Tryptophan-rich diet and its effects on brain Tregs in alleviating neuroinflammation and cognitive impairment

doi:10.21203/rs.3.rs-3891460/v1

Download PDF

Research Article

Tryptophan-rich diet and its effects on brain Tregs in alleviating neuroinflammation and cognitive impairment

https://doi.org/10.21203/rs.3.rs-3891460/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 27 Sep, 2024

Read the published version in Journal of Neuroinflammation →

You are reading this latest preprint version

Background: Perioperative neurocognitive disorders (PND) present a common challenge for elderly people and contribute to increased postoperative dementia rates and mortality. Notably, neuroinflammation serves as a pivotal pathogenic mechanism for PND. Regulatory T cells (Tregs) exhibit potent anti-inflammatory properties and can modulate neurodegenerative diseases arising from central nervous system inflammatory responses. However, the role of Tregs in neuroinflammation-related PND remains unclear. It is highly plausible that brain Tregs expressing unique genes associated with the nervous system, including the Htr7 gene encoding the serotonin receptor 5-HT7, play a pivotal role.

Methods: A model of neuroinflammation-mediated cognitive dysfunction was established via intracerebroventricular injection of lipopolysaccharide (LPS). The activation and infiltration of Tregs were measured using flow cytometry. Metagenomic sequencing of fecal samples was employed to investigate alterations in gut bacterial abundance following LPS administration. TMT-based quantitative proteomics analysis was employed to detect the proteomes of hippocampal tissue following LPS treatment. Primary Tregs were cocultured separately with primary CD8⁺T cells and primary microglia for in vitro validation of the impact of 5-HT and Htr7 on Tregs. Prior to their transfer into recombination activating gene 1 (Rag1^−/−) mice, Tregs were ex vivo transfected with lentivirus to knock down the expression of Htr7.

Results:

In this study, the tryptophan-rich diet was found to reverse LPS-induced cognitive impairment and reduce the levels of 5-HT in peripheral blood. Following intracerebroventricular microinjection of LPS, there was a decrease in the abundance of 5-HT-related gut microbiota. The tryptophan-rich diet led to increased levels of 5-HT in peripheral blood, which in turn promoted the proliferation and activation of brain Tregs. Additionally, proteomic alterations were observed in the hippocampus of the mice following LPS treatment. The tryptophan-rich diet was also shown to attenuate LPS-mediated neuroinflammation by activating brain Tregs. Furthermore, 5-HT and Htr7 were found to enhance the immunosuppressive effect of Tregs on CD8⁺ T cells and microglia. In Rag1^-/- mice, brain Tregs were shown to alleviate LPS-induced neuroinflammation and cognitive impairment.

Conclusions:

Our research revealed the ability of brain Tregs to mitigate neuroinflammation and prevent neuronal damage by suppressing the infiltration of CD8⁺T cells into the brain and excessive activation of microglia, thereby ameliorating LPS-induced cognitive impairment. These insights may offer novel therapeutic targets involving Tregs for PND.

Postoperative cognitive dysfunction

Regulatory T cells

Serotonin

Neuroinflammation

Lipopolysaccharide

Gut microbiota

Perioperative neurocognitive disorders (PND), also known as cognitive impairments that occur during the perioperative period, encompass a range of central nervous system complications induced by anesthesia and surgery. The primary manifestations of PND include diminished learning and memory capabilities, as well as impaired cognitive functions such as reduced mental focus. In severe cases, individuals may exhibit personality changes and a decline in social adaptability, with some even experiencing irreversible cognitive impairments [1–3]. Although PND can affect patients of all ages, it is more prevalent among elderly people. Research indicates that the incidence of cognitive impairment one week after noncardiac surgery in elderly patients reaches 25.8%, with a persistently high rate of 9.9% after three months. In certain cases, these impairments may persist for more than a year, ultimately leading to progression to elderly dementia [4]. The impact of PND, especially long-term PND, extends beyond the individual's health, affecting the rehabilitation process and resulting in prolonged hospital stays, increased medical expenses, diminished quality of life, and elevated mortality rates. Consequently, this places a significant burden on both healthcare and human resources, exerting a substantial toll on individuals and society at large. Given the intensifying aging population, advancements in modern medical technologies, and evolving healthcare philosophies, the number of elderly patients undergoing surgical procedures is steadily increasing. Consequently, PND has emerged as an increasingly significant concern during the early postoperative period among elderly patients. Therefore, further investigations into the mechanisms and preventive measures of PND hold immense social significance.

Neuroinflammation plays a crucial role in neurodegenerative diseases such as Alzheimer's disease, schizophrenia, and age-related cognitive decline. This conclusion has been substantiated through both animal experiments and clinical trials. The stimulation of anesthesia and surgical procedures can trigger robust central and peripheral inflammatory responses, concurrently eliciting the migration of peripheral immune cells into the brain [5, 6]. However, the exact pathogenic mechanism by which neuroinflammation impacts cognitive function remains elusive. Administration of lipopolysaccharide (LPS) systemically and through intracerebroventricular microinjection can induce cognitive impairment in animal models [7, 8]. Studies have demonstrated the pivotal role of the neuroinflammatory response in the LPS-induced murine model of memory impairment. Alleviating the inflammatory response in the hippocampus can ameliorate learning and memory deficits in mice [9].

Regulatory T cells (Tregs) are vital components of the immune system that function to maintain immune homeostasis and limit excessive inflammatory responses, thereby protecting the body from inflammatory damage [10]. Tregs secrete inhibitory inflammatory factors such as IL-10 and TGF-β, which suppress the activation of effector cells, including T and B lymphocytes, macrophages, and dendritic cells, and inhibit their infiltration into effector brain regions. Tregs that infiltrate the brain can also exert immunosuppressive effects, possibly by modulating the survival conditions of cytotoxic T cells, leading to their apoptosis [11]. In their study, Tian et al. discovered significant infiltration of Tregs in the hippocampal region, with upregulation of IL-10 mRNA expression, indicating their role in suppressing immune responses [12]. Minako et al. conducted a systematic investigation of Tregs and reported that a specific subtype of brain Tregs (Htr7⁺ Tregs) express the receptor Htr7 for 5-hydroxytryptamine (5-HT), also known as serotonin. Binding of 5-HT to Htr7 stimulates the proliferation of brain Tregs and enhances their immune activity. Brain Tregs proliferate in the cervical lymph nodes and can also infiltrate effector brain regions, where they exert immunosuppressive effects [13]. Approximately 90% of 5-HT in the human body is produced in the intestine, and 5-HT is synthesized in enterochromaffin cells and the enteric nervous plexus. Through enzymatic catalysis in intestinal cells, tryptophan is converted to 5-HT in the gut. Furthermore, studies have indicated that certain microbial communities in the gut play pivotal roles in the synthesis of 5-HT [14, 15]. A diet rich in tryptophan can increase 5-HT levels in peripheral blood [16]. Hence, we can try to modulate brain Tregs by augmenting peripheral blood levels of 5-HT in mice through the administration of a diet rich in tryptophan.

Accordingly, we hypothesize that brain Tregs is associated with neuroinflammation induced cognitive dysfunction. To test this hypothesis, we assessed the impact of brain Tregs on neuroinflammation and cognitive function following intracerebroventricular administration of LPS, aiming to explore potential Tregs-based therapeutic strategies for PND.

Animals

Male C57BL/6J WT mice aged 8 weeks (weighing 22 to 25 g) and C57BL/6J WT neonatal mice aged 24 h (weighing 9 to 11 g) were obtained from the Beijing SPF Animal Technology Company. Rag1^−/− mice (B6.129S7-Rag1tm1Mom/J) on the C57BL/6 background were originally purchased from The Jackson Laboratory. The mice were housed in a carefully regulated environment with standard temperature and humidity levels to ensure their well-being and minimize any potential stressors. The animals were given free access to food and water throughout the duration of the experiment, allowing them to maintain their normal nutritional intake. All animal experiments were approved by the Ethics Committee for Animal Experimentation of the Chinese PLA General Hospital.

Intracerebroventricular Microinjection Surgery

Prior to the stereotactic injection of LPS (Escherichia coli serotype 0111:B4, Sigma‒Aldrich, St. Louis, USA) or artificial cerebrospinal fluid (aCSF), the mice were anesthetized by an intraperitoneal injection of Avertin (200 mg/kg). Specific coordinates (posterior: 0.5, lateral: ± 1.0, and ventral: 2.0 from the bregma [in mm]) were used to precisely locate the lateral ventricle for the injection. Subsequently, LPS or aCSF was administered at a consistent rate within a 3-minute time frame. Following the injection, the needle was maintained in position for a minimum of 3 minutes.

