<<

726 Cell Research (2017) 27:726-727. © 2017 IBCB, SIBS, CAS All rights reserved 1001-0602/17 $ 32.00 RESEARCH HIGHLIGHT www.nature.com/cr

Seducing astrocytes to the dark side Cell Research (2017) 27:726-727. doi:10.1038/cr.2017.37; published online 17 March 2017

After injury and in disease of the trocyte reactivity for CNS disorders, de force of trial and error experi- central nervous system (CNS), local remain central questions in the study of ments to identify the microglia-derived cells called astrocytes respond with the neurobiology of disease. mediator(s) of the A1 phenotype. By diverse molecular changes whose func- In a recent study from the Barres screening numerous combinations of tional consequences are incompletely group, Liddelow et al. [6] identify a from diverse functional groups understood. A combined genomic and mechanism through which classically for their ability to induce the A1 reactive experimental analysis shows that clas- activated microglia, innate inflammatory gene expression profile in cultured astro- sically-activated microglia, which are cells that are resident in CNS neural tis- cytes, the authors identified the innate immune cells resident in CNS sue, can drive reactive astrocytes into a Interleukin 1 alpha (Il1α) and Tumor neural tissue, release molecules that neurotoxic state (Figure 1). The authors Necrosis Factor (TNF), in combination drive astrocytes into a neurotoxic state, investigated the regulation and function with compliment subcomponent C1q, raising important questions about of two markedly different subtypes of as required and sufficient to recapitulate potential adaptive and maladaptive reactive astrocytes previously character- the A1 expression profile induced by mi- functions of such a mechanism. ized in the Barres lab by genomic analy- croglia. Astrocyte transformation to A1 Astrocytes, a type of glial cell found sis of mouse models of two divergent could be rapidly reversed by treatment throughout the brain and spinal cord, play forms of CNS insult: (i) neuroinflam- with Transforming Growth Factor beta indispensable roles in sustaining normal mation due to systemic injection of a (TGFβ), but not simply by the removal neurologic function and responding bacterial endotoxin, of A1 mediators Il1α, TNF and C1q, to all forms of central nervous system (LPS), and (ii) ischemic brain injury [3]. suggesting that states of reactivity may (CNS) injury and disease. Following Neuroinflammation-reactive astrocytes, endure until counteracted by specific CNS insults, astrocytes undergo rapid which in the previous study were found molecular signals in the injury micro- changes in gene expression, morphology to exhibit a potentially noxious pro- environment. This may be particularly and function collectively referred to as inflammatory genomic profile [3], were important for understanding persistent astrocyte reactivity. Astrocyte reactivity in the present study termed ‘A1’ [6]. In astrocyte reactivity in the context of is not a single all-or-none response, but contrast, ischemia-reactive astrocytes, chronic neurological disorders. instead is a diverse continuum of graded which exhibit a more neuro-supportive Next, using A1 astrocyte cultures, the and context-specific responses that may gene profile with enhanced expression of authors found that multiple well-estab- result in either adaptive or maladaptive metabolism-associated genes and neuro- lished astrocyte functions were severely effects [1, 2]. To begin dissecting and trophic factors, were termed ‘A2’. In their impaired in A1 cells, including the ability understanding the diversity, regulation first set of experiments, Liddelow et al. to promote structural and functional syn- and functions of reactive astrocyte re- [6] used a combination of in vivo quan- apse formation [7, 8], and phagocytose sponses, several studies have employed titative genomic profiling, genetically synaptic debris [9]. Remarkably, the loss quantitative genomic analyses to charac- modified mice and purified rodent astro- of various normal functions was accom- terize astrocyte gene expression profiles cyte cultures, and found that LPS acted panied by an unusual gain of function in various CNS disorders. The results of indirectly through microglia to induce — the ability to selectively kill neurons these studies highlights the existence of A1 astrocytes. Indeed, simply treating via an as yet unidentified secreted mol- molecularly diverse subtypes of reactive astrocyte cultures with LPS-stimulated ecule. Moreover, evidence is provided astrocytes that may serve a multitude microglia conditioned culture media was that the neurotoxic A1 astrocytes can of beneficial or deleterious roles [3-5]. sufficient to drive A1-associated gene play a central role in mediating the death Nevertheless, the identity of key cellular expression, while direct stimulation of of damaged neurons following CNS and molecular mechanisms driving such astrocytes with LPS was unable to induce traumatic injury in vivo. Preventing A1 diverse responses, and the functional the A1 reactivity profile. reactive astrocyte formation in response implications of different forms of as- The authors then conducted a tour to optic nerve crush injury, either by in

