Lactobacillus casei and Its Supplement Alleviate Stress-Induced Depression and Anxiety in Mice by the Regulation of BDNF Expression and NF-κB Activation

Stress-induced depression and anxiety (DA) are closely connected to gastrointestinal inflammation and dysbiosis, which can suppress brain-derived neurotrophic factor (BDNF) in the brain. Herein, we isolated the BDNF expression-inducing probiotics Lactobacillus casei HY2782 and Bifidobacterium lactis HY8002 in lipopolysaccharide-stimulated SH-SY5Y cells. Then, we investigated the effects of HY2782, HY8002, anti-inflammatory L-theanine, and their supplement (PfS, probiotics-fermented L-theanine-containing supplement) on DA in mice exposed to restraint stress (RS) or the fecal microbiota of patients with inflammatory bowel disease and depression (FMd). Oral administration of HY2782, HY8002, or L-theanine alleviated RS-induced DA-like behaviors. They also decreased RS-induced hippocampal interleukin (IL)-1β and IL-6 levels, as well as NF-κB-positive cell numbers, blood corticosterone level, and colonic IL-1β and IL-6 levels and NF-κB-positive cell numbers. L-theanine more potently suppressed DA-like behaviors and inflammation-related marker levels than probiotics. However, these probiotics more potently increased RS-suppressed hippocampal BDNF level and BDNF+NeuN+ cell numbers than L-theanine. Furthermore, HY2782 and HY8002 suppressed RS-increased Proteobacteria and Verrucomicrobia populations in gut microbiota. In particular, they increased Lachnospiraceae and Lactobacillacease populations, which are closely positively associated with hippocampal BDNF expression, and suppressed Sutterellaceae, Helicobacteriaceae, Akkermansiaceae, and Enterobacteriaceae populations, which are closely positively associated with hippocampal IL-1β expression. HY2782 and HY8002 potently alleviated FMd-induced DA-like behaviors and increased FMd-suppressed BDNF, serotonin levels, and BDNF-positive neuronal cell numbers in the brain. They alleviated blood corticosterone level and colonic IL-1β α and IL-6 levels. However, L-theanine weakly, but not significantly, alleviated FMd-induced DA-like behaviors and gut inflammation. BDNF expression-inducing probiotic (HY2782, HY8002, Streptococcus thermophilus, and Lactobacillus acidophilus)-fermented and anti-inflammatory L-theanine-containing supplement PfS alleviated DA-like behaviors, inflammation-related biomarker levels, and gut dysbiosis more than probiotics or L-theanine. Based on these findings, a combination of BDNF expression-inducing probiotics with anti-inflammatory L-theanine may additively or synergistically alleviate DA and gut dysbiosis by regulating gut microbiota-mediated inflammation and BDNF expression, thereby being beneficial for DA.


Bacterial Strains
HY2782 and HY8002 were isolated from human fecal lactic acid bacteria collection, and Streptococcus thermophilus and Lactobacillus acidophilus strains were purchased from CSI (Culture System Inc., Mishawaka, IN, USA) as commercial strains for starter culture. Probiotics were cultured in MRS broth (BD, Franklin Lakes, NJ, USA) at 37 • C for 18 h, centrifuged (4000× g, 4 • C, 10 min), and washed with saline twice. The collected cells were resuspended in saline for in vivo experiments and heat (75 • C for 15 min)-tyndallized for in vitro experiments.

Animals
C57BL/6 mice (male, 7 weeks old) were provided from Samtaco Inc. (Osan-shi, Seoul, Republic of Korea), maintained in a controlled room with water and food ad libitum, and acclimatized for 7 days before the use of experiment. All animal experiments were approved by the Committee for the Care and Use of Laboratory Animals in Kyung Hee University (IACUC No, KHUASP(SE)-23005) and were ethically carried out in accordance with the Guidelines of the University for Laboratory Animals Care and Use.
DA-like behaviors (one task in one day) were assessed from the next day after the final probiotic treatment. Mice were euthanized by exposure to CO 2 in a chamber and then sacrificed by cervical dislocation. Sera, brains, colons, and feces were collected and stored at −80 • C for biomarker assays.

Behavioral Tasks
The open field test (OFT) was performed in a chamber (40 × 40 cm; center zone, 20 × 20 cm) equipped with a record camera for 10 min and quantified using the EthoVision XT software [26]. The elevated plus maze task (EPMT) and tail suspension test (TST) were assessed in the plus-maze apparatus and table edge, as previously reported [25].

