Gestational Hypothyroxinemia Shifts Th1/Th17 Immunity and Innate Lymphoid Cell Balance in the Adult Offspring during the Presymptomatic Stage of Experimental Autoimmune Encephalomyelitis Open Access

Introduction: Thyroid hormone homeostasis during pregnancy is crucial for proper neurodevelopment and cognitive capacity during adulthood. Accumulating evidence reveals that gestational hypothyroxinemia (HTX) modulates the immune response of the adult offspring. Methods: In the present study, adult mice gestated in HTX and their euthyroid counterparts were induced with a mild form of experimental autoimmune encephalomyelitis (EAE), a widespread model of multiple sclerosis, and analyzed at baseline and 7 days after EAE induction. Results: Levels of circulating IL-17 were significantly lower in mice gestated in HTX at both timepoints, while circulating IFN-γ was significantly higher only in mice gestated in HTX, 7 days after EAE induction. A significant increase in type 1 innate lymphoid cells (ILC1) was found only in mice gestated in HTX 7 days after EAE induction, while type 3 innate lymphoid cells (ILC3) populations showed no variation. Interestingly, a significant increase of Th17 CD4⁺ cells was found only in mice of euthyroid gestation, 7 days after EAE induction. Conclusion: These results highlight the repercussions of thyroid hormone impairment in utero at adult ages while dissecting on the pathogenesis of EAE in terms of Th1/Th17 balance from an innate immune perspective. These findings contribute to the advancement of our comprehension of the presymptomatic stage of EAE, unveiling new paths for basic and translational research in the field of neuroinflammation.

Iodine is essential for thyroid hormone homeostasis, which continually shifts during pregnancy adapting to the physiological needs of the developing fetus. Trimester-specific reference ranges have been identified by the American Thyroid Association (ATA) for thyroid hormones thyroid-stimulating hormone (TSH), tri-iodothyronine (T₃), and thyroxine (T₄) in the 2017 Guidelines for the Diagnosis and Management of Thyroid Disease During Pregnancy and the Postpartum [1]. There, a state of gestational hypothyroxinemia (HTX) is defined as a free T₄ concentration in the lower 2.5th–5th percentile of a given population together with normal maternal TSH concentrations. According to the latest report on maternal thyroid function, the incidence of isolated HTX is 2.2% worldwide [2].

The fetal thyroid gland reaches its final anatomical position by the end of the 7th week of development and begins to function at the end of the 12th week [3]. Therefore, fetal reliance on maternal thyroid hormone homeostasis is exclusive throughout the first trimester. Early reports indicate that maternal T₄ is transferred into the exocoelomic cavity, the fetal gut, and reaches circulation via the secondary yolk sac [4]. There, an orchestrated expression and enzymatic activity of type II 5′-iodothyronine deiodinase (5′D-II) and inner ring iodothyronine deiodinase (5D-III) achieve tightly controlled concentrations of T₃ in fetal structures [4]. Interestingly, it is suggested that T₄ levels in fetal fluids directly relate to the respective levels of maternal T₄ and not T₃. Thus, T₃ in early fetal fluids are not transferred from the mother but result from the metabolization of the maternal T₄ that reaches fetal tissues [4]. Consequently, while gestational HTX may proceed subclinically in the mother, it has been shown to induce irreparable damage in the offspring, resulting in adverse cognitive function [5], abnormal birthweight [2], intellectual disability [6], and development of autism spectrum disorders [7].

