Introduction

Multiple sclerosis (MS) is a chronic inflammatory life-threatening autoimmune disease of the central nervous system (CNS), causing progressive neurological disability and considerably undermining the quality and longevity.1 Molecular mimicry has been implicated in the initiation of MS,2 but the underlying cause and the complex immune regulation of MS pathology remain elusive.3 Although available disease-modifying therapies can alter the course of the disease, their efficacy over a long-term period is still insufficient and followed by adverse effects,4 emphasising the need for developing more efficient immunotherapies for MS. The driving mechanisms of MS pathogenesis, and its animal model experimental autoimmune encephalomyelitis (EAE), were shown to include a disbalance in the effector and regulatory arms of multiple immune cell subsets, including T, B, natural killer (NK) cells, and particularly myeloid cells.3,5,6 Moreover, a dysregulated interplay between gut microbiota and the immune system was described as critical for the regulation of MS and EAE pathogenesis.7,8 The gut microbiome is a community of about 1014 different populations of microorganisms inhabiting the intestinal tract (bacteria, viruses, fungi, etc). with all its interactions and activities,9 thus representing the largest environmental factor for autoimmune diseases in genetically susceptible individuals.10 Accordingly, the hygiene hypothesis may well frame an increase in autoimmune disease incidence in environments with less infectious agents and poorer microbial diversity, causing dysregulated immune responses.10

Lack of exposure to infectious agents, like parasite worms - helminths, has been proposed as one of the important causes for dysregulation of the immune system in MS.11,12Trichinella spiralis is a unique parasitic worm that completes its whole life cycle in one host, residing in the Nurse cell (transformed muscle cell) and communicating with the host organism through excretory-secretory products of the muscle larvae (ESL1).11 Through ESL1 antigens, this parasite creates an environment dominated by anti-inflammatory and regulatory immune responses, unfavourable for developing autoimmune diseases.13–15 We have found previously that the infection with T. spiralis, by applying its ESL1 products, alone or via dendritic cells (DCs), can effectively ameliorate EAE.14,16,17 By triggering the regulatory cytokines (IL-10 and TGF-β) in tolerogenic DCs via different innate receptors and lowering the pro-pathogenic T helper (Th)1 and Th17 cells responses while increasing the number and functions of regulatory T cells (Tregs),18,19 ESL1 antigens can restore dysregulated immune response.14,16,17,20 ESL1 products represent a complex mixture of at least 43 major glycoproteins, including their post-translationally modified isoforms.21 The three glycoproteins (53, 49 and 43 kDa) bearing immunodominant epitopes were demonstrated as potent anti-inflammatory proteins able to induce tolerogenic DCs,22 via toll-like receptor (TLR)2, TLR418 and DC-SIGN.19 ESL1 also contains galectin-1-like proteins which, similar to Galectin-1, increase the capacity of DCs to induce Tregs.23 Furthermore, we were the first to demonstrate that ESL1 also contains 30–80 nm extracellular vesicles (EVs) with a potent immunoregulatory capacity in human cell cultures in vitro.24T. spiralis-derived EVs alone contain complex structure,25 and they demonstrated the capacity to suppress inflammation in a mouse model of colitis.26 These results illustrate that ESL1 antigens contain different complex components that could be harnessed to effectively ameliorate inflammation in autoimmune diseases, including MS.20 However, this complexity of ESL1 antigens represents a great challenge for their massive production, which is required for their clinical use. Therefore, new innovative approaches are needed to enhance the therapeutic potency of ESL1 products.

Advances in nanotechnologies enabled the development of highly biocompatible drug delivery systems, which can enhance the efficacy of drugs while minimising their toxicity, overall improving drug formulation, delivery efficiency, and manufacturing processes.27 Thereby, biodegradable polymer-based nanomaterials, such as poly (lactic-co-glycolic) acid (PLGA), were shown to improve biodistribution and absorption of therapeutics to the diseased site.28 In contrast to widely studied PLGA micro/nanoparticles (NPs), displaying either anti-inflammatory or proinflammatory effects,29 the immunomodulatory properties of electrospun PLGA nanofibers are less well known. Biodegradable polymers exhibit excellent properties for drug delivery systems, including a high surface-to-volume ratio, high potential for drug encapsulation, easy and low-cost production, and simple surface functionalization chemistry. Nanofibers offer several advantages over nanoparticles, as they have a significantly larger surface area-to-volume ratio, lightweight with small diameters, and controllable pore structures, and generally have better mechanical properties, including modulus elasticity, strength, flexibility, and ease of production in large quantities making nanofibers more accessible for industrial applications.30 Nanofibers with core-shell structure fabricated by emulsion or co-axial electrospinning process31,32 can provide a better controlled and sustained release behaviour than those by emulsification process of the nanoparticles that have a burst drug release profiles.30 Moreover, we showed previously that nanofibers made from poly (ε-caprolactone) and PLGA-based nanofibers prepared by sequential electrospinning are excellent drug delivery systems for timely-controlled drug release, enabling 9% of the drug release within 60 days,33 suggesting the possibility of a continuous drug release for up to 2 years. The approval of PLGA-based biodegradable polymers for human application34 opened the opportunity to prepare an ESL1-based implant that could continuously modulate the autoimmune process in patients. The emulsion electrospinning creates nanofibers with a core-shell structure through a single nozzle. This method uses a stable mixture of two or more immiscible liquids, stabilized by surfactants. In the process, one liquid forms the outer shell of the fibers, while the other forms the core. The processing conditions for emulsion electrospinning differ from blend and co-axial methods, leading to unique nanofiber properties.35,36 Therefore, this study aimed to prepare PLGA/ESL1-based nanofiber implants by emulsion electrospinning and investigate whether they can regulate inflammatory processes in an animal model of relapsing EAE. Additionally, we evaluated the effects of PLGA/ESL1 implants on pathogenic and regulatory immune cell subsets, both at the systemic and CNS levels, as well as their effects on gut microbiota composition, all of which represent promising therapeutic targets of imbalanced immune responses in autoimmune diseases such as MS.

