Chronic activation of the small-conductance, calcium-activated potassium channel precipitates age-dependent depressive-like behavior and cognitive deficits and reduces klotho concentration.

Abstract

Depression is a common mood disorder with a multifaceted, complex pathophysiology. An abnormal level of corticosteroids in the brain can cause structural change in the hippocampus. It may cause an imbalance in calcium homeostasis leading to functional impairment in the activity of the voltage-insensitive, small conductance, calcium-activated potassium channel (SKC). Deficit in the longevity protein klotho is also implicated in stress-induced depression. We ascertained whether chronic activation of SKCs can cause age-dependent depressive-like behavior and cognitive deficits, accompanied by disturbances in klotho concentration. We tested the effect of repeated activation of SKCs by the potent SKC agonist, 1-EBIO (1.0 mg/kg, IP, once daily for 15 days) in young (3 months) and mid-life (12 months) male BALB/c mice. We conducted a battery of behavioral tests to investigate depressive-like behavior and cognitive functions. Then, we used an enzyme-linked immunosorbent assay (ELISA) on brain homogenates to determine the change in total klotho concentration in the prefrontal cortex, hippocampus, and dorsal raphe. We found that chronic 1-EBIO treatment decreased locomotor activity, sucrose preference, and alternation index in an age-dependent manner. The drug does not affect stress-coping behavior in the forced swim test. The behavioral deficits were accompanied by a significant decrease in total klotho concentration in the hippocampus but not in the prefrontal cortex or dorsal raphe, observed in an age-dependent manner. Based on these results, we surmise that chronic activation of SKCs results in concurrent cognitive and depressive-like phenotypes in mid-aged mice. Therefore, this could represent a putative therapeutic target for age-related psycho-affective disorders.

Highlights

• •SKCs and klotho protein may have an undiscovered causal relationship in the course of depressive pathophysiology.
• •We investigated the age-dependent effect of SKC activation on depressive-like behaviors and cognitive functions using 1-EBIO
• •1-EBIO treatment decreased locomotor activity, sucrose preference and alternation index in an age dependent manner.
• •We observed significant reduction in total klotho protein in HPC, but not in PFC and DRN.
• •The study indicates activity of SKCS in the pathology of depression and cognitive deficit, and interaction with klotho.

1. Introduction

Depression is a common type of affective disorder that is characterized by low mood, loss of interest and long-term disability, leading to serious functional disability in patients. The pathology of depression is multi-faceted and complex. Among the established pathologies of depression, chronic stress exposure can recruit factors like inflammatory processes and failed neuroplasticity, along with genetic and epigenetic modifications (Bambico and Belzung, 2013; Price and Duman, 2020; For a detailed review, please see Willner, 2017) to produce the behavioral and neuropathological symptoms of depression. However, current empirical knowledge on how stress recruits and weighs between the pathogenic mechanisms to determine disease trajectory and progression is somewhat limited.

Widespread disarray in the network dynamics and information processing throughout the limbic system-the prefrontal cortex (PFC), the hippocampus (HPC) and the amygdala- are widely recognized (Price and Duman, 2020). The hippocampal formation is the central brain region that processes many types of learning and memory (Bartsch and Wulff, 2015). The brain structure contains a high number of glucocorticoid receptors, and it regulates the hypothalamus-pituitary-adrenal (HPA) axis, which renders the structure more susceptible to stress and stress-induced pathologies, e.g., depression. Repeated exposure to stress activates the HPA axis, leading to increased glucocorticoid release, and activation of glucocorticoid receptors. This, in turn, leads to increased metabolic activity in the cell and increased release of glutamate. Upon release, glutamate activates N-methyl-D-aspartate receptors (NMDARs) and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs), which can potentially lead to cellular hyperactivity and excitotoxicity. This insidious effect is also compounded by the activation of microglia due to increased exposure to glucocorticoids (Walker et al., 2013). Moreover, chronic exposure to stress decreases the expression of neurotrophic factors like brain-derived neurotrophic factor (BDNF) (Song et al., 2006). The loss of neurotrophic support leads to the shrinkage of the dendritic tree of hippocampal neurons (Bessa et al., 2009) and, ultimately, the loss of granule cells (Jayatissa et al., 2010). The profound negative effect of increased glucocorticoids on hippocampal neurogenesis also contributes to morphological abnormalities.

Activation of NMDA and AMPA receptors by glutamate causes calcium ion mobilization. Elevated concentration of circulating glucocorticoids thus may facilitate the activation of NMDARs as well (Weiland et al., 1997), which ultimately causes an imbalance in calcium homeostasis, abnormalities of which is heavily implicated in depression (Ureshino et al., 2019). In a rat model of restraint stress, an elevation in the extracellular level of glutamate has been reported (Lowy et al., 1993), and circulating glucocorticoids seem to be potentiating the increased extracellular level of glutamate (Moghaddam et al., 1994; Schasfoort et al., 1988). Interestingly, in aging animals, glutamate release after exposure to a stressful stimulus is markedly increased (Lowy et al., 1995).