Immunofluorescence Staining
Mice were transcardially perfused with paraformaldehyde. Their hypothalamus and colon tissues were sectioned and incubated with primary antibodies for 12 h, then treated with secondary antibodies and observed using a confocal microscope, as previously reported [28]. The sections were incubated with primary antibodies against BDNF, NeuN, NF-κB, Iba1, and/or CD11c for 12 h, then treated with secondary antibodies conjugated with Alexa Fluor 594 or Alexa Fluor 488 (1:200, Invitrogen, Waltham, MA, USA) for 2 h, and observed using a confocal microscope.

Microbiota Analysis
Microbiota genomic DNA was extracted from the stool of mice using a QIAamp DNA stool mini kit. Next, 16S rRNA genes were amplified and sequenced, as previously reported [29]. Sequenced data were deposited in NCBI (PRJNA962265).

Statistics
Data are expressed as mean ± S.D. using GraphPad Prism 9. The significant differences were analyzed using one-way ANOVA followed by Duncan's multiple range test (p < 0.05). The correlation between gut microbiota and BDNF expression or IL-1β level was analyzed using the Spearman correlation coefficient.

L. casei and B. lactis Up-Regulated LPS-Suppressed BDNF Expression in SH-SY5Y Cells
To select stress-induced DA-ameliorating probiotics, we screened BDNF expressioninducing probiotics from our lactic acid bacteria collection. Of the tested bacteria, L. casei HY2782 and B. lactis HY8002 potently increased LPS-suppressed BDNF expression in SH-SY5Y cells compared to that of the normal control (NC) ( Figure 1).
with secondary antibodies and observed using a confocal microscope, as previously reported [28]. The sections were incubated with primary antibodies against BDNF, NeuN, NF-κB, Iba1, and/or CD11c for 12 h, then treated with secondary antibodies conjugated with Alexa Fluor 594 or Alexa Fluor 488 (1:200, Invitrogen, Waltham, MA, USA) for 2 h, and observed using a confocal microscope.

Microbiota Analysis
Microbiota genomic DNA was extracted from the stool of mice using a QIAamp DNA stool mini kit. Next, 16S rRNA genes were amplified and sequenced, as previously reported [29]. Sequenced data were deposited in NCBI (PRJNA962265).

Statistics
Data are expressed as mean ± S.D. using GraphPad Prism 9. The significant differences were analyzed using one-way ANOVA followed by Duncan's multiple range test (p < 0.05). The correlation between gut microbiota and BDNF expression or IL-1β level was analyzed using the Spearman correlation coefficient.

L. casei, B. lactis, L-Theanine, and Their Supplement Improved RS-Induced GI and GD in Mice
The effects of HY2782 and HY8002 on RS-induced GI were investigated in the colon of mice ( Figure 5). Exposure to RS induced colon shortening and increased IL-1β, IL-2, IL-6, and TNF-α levels, as well as NF-κB-positive cell numbers. However, HY2782 and HY8002 decreased RS-increased IL-1β and IL-6 levels, and NF-κB-positive cell numbers. L-theanine did not affect IL-1β, IL-2, IL-6, or TNF-α levels. PfS strongly suppressed IL-1β, IL-6, and TNF-α levels, as well as NF-κB-positive cell numbers, compared to those of Ltheanine or probiotics.  The effects of probiotics, L-theanine, and their supplement on RS-induced gut microbiota alteration were investigated in mice ( Figure 6, Supplementary Materials, Tables S1-S3). Exposure to RS-altered fecal microbiota composition: Exposure weakly, but not significantly, decreased α-diversity (OTUs) and shifted β-diversity (PCoA). RS increased Proteobacteria, Verrucomicrobia, and Actinobacteria populations, and decreased Firmicutes and Deferribacteres populations. However, HY2783, HY8002, and PfS suppressed RS-increased Proteobacteria and Verrucomicrobia populations and increased Firmicutes and Deferribacteres populations. Although L-theanine weakly, but not significantly, increased RS-suppressed Firmicutes population, it did not significantly affect RS-increased Verrucomicrobia or Actinobacteria populations. Of the gut microbiota, RS-decreased Ruminococcaceae and Lactobacillacease populations showed a positive correlation with BDNF expression levels, while RS-increased Sutterellaceae, Helicobacteriaceae, Akkermansiaceae, Enterobacteriaceae, and Bifidobacteriaceae numbers had a negative correlation. IL-1β expression levels showed a positive correlation with Sutterellaceae, Bifidobacteiralceae, and Helicobacteriaceae populations, while Lactobacillaceae and Peptococcaceae populations had a negative correlation.
Firmicutes population, it did not significantly affect RS-increased Verrucomicrobia or Actinobacteria populations. Of the gut microbiota, RS-decreased Ruminococcaceae and Lactobacillacease populations showed a positive correlation with BDNF expression levels, while RSincreased Sutterellaceae, Helicobacteriaceae, Akkermansiaceae, Enterobacteriaceae, and Bifidobacteriaceae numbers had a negative correlation. IL-1β expression levels showed a positive correlation with Sutterellaceae, Bifidobacteiralceae, and Helicobacteriaceae populations, while Lactobacillaceae and Peptococcaceae populations had a negative correlation.

