The brain accounts for about 2 % of our body weight (about 1400 g) but consumes up to 20 % of our total energy requirement. Only due to sufficient uptake of protein- and sugar-rich food was our brain able to develop into a high-performance organ during the evolution of man. Functions of our brain directly correlate with a sufficient supply of oxygen and nutrient-rich food. These substances are transported via the vasculature, which supplies the brain with sufficient blood and ensures that all nutrients are distributed to different regions of the brain. The circulation system of our brain consists of about 600 km of vessels and almost accounts for one third of the total brain volume. Still, blood capillaries are unable to reach all cells in the brain, which is made up of hundred billion nerve cells (neurons) and at least the same number of glia cells. Every single cell needs to be supplied with enough energy to exert its function without impairment. This is an enormous logistic challenge, which requires a complex network of distribution and utilisation of energy-rich substrates. This logistic task comprises regulation of the cerebral blood flow and efficient transport and utilisation of substrates by all cells and their anatomic specialisations like synapses and nerve fibres. How does our brain cope with this enormous task?
Is task sharing the secret of energy metabolism?
Brain cells are divided into nerve cells and glial cells, of which there are distinct types and many different ”task bearer”. Neurons consume a considerably higher amount of energy than glial cells because functional tasks of synapses, the chemical-electric links between neurons consume a lot of energy. Conduction of electric impulses along nerve fibres also requires a considerable amount of energy, not because one single impulse needs a lot of energy but because of the high number of impulses, which are fired simultaneously to send information from one part of our brain to the other. There are small electric currents at both synapses and nerve fibres, which are generated by ions and which flow along their electrochemical gradient. It is especially the sodium gradient across the cell membrane, which is used energetically to take up the neurotransmitters, which are released at synapses between neurons, in order to transmit certain information. In order to maintain the ion gradients, especially sodium, potassium, hydrogen, calcium, chloride and hydrogencarbonate (bicarbonate) have to be transported against their gradient so that the initial ion distribution can be restored (”ionic homeostasis”). It is this transport of ions against their gradient requires most of the energy in our brain – and also in most other organs. Ultimately, most of the energy, which is stored as ATP, is consumed by ATP-cleaving transport enzymes, notably the ”sodium-potassium pump” (also known as Na-K-ATPase or Na-K-pump), and the calcium-pump (or Ca-ATPase). The Na-K-pump in the cell membrane escorts sodium ions against their electrochemical gradient out of the cell and exchanges them with potassium ions (three sodium ions per two potassium ions). This process is driven by a complex cycle, which requires cleavage of ATP. In contrast to sodium, calcium is not only pumped out of the cell, but also into intracellular stores (endoplasmatic reticulum, mitochondria) against large gradients, often in exchange with protons (Deitmer and Rose, 2010).
Glial cells aid neurons to maintain the ionic homeostasis, take up neurotransmitters from the synaptic cleft and, probably in a unique manner, supply neurons with energy-rich substrates. Based on histological findings, Camillo Golgi (Pavia, Italy) already proposed at the end of the 19th century that glial cells provide nutrimental support to neurons. In this article we want to shed light on some aspects of this metabolic interaction of nerve- and glial cells.
Sugar and lactate are important energy substrates in our brain
The systemic blood supply provides the two most important energy substrates for our brain: glucose and oxygen. In the presence of oxygen, glucose can efficiently be metabolised into energy in form of ATP (”oxidative phosphorylation”). Glucose is transported from blood into brain tissue via glucose transporters (GLUT) in blood endothelial cells. Since those GLUTs merely favour transport down the chemical gradient of glucose (”facilitating transport”) glucose concentration in brain tissue is never higher than in blood, where it is stabilised at approximately 5 mM. In fact, the glucose concentration in brain tissue is a lot lower, usually between 1–2 mM, because it is rapidly taken up into cells by more GLUTs located in the cell membrane of nerve- and glial cells. Once taken up, glucose gets immediately phosphorylated (to glucose-6-phosphate) so that there is a constant gradient of glucose from both blood to brain tissue and also from there to the cell interior. This ensures a constant uptake of glucose into the brain and into cells. Thus, astrocytes, which cover more than 99 % of all blood capillaries with their processes (cell processes, which extensively cover blood vessels), seem to have the best access to glucose, which is provided by blood vessels and take it up via GLUT1. Neurons take up glucose from brain tissue via GLUT3. After being phosphorylated, glucose can directly be fed into glycolysis (cleavage of glucose) inside the cell (Fig. 1). In astrocytes, possibly also in a few neurons, glucose can be used to generate energy reservoirs in form of glycogen. Glycogen is a large molecule, which consists of thousands of interlinked glucose molecules. In the case of sugar deprivation in the brain, glycogen can be degraded to glucose-6-phosphate and then fed into glycolysis.
