Clec7a Drives Microglial Activation–Mediated Myelin Degradation in Sepsis-Associated Encephalopathy

Abstract

Sepsis-associated encephalopathy (SAE) is a frequent and devastating central nervous system complication of sepsis and portends a poor prognosis. Accumulating data implicate C-type lectin domain family 7 member A (Clec7a) in neuroinflammation and cognitive impairment, however, its specific role in SAE-linked cognitive impairment remains to be defined. Adult male mice received a single intraperitoneal injection of lipopolysaccharide (LPS, 5 mg·kg⁻¹ i.p.) to induce SAE. Four hours later, animals were treated with the Clec7a antagonist laminarin (250 mg·kg⁻¹ i.p.) once daily for three days. Cognitive performance was assessed on days 6–9 post-LPS using the open field, novel object recognition, and Y-maze tests. Hippocampal Clec7a expression was evaluated by Western blotting; microglial number and morphology, myelin integrity, and neuronal activity were assessed by immunofluorescence, Whole-Cell Recordings and transmission electron microscopy. SAE was characterized by marked upregulation of Clec7a and robust microglial activation in the CA1 region of the hippocampus. The LPS challenge increased Clec7a and CD68 expression, triggered excessive microglial phagocytosis of myelin, provoked axonal degeneration and nodal elongation, and reduced neuronal activity - as electrophysiologically demonstrated by hyperpolarized resting membrane potential(RMP), a trend toward an elevated rheobase current, and a decrease in the number of action potentials - along with decreased c-Fos⁺ cell numbers, ultimately leading to cognitive impairment. Laminarin treatment substantially attenuated microglial activation, preserved myelin architecture, restored neuronal activity (evidenced by normalized electrophysiological parameters), increased c-Fos⁺ cell numbers, and rescued cognitive impairment. Clec7a drives microglial activation, leading to aberrant myelin phagocytosis, neuronal hypoactivity, and cognitive impairment in SAE. Pharmacological inhibition of Clec7a reverses these alterations, highlighting Clec7a as a potential therapeutic target for SAE.

Introduction

Sepsis-associated encephalopathy (SAE) manifests as diffuse cerebral dysregulation secondary to systemic infection, characterized by neurological dysfunction in the absence of clinically evident central nervous system (CNS) infection [1]. Clinically, patients exhibit a spectrum of neurological manifestations, including altered consciousness, disorientation, cognitive deficits, seizures, and even coma [2]. In the intensive care setting, the incidence of SAE approaches 70%, and its presence markedly increases sepsis-related mortality [3, 4]. Long-term outcomes are equally concerning: roughly one-third of SAE survivors still demonstrate measurable cognitive impairment one year after hospital discharge [5]. These deficits substantially diminish quality of life and impose a heavy socioeconomic burden. Current mechanistic studies of SAE focus on central neuroinflammation, oxidative stress, circuit dysfunction, and neurotransmitter imbalance [6–9]; yet the precise pathogenesis remains elusive, and translational progress toward effective interventions has been slow. Identifying disease-specific therapeutic targets for SAE-related cognitive decline is therefore an urgent priority.

C-type lectin domain family 7 member A (Clec7a), a C-type lectin domain-containing pattern recognition receptor, exhibits predominant expression on myeloid lineage cells [10]. In the periphery, Clec7a signals through an immunoreceptor tyrosine-based activation motif–like domain to recruit spleen tyrosine kinase (SYK), triggering downstream cascades that promote proinflammatory cytokines and reactive oxygen species and bolster host defense [11–13]. Within the CNS, Clec7a is highly enriched in microglia, where it plays a pivotal role in neuroinflammation [14]. Excessive Clec7a activation amplifies the microglial inflammatory response and exacerbates neuronal injury via SYK-dependent pathways [15]. Clec7a also mediates microglial recognition and phagocytosis of β-amyloid, thereby accelerating the pathogenesis of Alzheimer’s disease (AD) [16], and interacts with synaptic MD2 to drive aberrant synapse elimination and behavioral deficits after stroke [17]. Whether Clec7a participates in the pathophysiology of SAE, however, is unknown.

As specialized innate immune sentinels within the CNS, microglia critically regulate neural homeostasis maintenance. Under physiological conditions, they clear cellular debris, secrete neurotrophic factors, and refine developing synapses—actions crucial for circuit stability and cognition. Pathological activation, in contrast, can lead to inappropriate synapse or even neuron loss with ensuing cognitive decline [18–21]. In AD, for example, hyperactivated microglia engulf β-amyloid and synaptic elements, driving inflammation and cognitive impairment [22]. Myelin is another potential target: in experimental cerebral ischemia, activated striatal microglia phagocytose and degrade myelin sheaths, provoking neuroinflammation and both cognitive and motor deficits [23]. Myelin, a lipid-rich multilamellar membrane wrapping axons, is indispensable for rapid impulse propagation [24]. Myelin plasticity—manifested as changes in sheath number, thickness, or nodal length—fine-tunes conduction velocity, network synchrony, and ultimately neuronal excitability, thereby contributing to cognitive processing [25, 26]. Conversely, myelin loss aggravates AD pathology: demyelinated axons preferentially accumulate β-amyloid, fueling neuroinflammation and cognitive decline [27].

