Seminars in Cell & Developmental Biology 35 (2014) 165–172

Contents lists available at ScienceDirect

Seminars in Cell & Developmental Biology journal homepage: www.elsevier.com/locate/semcdb

Review

Distinct roles of homeoproteins in brain topographic mapping and in neural circuit formation Olivier Stettler a,∗ , Kenneth L. Moya b,c a b c

Laboratoire CRRET EAC 7149, Université Paris-Est Créteil, 61, Av. du Général de Gaulle, 94010 Créteil Cedex, France Collège de France, Center for Interdisciplinary Research in Biology, UMR CNRS 7241/INSERM U1050, 11 place Marcelin Berthelot, 75005 Paris, France Labex Memolife, PSL Research University, France

a r t i c l e

i n f o

Article history: Available online 18 July 2014 Keywords: Homeoproteins Cortical regionalisation Retino-tectal Barrel field Otx2 Engrailed

a b s t r a c t The construction of the brain is a highly regulated process, requiring coordination of various cellular and molecular mechanisms that together ensure the stability of the cerebrum architecture and functions. The mature brain is an organ that performs complex computational operations using specific sensory information from the outside world and this requires precise organization within sensory networks and a separation of sensory modalities during development. We review here the role of homeoproteins in the arealization of the brain according to sensorimotor functions, the micropartition of its cytoarchitecture, and the maturation of its sensory circuitry. One of the most interesting observation about homeoproteins in recent years concerns their ability to act both in a cell-autonomous and non-cell-autonomous manner. The highlights in the present review collectively show how these two modes of action of homeoproteins confer various functions in shaping cortical maps. © 2014 Elsevier Ltd. All rights reserved.

Contents 1. 2. 3. 4. 5.

Introduction: canonical and non-canonical functions of homeoproteins during building of the brain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Homeoproteins define neocortical territories and neural circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Signaling with Engrailed during retino-tectal mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Otx2 stabilizes ocular dominance in the visual cortex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding remarks and perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction: canonical and non-canonical functions of homeoproteins during building of the brain The formation of functional brain areas is a process that starts with the regional specification of the neuroepithelium, followed by differentiation of neuronal types according to their lineage, and finally by the establishment of precise connections between areas that are functionally linked [1–4]. Homeoproteins (HPs) have long been known as key molecular determinants capable of specifying distinct embryonic territories during early body development [5–14]. This is exemplified in the nervous system where the HPs function as transcription factors that can early (i.e. before

∗ Corresponding author at: EAC 7149, Laboratoire CRRET, 61, Av. du Général de Gaulle, 94010 Créteil Cedex, France. Tel.: +33 1 45 17 18 13. E-mail address: [email protected] (O. Stettler). http://dx.doi.org/10.1016/j.semcdb.2014.07.004 1084-9521/© 2014 Elsevier Ltd. All rights reserved.

165 166 168 169 170 171

embryonic day 12 in the mouse) control specific cell differentiation programs and the size and physiological fate of brain areas [3,5,7,15–20] (see Section 2). However, HPs can also have later effects during the course of the development of the brain (i.e. at late embryonic and perinatal stages in the mouse) by controlling the proper formation and the stabilization of neuronal connections [20–28]. For instance and as shown below (Section 2), the HP Lhx2 drives the precise connections of thalamocortical axons within the somatosensory barrel field cortex [20]. Two other recent examples of HPs that participate in neuronal circuits development are provided by Engrailed which guides retinal axons and regulates the retinocollicular map formation [21,24,25] (Section 3), and by Otx2 which stabilizes the connections from the two eyes in the binocular visual cortex [26,27] (Section 4). A number of data now support the notion that, in a developmental context, HPs can function in two different ways, as gene transcription factors, and less classically, as protein

166

O. Stettler, K.L. Moya / Seminars in Cell & Developmental Biology 35 (2014) 165–172

translational modulators [29,30]. Homeoprotein transcription factors bind to DNA through a highly conserved structure, called the homeodomain, which is 60 amino acids in length and is structured in three alpha helices [30]. Additionally, interaction of the eukaryotic translation initiation factor eIF4E, with one or another of the putative eIF4E-binding sites flanking the homeodomain is thought to be responsible for HP translational activities [21,29,31]. Translational activity was first demonstrated for bicoid which regulates the translation of caudal mRNA from the anterior pole of the fly embryo through eIF4E binding [32]. The canonical eIF4E binding motif (YXXXXLФ)[33] was subsequently found in the homeodomain protein PRH that inhibits eIF4E-dependent mRNA transport, and in 199 other HPs, which thus potentially act as regulators of eIF4E-dependent mRNA translation. In addition to bicoid and PRH, a direct interaction with eIF4E has so far been demonstrated for the HPs HoxA9, Emx2, OTX2 and Engrailed-1 and Engrailed-2 [24,29]. Another non-classical trait of HPs of potential importance, in particular during late development of the brain, is their capability to act as signaling molecules between cells. Thus far, all HP that have been tested for their signaling property (i.e. 10 of more than 200) [30] have been shown to pass from cell to cell, allowing direct access to the cytosol and eventually the nucleus of recipient cell [30]. Most neurodevelopmental studies related to HPs have nonetheless focused on their classical cell autonomous functions; the possibility that HPs act both at a cell autonomous and non-cell autonomous level in a same developmental process is not excluded. A future challenge will be to examine or re-examine how these two HPs modus operandi are coordinated during brain development (Section 5). This review aims to highlight recent data that extend our knowledge of the role of these proteins, regardless of their mode of action, as organizers of neural circuitry and neural maps. In order to focus our discussion on brain areas where both cell autonomous and non-cell autonomous functions of HPs have been described, the present review is limited to the role of HPs in the forebrain and midbrain regions. Consequently, the cerebellum for which the HPs Engrailed-1 and -2 act nonetheless as master transcriptional regulators of the patterning of gene expression and of afferent topography, is considered elsewhere [23,34]. 2. Homeoproteins define neocortical territories and neural circuitry The position of borders between brain regions is of primary physiological importance as it determines the neural tissue that will be allocated to specific brain functions. These borders are local transitions within the cerebral microarchitecture, the latter being formed of distinct populations of neurons and afferent inputs. Many studies aimed at understanding how boundaries are implemented have focused on the neocortex [1,4,15,20,35,36], a part of the brain with functional specializations that are highly conserved in mammals. The neocortex is organized into distinct primary sensory subdivisions (S1, A1, V1, M1 for somatosensory, auditory, visual, and motor cortex respectively) referred to as cortical areas or fields. These areas reproduce locally, through specific cytoarchitecture and/or chemoarchitecture and gene expression patterns, the topographic organization of peripheral sensory receptors to which they are connected [1,35] (see Lockmane and Garel, in this issue). The central representation of the periphery is very precise so that nearest neighbor relationships between primary sensory fields are maintained, in the subcortical relays and in the neocortex, thus forming the so-called topographic maps. The specification and differentiation of neocortical areas arises under the combined influences of extrinsic mechanisms, driven by afferent pathways that convey sensory information from the periphery (e.g. from the brainstem and thalamic relays), and genetic regulation, intrinsic to the neocortex. A significant number of studies have

