autoimmune encephalomyelitis experimental models

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 25 July 2024

Apelin modulates inflammation and leukocyte recruitment in experimental autoimmune encephalomyelitis

• Hongryeol Park   ORCID: orcid.org/0000-0001-6717-0432 1 ,
• Jian Song   ORCID: orcid.org/0000-0002-2459-7046 2 ,
• Hyun-Woo Jeong   ORCID: orcid.org/0000-0002-6976-6739 1 ,
• Max L. B. Grönloh   ORCID: orcid.org/0000-0003-0109-8225 3 ,
• Bong Ihn Koh   ORCID: orcid.org/0000-0002-3636-0492 1 ,
• Esther Bovay 1 ,
• Kee-Pyo Kim   ORCID: orcid.org/0000-0002-8666-8444 4 ,
• Luisa Klotz   ORCID: orcid.org/0000-0001-5439-9633 5 ,
• Patricia A. Thistlethwaite   ORCID: orcid.org/0000-0002-1986-7304 6 ,
• Jaap D. van Buul   ORCID: orcid.org/0000-0003-0054-7949 3 ,
• Lydia Sorokin   ORCID: orcid.org/0000-0001-7704-7921 2 &
• Ralf H. Adams   ORCID: orcid.org/0000-0003-3031-7677 1  

Nature Communications volume  15 , Article number:  6282 ( 2024 ) Cite this article

828 Accesses

13 Altmetric

Metrics details

• Cardiovascular biology
• Inflammatory diseases

Demyelination due to autoreactive T cells and inflammation in the central nervous system are principal features of multiple sclerosis (MS), a chronic and highly disabling human disease affecting brain and spinal cord. Here, we show that treatment with apelin, a secreted peptide ligand for the G protein-coupled receptor APJ/Aplnr, is protective in experimental autoimmune encephalomyelitis (EAE), an animal model of MS. Apelin reduces immune cell entry into the brain, delays the onset and reduces the severity of EAE. Apelin affects the trafficking of leukocytes through the lung by modulating the expression of cell adhesion molecules that mediate leukocyte recruitment. In addition, apelin induces the internalization and desensitization of its receptor in endothelial cells (ECs). Accordingly, protection against EAE major outcomes of apelin treatment are phenocopied by loss of APJ/Aplnr function, achieved by EC-specific gene inactivation in mice or knockdown experiments in cultured primary endothelial cells. Our findings highlight the importance of the lung-brain axis in neuroinflammation and indicate that apelin targets the transendothelial migration of immune cells into the lung during acute inflammation.

Similar content being viewed by others

Blocking PDGF-CC signaling ameliorates multiple sclerosis-like neuroinflammation by inhibiting disruption of the blood–brain barrier

Microglia and meningeal macrophages depletion delays the onset of experimental autoimmune encephalomyelitis

Guanabenz modulates microglia and macrophages during demyelination

Introduction.

Multiple sclerosis (MS) is a complex and chronic autoimmune disease with unknown etiology. The prevailing view is that the pathogenesis of MS involves autoreactive T cells, but also B lymphocytes, leading to immune responses against myelin or other antigens, inflammation, demyelination, and neuronal damage 1 , 2 . Experimental autoimmune encephalomyelitis (EAE), evoked by immunization with myelin proteins, is a widely used animal model for MS, which has been used to characterize the migration of autoreactive T cells across the blood–brain barrier and the demyelination of nerve axons 3 , 4 . EAE provides valuable insight into the early stages of immune cell activation and leukocyte extravasation across cerebral blood vessels that is relevant to MS and allows the explorative analysis of treatment options for acute MS and disease-modifying therapies. It is also established that the generation of encephalitogenic T cells capable of penetrating the blood-brain barrier involves activation steps in lymph nodes, spleen, intestine, and lung 5 , 6 , 7 , 8 , 9 .

Endothelial cells (ECs), which form the innermost lining of the vascular network, are central to inflammatory processes inside and outside of the central nervous system. In the brain, the initial loose interaction of circulating leukocytes with the activated endothelium is mediated mainly by integrin binding to vascular cell adhesion molecule 1 (VCAM-1) on the endothelial surface 10 , 11 . Chemokine signals presented on the luminal endothelial surface promote the activation of leukocyte integrins and, thereby, high affinity binding to endothelial adhesion molecules such as ICAM-1 (Intercellular Adhesion Molecule 1 or CD54), which induces arrest of the immune cells prior to transendothelial migration (TEM). To transmigrate through the endothelium and enter into the adjacent tissue, leukocytes migrate mainly paracellularly through endothelial junctions but, in some cases, also transcellularly through the cell body 12 .

Recent work indicates that inflammatory processes can be modulated by apelin, which is a peptide ligand for the G protein-coupled receptor (GPCR) APJ/Aplnr. Apelin is proteolytically processed, converting a larger 77-amino acid (aa) (in humans) preproprotein into various shorter products including a biologically active 13-aa isoform 13 . This short peptide is further activated by a pyroglutamyl modification at its N-terminus, which generates the highly potent [Pyr1]-apelin-13 (termed A13 in the article). Apelin, the related peptide ligand apela/elabela, and their common receptor APJ/Aplnr regulate cardiovascular morphogenesis during development, but there is increasing evidence linking these molecules to cancer and other pathobiological processes including inflammation-related diseases 14 , 15 , 16 .

In the present study, we show that administration of A13 delays and suppresses the development of EAE in mice. Analysis of apelin receptor expression by single cell RNA sequencing (scRNA-seq) and genetic labeling indicate that the GPCR is largely absent from adult brain endothelium, whereas expression is prominent in the lung and a few other peripheral organs. We find that A13 interferes with leukocyte transendothelial migration and the formation of immune cell clusters in the lung, a step that was previously shown to endow T cells with the capacity to enter the central nervous system 8 , 17 , 18 , 19 . Our data also indicate that apelin treatment induces desensitization and internalization of the receptor APJ/Aplnr and, accordingly, major outcomes of A13 administration are phenocopied by EC-specific inactivation of the Aplnr gene in adult mice or siRNA-mediated gene knockdown in cultured cells.

A13 suppresses development of experimental autoimmune encephalomyelitis

Mice were immunized with myelin oligodendrocyte glycoprotein peptide (MOG35-55 and injected intraperitoneally (i.p) with either A13 or vehicle (PBS) every other day commencing on the day of immunization (Fig.  1a ). Disease onset and severity were assessed daily and brains were removed for immunofluorescence and/or flow cytometry analyses on day (D) 16, corresponding to peak disease severity in PBS-treated mice. EAE incidence was significantly lower in A13-treated animals (Fig.  1a ). While EAE symptoms were detected in most of vehicle-treated animals by day (D) 13, less than 20% of mice in the A13 group developed symptoms by D16. Furthermore, the average disease score was significantly lower in A13-treated mice (Fig.  1a ), which continued to show lower disease severity until D30. Consistent with these findings, immunofluorescence staining of brains of A13-treated mice at D16 showed a lack of CD45+ immune cell infiltration (Fig.  1b, c ). Expression of the cell adhesion molecule ICAM-1 in the endothelium, an indicator of the inflammatory response 20 , was also strongly reduced in A13 brain samples (Fig.  1b ). Flow cytometry confirmed substantially lower numbers of CD4+ T cells, CD8+ T cells, Th1 helper cells, and IL-17-secreting (Th17) helper cells, important disease drivers in MS and EAE 21 , 22 , in EAE brains and spinal cords at D16 after A13 administration (Fig.  1e and Supplementary Fig.  1a, b ).

a Time course of MOG 35-55 immunization and A13 administration and analysis. Female C57Bl/6 mice aged 8-12 weeks received intraperitoneal (i.p.) injections of A13 or vehicle every other day commencing on the day of immunization (D0). Incidence rate and disease scores are shown up to day 30 (D30). Error bars, s.e.m. Two-way ANOVA. n  = 27 mice/group (D0-D12, 4 independent experiments), 21 mice/group (D13-D15, 3 experiments), 14 mice/group (D16, 2 experiments), 8 mice/group (D17-D30, 1 experiment). b Representative confocal images showing sagittal sections of brain cortex from vehicle or A13-treated (+A13) mice at peak EAE (D16) immunostained for CD31 (white), ICAM-1 (green) and CD45 (red). Scale bar, 50 µm. Disease score, vehicle=3, +A13 = 1. c Quantification of immune cells infiltrating brain cortex or lung per vascular area at D16 ( n  = 6 mice each). Error bars, s.e.m. Two-sided Student’s t -test. d Overview confocal images showing lung lobe sections at D16 EAE from vehicle control and A13-treated mice immunostained for CD31 (white) and CD45+ (red) immune cell clusters (arrowheads). Scale bar, 1 mm. Quantitation of CD45 + cells, CD3e + T cells, CD4 + T cells, CD8 + T cells, Th1 cells, and Th17 cells in brain ( e ) and lung ( f ) by flow cytometry at EAE onset (D11) and peak (D16). Error bars, s.e.m. Two-sided Student’s t -test. 5 mice per group.

Previous work has shown that the development of EAE pathology involves the reprogramming of T cells from an activated into a migratory mode in the lung, which is required to provide the cells with the capacity to enter the CNS 8 , 17 , 18 . Strikingly, immunofluorescence staining of tissue sections revealed changes in immune cell recruitment to the lung in A13-treated mice. While the initial accumulation of CD45+ immune cells and cluster formation was substantially delayed by A13 at D7, the resolution of these clusters, visible in vehicle-treated controls at peak EAE, occurred also significantly slower, leading to a much larger number of CD45+ foci in A13 lungs at D16 (Fig.  1d ; Supplementary Fig.  2a, b ). Likewise, flow cytometry indicated higher numbers of CD4+ T cells, CD8+ T cells, Th1 and Th17 in lungs of A13-treated mice at peak EAE (D16) but not at disease onset (D11) (Fig.  1f ). The weight of freshly isolated lungs was significantly higher in A13 treated mice than vehicle controls at peak EAE, but a smaller difference was already seen at D11 (Supplementary Fig.  3a ). As these data indicated that the trafficking of immune cells through the lung is compromised by A13, we used flow cytometry to measure total CD45+ immune cells in peripheral blood. The number of CD45+ and different T cell populations in peripheral blood were highest at disease onset (D11) in the control EAE group, and were substantially lower in A13-treated EAE animals at the same time point (Supplementary Fig.  3b ). Moreover, FACS analysis of the mediastinal lymph node, which is draining lymph and immune cells from the lung and other nearby organs, shows that A13 treatment leads to significant reduction of multiple immune cell populations at D11 but no longer at peak EAE (D16) (Supplementary Fig.  1a, b ). In contrast, immune cells in the spleen showed no significant differences between A13 and vehicle control. Moreover, analysis of inguinal lymph nodes (near the site of MOG35-55 injection) at D7 also showed no differences in immune cell content, indicating that the initial immune response is not compromised by A13 (Supplementary Fig.  1c ). Taken together, these data indicate that A13 treatment affects immune cell trafficking and thereby reduces the entry pathogenic cells into the central nervous system, which results in delayed EAE onset and reduced disease severity.

Next, we addressed the question whether A13 acts in an organ-specific fashion. While apelin receptor is broadly expressed by endothelial cells (ECs) in the developing mouse 23 , expression in the adult is more restricted and, as previously published scRNA-seq data show 24 , confined to certain organs such as heart, gut, kidney, testis and lung (Supplementary Fig.  4a ). To validate this data, we employed Aplnr-CreERT2 transgenic mice in a R26-mTmG Cre reporter background, which results in expression of green fluorescent protein (GFP) by Aplnr + cells upon tamoxifen injection. Shortly after tamoxifen administration to adult mice, GFP was detectable throughout the microvasculature of the adult heart and kidney, but was limited to only a few vessels in spleen and liver (Supplementary Fig.  4b ). The same strategy, analysis of Aplnr-CreERT2 -expressing cells shortly after tamoxifen administration, labeled lung microvascular ECs in early postnatal stages but also in adult (8-week-old) and aged (78-week-old) mice (Supplementary Fig.  4c ), respectively. In contrast, Aplnr expression was largely absent in adult brain, whereas GFP signal was readily detectable in postnatal brain endothelium where it also decorated a subset of Olig2+ oligodendrocyte-lineage cells (Supplementary Fig.  5a–d ). Downregulation of Aplnr transcripts in adult relative to postnatal brain ECs is also supported by scRNA-seq data 25 and transcripts for the GPCR were also not upregulated in EAE (Supplementary Fig.  5e ). These results argue that A13 is unlikely to directly affect brain endothelium in EAE, but, instead, might act through ECs in the lung or other organs.

We also analyzed intestine because this organ shows endothelial Aplnr expression in the adult (Supplementary Fig.  4a ) and is an important site of immune surveillance with established relevance for MS and EAE 9 , 26 , 27 , 28 . Analysis of intestinal sections, however, revealed no significant changes in tissue morphology or the abundance of CD45+ immune cells in A13-treated EAE mice (Supplementary Fig.  6a, b ).

A13 reduces the endothelial inflammatory response

Next, we investigated how A13 affects the accumulation of immune cells in the lung. Previous studies have shown that A13 treatment attenuates lipopolysaccharide (LPS)-induced acute lung injury 29 , 30 , 31 . To limit systemic effects and induce localized inflammation in the lung, we administered LPS (100 µg/40 µl) intranasally either alone or in conjunction with A13 (100 nM). Immunofluorescence staining revealed that A13 reduces the number of CD45+ cells in the lungs of treated mice relative to non-treated mice at 18 hours after LPS challenge (Supplementary Fig.  7a ). Four hours after treatment, flow cytometry showed that A13 strongly reduces the number of immune cells in bronchoalveolar lavage fluid (BALF) relative to controls, while immune cell numbers in peripheral blood were not significantly changed (Supplementary Fig.  7b and c). Consistent with the short A13 half-life 32 , differences between treated and non-treated mice were less apparent at 24 hours after treatment (Supplementary Fig.  7b ). Thus, a single treatment with A13 is sufficient to reduce acute immune cell recruitment to the lung.

Given that Aplnr expression is restricted to ECs, we investigated whether A13 might affect transendothelial migration of leukocytes. The cell surface glycoprotein ICAM-1 is expressed by ECs but also by epithelial and some immune cells. In ECs, ICAM-1 enables integrin-mediated stable adhesion of leukocytes and promotes inflammatory processes 33 . It is also known that endothelial ICAM-1 expression is strongly increased by inflammation and flow-induced shear stress 33 , 34 . Confocal imaging of immunostained tissue sections revealed that endothelial but not epithelial ICAM-1 staining intensity was significantly lower in A13-treated lungs relative to vehicle control at peak EAE (Fig.  2a ). qPCR analysis of extracted RNA from whole lungs and from FACS-isolated ECs confirmed the downregulation of Icam1 transcripts after A13 treatment (Fig.  2b ). In addition, transcripts encoding chemokine ligand 2 ( Ccl2 ) and interleukin 6 ( Il6 ), important mediators of inflammation, were also downregulated in A13-treated lungs (Fig.  2b ). In contrast, transcripts for the cell adhesion molecules VE-cadherin ( Cdh5 ) and Claudin 5 ( Cldn5 ), and for Gap Junction Protein Alpha 4 ( Gja4 ), which mediate EC-EC junctional interactions, were not significantly altered. Interestingly, the endogenous expression of apelin ( Apln ) was also significantly reduced by A13, whereas Aplnr transcript levels were slightly decreased in whole lung without a significant change in sorted ECs (Fig.  2b ). These results indicate that A13 treatment lowers endothelial ICAM-1 in the pulmonary endothelium but also reduces the expression of apelin itself and of cytokines that promote tissue inflammation.

a Lung sections from A13-treated (+A13) or vehicle control EAE mice stained for CD31 + (red) ECs (arrowheads) and ICAM-1 (green). Note reduction of endothelial but not epithelial ICAM-1 by A13. Graphs on the right show quantitation of CD31 + area and ICAM-1 + area as a fraction of total EC area in randomly chosen areas. Scale bar, 20 µm. Error bars, s.e.m. Two-sided Student’s t -test. n  = 21 (EAE) and 26 (+A13) areas from 5 mice in each group. b qPCR of whole lung and sorted lung ECs after 7 days of A13 treatment. Data normalized to control (Ctrl). Error bars, s.e.m. Two-sided Student’s t -test. n  = 6 (2 technical replicates for 3 independent samples). c Volcano plots comparing gene expression differences in HUVECs with or without A13 treatment (Ctrl, A13) both under static conditions and flow (15 dyn/cm²). DEG analysis was performed based on two-sided Negative Binomial model. d Enrichment plots for inflammatory or TNF-α response genes after GSEA analysis for each comparison. The green curves represent the running sum of the weighted enrichment score, which is higher in the flow (15 dyn/cm 2 ) control compared to the static control or A13-treated HUVEC. e , Heatmap of gene expression related to inflammation, comparing static and flow conditions (15 dyn/cm 2 ) with/without A13 treatment (upper panel). GSEA Hallmark analysis showing enriched gene sets (FDR ≤ 0.25) in HUVEC treated with A13 under flow (lower panel).

As fundamental endothelial physiological functions, such as vascular permeability, are influenced by fluid shear stress, we investigated whether A13 treatment alters the response to shear (15 dyn/cm 2 for 18 hours) in cultured human umbilical cord venous endothelial cells (HUVECs) by bulk RNA sequencing (Fig.  2c-e ). As expected, shear stress-induced a strong upregulation of the transcription factors Krüppel-like Factor 2 ( KLF2 ) and KLF4 , which was not affected by A13 (Supplementary Fig.  8 ). Apelin ( APLN ) was downregulated by fluid shear stress, whereas levels of apelin receptor transcripts ( APLNR ) were strongly increased, consistent with previous findings 35 . APLNR expression was further enhanced by A13 treatment (Supplementary Fig.  8 ). Importantly, shear stress-induced increases in the transcripts for ICAM-1 and VCAM were suppressed by A13 (Fig.  2c, e ; Supplementary Fig.  8 ). A13 also reduced the expression of E-selectin ( SELE ), a molecule that enables the initial rolling and loose attachment of leukocytes from the bloodstream in peripheral tissues, but had no effect on the related P-selectin molecule ( SELP ) (Fig.  2c, e ; Supplementary Fig.  8 ). Other regulators of the inflammatory response, including interleukin 6 ( IL6 ), interleukin 8 ( IL8 ), interleukin 1α ( IL1A ) and CXCL2, that have been reported to be upregulated by shear stress in human umbilical vein endothelial cells (HUVECs) 36 , 37 , were downregulated by A13 to levels seen without flow (Fig.  2c, e ; Supplementary Fig.  8 ). Gene set enrichment analysis (GSEA) confirmed that gene sets associated with the inflammatory response, tumor necrosis factor α (TNFα) signaling, IL6-STAT3 and IL2-STAT5 signaling were downregulated in response to A13 (Fig.  2d ). These results argue that A13 affects leukocyte extravasation through gene expression changes in ECs.

A13 suppresses inflammatory gene expression in Aplnr+ ECs, but not T cells

To systematically investigate EAE and A13-induced changes in different cell populations of the lung in vivo, we performed scRNA-seq analysis of untreated controls (8 to 9-week-old C57BL6 naïve females) and sex and age-matched mice at peak EAE after treatment with vehicle (PBS) or A13, respectively. Red blood cells were depleted prior to barcoding with the BD Rhapsody platform, library generation, and sequencing. The resulting data sets contain ECs, epithelial cells (type I and II), mesenchymal cells, monocytes, dendritic cells (DCs), neutrophils, T cells, B cells, and NK cells (Fig.  3a, b ; Supplementary Fig.  9a ). EAE and A13 treatment-induced gene expression changes in a variety of cell types including type I epithelial cells, matrix fibroblasts, and Aplnr + ECs, according to the analysis of differentially expressed genes (DEGs) (Supplementary Fig.  9b ). We initially focused on ECs to gain insight into the alterations in response to EAE and A13 treatment. By sub-clustering, we distinguished six endothelial subsets, which include arterial, venous, and lymphatic ECs together with several capillary EC subpopulations. Apln expression is prominent in Car4+ aerocytes 38 , 39 , 40 , which have been previously associated with gas exchange in the lung. Aplnr is found in general capillary (gCap) ECs as well as in a smaller EC subpopulation expressing glutathione peroxidase 3 ( Gpx3 ), an enzyme that protects against oxidative damage by catalyzing the reduction of hydrogen peroxide (Fig.  3c ; Supplementary Fig.  10a, b ). EAE caused a threefold increase in the relative fraction of lymphatic ECs, whereas the other EC subpopulations showed smaller changes (Supplementary Fig.  10a, b ). gCap ECs were the most responsive endothelial population in terms of DEGs for both EAE and A13 treatment (Supplementary Fig.  10b ). Gene Set Enrichment analysis using the Hallmark gene set 41 showed strong A13-induced reductions of TNFα/NFKB and transforming growth factor-β (TGFβ) signaling genes relative to vehicle-treated EAE lungs (Fig.  3d ). Inflammation-associated genes, including Cyp26b1 (cytochrome P450 family 26 subfamily B member 1), Csf1 (colony-stimulating factor 1), Sgms1 (sphingomyelin synthase 1), and Zfp36 (zinc finger protein 36), were downregulated by A13 in EAE (Fig.  3d ; Supplementary Fig.  10c ).

a UMAP plot of total cells from lung with color-coded cell types. Single cells were dissociated from three mouse lungs in each group (Naïve, EAE, EAE + A13) at D16 without any depletion, except red blood cell lysis. scRNA-seq was performed using BD Rhapsody and Illumina NextSeq 500 platforms. b UMAP plots of individual samples and bar graphs showing the population ratio of each cell type. c UMAP plot of the EC subset. d GSEA Hallmark analysis of enriched gene sets (FDR ≤ 0.25) in gCap ECs, comparing +A13 vs. EAE. Dot plot shows the expression and percent expression of inflammatory-related genes in gCap ECs for each condition. e UMAP plots of T cell subsets. f Population ratio of each T cell subset in each condition (upper panel) and a number of differentially expressed genes (DEGs) (adjusted p -value < 0.05, log2 fold-change>0.25) comparing Naïve vs. EAE or +A13 vs. EAE. DEG analysis was performed based on a two-sided Negative Binomial model.

