Astrocyte-dependent protective effect of quetiapine on GABAergic neuron is associated with the prevention of anxiety-like behaviors in aging mice after long-term treatment

Authors

  • Junhui Wang,

    1. Mental Health Center, Shantou University, Shantou, Guangdong, China
    2. Department of Psychiatry, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Shenghua Zhu,

    1. Department of Pharmacology and Therapeutics, University of Manitoba, Winnipeg, Manitoba, Canada
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Hongxing Wang,

    1. Beijing Anding Hospital, Capital Medical University, Beijing, China
    Search for more papers by this author
  • Jue He,

    1. First Affiliated Hospital, Henan University, Kaifeng, Henan, China
    Search for more papers by this author
  • Yanbo Zhang,

    1. Department of Psychiatry, College of Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
    Search for more papers by this author
  • Abulimiti Adilijiang,

    1. Department of Psychiatry, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada
    Search for more papers by this author
  • Handi Zhang,

    1. Mental Health Center, Shantou University, Shantou, Guangdong, China
    Search for more papers by this author
  • Kelly Hartle,

    1. Department of Psychiatry, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada
    Search for more papers by this author
  • Huining Guo,

    1. Department of Psychiatry, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada
    Search for more papers by this author
  • Jiming Kong,

    1. Department of Human Anatomy and Cell Science, University of Manitoba, Winnipeg, Manitoba, Canada
    Search for more papers by this author
  • Qingjun Huang,

    Corresponding author
    1. Mental Health Center, Shantou University, Shantou, Guangdong, China
    • Address correspondence and reprint requests to Dr Xin-Min Li, Department of Psychiatry, Faculty of Medicine and Dentistry, University of Alberta, 1E7.31 Walter C. Mackenzie Health Sciences Centre, Edmonton, Alberta T6G 2B7, Canada. E-mail: xinmin@ualberta.ca (or) Dr Qingjun Huang, Mental Health Center, Shantou University, North Taishan Road, Shantou, Guangdong 515063 China. E-mail: huangqj@stumhc.cn

    Search for more papers by this author
  • Xin-Min Li

    Corresponding author
    1. Department of Psychiatry, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada
    • Address correspondence and reprint requests to Dr Xin-Min Li, Department of Psychiatry, Faculty of Medicine and Dentistry, University of Alberta, 1E7.31 Walter C. Mackenzie Health Sciences Centre, Edmonton, Alberta T6G 2B7, Canada. E-mail: xinmin@ualberta.ca (or) Dr Qingjun Huang, Mental Health Center, Shantou University, North Taishan Road, Shantou, Guangdong 515063 China. E-mail: huangqj@stumhc.cn

    Search for more papers by this author

Abstract

Previous studies have demonstrated that quetiapine (QTP) may have neuroprotective properties; however, the underlying mechanisms have not been fully elucidated. In this study, we identified a novel mechanism by which QTP increased the synthesis of ATP in astrocytes and protected GABAergic neurons from aging-induced death. In 12-month-old mice, QTP significantly improved cell number of GABAegic neurons in the cortex and ameliorated anxiety-like behaviors compared to control group. Complimentary in vitro studies showed that QTP had no direct effect on the survival of aging GABAergic neurons in culture. Astrocyte-conditioned medium (ACM) pretreated with QTP (ACMQTP) for 24 h effectively protected GABAergic neurons against aging-induced spontaneous cell death. It was also found that QTP boosted the synthesis of ATP from cultured astrocytes after 24 h of treatment, which might be responsible for the protective effects on neurons. Consistent with the above findings, a Rhodamine 123 test showed that ACMQTP, not QTP itself, was able to prevent the decrease in mitochondrial membrane potential in the aging neurons. For the first time, our study has provided evidence that astrocytes may be the conduit through which QTP is able to exert its neuroprotective effects on GABAergic neurons.

image

The neuroprotective properties of quetiapine (QTP) have not been fully understood. Here, we identify a novel mechanism by which QTP increases the synthesis of ATP in astrocytes and protects GABAergic neurons from aging-induced death in a primary cell culture model. In 12-month-old mice, QTP significantly improves cell number of GABAegic neurons and ameliorates anxiety-like behaviors. Our study indicates that astrocytes may be the conduit through which QTP exerts its neuroprotective effects on GABAergic neurons.