Materials and Methods Preparation of ESL1 Antigens

Adult male Wistar rats and Dark Agouti (DA) rats were obtained from the Military Medical Academy (MMA, Belgrade, Serbia) and housed under standard conditions of a 12/12 light/dark cycle in an animal facility with ad libitum access to water and food. All procedures and experiments involving animals were conducted approved and performed following the guidelines and regulations of the local ethics committee of the Institute for the Application of Nuclear Energy—INEP, University of Belgrade, Serbia, Belgrade (No. 323–07-11,160/2019-05/1 and No. 323–07-00758/2021-05/1). Wistar rats, 2–4 months old were used to propagate T. spiralis strain ISS 7564. Rats were infected by gastric inoculation with 8500 T. spiralis muscle larvae (L1 larvae). L1 larvae were recovered after 2 months by digesting the carcasses in pre-warmed gastric juice (1% pepsin in 1% HCl, pH 1.6–1.8).37 Muscle larvae were kept under controlled conditions (37 °C, 5% CO2) in complete Dulbecco’s modified Eagle medium (Sigma-Aldrich), supplemented with 10 mm HEPES, 2 mm L-glutamine, 1 mm sodium pyruvate, and 50 U/ mL of penicillin–streptomycin (all from Sigma-Aldrich) for 18 h. ESL1 products were purified from culture supernatants according to a previous procedure,18,19,22–24,38 and kept at −20°C for further use.

For tracking experiments, ESL1 in a quantity of 2 mg was labelled with Fluorescein 5-isothiocyanate (FITC) (Sigma) using 15 µg of FITC per 1 mg of protein. FITC was dissolved in 0.15 M Na2HPO4 x H2O buffer (pH 9.0) and slowly added to the ESL1 solution with continual mixing on the magnetic stirrer device, providing labelling of the protein for 90 min at room temperature. Purification of the labelled ESL1 was performed by gel filtration using a Sephadex G-25 column. Upon separation, the absorbance of FITC-labelled ESL1 (ESL1-FITC) was measured at 280 nm and 495 nm using a spectrophotometer (Ultrospec, Pharmacia Biotech), thus providing information on the protein concentration (Cp=1.4 mg/ mL) and fluorescein/protein ratio (F/P=3.0).

Preparation and Characterization of PLGA/ESL1 Implants

PLGA (average molecular weight MW 30,000–60,000 g/mol), with a mass ratio of lactide: glycolic acid units of 50:50, span-80, N, N-dimethylformamide (DMF), and chloroform (CHCl3) were purchased from Sigma-Aldrich (St. Louis, MO, USA). PLGA 50:50 (PLA/PGA) was chosen because of its predicted degradation rate of 1–2 months,33 which is suitable for the EAE model (lasting up to 1–1.5 months) and can be controlled by the amount of glycolic acid and hydrophilicity of the matrix.33,39 First, PLGA (2.0 g) was dissolved in a chloroform/DMF mixture (8.25/2.75 g) and magnetically stirred at 200 rpm at room temperature for 24 h. After adding biodegradable surfactant Span-80 (33.0 mg) to this solution, ESL1 (or ESL1-FITC) products in 2 mL PBS (2.5 mg/mL) were added, and the mixture was stirred for 2 h. Water-in-oil (W/O) emulsion was prepared as an electrospinning solution, and the oil and water phases were prepared separately. Thus, the emulsion containing 20 wt. % PLGA and 0.25 wt.% ESL1 was prepared (relative to the weight of the polymer). Control PLGA implants were prepared similarly, but without the ESL1 products, by dissolving the polymer in a suitable solvent (solution electrospinning). A vertical electrospinning experimental setup (CH-01, Linari Engineering, Pisa, Italy) was used to prepare nanofibers. To obtain fine nanofibers, the electrospinning conditions were optimised as follows: polymer solutions were ejected from a 20 mL plastic syringe into a metallic needle (1 mm inner diameter) at a distance of 10 cm from the collector, with a flow rate of 3 mL/h and an applied voltage of 20 kV. Electrospun nanofibers were collected on a piece of grounded aluminium foil on a surface of ~100 cm2 containing a total of 5 mg ESL1 (50 µg/cm2).

The morphology of the nanofibers was examined using a field-emission scanning electron microscope (FE-SEM) at 20 kV in conjunction with a Tescan Mira3 XMU microanalysis system (Tescan Orsay Holding AS, Brno, Czech Republic). In each system, 50 nanofibers from five FE-SEM images were selected, and the micrographs were analysed using Image-Pro Plus 6.0 software (Media Cybernetics, Rockville, MD, USA) to obtain their diameter distribution.

The surface chemistry of the prepared nanofibrous material and ESL1 was analysed using Attenuated Total Reflection Fourier-Transform Infrared (ATR–FTIR) spectrometry (Nicolet iS10 (Thermo Fisher Scientific Inc., Waltham, MA, USA). The transmittance spectra were measured in the range of 4000–500 cm−1 with a spectral resolution of 4 cm−1 following 32 scans. The results obtained from ESL1 were analysed using the spectral analysis software OMNICTM (Thermo Fisher Scientific Inc). in the range 1800 cm−1–1400 cm−1.

The glass transition temperatures (Tg) of the PLGA and PLGA/ESL1 nanofibers were determined using differential scanning calorimetry (DSC) (Q10, TA Instruments, USA). The measurements were conducted in the temperature range of 30 °C to 70 °C at a heating rate of 10 °C/min. The sample weight used was 4 ± 0.5 mg, and the heat flow was monitored under a nitrogen purge gas flow of 50 mL min−1. The Tg values were obtained by determining the midpoint between the intersection of the two parallel baselines before and after Tg. Universal Analysis 2000 software was used to calculate the Tg, and a minimum of three measurements were performed for each sample to obtain the average value. The presence of ESL1-FITC in the PLGA nanofibers was also analysed by fluorescence microscopy (Carl Zeiss AxioImager A1, Jena, Germany), in which the ESL1-FITC/PLGA nanofibers were collected on microscopic slides during electrospinning.

ESL1-FITC Release From PLGA in vitro

To measure in vitro release of ESL1-FITC from PLGA nanofibers, nanofiber sheets were cut into squares (1 cm × 1 cm) containing approximately 50 µg of FITC-ESL1 and placed in 24-well sterile microplates containing 1 mL of phosphate buffer (PBS, pH 7.4): wound exudate mixed in a 1:1 ratio (1 mL/well) (WE) to model the inflammatory microenvironment of the implant. The wound exudate was isolated from a sterile inflammation site induced by the subcutaneous implantation of polyvinyl sponges into DA rats (1 cm x 1.5 cm x 0.25 cm), as described previously.40 Briefly, two sponges per animal were implanted at the dorsal site of the skin under general ketamine/xylazine anaesthesia. After 3 days, the animals were euthanised, the sponges were harvested and squeezed with a syringe, and cell-free exudates were collected by pelleting the cells by centrifugation at 2000 rpm for 10 min.

The PLGA/ESL1 nanofibers were incubated at 37 °C and every 3 days for a total of 42 days, 0.5 mL of 50% WE from each well was removed for the analysis and an equal volume of fresh WE mixture was replenished. For blank controls, an equivalent amount of control PBS/PLGA nanofibers was incubated and the supernatants were collected. The fluorescence emission was assessed at 519 nm using a spectrofluorometer (Shimadzu). The concentration of ESL1 was calculated based on the measured fluorescence, according to the standard curve of ESL1-FITC. The percentage of ESL1 released from PLGA nanofibers per 3 days and the cumulative release were calculated accordingly.