The small-conductance calcium-activated potassium channel subfamily (SKC/KCa2) is a crucial player in calcium homeostasis. These inhibitory channels are known to generate the medium afterhyperpolarization component of an action potential, essentially comprising the refractory period (Adelman et al., 2012; Faber, 2010). Three subtypes (SK1/KCa2.1-SK3/KCa2.3) have been identified and are widely expressed within the CNS, especially in the limbic system and basal ganglia (Stocker and Pedarzani, 2000). Each subunit has six transmembrane hydrophobic alpha-helical domains and three associated subunits: the protein phosphatase 2A (PP2A), casein kinase 2 (CK2), and calmodulin bound to calmodulin-binding domain (CaMBD), which is responsible for the sensitivity to calcium transients within the intracellular space (Adelman et al., 2012). This enables SKCs to have a unique potential to couple intracellular calcium concentrations with very low changes in potassium conductance and membrane potential. The activity of the SKCs can be modulated via postsynaptic muscarinic receptors or NMDARs (Faber, 2010; Ngo-Anh et al., 2005), which in turn regulates excitatory postsynaptic potentials, inter-spike intervals, and spike frequency adaptions. Through these mechanisms, SKCs mediate activity-dependent plasticity and long-term potential (LTP)-like changes that affect memory-related and cognitive functions (Faber, 2010; Kshatri et al., 2018).

Klotho is a recently discovered protein that is associated with life extension. The protein is highly expressed in the choroid plexus and HPC (Pavlatou et al., 2016). Overexpression of klotho has been associated with increased cognitive function (Dubal et al., 2014), while decreased expression shows the opposite effect (Kuro-o et al., 1997). Interestingly, the memory impairment seen in klotho knockout (−/−) mice is age dependent (Nagai et al., 2003). The effect is thought to be mediated via the enrichment of the synaptic GluN2B subunit of the NMDAR in the postsynaptic density of hippocampal neurons (Wu et al., 2022) On the other hand, in clinical experiments, electroconvulsive therapy -a highly effective antidepressant therapy- significantly increases klotho level in the cerebrospinal fluid (CSF) of geriatric patients (Hoyer et al., 2018). Additionally, patients with major depressive disorder have low klotho levels in CSF (Hoyer et al., 2018), and klotho levels positively correlate with the severity of depressive symptoms (Prather et al., 2015). In preclinical settings, the use of the rodent chronic unpredictable stress model of depression has shown that klotho mRNA is decreased in the choroid plexus (Sathyanesan et al., 2012). Klotho also regulates behavioral response to stress via its activity on GluN2B function in the nucleus accumbens of mice (Wu et al., 2022). Even though studies investigating klotho in depression is scarce, these abovementioned findings indicate a critical role of klotho in stress and depression. However, the precise mechanism undergirding klotho's involvement is still unclear.

Therapeutically targeting SKCs has gained traction in recent years. Clinical studies have implicated the SK3 channel in cognitive disorders, depression, and aging (Chandy et al., 1998; Tomita et al., 2003). In addition, SK3 overexpression has been associated with hippocampal shrinkage and LTP deficits (Blank et al., 2003; Martin et al., 2017). Interestingly, SK1-3 is expressed in structures implicated in depression and rapid antidepressant response, e.g., the cingulate cortex and the serotonin-producing dorsal raphe nucleus (DRN) (Sailer et al., 2002; Stocker and Pedarzani, 2000). We have recently shown that in a rat model of chronic unpredictable mild stress (UCMS), pharmacological blockade of SKCs leads to positive neuroplastic changes in the medial prefrontal cortex (mPFC), and the effect was associated with a rapid antidepressant response (Bambico et al., 2020). On the other hand, exposure to stress is associated with SK3 overexpression in the DRN (Sargin et al., 2016). In addition, stress is associated with induced glucocorticoid release that rapidly mobilizes calcium influx and enhances the transcription and insertion of SKCs via GR2 receptors (Levitan et al., 1991; Shipston et al., 1996). As such, there could be an undiscovered causal relationship between the activity of SKCs and the concentration of klotho in stress-induced conditions.

The aim of our current study was to investigate whether manipulation of SKCs at different age points could precipitate depressive-like behavior and cognitive deficits, and whether this could be associated with klotho activity changes. Using two different age groups and pharmacological treatments in mice, we ascertained that pharmacological inhibition of SKCs could precipitate depressive-like behavior and cognitive deficits. Indeed, there seems to be an age-dependent interplay between SKCs and klotho expression differentially affecting cognitive-affective functioning at different age groups.

