Cathelicidin antimicrobial peptide expression in neutrophils and neurons antagonistically modulates neuroinflammation

Abstract

Multiple sclerosis (MS) is an autoimmune disease that affects the CNS, the pathophysiology of which remains unclear and for which there is no definitive cure. Antimicrobial peptides (AMPs) are immunomodulatory molecules expressed in various tissues, including the CNS. Here, we investigated whether the cathelicidin-related AMP (CRAMP) modulated the development of experimental autoimmune encephalomyelitis (EAE), a mouse model of MS. We showed that, at an early stage, CNS-recruited neutrophils produced neutrophil extracellular traps (NETs) rich in CRAMP that were required for EAE initiation. NET-associated CRAMP stimulated IL-6 production by dendritic cells via the cGAS/STING pathway, thereby promoting encephalitogenic Th17 response. However, at a later disease stage, neurons also expressed CRAMP that reduced EAE severity. Camp knockdown in neurons led to disease exacerbation, while local injection of CRAMP1-39 at the peak of EAE promoted disease remission. In vitro, CRAMP1-39 regulated the activation of microglia and astrocytes through the formyl peptide receptor (FPR) 2. Finally, administration of butyrate, a gut microbiota-derived metabolite, stimulated the expression of neural CRAMP via the free fatty acids receptors 2/3 (FFAR2/3), and prevented EAE. This study shows that CRAMP produced by different cell types has opposing effects on neuroinflammation, offering therapeutic opportunities for MS and other neuroinflammatory disorders.

Keywords: Autoimmunity; Immunology; Innate immunity; Multiple sclerosis; Neutrophils.

Figure 1. Opposite role of CRAMP during EAE.

(A) mRNA expression of CRAMP was analyzed by RT-qPCR in SC from C57BL/6 WT mice at different days after EAE induction. Data are the median plus or minus interquartile range of 5 to 11 independent mice per group from 4 independent experiments. (B) Camp^–/– C57BL/6 mice and WT littermate controls (n = 30 mice per group from 6 independent experiments) were immunized with MOG_35-55 to induce EAE. Clinical scores are shown (data are represented as mean ± SEM). (C and D) Cells from draining lymph nodes (C) or SC (D) were recovered on day 7 or 12, respectively, after EAE induction in Camp^–/– and WT C57BL/6 mice and stained with I-A^b MOG_35-55 tetramer. Data are the frequency and number of tetramer⁺ cells among CD45⁺βTCR⁺CD4⁺ cells. Median plus or minus interquartile range of 6 independent mice per group from 3 independent experiments is shown. (E) Frequency of IFN-γ⁺ and IL-17⁺ cells in CD4⁺ T cells recovered farom the SC at day 12 after EAE induction in Camp^–/– and WT C57BL/6 mice and restimulated for 6 hours with PMA/ionomycin. Median plus or minus interquartile range of 6 independent mice per group is shown from 3 independent experiments. (F−H) WT mice (n = 15 mice per group from 3 independent experiments) were immunized with MOG_35-55 to induce EAE. Mice were treated with CRAMP or scCRAMP i.p. at day 7 (F) or at day 15 (G) or i.t. at day 9 (H) after EAE induction. Clinical scores are shown (data are represented as mean ± SEM). (I) NF-L protein levels were measured in the serum 15 days after EAE induction in WT mice treated as in G and in unmanipulated (naive) WT mice. Data are the median plus or minus interquartile range of 7 independent mice per group from 3 independent experiments. (J and K) Expression of CD44 on astrocytes (J) and CD86 on microglia (K) was determined by flow cytometry at day 15 after EAE induction in the SC of WT mice treated as in G and in unmanipulated (naive) WT mice. Data are the median plus or minus interquartile range of 6 independent mice per group from 3 independent experiments. Comparison between each group was performed using the nonparametric Mann-Whitney U test or Kruskal-Wallis test followed up by Dunn’s post test when more than 2 groups were compared. For evaluating differences between EAE scores over time, a 2-way ANOVA with Tukey’s multiple-comparisons test was used.

Figure 2. CRAMP is expressed in different cell types within the CNS during EAE.

