- Email: [email protected]
Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2
Biochemical and Biophysical Research Communications xxx (2017) 1e7
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2 Tetsushi Yamagata a, Ikuo Ogiwara a, b, Emi Mazaki a, Yuchio Yanagawa c, Kazuhiro Yamakawa a, * a b c
Laboratory for Neurogenetics, RIKEN Brain Science Institute, Wako, Saitama 351-0198, Japan Department of Physiology, Nippon Medical School, Tokyo 113-8602, Japan Department of Genetic and Behavioral Neuroscience, Gunma University Graduate School of Medicine, Maebashi 371-8511, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 July 2017 Accepted 2 August 2017 Available online xxx
Nav1.1 and Nav1.2 are the voltage-gated sodium channel pore-forming alpha I and II subunits, encoded by the genes SCN1A and SCN2A. Although mutations of both genes have similarly been described in patients with epilepsy, autism and/or intellectual disability, their expression sites in brain are largely distinct. Nav1.1 was shown to be expressed dominantly in parvalbumin (PV)-positive or somatostatin (SST)-positive inhibitory neurons and in a sparsely-distributed subpopulation of excitatory neurons. In contrast, Nav1.2 has been reported to be dominantly expressed in excitatory neurons. Here we show that Nav1.2 is also expressed in caudal ganglionic eminence (CGE)-derived inhibitory neurons, and expressions of Nav1.1 and Nav1.2 are mutually-exclusive in many of brain regions including neocortex, hippocampus, cerebellum, striatum and globus pallidus. In neocortex at postnatal day 15, in addition to the expression in excitatory neurons we show that Nav1.2 is expressed in reelin (RLN)-positive/SST-negative inhibitory neurons that are presumably single-bouquet cells because of their cortical layer I-limited distribution, and vasoactive intestinal peptide (VIP)-positive neurons that would be multipolar cell because of their layer I/II margin and layer VI distribution. Although Nav1.2 has previously been reported to be expressed in SST-positive cells, we here show that Nav1.2 is not expressed in either of PV-positive or SST-positive inhibitory neurons. PV-positive and SST-positive inhibitory neurons derive from medial ganglionic eminence (MGE) and innervate excitatory neurons, while VIP-positive and RLN-positive/SSTnegative inhibitory neurons derive from CGE, innervate on inhibitory neurons and play disinhibitory roles in the neural network. Our results therefore indicate that, while Nav1.1 is expressed in MEG-derived inhibitory neurons, Nav1.2 is expressed in CGE-derived disinhibitory interneurons in addition to excitatory neurons. These findings should contribute to understanding of the pathology of neurodevelopmental diseases caused by SCN2A mutations. © 2017 Elsevier Inc. All rights reserved.
Keywords: Sodium channel Nav1.2 Nav1.1 Interneuron Disinhibitory neuron
1. Introduction Voltage-gated sodium channels are transmembrane glycoprotein complexes that play a critical role in generation and propagation of action potentials in excitable cells including neurons, and consist of one pore-forming alpha-subunit and one or two accessory beta-subunits. In mammalian brain, four a subunits, namely,
* Corresponding author. Laboratory for Neurogenetics, RIKEN Brain Science Institute, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan. E-mail address: [email protected] (K. Yamakawa).
Nav1.1, 1.2, 1.3, and 1.6 encoded by SCN1A, 2A, 3A, and 8A, respectively, are heterogeneously expressed mainly at axon initial segments (AISs) and nodes of Ranvier. Although some previous studies proposed dominant somatodendritic distribution of Nav1.1 in excitatory neurons [1e3], we and others revealed that axons of PVor SST-positive inhibitory interneurons in neocortex and hippocampus heavily express Nav1.1 [4e9]. We also showed that Nav1.1 is sparsely expressed in axons of a distinct subpopulation of neocortical layer V excitatory neurons [6]. In contrast, dominant Nav1.2 expression in excitatory neurons has been widely accepted. In neocortex, Nav1.2 is expressed at proximal AISs and nodes of
http://dx.doi.org/10.1016/j.bbrc.2017.08.013 0006-291X/© 2017 Elsevier Inc. All rights reserved.
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
2
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
Ranvier of pyramidal cells [9,10]. In hippocampus, Nav1.2 is densely observed in unmyelinated mossy fibers of hippocampal granule cells [1,3] and in axons of CA1 and CA3 pyramidal cells [3,11]. In cerebellum, parallel fibers of granule cells densely express Nav1.2 [1,3,12]. Nav1.2 has been reported to be detected in GABAergic neurons such as striatal medial spiny neurons [13] and neocortical SST-positive interneurons [7], though the latter reveals to be contradictory to our present study. Here, we report mutually-exclusive expressions of Nav.1.1 and Nav1.2 in multiple brain regions such as neocortex, hippocampus, cerebellum, striatum and globus pallidus. Moreover, in neocortex and hippocampus we show that Nav1.2 is expressed in CGE-derived GABAergic interneurons such as RLN-positive/SST-negative singlebouquet cells and VIP-positive multipolar cells. We do not observe Nav1.2 in either of PV-positive or SST-positive neurons, both are MGE-derived GABAergic interneurons.
England Biolabs, Ipswich, MA, USA) to enhance the binding of the rabbit anti-Nav1.2 (ASC-002, Alomone) to Nav1.2 by removing the phosphate group on the antigen. The sections were then incubated with the rabbit anti-Nav1.2 (1:500; ASC-002, Alomone), the mouse anti-ankyrinG (1:250; SC-12719, Santa Cruz Biotechnology), and the goat anti-Nav1.1 antibody (1:500, C-18, Santa Cruz Biotechnology), and incubated with the secondary antibodies Alexa Flour 594, 647 (1:1000; Thermo Fisher Scientific, Waltham, MA USA) and Biotin conjugated (1:200; Vector Laboratories). Biotinylated antirabbit IgG antibody was detected using the Streptavidin conjugated Alexa Flour 488 (Thermo Fisher Scientific). Sections were mounted with Antifade Vectashield mounting medium containing 40 -6-diamidino-2-phenylindole (DAPI) (Vector Laboratories) to stain nuclei. Images were captured using Biozero BZ-8100, BZX-710 microscope (Keyence, Osaka, Japan) and TCS SP2 microscope (Leica Microsystems, Wetzler, Germany), and processed using Adobe Photoshop Elements 10 (Adobe Systems, San Jose, CA, USA).