Here, we propose that LPS upregulates Clec7a expression in microglia within the hippocampal CA1 region of mice, thereby inducing microglial activation and subsequent phagocytosis-mediated myelin damage, ultimately contributing to cognitive impairment.

Materials and Methods

Animals

Male C57BL/6 mice (age: 8 weeks; weight: 18–24 g) were sourced from the Comparative Medicine Center of Jinling Hospital, Nanjing University School of Medicine (Nanjing, China). Animals were maintained under standard conditions with unrestricted food and water access, followed by a 7-day acclimation period before experiments.

Induction of the SAE Model

The SAE model was induced in mice by a single intraperitoneal injection of lipopolysaccharide (LPS; 0.4 mg·mL⁻¹ in sterile saline) at 5 mg·kg⁻¹ body weight, following established protocols [8]. Control counterparts were administered equivalent volumes of sterile saline.

Experimental Design and Drug Administration

A total of 120 mice were randomly divided into four main groups (n = 30 per group): (1) NS + PBS, (2) NS + LAM, (3) LPS + PBS, (4) LPS + LAM. At 4 h post-LPS or saline injection, mice received i.p. injections of the Clec7a inhibitor Laminarin (LAM; 250 mg/kg, Merck, USA) or phosphate-buffered saline (PBS), administered once daily for 3 consecutive days [14].

Open Field Test (OFT)

To measure locomotor activity, the total distance traveled by each mouse during the OFT was recorded. Anxiety-like behavior was assessed based on both the duration spent and the distance moved within the central zone. The testing arena was a white, open box (50 × 50 × 50 cm) partitioned into 16 identical squares, with the innermost four squares designated as the center. Mice were individually introduced to the arena’s center and allowed 5 min of unrestricted exploration. A video tracking system (SuperMaze, Shanghai Xinruan Information Technology Co., Ltd., China) automatically quantified the total distance traveled, central zone duration, and central zone distance. Between each test session, the arena was thoroughly cleaned with 75% ethanol to remove residual odors.

Novel Object Recognition (NOR)

The open-field arena served as the testing apparatus for NOR. After 10 min of habituation to the empty chamber, mice were returned to their home cages. Twenty-four hours later, two identical objects were introduced for a 10-minute familiarization trial. Following an additional 24 h interval, subjects were re-exposed to the arena for 10 min, wherein one familiar object had been replaced by a novel counterpart. Exploration behavior (sniffing or nose-touching within 2 cm of objects) was recorded automatically via the SuperMaze tracking platform. Discrimination indices (DI) were derived from the formula: (Novel object exploration time/Total exploration time for both objects) × 100% [28]. Inter-trial sanitization employed 75% ethanol for both arena surfaces and objects.

Y-Maze

Three uniform opaque arms comprised the Y-maze apparatus. Each mouse was released centrally for an 8-minute exploration session. Consecutive entries into three separate arms defined spontaneous alternation, with its percentage computed as: [Alternations/(Total arm entries − 2)] × 100% [29]. Ethanol (75%) sanitization occurred between trials.

Western Blotting

Bilateral hippocampal tissues were dissected immediately post-behavior and homogenized in RIPA lysis buffer supplemented with protease inhibitors. Protein concentrations were quantified using a BCA assay. Equal protein aliquots (20 µg) underwent electrophoretic separation on 12.5% SDS-polyacrylamide gels followed by transfer to PVDF membranes. After blocking with 5% non-fat milk in TBST for 1 h at room temperature, membranes were incubated overnight at 4 °C with primary antibodies against Clec7a (Abcam, ab140039,1:1000), SYK(Cell Signaling Technology,13198T,1:1000), MBP(Abclonal, A1664,1:1000), β-tubulin(Abclonal, AC008,1:1000), and β-actin (Abclonal, AC026,1:1000). Following three washes with TBST, membranes were probed with HRP-conjugated goat anti-rabbit IgG for 2 h at room temperature. Protein bands were visualized using enhanced chemiluminescence (ECL) and quantified densitometrically with ImageJ software.

