Synopsis

Keywords: Contrast mechanisms: Spectroscopy, Neuro: Brain function, Contrast mechanisms: fMRI

fMRS acquisitions are gradually becoming complementary to fMRI acquisitions providing crucial insights in various functions of the normal and altered human brain. I will summarize the main findings on functional magnetic resonance spectroscopy (fMRS) from its birth until today and demonstrate the power of this technique for investigating the complexities of brain activity showing the importance of the investigations of the neurometabolic properties for a better understanding of the underlying physiological mechanisms through many examples. I will also emphasize the necessity of performing fMRS in rodent models thereby showing the enormous possibilities available from fMRS coupling with optogenetics and chemogenetics.

Syllabus on functional Magnetic Resonance Spectroscopy

The human brain is complex and its exploration during activity remains poorly understood. However, the enormous and crucial continuing developments accomplished over the past 20years in neuroimaging enable us a better understanding of the underpinnings of brain function paving the way to a better assessment of its malfunction. During this week, we learned all about the biophysical origins of the fMRI signals and went from myth to reality about BOLD fMRI responses. By now, we know a lot more about the underpinnings of neurovascular coupling mechanisms. Nevertheless, the underlying processes remain difficult to describe because the brain consists of an enormous number of different cells (neurons and glial cells mainly) that have a significant, specific and sometimes selective impact on brain activity. With this, we touch upon a novel area of research represented by the neurometabolic coupling mechanisms. Catabolic and anabolic pathways ensure that cells release and supply, the necessary energy for synaptic neurotransmission, neurotransmitter cycling or energy storage, respectively. Within this framework, understanding the mechanisms of glucose metabolism is of paramount importance as it is established that it is the major energy source for the brain, fulfilling many critical functions such as ATP production, synthesis of neurotransmitters and neuromodulators, oxidative stress etc…Although glucose is the main fuel oxidized by neurons, other fuels such as lactate have been proposed through the Astrocyte-Neuron Lactate Shuttle hypothesis (ANLSH) thereby explaining some uncoupling processes between the cerebral metabolic rate of glucose (CMRGlc) and the cerebral metabolic rate of oxygen (CMRO2). Despite many trials in favor or against this theory for the past 30 years, the topic remains highly controversial with valuable arguments in support or against it on both sides. In this context, functional Magnetic Resonance Spectroscopy (fMRS), which early steps were performed in the early 90s, has seen increasing interests across the past 10 years becoming in some cases a complementary technique to BOLD fMRI in the human brain. fMRS enables the measurement and quantification of neurochemical changes induced by localized brain activation. Early studies both in the human and rodent somatosensory cortex supported the ANLSH theory. However, fMRS has since demonstrated to be a powerful technique to investigate many other aspects of neurometabolic coupling mechanisms, including the impact of functional metabolic changes in cognition and neurodegenerative and neuropsychiatric diseases. The initial in-vivo and non-invasive measurements of Mangia and colleagues (Mangia S, 2006, 2007) in the activated human visual cortex demonstrated for the first time, the possibility to obtain quantitative changes in glutamate, lactate, glucose and aspartate levels at 7T. These measurements were then largely reproduced by several groups and were also extended to other areas of the brain such as the motor cortex. Moreover, at high field strength (>= 7T), the influence of the blood oxygen level dependent (BOLD) T2* changes were detected as a small line-narrowing effect. Although the change in metabolite concentration due to T2*-induced effects is low (less than 1 %), if left uncorrected, it may lead to a high degree of false-discovery rate. Errors may become particularly important when absolute changes as low as 0.2 µmol/g are expected. For an accurate estimation of metabolite concentration changes, it was proposed to correct for the BOLD effects by line -broadening the population-averaged stimulated spectrum to match the linewidth of the corresponding population-averaged REST spectrum. The corrected stimulated and REST spectra were subsequently subtracted resulting in BOLD-free difference spectrum. Positive glutamate and Lac peaks were visually identified and subsequently quantified using a simulated difference basis set within LCModel. This procedure demonstrated reproducible outcome within the human primary visual cortex at a high magnetic field owing to the large SNR available. In rodents, 1H-fMRS studies remain challenging and have shown lower quantitative reproducibility. The methodology for obtaining accurate estimates of metabolite concentration changes during brain activation remains difficult to replicate in animal models with a voxel of interest more than 500 times smaller than in the human brain. The advent of revolutionary techniques such as optogenetics and chemogenetics that can be coupled to 1H-fMRS for the specific and selective stimulation of excitatory or inhibitory cell populations could increase its potential if accurate quantification of metabolic concentration changes can be achieved. While high field strength fMRS is of paramount interest for a better understanding of normal brain activity, interesting fMRS studies have also been conducted at clinical field strengths (3T) allowing the study of normal and impaired functions in patients. These studies focus on the temporal dynamics of glutamate and gamma-aminobutyric acid (GABA) enabling to evaluate the impact of the Excitatory/Inhibitory balance for understanding behaviorally relevant neural outputs. Different studies have notably and reproducibly found positive correlations between BOLD and stimulated levels of glutamate and negative correlations with stimulated GABA levels. Earlier studies also showed negative correlations between resting GABA levels and BOLD responses. In addition, edited fMRS also enabled investigations of GABAergic mechanisms and remains an important driver within the field since GABA is the major inhibitory neurotransmitter in the brain and plays a key role in various functional aspects of the brain such as motor learning. Studies on pain mechanisms have notably largely benefited from the development of GABA fMRS. During the past two years, major advances have also been conducted regarding a better processing of fMRS data. Currently, very few models exist to simulate and adjust to current fMRS findings but many efforts are undergoing paving the way to major breakthroughs in the assessment and understanding of functional metabolism and neurometabolic mechanisms.

Acknowledgements

Nathalie Just receives funding from the Lundbeck Foundation (Experiment grant, grant nr R370-2021-402).