2. Materials and methods 3. Results 2.1. Animals 3.1. Nav1.2 and Nav1.1 are distributed in mutually-excusive manner GAD67-GFP knock-in and Vgat-Venus transgenic mice expressing a fluorescent protein specifically in inhibitory cells were described previously [14,15], and maintained on a C57BL/6 J background. All animal experimental protocols were approved by the Animal Experiment Committee of RIKEN Institute. 2.2. Tissue preparation and immunohistochemistry Mouse tissues were fixed using periodate-lysine-4% paraformaldehyde (PLP), postfixed in PLP for overnight at 4 C, embedded in paraffin, and cut in 6 mm sections. The sections were deparaffinized, rehydrated, microwaved in 1 mM EDTA, pH8.0, and blocked in phosphate-buffered saline (PBS) containing 0.05% Tween 20, 4% BlockAce (Dainippon Sumitomo Pharma, Osaka, Japan) and endogenous avidin and biotin blocker (Avidin/Biotin or Streptavidin/Biotin Blocking Kit: Vector Laboratories, Burlingame, CA, USA) for 1 h at room temperature. The sections were then incubated with the goat anti-internal-region Nav1.2 antibody (1:500; SC-31371, G-20, Santa Cruz Biotechnology, Santa Cruz, CA, USA), the rabbit anti-internal-region Nav1.2 (1:500; ASC-002, Alomone, Jerusalem, Israel) or the goat anti-Nav1.1 antibody (1:500; SC-16031, C-18, Santa Cruz Biotechnology) for 12e15 h at 4 C. Endogenous peroxidases were quenched by incubation with 0.3% hydrogen peroxide in PBS. The sections were further incubated with biotinylated goat polyclonal secondary antibody (1:200; BA9500, Vector Laboratories). Detection of antibodyeantigen complexes was accomplished using the Vectastain Elite ABC kit (PK6100, Vector Laboratories) and the NovaRed substrate kit (SK-4800, Vector Laboratories). For the double staining, the sections, in which the antibody-antigen complexes were visualized with the NovaRed substrate (Vector Laboratories), were subsequently incubated with the mouse anti-GFP antibodies (1:500; 11814460001, clones 7.1 and 13.1, Roche Diagnostics, Indianapolis, IN, USA), the mouse antireelin antibody (1:1000; ab78540, G-10, Abcam, Cambridge, United Kingdom), the rabbit anti-vasoactive intestinal peptide antibody (1:1000; 20077, ImmunoStar, Inc., Hudson,WI, USA), the rabbit anti-somatostatin antibody (1:1000; T-4103, Peninsula Laboratories, San Carlos, CA, USA) or the goat anti-Nav1.1 antibody (1:500; C-18, Santa Cruz Biotechnology). Detection of the second antibodyeantigen complexes was accomplished using the Vectastain ABC-AP kit (AK-5000, Vector Laboratories) and the alkaline phosphatase substrate kit III (SK-5300, Vector Laboratories). For immunofluorescence histochemistry, the sections were pretreated with Lambda Protein Phosphatase (P0753S, New
We immunohistochemically investigated expression sites of Nav1.2 and Nav1.1. Because both Nav1.1 and Nav1.2 immunosignals at AISs have been most dense at around postnatal day (P) 15 during development in previous studies [6,11], we focused on P14.5-P15.5. At first, we performed single labeling of Nav1.2 or Nav1.1 by using the Vgat-Venus transgenic mice, which expressed GFP-derived fluorescent protein Venus in global inhibitory neurons [15]. In neocortex, Nav1.2-immunoreactivity was pronounced in most axons of Venusenegative excitatory neurons such as neocortical pyramidal cells (Fig. 1A and B). While Nav1.2-imunoreactivity was indiscernible in AISs of most Venus-positive neurons, it was sparsely observed in a minor population of Venus-positive cells such as those in neocortical layer I (Fig. 1B). Meanwhile, Nav1.1immunoreactivity was dense at AISs of a major subpopulation of GABAergic neurons (Fig. 1D and E), which are assumed to be PVpositive basket [4,6] or SST-positive Martinotti [7,8] inhibitory neurons, though we previously showed that a distinct subpopulation of excitatory neurons are Nav1.1-positive [6]. In hippocampus, Nav1.2-immunoractivity was apparent in most axons of excitatory neurons such as hippocampal CA pyramidal and dentate granule cells (Fig. 1A,C, Supplementary Figs. 1AeC). Venus-negative cells in hilus, presumably mossy cells that are excitatory neurons, were also Nav1.2-positive (Supplementary Fig. 1C). Nav1.2immunoreactivity was not observed in AISs of most Venuspositive neurons, but observed in a minor population of Venuspositive neurons in lacnosum-moleculare/deep radiatum. Meanwhile, Nav1.1-imunoreactivity was observed at AISs of a major subpopulation of GABAergic neurons (Fig. 1D, F and Supplementary Figs. 1EeG), which are assumed to be PV-positive basket [4,6] or SST-positive HIPP or O-LM [7,8] inhibitory neurons, while it was not observed in excitatory neurons. In cerebellum, Nav1.2immmunoreactivity was observed in parallel fibers of granule cells which are excitatory neurons (Supplementary Fig. 1D). As we showed previously [6], Nav1.1-immunoreactivity was apparent in axons of basket cells, but here we newly found that Venus-positive neurons in granule cell layer, presumably Golgi cells that are GABAergic cells, are also Nav1.1-positive (Supplementary Fig. 1H). In our previously study, we showed Nav1.1 is expressed in Purkinje cells [6] which are again GABAergic neurons. Double labeling (Fig. 2) further confirmed that Nav1.2 and Nav1.1-immunosignals were distributed in a mutually-exclusive manner in neocortex (Fig. 2B and H), hippocampus (Fig. 2C, D and I) and other brain regions including cerebellum (Fig. 2E),
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
3
Fig. 1. Single staining of Nav1.2 and Nav1.1 showed their distinct distributions in mouse brain. Parasagittal sections of P15.5 Vgat-Venus, an yellow emission GFP variant, transgenic mouse brains stained with anti-GFP (blue) antibody and anti-Nav1.2 (G-20) (red in A-C) or anti-Nav1.1 (C-18) (red in D-F) antibodies. Higher-magnified images outlined in A, D are shown in B, C, E, F. Note that Nav1.2-immunoreactivity is expressed in AISs of most of Venus-negative excitatory neurons (white arrowheads in B, C) and also in a subpopulation of Venus-positive inhibitory neurons in neocortical layer I (white arrows in B). Nav1.1-immunoreactivity is expressed in AISs of a subpopulation of Venus-positive inhibitory neurons in neocortical deeper layers and hippocampus (black arrowheads in E, F). Scale bars: (A, D) 1 mm; (B, C, E, F) 50 mm.
striatum (Fig. 2F) and globus pallidus (Fig. 2G). Although we previously showed that Nav1.1 is expressed in a sparsely-distributed distinct subpopulation of excitatory neurons in neocortex [6], here we did not observe any neocortical neurons co-expressing Nav1.2 and Nav1.1. Thus, Nav1.1 or Nav1.2 seemed not to be coexpressed in any of neurons in the brain regions so far investigated. 3.2. Nav1.2 is expressed in CGE-derived but not MGE-derived inhibitory neurons The Nav1.2 immunosignals at AISs of GABAergic neurons in neocortical layer I were reproducibly observed in the Vgat-Venus mice (Fig. 1B and Supplementary Figs. 2AeC) as well as in GAD67GFP knock-in mice [14] (Supplementary Figs. 2FeH). In addition, Nav1.2-positive AIS of GABAergic neurons was also observed in hippocampal lacunosum-moleculare (Supplementary Figs. 2A, D, E, F, I, J). In order to further investigate the nature of GABAergic neurons expressing Nav1.2-immunoreactivity, we performed doublelabeling of Nav1.2 and some inhibitory neuron markers (Fig. 3 and Supplementary Fig. 3). In the double-labeling of Nav1.2 and PV, in neocortex and hippocampus we found that none of AISs of PV-positive neurons seemed Nav1.2-positive (Supplementary Fig. 3), which was consistent to the previous report [7]. In the double-labeling of Nav1.2 and RLN (Fig. 3AeF), we found that AISs of some RLN-positive cells in neocortical layer I and hippocampal stratum radiatum-lacunosum-moleculare were apparently Nav1.2positive (Fig. 3A, B, D and E), while those in the other regions such as neocortical deeper layer and hippocampal striatum oriens seemed Nav1.2-negative (Fig. 3A, C and F). Although it has previously been reported that Nav1.2 is expressed in SST-positive
neocortical neurons [7], in our double-labeling of Nav1.2 and SST we found that none of SST-positive cells were Nav1.2-positive (Fig. 3GeK). These observations indicate that the RLN-positive cells in neocortical layer I and hippocampal lacunosummoleculare (Fig. 3A, B, D and E) were SST-negative. RLN-positive inhibitory neurons can be divided to SST-positive or -negative cells, and the RLN-positive/SST-negative cells have been known to be preferentially locating in neocortical layer I or hippocampal stratum lacunosum-moleculare/deep radiatum, while RLN-positive/ SST-positive cells, that would be Martinotti cells, have been known to be locating in deeper neocortical layers (II/III, V/VI) or hippocampal striatum oriens [16,17] (Fig. 4). The RLN-positive/SSTnegative interneurons can further be divided to two cell types; single bouquet cells (SBCs), that are most exclusively locating at neocortical layer I, and neurogliaform cells (NGFCs) that are a major component of neocortical layer I but also have been found in all layers [18e20]. Because the RLN-positive/SST-negative cells with Nav1.2-positive AISs were a minor population of the whole RLNpositive cell population in neocortical layer I and because their distribution is limited in layer I (Fig. 3A and B), these Nav1.2expressing RLN-positive/SST-negative neurons seem to be SBCs rather than NGFCs (Fig. 4). In the double-labeling of Nav1.2 and VIP (Fig. 3LeP), Nav1.2-immunoreactivity was obvious in VIP-positive neurons at layer I-II border and layer VI (Fig. 3M and N), but was not found those in hippocampus (Fig. 3O and P). VIP-positive neurons can be divided into two types; bipolar cells locating in neocortical layer II-VI and enriched at layer II/III and multipolar cells locating at neocortical layers I/II border and VI [19e21] (Fig. 4). Because of the distribution, Nav1.2-positive VIP neurons may correspond to the multipolar VIP interneurons. PV-positive and SST-positive inhibitory interneurons are both
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
4
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
Fig. 2. Double staining of Nav1.2 and Nav1.1 revealed their mutually-exclusive distributions. Parasagittal sections of P15.5 wild-type brain, double-stained with anti-Nav1.1 (C18) (blue) and anti-Nav1.2 (ASC-002) (red) antibodies. Higher-magnification images outlined in A are shown in B-G. White and black arrowheads indicate AISs positive for Nav1.2 and Nav1.1, respectively. Black arrows indicate Nav1.1-positive somata. Note that Nav1.2-immunosignals does not co-localize with Nav1.1-immunosignals in neocortex (B), hippocampus (C, D), cerebellum (E), and striatum (F). (G) In globus pallidus, Nav1.1-immunoreactivity was observed in somata (black arrows) but not in their AIS. Nav1.2immunoreactivity was not observed at AISs nor somata. (H, I) Immunofluorescence histochemistry of P15.5 wild-type neocortical layer V (H) and hippocampus CA1 (I), simultaneously stained with anti-Nav1.2 (ASC-002, green), anti-Nav1.1 (C-18, red), and anti-ankyrinG (pale blue) antibodies together with DAPI (blue). Merged images are shown in rightend panels. Arrows or arrowheads indicate Nav1.2- or Nav1.1-positive AISs or their corresponding positions in the panels, respectively. Note that Nav1.1- and Nav1.2-positive AISs are distinct each other. o, stratum oriens; p, stratum pyramidale. o, stratum oriens; p, stratum pyramidale; r, stratum radiatum; GL, granule cell layer; PC, Purkinje cell; ML, molecular layer; Shown are representative images. Scale bars: (BeG) 25 mm; (H, I) 20 mm.
