Interleukin-6 deficiency reduces neuroinflammation by inhibiting the STAT3-cGAS-STING pathway in Alzheimer’s disease mice

Abstract

Background

The Interleukin-6 (IL-6)-signal transducer and activator of transcription 3 (STAT3) pathway, along with the cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) pathway, are critical contributors to neuroinflammation in Alzheimer’s disease (AD). Although previous research outside the context of AD has indicated that the IL-6-STAT3 pathway may regulate the cGAS-STING pathway, the exact molecular mechanisms through which IL-6-STAT3 influences cGAS-STING in AD are still not well understood.

Methods

The activation of the IL-6-STAT3 and cGAS-STING pathways in the hippocampus of 5×FAD and WT mice was analyzed using WB and qRT-PCR. To explore the effects of IL-6 deficiency, Il6^+/− mice were crossed with 5×FAD mice, and the subsequent impact on hippocampal STAT3 pathway activity, cGAS-STING pathway activation, amyloid pathology, neuroinflammation, and cognitive function was evaluated through WB, qRT-PCR, immunohistochemistry, ThS staining, ELISA, and behavioral tests. The regulatory role of STAT3 in the transcription of the Cgas and Sting genes was further validated using ChIP-seq and ChIP-qPCR on hippocampal tissue from 5×FAD and Il6^−/−: 5×FAD mice. Additionally, in the BV2 microglial cell line, the impact of STAT3 activation on the transcriptional regulation of Cgas and Sting genes, as well as the production of inflammatory mediators, was examined through WB and qRT-PCR.

Results

We observed marked activation of the IL-6-STAT3 and cGAS-STING pathways in the hippocampus of AD mice, which was attenuated in the absence of IL-6. IL-6 deficiency reduced beta-amyloid deposition and neuroinflammation in the hippocampus of AD mice, contributing to cognitive improvements. Further analysis revealed that STAT3 directly regulates the transcription of both the Cgas and Sting genes. These findings suggest a potential mechanism involving the STAT3-cGAS-STING pathway, wherein IL-6 deficiency mitigates neuroinflammation in AD mice by modulating this pathway.

Conclusion

These findings indicate that the STAT3-cGAS-STING pathway is critical in mediating neuroinflammation associated with AD and may represent a potential therapeutic target for modulating this inflammatory process in AD.

Introduction

Alzheimer’s disease (AD) is the leading cause of dementia in the elderly, characterized by two key pathological hallmarks: the formation of plaques from aggregated beta-amyloid (Aβ) and neurofibrillary tangles resulting from the accumulation of hyperphosphorylated tau protein [1]. Recent research has emphasized the critical role of neuroinflammation in the progression of AD. Both Aβ and hyperphosphorylated tau protein can activate microglia and astrocytes in the brain, triggering the release of inflammatory mediators such as interleukin-1β (IL-1β), interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-α). These inflammatory mediators not only directly damage neurons but also disrupt the blood-brain barrier, allowing additional inflammatory factors and peripheral immune cells to infiltrate the brain, exacerbating neuroinflammation and accelerating AD progression [2,3,4,5,6,7].

Multiple studies have shown that IL-6 expression levels are significantly elevated in both the brain parenchyma [8, 9] and peripheral blood of AD patients [10,11,12], as well as in the brain parenchyma of AD animal models [13,14,15]. As a pro-inflammatory cytokine, IL-6 is a key activator of the signal transducer and activator of transcription 3 (STAT3) pathway [16]. Research suggests that STAT3 plays a role in mediating Aβ-induced neuronal death [17]. Various drugs have been shown to target the STAT3 pathway in astrocytes or microglia, reducing the production of inflammatory mediators [18,19,20,21,22]. Moreover, the STAT3 pathway is crucial for the activation of astrocytes in the hippocampus of AD mice [23,24,25]. Inhibition of STAT3 phosphorylation in astrocytes [23] or conditional deletion of STAT3 in astrocytes [24] can attenuate astrocyte overactivation and significantly improve cognitive deficits in AD mice. These findings highlight the pivotal role of the IL-6-STAT3 pathway in AD-related neuroinflammation.

The cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) pathway is a vital immune response mechanism activated by the presence of cytosolic DNA, which can originate from viral and bacterial sources or aberrantly released mitochondrial DNA. Upon activation, this pathway triggers the upregulation of various inflammatory mediators, including interferons (IFNs), interferon-stimulated genes (ISGs), and cytokines [26, 27]. Studies have shown that the cGAS-STING pathway is activated in the brains of AD patients [28], AD mouse models [29, 30], and aged mice [28, 31]. In AD mouse models, deficiency in the Cgas gene or treatment with the STING inhibitor H-151 significantly reduces neuroinflammation and improves cognitive deficits [28]. Furthermore, modulating the cGAS-STING pathway has been shown to alleviate neuroinflammation in AD mice [32, 33]. These findings suggest that the cGAS-STING pathway plays a pivotal role in AD-related neuroinflammation.

Interestingly, research in tumor biology has indicated that the IL-6-STAT3 pathway can influence the cGAS-STING pathway’s activity. Specifically, activation of the IL-6-STAT3 pathway affects the cGAS-STING-mediated anti-tumor response [34,35,36]. Additionally, studies on stroke have shown that inhibiting the STAT3 pathway suppresses the expression of cGAS and STING in mouse models of the disease [37]. However, it remains unclear how the IL-6-STAT3 pathway modulates the cGAS-STING pathway in the context of AD.

In this study, we identified the activation of both the IL-6-STAT3 and cGAS-STING pathways in AD mice. IL-6 deficiency in these mice resulted in a significant reduction in STAT3 activation, suppression of cGAS-STING pathway activity, decreased Aβ deposition, alleviated neuroinflammation, and improved cognitive function. Mechanistically, we demonstrated that STAT3 binds to the promoter regions of Cgas and Sting, directly regulating their transcription. This finding suggests a novel mechanism involving the STAT3-cGAS-STING pathway, which was further validated through in vitro experiments showing activation of this pathway by both IL-6 and Aβ42. Moreover, the absence of IL-6 in AD mice led to reduced neuroinflammation via this pathway. Collectively, these results highlight the critical role of the STAT3-cGAS-STING pathway in AD pathology, providing important insights into the contribution of neuroinflammation to disease progression.

Results

In the hippocampus of 5×FAD mice, the IL-6-STAT3 pathway is activated and is associated with disease progression

To investigate the role of the IL-6-STAT3 pathway in AD progression, we employed the 5×FAD mouse model. We quantified the expression levels of IL-6, phosphorylated STAT3 (p-STAT3), and total STAT3 in the hippocampus of 5×FAD mice and their wild-type (WT) littermates at 4, 6, 9, and 12 months of age. The activation of the IL-6-STAT3 pathway was determined by calculating the p-STAT3/STAT3 ratio. Our results revealed that IL-6 expression in the hippocampus of 5×FAD mice was consistently and significantly higher than in age- and sex-matched WT mice at all examined time points (Fig. 1A, B), suggesting that IL-6 may play a role in AD pathology. Consistently, the p-STAT3/STAT3 ratio in the hippocampus of 5×FAD mice was markedly elevated compared to WT mice at 4 months (Fig. 1C, D), 6 months (Fig. 1E, F), 9 months (Fig. 1G, H), and 12 months (Fig. 1I, J). However, no significant differences were observed in total STAT3 protein levels between 5×FAD and WT mice (Fig. 1C-J). These findings suggest that while total STAT3 levels remain unchanged, the significantly increased p-STAT3/STAT3 ratio in the AD mouse model indicates IL-6-STAT3 pathway activation in AD.

