In vivo assessment of cortical astrocyte network dysfunction during autoimmune demyelination

1 of 10 Journal of Neurochemistry, 2025; 169:e16305 https://doi.org/10.1111/jnc.16305 Journal of Neurochemistry SPECIAL ISSUE Astrocytes in Cognition ORIGINAL ARTICLE OPEN ACCESS In Vivo Assessment of Cortical Astrocyte Network Dysfunction During Autoimmune Demyelination: Correlation With Disease Severity A.MorenoGarcía1,2,3 | R.Serrat4,5 | F.JulioKalajzic4,5 | A.BernalChico1,2,3 | A.M.Baraibar1,2,3 | C.Matute1,6 | G.Marsicano4,5 | S.Mato1,2,3 1Department of Neurosciences, University of the Basque Country UPV/EHU, Leioa, Spain | 2Achucarro Basque Center for Neuroscience, Leioa, Spain | 3Neuroinmunology Group, Biobizkaia Health Research Institute, Barakaldo, Spain | 4Centro de Investigación Biomédica en Red Sobre Enfermedades Neurodegenerativas (CIBERNED), Madrid, Spain | 5INSERM, U1215 NeuroCentre Magendie, Bordeaux, France | 6University of Bordeaux, Bordeaux, France Correspondence: G. Marsicano ([email protected]) | S. Mato ([email protected]) Received: 11 October 2024 | Revised: 2 January 2025 | Accepted: 3 January 2025 Funding: This work was funded by the Instituto de Salud Carlos III (PI21/00629, to S.M.) and cofounded by the European Union, Basque Government (PIBA_2023_1_0046; 2023111031; IT147322, to S.M.; CannaMetHD to S.M. and G.M.; IT120319, to C.M.), ARSEP Foundation (ARSEP1310 to S.M. and G.M.), INSERM (to G.M.), the European Research Council (MiCaBra, ERC2017AdG786467; CaMeLS, ANR23CE16002201; to G.M. to G.M.), Fondation pour la Recherche Medicale (FRM, DRM20101220445 to G.M.), Region Aquitaine (CanBrain, AAP2022A202116763610 and - 17219710 to G.M.); French State/Agence Nationale de la Recherche (HippObese, ANR23ce14000403; ERANet Neuron CanShank, ANR21NEU2000104, to G.M.), La Caixa Research Health 2023 (PsychoCannabis, HR2300793, to G.M.), Ministerio de Ciencia e Innovacion(PID2019109724RB100 to C.M.), and the Postdoctoral Program of the Basque Government (to A.M.B.). Keywords: astrocyte| calcium| multiple sclerosis| somatosensory cortex ABSTRACT Cortical damage and dysfunction is a pathological hallmark of multiple sclerosis (MS) that correlates with the severity of physical and cognitive disability. Astrocytes participate in MS pathobiology through a variety of mechanisms, and abnormal astrocytic calcium signaling has been pointed as a pathogenic mechanism of cortical dysfunction in MS. However, invivo evidence supporting deregulation of astrocyte calciumdependent mechanisms in cortical MS is still limited. Here, we applied fiber photometry to the longitudinal analysis of spontaneous and sensoryevoked astrocyte network activity in the somatosensory cortex of mice in an experimental autoimmune encephalomyelitis (EAE). We found that freely moving EAE mice exhibit spontaneously occurring astrocyte calcium signals of increased duration and reduced amplitude. Concomitantly, cortical astrocytes in EAE mice responded to sensory stimulation with calcium events of decreased amplitude. The emergence of aberrant astrocyte calcium signals in the somatosensory cortex paralleled the onset of neurological symptomatology, and changes in the amplitude of both spontaneous and evoked responses were selectively correlated to the severity of neurological deficits. These results highlight the imbalance of astrocyte network activity in the brain cortex during autoimmune inflammation and further support the relevance of astrocytebased pathobiology as an underlying mechanism of cortical dysfunction in MS. A. MorenoGarcía and R. Serrat shared first authorship. G. Marsicano and S. Mato shared senior authorship. [Correction added on 14 May 2025, after first online publication: The copyright line was changed.] Abbreviations: Cat. no, Catalog number; CNS, Central nervous system; Dpi, Days postimmunization; EAE, Experimental autoimmune encephalomyelitis; GFAP, Glial fibrillary acidic protein; GFP, Green fluorescent protein; MAD, Median absolute deviation; MOG, Myelin oligodendrocyte glycoprotein; MS, Multiple sclerosis; PBS, Phosphatebuffered saline; RRID, Research Resource Identifier; SEM, Standard error of the mean. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2025 The Author(s). Journal of Neurochemistry published by John Wiley & Sons Ltd on behalf of International Society for Neurochemistry. 2 of 10 Journal of Neurochemistry, 2025 1 | Introduction Multiple sclerosis (MS) is a widely prevalent neurological disorder of the central nervous system (CNS) that involves neuroinflammatory and neurodegenerative mechanisms affecting both the brain and the spinal cord. The main neuropathological hallmarks of MS include demyelination, inflammation, astrocytic gliosis, and neurodegeneration, with irreversible neuroaxonal damage underlying disease symptomatology and clinical progression (Mahad, Trapp, and Lassmann2015). Patients with MS manifest a broad range of symptoms that include abnormalities in sensory performance, including numbness, burning and increased pain sensitivity, fatigue, spasticity, urinary incontinence, and visual defects, among others, whose neuroanatomical and cellular basis remains poorly understood. Available neuropathological and neuroimaging evidence points to the cerebral cortex as a particularly vulnerable region for MS and possible substratum of disability worsening: cortical pathology is present at the earliest stages of the disease, correlates with the severity of physical disability, and is associated with the cognitive impairments that affect 35%–90% of the patients according to variables such as age, gender, and stage of disease progression (Calabrese etal.2012; Lazzarotto etal.2024; Margoni etal.2023). Defects in somatosensory cortical activity, in particular, have been identified both in MS patients and animal models of the disease and is associated with diverse clinical signatures such as impaired limb function, fatigue, and pain (Madsen etal.2022; Latypov etal.2024; Tecchio etal.2014; Potter etal.2016). Astrocytes are a functionally complex glial cell population that supports neuronal function through a variety of mechanisms, including the release of molecules that finetune synaptic structure and function and the homeostatic control of the extracellular milieu during fastsynaptic