Combined and Separate Pretreatments with L-Theanine and Aerobic Exercise Modulate Cognitive Decline Following Chronic Neuroinflammation in Rats Exposed to Lipopolysaccharide

TL;DR

This study shows that pretreating rats with L-theanine and aerobic exercise can help reduce cognitive decline caused by chronic brain inflammation.

Contribution

The study demonstrates the combined and individual effects of L-theanine and aerobic exercise in mitigating cognitive impairments from neuroinflammation.

Findings

Pretreatment with L-theanine or exercise improved recognition memory and CREB phosphorylation in rats with LPS-induced inflammation.

Exercise reduced Aβ1–42 levels and neuroinflammatory cytokines, but L-theanine did not.

Combined treatment showed additive benefits for some cognitive measures but not spatial memory.

Abstract

Chronic neuroinflammation is a prominent feature of several central nervous system disorders and contributes significantly to cognitive impairment. The present study aimed to investigate the effects of pretreatment with L-theanine (LT), aerobic exercise (ex), and their combination on cognitive deficits induced by subchronic lipopolysaccharide (LPS) administration in rats. Male Wistar rats were assigned to the following groups: control; veh-sed-LPS, sedentary (sed) rats treated with vehicle (veh) and LPS; LT-sed-LPS; veh-ex-LPS; and LT-ex-LPS. L-theanine treatment and/or treadmill running were administered for 5 weeks. Following these interventions, neuroinflammation was induced by LPS injections for 7 days, while the control group received veh treatment. Cognitive function was assessed using Y-maze, object recognition, and object location tests. Hippocampal cAMP response element-binding protein (CREB) phosphorylation status, β-amyloid (Aβ1–42) accumulation, and pro-inflammatory cytokines (TNF-α, IL-1β) were measured by ELISA. Pretreatment with LT, ex, or their combination improved Y-maze performance and recognition memory, partly restoring the LPS-induced reduction in the pCREB/CREB ratio. Exercise, but not LT, reduced Aβ1–42 levels and neuroinflammatory cytokine expression. Combined treatment produced additive benefits for some cognitive measures but not for spatial memory. These findings suggest that the prophylactic combination of LT and ex can partially attenuate cognitive impairments associated with subchronic neuroinflammation in this model.

Figures

Full text

1. Introduction

Neuroinflammation has emerged as a critical factor in the development of various neuropsychiatric disorders, including multiple sclerosis [1], stroke [2], myalgic encephalomyelitis/chronic fatigue syndrome [3], major depressive disorder [4], schizophrenia [5], anxiety disorders [6] and certain forms of obsessive–compulsive disorder [7]. Furthermore, substantial research has highlighted the contribution of neuroinflammatory processes to the deterioration of cognitive and other higher-order brain functions across a range of neurological conditions. Neuroinflammation is frequently associated with neurodegenerative disorders such as Alzheimer’s disease (AD), Parkinson’s disease, and amyotrophic lateral sclerosis, where it is thought to contribute to disease progression and cognitive decline [8].

Chronic neuroinflammation is characterized by continuous microglial activation, leading to increased production of pro-inflammatory mediators within the central nervous system [8]. Dysregulation of these inflammatory processes can disrupt neuronal function and impair cognitive performance, emphasizing the importance of understanding neuroinflammation as a modifiable contributor to cognitive dysfunction.

Inflammatory processes in the central nervous system can exert both adaptive and maladaptive effects. While an acute inflammatory response may facilitate resolution and recovery, sustained or dysregulated inflammation has been associated with alterations in neuronal function and cognitive performance [8]. This functional duality underscores the importance of investigating interventions that may modulate cognitive impairment associated with prolonged inflammatory states.

Aerobic exercise (ex) is widely used as a non-pharmacological intervention in experimental models examining cognitive impairment associated with inflammatory challenges, including lipopolysaccharide (LPS) exposure. Previous studies have shown that aerobic exercise ameliorates LPS-induced learning and memory impairments and reduces associated neuroinflammatory responses in rodents [9].

L-theanine (LT), a non-proteinogenic amino acid abundant in tea leaves, has garnered attention for its reported effects on cognition and behavior. Previous animal studies have shown that LT administration can modulate cognitive performance in models involving inflammatory or stress-related challenges, including LPS-based paradigms [10,11,12].

