- Email: [email protected]
Perception without a Thalamus
Neuron 166
the formation of new axon extensions. An intriguing possibility is that the exocyst might provide another link between guidance receptors, small GTPases, and the downstream changes in growth cone shape that ultimately determine the morphology and connectivity of a neuron. Thomas R. Clandinin Department of Neurobiology 299 West Campus Drive Stanford University Stanford, California 94305
Selected Reading Beronja, S., Laprise, P., Papoulas, O., Pellikka, M., Sisson, J., and Tepaass, U. (2005). Essential function of Drosophila Sec6 in apical exocytosis of epithelial photoreceptor cells. J. Cell Biol., in press. Clandinin, T.R., and Zipursky, S.L. (2002). Making connections in the fly visual system. Neuron 35, 827–841. Grindstaff, K.K., Yeaman, C., Anandasabapathy, N., Hsu, S.C., Rodriguez-Boulan, E., Scheller, R.H., and Nelson, W.J. (1998). Sec6/8 complex is recruited to cell-cell contacts and specifies transport vesicle delivery to the basal-lateral membrane in epithelial cells. Cell 93, 731–740. Hsu, S.C., Hazuka, C.D., Foletti, D.L., and Scheller, R.H. (1999). Targeting vesicles to specific sites on the plasma membrane: the role of the Sec6/8 complex. Trends Cell Biol. 9, 150–153. Mehta, S.Q., Hiesinger, P.R., Beronja, S., Zhai, R.G., Schulze, K.L., Verstreken, P., Cao, Y., Zhou, Y., Tepass, U., Crair, M.C., and Bellen, H.J. (2005). Mutations in Drosophila sec15 reveal a function in neuronal targeting for a subset of exocyst components. Neuron 46, this issue, 219–232. Moskalenko, S., Henry, D.O., Rosse, C., Mirey, G., Camonis, J.H., and White, M.A. (2002). The exocyst is a Ral effector complex. Nat. Cell Biol. 4, 66–72. Murthy, M., and Schwarz, T.L. (2004). The exocyst component Sec5 is required for membrane traffic and polarity in the Drosophila ovary. Development 131, 377–388. Murthy, M., Garza, D., Scheller, R.H., and Schwarz, T.L. (2003). Mutations in the exocyst component Sec5 disrupt neuronal membrane traffic, but neurotransmitter release persists. Neuron 37, 433–447. Murthy, M., Ranjan, R., Denef, N., Higashi, M.E., Schupbach, T., and Schwarz, T.L. (2005). Sec6 mutations and the Drosophila exocyst complex. J. Cell Sci. 15, 1139–1150. Prakash, S., Caldwell, J.C., Eberl, D.F., and Clandinin, T.R. (2005). Drosophila N-cadherin mediates an attractive interaction between photoreceptor axons and their targets. Nat. Neurosci. 8, 443–450. Satoh, A.K., O’tousa, J.E., Ozaki, K., and Ready, D.F. (2005). Rab11 mediates post-Golgi trafficking of rhodopsin to the photosensitive apical membrane of Drosophila photoreceptors. Development 132, 1487–1497. Shipitsin, M., and Feig, L.A. (2004). RalA but not RalB enhances polarized delivery of membrane proteins to the basolateral surface of epithelial cells. Mol. Cell. Biol. 24, 5746–5756. Sugihara, K., Asano, S., Tanaka, K., Iwamatsu, A., Okawa, K., and Ohta, Y. (2002). The exocyst complex binds the small GTPase RalA to mediate filopodia formation. Nat. Cell Biol. 4, 73–78. Yeaman, C., Grindstaff, K.K., and Nelson, W.J. (2004). Mechanism of recruiting Sec6/8 (exocyst) complex to the apical junctional complex during polarization of epithelial cells. J. Cell Sci. 117, 559–570. Zhang, X.M., Ellis, S., Sriratana, A., Mitchell, C.A., and Rowe, T. (2004). Sec15 is an effector for the Rab11 GTPase in mammalian cells. J. Biol. Chem. 279, 43027–43034. DOI 10.1016/j.neuron.2005.04.003
Perception without a Thalamus: How Does Olfaction Do It? The olfactory system has generated considerable interest in recent years, mainly focused on receptor genes and early olfactory processing. In this issue of Neuron, Mori et al. focus centrally, providing evidence for slow- and fast-wave states in olfactory cortex that appear to gate the inflow of information underlying conscious smell perception. A consensus has been emerging in recent years on the steps involved in what may be called early olfaction (Wilson and Stevenson, 2003; Shepherd, 2005). These begin with the combinatorial transduction of odor molecules by a large family of olfactory receptors (Buck and Axel, 1991; Malnic et al., 1999); conversion of those responses into odor maps (“odor images”) in the glomerular layer of the olfactory bulb (summarized