Immunofluorescence

Under sevoflurane anesthesia, mice were transcardially perfused with PBS and then 4% paraformaldehyde (PFA). Brains were post-fixed in 4% PFA for 6–8 h and cryoprotected in 30% sucrose. Coronal Sect. (30 μm thickness) were prepared using a cryostat. After three washes in PBS, sections were blocked for 1 h at room temperature with 10% goat serum and 0.3% Triton X-100 in TBST, then incubated overnight at 4 °C with primary antibodies against the following targets: Iba1 (SYSY,234308,1:500), GFAP (Cell Signaling Technology,3670T,1:200), NeuN (Cell Signaling Technology,20307T,1:200), Clec7a (Abcam, ab140039,1:200), MBP (Abcam, ab7349,1:200), PLP (Abcam, ab9311,1:500); CD68 (BioLegend,137001,1:500); dMBP (Merck, AB5864,1:200); NF-H(Cell Signaling Technology, 30564 T,1:200), Caspr(Cell Signaling Technology, 97736 T,1:200), c-Fos (Cell Signaling Technology,2250T,1:200). Sections were subsequently incubated with fluorescent secondary antibodies (room temperature, 1 h) and counterstained with DAPI (Beyotime). Hippocampal CA1 regions were imaged by using an Olympus FV3000 confocal microscope. For c-Fos immunofluorescence staining, brain tissue was collected 1–1.5 h post-NOR test, following which all processing steps were performed according to the standard immunofluorescence protocol detailed previously [30]. Internodal length was defined and measured as the calibrated distance between adjacent Caspr clusters, using the line tool in ImageJ. For three-dimensional analysis of microglial morphology and Iba1 colocalization with CD68, MBP, or dMBP, image z-stacks were first acquired under identical conditions using a confocal microscope with a z-step size of 0.75 μm (typically encompassing ~ 12 optical sections per stack). The resulting stacks were then processed using ImageJ to reconstruct 3D surface models, enabling quantitative analysis via the Manders’ co-localization coefficient. Additionally, Microglia solidity was defined as the ratio of the microglia’s actual area to the area of its convex hull (Solidity = Area/Convex Area).

Transmission Electron Microscopy

Following isoflurane anesthesia and decapitation, whole brains were excised, rinsed in deionized water, and blotted dry. Hippocampal tissue was dissected into ~ 1 mm³ blocks. Samples underwent overnight fixation (4 °C) in 2.5% glutaraldehyde, followed by three 15-minute PBS rinses. Post-fixation used 1% osmium tetroxide (2 h) with subsequent washes. Dehydration progressed through graded ethanol to 100% acetone. Infiltration occurred in acetone:812 resin (1:1, 3 h at 37 °C), then pure resin (overnight + 6 h). Embedding and polymerization (60 °C, 48 h) preceded ultrathin sectioning (80 nm). Sections were stained with uranyl acetate/lead citrate, air-dried, and imaged by transmission electron microscopy. In transmission electron micrographs, a myelinated axon is defined as a nerve fiber structure consisting of a central axon surrounded by concentrically arranged myelin lamellae. Myelin sheath dimensions from TEM images were quantified in ImageJ after scale calibration. Using the ellipse tool, the inner (corresponding to axon caliber) and outer diameters (total fiber diameter) were derived from perimeters/areas measured by circle-fitting to the myelin boundaries. The g-ratio (Inner/Outer Diameter) was calculated to evaluate myelination [31].

Brain Slice Preparation and Whole-Cell Recordings

Following deep anesthesia with isoflurane, mice were rapidly decapitated, and the brain was dissected and placed in ice-cold slicing solution continuously bubbled with carbogen (95% O₂, 5% CO₂). Coronal brain slices (350 μm thick) were prepared using a vibrating microtome. Slices were subsequently transferred to a recovery chamber containing standard artificial cerebrospinal fluid (ACSF), maintained at 32 °C and continuously oxygenated with carbogen, for at least 45 min before recording. Whole-cell patch-clamp recordings were performed using a MultiClamp 700B amplifier. Recording pipettes, fabricated from borosilicate glass, had a resistance of 2–3 MΩ when filled with the internal solution. All recordings were conducted in current-clamp mode, and the following specific protocols were employed to assess electrophysiological properties: (1) The resting membrane potential (RMP) was measured directly under zero-current clamp (0 pA) conditions after the membrane potential had stabilized. (2) Subsequently, a series of depolarizing current injections (from 0 pA to 200 pA in 25 pA increments, 400 ms pulse duration, 5 s inter-stimulus interval) were applied to the neuron. The action potential (AP) rheobase current was defined as the minimal injected current amplitude required to elicit the first AP. (3) Within the same current injection protocol, the number of action potentials fired at each stimulation intensity was precisely counted to construct the neuronal input-output relationship curve. All data were analyzed offline using Clampfit software.

Statistical Analysis

Statistical analyses were conducted using GraphPad Prism 9.0 (GraphPad Software). Results are expressed as mean ± SEM. Normality testing employed the Shapiro-Wilk test. Unpaired Student’s t tests compared two groups. Multiple group comparisons utilized one-way or two-way ANOVA with Tukey’s post hoc testing. Body weight longitudinal data were evaluated by repeated measures ANOVA. Kaplan-Meier survival curves underwent log-rank testing. Statistical significance threshold was set at p < 0.05.

Results

Clec7a Signaling was Predominantly Activated in Microglia in SAE Mice

We performed immunofluorescence co-staining, which confirmed predominant Clec7a localization to microglia (Iba-1⁺ cells) in the hippocampal CA1 region and minimal expression in astrocytes (GFAP⁺) or neurons (NeuN⁺) (Fig. 1A-D; Iba-1 vs. NeuN: P < 0.0001; Iba-1 vs. GFAP: P < 0.0001). Following LPS challenge, we observed a specific upregulation of Clec7a in microglia versus NS controls, with no alteration in other cell types (Fig. 1A-C, E; LPS vs. NS: Iba-1, t = 16.27, P < 0.0001; GFAP, t = 1.340, P = 0.4635; NeuN, t = 0.1914, P = 0.9966).

Fig. 1.

