Podoplanin expands and interacts with myeloid cells at the cribriform plate during EAE.

(A) : Cartoon schematic of the cribriform plate lymphatic vessels (cpLVs) which associate the perineural environment of olfactory nerves (ON) between the olfactory bulbs (OB) in mice. Lymphatic vessels are directly next to ONs as they project through the cribriform plate (CP) into the nasal epithelium (NE). (B) : Healthy and EAE 3.0 CSFR1R-GFP mouse CP sections were stained for podoplanin (PDPN) PE and Lyve-1 Alexa 660 to visualize the relationship of myeloid cells (CSF1R-GFP+), lymphatics (Lyve-1+, PDPN+), and meningeal stroma (Lyve-1negative, PDPN+). Zoomed panels show the perineural immune-stromal-lymphatic environment around ONs during healthy and EAE. (C) : Quantification of MHCII MFI at the CP during healthy and EAE 3.0 conditions using identical imaging conditions. MHCII (eFlour 450) increases in PDPN+ regions during EAE. Unpaired student’s t-test. **P < 0.01. Data are represented as mean ± SEM. (D) : Representative image of perineural CSF1R-GFP cells adhered to PDPN+ region of ONs. (E) : Quantification of average PDPN+ thickness (µm) around ONs during healthy and EAE conditions. Intensity plot profiles (yellow dotted line) were used to estimate PDPN+ regions. n=3-5 mice. *P<0.05. Data are represented as mean ± SEM (F) : Representative image of immune response at cribriform plate in EAE 3.0 CD11c-eYFP mice stained for Lyve-1 and PDPN. (G) : Non-Lymphatic (PDPNnegative, Lyve1negative) associated CD11c+ dendritic cells are in the olfactory bulb brain tissue. The cells likely represent infiltrating DCs, macrophages, and activated Microglia. (H) : Meningeal-Lymphatic (PDPN+) associated CD11c+ are at the cribriform plate lymphatics. The cells likely represent infiltrating DCs, macrophages, and monocytes (I-J): Representative image of immune response at cribriform plate in EAE 3.0 CD11c-eYFP mice stained for Lyve-1 and PDPN. CD11c-eYFP+ embed into dense PDPN+ ECM) near cribriform lymphatics (Lyve-1).

Elevation of CHI3L3 and ARG1 at cribriform lymphatics during EAE

(A-C): Magnification between an EAE 3.0 mouse olfactory bulb (OB) at cribriform plate (CP) showing infiltration of CHI3L3⁺ cells (green) surrounding olfactory nerves (ON) and localizing within PDPN⁺ stroma networks (red) and LYVE-1⁺ lymphatic vessels (white). DAPI marks nuclei (blue). (B-C) High-magnification views of boxed regions showing accumulation of CHI3L3⁺ cells in proximity to ONs and embedded within PDPN⁺ stromal-lymphatic regions and LYVE-1⁺ lymphatic zones. (D-E): (D) Representative confocal images of healthy and EAE 3.0 tissues showing increase of CHI3L3⁺ cells (green) in PDPN⁺ regions during inflammation. (E) Quantification of CHI3L3⁺ cells within PDPN-rich areas, demonstrating significantly elevated numbers in EAE versus healthy controls. Data are represented as mean ± SEM. n = 3 per group (F-G): (F) Representative images of ARG1⁺ (green) immunoregulatory macrophages in healthy and EAE 3.0 mice, showing spatial colocalization with PDPN⁺ (red) and LYVE-1⁺ structures (white) near olfactory nerves. Insets highlight increased density and clustering of ARG1⁺ cells in EAE. (G) Quantification of ARG1⁺ macrophages in PDPN⁺ regions, showing a significant increase in EAE mice compared to healthy controls (p = 0.0116). Data are represented as mean ± SEM. n=6-9 (H) High-resolution confocal image of EAE 3.0 olfactory nerve region showing ARG1⁺ macrophages interacting closely with PDPN⁺ and LYVE-1⁺ lymphatics around olfactory nerve bundles (ON). Individual channels shown for ARG1, PDPN, and LYVE-1. Zoomed panel shows perineural ARG1.