transmission. Studies over the past two decades have endowed astrocytes with the ability to modulate neuron excitability through the release of neuroactive molecules based on intracellular calcium variations induced by incoming neurotransmitters (Araque etal.2014). Through this physiological mechanism, astrocytes locally adjust synaptic functions and modulate neuronal network states, thereby shaping a variety of behaviors (Santello, Toni, and Volterra 2019; Oliveira and Araque2022). At the cortical level, calciumbased astrocyte signaling controls sensoryevoked neuronal network activity (Lines etal.2020; Wang etal.2006; MiguelQuesada etal.2023) and modulates the generation of slowwave oscillatory patterns involved in cognitive performance (Poskanzer and Yuste2016; Fellin etal.2009). Astrocyte–neuron interactions undergo deleterious transformations in CNS pathologies with emerging impacts on disease symptomatology. Combined calcium imaging and electrophysiological studies have revealed diseasespecific astrocyte alterations affecting brain circuits relevant to cognitive function in mouse models of various neurodegenerative disorders (Delekate etal.2014; Nanclares etal.2023; Yu etal.2020; Shah etal.2022). In the experimental autoimmune encephalomyelitis (EAE) mouse model of MS, astrocytes display aberrant calcium activity causing exacerbated glutamate gliotransmission and longlasting alterations of excitatory synaptic transmission associated with memory deficits (Baraibar etal.2024; Habbas et al. 2015). These ex vivo studies showed cellautonomous functional deficits mediated by local intraparenchymal inflammation and support the notion that deregulated astrocyte network activity affects the brain cortex in MS. However, invivo studies assessing the emergence of aberrant astrocyte activity patterns in experimental MS are lacking. Here, we applied fiber photometry to the longitudinal analysis of the astrocytic network in the brain cortex of freely moving mice during the onset and progression of EAE. We show that the population of cortical astrocytes displays spontaneously occurring calcium events of enhanced duration at acute EAE disease. Furthermore, both spontaneous and sensoryevoked astrocyte calcium signals showed reduced amplitudes that correlated to the severity of clinical symptomatology. These results point to the involvement of cellautonomous and nonautonomous mechanisms underlying astrocyte network dysfunction in cortical MS with potential implications in disease symptomatology and progression. 2 | Methods 2.1 | Mice All experiments were performed in accordance with the Guidelines for the Animal Care and Use and the European Communities Council Directive of September 22th 2010 (2010/63/EU74). Experiments were approved by the local ethical committees of the University of the Basque Country (approval numbers 2017140, M202017144, and 2022245), the University of Bordeaux (approval number A33063098), and the French Ministry of Agriculture and Forestry (authorization number 3306369). Naive mice on a C57BL/6N background (Janvier, France; RRID: IMSR_JAX:000664) were used. Cages were enriched and mice were maintained under standard conditions (food and water adlibitum; 12–12 h light–dark cycle; two to five mice per cage). Experiments were performed during dark cycle (light off at 8:00 a.m.). EAE was induced in female mice based on epidemiological evidence that MS affects two to four times more women than men (Walton etal.2020). 2.2 | Surgery for AAV Administration and Fiber Implantation A total of 38 mice (8–9 weeks; 19–22 g) were anesthetized with isoflurane (4% for induction and 2.5% for maintenance) and placed on a heating pad to keep their body temperature at 37°C. Eye dehydration was prevented by topical application of ophthalmic gel, and analgesia was achieved by subcutaneous (s.c.) injection of buprenorphine (Buprecare, 0.05 mg/kg). The skin above the skull was shaved with a razor and disinfected with modified 70% ethanol and betadine before an incision was made. Mice were placed in a stereotaxic apparatus (David Kopf Instruments) with a mouse adaptor and lateral ear bars. Mice were injected with ssAAV9/2hGFAPhHBbI/EGCaMP6fbGHp(A) (ETH viral vector facility—ETH Zürich; cat. no. v275) for fiber photometry imaging of astrocytes. Virus titers were between 1010 and 1012 genomic copies per milliliter. Stereotaxic injections were targeted to the mouse somatosensory cortex according to the following coordinates (from bregma): anterior–posterior −1.5, medial–lateral ±2.5, and dorsal–ventral 14714159, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jnc.16305 by Universidad Del Paã-S Vasco, Wiley Online Library on [07/01/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 3 of 10 −1.5. Viral particles were injected at 400–500 nL alone or in combination at a maximum rate of 100 nL/min using a glass pipette attached to a Nanojet III (Drummond, Broomall, USA). Following virus delivery, the syringe was left in place for 10 min before being slowly withdrawn from the brain. The optical fiber (400 μm diameter) was placed 250 μm above the injection site during the same surgical session. Mice were weighed daily, and individuals that failed to return to their presurgery body weight were excluded from subsequent experiments. Mice were used for EAE induction 2 weeks after surgery. 2.3 | EAE Model Mice were arbitrarily assigned to control and EAE groups. Animals were immunized in the flank by s.c. injection of 200 μg myelin oligodendrocyte glycoprotein (MOG30–55) peptide (MEVGWYRSPFSRVVHLYRNGK) (Peptide Synthesis Core Facilities of the Pompeu Fabra University, Barcelona, Spain) in incomplete Freund's adjuvant supplemented with 8 mg/mL Mycobacterium tuberculosis H37Ra (Difco Laboratories; cat. no. 231141). Pertussis toxin (500 ng; Millipore; cat. no. 516560) was injected intraperitoneally on the day of immunization and again 2 days later. Body weight and neurological symptomatology were recorded daily and scored from 0 to 8 as follows: 0, no detectable changes in muscle tone and behavior; 1, flaccid tail; 2, paralyzed tail; 3, impairment or loss of muscle tone in hindlimbs; 4, hindlimb hemiparalysis; 5, complete hindlimb paralysis; 6, complete hindlimb paralysis and loss of muscle tone in forelimbs; 7, tetraplegia; and 8, moribund. A total of six EAE mice displaying neurological scores < 1 were excluded from the analysis. 