|Home | About | Journals | Submit | Contact Us | Français|
Astrocytes are often referred to, and historically have been regarded as, support cells of the mammalian CNS. Work over the last decade suggests otherwise, that astrocytes may in fact play a more active role in higher neural processing than previously recognized. Because astrocytes can potentially serve as novel therapeutic targets, it is critical to understand how astrocytes execute their diverse supportive tasks while maintaining neuronal health. To that end, this review will focus on the supportive roles of astrocytes, a line of study relevant to essentially all acute and chronic neurological diseases. Furthermore, this review will critically re-evaluate our concepts of the functional properties of astrocytes and relate these tasks to their intricate morphology.
The human brain consists of about 1012 cells, of which around 1011 are considered to be neurons with the remaining 9 × 1011 being glia 1. Astroglia, often called astrocytes, are one of the major types of glial cell2. The brains of smaller mammalian creatures, naturally, have a smaller number of cells and the ratio of astrocytes to neurons also declines 3. The term astrocyte derives from a combination of the Latin word for stars (astra, singular astrum) and the word for cell (cyte, which is in turn derived from the Greek word kytos, meaning vessel). As can be seen in Fig.1., astrocytes as they were originally visualized in the late 19th and early 20th century by Golgi s reazione nera (Fig. 1A,B,C), do not look much like the stars as we see them, of course misleadingly, with the naked eye. These drawings of protoplasmic astrocytes are exactly the same as we see them now with modern techniques of visualization, such as by filling the cells with dye, as shown in Fig. 1.D. Some are rather like a tree s root ball or an ornamental bushy shrub. Some astroglia do not look even most fancifully look like stars but are elongate with a number of branches that retain the elongate appearance and orientate along the main axis of the cell (see Fig. 1B,C). However, the etymology of the word used to describe something is, after all, not that significant, although it is helpful if it has some connection with the thing itself as invoking some clear aspect of it and as an aide-mémoire. The key question is whether the definition of the word used precisely describes its characteristics accurately and we can all agree on them. So why are all the cells that are termed astroglia considered to be astroglia, and the corollary, what cells should be termed astroglia? The term glia, a term having its origin in the 1850s 4 is now quite misleading, and therefore unhelpful. If it means anything at all it refers to all the electrically non excitable (i.e. non neuronal) cells, but the only thing they have in common is that they are not neurons and therefore, by definition, non excitable. It is odd that in the experimental sciences tradition and history are so strong, otherwise we could more informatively replace neurons with excitable neural cells and glia with non excitable neural cells, neural referring to any cell of the CNS.
The objective of this review, for now, is to communicate to you, the reader, our view of what mature mammalian astrocytes are and what they most likely do. The latter, naturally, relates to the question of whether astrocytes could be useful or effective targets for drugs. It is important to point out upfront that most of our concepts regarding astrocytic functions are based on analysis of cultured astrocytes. Astrocytes in cultures are flat cells, resembling epithelioids, that differ fundamentally from astrocytes in situ. Many of the properties of astrocytes in situ are still debated, mainly because it is difficult to study these non-excitable cells, intermingled as they are with the multiple other cell types of the CNS, including neurons, oligodendrocytes, microglial cells, vascular cells, and more. Another problem is that we do not have a good functional endpoint by which we can assess astrocytic function. Moreover, a defining feature of astrocytes is that they are plastic cells that rapidly change in response to alterations in their environment. For example, astrocytes display reactive changes shortly after preparing brain slices and re-express nestin, a marker of immature glial cells, a mere few hours after the slices are prepared. Astrocytes express a large number of receptors, which are primarily G-protein linked, and respond to essentially all neurotransmitters, neuromodulators, and hormones by changes in cytosolic Ca2+ or cAMP. However, it is difficult to target astrocytes specifically, because most of the receptors are expressed by other cell types. We will here give a short update on current concepts with regard to the function of astrocytes and emphasize the functions that have been demonstrated definitively in the intact CNS.
Mature astrocytes come in three forms, the spherically bushy variant, the less bushy process-bearing form, and the elongated, non-excitable version. The former two are the protoplasmic astrocytes of grey matter (Figs 1A and 1D) and the fibrous astrocytes of white matter (Fig. 1C). The elongated cells are the Bergmann glia of the cerebellum (Fig. 1C.W) and the Muller cells of the retina (Fig. 1B, also see Table. 1). This description of mature astrocytes dates back to the first great period of neurohistology, being essentially unchanged since the latter decades of the nineteenth and the early part of the twentieth century5,6.