Expression of Clec7a in microglia. (A) Representative immunofluorescence images showing co-localization of Iba-1 (red), Clec7a (green), and DAPI (blue) in the hippocampal CA1 region of mice across groups. Scale bar = 20 μm. (B) Representative immunofluorescence images showing co-localization of GFAP (red), Clec7a (green), and DAPI (blue) in the hippocampal CA1 region. Scale bar = 20 μm. (C) Representative immunofluorescence images showing co-localization of NeuN (red), Clec7a (green), and DAPI (blue) in the hippocampal CA1 region. Scale bar = 20 μm. (D) Proportion of Clec7a-positive cells after intraperitoneal LPS injection. (E) Quantification of cells co-expressing Clec7a with Iba-1, GFAP, or NeuN in the hippocampal CA1 region. Data are presented as mean ± SEM (n = 4–5 mice/group). **** P < 0.0001

Inhibition of Clec7a Ameliorates Sepsis-Induced Cognitive Impairment

Consistent with our experimental timeline (Fig. 2A), no mortality occurred in the NS + PBS or NS + LAM groups during the 6 days following LPS/saline injection. In contrast, survival in the LPS + PBS group was significantly reduced (81.33%); Treatment with the Clec7a inhibitor Laminarin improved survival in SAE mice, with the LPS + LAM group reaching 91.67% survival (Fig. 2B). Furthermore, LPS + LAM mice exhibited attenuated body weight loss compared to the LPS + PBS group (Fig. 2C). Behavioral assessments on days 6–9 post-LPS revealed no significant differences among groups in OFT parameters: total distance traveled (Fig. 2D; interaction: F(1,41) = 0.2041, P = 0.6538; LPS: F(1,41) = 1.301, P = 0.2606; LAM: F(1,41) = 1.182, P = 0.2833), time spend in the center (Fig. 2E; interaction: F(1,41) = 1.936, P = 0.1716; LPS: F(1,41) = 0.00187, P = 0.9657; LAM: F(1,41) = 0.3539, P = 0.5552), or distance traveled in the center (Fig. 2F; interaction: F(1,41) = 2.515, P = 0.1205; LPS: F(1,41) = 0.3569, P = 0.5535; LAM: F(1,41) = 0.8572, P = 0.3600). In the Y-maze test, LPS + PBS mice showed unchanged total arm entries versus NS + PBS controls (Fig. 2G; interaction: F(1,41) = 0.0680, P = 0.7956; LPS: F(1,41) = 1.742, P = 0.1942; LAM: F(1,41) = 1.069, P = 0.3073), but exhibited significantly impaired spontaneous alternation (Fig. 2H; interaction: F(1,41) = 10.87, P = 0.0020; LPS: F(1,41) = 7.650, P = 0.0085; LAM: F(1,41) = 6.016, P = 0.0185); LAM treatment restored alternation performance without affecting locomotion. Critically, NOR testing demonstrated that LPS significantly reduced the discrimination index (Fig. 2I; interaction: F(1,41) = 8.131, P = 0.0068; LPS: F(1,41) = 15.72, P = 0.0003; LAM: F(1,41) = 5.683, P = 0.0218), an effect fully rescued by Laminarin. These findings demonstrate that Clec7a inhibition ameliorates LPS-induced cognitive impairment.

Fig. 2.

Inhibition of Clec7a ameliorates sepsis-induced cognitive impairment. (A) Experimental flowchart. (B) Survival analysis of mice in the four groups. (C) Body weight comparisons among the four groups. (D) Total distance traveled in the OFT for the four groups. (E) Time spent in the center during OFT for the four groups. (F) Distance traveled in the center during OFT for the four groups. (G) Total arm entries in the Y maze test for the four groups. (H) Spontaneous alternation rate in the Y maze test for the four groups. (I) Discrimination index in the NOR test for the four groups. Data are presented as mean ± SEM (n = 10–12 mice/group). ** P < 0.01, *** P < 0.001, **** P < 0.0001

Clec7a Inhibitor Laminarin Attenuates Aberrant Clec7a Upregulation in Hippocampal Microglia of SAE Mice

Western blot revealed elevated whole-hippocampus Clec7a protein in LPS + PBS mice (Fig. 3A, B; interaction: F(1,20) = 11.78, P = 0.0026; LPS: F(1,20) = 16.67, P = 0.0006; LAM: F(1,20) = 12.12, P = 0.0024), which was decreased by LAM. CA1-focused immunofluorescence further demonstrated LPS-induced increases in Clec7a⁺/Iba-1⁺ microglial density (Fig. 3E, F; interaction: F(1,28) = 28.81, P < 0.0001; LPS: F(1,28) = 24.84, P < 0.0001; LAM: F(1,28) = 26.79, P < 0.0001) and Clec7a mean fluorescence intensity (Fig. 3E, G; interaction: F(1,60) = 4.765, P = 0.0330; LPS: F(1,60) = 9.411, P = 0.0032; LAM: F(1,60) = 7.294, P = 0.0090). Given that SYK is a critical downstream signaling molecule mediating Clec7a-dependent microglial functions as reported in previous studies, we further assessed SYK expression and found a significant upregulation in the LPS group by Western blot analysis (Fig. 3C, D; interaction: F (1, 12) = 8.492, P = 0.0130; LPS: F(1,12) = 32.19, P = 0.0001; LAM: F(1,12) = 16.76, P = 0.0015). Laminarin treatment reversed these changes, indicating microglial Clec7a signaling overexpression contributes to SAE pathology.

