Magnetic Resonance Spectroscopy and Imaging Studies of Brain Functions

National Institute Of Mental Health

Investigators

Linked publications & trials

Abstract

The goal of this research is to develop advanced magnetic resonance spectroscopy (MRS) and imaging techniques and to apply them and other complementary methods to study brain metabolism, neurotransmission, and enzyme activity. MRS allows in vivo measurement of the neurotransmission of glutamate and GABA, which play important roles in many major psychiatric diseases, including depression and schizophrenia. Magnetic resonance spectroscopy can measure multiple brain chemicals simultaneously. It offers a pathway to characterize metabolic connections or disconnections in vivo in the brain of both healthy individuals and patients with brain disorders. With the commonly used short echo time magnetic resonance spectroscopy techniques signals of metabolites overlap with each other. This overlap is particularly severe at 3 Tesla, the main workhorse scanner for clinical studies. To characterize the magnitude and polarity of metabolite correlations originating from spectral overlap we performed Monte Carlo studies in the absence of any biological correlations. We devised a mathematical model of spectral fitting to predict cross-correlation coefficients based on metabolite basis spectra. We found that metabolite cross-correlation coefficient can be positive or negative, depending on the way their signals overlap. In particular, we found conditions where the correlation coefficient becomes zero or close to zero. The existence of conditions with close to zero correlations due to spectral overlap (e.g., between N-acetylaspartate, a neuronal marker, and glutamate, the major excitatory neurotransmitter and an important metabolite linking carbon and nitrogen metabolism) allows experimental determination of metabolic connection or disconnections in the brain using magnetic resonance spectroscopy. Our results indicate that the confounding correlation between metabolite originating from spectral overlap can be minimized by pulse-sequence design and optimization of data acquisition parameters, even when it is not possible to isolate the magnetic resonance spectroscopy signal of interest spectrally (S. Hong, L. An, and J. Shen, Monte Carlo study of metabolite correlations originating from spectral overlap, J. Magn. Reson., 341:107257, 2022. The elimination or minimization of unwanted correlations due to spectral overlap facilitates the MRS mapping of metabolic connections. Using the high-speed full density matrix simulation algorithm developed by our group (Y. Zhang, L. An, and J. Shen, Fast computation of full density matrix of multispin systems for spatially localized in vivo magnetic resonance spectroscopy, Med. Phys., 44:4169-4178 (2017), we have studied the roles of internal and external strong couplings on the formation of glutamate, glutamine, and glutathione pseudo singlets at 7 Tesla. We found that there exists an optimal placement of the slice-selective refocusing pulses such that the signal intensity of the glutamate, glutamine and glutathione pseudo singlets is maximized almost simultaneously (L. An, J.W. Evans, C. Burton, J.S. Tomar, M. Ferraris Araneta, C.A. Zarate Jr., and J. Shen, Roles of strong scalar couplings in maximizing glutamate, glutamine and glutathione pseudo singlets at 7 Tesla, Front. Phys., 10:927162, 2022). Using this technique, we have measured glutamate, glutamine, and glutathione in the pregenual anterior cingulate cortex of healthy participants. In collaboration with Dr. Carlos Zarate we are also applying this technique to study patients with major depression.