The illustrations in Figure 1 are all of rat brains, except figure 1A, which is an older drawing of a slice prepared from a human infant. Figure 2 shows protoplasmic astrocytes from the cerebral cortices of an adult mouse brain, a rhesus monkey, and human. The bushy morphology of the dye labeled cells is the same but it was found that the mouse astrocyte domain (soma plus processes) was about 2.5 times smaller than the human7. This was based on dye labeling, as well as GFAP staining which in the rodent only shows the soma and largest parts of the process most proximate to the soma (see Fig.1D). Also characteristic of the primate astrocytes are long, varicosity-bearing processes that extend beyond the domains of the bushy processes and extend between the cortical lamina, and have been suggested to optimize the columnar organization of the cerebral cortex of primates8 . Whether this is to effect some higher level of coordination within the cortical columns has not yet been, and will be, difficult to determine. The apparent larger size of the domains of the protoplasmic astrocytes of larger–brained primates could be, at its simplest, to limit the expansion of the astrocyte population; namely to control more synapses with fewer astrocyte soma.
Presumed astrocytic-specific proteins are used as markers to identify astrocytes, especially to identify cells recorded in situ after filling the cell with a dye present in the microelectrode to identify the recorded cell. This is now increasingly being supplemented by use of presumed astrocyte-specific promoters to drive synthesis of fluorescent proteins. This has the advantage that the cells to be studied can be preselected in living tissue. These techniques have recently been reviewed for neurons 9 and for astrocytes 10 . However, proteins are responsible, for most cellular activities and are therefore also subject to change. Similarly, promoter activation can also change. Protein markers that are generally used for astrocytes are the intermediate filament protein GFAP and the glutamate transporters named GLAST and GLT-1 for rats, and EAAT, after EAA for excitatory amino acids, 1 and 2 for human brain. There are other EAAT isoforms found on neurons. GFAP was the first astrocyte marker widely used and has stood the test of time well, such that all cells that robustly express GFAP are astrocytes. However, the converse does not appear to hold and it is recognized that there a number of other proteins and physiological properties that define astrocytes, so that one can have a GFAP(−) cell that one should call an astrocyte because it has these other properties. Of course, it also has to be excluded that it is not another non excitable neural cell like the NG2 positive cells, that were at one time considered to be smooth protoplasmic astrocytes based on their morphology and location in grey and white matter 11.
The water channel protein Aquaporin 4 is also a good marker for astrocytes and is predominantly localized in the perivascular membranes of protoplasmic astrocytes. This protein facilitates water transport but under what conditions it is needed is less clear12. Recent microarray studies for isolated astrocytes13, 14 has turned up some other good candidates, especially Aldh1L1 (aldehyde dehydrogenase 1 family, member L1), which was quite unexpected.
The passive electrical properties of mature astrocytes due to their highly selective leak K+ channels are a well-established characteristic of astrocytes which, after a confusing period when it was questioned due to results from primary cultures and recordings from immature astrocytes 5, has been reestablished and is now used as diagnostic of mature protoplasmic astrocytes in situ (e.g. ref 16,17). It has not been so clearly established for the other astrocytes. It confers on the cells a linear I-V curve with a reversal potential close to the K+ equilibrium potential and a very low membrane resistance of a few megohms that makes single electrode whole cell voltage clamping impossible, not in the sense of doing it, but in the interpretation of the results6,18,19. It also results in current not spreading further than closest (nearest neighbor) astrocytes20.
Other characteristics that are used to identify astrocytes are one or more processes touching a blood vessel (also see Fig.1A) and communication between up to 100 astrocytes by gap junctions located at the tips of joined processes (e.g.21).
A long-held principle in biology is that form and function are closely interrelated. It is often expressed as function follows form. Thus the large number of fine processes of especially the protoplasmic astrocytes and their endings on blood vessels and synapses (see Fig.1A, D) is an important jumping off place when considering astrocyte function. In Table 2, we summarize the basic support functions of astrocytes, defined as functions that allow neurons and the brain in general to function optimally (see 5,6 for more discussion of this issue). Some original references are noted to give the approximate time when the idea was first proposed. This is a difficult task, especially to correctly identify who really had priority and then to not give offence by omission. Many of the original suggestions are from cultures, which are correct in some generalities but unpredictably in error for specifics6. The later reviews can be consulted to sort out all the details. We will now cover the topics summarized in table 1, item by item.
The first dynamic studies on glial cells were electrophysiological, as this was the only cell-specific physiological technique that could be used in situ, and were done in the amphibian optic nerve in the mid 1960s. Only cell bodies could be impaled, and this was actually an advantage as all the cell bodies in the optic nerve are glia. Also amphibian tissue was easier to work with compared to mammalian tissue. Two important findings emerged from these studies. One was that these glia were found to be non-excitable and two, that they had very negative membrane potentials determined essentially exclusively by the transmembrane K+ gradient, 22. These amphibian optic glia were likely best described as astrocytes, but important later work in cats showed that the same characteristics applied to recorded astrocytes, identified histologically after recording, in the adult mammalian cortex 3. Concurrently, studies of cultured cells indicated that astrocytes have a much higher capacity for [K+]o uptake than neurons. Increasing [K+]o lead to a 50% increase in K+ content in astrocytes within seconds and to a doubling within 1–2 min 24.
