Positive modulation of SK channel impedes neuron-specific cytoskeletal organization and maturation

Abstract

N-Methyl-D-Aspartate receptor (NMDAR) modulates the structural plasticity of dendritic spines by impacting cytoskeletal organization and kinase signaling. In the developing nervous system, activation of NMDAR is pertinent for neuronal migration, neurite differentiation, and cellular organization. Given that small conductance potassium channels (SK2/3) repress NMDAR ionotropic signaling, the current study highlights the impact of neonatal SK channel potentiation on adult cortical and hippocampal organization. In the current study, neonatal SK channel potentiation was performed by a single time point injection of SK2/3 agonist (CyPPA) into the pallium of mice at postnatal day 2 (P2). When the animals reached adulthood (P55), the hippocampus and cortex were examined to assess neuronal maturation, lamination, and the distribution of synaptic cytoskeletal proteins. Immunodetection of neuronal markers in the brain of P0 treated P55 mice revealed the presence of immature neurons in the upper cortical layers (layer II-IV) and CA1 (hippocampus). Also, layer-dependent cortical cell density was attenuated as a result of the ectopic localization of mature (NeuN⁺) and immature (Doublecortin+) neurons in layer II-IV. Similarly, the decreased count of adult neurons (NeuN⁺) in the CA1 is accompanied by an increase in the number of immature (Doublecortin⁺) neurons. Ectopic localization of neurons in the upper cortex and CA1 caused the dramatic expression of neuron-specific cytoskeletal proteins. In the same vein, structural deformity of neuronal projections, and loss of post-synaptic densities suggests that post-synaptic integrity is compromised in the SK2/3⁺ brain. From these results, we deduced that SK channel activity in the developing brain likely impacts neuronal maturation through its effects on cytoskeletal formation.

INTRODUCTION

Transient Ca⁺⁺ current generated by N-methyl-D-Aspartate receptor (NMDAR)-GluN1 potentiation constitutes approximately 75% of post-synaptic Ca⁺⁺ that modulates synaptic plasticity ^[1–4]. By activating several synaptic downstream kinases, synaptic potentiation events are further coupled to cellular regulation and gene expression^[5–11]. This process represents an ordered cue of signaling events that determine protein recruitment to synapses, cell death, synaptic potentiation or inhibition, and synaptic cytoskeletal organization ^{[9, 10, 12–15]}. In the developing nervous system, potentiation of NMDAR directs neuronal migration and cellular organization. Genetic ablation or pharmacological inhibition of NMDAR leads to erroneous neural migration patterns that alter neural organization^{[14, 16–20]}. Loss of NMDAR function during the critical periods of brain development results in perturbations of neural circuits and region-dependent cell density ^{[17, 21–23]}. Substantive evidence also exists to suggest that the ablation of NMDAR function in newly formed synapses can lead to defects in synaptic pruning, delayed maturation, and aberrant distribution of synapses ^[24–26]. Loss of NMDAR function and aberrant neuronal migration patterns are implicated in the etiology and progression of developmental neuropsychiatric disorders like autism ^{[21, 24, 25, 27–33]}.

The regulation of long-term synaptic potentiation can offer significant insights into the concept of NMDAR hypofunction in developmental cognitive defects. NMDAR-mediated synaptic potentiation is dependent on the activation of kinases that facilitate high-frequency neuronal firing and re-arrangement of the synaptic cytoskeleton. Accordingly, the synaptic potentiation rate and structure of the dendritic spine cytoskeleton underlie structural and functional plasticity of fully developed synapses ^{[2, 3, 31, 34–39]}.