(A and C) Confocal microscopy images of SC section from WT mice immunized with MOG_35-55 12 days earlier. Sections were stained for CRAMP (red), MAP2 (green), GFAP (blue), and DNA (gray) in A or for CRAMP (red), Ly6G (blue), and DNA (gray) in C. Arrows indicate NETs. Data are representative of 3 independent experiments. Original magnification, ×40. Scale bars: 15 μm. (B) Confocal microscopy images of motor neuron-like NSC-34 cells stained for CRAMP (red), MAP2 (green), synaptophysin (blue), and DNA (gray). Data are representative of 3 independent experiments. Original magnification, ×63. (D and E) Flow cytometry analysis of SC after EAE induction in WT mice. In D, 12 days after EAE induction, frequency of neutrophils (CD45⁺CD11b⁺Ly6G⁺Ly6C^lo) is shown in the left panel and the expression of CRAMP and CD62L by neutrophils is shown in the right panel. BM from tibia rich in immature neutrophils is shown as control. In E, the MFI of surface CRAMP expression on neutrophils, neurons (CD45⁻NeuN⁺), astrocytes (CD45⁻GFAP⁺), and microglia (CD45⁺CD11b⁺Tmem119⁺) is shown at different days after EAE induction. Data are the median plus or minus interquartile range of 8 independent mice per group from 3 independent experiments. Comparison between each group was performed using the nonparametric Mann-Whitney U test or Kruskal-Wallis test followed up by Dunn’s post test when more than 2 groups were compared.

Figure 3. CRAMP from neutrophils is essential for EAE.

(A) Mrp8-Cre^neg.Camp^fl/fl and Mrp8-Cre^Tg.Camp^fl/fl mice (n = 30 mice per group; 6 independent experiments) were immunized with MOG_35-55 to induce EAE. Clinical scores are shown (data are represented as mean ± SEM). (B) Flow cytometry analysis of SC 15 days after EAE induction. The absolute number of immune cells (CD45⁺), B cells (CD45⁺CD19⁺), T cells (CD45⁺βTCR⁺), neutrophils (CD45⁺CD11b⁺Ly6G⁺Ly6C^lo), macrophages (CD45⁺CD11b⁺Ly6G^–F4/80⁺), and dendritic cells (CD45⁺Ly6G^–F4/80^–CD11c⁺) is shown. Data are the median plus or minus interquartile range of 6 independent mice per group from 3 independent experiments. (C) mRNA expression of cytokines was analyzed by RT-qPCR in SC from Mrp8-Cre^neg.Camp^fl/fl and Mrp8-Cre^Tg.Camp^fl/fl mice 15 days after EAE induction. Data are the median plus or minus interquartile range of 6 independent mice per group from 3 independent experiments. (D−F) Splenocytes were recovered 10 days after EAE induction in Mrp8-Cre^neg.Camp^fl/fl and Mrp8-Cre^Tg.Camp^fl/fl mice (D), C57BL/6 WT mice (E), or C57BL/6 CD45.1 mice (F). Cells were cultured for 3 days with MOG_35-55 in pro-Th17 conditions and transferred in C57BL/6 WT mice (D) or in Mrp8-Cre^neg.Camp^fl/fl or Mrp8-Cre^Tg.Camp^fl/fl mice (E and F). Clinical scores are shown (data are represented as mean ± SEM) (n = 15 mice per group from 3 independent experiments) in D and E. Number of CD45.1⁺CD4⁺ T cells in the SC and the brain 7 days after transfer is shown in F. Comparison between each group was performed using the nonparametric Mann-Whitney U test or Kruskal-Wallis test followed by Dunn’s post test when more than 2 groups were compared. For evaluating differences between EAE scores over time, a 2-way ANOVA with Tukey’s multiple comparisons test was used.

Figure 4. CRAMP from NETs favors encephalitogenic T cell response.

(A and C) WT mice (n = 10 mice/group from 2 independent experiments) were immunized with MOG_35-55 to induce EAE. Mice were then treated s.c. with Cl-amidine daily from day 7 to 17 after EAE induction (A) or treated i.p. with neutrophil-depleting αLy6G mAb or isotype control mAb every 3 days from day 7 to 17 after EAE induction (C). Clinical scores are shown (mean ± SEM). (B) NF-L protein levels were measured in the serum of Cl-amidine-treated mice 15 days after EAE induction. Data are the median plus or minus interquartile range of 7 independent mice per group from 3 independent experiments. (D) Neutrophils were isolated from bone marrow of tibia of WT and Camp^–/– mice and activated with A23187 ionophore to induce NETs. Visualization of NETs was performed by confocal microscopy after DNA (SYTOX green) and CRAMP (red) staining. Data are representative of 3 independent experiments. Original magnification, ×40. (E and F) NETs were prepared from WT and Camp^–/– mice as in E and added to splenocyte culture isolated from WT mice immunized with MOG_35-55 10 days earlier. (G) BMDCs from WT mice were stimulated for 18 hours with NETs. In some conditions, the cGAS inhibitor TDI-6570 or a multi-specific TLR9, AIM2, and cGAS antagonist ODN A151 was added to the culture 3 hours before stimulation. Cytokine levels in the supernatant were measured by multiplex ELISA. Data are the median plus or minus interquartile range of 6–8 independent wells from 3 independent experiments. Comparison between each group was performed using the nonparametric Mann-Whitney U test or Kruskal-Wallis test followed by Dunn’s post test when more than 2 groups were compared. For evaluating differences between EAE scores over time, a 2-way ANOVA with Tukey’s multiple comparisons test was used.

