- Email: [email protected]
Location, Location, Location: Microglia Are Where They Live
Neuron
Previews Location, Location, Location: Microglia Are Where They Live Margaret M. McCarthy1,* 1Department of Pharmacology, University of Maryland School of Medicine, Baltimore, MD 21230, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.neuron.2017.07.005
A deep dive into microglia form and function reveals startling regional heterogeneity in number, morphology, activity, and transcriptomics in nuclei relevant to motor control and motivation (De Biase et al., 2017). Differences appear 2 weeks after birth, and depletion and recolonization in adulthood recapitulate the original phenotype, implicating local environment as mediators of microglial phenotype. A deep dive into microglia form and function reveals startling specific regional heterogeneity in terms of number, morphology, activity, and transcriptomics in deep brain nuclei with relevance to motor control and motivation and reward. Some regions, such as the substantia nigra reticulata (SNr), have markedly higher density of microglia overall and exist in a highly active state compared to closely related nuclei, either neuroanatomically or functionally, that have far fewer and simpler microglia. While the innate immune cells of the SNr are particularly metabolically and electrically active, those of the ventral tegmental area (VTA) show the most unique transcriptome. Depletion of microglia either chemically or genetically confirms that during recolonization the original phenotype is recapitulated, suggesting local environment is the driving force of microglial phenotype. Strong correlations between the density of astrocytes and microglia but a disconnect between both those populations and neuronal density raises multiple interesting questions about who is the master of the house. Collectively, these nuclei subserve some of the most important neurological functions of any organism, ranging from motor control to motivation and reward. Microglia are neither micro nor glia. Outsized in their impact, these innate immune cells of the brain resemble glia in appearance and partially in function but are unique influencers of the brain in both health and disease. The mysteries surrounding these cells abound, beginning with their surprising origins in the embryonic yolk sac, followed by migration into the brain before the blood-brain
barrier closes (Ginhoux et al., 2013; Nayak et al., 2014). Once safely ensconced in the CNS, they disperse and propagate until the tiling of the entire neuropil is complete. From there they grow and multiply in tandem with the developing brain and actively participate in sculpting the formation of neural circuits by regulating cell number and synaptic profiles (Hong et al., 2016). An essential feature of the brain is regional heterogeneity achieved via variance in cell number, density, phenotype, and connectivity. Neurons and glia (astrocytes and oligodendrocytes) are derived from multiple progenitor pools and diverge into hundreds of different, mostly post-mitotic phenotypes. By contrast, microglia are a unique mono-lineage derived from a macrophage precursor but found only in the CNS and are readily proliferative in response to a challenge such as injury (Cuadros and Navascue´s, 1998; Sierra et al., 2013; Streit, 2000). A natural assumption is that microglia are also mono-phenotypic, and if they do vary regionally, it is a transient byproduct of the environment in which they find themselves. But is this true? The paper by De Biase and colleagues in this issue directly addresses this question with in-depth analyses of three basic parameters: (1) density and morphology, (2) metabolic and activational state, and (3) transcriptomics of microglia of the basal ganglia, a collection of subcortical nuclei critical to motor control as well as motivation and reward (Figure 1). Regional comparison of microglia begins with simple questions: Is the density the same or different? Is the shape same or different? De Biase and colleagues
found both differed dramatically within the basal ganglia. The VTA is the source of dopamine neurons projecting to the nucleus accumbens (NAc), but it hosts on average 45% fewer microglia per unit area than its target. Both are dwarfed by the adjacent SNr, which has almost three times the density of microglia as the VTA. The dopamine neuron component of the SN, the pars compacta (SNc), resembles the VTA, another dopamine projection region. Intriguingly, the ratio of microglia to astrocytes remains constant across regions, suggesting either a common factor(s) determines their density or the existence of a hidden mutual co-dependency of these divergent yet functionally connected cells. The shape of microglia can be an indicator of function as these cells often physically contact others in order to influence them via highly localized release of signaling molecules, trophic factors, or phagocytosis of part or all of a cell. In the healthy adult basal ganglia, the microglia are ramified in appearance with long, thin, highly branched processes, much the same as elsewhere in the brain. The length and number of the processes are indirect indicators of the number of neurons potentially impacted by a given microglia. Here, too, regional variation was found, with microglia in the NAc having the longest processes compared to other basal ganglia nuclei, while those in the SNr were the most complexly branched. Morphology is important, but one of the more startling discoveries about microglia was their ongoing motility. Originally thought to be ‘‘quiescent,’’ we now know these cells are perpetual motion machines constantly ‘‘surveying’’ their
Neuron 95, July 19, 2017 ª 2017 Published by Elsevier Inc. 233
Neuron
Previews
Figure 1. Microglia Phenotype in the Basal Ganglia Comparison of microglia in terms of density, morphology, activational state, membrane properties, and transcriptomics revealed startling sharp regional heterogeneity in a group of functionally related nuclei comprising a portion of the basal ganglia. Some regions, such as the SNr, have markedly higher density of microglia overall, and they appear to be in a highly active state compared to closely related nuclei, either neuroanatomically or functionally, that have far fewer and simpler microglia.
