“How does the brain work?” is one of the most frequently asked questions. By and large, the human brain consists of approx-imately 160 billion cells that mainly repre-sent two types, classified as neurons and glia, each contributing about 50 % to the total cell number. Modern imaging tech-niques such as functional magnetic reso-nance imaging (fMRI) or diffusion-tensor imaging (DTI) show us many functional details of our brain’s plastic activities and have provided exciting new insights of how and where the brain processes spe-cific information (. Fig. 1a). Interestingly, neither fMRI nor DTI are mechanistical-ly based on the electrical activity of neu-rons, classically thought to be exclusively responsible for information processing in the brain. Instead, these techniques enable the visualization of brain activity based on changes of cerebral blood flow, i.e. the ox-ygen content supplied by blood capillaries, or by anisotropic water diffusion. The cel-lular building blocks of the smallest fMRI and DTI signals have already been identi-fied as the neurovascular unit and the my-elin-axon unit, respectively (. Fig. 1b and c). A single fMRI voxel integrates the oxy-gen consumption of several cell types with vascular, glial and neuronal origin: endo-thelial cells, neutrophils, pericytes, astro-cytes, microglia, NG2 glia, oligodendro-cytes and neurons. The myelin-axon unit that gives rise to the DTI signal appears much less complicated: neuronal fiber tracts, the axons, are surrounded by the insulating and support-providing myelin sheaths of the oligodendrocytes. Both im-aging techniques take thus advantage of the strategic positions of glial cells which enable them to power neurons and to

maintain long-range connections, high-lighting the prominent role of neuroglia in neuronal performance.

The two-dimensional space of his-tological or vital brain sections already demonstrated that cells were organized in functional and morphological networks. Among these are the columnar modules of the cortex that contribute to sensory in-tegration and cognition, the layers of the hippocampus involved in learning and memory, the rhythmic centers of the brain stem which regulate breathing, or the cer-ebellar circuits responsible for fine-tuning motor coordination. While excitatory and inhibitory neurons are the main relay sta-tions for the input, processing and output of electrical signals, the macroglial cells execute quite different tasks. Astrocytes are polarized cells that represent a bridge between blood vessels and neurons. They take up nutrients from the blood, metab-olize them and provide them to neurons. Astrocytes also regulate extracellular ion and transmitter homeostasis from the so-called “tri-partite” synapse where they are in direct contact with neuronal synapses. Furthermore, by secreting transmitters and peptide hormones, they can directly modulate synaptic transmission. Oligo-dendrocytes insulate neuronal axons with a lipid-rich structure, the myelin sheath, to accelerate action potential propagation and to electrically insulate axons. Recent data has additionally demonstrated that oligodendrocytes metabolically support axons, the long-range links of neural cir-cuits. Glial cells expressing the proteogly-can NG2 (NG2 glia) are a relatively nov-el class of macroglia and were originally identified as oligodendroglial progenitor

cells, but appear to represent a more versa-tile cell reservoir in the adult brain.

Present-day research provides com-pelling evidence that a neuron-centered picture of the brain is way too simplis-tic, indicating that each class of glial cells is much more diverse than common-ly thought. Glial cells appear to have dis-tinct physiological properties in different brain regions, at different developmen-tal stages and at different activity levels of the organism. These observations sug-gest that functional specializations of glia might have developed to meet the spe-cific requirements of distinct networks which might as such be critical determi-nants of brain activity. This new concept will change the way we think about brain function and put glial cells into an even more prominent focus of attention.

Astrocytes are probably the most ver-satile class of neuroglia. Functionally po-sitioned between the pia mater, blood vessels and neuronal synapses, they dis-play a plethora of properties. Astrocytes contribute to the blood-brain barrier [3], take up nutrients from the blood, metab-olize them and provide energy substrates to neurons [26]. They link neuronal ac-tivity to blood circulation [2, 15], promote synapse formation [5, 8], and determine the properties of the extracellular ma-trix [9, 16, 27]. Furthermore, they regu-late extracellular ion and transmitter lev-els thereby regulating synaptic transmis-sion. Last but not least, astrocytes secrete compounds which modulate neurotrans-mission [1, 22]. This impressive list dem-onstrates via how many routes astrocytes interact with neurons and influence brain activity. Astrocytes must be particular-

Christine R. Rose1 · Frank Kirchhoff2

1  Institute of Neurobiology, Heinrich Heine University Düsseldorf, Düsseldorf, Germany2  Molecular Physiology, Center for Integrative Physiology and Molecular

Medicine (CIPMM), University of Saarland, Homburg, Germany

Glial heterogeneity: the increasing complexity of the brain

e-Neuroforum 2015 · 6:59–62DOI 10.1007/s13295-015-0012-0Published online: 31 July 2015© Springer-Verlag Berlin Heidelberg 2015

59e-Neuroforum 3 · 2015 |

Editorial

ly tailored to fulfil these functions and be able to adapt to the developmental stage, the brain region and the activity phases. Indeed, numerous examples of glial het-erogeneity exist. This is especially con-spicuous in the morphological specializa-tion of astrocytes (. Fig. 2). In some re-gions such as the cerebellar cortex and the retina, they exhibit a radial orientation. In contrast, astrocytes in the cortex or hip-pocampus extend processes in all direc-tions displaying a star-like appearance, while those of the white matter are less frequently branched and largely lack thin membrane protrusions. Not surprising-ly, first profiling studies of astrocytes iso-lated from different brain regions display substantial differences in gene expression [4, 7, 23]. These include cell surface gly-coproteins and components of the extra-cellular matrix, ion channels, neurotrans-mitter receptors and transporters, con-nexins, Eph receptors, and many more [10, 20, 28]. More recent studies show that as-troglia sense and compute neuronal activ-ity to feed back to neurones, thereby even modulating the most visible brain output, the behaviour. Astroglial cannabinoid re-ceptors in the hippocampus, for example, are involved in the acquisition of spatial working memory [14], and in the cerebel-lum, Bergmann glial AMPA receptors are important determinants of fine motor co-ordination [25].

The second class of macroglia, the oli-godendrocytes, are the myelin-forming cells of the CNS. A single oligodendro-

cyte enwraps up to 50 axons, and myelin-ating segments can vary in length from 50 to 400 µm. Their morphological heteroge-neity has already been described by Rio-Hortega, who distinguished four types. In white matter fiber tracts such as the optic nerve or the corpus callosum, axons are mainly oriented in parallel, and so are the processes of the oligodendrocytes. In con-trast, in grey matter regions where axons traverse the parenchyma irregularly, oli-godendroglial processes point into all di-rections. The functional characterization of oligodendroglial heterogeneity is still in its infancy. They are equipped with a vari-ety of receptors to sense the extracellular level of transmitters released by neurones [17]. As a consequence, myelination is reg-ulated by neuronal activity, but is also de-termined by axon diameter. In addition to their role in myelination, recent studies highlight the important role of oligoden-drocytes in supporting axons also meta-bolically [11, 19].

NG2 glial cells constitute less than ten percent of glial cells in the developing and adult CNS [6]. They have originally been identified by the expression of the chon-droitinsulfate proteoglycan NG2, and functionally been characterized as oligo-dendrocyte progenitor cells. Recent stud-ies demonstrate that outside of neurogen-ic niches, NG2 cells are the most prolif-erative cells of the adult CNS. NG2 glial cells are the only glial cells directly inner-vated by neurons. While in the hippocam-pus and cerebellum NG2 cells receive both

glutamatergic and GABAergic input, NG2 cells in the medial nucleus of the trape-zoid body only receive excitatory glutama-tergic synaptic input in parallel to the Ca-lyx of Held [21]. The neuron-glia synapses even persist during cell division. Although all NG2 cells can develop into oligoden-drocytes, there exist strong differences be-tween white and grey matter. While in the corpus callosum, almost two third of NG2 glia become oligodendrocytes, in the cor-tex 90 % remain NG2 glia. Cell prolifera-tion was suggested to be regulated by volt-age-gated sodium and potassium chan-nels heterogeneously expressed on NG2 glia during differentiation [12, 18].

The Deutsche Forschungsgemein-schaft acknowledges the demand for fur-ther research on the heterogeneity and funds the special priority program SPP 1757 “Functional specializations of neu-roglia ascritical determinants of brain ac-tivity”. The Priority Program will pave the way for a better understanding of the mo-lecular and cellular role of glia in brain pa-thologies that is urgently needed to devel-op novel, more customized and targeted strategies for the treatment of brain inju-ry and disease.

In this special issue of Neuroforum members of the SPP 1757 will present re-cent highlights of glia research addressing the heterogeneity of astrocytes (Christian Henneberger), NG2 glia (Dirk Dietrich and Christian Steinhäuser) and oligoden-drocytes (Leda Dimou and Michael Weg-ner). These reviews will be complement-

Fig. 1 8 Modern brain imaging techniques are mechanistically based on activity of macroglial cells. a Functional magnet-ic resonance imaging (fMRI) is able to visualize different functional units (1–3) of the brain in its idle mode (default mode net-work). A variant of fMRI, diffusion tensor imaging (DTI) can selectively visualize connecting fibre tracts between these centres (4–6). b The cellular basis of the fMRI signal is the change in oxygen levels in the neurovascular unit composed of capillaries, astrocytes and neurones. c The axon-myelin unit of fibre tracts causes anisotropic water diffusion that gives rise to DTI signals. Figure A modified from [13]

60 | e-Neuroforum 3 · 2015

Editorial

ed by an article of Daniela Dieterich and Moritz Rossner who describe novel high-throughput approaches with significant impact for the identification of cellular specializations in the brain and among gli-al cells. The SPP 1757 moreover will serve as a platform to enhance inter-lab com-munication not only at the national, but also at the international level. Especially strong ties have been established with gli-al research colleagues in Japan. Kazuhi-ro Ikenaka will describe the Japanese pro-gram to foster glial research.

Corresponding addresses

C.R. RoseInstitute of NeurobiologyHeinrich Heine University Düsseldorf Universitätsstr. 1; Building 26.02.00, 40225 Düsseldorfrose@uni-duesseldorf.de

F. KirchhoffMolecular Physiology, Center for Integrative Physiology and Molecular Medicine (CIPMM)University of Saarland Building 48, 66421 Homburgfrank.kirchhoff@uks.eu

Christine R. Rose Diploma in Biology (University of Konstanz); PhD 1990–1993 University of Kaiserslautern (Dept. of General Zoology, Prof. J. Deitmer); Postdoctoral Fellow (DFG Fellowship) at the Dept. of Neurology, Yale University School of Medicine (Bruce Ransom and Steve Waxman). 1997–1999 Research Assistant at the Institute of Physiology (University of Saarland, Homburg; Prof. A. Konnerth), 1999–2003: Research Assistant at the Institute of Physiology TU/LMU Munich (A. Konnerth); 2001: Habilitation in Physiology (LMU Munich), 2003–2005 Heisenberg Fellow of the DFG DFG. Since 2005: Director of the Institute of Neurobiology (University Düsseldorf). Since 2012 Subject Reviewer of the DFG.

Frank Kirchhoff 1981–1985: Diploma in Biochemistry at the University of Hannover; 1986–1990 PhD University Heidelberg (Institute of Neurobiology, Prof. M. Schachner); 1991–1994 Postdoktoral Fellow (Institute of Neurobiology, Prof. Helmut Kettenmann); 1995–1999 Max Delbrück Center for Molecular Medicine Berlin (Cellular Neurosciences, Prof. H. Kettenmann), 1997 Habilitation in Biochemistry at the Free University of Berlin; 2000–2008 research group leader at the Max Planck Institute of Experimental Medicine (Neurogenetics, Prof. K.-A. Nave). Since 2009 University Professor (Molecular Physiology, University of Saarland, Homburg).

References

1. Araque A, Carmignoto G, Haydon PG, Oliet SH, Ro-bitaille R, Volterra A (2014) Gliotransmitters travel in time and space. Neuron 81:728–739

2. Attwell D, Buchan AM, Charpak S, Lauritzen M, Macvicar BA, Newman EA (2010) Glial and neuro-nal control of brain blood flow. Nature 468:232–243

3. Ballabh P, Braun A, Nedergaard M (2004) The blood-brain barrier: an overview: structure, regula-tion, and clinical implications. Neurobiol Dis 16:1–13

4. Beckervordersandforth R, Tripathi P, Ninkovic J, Bayam E, Lepier A, Stempfhuber B, Kirchhoff F, Hirrlinger J, Haslinger A, Lie DC, Beckers J, Yoder B, Irmler M, Gotz M (2010) In vivo fate mapping and expression analysis reveals molecular hallmarks of prospectively isolated adult neural stem cells. Cell Stem Cell 7:744–758

5. Chung WS, Allen NJ, Eroglu C (2015) Astrocytes Control Synapse Formation, Function, and Elimina-tion. Cold Spring Harb Perspect Biol a020370

6. Dimou L, Gallo V (2015) NG2-glia and their func-tions in the central nervous system. Glia 63:1429–1451

7. Doyle JP, Dougherty JD, Heiman M, Schmidt EF, Stevens TR, Ma G, Bupp S, Shrestha P, Shah RD, Doughty ML, Gong S, Greengard P, Heintz N (2008) Application of a translational profiling approach for the comparative analysis of CNS cell types. Cell 135:749–762

8. Eroglu C, Barres BA (2010) Regulation of synaptic connectivity by glia. Nature 468:223–231

9. Faissner A, Pyka M, Geissler M, Sobik T, Frisch-knecht R, Gundelfinger ED, Seidenbecher C (2010) Contributions of astrocytes to synapse formation and maturation – potential functions of the peri-synaptic extracellular matrix. Brain Res Rev 63:26–38

10. Freeman MR (2010) Specification and morpho-genesis of astrocytes. Science 330:774–778

11. Fünfschilling U, Supplie LM, Mahad D, Boretius S, Saab AS, Edgar J, Brinkmann BG, Kassmann CM, Tzvetanova ID, Möbius W, Diaz F, Meijer D, Suter U, Hamprecht B, Sereda MW, Moraes CT, Frahm J, Goebbels S, Nave KA (2012) Glycolytic oligoden-drocytes maintain myelin and long-term axonal in-tegrity. Nature 485:517–521

12. Gallo V, Zhou JM, Mcbain CJ, Wright P, Knutson PL, Armstrong RC (1996) Oligodendrocyte progenitor cell proliferation and lineage progression are regu-lated by glutamate receptor-mediated K + channel block. J Neurosci 16:2659–2670