Although a growing body of evidence supports the individual neuroprotective effects of LT and ex in experimental models involving inflammatory challenges, their combined prophylactic effects on LPS-induced cognitive impairment have not been systematically evaluated. Previous studies have largely examined these interventions in isolation, making it unclear whether concurrent administration confers comparable or additional benefits for cognitive function. Therefore, the present study aimed to assess and compare the effects of LT, ex, and their combined administration on LPS-induced cognitive impairment in young adult rats under identical experimental conditions. By focusing on behavioral outcomes in a validated model of subchronic neuroinflammation, this study emphasizes the functional cognitive impact of these non-pharmacological interventions, while assessing selected hippocampal molecular markers associated with inflammation and CREB-mediated signaling.

2. Results

The experimental design is described in Figure 1.

2.1. Body Weight Measured During Prophylactic Pretreatment with L-Theanine (LT) and/or Physical Exercise (ex) and During the Lipopolysaccharide (LPS) Treatment

Weight changes in rats resulting from either ex and LT treatment or LPS exposure were monitored both to ensure overall health and assess potential effects of the interventions (Figure 2).

The body weight significantly decreased as a result of LPS injection for 7 days in young adult rats (p < 0.05, veh-sed-LPS vs. control) (Figure 3). Furthermore, the treatment with LT, ex or their combination was unable to restore LPS-induced body weight drop to control levels.

2.2. Working and Spatial Short-Term Memory, Measured in the Y-Maze Test

A significant difference between groups was found for spontaneous alternation behavior (SAB) in young adult rats (p < 0.001). The 3-month-old veh-sed-LPS rats showed reduced SAB compared to the control group (p < 0.001) (Figure 4A). The three pretreatment procedures corrected the LPS-induced working memory impairment in the first protocol of the Y-maze test (p < 0.001, LT-sed-LPS vs. veh-sed-LPS; p < 0.001, veh-ex-LPS vs. veh-sed-LPS; p < 0.001, LT-ex-LPS vs. veh-sed-LPS).

Motor activity was assessed by the number of arm entries in the first protocol of the Y-maze. There was a significantly reduced number of arm entries in the veh-sed-LPS groups compared to the control group (p < 0.001) (Figure 4B).

Significant between-group differences were found for the second protocol of the Y-maze (time) (p < 0.001) (count) (p = 0.007). In the young adult rats, the veh-sed-LPS group showed impaired short-term memory compared to the control group (Discrimination Index (DI) time and count: p < 0.001) (Figure 5A,B). However, only the pre-exercise training procedure had a beneficial effect on LPS-induced short-term memory impairment in 3-month-old rats (p = 0.008 vs. veh-sed-LPS group) (Figure 5A).

2.3. Short-Term Spatial Memory Measured in the Object Location Test (OLT)

As in the Y-maze test, significant between-group differences were found for the spatial memory measured in the OLT (DI time %: p < 0.001 and DI count %: p = 0.003). The veh-sed-LPS group was characterized by impaired spatial memory compared to the control group, (p < 0.001) (Figure 6A) and (p = 0.0011) (Figure 6B), respectively. Pretreatment with LT and the combination LT-ex attenuated LPS-induced memory decline (p < 0.001, vs. veh-sed-LPS, Figure 6A; p = 0.003 and p < 0.001, vs. veh-sed-LPS; Figure 6B, respectively).

2.4. Recognition Short-Term Memory Measured in the Object Recognition Test (ORT)

When recognition memory was assessed 60 min after the acquisition phase, analysis of the ORT revealed significant differences between groups in young adult rats (p = 0.010). The veh-sed-LPS group showed impaired recognition memory compared to the control group, (p = 0.0247) (DI time %) (Figure 7A) and (p = 0.003) (DI count %) (Figure 7B), respectively. Pretreatment with LT, ex and the combination ex-LT showed similar efficacy in alleviating the LPS-induced decline in short-term memory as measured by the ORT (p < 0.001; p = 0.0378; p = 0.0137, respectively) (Figure 7A).

2.5. Effect of LT, ex and Their Combination (LT-ex) on LPS-Induced Changes in pCREB/CREB Ratio in the Hippocampus

LPS administration for 7 days induced a significant decrease in the pCREB/CREB ratio in the hippocampus of young adult rats compared to the control group (p < 0.05, veh-sed-LPS vs. control) (Figure 8).