shown that sensory deprivation by removal of sensory innervation from a body part during a critical period of development alters both functionally and physically the cortical representation of the body part [37–40]. Manipulating the periphery or its subcortical relays produces highly stereotyped changes in the organization of the neocortex in various mammals suggesting that the extrinsic mechanisms themselves are under a tight genetic control [28]. Among the various transcription factors (TFs), morphogens, and signaling molecules participating in neocortical arealization and compartmentalization are the HPs and TFs Emx2 (empty spiracles homeobox 2) and Pax6 (paired box 6) [4,41]. These two proteins have important functions along with the non-HP TFs COUP-TFI (also known as NR2f1) and sp8 in establishing the layout of neocortex subfields in the rostrocaudal axis [42–44]. Gradients of Emx2 and Pax6 expressed by cortical progenitor cells in the ventricular zone (VZ) of the neocortical primordium may act primarily in the formation of a cortical protomap [41,45,46]. Neuroepithelial cells specified by TFs in the cortical primordium then proliferate and differentiate to form a complex six-layered neocortical structure with regionally diverse cytoarchitectures. Positional informations initially driven within the protomap by Emx2 and Pax6 in particular, foreshadow the location and size of cortical subfields that will subsequently form. Acting as cortical field organizer, and/or by conferring specific identities to progenitors cells, HPs are involved in the final targeting of thalamo-cortical axons (TCA) inputs that innervate distinct cortical areas [20,28,36,47]. In the rodent VZ, Emx2 is expressed in a high Posterior-Medial to low Anterior-Lateral gradient and impart caudal areal identity [48,49], while Pax6 is expressed in an opposing pattern with a low P-M to high A-L gradient, and impart rostral identity to the cortical primordium [50] (Fig. 1a). In the mouse, reduction of V1 has been proposed as a possible consequence of Emx2 loss of function [45,50,51] (Fig. 1). And, V1 extension anteriorly at the expense of S1 and the fronto-motor area has been observed after Emx2 gain of function [15,45]. The same V1 shift was predicted after loss of Pax6 function [15,42,41], but this has not been confirmed in a conditional KO of Pax6 that does however show the expected reduction of S1 size [36] (Fig. 1). Under physiological conditions, EMX2 repression of PAX6 specification of rostral identity contributes to reduced rostral areas [45]. Thus Emx2 and Pax6 operate by concentration-dependent mechanisms in cortical progenitors to specify the sizes and positioning of the primary cortical areas that establish area-specific TCA projections. A recent study using viable Pax6 conditional knockout (cKO) showed that mice with a cortex-specific Pax6 deletion not only displayed a substantially reduced S1 but also a partial loss of the body sensory representation [36]. Tactile receptors distributed over the body are represented in somatotopic maps within S1. Rodent S1 has a large area allocated to posterior medial barrel subfield (PMBSF), and the anterior lateral barrel subfield (ALBSF) which receive sensory inputs from the facial whiskers [52]. Barrels in these fields consist in clusters of terminal arbors (the barrel core) of VPN (ventro posterior thalamic nucleus) afferents, synapsing onto the dendrites of spiny stellate neurons that form the barrel wall in layer 4 of the S1. Each barrel receives input from a single whisker and the barrel field is organized to represent the topographic distribution of the facial whiskers (see Lockmane and Garel, and Vitali and Jabaudon, in this issue). Pax6 cKO show sharply reduced PMBSF and ALBSF (Fig. 1b); the magnitude of this reduction is even greater than observed for the whole S1 suggesting that the portion of S1 allocated to the barrel field is specifically reduced in these TG mice [36]. These mice also show the loss of specific parts of the cortical barrel field (Fig. 1b) that may be due to an exaggerated competition among VPN TCA for limited cortical space. Finally, the reduced S1 in Pax6 cKO alters the VPN thalamic relay resulting in its re-patterning to match the

O. Stettler, K.L. Moya / Seminars in Cell & Developmental Biology 35 (2014) 165–172

167

Fig. 1. Emx2 and Pax6 modulate the size of sensory fields and Pax6 is active in the completion of the sensory maps. (a) (Upper two panels) graded expression patterns of the transcription factors Emx2 and Pax6 in the mouse embryonic neocortex. Emx2 is expressed in a high medioposterior to low rostrolateral gradient, and Pax6 is expressed in an opposing gradient to this of Emx2. (Lower panels) predicted changes in area extent in neonatal neocortex in the mutants Emx2 KO (Emx2−/−) and sey/sey (small eye mutant) that lack functional Pax6 protein. Changes are presented relative to control wild type (WT, middle panel). The shifted areas are predicted only because Emx2 and sey/sey mutants die after birth, however reduced S1 was confirmed recently with a conditional KO for Pax6 [36] (a, adapted from [15]). V1, visual, M1, motor, and somatosensory (S1) areas. Abbreviations: C, caudal; L, lateral; M, medial; R, rostral. (b) (Upper two panels) schemas displaying the respective position of the primary components of the S1 body map in wild type and Pax6 cKO mice on tangential sections of flattened cortex at PND7. Barrel rows in PMBSF and ALBSF of wild type and Pax6 cKO are illustrated in the lower two drawings. All primary components of the S1 body map and barrels were miniaturized in cKO mice. Most of the ALBSF barrel subfield and the entire PMBSF barrel row ‘A’ were also absent in S1 of Pax6 cKO mice (adapted from [36]). Abbreviations: PMBSF: posterior medial barrel subfield, ALBSF: anterolateral barrel subfield, FP: forepaw map, HP: hindpaw map, T: map of the trunk/rest of the body, LJ: lower jaw map.

