Vitamin D and 17β-estradiol upregulate each other's receptors and regulating the AMPK/NF-κB pathway to relieve depressive-like behaviors in female ovariectomized rats

Background: A deciency of vitamin D (VD) or 17β-estradiol (E2) is associated with increased risk of mood disorders such as depression in menopausal females, but the mechanism underlying is still elusive. The present study aims to evaluate whether vitamin D and 17β-estradiol could relieve a depressive-like state through neuroinammatory regulation in ovariectomized (OVX) rats. Methods: Female SD rats were randomly divided into four groups, namely, control (SHAM), OVX, OVX+VD, and OVX+E2. The treatment procedure was performed for 10 weeks until sacrice. Results: The chronic administration of vitamin D and 17β-estradiol showed anti-depressive-like activity in the OVX rats. Additionally, vitamin D and 17β-estradiol upregulated each other's receptors, including VDR, ERα, and ERβ in the hippocampus of OVX rats. Vitamin D and 17β-estradiol showed neuroprotective effects by decreasing OVX-induced apoptosis and neuronal damage, regulating the AMPK/NF-κB signaling pathway, and reducing the proinammatory cytokines (IL-1β, IL-6, and TNFα), as well as iNOS and COX-2 in the hippocampus of OVX rats. Conclusions: The present study demonstrated that vitamin D and 17β-estradiol could upregulate each other's receptors and regulate the AMPK/NF-κB pathway to relieve the OVX-induced depressive-like state. The results should stimulate translational research towards the vitamin D potential for prevention or treatment of menopause-related depression. Collectively, the present study demonstrates that vitamin D and 17β-estradiol exert an anti-depressive-like effect by regulating the AMPK/NF-κB pathway, and decreasing proinammatory cytokines (IL–1β, IL–6, and TNFα), as well as iNOS and COX–2 in the hippocampus of OVX rats. Additionally, vitamin D and 17β-estradiol could upregulate each other’s receptors in the hippocampus of OVX rats, indicating a crosstalk between vitamin D and 17β-estradiol in relieving depressive-like behaviors. These ndings are supporting the evidence that vitamin D and 17β-estradiol might exert an anti-depressive-like function through neuroinammatory regulation. Furthermore, these ndings should stimulate future translational research towards the vitamin D potential for prevention or treatment of menopause-related depression.


Introduction
The prevalence of depression is dramatically increased in menopausal women [1]. Menopausal disorders, including depression, are strictly related to the loss of ovarian function and a chronic hypo-estrogenic state. One strategy to relieve menopausal disorders is to supply estrogens. However, estrogens increase the risk of endometrial hyperplasia or cancer, which hinders their clinical use [2]. Therefore, it becomes urgent to seek an alternative treatment strategy to relieve menopause-related depression. Vitamin D (VD) supplementation seems like a promising strategy. Vitamin D intake is considered to be safe and is recommended to prevent bone loss in menopausal women. One study found that healthy postmenopausal older women supplemented with vitamin D 3 and calcium did not result in a signi cant risk change of all-type cancer [3]. Furthermore, low vitamin D level is noted in depressive patients, and vitamin D supplementation has shown bene cial in improving mood [4,5]. Although vitamin D and 17βestradiol (E2) is likely to be improving mood in menopausal women with depressive disorder, the mechanisms underlying remain equivocal.
Neuroin ammation is one of the most essential contributors to depression. Patients with depression are more likely to have a higher status of proin ammatory cytokines in the periphery, cerebrospinal uid, and hippocampus [6]. Animal studies nd that proin ammatory cytokines could induce depressive-like behaviors, and depressive-like models have elevated levels of proin ammatory cytokines [7]. Ovariectomized (OVX) rodent is a widely used menopausal model [8]. Long-term after OVX, animals not only developed depressive-like behaviors but also showed higher neuroin ammatory levels [9]. Furthermore, estrogen supplementation, exercise, and in ammasome inhibition in the OVX rodents could exert an anti-depressive-like effect by immune-regulating [9,10], strongly suggesting an important role of the immune system in the OVX-induced depressive-like behaviors.
Nuclear factor-kappa B (NF-κB) is a protein complex that plays a crucial role in in ammatory regulation, and NF-κB could be negatively controlled by AMP-activated protein kinase (AMPK) [11]. Under physiological condition, NF-κB bounding with IκB maintains in an inactive state in the cytoplasm. Once activated, NF-κB's inhibitor IκB is phosphorylated and degraded, resulting in p65 phosphorylation and transcriptional regulation of proin ammatory cytokines and neurotoxic mediators, such as interleukin (IL)-1β, IL-6, tumor necrosis factor (TNF)-α, inducible nitric oxide synthase (iNOS), and cyclooxygenase-2 (COX-2) [12] [12]. These proin ammatory cytokines and neurotoxic mediators have been proven to involve in the pathophysiology of depression [13]. Therefore, the present study aims to evaluate whether vitamin D and 17β-estradiol relieve a depressive-like state through neuroin ammatory regulation in the OVX rats.