To further explore the relationship between IL-6-STAT3 pathway activation and AD progression, we conducted a longitudinal analysis comparing 5×FAD and WT mice across the same age groups. The study primarily focused on changes in the p-STAT3/STAT3 ratio and STAT3 expression levels. No significant differences in the hippocampal p-STAT3/STAT3 ratio were observed between 4- and 6-month-old AD mice (Fig. 1K, L). However, a marked increase was detected at 9 months compared to 6 months, with a further elevation at 12 months (Fig. 1K, L). These results suggest a progressive increase in the p-STAT3/STAT3 ratio with advancing age in 5×FAD mice. In contrast, total STAT3 protein levels remained consistent across all age comparisons (4 to 6 months, 6 to 9 months, and 9 to 12 months) in AD mice (Fig. 1K, L). Similarly, WT mice showed no significant changes in either the p-STAT3/STAT3 ratio or STAT3 expression levels at any of the examined time points (Fig. 1M, N). Overall, these results indicate that the IL-6-STAT3 pathway is progressively activated in 5×FAD mice as the disease progresses, whereas no such activation occurs in WT mice.

Fig. 1

Activation of the IL-6-STAT3 pathway in the hippocampus of 5×FAD mice. (A) Western blot analysis of IL-6 expression in hippocampal tissue from WT and 5×FAD mice at different ages. (B) Quantification of IL-6 levels shown in panel (A). Western blot analysis of p-STAT3 and total STAT3 in hippocampal tissue from WT and 5×FAD mice at 4 months (C), 6 months (E), 9 months (G), 12 months (I) of age. (D), (F), (H), (J) Quantification of the p-STAT3/STAT3 ratio and total STAT3 levels corresponding to panels (C), (E), (G), and (I). Western blot analysis of p-STAT3 and STAT3 in hippocampal tissue from 5×FAD (K) and WT (M) mice across different ages. (L), (N) Quantification of the p-STAT3/STAT3 ratio and total STAT3 levels corresponding to panels (K) and (M). Each data point represents an individual mouse. Panels (A-J): n = 5 mice per group (mixed sexes). Panels (K-N): n = 3 mice per group (mixed sexes). Statistical significance for panels (B), (D), (F), (H), and (J) was determined using Student’s t-test or Welch’s t-test. For panels (L) and (N), statistical significance was assessed using one-way ANOVA followed by Tukey’s multiple comparisons test or the Games-Howell test. Data are presented as mean ± S.E.M. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Ns: not significant

In the hippocampus of 5×FAD mice, the cGAS-STING pathway is activated and is associated with disease progression

To investigate the role of the cGAS-STING pathway in AD progression, we analyzed the expression levels of cGAS and STING in the hippocampus of 5×FAD and WT mice at 4, 6, 9, and 12 months of age. At 4 months, no significant differences in cGAS and STING expression were observed between 5×FAD and WT mice (Fig. 2A, B). However, by 6 months (Fig. 2C, D), and continuing through 9 months (Fig. 2E, F) and 12 months (Fig. 2G, H), both cGAS and STING expression levels were substantially elevated in the hippocampus of 5×FAD mice compared to WT mice.

A longitudinal analysis of cGAS and STING expression revealed a slight, but not statistically significant, increase in 6-month-old 5×FAD mice relative to their 4-month-old counterparts (Fig. 2I, J). In contrast, a significant elevation in both cGAS and STING expression was detected in 9-month-old 5×FAD mice compared to the 6-month-old group (Fig. 2I, J). By 12 months, STING expression levels continued to rise, while cGAS levels plateaued relative to those at 9 months (Fig. 2I, J). These findings suggest that during AD progression in 5×FAD mice, cGAS expression gradually increases, reaching a peak in the mid-stage of the disease before stabilizing, while STING expression shows a continuous escalation with age. In contrast, WT mice maintained stable cGAS and STING expression levels across all examined time points (Fig. 2K, L). Overall, these results indicate that the activation of the cGAS-STING pathway intensifies progressively with disease progression in 5×FAD mice, a pattern that is not observed in WT mice.

Given that activation of the cGAS-STING pathway can induce the expression of ISGs, we examined the mRNA expression levels of several ISGs, including Isg15, Isg20, Ifit1, Ifit2 and Irf7 [38, 39], to further elucidate the activation of the cGAS-STING pathway in 5×FAD mice. Building on previous findings that cGAS and STING expression levels are substantially elevated in 6-month-old 5×FAD mice compared to WT mice, we focused on the hippocampus of 6-month-old 5×FAD and WT mice to assess ISG expression. Our results demonstrated that the mRNA levels of Isg15, Isg20, Ifit1, Ifit2 and Irf7 were considerably higher in the hippocampus of 6-month-old 5×FAD mice compared to age-matched WT controls (Fig. 2M), further confirming the activation of the cGAS-STING pathway in this AD mouse model.

Fig. 2

Activation of the cGAS-STING pathway in the hippocampus of 5×FAD mice. Western blot analysis of cGAS and STING expression in hippocampal tissue from WT and 5×FAD mice at 4 months (A), 6 months (C), 9 months (E), and 12 months (G) of age. (B), (D), (F), (H) Quantification of cGAS and STING levels shown in panels (A), (C), (E), and (G). Western blot analysis of cGAS and STING at different ages in hippocampal tissue from 5×FAD (I) and WT (K) mice. (J), (L) Quantification of cGAS and STING levels corresponding to panels (I) and (K). (M) qRT-PCR analysis of Isg15, Isg20, Ifit1, Ifit2, and Irf7 mRNA levels relative to Gapdh in 6-month-old WT and 5×FAD mice. Each dot represents an individual mouse. Panels (A-H) and (M): n = 5 mice per group (mixed sexes). Panels (I-L): n = 3 mice per group (mixed sexes). Statistical significance for panels (B), (D), (F), (H), and (M) was determined using Student’s t-test or Welch’s t-test. For panels (J) and (L), one-way ANOVA followed by Tukey’s multiple comparisons test or the Games-Howell test was employed. Data are presented as mean ± S.E.M. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Ns: not significant

IL-6 deficiency attenuates the activation of the STAT3 and cGAS-STING pathways in the hippocampus of 5×FAD mice

Based on our findings, we observed that in the hippocampus of 4-month-old 5×FAD mice, the IL-6-STAT3 pathway was already activated, while the expression levels of cGAS and STING remained largely unchanged. To explore whether and how the IL-6-STAT3 pathway influences the activation of the cGAS-STING pathway, we generated IL-6 deficient 5×FAD mice (Il6^−/−:5×FAD) by crossing heterozygous Il6 gene knockout mice (Il6^+/−) with 5×FAD mice. In the hippocampus of 6-month-old Il6^−/−:5×FAD mice, IL-6 expression was nearly undetectable compared to age- and sex-matched 5×FAD mice (Fig. 3A), confirming the successful IL-6 deficiency. Further analysis revealed a significant reduction in the p-STAT3/STAT3 ratio, along with a marked decrease in total STAT3 protein levels (Fig. 3B, C). These findings indicate that IL-6 deficiency substantially attenuates STAT3 pathway activation in the hippocampus of 5×FAD mice.

