Myeloid-derived suppressor cells expressing a self-antigen ameliorate experimental autoimmune encephalomyelitis

Myeloid-derived suppressor cells expressing a self-antigen ameliorate experimental autoimmune encephalomyelitis

    Myeloid-derived suppressor cells expressing a self-antigen ameliorate experimental autoimmune encephalomyelitis Silvia Casacuberta-Se...

1MB Sizes 0 Downloads 0 Views

    Myeloid-derived suppressor cells expressing a self-antigen ameliorate experimental autoimmune encephalomyelitis Silvia Casacuberta-Serra, Carme Costa, Herena Eixarch, Maria Jose Mansilla, Sergio Lopez-Est´evez, Llu´ıs Martorell Cedr´es, Marta Par´es, Xavier Montalban, Carmen Espejo, Jordi Barquinero PII: DOI: Reference:

S0014-4886(16)30291-6 doi:10.1016/j.expneurol.2016.09.012 YEXNR 12406

To appear in:

Experimental Neurology

Received date: Revised date: Accepted date:

19 May 2016 5 September 2016 20 September 2016

Please cite this article as: Casacuberta-Serra, Silvia, Costa, Carme, Eixarch, Herena, Mansilla, Maria Jose, Lopez-Est´evez, Sergio, Cedr´es, Llu´ıs Martorell, Par´es, Marta, Montalban, Xavier, Espejo, Carmen, Barquinero, Jordi, Myeloid-derived suppressor cells expressing a self-antigen ameliorate experimental autoimmune encephalomyelitis, Experimental Neurology (2016), doi:10.1016/j.expneurol.2016.09.012

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Title Myeloid-derived suppressor cells expressing a self-antigen ameliorate experimental

RI

PT

autoimmune encephalomyelitis

Authors

NU

Carme Costab c ([email protected])

SC

Silvia Casacuberta-Serraa b ([email protected])

Herena Eixarchb c ([email protected])

MA

Maria Jose Mansillaa b ([email protected])

Sergio Lopez-Estéveza b ([email protected])

D

Lluís Martorell Cedrésa b ([email protected])

TE

Marta Parésa b ([email protected])

AC CE P

Xavier Montalbanb c ([email protected]) Carmen Espejob c d ([email protected]) Jordi Barquineroa b d ([email protected])

Authors' affiliations a

Gene and Cell Therapy Laboratory. Vall d’Hebron Institut de Recerca (VHIR). Passeig

Vall d'Hebron 119-129, Barcelona 08035, Spain. b

Universitat Autònoma de Barcelona, 08193 Bellaterra (Cerdanyola del Vallès), Spain.

c

Servei de Neurologia-Neuroimmunologia Clínica, Centre d’Esclerosi Múltiple de

Catalunya (Cemcat). VHIR, Hospital Universitari Vall d’Hebron. Passeig Vall d'Hebron 119-129, Barcelona 08035, Spain. d

These authors contributed equally.

1

ACCEPTED MANUSCRIPT

Corresponding authors:

PT

Jordi Barquinero, M.D., Ph.D.

RI

Laboratory of Gene and Cell Therapy. Vall d’Hebron Institut de Recerca (VHIR), Passeig

Telephone: 34 932746726. Fax: 34 932746727

NU

e-mail: [email protected]

SC

Vall d’Hebron 119-129. Barcelona 08035, Spain.

MA

Carmen Espejo, Ph.D.

Servei de Neurologia-Neuroimmunologia, Centre d’Esclerosi Múltiple de Catalunya

D

(Cemcat), Vall d’Hebron Institut de Recerca, Hospital Universitari Vall d’Hebron, Passeig

Telephone: 34 934893599.

TE

Vall d'Hebron 119-129, Barcelona 08035, Spain.

AC CE P

e-mail: [email protected]

2

ACCEPTED MANUSCRIPT

ABSTRACT

PT

Previous work by our group showed that transferring bone marrow cells transduced with a self-antigen induced immune tolerance and ameliorated experimental autoimmune

RI

encephalomyelitis (EAE), a model of multiple sclerosis (MS). We also found that following

SC

retroviral transduction of murine bone marrow (BM) cells, the majority of cells generated and transduced were myeloid-derived suppressor cells (MDSCs). Here, we aimed to

NU

determine whether purified antigen-expressing MDSCs have similar therapeutic effects

MA

than those of unfractionated BM, and to investigate their potential mechanisms. We performed phenotypic and functional analyses in these cells using the same animal model, and we used purified antigen-expressing MDSCs in preventive and therapeutic

TE

D

approaches. These cells exerted therapeutic effects similar to those of BM cells, which depended upon self-antigen expression. The majority of monocytic (M)-MDSCs expressed

AC CE P

the immunosuppressive molecule Programmed Death Ligand-1 (PD-L1), CD80, CD86 and MHC class II molecules. Additionally, the animals infused with antigen-expressing cells exhibited lower percentages of activated T cells and higher percentages of B cells with a regulatory phenotype (B220+CD1dhighCD5+) in the spleen than their respective controls. MDSCs expressing self-antigens, alloantigens or therapeutic transgenes are tolerogenic and can be exploited therapeutically in autoimmune diseases, transplantation and in gene therapy, respectively.

HIGHLIGHTS - Ex vivo-generated MDSCs expressing a self-antigen (MOG40-55) have preventive and therapeutic effects in a murine model of multiple sclerosis, which are strictly dependent on antigen expression by the MDSCs.

3

ACCEPTED MANUSCRIPT

- A vast majority of the ex vivo-generated M-MDSCs expressed PD-L1, CD80, CD86 and MHC class II molecules.

PT

- In the preventive experiments, infusion of MDSCs expressing the self-antigen was

RI

associated with a reduced percentage of activated T cells and an increased percentage of B cells with a regulatory phenotype in the spleen.

KEYWORDS:

Myeloid-derived

NU

SC

- MDSCs expressing MOG40-55 induce apoptosis in total CD3+ and CD4+ T cells.

suppressor

cells,

Multiple

sclerosis,

Experimental

MA

autoimmune encephalomyelitis, Immune tolerance, Cell therapy, Regulatory B cells, PD-1,

D

PD-L1, MHC class II.

TE

ABBREVIATIONS: 7-AAD, 7-aminoactinomycin D; DMEM, Dulbecco’s modified Eagle’s medium; EGFP, enhanced green fluorescent protein; EAE, experimental autoimmune

estimating

AC CE P

encephalomyelitis; EU: European Union; FBS, fetal bovine serum; GEE, generalized equation;

GFAP,

glial

fibrillary

acidic

protein;

GM-CSF,

granulocyte/macrophage colony stimulating factor; G-MDSCs, granulocytic myeloidderived suppressor cells; HE, hematoxylin and eosin; Ig, immunoglobulin; iNOS, inducible nitric oxide synthase; IFN-γ, interferon-γ; i.v. intravenous; IL, interleukin; Ii, invariant chain; KB, Klüver-Barrera; LPS, lipopolysaccharide; LEA, Lycopersicon esculentum agglutinin; MHC, major histocompatibility complex; MFI, mean fluorescence intensity; MS, multiple sclerosis; BM, bone marrow; MBP, myelin basic protein; MOG, myelin oligodendrocyte protein; M-MDSCs, monocytic myeloid-derived suppressor cells; MDSCs, myeloid-derived suppressor cells; NI, non-immunized; NT, non-treated mice; OD, optical density ; NX-e, Phoenix ecotropic packaging cell line; PBS, phosphate buffered saline; PD-L1, programmed death ligand-1; PG, prostaglandins; PLP, proteolipidic protein; ROS, reactive

4

ACCEPTED MANUSCRIPT

oxygen species; Tregs, regulatory T cells; RT, room temperature; s.c., subcutaneous; SD, standard deviation; SEM, standard error of the mean; Th, T helper; TGF-β, transforming

AC CE P

TE

D

MA

NU

SC

RI

PT

growth factor beta; VEGF, vascular endothelial growth factor.