Gestational HTX can be reproduced in vivo through the administration of anti-thyroid drug methimazole with low birthweight and delayed maturation of glial cells of the CA1 region of the hippocampus [8], abnormal neurobehavioral and physiological development [9], impaired learning and memory [10], delayed offspring hippocampal growth [11, 12], abnormal mRNA and microRNA patterns in the cerebral cortex [13], decreased microglia and increased astrocyte reactivity [14], and impaired number of synaptic Glun1 and CD3ζ clusters and glycine-induced long-term potentiation in hippocampal neurons [15], as outcomes in the exposed offspring. Surprisingly, only three studies have assessed the effects of gestational HTX over the immune system in the adult offspring, with findings of a stronger response to human metapneumovirus [16] and an earlier and stronger response to both ulcerative colitis [17] and to a mild form of experimental autoimmune encephalomyelitis (EAE) [18], a widespread mouse model that recapitulates most aspects of multiple sclerosis (MS) of widespread use. While the former study describes an increase in the quantity of activated CD8⁺ T cells in the lungs of the HTX offspring [16], the latter studies coincide in describing clinical progressions of earlier and stronger presentation, with a significant increase of CD8⁺ and regulatory T (Treg) cells in tissue after immune challenge, compared to their euthyroid (EU) counterparts. EAE in the HTX offspring was further characterized with an increase in CD4⁺ T cells in the spinal cord [18]. Therefore, gestational HTX represents an interesting condition with significant effects on the immune response of the adult offspring that remains to be explored fully.

Recently, innate lymphoid cells (ILCs) have been deemed key coordinators of EAE-induced neuroinflammation, with a unique population of ILC3 cells promoting T-cell responses in the central nervous system (CNS) in mice [19] and another unique population of T-bet-dependent NKp46⁺ ILCs initiating myelin-reactive Th17 CD4⁺ cell-mediated responses [20]. Despite ILCs being considered first responders in this disease [21, 22] and despite the presymptomatic stage of EAE being considered a time window of interest for the discovery of novel early biomarkers and novel therapeutic target [23], to our knowledge, no study to date has characterized EAE-induced neuroinflammation within the first 7 days of its progression. Moreover, to our knowledge, there is no account of ILCs in mice gestated in HTX. Thus, the present work is aimed at (1) assessing systemic levels of proinflammatory cytokines IL-17 and IFN-γ, (2) evaluating the differential distribution of ILC1, ILC3, and Th17 CD4⁺ cells, in the CNS, gut, spleen, and mesenteric lymph nodes (mLNs), of mice at baseline and 7 days after the induction of EAE, and (3) assessing how gestational HTX exacerbates and propels this immune response in the adult offspring.

All procedures were approved by the Bioethics Committee from Universidad Andrés Bello (protocol number 008/2022). Wild-type C57BL/6 mice were housed individually in cages under a 12 h/12 h light/dark cycle and allowed food and water ad libitum. Timed-pregnant dams were randomly assigned to the HTX group and were provided with drinking water containing 0.025% methimazole and 2% sucrose (to mask bitterness) only between gestational days 10 and 14, while their EU counterpart was provided with drinking water containing only 2% sucrose for the same gestational period.

On gestational day 14, blood samples from the facial vein were collected from all pregnant dams to assess thyroid hormone levels to confirm gestational HTX. Serum was separated by centrifugation at 1,000 g for 15 min at 4°C. Levels of total T₃ and TSH were quantified by ELISA (CUSABIO, #E05086m and #E05116m, respectively) following manufacturer instructions. Total T₄ levels were determined by chemiluminescence at an external certified veterinary laboratory (LQCE, Santiago, Chile).

The marble burying test is designed to analyze repetitive/persistent behavior [24, 25]. Each mouse was placed inside the testing cage (arena size: 24 cm × 17.2 cm, wood chips bedding depth: 5 cm) containing 12 marbles arranged in four columns of three marbles at equidistant distances. After 30 min of exploration, each mouse was returned to its respective cage. The number of buried marbles was counted, considering them buried when 2/3 of their volume was concealed within the wood chips.

Fifty-five days after parturition, EU and HTX adult female offspring weighting between 16 and 19 g were induced with a mild form of EAE. Briefly, mice were immunized with 50 µg of myelin oligodendrocyte glycoprotein peptide (MOGp_35–55, MEVGWYRSPFSRVVHLYRNGK) s.c., emulsified in complete Freund’s adjuvant supplemented with heat-inactivated Mycobacterium tuberculosis H37Ra. On days 0 and 2 after immunization, mice were injected with 350 ng of pertussis toxin i.p. Mice were followed for standard EAE scoring (0, no detectable clinical signs; 0.5, partial tail weakness; 1, tail paralysis; gait instability or impaired righting ability; 2, hind limb paresis or partial paralysis 3, full hind limb paralysis with partial fore limb paresis or paralysis; 4, hind limb and fore limb paralysis; 5, moribund) and weight on a daily basis. On day 7 after induction, mice were euthanized by gaseous anesthetic overdose (5% isoflurane) and tissues were harvested.