During glycolysis, enzymatic processes take place, which are extremely pH-dependent and directly regulated by H+, especially processes in which phosphofructokinase is involved (PFK, Fig. 1). End product of glycolysis is pyruvate, a molecule with three C-atoms, which means that one glucose molecule gets metabolised into two pyruvate molecules during glycolysis. Pyruvate is taken up into mitochondria, where it undergoes more enzymatic steps, which make it accessible for oxidative phosphorylation, which, while using oxygen, produces most of energy-rich ATP. While the net yield of glycolysis are 2 ATP-molecules, it is 34 by oxidative phosphorylation.
Instead of being transported into mitochondria, pyruvate can also be metabolised into a very similar substance in the cytosol, namely lactate, the anion of lactic acid. Lactate is also a molecule carrying three C-atoms and, similar to pyruvate, it can be metabolised into 16–17 ATP-molecules (this would mean up to 34 ATP-molecules per one glucose molecule). The conversion of pyruvate into lactate occurs in a single reversible enzymatic step with the help of lactate dehydrogenase (LDH). This enzyme exists in several isoforms, which determine whether pyruvate is metabolised into lactate or lactate into pyruvate. This is an important criterion in order to explain if and why some cells accumulate lactate also under aerobic conditions and release it to adjacent cells. This transfer of lactate from one cell to the other seems to play an important role in energy distribution in general and is not restricted to the brain. Based on the astrocyte-neuron-lactate-shuttle (ANLS)-hypothesis (Brooks, 2009; Magistretti et al., 1999), the glucose in astrocytes is mainly degraded to pyruvate, which in turn is converted into lactate by LDH-5 and released to neurons (Fig. 2). In neurons, lactate is converted by LDH-1 into pyruvate again which is fed into oxidative phosphorylation for further energy production. The processes involved in lactate transfer shall now be discussed in more detail.
The ANLS-hypothesis and lactate transport
Recent findings largely agree that neurons consume more energy than glial cells (Harris and Atwell, 2012). Therefore, it has long been assumed that they also consume most of the glucose. However, there has been a discussion for about 20 years now about whether neurons metabolise glucose directly to suffice their energy demands or if they partly utilise lactate, which is provided by astrocytes. Although the ANLS-hypothesis (Fig. 2) is not supported by everyone (Dienel, 2014), there has been growing evidence over the last few years, which supports this hypothesis (summarised in Magistretti and Allaman, 2015; Barros and Deitmer, 2010). The most important evidence is (1) astrocytes have more direct access to nutrients provided by the blood, including glucose; (2) astrocytes are equipped with the LDH-isoform 5, which converts pyruvate into lactate whereas neurons mainly have LDH-isoform 1, which converts lactate into pyruvate; (3) cell-dependent expression of various isoforms of lactate transporters in astrocytes and neurons favour transfer of lactate from astrocytes to neurons (more about these transporters below); (4) lactate can maintain or restore synaptic and other neuronal functions in the absence of glucose (Schurr et al., 1988). The advantage for neurons in taking up lactate, converting it into pyruvate and metabolising it inside the mitochondria would be that they could at least partly bypass the lavish glycolysis. Large amounts of ATP are produced upon oxidative phosphorylation and not upon glycolysis, as long as oxygen supply is secured. Astrocytes on the other hand could cover their low energy demand largely by performing glycolysis. Further, energy production from lactate allows neurons to not “burn” some of the consumed glucose for glycolysis, but to use it to produce antioxidants and basic substances for biosynthesis via the pentose phosphate pathway (Fig. 1).
It seems confirmed that both cell types, neurons and astrocytes, perform both glycolysis and oxidative phosphorylation but, according to the ANLS-hypothesis, they perform it to different extents; neurons mainly operate oxidative and astrocytes mainly glycolytic. The exact balance can vary between the numerous subtypes of neurons and astrocytes, and also between different regions of the brain, more or less oxidative neurons and more or less glycolytic astrocytes with glycolytic neurons and oxidative astrocytes being rather exceptions.
