<<

Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

Astrogliosis

Michael V. Sofroniew

Department of Neurobiology, David Geffen School of Medicine, University of California, Los Angeles, California 90095 Correspondence: [email protected]

In addition to their many functions in the healthy (CNS), respond to CNS damage and disease through a process called . For many decades, astrogliosis was sparsely studied and enigmatic. This article examines recent evidence sup- porting a definition of astrogliosis as a spectrum of heterogeneous potential changes in astrocytes that occur in a context-specific manner as determined by diverse signaling events that vary with the nature and severity of different CNS insults. Astrogliosis is associated with essential beneficial functions, but under specific circumstances can lead to harmful effects. Potential dysfunctions of astrocytes and astrogliosis are being identified that can contribute to, or be primary causes of, CNS disorders, leading to the notion of astrocytopa- thies. A conceptual framework is presented that allows consideration of normally occurring and dysfunctional astrogliosis and their different roles in CNS disorders.

strocytes exert many essential functions in mation has now accrued regarding the mecha- Athe healthy central nervous system (CNS) as nisms, functions, and impact of astrogliosis. It is reviewed and discussed in other articles in this now clear that astrogliosis plays fundamental volume. In addition, astrocytes respond to all roles in determining tissue repair and outcome forms of CNS damage and disease with a variety after injury or disease, and has the potential to of potential changes in gene expression, cellu- influence neural function. lar structure, and function. Such responses are commonly referred to as astrogliosis. Although astrogliosis has a long history of descriptive WHAT IS ASTROGLIOSIS? analysis beginning in the late 19th century, there The term astrogliosis dates back to late 19th and was only sparse interest in, or investigation of, its early 20th century neuroanatomists who re- mechanisms or functions for much of the 20th cognized that astroglia underwent pronounced century. This situation changed gradually over structural changes in response to CNS damage the last two decades when, in parallel with the and disease. At different times, astrogliosis has explosion of information regarding essential been assigned various definitions but it has al- roles in the healthy CNS, interest has ways referred to astrocyte responses to CNS in- grown in understanding astrogliosis (Pekny sults. Based on a large cross section of studies in and Nilsson 2005; Sofroniew 2009; Sofroniew experimental animals and human pathological and Vinters 2010; Kang and Hebert 2011). As specimens, we have recently proposed a more summarized in this article, considerable infor- detailed working definition of astrogliosis that

Editors: Ben A. Barres, Marc R. Freeman, and Beth Stevens Additional Perspectives on available at www.cshperspectives.org Copyright # 2014 Cold Spring Harbor Laboratory Press; all rights reserved Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420

1 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

M.V. Sofroniew

encompasses four key features: (1) astrogliosis does not occur in isolation, but is part is a spectrum of potential molecular, cellular, of a coordinated multicellular response to CNS and functional changes in astrocytes that occur insults that includes multiple types of glia as in response to all forms and severities of CNS well as and different types of non-neu- injury and disease; (2) changes undergone by ral cells that are intrinsic to the CNS or that reactive astrocytes vary with severity of the in- enter from the bloodstream (Burda and Sofro- sult along a graded continuum; (3) changes niew 2014). Different types of glia that respond associated with astrogliosis are regulated in a to CNS insults include , astrocytes, context-specific manner by many different and NG2-positive progenitors inter- and intracellular signaling molecules; (NG2-OPC). Non-neural cells intrinsic to the and (4) changes undergone during astrogliosis CNS that respond to insults include endothelia, have the potential to alter astrocyte activities perivascular fibroblasts, , and menin- both through gain and loss of functions (So- geal cells. Cells that enter from the bloodstream froniew 2009; Sofroniew and Vinters 2010). after CNS insults include leukocytes, platelets, Specific aspects of these features are discussed and other bone marrow–derived cells. Together, below. these cells have the capacity to produce a vast Various other terms are sometimes used to array of intercellular signaling molecules, and refer to astrocyte responses to CNS damage or these molecules have the potential to influence disease, and use of certain terms can vary among the activities and functions of different cell investigators. Wewill use “astrogliosis” and “re- types, including astrocytes. Thus, the response active astrocytes” as general all-inclusive de- to CNS insults is a complex mixture of events scriptors of all forms of astrocyte responses involving the interactions of multiple cell types associated with any form of CNS damage or dis- that change over time (Burda and Sofroniew ease. As discussed in more detail below, these 2014). Astrocytes take part in these interactions terms encompass astrocyte responses of con- both by receiving instructive signals from other siderable diversity and heterogeneity. We will cells and by sending instructive signals that in- not use “activation” or “activated astrocytes” as fluence other cells (Fig. 1). termsthatreferexclusively toastrocyteresponses to injury or disease. Astrocytes in healthy tissue ASTROGLIOSIS AS A CONTINUUM continually show physiological activation in the OF POTENTIAL RESPONSES form of transient, ligand-evoked elevations in 2þ intracellular calcium ([Ca ]i) that represent a Astrogliosis has often been referred to as if type of astrocyte excitability, involved in medi- it were a single uniform entity that is essential- ating many critical dynamic astrocyte functions, ly the same in all situations. This is emphatically including interactions with and regu- not the case. A multitude of studies over the lation of blood flow as discussed in other articles past 20 yr have provided compelling evidence in this volume. Thus, astrocyte activation can that astrogliosis is not a simple stereotypic re- range from physiological contexts in healthy sponse, but instead is a finely tuned spectrum CNS to pathophysiologic contexts that involve of potential changes that range from reversible responding to CNS injury or disease. In keeping alterations in gene expression and cellular hy- with these definitions, there seems to be no place pertrophy to pronounced cell proliferation with for the notion of an “activated astrocyte” as a compact scar formation and permanent tissue single uniform entity. rearrangement. Available evidence points to- ward a wide range of specific molecular sig- naling mechanisms that regulate specific aspects MULTICELLULAR RESPONSES of astrogliosis, rather than a single stereotypic TO CNS INSULTS program that is regulated in an on–off manner Although the focus of this article is on astro- (Pekny and Nilsson 2005; Sofroniew 2009; So- gliosis, it is important to emphasize that astro- froniew and Vinters 2010; Kang and Hebert

2 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

Astrogliosis

ABAstrocytes receive instructions Astrocytes send instructions Microglia Synapse Microglia FGF-2 IL-10 + NO ATP PT3 K ATP FGF-2 EGF TNF-α TNF-α KYN NO LCN2 TNFα Glut INF-γ Glut IL-11IL-15 IL-1β β IL1- CXCL10 THBS1 NE Reactive Astrocyte Reactive Astrocyte CCL2 GPC4 CCL7 wbc ROS IL-6 wbc CSPG CXCL1 IL-10 LIF BDNF IL-6 CCL8 GDNF TGF-β GDNF CCL5 β NGF CXCL9 A Estrogen CXCL16 NH + bv ??? bv 4 GFs NO BBB CNTF Albumin END LIF MMP3 PGE repair SHH Thrombin LPS IGF-1 VEGF-A PAMPs BBB leak Blood BBB leak Neuron flow

Figure 1. Reactive astrocytes receive and send diverse molecular instructions. Schematic drawings depict exam- ples of diverse molecular signals that instruct reactive astrocytes (A), or that reactive astrocytes release to instruct other cells (B). For abbreviations, see Table 1.

2011). Although there are common features very intense changes (Fig. 2). Thus, astrogliosis across different forms and intensities of astro- is complex and multivariate, and the changes gliosis, such as up-regulation of glial fibrillary in astrocytes that are associated with astro- acidic protein (GFAP) and cellular hypertro- gliosis are induced in a context-dependent phy, even these do not occur in all or none fash- manner as regulated by many different potential ions, but are finely regulated along a gradient signaling events as summarized in the next of intensity that can vary from very small to sections.

Astrogliosis gradient Healthy Mild Moderate Severe diffuse Scar Tissue lesion Domain Overlapping preservation processes

Cytoskeletal Cytoskeleton hypertrophy Proliferated Inflammatory scar astrocytes cells

Figure 2. Schema of astrogliosis gradient from mild to moderate to scar. In healthy CNS tissue, many astrocytes do not express detectable levels of the cytoskeletal protein, GFAP. In mild to moderate astrogliosis most astrocytes up-regulate GFAP and hypertrophy their cytoskeleton but preserve individual domains. In severe diffuse astrogliosis, there is also proliferation (depicted by red nuclei). Compact astroglial scars are comprised of newly proliferated astrocytes with densely overlapping processes that form borders to damaged tissue and inflammation.