Abbreviations used
ACM

astrocyte-conditioned medium

CNS

central nerve system

DIV

days in vitro

GAD

glutamic acid decarboxylase

MMP

mitochondrial membrane potential

QTP

quetiapine

Quetiapine (QTP) has been widely used in the treatment of schizophrenia and mood disorders (Yamamura et al. 2009). Although QTP is thought to interact with neurotransmitter receptors, the drug has also been known to significantly affect other systems in central nerve system (CNS). Our previous studies have shown that QTP may be a neuroprotectant (Bai et al. 2002; Wang et al. 2005). In a bilateral common carotid artery occlusion (CCAO) mouse model, we found QTP to significantly attenuate spatial memory impairment and neuronal cell loss in hippocampus (Yan et al. 2007a). Although the critical neuroprotective and neurotrophic role of QTP treatment has been well established at the experimental level, very little is known about the downstream mechanisms responsible for the beneficial influences of the drug.

A growing body of literature has shown that a deficit in the GABAergic system is involved in a variety of psychological disorders, including schizophrenia, anxiety, and depression (Lydiard 2003; Costa et al. 2004). The decrease of GABAergic signaling was considered to be one of the most robust pathologies observed in schizophrenia (Lodge et al. 2009). In addition, glutamic acid decarboxylase (GAD), which is an enzyme that catalyzes the decarboxylation of glutamate to GABA in interneurons, was markedly reduced in schizophrenia patients (Hashimoto et al. 2003). Dysfunction of GABAergic neurons has also been implicated in mood disorders. Animal studies have demonstrated GABA dysfunction as a link between depression and anxiety, and is now considered a common pathophysiology of both diseases (Mohler 2012). Abnormal GABA level and reduced GABAergic neurons were found in patients diagnosed with major depression (Sanacora et al. 2003; Rajkowska et al. 2007). Reduction of GABA transmission induces anxiety and correlates well with the severity of anxiety symptoms observed in panic disorder and post-traumatic stress disorder (Malizia et al. 1998; Bremner et al. 2000; Horowski and Dorow 2002). However, less is known about the effects of QTP on the GABA network in CNS. A recent epigenetic study showed that QTP could facilitate GABAergic promoter demethylation (Guidotti et al. 2011). Given these findings, the GABAergic neuronal system may be a possible target of QTP.

Astrocytes are increasingly acknowledged as being involved in some important functions in CNS. They actively constitute the ‘tripartite synapse’ with neuronal cells, including GABAergic neurons (Haydon and Carmignoto 2006; Egawa et al. 2013). In addition, astrocytes have specific relevance to the pathophysiology and treatment of mental disorders (Chen et al. 2006; Di Benedetto and Rupprecht 2013). Structurally, astrocytes have long processes that extend from the cell body, allowing the cell to interact with blood vessels which provide glucose, glutamine, and other nutrients to the brain (Choi et al. 2012; Wang et al. 2013). Because of their close proximity to blood vessels, astrocytes come into contact with infiltrated medications earlier than neuron, which validates the possible involvement of astrocytes during QTP treatment. However, most of the previous studies concerning the neuroprotection of QTP focus on its direct neuronal effects and thus may exclude the possible role of astrocytes.

This study was designed to explore whether astrocyte activity is responsible for the neuroprotective effects of QTP on aging GABAergic neurons. Consistent with a previous study (Albayram et al. 2011), we found a significant reduction of GABAergic neurons in the cortex of mice at 12- and 24-month-old compared to 9-month-old mice. Chronic treatment with QTP effectively increased the number of GABAergic neurons and decreased the coincidental anxiety-like behavior in aging mice. Using cell culture models, we found that the neuroprotective effects of QTP might be initiated by QTP-stimulated ATP release from astrocytes instead of a direct interaction with GABAergic neurons. These results suggest a critical role for astrocytes in QTP-induced neuroprotection of GABAergic neurons.

