European Journal of Neuroscience, Vol. 39, pp. 883–892, 2014

doi:10.1111/ejn.12474

NEUROSYSTEMS

Differential subcellular localization of SK3-containing channels in the hippocampus Carmen Ballesteros-Merino,1 Masahiko Watanabe,2 Ryuichi Shigemoto,3 Yugo Fukazawa,3 John P. Adelman4 and
n1 Rafael Luja 1

Instituto de Investigacio´n en Discapacidades Neurolo´gicas (IDINE), Departamento de Ciencias Me´dicas, Facultad de Medicina, Universidad Castilla-La Mancha, Campus Biosanitario, Albacete, Spain 2 Department of Anatomy, Hokkaido University School of Medicine, Sapporo, Japan 3 Division of Cerebral Structure, National Institute for Physiological Sciences, Okazaki, Japan 4 Vollum Institute, Oregon Health & Science University, Portland, OR, USA Keywords: cell surface distribution, development, electron microscopy, immunohistochemistry, mouse, potassium channel

Abstract Small-conductance, Ca2+-activated K+ (SK) channels are expressed in the hippocampus where they regulate synaptic responses, plasticity, and learning and memory. To investigate the expression of SK3 (KCNN3) subunits, we determined the developmental profile and subcellular distribution of SK3 in the developing mouse hippocampus using western blots, immunohistochemistry and high-resolution immunoelectron microscopy. The results showed that SK3 expression increased during postnatal development, and that the localization of SK3 changed from being mainly associated with the endoplasmic reticulum and intracellular sites during the first postnatal week to being progressively concentrated in dendritic spines during later stages. In the adult, SK3 was localized mainly in postsynaptic compartments, both at extrasynaptic sites and along the postsynaptic density of excitatory synapses. Double labelling showed that SK3 co-localized with SK2 (KCNN2) and with N-methyl-D-aspartate receptors. Finally, quantitative analysis of SK3 density revealed two subcellular distribution patterns in different hippocampal layers, with SK3 being unevenly distributed in CA1 region of the hippocampus pyramidal cells and homogeneously distributed in dentate gyrus granule cells. Our results revealed a complex cell surface distribution of SK3-containing channels and a distinct developmental program that may influence different hippocampal functions.

Introduction Principal cells of the hippocampal formation (pyramidal and granule cells) are required for hippocampal information processing (Anderson et al., 2007). The main trisynaptic circuit in the hippocampus that relays cortical input to granule cells in the dentate gyrus (DG) on to pyramidal neurons in the CA3 region of the hippocampus (CA3) and then to pyramidal neurons in the CA1 region of the hippocampus (CA1) is important for several forms of learning and memory. These and other functions of principal cells depend on electrical signals, the generation, patterns, and propagation of which are determined largely by the complement of ion channels that they express. Indeed, the dysfunction of ion channels known to regulate hippocampal neurophysiology is linked to pathologies including schizophrenia, epilepsy and Alzheimer’s disease (Poolos & Johnston, 2012; MacDonald et al., 2013). Small-conductance Ca2+-activated K+ (SK; KCNN) channels are one type of ion channel shaping the synaptic responses of both pyramidal and granule cells. SK channels are voltage-independent and directly

Correspondence: Rafael Luj
an, as above. E-mail: [email protected] Received 6 November 2013, revised 3 December 2013, accepted 3 December 2013

gated by changes in intracellular Ca2+ concentration and hence function as feedback regulators of neuronal activity (Kohler et al., 1996; Bond et al., 2005; Ngo-Anh et al., 2005; Luján et al., 2009). There are four members of the mammalian SK gene family (KCNN1/KCa2.1/SK1, KCNN2/KCa2.2/SK2, KCNN3/KCa2.3/SK3, KCNN4/KCa3.1/SK4), with SK1, SK2 and SK3 being expressed in the brain with overlapping yet distinct expression profiles (Kohler et al., 1996; Ishii et al., 1997; Stocker & Pedarzani, 2000; Sailer et al., 2004; Armstrong et al., 2005). Indeed, SK2 and SK3 subunits co-assemble to form heteromeric channels in the brain (Strassmaier et al., 2005). In situ hybridization and light-level immunohistochemistry show that SK3 expression is regionally restricted in the brain and is highly enriched in dopamine neurons of the ventral midbrain (Stocker & Pedarzani, 2000; Tacconi et al., 2001; Sarpal et al., 2004), where it is an important regulator of spike pattern generation in dopamine neurons (Deignan et al., 2012; Soden et al., 2013). The functional significance of the SK3 channel protein has also been investigated in the context of hippocampal function. For instance, gene-silencing of SK3 in mice resulted in short-term memory and long-term potentiation deficits (Jacobsen et al., 2009). SK3 is up-regulated in the hippocampus of aged mice, and its down-regulation by antisense oligonucleotides has been reported to reverse age-related deficits in hippocampus-dependent memory tasks and long-term potentiation

© 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd

884 C. Ballesteros-Merino et al. (Blank et al., 2003). Interestingly, SK2 is expressed in hippocampal dendrites and spines, close to N-methyl-D-aspartate (NMDA) receptors (Lin et al., 2008; Ballesteros-Merino et al., 2012). However, there is little information about the expression profile and subcellular localization of SK3 in hippocampal neurons. To understand the molecular mechanisms underlying the function and dysfunction of SK channels in the hippocampus, it is essential to identify the cellular and subcellular compartments in which SK3containing channels are operative. Therefore, in the present study, we used biochemical and high-resolution immunoelectron microscopy approaches to determine the developmental expression profile and surface distribution of SK3 in two populations of principal cells in the hippocampus, CA1 pyramidal cells and DG granule cells. Our findings show that synaptic localization is acquired during postnatal development and reveal a gradient of SK3 along the somato-dendritic domains of CA1 pyramidal neurons in the adult. The findings presented in this article represent the first direct demonstration that SK3-containing channels are located in both the pre- and postsynaptic compartments of hippocampal pyramidal neurons.

Materials and methods Tissue preparation The OF-1 mice, from the day of birth [postnatal day (P)0] to adulthood (obtained from the Animal House Facility, School of Medicine, University of Castilla-La Mancha), were used in this study for western blots and pre-embedding immunohistochemical analyses. The care and handling of animals prior to and during the experimental procedures were in accordance with Spanish (RD 1201/2005) and European Union (86/609/EC) regulations, and the protocols were approved by the University’s Animal Care and Use Committee. For each developmental stage, the animals used were from different litters and were grouped as follows: P0, P5, P7, P10, P12, P15, P21, P60 and P90, n = 3 per group for immunoblots; P0, P5, P10, P12, P15, P21 and P60, n = 3 per group for light microscopic immunohistochemistry; and P60, n = 3 animals for each of the three techniques at the electron microscopic level. For immunoblotting, animals were deeply anaesthetized by hypothermia (P0-P5) or by intraperitoneal injection of ketamine–xylazine 1 : 1 (0.1 mL/kg b.w.) and the brains were quickly frozen. For immunohistochemistry, animals were anaesthetized by intraperitoneal injection of ketamine–xylazine 1 : 1 (0.1 mL/kg b.w.) and transcardially perfused with ice-cold fixative containing 4% paraformaldehyde, with or without 0.05% glutaraldehyde and 15% (v/v) saturated picric acid made up in 0.1 M phosphate buffer (PB) (pH 7.4). After perfusion, the brains were removed and immersed in the same fixative for 2 h or overnight at 4 °C. Tissue blocks were washed thoroughly in 0.1 M PB. Coronal 60-lm-thick sections were cut on a Vibratome (Leica V1000). For sodium dodecyl sulphate– freeze-fracture replica labelling (SDS-FRL), animals were anaesthetized by intraperitoneal injection of ketamine–xylazine 1 : 1 (0.1 mL/ kg b.w.) and transcardially perfused with ice-cold fixative containing 2% paraformaldehyde made up in 0.1 M PB for 12 min. Coronal 130-lm-thick sections were cut on a Vibratome before the immersion of fixed slices in 30% glycerol in 0.1 M PB. Replicas were obtained as described previously (Tarusawa et al., 2009).

