Rice Germ Ameliorated Chronic Unpredictable Mild Stress-Induced Depressive-like Behavior by Reducing Neuroinflammation

Stress-induced neuroinflammation is widely regarded as one of the primary causes of depression. Gamma-aminobutyric acid (GABA)-enriched foods relieve stress and reduce inflammatory reactions. This study aimed to evaluate whether rice germ with 30% GABA (RG) reduced neuroinflammation in mice exposed to chronic unpredictable mild stress (CUMS). CUMS mice were administered 40, 90, and 140 mg/kg of RG. CUMS increased serum and hypothalamic pro-inflammatory cytokine (TNF-α and IL-6) levels, which were decreased by RG. In the hypothalamus, CUMS elevated M1-type microglia markers of CD86 and NF-κB, whereas RG lowered these levels. The expression levels of NLRP3 inflammasome complex (NLRP3, apoptosis-associated speck-like protein containing a caspase recruitment domain, and caspase-1), IL-1β, and IL-18 were increased in the hypothalamus of CUMS mice and decreased by RG. RG attenuated depressive-like behaviors in CUMS mice, as measured by the forced swim test and tail suspension test. In conclusion, RG decreased hypothalamic inflammation-related signals, such as TNF-α, IL-6, M1 polarization, NF-κB, NLRP3 inflammasome complex, caspase-1, IL-1β, and IL-18, to diminish depressive-like behavior.


Introduction
Stress causes aberrant regulation or malfunction of the hypothalamic-pituitary-adrenal (HPA) axis, resulting in cognitive impairment, depression, and anxiety [1,2]. Stress-induced neuroinflammation is considered one of the primary causes of depression. The expression levels of pro-inflammatory cytokines, such as tumor necrosis factor alpha (TNF-α), interleukin (IL)-6 and IL-1β have been found to be elevated in the peripheral blood or central nervous system (CNS) of patients suffering from major depressive disorder (MDD) [3,4]. Furthermore, these pro-inflammatory cytokines were found to be elevated in numerous animal models of stress, and increased pro-inflammatory cytokines are associated with depressive-like behaviors [5][6][7][8].

Standard Solution and Sample Preparation
A 0.1 g standard sample was dissolved in 100 mL of DW in a volumetric flask to prepare a standard solution, filtered through a polytetrafluoroethylene syringe filter (25 mm/0.2 µm), and kept at −80 • C. For the 5% aqueous sample solution, 5 g of RG was dissolved in DW in a 100 mL volumetric flask and filtered using a polytetrafluoroethylene syringe filter.

HPLC Analysis Method
GABA content was determined using the method described [36], with minor changes. In our study, a Dionex U3000 series HPLC system (Thermo Fisher, Waltham, MA, USA) equipped with a UV detector was utilized, and the flow rate was lowered to 1 mL/min. The samples were analyzed using UV-Vis spectrophotometry at a wavelength of 338 nm. The amount of GABA in the RG was calculated using the following Equation: Substance (mg/g) = Measurement (mg/mL) × Dilution factor ÷ Amount(g) × 100 (mL)

Animals and CUMS Procedure
C57BL/6 (Male) mice were provided by Orient Bio (Seongnam, Republic of Korea). The animals were bred under standard conditions (23 ± 2 • C, 12 h light/dark cycle, humidity of 50 ± 3%). All animal experiments were conducted in accordance with the ethical principles of the Institutional Animal Care and Use Committee of Gachon University (approval No. LCDI-2021-0170).
After an acclimatization period of 1 week, the CUMS procedure was conducted for 5 weeks. In our study, 24 h of food deprivation, rein for 24 h in an empty water bottle, exposure to foreign object for 24 h and wet bedding for 24 h were a stressor ( Figure S1A).

Animal Experimental Design
The mice were randomly separated into seven groups (n = 6): (1) Control: The group was administered orally the same volume of saline as that administered in the other groups without the CUMS procedure. (2) CUMS: The group was administered orally the same volume of saline as that administered in the CUMS procedure. (3) CUMS/RG40: The group was administered orally RG at 40 mg/kg daily with the CUMS procedure. (4) CUMS/RG90: The group was administered orally RG at 90 mg/kg daily with the CUMS procedure. (5) CUMS/RG140: The group was administered orally RG at 140 mg/kg daily with the CUMS procedure. (6) CUMS/GABA: The group was administered orally GABA at 30 mg/kg daily with the CUMS procedure. After the CUMS procedure for 5 weeks, RG, GABA, and theanine were administered orally at the same time as the CUMS procedure over 4 weeks.
Behavioral tests were performed at the last oral administration, and the mice were sacrificed under respiratory anesthesia with isoflurane ( Figure S1B). Blood and brain samples were collected for this study.

