- Research article
- Open Access
Characterization of the axon initial segment (AIS) of motor neurons and identification of a para-AIS and a juxtapara-AIS, organized by protein 4.1B
© Duflocq et al; licensee BioMed Central Ltd. 2011
- Received: 4 May 2011
- Accepted: 29 September 2011
- Published: 29 September 2011
The axon initial segment (AIS) plays a crucial role: it is the site where neurons initiate their electrical outputs. Its composition in terms of voltage-gated sodium (Nav) and voltage-gated potassium (Kv) channels, as well as its length and localization determine the neuron's spiking properties. Some neurons are able to modulate their AIS length or distance from the soma in order to adapt their excitability properties to their activity level. It is therefore crucial to characterize all these parameters and determine where the myelin sheath begins in order to assess a neuron's excitability properties and ability to display such plasticity mechanisms. If the myelin sheath starts immediately after the AIS, another question then arises as to how would the axon be organized at its first myelin attachment site; since AISs are different from nodes of Ranvier, would this particular axonal region resemble a hemi-node of Ranvier?
We have characterized the AIS of mouse somatic motor neurons. In addition to constant determinants of excitability properties, we found heterogeneities, in terms of AIS localization and Nav composition. We also identified in all α motor neurons a hemi-node-type organization, with a contactin-associated protein (Caspr)+ paranode-type, as well as a Caspr2+ and Kv1+ juxtaparanode-type compartment, referred to as a para-AIS and a juxtapara (JXP)-AIS, adjacent to the AIS, where the myelin sheath begins. We found that Kv1 channels appear in the AIS, para-AIS and JXP-AIS concomitantly with myelination and are progressively excluded from the para-AIS. Their expression in the AIS and JXP-AIS is independent from transient axonal glycoprotein-1 (TAG-1)/Caspr2, in contrast to juxtaparanodes, and independent from PSD-93. Data from mice lacking the cytoskeletal linker protein 4.1B show that this protein is necessary to form the Caspr+ para-AIS barrier, ensuring the compartmentalization of Kv1 channels and the segregation of the AIS, para-AIS and JXP-AIS.
α Motor neurons have heterogeneous AISs, which underlie different spiking properties. However, they all have a para-AIS and a JXP-AIS contiguous to their AIS, where the myelin sheath begins, which might limit some AIS plasticity. Protein 4.1B plays a key role in ensuring the proper molecular compartmentalization of this hemi-node-type region.
- Myelin Sheath
- Ventral Root
- Axon Initial Segment
- KCNQ2 Channel
- Immunofluorescence Intensity
Several studies have identified Nav and Kv isoforms that can be expressed specifically at the AIS: (i) Nav1.1, Nav1.2 and Nav1.6 [5–7], clustered at the AIS by ankyrin G (AnkG) [8, 9], a cytoskeleton-linked scaffolding protein expressed throughout all AISs ; (ii) Kv1.1 and Kv1.2 [2, 7, 11–15], as well as Kv7.2/KCNQ2 and Kv7.3/KCNQ3 [16–18].
While AnkG also clusters KCNQ2/3 at the AIS , the mechanisms controlling expression of Kv1 channels at the AIS remain unresolved. PSD-93 was shown to play a role, although it may be compensated for in vivo [15, 19]. Clustering of Kv1 channels has been mostly studied in juxtaparanodes, which are separated from nodes of Ranvier by paranodes: clustering of Kv1 channels requires the juxtaparanodal cell adhesion complex consisting of the axonal proteins transient axonal glycoprotein-1 (TAG-1) and contactin-associated protein-like 2 (Caspr2) [20–22], controlled by glial TAG-1 . The contiguous Caspr-expressing paranode is also necessary: it forms a barrier, which excludes Kv1 channels from paranodes and confines them to juxtaparanodes [24–27]. The cytoskeletal linker protein, protein 4.1B, which binds both Caspr and Caspr2 , might play a key role in stabilizing their expression in paranodes and juxtaparanodes, respectively, thus ensuring the proper compartmentalization of Kv1 channels. However, 4.1B has been shown to be necessary in juxtaparanodes only, controlling the distribution of Caspr2 and thus the expression of Kv1 channels .
Somatic motor neurons (MNs) constitute a functionally heterogeneous population of neurons that can be subdivided into different functional subgroups, the most prominent division being between α and γ MNs . Despite this functional heterogeneity, the above-mentioned AIS determinants of excitability properties have hardly been investigated [17, 31].
We have analyzed the AIS of somatic MNs and identified heterogeneous as well as homogeneous determinants of AIS spiking properties. We also provide, to our knowledge, the first molecular characterization of the axonal region directly following the AIS. We identified a paranode-like compartment and a juxtaparanode-like compartment, which support the beginning of the myelin sheath immediately after the AIS, and that, given the differences between nodes of Ranvier and AISs, we refer to as a 'para-AIS' and a 'juxtapara-AIS'. We also demonstrate that Kv1 channels appear at the AIS, para-AIS, and juxtapara-AIS concomitantly with myelination, and are progressively excluded from the para-AIS. We found that expression of Kv1 channels in the AIS and JXP-AIS is independent from TAG-1/Caspr2, in contrast to juxtaparanodes, and from PSD-93. Finally, we demonstrate that protein 4.1B is necessary to form the Caspr-expressing para-AIS barrier and to ensure the proper compartmentalization of Kv1 channels as well as the segregation of the AIS, para-AIS and juxtapara-AIS.
AIS localization and length in motor neurons
Axon initial segment (AIS) origin, length, distance from the soma and voltage-gated sodium (Nav) channel composition in α and γ motor neurons (MNs)
Distance from the soma (μm)
(n = 19)
29.5 ± 3.2
(n = 6)
16.9 ± 13.2
(n = 6)
(n = 18)
(n = 18)
(n = 61)
29.3 ± 5.4
(n = 26)
5.6 ± 3.8
(n = 26)
(n = 35)
(n = 35)
(n = 69)
28 ± 8.7
(n = 69)
8.1 ± 3.5
(n = 69)
(n = 43)
(n = 43)
We next measured the length of the AIS and its distance from the soma. We found that overall MN AISs had a very uniform length (30 ± 4.4 μm; n = 32; Table 1), which does not show statistically significant differences when comparing (for α MNs) dendrite-derived (29.5 ± 3.2 μm; n = 6) and soma-derived AISs (29.3 ± 5.4 μm; n = 26) or when comparing soma-derived AISs from α (29.3 ± 5.4 μm; n = 26) and γ MNs (28 μm ± 8.7; n = 69). The distance between the soma and the AIS showed a greater variability in particular for dendrite-derived AISs, whose distance from the soma was nonetheless longer in average (16.9 ± 13.2 μm; n = 6) than soma-derived AISs (for α MNs: 5.6 μm ± 3.8; n = 26; P < 0.05).
