How Controlling a Signaling Pathway Affects Support Cell Gathering Around Amyloid Beta Plaques

Our previous studies have shown that CMKLR1 participates in the AD pathological process by mediating neuronal tau propagation [26] and microglia migration [28]. Here we further explored the role and mechanism of the astrocyte-expressed CMKLR1 in AD pathology. We first examined the expression of CMKLR1 and its colocalization with astrocytes in the brains of 9-month-old APP/PS1 transgenic mice. Immunofluorescence staining was conducted with anti-CMKLR1 and anti-glial fibrillary acidic protein (GFAP) antibodies to detect CMKLR1 and astrocytes, respectively. As shown in Figure 1A–D, significantly elevated expressions of CMKLR1 (green fluorescence) and GFAP (red fluorescence) were observed in the cortex as well as in the dentate gyrus (DG), cornu ammonis-1 (CA1), and stratum lacunosum-moleculare (SLM) regions of the hippocampus of APP/PS1 mice, compared to findings in WT littermates. Furthermore, the number of astrocytes co-localized with CMKRL1 (yellow fluorescence) increased markedly in these brain regions, especially the SLM region of the hippocampus, in APP/PS1 mice, compared to WT mice (Figure 1E).

We further tested the expression of chemerin, the endogenous ligand of CMKLR1, in AD mouse brains. Compared to WT mice, chemerin was significantly reduced in APP/PS1 mouse brains (Figure 2A,B, Supplementary Figure S1). To characterize the cellular distribution of chemerin, double immunofluorescence staining was conducted for chemerin and MAP2 (the neuron-specific molecular). Scatter plots generated from the images demonstrated strong co-localization between chemerin and MAP2, especially in the CA1 regions and cortices of wild-type mice (Figure 2C,D). Chemerin showed significantly less co-localization with neurons, as indicated by MAP2, in the CA1 and cortex of APP/PS1 mice compared to wild-type mice (Figure 2C,D). Additionally, the same trend was observed in the DG and SLM of the hippocampus (Supplementary Figure S2). This result indicates transcellular regulation of the chemerin/CMKLR1 axis in the brain. Additionally, Pearson correlation coefficients between chemerin and MAP2 expression revealed that chemerin was predominantly co-localized with neurons in WT mice (Figure 2D), indicating that neuronally derived chemerin may target astrocyte-expressed CMKLR1 to exert immune functions in the CNS. We have shown that Aβ is another endogenous ligand of CMKLR1 [25], suggesting that the Aβ/CMKLR1 axis is enhanced in AD pathology. These findings suggest the imbalance of the chemerin/CMKLR1 and Aβ/CMKLR1 axes in AD mice, and astrocyte-expressed CMKLR1 may contribute to the Aβ pathology.

CMKLR1 participates in immune responses and has the properties of chemokine receptors [18]. We sought to determine whether CMKLR1 contributes to the recruitment of astrocytes toward Aβ plaques in AD mice. We generated APP/PS1 mice with CMKLR1 deficiency (APP/PS1-Cmklr1^−/− mice) by crossing APP/PS1 mice with Cmklr1^−/− mice [26], both on a C57BL/6J background. Double immunofluorescence staining for GFAP and Aβ plaques was performed. Astrocytes (red fluorescence) within 25 µm of an Aβ plaque (green fluorescence) were considered co-localized [41]. A significant reduction in the number of reactive astrocytes co-localized with Aβ deposits was observed in the cortices and hippocampi of APP/PS1-Cmklr1^−/− mice, compared to APP/PS1 mice (Figure 3A,B), suggesting that CMKLR1 deficiency suppressed astrocyte migration and aggregation towards Aβ deposits. These findings indicate that astrocyte-expressed CMKLR1 may participate in AD pathology by regulating the recruitment and migration of astrocytes to Aβ.