We also analyzed immune cells in our lung scRNA-seq data (Supplementary Fig.  11a ). EAE raises the relative abundance of neutrophils 6-fold, classical monocytes 2-fold and T-helper cells (Th) cells 4-fold, whereas B cells decrease by more than 60% (Fig.  3b ). The abundance of Th cells as well as classical and non-classical monocytes was further elevated in EAE lungs treated with A13 (Fig.  3b, e, f ; Supplementary Fig.  11a, b ). The degree of myelin phagocytosis in MS coincides with CCR2+ monocyte invasion into the CNS 42 . Numerous mediators of inflammation, including C-C motif chemokine receptor 1 ( Ccr1 ), matrix metalloproteinase-8 ( Mmp8 ), S100 calcium-binding protein A9 ( S100a9 ), and chemokine (C-X-C motif) ligand 2 ( Cxcl2 ), were upregulated in classic monocytes in EAE lungs (Supplementary Fig.  11c, d ). Ccr1 transcripts and other inflammation-associated genes, such as Fn1 (encoding fibronectin 1), were reduced after A13 treatment, but changes in gene expression relative to vehicle-treated EAE lungs were small (Supplementary Fig.  11c, d ). We also looked into the changes affecting helper T cells (Th cells) and γδ T cells, which are critical players in the development of autoimmune diseases 43 . According to DEG analysis, EAE caused naïve T cells to acquire T cell activation-related factors such as galectin 1 ( Lgals ), Thy-1 cell surface antigen ( Thy1 ), the cell surface adhesion receptor CD44 ( Cd44 ), and B-cell lymphoma 2 ( Bcl2 ), a regulator of cell death (Supplementary Fig.  12a, b ). Th cells also acquired expression of markers associated with T cell activation, including Ccr2 ( Ccr2 ), galectin 1 ( Lgals ), and ADAM metallopeptidase domain 8 ( Adam8 ) (Supplementary Fig.  12a, c ). Notably, administration of A13 had no or only small effects on the expression of these genes, indicating normal T cell representation and activation in A13-treated lungs. Finally, we compared gene expression in T cells from the lung and brain at peak EAE (Supplementary Fig.  13a, b ). This revealed that T cell subpopulations isolated from the brain show much higher expression of activation markers 44 , 45 , 46 relative to lung. Some examples include interferon γ ( Ifng ), programmed cell death protein 1, also known as PD-1 or CD279 ( Pdcd1 ), transforming growth factor β1 ( Tgfb1 ), and chemokine ligand 5 ( Ccl5 ) (Supplementary Fig.  13b ). This result is consistent with the concept that T cells undergo multiple activation steps before exhibiting their full pathogenic potential inside the CNS. Indicating that the accumulation of immune cells in A13-treated lungs at peak EAE is not inducing pulmonary fibrosis, analysis of immunostained tissue sections revealed only a marginal elevation in collagen type I and immune cell cluster-containing areas of EAE showed no accumulation of α-smooth muscle actin (αSMA) in A13-treated samples relative to naïve lung (Supplementary Fig.  14a ). At D30, the predominant expression of collagen type I and αSMA was observed in the bronchial and large arterial regions, reflecting coverage by smooth muscle cells, whereas only little signal was visible across the lung parenchyma, except for the few remaining immune cell clusters (Supplementary Fig.  14b ).

Immune cells in the lung form clusters near apelin receptor-positive gCap ECs

As mentioned above, immune cells accumulated in the lung before the onset of EAE symptoms and formed clusters, which were already visible at seven days after MOG35-55 immunization (Supplementary Fig.  1a ). A13 treatment delayed this early formation of immune cell clusters, whereas the number of immune cells that were retained in the lung at EAE peak was strongly increased (Fig.  1f and Supplement Fig.  1a ). These immune cell clusters resemble inducible bronchus-associated lymphoid tissue (iBALT), which are tertiary lymphoid structures generated by immune cell aggregation in non-lymphoid organs 47 . iBALT is frequently seen in the lungs of smokers and other patients with chronic pulmonary diseases, and is associated with the activation of the adaptive immune system 48 , 49 . Within these structures, which are often associated with bronchial epithelium, B cells are selected, T cells are primed, and both cell populations increase in number during inflammation 50 , 51 . CD45+ clusters, similar to iBALT structures, were detectable later in immunostained lung sections from A13-treated EAE mice compared to controls (Supplementary Fig.  1 ) and tended to be located further away from bronchi (Fig.  4a, b ). Furthermore, A13 lungs contained higher proportions of T cells and less B lymphocytes compared to non-treated animals (Fig.  4c, d ; Supplementary Fig.  15a, b ). Relatively little is known about T cell migration within the pulmonary interstitial tissue, but previous work has proposed that the pulmonary vasculature is guiding this process 52 . Accordingly, we find that CD45+ immune cells were frequently detected near ECs in the alveolar microvasculature, which was enhanced by A13 treatment (Fig.  4e, f ). T cell receptor (TCR) complexes are typically located at the rear end (uropod) of migrating T cells 53 , 54 . Analysis of CD3e distribution by immunostaining shows that the subunit of the TCR complex was more polarized in T cells of vehicle-treated EAE lungs compared to A13 samples (Fig.  4f, g ). Taken together, these data suggest that A13 delays immune cell entry into the lung with implications for T cell polarization, the cellular composition of iBALT-like structures, and the resolution of immune cell cluster later during EAE.

a Immunostaining of CD45+ (red) immune cell clusters (arrowheads) and CD31 (white) in lung sections at D11 of EAE in vehicle-treated (EAE) and A13-injected mice (+A13). Scale bar, 1 mm. b Quantification of the distance between immune cell clusters and nearest bronchus in lung sections ( n  = 5 mice). Lower panel shows the number of immune cell clusters per distance and upper panel shows the ratio of clusters per distance relative to all clusters for each condition. c Immunostaining of EAE and +A13 lung sections at D11 (EAE onset) for CD31 (blue), CD45 (white), CD3e (green), and B220 (red). White lines outline immune cell clusters. Note low abundance of B220+ B cells in +A13 samples. Scale bar, 100μm. d Quantitation of CD3e+ or B220+ area normalized to CD45+ or CD31+ area in EAE and +A13 immune cell clusters (as shown in c ). Error bars, s.e.m. Two-sided Student’s t -test. n  = 28 (EAE) and 41 (+A13) areas from 5 mice for each condition. e Representative images of lung sections stained for CD31 (white) and CD45 (red) for EAE and +A13 at D16 (peak EAE). Yellow arrowheads indicate immune cells located in alveolar area. Scale bar, 100 μm (left) and 20μm (right). f Quantitation of CD45+ cells per alveolar area (as shown in e and ratio of T cells with polarized CD3e staining (green, as shown in g ). Error bars, s.e.m. Two-sided Student’s t -test. n  = 6 mice (EAE) and 7 mice (+A13) for e , n  = 12 each for g . Scale bar, 5μm (left) and 1μm (right). g Representative overview images and higher magnifications of insets showing CD3e immunostaining (green) with dashed lines marking outline of cells based on CD45 signal (red).

Next, we investigated the spatial relationship between immune cells and APJ/Aplnr-expressing ECs in mice carrying the Aplnr-CreERT2 allele together with R26-mTmG Cre reporter. Aplnr-CreERT2 -mediated labeling of Aplnr+ gCap ECs showed that these cells are strongly associated with iBALT-like structures and scattered interstitial CD45+ immune cells in EAE mice (Supplementary Fig.  15c ). Strikingly, more than 90% of CD45+ immune cells were already in contact with genetically labeled Aplnr - GFP  + ECs in naïve conditions (Supplementary Fig.  16a, b ). Another model of tissue inflammation, LPS treatment, increased the fraction of CD45+ cells in the proximity of GFP-negative CD31+ ECs and CD31-negative structures, but the majority of immune cells (70%) remained associated with Aplnr-GFP  + ECs under these conditions (Supplementary Fig.  16a, b ). Higher magnification images confirmed that Aplnr-GFP  + ECs were associated with CD45+ leukocytes, which is likely to indicate sites of leukocyte extravasation (Supplementary Fig.  16c ). Conversely, a similar genetic strategy using Apln-CreERT2 + for the labeling of ECs shows that CD45+ cells were not preferentially associated with apelin+ (GFP+) aerocytes (Supplementary Fig.  16d ).

Regulation of leukocyte recruitment by apelin receptor

During our analysis of immune cell association with Aplnr-CreERT2 -labelled cells, we observed that VE-cadherin immunostaining at junctions surrounding GFP+ ECs in EAE lungs was weaker relative to GFP-negative cells (Fig.  5a ). This difference between GFP+ and unlabelled ECs was eliminated in A13-treated EAE lungs (Fig.  5a ). To get further insight into the effect of apelin on EC behavior, we performed a series of in vitro experiments. As APLNR expression in HUVECs is low in the absence of flow, a receptor fusion construct carrying GFP at its carboxyterminus (APJ-GFP) was expressed by lentiviral infection and a similar strategy was used to generate control HUVECs expressing tdTomato fluorescent protein without APJ. In mixed cultures of tdTomato+ control and APJ-GFP+ HUVECs, VE-cadherin staining at junctions between APJ-GFP+ cells was less intense relative to junctions between tdTomato+ ECs, and this difference was reduced by A13 administration (Supplementary Fig.  17a ). VE-cadherin function is controlled through the phosphorylation tyrosine residues in the cytoplasmic region of the cell adhesion molecule. While tyrosine 731 (Y731) in VE-cadherin is constitutively phosphorylated by Src family kinases (SFKs) and, in particular, the kinase Yes 55 , dephosphorylation of Y731 is induced by interactions of ECs with T cells, which promotes T cell transendothelial migration 56 . Western blot analysis and immunostaining showed that A13 increased Y731 phosphorylation of VE-cadherin relative to vehicle control (Fig.  5b–d ). The A13-induced increase in phospho-Y731 was abolished by treatment with PP1 (Fig.  5d ), an inhibitor of all SFKs, consistent with previous reports linking Src family kinase activity to apelin 57 , 58 .

a High magnification confocal image of immune cell cluster in tamoxifen-treated Aplnr-CreERT2 R26-mTmG lung stained for VE-cadherin (red). White arrowheads mark VE-cadherin+ junction of APJ+ ECs. Scale bar, 50 μm. b Confocal image of HUVEC expressing apelin receptor with c-terminal GFP-tag (APJ-GFP) with or without A13 treatment, stained for VE-cadherin (red), phospho-VE-cadherin Y731 (white), APJ-GFP, and DAPI (blue). Note weak phospho-Y731 signal surrounding APJ-GFP+ cells (white arrowheads) but increase in response to A13 (yellow arrowheads), which also reduces APJ-GFP at cell-cell borders. Scale bar, 20 μm. c Quantification of phospho-VE-cadherin Y731 and APJ-GFP signal at cell perimeter, as indicated. Error bars, s.e.m. Two-sided Student’s t -test. n  = 157 (vehicle treated), n  = 153 (A13 treated) for both phospho-VE-cadherin Y731 and APJ-GFP. d Western blots of phospho-VE-cadherin Y658 , phospho-VE-cadherin Y731 , total VE-cadherin, and β-actin (loading control). APJ-GFP expressing HUVEC were treated with A13 and/or PP1, as indicated. Results from two separate stimulation experiments (#1, #2) are shown. e Quantitation of VE-cadherin at cell-cell contacts in tdTomato+ control HUVECs (T) and APJ-GFP+ HUVECs (G) (both pre-treated with LPS, with or without A13 treatment, as indicated). Error bars, s.e.m. Two-sided Student’s t -test. n  = 51 for tdTomato+ HUVECs and n  = 75 for APJ-GFP+ HUVECs treated with vehicle; n  = 58 for tdTomato+ HUVECs and n  = 64 for APJ-GFP+ HUVECs treated with A13. f Merged images showing phase-contrast of HUVECs (pretreated with LPS) and fluorescently labelled CD4+ T cells (top). Bottom panels show isolated fluorescent channel highlighting A13-induced reduction of adherent CD4 + T cells (purple). Representative movies of brightfield channels are provided as Supplementary Movies  1 - 4 . Quantitation of adherent CD4+ T cells, crawling distance, and ratio of transcellular vs. paracellular trans-endothelial migration on tdTomato+ control HUVECs (T) and APJ-GFP HUVECs (G) with or without A13 treatment, as indicated. Error bars, s.e.m. Two-sided Student’s t -test. Scale bar, 50 μm. n  = 6 areas for each group from 2 independent experiments ( g , left panel). n  = 244 cells for all conditions except APJ-GFP+ HUVECs +A13 ( n  = 104) ( g , right panel). n  = 9 areas for each group from 2 independent experiments ( h ). i Quantification of CD4+ T cell migration across APJ-GFP HUVECs in transwell assay. 6 individual wells/condition from 2 independent experiments. Error bars, s.e.m. Two-sided Student’s t -test.

Immunostaining also revealed localization of APJ-GFP at regions of cell-cell contact in vehicle control-treated cells, which was lost in A13-treated cells, resulting in perinuclear accumulation inside ECs (Fig.  5b ; Supplementary Fig.  17a ). To exclude the possibility that the C-terminal GFP fusion might alter the function of the GPCR, we also constructed an APJ-T2A-GFP lentiviral construct, in which self-cleavage generates separate APJ and GFP protein products. Coculture of APJ-T2A-GFP and tdTomato HUVECs confirmed weaker VE-cadherin staining at junctions between GFP+ cells relative to tdTomato+ cells, and this difference was diminished by A13 treatment (Supplementary Fig.  17b ). As no major changes in Cdh5 transcript level were observable in pulmonary EC subpopulations in vivo or in HUVECs in vitro, the effects of A13 on VE-cadherin are likely to be posttrancriptional (Supplementary Fig.  17c, d ).

ICAM-1, a known mediator of leukocyte arrest, intravascular crawling and transendothelial migration, is upregulated in EAE 12 , 20 , 33 , but was reduced by A13 treatment in pulmonary ECs in vivo and in HUVECs in vitro (Fig.  2a-c ). We therefore employed in vitro assays to directly assess the effects of A13 on T cell adhesion and transmigration across HUVECs and on ICAM-1 expression. Analysis of transmigration under flow confirmed that APJ-GFP expression by HUVECs increased T cell adhesion relative to tdTomato+ control HUVECs. A13 reversed this effect leading to low levels of T cell adhesion to APJ-GFP+ cells and had no significant effect on tdTomato+ control HUVECs (Fig.  5f, g ). Live imaging analysis showed that crawling distances of T cells on HUVECs were slightly increased by A13 administration (Fig.  5g ). The same approach also indicated that A13 treatment of APJ-GFP+ cells led to a small but significant increase in transcellular migration events relative to paracellular (junction-mediated) migration (Fig.  5g, h ). In transwell assays, the migration of T cells across an APJ-GFP HUVEC monolayer in response to CXCL12/SDF-1 was also reduced by A13 (Fig.  5i ).

Together, these results show that apelin reduces the transendothelial migration of T cells, but possibly also of other immune cells, which involves reduced adhesion to ECs.

Ablation of apelin receptor in ECs recapitulates the effect of A13 in EAE

Ligand binding induces signaling by G protein-coupled receptors but, depending on ligand concentration and other factors, it can also limit signal transduction by triggering GPCR internalization and removal from the cell surface in a process termed desensitization 59 . Indeed, it was shown that A13 can desensitize apelin receptor, which involves internalization via clathrin-coated vesicles 60 . As our own in vitro experiments supported the possibility of A13-induced APJ/Aplnr internalization (Fig.  5b ), we generated Cy5-tagged fluorescent version of A13 (A13-Cy5) for stimulation experiments. At 2 hours after treatment, internalized A13-Cy5 speckles were visible inside APJ-GFP-expressing HUVECs but not in tdTomato+ control cells. In vehicle-treated HUVECs, APJ-GFP was readily detectable at cell-cell contact sites and this signal was lost after A13-Cy5 administration and APJ internalization (Supplementary Fig.  17e ). The sum of the data above indicates that A13 removes its receptor from the cell surface through internalization, resulting in a transient loss-of-function situation limiting inflammatory leukocyte transmigration across the endothelial monolayer.

Next, we performed bulk RNA-sequencing of cultured HUVECs in a loss-of-function setting, namely after siRNA-mediated knockdown of apelin receptor expression ( siAPLNR ). This approach led to the downregulation of inflammatory and TGFβ signaling-related gene expression both under fluid shear stress and static conditions (Fig.  6a, b ). Similar to A13 treatment (Supplementary Fig.  6 ), siAPLNR reduced the low baseline expression of ICAM1 , VCAM and SELE , but the effects were not as pronounced as those measured after A13 administration (Supplementary Fig.  18a ). Moreover, IL8 , IL6 and CCL2 , which were downregulated by A13 administration, were increased in siAPLNR HUVECs (Supplementary Fig.  18a ).

a Heatmap showing expression of inflammation-related genes in HUVECs treated with control siRNA (siCtrl) or APLNR siRNA (si APLNR ) under flow conditions. Red arrowheads indicate genes mentioned in text. b Volcano plot comparing gene expression of siCtrl and si APLNR treated HUVECs under flow (left panel). Plots show enrichment of TGF-β or inflammatory response genes in GSEA analysis (right panels). DEG analysis was performed based on two-sided Negative Binomial model. c Confocal images of lung sections from EC-specific Aplnr knockout mice ( Aplnr iECKO ) and littermate controls at EAE peak stained for ICAM-1 (green), CD31 (red) and CD45 (white). Arrowheads indicate endothelial ICAM-1 and downregulation in Aplnr iECKO lung. Scale bar, 20 μm. d Quantitation of vessel density (CD31+ area per unit area) and ratio of ICAM-1+ (green) area per total area or CD31+ area. Error bars, s.e.m. Two-sided Student’s t -test. n  = 7 for Ctrl, n  = 10 for Aplnr iECKO . e Disease score and incidence rate of EAE in 8-12 week-old female Aplnr iECKO mice and age and sex-matched littermate controls (Ctrl). Error bars, s.e.m. two-way ANOVA. n  = 6 mice for each condition. Number of CD45+ cells, CD3e+ T cells, CD4+ T cells, CD8+ T cells, Th1 cells, and Th17 cells in Ctrl and Aplnr iECKO whole brain ( f ) and lung ( g ) at peak EAE. Error bars, s.e.m. Two-sided Student’s t -test. n  = 5 mice/group.

For functional studies in vivo, we generated EC-specific Aplnr knockouts ( Aplnr iECKO ) by interbreeding of Cdh5-CreERT2 mice 61 with animals carrying a loxP-flanked Aplnr gene 62 and tamoxifen treatment. In contrast to brain, both the endothelium and the epithelium in the lung shows constitutive expression of ICAM-1, which is further upregulated in response to inflammation 63 , 64 . Staining of lung sections confirmed constitutive ICAM-1 expression in control littermates, whereas ICAM-1 immunofluorescence was strongly reduced in Aplnr iECKO mutants (Supplementary Fig.  18b ). Decrease of ICAM-1 was also seen in pulmonary ECs of adult (8-week-old) Aplnr iECKO mutants (treated with tamoxifen 1 week earlier), without alterations in the density of pulmonary vessels (Fig.  6c, d ). Aplnr iECKO mutants showed reduced accumulation of CD45+ cells in response to LPS exposure (Supplementary Fig.  19a, b ). Importantly, EAE progression was delayed and disease severity reduced in Aplnr iECKO mutants relative littermate controls (Fig.  6e ). Flow cytometry showed decreased immune cell infiltration into Aplnr iECKO brains and elevated accumulation of immune cells in the lung compared with control littermates (Fig.  6f, g ). Thus, genetic inactivation of apelin receptor expression in ECs recapitulates critical aspects of the A13-induced protective phenotype, arguing that decreased apelin receptor function is a common feature in both conditions. The sum of these data argues that Aplnr + gCap ECs facilitate immune cell entry into the lung in the EAE model but also in response to LPS treatment. Immune cell recruitment into the lung is impaired by A13, which induces internalization of apelin receptor, or by genetic inactivation of the Aplnr gene in ECs.

A13 treatment during EAE development ameliorates disease symptoms

To investigate the therapeutic potential of A13 in EAE and also rule out potential effects of the peptide on the initial immune response after MOG35-55 administration, mice were treated with A13 every second day from D7 or daily from D11 (at disease onset) (Fig.  7a ). Strikingly, both late treatment regimens significantly reduced disease severity relative to vehicle control animals (Fig.  7a ). A13 treatment from D7 also delayed the onset of disease symptoms. Consistent with our previous results, lung lobe sections of mice treated with A13 from D11 presented a higher number of CD45+ immune clusters relative to vehicle control at D20 (Fig.  7b, c ). Moreover, the cerebellum of A13 mice at D20 showed reduced immune cell infiltration together with lower ICAM-1 expression (Fig.  7b, d ). These results support that the benefits of A13 are not caused by the suppression of the initial immune response in the EAE model (before D7 or D11, respectively), but they also raise the possibility that A13 might allow therapeutic intervention in an early stage of disease development with potential relevance for MS.

a Time course of MOG 35-55 immunization and A13 administration and analysis. Female C57Bl/6 mice aged 8-12 weeks received i.p. A13 or vehicle injections every other day from D7 or daily from D11 (Onset). Shown are incidence rate and disease scores on the indicated days after immunization. Error bars, s.e.m. two-way ANOVA, n  = 7 mice per group. b Quantification of immune cells infiltrating the brain and lung per area, number of clusters in the lung and ICAM-1 + EC ratio in the brain at D20 of EAE and +A13 from D11 onwards ( n  = 6 mice per group). Error bars, s.e.m. Two-sided Student’s t -test. c Overview confocal images of EAE or +A13 (D11 onwards) lung lobe sections from D20 immunostained for CD31 (white) and CD45 (red). Scale bar, 1 mm. White arrowheads indicate immune cell clusters. d Representative confocal images of control (EAE) and +A13 (D11 onwards) cerebellum at D20 immunostained for CD45 (white), ICAM-1 (green) and CD31+ (red). Scale bar, 100 μm. Disease score, 2 for EAE and 1 for +A13.

It has been previously suggested that A13 might be a disease biomarker that is elevated in the serum of MS patients relative to healthy controls 65 . We, therefore, analyzed serum samples of healthy controls and MS patients in relapse or remission (Supplementary Fig.  20a ). This revealed significantly higher levels of serum A13 in healthy males relative to healthy women. Female MS patients exhibited an increase in serum A13 during relapse and a smaller, statistically not significant elevation during remission, which raised levels to those seen in healthy males. Disease status had no measurable impact in male samples (Supplementary Fig.  20a ). Given that MS is more prevalent in women than men 66 , these findings raise the interesting possibility that higher levels of A13 in male serum might be potentially relevant for protection against the disease in humans.

The etiology of MS in patients remains incompletely understood, but research using EAE as an animal model has indicated a role of T cell reprogramming in the lung, which induces a migratory phenotype and enables the trafficking of these cells into the CNS 8 , 18 . Within the lung, chemokine signals direct T cells into bronchus-associated lymphoid tissue before they re-enter the blood circulation through lung-draining lymph nodes 8 . These findings are consistent with other reports showing that iBALT in the lung can mediate substantial respiratory immune responses even without an essential role of traditional secondary lymphoid organs 47 , 67 . Our study shows that A13 delayed immune cell recruitment to the lung but also the subsequent resolution of immune cell clusters in the pulmonary parenchyma, which led to reduction of the number of immune cells in peripheral blood and in mediastinal lymph nodes. These changes were associated with reduced entry of pathogenic T cells into the CNS, resulting in delayed EAE progression and milder disease symptoms relative to vehicle-treated EAE animals. We also provide evidence that treatment at the onset of EAE symptoms can reduce disease score, which suggests that further exploration of the therapeutic potential of A13 could be worthwhile. At the same time, limitations of EAE as a transient model of MS need to be considered. EAE either leads to death or partial recovery of the affected animals 68 , 69 . The latter is associated with the deletion of encephalitogenic T cells and, in models involving active immunization, resistance to subsequent induction of EAE. Therefore, the role of A13 in chronic neuroinflammation cannot be studied. The primary immune cell types implicated in the manifestation of symptoms are also not identical for the human disease and its animal model. CD8+ T-cells and B-lymphocytes have been associated with progressing inflammation and tissue damage in MS, whereas pathogenesis in EAE is mediated by auto-reactive CD4+ T-cells 70 .