aberrant body map in S1 [36]. Thus, Pax6 and S1 size determine the completeness of body maps and engages ‘top-down’ plasticity to physically and functionally match specific thalamus and cortical areas [36]. The HP, coded by the LIM homeobox gene Lhx2, is expressed by neocortical progenitors within the ventricular zone of the dorsal telencephalon throughout cortical neurogenesis [3]. Early loss of Lhx2 function results in severe defects that prevent the formation of the neocortex [53,54]. Conditional deletion of Lhx2 at embryonic day (E) 10.5 using an Emx1Cre driver produces ectopic paleocortex ressembling the three-layer olfactory cortex instead of the six-layers neocortex [55]. Shetty and collaborators [20] used an Emx1Cre line acting in the dorsal telencephalon starting at E11.5. Despite the absence of Lhx2 after E11.5, a true neocortex did form but the thickness of neocortical layers was reduced. In these mice, the authors found that Lhx2 was essential to the formation of the barrel field [20]. Indeed, the barrels of Lhx2 cKO mice, as visualized by cytochrome oxidase staining, are completely missing. Yet the projections between the thalamus and cortex are maintained in the appropriate location suggesting that the growth TCA into the cortex is not affected by Lhx2 silencing in the Emx1Cre line [20]. This outcome appears different from that observed after silencing Lhx2 in the thalamus, which results in the aberrant targeting of TCAs to cortical areas consistent with a role for Lhx2 in TCA guidance [28]. When Lhx2 is silenced in the Emx1Cre cortical cell line mice, stimulation of individual whiskers failed to evoke detectable response. However, cortical responses were obtained after stimulation of multiple whiskers [20]. In these mice, the response in S1 to stimulation of other body parts such as hind limb and tail appear normal. Thus, Lhx2 expression by

cortical progenitors may be necessary for a point-to-point and whisker-barrel specific input relationship, rather than for the specification of the S1 area as a whole and its corresponding TCA inputs. The KO mice for Pax6, Emx2, and Lhx2 as described above reveal that HPs are not only capable of influencing the early primary sensory field location and size (i.e. arealization or area identity) but also contribute to the later modular organization of cortical connectivity (i.e. the functional local circuitry). What are the molecular correlates of HP functions during these processes? HPs regulate cortical progenitor number, which is critical for determining the number of cortical neurons and thus the size of distinct cortical areas. For instance, Lhx2 is needed for the expression of Hes1, a key effector in the Notch signaling pathway that maintains the proliferative state of neocortical progenitors during corticogenesis [3]. Among molecules that may influence neuronal circuit patterning under the control of HPs are adhesion and axonal guidance proteins. Emx2 and Pax6 regulate the cortical expression pattern of several cadherins (Cad6, Cad8, Cad11) [50]. These are part of large family of adhesion proteins involved in synaptogenesis [56], known to participate in the formation of precise thalamocortical circuitry [57]. Ephrin-A5, a guidance molecule implicated in TCA branching [58] and synaptogenesis [59] is expressed in S1 cortex and is greatly reduced in Lhx2 cKO [20]. In thalamic cells, the guidance receptors Robo1 and Robo2 are downstream targets of the Lhx2 activity and modulate thalamocortical axon pathfinding [28]. Thus, HPs not only function in cortical cells to regulate proliferation and differentiation, they also act to maintain spatial and functional relationships between subcortical neuronal populations and their target fields in the cortex.

168

O. Stettler, K.L. Moya / Seminars in Cell & Developmental Biology 35 (2014) 165–172

Fig. 2. Engrailed sensitizes temporal RGC growth cones to the collapsing effect of Ephrin-A5. (a) Engrailed (En) increases ATP production and secretion from the growth cone via a translation-dependent mechanism. Extracellular ATP is converted in adenosine that interacts with adenosine A1 receptor (A1R), and A1R potentiates the signaling effect of Ephrin through Eph receptor. (b) Drawing showing the retinotectal projections of RGCs axons from the nasal retina (N) to the posterior tectum (P) and of RGCs axons from the temporal retina (T) to the anterior tectum (A). Engrailed and Ephrin-A5 are both expressed in an increasing gradients along the A–P axis of the tectum. Engrailed by co-signaling with Ephrin may enhance the precision of connections of RGC axons within the gradients (black solid arrows) compared to less precise connections in the absence of Engrailed (dashed red arrows). Source: Adapted from [24].

3. Signaling with Engrailed during retino-tectal mapping One of the most studied systems used to understand general mapping processes and central sensory representation, is the retinotopic map formed in the lateral geniculate nucleus (thalamus, LGN), the dorsal midbrain (chick tectum or mouse superior colliculus, SC), and in the primary visual cortex V1 [60]. In the visual system, the Cartesian coordinates of the retina are mapped onto the LGN/SC/V1. Retinal ganglion cell (RGC) axons from the nasal and temporal retina project to the posterior and anterior SC respectively so that the temporal–nasal (T–N) axis of the retina maps along the anterior–posterior (A–P) axis of the tectum/superior colliculus (see Franco Weth et al., in this issue). Similarly the axons of dorsal and ventral RGC project on the medial and lateral regions of the tectum/SC. In this manner, a point-to-point sensory map is formed in the tectum/SC during the period of RGC axon growth. This is achieved principally with the help of gradients of guidance molecules that repel advancing RGC axonal growth cones from inappropriate areas and/or attract RGC growth cones to appropriate target areas [61]. The accurate positioning of RGC axon terminals at a given location in the tectal map is largely controlled by the Ephrin-A/B guidance molecules and their receptors EphA/B [61,62]. A diminishing T–N gradient of EphA receptors on RGC axons and an increasing gradient of Ephrin-As in the tectum generated by forward signaling underlies the low-to-high antero-posterior repulsion of axons from temporal retina [61,62]. Engrailed homeodomain transcription factors regulate Ephrin-A expression in the chick optic tectum [63] and Engrailed and Ephrin-A5 have the same graded expression in this structure [64,65] (Fig. 2). In addition to the indirect role of Engrailed in RGC axonal guidance via regulation of Ephrin expression, a series of studies have shown that Engrailed directly guides retinal axons in vitro and in vivo [21,24,25]. In cultures of Xenopous retinal explants, extracellular Engrailed attracts axons from nasal retina and repels axons from temporal retina [21]. In the chick optic tectum, about 5% of Engrailed is in the extracellular space and its neutralization with simple chain antibodies leads