Five days after the Morris water maze (MWM) training phase, LPS (2 µg of LPS dissolved in 2 µL of aCSF [containing 140 mM NaCl, 3.0 mM KCl, 2.5 mM CaCl₂, 1.2 mM Na₂HPO, and 41.0 mM MgCl₂]) was intracerebroventricularly injected. The control groups were administered an equivalent volume of aCSF.

Experimental diets

Based on previous studies [16], we fed the mice a similar tryptophan-rich diet. All mice were nourished with diets conforming to the AIN-93 M (3.872 kcal/g), comprising 17.56% kcal of protein, 70.82% kcal of carbohydrates, and 11.62% kcal of fat (HFK Bioscience, Beijing, China). The Sham group, LPS group, and the LPS + Trp group were treated as follows: the Sham group and the LPS group contained 0.16% tryptophan, and the LPS + Trp group contained 0.6% tryptophan.

MWM test

The Morris water maze test was conducted in a large circular pool filled with water and featuring a circular hidden platform, serving as the basis for evaluating spatial learning and memory in mice. The pool was then provided visual cues to aid the mice in remembering the platform's location. The test comprised place navigation trials and a spatial probe test. Following daily training sessions, the latency of the mice to find the platform was recorded and averaged. After five days of navigation training, the mice received intracerebroventricular microinjections. The following day, the hidden platforms were removed, and the spatial probe test began, during which the mice were allowed to access the platform. Various parameters were recorded throughout the experiment.

Primary cell coculture

Primary Tregs and CD8⁺ T cells were isolated from the cervical lymph nodes of healthy mice via magnetic sorting (Miltenyi, Germany) and were separately prepared and cocultured.

Primary Tregs isolated from healthy mice cervical lymph nodes and microglia isolated from neonatal mice brains using magnetic sorting (Miltenyi, Germany) were separately prepared and cocultured.

After 4 h of coculture in a cell culture incubator, the cells were divided into PBS, 5-HT, and 5-HT + SB269970 (5-HT + SB) groups. SB269970 (Abcam, ab120508, Cambridge, UK) is a selective inhibitor of Htr7 [17]. The PBS group received PBS, the 5-HT group received 5-HT, and the 5-HT + SB group received both 5-HT and SB269970 for an additional 24 h. Drug concentrations were 8.3 nM/mL for 5-HT and 200 ng/mL for SB269970.

Flow Cytometry

The tissues were digested and made into single-cell suspensions for flow cytometry. The antibodies and detection kits used were from several companies: Mouse Regulatory T Cell Staining Kit Anti-Mouse CD206 PE, Anti-Mouse MHC-II FITC, Anti-Mouse MHC-I FITC, Rat IgG2a Isotype Control PE, Rat IgG2a Isotype Control FITC, Rat IgG1 Isotype Control APC, Armenian Hamster IgG Isotype Control Percp-Cyanine 5.5, Anti-mouse/human CD44, Anti-mouse CD62L, Anti-Mouse CD45 Percp-Cyanine 5.5, and Anti-Mouse CD11b APC were from eBioscience (Waltham, MA, USA); Carboxyfluorescein diacetate succinimidyl ester (CFSE) was from Absin Biology (Shanghai, China); Anti-Mouse 5-HT7 was from Novus Biologicals (Littleton, CO, USA); and the Annexin V-FITC Apoptosis Staining/Detection Kit was from Abcam (Cambridge, UK). Flow cytometry data acquired from a FACS Calibur (BD Biosciences, FACSCanto II, San Jose, CA, USA) were subjected to analysis using FlowJo V10 software.

Lentiviral transfection of Tregs

Htr7 downregulation was achieved by transfecting lentiviral particles expressing a Flag epitope-tagged form of murine protein into Tregs from cervical lymph nodes 72 h before the experiments. pSLenti-EF1-EGFP-CMVCHRNA7(GV417)-3FLAG (denoted as the empty vector) was used as a control lentivirus. Htr7 downregulation was achieved by transfecting Tregs with lentivirus expressing Htr7-specific short hairpin RNAs (shHtr7) 72 h before the experiments. A lentiviral vector expressing a scrambled sequence (shScramble) was used as a control shRNA. The lentiviral vectors containing Htr7-Flag, the empty vector, shHtr7, and shScramble were purchased from Obio (Shanghai, China).

Adoptive cell transfer in Rag1^-/- mice

Tregs isolated from healthy mice cervical lymph nodes via magnetic sorting (Miltenyi, Germany) were plated at 5 × 10⁵ cells/well in a 24-well plate and stimulated with anti-CD3/CD28 beads (Miltenyi, Germany) for 48 h. After transfection with the Htr7-Flag lentivirus to downregulate Htr7 expression, 2×10⁶ transfected brain Tregs and CD4⁺CD25⁻T cells were injected via the femoral vein into recipient Rag1^−/− mice before the MWM test. The Rag1^−/− mice were randomly divided into four groups: the Sham group (aCSF + CD4⁺CD25⁻T cells), the LPS group (LPS + CD4⁺CD25⁻T cells), the LPS + Treg group (LPS + Tregs transfected with empty vector + CD4⁺CD25⁻T cells), and the LPS + Treg^Lv group (LPS + Tregs transfected with Htr7-Flag + CD4⁺CD25⁻T cells). Mice were treated with LPS or aCSF via intracerebroventricular injection. The premixes of transfected Tregs and CD4⁺CD25⁻T cells were injected via the femoral vein.

Western blotting

For the Western blot analysis, equimolar quantities (40 µg) of total protein per sample were separated via SDS‒PAGE on a 12.5% gel. The proteins were subsequently transferred onto a polyvinylidene fluoride membrane using the semidry blotting technique. The membrane was then subjected to an extended overnight blocking phase at 4°C utilizing a 5% BSA solution. Subsequently, the membrane was incubated with primary antibodies against the proteins of interest: perforin (1:1000, Abcam, ab97305, Cambridge, UK) and granzyme B (1:3000, Abcam, ab255598, Cambridge, UK). The membranes were incubated with these antibodies at ambient temperature for 2.5 h. After the necessary washes were performed, the membranes were incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit secondary antibody (1:10,000, Abcam, ab107866, Cambridge, UK). The images were subjected to analysis using HIN ImageJ software and were presented as the density ratio relative to that of β-actin.

Antibodies for immunofluorescence staining

The primary antibodies used for immunofluorescence staining were as follows: rabbit anti-CD8 (1:500, Servicebio, P01731), mouse anti-Iba-1 (1:500, Servicebio, O70200), rabbit anti-Ly6g (1:200, Servicebio, P35461), rabbit anti-CD3 (1:50, Abcam, ab135372), rabbit anti-myelin basic protein (MBP, 1:500, Abcam, ab40390), and mouse anti-nonphosphorylated neurofilaments (SMI32, 1:1000, Biolegend, 801701). The secondary antibodies used for immunohistochemical staining were purchased from Invitrogen.

Quantitative real-time PCR (qRT‒PCR)

The cellular specimens were subjected to total RNA extraction using an RNA extraction kit (Solarbio, R1200, Beijing, China). A universal reverse transcription kit (Solarbio, RP1105, Beijing, China) was subsequently used for reverse transcription. Real-time PCR was carried out using a universal RT‒PCR kit (Solarbio, RP1200, Beijing, China) on a Step One Plus thermal cycler (Thermo Fisher, 4376592, Waltham, MA, USA). The primers used were designed according to the principles of primer design, and their specific sequences can be found in Table 1.

Table 1

Primer sequences used for qRT‒PCR analysis.
Primer Name	Forward Primer (5’-3’)	Reverse Primer (5’-3’)
Perforin	CAAGGTAGCCAATTTTGCAGC	GTACATGCGACACTCTACTGTG
Granzyme B	GCCCCACTCTCGACCCTA	AGCACAAAGTCCTCTCGAAT
IL-1β	ACTCATTGTGGCTGTGGAGA	TTGTTCATCTCGGAGCCTGT
iNOS	CACAGTGTCGCTGGTTTGAA	TCTCCGTGGGGCTTGTAGT
IL-4	GGTCTCAACCCCCAGCTAGT	GCCGATGATCTCTCTCAAGTGAT
TNF-α	CTCATGCACCACCATCAAGG	ACCTGACCACTCTCCCTTTG

Metagenomic sequencing

Fecal samples were collected and preserved in tubes containing a DNA stabilizer (Sarstedt) and stored at − 80°C. Total fecal DNA was extracted using a QIAamp DNA Stool Mini Kit, and the purity and quantity of the DNA were examined using Epoch Microplate Spectrophotometer (BioTek, USA). For metagenome sequencing, libraries were constructed using Illumina DNA Prep kits (Illumina, USA), and prepared libraries with target insert sizes of approximately 350 bp were sequenced on an Illumina NovaSeq 6000 sequencer using S4 flow cells by Oebiotech Co., Ltd. (Shanghai, China).