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
5
Fig. 3. Nav1.2 is expressed in VIP-positive and RLN-positive/SST-negative GABAergic neurons. Double-staining of Nav1.2 and GABAergic neuron markers on parasagittal sections of P14.5 wild-type mouse brains. Arrowheads indicate Nav1.2-positve AISs. (A-F) Double-staining of Nav1.2 and RLN. RLN-positive neurons with Nav1.2-positive AISs were observed in the neocortical layer I and hippocampal lacunosum-moleculare (A, B, D, E). (G-K) Double-staining of Nav1.2 and SST. AISs of any of SST-expressing cells were Nav1.2-negative. SSTpositive somata did not appear in neocortical layer I. In the hippocampal lacunosum-moleculare, defused SST-immunosignals were detected but SST-positive somata were not observed. (L-P) Double-staining of Nav1.2 and VIP. VIP-positive neurons with Nav1.2-positive AISs were observed at neocortical layer I/II border and in layer VI (L, M, N). AISs of VIPpositive neurons in hippocampus were Nav1.2-negative (L, O, P). Higher-magnified images outlined in A, G, L are shown in B-F, H-K, M P, respectively. DG, dentate gyrus. Scale bars: (A, G, L) 200 mm; (B-F, H-K, M P) 25 mm.
MGE-derived, while VIP-positive and RLN-positive/SST-negative inhibitory neurons are both CGE-derived [16,17,22]. Nav1.1 have been found dominantly in PV- or SST-positive inhibitory interneurons in neocortex and hippocampus [4,6e8]. While PVpositive and SST-positive inhibitory neurons directly enervate onto exitatory neurons, targeting of axons of SBCs and VIP-positive cells including the multipolar cells preferentially onto interneurons in neocortex and their electrophysiological data have indicated that these cells play disinhibitory roles for neocortical pyramidal
neurons [19,23e25]. Our present data together with these previous studies indicate that Nav1.2 is expressed in CGE-derived disinhibitory interneurons in addition to excitatory neurons, which is in a good contrast to the dominant Nav1.1 expression in MGEderived inhibitory neurons (Fig. 4). 4. Discussion We have here shown that Nav1.1 and Nav1.2 are mutually
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
6
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
Fig. 4. Nav1.2 is expressed in CGE-derived disinhibitory interneurons. Medial ganglionic eminence (MGE) and caudal ganglionic eminence (CGE) generate the majority of neocortical interneurons; MGE generate PV-positive and SST-positive inhibitory neurons, and CGE generate VIP-positive and RLN-positive/SST-negative inhibitory neurons [16,17,19,20,22]. Some of CGE-derived inhibitory neurons such as VIP-positive multipolar or RLN-positive/SST-negative single-bouquet cells are know to play disinhibitory roles in the neural network [19,20,23e25]. While Nav1.1 is dominantly expressed in MGE-derived inhibitory interneurons such as PV-positive fast-spiking basket cells or SST-positive Martinotti cells [4,6e8], the present study indicates that Nav1.2 is expressed in CGE-derived disinhibitory interneurons such as multipolar or single-bouquet cells and not expressed in MGE-derived inhibitory neurons.
exclusive in both excitatory and inhibitory neurons focusing at ~ P15 when the expressions of both channels become most dense at AISs. Mutual exclusions of Nav1.1 and Nav1.2 were observed in many of brain regions including neocortex, hippocampus, striatum, cerebellum, and globus pallidus. In neocortex, while Nav1.2 is expressed in most of pyramidal cells, we did not observe any coexpressions of Nav1.1 and Nav1.2 even though we previously reported the Nav1.1 expression in a subpopulation of neocortical pyramidal cells [6]. These suggest that Nav1.1 is expressed in a minor distinct subpopulation of pyramidal cells which do not express Nav1.2. In inhibitory neurons, we found that Nav1.2 was expressed in RLN-positive/SST-negative inhibitory neurons in neocortical layer I, presumably single-bouquet cells, and hippocampal stratum lacunosum-moleculare/deep radiatum as well as VIP-positive inhibitory neurons at layer I/II margin and layer VI, presumably multipolar cells (Fig. 4). All of these inhibitory neurons are CGE-derived [16,17,20], and disinhibitory interneurons that innervate onto other inhibitory neurons [19,23e25], which is in a good contrast to that Nav1.1 is expressed in PV-positive [4,6] or SSTpositive [7,8] inhibitory neurons those are both MGE-derived inhibitory interneurons [16] that innervate onto primarily excitatory neurons. A previous study [7] reported that Nav1.2 is expressed in SST-positive cortical inhibitory neurons and proposed that loss of function mutations of SCN2A may impair those SSTpositive interneurons, decreases their inhibition onto pyramidal neurons, and result in epileptic seizures. However, we here showed that Nav1.2 were negative in either of PV- or SST-positive MGEderived inhibitory neurons and suggest alternative pathologies as described below. We and others have reported SCN2A mutations in patients with a wide spectrum of epilepsies, intellectual disability (ID) and autism spectrum disorder (ASD) [26e32]. Notably and contrarily to the case of SCN1A, SCN2A mutations appeared in patients with severer end of epilepsies such as early-infantile epileptic encephalopathy (EIEE) or Ohtahara syndrome and West syndrome are exclusively missense, while a majority of those appeared in patients with autism with milder or later-onset epilepsies or those even without
epilepsies were truncation mutations, and these suggested gain-offunction effects for the missense mutations appeared in severe early-onset epilepsies while loss-of-function effects for mutations appeared in ASDs and milder or later-onset epilepsies [32]. Actually, recent patch-clamp analyses surely showed that Nav1.2 channels with missense mutations found in patients with early-infantile severe epilepsies had increased channel activity with gain-offunction, while those found in patients with ASDs or late-onset epilepsies had loss-of-function effects [33,34]. Nav1.2 is actually well expressed in excitatory neurons such as neocortical and hippocampal pyramidal cells, and therefore it can be naturally assumed that gain-of-function mutations of SCN2A lead to hyperexcitabilities of those excitatory neurons and result in epileptic seizures. However, here we showed that Nav1.2 is also expressed in disinhibitory interneurons and therefore it would also be possible that Nav1.2 with such gain-of-function mutations in those disinhibitory interneurons may additionally contribute to epileptic seizures. It is now of interest whether and if so how and how much Nav1.2 with gain-of-function mutations in disinhibitory interneurons contribute to those severe epilepsies and also whether those with loss-of-function mutations in those interneurons contribute to ASDs. In summary, we showed that expression sites of Nav1.1 and Nav1.2 are mutually exclusive and Nav1.2 is expressed in CGEderived disinhibitory interneurons in addition to excitatory neurons. These results may contribute to the understanding of pathologies for neurodevelopmental and neuropsychiatric diseases caused by SCN2A mutations. Acknowledgement We are grateful to all members of the Laboratory for Neurogenetics (RIKEN Brain Science Institute (BSI)) for helpful discussion, and to the Research Resources Center (RIKEN-BSI) for technical assistances. We also thank Dr. Atsushi Miyawaki (RIKEN-BSI) for the Venus clone, and Dr. Makoto Kaneda (Nippon Medical School) for his support. This study was supported by RIKEN-BSI, the Japanese