5

ACCEPTED MANUSCRIPT

INTRODUCTION

PT

Myeloid-derived suppressor cells (MDSCs) comprise highly heterogeneous populations of immature myeloid cells that inhibit adaptive immune responses. They can be divided in

RI

two main types, granulocytic (G-MDSCs) and monocytic MDSCs (M-MDSCs), which differ

SC

in their morphology, phenotype and function. Murine MDSCs express CD11b (αM-integrin, Mac-1) and Gr-1 (Bronte et al., 1998). Specifically, G-MDSCs are CD11b+Gr-1high, whereas

NU

M-MDSCs are CD11b+Gr-1low (Movahedi et al., 2008). Although MDSCs belong to the

MA

innate immune system, their suppressive effects are mainly exerted on adaptive immune responses (Gabrilovich and Nagaraj, 2009) through several mechanisms, including the expression of arginase-1, inducible nitric oxide synthase (iNOS), production of reactive

TE

D

oxygen species (ROS), expression of programmed death ligand-1 (PD-L1, B7-H1, CD274) and the induction of regulatory T cells (Tregs) (Condamine and Gabrilovich, 2011; Huang

AC CE P

et al., 2006). Although MDSCs were originally described in cancer (Gabrilovich et al., 2012), in which they exert a clear detrimental effect, they also play important physiological roles in acute inflammation processes (such as burns and sepsis) in which they enhance innate immune responses (Cuenca et al., 2011). In individuals with cancer, these cells expand and accumulate under the influence of soluble factors such as granulocyte/macrophage colony stimulating factor (GM-CSF), vascular endothelial growth factor (VEGF) or prostaglandins (PG), among others, which can be secreted by the tumor cells (Serafini et al., 2006). Additional factors such as hypoxia (i.e., within the tumor), PGE2, interferon-γ (IFN-γ), transforming growth factor beta (TGF-β), ligands for Toll-like receptors, interleukin (IL)-4 or IL-13 can mediate their activation (Gabrilovich and Nagaraj, 2009; Serafini, 2013). Expansion and activation of MDSCs also occur in chronic inflammatory diseases, in which their suppressive properties

6

ACCEPTED MANUSCRIPT

may limit potential deleterious effects that are associated with persistent activation of inflammatory cells (Serafini, 2013), although this may contribute to pathogen persistence

PT

(i.e., in chronic infections) (Cai et al., 2013; Qin et al., 2013). In such scenarios,

RI

pharmacological inhibition of MDSCs has shown great therapeutic value, for instance in

SC

cancer immunotherapy (Draghiciu et al., 2015). In other pathological situations, such as autoimmune diseases, organ transplantation or graft versus host disease these cells are

NU

beneficial and can be exploited therapeutically (Garcia et al., 2010; Highfill et al., 2010). Although MDSCs belong to the innate immune system, they can produce both non-specific

MA

and antigen-specific immune suppression, and to which extent one predominates over the other depends on many factors, including the amount and type of MDSCs, their activation

D

level, their localization and environmental milieu, the experimental model used or their

TE

pathological context, which can be antigen-specific (Solito et al., 2011).

AC CE P

MDSCs have been investigated and used therapeutically in preclinical models of autoimmunity and chronic inflammation (Crook and Liu, 2014; Drujont et al., 2014; Yin et al., 2010). Experimental autoimmune encephalomyelitis (EAE) is induced in rodents upon immunization with autoantigens presented to MHC-II-restricted CD4+ T cells, and involve both T helper (Th)1 and Th17 cells. In mice with EAE, MDSCs usually play a beneficial role (Moline-Velazquez et al., 2011; Zhu et al., 2007), although a pathogenic effect has also been reported for some subtypes (King et al., 2009). In addition, different MDSC types have been employed with variable results (Crook and Liu, 2014; Ioannou et al., 2012; Yi et al., 2012; Zhu et al., 2011). We previously reported that a single infusion of bone marrow (BM) cells transduced with an encephalitogenic peptide (myelin oligodendrocyte glycoprotein [MOG]40-55) ameliorated EAE (Eixarch et al., 2009). This effect was strictly dependent upon MOG40-55 expression by the BM cells. We also

7

ACCEPTED MANUSCRIPT

demonstrated that both G- and M-MDSCs constituted the most abundant cell types that were generated in retroviral transduction of murine BM cells (Gomez et al., 2014). In this

PT

work, we further investigate the phenotype and function of these ex vivo generated

RI

antigen-expressing MDSCs and we analyze their preventive and therapeutic effects in the

NU

SC

MOG40-55-induced EAE model.

MATERIALS & METHODS

MA

Retroviral vectors and producer cell lines

Generation of the murine leukemia virus–based retroviral vectors pSF91-Ii-IRES-EGFP (Ii)

TE

D

and pSF91-IiMOG-IRES-EGFP (IiMOG) has been described elsewhere (Eixarch et al., 2009). The stable human vector-producing cell lines Phoenix ecotropic (NX-e)/Ii and NX-

AC CE P

e/IiMOG were also generated in our laboratory from the NX-e packaging cell line (a gift from G. Nolan, Stanford University, CA, USA), as described previously (Eixarch et al., 2009). The cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; L0101500) that was supplemented with 10% fetal bovine serum (FBS; 518B-500), 2 mM Lglutamine (X0550-100), 50 IU/ml penicillin and 50 μg/ml streptomycin (L0022-100) (all from Biowest, Nuaillé, France). BM cell transduction and the isolation of total MDSCs BM cells were obtained, cultured and transduced as previously described (Gomez et al., 2014). To isolate the total MDSCs, a fraction of these cells was magnetically labeled with CD11b-coated microbeads (130-049-601; Miltenyi Biotec, Teterow, Germany) and the MDSCs were isolated by positive selection according to the manufacturer’s instructions.

8

ACCEPTED MANUSCRIPT

The transduction efficiency, purity of the isolated MDSCs and the characterization of the isolated MDSCs were assessed by flow cytometry. Both purified MDSCs or unfractionated

RI

BM cells and MDSC transfer

PT

BM cells were then used for in vitro and in vivo studies.