The effect of A13 is not restricted to EAE and the ligand also successfully reduced pulmonary inflammation in response to intranasally administered LPS, which is consistent with various previous reports showing that A13 can protect against acute lung injury 29 , 31 , 71 , 72 . Most of these studies have attributed the beneficial effects of apelin to altered signaling in immune cells 29 , 31 , 72 , whereas the EC-specific knockout of Aplnr and in vitro experiments in our study indicate an important and direct role of ECs through the regulation of leukocyte transendothelial migration. In our in vivo experiments, we noted that the protection against EAE symptoms is more efficient in A13-treated mice than after EC-specific Aplnr inactivation. One possible explanation could be effects on other cell populations even though Aplnr expression is largely, but not completely, confined to ECs. Moreover, A13 might also trigger APJ signaling and not only desensitization of the GPCR, whereas the genetic approach generates an irreversible loss of apelin receptor function without active signal transduction. Finally, it should also be considered that systemic administration of A13 can exert vasodilatory and anti-hypertensive effects, which has been, at least in part, attributed to the modulation of vascular smooth cell function 73 , 74 .

Our data on serum levels of A13 in MS patients support its use as a biomarker at least in female patients. Two previous studies had reported variable results between male 57 and female MS patients 75 , but it has to be considered that sample numbers were small in both studies and that disease state, comorbidities and treatment differences might result in substantial variability. Given that MS is more prevalent in women than men 66 , these findings raise the interesting possibility that higher levels of apelin in male serum might be potentially relevant for protection against the disease. However, it also needs to be considered that the biological effect of constant exposure to endogenous apelin might not be equivalent to acute A13 administration. The functional modulation of Aplnr + ECs, which are, as our data show, associated with leukocyte transendothelial migration in the lung, by apelin and other factors deserves further investigation in future studies.

At the mechanistic level, we link the biological effect of A13 to reduced T cell adhesion to ECs (Supplementary Fig.  20b ). E-selectin, one of the critical TEM mediators in peripheral tissues that is downregulated by A13, functions in the rolling interactions that decelerate leukocytes and position them in close proximity to the endothelial monolayer 76 . A13 also reduces the endothelial expression of ICAM-1 and VCAM-1, which mediate leukocyte adhesion through interactions with integrin receptors and are both upregulated by inflammation. Interestingly, previous work has shown that effector T cells are temporarily entrapped in the pulmonary vasculature on their way to lymph nodes during systemic inflammation 63 . T cell entrapment in the lung is reduced in mice lacking ICAM-1 and the related ICAM-2 molecule, whereas neutrophil recruitment is unaffected 63 . Endothelial VCAM-1 participates in leukocyte diapedesis, which also involves downstream signalling events that help to open endothelial junctional cell-cell contacts 77 , 78 , 79 . This might also explain why some of our findings link A13 and APJ/Aplnr internalization to junctional localization of VE-cadherin, whereas Cdh5 transcript levels are not altered. Other processes, including altered expression of the inflammatory cytokines CCL2 and IL6 (Supplementary Fig.  20b ), might also help to explain why A13 administration delays immune cell trafficking through the lung and thereby suppresses the development of EAE. Similarly, the persistence of immune cell clusters in the lung of A13-treated mice at peak EAE might be caused by altered cytokine expression, the initial delay in immune cell accumulation in the lung or impaired migration towards lymphatic vessels. Regarding the latter, it should be noted that various studies have revealed important roles of apelin signaling in the growth and remodeling of the lymphatic vasculature 80 , 81 , 82 . Genetic approaches could be used to distinguish between the effects of A13 on the ECs of blood vessels or those of the lymphatic vasculature in future studies.

There is currently no cure for autoimmune disorders such as MS, and available treatments can have harmful side effects and need to be carefully monitored because they generate an increased risk of infection 83 , 84 . Further research will be needed to address whether A13 or other mediators of apelin signaling might be useful for the suppression of acute or chronic inflammation either alone or in combination with other treatment options.

Mice and EAE mouse model

To generate EC-specific Aplnr knockouts ( Aplnr iECKO ), mice carrying a loxP-flanked Aplnr allele 62 were interbred with Cdh5-CreERT2   61 transgenic animals. For analysis at postnatal day 6 (P6), 50 μl of 1 mg/ml tamoxifen (Sigma T5648) was administered at P1, P2, and P3. For studies in adult mice, 200 μl of 10 mg/ml tamoxifen were given on 3 consecutive days, 1-2 weeks prior to analysis or EAE induction. For the analysis of interactions between ECs and immune cells, Aplnr-CreERT2 85 or Apln - CreERT2 mice 86 were bred to R26-mTmG Cre reporter animals 87 . The resulting offspring were treated with tamoxifen as described above. EAE was induced in 8-12-week-old C57Bl/6 wild-type females or in Aplnr iECKO or Aplnr-CreERT2 R26-mTmG females and littermate controls, respectively. Experimental procedures adhered to previous studies 88 . Briefly, a subcutaneous injection of 150 μl of MOG 35-55 (112 μg/ml in emulsion with Complete freunds adjuvant (CFA)) was administered near the tail base, along with 100 μl of Pertussis toxin (Sigma P7208, 2 μg/ml) via the tail vein. The following day, 100 μl of Pertussis toxin (2 μg/ml) was administered once again. Throughout the experiment, 100 μl of A13 (10 nM, Bio-techne, #2420) or PBS was administered intraperitoneally to the mice, and body weight and disease severity according to the EAE clinical severity scale 89 were assessed daily. During EAE experiments, vehicle control and A13-treated mice were kept in separate cages within the same room. Animal experiments were not blinded because treatments had to be indicated on cage labels and in our animal database for legal reasons. The size of experimental groups was selected on the basis of previous experiments 88 . Moreover, A13-treated animals showed obvious phenotypic differences during the course of the EAE experiments. Mice were euthanized and dissected on the designated day as detailed in the main text, following terminal anesthesia. All animals were housed in a dedicated pathogen-free facility, and experiments were conducted in compliance with applicable laws and institutional guidelines. The experiments were performed after ethical review and with the necessary permissions granted by the Landesamt für Natur, Umwelt und Verbraucherschutz (LANUV) of North Rhine-Westphalia, Germany.

Intranasal LPS challenge

A concentration of 2.5 mg/ml of LPS (Sigma, L2630) was dissolved in PBS, with or without A13 (10 nM). 8-12 week-old female C57Bl/6 wild-type mice were sedated using isoflurane. Subsequently, 20 μl of LPS solution was administered to the nostrils of each mouse under anesthesia. Mice were kept in an upright position until they regained consciousness. Animals were sacrificed and lungs were prepared for further processing after 18 h for section staining and 4 or 24 h for BALF analysis.

Organ isolation and immunostaining

Mice were examined for reflexes following anesthesia induced by Rompun and Ketamine. Anesthetized mice had their caudal vein cut and the chest was opened. DPBS (Dulbecco’s Phosphate-Buffered Saline) was injected through the right ventricle until the lung became pale. Subsequently, 0.1% low melting agarose was administered via the trachea, and the trachea was clamped for 15 minutes. The organs were then prepared and fixed in 4% PFA (Paraformaldehyde) overnight at 4 °C. Vibratome (Leica VT1200S) was used to cut the organs into 100μm or 200μm slices. Sections were stored in an anti-freezing media at −20 °C until use. To remove the anti-freezing media prior to staining, sections were rinsed three times with PBS (Phosphate-Buffered Saline). A blocking buffer and antibody diluent containing 5% Donkey serum in 0.1% Triton X-100 PBST (Phosphate-Buffered Saline with 0.1% Tween-20) were used. The blocking and antibody treatment was performed overnight at 4 °C. Finally, samples were washed three times for 15 minutes each with 0.1% Triton X-100 PBST at room temperature.

Analysis of CD3e polarity

Utilizing Fiji, individual cells identified through CD45 staining were delineated and a macro was employed to partition these cell into semicircles, measuring CD3e staining intensity on each side separately. The ratio of intensity between one semicircle and the other was computed. Subsequently, the dividing line oft he semicircle was rotated by 45 degrees and intensity measurement and calculation was repeated three more times. The results from these four calculations were aggregated for each cell and cells with calculated values exceeding 2 or below 0.5 were classified as polarized.

A list of antibodies used in the study is shown in Table  1 .

Lung dissociation for the scRNA sequencing and EC isolation

Lung tissue was dissociated using a solution containing papain (Worthington, LK003150) and liberase (Roche, 5401054001). Dissociation was carried out for 40 minutes at 37 °C in a water shaker. Cells were collected by centrifugation and the pellet was resuspended in MEM. The cell suspension was then filtered through a 100μm mesh to remove large debris, followed by treatment with RBC lysis buffer (Sigma, R7757) to eliminate red blood cells. A second filtration step using a 40-μm mesh was performed. The cell count was determined using a cell counter (Logos, L10001). For single-cell RNA sequencing, the cell suspension was diluted and aliquoted at a concentration of 40,000 cells per 100 μl in FACS buffer (0.2% FBS, 2 mM EDTA in PBS). The BD Rhapsody single-cell analysis system was utilized to prepare the single-cell RNA libraries, which were subsequently sequenced using the Illumina Next-seq 500 platform with a mid-output setting. To isolate ECs, CD45+ cells were depleted using the MACS method and CD45 microbeads (Miltenyi Biotec, 130-052-301). Next, cells were stained with PDPN-PE (12-5381, eBioscience), CD45-Pecy7 (25-0451-82, eBioscience), and CD31-APC (FAB3628A, R&D) for 30 minutes at 4 °C. The stained cells were washed three times with FACS buffer. Using FACS ARIA (BD, FACSARIA IIIU), CD45-negative cells were gated, and CD31 + PDPN- cells were sorted for further applications.

FACS analysis for the brain and lung-infiltrating immune cells

Mice were anesthetized using Ketamine and Rompun. Peripheral blood was collected from the caudal vein using a needle and syringe into an anticoagulant tube. To remove red blood cells, the collected blood was treated three times with RBC lysis buffer (Sigma cat. R7757). Cells were then kept in FACS buffer on ice for further FACS analysis. For the analysis of brain and lung tissue, mice were perfused with DPBS as described in the method section on organ isolation. Lungs and brains were minced in a 2 ml eppendorf tube, and 1 ml of FACS buffer was added until the color became homogeneous. Lung and brain pieces were dissociated in FACS buffer in a 6-well plate using a 100μm mesh and plunger. The cell suspension was collected with 10 ml of FACS buffer and centrifuged to obtain a cell pellet. Cell pellets were resuspended in 37% Percoll and loaded on top of 70% Percoll (Cytiva, 17089102). After centrifugation, the mononuclear cell layer was collected. Cell pellets were resuspended in 1 ml of FACS buffer and the cell concentration was measured using a cell counter. For CD45, CD4, and CD8 T cell analysis, 10 5 cells were stained directly with CD45-APCcy7, CD3e-FITC, CD4-APC, and CD8-PE at 4 °C for 20 minutes, followed by three washes with FACS buffer. The stained cells were then analyzed using a FACS analyzer (FACSymphony A3). The remaining cells were cultured with 10% FBS RPMI 1640 medium supplemented with 10 ng/ml PMA (Sigma, P1585), 1 μg/ml ionomycin (Sigma, IO634), and 10 μg/ml Brefeldin A (Sigma, B6542) for 6 hours. For IFNγ-PECy7 and IL-17A Alexa Fluor 488 staining, the BD Cytofix/Cytoperm kit (BD, VDV554715) was used according to the manufacturer’s instructions, followed by staining with CD45-PE and CD4-APC. The stained cells were then analyzed using a FACS analyzer (FACSymphony A3). Gating strategies are shown in Supplementary Figs.  21 and 22 .

Cell culture and flow experiment

HUVECs were cultured in EC proliferation medium (Provitro, 201 0001) on dishes coated with 0.5% gelatin. Cells were used in experiments until passage 5. For the flow experiment, 10 5 cells were seeded on gelatin-coated 0.6 mm u-slide I Luer chambers (ibidi 80186). The media was changed 4 hours after seeding to complete media before cells were incubated overnight in a 5% CO2, incubator at 37 °C. Subsequently, cells were starved for 12 hours using starvation media. Slides were then connected to the flow system (Ibidi, 10902). Shear stress (15 dyn/cm 2 ) was applied to cells for 12 hours, followed by an additional 6 hours of shear stress exposure with or without 1 μM A13.

Transendothelial cell migration assay

10 5 HUVECs were seeded on Fibronectin-coated inserts of transwells (Corning, 3415) and cultured for 24-36 hours until they reached confluence. After the medium was then replaced by medium with or without 1 μM A13, cells were placed in the cell culture incubator for an additional 48 hours. After removing the medium, 10 5 human CD4 T cells in 0.5% BSA/RPMI medium were added to the insert, which was placed in a well filled with chemoattractant medium (50 ng/ml Cxcl12, R&D, 350-NS-010/CF and 0.5% BSA/RPMI). After six hours, cells in the suspension of the lower chamber were counted.

Lentivirus generation and infection

The lentiviral APJ-GFP vector was purchased from Origene (RC207576L2) and coding sequences for tdTomato and APJ-T2A-GFP were inserted into the same lentiviral vector backbone. Vectors were expanded using DH5a cells and a Midiprep kit (Macherey-Nagel, MN740410.50). Lentiviruses were generated by transfecting 293 T cells with the lentiviral vector and packaging vectors (pMD2.G and psPAX2). The lentivirus-containing media were concentrated using Lenti X concentrator (Takara, PT4421-2) to 100x and stored at −80 °C. The concentrated lentiviruses were used to treat P1 HUVECs with 8 µg/mL polybrene overnight. The following day, the medium was replaced and the lentivirus-treated HUVECs were used for further assays and analysis.

BALF cell isolation and blood cell contents measuring

Mice were fully anesthetized with Rompun and Ketamin before an incision was made in the neck to remove the salivary gland and expose the trachea. The head was fixed to straighten the trachea and intubation needles were inserted. Needles were slowly withdrawn while keeping the tube inside the trachea. The trachea and tube were tightly secured with nylon threads. A 1 ml syringe filled with 600 μl of DPBS was connected to the tube. DPBS was injected slowly and drained three times by manipulating the plunger, with the collected fluid being transferred to a 2 ml Eppendorf tube. This DPBS collection process was repeated three times. Blood was obtained from the abdominal vein and collected in a heparin-coated tube. The collected DPBS volume was measured using a pipette, and cell concentration was determined using a Luna cell counter. The BALF and blood cells were analyzed using the Scil Vet abc Plus+ system following the manufacturer’s instructions.

Knockdown of APLNR

For the downregulation of the apelin receptor in HUVECs, siAPLNR (Invitrogen, HSS100324) was used with siCont (Invitrogen, 12935200) serving as control. The transfection reagent was prepared by mixing 3.6 µL of siRNA (20pmol/µL) and 9 µL of RNAiMax (Invitrogen, 13778075) diluted in Opti-MEM before the mixture was incubated for 20 minutes. The transfection reagent was then added to 10 5 adherent HUVECs and allowed to incubate overnight. The assay was conducted between 24-48 hours post-transfection.

Bulk RNA seq and qPCR

RNA from HUVECs, tissues, and sorted cells was isolated using the RNeasy Plus Micro kit (Qiagen, 74034). For qPCR analysis, cDNA was synthesized using the iScript cDNA synthesis Kit (Bio-Rad, 1708891), and qPCR was carried out using TaqMan primers (listed below) and the SsoAdvanced Universal Probes Supermix (Bio-Rad, 1725284). For bulk RNA-seq, libraries were generated using NEBNext (NEB, E7760L), and sequencing was performed on an Illumina NextSeq 500 with mid output.

The following primers were used for qPCR: Cdh5(Mm00486938_m1), Cldn5(Mm00727012_s1), Klf4(Mm00516104_m1), Apln(Mm00443562_m1), Aplnr(Mm00442191_s1), Icam1(Mm00516023_m1), Il6(Mm00446190_m1), Ccl2(Mm00441242_m1), Gja4(Mm01179783_m1), Actb(Mm00607939_s1).

Bioinformatics analysis

The sequencing data in FASTQ format was processed using the BD Rhapsody WTA Analysis pipeline (version 1.0) on the SevenBridges Genomics online platform. The resulting expression matrix was used for further data analysis. Data normalization, dimensionality reduction, and visualization were performed using Seurat (version 4.3.0) unless specified otherwise.

For initial quality control of the extracted gene-cell matrices, cells were filtered using the following parameters: nFeature_RNA > 500 & nFeature_RNA < 6000 for the number of genes per cell, percent.mito <25 for the percentage of mitochondrial genes, and genes with parameter min.cells = 3. Filtered matrices were normalized using the LogNormalize method with a scale factor of 10,000. Variable genes were identified using the FindVariableFeatures function with the following parameters: selection.method = “vst”, nfeatures = 2000, trimmed for the genes related to cell cycle (GO:0007049), and then used. Data integration was performed using the FindIntegrationAnchors and IntegrateData functions with default options. Statistically significant principal components were determined using the JackStraw method, and the first 9 principal components were used for UMAP non-linear dimensional reduction.

Unsupervised hierarchical clustering analysis was performed using the FindClusters function in the Seurat package. We tested different resolutions between 0.1 ~ 0.9 and selected the final resolution using the clustree R package to determine the most stable and relevant resolution for our previous knowledge. The cellular identity of each cluster was determined by finding cluster-specific marker genes using the FindAllMarkers function with a minimum fraction of cells expressing the gene over 25% (min.pct=0.25) and a log2 fold change threshold of 0.25 (logfc.threshold=0.25).

For subclustering analysis, specific cluster(s) were isolated using the subset function, and the data matrix was extracted from the Seurat object using the GetAssayData function. The whole analysis pipeline was then repeated from data normalization. The FeaturePlot, VlnPlot, and DoHeatmap functions of the Seurat package were used for visualization of selected genes. Gene set enrichment analysis (GSEA) was performed using the Molecular Signatures Database (MSigDB) v7.1 hallmark gene sets (mouse version) with the fgsea R package (version 1.24.0).

For bulk RNA-seq analysis, total RNAs were extracted using the RNeasy Plus Micro kit (Qiagen). The quality and quantity of RNA samples were analyzed with a Bioanalyzer and RNA 6000 pico kit (Agilent). Double-stranded cDNA was synthesized using the SMART-Seq v4 Ultra Low Input RNA kit for Sequencing (Takara), and sequencing libraries were constructed with the Nextera XT DNA Library Preparation Kit (Illumina). The resulting sequencing libraries were sequenced with 2 × 75 bp paired-end reads on the NextSeq 500 sequencer (Illumina). Sequenced reads were aligned to the human (hg38) reference genome using TopHat (version 2.1.1), and the aligned reads were used for transcript quantification by using HTSeq-count (version 0.6.1). DESeq2 (version 1.44.0) was used to identify differentially expressed genes across the samples.

T cell isolation

In line with Dutch legislation and the Declaration of Helsinki, 50 mL peripheral whole blood was drawn from healthy volunteers. All volunteers provided informed consent and all protocols were approved by the Amsterdam University Medical Centre ethical committee (METC). Blood was processed within 2 hours of donation.

First, 10–20 mL of blood was mixed 1:1 with PBS + 5% TNC (tri-sodiumcitrate, Merck, 1.06447.5000) and gently layered on top of 12.5 mL Ficoll (Cytiva, 17144003) in a 50 mL tube. This tube was centrifuged at 800 x g for 20 minutes with slow start and no brake. After centrifugation, the PBMC ring was pipetted out carefully and pipetted into a fresh 50 mL tube. The PMBC fraction was washed once in PBS + 5% TNC, after which they were centrifuged for 10 min at 300x g. Remaining erythrocytes were then lysed with 45 mL ice-cold lysis buffer (155 mM NH4CL, 10 mM KHCO 3 , 0.1 mM EDTA, pH7.4 in Milli-Q (Gibco, A1283-01) for 15 min at ice, with a 10 min 300 x g centrifuge step after lysis. Cells were then resuspended in RT HEPES+ (20 mM HEPES, 132 mM NaCl, 6 mM KCL, 1 mM CaCl 2 , 1 mM MgSO 4 , 1.2 mM K2HPO4, 5 mM glucose (All Sigma-Aldrich), and 0.4% (w/v) human serum albumin (Sanquin Reagents), pH7.4) and total cell numbers were counted.

To isolate CD4+ T cells from the PBMC fraction, a negative selection separation kit (Miltenyi Biotec, 130-096-533) was used according to manufacturer’s instructions. After isolation, CD4+ T cells were resuspended 2 mil/mL in RPMI 1640 medium (Gibco, 11875093) with 100 U/mL penicillin and streptomycin (P/S) and were kept in a 12 well plate overnight at 37 °C in 5% CO2.

The protocol for the leukocyte flow assays is based on a previous study 90 . 30000 HUVECs, expressing either APJ-GFP or tdTomato, per lane were seeded into a FN-coated Ibidi μ-slides VI0.4 (Ibidi, Munich, Germany) and grown for 72 hours. Medium was refresh twice daily. 24 hours before the experiment, 1/1000 A13, or DMSO was added to the HUVECs. 4 hours before the experiment, 10 ng/mL LPS (Sigma, L2880) was added to all HUVEC to mimic inflammation. A13 and DMSO were refreshed also.

CD4 + T cells were resuspended 10 6 /mL in 37°C HEPES+ and labelled using VybrantTM DiD Cell-labeling solution (1/6000) for 20 minutes. The ibidi flow slides were connected to a perfusion system and underwent shear flow of 0.8 dyn/cm 2 . Flow was turned on 3 minutes before 700.000 CD4+ T cells were injected upstream of the ibidi flow chamber. CD4 + T cell TEM (transendothelial migration) dynamics were recorded using an Axiovert 200 M widefield microscope, equipped with a 10x NA 0.30 DIC Air objective (Zeiss). Fluorescent excitation was induced by a HXP 120 X light source (100% intensity). For transmitting light, a TL Halogen Lamp at 6.06 V. An AxioCam Icc 3 (Zeiss) camera was used for detection. In the phase-contrast channel, an exposure of 32 ms was used. In the DiD/Far-red channel, an exposure of 3000 ms with an 625-655 excitation filter, a 660 beam splitter and a 655-715 emission filter were used.

To analyse CD4 + T cell crawling dynamics and preferred route of diapedesis, images were taken approximately every 5 seconds for 20 minutes at 3 positions in the middle of the ibidi lane. Immediately after time-lapse acquisition, tile-scans were performed to later quantify total adhesion and transmigration efficacy. Images were taken using Zen Blue software from Zeiss. The tile-scan was stitched in this software too, based on the phase-contrast image.

Flow assay analysis

All analyses were performed in Imaris version (10.0.0) and were based on earlier publications 90 . Total adhesion and diapedesis efficacy were measured by performed a spot analysis on the tile-scans. Spots were detected based on DiD signal, with an estimated diameter of 8 μm. Spots were manually thresholded based on Imaris’ quality parameter. Transmigrated and non-transmigrated CD4+ T cells were distinguished based on their intensity in the phase contrast channel, where non-transmigrated CD4+ T cells were white and transmigrated CD4+ T cells black. Total adhesion was calculated as # non-transmigrated CD4+ T cells + transmigrated CD4+ T cells. CD4+ T cell diapedesis efficacy was quantified as (# transmigrated CD4+ T cells) / (# non-transmigrated CD4+ T cells + transmigrated CD4+ T cells) * 100%.