At the time of euthanasia, whole blood was collected. Serum was separated by centrifugation at 1,000 g for 10 min at 4°C. Circulating levels of IL-17A (BioLegend, #432501) and IFN-γ (BD #555138) were quantified by ELISA following manufacturer instructions.

Tissues were minced and homogenized using a sterile razor blade and a syringe plunger in cold PBS-EDTA 2 mm. Homogenates were then filtered through a 70-µm cell strainer, collected in a 50 mL conical tube, and centrifuged 300 g for 5 min at 4°C. Resulting pellet was resuspended in 5 mL Percoll 30% v/v, carefully transferred into a 15 mL conical tube already containing 5 mL Percoll 70% v/v, and centrifuged 700 g for 20 min at 4°C without brakes. Cells in the resulting interphase were transferred into a 15 mL conical tube, resuspended in 10 mL PBS-EDTA 2 mm, and centrifuged 300 g for 5 min at 4°C. Resulting pellet was resuspended in PBS-EDTA 1 mM–2% FBS (FACS buffer) and was kept at 4°C until flow cytometry preparation.

Cell suspensions were prepared following protocol by Melo-Gonzalez and Hepworth [26]. Briefly, tissues were stripped of associated fat, Peyer’s patches, mucus, epithelial cells, and intraepithelial lymphocytes, by mechanical traction, razor blade cuts, and by incubation with a 1 mm EDTA-1 mM DTT-5% FBS stripping buffer. The tissues were washed using cold PBS and digested with 10 mL of 1 mg/mL collagenase/dispase (Roche #10269638001) and 20 µg/mL DNAse I (Roche # 10104159001) for 45 min at 37°C in a rotating shaker. Liberated lamina propria lymphocytes were filtered through a 70-µm cell strainer, washed with fresh RPMI-1% l-glutamine-1x Pen/Strep-10% FBS, and centrifuged 450 g for 5 min at 4°C. Resulting pellet was resuspended in FACS buffer and was kept at 4°C until flow cytometry preparation.

Tissues were homogenized using a syringe plunger in cold RPMI-2% FBS. The resulting homogenate was filtered through a 70-µm cell strainer, collected in a 15 mL conical tube, and centrifuged at 210 g for 5 min at 4°C. Red blood cells were lysed only in spleen suspensions by incubation with 0.8% NH₄Cl for 5 min. Resulting pellet was resuspended in FACS buffer and was kept at 4°C until flow cytometry preparation.

Dead cells were excluded with BD Horizon Fixable Viability Stain 700 (BD Biosciences #564997). Antibodies used were as follows: CD4 (RM4–5, BD Biosciences #563747), CD5 (53–7.3, BD Biosciences #563069), CD11b (M1/70, BD #561039), CD45 (30-F11, BioLegend #103149), NK1.1 (PK136, BioLegend #108739), CD11c (N418, BioLegend #117323), CD19 (6D5, BioLegend #115529), CD127 (A7R34, BioLegend #135017), CD90.2 (30-H12, BioLegend #105337), TCRβ (H57-597, Invitrogen #17-5961-82), T-bet (4B10, BioLegend #644815), and ROR-gamma (RORγt) (B2D, Invitrogen #12-6981-82). For intracellular transcription factors, cells were stained for surface markers, followed by fixation and permeabilization with eBioscience Foxp3/Transcription Factor Staining Buffer Set (ThermoFisher # 00-5523-00). All flow cytometry experiments were performed using a LSRFortessa X-20 Cell Analyzer (BD Biosciences) running FACS Diva (BD Biosciences) and analyzed with FlowJo v.10.10 (BD Biosciences). Lineage was defined as CD11b⁺CD11c⁺CD19⁺CD4⁺CD5⁺TCRβ⁺. ILC were gated according to their phenotype as Lin⁻CD45⁺CD90.2⁺CD127⁺, ILC1 were gated according to their phenotype as Lin⁻CD45⁺CD90.2⁺CD127⁺T-bet⁺NK1.1⁺, ILC3 were gated as Lin⁻CD45⁺CD90.2⁺CD127⁺RORγt⁺KLRG1⁻, and Th17 CD4⁺ cells were gated as CD11b⁻CD11c⁻CD19⁻CD45⁺CD4⁺CD5⁺TCRβ⁺RORγt⁺.