2. Materials and methods

2.1. Animals

All procedures of our experiment followed the guidelines of the Canadian Council on Animal Care (CCAC) and Memorial University of Newfoundland's Animal Care Committee. Ninety male BALB/c mice from two different age groups: 10–11 weeks (n = 50) and 45–46 weeks (n = 40) were obtained from Charles River (Quebec, Canada) and weighed 25–30 g at the beginning of the experiment. Animals were single-housed and kept under standard vivarium conditions (12-h light–dark cycle, lights on at 7:00 a.m. and off at 7:00 p.m.; temperature at 20 ± 2 °C; 50–60% relative humidity) and had access to ad libitum food and water. After 1 week of acclimatization, animals were grouped into four groups: control and treatment cohorts for young (3 months) and mid-aged (12 months) groups. Schematic representation of the experimental design and the chronology of procedures is provided in Fig. 1.

Fig. 1.

Schematic representation of the experimental design.

2.2. Pharmacological treatment

1-ethyl-2-benzimidazolinone (1-EBIO), a potent positive allosteric modulator of SKCs (Faber, 2010) was purchased from Sigma Milipore (Oakville, ON). It was dissolved in a vehicle containing 1% di-methyl sulfoxide (DMSO) in saline for IP injections. The control treatment was saline. The drug treatment groups were injected with 1-EBIO at a dose of 1.0 mg/kg, 30 min before the scheduled trial. The dose was selected based on data from our pilot experiment, where we found that 1.0 mg/kg was sufficient to elicit a significant behavioral outcome. The control animals were injected with saline. The total duration of daily treatments was 15 days.

2.3. Behavioral tests

2.3.1. Open field test (OFT)

Open field testing for locomotor activity was performed as described earlier (Al-Amin et al., 2014). Square wooden boxes, painted with a non-absorbent, non-reactive and non-reflective primer, and measuring 50 cm (length) x 50 cm (width) x 38 cm (height) were used for the test. Animals were brought to the test room in their home cages and left undisturbed for atleast 30 min to habituate them to the testing environment. After treatment, as described in the previous section, mice were put in the open field, and locomotor activity was recorded for 10 min. The apparatus was wiped clean with 70% ethanol and air-dried between trials. Trials were recorded with a ceiling-mounted video camera that ran the Dazzle DVD Recorder software (Pinnacle Studio by Correl, USA). The records were processed and used for offline analysis. Locomotor activity was determined by two dependent measures: 1) Total distance traveled (in cm), defined as the sum of the recorded movements of the animal's center of gravity, and 2) thigmotaxis, defined as the time spent in the central area divided by the sum of time spent in the central area and peripheries:

$$\text{Thigmotaxis}\text{time}(\%)=\frac{\text{central}\text{area}(\mathrm{s})}{\text{central}\text{area}(\mathrm{s})+\text{peripheral}\text{area}(\mathrm{s})}\mathrm{x}100$$

Thigmotaxis is an indicator of anxiety-like behavior. A lower thigmotaxis score indicates more time is spent in the peripheral areas. The more anxious the mouse is, the more time the mouse will spend in the peripheral areas (closer to the wall).

2.3.2. Sucrose preference test (SPT)

The test was carried out overnight in the mouse's home cage. Two plastic cups filled with 4% sucrose solution and tap water, respectively, were placed inside the cage. The cups were glued to a tile block that was taped to the floor to avoid accidental spillage. The sucrose preference index was calculated as the ratio of sucrose solution consumption to the total fluid intake, as described by the equation below:

$$\text{Sucrose}\text{preference}=\frac{\text{Sucrose}\text{consumption}(\text{gm})}{\text{Sucrose}\text{consumption}(\text{gm})+\text{water}\text{consumption}(\text{gm})}\mathrm{x}100$$

2.3.3. Forced swim test (FST)

The test was carried out as described elsewhere (Porsolt et al., 1977). Mice were individually placed in clear glass cylinders measuring 25 cm (height) x 9 cm (diameter). They contained approximately 18 cm of water (temperature maintained at 20-25 °C), from which they could not escape. The experiment lasted 5 min and was recorded using a video camera. After the procedure, mice were removed from the water cylinder, dried with a paper towel, then returned to the home cage, which was placed over a heat source. The last 4 min of the video were used to analyze active and passive coping behavior.

2.3.4. Spontaneous alteration in the Y-maze

Spontaneous alteration test was performed as described elsewhere (Garcia and Esquivel, 2018). Three acrylic dividers were used to convert an 8-arm radial maze into a Y-maze. Each arm of the Y-maze was 30 cm × 10 cm x 15 cm in dimension. The arms were interconnected with 120° angles between them. The exterior walls of each arm were covered with white paper to prevent the animals from using visual cues to navigate the maze. During the test, animals were each placed at the center of the maze and allowed to navigate the maze for 5 min. Each trial was recorded for further offline analysis. Spontaneous alteration was calculated using the equation below:

$$\text{Spontenous}\text{Alternation}=\frac{\text{Alternation}}{\text{Total}\text{number}\text{of}\text{entry}-2}\mathrm{x}100$$

2.4. ELISA for klotho concentration

For klotho ELISA, samples were extracted from flash-frozen brains and homogenized in ice-cold PBS buffer (pH = 7.4) using a Benchmark D1000 handheld homogenizer. The samples were then centrifuged at 5000 x g and at 4 °C for 5 min. Supernatants were collected and used for quantification of total klotho protein. Klotho proteins were analyzed in the PFC, HPC, and DRN using an ELISA kit (Biomatik, Kitchener, ON) as per the manufacturer's instructions. The developed plate was read at 450 nm using an Epoch2 plate reader (Biotek, Winooski, VT).