Materials and methods

Animals and drugs

Female C57BL mice were purchased from Charles River Laboratories Inc. (St. Constant, QC, Canada). The mice were group housed and maintained on a 12-h light/12-h dark cycle with food and water for a 1 week of acclimation period. All mice were treated according to the guidelines established by the Canadian Council on Animal Care and all procedures were approved by the Animal Care Committee of the University of Manitoba. QTP was obtained from AstraZeneca Pharmaceuticals (Macclesfield, UK) and dissolved in distilled drinking water. The mice began treatment with QTP at the age of 2 months and were treated until the age of 12 months. Three different doses of QTP were used: 0, 2.5, or 5 mg/kg/day.

Open-field test

The open-field test was performed in a square box (36” × 36”) made of compressed wood and painted gray. The test procedure was the same as previously described (He et al. 2005). Briefly, each mouse was placed in a particular corner of the apparatus. The total distance travelled and time spent in the inner border was measured in a 5-min session. The behaviors of the mice were recorded using a camcorder mounted above the open field. The data were analyzed by a blind observer.

Elevated plus maze test

The elevated plus maze consisted two open and two closed arms (50 × 10 cm). Closed arms had 45 cm high walls at the sides and end. The entire maze was elevated to a height of 50 cm above the floor. Each mouse was placed in the central square (10 × 10 cm) facing an open arm and allowed to explore the maze for 5 min (He et al. 2005). One valid entry into any of the four arms was a measure as all four paws of a mouse crossed from the central region into an arm, and the number of total open arm entries and the amount of time spent in the open arms were recorded. The amount of time spent in open arms was measured for 5 min.

Light/Dark transition test

Light/dark transition test was conducted as previously described (Yan et al. 2007b). The light box was open at the top and painted white, and the dark box was painted black and had a removable black lid. Initially, the shuttle door was closed and the mouse was placed into the dark box. The lid was replaced and after 1 min, the shuttle door was opened. First latency to enter the light side and total number of transitions between two chambers was measured for 5 min.

Primary cultures of GABAergic neurons and astrocytes

C57BL mouse pups were obtained from Charles River. Cortical astrocytes were prepared from 1-day-old C57BL mouse pups according to Yu (Yu et al. 1986), with minor modifications. All cultures were incubated in a CO2 incubator at 37°C with 95% air/5% CO2 (vol/vol) and 95% humidity. Cultured astrocytes were used at 4 weeks. Cells in the culture were over 95% positive to glial fibrillary acidic protein immunostaining. Cortical GABAergic neurons were prepared from embryonic day 16 C57BL mice as previously described (Yu et al. 1986; Li et al. 2002). Cytosine arabinoside (40 μM) (Sigma, MO, USA) was added on day 3 to inhibit the growth of the astrocytes. GABAergic neurons were identified with GAD65/67 staining in cultures and showed more than 95% positive (data not shown). The cultures were used for QTP treatment on Day 7 when they became mature as previously described (Yu et al. 1986).

Preparation of astrocyte condition medium

To generate astrocyte condition medium (ACM), 4-week-old astrocyte cultures were incubated with QTP (0.1, 1 and 10 μM) in serum-free Dulbecco's modified Eagle's medium (DMEM). The ACM was collected 24 h after treatment with QTP (ACMQTP). Control ACM was collected from astrocyte cultures that were incubated in serum-free DMEM.

Cell viability assay

Cell viability was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Cells were plated in 96-well plates at a density of 1.0–1.5 × 105 cells per well. At the endpoint of each treatment, MTT solution was added into the culture media at a final concentration of 10% and incubated for 4 h at 37°C. The MTT crystals and the optical density were measured at 570 nm against the background (670 nm). Cell viability was expressed as percentage over the respective control.

Hochest staining

Hochest staining was performed at 22°C. Before staining, cell cultures were washed twice with phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde for 30 min. After washing twice with PBS, the cells were permeabilized with 0.2% Triton X-100 for 30 min and then incubated for 5 min in Hochest 33258 to stain nuclei. Micrographs were taken with a fluorescence microscope (Leica, Solms, Germany).