The amount of ESL1 encapsulated in the PLGA nanofibers was measured to assess the encapsulation efficiency, defined as the ratio of ESL1 encapsulated in the PLGA nanofibers to the initial quantity of ESL1 added to the spinning solution. Furthermore, the drug loading was calculated by determining the amount of ESL1 encapsulated in the PLGA nanofibers relative to the total amount of PLGA ESL1 nanofibers.

ESL1-FITC Uptake in vitro

Splenocytes (1x107) isolated from DA rats were placed in 2 mL of 6-well plates in complete RPMI 1640 medium (supplemented with 10% FCS, l-glutamine, 2-mercaptoethanol (all from Sigma), and antibiotics (penicillin, streptomycin, Pan Biotech)). The cells were incubated at 37 °C in a humidified incubator with a 5% CO2 atmosphere for 3 h and then treated with ESL1-FITC or non-labelled ESL1 (50 µg/ mL) for the next 24 h. After each time point, the were washed in PBS/0.01% NaN3 and stained with anti-CD11b-APC and anti-MHC class II (I-A) (OX-6)-phycoerythrin (PE)(both from Bio-Rad, Hercules, California, USA), as described by flow cytometry. The percentage of ESL1-FITC positive cells within the CD11b+ MHC class II+ population (myeloid antigen-presenting cells (APC)) was evaluated in splenocytes treated with unlabelled ESL1 at each time point.

Experimental Treatments and EAE Induction

PLGA/ESL1 nanofiber sheets were cut into squares of 1×1 cm (thickness ~ 0.1 cm) with approximately 50 µg of ESL1 per membrane piece. Seven pieces were slightly compressed and implanted subcutaneously in the suprascapular region under sterile conditions; therefore, each treated animal received 350 µg ESL1 (PLGA/ESL1 group). Namely, DA rats were completely anaesthetised with ketamine (200 mg/kg) and their backs were shaved and disinfected with 70% alcohol. A small incision was made in the lower back and the implants were placed deep in the suprascapular region with tweezers. The skin was sutured and disinfected and the animals were monitored daily by a veterinarian. Control groups included animals treated with control PBS/PLGA nanofibers (PLGA group), sham-operated animals treated subcutaneously with ESL1 (350 μg/animal in 200 µL PBS) at the same time (ESL1 group), and sham-operated animals (control group).

EAE was induced as described previously,16,41 7 days after the experimental treatments. Briefly, DA rats were treated with a single intradermal injection (0.1 mL) into the right hind footpad with an encephalitogenic emulsion prepared by mixing rat spinal cord (SC) tissue homogenate (50% w/v in saline) emulsified in an equal volume of complete Freund’s adjuvant (CFA; Difco, Detroit, MI) supplemented with 4 mg/ mL Mycobacterium tuberculosis (Difco). Eight days post-immunization (pi), rats were weighed and scored daily for signs of the disease according to a scheme, 0 - no clinical signs, 1 - flaccid tail, 2 - hind limb paresis, 3 - complete bilateral hind limb paralysis and incontinence, and 4 - moribund state or death. Intermediate scores (± 0.5) were assigned if neurological signs were between typically observed symptoms. Several disease parameters were examined to evaluate the severity of EAE: the cumulative index (the sum of the daily mean clinical scores for each animal over a given number of days throughout the monitoring period), maximal clinical score (the maximal clinical score that each animal developed throughout the experiment), day of onset, and duration of the disease (total number of days with EAE symptoms). Additionally, the total incidence of EAE, as well as the incidence of a particular score (ie 2-paresis or 3-paralysis), was calculated using log-rank analysis. Day 14th upon EAE induction showed a peak of the disease, whereas on day 20th the animals entered the remission phase. An additional peak, relapse, was observed between days 24–28. For each experiment, 5 animals per group were euthanized for cell isolation from spleens and CNS, either at the peak of the disease (day 14th) or during the recovery phase (day 20th) for further analysis, and at least two independent experiments were carried out, as indicated in figure legends.

Cell Isolation and Cultivation

Single-cell suspensions from the spleen, lymph nodes, and CNS isolated at the peak and recovery of EAE were prepared. The cells infiltrating CNS were obtained from the spinal cords (SC) of rats perfused with sterile PBS, as described.41 The spinal cords were homogenized, centrifuged at 100 g, 10 min, +4°C, and the pellet was resuspended in 3 mL of 40% Percoll (Sigma-Aldrich) and overlaid on 3 mL of 70% Percoll gradient. Following centrifugation at 850 × g for 50 min, mononuclear cells were recovered from the 40% / 70% Percoll interface and washed with RPMI medium (Sigma-Aldrich). Spleen and lymph node cells were isolated aseptically, and the spleen cells were lysed with a lysis buffer (155 mm NH4Cl, 12 mm NaHCO3, 0.1 mm EDTA in ddH20) for 5 minutes, followed by washing in PBS. The total number and viability of the cells were determined using a Cell Count Viability kit on a Cell Muse analyser (Cytek Biosciences), and the cells were prepared for flow cytometry analysis. Additionally, splenocytes isolated from PLGA/ESL1 or sham-operated control animals at the peak or recovery of EAE were seeded in 24-well plates (2 x 106/ mL) and cultivated in complete RPMI-1640 medium for 48h to assess the spontaneous production of cytokines in culture. After culturing, cell-free supernatants were collected and stored at −20 °C until analysis by ELISA.

ESL1-FITC/PLGA Tracking in vivo

To assess the release of ESL1-FITC from PLGA nanofibers in vivo, DA rats were implanted with either ESL1-FITC/PLGA or PLGA/ESL1 as a control (21 animals/group). After 3, 4, 7 or 14 days, the animals were euthanized for collection of the peri-implant region (on day 4, 2 animals per group), draining lymph nodes from the axillary (AxLN), cervical (CLN), and inguinal (IngLN) regions, spleen, Payer’s patches, and spinal cords (on days 3, 7 and 14 after implantation, 4 animals per group in each term). Additionally, four animals per group were immunised 7 days after implantation with SC homogenate in complete Freund’s adjuvant (CFA) to induce EAE, and the organs were collected 7 days after immunisation (ie 14 days after implantation). Single-cell suspensions were prepared as described previously, and the cells were stained with anti-CD11b-APC and anti-MHC class II (I-A)-PE (Bio-Rad) to monitor ESL1-FITC presence within myeloid antigen-presenting cells by flow cytometry. To compare the trafficking of released ESL1-FITC with the trafficking of free ESL1-FITC, 3 animals/group were injected with either 350 µg/mL of ESL1-FITC or non-labelled ESL1 subcutaneously in the same suprascapular region, and the myeloid APCs were analysed 3, 7 or 14 days after the injection. The percentage of FITC-positive cells within the CD11b+ MHC II+ population was evaluated according to the corresponding cell isolates from PLGA/ESL1- or free ESL1-treated animals.