SC

In the preventive approach, a single intravenous (i.v.) infusion of 1x106 transduced BM

NU

cells or 0.5x106 purified MDSCs was administered into recipient mice seven days before EAE induction. In the therapeutic experiments, 1x106 BM cells or 0.5-1x106 MDSCs were

MA

infused once the clinical signs of disease were already present in the vast majority of the animals (13-14 days after EAE induction). To avoid bias, the mice were randomly allocated

D

and appropriately matched into five experimental groups with similar clinical characteristics

TE

(randomization was based on both the clinical and the cumulative score at the day of

AC CE P

treatment). Mice and EAE induction

Five- to six-week-old (donor mice) and seven- to eight-week-old (receptor mice) female C57BL6/J mice were purchased from Harlan Laboratories (Udine, Italy). All of the experimental procedures were approved by our institutional Ethics Committee on Animal Experimentation and were performed according to European Union (EU) and governmental regulations (Generalitat de Catalunya, Decret 214/97 July 30th and EU Directive 2010/63/EU). Anesthetized mice were immunized with subcutaneous (s.c.) injections of saline containing 100 µg of MOG40-55 (Proteomics Section, Universitat Pompeu Fabra, Barcelona, Spain) that was emulsified in incomplete Freund's adjuvant (F5506; Sigma-Aldrich, St Louis, MO, USA) containing 4 mg/ml of Mycobacterium tuberculosis H37RA (231141; Becton, Dickinson and Company, Sparks, MD, USA). At

9

ACCEPTED MANUSCRIPT

days 0 and 2 post-immunization (p.i.), the mice received 250 ng of pertussis toxin (P7208; Sigma-Aldrich) i.v. Mice that were immunized in the same way but without the MOG40-55

PT

peptide were included as controls for the immunization process.

RI

Clinical follow-up and motor performance test

SC

The mice were weighed and examined daily for neurological signs using the following

NU

criteria: 0, no clinical signs; 0.5, partial paresis of tail; 1, paralysis of whole tail; 2, mild paraparesis of one or both hind limbs; 2.5, severe paraparesis or paraplegia of hind limbs;

MA

3, mild tetraparesis; 4, tetraparesis (severe in hind limbs); 4.5, severe tetraparesis; 5, tetraplegia; and 6, death (modified from Espejo et al. (Espejo et al., 2001)). Weight loss

D

was calculated as the percentage of the variation in daily weight compared with the weight

TE

of each animal on the day of immunization. Score 5 and weight loss >30% were defined as

AC CE P

endpoint criteria to minimize the suffering and to guarantee the animal welfare. At the end of each experiment (only in the therapeutic approaches), the motor performance was evaluated using a rotarod apparatus (Ugo Basile, Comerio, Italy) that was set to accelerate from a speed of 4 to 40 rotations per minute in a 300-second time trial. Each mouse was then placed on the rotating cylinder and the amount of time the animals remained walking on the cylinder without falling (rotarod latency) was recorded. Each mouse was given two trials on the rotarod, which were averaged for each individual mouse and group averages were calculated for all animals within a given treatment group. All of the experiments were performed in a blinded manner in such a way that the investigator examining and evaluating the animals was kept unaware of the treatment that was administered to each subject. All data presented here are in accordance with the guidelines suggested for EAE publication (Baker and Amor, 2012).

10

ACCEPTED MANUSCRIPT

Immunophenotypic analysis

PT

The phenotypic studies of myeloid and lymphoid cells were performed using the antimouse antibodies that are listed in Table S1. MDSCs were analyzed at baseline and in an

RI

activated state after 18 hours of stimulation with 2 ng/ml of IFN-γ (315-05; Peprotech,

SC

Rocky Hill, NJ, USA) and/or 100 ng/ml of lipopolysaccharide (LPS; L4391; Sigma-Aldrich). In all cases, the antibody against mouse CD16/32 (101310; clone 93, BioLegend, San

NU

Diego, CA, USA) was used to block non-specific unions. The data acquisition was performed on a FACSCanto™ flow cytometer (BD Biosciences, San Jose, CA, USA) and

MA

analyzed with FCS Express software, v4 (De Novo Software, Los Angeles, CA, USA).

D

Histopathology and immunostaining of the central nervous system (CNS)

TE

At the end of the EAE experiments (33-35 days after immunization), the animals were

AC CE P

euthanized by CO2 asphyxiation and the brains and spinal cords were removed, fixed with 4% paraformaldehyde (141451; Panreac, Castellar del Vallès, Spain) for 24 hours and embedded in paraffin wax. The brain and spinal cords were cut into 4-μm-thick serial sections. For histopathology, the sections were stained with hematoxylin and eosin (HE) and Klüver-Barrera (KB) to assess the degree of inflammation and demyelination, respectively. For immunostaining, the sections were attached to glass slides that were pretreated with poly-L-lysine (S21.2113.A; Leica Microsystems Plus Slides, Wetzlar, Germany). Subsequently, the tissues were deparaffinized in xylene and rehydrated through a descending alcohol battery, ending in phosphate buffered saline (PBS). Afterwards, antigen retrieval was performed in 10 mM citrate (pH=6) for SMI-32 or in protease type XIV (Sigma-Aldrich) for CD3. Non-specific protein binding was blocked by incubating the sections with blocking solution (0.2% BSA in PBS) for 1 hour at room

11

ACCEPTED MANUSCRIPT

temperature (RT). The tissues were then incubated overnight at 4ºC with primary antibodies (Table S2) that were diluted in blocking solution. After several washes with

PT

PBS, the sections were incubated at RT with their respective secondary antibodies or

RI

streptavidin diluted in PBS (Table S2).

SC

Cell infiltration (HE) was evaluated according to the following criteria: 0, no lesion; 1, cellular infiltration only in the meninges; 2, very discrete and superficial infiltrates in the

NU

parenchyma; 3, moderate infiltrates (< 25%) in the white matter; 4, severe infiltrates (25 50%) in the white matter; and 5, more severe infiltrates (> 50%) in the white matter.

MA

Demyelination (KB staining) was scored as follows: 0, no demyelination; 1, little demyelination only around infiltrates and involving less than 25% of the white matter; 2,

D

demyelination involving less than 50% of the white matter; and 3, diffuse and widespread

TE

demyelination involving more than 50% of the white matter. Three randomly chosen areas

AC CE P

(1 mm2) along the spinal cord were analyzed. The CD3-positive cells in infiltrates were counted manually. Lycopersicon esculentum agglutinin (LEA), glial fibrilary acidic protein (GFAP), SMI-32 and myelin basic protein (MBP) quantifications were performed with Image J analysis software (http://imagej.nih.gov/ij/). All analyses were performed in a blinded manner, in such a way that the investigator examining and evaluating the samples was kept unaware of the treatment administrated to each subject. Apoptosis assays In order to determine if the ex vivo generated MDSCs could induce apoptosis to T cells, both antigen- and Ii-expressing MDSCs were incubated with splenocytes from EAE mice (1:1) and stimulated with the MOG40-55 peptide (5 μg/ml). After 18 h the cells were harvested and stained with anti-CD3e and anti-CD4 antibodies (both listed in Table S1)

12

ACCEPTED MANUSCRIPT

followed by PE Annexin V and 7-aminoactinomycin D (7-AAD) staining using the Apoptosis Detection Kit I (559763; BD Biosciences) following the manufacturer’s

PT

instructions. The data acquisition was performed on an LSR Fortessa™ flow cytometer

RI

(BD Biosciences) and analyzed with the FCS Express software, v4.