Both prophylactic treatments, LT and ex, had a beneficial effect on the LPS-reduced pCREB/CREB ratio in 3-month-old rats (p < 0.05, LT-veh-LPS vs. veh-sed-LPS; p < 0.05, veh-ex-LPS vs. veh-sed-LPS). However, the combination of the two treatment approaches was ineffective (p > 0.05, LT-ex-LPS vs. veh-sed-LPS).

2.6. Effect of LT, ex and LT-ex on LPS-Induced Changes in Aβ1–42 Levels in the Hippocampus

Subchronic LPS treatment increased Aβ_1–42_ expression in the hippocampus of 3-month-old rats (p < 0.001, veh-sed-LPS vs. C) (Figure 9). Exercise prior to LPS with veh or in combination with LT prevented the toxin-induced increase in Aβ_1–42_ levels in young adult rats (p = 0.0036, veh-ex-LPS vs. veh-sed-LPS; p = 0.0261, LT-ex-LPS vs. veh-sed-LPS) (Figure 9).

2.7. Effect of Treatment with LT, ex, and LT-ex of the Two Approaches (Pharmacological and Alternative) on LPS-Induced Inflammation in the Hippocampus and Serum

The inflammatory response induced by LPS injection for 7 days was expressed by significantly elevated pro-inflammatory markers interleukin (IL)-β and tumor-necrotic factor (TNF)-α in the hippocampus (p < 0.05, LPS-sed-veh vs. control) (Figure 10A,B). In addition, C-reactive protein (CRP) was also significantly elevated in the serum of the LPS-treated control group (p < 0.05, LPS-sed-veh vs. control) (Figure 10C). The prophylactic ex pretreatment and its combination with LT mitigated LPS-induced elevation of IL-β in the hippocampus (p < 0.05, vs. veh-sed-LPS) (Figure 10A). The LPS-induced elevation of TNF-α in the hippocampus was attenuated by the combination LT + ex (p < 0.01, LT-ex-LPS vs. veh-sed-LPS) (Figure 10B). Furthermore, the CRP level in the serum was reduced by the LT treatment (p < 0.01, LT-sed-LPS vs. veh-sed-LPS) (Figure 10C).

3. Discussion

In the present study, we found that prophylactic treatment with LT, ex, or their combination improved several indices of cognition in rats subjected to subchronic LPS administration. LPS exposure produced robust neuroinflammatory responses (increased hippocampal IL-1β and TNF-α, elevated serum CRP), impaired SAB, recognition memory, and short-term spatial memory, reduced exploratory activity, and was accompanied by a reduction in the hippocampal pCREB/CREB ratio and an increase in hippocampal Aβ_1–42_. Overall, LT and ex each produced partial behavioral recovery, with ex but not LT associated with lower hippocampal Aβ_1–42_ and reduced hippocampal IL-1β; LT reduced serum CRP but did not alter hippocampal cytokines. The combined treatment conferred some additional task-specific benefits but did not produce broadly synergistic effects across endpoints. These findings demonstrate that both interventions can ameliorate select functional deficits following an inflammatory challenge, while also highlighting differences in their measurable molecular correlates.

Beyond classical toll-like receptor 4 signaling, LPS activates both canonical and non-canonical innate immune pathways that contribute to neuroinflammation and downstream neural dysfunction. Intracellular LPS can engage inflammatory caspases, particularly caspase-4 (the human equivalent of murine caspase-11), initiating inflammasome-linked signaling and amplifying cytokine release and pyroptotic responses in neural tissue [13,14]. Canonical inflammasome activation, such as NLR family pyrin domain containing 3-dependent signaling, has similarly been implicated in LPS-driven neuroinflammatory pathology and cognitive impairment [15]. Furthermore, CREB has been identified as a caspase substrate in neuronal cells, providing a possible mechanistic connection between inflammatory signaling and the reduced hippocampal pCREB/CREB ratio [16].

Previous studies have established that phosphorylation of CREB at Ser133 links neuronal activity with transcriptional programs important for synaptic plasticity and memory consolidation [17,18,19,20]. Reduced hippocampal pCREB has been observed in multiple models of inflammatory challenge and is commonly associated with impaired learning and memory [21,22]. In the present study, pretreatment with LT and/or ex was associated with partial restoration of the pCREB/CREB ratio alongside behavioral improvements, a result consistent with prior reports that aerobic training and other interventions can increase hippocampal CREB phosphorylation and support cognitive function [23,24,25]. That said, pCREB should be interpreted here as a supportive molecular correlate rather than definitive evidence of a CREB-mediated causal mechanism; we measured phosphorylation by ELISA and did not assay downstream transcriptional targets (e.g., BDNF, Arc, cFos) or cell-type-specific responses, which would be required to test causal links between CREB activation and synaptic or structural plasticity [17,18,19,26].