environment and poking their processes into everybody else’s business (Nimmerjahn et al., 2005). Lysosome content is a measure of both phagocytic capacity and metabolic state, sort of a busy body indicator. The SNr is busy body central, with the most densely packed and metabolically active microglia in the basal ganglia. This is surprising given that the density of neurons here is the lowest of all the regions examined, raising the interesting question of whether neurons suppress microglia proliferation and/or survival, or vice versa. For now, the answer remains one of the many mysteries surrounding microglia. The innate immune cells of SNr further distinguish themselves by exhibiting unique membrane properties compared to their immediate neighbors in the SNc or more distant relatives in the VTA. Consistent with their apparent higher metabolic rate and activational state, the microglia of the SNr have significantly more hyperpolarized resting potentials and larger membrane capacitance. Over 60% of them displayed voltage-activated currents upon depolarization that resembled the delayed rectifier potassium currents seen in early postnatal or injuryresponsive microglia. By contrast, membranes of microglia in the VTA and SNc 234 Neuron 95, July 19, 2017
passively responded to injection of hyperpolarizing or depolarizing currents. The tonically active high-frequency firing rate of SNr neurons may provide the juice that makes the microglia there so reactive, but that possibility remains to be determined. Lastly, there is gene expression. The gene expression profile of microglia changes as they mature (Hanamsagar et al., 2017), and multiple studies reviewed by De Biase et al. have reported regional heterogeneity in the mature brain. But there has always been the nagging question of the biological version of the Heisenberg uncertainty principle: by isolating the microglia in order to measure them, are you changing them in unanticipated ways? We will never know for sure, but De Biase and colleagues make a convincing case that the microdissection of subnuclei and subsequent cell sorting conducted under ice-cold conditions from individual animals (i.e., no pooling) followed by RNA sequencing (RNA-seq) or RT-PCR provided reliable profiles of region-specific gene expression by microglia. And what they found is interesting. First, there is a high degree of overlap in the microglial transcriptome of the cortex and NAc (84%), whereas the VTA had the least amount of overlap with any of
the other regions, in some cases approaching only 50%. They further found that a classic cohort of genes associate with microglia functioning (i.e., process dynamics, inflammation, and immune function) was highly conserved across all regions, whereas genes involved in mitochondrial function, oxidative stress, and glucose metabolism differed across regions, most notably in the VTA. Now that clear and convincing evidence for regional heterogeneity in number, morphology, activational state, and gene expression is established, the next question is how these striking regional differences come about. There are basically two possibilities. Either the microglia are the marionettes of the neuronal puppeteers, i.e., respond to local cell extrinsic cues, or they establish themselves as sovereign states following colonization of virgin territory, i.e., epigenetic programming early in life determines enduring properties later. Within the first week of life, De Biase et al. found no regional differences in microglia morphology, but less than 1 week later they had begun to emerge, creating the potential for epigenetic programming. However, when microglia were depleted in the mature adult, either chemically or genetically, the re-colonization began with homogeneity and gradually progressed to regional heterogeneity such that the original pattern of specificity was recapitulated, including across the very small divide of the SNr and the SNc. This highly local heterogeneity, with marked variance across a span of tens of microns, places the preponderance of evidence in favor of local cues. It appears that microglia respond to the stable of neurons they are charged with protecting and the level of vigilance that requires depends upon the nature of the horses in the stable. Future studies in which microglia from one region are isolated and transplanted into another will firmly establish the pecking order between neuron, astrocyte, and immune cell, or new data may upend the order all together, for these cells never cease to amaze. REFERENCES Cuadros, M.A., and Navascue´s, J. (1998). Prog. Neurobiol. 56, 173–189. De Biase, L.M., Schuebel, K.E., Fusfeld, Z.H., Jair, K., Hawes, I.A., Cimbro, R., Zhang, H.-Y.,
Neuron
Previews Liu, Q.-R., Shen, H., Xi, Z.-H., et al. (2017). Neuron 95, this issue, 341–356.