Fig. 3.

Clec7a inhibitor Laminarin attenuates aberrant Clec7a upregulation in hippocampal microglia of SAE mice. (A) Representative Western blot bands of Clec7a in hippocampal tissues across groups. (B) Quantitative analysis of Clec7a protein expression via Western blot. (C) Representative Western blot bands of SYK in hippocampal tissues across groups. (D) Quantitative analysis of SYK protein expression via Western blot. (E) Immunofluorescence images showing co-localization of Iba-1 (red), Clec7a (green), and DAPI (blue) in the hippocampal CA1 region. Scale bar = 30 μm. (F) Quantification of cells co-expressing Clec7a and Iba-1 in the CA1 region. (G) Analysis of mean fluorescence intensity of Clec7a in microglia. Data are presented as mean ± SEM (n = 3–5 mice/group). ** P < 0.01, *** P < 0.001, **** P < 0.0001

Inhibition of Clec7a Reduces Microglial Numbers and Suppresses Their Abnormal Activation in the Hippocampal CA1 Region of SAE Mice

LPS challenge significantly increased microglial density in the hippocampal CA1 region (Fig. 4A, B; interaction: F(1,20) = 8.000, P = 0.0104; LPS: F(1,20) = 5.556, P = 0.0287; LAM: F(1,20) = 10.89, P = 0.0036) and induced morphological activation: shortened branch length (Fig. 4A, C; interaction: F(1,60) = 8.046, P = 0.0062; LPS: F(1,60) = 7.529, P = 0.0080; LAM: F(1,60) = 3.124, P = 0.0822) and increased solidity (Fig. 4A, D; interaction: F(1,60) = 9.998, P = 0.0025; LPS: F(1,60) = 9.179, P = 0.0036; LAM: F(1,60) = 0.8531, P = 0.3594). Phagocytic activity, measured by CD68⁺ volume within microglia, was also markedly elevated (Fig. 4E, F; interaction: F(1,28) = 5.044, P = 0.0328; LPS: F(1,28) = 14.07, P = 0.0008; LAM: F(1,28) = 22.52, P < 0.0001). Laminarin treatment normalized all parameters, demonstrating its efficacy in reversing aberrant microglial activation and phagocytosis.

Fig. 4.

Inhibition of Clec7a reduces microglial numbers and suppresses their abnormal activation in the hippocampal CA1 region of SAE mice. (A) Representative immunofluorescence images of Iba-1⁺ microglia (red) with morphological reconstructions in the CA1 region. Scale bar = 30 μm. (B) Quantitative analysis of microglial density in the CA1 region. (C) Analysis of mean branch length of microglia. (D) Measurement of microglial solidity. (E) Representative immunofluorescence images showing co-localization of Iba-1 (green), CD68 (red), and DAPI (blue) with 3D reconstructions of phagocytic lysosomes. Scale bar = 10 μm. (F) Quantification of CD68⁺ volume fraction within Iba-1⁺ microglia. Data are presented as mean ± SEM (n = 4–5 mice/group). * P < 0.05, ** P < 0.01, *** P < 0.001

Inhibition of Clec7a Reduces Phagocytosis of Myelin By Microglia in the Hippocampal CA1 Region of SAE Mice

Volumetric analysis confirmed significantly increased microglial engulfment of both intact myelin basic protein (MBP; Fig. 5A, B; interaction: F(1,28) = 7.241, P = 0.0119; LPS: F(1,28) = 9.251, P = 0.0051; LAM: F(1,28) = 14.64, P = 0.0007) and degraded MBP (dMBP; Fig. 5C, D; interaction: F(1,28) = 7.362, P = 0.0113; LPS: F(1,28) = 9.609, P = 0.0044; LAM: F(1,28) = 16.00, P = 0.0004) within the hippocampal CA1 region of LPS + PBS mice. LAM treatment effectively suppressed this pathological phagocytosis.

Fig. 5.

Inhibition of Clec7a reduces phagocytosis of myelin by microglia in the hippocampal CA1 region of SAE mice. (A) Representative immunofluorescence images with 3D reconstructions showing co-localization of MBP (green) and Iba-1⁺ microglia (red) in the CA1 region. Scale bar = 10 μm. (B) Quantification of MBP⁺ volume fraction within microglia. (C) Representative immunofluorescence images with 3D reconstructions showing co-localization of degenerated MBP (dMBP, green) and Iba-1⁺ microglia (red). Scale bar = 10 μm. (D) Quantification of dMBP⁺ volume fraction within microglia. Data are presented as mean ± SEM (n = 4–5 mice/group). ** P < 0.01, *** P < 0.001

Inhibition of Clec7a Reverses Myelin Loss in the Hippocampal CA1 Region of SAE Mice