Our observation that LPS increased hippocampal Aβ_1–42_ is consistent with data showing that systemic inflammatory stimuli can promote amyloidogenic pathways by upregulating secretase activity and impairing Aβ clearance mechanisms [22,27,28]. LPS-induced elevations in Aβ can reflect transient shifts in amyloid-beta precursor protein processing or impaired efflux/clearance [22,28] and may contribute to short-term cognitive dysfunction in non-degenerative contexts. In the present model, ex pretreatment reduced hippocampal Aβ_1–42_ levels, while LT did not, suggesting that exercise may more directly influence amyloidogenic processes or clearance. However, we did not evaluate tau pathology, progressive synaptic loss, or longitudinal accumulation of Aβ; therefore, caution is warranted in extrapolating these findings to neurodegenerative disease mechanisms.

Although both interventions improved behavioral outcomes, their molecular fingerprints differed. Exercise attenuated hippocampal Aβ_1–42_ and reduced IL-1β in the hippocampus, whereas LT lowered systemic CRP without detectable reductions in hippocampal Aβ_1–42_ or cytokines. Such dissociation suggests that LT and ex may act through partially distinct processes: LT may modulate systemic inflammatory tone or stress responsivity, thereby indirectly influencing brain function, while ex may engage mechanisms that more directly alter central inflammatory signaling and amyloid handling. These interpretations are consistent with prior reports describing exercise-related modulation of neurotrophic and metabolic pathways and with studies reporting LT effects on stress responses and inflammation in other models [10,23,24,25,29]. Importantly, this study was not designed to delineate precise molecular pathways; rather, it provides comparative functional evidence that informs hypotheses for future mechanistic work.

The combination of LT and ex produced some additive benefits in specific tasks. Still, they did not show broad synergistic restoration of all endpoints (notably, pCREB was not more robustly restored by the combination than by each intervention alone). This non-additive outcome is informative: it suggests that concurrent prophylaxis does not necessarily produce linear enhancement across molecular and behavioral endpoints and may reflect ceiling effects, timing or dosing interactions, or overlapping functional targets. We have reported this negative result transparently and discussed potential explanations (e.g., interference, saturation of specific pathways, or differential central vs. peripheral actions) as hypotheses rather than conclusions. Clarifying the interaction between these interventions will require studies explicitly powered to test interaction effects and incorporate time-course, dose–response, and mechanistic readouts.

Study Limitations

This study has several limitations. First, only male rats were used to minimize biological variability; however, sex differences in neuroinflammatory responses and in the effects of exercise or nutraceutical interventions are well documented. Future studies should include both sexes to assess sex-specific effects. Second, a single subchronic LPS protocol was employed, which models inflammation-associated cognitive disruption but does not capture other sources of chronic neuroinflammation, such as aging, neurodegenerative disease, or brain injury. Third, interventions were administered for five weeks, with cognitive testing conducted shortly thereafter; longer treatment periods and delayed or longitudinal assessments will be needed to determine the persistence of effects. Fourth, dose–response relationships for LT and systematic variation in exercise intensity or duration were not examined and warrant future investigation. Finally, molecular analyses were limited in scope: CREB phosphorylation was assessed as a bulk hippocampal measure, without evaluation of downstream CREB-dependent targets, synaptic markers, or direct indices of neurogenesis. Likewise, hippocampal Aβ_1–42_ was measured as an inflammation-sensitive biochemical marker rather than as evidence of progressive neurodegenerative pathology. Although hippocampal CREB activation and inflammatory markers were assessed, direct measures of synaptic plasticity and neurogenesis were beyond the scope of the present study and should be addressed in future investigations.

4. Materials and Methods

4.1. Experimental Animals

Young 3-month-old male Wistar rats (240–325 g, n = 55), were used in the experiment. Animals were obtained and housed in the vivarium of the Department of Physiology, Medical University of Plovdiv. The rats were acclimated to the housing facility for one week prior to the start of the experiment. Animals were group-housed (five animals per cage) with ad libitum access to water and standard chow, under a 12 h light/12 h dark cycle and a room temperature of 23 ± 1 °C. All procedures conformed to the European Commission recommendations for the protection and welfare of laboratory animals and were approved by the Committee of Scientific Ethics at the Medical University of Plovdiv and the Committee of Ethical Treatment of Animals at the Bulgarian Food Safety Agency (BFSA).

