https://doi.org/10.1093/braincomms/fcad109 BRAIN COMMUNICATIONS 2023: Page 1 of 10 | 1 BRAIN COMMUNICATIONS Antibodies against the flotillin-1/2 complex in patients with multiple sclerosis Cinta Lleixà,1,2,† Marta Caballero-Ávila,1,† Elba Pascual-Goñi,1 Lorena Martín-Aguilar,1 Nuria Vidal,1 Clara Tejada,1 Eduardo Valdés-Hevia,1 Elisa Zárate,1 Ana Vesperinas,1 Roger Collet,1 Teresa Franco-Leyva,3 Laura Martínez-Martínez,3 Esther Moga,3 Elena Cortés-Vicente,1,2 Ricard Rojas-García,1,2 Beatriz Gómez-Anson,4 Anna Gil,5 Cristina González-Mingot,5 Luis Brieva,5 Sergio Martínez-Yélamos6 and Luis Querol1,2 † These authors contributed equally to this work. Multiple sclerosis is a tissue-specific autoimmune disease of the central nervous system in which the antigen(s) remains elusive. Antibodies targeting the flotillin-1/2 complex have been described in 1–2% of the patients in a recent study. Other candidate antigens as anoctamin-2 or neurofascin-155 have been previously described in multiple sclerosis patients, although their clinical relevance re­ mains uncertain. Our study aims to analyse the frequency and clinical relevance of antibodies against neurofascin-155, anoctamin-2 and flotillin-1/2 complex in multiple sclerosis. Serum (n = 252) and CSF (n = 50) samples from 282 multiple sclerosis patients were included in the study. The control group was composed of 260 serum samples (71 healthy donors and 189 with other neuroinflam­ matory disorders). Anti-flotillin-1/2, anti-anoctamin-2 and anti-neurofascin-155 antibodies were tested by cell-based assays using transfected cells. We identified six multiple sclerosis patients with antibodies against the flotillin-1/2 complex (2.1%) and one multiple sclerosis patient with antibodies against anoctamin-2 (0.35%). All multiple sclerosis patients were negative for anti-neurofascin-155 antibodies. Three of the anti-flotillin-1/2 positive patients showed anti-flotillin-1/2 positivity in other serum samples extracted at dif­ ferent moments of their disease. Immunoglobulin G subclasses of anti-flotillin-1/2 antibodies were predominantly one and three. We confirm that antibodies targeting the flotillin-1/2 complex are present in a subgroup of patients with multiple sclerosis. Further studies are needed to understand the clinical and pathological relevance of anti-flotillin-1/2 autoantibodies in multiple sclerosis. 1 Neuromuscular Diseases Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Institut d’Investigació Biomèdica Sant Pau, Universitat Autònoma de Barcelona, 08041 Barcelona, Spain 2 Neuromuscular Diseases, Centro para la Investigación Biomédica en Red en Enfermedades Raras (CIBERER), 28029 Madrid, Spain 3 Immunology Department, Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, 08025 Barcelona, Spain 4 Department of Radiology, Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, 08025 Barcelona, Spain 5 Department of Neurology, Hospital Arnau de Vilanova, 25198 Lleida, Spain 6 Multiple Sclerosis Unit, Department of Neurology, Hospital Universitari de Bellvitge, Institut d’Investigació Biomèdica de Bellvitge (IDIBELL), Departament de Ciències Clíniques, Facultat de Medicina, Universitat de Barcelona, 08907 Barcelona, Spain Correspondence to: Cinta Lleixà Hospital de la Santa Creu i Sant Pau Institut d’Investigació Biomèdica Sant Pau Sant Quintí 77, 08041 Barcelona, Spain E-mail: mlleixa@santpau.cat Keywords: autoantibodies; multiple sclerosis; antigens Received September 14, 2022. Revised January 16, 2023. Accepted April 03, 2023. Advance access publication April 4, 2023 © The Author(s) 2023. Published by Oxford University Press on behalf of the Guarantors of Brain. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 2 | BRAIN COMMUNICATIONS 2023: Page 2 of 10 C. Lleixà et al. Abbreviations: ALS = amyotrophic lateral sclerosis; ANO2 = anoctamin-2; CIDP = chronic inflammatory demyelinating polyneuropathy; CMT = Charcot–Marie–Tooth neuropathy; CNS = central nervous system; CSF = cerebrospinal fluid; DMEM = Dulbecco’s Modified Eagle Medium; FITC = fluorescein isothiocyanate; FLOT1/2 = flotillin-1/2; GBS = Guillain–Barré syndrome; GlialCAM/HEPACAM = hepatocyte cell adhesion molecule; HEK293 = human embryonic kidney 293; ICC = immunocytochemistry; IgG = immunoglobulin G; MG = myasthenia gravis; MMN = multifocal motor neuropathy; NF155 = neurofascin-155; PPMS = primary-progressive multiple sclerosis; RRMS = relapsing–remitting multiple sclerosis; SPMS = secondary-progressive multiple sclerosis Graphical Abstract The only validated laboratory biomarker for multiple Introduction sclerosis diagnosis is the detection of oligoclonal immuno­ Multiple sclerosis is an autoimmune and neurodegenerative globulin G (IgG) bands in the CSF3 that support a prominent disease of the central nervous system that has a heteroge­ role of humoral immunity in this disease.4 Moreover, the epi­ neous clinical presentation and course. The aetiology of mul­ demiological role of the Epstein–Barr virus,5 the presence of tiple sclerosis is likely multifactorial; evidence suggests that B cells and antibody-secreting plasma cells in CNS demyeli­ various genetic and environmental risk factors contribute nated lesions and the efficacy of B cell-depleting treatments6 to disease manifestation.1 Pathologically, multiple sclerosis provide strong evidence for the involvement of B cells and is characterized by demyelination, multifocal inflammation, antibodies in multiple sclerosis pathogenesis.7,8 reactive gliosis and oligodendrocyte and axonal loss, leading Despite exhaustive research in autoantibody discovery in to neurological disability.2 multiple sclerosis, the target antigen(s) of the humoral Antibodies against FLOT1/2 in multiple sclerosis BRAIN COMMUNICATIONS 2023: Page 3 of 10 | 3 immune response in multiple sclerosis patients remains elu­ vectors encoding human NF155 (EX-Z7183-M02, sive. A long list of candidate antigens have been found,8 in­ Genecopoeia, MD, USA) or ANO2 (RC222601, Origene, cluding the glial potassium channel KIR4.1,9 the flotillin-1/2 MD, USA) with Lipofectamine 2000 (Invitrogen, CA, complex (FLOT1/2),10 anoctamin-2 (ANO2)11 or USA). The day after transfection, cells were fixed with 4% neurofascin-155 (NF155)12,13 and most recently, antibodies paraformaldehyde (Santa Cruz Biotechnology, DA, USA) targeting hepatocyte cell adhesion molecule (GlialCAM/ and blocked for 1 hour with 5% goat serum. Double immuno­ HEPACAM);14 although most of them have not been repli­ cytochemistry (ICC) was performed using patients’ sera cated in independent cohorts, and their clinical relevance re­ (1:100) and chicken monoclonal antibody against pan-NF mains uncertain.15,16 (R&D systems, MI, USA) at 1:1000 or rabbit monoclonal Our study aims to analyse the presence of antibodies antibody against ANO2 (Origene) at 1:500. Goat anti-human against NF155, ANO2 and FLOT-1/2 complex in our mul­ IgG AF594 and goat anti-chicken IgG AF488 (for NF155) or tiple sclerosis cohort, and to analyse if there are clinical fea­ goat anti-rabbit IgG AF488 (for ANO2) (Molecular Probes, tures associated with these autoantibodies that could suggest Eugene, OR, USA) were used as secondary antibodies at that patients harbouring any of these antibodies constitute a 1:1000 concentration. Finally, culture slides were mounted differentiated multiple sclerosis subset. with Fluoromount medium (Invitrogen) and examined by two independent observers. Images were obtained with an Olympus BX51 Fluorescence Microscope (Olympus Materials and methods Corporation, Tokyo, Japan). Patients and protocol approvals Testing for anti-FLOT-1/2 antibodies Serum (n = 252) and CSF (n = 50) samples from 282 patients Anti-flotillin-1/2 antibodies were also tested by cell-based as­ fulfilling the McDonald diagnostic criteria for multiple scler­ says using HEK293 cells co-transfected with mammalian- osis17 and followed-up in Hospital de la Santa Creu i Sant expression vectors encoding human FLOT1 and FLOT2. Pau (n = 188) and Hospital Arnau de Vilanova (n = 94) Our screening study was organized in two steps. First, all were included in the initial antibody screening. The control the samples were analysed for IgG reactivity on commercial group was composed of serum samples from 71 healthy do­ slides containing