Consistent with excessive phagocytosis, LPS + PBS mice exhibited profound myelin loss in CA1: reduced MBP protein levels (Fig. 6A, B; interaction: F(1,20) = 23.23, P = 0.0001; LPS: F(1,20) = 11.00, P = 0.0034; LAM: F(1,20) = 2.743, P = 0.1133), decreased MBP (Fig. 6C, D; interaction: F(1,28) = 7.756, P = 0.0095; LPS: F(1,28) = 8.150, P = 0.0080; LAM: F(1,28) = 6.116, P = 0.0197) and PLP immunoreactive area fractions (Fig. 6C, E; interaction: F(1,28) = 13.48, P = 0.0010; LPS: F(1,28) = 6.020, P = 0.0206; LAM: F(1,28) = 6.669, P = 0.0153). TEM further revealed ultrastructural damage: thinner myelin sheaths (Fig. 6F, G; interaction: F(1,128) = 5.865, P = 0.0168; LPS: F(1,128) = 5.848, P = 0.0170; LAM: F(1,128) = 10.81, P = 0.0013), fewer myelinated axons (Fig. 6F, H; interaction: F(1,20) = 12.55, P = 0.0020; LPS: F(1,20) = 14.74, P = 0.0010; LAM: F(1,20) = 5.087, P = 0.0355), and increased g-ratios (Fig. 6F, I). Laminarin treatment consistently rescued these deficits.

Fig. 6.

Inhibition of Clec7a reverses myelin loss in the hippocampal CA1 region of SAE mice. (A) Representative Western blot bands of MBP in hippocampal tissues. (B) Quantitative analysis of MBP protein expression. (C) Immunofluorescence images showing MBP⁺ (red) and PLP⁺ (red) signals with DAPI (blue) in the CA1 region. Scale bar = 30 μm. (D) Quantification of MBP⁺ area fraction in the CA1 region. (E) Quantification of PLP⁺ area fraction in the CA1 region. (F) Representative transmission electron microscopy (TEM) images of myelinated axons in the CA1 region. Scale bar = 2 μm.(G) Statistical analysis of myelin sheath thickness. (H) Percentage of myelinated axons. (I) Scatter plots with linear fit showing g-ratio values: NS + PBS (35 axons), LPS + PBS (35 axons), NS + LAM (33 axons), LPS + LAM (32 axons). Data are presented as mean ± SEM (n = 3–5 mice/group). ** P < 0.01, *** P < 0.001, **** P < 0.0001

Inhibition of Clec7a Reverses Axonal Degeneration and Nodal Lengthening in the Hippocampal CA1 Region of SAE Mice

As myelin provides critical axonal support [32], we assessed axonal integrity. LPS challenge reduced the NF-H⁺ area fraction in CA1, indicating axonal degeneration (Fig. 7A, B; interaction: F(1,28) = 13.58, P = 0.0010; LPS: F(1,28) = 8.430, P = 0.0071; LAM: F(1,28) = 4.810, P = 0.0368). In addition, Caspr, a well-established marker of the paranodal region, was labeled by immunofluorescence, and the nodal length was indirectly determined by measuring the distance between the inner boundaries of the bilateral Caspr-positive domains [33]. Our results showed that the nodal length in the hippocampal CA1 region was significantly increased in SAE mice (Fig. 7C, D; interaction: F(1,156) = 5.846, P = 0.0168; LPS: F(1,156) = 5.706, P = 0.0181; LAM: F(1,156) = 12.42, P = 0.0006). Moreover, the distribution of nodal lengths was markedly broader in SAE mice (Fig. 7E–H), indicating greater variability, and this abnormal elongation and dispersion were effectively reversed by laminarin treatment.

Fig. 7.

Inhibition of Clec7a reverses axonal degeneration and nodal lengthening in the hippocampal CA1 region of SAE mice. (A) Immunofluorescence images of NF-H⁺ axons (green) with DAPI (blue) in the CA1 region. Scale bar = 30 μm. (B) Quantification of NF-H⁺ area fraction. (C) Immunofluorescence images of Caspr⁺ paranodes (red) with DAPI (blue). Scale bar = 30 μm. (D) Statistical analysis of nodal length. (E-H) Frequency distribution histograms of nodal lengths (40 nodes/mouse from 3 mice/group). Data are presented as mean ± SEM (n = 3–5 mice/group). ** P < 0.01, *** P < 0.001

Inhibition of Clec7a Restores Neuronal Activity in the Hippocampal CA1 Region of SAE Mice

To systematically evaluate the effects of LPS on neuronal activity in the hippocampal CA1 region of mice, we further employed ex vivo electrophysiological recordings and immunofluorescence staining targeting this specific area. Electrophysiological results revealed that compared to the NS + PBS group, the LPS + PBS group exhibited a significant increase in the absolute value of the RMP, indicating a hyperpolarized state (Fig. 8A, B; interaction: F (1, 32) = 18.75, P = 0.0001; LPS: F (1, 32) = 2.139, P = 0.1534; LAM: F (1, 32) = 1.561, P = 0.2206). Furthermore, with increasing current stimulation intensities, the number of action potentials elicited in the NS + PBS group was significantly higher than that in the LPS + PBS group (Fig. 8D). Although non-significant, the LPS + PBS group showed a trend toward an elevated rheobase (Fig. 8A, C; interaction: F (1, 32) = 0.3062, P = 0.5839; LPS: F (1, 32) = 4.444, P = 0.429; LAM: F (1, 32) = 1.790, P = 0.1904). Immunofluorescence results further demonstrated a significant reduction in the number of c-Fos-positive neurons in the hippocampal CA1 region of LPS-treated mice (Fig. 8E, F; interaction: F(1,28) = 5.413, P = 0.0274; LPS: F(1,28) = 6.467, P = 0.0168; LAM: F(1,28) = 7.615, P = 0.0101). Importantly, these aberrant changes in both electrophysiological and immunofluorescence indicators were reversed by LAM treatment.