4.2. Grouping

Rats were randomly assigned to one of five experimental groups:

• control: sedentary (sed) rats treated with vehicle (veh);
• veh-sed-LPS: sedentary rats pretreated with vehicle followed by LPS injections;
• LT-sed-LPS: sedentary rats treated with LT followed by LPS injections;
• veh-ex-LPS: exercised rats treated with vehicle followed by LPS injections;
• LT-ex-LPS: exercised rats treated with LT followed by LPS injections.

4.3. Treatment with L-Theanine

The rats from LT-sed-LPS and LT-ex-LPS groups were treated with LT (4 mg/kg/day; p.o.) for 5 weeks. The selected dose falls within the 2–4 mg/kg/day range shown to produce cognitive and neuroprotective effects in prior rodent studies and has been used safely in comparable experimental paradigms [30,31,32]. In the groups not receiving LT, animals underwent a matched oral administration procedure using saline (veh).

4.4. Aerobic Training

Before the experiments, screening tests were performed to replace the individuals that refused to run on the treadmill. Owing to the fact that running on a treadmill is a skill that must be acquired by the experimental animals, prior to the commencement of the experiment, all rats were subjected to a daily regimen of 5 min sessions on the EXER-3R-Treadmill for small experimental animals (Columbus Instruments, Columbus, OH, USA), three days per week for a period of two weeks. This has the advantage of enabling the animals to adjust to the task of running on a treadmill without causing any significant adaptive changes. It also facilitates the exclusion of rats who are unable to run. According to both best practices and our previous experience, approximately 15% of Wistar rats refuse to run on a treadmill, regardless of the training process, and at the end of the preparatory period these animals are removed and replaced from the experiment [33].

All the rats from groups veh-ex-LPS and LT-ex-LPS were subjected to dosed systemic submaximal exercise on a treadmill at a fixed treadmill speed of 16 m/min and an incline of 5° (about 50–60% VO_2max_), 5 days per week for 5 weeks. On the first day, the training duration was 20 min and gradually increased by 5 min every day. At the end of the second week, it reached 40 min, and this duration was maintained until the end of the experiment. The exercise intensity was set below the previously established maximal lactate steady state for treadmill running in Wistar rats [34], indicating that the training is aerobic. Our previous data show that the same intensity of aerobic training has an enhancing effect on learning and memory performance and increases the hippocampal BDNF expression in rats [34].

4.5. Induction of Chronic Neuroinflammation Via LPS Administration

Following the pretreatment phase, animals in all groups except control received intraperitoneal injections of LPS (250 µg/kg) once daily for 7 consecutive days, beginning 24 h after the last LT administration and/or ex session. Control animals received vehicle injections on the same schedule. This subchronic LPS regimen has been widely used to induce sustained neuroinflammatory responses and cognitive alterations without causing overt systemic toxicity in rodents [22,32,35].

4.6. Behavioral Tests

Following the induction of neuroinflammation, the rats underwent a comprehensive learning and memory training program. This program commenced 5 days after the start of LPS induction with the first Y-maze trial. One day later, the object recognition test was conducted, followed by the object location test two days after that. Five days subsequent to the object location test, the second Y-maze trial was performed.

4.6.1. Y-Maze Test

The Y-maze apparatus comprises three steel arms that are positioned at an angle of 120° relative to one another. In the initial trial, which assessed working memory, the rats were allowed to roam freely through the arms for a period of 8 min. Two impartial observers, who were blind to the experiment, diligently documented the animals’ entry patterns. The SAB was calculated by taking into account the number of visits to the different arms. SAB (%) = (correct entries × 100)/(total entries − 2). In the second study, a pretest was administered at least five days later. One arm was closed off, leaving only two arms available for exploration for periods of 5, 10, and 15 min each. After a 30 min break, the rat was placed in the arm that had not yet been explored. The time spent and number of entries into both the familiar and novel arms were carefully measured and documented. The DI, calculated as (N)/(N + F), assessed the preference for the novel arm in the Y-maze test.