SPRINGER NATURE | Cell Research | www.cell-research.com 727 function, or is A1 function maladaptively removing healthy neurons that are being inappropriately targeted? Understanding this distinction is critical to determine whether inhibiting A1 astrocyte activity will be beneficial or not. Lastly, the ground-breaking study by Liddelow et al. underscores the notion that understanding and modulating as- trocyte reactivity represents a rich new area for identifying potential therapeutic targets for neurological disorders. In this regard, further research is also warranted to define astrocyte intracellular signaling Figure 1 Activated microglia can drive reactive astrocytes into a neurotoxic state. Lipopolysaccharide (LPS), a microbial endotoxin, activates microglia to release In- pathways that drive reactive changes in terleukin 1 alpha (Il1α), (TNF), and compliment subcompo- gene expression and thereby cell func- nent C1q. These microglia-derived molecular signals are required and sufficient to tion. Transcriptional regulatory mol- drive astrocytes into an ‘A1’ reactive state characterized by upregulated expression ecules ultimately transmit extracellular of A1-specific genes and the release of a still unidentified, but potent, that se- stimuli into changes in gene expression, lectively kills neurons. It is not yet clear whether this neurotoxic astrocyte function positioning them as promising targets to acts adaptively to remove irreversibly damaged and dysfunctional neurons, or acts selectively manipulate subtypes of reac- maladaptively to remove healthy neurons that are being inappropriately targeted. tive astrocytes across a broad range of CNS disorders. vivo administration of Il1α, TNF and C1q where naturally occurring cell death re- Joshua E Burda1, function-blocking , or using moves large numbers of neurons that are Michael V Sofroniew1 Il1α–/–TNF–/–C1q triple-knockout mice, either dysfunctional or inappropriately greatly attenuated the loss of axotomized wired [10]. Whether or not A1 astrocytes 1Department of Neurobiology, David Geffen School retinal ganglion neurons. Lastly, using participate in such mechanisms awaits of Medicine, University of California, Los Angeles, expression of complement component further study. Interestingly, the observa- Los Angeles, CA 90095, USA a 3 (C3) as a marker, A1 astrocytes were tion by Liddelow et al. that A1 astrocytes Correspondence: Joshua E Burda , Michael V Sofroniewb found to be present in tissue sections mediate the death of axotomized neurons aE-mail: [email protected] derived from patients with common might be consistent with such a role. An bE-mail: [email protected] neurodegenerative diseases, including equally important third question is to Huntington’s, Alzheimer’s, multiple scle- what degree do A1 astrocytes partici- References rosis, Parkinson’s and ALS, suggesting a pate in maladaptive neurodegenerative potential role for A1 astrocytes in human events. Neuronal loss is a feature of 1 Burda JE, Sofroniew MV. Neuron 2014; 81:229-248. CNS disorders. many neurodegenerative conditions and 2 Khakh BS, Sofroniew MV. Nat Neurosci These findings raise important ques- the underlying mechanisms are not well 2015; 18:942-952. tions. First, what is the identity of the understood. The observation by Lid- 3 Zamanian JL, Xu L, Foo LC, et al. J Neurosci produced by A1 astrocytes? delow et al. that astrocytes expressing the 2012; 32:6391-6410. 4 Zhang Y, Sloan SA, Clarke LE, et al. Neuron Work is underway in this regard. Second, A1 marker C3 are present in tissue from 2016; 89:37-53. what purpose(s) might be served by the a variety of human neurodegenerative 5 Anderson MA, Burda JE, Ren Y, et al. Nature existence and evolutionary conservation conditions suggests that A1 astrocytes 2016; 532:195-200. of a mechanism to generate neurotoxic may be exerting neurotoxic functions in 6 Liddelow SA, Guttenplan KA, Clarke LE, et al. Nature 2017; 541:481-487. astrocytes? One possibility could be the such conditions. Further work is called 7 Kucukdereli H, Allen NJ, Lee AT, et al. Proc adaptive function of removing malfunc- for. In particular, it will be interesting Natl Acad Sci USA 2011; 108:E440-E449. tioning neurons from neural networks to and critical to determine whether or 8 Allen NJ, Bennett ML, Foo LC, et al. Nature prevent circuit disruption and neurologic not potential A1 astrocyte neurotoxic 2012; 486:410-414. 9 Chung WS, Clarke LE, Wang GX, et al. Na- dysfunction after either injury, microbial activities are adaptive or maladaptive. ture 2013; 504:394-400. infection or neurodegenerative disease. Is A1 function adaptively removing 10 Cowan WM, Fawcett JW, O’Leary DD, et al. Another potential adaptive function irreversibly damaged or dysfunctional Science 1984; 225:1258-1265. might be during neurodevelopment, neurons that might disrupt neural circuit www.cell-research.com | Cell Research | SPRINGER NATURE