13. Greicius MD, Supekar K, Menon V, Dougherty RF (2009) Resting-state functional connectivity re-flects structural connectivity in the default mode network. Cereb Cortex 19:72–78

14. Han J, Kesner P, Metna-Laurent M, Duan T, Xu L, Georges F, Koehl M, Abrous DN, Mendizabal-Zu-biaga J, Grandes P, Liu Q, Bai G, Wang W, Xiong L, Ren W, Marsicano G, Zhang X (2012) Acute canna-binoids impair working memory through astrogli-al CB1 receptor modulation of hippocampal LTD. Cell 148:1039–1050

15. Haydon PG, Carmignoto G (2006) Astrocyte con-trol of synaptic transmission and neurovascular coupling. Physiol Rev 86:1009–1031

16. Jones EV, Bouvier DS (2014) Astrocyte-secret-ed matricellular proteins in CNS remodelling during development and disease. Neural Plast 2014:321209

17. Karadottir R, Attwell D (2007) Neurotransmitter re-ceptors in the life and death of oligodendrocytes. Neuroscience 145:1426–1438

18. Karadottir R, Hamilton NB, Bakiri Y, Attwell D (2008) Spiking and nonspiking classes of oligo-dendrocyte precursor glia in CNS white matter. Nat Neurosci 11:450–456

Fig. 2 8 Heterogeneity of human astroglia. Semi-schematic survey of the main types of astroglia and related glial cells, and their localization in differ-ent layers/specialized regions of the human brain. I: tanicyte (a: pial; b: vas-cular); II: radial astrocyte (Bergmann glial cell); III: marginal astrocyte; IV: pro-toplasmic astrocyte; V: velate astrocyte; VI: fibrous astrocyte; VII: perivascu-lar astrocyte; VIII: inter-laminar astrocyte; IX: immature astrocyte/glioblast; X: ependymocyte; XI: choroid plexus cell. From: Reichenbach and Wolburg [24]

61e-Neuroforum 3 · 2015 |

19. Lee Y, Morrison BM, Li Y, Lengacher S, Farah MH, Hoffman PN, Liu Y, Tsingalia A, Jin L, Zhang PW, Pellerin L, Magistretti PJ, Rothstein JD (2012) Oligo-dendroglia metabolically support axons and con-tribute to neurodegeneration. Nature 487:443–448

20. Matyash V, Kettenmann H (2010) Heterogeneity in astrocyte morphology and physiology. Brain Res Rev 63:2–10

21. Müller J, Reyes-Haro D, Pivneva T, Nolte C, Schaette R, Lubke J, Kettenmann H (2009) The principal neurons of the medial nucleus of the trapezoid body and NG2( + ) glial cells receive co-ordinated excitatory synaptic input. J Gen Physiol 134:115–127

22. Perea G, Navarrete M, Araque A (2009) Tripartite synapses: astrocytes process and control synaptic information. Trends Neurosci 32:421–431

23. Pinto L, Mader MT, Irmler M, Gentilini M, Santoni F, Drechsel D, Blum R, Stahl R, Bulfone A, Malates-ta P, Beckers J, Gotz M (2008) Prospective isolation of functionally distinct radial glial subtypes–lin-eage and transcriptome analysis. Mol Cell Neurosci 38:15–42

24. Reichenbach A, Wolburg H (2005) Astrocytes and ependymal glia. In: Kettenmann H & Ransom BR (eds) Neuroglia, 2nd edn. Oxford University Press, New York

25. Saab AS, Neumeyer A, Jahn HM, Cupido A, Simek AA, Boele HJ, Scheller A, Le Meur K, Gotz M, Mony-er H, Sprengel R, Rubio ME, Deitmer JW, De Zeeuw CI, Kirchhoff F (2012) Bergmann glial AMPA recep-tors are required for fine motor coordination. Sci-ence 337:749–753

26. Suzuki A, Stern SA, Bozdagi O, Huntley GW, Walk-er RH, Magistretti PJ, Alberini CM (2011) Astrocyte-neuron lactate transport is required for long-term memory formation. Cell 144:810–823

27. Wiese S, Karus M, Faissner A (2012) Astrocytes as a source for extracellular matrix molecules and cyto-kines. Front Pharmacol 3:120

28. Zhang Y, Barres BA (2010) Astrocyte heterogene-ity: an underappreciated topic in neurobiology. Curr Opin Neurobiol 20:588–594

62 | e-Neuroforum 3 · 2015

Editorial

Background

Increasing evidence suggest that both neuronal and glial heterogeneity contrib-ute to and are essential for higher order brain functions. A plethora of observa-tions have been made showing that neu-rons and glial cells are not only physically highly intermingled but are physiological-ly tightly connected and mutually depend at various levels on each other. For exam-ple, astroglial cells are essential for main-taining the metabolic integrity of neurons and participate in synaptic transmission (“tripartite synapse”) [1, 4]. Therefore, the view emerges that disturbed neuron–glia interactions may contribute to the initia-tion and progression of neurodegenera-tive and psychiatric diseases. Intriguing-ly and similarly to neurons, macroglia classes like astrocytes, NG2 cells, and oli-godendrocytes are not at all homogenous cell populations but do possess a marked-ly heterogeneity in morphology, function-ality, and cellular activity as well, that on-ly recently has been started to be acknowl-edged and integrated into a concept of brain function that pictures a neural world rather than a puristical neuronal world. In the light of these findings, it is not too sur-prising that glial cells are also increasingly considered as attractive targets for clinical intervention strategies aiming at re-bal-ancing disturbed brain function.

So far, the application of convention-al approaches to characterize neurons and more recently also glial cells at global lev-els by applying so-called -omic technolo-gies has been limited for the brain [8]. This has several reasons: (1) the cellular com-

plexity of the nervous system is extreme, with a high degree of regional variability of the different cell types and, hence, sus-ceptibility in different diseases. (2) Under physiological and pathological conditions, the proper function of neurons cannot be dissociated from supporting glial cells—and vice versa. (3) Most neurodegenera-tive and psychiatric disorders are late-on-set chronic “systems-diseases” that affect different neuron–glia networks, which complicates comparison between differ-ent studies (4). The analysis of brain tis-sue is still state of the art for molecular and biochemical studies, although cell-type relevant processes are likely to be masked due to the fact that all cells share a rath-er large pool of identical proteins rang-ing from housekeeping proteins, protein signaling cascades, adhesion molecules, and receptors and ion channels. Because of these facts, several methods needed to be developed and are still in progress of refined development that allow to specif-ically label and isolate either different gli-al and neuronal cells or cell-specific sub-cellular compartments to high purity and at conditions that allow for the application of state-of-the art -omic approaches. The appendix “omics” added to the class of bi-ological molecules under investigation circumscribes respective globally scaled technologies, such as Genomics, Tran-scriptomics, Proteomics, Lipidomics, and Metabolomics. By definition, omics tech-nologies claim to cover the majority of dif-ferent molecules of a given subclass within one experiment providing a systems-level view onto a biological sample. Thus, un-biased and exploratory (“hypothesis-gen-

erating”) approaches toward an enhanced understanding of physiological or patho-logical cellular states are possible. Subse-quently, we will provide an overview on current technical approaches used to per-form transcriptomics and proteomics to dissect glial heterogeneity of the brain.

Strategies to isolate neuronal and glial cell types of the brain

Defined ex vivo culture conditions have been developed for all major cell types of the brain, such as astrocytes, oligodendro-cyte precursor cells (OPCs or NG2 glia), mature oligodendrocytes, and different neuronal subtypes as well as microglia. Thus, it is possible to culture large quanti-ties of defined cell types for extended pe-riods in vitro. Although the basic electro-physiological and molecular character-istics of cultured neuronal and glial cells appear to be sustained in vitro, it is wide-ly accepted that ex vivo cultures will on-ly partially reflect the in vivo situation even when different culture conditions are employed and compared. The mo-lecular identity of a given cell type de-pends not only on the availability of nu-trients and soluble survival factors (“me-dia and supplements”) but also critically on regional cues depending for example on the nature and activity of surrounding cells, composition of the local extracellu-lar matrix as well as many other factors. Thus, the local microenvironment deter-mines the molecular and structural fea-tures of cells in vivo, which cannot be ex-actly mirrored in vitro. For globally scaled proteomics and also the analysis of non-

Daniela C. Dieterich1,2,3 · Moritz J. Rossner4

1  Institute for Pharmacology and Toxicology, Otto-von-Guericke University Magdeburg, Magdeburg, Germany2  Leibniz Institute for Neurobiology, Magdeburg, Germany3  Center for Behavioral Brain Sciences, Magdeburg, Germany4  Laboratory of Molecular and Behavioral Neurobiology, Department of Psychiatry, Munich, Germany

Dissecting the regional diversity of glial cells by applying -omic technologies

e-Neuroforum 2015 · 6:63–68DOI 10.1007/s13295-015-0009-8Published online: 11 August 2015© Springer-Verlag Berlin Heidelberg 2015

63e-Neuroforum 3 · 2015 |

Review article

coding RNAs (ncRNAs), however, large homogenous pools of intact cells includ-ing all compartments and membranous extrusions (filopodia, neurites, etc.) may be the preferred substrate for analysis. Un-der these circumstances, ex vivo cultures for example of different glial cells—both primary and cell lines—at high purity are still the substrate of choice to date.

To overcome the limitations of ex vi-vo culturing and enrichment, several iso-lation techniques have been developed. A powerful method for isolating small de-fined brain regions and individual cell-types is “Laser-capture microdissection” (LCM), which combines high-resolution

microscopy with precise microlaser cut-ting in one single device. Simple Nissl-staining can be applied to precisely guide regional selection and to identify a few neuronal cell types such as large-sized Purkinje cells of the cerebellum and mo-tor neurons of the spinal cord [13, 19]. We have adapted this technique by using a nu-clear targeted fluorescent protein to genet-ically label projection neurons of the adult cortex and could show that samples with cellular resolution revealed transcriptom-ic signatures, which were masked even in layer-specific cortical microregions [19]. Nonetheless, LCM is considered rather an enrichment than purification strate-

gy given the fact that for example minor proportions of glial transcripts originating most likely from cellular processes will be co-isolated from semithin cryosections. Small-sized glial cells are at the limit of the resolution of the technique. Moreover, due to the low amount of isolated sample globally scaled proteomics has not been combined with LCM purification.

A more widely applied and less labor intensive isolation technique, particularly for small-sized glial cells, relies on tissue trituration followed by fluorescence-ac-tivated cell sorting (FACS). This strategy has been successfully used by us and oth-ers in combination with quantitative real-

Fig. 1 8 Identification of transporter genes enriched in forebrain astrocytes by transcriptome analysis. a Experimental strat-egy: Cortex (Cx), Hippocampus (Hi), and Brainstem (Bs) tissues from brains of mice expressing enhanced green fluorescent protein (EGFP) under the control of the glial-specific promoter glial fibrillary acidic protein (hGFAP-EGFP) were triturated and 50k EGFP positive cells were purified by fluorescence-activated cell sorting (FACS), followed by RNA isolation and transcrip-tome profiling with RNAseq. SR101 candidate uptake transporter genes (of the solute carrier, Slc family) were hypothesized to display a higher expression in Cx and Hi versus Bs astrocytes (Cx/Hi > Bs). b Scatter plots of FACS analysis with green fluores-cent protein intensities plotted versus forward scatter (FSC) given as arbitrary units. Blue dots represent Hoechst 33342 pos-itive and GFP negative cells, green dots represent Hoechst 33342 and GFP double positive cells representing different astro-cyte populations. c Venn diagram of all forebrain-enriched transporter genes. Seven mRNAs were significantly elevated both in Cx and Hi. d Differentially expressed mRNAs coding for solute carriers (Slc and Slco gene families). Depicted are the gene symbols, accession, and average read numbers in Cx, Hi, and Bs samples. The seven detected forebrain-enriched mRNAs cor-respond to five genes (Slc1a2 and Slco1c1 annotated by two mRNAs each). Detection cut-off was determined with an adjust-ed p value cut-off < 0.05. Cx (n = 2), Hi (n = 3), Bs (n = 4) biological replicates. e Scatter plot analysis shows high degree of re-producibility of technical replicates, biological replicates, and astrocyte from different region as well as comparison of astro-cytes and neurons display as expected increasing levels of divergence. f Inspection of ISH data (from Allen Brain Atlas) for top candidates being differentially expressed between Hi and Bs validates regional NGS profiles (a–d). (modified from (Schnell et al. 2013)

64 | e-Neuroforum 3 · 2015

Review article

time polymerase chain reaction, microar-ray analysis, and more recently RNAseq to profile mRNA expression of glial cell pop-ulations [3, 12, 14, 15]. Both, transgenic la-bels introducing for example green flu-orescent protein in astrocytes and anti-bodies directed against extracellular epi-topes have been applied [14]. The caveats of FACS mediated approaches are ex vi-vo incubation steps that need to be strict-ly standardized and which could poten-tially cause artifacts. The induction of im-mediate-early or stress-associated genes has, however, not been observed [14]. The advantage of this method is that all RNA species, mRNAs as well as noncod-ing transcripts, can be harvested simulta-neously (see below). Of particular impor-tance for regulatory mechanisms in nucle-ar organization and transcriptional, post-transcriptional as well as epigenetic pro-cesses in the brain appear to be small and long ncRNAs [18]. Comprehensive char-acterizations of these molecules have not been described in different glia popula-tions so far. Given the enormous progress in the sensitivity and coverage of mass spectometry based technologies, globally scaled proteomic analyses for example of different FACS enriched cell populations are already on the way to deliver the full complement of all expressed proteins in neuronal and glial cell types.

A partially complementary approach relies on ribosome pull-down protocols that allow, by using a transgenically tagged ribosome component, to study selective-ly the translated transcriptome of a pool of defined cell types from different brain regions, however, ncRNAs cannot be an-alyzed [6].