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 3 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

M.V. Sofroniew

SIMPLE CONVENIENT CATEGORIES with overlapping of neighboring astrocyte pro- OF ASTROGLIOSIS cesses (Fig. 2) (Sofroniew 2009; Sofroniew and Vinters 2010). These changes can extend dif- Changes in astrocyte appearance, molecular ex- fusely over substantive areas, and can occur in pression, and function that are associated with various situations, such as chronic neurodegen- astrogliosis occur along a seamless graded con- erative insults, diffuse trauma, diffuse , tinuum of intensity. Nevertheless, for practical or certain types of . Because there can purposes of description, comparison, and dis- be considerable tissue reorganization, the po- cussion, it is convenient to distinguish among tential for resolution and return to normal several simple categories of astrogliosis as delin- structure is reduced and there is a high tendency eated on the basis of changes in structure, mo- toward long-lasting tissue reorganization (So- lecular expression, and proliferation. These cat- froniew and Vinters 2010). egories include mild to moderate astrogliosis, severe diffuse astrogliosis, and severe astroglio- sis with compact scar formation (Fig. 2). Severe Astrogliosis with Compact Scar Formation Mild to Moderate Astrogliosis Compact astroglial scars derive almost entirely Mild to moderate astrogliosis consists of from newly proliferated astrocytes with elongat- changes (up or down) in astrocyte gene expres- ed shapes (Wanner et al. 2013), whose cell pro- sion together with variable degrees of hypertro- cesses overlap and intertwine extensively to phy of cell body and stem processes, without form compact borders that surround and de- substantive loss of individual astrocyte domains marcate areas of tissue damage, , and and without astrocyte proliferation (Sofroniew inflammation after trauma, , infection, 2009; Sofroniew and Vinters 2010). There is autoimmune-triggered inflammatory infiltra- increased expression of the astrocyte intermedi- tion, or neurodegenerative disease (Bush et al. ate filament protein, GFAP, which is somewhat 1999; Faulkner et al. 2004; Drogemuller et al. proportional to the degree of reactivity. The 2008; Voskuhl et al. 2009; White et al. 2010; essentially nonoverlapping astrocyte domains Wanneret al. 2013). There may be several possi- found in healthy gray matter are more or less ble sources of newly divided scar-forming astro- preserved in mild to moderate reactive astroglio- cytes including: (1) mature astrocytes that re- sis in and (Fig. 2) (Wilhelms- enter the cell cycle and proliferate (Bush et al. son et al. 2006; Wanner et al. 2013). Mild to 1999; Buffo et al. 2008; Gadea et al. 2008; Bar- moderate astrogliosis is generally associated dehle et al. 2013), (2) NG2 progenitorcells in the with mild nonpenetrating and noncontusive local parenchyma (Magnus et al. 2008), or (3) trauma, or with diffuse innate immune activa- ependymal cell progenitors (Meletis et al. 2008; tion (viral , system bacterial infec- Barnabe-Heider et al. 2010). Astrocyte scar bor- tions), or occurs in areas that are some distance ders directly interface with and surround a vari- to focal CNS lesions. After insults that are acute ety of non-neural cell types in the central core single events, such as traumatic or ischemic in- of tissue lesions, including perivascular-derived juries, mild to moderate astrogliosis has consid- fibroblasts, fibrocytes, pericytes, and other glial erable potential to resolve. cells (Fig. 2) (Bundesen et al. 2003; Herrmann et al. 2008; Sofroniew 2009; Sofroniew and Vinters 2010; Aldrich and Kielian 2011; Reilkoff Severe Diffuse Astrogliosis et al. 2011; Burda and Sofroniew 2014). It is Severe diffuse reactive astrogliosis consists of noteworthy that astroglial scar formation is as- changes (up or down) in gene expression with sociated with substantive tissue reorganization pronounced up-regulation of GFAP, cellular and structural changes that are essentially per- hypertrophy, dispersed astrocyte proliferation, manent and persist even when triggering insults and some loss of individual astrocyte domains may have resolved.

4 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

Astrogliosis

HETEROGENEITY OF ASTROCYTES is a spectrum of potential changes that occur AND REACTIVE ASTROCYTES in a context-specific manner as determined by a myriad of specific potential signaling events There is growing evidence for and interest in the that vary with the nature and severity of specific potential for heterogeneity among astrocytes in CNS insults. Current evidence indicates that healthy CNS not only across different CNS re- specific aspects of reactive astrogliosis can be gions, but also locally within the same regions regulated separately and individually by a wide (Zhang and Barres 2010; Tsai et al. 2012). Given variety of intercellular and intracellular signal- that astrogliosis is not a single stereotypic re- ing and regulatory molecules, as called for in sponse but is regulated by multiple specific sig- different situations. Combinatorial exposure to naling events, there is also growing awareness of multiple signals can markedly alter astrocyte heterogeneity among reactive astrocytes at mul- transcriptome profiles and give rise to synergis- tiple levels (Anderson et al. 2014). For example, tic effects not predicted by exposure to individ- adjacent to focal traumatic or ischemic lesions, ual stimuli (Hamby et al. 2012). there is topographic heterogeneity of astroglio- sis as regards astrocyte proliferation, morphol- ogy, and gene expressionwith respect to distance Extracellular Signals from the insult (Sofroniew and Vinters 2010; Wanner et al. 2013). In addition, analysis at the Astrogliosis can be induced, regulated, or mod- single cell in vivo shows that intermingled re- ulated by a wide variety of extracellular mole- active astrocytes can show different expression cules ranging from small molecules, such as levels of (1) chemokines or (Hamby purines, transmitters, and steroid hormones, et al. 2012), (2) signaling molecules, such as to large polypeptide growth factors, cytokines, pSTAT3 (Herrmann et al. 2008), or (3) tran- serum proteins, or associ- b scription factors that regulate Sonic Hedgehog ated molecules like - (Table 1). These signaling (SHH) (Garcia et al. 2010). It is also instructive signals can derive from many differ- noteworthy that different stimuli of astrocyte ent sources (Fig. 1A) and can be released by cell reactivity can lead to similar degrees of GFAP damage or cell death, or via specific signaling up-regulation, whereas causing substantially mechanisms. Many cell types can release molec- different changes in transcriptome profiles and ular regulators of astrogliosis, including (1) lo- cell function (Hambyet al. 2012; Zamanian et al. cal neural and non-neural cells intrinsic to CNS 2012). Thus, it is not possible to equate simple tissue, such as neurons, microglia, oligodendro- and uniform measures, such as cell hypertrophy cyte lineage cells, endothelia, pericytes, fibro- and up-regulation of GFAP expression with a meningeal cells, and other astrocytes, as well single, uniform concept of astrocyte reactivity. as (2) non-neural cells that gain entry into the Indeed, certain inflammatory signals can sub- CNS, such as bone marrow–derived leukocytes, stantively alter reactive astrocyte transcriptome fibrocytes, and microbial infectious agents (Fig. profiles without inducing further changes in 1A) (Burda and Sofroniew 2014). Last but not GFAP (Hamby et al. 2012). These findings un- least, astrogliosis can be regulated by molecules, derscore the considerable potential for pheno- such as serum proteins, cytokines, or steroid typic and functional heterogeneity among reac- hormones that are produced by cells outside tive astrocytes as regulated by specific signaling of the CNS, including microbial endotoxins, events. such as lipopolysaccharides (LPS) (Burda and Sofroniew 2014).

MULTIPLE MOLECULAR REGULATORS OF ASTROGLIOSIS Intracellular Signal Transducers and Regulators Astrogliosis is not an all-or-none response or stereotypic program regulated by a few simple Many extracellular molecular regulators of as- on–off molecular switches. Instead, astrogliosis trogliosis have receptor targets that initiate in-

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 5 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

M.V. Sofroniew

Table 1. Selected examples of molecular regulators of astrogliosis Categories Source Molecules Extracellular molecular signals DAMPs Cell damage triphosphate (ATP), (ROS), (NO), high-mobility group box 1 (HGMB1), etc. Hypoxia and Ischemia Oxygen deprivation, glucose deprivation metabolic stress Transmitters Local neurons Glutamate (Glut), noradrenalin (NE) Hormones Endocrine glands Estrogens, glucocorticoids Neurodegeneration b-amyloid (Ab) PAMPs Microbes LPS, zymosan, dsRNA (viral) þ Systemic metabolic toxicity Ammonium (NH4 ) (liver failure) Serum proteins Bloodstream Thrombin, albumin, fibrin, etc. Cytokines and Other glia, local non-neural IL-1b, (TNF)-a, interferon growth factors cells, infiltrating leukocytes (INF)-g, IL-6, ciliary neurotrophic factor (CNTF), leukemia inhibitory factor (LIF), transforming growth factor (TGF)-b, IL-10, (EGF), fibroblast growth factor (FGF)-2, Sonic Hedgehog (SHH), (END) Astrocyte intrinsic signaling pathways and regulatory mechanisms Signal transducers STAT3, STAT2, STAT2, JAK2, NF-kB, IRAK, SOC3, MAP-kinase, SOX9, Nrf2, Olig2, cAMP, Nurr1, SMAD3, SMAD4, mTOR, c-FOS, c-JUN, PKA, PKC, ERK, MyDD8, IRF1, G proteins, etc. microRNAs miR-21, miR-181, Dicer (ribonuclease) identified thus far Nomenclature for growth factors, cytokines, chemokines, receptors, and transcription factors are per The Human Gene Compendium, GeneCards (see www.genecards.org). DAMPs, danger-associated molecular patterns; PAMPs, pathogen-associated molecular patterns.

