Magnetic Resonance Spectroscopy Spectral Registration Using Deep Learning

BACKGROUND

Deep learning-based methods have been successfully applied to MRI image registration. However, there is a lack of deep learning-based registration methods for magnetic resonance spectroscopy (MRS) spectral registration (SR).

PURPOSE

To investigate a convolutional neural network-based SR (CNN-SR) approach for simultaneous frequency-and-phase correction (FPC) of single-voxel Meshcher-Garwood point-resolved spectroscopy (MEGA-PRESS) MRS data.

STUDY TYPE

Retrospective.

SUBJECTS

Forty thousand simulated MEGA-PRESS datasets generated from FID Appliance (FID-A) were used and split into the following: 32,000/4000/4000 for training/validation/testing. A 101 MEGA-PRESS medial parietal lobe data retrieved from the Big GABA were used as the in vivo datasets.

FIELD STRENGTH/SEQUENCE

3T, MEGA-PRESS.

ASSESSMENT

Evaluation of frequency and phase offsets mean absolute errors were performed for the simulation dataset. Evaluation of the choline interval variance was performed for the in vivo dataset. The magnitudes of the offsets introduced were -20 to 20 Hz and -90° to 90° and were uniformly distributed for the simulation dataset at different signal-to-noise ratio (SNR) levels. For the in vivo dataset, different additional magnitudes of offsets were introduced: small offsets (0-5 Hz; 0-20°), medium offsets (5-10 Hz; 20-45°), and large offsets (10-20 Hz; 45-90°).

STATISTICAL TESTS

Two-tailed paired t-tests for model performances in the simulation and in vivo datasets were used and a P-value <0.05 was considered statistically significant.

RESULTS

CNN-SR model was capable of correcting frequency offsets (0.014 ± 0.010 Hz at SNR 20 and 0.058 ± 0.050 Hz at SNR 2.5 with line broadening) and phase offsets (0.104 ± 0.076° at SNR 20 and 0.416 ± 0.317° at SNR 2.5 with line broadening). Using in vivo datasets, CNN-SR achieved the best performance without (0.000055 ± 0.000054) and with different magnitudes of additional frequency and phase offsets (i.e., 0.000062 ± 0.000068 at small, -0.000033 ± 0.000023 at medium, 0.000067 ± 0.000102 at large) applied.

DATA CONCLUSION

The proposed CNN-SR method is an efficient and accurate approach for simultaneous FPC of single-voxel MEGA-PRESS MRS data.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: Stage 2.

Brain energy metabolism: A roadmap for future research

Although we have learned much about how the brain fuels its functions over the last decades, there remains much still to discover in an organ that is so complex. This article lays out major gaps in our knowledge of interrelationships between brain metabolism and brain function, including biochemical, cellular, and subcellular aspects of functional metabolism and its imaging in adult brain, as well as during development, aging, and disease. The focus is on unknowns in metabolism of major brain substrates and associated transporters, the roles of insulin and of lipid droplets, the emerging role of metabolism in microglia, mysteries about the major brain cofactor and signaling molecule NAD+ , as well as unsolved problems underlying brain metabolism in pathologies such as traumatic brain injury, epilepsy, and metabolic downregulation during hibernation. It describes our current level of understanding of these facets of brain energy metabolism as well as a roadmap for future research.

The impact of metabolism on the adaptation of organisms to environmental change

Since Jacob and Monod's discovery of the lac operon ∼1960, the explanations offered for most metabolic adaptations have been genetic. The focus has been on the adaptive changes in gene expression that occur, which are often referred to as "metabolic reprogramming." The contributions metabolism makes to adaptation have been largely ignored. Here we point out that metabolic adaptations, including the associated changes in gene expression, are highly dependent on the metabolic state of an organism prior to the environmental change to which it is adapting, and on the plasticity of that state. In support of this hypothesis, we examine the paradigmatic example of a genetically driven adaptation, the adaptation of E. coli to growth on lactose, and the paradigmatic example of a metabolic driven adaptation, the Crabtree effect in yeast. Using a framework based on metabolic control analysis, we have reevaluated what is known about both adaptations, and conclude that knowledge of the metabolic properties of these organisms prior to environmental change is critical for understanding not only how they survive long enough to adapt, but also how the ensuing changes in gene expression occur, and their phenotypes post-adaptation. It would be useful if future explanations for metabolic adaptations acknowledged the contributions made to them by metabolism, and described the complex interplay between metabolic systems and genetic systems that make these adaptations possible.

Neurovascular coupling is optimized to compensate for the increase in proton production from nonoxidative glycolysis and glycogenolysis during brain activation and maintain homeostasis of pH, pCO2 , and pO2

During transient brain activation cerebral blood flow (CBF) increases substantially more than cerebral metabolic rate of oxygen consumption (CMRO2 ) resulting in blood hyperoxygenation, the basis of BOLD fMRI contrast. Explanations for the high CBF vs. CMRO2 slope, termed neurovascular coupling (NVC) constant, focused on maintainenance of tissue oxygenation to support mitochondrial ATP production. However, paradoxically the brain has a 3-fold lower oxygen extraction fraction (OEF) than other organs with high energy requirements, like heart and muscle during exercise. Here, we hypothesize that the NVC constant and the capillary oxygen mass transfer coefficient (which in combination determine OEF) are co-regulated during activation to maintain simultaneous homeostasis of pH and partial pressure of CO2 and O2 (pCO2 and pO2 ). To test our hypothesis, we developed an arteriovenous flux balance model for calculating blood and brain pH, pCO2 , and pO2 as a function of baseline OEF (OEF0 ), CBF, CMRO2 , and proton production by nonoxidative metabolism coupled to ATP hydrolysis. Our model was validated against published brain arteriovenous difference studies and then used to calculate pH, pCO2, and pO2 in activated human cortex from published calibrated fMRI and PET measurements. In agreement with our hypothesis, calculated pH, pCO2, and pO2 remained close to constant independently of CMRO2 in correspondence to experimental measurements of NVC and OEF0 . We also found that the optimum values of the NVC constant and OEF0 that ensure simultaneous homeostasis of pH, pCO2, and pO2 were remarkably similar to their experimental values. Thus, the high NVC constant is overall determined by proton removal by CBF due to increases in nonoxidative glycolysis and glycogenolysis. These findings resolve the paradox of the brain's high CBF yet low OEF during activation, and may contribute to explaining the vulnerability of brain function to reductions in blood flow and capillary density with aging and neurovascular disease.

Measurement of neuro-energetics and neurotransmission in the rat olfactory bulb using 1 H and 1 H-[13 C] NMR spectroscopy

The olfactory bulb (OB) plays a fundamental role in the sense of smell and has been implicated in several pathologies, including Alzheimer's disease. Despite its importance, high metabolic activity and unique laminar architecture, the OB is not frequently studied using MRS methods, likely due to the small size and challenging location. Here we present a detailed metabolic characterization of OB metabolism, in terms of both static metabolite concentrations using 1 H MRS and metabolic fluxes associated with neuro-energetics and neurotransmission by tracing the dynamic 13 C flow from intravenously administered [1,6-13 C2 ]-glucose, [2-13 C]-glucose and [2-13 C]-acetate to downstream metabolites, including [4-13 C]-glutamate, [4-13 C]-glutamine and [2-13 C]-GABA. The unique laminar architecture and associated metabolism of the OB, distinctly different from that of the cerebral cortex, is characterized by elevated GABA and glutamine levels, as well as increased GABAergic and astroglial energy metabolism and neurotransmission. The results show that, despite the technical challenges, high-quality 1 H and 1 H-[13 C] MR spectra can be obtained from the rat OB in vivo. The derived metabolite concentrations and metabolic rates demonstrate a unique metabolic profile for the OB. The metabolic model provides a solid basis for future OB studies on functional activation or pathological conditions.

The early days of ex vivo 1 H, 13 C, and 31 P nuclear magnetic resonance in the laboratory of Dr. Robert G. Shulman from 1975 to 1995

This paper provides a brief description of the early use of ex vivo nuclear magnetic resonance (NMR) studies of tissue and tissue extracts performed in the laboratory of Dr. Robert G. Shulman from 1975 through 1995 at Bell Laboratories, then later at Yale University. During that period, ex vivo NMR provided critical information in support of resonance assignments and the quantitation of concentrations for magnetic resonance spectroscopy studies. The period covered saw rapid advances in magnet technology, starting with studies of microorganisms in vertical bore high-resolution NMR studies, then by 1981 studies of small mammals in a horizontal bore magnet, and then studies of humans in 1984. Ex vivo NMR played a critical role in all these studies. A general strategy developed in the lab for using ex vivo NMR to support in vivo studies is presented, as well as illustrative examples.