Flow Cytometry

Single-cell suspensions prepared from the spleen and SCs were analysed by flow cytometry for surface and intracellular antigens. The cells were washed once in PBS containing 0.01% Na-azide (Sigma-Aldrich), followed by blocking Fc receptors with 2% normal rat serum (eBioscience, Frankfurt am Main, Germany) for 15 minutes, followed by labelling with fluorochrome-conjugated monoclonal antibodies (mAbs) at dilutions recommended by the manufacturer. The incubation for surface staining lasted 30 minutes at +4°C. After the surface staining, the cells were fixed and permeabilised using the fixation/permeabilization kit (R&D Systems) according to the manufacturer’s recommendation, followed by intracellular staining. Intracellular staining of IFN-γ, IL-10 and IL-17 was performed after 6-hour activation of the cell with phorbol-12-myristate-13-acetate (PMA, 20  ng/mL) and ionomycin (500  ng/mL) in the presence of monensin (2 µM) (all from Sigma-Aldrich). The following mAbs were used: mouse IgG1 negative control-Alexa Fluor 488 (F8-11-13), mouse IgG1 negative control-phycoerythrin (PE) (F8-11-13), anti CD68-Alexa Fluor 647 (ED1), anti-CD4-biotin (clone W3/25), anti-CD3-Alexa Fluor 647 (1F4), anti-CD11b-FITC (OX-42), anti-CD11b-APC (OX42), anti-MHC class II-PE (OX-6), (all from Bio-Rad, Hercules, California, USA); mouse IgG1 negative control-biotin (AB_2550616), anti-IL-17A-FITC (eBio17B7), anti-CD45R antibody-FITC (HIS24), anti-CD161-Biotin (10/78), CD86-Pe/Cyanin 5 (GL1), mouse IgG1 negative control (MOPC-21)-Alexa Fluor 647 (all form Thermo Fisher Scientific, Waltham, Massachusetts, USA); anti-CD25-PE (OX-39), anti-IL-10-PE (A5-4), anti-IL-4-PE (OX-81), anti-CD103-biotin (OX-62) (all form BD Pharmingen, San Diego, California, USA); anti-FoxP3-APC (FJK-16s), mouse IgG1 kappa negative control-Allophycocyanin (APC) (P3.6.2.8.1) (eBioscience); anti-IFN-γ-FITC (DB-1), anti-CD8-PECy7 (OX-8), anti-CD45-PE (OX1) PE/Cyanine7 Streptavidin, PerCP-Cyanine 5.5 streptavidin, APC/Cyanine7 Streptavidin, 7AAD viability dye (BioLegend, San Diego, California, USA); rabbit polyclonal anti-mouse IgG (whole molecule)-FITC (Sigma-Aldrich). For each analysis, at least 10,000 cells in the leukocytes gate were acquired on an 8-colours BD LSR II, gated according to their specific side-scatter (SSC)/forward-scatter (FSC) properties, 7-AAD- (viable cells), and avoiding cells with low SSC/FSC properties (predominantly dead cell debris). Non-specific fluorescence was determined using fluorescently-labelled isotype antibodies and fluorescent minus one-negative controls and the signal overlap was compensated for before each experiment using single-labelled cells.

Microscopy Analyses

The spinal cords (SCs) were isolated from 4 animals per group at the peak and recovery phases of EAE by opening the spine on the front and caudal sides of euthanised animals and then expulsing the SCs by strong aspiration with PBS. The lumbar region of the spinal cord was collected using a scalpel, submerged in buffered formalin (Sigma) for 48–72h, and then embedded in paraffin using the Modular Tissue Embedding Centre EC 350 (Especialidades Médicas Myr, Spain) after dehydration in ethanol and xylol. In addition, peri-implant regions and draining lymph nodes were isolated during ESL1-FTIC/PLGA tracking and formalin fixation experiments. Thin cross-sections (~ 5μm) were prepared using a Manual Rotary Microtome (Biobase, China) and the tissues collected on the slides were rehydrated and stained with Mayer’s Haematoxylin and Eosin Y (both from Fisher Scientific). Histological quantification of cellular infiltrates in SCs was performed using H&E-stained sections. The number of inflammatory lesions (infiltrates) and cells per infiltrate were counted per cross section. Inflammatory infiltrates were considered positive only if they were present in three or more consecutive cross sections. At least five slides, each containing 3–4 cross-sections, separated by ~50μm, were analysed for each animal.

Additionally, the slides were prepared for epi-fluorescent microscopy by antigen retrieval in citrate buffer (10mM Citric Acid, 0.05% Tween 20, pH 6.0) at 95°C for 20 min, followed by spontaneous cooling to room temperature. Rabbit anti-myelin base protein (MBP, 1:100; Abcam) and anti-rat CD45-biotin (clone 30-F11, BioLegend, 1:50) were incubated on slides overnight at 4°C, the slides were washed in PBS, and then incubated with secondary anti-rabbit IgG (H+L) -Alexa 488 (1:500) for 2 h and streptavidin – Alexa 546 (1:500) for 30 minutes (both from Thermo Fisher). Alternatively, peri-implant tissues from PLGA/ESL1- and PLGA/ESL1-FITC-treated animals were either H&E stained or incubated with anti-FITC antibody-biotin (BioLegend) (1:200) overnight and after washing in PBS, with streptavidin-Alexa 546. The slides for epifluorescent microscopy were then incubated with DAPI nuclear stain (Thermo Fisher 1:10000) for 15 minutes, washed in PBS, and mounted in a fluorescence mounting medium (Sigma) before analysis. The slides were analysed using a Zeiss AxioImager A1 under light microscopy or under a UV filter set for DAPI (UV-2B, ex: 330–380 nm, DM 400, BA 435), a green filter set for detection of MBP (B-2A, ex: 450–490 nm, DM 505, BA 520), and a red filter set for detection of CD45 or ESL1 antigens (G-2A, ex: 510–560 nm, DM 575, BA 590) in epifluorescent microscopy. The images for epifluorescent microscopy were acquired as monochromatic and later merged and analyzed offline using the ImageJ software (National Institutes of Health, Bethesda, Maryland, USA).