We also observed a significant reduction in the expression of cGAS and STING in the hippocampus of 6-month-old Il6^−/−:5×FAD mice (Fig. 3D, E), indicating a corresponding decrease in cGAS-STING pathway activity. To further assess the impact of IL-6 deficiency on this pathway, we compared the mRNA expression levels of several ISGs, including Isg15, Isg20, Ifit1, Ifit2 and Irf7, between 6-month-old 5×FAD mice and Il6^−/−:5×FAD mice. The results demonstrated a substantial reduction in the mRNA levels of these ISGs in the hippocampus of Il6^−/−:5×FAD mice compared to their 5×FAD counterparts (Fig. 3F). These findings collectively suggest that IL-6 deficiency significantly diminishes the activity of the cGAS-STING pathway in 5×FAD mice.

Fig. 3

IL-6 deficiency reduces activation of the STAT3 and cGAS-STING pathways in the hippocampus of 5×FAD mice. (A) Western blot analysis of IL-6 expression in hippocampal tissue from 6-month-old 5×FAD and Il6^−/−:5×FAD mice. (B) Western blot analysis of p-STAT3 and total STAT3 in the same tissue samples. (C) Quantification of the p-STAT3/STAT3 ratio and total STAT3 levels shown in panel (B). (D) Western blot analysis of cGAS and STING expression in hippocampal tissue from the same groups of mice. (E) Quantification of cGAS and STING levels corresponding to panel (D). (F) qRT-PCR analysis of Isg15, Isg20, Ifit1, Ifit2, and Irf7 mRNA levels relative to Gapdh in 5×FAD and Il6^−/−:5×FAD mice. Each dot represents an individual mouse. Panels (A-E): n = 6 mice per group (mixed sexes). Panel (F): n = 5 mice per group (mixed sexes). Statistical significance was determined using Student’s t-test or Welch’s t-test. Data are presented as mean ± S.E.M. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Ns: not significant

IL-6 deficiency improves cognitive deficits in 5×FAD mice

Building on our findings that IL-6 deficiency substantially reduces the activity of the STAT3 and cGAS-STING pathways in 5×FAD mice, we hypothesized that IL-6 deficiency might improve cognitive deficits in these mice. To test this, we divided the mice into four groups: WT, Il6^−/−, Il6^−/−:5×FAD, and 5×FAD. To eliminate potential confounding effects of motor function on cognitive assessments, we first evaluated the impact of IL-6 deficiency on motor abilities in 5×FAD mice using the open field test. The results showed no significant differences in distance traveled (Fig. 4A) or movement speed (Fig. 4B) among the groups, indicating that IL-6 deficiency does not impair motor abilities in either 5×FAD or WT mice.

Next, we investigated whether IL-6 deficiency influences emotional behavior, particularly anxiety-like behavior, in 5×FAD and WT mice. The distance traveled and time spent in the central area of the open field are commonly used indicators of anxiety-like behavior in mice [40]. The results showed no significant differences in distance traveled (Fig. 4C) or time spent in the central area (Fig. 4D) among the four groups. To further assess anxiety-like behavior, we used the elevated plus maze test, where the time spent in the open arms and the number of entries into the open arms are commonly used indicators of anxiety [41]. The results indicated no significant differences in these anxiety-related measures among the groups (Fig. 4E and F). Based on findings from both the open field and elevated plus maze tests, we concluded that IL-6 deficiency does not affect anxiety-like behavior in either WT or 5×FAD mice.

The Morris water maze (MWM) test, which includes both training and testing phases, is a widely used method for evaluating cognitive function in mice [42, 43]. During the training phase, mice learn to locate a submerged, invisible platform through repeated trials, with spatial learning and memory assessed by measuring escape latency—the time it takes to find the platform. In the testing phase, the platform is removed, and memory is evaluated by recording the number of times the mice cross the original platform location. In this study, we used the MWM to examine the effect of IL-6 deficiency on cognitive function in 5×FAD mice. The four groups of mice underwent five days of training, with daily measurements of swimming speed and escape latency. The results showed no significant differences in swimming speed among the groups during the training period (Fig. 4G), indicating that IL-6 deficiency does not affect motor abilities in either WT or 5×FAD mice. Additionally, Il6^−/− mice showed no significant differences in escape latency compared to WT mice on any training day (Fig. 4H), suggesting that IL-6 deficiency does not impair cognitive function in WT mice. However, 5×FAD mice exhibited considerably longer escape latencies on days 2, 3, 4, and 5 compared to WT mice (Fig. 4H), reflecting cognitive deficits in the 5×FAD group. Notably, on the fifth day, Il6^−/−:5×FAD mice demonstrated shorter escape latencies compared to 5×FAD mice (Fig. 4H), suggesting that IL-6 deficiency can enhance cognitive function in 5×FAD mice. Following the five-day training period, we removed the platform and assessed swimming speed and the number of platform crossings among the four groups. Consistent with the training phase results, there were no significant differences in swimming speed among the groups (Fig. 4I). Additionally, there were no significant differences in the number of platform crossings between WT and Il6^−/− mice (Fig. 4J). However, 5×FAD mice exhibited a significant reduction in platform crossings compared to WT mice (Fig. 4J), indicating impaired cognitive function. Importantly, Il6^−/−:5×FAD mice had significantly more platform crossings than 5×FAD mice (Fig. 4J), further suggesting that IL-6 deficiency can ameliorate cognitive deficits in 5×FAD mice.

Fig. 4

IL-6 deficiency mitigates cognitive impairment in 5×FAD mice. (A-B) Total distance traveled (A) and movement speed (B) in the open field test for WT mice, Il6^−/− mice, Il6^−/−:5×FAD mice, and 5×FAD mice. (C-D) Distance traveled (C) and time spent (D) in the central area during the open field test for the same groups. (E-F) Time spent in the open arms (E) and entries into the open arms (F) in the elevated plus maze test for the same groups. (G-H) Swimming speed (G) and escape latency (H) during the five-day training period in the Morris water maze for the same groups. (I-J) Swimming speed (I) and platform crossings (J) during the probe phase of the Morris water maze for the same groups. Each dot represents an individual mouse: WT and Il6^−/− mice (n = 8, mixed sexes); Il6^−/−:5×FAD and 5×FAD mice (n = 10, mixed sexes). Statistical analysis was performed using one-way ANOVA followed by Scheffe’s or Games-Howell tests. Data are presented as mean ± S.E.M. Significance levels are indicated as follows: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ^★P < 0.05; ns denotes not significant. In panel (H), significance is indicated as * (WT vs. 5×FAD) and ^★ (5×FAD vs. Il6^−/−:5×FAD), with ns denoting no significant difference between WT and Il6^−/− mice

IL-6 deficiency alleviates Aβ burden in the hippocampus of 5×FAD mice

Given the cognitive improvements observed in Il6^−/−:5×FAD mice, we next explored whether IL-6 deficiency influences Aβ accumulation, a hallmark of AD pathology. To address this, we employed immunostaining techniques to quantify the fraction of Aβ-positive areas in the hippocampus using various antibodies. We then compared the Aβ burden between 5×FAD mice and Il6^−/−:5×FAD mice. Aβ is a peptide generated through the enzymatic cleavage of amyloid precursor protein (APP) [44, 45]. The length of Aβ varies depending on the cleavage sites, with Aβ40 and Aβ42 being the most common isoforms [44, 45]. To detect Aβ, we used the 6E10 monoclonal antibody, which specifically recognizes the amino terminus of Aβ and can detect multiple isoforms, including Aβ40 and Aβ42. Additionally, to distinguish between Aβ isoforms and their distributions, we utilized antibodies specific to Aβ40 and Aβ42. Our results demonstrated a marked reduction in the 6E10 antibody-positive staining area in Il6^−/−:5×FAD mice compared to 5×FAD mice (Fig. 5A, D), indicating a decrease in overall Aβ burden. Furthermore, the reduction in staining with the Aβ40-specific antibody suggests a notable decrease in Aβ40 accumulation in Il6^−/−:5×FAD mice (Fig. 5B, E). Similarly, decreased Aβ42 accumulation was observed using the Aβ42-specific antibody (Fig. 5C, F). Collectively, these findings indicate that IL-6 deficiency substantially reduces Aβ burden in 5×FAD mice, lowering not only total Aβ accumulation (detected by the 6E10 antibody) but also the specific accumulation of Aβ40 and Aβ42 (detected by their respective antibodies).