Animals
Experiments were conducted on female Sprague-Dawley (SD) rats (12-week old). All rats were housed under standard laboratory conditions of temperature (23±2℃), light (12h light/dark cycle), and a relative humidity (55±10%), with free access to food and water. The rats were habituated for 7 days before the experiment. All animal use procedures were conducted by the ethical standards in the 1964 Declaration of Helsinki and its later amendments, with the approval of the Ethics Committee of the Zhongshan A liated Hospital of Zhongshan University.

Drug supplementation and ovariectomy
The rats were randomly assigned into the following four groups (7-8 rats in each group): the shamoperated control (SHAM), the OVX, the OVX+VD, and the OVX+E2 group. The bodyweight of the rats was measured in the experiment. In the OVX group, after full anesthesia with sodium pentobarbital (50 mg/kg, i.p.), the ovaries, oviducts, and top of the fallopian tubes were bilaterally removed through two small incisions. In the SHAM group, similar protocols were conducted without ovariectomy. Since the day of OVX surgery, vitamin D (calcitriol, 100 ng/kg, Roche, China) and 17β-estradiol (30 μg/kg, Macklin, Shanghai, China) had been daily gavaged in the OVX+VD and OVX+E2 group, respectively. Calcitriol at a dose of 100ng/kg has been proven to enhance the VDR protein level without affecting serum calcium and phosphate status in our previous study [14]. The treatment procedure was performed for 10 weeks until sacri ce. The animals in the SHAM group were sacri ced at the diestrus phase to avoid the effects of the estrus cycle. Ten weeks later, the forced swim test (FST) and novelty-suppressed feeding test (NSFT) were carried out. Twenty-four hours after the behavioral tests, the rats were anesthetized with a peritoneal injection of sodium pentobarbital (50mg/kg). Serum was separated by centrifuging at 4000 rpm for 10min at 4 °C. The uterine were collected and weighted. Hippocampus samples were thoroughly washed with cold physiological saline. All the samples were then stored at −80 °C until analysis.

Behavioral tests
The FST was conducted according to a classic paradigm with minor modi cations to measure behavioral despair in stressful situations [15]. The FST contained two trials. In the rst trial, the rats were individually forced to swim for 15 min in a Plexiglas cylinder (45cm height, 25cm diameter) containing approximately 35cm of water (24±1℃). The rats were dried and removed to their home cages. In the second trial, the rats were placed again in the cylinders for 5 min 24 h later, and this test session was videotaped.
Immobility is de ned as oating motionless or making slight movements to keep its head just above water. The duration of immobility was measured by an experienced observer who was blinded to the experimental design.
The NSFT was adapted from a previous study [16]. Brie y, the rats were deprived food for 24 h before the test. The rats were then placed in an open eld (75×75×40 cm), where the rats were free to explore for 8min. There is a small amount of food placed on a piece of white paper (10×10 cm) in the center of the eld. The latency to feed is de ned as the duration between placing the rats in the open eld and they take the rst bite of the food. After the test, the rats were immediately sent back to their home cage, and the total food consumption for the next 5 min was weighed to avoid the influence of the rats' appetite.

Hormone assay
Concentrations of 17β-estradiol in the serum were measured using an enzyme-linked immunosorbent assay (ELISA) kit (Guxi Biotech, China) following the manufacturer's protocol.

Quantitative real-time PCR
Total RNA was isolated from the hippocampal homogenates using Trizol reagent (Invitrogen, USA). The mRNA expression of calbindin-d28k (cabp-d28k), calbindin-d9k (cabp-d9k), IL-1β, IL-6, TNF-α, iNOS, and Cox2 was detected. Their oligonucleotide primers speci c for rats are listed in table 1. Quantitative RT-PCR was performed on a Bio-Rad Cx96 Detection System (Bio-Rad, USA) using a SYBR green PCR kit (Applied Bio-systems, USA). The 5 ng cDNA samples received 40 cycles of ampli cation. Each cDNA was examined in triplicate. Relative quantitation for PCR product was normalized to β-actin as an internal standard.

Apoptosis Analysis
Apoptosis examination was carried out using a commercially kit (Keygen Biotech, Nanjing, China). The hippocampal samples were xed in 10% phosphate-buffered paraformaldehyde, embedded in para n, and then cut into 6 mm thick sections. The apoptosis in the hippocampal CA1 region was detected using the Tunel method, which checks the fragmentation of DNA in the nucleus during apoptotic cell death in situ.