To quantify crawling dynamics, the same spot analysis was performed on CD4+ T cells during the time-lapses. CD4+ T cells were detected up until their moment of diapedesis, again based on their intensity in the phase-contrast channel. An auto-regressive motion tracking algorithm in Imaris was added to connect all detected spots in the video, allowing a maximum distance of 20 μm between frames. A gap size (a frame without spot detected) of 1 was allowed in this algorithm. All tracks with less than 4 spots were filtered out to ensure only proper crawling tracks were measured. Crawling speed, duration and length were extracted from these data.

In the time-lapses, we manually counted for each diapedesis event if it took place at a cell-cell junction (paracellular diapedesis), or through the cell body (transcellular diapedesis). Data were displayed as percentages for each donor.

Serum sample collection and ELISA analysis from MS patients

A total of 108 serum samples were collected from 36 male and 72 female participants (Supplementary Data  1 ). Samples were divided into three groups based on health status: healthy donors, multiple sclerosis (MS) relapse, and MS remission. The study was approved by the local ethics committee Ärztekammer Westfalen-Lippe under the approval numbers 2016-053-f-S and 2010-262-f-S. All patients provided written consent. The human A13 level in the collected serum samples was examined using the CUSABIO ELISA Kit (CSB-E13072h) according to the manufacturer’s instructions. Sample selection and serum analysis were done independently. Health status information was unblinded after ELISA assays.

Reproducibility was ensured by analyzing 3 independent samples or by conducting 3 independent experiments, unless indicated otherwise. No statistical method was used to predetermine sample size, and no animals were excluded from the analysis. All quantifications were performed using GraphPad Prism (GraphStats Technologies, version 9). For disease incidence and score, a two-way ANOVA with Geisser-Greenhouse correction was applied. Other quantifications were analyzed with a two-sided Student’s t -test. Differential expression testing for DEG analysis was performed based on the two-sided Negative Binomial model for bulk RNA-seq (Figs.  2 c and 6b) and on the two-sided non-parametric Wilcoxon rank sum test for scRNA-seq (Fig.  3f , Supplementary Fig.  10c , 11c , 12a ). DEGs were visualized with volcano plots using the EnhancedVolcano R package (version 1.10.0).

Data availability

The scRNA-seq data generated in this study have been deposited in the gene expression omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/ ) under the accession number GSE230551 ( https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE230551 ). All other relevant data supporting the key findings of this study are available within the article and its Supplementary Information files.  Source data are provided with this paper.

Klineova, S. & Lublin, F. D. Clinical course of multiple sclerosis. Cold Spring Harb. Perspect. Med. 8 https://doi.org/10.1101/cshperspect.a028928 (2018).

Noseworthy, J. H., Lucchinetti, C., Rodriguez, M. & Weinshenker, B. G. Multiple sclerosis. N. Engl. J. Med 343 , 938–952 (2000).

Article   CAS   PubMed   Google Scholar  

Baxter, A. G. The origin and application of experimental autoimmune encephalomyelitis. Nat. Rev. Immunol. 7 , 904–912 (2007).

Ransohoff, R. M. Animal models of multiple sclerosis: the good, the bad and the bottom line. Nat. Neurosci. 15 , 1074–1077 (2012).

Article   CAS   PubMed   PubMed Central   Google Scholar  

van Zwam, M. et al. Brain antigens in functionally distinct antigen-presenting cell populations in cervical lymph nodes in MS and EAE. J. Mol. Med. 87 , 273–286 (2009).

Article   PubMed   Google Scholar  

Furtado, G. C. et al. Swift entry of myelin-specific T lymphocytes into the central nervous system in spontaneous autoimmune encephalomyelitis. J. Immunol. 181 , 4648–4655 (2008).

Tischner, D. et al. Polyclonal expansion of regulatory T cells interferes with effector cell migration in a model of multiple sclerosis. Brain 129 , 2635–2647 (2006).

Odoardi, F. et al. T cells become licensed in the lung to enter the central nervous system. Nature 488 , 675–679 (2012).

Article   ADS   CAS   PubMed   Google Scholar  

Duc, D. et al. Disrupting Myelin-Specific Th17 Cell Gut Homing Confers Protection in an Adoptive Transfer Experimental Autoimmune Encephalomyelitis. Cell Rep. 29 , 378–390 e374 (2019).

Article   MathSciNet   CAS   PubMed   Google Scholar  

Yednock, T. A. et al. Prevention of experimental autoimmune encephalomyelitis by antibodies against alpha 4 beta 1 integrin. Nature 356 , 63–66 (1992).

Kerfoot, S. M. & Kubes, P. Overlapping roles of P-selectin and alpha 4 integrin to recruit leukocytes to the central nervous system in experimental autoimmune encephalomyelitis. J. Immunol. 169 , 1000–1006 (2002).

Vestweber, D. How leukocytes cross the vascular endothelium. Nat. Rev. Immunol. 15 , 692–704 (2015).

Shin, K. et al. Proapelin is processed extracellularly in a cell line-dependent manner with clear modulation by proprotein convertases. Amino Acids 51 , 395–405 (2019).

Kuba, K., Sato, T., Imai, Y. & Yamaguchi, T. Apelin and Elabela/Toddler; double ligands for APJ/Apelin receptor in heart development, physiology, and pathology. Peptides 111 , 62–70 (2019).

Antushevich, H. & Wojcik, M. Review: Apelin in disease. Clin. Chim. Acta 483 , 241–248 (2018).

Wang, X. et al. Apelin/APJ system in inflammation. Int Immunopharmacol. 109 , 108822 (2022).

Vijitha, N. & Engel, D. R. Remote control of Th 17 responses: The lung-CNS axis during EAE. J. Leukoc. Biol. 105 , 827–828 (2019).

Kanayama, M., Danzaki, K., He, Y. W. & Shinohara, M. L. Lung inflammation stalls Th17-cell migration en route to the central nervous system during the development of experimental autoimmune encephalomyelitis. Int Immunol. 28 , 463–469 (2016).

Hosang, L. et al. The lung microbiome regulates brain autoimmunity. Nature 603 , 138–144 (2022).

Dietrich, J. B. The adhesion molecule ICAM-1 and its regulation in relation with the blood-brain barrier. J. Neuroimmunol. 128 , 58–68 (2002).

McGinley, A. M., Edwards, S. C., Raverdeau, M. & Mills, K. H. G. Th17cells, gammadelta T cells and their interplay in EAE and multiple sclerosis. J Autoimmun https://doi.org/10.1016/j.jaut.2018.01.001 (2018).

Goverman, J. Autoimmune T cell responses in the central nervous system. Nat. Rev. Immunol. 9 , 393–407 (2009).

Franzen, O., Gan, L. M. & Bjorkegren, J. L. M. PanglaoDB: a web server for exploration of mouse and human single-cell RNA sequencing data. Database 2019 https://doi.org/10.1093/database/baz046 (2019).

Kalucka, J. et al. Single-Cell Transcriptome Atlas of Murine Endothelial Cells. Cell 180 , 764–779.e720 (2020).

Jeong, H. W. et al. Single-cell transcriptomics reveals functionally specialized vascular endothelium in brain. Elife 11 https://doi.org/10.7554/eLife.57520 (2022).

Kayama, H. & Takeda, K. Regulation of intestinal homeostasis by innate and adaptive immunity. Int Immunol. 24 , 673–680 (2012).

Preziosi, G. et al. Gut dysfunction in patients with multiple sclerosis and the role of spinal cord involvement in the disease. Eur. J. Gastroenterol. Hepatol. 25 , 1044–1050 (2013).

Article   PubMed   PubMed Central   Google Scholar  

Haghikia, A. et al. Dietary Fatty Acids Directly Impact Central Nervous System Autoimmunity via the Small Intestine. Immunity 43 , 817–829 (2015).

Zhang, H. et al. Apelin-13 Administration Protects Against LPS-Induced Acute Lung Injury by Inhibiting NF-kappaB Pathway and NLRP3 Inflammasome Activation. Cell Physiol. Biochem 49 , 1918–1932 (2018).

He, Q. et al. Apelin36 protects against lipopolysaccharideinduced acute lung injury by inhibiting the ASK1/MAPK signaling pathway. Mol Med Rep 23 https://doi.org/10.3892/mmr.2020.11644 (2021).

Yuan, Y. et al. Apelin-13 Attenuates Lipopolysaccharide-Induced Inflammatory Responses and Acute Lung Injury by Regulating PFKFB3-Driven Glycolysis Induced by NOX4-Dependent ROS. J. Inflamm. Res 15 , 2121–2139 (2022).

Yang, P. et al. Pyr(1)]Apelin-13(1-12) Is a Biologically Active ACE2 Metabolite of the Endogenous Cardiovascular Peptide [Pyr(1)]Apelin-13. Front Neurosci. 11 , 92 (2017).

Bui, T. M., Wiesolek, H. L. & Sumagin, R. ICAM-1: A master regulator of cellular responses in inflammation, injury resolution, and tumorigenesis. J. Leukoc. Biol. 108 , 787–799 (2020).

Tsuboi, H., Ando, J., Korenaga, R., Takada, Y. & Kamiya, A. Flow stimulates ICAM-1 expression time and shear stress dependently in cultured human endothelial cells. Biochem Biophys. Res Commun. 206 , 988–996 (1995).

Papangeli, I. et al. MicroRNA 139-5p coordinates APLNR-CXCR4 crosstalk during vascular maturation. Nat. Commun. 7 , 11268 (2016).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Ward, A. O. et al. NF-kappaB inhibition prevents acute shear stress-induced inflammation in the saphenous vein graft endothelium. Sci. Rep. 10 , 15133 (2020).

Cunningham, K. S. & Gotlieb, A. I. The role of shear stress in the pathogenesis of atherosclerosis. Lab Invest 85 , 9–23 (2005).

Gillich, A. et al. Capillary cell-type specialization in the alveolus. Nature 586 , 785–789 (2020).

Niethamer, T. K. et al. Defining the role of pulmonary endothelial cell heterogeneity in the response to acute lung injury. Elife 9 , e53072 (2020).

Vila Ellis, L. et al. Epithelial Vegfa Specifies a Distinct Endothelial Population in the Mouse Lung. Dev. Cell 52 , 617–630.e616 (2020).

Liberzon, A. et al. The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst. 1 , 417–425 (2015).

Ajami, B., Bennett, J. L., Krieger, C., McNagny, K. M. & Rossi, F. M. Infiltrating monocytes trigger EAE progression, but do not contribute to the resident microglia pool. Nat. Neurosci. 14 , 1142–1149 (2011).

Fletcher, J. M., Lalor, S. J., Sweeney, C. M., Tubridy, N. & Mills, K. H. T cells in multiple sclerosis and experimental autoimmune encephalomyelitis. Clin. Exp. Immunol. 162 , 1–11 (2010).

Agata, Y. et al. Expression of the PD-1 antigen on the surface of stimulated mouse T and B lymphocytes. Int Immunol. 8 , 765–772 (1996).

Castro, F., Cardoso, A. P., Goncalves, R. M., Serre, K. & Oliveira, M. J. Interferon-Gamma at the Crossroads of Tumor Immune Surveillance or Evasion. Front Immunol. 9 , 847 (2018).

Jeong, H. W. et al. Single cell transcriptomics reveals functionally specialized vascular endothelium in brain. Elife 11 , e57520 (2022).

Moyron-Quiroz, J. E. et al. Role of inducible bronchus associated lymphoid tissue (iBALT) in respiratory immunity. Nat. Med 10 , 927–934 (2004).

Rangel-Moreno, J. et al. The development of inducible bronchus-associated lymphoid tissue depends on IL-17. Nat. Immunol. 12 , 639–646 (2011).

Foo, S. Y. & Phipps, S. Regulation of inducible BALT formation and contribution to immunity and pathology. Mucosal. Immunol. 3 , 537–544 (2010).

Tan, H. X. et al. Inducible Bronchus-Associated Lymphoid Tissues (iBALT) Serve as Sites of B Cell Selection and Maturation Following Influenza Infection in Mice. Front. Immunol. 10 , 611 (2019).

Halle, S. et al. Induced bronchus-associated lymphoid tissue serves as a general priming site for T cells and is maintained by dendritic cells. J. Exp. Med. 206 , 2593–2601 (2009).

Mrass, P. et al. ROCK regulates the intermittent mode of interstitial T cell migration in inflamed lungs. Nat. Commun. 8 , 1010 (2017).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Yang, J. Q., Leitges, M., Duran, A., Diaz-Meco, M. T. & Moscat, J. Loss of PKC lambda/iota impairs Th2 establishment and allergic airway inflammation in vivo. Proc. Natl Acad. Sci. USA 106 , 1099–1104 (2009).

Tibaldi, E. V., Salgia, R. & Reinherz, E. L. CD2 molecules redistribute to the uropod during T cell scanning: implications for cellular activation and immune surveillance. Proc. Natl Acad. Sci. USA 99 , 7582–7587 (2002).

Jin, Y. et al. Tyrosine-protein kinase Yes controls endothelial junctional plasticity and barrier integrity by regulating VE-cadherin phosphorylation and endocytosis. Nat. Cardiovasc. Res. 1 , 1156–1173 (2022).

Wessel, F. et al. Leukocyte extravasation and vascular permeability are each controlled in vivo by different tyrosine residues of VE-cadherin. Nat. Immunol. 15 , 223–230 (2014).

Folino, A. et al. Apelin-induced cardioprotection against ischaemia/reperfusion injury: roles of epidermal growth factor and Src. Acta Physiol. 222 https://doi.org/10.1111/apha.12924 (2018).

Huang, T. et al. Effect and mechanism of apelin on lipopolysaccharide induced acute pulmonary vascular endothelial barrier dysfunction. Sci. Rep. 13 , 1560 (2023).

Rajagopal, S. & Shenoy, S. K. GPCR desensitization: Acute and prolonged phases. Cell Signal 41 , 9–16 (2018).

Pope, G. R., Tilve, S., McArdle, C. A., Lolait, S. J. & O’Carroll, A. M. Agonist-induced internalization and desensitization of the apelin receptor. Mol. Cell Endocrinol. 437 , 108–119 (2016).

Wang, Y. et al. Ephrin-B2 controls VEGF-induced angiogenesis and lymphangiogenesis. Nature 465 , 483–486 (2010).

Lathen, C. et al. ERG-APLNR axis controls pulmonary venule endothelial proliferation in pulmonary veno-occlusive disease. Circulation 130 , 1179–1191 (2014).

Petrovich, E. et al. Lung ICAM-1 and ICAM-2 support spontaneous intravascular effector lymphocyte entrapment but are not required for neutrophil entrapment or emigration inside endotoxin-inflamed lungs. FASEB J. 30 , 1767–1778 (2016).

Chong, D. L. W. et al. ICAM-1 and ICAM-2 Are Differentially Expressed and Up-Regulated on Inflamed Pulmonary Epithelium, but Neither ICAM-2 nor LFA-1: ICAM-1 Are Required for Neutrophil Migration Into the Airways In Vivo. Front Immunol. 12 , 691957 (2021).

Alpua, M., Turkel, Y., Dag, E. & Kisa, U. Apelin-13: A Promising Biomarker for Multiple Sclerosis? Ann. Indian Acad. Neurol. 21 , 126–129 (2018).

Harbo, H. F., Gold, R. & Tintore, M. Sex and gender issues in multiple sclerosis. Ther. Adv. Neurol. Disord. 6 , 237–248 (2013).

Constant, S. L. et al. Resident lung antigen-presenting cells have the capacity to promote Th2 T cell differentiation in situ. J. Clin. Invest. 110 , 1441–1448 (2002).

Chen, Y., Hancock, W. W., Marks, R., Gonnella, P. & Weiner, H. L. Mechanisms of recovery from experimental autoimmune encephalomyelitis: T cell deletion and immune deviation in myelin basic protein T cell receptor transgenic mice. J. Neuroimmunol. 82 , 149–159 (1998).

Willenborg, D. O. Experimental allergic encephalomyelitis in the Lewis rat: studies on the mechanism of recovery from disease and acquired resistance to reinduction. J. Immunol. 123 , 1145–1150 (1979).

Lassmann, H. & Bradl, M. Multiple sclerosis: experimental models and reality. Acta Neuropathol. 133 , 223–244 (2017).

Wang, H. et al. The Apelin-APJ axis alleviates LPS-induced pulmonary fibrosis and endothelial mesenchymal transformation in mice by promoting Angiotensin-Converting Enzyme 2. Cell Signal 98 , 110418 (2022).

Kong, X. et al. Apelin-13-Mediated AMPK ameliorates endothelial barrier dysfunction in acute lung injury mice via improvement of mitochondrial function and autophagy. Int Immunopharmacol. 101 , 108230 (2021).

Luo, X., Liu, J., Zhou, H. & Chen, L. Apelin/APJ system: A critical regulator of vascular smooth muscle cell. J. Cell Physiol. 233 , 5180–5188 (2018).

Nagano, K., Ishida, J., Unno, M., Matsukura, T. & Fukamizu, A. Apelin elevates blood pressure in ICR mice with L‑NAME‑induced endothelial dysfunction. Mol. Med Rep. 7 , 1371–1375 (2013).

Rasooli Tehrani, A. et al. Plasma levels of CTRP-3, CTRP-9 and apelin in women with multiple sclerosis. J. Neuroimmunol. 333 , 576968 (2019).

Bevilacqua, M. P. & Nelson, R. M. Selectins. J. Clin. Invest. 91 , 379–387 (1993).

Weber, C. & Springer, T. A. Interaction of very late antigen-4 with VCAM-1 supports transendothelial chemotaxis of monocytes by facilitating lateral migration. J. Immunol. 161 , 6825–6834 (1998).

Ronald, J. A., Ionescu, C. V., Rogers, K. A. & Sandig, M. Differential regulation of transendothelial migration of THP-1 cells by ICAM-1/LFA-1 and VCAM-1/VLA-4. J. Leukoc. Biol. 70 , 601–609 (2001).

van Wetering, S. et al. VCAM-1-mediated Rac signaling controls endothelial cell-cell contacts and leukocyte transmigration. Am. J. Physiol. Cell Physiol. 285 , C343–C352 (2003).

Tatin, F. et al. Apelin modulates pathological remodeling of lymphatic endothelium after myocardial infarction. JCI Insight . 2 , e93887 (2017).

Kim, J. D. et al. Essential role of Apelin signaling during lymphatic development in zebrafish. Arterioscler Thromb. Vasc. Biol. 34 , 338–345 (2014).

Berta, J. et al. Apelin promotes blood and lymph vessel formation and the growth of melanoma lung metastasis. Sci. Rep. 11 , 5798 (2021).

Diehl, R. et al. Immunosuppression for in vivo research: state-of-the-art protocols and experimental approaches. Cell Mol. Immunol. 14 , 146–179 (2017).

Winkelmann, A., Loebermann, M., Reisinger, E. C., Hartung, H. P. & Zettl, U. K. Disease-modifying therapies and infectious risks in multiple sclerosis. Nat. Rev. Neurol. 12 , 217–233 (2016).

Chen, H. I. et al. VEGF-C and aortic cardiomyocytes guide coronary artery stem development. J. Clin. Invest. 124 , 4899–4914 (2014).

Tian, X. et al. Subepicardial endothelial cells invade the embryonic ventricle wall to form coronary arteries. Cell Res. 23 , 1075–1090 (2013).

Muzumdar, M. D., Tasic, B., Miyamichi, K., Li, L. & Luo, L. A global double-fluorescent Cre reporter mouse. Genesis 45 , 593–605 (2007).

Sixt, M. et al. Endothelial cell laminin isoforms, laminins 8 and 10, play decisive roles in T cell recruitment across the blood-brain barrier in experimental autoimmune encephalomyelitis. J. Cell Biol. 153 , 933–946 (2001).

Kassis, I. et al. Neuroprotection and immunomodulation with mesenchymal stem cells in chronic experimental autoimmune encephalomyelitis. Arch. Neurol. 65 , 753–761 (2008).

Gronloh, M. L. B. et al. Endothelial transmigration hotspots limit vascular leakage through heterogeneous expression of ICAM-1. EMBO Rep. 24 , e55483 (2023).

Download references

Acknowledgements

We thank Martin Stehling (Max Planck Institute for Molecular Miomedicine, Germany) for cell sorting, our animal facility for excellent animal care, Eva Maria Schumann (Biobank management of Universitätsklinikum Münster, Germany) for serum preparation, Kristy Red-Horse (Stanford University) for Aplnr-CreERT2 mice, and Prof. Gou-Young Koh (Institute for Basic Science, Korea) for fruitful scientific discussions. This study was supported by the Max Planck Society, the University of Münster, the DFG (CRC 1366 and CRC 1009), and the Leducq Foundation.

Open Access funding enabled and organized by Projekt DEAL.

Author information

Authors and affiliations.

Max Planck Institute for Molecular Biomedicine, Department of Tissue Morphogenesis, Münster, Germany

Hongryeol Park, Hyun-Woo Jeong, Bong Ihn Koh, Esther Bovay & Ralf H. Adams

Institute of Physiological Chemistry and Pathobiochemistry and Cells-in-Motion Interfaculty Centre (CIMIC), University of Münster, Münster, Germany

Jian Song & Lydia Sorokin

Vascular Cell Biology Lab, Department of Medical Biochemistry, Amsterdam UMC, and Section Molecular Cytology at Swammerdam Institute for Life Sciences, Leeuwenhoek Centre for Advanced Microscopy, University of Amsterdam, Amsterdam, The Netherlands

Max L. B. Grönloh & Jaap D. van Buul

Department of Medical Life Sciences, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

Kee-Pyo Kim

Department of Neurology, University of Münster, Münster, Germany

Luisa Klotz

Division of Cardiothoracic Surgery, University of California, San Diego, CA, USA

Patricia A. Thistlethwaite

You can also search for this author in PubMed   Google Scholar

Contributions

H.P., L.S., L.K., J.D.vB., and R.H.A. designed experiments and interpreted results. H.P., J.S., M.L.B.G., B.I.K., E.B., and K.P.K. conducted all experiments including animal models, cell culture, imaging, quantifications, and analysis of human serum samples. H.P., B.I.K. and H.W.J. generated and analysed the scRNA-sequencing data. P.AK. generated critical genetic tools. H.P. and R.H.A. wrote the manuscript.

