Patching the glia reveals the functional organisation of the brain
- 629 Downloads
The neuroglia was initially conceived by Rudolf Virchow as a non-cellular connective tissue holding neurones together. In 1894, Carl Ludwig Schleich proposed a hypothesis of fully integrated and interconnected neuronal-glial circuits as a substrate for brain function. This hypothesis received direct experimental support only hundred years later, after several physiological techniques, and most notably the patch-clamp method, were applied to glial cells. These experiments have demonstrated the existence of active and bi-directional neuronal–glial communications, integrating neuronal networks and glial syncytium into one functional circuit. The data accumulated during last 15 years prompt rethinking of the neuronal doctrine towards more inclusive concept, which regards both neurones and glia as equally responsible for information processing in the brain.
KeywordsGlia Neuronal–glial interactions Ion currents Neurotransmitter receptors Calcium signalling Patch-clamp History
The birth of glia
The history of glial research is confusing and painful. For a good part of the 20th century, glial cells were very much neglected by neurobiology, and the real advance of glial physiology begun only in the mid-1980s. Yet, the concept of glia appeared almost 50 years before the neuronal doctrine of nervous system organisation was conceived, and by the end of 19th century, a great variety of glial cells were identified and their morphology was very precisely characterised.
The father of glia, Rudolf Virchow, invented the idea of nervous ‘connective tissue’ in 1846, when he wrote “...this connective substance forms in the brain, in the spinal cord, and in the higher sensory nerves a sort of Nervenkitt (neuroglia), in which the nervous system elements are embedded” (; quoted from ). The word ‘glia’ has a dubious origin, initially appearing in the ancient Greek texts in a form of ‘γλοιοσ’, which meant ‘oily sediment’ used for taking baths (Semonides, VII century b.c.), ‘gum’ (Herodotus, V century b.c.), or ‘slippery, knavish person’ in Aristophanes (∼V century b.c.) tragedies. The latter meaning remained in Modern Greek, where the word ‘γλιοδισ’ stands for a morally debased and filthy person. Virchow further elaborated the concept of glia in his classical book Die Cellularpathologie in ihrer Begründung auf physiologische and pathologische Gewebelehre ; for Virchow, neuroglia was a true connective tissue of a non-cellular nature.
The first description of neuroglial cell, the radial glial cell (which is now known as Müller cell) in the retina was made by Heinrich Müller in 1851. Several years later, these cells were also characterised by Max Schultze (for relevant referencing see historical reviews [41, 79]). Around 1865, Otto Deiters described the stellate cells in white and grey matter . These cells closely resembled what we now know as astrocytes and in 1869, Jakob Henle has published the first image of cellular networks formed by stellate cells (i.e. astrocytes) in both grey and white matter of the spinal cord. Further discoveries in the field of the cellular origin of glial cells resulted from the efforts of many prominent histologists, in particular Camillo Golgi (1843–1926), Santiago Ramon y Cajal (1852–1934) and Pio Del Rio Hortega (1882–1945). Using a variety of microscopic techniques, Golgi and Ramon y Cajal discovered a wide diversity of glial cells in the brain, found the contacts formed between glial cells and blood vessels and observed cells located in closely aligned groups in between nerve fibres (this being the first observation of oligodendrocytes). Further advances in morphological characterisation of glia appeared after Golgi developed famous black (silver chromate) reaction, and Ramon y Cajal invented gold–chloride sublimate staining technique, which significantly improved microscopic visualisation of cells (and neuroglial cells in particular) in brain tissues. Closer to the end of 19th century, Michael von Lenhossek proposed the term astrocyte (1893), which within the next two decades gained universal acceptance. The name ‘oligodendrocyte’ appeared slightly later, after Pio Del Rio-Hortega introduced silver carbonate staining technique, which selectively labelled these cells (1921). It was also Del Rio-Hortega who proposed the term ‘microglia’ to characterise the distinct cellular population of mesodermal origin; he was one of the first to understand that these cells can migrate and can act as phagocytes.
The main peripheral glial element, the Schwann cell, was named by Louis Antoine Ranvier (1871), after earlier discoveries of Robert Remak, who described the myelin sheath around peripheral nerve fibres (1838) and Theodor Schwann, who suggested that myelin sheath was a product of specialised cells (1839).
At the end of 19th century, several possible functional roles of glial cells were considered. Camillo Golgi, for example, believed that glial cells are mainly responsible for feeding neurones, by virtue of their processes contacting both blood vessels and nerve cells. This theory was, however, opposed by Santiago Ramon y Cajal. Another theory (proposed by Carl Weigert) considered glial cells as mere structural elements of the brain, which filled space not occupied by neurones. Finally, Ramon-y-Cajal brother, Pedro, considered astrocytes as insulators, which prevented undesirable spread of neuronal impulses.