According to the ANLS-hypothesis, transport of lactate across the cell membrane, out of astrocytes and into neurons, is critical. This transport is performed by the three isoforms of the monocarboxylate transporterfamily (MCT), MCT1, MCT2, and MCT4, which have been extensively studied by our group for the last 20 years (Deitmer et al., 2015). The MCT family (SLC16) comprises 14 isoforms, of which MCT1-4 transport monocarboxylate anions 1:1 with H+ (Fig. 3A). Thus, this transport is electroneutral, i. e. it does not generate any current across the cell membrane. Chaperone proteins, namely basigin (CD147) and embigin (GP70) support integration of these MCTs into the cell membrane and their transport activity.
At first, transport properties of MCTs for lactate and other substances needed to be characterised, which we did by heterologous expression of MCT isoforms in frog egg cells (oocytes of Xenopus laevis) in cooperation with Stefan Bröer (Bröer et al., 1998, 1999; Dimmer et al., 2000). In addition to that, several research groups examined the importance of MCTs for the energy metabolism in the brain. Based on their findings, the ubiquitous lactate transporter in almost all tissues and organs is MCT1 with an average affinity for substrates (Km for lactate 3–5mM). MCT2 is a highly affine lactate transporter (Km for lactate 0.5–1mM) and is expressed in the brain mainly, possibly exclusively, by neurons located primarily in postsynaptic regions. MCT4 is a lactate transporter with low affinity (Km for lactate 17–35mM) and high capacity, which is expressed in the brain mostly by astrocytes. Km values suggest that MCT1 can be responsible for both import and export of lactate into and out of cells, while uptake of lactate could occur mainly via MCT2 and release rather via MCT4.
When looking at the transport of lactate in the brain, both from blood into the brain parenchyma and from astrocytes into neurons, lactate could be transported across endothelial cells of blood capillaries via MCT1 both from blood into the brain and also vice versa. MCT2 in neurons would favour uptake of lactate, where as MCT1 and MCT4 in astrocytes allow release and uptake of lactate.
pH and carbonic anhydrases play important in MCT transport activity
Based on co-transport of organic anions with protons, pH in the tissue, inside the cells, and also the H+ gradient across the cell membrane play important regulatory for the activity of MCT1-4. Thus, extracellular acidification favours lactate transport into the cell, whereas transport out of the cell can often only occur against the H+ gradient, hence requires a robust lactate gradient from the inside to the outside. Regulation of cytosolic pH plays an important role for the activity of MCTs and also for all H+-dependent transport processes. Almost all cell types require a sodium-proton-exchange (Na+/H+) for these processes, supported by several HCO3-dependent transporters. An essential transporter in glial cells is the electrogenic sodium-bicarbonate (Na+-HCO3 -) co-transporter (NBCe1, SLC4A4), which transports 2 HCO3 -with 1 Na+, dependent on the membrane potential, in both directions across the cell membrane. In cortical astrocytes we noticed a paramount involvement of NBCe1 in cytosolic pH-regulation and pH-buffering (Theparambil et al., 2014; Theparambil and Deitmer, 2015) as, dependent on the electrochemical gradient for this carrier, NBCe1 transports HCO3 - highly efficiently into or out of the cells. We were able to demonstrate that NBCe1, when it is co-expressed in Xenopus oocytes, increases lactate transport by MCT1 (Becker et al., 2004) and modulates glycolysis in astrocytes (Rumino et al., 2011).
H+ buffer capacity is a measure in order to avoid rapid accumulation of protons during the transport of lactate on either one or the other side of the cell membrane. Since there is only a small amount of H+ in form of free ions in the cytosol (30–100nM), but they are rather bound (10–30mM), also as intracellular mobile H+buffer molecules (HB), which undergo slow diffusion, local accumulation of H+/HB can occur quickly. Such local accumulation in turn inhibits further transport of H+ and lactate, which would be increasingly reduced by the low buffer capacity. In contrast, all factors, which increase H+ buffer capacity also increase the capacity of the lactate transporter. An important factor for the buffer capacity in astrocytes is the rapid transport of HCO3 - via NBCe1. At this point carbonic anhydrases cooperate with transport activities of MCTs in a unique manner (Becker et al., 2011, 2014).
Carbonic anhydrases (CA) exist in numerous isoforms in almost all cells and organisms and catalyse the reversible hydration of carbon dioxide (CO2), one of the most important biochemical reactions (CO2 + H2O ←→ H+ + HCO3 -). In humans, there are 15 CA-isoforms, of which some show cytosolic activity on and inside cell organelles, and some are active inside the cell membrane and outside of the cell. Apparently, accelerating the reaction time plays a very important role in vital functions like respiration, renal function and in many epithelial tissues as well as in the brain, where mainly CA-isoforms II, IV, XII and XIV exist.