tracellular second messenger signaling cascades Selective Regulation of Specific Functions (Table 1), as reviewed elsewhere (Pekny and and Effects of Astrogliosis Nilsson 2005; Sofroniew 2009; Sofroniew and Vinters 2010; Kang and Hebert 2011). There is There is increasing interest in dissecting molec- also emerging evidence that microRNAs (miR), ular signaling mechanisms that regulate specific such as miR-21 and miR-181 (Bhalala et al.2012; functions or effects of astrogliosis so that these Hutchison et al. 2013), and miR regulatory en- might be targeted by specific therapeutic inter- zymes, such as Dicer (Tao et al. 2011), can mod- ventions (Sofroniew 2009; Hamby and Sofro- ulate astrogliosis and reactive astrocyte func- niew 2010). Considerable information is now tions, adding yet another level of potential available, which reveals that some aspects of as- regulation and specification of functions (Table trogliosis can be regulated by many pathways, 1). Thus, intrinsic properties of individual as- whereas other aspects are regulated more selec- trocytes, such as epigenetic mechanisms or ge- tively. For example, expression of structural fil- netic polymorphisms of receptors and second aments, such as GFAP or vimentin can be in- messengers and other regulators, may impact duced by signaling pathways associated with astrogliosis. cAMP, STAT3, NF-kB, Rho-kinase, JNK, calci-

6 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

Astrogliosis

um, and others (Pekny and Nilsson 2005; Sofro- 2010; Ji et al. 2013). Findings such as these sup- niew 2009; Middeldorp and Hol 2011; Gao et al. port a model in which many different specific 2013). Similarly, astrocyte proliferation can be signaling mechanisms can trigger different spe- regulated by different extracellular signals in- cific molecular, morphological, and functional cluding EGF, FGF, endothelin 1, Sonic Hedge- changes in reactive astrocytes. This model also hog, the serum proteins thrombin and albumin, predicts that the impact of astrogliosis on its and others, and by intracellular signals, such as host neural tissue and associated neural func- Olig2, JNK pathway, and others (Levison et al. tions will not be uniform or stereotypic, but 2000; Gadea et al. 2008; Sofroniew 2009; Sirko will vary in a context-specific manner as deter- et al. 2013). Other aspects of astrogliosis are mined by the specific signaling mechanisms as- regulated more selectively. For example, certain sociated with specific insults and the environ- pro- and anti-inflammatory functions of astro- ment in which they are occurring. cytes are regulated separately. Deletion or dis- ruption of NF-kB or SOC3 signaling pathways BENEFICIAL FUNCTIONS selectively in astrocytes diminishes recruitment AND MALADAPTIVE EFFECTS of inflammatory cells after traumatic injury and OF ASTROGLIOSIS autoimmune disease (Brambilla et al. 2005, 2009; Okada et al. 2006). Deletion of estrogen In response to the many signaling events just receptor-a, but not estrogen receptor-b, selec- described, reactive astrocytes have the potential tively from astrocytes diminishes the anti- to release a large variety of molecules that im- inflammatory and neuroprotective effects of pact nearby cells (Fig. 1B; Table 2). These mol- estrogenonautoimmuneinflammation(Spence ecules can exert many different functions and et al. 2011, 2013). In contrast, deletion of STAT3 effects. For much of its greater than 100-year or its associated membrane receptor GP130, history, astrogliosis has been regarded by some markedly increases the spread of inflammation after traumatic injury, autoimmune disease, or Table 2. Selected examples of effector molecular infection (Okada et al. 2006; Drogemuller et al. released by reactive astrocytes 2008; Herrmann et al. 2008; Haroon et al. 2011; Categories Molecules Wanner et al. 2013). Deletion of YLK-40/BRP- b a g 39, an astrocyte produced anti-inflammatory Cytokines IL-6, IL-1 , TNF- , INF- , CNTF, LIF, TGF-b, IL-11, protein, exacerbates autoimmune disease in IL-15 mice (Bonneh-Barkay et al. 2012). miR-181 de- Chemokines CXCL1, CXCL9, CXCL10, creases astrocyte production of inflammatory CXCL12, CXCL16, CCL2, cytokines (TNF-a, IL-1b) and increases pro- CCL5, CCL7, CCL8 duction of anti-inflammatory cytokines (IL- Growth factors and BDNF, NGF, GDNF, THSP1, 10) (Hutchison et al. 2013). Thus, different sig- other proteins GPC4, GPC6, FGF2, naling mechanisms regulate different aspects of PDGFb, BMP1, VEGF-A, pro- or anti-inflammatory functions of reactive pentraxin 3 (PT3), astrocytes. From a different perspective, expres- a-crystallin B, etc. sion and function of the MMP3, SEMA4A, TIMP1, GLT-1 can be modulated up or down during chondroitin sulfate (CSPG) astrogliosis depending on the nature of the in- Intermediate GFAP, vimentin, nestin sult and time after the insult, and this regulation filaments can impact outcome by influencing extracellular Transmitters Glutamate (Glut), kynurenic glutamate levels and the potential for neuronal acid (KYN), D-serine (Bianchi et al. 2013). The mTOR- Small molecules NO, prostaglandin E (PGE), Akt-NF-kB pathway is one molecular signaling AT P mechanism that can modulate astrocyte expres- Nomenclature as per The Human Gene Compendium, sion of Glt-1 (Pawlak et al. 2005; Faideau et al. GeneCards (see www.genecards.org).

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 7 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

M.V. Sofroniew

investigators as a reaction that is fundamentally disease, or autoimmune attack, as discussed be- harmful to functional recovery. This notion is low (Sofroniew 2013). refuted bya large volume of in vivo experimental Perhaps the most widely regarded potential studies from many laboratories, which show that detrimental effect of astrogliosis is the inhibi- in response to most if not all CNS insults, nor- tion of regeneration by astrocyte scars after mally occurring astrogliosis is initiated as an CNS damage (Silver and Miller 2004). Never- adaptive, beneficial process directed at wound theless, this potential detrimental effect must be repair and tissue protection. Nevertheless, ex- viewed in the context of the essential protective perimental studies have also established the po- roles that astrocyte scars perform in restricting tential for potentially harmful effects of astro- inflammation and protecting healthy tissue ad- gliosis under specific circumstances. Thus, it is jacent to lesions as discussed above. Simply pre- important to understand that astrogliosis, like venting or removing astrocyte scars is likely to inflammation, exerts essential beneficial func- do more harm than good. Attempts to manip- tions but can alsogiverise to maladaptive effects, ulate scar formation with the intention of and that these are regulated by specific signaling improving axon regeneration will likely need mechanisms that occur in specific contexts. to be very specifically targeted and controlled, Transgenic loss-of-function models have and, given the multiple and complex causes for been particularly useful in identifying beneficial failure of CNS axon regeneration (Brosius Lutz functions and harmful effects of astrogliosis and Barres 2014), are unlikely to be effective on and astrocyte scar formation (Sofroniew 2005; their own. Sofroniew and Vinters 2010). For example, Together, such observations provide com- transgenic ablation or prevention of astrogliosis pelling evidence that astrogliosis exerts a variety or astrocyte scar formation causes increased in- of beneficial functions that are essential for lim- flammation and tissue damage and worsens iting tissue damage and preserving neurological functional outcome in all CNS insult models function after CNS insults but that astrogliosis studies thus far, including traumatic injury, is- also has the potential to exert detrimental effects chemic injury (stroke), infection, autoimmune as determined byspecific signaling mechanisms. inflammation, and neurodegenerative disorders Recognition of these concepts sets the stage for (Nawashiro et al. 1998; Bush et al. 1999; Faulk- dissecting the physiology and pathophysiology ner et al. 2004; Myer et al. 2006; Drogemuller of astrogliosis to identify and ameliorate the et al. 2008; Herrmann et al. 2008; Li et al. 2008; consequences of astrocyte dysfunction. Voskuhl et al. 2009; Haroon et al. 2011; Macau- ley et al. 2011; Wanneret al. 2013). Nevertheless, transgenic studies also reveal the potential for DYSFUNCTIONS OF ASTROCYTES AND certain aspects of astrogliosis to exacerbate in- ASTROGLIOSIS: TOWARD A CONCEPT OF ASTROCYTE PATHOPHYSIOLOGY flammation after traumatic injury or autoim- AND ASTROCYTOPATHIES mune challenge (Brambilla et al. 2005, 2009; Spence et al. 2011, 2013). Other studies show There is increasing evidence for a concept of that controlling astrocyte pH can reduce gluta- astrocytopathies in which a disruption of nor- mate excitotoxicity during ischemia (Beppu mal astrocyte functions in healthy tissue is the et al. 2014; Sloan and Barres 2014). Large-scale primary cause of neurological dysfunction and gene expression evaluations show that inflam- disease (Verkhratsky et al. 2012). The argument matory mediators can drive astrocyte transcrip- can be made that a full concept of astrocytopa- tome profiles toward proinflammatory and po- thies will also include disorders caused by alter- tentially cytotoxic phenotypes (Hamby et al. ing the process of astrogliosis (Fig. 3). As dis- 2012; Zamanian et al. 2012) that may be bene- cussed above, a strictly negative connotation of ficial in microbial infection but may be detri- astrogliosis is being replaced by a balanced view mental if triggered during sterile (uninfected) of astrogliosis as an adaptive process generated tissue responses to trauma, stroke, degenerative and conserved by evolution, which serves es-