Nav channel distribution in MN AISs
We analyzed whether this new heterogeneity of AISs correlated with different functional types of MNs (α or γ) or with the first subdivision we observed into soma-derived or dendrite-derived MN AISs. We found that soma-derived AISs from α and γ MNs both display the two types of Nav composition in the same proportion (Table 1): about 80% of them express Nav1.1 and Nav1.6, while about 20% of them express only Nav1.6. As for dendrite-derived AISs (found only in 15% of α MNs), 100% of them express both Nav1.1 and Nav1.6 in two complementary AIS subcompartments (Table 1). Heterogeneous AISs in terms of localization and Nav channel composition can thus be found within a single neuronal population, and this heterogeneity differs among different functional types of MNs.
Kv channel distribution in MN AISs
Identification of a 'para-AIS'
Identification of a 'juxtapara-AIS'
In conclusion, all MNs have their AIS immediately flanked by a para-AIS and a JXP-AIS, which together strongly suggest the presence of a hemi-node type organization at the distal AIS and therefore the presence of a myelin sheath starting immediately after the AIS. However, given the differences between AISs and nodes of Ranvier, this hemi-node-type organization may display differences compared to a real hemi-node.
We investigated whether para-AISs and JXP-AISs can also be found in other neuronal types. In the cortex, we found, adjacent to the AIS, a Caspr+ para-AIS and a contiguous Kv1.1+ JXP-AIS in two neuronal types (see Additional file 1, Figure S1A-J), one of which being presumably a pyramidal cell (from the descending orientation and characteristic arrowhead-like morphology of its AIS: Additional file 1, Figure S1A-E). We also found a Caspr+ para-AIS and a Kv1.1+ JXP-AIS, adjacent to the AIS of Purkinje cells, which was surrounded by Kv1.1+ basket cell terminals (Additional file 1, Figure S1K-O).
Developmental expression of Kv1 channels at the MN AIS and JXP-AIS
At P1, we did not find any expression of Kv1.1 in AnkG+ AISs of Peripherin+ MNs (n = 46) or in their presumptive JXP-AISs (Figure 7A-D). At P3, 83.3% of MNs (n = 38) were still not expressing Kv1.1 in their AIS or JXP-AIS, while 16.7% showed a weak expression of Kv1.1 in their AnkG+ AIS, but did not yet express Caspr (as for the following stages/populations, a representative example is shown, in Figure 7E-H). At P3, expression of the myelin basic protein (MBP; a compact myelin marker) could not yet be detected (Figure 7A'-D'), which does not exclude the possibility that at this stage oligodendrocytes already start myelinating the axon of MNs at their distal AIS.
At P5, 67.3% of Peripherin+ MNs (n = 36) were expressing Kv1.1. Strikingly, at this stage, Kv1.1 was never found expressed in the AIS or the presumptive JXP-AIS alone, but was always found expressed in both compartments together. Moreover, as soon as Kv1.1 could be detected at MN AISs, its distribution appeared restricted to the distal AIS (Figure 7G), as in adult MN AISs (Figure 4H), suggesting that the molecules responsible for segregating Kv1.1 within the AIS are already present and active at this early stage. Among the MNs expressing Kv1.1 at P5, 16.8% expressed Kv1.1 at a low level in both AIS and JXP-AIS, but did not yet express Caspr in their presumptive para-AIS; while 50.5% of MNs expressed Kv1.1 at a low to moderate level in both AIS and JXP-AIS and also expressed Caspr at the para-AIS (Figure 7I-L). Interestingly, at this early stage of Kv1.1 expression in AIS and JXP-AIS, the distribution of AnkG overlapped with that of Caspr (Figure 7I, J), suggesting that the AIS and para-AIS are not yet properly segregated. Similarly, the distribution of Kv1.1 was not well restricted to the AIS and the JXP-AIS and often extended into the para-AIS, where it overlapped with Caspr distribution (Figure 7J-L, 7N-P), resembling the initial stages of Kv1.1 expression in sciatic nerve JXP-nodes, where Kv1.1 extended into the Caspr+ paranodal domain . In addition, at P5, MNs that express Kv1.1 also display MBP staining along their axon, adjacent to the AIS (Figure 7E'-H'). This demonstrates the presence of a compact myelin sheath, which must be attached to the axon by adjacent paranodal loops (not labeled here by the anti-MBP antibody, as is often the case for paranodal loops) apposed to the adjacent Caspr+ presumptive para-AIS. This result supports the previous conclusion that the para-AIS constitutes the first site of myelin attachment.
At P7, almost all MNs (92.2%, n = 37) now showed a significant expression level of Kv1.1 at both AIS and JXP-AIS, but AnkG and Kv1.1 were still overlapping with Caspr (Figure 7M-P). The remaining 7.8% of MNs expressed Kv1.1 at the AIS and JXP-AIS without Caspr. The segregation of the AIS, para-AIS and JXP-AIS into three mutually exclusive compartments occurred very progressively: at P14, 90% of MNs (n = 38) showed well segregated AnkG+, Caspr+ and Kv1.1+ compartments (Figure 7Q-T), similarly to Kv1.1 expression in sciatic nerve JXP-nodes, where it does not much extend any further within Caspr+ paranodes [19, 27]. MBP staining at P7 and P14 still reveals the presence of the myelin sheath adjacent to AIS (Figure 7I'-P').
The TAG-1/Caspr2 complex is not required for Kv1 channels expression at the MN AIS and JXP-AIS in contrast to MN JXP-nodes
What controls the clustering of Kv1 channels at the AIS and JXP-AIS? Concerning the AIS and JXP-nodes, two different mechanisms have been proposed, respectively.
At the AIS, the membrane-associated guanylate kinase (MAGUK) PSD-93 was shown, using extinction with small hairpin (sh)RNAs in cultured hippocampal neurons, to control the clustering of Kv1 channels , even though analysis in PSD-93-/- mice indicated that the lack of PSD-93 is compensated for . At JXP-nodes, the clustering of Kv1 channels relies on two processes: it first requires an interaction with the JXP-nodal TAG-1/Caspr2 complex: in the absence of JXP-nodal Caspr2 (in Caspr2-/- or TAG-1-/- mice; ) or TAG-1 (in TAG-1-/- mice; ), Kv1 channels failed to accumulate at JXP-nodes in sciatic and optic nerves. Second, paranodal axoglial junctions form a barrier that excludes Kv1 channels and restrict their clustering to JXP-nodes [24–27].
With regard to the clustering of Kv1 channels at the MN AIS, expression of Kv1.1 in TAG-1-/- mice displayed no statistically significant difference of immunofluorescence intensity compared to WT littermates (Figure 8B, E; respectively 32 and 15 MN AISs analyzed from 3 TAG-1-/- and 3 WT mice), as was shown at the AIS of hippocampal and cortical neurons . Thus, in contrast to JXP-nodes, the clustering of Kv1 channels at AISs, as at JXP-AISs, does not seem to require the TAG-1/Caspr2 complex in MNs.