A 3D atlas of functional human brain energetic connectome based on neuropil distribution

The human brain is energetically expensive, yet the key factors governing its heterogeneous energy distributions across cortical regions to support its diversity of functions remain unexplored. Here, we built up a 3D digital cortical energy atlas based on the energetic costs of all neuropil activities into a high-resolution stereological map of the human cortex with cellular and synaptic densities derived, respectively, from ex vivo histological staining and in vivo PET imaging. The atlas was validated with PET-measured glucose oxidation at the voxel level. A 3D cortical activity map was calculated to predict the heterogeneous activity rates across all cortical regions, which revealed that resting brain is indeed active with heterogeneous neuronal activity rates averaging around 1.2 Hz, comprising around 70% of the glucose oxidation of the cortex. Additionally, synaptic density dominates spatial patterns of energetics, suggesting that the cortical energetics rely heavily on the distribution of synaptic connections. Recent evidence from functional imaging studies suggests that some cortical areas act as hubs (i.e., interconnecting distinct and functionally active regions). An inverse allometric relationship was observed between hub metabolic rates versus hub volumes. Hubs with smaller volumes have higher synapse density, metabolic rate, and activity rates compared to nonhubs. The open-source BrainEnergyAtlas provides a granular framework for exploring revealing design principles in energy-constrained human cortical circuits across multiple spatial scales.

A tribute to Leif Hertz: The historical context of his pioneering studies of the roles of astrocytes in brain energy metabolism, neurotransmission, cognitive functions, and pharmacology identifies important, unresolved topics for future studies

Leif Hertz, M.D., D.Sc. (honōris causā) (1930-2018) was one of the original and noteworthy participants in the International Conference on Brain Energy Metabolism (ICBEM) series since its inception in 1993. The biennial ICBEM conferences are organized by neuroscientists interested in energetics and metabolism underlying neural functions; they have had a high impact on conceptual and experimental advances in these fields and on promoting collaborative interactions among neuroscientists. Leif made major contributions to ICBEM discussions and understanding of metabolic and signaling characteristics of astrocytes and their roles in brain function. His studies ranged from uptake of K⁺ from extracellular fluid and its stimulation of astrocytic respiration, identification and regulation of enzymes specifically or preferentially expressed in astrocytes in the glutamate-glutamine cycle of excitatory neurotransmission, a requirement for astrocytic glycogenolysis for fueling K⁺ uptake, involvement of glycogen in memory consolidation in the chick, and pharmacology of astrocytes. This tribute to Leif Hertz highlights his major discoveries, the high impact of his work on astrocyte-neuron interactions, and his unparalleled influence on understanding the cellular basis of brain energy metabolism. His work over seven decades has helped integrate the roles of astrocytes into neurotransmission where oxidative and glycogenolytic metabolism during neurotransmitter glutamate turnover are key aspects of astrocytic energetics. Leif recognized that brain astrocytic metabolism is greatly underestimated unless the volume fraction of astrocytes is taken into account. Adjustment for pathway rates expressed per gram tissue for volume fraction indicates that astrocytes have much higher oxidative rates than neurons and astrocytic glycogen concentrations and glycogenolytic rates during sensory stimulation in vivo are similar to those in resting and exercising muscle, respectively. These novel insights are typical of Leif's astute contributions to the energy metabolism field, and his publications have identified unresolved topics that provide the neuroscience community with challenges and opportunities for future research.

Why Hippocampal Glutamate Levels Are Elevated in Schizophrenia

This article discusses why glutamate levels are abnormally elevated in the hippocampus of patients with schizophrenia and related disorders.

In vivo calibration of genetically-encoded metabolite biosensors must account for metabolite metabolism during calibration and cellular volume

Isotopic assays of brain glucose utilization rates have been used for more than four decades to establish relationships between energetics, functional activity, and neurotransmitter cycling. Limitations of these methods include the relatively long time (1-60 min) for determination of labeled metabolite levels and the lack of cellular resolution. Identification and quantification of fuels for neurons and astrocytes that support activation and higher brain functions are a major, unresolved issues. Glycolysis is preferentially upregulated during activation even though oxygen level and supply are adequate, causing lactate concentrations to quickly rise during alerting, sensory processing, cognitive tasks, and memory consolidation. However, the fate of lactate (rapid release from brain or cell-cell shuttling coupled with local oxidation) is long-disputed. Genetically-encoded biosensors can determine intracellular metabolite concentrations and report real-time lactate level responses to sensory, behavioral, and biochemical challenges at the cellular level. Kinetics and time courses of cellular lactate concentration changes are informative, but accurate biosensor calibration is required for quantitative comparisons of lactate levels in astrocytes and neurons. An in vivo calibration procedure for the Laconic lactate biosensor involves intracellular lactate depletion by intravenous pyruvate-mediated trans-acceleration of lactate efflux followed by sensor saturation by intravenous infusion of high doses of lactate plus ammonium chloride. In the present paper, the validity of this procedure is questioned because rapid lactate-pyruvate interconversion in blood, preferential neuronal oxidation of both monocarboxylates, on-going glycolytic metabolism, and cellular volumes were not taken into account. Calibration pitfalls for the Laconic lactate biosensor also apply to other metabolite biosensors that are standardized in vivo by infusion of substrates that can be metabolized in peripheral tissues. We discuss how technical shortcomings negate the conclusion that Laconic sensor calibrations support the existence of an in vivo astrocyte-neuron lactate concentration gradient linked to lactate shuttling from astrocytes to neurons to fuel neuronal activity.

Overexpression of UCP3 decreases mitochondrial efficiency in mouse skeletal muscle in vivo

Uncoupling protein-3 (UCP3) is a mitochondrial transmembrane protein highly expressed in the muscle that has been implicated in regulating the efficiency of mitochondrial oxidative phosphorylation. Increasing UCP3 expression in skeletal muscle enhances proton leak across the inner mitochondrial membrane and increases oxygen consumption in isolated mitochondria, but its precise function in vivo has yet to be fully elucidated. To examine whether muscle-specific overexpression of UCP3 modulates muscle mitochondrial oxidation in vivo, rates of ATP synthesis were assessed by ³¹ P magnetic resonance spectroscopy (MRS), and rates of mitochondrial oxidative metabolism were measured by assessing the rate of [2-¹³ C]acetate incorporation into muscle [4-¹³ C]-, [3-¹³ C]-glutamate, and [4-¹³ C]-glutamine by high-resolution ¹³ C/¹ H MRS. Using this approach, we found that the overexpression of UCP3 in skeletal muscle was accompanied by increased muscle mitochondrial inefficiency in vivo as reflected by a 42% reduction in the ratio of ATP synthesis to mitochondrial oxidation.

Mechanistic stoichiometric relationship between the rates of neurotransmission and neuronal glucose oxidation: Reevaluation of and alternatives to the pseudo-malate-aspartate shuttle model

The ~1:1 stoichiometry between the rates of neuronal glucose oxidation (CMRglc-ox-N ) and glutamate (Glu)/γ-aminobutyric acid (GABA)-glutamine (Gln) neurotransmitter (NT) cycling between neurons and astrocytes (VNTcycle ) has been firmly established. However, the mechanistic basis for this relationship is not fully understood, and this knowledge is critical for the interpretation of metabolic and brain imaging studies in normal and diseased brain. The pseudo-malate-aspartate shuttle (pseudo-MAS) model established the requirement for glycolytic metabolism in cultured glutamatergic neurons to produce NADH that is shuttled into mitochondria to support conversion of extracellular Gln (i.e., astrocyte-derived Gln in vivo) into vesicular neurotransmitter Glu. The evaluation of this model revealed that it could explain half of the 1:1 stoichiometry and it has limitations. Modifications of the pseudo-MAS model were, therefore, devised to address major knowledge gaps, that is, submitochondrial glutaminase location, identities of mitochondrial carriers for Gln and other model components, alternative mechanisms to transaminate α-ketoglutarate to form Glu and shuttle glutamine-derived ammonia while maintaining mass balance. All modified models had a similar 0.5 to 1.0 predicted mechanistic stoichiometry between VNTcycle and the rate of glucose oxidation. Based on studies of brain β-hydroxybutyrate oxidation, about half of CMRglc-ox-N may be linked to glutamatergic neurotransmission and localized in pre-synaptic structures that use pseudo-MAS type mechanisms for Glu-Gln cycling. In contrast, neuronal compartments that do not participate in transmitter cycling may use the MAS to sustain glucose oxidation. The evaluation of subcellular compartmentation of neuronal glucose metabolism in vivo is a critically important topic for future studies to understand glutamatergic and GABAergic neurotransmission.