Transcriptomic approaches to unravel glial heterogeneity

Transcriptomic analyses with DNA mi-croarrays were among the most widely applied globally scaled genomic technol-ogies and have been successfully used to profile neural gene expression in the con-text of normal brain physiology and dis-ease, however, mostly not at the cell-type-specific level [8]. More recently, second or next generation sequencing technolo-gies (“Deep-Sequencing”) principally al-low obtaining a much deeper insight in-

to the “whole transcriptome” in an unbi-ased matter [22]. With RNA-Seq, the de-tection and quantification of all expressed mRNA splice variants and ncRNAs is pos-sible without the a priori knowledge of their exact nature. It has been shown that RNAseq provides a more accurate, quan-titative, and sensitive analysis of global transcriptomes as compared with hybrid-ization based methods such as microar-rays [22]. The maximally meaningful ap-plication of these globally scaled technol-ogies most likely requires a coordinated approach since particularly ncRNAs are essential players in the control of mRNA abundance and initiation of protein trans-lation.

As a first step toward the character-ization of region-specific molecular sig-natures of astrocytes, we have combined FACS with 3’ digital gene expression [20]. In this RNAseq approach, we selectively focused on mRNAs expression profiles of astrocytes FACS isolated from the cor-tex (Cx), hippocampus (Hi), and brain-stem (Bs). We hypothesized that differen-tial expression levels of genes encoding for transporter proteins correlating with the Cx- and Hi-specific uptake of sulforho-damine (SR101) into astrocytes would al-low us to identify the SR101 transporter. From the identified candidates, Slco1c1 was validated with corresponding bio-chemical assays and mouse mutants as the Cx/Hi-specific astrocyte SR101 trans-porter (. Fig. 1a–d) [20]. This study was purely hypothesis driven and utilized on-ly a small fraction of the obtained data. Nonetheless, it clearly showed that tran-scriptomic datasets from different astro-cytes population can help to better un-derstand functional consequences of ex-pression differences. By performing qual-ity controls at all steps of the protocol (av-erage r2 values of > 0.99 for technical repli-cates indicate a high level of reproducibil-ity, see . Fig. 1e) and by applying strin-gent statistical criteria, we are convinced that an unbiased bioinformatic analysis of the complete data set will reveal a deep in-sight into the regional differences of the transcriptomes of astrocytes isolated at P10 from Cx, Hi, and Bs. We can conclude from a preliminary analysis that astro-cytes from these regions are highly heter-ogenous with respect to differential gene

expression beyond transporter. The num-ber of differentially expressed genes at a stringent cutoff (p value > 10−5) is high (Hi vs. Bs = 277, Hi vs. Cx = 1105, and Cx vs. Bs = 724), which most likely reflects sev-eral additional subregion-specific mech-anisms operating in these astrocyte pop-ulations. A preliminary pathway analysis shows, for example, that—among many others—Ca2+ signaling-related gene sets are enriched in forebrain astrocytes (Kan-nayian and Rossner, unpublished).

Proteomic approaches to unravel glial heterogeneity

With an estimated number of approx-imately 10,000 different proteins in each single mammalian cell [16], in-depth identification of a cell’s entire proteome

Abstract

e-Neuroforum 2015 · 6:63–68DOI 10.1007/s13295-015-0009-8© Springer-Verlag Berlin Heidelberg 2015

D.C. Dieterich · M.J. Rossner

Dissecting the regional diversity of glial cells by applying -omic technologies

AbstractNeuronal as well as glial cells contribute to higher order brain functions. Many obser-vations show that neurons and glial cells are not only physically highly intermin-gled but are physiologically tightly con-nected and mutually depend at various levels on each other. Moreover, macrog-lia classes like astrocytes, NG2 cells and oli-godendrocytes are not at all homogenous cell populations but do possess a marked-ly heterogeneity in various aspects simi-lar to neurons. The diversity of differenc-es in morphology, functionality and, cellu-lar activity has been acknowledged recent-ly and will be integrated into a concept of brain function that pictures a neural rath-er than a puristical neuronal world. With the recent progress in “omic” technologies, an unbiased and exploratory approach to-ward an enhanced understanding of glial heterogeneity has become possible. Here, we provide an overview on current tech-nical transcriptomic and proteomic ap-proaches used to dissect glial heterogene-ity of the brain.

KeywordsGlial cell types · Brain · Transcriptomics · Proteomics · Cell isolation

65e-Neuroforum 3 · 2015 |

is still a major challenge for modern pro-teomic technologies. In addition, the total number of different/unique proteins per synapse is estimated to be at approximate-ly 2000–2500 [17], which are often under control of different activity-dependent turnover rates. While today’s state-of-the-art MS instruments routinely sequence single purified proteins with subfemtomo-lar sensitivity, the effective identification of low-abundance proteins is orders of magnitude lower in complex mixtures due to limited dynamic range and sequencing speed and due to the common, strong bi-as toward acquiring MS/MS data on high-er abundance molecules. Hence, to tackle activity-dependent proteome alterations entire, functionally heterogeneous brain regions with different subtypes of neu-rons and macroglial cells are usually be-ing sampled at the loss of cell-type spec-ificity and spatial resolution. The charac-

terization of a proteome is an even more difficult challenge if temporal and spatial aspects of a proteome or a subpopulation of the proteome have to be taken into con-sideration. Thus, the separation and en-richment of the subproteome in question is key for its successful characterization. To overcome the above mentioned limi-tations, several cellular and biochemical (subcellular) enrichment strategies com-bined with proteomics have been devel-oped to increase sensitivity and selectiv-ity for the analysis of neuronal and glial proteomes, and to finally dissolve cellular heterogeneity of neural cells in the brain (. Fig. 2a-b).

Concerning cellular selectivity, intra-cellular proteomes and secretomes from cultured primary astrocytes and astrogli-al cell lines have been characterized in de-tail by several labs as proteomic approach-es as indicated above require a much larg-

er quantity than transcriptomic approach-es. Analysis of primary cortical astrocytes revealed the astonishingly high number of proteins in conditioned media including 1247 putative secreted proteins [21]. More recently 6000 unique protein groups be-longing to the secretome and 7265 unique intracellular protein groups were iden-tified from cultured C8-D1A astrocytes [10]. Moreover, these experiments dem-onstrate activity-dependent changes in intracellular and secreted proteomes even under ex vitro conditions. Similar to in-depth analysis of postsynaptic and presyn-aptic protein fractions, biochemical frac-tionation approaches have been recent-ly used to attain detailed pictures of de-fined subcellular glial proteomes, such as myelin membranes of oligodendrocytes or astroglial membrane fractions. Analy-sis of freshly purified human and murine myelin fractions revealed more than 1000

Fig. 2 8 Marcoglial proteomics approaches. Schematic strategy for the investigation of marcoglial proteomes. a In vitro culti-vation of purified primary marcoglia cells or respective cell lines enables the analysis of the intracellular protein entity as well as secretomes, whereas fractionation and extraction procedures of brain homogenates b allows the identification of subcellu-lar proteomes such as astroglial gliosomes, myelin, and others, however, at the expense of true cell-type specificity. Metabolic labeling of newly synthesized proteins using the noncanonical amino acid azidonorleucine (ANL) and a mutated methionine tRNA synthetase expressed in selected cell types may facility cell-type-specific proteome analysis in combination with “click chemistry” under in situ conditions prior biochemical and mass spectrometrical analyses

66 | e-Neuroforum 3 · 2015

Review article

proteins and about 700 different lipids [9]. In a very recent study, Carney et al. char-acterized the proteome of gliosomes puri-fied from murine brain tissue. Gliosomes are enriched for proteins of the astrocyt-ic exocytosis machinery (marker protein VAMP3), perisynaptic astrocyte process-es (marker protein Ezrin), and astrocyte plasma membrane proteins (e.g.astrocyte membrane glycoprotein Basigin44). While common synaptosomal proteins (such as the presynaptic SNARE proteins and postsynaptic proteins NR1 and PSD-95) were depleted in the gliosome frac-tion, the fraction contained glutamate re-ceptors and a plethora of heteromeric G-protein subunits and small GTPases re-lated to perisynaptic function. In combi-nation with fluorescence-activated sort-ing, such as recently done for VGLUT1-Venus-labeled synaptosomes [2], special-ized astrocytic gliosomes could be inves-tigated in unprecedented detail in the fu-ture. Another approach to decipher glial proteomic heterogeneity might employ bioorthogonal metabolic protein labeling (. Fig. 2c) as recently shown by us for dif-ferent cell types including glial and neu-ronal cells in Drosophila melanogaster [7]. This technique relies on a mutated me-thionine tRNA synthetase (MetRSLtoG), which incorporates, instead of methio-nine, the noncanonical amino acid azido-norleucine (ANL) into nascent proteins. By transgenically expressing the MetRSL-toG cell-specificity can be achieved, whereas restricting feeding ANL to trans-genic flies within a desired time enables temporal specificity. Thus, only in the cells expressing the mutated MetRSLtoG and only during the ANL feeding, the syn-thesized proteins incorporate ANL. Sub-sequently and depending of the type of analysis, ANL is clicked either to a fluo-rescent tag for visualization of ANL-har-boring proteins (FUNCAT; [5]) or clicked to an affinity tag with subsequent analysis of tagged proteins using immunoblot or bulk identification via mass spectrometry (BONCAT [11]). Notably, this approach revealed for the first time the expression of the scaffolding protein Dlg in glia [5, 7] pointing once more to the importance of sub-type-specific proteome analysis for understanding brain function on the global and on the regional level.

Conclusions

Because of its apparent limitations, none of these methods qualifies as “the unique” and generally applicable strategy toward transcriptomic and proteomic approach-es to unravel the molecular complexity for example of regional glial cell popula-tions of the brain. Unresolved issues for proteomic approaches include the analy-sis of splice variants, alternative transla-tion initiation sites, and point mutations of proteins due to both dynamic range and sequence coverage limitations. Final-ly, current proteomics is unable to simul-taneously unlock all the critical determi-nants of cellular proteostasis because spe-cific purification and separation proce-dures have to be employed for each sin-gle protein modifications making it nec-essary to address posttranslational modi-fications one by one. In parallel to the lim-itations mentioned for the proteomics, ex-ploring the whole universe of RNA mol-ecules is still challenging although se-quencing methodologies have dramatical-ly improved throughout the last decade. Nonetheless, kinetic and structural varia-tions as well as for example allele-specific expression require sophisticated bioinfor-matics analyses, which are not yet avail-able for all purposes.

Finally, to obtain a more complete pic-ture and to have the opportunity to iden-tify different technical biases, most like-ly several -omic and isolation strategies should be combined and analyzed in an integrated fashion to increase our un-derstanding of glial heterogeneity of the brain.

Corresponding address

D. C. DieterichInstitute for Pharmacology and ToxicologyOtto-von-Guericke University Magdeburg Leipziger Strasse 44, 39120 MagdeburgDaniela.Dieterich@med.ovgu.de

M.J. RossnerLaboratory of Molecular and Behavioral Neurobiology Department of Psychiatry Nussbaumstr. 7, 80336 MunichMoritz.Rossner@med.uni-muenchen.de

Daniela C. Dieterich studied Biochemistry at the University of Hannover and did her PhD. under the supervision of Dr. Michael Kreutz in the Neuroplasticity research group at the Leibniz Institute for Neurobiology (LIN), Magdeburg. She was a Leopoldina-postdoctoral fellow from 2004–2006 in Erin Schuman’s lab at the California Institute of Technology, Pasadena, USA. In 2008 she was granted a research group at the LIN within the Emmy Noether program of the DFG. In 2011 she was offered the W3 professorship in Pharmacology and Toxicology at the Otto-von-Guericke University in Magdeburg, where she heads the research group Neural Communication and Plasticity as well as the Institute for Pharmacology and Toxicology since spring 2012. Her research group is associated with the Leibniz Institute for Neurobiology, Magdeburg.

Moritz J. Rossner studied Biology at the University of Heidelberg. He performed his PhD work at the Center of Molecular Biology in Heidelberg and worked afterwards as scientific employee at Axaron Bioscience AG. In 2002, he moved as a group leader to the Max-Planck-Institute of Experimental Medicine in Göttingen and did his Habilitation in 2010 in the field of Molecular and Cellular Neurobiology at the Faculty of Biology of the Georg-August-University in Göttingen. In 2013 he took over a position as W2 professor at the Ludwig-Maximilians-University in Munich where he leads the laboratory of Molecular Neurobiology at the Department of Psychiatry.

References

1. Allen NJ, Barres BA (2005) Signaling between glia and neurons: focus on synaptic plasticity. Curr Opin Neurobiol 15:542–548

2. Biesemann C, Grønborg M, Luquet E, Wichert SP, Bernard V, Bungers SR, Cooper B, Varoqueaux F, Li L, Byrne JA et al (2014) Proteomic screening of glu-tamatergic mouse brain synaptosomes isolated by fluorescence activated sorting. EMBO J 33:157–170

3. Cahoy JD, Emery B, Kaushal A, Foo LC, Zamanian JL, Christopherson KS, Xing Y, Lubischer JL, Krieg PA, Krupenko SA et al (2008) A transcriptome da-tabase for astrocytes, neurons, and oligodendro-cytes: a new resource for understanding brain de-velopment and function. J Neurosci 28:264–278

4. Clarke LE, Barres BA (2013) Emerging roles of as-trocytes in neural circuit development. Nat Rev Neurosci 14:311–321

5. Dieterich DC, Hodas JJL, Gouzer G, Shadrin IY, Ngo JT, Triller A, Tirrell DA, Schuman EM (2010) In situ visualization and dynamics of newly synthesized proteins in rat hippocampal neurons. Nat Neurosci 13:897–905

6. Doyle JP, Dougherty JD, Heiman M, Schmidt EF, Stevens TR, Ma G, Bupp S, Shrestha P, Shah RD, Doughty ML et al (2008) Application of a transla-tional profiling approach for the comparative anal-ysis of CNS cell types. Cell 135:749–762

7. Erdmann I, Marter K, Kobler O, Niehues S, Abele J, Müller A, Bussmann J, Storkebaum E, Ziv T, Thom-as U et al (2015) Cell-selective labeling of pro-teomes in Drosophila melanogaster. Nat Com-mun doi:10.1038/ncomms8521

8. Geschwind DH, Konopka G (2009) Neuroscience in the era of functional genomics and systems biolo-gy. Nature 461:908–915

67e-Neuroforum 3 · 2015 |

9. Gopalakrishnan G, Awasthi A, Belkaid W, De Far-ia O, Liazoghli D, Colman DR, Dhaunchak AS (2013) Lipidome and proteome map of myelin mem-branes. J Neurosci Res 91:321–334

10. Han D, Jin J, Woo J, Min H, Kim Y (2014) Proteomic analysis of mouse astrocytes and their secretome by a combination of FASP and StageTip-based, high pH, reversed-phase fractionation. Proteomics 14:1604–1609