in this study. An affinity-purified polyclonal antibody against SK2 was raised in guinea pig and characterized previously (Cueni et al., 2008; Lin et al., 2008). The monoclonal antibody against GluN1 (clone 54.1 MAB363) was obtained from Millipore (Germany), the monoclonal antibody against Postsynaptic density protein 95 (PSD95) was obtained from Abcam (Cambridge, UK) and the antibodies against calbindin were from Chemicon (Temecula, CA, USA). Western blots Hippocampi were homogenized in 320 mM sucrose, 2 mM EDTA, 10 mM HEPES, pH 7.4, and protease inhibitor cocktail (104 mM 4-(2Aminoethyl) benzenesulfonyl fluoride (AEBSF), 80 lM aprotinin, 4 mM bestatin, 1.4 mM E-64, 2 mM leupeptin and 1.5 mM pepstatin A; Sigma–Aldrich) with a pestle motor (Sigma–Aldrich). The homogenized tissue was centrifuged at 1900 g at 4 °C and the supernatant was ultracentrifuged at 47 000 g at 4 °C (Optima L-90K Ultracentrifuge, Beckman Coulter, CA, USA) using rotor SW40Ti (Beckman Coulter). The membrane protein (100 lg) was prepared as western blots and probed with anti-SK3 (1 : 500). Protein bands were visualized after the application of goat anti-rabbit secondary antiserum coupled to horseradish peroxidase (1 : 3000) using the ECL blotting detection kit (SuperSignal West Dura, Pierce, Rockford, USA). Blots were quantified by densitometry using an LAS4000 MINI (Fujifilm, Japan). A series of primary and secondary antibody dilutions and incubation times were used to optimize the experimental conditions for the linear sensitivity range, confirming that our labelling was well below saturation levels (Schilling & Aletsee-Ufrecht, 1998). To normalize the SK3 expression at different developmental ages, we used the band for a-tubulin as control [these blots were included in a previous work (Ballesteros-Merino et al., 2012)]. Immunohistochemistry for light microscopy Immunohistochemical reactions at the light microscopic level were carried out using the immunoperoxidase method as described previously (Luj
an et al., 1996). Briefly, sections were incubated in 10% normal goat serum (NGS) diluted in 50 mM Tris buffer (pH 7.4) containing 0.9% NaCl [Tris-buffered saline (TBS)], with 0.2% Triton X-100, for 1 h. Sections were incubated in anti-SK3 (1–2 lg/ mL diluted in TBS containing 1% NGS), followed by incubation in biotinylated goat anti-rabbit IgG (Vector Laboratories, Burlingame, CA, USA) in TBS containing 1% NGS. Sections were then transferred into avidin–biotin–peroxidase complex (ABC kit, Vector Laboratories). Bound peroxidase enzyme activity was revealed using 3,3′-diaminobenzidine tetrahydrochloride [0.05% in Tris-Cl buffer (TB), pH 7.4] as the chromogen and 0.01% H2O2 as the substrate. Finally, sections were air-dried and coverslipped prior to observation in a photomicroscope (Eclipse 80i, Nikon) equipped with differential interference contrast optics and a digital imaging camera. Immunohistochemistry for electron microscopy Immunohistochemical reactions at the electron microscopic level were carried out using the immunogold methods as described previously (Luj
an et al., 1996). Ultrastructural analyses were performed in a Jeol-1010 electron microscope.

Antibodies and chemicals

Pre-embedding immunogold method

The characteristics and specificity of the anti-SK3 antibodies have been described elsewhere (Deignan et al., 2012) and are also provided

Briefly, free-floating sections were incubated in 10% NGS diluted in TBS. Sections were then incubated in anti-SK3 antibodies (1–2

© 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd European Journal of Neuroscience, 39, 883–892

Localization of the SK3 subunit 885 lg/mL diluted in TBS containing 1% NGS), followed by incubation in goat anti-rabbit IgG coupled to 1.4 nm gold (Nanoprobes Inc., Stony Brook, NY, USA). Sections were postfixed in 1% glutaraldehyde and washed in double-distilled water, followed by silver enhancement of the gold particles with an HQ Silver kit (Nanoprobes Inc.). Sections were then treated with osmium tetraoxide (1% in 0.1 M PB), block-stained with uranyl acetate, dehydrated in graded series of ethanol and flat-embedded on glass slides in Durcupan (Fluka) resin. Regions of interest were cut at 70–90 nm on an ultramicrotome (Reichert Ultracut E, Leica, Austria) and collected on 200-mesh copper grids. Staining was performed on drops of 1% aqueous uranyl acetate followed by Reynolds’s lead citrate. Postembedding immunogold method Briefly, ultrathin 80-nm-thick sections from Lowicryl-embedded blocks of the hippocampus were picked up on coated nickel grids and incubated on drops of a blocking solution consisting of 2% human serum albumin in 0.05 M TBS and 0.03% Triton X-100. The grids were incubated with SK3 antibodies (10 lg/mL in 0.05 M TBS and 0.03% Triton X-100 with 2% human serum albumin) at 28 °C overnight. The grids were incubated on drops of goat antirabbit IgG conjugated to 10 nm colloidal gold particles (Nanoprobes Inc.) in 2% human serum albumin and 0.5% polyethylene glycol in 0.05 M TBS and 0.03% Triton X-100. The grids were then washed in TBS and counterstained for electron microscopy with saturated aqueous uranyl acetate followed by lead citrate. Sodium dodecyl sulphate–freeze-fracture replica labelling technique SDS-FRL was performed with some modifications to the original method described by Fujimoto (1995). Replicas were transferred to 2.5% sodium dodecyl sulphate containing 0.0625 M Tris and 10% glycerol (pH 6.8) for 16 h at 80 °C with shaking, and then washed and reacted with a mixture of polyclonal rabbit antibody for SK3 and monoclonal mouse antibody for PSD-95 at 15 °C overnight. Following three washes in 0.1% Bovine Serum Albumin in TBS and blocking in 5% Bovine Serum Albumin/TBS, replicas were incubated in a mixture of secondary antibodies coupled to gold particles (British Biocell International, UK) overnight at 4 °C. When one of the primary antibodies was omitted, no labelling activity of the secondary antibodies was observed. After immunogold labelling, the replicas were immediately rinsed three times with 0.1% Bovine Serum Albumin/TBS, washed twice with distilled water, and picked up onto grids coated with pioloform (Agar Scientific, Stansted, Essex, UK). The specificity of SDS-FRL immunolabelling was confirmed in samples from SK3 null mice. Quantification of SK3 channel immunoreactivity To establish the relative abundance of SK3 channel immunoreactivity in different compartments of pyramidal cells during development (P5, P15 and P21) and in the adult (P60), we used 60-lm-thick coronal slices processed for pre-embedding immunogold immunohistochemistry. The procedure was similar to that used previously (Luj
an et al., 1996; Luj
an & Shigemoto, 2006). Briefly, for each of three animals of different postnatal ages and adult, three samples of tissue were obtained for the preparation of embedding blocks (totalling nine blocks for each age). To minimize false negatives, electron microscopic serial ultrathin sections were cut close to the surface of each block, as immunoreactivity decreased with depth. We estimated the quality of immunolabelling by always selecting areas with opti-