Forced Swimming Test (FST)
FST was performed with slight changes based on what was described in previous study [37,38]. Briefly, before the experiment, an open cylindrical container with a height of 45 cm and a diameter of 25 cm was filed water at 26 ± 1 • C to a depth of 35 cm.
The experimenter holds the mice by the tail and slowly placed the mouse into the container with water. The mice were adapted to the water for 15 min and the main test was performed after 24 h. After filling the open cylindrical container with water under the same conditions as the previous day, the mice were placed in the water and FST was performed for 7 min. During the FST, the experimenter observed while maintaining an appropriate distance, and when the mouse submerged, it was removed from the container. After 7 min of FST, the recording was stopped, and the mice were taken out of the water, gently dried with dry paper, and put back into the cage.
For FST analysis, only the last 5 min of the 7 min-FST were analyzed. This is because most mice were active at the beginning of the FST, so the potential effects of the treatment may be masked during the first 2 min. Immobility time was analyzed using Smart 3.0. program (PanLab Harvard Apparatus, MA, USA).

Tail-Suspension Test (TST)
TST was conducted with minor modifications, referring to the content described in previous study [39]. Briefly, the tail of the mice was hung upside down in a suspension box (45 cm height, 25 cm diameter). After the tail was wrapped with tape, the head of each mouse was approximately 20 cm from the floor. And TST was performed for 6 min. After 6 min of TST, the recording was stopped, and the mice were placed in a cage with the tape on its tail removed.
For TST analysis, last 4 min of the suspension time of 6 min was analyzed. This is due to most mice were active during first 2 min. Immobility time was measured and analyzed using Smart 3.0. program (PanLab Harvard Apparatus).

Enzyme-Linked Immunosorbent Assay
To measure the expression levels of TNF-α and IL-6 in the serum, collected blood was placed in a serum separator tubes (Becton Dickinson, Franklin Lakes, NJ, USA) and incubated at room temperature for 20 min. Thereafter, the blood samples were centrifuged at 2000× g for 20 min, and the supernatant was separated and put to a new tube.
The 96-well microplates coated with a coating buffer (pH 9.6) containing 100 nM sodium carbonate and sodium bicarbonate was incubated with 5% skim milk at room temperature for 2 h to block unnecessary protein binding. After rinsing with phosphate buffered saline containing tween-20 (TPBS), equal amounts (100 µg) of samples were placed in each well and incubated 12 h at 4 • C. And then washed with TPBS and incubated for 12 h at 4 • C with appropriate aliquots of anti-TNF-α and anti-IL-6 antibodies (Table S1), respectively. Thereafter, each well was washed again and incubated for 2 h at room temperature with peroxidase-conjugated antibody. After rinsing with TPBS, a tetramethylbenzidine solution (100 µL/well; Sigma-Aldrich) was added for color development, and incubation was performed at room temperature for 15-20 min while blocking the light. After color development, the same volume of stop solution (100 µL/well; sulfuric acid, 2N) was dispensed, and optical density was analyzed at a wavelength of 450 nm using a Multiskan SkyHigh Microplate Spectrometer (ThermoFisher Scientific, Waltham, MA, USA).

3,3 -Diaminobenzidine (DAB) Staining
The brain tissue of mice was fixed in cold 4% paraformaldehyde (Sigma-Aldrich) for 4 h. Paraffin-embedded tissue blocks were prepared using a tissue processor (Tissue-Tek VIP ® 5 Jr, SAKURA Finetek, Tokyo, Japan) and an embedding machine (Tissue-Tek ® TEC™ 6, SAKURA Finetek) from fixed brain tissue. The paraffin-embedded brain blocks were cut into 5 µm thickness using a microtome (ThermoFisher Scientific, Waltham, MA, USA) and dried at 60 • C for 24 h to adhere well to slides. Before starting DAB staining, paraffin was removed from the slides and the tissues were boiled in the sodium citrate buffer (pH 6.0) using a microwave oven, and then cooled in cold distilled water for antigen retrieval. To prevent non-specific antibody binding, after incubation in normal serum (Vector Laboratories, Burlingame, CA, USA) for 1 h, appropriate concentration of primary antibodies in normal serum (Table S1) was incubated followed by rinsing with PBS.
Then, the primary antibodies-tagged brain tissues were incubated with biotinylated secondary antibodies (Vector Laboratories) and washed with PBS. For brown color development, the slides were reacted with DAB solution activated with H 2 O 2 for 15 min. The stained tissues were mounted with coverslip and DPX solution (Sigma-Aldrich). Images were obtained using a slide scanner (Motic, Kowloon, Hong Kong), and the brown color intensity was analyzed with ImageJ software (NIH, Bethesda, MD, USA).