2.5. Behavioral video and data analysis

All behavioral videos were analyzed using Ethovision XT Version 14.02 by Noldus (Wageningen, the Netherlands). All data were statistically analyzed using mixed-design ANOVAs followed by the Student Newman-Keuls (SNK) post hoc test. A probability value of p ≤ 0.05 was considered to be statistically significant. Non-parametric analyses were used when assumptions of normality and homogeneity of variance were not met. Excel 2016 (Microsoft Corporation, WA, USA), Sigma Plot version 12.3 and Jamovi version 0.9 (the Jamovi project, 2019; retrieved from https://www.jamovi.org) were used for data analyses. The ELISA data set was analyzed using the Gen5 imaging software developed by Biotek for Epoch2 plate reader.

3. Results

3.1. 1-EBIO affected locomotor activity in the mid-life cohort, but not in the young cohort

A 2 × 2 between-groups ANOVA on the total distance of travel showed that using age and treatment as main factors resulted in a significant effect of age [F_(1,69) = 12.318, p < 0.001] and treatment [F_(1,69) = 7.101, p = 0.01], but without a significant interaction [F_(1,69) = 1.010, p = 0.318]. Post hoc analysis revealed that 1-EBIO treatment (M = 1934.13 ± 160.73) decreased the total distance traveled in the mid-life animals compared to mid-life controls (M = 1769.50 ± 188.68; p = 0.024). However, this effect was not seen in the 1-EBIO-treated young cohort (M = 1934.13 ± 180.30) compared to age-matched controls (M = 2257.64 ± 186.86; p = 0.174) (see Fig. 2 A).

Fig. 2.

A) Mean total distance traveled and B) mean thigmotaxis time in the open field test. Young: n = 23–25/group, mid-life: n = 13–15/group. All data are presented as mean ± SE (standard error of the mean). *p ≤ 0.05.

Additionally, statistical analysis of thigmotaxis time percentage found a significant main effect of age [F_(1,67) = 4.588, p = 0.031), without a significant main effect of treatment F_(1,67) = 0.00061, p = 0.980] and no significant interaction [F_(1,67) = 0.0137, p = 0.907]. The follow-up post hoc test showed an overall decrease in the thigmotaxis time of the mid-life cohort (M = 2.77 ± 0.62) compared to their young cohort (M = 9.12 ± 1.93; p = 0.031). No other pairwise comparisons were significant (Fig. 2 B).

3.2. 1-EBIO treatment does not affect coping behavior in forced swim test

A 2 × 2 between-groups ANOVA on the immobility time in the forced swim test showed a significant main effect of age [F_(1,62) = 8.074, p = 0.006] without a significant effect of treatment [F_(1,67) = 3.514, p = 0.066]. However, follow up post hoc analysis failed to reveal a significant difference in the immobility time of the 1-EBIO-treated mid-life cohort (M = 156.21 ± 11.66) compared to the mid-life control cohort (M = 179.64 ± 31.07; p = 0.064). Similarly, no significant difference was observed in the immobility time when the young control cohort (M = 195.44 ± 27.66) was compared to the young 1-EBIO-treated cohort (M = 187.68 ± 31.59; p = 0.485) (see Fig. 3 A)

Fig. 3.

A) Mean immobility time and B) mean swimming time in the forced swim test. Young: n = 17–20/group, mid-aged: n = 14–15/group. All data are presented as mean ± SE (standard error of the mean). *p ≤ 0.05.

Similar statistical analyses were performed on the swimming time using age and treatment as the main factors. Results indicated a significant main effect of age [F_(1,62) = 6.866, p = 0.011)] and treatment [F_(1,62) = 5.074, p = 0.028)] but a non-significant interaction between the factors [F_(1,62) = 0.358, p = 0.552)]. But, as observed in the case of immobility time, follow up post hoc analysis failed to reveal a significant difference in the swimming time between the 1-EBIO-treated mid-life cohort (M = 42.46 ± 21.92) and the mid-life control cohort (M = 30.66 ± 12.94; p = 0.062). Similarly, no significant difference was observed in the swimming time when the young control cohort (M = 22.27 ± 13.83) was compared to the 1-EBIO-treated young cohort (M = 29.11 ± 17.98; p = 0.219) (see Fig. 3 B).