2.4 | Fiber Photometry Imaging Freely moving mice were imaged after 3 days of handling habituation. On the day of recording, each mouse was placed in a rectangular chamber, and its behavior was recorded using a camera placed above the chamber. Baseline recordings of spontaneous astrocyte activity were made for 15 min every 1–2 days, starting 3 days before MOG administration. The calcium signal evoked by sensory stimulation of the tail was assessed at the end of the baseline period. Control and EAE mice were imaged in parallel during the time course of three independent experiments, including similar numbers of mice per experimental group. Cortical astrocyte GCaMP6f was imaged invivo using 470 and 405 nm LEDs. The emitted fluorescence is proportional to the calcium concentration for stimulation at 470 nm (Akerboom etal.2013; Ohkura etal.2012). The isosbestic 405 nm stimulation (UV light) was used in alternation with the blue light (470 nm) for analysis purposes, as the fluorescence emitted after this stimulation is not dependent on calcium (Lütcke etal.2010). The GCaMP6f fluorescence from the astrocytes was collected with an sCMOS camera through an optic fiber divided into two sections: a short fiber implanted in the brain of the mouse and a long fiber (modified patch cord), both connected through a ferrule–ferrule (1.25 mm) connection. A MATLAB program (MatlabWorks; RRID: SCR_001622) was used to synchronize each image recording made by the camera and the GCaMP6f light excitation was made by the LEDs (470 and 405 nm). The two wavelengths of 470 and 405 nm, at a power of 0.1 mW, were alternated at a frequency of 20 Hz each (40 Hz alternated light stimulations). To calculate fluorescence specifically due to calcium fluctuations and to remove bleaching and movement artifacts, the isosbestic 405 nm signal was subtracted from the 470 nm calcium signal. Specifically, normalized fluorescence changes (∆F/F0) were calculated by subtracting the mean fluorescence (2 min sliding window average) from the fluorescence recorded by the fiber at each time point and dividing this value by the mean fluorescence ((F−Fmean)/Fmean) using a customized Matlab software. Subsequently, the calciumindependent isosbestic signal was subtracted from the raw signal emitted after the 470 nm excitation to eliminate unspecific fluorescence. The result is the global calcium signal (∆F/F (%) = ∆FCa−∆Fisos), which was used as an estimate of tonic activity of the astrocytes. Calcium transients were detected on the filtered trace (high filter) using a threshold to identify them (2 median absolute deviation [MAD] of the entire trace). Duration and frequency were calculated for the detected transients. Amplitude was determined as the MAD of each studied period (Serrat etal.2021; Baraibar etal.2024). 2.5 | Fluorescence Immunohistochemistry Mice were deeply anesthetized by intraperitoneal injection of pentobarbital (Exagon, Axience SAS, 400 mg/kg body weight) and transcardially perfused with phosphatebuffered saline (PBS, 0.1 M, pH 7.4) before being fixed with 4% formaldehyde (Sigma, HT501128). Brains were isolated, incubated in the same fixative solution overnight at 4°C, and subsequently maintained in PBS + 0.025% azide at 4°C until use. Freefloating frozen coronal sections (40 μm) were cut out using a cryostat (Leica Biosystems CM1950S; RRID: SCR_018061), collected in antifreeze solution, and conserved at −20°C until further use. Floating sections were washed three times for 5 min in PBS (0.1 M, pH 7.4) and incubated with rabbit antiGFP (1:500; Invitrogen; RRID: AB_221569) and chicken antiGFAP (1:500; Abcam; RRID: AB_304558) overnight at 4°C in a blocking solution containing 10% donkey serum and 0.3% Triton X100 in PBS. The sections were then washed in PBS for 30 min at room temperature (RT), and primary antibodies were detected by incubation with donkey antirabbit Alexa 647 (1:500; Invitrogen; RRID: AB_2536183) and donkey antichicken Rhodamine Red (1:500; Jackson ImmunoResearch; RRID: AB_2340371) for 2 h at RT. Then, sections were washed for 15 min in PBS, mounted, dried, and mounted on microscope slides in FluoromountG (Invitrogen; cat. no. 00495802). Optical images from tissue sections processed in parallel were acquired in the same session using a 20× oil lens on a fluorescence microscope (Leica Microsystems CMS GmBbH, Type: 11504197). Image acquisition was carried out using fluorescence intensity settings at which the control sections without primary antibodies gave no signal. Immunolabeling was examined in the tissue area surrounding the probe using opensource image analysis software Fiji ImageJ (Schindelin etal.2012). GFP+ and GFAP+ cells were quantified by cell counting in two optical sections per tissue slice, and data were expressed as the mean cell number per square millimeter (mm2) of tissue area. 14714159, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jnc.16305 by Universidad Del Paã-S Vasco, Wiley Online Library on [07/01/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4 of 10 Journal of Neurochemistry, 2025 2.6 | Statistical Analyses No statistical methods were used to predetermine sample size. The number of subjects included in the study was determined based on previous studies of a similar nature (Baraibar etal.2024). No randomization was performed to allocate subjects to the study, and no blinding was performed. Summary results are presented as the mean of independent data points ± SEM, representing the number of animals tested. Datasets were initially tested for normal distribution using the Shapiro–Wilk test, and statistical analysis of the differences between groups was determined by a twotailed unpaired Student ttest or Mann–Whitney test. Correlation analysis was performed using Pearson's or Spearman's test. No test for outliers was conducted. Statistical analyses were performed using GraphPad Prism software (version 10.1.2). Differences were considered significantly different when p < 0.05. 