SC

Anti-MOG40-55 antibody detection

NU

Anti-MOG40-55 antibody detection was performed as previously described (Eixarch et al., 2009). Briefly, 96-well flat-bottom plates (3590; Costar, Sigma-Aldrich, Saint Louis, MO,

MA

USA) were coated overnight with 0.1 μg/well of MOG40–55. Serum samples were added in duplicate and a goat anti-mouse IgG (H+L) horseradish peroxidase-conjugated antibody

D

(550337; BD Pharmingen, San Jose, CA, USA) was used to reveal the binding of the

TE

eventual anti-MOG antibodies. After the addition of the TMB Substrate Reagent Set (555214; BD Pharmingen), the plates were read at 450 nm. The results are presented as

AC CE P

the mean optical density (OD) of each sample. Positivity was defined as an OD greater than the mean OD plus two standard deviations (SD) of sera from saline-immunized control mice.

Statistical analysis

Statistical analysis was performed using the SAS 9.3 program (SAS Institute Inc., Cary, NC, USA), the SPSS v21.0 program (SPSS Inc., Chicago, IL, USA) for Windows or the GraphPad Prism program v5.1 (GraphPad Inc., La Jolla, CA, USA). Depending on the applicable conditions, Mann-Whitney, Student’s t- and Wilcoxon rank-sum tests were utilized to report exact p-values, and they were paired or unpaired, depending on the type of data, with one or two tails (depending on whether the hypothesis was uni- or bidirectional) and were used for mean value comparisons between the groups. For mean

13

ACCEPTED MANUSCRIPT

comparisons of three or more groups, a generalized estimating equation (GEE) method was used to account for the repeated measures design, and a Bonferroni correction was

PT

applied to adjust for multiple comparisons. In all cases, differences were considered

RI

statistically significant when p-values were below 0.05. Quantitative data are presented as

SC

mean values±standard deviation (SD) in the text and tables, and as mean values±standard error of the mean (SEM) in the figures.

NU

RESULTS

MA

Phenotypic characterization of transduced MDSCs

We previously showed that the two main cell populations that were generated during

D

retroviral transduction of murine BM cells consisted of G- and M-MDSCs, which exhibited

TE

arginase-1 and iNOS activities and produced ROS (Gomez et al., 2014). In this work, we

AC CE P

performed a more detailed phenotypic analysis of the two MDSC subsets, which were generated ex vivo. The expression of ligands of the B7 family PD-L1, PD-L2, CD80, CD86 as well as MHC-II (I-Ab) were analyzed by flow cytometry in both the M- and G-MDSC subsets at day four of culture, both at baseline and after exposure to inflammatory stimuli (IFN-γ and/or LPS).

Both M- and G-MDSCs expressed the immunomodulatory molecule PD-L1 (Figure 1a-b), and its expression level was significantly increased upon stimulation with IFN-γ, LPS and with IFN-γ plus LPS, which had a synergistic effect evidenced by the increases in the mean fluorescence intensity (MFI) (Figure 1a-b). The cells did not express PD-L2, even after stimulation (data not shown). CD80 and CD86 were expressed in a higher proportion of M-MDSCs than in G-MDSCs, and these proportions increased upon stimulation in parallel with increases in the levels of MFI (Figure 1c-d). Similarly, MHC-II molecule

14

ACCEPTED MANUSCRIPT

expression (I-Ab) was detected in 45.2% of the M-MDSCs but only in 2% of the G-MDSCs, with significant increases upon stimulation. In this case, MFI increased in M-MDSCs but

PT

not in G-MDSCs (Figure 1e).

RI

Isolated MDSC characterization

SC

To determine whether the ex vivo generated MDSCs were responsible or contributed to

NU

the therapeutic effects observed in the EAE model that we previously reported (Eixarch et al., 2009), these cells were isolated from total BM cells at day 4 of the transduction culture

MA

via the immunomagnetic selection of CD11b+ cells. The purity of the isolated cells was assessed by flow cytometry prior to each experiment by staining the isolated cells with

D

anti-CD11b and anti-Gr-1 antibodies. Virtually all isolated cells expressed Gr-1 (data not

TE

shown). The average purity of the isolated cells in all these experiments was 98.2±1.6%

AC CE P

for the IiMOG-MDSCs and 98.2±1.3% for the Ii-MDSCs. To better characterize which cell populations were present in the cell products after CD11b positive selection, the cells were stained with myeloid, T- and B-lineage markers and analyzed by flow cytometry. The vast majority of cells corresponded to MDSCs, of which 19.7±3.2% were M-MDSCs and 68.7±4.8% were G-MDSCs. As we expected, the CD11benriched cell products contained very small proportions of Lin- cells or lymphocytes, and there was a small percentage of cells that were positive for the CD11c marker (5.1±1.9%) that were also Gr-1low. The effectively transduced cells (positive for the EGFP marker) primarily consisted of M- and G-MDSCs and represented 40.4±10% and 38.9±8.6% of the total transduced cells, respectively. Infusion of IiMOG-expressing BM cells and MDSCs protects against EAE

15

ACCEPTED MANUSCRIPT

To evaluate whether the BM cells or isolated MDSCs were capable of protecting mice from EAE, normal non-myeloablated C57BL6/J mice were infused with 1x106 unfractionated BM

PT

cells or 0.5x106 total MDSCs that were transduced with either the control (Ii-treated mice)

RI

or IiMOG vectors (IiMOG-treated mice) seven days before EAE induction. Moreover, a

SC

group of untreated mice was included as an additional control (NT). Upon EAE induction, both of the IiMOG-treated groups were significantly protected against

NU

the disease in comparison with their respective controls. These IiMOG-treated mice exhibited significantly lower maximum and cumulative clinical scores than their

MA

counterparts (Table 1). A single infusion of IiMOG-expressing MDSCs ameliorated the clinical course of the disease to a similar extent as the unfractionated IiMOG-expressing

D

BM cells (Figure 2a-b), in comparison with that of their respective controls, which did not

TE

differ from that of untreated animals, suggesting that MDSCs were the main contributors to

AC CE P

the therapeutic effects. Consistently, this improvement in the clinical score was paralleled by lower weight loss in the IiMOG-treated animals in comparison with their respective controls (Figure S1a-b).

Infusion of IiMOG-specific transduced BM cells and MDSCs improves established EAE

In the therapeutic approaches, the animals received the transduced cells once the majority of them had developed clinical signs of the disease (day 13-14 p.i.). At this point, the mice were randomized into different experimental groups (NT, Ii-BM, IiMOG-BM, Ii-MDSCs and IiMOG-MDSCs) in such a way that the clinical parameters were comparable between all of the groups (Table S3). The mice received either 1x106 transduced BM cells or two different doses of transduced MDSCs (0.5x106 or 1x106). The animals infused with 1x106 IiMOG-

16

ACCEPTED MANUSCRIPT

BM cells or 1x106 IiMOG-MDSCs presented a significantly lower cumulative clinical score in comparison with their counterparts (Table 1 and Figure 2c-d), a difference that was not

PT

observed in the mice receiving only 0.5x106 IiMOG-MDSCs (data not shown). Moreover, in

RI

contrast to the Ii-treated controls that continued to develop a chronic EAE after cell

SC

infusion, most of the IiMOG-treated animals did not develop the typical clinical course of this chronic non-remitting model. Additionally, 37.5% of the animals from the IiMOG-BM

NU

cell group and 28.57% from the IiMOG-MDSC-treated group presented disease remission, which was not observed in the control groups. In this study, a remission was defined as an

MA

improvement in the clinical score for three consecutive days with respect to the score on the day of the cell infusion. Additionally, in accordance with the clinical outcome, the

D

IiMOG-treated animals presented significantly lower weight loss than their controls (Figure

TE

S2c-d).