PSD-93 is not required for Kv1 channels expression at the MN AIS and JXP-AIS
Of note, we found PSD-93 being also expressed in the JXP-AIS, again matching exactly with Kv1.1 distribution (Figure 9B, C). In contrast to ventral roots contributing to the sciatic nerve, where PSD-93 was expressed in a minority of MN peripheral JXP-nodes (about 10% of them ; data not shown), we found PSD-93 expression in 100% of MN JXP-AISs (53 MN JXP-AISs analyzed), suggesting again that spinal somatic MNs have a JXP-AIS that differs from their peripheral JXP-nodes. This is further illustrated by the fact that the TAG-1/Caspr2 complex did not appear to control PSD-93 distribution in JXP-AISs (nor at the AIS; see Additional file 3, Figure S3), in contrast to sciatic nerve JXP-nodes  or more specifically to MN peripheral JXP-nodes (data not shown): PSD-93 expression was normal in all MN JXP-AISs analyzed in TAG-1-/- mice (respectively 22 and 53 MN JXP-AISs analyzed from 3 TAG-1-/- and 5 WT mice; Additional file 3, Figure S3). However, despite these differences between MN JXP-AISs and peripheral JXP-nodes, in both compartments PSD-93 did not appear to control the clustering of Kv1 channels. As in sciatic nerve JXP-nodes  and more specifically in MN peripheral JXP-nodes from teased ventral roots (respectively 123 and 74 MN JXP-nodes analyzed from 3 PSD-93-/- and 3 WT mice; see Additional file 2, Figure S2), Kv1.1 distribution at MN JXP-AISs was not changed in PSD-93-/- mice as compared to WT littermate controls (Figure 9B, E). Kv1.1 at the MN JXP-AIS in PSD-93-/- mice displayed no statistically significant difference of immunofluorescence intensity compared to WT mice (Figure 9E; respectively 14 and 11 MN JXP-AISs analyzed from 3 PSD-93-/- and 3 WT mice). Thus, the clustering of Kv1 channels in MN JXP-AISs, as in MN AISs, did not require PSD-93.
Protein 4.1B is required to maintain a barrier at the MN para-AIS, necessary to cluster Kv1 channels and delimit the AIS, para-AIS and JXP-AIS
Finally, we tested the role of protein 4.1B in controlling both the hemi-node-like organization at the distal tip of the MN AIS and the clustering of Kv1 channels. 4.1 proteins are cytoskeletal proteins that play a key role in organizing membrane domains: they stabilize the membrane expression of a wide range of proteins by linking them to the actin/spectrin cytoskeleton . Of the four members of the protein 4.1 family, protein 4.1B is expressed in myelinated axons at paranodes and JXP-nodes, where it binds Caspr and Caspr2, respectively [28, 42]. Protein 4.1B has been shown to be necessary in JXP-nodes, for Caspr2 and thus Kv1 channels expression .
At the MN JXP-AIS, Kv1.1 expression was diminished in 4.1B-/- mice; its expression level in the JXP-AIS was not any higher than in the distal AIS, as in WT mice (Figure 10C, F); instead, in most cases it became even lower (Figure 10I, L, O). PSD-93 and Caspr2 were found to display the same distribution as Kv1.1: in contrast to WT mice (Figures 9C and 8C, respectively), PSD-93 and Caspr2 expression in 4.1B-/- mice also invaded the MN para-AIS and displayed a continuous distribution from the distal AIS to the JXP-AIS (Figure 10M-O and 10P-R, respectively).
Heterogeneous AISs within the single neuronal population of MNs
The spiking properties of a neuron depend on its AIS ion channel composition, length and distance from the soma [1–4]. We have shown that within a single neuronal population, the somatic lumbar MNs, AISs can be heterogeneous both in terms of ion channel composition and localization, suggesting that the AIS may contribute to defining different functional cell types within a single neuronal population.
Finally, AISs of MNs also differ with respect to their Nav channel composition, expressing either Nav1.6 only, or both Nav1.1 and Nav1.6 (Figure 11A). Nav1.1 may fulfill a similar role as Nav1.2 in layer 5 pyramidal neurons: due to their lower activation threshold, distal AIS Nav1.6 channels initiate action potentials, while higher threshold proximal AIS Nav1.2 channels ensure their backpropagation towards the soma . This scenario was based on a comparison between Nav1.2 and Nav1.6 activation thresholds in neurons . Such a comparison between Nav1.1 and Nav1.6 would help supporting a similar scenario for MNs. The higher impedance mismatch, that action potentials backpropagating from dendrite-derived AISs might have to overcome due to the larger proximal dendritic compartment, could explain why such AISs in MNs always express Nav1.1 in their proximal part.
These heterogeneous AISs, in terms of localization and Nav channel composition, may thus confer different excitability properties and contribute to define subsets of MNs with different firing properties . Since intracellular recordings of adult mouse spinal MNs can now be performed [50, 51], it will be possible to analyze the functional impact of these variable AIS parameters.
Of note, we found some AIS parameters that were instead constant across MNs, and may provide MNs with a common set of firing properties. This is the case for Kv channels distribution, even though regulatory mechanisms (for example, phosphorylation status, binding to interacting partners and in particular homo/heteromultimerization, binding to β subunits) could bring about differences between MNs. First, expression of slow subthreshold M-type K+ current-mediating KCNQ2 channels , along the entire AIS of all α MNs (Figure 11A), might serve to control both the resting membrane potential and the action potential threshold, as they do in CA1 hippocampal pyramidal neurons . Second, shaker-like Kv1.1 and Kv1.2 channels as well as the auxiliary Kvβ2 subunit were found to be colocalized at the distal AIS in all α MNs (Figure 11A), suggesting that they coassemble into heteromultimeric channels, as in juxtaparanodes . These AIS Kv1 channels may function in MNs, as in layer 5 pyramidal neurons, to regulate the duration of action potentials  or, as in neocortical fast-spiking γ-aminobutyric (GABA)ergic interneurons, to control the action potential threshold and thus the firing rate . Finally, AIS length was also rather constant among MNs, yet slightly different from other neuronal types (data not shown), suggesting that mechanisms must exist to regulate AIS length in a neuronal type-dependent manner.
Identification of a para-AIS and a juxtapara-AIS
We chose to use here the terms 'para-AIS' and 'JXP-AIS' (instead of 'first paranode' and 'first JXP-node' for instance) for several reasons: these compartments abut an AIS, which displays important differences compared to a node of Ranvier; in addition these differences may well cause the para-AIS and JXP-AIS to differ from paranodes and JXP-nodes. Despite similarities, AISs and nodes indeed differ in terms of molecular composition (for instance, Kv1 channels and their binding partners Caspr2, TAG-1, PSD-93 and A Disintegrin And Metalloproteinase 22 (ADAM22) are expressed in AISs only), which implies different clustering/stabilizing mechanism. In addition, in contrast to nodes, AISs develop long before myelination. Thus, the mechanisms that cluster axonal proteins at the para-AIS and JXP-AIS upon myelination, at the distal tip of the preorganized AIS, might well differ from those operating at paranodes and JXP-nodes. Our results with TAG-1 -/- mice may illustrate such a difference.
