Animals

Female, 8- to 12-week-old C57BL/6J mice were purchased from Envigo (Indianapolis, IN, USA). Mice were kept in individually ventilated cages under specific pathogen-free conditions and fed ad libitum. All animal studies were approved by institutional care committee and state committees for animal welfare (84-02.04.2015.A585). Animal experiments were conducted in accordance with the European Union normative for care and use of experimental animals and the German Animal Protection Law.

Combined active and focal experimental autoimmune encephalomyelitis model

Induction of EAE was performed in 8- to 12-week-old female C57BL/6J mice as previously described [15]. Briefly, 10 days prior to focal EAE induction, mice were subcutaneously immunized with 200 µg of murine MOG35–55 peptide (Charité, Berlin, Germany) dissolved in phosphate-buffered saline (PBS, 2 mg/ml) and homogenized with complete Freund’s Adjuvant (CFA, 2 mg/ml; Merck, Darmstadt, Germany) in a 1:1 ratio. 100 μl of the resulting MOG35–55 emulsion was injected in each flank of anesthetized mice (isoflurane). Injection of pertussis toxin (200 ng in 100 µl PBS; Enzo Life Sciences, Farmingdale, NY, USA) was performed on day 0 and day 2 after MOG35–55 immunization, intraperitoneally.

Ten days after MOG35–55 immunization, mice were deeply anesthetized and mounted on a stereotactic device. Using the following coordinates (anteroposterior, − 2.18 mm; lateral, 4.2 mm from bregma; and dorsoventral, 1 mm from the brain surface for the auditory cortex), a hole was drilled through the skull. Two μl of a solution containing the proinflammatory cytokines TNF-α (150 U; Merck) and IFN-γ (800 U; Merck) dissolved in PBS were slowly injected into the left auditory cortex. The contralateral hemisphere (right side) served as control. Mice were killed at the day of maximal clinical deterioration (dmax, Additional file 1: Fig. S1) for experimental setting one (Figs. 1, 2, 3) or 17 days post-injection (day 27) to determine long-term effects of cladribine treatment in a chronic EAE state (Fig. 4). In all experimental steps, mice were randomly assigned to the operators by an independent person not involved in data analysis. Surgery and evaluation of all read-out parameters were performed in a blinded manner.

Fig. 1

In vivo effects of cladribine on EAE score. A Oral treatment with cladribine (for 5 consecutive days—day 5 to 9, as indicated in the figure) induced a less severe disease course in EAE mice compared to EAE mice receiving vehicle. Statistical differences were observed in terms of development of neurological signs over the whole observation period until day 27 (two-way ANOVA, F (1, 22) = 9.446, p = 0.0056, n = 12 for each group). B The bar graph depicts the cumulative EAE score of both experimental groups on the day of maximal clinical deterioration (dmax, unpaired Mann–Whitney U test, U = 34.50, p = 0.0295, n = 12 for each group). C Immunophenotypings of the peripheral blood (a), LNs (b), spleen (c), thymus (d) and bone marrow (e) at dmax were performed by flow cytometry. Immune cell profiles of mice treated with cladribine were compared to those receiving vehicle. Statistically significant differences were obtained by performing two-way ANOVAs complemented by Bonferroni test for multiple comparisons, n = 3 for each group. p-values: 0.0019 (a, blood), 0.0003 (b, LN), 0.6227 (c, spleen), 0.0016 (d, thymus) and 0.3734 (e, bone marrow). D Flow cytometric analyses were used, to evaluate the immune cell distribution in the brain (a) and spinal cord (b) of EAE mice, either treated with cladribine or vehicle at dmax. Statistically significant differences were obtained by performing two-way ANOVAs complemented by Bonferroni test for multiple comparisons, n = 3 for each group. p-value was 0.2486 for brain, < 0.0001 for spinal cord. p > 0.05 = ns, p < 0.05 = *, p < 0.01 = **, p < 0.001 = ***, p < 0.0001 = ****