Published online June 15, 2017. http://dx.doi.org/ 10.1002/glia.23176.
Ginhoux, F., Lim, S., Hoeffel, G., Low, D., and Huber, T. (2013). Front. Cell. Neurosci. 7, 45.
Hong, S., Dissing-Olesen, L., and Stevens, B. (2016). Curr. Opin. Neurobiol. 36, 128–134.
Hanamsagar, R., Alter, M.D., Block, C.S., Sullivan, H., Bolton, J.L., and Bilbo, S.D. (2017). Glia.
Nayak, D., Roth, T.L., and McGavern, D.B. (2014). Annu. Rev. Immunol. 32, 367–402.
Nimmerjahn, A., Kirchhoff, F., and Helmchen, F. (2005). Science 308, 1314–1318. Sierra, A., Abiega, O., Shahraz, A., and Neumann, H. (2013). Front. Cell. Neurosci. 7, 6. Streit, W.J. (2000). Toxicol. Pathol. 28, 28–30.
Oxytocin Mobilizes Midbrain Dopamine toward Sociality Alexandre Charlet1,2,* and Valery Grinevich3,4,5,* 1Centre National de la Recherche Scientifique and University of Strasbourg, Institute of Cellular and Integrative Neurosciences, 67084 Strasbourg, France 2University of Strasbourg Institute for Advanced Study (USIAS), 67083 Strasbourg, France 3Schaller Research Group on Neuropeptides, German Cancer Research Center (DKFZ), 69120 Heidelberg, Germany 4CellNetwork Cluster of Excellence, University of Heidelberg, 69120 Heidelberg, Germany 5Central Institute of Mental Health, 68159 Mannheim, Germany *Correspondence: [email protected] (A.C.), [email protected] (V.G.) http://dx.doi.org/10.1016/j.neuron.2017.07.002
Oxytocin and dopamine possess significant overlap in the modulation of life-essential behaviors. Here, Xiao et al. (2017) show that the activity of dopamine neurons of the ventral tegmental area and the substantia nigra is finely tuned by axonal release of oxytocin. Oxytocin (OT) and dopamine (DA) systems are two neuromodulators of brain functions that exhibit similarity in organization and project sites. In the brain, OT neurons are exclusively present in the hypothalamus, while DA neurons are primarily found in the ventral tegmental area (VTA) and substantia nigra (SN). Interestingly, OT and DA neurons project to similar forebrain regions, including prefrontal cortex, nucleus accumbens, and striatum, to control social and affiliative behaviors, such as sexual behavior and pair bonding (Neumann, 2009). Given that DA receptors are present on OT neurons and OT receptors (OTRs) are detected in both VTA and SN, it is tempting to hypothesize that both systems are regulating each other. However, physiological relevance, anatomical connections, and intracellular signaling between OT and DA systems have not been dissected. In this issue of Neuron, Xiao et al. first used viral-based and latex bead-based retrograde tracing to identify two small subsets of parvocellular OT neurons separately projecting to the VTA or SN. Next,
the authors detected both OTR immunoreactivity and OTR mRNA in DA neurons of the VTA, and in GABAergic neurons of the SN. Finally, using ex vivo pharmacological and optogenetic approaches applied on acute midbrain slices, Xiao et al. showed that OT distinctly modulates DA neuronal populations: while OT seems to directly increase VTA neuron activity, this neuropeptide indirectly inhibits SN neurons via local recruitment of GABAergic interneurons (Figure 1). Hence, Xiao et al. provided a comprehensive set of data that bring a new level of complexity in the OT system and raise numerous basic questions. First, the authors demonstrated that pharmacological block of OTRs, and to a lesser extent vasopressin (VP) receptors (V1aRs), prevents the OT-induced increase in firing rate of DA neurons of the VTA. This OTR and V1aR dual action of OT is currently under debate as physiologically relevant high concentrations of OT seem to be capable of activating V1aR in several brain and spinal cord regions (Busnelli et al., 2012). This is particularly important
