Sleep Dependent Changes of Lactate Concentration in Human Brain

Lactate is an important cellular metabolite that is present at high concentrations in the brain, both within cells and in the extracellular space between cells. Small animal studies demonstrated high extracellular concentrations of lactate during wakefulness with reductions during sleep and/or anesthesia with a recent study suggesting the glymphatic activity as the mechanism for the reduction of lactate concentrations. We have recently developed a rigorous non-invasive imaging approach combining simultaneous magnetic resonance spectroscopy (MRS) and polysomnography (PSG) measurements, and here, we present the first in-vivo evaluation of brain lactate levels during sleep-wake cycles in young healthy humans. First, we collected single voxel proton MRS (1H-MRS) data at the posterior cingulate with high temporal resolution (every 7.5 sec), and simultaneously recorded PSG data while temporally registering with 1H-MRS time-series. Second, we evaluated PSG data in 30 s epochs, and classified into four stages Wake (W), Non-REM sleep stage 1 (N1), Non-REM sleep stage 2 (N2), and Non-REM sleep stage 3 (N3). Third, we determined lactate signal intensity from each 7.5-s spectrum, normalized to corresponding water signal, and averaged over 30-s for each PSG epoch. In examinations of nine healthy participants (four females, five males; mean age 24.2 (±2; SD) years; age range: 21-27 years) undergoing up to 3-hour simultaneous MRS/PSG recordings, we observed a group mean reduction of [4.9 ± 4.9] % in N1, [10.4 ± 5.2] % in N2, and [24.0 ± 5.8] % in N3 when compared to W. Our finding is consistent with more than 70 years of invasive lactate measurements from small animal studies. In addition, reduced brain lactate was accompanied by a significant reduction the apparent diffusion coefficient of brain lactate. Taken together, these findings are consistent with the loss of lactate from the extracellular space during sleep while suggesting lactate metabolism is altered and/or lactate clearance via glymphatic exchange is increased during sleep.Significance StatementThis study describes a non-invasive magnetic resonance spectroscopy/polysomnography technique that allows rigorous measurement of brain metabolite levels together with simultaneous characterization of brain arousal state as either wakeful or one of the several sleep states. The results provide the first in-vivo demonstration of reductions in brain lactate concentration and diffusivity during sleep versus wakefulness in young healthy human brain. These findings are consistent with invasive small-animal studies showing the loss of extracellular lactate during sleep, and support the notion of altered lactate metabolism and/or increased glymphatic activity in sleeping human brain.

252 Non-Invasive Quantification of Human Brain Lactate Concentrations Across Sleep-Wake Cycles

Introduction Cellular mechanisms underlying changes in small animal brain lactate concentrations have been investigated for more than 70 years and report sharp reductions in lactate (12-35%) during sleep or anesthesia relative to wakefulness. The goal of this study was to investigate alterations in human cerebral lactate concentrations across sleep-wake cycles. Toward this goal, we developed a novel non-invasive methodology, quantified changes in human cerebral lactate during sleep stages, and investigated potential mechanisms associated with changes in lactate. Methods Nine subjects (four females, five males; 21-27 y-o, mean age 24.2 ±2) were sleep deprived overnight, and underwent (5:45~11:00 am) experiments combining simultaneous MR-spectroscopy (MRS) and polysomnography (PSG) in a 3 T MR instrument using a 64-channel head/neck coil. A single voxel MRS (1H-MRS) acquired signals from a volume of interest (12~24 cm3) for every 7.5-s for 88~180-min. Lactate signal intensity was determined from each 7.5-s spectrum, normalized to corresponding water signal, and averaged over 30-s for each PSG epochs. Artifact corrected PSG data were scored for each 30-s epoch using the standard criteria and classified into one of four stages: W, N1, N2 and N3. Group mean lactate levels were quantified using LCModel. Three subjects returned for lactate diffusivity measurements using diffusion-sensitized PRESS MRS sequence. Results Compared to W, group mean lactate levels within each sleep stage showed a reduction of [4.9 ± 4.9] % in N1, [10.4 ± 5.2] % in N2, and [24.0 ± 5.8] % in N3. We observed a significant decrease in lactate apparent diffusion coefficient (ADC) accompanied by reduced brain lactate in sleep compared to wake (P<0.002). There were no differences in ADC values between wake and sleep for H2O, NAA, tCr, or Cho. Conclusion This is the first in-vivo report of alterations in human brain lactate concentrations across sleep-wake cycles. Observed decline in lactate levels during sleep compared to wakefulness is consistent with, and extends results from invasive small animal brain studies first reported more than 70 years ago, and support the notion of altered lactate metabolism and/or increased glymphatic activity in sleeping human brain. Support (if any) The Paul. G. Allen Family Foundation funded the study.

Systematic analysis of brain lactate and pH levels in 65 animal models related to neuropsychiatric conditions

Altered brain energy metabolism associated with increase in lactate levels and the resultant decrease in pH have been increasingly implicated in multiple neuropsychiatric disorders, such as schizophrenia, bipolar disorder, autism spectrum disorder and neurodegenerative disorders. Although it is controversial, change of pH/ lactate level as a primary feature of these diseases, rather than a result of confounding factors such as medication and agonal state, has been evidenced. Animal models that can be studied without such confounding factors inherent to humans are a suitable alternative to understand the controversy. However, the knowledge in animal models regarding brain pH and lactate and their relation to behavioral outcomes is limited in the context of neuropsychiatric disease conditions. In this study, we investigated the common occurrence of changes in the pH and lactate levels in the brain in animal models by analyzing 65 animal models related to neuropsychiatric and neurodegenerative diseases with 1,239 animals. Additionally, we evaluated the behavioral phenotypes relative to the chemical changes in the brain. Among the models, 27 and 24 had significant changes in brain pH and lactate levels, respectively, including Shank2 KO mice, Clock mutant mice, serotonin transporter KO mice, mice with a paternal duplication of human chromosome 15q11-13, Fmr1 KO mice, BTBR mice, APP-J20 Tg mice, social defeat stress-exposed mice, corticosterone-treated mice, and streptozotocin-induced diabetic mice. Meta-analysis of the data revealed a highly significant negative correlation between brain pH and lactate levels, suggestive of increased lactate levels causing decreased brain pH. Statistical learning algorithm based on the comprehensive data has revealed that the increased brain lactate levels can be predominantly predicted by the indices for the percentage of correct response in working memory test, with a significant simple, negative correlation. Our results suggest that brain energy metabolism is commonly altered in many animal models of neuropsychiatric and neurodegenerative diseases, which may be associated with working memory performance. We consider our study to be an essential step suggesting that the brain endophenotypes serve as a basis for the transdiagnostic characterization of the biologically heterogeneous and debilitating cognitive illnesses. Based on these results, we are openly accepting collaborations to extend these findings and to test the hypotheses generated in this study using more animal models. We welcome any mice/rat models of diseases with or without any behavioral phenotypes.