11. Hodas JJL, Nehring A, Höche N, Sweredoski MJ, Pielot R, Hess S, Tirrell DA, Dieterich DC, Schuman EM (2012) Dopaminergic modulation of the hip-pocampal neuropil proteome identified by bioor-thogonal noncanonical amino acid tagging (BON-CAT). Proteomics 12:2464–2476

12. Lalo U, Pankratov Y, Wichert SP, Rossner MJ, North RA, Kirchhoff F, Verkhratsky A (2008) P2 × 1 and P2 × 5 subunits form the functional P2X receptor in mouse cortical astrocytes. J Neurosci 28:5473–5480

13. Lobsiger CS, Boillée S, Cleveland DW (2007) Tox-icity from different SOD1 mutants dysregulates the complement system and the neuronal regen-erative response in ALS motor neurons. Proc Natl Acad Sci U S A 104:7319–7326

14. Lovatt D, Sonnewald U, Waagepetersen HS, Schousboe A, He W, Lin JH-C, Han X, Takano T, Wang S, Sim FJ et al (2007) The transcriptome and metabolic gene signature of protoplasmic as-trocytes in the adult murine cortex. J Neurosci 27:12255–12266

15. Malatesta P, Hartfuss E, Götz M (2000) Isolation of radial glial cells by fluorescent-activated cell sorting reveals a neuronal lineage. Development 127:5253–5263

16. Pandey A, Mann M (2000) Proteomics to study genes and genomes. Nature 405:837–846

17. Pielot R, Smalla K-H, Müller A, Landgraf P, Lehm-ann A-C, Eisenschmidt E, Haus U-U, Weismantel R, Gundelfinger ED, Dieterich DC (2012) SynProt: a database for proteins of detergent-resistant syn-aptic protein preparations. Front Synaptic Neurosci 4:1

18. Qureshi IA, Mehler MF (2012) Emerging roles of non-coding RNAs in brain evolution, develop-ment, plasticity and disease. Nat Rev Neurosci 13:528–541

19. Rossner MJ, Hirrlinger J, Wichert SP, Boehm C, Ne-wrzella D, Hiemisch H, Eisenhardt G, Stuenkel C, von Ahsen O, Nave K-A (2006) Global transcrip-tome analysis of genetically identified neurons in the adult cortex. J Neurosci 26:9956–9966

20. Schnell C, Shahmoradi A, Wichert SP, Mayerl S, Ha-gos Y, Heuer H, Rossner MJ, Hülsmann S (2015). The multispecific thyroid hormone transporter OATP1C1 mediates cell-specific sulforhodamine 101-labeling of hippocampal astrocytes. Brain Struct Funct 220:193–203. doi: 10.1007/s00429-013-0645-0

21. Skorupa A, Urbach S, Vigy O, King MA, Chaumont-Dubel S, Prehn JHM, Marin P (2013) Angiogenin in-duces modifications in the astrocyte secretome: relevance to amyotrophic lateral sclerosis. J Pro-teomics 91:274–285

22. Wang Z, Gerstein M, Snyder M (2009) RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet 10:57–63

Abstract

68 | e-Neuroforum 3 · 2015

Review article

Introduction

Oligodendrocytes are an important cell type in the central nervous system (CNS). They generate myelin, which is responsi-ble for the fast, saltatory nerve conduction. Until a few years ago it was believed that myelination takes place only during devel-opment and is completed at young adoles-cence. However, the progenitor cells that generate oligodendrocytes during devel-opment are also present in the adult brain and spinal cord. These oligodendrocyte progenitor cells are also known as poly-dendrocytes or NG2 glia, because of their appearance and the expression of the pro-teoglycan NG2. They generate not only oli-godendrocytes but also astrocytes in a re-gion-dependant manner during develop-ment [12]. Further, they are the only pro-liferating cell population outside the neuro-genic niches in the adult brain and are able not only to renew themselves, but also to differentiate into mature, myelin-produc-ing oligodendrocytes [4]. During the last years it became obvious that NG2 glia have even more functions in the brain. Thus, NG2 glia express a variety of ion channels and can form excitatory as well as inhibi-tory synapses with neurons (see article by C. Steinhäuser and D. Dieterich in this is-sue). Despite the high number of NG2 glia in the adult brain (5–10 % of all cells) und their fundamental ability to generate oligo-dendrocytes, this maturation process is not sufficient to ensure efficient myelin repair after disease or injury. A targeted increase

in the maturation of NG2 glia under path-ological conditions like Multiple Sclerosis or injury would be beneficial, but first more detailed information about their differen-tiation, the signalling processes during de-velopment and the differences between NG2 glia in young and aged individuals are required. Another heavily discussed question of the last years with relevance for repair processes in the brain concerns the heterogeneity of the NG2 glia population. This aspect will be investigated as part of the SPP 1757 (Glial heterogeneity).

NG2 glia in the developing CNS

NG2 glia are first detected rather late dur-ing ontogenesis. In the murine spinal cord NG2 glia develop after several neuronal subpopulations have already been gen-erated. As cells of neuroectodermal or-igin, NG2 glia develop from the neuro-epithelial progenitor cells of the ventric-ular zone [8]. In the spinal cord NG2 glia originate in the ventral area from precur-sor cells that express the transcription fac-tor Olig2 and have generated motor neu-rons before. The Olig2 expression is sus-tained in NG2 glia. Remarkably addition-al NG2 glia are formed later in the dor-sal region of the ventricular zone. Also in other regions of the CNS NG2 glia de-velop in different areas of the ventricular zone in a distinct temporal order. In the forebrain the medial ganglionic eminence is the earliest origin for NG2 glia, followed by the lateral ganglionic eminence and fi-nally the dorsal telencephalon. As a con-sequence NG2 glia have various origins along the dorso-ventral and anterior-pos-

terior axes. As the cells leave the ventricu-lar zone, the growth factor receptor Pdg-fra and the transcription factor Sox10 are induced. The expression of the proteogly-can NG2 follows with a slight delay. De-spite their heterogeneity in origin most NG2 glia are characterised by the joint expression of Olig2, Sox10, Pdgfra and the eponymic NG2 marker protein. NG2 glia retain their ability to undergo cell di-vision also in the parenchyma and are able to efficiently populate the whole CNS from their defined site of origin, due to their high migratory activity. As a conse-quence, many CNS areas are already pop-ulated by “ventral” NG2 glia, before “dor-sal” NG2 glia are even generated. In line with this, NG2 glia with dorsal origin on-ly represent a small proportion of the to-tal population in the spinal cord. Howev-er, in the forebrain many of the early NG2 glia with ventral origin are later replaced by NG2 glia with dorsal origin. These ob-servations argue that “ventral” and “dor-sal” NG2 glia may have functional dif-ferences. However, an argument against this hypothesis is that so far no differenc-es in the properties of "ventral" and "dor-sal" NG2 glia or mature oligodendrocytes could be demonstrated and that the selec-tive ablation of NG2 glia of a certain origin can be compensated in the mouse brain by remaining NG2 glia without phenotypical consequences. However, it needs to be stat-ed, that the missing proof of a phenotype does not necessarily mean that this pheno-type does not exist. It cannot be excluded that NG2 glia from different origins are het-erogeneous under normal conditions, but show a high rate of plasticity in case of dis-

Leda Dimou1 · Michael Wegner2

1  Physiological Genomics, Biomedical Center (BMC),

Ludwig-Maximilians University Munich, Munich, Germany2  Institute of Biochemistry, Emil-Fischer-Zentrum,

Friedrich-Alexander-University Erlangen-Nürnberg, Erlangen, Germany

Oligodendroglial heterogeneity in time and space (NG2 glia in the CNS)

e-Neuroforum 2015 · 6:69–72DOI 10.1007/s13295-015-0014-yPublished online: 27 August 2015© Springer-Verlag Berlin Heidelberg 2015

69e-Neuroforum 3 · 2015 |

Review article

A complete list of literature can be requested from the authors.

turbances in CNS development or injury so they are able to compensate for each other.

Relation of NG2 glia in the developing and adult CNS

During CNS development most NG2 glia differentiate to oligodendrocytes. Due to this, NG2 glia are usually referred to as ol-igodendrocyte progenitor cells (OPCs) in developmental studies. Disturbances in development and differentiation of NG2 glia are related to demyelinating diseas-es and leukodystrophies. Part of the NG2 glia population does not differentiate, but remains as an autonomous cell popula-

tion in the adult CNS. At the moment it is unclear, which signals and mechanisms decide, whether a NG2 glia cell differenti-ates further or stays at a progenitor state. Further, NG2 glia have to adapt to the modified situation in the adult brain in various characteristics and abilities. How this is achieved on mechanistic and mo-lecular levels is currently unclear and part of our research in the SPP 1757 (Glial het-erogeneity).

NG2 glia in the adult brain

In the adult brain NG2 glia are able to proliferate and to renew themselves, but

in contrast to developmental states, the cell cycle is rather slow. In the last years it has been shown, that NG2 glia gener-ate mature, myelin-generating oligoden-drocytes also in adulthood and hence at a time point, when myelination has already been completed. The role of these newly generated oligodendrocytes is not known. Recent studies lead to the assumption, that these oligodendrocytes could be im-portant for complex, motor learning tasks [7]. Interestingly, for adult NG2 glia the ability to proliferate and divide is strong-ly region-dependent. While the majority of NG2 glia in the white matter of the ce-rebral cortex (e.g. in the corpus callosum) generates mature, myelin-producing oli-godendrocytes, most of the NG2 glia in the grey matter keep their identity as progenitor cells. Homo- and heterotop-ic transplantations in the brain of adult mice point to a substantial heterogene-ity of NG2 glia in different brain regions, which can either be explained by different intrinsic determinants or by divergent en-vironmental stimuli [9]. Further, the pro-portion of proliferating NG2 glia is small-er in the grey matter and cell cycle length is increased, compared to NG2 glia resid-ing in the white matter (. Fig. 1). Prolif-eration and differentiation can also be in-fluenced by neuronal modulation. This can be explained by the fact that NG2 glia interact closely with synapses and nodes of Ranvier and some NG2 glia are al-so in contact with neurons via synapses. All of these observations underline, that under physiological conditions NG2 glia act as a heterogeneous population, which can react differently to a variety of stimu-li. This heterogeneity needs to be under-stood, before therapeutics based on NG2 glia can be developed and applied to re-myelination and repair (for a review ar-ticle see [2]).

Also within the same region heteroge-neity of NG2 glia was observed. The tran-scription factor Mash1/Ascl1 was on-ly detected in a subpopulation of NG2 glia in the cerebral cortex. It remains un-clear, whether Ascl1-positive and –nega-tive NG2 glia show differences in prolifer-ation- and differentiation behaviour. Also the G-protein coupled membrane recep-tor 17 (GPR17) is only expressed in a sub-population of NG2 glia in the brain. Sur-

Fig. 1 8 NG2 glia are a heterogeneous population under physiological (left) and pathological (right) conditions: NG2 glia in the grey matter have a long cell cycle, differentiate slowly and only partially to mature oligodendrocytes with most cells remaining as NG2 glia. In contrast, NG2 glia located in the white matter have a shorter cell cycle and differentiate faster and in a higher amount into oligoden-drocytes. After injury (. Fig. 1, right) NG2 glia in the grey matter show a very fast and heterogeneous reaction. They become hypertrophic, polarize, migrate and proliferate more strongly with a shorter cell cycle

70 | e-Neuroforum 3 · 2015

Review article

prisingly, NG2 glia expressing GPR17 do not develop to mature oligodendrocytes in an investigated period of 3 months. As part of SPP1757, the molecular differences between GPR17-positive and GPR17-neg-ative NG2 glia will be analysed to identify molecules and signalling pathways that are important and necessary for the differenti-ation of NG2 glia.

Regulatory mechanisms of NG2 glia

Different properties of NG2 glia in the de-veloping and in the adult brain are caused by alterations of the gene regulatory net-work. For NG2 glia in the developing CNS the main components of the net-work are known [5]. Among those are the two transcription factors and marker pro-teins for NG2 glia Olig2 and Sox10. Both are supported in their function by close-ly related proteins (Olig1 for Olig2; and Sox8, Sox9 support Sox10), but are func-tionally dominant compared to the cor-responding paralogous proteins. Their effect is further modulated by addition-al transcription factors. The helix-loop-helix proteins Id2 and Id4 inhibit Olig2 and the transcription factor Sox10 is reg-ulated in its activity by the related pro-teins Sox5 and Sox6. Additional crucial regulators are the helix-loop-helix pro-tein Mash1/Ascl1, the zinc finger pro-tein Sip1 and the homeodomain protein Nkx2.2. New studies give first insights in-to the complex interactions of these tran-scription factors inside the gene-regulato-ry network that involves antagonistic as well as synergistic and inductive relations and establish a variety of control circuits, which additionally are influenced by the exact amount of the corresponding tran-scription factor, post-translational modi-fications and epigenetic factors [10, 11]. Currently, there is insufficient knowledge how the gene regulatory network in NG2 glia during development differs from the adult CNS. However, current research hints at some functionally relevant differ-ences. The influence of Olig2 and Olig1 on the differentiation of NG2 glia appears to be interchanged between developing and adult CNS. Whereas Olig2 is respon-sible for many developmental aspects und properties of NG2 glia in the developing

CNS and can not be substituted by Olig1, regeneration and differentiation of adult NG2 glia depends on Olig1 and is drasti-cally reduced if only Olig2 is present [1]. By using comparative genomics to inves-tigate the regulatory network we hope to receive important information about the causes for differences and similarities be-tween NG2 glia in the developing and in the adult brain. Further, we will tackle the question, how the modulation of the net-work influences the heterogeneity of NG2 glia during adulthood.

Response of NG2 glia after injury

NG2 glia react towards injurý or other pathological conditions with a change in their morphology and proliferation rate. The type and time course of this reaction is dependent on the nature of the insult. Af-ter demyelination NG2 glia become hyper-trophic, more cells proliferate with a short-ened cell cycle and differentiate finally to mature oligodendrocytes to repair myelin. NG2 glia react similarly in neurodegenera-tive diseases such as Morbus Alzheimer or Amyotrophic Lateral Sclerosis (ALS), but no hypertrophy can be detected in these cases. It remains unclear, whether the re-action of NG2 glia also differs between dif-ferent kinds of injury and other patholog-ical states. Using repetitive in vivo 2-pho-ton microscopy, first evidences for a het-erogeneous reaction of NG2 glia towards acute injury (e.g. focal laser lesion and stab wound) was detected ([6]; von Streitberg, Dimou et al., unpublished data). Some NG2 glia polarize and migrate towards the site of injury, while others proliferate and/or become hypertrophic (. Fig. 1). All these events lead to an accumulation of NG2 glia at the lesion site. The relevance of this be-haviour remains unclear (for a review see [3]). Speed and strength of the reaction lead to the assumption that NG2 glia are respon-sible for wound closure and scar formation.