in Xu et al., 2000); extraction of features of the odor maps by synaptic microcircuits in the bulb (Mori and Yoshihara, 1995); and processing of the maps into a contentaddressable memory representation in the olfactory cortex (Haberly, 2001; Wilson and Stevenson, 2003). This combination of evidence represents a tremendous advance for the field, putting our understanding of early olfaction on par with early vision (Tsodyks and Gilbert, 2004) and initial processing in other sensory systems. At the olfactory cortex, however, we reach an impasse. There is a common assumption, explicit or implicit, that conscious perception of smell may arise in this three-layered cortex. However, in other sensory systems, conscious perception depends on a pathway to the level of the neocortex, and in those systems this requires going through thalamus (see, for example, Pinault, 2004). For many years it was thought that the olfactory pathway also passes through the thalamus, from olfactory cortex through the mediodorsal thalamic nucleus to prefrontal cortex. However, recent careful anatomical studies have shown that the pathway between olfactory cortex and prefrontal cortex is mostly direct, with only a small contingent of fibers going to mediodorsal thalamus (Ongur and Price, 2000). Within prefrontal cortex, the primary olfactory area consists of the medial and lateral orbitofrontal cortex. This mainly direct pathway to the neocortex, for the most part bypassing the thalamus, raises a host of questions regarding the neural substrate for conscious smell perception. Where does conscious perception arise? At the level of the olfactory cortex or orbitofrontal cortex? How does activity in the olfactory pathway relate to the alternating levels of activity between waking and deep sleep that are found in all other systems? How does synchronization of activity between olfactory and nonolfactory systems occur? How can conscious perception of odors arise without the participation of the thalamus? If olfaction does not require a thalamic relay, what does this tell us about the presumably critical role that the thalamus plays for the conscious state in other sensory systems? In this issue of Neuron, Murakami et al. (2005) have taken the first step toward answering these questions
Previews 167
by asking: is there evidence for waking and deep sleep states in the cellular activity of the olfactory cortex? They carried out experiments in the urethane-anesthetized rat, in which the cortical EEG showed the wellknown spontaneous alternations between a fast-wave state (FWS) and slow-wave state (SWS). Single-cell recordings from the olfactory cortex showed vigorous spike discharge responses to odors during FWS but not SWS. This indicated that the flow of activity through the olfactory pathway was gated in relation to behavioral state in a manner similar to other sensory systems (Steriade and Llinas, 1988). The authors carried out further experiments to document this finding. The gating applied across the odors tested, and thus was not odor specific. It was found with both natural and artifical respiration. It was particularly prominent in the olfactory cortex, including the anterior pyriform area and the olfactory tubercle, but was largely absent from the olfactory bulb; the small degree of gating found there may reflect the action of the long association fibers in the cortex recurring to the olfactory bulb. Membrane mechanisms were analyzed with intracellular recordings, which showed that during SWS the membrane potential oscillated between up (depolarized) and down (hyperpolarized) states, changing to an up, depolarized state during FWS. The olfactory SWS oscillations were synchronized with the generalized SWS oscillations of the cortical EEG. Electrical stimulation of the olfactory bulb evoked excitatory postsynaptic potentials (EPSPs) in olfactory cortical neurons. An intriguing finding was that this EPSP is larger during the hyperpolarizing phase, but does not reach spike threshold because of the hyperpolarizing shift. This suggests that gating does not block the input from the olfactory bulb or the EPSP response, but acts through a mechanism of disfacilitation similar to that shown in neocortical neurons during SWS. Further