as OT and VP are often presented as antagonists in their function. It is particularly well described in the amygdala-mediated control of anxiety, where VP enhances fear, while OT exerts an anxiolytic function. Given that the VP system also projects to the VTA (Beier et al., 2015), the VP-induced modulation of the DA system may be as important as the OT one. Hence, the interplay between OT and VP in the regulation of the DA system has to be deciphered. Second, OT neurons express glutamate as the main neurotransmitter. While an evoked release of OT also results in a release of glutamate (Knobloch et al., 2012; Eliava et al., 2016), the physiological relevance of such a corelease is still unknown. Hence, the general function of glutamate and OT axonal co-release should be investigated, and particularly in the context of the proposed complex and dual DA system modulation. Finally, although the authors demonstrated the contribution of Gq-mediated pathway in excitatory effects of OT, other subunits such as Gi and Go can be recruited by the OTR (Busnelli et al., 2012)
Neuron 95, July 19, 2017 ª 2017 Elsevier Inc. 235
Previews Location, Location, Location: Microglia Are Where They Live Margaret M. McCarthy1,* 1Department of Pharmacology, University of Maryland School of Medicine, Baltimore, MD 21230, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.neuron.2017.07.005
A deep dive into microglia form and function reveals startling regional heterogeneity in number, morphology, activity, and transcriptomics in nuclei relevant to motor control and motivation (De Biase et al., 2017). Differences appear 2 weeks after birth, and depletion and recolonization in adulthood recapitulate the original phenotype, implicating local environment as mediators of microglial phenotype. A deep dive into microglia form and function reveals startling specific regional heterogeneity in terms of number, morphology, activity, and transcriptomics in deep brain nuclei with relevance to motor control and motivation and reward. Some regions, such as the substantia nigra reticulata (SNr), have markedly higher density of microglia overall and exist in a highly active state compared to closely related nuclei, either neuroanatomically or functionally, that have far fewer and simpler microglia. While the innate immune cells of the SNr are particularly metabolically and electrically active, those of the ventral tegmental area (VTA) show the most unique transcriptome. Depletion of microglia either chemically or genetically confirms that during recolonization the original phenotype is recapitulated, suggesting local environment is the driving force of microglial phenotype. Strong correlations between the density of astrocytes and microglia but a disconnect between both those populations and neuronal density raises multiple interesting questions about who is the master of the house. Collectively, these nuclei subserve some of the most important neurological functions of any organism, ranging from motor control to motivation and reward. Microglia are neither micro nor glia. Outsized in their impact, these innate immune cells of the brain resemble glia in appearance and partially in function but are unique influencers of the brain in both health and disease. The mysteries surrounding these cells abound, beginning with their surprising origins in the embryonic yolk sac, followed by migration into the brain before the blood-brain