Fig. 2

In vivo cladribine treatment does not affect the number of inflammatory foci in the brain of EAE mice. Hematoxylin and eosin (H&E) staining of coronal sections comparing inflammatory CNS lesions in vehicle- and cladribine-treated EAE mice. Images were acquired with a Zeiss Axio Scope.A1 (Zeiss, Göttingen, Germany) using tenfold objectives. Exemplary H&E stainings of the ipsilateral hemisphere (A) and the contralateral hemisphere (without stereotactic cytokine injection, B) are shown. Scale bar 100 µm. Cc corpus callosum, Cx cortex, Hip hippocampus, Str striatum. C Representative lesions of the ipsilateral (a) and the contralateral (b) hemisphere are magnified. Scale bar 50 μm. D Histograms showing the quantification of inflammatory lesions in both hemispheres in coronal sections of vehicle- and cladribine-treated EAE mice. Lesions are classified topologically (grey (GM) and white (WM) matter). No statistic difference was obtained following unpaired Mann–Whitney U test (p > 0.05). n = 3, 35 slices per animal were analyzed and the mean per slice and mouse was used for statistical analysis

Fig. 3

The capacity of cladribine to affect neurons of the primary auditory cortex. A mRNA expression of deoxycytidine kinase (DCK) was quantified by quantitative polymerase chain reaction (qPCR) in magnetic-associated cell-sorted (MACS) neurons, compared to macrophages (macroph.; CD45highCD11bhigh), CD3+CD4+ and CD19+ lymphocytes. Platelets served as negative control. Results were normalized to expression levels in macrophages and are depicted as 2−∆∆Ct values (n = 9 for macrophages, n = 5 for all other cell types, Kruskal–Wallis test). B–F All electrophysiological recordings were obtained at dmax by current-clamp mode (B–D) or voltage-clamp mode (E + F). Four mice were examined per experimental group. B Representative traces recorded at a potential of − 60 mV (set by DC current injection) in current-clamp mode from one animal. A depolarizing current step of + 160 pA triggered the generation of action potentials (APs) in all experimental groups. C Mean bar graph indicating the number of APs generated in response to depolarizing steps of increased intensity ranging from + 20 to + 160 pA. The number of generated APs increased with increasing depolarization under all experimental conditions. Statistically significant differences were obtained starting from + 100 pA (two-way ANOVA, F (2.59) = 5.061, p = 0.0094, n = 23—naïve ctrl, 14—vehicle, 25—cladribine). D Resting membrane potential (in mV; a) and input resistance (in MΩ; b) of neurons of the auditory cortex are depicted indicating no differences between vehicle- vs. cladribine-treated EAE mice compared to naïve controls (unpaired Mann–Whitney U test, p > 0.05, n = 23—naïve ctrl, 14—vehicle, 24—cladribine). E Effects of oral treatment with cladribine on glutamatergic transmission in voltage-clamp mode recorded as excitatory postsynaptic currents (EPSCs). Upon EAE induction, the number of EPSCs (a) increased significantly compared to naïve control (unpaired Mann–Whitney U test, p = 0.0401, n = 28—naïve ctrl, 16—vehicle). Upon cladribine treatment, a trend for a reduction in EPSCs compared to vehicle-treated mice was observed, although not reaching significance (unpaired t test, p = 0.0751, n = 16—vehicle, 19—cladribine). Exemplary electrophysiological traces of voltage-clamp recordings (b) show exemplary EPSC events in recorded neurons from naïve control, vehicle-treated and cladribine-treated EAE mice. F Bar graph illustrating the number of inhibitory postsynaptic currents (IPSCs; a indicating that neither the experimental EAE itself nor oral cladribine treatment did significantly affect the GABAergic transmission (unpaired Mann–Whitney U test, p > 0.05, n = 19—naïve ctrl, 11—vehicle, 16—cladribine). Representative electrophysiological traces of voltage-clamp recordings of IPSCs (b) from all three experimental groups. p > 0.05 = ns, p < 0.05 = *, p < 0.01 = **