It was proposed that the exclusive K+ permeability conferred on glia a function of maintaining a constant extracellular K+ concentration in the face of neuronal activity which would tend to increase it 25. The process was termed K+ spatial buffering and involves redistribution of increased [K+]o by a current loop set up by a membrane potential difference which is due to locally increased [K+]o , and serves to dissipate the increased [K+]o to distant sites; hence the term “spatial buffering”26,27. It was an ingenious speculation giving a function for the exclusive K+ conductance of the astroglial membrane and seemingly further supported by the organization of the cells into syncytia which could, in theory, carry the currents for long distances. However, the evidence was actually very indirect. A major critique of the model is that the current cannot spread, given the low resistance of the astrocyte18, in relation to the gradients of increased [K+]o which can be generated by neuronal activity. In fact, recent studies were not able to confirm that spatial buffering of [K+]o plays a role in neurovascular coupling in retina28, or that gap junctions contribute to K+ buffering in hippocampus29.
The clearance of increasing [K+]o due to neuronal activity may however not be limited to passive influx of K+ driven by its electrochemical gradient, but could include other K+ uptake mechanisms. These included K+ uptake by the Na+ /K+ pump, and co-transporters belonging to the Slc12a gene family. D Ambriosio and co-workers concluded that KIR channels are primarily responsible for regulation of baseline [K+]o , whereas the Na+ /K+ pump determines the rate of [K+]o recovery following excessive neuronal firing in hippocampal slices30. A similar conclusion was reach by Walz and Wuttke, which analysis of reactive astrocytes in hippocampus suggested that astrocytes limit increases in [K+]o by a combination of uptake via the Na+ /K+ pump and passive influx 31–32. A potential critique of the proposed role of the Na+ /K+ pump, is that the pump respond most effectively to changes in [Na+]i to pump out, whereas its outside K+ binding site is ~90% saturated at the normal [K+]o of 3 mM 31. Thus, clearance of [K+]o by the Na+ /K+ pump must be accompanied by increases in intracellular Na+, which is a possibility since several transporters and receptors, including GLT1, GT1, NMDA, and P2X receptors, facilitate Na+ influx and are activated during synaptic activity33,34,35,36.
Another proposed mechanism is channel-mediated uptake of increased [K+]o and Cl− driven by the Donnan potential 37. However, this mechanism requires Cl− channels in addition to K+ channels38, and there is little evidence for open Cl− channels under the mildly elevated [K+]o levels seen, for example, during normal neuronal activity39. However, reactive astrocytes have a significant resting Cl- conductance and passive uptake of K+ may contribute to K+ buffering in gliotic scares 31.
In conclusion, strong evidence obtained in in vitro systems support the concept that astrocytes are engineered to buffer the increases in extracellular K+ that may occur in connection with neural activity, whereas the [K+]o re-uptake mechanisms in neurons seem less efficient and slower. Intact brain also has a very high capacity for rapid normalizing of excess [K+]o, but it has been difficult to separate the roles of astrocytes versus neurons in [K+]o buffering. The pharmacological approaches available to study [K+]o buffering do not enable specific targeting a single cell type. Nevertheless, the very large number of membrane transporter expressed by astrocytes, which facilitate fluxes of ions across the plasma membrane suggest that astrocytes are heavily involved in regulation of extracellular ion homeostasis. However, [K+]o buffering is at this point not an attractive target for prevention of for example epilepsy or post-ischemic spreading depression waves, due to our poor understanding of the molecular mechanisms involved in [K+]o uptake.
An involvement of astrocytes in brain pH control mechanisms was first proposed based on a localization of carbonic anhydrase (Car) in astrocytes at the blood brain barrier, involving transport of HCO3− linked to acceleration of intra-astrocytic CO2 hydration 40 . Subsequent studies showed Na+/H+ and Cl−/HCO3− exchangers and Car activity in primary astrocyte cultures 41–43, and a model of coupled Na+/H+ and Cl−/HCO3− exchange was proposed to explain astrocytic swelling under pathological conditions. These studies were what now would be consider dated culture studies and this model is very difficult to test in situ. Other findings in primary astrocyte cultures are an electrogenic sodium proton co-transporter(3 HCO3− plus 2 Na+), which can acidify the extracellular space when stimulated by an increased K+-dependent depolarization and alkalinize the cell interior 44, and sodium dependent Cl−/HCO3− exchange, as well as the sodium independent exchanger mentioned above 45–47. All these are members of the bicarbonate transporter (BT) super family. Inspection of the transcriptosome from isolated astrocytes13 14 shows that many of these transporters (gene symbols starting with Slc), as well as other acid homeostasis related transporting systems, are preferentially expressed in astrocytes.