experiments will be needed to test for this mechanism. How is gating in the olfactory pathway coordinated with gating in other sensory systems? To test for this, the authors carried out electrical stimulation of the brainstem interpeduncular nucleus during SWS to mimick the action of the reticular activating system that is known to underlie the FWS. By this route they converted the cortical EEG from SWS to FWS and concurrently changed a weak odor response to a strong one. This suggested that the gating control originates in the brainstem ascending reticular formation and broadly affects all cortical areas, including the olfactory areas, in synchrony with thalamic gating of the other systems. Like any pioneering study, this report only scratches the surface. Other systems that may contribute to the synchronizing action with the rest of the brain are the various transmitter systems that project widely throughout the cortex, including serotonergic and noradrenergic projections from the brainstem, and cholinergic projections from the basal forebrain. The most pressing need now is to understand the processing step from olfactory cortex to orbitofrontal cortex at the neocortical level. We've arrived at the gate, but what lies beyond? The first functional study to address this problem was carried out many years ago by Mori's mentor, Sadayuki Takagi. In a tour de force, he and his colleagues (Tanabe et al., 1975) made
single-cell recordings in the monkey and showed that there is a progressive sharpening of the response spectrum from olfactory bulb through olfactory cortex to orbitofrontal cortex, reflecting a type of feature extraction at the highest cortical level. In awake behaving monkeys, most neurons in the olfactory region of the orbitofrontal cortex decrease their responses to an odor of a food to which the monkey is fed to satiety (Critchley and Rolls, 1996), indicating that these neurons encode the reward value and relative pleasantness or unpleasantness of a stimulus within its behavioral context. This property, however, is not exclusive to orbitofrontal cortex; it has also been seen in recordings from mitral cells in the rat olfactory bulb (Pager, 1974) and in some neurons in olfactory cortex (Schoenbaum and Eichenbaum, 1995). It appears that behavioral context is communicated to multiple levels of the olfactory pathway. These multiple levels apparently bridge across the gating between regions such as olfactory cortex and orbitofrontal cortex. The fact that olfactory processing begins in the olfactory bulb with “odor images” analogous to visual images (Haberly, 1985; Haberly, 2001; Shepherd, 2005; Shepherd, 1991; Wilson and Stevenson, 2003) suggests possible parallels with central vision. It will be important to identify exactly what kind of higher level of processing of these images takes place in orbitofrontal cortex compared to olfactory cortex, and how it relates to specific psychophysical attributes of conscious smell perception. As these experiments are undertaken, the knowledge that the olfactory pathway is subject to gating of sensory inflow similar to that which occurs in other sensory systems will be critical to planning the strategy and interpreting the results. Gordon M. Shepherd Department of Neurobiology Yale University School of Medicine 333 Cedar Street New Haven, Connecticut 06510
Selected Reading Buck, L.B., and Axel, R. (1991). Cell 65, 175–189. Critchley, H.D., and Rolls, E.T. (1996). J. Neurophysiol. 75, 1673– 1686. Haberly, L.B. (1985). Chem. Senses 10, 219–238. Haberly, L.B. (2001). Chem. Senses 26, 551–576. Malnic, B., Hrono, J., Sato, T., and Buck, L.B. (1999). Cell 96, 713– 723. Mori, K., and Yoshihara, Y. (1995). Prog. Neurobiol. 45, 585–619. Murakami, M., Kashiwadani, H., Kirino, Y., and Mori, K. (2005). Neuron 46, this issue, 285–296. Ongur, D., and Price, J.L. (2000). Cereb. Cortex 10, 206–219. Pager, J. (1974). Physiol. Behav. 12, 189–195. Pinault, D. (2004). Brain Res. Brain Res. Rev. 46, 1–31. Schoenbaum, G., and Eichenbaum, H. (1995). J. Neurophysiol. 74, 733–750. Shepherd, G.M. (1991). In Olfaction: A Model for Computational Neuroscience, H. Eichenbaum and J. Davis, eds. (Cambridge, MA: MIT Press), pp. 3–42. Shepherd, G.M. (2005). Chem. Senses 30, i3–i5.