3.3. 1-EBIO treatment decreased sucrose preference in mid-life cohort, but not in the young cohort

A 2 × 2 between-groups ANOVA using age and treatment as main factors revealed significant main effects of age [F_(1,80) = 7.047, p = 0.010] and treatment [F_(1,80) = 4.317, p = 0.041], but with a non-significant interaction between the factors [F_(1,80) = 3.265, p = 0.075]. SNK post hoc analysis revealed that 1-EBIO-treated mid-life animals (M = 47.0 ± 2.67) had a significantly lower sucrose preference than the age-matched, vehicle-treated animals (M = 57.3 ± 3.17; p = 0.002). This effect was not seen in the 1-EBIO-treated young animals (M = 58.8 ± 2.25) when compared to their age-matched controls (M = 59.5 ± 3.17; p = 0.843) (Fig. 4 A).

Fig. 4.

A) Mean sucrose preference index and B) mean alternation index in the sucrose preference test and spontaneous alternation test, respectively. SPT: young: n = 23–24/group, mid-aged: n = 19–20/group; Spontaneous alternation test: young: n = 23–24/group, mid-aged: n = 14–15/group. All data are presented as mean ± SE (standard error of the mean). *p ≤ 0.05.

3.4. Treatment with 1-EBIO affects spatial working memory of only the mid-life cohort

A 2 × 2 between-groups ANOVA design using age and treatment as main factors resulted in non-significant main effects of age [F_(1,73) = 0.115, p = 0.735)] and treatment [F_(1,73) = 3.614, p = 0.061]. However, a significant interaction between these two factors was found [F_(1,73) = 4.617, p = 0.035]. Post hoc analysis revealed that 1-EBIO treatment significantly decreased the alternation index in the mid-life cohort (M = 34.74 ± 2.42) compared to the mid-life control cohort (M = 47.58 ± 3.37; p = 0.010). The effect was not seen in the young cohort: the mean alternation index of the 1-EBIO treated animals (M = 39.26 ± 3.62) was not significantly different from the control animals (M = 35.86 ± 3.79; p = 0.848) (Fig. 4 B).

3.5. 1-EBIO treatment in the mid-life cohort significantly reduces klotho in the HPC but not in the PFC or DRN

Three separate 2 × 2 between-groups ANOVAs with age and treatment as main factors were run to investigate the effect of 1-EBIO treatment on klotho concentration in the PFC, HPC and DRN.

In the PFC, there was a significant main effect of age [F_(1,19) = 31.579, p < 0.001], but no significant effect of treatment [F_(1,19) = 0.531, p = 0.475] and no significant interaction between age and treatment [F_(1,19) = 0.327, p = 0.574]. Post hoc analysis failed to show a significant difference between the 1-EBIO- treated mid-life cohort (M = 18.02 ± 8.64) and the mid-life control cohort (M = 15.80 ± 6.07, p = 0.907). Additionally, there was no significant difference in klotho level between the 1-EBIO-treated young cohort (M = 105.88 ± 13.28) and the youngcontrol cohort (M = 105.88 ± 13.28, p = 0.397) (Fig. 5 A)

Fig. 5.

Mean klotho concentration in the A) PFC B) HPC and C) DRN. Protein concentration is expressed as ng of protein per mg of brain tissue. N = 4–7/group. All data are presented as mean ± SE (standard error of the mean). *p ≤ 0.05.

In the HPC, a significant main effect of age was seen [F_(1,18) = 4.234, p = 0.05], but the effect of treatment was not significant [F_(1,18) = 2.242, p = 0.152]. Also, no significant interaction between the factors was seen [F_(1,18) = 1.727, p = 0.205]. Post hoc analysis showed that in the mid-life cohort, 1-EBIO treatment significantly reduced the concentration of klotho (M = 14.54 ± 6.96) when compared to the vehicle-treated, mid-life control cohort (M = 58.80 ± 25.15; p = 0.05). Significant effects were also absent when comparing the 1-EBIO-treated young cohort (M = 67.61 ± 14.80) with the vehicle-treated young control cohort (M = 70.49 ± 27.43, p = 0.903) (Fig. 5 B).

In the DRN, a significant main effect of age was seen on the concentration of klotho [F_(1,13) = 11.45, p = 0.005], but no significant effect of treatment was found [F_(1,13) = 0.327, p = 0.574]. The interaction between the main factors was not significant as well [F_(1,13) = 0.227, p = 0.608]. SNK post hoc analysis didn't show any significant difference between the mid-life control cohort (M = 50.03 ± 8.44) and the 1-EBIO-treated mid-life cohort (M = 23.87 ± 17.80, p = 0.583). Pairwise comparison between young control animals (M = 165.68 ± 105.44) and age-matched, 1-EBIO-treated animals (M = 172.66 ± 88.47, p = 0.883) was also not significant (Fig. 5 C).