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
Ministry of Education, Culture, Sports, Sciences and Technology (MEXT) Strategic Research Program for Brain Sciences (SRPBS) and Japan Agency for Medical Research and Development (AMED) (K.Y.); JSPS Grant-in-aid for Young Scientist (B) (21791020) and MEXT Grants-in-Aid for Scientific Research (C) (25461572) (I.O.) Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.bbrc.2017.08.013. Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.bbrc.2017.08.013. Conflict of interest The authors declare no conflict interest. References [1] R.E. Westenbroek, D.K. Merrick, W.A. Catterall, Differential subcellular localization of the RI and RII Naþ channel subtypes in central neurons, Neuron 3 (1989) 695e704. [2] R.E. Westenbroek, J.L. Noebels, W.A. Catterall, Elevated expression of type II Naþ channels in hypomyelinated axons of shiverer mouse brain, J. Neurosci. 12 (1992) 2259e2267. [3] B. Gong, K.J. Rhodes, Z. Bekele-Arcuri, J.S. Trimmer, Type I and type II Na(þ) channel alpha-subunit polypeptides exhibit distinct spatial and temporal patterning, and association with auxiliary subunits in rat brain, J. Comp. Neurol. 412 (1999) 342e352. [4] I. Ogiwara, H. Miyamoto, N. Morita, et al., Nav1.1 localizes to axons of parvalbumin-positive inhibitory interneurons: a circuit basis for epileptic seizures in mice carrying an Scn1a gene mutation, J. Neurosci. 27 (2007) 5903e5914. [5] A. Lorincz, Z. Nusser, Cell-type-dependent molecular composition of the axon initial segment, J. Neurosci. 28 (2008) 14329e14340. [6] I. Ogiwara, T. Iwasato, H. Miyamoto, et al., Nav1.1 haploinsufficiency in excitatory neurons ameliorates seizure-associated sudden death in a mouse model of Dravet syndrome, Hum. Mol. Genet. 22 (2013) 4784e4804. [7] T. Li, C. Tian, P. Scalmani, et al., Action potential initiation in neocortical inhibitory interneurons, PLoS Biol. 12 (2014) e1001944. [8] C. Tai, Y. Abe, R.E. Westenbroek, et al., Impaired excitability of somatostatinand parvalbumin-expressing cortical interneurons in a mouse model of Dravet syndrome, Proc. Natl. Acad. Sci. U. S. A. 111 (2014) E3139eE3148. [9] C. Tian, K. Wang, W. Ke, et al., Molecular identity of axonal sodium channels in human cortical pyramidal cells, Front. Cell. Neurosci. 8 (2014) 297. [10] W. Hu, C. Tian, T. Li, et al., Distinct contributions of Nav1.6 and Nav1.2 in action potential initiation and backpropagation, Nat. Neurosci. 12 (2009) 996e1002. [11] Y. Liao, L. Deprez, S. Maljevic, et al., Molecular correlates of age dependent seizures in an inherited neonatal-infantile epilepsy, Brain 133 (2010) 1403e1414. [12] Y. Liao, A.K. Anttonen, E. Liukkonen, et al., SCN2A mutation associated with neonatal epilepsy, late-onset episodic ataxia, myoclonus, and pain, Neurology 75 (2010b) 1454e1458. [13] H. Miyazaki, F. Oyama, R. Inoue, et al., Singular localization of sodium channel
7
b4 subunit in unmyelinated fibres and its role in the striatum, Nat. Commun. 21 (2014) 5525. [14] N. Tamamaki, Y. Yanagawa, R. Tomioka, et al., Green fluorescent protein expression and colocalization with calretinin, parvalbumin, and somatostatin in the GAD67-GFP knock-in mouse, J. Comp. Neurol. 467 (2003) 60e79. [15] Y. Wang, T. Kakizaki, H. Sakagami, et al., Fluorescents labeling of both GABAergic and glycinergic neurons in vesicular GABA transporter (VGAT)venus transgenic mouse, Neuroscience 64 (2009) 1031e1043. [16] G. Miyoshi, J. Hjerling-Leffler, T. Karayannis, et al., Genetic fate mapping reveals that the caudal ganglionic eminence produces a large and diverse population of superficial cortical interneurons, J. Neurosci. 30 (2010) 1582e1594. [17] L. Tricoire, K.A. Pelkey, B.E. Erkkila, et al., A blueprint for the spatiotemporal origins of mouse hippocampal interneuron diversity, J. Neurosci. 31 (2011) 10948e10970. [18] Y. Kubota, N. Shigematsu, F. Karube, et al., Selective coexpression of multiple chemical markers defines discrete populations of neocortical GABAergic neurons, Cereb. Cortex 21 (2011) 1803e1817. [19] R. Tremblay, S. Lee, B. Rudy, GABAergic interneurons in the neocortex: from cellular properties to circuits, Neuron 91 (2016) 260e292. [20] B. Wamsley, G. Fishell, Genetic and activity-dependent mechanisms underlying interneuron diversity, Nat. Rev. 18 (2017) 299e309. €nneke, B. Scheuer, R.J. Wagener, et al., Staiger. Characterizing VIP neu[21] A. Pro rons in the barrel cortex of VIPcre/tdTomato mice reveals layer-specific differences, Cereb. Cortex 25 (2015) 4854e4868. [22] A. Kepecs, G. Fishell, Interneuron cell types are fit to function, Nature 505 (2014) 318e326. [23] S. Lee, I. Kruglikov, Z.J. Huang, et al., A disinhibitory circuit mediates motor integration in the somatosensory cortex, Nat. Neurosci. 16 (2013) 1662e1670. [24] H.J. Pi, B. Hangya, D. Kvitsiani, et al., Cortical interneurons that specialize in disinhibitory control, Nature 503 (2013) 521e524. [25] X. Jiang, G. Wang, A.J. Lee, et al., The organization of two new cortical interneuronal circuits, Nat. Neurosci. 16 (2013) 210e218. [26] T. Sugawara, Y. Tsurubuchi, K.L. Agarwala, et al., A missense mutation of the Naþ channel aII subunit gene Nav1.2 in a patient with febrile and afebrile seizures causes channel dysfunction, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 6384e6389. [27] K. Kamiya, M. Kaneda, T. Sugawara, et al., A nonsense mutation of the sodium channel gene SCN2A in a patient with intractable epilepsy and mental decline, J. Neurosci. 24 (2004) 2690e2698. [28] I. Ogiwara, K. Ito, Y. Sawaishi, et al., De novo mutations of voltage-gated sodium channel aII gene SCN2A in intractable epilepsies, Neurol. 73 (2009) 1046e1053. [29] K. Nakamura, M. Kato, H. Osaka, et al., Clinical spectrum of SCN2A mutations expanding to Ohtahara syndrome, Neurology. 81 (2013) 992e998. [30] A. Hoischen, N. Krumm, E.E. Eichler, Prioritization of neurodevelopmental disease genes by discovery of new mutations, Nat. Neurosci. 17 (2014) 764e772. [31] M.R. Johnson, K. Shkura, S.R. Langley, et al., Systems genetics identifies a convergent gene network for cognition and neurodevelopmental disease, Nat. Neurosci. 19 (2016) 223e232. [32] K. Yamakawa, Mutations of voltage-gated sodium channel genes SCN1A and SCN2A in epilepsy, intellectual disability, and autism, in: C. Sara, C. Verpelli (Eds.), Synaptic Dysfunction in Autism Spectrum Disorder and Intellectual Disability, Elsevier, Amsterdam, Netherlands, 2016. [33] M. Wolff, K.M. Johannesen, U.B. Hedrich, et al., Genetic and phenotypic heterogeneity suggest therapeutic implications in SCN2A-related disorders, Brain 140 (2017) 1326e1336. [34] R. Ben-Shalom, C.M. Keeshen, K.N. Berrios, et al., Opposing effects on NaV1.2 function underlie differences between SCN2A variants observed in individuals with autism spectrum disorder or infantile seizures, Biol. Psychiatry 82 (2017) 224e232.