Rates of pyruvate carboxylase, glutamate and GABA neurotransmitter cycling, and glucose oxidation in multiple brain regions of the awake rat using a combination of [2-13C]/[1-13C]glucose infusion and 1H-[13C]NMR ex vivo

Anaplerosis occurs predominately in astroglia through the action of pyruvate carboxylase (PC). The rate of PC (Vpc) has been reported for cerebral cortex (or whole brain) of awake humans and anesthetized rodents, but regional brain rates remain largely unknown and, hence, were subjected to investigation in the current study. Awake male rats were infused with either [2-¹³C]glucose or [1-¹³C]glucose (n = 27/30) for 8, 15, 30, 60 or 120 min, followed by rapid euthanasia with focused-beam microwave irradiation to the brain. Blood plasma and extracts of cerebellum, hippocampus, striatum, and cerebral cortex were analyzed by ¹H-[¹³C]-NMR to establish ¹³C-enrichment time courses for glutamate-C4,C3,C2, glutamine-C4,C3, GABA-C2,C3,C4 and aspartate-C2,C3. Metabolic rates were determined by fitting a three-compartment metabolic model (glutamatergic and GABAergic neurons and astroglia) to the eighteen time courses. Vpc varied by 44% across brain regions, being lowest in the cerebellum (0.087 ± 0.004 µmol/g/min) and highest in striatum (0.125 ± 0.009) with intermediate values in cerebral cortex (0.106 ± 0.005) and hippocampus (0.114 ± 0.005). Vpc constituted 13-19% of the oxidative glucose consumption rate. Combination of cerebral cortical data with literature values revealed a positive correlation between Vpc and the rates of glutamate/glutamine-cycling and oxidative glucose consumption, respectively, consistent with earlier observations.

Human brain functional MRS reveals interplay of metabolites implicated in neurotransmission and neuroenergetics

While functional MRI (fMRI) localizes brain activation and deactivation, functional MRS (fMRS) provides insights into the underlying metabolic conditions. There is much interest in measuring task-induced and resting levels of metabolites implicated in neuroenergetics (e.g., lactate, glucose, or β-hydroxybutyrate (BHB)) and neurotransmission (e.g., γ-aminobutyric acid (GABA) or pooled glutamate and glutamine (Glx)). Ultra-high magnetic field (e.g., 7T) has boosted the fMRS quantification precision, reliability, and stability of spectroscopic observations using short echo-time (TE) ¹H-MRS techniques. While short TE ¹H-MRS lacks sensitivity and specificity for fMRS at lower magnetic fields (e.g., 3T or 4T), most of these metabolites can also be detected by J-difference editing (JDE) ¹H-MRS with longer TE to filter overlapping resonances. The ¹H-MRS studies show that JDE can detect GABA, Glx, lactate, and BHB at 3T, 4T and 7T. Most recently, it has also been demonstrated that JDE ¹H-MRS is capable of reliable detection of metabolic changes in different brain areas at various magnetic fields. Combining fMRS measurements with fMRI is important for understanding normal brain function, but also clinically relevant for mechanisms and/or biomarkers of neurological and neuropsychiatric disorders. We provide an up-to-date overview of fMRS research in the last three decades, both in terms of applications and technological advances. Overall the emerging fMRS techniques can be expected to contribute substantially to our understanding of metabolism for brain function and dysfunction.

A Novel Biomarker of Neuronal Glutamate Metabolism in Nonhuman Primates Using Localized 1H-Magnetic Resonance Spectroscopy: Development and Effects of BNC375, an α7 Nicotinic Acetylcholine Receptor Positive Allosteric Modulator

BACKGROUND

The development of treatments for cognitive deficits associated with central nervous system disorders is currently a significant medical need. Despite the great need for such therapeutics, a significant challenge in the drug development process is the paucity of robust biomarkers to assess target modulation and guide clinical decisions. We developed a novel, translatable biomarker of neuronal glutamate metabolism, the ¹³C-glutamate+glutamine (Glx) H3:H4 labeling ratio, in nonhuman primates using localized ¹H-magnetic resonance spectroscopy combined with ¹³C-glucose infusions.

METHODS

We began with numerical simulations in an established model of brain glutamate metabolism, showing that the ¹³C-Glx H3:H4 ratio should be a sensitive biomarker of neuronal tricarboxylic acid cycle activity, a key measure of overall neuronal metabolism. We showed that this biomarker can be measured reliably using a standard ¹H-magnetic resonance spectroscopy method (point-resolved spectroscopy sequence/echo time = 20 ms), obviating the need for specialized hardware and pulse sequences typically used with ¹³C-magnetic resonance spectroscopy, thus improving overall clinical translatability. Finally, we used this biomarker in 8 male rhesus macaques before and after administration of the compound BNC375, a positive allosteric modulator of the α7 nicotinic acetylcholine receptor that enhances glutamate signaling ex vivo and elicits procognitive effects in preclinical species.

RESULTS

The ¹³C-Glx H3:H4 ratios in the monkeys showed that BNC375 increases neuronal metabolism in nonhuman primates in vivo, detectable on an individual basis.

CONCLUSIONS

This study demonstrates that the ratio of ¹³C-Glx H3:H4 labeling is a biomarker that may provide an objective readout of compounds affecting glutamatergic neurotransmission and could improve decision making for the development of therapeutic agents.

Glucose sparing by glycogenolysis (GSG) determines the relationship between brain metabolism and neurotransmission

Over the last two decades, it has been established that glucose metabolic fluxes in neurons and astrocytes are proportional to the rates of the glutamate/GABA-glutamine neurotransmitter cycles in close to 1:1 stoichiometries across a wide range of functional energy demands. However, there is presently no mechanistic explanation for these relationships. We present here a theoretical meta-analysis that tests whether the brain's unique compartmentation of glycogen metabolism in the astrocyte and the requirement for neuronal glucose homeostasis lead to the observed stoichiometries. We found that blood-brain barrier glucose transport can be limiting during activation and that the energy demand could only be met if glycogenolysis supports neuronal glucose metabolism by replacing the glucose consumed by astrocytes, a mechanism we call Glucose Sparing by Glycogenolysis (GSG). The predictions of the GSG model are in excellent agreement with a wide range of experimental results from rats, mice, tree shrews, and humans, which were previously unexplained. Glycogenolysis and glucose sparing dictate the energy available to support neuronal activity, thus playing a fundamental role in brain function in health and disease.

Reversibility of brain glucose kinetics in type 2 diabetes mellitus

AIMS/HYPOTHESIS

We have previously shown that individuals with uncontrolled type 2 diabetes have a blunted rise in brain glucose levels measured by ¹H magnetic resonance spectroscopy. Here, we investigate whether reductions in HbA_1c normalise intracerebral glucose levels.

METHODS

Eight individuals (two men, six women) with poorly controlled type 2 diabetes and mean ± SD age 44.8 ± 8.3 years, BMI 31.4 ± 6.1 kg/m² and HbA_1c 84.1 ± 16.2 mmol/mol (9.8 ± 1.4%) underwent ¹H MRS scanning at 4 Tesla during a hyperglycaemic clamp (~12.21 mmol/l) to measure changes in cerebral glucose at baseline and after a 12 week intervention that improved glycaemic control through the use of continuous glucose monitoring, diabetes regimen intensification and frequent visits to an endocrinologist and nutritionist.

RESULTS

Following the intervention, mean ± SD HbA_1c decreased by 24.3 ± 15.3 mmol/mol (2.1 ± 1.5%) (p=0.006), with minimal weight changes (p=0.242). Using a linear mixed-effects regression model to compare glucose time courses during the clamp pre and post intervention, the pre-intervention brain glucose level during the hyperglycaemic clamp was significantly lower than the post-intervention brain glucose (p<0.001) despite plasma glucose levels during the hyperglycaemic clamp being similar (p=0.266). Furthermore, the increases in brain glucose were correlated with the magnitude of improvement in HbA_1c (r = 0.71, p=0.048).

CONCLUSION/INTERPRETATION

These findings highlight the potential reversibility of cerebral glucose transport capacity and metabolism that can occur in individuals with type 2 diabetes following improvement of glycaemic control. Trial registration ClinicalTrials.gov NCT03469492.

Prefrontal Glutamate Neurotransmission in PTSD: A Novel Approach to Estimate Synaptic Strength in Vivo in Humans

Background

Trauma and chronic stress are believed to induce and exacerbate psychopathology by disrupting glutamate synaptic strength. However, in vivo in human methods to estimate synaptic strength are limited. In this study, we established a novel putative biomarker of glutamatergic synaptic strength, termed energy-per-cycle (EPC). Then, we used EPC to investigate the role of prefrontal neurotransmission in trauma-related psychopathology.

Methods

Healthy controls (n = 18) and patients with posttraumatic stress (PTSD; n = 16) completed ¹³C-acetate magnetic resonance spectroscopy (MRS) scans to estimate prefrontal EPC, which is the ratio of neuronal energetic needs per glutamate neurotransmission cycle (V_TCA/V_Cycle).

Results

Patients with PTSD were found to have 28% reduction in prefrontal EPC (t = 3.0; df = 32, P = .005). There was no effect of sex on EPC, but age was negatively associated with prefrontal EPC across groups (r = -0.46, n = 34, P = .006). Controlling for age did not affect the study results.