First, the Boyden chamber assay was conducted to explore the effect of chemerin/CMKLR1 on the migration of astrocytes. Primary cultures of mouse astrocytes and human glioma U251 cells were seeded to the upper chamber wells, and chemerin (0.1–20 nM) was added to the lower wells as the chemoattractant. The cells that migrated through the filter were counted after 12 h. Chemerin significantly induced the migration of primary astrocytes and U251 cells in a dose-dependent manner (Figure 4A,B, and Supplementary Figure S3A,B). The treatment of C9 (an agonist of CMKLR1 derived from chemerin) and FBS (10%, as a positive control) also promoted the migration of primary astrocytes and U251 cells (Supplementary Figure S4A–C). Additionally, the MTT assay was performed to detect the effects of chemerin and C9 on the viability of primary astrocytes and U251 cells. The results showed that treatments of chemerin (20 nM) or C9 (100 nM) for 16 h did not affect the viability of these cells (Supplementary Figure S5A,B).

The scratch-wound assay was also performed to verify the promotion effect of chemerin on astrocyte migration. Primary cultured astrocytes and U251 cells were treated with 5 nM chemerin, 100 nM C9, or 10% FBS. After 8 h, 12 h, and 16 h, images were taken, and wound closure was quantified. A marked promotion of wound closure was observed in primary astrocytes and U251 cells treated with chemerin, C9, or 10% FBS compared to untreated controls (Figure 4C,D, and Supplementary Figure S3C,D), consistent with the results of the Boyden chamber assay. All of these results suggest that chemerin induces the migration of astrocytes.

Furthermore, to verify that the chemerin-induced migration of astrocytes is via CMKLR1, C15, a chemerin-derived 15-residue peptide lacking chemerin agonistic activity, was used [42]. The pre-treatment of C15 inhibited astrocyte migration induced by chemerin significantly in the Boyden chamber and scratch-wound assays (Figure 4E–H), indicating that the activation of the chemerin/CMKLR1 axis contributes to the migration of astrocytes.

We next explored the effects of chemerin on the recruitment and aggregation of astrocytes towards Aβ in vitro. U251 cells were incubated with Aβ₄₂ (3 μM) for 24 h, with and without a 15 min pre-treatment of chemerin (5 nM) or C9 (100 nM). As shown in Figure 5A,B, Aβ₄₂ induced the aggregation of U251 cells, which was significantly inhibited by the treatment of chemerin or C9. These results suggest that chemerin/CMKLR1 could suppress the Aβ-induced accumulation of astrocytes.

Previous studies have shown that STING is involved in cell migration [38,43]. Here, we determined whether STING participates in the migration of astrocytes induced by the chemerin/CMKLR1 axis. The Western blot assay was first performed to test the effect of chemerin on the phosphorylation of STING. Primary cultured astrocytes were incubated with 5 nM chemerin for 5, 15, and 30 min. The results showed that 15 min of chemerin incubation significantly increased the phosphorylation of STING (Figure 6A–C), indicating that chemerin induces STING activation. Next, we examined whether chemerin induces the activation of STING through CMKLR1. Primary cultured astrocytes were incubated with 5 nM chemerin for 15 min with or without pre-treatment of C15 (1 μM). As shown in Figure 6D–F, C15 pre-treatment inhibited the chemerin-induced increase in STING phosphorylation. These findings suggest that STING is downstream of CMKLR1, and chemerin can activate the CMKLR1/STING signaling pathway in astrocytes.

We then explored whether the migration of astrocytes induced by chemerin/CMKLR1 is dependent on STING activation. The Boyden chamber and scratch-wound assays were conducted using primary astrocytes obtained from STING knockout (Sting^−/−) and WT mice. The results showed that treatment with chemerin (5 nM) enhanced the migration of astrocytes from WT mice, while no effect was observed on astrocytes from the Sting^−/− mice (Figure 6G–J), suggesting that STING deficiency eliminated the promotion of chemerin-induced migration of astrocytes. These findings imply that STING plays a crucial role in regulating chemerin/CMKLR1-mediated astrocyte migration.

Since we have confirmed that the chemerin/CMKLR1 axis is involved in the Aβ-induced migration and aggregation of astrocytes, we further attempted to detect whether STING is required for this effect. Primary cultured astrocytes from Sting^−/− and WT mice were incubated with Aβ₄₂ (3 μM) for 24 h, with and without a 15 min pre-treatment of chemerin (5 nM). As shown in Figure 7A,B, Aβ induced the aggregation of primary astrocytes from the WT mouse brain, while chemerin treatment suppressed the accumulation of astrocytes induced by Aβ. In comparison, STING deficiency abrogated the suppressive effect of chemerin on astrocyte aggregation induced by Aβ. These findings suggest that the chemerin/CMKLR1/STING pathway contributes to Aβ-induced astrocyte aggregation.