Corresponding authors

Correspondence to Hongryeol Park or Ralf H. Adams .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

Nature Communications thanks Francis Luscinskas and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, peer review file, description of additional supplementary files, supplementary movie 1, supplementary movie 2, supplementary movie 3, supplementary movie 4, supplementary data 1, source data, source data, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Park, H., Song, J., Jeong, HW. et al. Apelin modulates inflammation and leukocyte recruitment in experimental autoimmune encephalomyelitis. Nat Commun 15 , 6282 (2024). https://doi.org/10.1038/s41467-024-50540-5

Download citation

Received : 21 July 2023

Accepted : 08 July 2024

Published : 25 July 2024

DOI : https://doi.org/10.1038/s41467-024-50540-5

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Experimental Autoimmune Encephalomyelitis

• First Online: 30 April 2021

Cite this protocol

• Clara Ballerini 5  

Part of the book series: Methods in Molecular Biology ((MIMB,volume 2285))

2569 Accesses

8 Citations

7 Altmetric

Experimental autoimmune encephalomyelitis, originally experimental allergic encephalomyelitis, is the well-known animal model of multiple sclerosis, an immune- mediated, demyelinating, inflammatory chronic disease of the central nervous system. The experimental disease is widely utilized to test new therapies in preclinical studies, to investigate new hypothesis on the possible pathogenic mechanisms of autoimmune reaction directed against the central nervous system or more generally to investigate the interactions between the immune system and the central nervous system that lead to neuroinflammation. The experimental autoimmune encephalomyelitis may be induced following different protocols in mammals, including nonhuman primates, and autoreactive CD4+ T-lymphocytes directed against myelin antigens are the main factors. Here, after introducing the model, we describe the protocol to induce active EAE in inbred mice, we report on a table the different clinical courses of EAE depending on the combination of antigen /mouse strain and we provide indications on how to evaluate the clinics and pathology of this induced disease.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime
• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Active Induction of Experimental Autoimmune Encephalomyelitis in C57BL/6 Mice

Experimental Autoimmune Encephalomyelitis in Mice

Active Induction of Experimental Autoimmune Encephalomyelitis (EAE) with MOG35–55 in the Mouse

Van Epps VL (2005) Thomas River and the EAE model. J Exp Med 202(1):4. https://doi.org/10.1084/jem.2021fta

Article   CAS   PubMed Central   Google Scholar  

Procaccini C, De Rosa V, Pucino V et al (2015) Animal models of multiple sclerosis. Eur J Pharmacol 759:182–191. https://doi.org/10.1016/j.ejphar.2015.03.042

Article   CAS   PubMed   PubMed Central   Google Scholar  

Costantinescu CS, Farooqi N, O’Briejn K et al (2011) Experimental autoimmune encephalomyelitis (EAE) as a model of multiple sclerosis (MS). Br J Pharmacol 164:1079–1106. https://doi.org/10.1111/j.1476-5381.2011.01302.x

Article   CAS   Google Scholar  

Kuchroo VK, Anderson AC, Waldner H et al (2002) T-cell response in experimental autoimmune encephalomyelitis (EAE): role of self and cross-reactive antigens in shaping, tuning and regulating the autopathogenic T-cell repertoire. Annu Rev Immunol 20:101–123. https://doi.org/10.1146/annurev.immunol.20.081701.141316

Article   CAS   PubMed   Google Scholar  

Gran B, Zhang GX, Yu S et al (2002) IL-12p35-deficient mice are susceptible to experimental autoimmune encephalomyelitis: evidence for redundancy in the IL-12 system in the induction of central nervous system demyelination. J Immunol 169:7104–7110. https://doi.org/10.4049/jimmunol.169.12.7104

Langrish CL, Chen Y, Blumenschein WM et al (2005) IL-23 drives a pathogenic T- cell population that induces autoimmune inflammation. J Exp Med 201:233–240. https://doi.org/10.1084/jem.20041257

Paterson PY (1960) Transfer of allergic encephalomyelitis in rats by means of lymph node cells. J Exp Med 111:119–135. https://doi.org/10.1084/jem.111.1.119

Furlan R, Cuomo C, Martino G (2009) Animal models of multiple sclerosis. Methods Mol Biol 549:157–173

Bjielobaba I, Begovic-Kupresanin V, Pekovic S et al (2018) Animal models of multiple sclerosis: focus on experimental autoimmune encephalomyelitis. J Neurosci Res 96:1021–1042. https://doi.org/10.1002/jnr.24224

Fritz RB, Chou CH, McFarlin DE (1983) Relapsing murine experimental allergic encephalomyelitis induced by myelin basic protein. J Immunol 130:1024–1026

CAS   PubMed   Google Scholar  

Amor S, Groome N, Linington C et al (1994) Identification of epitopes of myelin oligodendrocyte glycoprotein for the induction of experimental allergic encephalomyelitis in SJL and Biozzi AB/H mice. J Immunol 153:4349–4356

Oliver AR, Lyon GM, Ruddle NH (2003) Rat and human myelin oligodendrocyte glycoproteins induce experimental autoimmune encephalomyelitis by different mechanisms in C57BL/6 mice. J Immunol 17:462–468. https://doi.org/10.4049/jimmunol.171.1.462

Article   Google Scholar  

Sundvall M, Jirholt J, Yang HT et al (1995) Identification of murine loci associated with susceptibility to chronic experimental autoimmune encephalomyelitis. Nat Genet 10:313–317. https://doi.org/10.1038/ng0795-313

Kipp M, van der Star B, Vogel DYS et al (2012) Experimental in vivo and in vitro models of multiple sclerosis: EAE and beyond. Mult Scler Relat Disord 1:15–28. https://doi.org/10.1016/j.msard.2011.09.002

Article   PubMed   Google Scholar  

Giuliani F, Fu SA, Metz LM et al (2005) Effective combination of minocycline and interferon beta in a model of multiple sclerosis. J Neuroimmunol 175:83–91. https://doi.org/10.1016/j.jneuroim.2005.04.020

Download references

Author information

Authors and affiliations.

Laboratory of Neuroimmunology, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy

Clara Ballerini

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Clara Ballerini .

Editor information

Editors and affiliations.

Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy

Francesco Annunziato

Laura Maggi

Alessio Mazzoni

Rights and permissions

Reprints and permissions

Copyright information

© 2021 Springer Science+Business Media, LLC, part of Springer Nature

About this protocol

Ballerini, C. (2021). Experimental Autoimmune Encephalomyelitis. In: Annunziato, F., Maggi, L., Mazzoni, A. (eds) T-Helper Cells. Methods in Molecular Biology, vol 2285. Humana, New York, NY. https://doi.org/10.1007/978-1-0716-1311-5_27

Download citation

DOI : https://doi.org/10.1007/978-1-0716-1311-5_27

Published : 30 April 2021

Publisher Name : Humana, New York, NY

Print ISBN : 978-1-0716-1310-8

Online ISBN : 978-1-0716-1311-5

eBook Packages : Springer Protocols

• Publish with us

Policies and ethics

• Find a journal
• Track your research

• Vision, Mission and Purpose
• Focus & Scope
• Editorial Info
• Open Access Policy
• Editing Services
• Article Outreach
• Why with us
• Focused Topics
• Manuscript Guidelines
• Ethics & Disclosures
• What happens next to your Submission
• Submission Link
• Special Issues
• Latest Articles
• All Articles
• Article Details

Experimental Autoimmune Encephalomyelitis Animal Models Induced by Different Myelin Antigens Exhibit Differential Pharmacologic Responses to Anti-Inflammatory Drugs

Yuxi Yan 1# , Quan Zhao 1# , Ya Huang 1 , Janine Y. Yang 2 , Jie Zou 1 , Chunxia Ao 1 , Xiaojuan Chai 1 , Renhong Tang 1 and WenQing Yang 1* 1 State Key Laboratory of Translational Medicine and Innovative Drug Development, Jiangsu Simcere Pharmaceutical Co., Ltd., Nanjing, Jiangsu, China. 2 Massachusetts Eye and Ear, Harvard Medical School, USA

Background and objective

Experimental autoimmune encephalomyelitis (EAE) is the most commonly used model for studying autoimmune-mediated myelin degradation in multiple sclerosis (MS). Here, we evaluated the pharmacologic responses of several anti-inflammatory drugs with varying mechanisms of actions (MOAs) using EAE models induced by different MOG immunogens to reveal differential pharmacologic characteristics of the disease models and provide a general guidance in animal model selection for MS research.

The pharmacologic responses of anti-inflammatory drugs with different mechanisms of actions (MOAs) were evaluated using EAE models induced by either myelin oligodendrocyte glycoprotein p35-55 (MOG 35-55 )  or p1-128 (MOG 1-128 ). EAE animal models were developed in mice with C57BL/6 background. The animals were treated with different anti-MS medications, including 3 B cell-mediated agents and 2 T cell-mediated agents, respectively. Clinical symptoms were monitored and scored, and pharmacodynamic markers including cytokine secretion, inflammatory cell infiltration, and demyelination in spinal cord were analyzed.

In MOG 35-55 peptide-induced EAE model, T cell modulating agents Secukinumab and Fingolimod significantly alleviated clinical symptoms, while B cell-depleting agents, BTK inhibitors PRN2246 and Telitacicept, displayed minimal therapeutic effects or even exacerbated disease progression. In contrast, both T cell-modulating agents and B cell-depleting agents ameliorated disease severity in MOG 1-128 -induced EAE model. T cell and B cell infiltration in spinal cord increased with disease progression in MOG 1-128 -induced EAE model.

Conclusions

Our results demonstrated that induction of EAE by different myelin antigens resulted in differential pharmacologic responses to drugs with specific MOAs. The MOG 35-55 peptide-induced EAE model only responded to T cell-modulating drugs, whereas the MOG 1-128 protein-induced EAE model exhibited therapeutic sensitivity to both T cell- and B cell-modulating agents. These data suggest the MOG 35-55 peptide-induced EAE model is suitable for assessing T cell-modulating agents while MOG 1-128 protein-induced model can be employed to evaluate both T cell- and B cell-modulating agents.

Introduction

Multiple sclerosis (MS) is an autoimmune and neurological disease characterized by demyelination of the central nervous system (CNS). Due to the long course of the disease and high disability rate, MS is a huge socio-economic burden 1 . Although a few agents, including kinase inhibitors and monoclonal antibodies, have shown promising results in the clinic, MS remains one of the most common causes of neurological dysfunction, highlighting the need to develop novel therapies for MS patients 2 .

The experimental allergic encephalomyelitis (EAE) mouse model has been widely used in studying the mechanisms of autoimmune-mediated myelin degradation and testing new therapies for MS 3 . Two commonly used animal models of myelin oligodendrocyte glycoprotein (MOG) antigen-induced EAE include MOG 35-55 peptide- and MOG 1-125 (or MOG 1–128 ) protein-induced models 4 . Though both MOG 35–55 and MOG 1–128 can induce similar clinical symptoms in mice, there are different immunological mechanisms for disease occurrence and progression between these two models. The short peptide MOG 35–55 is the dominant MHC II class restricted epitope of myelin proteins, and its pathogenic mechanism for EAE is largely mediated by CD4 + T help (Th) cells. The extracellular domain of MOG (MOG 1–128 ) contains conformational epitopes that can be recognized by B cell receptor (BCR) 4-6 . As for MOG 1-128 -induced EAE model, both antigen-activated B cells and antibodies against MOG protein are involved in the pathogenic processes due to the recognition of conformational epitopes on MOG 1-128 protein by BCR.

In pharmacological studies, choosing appropriate animal models based on specific MOAs of test agents is a critical process as this may influence data interpretation, especially for a disease model like EAE where the pathogenesis and pathophysiology are complex 7 . It has been suggested that both T cell- and B cell-mediated inflammatory responses contribute to the initiation and progression of EAE 8 . However, to our knowledge, there has not been a direct comparison or validation published to systemically investigate the pharmacological effects of commonly used anti-inflammatory drugs in these models.

Food and Drug Administration (FDA)-approved drugs and agents under investigation for MS include corticosteroids, immune-modulating agents, neuroprotective agents, and agents that help improve neuromuscular/synaptic transmission. Based on MOA, immune-modulating agents can be roughly classified into two subsets, B cell modulating/depleting agents and T cell modulating molecules. B cell-acting drugs include Bruton’s tyrosine kinase inhibitors (BTKi) PRN2246 9 and Evobrutinib 10 , and the human recombinant fusion protein Telitacicept 11 . T cell-modulating agents include Secukinumab (an anti-human interleukin-17 biologic 12 ) and Fingolimod (a sphingosine-1-phosphate receptor modulator 13 ), which sequester lymphocytes in the lymph nodes, preventing them from contributing to an autoimmune reaction.

In the present study, we evaluated the pharmacologic responses of several anti-inflammatory drugs with different MOAs using animal models of EAE induced by MOG 35-55 or MOG 1-128 . The results of this study may shed light on the biology of EAE models and guide EAE animal model selection for novel drug candidate profiling or mechanistic investigations in MS research.

Methods and materials

6- to 8-week-old female C57BL/6 mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. C57BL/6- Il17a em2(hIL17A) Il17f em2(hIL17F)Smoc (hIL17A/hIL17F, Cat No: NM-HU-200281) mice were purchased from Shanghai Model Organisms Center, Inc. Animals were housed in a specific pathogen-free barrier condition at OBiO Technology (Shanghai) Corp., Ltd. Animal welfare and experimental procedures were carried out strictly in accordance with animal care and protocol approved by the Institutional Animal Care and Use Committee (IACUC) at OBiO Technology (Shanghai) Corp., Ltd.

Reagents and proteins

PRN2246, Evobrutinib, and Fingolimod were purchased from MedChemExpress Co., Ltd (Princeton, NJ, USA). Secukinumab and Telitacicept were synthesized by Simcere Pharmaceutical (Nanjing, China). MOG 35-55 peptide (MEVGWYRSPFSRVVHLYRNGK; Cat. No.: T510219) was purchased from Sangon Biotech (Shanghai) Co., Ltd. MOG 1-128 protein was developed by genetically synthesizing the extracellular domain (aa1-128) of mouse MOG based on amino acid sequence from Uniprot database, cloning it into mammalian expression vector with His-tag expressed in Expi293 cells via transient transfection, and purifying using HisTrap excel column (GE).

Model establishment and in vivo efficacy studies with EAE models

For MOG 35-55 peptide-induced EAE model, 6- to 8-week-old female mice were anesthetized with isoflurane and immunized subcutaneously in three sites on the back with a total of 150 μg MOG completely emulsified with equal volume of Incomplete Freund's Adjuvant (Sigma-Aldrich) containing 2.5mg/mL of Mycobacterium tuberculosis H37 Ra (BD DIFCO). This was compared to the immunization agent in MOG 1-128 protein-induced EAE model with 225 μg MOG emulsified with Complete Freund's Adjuvant (Sigma-Aldrich). Simultaneously, the mice were administered 200 ng of pertussis toxin intraperitoneally and again at 48 hours after the immunization. The severity of EAE features on these mice was monitored and scored daily for 28 days according to the following standard: 0, no clinical signs; 1, mild tail paralysis or hind limb weakness; 2, tail paralysis and hind limb weakness; 3, partially hind limb paralysis; 4, complete hind limb paralysis and front limb weakness; and 5, near death or death 3 . Mice were randomly divided into vehicle control or drug treatment groups. For treatment, Fingolimod, PRN2246, and Evobrutinib were administered intragastrically at 0.5, 5, and 10mg/kg/d, respectively. Secukinumab and Telitacicept were diluted in PBS and administered intraperitoneally at 15mpk and 7.5mpk (MOG 35-55 induced EAE) and 10mpk and 5mpk (MOG 1-121 induced EAE) every 3 days from day 0.

Histopathology and immunohistochemistry (IHC)

Mice were anesthetized and sacrificed at experimental endpoints, and spinal cords were collected and fixed in 10% formalin at the end of the experiment. The cervical, thoracic, and lumbar regions of the spinal cord were separated and embedded in paraffin transversely for four-micrometer histological section, and the rehydrated sections were then stained by Luxol fast blue as well as hematoxylin and eosin (H&E) according to previously described protocols 14 . The severity of inflammatory cell infiltration and demyelination in stained spinal cord sections was evaluated in double-blinded manner as follows: 0, no cell infiltration or inflammation; 1, little demyelination with few infiltrated cells around the perivascular and meninges area; 2, less than one third part of the white matter is infiltrated by inflammatory cells, and demyelination appears in a few areas; 3, more than one third part of the white matter is infiltrated by inflammatory cells with large areas of demyelination; and 4, the whole white matter is infiltrated by inflammatory cells 15 . IHC staining of CD4 + T cells and B cells in spinal cords was performed using an automated staining system (BOND-III; Leica Microsystems, Vista, CA) with antibodies against mouse CD4 (Abcam TM , dilution 1:200) and CD45R (eBioscience, dilution 1:300). Specificity of the primary antibodies was cross-validated. Omission of the primary antibodies was also used to test the specificity of the secondary antibodies. Isotype controls and secondary antibody only controls were included in each experiment. Digital images were acquired by Aperio CS2 using ImageScope viewing software, and the infiltration of T-cells and B-cells in spinal cord was quantified by counting positive cells using Aperio Image Analysis Systems.

Cytokine analysis

Whole blood of normal control and MOG 1-128 -induced EAE mice was collected by cardiac puncture and centrifuged at 1000g for five minutes to extract the serum at the endpoint of animal experiment. The serum levels of IgA and IL-17 were quantified by ELISA kits (Bio-techne, R&D systems) according to manufacturer’s instructions. In ELISA assay, serum was diluted by 1:200 and 1:1 for IgA and IL-17 detection, respectively. ODs at 450nm were measured by multimode plate reader (PerkinElmer EVision 2105).

Statistical analysis

Statistical analysis was conducted by GraphPad Prism 7.0 software. Data were expressed as mean ± standard error of mean (SEM). Normality and variance homogeneity of the data were assessed by shapiro-wilk test and kolmogorow smirnov test. Statistical comparisons of EAE clinical scores were conducted using two-way ANOVA, followed by Dunnett’s multiple comparisons test. A P -value of <0.05 was defined as statistically significant.

T cell-modulating drugs and B cell-depleting agents exhibit distinct pharmacologic responses in MOG 35-55 peptide-induced EAE model

MOG 35-55 peptide-induced EAE models in wild-type C57BL/6 mice were utilized to evaluate the pharmacologic activities of BTKi-PRN2246 and S1P1 inhibitor Fingolimod. Clinical evaluation revealed that mice in vehicle control group of MOG 35-55 peptide-induced EAE displayed typical clinical symptoms, such as severe tail paralysis and total limb weakness, while Fingolimod treatment significantly reduced the clinical severity of EAE. PRN2246 exhibited relatively weaker therapeutic effect on EAE mice compared with Fingolimod (Figure 1A). Consistently, histological analysis demonstrated inflammatory cell infiltration and demyelination in the spinal cord of EAE mice (Figure 1B & 1C). In addition, hIL17A/hIL17F mice were immunized with MOG 35-55 peptide to evaluate the pharmacologic activities of Secukinumab and Telitacicept. Secukinumab was found to significantly ameliorate EAE clinical score, possibly through modulation of T cell function by neutralizing IL-17A. However, the clinical symptoms of EAE mouse treated with Telitacicept were significantly aggravated compared with vehicle control (Figure 1D).

Figure 1: Pharmacologic responses and histological findings in MOG35-55-induced EAE mice treated with various anti-inflammatory agents.

(A) Clinical score of EAE mice receiving daily treatment with Fingolimod, PRN2246, or corresponding solvent. Asterisks indicate a significant difference between vehicle and Fingolimod or PRN2246 treatment groups. **p<0.01, ***p<0.005.

(B) Representative figures of H&E & fast blue staining of spinal cord in normal and EAE mice treated with Fingolimod, PRN2246, or corresponding solvent.

(C) Evaluation of inflammatory infiltration score in spinal cord by H&E staining. Graph represents the average score of five mice in each group. Asterisks indicate a significant difference between vehicle and Fingolimod or PRN2246 treatment groups. *p<0.05, ***p<0.005.

(D) Clinical score of EAE mice receiving daily treatment with Secukinumab, Telitacicept, or corresponding solvent. Graph represents the average score of 10 mice in each group. Asterisks indicate the significant difference between vehicle and Fingolimod, Secukinumab or Telitacicept treatment groups. ***p<0.005.

Collectively, all the results above demonstrated that drugs targeting T cells rather than B cells were therapeutically effective in MOG 35-55 -induced EAE model, suggesting that the pathogenesis of this model is dominated by T cells and it is unsuitable for evaluating pharmacological activities of B cell-depleting agents

Both B cell-depleting agents and T cell-modulating drugs show therapeutic response in MOG 1-128 -induced EAE model

In order to elucidate how drugs targeting different lymphocytes play a role during EAE induction by MOG 1-128 protein, mice were independently treated with Evobrutinib and Fingolimod at the beginning of immunization process. As shown in Figure 2A, both Evobrutinib and Fingolimod had significant effects on relieving clinical symptoms of EAE. In addition, serum cytokine levels of EAE mice were also measured at the endpoint of the experiment. ELISA analysis revealed significantly increased IL-17 and IgA levels in vehicle control group compared to normal mice (Figure 2B), suggesting that MOG 1-128 protein may trigger the upregulation of inflammatory factors, including IL-17 and IgA, which may be secreted by activated B cells and T cells.

Because innate and adaptive immune cells are thought to contribute to neuro-axonal injury and demyelination through the secretion of soluble factors during MS relapses [22], the pathogenesis of MOG 1-128 -induced EAE model may reflect real-life clinical situations in MS, while the EAE induction in humanized mice may be optimal for drug testing. To confirm this, we immunized IL-17A humanized mouse with MOG 1-128 protein to induce EAE, and evaluated the pharmacological effects of Secukinumab and Telitacicept. Since human TACI protein is homologous in mice, the IL-17A humanized mouse is also suitable for testing the drug effect of Telitacicept on EAE model. Continuous observation of clinical symptoms revealed that both Secukinumab and Telitacicept reduced the severity of tail paralyses and limb weakness in mice, but Secukinumab showed higher efficacy than equal dose of Telitacicept (Figure 2C). Histological analysis also revealed that the number of infiltrating inflammatory cells around the white matter in spinal cord was significantly decreased after treatment with Secukinumab or Telitacicept (Figure 2D & 2E).

Figure 2: Pharmacologic responses, histological findings, and serum cytokine levels in MOG1-128 -induced EAE mice treated with various anti-inflammatory agents.

(A) Clinical score of EAE mice treated with Fingolimod, Evobrutinib, or corresponding solvent daily. Values are mean ± SEM of 10 mice/group. Asterisks indicate a significant difference between vehicle and Fingolimod or Evobrutinib treatment groups. ***p<0.005.

(B) Protein levels of IL-17 and IgA in serum of EAE mice. Values are mean ± SEM of five mice/group. *p<0.05 between vehicle control versus normal control.

(C) Clinical score of MOG1-128-induced EAE in hIL17A/hIL17F mice treated with Secukinumab, Telitacicept, or corresponding solvent daily. Values are mean ± SEM of 10 mice/group. Asterisks indicate a significant difference between vehicle and Secukinumab or Telitacicept treatment groups. ***p<0.005

(D) Representative figures of H&E staining of spinal cord in normal and EAE mice treated with Secukinumab, Telitacicept, or corresponding solvent.

(E) Evaluation of inflammatory infiltration score in spinal cord by H&E staining. Values are mean ± SEM of 10 mice/group. Asterisks indicate a significant difference between vehicle and dosing groups. *p<0.05, ***p<0.005.

Overall, the results suggest that molecules that suppress both B cell and T cell functions may have a considerable therapeutic effect on MOG protein-immunized mice, and the higher efficacy of Secukinumab and Telitacicept is suggestive of the vital function of Th17 cells and B cells in MOG 1-128 -induced EAE model, which may be used as an ideal model for screening and testing drugs for clinical treatment.