Another H+ transporting system is the lactate + H+ transporter, different isozymes of which are present in astrocytes and neurons. These systems are responsible for the efflux of lactate from astrocyte and uptake of lactate into neurons as proposed in the astrocyte neuronal lactate shuttle hypothesis (ANLSH) 48. Thus, lactate transport affects pH levels or, viewed as the converse, pH can affect lactate exchange and therefore link the ANLS to changes in pHe and pHi. Note that lactate transport is directly linked to pH because of the H+ being co-transported with lactate. The end product of glycolysis is the lactate anion and not lactic acid, so does not directly change pH, as is often incorrectly implied in the term lactic acidosis 49.
For some time astrocyte depolarization was not thought to occur by activation of ionotropic receptors, as an early study had shown that depolarization of astrocytes due to application of GABA was better explained by electrogenic uptake of GABA because no conductance changes could be measured, as expected if ionotropic astrocytic receptors were being activated. But this is an argument by omission and was not otherwise directly shown 50. It was similarly concluded from work on neuroglia in slices that depolarization due to added glutamate was because of raised [K]o due to neuronal activation 51. Thus, because of these and other studies, the view arose that astroglia lacked receptors, and only responded to changes in neuronal activity through depolarization of their membrane potentials in response to raised [K]o. This also simplified neuronal research, because receptor mediated effects of neurotransmitters in intact tissue could then be attributed only to neurons, so maybe it was accepted more readily than it should have been. The major problem, however, is that this view is wrong. It would be a very unusual cell that lacked receptors, for all cells have receptors to respond to their environments, albeit many of these are metabotropic rather than ionotropic. The more reasonable view is that astrocytes only express a limited number of ionotropic receptors which would make more biological sense as astrocytes, being non-excitable, do not need to change their membrane potentials for the same purposes as neurons. Later studies in primary astrocyte cultures showed that they had glutamate and GABA ionotropic receptors 52–53, as well as a variety of metabotropic receptors54.
However, the cultures may, because of alterations in their gene expression, express more receptors in culture than in situ, and the existence of ionotropic receptors in astrocytes in situ, is more limited 55. One of these exceptions is the AMPA type receptors of the Ca2+ permeable variety found on Bergmann glia in situ56. In these cells there are concentrations of vesicles in the boutons of parallel fibers abutting BG. These give fast inward currents in the BG due to release of glutamate from these vesicles, a process termed ectopic release. Other than a role in maintaining excitatory synapses on the Purkinje neurons for the Ca2+ influx associated with this activation57, other roles have not yet been described. A possible function could be an increase in intracellular Na, which would increase the activity of the Na/K ATPase and in turn lower [K]o since the low concentration of Na+ in many cells is the rate limiting step for the Na/K ATPase.
Specifically targeting astrocytic receptors is, perhaps with a few exceptions, not possible, since neurons and astrocytes express receptors for the same ligand. In fact, one characteristic of astrocytes is that their receptor expression mimics surrounding neurons. Often, neurons will express both ionotropic and metabotropic receptors, whereas astrocytes express primarily metabotropic receptors.
Uptake of the excitatory amino acid transmitter glutamate is arguably the best established and important property of mature protoplasmic astrocytes. Inactivation of glutamate then occurs by conversion to glutamine by the astrocyte-specific, intracellular enzyme, glutamine synthetase, consuming ATP and ammonia. Glutamine leaves the astrocyte, in the same region as it is normally taken up, namely the perisynaptic processes, by an amino acid carrier, and is taken up by neighboring neurons, where it is reconverted to glutamate via glutaminase 58.
The uptake of glutamate into a small second compartment was first proposed based on biochemical evidence that the specific activity (SA) of its product glutamine was greater than the SA of the injected radiolabeled glutamate, its immediate precursor58. Obviously if the injected radiolabeled glutamate had equilibrated with all the unlabeled glutamate present in the brain, the SA of its product should have decreased. By now a wealth of evidence has shown that specific EAA carriers are specifically expressed in astrocytes (also shown by high and specific levels of mRNA, see table 1) and that their knockdown leads to increased glutamate levels in the ECS 59.