to misrouting of RGC axons [25]. Concordantly in vitro, Engrailed acts synergistically with Ephrin-A5 during axonal guidance [24,25] (Fig. 2). Subthreshold concentrations of soluble Ephrin-A5 that do not induce collapse of RGC temporal growth cones alone, provoke the collapse of growth cones when added in combination with nanomolar concentrations of soluble Engrailed [24,25]. The collapse/guidance effect of Engrailed requires its internalization by RGC growth cones and local protein translation. Thus, during retinotectal map formation Engrailed can act as a direct signaling molecule. HP cell-to-cell signaling during development may be more general than previously thought. HP transfer is supported by the presence of secretion and internalization motifs in the highly conserved HP homeodomain [66], and Engrailed and two other HPs (i.e. Otx2 and Pax6) have been shown to signal between cells in other developmental contexts. In drosophila, secreted Engrailed participates in the anterior wing cross vein development via interaction with the Dpp signaling pathway [67]. In the chicken neural tube, extracellular Pax6 promotes oligodendrocytes precursor cell migration [68], and in the mouse, specific transfer and accumulation of Otx2 in parvalbumin (PV) expressing GABAergic interneurons is necessary and sufficient to open, then close, a critical period (CP) of plasticity in the developing visual cortex [26] (see Section 4). Considering again the retinotectal system, one may wonder about the physiological significance of the non-cell autonomous function of Engrailed in brain map formation that is dominated by the Ephrins. This question cannot be answered simply by a genetic approach since En1/2 double mutants lack the mid/hindbrain region [69]. We found that the guidance activity of Engrailed, which has no known membrane receptor, requires the activation of purinergic signaling and of the adenosine A1 receptor (A1R) (Fig. 2) [24]. How this signaling cooperates with the classical signaling of Ephrin-A5 is not known, but we observed that the pharmacological manipulation of A1R signaling impacts Ephrin-A5 guidance activity [24]. Thus, an antagonist of A1R (DPCPX) abrogates the sensitization by Engrailed of the Ephrin collapse activity on temporal

O. Stettler, K.L. Moya / Seminars in Cell & Developmental Biology 35 (2014) 165–172

169

Fig. 3. Otx2 governs the ocular dominance plasticity during the binocular visual field completion. (a) Otx2 opens, then closes the critical period of ocular dominance plasticity in the visual binocular cortex (red solid curve). Cortical infusion of Otx2 in the visual cortex in the early postnatal period prematurely opens and closes the critical period (red dashed curve) (based on [27]). In conditional Otx2 KO, the critical period does not open. (b) Illustration of the binocular visual cortex where, during the CP, competitive mechanisms depending on visual experience between the inputs from the ipsilateral and controlateral eyes are required for the strengthening and consolidation of the visual receptive field architecture. (c) A simplified functional circuit linking RGC neurons in the retina to layer II/III pyramidal cells and to neighboring parvalbumin expressing interneurons in the visual cortex. Otx2 transferred from the retina and from the choroid plexus is internalized by PV interneurons. Internalized Otx2 stimulates their maturation and the formation of a perineuronal net (PNN). Maturation of PV interneurons stabilizes synaptic connections within the visual binocular cortex. Abbreviations: RGC, retinal ganglionic cells; dLGN, dorsolateral geniculate nucleus; PV, parvalbumin; II–III–IV, layers 2, 3, and 4 of the visual cortex.

growth cones, while an agonist (CCPA) mimics this sensitization. Furthermore, by using a classical stripe assay [25], it could be possible to show that DPCPX completely disturbs the guidance of RGC temporal growth cones [24]. On the basis of our finding that A1R showed, like EphA2 (i.e. the receptor for Ephrin-A5), a high temporal, low-nasal expression in RGC growth cones, we developed a mathematical model that explains how Engrailed signaling can increase the sensitivity of RGC temporal growth cones to EphrinA5 and the precision of their navigation [24] (Fig. 2; see Reingruber and Holcman, in this issue). Thus, Engrailed’s main function during development of the retinocollicular system could be to adjust the

sensitivity of axons to Ephrin-A5 in order to increase the precision of the retino-tectal map. 4. Otx2 stabilizes ocular dominance in the visual cortex Three consecutive phases have been characterized in the development of the V1 visual cortical field, and only the last one is dependent of visual experience [for review see 71]. The first phase is the formation of a precise visual topographic map. As seen above for the superior colliculus, a retinal map is formed in the LGNd (dorso lateral geniculate nucleus) and this map is reproduced into