Characterization of post-contact CD11c+CD11b+ cells which previously interacted with cribriform plate meningeal-lymphatic niche

(A-B): Cartoon scheme outlining “PostContact-seq” sorting procedure of cribriform plate + Olfactory bulb tissue preparations from EAE 3.0. Sorted singlets from CD11c+CD11b+ (Myeloid cells) and CD31+PDPN+ (Meningeal-Lymphatic niche) gates were collected into a no-contact tube. Simultaneously, quadruple positive CD11c+CD11b+CD31+PDPN+ (Myeloid cell+Meningeal Lymphatic Niche) aggregating cells were sorted into the “interactor” tube. Prior to scRNAseq, both tubes underwent short liberase treatment to dissociate interacting cells and generate single cell suspensions of post-contact, and sequenced. (C): Combined UMAP plots show the 12 cluster identities from post-contact and no-contact scRNAseq (Left). Overlaid UMAP plots show the 12 cluster identities from post-contact and non-contact scRNAseq (Right). (D): Vertical slice graph shows proportion of each cell cluster between no contact and post-contact tubes. (E): Dot plot displays the expression of selected marker genes across cell clusters identified by scRNA sequencing.

Analysis of post-contact cells reveals immune cell and extracellular matrix crosstalk

(A) Heatmap displaying scaled expression (z-score) of highly expressed representative genes involved in immune regulation, immune activation (e.g., Pdcd1) across four cell types: macrophages (MΦ), T cells, migratory dendritic cells (Mig DCs), and fibroblasts (e.g., Pdpn). Color scale indicates expression levels. (B) Quantification and network map of predicted ligand-receptor interactions between fibroblasts and immune populations (macrophages, Mig DCs, and T cells) using integrated analysis from CellChat, CellPhoneDB, iCellNet, and the Ramilowsky dataset. The bar graph shows the number of unique ligand-receptor interactions per pairwise combination, with the most extensive communication observed between macrophages and fibroblasts. Analyzed with crossTALK IntEraction Network (TALKIEN) (Moratalla-Navarro et al., 2023) with microglia, neutrophils, and proliferating gene sets excluded. (C) Network plot illustrating specific ligand-receptor pairs involved in fibroblast-immune cell communication. Arrows point from ligand-expressing to receptor-expressing cell types. Notable interactions include Spp1-CD44, Pdpn-Clec2d, and Thbs1-CD47. (D) Immunofluorescence imaging of the cribriform plate region in EAE day 3.0 mice showing spatial organization of LYVE-1+ lymphatics (white), PDPN+ lymphatics and ECM (red), and CSF1R-GFP+ myeloid cells (green). DAPI (blue) marks nuclei. Scale bars: 30 µm. (E) Pathway enrichment analysis of ligand-receptor pairs among macrophages, Mig DCs, T cells, and fibroblasts. Enriched pathways include extracellular matrix organization, integrin-mediated interactions, interleukin 10 signaling. Gene ratio represents the proportion of identified genes contributing to each pathway; dot size reflects the number of genes, and color indicates adjusted p-value. (F) : UMAP projection shows major post contact cell types, including macrophages (purple), microglia (green), neutrophils (blue), T cells (orange), proliferating cells (light blue), migratory dendritic cells (yellow-orange), and fibroblasts (aqua). Feature plots highlight expression of Chil3 (CHI3L3), Arg1, Spp1, Ccr2, Cxcr4, Csf1r (G) : Volcano plot comparing post-contact vs no-contact cluster 2 macrophages (Highest Arg1 expressing cluster). Analysis reveals post contact macrophage have downregulated MHC-II processing and presentation genes (H2-DMa, H2-Aa, H2-Ab1, H2-DMb1, H2-Eb1, Cd74, Ciita) and elevated expression of Chil3 (CHI3L3), another alternatively activated macrophage marker and ECM binding protein. (H) : Gene ontology enrichment analysis (bottom) demonstrates post contact Macrophages are enriched for pathways in Fcγ receptor and complement receptor signaling, negative regulation of IL-12 production, homotypic cell-cell adhesion, and hemostasis.