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2 Tetsushi Yamagata a, Ikuo Ogiwara a, b, Emi Mazaki a, Yuchio Yanagawa c, Kazuhiro Yamakawa a, * a b c
Laboratory for Neurogenetics, RIKEN Brain Science Institute, Wako, Saitama 351-0198, Japan Department of Physiology, Nippon Medical School, Tokyo 113-8602, Japan Department of Genetic and Behavioral Neuroscience, Gunma University Graduate School of Medicine, Maebashi 371-8511, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 July 2017 Accepted 2 August 2017 Available online xxx
Nav1.1 and Nav1.2 are the voltage-gated sodium channel pore-forming alpha I and II subunits, encoded by the genes SCN1A and SCN2A. Although mutations of both genes have similarly been described in patients with epilepsy, autism and/or intellectual disability, their expression sites in brain are largely distinct. Nav1.1 was shown to be expressed dominantly in parvalbumin (PV)-positive or somatostatin (SST)-positive inhibitory neurons and in a sparsely-distributed subpopulation of excitatory neurons. In contrast, Nav1.2 has been reported to be dominantly expressed in excitatory neurons. Here we show that Nav1.2 is also expressed in caudal ganglionic eminence (CGE)-derived inhibitory neurons, and expressions of Nav1.1 and Nav1.2 are mutually-exclusive in many of brain regions including neocortex, hippocampus, cerebellum, striatum and globus pallidus. In neocortex at postnatal day 15, in addition to the expression in excitatory neurons we show that Nav1.2 is expressed in reelin (RLN)-positive/SST-negative inhibitory neurons that are presumably single-bouquet cells because of their cortical layer I-limited distribution, and vasoactive intestinal peptide (VIP)-positive neurons that would be multipolar cell because of their layer I/II margin and layer VI distribution. Although Nav1.2 has previously been reported to be expressed in SST-positive cells, we here show that Nav1.2 is not expressed in either of PV-positive or SST-positive inhibitory neurons. PV-positive and SST-positive inhibitory neurons derive from medial ganglionic eminence (MGE) and innervate excitatory neurons, while VIP-positive and RLN-positive/SSTnegative inhibitory neurons derive from CGE, innervate on inhibitory neurons and play disinhibitory roles in the neural network. Our results therefore indicate that, while Nav1.1 is expressed in MEG-derived inhibitory neurons, Nav1.2 is expressed in CGE-derived disinhibitory interneurons in addition to excitatory neurons. These findings should contribute to understanding of the pathology of neurodevelopmental diseases caused by SCN2A mutations. © 2017 Elsevier Inc. All rights reserved.
Keywords: Sodium channel Nav1.2 Nav1.1 Interneuron Disinhibitory neuron
1. Introduction Voltage-gated sodium channels are transmembrane glycoprotein complexes that play a critical role in generation and propagation of action potentials in excitable cells including neurons, and consist of one pore-forming alpha-subunit and one or two accessory beta-subunits. In mammalian brain, four a subunits, namely,
* Corresponding author. Laboratory for Neurogenetics, RIKEN Brain Science Institute, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan. E-mail address: [email protected] (K. Yamakawa).
Nav1.1, 1.2, 1.3, and 1.6 encoded by SCN1A, 2A, 3A, and 8A, respectively, are heterogeneously expressed mainly at axon initial segments (AISs) and nodes of Ranvier. Although some previous studies proposed dominant somatodendritic distribution of Nav1.1 in excitatory neurons [1e3], we and others revealed that axons of PVor SST-positive inhibitory interneurons in neocortex and hippocampus heavily express Nav1.1 [4e9]. We also showed that Nav1.1 is sparsely expressed in axons of a distinct subpopulation of neocortical layer V excitatory neurons [6]. In contrast, dominant Nav1.2 expression in excitatory neurons has been widely accepted. In neocortex, Nav1.2 is expressed at proximal AISs and nodes of
http://dx.doi.org/10.1016/j.bbrc.2017.08.013 0006-291X/© 2017 Elsevier Inc. All rights reserved.
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
2
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
Ranvier of pyramidal cells [9,10]. In hippocampus, Nav1.2 is densely observed in unmyelinated mossy fibers of hippocampal granule cells [1,3] and in axons of CA1 and CA3 pyramidal cells [3,11]. In cerebellum, parallel fibers of granule cells densely express Nav1.2 [1,3,12]. Nav1.2 has been reported to be detected in GABAergic neurons such as striatal medial spiny neurons [13] and neocortical SST-positive interneurons [7], though the latter reveals to be contradictory to our present study. Here, we report mutually-exclusive expressions of Nav.1.1 and Nav1.2 in multiple brain regions such as neocortex, hippocampus, cerebellum, striatum and globus pallidus. Moreover, in neocortex and hippocampus we show that Nav1.2 is expressed in CGE-derived GABAergic interneurons such as RLN-positive/SST-negative singlebouquet cells and VIP-positive multipolar cells. We do not observe Nav1.2 in either of PV-positive or SST-positive neurons, both are MGE-derived GABAergic interneurons.
England Biolabs, Ipswich, MA, USA) to enhance the binding of the rabbit anti-Nav1.2 (ASC-002, Alomone) to Nav1.2 by removing the phosphate group on the antigen. The sections were then incubated with the rabbit anti-Nav1.2 (1:500; ASC-002, Alomone), the mouse anti-ankyrinG (1:250; SC-12719, Santa Cruz Biotechnology), and the goat anti-Nav1.1 antibody (1:500, C-18, Santa Cruz Biotechnology), and incubated with the secondary antibodies Alexa Flour 594, 647 (1:1000; Thermo Fisher Scientific, Waltham, MA USA) and Biotin conjugated (1:200; Vector Laboratories). Biotinylated antirabbit IgG antibody was detected using the Streptavidin conjugated Alexa Flour 488 (Thermo Fisher Scientific). Sections were mounted with Antifade Vectashield mounting medium containing 40 -6-diamidino-2-phenylindole (DAPI) (Vector Laboratories) to stain nuclei. Images were captured using Biozero BZ-8100, BZX-710 microscope (Keyence, Osaka, Japan) and TCS SP2 microscope (Leica Microsystems, Wetzler, Germany), and processed using Adobe Photoshop Elements 10 (Adobe Systems, San Jose, CA, USA).