4.6.2. Object Recognition Test

For the ORT, we followed the methodology outlined in our previous study [36]. After 24 h of familiarization with an empty open field apparatus of 50 × 50 × 50 cm^3^, rats were placed next to two identical plastic objects labeled ‘F’ (familiar) for a 5 min training period. To facilitate habituation, each rat was placed alone in the box for 10 min the day before the training period. After a 60 min interval, the rats were returned to the same box for the test phase, in which one object was replaced by a novel object labeled ‘N,’ also lasting 5 min. During both phases, the time spent exploring each object and the number of sniffs were carefully observed and recorded. Exploration time was quantified in seconds, and the number of sniff instances was recorded for each object. The DI, calculated as (N)/(N + F), was used to assess recognition performance. This calculation was performed before and after any experimental treatments to assess changes in cognitive function.

4.6.3. Object Location Test

The OLT entails several critical stages for evaluating spatial memory in animals. During the adaptation phase, each animal is placed in a 50 × 50 cm enclosure without cameras and is allowed to become familiar with its surroundings for 10 min without being timed. The animals are then moved to a dimly lit test room 30 min prior to the start of the test. The test itself occurs 24 h after the pretest. Two identical objects are positioned in opposite corners of a 50 × 50 cm enclosure, and the animals are granted 5 min to freely explore and interact with them. After a 60 min pause, the animals progress to the test phase. One of the objects is relocated to the opposite corner, creating a familiar and novel location. The animals’ behavior during this 5 min period is closely observed and documented. Parameters such as the number and duration (in seconds) of sniffs at both familiar and novel locations are among those recorded. Additionally, the DI is calculated using the formula DI = (NO × 100)/(NO + FO). This index provides valuable insight into the animals’ capacity to differentiate between familiar and novel object locations. To ensure consistency and eliminate potential confounding factors, such as odor, the maze arms are wiped with alcohol between each test.

4.7. Biochemical Analysis

Following the completion of behavioral testing, rats were humanely euthanized via carbon dioxide inhalation. The hippocampi were rapidly dissected under cold conditions, immediately flash-frozen in liquid nitrogen, and stored at −80 °C until biochemical analysis.

4.7.1. Tissue Homogenization

Hippocampal tissue samples were homogenized following a protocol adapted from our previous studies [37]. Briefly, 10% (w/v) tissue homogenates were prepared in 0.1 M phosphate buffer (pH 7.4). The homogenates were then centrifuged at 10,000× g for 15 min at 4 °C. The resulting supernatants were carefully separated and used for subsequent biochemical analyses.

4.7.2. Enzyme-Linked Immunosorbent Assay (ELISA)

The expression levels of pCREB, total CREB (CREB1), Aβ_1–42_, IL-β and TNF-α were quantified using commercially available ELISA kits (Elabscience: Rat Aβ_1–42_ (Amyloid Beta_1–42_) ELISA Kit, E-EL-R1402; Rat CREB1 (Cyclic AMP Response Element Binding Protein 1) CLIA Kit, E-CL-R0192; Rat Phospho cAMP response element binding protein (P-CREB) ELISA Kit). All assays were performed according to the manufacturer’s instructions. For pCREB, and CREB1, results were expressed as ng/mg of total protein. Total protein content was determined using the Bradford method [38]. All biochemical measurements were performed in duplicate, and the mean values were used for statistical analysis.

4.7.3. Determination of C-Reactive Protein

The expression of CRP was quantified via immunoturbidimetric biochemical kit in blood plasma (Fortress diagnostics) according to the manufacturer’s instructions.

4.8. Statistical Analysis

The experimental data are presented as mean ± SEM. First, experimental data were checked for normality to estimate their appropriateness for parametric statistical tests. Data were assessed by one-way ANOVA or Kruskal–Wallis test. When the F-ratio was significant, the between-group differences were assessed by Tukey’s or Games–Howell post hoc tests depending on the homogeneity of the dispersions (found by using the Levene’s test). p ≤ 0.05 was considered statistically significant.

5. Conclusions

These findings highlight the potential of prophylactic L-theanine therapy administration and moderate aerobic exercise to preserve cognitive function in rats exposed to subchronic LPS-induced neuroinflammation. Both LT and moderate aerobic exercise independently mitigated LPS-associated cognitive impairments, supporting their roles as non-pharmacological interventions. Notably, their effects were associated with partially distinct molecular profiles. L-theanine may serve as a practical intervention in situations where regular physical activity is not feasible, and it confers systemic benefits that extend beyond cognitive outcomes.

Therefore, LT and exercise should not be regarded as interchangeable but rather as complementary strategies for maintaining cognitive performance under inflammatory conditions. Further research is needed to elucidate their respective contributions and potential interactions.