T cell and B cell infiltration in spinal cord increases with disease progression in MOG 1-128 protein-induced EAE model

To further investigate the possible responding mechanism of T cells and B cells in MOG 1-128 protein-induced EAE models, mice were sacrificed at different time points of the pathogenic process and the spinal cords were collected, sectioned, and stained. Hematoxylin & eosin (H&E) staining analysis revealed that only a few inflammatory cells had infiltrated into the gray matter of spinal cord at day 9 after immunization. However, the number of infiltrating cells had significantly increased 13 days post modeling. High-level cell infiltration remained constant until day 22 with the increased presence of inflammatory cells scattered in the central area of the gray matter (Figure 3A & 3B). We also calculated the total counts of T helper cells and B cells in the cervical, thoracic, and lumbar regions of spinal cord sections. As shown in Figure 3C, both CD4 + and CD45R + cells started to accumulate since day 9, peaked at day 13, and remained constant until day 22 post immunization, which was similar to the H&E staining results. Collectively, the data show that T helper cell and B cell infiltration in the spinal cord increased with disease progression in MOG 1-128 protein-induced EAE model, suggesting that both T and B cells contribute to the initiation and progression of this EAE model. Furthermore, T cell and B cell infiltration in the spinal cord from MOG 35-55 -induced EAE model mice was analyzed and we found that there is almost no B cell infiltration at day 9 (data not shown), suggesting B cells are not involved the pathogenic mechanism at the early stage of this model.

Figure 3: T helper cell and B cell infiltration analysis in the spinal cord of MOG1-128-induced EAE model.

(A) Representative figures of H&E and IHC staining at 9, 13, and 22 days after first immunization in the spinal cord of MOG1-128-induced EAE model.

(B) Evaluation of inflammatory infiltration score in spinal cord by H&E staining. Data shown on the graph represent the average score of five mice in each group. ***p<0.005

(C) Statistical IHC analysis of T helper cells (CD4+) and B cells (CD45R+) 9, 13, and 22 days after first immunization in the spinal cord of EAE mice. Values are mean ± SD of 5 mice/group. ***p<0.005.

The immunological and neurobiological mechanisms underlying the pathogenesis, progression, and prognosis of MS are complicated. It is widely accepted that the hallmark of MS is the formation of demyelinating lesions in the brain and spinal cord, which are thought to be caused by infiltration of leukocytes including T cells, B cells, and myeloid cells into the lesion sites 8, 16 .

Animal models are important tools in research and pre-clinical drug development 17 . In this study, we established EAE models via induction with MOG p35-55 or MOG p1-128, and evaluated the pharmacological activities of a series of drugs for autoimmunity disease. These agents with distinct MOAs showed varied results in the two models due to differences in immunological mechanisms. Drugs modulating T cells exhibited therapeutic effects on both MOG 35-55 peptide- and MOG 1-128 protein-induced EAE models. However, agents targeting B cells only responded to MOG 1-128 -induced EAE model.

The MOG 35-55 -EAE C57BL/6 model is widely used due to the accessibility of short peptide and mature modeling methods. Consistent with previous study, our research demonstrated that Fingolimod, which sequesters T cells in lymph nodes, reduced EAE clinical score and inflammation of spinal cord. The serum level of the pro-inflammatory cytokine IL-17, mainly secreted by Th17 (a major subset of CD4 + T cell), was significantly increased in MS patients and EAE model 18 . In IL-17A humanized mouse, blocking the activity of IL-17A by Secukinumab alleviated clinical symptoms induced by MOG 35-55 . All these data suggested MOG 35-55 -induced EAE as an appropriate model for evaluating the pharmacological activities of agents targeting T cells activities. Nevertheless, this type of EAE model may not be a rationale choice when assessing more complex pathogenic immune responses involving both T and B cell recognition of autoantigen, since the antigen recognition of B cells depends on conformational epitopes of whole proteins.

The infiltration of B cells in active MS lesions 19 and the success of B cell-depleting agents in MS therapy 16, 20 confirm that B cells play a significant role in the pathogenesis of MS patients. In order to evaluate the therapeutic effects of agents targeting B cells, we immunized C57BL/6 mouse with the protein MOG 1-128 , which can be recognized by B cells. In this model, both B cell-depleting agents Telitacicept and Evobrutinib ameliorated clinical symptoms by suppressing B cell functions using different mechanism of actions. Notably, treatment with Telitacicept significantly aggravated MOG 35-55 -induced mouse EAE clinical symptoms, opposite to the effects in MOG 1-128 model. The exacerbation of MOG 35-55 -induced EAE may relate to the depletion of antigen-naïve B cells, which exert anti-inflammatory properties 21, 22 . The paradoxical outcomes of Telitacicept in the two EAE models are similar to the results of a previous study on anti-CD20 agents 23 . BTK inhibitors, which may suppress both B cells and myeloid cells 24 , showed positive results in these two models, and the therapeutic effect in MOG 35-55 -induced EAE was mainly due to suppression of myeloid cells.

This study has three major limitations. First, the differences in immune responses to immunogenic MOG protein and peptide were not investigated in depth to clarify the complicated pathogenic mechanisms of these two EAE models. Second, the drugs/agents used in this study were limited and could therefore be biased. These agents may not completely represent the pharmacological mechanisms of therapeutic agents for MS treatment. Third, precautions should be applied when implying the conclusions obtained from this paper to addressing different aspects of questions in MS human patients. MS is a very complex disease and the pathogenesis is not completely known. The data from this paper provide preliminary guidance for animal model selection when addressing very specific mechanistic or pharmacological questions in preclinical studies, with the rodent–human dissimilarities and other potentially impactful factors born in mind. Therefore, more in-depth and systematic studies are needed to characterize the immunological mechanisms of these two paradigms of EAE models.

This study concludes that anti-inflammatory drugs with different MOAs exhibit differential pharmacological responses in EAE models induced by distinct immunogens. The MOG 35-55 peptide-induced EAE model only responds to T cell modulating drugs, whereas the MOG 1-128 protein-induced EAE model exhibits therapeutic sensitivity to both T cell- and B cell-modulating agents. These data indicate that it is critical to understand the pathogenesis of EAE animal models and provide preliminary rationale for model selection in MS research.

• Reich DS, Lucchinetti CF, Calabresi PA. Multiple Sclerosis. N Engl J Med . 2018; 378: 169-180.
• Wei W, Ma D, Li L, et al. Progress in the Application of Drugs for the Treatment of Multiple Sclerosis. Front Pharmacol . 2021; 12: 724718.
• Stromnes IM, Goverman JM. Active induction of experimental allergic encephalomyelitis. Nat Protoc . 2006; 1: 1810-1819.
• Lyons JA, San M, Happ MP, et al. B cells are critical to induction of experimental allergic encephalomyelitis by protein but not by a short encephalitogenic peptide. Eur J Immunol . 1999; 29: 3432-3439.
• Dang AK, Jain RW, Craig HC, et al. B cell recognition of myelin oligodendrocyte glycoprotein autoantigen depends on immunization with protein rather than short peptide, while B cell invasion of the CNS in autoimmunity does not. J Neuroimmunol . 2015; 278: 73-84.
• von Budingen HC, Tanuma N, Villoslada P, et al. Immune responses against the myelin/oligodendrocyte glycoprotein in experimental autoimmune demyelination. J Clin Immunol . 2001; 21: 155-170.
• Constantinescu CS, Farooqi N, O'Brien K, et al. Experimental autoimmune encephalomyelitis (EAE) as a model for multiple sclerosis (MS). Br J Pharmacol . 2011; 164: 1079-1106.
• Dendrou CA, Fugger L, Friese MA. Immunopathology of multiple sclerosis. Nat Rev Immunol . 2015; 15: 545-558.
• Owens TD, Smith PF, Redfern A, et al. Phase 1 clinical trial evaluating safety, exposure and pharmacodynamics of BTK inhibitor tolebrutinib (PRN2246, SAR442168). Clin Transl Sci . 2021.
• Caldwell RD, Qiu H, Askew BC, et al. Discovery of Evobrutinib: An Oral, Potent, and Highly Selective, Covalent Bruton's Tyrosine Kinase (BTK) Inhibitor for the Treatment of Immunological Diseases. J Med Chem . 2019; 62: 7643-7655.
• Ding J, Cai Y, Deng Y, et al. Telitacicept Following Plasma Exchange in the Treatment of Subjects With Recurrent NMOSD: Study Protocol for a Single-Center, Single-Arm, Open-Label Study. Front Neurol . 2021; 12: 596791.
• Frieder J, Kivelevitch D, Menter A. Secukinumab: a review of the anti-IL-17A biologic for the treatment of psoriasis. Ther Adv Chronic Dis . 2018; 9: 5-21.
• Brinkmann V, Billich A, Baumruker T, et al. Fingolimod (FTY720): discovery and development of an oral drug to treat multiple sclerosis. Nat Rev Drug Discov . 2010; 9: 883-897.
• Tan YV, Abad C, Lopez R, et al. Pituitary adenylyl cyclase-activating polypeptide is an intrinsic regulator of Treg abundance and protects against experimental autoimmune encephalomyelitis. Proc Natl Acad Sci U S A . 2009; 106: 2012-2017.
• Seno A, Maruhashi T, Kaifu T, et al. Exacerbation of experimental autoimmune encephalomyelitis in mice deficient for DCIR, an inhibitory C-type lectin receptor. Exp Anim . 2015; 64: 109-119.
• Krumbholz M, Derfuss T, Hohlfeld R, et al. B cells and antibodies in multiple sclerosis pathogenesis and therapy. Nat Rev Neurol . 2012; 8: 613-623.
• Denayer T, Stöhr T, Roy MV. Animal models in translational medicine: Validation and prediction. European Journal of Molecular & Clinical Medicine . 2014; 2.
• Babaloo Z, Aliparasti MR, Babaiea F, et al. The role of Th17 cells in patients with relapsing-remitting multiple sclerosis: interleukin-17A and interleukin-17F serum levels. Immunol Lett . 2015; 164: 76-80.
• Prineas JW, Connell F. The fine structure of chronically active multiple sclerosis plaques. Neurology . 1978; 28: 68-75.
• Chisari CG, Sgarlata E, Arena S, et al. Rituximab for the treatment of multiple sclerosis: a review. J Neurol . 2022; 269: 159-183.
• Evans JG, Chavez-Rueda KA, Eddaoudi A, et al. Novel suppressive function of transitional 2 B cells in experimental arthritis. J Immunol . 2007; 178: 7868-7878.
• Fillatreau S, Sweenie CH, McGeachy MJ, et al. B cells regulate autoimmunity by provision of IL-10. Nat Immunol . 2002; 3: 944-950.
• Weber MS, Prod'homme T, Patarroyo JC, et al. B-cell activation influences T-cell polarization and outcome of anti-CD20 B-cell depletion in central nervous system autoimmunity. Ann Neurol . 2010; 68: 369-383.
• Burger JA. Bruton Tyrosine Kinase Inhibitors: Present and Future. Cancer J . 2019; 25: 386-393.

Article Info

• Journal of Immunological Sciences
• Article Type : Original Research Article
• View/Download pdf
• DOI : 10.29245/2578-3009/2022/1.1231

Article Notes

• Published on: March 31, 2022
• Multiple sclerosis
• Experimental autoimmune encephalomyelitis
• Myelin oligodendrocyte glycoprotein
• Mechanisms of action

*Correspondence:

# Yuxi Yan and Quan Zhao contributed equally to this work.

Copyright: ©2022 Yang W . This article is distributed under the terms of the Creative Commons Attribution 4.0 International License.

Multiple Sclerosis Discovery Forum

Inspiring connections, utility navigation, user top menu.

• Welcome, guest

Search form

You are here, experimental autoimmune encephalomyelitis, animal description.

Experimental autoimmune encephalomyelitis (EAE) is the oldest and most frequently used model system for studying MS in laboratory animals. Rather than a single model, EAE is a family of models in which central nervous system inflammation occurs after immunization against CNS-specific antigen. In its classic form, EAE is a CD4+ T cell–mediated autoimmune disease in which immunization with myelin proteins or peptides induces the migration of activated autoreactive T cells across the blood-brain barrier and into the CNS; alternatively, transfer of autoreactive T cells activated by such antigens can achieve the same result (Miller and Karpus, 2007). In the past 20 years, researchers have used transgenic techniques to develop new versions of the model and to expand its utility. For example, T-cell receptor (TCR) transgenic mice carry T cells engineered to respond to specific brain antigens. Similarly, genetic engineering also allows researchers to induce EAE in animals that lack a particular cytokine or immune molecule to test whether it is involved in the disease.

Initially, rats, guinea pigs, and nonhuman primates comprised the largest contingent of EAE animals. Today, however, most researchers use mice—and two strains in particular power the bulk of EAE studies. C57BL/6 mice (sometimes called B6), the most commonly used strain in the lab, are usually used to created transgenic animals, so if your plans include genetic engineering, stick with them. These mice, when immunized with peptides of myelin oligodendrocyte glycoprotein (MOG) along with an adjuvant (usually Complete Freund’s Adjuvant, pertussis toxin, or both), generally develop a chronic form of EAE, with inflammation, demyelination, and oligodendrocyte and neuronal death occurring about 2 weeks after disease induction.

A different mouse, the SJL/J strain, has played a crucial role in research because it develops a relapsing-remitting form of the disease—the most commonly seen version of MS. In these animals, immunization with myelin basic protein (MBP) or with proteolipid protein (or peptides of it)—also with an adjuvant—spurs the first signs of disease within 7 to 14 days. After the initial inflammatory attack subsides, the animals go into remission and later experience episodes that involve inflammation, demyelination, and axonal loss (Merrill, 2008). Other animals can experience different clinical and pathological manifestations of the disease; in Lewis rats, a commonly used lab strain, for example, EAE provokes no demyelination, inflammation localizes primarily to the spinal cord, and the animals experience acute paralytic disease and then recover completely (Baxter, 2007; Croxford et al. , 2011).

Induction method

Active induction—direct immunization with myelin protein or peptides thereof—is the easiest and fastest way to induce EAE. In the alternative approach, called adoptive transfer, animals are immunized against a particular antigen and then sacrificed. Their T cells are then harvested, reactivated with the immunizing antigen in culture, and injected into recipient animals. Which strategy you use depends on the question you’re investigating, says immunologist Burkhard Becher of the University of Zurich in Switzerland. If you’re studying the initial phase of the immune response—how T cells are activated—you’ll probably want to employ an active induction disease model, in which the injected antigen kicks off this event.

If you’re primarily interested in the effector phase—that is, how the activated T cells encounter and attack cells they deem pathogenic—an adoptive transfer model is appropriate. Because an experimenter can choose the type of introduced T cell, the latter approach offers the opportunity to tease apart the roles of different T cell lineages in the immune response, for example. A particular mouse strain might resist EAE induced by one approach but succumb to the other—and figuring out why can reveal important molecular mechanisms, Becher adds. “But one way or the other, if you wanted to look at therapeutic efficacy, then you’d have to look at both for sure,” he says.

Mice engineered to carry TCRs specific for certain brain proteins often develop spontaneous EAE, with varying frequency, depending on the particular TCR and the animal’s genetic background (Krishnamoorthy et al. , 2007). Removing the need to inject a bolus of antigen or T cells into the animals, as in active induction and passive transfer, makes the process of disease onset much less artificial, says Vijay Kuchroo, an immunologist at Harvard Medical School in Boston. However, cautions Stefanie Kuerten, an anatomist at the University Hospitals of Cologne in Germany, a spontaneous model doesn’t allow researchers to predict when—or if—mice will develop disease, so if you want to study the proliferation, turnover, pathogenicity, and other characteristics of different T cell clones, then mice that spontaneously develop EAE might not be ideal subjects because you won’t know when to look. One model—a cross between two different transgenic C57B/6 lines, one engineered to carry MOG-specific T cells and the other to carry MOG-specific B cells—is becoming the spontaneous EAE model of choice (Krishnamoorthy et al ., 2006; Bettelli et al ., 2006).. Additionally, unlike most EAE models, in which T cell responses drive pathology, B cell activation in these animals contributes significantly to disease development. 

Key pathological features

Most variants of the EAE model are thought to reflect a CD4+-mediated immune response. Pathological elements manifest in about 7 to 14 days when EAE is actively induced but appear more quickly upon adoptive transfer of activated myelin-specific T cells. EAE generally targets the spinal cord and sometimes the cerebellum, causing inflammation followed by demyelination and axonal damage. In monophasic or relapsing-remitting EAE, varying degrees of remyelination occur. The specific pathological features vary dramatically depending on the animal, genetic strain, induction method, and autoantigen used (Miller and Karpus, 2007; Pachner 2011; Baxter 2007).

Key clinical features

Classic EAE in mice causes so-called flaccid paralysis characterized by decreased muscle tone that progresses from the tail upward along the body. A six-point scale (0-5) reflects the severity of symptoms, with a score of 1 being tail paralysis, a score of 4 indicating quadriplegia, and a score of 5 defined as death. In some models, strains, and species—particularly those in which disease pathology reaches the cerebellum—animals may instead or additionally experience lack of coordination, or ataxia. Symptoms can be chronic, monophasic, or relapsing-remitting. This disease-progression pattern and other clinical features, such as the type of paralysis that occurs, depend on the animal, genetic strain, induction method, and autoantigen used (Merrill, 2008; Miller and Karpus, 2007; Baxter, 2007). 

EAE provides a means for investigating mechanisms of autoimmune-related CNS damage and demyelination. It has broad similarities to MS—for example, the characteristics of demyelination and partial remyelination, the distribution of lesions around blood vessels, and the presence of immunoglobulin G in the cerebrospinal fluid (Baxter, 2007). As the most widely used model system for studying MS, with many variations (see above for types of variation), it offers an array of possibilities for matching an experimental question with a physiological profile. 

Weaknesses/Caveats

The biology of all EAE models diverges significantly from that of MS. For example, clear autoimmunity underlies EAE, but MS lacks some features of classic autoimmunity, such as a known autoantigen that sparks the disease (Denic et al. , 2011). And whereas CD4+ T cells play the primary role in prompting pathology in most versions of EAE, a wider spectrum of T cells as well as other immune cells such as B cells contribute significantly to MS (see " B Cells Step Into the Limelight ").  

Researchers therefore caution against drawing direct parallels between MS and EAE. This caveat applies especially in drug testing, they say. Work in EAE has led to the approval of four drug therapies for MS: natalizumab , mitoxantrone , glatiramer acetate , and fingolimod . However, many prospective treatments that relieve symptoms in EAE mice have not worked in human trials, and some, such as tumor necrosis factor-α, have exacerbated MS (Baxter, 2007). “ ’Oh, the mice do better, so this could be a therapy’—that is something I don’t even look at anymore,” Becher says.

Part of the mismatch, he says, stems from the nature of the experimental manipulations: They create a situation that is so extreme on the cellular level that the condition they trigger differs in crucial ways from MS. The spontaneous B- and T-cell model described above, for example, “has been discussed as the supermodel for spontaneous-onset EAE, but that’s a bit unfair because if all of your B cells and all of your T cells recognize brain antigen, then it isn’t surprising that these mutants develop EAE,” Becher says. “This is a very valuable addition to the toolbox available to date, but discoveries based on one single EAE model don’t necessarily translate to the human disease.”

Regardless of which EAE model you use, it’s not just the strain, the antigen, and the induction method that determines animals’ disease course but also the vivarium in which they are raised. It’s very tough to induce EAE in animals raised in a “dirty” environment; pathogen-free status is best. “But even if you have such a facility, there are differences depending on animal houses”—for example, in the time to disease onset, Kuerten says. Becher agrees that such environmental variations are a major issue but notes that if EAE experiments more often included the necessary controls, such as tests in both actively induced and adoptive transfer models, the noise generated by animal husbandry variations would matter less (see also Tips section below). 

Disease processes that can be studied

EAE provides a powerful framework for investigating the inflammatory elements of MS. Researchers have used the model to study processes such as tolerance, immune surveillance, molecular mimicry, epitope spreading, environmental triggers and genetics of autoimmune disease, inflammation, lymphocyte entry into the CNS through the blood-brain barrier, functional differences between T-cell clones; relapse mechanisms, and immune-mediated demyelination and tissue injury (Pachner, 2011; Krishnamoorthy and Wekerle, 2009). A handful of EAE models, including the TCR transgenic relapse-remitting mouse model described above, also allow researchers to explore the role of B cells in EAE and, by extension, in MS.

Disease processes that cannot be studied

Because it’s impossible to predict the timing and location of lesions in EAE, the model is not ideal for studying the complete cycle of demyelination and partial remyelination, says Richard Ransohoff, an immunologist at the Cleveland Clinic in Ohio. What’s more, says Stephen Miller, an immunologist at Northwestern University Feinberg School of Medicine in Chicago, Illinois, assessing whether a compound directly guards against or even ameliorates neuronal damage in EAE or other models with an immune component is challenging. Many drugs that researchers are testing as neuroprotective agents might also exert immunosuppressive effects; that characteristic makes it difficult to determine whether a compound acts directly on neurons or simply quells the inflammation that damages them. To separate the two processes, a toxic agent such as cuprizone to elicit demyelination but not inflammation might be useful and more appropriate (see " Animal Arsenal ").  

The sheer number of permutations of EAE models makes their use a minefield for experimental error, and diving into the literature is often not enough to develop the needed level of expertise to use EAE most effectively, Becher says. Apprentice yourself to an experienced EAE researcher who can critique your scientific plan, Becher suggests: “When I started on EAE, I latched onto a couple of people, and it was very helpful.”

If you are using EAE to test experimental therapeutic compounds, says David Baker, an immunologist at Queen Mary, University of London, make sure to treat the animals with your compound of interest after immunizing the animals rather than before. This order more accurately reflects the experience of MS patients, who don’t receive drugs until they experience symptoms. Baker and his colleague Sandra Amor at VU University Medical Center in the Netherlands laid out a set of guidelines that proposes minimally acceptable standards for publishing studies that use EAE as a preclinical model (Baker and Amor, 2011). 

Researchers also say that it’s important to confirm findings in different models—especially when testing investigational compounds. Experiments should be done in an actively induced model and a passive transfer model. As with all types of experiments, controls are crucial. If you’re testing a compound that blocks some immune factor, make sure that inducing EAE in a knockout mouse that lacks the factor produces the same result, suggests Cris Constantinescu, a neurologist at the University of Nottingham in the U.K.

Utility for probing relevant biology

Researchers widely agree that EAE has played a significant role in uncovering basic immunological features of multiple sclerosis—as well as immune response more generally. In the early days of EAE, studying disease states caused by injecting brain homogenate in different species helped pinpoint how particular tissue-specific molecules produce autoimmune reactions (Krishnamoorthy and Wekerle, 2009). A landmark 1981 study reported that injecting T cells that target MBP into rats causes EAE and thus propelled T cells into their starring role in MS research (Ben-Nun et al. , 1981). Later work expanded on the inflammatory mechanism such as the role of CD4+ regulatory T cells (Olivares-Villagomez et al. , 1998; McGreachy et al ., 2005) and epitope spreading (McRae et al ., 2005).

In 2005, researchers identified a novel class of helper T cells called Th17 cells that seem to contribute to the pathogenesis of EAE (and possibly MS), although the precise mechanism remains unclear (Korn et al ., 2009; Becher and Segal, 2011). Other experiments have riffed on classic EAE models to study the involvement of specific cytokines such as GM-CSF (McQualter et al ., 2001; Codarri et al., 2011) and have also panned out to probe nonclassic contributors to MS such as B cells (Pollinger et al ., 2009).  

EAE is also the go-to model for testing the effectiveness of potential MS therapeutics. However, its contribution to this endeavor is controversial: Most researchers support its utility to at least some extent, but a few dismiss it. 