In this study, we employed not only specific antibody staining techniques but also Thioflavin S (ThS) to quantify Aβ burden. ThS is a widely used fluorescent dye that binds specifically to Aβ fibrils, emitting fluorescence upon binding, thereby allowing visualization of Aβ deposition [46, 47]. Aβ fibrils accumulate in the extracellular space between neurons, eventually forming plaques—one of the key pathological hallmarks of AD [48, 49]. By employing ThS staining, we quantitatively assessed changes in Aβ burden under different experimental conditions to evaluate the impact of IL-6 deficiency on Aβ accumulation. The results revealed a marked reduction in the ThS-positive staining area fraction in the hippocampus of Il6^−/−:5×FAD mice compared to 5×FAD mice (Fig. 5G, H), indicating a substantial decrease in Aβ fibril deposition. Combined with the antibody staining data, these findings suggest that IL-6 deficiency effectively reduces Aβ burden, including Aβ fibril deposition, in the hippocampus of 5×FAD mice.

Fig. 5

IL-6 deficiency alleviates Aβ burden in 5×FAD mice. (A-C) Representative images showing antibody-positive staining areas for 6E10 (A), Aβ40 (B), and Aβ42 (C) in hippocampal tissue from 6-month-old 5×FAD and Il6^−/−:5×FAD mice. (D-F) Quantification of antibody-positive staining areas for 6E10 (D), Aβ40 (E), and Aβ42 (F). (G) Representative images showing ThS-positive staining areas in hippocampal tissue from the same groups. (H) Quantification of the ThS-positive staining areas in hippocampal tissue from the same groups. Each dot represents an individual mouse. Panels (D), (E), (F), and (H): n = 5 mice per group (mixed sexes). Statistical analysis was performed using Student’s t-test or Welch’s t-test. Data are presented as mean ± S.E.M. Significance levels: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns indicates not significant. (A-C) Scale bar: 250 μm and 25 μm. (G) Scale bar: 50 μm

IL-6 deficiency attenuates neuroinflammation in the hippocampus of 5×FAD mice

Emerging evidence indicates that reactive microgliosis and astrogliosis are central pathological features of AD [50]. These glial cells secrete a range of inflammatory mediators, including cytokines and chemokines, which amplify neuroinflammatory responses. In AD research, Ionized Calcium Binding Adaptor Molecule 1 (IBA1) and Glial Fibrillary Acidic Protein (GFAP) are widely used as markers for microglia and astrocytes, respectively [32]. Elevated levels of IBA1 and GFAP are typically indicative of reactive microgliosis and astrogliosis, which are key components of neuroinflammation. In this study, we assessed neuroinflammatory changes in 5×FAD mice, both with and without IL-6 deficiency, by analyzing IBA1 and GFAP expression in the hippocampus.

Immunohistochemical staining for IBA1 and GFAP was performed on brain tissue from both 5×FAD and Il6^−/−:5×FAD mice, followed by quantitative analysis using Western blotting (WB). The results showed a significant reduction in IBA1-positive microglia within the dentate gyrus (DG) and CA1 regions of the hippocampus in Il6^−/−:5×FAD mice compared to 5×FAD mice (Fig. 6A). Additionally, IBA1 protein levels in hippocampal tissue were markedly lower in Il6^−/−:5×FAD mice than in 5×FAD mice (Fig. 6C and E). However, the number of GFAP-positive astrocytes remained unchanged (Fig. 6B), a finding corroborated by WB analysis (Fig. 6D and F). Given the reduction in IBA1-positive cells, we hypothesized that this decrease might influence cytokine production. To investigate this, we measured the expression levels of the inflammatory cytokine IL-1β. The results indicated a substantial increase in IL-1β expression in the hippocampal tissue of 5×FAD mice compared to WT mice (Fig. 6G). However, in Il6^−/−:5×FAD mice, IL-1β expression in the hippocampus was markedly reduced compared to 5×FAD mice (Fig. 6G). These findings, combined with the results from Fig. 3G, which demonstrate that IL-6 deficiency reduces mRNA levels of several ISGs, suggest that IL-6 deficiency mitigates the neuroinflammatory response in the hippocampus of 5×FAD mice.

Fig. 6

IL-6 deficiency attenuates neuroinflammation in 5×FAD mice. (A-B) Representative images showing IBA1 (A) and GFAP (B) staining in the dentate gyrus (DG) and CA1 regions of hippocampal tissue from 5×FAD and Il6^−/−:5×FAD mice. (C-D) Western blot analysis of IBA1 (C) and GFAP (D) in hippocampal tissue from the same groups. (E-F) Quantification of IBA1 (E) and GFAP (F) levels shown in panels C and D. (G) ELISA analysis of IL-1β in hippocampal tissue from the same groups. Each dot represents an individual mouse. Panels (C-F): n = 6 mice per group (mixed sexes). Panel (G): n = 5 mice per group (mixed sexes). Statistical significance for panels (E-F) was determined using Student’s t-test or Welch’s t-test. For panels (G), one-way ANOVA followed by Tukey’s multiple comparisons test was employed. Data are presented as mean ± S.E.M. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Ns: not significant. Scale bars: (A-B) 25 μm

IL-6 deficiency attenuates neuroinflammation in the hippocampus of 5×FAD mice via the STAT3-cGAS-STING pathway

We investigated the underlying mechanisms contributing to the observed reduction in neuroinflammation in Il6^−/−:5×FAD mice. Previous findings have shown substantially decreased activity in the STAT3 and cGAS-STING pathways in these mice. Based on this evidence, we hypothesize that the reduction in neuroinflammation is closely linked to the diminished activity of these two pathways. STAT3, a known transcription factor, may regulate the cGAS-STING pathway by modulating the expression of cGAS and STING. We propose that IL-6 deficiency leads to decreased STAT3 activity, which in turn reduces cGAS-STING pathway activity, ultimately alleviating neuroinflammation in Il6^−/−:5×FAD mice. To test this hypothesis, we analyzed the mRNA expression levels of Cgas and Sting in the hippocampus of 6-month-old 5×FAD mice, both prior to and following IL-6 deficiency. Our results showed that, compared to WT mice, the mRNA expression levels of Cgas (Fig. 7A) and Sting (Fig. 7B) were markedly elevated in the hippocampus of 6-month-old 5×FAD mice, consistent with the increased protein levels of cGAS and STING observed in these mice. Conversely, the mRNA expression levels of Cgas (Fig. 7C) and Sting (Fig. 7D) were significantly reduced in the hippocampus of 6-month-old Il6^−/−:5×FAD mice compared to 5×FAD mice, corresponding to the decrease in cGAS and STING protein levels following IL-6 deficiency. These changes suggest that STAT3 may play a crucial role in the transcriptional regulation of these genes.