mal gold labelling at approximately the same distance from the cutting surface. Randomly selected areas were then photographed from the selected ultrathin sections and printed with a final magnification of 45 000 9. Quantification of immunogold labelling was carried out in reference areas totalling approximately 1800 lm2 for each age. Immunoparticles identified in each reference area and present in different subcellular compartments (dendritic spines, dendritic shafts, axon terminals and somata) were counted. The data were expressed as a percentage of immunoparticles in each subcellular compartment and/or at different dendritic layers. To establish the density of SK3 at extrasynaptic sites in dendritic spines of CA1 pyramidal cells and granule cells in the adult, quantification of immunolabelling was performed in 60-lm-thick coronal slices processed for pre-embedding immunogold in the following layers: the distal part of the stratum radiatum (SR) and the stratum lacunosum-moleculare (SLM) of CA1 pyramidal cells, and the dendrites of the inner one-third and outer two-thirds of granule cells in the molecular layer of the dendate gyrus. For each of the three adult animals, three samples of tissue were obtained for the preparation of embedding blocks (totalling nine blocks for each age). Randomly selected areas were then photographed from the selected ultrathin sections and printed with a final magnification of 45 000 9. Quantification of immunogold labelling was carried out in reference areas totalling approximately 2000 lm2 for each age. Immunoparticles identified in dendritic spines were counted and the area of each spine was measured. The data (density of SK3 in spines in each dendritic subfield) were expressed as the number of immunoparticles/lm2 . To establish the tangential distribution of SK3 at postsynaptic densities (PSDs) of excitatory synapses in spines of CA1 pyramidal cells, quantification of immunolabelling was performed on ultrathin sections processed for postembedding immunogold. The radial location of immunoparticles was measured from the midline of the PSD and normalized across the synapse population to control for variable size. The distribution obtained was mirrored across the midline for display. For the analysis, 174 immunoparticles for SK3 were identified in 83 synapses, 169 immunoparticles for SK2 were identified in 79 synapses and 258 immunoparticles for GluN1 were identified in 81 synapses. The data were expressed as a percentage of immunoparticles along the PSD length. Controls To test method specificity in the procedures for both light and electron microscopy, antiserum against SK3 was tested on brain slices of SK3 knockout mice (Fig. S1). After immunohistochemical experiments, the signal disappeared completely in areas where a strong signal was present in control sections (Fig. S1). Furthermore, the primary antibody was either omitted or replaced with 5% (v/v) normal serum of the species of the primary antibody, resulting in total loss of the signal. Labelling patterns were also compared with those obtained by calbindin; only the antibodies against SK3 consistently labelled the plasma membrane (PM). Data analysis Statistical analyses for morphological data were performed using SigmaStat Pro (Jandel Scientific) and data were presented as mean  SEM. The statistical significance was defined as P < 0.05, as determined using ANOVA followed by the Bonferroni test for multiple comparisons. For the electron microscopic data, statistical significance in the distribution of gold particles among

© 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd European Journal of Neuroscience, 39, 883–892

886 C. Ballesteros-Merino et al. samples was assessed with the Kolmogorov–Smirnov non-parametric test.

Results

A1

A2

Developmental profile of SK3 protein in the hippocampus The developmental profile of SK3 protein expression was determined during postnatal development in the hippocampus. Western blots were prepared using equal amounts of membrane protein obtained from the hippocampi of individual mice at different developmental ages. Probing the blots with SK3 antibody labelled a band of approximately 87 kDa at each developmental age (Fig. 1A), consistent with the size of SK3 (Strassmaier et al., 2005), and showed that SK3 expression increased after birth, reaching a steady state by P60 (Figs 1A and B). Densitometry measurements from six different experiments were averaged to compare protein expression at each age and revealed a ninefold increase in SK3 protein expression between P0 and P21, and a sevenfold increase between P0 and P60. In the adult (P60-P90), SK3 expression decreased significantly to that detected at P21 (Fig. 1B). The band for a-tubulin was used as a control for normalizing SK3 expression at different developmental ages.

B1

C1

D1

B2

C2

D2

SK3 distribution profile is dynamic during development To investigate the expression and distribution profiles of SK3 during postnatal development of the hippocampus, we carried out immunohistochemical studies using light microscopy (Fig. 2). At birth (P0) and at P5, SK3 was intensely expressed in the principal cell layers of all hippocampal areas, whereas very weak labelling was detected in the dendritic layers (Figs 2A and B). During the second postnatal week (P10 and P12), the distribution profile of SK3 changed, show-

E1

E2

A

F1

F2

B

G1

Fig. 1. Quantitative immunoblot analysis of SK3 protein expression in the mouse hippocampus during postnatal development. (A and B) Using western blots, the SK3 antibody recognized a band of approximately 87 kDa. During postnatal development, SK3 protein was detected at birth, and increased ninefold between P0 and P21 but only sevenfold between P10 and P60. The position of molecular weight markers corresponding to 75 and 100 kDa is indicated. Densitometry measurements from nine independent experiments were averaged for each developmental age and normalized to P60. The band for a-tubulin, shown in Ballesteros-Merino et al. (2012), was used as a control for normalizing SK3 expression at different developmental ages. Error bars indicate SEM. ***P < 0.01.

G2

Fig. 2. Immunoreactivity for SK3 in the hippocampus during postnatal development using a pre-embedding immunoperoxidase method. (A and B) At P0 and P5, immunoreactivity for SK3 was strong in the principal cell layers of all hippocampal areas, and very weak in the dendritic layers. (C and D) At P10 and P12, the distribution pattern of SK3 changed, showing weaker immunoreactivity in the pyramidal and granule cell layers, and stronger immunoreactivity in the neuropile of the SLM (slm) of the CA1 region and molecular layer (ml) of the DG. (E) At P15, a similar distribution for SK3 as at P10-P12 was observed but showing stronger neuropilar labelling of the SLM (slm) of the CA1 region and molecular layer (ml) of the DG, which was more evident at P12. (F and G) At P21 and P60, immunoreactivity for SK3 was very weak in the pyramidal cell layer (sp) and granule cell layer (gc) and intense in the neuropile of the SLM (slm) of the CA1 region and molecular layer (ml) of the DG. Immunoreactivity for SK3 in the strata oriens (so) and SR (sr) was consistently lower than in the SLM (slm) throughout postnatal development. Frames indicate the region selected for panels in the right column. h, hilus. Scale bars: A1–G1, 0.5 lm; A2–G2, 0.1 lm.