Fig. 4

Immunosuppressive and neuroprotective effects of cladribine are transient on cellular level. A Oral treatment with cladribine induced a less severe disease score on day 27 in EAE mice compared to EAE mice receiving vehicle treatment (unpaired Mann–Whitney U test, p = 0.0386, cladribine-treated (n = 12) vs. vehicle-treated (n = 13)). B Immunophenotypings of the peripheral blood (a), LNs (b), spleen (c), thymus (d) and bone marrow (e) at day 27 post-EAE induction were performed by flow cytometry. Immune cell profiles of mice treated with cladribine were compared to those receiving vehicle. No statistically significant differences could be observed (two-way ANOVAs, n = 3 for each group. p-values > 0.05). C: Flow cytometric analysis of the immune cell distribution in the brain (a) and spinal cord (b) of EAE mice on day 27 post-immunization, either treated with cladribine or vehicle, shows no significant differences between both experimental groups (two-way ANOVAs, n = 3 for each group. p-values > 0.05). D mRNA expression of deoxycytidine kinase (DCK) was quantified by quantitative polymerase chain reaction (qPCR) in murine PBMCs and normalized to human PBMCs. Results are depicted as 2−∆∆Ct values (n = 6 for both species, unpaired Mann–Whitney U test, p = 0.0022). E Electrophysiological recordings of APs were obtained by recording in current-clamp mode. Mean bar graph indicating the number of APs recorded on day 27 post-EAE induction in response to depolarizing steps of increased intensity ranging from + 20 to + 160 pA. No significant alterations could be observed between groups (two-way ANOVA, n = 23—naïve ctrl, 10—vehicle, 9—cladribine. p-values > 0.05). p > 0.05 = ns, p < 0.05 = *, p < 0.01 = **

Oral cladribine treatment

Five days post-immunization, cladribine (10 mg/kg; Merck) was administered daily via oral gavage in 0.5% aqueous carboxymethylcellulose for 5 days (further referred to as ‘cladribine-treated EAE group’). The vehicle group received 0.5% aqueous carboxymethylcellulose for 5 days at the same period (further referred to as ‘vehicle-treated EAE group’). Moreover, an additional control group of naïve mice (without EAE or any oral treatment), further referred to as ‘controls’, was used for ex vivo electrophysiological analysis. Prior to the EAE experiments, titration experiments of cladribine were conducted, indicating 10 mg/kg as the appropriate dose to reach an adequate brain concentration of approximately 80–100 ng/g over a period of 2.5 h (Additional file 2: Fig. S2). This time interval was chosen in advance, as cladribine is rapidly available following oral administration and maximum plasma levels are found within 1 to 2 h of ingestion [8].

Functional outcome tests

Health status and disease progression (weight, disease score, general appearance and performance) of mice was monitored on a daily basis by two blinded investigators. Disease severity was defined using the following scoring system: grade 0, no abnormality; grade 1, limp tail tip; grade 2, limp tail; grade 3, moderate hindlimb weakness; grade 4, complete hindlimb weakness; grade 5, mild paraparesis; grade 6, paraparesis; grade 7, heavy paraparesis or paraplegia; grade 8, tetraparesis; grade 9, quadriplegia or premoribund state; or grade 10, death. Animals presenting with a score ≥ 7 for more than three consecutive days or showing a score of 8 (independent of the time period) were killed and the last score observed was included in the analysis until the end of the experiment, respectively. Mice showing a bodyweight reduction of more than 20% compared to their starting bodyweight (bw) were excluded from experiments as well. The cumulative disease score was calculated as the sum of the daily clinical scores of each mouse during the EAE observation period and reported as an average within each group (mean ± SEM). dmax refers to the day of maximum clinical deterioration representing the day on which the majority of mice in the experiment had the highest clinical score (i.e., day with highest cumulative disease score). We determined dmax based on our long-term experience with the EAE model [15, 18, 20,21,22,23,24]. Using our standardized EAE protocol, we observed a typical chronic course over 25–30 days (EAE duration depending on experimental setup) with onset of clinical symptoms about day 10 and disease maximum (dmax) about day 15–18 post-immunization followed by a remission of symptoms [18, 20]. In the current study, dmax was reached on day 17 and maximum EAE duration was set to 27 days.