SK2 is a small conductance channel that is localized with the NMDAR at postsynaptic densities (PSDs) ^[40–49]. Because of the proximity to NMDAR, glutamatergic ionotropic neurotransmission produces most of the post-synaptic Ca⁺⁺ transients that activate SK ^{[44, 46]}. SK channel is a homomeric tetramer that assembles pore-forming subunits when Ca++ binds to the Calmodulin (CaM) on the CaM lobe ^{[50, 51]}. Activation of SK2 represents an ion channel gating mechanism that promotes K⁺ after-hyperpolarization currents which abrogate synaptic potentiation ^{[42, 46–48, 52–54]}. As a result, SK2 regulates “excitability” by enhancing “synaptic adaptation”. Mechanistically, NMDAR-mediated dendritic spine plasticity is tuned by repetitive activation of SK by transient Ca⁺⁺ currents. This leads to the suppression of synaptic plasticity ^{[41–43, 50, 51, 55–59]}.

Activation of NMDAR in developing in the developing brain is pertinent to cytoskeletal differentiation and maturation ^[60–64]. Similar to plasticity in the adult brain, NMDAR signaling in the developing brain directs kinase activity that facilitates cytoskeletal assembly for dendrite branching, axon guidance, and neuronal migration ^[65–68]. Given that SK refines the activity of NMDAR in dendritic spines ^{[49, 56]}, there is a need to investigate its (i.e. SK2) role in the developmental organization of pallial neurons destined for the cortex and hippocampus.

Our results revealed that neonatal positive modulation of SK channel impaired maturation and cytoskeletal organization of pallial neurons in the adult cortex and hippocampus. Specifically, neonatal positive modulation of SK channel caused the ectopic localization of adult neurons and overexpression of cytoskeletal proteins in cortical layers II-IV and hippocampal CA1-subiculum area. In addition to the aberrant distribution of neuron-specific cytoskeletal proteins, some of the ectopic neurons exhibit delayed maturation.

METHODS

Animals

Adult male and female C57BL/6 mice were housed as mating pairs. Animals were housed under standard laboratory conditions of 12 hours alternating light and dark cycle with food and water provided ad libitum. All animal handling procedures were approved by the Institutional Animal Care and Use Committee of the Louisiana State University School of Veterinary Medicine.

Intracranial CyPPA injection (P2)

In a separate set of liters, CyPPA (Tocris #2953) or saline was injected into the pallium. Pups were sedated with isoflurane, with the head and body held in position by a rigid foam. A hydraulic ultra-fine micromanipulator (Narishige) was used to determine stereotaxic coordinates relative to the bregma. The coordinate for the medial pallium (AP: −0.5mm, ML: 0.2 mm, DV: 1.0 mm) were determined using the Allen Developing Mouse Brain Atlas. The micromanipulator probe holder, carrying a 32-gauge Hamilton syringe, was gently lowered to penetrate the skin and cranium. CyPPA was injected gradually at the rate of 2μL/min. The injector was left in place for 5 minutes after the drug had been delivered. The treated animals (SK2/3⁺) were returned to the maternal cage and monitored thereafter. Mice were weaned on P28 and housed based on sex.

Western blotting

Frozen hippocampal tissue was incubated on ice with RIPA lysis buffer containing protease and phosphatases inhibitor cocktail. After 30 minutes, the incubated tissue was rapidly homogenized to obtain whole tissue lysate. 10μg of protein was processed for SDS-PAGE electrophoresis. After western blotting (wet transfer), Polyvinylidene fluoride membrane (PVDF) was incubated in Tris-buffered saline (with 0.01% Tween 20) for 15 minutes (TBST) at room temperature. Afterward, the membrane was blocked in 3% bovine serum albumin (prepared in TBST) for 50 minutes at room temperature. The level of cytoskeletal proteins in the hippocampus was determined using Rabbit anti-Doublecortin (DCX) antibody (Cell Signaling #14802S), Rabbit anti -Neurofilament antibody (Cell Signaling #2837S), Rabbit anti-β-Tubulin (III) (Cell Signaling #5568S), and Rabbit anti-PSD-95 (Cell Signaling #3450) antibody at a dilution of 1:1,000. The membrane was incubated in the primary antibody solution overnight at 4oC. The primary antibodies were detected using Chicken anti-Rabbit-HRP (Thermofisher Scientific #A15987) a dilution of 1:5,000. The reaction was developed using a chemiluminescence substrate (Thermofisher-#34579). To normalize protein expression, the membranes were treated with Restore PLUS Western Blot Stripping Buffer (Thermofisher Scientific #46430), and re-probed (overnight at 4oC) with HRP-Conjugated GAPDH Polyclonal Antibody (Thermofisher Scientific # PA1–987-HRP) prepared in blocking solution. Protein expression was normalized per lane using the corresponding GAPDH expression.