© 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd European Journal of Neuroscience, 39, 883–892

Localization of the SK3 subunit 887 ing decreased labelling at the pyramidal cell layer but intense labelling in the neuropil of all dendritic layers, particularly in the SLM of CA1 and CA3, and the molecular layer of the DG (Figs 2C and D). This profile persisted at the beginning of the third postnatal week (P15), but expression was even stronger in the SLM of CA1 and CA3, and the molecular layer of the DG (Fig. 2E). However, during the fourth postnatal week (P21), SK3 immunolabelling was increased throughout the dendritic layers of the hippocampus (Fig. 2F), and this profile was unchanged to adulthood (P60). Thus, in the CA1 and CA3 regions, there was weak labelling in the pyramidal cell layer, low to moderate labelling in the neuropil of the stratum oriens and SR and intense labelling in the neuropil. There was very weak labelling in the granule cell layer, and intense labelling in the neuropil of the molecular layer (Figs 2F and G). Developmental profile of SK3 plasma membrane expression The results presented above show that SK3 expression progressively shifted into the dendrites. To determine the precise subcellular localization of SK3 during development, we carried out immnunoelectron microscopic studies at different times (P5, P15 and P21) in pyramidal cells of the CA1 region using pre-embedding immunogold labelling; we restricted quantitative analysis to the somata and dendritic spines. In spines, this included all particles localized to the extrasynaptic membrane, not localized within the main body of the PSD. Consistent with the light microscopic data, SK3 was expressed in pyramidal cells early in development, and the cell surface distribution changed significantly during postnatal development. Thus, at P5, the majority of immunoparticles were in the soma (96% of all immunoparticles) and essentially all of those (99%) were associated with the rough endoplasmic reticulum (Fig. 3A; Table 1). Of the immunoparticles observed in the neuropil (SR: 1.5% of all immunoparticles; SLM: 2.4% of all immunoparticles), most were detected at intracellular sites (SR: 92%; SLM: 89%) and some (SR: 8%; SLM: 11%) were found along the PM of the spines, as well as on the dendritic shafts (Table 1). Thus, at P5, SK3 was largely intracellular. At P15, fewer immunoparticles for SK3 were associated with the rough endoplasmic reticulum than at P5, with 78% of immunoparticles localized to the soma, with 76% of those associated with the rough endoplasmic reticulum and 24% on the PM (Table 1). In the dendritic fields, 8% of all SK3 immunoparticles were detected in dendrites and spines in the SR and 34% of those were on the PM, whereas in the SLM, 14% of all SK3 immunoparticles were found in dendrites and spines, with 35% of them being located in the PM

A

(Figs 3B and C; Table 1). This dynamic developmental shift continued such that by P21 relatively few immunoparticles for SK3 were still associated with the rough endoplasmic reticulum; only 32% were in the soma and, of those, only 6% were on the PM (Fig. 3D; Table 1). In the SR (25% of all immunoparticles), 28% were on the PM, whereas in the SLM (43% of all immunoparticles) 47% were on the PM (Fig. 3D; Table 1). Synaptic and extrasynaptic localization of SK3 channels in mature hippocampal cells Pre-embedding immunogold labelling permits high-resolution subcellular localization, but it does not efficiently detect proteins localized to the dense matrix of the postsynaptic membrane. Therefore, we expanded our technical approach to additionally include postembedding and SDS-FRL, and examined SK3 expression in CA1 pyramidal neurons and DG granule cells, the two hippocampal neuronal subtypes showing the strongest SK3 immunoreactivity in adult (P60) mice. CA1 pyramidal cells Using pre-embedding, SK3 was primarily found at postsynaptic sites along the extrasynaptic PM and also at intracellular sites associated with the cisterna of dendritic shafts and the spine apparatus in the SLM (Fig. 4A–C). To a lesser extent, immunoparticles for SK3 were also detected at presynaptic axon terminals (Fig. 4B), always outside the active zone. Quantitative analysis performed on the postsynaptic compartments showed that only 32% of all SK3 immunoparticles were in the soma and, of those, only 6% were on the PM (Table 2). In the SR (25% of all immunoparticles), 28% were on the spine PM, whereas in the SLM (43% of all immunoparticles), 47% were on the spine PM (Table 2). Overall, the distribution of immunoparticles for SK3 at P60 was very similar to that described for P21. We used the postembedding immunogold technique to determine whether SK3 was localized at synaptic sites in excitatory synapses of pyramidal cells. The results showed that immunoparticles for SK3 were present within the postsynaptic specialization, as well as along the extrasynaptic and perisynaptic PM of dendritic spines establishing asymmetrical synapses with axon terminals (Fig. 4D). To further define SK3 localization within the synaptic specialization and to investigate if SK channels were co-localized with NMDA receptors, we used double-labelling approaches for SK3 and GluN1, the obligate NMDA receptor subunit. Of 105 spines with SK3 im-

B

C

D

Fig. 3. Electron micrographs of the CA1 region of the hippocampus showing immunogold particles for SK3 during postnatal development, using a pre-embedding immunogold method. (A) At P5, immunoparticles for SK3 were mainly associated with the rough endoplasmic reticulum (ER) in the cytoplasm (crossed arrows) of pyramidal cells. (B and C) At P15, in the SLM most immunoparticles for SK3 were mainly detected at intracellular sites (crossed arrows) in the dendritic shafts (Den) and spines (s) of pyramidal cells, and only a few immunoparticles were found along the extrasynaptic PM of dendritic spines (s) (arrow in C). (D) At P21, in the SLM most immunoparticles for SK3 were mainly localized along the extrasynaptic PM (arrow) of dendritic spines (s). A few immunoparticles for SK3 (arrowhead) were also observed at presynaptic sites in axon terminals (at), presumably from the enthorinal cortex, establishing excitatory synapses with dendritic spines. Scale bar: A–D, 0.5 lm. © 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd European Journal of Neuroscience, 39, 883–892

888 C. Ballesteros-Merino et al. Table 1. Summary of immunogold labelling for extrasynaptic SK3 during postnatal development in CA1 pyramidal cells

PM gold

% PM

Intracellular gold

% Intracellular

Total gold

P5 Soma SR SLM

23 3 7

0.9 7.9 11.3

2436 35 55

99.1 92.1 88.7

2459 38 62

P15 Soma SR SLM

51 65 113

2.7 34.2 35.2

1811 125 208

97.3 65.8 64.8

1862 190 321

P21 Soma SR SLM

21 76 213

6.2 28.3 46.9

315 193 241

93.8 71.7 53.1

336 269 454

expressed in the dendrites and spines, including the postsynaptic membrane, of CA1 pyramidal cells (Lin et al., 2008), and SK2 and SK3 co-immmunoprecipitated from whole-brain lysates. Therefore, we investigated whether SK3 and SK2 were co-expressed in the same postsynaptic specialization. Among 102 spines that were immunoreactive for SK3, 95 were also immunolabelled for SK2 (Fig. 4F). Moreover, quantitative analysis of the tangential distribution of SK3-, SK2- and GluN1-specific gold particles along the PSD showed a similar distribution, with the overlap being strongest toward the middle of the PSD (Fig. 4G–I). Therefore, SK3, SK2, and NMDA receptors co-habited the same microdomain within the PSD.