Corresponding address

L. DimouPhysiological Genomics, Biomedical Center (BMC)Ludwig-Maximilians University Munich 80336 Munichleda.dimou@lrz.uni-muenchen.de

Leda Dimou studied biology at the Ruprecht-Karls-University Heidelberg and did her PhD at the Center for Molecular Biology (ZMBH) in Heidelberg and at the Max-Planck-Institute for Experimental Medicine in Göttingen focussing on the homologs of the main myelin protein PLP in the brain. After a postdoctoral fellowship at the Institute for Brain Research in Zurich she moved to the Department of Physiological Genomics at the Biomedical Center of the Medical Faculty (LMU) in Munich, where she leads her own research group since 2012. Her research group investigates the role of NG2 glia in the adult brain under physiological and pathological conditions using molecular and cell biological- as well as imaging methods.

Michael Wegner studied biology at the universities of Münster and Würzburg and did his PhD from 1987 to 1990 at the Institute of Biochemistry in Würzburg. After postdoctoral work at the University of California at San Diego (UCSD) he took over a position as leader of a junior research group at the Center for Molecular Neurobiology at Hamburg University (ZMNH) in 1994 and was appointed to the chair of Biochemistry and Pathobiochemisty at the Institute of Biochemistry of the Medical Faculty at the Friedrich-Alexander-University Erlangen-Nürnberg in 2000. His research is dealing with the transcriptional and epigenetic control of glial cells and myelin formation.

Abstract

e-Neuroforum 2015 · 6:69–72DOI 10.1007/s13295-015-0014-y© Springer-Verlag Berlin Heidelberg 2015

L. Dimou · M. Wegner

Oligodendroglial heterogeneity in time and space (NG2 glia in the CNS)

AbstractNG2 glia represent a neural cell population that expresses the proteoglycan NG2 and is distinct from other cell types of the central nervous system. While they generate oligo-dendrocytes and a subset of astrocytes dur-ing development, their progeny in the adult brain solely consists of oligodendrocytes and further NG2 glia. In the last years, it has be-come clear that NG2 glia represent a hetero-geneous population of cells with different properties and potential. In this review we will first discuss the similarities and differenc-es between NG2 glia of the developing and adult CNS, before we will describe the regula-tory mechanisms in these cells to finally con-centrate on the heterogeneity of NG2 glia under physiological and pathological con-ditions.

KeywordsHeterogeneity · Differentiation · Proliferation · Transcription factors · Injury

71e-Neuroforum 3 · 2015 |

Acknowledgment. The work of the authors is supported by the DFG (We1326-8, We1326-11 and We1326-12 for MW & DI 1763/1–1 and SFB870 for LD) and the EU/DLR (01EW1306A for LD). We thank Frau Dr. Francesca Vigano for the graphic illustration.

References

1. Arnett HA, Fancy SP, Alberta JA, Zhao C, Plant SR, Kaing S, Raine CS, Rowitch DH, Franklin RJ, Stiles CD (2004) bHLH transcription factor Olig1 is re-quired to repair demyelinated lesions in the CNS. Science 306(5704):2111–2115

2. Dimou L, Gallo V (2015) NG2-glia and their func-tions in the CNS. Glia 63(8):1429–1451

3. Dimou L, Götz M (2014) Glial cells as progenitors and stem cells: new roles in the healthy and dis-eased brain. Physiol Rev 94(3):709–737

4. Dimou L, Simon C, Kirchhoff F, Takebayashi H, Götz M (2008) Progeny of Olig2-expressing progenitors in the gray and white matter of the adult mouse cerebral cortex. J Neurosci 28(41):10434–10442

5. Hernandez M, Casaccia P (2015) Interplay between transcriptional control and chromatin regulation in the oligodendrocyte lineage. Glia 63(8):1357–1375

6. Hughes EG, Kang SH, Fukaya M, Bergles DE (2013) Oligodendrocyte progenitors balance growth with self-repulsion to achieve homeostasis in the adult brain. Nat Neurosci 16(6):668–676

7. McKenzie IA, Ohayon D, Li H, de Faria JP, Emery B, Tohyama K, Richardson WD (2014) Motor skill learning requires active central myelination. Sci-ence 346(6207):318–322

8. Richardson WD, Kessaris N, Pringle N (2006) Oligo-dendrocyte wars. Nat Rev Neurosci 7(1):11–18

9. Viganò F, Möbius W, Götz M, Dimou L (2013) Trans-plantation reveals regional differences in oligo-dendrocyte differentiation in the adult brain. Nat Neurosci 16(10):1370–1372

10. Weider M, Küspert M, Bischof M, Vogl MR, Hornig J, Loy K, Kosian T, Müller J, Hillgärtner S, Tamm ER, Metzger D, Wegner M (2012) Chromatin-remodel-ing factor Brg1 is required for Schwann cell differ-entiation and myelination. Dev Cell 23(1):193–201

11. Weider M, Wegener A, Schmitt C, Küspert M, Hill-gärtner S, Bösl MR, Hermans-Borgmeyer I, Nait-Oumesmar B, Wegner M (2015) Elevated in vivo levels of a single transcription factor directly con-vert satellite glia into oligodendrocyte-like cells. PLoS Genet 11(2):e1005008

12. Zhu X, Bergles DE, Nishiyama A (2008) NG2 cells generate both oligodendrocytes and gray matter astrocytes. Development 135(1):145–157

72 | e-Neuroforum 3 · 2015

Review article

Introduction

NG2 glia represent a subgroup of glial cells in the brain which express the NG2 pro-teoglycan. Earlier research assumed that these cells are immature astrocytes, but it is now well established that they comprise a distinct cell type. Owing to their specif-ic functional and morphological proper-ties, NG2 glial cells have also been termed complex cells, oligodendroglial precursor cells (OPCs), GluR cells or polydendro-cytes [1]. NG2 glial cells make up approx-imately 5–10 % of the total cell number in the brain, and many of these cells under-go cell division throughout postnatal life. During brain development, the vast ma-jority of white matter NG2 glia differen-tiate into NG2-negative oligodendrocytes and thus can be considered immature pre-cursors. In contrast, in gray matter, many cells do not become oligodendrocytes but keep their NG2 phenotype throughout life (see article by L. Dimou and M. Wegner in this issue). Upon injury to the adult brain, NG2 glia have been observed to differen-tiate into oligodendrocytes or astrocytes, that is, they show properties characteris-tic of progenitors. However, it is unclear whether all NG2 glial cells can undergo such a type of differentiation. NG2 glia ex-press a plethora of ion channels and plas-ma membrane receptors with properties similar to those found in neurons. In con-trast to astrocytes, which form abundant inter-cellular networks among each oth-er and with oligodendrocytes (so-called panglial networks), in most brain regions, NG2 glial cells lack coupling. Intriguing-ly, these glial cells receive direct synaptic input from GABAergic and glutamater-gic neurons.

NG2 glia possess heterogeneous functional properties

Ion channels

So far, NG2 glial cells have been electro-physiologically analysed only in a few brain regions, such as the cerebellum, neocortex, hippocampus and corpus cal-losum. These cells possess voltage-gated Na+ channels, but they cannot fire regen-erative action potentials. The lack of ac-tion potentials is due to the low number of plasma membrane Na+ channels and the much higher densities of voltage-gat-ed delayed rectifier and transient K+ chan-nels because opening of the latter on de-polarization produces outward currents that largely compensate the Na+ inward currents. NG2 glia also express inwardly rectifying K+ (Kir) channels (. Fig. 1a). While Kir channels are instrumental in stabilizing the cells’ negative resting po-tential, the density of the delayed rectifi-er K+ channels (subunits Kv1.3, Kv1.5) was reported to correlate with cell division: High expression of the channels was ob-served in proliferating cells, and the block of Kv1.3 inhibited the transition from the G1 to the S phase of the cell cycle [2]. Analyses of mice with gene deletions, an-tibody staining and single-cell transcript profiles have shown that the Kir channels in NG2 glia are mainly composed of the Kir4.1 subunit, which is also abundant-ly expressed by astrocytes. In the hippo-campus, Na+ channels are downregulated during postnatal development while the density of Kir channels increases.

In addition to Na+ and K+ channels, different types of voltage-gated Ca2+ chan-nels have been described in NG2 glia,

which together with transmitter receptors (see below) might be important to regu-late Ca2+ -dependent cellular processes, such as motility, proliferation or migration [3, 4]. Independent of the developmental stage, NG2 glia in gray and white matter differ with respect to their morpho logical and physiological properties. For exam-ple, compared to subcortical white mat-ter, NG2 glial cells in the neocortex ex-press a lower density of delayed rectifi-er K+ channels but higher densities of Kir and Na+ channels [5]. The physiological impact of this diversity is still unclear.

Transmitter receptors

NG2 glial cells possess different types of ionotropic glutamate receptors (main-ly AMPA subtype), which mediate de-polarization and Ca2+ influx into the cell [1]. Significant regional differences have been reported regarding the expression strength and properties of glial AMPA re-ceptors. For example, Ca2+ permeability and density of the receptors in NG2 glia of the cerebellum are much higher com-pared with those of the hippocampus, en-tailing a higher efficiency of neuron–glia synaptic signaling. The question whether differences in the molecular composition of the receptors add to the observed func-tional heterogeneity is addressed by our research project within the framework of the DFG Priority Program 1757. We have found that hippocampal glial AMPA re-ceptors mainly express the GluA2 subunit, which is less Ca2+ permeable. The Na+ in-flux through the receptors leads to a block of Kir channels and enhanced receptor-mediated depolarization of the postsyn-aptic glial membrane [1].

Christian Steinhäuser1 · Dirk Dietrich2

1  Institute of Cellular Neurosciences, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany2  Experimental Neurophysiology, Neurosurgery Hospital, University of Bonn, Bonn, Germany

Neuron–glia synapses in the brain: properties, diversity and functions of NG2 glia

e-Neuroforum 2015 · 6:73–77DOI 10.1007/s13295-015-0010-2Published online: 25 July 2015© Springer-Verlag Berlin Heidelberg 2015

73e-Neuroforum 3 · 2015 |

Review article

NG2 glia bear GABAA receptors which also mediate depolarization of the mem-brane through the efflux of Cl−. This ef-

flux is due to a relatively high intracellular Cl− concentration resulting from the ab-sence of transporters (e.g. KCC2), which

move Cl− outwardly in neurons. Com-bined pharmacological and transcript analyses in single cells unraveled the sub-

Fig. 1 8 Properties of synaptic inputs to NG2 cells in acute slices of the hippocampus from juvenile transgenic hGFAP/EGFP mice. a The morphology of NG2 cells was visualized by Texas Red dextran filling during whole-cell recording. Subsequent con-focal analysis and 2D maximum projection showed the extensive arborisation of their processes. Note the typical nodules ap-pearing as dots along the fine processes (bar 10 µm). A typical current pattern is given in the middle panel. Current respons-es were evoked by de- and hyperpolarizing the membrane between + 20 and − 160 mV (holding potential − 80 mV, 10 mV steps, bars 1 nA, 10 ms). Post-recording immunostaining and triple fluorescence confocal analysis were applied to test for NG2 immunoreactivity. The middle panel shows three separated colour channels of one confocal plane. Texas Red dextran la-belling is given in green (g), NG2 immunoreactivity in red (r) and EGFP expression in blue (b). The superimposed RGB picture (right) shows the membrane-associated distribution of the NG2 immunoreactivity of the recorded cell (yellow details). b Post-synaptic currents were recorded from an NG2 cell upon short single pulse stimulation. At least two types of presynaptic neu-rons, glutamatergic CA3 pyramidal neurons and GABAergic interneurons, innervate hippocampal NG2 cells. Corresponding glial postsynaptic currents can be distinguished by their current kinetics and pharmacological profiles. [1]

74 | e-Neuroforum 3 · 2015

Review article

unit composition of the glial GABAA re-ceptors in the hippocampus and neocor-tex. In addition to various α and β iso-forms, the cells also have the γ2 subunit, although its expression is confined to the postsynaptic membrane; on the oth-er hand, extrasynaptic GABAA receptors lack this subunit [6]. In the neocortex, γ2 is downregulated during postnatal devel-opment and this change is accompanied by a loss of the GABAergic synaptic inner-vation of NG2 glia [7].

Besides ionotropic receptors, NG2 gli-al cells also carry metabotropic glutamate receptors (mGluR1, mGluR5), whose acti-vation entails Ca2+ -induced release of in-tracellular Ca2+ [3]. Other studies have re-ported functional expression of puriner-gic and nicotinic acetylcholine receptors in NG2 glia, which also mediate increase in the intracellular Ca2+ concentration.

Synaptic signaling between neurons and NG2 glial cells

NG2 glia receive synaptic input from neurons

Over the past years, NG2 glial cells have attracted much attention for two main reasons: (i) As described earlier, these cells express a plethora of voltage-gated ion channels, and (ii) intriguingly, almost all NG2 glial cells receive synaptic input from neighboring neurons. The synaptic input is mediated through classical vesicular transmitter release at morphological con-tact points, similar to classical neuron–neuron synapses (. Fig. 2). The proper-ties of postsynaptic currents in NG2 glia are also indistinguishable from those re-corded in neurons (. Fig. 1b). NG2 glial cells express a variety of neurotransmitter receptors (cf. above), but synaptic release activates virtually exclusively AMPA and GABAA receptors, as could be demon-strated by the complete block of the post-synaptic receptors with specific antago-nists (. Fig. 1b). Dwight Bergles was the first to report functional synapses on NG2 glia in the hippocampus [8]. Since then, several studies have shown that in vari-ous gray and white matter regions, for ex-ample, the neocortex, hippocampus, cer-ebellum, corpus callosum and optic nerve, NG2 glial cells receive glutamatergic and

GABAergic synaptic inputs. Quantal syn-aptic currents in NG2 glia (due to exocy-tosis of a single transmitter-filled vesicle) have amplitudes of approximately 10 pA and very fast (millisecond range) rising and decaying phases. As in neurons, glu-tamatergic synaptic currents reverse di-rection at approximately 0 mV, while GA-BA-induced postsynaptic currents reverse at approximately − 40 mV (cf. above). Ac-cordingly, in NG2 glia, GABA is not an inhibitory transmitter but depolarizes the cells similarly to glutamate. NG2 glial cells receive much fewer synapses than neu-rons, with the innervation in the cerebel-lum being stronger than that in the hip-pocampus. Glial glutamatergic synapses outnumber GABAergic synaptic inputs by approximately fivefold.

