1. IntroductionBrain function is dependent on the oxidative metabolism of glucose. Therefore, a human adult brain weighing approximately 2% (1400 g) of the body weight (70 kg) is responsible for 25% of the total glucose consumption of the whole body [1,2]. Importantly, most glucose is oxidized by oxygen to generate carbon dioxide (CO2) and adenosine triphosphate (ATP). This high rate of oxidative metabolism of glucose cannot be replaced by any other energy substrates under normal physiological conditions [3,4,5]. Because glucose and oxygen are not stored in brain tissue, these essential substrates must be supplied by cerebral blood flow, and the cessation of a continuous supply causes irreversible cell damage within a short period of time [6,7,8]. Efficient ATP production is the basis of the generation of action potentials, and normal mitochondrial function in neurons is essential for this purpose. Unfortunately, however, high rates of oxidative metabolism of glucose always generate reactive oxygen species (ROSs) at a certain rate: 0.1–0.2% of the oxygen that is consumed is converted to ROSs [3,9]. In addition to mitochondria, peroxisomes are thought to be another important source of ROSs. Peroxisome is an organelle where very long chain fatty acids (>C22) are metabolized [10,11]. Although ROSs act as important signal molecules, in general, ROSs are regarded to play detrimental roles in cell injury [9,12,13,14]. Thus, several intrinsic protective mechanisms operate to reduce the toxic effects of ROSs in the brain. The dysfunction of these intrinsic mechanisms by which the brain eliminates ROSs leads to both acute [9,12,13,14] and chronic [15,16] neuronal damage. In ischemic stroke, for example, the elimination of ROSs during reperfusion therapy by either pharmacological thrombolysis or a mechanical thrombectomy is an essential therapeutic strategy, since damaged mitochondria during ischemia serve as a potential source of massive ROS production, especially after reperfusion. Thus, the intravenous administration of edaravone, a free radical scavenger, has been an established therapy for over 20 years [17,18]. Reactive oxygen species also play important roles in triggering both cytotoxic and vasogenic edema. Even with edaravone administration, formation of brain edema is sometimes unavoidable after recanalization therapy and massive brain edema results in death in the acute phase of cerebral infarction. Initial ischemia-induced cytotoxic edema, which is a reflection of energy failure, is followed by vasogenic edema that reflects an increased permeability of blood–brain barrier. In addition to ROS elimination, the development of effective treatment of brain edema is imperative [12,13,14,19,20,21,22]. Likewise, accumulated cellular damage caused by chronic ROS production might also be a potential mechanism of age-related neuronal damage, i.e., neurodegenerative diseases including Parkinson’s disease, Alzheimer’s disease, and amyotrophic lateral sclerosis (ALS). Recently, edaravone has also been added as a therapeutic option for the treatment of ALS [17,18]. Thus, enhancement of the intrinsic mechanism of ROS elimination seems to be a promising therapeutic strategy.The high rate of glucose consumption in the brain is thought to mainly reflect neuronal glucose utilization. It is of note, however, that glucose consumption by glial cells, which outnumber neurons by a factor of 10, is not negligible, even though glial cells do not generate action potentials [23,24]. Among glial cells, astrocytes (or astroglia) are the most abundant glial cells in the brain, and their anatomical location allows astroglia to take up glucose directly from capillaries [23,24]. Astrocytes are interposed between neurons and capillaries. Astrocytic endfeet, therefore, play pivotal roles in the initiation of vasogenic edema through their water channel protein aquaporin-4 (AQP4) [21,22]. Neuronal synapses are enveloped by astrocytic endfeet to create tripartite synapses, and 99% of the capillary surface is also covered by the endfeet of astrocytes. In vitro data obtained using cultured rodent astroglia demonstrated that glucose utilization in astroglia is comparable with that in neurons, or even higher [3,4,5]. Importantly, glucose consumption by astrocytes may not consist of complete oxidation to generate CO2. In fact, astrocytes produce huge amounts of lactate even under a sufficient supply of oxygen (aerobic glycolysis). Their high glucose consumption and high glycolytic activity leads to lactate production, and lactate, in turn, is transferred to neurons to serve as a tricarboxylic acid (TCA) cycle substrate instead of glucose (astrocyte–neuron lactate shuttle model) [25,26]. Regardless of the long-lasting debate over this model, it is now widely accepted as an example of metabolic compartmentalization between astroglia and neurons [27,28,29,30,31,32,33]. One of the benefits of using lactate instead of glucose is that lactate enters directly into the TCA cycle after conversion to pyruvate, which generates 36 ATPs, while glucose needs to be metabolized to pyruvate through glycolysis and generates only 2 ATPs. Pyruvate is then transferred to mitochondria via the pyruvate dehydrogenase complex (PDHC) [1,2,3,4,5]. Both astrocytes and neurons possess monocarboxylate transporters (MCTs), and lactate is released from astrocytes and enters neurons through these transporters [3,4,5]. Besides lactate production, the astrocytic glycolytic pathway, per se, seems to have additional important roles [3,4,5]. More precisely, the pentose-phosphate pathway (PPP), a shunt pathway of glycolysis, plays a pivotal role in protecting neurons against oxidative stress [3,4,5,34,35,36]. This review will focus on the astrocytic PPP and its common beneficial roles in stroke and neurodegenerative diseases. Most of the data presented here were obtained using 14C-labeled tracer assays and cultured rodent astroglia and neurons (Figure 1) [3,4,5,37,38,39,40,41,42,43,44]. In addition to glucose metabolism, we also used 14C-labeled palmitic acid (long chain fatty acid; C16) and 14C-labeled β-hydroxybutyrate (BHB, a ketone body) to evaluate fatty acid oxidation and ketone body metabolism , respectively. We also discuss fatty acid metabolism in peroxisomes where very long chain fatty acids are metabolized.