8 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

Astrogliosis

Healthy tissue Tissue CNS insult Normal Outcome protection Normal astrogliosis astrocyte and repair CNS insult

Abnormal astrogliosis Outcome Astrocytopathy mutations, with polymorphisms, secondary chronic insult, neural infections, dysfunction CNS insult high cytokines, and Astrocyte autoimmune with mutation attack, degeneration GFAP-AlexD, etc. SOD-fALS, mHtt-HD, Astrocytopathy etc. Outcome

Figure 3. Schema that distinguishes normal from abnormal astrogliosis and depicts different pathways that can lead to astrocytopathies.

sential repair and tissue protective functions, show macrocephaly, severe neurological dys- but that in certain situations can show harm- function, , psychomotor disturbances, ful effects as a direct consequence of context- and premature death (Brenner et al. 2001; Mess- inappropriate signaling interactions. This per- ing et al. 2012). A second well, studied example is spective allows a conceptual framework for a familial form of amyotrophic lateral sclerosis astrogliosis that includes both adaptive (benefi- (ALS) caused by a dominant gain-of-function cial) and maladaptive (harmful) forms of astro- mutation of the gene-encoding superoxide dis- gliosis, and allows for consideration of multiple mutase (SOD) that leads to the production of ways in which dysfunctions of astrogliosis could molecules that are toxic to motor neurons. Tar- be primary causes of, or major contributors geting mutant SOD selectively to astrocytes in to,CNSdisorders.Recognizingandunderstand- transgenic mice is sufficient to cause neuronal ing the potential for dysfunction of astrogliosis dysfunction and degeneration (Lobsiger and (Fig. 3) is likely to be fundamental to dissecting Cleveland 2007; Nagai et al. 2007; Yamanaka the cellular mechanisms underlying a variety of et al. 2008). Nevertheless, in the full disease, mu- CNS disorders. The next sections discuss clini- tant SOD produced in neurons also contributes cal and experimental examples of ways in which to neuronal degeneration. In addition to these primary dysfunctions of astrocytes or of astro- examples, there is recent evidence from experi- gliosis could precipitate, or contribute to, neu- mental animal models that the genetic mutation rological dysfunction and disease. associated with Huntington’s disease causes dys- function of astrocytes in ways that may contrib- ute significantly to neurological symptoms and ASTROCYTE GENE MUTATIONS neurodegeneration. In Huntington’s disease, as- There are now several examples of genetic mu- trocytes as well as neurons accumulate nuclear tations that cause astrocyte dysfunctions that inclusionsofmutanthuntingtinprotein(mHtt). can lead, or contribute significantly, to neuronal In transgenic mouse models, astrocytes with dysfunction and degeneration, which in turn mHtt down-regulate the potassium channel lead to neurological symptoms and disease. One Kir4.1, leading to increased extracellular potas- first example, , is caused by a sium and increased neuronal excitability, which dominant mutation of the GFAP gene, which is is likely to contribute to and exacerbate direct expressed in the CNS only by astroglia. Patients effects of mHtt on neurons (Tonget al. 2014). It

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 9 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

M.V. Sofroniew

is noteworthy that these effects occur at early CNS autoimmune disease that selectively dam- stages of symptom onset and in the absence of ages astrocytes has a more severe course than astrocyte changes associated with astrogliosis autoimmune disease that involves other cellular (Tong et al. 2014). It is important to emphasize targets. Neuromyelitis optica (NMO) is a CNS that in all of these conditions, the detrimental inflammatory demyelinating disease with pro- effects on neuronal function and nounced tissue destruction and severe symp- are caused by genetically mutated and abnormal toms that can include vision loss and paralysis. astrocytes rather than by the normal process of NMO has been causally associated with autoan- astrogliosis. Thus, astrocyte dysfunction that is tibodies to aquaporin-4 (AQP4) on astrocytes harmful can occur in the absence of astrogliosis, that mediate compliment-mediated astrocyte and astrogliosis can occur without causing harm lysis (Lennon et al. 2005; Roemer et al. 2007). (Fig. 3). Both clinical course and tissue lesions are more severe in patients that have autoimmune-medi- ated CNS demyelination associated with anti- ASTROCYTE MOLECULAR AQP4 autoantibodies compared with patients POLYMORPHISMS with antimyelin autoantibodies (Kitley et al. An intriguing and potentially important ques- 2014; Sato et al. 2014). These findings strongly tion is the degree to which genetic polymor- suggest that AQP4 is not simply a passive auto- phisms could alter functions of astrocytes and immune CNS , but that disruption of astrogliosis. Available evidence suggests that astrocyte functions by AQP4 antibodies exacer- this is likely to be the case. Apolipoprotein E bates the disease process and worsens the out- (APOE) polymorphisms influence the risk of come. This notion is consistent with experimen- Alzheimer’s disease, with isoform APOE4 in- tal evidence that astrocytes critically restrict the creasing risk and APOE2 reducing risk. Stud- spread of inflammation during autoimmune ies using various transgenic mouse models CNS attack (Voskuhl et al. 2009; Haroon et al. show that astrocyte-secreted APOE4, but not 2011). From an experimental perspective, it is APOE3 or APOE2, is associated with increased well documented that astrocytes can produce a blood–brain barrier leak and increased neuro- wide variety of pro- or anti-inflammatory mol- nal degeneration (Bell et al. 2012) in a manner ecules and that specific molecular regulators that may be relevant to Alzheimer’s disease pa- modulate specific astrocyte functions in ways thology (Zlokovic 2011). The notion that astro- can increase or limit CNS inflammation. Astro- cyte intrinsic molecular polymorphisms might cytes produce not only a wide range of chemo- impact astrocyte functions in disease-relevant kines that open the blood–brain barrier (Argaw ways is further supported by the example of et al. 2012) and attract inflammatory cells genetic mutations or experimental gene manip- (Hamby et al. 2012; Zamanian et al. 2012; So- ulations that modulate astrocytes and astroglio- froniew 2013) but also molecules that can exert sis in ways that alter neuronal function or cause potent suppressive effects on inflammatory cells neurodegeneration as discussed above. (Kostianovsky et al. 2008). Astrocyte scars play critical roles in limiting the spread of inflamma- tory cells from damage areas into neighboring ASTROGLIOSIS AND CNS healthy tissue after trauma or autoimmune at- Inflammation is an important component of tack (Bush et al. 1999; Voskuhl et al. 2009; Wan- the multicellular response to CNS damage and ner et al. 2013). Modulation of specific astrocyte disease that can markedly influence the balance functions can eitherattenuateorexacerbateCNS between tissue loss and preservation (Burda and inflammation (Okada et al. 2006; Herrmann Sofroniew 2014). There is now both clinical and et al. 2008; Brambilla et al. 2009), opening the experimental evidence that astrocytes are criti- door for gain or loss of astrocyte functions to cal regulators of CNS inflammation. From a impact CNS inflammation (Dong and Benve- clinical perspective, recent findings show that niste 2001; Sofroniew 2009). Together, these