The representative sequences of the gene sets (amino acid sequences) were compared to those in the GO database using DIAMOND (v0.9.7) software. The BLAST comparison parameters were set with an expected value (e-value) of 1e-5. Taxonomic information was obtained by aligning the sequences to the NR database, and the gene abundances were calculated based on the corresponding species. The species abundances were then integrated to calculate the overall abundance of each species. The abundances of species at different taxonomic levels, including domain, kingdom, phylum, class, order, family, genus, and species, were calculated to construct abundance profiles at each taxonomic level across the samples.

TMT-based quantitative proteomics analysis

Total protein was extracted from the samples. A portion of the extracted protein was subjected to protein concentration determination and SDS‒PAGE analysis, while the other portion was subjected to trypsin digestion and labeling. Subsequently, equal amounts of the labeled samples were mixed and subjected to chromatographic separation. Finally, the samples were analyzed using LC–MS/MS, and the data were analyzed.

The basic process of bioinformatics analysis involved database searching for qualitative and quantitative data. After quality assessment and preprocessing, expression level analysis and functional analysis were conducted. Common databases were used for functional annotation analysis of the identified proteins. The differentially expressed proteins were subjected to GO analysis. Additionally, further research and validation were conducted on key proteins and their functions or pathways of interest.

Enzyme-linked immunosorbent assay (ELISA)

The concentrations of IL-10 (Thermo Fisher Scientific, 88-7105-22, Minneapolis, MN, USA), TGF-β (Thermo Fisher Scientific, 88-8350-22, Minneapolis, MN, USA), TNF-α (Thermo Fisher Scientific, 88-7324-22, Minneapolis, MN, USA), and IL-1β (Thermo Fisher Scientific, 88-7013-22, Minneapolis, MN, USA) were determined using ELISA kits. Hippocampal tissues were carefully disrupted in a solution of RIPA lysis buffer. Subsequently, the resultant mixture was subjected to centrifugation at a speed of 12,000 rotations/min for 5 minutes, maintaining a frigid temperature of 4°C. Through this process, the protein fraction in the supernatant was effectively isolated. The subsequent experimental procedures strictly adhered to the guidelines provided by the manufacturer. To determine the concentrations of IL-10, TGF-β, TNF-α, and IL-1β, spectrophotometric analysis was conducted, and the absorbance was measured at a wavelength of 450 nm. Utilizing a standard curve, the exact concentrations of the aforementioned proteins were accurately determined.

Statistical analyses

All the results were meticulously scrutinized by an assessor who was blinded to the experimental design. The findings are presented as the mean ± standard error of the mean (SEM) for continuous numerical variables. The difference between two groups was evaluated using Student's t test. For comparisons among multiple groups, either a one- or two-way analysis of variance (ANOVA) was used, followed by a Tukey post hoc analysis. To evaluate the differences between various groups at each time point, two-way ANOVA with a subsequent Tukey post hoc test was conducted. The statistical analyses were performed using GraphPad Prism 9.5 (GraphPad Software, Inc., San Diego, CA, USA). All the statistical tests were two-tailed, and a significance level of P < 0.05 was considered to indicate statistical significance.

Tryptophan-rich diet reversed LPS-induced cognitive impairment and decreased 5-HT in peripheral blood

Our prior work has demonstrated that intracerebroventricular injection of LPS results in learning and memory deficits [18]. Therefore, the present study assessed the protective effects of the tryptophan-rich diet against LPS-induced memory impairment in mice.

Mice were administered either the tryptophan-rich diet or the normal diet for 41 days after two weeks of adapt feeding. After a 35-day dietary intervention involving a diet rich in tryptophan in the LPS + Trp group and a normal diet in the Sham group and LPS group, we performed a five-day MWM test training phase. The LPS group and LPS + Trp group of mice were subsequently subjected to intracerebroventricular injection of LPS. The Sham group received an equal volume of aCSF. The MWM probe test was performed 24 h after intracerebroventricular microinjection (Fig. 1A).

The results of the MWM test revealed that during the training phase of the MWM test, the proficiency of the mice in locating the hidden platform was enhanced with training. There were no notable distinctions in the latency to reach the platform across the groups on a daily basis (F_{(2, 18)} = 0.3505, P = 0.7019, Fig. 1B, E). The probe test was performed 24 h after LPS administration. During the probe test, all three groups of mice displayed a noteworthy preference for the target quadrant, dedicating significantly more time to the target quadrant than to the opposite quadrant. However, the LPS group exhibited a discernible decrease in the duration spent in the target quadrant, which was different from that of the other two groups (F_{(2, 18)} = 7.444, P = 0.0044, Fig. 1C, E). The swimming velocities exhibited by the mice were comparable across all groups, effectively excluding any potential effect of motor or perceptual factors on spatial learning and memory (F_{(2, 18)} = 0.3656, P = 0.6988, Fig. 1D). This suggested that intracerebroventricular LPS treatment disrupted the memory of the shortest path and the location of the platform in mice. Moreover, a tryptophan-rich diet was found to alleviate the detrimental effects of LPS-induced cognitive impairment in mice.

Considering the profound effects of LPS-induced neuroinflammation on the intricate mechanisms of the gut-brain axis, we evaluated the concentrations of 5-HT, the tryptophan metabolite synthesized in the intestines of mice. The HPLC‒MS technique was used to assess the levels of 5-HT in the peripheral blood and hippocampal regions of mice following intracerebroventricular injection. Notably, the serum levels of 5-HT were lower in the LPS group than in the other two groups (F_{(2, 18)} = 7.089, P = 0.0054, Fig. 1F). However, there was no significant difference in the concentration of 5-HT in the hippocampal region among the three groups of mice (F_{(2, 18)} = 0.4304 P = 0.6567, Fig. 1G). These findings suggested that LPS may impact cognitive function by affecting peripheral blood 5-HT levels rather than hippocampal 5-HT levels.

The abundance of 5-HT-related gut microbiota declined after LPS treatment

The gut-brain axis refers to the bidirectional communication between the gut and the brain through neural, endocrine, and immune pathways to maintain normal brain and gut functions. As the gut microbiota primarily exerts its effects through the gut-brain axis, it has now evolved into the concept of the microbiota-gut-brain axis [19]. Over 90% of 5-HT is synthesized from tryptophan in enterochromaffin cells, a process directly regulated by the gut microbiota [20]. To investigate the impact of intracerebroventricular injection of LPS on the gut microbiota, metagenomic sequencing of fecal samples was utilized to probe changes in gut bacterial abundance 24 h after LPS administration. At the genus level, a marked shift in the intestinal microbiota structure was evident between the Sham and LPS groups (Fig. 2A), and at the species level, there was also a notable difference in the relative abundance of intestinal microbiota between the two groups (Fig. 2B). Notably, the relative abundance of 5-HT-associated intestinal microbes (s_Escherichia_coli, s_Klebsiella_pneumoniae, s_Klebsiella_quasipneumoniae, and s_Klebsiella_variicola) in the LPS group was significantly lower than that in the Sham group (all P < 0.0001, Fig. 2C-F). To delineate the distinct bacterial species in the Sham and LPS groups, we conducted linear discriminant analysis coupled with effect size measurements (LEfSe). The LPS group exhibited a greater abundance of species changes than the Sham group, as indicated by both the taxonomic representation and LDA score (Fig. 2G, H). Moreover, the average and median relative abundance values of s_Klebsiella_huaxiensis were greater in the Sham group than in the LPS group (Fig. 2I). In summary, these data indicated that the intraventricular injection of LPS indeed alters the gut microbiota of mice, particularly affecting the flora responsible for gut 5-HT production.

