in human brain. For rodents, only minor regional and species (rat vs mouse) differences have been found, which may due to the fact that rodent brains contain mainly grey matter. In addition, age dependence was observed for MM in human brain(6) and rodent brain(7), which could be ascribed to the age-related tissue content change.

Moreover, MM itself may serve as a potential biomarker for certain pathologies. Studies in brain tumors, stroke and multiple sclerosis have observed alterations in MM but dominantly in mobile lipid, which can contribute to brain tumor classification. Future studies are encouraged towards the understanding of the origin of these changes in MM components and its role in disease pathophysiology.

Taken together, many factors could affect the amplitude or pattern of MM spectrum. Therefore, a comprehensive characterization of MM with the sequence parameters used at the respective B0 and pathological state is an important prerequisite for metabolite quantification and the discovery of potential disease biomarker.

MM estimation for MRS quantification

Two approaches are generally used:
1. Measuring MM spectrum in vivo and incorporating it in the fitting prior-knowledge.
The measurement of metabolite-nulled MM spectra in vivo relies on distinct characteristics of MM from metabolites, i.e. short T1 relaxation time and low apparent diffusion coefficient. Metabolite-nulling by inversion recovery(IR), either single IR or double IR, is generally used for MM measurement with the following key steps:
  • Vary a range of TIs to determine the optimal TI with minimal metabolite residuals.
  • Acquire spectra at this TI with a longer TE to determine the residual metabolites.
  • Use a short TR(in single IR experiment): improve temporal SNR and further saturate metabolite signals.
  • Measurement within a relatively small VOI to insure the narrowest linewidth possible.
  • Optimal water and lipid suppression to avoid baseline distortion.
  • Post-processing: frequency/phase alignment, removal of residual metabolites (e.g. jMRUI: HLSVD, AMARES), the procedure of residual removal should be proceed carefully to avoid introducing variability in MM.
Moreover, when sufficient gradient strength is available, diffusion weighted MRS is an alternative method to measure MM in combination with IR(8). A better metabolites attenuation can be achieved with diffusion MRS which does not require further post-processing for residual metabolites removal.

2. Mathematical estimation of MM spectrum
MM spectrum can be parameterized by splines, or other lineshape functions such as lorentzian, gaussian and voigt. However, overparameterization should be taken care of. More importantly, such an approach that lack of physical/chemical basis could not predict the modulation of MM spectral pattern with TE or other parameters.
The direct simulation of MM spectrum is not applicable at present, as the number of amino acids and their chemical-environmental information within different proteins are still not completely available. Pioneer work from Behar et al. identified chemical shifts and J-coupling constants of some amino acids using 2D-COSY MRS in dialyzed brain cytosol(1). Recently, Borbáth et al. attempted to use chemical information provided by Biological Magnetic Resonance Data Bank to generate chemical shift histograms as amino acid basis sets(9). Future work towards this direction may provide a physical estimation of MM spectrum.

Co-edited MM in GABA spectral editing

Gamma-aminobutyric acid(GABA) detection is hampered by its low abundance and spectral overlap with other intensive metabolites and MM. To achieve distinct measurement, spectral editing can be used including J-difference editing (e.g. MEGA) and multiple quantum filters. MEGA is the most common editing method used for GABA editing. Due to limited spectral dispersion at B0<7T, when applying editing pulse at 1.9ppm, the MM resonance at 1.7 ppm is often affected by this editing pulse and its coupling partner at 3ppm gets refocused, and thus co-edited with GABA leading to a GABA+ (GABA+MM) measurement at 3ppm. The potential alterations in MM such as with different tissue composition, diseases or ages, would lead to quantification issues for GABA.

MM suppression scheme was proposed, one may apply editing pulse symmetrical around 1.7ppm (i.e. 1.9 and 1.5ppm)(10). Note that the suppression efficiency is very sensitive to frequency drift during the acquisition. Another way to suppress co-edited MM is to use inversion recovery, while this requires the knowledge of T1 relaxation and suffers simultaneous GABA signal attenuation. Lastly, one can use extend TE allowing to use a highly selective editing pulse(11).

Acknowledgements

No acknowledgement found.

References

  1. Behar KL, Ogino T. Characterization of macromolecule resonances in the 1H NMR spectrum of rat brain. Magn Reson Med. 1993; 30(1):38-44.
  2. Cudalbu C, Behar KL, Pallab KB. Wolfgang B. et al. Contribution of macromolecules to brain 1H MR spectra: Experts' consensus recommendations. NMR Biomed. 2020; in press.
  3. Považan M, Strasser B, Hangel G, et al. Simultaneous mapping of metabolites and individual macromolecular components via ultra-short acquisition delay 1 H MRSI in the brain at 7T. Magn Reson Med. 2017;79(3):1231-1240.
  4. Schaller B, Xin L, Gruetter R. Is the macromolecule signal tissue-specific in healthy human brain? a 1H MRS study at 7 tesla in the occipital lobe. Magn Reson Med.2014;72(4):934-940.
  5. Hofmann L, Slotboom J, Boesch C, Kreis R. Characterization of the macromolecule baseline in localized 1H-MR spectra of human brain. Magn Reson Med. 2001;46(5):855-863.
  6. Marjańska M, Deelchand DK, Hodges JS, et al. Altered macromolecular pattern and content in the aging human brain. NMR Biomed. 2018;31(2):1-8.
  7. Tkac I, Rao R, Georgieff MK, Gruetter R. Developmental and regional changes in the neurochemical profile of the rat brain determined by in vivo 1H NMR spectroscopy. Magn Reson Med. 2003;50(1):24-32.
  8. Kunz N, Cudalbu C, Mlynarik V, Hüppi PS, Sizonenko SV, Gruetter R. Diffusion weighted spectroscopy: A novel approach to determine macromolecule resonances in short-echo time 1H-MRS. Magn Reson Med. 2010;64(4).
  9. Borbáth T, Murali-Manohar S, Henning A. Towards a Fitting Model of Macromolecular Spectra: Amino Acids. In: ISMRM 2019: #1068.
  10. Henry PG, Dautry C, Hantraye P, Bloch G. Brain GABA editing without macromolecule contamination. Magn Reson Med. 2001;45(3):517-20.
  11. Edden RA, Puts NA, Barker PB. Macromolecule-suppressed GABA-edited magnetic resonance spectroscopy at 3T. Magn Reson Med. 2012;68(3):657-61.
Proc. Intl. Soc. Mag. Reson. Med. 28 (2020)