10 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

Astrogliosis

findings provide compelling clinical and exper- et al. 2006; Drogemuller et al. 2008; Herrmann imental evidence that dysfunction of astroglio- et al. 2008; Li et al. 2008; Haroon et al. 2011). It is sis can exacerbate CNS inflammation and the also intriguing to consider whether chronic as- spread of tissue damage in various ways and trogliosis with long-term overexpression of contexts. GFAP may be sufficient to cause cellular dys- function and degeneration. This possibility is raised not only by the clinical example Alexan- ASTROCYTE-MEDIATED NON-CELL- der disease with mutated GFAP discussed above, AUTONOMOUS NEURONAL but also by evidence that chronic high level ex- DYSFUNCTION OR DEGENERATION pression of normal GFAP may cause pathology There are many ways that loss of astrocyte func- (Messing et al. 1998). Chronic astrogliosis can tions or disruption of astrogliosis could lead occur in neurodegenerative disorders or when to non-cell-autonomous neuronal dysfunction CNS insults are comorbid with infections as or degeneration. In healthy neural tissue, astro- discussed below, and its effects warrant further cytes play critical roles for normal neuronal study. Nonetheless, it deserves emphasis that, function including of extracellu- after acute insults, such as single traumatic or lar fluid, ions, and transmitters, regulation of ischemic events that are not comorbid with in- blood flow, energy provision, and interactions fections, astrogliosis in perilesion tissue resolves with synapses as reviewed elsewhere (Barres relatively rapidly (within weeks) and GFAP lev- 2008; Sofroniew and Vinters 2010). In damaged els return to normal or low levels. Last, an excit- neural tissue, reactive astrocytes play critical ing new area pioneered by the Barres laboratory roles in , blood–brain barrier is astrocyte induction and pruning of neuronal repair, and regulation of inflammation (Sofro- synapses during development and in adult neu- niew and Vinters 2010; Burda and Sofroniew ral plasticity (Clarke and Barres 2013). Mice de- 2014). Thus, perhaps not surprisingly, experi- ficient in astrocyte-specific phagocytic pathways mentally induced loss of specific functions of fail to develop normal connections and retain astrocytes or astrogliosis can cause neuronal excess synapses (Chung et al. 2013). The impact dysfunction and degeneration and worsen out- of normal or anomalous astrogliosis on astro- come after CNS trauma, ischemia, or autoim- cyte interactions with synapses remains to be mune attack. A pioneering example is that selec- determined but has the potential to open new tive deletion of the astrocyte glutamate uptake ways of thinking about certain CNS disorders. transporter, Glt-1 (EAAT2) will lead to seizures and excitotoxic neuronal death (Rothstein et al. 1996). Further examples include (1) astrocyte- PERIPHERAL INFECTIONS AND MOLECULAR SIGNALS THAT SKEW selective expression of mutant SOD causes neu- ASTROGLIOSIS TOWARD CYTOTOXICITY ronal dysfunction and degeneration (Lobsiger and Cleveland 2007; Nagai et al. 2007; Yamanaka Peripheral infections can generate high levels et al. 2008), (2) astrocyte-selective deletion of circulating cytokines and inflammatory media- the endoribonuclease, Dicer, cases of severe tors, such as microbial PAMPs, which can gain ataxia, progressive cerebellar degeneration, sei- access to CNS lesions with compromised zures, and premature death (Taoet al. 2011), and blood–brain barrier (Burda and Sofroniew (3) astrocyte-selective deletion of the Wnt sig- 2014). Exposure of astrocytes to PAMPs, such naling molecule adenopolyposis coli causes as lipopolysaccharide and certain cytokines, delayed degeneration of cerebellar Purkinje neu- drives astrocyte transcriptome profiles toward rons (Wang et al. 2011). Experimental disrup- a proinflammatory and cytotoxic potential tion of various astrocyte signaling molecules can (Hamby et al. 2012; Zamanian et al. 2012; So- alter astrogliosis and cause neuronal dysfunc- froniew 2013). This response is not surprising tion and tissue degeneration during traumatic because limiting infection spread is likely to injury,ischemia,andautoimmuneattack(Okada have shaped the evolution of injury responses

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 11 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

M.V. Sofroniew

in all tissues including the CNS (Burda and So- lassa and Haydon 2010; Paradise et al. 2012; froniew 2014). If sterile CNS insults are comor- Czeh and Di Benedetto 2013; Martin et al. bid with peripheral infections, there is a strong 2013), and certain childhood behavioral disor- possibility that the molecular expression profile ders with altered synapse development (Ste- and function of reactive glia may be altered in phan et al. 2012; Clarke and Barres 2013). As- ways that can exacerbate tissue damage and trocytes are primary targets of cytokines and compromise neural repair (Burda and Sofro- inflammatory mediators increasingly implicat- niew 2014). In keeping with these experimental ed not only in sickness behaviors, such as social findings, clinical epidemiological observations withdrawal, reduced activity, loss of appetite, indicate that peripheral infections have a nega- and cognitive disturbances, but also in certain tive impact on neurological outcome after spi- kinds of sleep disturbances, mood disturbances, nal cord injury (Failli et al. 2012). The potential major depressive disorders, and CNS develop- impact of concurrent low-grade infections, or mental disorders (Dantzer et al. 2008; Dowlati even of microbiome (Heintz and Mair 2014), et al. 2010; Irwin and Cole 2011; Jurgens et al. on astrogliosis and responses to CNS insults is 2012; Patterson 2012; Sofroniew 2013). Mod- likely to be an interesting area of future research. ulation of astrocyte physiology and function by cytokines and inflammatory mediators has the potential to impact such behaviors (Miller ASTROGLIOSIS AND NEURONAL and O’Callaghan 2005; Sofroniew 2013). Al- FUNCTION AND BEHAVIOR though such interactions are only beginning to Astrogliosis has the potential to modify astro- be studied, early information suggests a strong cyte functions in ways that influence neuronal potential for identifying new ways in which CNS functions, including behaviors. In the healthy cellular responses to injury or disease impact CNS, astrocytes (1) maintain the extracellular systems-level neural functions and behaviors homeostasis of Kþ, water, pH, and transmitter (Sofroniew 2013). uptake (Halassa and Haydon 2010; Jin et al. 2013), (2) may modulate synaptic activity di- ASTROGLIOSIS AS A THERAPEUTIC TARGET rectly via release of “gliotransmitters” (Volterra and Meldolesi 2005; Halassa and Haydon 2010; There is increasing recognition that molecular Hamilton and Attwell 2010; Henneberger et al. mechanisms associated with specific functions 2010), and (3) take part in the formation and and effects of astrocytes and astrogliosis may pruning of synapses (Chung et al. 2013; Clarke be potential targets for novel therapeutic strat- and Barres 2013). Functional changes under- egies for CNS disorders (Hamby and Sofroniew gone by reactive astrocytes in response to spe- 2010). As emphasized throughout this article, cific molecular triggers can impact neurons via astrogliosis exerts essential beneficial functions (1) down-regulation of glutamine synthase as- but can also give rise to harmful effects in spe- sociated with reduced inhibitory synaptic cur- cific contexts as determined by specific signaling rents in local neurons (Ortinski et al. 2010), (2) events. Thus, useful therapeutic strategies will increased expression of xCT (Slc7a11), a cys- need to be specifically targeted so as to preserve teine-glutamate transporter associated with in- or augment beneficial effects of astrogliosis creased glutamate signaling, seizures, and exci- while blocking or reducing harmful ones. The totoxicity (Jackman et al. 2010; Buckingham once prevalent view that wholesale blockade of et al. 2011), and (3) changes in the expression astrogliosis or astroglial scar formation could be of multiple GPCRs and G proteins and calcium a therapeutic strategy is no longer tenable and signaling evoked by their ligands (Hamby et al. would likely do more harm than good. Specific 2012). There is increasing evidence for astrocyte aspects of astrogliosis that are being explored as participation in complex behaviors including potential targets for therapeutic manipulations sleep (Halassa and Haydon 2010), pain (Hansen include mechanisms that regulate glutamate, and Malcangio 2013), mood, depression (Ha- enzymes that generate or neutralize reactive ox-