The elevation of peripheral blood 5-HT levels induced by the tryptophan-rich diet promoted the proliferation and activation of brain Tregs

Given the profound impact of intracerebroventricular injection of LPS on immune function, we evaluated the percentage of Tregs in the cervical lymph nodes and brain 24 h after LPS treatment by flow cytometry. There was no discernible difference in the proportion of Foxp3⁺CD25⁺ Tregs among the three groups (F_{(2, 18)} = 0.09447, P = 0.9103, Fig. 3A, B). However, compared with those in the other groups, the fraction of Htr7⁺ Tregs in the cervical lymph nodes in the LPS group was lower (F_{(2, 18)} = 5.511, P = 0.0136, Fig. 3A, C). Flow cytometry analysis of brain-infiltrating Tregs revealed a decrease in Foxp3⁺CD25⁺ Tregs (t₍₁₂₎ = 3.841, P = 0.0023) and Htr7⁺ Tregs (t₍₁₂₎ = 2.821, P = 0.0154) in the LPS group compared to those in the LPS + Trp group (Fig. 3D-F). These findings indicated that microinjection of LPS leads to central inflammatory responses and disruptions in the gut microbiota, which inhibit the proliferation and activation of peripheral brain Tregs while also suppressing their infiltration into the brain. However, the administration of the tryptophan-rich diet to mice to increase peripheral blood 5-HT levels can promote the proliferation of peripheral brain Tregs and their infiltration into the central nervous system.

To assess the immunosuppressive activity of infiltrating brain Tregs in the brain, we employed immunofluorescence and flow cytometry to examine the infiltration of CD8⁺ T lymphocytes in the brains of the mice. Immunofluorescence revealed a significant increase in the number of infiltrating CD8⁺ T cells in the dentate gyrus (DG) of the hippocampus in the LPS group compared to that in the LPS + Trp group (t₍₁₂₎ = 4.166, P = 0.0013, Fig. 3G, H). Similar findings were observed via flow cytometry, which also demonstrated a markedly greater proportion of infiltrating CD44^highCD62L^lowCD8⁺ T lymphocytes in the hippocampal DG of mice in the LPS group than in those in the LPS + Trp group (t₍₁₂₎ = 2.754, P = 0.0175; Fig. 3I, J). These findings suggested that the increase in peripheral blood 5-HT levels induced by the tryptophan-rich diet facilitated the activation of brain Tregs, enhancing their immunosuppressive activity.

Proteomic alterations in the hippocampus of mice induced by intracerebroventricular administration of LPS

The hippocampus, as a crucial brain region for memory formation and storage, plays a pivotal role in the occurrence and progression of postoperative cognitive dysfunction [21–23]. Therefore, alterations in the protein expression profile of hippocampal tissue in mice with postoperative cognitive dysfunction can be screened using proteomic techniques to further understand the mechanism underlying the occurrence of postoperative cognitive dysfunction. To investigate this phenomenon, we established the model of LPS-induced cognitive impairment and conducted TMT-based quantitative proteomics analysis of hippocampal tissues 24 h after LPS injection. Both partial least-squares discrimination analysis (PLS-DA) (data not shown) and unsupervised hierarchical clustering revealed a difference in the protein profiles of the LPS group compared to those of the Sham group (Fig. 4A). Among the 6403 protein species examined, 42 were upregulated and 4 were downregulated in the LPS group. A volcano plot further illustrated the upregulation and downregulation of several inflammation-related proteins, including Hspb1, C3, SIc12a2, Il33, Nos2, Cdk5r1, S100a10, Htr7, and Il-10, in the LPS group compared to the sham group (Fig. 4B). Gene Ontology (GO) enrichment analysis of the differentially expressed proteins revealed that a certain proportion of these proteins were involved in biological processes, including positive regulation of MHC class II and negative regulation of regulatory T cell differentiation (Fig. 4C, D). Notably, flow cytometry and immunofluorescence confirmed the proliferation of Htr7⁺ Tregs and the increase in inflammatory cells in the brain after LPS treatment. This is consistent with the increase in proinflammatory cytokines and decrease in anti-inflammatory cytokine protein expression in the hippocampal region revealed by proteomic analysis, as well as the positive regulation of MHC class II and negative regulation of regulatory T cell differentiation biological processes.

Tryptophan-rich diet attenuated LPS-mediated neuroinflammation by activating brain Tregs

As is well-known, the activation of mature T lymphocytes, neutrophils, and microglia constitutes a response to neuroinflammation, and intense neuroinflammatory damage can result in demyelination of neurons. In light of the preceding proteomic results, we utilized immunofluorescence and ELISA to assess the inflammatory response in the hippocampal region of mice 24 h after LPS injection. CD3 is expressed on the surface of almost all mature T lymphocytes. The immunofluorescence results revealed a significant increase in the activation of CD3⁺ T cells in the DG of the hippocampus in the LPS group compared to the other two groups (t₍₁₂₎ = 2.550, P = 0.0255, Fig. 5A, B). Additionally, the accumulation of neutrophils in the hippocampal DG was more pronounced in the LPS group than in the other two groups (t₍₁₂₎ = 2.495, P = 0.0282, Fig. 5A, C). Moreover, compared to those in the remaining two groups, the LPS group exhibited increased Iba-1 staining in the DG of the hippocampus, which was indicative of a noteworthy increase in microglial activation (t₍₁₂₎ = 2.251, P = 0.0439, Fig. 5A, D). The SMI-32/MBP protein ratio is a widely acknowledged parameter for assessing demyelination. We observed a significant increase in demyelination of neurons in the hippocampal region in mice in the LPS group compared to those in the other two groups (t₍₁₂₎ = 2.263, P = 0.0430, Fig. 5A, E).

The levels of two common anti-inflammatory cytokines (IL-10 and TGF-β) released by Tregs in conjunction with TNF-α and IL-1β (two common proinflammatory cytokines) in the hippocampal region were evaluated through ELISA. Compared to those in the other two groups, the IL-10 levels in the hippocampal region were significantly lower in the LPS group (F_{(2, 18)} = 8.631, P = 0.0024, Fig. 5F). Furthermore, in comparison to those in the other two groups, the hippocampal TGF-β levels in the LPS group were notably lower (F_{(2, 18)} = 5.394, P = 0.0146, Fig. 5G). Conversely, in contrast to those in the other two groups, TNF-α levels in the hippocampal region were greater in the LPS group (F_{(2, 18)} = 6.834, P = 0.0062, Fig. 5H). Similarly, compared to those in the other two groups, the IL-1β levels in the hippocampal region were elevated in the LPS group (F_{(2, 18)} = 11.78, P = 0.0005, Fig. 5I). Collectively, our results suggested that LPS injection induces central inflammatory injury, which can be mitigated through the activation of brain Tregs.

5-HT and Htr7 enhanced the immunosuppressive effect of Tregs on CD8⁺ T cells and microglia

To determine the mechanism by which 5-HT activates the immunosuppressive effect of Tregs, a series of cellular experiments was conducted. Primary Tregs and primary CD8⁺ T cells obtained from cervical lymph nodes were cocultured in a 12-well plate. The cells were divided into PBS, 5-HT, and 5-HT + SB groups according to the different stimulus factors added to the medium. We utilized CFSE to label primary CD8⁺ T cells to track the proliferation of these cells. Flow cytometry analysis revealed that after 24 h of coculture (P1), the proportion of CD8⁺ T cells in the 5-HT group was significantly greater than that in the other two groups (F_{(2, 15)} = 11.66, P = 0.0009). However, after 48 h of coculture (P2), the percentage of CD8⁺ T cells in the 5-HT group was markedly lower than that in the other two groups (F_{(2, 15)} = 20.52, P < 0.0001). In contrast, no significant differences in the proportions of CD8⁺ T cells were observed among the three groups following 72 h of coculture (P3) (F_{(2, 15)} = 0.4412, P = 0.6513, Fig. 6A, B). These findings indicated that the binding of 5-HT to Htr7 can enhance the inhibitory effect of Tregs on the proliferation of CD8⁺ T cells. We also employed flow cytometry to assess the apoptosis of primary CD8⁺ T cells cocultured with primary Tregs (Fig. 6C). The percentage of apoptotic CD8⁺ T cells in the 5-HT group was significantly greater than that in the PBS group and 5-HT + SB group (F_{(2, 18)} = 12.37, P = 0.0004, Fig. 6D). Additionally, we evaluated the alteration in the cytotoxic capacity of CD8⁺ T cells by assessing the expression of perforin (Prf) and granzyme B (GranzB). Similarly, compared with those in both the PBS group and the 5-HT + SB group, the relative expression of Prf mRNA in the 5-HT group was significantly lower (F_{(2, 18)} = 15.26, P = 0.0001, Fig. 6E). Moreover, compared with those in both the PBS group and the 5-HT + SB group, the relative expression levels of GranzB mRNA in the 5-HT group were lower (F_{(2, 18)} = 6.233, P = 0.0088, Fig. 6F). As depicted in Fig. 6G, the relative expression level of the Prf protein in the 5-HT group was markedly lower than that in the PBS group and the 5-HT + SB group (F_{(2, 18)} = 8.173, P = 0.0030, Fig. 6H). Similarly, compared with those in the PBS group and the 5-HT + SB group, the relative expression levels of the GranzB protein in the 5-HT group were significantly lower (F_{(2, 18)} = 6.664, P = 0.0068, Fig. 6G, I).