AC CE P

To determine whether the clinical score improvements were associated with motor function improvements, the mice were evaluated for motor coordination and balance using a rotarod apparatus at the end of each experiment. In accordance with the lower clinical scores, the IiMOG-treated mice remained on the rotating cylinder for longer periods than their respective controls, although the differences were only statistically significant between the BM cell-treated mice (Figure S2e-f). CNS pathology is improved in IiMOG-treated animals To assess whether EAE improvement was associated with decreased neuropathology, histopathological studies were performed on the brains and spinal cords at the end of the experiments (day 33-35 p.i.). HE and KB stainings were performed in the CNS of all the experimental animals. The brains of the mice from the IiMOG-treated animals showed less

17

ACCEPTED MANUSCRIPT

infiltration in the HE staining and less demyelination in the KB staining with respect to the Ii-treated controls (data not shown). The spinal cords of three representative mice from

PT

each group in one of the experiments were also immunostained to evaluate T-cell

RI

infiltration (CD3), microglia activation (LEA), reactive astrogliosis (GFAP), axonal damage

SC

through specific detection of non-phosphorylated neurofilaments (SMI-32) and myelin (MBP). In both the preventive and therapeutic approaches, all of the IiMOG-treated animal

NU

groups analyzed presented milder histopathological findings than their respective controls, which showed extensive inflammatory infiltrates in the spinal cord white matter and

MA

demyelination in areas with moderate to severe inflammatory infiltration (data not shown). Moreover, the mice that were preventively and therapeutically treated with IiMOG-

D

expressing cells presented significantly less T-cell infiltration, microglia activation, reactive

TE

astrogliosis, axonal damage and demyelination (Table 2, Figure 3).

AC CE P

Treatment with IiMOG-MDSCs does not change the proportion of Tregs, but reduces that of activated T cells and increases B cells with a regulatory phenotype in the spleens

T- and B-splenocyte populations were studied at the end of each experiment (day 33-35 p.i.) with flow cytometry. Regarding the percentages of Tregs with the CD4+CD25+FoxP3+ phenotype, no statistically significant differences were observed between the different groups (data not shown). In the preventive approach there were no significant differences in the percentage of activated T cells (CD3+CD4+CD25+FoxP3-) between the two groups treated with BM cells (Ii-BM: 2.3±0.8 vs. IiMOG-BM: 2.0±0.8), but the IiMOG-MDSC treated mice exhibited a significantly lower percentage (IiMOG-MDSCs: 2.1%±1.1 vs. IiMDSCs: 3.5%±1.1, p=0.043) than their counterparts (Figure 4a). Moreover, the mice treated with both IiMOG-cells exhibited an increased percentage of B cells with a

18

ACCEPTED MANUSCRIPT

regulatory phenotype (B220+CD1dhighCD5+) in comparison with their respective controls, although it only reached statistical significance in the MDSCs treated animals (Ii-BM:

PT

2.2%±0.6 vs. IiMOG-BM: 2.9%±0.8, p=0.06; Ii-MDSCs: 2%±0.5 vs. IiMOG-MDSCs:

RI

2.7%±0.6, p=0.03) (Figure 4b).

SC

In the therapeutic experiments, the IiMOG-BM cell mice presented a significantly lower proportion of activated CD4+ T cells than their respective controls (Ii-BM: 2.8%±0.7 vs.

NU

IiMOG-BM: 1.9%±0.4, p=0.021), a difference that was not observed between the MDSCtreated groups (Ii-MDSCs: 3.2±0.8 vs. IiMOG-MDSCs: 3.0±0.8) (Figure 4c). However, as

MA

with the preventive approach, the IiMOG-MDSC-treated mice had a higher proportion of B cells with a regulatory phenotype than their Ii-MDSCs-treated counterparts, although this

TE

5.0%±1.6, p=0.1) (Figure 4d).

D

difference did not reach statistical significance (Ii-MDSCs: 3.5%±1.6 vs. IiMOG-MDSCs:

AC CE P

Antigen-expressing MDSCs induce apoptosis in CD4+ T cells from EAE mice We analyzed whether the ex vivo generated, transduced MDSCs could induce apoptosis in T cells from mice with EAE. To this end, both IiMOG-expressing and control MDSCs were incubated with splenocytes for 18 h and apoptosis was measured in the different T cell subpopulations by flow cytometry using the appropriate gating. We observed that apoptosis in total (CD3+) and CD4+ T cells was significantly higher when the cells were incubated with IiMOG-expressing MDSCs than with their controls (Figure 5). The presence of specific antibodies against MOG40-55 does not affect EAE outcomes Serum samples were collected at the end of the experiment and the presence of specific IgG antibodies to the encephalitogenic peptide was assessed using an ELISA technique. No significant differences were found in the specific antibody levels between the IiMOG-

19

ACCEPTED MANUSCRIPT

treated and Ii-treated groups, neither in the preventive nor in the therapeutic experiments (Figure 6). Moreover, no differences were observed between the animals from the IiMOG-

PT

treated groups that improved and those that did not improve (data not shown), which

RI

suggests that anti-MOG IgG antibodies do not play a major role in the therapeutic effect

NU

SC

observed in this model.

DISCUSSION

MA

In this study, we showed that MDSCs expressing a self-antigen can induce specific immune tolerance in a murine model of autoimmunity. In a previous work, we

D

hypothesized that creating a molecular chimerism with syngeneic hematopoietic cells

TE

expressing MOG40-55 targeted to the MHC-II antigen presentation pathway could induce

AC CE P

tolerance and prevent and improve MOG-induced EAE. With the preventive approach, stable molecular chimerism was associated with robust immune tolerance. However, with the therapeutic approach, even after partial myeloablation, molecular chimerism was observed in the control animals treated with Ii-transduced BM cells but not in those infused with IiMOG-expressing cells. This result was anticipated given that the animals had been previously immunized with the MOG40-55 and the same autoimmune response causing EAE most likely resulted in the rejection of the infused IiMOG-expressing cells. However, EAE was significantly improved upon the infusion of IiMOG-expressing BM cells but not with control cells (Eixarch et al., 2009). This lack of engraftment of antigen-expressing cells made us hypothesize that myeloablation was dispensable for the therapeutic effect, which was confirmed in non-myeloablative experiments, and that the therapeutic effect was most likely related to mature cells that were present in the cultures and presented the MOG40-55

20

ACCEPTED MANUSCRIPT

in a tolerogenic manner. We found that the most abundant cell populations after retroviral transduction of murine BM were of the myeloid lineage, and that they fulfilled the