Conclusion

The feasibility and utility of estimating prefrontal EPC using ¹³C-acetate MRS were established. Patients with PTSD were found to have reduced prefrontal glutamatergic synaptic strength. These findings suggest that reduced glutamatergic synaptic strength may contribute to the pathophysiology of PTSD and could be targeted by new treatments.

Ethnic and sex differences in hepatic lipid content and related cardiometabolic parameters in lean individuals

BackgroundNonalcoholic fatty liver affects 25% to 30% of the US and European populations; is associated with insulin resistance (IR), type 2 diabetes, and increased cardiovascular risk; and is defined by hepatic triglyceride (HTG) content greater than 5.56%. However, it is unknown whether HTG content less than 5.56% is associated with cardiometabolic risk factors and whether there are ethnic (Asian Indian, AI, versus non-AI) and/or sex differences in these parameters in lean individuals.MethodsWe prospectively recruited 2331 individuals and measured HTG, using 1H magnetic resonance spectroscopy, and plasma concentrations of triglycerides, total cholesterol, LDL-cholesterol, HDL-cholesterol, and uric acid. Insulin sensitivity was assessed using Homeostatic Model Assessment of Insulin Resistance and the Matsuda Insulin Sensitivity Index.ResultsThe 95th percentile for HTG in lean non-AI individuals was 1.85%. Plasma insulin, triglycerides, total cholesterol, LDL-cholesterol, and uric acid concentrations were increased and HDL-cholesterol was decreased in individuals with HTG content > 1.85% and ≤ 5.56% compared with those individuals with HTG content ≤ 1.85%, and these altered parameters were associated with increased IR. Mean HTG was lower in lean non-AI women compared with lean non-AI men, whereas lean AI men and women had a 40% to 100% increase in HTG when compared with non-AI men and women, which was associated with increased cardiometabolic risk factors.ConclusionWe found that the 95th percentile of HTG in lean non-AI individuals was 1.85% and that HTG concentrations above this threshold were associated with IR and cardiovascular risk factors. Premenopausal women were protected from these changes whereas young, lean AI men and women manifested increased HTG content and associated cardiometabolic risk factors.FundingGrants from the United States Department of Health and Human Resources (NIH/National Institute of Diabetes and Digestive and Kidney Diseases): R01 DK113984, P30 DK45735, U24 DK59635, and UL1 RR024139; and the Novo Nordisk Foundation (NNF18CC0034900).

From a Demand-Based to a Supply-Limited Framework of Brain Metabolism

What defines the rate of energy use by the brain, as well as per neurons of different sizes in different structures and animals, is one fundamental aspect of neuroscience for which much has been theorized, but very little data are available. The prevalent theories and models consider that energy supply from the vascular system to different brain regions is adjusted both dynamically and in the course of development and evolution to meet the demands of neuronal activity. In this perspective, we offer an alternative view: that regional rates of energy use might be mostly constrained by supply, given the properties of the brain capillary network, the highly stable rate of oxygen delivery to the whole brain under physiological conditions, and homeostatic constraints. We present evidence that these constraints, based on capillary density and tissue oxygen homeostasis, are similar between brain regions and mammalian species, suggesting they derive from fundamental biophysical limitations. The same constraints also determine the relationship between regional rates of brain oxygen supply and usage over the full physiological range of brain activity, from deep sleep to intense sensory stimulation, during which the apparent uncoupling of blood flow and oxygen use is still a predicted consequence of supply limitation. By carefully separating "energy cost" into energy supply and energy use, and doing away with the problematic concept of energetic "demands," our new framework should help shine a new light on the neurovascular bases of metabolic support of brain function and brain functional imaging. We speculate that the trade-offs between functional systems and even the limitation to a single attentional spot at a time might be consequences of a strongly supply-limited brain economy. We propose that a deeper understanding of brain energy supply constraints will provide a new evolutionary understanding of constraints on brain function due to energetics; offer new diagnostic insight to disturbances of brain metabolism; lead to clear, testable predictions on the scaling of brain metabolic cost and the evolution of brains of different sizes; and open new lines of investigation into the microvascular bases of progressive cognitive loss in normal aging as well as metabolic diseases.

MR spectroscopy frequency and phase correction using convolutional neural networks

PURPOSE

To introduce a novel convolutional neural network (CNN)-based approach for frequency-and-phase correction (FPC) of MR spectroscopy (MRS) spectra to achieve fast and accurate FPC of single-voxel MEGA-PRESS MRS data.

METHODS

Two neural networks (one for frequency and one for phase) were trained and validated using published simulated and in vivo MEGA-PRESS MRS dataset with wide-range artificial frequency and phase offsets applied. The CNN-based approach was subsequently tested and compared to the current deep learning solution: multilayer perceptrons (MLP). Furthermore, random noise was added to the original simulated dataset to further investigate the model performance at varied signal-to-noise ratio (SNR) levels (i.e., 10, 5, and 2.5). Additional frequency and phase offsets (i.e., small, moderate, large) were also applied to the in vivo dataset, and the CNN model was compared to the conventional approach SR and model-based SR implementation (mSR).

RESULTS

The CNN model is more robust to noise compared to the MLP-based approach due to having smaller mean absolute errors in both frequency (0.01 ± 0.01 Hz at SNR = 10 and 0.01 ± 0.02 Hz at SNR = 2.5) and phase (0.12 ± 0.09° at SNR = 10 and -0.07 ± 0.44° at SNR = 2.5) offset prediction. Furthermore, better performance was demonstrated for FPC when compared to the MLP-based approach, and SR when applied to the in vivo dataset for both with and without additional offsets.

CONCLUSION

A CNN-based approach provides a solution to the automated preprocessing of MRS data, and the experimental results demonstrate the quantitatively improved spectra quality compared to the state-of-the-art approach.

1H Magnetic Resonance Spectroscopy to Understand the Biological Basis of ALS, Diagnose Patients Earlier, and Monitor Disease Progression

At present, limited biomarkers exist to reliably understand, diagnose, and monitor the progression of amyotrophic lateral sclerosis (ALS), a fatal neurological disease characterized by motor neuron death. Standard MRI technology can only be used to exclude a diagnosis of ALS, but ¹H-MRS technology, which measures neurochemical composition, may provide the unique ability to reveal biomarkers that are specific to ALS and sensitive enough to diagnose patients at early stages in disease progression. In this review, we present a summary of current theories of how mitochondrial energetics and an altered glutamate/GABA neurotransmitter flux balance play a role in the pathogenesis of ALS. The theories are synthesized into a model that predicts how pathogenesis impacts glutamate and GABA concentrations. When compared with the results of all MRS studies published to date that measure the absolute concentrations of these neurochemicals in ALS patients, results were variable. However, when normalized for neuronal volume using the MRS biomarker N-acetyl aspartate (NAA), there is clear evidence for an elevation of neuronal glutamate in nine out of thirteen studies reviewed, an observation consistent with the predictions of the model of increased activity of glutamatergic neurons and excitotoxicity. We propose that this increase in neuronal glutamate concentration, in combination with decreased neuronal volume, is specific to the pathology of ALS. In addition, when normalized to glutamate levels, there is clear evidence for a decrease in neuronal GABA in three out of four possible studies reviewed, a finding consistent with a loss of inhibitory regulation contributing to excessive neuronal excitability. The combination of a decreased GABA/Glx ratio with an elevated Glx/NAA ratio may enhance the specificity for ¹H-MRS detection of ALS and ability to monitor glutamatergic and GABAergic targeted therapeutics. Additional longitudinal studies calculating the exact value of these ratios are needed to test these hypotheses and understand how ratios may change over the course of disease progression. Proposed modifications to the experimental design of the reviewed ¹H MRS studies may also increase the sensitivity of the technology to changes in these neurochemicals, particularly in early stages of disease progression.

Analysis of the time course of COVID-19 cases and deaths from countries with extensive testing allows accurate early estimates of the age specific symptomatic CFR values

BACKGROUND

Knowing the true infected and symptomatic case fatality ratios (IFR and CFR) for COVID-19 is of high importance for epidemiological model projections. Early in the pandemic many locations had limited testing and reporting, so that standard methods for determining IFR and CFR required large adjustments for missed cases. We present an alternate approach, based on results from the countries at the time that had a high test to positive case ratio to estimate symptomatic CFR.

METHODS

We calculated age specific (0-69, 70-79, 80+ years old) time corrected crude symptomatic CFR values from 7 countries using two independent time to fatality correction methods. Data was obtained through May 7, 2020. We applied linear regression to determine whether the mean of these coefficients had converged to the true symptomatic CFR values. We then tested these coefficients against values derived in later studies as well as a large random serological study in NYC at that time.

RESULTS

The age dependent symptomatic CFR values accurately predicted the percentage of the population infected as reported by two random testing studies in NYC. They also were in good agreement with later studies that estimated age specific IFR and CFR values from serological studies and more extensive data sets available later in the pandemic.