biochips with cells co-expressing FLOT1 nors and 189 patients with other neuroinflammatory and de­ and FLOT2 proteins (Euroimmun, Lübeck, Germany). generative disorders: chronic inflammatory demyelinating Each biochip contains one area with transfected cells and an­ polyneuropathy (CIDP, n = 30), Guillain–Barré syndrome other with non-transfected cells. Slides were incubated with (GBS, n = 26), myasthenia gravis (MG, n = 44), amyotrophic patient’s sera diluted 1:10 in PBS-Tween 0,2% at room tem­ lateral sclerosis (ALS, n = 44), Charcot–Marie–Tooth neur­ perature for 30 min; followed by incubation with polyclonal opathy (CMT, n = 30) and multifocal motor neuropathy goat anti-human IgG fluorescein isothiocyanate (FITC) (MMN, n = 15) obtained from our Unit Biobank (code (Euroimmun) at room temperature for 30 min. The slides C.0002365). Serum and CSF samples were obtained, pro­ were then mounted with Fluoromount medium and exam­ cessed and frozen at −80°C until needed. ined by two independent observers. Samples were categor­ In addition, serum samples from 49 multiple sclerosis pa­ ized based on fluorescence intensity of transfected cells in tients from Hospital de Bellvitge without oligoclonal bands comparison with non-transfected cells and control samples. in the CSF were used as confirmation cohort. Sera from patients showing strong reactivity against the Written informed consents were obtained from all sub­ commercial biochips were used for immunocytochemistry jects for sample handling and data collection. Participation over in house co-transfected cells, as explained before. in the study was conducted under a protocol approved by Briefly, mammalian expression vectors encoding human the Institutional Ethics Committee of the Hospital de la FLOT1 (SC319463, Origene) and FLOT2 (EX-H1820-M02, Santa Creu i Sant Pau. Genecopoeia) were transfected into HEK293 cells using All methods were carried out in accordance with relevant Lipofectamine 2000. Cells were then fixed with 4% parafor­ guidelines and regulations. maldehyde and blocked with 5% goat serum. Double ICC was performed using patients’ sera (1:10) and polyclonal rabbit Testing for anti-NF155 and antibodies against FLOT1 (1:100) or FLOT2 (1:250) (Sigma). Goat anti-human IgG AF594 and goat anti-rabbit IgG AF488 anti-ANO2 antibodies (Molecular Probes) were used as secondary antibodies at Anti-NF155 and anti-ANO2 antibodies were tested by cell- 1:1000 concentration. To determine the autoantibody sub­ based assays using human embryonic kidney 293 class, goat anti-human IgG was substituted by mouse anti- (HEK293) transfected cells. Cells were grown in culture human IgG1, IgG2, IgG3 or IgG4 (Southern Biotech). dishes with coverslips coated with Poly-D-lysine (Corning, Finally, culture slides were mounted with Fluoromount NY, USA), in foetal bovine serum-supplemented medium and examined by two independent observers. Dulbecco’s Modified Eagle Medium (DMEM) culture me­ Images were obtained with an Olympus BX51 dium. After 24 hours, cells were transfected overnight with Fluorescence Microscope. 4 | BRAIN COMMUNICATIONS 2023: Page 4 of 10 C. Lleixà et al. FLOT-1/2 and ANO2 (n = 239, 85%), 25 patients were diagnosed of primary- progressive multiple sclerosis (PPMS) (9%) and 18 of immunoadsorption secondary-progressive multiple sclerosis (SPMS) (6%). CSF Non-transfected, FLOT1-, FLOT2- or FLOT-1/ study was available in 234 patients, and 170 had positive oli­ 2-co-transfected HEK293 cells (for FLOT-1/2 positive pa­ goclonal IgG bands (73%). tients) and non-transfected or ANO2-transfected cells (for Demographic and clinical data of the three cohorts ana­ ANO2 positive patients) were grown in 6-well plates and lysed are shown in Table 1. fixed with 4% paraformaldehyde. Sera showing strong re­ activity in immunocytochemistry experiments were diluted Autoantibodies against NF155 and in 5% goat serum in PBS, and serially incubated for 1 hour in each of the 6 wells. Then, supernatant was collected, ANO2 and its reactivity was tested by immunocytochemistry on We tested 252 sera and 50 CSF samples from 282 patients. FLOT-1/2 co-transfected or ANO2 