Dentate gyrus granule cells

munoparticles, 103 were also immunolabelled for GluN1 (Fig. 4E), indicating that most, if not all, dendritic spines co-expressed SK3 and NMDA receptors. We have previously shown that SK2 was

The analysis was performed in the inner one-third and outer twothirds of the molecular layer. Similar to pyramidal cells, SK3 immunoreactivity in granule cells was primarily found at postsynaptic sites along the extrasynaptic PM, both within the PM and in intracellular membranes of dendritic shafts and spines (Figs 5A and B). SK3 was also located at presynaptic sites, in axon terminals

C

A

B

D

G

E

H

F

I

Fig. 4. Electron micrographs of the hippocampus showing immunoparticles for SK3 in the SLM of the CA1 region, as detected using immunogold methods in the adult (P60) mice. (A–C) Using the pre-embedding inmmunogold method, immunoparticles for SK3 were mainly detected along the extrasynaptic PM (arrows) of dendritic spines (s) of CA1 pyramidal cells establishing asymmetrical synapses with axon terminals (at). SK3 immunoparticles were also observed intracellularly associated with intracellular membranes (crossed arrows) of the dendritic spines (s) and dendritic shafts (Den), as well as at presynaptic sites (arrowheads) along the extrasynaptic PM of axon terminals (at) establishing excitatory synapses with dendritic spines (s). (D–F) Synaptic localization of SK3 using the postembedding inmmunogold method. SK3 immunoparticles were detected along the PSD of dendritic spines (s) of CA1 pyramidal cells (arrows) establishing asymmetrical synapses with axon terminals (at), as well as at extrasynaptic sites (crossed arrow). Double-labelling immunogold methods showing immunoreactivity for SK3 (10 nm particles) and the GluN1 subunit of NMDA receptors (20 nm particles, E) or SK2 channels (20 nm particles, F). Immunoparticles for SK3 co-localized with GluN1- or SK2-labelled immunoparticles at postsynaptic densities of excitatory synapses in dendritic spines (s) of pyramidal cells establishing synapses with axon terminals (at), presumably from the entorhinal cortex. Quantitative analysis showing the tangential distribution of immunoparticles for SK3 (G), SK2 (H) and GluN1 (I) across the PSD of CA1 excitatory synapses. The radial location of immunoparticles was measured from the midline of the PSD and the distribution was mirrored across the midline for display. Labelling for SK3, SK2 and NR1 was preferentially located at the middle of the PSD, and their densities decreased toward the edges of the synapse. Scale bars: A–C, 0.5 lm; D–F, 0.2 lm. © 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd European Journal of Neuroscience, 39, 883–892

Localization of the SK3 subunit 889 Table 2. Summary of immunogold labelling for extrasynaptic SK3 in adult (P60) CA1 pyramidal cells and DG granule cells

% PM

Intracellular gold

% Intracellular

Total

CA1 pyramidal cells Soma 22 SR 75 SLM 212

7.2 29.1 45.9

284 183 250

92.8 70.9 54.1

306 258 462

DG granule Soma iML oML

6.7 44.6 43.9

265 229 245

93.3 55.4 56.1

284 413 436

PM gold

cells 19 184 191

iML, inner one-third of the molecular layer; oML, outer two-thirds of the molecular layer.

establishing excitatory synapses with dendritic spines of granule cells (Figs 5A and B). Quantitative analysis showed that 36% of all SK3 immunoparticles were located in dendrites and spines in the inner one-third of the molecular layer and, of those, 47% were on the PM (Table 2). In the outer two-thirds of the molecular layer, 38% of all SK3 immunoparticles were located in dendrites and spines and, of those, 44% were on the PM (Table 2). The soma of granule cells showed the lowest immunoreactivity for SK3 (25% of all immunoparticles), with 93% being associated with the rough endoplasmic reticulum and 7% with the PM (Table 2). Therefore, the distribution of SK3 was virtually identical between the inner one-third and outer two-thirds of the molecular layer. Using the postembedding immunogold technique, immunoparticles for SK3 were seen within the postsynaptic specialization, as well as along the extrasynaptic and perisynaptic PM of dendritic spines establishing asymmetrical synapses with axon terminals (Figs 5C and D). Clustering of SK3 in spines of hippocampal cells To investigate more precisely the location of SK3 along the postsynaptic PM of CA1 and DG spines in the adult, we used the SDSFRL technique. SDS-FRL permits the unparallelled localization of

A

membrane protein, beyond the limitations of standard thin-section electron microscopy (Fujimoto, 1995). Immunoparticles for SK3 were clustered in the dendritic spine PM of pyramidal cells and granule cells (Figs 6A and C). To define the synaptic specialization of excitatory synapses, immunogold labelling for PSD-95, a specific marker of the postsynaptic membrane, was performed in concert with SK3 immunogold labelling. Both sets of gold particles were localized to the protoplasmic face (Fig. 6), reflecting the intracellular location of the epitopes detected by the antibodies for the two proteins. Consistent with the results of pre- and postembedding experiments at P60, immunolabelling revealed co-clustering of SK3 and PSD-95 immunoparticles, demonstrating synaptic SK3 location (Figs 6B and D). In summary, SK3 channels were expressed in the PSD as well as on the extrasynaptic spine membrane. The specificity of immunolabelling using the SDS-FRL technique was confirmed in samples from SK3 null mice (Fig. S1). Uniform vs. non-uniform gradient of SK3 in the cell surface of hippocampal cells The laminar arrangement of pyramidal and granule cells allowed us to investigate the distribution of SK3 along the dendritic tree of principal cells as it ramified through the different subfields in the adult hippocampus. Therefore, we analysed SK3 PM distribution in spines as a function of distance from the soma (immunoparticles/ lm2; Fig. 7) in two neuronal populations, i.e. CA1 pyramidal cells and granule cells of the DG (Fig. 7A). In this pre-embedding study, we restricted quantitative analysis to the somata and dendritic spines. In CA1 pyramidal cells, the density of SK3 was low in the somata (0.2  0.01 immunoparticles/lm2) and in the SR (1.4  0.2 immunoparticles/lm2) compared with the highest density detected in the SLM (10.3  0.6 immunoparticles/lm2) (Fig. 7B). In granule cells, the density of SK3 was low in the somata (0.3  0.01 immunoparticles/lm2), but it was similar in the inner one-third (11.6  0.6 immunoparticles/lm2) and outer two-thirds (11.4  0.7 immunoparticles/lm2) of the molecular layer (Fig. 7C). We also calculated the non-specific labelling density in every reaction in the nuclei of pyramidal and granule cells, a subcellular compartment that should not contain any SK3; the immunoparticle

B

C

D

Fig. 5. Electron micrographs of the hippocampus showing immunoparticles for SK3 in the molecular layer of the DG, as detected using the pre- and postembedding immunogold methods in the adult (P60) mice. (A and B) Using the pre-embedding inmmunogold method, SK3 immunoparticles were more frequently detected along the extrasynaptic PM (arrows) of granule cell dendritic spines (s) establishing asymmetrical synapses with axon terminals (at), in both the inner one-third (A) and outer two-thirds (B) of the molecular layer. Immunoparticles were also detected at intracellular sites (crossed arrows) of dendritic spines (s) and dendritic shafts (Den) of granule cells. To a lesser extent, SK3 immunoparticles were detected at presynaptic sites along the extrasynaptic PM (arrowheads) of axon terminals (at). (C and D) Synaptic localization of SK3 using the postembedding inmmunogold method. SK3 immunoparticles were detected along the PSD of dendritic spines (s) of granule cells establishing asymmetrical synapses with axon terminals (at), as well as at extrasynaptic sites (arrows) of the spine and spine neck (sn). Scale bars: A–E, 0.2 lm. © 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd European Journal of Neuroscience, 39, 883–892