Primary microglia were subsequently isolated from neonatal mouse brains, and primary Tregs were isolated from lymph nodes and cocultured in a 12-well plate. These cells were also divided into PBS, 5-HT, and 5-HT + SB groups according to the different stimulus factors added to the medium. We employed flow cytometry to assess the polarization of primary microglia cocultured with primary Tregs. MHC-II was used as a marker of M1 polarization, and the mean fluorescence intensity (MFI) of MHC-II revealed a notable reduction in M1 polarization in the 5-HT group compared to both the PBS group and the 5-HT + SB group (F_{(2, 18)} = 8.428, P = 0.0026, Fig. 6J). Furthermore, CD206 was utilized as a marker of M2 polarization, and the MFI of CD206 indicated a substantial increase in M2 polarization in the 5-HT group compared with both the PBS group and the 5-HT + SB group (F_{(2, 18)} = 9.470, P = 0.0015, Fig. 6K). Furthermore, we assessed microglial immune function by examining the expression of inflammatory factor mRNAs in microglia. Compared to those in the PBS group and the 5-HT + SB group, the 5-HT group exhibited significant decreases in the mRNA levels of IL-1β (F_{(2, 18)} = 7.989, P = 0.0033, Fig. 6L). The relative expression level of IL-4 mRNA in the 5-HT group was significantly lower than that in both the PBS group and the 5-HT + SB group (F_{(2, 18)} = 8.400, P = 0.0026, Fig. 6M). Moreover, compared with those in both the PBS group and the 5-HT + SB group, the relative expression levels of inducible nitric oxide synthase (iNOS) mRNA in the 5-HT group were significantly lower (F_{(2, 18)} = 6.866, P = 0.0061, Fig. 6N). Similarly, the mRNA levels of TNF-α were significantly lower in the 5-HT group than in both the PBS group and the 5-HT + SB group (F_{(2, 18)} = 11.33, P = 0.0007, Fig. 6O).

These findings showed that primary brain Tregs exert immunosuppressive effects on primary CD8⁺ T cells and primary microglia by binding to 5-HT and the Htr7 receptor on the cell surface.

Brain Tregs alleviated LPS-induced cognitive impairment in Rag1^-/- mice

We further examined whether brain Tregs mitigated neuroinflammation-induced cognitive impairment (Fig. 7A). After isolating and transfecting the lentivirus to downregulate Htr7 in the cervical lymph nodes of healthy C57BL/6 mice, we confirmed the significant reduction in Htr7 in Tregs through LC‒MS analysis (t₍₁₂₎ = 5.428, P = 0.0002, Fig. 7B). Subsequently, 2×10⁶ transfected Tregs and CD4⁺CD25⁻T cells were intravenously injected into recipient Rag1^−/− mice. Rag1^−/− mice lack mature T and B cells, preventing consideration of the impact of other T and B lymphocytes on the experiment. The orbital blood of Rag1^−/− mice was collected for flow cytometry analysis of CD4⁺ T lymphocytes and Tregs levels in the peripheral blood. The results demonstrated that the levels of CD4⁺ T lymphocytes and Tregs in the peripheral blood of Rag1^−/− mice without Tregs and CD4⁺CD25⁻T-cell infusion were significantly lower than those in the Tregs infusion group. These findings indicate the successful establishment of both the Rag1^−/− immunodeficient mouse model and the animal model for reinfusing lentivirus-transfected Tregs into Rag1^−/− mice (t₍₁₂₎ = 0.8154, P = 0.4307, Fig. 7C, D).

The MWM training was performed 24 h after femoral vein injection. The MWM training results revealed an increase in the proficiency of the mice in locating the concealed platform during the training phase, with no discernible differences in the daily latency to reach the platform across the groups (F_{(3, 24)} = 0.1867, P = 0.9044, Fig. 7E, H). The probe test was conducted 24 h after LPS administration, during which all four groups of mice demonstrated a notable preference for the target quadrant, in which they spent significantly more time in the target quadrant than in the opposite quadrant. Notably, the LPS group and LPS + Treg^Lv group exhibited a distinct decrease in the duration spent in the target quadrant, setting them apart from the other two groups (F_{(3, 24)} = 7.875, P = 0.0008, Fig. 7F, H). The swimming velocities displayed by the mice were comparable across all groups, effectively excluding any potential effect of motor or perceptual factors on spatial learning and memory (F_{(3, 24)} = 0.1774, P = 0.9106, Fig. 7G). These findings suggested that Htr7⁺ Tregs mitigate the deleterious effects of LPS-induced cognitive impairment in mice.

The HPLC‒MS technique was also used to evaluate the levels of 5-HT in the peripheral blood and hippocampal regions of mice following intracerebroventricular injection. The results revealed that, in comparison with those in the Sham group, the other three groups exhibited markedly lower levels of 5-HT in the peripheral blood (F_{(3, 24)} = 5.024, P = 0.0076, Fig. 7I). Furthermore, there was no significant disparity in the concentration of 5-HT in the hippocampal region among the four groups of mice (F_{(3, 24)} = 0.0363, P = 0.9905, Fig. 7J). These findings were consistent with the outcomes of previous experiments, indicating that intracerebroventricular injection of LPS disrupted 5-HT biosynthesis.

Brain Tregs ameliorated LPS-induced neuroinflammation in Rag1^-/- mice

The preceding findings demonstrated that in Rag1^−/− mice, brain Tregs alleviated LPS-induced cognitive impairment and decreased peripheral 5-HT levels. However, whether brain Tregs achieve this effect by mitigating central inflammation remains to be elucidated. Therefore, flow cytometry was used to evaluate the expression of CD206⁺ and MHC-II⁺ microglia in the hippocampus 24 h after LPS treatment (Fig. 8A). In the LPS and LPS + Treg^Lv groups, the expression of CD206⁺ microglia in the hippocampus was lower than that in the other two groups (F_{(3. 24)} = 9.930, P = 0.0002, Fig. 8B). Moreover, the number of MHC-II⁺ microglia in the hippocampus of mice in the LPS and LPS + Treg^Lv groups was significantly greater than that in the other two groups (F_{(2. 18)} = 11.71, P = 0.0006, Fig. 8C). Moreover, the expression of CD206⁺ macrophage in the hippocampal tissue of mice in the LPS group was lower than that in the Sham and LPS + Treg groups (F_{(3. 24)} = 7.313, P = 0.0012, Fig. 8D). Similarly, the number of MHC-II⁺ macrophage in the hippocampal tissue of mice in the LPS and LPS + Treg^Lv groups was greater than that in the other two groups (F_{(2. 18)} = 9.601, P = 0.0015, Fig. 8E). We also found that the upregulation of MHC-I⁺ microglia (F_{(2. 18)} = 7.583, P = 0.0041) and MHC-I⁺ macrophage (F_{(2. 18)} = 5.511, P = 0.0136) in the hippocampal tissue of mice in the LPS and LPS + Treg^Lv groups was evident in comparison to that in the other two groups (Fig. 8F, G). These results demonstrated that brain Tregs promote the protective polarization of microglia and macrophages in Rag1^−/− mice. Notably, immunofluorescence staining of the hippocampal DG region revealed enhanced Iba-1 staining in the LPS group and LPS + Treg^Lv group compared to the other two groups, which similarly indicated a significant increase in microglial activation (F_{(3. 24)} = 12.54, P < 0.0001, Fig. 8H, I). We employed ELISA to evaluate the levels of inflammatory factors in the hippocampal region. Compared to those in the other two groups, the IL-10 levels in the hippocampal region were lower in the LPS group and the LPS + TregLv group (F_{(3. 24)} = 11.05, P < 0.0001, Fig. 8J). Similarly, in contrast to those in the Sham group, the LPS + Treg^Lv group exhibited a notable reduction in the hippocampal TGF-β concentration (F_{(3. 24)} = 4.654, P = 0.0106, Fig. 8K). Moreover, the LPS + Treg^Lv group displayed a substantial increase in TNF-α levels in the hippocampal region compared to those in the Sham and LPS + Treg groups (F_{(3. 24)} = 4.493, P = 0.0122, Fig. 8L). Additionally, the LPS + Treg^Lv group exhibited significantly greater IL-1β levels in the hippocampal region than did the Sham and LPS + Treg groups (F_{(3. 24)} = 4.995, P = 0.0078, Fig. 8M). These findings indicated that brain Tregs alleviate cognitive dysfunction by mitigating central inflammation.