The site where the axon starts being myelinated is still an open question for many neuronal populations, including mouse MNs: some chicken spinal MNs have a myelin sheath starting immediately after the AIS , while cat visceral preganglionic MNs have a myelin sheath starting several hundred microns after the AIS . Our findings show that the axon of mouse somatic α MNs starts being myelinated immediately after the AIS and displays a hemi-node-type organization contiguous to the AIS, with a para-AIS and a JXP-AIS. The presence of the myelin sheath immediately after the AIS will allow action potentials to start their high velocity propagation along the axon and synchronously along its collaterals right from their site of initiation. In addition, it might change the ability of the AIS to increase either its length or its distance from the soma as a way for the neuron to adapt its excitability in response to changes in neuronal activity [3, 4]. If MNs are able to display such homeostatic plasticity mechanisms, these might take place more easily during a critical period or during developmental stages preceding myelination, when MNs might need to adjust their intrinsic excitability to their changing synaptic inputs. Finally, given the presence of the myelin sheath immediately after the AIS (and the expression of Kv1 channels at the JXP-AIS), it is conceivable that demyelinating conditions may affect AIS spiking properties.
Mechanisms that control Kv1 channel distribution in the AIS and JXP-AIS
The homogeneous distribution of KCNQ2 channels along the entire AIS, matching that of AnkG, reflects their AnkG-dependent clustering . The mechanisms that segregate Nav channels in the AIS proximal (Nav1.1) or distal (Nav1.6) subcompartment, in addition to their AnkG-mediated membrane clustering, are currently unknown. Nav auxiliary β subunits (β1-4), which regulate Nav channel cell surface expression, interact with cell adhesion molecules and components of the extracellular matrix , could be potential candidates.
As for Kv1.1, Kv1.2 and Kvβ2, what controls their restricted expression in the distal AIS and in the newly characterized JXP-AIS? In TAG-1 -/- mice, their normal expression in both compartments indicates that TAG-1/Caspr2-independent mechanisms can control the distribution of Kv1 channels in the AIS and JXP-AIS, in contrast to the commonly accepted mechanism operating in JXP-nodes. Is this difference due to the presence of PSD-93 in all AISs and JXP-AISs (in contrast to only 7% of JXP-nodes in MNs ), and the fact that in both compartments PSD-93 distribution is not controlled by TAG-1/Caspr2 either (in contrast to JXP-nodes )?
PSD-93 was indeed shown, with shRNAs, to control Kv1 channels in the AIS of cultured hippocampal neurons . However, Kv1 channels distribution was normal in the AIS of PSD-93 -/- mice . The normal distribution of Kv1 channels found also in the AIS and JXP-AIS of PSD-93 -/- mice leaves open the possibility that PSD-93 plays a role in vivo: its absence in KO mice may lead to compensatory mechanisms , possibly mediated by other MAGUKs.
These results together with those from TAG-1 -/- mice support the existence of multifactorial mechanisms that can compensate for the absence of either PSD-93 or TAG-1/Caspr2 in the AIS and JXP-AIS (Figure 11B), but also in some JXP-nodes, where Kv1 channels expression was normal in Caspr2-/- mice [19, 56]. In Caspr2 -/- and TAG-1 -/- mice, the normal distribution of protein 4.1B respectively in JXP-nodes  and JXP-AISs (data not shown) could stabilize the expression of scaffolding proteins containing a PDZ domain and a protein 4.1-binding motif. Such scaffolds could allow PDZ-mediated recruitment of Kv1 channels. MPP2 and MPP6, members of the Membrane Palmitoylated Protein (MPP) family of MAGUKs , which can also bind Caspr2 , could play such a role and compensate for the absence of TAG-1/Caspr2 or PSD-93. Nectin-like proteins (Necls), which can bind PDZ-containing proteins (including MPP members ) and 4.1 proteins  may also contribute to such compensatory mechanisms. In AISs, other proteins may fulfill a similar role as protein 4.1B, or other mechanisms might exist to recruit Kv1 channels.
Protein 4.1B is necessary to maintain a barrier at the para-AIS that compartmentalizes Kv1 channels and delimits the AIS, para-AIS and JXP-AIS
Recruitment of Kv1 channels at the AIS appears concomitantly with myelination and the formation of axoglial contacts at the AIS distal tip, as revealed by Caspr expression. Myelination may thus trigger the recruitment of Kv1 channels at the AIS, which may in part serve to adapt spike initiation properties to the new saltatory propagation and concomitant changes in firing properties .
Clustering of Caspr in the presumptive para-AIS indicates the formation of a membrane barrier, which progressively excludes Kv1 channels from this compartment and restricts their distribution to the AIS and JXP-AIS. The stabilization of the Caspr-expressing barrier is thus crucial for controlling the final distribution of Kv1 channels. Protein 4.1B, by linking Caspr to the cytoskeleton in the para-AIS, may play such a crucial role (Figure 11B): in its absence, Caspr and Kv1 channels distribution recapitulates an early developmental stage (such as P7), when Caspr is not yet properly clustered and does not form a complete barrier able to exclude all (JXP-AIS and/or AIS) Kv1 channels from the presumptive para-AIS. In 4.1B-/- mice, some Kv1 channels are found in the JXP-AIS (probably resulting from a weak barrier function at the para-AIS), where they might still be stabilized by 4.1B-independent mechanisms, as in the AIS. These results differ from the normal appearance of Caspr+ peripheral paranodes observed in other 4.1B-/- mice , and suggest that the lack of protein 4.1B in the para-AIS is not compensated for by another 4.1 protein able to maintain a normal Caspr distribution (two new studies have analyzed the role of protein 4.1B in paranodes: one was published while our manuscript was under review , the other one is in press ).
Of note, the protein 4.1B-mediated consolidation of the Caspr-expressing barrier at the para-AIS ensures the proper segregation of Caspr not only from Kv1 channels, but also from AnkG: in 4.1B -/- mice, AnkG overlaps with Caspr (as at P7). Protein 4.1B thus plays a key role in controlling the hemi-nodal organization at the AIS distal tip and ensures the proper delimitation of the three compartments, the AIS, para-AIS and JXP-AIS.
AISs from α MNs are homogeneous in terms of Kv1 and Kv7 channels composition but are heterogeneous in terms of axonal versus dendritic origin and Nav channel composition (Figure 11A). These heterogeneities suggest an important role for the AIS in defining subsets of motor neurons with different spiking properties. Despite these heterogeneities, all α MNs have their AIS immediately followed by a Caspr+ paranode-type and a Caspr2+ and Kv1 channels+ JXP-node-type compartment (Figure 11A), which underlie the beginning of the myelin sheath and which might limit plasticity of AIS length or distance from the soma. Differences are observed at this first myelin attachment site compared with a hemi-node of Ranvier: in particular, Kv1 channels appear to be clustered in the JXP-AIS and AIS by TAG-1/Caspr2-independent mechanisms, while the cytoskeletal linker, protein 4.1B plays a crucial role in controlling the molecular organization of this region (Figure 11B): it is necessary to form the Caspr-expressing barrier at the para-AIS and to ensure the proper compartmentalization of Kv1 channels and the segregation of the AIS, para-AIS and JXP-AIS.