12 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

Astrogliosis

ygen species, and production of certain cyto- augmenting or attenuating specific aspects of kines (Hamby and Sofroniew 2010). Although astrogliosis. detailed examination is beyond the scope of this article, astrocyte regulation of glutamate clearance and the potential role of astrocytes ACKNOWLEDGMENTS and astrogliosis in glutamate excitotoxicity after Work in the author’s laboratory is supported by ischemia (Beppu et al. 2014; Sloan and Barres grants from the National Institutes of Health, 2014) and during neurodegenerative disease The Dr. Miriam and Sheldon G. Adelson Med- (Maragakis and Rothstein 2006) are of particu- ical Foundation, and Wings for Life. lar interest. Pharmacological manipulation of expression or activity of astrocyte glutamate transporters is being explored, and molecules REFERENCES have been identified that enhance astrocyte-me- Aldrich A, Kielian T.2011. Central nervous system fibrosis is diated glutamate uptake sufficiently to provide associated with fibrocyte-like infiltrates. Am J Pathol 179: neuroprotection in animal models of stroke and 2952–2962. neurodegenerative disorders (Rothstein et al. Anderson MA, Ao Y, Sofroniew MV.2014. Heterogeneity of 2005; Fontana et al. 2007). reactive astrocytes. Neurosci Lett 565C: 23–29. Argaw AT, Asp L, Zhang J, Navrazhina K, Pham T, Mariani JN, Mahase S, Dutta DJ, Seto J, Kramer EG, et al. 2012. Astrocyte-derived VEGF-A drives blood–brain barrier CONCLUDING REMARKS disruption in CNS inflammatory disease. J Clin Invest 122: 2454–2468. Astrogliosis in response to CNS damage and Bardehle S, Kruger M, Buggenthin F,Schwausch J, Ninkovic disease has been recognized for well over 100 J, Clevers H, Snippert HJ, Theis FJ, Meyer-Luehmann M, years. Recent progress shows that astrogliosis is Bechmann I, et al. 2013. Live imaging of astrocyte re- a complex and multifaceted process that can sponses to acute injury reveals selective juxtavascular proliferation. Nat Neurosci 16: 580–586. range from subtle and reversible alterations in Barnabe-Heider F, Goritz C, Sabelstrom H, Takebayashi H, gene expression and morphology to the pro- Pfrieger FW,Meletis K, Frisen J. 2010. Origin of new glial nounced and long-lasting changes associated cells in intact and injured adult spinal cord. Cell Stem Cell with scar formation. Responses of astrocytes to 7: 470–482. Barres BA. 2008. The mystery and magic of glia: A perspec- CNS insults are controlled in a context-depen- tive on their roles in health and disease. Neuron 60: 430– dent manner by specific signaling mechanisms 440. that mediate numerous essential beneficial Bell RD, Winkler EA, Singh I, Sagare AP, Deane R, Wu Z, functions, but under certain circumstances can Holtzman DM, Betsholtz C, Armulik A, Sallstrom J, et al. 2012. Apolipoprotein E controls cerebrovascular integri- lead to harmful effects. Simplistic notions that ty via cyclophilin A. Nature 485: 512–516. astrogliosis and scar formation are maladaptive, Beppu K, Sasaki T, Tanaka KF, Yamanaka A, Fukazawa Y, uniformly detrimental responses, and that com- Shigemoto R, Matsui K. 2014. Optogenetic countering of glial acidosis suppresses glial glutamate release and plete blockade of reactive astrogliosis per se will ischemic brain damage. Neuron 81: 314–320. be beneficial, are refuted by findings from many Bhalala OG, Pan L, Sahni V, McGuire TL, Gruner K, Tour- laboratories. Essential functions of astrogliosis tellotte WG, Kessler JA. 2012. microRNA-21 regulates and scar formation include the protection of astrocytic response following spinal cord injury. J Neuro- sci 32: 17935–17947. neural cells, the restriction of inflammation Bianchi MG, Bardelli D, Chiu M, Bussolati O. 2013. Chang- and infection, and the preservation of tissue es in the expression of the glutamate transporter EAAT3/ and function. Dysfunctions of astrocytes and EAAC1 in health and disease. Cell Mol Life Sci 71: 2001– astrogliosis can also occur, either through gain 2015. Bonneh-Barkay D, Wang G, Laframboise WA, Wiley CA, of abnormal effects or loss of normal functions, Bissel SJ. 2012. Exacerbation of experimental autoim- and contribute to, or be primary causes of, a mune in the absence of breast regres- variety of CNS disorders. As a result, astrocytes sion protein 39/chitinase 3-like 1. J Neuropathol Exp and astrogliosis are increasingly recognized as Neurol 71: 948–958. Brambilla R, Bracchi-Ricard V,Hu WH, Frydel B, Bramwell potential targets for novel therapeutic strategies. A, Karmally S, Green EJ, Bethea JR. 2005. Inhibition of Effective therapies will need to be directed at astroglial nuclear factor-kB reduces inflammation and

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 13 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

M.V. Sofroniew

improves functional recovery after spinal cord injury. J Faideau M, Kim J, Cormier K, Gilmore R, WelchM, Auregan Exp Med 202: 145–156. G, Dufour N, Guillermier M, Brouillet E, Hantraye P, et Brambilla R, Persaud T, Hu X, Karmally S, Shestopalov VI, al. 2010. In vivo expression of polyglutamine-expanded Dvoriantchikova G, Ivanov D, Nathanson L, Barnum huntingtin by mouse striatal astrocytes impairs gluta- SR, Bethea JR. 2009. Transgenic inhibition of astroglial mate transport: A correlation with Huntington’s disease NF-kB improves functional outcome in experimental subjects. Hum Mol Genet 19: 3053–3067. autoimmune encephalomyelitis by suppressing chronic Failli V, Kopp MA, Gericke C, Martus P, Klingbeil S, central nervous system inflammation. J Immunol 182: Brommer B, Laginha I, Chen Y, DeVivo MJ, Dirnagl U, 2628–2640. et al. 2012. Functional neurological recovery after spinal Brenner M, Johnson AB, Boespflug-Tanguy O, Rodriguez D, cord injury is impaired in patients with infections. Brain Goldman JE, Messing A. 2001. Mutations in GFAP, en- 135: 3238–3250. coding glial fibrillary acidic protein, are associated with Faulkner JR, Herrmann JE, Woo MJ, Tansey KE, Doan NB, Alexander disease. Nat Genet 27: 117–120. Sofroniew MV. 2004. Reactive astrocytes protect tissue Brosius Lutz A, Barres BA. 2014. Contrasting the glial re- and preserve function after spinal cord injury. J Neurosci sponse to axon injury in the central and peripheral ner- 24: 2143–2155. vous systems. Dev Cell 28: 7–17. Fontana AC, de Oliveira Beleboni R, Wojewodzic MW, Fer- Buckingham SC, Campbell SL, Haas BR, Montana V, Robel reira Dos Santos W, Coutinho-Netto J, Grutle NJ, Watts S, Ogunrinu T,Sontheimer H. 2011. Glutamate release by SD, Danbolt NC, Amara SG. 2007. Enhancing glutamate primary brain tumors induces epileptic activity. Nat Med transport: mechanism of action of Parawixin1, a neuro- 17: 1269–1274. protective compound from Parawixia bistriata spider venom. Mol Pharmacol 72: 1228–1237. Buffo A, Rite I, Tripathi P,Lepier A, Colak D, Horn AP,Mori T,Gotz M. 2008. Origin and progeny of reactive gliosis: A Gadea A, Schinelli S, Gallo V. 2008. Endothelin-1 regulates source of multipotent cells in the injured brain. Proc Natl astrocyte proliferation and reactive gliosis via a JNK/ Acad Sci 105: 3581–3586. c-Jun signaling pathway. J Neurosci 28: 2394–2408. Bundesen LQ, Scheel TA, Bregman BS, Kromer LF. 2003. Gao K, WangCR, Jiang F,WongAY,Su N, Jiang JH, Chai RC, Ephrin-B2 and EphB2 regulation of astrocyte-meningeal Vatcher G, Teng J, Chen J, et al. 2013. Traumatic scratch fibroblast interactions in response to spinal cord lesions injury in astrocytes triggers calcium influx to activate the in adult rats. J Neurosci 23: 7789–7800. JNK/c-Jun/AP-1 pathway and switch on GFAP expres- Burda JE, Sofroniew MV.2014. Reactive gliosis and the mul- sion. Glia 61: 2063–2077. ticellular response to CNS damage and disease. Neuron Garcia AD, Petrova R, Eng L, Joyner AL. 2010. Sonic Hedge- 81: 229–248. hog regulates discrete populations of astrocytes in the adult mouse forebrain. J Neurosci 30: 13597–13608. Bush TGNP, Horner CH, Polito A, Ostenfeld T, Svendsen CN, Mucke L, Johnson MH, Sofroniew MV. 1999. Leu- Halassa MM, Haydon PG. 2010. Integrated brain circuits: kocyte infiltration, neuronal degeneration and neurite Astrocytic networks modulate neuronal activity and be- outgrowth after ablation of scar-forming, reactive astro- havior. Annu Rev Physiol 72: 335–355. cytes in adult transgenic mice. Neuron 23: 297–308. Hamby ME, Sofroniew MV. 2010. Reactive astrocytes as Chung WS, Clarke LE, WangGX, Stafford BK, Sher A, Chak- therapeutic targets for CNS disorders. Neurotherapeutics raborty C, Joung J, Foo LC, Thompson A, Chen C, et al. 7: 494–506. 2013. Astrocytes mediate synapse elimination through Hamby ME, Coppola G, Ao Y, Geschwind DH, Khakh BS, MEGF10 and MERTK pathways. Nature 504: 394–400. Sofroniew MV. 2012. Inflammatory mediators alter the Clarke LE, Barres BA. 2013. Emerging roles of astrocytes in astrocyte transcriptome and calcium signaling elicited by neural circuit development. Nat Rev Neurosci 14: 311– multiple G-protein-coupled receptors. J Neurosci 32: 321. 14489–14510. Czeh B, Di Benedetto B. 2013. Antidepressants act directly Hamilton NB, Attwell D. 2010. Do astrocytes really exocy- on astrocytes: Evidences and functional consequences. tose ? Nat Rev Neurosci 11: 227–238. Eur Neuropsychopharmacol 23: 171–185. Hansen RR, Malcangio M. 2013. Astrocytes-multitaskers in Dantzer R, O’Connor JC, Freund GG, Johnson RW, Kelley chronic pain. Eur J Pharmacol 716: 120–128. KW. 2008. From inflammation to sickness and depres- Haroon F, Drogemuller K, Handel U, Brunn A, Reinhold D, sion: When the immune system subjugates the brain. Nat Nishanth G, Mueller W, Trautwein C, Ernst M, Deckert Rev Neurosci 9: 46–56. M, et al. 2011. Gp130-dependent astrocytic survival is Dong Y, Benveniste EN. 2001. Immune function of astro- critical for the control of autoimmune central nervous cytes. Glia 36: 180–190. system inflammation. J Immunol 186: 6521–6531. Dowlati Y, Herrmann N, Swardfager W, Liu H, Sham L, Heintz C, Mair W.2014. Youare what you host: Microbiome Reim EK, Lanctot KL. 2010. A meta-analysis of cytokines modulation of the aging process. Cell 156: 408–411. in major depression. Biol Psychiatry 67: 446–457. Henneberger C, Papouin T, Oliet SH, Rusakov DA. 2010. Drogemuller K, Helmuth U, Brunn A, Sakowicz-Burkiewicz Long-term potentiation depends on release of D-serine M, Gutmann DH, Mueller W, Deckert M, Schluter D. from astrocytes. Nature 463: 232–236. 2008. Astrocyte gp130 expression is critical for the con- Herrmann JE, Imura T, Song B, Qi J, Ao Y, Nguyen TK, trol of Toxoplasma encephalitis. J Immunol 181: 2683– Korsak RA, Takeda K, Akira S, Sofroniew MV. 2008. 2693. STAT3 is a critical regulator of astrogliosis and scar for-