Immunohistochemistry

Immunohistochemistry was stained with the ABC Peroxidase Staining Kit (Pierce, Rockford, IL, USA) in accordance with the manufacturer's instructions. Briefly, endogenous peroxidase activity was blocked by 3% hydrogen peroxide in PBS for 20 min. Sections were blocked at 22°C for 30 min in 5% serum and 0.3% Triton X-100 in PBS, and incubated overnight at 4°C with series of primary antibodies: mouse anti-GAD65/67 (1 : 8000; Millipore Corporate, Billerica, MA, USA). Secondary biotinylated antibody was used at a dilution of 1 : 1000 (Vector Laboratories, Burlingame, CA, USA). After washing three times with phosphate buffered saline with Tween-20, the slides were incubated with ABC reagents for 30 min. The chromogen was diaminobenzidine (Thermo Scientific, Waltham, MA, USA). Slides were then air dried in the dark, mounted, and viewed with an Olympus BH-2 microscope equipped for bright and dark fields (Olympus, Tokyo, Japan).

Neuronal cell counting

GAD65/67-positive neuron counting in the cortex was performed at a magnification of 10 × on 30 μm coronal sections 180 μm apart (four sections per mouse). The counts for the four sections were averaged and used as a total value for each mouse. All counts were performed by people who were blind to the experimental design.

Western blot analysis

Total protein samples were boiled and resolved on 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) mini-gels under reducing conditions. They were then electrophoretically transferred onto nitrocellulose membranes. Membranes were blocked with 5% (w/v) non-fat dried milk in tris-buffered saline with Tween-20 (TBST) buffer and were probed with polyclonal rabbit antibody to GAD65/67 (1 : 1000; Millipore Corporate) or monoclonal mouse ATP synthase subunit β (1 : 1000; MitoSciences, Eugene, OR, USA) in TBST milk overnight at 4°C. The membranes were also probed with mouse monoclonal antibody to β-actin as a loading control (1 : 5000; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). After incubation with the secondary antibodies for 2 h at 22°C, antigens were revealed by chemiluminescence reaction (Amersham Biosciences, Piscataway, NJ, USA). Quantitative results were expressed as a ratio of target protein to β-actin.

Intracelluar ATP determination

A firefly luciferase-based ATP assay kit (Invitrogen, Carlsbad, CA, USA) was used to measure the ATP levels in permeabilized cells (Wang et al. 2011). Cultured astrocytes were treated with or without QTP for 24 h in 24-well plates at different concentrations. After the culture medium was removed, ice-cold solution containing 10% trichloroacetic acid and 2 mM EDTA was used to extract intracellular ATP (Stanley 1986). The extracted ATP sample was diluted 1 : 40 with ice-cold sterile 0.1 M Tris-acetate buffer with 2 mM EDTA at pH 7.75 to prevent trichloroacetic acid-induced inhibition of firefly luciferase. Standard ATP solution (10 μL) or sample was mixed with 100 μL of standard reaction solution at 20°C for 4 min in a sterile non-transparent white 96-well microplate (Fisher Scientific, Pittsburgh, PA, USA). The bioluminescence was read with a microplate reader. The amount of ATP in each sample was calculated from the ATP standard curve.

Rhodamine 123 assay

To investigate the mitochondrial compartment of neurons during the spontaneous cell death, Rhodamine 123 was used to measure mitochondrial membrane potential (MMP) as previously described (Wang et al. 2011). Cultured neurons were incubated for 30 min in a 24-well plate with 2 μM Rhodamine123 at 37°C. After the medium was removed, cultures were washed with cold Hank's balanced salt solution three times and then lysed with 500 μL of radioimmunoprecipitation assay solution (150 mM sodium chloride, 1.0% Triton X-100, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate and 50 mM Tris at pH 8.0). Cell lysates were transferred into a non-transparent black 96-well plate. Fluorescence was read with a microplate reader with the parameters of 480 nm for excitation and 530 nm for emission.

Statistical analysis

All data were represented as mean ± SEM (*< 0.05, **< 0.01, ***< 0.001). Comparisons between groups were performed using analysis of variance (one-way anova) for repeated measurements followed by Newman–Keuls multiple comparison test or Student's t-test using PRISM 4.0 software (GraphPad Software, Inc, La Jolla, CA, USA).