PT

morphologic, phenotypic and functional criteria of MDSCs (Gomez et al., 2014). Herein,

RI

we further characterized the phenotype of these ex vivo generated MDSCs and we

SC

evaluated their therapeutic potential in the EAE model. In our in vitro experiments we stimulated the cells with IFN- and LPS, a widely accepted method for T-cell activation in

NU

vitro, although it does not mimic what occurs in vivo, in which MDSCs are stimulated by activated T cells and the inflammatory milieu. We show that the majority of ex vivo

MA

generated M-MDSCs (and to a lesser extent the G-MDSCs) expressed CD80 (B7.1), CD86 (B7.2), PD-L1, all of them molecules of the B7 superfamily and ligands of immune

D

inhibitory receptors that are critical for EAE induction and development (Chang et al.,

TE

1999). CD80 and CD86 can bind either to the co-stimulatory molecule CD28 or the

AC CE P

inhibitory CTLA-4 on the T-cell surface, while PD-L1 is the main ligand of PD-1, one of the most important effectors that maintain peripheral tolerance in tissues and prevents from autoimmunity (Dai et al., 2014). Expression of these molecules increased upon stimulation with IFN-γ plus LPS, whereas PD-L2 expression was undetectable in our MDSCs, in agreement with previous reports (Topalian et al., 2015). The importance of the PD-1/PDL1 pathway in immune regulation has been recognized for many years (Freeman et al., 2000; Gianchecchi et al., 2013). In cancer patients, its blockade can successfully revert immunosuppression and synergize with other therapies (Sunshine and Taube, 2015). In autoimmune diseases, the importance of MDSCs is still controversial. Published data establishes a crucial role for PD-1/PD-L1 engagement in the modulation of lymphocyte activation and autoimmune responses (Dai et al., 2014). Adoptive transfer of G-MDSCs from PD-L1-deficient mice, but not from wild-type G-MDSCs, failed to suppress EAE pathology, suggesting a PD-L1-dependent mediated regulation of the immune responses

21

ACCEPTED MANUSCRIPT

in EAE (Ioannou et al., 2012). On the other hand, MDSCs were found to contribute to the pathogenesis of EAE by driving Th17 responses from naive CD4+ T cell precursors (Yi et

PT

al., 2012).

in

cancer

patients,

MDSC

expansions

rarely

produce

systemic

SC

However,

RI

In tumors, MDSCs can produce both non-specific and antigen-specific suppression.

immunosuppression, but rather a selective suppression to tumor-associated antigens. This

NU

is due to the relative abundance of these antigens, which can be processed and presented by MDSCs. This also helps explaining the difficulties in generating effective tumor-specific

MA

responses using vaccines in cancer immunotherapy. In a previous report, G-MDSCs isolated from the spleens of mice with EAE suppressed antigen-specific Th1 and Th17

D

immune responses (both of which are critical on EAE pathogenesis) and had a therapeutic

TE

effect on the disease, which was dependent on PD-L1 expression by the G-MDSCs

AC CE P

(Ioannou et al., 2012). In our study, Ii-expressing MDSCs did not have any therapeutic effect, but this can be explained by the higher cell dose and the earlier time of infusion used by these authors (2x106 G-MDSCs on days 4 and 7 p.i.), prior to the clinical onset of the disease. Regarding the cell doses required to obtain therapeutic effects in our model, we initially used the dose that worked preventively. We transferred 0.5x106 purified MDSCs, as we estimated that MDSCs represented at least half of the total cell content in our BM cultures. However, as this failed to improve the disease we decided to double this dose. With 1x106 IiMOG-expressing MDSCs, the clinical efficacy increased and resembled that obtained with the total BM cells. The lack of effect of the low cell dose might be related to differences in the transduction efficiency, which varied among the different experiments, or to the manipulation associated with the MDSC purification, which could have reduced their viability and/or functionality.

22

ACCEPTED MANUSCRIPT

Antigen-specific CD8+ T cell suppression by MDSCs is fundamentally exerted by GMDSCs, as they provide cell-to-cell contact, express MHC-I and produce high levels of

PT

ROS, which can nitrosylate and alter the specificity of T cell receptors (TCRs) and the

RI

functionality of the CD8 molecule in these cells. On the other hand, M-MDSCs can induce both antigen-specific and non-specific suppression, which are mediated by soluble factors

SC

and do not require cell-to-cell contact. On the CD4+ T cell compartment, antigen-specific

NU

suppression by MDSCs is still controversial (Solito et al., 2011). In murine tumor models, MHC-II expression by MDSCs is usually low. However, MDSCs expressing MHC-II were

MA

found to suppress antigen-specific CD4+ T cell responses in tumor-bearing mice (Nagaraj et al., 2012). In addition, antigen-specific CD4+ T cells enhanced MDSCs suppressive

D

activity, which became non-specific. This highlighted the importance of a crosstalk

TE

between these two immune cell populations (Nagaraj et al., 2013). In our experimental model, the sequence encoding MOG40-55 replaced the CLIP region of the murine invariant

AC CE P

chain (Ii) to target its expression to the MHC-II antigen -pathway (Bischof et al., 2001; Eixarch et al., 2009). This selective targeting, together with the fact that a significant fraction of our ex vivo generated cells (mainly M-MDSCs) expressed I-Ab and the reduced proportion of activated CD4+ T cells in the spleens of IiMOG-treated mice suggest that the therapeutic effects are related to a MHC-II-mediated suppression on CD4+ T cells. This is also supported by a previous work in a proteolipidic protein peptide (PLP)139-151-induced EAE model in which i.v. injection of a recombinant Ii carrying this peptide instead of the CLIP region (Ii-PLP139–151) improved the disease and induced apoptosis and deletion of antigen-specific activated T cells, while the proportion of antigen-specific FoxP3+ Tregs was increased in the treated mice (Lange et al., 2009).

23

ACCEPTED MANUSCRIPT

Although MDSCs can induce the generation of Tregs in vitro and in vivo (Serafini et al., 2008), we did not observe increased proportions of CD4+CD25+FoxP3+ Tregs in the

PT

IiMOG-treated animals. This could be related to the relatively long interval between the cell

RI

infusions and the analyses, especially in the preventive experiments. However, we cannot rule out potential contributions of other regulatory cell types (such as Tr-1 or CD8+

SC

regulatory T cells). On the other hand, B220+CD1dhighCD5+ cell proportions were increased

NU

in mice preventively treated with IiMOG-MDSCs. This phenotype includes a type of regulatory B cells, also known as Bregs or B10, which represent 1-3% of B cells in the

MA

spleens of normal adult mice (Tedder, 2015), and are functionally defined by their ability to differentiate and secrete IL-10, which mediates their immunosuppressive effects. This

D

occurs upon T-cell activation by the antigen and release of IL-21 (Yoshizaki and Tedder,

TE

2015). B10 cells can modulate antigen presentation by dendritic cells, inhibit T-cell responses and induce antigen-specific Tregs (Tedder, 2015). In the EAE model, the

AC CE P

transfer of these B cells prevents and ameliorates the disease (Matsushita et al., 2008). In our experiments, the relative increase of a cell population with this phenotype in the animals treated with the antigen-expressing cells suggests that B10 cells could participate in the therapeutic effect. This is also supported by the increased IL-10 production by the splenocytes of IiMOG-treated mice upon antigen exposure that we found in our previous studies using BM cells (Eixarch et al., 2009). All of the mouse groups in which EAE was induced developed low levels of anti-MOG40-55 IgG antibodies, although no significant differences were found between the groups. This observation confirms our previous results (Eixarch et al., 2009) and further supports the notion of a small role, if any, for these antibodies in the pathogenesis of this experimental disease, or at least in the therapeutic mechanisms of the antigen-expressing cells.