L. casei, B. lactis, L-Theanine, and Their Supplement Improved FMd Transplantation-Induced GI in Mice
The effects of HY2782 and HY8002 on FMd-induced GI were investigated in the colon of mice ( Figure 10). FMd transplantation increased IL-1β, IL-2, IL-6, and TNF-α levels, as well as NF-κB-positive cell numbers. However, HY2782 and HY8002 decreased IL-1β, IL-6, and TNF-α levels, as well as NF-κB-positive cell populations. However, L-theanine weakly suppressed these compared to probiotics. Treatment with PfS strongly suppressed the above levels compared to L-theanine or probiotics alone.

Discussion
Exposure to stressors induces the secretion of adrenocorticotropic hormone from the pituitary gland, which the stimulates the excretion of cortisol (corticosterone) from the adrenal gland through the hypothalamic-pituitary-adrenal axis and activates NF-κB signaling [30,31], while serotonin and BDNF levels decrease in the central nervous system and gastrointestinal tract, resulting in DA with neuroinflammation and colitis [32,33]. In the present study, exposure to RS increased blood corticosterone and IL-6 levels, and decreased hippocampal BDNF and serotonin levels, as well as BDNF-positive neuron cells. Moreover, exposure to RS increased IL-1β levels and NF-κB-positive cells in the brain and colon. RS increased DA-like behaviors. Furthermore, RS increased the gut Proteobacteria and Verrucomicrobia populations. Hippocampal BDNF expression levels were positively correlated with Ruminococcaceae and Lactobacillacease populations, which were negatively correlated with the expression levels of inflammatory cytokines such as IL-1β. Hippocampal BDNF expression levels were negatively correlated with Sutterellaceae, Helicobacteri-aceae, Akkermansiaceae, and Enterobacteriaceae populations, which were positively correlated with the expression levels of inflammatory cytokines such as IL-1β. Peirce and Alvina suggested that RS-induced depression could cause GI and GD [34]. Jang et al. reported that RS increased blood and fecal LPS levels through GD [3]. These observations suggest that RS may cause GI and DA by suppressing corticosterone-mediated BDNF and serotonin expression, and inducing corticosterone-mediated NF-κB signaling through LPS-overexpressed GD.
Here, we selected BDNF expression-increasing probiotics L casei HY2782 and B. lactis HY8002 in LPS-treated SH-SY5Y cells. These probiotics increased RS-suppressed BDNF and serotonin levels in the brain. However, they weakly suppressed RS-induced hippocampal and colonic proinflammatory cytokine and NF-κB-positive cell levels. Moreover, they weakly reduced blood corticosterone levels, and weakly alleviated RS-induced DA-like behaviors. Nevertheless, they shifted RS-fluctuated gut microbiota composition to that of NC. In particular, these probiotics suppressed RS-increased Proteobacteria and Verrucomicrobia populations. Furthermore, they increased Lachnospiraceae and Lactobacillacease populations, which are closely positively associated with hippocampal BDNF expression, and suppressed Sutterellaceae, Helicobacteriaceae, Akkermansiaceae, and Enterobacteriaceae populations, which are closely positively associated with hippocampal IL-1β expression. L-theanine also increased serotonin and BDNF levels, similarly to HY2782. However, L-theanine showed an anti-depressive effect. These results suggest that L-theanine may express its anti-depressive effects by regulating depressive factors such as neuropeptide Y and adrenaline [35], differently from probiotics.
Interestingly, HY2782 and HY8002 potently alleviated FMd-increased DA-like behaviors. They decreased FMd-induced hippocampal and colonic proinflammatory cytokine expression, and increased FMd-suppressed hippocampal BDNF expression and BDNFpositive neuron cell numbers. L-theanine strongly alleviated RS-induced DA-like behaviors associated with neuroinflammation and colitis. Its effects were more potent than those of probiotics in mice with RS-induced DA. However, L-theanine treatment hardly affected RS-fluctuated gut microbiota or hippocampal BDNF level. L-theanine hardly induced LPS-suppressed BDNF expression in vitro (Supplementary Materials, Figure S1). However, L-theanine weakly alleviated FMd-induced DA-like behaviors. L-theanine also weakly decreased the expression of IL-1β, IL-6, and TNF-α, which are neuroinflammation-and colitis-related biomarkers [36,37]. Zhang et al. found that L-theanine suppressed colitis in rodents by suppressing NF-κB signaling [38]. Gut inflammation is closely associated with the outbreak of psychiatric disorders, including DA and gut dysbiosis [4,39,40]. DA causes gut dysbiosis [41], which alters gut microbiota composition and overexpresses microbiota toxins such as LPS, and inflammation [25], which further induces DA [42]. Proinflammatory cytokines suppress BDNF and serotonin production in neuron cells [43]. The suppression of systemic inflammation alleviates DA in vivo [25]. We found that RS and FMd transplantation increased hippocampal and colonic IL-1β and IL-6 levels, as well as NF-κB-positive cell numbers, while hippocampal serotonin and BDNF levels decreased, as previously reported [25]. However, FMd transplantation more potently increased colonic IL-1β and IL-6 levels, as well as blood corticosterone levels, in mice that had been exposed to RS. These observations imply that BDNF expression-inducing probiotics, in particular L. casei, may alleviate RS-induced DA with gut dysbiosis through the modulation of gut microbiota composition and BDNF expression, while L-theanine may alleviate DA with neuroinflammation and colitis through the suppression of NF-κB activation. Treatment with PfS, which is a probiotics-fermented L-theanine-containing supplement, alleviated RS-or FMd-induced DA-like behaviors, decreased hippocampal IL-1β and IL-6 levels, and increased RS-or FMd-suppressed hippocampal BDNF, serotonin levels, and BDNF-positive neuron cell numbers more potently than probiotics or L-theanine alone. Treatment also suppressed RS-or FMd-increased blood corticosterone levels, colonic IL-1β and IL-6 levels, and NF-κB-positive cell numbers more than probiotics or L-theanine alone. Furthermore, PfS suppressed Proteobacteria and Verrucomicrobia populations in RS-exposed mice. PfS also increased Lactobacillacease populations, which is closely positively associated with hippocampal BDNF expression, and suppressed Sutterellaceae, Helicobacteriaceae, Akkermansiaceae, and Enterobacteriaceae populations, which are closely positively associated with hippocampal IL-1β expression.
These observations suggest that the combination of BDNF expression-inducing probiotics, including HY2782 and HY8002, with anti-inflammatory L-theanine may additively or synergistically alleviate stress-induced DA by modulating gut microbiota-mediated inflammation and BDNF expression.