Immunofluorescence

Adult mice were transcardially perfused with 10mM PBS, then 4% phosphate-buffered paraformaldehyde (4% PB-PFA). The whole brain was removed and fixed in 4% PB-PFA overnight. Subsequently, the brain was transferred into 4% PB-PFA containing 30% sucrose for cryopreservation (36–72 hours). Free-floating cryostat sections (40μm) were obtained and preserved in 48-well plates containing 10mM PBS at 4°C. The sections were washed three times (5 minutes each) in 10mM PBS (pH 7.4) on a slow orbital shaker. Blocking of non-specific protein interaction was performed in 5% Normal Goat Serum (Vector Labs #S-1000), prepared in 10mM PBS + 0.03% Triton-X100, for 1 hour at room temperature. The sections were incubated overnight at 40C in the following primary antibodies; Rabbit anti-NeuN Alexa-488 Conjugate (EMD Millipore #MAB377XM), Rabbit anti-Doublecortin (Dcx) antibody (Cell Signaling #14802S ), Rabbit anti-Neurofilament antibody (Cell Signaling #2837S), and Rabbit anti-β-Tubulin (Type III) antibody (Cell Signaling #5568S). The primary antibodies were diluted in blocking solution (10mM PBS+0.03% Triton-X 100 and 5% Normal Goat serum). After primary antibody incubation, the sections were washed two times in 10mM PBS, then incubated in a secondary antibody - Goat anti-Rabbit Alexa 568 (Thermofisher Scientific #A-11036) or Goat anti-Mouse Alexa 488 (Cell Signaling #4408S) - diluted in the blocking solution. Secondary antibody incubation was done for 1 hour at room temperature, with gentle shaking (35rpm). Immunolabeled sections were washed and mounted on gelatin-coated slides using ProLong™ Diamond Antifade Mountant containing DAPI (Thermofisher Scientific #P36971).

Expansion microscopy

Cryostat sectioned 40μm brain slice was incubated in Rabbit anti Neurofilament primary antibody (Cell Signaling # 2837) overnight at 4°C. After washing in the blocking buffer, the section was incubated in Goat anti-Rabbit Alexa 568 (Thermo Fisher Scientific Cat# A-11036, RRID: AB_10563566) secondary antibody diluted in 10mM PBS+0.03% Triton-X 100 and 5% Normal Goat serum. Anchoring treatment was done overnight in 0.1mg/ml Acryloyl-X. Gelation, overnight digestion, and mounting were done using previously described methods ^{[69, 70]}.

Quantification

Fluorescence imaging was performed using a Nikon-NiU fluorescence upright microscope configured for 3D imaging. Z-stacks were obtained and converted into 2D images through the “extended depth of focus (EDF)” option in the Nikon Element Advanced Research software. Normalized fluorescence intensity for immunolabeled proteins in the hippocampus and cortex were determined in optical slices for serial section images (n=4 per group). Fluorescence intensity and cell count were determined using Image J software. Grids were positioned on images using the Grid2 plugin in Image J. We determined the mean fluorescence intensity per unit area using the distribution of the grids on anatomically defined cortical layers and regions of the hippocampus. The mean intensity in each grid was normalized as a percentage of the total mean intensities for all grid. Similarly, cell counting was performed to determine the distribution of cells per unit area (grid). In subsequent analysis, each grid was assigned to an anatomical layer (cortex) or region (hippocampus).