Tissue preparation

At dmax and day 27 post-EAE induction, blood was taken by heart puncture under deep isoflurane anesthesia. Afterwards, mice were perfused via the left ventricle with PBS. Directly after cardiac perfusion, brains, spinal cords, lymph nodes (LN), spleens, thymi and femur bones were removed and single-cell suspensions were prepared.

Blood was treated with erythrocyte lysis buffer (150 mM NH4Cl, 10 mM KHCO3, 0.1 mM EDTA; pH 7.3). Organs (mouse spleen, LN, thymus and bone marrow) were homogenized by a 40-μm cell strainer (BD Biosciences, Germany). Homogenates were rinsed with washing medium (Dulbecco’s modified Eagle’s medium, DMEM; Invitrogen, Waltham, MA, USA) containing 1% fetal bovine serum (FBS, ScienCell, Carlsbad, CA, USA) and 1% antibiotics (penicillin/streptomycin, Sigma-Aldrich, St. Louis, MO, USA). Erythrocytes in the splenocyte suspension were lysed with erythrocyte lysis buffer for 30 s. Then, lysis was stopped by addition of washing medium. Single cell suspensions were washed once again and resuspended in the desired buffer for subsequent applications.

For purification of CNS-infiltrating leukocytes, brain and spinal cord tissues were cut into pieces, and mechanically homogenized in HBSS by an insulin syringe. After a 30-min enzymatic digestion with a collagenases, DNase and trypsin-inhibitor mix, suspension was layered on a density gradient using Lymphoprep™ (Fresenius, Germany), and separated by centrifugation (18 min at 790g without acceleration and break). After isolation, the pellet was washed and resuspended in the respective staining buffer. To quantify numbers of cells isolated from the CNS, beads (Beckman Coulter, Brea, CA, USA) were added.

Flow cytometry

Phenotyping of different immune cell subsets was performed using flow cytometry. Therefore, immune cells isolated from blood, spleen, thymus, LN, bone marrow and CNS (brain and spinal cord) were characterized by staining for CD3, CD4, CD8, CD11a/b, CD11c, CD25, CD44, CD45, CD45R/B220, CD49d, CD62L, CD69 and NK1.1 (for details see Table 1 and Additional file 1: Fig. S3). Briefly, single-cell suspensions were stained for 30 min at 4 °C with the appropriate combination of indicated fluorescence-labeled monoclonal antibodies in PBS, containing 2 mM EDTA and 0.1% bovine serum albumin (Sigma-Aldrich). Corresponding isotype controls were used for all stainings. For blocking of Fc receptors, cells were preincubated with purified anti-CD16/CD32 antibody for 10 min on ice prior to immunostaining. Fixable viability dye eFluor (Thermo scientific, Waltham, MA, USA) was used for live/dead staining. Concentrations of antibodies were carefully titrated prior to experiments. Flow cytometric analysis of stained cells was performed following standard protocols. Cells were analyzed on a Gallios Flow Cytometer (Beckman Coulter) and a CytoFLEX S (Beckman Coulter) using Kaluza Analysis Software (Beckman Coulter).

Table 1 Antibodies used for flow cytometryIsolation of murine peripheral blood mononuclear cells (PBMCs)

Female mice (6 replicates, 3 mice per replicate) were anesthetized with isoflurane. Blood was withdrawn directly from the heart using a 1-ml syringe filled with 50 µl of EDTA (2 mM). Blood from 3 mice was pooled in one 2-ml tube and transferred to a 50-ml conical tube. The remaining blood in the 2-ml tubes was washed out with 2 mL D-PBS + 2% FCS. 5 ml of D-PBS + 2% FCS were added to the blood/D-PBS mixture to obtain a final volume of 13 ml. Next, the blood was layered in SepMate™-50 IVD tubes (STEMCELL Technologies, France) pre-filled with Cytiva Ficoll-Paque™ PLUS (density 1.077 ± 0.001 g/ml; Thermo scientific). Samples were centrifuged for 20 min at 12,000 rpm at room temperature. The upper phase with PMBCs was transferred to a new 50-ml conical tube. Samples were filled with wash medium (DMEM supplemented with 1% FCS, 1% l-glutamine and 1% penicillin/streptomycin) to a total volume of 50 ml. Subsequently, centrifugation for 10 min at 300g and 4 °C was performed. The supernatant was discarded, and the pellet was resuspended in DMEM for cell counting using a Neubauer chamber. We acquired approximately 5 × 106 PBMCs per replicate.