barrier closes (Ginhoux et al., 2013; Nayak et al., 2014). Once safely ensconced in the CNS, they disperse and propagate until the tiling of the entire neuropil is complete. From there they grow and multiply in tandem with the developing brain and actively participate in sculpting the formation of neural circuits by regulating cell number and synaptic profiles (Hong et al., 2016). An essential feature of the brain is regional heterogeneity achieved via variance in cell number, density, phenotype, and connectivity. Neurons and glia (astrocytes and oligodendrocytes) are derived from multiple progenitor pools and diverge into hundreds of different, mostly post-mitotic phenotypes. By contrast, microglia are a unique mono-lineage derived from a macrophage precursor but found only in the CNS and are readily proliferative in response to a challenge such as injury (Cuadros and Navascue´s, 1998; Sierra et al., 2013; Streit, 2000). A natural assumption is that microglia are also mono-phenotypic, and if they do vary regionally, it is a transient byproduct of the environment in which they find themselves. But is this true? The paper by De Biase and colleagues in this issue directly addresses this question with in-depth analyses of three basic parameters: (1) density and morphology, (2) metabolic and activational state, and (3) transcriptomics of microglia of the basal ganglia, a collection of subcortical nuclei critical to motor control as well as motivation and reward (Figure 1). Regional comparison of microglia begins with simple questions: Is the density the same or different? Is the shape same or different? De Biase and colleagues
found both differed dramatically within the basal ganglia. The VTA is the source of dopamine neurons projecting to the nucleus accumbens (NAc), but it hosts on average 45% fewer microglia per unit area than its target. Both are dwarfed by the adjacent SNr, which has almost three times the density of microglia as the VTA. The dopamine neuron component of the SN, the pars compacta (SNc), resembles the VTA, another dopamine projection region. Intriguingly, the ratio of microglia to astrocytes remains constant across regions, suggesting either a common factor(s) determines their density or the existence of a hidden mutual co-dependency of these divergent yet functionally connected cells. The shape of microglia can be an indicator of function as these cells often physically contact others in order to influence them via highly localized release of signaling molecules, trophic factors, or phagocytosis of part or all of a cell. In the healthy adult basal ganglia, the microglia are ramified in appearance with long, thin, highly branched processes, much the same as elsewhere in the brain. The length and number of the processes are indirect indicators of the number of neurons potentially impacted by a given microglia. Here, too, regional variation was found, with microglia in the NAc having the longest processes compared to other basal ganglia nuclei, while those in the SNr were the most complexly branched. Morphology is important, but one of the more startling discoveries about microglia was their ongoing motility. Originally thought to be ‘‘quiescent,’’ we now know these cells are perpetual motion machines constantly ‘‘surveying’’ their
Neuron 95, July 19, 2017 ª 2017 Published by Elsevier Inc. 233
Neuron
Previews
Figure 1. Microglia Phenotype in the Basal Ganglia Comparison of microglia in terms of density, morphology, activational state, membrane properties, and transcriptomics revealed startling sharp regional heterogeneity in a group of functionally related nuclei comprising a portion of the basal ganglia. Some regions, such as the SNr, have markedly higher density of microglia overall, and they appear to be in a highly active state compared to closely related nuclei, either neuroanatomically or functionally, that have far fewer and simpler microglia.