Neuron 168
Steriade, M., and Llinas, R.R. (1988). Physiol. Rev. 68, 649–742. Tanabe, T., Iino, M., and Takagi, S.F. (1975). J. Neurophysiol. 38, 1284–1296. Tsodyks, M., and Gilbert, C. (2004). Nature 431, 775–781. Wilson, D.A., and Stevenson, R.J. (2003). Neurosci. Biobehav. Rev. 27, 307–328. Xu, F., Greer, C.A., and Shepherd, G.M. (2000). J. Comp. Neurol. 422, 489–495. DOI 10.1016/j.neuron.2005.03.012
the formation of new axon extensions. An intriguing possibility is that the exocyst might provide another link between guidance receptors, small GTPases, and the downstream changes in growth cone shape that ultimately determine the morphology and connectivity of a neuron. Thomas R. Clandinin Department of Neurobiology 299 West Campus Drive Stanford University Stanford, California 94305
Selected Reading Beronja, S., Laprise, P., Papoulas, O., Pellikka, M., Sisson, J., and Tepaass, U. (2005). Essential function of Drosophila Sec6 in apical exocytosis of epithelial photoreceptor cells. J. Cell Biol., in press. Clandinin, T.R., and Zipursky, S.L. (2002). Making connections in the fly visual system. Neuron 35, 827–841. Grindstaff, K.K., Yeaman, C., Anandasabapathy, N., Hsu, S.C., Rodriguez-Boulan, E., Scheller, R.H., and Nelson, W.J. (1998). Sec6/8 complex is recruited to cell-cell contacts and specifies transport vesicle delivery to the basal-lateral membrane in epithelial cells. Cell 93, 731–740. Hsu, S.C., Hazuka, C.D., Foletti, D.L., and Scheller, R.H. (1999). Targeting vesicles to specific sites on the plasma membrane: the role of the Sec6/8 complex. Trends Cell Biol. 9, 150–153. Mehta, S.Q., Hiesinger, P.R., Beronja, S., Zhai, R.G., Schulze, K.L., Verstreken, P., Cao, Y., Zhou, Y., Tepass, U., Crair, M.C., and Bellen, H.J. (2005). Mutations in Drosophila sec15 reveal a function in neuronal targeting for a subset of exocyst components. Neuron 46, this issue, 219–232. Moskalenko, S., Henry, D.O., Rosse, C., Mirey, G., Camonis, J.H., and White, M.A. (2002). The exocyst is a Ral effector complex. Nat. Cell Biol. 4, 66–72. Murthy, M., and Schwarz, T.L. (2004). The exocyst component Sec5 is required for membrane traffic and polarity in the Drosophila ovary. Development 131, 377–388. Murthy, M., Garza, D., Scheller, R.H., and Schwarz, T.L. (2003). Mutations in the exocyst component Sec5 disrupt neuronal membrane traffic, but neurotransmitter release persists. Neuron 37, 433–447. Murthy, M., Ranjan, R., Denef, N., Higashi, M.E., Schupbach, T., and Schwarz, T.L. (2005). Sec6 mutations and the Drosophila exocyst complex. J. Cell Sci. 15, 1139–1150. Prakash, S., Caldwell, J.C., Eberl, D.F., and Clandinin, T.R. (2005). Drosophila N-cadherin mediates an attractive interaction between photoreceptor axons and their targets. Nat. Neurosci. 8, 443–450. Satoh, A.K., O’tousa, J.E., Ozaki, K., and Ready, D.F. (2005). Rab11 mediates post-Golgi trafficking of rhodopsin to the photosensitive apical membrane of Drosophila photoreceptors. Development 132, 1487–1497. Shipitsin, M., and Feig, L.A. (2004). RalA but not RalB enhances polarized delivery of membrane proteins to the basolateral surface of epithelial cells. Mol. Cell. Biol. 24, 5746–5756. Sugihara, K., Asano, S., Tanaka, K., Iwamatsu, A., Okawa, K., and Ohta, Y. (2002). The exocyst complex binds the small GTPase RalA to mediate filopodia formation. Nat. Cell Biol. 4, 73–78. Yeaman, C., Grindstaff, K.K., and Nelson, W.J. (2004). Mechanism of recruiting Sec6/8 (exocyst) complex to the apical junctional complex during polarization of epithelial cells. J. Cell Sci. 117, 559–570. Zhang, X.M., Ellis, S., Sriratana, A., Mitchell, C.A., and Rowe, T. (2004). Sec15 is an effector for the Rab11 GTPase in mammalian cells. J. Biol. Chem. 279, 43027–43034. DOI 10.1016/j.neuron.2005.04.003
Perception without a Thalamus: How Does Olfaction Do It? The olfactory system has generated considerable interest in recent years, mainly focused on receptor genes and early olfactory processing. In this issue of Neuron, Mori et al. focus centrally, providing evidence for slow- and fast-wave states in olfactory cortex that appear to gate the inflow of information underlying conscious smell perception. A consensus has been emerging in recent years on the steps involved in what may be called early olfaction (Wilson and Stevenson, 2003; Shepherd, 2005). These begin with the combinatorial transduction of odor molecules by a large family of olfactory receptors (Buck and Axel, 1991; Malnic et al., 1999); conversion of those responses