To directly assess the involvement of STAT3 in this regulatory process, we performed chromatin immunoprecipitation sequencing (ChIP-Seq) on the hippocampus of 5×FAD and Il6^−/−:5×FAD mice. Using a STAT3-specific antibody, we isolated DNA fragments bound by STAT3 and conducted high-throughput sequencing. The subsequent analysis of the sequencing data identified STAT3 binding peaks across the genome, corresponding to specific STAT3 binding sites. This approach enabled us to determine whether STAT3 binds to the promoter regions of the Cgas and Sting genes, thereby confirming its role in directly regulating their transcription. We detected significant STAT3 binding peaks in the promoter regions of both Cgas and Sting genes in the hippocampus of 5×FAD mice, indicating that STAT3 directly regulates the transcription of these genes (Fig. 7E, F). However, in the hippocampus of Il6^−/−:5×FAD mice, although there was a marked alteration in the STAT3 binding peak at the Sting gene promoter, the binding peak at the Cgas gene promoter remained largely unchanged (Fig. 7E, F). To precisely quantify these changes, we identified the DNA sequences corresponding to the STAT3 binding peaks in the promoter regions of Cgas and Sting from the ChIP-seq data and designed specific qPCR primers targeting these sequences. We then conducted ChIP-qPCR experiments to measure the relative fold enrichment of STAT3 binding to the Cgas and Sting genes and to evaluate the impact of IL-6 deficiency on STAT3 binding abundance. In the hippocampus of 5×FAD mice, we observed substantial fold enrichment of STAT3 binding at the Cgas and Sting genes compared to isotype control antibodies (Fig. 7G). However, in the hippocampus of Il6^−/−:5×FAD mice, the fold enrichment of STAT3 at these genes did not differ from that of the isotype control antibodies (Fig. 7H). Combining the ChIP-seq and ChIP-qPCR results, we demonstrate that STAT3 directly binds to the promoter regions of the Cgas and Sting genes, thereby regulating their transcription. These findings suggest the presence of a novel STAT3-cGAS-STING regulatory pathway.

To further validate the involvement of this pathway, we conducted experiments on BV2 microglial cells, exposing them to varying concentrations and treatment durations of mouse recombinant IL-6 protein (mIL-6). In untreated BV2 cells, the expression levels of p-STAT3, STAT3, cGAS, and STING were nearly undetectable (Fig. 7I). However, after 48 h of mIL-6 treatment at a concentration of 5000 pg/mL, there was a pronounced increase in the expression levels of p-STAT3, STAT3, cGAS, and STING (Fig. 7I). This finding supports the idea that IL-6 can activate the STAT3-cGAS-STING pathway in microglia. In the context of AD, Aβ is recognized as a major source of damage-associated molecular patterns (DAMPs) [51]. As a DAMP, Aβ interacts with pattern recognition receptors (PRRs) on the surface of microglia, triggering their activation and the release of inflammatory cytokines [52]. To investigate whether Aβ can activate the STAT3-cGAS-STING pathway, we treated BV2 microglial cells with different concentrations of oligomeric Aβ42 and measured the expression levels of p-STAT3, STAT3, cGAS, and STING after 24 and 48 h of treatment. The results showed that after 48 h of treatment with Aβ42 at concentrations of 10 µM and 20 µM, these protein levels were considerably upregulated compared to untreated cells (Fig. 7I). This finding suggests that Aβ42 can indeed activate the STAT3-cGAS-STING pathway in microglia. We also evaluated changes in mRNA expression levels of Cgas and Sting in BV2 cells after 48 h of treatment with Aβ42 (10 µM and 20 µM) and mIL-6 (5000 pg/mL). The results demonstrated a marked increase in Cgas and Sting mRNA levels in the treated BV2 cells compared to untreated controls (Fig. 7J). This elevation in mRNA levels further supports the role of STAT3 in regulating the transcription of Cgas and Sting, thereby activating the STAT3-cGAS-STING pathway.

Given the coexistence of IL-6 and Aβ42 in AD model mice, both factors may trigger the STAT3-cGAS-STING pathway, potentially amplifying the neuroinflammatory response associated with AD. To investigate the impact of this pathway on inflammatory changes, we treated BV2 cells with mIL-6 (5000 pg/mL) and Aβ42 (10 µM and 20 µM). After 48 h of treatment, we observed a significant upregulation in the mRNA levels of Il1b in BV2 cells treated with mIL-6 and Aβ42 compared to untreated controls (Fig. 7K). Moreover, the mRNA levels of downstream targets of the cGAS-STING pathway, including Ifnb1 (Fig. 7L) and several ISGs such as Isg15, Isg20, Ifit1, and Ifit2 (Fig. 7L), were substantially elevated. These results suggest that activation of the STAT3-cGAS-STING pathway leads to a pronounced increase in inflammatory mediators. Conversely, IL-6 deficiency appears to suppress this pathway’s activation, thereby reducing the neuroinflammatory response in 5×FAD mice.

Fig. 7

IL-6 deficiency attenuates neuroinflammation in 5×FAD mice via the STAT3-cGAS-STING pathway. (A-B) qRT-PCR analysis of Cgas (A) and Sting (B) mRNA levels relative to Gapdh in hippocampal tissue from 6-month-old 5×FAD mice compared to WT mice. (C-D) qRT-PCR analysis of Cgas (C) and Sting (D) mRNA levels relative to Gapdh in hippocampal tissue from 6-month-old Il6^−/−:5×FAD mice compared to 5×FAD mice. (E-F) Peak distribution of STAT3 binding within the promoter regions of Cgas (E) and Sting (F), visualized using the Integrative Genomics Viewer (IGV) after ChIP-seq experiments on hippocampal tissues from 5×FAD and Il6^−/−:5×FAD mice. (G-H) ChIP-qPCR of hippocampal tissues from 5×FAD (G) and Il6^−/−:5×FAD mice (H) using a STAT3 antibody and isotype control IgG to evaluate fold enrichment levels on the Cgas and Sting gene fragments. (I) Western blot analysis of p-STAT3, STAT3, cGAS, and STING in BV2 cells treated with different concentrations of mIL-6 or Aβ42 for 24–48 h. (J-L) qRT-PCR analysis of mRNA levels of Cgas (J), Sting (J), Il1b (K), Ifnb1 (L), and ISGs (Isg15, Isg20, Ifit1, Ifit2) (L) relative to Gapdh in BV2 cells treated with mIL-6 (5000 pg/mL) and Aβ42 (10 µM or 20 µM) for 48 h. For panels (A-D), n = 5 mice per group (mixed sexes), with each dot representing a mouse. For panels (E-H), n = 1 mouse per group. For panels (G-H) and (J-L), each dot represents a well. For panel (I), each lane represents a group. Data are presented as mean ± S.E.M. Statistical significance was determined using Student’s t-test or Welch’s t-test for panels (A-D) and (G-H), and one-way ANOVA followed by Tukey’s multiple comparisons test or the Games-Howell test for panels (J-L). Significance levels are indicated as follows: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns indicates not significant

Discussion

Recent studies have indicated that neuroinflammation not only plays a crucial role in the pathological progression of AD but may also serve as a key pathogenic mechanism. Our research shows that the IL-6-STAT3 and cGAS-STING pathways are co-activated in AD mice and closely associated with disease progression, emphasizing their role in neuroinflammation. In IL-6-deficient AD mice, we observed a marked reduction in the activation of both the STAT3 and cGAS-STING pathways, which correlated with improved cognitive function, reduced Aβ deposition, and decreased neuroinflammation. Additionally, our study showed that STAT3 directly regulates the transcription of Cgas and Sting. Based on this discovery, we propose the novel concept of the STAT3-cGAS-STING pathway. These findings suggest that targeting this pathway may represent a promising approach for mitigating neuroinflammation in AD.