CONCLUSIONS

We found that for regions with extensive testing it is possible to get early accurate symptomatic CFR coefficients. These values, in combination with an estimate of the age dependence of infection, allows symptomatic CFR values and percentage of the population that is infected to be determined in similar regions with limited testing.

Two transition states of the glycogen shunt and two steady states of gene expression support metabolic flexibility and the Warburg effect in cancer

Previously we suggested that the early Warburg effect can be explained by the use by cancer cells the glycogen shunt during a rapid increase in glucose concentration. In analogy to the Crabtree effect in yeast, the shunt plays a critical role in maintaining homeostasis of glycolytic intermediate levels during these transitions. We extend this analysis here, and propose that the recently appreciated flexibility of cancer cell glucose and glycogen metabolism involves 4 metabolic states that we recently identified in metabolic control analysis studies of yeast. Under stable conditions of low glucose and normal O₂ yeast, and by analogy cancer, cells are in the Respiration State in which through gene expression for oxidizing non glucose substrates. When their environment changes to high glucose with reduced O₂ levels, such as occur in tumors, they transition to the Glycolysis State due to gene expression of new glycolytic enzyme isoforms such as PKM2. These isoforms optimize metabolism to sustain the Warburg effect. When the changes in glucose and O₂ levels are rapid there may be insufficient time for gene expression to adapt. The metabolic flexibility conferred by 2 states of the glycogen shunt allow the cells to survive these transitions. The model explains experimental observations in cancer such as the function of the glycogen shunt and the frequent expression of PKM2 in cells undergoing the Warburg Effect. A surprising conclusion is that the function of PKM2 is to maintain glycolytic intermediate homeostasis rather than controlling the glycolytic flux. The glycogen shunt may also have an important role in cancer metabolic reprogramming by allowing cancer cells to survive large glucose and oxygen changes during the selection of mutations that lead to the Warburg phenotype

Diffuse white matter loss in a transgenic rat model of cerebral amyloid angiopathy

Diffuse white matter (WM) disease is highly prevalent in elderly with cerebral small vessel disease (cSVD). In humans, cSVD such as cerebral amyloid angiopathy (CAA) often coexists with Alzheimer's disease imposing a significant impediment for characterizing their distinct effects on WM. Here we studied the burden of age-related CAA pathology on WM disease in a novel transgenic rat model of CAA type 1 (rTg-DI). A cohort of rTg-DI and wild-type rats was scanned longitudinally using MRI for characterization of morphometry, cerebral microbleeds (CMB) and WM integrity. In rTg-DI rats, a distinct pattern of WM loss was observed at 9 M and 11 M. MRI also revealed manifestation of small CMB in thalamus at 6 M, which preceded WM loss and progressively enlarged until the moribund disease stage. Histology revealed myelin loss in the corpus callosum and thalamic CMB in all rTg-DI rats, the latter of which manifested in close proximity to occluded and calcified microvessels. The quantitation of CAA load in rTg-DI rats revealed that the most extensive microvascular Aβ deposition occurred in the thalamus. For the first time using in vivo MRI, we show that CAA type 1 pathology alone is associated with a distinct pattern of WM loss.

Metabolic underpinnings of activated and deactivated cortical areas in human brain

Neuroimaging with functional MRI (fMRI) identifies activated and deactivated brain regions in task-based paradigms. These patterns of (de)activation are altered in diseases, motivating research to understand their underlying biochemical/biophysical mechanisms. Essentially, it remains unknown how aerobic metabolism of glucose to lactate (aerobic glycolysis) and excitatory-inhibitory balance of glutamatergic and GABAergic neuronal activities vary in these areas. In healthy volunteers, we investigated metabolic distinctions of activating visual cortex (VC, a task-positive area) using a visual task and deactivating posterior cingulate cortex (PCC, a task-negative area) using a cognitive task. We used fMRI-guided J-edited functional MRS (fMRS) to measure lactate, glutamate plus glutamine (Glx) and γ-aminobutyric acid (GABA), as indicators of aerobic glycolysis and excitatory-inhibitory balance, respectively. Both lactate and Glx increased upon activating VC, but did not change upon deactivating PCC. Basal GABA was negatively correlated with BOLD responses in both brain areas, but during functional tasks GABA decreased in VC upon activation and GABA increased in PCC upon deactivation, suggesting BOLD responses in relation to baseline are impacted oppositely by task-induced inhibition. In summary, opposite relations between BOLD response and GABAergic inhibition, and increases in aerobic glycolysis and glutamatergic activity distinguish the BOLD response in (de)activated areas.

Dissociation of Muscle Insulin Resistance from Alterations in Mitochondrial Substrate Preference

Alterations in muscle mitochondrial substrate preference have been postulated to play a major role in the pathogenesis of muscle insulin resistance. In order to examine this hypothesis, we assessed the ratio of mitochondrial pyruvate oxidation (V_PDH) to rates of mitochondrial citrate synthase flux (V_CS) in muscle. Contrary to this hypothesis, we found that high-fat-diet (HFD)-fed insulin-resistant rats did not manifest altered muscle substrate preference (V_PDH/V_CS) in soleus or quadriceps muscles in the fasting state. Furthermore, hyperinsulinemic-euglycemic (HE) clamps increased V_PDH/V_CS in both muscles in normal and insulin-resistant rats. We then examined the muscle V_PDH/V_CS flux in insulin-sensitive and insulin-resistant humans and found similar relative rates of V_PDH/V_CS, following an overnight fast (∼20%), and similar increases in V_PDH/V_CS fluxes during a HE clamp. Altogether, these findings demonstrate that alterations in mitochondrial substrate preference are not an essential step in the pathogenesis of muscle insulin resistance.

Elevated homocarnosine and GABA in subject on isoniazid as assessed through 1H MRS at 7T

Typical magnetic resonance spectroscopy J-editing methods designed to quantify GABA suffer from contamination of both overlapping macromolecules and homocarnosine signal, introducing potential confounds. The aim of this study was to develop a novel method to assess accurately both the relative concentrations of homocarnosine as well as GABA free from overlapping creatine, homocarnosine and macromolecule signal. A novel method which utilized the combination of echo time STEAM and MEGA-sLASER magnetic resonance spectroscopy experiments at 7T were used to quantify the concentration of GABA and homocarnsoine independently, which are typically quantified in tandem. The metabolites GABA and homocarnosine were measured in brain of 6 healthy control subjects, and in a single subject medicated with isoniazid. It was found that (16.6±10.2)% of the supposed GABA signal in the brain originated from homocarnosine, and that isoniazid caused significantly elevated concentration of GABA and homocarnosine in a single subject compared to controls.

Early life stress and glutamate neurotransmission in major depressive disorder

Early life stress (ELS) and glutamate neurotransmission have been implicated in the pathophysiology of major depressive disorder (MDD). In non-human primates, ELS was positively correlated with cortical Glx (i.e., glutamate + glutamine). However, the relationship between ELS and cortical glutamate in adult patients with MDD is not fully known. Using ¹H Magnetic Resonance Spectroscopy (MRS), we conducted exploratory analyses measuring occipital cortical glutamate and glutamine levels in 36 medication-free patients with MDD. In a subsample (n=11), we measured dynamic glutamate/glutamine cycling (V_cycle) using advanced ¹³C MRS methods. ELS history was assessed using Early-life Trauma Inventory (ETI). Exploratory analyses suggest a relationship between ETI and glutamine as reflected by a significant positive correlation between ETI scores and occipital glutamine (r_s=0.39, p=0.017) but not glutamate. Post-hoc analyses showed that the association with glutamine was driven by the ETI emotional abuse (ETI-EA) subscale (r_s=0.39, p=0.02). V_cycle correlation with ETI was at trend level (r_s=0.55, p=0.087) and significantly correlated with ETI-EA (r_s=0.67, p=0.03). In this small sample of patients with MDD, those with childhood emotional abuse appear to have increased occipital glutamate neurotransmission as reflected by increased glutamate/glutamine cycling and glutamine level. Future studies would be needed to confirm this pilot evidence and to examine whether ELS effects on glutamate neurotransmission underlie the relationship between ELS and psychopathology.

Hippocampal Pathology in Clinical High-Risk Patients and the Onset of Schizophrenia

BACKGROUND

We examined neuroimaging-derived hippocampal biomarkers in subjects at clinical high risk (CHR) for psychosis to further characterize the pathophysiology of early psychosis. We hypothesized that glutamate hyperactivity, reflected by increased metabolic activity derived from functional magnetic resonance imaging in the CA1 hippocampal subregion and from proton magnetic resonance spectroscopy-derived hippocampal levels of glutamate/glutamine, represents early hippocampal dysfunction in CHR subjects and is predictive of conversion to syndromal psychosis.