14 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

Astrogliosis

mation after spinal cord injury. J Neurosci 28: 7231– Maragakis NJ, Rothstein JD. 2006. Mechanisms of Disease: 7243. Astrocytes in neurodegenerative disease. Nat Clin Pract Hutchison ER, Kawamoto EM, Taub DD, Lal A, Abdelmoh- Neurol 2: 679–689. sen K, Zhang Y,WoodWH 3rd, Lehrmann E, Camandola Martin JL, Magistretti PJ, Allaman I. 2013. Regulation of S, Becker KG, et al. 2013. Evidence for miR-181 involve- and energy metabolism by antide- ment in neuroinflammatory responses of astrocytes. Glia pressants in astrocytes. Curr Drug Targets 14: 1308–1321. 61: 1018–1028. Meletis K, Barnabe-Heider F,Carlen M, Evergren E, Tomilin Irwin MR, Cole SW.2011. Reciprocal regulation of the neu- N, Shupliakov O, Frisen J. 2008. Spinal cord injury reveals ral and innate immune systems. Nat Rev Immunol 11: multilineage differentiation of ependymal cells. PLoS Biol 625–632. 6: e182. Messing A, Head MW, Galles K, Galbreath EJ, Goldman JE, Jackman NA, Uliasz TF, Hewett JA, Hewett SJ. 2010. Regu- Brenner M. 1998. Fatal with astrocyte lation of system x – activity and expression in astrocytes c inclusions in GFAP transgenic mice. Am J Pathol 152: by interleukin-1b: Implications for hypoxic neuronal in- 391–398. jury. Glia 58: 1806–1815. Messing A, Brenner M, Feany MB, Nedergaard M, Goldman Ji YF,Zhou L, Xie YJ, Xu SM, Zhu J, TengP,Shao CY,WangY, JE. 2012. Alexander disease. J Neurosci 32: 5017–5023. Luo JH, Shen Y. 2013. Upregulation of glutamate trans- Middeldorp J, Hol EM. 2011. GFAP in health and disease. porter GLT-1 by mTOR-Akt-NF-kB cascade in astrocytic Prog Neurobiol 93: 421–443. oxygen-glucose deprivation. Glia 61: 1959–1975. Miller DB, O’Callaghan JP.2005. Depression, cytokines, and Jin BJ, Zhang H, Binder DK, VerkmanAS. 2013. Aquaporin- glial function. Metabolism 54: 33–38. 4-dependent Kþ and water transport modeled in brain Myer DJ, Gurkoff GG, Lee SM, Hovda DA, Sofroniew MV. extracellular space following neuroexcitation. J Gen Phys- 2006. Essential protective roles of reactive astrocytes in iol 141: 119–132. . Brain 129: 2761–2772. Jurgens HA, Amancherla K, Johnson RW. 2012. Influenza Nagai M, Re DB, Nagata T,Chalazonitis A, Jessell TM, Wich- infection induces neuroinflammation, alters hippocam- terle H, Przedborski S. 2007. Astrocytes expressing ALS- pal neuron morphology, and impairs cognition in adult linked mutated SOD1 release factors selectively toxic to mice. J Neurosci 32: 3958–3968. motor neurons. Nat Neurosci 10: 615–622. Kang W, Hebert JM. 2011. Signaling pathways in reactive Nawashiro H, Messing A, Azzam N, Brenner M. 1998. Mice astrocytes, a genetic perspective. Mol Neurobiol 43: 147– lacking GFAP are hypersensitive to traumatic cerebrospi- 154. nal injury. Neuroreport 9: 1691–1696. Kitley J, Waters P, Woodhall M, Leite MI, Murchison A, Okada S, Nakamura M, Katoh H, Miyao T, Shimazaki T, George J, Kuker W, Chandratre S, Vincent A, Palace J. Ishii K, Yamane J, Yoshimura A, Iwamoto Y, Toyama Y, 2014. Neuromyelitis optica spectrum disorders with et al. 2006. Conditional ablation of Stat3 or Socs3 dis- aquaporin-4 and -oligodendrocyte glycoprotein closes a dual role for reactive astrocytes after spinal cord antibodies: A comparative study. JAMA Neurol 71: injury. Nature Med 12: 829–834. 273–283. Ortinski PI, Dong J, Mungenast A, Yue C, TakanoH, Watson Kostianovsky AM, Maier LM, Anderson RC, Bruce JN, An- DJ, Haydon PG, Coulter DA. 2010. Selective induction of derson DE. 2008. Astrocytic regulation of human mono- astrocytic gliosis generates deficits in neuronal inhibi- cytic/microglial activation. J Immunol 181: 5425–5432. tion. Nat Neurosci 13: 584–591. Lennon VA,Kryzer TJ, Pittock SJ, Verkman AS, Hinson SR. Paradise MB, Naismith SL, Norrie LM, Graeber MB, Hickie 2005. IgG marker of optic-spinal binds IB. 2012. The role of glia in late-life depression. Int Psy- to the aquaporin-4 water channel. J Exp Med 202: 473– chogeriatr 24: 1878–1890. 477. Patterson PH. 2012. Maternal infection and autism. Brain Behav Immun 26: 393. Levison SW, Jiang FJ, Stoltzfus OK, Ducceschi MH. 2000. IL-6-type cytokines enhance epidermal growth factor- Pawlak J, Brito V, Kuppers E, Beyer C. 2005. Regulation of stimulated astrocyte proliferation. Glia 32: 328–337. glutamate transporter GLAST and GLT-1 expression in astrocytes by estrogen. Brain Res Mol Brain Res 138: 1–7. Li L, Lundkvist A, Andersson D, Wilhelmsson U, Nagai N, Pekny M, Nilsson M. 2005. Astrocyte activation and reactive Pardo AC, Nodin C, Stahlberg A, Aprico K, Larsson K, et gliosis. Glia 50: 427–434. al. 2008. Protective role of reactive astrocytes in brain ischemia. J Cereb Blood Flow Metab 28: 468–481. Reilkoff RA, Bucala R, Herzog EL. 2011. Fibrocytes: Emerg- ing effector cells in chronic inflammation. Nat Rev Im- Lobsiger CS, Cleveland DW. 2007. Glial cells as intrinsic munol 11: 427–435. components of non-cell-autonomous neurodegenerative Roemer SF, Parisi JE, Lennon VA,Benarroch EE, Lassmann disease. Nat Neurosci 10: 1355–1360. H, Bruck W, Mandler RN, Weinshenker BG, Pittock SJ, Macauley SL, Pekny M, Sands MS. 2011. The role of atten- Wingerchuk DM, et al. 2007. Pattern-specific loss of uated astrocyte activation in infantile neuronal ceroid aquaporin-4 immunoreactivity distinguishes neuromye- lipofuscinosis. J Neurosci 31: 15575–15585. litis optica from multiple sclerosis. Brain 130: 1194– Magnus T,Carmen J, Deleon J, Xue H, Pardo AC, Lepore AC, 1205. Mattson MP, Rao MS, Maragakis NJ. 2008. Adult glial Rothstein JD, Dykes-Hoberg M, Pardo CA, Bristol LA, Jin L, precursor proliferation in mutant SOD1G93A mice. Glia Kuncl RW, Kanai Y, Hediger MA, Wang Y, Schielke JP, et 56: 200–208. al. 1996. Knockout of glutamate transporters reveals a