24

ACCEPTED MANUSCRIPT

In the therapeutic experiments the IiMOG-expressing cells infused had a fast, potent and long-lasting therapeutic effect, in spite of the fact that they were most likely rejected after

PT

infusion, as we documented in our previous experiments with total BM cells (Eixarch et al.,

RI

2009). Whether other cell subpopulations different from MDSCs but also present in the

SC

cultured BM cells such as the mesenchymal stromal cells (MSCs) could have contributed to the therapeutic effect of these BM cells cannot completely be ruled out. MSCs also have

NU

immunomodulatory properties, but in our model the therapeutic effect must be mediated by antigen-expressing cells, since their Ii-expressing counterparts did not have any effect. For

MA

this reason, we believe that the therapeutic effect of the antigen-expressing BM cells was mainly due to the IiMOG-expressing MDSCs present in the cultured BM cells. That G- and

D

M-MDSCs were the most abundant and the most efficiently transduced cell subpopulations

TE

in the total BM cells (e.g. in comparison with MSCs) (Gomez et al., 2012), and the similar type and kinetics of clinical improvement observed upon infusion of purified MDSCs and

AC CE P

total BM cells strongly support this notion. On the other hand, T-cell apoptosis induced by immature myeloid progenitors were reported to modulate EAE (Zhu et al., 2011). Our finding of a higher proapoptotic effect in vitro of antigen-expressing MDSCs on CD4+ T cells from EAE mice in comparison with the control MDSCs may help explaining the antigen specificity of the therapeutic effect observed. We cannot rule out the possibility that IiMOG-expressing MDSCs become activated to a higher extent than Ii-expressing MDSCs upon infusion in vivo by activated MOG-expressing CD4+ T cells. In addition, antigen-specific regulatory cells such as B10 cells are probably induced, thus contributing to maintain tolerance after the IiMOG-expressing MDSCs have disappeared. Finally, we think that ex vivo generated MDSCs expressing self-antigens deserve further investigation as a potential tool to induce tolerance in autoimmune diseases.

25

ACCEPTED MANUSCRIPT

PT

CONCLUSIONS

RI

Our results demonstrate that in vitro-generated BM-derived MDSCs expressing a selfantigen (MOG40-55) targeted to the class II antigen presentation pathway have preventive

SC

and therapeutic effects in MOG-induced EAE, an experimental model of multiple sclerosis.

NU

These effects are antigen-specific and uniquely depend on self-antigen expression by the MDSCs. Phenotypically, the cells (and in particular the M-MDSCs) expressed PD-L1,

MA

CD80, CD86 and the MHC class II molecule I-Ab, which were further induced upon activation. The analysis of lymphoid cell subpopulations in the spleens of treated mice

D

infused points to a reduction in the numbers of effector T cells which can be partly due to

TE

apoptosis induction by the antigen expressing myeloid cells and is also associated with an increase in the number of B cells with a regulatory phenotype (B220+CD1dhighCD5+). We

AC CE P

propose that MDSCs expressing self-antigens can be used therapeutically to induce tolerance in antigen-specific autoimmune diseases.

26

ACCEPTED MANUSCRIPT

ACKNOWLEDGMENTS

PT

We thank C. Baum (Hannover Medical School, Hannover, Germany) for providing the retroviral vector backbones. This work was supported by grants from the "Fondo de

RI

Investigación Sanitaria" (FIS), "Instituto de Salud Carlos III" (ISCIII), and Ministry of

SC

Economy and Competitivity (MEC), Spain (PI 12/01001). The authors thank the “Red Española de Esclerosis Múltiple” and the “Agència de Gestió d’Ajuts Universitaris i de

NU

Recerca” (AGAUR), Generalitat de Catalunya (2014 SGR 1082)”. SC was supported by a VHIR predoctoral fellowship. CC was supported by "Fundació la Marató de TV3"

MA

(493/C/2012). CE was partially supported by the Miguel Servet program (CP13/00028) of

AC CE P

TE

D

the ISCIII, MEC.

27

ACCEPTED MANUSCRIPT

FIGURE LEGENDS

PT

Figure 1) Phenotypic characterization of the in vitro generated M- and G-MDSCs. The phenotype of MDSCs was not influenced by the type of vector used for transduction,

RI

so for practical reasons only data from IiMOG-expressing cells is shown. Expression of

SC

PD-L1 in M- and G-MDSC subsets, both the percentage of positive cells and the MFIs (a, b), are shown, at baseline and after exposure to inflammatory stimuli (IFN-γ or/and LPS).

NU

Expression of CD80, CD86, and MHC-II molecules was assessed at baseline (empty bars) and after 18 h of IFN-γ and LPS stimulation (solid bars) (c-e). IFN-γ, interferon gamma;

MA

LPS, lipopolysaccharide; MFI, mean fluorescence intensity. * p<0.05, ** p<0.01, ***

TE

D

p<0.001. (n=3-4).

AC CE P

Figure 2) Preventive and therapeutic effects of IiMOG-expressing cells on EAE. Mice that were treated preventively with IiMOG-BM cells (a) or IiMOG-MDSCs (b) were protected against EAE (n=1). In the therapeutic approach, mice were infused either with 1x106 BM cells (n=1) (c) or with 1x106 MDSCs (n=2) (d) after the appearance of the first neurological signs (results correspond to a representative experiment). Mice infused with IiMOG-expressing BM cells or MDSCs developed a milder EAE in comparison with their controls. The charts represent the mean±SEM of the daily clinical score for every experimental group. The p values correspond to the statistical significance of the differences between Ii- and IiMOG-treated groups. NT animals are shown as positive controls of the disease. The arrows indicate the day of cell infusion. NT, non-treated. * p<0.05.

28

ACCEPTED MANUSCRIPT

Figure 3) Histologic parameters in the CNS of mice. T-cell infiltration (a) was significantly reduced in the CNS samples of all the IiMOG-treated mice that were

PT

analyzed. Moreover, these mice presented less microglia activation (b), reactive

SC

RI

astrogliosis (c) axonal damage (d) and demyelination (e). * p<0.05, ** p<0.01, *** p<0.001.

NU

Figure 4) Frequency of the splenic activated T cells and regulatory B cell populations in the mice that were treated with IiMOG-BM cells and IiMOG-MDSCs.

MA

Charts (a) and (b) show the percentages of activated T cells (CD25+FoxP3-) and B cells with a regulatory phenotype (CD1dhighCD5+), respectively, in the preventively treated

D

animals. Charts (c) and (d) show the corresponding percentages in the therapeutic

TE

experiments. The activated T-cell percentages were calculated by gating on the CD3+CD4+ T-cell population. The regulatory B-cell percentages were calculated by gating on the

AC CE P

CD45+B220+ B-cell population. The data are represented as the mean±SEM. * p<0.05.