Dulbecco’s modified Eagle’s medium (DMEM), trypsin-ethylenediaminetetraacetic acid (trypsin-EDTA), and fetal bovine serum (FBS) were obtained from Gibco (Invitrogen, Carlsbad, CA, USA). Chemerin (recombinant human chemerin) was purchased from R&D systems (Minneapolis, MN, USA). The BCA protein assay kit and 4,6-diamidino-2-phenylindole (DAPI) were obtained from the Beyotime Institute of Biotechnology (Nantong, China). Chemerin9 (C9), chemerin15 (C15) peptides, and Aβ₄₂ (≥95% purity) were synthesized at Shanghai Science Peptide Biologic Technology Co., Ltd. (Shanghai, China). Mouse monoclonal anti-GFAP-Cy3™ (catalog No. C9205) and rabbit polyclonal anti-MAP2 (catalog No. M3696) were obtained from Sigma-Aldrich, Inc. (St. Louis, MO, USA). Rabbit polyclonal anti-CMKLR1/AF488 (catalog No. BS-20043R) and anti-chemerin/AF647 antibodies (catalog No. BS-10410R) were purchased from Beijing Biosynthesis Biotechnology Co., Ltd. (Beijing, China). Rabbit polyclonal anti-Aβ (catalog No. 14975), anti-β-tubulin (catalog No. 2128), rabbit monoclonal anti-STING (catalog No. 13647), and anti-phospho-STING (catalog No. 72971) antibodies were from Cell Signaling Technology (Danvers, MA, USA). Rabbit and mouse polyclonal HRP-labeled IgG (catalog No. AW-WB1067 and AW-WB1064, respectively) were purchased from ALLWIN Biotechnology Co., Ltd. (Shanghai, China). AlexaFluor-488-conjugated anti-rabbit IgG secondary antibody (catalog No. A21206) was purchased from Invitrogen. AlexaFluor-647-conjugated anti-rabbit IgG secondary antibody (catalog No. Ab150079↗) was from Abcam (Cambridge, UK). Other reagents were purchased from Sigma-Aldrich.

The CMKLR1 knockout (Cmklr1^−/−) mice in C57BL/6J background were generated by BRL Medicine Inc. (Shanghai, China). The STING knockout (Sting^−/−, obtained from GemPharmatech, Nanjing, China) mice in C57BL/6J background were kindly provided by Dr. Ao Zhang and Dr. Chunyong Ding (Shanghai Jiao Tong University, Shanghai, China). The APP/PS1 double transgenic mice in C57BL/6J background (APP_SWE/PS1ΔE9^+/−, stock number 005864) were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). All mice were group-housed (4–5 mice per cage) under a 12/12 h light and dark cycle, with free access to food and water. The housing, breeding, and animal experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and approved by the Biological Research Ethics Committee of Shanghai Jiao Tong University (Protocol number: 202101323/Approval date: 14 October 2021). The breeding strategy involved crossing Cmklr1^−/− mice with APP/PS1 mice to generate APP/PS1-Cmklr1^+/− mice. Subsequently, these offspring were further bred with Cmklr1^+/− mice to establish three experimental groups: WT (APP/PS1^−/−-Cmklr1^+/+), APP/PS1 (APP/PS1^+/−-Cmklr1^+/+), and APP/PS1-Cmklr1^−/− (APP/PS1^+/--Cmklr1^−/−) [26]. Genotypes of mice were examined using PCR. At 9 months of age, male and female mice from all three groups were euthanized by decapitation, and brains were promptly removed. The right hemispheres were fixed in 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS) and cryoprotected in 30% sucrose. Coronal brain sections of 30 μm thickness were obtained using a freezing sliding microtome and stored in a glycol anti-freeze solution at −20 °C until immunofluorescence staining. The cerebral cortices and hippocampi of the left hemispheres were dissected, flash-frozen in dry ice, and stored at −80 °C for Western blot analysis.