METHODS

We enrolled 75 CHR individuals with attenuated positive symptom psychosis-risk syndrome as defined by the Structured Interview for Psychosis-risk Syndromes. We used optimized magnetic resonance imaging techniques to measure 3 validated in vivo pathologies of hippocampal dysfunction-focal cerebral blood volume, focal atrophy, and evidence of elevated glutamate concentrations. All patients were imaged at baseline and were followed for up to 2 years to assess for conversion to psychosis.

RESULTS

At baseline, compared with control subjects, CHR individuals had high glutamate/glutamine and elevated focal cerebral blood volume on functional magnetic resonance imaging, but only baseline focal hippocampal atrophy predicted progression to syndromal psychosis.

CONCLUSIONS

These findings provide evidence that CHR patients with attenuated psychotic symptoms have glutamatergic abnormalities, although only CHR patients who develop syndromal psychosis exhibit focal hippocampal atrophy. Furthermore, these results support the growing evidence that hippocampal dysfunction is an early feature of schizophrenia and related psychotic disorders.

Glymphatic System Function in Relation to Anesthesia and Sleep States

The brain is one of the most metabolically active organs in the body. The brain's high energy demand associated with wakefulness persists during rapid eye movement sleep, and even during non-rapid eye movement sleep, cerebral oxygen consumption is only reduced by 20%. The active bioenergetic state parallels metabolic waste production at a higher rate than in other organs, and the lack of lymphatic vasculature in brain parenchyma is therefore a conundrum. A common assumption has been that with a tight blood-brain barrier restricting solute and fluid movements, a lymphatic system is superfluous in the central nervous system. Cerebrospinal fluid (CSF) flow has long been thought to facilitate central nervous system tissue "detoxification" in place of lymphatics. Nonetheless, while CSF production and transport have been studied for decades, the exact processes involved in toxic waste clearance remain poorly understood. Over the past 5 years, emerging data have begun to shed new light on these processes in the form of the "glymphatic system," a novel brain-wide perivascular transit passageway dedicated to CSF transport and metabolic waste drainage from the brain. Here, we review the key anatomical components and operational drivers of the brain's glymphatic system, with a focus on its unique functional dependence on the state of arousal and anesthetic regimens. We also discuss evidence for why clinical exploration of this novel system may in the future provide valuable insight into new strategies for preventing delirium and cognitive dysfunction in perioperative and critical care settings.

Extracellular pH mapping of liver cancer on a clinical 3T MRI scanner

PURPOSE

To demonstrate feasibility of developing a noninvasive extracellular pH (pH_e ) mapping method on a clinical MRI scanner for molecular imaging of liver cancer.

METHODS

In vivo pH_e mapping has been demonstrated on preclinical scanners (e.g., 9.4T, 11.7T) with Biosensor Imaging of Redundant Deviation in Shifts (BIRDS), where the pH_e readout by 3D chemical shift imaging (CSI) depends on hyperfine shifts emanating from paramagnetic macrocyclic chelates like TmDOTP^5- which upon extravasation from blood resides in the extracellular space. We implemented BIRDS-based pH_e mapping on a clinical 3T Siemens scanner, where typically diamagnetic ¹ H signals are detected using millisecond-long radiofrequency (RF) pulses, and ¹ H shifts span over ±10 ppm with long transverse (T₂ , 10² ms) and longitudinal (T₁ , 10³ ms) relaxation times. We modified this 3D-CSI method for ultra-fast acquisition with microsecond-long RF pulses, because even at 3T the paramagnetic ¹ H shifts of TmDOTP^5- have millisecond-long T₂ and T₁ and ultra-wide chemical shifts (±200 ppm) as previously observed in ultra-high magnetic fields.

RESULTS

We validated BIRDS-based pH in vitro with a pH electrode. We measured pH_e in a rabbit model for liver cancer using VX2 tumors, which are highly vascularized and hyperglycolytic. Compared to intratumoral pH_e (6.8 ± 0.1; P < 10^-9 ) and tumor's edge pH_e (6.9 ± 0.1; P < 10^-7 ), liver parenchyma pH_e was significantly higher (7.2 ± 0.1). Tumor localization was confirmed with histopathological markers of necrosis (hematoxylin and eosin), glucose uptake (glucose transporter 1), and tissue acidosis (lysosome-associated membrane protein 2).

CONCLUSION

This work demonstrates feasibility and potential clinical translatability of high-resolution pH_e mapping to monitor tumor aggressiveness and therapeutic outcome, all to improve personalized cancer treatment planning.

Prognosticating brain tumor patient survival after laser thermotherapy: Comparison between neuroradiological reading and semi-quantitative analysis of MRI data

BACKGROUND AND PURPOSE

Given increasing interest in laser interstitial thermotherapy (LITT) to treat brain tumor patients, we explored if examining multiple MRI contrasts per brain tumor patient undergoing surgery can impact predictive accuracy of survival post-LITT.

MATERIALS AND METHODS

MRI contrasts included fluid-attenuated inversion recovery (FLAIR), T1 pre-gadolinium (T1pre), T1 post-gadolinium (T1Gd), T2, diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC), susceptibility weighted images (SWI), and magnetization-prepared rapid gradient-echo (MPRAGE). The latter was used for MRI data registration across preoperative to postoperative scans. Two ROIs were identified by thresholding preoperative FLAIR (large ROI) and T1Gd (small ROI) images. For each MRI contrast, a numerical score was assigned based on changing image intensity of both ROIs (vs. a normal ROI) from preoperative to postoperative stages. The fully-quantitative method was based on changing image intensity across scans at different stages without any human intervention, whereas the semi-quantitative method was based on subjective criteria of cumulative trends across scans at different stages. A fully-quantitative/semi-quantitative score per patient was obtained by averaging scores for each MRI contrast. A standard neuroradiological reading score per patient was obtained from radiological interpretation of MRI data. Scores from all 3 methods per patient were compared against patient survival, and re-examined for comorbidity and pathology effects.

RESULTS

Patient survival correlated best with semi-quantitative scores obtained from T1Gd, ADC, and T2 data, and these correlations improved when biopsy and comorbidity were included.

CONCLUSION

These results suggest interfacing neuroradiological readings with semi-quantitative image analysis can improve predictive accuracy of patient survival.

In vivo 13 C and 1 H-[13 C] MRS studies of neuroenergetics and neurotransmitter cycling, applications to neurological and psychiatric disease and brain cancer

In the last 25 years ¹³ C MRS has been established as the only noninvasive method for measuring glutamate neurotransmission and cell specific neuroenergetics. Although technically and experimentally challenging ¹³ C MRS has already provided important new information on the relationship between neuroenergetics and neuronal function, the high energy cost of brain function in the resting state and the role of altered neuroenergetics and neurotransmitter cycling in disease. In this paper we review the metabolic and neurotransmitter pathways that can be measured by ¹³ C MRS and key findings on the linkage between neuroenergetics, neurotransmitter cycling, and brain function. Applications of ¹³ C MRS to neurological and psychiatric disease as well as brain cancer are reviewed. Recent technological developments that may help to overcome spatial resolution and brain coverage limitations of ¹³ C MRS are discussed.

Dynamic Thermal Mapping of Localized Therapeutic Hypothermia in the Brain

Although whole body cooling is used widely to provide therapeutic hypothermia for the brain, there are undesirable clinical side effects. Selective brain cooling may allow for rapid and controllable neuroprotection while mitigating these undesirable side effects. We evaluated an innovative cerebrospinal fluid (CSF) cooling platform that utilizes chilled saline pumped through surgically implanted intraventricular catheters to induce hypothermia. Magnetic resonance thermal imaging of the healthy sheep brain (n = 4) at 7.0T provided dynamic temperature measurements from the whole brain. Global brain temperature was 38.5 ± 0.8°C at baseline (body temperature of 39.2 ± 0.4°C), and decreased by 3.1 ± 0.3°C over ∼30 min of cooling (p < 0.0001). Significant cooling was achieved in all defined regions across both the ipsilateral and contralateral hemispheres relative to catheter placement. On cooling cessation, global brain temperature increased by 3.1 ± 0.2°C over ∼20 min (p < 0.0001). Rapid and synchronized temperature fall/rise on cooling onset/offset was observed reproducibly with rates ranging from 0.06-0.21°C/min, where rewarming was faster than cooling (p < 0.0001) signifying the importance of thermoregulation in the brain. Although core regions (including the subcortex, midbrain, olfactory tract, temporal lobe, occipital lobe, and parahippocampal cortex) had slightly warmer (∼0.2°C) baseline temperatures, after cooling, temperatures reached the same level as the non-core regions (35.6 ± 0.2°C), indicating the cooling effectiveness of the CSF-based cooling device. In summary, CSF-based intraventricular cooling reliably reduces temperature in all identified brain regions to levels known to be neuroprotective, while maintaining overall systemic normothermia. Dynamic thermal mapping provides high spatiotemporal temperature measurements that can aid in optimizing selective neuroprotective protocols.