170

O. Stettler, K.L. Moya / Seminars in Cell & Developmental Biology 35 (2014) 165–172

V1 by the orderly and precise connections from LGNd axons. The resulting topographic map in V1, is formed using both Ephrin/Eph signaling and spontaneous activity. This phase prepares neurons to be receptive to, and to encode changes occurring in visual space. The second developmental phase confers orientation selectivity and ocular dominance (i.e. the preferential response of a neuron to one eye or the other) to V1b (V1 binocular field) neurons. The last phase is referred to as the “critical period” (CP). The CP is a phase of augmented “plasticity”, during which neural circuitry in V1b is sensitive to peripheral sensory deprivation [71]. During the CP individual cells in mouse V1b integrate inputs from the two eyes [71] leading to increased visual acuity. During this “plastic phase”, competitive mechanisms between the visual inputs from the ipsilateral and controlateral eyes are required for the strengthening and consolidation of the visual receptive field architecture in the visual binocular cortex (Fig. 3). The closure of one eye during CP can permanently shift the response properties of neurons in the primary visual cortex to favor inputs from the open eye, i.e. ocular dominance [OD] shift [72]. This occurs with a rapid pruning of dendritic spines and later rewiring of thalamo-cortical afferents [71,73]. Raising animals in the dark or depriving them of binocular vision from birth displaced the onset of CP into adulthood [74,75] by a specific reduction of somatic inhibition [76,77]. Indeed, for a reason that is not yet fully understood, normal CP opening (P21 in the mouse) corresponds to an increase in maturation of local inhibitory circuits [71]. This maturation is driven primarily by a subclass of GABAergic basket cells expressing parvalbumin (PV) and enwrapped by glycosaminoglycan matrix molecules forming a perineuronal net (PNN). Recent studies have reported that Otx2 (orthodenticle homeobox protein 2) coordinates the PV cell maturation, and thus the timing of the CP plasticity in mouse V1b [26,27,78,79]. Conditional knockout of Otx2 prevents the maturation of PV cells and prevents the opening of the critical period (Fig. 3). Dark-rearing from birth strongly reduced Otx2 and PV signals while PV expression and plasticity are rescued by cortical infusion of recombinant Otx2 protein which accumulated in PV cells [26]. This rescue experiment clearly suggested the possibility of a cellular transfer of endogenous Otx2. Several lines of evidence indicate indeed that PV cortical cells do not produce Otx2 themselves although they accumulate Otx2 as soon as visual activity increases. First of all, Otx2 mRNA is not expressed in the visual cortex (nor in any other cortices, [27]). Second, Otx2-targeted siRNA retard the activity-dependent opening of CP in the visual cortex when they are delivered into the retina, but not when injected into the cortex. Finally, PV-cell maturation and critical period plasticity were impaired in transgenic Otx2 conditional knockout mouse line driven by a CRE recombinase that is not expressed in PV cells. Collectively these experiments demonstrate [26] that Otx2 acts non-cell autonomously, and is synthesized and transferred from outside V1 to visual cortex by an activity dependent mechanism to enable CP-associated plasticity. This intriguing finding raises the question of the source of synthesis of Otx2. Activity dependent CP opening may require the transport of Otx2 from particular sites along the visual pathway to initiate in the cortex the maturation of PV cells marked by PNN assembly. A retinal source of transportable Otx2 to visual relays could arise by Otx2 protein accumulation in ganglion cells which are contacted by bipolar cells that express Otx2 mRNA and protein [26]. In this respect, blocking Otx2 extracellular transfer (i.e. between bipolar cells and retinal ganglion cells) by infusing Otx2 antibodies in the retina retards the activity-dependent opening of CP in visual cortex. In addition, tagged-Otx2 injected in the eye can travel along the visual pathway and terminates in PV cells [26], confirming that Otx2 can be transported from the retina to V1. An alternative and much more important source of Otx2 that has been recently identified is the choroid plexus [27]. Knocking

down Otx2 specifically in the choroid plexus decreases the number of Otx2-positive cells and of PV cells surrounded by a PNN in the visual cortex [27]. Ablating Otx2 in the choroid plexus of adult mice also rejuvenates their visual cortex as illustrated by decreased visual acuity in these animals, an index of the reactivation of cortical plasticity [27]. PNNs gradually form around PV during the CP in parallel with the decline of critical period plasticity [80,81]. The ability of PV cells to capture Otx2 after initial PNN assembly seems to be the key event regulating plasticity. Otx2 internalization enhances expression of several markers of PV-cell maturation, including PNN formation itself, and alteration or destruction of these ECM scaffolds interferes with Otx2 internalization and CP closure [78,79]. Thus, Otx2 by both opening and closing the CP behaves as a central regulator of ocular dominance and of the maturation of the visual map within the visual cortex. 5. Concluding remarks and perspectives The results discussed in this review indicate that two different functional modes may characterize HP function during the development of brain maps. The first one is cell-autonomous, requires transcription, and is necessary for the early specification of cortical areas, or for thalamic and retinal axonal guidance [15,20,28,45,41]. The second mode is non-cell-autonomous, requires intercellular transfer of HPs, can in addition activate translation, and has been implicated in segregation of retinal inputs, axonal stability, and visual map completion in the cortex [24–27,78,82]. Significantly, non-cell autonomous and cell-autonomous functions of HPs do not appear to be mutually exclusive at spatial or temporal levels during brain development. This observation must be tempered by the fact that relatively few studies have yet addressed the issue of a non-cell autonomous function of HPs during the formation of connective maps. One reason for this could be that the differentiation of brain areas has been principally seen for years, and in line with the protomap model [2,46], as a delayed manifestation of early transcriptional activities of HPs in progenitor cells. This explanation is clear in the work of Shetty and collaborators [20] that implicitly links the lack of clustering of thalamic afferent in the barrel field in Emx1Cre line (see Section 2) to the lack of Lhx2 expression in the early prenatal period. However, examination of Lhx2 expression pattern in a brain atlas [http://www.gensat.org] reveals that Lhx2 is susceptible to play a role at later stages of development. Specifically, Lhx2 is expressed in several cortical layers in the mouse by postnatal day 7 that corresponds to the completion of barrel field formation. Evidence showing that Otx2 signaling affects the cortical visual map organization during the postnatal period raises the possibility that Lhx2 or another HP may also act noncell-autonomously to modulate the morphology of TCA inputs in the barrel field. The persistence of HP functions beyond the early embryonic period may be more general than shown for Otx2 considering the dozens of HPs including Cux1/2, Arx, Engrailed-1/2, Dlx1/2 (a complete list can be found at: http://www.gensat.org/searchgenes.jsp) whose expression persists in neonatal and adult forebrain and midbrain regions. It seems possible in this context that the role of HP proteins would not only be to create the conditions for gross architectural changes in the neocortex but also to support and/or stabilize these changes at a finer level. This finer level could be the cell itself. As originally suggested and demonstrated by Alain Prochiantz and colleagues [29], homeogenes, encoding positional information, can regulate the shape of single cells in addition to that of organs [83]. Many molecules that participate in brain mapping (purinergic, adhesion, guidance, and cytoskeletal molecules) that are transcriptional and translational targets of HPs also contribute to cell morphogenesis. A adenosine triphosphate secretion