The primary astrocyte cultures were prepared from 1-day-old WT or Sting^−/− mouse pups in C57BL/6J background, as previously described [56]. Briefly, the cerebral cortices were dissected from the brains of mice, followed by the removal of the meninges and microvessels. The tissues were minced using sterile ophthalmic scissors and then digested with 0.05% trypsin at 37 °C for 10 min. The cell suspension was filtered through a 40 μm sieve, and the cells were subsequently seeded onto poly-D-lysine-coated 75 cm² flasks with DMEM supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin sulfate. The culture medium was refreshed on days 1 and 3. After 7 days, the microglia in the culture flasks were detached by shaking at 240 rpm for 3 h, while the remaining astrocytes were maintained in DMEM supplemented with 10% FBS. Experimental procedures were carried out following at least one passage of the cells.

Human glioma U251 cells were generously provided by Dr. Zejian Wang (Shanghai Jiao Tong University, Shanghai, China). The U251 cells were cultured in DMEM with 2 mM glutamine, 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin sulfate.

Brain sections were rinsed with 0.05 M TBS, treated with 0.1% Triton X-100 in TBS for 10 min, and then incubated in 5% normal goat serum in TBS (0.1% Tween-20) for 30 min to block non-specific binding sites. For double immunofluorescence staining of CMKLR1 and GFAP or Aβ and GFAP, the sections were incubated for 1 h at room temperature or overnight at 4 °C with anti-CMKLR1/AF488 (1:200) or anti-Aβ (1:200) in TBS, respectively. After rinsing with TBS, the sections stained with anti-Aβ antibody were incubated with Alexa Fluor 647-conjugated anti-rabbit secondary antibody (1:500) at room temperature for 1 h. Then, the sections were incubated with anti-GFAP-Cy3™ antibody (1:200) in TBS for 1 h at room temperature. After washing in TBS, the sections were stained for nuclei with 100 ng/mL of DAPI for 10 min and mounted with 90% glycerol. The fluorescent confocal images were captured using a laser scanning confocal fluorescent microscope (TCS SP8, Leica Microsystems, Wetzlar, Germany). The expression of CMKLR1 and GFAP was quantified by analyzing immunofluorescence intensity using ImageJ/FIJI software 1.54f (NIH, Bethesda, MD, USA). The number of astrocytes (GFAP staining) colocalized with CMKLR1 or Aβ plaques (within 25 µm surrounding Aβ plaque [41]) was counted. Data are presented as the mean ± SEM based on 2 individual fields for each region, using three mice in each group.

To identify the expression of chemerin and its colocalization with microtubule-associated protein 2 (MAP2, neuron-specific molecule), brain sections were initially stained with anti-MAP2 (1:500) overnight. After three washes with TBS, the sections were incubated in Alexa Fluor 488-conjugated anti-rabbit secondary antibody (1:500) at room temperature for 1 h. The sections were rinsed in TBS and stained with anti-chemerin/AF647 (1:200) for 1 h at room temperature. Then, sections were counterstained with DAPI for 10 min and mounted with 90% glycerol. The fluorescent confocal images were taken on a laser scanning confocal microscope (A1 HD25, Nikon, Tokyo, Japan). Z-stack images were acquired using 20× water immersion objectives with a z-stack interval of 1.63 μm. The z-stack capture was performed by setting the bottom position at the appearance of the object of interest (first slice image) and the top position at the disappearance of the object of interest (last slice image). Merged maximum intensity projections (DAPI and chemerin) of the images were processed to quantify the expression of chemerin. The immunofluorescence intensity was analyzed using the ImageJ/FIJI software 1.54f. The colocalization analysis of chemerin in neurons stained with MAP2 was conducted using the Coloc2 plugin in ImageJ/FIJI software 1.54f. Pearson’s correlation coefficient was utilized to calculate the double fluorescence correlation coefficients. Scatter plots were employed to illustrate changes in the quantification of co-localized fluorescence [57]. Data are presented as the mean ± SEM based on 2 individual fields for each region, using three mice in each group.