Isolation of human PBMCs

Human PBMCs were isolated from 6 healthy donors. Blood samples of control subjects without an autoimmune or neuroinflammatory disorders (n = 6) were included in this study. All cases presented with non-specific complaints and underwent blood sampling during a routine diagnostic examination conducted to rule out any neurological condition. None of the healthy controls suffered from a neurological disorder, nor did they show any specific abnormalities during the neurological examination. All patients included in this study gave their written informed consent in accordance with the Declaration of Helsinki and a protocol approved by the Ethics Committee of the University of Duesseldorf (5951R). 

On the day of experiment, blood was drawn from healthy donors, collected in 10-ml EDTA tubes (BD Diagnostics Systems, Franklin Lakes, NJ, USA), and diluted with PBS in a 1:1 ratio. Isolation was performed using Cytiva Ficoll-Paque™ PLUS (density 1.077 ± 0.001 g/ml; Thermo scientific) within a SepMate™-50 IVD tube (STEMCELL Technologies) according to the manufacturer’s instructions. Cells were counted and cryopreserved at approximately 1 × 107 cells per 1 ml.

Fluorescence-activated cell sorting (FACS) of macrophages

First, CD11b+ cells were isolated from murine adult brain and spinal cord as previously described [25]. Nine biological replicates were collected. For each biological replicate, 3 naïve C57BL/6J mice (female, 15–25 weeks old, 18–35 g bw) were pooled. Second, CD11b+ cells were stained with FITC anti-mouse/human CD11b antibody (clone M1/70; BioLegend, San Diego. CA, USA) 1:50 and APC/Cyanine7 anti-mouse CD45 antibody (clone 30-F11; BioLegend) 1:200 in PBS for 15 min at room temperature. Concentrations of antibodies were carefully titrated prior to experiments. FACS staining was performed following standard protocols. Afterwards, macrophages were analyzed and sorted by use of a MoFlo XDP, Cell Sorter (Beckman Coulter; 100 m\(\upmu\) nozzle, pressure: 26 psi) using Summit Analysis Software version 5.4.0 (Beckman Coulter). Macrophages were identified as CD45highCD11bhigh, while microglia were suspected to be CD45intCD11bhigh [25] (for gating strategy see Additional file 4: Fig. S4).

RNA isolation and real-time quantitative PCR (qPCR)

CD3+CD4+ and CD19+ lymphocytes were isolated via magnetic-activated cell sorting (MACS) from cervical LN of adult naïve mice according to the manufacturer’s instructions (Miltenyi Biotec, Germany). Adult neurons were isolated as previously described [25]. Subsequently, the expression of DCK was quantified by qPCR. Platelets were isolated as previously described and served as negative control, as they have no nucleus and showed no DCK expression [26].

RNA was isolated with the Quick-RNA Microprep Kit (Zymo Research) following the manufacturer’s protocol. Tissue homogenates and cells were lysed in 300 μl RNA lysis buffer, followed by sample clearing. The supernatant was mixed with 95–100% ethanol and transferred to the column. In-column DNAse treatment was performed. After washing and drying the column, RNA was eluted by pre-warmed DNase/RNase-free water (15 μl). RNA quality was measured with Nanodrop by A260/A280 and A260/A230 ratios.

Reverse transcription was performed with Maxima Reverse Transcriptase (Thermo scientific) and random hexamer primers. One-hundred ng cDNA was used for real-time qPCR with TaqMan Master Mix (Maxima probe/ROX, Applied Biosystem). To this end, 1 μM of each primer (target primers: murine DCK, Mm00432794_m1; human DCK, #4331182; Thermo scientific) or 1 μM house-keeping primer for the respective control (18 s, #4333760 T), 10 μl of maxima probe/carboxyrhodamine (ROX) fluorescent dye, 4 µl of DNA-free aqua and 100 ng cDNA (4 µl) were mixed. Run was performed on a StepOnePlus™ Real-Time PCR System (Applied Biosystems) according to the following steps: hold—2 min 50 °C, initial denaturation—10 min 95 °C, amplification—(40x) 10 s 95 °C—45 s 58 °C—1 min 72 °C. Data were analyzed with the StepOne software (Applied Biosystems, v2.1) calculating 2−∆∆Ct values (the ratio of DCK expression to the house-keeping gene, normalized to its expression in macrophages or human PBMCs, respectively).