2. Materials and methods 3. Results 2.1. Animals 3.1. Nav1.2 and Nav1.1 are distributed in mutually-excusive manner GAD67-GFP knock-in and Vgat-Venus transgenic mice expressing a fluorescent protein specifically in inhibitory cells were described previously [14,15], and maintained on a C57BL/6 J background. All animal experimental protocols were approved by the Animal Experiment Committee of RIKEN Institute. 2.2. Tissue preparation and immunohistochemistry Mouse tissues were fixed using periodate-lysine-4% paraformaldehyde (PLP), postfixed in PLP for overnight at 4 C, embedded in paraffin, and cut in 6 mm sections. The sections were deparaffinized, rehydrated, microwaved in 1 mM EDTA, pH8.0, and blocked in phosphate-buffered saline (PBS) containing 0.05% Tween 20, 4% BlockAce (Dainippon Sumitomo Pharma, Osaka, Japan) and endogenous avidin and biotin blocker (Avidin/Biotin or Streptavidin/Biotin Blocking Kit: Vector Laboratories, Burlingame, CA, USA) for 1 h at room temperature. The sections were then incubated with the goat anti-internal-region Nav1.2 antibody (1:500; SC-31371, G-20, Santa Cruz Biotechnology, Santa Cruz, CA, USA), the rabbit anti-internal-region Nav1.2 (1:500; ASC-002, Alomone, Jerusalem, Israel) or the goat anti-Nav1.1 antibody (1:500; SC-16031, C-18, Santa Cruz Biotechnology) for 12e15 h at 4 C. Endogenous peroxidases were quenched by incubation with 0.3% hydrogen peroxide in PBS. The sections were further incubated with biotinylated goat polyclonal secondary antibody (1:200; BA9500, Vector Laboratories). Detection of antibodyeantigen complexes was accomplished using the Vectastain Elite ABC kit (PK6100, Vector Laboratories) and the NovaRed substrate kit (SK-4800, Vector Laboratories). For the double staining, the sections, in which the antibody-antigen complexes were visualized with the NovaRed substrate (Vector Laboratories), were subsequently incubated with the mouse anti-GFP antibodies (1:500; 11814460001, clones 7.1 and 13.1, Roche Diagnostics, Indianapolis, IN, USA), the mouse antireelin antibody (1:1000; ab78540, G-10, Abcam, Cambridge, United Kingdom), the rabbit anti-vasoactive intestinal peptide antibody (1:1000; 20077, ImmunoStar, Inc., Hudson,WI, USA), the rabbit anti-somatostatin antibody (1:1000; T-4103, Peninsula Laboratories, San Carlos, CA, USA) or the goat anti-Nav1.1 antibody (1:500; C-18, Santa Cruz Biotechnology). Detection of the second antibodyeantigen complexes was accomplished using the Vectastain ABC-AP kit (AK-5000, Vector Laboratories) and the alkaline phosphatase substrate kit III (SK-5300, Vector Laboratories). For immunofluorescence histochemistry, the sections were pretreated with Lambda Protein Phosphatase (P0753S, New
We immunohistochemically investigated expression sites of Nav1.2 and Nav1.1. Because both Nav1.1 and Nav1.2 immunosignals at AISs have been most dense at around postnatal day (P) 15 during development in previous studies [6,11], we focused on P14.5-P15.5. At first, we performed single labeling of Nav1.2 or Nav1.1 by using the Vgat-Venus transgenic mice, which expressed GFP-derived fluorescent protein Venus in global inhibitory neurons [15]. In neocortex, Nav1.2-immunoreactivity was pronounced in most axons of Venusenegative excitatory neurons such as neocortical pyramidal cells (Fig. 1A and B). While Nav1.2-imunoreactivity was indiscernible in AISs of most Venus-positive neurons, it was sparsely observed in a minor population of Venus-positive cells such as those in neocortical layer I (Fig. 1B). Meanwhile, Nav1.1immunoreactivity was dense at AISs of a major subpopulation of GABAergic neurons (Fig. 1D and E), which are assumed to be PVpositive basket [4,6] or SST-positive Martinotti [7,8] inhibitory neurons, though we previously showed that a distinct subpopulation of excitatory neurons are Nav1.1-positive [6]. In hippocampus, Nav1.2-immunoractivity was apparent in most axons of excitatory neurons such as hippocampal CA pyramidal and dentate granule cells (Fig. 1A,C, Supplementary Figs. 1AeC). Venus-negative cells in hilus, presumably mossy cells that are excitatory neurons, were also Nav1.2-positive (Supplementary Fig. 1C). Nav1.2immunoreactivity was not observed in AISs of most Venuspositive neurons, but observed in a minor population of Venuspositive neurons in lacnosum-moleculare/deep radiatum. Meanwhile, Nav1.1-imunoreactivity was observed at AISs of a major subpopulation of GABAergic neurons (Fig. 1D, F and Supplementary Figs. 1EeG), which are assumed to be PV-positive basket [4,6] or SST-positive HIPP or O-LM [7,8] inhibitory neurons, while it was not observed in excitatory neurons. In cerebellum, Nav1.2immmunoreactivity was observed in parallel fibers of granule cells which are excitatory neurons (Supplementary Fig. 1D). As we showed previously [6], Nav1.1-immunoreactivity was apparent in axons of basket cells, but here we newly found that Venus-positive neurons in granule cell layer, presumably Golgi cells that are GABAergic cells, are also Nav1.1-positive (Supplementary Fig. 1H). In our previously study, we showed Nav1.1 is expressed in Purkinje cells [6] which are again GABAergic neurons. Double labeling (Fig. 2) further confirmed that Nav1.2 and Nav1.1-immunosignals were distributed in a mutually-exclusive manner in neocortex (Fig. 2B and H), hippocampus (Fig. 2C, D and I) and other brain regions including cerebellum (Fig. 2E),
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
3
Fig. 1. Single staining of Nav1.2 and Nav1.1 showed their distinct distributions in mouse brain. Parasagittal sections of P15.5 Vgat-Venus, an yellow emission GFP variant, transgenic mouse brains stained with anti-GFP (blue) antibody and anti-Nav1.2 (G-20) (red in A-C) or anti-Nav1.1 (C-18) (red in D-F) antibodies. Higher-magnified images outlined in A, D are shown in B, C, E, F. Note that Nav1.2-immunoreactivity is expressed in AISs of most of Venus-negative excitatory neurons (white arrowheads in B, C) and also in a subpopulation of Venus-positive inhibitory neurons in neocortical layer I (white arrows in B). Nav1.1-immunoreactivity is expressed in AISs of a subpopulation of Venus-positive inhibitory neurons in neocortical deeper layers and hippocampus (black arrowheads in E, F). Scale bars: (A, D) 1 mm; (B, C, E, F) 50 mm.
striatum (Fig. 2F) and globus pallidus (Fig. 2G). Although we previously showed that Nav1.1 is expressed in a sparsely-distributed distinct subpopulation of excitatory neurons in neocortex [6], here we did not observe any neocortical neurons co-expressing Nav1.2 and Nav1.1. Thus, Nav1.1 or Nav1.2 seemed not to be coexpressed in any of neurons in the brain regions so far investigated. 3.2. Nav1.2 is expressed in CGE-derived but not MGE-derived inhibitory neurons The Nav1.2 immunosignals at AISs of GABAergic neurons in neocortical layer I were reproducibly observed in the Vgat-Venus mice (Fig. 1B and Supplementary Figs. 2AeC) as well as in GAD67GFP knock-in mice [14] (Supplementary Figs. 2FeH). In addition, Nav1.2-positive AIS of GABAergic neurons was also observed in hippocampal lacunosum-moleculare (Supplementary Figs. 2A, D, E, F, I, J). In order to further investigate the nature of GABAergic neurons expressing Nav1.2-immunoreactivity, we performed doublelabeling of Nav1.2 and some inhibitory neuron markers (Fig. 3 and Supplementary Fig. 3). In the double-labeling of Nav1.2 and PV, in neocortex and hippocampus we found that none of AISs of PV-positive neurons seemed Nav1.2-positive (Supplementary Fig. 3), which was consistent to the previous report [7]. In the double-labeling of Nav1.2 and RLN (Fig. 3AeF), we found that AISs of some RLN-positive cells in neocortical layer I and hippocampal stratum radiatum-lacunosum-moleculare were apparently Nav1.2positive (Fig. 3A, B, D and E), while those in the other regions such as neocortical deeper layer and hippocampal striatum oriens seemed Nav1.2-negative (Fig. 3A, C and F). Although it has previously been reported that Nav1.2 is expressed in SST-positive
neocortical neurons [7], in our double-labeling of Nav1.2 and SST we found that none of SST-positive cells were Nav1.2-positive (Fig. 3GeK). These observations indicate that the RLN-positive cells in neocortical layer I and hippocampal lacunosummoleculare (Fig. 3A, B, D and E) were SST-negative. RLN-positive inhibitory neurons can be divided to SST-positive or -negative cells, and the RLN-positive/SST-negative cells have been known to be preferentially locating in neocortical layer I or hippocampal stratum lacunosum-moleculare/deep radiatum, while RLN-positive/ SST-positive cells, that would be Martinotti cells, have been known to be locating in deeper neocortical layers (II/III, V/VI) or hippocampal striatum oriens [16,17] (Fig. 4). The RLN-positive/SSTnegative interneurons can further be divided to two cell types; single bouquet cells (SBCs), that are most exclusively locating at neocortical layer I, and neurogliaform cells (NGFCs) that are a major component of neocortical layer I but also have been found in all layers [18e20]. Because the RLN-positive/SST-negative cells with Nav1.2-positive AISs were a minor population of the whole RLNpositive cell population in neocortical layer I and because their distribution is limited in layer I (Fig. 3A and B), these Nav1.2expressing RLN-positive/SST-negative neurons seem to be SBCs rather than NGFCs (Fig. 4). In the double-labeling of Nav1.2 and VIP (Fig. 3LeP), Nav1.2-immunoreactivity was obvious in VIP-positive neurons at layer I-II border and layer VI (Fig. 3M and N), but was not found those in hippocampus (Fig. 3O and P). VIP-positive neurons can be divided into two types; bipolar cells locating in neocortical layer II-VI and enriched at layer II/III and multipolar cells locating at neocortical layers I/II border and VI [19e21] (Fig. 4). Because of the distribution, Nav1.2-positive VIP neurons may correspond to the multipolar VIP interneurons. PV-positive and SST-positive inhibitory interneurons are both
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
4
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
Fig. 2. Double staining of Nav1.2 and Nav1.1 revealed their mutually-exclusive distributions. Parasagittal sections of P15.5 wild-type brain, double-stained with anti-Nav1.1 (C18) (blue) and anti-Nav1.2 (ASC-002) (red) antibodies. Higher-magnification images outlined in A are shown in B-G. White and black arrowheads indicate AISs positive for Nav1.2 and Nav1.1, respectively. Black arrows indicate Nav1.1-positive somata. Note that Nav1.2-immunosignals does not co-localize with Nav1.1-immunosignals in neocortex (B), hippocampus (C, D), cerebellum (E), and striatum (F). (G) In globus pallidus, Nav1.1-immunoreactivity was observed in somata (black arrows) but not in their AIS. Nav1.2immunoreactivity was not observed at AISs nor somata. (H, I) Immunofluorescence histochemistry of P15.5 wild-type neocortical layer V (H) and hippocampus CA1 (I), simultaneously stained with anti-Nav1.2 (ASC-002, green), anti-Nav1.1 (C-18, red), and anti-ankyrinG (pale blue) antibodies together with DAPI (blue). Merged images are shown in rightend panels. Arrows or arrowheads indicate Nav1.2- or Nav1.1-positive AISs or their corresponding positions in the panels, respectively. Note that Nav1.1- and Nav1.2-positive AISs are distinct each other. o, stratum oriens; p, stratum pyramidale. o, stratum oriens; p, stratum pyramidale; r, stratum radiatum; GL, granule cell layer; PC, Purkinje cell; ML, molecular layer; Shown are representative images. Scale bars: (BeG) 25 mm; (H, I) 20 mm.
