Our coordinate movements are controlled by individual muscles within specified circuitries of sensory neurons, interneurons and spinal motor neurons (Stafini, 2014). Motor neurons are classified into two groups, the upper motor neurons settle in the cerebral cortex, and the lower motor neurons located in the cranial nerve nuclei or spinal cord. The lower motor neurons commanded by the upper motor neurons link the central nervous system to the peripheral nervous system, the irreplaceable role of which has made it been studied most intensively over decades. The motor neuronal circuit, like the rest of the nervous system, is a spatially organized structure and the development of which requires a temporospatial regulation of molecular gradients, dynamic gene expression, cell adhesion and migration, nucleogenesis and axon guidance. It is believed that the identity and positioning of motor neurons are regulated by a complicated interplay of Hox transcription factors and its downstream effectors, which specifies cadherin expression to control motor pool characterization (Price et al., 2002; Demireva et al., 2011). Cadherins are a family of cell adhesion molecules which play important roles in motor neuron migration and motor pool establishment. Classical cadherins share a common binding partner called catenins which attach them to the actin filaments in the cell adhesion junctions (Hirohashi and Kanai, 2003). Nucleogenesis is a highly conserved feature that builds the topographic map of motor neurons aggregated in the ventral spinal cord to guarantee the functionally co-tuned neurons could receive equal sensory inputs (Surmeli et al., 2011). Outside the spinal cord, individual limb muscle innervates to the specific motor pool to organize a unique anatomical connection between motor neurons and their target organs. In this essay, I will discuss two papers that had advanced a fundamental discovery of cadherin function in our understanding of the development of motor neuronal circuitry, and how they pose further questions in the field. However, a dissection of the underlying molecular mechanisms and experimental protocol is beyond this argument. Compartmental nature of spinal motor neuron organization and target projectionNeurons grouped themselves with others that sharing the same function by either organized into stratified layers or forming clusters of nuclei by nucleogenesis (Pearson et al., 2014). Yet, whether an accurate topographic arrangement of neurons is prerequisite for proper muscle innervation remains largely unknown. Demireva first found that from embryonic to postnatal stages in mice, there is a broad co-expression of ?- and ?-catenin proteins in the spinal motor neurons. By genetically ablating both ?- and ?-catenin from spinal motor neurons using single or double conditional mutant Olig2::Cre line, they observed a disruption of motor column organization of median motor and preganglionic columns, and unequal divisions of lateral motor column neurons and motor pool segregation failure. Surprisingly, these perturbations of motor neurons have no influence on transcription factor expression and the neuronal projections toward their muscle target as illustrated in Figure1. Figure1. The activity of ?- and ?-catenin in motor pools sorting and muscle target mapping (adapted from Demireva et al., 2011). The motor pool clustering is cadherin/catenin signalling-dependent regardless of muscle target selection, while the outgrowth of motor neuron axons and effector muscle innervation rely on the transcriptional identities which are maintained in disrupted cadherin/catenin pathway. Since the innervation is independent of motor neuron positioning, it questions the reason of consuming energy on the topographic arrangement of motor neurons during development (Demireva et al., 2011) as well as the necessity of neuronal migration during spinal cord circuitry development (Kania, 2014; McArthur and Fetcho, 2017). As Demireva hypothesized, the topographic map of motor neurons is a representative of non-linear framework found exclusively in the central nervous system to avoid confounding the various inputs to the output pathways (Demireva et al., 2011). It then seems reasonable to reckon this topographic organization must possess additional functions to be evolutionarily conserved. Indeed, the motor pool sorting is believed to enhance neuromuscular connections by separating neurons that share the same muscle target to ensure a coherent neuron firing (Personius et al., 2007). Furthermore, as in the canonical reflex arc, afferent sensory neurons in the spinal cord link the efferent motor neurons via interneurons in between, the misplaced motor neurons could lead to position modification in sensory and interneurons. Evidence of this was claimed in the research by