Cytokines Measurements

Cytokines were measured in the supernatants of 48h spleen cell cultures isolated from animals at the peak and recovery phases of EAE. The levels of IL-17, IL-6, IFN-γ, TNF-α, IL-4, and IL-10 were determined using sandwich ELISA kits (R&D Systems) according to the manufacturer’s protocol. The unknown concentrations of cytokines were determined from standard curves using 5-parameter non-linear fit curves (GraphPad Prism).

RNA Isolation and Quantitative Real-Time PCR (qRT-PCR)

Total RNA extraction from intestinal samples collected at the peak of EAE was performed using liquid nitrogen homogenization of tissue, followed by resuspension in 500 μL denaturing solution as described previously.42 The next steps included phenol/chloroform extraction, repeated twice, and RNA precipitation with isopropanol. The isolated RNA was treated with the Rapid Out DNA Removal Kit according to the manufacturer’s protocol (Thermo Fisher Scientific). The Revert Aid Reverse Transcriptase (Thermo Fisher Scientific) with random hexamers (Thermo Fisher Scientific) and Ribo Lock RNase inhibitor (Thermo Fisher Scientific) were used to transcribe 0.5 µg of isolated RNA. The target gene expression analysis was performed using 7500 real-time PCR system (Applied Biosystems) with IC Green qPCR Universal Kit (NIPPON Genetics) under the following conditions: 2 min at 95 °C activation, 40 cycles of 5s at 95 °C and 30s at 60 °C. All results were normalized against the β-actin (Actb) gene and expressed as relative target abundance by using the 2−ΔΔCt method.43 All primers for gene expression analysis, presented in Supplementary Table S1, were purchased from Thermo Fisher Scientific.

DNA Extraction, Determination of Segmented Filamentous Bacteria Abundance and Sequencing

The stool samples were collected from all rats before the EAE induction (ie 7 days after the implantation), and at the peak of the disease, and they were stored at −80 °C before extraction. Genomic DNA was extracted using ZR Fecal DNA MiniPrep Kit (Zymo Research, Orange, CA, USA) according to the manufacturer’s instructions. The integrity and concentration of extracted DNA were measured using a BioSpec-nano spectrophotometer (Shimadzu, Kyoto, Japan) and Qubit fluorometer (Thermo Fisher Scientific, USA). DNA concentrations were uniformed to evaluate the relative content of segmented filamentous bacteria (SFB) across animals. To assess the relative quantification of amplicons specific for SFB from isolated DNA, we performed qPCR analysis using primers listed in Supplementary Table S1. The relative SFB abundance was normalized to the total bacteria and expressed as a fold change of SFB in each animal before the induction of EAE and at the peak of the disease. Hypervariable V3-V4 regions of the bacterial 16S rRNA gene were amplified using the 341 F/806 R primer set as previously described,41 followed by library preparation, quality checking, and sequencing on the Illumina NovaSeq 6000 platform at Novogene (Beijing, China) in PE250 mode with 30000 tags per sample.

Microbiota Data Processing and Analyses

Based on their barcode sequences, paired-end raw reads were demultiplexed and truncated by removing barcodes and primers. High-quality demultiplexed raw reads were processed using QIIME2 v2020-844 to obtain clean reads for taxonomy analysis. We performed taxonomy assignment using the classify-sklearn plugin with SILVA 132 database as reference (99% sequence identity). The resulting files, together with the metadata, were imported into R Studio (v2022.07.2) (RStudio team) using the functions from package qiime2R (github.com/jbisanz/qiime2R). The function set seed was set at 999 to ensure reproducibility of the results. Phyloseq45 functions were used to calculate the alpha (Shannon index) diversity metric on a rarified sample size of 9400. Significant differences between groups were examined using one-way ANOVA followed by Tukey’s post-hoc test in GraphPad version 9.0.0 software for Windows (San Diego, California, USA).

Analysis of the composition of microbiomes (ANCOM) protocol46 was used to identify differentially abundant taxa at the different taxonomy levels. Species and genera with W scores above 1 were chosen for additional testing with a paired T-test or Pairwise Wilcoxon test in R Studio, and only statistically significant differences are shown.

Predictive functional profiles of the gut microbiota were generated using the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt)247 pipeline, based on the feature table and representative sequences with default options. To unveil discriminative predicted functional profiles before immunisation and at the peak of disease, PICRUSt2 results were examined using the linear discriminant analysis Effect Size (LEfSe) pipeline on the Galaxy platform,48 with parameters set to α<0.05, for the factorial Kruskal–Wallis test and all-against-all strategies for analysis between classes and only predicted pathways above the LDA threshold 2.5 were plotted. The normality of the distribution, for continuous variables, was checked using the Shapiro–Wilk test. All p values less than 0.05 were considered statistically significant (*p < 0.05, **p < 0.01, ***p < 0.001). The Benjamini-Hochberg method was used to correct the p-values for multiple comparisons. All plots were generated using the ggplot2 (v3.3.5) package and GraphPad software unless otherwise stated.

Raw sequencing data were deposited in the European Nucleotide Archive (ENA) (https://www.ebi.ac.uk/ena) under the accession number PRJEB79235 and secondary accession number ERP163422.

Statistical Analysis

The distribution of data points was analysed using the Shapiro–Wilk test, and then either one-way ANOVA with Tukey’s multiple comparison test or Kruskal–Wallis test was used to compare the relative mRNA expression and the percentage of immune cells within the tested groups at the peak or recovery of EAE. The relative distribution of immune cells was shown as obtained from flow cytometry analysis or normalized (in stack bars) to total gated cells for each analysis panel. The statistical significance in clinical scores between the time points across the tested groups was determined using two-way ANOVA with Dunnett’s post-hoc test. Disease severity parameters between the groups were examined using one-way ANOVA with Tukey’s multiple comparison test. The Log rank test was used to compare the individual EAE scores between the experimental groups. All statistical analyses were performed using GraphPad Prism (v9.1.0) (GraphPad Software).

Results Preparation, Characterisation and Release of ESL1 from PLGA Nanofibers

Due to easily controllable release properties33 and good biocompatibility,49 PLGA nanofibers have been used as platforms to deliver immunomodulatory ESL1 antigens. Therefore, according to our previous findings on the release dynamics of a model drug,33,50 the PLA: PGA 50:50 mixture was chosen for emulsion electrospinning with the aqueous phase of ESL1 antigens (2.5 mg/mL in PBS) for a targeted dose of 50 µg ESL1 per cm2 of PLGA.