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
5
Fig. 3. Nav1.2 is expressed in VIP-positive and RLN-positive/SST-negative GABAergic neurons. Double-staining of Nav1.2 and GABAergic neuron markers on parasagittal sections of P14.5 wild-type mouse brains. Arrowheads indicate Nav1.2-positve AISs. (A-F) Double-staining of Nav1.2 and RLN. RLN-positive neurons with Nav1.2-positive AISs were observed in the neocortical layer I and hippocampal lacunosum-moleculare (A, B, D, E). (G-K) Double-staining of Nav1.2 and SST. AISs of any of SST-expressing cells were Nav1.2-negative. SSTpositive somata did not appear in neocortical layer I. In the hippocampal lacunosum-moleculare, defused SST-immunosignals were detected but SST-positive somata were not observed. (L-P) Double-staining of Nav1.2 and VIP. VIP-positive neurons with Nav1.2-positive AISs were observed at neocortical layer I/II border and in layer VI (L, M, N). AISs of VIPpositive neurons in hippocampus were Nav1.2-negative (L, O, P). Higher-magnified images outlined in A, G, L are shown in B-F, H-K, M P, respectively. DG, dentate gyrus. Scale bars: (A, G, L) 200 mm; (B-F, H-K, M P) 25 mm.
MGE-derived, while VIP-positive and RLN-positive/SST-negative inhibitory neurons are both CGE-derived [16,17,22]. Nav1.1 have been found dominantly in PV- or SST-positive inhibitory interneurons in neocortex and hippocampus [4,6e8]. While PVpositive and SST-positive inhibitory neurons directly enervate onto exitatory neurons, targeting of axons of SBCs and VIP-positive cells including the multipolar cells preferentially onto interneurons in neocortex and their electrophysiological data have indicated that these cells play disinhibitory roles for neocortical pyramidal
neurons [19,23e25]. Our present data together with these previous studies indicate that Nav1.2 is expressed in CGE-derived disinhibitory interneurons in addition to excitatory neurons, which is in a good contrast to the dominant Nav1.1 expression in MGEderived inhibitory neurons (Fig. 4). 4. Discussion We have here shown that Nav1.1 and Nav1.2 are mutually
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
6
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
Fig. 4. Nav1.2 is expressed in CGE-derived disinhibitory interneurons. Medial ganglionic eminence (MGE) and caudal ganglionic eminence (CGE) generate the majority of neocortical interneurons; MGE generate PV-positive and SST-positive inhibitory neurons, and CGE generate VIP-positive and RLN-positive/SST-negative inhibitory neurons [16,17,19,20,22]. Some of CGE-derived inhibitory neurons such as VIP-positive multipolar or RLN-positive/SST-negative single-bouquet cells are know to play disinhibitory roles in the neural network [19,20,23e25]. While Nav1.1 is dominantly expressed in MGE-derived inhibitory interneurons such as PV-positive fast-spiking basket cells or SST-positive Martinotti cells [4,6e8], the present study indicates that Nav1.2 is expressed in CGE-derived disinhibitory interneurons such as multipolar or single-bouquet cells and not expressed in MGE-derived inhibitory neurons.
exclusive in both excitatory and inhibitory neurons focusing at ~ P15 when the expressions of both channels become most dense at AISs. Mutual exclusions of Nav1.1 and Nav1.2 were observed in many of brain regions including neocortex, hippocampus, striatum, cerebellum, and globus pallidus. In neocortex, while Nav1.2 is expressed in most of pyramidal cells, we did not observe any coexpressions of Nav1.1 and Nav1.2 even though we previously reported the Nav1.1 expression in a subpopulation of neocortical pyramidal cells [6]. These suggest that Nav1.1 is expressed in a minor distinct subpopulation of pyramidal cells which do not express Nav1.2. In inhibitory neurons, we found that Nav1.2 was expressed in RLN-positive/SST-negative inhibitory neurons in neocortical layer I, presumably single-bouquet cells, and hippocampal stratum lacunosum-moleculare/deep radiatum as well as VIP-positive inhibitory neurons at layer I/II margin and layer VI, presumably multipolar cells (Fig. 4). All of these inhibitory neurons are CGE-derived [16,17,20], and disinhibitory interneurons that innervate onto other inhibitory neurons [19,23e25], which is in a good contrast to that Nav1.1 is expressed in PV-positive [4,6] or SSTpositive [7,8] inhibitory neurons those are both MGE-derived inhibitory interneurons [16] that innervate onto primarily excitatory neurons. A previous study [7] reported that Nav1.2 is expressed in SST-positive cortical inhibitory neurons and proposed that loss of function mutations of SCN2A may impair those SSTpositive interneurons, decreases their inhibition onto pyramidal neurons, and result in epileptic seizures. However, we here showed that Nav1.2 were negative in either of PV- or SST-positive MGEderived inhibitory neurons and suggest alternative pathologies as described below. We and others have reported SCN2A mutations in patients with a wide spectrum of epilepsies, intellectual disability (ID) and autism spectrum disorder (ASD) [26e32]. Notably and contrarily to the case of SCN1A, SCN2A mutations appeared in patients with severer end of epilepsies such as early-infantile epileptic encephalopathy (EIEE) or Ohtahara syndrome and West syndrome are exclusively missense, while a majority of those appeared in patients with autism with milder or later-onset epilepsies or those even without
epilepsies were truncation mutations, and these suggested gain-offunction effects for the missense mutations appeared in severe early-onset epilepsies while loss-of-function effects for mutations appeared in ASDs and milder or later-onset epilepsies [32]. Actually, recent patch-clamp analyses surely showed that Nav1.2 channels with missense mutations found in patients with early-infantile severe epilepsies had increased channel activity with gain-offunction, while those found in patients with ASDs or late-onset epilepsies had loss-of-function effects [33,34]. Nav1.2 is actually well expressed in excitatory neurons such as neocortical and hippocampal pyramidal cells, and therefore it can be naturally assumed that gain-of-function mutations of SCN2A lead to hyperexcitabilities of those excitatory neurons and result in epileptic seizures. However, here we showed that Nav1.2 is also expressed in disinhibitory interneurons and therefore it would also be possible that Nav1.2 with such gain-of-function mutations in those disinhibitory interneurons may additionally contribute to epileptic seizures. It is now of interest whether and if so how and how much Nav1.2 with gain-of-function mutations in disinhibitory interneurons contribute to those severe epilepsies and also whether those with loss-of-function mutations in those interneurons contribute to ASDs. In summary, we showed that expression sites of Nav1.1 and Nav1.2 are mutually exclusive and Nav1.2 is expressed in CGEderived disinhibitory interneurons in addition to excitatory neurons. These results may contribute to the understanding of pathologies for neurodevelopmental and neuropsychiatric diseases caused by SCN2A mutations. Acknowledgement We are grateful to all members of the Laboratory for Neurogenetics (RIKEN Brain Science Institute (BSI)) for helpful discussion, and to the Research Resources Center (RIKEN-BSI) for technical assistances. We also thank Dr. Atsushi Miyawaki (RIKEN-BSI) for the Venus clone, and Dr. Makoto Kaneda (Nippon Medical School) for his support. This study was supported by RIKEN-BSI, the Japanese