Post-Contact DCs have pathways related to reduced pro-inflammatory signature and Pdcd1 (PD-1) expression

(A) High-resolution confocal image from EAE 3.0 CP showing MHCII⁺ (green) cells in close proximity to PDPN+ stroma (red). (B) Gene ontology enrichment analysis of cluster 6 migratory DCs reveals pathways related to actin dynamics, interferon signaling, leukocyte activation, and programmed cell death. Red text indicates immune pathways of interest. (C) Comparison of unique enrichment pathways from top 300 genes compared between no-contact and post-contact DCs. (D) UMAP visualization of scRNA-seq data highlighting Pdcd1 (PD-1) expression in cluster 6 CCR7⁺ migratory DCs and cluster 8 T cells (E) Violin plot showing Pdcd1 expression in post contact DCs versus no-contact, suggesting PME influence on DC checkpoint expression. (F-G) Quantification of PD-1High single and doublet DCs at the olfactory-cribriform junction, showing significant increase in EAE mice compared to controls. n = 3, unpaired Student’s t-test. Data are represented as mean ± SEM. Singlet DCs (p=0.0322), Doublet DCs (p=0.0084). (H) Heatmaps of highest genes expressed genes in PD-1⁺ vs PD-1⁻ DCs within CD31+PDPN+ post-contact and no-contact subsets. Pro-inflammatory genes (Cxcl10, Il1b, Ifit1, Ifit2, Ifit3.) are high in PD-1 negative DCs.

Blood-derived CD11c+CD11b+ cells migrate to cribriform plate lymphatics in a time dependent manner and undergo cell death

(A) Schematic overview of experimental design. Mice were intravenously injected with Alexa Fluor 647-labeled CD45 antibody (CD45-IV) at multiple timepoints following EAE induction to label blood-accessible leukocytes. Diagram (left) shows DC trafficking to the cribriform plate niche. Representative flow plots (right) show the lack of CD45-IV⁺ doublets at 3 minutes and emergence of IV+ cells at 24 hours post-injection. (B) Quantification of the percentage of CD45-IV⁺ CD11c+CD11b+CD31+PDPN+ doublets over time, showing progressive accumulation of stained doublets, peaking at 48 hours post-injection. 24 hr (p=0.0227) and 48 hr (p=0.0001) time points are significantly elevated when compared 6 hr. One way anova with Tukey’s multiple comparison. Data combined from 3 independent experiments. Data are represented as mean ± SEM n=2-6 per group (C) Viability of CD45-IV⁺ doublets assessed by live/dead staining at different timepoints, showing trend of cell death (Ghost+) over time. One-way anova (P = ns). data are represented as mean ± SEM n=2-6 per group (D) Representative confocal images from EAE 3.0 mice showing cleaved caspase-3⁺ (red) CD11c-EYFP⁺ DCs (green) interacting with LYVE-1⁺ lymphatics (white). Yellow arrowheads highlight apoptotic DCs. (E) Quantification of Ghost+ cells among PD-1⁻ and PD-1⁺ DCs identified as interactors, showing significantly higher cell death among PD-1⁺ interactor doublets. n = 3, P = 0.0008, unpaired Student’s t-test. Data are represented as mean ± SEM.

Strategy to isolate CD11b+CD11c+PDPN+CD11b+ “No contact” and “Post-Contact” populations

(A) : Cartoon scheme outlining “PostContact-seq” sorting procedure of cribriform plate + Olfactory bulb tissue preparations from EAE 3.0. Sorted singlets from CD11c+CD11b+ (Myeloid cells) and CD31+PDPN+ (Meningeal-Lymphatic niche) gates were collected into a no-contact tube. Simultaneously, quadruple positive CD11c+CD11b+CD31+PDPN+ (Myeloid cell+Meningeal Lymphatic Niche) aggregating cells were sorted into the “interactor” tube. Prior to scRNAseq, both tubes underwent short liberase treatment to dissociate interacting cells and generate single cell suspensions of post-contact, and sequenced. (B) : Cutoffs for filtered cells for differential analysis using the following criteria: Cells have > 200 genes expressed, Cells have < 3,500 genes expressed, Cells have < 5% mitochondrial genes expressed. Strategy limits doublet aggregates at time of sequencing (High nFeatureRNA) and dead cells (High Percentage Mt).