Statistical analysis

Fluorescence intensity, cell count, and western blot protein level for control and SK2/3⁺ group were compared by T-test analysis. Upper and lower cortical cell count or fluorescence intensity was also determined by T-test analysis across groups (paired and unpaired). Statistical comparison of multiple cortical layers or hippocampal region was performed in One-Way ANOVA. All statistical analysis was carried out in GraphPad Prism version 8.0. Here, we presented the results as point graphs with error bars depicting the mean and standard error of mean (SEM) respectively.

RESULTS

Cortical and hippocampal distribution of adult neurons

In the P55 brain, the distribution of adult neurons in the cortex and hippocampal CA1 region were determined by NeuN+ immunostaining (Fig. 1a). Our results showed that neonatal SK channel potentiation led to an increase in adult neuron cell count in the cortex. NeuN⁺ cells increased significantly in SK2/3⁺ cortical layer II/III when compared with the control (p=0.005; Fig. 1b). Similarly, SK2/3⁺ cortical layer IV recorded an increase in NeuN+ cell count when compared with the control (p<0.0001; Fig. 1c ). No significant difference was observed for SK2/3⁺ cortical layer V-VI NeuN⁺ cell count when compared with the and control (p=0.3902; Fig. 1d). Sub sequent comparison of layer dependent NeuN+ cell count showed that layer II/III cell count is higher than layer IV in the control cortex (p<0.0001; Fig. 1e). The same is the case for the SK2/3⁺ cortex, but the level of significance was lower (p=0.0002; Fig. 1f). In support of this outcome, a comparison of NeuN+ cell count for control cortical layers IV and V-VI showed no significant difference (p=0.10; Fig. 1e). However, in the SK2/3⁺ cortex, layer IV NeuN⁺ cell count was significantly lower when compared with layer V-VI (p=0.0033; Fig. 1f).

Figure 1: Mature cell count in the cortex and hippocampus of control and SK2/3⁺ mice (P55).

(A). Low magnification fluorescence images demonstrating the distribution of adult neurons (NeuN+) in the cortex of control and SK2/3⁺ mice (scale bar=50μm, 25μm).

(B). Graph showing increased NeuN⁺ cell count in layer II/III of the SK2/3⁺ cortex (**p=0.005).

(C). Graph showing increased NeuN⁺ cell count in layer IV of the SK2/3⁺ cortex (****p<0.0001).

(D). Graph illustrating the comparative count of NeuN⁺ cell in cortical layer V-VI (ns; p=0.3902).

(E). Graphical representation of NeuN⁺ cell distribution for the layers of the control (****p<0.0001) and

(F). SK2/3⁺ cortex (***p=0.0002, **p=0.0033).

Graphs demonstrating NeuN⁺ cell count for border region between layer II/III and layer IV.

(G). Control: layer II/III count is higher than layer IV (****p<0.0001).

(H). SK2/3⁺: layer IV count is higher than layer II/III (**p=0.0014).

(I-J). Low (scale bar=100μm) and high magnification Low (scale bar=25μm) fluorescence images demonstrating the distribution of NeuN⁺ cells in the hippocampus (so: stratum oriens; pyr: pyramidal cell layer; rad: stratum radiatum).

(K). Graph demonstrating a reduction in CA1 NeuN⁺ cell count in the SK2/3⁺ hippocampus (p=0.0036).

Together, these results demonstrate an increase in upper cortical (layer II-IV) NeuN⁺ count for the SK2/3⁺ cortex and are supported by the loss of layer-dependent cell density (yellow arrows; Fig. 1a). To detect ectopic distribution in cortical layer IV, we analyzed high magnification images showing the lower region of layer II/III and the upper part of layer IV (Fig. 1a; lower panel). As expected, the control recorded significantly higher cell count for layer II/III when compared to layer IV (Fig. 1g; p<0.0001). Conversely, the NeuN⁺ cell count was significantly higher in SK2/3⁺ cortical layer IV compared with layer II/III (p=0.0014;Fig. 1h ). Analysis of NeuN⁺ cell distribution in the hippocampus revealed a change in cell count in parts of the hippocampus (Fig. 1i). Notably, there was a significant decrease in CA1 NeuN⁺ count in the SK2/3⁺ hippocampus (Fig. 1j–k; p=0.0036) when compared with the control.