Results

QTP attenuates anxiety-like behaviors and improved the number of GABAergic neurons in aging mice

Clinical data have demonstrated that the prevalence of anxiety disorders changes with age (Joshi and Pratico 2011). To investigate whether aging mice showed an increase in anxiety-like behaviors, an open-field test was conducted. As shown in Fig. 1a, a significant decrease in the time spent in center area of the open field was observed in 12- and 22-month-old mice compared to younger mice (9 months old). Correspondingly, an increased total travel distance was recorded in the older mice (Fig. 1b). These results indicate a higher anxiety level in older mice.

Figure 1.

Quetiapine (QTP) attenuated anxiety-like behaviors in aging mice. (a) Total time spent in center area was reduced in 12- and 22-month-old mice in open-field test. (b) Total travel distance was increased in 12- and 22-month-old mice in open-field test. (c) Quantitative analysis revealed a decrease in the number of GAD65/67-positive cells in the cortex of 12- and 22-month-old mice compared to 9-month-old. (d) QTP improved the total time spent in center area of 12-month-old mice in open-field test. (e) QTP had no effects on the total travel distance of 12-month-old mice in open-field test. (f) QTP improved the total time spent in open arms of 12-month-old mice in elevated plus maze test. (g) QTP reduced the latency time for first entrance to light box of 12-month-old mice in light/dark transition test. (h) QTP improved the total number of transition between dark and light boxes of 12-month-old mice in light/dark transition test. Data are expressed as means ± SEM. n = 5–10. *< 0.05 versus 9 months old or Cont.; #< 0.05 versus 2.5 mg.

Next, we asked whether QTP was able to prevent the anxiety-like behavior observed in mice as they aged. QTP was administrated to the mice at 5 mg/kg/day in drinking water. We found 12-month-old mice treated with 5 mg/kg/day QTP showed a significant increase in time spent in the center area of the open-field test, although no significant changes were observed in distance travelled (Fig. 1d and e). The reduction of anxiety in the older mice was further confirmed by the elevated plus maze test and light/dark transition test. QTP increased the total time mice spent in the open arms of the elevated plus maze (Fig. 1f). In light/dark transition test, the latency for entering the light area was notably reduced in the QTP treated group (Fig. 1g); and the number of transitions between the light and dark box was also significantly increased in the treatment group (Fig. 1h).

We then tested whether the GABA system was involved in pathogenesis and treatment. GAD is the key enzyme that catalyzes the decarboxylation of glutamate to GABA and has been widely used as a specific marker for GABAergic neurons (Huang et al. 1990). We used GAD65/67 antibody to detect the immunoreactivity of the protein in the mouse brain. We found 12- and 22-month-old mice had a dramatic decrease in the number of GAD positive cells in their cortex (Fig. 1c). Importantly, after 10 months of chronic treatment with QTP, an increased density of GAD65/67 immunoreactivity was observed in the cortex of 12-month- old mice compared to their counterparts in the control group (Fig. 2a). Quantification analysis revealed QTP significantly up-regulated the number of GABAergic neurons in the cortex (Fig. 2b).

Figure 2.

Quetiapine (QTP) improved cell number of GABAergic neurons in brain cortex of aging mice and a validated in vitro model of aging-induced GABAergic neuron death. (a) Images showing GAD65/67-positive cells in brain sections of mice of 12 months old. Those in the bottom panel are magnified from the marked areas in the images of the upper panel. (b) Bar graphs represented the average total number of GAD65/67-positive cells per mm2. (c) A representative phase-contrast image showing the spontaneous cell death of cultured neurons when they reach 10 DIV. (d) Bar graphs represented the relative cell viability in 7 and 10 DIV in 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) test. (e) A representative nuclei staining with Hochest 33258 image showing the apoptotic-like cell death with condensed cell nuclei. The bottom panel is magnified image for the shrinking nuclei. (f) Bar graphs represented the percentage of apoptotic cell in cultures of 7 and 10 DIV. Scale bars represent 500 μm (a) or 40 μm (c and e). Data are expressed as means ± SEM. n = 5–6. *< 0.05 versus Cont. or 7 day.