Analysis of PD-1+ immune subsets in isolated cribriform plate cell suspensions

(A) Violin plot showing Pdcd1 (PD-1) expression in T cells (Cluster 8) stratified by inferred cell-cell interactions, with higher expression observed in post contact T cells. (B) Volcano plots of differential gene expression between post-contact and no-contact cells in Cluster 6 migratory dendritic cells (left) and Cluster 8 T cells (right) (C) Flow cytometry gating strategy for identifying singlet and doublet populations of CD45+ leukocytes, dendritic cells (CD11c+), and CD4+ T cells, followed by assessment of PD-1 expression across subsets. (D) Representative flow plots and quantification of PD-1high singlet and doublet dendritic cells and T cells in healthy control and EAE day 3.0 mice. EAE mice show significantly elevated PD-1high populations across all subsets, with the largest increase observed in doublet T cells, suggesting enhanced immune interaction and checkpoint activation during neuroinflammation. Singlet DCs (p=0.0322), Doublet DCs (p=0.0084), Singlet T cells (p=0.0005), Doublet T cells (p=0.0007). Unpaired two-tailed t test. Data are represented as mean ± SEM

Cell cluster characterization of “post contact” CD11b+CD11c+ PDPN+CD31+ interacting cells

UMAP projection (top left) shows major post contact cell types isolated, including macrophages (purple), microglia (green), neutrophils (blue), T cells (orange), proliferating cells (light blue), migratory dendritic cells (yellow-orange), and fibroblasts (aqua). Feature plots highlight expression of Arg1, Ly6c2, Ccr2, Chil3 (CHI3L3), and Pdpn, identifying infiltrating monocyte-derived, alternatively activated macrophages, distinct from T cells (Cd3e, Ifng), DCs (CCR7, Cd80), microglia (Tmem119, P2ry12), neutrophils (Ly6g, Cxcr2), and proliferating cells (Mki67, Top2a). Gene ontology enrichment analysis (bottom) demonstrates that these post contact macrophages are enriched for pathways in Fcγ receptor and complement receptor signaling, negative regulation of IL-12 production, homotypic cell-cell adhesion, and hemostasis.

Accumulation of CCR2+ immune cells in PDPN+ cribriform regions during EAE

(A) Immunofluorescence images of healthy and EAE 3.0 olfactory bulbs (OBs), showing increased CCR2⁺ macrophage accumulation at the olfactory-cribriform junction. Insets show higher magnification of boxed areas with increased CCR2⁺ cell localization to PDPN-rich regions in EAE. (B) Quantification of CCR2⁺ cell numbers and their proportion in contact with PDPN-rich zones at the olfactory-cribriform interface. EAE mice show significantly more CCR2⁺ cells and higher PDPN-association. Unpaired two-tailed t test. Data are represented as mean ± SEM (p = 0.0027). (C) High-magnification image showing CCR2⁺ cells (green) embedded within cribriform plate lymphatic vessels (CPLVs, marked by LYVE-1, white) and adjacent to PDPN⁺ region (red) at the olfactory nerve (ON) interface in EAE 3.0 mice. Arrowheads indicate CCR2⁺ cells associated with CPLVs. Unpaired two-tailed t test. Data are represented as mean ± SEM (p = 0.0064)

Graphical Summary.

Perineural Microenvironment (PME) has Expanded Immune-Stromal-Lymphatic Niche During EAE. In the healthy state the PME of olfactory nerve bundles (ON) have lower levels of myeloid cells and lymphatics but still have access to draining cerebrospinal fluid (CSF), waste, antigen, and immune cells like DCs. During EAE neuroinflammation the PME becomes remodeled with higher myeloid cell accumulation within PDPN+ regions: fibroblasts, lymphatics, ECM, and macrophages. This creates an assembled immunoregulatory environment around olfactory nerve bundles where tolerogenic myeloid cells engage in cell-cell interactions alongside nerve bundles and cribriform lymphatics.