Differences were assessed by unpaired, or paired (where applicable), two-tailed t test with a 95% confidence interval. Alternatively, a two-way ANOVA followed by Tukey’s post hoc test were performed. Differences with a p value <0.05 were considered significant and were indicated as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Consistent with previous publications [16‒18], pregnant dams assigned to the HTX group developed a profile in which TSH and T₃ remained steady and only T₄ showed a significant decrease on gestational day 14, while in their EU counterparts all thyroid hormones remained steady (online suppl. Fig. 1; for all online suppl. material, see https://doi.org/10.1159/000545578). To corroborate gestational HTX effect on the offspring, hippocampal function was evaluated using the marble burying test. Percentage of marbles buried by each group revealed that the HTX offspring buried significantly less marbles than their EU counterpart (Fig. 1), a pattern consistent with previous reports [27] and compatible with reduced hippocampal activity [28]. Subsequently, EAE was induced, and mice were followed for EAE scoring until day EAE 7. Standard EAE scoring showed no difference between the HTX and EU groups until day EAE 7, confirming a presymptomatic stage for both groups (data not shown).

To assess the inflammatory environment surrounding the presymptomatic stage of EAE we assessed circulatory levels of Th17 signature cytokine IL-17 and Th1 signature cytokine IFN-γ at baseline (day 0) and 7 days after EAE induction (Fig. 2). Measurements of IL-17 revealed a significant difference at baseline and 7 days after EAE induction, with the HTX offspring displaying significantly lower levels, compared to their EU counterpart (Fig. 2a). Moreover, no significant differences were observed for either the HTX or EU offspring at the same time points of the presymptomatic stage. Conversely, while measurements of IFN-γ revealed no significant differences at baseline, levels of circulating IFN-γ showed a 5-fold increase in the HTX offspring 7 days after EAE induction, with no significant change observed in the EU group induced with EAE (Fig. 2b). These results uncover a balance that favors Th1-associated cytokine IFN-γ over Th17 signature cytokine IL-17 at the systemic level in the HTX offspring 7 days after EAE induction.

To further elucidate Th1/Th17 balance and ILC contribution during the presymptomatic stage of EAE, and because MS and EAE pathogenesis have been linked to the gut microbiome composition as a source of immunoregulation to the progression of neuroinflammation [29, 30], ILC1 and ILC3 were quantified in the CNS, the intestinal tract, and secondary lymphoid organs, with significant differences evidenced only in the brain and the spinal cord (online suppl. Fig. 2A, B). Quantification of ILC1 in the brain and spinal cord from EU and HTX offspring revealed no significant differences at baseline (Fig. 3). Then, 7 days after EAE induction, a comparable increase in ILC1 content was observed in the brain (Fig. 3a) and spinal cord (Fig. 3b) of the EU offspring and a significant increase was observed in both organs of the HTX offspring (Fig. 3). In fact, ILC1 content in the spinal cord of the HTX offspring induced with EAE was observed to be significantly higher than in the rest of the groups (Fig. 3b). Quantification of ILC3 in the brain and spinal cord in EU and HTX offspring revealed a comparable but not statistically significant decrease 7 days after EAE induction (Fig. 4). Due to their pivotal role in EAE pathogenesis [31, 32], Th17 CD4⁺ cells were also quantified, for which significant differences were also evidenced only in the brain and the spinal cord (online suppl. Fig. 2C). Quantification of Th17 CD4⁺ cells in the brain and spinal cord in EU and HTX offspring revealed no differences at baseline. However, a significant increase was evidenced only in the brain of EU animals induced with EAE (Fig. 5a). This response was not observed in their EU counterparts (Fig. 5b). Gating strategy and representative event differences between EU and HTX offspring are presented in online supplementary Figures 3 and 4. Taken together, these results reveal that significant changes in ILC1 and Th17 CD4⁺ cell populations occur only in the CNS, with a marked presence of only ILC1 – and not ILC3 or Th17 CD4⁺ cells – in the brain and spinal cord of the HTX adult offspring, suggesting a Th1 shift in the Th1/Th17 balance during the presymptomatic stage of EAE.