4. Discussion

In this study, we have demonstrated that repeated activation of SKCs by 1-EBIO administration robustly precipitated depressive-like behavior. Additionally, the drug treatment also decreased klotho concentration in the HPC in an age-dependent manner, and the effect of treatment was not seen in either the PFC or the DRN. Systemic administration of 1-EBIO decreased total distance traveled in the mid-life cohort but not in the young cohort. In addition, the drug did not seem to affect thigmotaxis in any of the cohorts. In the forced swim test, 1-EBIO treatment did not seem to affect immobility time or swimming time in any of the cohorts, therefore indicating an absence of drug treatment effects on stress-coping behavior. 1-EBIO also decreased sucrose preference in mid-life animals, along with an age-dependent decrease of alternation index in the Y-maze spontaneous alternation test. The behavioral and molecular effects were likely due to the drug's ability to activate SKCs. To our knowledge, this study is the first reported evidence of the age-dependent impact of pharmacological SKC activation on cognitive and depression-related phenotypes in an animal model. It is also the first to demonstrate that prolonged pharmacological SKC activation can affect klotho concentration in a region-specific manner.

Our findings indicated an age-dependent effect of 1-EBIO treatment on locomotor activity, as it was seen in the mid-life cohort only. The precise mechanism or site of action of the motor-impairing effect of 1-EBIO is yet to be identified. SK3 is highly expressed in the substantia nigra, one of the major dopaminergic nuclei in the brain (Sailer et al., 2002). Brain dopamine level is crucial for normal motor control, and preclinical studies have shown that pharmacological inhibition of SKCs by apamin, a classic SKC blocker, improves motor deficits in rodent models (Jd and Pw, 1990; Chen et al., 2014). On the other hand, activation of SKCs using systemic 1-EBIO treatment have been shown to result in decreased burst firing of dopaminergic neurons in the substantia neurons (Ji and Shepard, 2006). These findings indicate that the effect seen in our open-field test may have resulted from suppressing dopaminergic neurons in the nigrostriatal dopaminergic, the mesolimbic-mesocortical pathway, or both. The observed decrease in the locomotor movement of the mid-life cohort may have been a result of age-dependent increased expression of SK3 channels in the substantia nigra. This increased expression could lead to greater suppression of dopaminergic neurons in the presence of an SKC activator, such as 1-EBIO. However, our findings contrast with the report of Anderson and colleagues, who have found that 1-EBIO injection at (>20 mg/kg) produced motor ataxia on the rotarod test, but doses up to 40 mg/kg did not affect the exploratory motor activity in the open field test (Anderson et al., 2006). On the contrary, in an experiment where different 1-EBIO doses were tested on motor activity, Vick et al. (2010) reported that ≥10 mg/mg resulted in a significant decrease in locomotor activity (Vick et al., 2010). However, the effect was diminished within 30 min. Differences in the mouse strain, route of drug administration, and age during testing may have contributed to the disparity between these studies.

The forced swim test (FST) data are difficult to interpret. Multiple preclinical studies have shown the antidepressant activity of potassium channel blockers in the FST (Galeotti et al., 1999; Inan et al., 2004; Kaster et al., 2005). In addition, we have recently demonstrated that genetic deletion or pharmacological inhibition of SK3 produces an antidepressant-like effect in the FST (Nashed et al., 2022). On the other hand, potassium channel openers induce depressive-like behavior in this test (Galeotti et al., 1999). In our experiment, we observed a non-significant effect of 1-EBIO on FST behavior in both young and mid-life cohorts. While we do not have a definite answer for this, it is possible that the effects on neural pathways mediating FST behaviour are not particulary profound or are masked by the drug's effects on motor activity that could confound the FST phenotype. Previously, our lab has reported that 4 mg/kg IV 1-EBIO treatment in rats exposed to UCMS also did not show decreased immobility time when compared to controls (Bambico et al., 2020), which is in line with our current findings.

Treatment with 1-EBIO showed a pro-depressive-like effect in the sucrose preference test evident in the mid-life cohort, as it reduced sucrose preference scores. Anhedonia-like behavior, indicated by the reduction in sucrose preference scores, is thought to be mediated via the brain's reward system, which includes the ventral tegmental area (VTA), amygdala, PFC, and HPC- all of which are implicated in depression pathophysiology. Depression paradigms like UCMS are known to negatively alter the excitability of neurons in brainstem monoamine nuclei like VTA dopaminergic neurons (Chang and Grace, 2014) and DRN 5-HT neurons (Qu et al., 2019). This suppression in excitability may be a result of increased expression of SKCs, among other potenital mechanisms (Bambico et al., 2020; Nashed et al., 2022). The pharmacological activation of SKCs may forge effects that are similar to those by UCMS. The non-significant effect of 1-EBIO treatment in the young cohort may be attributed to the younger age and the comparatively lower expression of SKCs at this age. In the mid-life cohort, where expression of SKCs may have increased due to aging, activation of SKCs by 1-EBIO could have substantial adverse effects on the neuronal firing activity.