O. Stettler, K.L. Moya / Seminars in Cell & Developmental Biology 35 (2014) 165–172

by neural cells is augmented by Engrailed signaling [24], and extracellular ATP is known to modulate NCAM-mediated cell adhesion and axonal growth [84,85]. We also observed recently, that extracellular Engrailed increases the complexity of the arborization of hippocampal cell dendrites and the plasticity of dendritic spines (Soltani et al., submitted for publication). Another example of how a HP may contribute to both brain arealization and cellular differentiation is given by the Cux1/Cux2 homeoproteins. By controlling cell cycle exit in a cell-autonomous manner Cux-2 is a key element in the control of the proliferation rates of the subventricular zone (SVZ) neural precursors in the developing cortex. By this means Cux-2 regulates the number of upper layer (I–IV) neurons [86], but at later stages of the development, Cux1 and Cux2 complementarily regulate dendrite branching, dendritic spine development, and synapse formation in layer II–III neurons of the cerebral cortex [87]. Dendrite branching and dendritic spines formation determine the function of morphologically distinct and specialized neuronal subclasses. Thus, Cux HPs may participate in the construction of upper cortical layer identities not only through the control of progenitor proliferation, but also by allowing post-mitotic neural cells to differentiate and integrate in a specific neuronal network. Collectively, these observations open new territories to explore the roles of HPs in sensory map construction in the brain. We suggest two general objectives. On the basis of the signaling functions of HPs and the example provided by Otx2 in visual cortex, future studies should consider the possibility that sensory maps may be influenced by various sources, even remote, of HP expression. Also, characterizing the functions of the HPs in the adult cortex by conditional KO could help to understand the exact influence of this class of proteins on the stability and/or plasticity of distinct cortical fields. References [1] Huffman K. The developing, aging neocortex: how genetics and epigenetics influence early developmental patterning and age-related change. Front Genet 2012;3:212. [2] Rakic P, Ayoub AE, Breunig JJ, Dominguez MH. Decision by division: making cortical maps. Trends Neurosci 2009;32(5):291–301. [3] Chou S-J, O’Leary DDM. Role for Lhx2 in corticogenesis through regulation of progenitor differentiation. Mol Cell Neurosci 2013;56(1):9. [4] O’Leary DDM, Chou S-J, Sahara S. Area patterning of the mammalian cortex. Neuron 2007;56(2):252–69. [5] Kayam G, Kohl A, Magen Z, Peretz Y, Weisinger K, Bar A, et al. A novel role for Pax6 in the segmental organization of the hindbrain. Development 2013;140(10):2190–202. [6] Coffinier C, Thépot D, Babinet C, Yaniv M, Barra J. Essential role for the homeoprotein vHNF1/HNF1beta in visceral endoderm differentiation. Development 1999;126(21):4785–94. [7] Boyl PP, Signore M, Annino A, Barbera JP, Acampora D, Simeone A. Otx genes in the development and evolution of the vertebrate brain. Int J Dev Neurosci 2001;19(4):353–63. [8] Deckelbaum RA, Majithia A, Booker T, Henderson JE, Loomis CA. The homeoprotein engrailed 1 has pleiotropic functions in calvarial intramembranous bone formation and remodeling. Development 2006;133(1):63–74. [9] Hale MA, Kagami H, Shi L, Holland AM, Elsässer H-P, Hammer RE, et al. The homeodomain protein PDX1 is required at mid-pancreatic development for the formation of the exocrine pancreas. Dev Biol 2005;286(1):225–37. [10] Harvey RP. NK-2 homeobox genes and heart development. Dev Biol 1996;178(2):203–16. [11] Mankoo BS, Collins NS, Ashby P, Grigorieva E, Pevny LH, Candia A, et al. Mox2 is a component of the genetic hierarchy controlling limb muscle development. Nature 1999;400(6739):69–73. [12] Orkin SH. Development of the hematopoietic system. Curr Opin Genet Dev 1996;6(5):597–602. [13] Li Z, Deng D, Huang H, Tian L, Chen Z, Zou Y, et al. Overexpression of Six1 leads to retardation of myogenic differentiation in C2C12 myoblasts. Mol Biol Rep 2013;40(1):217–23. [14] Bendall AJ, Abate-Shen C. Roles for Msx and Dlx homeoproteins in vertebrate development. Gene 2000;247(1-2):17–31. [15] Bishop KM, Rubenstein JLR, O’Leary DDM. Distinct actions of Emx1 Emx2, and Pax6 in regulating the specification of areas in the developing neocortex. J Neurosci 2002;22(17):7627–38. [16] Joyner AL. Engrailed, Wnt and Pax genes regulate midbrain–hindbrain development. Trends Genet TIG 1996;12(1):15–20. [17] Narita Y, Rijli FM. Hox genes in neural patterning and circuit formation in the mouse hindbrain. Curr Top Dev Biol 2009;88:139–67.