Synaptic innervation as a signal regulating glial differentiation

Increasing evidence suggests that synaps-es on NG2 glia might adjust the develop-ment of oligodendrocytes to neuronal ac-tivity. This idea is still speculative because convincing experimental evidence is still missing, and there are many possibilities regarding how neuronal activity might influence the differentiation of NG2 cells. However, considering the enormous bio-logical effort needed to establish and sus-tain this non-conventional signal pathway between neurons and NG2 cells, it appears likely to assume a specific impact of this mechanism. Actually, (i) synapses on NG2 glial cells exist in white and gray matter throughout the life span; (ii) establish-ing synapses, particularly with projecting axons, requires significant transport ef-forts; (iii) synapses are already established shortly after birth; (iv) surprisingly, neu-ron–NG2 glia synapses are retained dur-ing cell division and passed on to daugh-ter cells; (v) during differentiation into oli-godendrocytes, the release machinery of NG2 cells is rapidly degraded; (vi) even adult-born NG2 glia, which upon white matter injury migrate to the lesion site and contribute to regeneration, receive synap-tic input.

So far, most properties of neuron–glia communication have been determined in rodents during the first two postna-tal weeks. It is possible that in the third

and fourth weeks, when myelination cul-minates, the electrical impact of released transmitter molecules increases owing to enhanced synaptic strength. This corre-lation suggests that the synaptic input to NG2 glia modulates oligodendrogenesis to adjust it to the needs of the neuronal electrical circuitry.

Synaptic integration in NG2 glia

Electrical integration of synaptic depolari-sation through spatial and temporal sum-mation of postsynaptic events represents a classical mechanism of signal processing in neurons. These cells integrate the in-put in their dendrites and conduct the re-sulting signal to their axon hillock, where eventually action potentials are generat-ed. However, NG2 glial cells lack axons and do not generate action potentials, in-

Abstract

e-Neuroforum 2015 · 6:73–77DOI 10.1007/s13295-015-0010-2© Springer-Verlag Berlin Heidelberg 2015

C. Steinhäuser · D. Dietrich

Neuron–glia synapses in the brain: properties, diversity and functions of NG2 glia

AbstractAlthough NG2 glial cells represent a fre-quent glial cell type in the brain, character-ized by expression of the NG2 proteogly-can, the functional impact of these cells is still enigmatic. A large proportion of NG2 glia are proliferatively active throughout life. These cells express a plethora of ion channels and transmitter receptors, which enable them to detect neuronal activi-ty. Intriguingly, NG2 glial cells receive syn-aptic input from glutamatergic and GA-BAergic neurons. Since these postsynap-tic glial currents are very small, their spa-tial and temporal integration might play an important role. In white matter, most NG2 glial cells differentiate into oligoden-drocytes and this process might be influ-enced through the activity of the afore-mentioned neuron–glia synapses. Increas-ing evidence suggests that the properties of NG2 glia vary across brain regions; how-ever, the impact of this variability is not understood yet.

KeywordsNG2 cells · Neuron–glia synapse · Synaptic integration · Glial differentiation

75e-Neuroforum 3 · 2015 |

dicating that their signal integration fun-damentally differs from neurons. Despite the lack of action potentials, the fast gli-al Na+ channels may amplify or accelerate synaptic depolarisation. It remains to be shown whether these Na+ channels open rapidly enough and if their number is suf-ficient to amplify synaptic depolarisation. In reality, the large number of voltage-gat-ed K+ channels counteracts this depolar-isation, and the net effect will depend on the activation kinetics and amplitudes of the latter.

NG2 glia possess a complex process tree with a surface area of approximate-ly 2000 µm2. Assuming that the diameter of the cell body is 6–7 µm, it follows that the processes make up more than 90 % of the total cell surface and probably receive most of the synaptic input. This raises the question whether NG2 glial cell process-es are competent of local synaptic integra-tion. This has not yet been investigated. However, using a simple “ball and stick” cable model and considering that the length of NG2 glia processes is only one

tenth that of typical principal neurons, we could recently show that the spatial volt-age spread is similar in both cell types. [9]. Accordingly, regarding electrical com-pactness, NG2 glia might be considered a miniature edition of neurons and able to process the input, which could be crucial for initiating activity-dependent myelina-tion. Within this DFG Priority program, we will investigate the role of ion chan-nels and altered intracellular Ca2+ con-centrations and their subcellular localiza-tion in the integration of synaptic signals in NG2 glia.

Functions of NG2 glial cells and neuron–glia synapses

The consequences of receptor activation in NG2 glia through presynaptic trans-mitter release are largely unknown until now because the resulting depolarisations are very small (only a few millivolts in the hippocampus). However, most analyses so far were confined to the cell body and it is possible that receptor activation in the thin peripheral processes entails much stronger depolarization or intracellular Ca2+ increase. Accordingly, it is tempting to speculate that NG2 glial cells command complex local signal integration, which al-lows them to respond to the specific ac-tivity patterns of single presynaptic axons. Indeed, increasing evidence suggests that these non-conventional synapses between neurons and NG2 glia encode important information for cell differentiation. It can be expected that identifying the function-al role of these synapses will be crucial for our general appreciation of the enigmat-ic NG2 cell population within the central nervous system.

Corresponding address

C. SteinhäuserInstitute of Cellular NeurosciencesRheinische Friedrich-Wilhelms-Universität Bonn Sigmund-Freud Str. 25, 53105 Bonnchristian.steinhaeuser@ukb.uni-bonn.de

Christian Steinhäuser studied physics at the Friedrich-Schiller Universität Jena. He received a PhD at the Institute for Neurobiology and Brain Research, Academy of Sciences, Magdeburg and the Faculty of Biology, University of Jena on the functional analysis of Na+ channels in pyramidal

Fig. 2 9 Morphologi-cal properties of neu-ron–glia synapses in acute brain slices of the juvenile hippocam-pus. a Electron micro-scopic visualization of an asymmetric neu-ron–glia synapse. The cell was first function-ally characterized (in-set, pulse protocol as in . Fig. 1b), there-by filled with biocytin, and then fixed and an-alysed ultrastructural-ly. Note the neuronal presynaptic vesicles, synaptic cleft and gli-al postsynaptic struc-ture. b is an enlarged view of the boxed area; the arrow heads indi-cate vesicles. Scale bar: 200 nm. [3]

76 | e-Neuroforum 3 · 2015

Review article

neurons freshly isolated from the hippocampus. After habilitation in physiology, in 1997 he moved to Bonn as a professor for Experimental Neurobiology at the Medical Faculty of the Rheinische Friedrich-Wilhelms-Universität. In 2007, he founded the Institute of Cellular Neurosciences at University of Bonn that he chairs since. He employs molecular, electrophysiological and imaging techniques to investigate the role of glial cells in information processing in the normal and epileptic brain

Dirk Dietrich studied medicine and informatics in Bonn, Hagen and London and received an MD at the Medical Faculty of the Rheinische Friedrich-Wilhelms-Universität Bonn. After habilitation in neurophysiology he accepted a W2 professorship for Experimental Neurophysiology at Neurosurgery Hospital, Bonn. Since 2014 he is a Research Chair at the Neurosurgery Hospital. His research is focused on the analysis of neuron-glia signaling, pre- and extrasynaptic regulation of glutamatergic neurotransmission, Ca2+ -induced vesicular release and synaptic plasticity.

Acknowledgements. Work of the authors is sup-ported by Deutsche Forschungsgemeinschaft (STE 552/5, STE 552/4, DI 853/5-1, DI 853/3-1, SFB 1089). We thank Dr. Ines Heuer for technical support.

References

1. Bergles DE, Jabs R, Steinhäuser C (2010) Neuron-glia synapses in the brain. Brain Res Rev 63:130–137

2. Chittajallu R, Chen Y, Wang H, Yuan X, Ghiani CA, Heckman T, McBain CJ, Gallo V (2002) Regula-tion of Kv1 subunit expression in oligodendrocyte progenitor cells and their role in G1/S phase pro-gression of the cell cycle. Proc Natl Acad Sci U S A 99:2350–2355

3. Haberlandt C, Derouiche A, Wyczynski A, Haseleu J, Pohle J, Karram K, Trotter J, Seifert G, Frotscher M, Steinhäuser C, Jabs R (2011) Gray matter NG2 cells display multiple Ca-signaling pathways and highly motile processes. PLoS One 6:e17575

4. Hughes EG, Kang SH, Fukaya M, Bergles DE (2013) Oligodendrocyte progenitors balance growth with self-repulsion to achieve homeostasis in the adult brain. Nat Neurosci 16:668–676

5. Chittajallu R, Aguirre A, Gallo V (2004) NG2-posi-tive cells in the mouse white and grey matter dis-play distinct physiological properties. J Physiol 561:109–122

6. Passlick S, Grauer M, Schäfer C, Jabs R, Seifert G, Steinhäuser C (2013) Expression of the gamma2-subunit distinguishes synaptic and extrasynaptic GABA(A) receptors in NG2 cells of the hippocam-pus. J Neurosci 33:12030–12040

7. Balia M, Velez-Fort M, Passlick S, Schafer C, Audi-nat E, Steinhäuser C, Seifert G, Angulo MC (2015) Postnatal down-regulation of the GABAA recep-tor gamma2 subunit in neocortical NG2 cells ac-companies synaptic-to-extrasynaptic switch in the GABAergic transmission mode. Cereb Cortex 25:1114–1123

8. Bergles DE, Roberts JD, Somogyi P, Jahr CE (2000) Glutamatergic synapses on oligodendrocyte pre-cursor cells in the hippocampus. Nature 405:187–191

9. Sun W, Dietrich D (2013) Synaptic integration by NG2 cells. Front Cell Neurosci 7:255

77e-Neuroforum 3 · 2015 |

Introduction

Synaptic transmission is the primary means of information transfer in the ner-vous system, and its long-term plastici-ty is thought to underlie acquisition and storage of new information. While these processes are primarily neuronal, they do occur in an environment that contains both extracellular matrix and nonneuro-nal cells such as glia. Especially astrocytes, a subtype of glia cell with a characteristic star-like morphology, have been the sub-ject of intense study over the past decades. An important feature of astrocytes is that, unlike neurons, the branches of an indi-vidual cell occupy clearly delineated terri-tories with little overlap with their neigh-bors under normal conditions. Within these domains, fine astrocyte branches approach and enwrap synapses (. Fig. 1, [20]). A single astrocyte can cover a total of approximately 100,000 synapses arising from many different neuronal connec-tions in the rat hippocampus [6]. As as-trocytes are capable of sensing and modu-lating synapse activity, they are potential-ly powerful modulators of large synaptic populations. Therefore, their role as key players in neuronal network function has received a lot of attention but is also in-tensely debated.

The possibility of fast reciprocal signal-ing between neurons and astrocytes was first suggested in the 1990s. Among the many important findings was the obser-vation of Ca2+ responses in astrocytes in-duced by the neurotransmitter glutamate in cultured astrocytes and in situ [7, 35]. At the same time, an increase of cytosol-ic Ca2+ was shown, for example, to trig-ger release of glutamate from astrocytes, thereby exciting nearby neurons [31].

These and other findings challenged the classical view that the electrically passive astrocytes only act as scaffolding for neu-rons, maintaining ion homeostasis and providing metabolic support and neu-rotransmitter clearance, and gave rise to the concept of the tripartite synapse [3]. A vast number of signaling cascades en-abling neuron–astrocyte communication is well documented now. They are trig-gered by neuronal activity and appear to converge on astrocyte Ca2+ signaling while astrocyte Ca2+ signaling was shown to have a plethora of reciprocal effects on neuronal network function (for review [36] and references therein). For this rea-son, understanding astrocyte Ca2+ signal-ing is critical for establishing the function-al and computational role of astrocyte–neuron interactions, although astrocytes have other means of sensing and integrat-ing neuronal activity, sodium signaling for instance [16, 36].

Astrocyte Ca2+ signaling in astrocyte–neuron communication

Metabotropic neurotransmitter receptors represent a common link between neuro-nal activity and astrocyte Ca2+ signaling. In the juvenile hippocampus for instance, group I metabotropic glutamate receptors mediate somatic store-dependent Ca2+ transients in response to Schaffer collater-al stimulation [35]. Similar pathways exist in hippocampal astrocytes for other pre-synaptically released neurotransmitters such as acetylcholine [25], while multiple mechanisms underlie astrocyte Ca2+ re-sponses to the inhibitory neurotransmit-ter GABA [21]. These examples highlight that astrocytes appear to respond with a

transient Ca2+ increase regardless of the identity of the neurotransmitter and its excitatory or inhibitory action on neu-rons. It raises the question if and how as-trocytes discriminate between activity of the various different presynaptic termi-nals they cover. In addition, astrocyte Ca2+ transients also occur in response to post-synaptic depolarization. For instance, en-docannabinoid release after depolariza-tion of hippocampal CA1 pyramidal cells activates store-dependent Ca2+ signaling in astrocytes [23]. This reduction of ma-ny different incoming signals into a sin-gle type of response, a Ca2+ transient of a hippocampal astrocyte, implies that strik-ingly different neuronal network activity may be translated into apparently similar astrocyte Ca2+ signals.