Spontaneous cell death in cultured cerebral cortical GABAergic neurons

Cultured cortical GABAergic neurons incubated with DMEM and fetal bovine serum (FBS) undergo spontaneous degeneration as they age (Chen et al. 2005a). We employed this model to mimic aging-induced neuronal cell death in vitro. As shown in Fig. 2c, a noticeable morphological change in the cultured neurons was observed using phase-contrast microscopy. After 10 days in vitro (DIV), most of the neurons shrunk and became condensed. A cell viability study with MTT showed the survival ratio of neurons at 10 DIV was reduced compared to the survival at 7 DIV (Fig. 2d). Cortical neurons that underwent spontaneous cell death showed highly condensed nuclei (arrow) (Fig. 2e), a hallmark of apoptosis (Chen et al. 2005b). The percentage of apoptotic cells to total cells was quantified by counting the number of cells in nine random fields (Chen et al. 2005b). Statistical analysis showed that there was a 4-fold increase in apoptotic cells in 10 DIV cultures compared to 7 DIV cultures (Fig. 2f).

ACMQTP not QTP prolongs the survival of GABAergic neurons in primary cultures

Cytotoxicity of QTP was measured with MTT assay in primary cultures of astrocytes and neurons. QTP at the concentrations of 0.01, 0.1, 1, and 10 μM had no toxic effect on cultured 4-week-old astrocytes (Fig. 3a) and 7-day-old neurons (Fig. 3b) after 24 h of incubation. Western blot analysis showed 1 μM QTP treatment for 24 h induced a significant increase of GAD65/67 expression in the 7-day-old neurons (Fig. 3c, d), suggesting a beneficial influence at this concentration. Conversely, QTP at 1 μM did not protect neurons against aging-induced cell death after being added to the medium on the beginning of Day 8. On Day 10, after 3-day treatment, QTP failed to prevent the reduced cell viability in the spontaneous cell death model (Fig. 3f). Neither 0.1 nor 100 μM was found to be effective in subsequent studies (Fig. 3g). However, ACM collected from astrocyte cultures that were pre-incubated with 1 μM QTP for 24 h (ACMQTP), significantly improved the survival of neurons on Day 10 after 3 days of treatment (Fig. 3f). ACM without QTP had no effect on cell viability (Fig. 3f). To exclude the role of remaining degradation product of QTP, fresh medium which was pre-incubated with 1 μM QTP for 24 h was added to incubate the cultures as another control (data not shown). Morphological examination with phase-contrast microscopy showed a higher density of live cells in ACMQTP treated cultures than the QTP treated ones (Fig. 3e).

Figure 3.

ACMQTP not quetiapine (QTP) prolonged the survival of GABAergic neurons in primary cultures. (a) Cell viability was measured with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay in cultured astrocytes after treatment with QTP (0.01, 0.1, 1, 10 μm). (b) Cell viability was measured with MTT assay in cultured neurons after treatment with QTP (0.01, 0.1, 1, and 10 μm). (c) A representative western blot image showing the immunoreactive complex of GAD65/67 protein in cultured neurons with and without 1 μM QTP incubation. (d) The relative amounts of GAD65/67 protein in all groups are presented. (e) Representative phase-contrast images of GABAergic neurons at 7 and 10 DIV with QTP or ACMQTP treatment. (f) Graph bar represented the relative cell viability of GABAergic neurons in cultures. (g) QTP did not protect GABAegic neurons from spontaneous cell death in the cultures. Cell viability was measured with MTT assay in cultured neurons without or with QTP (0.01, 0.1, 1 and 10 μM). (h) ACMQTP not QTP improved mitochondrial membrane potential (MMP) of cultured neurons at 10 DIV. Relative rhodamine 123 uptake was measured in neurons with or without treatment. Scale bar represents 40 μm. Data are expressed as means ± SEM. n = 4–6, *< 0.05 versus Cont. or; #< 0.05 versus QTP.