Doublecortin (DCX) and delayed maturation

Immunofluorescence detection of DCX demonstrates the ectopic localization of immature neurons in the SK2/3⁺ upper cortical layers at P55 (Fig. 2a). Clusters of DCX⁺ neurons were found in the SK2/3⁺ cortex and not the control (yellow arrowhead). Co-localization of DCX and adult neuron markers (NeuN) suggests that immature neurons were ectopic in the upper layers (II-IV) of the SK2/3⁺ cortex, and absent in the control cortex (Fig. 2b; p<0.0001).

Figure 2: Distribution of DCX⁺ cells in the cortex and hippocampus of adult mice (P55) treated in early neonatal development (P0-P3).

(A). Representative fluorescence images demonstrating the expression of DCX in the cortex (scale bar=40μm, 10μm).

(B). Graph demonstrating higher DCX+ cell count in the SK2/3⁺ cortical layer II/III (****p<0.0001).

(C). Western blot detection of DCX level in whole hippocampal lysate. This demonstrates the DCX loss in the SK2/3⁺ hippocampus (see also Fig. 3e).

(D). Graph showing a lower DCX level in the SK2/3⁺ hippocampus (*p=0.0147).

(E). Representative fluorescence images demonstrating the expression of DCX in the DG (scale bar=20μm).

(F). Higher magnification images of DCX expressing neurons (yellow arrow heads) in the DG.

(G). Graph showing a decrease in DCX+ cell count per unit area of the SK2/3⁺ DG (****p<0.0001).

(H). Immunofluorescence labeling of DCX in the CA1 (so: stratum oriens; pyr: pyramidal cell layer; rad: stratum radiatum; scale bar=20μm).

(I). Graph demonstrating an increase in DCX⁺ cell count in the CA1 of SK2/3⁺ hippocampus (**p=0.01).

Immunoblot analysis of DCX (Fig. 2c) revealed a decrease in hippocampal protein expression for the SK2/3⁺ group (Fig. 2c–d; p=0.01). This is further evident by a decrease in DCX immunofluorescence and cell count in the DG of SK2/3⁺ mice (Fig. 2e–f). Compared with the control, the SK2/3⁺ group recorded a decrease in DCX⁺ cell count in the DG (Fig. 2g; p<0.0001). Interestingly, there is a significant increase in the count of DCX⁺ neurons in the CA1 of SK2/3⁺ mice (Fig. 2h). This result demonstrates ectopic localization of immature neurons (Fig. 2i; p=0.01) and supports the observed decrease in mature (NeuN⁺) neuron cell count in the CA1.

Neonatal SK channel potentiation abrogates cytoskeletal maturation

Neurofilament

Layer dependent distribution of neurofilament in the cortex further supports the results for SK2/3⁺-induced aberrant neuron distribution and cytoskeletal maturation (Fig. 3a). The expression of neurofilament is higher in the control upper cortical layers (II-IV) when compared with the lower cortex (V-VI) (Fig. 3b; p=0.045). While this is the case in the SK2/3⁺ brain, the upper cortex exhibited overexpression of neurofilament and may indicate abnormal neural projections in this brain area. To this effect, a comparison of layer II-IV with layer V-VI neurofilament expression in the SK2/3⁺ cortex produced a higher level of significance (Fig. 3c; p<0.0001). Similarly, normalized fluorescence intensity for neurofilament in the SK2/3⁺ upper cortex was empirically higher than the control (Fig. 3d; p=0.27). Comparatively, in the lower cortical (layer V-VI) region, fluorescence detection decreased for the SK2/3⁺ brain (Fig. 3e; p=0.026).

Figure 3: Aberrant expression of neurofilament (NF) in the cortex and hippocampus of SK2/3⁺ mice.

(A). Representative fluorescence images demonstrating NF expression in control and SK2/3⁺ cortical layers (scale bar=100μm, 10μm).