Mature astrocytes in situ are also known to contain GABA transporters, specifically GAT1 and especially GAT3 60–62. However, these are not specific to astrocytes as they are also found in neurons. The metabolic consequences of GABA uptake can be quite varied 63. Also there are transporters for glycine especially in the glycine-rich posterior regions of the CNS. Of the two glycine transporters, GlyT-1 and 2, Glyt-1 predominates in astrocytes, and also seems to have the more crucial role in lowering glycine levels in the CNS (see 64 for a fuller account of the neurotransmitter transporters). Taurine is also taken up by cultured astrocytes, but the role for this transmitter, also thought to be have a cell osmolyte function in the brain because it is markedly released upon astrocytic swelling, is not well-elucidated (see 65).
The consequences of targeting astrocytic uptake of glutamate have already been studied using several approaches. For example, mice lacking the astrocyte specific glutamate transporter, GLT1, die within weeks after birth as a consequence of repeated seizures, suggesting that pharmacological manipulation of transmitter also represent a potent therapeutic target59. In fact, Rothstein and co-workers screened 1,040 FDA-approved drugs and nutritionals, and found that many beta-lactam antibiotics are potent stimulators of GLT1 expression. When used in an animal model of amyotrophic lateral sclerosis, the drug delayed loss of neurons increased mouse survival. These experiments provide an example on how known function of astrocytes can be manipulated and novel classes of potential neurotherapeutics can be useful to treat a fatal disease 66. Beta-lactam antibiotics have also shown to reduce the severity of experimental stroke although their mechanism of action are debated 67,68.
Astrocyte foot processes surround all blood vessels in the brain including the precapillary arterioles which are the basic regulators of blood flow 69, and flow effects have been shown to occur 70,71. It is necessary for the control to be exerted at the level of the arterioles as the capillaries lack the smooth muscle needed to cause contraction or relaxation to change the blood vessel diameter 72,73. The astrocytic ensheathment of CNS blood vessels originates as blood vessels penetrate the brain parenchyma early in development from the arachnoid to the brain parenchyma and carry with them the glia limitans that consists of the end feet of astrocytic processes 74. Interactions between the astrocytic sheath and the vascular endothelial cells are thought to be responsible for the formation of the interendothelial tight junctions that form the BBB , which occurs at around the end of the first trimester in man 75. However, if the role of the astrocyte is purely developmental why does the vascular ensheathment persist through adulthood? Presumably they are then converted to physiological roles, or a constant astrocytic influence is needed to maintain the BBB. Clearly the former can be the lactate shuttle hypothesis (see below), control of ingress of compounds such as glucose and amino acids or egress of waste metabolites 48, and control of [K+]o and blood flow 76. A reasonable view would be that one reason astrocyte processes continue to surround blood vessels in the mature animal is that they are transducers of changes in neuronal activity affecting blood vessel (arteriole) diameter and therefore flow. It has been recently suggested that one of the major roles of the mGluR-related increases in intracellular [Ca2+] in astrocytes is to activate phospholipase 2 to generate arachidonic acid and then prostaglandin 2 via COX-1 to dilate vascular smooth muscle 77,78. The astrocytes can also synthesize vasoactive epoxyeicosatrienoic acids (EETs) 79. It is possible that astrocytes act as transducers, which integrate local neuronal activity into either vasodilatation or constriction. Nevertheless, there is little doubt that direct innervations also play a role in regulation of vascular tone 80. Also see 78,81 for a recent summary of the possible neuronal, smooth muscle and endothelial cell, as well as astrocyte, influences on vascular smooth muscle tone.
The details of astrocyte regulation of blood vessel diameter and flow are now being worked out in both brain slices 77, and in vivo by 2 photon imaging 80. Differences have been observed between the two experimental systems which may depend on the degree of oxygen levels of the tissue vs. the slices 77. In this regard the intact animal would, by definition, provide the normal oxygenation, with the caveat of it being anesthetized. Recently an in vivo study has shown fine tuning of astrocyte Ca2+ responses to responses of contiguous neurons in the visual cortex of isoflurane-anesthetized ferrets in response to visual stimulation 82, which could then affect vessel diameter by Ca2+ dependent mechanisms, as reported in the recent studies just mentioned. The astrocyte response was more sensitive than neurons to increasing levels of isoflurane and this was used to show that the map of the hemodynamic response depended on the astrocyte Ca2+ response. Understanding which processes control the microcirculation is clearly an area of emerging interest and a topic of importance in several diseases, including cerebrovascular diseases and Alzheimer disease.
The first indication that astrocytes express high level of water channels or aquaporins was the discovery of intramembranous particles forming orthogonal arrays predominantly localized at the perivascular membranes of astrocytes around both capillaries and arterioles 83, contain a specific isoform of the recently discovered water channels or aquaporins, namely AQP4 84. These orthogonal arrays are now a defining characteristic of perivascular astrocytic processes. The intramembranous proteins that form the arrays were originally supposed to be K+ channels, in keeping with the K+ spatial buffering or siphoning concept, where increased [K+]o due to increased neuronal activity would be removed from the CNS by efflux from the astrocyte to the blood 85. Recent work has shown that Kir4.1 channels, whose RNA and protein 13,14 are well-expressed in astrocytes, are also part of the assemblies 85.