OF1 adult and postnatal mice (obtained from Charles River; L'Arbresle, France) were housed under standard laboratory conditions. PSD93-null, TAG-1-null and 4.1B-null mice have already been described [38, 40, 62]. All animal experiments were performed in compliance with European Community guiding principles on the care and use of animals (86/609/EEC, EC off. J. no. L358, 18 December 1986), the French decree no. 97/748 of October 19, 1987 (J. Off. République Française, 20 October 1987) and recommendations from the CNRS and Université Pierre et Marie Curie.
The following antibodies were used: anti-Nav1.1 rabbit polyclonal (AB5204, Millipore; Molsheim, France); anti-Nav1.6 mouse monoclonal (clone K87A/10, UC Davis/NIH NeuroMab Facility; Davis, CA, USA); anti-Peripherin rabbit polyclonal (AB1530, Millipore); anti-Kv1.1 mouse monoclonal (clones K20/78 and K36/15, UC Davis/NIH NeuroMab Facility); anti-Kv1.2 mouse monoclonal (clone K14/16, UC Davis/NIH NeuroMab Facility); anti-Kvβ2 mouse monoclonal (clone K17/70, UC Davis/NIH NeuroMab Facility); anti-Caspr (Paranodin) mouse monoclonal (clone K65/35, UC Davis/NIH NeuroMab Facility); anti-Caspr2 rabbit polyclonal (C8737, Sigma-Aldrich; Lyon, France); anti-Chapsyn-110 (PSD-93) mouse monoclonal (clone N18/30, UC Davis/NIH NeuroMab Facility), anti-MAP2 mouse monoclonal (M4403, Sigma-Aldrich), anti-Errγ. Err3) mouse monoclonal (H6812, PPMX; Tokyo, Japan), anti-NeuN mouse monoclonal (MAB377, Millipore) and anti-MBP rat monoclonal (MAB386, Millipore) antibodies. We also generated anti-KCNQ2 guinea pig polyclonal antibodies directed against residues 16-37 and 398-420 of human KCNQ2 and an anti-AnkG guinea pig polyclonal antibody directed against residues 1, 633-1, 650 of human AnkG. The anti-protein 4.1B antibody was a generous gift from Dr Laurence Goutebroze (INSERM UMR-S 839; Paris, France).
Alexa 405/Alexa 488/Alexa 594-conjugated secondary antibodies (Invitrogen; Villebon sur Yvette, France) were used to detect rabbit polyclonal or mouse monoclonal primary antibodies. Alexa 594-conjugated or Cy5-conjugated secondary antibodies (Jackson ImmunoResearch; Newmarket, Suffolk, UK) were used to detect guinea pig polyclonal primary antibodies.
Adult mice were deeply anesthetized and fixed by intracardiac perfusion with 20 ml of 1% paraformaldehyde (PFA) freshly prepared in phosphate-buffered saline (PBS, pH 7.4) at 4°C. Lumbar spinal cords with their dorsal and ventral roots and brains were then immediately dissected out, further immersion fixed for 1 h in 1% PFA at 4°C, rinsed in PBS and cryoprotected by overnight immersion in 20% sucrose in PBS at 4°C. Spinal cords and brains from P3 to P21 were directly dissected out from deeply anesthetized mice, immersion fixed for 1 h in 1% PFA at 4°C and then rinsed and cryoprotected as adult tissues. Tissue samples were then frozen in OCT medium (VWR; Fontenay-sous-bois, France) and 20 μm cryosections were collected onto Superfrost Plus slides (VWR), and stored at -20°C. Ventral root sciatic nerves from spinal L3-L6 segments were desheated and teased with fine forceps on Superfrost Plus slides, air dried for 2 h and also stored at -20°C. Both types of slides were first thawed at room temperature (RT), washed in Tris-buffered saline (TBS), and incubated for 1 h in a blocking solution (10% goat serum in TBS) with 0.4% Triton X-100. They were then incubated overnight at 4°C with primary antibodies, diluted in the blocking solution with 0.2% Triton X-100, washed in TBS and incubated again for 2 h with secondary antibodies, diluted in the blocking solution with 0.2% Triton X-100. After washing in TBS, slides were dried out and mounted in Mowiol medium (Merk; Lyon, France). When primary antibodies generated in the mouse were used, prior to the primary antibody incubation step, slides were incubated for 3 h in goat anti-mouse F(ab)'2 (Jackson ImmunoResearch; Newmarket, Suffolk, UK), diluted in the blocking solution with 0.2% Triton X-100 and washed in TBS. This supplementary step reduced non-specific fixation of anti-mouse secondary antibodies onto mouse tissue.
When two rabbit-generated primary antibodies were combined in the same experiment, these were used sequentially: a first immunohistochemistry experiment was performed with one of them and after the incubation with its secondary antibody and washes, slides were incubated for 3 h in goat anti-rabbit F(ab)'2 (Jackson ImmunoResearch), diluted in TBS. Slides were then washed in TBS, incubated overnight at 4°C with the second rabbit-generated primary antibody, and processed normally until the end of this second immunohistochemistry. An identical experiment with the two primary antibodies used in the opposite order was carried out to confirm the absence of crossreactivity.
Images were acquired using a fluorescence microscope equipped with an Apotome module (Axiovert 200 M, Zeiss; Le Pecq, France) and a 63 × objective with non-saturating exposure times. Images were processed with the NIH ImageJ software (Bethesda, MD, USA). AIS length was measured with the 'Simple neurite tracer' plugin. Each figure corresponds to a projection image from a z stack of 0.6 μm distant optical sections, limited to the object of interest. Immunofluorescence intensity profiles were obtained with ImageJ, from a thick line (almost as large as the axon) traced along the axon segment, on a stack-of-interest projected image. For each antibody staining, the background intensity was first subtracted, and immunofluorescence intensity values along each axon segment were then normalized relative to the maximum intensity found along that segment. For each antibody, the mean intensity ± SEM from several axon segments is shown. In Figures 3 and 4, given the variable length of AISs and the consequent difficulty to align immunostaining profiles from several AISs, the positions along each AIS were first normalized relative to the length of that AIS. For Figure 10S, T, given the additional variability in the length of para-AISs and JXP-AISs, profiles were instead aligned at the end of the AIS, defined as the position along the axon where half of the maximum drop of AnkG immunostaining intensity (over 2 μm) is reached. In Figure 10U, the start and end positions of both AnkG and Caspr were similarly defined as the positions where half the maximum change of the respective immunostaining (over 2 μm) is reached. As for the analysis of Kv1.1 immunofluorescence intensity at the AIS and juxtapara-AIS in WT, TAG-1-/- and PSD-93-/- mice, the mean fluorescence intensity was measured with ImageJ on a selection square adjusted to the surface of the distal AIS and of the juxtapara-AIS, respectively. WT and mutant tissue sections were collected together on the same slide, in order for them to be processed in the same immunostaining experiment, thus with identical immunostaining conditions, and were analyzed in identical acquisition conditions (with the same exposure time).