transfected HEK293 None of the patients included in the study had autoanti­ cells, as described above. bodies against the NF155 protein. Regarding ANO2 anti­ bodies, we only detected one positive serum (1/252, Statistical analysis 0.40%). We also tested the CSF and another serum from the same patient but extracted at different moments of the Results were analysed by GraphPad Prism v8.0 (GraphPad disease, and both samples also showed anti-ANO2 anti­ Software). Statistical comparison of proportions among bodies (Fig. 1). groups was performed using contingency analysis with the application of a two-tailed Fisher’s exact test, accepting an alpha-level <0.05 for statistical significance. Autoantibodies against the FLOT-1/2 complex Data availability We screened the presence of anti-FLOT-1/2 antibodies in 282 patients fulfilling diagnostic criteria for multiple scler­ The datasets used and/or analysed during the current study osis (252 sera and 50 CSF), and 260 controls (including are available from the corresponding author on reasonable healthy donors and patients with other neurological disor­ request. ders). From the 512 serums tested by immunocytochemistry on commercial biochips containing transfected HEK293 cells, we identified five multiple sclerosis patients with anti­ Results bodies against the flotillin-1/2 complex (5/252, 1.98%), whereas none of the control sera tested showed reactivity Baseline characteristics in cell-based assays (0/260, 0%) (Fig. 2). When comparing A total of 282 patients fulfilling diagnostic criteria for mul­ the results obtained in the two groups, the differences ob­ tiple sclerosis were included in the initial antibody screening served were statistically significant (P = 0.0283). Moreover, (188 from Hospital de la Santa Creu i Sant Pau and 94 from three of the anti-FLOT1/2 positive patients showed this posi­ Hospital Arnau de Vilanova). Most patients were female tivity in serum samples extracted at different moments of (n = 202, 72%), and mean age at diagnosis was 37.19 ± their disease. 11.7 years (n = 188). Relapsing–remitting multiple sclerosis Regarding CSF samples, we detected anti-FLOT1/2 anti­ (RRMS) was the most common subtype of multiple sclerosis bodies in one more multiple sclerosis patient (1/50, 2%). In Table 1 Demographic and clinical data of MS patients of the three cohorts analysed Hospital de Sant Pau cohort Hospital Arnau de Vilanova cohort Hospital de Bellvitge cohort (n = 188) (n = 94) (n = 49) Sex Female, n (%) 137 (73%) 65 (69%) 26 (53%) Male, n (%) 51 (27%) 29 (31%) 23 (47%) MS subtype RRMS, n (%) 155 (83%) 85 (90%) 45 (92%) PPMS, n (%) 19 (10%) 5 (5%) 0 SPMS, n (%) 14 (7%) 4 (5%) 4 (8%) OCB Positive, n (%) 107 (76%) 63 (68%) 0 (0%) Negative, n (%) 34 (24%) 30 (35%) 49 (100%) MS, multiple sclerosis; RRMS, relapsing–remitting multiple sclerosis; PPMS, primary-progressive multiple sclerosis; SPMS, secondary-progressive multiple sclerosis; OCB, oligoclonal IgG bands. Antibodies against FLOT1/2 in multiple sclerosis BRAIN COMMUNICATIONS 2023: Page 5 of 10 | 5 Figure 1 Immunocytochemistry of anti-ANO2 antibodies. HEK293 cells transfected with mammalian-expression vectors encoding human ANO2 using Lipofectamine 2000 (A–F) or HEK293 cells non-transfected (G–I); double-stained with serum (B, E, H) and with commercial antibody against ANO2 (A, D, G). MS patient’s IgG bind to transfected cells (B) and colocalize with ANO2 ab (C); in contrast with the healthy control (E) that does not show any reactivity against ANO2 antibodies. this case, the autoantibodies were not found in a serum sam­ multiple sclerosis and a possible correlation between the ab­ ple of the same patient that was extracted at a different mo­ sence of IgG oligoclonal bands in CSF and the presence of ment of the disease. anti-FLOT-1/2 antibodies, we tested these autoantibodies When assessing the presence of autoantibodies against in a cohort from Hospital de Bellvitge including 49 serum FLOT-1/2 by immunocytochemistry over ‘in house’ co- samples from patients fulfilling multiple sclerosis criteria in transfected HEK293 cells, we confirmed the positivity of which oligoclonal bands