3 | Results 3.1 | Cortical Astrocytes Show Deregulated Spontaneous Calcium Signals at Early Stages of EAE Disease Previous results in brain slices show that astrocytes of the mouse somatosensory cortex display spontaneous calcium hyperactivity at acute EAE disease (Baraibar etal.2024). To decipher the in vivo features of cortical astrocyte network dysfunction in MS, in this study we applied GCaMP6fbased fiber photometry to the analysis of astrocyte activity over the time course of EAE. The astrocyte calcium signals of freely behaving EAE and nonimmunized mice were monitored in parallel for 3 days before MOG35–55 administration and every 2–3 days thereafter starting at 3 days postimmunization (dpi) (Figure1a). The first clinical manifestations of EAE occurred at 10 dpi (score = 0.173 ± 0.072) with acute disease peaking around 18 dpi (score = 3.404 ± 0.424) and attenuation of neurological severity taking place at later time points (Figure1b). Histological analysis of postmortem tissues showed specific expression of the calcium sensor in astrocytes from control and EAE mice at 21 dpi following chronic fiber photometry imaging (FigureS1). As first approach to investigate the impact of autoimmune inflammation on the calcium activity of cortical astrocytes, we sought for possible differences in naturally occurring calcium fluctuations, which reflect the integration of both activitydependent and - independent cellular signals (Wang etal.2006; Zur Nieden and Deitmer2006), in EAE mice as compared to control animals. Consistent with previous exvivo reports of exacerbated astrocyte activity during EAE (Baraibar etal.2024), imaging of GCaMP6f expressing astrocytes in freely behaving mice shows a sustained increase in the duration of spontaneous calcium signals during acute disease (Figure1c–e). The frequency of astrocyte calcium oscillations was concomitantly reduced at earliest stages of acute EAE (Figure1c,d,f). Further analysis showed that spontaneous astrocyte calcium activity displays events of reduced amplitude in EAE mice as compared to control animals recorded in parallel (Figure1c,d,g). Remarkably, deregulation of astrocyte calcium signals in terms of duration, frequency, and amplitude emerged at early stages of acute disease (12 dpi), thus encompassing the onset of EAE symptomatology (Figure1b). Hence, EAE causes early impairments of spontaneously occurring astrocyte network activity in the brain cortex of freely moving mice. We next investigated the possible relationship between deregulated spontaneous calcium activity in cortical astrocytes and disease scores, which reflects spinal cord axon loss associated with the emergence of inflammatory demyelinating lesions during EAE progression (Wujek etal.2002; Liu etal.2008). Changes in the duration and frequency of astrocyte calcium signals did not correlate with the severity of clinical disability at any of the time points tested (Figure2a,b). However, we found a negative correlation between the amplitude deficits affecting spontaneously occurring astrocyte calcium signals and neurological disability in EAE mice that reached statistical significance at 12 and 14 dpi (Figure2c). These observations suggest that the neurodegenerative and inflammatory mechanisms that determine the severity of disease symptomatology in EAE mice are translated into a diminished amplitude of spontaneous astrocyte calcium signals at the cortical network level. 3.2 | Impaired SensoryEvoked Astrocyte Calcium Responses of Cortical Astrocytes Correlate With Disease Severity The population of cortical astrocytes responds to sensory stimulation invivo with a high degree of network reliability (Lines etal.2020; Wang etal.2006). In turn, astrocyte calcium signaling regulates neuronal network activity and sensoryevoked behaviors (Lines et al. 2020; MiguelQuesada et al. 2023). Thus, to further investigate the relationship between cortical astrocyte dysfunction and sensory information processing in MS, we next addressed changes in astrocytic calcium signals evoked by sensory stimulation of the tail in EAE mice. Under our experimental settings, suspension by the tail induced reliable calcium responses in cortical astrocytes from naive mice that remained stable in size during the chronic recording period (Figure3a,b). EAE induction was associated with a significantly decreased amplitude, but not duration, of astrocytic calcium responses evoked by tailholding that paralleled symptom onset (Figure 3a–c). Reminiscent of our observations concerning spontaneously occurring calcium oscillations in freely behaving mice, we observed an inverse correlation between the amplitude of astrocyte calcium transients evoked by tailholding and neurological disability score values during EAE progression that reached statistical significance at 17 and 19 dpi (Figure3d). Thus, EAE causes hyporesponsiveness of cortical astrocytes to sensory stimulation associated with clinical disability at acute disease. 4 | Discussion Neuroinflammatory mechanisms affecting the brain cortex play a major role in relevant clinical aspects of MS with both physical and cognitive rebounds. Reactive astrocytes produce aberrant signals that lead to synaptic and behavioral alterations during autoimmune cortical inflammation (Baraibar et al. 2024; 14714159, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jnc.16305 by Universidad Del Paã-S Vasco, Wiley Online Library on [07/01/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 5 of 10 Habbas etal.2015). However, the functional implications of cortical MS at the astrocyte network level remain largely defined. In this study, we asked whether the population of cortical astrocytes displays activity deficits in MS by applying fiber photometry imaging in freely behaving mice to the analysis of calcium signals in the EAE model of autoimmune demyelination. Our results link aberrant astrocyte network function in the brain cortex to neurological deficits and shed light on the mechanism underlying cortical dysfunction in MS. Although previous ex vivo studies have reported on aberrant calcium signals of cortical astrocytes in the EAE model (Baraibar etal.2024), here we provide the first quantification of spontaneous and peripherally evoked astrocyte network FIGURE 1 | Cortical astrocytes show dysregulated spontaneous calcium activity at early EAE stages. (a) Experimental approach for invivo timecourse analysis of astrocytic calcium activity during EAE. Fiber photometry imaging was