For the first time, the idea of active neuronal–glial interactions as a substrate for brain functions was voiced by Carl Ludwig Schleich (1859–1922) in 1894 in his book Schmerzlose Operationen . Incidentally, this happened at the same year when the ‘neuronal’ doctrine was introduced by Sigmund Exner  and only 3 years after the term ‘neurone’ was coined by Wilhelm Gottfried von Waldeyer . Schleich believed that glia and neurones both act as active cellular elements of the brain. He thought that glial cells represent fundamental inhibitory mechanism in the brain. According to Schleich, neuronal excitation is transmitted from neurone to neurone through intercellular gap; this inter-neuronal gap, surrounded by glial cells is the anatomical substrate for controlling excitation or inhibition in neural networks. The constantly changing volume of glial cell represents the mechanism for this control: swelling of glial cells inhibits neuronal communications, when glia shrinks, the impulse propagation is facilitated. The ideas of Schleich were never fully appreciated by a neuroscience community and from the beginning of 20th century glial cells were firmly believed to represent a supportive and passive element of the brain, completely devoid of any role in the integration and processing of information.
Beginning of glial electrophysiology
The turn of the tide-identification of glial excitation
The modern era of glial physiology was triggered by several important technical advances, which arrived almost simultaneously at the end of 1970s to the beginning of 1980s. These were the patch-clamp, developed by Bert Sakmann and Ervin Neher in collaboration with Owen Hamill, Fred Sigworth and Alan Marty ; the florescent calcium indicators invented by Roger Tsien [83, 84] and the technique of preparation and maintenance of glial cell cultures [57, 58]. These techniques used alone or in combination were instrumental for many fundamental discoveries, which completely change our comprehension of the importance of glial cells in the nervous system.
In parallel, the cytoplasmic calcium fluctuations were intensively studied in cultured glial cells loaded with fluorescent Ca2+ indicators. These studies have found that (1) stimulation of glia with neurotransmitters or hormones almost invariably resulted in generation of intracellular Ca2+ signals driven predominantly through the metabotropic route, which involved phospholipase C/InsP3-evoked Ca2+ release from the endoplasmic reticulum Ca2+ store [30, 46, 72, 85] and (2) that the Ca2+ signals triggered in one glial cell may travel though the syncytium formed within monolayers of cultured glial cells connected with gap junctions [21, 22, 23, 24, 34]. These findings led to formulation of hypothesis of special form of glial excitability, which is governed by Ca2+ release from the intracellular excitable media, represented by the membrane of the ER .
In cerebellum, stimulation of parallel fibres induced a complex membrane current response in Bergmann glial cells (which intimately enwrap synapses formed by parallel fibres on Purkinje neurones), which was mediated by both activation of ionotropic AMPA-type glutamate receptors and activation of Na+-dependent glutamate transporters . Most interestingly, the glial AMPA receptor-mediated responses show both short-term and long-term plasticity. The former is manifested by prominent paired-pulse facilitation , whereas the latter appears in a form of long-term depression of synaptically induced glial responses following low-frequency repetitive stimulation of parallel fibres . Bergmann glial cells also demonstrated clear elementary ‘postsynaptic’ currents in response to quantal release of glutamate from presynaptic terminals , suggesting very close apposition of glial and presynaptic membranes. Electrical stimulation of parallel fibres also triggered cytoplasmic calcium responses in Bergmann glia; these calcium responses were very much localised, being limited to a small compartments within glial cell processes . It is interesting to note that the size of these compartments, where local Ca2+ signals occurred, was similar to morphologically distinct appendage-like structures, protruding from the main process and enwrapping a cluster of synapses formed by parallel fibres. This high compartmentalisation of synaptically induced glial Ca2+ responses can be important for spatial discrimination of the incoming signals.
Further experiments revealed yet another important fact: glial cells can signal back to neurones. This type of glia to neurone signalling was detected in both glial–neuronal co-cultures and in acutely isolated brain slices. Using the in vitro model, several groups have demonstrated that stimulation of astrocytes evoked direct excitation of nearby neurones [34, 75] and affected spontaneous  and evoked  postsynaptic currents. Inhibition of astroglial calcium signals (either by intracellular administration of BAPTA or by suppression of ER Ca2+ release by thapsigargin) eliminated all neuronal responses. In acute slices, activation of glia was also shown to trigger direct neuronal responses. Spontaneous Ca2+ oscillations in astrocytes, for example, excited neighbouring neurones in hippocampus, cortex and thalamus [66, 68]. When astrocytes in CA1 hippocampal slices were activated by prostaglandin E2, adjacent neurones generated Ca2+ signals; this action was mediated through glutamate receptors . Activation of hippocampal astrocytes was also reported to produce tonic suppression of synaptic transmission; this inhibition was mediated by adenosine accumulated after the release of ATP from the astroglia .