171

[18] Nunes I, Tovmasian LT, Silva RM, Burke RE, Goff SP. Pitx3 is required for development of substantia nigra dopaminergic neurons. Proc Natl Acad Sci USA 2003;100(7):4245–50. [19] Veenvliet JV, Dos Santos MTMA, Kouwenhoven WM, von Oerthel L, Lim JL, van der Linden AJA, et al. Specification of dopaminergic subsets involves interplay of En1 and Pitx3. Development 2013;140(16):3373–84. [20] Shetty AS, Godbole G, Maheshwari U, Padmanabhan H, Chaudhary R, Muralidharan B, et al. Lhx2 regulates a cortex-specific mechanism for barrel formation. Proc Natl Acad Sci USA 2013;110(50):E4913–21. [21] Brunet I, Weinl C, Piper M, Trembleau A, Volovitch M, Harris W, et al. The transcription factor Engrailed-2 guides retinal axons. Nature 2005;438(7064):94–8. [22] Fox SR, Deneris ES. Engrailed is required in maturing serotonin neurons to regulate the cytoarchitecture and survival of the dorsal raphe nucleus. J Neurosci 2012;32(23):7832–42. [23] Sillitoe RV, Vogel MW, Joyner AL. Engrailed homeobox genes regulate establishment of the cerebellar afferent circuit map. J Neurosci 2010;28(30):10015–24. [24] Stettler O, Joshi RL, Wizenmann A, Reingruber J, Holcman D, Bouillot C, et al. Engrailed homeoprotein recruits the adenosine A1 receptor to potentiate ephrin A5 function in retinal growth cones. Development 2012;139(1):215–24. [25] Wizenmann A, Brunet I, Lam JSY, Sonnier L, Beurdeley M, Zarbalis K, et al. Extracellular Engrailed participates in the topographic guidance of retinal axons in vivo. Neuron 2009;64(3):355–66. [26] Sugiyama S, Di Nardo AA, Aizawa S, Matsuo I, Volovitch M, Prochiantz A, et al. Experience-dependent transfer of Otx2 homeoprotein into the visual cortex activates postnatal plasticity. Cell 2008;134(3):508–20. [27] Spatazza J, Lee HHC, Di Nardo AA, Tibaldi L, Joliot A, Hensch TK, et al. Choroidplexus-derived Otx2 homeoprotein constrains adult cortical plasticity. Cell Rep 2013;3(6):1815–23. [28] Marcos-Mondéjar P, Peregrín S, Li JY, Carlsson L, Tole S, López-Bendito G. The lhx2 transcription factor controls thalamocortical axonal guidance by specific regulation of robo1 and robo2 receptors. J Neurosci 2012;32(13):4372–85. [29] Prochiantz A. Signaling with homeoprotein transcription factors in development and throughout adulthood. Curr Genomics 2013;14(6):361–70. [30] Spatazza J, Di Lullo E, Joliot A, Dupont E, Moya KL, Prochiantz A. Homeoprotein signaling in development, health, and disease: a shaking of Dogmas offers challenges and promises from bench to bed. Pharmacol Rev 2013;65(1):90–104. [31] Nédélec S, Foucher I, Brunet I, Bouillot C, Prochiantz A, Trembleau A. Emx2 homeodomain transcription factor interacts with eukaryotic translation initiation factor 4E [eIF4E] in the axons of olfactory sensory neurons. Proc Natl Acad Sci USA 2004;101(29):10815–20. [32] Dubnau J, Struhl G. RNA recognition and translational regulation by a homeodomain protein. Nature 1996;379(6567):694–9. [33] Topisirovic I, Culjkovic B, Cohen N, Perez JM, Skrabanek L, Borden KLB. The proline-rich homeodomain protein PRH, is a tissue-specific inhibitor of eIF4E-dependent cyclin D1 mRNA transport and growth. EMBO J 2003;22(3):689–703. [34] Orvis GD, Hartzell AL, Smith JB, Barraza LH, Wilson SL, Szulc KU, et al. The engrailed homeobox genes are required in multiple cell lineages to coordinate sequential formation of fissures and growth of the cerebellum. Dev Biol 2012;367(1):25–39. [35] Krubitzer L, Kaas J. The evolution of the neocortex in mammals: how is phenotypic diversity generated? Curr Opin Neurobiol 2005;15(4):444–53. [36] Zembrzycki A, Chou S-J, Ashery-Padan R, Stoykova A, O’Leary DDM. Sensory cortex limits cortical maps and drives top-down plasticity in thalamocortical circuits. Nat Neurosci 2013;16(8):1060–7. ˜ C. The size of the whisker barrel field [37] Machín R, Blasco B, Bjugn R, Avendano in adult rats: minimal nondirectional asymmetry and limited modifiability by chronic changes of the sensory input. Brain Res 2004;1025(1–2):130–8. [38] Stettler O, Tavitian B, Moya KL. Differential synaptic vesicle protein expression in the barrel field of developing cortex. J Comp Neurol 1996;375(2):321–32. [39] Rosenkranz K, Seibel J, Kacar A, Rothwell J. Sensorimotor deprivation induces interdependent changes in excitability and plasticity of the human hand motor cortex. J Neurosci 2014;34(21):7375–82. [40] Wong C, Chabot N, Kok MA, Lomber SG. Modified areal cartography in auditory cortex following early- and late-onset deafness. Cereb Cortex 2014;24(7):1778–92. [41] Muzio L, Di Benedetto B, DiBenedetto B, Stoykova A, Boncinelli E, Gruss P, et al. Emx2 and Pax6 control regionalization of the pre-neuronogenic cortical primordium. Cereb Cortex 2002;12(2):129–39. [42] O’Leary DD, Sahara S. Genetic regulation of arealization of the neocortex. Curr Opin Neurobiol 2008;18(1):90–100. [43] Armentano M, Chou S-J, Tomassy GS, Leingärtner A, O’Leary DDM, Studer M. COUP-TFI regulates the balance of cortical patterning between frontal/motor and sensory areas. Nat Neurosci 2007;10(10):1277–86. [44] Zembrzycki A, Griesel G, Stoykova A, Mansouri A. Genetic interplay between the transcription factors Sp8 and Emx2 in the patterning of the forebrain. Neural Dev 2007;2:8. [45] Hamasaki T, Leingärtner A, Ringstedt T, O’Leary DDM. EMX2 regulates sizes and positioning of the primary sensory and motor areas in neocortex by direct specification of cortical progenitors. Neuron 2004;43(3):359–72. [46] Elsen GE, Hodge RD, Bedogni F, Daza RAM, Nelson BR, Shiba N, et al. The protomap is propagated to cortical plate neurons through an Eomes-dependent intermediate map. Proc Natl Acad Sci USA 2013;110(10):4081–6. [47] López-Bendito G, Chan C-H, Mallamaci A, Parnavelas J, Molnár Z. Role of Emx2 in the development of the reciprocal connectivity between cortex and thalamus. J Comp Neurol 2002;451(2):153–69.