Conclusions
BDNF expression-inducing L. casei HY2782 and B. lactis HY8002 alleviated RD-or fecal microbiota-induced DA-like behaviors, neuroinflammation, and GI with GD by suppressing gut microbiota-mediated NF-κB activation and inducing BDNF expression. L-theanine alleviated RS-induced DA-like behaviors by suppressing NF-κB signaling. The combination of BDNF expression-inducing probiotics, including HY2782 and HY8002, with anti-inflammatory L-theanine may additively alleviate DA by modulating gut microbiotamediated inflammation and BDNF expression, thereby being beneficial for DA.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/nu15112488/s1, Table S1. Effects of HY2782 and its supplements on the fecal microbiota composition (at the phylum level). Table S2. Effects of HY2782 and its supplements on the fecal microbiota composition (at the family level). Table S3. Effects of HY2782 and its supplements on the fecal microbiota composition (at the genus level). Figure S1. Effect of L-theanine on LPS-suppressed BDNF expression in SH-SY5Y cells.

Data Availability Statement:
The data analyzed during the present study are available from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.

Abbreviations
BDNF, brain-derived neurotropic factor; CD, central distance; CN number of entries in the center; CT, central area-spent time; CUMS, chronic unpredictable mild stress; DA, depression and anxiety; ELISA, enzyme-linked immunosorbent assay; EPMT, elevated plus maze task; GAM, general anaerobic medium; LPS, lipopolysaccharide; OFT, open field test; NC, normal control; SD, standard deviation; TD, distance travelled; TST, tail suspension test.