Primary astrocyte cultures prepared from AQP-4 knockout mice have a seven-fold reduced water permeability, whereas water transport of some neurons are very slow 86,87, consisting with the limited expression of AQPs in neurons. The function of the astrocytic perivascular AQP-4 channels remains unclear. Obviously, their polarized expression in the vascular endfeet of astrocytes suggest that water transport is high and that astrocytes constitute the major route for water transport into and out of the brain. However, endothelial cells have few, if any water channels 88. Mice with deletion of AQP4 developed strikingly smaller infarct and significant less edema, suggesting that pharmacological inhibition of AQP4 channels could represent a powerful neuroprotective strategy89. However, AQPs are now being shown to have other functions, most importantly to also facilitate the flux of gasses, including O2, CO2 and NO 90. In fact, the polarized expression of AQP4 in the endfeet facing the vasculature fits well with a channel that facilitates rapid exchange of vital gasses. Thus, the neuroprotective action of AQP4 deletion cannot be unequivocally attributed to reduced water fluxes. Unfortunately, these issues are not easily resolved, because agents that blocks AQP4 water permeability have not yet been identified91,92. Another concern related to the neuroprotective effects of AQP4 is that we must take into the account the extraordinary broad pleiotropic effects of deletion of protein that is essential to astrocytic function. A parallel example is the impact of deletion of the gap junction protein, Cx 43, that have been reported 93.
Astrocyte Neuron-Lactate-Shuttle-Hypothesis Magistretti and colleagues proposed an hypothesis, termed the astrocyte neuron-lactate-shuttle-hypothesis (ANLSH), in which glucose enters the CNS via the perivascular end feet of astrocytic processes and is there converted by aerobic glycolysis to lactate which then serves as the principle food for neurons94–96. Like regulation of blood vessel diameter and thence flow (see item 5 above), the underlying morphological characteristics have been recognized since the times of Golgi 97 and Ramón y Cajal 98. Namely, that blood vessels in the CNS are surrounded by astrocytic processes, and we now know that this can be close to 100% 99. This fundamental fact has given rise to many hypotheses ranging from development of the BBB due to signals derived from the astrocytes as the CNS develops, to the ANLSH. It seems that all material that does not diffuse between the astrocytic endfeet will have to pass initially through them, and then be transported out at some location or diffusing through the entire astrocyte and into neighboring ones via the gap junctions. The spaces between the astrocytic processes are of the order of a few hundreds of angstroms wide, so this could still form a major short circuit path as the astrocyte membranes would be a major resistance pathway for polar substances unless there are carriers on the astrocytic membranes.
Most of the supporting observations have been made in primary astrocyte cultures for the usual reasons, but these are convincing, only for the cultures. Astrocytes in situ do have the LDH isoform that favors the reductive production of lactate, and neurons the form that favors oxidation of lactate. However , although astrocytes have the glucose (Glu)-1 carrier, there is also a high density of high affinity Glu-3 transporters on neuronal membranes 100. The distribution of the 14 monocarboxylic transporters (MCT) isoforms are yet to be consistently correlated with export of lactate from astrocytes and uptake into neurons. However, in vivo confirmation is still lacking, except for one analysis which results can be interpreted in several different ways 101. A recent transcriptosome paper focused on enzymes and other proteins related to those processes principally involved in producing metabolic energy in astrocytes. The astrocytes were labeled by pGFAP driven GFP expression stained for GLT-1 after isolation to also cover a large pool of cells that were not GFAP-GFP positive, and then FACs-sorted. The pGFAP-GFP (−)/GLT-1(+) mice had only 2-fold less of the reliable astrocyte marker AQP-4 than the cells that were positive for both markers, but 236-fold more than Thy1 (+) neurons, as measured by qPCR. The cells were isolated from adult 10- to 12-week old mice using papain14. It was found that astrocytes contained transcripts for enzymes involved with glycolytic conversion of glucose to lactate, particularly >13-fold more lactic dehydrogenase b, which converts pyruvate to lactate, relative to neurons. The LDHa isozyme, that principally converts lactate to pyruvate, was 8 fold enriched in neurons. However, there were also high amounts of all the enzymes involved in the TCA cycle in astrocytes and mass spectrometric measurements showed that these cycles were active, so the cells do appear to be set up for aerobic glycolysis. Electron micrographs were also presented that showed high density of mitochondria in GFP(+) astrocytic processes, around or close to blood vessels. In terms of the reproducibility of the transcriptosome studies, of which there are at present only two for isolated astrocytes, the other study 13 showed equivalent expression of LDHa in isolated neurons , astrocytes and even primary cultures. However, the LDHb form was several fold higher in isolated astrocytes as compared to neurons and also quite high in the primary cultures. Plausible differences are that the oldest mice in 13 were only 17 days old, while the other study14 used 10–12 week old mice, or that the former study used the S100β promoter, a less selective astrocyte marker, to label the cells.