into odor maps (“odor images”) in the glomerular layer of the olfactory bulb (summarized in Xu et al., 2000); extraction of features of the odor maps by synaptic microcircuits in the bulb (Mori and Yoshihara, 1995); and processing of the maps into a contentaddressable memory representation in the olfactory cortex (Haberly, 2001; Wilson and Stevenson, 2003). This combination of evidence represents a tremendous advance for the field, putting our understanding of early olfaction on par with early vision (Tsodyks and Gilbert, 2004) and initial processing in other sensory systems. At the olfactory cortex, however, we reach an impasse. There is a common assumption, explicit or implicit, that conscious perception of smell may arise in this three-layered cortex. However, in other sensory systems, conscious perception depends on a pathway to the level of the neocortex, and in those systems this requires going through thalamus (see, for example, Pinault, 2004). For many years it was thought that the olfactory pathway also passes through the thalamus, from olfactory cortex through the mediodorsal thalamic nucleus to prefrontal cortex. However, recent careful anatomical studies have shown that the pathway between olfactory cortex and prefrontal cortex is mostly direct, with only a small contingent of fibers going to mediodorsal thalamus (Ongur and Price, 2000). Within prefrontal cortex, the primary olfactory area consists of the medial and lateral orbitofrontal cortex. This mainly direct pathway to the neocortex, for the most part bypassing the thalamus, raises a host of questions regarding the neural substrate for conscious smell perception. Where does conscious perception arise? At the level of the olfactory cortex or orbitofrontal cortex? How does activity in the olfactory pathway relate to the alternating levels of activity between waking and deep sleep that are found in all other systems? How does synchronization of activity between olfactory and nonolfactory systems occur? How can conscious perception of odors arise without the participation of the thalamus? If olfaction does not require a thalamic relay, what does this tell us about the presumably critical role that the thalamus plays for the conscious state in other sensory systems? In this issue of Neuron, Murakami et al. (2005) have taken the first step toward answering these questions
Previews 167
by asking: is there evidence for waking and deep sleep states in the cellular activity of the olfactory cortex? They carried out experiments in the urethane-anesthetized rat, in which the cortical EEG showed the wellknown spontaneous alternations between a fast-wave state (FWS) and slow-wave state (SWS). Single-cell recordings from the olfactory cortex showed vigorous spike discharge responses to odors during FWS but not SWS. This indicated that the flow of activity through the olfactory pathway was gated in relation to behavioral state in a manner similar to other sensory systems (Steriade and Llinas, 1988). The authors carried out further experiments to document this finding. The gating applied across the odors tested, and thus was not odor specific. It was found with both natural and artifical respiration. It was particularly prominent in the olfactory cortex, including the anterior pyriform area and the olfactory tubercle, but was largely absent from the olfactory bulb; the small degree of gating found there may reflect the action of the long association fibers in the cortex recurring to the olfactory bulb. Membrane mechanisms were analyzed with intracellular recordings, which showed that during SWS the membrane potential oscillated between up (depolarized) and down (hyperpolarized) states, changing to an up, depolarized state during FWS. The olfactory SWS oscillations were synchronized with the generalized SWS oscillations of the cortical EEG. Electrical stimulation of the olfactory bulb evoked excitatory postsynaptic potentials (EPSPs) in olfactory cortical neurons. An intriguing finding was that this EPSP is larger during the hyperpolarizing phase, but does not reach spike threshold because of the hyperpolarizing shift. This suggests that gating does not block the input from the olfactory bulb or the EPSP response, but acts through a mechanism of disfacilitation similar to that shown in neocortical neurons during SWS. Further experiments will be needed to test for this mechanism. How is gating in the olfactory pathway coordinated with gating in other sensory systems? To test for