Standard 48-well chemotaxis chambers (Neuro Probe, Gaithersburg, MD, USA) were used to detect the migration of astrocytes. The upper and lower wells of the chamber were separated by a polycarbonate membrane (10-μm pore size, Neuro Probe). The Boyden chamber assay was conducted as described previously [58]. Briefly, either DMEM or DMEM containing various concentrations of chemerin, C9, or 10% FBS were added to the lower wells of the chamber. The polycarbonate membrane was placed over the lower wells of the chamber. The upper chamber wells were filled with primary cultures of astrocytes or U251 cells with or without 15 min pretreatment of 1 μM C15. Following the addition of cells to the upper wells, the chamber was placed in a humidified incubator with 5% CO₂ for 12 h. Subsequently, the filter was delicately extracted, and any non-migrating astrocytes on the upper surface were carefully removed using a cotton-tipped swab. The astrocytes that successfully migrated to the lower surface of the filter were fixed with methanol and stained with 0.1% crystal violet. Cell counting was performed using a phase-contrast inverted microscope (IX51, Olympus Optical Co., Ltd., Tokyo, Japan) equipped with a digital camera (EOS 1100D, Canon Inc., Tokyo, Japan). The results were quantified as the chemotaxis index, indicating the fold change in migrated cell numbers in response to the chemoattractant, compared to the control medium. Data are shown as the mean ± SEM from three independent experiments, each with three wells for each group.

The scratch-wound assay for evaluating cell mobility was performed [59]. Specifically, a scratch was created in confluent primary astrocytes or U251 cells cultured on coverslips precoated with poly-L-ornithine using a 10-μL pipette tip, resulting in the formation of a cell-free region measuring 500 μm in width. Subsequent movement of primary astrocytes and U251 cells was captured through light microscopy (Olympus Optical Co., Ltd.). Pictures were captured at 0 h as a control. The cells were incubated with chemerin (5 nM) with or without 15-min pretreatments with C15 for 8 h, 12 h, or 16 h. Then the pictures of the cells were taken at each time point, and the results were quantified by calculating the mean migrated distance of leading cells in the scratched area. Data are presented as the mean ± SEM from three independent experiments, each with three samples per group.

Astrocyte aggregation was measured as described previously [60]. Briefly, the primary astrocyte cultures or U251 cells (1 × 10⁴) were seeded in a 24-well plate and incubated with 3 µM fibrillar Aβ₄₂ with or without 5 nM chemerin or 100 nM C9 for 24 h. The cluster of astrocytes was defined as at least ten cells aggregated together around Aβ₄₂. The distribution of aggregated astrocytes was captured with a phase contrast inverted microscope with a digital camera. The data are presented as the means ± SEM from three independent experiments, each with three wells for each group.

The tissues of the mouse brain were homogenized in lysis buffer containing 50 mM Tris-HCL (pH 7.4), 2 mM EDTA, 100 mM NaF, 2 mM sodium vanadate, 10 mM β-mercaptoethanol, 8.5% sucrose, 100 µg/mL leupeptin, 5 µg/mL aprotinin, and 5 µg/mL pepstatin. Protein concentrations were validated using BCA kits. The samples were mixed with 5× sodium dodecyl sulfate (SDS)-PAGE loading buffer, heated at 99 °C for 10 min, and separated by 10% or 12% SDS-PAGE. Subsequently, the proteins were transferred to nitrocellulose membranes (GE Healthcare, Wauwatosa, WI, USA). The membranes were blocked for 1 h at room temperature with 5% non-fat milk and then incubated at 4 °C overnight with anti-STING (1:1000), anti-phospho-STING (1:1000), or anti-β-tubulin (1:1000) antibodies, followed with corresponding HRP-labeled IgG secondary antibodies. The membranes were visualized using a Super ECL Plus (US Everbright, Suzhou, China) under the ChemiDoc MP Imaging System (Bio-Rad, Richmond, CA, USA). Densitometric quantification of protein bands was determined using the ImageJ software 1.54f (NIH, Bethesda, MD, USA).

The data are shown as means ± SEM. Two-group comparisons were assessed using the two-tailed t-test. For multiple comparisons, one-way ANOVA with Tukey’s post hoc test was employed. Statistical analyses were conducted using GraphPad Prism 8 (San Diego, CA, USA). A p-value smaller than 0.05 was considered statistically significant.

All datasets generated or analyzed during this study are available from the corresponding author on reasonable request.