Thus, although several lines of work points to the existence of the astrocyte neuron-lactate-shuttle, definitive in vivo evidence is still lacking. It is also important to remember that neurons express glucose transporters and it is likely that multiple alternative pathways exist for something as important as fueling neural activity. Manipulation of key proteins involved in the astrocyte neuron-lactate-shuttle may, however, constitute a potential target for limiting neuronal loss in neurodegenerative diseases.
The mammalian CNS is particularly subject to the damaging effects of reactive oxygen species (ROSs) because of its high rate of oxidative metabolic activity and its high fatty acid content in the large quantities of myelin and other membranes. The unsaturated carbon-carbon bonds, needed to ensure sufficient fluidity of the fatty acid side chains of the phospholipids, are the ones most susceptible to oxidative damage by ROSs 102. ROSs are the unpaired electron versions of atomic and molecular oxygen that cause the breakdown of a large number of lipid and proteins in an autocatalytic manner. Because the full reduction of molecular oxygen by the respiratory chain will never be complete these ROSs need to be neutralized, and antioxidants and enzymes such as catalase and peroxidases are present to neutralize these. However, under pathological conditions these protective pathways are overwhelmed and ROS-induced damage becomes a major source of cellular injury, for example in the reperfusion phase of cerebral ischemia.
Astrocytes have a number of antioxidant systems such as the GSSG- GSH system 103-104. Isolated astrocytes also contain messages for the enzymes superoxide dismutase (Sod) and catalase (Cat), but these are not specifically enriched in astrocytes but found in neurons at reasonably high levels 13, and likely in other neural cells too as they are quite ubiquitous. However, improving antioxidation in astrocytes or limit their own production of ROSs is clearly a powerful strategy for reducing injury in a number of neurological diseases.
An excellent example on a supportive function of astrocytes is the recent suggestion that Muller cells act as light guides. Muller cells are specialized astrocytes of the retina, which similar to other astrocytes are responsible for glutamate uptake, K+ homeostasis, and pH control. As pointed out in the paper describing this phenomenon 105, the histology of the mammalian eye has always been paradoxical in that the photoreceptors are located at the back of the retina where the light s transmission would be degraded by scattering. The Muller cells orientation and low scattering makes them able to conduct the light like optical fibers to the interior of the retina so that it falls on the photoreceptors with less degradation (see rightmost image in fig 1C). This property can be best viewed as a support role; allowing light to reach the photoreceptors where it starts the complex process of being transformed to vision. It is a vital role given the structure of the retina (why the retina is built this way is another question), and would seem to be best described as facilitative, i.e. helps the eye s neuronal photoreceptors do their job of performing the initial step of translating light into vision.
It has long been known that the processes of astrocytes wall off groups of neurons and their synapses, as in glomeruli 69. Retraction of such ensheathment results in increased interaction levels between neurons, and increases the amount of hormones secreted from the neurons, such as the oxytocin/vasopressin-secreting neurons in the supraoptic nucleus106–107. A much earlier related suggestion was that this was, by insertion of astrocytic processes into the synaptic cleft, inhibiting synaptic activity 108. Pharmacological manipulation of astrocytic coverage of synapses constitutes a potential powerful therapeutic target. However, so far little is known with regard to what controls the plastic alteration in the fine structure of astrocytes.
What would a current general support theory of astrocytic function look like? Bushong (et al, 109) emphasized in his combined immunohistochemical and dye-filling studies of the CA1 region of 1 month old Sprague-Dawley rats that the dye-filled astrocytes were morphologically homogenous and their highly bushy processes occupied separate territories of ~ 70,000 μm3. See also Ogata and Kasaka110. Another study has showed that the domain concept also applies to human cortical protoplasmic astrocytes, with a bit more overlap7 . This is an important concept for protoplasmic astrocyte function. The domain arrangement of astrocytes can form the lynch pin of a general support theory of astrocyte function in the mature brain. This is that the vascular and synaptic end-feet function as autonomous units, first responding to specific events at these loci as independent entities driven by local feedback signals. Independent domains for intracellular [Ca2+] increases was shown some years back for the Bergmann glia lamellae and filipodia enwrapping the synapses that the Purkinje dendritic spines make with the parallel fibers 111. Also see 69 for a discussion on micro- to macrodomains based on current detailed morphological understanding. The simplest mechanism for how astrocytes specialize for their synaptic vs. vascular properties, would be segregation of functional membrane proteins within the plasma membrane with different transporters, channels and other membrane protein classes on astrocyte endfeet, lamellipodia and filipodia contacting synapses and blood vessels.