890 C. Ballesteros-Merino et al.

A

C

B

D

Fig. 6. Localization of SK3 channels along the surface of CA1 pyramidal cells and granule cells using the SDS-FRL technique. Freeze-fracture replicas prepared from mouse hippocampus were labelled with 5-nm (small black dots) immunoparticles to detect SK3 (A and C), or 5-nm (small black dots) and 10-nm (bold black dots) immunoparticles to detect PSD-95 and SK3 (B and D), respectively. SK3 is immunolabelled with an antibody directed against an epitope at the intracellular C-terminus, so the two proteins can be detected at the protoplasmic face (P-face) of the PM. The Exoplasmic face (E-face) is free of any immunolabelling. (A and C) Clusters of immunoparticles for SK3 were detected in dendritic spines of pyramidal cells in the SLM (A) and in dendritic spines of granule cells in the molecular layer of the DG establishing synapses with axon terminals (at). (B and D) In both cell types, clusters of SK3 were observed in close proximity to or intermingled with PSD-95. Scale bar: A–D, 0.2 lm.

density over the nuclei was 0.05  0.01 immunoparticles/lm2. Altogether, the present data showed that the localization of SK3 was not uniform over the dendritic surface of pyramidal cells, but was uniform over the dendritic surface of granule cells.

Discussion In the hippocampus, SK channels regulate neurotransmission, synaptic plasticity and learning (Behnisch & Reymann, 1998; Stackman et al., 2002; Kramar et al., 2004; Ngo-Anh et al., 2005; Lin et al., 2008). Although the mRNAs and proteins for the three SK subunits (SK1, SK2, and SK3) are expressed in the hippocampus, SK3 expression is low compared with SK2 or SK1 (Stocker & Pedarzani, 2000; Sailer et al., 2002; Bond et al., 2004). In the present study, we detected SK3 protein in pyramidal cells and granule cells of the hippocampal formation both during postnatal development and in the adult. This protein expression pattern is consistent with in situ hybridization, showing low SK3 mRNA levels in the CA1 and CA3 regions and moderate levels in granule cells of the DG (Stocker & Pedarzani, 2000; Tacconi et al., 2001), but not as congruous with previous immunohistochemical studies that showed an enrichment in the mossy fibre system (Sailer et al., 2004). The reason for this discrepancy is not clear but might reflect differences in the immunohistochemical techniques and the different SK3 antibodies employed. The results presented here rely on SK3 antibodies, the specificity of

which was validated using membranes or tissue derived from SK3 knockout mouse brain, engendering confidence in the results. An important component for understanding the molecular basis of SK channel function is the determination of the spatial and temporal appearance of the subtypes underlying the electrical activity that regulates hippocampal function. The development of the hippocampal network requires neuronal activity, which is shaped by the differential expression and subcellular sorting of neurotransmitter receptors and ion channels. Parallel to their anatomical and functional maturation, hippocampal neurons undergo distinct development of their complement of neurotransmitter receptor/ion channel profiles. However, data regarding the exact temporal expression of potassium channels are scarce (Petralia et al., 1999; L
opez-Bendito et al., 2002; Brewster et al., 2007). Recently, we reported a thorough description of the developmental expression profile of SK2 in the mouse hippocampus, showing temporal changes in subcellular localization (Ballesteros-Merino et al., 2012). Here, SK3 channels were detected at birth and their expression increased progressively with age during postnatal development up to adult levels. High-resolution immunoelectron microscopy revealed differential trafficking of SK3 to the dendritic compartment of CA1 pyramidal cells as a function of age, changing from predominantly rough endoplasmic reticulum localization in the somata during the first week of development, to PM expression in dendritic shafts and spines at later stages. This developmental distribution pattern is markedly similar to that described for SK2 (Ballesteros-Merino et al., 2012), suggesting that the two SK channel subunits are subjected to similar trafficking regulatory processes during development. In mature principal cells of CA1 and the DG, our data demonstrate that SK3 is predominantly postsynaptic. In addition, the results show two subcellular pools of SK3 in the adult. One pool is clustered along the PSD of excitatory synapses and the other pool is found along the extrasynaptic (dendritic) PM of spines, always close to asymmetrical, presumably excitatory, synapses. Two important determinants for the physiological impact of ion channels in central neurons are location and density on the neuronal surface. As discussed above, high-resolution immunolocalization revealed a differential distance-specific and subcellular compartment-specific distribution pattern for SK3 on the PM of two distinct hippocampal principal cells. Previously, we showed that the density of G protein-gated inwardly rectifying K+ (GIRK) channel subunits GIRK1, GIRK2 and GIRK3 increases as a function of distance along the proximo-distal axis of the CA1 pyramidal cell dendrite, with a progressive increase in the SLM (Fern
andez-Alacid et al., 2011), and similar non-uniform subcellular localization along the neuronal surface has been described for other ion channels (Cooper et al., 1998; Jinno et al., 2005; Kollo et al., 2006; Bourdeau et al., 2007). SK3 now joins this group. In the adult, quantitative comparisons of immunogold densities showed a 51-fold increase for SK3 from somatic to distal apical spine membranes, with the somata having almost no SK3 and proximal spines of similar size in the SR having sevenfold less SK3 than distal spines. These findings show distinct SK3 densities in three different types of synapses in the CA1 region: inhibitory synapses from basket cells, excitatory synapses from CA3 pyramidal cells and excitatory synapses from entorhinal cortex neurons. This is similar to the increase as a function of distance along the proximo-distal axis seen for SK2 expression in the same neuronal population (Ballesteros-Merino et al., 2012). In contrast, for SK3 expression in DG granule cells, quantitative comparison of immunogold densities showed a 38-fold increase from the soma to the dendrites, but SK3 density did not vary along the dendritic surface. Thus, SK3 density

© 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd European Journal of Neuroscience, 39, 883–892

Localization of the SK3 subunit 891 A

B

C

Fig. 7. Change in the density of SK3 in CA1 pyramidal cells and granule cells as a function of distance from the soma. (A) Drawing of the hippocampus showing the regions and cell types chosen for the quantitative analysis. (B) The SLM showed the highest density for SK3 in dendritic spines of pyramidal cells compared with the SR. (C) In the molecular layer of the DG, a similar density of SK3 immunoparticles was detected in the dendritic spines of granule cell commissural/associational inputs and the entorhinal inputs. h, hilus; so, stratum oriens; sp, stratum pyramidale; sr, SR; slm, SLM; I1/3, inner one-third of the molecular layer; O2/3, outer two-thirds of the molecular layer.