This convergence onto a seeming-ly uniform Ca2+ signaling is contrast-ed by highly diverging effects that an as-trocyte Ca2+ increase can have on synap-tic transmission (. Fig. 2). Targets of as-trocyte–neuron communication include axonal action potential propagation. For instance, action potentials of CA3 pyra-midal cells are broadened while traveling along their axons and release from their synapses is facilitated if the Ca2+ concen-tration in nearby astrocytes is increased by Ca2+ uncaging [38]. Neurotransmitter release from these synapses is also under direct control of astrocyte Ca2+ signaling. Increasing astrocyte Ca2+ levels can tran-siently or permanently potentiate the re-lease probability of CA3-CA1 synapses by triggering glutamate release from as-trocytes [24, 32] or by adenosine recep-tor signaling [30]. However, intact astro-cyte Ca2+ signaling is also required for the transient postburst depression of release from these synapses [2] and spontaneous

Christian Henneberger1,2 · Gabor C. Petzold3

1  Institute of Cellular Neurosciences, University of Bonn Medical School, Bonn, Germany2  UCL Institute of Neurology, London, UK3  German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany

Diversity of synaptic astrocyte–neuron signaling

e-Neuroforum 2015 · 6:79–83DOI 10.1007/s13295-015-0011-1Published online: 11 August 2015© Springer-Verlag Berlin Heidelberg 2015

79e-Neuroforum 3 · 2015 |

Review article

excitatory synaptic transmission can be inhibited by a Ca2+-dependent boost of astrocyte potassium uptake [43]. Further-

more, postsynaptic neuronal N-methyl-D-aspartate receptors (NMDARs) are ac-tivated by Ca2+-dependent release of glu-tamate from astrocytes [23] and support-ed by Ca2+-dependent NMDAR co-ago-nist supply by astrocytes [13]. As a con-sequence of this diversity, the net ef-fect of an astrocyte Ca2+ signal on excit-atory synaptic transmission is far from clear (. Fig. 2, upper half). Another tar-get of astrocyte–neuron communication is GABAergic inhibition (for review [18]). For instance, induction of astrocyte Ca2+ transients increases GABAergic trans-mission pre and postsynaptically ([15], . Fig. 2, lower half). These examples of astrocyte modulation of synaptic trans-mission highlight the diverse and possi-bly opposing effects an astrocyte Ca2+ el-evation could trigger locally at the synaps-es received by a CA1 pyramidal cells. De-velopmental changes of glutamate recep-tor expression by astrocytes [40] and pro-found alterations of astrocyte Ca2+ sig-naling in an animal model of Alzheimer’s disease [9] are just two important exam-ples of additional layers of complexity of astrocyte–neuron communication that al-so emphasize the need to carefully consid-er experimental conditions.

Synaptic specificity of astrocyte–neuron communication

This puzzling diversity of astrocyte–neu-ron communication at synapses received by a single neuron subtype, here the CA1 pyramidal cell, poses questions about its physiological and computational func-tion. Are individual pathways spatially segregated? Are they simultaneously re-cruited at single synapses and across syn-apse populations? For instance, a glob-al astrocyte Ca2+ transient propagating through an entire astrocyte and possibly invading neighbouring gap-junction cou-pled astrocytes is likely to engage astro-cyte–neuron signaling at a vast number of different synapses and thus may rep-resent a homeostatic mechanism. In con-trast, a local and spatially confined Ca2+ transient could serve to fine-tune infor-mation transfer at single synaptic connec-tions or gate their plasticity. Indeed, as-trocyte processes can generate Ca2+ tran-sients locally and independently as dem-onstrated more than a decade ago in si-tu [26]. More recent advances in Ca2+ im-aging techniques and development of ge-netically encoded Ca2+ sensors allowed detection of a striking degree of com-partmentalization of astrocyte Ca2+ tran-sients in situ and in vivo [10, 14, 39]. The vast majority of these Ca2+ transient were highly localized events often not extend-ing more than a micrometre and only a minority of events invaded larger parts of the astrocyte or its soma. Important-ly, local Ca2+ transients could be induced by synaptic stimulation and depended on synaptic activity [10, 30]. This implies that astrocytes can translate neuronal ac-tivity into the small scale Ca2+ transients that could locally modulate the function of single or few synapses through a spe-cific mechanism. Demonstrating that this is indeed the case will be a significant ex-perimental challenge for several reasons. It requires methods to manipulate and monitor synapses and astrocytes simulta-neously at the micrometre scale. In addi-tion, it is for instance unclear if all astro-cytes and all processes of a single astro-cyte are equipped with the same molecu-lar machinery for detection of synaptic ac-tivity and its modulation. Thus, the ques-tion arises which astrocyte and which part

Fig. 2 8 Astrocyte Ca2+ signaling—one signal with too many targets? Cytosolic Ca2+ rises in as-trocytes are a potent trigger of astrocyte to neu-ron communication. A vast body of experimen-tal evidence reports a highly diverse set of neu-ronal targets. For instance, glutamatergic excit-atory synaptic transmission (orange, upper half) received by CA1 pyramidal cells of the hippo-campus is modulated presynaptically at the lev-el of axonal action potential propagation and neurotransmitter release and by activation of postsynaptic receptors. At the same time, GAB-Aergic transmission (green, lower half) is also al-tered in response to similar Ca2+ transients. The net effect on the synaptic or cellular level will depend on the specific set of mechanisms avail-able at an individual synapse, the spatial extent of the astrocyte Ca2+ signal and the range of ac-tion of a signaling molecules released from as-trocytes

Fig. 1 8 Synapses covered by a single astrocyte. Three-dimensional (3D) reconstruction after seri-al section electron microscopy of a small astrocyte fragment reveals the elaborate 3D fine structure of astrocytes with their leaf-like terminal processes. (a turquoise, rat hippocampus). Postsynaptic spines (b, white and green) with their postsynaptic densities (red) are embedded in the meshwork created by fine astrocyte processes. Terminal astrocyte branches approaching and covering synaptic contacts are often as thin as 100–200 nm. (scale cube, 1 µm3, figure adopted from [20]). An individual hippocampal astrocyte covers approximately 100,000 synapses [6].

80 | e-Neuroforum 3 · 2015

Review article

of the astrocyte to probe for local synaptic astrocyte–neuron signaling. While differ-ences of astrocyte–neuron signaling be-tween brain regions have been established [19], it is largely unexplored if individu-al astrocytes in a subregion like the CA1 stratum represent a homogeneous pop-ulation or could be as diverse as, for in-stance, interneurons [17]. The gradient of astrocyte gap junction coupling anisotro-py and morphology within the CA1 stra-tum radiatum could indicate that such local heterogeneity exists [1]. Also, astro-cytes can potentially control the supply of the NMDAR co-agonist D-serine individ-ually rather than acting as a gap junction-coupled functional syncytium [13]. It re-mains unknown however what the phys-iological granularity of NMDAR co-ago-nist supply is and if variability of astro-cyte morphology is functionally relevant for astrocyte–neuron communication.

Significant heterogeneity could al-so exist on the level of a single astrocyte. It is conceivable that certain compart-ments of an astrocyte would only be ca-pable of modulation of inhibitory or ex-citatory synapse or be exclusively sen-sitive to their activity. Identification of such subcellular specializations in astro-cytes is complicated by the lack of a clear-ly definable cellular polarity. In neurons, localization of a protein already gives a strong clue as to its functions on the cel-lular level. Presence or absence of a partic-ular signaling cascade in synaptic spines, dendritic shafts, soma, axon, or synaptic boutons guides interpretation of exper-imental results regarding the computa-tional processes it might be relevant for. Such structural features with firmly es-tablished functions are largely undefined for astrocyte with the exception of astro-cyte endfeet. These are astrocyte process-es that contact and encase blood vessels and are thought to play a role in neurovas-cular coupling [34]. Interestingly, astro-cyte endfeet are compartments where dif-fusion is slowed considerably compared with the rest of the cell thus limiting the spread of diffusible signals [27]. They are also regions of the astrocyte enriched with Ca2+-permeable TRPV4 channels [4] that trigger store-amplified Ca2+ transients in endfeet and may contribute to neurovas-cular coupling [11]. Whether other simi-

larly specialized compartments exist with-in single astrocytes that control astrocyte–neuron signaling on a synaptic scale re-mains to be established. Several recent studies indicate that at least the signaling cascades that generate Ca2+ transients are unevenly distributed throughout the as-trocytes. Genetic deletion of the IP3 re-ceptor expressed by astrocytes revealed that a significant fraction of spontaneous Ca2+ transients do not rely on IP3-depen-dent signaling, especially those generated in the astrocyte periphery [14, 39]. The lo-calization of astrocyte Ca2+ transients trig-gered by synaptic stimulation also deter-mines their dependence on metabotropic glutamate receptors [41]. Thus, the spatial distribution of different astrocyte Ca2+ signaling mechanisms could ensure spe-cific neuron–astrocyte communication on the synaptic scale.

Astrocyte coverage of synapses

Most of the well-established mecha-nisms of astrocyte–neuron signaling re-ly on diffusion of signaling molecules be-tween neurons and astrocytes. As a con-sequence, the efficiency of interactions is determined by the distance between a site of release and the site of action and there-fore the spatial configuration of synapses and the neighbouring astrocyte branches (. Fig. 1). Physiological changes of astro-cyte coverage of neurons during lactation affect glutamate clearance and supply of the NMDAR co-agonists D-serine in the supraoptic nucleus [28, 29]. In the hippo-campus, the coverage of excitatory syn-apses by astrocytes also varies consider-ably between individual synapses [20, 42]. A functional implication could be that as-trocyte–neuron communication shows a similar degree of variability or selectivity, thereby increasing its synaptic specifici-ty. In that regard, it is of particular inter-est that the morphology of astrocytes and their fine terminal branches is not static but dynamic [5, 12, 33, 44]. Such dynam-ic changes of astrocyte morphology and therefore reconfiguration of the spatial relationship between astrocyte branches and synaptic contacts could be a critical determinant of astrocyte–neuron com-munication. For example, an astrocyte process retraction from a synaptic con-

nection could affect synaptic transmission in a number of ways. The astrocyte’s abil-ity to sense activity of that particular syn-aptic connection could be impaired while transmitters released from the astrocyte may reach a much lower concentrations at this synapse. On the other hand, astro-cytes mediate most of the glutamate up-take [8]. An astrocyte process retraction could therefore reduce clearance of neu-rotransmitter released from neurons and as a consequence effectively boost syn-aptic transmission through extrasynap-tic high affinity glutamate receptors. It is noteworthy that induction of long-term potentiation (LTP) appears to be a par-ticularly robust trigger of astrocyte mor-phology changes [5, 12, 33, 44]. Coordi-nated morphology changes of presynap-tic boutons and postsynaptic spines in parallel to LTP [22] accompanied by as-trocyte restructuring could therefore per-sistently enable or disable astrocyte–neu-ron communication at potentiated syn-apses. Indeed the diffusion weighted dis-tance from a postsynaptic density to the surrounding astrocyte processes is small-er at thin spines compared to mushroom spines, which possibly have undergone previous potentiation in the dentate gy-rus [20]. While this suggests that synap-tic plasticity can profoundly alter astro-cyte–neuron communication on the lev-el of single synapses by structural change

Abstract

e-Neuroforum 2015 · 6:79–83DOI 10.1007/s13295-015-0011-1© Springer-Verlag Berlin Heidelberg 2015

C. Henneberger · G.C. Petzold

Diversity of synaptic astrocyte–neuron signaling

AbstractFast signal exchange between neurons and astrocytes at the synaptic level has attracted considerable attention. Astrocytes often re-spond with Ca2+ transients to widely differ-ent neuronal synaptic activity. At the same time, astrocyte Ca2+ elevations trigger pro-found and diverse changes of both excitato-ry and inhibitory synaptic transmission. Here, we briefly review examples of the heteroge-neity of Ca2+-dependent astrocyte–neuron communication in the rodent hippocampus and discuss mechanisms that could maintain specificity of synaptic astrocyte–neuron sig-naling in the face of its diversity.

81e-Neuroforum 3 · 2015 |

such modulation could also occur as a re-sult of, for instance, TNFα signaling [37].

Conclusions

Astrocytes sense ongoing neuronal ac-tivity by various mechanisms that ap-pear to trigger qualitatively similar Ca2+ transients. At the same time, Ca2+ signal-ing in astrocytes is a powerful modulator of synapse function with diverse and po-tentially opposing net effects on neuro-nal activity even within a subregion like the CA1 stratum radiatum. To what de-gree and how synapse specificity of as-trocyte–neuron communication is main-tained in the face of such diversity is on-ly beginning to emerge. Establishing if regional astrocyte heterogeneity or sub-cellular compartmentalization of signal-ing mechanisms within single astrocytes does constrain neuron–astrocyte com-munication will be a technical challenge and conceptually demanding because principles established for neurons may not fully apply. The specific geometry of the tripartite synapse and its dynamic changes are additional important deter-minants of signal exchange that will re-quire careful consideration. Characteriz-ing the spatial scale of a documented sig-naling pathway, if it tunes single synaptic connections or homeostatically modu-lates a vast number of synapses, and the context in which it is activated appears to be required before the computational role of astrocyte–neuron communication can be fully appreciated.

Corresponding address

C. HennebergerInstitute of Cellular NeurosciencesUniversity of Bonn Medical School Sigmund-Freud-Str. 25, 53105 Bonnchristian.henneberger@uni-bonn.de

Christian Henneberger studied medicine at the Humboldt and Free University Berlin (Germany) where he obtained his degree and defended his thesis in neurophysiology both in 2003. He continued his work on the properties of synaptic transmission in the developing visual system as a postdoctoral fellow at the Institute of Neurophysiology at the Charité (Berlin, AiP research fellowship). After moving to the UCL Institute of Neurology (London, UK), he focused on hippocampal synaptic transmission and plasticity and its dependence on components of the

extracellular matrix and astrocyte Ca2+ signaling. Obtaining a UCL Excellence Fellowship allowed him to continue this line of work as a principal investigator at UCL before establishing his own lab in Bonn in 2011. Funded by the NRW-Rückkehrerprogramm, DFG and HFSP his lab (www.henneberger-lab.com, Institute of Cellular Neuroscience) investigates how dynamic changes of astrocyte morphology affect astrocyte–neuron interactions at the cellular and synaptic level in healthy brain tissue and in disease.

Gabor C. Petzold studied medicine in Düsseldorf, Budapest, New York and London. He worked as a postdoc and resident in clinical neurology at the Departments of Neurology and Experimental Neurology, Charité Berlin. He continued to work as a postdoc, funded by an EU Marie Curie fellowship and the DFG, at the Department of Molecular and Cellular Biology, Harvard University. He returned to Berlin to work as a principal investigator on the role of astrocytes in neurovascular coupling and neurodegenerative diseases, and as a senior clinical fellow in clinical neurology. In 2011, he joined the German Center for Neurodegenerative Diseases (DZNE) in Bonn as a group leader, and the University Hospital Bonn as a consultant. In 2013, he was jointly appointed full tenured professor for vascular neurology by the DZNE and Bonn University. His lab investigates the contribution of astrocytes, neuron–astrocyte communication and blood flow changes to neurodegenerative diseases and stroke.