(B-C). Graphical illustration of NF expression across cortical layers. Layer II-IV expression is higher than layer V-VI for both the control (*p=0.045) and SK2/3+ cortex (****p<0.0001).

(D). Percentage fluorescence intensity per unit area (normalized) is not significantly different for control SK2/3⁺ layer II-IV (p=0.27).

(E). Graph showing a decreased normalized fluorescence intensity in the SK2/3⁺ layer V-VI (*p=0.026).

(F). Fluorescence images showing the relative distribution of neurofilament in control SK2/3+ layer hippocampus (CA1, CA3, DG, and sub; scale bar=100μm)

(G-H). Graphs showing normalized neurofilament expression in the CA1, CA3, and DG (****p<0.0001; of control and SK2/3⁺ mice (**p=0.047; ***p=0.0005).

(I-J). Western blot and graph demonstrating a significant loss of neurofilament in SK2/3⁺ whole hippocampal lysate (p=0.0035).

(K). Representative high magnification image demonstrating loss of neurofilament in the SK2/3⁺ hippocampal CA1 region (scale bar=20μm).

(L). Graph illustrating decreased CA1 neurofilament for SK2/3⁺ group (**p=0.0016).

(M). Graph illustrating increased neurofilament expression in the SK2/3⁺ subiculum (**p=0.0041).

(N-O). Graphs illustrating significant loss of neurofilament in the CA3 (*p=0.027) and DG (*p=0.017) of SK2/3⁺ hippocampus.

Aberrant distribution of neurofilament was also recorded in the SK2/3⁺ hippocampus (Fig. 3f). There was no significant difference in neurofilament expression for control CA1, CA3, DG, and sub (Fig. 3g). In contrast, the SK2/3⁺ subiculum (Fig. 3h) recorded significantly higher neurofilament level when compared with the CA1 (p<0.0001), CA3 (p=0.047), and DG (p=0.0005). In support of this outcome, immunoblot analysis of whole hippocampal lysate revealed a significant loss of neurofilament in the SK2/3⁺ hippocampus (Fig. 3i–j; p=0.0035). In comparison with the control, the SK2/3⁺ hippocampus showed a prominent loss of CA1 neurofilament (Fig. 3k–l; p=0.0016). Conversely, the subiculum was characterized by an overexpression of neurofilament (Fig. 3m; p=0.0041). In addition to the CA1, Neurofilament expression decreased significantly in the CA3 (Fig. 3n; p=0.027) and DG (Fig. 3o; p=0.017) regions of the SK2/3⁺ hippocampus.

Type III β-Tubulin

Dramatic distribution of Type III β-Tubulin in the SK2/3⁺ cortex supports the presence of aberrant neuronal projections (Fig. 4a). The SK2/3⁺ upper cortex (layer II-IV) recorded an overexpression of Type III β-Tubulin because of heavily labeled layer IV neuronal cell bodies, and their projections to layer II/III (Fig. 4b; white arrowheads). In the control cortex, expression of Type III β-Tubulin in layer II-IV is higher when compared with layer V-VI (p=0.0268; Fig. 4c). Overexpression of Type III β-Tubulin in the SK2/3⁺ upper cortex is evident as a higher level of significance when layer II-IV expression is compared with layer V-VI (p<0.0001; Fig. 4d). As illustrated in Fig. 4a–b, normalized fluorescence expression of Type III β-Tubulin is significantly higher for the SK2/3⁺ layer II-IV versus the control (p<0.0001; Fig. 4e). Compared with the control, expression of Type III β-Tubulin did not significantly change for SK2/3⁺ cortical layer V-VI (Fig. 4f; p=0.017).

Figure 4: Distribution of Type III β-Tubulin in the cortex and hippocampus.

(A). Fluorescence images showing the expression of β-Tubulin (III) in the cortex (scale bar=50μm).

(B). Representative fluorescence images demonstrating layer IV projections to layer II/III (scale bar=20μm).

(C-D). Graphs illustrating higher β-Tubulin (III) expression in cortical layer II/III versus layer IV in control (*p=0.0268) and SK2/3+ cortex (****p<0.0001).