Strength of resting state functional connectivity and local GABA concentrations predict oral reading of real and pseudo-words

Reading is a learned activity that engages multiple cognitive systems. In a cohort of typical and struggling adult readers we show evidence that successful oral reading of real words is related to gamma-amino-butyric acid (GABA) concentration in the higher-order language system, whereas reading of unfamiliar pseudo-words is not related to GABA in this system. We also demonstrate the capability of resting state functional connectivity (rsFC) combined with GABA measures to predict single real word compared to pseudo-word reading performance. Results show that the strength of rsFC between left fusiform gyrus (L-FG) and higher-order language systems predicts oral reading behavior of real words, irrespective of the local concentration of GABA. On the other hand, pseudo-words, which require grapheme-to-phoneme conversion, are not predicted by the connection between L-FG and higher-order language system. This suggests that L-FG may have a multi-functional role: lexical processing of real words and grapheme-to-phoneme processing of pseudo-words. Additionally, rsFC between L-FG, pre-motor, and putamen areas are positively related to the oral reading of both real and pseudo-words, suggesting that text may be converted into a phoneme sequence for speech initiation and production regardless of whether the stimulus is a real word or pseudo-word. In summary, from a systems neuroscience perspective, we show that: (i) strong rsFC between higher order visual, language, and pre-motor areas can predict and differentiate efficient oral reading of real and pseudo-words. (ii) GABA measures, along with rsFC, help to further differentiate the neural pathways for previously learned real words versus unfamiliar pseudo-words.

Glutaminase activity in GLS1 Het mouse brain compared to putative pharmacological inhibition by ebselen using ex vivo MRS

Glutaminase mediates the recycling of neurotransmitter glutamate, supporting most excitatory neurotransmission in the mammalian central nervous system. A constitutive heterozygous reduction in GLS1 engenders in mice a model of schizophrenia resilience and associated increases in Gln, reductions in Glu and activity-dependent attenuation of excitatory synaptic transmission. Hippocampal brain slices from GLS1 heterozygous mice metabolize less Gln to Glu. Whether glutaminase activity is diminished in the intact brain in GLS1 heterozygous mice has not been assessed, nor the regional impact. Moreover, it is not known whether pharmacological inhibition would mimic the genetic reduction. We addressed this using magnetic resonance spectroscopy to assess amino acid content and ¹³C-acetate loading to assess glutaminase activity, in multiple brain regions. Glutaminase activity was reduced significantly in the hippocampus of GLS1 heterozygous mice, while acute treatment with the putative glutaminase inhibitor ebselen did not impact glutaminase activity, but did significantly increase GABA. This approach identifies a molecular imaging strategy for testing target engagement by comparing genetic and pharmacological inhibition, across brain regions.

A Non-cognitive Behavioral Model for Interpreting Functional Neuroimaging Studies

The dominant model for interpreting brain imaging experiments, which we refer to as the Standard Cognitive Model (SCM), assumes that the brain is organized in support of mental processes that control behavior. However, functional neuroimaging experiments of cognitive tasks have not shown clear anatomic segregation between mental processes originally proposed by this model. This failing has been blamed on limitations in imaging technology and non-linearity in the brain’s implementation of these processes. However, the validity of the underlying cognitive models used to describe the brain has rarely been questioned or directly tested against imaging results. We propose an alternative model of brain function, that we term the Non-cognitive Behavioral Model (NBM), which correlates observed human behavior directly with measured brain activity without making assumptions about intervening cognitive processes. Our model derives from behavioral psychology but is extended to include brain activity, in addition to behavior, as observables. A further extension is the role of neuroplasticity, as opposed to innate cognitive processes, in developing the brain’s support of cognitive behavior. We present the theoretical basis with which the SCM maps cognitive processes onto functional magnetic resonance and positron emission tomography images and compare and contrast with the NBM. We also describe how the NBM can be used experimentally to study how the brain supports behavior. Two applications are presented that support the usefulness of the NBM. In one, the NBM use of the total functional imaging signal (not just the differences between states) provides a stronger correlation of neural activity with the behavioral state of consciousness than the SCM approach in both anesthesia and coma. The second example reviews studies of facial and object recognition that provide evidence for the NBM proposal that neuroplasticity and experience play key roles in the brain’s support of recognition and other behaviors. The conclusions regarding neuroplasticity are then generalized to explain the incomplete functional segregation observed in the application of the SCM to neuroimaging.

Glycemic Variability and Brain Glucose Levels in Type 1 Diabetes

The impact of glycemic variability on brain glucose transport kinetics among individuals with type 1 diabetes mellitus (T1DM) remains unclear. Fourteen individuals with T1DM (age 35 ± 4 years; BMI 26.0 ± 1.4 kg/m²; HbA_1c 7.6 ± 0.3) and nine healthy control participants (age 32 ± 4; BMI 23.1 ± 0.8; HbA_1c 5.0 ± 0.1) wore a continuous glucose monitor (Dexcom) to measure hypoglycemia, hyperglycemia, and glycemic variability for 5 days followed by ¹H MRS scanning in the occipital lobe to measure the change in intracerebral glucose levels during a 2-h glucose clamp (target glucose concentration 220 mg/dL). Hyperglycemic clamps were also performed in a rat model of T1DM to assess regional differences in brain glucose transport and metabolism. Despite a similar change in plasma glucose levels during the hyperglycemic clamp, individuals with T1DM had significantly smaller increments in intracerebral glucose levels (P = 0.0002). Moreover, among individuals with T1DM, the change in brain glucose correlated positively with the lability index (r = 0.67, P = 0.006). Consistent with findings in humans, streptozotocin-treated rats had lower brain glucose levels in the cortex, hippocampus, and striatum compared with control rats. These findings that glycemic variability is associated with brain glucose levels highlight the need for future studies to investigate the impact of glycemic variability on brain glucose kinetics.

Development of a Model to Test Whether Glycogenolysis Can Support Astrocytic Energy Demands of Na+, K+-ATPase and Glutamate-Glutamine Cycling, Sparing an Equivalent Amount of Glucose for Neurons

Recent studies of glycogen in brain have suggested a much more important role in brain energy metabolism and function than previously recognized, including findings of much higher than previously recognized concentrations, consumption at substantial rates compared with utilization of blood-borne glucose, and involvement in ion pumping and in neurotransmission and memory. However, it remains unclear how glycogenolysis is coupled to neuronal activity and provides support for neuronal as well as astroglial function. At present, quantitative aspects of glycogenolysis in brain functions are very difficult to assess due to its metabolic lability, heterogeneous distributions within and among cells, and extreme sensitivity to physiological stimuli. To begin to address this problem, the present study develops a model based on pathway fluxes, mass balance, and literature relevant to functions and turnover of pathways that intersect with glycogen mobilization. A series of equations is developed to describe the stoichiometric relationships between net glycogen consumption that is predominantly in astrocytes with the rate of the glutamate-glutamine cycle, rates of astrocytic and neuronal glycolytic and oxidative metabolism, and the energetics of sodium/potassium pumping in astrocytes and neurons during brain activation. Literature supporting the assumptions of the model is discussed in detail. The overall conclusion is that astrocyte glycogen metabolism is primarily coupled to neuronal function via fueling glycolytically pumping of Na⁺ and K⁺ and sparing glucose for neuronal oxidation, as opposed to previous proposals of coupling neurotransmission via glutamate transport, lactate shuttling, and neuronal oxidation of lactate.

Trajectories of Brain Lactate and Re-visited Oxygen-Glucose Index Calculations Do Not Support Elevated Non-oxidative Metabolism of Glucose Across Childhood

Brain growth across childhood is a dynamic process associated with specific energy requirements. A disproportionately higher rate of glucose utilization (CMR_glucose) compared with oxygen consumption (CMR_O2) was documented in children's brain and suggestive of non-oxidative metabolism of glucose. Several candidate metabolic pathways may explain the CMR_glucose-CMR_O2 mismatch, and lactate production is considered a major contender. The ~33% excess CMR_glucose equals 0.18 μmol glucose/g/min and predicts lactate release of 0.36 μmol/g/min. To validate such scenario, we measured the brain lactate concentration ([Lac]) in 65 children to determine if indeed lactate accumulates and is high enough to (1) account for the glucose consumed in excess of oxygen and (2) support a high rate of lactate efflux from the young brain. Across childhood, brain [Lac] was lower than predicted, and below the range for adult brain. In addition, we re-calculated the CMR_glucose-CMR_O2 mismatch itself by using updated lumped constant values. The calculated cerebral metabolic rate of lactate indicated a net influx of 0.04 μmol/g/min, or in terms of CMR_glucose, of 0.02 μmol glucose/g/min. Accumulation of [Lac] and calculated efflux of lactate from brain are not consistent with the increase in non-oxidative metabolism of glucose. In addition, the value for the lumped constant for [¹⁸F]fluorodeoxyglucose has a high impact on calculated CMR_glucose and use of updated values alters or eliminates the CMR_glucose-CMR_O2 mismatch in developing brain. We conclude that the presently-accepted notion of non-oxidative metabolism of glucose during childhood must be revisited and deserves further investigations.