Fig. 8.

Inhibition of Clec7a restores neuronal activity in the hippocampal CA1 region of SAE mice. (A)Representative APs of pyramidal neurons in the CA1 region. Horizontal calibration bars, 200 ms; vertical bars, 20 mV. (B)The RMP of pyramidal neurons in the CA1 region. (C) The Rheobase current of pyramidal neurons in the CA1 region. (D) The APs numbers of pyramidal neurons in the CA1 region; NS + PBS vs. NS + LAM, ^&P<0.05; NS + PBS vs. LPS + PBS, **P<0.01, ****P<0.0001; LPS + PBS vs. LPS + LAM, ^##P<0.01, ^###P<0.01, ^####P<0.01. (E) Representative immunofluorescence images of c-Fos⁺ cells (green) with DAPI (blue) in the CA1 region. Scale bar = 30 μm. (F) Quantification of c-Fos⁺ cells. Data are presented as mean ± SEM (n = 3–4 mice/group). ** P < 0.01

Discussion

In this study, we established a murine model of SAE by administering a single intraperitoneal LPS injection. Our findings demonstrate that LPS challenge significantly upregulated Clec7a expression in hippocampal microglia, triggering aberrant microglial activation and excessive myelin phagocytosis. This cascade subsequently induced axonal degeneration and abnormal nodal lengthening, ultimately leading to suppressed neuronal activity and SAE manifestations. Critically, administration of laminarin, a specific Clec7a inhibitor, effectively reversed these pathological alterations and improved cognitive function in SAE mice (Fig. 9).

Fig. 9.

The schematic diagram illustrates that LPS induces Clec7a overexpression in microglia, promoting myelin phagocytosis, ultimately reducing neuronal activity (c-Fos⁺ cells) and cognitive function

SAE constitutes a life-threatening neurological complication of systemic infection, manifesting clinically as diffuse neurophysiological derangement with heterogeneous neurological manifestations. Notably, over 80% of affected patients exhibit abnormal electroencephalogram findings [34–36]. SAE is strongly associated with persistent cognitive impairment, irreversible organ damage, and increased mortality [37]. LPS, a gram-negative bacterial membrane endotoxin, functions as a potent pathogen-associated molecular pattern that triggers systemic inflammation, recapitulating hallmark pathophysiological manifestations of clinical sepsis [38]. Consistent with studies reporting LPS-induced cognitive impairment in mice [39, 40], our behavioral assessments confirmed significant cognitive impairment in LPS-challenged animals.

While mechanistic insights into sepsis-mediated cognitive impairment are still evolving, the myeloid-restricted C-type lectin receptor Clec7a emerges as a critical modulator of pathogen sensing, immunoregulatory cascades, and inflammatory signaling pathways [41]. Evidence suggests it contributes to the pathogenesis of Alzheimer’s disease (AD), neuropathic pain, Parkinson’s disease, and post-stroke cognitive impairment, primarily through microglial activation [42–45]. Paradoxically, some studies suggest a neuroprotective anti-inflammatory function; for instance, astrocyte-derived galectin-9 signaling via Clec7a on neutrophils can limit autoimmune neuroinflammation [46]. This context-dependent duality may be attributed to the specific inflammatory milieu; in the hyperinflammatory environment of SAE, sustained Clec7a signaling on microglia appears to be co-opted to drive a maladaptive response, shifting its role from homeostasis maintenance to pathological amplification. Here, we report for the first time that Clec7a is significantly upregulated specifically in hippocampal CA1 microglia in SAE. This upregulation correlates with microglial proliferation, morphological activation, and enhanced phagocytosis, suggesting that the inflammatory environment promotes a pathogenic shift in microglial function. Interestingly, laminarin treatment in healthy mice suppressed neuronal activity (reduced action potential firing) without causing behavioral deficits, implying that basal Clec7a signaling may help maintain microglial homeostasis and neuronal circuit function - a finding that warrants further investigation into its physiological roles.

Myelin sheaths dramatically increase action potential conduction velocity via saltatory conduction, enabling rapid signal transmission within neural networks [47]. Demyelination or myelin damage is a hallmark pathological feature in numerous CNS disorders [48, 49], leading to delayed signal conduction, impaired network connectivity, and consequent cognitive, motor, and sensory impairments [31]. Under physiological conditions, microglia support myelin homeostasis by clearing degraded myelin debris, facilitating remyelination. However, pathologically activated microglia may aberrantly phagocytose intact myelin structures, causing persistent damage and impaired regeneration. We demonstrate for the first time in a SAE model that microglia in the CA1 region drive subsequent pathological progression by phagocytosing intact myelin.