performed in control and EAE mice for 3 consecutive days before EAE induction by immunization with MOG3055 and at 3, 5, 7, 10, 12, 14, 17, 19, and 21 days postimmunization (dpi). (b) Neurological score of mice included in the invivo analysis of cortical astrocyte calcium signals during EAE time course. (c) Representative traces of spontaneous activity in the population of cortical astrocytes during the preimmunization phase (−1 dpi) (top) and during acute EAE disease (14 dpi) (bottom). Dots correspond to transients detected above the threshold (median + 2*MAD). (d) Time course of normalized duration, frequency, and amplitude of cortical astrocyte calcium signals recorded during EAE progression. Raw data recorded from EAE mice were expressed relative to values from control animals recorded in parallel during EAE progression (n = 13–19 mice). (e–g) Comparison between spontaneous astrocyte calcium responses in freely behaving naïve and EAE mice at different time points of acute disease progression. Data were analyzed by twotailed unpaired ttest or Mann–Whitney test. (e) Duration: 12 dpi (t, df = 3.948, 30; p = 0.00044), 14 dpi (t, df = 2.865, 30; p = 0.00754), 17 dpi (t, df = 3.284, 30; p = 0.00260), 19 dpi (t, df = 2.232, 30; p = 0.03325), and 21 dpi (U = 117; p = 0.82055). (f) Events per minute: 12 dpi (t, df = 4.725, 30; p = 0.00005), 14 dpi (t, df = 3.254, 30; p = 0.00282), 17 dpi (U = 62; p = 0.01765), 19 dpi (t, df = 1.232, 30; p = 0.2274), and 21 dpi (U = 91; p = 0.21590). (g) Amplitude: 12 dpi (U = 44; p = 0.00165), 14 dpi (U = 26; p = 0.00006), 17 dpi (U = 38; p = 0.00062), 19 dpi (t, df = 4.520, 30; p = 0.00009), and 21 dpi (t, df = 4.481, 30; p = 0.00010). 0510 15 20 0 1 2 3 4 5 6 Dpi Mean clinical score 12 14 17 19 21 0 1 2 3 4 Days post-immunization Events per minute Control EAE *** ** * n.s. n.s. 12 14 17 19 21 0 1 2 3 4 5 Days post-immunization Amplitude (ΔF/F 0 ) Control EAE ** *** ****** *** Stimulation Fiber photometry imaging 2 weeks ca Pre-EAE EAE Day -1 Day -2 Day -3 … Dpi 3 Dpi 5 … Dpi 21 EAE GFAP GCaMP6f 15 min Spontaneous de f b g 12 14 17 19 21 0 10 20 30 40 50 Days post-immunization Duration (s) Control EAE *** ** ** ** * 0510 15 20 0 1 2 3 4 Days post-immunization Spontaneous astrocyte calcium signals Frequency Duration Amplitude 30 s 2.5 (ΔF/F0) 14714159, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jnc.16305 by Universidad Del Paã-S Vasco, Wiley Online Library on [07/01/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 6 of 10 Journal of Neurochemistry, 2025 activities during the emergence of disease symptomatology. Astrocytes of the somatosensory cortex displayed naturally occurring calcium signals of increased duration but reduced amplitude and event frequency during acute disease. These observations highlight the complexity of aberrant astrocyte network activity in freely behaving EAE mice and suggest that the dysfunctional properties of astrocytic calcium signals in the brain cortex may emerge from diverse physiopathological mechanisms. In this regard, our recent study in cortical slices showed that EAE promotes astrocyte calcium hyperactivity consisting of spontaneous events of enhanced duration and frequency that might potentially translate into invivo signals with an extended duration during acute disease (Baraibar etal.2024). At a mechanistic level, calcium hyperexcitability of cortical astrocytes recorded exvivo was independent of neuronal activity and emerged from reactivityrelated adaptations during autoimmune inflammation (Baraibar etal.2024). In the context of these previous results, our invivo observations suggest that the dysfunctional spontaneous astrocyte network activity in the somatosensory cortex results, in part, from intrinsic mechanisms associated with the pathogenic transformation of astroglial cells in the local inflammatory milieu. The present observation that astrocytes of the somatosensory cortex display spontaneous calcium events of reduced amplitude when recorded invivo is, however, in apparent contrast to exvivo slice imaging results of astrocyte hyperactivity at acute FIGURE 2 | Dysregulation of spontaneous astrocyte calcium signals in the EAE cortex: Correlation to neurological deficits. Correlation analysis between the duration (a), frequency (b), and amplitude (c) of spontaneous astrocyte calcium signals in the mouse somatosensory cortex and clinical disability scores at different time points of EAE disease progression (n = 13 mice). Raw calcium imaging data from EAE mice were normalized to values recorded during the presymptomatic phase and plotted against neurological scores. Pearson's or Spearman's correlation coefficients and p values are indicated in each dot plot. 02 46 0 1 2 3 4 5 Neurological score Events per minute 21 dpi r=0.2017 p=0.5059 a b c 0246 0 1 2 3 417 dpi r=-0.5450 p=0.0570 Neurological score Amplitude 0246 0 1 2 3 4 5 Neurological score Duration 19 dpi r=-0.0504 p=0.8713 0123 0 1 2 3 412 dpi r=-0.5705 p=0.0418 Neurological score Amplitude 0246 0 1 2 3 4 Neurological score Amplitude 19 dpi r=-0.5446 p=0.0543 02 46 0 1 2 3 4 Neurological score Amplitude 21 dpi r=-0.5290 p=0.0630 0123 0 1 2 3 4 512 dpi r=-0.1981 p=0.5136 Neurological score Eventsperminute 0246 0 1 2 3 4 5 Neurological score Eventsperminute 19 dpi r=0.0707 p=0.8185 0246 0 1 2 3 4 517 dpi r=0.3126 p=0.2964 Neurological score Events per minute 0246 0 1 2 3 414 dpi r=-0.6851 p=0.0120 Neurological score Amplitude 0246 0 1 2 3 4 514 dpi r=-0.1934 p=0.5270 Neurological score Eventsper minute 0123 0 1 2 3 4 512 dpi r=-0.0650 p=0.8337 Neurological score Duration 0246 0 1 2 3 4 514 dpi r=0.2093 p=0.4925 Neurological score Duration 0246 0 1 2 3 4 517 dpi r=-0.2265 p=0.4569 Neurological score Duration 02 46 0 1 2 3 4 5 Neurological score Duration 21 dpi r=-0.1740 p=0.5670 14714159, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jnc.16305 by Universidad Del Paã-S Vasco, Wiley Online Library on [07/01/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 7 of 10 EAE disease (Baraibar etal.2024). Discrepancies concerning spontaneous activity measured using fiber photometry and two photon slice imaging of astrocytes in the somatosensory cortex of EAE mice might be related to disparities between both experimental procedures, such as fiber implantation, and