Functional MRS with J-edited lactate in human motor cortex at 4 T

While functional MRI (fMRI) localizes regions of brain activation, functional MRS (fMRS) provides insights into metabolic underpinnings. Previous fMRS studies detected task-induced lactate increase using short echo-time non-edited ¹H-MRS protocols, where lactate changes depended on accurate exclusion of overlapping lactate and lipid/macromolecule signals. Because long echo-time J-difference ¹H-MRS detection of lactate is less susceptible to this shortcoming, we posited if J-edited fMRS protocol could reliably detect metabolic changes in the human motor cortex during a finger-tapping paradigm in relation to a reliable measure of basal lactate. Our J-edited fMRS protocol at 4T was guided by an fMRI pre-scan to determine the ¹H-MRS voxel placement in the motor cortex. Because lactate and β-hydroxybutyrate (BHB) follow similar J-evolution profiles we observed both metabolites in all spectra, but only lactate showed reproducible task-induced modulation by 0.07 mM from a basal value of 0.82 mM. These J-edited fMRS results demonstrate good sensitivity and specificity for task-induced lactate modulation, suggesting that J-edited fMRS studies can be used to investigate the metabolic underpinning of human cognition by measuring lactate dynamics associated with activation and deactivation fMRI paradigms across brain regions at magnetic field lower than 7T.

Neurochemistry Predicts Convergence of Written and Spoken Language: A Proton Magnetic Resonance Spectroscopy Study of Cross-Modal Language Integration

Recent studies have provided evidence of associations between neurochemistry and reading (dis)ability (Pugh et al., 2014). Based on a long history of studies indicating that fluent reading entails the automatic convergence of the written and spoken forms of language and our recently proposed Neural Noise Hypothesis (Hancock et al., 2017), we hypothesized that individual differences in cross-modal integration would mediate, at least partially, the relationship between neurochemical concentrations and reading. Cross-modal integration was measured in 231 children using a two-alternative forced choice cross-modal matching task with three language conditions (letters, words, and pseudowords) and two levels of difficulty within each language condition. Neurometabolite concentrations of Choline (Cho), Glutamate (Glu), gamma-Aminobutyric (GABA), and N- acetyl-aspartate (NAA) were then measured in a subset of this sample (n = 70) with Magnetic Resonance Spectroscopy (MRS). A structural equation mediation model revealed that the effect of cross-modal word matching mediated the relationship between increased Glu (which has been proposed to be an index of neural noise) and poorer reading ability. In addition, the effect of cross-modal word matching fully mediated a relationship between increased Cho and poorer reading ability. Multilevel mixed effects models confirmed that lower Cho predicted faster cross-modal matching reaction time, specifically in the hard word condition. These Cho findings are consistent with previous work in both adults and children showing a negative association between Cho and reading ability. We also found two novel neurochemical relationships. Specifically, lower GABA and higher NAA predicted faster cross-modal matching reaction times. We interpret these results within a biochemical framework in which the ability of neurochemistry to predict reading ability may at least partially be explained by cross-modal integration.

The effects of ketamine on prefrontal glutamate neurotransmission in healthy and depressed subjects

The ability of ketamine administration to activate prefrontal glutamate neurotransmission is thought to be a key mechanism contributing to its transient psychotomimetic effects and its delayed and sustained antidepressant effects. Rodent studies employing carbon-13 magnetic resonance spectroscopy (13C MRS) methods have shown ketamine and other N-methyl-D-aspartate (NMDA) receptor antagonists to transiently increase measures reflecting glutamate-glutamine cycling and glutamate neurotransmission in the frontal cortex. However, there are not yet direct measures of glutamate neurotransmission in vivo in humans to support these hypotheses. The current first-level pilot study employed a novel prefrontal 13C MRS approach similar to that used in the rodent studies for direct measurement of ketamine effects on glutamate-glutamine cycling. Twenty-one participants (14 healthy and 7 depressed) completed two 13C MRS scans during infusion of normal saline or subanesthetic doses of ketamine. Compared to placebo, ketamine increased prefrontal glutamate-glutamine cycling, as indicated by a 13% increase in 13C glutamine enrichment (t = 2.4, p = 0.02). We found no evidence of ketamine effects on oxidative energy production, as reflected by 13C glutamate enrichment. During ketamine infusion, the ratio of 13C glutamate/glutamine enrichments, a putative measure of neurotransmission strength, was correlated with the Clinician-Administered Dissociative States Scale (r = -0.54, p = 0.048). These findings provide the most direct evidence in humans to date that ketamine increases glutamate release in the prefrontal cortex, a mechanism previously linked to schizophrenia pathophysiology and implicated in the induction of rapid antidepressant effects.

Evaluating the gray and white matter energy budgets of human brain function

The insatiable appetite for energy to support human brain function is mainly supplied by glucose oxidation (CMR_glc(ox)). But how much energy is consumed for signaling and nonsignaling processes in gray/white matter is highly debated. We examined this issue by combining metabolic measurements of gray/white matter and a theoretical calculation of bottom-up energy budget using biophysical properties of neuronal/glial cells in conjunction with species-exclusive electrophysiological and morphological data. We calculated a CMR_glc(ox)-derived budget and confirmed it with experimental results measured by PET, autoradiography, ¹³C-MRS, and electrophysiology. Several conserved principles were observed regarding the energy costs for brain's signaling and nonsignaling components in both human and rat. The awake resting cortical signaling processes and mass-dependent nonsignaling processes, respectively, demand ∼70% and ∼30% of CMR_glc(ox). Inhibitory neurons and glia need 15-20% of CMR_glc(ox), with the rest demanded by excitatory neurons. Nonsignaling demands dominate in white matter, in near opposite contrast to gray matter demands. Comparison between ¹³C-MRS data and calculations suggests ∼1.2 Hz glutamatergic signaling rate in the awake human cortex, which is ∼4 times lower than signaling in the rat cortex. Top-down validated bottom-up budgets could allow computation of anatomy-based CMR_glc(ox) maps and accurate cellular level interpretation of brain metabolic imaging.

Deuterium metabolic imaging (DMI) for MRI-based 3D mapping of metabolism in vivo

Currently, the only widely available metabolic imaging technique in the clinic is positron emission tomography (PET) detection of the radioactive glucose analog 2-¹⁸F-fluoro-2-deoxy-d-glucose (¹⁸FDG). However, ¹⁸FDG-PET does not inform on metabolism downstream of glucose uptake and often provides ambiguous results in organs with intrinsic high glucose uptake, such as the brain. Deuterium metabolic imaging (DMI) is a novel, noninvasive approach that combines deuterium magnetic resonance spectroscopic imaging with oral intake or intravenous infusion of nonradioactive ²H-labeled substrates to generate three-dimensional metabolic maps. DMI can reveal glucose metabolism beyond mere uptake and can be used with other ²H-labeled substrates as well. We demonstrate DMI by mapping metabolism in the brain and liver of animal models and human subjects using [6,6'-²H₂]glucose or [²H₃]acetate. In a rat glioma model, DMI revealed pronounced metabolic differences between normal brain and tumor tissue, with high-contrast metabolic maps depicting the Warburg effect. We observed similar metabolic patterns and image contrast in two patients with a high-grade brain tumor after oral intake of ²H-labeled glucose. Further, DMI used in rat and human livers showed [6,6'-²H₂]glucose stored as labeled glycogen. DMI is a versatile, robust, and easy-to-implement technique that requires minimal modifications to existing clinical magnetic resonance imaging scanners. DMI has great potential to become a widespread method for metabolic imaging in both (pre)clinical research and the clinic.

Advances in Imaging Brain Metabolism

Metabolism is central to neuroimaging because it can reveal pathways by which neuronal and glial cells use nutrients to fuel their growth and function. We focus on advanced magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) methods used in brain metabolic studies. ¹⁷O-MRS and ³¹P-MRS, respectively, provide rates of oxygen use and ATP synthesis inside mitochondria, whereas ¹⁹F-MRS enables measurement of cytosolic glucose metabolism. Calibrated functional MRI (fMRI), an advanced form of fMRI that uses contrast generated by deoxyhemoglobin, provides maps of oxygen use that track neuronal firing across brain regions. ¹³C-MRS is the only noninvasive method of measuring both glutamatergic neurotransmission and cell-specific energetics with signaling and nonsignaling purposes. Novel MRI contrasts, arising from endogenous diamagnetic agents and exogenous paramagnetic agents, permit pH imaging of glioma. Overall, these magnetic resonance methods for imaging brain metabolism demonstrate translational potential to better understand brain disorders and guide diagnosis and treatment.