Co-expression of MCT1 and carbonic anhydrase isoform II (CAII), the most active and most common isoform, which exists in almost all cells in the cytosol, in Xenopus oocytes, surprisingly resulted in a significant increase of lactate transport. Transport rate was increased 2–3 fold and even inhibition of the catalytic activity of CAII was not able to diminish this increase. An enzymatically inactive CAII mutant lead to an increased MCT1 transport rate, similar to wildtype CAII; herewith we were able to demonstrate a novel non-enzymatic function of CA (Becker et al., 2011, 2014). Next, we investigated the specificity of the interaction between MCT and CA and found that also MCT2 and MCT4 interact with some but not with all of the analysed CA isoforms. We noticed that these interactions are isoform-specific; therefore, CAII interacts with MCT1 and MCT4, but not with MCT2, whereas extracellular CAIV and CAIX increase transport activity of all three MCT-isoforms, and CAI and CAIII interact with none of the analysed MCT-isoforms.
While elucidating mechanisms of this functional “transport metabolon” (Becker et al., 2014; Deitmer et al., 2015), it was found that CA binds to MCT and thereby acts as kind of a “proton antenna” and increases the local H+ buffer capacity (Fig. 3B) (Becker et al., 2011; Noor et al., 2015; Deitmer et al., 2015). We then showed that this non-enzymatic interaction between MCT and CA not only occurs in heterologous expression systems (frog oocyte), but also in murine astrocytes (Stridh et al., 2012).
The global energy supply for the brain, and also at the cellular level, are complex systems consisting of metabolic and transport processes, which are tightly regulated. The pH in the brain parenchyma as well as inside cells plays a critical role for these processes. We have only just started to understand the protein networks, which enable efficient distribution of nutrients and energy carriers. Since many neurological disorders are directly or indirectly linked with the energy supply of cells and whole brain areas, metabolite transport, metabolic pathways and pH regulation could represent potential targets for therapeutic drugs. We yet have to understand the described processes in more detail, especially to what extent they occur in different brain regions, and how targeted interventions into these processes are possible. It seems certain that energy metabolism and energy distribution in the brain will be of preeminent interest in the field of neurobiology and clinical neurology in the future.
Becker, H., Bröer, S., Deitmer, J.W. (2004). Facilitated lactate transport by MCT1 when coexpressed with the sodium bicarbonate cotransporter (NBC) in Xenopus oocytes. Biophys. J. 86, 235–247. Google Scholar
Becker, H.M., Klier, M., Schüler, C., McKenna, R., Deitmer, J.W. (2011). The intramolecular proton shuttle supports not only catalytic, but also non-catalytic function of carbonic anhydrase II. Proc. Natl. Acad. Sci. U.S.A. 108, 3071–3076. Google Scholar
Becker, H.M., Klier, M., Deitmer, J.W. (2014). Carbonic anhydrase and their interplay with acid/base-coupled membrane transporters. In: Carbonic Anhydrase: Mechanism, Regulation, Links to Disease and Industrial Applications. (eds. Susan Frost, Robert McKenna) Subcell. Biochem. 75, 105–134. Google Scholar
Bröer, S., Schneider, H., Bröer, A., Deitmer, J.W. (1998). Characterization of the monocarboxylate transporter 1 expressed in Xenopus laevis oocytes by changes in cytosolic pH. Biochem. J. 174, 167–174. Google Scholar
Bröer, S., Bröer, A., Schneider, H., Stegen, C., Halestrap, A.P., Deitmer, J.W. (1999). Characterization of the high-affinity monocarboxylate transporter MCT2 in Xenopus laevis oocytes. Biochem. J. 535, 529–535. Google Scholar
Deitmer, J.W., Rose, C.R. (2010). Ion changes and signaling perisynaptic glia. Brain Res. Rev. 63, 113–129. Google Scholar
Deitmer, J.W., Theparambil, S.M., Ruminot, I., Becker, H.M. (2015). The role of membrane acid/base transporters and carbonic anhydrases for cellular pH and metabolic processes. Front. Neurosci. 8: 430, doi: . CrossrefWeb of ScienceGoogle Scholar