the use of TTX during exvivo analysis (Baraibar etal.2024). An additional confounding factor is that fiber photometry recordings represent the added signal from a population of astrocytes—including somas and processes—whereas ex vivo slice imaging monitors individual cells and processes. The signal measured invivo, equivalent to the population average, could result in filtering some events, for example, those very slow or with relatively low amplitudes. Further studies are needed to decipher how deregulated calcium signaling at the single cell level contributes to complex changes in the spatial and temporal properties of astrocyte population activity during autoimmune inflammation. The calcium dynamics of cortical astrocytes in the intact brain are modulated by local and peripherally evoked neuronal network activities via neurontoastrocyte communication (Hirase etal.2004; Wang etal.2006; Lines etal.2020). In this study, we show that reductions in the amplitude of spontaneous astrocyte signals during acute EAE symptomatology are mirrored by deficits in peripherally evoked calcium responses, which reflect the activation of cortical neurons during information processing (Lines etal.2020; Wang etal.2006). Although caution is needed when attempting to compare data from exvivo and invivo approaches, this observation sits well with previous findings that astrocytes in the EAE cortex demonstrate decreased calcium signaling to neurotransmitter receptor agonists and chemogenetic activation of Gq and Gi proteins despite being spontaneously hyperactive (Baraibar etal.2024). Both mechanisms of astrocyte dysfunction are not mutually exclusive and may act independently, synergistically, or in opposition to modulate specific patterns of astrocyte calcium activity invivo, thus providing an explanation for the apparently contradictory findings regarding the modulation of spontaneous calcium activity in terms of duration and amplitude during EAE. Beyond mechanistic implications, the cellautonomous hyporesponsiveness of astrocytes to acute pharmacological stimuli points to intrinsic deficits in astrocyte feedbacks to neuronal activity at the cortical level that would be translated into a reduced calcium population activity in freely moving mice. FIGURE 3 | Impaired astrocyte calcium responses to sensory stimulation in the EAE cortex correlate to disease severity. (a) Representative traces showing calcium responses of cortical astrocytes evoked by tailholding in control (top) and EAE (17 dpi) (bottom) mice. (b) Normalized amplitude and duration of evoked cortical astrocyte calcium responses during EAE progression. Raw data recorded from EAE mice were expressed relative to values from control animals recorded in parallel during EAE progression (n = 13–19 mice). (c) Amplitude of calcium transients evoked by sensory stimulation of the tail in immunized mice at different time points of disease progression and control animals recorded in parallel. Data were analyzed by twotailed unpaired ttest or Mann–Whitney test: 12 dpi (U = 37; p = 0.00052), 14 dpi (U = 34; p = 0.00030), 17 dpi (t, df = 4.193, 30; p = 0.00022), 19 dpi (t, df = 3.583, 30; p = 0.00118), and 21 dpi (t, df = 3.872, 30; p = 0.00054). (d) Correlation analysis between the amplitude of sensoryevoked calcium responses in cortical astrocytes and clinical severity during EAE. Raw data from EAE mice were normalized to values recorded during the presymptomatic phase and plotted against neurological scores. Pearson's or Spearman's correlation coefficients and p values are indicated in each dot plot. a d Control EAE 2.5 (ΔF/F0) 30 s bc 12 14 17 19 21 0 5 10 15 20 Days post-immunization Amplitude (ΔF/F 0 ) Control EAE *** *** ***** *** 0510 15 20 0 1 2 3 4 Dpi Evoked astrocytic Ca 2+ transients Duration Amplitude 0246 0 1 2 3 417 dpi r = - 0.6058 p = 0.0311 Neurological score Amplitude (ΔF/F 0 ) 0123 0 1 2 3 412 dpi r = - 0.3849 p = 0.1936 Neurological score Amplitude (ΔF/F 0 ) 0246 0 1 2 3 414 dpi r = - 0.5148 p = 0.0718 Neurological score Amplitude (ΔF/F 0 ) 0246 0 1 2 3 4 Neurological score Amplitude (ΔF/F 0 ) 21 dpi r = - 0.5188 p = 0.0693 0246 0 1 2 3 4 Neurological score Amplitude (ΔF/F 0 ) 19 dpi r = - 0.5714 p = 0.0413 14714159, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jnc.16305 by Universidad Del Paã-S Vasco, Wiley Online Library on [07/01/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 8 of 10 Journal of Neurochemistry, 2025 Present results unveil interesting aspects of invivo astrocyte pathology in the MS cortex with potential implications for disease symptomatology. First, astrocyte network activity deficits encompass the appearance of neurological disability at acute EAE disease. Second, impairments of spontaneous activity and peripherally evoked astrocyte responses in terms of amplitude of the calcium signals correlate to the severity of clinical scores at different time points of disease progression. Although a causal relationship between neuropathology and astrocyte dysfunction in the somatosensory cortex cannot be firmly established based on our correlation analysis, these observations support the possibility that the reductions in cortical astrocyte network activity and neurological disability scores reflect common pathological mechanisms. The clinical deficits that characterize EAE manifest as an ascending spasticity and paralysis generally graded on a point scale based on the visual scoring of motor signs, although disease scores most likely reflect a combination of sensorimotor abnormalities progressively affecting the tail and limbs (Thibault, Calvino, and Pezet 2011; Potter et al. 2016; Segal etal.2020). Although a number of studies have highlighted inflammatory mechanisms and functional deficits that affect the somatosensory cortex during EAE (Potter etal.2016; Baraibar etal.2024; Yang etal.2013), the ascending paralysis that typifies disease progression seems to emerge predominantly from the loss of spinal cord axons associated with the occurrence of inflammatory, demyelinating lesions in white matter tracks, which mimics the degeneration of sensory and corticospinal tracts in MS patients (Wujek etal.2002; Liu etal.2008; DeLuca, Ebers, and Esiri2004). Indeed, the extent of spinal cord axonal degeneration correlates