To analyse the morphology of PLGA/ESL1 and control PLGA nanofibers, the diameters were measured from FE-SEM images (Figure 1A), as indicated previously.50 The data suggested that the average diameter of PLGA/ESL1 nanofibers was 0.431 ± 0.02 µm, with a size range between 0.1 µm and 1.2 µm. The microstructure of the control PLGA nanofibers consisted of a relatively uniform cylindrical shape with an average diameter of 0.387 ± 0.03 µm, with the same size range. A comparison of the ATR-FTIR spectra between PLGA/ESL1 and PLGA confirmed the presence of ESL1 within the nanofibers. The PLGA nanofibers spectrum showed low-intensity signals at 2998 and 2926 cm−1, which were attributed to asymmetric stretching of the C−H bond of the −CH2− and −CH−groups of the aliphatic chain of PLGA, respectively. The intense signal at 1747 cm−1 was attributed to the C=O stretching of the ester, and the signals at 1166 and 1087 cm−1 corresponded to the C−O single bond stretching of the ester group.51 The spectra of the ESL1 antigens showed a characteristic FT-IR spectrum with a broad signal at 3366 cm−1 attributed to the OH stretching of water. The low-intensity signals at 2955 cm−1 and 2929 cm−1 were attributed to the asymmetric and symmetric stretching of methylene C-H bonds, respectively. The absorption signal at 1637 cm−1 is representative of amide I owing to the stretching vibrations of the protein carbonyl group C=O groups. The signal at 1545 cm−1 is attributed to amide II due to N−H mixed with C−N bending. The weaker signal centred at 1263 cm−1 represents the COO stretching vibration of proteins. The symmetric stretching vibration of the O–C group at 1186 cm−1 was attributed to carbohydrates such as glycogen.52 The FT-IR of PLGA/ESL1 spectra did not reveal any significant difference between the functional groups of PLGA nanofibers. However, low-intensity signals at 1637 and 1545 cm−1 due to amides I and II confirmed the successful incorporation of ESL1 antigens in the nanofibers without losing their structural nature during emulsion electrospinning (Figure 1B). The DSC results were used to examine changes in the thermal behaviour of the PLGA nanofibers (Figure 1C). The results indicated that the glass transition temperature of PLGA nanofibers decreased from 47.5 ± 0.21 °C to 43.7 ± 0.42 °C compared to PLGA/ESL1 nanofibers. Thermal analysis of the emulsion nanofibers showed a small decrease in Tg values which demonstrated the plasticising properties of a non-ionic surfactant and the possibility of interactions between Span-80 and PLGA.53 This phenomenon also confirms the filling of PLGA nanofibers during coaxial electrospinning, according to previous findings.54

Figure 1 Characteristics of PLGA/ESL1 nanofibers and release dynamics of ESL1. (A) Representative FEI-SEM images of PLGA/ESL1 and PBS/PLGA nanofibers are shown, which were used for the measurements of fibre diameter (upper row). The frequency distribution of nanofiber diameters is shown with the fitting curve (lower row). (B) FTIR spectra of PLGA/ESL1 and PLGA are shown with peak transmittance indicated for relevant bonds. (C) DSC analysis of PLGA/ESL1 and PLGA are shown.

To evaluate the release dynamics of ESL1 from PLGA nanofibers, ESL1-FITC was encapsulated into PLGA fibres as described to obtain 50 µg of ESL1-FITC /cm2 of PLGA. The presence of ESL1-FITC in nanofibers was confirmed using epifluorescence microscopy (Figure 2A). The fluorometry measurements of WE collected every 3 days indicated that the released ESL1 was detectable after the first measurement, and its release increased during the first 27 days, up to a level of 1μg/ mL/day (Figure 2B). Subsequently, ESL1 was still released from PLGA but at a slower rate. Cumulatively, ~90% of the encapsulated ESL1 was released from PLGA in WE during the 40 days of monitoring, with a greater slope during the first period of monitoring (Figure 2C). Based on the obtained results, it was determined that the encapsulation efficiency and drug loading was 99 ± 0.5% and 0.2 ± 0.03%, respectively.

Figure 2 Release dynamics of ESL1 from PLGA in vitro. (A) Representative phase-contrast (upper) and fluorescence (lower) images of PLGA loaded with ESL1-FITC (40x magnification) are shown. (B) Release dynamics of ESL1-FITC (50 µg) from PLGA (1 cm2) incubated in 50% wound exudate in PBS over 44 days. Each dot represents the mean amount ± SD (from triplicates of one out of four independent experiments) of ESL1 collected every 3 days during the incubation. The amount of ESL1 was calculated from the standard curve (inset in the upper right corner). (C) A cumulative release of ESL1-FITC over 44 days is shown as mean ± SD (n=3).

ESL1 Released from PLGA Effectively Ameliorate EAE

We previously demonstrated that ESL1 antigens were most effective in alleviating relapse-remitting EAE symptoms induced in DA rats, when administered prophylactically in multiple doses before disease induction, compared to single-dose treatment.14 Therefore, to test whether the delivery of ESL1 antigens via degradable PLGA fibres could reproduce this phenomenon, a group of DA rats was implanted subcutaneously with PLGA/ESL1 fibres, containing a total of 350 μg ESL1 /implant, 7 days before disease induction (EAE_PLGA/ESL1). Control groups of animals were at the same time sham-operated and then treated with the equivalent amount of ESL1 subcutaneously (350 μg/rat) (EAE_ESL1), or implanted with control PLGA nanofibers (EAE_PLGA), or just sham-operated (EAE_ctrl), 7 days before disease induction.

After immunisation, the symptoms and weights of the animals were monitored daily for 38 days (Figure 3A and B). The control group (EAE_ctrl) developed a relapsing course of EAE, with initial symptoms appearing between 7 and 10 days after immunisation. The peak of the disease was evident on day 14, followed by partial remission by day 20th, and an additional relapse with a peak on days 24–27. In line with this, the greatest weight loss was observed in the EAE_ctrl group during the first 8 days following immunisation. A similar course of disease and weight changes were observed in animals from the EAE_ESL1 and EAE_PLGA groups. In contrast, the group of animals implanted with PLGA/ESL1 displayed the lowest reduction in weight %, and the lowest severity of EAE scores (Figure 3A and B). According to the area under the curve, only the EAE_ESL1/PGLA group displayed a significant difference compared with the other tested groups. A cumulative index score, as a measure of the average EAE score over the number of days of monitoring, also suggested that the EAE_ESL1/PGLA group displayed significantly lower EAE symptoms than the EAE_ctrl, EAE_ESL1, and EAE_PLGA groups (Figure 3C). The animals from the EAE_PLGA/ESL1 group displayed the lowest maximal clinical score, initial EAE symptoms appearance (day of EAE onset), and total duration of the disease (Figure 3C), all of which suggested a protective effect of PLGA/ESL1 implants in this EAE model.