Mitochondria are cellular organelles which are responsible for regulating metabolism and cell death pathways (Acton et al. 2004). It is widely acknowledged that loss of MMP is an important mechanism of mitochondria-mediated cell apoptosis (Wang et al. 2011). We asked if MMP was a possible mechanism by which the astrocyte-dependent neuroprotection was fulfilled. Rhodamine 123 (RH-123) was used to detect the MMP in 10 DIV GABAergic neurons with or without treatment (Fig. 3h). The ACMQTP treatment for 24 h significantly increased the RH-123 uptake in 10 DIV neurons, which indicates an improved MMP. However, this phenomenon was not observed in QTP-treated cultures.

QTP up-regulated the synthesis of ATP in astrocytes and down-regulated the protein level of ATP synthase in vitro and in vivo

Astrocytes represent an important source of ATP release in the CNS (Lee et al. 2013). It has been well established that ATP has beneficial effects in terms of ameliorating damage after CNS injury (Jia et al. 2011). In light of this finding, we proposed that ATP might be one of the candidates for the unidentified QTP-induced neuroprotective substances generated from astrocytes. To test the possibility, the content of intracellular ATP in cultured astrocytes with or without QTP treatment was measured. As shown in Fig. 4a, cultures treated with 1 μM QTP and incubated for 24 h showed increased intracellular levels of ATP. In addition, QTP inhibited the expression of the β-subunit of ATP synthase, which is one of the key enzymes for ATP synthesis (Fig. 4b). The inhibitory effect of QTP on β-subunit of ATP synthase was also observed in the mouse study with 5 mg QTP (Fig. 4c).

Figure 4.

Quetiapine (QTP) up-regulated the synthesis of ATP in astrocytes and down-regulated the protein level of ATP synthase in vitro and in vivo. (a) 1 μM QTP significantly increased ATP content in cultured astrocytes. (b) 1 μM QTP inhibited the protein level of ATP synthase β-subunit (AtpB) in cultured astrocytes. (c) 5 mg QTP inhibited the protein level of ATP synthase β-subunit (AtpB) in mice of 12 months old. Data are expressed as means ± SEM. n = 4–5, *< 0.05 versus Cont.

Discussion

Recent evidence from preclinical studies has shown that QTP has some neuroprotective capabilities, but the underlying mechanism remains unclear. In this study, we demonstrated that QTP treatment improved the density of GABAergic neurons and attenuated anxiety-like behaviors in aging mice. In an in vitro spontaneous cell death model, we found that QTP was not able to directly protect GABAergic neurons from aging-induced cell death. However, astrocyte-conditioned medium that was pre-incubated with QTP for 24 h effectively improved mitochondrial membrane potential and prolonged the survival of GABAergic neurons in culture. In addition, we found that QTP induced a notable increase in intracellular ATP content in cultured astrocytes, which might be the mechanism involved in the interplay between astrocytes and GABAergic neurons after QTP exposure.