closely to neurological disability both in patients with MS and in the EAE model of the disease (Liu etal.2008; Bjartmar etal.2000). On the other hand, the primary somatosensory cortex computes sensory information from different modalities via thalamic inputs to layer IV and eventually transmits output signals to the spinal cord through the corticospinal tract, which originates from the pyramidal neurons in layer V and constitutes the major cortical output to control motor behavior and sensory ascending feedback inputs (Macías etal. 2022; MorenoLópez etal. 2016; Lemon 2008). Finally, it is well established that astrocyte network signals in the somatosensory cortical circuits are positively modulated by peripheral stimuli that elicit neuronal activity in a stimulusdependent manner and thus follow the information encoded by neuronal inputs to the cortex (MiguelQuesada etal.2023; Lines etal.2020; Zhao, Wang, and Wang2012). Thus, in the context of these previous observations, our present results suggest that deficits in the transmission of peripherally evoked neuronal signals emerging, at least in part, from spinal cord pathology, may contribute to astrocyte network activity impairments in the EAE cortex. It should be noted, however, that in this study we performed fiber photometry imaging of astrocyte activity in the whiskerrelated primary somatosensory cortex while applying sensory stimulation in the mouse tail. This mechanical manipulation induces an indirect sensory stimulation effect within the somatosensory barrel cortex linked to the activation of different brain structures as neuromodulatory systems (Aronoff etal.2010) rather than a direct sensory input from the periphery to the cortex. Furthermore, suspension by the tail might potentially activate astrocytes within somatosensory cortical circuits due to stressrelated delivery of acetylcholine and/or norepinephrine (Rasmussen etal.2023; Gau etal.2024; Wang etal.2023), thus adding further complexity to the interpretation of our present results. Future studies combining sensory stimulation of increasing intensities with activity readouts of whisker and/or limb representation of the mouse somatosensory cortex may help clarify the mechanistic implications of astrocyte dysfunction in the pathological modulationspecific sensory modalities during EAE (Ziegler et al. 2023; MiguelQuesada etal.2023). Sensory perception, locomotion, and arousal are directly associated with increases in cortical astrocytic calcium activity invivo (Bojarskaite etal.2020; Wang etal.2006, 2023; Paukert etal.2014; Rasmussen etal.2023; Gau etal.2024). However, the consequences of astrocyte network activity on cortical output function have been best studied in the context of sensory information processing. Recent studies in the field demonstrate that astrocyte responses to peripherally evoked neuronal activity modulate the dynamic range for which cortical networks respond to sensory stimuli to control behavior (Lines etal.2020; MiguelQuesada etal.2023; Wang etal.2023). The corollary of these observations is that astrocyte network dysfunction may significantly contribute to sensory symptoms and potentially other clinically relevant aspects related to cortical dysfunction in neurological disorders. Although in vivo studies assessing the implications of astrocytetoneuron communication deficits in neuroinflammatory and neurodegenerative pathologies are still scarce, available evidence supports the hypothesis that aberrant astrocyte calcium signaling in the brain cortex causes neuronal network dysfunction and behavioral impairments in Alzheimer's disease (Lines etal.2022; Åbjørsbråten etal.2022; Shah et al. 2022). In this context, it seems plausible that the functional disturbances of cortical astrocytes dysregulate neuron network operation, leading to sensorimotor pathology and cognitive impairments in MS patients (Margoni et al. 2023; Madsen etal.2022). We need to remark, however, that in this study we did not monitor neuronal activity and our results do not provide definitive evidence linking deregulation of astrocyte calcium activity to deficits in cortical neuron network responsiveness during sensory information processing. Nonetheless, recent electrophysiological results in the EAE mouse model demonstrate that aberrant astrocyte calcium signaling is associated with exacerbated glutamatemediated gliotransmission and with synaptic plasticity deficits affecting layer V pyramidal neurons of the somatosensory cortex during acute disease (Baraibar et al. 2024). Together with these previous observations, our present findings strongly suggest that defects in the astrocyte calcium population activity may indeed translate into aberrant cortical neuronal network function during EAE. Future invivo studies combining electrophysiological neuronal recordings in response to sensory stimulation while simultaneously monitoring and/or manipulating astrocyte calcium activity will help identify mechanistic associations between deficits in astrocyte and neuronal network responsiveness in cortical MS, as well as their impact on disease symptomatology. In conclusion, the present study provides several key findings on the role of astrocytes in MS cortical dysfunction, which is one of the earliest and most consistent findings in patients. Our results demonstrate that naturally occurring and evoked astrocyte population activities are altered in the cortex of EAE mice at early stages of disability and link astrocyte network dysfunction 14714159, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jnc.16305 by Universidad Del Paã-S Vasco, Wiley Online Library on [07/01/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 9 of 10 at the cortical level to neurological deficits emerging from autoimmune demyelination. In the context of previous studies, these findings indicate that aberrant astrocyte activity is a consistent observation in cortical MS that emerges in part from deficits in population responses to peripheral inputs. These results support the existence of a disrupted astrocyte–neuronal network interplay, contributing to cortical dysfunction and associated sensorimotor symptomatology in MS patients. Author Contributions A. MorenoGarcía: formal analysis, investigation. R. Serrat: formal