To our knowledge, this is the first report not only describing the balance between Th1 and Th17 responses in terms of ILCs and Th17 CD4⁺ cells but also through surrogate cytokines at a presymptomatic stage of EAE. This also accounts as the first report describing such cellular and cytokine balance in animals exposed to low levels of T₄ in utero and their EU counterparts.

Consistent with the theory of fetal programming [33], our results iterate on warning about the long developing effects that HTX has over the offspring. In fact, behavioral effects in the unchallenged HTX offspring include anxious-like behavior, a subordinate state, and impaired social interaction, as measured using marble burying, elevated plus maze, tube dominance, and three-chamber social preference tests [27], and immune effects include a stronger response to human metapneumovirus [16] and an earlier and stronger response to both ulcerative colitis [17] and to a mild form of EAE [18]. Although the biological mechanisms underlying such effects have not been elucidated yet, our hypothesis relates thyroid hormone receptors TRα1 and TRβ1 binding as heterodimers (with the retinoid X receptor) with thyroid hormone response elements [34] to regulate the expression of genes directly related to the Treg/Th17 balance [35]. Moreover, such effect should take place during the early phases of the immune ontogenesis [36] and should be able to endure over time, possibly through an epigenetic mechanism [37], creeping until challenged, and showing with varying degrees of severity. If thyroid homeostasis is at the root, and because pregnancy awareness [38] overlaps late with the immune ontogenesis of the fetus [39], then adequate screening for thyroid hormones that may proceed asymptomatic for the mother when out of range [1] should be at the center of any public health program.

The marked and converse effect observed in the CNS between ILC1 (Fig. 3) and ILC3 (Fig. 4) on day 7 of EAE induction represents an additional hint about the mechanism by which HTX exerts its immunomodulatory action over the offspring. Recent studies describe a complex dependency on traditional transcription factors T-box transcription factor TBX21 (T-bet) and nuclear receptor RORγt, and less described transcription factors tissue-resident T-cell transcription regulator protein ZNF683 (Hobit), zinc finger protein Aiolos (Aiolos), B-cell lymphoma 6 protein homolog (Bcl6), nuclear receptor ROR-alpha (RORα), transcription factor Maf (c-Maf), and hypoxia-inducible factor 1-alpha (HIF-1α) [40, 41] underlying ILC1/ILC3 cross-plasticity. While ILC1 phenotype is conventionally maintained by T-bet and Hobit and unconventionally by Aiolos and Bcl6, ILC3 phenotype is maintained by RORγt, RORα, c-Maf, and HIF-1α [21, 42‒44]. Interestingly, expression of these transcription factors not only maintains independent phenotype balance but also underlies ILC1/ILC3 counterbalance. Thus, by extension, HTX immunomodulation legacy may involve in utero programming of the cross-repressing mechanism between RORα/RORγt against T-bet [45] and/or c-Maf/HIF-1α against T-bet/Bcl6 [46‒49], significantly increasing the degree of complexity of ILC1/ILC3 inflammatory responses at adult ages. Interestingly, ILC1/ILC3 counterbalance has been investigated using animal models under the context of ulcerative colitis [50], hypoxia [48, 49], bacterial challenge (Citrobacter rodentium [45] and Clostridium difficile [49]), and straight loss-of-function [45‒47], while, to the best of our knowledge, EAE and its presymptomatic stage remains a model yet to be explored for ILC1/ILC3 transcription factor counteraction. Thus, although broad and blunt, our results constitute the first step towards a comprehensive account of ILC1/ILC3 control during the presymptomatic stage of EAE.