172

O. Stettler, K.L. Moya / Seminars in Cell & Developmental Biology 35 (2014) 165–172

[48] Simeone A, Gulisano M, Acampora D, Stornaiuolo A, Rambaldi M, Boncinelli E. Two vertebrate homeobox genes related to the Drosophila empty spiracles gene are expressed in the embryonic cerebral cortex. EMBO J 1992;11(7):2541–50. [49] Simeone A, Acampora D, Gulisano M, Stornaiuolo A, Boncinelli E. Nested expression domains of four homeobox genes in developing rostral brain. Nature 1992;358(6388):687–90. [50] Bishop KM, Goudreau G, O’Leary DD. Regulation of area identity in the mammalian neocortex by Emx2 and Pax6. Science 2000;288(5464): 344–9. [51] Mallamaci A, Muzio L, Chan CH, Parnavelas J, Boncinelli E. Area identity shifts in the early cerebral cortex of Emx2−/− mutant mice. Nat Neurosci 2000;3(7):679–86. [52] Woolsey TA, Van der Loos H. The structural organization of layer IV in the somatosensory region [SI] of mouse cerebral cortex. The description of a cortical field composed of discrete cytoarchitectonic units. Brain Res 1970;17(2):205–42. [53] Bulchand S, Grove EA, Porter FD, Tole S. LIM-homeodomain gene Lhx2 regulates the formation of the cortical hem. Mech Dev 2001;100(2):165–75. [54] Mangale VS, Hirokawa KE, Satyaki PRV, Gokulchandran N, Chikbire S, Subramanian L, et al. Lhx2 selector activity specifies cortical identity and suppresses hippocampal organizer fate. Science 2008;319(5861):304–9. [55] Chou S-J, Perez-Garcia CG, Kroll TT, O’Leary DDM. Lhx2 specifies regional fate in Emx1 lineage of telencephalic progenitors generating cerebral cortex. Nat Neurosci 2009;12(11):1381–9. [56] Arikkath J, Reichardt LF. Cadherins and catenins at synapses: roles in synaptogenesis and synaptic plasticity. Trends Neurosci 2008;31(9):487–94. [57] Gil OD, Needleman L, Huntley GW. Developmental patterns of cadherin expression and localization in relation to compartmentalized thalamocortical terminations in rat barrel cortex. J Comp Neurol 2002;453(4):372–88. [58] Uziel D, Mühlfriedel S, Bolz J. Ephrin-A5 promotes the formation of terminal thalamocortical arbors. Neuroreport 2008;19(8):877–81. [59] Akaneya Y, Sohya K, Kitamura A, Kimura F, Washburn C, Zhou R, et al. EphrinA5 and EphA5 interaction induces synaptogenesis during early hippocampal development. PLoS ONE 2010;5(8):e12486. [60] Erskine L, Herrera E. The retinal ganglion cell axon’s journey: insights into molecular mechanisms of axon guidance. Dev Biol 2007;308(1):1–14. [61] Flanagan JG. Neural map specification by gradients. Curr Opin Neurobiol 2006;16(1):59–66. [62] Scicolone G, Ortalli AL, Carri NG. Key roles of Ephs and ephrins in retinotectal topographic map formation. Brain Res Bull 2009;79(5):227–47. [63] Logan C, Wizenmann A, Drescher U, Monschau B, Bonhoeffer F, Lumsden A. Rostral optic tectum acquires caudal characteristics following ectopic engrailed expression. Curr Biol 1996;6(8):1006–14. [64] Itasaki N, Nakamura H. A role for gradient en expression in positional specification on the optic tectum. Neuron 1996;16(1):55–62. [65] Rétaux S, McNeill L, Harris WA. Engrailed, retinotectal targeting, and axonal patterning in the midbrain during Xenopus development: an antisense study. Neuron 1996;16(1):63–75. [66] Joliot A, Prochiantz A. Transduction peptides: from technology to physiology. Nat Cell Biol 2004;6(3):189–96. [67] Layalle S, Volovitch M, Mugat B, Bonneaud N, Parmentier M-L, Prochiantz A, et al. Engrailed homeoprotein acts as a signaling molecule in the developing fly. Development 2011;138(11):2315–23.

[68] Di Lullo E, Haton C, Le Poupon C, Volovitch M, Joliot A, Thomas J-L, et al. Paracrine Pax6 activity regulates oligodendrocyte precursor cell migration in the chick embryonic neural tube. Development 2011;138(22):4991–5001. [69] Liu A, Joyner AL. EN and GBX2 play essential roles downstream of FGF8 in patterning the mouse mid/hindbrain region. Development 2001;128(2): 181–91. [70] Espinosa JS, Stryker MP. Development and plasticity of the primary visual cortex. Neuron 2012;75(2):230–49. [71] Hensch TK. Critical period plasticity in local cortical circuits. Nat Rev Neurosci 2005;6(11):877–88. [72] Hubel DH, Wiesel TN. Stereoscopic vision in macaque monkey cells sensitive to binocular depth in area 18 of the macaque monkey cortex. Nature 1970;225(5227):41–2. [73] Antonini A, Fagiolini M, Stryker MP. Anatomical correlates of functional plasticity in mouse visual cortex. J Neurosci 1999;19(11):4388–406. [74] Mower GD. The effect of dark rearing on the time course of the critical period in cat visual cortex. Brain Res Dev Brain Res 1991;58(2):151–8. [75] Fagiolini M, Katagiri H, Miyamoto H, Mori H, Grant SGN, Mishina M, et al. Separable features of visual cortical plasticity revealed by N-methyl-d-aspartate receptor 2A signaling. Proc Natl Acad Sci USA 2003;100(5):2854–9. [76] Iwai Y, Fagiolini M, Obata K, Hensch TK. Rapid critical period induction by tonic inhibition in visual cortex. J Neurosci 2003;23(17):6695–702. [77] Katagiri H, Fagiolini M, Hensch TK. Optimization of somatic inhibition at critical period onset in mouse visual cortex. Neuron 2007;53(6):805–12. [78] Beurdeley M, Spatazza J, Lee HHC, Sugiyama S, Bernard C, Di Nardo AA, et al. Otx2 binding to perineuronal nets persistently regulates plasticity in the mature visual cortex. J Neurosci 2012;32(27):9429–37. [79] Miyata S, Komatsu Y, Yoshimura Y, Taya C, Kitagawa H. Persistent cortical plasticity by upregulation of chondroitin 6-sulfation. Nat Neurosci 2012;15(3), 414–22, S1–2. [80] Celio MR, Spreafico R, De Biasi S, Vitellaro-Zuccarello L. Perineuronal nets: past and present. Trends Neurosci 1998;21(12):510–5. [81] Pizzorusso T, Medini P, Berardi N, Chierzi S, Fawcett JW, Maffei L. Reactivation of ocular dominance plasticity in the adult visual cortex. Science 2002;298(5596):1248–51. [82] Yoon BC, Jung H, Dwivedy A, O’Hare CM, Zivraj KH, Holt CE. Local translation of extranuclear lamin B promotes axon maintenance. Cell 2012;148(4): 752–64. [83] Bloch-Gallego E, Le Roux I, Joliot AH, Volovitch M, Henderson CE, Prochiantz A. Antennapedia homeobox peptide enhances growth and branching of embryonic chicken motoneurons in vitro. J Cell Biol 1993;120(2):485–92. [84] Skladchikova G, Ronn LC, Berezin V, Bock E. Extracellular adenosine triphosphate affects neural cell adhesion molecule [NCAM]-mediated cell adhesion and neurite outgrowth. J Neurosci Res 1999;57(2):207–18. [85] Bork K, Hoffmann M, Horstkorte R. ATP interferes with neural cell adhesion molecule-induced neurite outgrowth. Neuroreport 2013;24(11):616–9. [86] Zimmer C, Tiveron M-C, Bodmer R, Cremer H. Dynamics of Cux2 expression suggests that an early pool of SVZ precursors is fated to become upper cortical layer neurons. Cereb Cortex 2004;14(12):1408–20. [87] Cubelos B, Sebastián-Serrano A, Beccari L, Calcagnotto ME, Cisneros E, Kim S, et al. Cux1 and Cux2 regulate dendritic branching, spine morphology, and synapses of the upper layer neurons of the cortex. Neuron 2010;66(4): 523–35.