PND is a poorly understood, catastrophic complication with high mortality and disability rates [24]. However, the mechanisms underlying PND remain largely uncertain. In this study, we focused on neuroinflammation-induced cognitive impairment and demonstrated that brain Tregs alleviate cognitive dysfunction by inhibiting central nervous system inflammatory responses. The potential clinical significance of our findings is that brain Tregs may provide a scientific basis for therapeutic strategies for PND.

The pathogenesis of PND involves multiple mechanisms, such as neuroinflammation, oxidative stress, and neurodegenerative changes [25–28]. Tregs play a pivotal role in the intricate process of effector T cell suppression, thereby maintaining self-tolerance and ensuring immune system homeostasis [29]. Moreover, the impairment of Tregs under neurodegenerative conditions leads to a loss of inflammation, ultimately resulting in the persistence of inflammatory environments within the central nervous system. This study investigated the impact of brain Tregs on neuroinflammation and inflammation-related PND in a mouse model of LPS-induced memory impairment.

The gut-brain axis refers to the interaction and communication between the gastrointestinal tract and the brain and encompasses the nervous, endocrine, and immune systems. Central inflammation and disruption of the gut microbiota can influence each other through the gut-brain axis. On the one hand, central inflammation can trigger excessive activation of the immune system, leading to increased release of inflammatory factors. These inflammatory factors may directly or indirectly impact the balance of the gut microbiota, suppressing the growth of beneficial bacteria and promoting the proliferation of harmful bacteria, thereby causing dysbiosis of the gut microbiota. On the other hand, dysbiosis of the gut microbiota may weaken intestinal barrier function, allowing harmful microorganisms and toxins to enter the bloodstream through the intestinal mucosa, triggering peripheral inflammatory responses. These inflammatory factors and metabolites can be transmitted to the brain through the gut-brain axis, leading to neuroinflammation and inflammation-related changes in neural transmission. 5-HT is synthesized primarily in enteric neurons and enterochromaffin cells located in the gastrointestinal muscularis. This intricate process involves the conversion of tryptophan, an amino acid obtained through the diet, into serotonin [30]. On the basis of previous studies on tryptophan-rich diet strategies to increase peripheral blood 5-HT levels in mice, our results showed that the tryptophan-rich diet can significantly reverse the decrease in peripheral blood serum 5-HT concentration and cognitive dysfunction caused by intracerebroventricular injection of LPS. This provides a method for activating brain Tregs in vivo. Recent studies have revealed the involvement of certain gut microbiota, such as members of the Klebsiella, Escherichia, Streptococcus, and Enterococcus genera, in the synthesis of 5-HT [15, 31]. Previous studies on the gut microbiota composition in mice with cognitive impairment have primarily utilized 16S rRNA analysis, limited to the identification of microorganisms at the genus or family level. In our investigation, we employed metagenomic sequencing to comprehensively investigate the gut microbiota at the species level. Our findings revealed a relative increase in the abundance of specific bacterial species, such as s_Escherichia coli, s_Klebsiella_pneumoniae, s_Klebsiella_quasipneumoniae, and s_Klebsiella_variicola, in mice following microinjection of LPS. This provides a direction for future research on the impact of specific bacterial species on PND.

Tregs, by their ability to abrogate the pathogenic activities of immune cells and sustain immunological tolerance toward self-antigens, have garnered considerable attention over the years. Tregs exert immune suppressive effects through three main pathways: 1) inhibiting cytokine production or promoting effector T cell proliferation; 2) engaging in the direct secretion of cytokines, notably TGF-β and IL-10, thus orchestrating the cytokine milieu at the inflammatory site; and 3) directly killing cytotoxic cells [32]. Based on the systematic study of Tregs by Minako et al. [13], we proposed that the reduced activation of brain Tregs is due to a decrease in peripheral blood 5-HT levels. Consequently, when we increased the peripheral blood 5-HT concentration by providing the diet rich in tryptophan, we observed a significant improvement in cognitive function in mice following intracerebroventricular microinjection of LPS. Building upon the comprehensive investigation of Tregs by Minako et al. [13], we posited that the diminished activation of brain Tregs stems from reduced peripheral blood 5-HT levels, while augmenting peripheral blood 5-HT can stimulate the proliferation and activation of brain Tregs. Thus, we observed that mice receiving the tryptophan-rich diet to increase peripheral blood 5-HT levels after intracerebroventricular injection of LPS exhibited significantly greater proportions of brain Tregs in the cervical lymph nodes and hippocampus than mice receiving only intracerebroventricular injection of LPS. Additionally, there was an elevation in central anti-inflammatory factor levels, a reduction in inflammatory cell and proinflammatory factor levels, and alleviation of demyelination in neurons. Moreover, the cognitive function of the mice improved.

Current evidence suggests the potential interplay between immune dysregulation and cognitive decline in individuals with mild cognitive impairment or mild AD. Studies have revealed elevated levels of activated CD4⁺ T and CD8⁺ T cells in these individuals, indicating an aberrant immune response [33]. Notably, increased activation of CD8⁺ T cells has been linked to compromised language acquisition, visual-spatial abilities, and hippocampal degradation, which are hallmark features of cognitive impairment [34]. This finding points toward a potential association between the cognitive manifestations of AD and the heightened activation of CD8⁺ T cells. Furthermore, investigations have shown elevated quantities of effector memory CD8⁺ T cells in the CSF of AD patients, highlighting their involvement in central inflammation [35]. Previous investigations conducted by our research group and other scholars have revealed notable infiltration of CD8⁺ T cells within the cerebral parenchyma of mice afflicted with ischemic stroke. These CD8⁺ T-cell-derived Prf and GranzB compounds have been determined to play a significant role in neurotoxicity, exacerbating perioperative ischemic brain injury and subsequently leading to cognitive impairment [11, 36]. These findings suggested that CD8⁺ T cells play a pivotal role in mediating inflammation and cognitive impairment. In coculture experiments, 5-HT was shown to enhance the immunosuppressive effect of primary Tregs on primary CD8⁺ T cells through activation of the Htr7 receptor. These findings indicated that 5-HT amplifies the immunosuppressive effects of Tregs on primary CD8⁺ T cells through the activation of the Htr7 receptor.

The activation of microglia plays a key role in promoting neuroinflammation, a process implicated in various neurological disorders [37]. Currently, an increasing number of studies are related to the M1/M2 paradigm of microglial activation, wherein the M1 phenotype of microglia represents a proinflammatory state, while the M2 phenotype embodies an anti-inflammatory state [38]. Notably, exposure of microglia to LPS and sevoflurane, a commonly used anesthetic, stimulates the expression of the proinflammatory cytokines IL-1β and IL-6 [39]. Similarly, the administration of isoflurane, another anesthetic, has been shown to promote microglial inflammation and induce cognitive decline in elderly mice [40]. In an effort to mitigate these detrimental effects, researchers have focused on upregulating the expression of IL-10 in microglia, which can inhibit the NF-κB/MAPK pathway and alleviate POCD [41]. Additionally, the activation of 5-HT receptors has emerged as a promising strategy for limiting neuroinflammation. By reducing astrocyte and microglial reactivity, 5-HT receptor activation protects the brain from inflammation-induced neurodegenerative changes [42, 43]. The aforementioned studies primarily concentrated on the activation of hippocampal 5-HT receptors. In the present study, we investigated the importance of Htr7 receptors positioned on the surface of peripheral Tregs. We cocultured primary Tregs with primary microglia and found that in the 5-HT group, microglia exhibited a decrease in M1 polarization, which enhanced the inflammatory response, and an increase in M2 polarization, which promoted neural repair. Furthermore, the 5-HT group exhibited significant decreases in the concentrations of IL-1β and TNF-α. However, when 5-HT and Htr7 receptor antagonists were added to the culture medium, this enhanced immunosuppressive effect was blocked. These findings suggested that 5-HT enhances the immunosuppressive effect of primary Tregs on primary microglia through the Htr7 receptor.