The significant expansion of ILC1 observed in the CNS 7 after EAE induction (Fig. 3) represents an interesting readout where input from local expansion of tissue-resident niches of ILC meet input from distant organs. For the first half of the last decade, mature ILC were considered as tissue-resident, «sedentarily» residing within tissues without any possibility for migration [51‒53]. However, more recent observations challenge this notion with notable examples highlighting the mechanistic capacity for migration of ILC by showing a tissue- and subset-specific expression of homing receptors [54], directly demonstrating ILC migration capacity from the intestine to the mLN [55, 56], and directly demonstrating ILC migration capacity from circulation to the CNS during EAE-induced neuroinflammation [19]. In fact, ILC1 have been proposed as the migratory ILC subset displaying behavioral similarities to conventional NK cells, migrating into peripheral lymph nodes in a CCR7-dependent manner, egressing from peripheral lymph nodes in a S1P-dependent manner, and homing into target tissues in a CCR6-dependent manner [57]. Notably, CCR6 ligand CCL20 has been found elevated in the spinal cord and lymph nodes of EAE mice (and in peripheral blood of MS patients [58]) in a severity-dependent manner [59] and has been deemed fundamental to initiate EAE [60], as Ccr6^−/− mice or mice treated with neutralizing anti-CCR6 antibodies showed resistant to EAE progression [59]. Thus, it is plausible to propose early CNS inflammation and CCL20 production at blood-cerebrospinal fluid barriers, with the choroid plexus acting as entry site [61], as one of the initial events in EAE progression, leading chemotactic ILC1 recruitment in a CCR6-dependent manner. However, the absence of differences observed in peripheral tissues (online suppl. Fig. 2A, B) suggests that the differential ILC response in the CNS of the HTX offspring 7 days after EAE induction is probably due to an increased recruitment of ILC precursors exhibiting a phenotype dissimilar to mature populations [62], as hematopoietic stem/progenitor cells are attracted to inflammatory environments and have the capacity to differentiate in situ into ILC phenotypes [63]. Although hints of migration of ILC populations may be drawn from our tissue- and subset-specific assessment (Fig. 3, 4; online suppl. Fig. 2), further research using a different strategy under the scope of cell trafficking of both mature and progenitor ILC is required to better explain the differences in ILC1 and ILC3 presented here for the CNS of the HTX offspring (Fig. 3, 4).

Canonical theories on the pathogenesis of EAE strongly connect the Th17 pathway to clinical progression through IL-17 and autoantigen-specific encephalitogenic Th17 CD4⁺ cells [64, 65], which are considered essential for its development [31]. However, such progression has been widely accepted to affect the spinal cord white matter during EAE, as opposed to the cerebral and cerebellar cortex of the brain during MS [66]. In light of our results, it may be plausible to propose a modification to the three-compartment hypothesis [67] considering day 7 after EAE induction as the time point past afferent compartment Th17 CD4⁺ cell expansion (i.e., the spleen, online suppl. Fig. 2C) and the time point of proximal target compartment accumulation (i.e., the brain, Fig. 5a) prior to migration to distal CNS compartments (i.e., the spinal cord, Fig. 5b) and draining compartments (i.e., draining lymph nodes). The latter while circulating levels of IL-17 and IFN-γ (Fig. 2), and CNS ILC1 (Fig. 3) and ILC3 (Fig. 4) remain steady. Interestingly, when HTX is used to exacerbate and propel this immune response, the three-compartment hypothesis appears to shift from Th17 to Th1, with no proximal target compartment accumulation of Th17 CD4⁺ cells (Fig. 5), a significant accumulation of ILC1 (Fig. 3) and a virtual effacement of ILC3 (Fig. 4) in both target compartments. The latter while circulating levels of IL-17 are significantly reduced (Fig. 2a) and levels of IFN-γ are significantly increased (Fig. 2b).