analysis, investigation, methodology, software, supervision, writing – review and editing. F. JulioKalajzic: investigation. A. BernalChico: investigation. A. M. Baraibar: writing – original draft, writing – review and editing. C. Matute: investigation. G. Marsicano: conceptualization, data curation, funding acquisition, project administration, resources, supervision, writing – review and editing. S. Mato: conceptualization, data curation, funding acquisition, project administration, resources, supervision, visualization, writing – original draft, writing – review and editing. Acknowledgements We would like to thank the personnel of the Animal Facilities of the University of the Basque Country and Neurocentre Magendie for mouse care. We also thank all members of Mato's and Marsicano's labs for useful discussions and advice. This work was funded by the Instituto de Salud Carlos III (PI21/00629, to S.M.) and cofounded by the European Union, Basque Government (PIBA_2023_1_0046; 2023111031; IT147322, to S.M.; CannaMetHD, to S.M. and G.M.; IT120319, to C.M.), ARSEP Foundation (ARSEP1310 to S.M. and G.M.), INSERM (to G.M.), the European Research Council (MiCaBra, ERC2017AdG786467, to G.M.), Fondation pour la Recherche Medicale (FRM, DRM20101220445 to G.M.), Region Aquitaine (CanBrain, AAP2022A202116763610 and - 17219710 to G.M.); French State/Agence Nationale de la Recherche (HippObese, ANR23ce14000403; ERANet Neuron CanShank, ANR21NEU2000104, to G.M.), La Caixa Research Health 2023 (PsychoCannabis, HR2300793, to G.M.), Ministerio de Ciencia e Innovacion (PID2019109724RB100 to C.M.), and the Postdoctoral Program of the Basque Government (to A.M.B.). The cartoon in Figure1a was created using BioRe nder. com. Conflicts of Interest The authors declare no conflicts of interest. Data Availability Statement All data are available in the main text and the Supporting Information, or from the corresponding author on reasonable request. Peer Review The peer review history for this article is available at https:// www. webof scien ce. com/ api/ gatew ay/ wos/ peerreview/ 10. 1111/ jnc. 16305 . References Åbjørsbråten, K. S., G. H. E. S. Skaaraas, C. Cunen, et al. 2022. “Impaired Astrocytic Ca2+ Signaling in AwakeBehaving Alzheimer's Disease Transgenic Mice.” eLife 11: 75055. Akerboom, J., N. Carreras Calderón, L. Tian, etal. 2013. “Genetically Encoded Calcium Indicators for MultiColor Neural Activity Imaging and Combination With Optogenetics.” Frontiers in Molecular Neuroscience 6: 2. Araque, A., G. Carmignoto, P. G. Haydon, S. H. Oliet, R. Robitaille, and A. Volterra. 2014. “Gliotransmitters Travel in Time and Space.” Neuron 81: 728–739. Aronoff, R., F. Matyas, C. Mateo, C. Ciron, B. Schneider, and C. C. Petersen. 2010. “LongRange Connectivity of Mouse Primary Somatosensory Barrel Cortex.” European Journal of Neuroscience 31: 2221–2233. Baraibar, A. M., T. Colomer, A. MorenoGarcía, et al. 2024. “Autoimmune Inflammation Triggers Aberrant Astrocytic Calcium Signaling to Impair Synaptic Plasticity.” Brain, Behavior, and Immunity 121: 192–210. Bjartmar, C., G. Kidd, S. Mörk, R. Rudick, and B. D. Trapp. 2000. “Neurological Disability Correlates With Spinal Cord Axonal Loss and Reduced NAcetyl Aspartate in Chronic Multiple Sclerosis Patients.” Annals of Neurology 48: 893–901. Bojarskaite, L., D. M. Bjørnstad, K. H. Pettersen, etal. 2020. “Astrocytic Ca2+ Signaling Is Reduced During Sleep and Is Involved in the Regulation of Slow Wave Sleep.” Nature Communications 11: 3240. Calabrese, M., V. Poretto, A. Favaretto, et al. 2012. “Cortical Lesion Load Associates With Progression of Disability in Multiple Sclerosis.” Brain 135: 2952–2961. Delekate, A., M. Füchtemeier, T. Schumacher, C. Ulbrich, M. Foddis, and G. C. Petzold. 2014. “Metabotropic P2Y1 Receptor Signalling Mediates Astrocytic Hyperactivity InVivo in an Alzheimer's Disease Mouse Model.” Nature Communications 5: 5422. DeLuca, G. C., G. C. Ebers, and M. M. Esiri. 2004. “Axonal Loss in Multiple Sclerosis: A Pathological Survey of the Corticospinal and Sensory Tracts.” Brain 127: 1009–1018. Fellin, T., M. M. Halassa, M. Terunuma, et al. 2009. “Endogenous Nonneuronal Modulators of Synaptic Transmission Control Cortical Slow Oscillations In Vivo.” Proceedings of the National Academy of Sciences of the United States of America 106: 15037–15042. Gau, Y. A., E. T. Hsu, R. J. Cha, etal. 2024. “Multicore Fiber Optic Imaging Reveals That Astrocyte Calcium Activity in the Mouse Cerebral Cortex Is Modulated by Internal Motivational State.” Nature Communications 15: 3039. Habbas, S., M. Santello, D. Becker, etal. 2015. “Neuroinflammatory TNFα Impairs Memory via Astrocyte Signaling.” Cell 163: 1730–1741. Hirase, H., L. Qian, P. Barthó, and G. Buzsáki. 2004. “Calcium Dynamics of Cortical Astrocytic Networks InVivo.” PLoS Biology 2: E96. Latypov, T. H., A. Wolfensohn, R. Yakubov, etal. 2024. “Signatures of Chronic Pain in Multiple Sclerosis: A Machine Learning Approach to Investigate Trigeminal Neuralgia.” Pain. https:// doi. org/ 10. 1097/j. pain. 00000 00000 003497 (in press). Lazzarotto, A., M. Hamzaoui, M. Tonietto, etal. 2024. “Time is Myelin: Early Cortical Myelin Repair Prevents Atrophy and Clinical Progression in Multiple Sclerosis.” Brain 147: 1331–1343. Lemon, R. N. 2008. “Descending Pathways in Motor Control.” Annual Review of Neuroscience 31: 195–218. Lines, J., A. M. Baraibar, C. Fang, et al. 2022. “AstrocyteNeuronal Network Interplay is Disrupted in Alzheimer's Disease Mice.” Glia 70: 368–378. Lines, J., E. D. Martin, P. Kofuji, J. Aguilar, and A. Araque. 2020. “Astrocytes Modulate SensoryEvoked Neuronal Network Activity.” Nature Communications 11: 3689. Liu, Z., Y. Li, J. Zhang, S. Elias, and M. Chopp. 2008. “Evaluation of Corticospinal Axon Loss by Fluorescent Dye Tracing in Mice With Experimental Autoimmune Encephalomyelitis.” Journal of Neuroscience Methods 167: 191–197. Lütcke, H., M. Murayama, T. Hahn, etal. 2010. “Optical Recording of Neuronal Activity With a GeneticallyEncoded Calcium indicator in Anesthetized and Freely Moving Mice.” Frontiers in Neural Circuits 4: 9. 14714159, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jnc.16305 by Universidad Del Paã-S Vasco, Wiley Online Library on [07/01/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License