is similar for two different inputs in the molecular layer of the DG: the commissural/associational and the perforant pathway from the entorhinal cortex. Although the impact of this is not yet known, the data suggest that the level of SK3 synaptic expression may be regulated at the level of single synapses according to the different presynaptic inputs. Taken together with previous studies of SK2 expression in the hippocampus, the present findings for SK3 strongly suggest that at least some SK channels are heteromeric assemblies of the two subunits. This is supported by the co-immunoprecipitation of SK2 and SK3 from whole brain or following heterologous co-expression (Strassmaier et al., 2005), and by the sensitivity of native SK currents to block by apamin (Stocker et al., 1999; Weatherall et al., 2011). Based upon in situ hybridization for the respective mRNAs and by immunotechniques detecting the proteins (Stocker & Pedarzani, 2000; Sailer et al., 2002, 2004), it is interesting to note that, in CA1 pyramidal neurons, SK3 expression is substantially lower than SK2 expression. Moreover, the apamin-sensitive current in CA1 pyramidal neurons is lost in SK2, but not SK3, knockout mice (Bond et al., 2004). Both SK2 and SK3 are localized to the PSD. These synaptic SK channels in CA1 spines are activated by Ca2+ entry through NMDA receptors modulating synaptic responses to Shaffer collateral stimulations and influencing the induction of synaptic plasticity (Stackman et al., 2002). Upon the induction of long-term potentiation, synaptic SK channel function is lost and this correlates with the Protein kinase A -dependent endocytosis of SK2 (Ngo-Anh et al., 2005; Lin et al., 2008). In this way, activity-dependent trafficking of SK channels contributes to the expression of long-term potentiation. The SK2-L isoform is required for synaptic localization and function, and the unique N-terminal domain of SK2-L contains several regions of homology to the N-terminal domain of SK3 (Allen

et al., 2011). Notably, in the postsynaptic membrane, both SK2 and SK3 immunoparticles are expressed in close anatomical proximity to NMDA receptors. A similar proximity and function are found for SK3 in dopamine neurons of the substantia nigra (Soden et al., 2013). Thus, the present findings suggest that it will be interesting to examine synaptic SK channel function, both for basal neurotransmission and synaptic plasticity, in SK3 knockout mice.

Supporting Information Additional supporting information can be found in the online version of this article: Fig. S1. Absence of SK3 immunolabelling in SK3 / mice.

Acknowledgements The authors would like to thank members of the laboratory of R.L. for their comments on the manuscript. We would also like to thank Mrs Mercedes Gil for the excellent technical assistance. This work was supported by grants from the Spanish Ministry of Education and Science (BFU-2012-38348) and CONSOLIDER (CSD2008-00005) to R.L. All authors declare no conflict of interest.

Abbreviations CA1, CA1 region of the hippocampus; CA3, CA3 region of the hippocampus; DG, dentate gyrus; GIRK, G protein-gated inwardly rectifying K+; NGS, normal goat serum; NMDA, N-methyl-D-aspartate; P, postnatal day; PB, phosphate buffer; PM, plasma membrane; PSD, Postsynaptic density; PSD-95, Postsynaptic density protein 95; SDS-FRL, sodium dodecyl sulphate–freeze-fracture replica labelling; SK (KCNN), small-conductance Ca2+-activated K+; SLM, stratum lacunosum-moleculare; SR, stratum radiatum; TBS, Tris-buffered saline.

© 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd European Journal of Neuroscience, 39, 883–892

892 C. Ballesteros-Merino et al.

References Allen, D., Bond, C.T., Luj
an, R., Ballesteros-Merino, C., Lin, M.T., Wang, K., Klett, N., Watanabe, M., Shigemoto, R., Stackman, R.W. Jr., Maylie, J. & Adelman, J.P. (2011) The SK2-long isoform directs synaptic localization and function of SK2-containing channels. Nat. Neurosci., 14, 744–749. Anderson, P., Morris, R., Amaral, D., Bliss, T. & O’Kefefe, J. (2007) The Hippocampus Book. Oxford University Press Inc, New York. Armstrong, W.E., Rubrum, A., Teruyama, R., Bond, C.T. & Adelman, J.P. (2005) Immunocytochemical localization of small-conductance, calciumdependent potassium channel in astrocytes of the rat supraoptic nucleus. J. Comp. Neurol., 491, 175–185. Ballesteros-Merino, C., Lin, M., Wu, W.W., Ferrandiz-Huertas, C., Caba~ nero, M.J., Watanabe, M., Fukazawa, Y., Shigemoto, R., Maylie, J., Adelman, J.P. & Luj
an, R. (2012) Developmental profile of SK2 channel expression and function in CA1 neurons. Hippocampus, 22, 1467–1480. Behnisch, T. & Reymann, K.G. (1998) Inhibition of apamin-sensitive calcium dependent potassium channels facilitate the induction of long-term potentiation in the CA1 region of rat hippocampus in vitro. Neurosci. Lett., 253, 91–94. Blank, T., Nijholt, I., Kye, M.J., Radulovic, J. & Spiess, J. (2003) Smallconductance, Ca2+-activated K+ channel SK3 generates age-related memory and LTP deficits. Nat. Neurosci., 6, 911–912. Bond, C.T., Herson, P.S., Strassmaier, T., Hammond, R., Stackman, R., Maylie, J. & Adelman, J.P. (2004) Small conductance Ca2+-activated K+ channel knock-out mice reveal the identity of calcium-dependent afterhyperpolarization currents. J. Neurosci., 24, 5301–5306. Bond, C.T., Maylie, J. & Adelman, J.P. (2005) SK channels in excitability, pacemaking and synaptic integration. Curr. Opin. Neurobiol., 15, 305–311. Bourdeau, M.L., Morin, F., Laurent, C.E., Azzi, M. & Lacaille, J.C. (2007) Kv4.3-mediated A-type K+ currents underlie rhythmic activity in hippocampal interneurons. J. Neurosci., 27, 1942–1953. Brewster, A.L., Chen, Y., Bender, R.A., Yeh, A., Shigemoto, R. & Baram, T.Z. (2007) Quantitative analysis and subcellular distribution of mRNA and protein expression of the hyperpolarization-activated cyclic nucleotidegated channels throughout development in rat hippocampus. Cereb. Cortex, 17, 702–712. Cooper, E.C., Milroy, A., Jan, Y.N., Jan, L.Y. & Lowenstein, D.H. (1998) Presynaptic localization of Kv1.4-containing A-type potassium channels near excitatory synapses in the hippocampus. J. Neurosci., 18, 965–974. Cueni, L., Canepari, M., Luj
an, R., Emmenegger, Y., Watanabe, M., Bond, C.T., Franken, P., Adelman, J.P. & L€uthi, A. (2008) T-type Ca2+ channels, SK2 channels and SERCAs gate sleep-related oscillations in thalamic dendrites. Nat. Neurosci., 11, 683–692. Deignan, J., Luj
an, R., Bond, C., Riegel, A., Watanabe, M., Williams, J.T., Maylie, J. & Adelman, J.P. (2012) SK2 and SK3 expression differentially affect firing frequency and precision in dopamine neurons. Neuroscience, 217, 67–76. Fern
andez-Alacid, L., Watanabe, M., Moln
ar, E., Wickman, K. & Luj
an, R. (2011) Developmental regulation of G protein-gated inwardly-rectifying K+ (GIRK/Kir3) channel subunits in the brain. Eur. J. Neurosci., 34, 1724–1736. Fujimoto, K. (1995) Freeze-fracture replica electron microscopy combined with SDS digestion for cytochemical labeling for integral membrane proteins. Application to the immunogold labeling of intercellular junctional complexes. J. Cell. Sci., 108, 3443–3449. Ishii, T.M., Maylie, J. & Adelman, J.P. (1997) Determinants of apamin and d-tubocurarine block in SK potassium channels. J. Biol. Chem., 272, 23195–23200. Jacobsen, J.P., Redrobe, J.P., Hansen, H.H., Petersen, S., Bond, C.T., Adelman, J.P., Mikkelsen, J.D. & Mirza, N.R. (2009) Selective cognitive deficits and reduced hippocampal brain-derived neurotrophic factor mRNA expression in small-conductance calcium-activated K+ channel deficient mice. Neuroscience, 163, 73–81. Jinno, S., Jeromin, A. & Kosaka, T. (2005) Postsynaptic and extrasynaptic localization of Kv4.2 channels in the mouse hippocampal region, with special reference to targeted clustering at gabaergic synapses. Neuroscience, 134, 483–494. Kohler, M., Hirschberg, B., Bond, C.T., Kinzie, J.M., Marrion, N.V., Maylie, J. & Adelman, J.P. (1996) Small-conductance, calcium-activated potassium channels from mammalian brain. Science, 273, 1709–1714. Kollo, M., Holderith, N.B. & Nusser, Z. (2006) Novel subcellular distribution pattern of A-type K+ channels on neuronal surface. J. Neurosci., 26, 2684–2691. Kramar, E.A., Lin, B., Lin, C.Y., Arai, A.C., Gall, C.M. & Lynch, G. (2004) A novel mechanism for the facilitation of theta-induced long-term