were not detected. However, we the five serum and one CSF samples in which we had previ­ did not detect any other patient with antibodies against the ously detected anti-FLOT1/2 antibodies with commercial FLOT-1/2 complex in this confirmatory cohort. slides (Fig. 3). The autoantibody subclass of the positive sera was as­ sessed. The anti-FLOT-1/2 antibodies were exclusively Reactivity against FLOT-1/2 and IgG1 in two patients and were consistent in other serum sam­ ANO2 antibodies was abrogated after ples extracted at different moment of the disease. Two other patients’ antibodies were IgG3. Finally, in one patient of immunoadsorption whom four samples were tested, we found that three of We performed immunoadsorption experiments incubating them had IgG1 anti-FLOT-1/2, and the remaining one had anti-FLOT1/2 positive patients’ sera with non-transfected IgG3 antibodies. HEK293 cells and with HEK cells transfected with FLOT1, We observed that in three of four patients (75%) that har­ FLOT2 or co-transfected with FLOT1/2. Reactivity against boured antibodies against the FLOT-1/2 complex, CSF IgG the FLOT-1/2 complex was lost after serum pre-adsorption oligoclonal bands were not detected. In the other two posi­ with HEK cells co-expressing FLOT1 and FLOT2, but not tive patients, oligoclonal IgG bands were not assessed. For after pre-adsorption with cells transfected with FLOT1 or that reason, to confirm FLOT-1/2 as a target antigen in FLOT2 alone, or with non-transfected HEK cells (Fig.4). 6 | BRAIN COMMUNICATIONS 2023: Page 6 of 10 C. Lleixà et al. Figure 2 Immunocytochemistry of anti-FLOT1/2 antibodies (commercial slides). Commercial biochips (Euroimmun) containing FLOT-1/2 transfected HEK293 cells (A, C) or non-transfected HEK293 cells (B, D); incubated with serum from MS Patient 1 (A, B) and from a healthy donor (C, D). Patient 1 showed strong IgG reactivity against co-transfected cells (A) in comparison with non-transfected cells (B) and with the negative control (C). Figure 3 Immunocytochemistry of anti-FLOT1/2 antibodies (in house transfection). HEK293 cells co-transfected with mammalian-expression vectors encoding human FLOT1 and FLOT2 using Lipofectamine 2000; double-stained with serum (B, E) and with commercial antibody against FLOT1 (A, D). MS patient’s IgG bind to co-transfected cells (B) and colocalize with FLOT1 ab (C); in contrast with the healthy control (E) that doesn’t show any reactivity against FLOT-1/2 antibodies (F). Antibodies against FLOT1/2 in multiple sclerosis BRAIN COMMUNICATIONS 2023: Page 7 of 10 | 7 Figure 4 Immunoadsorption of anti-FLOT1/2 antibodies. HEK293 cells co-transfected with mammalian-expression vectors encoding human FLOT1 and FLOT2 double-stained with serum immunoadsorbed with non-transfected cells (A), FLOT1-transfected cells (D), FLOT2-transfected cells (G) or FLOT1 and FLOT2 co-transfected cells (J); and with commercial antibody against FLOT1 (B, E, K) or FLOT2 (H). Reactivity against the FLOT-1/2 complex was lost after serum pre-adsorption with HEK cells co-expressing FLOT1 and FLOT2 (L), but not after pre-adsorption with cells transfected with FLOT1 (F) or FLOT2 (I) alone or with non-transfected HEK cells (C). We also immunoadsorbed the anti-ANO2 positive patient fulfilled criteria of RRMS while one patient of PPMS. Four serum with non-transfected and ANO2-transfected HEK293 patients were female (67%). Only one patient had a history cells; and the reactivity against ANO2 was lost after incuba­ of optic neuritis. In two of the three patients treated with tion with cells expressing ANO2, but not after incubation anti-CD20 therapies (Patient 3 and Patient 5), positivity with non-transfected cells. against FLOT-1/2 persisted in serum after starting the treat­ ment (16 months in Patient 3 and 2 months in Patient 5). Clinical features Clinical and laboratory features of patients with Discussion anti-FLOT-1/2 and anti-ANO2 antibodies are summarized We have identified six patients with autoantibodies against in Table 2. Five out of six anti-FLOT1/2 positive patients the FLOT-1/2 complex and one patient with autoantibodies 8 | BRAIN COMMUNICATIONS 