Figure 5. CRAMP from neural cells dampens EAE.

(A, C, and D) WT mice were immunized with MOG_35-55 to induce EAE. C57BL/6 mice were treated with i.t. injection of AAV9-shCamp or AAV9-shScramble at day –7 (A) or with per os (p.o.) butyrate from day 7 to day 30 (C) after EAE induction or with both (D). Clinical scores are shown (data are represented as mean ± SEM). (n = 12 mice per group from 3 independent experiments). (B) WT mice were treated with drinking water with sodium butyrate (10 g/L) for 7 days. SC was recovered and Camp expression determined by RT-qPCR. Data are the median plus or minus interquartile range of 6 independent mice per group from 3 independent experiments. (E) Neuron NSC-34 cells were cultured for 24 hours in the presence of either butyrate, acetate, or propionate analysis of Camp expression by RT-qPCR. In some conditions, the FFAR2/3 antagonist GLPG 0974 was added to the cells 2 hours before the SCFAs. Data are the median plus or minus interquartile range of 3 independent experiments. (F) Splenocytes were recovered 10 days after EAE induction in WT mice and cultured for 3 days with MOG_35-55 in pro-Th17 conditions with growing doses of CRAMP_1-39 (μg/mL). Cytokine levels were measured in the supernatant by multiplex ELISA. Data are the median plus or minus interquartile range of 6 independent mice per group from 3 independent experiments. (G) Splenocytes were recovered 10 days after EAE induction in WT mice and were cultured for 3 days with MOG_35-55 in pro-Th17 conditions with either CRAMP_1-39 or scCRAMP_1-39 (10 μg/mL) and transferred in WT mice. Clinical scores are shown (data are represented as mean ± SEM) (n = 12 mice/group; 3 independent experiments). Comparisons between groups was performed using the nonparametric Mann-Whitney U test or Kruskal-Wallis test followed by Dunn’s post test when more than 2 groups were compared. For evaluating differences between EAE scores over time, a 2-way ANOVA with Tukey’s multiple-comparisons test was used.

Figure 6. CRAMP dampens microglia and astrocyte activation.

(A) Microglia culture was prepared from Camp^–/– mice and 10 μg/mL of CRAMP or vehicle was added for 24 hours before staining for CD11b (green), CRAMP (red), and DNA (blue) and analysis by confocal microscopy. Data are representative of 3 independent experiments. (B and C) Microglia culture was prepared from WT mice and activated or not with LPS for 24 hours before wash and addition of 10 μg/mL CRAMP_1-39 or vehicle for an additional 24 hours. In C, FPR2 antagonist WRW4 or P2X7R antagonist JNJ-47695567 was added 2 hours before CRAMP_1-39 addition. Cytokine levels were measured in the supernatant by multiplex ELISA. Data are the median plus or minus interquartile range of 6–8 independent mice per group from 3 independent experiments. (D) Astrocyte culture was prepared from Camp^–/– C57BL/6 mice and 10 μg/mL CRAMP or vehicle was added for 24 hours before staining for GFAP (green), CRAMP (red),and DNA (blue) and analysis by confocal microscopy. Data are representative of 3 independent experiments. Original magnification, ×40. (E) Astrocyte cultures were prepared from WT mice and activated or not with LPS for 24 hours before wash and addition of 10 μg/mL CRAMP_1-39 or vehicle for an additional 24 hours. The FPR2 antagonist WRW4 or P2X7R antagonist JNJ-47695567 was added 2 hours before CRAMP_1-39 addition. Cytokine levels were measured in the supernatant by multiplex ELISA. Data are the median plus or minus interquartile range of 6 independent mice per group from 3 independent experiments. Comparison between each group was performed using the nonparametric Mann-Whitney U test or Kruskal-Wallis test followed by Dunn’s post test when more than 2 groups were compared.