In the spontaneous alternation test in the Y-maze, our finding indicates that SKC activation affects working memory in mid-life animals, and this effect is absent in young animals. SKCs are known to mediate activity-dependent and LTP-like synaptic plasticity, which represents, albeit partly, the molecular mechanism of learning and memory (Adelman et al., 2012; Faber, 2010; Ngo-Anh et al., 2005). Thus, inhibition of SKCs may positively affect cognition, and activating SKCs could negatively affect learning and memory. Indeed, apamin increases LTP in the Schaffer's collateral after a brief tetanic stimulation and decreases the threshold of LTP generation in the CA1 region (Behnisch and Reymann, 1998; Kramár et al., 2004; Stackman et al., 2008). On the other hand, increased expression of the SK3 channel generates age-dependent LTP and memory deficits (Blank et al., 2003). Despite numerous studies on the influence of apamin on learning and memory, only a handful of studies have investigated the possible negative effect of SKC activation on them. The study by Vick et al. (2010) is the only study that has reported the effect of SKC activation by 1-EBIO and CyPPA on explicit memory, as tested by fear conditioning and novel object recognition tasks (Vick et al., 2010). The non-significant effect of 1-EBIO on young animals is supported by previous studies investigating the effect of apamin in spatial working memory. Systemic administration of apamin did not affect spontaneous alternation in animals with intact brains or in animals with medial septal lesions (Ikonen et al., 1998). Since hippocampal gamma-amino-butyic acid (GABA) neurons contain copious amounts of SKC, it could be possible that 1-EBIO may have decreased GABA activity within the HPC, which may have contributed towards intrinsic hippocampal excitability, resulting in a masked or constrained response of principal neurons to SKC activation at a young age. Indeed, elevated hippocampal activity, particularly along the ventral HPC-nucleus accumbens pathway, can predict depression-related vulnerabilities (Muir et al., 2020). Indeed, our earlier findings showed increased hippocampal expression of zif268, an immediate early gene marker of cell activation, following UCMS exposure in rats (Nashed et al., 2022). This could therefore highlight the differential expression of SKCs across different subpopulations of cells in the hippocampal formation, as well as in different neuronal compartments (soma vs dendrites) (Kuiper et al., 2012). As a consequence of aging, when SKC expression have increased significantly, the effect of 1-EBIO may be magnified by elevated activity of SKCs in all types of hippocampal neurons, e.g., pyramidal cells and GABAergic interneurons. Indeed, Martin and colleagues have reported that doxycycline-induced overexpression of SK3 in the HPC of transgenic mice causes cognitive impairments related to hippocampal atrophy (Martin et al., 2017).

Klotho is an anti-aging pleiotropic hormone primarily produced in the kidney and choroid plexus in the brain. When overexpressed, klotho extends life span (Kurosu et al., 2005), synaptic plasticity, and cognitive functions (Dubal et al., 2014). On the other hand, when klotho concentration is decreased, aging-related phenotypes are expressed (Kuro-o et al., 1997). The mechanism of klotho is multifaceted and complex. The protein regulates NMDAR signaling, enhances cognition (Dubal et al., 2014) and influences calcium homeostasis and ion channel clustering (Imura et al., 2007). Dubal et al. (2014) demonstrated that klotho almost doubled the GluN2B receptor subunit through a post-transcriptional mechanism in the HPC and PFC, along with elevated expression of fos (Dubal et al., 2014). Additionally, in humans, klotho level is reduced under high stress conditions (Prather et al., 2015). We found an overall decrease in klotho concentration in the PFC and DRN regardless of treatment. As the aging brain progressively expresses low levels of klotho, this regional decrease of klotho seems to be independent of SKC manipulation, but rather an indication of ongoing aging processes.

Our most striking finding is the interaction between klotho and 1-EBIO in the HPC. A positive correlation between hippocampal aging and increased plasma cortisol/corticosterone has already been established before (Prather et al., 2015). In addition, hippocampal neurons participate in the negative feedback control of the HPA axis, and it is postulated that HPC loses some of its negative feedback control due to aging, resulting in elevated glucocorticoids in plasma. It may be possible that an increased level of glucocorticoids in plasma acts on glucocorticoid type2 receptors in the HPC. This could result in rapid synthesis and insertion of SKCs in the cell membrane, possibly on GABAergic interneurons, leading to decreased GABAergic inhibitory activity in the HPC. It has been previously reported that at the beginning of mid-age, glutamic acid decarboxylase (GAD) expression is dramatically inhibited in many interneurons (Landfield et al., 1978). This may lead to a profound reduction in the synthesis of GABA and ultimately to increased excitability of the HPC. This aberrant excitability is believed to contribute to HPC-based memory impairments. The absence of strong GABAergic tone in the HPC may predispose the HPA stress axis towards allostatic overload, giving way to widespread hippocampal shrinkage and atrophy seen in stress-induced depression. These changes can ultimately be associated with decreased klotho activity. In addition, studies have shown that klotho exerts a neuroprotective effect on hippocampal neurons by recruiting anti-oxidant enzymes (Hanson et al., 2021; Stanley and Shetty, 2004). Oxidative stress is an important pathological mechanism related to cellular apoptosis, cognitive impairment, and depression. It may be possible that in the mid-life cohort, the already decreasing levels of klotho were magnifying the negative effects of SKC activation, potentiating further the decrease in klotho levels (see Fig. 6 for a schematic represantion of putative mechanisms)

Fig. 6.