Loss of GABAergic neurons is one of the most common features observed in aging animals and humans, and is believed to contribute to the development of abnormal behaviors related to advancing age (Dugan et al. 2009). The inhibitory function of GABAergic neurons in the cortex plays an important role in the pathophysiology of anxiety disorders (Quide et al. 2012). Disruption of inhibitory input in the cortex has been found to cause an anxiety-like phenotype in animals (Tan et al. 2012). A recent report suggested that GABA-A receptors mediated anxiety-like behavior in open field and elevated plus maze tests in a corticosterone-induced anxiety animal model (Skorzewska et al. 2014). In a chronic restraint stress model, the reduced function of hippocampal and cortical GABAergic neurotransmission in rats was found to have an important influence on the behavioral performance in open field and elevated plus maze tests (Wislowska-Stanek et al. 2013). Therefore, the GABA system could be an important therapeutic target for anxiety disorders. A significant loss of GABAergic cells was found in 12-month-old mice in our study (Fig. 1c). The cell loss seemed to reach a plateau at 12 months of age since there was no significant difference between 12- and 22-month-old mice. Importantly, QTP treatment effectively prevented cell loss in these aging mice (Fig. 2a). To our knowledge, this study is the first to demonstrate that QTP has the ability to increase the cell number of GABAergic neurons. The significant increase of GABAergic neurons in mouse cortex may account for the attenuate of anxiety-like behavioral performance after QTP treatment. This finding is consistent with a previous study that showed a reduction of GABAergic cortical neurons is responsible for the anxiety-like phenotype in mice (Tan et al. 2012). Although many studies have explored the possible mechanisms underlying the neuroprotective abilities of QTP, they mainly emphasized the direct neuronal effects. No effort has been made to examine the potential involvement of astrocytes in the process. As the most numerous cells in CNS, astrocytes are now recognized as integral part of ‘tripartite synapse’, and are tightly associated with memory, emotion, and other complex behaviors. Astrocytes have also been identified as an important target for the treatment of many mental disorders (Hamilton and Attwell 2010). Studies have also found that astrocytes extensively express 5-HT2A receptors, with which QTP has higher affinity for than D2 receptors (Maxishima et al. 2001; Tasman 2008). Here, we found astrocytes mediated the protection of QTP treatment on aging-induced GABAergic neuronal death. This finding seemed to be inconsistent with previous studies that implied direct protective effects of QTP in vitro (Qing et al. 2003; Wang et al. 2005). However, we noticed in all those studies, additional reagents were applied to induce significant cell death before QTP was added in. This ‘additive-containing’ model may make the reaction system quite complicated, and some effects observed might be the result of interactions between the added drugs. Primary neuronal cultures incubated with DMEM and FBS are considered as mature on Day 7 and thereafter start to degenerate (Chen et al. 2005a). This type of culture system provides a reliable way to mimic the process of aging-induced cell death in vitro. As a component of inhibitory circuits in the CNS, GABAergic neurons in culture may respond differently to a reduction in excitatory neurotransmitters resulting from the aging process. Additional studies are warranted in the future to elucidate how the spontaneous cell death occurs in these cultured neurons.

MMP is believed to be a critical factor for determining the fate of neurons (Li et al. 2013). We found that ACMQTP improved the MMP of neurons, while prolonging their life span. It is well established that dysfunctional MMP is closely associated with ATP depletion (Wu et al. 1990). Astrocytes release ATP as a ‘gliotransmitter’ to communicate with neurons by purinergic receptors which are extensively expressed in the latter and mediate intercellular communication between these two types of cells (Tozaki-Saitoh et al. 2011). ATP exerts neuroprotective effects on neurons by activation of purinergic receptors (Fujita et al. 2009). We postulated that ATP might be one of the substances released by astrocytes responsible for mediating neuroprotection in this study. As expected, we found that QTP at the doses of 1 and 10 μM induced obvious increase of ATP content in cultured astrocytes; however, 0.1 μM had no effects on the ATP level in astrocytes (Fig. 4a). Interestingly, the conditioned medium from both the higher doses of QTP-treated astrocytes significantly protected GABAergic neurons against aging-induced cell death, however, the beneficial effects was not seen for the 0.1 μM treated conditioned medium. These results suggested that ATP might be the key factor that released from astrocytes after QTP stimulation and mediated its neuroprotective effects of QTP.

As a neurotransmitter, ATP is an unstable molecule that is easily hydrolyzed to ADP and phosphate. The accumulating ADP has a strong inhibitory influence on ATP synthases (Feniouk and Junge 2005). We detected a significant decrease in the expression of ATP synthase β-subunit in astrocyte cultures and animals after QTP treatment (Fig. 4b and c).

In conclusion, for the first time we identified an astrocyte-dependent neuroprotective mechanism for QTP. QTP produces a significant increase in ATP products in astrocytes. The increased ATP content in astrocytes might play a role in QTP-induced neuroprotection of GABAergic neurons when released into the extracellular space. In addition to offering a new possible mechanism of action for QTP, this study also puts forth an important role for astrocytes in the pathogenesis of anxiety.

Acknowledgments and conflict of interest disclosure

We thank Nan Wu and Monica Novotny for their helpful comments during the preparation of this manuscript. This research was supported by grants from the Manitoba Health Research Council Foundation, the Canadian Institutes of Health Research Foundation, and the Winnipeg Health Science Centre Foundation. X.M.L has received research grants from AstraZeneca Canada, Inc, and Pfizer Canada, Inc.

All experiments were conducted in compliance with the ARRIVE guidelines. No conflicts of interest exist for any of the other authors.

Ancillary