Nissl's staining
Nissl's staining could detect the Nissl body in cytoplasm and dendrites of neurons, and this stain has been widely used to identify neuronal damage. The Nissl's staining was carried out according to a conventional method [17]. The hippocampal CA1 region was carefully observed in three randomly selected fields of view by an optical microscope (Leica DFC420, Germany) at a magnification of 200×.
Image J software was used to count the stained cells. According to a previous method [18], the average number of viable neurons per mm 2 was calculated.

Immuno uorescent staining
Para n-embedded sections of the hippocampus were dewaxed in xylol, rehydrated, and rinsed in PBS.
Immuno uorescent images were taken with an inverted uorescence microscope (Olympus, Japan), and images were then analyzed using the Image J software to obtain the mean uorescence density of each visual eld in the hippocampal CA1 region.

Statistical Analysis
Statistical procedures were conducted in SPSS version 18 software. Data were shown as the means ± SD.
Differences between groups were determined by one-way ANOVA. A prior level of signi cance was established at P < 0.05. Post hoc comparisons were made using the Tukey test.
Nonetheless, both vitamin D and 17β-estradiol alleviated the OVX-induced depressive-like behaviors with decreased immobility time (P < 0.05 and P < 0.01, respectively, Fig. 1e) in the FST and decreased latency time (P < 0.01, Fig. 1f) in the NSFT in the OVX rats.

Effects of vitamin D and 17β-estradiol on the AMPK/NF-κB signalingpathway in the hippocampus of OVX rats
To further con rm the immune-regulatory effect of vitamin D and 17β-estradiol, the AMPK/NF-κB signaling pathway was assessed. The results showed that OVX decreased the p-AMPK/AMPK ratio (P < 0.01, Fig. 4a), decreased the IκB protein expression (P < 0.01, Fig. 4b), and increased the NF-κB protein level (P < 0.01, Fig. 4c) in comparison to the SHAM group. The supplementation of vitamin D and 17βestradiol enhanced the p-AMPK/AMPK ratio (P < 0.01, Fig. 4a), suppressed the NF-κB protein level (P < 0.01, Fig. 4c), and increased the IκB protein expression (P < 0.05 and P < 0.01, respectively, Fig. 4b) in the OVX group.