Quite recently, it also became clear that the main mechanism of astroglial to neuronal signalling involves secretion of classical neurotransmitters, such as glutamate or ATP from the glial cells; this secretion can occur through several independent routes, most importantly though regulated exocytosis. It is interesting to note that the first indication that glial cells (e.g. Schwann cells and astrocytes) are capable of secreting neurotransmitters such as acetylcholine and GABA were obtained already from mid-1970s [27, 59, 78], although these data were not somehow generally acknowledged. In recent years, however, the concept of neurotransmitters release from glia gained firm support. Glial cells were found to contain secretory vesicles endowed with glutamate transporters and reach in glutamate; it was also shown that astroglial cells express numerous proteins responsible for exocytotic vesicles fusion, including cellubrevin (VAMP3), SNAP23, complexin 2, Munch 18a and synaptotagmin IV [15, 61, 92, 94]. Finally, the Ca2+-induced exocytosis of glutamate from astrocytes was directly identified by total internal reflection fluorescence imaging , and exocytotic fusion after [Ca2+]i signals was measured by membrane capacitance recordings . Besides glutamate and ATP astrocytes are able to release several other ‘glio’ transmitters, most notably taurine , aspartate  and D-serine [64, 76]. All these transmitters can be secreted either by exocytosis or through membrane channels, such as volume-sensitive Cl− channels or pores associated with activation of P2X7 purinoreceptors, or indeed through reversal of neurotransmitter transporters.
All these discoveries led to a complete rethinking of the role and place of neuroglia in brain circuits and initiated an appearance of a concept of a ‘tripartite’ synapse; this concept regards glial cell as legitimate part of the synapse, which is built from three functionally important compartments: presynaptic, postsynaptic and glial [3, 70, 91].
Brain integration as a function of neuronal–glial networking: can neuronal doctrine hold?
Our understanding of glial function changed dramatically over the last 15 years. This change concerns not only the physiology of glia but also the whole concept of the way how the brain is organised and how the development, life and death of neural circuits are controlled. There is a compelling evidence demonstrating that the radial glial cells serve as pluripotent neural progenitors and the cells with astroglial phenotype are the brain stem elements underlying adult neurogenesis. The mature astroglia defines the microarchitecture of the grey matter by dividing it into separate units, where the astrocyte rules over all neuronal membranes and synaptic contacts within this clearly defined territory. Within this territory, astrocytes control almost everything from regulating birth, maintenance and elimination of synapses to providing active neurones with energy substrates and linking neurones with local microcirculation. Although clearly delineated, the astrocytic domains are tightly integrated into the syncytium by gap junctions connecting the terminal processes of glial cells. This syncytial structure permits long-range signalling and intercellular exchange of metabolites and second messengers. At the same time, astrocytes are fully capable of extracellular signalling by virtue of ‘glio’ transmitters secreted in highly regulated fashion. These multiple signalling abilities, which include both intra-and extracellular routes, can potentially be very important for integrative processes within astroglial networks. Finally, the failure of astrocytes to perform their duty invariably results in neuronal death, and many types of brain pathology, from stroke to Alzheimer disease may be primarily glial diseases.
All in all, glial cells weave the canvass on top of which neurones perform their rapid signalling. The extremely elaborated panglial syncytium resembles the ‘reticular theory’ of brain organization so much championed by Golgi and Kolliker, which now can be safely reconciled with ‘neuronal’ theory, as both internally connected glial syncytium and discreet neuronal networks peacefully coexist within the brain. Shall we anticipate even more exciting discoveries which eventually can challenge the role of neurones as sole origins of cognition? Can neuronal doctrine withstand the challenge from glial cells? These questions are still open. Nonetheless, after 20 years of patching the glia, we already elevated these cells from a mere ‘nerve putty’ to an active element of the brain circuitry.
The author’s research was supported by The Wellcome Trust, The Alzheimer Research Trust, NIH, Royal Society and INTAS.
- 26.Deiters O (1865) Untersuchungen über Gehirn und Rückenmark des Menschen und der Säugethiere. Vieweg, BraunschweigGoogle Scholar
- 29.Exner S (1894) Entwurf zur physiologischen Erklärung der Psychischen Erscheinungen. Deiticke, Leipzig/ViennaGoogle Scholar
- 41.Kettenmann H, Ransom BR (2005) The concept of neuroglia: a historical perspective. In: Kettenmann H, Ransom BR (eds) Neuroglia. OUP, Oxford, pp 1–16Google Scholar
- 77.Schleich CL (1894) Schmerzlose Operationen: Örtliche Betäubung mit indiffrenten Flüssigkeiten. Psychophysik des natürlichen und künstlichen Schlafes. Springer, Berlin Heidelberg New YorkGoogle Scholar
- 89.Virchow R (1846) Uber das granulierte Ansehen der Wandungen der Gehirnventrikel. Allg Zsch Psychiat 3:242–250Google Scholar
- 90.Virchow R (1858) Die Cellularpathologie in ihrer Begründung auf physiologische and pathologische Gewebelehre. Zwanzig Vorlesungen gehalten während der Monate Februar, März und April 1858 im pathologischen Institut zu Berlin. August Hirschwald, BerlinGoogle Scholar
- 91.Volterra A, Haydon P, Magistretti P (2002) Glia in synaptic transmission. Oxford University Press, OxfordGoogle Scholar
- 93.Waldeyer von HWG (1891) Über einige neuere Forschungen im Gebiete der Anatomie des Centralnervensystems. Dtsch Med Wochenschr 44:1–44Google Scholar