The number of AISs analyzed for each figure (as indicated in the Results section) corresponds to the total number of AISs obtained from at least five animals, each analyzed in at least six independent experiments. We tested statistical significance using a Student's unpaired t test, and values are presented as mean ± SEM. For the comparison of distances from the soma to the AIS between soma-derived and dendrite-derived AISs, as well as for the comparison of AIS length and Caspr position between WT and 4.1B -/- mice, a Wilcoxon/Mann-Whitney test was used.
This work was supported by the Association Française contre les Myopathies (AFM), and the Fondation pour la Recherche Médicale (FRM). We thank Drs David Bredt and Kazutada Watanabe for generously providing PSD93-null and TAG-1-null mice, respectively, Erika Bullier and Philippe Mailly for help with genotyping and image analysis, respectively, Dr Laurence Goutebroze for generously providing the anti-protein 4.1B antibody, and Drs Catherine Faivre-Sarrailh, Laurence Goutebroze and Domna Karagogeos for critical reading of the manuscript.
- Bean BP: The action potential in mammalian central neurons. Nat Rev Neurosci. 2007, 8: 451-465.View ArticlePubMedGoogle Scholar
- Kuba H, Ishii TM, Ohmori H: Axonal site of spike initiation enhances auditory coincidence detection. Nature. 2006, 444: 1069-1072. 10.1038/nature05347.View ArticlePubMedGoogle Scholar
- Grubb MS, Burrone J: Activity-dependent relocation of the axon initial segment fine-tunes neuronal excitability. Nature. 2010, 465: 1070-1074. 10.1038/nature09160.PubMed CentralView ArticlePubMedGoogle Scholar
- Kuba H, Oichi Y, Ohmori H: Presynaptic activity regulates Na(+) channel distribution at the axon initial segment. Nature. 2010, 465: 1075-1078. 10.1038/nature09087.View ArticlePubMedGoogle Scholar
- Boiko T, Van Wart A, Caldwell JH, Levinson SR, Trimmer JS, Matthews G: Functional specialization of the axon initial segment by isoform-specific sodium channel targeting. J Neurosci. 2003, 23: 2306-2313.PubMedGoogle Scholar
- Jenkins SM, Bennett V: Ankyrin-G coordinates assembly of the spectrin-based membrane skeleton, voltage-gated sodium channels, and L1 CAMs at Purkinje neuron initial segments. J Cell Biol. 2001, 155: 739-746. 10.1083/jcb.200109026.PubMed CentralView ArticlePubMedGoogle Scholar
- Van Wart A, Trimmer JS, Matthews G: Polarized distribution of ion channels within microdomains of the axon initial segment. J Comp Neurol. 2007, 500: 339-352. 10.1002/cne.21173.View ArticlePubMedGoogle Scholar
- Garrido JJ, Giraud P, Carlier E, Fernandes F, Moussif A, Fache MP, Debanne D, Dargent B: A targeting motif involved in sodium channel clustering at the axonal initial segment. Science. 2003, 300: 2091-2094. 10.1126/science.1085167.View ArticlePubMedGoogle Scholar
- Lemaillet G, Walker B, Lambert S: Identification of a conserved ankyrin-binding motif in the family of sodium channel alpha subunits. J Biol Chem. 2003, 278: 27333-27339. 10.1074/jbc.M303327200.View ArticlePubMedGoogle Scholar
- Kordeli E, Lambert S, Bennett V: AnkyrinG. A new ankyrin gene with neural-specific isoforms localized at the axonal initial segment and node of Ranvier. J Biol Chem. 1995, 270: 2352-2359. 10.1074/jbc.270.5.2352.View ArticlePubMedGoogle Scholar
- Inda MC, DeFelipe J, Munoz A: Voltage-gated ion channels in the axon initial segment of human cortical pyramidal cells and their relationship with chandelier cells. Proc Natl Acad Sci USA. 2006, 103: 2920-2925. 10.1073/pnas.0511197103.PubMed CentralView ArticlePubMedGoogle Scholar
- Kole MH, Letzkus JJ, Stuart GJ: Axon initial segment Kv1 channels control axonal action potential waveform and synaptic efficacy. Neuron. 2007, 55: 633-647. 10.1016/j.neuron.2007.07.031.View ArticlePubMedGoogle Scholar
- Lorincz A, Nusser Z: Cell-type-dependent molecular composition of the axon initial segment. J Neurosci. 2008, 28: 14329-14340. 10.1523/JNEUROSCI.4833-08.2008.PubMed CentralView ArticlePubMedGoogle Scholar
- Goldberg EM, Clark BD, Zagha E, Nahmani M, Erisir A, Rudy B: K+ channels at the axon initial segment dampen near-threshold excitability of neocortical fast-spiking GABAergic interneurons. Neuron. 2008, 58: 387-400. 10.1016/j.neuron.2008.03.003.PubMed CentralView ArticlePubMedGoogle Scholar
- Ogawa Y, Horresh I, Trimmer JS, Bredt DS, Peles E, Rasband MN: Postsynaptic density-93 clusters Kv1 channels at axon initial segments independently of Caspr2. J Neurosci. 2008, 28: 5731-5739. 10.1523/JNEUROSCI.4431-07.2008.PubMed CentralView ArticlePubMedGoogle Scholar
- Devaux JJ, Kleopa KA, Cooper EC, Scherer SS: KCNQ2 is a nodal K+ channel. J Neurosci. 2004, 24: 1236-1244. 10.1523/JNEUROSCI.4512-03.2004.View ArticlePubMedGoogle Scholar
- Pan Z, Kao T, Horvath Z, Lemos J, Sul JY, Cranstoun SD, Bennett V, Scherer SS, Cooper EC: A common ankyrin-G-based mechanism retains KCNQ and NaV channels at electrically active domains of the axon. J Neurosci. 2006, 26: 2599-2613. 10.1523/JNEUROSCI.4314-05.2006.View ArticlePubMedGoogle Scholar
- Chung HJ, Jan YN, Jan LY: Polarized axonal surface expression of neuronal KCNQ channels is mediated by multiple signals in the KCNQ2 and KCNQ3 C-terminal domains. Proc Natl Acad Sci USA. 2006, 103: 8870-8875. 10.1073/pnas.0603376103.PubMed CentralView ArticlePubMedGoogle Scholar
- Ogawa Y, Oses-Prieto J, Kim MY, Horresh I, Peles E, Burlingame AL, Trimmer JS, Meijer D, Rasband MN: ADAM22, a Kv1 channel-interacting protein, recruits membrane-associated guanylate kinases to juxtaparanodes of myelinated axons. J Neurosci. 2010, 30: 1038-1048. 10.1523/JNEUROSCI.4661-09.2010.PubMed CentralView ArticlePubMedGoogle Scholar