We utilized the 5×FAD mouse model, a well-established model of AD that carries five human AD-related mutations in the APP and PS1 genes. These include the Swedish (K670N/M671L), Florida (I716V), and London (V717I) mutations in the APP gene, along with the M146L and L286V mutations in the PS1 gene [53].This model exhibits rapid and severe accumulation of Aβ plaques, accompanied by pronounced neurodegeneration [53]. Given that neuroinflammation is not only a hallmark of AD but also actively contributes to its onset and progression, we conducted a time-course analysis in 5×FAD mice at 4, 6, 9, and 12 months of age to dynamically assess the activation of the IL-6-STAT3 and cGAS-STING pathways. Additionally, we compared the activation of these pathways across these age groups. Our results collectively suggest that both the IL-6-STAT3 and cGAS-STING pathways play an active role in AD progression. These findings are consistent with other studies that have reported the activation of the IL-6-STAT3 pathway [24] and the cGAS-STING pathway [32] in AD. The co-activation of these pathways may establish a complex inflammatory network that amplifies the immune response and accelerates the pathological progression of AD. Subsequent research confirmed the existence of the STAT3-cGAS-STING pathway, offering a molecular mechanism for the increased expression of cGAS and STING observed in AD.

Moreover, our results showed that Il6 gene knockout reduces the Aβ burden in 5×FAD mice. Considering that Il6 gene knockout markedly weakens the STAT3-cGAS-STING pathway in these mice, and prior studies have indicated that cGAS deficiency can alleviate amyloid pathology in the 5×FAD model [28], this suggests that the attenuation of the STAT3-cGAS-STING pathway may underlie the observed reduction in amyloid pathology. However, further experiments are necessary to directly confirm whether the decreased Aβ load is specifically due to the weakening of this pathway, which will provide a more accurate understanding of the molecular mechanisms involved and their contributions to the observed effects. Additionally, as the reduction in amyloid burden could result from either decreased production or enhanced clearance of Aβ, our current findings do not fully clarify the underlying mechanism. Future studies should investigate both amyloid protein processing and the pathways involved in their clearance to better elucidate the factors responsible for the observed reduction in amyloid burden.

Additional animal studies have shown that inhibiting IL-6 trans-signaling reduces Aβ burden in the cortex and hippocampus of Tg2576 female mice [54], a finding that aligns with our experimental results. Given that our study utilized Il6 gene knockout mice, which abrogates both the classical and trans-signaling functions of IL-6 [55], the specific signaling mechanism through which Il6 knockout leads to the attenuation of the STAT3-cGAS-STING pathway and the subsequent reduction in Aβ burden in Il6^−/−:5×FAD mice remains to be elucidated. Further investigation is required to clarify the precise pathways involved in these observed effects.

Notably, our cellular study provided additional evidence that both IL-6 and Aβ42 can activate the STAT3-cGAS-STING pathway, resulting in the upregulation of various inflammatory mediators. In AD, the coexistence of IL-6 and Aβ42 may synergistically activate this pathway, thereby exacerbating neuroinflammation. This suggests that activation of the STAT3-cGAS-STING pathway plays a pivotal role in driving AD-related neuroinflammation. Additionally, glial cell proliferation and activation are critical components of neuroinflammation in AD. Our experiments proved that IL-6 deficiency significantly reduces the number of IBA1-positive microglia in 5×FAD mice, while leaving the number of GFAP-positive astrocytes unaffected. This observation aligns with previous research indicating that a transgenic mouse model with targeted IL-6 overexpression in astrocytes (GFAP-IL6Tg) exhibits increased microglial density [56]. Our findings further support this by demonstrating that IL-6 deficiency in 5×FAD mice leads to a reduction in IBA1-positive microglia. Additionally, several studies have emphasized the critical role of the cGAS-STING pathway in regulating microglial function [31, 57,58,59,60]. For example, in a cerebral venous sinus thrombosis (CVST) mouse model, administration of the selective cGAS inhibitor RU.521 substantially reduced the number of IBA1-positive microglia [61]. Considering the reduced activity of the STAT3-cGAS-STING pathway in Il6^−/−:5×FAD mice, the unchanged number of GFAP-positive astrocytes suggests the involvement of alternative mechanisms or compensatory pathways in astrocyte regulation. Further investigation is necessary to clarify the mechanisms underlying this differential response.

Existing literature reports that 5×FAD mice begin to exhibit Aβ deposition at 2 months of age [53, 62], with cognitive impairment becoming evident by 4 months [53, 62]. Consequently, we focused our study on 5-month-old IL-6-deficient AD mice. The experimental results demonstrate that IL-6 deficiency leads to significant improvements in both molecular pathology, including neuroinflammation, and behavioral outcomes. These findings suggest that neuroinflammation occurs in the early stages of AD, with IL-6 potentially playing a direct role in modulating the inflammatory response. Investigating the role of IL-6 during these initial stages may enhance our understanding of how neuroinflammation accelerates disease progression. However, since our experiments are limited to the early phases of AD, further research is essential to explore the role of IL-6 in later stages, which may impede a comprehensive understanding of its overall impact on the disease. Future studies should, therefore, consider extending the observation period to examine the effects of IL-6 deficiency on AD mice across various stages of the disease.

Overall, our research demonstrates that IL-6 deficiency improves cognitive function, reduces Aβ plaque accumulation, and alleviates neuroinflammation in 5×FAD mice. Notably, IL-6 deficiency was shown to mitigate neuroinflammation through the STAT3-cGAS-STING pathway. These findings enhance our understanding of the molecular mechanisms underlying neuroinflammation in AD, providing valuable insights for targeting neuroinflammation and contributing to the development of more effective therapeutic strategies for AD.

Materials and methods

Resource table

Antibody are listed in Table 1.

Table 1 Antibodies

Contact for reagent and resource sharing

zhangling@cnilas.org.

Experimental models and subject details

We utilized ChatGPT-4 for English editing during the writing of this article.

Generation of mouse lines

The Il6^+/− and 5×FAD mice used in this study were provided by Institute of Laboratory Animal Sciences, CAMS & Comparative Medicine Center, PUMC. To generate Il6^+/−:5×FAD offspring, Il6^+/− mice were crossed with 5×FAD mice. These Il6^+/−:5×FAD mice were then crossed with Il6^+/− mice to produce wild-type (WT) littermates, Il6^−/−, Il6^−/−:5×FAD, and 5×FAD mice. All animals were housed in specific pathogen-free (SPF) facilities under a 12-hour light/12-hour dark cycle, with a controlled temperature of 22–25 °C and humidity levels of 50–60%. They were housed in groups of 4–5 per cage and provided with standard mouse chow (Beijing Keao Xieli Feed Co., Ltd.) and sterile water.