The node of Ranvier, the unmyelinated axonal segment between two myelin sheaths, is densely packed with voltage-gated sodium channels essential for action potential regeneration during saltatory conduction [50]. Abnormal nodal lengthening, commonly observed in demyelinating diseases like multiple sclerosis, can cause conduction block and clinical manifestations such as motor and sensory deficits [51]. Furthermore, hippocampal myelin loss and nodal pathology are implicated as key mechanisms underlying depression [33]. Our findings corroborate this association in SAE, demonstrating that myelin loss precedes axonal degeneration and abnormal nodal lengthening in the hippocampal CA1 region – pathologies effectively mitigated by laminarin treatment.

The immediate-early gene product c-Fos serves as a reliable marker of neuronal activation. Its expression is rapidly and transiently induced by diverse stimuli (e.g., synaptic activity, growth factors, stress) [52], significantly upregulated in hippocampal neurons following novel learning experiences [53], and widely used as an indicator of neuronal activity due to its rapid induction kinetics (peaking 60–90 min post-stimulus) [54, 55]. Emerging evidence underscores a critical link between myelin integrity, neuronal activity, and cognitive function. On one hand, myelin serves as a crucial metabolic platform for axons, providing key energy substrates like lactate to maintain ionic homeostasis - myelin loss triggers axonal energy crisis, leading to impaired Na⁺/K⁺ ATPase function that directly causes resting membrane potential hyperpolarization and elevated action potential threshold, ultimately affecting memory function [56, 57]. On the other hand, conduction delays cause desynchronizes network activity, and disrupts information integration within neural circuits, impairing the activation of neuronal ensembles essential for learning, memory, and cognition [58]. Our study provides direct electrophysiological evidence supporting the role of this link in SAE. SAE suppressed hippocampal CA1 neuron activity, evidenced by hyperpolarized RMP, an increasing trend in the rheobase current, and reduced firing, alongside decreased c-Fos⁺ cells. Clec7a inhibition reversed both electrophysiological impairments and c-Fos expression, revealing a myelin-protective therapeutic avenue for SAE cognitive impairments.

Nevertheless, it is important to acknowledge certain limitations in the present study. The Clec7a antagonist employed here, Laminarin, is a natural polysaccharide compound derived from brown algae. Although widely recognized in the field as a classical pharmacological tool for inhibiting Clec7a signaling [14, 15] and providing key mechanistic insights in our experimental context, we must consider its potential for off-target effects. As a naturally occurring molecule, Laminarin may exhibit pleiotropic immunomodulatory activities independent of Clec7a blockade. In addition, while the current evidence strongly supports the microglial-specific role of Clec7a in our experimental setting, future studies employing microglial depletion via CSF1R inhibitors would provide more direct cellular validation. Similarly, although our data suggest the involvement of Syk downstream of Clec7a, further confirmation using specific Syk inhibitors would help establish a definitive causal link within this signaling axis. We commit to employing genetic knockout models as well as microglial ablation approaches in future investigations to provide more definitive validation of Clec7a’s specific role in SAE pathogenesis. Additionally, it should be noted that the evaluation of microglial inflammatory phenotype and phagocytic function in this study is primarily based on in vivo evidence. While these in vivo data provide important support for our conclusions, the observed alterations in these cellular functions still require further validation through in vitro experimental systems such as primary microglial cultures. Notwithstanding this methodological constraint, we believe our study systematically demonstrates the pivotal involvement of the Clec7a signaling pathway in the pathogenesis of SAE. The evidence presented herein offers valuable insights into the immune mechanisms underlying this condition and establishes a cornerstone for future research aimed at therapeutically targeting this pathway.

Conclusions

In summary, using the LPS-induced SAE model, this study delineates a pathogenic cascade initiated by Clec7a upregulation in hippocampal microglia. Clec7a activation drives aberrant microglial reactivity and excessive myelin phagocytosis, leading to subsequent myelin loss, axonal degeneration, abnormal nodal lengthening, and ultimately, suppressed neuronal activity and cognitive impairment. The compelling reversal of this entire pathological sequence and the amelioration of cognitive impairment by the Clec7a inhibitor laminarin highlight Clec7a as a critical mediator in SAE pathogenesis and a promising therapeutic target.

Author Contributions

JXJ and XYY designed this work. JXJ wrote the paper. JXJ and XYY performed the experiments. JXJ and XDM analyzed the data. JCS, JJY and MHJ directed the discussion. All authors read and approved the final manuscript.

Funding

This study was supported by grants from The Natural Science Foundation of Jiangsu Province (to J.C.S., NO.BK20242097).

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Consent for Publication

Not applicable.

Ethics Approval

All animal experiments and related operations were carried out in accordance with the Guideline for the Care and Use of Laboratory Animals from the Animal Center of Jinling Hospital and the National Institutes of Health.

Competing Interests

The authors declare no competing interests.

Clinical Trial Number

Not applicable.