- Traka M, Goutebroze L, Denisenko N, Bessa M, Nifli A, Havaki S, Iwakura Y, Fukamauchi F, Watanabe K, Soliven B, Girault JA, Karagogeos D: Association of TAG-1 with Caspr2 is essential for the molecular organization of juxtaparanodal regions of myelinated fibers. J Cell Biol. 2003, 162: 1161-1172. 10.1083/jcb.200305078.PubMed CentralView ArticlePubMedGoogle Scholar
- Poliak S, Salomon D, Elhanany H, Sabanay H, Kiernan B, Pevny L, Stewart CL, Xu X, Chiu SY, Shrager P, Furley AJ, Peles E: Juxtaparanodal clustering of Shaker-like K+ channels in myelinated axons depends on Caspr2 and TAG-1. J Cell Biol. 2003, 162: 1149-1160. 10.1083/jcb.200305018.PubMed CentralView ArticlePubMedGoogle Scholar
- Savvaki M, Panagiotaropoulos T, Stamatakis A, Sargiannidou I, Karatzioula P, Watanabe K, Stylianopoulou F, Karagogeos D, Kleopa KA: Impairment of learning and memory in TAG-1 deficient mice associated with shorter CNS internodes and disrupted juxtaparanodes. Mol Cell Neurosci. 2008, 39: 478-490. 10.1016/j.mcn.2008.07.025.View ArticlePubMedGoogle Scholar
- Savvaki M, Theodorakis K, Zoupi L, Stamatakis A, Tivodar S, Kyriacou K, Stylianopoulou F, Karagogeos D: The expression of TAG-1 in glial cells is sufficient for the formation of the juxtaparanodal complex and the phenotypic rescue of tag-1 homozygous mutants in the CNS. J Neurosci. 2010, 30: 13943-13954. 10.1523/JNEUROSCI.2574-10.2010.View ArticlePubMedGoogle Scholar
- Dupree JL, Girault JA, Popko B: Axo-glial interactions regulate the localization of axonal paranodal proteins. J Cell Biol. 1999, 147: 1145-1152. 10.1083/jcb.147.6.1145.PubMed CentralView ArticlePubMedGoogle Scholar
- Bhat MA, Rios JC, Lu Y, Garcia-Fresco GP, Ching W, St Martin M, Li J, Einheber S, Chesler M, Rosenbluth J, Salzer JL, Bellen HJ: Axon-glia interactions and the domain organization of myelinated axons requires neurexin IV/Caspr/Paranodin. Neuron. 2001, 30: 369-383. 10.1016/S0896-6273(01)00294-X.View ArticlePubMedGoogle Scholar
- Boyle ME, Berglund EO, Murai KK, Weber L, Peles E, Ranscht B: Contactin orchestrates assembly of the septate-like junctions at the paranode in myelinated peripheral nerve. Neuron. 2001, 30: 385-397. 10.1016/S0896-6273(01)00296-3.View ArticlePubMedGoogle Scholar
- Poliak S, Gollan L, Salomon D, Berglund EO, Ohara R, Ranscht B, Peles E: Localization of Caspr2 in myelinated nerves depends on axon-glia interactions and the generation of barriers along the axon. J Neurosci. 2001, 21: 7568-7575.PubMedGoogle Scholar
- Denisenko-Nehrbass N, Oguievetskaia K, Goutebroze L, Galvez T, Yamakawa H, Ohara O, Carnaud M, Girault JA: Protein 4.1B associates with both Caspr/paranodin and Caspr2 at paranodes and juxtaparanodes of myelinated fibres. Eur J Neurosci. 2003, 17: 411-416. 10.1046/j.1460-9568.2003.02441.x.View ArticlePubMedGoogle Scholar
- Horresh I, Bar V, Kissil JL, Peles E: Organization of myelinated axons by Caspr and Caspr2 requires the cytoskeletal adapter protein 4.1B. J Neurosci. 2010, 30: 2480-2489. 10.1523/JNEUROSCI.5225-09.2010.PubMed CentralView ArticlePubMedGoogle Scholar
- Burke R: Spinal cord: Ventral horn. The Synaptic Organization of the Brain. Edited by: Shepherd GM. 2004, New York, USA: Oxford University Press, 79-123.View ArticleGoogle Scholar
- Duflocq A, Le Bras B, Bullier E, Couraud F, Davenne M: Nav1.1 is predominantly expressed in nodes of Ranvier and axon initial segments. Mol Cell Neurosci. 2008, 39: 180-192. 10.1016/j.mcn.2008.06.008.View ArticlePubMedGoogle Scholar
- Greene LA: A new neuronal intermediate filament protein. Trends Neurosci. 1989, 12: 228-230. 10.1016/0166-2236(89)90127-6.View ArticlePubMedGoogle Scholar
- Houser CR, Crawford GD, Barber RP, Salvaterra PM, Vaughn JE: Organization and morphological characteristics of cholinergic neurons: an immunocytochemical study with a monoclonal antibody to choline acetyltransferase. Brain Res. 1983, 266: 97-119. 10.1016/0006-8993(83)91312-4.View ArticlePubMedGoogle Scholar
- Rasband MN: The axon initial segment and the maintenance of neuronal polarity. Nat Rev Neurosci. 2010, 11: 552-562. 10.1038/nrn2852.View ArticlePubMedGoogle Scholar
- Friese A, Kaltschmidt JA, Ladle DR, Sigrist M, Jessell TM, Arber S: Gamma and alpha motor neurons distinguished by expression of transcription factor Err3. Proc Natl Acad Sci USA. 2009, 106: 13588-13593. 10.1073/pnas.0906809106.PubMed CentralView ArticlePubMedGoogle Scholar
- Rhodes KJ, Strassle BW, Monaghan MM, Bekele-Arcuri Z, Matos MF, Trimmer JS: Association and colocalization of the Kvbeta1 and Kvbeta2 beta-subunits with Kv1 alpha-subunits in mammalian brain K+ channel complexes. J Neurosci. 1997, 17: 8246-8258.PubMedGoogle Scholar
- Vabnick I, Trimmer JS, Schwarz TL, Levinson SR, Risal D, Shrager P: Dynamic potassium channel distributions during axonal development prevent aberrant firing patterns. J Neurosci. 1999, 19: 747-758.PubMedGoogle Scholar
- Fukamauchi F, Aihara O, Wang YJ, Akasaka K, Takeda Y, Horie M, Kawano H, Sudo K, Asano M, Watanabe K, Iwakura Y: TAG-1-deficient mice have marked elevation of adenosine A1 receptors in the hippocampus. Biochem Biophys Res Commun. 2001, 281: 220-226. 10.1006/bbrc.2001.4334.View ArticlePubMedGoogle Scholar
- Horresh I, Poliak S, Grant S, Bredt D, Rasband MN, Peles E: Multiple molecular interactions determine the clustering of Caspr2 and Kv1 channels in myelinated axons. J Neurosci. 2008, 28: 14213-14222. 10.1523/JNEUROSCI.3398-08.2008.PubMed CentralView ArticlePubMedGoogle Scholar