Dienel, G.A. (2014). Lactate shuttling and lactate use as fuel after traumatic brain injury: metabolic considerations. J. Cereb. Blood Flow Metab. 34, 1736–1748. Google Scholar
Dimmer, K.S., Friedrich, B., Lang, F., Deitmer, J.W., Bröer, S. (2000). The low-affinity monocarboxylate transporter MCT4 is adapted to the export of lactate in highly glycolytic cells. Biochem. J. 350, 219–227. Google Scholar
Golgi, C. (1882–1883). Sulla fine anatomia degli organi centrali del sisterna nervoso.Google Scholar
Jakoby, P., Schmidt, E., Ruminot, I., Gutierrez, R., Barros, L.F., Deitmer, J.W. (2014). Higher transport and metabolism of glucose in astrocytes compared to neurons: A multiphoton study of hippocampal and cerebellar tissue slices. Cereb Cortex 24, 222–231.Web of ScienceGoogle Scholar
Magistretti, P.J., Pellerin, L., Rothman, D.L., Shulman, R.G. (1999). Energy on demand. Science 283, 496–497. Google Scholar
Noor, S.I., Dietz, S., Heidtmann, H., Boone, C.D., McKenna, R., Deitmer, J.W., Becker, H.M. (2015). Analysis of the Binding Moiety Mediating the Interaction between Monocarboxylate Transporters and Carbonic Anhydrase II. J. Biol. Chem. 290, 4476–4486. Web of ScienceGoogle Scholar
Ruminot, I., Gutiérrez, R., Peña-Münzenmayer, G., Añazco, C., Sotelo-Hitschfeld, T., Lerchundi, R., Niemeyer, M.I., Shull, G.E., Barros, L.F. (2011). NBCe1 mediates the acute stimulation of astrocytic glycolysis by extracellular K+. J. Neurosci. 31, 14264–71. Google Scholar
Schurr, A., West, C.A., Rigor, B.M. (1988). Lactate-supported synaptic function in the rat hippocampal slice preparation. Science 240, 1326–1328. Google Scholar
Stridh, M., Alt, M., Wittmann, S., Heidtmann, H., Aggarwal, A., Riederer, B., Seidler, U., Wennemuth, G., McKenna, R., Deitmer, J.W., Becker, H.M. (2012). Lactate fluxes in astrocytes enhanced by a non-catalytic action of carbonic anhydrase II. J. Physiol. 590.10, 2333–2351. Google Scholar
Theparambil, S.M., Ruminot, I., Schneider, H.P., Shull, G.E., Deitmer, J.W. (2014). The electrogenic sodium-bicarbonate cotransporter NBCe1 is a high-affinity bicarbonate carrier in mouse cortical astrocytes. J. Neurosci. 34, 1148–1157. Google Scholar
Theparambil, S.M., Deitmer, J.W. (2015). High effective cytosolic H+ buffering in mouse cortical astrocytes attributable to fast bicarbonate transport. Glia 63, 1581–1594. Web of ScienceGoogle Scholar
About the article
Joachim W. Deitmer
Joachim W. Deitmer is senior professor at the faculty of biology of TU Kaiserslautern. Between 1989 and 2015 he was head of the department of general zoology at the TU. His research focused on membrane transport of lactate, bicarbonate and protons as well as calcium, pH and bicarbonate regulation and signalling in glia cells. He especially focused on monocarboxylate transporters (MCT), sodium-bicarbonate co-transporter (NBC) and carbonic anhydrases.
Shefeeq M. Theparambil
Shefeeq M. Theparambil holds a Msc in biophysics (Bangalore, India), started his PhD in Deitmer`s lab in 2010 and worked there as a postdoc between 2014 and 2016. His research focuses on regulation of pH and bicarbonate in astrocytes and their functional impact on metabolism.
Iván Ruminot received his PhD in L.F. Barros` lab and after that worked as a postdoc in Deitmer`s lab between 2012 and 2015. He was working on functional interactions between pH and lactate as well as glucose metabolism in astrocytes. He now works as junior group leader at the CECS in Valdivia, Chile.
Holger M. Becker
Holger M. Becker received his PhD in Deitmer`s lab in 2005; after a few years of postdoc he became junior professor at the faculty of biology at the TU Kaiserslautern and habilitated there 2015. His research focused on functional protein interactions between monocarboxylate transporters and carbonic anhydrases and their effects on the energy metabolism, first in Xenopus oocytes and since a few years also in tumour cells.
Published Online: 2017-02-10
Published in Print: 2017-02-01