Discussion
In this study, the chronic administration of vitamin D and 17β-estradiol showed anti-depressive-like activity in the OVX rats. Additionally, we provide evidence that vitamin D and 17β-estradiol could upregulate each other's receptors, including VDR, ERα, and ERβ in the hippocampus of OVX rats. Vitamin D and 17β-estradiol showed neuroprotective effects by decreasing OVX-induced apoptosis and neuronal damage, regulating the AMPK/NF-κB signaling pathway, and decreasing proin ammatory cytokines (IL-1β, IL-6, and TNFα), as well as iNOS and COX-2 in the hippocampus of OVX rats. These changes might contribute to, at least in part, the anti-depressive-like activity of vitamin D and 17β-estradiol.
We found that chronic OVX induced depressive-like behaviors with decreased immobility time in the FST and decreased latency time in the NSFT, in line with a previous study that OVX developed depressive-like behaviors at more than 6 weeks [19]. The OVX model was successfully built, as evidenced by the increased bodyweight, decreased uterine weight, and decreased estrogen level. These basic characteristics were recovered by 17β-estradiol but not vitamin D, although they both showed an antidepressive-like effect. The results indicate that the anti-depressive-like effect of vitamin D is estrogenindependent.
The most striking nding in this study is that vitamin D and 17β-estradiol could upregulate each other's receptors in the hippocampus of OVX rats, indicating a crosstalk between vitamin D and 17β-estradiol in relieving depressive-like behaviors. A similar nding was reported in marrow stromal cells that the treatment of vitamin D and 17β-estradiol upregulated each other's receptors (VDR and ERα) [20]. Vitamin D has numerous biological functions and acts through its receptor (VDR) in most cells. The co-elevated expression of VDR and calcium-binding protein (cabp-d28k and cabp-d9k) suggests that the vitamin D signaling system was ampli ed by vitamin D and 17β-estradiol in the hippocampus of OVX rats. ERs are essential for memory consolidation [21]. In this study, we found that OVX caused an abrupt decrease of ERβ but not ERα. A similar nding was reported in a depressive rodent model that chronic unpredictable mild stress decreased ERβ protein level but not ERα in the hippocampus of mice [22]. Furthermore, OVXinduced depressive-like behaviors could be reversed by administration of 17β-estradiol and ERβ agonist but not ERα agonist in mice [9]. The results together indicate that the decreased ERβ level might be responsible for depressive-like behaviors in the rodents.
In morphology, OVX caused signi cantly neuronal apoptosis, damage, and neuroimmune overactivation, as evidenced by the fact that the Tunel-positive cells remarkably increased, viable neurons declined, and Iba-1 immuno uorescent enhanced, respectively. Vitamin D and 17β-estradiol ameliorated the OVXinduced neuronal apoptosis, damage, and neuroimmune overactivation in the OVX+VD and OVX+E2 group compared to the OVX group. The results provide visual evidence that vitamin D and 17β-estradiol have neuroprotective effects.
Neuroin ammation affects every pathological aspects of depression, including neuronal apoptosis, neurogenesis, neurotransmission, and neuroplasticity, playing a key role in the onset and development of depression [23]. Hippocampal in ammation has been widely reported in patients with depression and animal models of depression [24]. NF-κB serves as a dominated factor in regulating in ammation [25]. We found that the NF-κB was activated in the OVX group compared to the SHAM group. As expected, vitamin D and 17β-estradiol effectively re-balanced the AMPK/NF-κB signaling pathway. The results highlight that vitamin D and 17β-estradiol might exert an anti-in ammatory effect through the AMPK/NF-κB signaling pathway at the state of chronic hypoestrogenism.
Our study correlates with the results from previous literature that long-term OVX could induce a neuroimmune-activated state with increased expression of proin ammatory cytokines [26]. Furthermore, the current study demonstrates that Vitamin D and 17β-estradiol decreased the gene levels of proin ammatory cytokines (IL-1β, IL-6, and TNFα) in the hippocampus of OVX SD rats. Another study found that vitamin D only partially reversed the increase of IL-6 protein level in the hippocampus of OVX Wistar rats [27]. Although IL-1β, IL-6, and TNFα are both proin ammatory cytokines, they might respond diversely in depression. A meta-analysis found profoundly higher levels of TNF-α and IL-6 in the serum of patients with depression but not IL-1β [28]. Intriguingly, another meta-analysis reported that antidepressants decreased IL-1β level and possibly IL-6 but not TNFa [29].
In addition to the proin ammatory cytokines, iNOS and COX-2, which are both the downstream targets of the AMPK/NF-κB pathway, are also involved in the pathophysiology of depression [13]. iNOS is responsible for the nitric oxide (NO) formation in response to in ammatory mediators, and the inhibition of iNOS showed a potential anti-depressive-like effect in depressive models [30]. COX-2 is a key enzyme for the production of a series of in ammatory cytokines, and the inhibition of COX-2 by meloxicam could relieve depressive-like behaviors [31]. In this study, we found that vitamin D and 17β-estradiol signi cantly decreased the OVX-induced overexpression of iNOS and COX-2, which further support the neuroimmuneregulatory effect of vitamin D and 17β-estradiol.

Conclusions
Collectively, the present study demonstrates that vitamin D and 17β-estradiol exert an anti-depressive-like effect by regulating the AMPK/NF-κB pathway, and decreasing proin ammatory cytokines (IL-1β, IL-6, and TNFα), as well as iNOS and COX-2 in the hippocampus of OVX rats. Additionally, vitamin D and 17βestradiol could upregulate each other's receptors in the hippocampus of OVX rats, indicating a crosstalk between vitamin D and 17β-estradiol in relieving depressive-like behaviors. These ndings are supporting the evidence that vitamin D and 17β-estradiol might exert an anti-depressive-like function through neuroin ammatory regulation. Furthermore, these ndings should stimulate future translational research towards the vitamin D potential for prevention or treatment of menopause-related depression.    Representative blots and statistical graphs of VDR (B). Relative mRNA expression of cabp-d28k (C) and cabp-d9k (D). Representative blots (E) and statistical graphs of ERα (F) and ERβ (G). Data are the means ± SD (n=7-8). ** P < 0.01 compared to the SHAM group. + P < 0.05, ++ P < 0.01 compared to the OVX group.

Figure 3
Vitamin D and 17β-estradiol both show a neuroprotective effect in the hippocampus of OVX rats.
Representative images of Tunel staining (A) and Nissl staining (B) in the hippocampal CA1 region.
Statistical graphs of the apoptotic index (C) and the number of viable neurons (D). Representative blots and statistical graphs of cleaved caspase3 (E). Data are the means ± SD (n=7-8). ** P < 0.01 compared to the SHAM group. + P < 0.05, ++ P < 0.01 compared to the OVX group. Figure 4