Histology—hematoxylin–eosin (H&E) staining

In order to verify the injection site, and to quantify the amount of white and grey matter lesions in the cladribine-treated EAE compared to the vehicle-treated EAE group, brains were used for histopathological evaluation. Mice were perfused through the left ventricle with PBS for 5 min under deep isoflurane anesthesia. Brains were removed, and immediately frozen in embedding medium (Tissue-Tek® O.C.T.™ compound, Sakura Finetek, Germany). Cryopreserved brain slices (10 µm thick) were stained with hematoxylin and eosin (H&E) using standard protocols. Light microscopy and AxioVision software were used to determine the number of inflammatory lesions per animal. Lesions were classified by anatomy (grey and white matter) and location (ipsilateral (cytokine injection side) and contralateral hemisphere) for both groups. Thirty-five slices per mice were analyzed, three mice per experimental group. Means ± SEM (standard error of the mean) per slice and mouse were calculated and used for further statistical analyses.

Electrophysiological experiments

In order to investigate neuronal excitability, we analyzed the firing pattern of auditory pyramidal cortical neurons (layer 4) in current-clamp mode. In detail, at dmax and day 27 (for long-term studies, Additional file 1: Fig. S1), the animals were deeply anesthetized, and the brains quickly removed. Brains were glued onto a cutting plate with the help of an agar block to cut (vibratome from Leica) coronal slices (250–300 µm of thickness) containing preserved network structure of the auditory cortical network [27]. Cutting was performed in ice-cold artificial cerebrospinal fluid (ACSF) solution (200 mM sucrose, 10 mM glucose, 20 mM PIPES, 2.5 mM KCl, 10 mM MgSO4, 0.5 mM CaCl2; pH 7.35).

Electrical recordings were performed at room temperature in carbonated ACSF as extracellular solution containing the following (in mM): NaCl, 125; KCl, 2.5; NaH2PO4, 1.25; HEPES, 30; MgSO4, 2; CaCl2, 2; glucose, 10; pH 7.35 with NaOH; 305 mOsm/kg in a submerged chamber on an upright microscope (Zeiss, Germany). Intracellular recordings were performed in visually identified neurons of the layer 4 of the primary auditory cortex, using a ZEN 2.5 camera (Zeiss) and were governed by using the PatchMaster software (HEKA, Germany).

Recording pipettes were advanced towards individual neurons in the slice under positive pressure and visual control. The membrane patch was then ruptured by suction and the membrane potential was monitored using a double patch amplifier (HEKA EPC 10). Whole-cell patch clamp recordings were made with borosilicate glass pipettes (GC150TF-10, Harvard Bioscience, Holliston, MA, USA) for both voltage- and current-clamp mode.

Neurons were challenged with the administration of a series of depolarizing current steps of increasing intensity leading to the generation of action potentials (APs). A total of 8 current steps of increasing intensity (from + 20 to + 160 pA, 2.5 s duration) were applied and the number of APs was taken as read-out. APs were counted semi-automatically using the software PEAK (Meuth IT Consulting) and confirmed by visual inspection of each recorded step. To investigate changes in cellular excitability threshold and the number of fired APs for the same baseline, recordings were performed at a holding potential of − 60 mV that was set by DC current injection.

To study spontaneous postsynaptic currents, including EPSCs or inhibitory postsynaptic currents (IPSCs), the recording pipettes were filled with an internal solution of the following composition (in mM): NaCl, 10; K-gluconate, 88; K3-citrate, 20; HEPES, 10; BAPTA, 3; phosphocreatine, 15; MgCl2, 1; CaCl2, 0.5; Mg-ATP, 3; Na-GTP, 0.5; set to pH 7.25 with KOH and osmolality of 295 mOsmol/kg. To record both, spontaneous glutamatergic- and GABAergic currents in the same experiment, cells were clamped to a holding potential of − 60 mV (glutamatergic currents), and subsequently to 0 mV to record GABAergic currents.