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)
RNA of the hypothalamus was isolated using RNAiso (Takara, Tokyo, Japan) according to the manufacturer's instructions. The extracted RNA was synthesized with cDNA using a PrimeScript first-strand cDNA Synthesis Kit (Takara). qRT-PCR was performed using cDNA synthesized using the CFX384 TouchTM Real-Time PCR detection system. cDNA (300 ng), SYBR premix (5 µL; Takara), forward primer (0.4 µM), and reverse primer (0.4 µM) were mixed, and the number of threshold cycle were determined using CFX ManagerTM software (BioRad, CA, USA). All primer information is summarized in the Table S2.

Statistical Analysis
Non-parametric tests were conducted in this study. The Kruskal-Wallis test was used to confirm the importance of differences between the seven groups. Multiple comparisons were performed using Mann-Whitney U test if significant difference were found in the Kruskal-Wallis test. Results were expressed as mean ± standard deviation, and statistical significance was accepted as follows: * versus Non-CUMS; $ versus CUMS; # versus CUMS/RG90; † versus CUMS/GABA;ˆversus CUMS/Theanine. Statistical analyses were performed using SPSS version 22 (IBM Corporation, Armonk, NY, USA).

GABA Content in RG
To determine the GABA content in RG, we performed HPLC analysis. The presence of GABA after fermentation was confirmed by comparing the retention times of the standard and RG ( Figure S2 and Table S3). Furthermore, HPLC chromatography showed that the average percent content of GABA in RG was 31.04 ± 0.94%.

RG Decreased Expression Levels of TNF-α and IL-6 in Serum and Hypothalamic of CUMS Mice
First, we evaluated serum IL-6 and TNF-α levels in control, CUMS, CUMS/RG40, CUMS/RG90, CUMS/RG140, CUMS/GABA, and CUMS/Theanine groups. When RG 40 mg/kg was administered to the animal, the GABA content consumed by the animal was 12 mg/kg. When RG 90 and 140 mg/kg were administered to animal, the content of GABA which animal were consumed were 27 mg/kg and 42 mg/kg, respectively.
The serum TNF-α levels were significantly increased by CUMS. It was decreased by oral administration of RG 40 mg/kg (10.8% decreased compared with CUMS group), RG 90 mg/kg (18.5% decreased compared to CUMS group), and 140 mg/kg (21.6% decreased compared to CUMS group), GABA (17.2% decreased compared to CUMS group), and theanine (18.8% decreased compared with CUMS group). The reduction effects of RG (90 and 140 mg/kg), GABA, and theanine were not significantly different ( Figure 1A).
CUMS/RG90, CUMS/RG140, CUMS/GABA, and CUMS/Theanine groups. When RG mg/kg was administered to the animal, the GABA content consumed by the animal w 12 mg/kg. When RG 90 and 140 mg/kg were administered to animal, the content of GA which animal were consumed were 27 mg/kg and 42 mg/kg, respectively.
TNF-α expression in the hypothalamus increased significantly after CUMS treatment. It was decreased by oral administration of RG (40, 90, and 140 mg/kg), GABA, and theanine. The reduction effects of RG (40,90, and 140 mg/kg), GABA, and theanine did not differ significantly ( Figure 1C,D).
CUMS significantly increased the hypothalamic IL-6 expression. It was decreased by the administration of RG (90 and 140 mg/kg), GABA, and theanine. The reduction effects of RG (90 and 140 mg/kg), GABA, and theanine did not differ significantly ( Figure 1C,E).