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013
T. Yamagata et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7
Ministry of Education, Culture, Sports, Sciences and Technology (MEXT) Strategic Research Program for Brain Sciences (SRPBS) and Japan Agency for Medical Research and Development (AMED) (K.Y.); JSPS Grant-in-aid for Young Scientist (B) (21791020) and MEXT Grants-in-Aid for Scientific Research (C) (25461572) (I.O.) Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.bbrc.2017.08.013. Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.bbrc.2017.08.013. Conflict of interest The authors declare no conflict interest. References [1] R.E. Westenbroek, D.K. Merrick, W.A. Catterall, Differential subcellular localization of the RI and RII Naþ channel subtypes in central neurons, Neuron 3 (1989) 695e704. [2] R.E. Westenbroek, J.L. Noebels, W.A. Catterall, Elevated expression of type II Naþ channels in hypomyelinated axons of shiverer mouse brain, J. Neurosci. 12 (1992) 2259e2267. [3] B. Gong, K.J. Rhodes, Z. Bekele-Arcuri, J.S. Trimmer, Type I and type II Na(þ) channel alpha-subunit polypeptides exhibit distinct spatial and temporal patterning, and association with auxiliary subunits in rat brain, J. Comp. Neurol. 412 (1999) 342e352. [4] I. Ogiwara, H. Miyamoto, N. Morita, et al., Nav1.1 localizes to axons of parvalbumin-positive inhibitory interneurons: a circuit basis for epileptic seizures in mice carrying an Scn1a gene mutation, J. Neurosci. 27 (2007) 5903e5914. [5] A. Lorincz, Z. Nusser, Cell-type-dependent molecular composition of the axon initial segment, J. Neurosci. 28 (2008) 14329e14340. [6] I. Ogiwara, T. Iwasato, H. Miyamoto, et al., Nav1.1 haploinsufficiency in excitatory neurons ameliorates seizure-associated sudden death in a mouse model of Dravet syndrome, Hum. Mol. Genet. 22 (2013) 4784e4804. [7] T. Li, C. Tian, P. Scalmani, et al., Action potential initiation in neocortical inhibitory interneurons, PLoS Biol. 12 (2014) e1001944. [8] C. Tai, Y. Abe, R.E. Westenbroek, et al., Impaired excitability of somatostatinand parvalbumin-expressing cortical interneurons in a mouse model of Dravet syndrome, Proc. Natl. Acad. Sci. U. S. A. 111 (2014) E3139eE3148. [9] C. Tian, K. Wang, W. Ke, et al., Molecular identity of axonal sodium channels in human cortical pyramidal cells, Front. Cell. Neurosci. 8 (2014) 297. [10] W. Hu, C. Tian, T. Li, et al., Distinct contributions of Nav1.6 and Nav1.2 in action potential initiation and backpropagation, Nat. Neurosci. 12 (2009) 996e1002. [11] Y. Liao, L. Deprez, S. Maljevic, et al., Molecular correlates of age dependent seizures in an inherited neonatal-infantile epilepsy, Brain 133 (2010) 1403e1414. [12] Y. Liao, A.K. Anttonen, E. Liukkonen, et al., SCN2A mutation associated with neonatal epilepsy, late-onset episodic ataxia, myoclonus, and pain, Neurology 75 (2010b) 1454e1458. [13] H. Miyazaki, F. Oyama, R. Inoue, et al., Singular localization of sodium channel
7
b4 subunit in unmyelinated fibres and its role in the striatum, Nat. Commun. 21 (2014) 5525. [14] N. Tamamaki, Y. Yanagawa, R. Tomioka, et al., Green fluorescent protein expression and colocalization with calretinin, parvalbumin, and somatostatin in the GAD67-GFP knock-in mouse, J. Comp. Neurol. 467 (2003) 60e79. [15] Y. Wang, T. Kakizaki, H. Sakagami, et al., Fluorescents labeling of both GABAergic and glycinergic neurons in vesicular GABA transporter (VGAT)venus transgenic mouse, Neuroscience 64 (2009) 1031e1043. [16] G. Miyoshi, J. Hjerling-Leffler, T. Karayannis, et al., Genetic fate mapping reveals that the caudal ganglionic eminence produces a large and diverse population of superficial cortical interneurons, J. Neurosci. 30 (2010) 1582e1594. [17] L. Tricoire, K.A. Pelkey, B.E. Erkkila, et al., A blueprint for the spatiotemporal origins of mouse hippocampal interneuron diversity, J. Neurosci. 31 (2011) 10948e10970. [18] Y. Kubota, N. Shigematsu, F. Karube, et al., Selective coexpression of multiple chemical markers defines discrete populations of neocortical GABAergic neurons, Cereb. Cortex 21 (2011) 1803e1817. [19] R. Tremblay, S. Lee, B. Rudy, GABAergic interneurons in the neocortex: from cellular properties to circuits, Neuron 91 (2016) 260e292. [20] B. Wamsley, G. Fishell, Genetic and activity-dependent mechanisms underlying interneuron diversity, Nat. Rev. 18 (2017) 299e309. €nneke, B. Scheuer, R.J. Wagener, et al., Staiger. Characterizing VIP neu[21] A. Pro rons in the barrel cortex of VIPcre/tdTomato mice reveals layer-specific differences, Cereb. Cortex 25 (2015) 4854e4868. [22] A. Kepecs, G. Fishell, Interneuron cell types are fit to function, Nature 505 (2014) 318e326. [23] S. Lee, I. Kruglikov, Z.J. Huang, et al., A disinhibitory circuit mediates motor integration in the somatosensory cortex, Nat. Neurosci. 16 (2013) 1662e1670. [24] H.J. Pi, B. Hangya, D. Kvitsiani, et al., Cortical interneurons that specialize in disinhibitory control, Nature 503 (2013) 521e524. [25] X. Jiang, G. Wang, A.J. Lee, et al., The organization of two new cortical interneuronal circuits, Nat. Neurosci. 16 (2013) 210e218. [26] T. Sugawara, Y. Tsurubuchi, K.L. Agarwala, et al., A missense mutation of the Naþ channel aII subunit gene Nav1.2 in a patient with febrile and afebrile seizures causes channel dysfunction, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 6384e6389. [27] K. Kamiya, M. Kaneda, T. Sugawara, et al., A nonsense mutation of the sodium channel gene SCN2A in a patient with intractable epilepsy and mental decline, J. Neurosci. 24 (2004) 2690e2698. [28] I. Ogiwara, K. Ito, Y. Sawaishi, et al., De novo mutations of voltage-gated sodium channel aII gene SCN2A in intractable epilepsies, Neurol. 73 (2009) 1046e1053. [29] K. Nakamura, M. Kato, H. Osaka, et al., Clinical spectrum of SCN2A mutations expanding to Ohtahara syndrome, Neurology. 81 (2013) 992e998. [30] A. Hoischen, N. Krumm, E.E. Eichler, Prioritization of neurodevelopmental disease genes by discovery of new mutations, Nat. Neurosci. 17 (2014) 764e772. [31] M.R. Johnson, K. Shkura, S.R. Langley, et al., Systems genetics identifies a convergent gene network for cognition and neurodevelopmental disease, Nat. Neurosci. 19 (2016) 223e232. [32] K. Yamakawa, Mutations of voltage-gated sodium channel genes SCN1A and SCN2A in epilepsy, intellectual disability, and autism, in: C. Sara, C. Verpelli (Eds.), Synaptic Dysfunction in Autism Spectrum Disorder and Intellectual Disability, Elsevier, Amsterdam, Netherlands, 2016. [33] M. Wolff, K.M. Johannesen, U.B. Hedrich, et al., Genetic and phenotypic heterogeneity suggest therapeutic implications in SCN2A-related disorders, Brain 140 (2017) 1326e1336. [34] R. Ben-Shalom, C.M. Keeshen, K.N. Berrios, et al., Opposing effects on NaV1.2 function underlie differences between SCN2A variants observed in individuals with autism spectrum disorder or infantile seizures, Biol. Psychiatry 82 (2017) 224e232.
Please cite this article in press as: T. Yamagata, et al., Nav1.2 is expressed in caudal ganglionic eminence-derived disinhibitory interneurons: Mutually exclusive distributions of Nav1.1 and Nav1.2, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/ 10.1016/j.bbrc.2017.08.013