The addition of 5-HT to culture media enhances the immunosuppressive effects of primary Tregs on cocultured primary CD8⁺ T cells and primary microglia. However, the simultaneous addition of 5-HT and an Htr7 receptor antagonist to the culture medium abolished the suppressive effects of primary CD8⁺ T cells and primary microglia on cocultured primary Tregs. These findings further substantiated the crucial immunomodulatory role of brain Tregs.

Rag1^−/− mice lack mature T and B cells. Following intravenous injection of primary Tregs, the impact of Tregs on neuroinflammation and cognitive function in Rag1^−/− mice can be observed independently, thereby eliminating interference from other T and B lymphocytes. By selectively transferring brain Tregs into Rag1^−/− mice in vivo, the attenuation of Htr7 expression on the surface of brain Tregs intensified the cognitive impairment and neuroinflammatory responses induced by the administration of LPS. These data support our conclusion that brain Tregs can alleviate the cognitive impairments caused by LPS-induced neuroinflammation.

There are several limitations in our study. First, the pathogenesis of PND is complex, and neuroinflammation is one of the major contributing factors. It would be beneficial to conduct further investigations using surgical models in aged mice. Second, dysbiosis of the gut microbiota may also contribute to neuroinflammation, and the involvement of Tregs in suppressing gut dysbiosis warrants further exploration. Third, proteomic analysis revealed numerous inflammation-related proteins, the synergistic effects of which on the onset and progression of neuroinflammation-induced PND remain elusive, as they are beyond the scope of this study.

In conclusion, our study demonstrated the pivotal role of brain Tregs (Htr7⁺ Tregs) in LPS-induced cognitive impairment, suggesting that brain Tregs may be critical for alleviating central inflammation-associated PND. Brain Tregs alleviated the inflammatory response and prevent neuronal damage by suppressing the infiltration of CD8⁺ T cells into the brain and excessive activation of microglia, thereby ameliorating LPS-induced cognitive impairment. This study may offer new potential therapeutic targets based on Tregs for PND.

PND: Perioperative neurocognitive disorders; Tregs: Regulatory T cells; LPS: Lipopolysaccharide; 5-HT: 5-hydroxytryptamine; HPLC‒MS: High-performance liquid chromatography‒mass spectrometry; Trp: Tryptophan; Rag1: Recombination activating gene 1; aCSF: Artificial cerebrospinal fluid; MWM: Morris water maze; SYN: Synephrine; DHBA: Dihydroxybenzylamine; CFSE: Carboxyfluorescein diacetate succinimidyl ester; Prf: Perforin; GranzB: Granzyme B; MBP: Myelin basic protein; SB: SB269970; LEfSe: Linear discriminant analysis coupled with effect size measurements; DG: Dentate gyrus; PLS-DA: Partial least-squares discrimination analysis; GO: Gene Ontology; iNOS: Inducible nitric oxide synthase; MFI: Mean fluorescence intensity.

Acknowledgments

Not applicable.

Author contributions

Conceptualization, LW and QF; Data curation, YXL and LYL; Funding acquisition, LW and QF; Methodology, DHX, XG and JJL; Project administration, LW and QF; Resources, JBC and YHL; Software, YXL and LYL; Validation, GSL; Writing – original draft, DHX, XG and JJL; Writing – review & editing, JSL, HL and WDM.

Funding

The present research was supported by the National Natural Science Foundation of China (No. 82071178, 82271322).

Availability of data and materials

The data presented in this study are available upon request from the corresponding authors.

Ethics approval and consent to participate All animal experiments were performed in accordance with the National Institute of Health Guide for Care and Use of Laboratory Animals, with the approval of the Ethics Committee for Animal Experimentation of the Chinese PLA General Hospital.

Consent for publication

Not applicable.

Competing interests

The authors declare no conflicts of interest.

Author details

¹Department of Anesthesiology, The First Medical Center, Chinese PLA General Hospital, Beijing 100853, China. ²Department of Anesthesiology, Beijing Tongren Hospital, Capital Medical University, Beijing, 100730, China. ³Department of Anesthesiology, Chinese People's Armed Police Force Hospital of Beijing, Beijing 100027, China. ⁴Department of Pain Medicine, The First Medical Center, Chinese PLA General Hospital, Beijing 100853, China.