RG Decreased M1 Polarization and NF-κB Expression
CUMS significantly increased CD86 (M1 marker) expression in the hypothalamus. It was decreased by oral administration of RG (40,90, and 140 mg/kg), GABA, and theanine. This reduction was most prominent with RG at 140 mg/kg (Figure 2A,B). ment. It was decreased by oral administration of RG (40, 90, and 140 mg/kg), GABA, and theanine. The reduction effects of RG (40, 90, and 140 mg/kg), GABA, and theanine did not differ significantly ( Figure 1C,D).
CUMS significantly increased the hypothalamic IL-6 expression. It was decreased by the administration of RG (90 and 140 mg/kg), GABA, and theanine. The reduction effects of RG (90 and 140 mg/kg), GABA, and theanine did not differ significantly ( Figure 1C,E).
CUMS significantly increased the expression of NF-κB in the hypothalamus. It was decreased by oral administration of RG (90 and 140 mg/kg), GABA, and theanine. The effects of RG (90 and 140 mg/kg), GABA, and theanine on NF-κB expression did not differ significantly ( Figure 2D,E).
Caspase-1 expression in the hypothalamus increased significantly following CUMS exposure. It was decreased by oral administration of RG (40,90, and 140 mg/kg), GABA, and theanine. The inhibitory effect of RG at 140 mg/kg was greater than that of theanine ( Figure 3C). CUMS significantly increased the expression level of IL-1β in the hypothalamus. It was decreased by oral administration of RG (40,90, and 140 mg/kg), GABA, and theanine. RG at 140 mg/kg had a greater inhibitory effect than GABA and theanine ( Figure 3D).
CUMS significantly increased the expression of IL-18 in the hypothalamus. It was decreased by oral administration of RG (90 and 140 mg/kg), GABA, and theanine. The effects of RG at 90 and 140 mg/kg, GABA, and theanine on weight loss were not significantly different ( Figure 3E).
CUMS significantly increased the expression of GSDMDs in the hypothalamus. It was decreased by oral administration of RG (40,90, and 140 mg/kg), GABA, and theanine. The reduction effects of RG at 40, 90, and 140 mg/kg, GABA, and theanine were not significantly different ( Figure 3F).
CUMS increased the duration of immobility during the tail suspension test. It was decreased by oral administration of RG (40,90, and 140 mg/kg), GABA, and theanine. The reduction effects of RG at 40, 90, and 140 mg/kg, GABA, and theanine were not significantly different ( Figure 4B).

Discussion
Depression is characterized by various symptoms, including persistent low mood, anhedonia, loss of interest, and feelings of worthlessness [40]. Chronic stress induces mood, cognition, and memory abnormalities and leads to various brain diseases. The effects of stress on the brain have been evaluated using various animal models. Stress causes neuroinflammation and structural and functional alterations in neuronal networks [41,42].
The CUMS animal model, wherein animals are repeatedly exposed to varied unpredictable and uncontrollable stressors for days or weeks, is the most commonly used model [43][44][45][46][47]. CUMS is regarded as one of the most translationally relevant animal models for evaluating the pathophysiology of depression because of its reproducible neurochemical, neuroendocrine, and neuroinflammation outcomes [48,49]. Furthermore, CUMS also in-duces depressive behaviors; thus, it has been frequently used to evaluate the efficacy of antidepressants [49]. Thus, we evaluated the effect of RG on stress-induced depressive behaviors using the CUMS model.
Our study demonstrated that RG reduced expression levels of TNF-α and IL-6 in of serum CUMS mice. Additionally, RG reduced expression levels of TNF-α and IL-6 in the hypothalamic tissue.
Since GABA could not reach the brain directly, RG appeared to reduce pro-inflammatory cytokine release in peripheral blood across the BBB. Reduced levels of blood pro-inflammatory cytokines can attenuate neuroinflammation.
RG decreased hypothalamic M1 polarization and TNF-α and IL-6 levels. M1 polarization is reported to be elevated in MDD patients [50]. Furthermore, interferon-alpha-treated animals showed depressive-like behavior, which was accompanied by an increase in M1type microglia [51].
Microglia are involved in the activation of the NLRP3 inflammasome complex in depression. The NLRP3 inflammasome complex is activated by two signaling pathways [52]. First, priming signals, such as N-κB induce NLRP3 upregulation. Following priming, the activating signal increases the binding between NLRP3 and the remaining inflammasome machinery, such as ASC and pro-caspase-1 [53]. Numerous studies have linked NLRP3 inflammasome complex or pyroptosis to depression or stress-induced depressive behavior [54]. Antidepressants decreased IL-1β and IL-18 levels in serum and suppressed NLRP3 expression in MDD patients and mice with stress-induced depression [55]. Chronic stress induces astrocyte loss in the hippocampus via pyroptosis [56]. Inhibition of the NLRP3 inflammasome complex by caspase-1 inhibitors, purinergic 2X7 receptor antagonists, or genetic deletion of NLRP3 showed a protective effect against stress-induced IL-β elevation [57][58][59]. In our investigation, RG decreased the hypothalamic levels of NLRP3, ASC, caspa-se-1, IL-1β, and IL-18. Moreover, depressive-like behaviors, which were evaluated using the forced swim test and tail suspension test, were also attenuated by RG in CUMS mice.
Conventional antidepressants, such as selective serotonin reuptake inhibitors, have been used previously [60]. However, antidepressant-treated patients with chronic inflammation or elevated baseline IL-6 and TNF levels had poor treatment responses [61][62][63]. Thus, regulating inflammation as an immune-targeted therapeutic for depression has been widely studied [64]. Stress is a well-known contributor to depression. However, there is no effective treatment for reducing stress as a preventive measure against depression. We thought that RG treatment could be an effective method for reducing stress-induced neuroinflammation because neuroinflammation is the primary mechanism of stress-induced depression.