Spontaneous synaptic events were acquired using PatchMaster (HEKA) and analyzed using MiniAnalysis software 6.0. (Synaptosoft Inc., Fort Lee, NJ, USA). The detection threshold of EPSCs was set at twice the baseline noise and once the baseline for IPSCs. With this recording setting, all technically detected positive spikes represented GABAergic currents (IPSCs) and negative spikes constituted glutamatergic currents (EPSCs). Counts were confirmed for each cell and experiment by visual inspection of each recorded trace.

Cell isolation from spleen and LN

Spleens or LNs were homogenized by a 40-µm cell strainer and washed with 10 ml washing medium (DMEM, 1% FCS, 1% penicillin/streptomycin). Erythrocytes in the cell suspension were lysed with ACK buffer (150 mM NH4Cl, 10 mM KHCO3, 0.1 mM EDTA, pH 7.3) for 30 s, stopped by addition of washing medium. Single cell suspensions were washed once again and resuspended in splenocyte complete medium (DMEM, 10 mM HEPES, 25 μg/ml gentamicin, 50 μM mercaptoethanol, 5% fetal calf serum, 2 mM glutamine, and 1% non-essential amino acids (Cambrex, Verviers, Belgium)).

Leukocyte organ culture

Murine cervical LNs and small pieces of spleen were placed on 24-well plate Costar Transwell inserts with a 3.0 µm pore polyester membrane (Corning, Lowell, MA, USA). 600 µl organ culture medium (RPMI, 10% FCS, 1% penicillin/streptomycin) were placed in the lower chamber and 100 µl in upper chamber. The organs were treated either with 0.1 µM cladribine or with the responding vehicle in the control group. After 7 days, all migrated cells were collected and analyzed by flow cytometry. The organs were homogenized according to the protocol described for splenocyte isolation. Cells were analyzed by flow cytometry using the following markers: CD4, CD8, CD45R/B220, CD11a, CD25, CD44, CD49d, CD62L, CD69 (for details see Table 1 and “Flow cytometry” section).

Proliferation assay

Lymphocytes were isolated according to the protocol described above and labeled with VybrantTM CFDA SE Cell Tracer (12.5 µM) in 2 ml PBS + 2% FCS for 10 min at 37 °C, followed by addition of 10 ml cold washing buffer and incubation on ice for 10 min. After washing the cells, they were seeded into 96-well plates (U-bottom) coated with 1 µg/ml anti-CD3. Soluble anti-CD28 (2 µg/ml) was added to the splenocyte complete medium as indicated in the respective experiments. After plating the cells, the different groups were treated with either 0.1 µM cladribine, 1.0 µM cladribine or the vehicle, respectively. Cells were cultured for 3 days (37 °C, 5% CO2) prior to flow cytometry analysis. Cells were analyzed by flow cytometry using the following markers: CD4, CD8, CD45R/B220, CD11a, CD25, CD44, CD49d, CD62L, CD69 (for details see Table 1 and “Flow cytometry” section).

Statistical analysis

Results are displayed as means ± SEM unless indicated otherwise. For column-based data, Gaussian distribution was evaluated by D’Agostino–Pearson normality test. Dependent on normality for analysis of two groups, two-tailed t test (unpaired/paired) or Mann–Whitney U test was used as appropriate. If more groups were compared, we applied one-way ANOVA, complemented by Bonferroni test for multiple comparisons for parametric data, or the Kruskal–Wallis test including Dunn’s post-test for non-parametric data. Comparison of EAE data was performed using two-way ANOVA.

The level of significance was labeled according to the p-values: p values > 0.05 were classified as not significant, p < 0.05 (*) as significant, p < 0.01 (**), p < 0.001 (***) and p < 0.0001 (****) as highly significant. Analyses and graphs were prepared using Prism 9.1.2 (Graph Pad, San Diego, CA, USA).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.