Takazawa T, Horiuchi T, Orihara M, Nagumo K, Tomioka A, Ideno Y, et al. Prevention of Postoperative Cognitive Dysfunction by Minocycline in Elderly Patients after Total Knee Arthroplasty: A Randomized, Double-blind, Placebo-controlled Clinical Trial. Anesthesiology. 2023;138:172–83.
Moller JT, Cluitmans P, Rasmussen LS, Houx P, Rasmussen H, Canet J, et al. Long-term postoperative cognitive dysfunction in the elderly ISPOCD1 study. ISPOCD investigators. International Study of Post-Operative Cognitive Dysfunction. Lancet. 1998;351:857–61.
Liu Q, Sun Y-M, Huang H, Chen C, Wan J, Ma L-H, et al. Sirtuin 3 protects against anesthesia/surgery-induced cognitive decline in aged mice by suppressing hippocampal neuroinflammation. J Neuroinflamm. 2021;18:41.
Huang H, Chou J, Tang Y, Ouyang W, Wu X, Le Y. Nomogram to predict postoperative cognitive dysfunction in elderly patients undergoing gastrointestinal tumor resection. Front Aging Neurosci. 2022;14:1037852.
Ma G, Chen C, Jiang H, Qiu Y, Li Y, Li X, et al. Ribonuclease attenuates hepatic ischemia reperfusion induced cognitive impairment through the inhibition of inflammatory cytokines in aged mice. Biomedicine & Pharmacotherapy = Biomedecine & Pharmacotherapie. 2017;90:62–8.
Luo A, Yan J, Tang X, Zhao Y, Zhou B, Li S. Postoperative cognitive dysfunction in the aged: the collision of neuroinflammaging with perioperative neuroinflammation. Inflammopharmacology. 2019;27:27–37.
Liang Y, Kang X, Zhang H, Xu H, Wu X. Knockdown and inhibition of hippocampal GPR17 attenuates lipopolysaccharide-induced cognitive impairment in mice. J Neuroinflammation. 2023;20:271.
Liang X, Xue Z, Zheng Y, Li S, Zhou L, Cao L, et al. Selenium supplementation enhanced the expression of selenoproteins in hippocampus and played a neuroprotective role in LPS-induced neuroinflammation. Int J Biol Macromol. 2023;234:123740.
Sun M, Li Y, Liu M, Li Q, Shi L, Ruan X, et al. Insulin alleviates lipopolysaccharide-induced cognitive impairment via inhibiting neuroinflammation and ferroptosis. European Journal of Pharmacology. 2023;955:175929.
Barbi J, Pardoll D, Pan F. Treg functional stability and its responsiveness to the microenvironment. Immunol Rev. 2014;259:115–39.
Zhang F, Niu M, Guo K, Ma Y, Fu Q, Liu Y, et al. The immunometabolite S-2-hydroxyglutarate exacerbates perioperative ischemic brain injury and cognitive dysfunction by enhancing CD8⁺T lymphocyte-mediated neurotoxicity. J Neuroinflamm. 2022;19:176.
Tian A, Ma H, Cao X, Zhang R, Wang X, Wu B. Vitamin D improves cognitive function and modulates Th17/T reg cell balance after hepatectomy in mice. Inflammation. 2015;38:500–9.
Ito M, Komai K, Mise-Omata S, Iizuka-Koga M, Noguchi Y, Kondo T, et al. Brain regulatory T cells suppress astrogliosis and potentiate neurological recovery. Nature. 2019;565:246–50.
Margolis KG, Cryan JF, Mayer EA. The Microbiota-Gut-Brain Axis: From Motility to Mood. Gastroenterology. 2021;160:1486–501.
Wu W, Kong Q, Tian P, Zhai Q, Wang G, Liu X, et al. Targeting Gut Microbiota Dysbiosis: Potential Intervention Strategies for Neurological Disorders. Engineering. 2020;6:415–23.
Wang D, Wu J, Zhu P, Xie H, Lu L, Bai W, et al. Tryptophan-rich diet ameliorates chronic unpredictable mild stress induced depression- and anxiety-like behavior in mice: The potential involvement of gut-brain axis. Food Res Int. 2022;157:111289.
Nemoto W, Kozak D, Sotocinal SG, Tansley S, Bannister K, Mogil JS. Monoaminergic mediation of hyperalgesic and analgesic descending control of nociception in mice. Pain. 2023;164:1096–105.
Zhao W, Xu Z, Cao J, Fu Q, Wu Y, Zhang X, et al. Elamipretide (SS-31) improves mitochondrial dysfunction, synaptic and memory impairment induced by lipopolysaccharide in mice. J Neuroinflamm. 2019;16:230.
Wang Q, Yang Q, Liu X. The microbiota-gut-brain axis and neurodevelopmental disorders. Protein Cell. 2023;14:762–75.
Yano JM, Yu K, Donaldson GP, Shastri GG, Ann P, Ma L, et al. Indigenous Bacteria from the Gut Microbiota Regulate Host Serotonin Biosynthesis. Cell. 2015;161:264–76.
Schroer AB, Ventura PB, Sucharov J, Misra R, Chui MKK, Bieri G, et al. Platelet factors attenuate inflammation and rescue cognition in ageing. Nature. 2023;620:1071–9.
Leiter O, Brici D, Fletcher SJ, Yong XLH, Widagdo J, Matigian N, et al. Platelet-derived exerkine CXCL4/platelet factor 4 rejuvenates hippocampal neurogenesis and restores cognitive function in aged mice. Nat Commun. 2023;14:4375.
Meier MH, Caspi A, R Knodt A, Hall W, Ambler A, Harrington H, et al. Long-Term Cannabis Use and Cognitive Reserves and Hippocampal Volume in Midlife. Am J Psychiatry. 2022;179:362–74.
Wang P, Yin X, Chen G, Li L, Le Y, Xie Z, et al. Perioperative probiotic treatment decreased the incidence of postoperative cognitive impairment in elderly patients following non-cardiac surgery: A randomised double-blind and placebo-controlled trial. Clin Nutr. 2021;40:64–71.
Sun L, Yong Y, Wei P, Wang Y, Li H, Zhou Y, et al. Electroacupuncture ameliorates postoperative cognitive dysfunction and associated neuroinflammation via NLRP3 signal inhibition in aged mice. CNS Neurosci Ther. 2022;28:390–400.
Netto MB, De Oliveira Junior AN, Goldim M, Mathias K, Fileti ME, Da Rosa N, et al. Oxidative stress and mitochondrial dysfunction contributes to postoperative cognitive dysfunction in elderly rats. Brain, Behavior, and Immunity. 2018;73:661–9.
Granger KT, Barnett JH. Postoperative cognitive dysfunction: an acute approach for the development of novel treatments for neuroinflammation. Drug Discovery Today. 2021;26:1111–4.
Kant IMJ, De Bresser J, Van Montfort SJT, Slooter AJC, Hendrikse J. MRI Markers of Neurodegenerative and Neurovascular Changes in Relation to Postoperative Delirium and Postoperative Cognitive Decline. The American Journal of Geriatric Psychiatry. 2017;25:1048–61.
Ohkura N, Sakaguchi S. Transcriptional and epigenetic basis of Treg cell development and function: its genetic anomalies or variations in autoimmune diseases. Cell Res. 2020;30:465–74.
Höglund E, Øverli Ø, Winberg S. Tryptophan Metabolic Pathways and Brain Serotonergic Activity: A Comparative Review. Frontiers in Endocrinology. 2019;10:158.
O’Mahony SM, Clarke G, Borre YE, Dinan TG, Cryan JF. Serotonin, tryptophan metabolism and the brain-gut-microbiome axis. Behav Brain Res. 2015;277:32–48.
Yang Z, Yu A, Liu Y, Shen H, Lin C, Lin L, et al. Regulatory T cells inhibit microglia activation and protect against inflammatory injury in intracerebral hemorrhage. Int Immunopharmacol. 2014;22:522–5.
Reagin KL, Funk KE. The Role of Antiviral CD8⁺ T cells in Cognitive Impairment. Current opinion in neurobiology. 2022;76:102603.
Lueg G, Gross CC, Lohmann H, Johnen A, Kemmling A, Deppe M, et al. Clinical relevance of specific T-cell activation in the blood and cerebrospinal fluid of patients with mild Alzheimer’s disease. Neurobiol Aging. 2015;36:81–9.
Gate D, Saligrama N, Leventhal O, Yang AC, Unger MS, Middeldorp J, et al. Clonally expanded CD8 T cells patrol the cerebrospinal fluid in Alzheimer’s disease. Nature. 2020;577:399–404.
Lee PR, Johnson TP, Gnanapavan S, Giovannoni G, Wang T, Steiner JP, et al. Protease-activated receptor-1 activation by granzyme B causes neurotoxicity that is augmented by interleukin-1β. J Neuroinflamm. 2017;14:131.
Leng F, Edison P. Neuroinflammation and microglial activation in Alzheimer disease: where do we go from here? Nat Rev Neurol. 2021;17:157–72.
Lee JH, Kam EH, Kim SY, Cheon SY, Kim EJ, Chung S, et al. Erythropoietin Attenuates Postoperative Cognitive Dysfunction by Shifting Macrophage Activation toward the M2 Phenotype. Front Pharmacol. 2017;8:839.
Ye X, Lian Q, Eckenhoff MF, Eckenhoff RG, Pan JZ. Differential general anesthetic effects on microglial cytokine expression. PloS One. 2013;8:e52887.
Wang Z, Meng S, Cao L, Chen Y, Zuo Z, Peng S. Critical role of NLRP3-caspase-1 pathway in age-dependent isoflurane-induced microglial inflammatory response and cognitive impairment. J Neuroinflamm. 2018;15:109.
Zhang D, Li N, Wang Y, Lu W, Zhang Y, Chen Y, et al. Methane ameliorates post-operative cognitive dysfunction by inhibiting microglia NF-κB/MAPKs pathway and promoting IL-10 expression in aged mice. Int Immunopharmacol. 2019;71:52–60.
Bokobza C, Jacquens A, Guenoun D, Bianco B, Galland A, Pispisa M, et al. Targeting the brain 5-HT7 receptor to prevent hypomyelination in a rodent model of perinatal white matter injuries. J Neural Transm. 2023;130:281–97.
Costa L, Tempio A, Lacivita E, Leopoldo M, Ciranna L. Serotonin 5‐HT7 receptors require cyclin‐dependent kinase 5 to rescue hippocampal synaptic plasticity in a mouse model of Fragile X Syndrome. Eur J Neurosci. 2021;54:4124–32.

No competing interests reported.

GraphicalAbstract.jpg

Download PDF

Journal Publication

published 27 Sep, 2024

Read the published version in Journal of Neuroinflammation →

Editorial decision: Revision requested
22 Feb, 2024
Reviews received at journal
04 Feb, 2024
Reviewers agreed at journal
24 Jan, 2024
Reviewers invited by journal
24 Jan, 2024
Editor assigned by journal
24 Jan, 2024
Submission checks completed at journal
24 Jan, 2024
First submitted to journal
23 Jan, 2024

You are reading this latest preprint version

Tryptophan-rich diet and its effects on brain Tregs in alleviating neuroinflammation and cognitive impairment

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials and methods

Animals

Intracerebroventricular Microinjection Surgery

Experimental diets

MWM test

Primary cell coculture

Flow Cytometry

Lentiviral transfection of Tregs

Adoptive cell transfer in Rag1-/- mice

Western blotting

Antibodies for immunofluorescence staining

Quantitative real-time PCR (qRT‒PCR)

Metagenomic sequencing

TMT-based quantitative proteomics analysis

Enzyme-linked immunosorbent assay (ELISA)

Statistical analyses

Results

Tryptophan-rich diet reversed LPS-induced cognitive impairment and decreased 5-HT in peripheral blood

The abundance of 5-HT-related gut microbiota declined after LPS treatment

Proteomic alterations in the hippocampus of mice induced by intracerebroventricular administration of LPS

Tryptophan-rich diet attenuated LPS-mediated neuroinflammation by activating brain Tregs

5-HT and Htr7 enhanced the immunosuppressive effect of Tregs on CD8+ T cells and microglia

Brain Tregs alleviated LPS-induced cognitive impairment in Rag1-/- mice

Brain Tregs ameliorated LPS-induced neuroinflammation in Rag1-/- mice

Discussion

Conclusions

Abbreviations

Declarations

Acknowledgments

Author contributions

Funding

Availability of data and materials

Consent for publication

Competing interests

Author details

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1

Adoptive cell transfer in Rag1^-/- mice

5-HT and Htr7 enhanced the immunosuppressive effect of Tregs on CD8⁺ T cells and microglia

Brain Tregs alleviated LPS-induced cognitive impairment in Rag1^-/- mice

Brain Tregs ameliorated LPS-induced neuroinflammation in Rag1^-/- mice