EAE was born out of one investigator’s effort more than 80 years ago to determine why individuals who received an early form of the first rabies vaccine sometimes experienced attacks of paralysis. In the course of his experiments, the researcher, Thomas Rivers, injected control animals with emulsified brain tissue; they developed brain-specific antibodies and a proportion of them became transiently paralyzed. Based on Rivers’s work, other researchers began to probe the possibility that the antibodies were creating an immune response that led to demyelination. The experimental system thus became a model of demyelinating disease and human autoimmune disease (Baxter, 2007).

Mice (C57BL/6, SJL, other)

Jackson Laboratory ( http://www.jax.org/ )

Charles River Laboratories ( http://www.criver.com/ )

Harlan Laboratories ( http://www.harlan.com/ )

Myelin proteins or peptides

Hooke Laboratories ( http://hookelabs.com/ )

Adjuvant (pertussis toxin, Complete Freund’s Adjuvant)

Difco Microbiology ( http://www.vgdusa.com/DIFCO.htm )

Thanks to David Baker, Burkhard Becher, and Stephen Miller for reviewing this article.

Experimental Autoimmune Encephalomyelitis

Series: Methods In Molecular Biology > Book: T-Helper Cells

Protocol | DOI: 10.1007/978-1-0716-1311-5_27

• Laboratory of Neuroimmunology, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy

Experimental autoimmune encephalomyelitis, originally experimental allergic encephalomyelitis, is the well-known animal model of multiple sclerosis, an immune- mediated, demyelinating, inflammatory chronic disease of the central nervous systemCentral

Experimental autoimmune encephalomyelitis, originally experimental allergic encephalomyelitis, is the well-known animal model of multiple sclerosis, an immune- mediated, demyelinating, inflammatory chronic disease of the central nervous systemCentral nervous system (CNS). The experimental disease is widely utilized to test new therapies in preclinical studies, to investigate new hypothesis on the possible pathogenic mechanisms of autoimmune reaction directed against the central nervous systemCentral nervous system (CNS) or more generally to investigate the interactions between the immune system and the central nervous systemCentral nervous system (CNS) that lead to neuroinflammation. The experimental autoimmune encephalomyelitis may be induced following different protocols in mammals, including nonhuman primates, and autoreactive CD4+ T-lymphocytes directed against myelinantigens are the main factors. Here, after introducing the model, we describe the protocol to induce active EAEExperimental autoimmune encephalomyelitis (EAE) in inbred mice, we report on a table the different clinical courses of EAEExperimental autoimmune encephalomyelitis (EAE) depending on the combination of antigen/mouse strain and we provide indications on how to evaluate the clinics and pathology of this induced disease.

Figures ( 0 ) & Videos ( 0 )

Experimental specifications, other keywords.

Citations (8)

Associated articles.

• (This version), 2021
• Praveen Rao & Benjamin M. Segal, 2012
• Praveen Rao & Benjamin M. Segal, 2004
• Van Epps VL (2005) Thomas River and the EAE model. J Exp Med 202(1):4. https://doi.org/10.1084/jem.2021fta
• Procaccini C, De Rosa V, Pucino V et al (2015) Animal models of multiple sclerosis. Eur J Pharmacol 759:182–191. https://doi.org/10.1016/j.ejphar.2015.03.042
• Costantinescu CS, Farooqi N, O’Briejn K et al (2011) Experimental autoimmune encephalomyelitis (EAE) as a model of multiple sclerosis (MS). Br J Pharmacol 164:1079–1106. https://doi.org/10.1111/j.1476-5381.2011.01302.x
• Kuchroo VK, Anderson AC, Waldner H et al (2002) T-cell response in experimental autoimmune encephalomyelitis (EAE): role of self and cross-reactive antigens in shaping, tuning and regulating the autopathogenic T-cell repertoire. Annu Rev Immunol 20:101–123. https://doi.org/10.1146/annurev.immunol.20.081701.141316
• Gran B, Zhang GX, Yu S et al (2002) IL-12p35-deficient mice are susceptible to experimental autoimmune encephalomyelitis: evidence for redundancy in the IL-12 system in the induction of central nervous system demyelination. J Immunol 169:7104–7110. https://doi.org/10.4049/jimmunol.169.12.7104
• Langrish CL, Chen Y, Blumenschein WM et al (2005) IL-23 drives a pathogenic T- cell population that induces autoimmune inflammation. J Exp Med 201:233–240. https://doi.org/10.1084/jem.20041257
• Paterson PY (1960) Transfer of allergic encephalomyelitis in rats by means of lymph node cells. J Exp Med 111:119–135. https://doi.org/10.1084/jem.111.1.119
• Furlan R, Cuomo C, Martino G (2009) Animal models of multiple sclerosis. Methods Mol Biol 549:157–173
• Bjielobaba I, Begovic-Kupresanin V, Pekovic S et al (2018) Animal models of multiple sclerosis: focus on experimental autoimmune encephalomyelitis. J Neurosci Res 96:1021–1042. https://doi.org/10.1002/jnr.24224
• Fritz RB, Chou CH, McFarlin DE (1983) Relapsing murine experimental allergic encephalomyelitis induced by myelin basic protein. J Immunol 130:1024–1026
• Amor S, Groome N, Linington C et al (1994) Identification of epitopes of myelin oligodendrocyte glycoprotein for the induction of experimental allergic encephalomyelitis in SJL and Biozzi AB/H mice. J Immunol 153:4349–4356
• Oliver AR, Lyon GM, Ruddle NH (2003) Rat and human myelin oligodendrocyte glycoproteins induce experimental autoimmune encephalomyelitis by different mechanisms in C57BL/6 mice. J Immunol 17:462–468. https://doi.org/10.4049/jimmunol.171.1.462
• Sundvall M, Jirholt J, Yang HT et al (1995) Identification of murine loci associated with susceptibility to chronic experimental autoimmune encephalomyelitis. Nat Genet 10:313–317. https://doi.org/10.1038/ng0795-313
• Kipp M, van der Star B, Vogel DYS et al (2012) Experimental in vivo and in vitro models of multiple sclerosis: EAE and beyond. Mult Scler Relat Disord 1:15–28. https://doi.org/10.1016/j.msard.2011.09.002
• Giuliani F, Fu SA, Metz LM et al (2005) Effective combination of minocycline and interferon beta in a model of multiple sclerosis. J Neuroimmunol 175:83–91. https://doi.org/10.1016/j.jneuroim.2005.04.020

Advertisement

• Search Menu
• Sign in through your institution
• CNS Injury and Stroke
• Epilepsy and Sleep
• Movement Disorders
• Multiple Sclerosis/Neuroinflammation
• Neuro-oncology
• Neurodegeneration - Cellular & Molecular
• Neuromuscular Disease
• Neuropsychiatry
• Pain and Headache
• Advance articles
• Editor's Choice
• Author Guidelines
• Submission Site
• Why publish with this journal?
• Open Access
• About Brain
• Editorial Board
• Advertising and Corporate Services
• Journals Career Network
• Self-Archiving Policy
• Dispatch Dates
• Terms and Conditions
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

Competing interests.

• < Previous

Autoimmune ‘secondary synaptopathies’: do NMDAR antibodies cause a primary extra-synaptopathy?

• Article contents
• Figures & tables
• Supplementary Data

Meng Zhao, David R Lynch, Sarosh R Irani, Autoimmune ‘secondary synaptopathies’: do NMDAR antibodies cause a primary extra-synaptopathy?, Brain , Volume 147, Issue 8, August 2024, Pages 2601–2603, https://doi.org/10.1093/brain/awae236

• Permissions Icon Permissions

This scientific commentary refers to ‘NMDA receptor autoantibodies primarily impair the extrasynaptic compartment’ by Jamet et al. ( https://doi.org/10.1093/brain/awae163 ).

The field of autoantibody-mediated neurological illnesses has expanded rapidly over the past 15 years, with many new forms of encephalitis identified. Today, the two most commonly diagnosed autoimmune encephalitides are those associated with autoantibodies against leucine-rich glioma-inactivated protein 1 (LGI1) 1 or the N -methyl- D- aspartate receptor (NMDAR). 2 In both conditions, autoantibodies are exclusively directed against known brain antigens, and passive transfer of these autoantibodies to experimental rodents mimics key aspects of the disease phenotype, fulfilling Witebsky’s criteria for pathogenicity. 3 But despite these advances, the molecular mechanisms underlying pathogenicity remain poorly understood, limiting the development of new therapies for patients.

The presence of NMDAR antibodies has been associated with diverse psychiatric features, as well as with multifocal seizures, memory deficits, decreased consciousness and autonomic instability. 2 , 4 Studies have concluded that these bivalent antibodies trigger internalization of surface neuronal NMDARs within 2 h (without affecting neighbouring synaptic proteins), leading to synaptic NMDAR hypofunction. 2 , 5 , 6 This NMDAR internalization is not observed with monovalent Fab fragments. Autoimmune disease associated with NMDAR antibodies is therefore considered a pure NMDAR-IgG-opathy, with disruption of glutamatergic transmission leading to the induction of symptoms. Removal of autoantibodies from hippocampal neuronal cultures reverses this effect, restoring NMDARs to the cell surface 2 : this is considered the molecular mechanism for patient recovery. Current dogma thus upholds a simple pathophysiological model whereby bivalent autoantibodies cross-link synaptic NMDARs, leading to their internalization. In this issue of Brain , Jamet and colleagues 7 challenge the details of this mechanism by reframing the primacy of synaptic dysfunction, the purity of the dysfunction and the relevance of antibody bivalency. Their findings call for a change in the classification and molecular nosology of NMDAR-antibody encephalitis, and potentially for other syndromes associated with autoantibodies that target key neuronal autoantigens.

For several years, it has been recognized that the synaptic cleft, which is 10–40-nm wide and further constrained by the presence of abundant extracellular matrix proteins, 8 is an inaccessible and inhospitable space for ∼10–15-nm high IgG molecules. The question then arises, are these disorders truly ‘synapt’-opathies and, if so, how do autoantibodies infiltrate this narrow space to trigger dysfunction of synaptic autoantigens ( Fig. 1 )?

Dysregulation of the extrasynaptic compartment is proposed to result in secondary synaptopathy. Left : The synaptic cleft is a narrow space rich in extracellular matrix proteins; its small size means that molecules crossing it must also be small. Right : Overview of the findings of Jamet and colleagues. 7 Figure created with BioRender.com .

To address these questions, Jamet and colleagues 7 employed advanced single molecule-based imaging to observe the action of NMDAR antibodies on live dissociated hippocampal neurons. Their findings yield new insights which may reframe thinking in the field. First, they observed that within 30 min of incubation with neurons, patient CSF or monoclonal NMDAR antibodies markedly enhanced the dynamics of extrasynaptic NMDARs, without affecting synaptic (Homer-co-localized) NMDARs. These changes were accompanied by an increase in the area of the extrasynaptic surface protein interactome, with marked disorganization of membrane proteins in the extrasynaptic compartment.

Importantly, synaptic proteins were not altered over this time course. However, after 24 h of antibody incubation, synaptic dynamics and the synaptic interactome had also been disrupted. This suggests that the immediate effect of NMDAR antibodies was extrasynaptic, occurring in a subcellular region not constrained by the ∼20 nm synaptic cleft. Further, these rapid effects included modulation of proteins neighbouring the NMDAR, indicating a broader impact than previously understood. To complement co-localizations with canonical synaptic markers, the authors then conjugated NMDAR antibodies to 1-μm wide beads to definitively deny them access to the synaptic cleft. This purely extrasynaptic model effectively mimicked the effects of the patients’ antibodies over the 30-min and 24-h time courses. Finally, this study challenged the long-standing dogma of crosslinking-mediated internalization of NMDARs by showing that the number of GluN1-expressing clathrin-coated pits did not increase after antibody exposure; instead NMDARs were redistributed at the neuronal surface, as was also observed for Fab fragments.

The findings of Jamet and colleagues 7 (summarized in Fig. 1 ) represent molecular, clinical and nosological advances in our understanding of NMDAR-antibody encephalitis. They suggest that primary autoantibody-induced dysfunction is extrasynaptic, and that the synaptic effects are secondary, occurring over a delayed time course. This is consistent with early reports of an extrasynaptic mechanism based on autoantibody specificity. 9 According to this proposed mechanism, existing pathways of crosstalk between synaptic and extrasynaptic receptors serve as mediators of disease. The new findings also imply that clinical manifestations may be secondary to the disruption of other NMDAR-co-localizing proteins. This is intriguing given that some features seen in patients with NMDAR-antibody encephalitis are not easily explained by pure NMDAR dysfunction. For example, a neuroleptic malignant-like syndrome is classically considered secondary to blockade of dopaminergic pathways, while seizures are not easily squared with NMDAR hypofunction.

Overall, the current findings challenge the prevailing dogma in three ways, by suggesting that: (i) autoimmune channelopathies should potentially be reclassified as secondary synaptopathies occurring after a primary extra-synaptopathy; (ii) focusing exclusively on NMDAR dysfunction may be insufficient to fully explain the observed disease manifestations; and (iii) we should rethink the mechanisms by which molecules that target the NMDAR, such as allosteric inhibitors, 10 could have therapeutic effects in patients with NMDAR-antibody encephalitis.

Finally, if these findings are validated, the concept of secondary synaptopathies could also have implications for the nosology of other autoantibody-mediated diseases thought to manifest with primary synaptic dysfunction. It is possible that LGI1-, CASPR2- and other autoantibody-mediated diseases may also turn out to be secondary synaptopathies.

This work was funded in whole or in part by fellowships from UKRI, Medical Research Council [MR/V007173/1], Wellcome Trust [104079/Z/14/Z] and by the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre (BRC). For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript (AAM) version arising from this submission. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.

D.R.L. receives royalties from testing for anti-NMDA receptor encephalitis. S.R.I. receives licensed royalties on patent application WO/2010/046716 entitled ‘Neurological Autoimmune Disorders’, and has filed two other patents entitled ‘Diagnostic method and therapy’ (WO2019211633 and US app 17/051,930; PCT application WO202189788A1) and ‘Biomarkers’ (WO202189788A1, US App 18/279,624; PCT/GB2022/050614). S.R.I. has received honoraria/research support from UCB, Immunovant, MedImmun, Roche, Janssen, Cerebral therapeutics, ADC therapeutics, Brain, CSL Behring, and ONO Pharma.

Irani SR , Alexander S , Waters P , et al.  Antibodies to Kv1 potassium channel-complex proteins leucine-rich, glioma inactivated 1 protein and contactin-associated protein-2 in limbic encephalitis, Morvan's syndrome and acquired neuromyotonia . Brain . 2010 ; 133 : 2734 – 2748 .

Google Scholar

Dalmau J , Gleichman AJ , Hughes EG , et al.  Anti-NMDA-receptor encephalitis: Case series and analysis of the effects of antibodies . Lancet Neurol . 2008 ; 7 : 1091 – 1098 .

Rose NR , Bona C . Defining criteria for autoimmune diseases (Witebsky's postulates revisited) . Immunol Today . 1993 ; 14 : 426 – 430 .

Irani SR , Bera K , Waters P , et al.  N-methyl-D-aspartate antibody encephalitis: Temporal progression of clinical and paraclinical observations in a predominantly non-paraneoplastic disorder of both sexes . Brain . 2010 ; 133 ( Pt 6 ): 1655 – 1667 .

Moscato EH , Peng X , Jain A , Parsons TD , Dalmau J , Balice-Gordon RJ . Acute mechanisms underlying antibody effects in anti-N-methyl-D-aspartate receptor encephalitis . Ann Neurol . 2014 ; 76 : 108 – 119 .

Ladepeche L , Planaguma J , Thakur S , et al.  NMDA receptor autoantibodies in autoimmune encephalitis cause a subunit-specific nanoscale redistribution of NMDA receptors . Cell Rep . 2018 ; 23 : 3759 – 3768 .

Jamet Z , Mergaux C , Meras M , et al.  NMDA receptor autoantibodies primarily impair the extrasynaptic compartment . Brain . 2024;147:2745-2760.

Zuber B , Nikonenko I , Klauser P , Muller D , Dubochet J . The mammalian central nervous synaptic cleft contains a high density of periodically organized complexes . Proc Natl Acad Sci U S A . 2005 ; 102 : 19192 – 19197 .

Sharma R , Al-Saleem FH , Panzer J , et al.  Monoclonal antibodies from a patient with anti-NMDA receptor encephalitis . Ann Clin Transl Neurol . 2018 ; 5 : 935 – 951 .

Mannara F , Radosevic M , Planaguma J , et al.  Allosteric modulation of NMDA receptors prevents the antibody effects of patients with anti-NMDAR encephalitis . Brain . 2020 ; 143 : 2709 – 2720 .

Email alerts

Citing articles via, looking for your next opportunity.

• Contact the editorial office
• Guarantors of Brain
• Recommend to your Library

Affiliations

• Online ISSN 1460-2156
• Print ISSN 0006-8950
• Copyright © 2024 Guarantors of Brain
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

Experimental autoimmune encephalomyelitis as a model of immune-mediated CNS disease

Affiliation.

• 1 Max-Planck-Institut for Psychiatry, Martinsried, Germany.
• PMID: 8260829
• DOI: 10.1016/0959-4388(93)90153-p

Experimental autoimmune encephalomyelitis models are used to analyze the generation and organization of the myelin-specific autoimmune repertoire, and potential immunoregulatory loops preventing spontaneous activation of encephalitogenic T cells. These lymphocytes are profoundly modulated by infectious agents, which may trigger, or more commonly, prevent experimental autoimmune encephalomyelitis. The development and resolution of the pathogenic central nervous system infiltrations is controlled by locally produced cytokines that cause recruitment of infiltrate cells, and their disappearance. Several of the new findings seem now to be applicable for therapeutic strategies, especially with the aim of interfering with immunospecific recognition steps involved in disease generation.

PubMed Disclaimer

Similar articles

• Apoptotic elimination of V beta 8.2+ cells from the central nervous system during recovery from experimental autoimmune encephalomyelitis induced by the passive transfer of V beta 8.2+ encephalitogenic T cells. Tabi Z, McCombe PA, Pender MP. Tabi Z, et al. Eur J Immunol. 1994 Nov;24(11):2609-17. doi: 10.1002/eji.1830241107. Eur J Immunol. 1994. PMID: 7957554
• Special AT-rich sequence binding protein 1 is required for maintenance of T cell receptor responsiveness and development of experimental autoimmune encephalomyelitis. Akiba Y, Kuwabara T, Mukozu T, Mikami T, Kondo M. Akiba Y, et al. Microbiol Immunol. 2018 Apr;62(4):255-268. doi: 10.1111/1348-0421.12579. Epub 2018 Mar 30. Microbiol Immunol. 2018. PMID: 29388727 Free PMC article.
• Antigen-specific therapy of experimental allergic encephalomyelitis by soluble class II major histocompatibility complex-peptide complexes. Sharma SD, Nag B, Su XM, Green D, Spack E, Clark BR, Sriram S. Sharma SD, et al. Proc Natl Acad Sci U S A. 1991 Dec 15;88(24):11465-9. doi: 10.1073/pnas.88.24.11465. Proc Natl Acad Sci U S A. 1991. PMID: 1722335 Free PMC article.
• [Mechanisms involved in the regulation of immune response in experimental autoimmune encephalomyelitis in mice]. Tutaj M, Szczepanik M. Tutaj M, et al. Postepy Hig Med Dosw (Online). 2006;60:571-83. Postepy Hig Med Dosw (Online). 2006. PMID: 17115007 Review. Polish.
• [The pathogenic mechanism of experimental autoimmune encephalomyelitis and approach for new immunological therapy]. Yokoyama K, Yamamura T. Yokoyama K, et al. Nihon Rinsho. 1994 Nov;52(11):3013-8. Nihon Rinsho. 1994. PMID: 7996703 Review. Japanese.
• Gut-Brain Interactions and Their Impact on Astrocytes in the Context of Multiple Sclerosis and Beyond. Zißler J, Rothhammer V, Linnerbauer M. Zißler J, et al. Cells. 2024 Mar 13;13(6):497. doi: 10.3390/cells13060497. Cells. 2024. PMID: 38534341 Free PMC article. Review.
• Targeted antigen delivery to DEC-205⁺ dendritic cells for tolerogenic vaccination. Petzold C, Schallenberg S, Stern JN, Kretschmer K. Petzold C, et al. Rev Diabet Stud. 2012 Winter;9(4):305-18. doi: 10.1900/RDS.2012.9.305. Epub 2012 Dec 28. Rev Diabet Stud. 2012. PMID: 23804268 Free PMC article. Review.
• Mechanisms of dendritic cell trafficking across the blood-brain barrier. Sagar D, Foss C, El Baz R, Pomper MG, Khan ZK, Jain P. Sagar D, et al. J Neuroimmune Pharmacol. 2012 Mar;7(1):74-94. doi: 10.1007/s11481-011-9302-7. Epub 2011 Aug 6. J Neuroimmune Pharmacol. 2012. PMID: 21822588 Free PMC article. Review.
• Promoting tolerance to proteolipid protein-induced experimental autoimmune encephalomyelitis through targeting dendritic cells. Stern JN, Keskin DB, Kato Z, Waldner H, Schallenberg S, Anderson A, von Boehmer H, Kretschmer K, Strominger JL. Stern JN, et al. Proc Natl Acad Sci U S A. 2010 Oct 5;107(40):17280-5. doi: 10.1073/pnas.1010263107. Epub 2010 Sep 20. Proc Natl Acad Sci U S A. 2010. PMID: 20855626 Free PMC article.
• Carcinoembryonic antigen-related cell adhesion molecule 1 modulates experimental autoimmune encephalomyelitis via an iNKT cell-dependent mechanism. Fujita M, Otsuka T, Mizuno M, Tomi C, Yamamura T, Miyake S. Fujita M, et al. Am J Pathol. 2009 Sep;175(3):1116-23. doi: 10.2353/ajpath.2009.090265. Epub 2009 Aug 21. Am J Pathol. 2009. PMID: 19700760 Free PMC article.

Publication types

• Search in MeSH

Related information

• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• HHS Author Manuscripts

Experimental Autoimmune Encephalomyelitis in Mice

Experimental autoimmune encephalitis (EAE), the animal model of multiple sclerosis (MS), has provided significant insight into the mechanisms that initiate and drive autoimmunity. Several central nervous system proteins and peptides have been used to induce disease, in a number of different mouse strains, to model the diverse clinical presentations of MS. In this chapter, we detail the materials and methods used to induce active and adoptive EAE. We focus on disease induction in the SJL/J, C57BL/6, and BALB/c mouse strains, using peptides derived from proteolipid protein, myelin basic protein, and myelin oligodendrocyte glycoprotein. We also include a protocol for the isolation of leukocytes from the spinal cord and brain for flow cytometric analysis.

1 Introduction

Multiple sclerosis (MS) is a chronic and debilitating autoimmune disease of the central nervous system (CNS), characterized by lesion formation in the white matter of the brain, spinal cord, and optic nerve ( 1 , 2 ). The precise mechanisms that trigger and drive MS are not completely understood. However, it is clear that myelin antigens are key targets ( 3 ). Autoreactive T cells and other immune cells, particularly macrophages, infiltrate the CNS and cause significant damage to the myelin sheath and underlying axons, resulting in neuronal dysfunction and death ( 4 , 5 ). Depending on the severity and location of the immune insult, patients may present with a range of neurological symptoms that impact physical functioning (e.g., muscle weakness, numbness or spasms, impaired balance and coordination, fatigue, incontinence) or mental capacity (e.g., memory loss, depression, cognitive difficulties) ( 6 ).