2023: Page 8 of 10 C. Lleixà et al. Table 2 Clinical and laboratory features of patients with anti-FLOT-1/2 and anti-ANO2 antibodies Anti-FLOT-1/2 Anti-ANO2 Positive patients Positive patients P1 P2 P3 P4 P5 P6 P7 Gender Male Male Female Female Female Female Male Years at diagnosis 22 35 36 62 26 49 36 Disease duration at first 34 6 5 6 13 0 13 sampling (years) Optic neuritis Yes No No No No No No Diagnosis RRMS RRMS RRMS PPMS RRMS RRMS RRMS EDSS 4 4 2 6.5 1 1 1 Treatment Dymethilfumarate Natalizumab Rituximab Ocrelizumab Ocrelizumab None Fingolimod Previous treatment Interferon Interferon Interferon Interferon Interferon None Interferon Laquinimod Glatiramer Glatiramer Teriflunomide Glatiramer acetate acetate acetate Fingolimod Teriflunomide Natalizumab Fingolimod OCB in CSF na na Negative Positive Negative Negative Positive Number of samples with 1 2 4 1 4 − − anti-FLOT-1/2 Serum titre 1/1250 1/1250 >1/6250 1/50 >1/6250 − − Serum IgG subclass IgG3 IgG1 IgG1, IgG3 IgG3 IgG1 − − CSF titre na na na na na + − RRMS: relapsing–remitting multiple sclerosis; PPMS: primary progressive Multiple Sclerosis; EDSS: Expanded Disability Status Scale; OBC: oligoclonal bands, na = not available. against ANO2 protein in a study to screen for the presence of diseases are difficult to elucidate as only isolated cases are re­ antibodies against NF155, ANO2 and FLOT1/2, previously ported and further reports would be needed. These reports described as multiple sclerosis antigens. could also suggest that anti-FLOT1/2 antibodies may arise In 2017, Hahn et al.10 published a report describing the as a secondary immune response triggered upon brain dam­ presence of antibodies against the flotillin-1/2 complex in age in predisposed individuals. This is the reason why we in­ ∼1–2% of the multiple sclerosis patients they studied. In cluded controls with other autoimmune (MG, CIDP, MMN our cohort, multiple sclerosis patients with anti-FLOT-1/2 and GBS) and degenerative disorders (ALS and CMT). autoantibodies in serum are found in 2.1% of the patients. Flotillin-1 and flotillin-2 are homologous proteins that are Despite anti-FLOT1/2 antibodies are very infrequent in found in lipid raft microdomains at the plasma membrane.20 multiple sclerosis, we did not detect any positive patient in Both flotillins are preferentially associated with each other in the control group (that included 260 sera from healthy do­ hetero-oligomeric complexes (FLOT-1/2) ubiquitously ex­ nors and patients with other neurological disorders). This pressed.21 They are involved in axon regeneration and neur­ suggests that anti-FLOT-1/2 antibodies are specific of mul­ onal differentiation (especially in the optic nerve where these tiple sclerosis. The presence of these antibodies could define proteins are upregulated), endocytosis, T-lymphocyte activa­ a rare subset of multiple sclerosis patients, as occurs in other tion and membrane protein recruitment.22 One possible ex­ neuroinflammatory pathologies; however, despite our work planation for the pathogenicity of autoantibodies against validates the initial description, the clinical and pathophysio­ flotillin 1/2 could be that they might impair the recycling of logical relevance of these autoantibodies in multiple sclerosis the T-cell receptors (important for activating T-cells),23 remains to be elucidated in autoantibody passive-transfer and this failing T-cell activation might lead to cell death caus­ animal models, and larger prospective series are needed to ing CNS autoimmunity.24 characterize phenotypically anti-FLOT1/2 positive patients. In our cohort, no anti-NF155 antibodies were found in the Data about the presence of flotillin antibodies in other 282 multiple sclerosis patients. The association between the neurological disorders are scarce, with only a few case re­ anti-NF155 antibodies and a subgroup of patients with ports published. A case of limbic encephalitis with positive CIDP is well established25, and some patients with combined anti-FLOT-1/2 antibodies in serum and CSF is reported.18 central and peripheral demyelination are also positive to Clinical and radiological findings