Molecular experiments were conducted on WT and 5×FAD mice at 4, 6, 9, and 12 months of age. Previous studies have reported cognitive impairment in 5×FAD mice as early as 4 months old [53]. For this study, 5-month-old WT mice, Il6^−/−mice, Il6^−/−:5×FAD mice, and 5×FAD mice were selected for behavioral experiments, with molecular analyses performed when the mice reached 6 months of age.

Cell culture

BV2 cells, an immortalized mouse microglial cell line, were obtained from the Institute of Laboratory Animal Sciences, CAMS & Comparative Medicine Center, PUMC. BV2 cells are widely used as a substitute for primary microglial cells due to their similar functions and fundamental characteristics [63]. The cells were cultured in high-glucose DMEM medium (11995065, Gibco™) supplemented with 10% fetal bovine serum (FBS) (10091155, Gibco™) and 1% penicillin-streptomycin (PS) (15140122, Gibco™). Cultures were maintained in an incubator (BB150, Thermo Scientific™) at 37 °C with 5% CO2 and saturated humidity. When cell confluence reached 75–85%, cells were digested with 0.25% trypsin-EDTA (25200056, Gibco™). The digestion process was halted by adding DMEM medium containing 10% FBS and 1% PS. The cells were then resuspended by pipetting, centrifuged, and passaged at a 1:5 ratio.

Method details

Behavioral testing

Open field test (OFT)

The Open Field Test (OFT) was conducted in a box constructed from opaque white material, measuring 50 cm × 50 cm × 30 cm. The floor of the box was divided into central, intermediate, and peripheral zones based on proximity to the edges. To minimize anxiety, the mice were placed on a temporary platform for 1 h prior to the start of the experiment for acclimatization. During the test, each mouse was positioned in the center of the box, with its back facing the wall, and its spontaneous activity was recorded for 5 min using a camera mounted above the box. To ensure a stable experimental environment, the room was kept quiet during testing. After each test, the mouse was gently removed and returned to its cage for rest. Diffuse lighting was used in the laboratory, and the experimental box was thoroughly cleaned with 75% ethanol after each mouse to minimize odor transfer between animals. The EthoVision XT9 system was employed to analyze the mice’s locomotor activity, recording metrics such as total distance traveled, movement speed, distance traveled in the central arena, and time spent in the central arena.

Elevated plus maze test (EPM)

The Elevated Plus Maze (EPM) consists of two open arms and two closed arms, forming a cross-shaped structure with the intersection serving as the central zone. The entire apparatus is elevated 50 cm above the ground. At the start of the experiment, the test animal is placed in the central zone of the maze, facing the open arms, and allowed to explore freely for 5 min, with each animal starting from the same position. After each trial, the maze is thoroughly cleaned with 75% ethanol to eliminate odor residues that could influence subsequent experiments. The EthoVision XT9 system was employed to analyze the animals’ movements, recording the number of entries into the open arms as well as the time spent in the open arms.

Morris water maze test (MWM)

The Morris Water Maze (MWM) consists of a circular water tank with a diameter of 120 cm and a height of 40 cm. The inner walls and bottom of the tank are black, and it is filled with water dyed with edible white pigment, with the water surface positioned 30 cm below the edge. The water temperature is maintained at 22 ± 1 °C. A platform with a diameter of 10 cm and a height of 23.5 cm is submerged 1.5 cm below the water surface. The tank is evenly divided into four quadrants, each marked with distinctively shaped stickers as visual cues, with release points located at the center of each quadrant’s arc.

The experiment consists of two phases: training and probing. The training phase spans 5 to 7 days, during which each mouse is released daily from three different quadrants, completing three trials per day. Upon release, the mouse is placed into the water tank facing the wall, with each trial lasting 1 min. If the mouse successfully locates the platform and remains on it for 5 s, the time taken to find the platform is recorded as the escape latency. If the mouse fails to find the platform within 1 min, the escape latency is recorded as 1 min. Regardless of the outcome, the mouse stays on the platform for 20 s. The training phase concludes when the WT mice achieve a 100% success rate in locating the platform, after which the probing phase begins. During the probing phase, the platform is removed, and the mouse swims freely for 1 min. The EthoVision XT9 system is employed to analyze the mouse’s movements, recording metrics such as swimming speed, escape latency, and the number of platform crossings.

Immunohistochemical staining

Mice were first administered intraperitoneal injections of Zoletil 50 (Virbac), followed by perfusion through the left ventricle with cold phosphate-buffered saline (PBS, ZLI-9061, ZSGB-BIO). After perfusion, the entire brain was quickly extracted post-decapitation. The brain tissue was bisected along the midline; one half was placed in a 15 mL centrifuge tube (601052, NEST) containing 4% paraformaldehyde (BL539A, Biosharp) for 48 h of fixation. Following fixation, the tissue was dehydrated through a gradient of alcohols (Beijing Yili Fine Chemicals Co., Ltd), cleared with xylene (Beijing Yili Fine Chemicals Co., Ltd), and embedded in paraffin (Leica) to create paraffin blocks. These blocks were sectioned into slices approximately 3–5 micrometers thick and mounted on glass slides (188105 W, Citotest Scientific Co., Ltd) for subsequent staining and analysis. The hippocampal and cortical tissues were isolated from the other half of the brain, placed in 1.5 mL centrifuge tubes (MCT-150-C, Axygen), rapidly frozen in liquid nitrogen, and stored at -80 °C (905, Thermo Scientific™) for further experimental preparations.

The tissue sections were deparaffinized by immersion in xylene and subsequently dehydrated through a graded ethanol series (100% ethanol ×2, 95% ethanol, 80% ethanol, 50% ethanol) before rehydration in distilled water. Antigen retrieval was performed using Citrate Buffer (ZLI-9065, ZSGB-BIO). To block nonspecific binding, the sections were treated with goat serum (ZLI-9056, ZSGB-BIO). The sections were then incubated with the primary antibody (details provided in Table 1) at 4 °C for 16 h. Following incubation, the sections were washed three times with PBS, with each wash lasting 5 min—a protocol that was maintained for all subsequent steps. The sections were then incubated with a biotin-labeled secondary antibody and horseradish peroxidase-conjugated streptavidin, following the manufacturer’s instructions (PV-9001, PV-9002, ZSGB-BIO). Chromogenic detection was achieved using DAB (ZLI-9017, ZSGB-BIO), followed by counterstaining with hematoxylin (ZLI-9610, ZSGB-BIO). Finally, the sections underwent standard dehydration and clearing procedures before being mounted with neutral balsam, and the slides were scanned using a digital slide scanner (NanoZoomer S60).

Western blotting(WB)

The procedure for collecting hippocampal tissue for Western blot (WB) analysis follows the protocol outlined in the immunohistochemical staining section. Cells exposed to different treatments were digested with trypsin, neutralized with culture medium, and centrifuged at 3000 rpm to collect the cell pellet. Radio-immunoprecipitation assay (RIPA) lysis buffer (P0013B, Beyotime), supplemented with protease inhibitors (B14002, Selleck) and phosphatase inhibitors (B15002, Selleck), was added to the tissue or cell samples. The samples were then homogenized using an ultrasonic crusher (Qsonica Q700). After centrifugation at 12,000 g for 30 min at 4 °C (TOMY MX-307), the supernatant was collected. Protein concentration in the supernatant was determined using a BCA protein assay kit (23227, Thermo Scientific™) and adjusted to 5 µg/µL. SDS-PAGE sample loading buffer (P0015L, Beyotime) was added, and the samples were heated at 100 °C for 10 min in a dry thermostat metal bath (GL-1600) to ensure complete protein denaturation.