During the last decade, the canonical proinflammatory and pathogenic label bore by IFN-γ during EAE has been challenged [68, 69]. In fact, a protective role against EAE has been reported via direct systemic IFN-γ injection [70], through anti-IFN-γ monoclonal antibody neutralization and loss-of-function [71], using Ifng^−/− mice with IFN-γ loss-of-function [72, 73], and IFN-γ receptor knock-out loss-of-function [74]. However, this disease-limiting role has been restricted to later stages of EAE, while a pathogenic role is still attributed to this cytokine in the early stages of EAE [75]. Similar to observations of asymptomatic herpes simplex virus type 1 (HSV-1) infection [76, 77], the earlier and stronger responses reported previously [17, 18] and presented here elevate the HTX model as ideal for evaluating immune events from future stages. Thus, while pathogenic IFN-γ did not rise significantly in the EU offspring 7 days after EAE induction, circulating levels of IFN-γ appeared unrestrained in their HTX counterpart (Fig. 2). Given their potential to secrete IFN-γ in the CNS [78], ILC1 emerge as a potential source of this cytokine and a potentially new cellular player adding to the dynamic picture configured at presymptomatic stages of EAE. Hence, our results align with previous reports and include novel insights regarding ILC participation, which add to the participation and contribution of NK cells, M1 macrophages [75], microglia [79], neutrophils, and myeloid-derived suppressor cells [68] to the development of EAE and possibly to MS.

In this report, EAE is further characterized as a complex phenomenon in which a fine balance between Th1 and Th17 plays a fundamental role at the presymptomatic stage defining its pathogenic course. A significant participation of ILC1 in the CNS and Th17 CD4⁺ cells in the brain 7 days after EAE induction was identified. Additionally, using the HTX model to exacerbate and propel the immune response, a strong increase of pathogenic IFN-γ was detected in circulation, while a significant increase of ILC1 and an invariant response of Th17 CD4⁺ cells was identified in the CNS and the brain, respectively. Overall, while further warranting the detrimental effects of asymptomatic gestational HTX, the presymptomatic stage of EAE is presented as a neglected frame holding an enormous potential to determine its pathological course with ILCs presented as salient cellular players. Further research will be required to gain detailed insights about this stage, to extrapolate early observations to MS, to search for potential biomarkers, and to propose earlier therapies.

We are indebted to Mrs. Johana Santos for her dedicated support keeping the experimental animal subjects under optimal conditions.

This study protocol was reviewed and approved by the Bioethics Committee from Universidad Andrés Bello, protocol Approval No. 008/2022.

The authors have no conflicts of interest to declare.

This study was supported by the National Agency for Research and Development (Agencia Nacional de Investigación y Desarrollo, ANID) through grants FONDECYT Postdoctorado 3220565 (S.G.), FONDECYT Regular 1231851 (A.M.K.), FONDECYT Regular 1231905 (S.M.B.), FONDECYT Regular 1240971 (P.A.G.), FONDECYT Regular 1231866 (J.A.S.), PAI SA77210051 (J.A.S.), FONDECYT Regular 1250827 (F.M.-G.), FONDECYT Regular 1191300 (C.R.). Support was also granted received by grants from the Millennium Institute on Immunology and Immunotherapy (MIII),  Programa ANID-ICM, ICN2021_045 (former ICN09_016, P09-016-F) (A.M.K., C.A.R., F.S., L.J.C., P.A.G., S.M.B.).

Sebastian Gatica, Nicolas Paillal, Ma. Andreina Rangel-Ramírez, Luis Méndez, Alonso Fernández-Tello, Claudia A. Riedel, Felipe Melo-Gonzalez: formal analysis, data curation, visualization, original draft preparation, review, and editing. Alexis M. Kalergis, Susan M. Bueno, Pablo A. González, Jorge A. Soto, Felipe Simon, and Leandro J. Carreño: data curation, funding acquisition, project administration, supervision, writing, reviewing and editing.

Raw data supporting the findings of this study are not publicly available due to intellectual property reasons, but summaries are available from the corresponding authors upon reasonable request.