- McGee AW, Topinka JR, Hashimoto K, Petralia RS, Kakizawa S, Kauer FW, Aguilera-Moreno A, Wenthold RJ, Kano M, Bredt DS: PSD-93 knock-out mice reveal that neuronal MAGUKs are not required for development or function of parallel fiber synapses in cerebellum. J Neurosci. 2001, 21: 3085-3091.PubMedGoogle Scholar
- Baines AJ: The spectrin-ankyrin-4.1-adducin membrane skeleton: adapting eukaryotic cells to the demands of animal life. Protoplasma. 2010, 244: 99-131. 10.1007/s00709-010-0181-1.View ArticlePubMedGoogle Scholar
- Gollan L, Sabanay H, Poliak S, Berglund EO, Ranscht B, Peles E: Retention of a cell adhesion complex at the paranodal junction requires the cytoplasmic region of Caspr. J Cell Biol. 2002, 157: 1247-1256. 10.1083/jcb.200203050.PubMed CentralView ArticlePubMedGoogle Scholar
- Conradi S, Skoglund S: Observations on the ultrastructure of the initial motor axon segment and dorsal root boutons on the motoneurons in the lumbosacral spinal cord of the cat during postnatal development. Acta Physiol Scand Suppl. 1969, 333: 53-76.PubMedGoogle Scholar
- Hausser M, Stuart G, Racca C, Sakmann B: Axonal initiation and active dendritic propagation of action potentials in substantia nigra neurons. Neuron. 1995, 15: 637-647. 10.1016/0896-6273(95)90152-3.View ArticlePubMedGoogle Scholar
- Morgan CW: Axons of sacral preganglionic neurons in the cat: I. Origin, initial segment, and myelination. J Neurocytol. 2001, 30: 523-544. 10.1023/A:1015649419346.View ArticlePubMedGoogle Scholar
- Lee DC, Jensen AL, Schiefer MA, Morgan CW, Grill WM: Structural mechanisms to produce differential dendritic gains. Brain Res. 2005, 1033: 117-127. 10.1016/j.brainres.2004.11.034.View ArticlePubMedGoogle Scholar
- Hu W, Tian C, Li T, Yang M, Hou H, Shu Y: Distinct contributions of Na(v)1.6 and Na(v)1.2 in action potential initiation and backpropagation. Nat Neurosci. 2009, 12: 996-1002. 10.1038/nn.2359.View ArticlePubMedGoogle Scholar
- Rush AM, Dib-Hajj SD, Waxman SG: Electrophysiological properties of two axonal sodium channels, Nav1.2 and Nav1.6, expressed in mouse spinal sensory neurones. J Physiol. 2005, 564: 803-815. 10.1113/jphysiol.2005.083089.PubMed CentralView ArticlePubMedGoogle Scholar
- Kernell D: The Motoneuron and its Muscle Fibers. 2006, Oxford, UK: Oxford University PressView ArticleGoogle Scholar
- Manuel M, Iglesias C, Donnet M, Leroy F, Heckman CJ, Zytnicki D: Fast kinetics, high-frequency oscillations, and subprimary firing range in adult mouse spinal motoneurons. J Neurosci. 2009, 29: 11246-11256. 10.1523/JNEUROSCI.3260-09.2009.PubMed CentralView ArticlePubMedGoogle Scholar
- Meehan CF, Sukiasyan N, Zhang M, Nielsen JB, Hultborn H: Intrinsic properties of mouse lumbar motoneurons revealed by intracellular recording in vivo. J Neurophysiol. 2010, 103: 2599-2610. 10.1152/jn.00668.2009.View ArticlePubMedGoogle Scholar
- Brown DA, Passmore GM: Neural KCNQ (Kv7) channels. Br J Pharmacol. 2009, 156: 1185-1195. 10.1111/j.1476-5381.2009.00111.x.PubMed CentralView ArticlePubMedGoogle Scholar
- Shah MM, Migliore M, Valencia I, Cooper EC, Brown DA: Functional significance of axonal Kv7 channels in hippocampal pyramidal neurons. Proc Natl Acad Sci USA. 2008, 105: 7869-7874. 10.1073/pnas.0802805105.PubMed CentralView ArticlePubMedGoogle Scholar
- Li YC, Cheng CX, Li YN, Shimada O, Atsumi S: Beyond the initial axon segment of the spinal motor axon: fasciculated microtubules and polyribosomal clusters. J Anat. 2005, 206: 535-542. 10.1111/j.1469-7580.2005.00418.x.PubMed CentralView ArticlePubMedGoogle Scholar
- Patino GA, Isom LL: Electrophysiology and beyond: multiple roles of Na+ channel beta subunits in development and disease. Neurosci Lett. 2010, 486: 53-59. 10.1016/j.neulet.2010.06.050.PubMed CentralView ArticlePubMedGoogle Scholar
- Rasband MN: Clustered K+ channel complexes in axons. Neurosci Lett. 2010, 486: 101-106. 10.1016/j.neulet.2010.08.081.PubMed CentralView ArticlePubMedGoogle Scholar
- Karnak D, Lee S, Margolis B: Identification of multiple binding partners for the amino-terminal domain of synapse-associated protein 97. J Biol Chem. 2002, 277: 46730-46735. 10.1074/jbc.M208781200.View ArticlePubMedGoogle Scholar
- Shingai T, Ikeda W, Kakunaga S, Morimoto K, Takekuni K, Itoh S, Satoh K, Takeuchi M, Imai T, Monden M, Takai Y: Implications of nectin-like molecule-2/IGSF4/RA175/SgIGSF/TSLC1/SynCAM1 in cell-cell adhesion and transmembrane protein localization in epithelial cells. J Biol Chem. 2003, 278: 35421-35427. 10.1074/jbc.M305387200.View ArticlePubMedGoogle Scholar
- Yageta M, Kuramochi M, Masuda M, Fukami T, Fukuhara H, Maruyama T, Shibuya M, Murakami Y: Direct association of TSLC1 and DAL-1, two distinct tumor suppressor proteins in lung cancer. Cancer Res. 2002, 62: 5129-5133.PubMedGoogle Scholar
- Gao B-X, Ziskind-Conhaim L: Development of ionic currents underlying changes in action potential waveforms in rat spinal motoneurons. J Neurophysiol. 1998, 80: 3047-3061.PubMedGoogle Scholar
- Buttermore ED, Dupree JL, Cheng JG, An X, Tessarollo L, Bhat MA: The cytoskeletal adaptor protein band 4.1B is required for the maintenance of paranodal axoglial septate junctions in myelinated axons. J Neurosci. 2011, 31: 8013-8024. 10.1523/JNEUROSCI.1015-11.2011.PubMed CentralView ArticlePubMedGoogle Scholar
- Cifuentes-Diaz C, Chareyre F, Garcia M, Devaux J, Carnaud M, Levasseur G, Niwa-Kawakita M, Harroch S, Girault J-A, Giovannini M, Goutebroze L: Protein 4.1B contributes to the organization of peripheral myelinated axons. PLoS ONE. 2011, 6: e25043-PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.