The denatured protein samples were stored at -80 °C until use. Depending on the experimental requirements, 20–40 µg of protein was loaded onto polyacrylamide gels (PG111 and PG114, Epizyme) and separated by tris-glycine electrophoresis (P0014B, Beyotime). The proteins were then transferred to a PVDF membrane (IPVH00010 and ISEQ00010, Sigma) using transfer buffer (WB4600, NCM Biotech). To block nonspecific binding, the membranes were incubated with 5% non-fat milk (P0216-1500 g, Beyotime). Subsequently, the membranes were incubated with the primary antibody (details in Table 1) at 4 °C for 16 h, followed by washing with TBST buffer (T1082, Solarbio). HRP-conjugated secondary antibodies (ZB-2301 and ZB-2305, ZSGB-BIO) were then added, and after a 1-hour incubation at room temperature, the membranes were washed again with TBST. The target protein bands were detected using an ECL chemiluminescence kit (P10300, NCM Biotech) and visualized with a chemiluminescent imaging system (ChemiDoc™ XRS + System, Bio-Rad; Tanon-1600).

Real-time quantitative polymerase chain reaction (qRT-PCR)

The procedure for collecting hippocampal tissue for qRT-PCR follows the protocol outlined in the immunohistochemical staining section. Total RNA was extracted from hippocampal tissue or cells using the standard TRIzol method (15596026, Thermo Scientific™). RNA quantification was performed using a DeNovix DS-11 Plus Spectrophotometer, after which the RNA was reverse-transcribed into cDNA using a reverse transcription kit (RR047A, Takara). The primer sequences used for amplification are provided in Table 2. PCR amplification was conducted according to the manufacturer’s instructions (RR820A, Takara) using an Applied Biosystems QuantStudio 3 system.

Table 2 Primers used for qRT-PCR and Chip- qPCR

Following the amplification reaction, melting curve analysis was performed to verify the specificity of the primers. The qRT-PCR results were expressed as fold changes, calculated using the formula: Fold changes = 2^−Δ(ΔCt), where ΔCt = Ct(target gene) − Ct(gapdh), and Δ(ΔCt) = ΔCt(experimental group) − ΔCt(control group).

Enzyme-linked immunosorbent assay (ELISA)

The procedure for collecting hippocampal tissue for ELISA follows the protocol outlined in the immunohistochemical staining section. The tissue was treated with PBS (P0013B, Beyotime) containing protease inhibitors, lysed, and the supernatant was collected. Protein quantification was performed as described in the Western blot (WB) section, with concentrations standardized to 4 µg/µL. IL-1β levels in the tissue supernatant were measured using an ELISA kit (E-EL-M0037, Elabscience) according to the manufacturer’s instructions. The results were expressed as picograms of IL-1β per milligram of total protein (pg/mg).

Chromatin immunoprecipitation sequencing (ChIP-Seq)

Following the instructions provided with the ChIP kit (26156, Thermo Scientific™), the procedure is as follows: Transfer frozen hippocampal tissue from mice, stored at -80 °C, to a 100 mm culture dish (704201, NEST). Mince the tissue into 1–2 mm fragments using a scalpel or scissors, ensuring the tissue remains on ice throughout the process. Fix the minced tissue with formaldehyde solution (F8775, Sigma) to a final concentration of 1% at room temperature for 15 min. To quench the crosslinking reaction, add glycine (BS082, White Shark) to a final concentration of 125 mM and incubate at room temperature for 5 min.

Centrifuge the fixed tissue samples at 1,200 g for 5 min at 4 °C to collect the tissue. Wash the samples twice with pre-chilled PBS containing protease inhibitors, centrifuging after each wash. After removing the supernatant, proceed with chromatin fragmentation, immunoprecipitation, DNA recovery, and purification according to the kit instructions. The purified DNA can then be used for high-throughput sequencing on the Illumina Novaseq 6000 platform.

Chromatin immunoprecipitation quantitative PCR (ChIP-qPCR)

Using the purified DNA from the ChIP experiment as a template, perform a qPCR experiment. The primer sequences are provided in Table 2. A 10% input sample is used as the control. The results are expressed as fold enrichment relative to the % input, calculated using the following formula: fold enrichment = 10% × 2^[Ct(Input) - Ct(IgG/STAT3)].

Cell treatment

BV2 cells were treated with recombinant mouse IL-6 protein (216 − 16, Peprotech) at concentrations of 500, 1000, and 5000 pg/mL, or with oligomeric Aβ42 (10 µM and 20 µM, Fooyuelab) for 24–48 h. After treatment, cellular proteins or RNA were extracted for further analysis.

Preparation of Aβ42 oligomers

Following the method described in the literature [64], dissolve lyophilized Aβ42 powder (fooyuelab), previously stored at -20 °C, in hexafluoroisopropanol (HFIP, Sigma) while keeping the solution on ice. For each milligram of Aβ42, add 222 µl of HFIP, seal the tube, and vortex thoroughly. Allow the solution to rest at room temperature for 1 h until it becomes clear. Afterward, aliquot 55 µl of the solution into 1.5 ml EP tubes, and use a freeze dryer (CHRIST freeze dryer) to evaporate the HFIP, resulting in a colorless, transparent Aβ peptide film. This film can be stored at -20 °C or used for further experimental procedures.

To each aliquot, add 11 µl of dimethyl sulfoxide (DMSO, Sigma) and perform water bath sonication for 10 min to obtain a 5 mM Aβ-DMSO solution. Then, add 539 µl of pre-cooled PBS to achieve a final concentration of 100 µM, and vortex the solution. Incubate the mixture at 4 °C for 24 h, and then centrifuge at 14,000 rpm for 10 min to collect the supernatant containing soluble Aβ42 oligomers. To further promote oligomerization, incubate the solution at 4 °C for an additional week. The prepared Aβ oligomer solution can be stored at -80 °C. Prior to use, centrifuge the solution at 13,000 rpm for 10 min at 4 °C, and dilute it to the desired concentration (10 µM or 20 µM) using DMEM medium for experimental use.

Thioflavin S (ThS) staining

Begin by performing standard deparaffinization and rehydration of the paraffin sections. Permeabilize the sections with 0.1% Triton™ X-100 (9036-19-5, Sigma) at room temperature for 15 min. Prepare a 0.5% Thioflavin S (ThS) staining solution (T-1892, Sigma) in 50% ethanol, filtering the solution before use. Incubate the sections in the ThS staining solution at room temperature for 2 min. Wash the slides five times with 70% ethanol, each wash lasting 3 min, followed by three washes with 50% ethanol, each for 3 min. Finally, wash the slides twice with purified water for 15 min each. Air-dry the slides for 15 to 30 min. Once dried, mount the slides using an anti-fading mounting medium (P0126-25 ml, Beyotime) and capture images using a fluorescence microscope (Leica THUNDER DMi8).

Statistical analyses

Statistical analyses were conducted using SPSS Statistics version 26. For comparisons between two groups, either the two-tailed Student’s t-test or Welch’s t-test was used, depending on the equality of variances. Differences among multiple groups were evaluated using one-way ANOVA, followed by post hoc tests, including Tukey’s, Scheffé’s, or Games-Howell tests. A significance level of P < 0.05 was considered statistically significant. All data are presented as mean ± S.E.M.