were compatible with the anti-NF155 antibodies.26 In pure CNS demyelinating disor­ diagnosis of limbic encephalitis, and patient responded to ders such as multiple sclerosis, the proportion of patients with corticosteroids, which supported the autoimmune aetiology. anti-NF155 antibodies differs significantly in different pub­ In neurodegenerative disorders, a case of atypical dementia lished articles. Anti-NF155 antibodies were detected in 2% associated with anti-FLOT-1/2 antibodies is also pub­ of the 243 multiple sclerosis patients tested in the largest study lished,19 without other similar cases reported. The relevance assessing these autoantibodies in multiple sclerosis.12 and pathogenicity of these antibodies in this group of However, in a previous study, anti-NF155 reactivity was Antibodies against FLOT1/2 in multiple sclerosis BRAIN COMMUNICATIONS 2023: Page 9 of 10 | 9 higher than controls in a third of 26 multiple sclerosis pa­ Sanofi-Genzyme and Roche and is Principal Investigator tients;27 while in another publication with samples from 20 for UCB’s CIDP01 trial. multiple sclerosis patients, the investigators did not find anti­ bodies against NF155.28 Our report suggests that anti-NF155 antibodies in multiple sclerosis are, at best, infrequent. The presence of anti-ANO2 antibodies has been anecdotic in our cohort, being present in only one multiple sclerosis pa­ References 1. Nylander A, Hafler DA. Review series multiple sclerosis. J Clin tient. Anoctamin-2 is a calcium-activated chloride-channel Invest. 2012;122(4):1180-1188. protein, also named as transmembrane protein 16B which 2. Reich D. Multiple sclerosis. Nejm. 2017;176(3):139-148. is predominantly expressed in neuronal and muscle tissue.29 3. Gastaldi M, Zardini E, Leante R, et al. Cerebrospinal fluid analysis Antibodies against ANO2 in multiple sclerosis patients have and the determination of oligoclonal bands. Neurol Sci. 2017;38: been widely studied by Ayoglu et al.11,30 They used a non- 217-224. 4. D’Ambrosio A, Pontecorvo S, Colasanti T, et al. Peripheral blood conformational antigen array to discover the presence of biomarkers in multiple sclerosis [internet]. Autoimmun Rev. anti-ANO2 antibodies in the 15.5% of a 1063 multiple scler­ 2015;14(12):1097-1110. osis patients’ cohort. The expression and potential role of 5. Bjornevik K, Cortese M, Healy BC, et al. Longitudinal analysis re­ anti-ANO2 antibodies in multiple sclerosis remain to be elu­ veals high prevalence of Epstein–Barr virus associated with multiple cidated, but it seems that immunohistochemistry analysis sclerosis. Science (80-). 2022;375(6578):296-301. shows a clear increase in ANO2 intensity in the proximity 6. Hauser SL, Waubant E, Arnold DL, et al. B-cell depletion with ritux­ imab in relapsing–remitting multiple sclerosis. N Engl J Med. 2008; of multiple sclerosis lesions.11 358(7):676-688. In conclusion, although we found six patients with auto­ 7. Rahmanzadeh R, Weber MS, Brück W, et al. B cells in multiple antibodies against the FLOT-1/2 complex and one patient sclerosis therapy—A comprehensive review. Acta Neurol Scand. with autoantibodies against ANO2 protein, it is not clear if 2018;137(6):544-556. they associate with a specific disease subset or whether these 8. Comabella M, Khoury SJ. Immunopathogenesis of multiple scler­ osis [internet]. Clin Immunol. 2012;142(1):2-8. antibodies play a role in multiple sclerosis pathogenesis or re­ 9. Navas-Madroñal M, Valero-Mut A, Martínez-Zapata MJ, et al. present a non-pathogenic epiphenomenon. For these reasons, Absence of antibodies against KIR4.1 in multiple sclerosis: A three- further research in larger cohorts is needed to validate these technique approach and systematic review. PLoS One. 2017;12(4): target antigens as biomarkers for multiple sclerosis. 1-13. 10. Hahn S, Trendelenburg G, Scharf M, et al. Identification of the flotillin-1/2 heterocomplex as a target of autoantibodies in bona fide multiple sclerosis. J Neuroinflammation. 