This support model of the mature mammalian astrocyte is well-grounded in cell and biological principles; processes cannot exist independently of a cell body. The other general biological principle is economy of space. Up to 100,000 synaptic targets, and a lesser number of blood vessels, can be served by one cell soma and thus conserve space, which in the mammalian brains is at a premium and has also, for example, led to myelination to reduce axonal diameters to conserve space for a given action potential velocity 112. One could, of course, have more astrocytic cell bodies with fewer processes and the astrocytic soma are quite small, ~ 10μm diameter. Presumably, the packing that has evolved is, by definition, optimal. Oberheim et al1 have shown that the domain of protoplasmic astrocytes in the human cortex is 2.6 times larger than in rodents (see fig.2.). At a minimum this may be the optimal packing relation for astrocytes in the much larger human brain, but the larger volume of the astrocytic domains will also allow human astrocytes to integrate roughly 17 times more information than their rodent equivalents, and could be related to the greater processing power of the human brain.
Only one disease identified so far is caused by a defect in an astrocytic gene (GFAP) (See chapter by A. Messing). Another disease with a primary brain malfunction ascribed to astrocytes is hepatic encephalopathy (see chapter by A. Cooper). This condition is characterized by confusion, altered level of consciousness and coma as a consequence of increases in ammonia/glutamine. Astrocytes are the only cell type in the brain that can detoxify ammonia by conversion of glutamate to glutamine, which in turn interferes with neuronal function.
Astrocytes have been shown to contribute to rather than being the cause of other diseases. One example would be the dominant mutations in the gene for superoxide dismutase (SOD1). Dominant mutations in SOD1 are a frequent cause of inherited amyotroph lateral sclerosis (ALS). Similarly, expression of mutant SOD1 in rodents replicates the disease and leads to progressive, selective motor neuron degeneration113. Cleveland and co-workers generated mice carrying a deletable SOD1 gene. When these mice were crossed GFAP-Cre mice to delete the mutant gene in astrocytes only, disease progression was sharply reduced113. The onset of motor neuron loss and functional deficits were not affected by deletion of the mutant gene in astrocytes 113.
It is an open question how much the malfunction of astrocytes contributes to neurodegenerative diseases. However, based on the fact that astrocytes are the major supportive cell type in the brain, it seems logically that minor dysfunction of astrocytes over years can contribute to neuronal loss. Moreover, reactive changes of astrocytes may exacerbate pathological changes in other cell types and thereby accelerate the disease progression
For specific drug targeting, the astrocytes of interest are those of the human brain. Most studies have, of course, been done in mammals with much smaller brains than humans, such as rats and mice. The latter are now being increasingly used because they can be genetically manipulated. This is especially important for astrocytes because, as argued in more detail elsewhere 6, definitive work on astrocytes can only be done in situ and in vivo and, because astrocytes have many of the same receptors and other proteins as neurons and there are, of course, neuron-astrocyte interactions, one is hard-pressed to distinguish direct effects on astrocytes and indirect effects derived from primary effects due to any manipulation on neurons, in the complex cellular mosaic of the brain. Astrocyte-specific, promoter-driven genetically altered mice are presently the best approach to definitively alter an astrocyte property in vivo and evaluate functional alterations, especially if an inducible promoter system is used.
Current views of the functions and roles of mature astrocytes range from long and well-established support roles, such as glutamate uptake at synapses, to recent proposals such as the ability of astrocytes to modulate and control synaptic activity though the release of neurotransmitters such as glutamate and ATP 17,114,115 . The evidence supporting the latter view is still in its preliminary phases, and some of the observations used to support the concept are in dispute. How do we best target astrocytes pharmacologically? The concept of the astrocyte as a supportive component of the CNS is well-established and comprehensive, although we lack a detailed mechanistic understanding of for example the pathways by which astrocytes buffer potassium or control water homeostasis. Thus, a clear understanding of the molecular mechanisms underlying the well-established supportive function of astrocytes represent the fasted approach for efficient pharmacological targeting of astrocytes.
We thank Takahiro Takano for graphics and Jon Goldman for comments on the manuscript. Supported by NIH (NS50350)
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Harold K. Kimelberg, Ordway Research Institute, 150 New Scotland Avenue, Albany NY 12208, USA.
Maiken Nedergaard, Center for Translational Neuromedicine, Department of Neurosurgery, University of Rochester Medical School, 601 Elmwood Avenue, Rochester, NY 146.