Surmeli et al. which reported that motor pool positioning is crucial for the establishment of functional sensory-motor connectivity since the monosynaptic associations between motor neurons and the proprioceptors of muscle formed in a topographic manner rather than on neuronal identity base (Surneli et al., 2011; Bikoff et al., 2016). Besides, the relationship between interneuron arrangement and motor neuron target connectivity was answered in 2016 by Bikoff et al. where they demonstrated there is no predictable link between interneuron positions and the muscle targets whereas the input connectivity seems to confine the interneuron organizations.Cadherin/catenin signalling in motor neuronal circuitry patterningDemireva and colleagues also found similar motor pools mispositioning phenotype can be attributed to the inactivation of N-cadherin, which strongly suggests a prominent requirement of both ?-catenin and ?-catenin in the cadherin/catenin signalling pathway. These findings correspond to the report that ?-catenin recruited to N-cadherin is essential for the contraction of smooth muscle and this protein-protein interaction is mediated by actin polymerization (Wang et al., 2015). Conspicuously, this cooperative function of ?-catenin and ?-catenin offers a potential explanation of the condition when a stronger influence on neuronal differentiation is produced by N-cadherin mutants over ?-catenin mutants alone (Kadowaki et al., 2007). However, combined with the previous study by Price et al., 2002 which implied that both N-cadherin and type II cadherins regulate motor neuron settling, another concern would be the possibility of protein compensation during this genetic manipulation. One could argue that the disruption of cadherin/catenin might result in compensational effects caused by the activation of other cell-cell interaction pathways such as Notch signalling pathway, which is also believed to play important roles in spinal cord development and regeneration (Losada et al., 2017). Moreover, notwithstanding a universally accepted fact that motor pool formation is only affected by cadherins (Luxey et al., 2015), it could be that type II cadherins are facilitating N-cadherin to achieve motor pool coalescence (Bello et al., 2012). Accordingly, the function of cadherins in motor neuron assembling remains oblique considering the diversity of cadherin family (Demireva et al., 2011; SDasen, 2017)Nevertheless, this discovery produces widespread implications in genetic manipulation of motor neuron topography. Given that cadherin/catenin signalling is a core event of cell adhesion in neuronal body assembling, researchers could infer that developmental processes which highly rely on cell adhesions like migration and nucleogenesis are also based on somatic positioning (Montage et al., 2017). Another intriguing question this finding has led to is whether cadherin/catenin signalling can alter the formation of neuromuscular junction. This was somehow answered since the overexpression of ?-catenin in muscles was found to produce an abnormal distribution of receptors on the pre- and post-synapse during neuromuscular junction development (Wu et al., 2012). Notably, the conditional elimination of catenins in chicken showed a much worse phenotype in terms of lateral migration, suggesting that the impact of cadherin/catenin manipulations could vary in different species due to different gene expression profiles (Bello et al., 2012). In addition to the biological conclusion, the fluorophore-conjugated secondary antibodies and retrograde labelling used in Demireva’s study has been applied in other research to tag spinal cord neurons, besides, the columnar mixing index (Cmi) developed in this study allows a general calculation of the neuronal mixing level between different columns (Mendelsohn et al., 2017; Hanley et al., 2017). Differential cadherin expression compiles motor neuron pool identity and organizationIt is believed that understanding of molecular profile in the rhomboideus motor pool is critical to explore the spinal motor neuronal organization (Nicholas Stafini, 2014). Despite knowing different expression profile of transcription factors determines each class of motor neurons and facilitates motor pool segregation (Shirasaki and Pfaff, 2002; Pattyn et al., 2003). It is still astonishing to find that nature utilizes a combination of differential expression of a protein family to identify cell types during the development. In the study of Astick et al., 2014, they found individuals of eight distinct motor nuclei in rhombomere 5 (r5) and rhombomere 8 (r8) are identified by a combinational expression of six type II cadherins (shown in Figure2). Figure2. The specific combination of cadherins expression in cranial nuclei during the hindbrain motor neuron development