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 15 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

M.V. Sofroniew

major role for astroglial transport in excitotoxicity and 2014. Astrocyte Kir4.1 ion channel deficits contribute to clearance of glutamate. Neuron 16: 675–686. neuronal dysfunction in Huntington’s disease model Rothstein JD, Patel S, Regan MR, Haenggeli C, Huang YH, mice. Nat Neurosci 17: 694–703. Bergles DE, Jin L, Dykes Hoberg M, Vidensky S, Chung Tsai HH, Li H, Fuentealba LC, Molofsky AV, Taveira-Mar- DS, et al. 2005. b-Lactam antibiotics offer neuroprotec- ques R, Zhuang H, Tenney A, Murnen AT, Fancy SP,Mer- tion by increasing glutamate transporter expression. Na- kle F, et al. 2012. Regional astrocyte allocation regulates ture 433: 73–77. CNS synaptogenesis and repair. Science 337: 358–362. Sato DK, Callegaro D, Lana-Peixoto MA, Waters PJ, de Hai- Verkhratsky A, Sofroniew MV, Messing A, deLanerolle NC, dar Jorge FM, Takahashi T, Nakashima I, Apostolos-Per- Rempe D, Rodriguez JJ, Nedergaard M. 2012. Neurolog- eira SL, Talim N, Simm RF, et al. 2014. Distinction ical diseases as primary gliopathies: A reassessment of between MOG antibody-positive and AQP4 antibody- neurocentrism. ASN Neuro 4. positive NMO spectrum disorders. 82: 474– Volterra A, Meldolesi J. 2005. Astrocytes, from brain glue to 481. communication elements: The revolution continues. Nat Silver J, Miller JH. 2004. Regeneration beyond the . Rev Neurosci 6: 626–640. Nat Rev Neurosci 5: 146–156. VoskuhlRR, Peterson RS, Song B, Ao Y,Morales LB, Tiwari- Sirko S, Behrendt G, Johansson PA, Tripathi P,Costa M, Bek Woodruff S, Sofroniew MV. 2009. Reactive astrocytes S, Heinrich C, Tiedt S, Colak D, Dichgans M, et al. 2013. form scar-like perivascular barriers to leukocytes during Reactive glia in the injured brain acquire stem cell prop- adaptive immune inflammation of the CNS. J Neurosci erties in response to Sonic Hedgehog. Cell Stem Cell 12: 29: 11511–11522. 426–439. Wang X, Imura T, Sofroniew MV, Fushiki S. 2011. Loss of Sloan SA, Barres BA. 2014. The detrimental role of glial adenomatous polyposis coli in Bergmann glia disrupts acidification during ischemia. Neuron 81: 221–223. their unique architecture and leads to cell nonautono- Sofroniew MV.2005. Reactive astrocytes in neural repair and mous neurodegeneration of cerebellar Purkinje neurons. protection. Neuroscientist 5: 400–407. Glia 59: 857–868. Sofroniew MV.2009. Molecular dissection of reactive astro- Wanner IB, Anderson MA, Song B, Levine J, Fernandez A, gliosis and glial scar formation. Trends Neurosci 32: 638– Gray-Thompson Z, Ao Y, Sofroniew MV.2013. Glial scar 647. borders are formed by newly proliferated, elongated as- Sofroniew MV. 2013. Multiple roles for astrocytes as effec- trocytes that interact to corral inflammatory and fibrotic tors of cytokines and inflammatory mediators. Neurosci- cells via STAT3-dependent mechanisms after spinal cord entist 20: 160–172. injury. J Neurosci 33: 12870–12886. Sofroniew MV, Vinters HV. 2010. Astrocytes: Biology and White RE, McTigue DM, Jakeman LB. 2010. Regional het- pathology. Acta Neuropathol 119: 7–35. erogeneity in astrocyte responses following contusive spi- Spence RD, Hamby ME, Umeda E, Itoh N, Du S, Wisdom nal cord injury in mice. J Comp Neurol 518: 1370–1390. AJ, Cao Y, Bondar G, Lam J, Ao Y, et al. 2011. Neuro- Wilhelmsson U, Bushong EA, Price DL, Smarr BL, Phung V, protection mediated through estrogen receptor-a in as- Terada M, Ellisman MH, Pekny M. 2006. Redefining the trocytes. Proc Natl Acad Sci 108: 8867–8872. concept of reactive astrocytes as cells that remain within Spence RD, Wisdom AJ, Cao Y, Hill HM, Mongerson CR, their unique domains upon reaction to injury. Proc Natl Stapornkul B, Itoh N, Sofroniew MV, Voskuhl RR. 2013. Acad Sci 103: 17513–17518. Estrogen mediates neuroprotection and anti-inflamma- Yamanaka K, Chun SJ, Boillee S, Fujimori-Tonou N, Yama- tory effects during EAE through ERa signaling on astro- shita H, Gutmann DH, Takahashi R, Misawa H, Cleve- cytes but not through ERb signaling on astrocytes or land DW. 2008. Astrocytes as determinants of disease neurons. J Neurosci 33: 10924–10933. progression in inherited amyotrophic lateral sclerosis. Stephan AH, Barres BA, Stevens B. 2012. The complement Nat Neurosci 11: 251–253. system: An unexpected role in synaptic pruning during Zamanian JL, Xu L, Foo LC, Nouri N, Zhou L, Giffard RG, development and disease. Annu Rev Neurosci 35: 369– Barres BA. 2012. Genomic analysis of reactive astroglio- 389. sis. J Neurosci 32: 6391–6410. Tao J, Wu H, Lin Q, Wei W, Lu XH, Cantle JP, Ao Y, Olsen Zhang Y, Barres BA. 2010. Astrocyte heterogeneity: An un- RW, Yang XW, Mody I, et al. 2011. Deletion of astroglial derappreciated topic in neurobiology. Curr Opin Neuro- dicer causes non-cell-autonomous neuronal dysfunction biol 20: 588–594. and degeneration. J Neurosci 31: 8306–8319. Zlokovic BV.2011. Neurovascular pathways to neurodegen- Tong X, Ao Y, Faas GC, Nwaobi SE, Xu J, Haustein MD, eration in Alzheimer’s disease and other disorders. Nat Anderson MA, Mody I, Olsen ML, Sofroniew MV, et al. Rev Neurosci 12: 723–738.

16 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a020420 Downloaded from http://cshperspectives.cshlp.org/ on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press

Astrogliosis

Michael V. Sofroniew

Cold Spring Harb Perspect Biol published online November 7, 2014

Subject Collection Glia

The Nodes of Ranvier: Molecular Assembly and Oligodendrocyte Development and Plasticity Maintenance Dwight E. Bergles and William D. Richardson Matthew N. Rasband and Elior Peles Microglia in Health and Disease : Myelination and Axonal Richard M. Ransohoff and Joseph El Khoury Support Mikael Simons and Klaus-Armin Nave The Astrocyte: Powerhouse and Recycling Center Drosophila Central Nervous System Glia Bruno Weber and L. Felipe Barros Marc R. Freeman Microglia Function in Central Nervous System Perisynaptic Schwann Cells at the Neuromuscular Development and Plasticity Synapse: Adaptable, Multitasking Glial Cells Dorothy P. Schafer and Beth Stevens Chien-Ping Ko and Richard Robitaille Transcriptional and Epigenetic Regulation of Astrocytes Control Synapse Formation, Function, Oligodendrocyte Development and Myelination in and Elimination the Central Nervous System Won-Suk Chung, Nicola J. Allen and Cagla Eroglu Ben Emery and Q. Richard Lu Origin of Microglia: Current Concepts and Past Myelination Controversies James L. Salzer Florent Ginhoux and Marco Prinz Glia Disease and Repair−− Schwann Cells: Development and Role in Nerve Robin J.M. Franklin and Steven A. Goldman Repair Kristján R. Jessen, Rhona Mirsky and Alison C. Lloyd Astrocytes in Neurodegenerative Disease Perineurial Glia Hemali Phatnani and Tom Maniatis Sarah Kucenas

For additional articles in this collection, see http://cshperspectives.cshlp.org/cgi/collection/

Copyright © 2014 Cold Spring Harbor Laboratory Press; all rights reserved