(E). Graph illustrating increased β-Tubulin (III) expression in SK2/3+ cortical layer II-IV (****p<0.0001).

(F). Graph comparing β-Tubulin (III) expression in control and SK2/3+ cortical layer V-VI (p=0.1675).

(G). Graph illustrating the expression of β-Tubulin (III) in control CA1, CA3, DG, and subiculum (CA3 vs CA1, *p=0.021; CA3 vs DG, *p=0.017).

(H). Graph showing overexpression of β-Tubulin (III) in the SK2/3 subiculum (****p<0.0001).

(I-J). Western blot and graph demonstrating a decrease in hippocampal β-Tubulin (III) expression for the SK2/3⁺ group (*p=0.0434).

(K). Low magnification images showing β-Tubulin (III) expression in the hippocampus (scale bar=100μm).

(L-N). Graph demonstrating a decreased β-Tubulin (III) expression in the SK2/3⁺ CA1 (****p<0.0001), CA3 (**p=0.0016), and DG (**p=0.0021).

(O). Graph demonstrating overexpression of β-Tubulin (III) in the SK2/3⁺ subiculum (**p=0.0064).

Regional expression of Type III β-Tubulin was also altered in the hippocampus of mice following neonatal SK2/3⁺ potentiation. In control mice, the CA3 and subiculum areas showed higher Type III β-Tubulin expression. T he CA1 (p=0.0259) and DG (p=0.0146; Fig. 4g) recorded lower Type III β-Tubulin level compared with the control CA3. Conversely, SK2/3⁺ hippocampus was characterized by an overexpression of Type III β-Tubulin in the subiculum (p<0.0001), and a decrease in the CA1, CA3, and DG (Fig. 4h). The reduction of total Type III β-Tubulin level in the SK2/3⁺ hippocampus is further evident in immunoblot detection of the protein in tissue lysates (Fig. 4i). A comparison of normalized Type III β-Tubulin level in SK2/3⁺ and control whole hippocampal lysate revealed a significant decrease for the SK2/3⁺ group (Fig. 4j; p=0.043). In support of this outcome, the comparison of normalized Type III ß-Tubulin fluorescence for the control and SK2/3⁺ hippocampus (Fig. 4k–o) demonstrate prominent loss in the SK2/3⁺ CA1 (p<0.0001), CA3 (p=0.0016), DG (p=0.0021), and overexpression in the subiculum (p=0.0064).

Structural abnormalities in the hippocampus and cortex

Ectopic localization of neurons in cortical layer II-IV (Fig. 1a and Fig. 1c) was accompanied by an increase in neuron-specific cytoskeletal protein distribution in the upper cortex and CA1-subiculum area (Fig. 3 and Fig. 4). To examine the structure of neuronal processes (dendrites), we performed tissue expansion on brain slices following neurofilament immunolabeling. At high magnification, our results showed normal neurofilament distribution in labeled neuronal processes in cortical layer II-IV (Fig. 5a) and V-VI (Fig. 5b). In SK2/3⁺ brain, there were aggregations of neurofilament (white arrowheads) which supports a possibility for cytoskeletal defects in the ectopic neurons. Similarly, in the hippocampus (Fig. 5c), aggregates of neurofilament were also observed in the pyramidal cell layer of the CA1. In support of this result, immunoblot analysis of whole hippocampal lysate revealed a significant loss of post-synaptic protein marker (PSD-95) in the SK2/3⁺ hippocampus (Fig. 5d–e) when compared with the control (p=0.0016).

Figure 5: Disruption of neurofilament and PSD-95 in SK2/3+ brain.

(A-C). Fluorescence images (ExM) illustrating fragmentation of neurofilament in the SK2/3⁺ cortex and hippocampus (scale bar=20μm, 2μm).

(D-E). Western blot and graph demonstrating a reduction in PSD-95 in the SK2/3⁺ hippocampus (**p=0.0016).