potentiation by brain-derived neurotrophic factor. J. Neurosci., 24, 5151– 5161. Lin, M.T., Luj
an, R., Watanabe, M., Adelman, J.P. & Maylie, J. (2008) SK2 channel plasticity contributes to LTP at Schaffer collateral-CA1 synapses. Nat. Neurosci., 11, 170–177. L
opez-Bendito, G., Shigemoto, R., Kulik, A., Paulsen, O., Fair
en, A. & Luj
an, R. (2002) Differential distribution of group I metabotropic glutamate receptors during rat cortical development. Cereb. Cortex, 12, 625– 638. Luj
an, R. & Shigemoto, R. (2006) Localization of metabotropic GABA receptor subunits GABAB1 and GABAB2 relative to synaptic sites in the rat developing cerebellum. Eur. J. Neurosci., 23, 1479–1490. Luj
an, R., Nusser, Z., Roberts, J.D., Shigemoto, R. & Somogyi, P. (1996) Perisynaptic location of metabotropic glutamate receptors mGluR1 and mGluR5 on dendrites and dendritic spines in the rat hippocampus. Eur. J. Neurosci., 8, 1488–1500. Luj
an, R., Maylie, J. & Adelman, J.P. (2009) New sites of action for GIRK and SK channels. Nat. Rev. Neurosci., 10, 475–480. MacDonald, J.F., Belrose, J.C., Xie, Y.F. & Jackson, M.F. (2013) Nonselective cation channels and links to hippocampal ischemia, aging, and dementia. Adv. Exp. Med. Biol., 961, 433–447. Ngo-Anh, T.J., Bloodgood, B.L., Lin, M., Sabatini, B.L., Maylie, J. & Adelman, J.P. (2005) SK channels and NMDA receptors form a Ca2+-mediated feedback loop in dendritic spines. Nat. Neurosci., 8, 642–649. Petralia, R.S., Esteban, J.A., Wang, Y.X., Partridge, J.G., Zhao, H.M., Wenthold, R.J. & Malinow, R. (1999) Selective acquisition of AMPA receptors over postnatal development suggests a molecular basis for silent synapses. Nat. Neurosci., 2, 31–36. Poolos, N.P. & Johnston, D. (2012) Dendritic ion channelopathy in acquired epilepsy. Epilepsia, 53(Suppl. 9), 32–40. Sailer, C.A., Hu, H., Kaufmann, W.A., Trieb, M., Schwarzer, C., Storm, J.F. & Knaus, H.G. (2002) Regional differences in distribution and functional expression of small-conductance Ca2+-activated K+ channels in rat brain. J. Neurosci., 22, 9698–9707. Sailer, C.A., Kaufmann, W.A., Marksteiner, J. & Knaus, H.G. (2004) Comparative immunohistochemical distribution of three small-conductance Ca2+-activated potassium channel subunits, SK1, SK2, and SK3 in mouse brain. Mol. Cell. Neurosci., 26, 458–469. Sarpal, D., Koenig, J.I., Adelman, J.P., Brady, D., Prendeville, L.C. & Shepard, P.D. (2004) Regional distribution of SK3 mRNA-containing neurons in the adult and adolescent rat ventral midbrain and their relationship to dopamine-containing cells. Synapse, 53, 104–113. Schilling, K. & Aletsee-Ufrecht, M.C. (1998) An immunoblot assay for the simultaneous quantification of several antigens. Anal. Biochem., 177, 203– 206. Soden, M.E., Jones, G.L., Sanford, C.A., Chung, A.S., G€ uler, A.D., Chavkin, C., Luj
an, R. & Zweifel, L.S. (2013) Disruption of dopamine neuron activity pattern regulation through selective expression of a human KCNN3 mutation. Neuron, 80, 997–1009. Stackman, R.W., Hammond, R.S., Linardatos, E., Gerlach, A., Maylie, J., Adelman, J.P. & Tzounopoulos, T. (2002) Small conductance Ca2+-activated K+ channels modulate synaptic plasticity and memory encoding. J. Neurosci., 22, 10163–10171. Stocker, M. & Pedarzani, P. (2000) Differential distribution of three Ca(2+)activated K(+) channel subunits, SK1, SK2, and SK3, in the adult rat central nervous system. Mol. Cell. Neurosci., 15, 476–493. Stocker, M., Krause, M. & Pedarzani, P. (1999) An apamin-sensitive Ca2+activated K+ current in hippocampal pyramidal neurons. Proc. Natl. Acad. Sci. USA, 96, 4662–4667. Strassmaier, T., Bond, C.T., Sailer, C.A., Knaus, H.G., Maylie, J. & Adelman, J.P. (2005) A novel isoform of SK2 assembles with other SK subunits in mouse brain. J. Biol. Chem., 280, 21231–21236. Tacconi, S., Carletti, R., Bunnemann, B., Plumpton, C., Merlo, P.E. & Terstappen, G.C. (2001) Distribution of the messenger RNA for the small conductance calcium-activated potassium channel SK3 in the adult rat brain and correlation with immunoreactivity. Neuroscience, 102, 209–215. Tarusawa, E., Matsui, K., Budisantoso, T., Moln
ar, E., Watanabe, M., Matsui, M., Fukazawa, Y. & Shigemoto, R. (2009) Input-specific intrasynaptic arrangements of ionotropic glutamate receptors and their impact on postsynaptic responses. J. Neurosci., 29, 12896–12908. Weatherall, K.L., Seutin, V., Li
egeois, J.F. & Marrion, N.V. (2011) Crucial role of a shared extracellular loop in apamin sensitivity and maintenance of pore shape of small-conductance calcium-activated potassium (SK) channels. Proc. Natl. Acad. Sci. USA, 108, 18494–18499.

© 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd European Journal of Neuroscience, 39, 883–892