Physiologic Role for ‘‘Inducible’’ Nitric Oxide Synthase: A New Form of Astrocytic–Neuronal Interface YAEL AMITAI * Department of Physiology and Neurobiology, Faculty of Health Sciences, Ben-Gurion University, Beer-Sheva, Israel KEY WORDS excitatory synapses; DAF-2DA; synaptic release; LTP; nitric oxide ABSTRACT Nitric oxide (NO) has been long recognized as an atypical neu- ronal messenger affecting excitatory synaptic transmission, but its cellular source has remained unresolved as the neuro- nal isoform of NO synthase (nNOS) in many brain regions is expressed only by small subsets of inhibitory neurons. It is generally believed that the glial NO-producing isoform (iNOS) is not expressed in the normal brain, but rather it undergoes a transcription-mediated up-regulation following an immunolog- ical challenge. Therefore, the involvement of iNOS in modulat- ing normal neuronal functions has been largely ignored. Here I review evidence to the contrary: I summarize data pointing to the existence of a functioning iNOS in normal undisturbed mammalian brains, and experimental results tracing this expression to astrocytes. Finally, I review recent findings asserting that iNOS-dependent NO modulates synaptic release from presynaptic terminals. Based on these data, I propose that astrocytes express basal levels of iNOS. Flanking synaptic elements, astrocytes are perfectly positioned to release NO and affect synaptic transmission. V V C 2010 Wiley-Liss, Inc. INTRODUCTION Nitric oxide (NO), a small gaseous molecule, serves as a signaling molecule and participates in a plethora of biologi- cal functions. In the mammalian brain, NO has been defined as an atypical neurotransmitter (Boehning and Snyder, 2003), being secreted by neurons and having direct effects on their electrical activity. More specifically, it has been postulated that NO serves as a retrograde messenger between the post- and the presynaptic elements in the most abundant excitatory synapses. Accordingly, it has been demonstrated that NO participates in the induction of long- term potentiation (LTP) of the release machinery (reviewed by Garthwaite, 2008; Prast and Philippu, 2001). It has been largely assumed that a neuronal Ca 21 /calmodulin-regu- lated NO synthase (nNOS) consists in the principal isoform producing the NO that acts at the synapse (e.g. Arancio et al., 1996; Micheva et al., 2003; Prast and Philippu, 2001). However, in many brain regions such as the neocortex, nNOS is expressed only by small subpopulations of GABAergic neurons (Karagiannis et al., 2009; Tricoire et al., 2010). The field is therefore faced with a conundrum, on the one hand many experimental findings indicate that NO is involved in regulating synaptic transmission between glutamatergic neurons, while on the other hand principal excitatory neurons in many brain regions do not express the NO-synthesizing enzyme, implying nNOS is an unlikely source of modulation to these synapses. Another Ca 21 /calmodulin-regulated NOS isoform is expressed pri- marily by endothelial cells (eNOS) and is known to take part in blood flow regulation, although it has been proposed that NO produced by the cerebral vasculature also affects nearby neurons (Garthwaite et al., 2006). A third, ‘‘inducible’’ NOS isoform (iNOS) is expressed by glia cells, and acts in a Ca 21 -independent manner (reviewed by Aktan, 2004; Alderton et al., 2001). The involvement of this isoform in modulating neuronal activ- ity has been largely ignored, as the prevailing viewpoint maintains that this enzyme is not detectable under normal conditions, but rather it is up-regulated by gene induction following detrimental events such as ischemia or inflam- mation (reviewed by Aktan, 2004; Murphy, 2000; Saha and Pahan, 2006), where NO participates in the regulation of anti-inflammatory processes. Nonetheless, other findings cast doubt on this general contention regarding the role that iNOS plays in the central nervous system (CNS). Spe- cifically, several studies over the years have pointed to a weak constitutive expression of iNOS in various brain regions, and raised the possibility of astrocytic-derived NO participating in physiological brain processes (Table 1; Buskila et al., 2005; Chan et al., 2001, 2003; Starkey et al., 2001). These data imply an additional source for NO in the normally functioning brain, and a novel, entirely uncharted mode of astrocytic–neuronal interaction. In the following manuscript, I review the evidence for baseline iNOS expression by astrocytes in the CNS, dis- cuss its probable role in regulating synaptic transmission, and highlight future directions as well as technical hur- dles for the research of this new modulatory pathway. EVIDENCE FOR CONSTITUTIVELY ACTIVE iNOS The active iNOS enzyme was found in the normal developing rat brain during late embryonic and early Grant sponsor: Israel Science Foundation (ISF) (to Y.A.). *Correspondence to: Yael Amitai, Department of Physiology and Neurobiology, Faculty of Health Sciences, POB 653, Ben-Gurion University, Beer-Sheva, Israel 84105. E-mail: yaela@bgu.ac.il Received 9 April 2010; Accepted 6 July 2010 DOI 10.1002/glia.21057 Published online 24 August 2010 in Wiley Online Library (wileyonlinelibrary.com) GLIA 58:1775–1781 (2010) V V C 2010 Wiley-Liss, Inc. Review