at rhombomere 5 and 8 (Adapted from Pearson et al., 2014). Somatic cranial motor neurons are Ab (abducens), AcAb (accessory abducens), ventral hypoglossal a (vXIIa), ventral hypoglossal b (vXIIb); branchiomotor/visceromotor neurons are dorsal facial nucleus (dFM), ventral facial motor (vFM), glossopharyngeal (IX), vagal (X), dorsal hypoglossal (dXII).They reported that somatomotor and branchiomotor nuclei would first undergo a period where they mingle together prior to their segregation regardless of their birthplace. How could single cell distinguish from each other only by having different proteins expressed? And how do they recognize each other in such an intermingled phase? There must be some protein-protein interactions happening inside. Indeed, cadherins are reported to interact with adjacent cells in a homophilic fashion to achieve heterogeneous cell sorting (Price et al., 2002; Shirabe et al., 2005; Patel et al., 2005). To investigate the requirement of cadherins during cranial motor neuron coalescence, Astick and colleagues conditionally expressed a negative form of cadherin N?390 in cranial motor neuron progenitors by electroporation to disrupt cadherin-mediated functions. It turned out that cranial motor neurons without cadherin activity failed to aggregate to cluster pools in r5 and r8, while cellular differentiation and migration remain unchanged. Following the identification of cadherin code, they further test whether the ablation or insertion of the unique cadherin in other type motor neurons would affect their coalescence. As expected, abnormal cadherin expressions led to coalescence failure but had no effect on neuron generation and differentiation (Astick et al., 2014).Most interestingly, these cadherin codes are compiled before nucleogenesis (Pearson et al., 2014). This begs the question of how and when exactly is this code established? Does it happen in the phase where the molecular identity of spinal motor neurons matures? If this step is under the control of certain transcription factors, then this combination code would provide a clear direction for future research of gene expression dynamics. Moreover, beyond this cadherin code, will there be a more general principle that controls cadherin function in motor neuron organization? Undeniably, apart from defining motor pool identities, type II cadherins expressions can also be found in sensory neurons, which implicates potential roles of cadherin in establishing the sensory-motor circuits during development (Price et al., 2002). This discovery also provides clinical implications for motor neuron regeneration therapies since that requires particular motor neurons with desired identities to join the original neuronal circuit at proper positions and to project to their target organs (Stafini, 2014). However, mechanisms like the biochemical interactions of proteins behind this code remain elusive. Given that the EC1 domain of cadherins plays a key role in cadherin functional activity (Patel et al., 2005, what kind of interactions could these different motor neurons perform to achieve this identity-based clustering? A recent study by Montague et al., 2017 observed spontaneous calcium activity being a concomitant to nucleogenesis process, raising the possibility that neural circuitry is also governed by spontaneous activity and cadherin networks interplay. ConclusionOverall, these two papers built a framework of how cell adhesion molecules achieve motor neuron sorting and topography establishment in the central nervous system. They also forward a subtle method for future experiments to manipulate motor neuron positions while keeping the predictable links between neurons and muscle innervation intact. The important role of catenins demonstrated here affords a new way to examine cadherin pathway functions instead of directly studying the diverse cadherin families. Nevertheless, unsolved questions such as the relationship between the mechanism that drives neural lamination and that of neuronal nucleogenesis (Price, 2012), and what changes would happen to interneurons and sensory neurons in the spinal cord when the motor neuronal cell bodies are mispositioned are still open to answers. Not to mention that the inhibitory interneurons in spinal motor circuits exhibit massive diversities in transcriptional factors, positions and physiologies. Such unique settings of distinct interneuron circuits facilitate their control of motor pools-muscle innervation as well as specifying neuronal inputs (Bikoff et al., 2016) and this diversity is acquired gradually through development. A further understanding of the mechanisms underpinning cadherin family interactions is required to better probe the functional connectivity throughout the development of the nervous system (Sotomayor et al., 2014).