Discussion

In the current study, we showed that neonatal SK channel potentiation significantly impaired cortical and hippocampal organization and cytoskeletal maturation. Notable changes in adult neuron distribution were associated with the loss of layer-dependent cell density in the cortex. To this effect, an increased layer IV neuron count was recorded at P55 for mice treated during early neonatal development (P2). In support of this outcome, the distribution of neuron-specific cytoskeletal proteins increased significantly in the upper cortical layers II-IV, and parts of the hippocampus (CA1-subiculum) where the ectopic cells are located. Interestingly, some of the ectopic neurons show distinct characteristics of immature neurons with the absence of identifiable dendritic projections and DCX expression. Ultimately, there was evidence of post-synaptic cytoskeletal (neurofilament/PSD-95) damage in the cortex and hippocampus of adult mice following a neonatal SK channel potentiation.

Modification to synaptic structure and composition is characteristic of developmental neuropsychiatric disorders. In the etiology and progression of developmental cognitive defects, notable alterations in neural circuit organization and dendrite morphology have been identified. Notably, in autism, erroneous regulation of neuronal migration leads to mistargeting of intracortical and subcortical neural projections ^{[27, 71–75]}. Furthermore, genetic mutations linked with autism have shown to cause significant perturbations in spine morphology ^{[76, 77]}.

Regulated expression and activity of NMDAR are essential for the development and organization of the nervous system. NMDAR signaling directs neuronal migration, synaptogenesis, synaptic pruning, and maturation ^{[26, 30, 60, 61, 63, 64, 68, 78]}. For this, ionotropic and kinase signaling of NMDAR is pertinent for neurite formation and reorganization of the cytoskeleton during neuronal migration. As such, the ablation of NMDAR function in the developing nervous system causes an erroneous neural migration pattern and abolishes layer-dependent neuron density in the cortex ^[63,68]. Although SK channel potentiation attenuates NMDAR activity, how SK channel potentiation impacts neuronal development and cytoskeletal organization is yet to be investigated.

SK channels are calcium-activated potassium channels that are co-localized with NMDAR at postsynaptic densities. Although it is found in every part of the brain, the expression of SK channels is relatively high in the hippocampal and cortical dendrites ^{[41–44, 79, 80]}. Previous studies have established the role of SK in synaptic plasticity and memory encoding ^{[43, 47]}. Accordingly, increased potentiation of SK channels ablates postsynaptic calcium transient towards inhibiting synaptic plasticity. Consistent with its effect on calcium signaling, persistent activation of SK channels depresses synaptic function and learning in mice ^{[42, 81]}. Given that NMDAR signaling is required for structural plasticity of dendritic spines, and SK modulates NMDAR Ca++ transient, it is likely that neonatal SK activity directs the aspects of neuronal maturation and cytoskeletal differentiation.

In normal brain function, modulation of NMDAR alters the structure of the spine cytoskeleton in an activity-dependent manner. This process underlies memory encoding and long-term plasticity ^{[38, 82–84]}. To this effect, change in the shape and size of dendritic spines is directly related to the level of activity of the synapses. Dendritic spine dysgenesis has been identified in neuropsychiatric and neurodegenerative disorders whichinclude autism, schizophrenia, Rhett’s syndrome, Alzheimer’s disease, Down syndromes, among several others ^{[76, 77]}. Aberrations in dendritic spine morphology have been strongly linked to dysregulation of NMDAR signaling at glutamatergic synapses that are localized on these spines ^{[17, 83, 85–88]}. In developing neurons, the organization of various proteins that constitute neuronal cytoskeleton leads to neurite budding and differentiation. Disruption of cytoskeletal proteins during development significantly impacts neuronal migration and differentiation. As neurons mature, the transformation of cytoskeletal composition occurs and can be used to distinguish proliferating from fully formed neurons ^[89].

Summary

Our results showed that neonatal SK channel potentiation significantly impacts neuronal cytoskeletal organization and maturation. Notably, the cortex and hippocampus of neonatal CyPPA treated mice exhibits aberrant distribution of adult neurons and neuron-specific cytoskeletal proteins.