this, the authors carried out electrical stimulation of the brainstem interpeduncular nucleus during SWS to mimick the action of the reticular activating system that is known to underlie the FWS. By this route they converted the cortical EEG from SWS to FWS and concurrently changed a weak odor response to a strong one. This suggested that the gating control originates in the brainstem ascending reticular formation and broadly affects all cortical areas, including the olfactory areas, in synchrony with thalamic gating of the other systems. Like any pioneering study, this report only scratches the surface. Other systems that may contribute to the synchronizing action with the rest of the brain are the various transmitter systems that project widely throughout the cortex, including serotonergic and noradrenergic projections from the brainstem, and cholinergic projections from the basal forebrain. The most pressing need now is to understand the processing step from olfactory cortex to orbitofrontal cortex at the neocortical level. We've arrived at the gate, but what lies beyond? The first functional study to address this problem was carried out many years ago by Mori's mentor, Sadayuki Takagi. In a tour de force, he and his colleagues (Tanabe et al., 1975) made
single-cell recordings in the monkey and showed that there is a progressive sharpening of the response spectrum from olfactory bulb through olfactory cortex to orbitofrontal cortex, reflecting a type of feature extraction at the highest cortical level. In awake behaving monkeys, most neurons in the olfactory region of the orbitofrontal cortex decrease their responses to an odor of a food to which the monkey is fed to satiety (Critchley and Rolls, 1996), indicating that these neurons encode the reward value and relative pleasantness or unpleasantness of a stimulus within its behavioral context. This property, however, is not exclusive to orbitofrontal cortex; it has also been seen in recordings from mitral cells in the rat olfactory bulb (Pager, 1974) and in some neurons in olfactory cortex (Schoenbaum and Eichenbaum, 1995). It appears that behavioral context is communicated to multiple levels of the olfactory pathway. These multiple levels apparently bridge across the gating between regions such as olfactory cortex and orbitofrontal cortex. The fact that olfactory processing begins in the olfactory bulb with “odor images” analogous to visual images (Haberly, 1985; Haberly, 2001; Shepherd, 2005; Shepherd, 1991; Wilson and Stevenson, 2003) suggests possible parallels with central vision. It will be important to identify exactly what kind of higher level of processing of these images takes place in orbitofrontal cortex compared to olfactory cortex, and how it relates to specific psychophysical attributes of conscious smell perception. As these experiments are undertaken, the knowledge that the olfactory pathway is subject to gating of sensory inflow similar to that which occurs in other sensory systems will be critical to planning the strategy and interpreting the results. Gordon M. Shepherd Department of Neurobiology Yale University School of Medicine 333 Cedar Street New Haven, Connecticut 06510
Selected Reading Buck, L.B., and Axel, R. (1991). Cell 65, 175–189. Critchley, H.D., and Rolls, E.T. (1996). J. Neurophysiol. 75, 1673– 1686. Haberly, L.B. (1985). Chem. Senses 10, 219–238. Haberly, L.B. (2001). Chem. Senses 26, 551–576. Malnic, B., Hrono, J., Sato, T., and Buck, L.B. (1999). Cell 96, 713– 723. Mori, K., and Yoshihara, Y. (1995). Prog. Neurobiol. 45, 585–619. Murakami, M., Kashiwadani, H., Kirino, Y., and Mori, K. (2005). Neuron 46, this issue, 285–296. Ongur, D., and Price, J.L. (2000). Cereb. Cortex 10, 206–219. Pager, J. (1974). Physiol. Behav. 12, 189–195. Pinault, D. (2004). Brain Res. Brain Res. Rev. 46, 1–31. Schoenbaum, G., and Eichenbaum, H. (1995). J. Neurophysiol. 74, 733–750. Shepherd, G.M. (1991). In Olfaction: A Model for Computational Neuroscience, H. Eichenbaum and J. Davis, eds. (Cambridge, MA: MIT Press), pp. 3–42. Shepherd, G.M. (2005). Chem. Senses 30, i3–i5.
Neuron 168
Steriade, M., and Llinas, R.R. (1988). Physiol. Rev. 68, 649–742. Tanabe, T., Iino, M., and Takagi, S.F. (1975). J. Neurophysiol. 38, 1284–1296. Tsodyks, M., and Gilbert, C. (2004). Nature 431, 775–781. Wilson, D.A., and Stevenson, R.J. (2003). Neurosci. Biobehav. Rev. 27, 307–328. Xu, F., Greer, C.A., and Shepherd, G.M. (2000). J. Comp. Neurol. 422, 489–495. DOI 10.1016/j.neuron.2005.03.012