environment and poking their processes into everybody else’s business (Nimmerjahn et al., 2005). Lysosome content is a measure of both phagocytic capacity and metabolic state, sort of a busy body indicator. The SNr is busy body central, with the most densely packed and metabolically active microglia in the basal ganglia. This is surprising given that the density of neurons here is the lowest of all the regions examined, raising the interesting question of whether neurons suppress microglia proliferation and/or survival, or vice versa. For now, the answer remains one of the many mysteries surrounding microglia. The innate immune cells of SNr further distinguish themselves by exhibiting unique membrane properties compared to their immediate neighbors in the SNc or more distant relatives in the VTA. Consistent with their apparent higher metabolic rate and activational state, the microglia of the SNr have significantly more hyperpolarized resting potentials and larger membrane capacitance. Over 60% of them displayed voltage-activated currents upon depolarization that resembled the delayed rectifier potassium currents seen in early postnatal or injuryresponsive microglia. By contrast, membranes of microglia in the VTA and SNc 234 Neuron 95, July 19, 2017
passively responded to injection of hyperpolarizing or depolarizing currents. The tonically active high-frequency firing rate of SNr neurons may provide the juice that makes the microglia there so reactive, but that possibility remains to be determined. Lastly, there is gene expression. The gene expression profile of microglia changes as they mature (Hanamsagar et al., 2017), and multiple studies reviewed by De Biase et al. have reported regional heterogeneity in the mature brain. But there has always been the nagging question of the biological version of the Heisenberg uncertainty principle: by isolating the microglia in order to measure them, are you changing them in unanticipated ways? We will never know for sure, but De Biase and colleagues make a convincing case that the microdissection of subnuclei and subsequent cell sorting conducted under ice-cold conditions from individual animals (i.e., no pooling) followed by RNA sequencing (RNA-seq) or RT-PCR provided reliable profiles of region-specific gene expression by microglia. And what they found is interesting. First, there is a high degree of overlap in the microglial transcriptome of the cortex and NAc (84%), whereas the VTA had the least amount of overlap with any of
the other regions, in some cases approaching only 50%. They further found that a classic cohort of genes associate with microglia functioning (i.e., process dynamics, inflammation, and immune function) was highly conserved across all regions, whereas genes involved in mitochondrial function, oxidative stress, and glucose metabolism differed across regions, most notably in the VTA. Now that clear and convincing evidence for regional heterogeneity in number, morphology, activational state, and gene expression is established, the next question is how these striking regional differences come about. There are basically two possibilities. Either the microglia are the marionettes of the neuronal puppeteers, i.e., respond to local cell extrinsic cues, or they establish themselves as sovereign states following colonization of virgin territory, i.e., epigenetic programming early in life determines enduring properties later. Within the first week of life, De Biase et al. found no regional differences in microglia morphology, but less than 1 week later they had begun to emerge, creating the potential for epigenetic programming. However, when microglia were depleted in the mature adult, either chemically or genetically, the re-colonization began with homogeneity and gradually progressed to regional heterogeneity such that the original pattern of specificity was recapitulated, including across the very small divide of the SNr and the SNc. This highly local heterogeneity, with marked variance across a span of tens of microns, places the preponderance of evidence in favor of local cues. It appears that microglia respond to the stable of neurons they are charged with protecting and the level of vigilance that requires depends upon the nature of the horses in the stable. Future studies in which microglia from one region are isolated and transplanted into another will firmly establish the pecking order between neuron, astrocyte, and immune cell, or new data may upend the order all together, for these cells never cease to amaze. REFERENCES Cuadros, M.A., and Navascue´s, J. (1998). Prog. Neurobiol. 56, 173–189. De Biase, L.M., Schuebel, K.E., Fusfeld, Z.H., Jair, K., Hawes, I.A., Cimbro, R., Zhang, H.-Y.,
Neuron
Previews Liu, Q.-R., Shen, H., Xi, Z.-H., et al. (2017). Neuron 95, this issue, 341–356.
Published online June 15, 2017. http://dx.doi.org/ 10.1002/glia.23176.
Ginhoux, F., Lim, S., Hoeffel, G., Low, D., and Huber, T. (2013). Front. Cell. Neurosci. 7, 45.
Hong, S., Dissing-Olesen, L., and Stevens, B. (2016). Curr. Opin. Neurobiol. 36, 128–134.
Hanamsagar, R., Alter, M.D., Block, C.S., Sullivan, H., Bolton, J.L., and Bilbo, S.D. (2017). Glia.
Nayak, D., Roth, T.L., and McGavern, D.B. (2014). Annu. Rev. Immunol. 32, 367–402.
Nimmerjahn, A., Kirchhoff, F., and Helmchen, F. (2005). Science 308, 1314–1318. Sierra, A., Abiega, O., Shahraz, A., and Neumann, H. (2013). Front. Cell. Neurosci. 7, 6. Streit, W.J. (2000). Toxicol. Pathol. 28, 28–30.