2017;14(1):1-10. Acknowledgements 11. Ayoglu B, Mitsios N, Kockum I, et al. Anoctamin 2 identified as an autoimmune target in multiple sclerosis. Proc Natl Acad Sci U S A. The authors would like to acknowledge the Department of 2016;113(8):2188-2193. Medicine at the Universitat Autònoma de Barcelona. We 12. Stich O, Perera S, Berger B, et al. Prevalence of neurofascin-155 anti­ also would like to thank all our patients for their support bodies in patients with multiple sclerosis [internet]. J Neurol Sci. and collaboration. We want to particularly acknowledge the 2016;364:29-32. 13. Derfuss T, Linington C, Hohlfeld R, Meinl E. Axo-glial antigens as Biobank HUB-ICO-IDIBELL (PT17/0015/0024) integrated targets in multiple sclerosis: Implications for axonal and grey matter in the Spanish Biobank Network for their collaboration. injury. J Mol Med. 2010;88(8):753-761. 14. Lanz T V, Brewer RC, Ho PP, et al. Clonally expanded B cells in multiple sclerosis bind EBV EBNA1 and GlialCAM. Nature. Funding 2022;603(7900):321-327. 15. Owens GP, Bennett JL, Lassmann H, et al. Antibodies produced by This work is supported by Roche Spain. M.C.A. was sup­ clonally expanded plasma cells in multiple sclerosis cerebrospinal fluid. Ann Neurol. 2009;65(6):639-649. ported by a personal Rio Hortega grant CM21/00101. 16. DeMarshall C, Goldwaser EL, Sarkar A, et al. Autoantibodies L.M.A. was supported by a personal Juan Rodés grant as diagnostic biomarkers for the detection and subtyping of JR21/00060. L.Q. was supported by a personal clinical in­ multiple sclerosis [internet]. J Neuroimmunol. 2017;309­ tensification grant INT20/00080. (April):51-57. 17. Thompson AJ, Banwell BL, Barkhof F, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol. 2018;17(2):162-173. Competing interests 18. Videira G, Albuquerque L, Oliveira V, et al. Limbic encephalitis with positivity for anti-flotillin antibodies in serum and CSF. L.Q. received research grants from Instituto de Salud Carlos Neurol Clin Pract. 2021;11(1):e8-e10. III—Ministry of Economy and Innovation (Spain), 19. Hansen N, Bartels C, Stöcker W, et al. Case report: Anti-flotillin 1/2 CIBERER, Fundació La Marató, GBS-CIDP Foundation autoantibody-associated atypical dementia. Front Psychiatry. 2021; International, UCB and Grifols, received speaker or expert 12(June):1-5. 20. Zhao F, Zhang J, Liu YS, et al. Research advances on flotillins. Virol testimony honoraria from CSL Behring, Novartis, J. 2011;8:2-7. Sanofi-Genzyme, Merck, Annexon, Alnylam, Biogen, 21. Babuke T, Ruonala M, Meister M, et al. Hetero-oligomerization of Janssen, Lundbeck, ArgenX, UCB, LFB, Octapharma and reggie-1/flotillin-2 and reggie-2/flotillin-1 is required for their endo­ Roche, serves at Clinical Trial Steering Committee for cytosis. Cell Signal. 2009;21(8):1287-1297. 10 | BRAIN COMMUNICATIONS 2023: Page 10 of 10 C. Lleixà et al. 22. Langhorst MF, Reuter A, Stuermer CAO. Scaffolding microdo­ 26. Kawamura N. Anti-neurofascin antibody in combined central and mains and beyond: The function of reggie/flotillin proteins. Cell peripheral demyelination. Nihon Rinsho. 2015;73:347-351. Mol Life Sci. 2005;62(19–20):2228-2240. 27. Mathey EK, Derfuss T, Storch MK, et al. Neurofascin as a novel tar­ 23. Redpath GMI, Ecker M, Kapoor-Kaushik N, et al. Flotillins pro­ get for autoantibody-mediated axonal injury. J Exp Med. 2007; mote T cell receptor sorting through a fast Rab5–Rab11 endocytic 204(10):2363-2372. recycling axis [internet]. Nat Commun. 2019;10(1):1-14. 28. Devaux JJ, Odaka M, Yuki N. Nodal proteins are target antigens in 24. Badami E, Cexus ONF, Quaratino S. Activation-induced cell death Guillain–Barré syndrome. J Peripher Nerv Syst. 2012;17(1):62-71. of self-reactive regulatory T cells drives autoimmunity. Proc Natl 29. Schreiber R, Uliyakina I, Kongsuphol P, et al. Expression and function Acad Sci U S A. 2019;116(52):26788-26797. of epithelial anoctamins. J Biol Chem. 2010;285(10):7838-7845. 25. Querol L, Devaux J, Rojas-Garcia R, Illa I. Autoantibodies in chron­ 30. Ayoglu B, Häggmark A, Khademi M, et al. Autoantibody profiling ic inflammatory neuropathies: Diagnostic and therapeutic implica­ in multiple sclerosis using arrays of human protein fragments. Mol tions. Nat Rev Neurol. 2017;13(9):533-547. Cell Proteomics. 2013;12(9):2657-2672.