Figure 3 Effects of PLGA/ESL1 implants on EAE symptoms. The soluble ESL1 (350 µg/dose), or implants of PLGA/ESL1 (containing 350 µg ESL1 / 7 cm2 PLGA) or control PBS/PLGA (7 cm2), were applied subcutaneously in the suprascapular region of DA rats. Seven days after the implantation, the rats were immunised to develop EAE, as described. (A) The weight of each animal was monitored for 18 days, and the data is shown as mean weight % ± SE (N (EAE_PLGA, EAE_ESL1) = 10; N (EAE_PLGA/ESL1, EAE_ctrl) = 20), from two or four independent experiments, respectively. The area under the curve (AUC) for weight %, calculated for each animal, is shown as mean ± SEM. (B) EAE scores over 38 days of monitoring are shown as mean ± SEM (N= as in (A)), and AUC of EAE scores for each animal is presented as mean ± SEM. (C) Cumulative index, maximal clinical score, day of disease onset and duration of the disease are shown as mean ± SEM (N= as in (A), or 10 for duration of disease) and each animal is presented by a dot. (A-C) *p<0.05, **p<0.01, ***p<0.005, ****p<0.001, as indicated (Kruskal–Wallis test with Dunn’s multiple comparison test). (D) Incidence of EAE symptoms (limp tail ≥1, paresis ≥2, paralysis ≥3) in each group during the first 22 days is presented as the log-rank curve. Likewise, the incidence of relapse (score ≥1 after day 22) is shown. *p<0.05 Long-Rank (Mantel-Cox) test vs EAE_ctrl group.

Log-rank analyses of incidence were performed to assess differences in EAE severity between the tested groups. The results showed that the incidence of EAE (any score) in the EAE_ctrl group, EAE_PLGA, and EAE-ESL1 groups was 100%, whereas only ~75% of animals developed the symptom in the EAE_PLGA/ESL1 group. The incidences of paresis (score ≥2) and paralysis (score ≥3) were also significantly lower in the EAE_PLGA/ESL1 group than in the EAE_ctrl group, whereas the other groups did not differ significantly from EAE_ctrl (Figure 3D). Moreover, the incidence of relapse (ie second EAE peak ≥1) was below 40% in the EAE_PLGA/ESL1 group, as compared to over 80% in animals from other tested groups. These results suggest that the PLGA/ESL1 implant, but not PLGA implantation alone or ESL1 antigens in a single dose, could protect the animals from developing relapsing EAE symptoms.

ESL1 Released from PLGA Migrates Through Draining Lymph Nodes into the CNS and Guts

Implanted materials are sampled by local phagocytic APCs and drained to regional lymph nodes.55 Moreover, we showed previously that ESL1 antigens critically modulate myeloid APC in vitro (in particular DCs), thereby inducing immune tolerance in EAE.14,16 Indeed, when the total splenocytes were cultivated in the presence of FITC-labelled ESL1 antigens for 24h in vitro, the highest uptake of ESL1-FITC (~75%) and the highest fluorescence of FITC-positive cells were found in myeloid APCs (CD11b+ I-A+ cells), followed by CD11b+I-A− and non-myeloid cells (Figure 4A). Therefore, to investigate the migration of ESL1 released from the PLGA fibres in vivo, animals were implanted with ESL1-FITC/PLGA implants or control non-labelled PLGA/ESL1, 7 days before EAE induction to collect the target tissues 4, 7, or 14 days after the implantation for histological and epifluorescent microscopy (Figure 4B, Supplementary Figure S1) and flow cytometry analyses (Figure 4C, Supplementary Figure S2).

Figure 4 Tracking of ESL1-FITC released from PLGA in vivo. (A) ESL1 was labelled with FITC and used in 24h- cultures with rat splenocytes at 50 µg/mL. The live (7AAD-) cells were analysed for CD11b-APC and I-A-PE expression and the Q2 region (CD11b+I-A+) showed the highest fluorescence of ESL1-FITC. Ctrl represents the signal from CD11b/I-A-stained cells not treated with ESL1-FITC. (B) Epi-fluorescent microscopy analyses are shown of the peri-implant region collected 4 days after the implantation of PLGA/ESL1-FITC (upper row); axillary lymph nodes collected 7 days (middle row) after the implantation; the spinal cords collected on 14 days after the implantation and 7 days after EAE induction (lower row). To avoid bleaching and increase ESL1-FITC detection, ESL1-FITC was detected with anti-FITC/Alexa Fluor (AF) 546 mAb, and the nuclei were stained with DAPI (see Supplementary Figure S1). White arrows point to nucleated cells positive for ESL1. (C) The percentage of ESL1-FTIC+ myeloid (CD11b+I-A+) cells from cervical, axillary, inguinal lymph nodes, spleen, Payer’s patches and spinal cord (CNS) were analysed 7 days after the implantation of ESL1-FITC/PLGA, or 14 days after the implantation in immunised (EAE) or non-immunized (healthy) animals, as described. Non-specific fluorescence was determined by using FMO-controls and PLGA/ESL1-treated animals, and the data is shown as mean ± SEM (n=4 animals per each group) (see Supplementary Figure S2 for a representative analysis) from one out of three independent experiments. n.d.-not determined/detected *p<0.05, as indicated (Kruskal–Wallis test with Dunn’s multiple comparison test).

H&E staining of peri-implant regions of skin and subcutaneous tissues suggested the presence of diffuse immune infiltrates around the implanted material, composed of inflammatory cells, including polymorphonuclear cells, and giant cells, in contrast to the equivalent subcutaneous regions of control skin samples (Supplementary Figure S1A). Epi-fluorescent microscopic analyses of the peri-implant regions 4 days after implantation suggested that ESL1-FITC was released into the surrounding area and could be found within the local infiltrating cells (Figure 4B). The cells in the axillary lymph nodes (AxLN), which drain the suprascapular region of the implant, stained positive in the medullary sinuses and cortex 7 days after the implantation of PLGA/ESL1-FITC, unlike control samples from animals treated with non-fluorescent PLGA/ESL1. Thereby, 0.7% - 5.5% of CD11b+ I-A+ cells were stained over the background in CLN, AxLN, IngLN, spleen, and the Payer’s patches when analysing these lymphoid organs 7 days after the implantation (Figure 4C, Supplementary Figure S2), but not when analysing them 3 days after the implantation. In contrast, when soluble ESL1-FITC was injected directly in the suprascapular region, 0.5%-6.3% of CD11b+I-A+ cells stained positive to ESL1-FITC 3 days after the injection, particularly within CLN and AxLN. In contrast, insignificant level of these cells could be detected 7 or 14 days after the injection (Supplementary Figure S3). The number of mononuclear cells from the spinal cords of non-immunised animals was too low for ESL1 tracking analysis.