The frequency and severity in which MS symptoms occur differs greatly between individuals. Four main categories are used to classify the clinical course of MS in patients ( 7 ). Relapsing-remitting MS is the most common clinical pattern observed, and is characterized by recurrent attacks (relapses) followed by periods of remission in which little or no permanent neurological sequelae are evident. Secondary progressive MS describes the clinical course in which patients initially present with relapsing-remitting disease, but develop significant deficits that increase over time. Primary progressive MS describes the clinical course in which patients show progressive worsening of symptoms without relapse or remission phases. Progressive-relapsing MS is characterized by a steadily worsening disease state, in which patients undergo relapses without complete remission ( 7 ).

The animal model of MS, Experimental Autoimmune Encephalomyelitis (EAE), aims to replicate the clinical symptoms of disease in vivo, and has been induced in a range of species, including mice, rats, and hamsters ( 8 ). Two different methods of EAE induction have been described. Subcutaneous immunization of mice with an emulsion of myelin protein/peptide and complete Freund’s adjuvant (CFA) is referred to as “active EAE,” and models the induction and effector stages of disease ( 9 ). This process results in the direct priming of myelin epitope-specific CD4 + T cells in vivo, which migrate to the CNS and mediate autoimmune responses. In comparison, “adoptive EAE” models the effector phase of disease only. Activated CD4 + T cells are isolated from the draining lymph nodes of immunized mice, restimulated with the initiating myelin protein/peptide in vitro for several days, and then injected into naïve recipients. Disease is typically accelerated, more severe and uniform in comparison to active EAE, with higher incidence ( 9 ).

Depending on the initiating protein/peptide, animal species and strain used for the induction of EAE, different clinical etiologies may be mimicked ( Table 1 ). For example, induction of disease in the SJL/J mouse strain using the proteolipid protein (PLP) 139–155 peptide induces relapsing remitting disease ( Fig. 1 ). In comparison, immunization of C57BL/6 mice with the myelin oligodendrocyte glycoprotein (MOG) 35–55 peptide results in a chronic disease course ( Fig. 2 ). Immunization of BALB/c and C57BL/6 mice with the PLP 180–199 peptide also induces chronic disease, but causes relapsing-remitting disease in the SJL/J strain ( Fig. 3 ). The development of transgenic and humanized mice strains, outlined in Table 2 , have also proved to be invaluable tools in understanding the processes that drive autoimmunity. This chapter aims to provide a detailed protocol for inducing EAE in the mouse. Here, we focus on whole proteins and peptides derived from PLP, MOG, and myelin basic protein (MBP).We include the protocols to induce both active and adoptive disease, and the methodology used to isolate single-cell suspensions of leukocytes from brain tissue.

Induction of EAE in the SJL/J mouse using PLP 139–151 . Immunization of SJL/J mice with 50 µg of the PLP 139–151 peptide results in a relapsing-remitting course of EAE

Induction of EAE in the C57BL/6 mouse using MOG 35–55 . Immunization of C57BL/6 mice with 200 µg of the MOG 35–55 peptide results in a chronic course of EAE

Induction of EAE in the BALB/c, C57BL/6, and SJL/J mouse using PLP 180–199 . Immunization of BALB/c and C57BL/6 mice with 200 µg of the PLP 180–199 peptide results in a chronic course of EAE, but causes relapsing-remitting disease in SJL/J mice

Induction of active experimental autoimmune encephalomyelitis in mice

Transgenic mouse models of experimental autoimmune encephalomyelitis

AM acute monophasic, RR relapsing-remitting, PP primary progressive, tg transgenic, ODC oligodendrocytes, ND not defined

2 Materials

2.1 active induction of eae.

• Female SJL/J, C57BL/6, or BALB/c mice at 6–8 weeks old (Jackson Laboratories).
• Small animal clippers (e.g., model A5, blade size 50; Oster).
• Incomplete Freund’s Adjuvant (IFA; Difco).
• Mycobacterium Tuberculosis H37Ra, inactivated and desiccated (Difco).
• Desired myelin protein or peptide (see Table 1 ).
• Phosphate-buffered saline (PBS).
• Pertussis Toxin, if required (see Table 1 ; List Biologicals).
• 15 mL polystyrene test tubes.
• 18 and 25 G needles.
• 1 mL glass tuberculin syringes with Luer-Lok (VWR).

2.2 Adoptive Induction of EAE

• Materials listed in Section 2.1.
• Dissection forceps and scissors.
• 100 µm sieves (Becton Dickinson).
• 1 mL syringes.
• 50 mL conical polystyrene tubes.
• Recombinant mouse IL-12, if required (R&D Systems).
• 175 cm 2 culture flasks.
• 37 °C, 6 % CO 2 tissue culture incubator.
• 30.5 G needles.
• Complete Roswell Park Memorial Institute (cRPMI): 440 mL of calcium-free, l -glutamine-free RPMI medium, 5 mL of 100× l -glutamine, 5 mL of 100× Penicillin-Streptomycin, 500 µL of 55 µM 2-mercaptoethanol, and 50 mL of Fetal Bovine Serum (FBS).

2.3 Isolation of CNS Infiltrating Leukocytes

• Institution-approved anesthetic.
• 30 mL syringes with Luer-Lok.
• 21.5 G needles.
• Non-treated plastic petri dishes, 60 mm.
• 5 mL syringes.
• 18 G needle.
• Stainless steel wire mesh, 200–300 µm.
• 15 mL polystyrene tubes.
• Liberase, low Thermolysin concentration (Roche).
• DNase I (Sigma).
• Percoll (GE Healthcare).
• 10× no calcium, no magnesium, no phenol red Hank’s Balanced Salt Solution (HBSS).
• 10× no calcium, no magnesium HBSS.
• 0.5 M EDTA, pH 8.

3.1 Active Induction of EAE

• Shave the back of the mice using the small animal clippers ( see Note 1 ).
• Prepare complete Freund’s adjuvant (CFA) by combining 50 mL IFA and 200 mg M. tuberculosis H37Ra, resulting in a final concentration 4 mg/mL M. tuberculosis ( see Note 2 ).
• Prepare an emulsion of CFA and desired protein/peptide ( Table 1 ) by mixing 1 mL of CFA with 1 mL of desired peptide/protein diluted in PBS. Repeatedly draw up and expel the liquid from a 1 mL glass syringe into a 15 mL polystyrene test tube, using an 18 G needle ( see Notes 3 – 5 ).
• Slowly draw up the emulsion into a new 1 mL glass syringe using an 18 G needle, taking care not to introduce air bubbles. Replace the 18 G needle with a 25 G needle.
• Inject 100 µL of emulsion subcutaneously into the shaved backs of the mice, distributing evenly over three injection sites. One injection should be placed on the midline of the back just below the shoulders, and two on either side of the midline on the lower back.
• Refer to Table 1 to determine if pertussis toxin is needed for the mouse strain and initiating protein/peptide you are using. If required, dilute pertussis toxin to 1 µg/mL in sterile PBS ( see Note 6 ). Inject 200 µL of diluted pertussis toxin (200 ng per mouse) into the peritoneal cavity or intravenously on the day of disease induction, and again 48 h after induction ( see Note 7 ).

Clinical scoring of mice with experimental autoimmune encephalomyelitis

3.2 Adoptive Induction of EAE

• Immunize female donor SJL/J, C57BL/6, or BALB/c mice with the desired myelin protein/peptide as detailed in Section 3.1.
• Prepare cRPMI.

Induction of adoptive experimental autoimmune encephalomyelitis a

RR relapsing-remitting

• Place a 70 µm filter into a non-treated culture plate. Generate a single-cell suspension by pressing the lymph nodes in cRPMI through the filter with the plunger from a 1 mL syringe.
• Transfer the cRPMI solution containing the cells into a 50 mL polystyrene tube and centrifuge for 10 min at 300 × g .
• Remove the supernatant and raise the pellet in 1 mL of cRPMI and gently resuspend using a pipette. Add another 19 mL of cRPMI to increase the total volume to 20 mL.
• Count the number of viable white blood cells using trypan blue exclusion of dead cells on a hemocytometer slide.
• Culture the cells at a concentration of 8 × 10 6 cells per mL in cRPMI in 175 cm 2 flasks (up to 30 mL total volume per flask). Add in the required amount of initiating protein/peptide to restimulate the cells (see Table 4 ; Notes 8 – 9 ). Add in 10 ng/mL IL-12 if necessary (see Table 4 ; Notes 10 – 11 ). Incubate for 72 h in a 37 °C, 6 % CO 2 tissue culture incubator ( see Note 12 ).
• Harvest the cells by transferring into 50 mL conical tubes and centrifuging at 300 × g for 15 min. Count the number of live blasts by using trypan blue exclusion. Blasts will appear larger than other leukocytes in the culture and should comprise at least 10–30 % of the culture.
• Wash the cells twice in PBS and raise to the required concentration in PBS. Inject the recommended number of blasts in a final volume of 200 µL (see Table 4 ; Note 13 ) into recipient mice using a 1 mL syringe and 30.5 G needle via intravenous tail vein. Alternatively, blasts may be injected into the peritoneum.
• Inject 200 µL of diluted pertussis toxin (200 ng per mouse) into the peritoneal cavity or intravenously on the day of disease induction, and again 48 h after induction if necessary (refer to Table 4 ).

3.3 Isolation of CNS Infiltrating Leukocytes

• Induce deep anesthesia by injecting mice with approved anesthetic (e.g., Nembutal). Ensure deep anesthesia is achieved by testing reflex responses of the footpads.
• Draw up PBS into a 30 mL syringe with a 21.5 G needle.
• Using surgical scissors and forceps, expose the chest cavity by making an incision at the diaphragm and cutting upwards through the rib cage.
• Using the forceps, carefully hold the heart in place. Make a small incision in the right atrium of the heart.
• Immediately insert the 30 mL syringe with 21.5 G needle into the left atrium of the heart. To ensure correct placement, put a small amount of pressure on the syringe plunger. An efflux of dark red blood should immediately flow from the right atrium ( see Note 14 ). Slowly perfuse the animal by continuing to apply pressure on the plunger ( see Notes 15 , 16 ).
• Remove the head of the mouse by carefully cutting under the skull, through the neck. Using scissors or a scalpel, cut the skin and cutaneous muscle by making a single incision along the midline length of the head, from the posterior aspect of the skull to the nose of the mouse. Using the forceps, fold back the two flaps of skin to expose the skull.
• Carefully cut through the top of the skull along the midline, from the posterior aspect of the skull to the nose. Make two incisions from the posterior aspect of the skull to the eye sockets, and use the forceps to remove the pieces of skull from the top of the head. Carefully remove the brain from the cranial cavity and place in a 50 mL tube containing 20 mL of PBS.
• Using a scalpel or scissors, make a long incision through the midline of the mouse (from the neck to the base of the tail). Peel the skin back to expose the spinal column. Identify where the base of the spinal column attaches to the pelvis, and make a perpendicular cut through the spine at this point. Cut along each side of the column to the neck to remove the column.
• Attach an 18 G needle to a 5 mL syringe that has been filled with PBS. Hold the column in a petri dish with forceps, and insert the needle into the spinal column at the caudal end. Push the plunger of the syringe to expel PBS into the spinal column. Initial resistance should be felt, followed by release, in which the spinal cord will be flushed from the column. Transfer spinal cord into a 50 mL tube containing 20 mL PBS.
• Carefully pass the brain and spinal cord through the stainless steel wire mesh in the 60 mm petri dishes using the plungers taken from 10 mL syringes. Rinse all tissue off the mesh with additional PBS. Transfer back into 50 mL tubes and centrifuge for 20 min at 300 × g at 4 °C. Remove the supernatant.
• Prepare digestion enzyme mix by adding 1 g DNase I (50 µg/mL final concentration) and 800 U Liberase (40 U/mL final concentration) to 20 mL PBS ( see Note 17 ). Raise cells in 2 mL of digestion enzyme mix and incubate for 30 min at 37 °C.
• Prepare 500 mL FACS buffer by adding 10 mL FCS (2 % final concentration) and 2 mL 0.5M EDTA (2 mM final concentration) to 488 mL PBS.
• Add 30 mL FACS buffer to samples and pass through a 70 µM sieve into a new 50 mL tube. Centrifuge for 10 min at 300 × g at 4 °C. Wash cells again twice in FACS buffer.
• Prepare 30 % Percoll by adding 5 mL 10 % HBSS (with phenol red) and 15 mL Percoll to 35 mL of water. Prepare 70 % Percoll by adding 5 mL 10 % HBSS (without phenol red) and 35 mL Percoll to 15 mL of water ( see Note 18 ).
• Raise the CNS samples in 5mL of 30 % Percoll and transfer to a 15 mL tube. Underlay samples with 5 mL of 70 % Percoll. Centrifuge the samples at 1,000 × g for 25 min at room temperature with no brake.
• Using a pipette, remove the myelin layer from the top of the gradient. Using a new pipette tip, collect the mononuclear cells at the interface between the 30 and 70 % Percoll interface and transfer into a new 15 mL tube containing 5 mL FACS buffer. Centrifuge for 10 min at 300 × g at 4 °C. Wash cells again twice in FACS buffer.
• Count live cells using trypan blue exclusion.
• Single cells can now be further processed for the desired assay, e.g., flow cytometric analysis.

Abbreviations

1 It is recommended to shave the back of the mice at least 24 h before inducing disease. The animals will then be easier to handle on the day of induction.

2 It is critical to purchase IFA and M. tuberculosis separately and prepare CFA at a final concentration of 4 mg/mL. Commercially available CFA only contains 1 mg/mL M. tuberculosis .

3 The lowest concentration of a given myelin protein/peptide that can be used to reliably induce EAE may vary by source and batch number. The source and age of the mice used may also alter the disease course from that expected. Thus, the concentrations of protein/peptide indicated in Table 1 should be used as a guide only and should be confirmed by the individual investigator before conducting large-scale experiments.

4 CFA can be prepared up to 24 h in advance and stored it in polystyrene tubes or glass syringes at 4 °C until use.

5 To test the consistency of the emulsion, a small droplet should be expelled onto the surface of water in beaker. If the emulsion is stable, the droplet will remain in a bead on the water surface. If the droplet disperses across surface, further emulsification is required.

6 It is recommended to dissolve the lyophilized pertussis toxin in sterile PBS at least 24 h before injection and store at 4 °C until use.

7 We inject 200 ng of pertussis per mouse on day 0 and day 2 post-immunization where indicated on Table 2 . However, some papers report injection of up to 500 ng of pertussis per mouse. The individual investigator should determine the optimal dose for the protein/peptide and mouse strain used.

8 The amount of peptide needed to effectively restimulate T cells in vitro may vary by protein/peptide source and batch number. The source and age of the mice used may also affect the amount of protein/peptide needed for effective restimulation. Thus, the concentrations of protein/peptide indicated in Table 4 should be used as a guide only and should be confirmed by the individual investigator before conducting large-scale experiments.

9 Con A may be used at a dose of 1 µg/mL to stimulate T cells for 48 h in vitro in place of specific myelin protein/peptide for the induction of adoptive EAE. Con A activation, however, will reduce the frequency of myelin epitope-specific T cells in the culture, thus an increased number of total cells will need to be injected into recipient mice. The individual investigator will need to titrate these numbers in vivo to determine the lowest number of cells required to achieve severe and reliable EAE.

10 In addition to IL-12, IL-2 may also be added into the media at a concentration of 10 ng/mL to enhance T cell activation and proliferation. The individual investigator should confirm the optimum concentration of these cytokines before conducting large-scale experiments.

11 IL-23 may be added to the media at a concentration of 10 ng/mL to induce a Th17-skewed T cell phenotype. The individual investigator should confirm the optimum concentration of IL-23 before conducting large-scale experiments.

12 We typically incubate cells in vitro for 3 days before transferring into recipients. Most papers report incubation periods of 3–4 days. The individual investigator should determine the optimal incubation time for the protein/peptide and mouse strain used.

13 The lowest number of cells that are needed to reliably induce EAE may vary by myelin protein/peptide source and batch number. The source and age of the mice used may also alter the disease course from that expected. The numbers listed in Table 4 should be used as a guide only. The individual investigator will need to titrate these numbers in vivo to determine the lowest number of cells required to achieve severe and reliable EAE.

14 If the needle is not placed correctly, dark red blood will not flow from the right atrium and the lungs may inflate. Loss of red coloration of the liver is a good indicator of correct perfusion. The authors recommend practicing this technique before conducting large-scale experiments.

15 Perfusions should be conducted slowly (over a period of at least 3–5 min per mouse) to avoid tissue damage.

16 If the mouse is not well perfused after the initial procedure, the syringe may be refilled with 30 mL of PBS and perfusion repeated.

17 The protocol listed here using Liberase and DNase is optimized for the isolation of total leukocytes. Different enzymatic digestion may be performed on the CNS tissues to isolate different target cell populations. For example, we have found that digesting the CNS using Accutase (Millipore) in place of Liberase and DNase is optimal for the isolation of oligodendrocyte progenitor cell isolation.

18 The use of HBSS with and without phenol red for the 30 % Percoll and 70 % Percoll solutions, respectively, will increase the ease of which to see the interface between the two gradients and identify the mononuclear cell layer here.

Antibodies From Long COVID Patients Provide Clues to Autoimmunity Hypothesis

BY ISABELLA BACKMAN August 5, 2024

Promising new research supports that autoimmunity—in which the immune system targets its own body—may contribute to Long COVID symptoms in some patients.

As covered previously in this blog, researchers have several hypotheses to explain what causes Long COVID, including lingering viral remnants, the reactivation of latent viruses, tissue damage, and autoimmunity.

Now, in a recent study , when researchers gave healthy mice antibodies from patients with Long COVID, some of the animals began showing Long COVID symptoms—specifically heightened pain sensitivity and dizziness. It is among the first studies to offer enticing evidence for the autoimmunity hypothesis. The research was led by Akiko Iwasaki, PhD , Sterling Professor of Immunobiology at Yale School of Medicine (YSM).

“We believe this is a big step forward in trying to understand and provide treatment to patients with this subset of Long COVID,” Iwasaki said.

Iwasaki zeroed in on autoimmunity in this study for several reasons. First, Long COVID’s persistent nature suggested that a chronic triggering of the immune system might be at play. Second, women between ages 30 and 50, who are most susceptible to autoimmune diseases, are also at a heightened risk for Long COVID. Finally, some of Iwasaki’s previous research had detected heightened levels of antibodies in people infected with SARS-CoV-2.

Mice given antibodies show signs of Long COVID symptoms

Iwasaki’s team isolated antibodies from blood samples obtained from the Mount Sinai-Yale Long COVID study . They transferred these antibodies into mice and then conducted multiple experiments designed to look for changes in behavior that may indicate the presence of specific symptoms. For many of these experiments, mice that received antibodies [the experimental group] behaved no differently than mice that had not [the control group].

However, a few experiments revealed striking changes in the behavior of the experimental mice. These included:

• Pain sensitivity test: Some experimental mice were quicker to react after being placed on a heated plate.
• Coordination and balance test: Some experimental mice struggled to balance on a rotarod (rotating rod) compared to control mice.
• Grip strength test: Some of the experimental mice applied less force with their paws.

Among the mice that showed behavioral changes, the researchers identified which patients their antibodies came from and what symptoms they had experienced. Interestingly, of the mice that showed heightened pain, 85% received antibodies from patients that reported pain as one of their Long COVID symptoms. Additionally, 89% of mice that had demonstrated loss of balance and coordination on the rotarod test had received antibodies from patients who reported dizziness. Furthermore, 91% of mice that showed reduced strength and muscle weakness received antibodies from patients who reported headache and 55% from patients who reported tinnitus. More research is needed to better understand this correlation.

The autoimmunity hypothesis has recently been further supported by a research group in the Netherlands led by Jeroen den Dunnen, DRS , associate professor at Amsterdam University Medical Center, which also found a link between patients’ Long COVID antibodies and corresponding symptoms in mice.

Treatments for autoimmunity may help some Long COVID patients

Diagnosing and treating Long COVID requires doctors to understand what causes the disease. The new study suggests that treatments targeting autoimmunity, such as B cell depletion therapy or plasmapheresis, might alleviate symptoms in some patients by removing the disease-causing antibodies.

Intravenous immunoglobulin (IVIg) is another therapy used for treating autoimmune diseases like lupus in which patients receive antibodies from healthy donors. While its exact mechanism is still unclear, the treatment can help modulate the immune system and reduce inflammation. Could this treatment help cases of Long COVID that are caused by autoimmunity?

A 2024 study led by Lindsey McAlpine, MD , instructor at YSM and first author, and Serena Spudich, MD , Gilbert H. Glaser Professor of Neurology at YSM and principal investigator, found that IVIg might help improve small fiber neuropathy—a condition associated with numbness or painful sensations in the hands and feet—caused by Long COVID. Iwasaki is hopeful that future clinical trials might reveal the benefits of this treatment in helping some of the other painful symptoms of the diseases.

Other drugs are also in the pipeline, such as FcRn inhibitors. FcRn is a receptor that binds to antibodies and recycles them. Blocking this receptor could help bring down levels of circulating antibodies in the blood. An FcRn receptor was recently approved by the FDA for treating myasthenia gravis, another kind of autoimmune disease.

The study could also help researchers create diagnostic tools for evaluating which patients have Long COVID induced by autoimmunity so that doctors can identify who is most likely to benefit from treatments such as these.

Iwasaki plans to continue researching why and how autoantibodies might cause Long COVID, as well as conduct randomized clinical trials on promising treatments. She is also conducting similar antibody transfer studies in other post-acute infection syndromes, such as myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS).

In the meantime, she is excited about her team’s promising results. “Seeing this one-to-one correlation of antibodies that cause pain from patients who reported pain is really gratifying to me as it suggests a causal link,” she says. “It’s a first step, but I think it’s a big one.”

Isabella Backman is associate editor and writer at Yale School of Medicine.

The last word by Lisa Sanders, MD:

I am very excited by this research, which suggests that at least some of the symptoms of Long COVID are driven by autoimmunity. If so, then this suggests that there may be a way to test for some versions of Long COVID. And if we could identify the patients who have an autoimmune-driven disease, we have treatments to try that have been used with success in other autoimmune diseases. Many of the autoimmune diseases are treated with medications that suppress the immune system. These are powerful medicines that can leave an individual at risk for infection, so they must be thoughtfully applied to patients with evidence of immune system involvement.

I feel as though every blog post here ends with the possibility of better testing and better treatment, but what makes this different is that it points in a very specific direction and leads to the kind of specific questions that help get to useful answers. Which antibodies are involved? Which cells? And finally, can we develop treatments that are specific to those antibodies or to their targets? These are exciting questions, which will, I hope, lead to useful answers.

Read other installments of Long COVID Dispatches here .

If you’d like to share your experience with Long COVID for possible use in a future post (under a pseudonym), write to us at: [email protected]

Information provided in Yale Medicine content is for general informational purposes only. It should never be used as a substitute for medical advice from your doctor or other qualified clinician. Always seek the individual advice of your health care provider for any questions you have regarding a medical condition.

More news from Yale Medicine