Oxytocin Mobilizes Midbrain Dopamine toward Sociality Alexandre Charlet1,2,* and Valery Grinevich3,4,5,* 1Centre National de la Recherche Scientifique and University of Strasbourg, Institute of Cellular and Integrative Neurosciences, 67084 Strasbourg, France 2University of Strasbourg Institute for Advanced Study (USIAS), 67083 Strasbourg, France 3Schaller Research Group on Neuropeptides, German Cancer Research Center (DKFZ), 69120 Heidelberg, Germany 4CellNetwork Cluster of Excellence, University of Heidelberg, 69120 Heidelberg, Germany 5Central Institute of Mental Health, 68159 Mannheim, Germany *Correspondence: [email protected] (A.C.), [email protected] (V.G.) http://dx.doi.org/10.1016/j.neuron.2017.07.002
Oxytocin and dopamine possess significant overlap in the modulation of life-essential behaviors. Here, Xiao et al. (2017) show that the activity of dopamine neurons of the ventral tegmental area and the substantia nigra is finely tuned by axonal release of oxytocin. Oxytocin (OT) and dopamine (DA) systems are two neuromodulators of brain functions that exhibit similarity in organization and project sites. In the brain, OT neurons are exclusively present in the hypothalamus, while DA neurons are primarily found in the ventral tegmental area (VTA) and substantia nigra (SN). Interestingly, OT and DA neurons project to similar forebrain regions, including prefrontal cortex, nucleus accumbens, and striatum, to control social and affiliative behaviors, such as sexual behavior and pair bonding (Neumann, 2009). Given that DA receptors are present on OT neurons and OT receptors (OTRs) are detected in both VTA and SN, it is tempting to hypothesize that both systems are regulating each other. However, physiological relevance, anatomical connections, and intracellular signaling between OT and DA systems have not been dissected. In this issue of Neuron, Xiao et al. first used viral-based and latex bead-based retrograde tracing to identify two small subsets of parvocellular OT neurons separately projecting to the VTA or SN. Next,
the authors detected both OTR immunoreactivity and OTR mRNA in DA neurons of the VTA, and in GABAergic neurons of the SN. Finally, using ex vivo pharmacological and optogenetic approaches applied on acute midbrain slices, Xiao et al. showed that OT distinctly modulates DA neuronal populations: while OT seems to directly increase VTA neuron activity, this neuropeptide indirectly inhibits SN neurons via local recruitment of GABAergic interneurons (Figure 1). Hence, Xiao et al. provided a comprehensive set of data that bring a new level of complexity in the OT system and raise numerous basic questions. First, the authors demonstrated that pharmacological block of OTRs, and to a lesser extent vasopressin (VP) receptors (V1aRs), prevents the OT-induced increase in firing rate of DA neurons of the VTA. This OTR and V1aR dual action of OT is currently under debate as physiologically relevant high concentrations of OT seem to be capable of activating V1aR in several brain and spinal cord regions (Busnelli et al., 2012). This is particularly important
as OT and VP are often presented as antagonists in their function. It is particularly well described in the amygdala-mediated control of anxiety, where VP enhances fear, while OT exerts an anxiolytic function. Given that the VP system also projects to the VTA (Beier et al., 2015), the VP-induced modulation of the DA system may be as important as the OT one. Hence, the interplay between OT and VP in the regulation of the DA system has to be deciphered. Second, OT neurons express glutamate as the main neurotransmitter. While an evoked release of OT also results in a release of glutamate (Knobloch et al., 2012; Eliava et al., 2016), the physiological relevance of such a corelease is still unknown. Hence, the general function of glutamate and OT axonal co-release should be investigated, and particularly in the context of the proposed complex and dual DA system modulation. Finally, although the authors demonstrated the contribution of Gq-mediated pathway in excitatory effects of OT, other subunits such as Gi and Go can be recruited by the OTR (Busnelli et al., 2012)
Neuron 95, July 19, 2017 ª 2017 Elsevier Inc. 235