Target organ analyses were performed 14 days after implantation in immunised and non-immunized (healthy) animals. In the latter, ESL1 antigen further accumulated in the CLN, AxLN, and spleen (Figure 4C). In contrast, immunised animals displayed significantly higher accumulation of ESL1 within the CLN and AxLN compared to healthy controls and displayed an additional accumulation of ESL1 in the Payer’s patches. The accumulation of ESL1 in the spleens of PLGA/ESL1-FITC-treated animals was not significantly higher compared to healthy animals. Additionally, up to 9.5% of myeloid APCs within the spinal cords of immunised animals were found positive for ESL1 antigen. These results suggested that ESL1 antigens released from PLGA migrate through implant-draining lymph nodes and the spleen in healthy animals in a delayed and continuous pattern as compared to soluble ESL1, and that EAE-induced inflammation potentiated additional migration of the released ESL1 towards the gut and CNS.

PLGA/ESL1 Reduces Splenic Inflammation During EAE

As neither PLGA nor the ESL1 antigens alone displayed any protective effects against EAE in this model (Figure 3), only animals from the EAE_PLGA/ESL1 and EAE_ctrl groups were used in further experiments. Considering the effects of PLGA/ESL1 implants on EAE and the critical role of the spleen in the systemic immune response during EAE,56,57 we next investigated how immune cell subsets are affected in the spleen during EAE. Key cell populations in the spleen were analysed by flow cytometry (Supplementary Figure S4) at the peak of EAE (day 14th) and during the recovery phase (day 20th). It was found that the treated EAE_PLGA/ESL1 animals contained significantly lower proportions of OX62+ I-Ahi DCs at the peak and recovery phases, as well as macrophages (CD68+I-A+) in the recovery phase, compared to EAE_ctrl animals (Figure 5A). Moreover, myeloid cell populations demonstrated lower expression of CD86 and increased expression of IL-10 upon isolation from the EAE_PLGA/ESL1 group compared to the control EAE_ctrl group (Figure 5B). The proportions of NK and T cells (both CD4 and CD8), and B cells did not differ significantly between the groups. A reduced proportion of NK cells and an increased proportion of B cells were observed during the recovery phase, relative to the peak of EAE (Figure 5A and C, Supplementary Figure S5).

Figure 5 Effects of PLGA/ESL1 on immune response in spleen during EAE. Seven days before immunisation with spinal cord homogenate in CFA, DA rats were implanted with PLGA/ESL1 (350µg ESL1 / 7 cm2 PLGA) subcutaneously in the suprascapular region (EAE_PLGA/ESL1) or just sham-operated (EAE_ctrl). Splenocytes collected from animals at the peak of EAE (day 14), or in the recovery phase (day 20) were collected and analysed by flow cytometry, as described in Supplementary Figure S3. (A) The proportion of DCs (OX62+I-Ahi), macrophages (CD68+I-A+), T (CD3+CD161−) and NK cells (CD3−CD161+) cells is shown. (B) The expression of CD86 within DCs and macrophages, and IL-10 expression within total myeloid (CD11b+) cells are shown as mean fluorescence intensity (MFI). (C) The relative proportion of cell types is shown after the normalisation of the gated cells to 100%. See Supplementary Figure S4A for data on the proportion of B cells, CD4+ and CD8+ T cells. (D) The proportion of Th1 (IFN-γ+), Th17 (IL-17+), Th2 (IL-4+) cells within CD4+ T cells subsets, and IFN-γ producing NK cells are shown, as well as CD25hiFoxP3+ Tregs and IL-10-producing T cells. All T cell subsets in the spleen were gated as demonstrated in Figure 7A and 7E) The level of cytokines spontaneously produced by the splenocytes (2 x 106/ mL) in 48h-cultures upon the isolation from EAE_PLGA/ESL1 or EAE_ctrl animals at the EAE peak or recovery from one, out of two experiments is shown. (A, B, D, E) *p<0.05, **p<0.001, ***p<0.005, as indicated (Kruskal–Wallis test with Dunn’s multiple comparison test).

The functional properties of NK and T cells were significantly modulated by PLGA/ESL1 treatment irrespective of their unaltered proportions in the spleen. Namely, PLGA/ESL1-treated animals displayed a reduced proportion of IFN-γ-producing CD4+ (Th1) and NK cells, and IL-17-producing CD4+ T cells at the peak of EAE, as well as lower expression of CD86 on B cells, compared to EAE_ctrl animals (Figure 5D). These results correlated with the reduced levels of IFN-γ and IL-17 in 48 h cultures of ex vivo splenocytes isolated from PLGA/ESL1 animals at the peak of the disease (Figure 5E). In addition to the reduced proportion of pro-inflammatory lymphocytes in the spleen of these animals, an increased proportion of IL-10-producing CD8+ T cells was observed at the peak of EAE (Figure 5D). The recovery phase of EAE was generally characterised by a reduced proportion of Th1 and Th17 cells and an increased proportion of CD4+CD25hiFoxP3+ Tregs in the spleens of both groups. However, PLGA/ESL1-treated animals displayed a significantly increased proportion of Tregs during the recovery phase compared to the control EAE_ctrl group (Figure 5D). The recovery phase also displayed an increased proportion of IL-4-producing CD8 cells (ie Tc2), as well as IL-10-producing NK and B cells, although there were no differences between the groups (Supplementary Figure S5B). These results, along with the reduced levels of IL-6 in ex vivo splenocyte cultures (Figure 5D), indicated that PLGA/ESL1 implants reduced splenic inflammation in EAE and potentiated regulatory lymphocytes.

PLGA/ESL1 Implants Reduce Immune Cell Infiltration into the CNS

EAE symptoms in DA rats are critically linked to immune cell infiltration into the CNS, particularly in the spinal cords (SCs), which are the most commonly used pathological markers in this experimental model.58 Therefore, the SCs of animals from the EAE_ctrl and EAE_PLGA/ESL1 groups were collected at the peak and recovery phase of EAE and further analysed by microscopy and flow cytometry (Figures 6 and 7). Histological analysis of the H&E-stained lumbar SC cross-sections indicated a significantly higher number of infiltrates in the EAE_ctrl group than in the EAE_ PLGA/ESL1 group at the peak of EAE, although the average number of cells per infiltrate was similar between the groups (Figure 6A and B). Infiltrates were predominantly found in the white matter of SCs on the ventral, dorsal, and lateral horns, and their number was reduced during the recovery phase (Figure 6B). Epifluorescence microscopy analysis of CD45 / myelin-based protein (MBP) staining confirmed that the infiltrates in SCs were of leukocyte origin (Figure 6C). Moreover, the infiltrating areas showed reduced MBP expression at the infiltration sites, indicating tissue damage caused by the infiltrates (Figure 6C).