Acknowledgments. Work was supported by DFG (SFB1089 B03 to Christian Henneberger, SPP1757 HE6949/1 to Christian Henneberger, PE1193/2-1 to Gabor C. Petzold), NRW-Rückkehrerprogramm (Chris-tian Henneberger), Human Frontiers Science Program (Christian Henneberger), Else-Kröner Fresenius Foundation (Gabor C. Petzold), Network Of Centres Of Excellence In Neurodegeneration—CoEN (Gabor C. Petzold) and the DZNE (Gabor C. Petzold).

References

1. Anders S, Minge D, Griemsmann S, Herde MK, Steinhäuser C, Henneberger C (2014) Spatial prop-erties of astrocyte gap junction coupling in the rat hippocampus. Philos Trans R Soc Lond B Biol Sci 369:20130600

2. Andersson M, Hanse E (2010) Astrocytes im-pose postburst depression of release probabili-ty at hippocampal glutamate synapses. J Neurosci 30:5776–5780

3. Araque A, Parpura V, Sanzgiri RP, Haydon PG (1999) Tripartite synapses: glia, the unacknowledged partner. Trends Neurosci 22:208–215

4. Benfenati V, Amiry-Moghaddam M, Caprini M, My-lonakou MN, Rapisarda C, Ottersen OP, Ferroni S (2007) Expression and functional characterization of transient receptor potential vanilloid-related channel 4 (TRPV4) in rat cortical astrocytes. Neuro-science 148:876–892

5. Bernardinelli Y, Randall J, Janett E, Nikonenko I, König S, Jones EV, Flores CE, Murai KK, Bochet CG, Holtmaat A, Muller D (2014) Activity-dependent structural plasticity of perisynaptic astrocytic do-mains promotes excitatory synapse stability. Curr Biol 24:1679–1688

6. Bushong EA, Martone ME, Jones YZ, Ellisman MH (2002) Protoplasmic astrocytes in CA1 stratum ra-diatum occupy separate anatomical domains. J Neurosci 22:183–192

7. Cornell-Bell AH, Finkbeiner SM, Cooper MS, Smith SJ (1990) Glutamate induces calcium waves in cul-tured astrocytes: long-range glial signaling. Sci-ence 247:470–473

8. Danbolt NC (2001) Glutamate uptake. Prog Neuro-biol 65:1–105

9. Delekate A, Füchtemeier M, Schumacher T, Ulbrich C, Foddis M, Petzold GC (2014) Metabotropic P2Y1 receptor signalling mediates astrocytic hyperactiv-ity in vivo in an Alzheimer’s disease mouse model. Nat Commun 5:5422

10. Di Castro MA, Chuquet J, Liaudet N, Bhaukaural-ly K, Santello M, Bouvier D, Tiret P, Volterra A (2011) Local Ca2+ detection and modulation of synaptic release by astrocytes. Nat Neurosci 14:1276–1284

11. Dunn KM, Hill-Eubanks DC, Liedtke WB, Nelson MT (2013) TRPV4 channels stimulate Ca2+-induced Ca2+ release in astrocytic endfeet and amplify neu-rovascular coupling responses. Proc Natl Acad Sci U S A 110:6157–6162

12. Henneberger C, Medvedev NI, Stuart MG, Rusakov DA (2008). LTP induction changes the morphology of astrocytes in the CA1 region of the hippocam-pus in vitro. In Society for neuroscience meeting. Washington, p 242.7/I2

13. Henneberger C, Papouin T, Oliet SHR, Rusakov DA (2010) Long-term potentiation depends on release of D-serine from astrocytes. Nature 463:232–236

14. Kanemaru K, Sekiya H, Xu M, Satoh K, Kitajima N, Yoshida K, Okubo Y, Sasaki T, Moritoh S, Hasuwa H, Mimura M, Horikawa K, Matsui K, Nagai T, Iino M, Tanaka KF (2014) In vivo visualization of subtle, transient, and local activity of astrocytes using an ultrasensitive Ca2+ indicator. Cell Rep 8:311–318

15. Kang J, Jiang L, Goldman SA, Nedergaard M (1998) Astrocyte-mediated potentiation of inhibitory syn-aptic transmission. Nat Neurosci 1:683–692

16. Kirischuk S, Parpura V, Verkhratsky A (2012) Sodi-um dynamics: another key to astroglial excitabili-ty? Trends Neurosci 35:497–506

17. Klausberger T, Somogyi P (2008) Neuronal diversi-ty and temporal dynamics: the unity of hippocam-pal circuit operations. Science 321:53–57

18. Losi G, Mariotti L, Carmignoto G (2014) GABAer-gic interneuron to astrocyte signalling: a neglect-ed form of cell communication in the brain. Philos Trans R Soc Lond B Biol Sci 369:20130609

19. Matyash V, Kettenmann H (2010) Heterogeneity in astrocyte morphology and physiology. Brain Res Rev 63:2–10

20. Medvedev N, Popov V, Henneberger C, Kraev I, Ru-sakov DA, Stewart MG (2014) Glia selectively ap-proach synapses on thin dendritic spines. Philos Trans R Soc Lond B Biol Sci 369:20140047

21. Meier SD, Kafitz KW, Rose CR (2008) Developmen-tal profile and mechanisms of GABA-induced cal-cium signaling in hippocampal astrocytes. Glia 56:1127–1137

22. Meyer D, Bonhoeffer T, Scheuss V (2014) Balance and stability of synaptic structures during synaptic plasticity. Neuron 82:430–443

23. Navarrete M, Araque A (2008) Endocannabinoids mediate neuron–astrocyte communication. Neu-ron 57:883–893

24. Navarrete M, Araque A (2010) Endocannabinoids potentiate synaptic transmission through stimula-tion of astrocytes. Neuron 68:113–126

25. Navarrete M, Perea G, de Sevilla DF, Gómez-Gonza-lo M, Núñez A, Martín ED, Araque A (2012) Astro-cytes mediate in vivo cholinergic-induced synaptic plasticity. PLoS Biol 10:e1001259

82 | e-Neuroforum 3 · 2015

Review article

26. Nett WJ, Oloff SH, McCarthy KD (2002) Hippocam-pal astrocytes in situ exhibit calcium oscillations that occur independent of neuronal activity. J Neu-rophysiol 87:528–537

27. Nuriya M, Yasui M (2013) Endfeet serve as diffu-sion-limited subcellular compartments in astro-cytes. J Neurosci 33:3692–3698

28. Oliet SH, Piet R, Poulain DA (2001) Control of gluta-mate clearance and synaptic efficacy by glial cov-erage of neurons. Science 292:923–926

29. Panatier A, Theodosis DT, Mothet J-P, Touquet B, Pollegioni L, Poulain DA, Oliet SHR (2006) Glia-de-rived D-serine controls NMDA receptor activity and synaptic memory. Cell 125:775–784

30. Panatier A, Vallée J, Haber M, Murai KK, Lacaille J-C, Robitaille R (2011) Astrocytes are endogenous reg-ulators of basal transmission at central synapses. Cell 146:785–798

31. Parpura V, Basarsky TA, Liu F, Jeftinija K, Jeftinija S, Haydon PG (1994) Glutamate-mediated astro-cyte–neuron signalling. Nature 369:744–747

32. Perea G, Araque A (2007) Astrocytes potentiate transmitter release at single hippocampal synaps-es. Science 317:1083–1086

33. Perez-Alvarez A, Navarrete M, Covelo A, Martin ED, Araque A (2014) Structural and functional plastici-ty of astrocyte processes and dendritic spine inter-actions. J Neurosci 34:12738–12744

34. Petzold GC, Murthy VN (2011) Role of astrocytes in neurovascular coupling. Neuron 71:782–797

35. Porter JT, McCarthy KD (1996) Hippocampal astro-cytes in situ respond to glutamate released from synaptic terminals. J Neurosci 16:5073–5081

36. Rusakov DA, Bard L, Stewart MG, Henneberger C (2014) Diversity of astroglial functions alludes to subcellular specialisation. Trends Neurosci 37:228–242

37. Santello M, Bezzi P, Volterra A (2011) TNFa controls glutamatergic gliotransmission in the hippocam-pal dentate gyrus. Neuron 69:988–1001

38. Sasaki T, Beppu K, Tanaka KF, Fukazawa Y, Shige-moto R, Matsui K (2012) Application of an optoge-netic byway for perturbing neuronal activity via glial photostimulation. Proc Natl Acad Sci U S A 109:20720–20725

39. Srinivasan R, Huang BS, Venugopal S, Johnston AD, Chai H, Zeng H, Golshani P, Khakh BS (2015) Ca2+ signaling in astrocytes from Ip3r2(−/−) mice in brain slices and during startle responses in vivo. Nat Neurosci 18:708–717

40. Sun W, McConnell E, Pare J-F, Xu Q, Chen M, Peng W, Lovatt D, Han X, Smith Y, Nedergaard M (2013) Glutamate-dependent neuroglial calcium signal-ing differs between young and adult brain. Sci-ence 339:197–200

41. Tang W, Szokol K, Jensen V, Enger R, Trivedi CA, Hvalby Ø, Helm PJ, Looger LL, Sprengel R, Nagel-hus EA (2015) Stimulation-evoked Ca2+ signals in astrocytic processes at hippocampal CA3–CA1 synapses of adult mice are modulated by gluta-mate and ATP. J Neurosci 35:3016–3021

42. Ventura R, Harris KM (1999) Three-dimensional re-lationships between hippocampal synapses and astrocytes. J Neurosci 19:6897–6906

43. Wang F, Smith NA, Xu Q, Fujita T, Baba A, Matsu-da T, Takano T, Bekar L, Nedergaard M (2012). As-trocytes modulate neural network activity by Ca2+-dependent uptake of extracellular K+. Sci Signal 5:ra26

44. Wenzel J, Lammert G, Meyer U, Krug M (1991) The influence of long-term potentiation on the spatial relationship between astrocyte processes and po-tentiated synapses in the dentate gyrus neuropil of rat brain. Brain Res 560:122–131

83e-Neuroforum 3 · 2015 |

The Japanese government represented by the Japan Society for the Promotion of Science supports research in special pro-grams. The Grant-in-Aid for Scientific Research on Innovative Areas program supports research projects upon propos-al by a group of scientists. It is expected to strengthen the scientific standard of Japan on emerging fields and to develop collaborations within the group. It is al-so expected to educate and train young researchers who will further develop this scientific field in the future. The support will be given for 5 years with an expect- ed budget of up to 300 million Yen (about 2.2 million EUR) per year. The approval rate is about 10 %, and each year about 20 projects will be funded.

In 2013 the Grant-in-Aid program se-lected the proposal Glial Assembly: A new regulatory machinery of brain function (http://square.umin.ac.jp/glialassembl/en/) for funding, thereby supporting re-search of glia biology. The Glial Assembly research consortium will receive 1.2 bil-lion Yen (8.7 million EUR) for 5 years.

The general aim of the research project is a better understanding of neuron glia interactions in brain function (. Fig. 1). Neurons communicate with one anoth-er by forming neuronal circuits, which play a major role in expressing the brain function. Similarly, glial cells also com-municate with one another and form gli-al circuits. The communication pathways among neurons and those among glial cells are intrinsically different: glial com-munication being much slower and more gradual than neuronal communication. Long-range communication by glial cells often covers a macroscopic brain area, in-teracts with and affects the activity of neu-

ronal networks and circuits, and thereby controls the brain function.

The purpose of the project’s work pro-gram is to clarify the mechanisms that un-derlie the formation of the gigantic glial circuits, the glial assembly, and to under-stand how they control the brain function. Furthermore, it is aimed to clarify how ab-normalities of the glial assembly are relat-ed to the pathophysiology of neuropsychi-atric disorders.

Annual activities of the consortium are technical workshops, progress reports, general symposia and young investigator workshops. Financial support is allocated to foster collaborations within the consor-

tium and the attendance of young scien-tists at international glia meetings.

Thereby, the Grant-in-Aid appears to be very similar to the German DFG Pri-ority Programs.

List of core members:IKENAKA, Kazuhiro: National Insti-

tute for Physiological ScienceIINO, Masamitsu: The University of

TokyoKOIZUMI, Shuichi: University of Ya-

manashiOHKI, Kenichi: Kyushu UniversityOKABE, Shigeo: University of TokyoFUKUYAMA, Hidenao: Kyoto Uni-

versity

Kazuhiro IkenakaCentre for Multidisciplinary Brain Research and Department of Molecular Physiology,

National Institute for Physiological Sciences (NIPS), Okazaki, Japan

Funding of a glial research program in Japan: the glial assembly project

e-Neuroforum 2015 · 6:85–86DOI 10.1007/s13295-015-0013-zPublished online: 31 July 2015© Springer-Verlag Berlin Heidelberg 2015

neuronal circuits glial assembly

neurons glial cells

fast, �ring/resting, polarity slow, gradual, bidirectional

Fig. 1 8 Neuronal circuits and the glial assembly cooperate in distinct pattern for proper brain func-tion

85e-Neuroforum 3 · 2015 |

Special Issue Glia

KOHSAKA, Shinichi: National Insti-tute of Neuroscience

TAKEBAYASHI, Hirohide: Niigata University

OZAKI, Norio: Nagoya UniversityINOUE, Kazuhide: Kyushu UniversityKANBA, Shigenobu: Kyushu Univer-

sityKIRA, Junichi: Kyushu University

Corresponding address

Prof. K. IkenakaCentre for Multidisciplinary Brain Research and Department of Molecular PhysiologyNational Institute for Physiological Sciences (NIPS) 38 Nishigonaka Myodaiji 444–8585 Okazaki, Aichiikenaka@nips.ac.jp

Prof. Kazuhiro Ikenaka 1975 Graduated from Faculty of Science, Osaka University. 1980 Graduated from the doctoral course at Osaka University, PhD. 1980 Instructor at Institute for Protein Research, Osaka University. 1991 Associate Professor at Institute for Protein Research, Osaka University. 1992 Professor, NIPS. Major subject of research: Molecular Neurobiology.

86 | e-Neuroforum 3 · 2015

Special Issue Glia