Putative mechanism of 1-EBIO-induced klotho decrease. Aging and stress increasecirculating glucocorticoid levels, which then increase the activation of GR2-receptors. This activation causes rapid synthesis and insertion of SKCs on gamma-amino-butyric acid (GABA)ergic interneurons, which leads to decreased electrical excitability. Systemically administered 1-EBIOactivatesSKCs and cell hyperpolarization, leading to diminished GABAergic control on principal hippocampal cells. This then leads to widespread pyramidal disinhibition, and neuronal hyper-excitability. When this progresses, excitotoxicity can ensue, eventually causing decreased klotho concentrations and hippocampal atrophy and shrinkage. Image created with Biorender Online tool. © S M Nageeb Hasan.

With these data in our hands, we surmise that it is reasonable to consider that SKCs can be targeted to elicit an antidepressant and cognitive-enhancing response. Because the channel is directly involved in regulating neuronal activity, SKC inhibitors can elicit robust and fast-acting antidepressant action. Among the three subtypes of SKCs, it has been found that SK3 is predominantly expressed in the PFC and in hippocampal GABAergic interneurons, as well as, in monoaminergic neurons (Martin et al., 2017; Sailer et al., 2002). Indeed, both GABAergic and monoaminergic neurotransmission are compromised in depression. The compromised neurotransmission can be rescued using a specific SK3 inhibitor or negative allosteric modulator. A decrease in the intrinsic excitability of DRN serotonergic neurons, secondary to increased SK3 expression, was deemed directly related to stress and depression pathophysiology (Sargin et al., 2016). This may be reversed through SK3 inhibition, consequently re-eestablishing or normalizing serotonin release in the projection sites of the DRN, e.g., in the PFC, HPC and amygdala. Also, inhibition of SK3 on GABAergic interneurons in the HPC may result in the retention of the GABAergic tone in the HPC, which may regain the negative feedback control of imposed by the HPC on the HPA axis by decreasing the excitatory drive along hippocampal projections.

Moreover, some aspects of depression pathophysiology are present in some neurodegenerative disorders like Alzheimer's disease (AD). Strong clinical evidence suggests that depression may be a preclinical stage of incipient dementia (Diniz et al., 2013; Ownby et al., 2006). Depression is the most common psychiatric disorder, frequently comorbid with mild cognitive impairment (MCI) and AD. Reduced HPC volume is reported in both of these disorders, and SK3 overexpression may be a common pathology, as overexpression of the SK3 channel causes hippocampal shrinkage, LTP deficits, and upregulation of genes related to AD (Blank et al., 2003; Martin et al., 2017). Interestingly, recent accumulating evidence indicates the anti-AD effects of klotho. In clinical settings, klotho is significantly lower in AD patients (Massó et al., 2015), and in preclinical models, klotho overexpression is associated with enhanced cognitive performance and increaed clearance of AD pathological markers (Zeng et al., 2019; Zhao et al., 2020).

To build upon our current findings, future investigations on the therapeutic potential of SKC antagonists/negative allosteric modulators are warranted. One limitation of our study was that we could not identify the specific subtype of SKCs producing the observed behavioral deficits in 1-EBIO-treated animals. However, the extant literature suggests a role for SK3 (Martin et al., 2017, Tomita et al., 2003). Additionally, the effect of SK3 manipulation on age-dependent deficits were not investigated. Future investigation could focus on this subtype and could explore the interactions between SK3 overexpression and klotho activity. In addition, further investigation is required to see if SKC overexpression could also be verified and tested in preclinical models of AD, as well as, whether modulation of SKCs can produce any beneficial effects.

In summary, our findings indicate that chronic activation of SKCs precipitates depressive-like behavior and cognitive impairment in an age-dependent manner and is accompanied by decreased levels of the klotho protein in the HPC. Further studies are required to better understand the mechanism underlying the effect of SKC activation on the activity of klotho. Our findings provide further evidence for the role of SKCs in the pathology of depression and cognitive impairment. This study also provides a rationale for the future development of novel therapeutics modulating SKC activity for depression and other stress-related disorders.

Conflict of interest

• We declare that we have no conflict of interest.

Author contribution

NH and FRB conceptualized and planned the study. NH, AH, DW, RN ran the behavioral experiments. NH and KI performed the ELISAs. CL, AH, DW and KI assisted in data curation. NH analyzed all data and wrote the manuscript draft. FRB, and CL revised the manuscript. All authors approved the final version of the manuscript.

Funding

The study was funded by grants from Canadian Institutes of Health Research (CIHR), Canada (FRB), and National Science and Engineering Research Council of Canada (NSERC), Canada (FRB). The grants had no further role in the study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.