Fig. 5) Antigen specific MDSCs induce apoptosis in splenocytes from EAE mice. MDSCs transduced with Ii (empty bars) or IiMOG (solid bars) were exposed to splenocytes from EAE mice in the presence of the MOG40-55 peptide and apoptosis on T cells was analyzed by flow cytometry after labeling the cells with anti-CD3 and anti-CD4 antibodies. CD8+ cell population was measured indirectly by gating the CD3+ CD4- cell population. IiMOG-expressing MDSCs produced significantly higher levels of apoptosis in total CD3+ and CD4+ T cell populations than their controls. The data are represented as the mean±SEM; * p<0.05. (n=3).

29

ACCEPTED MANUSCRIPT

Figure 6) Anti-MOG40-55 antibodies levels in serum. The charts show the anti-MOG40–55 IgG reactivity in the preventively (a) or therapeutically (b) treated mouse sera. Both the

PT

prevalence and the mean anti-MOG40–55 antibody levels were similar in all of the

RI

experimental groups. The dotted lines represent the mean optical density (OD) of the

AC CE P

TE

D

MA

NU

SC

control sera plus 2 SD. NI, non-immunized; NT, non-treated.

30

ACCEPTED MANUSCRIPT

REFERENCES

PT

Baker, D., Amor, S., 2012. Publication guidelines for refereeing and reporting on animal use in experimental autoimmune encephalomyelitis. J Neuroimmunol 242, 78-83.

RI

Bischof, F., Wienhold, W., Wirblich, C., Malcherek, G., Zevering, O., Kruisbeek, A.M., Melms, A., 2001. Specific treatment of autoimmunity with recombinant invariant chains in which CLIP is replaced by self-epitopes. Proc Natl Acad Sci U S A 98, 12168-12173.

NU

SC

Bronte, V., Wang, M., Overwijk, W.W., Surman, D.R., Pericle, F., Rosenberg, S.A., Restifo, N.P., 1998. Apoptotic death of CD8+ T lymphocytes after immunization: induction of a suppressive population of Mac-1+/Gr-1+ cells. J Immunol 161, 5313-5320.

MA

Cai, W., Qin, A., Guo, P., Yan, D., Hu, F., Yang, Q., Xu, M., Fu, Y., Zhou, J., Tang, X., 2013. Clinical significance and functional studies of myeloid-derived suppressor cells in chronic hepatitis C patients. J Clin Immunol 33, 798-808.

D

Chang, T.T., Jabs, C., Sobel, R.A., Kuchroo, V.K., Sharpe, A.H., 1999. Studies in B7deficient mice reveal a critical role for B7 costimulation in both induction and effector phases of experimental autoimmune encephalomyelitis. J Exp Med 190, 733-740.

AC CE P

TE

Condamine, T., Gabrilovich, D.I., 2011. Molecular mechanisms regulating myeloid-derived suppressor cell differentiation and function. Trends in immunology 32, 19-25. Crook, K.R., Liu, P., 2014. Role of myeloid-derived suppressor cells in autoimmune disease. World J Immunol 4, 26-33. Cuenca, A.G., Delano, M.J., Kelly-Scumpia, K.M., Moreno, C., Scumpia, P.O., Laface, D.M., Heyworth, P.G., Efron, P.A., Moldawer, L.L., 2011. A paradoxical role for myeloidderived suppressor cells in sepsis and trauma. Mol Med 17, 281-292. Dai, S., Jia, R., Zhang, X., Fang, Q., Huang, L., 2014. The PD-1/PD-Ls pathway and autoimmune diseases. Cell Immunol 290, 72-79. Draghiciu, O., Lubbers, J., Nijman, H.W., Daemen, T., 2015. Myeloid derived suppressor cells-An overview of combat strategies to increase immunotherapy efficacy. Oncoimmunology 4, e954829. Drujont, L., Carretero-Iglesia, L., Bouchet-Delbos, L., Beriou, G., Merieau, E., Hill, M., Delneste, Y., Cuturi, M.C., Louvet, C., 2014. Evaluation of the therapeutic potential of bone marrow-derived myeloid suppressor cell (MDSC) adoptive transfer in mouse models of autoimmunity and allograft rejection. PloS one 9, e100013. Eixarch, H., Espejo, C., Gomez, A., Mansilla, M.J., Castillo, M., Mildner, A., Vidal, F., Gimeno, R., Prinz, M., Montalban, X., Barquinero, J., 2009. Tolerance induction in experimental autoimmune encephalomyelitis using non-myeloablative hematopoietic gene therapy with autoantigen. Mol Ther 17, 897-905.

31

ACCEPTED MANUSCRIPT

Espejo, C., Carrasco, J., Hidalgo, J., Penkowa, M., Garcia, A., Saez-Torres, I., MartinezCaceres, E.M., 2001. Differential expression of metallothioneins in the CNS of mice with experimental autoimmune encephalomyelitis. Neuroscience 105, 1055-1065.

SC

RI

PT

Freeman, G.J., Long, A.J., Iwai, Y., Bourque, K., Chernova, T., Nishimura, H., Fitz, L.J., Malenkovich, N., Okazaki, T., Byrne, M.C., Horton, H.F., Fouser, L., Carter, L., Ling, V., Bowman, M.R., Carreno, B.M., Collins, M., Wood, C.R., Honjo, T., 2000. Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J Exp Med 192, 1027-1034. Gabrilovich, D.I., Nagaraj, S., 2009. Myeloid-derived suppressor cells as regulators of the immune system. Nat Rev Immunol 9, 162-174.

NU

Gabrilovich, D.I., Ostrand-Rosenberg, S., Bronte, V., 2012. Coordinated regulation of myeloid cells by tumours. Nat Rev Immunol 12, 253-268.

D

MA

Garcia, M.R., Ledgerwood, L., Yang, Y., Xu, J., Lal, G., Burrell, B., Ma, G., Hashimoto, D., Li, Y., Boros, P., Grisotto, M., van Rooijen, N., Matesanz, R., Tacke, F., Ginhoux, F., Ding, Y., Chen, S.H., Randolph, G., Merad, M., Bromberg, J.S., Ochando, J.C., 2010. Monocytic suppressive cells mediate cardiovascular transplantation tolerance in mice. J Clin Invest 120, 2486-2496.

TE

Gianchecchi, E., Delfino, D.V., Fierabracci, A., 2013. Recent insights into the role of the PD-1/PD-L1 pathway in immunological tolerance and autoimmunity. Autoimmun Rev 12, 1091-1100.

AC CE P

Gomez, A., Espejo, C., Eixarch, H., Casacuberta-Serra, S., Mansilla, M.J., Sanchez, R., Pereira, S., Lopez-Estevez, S., Gimeno, R., Montalban, X., Barquinero, J., 2014. Myeloidderived suppressor cells are generated during retroviral transduction of murine bone marrow. Cell Transplant 23, 73-85. Highfill, S.L., Rodriguez, P.C., Zhou, Q., Goetz, C.A., Koehn, B.H., Veenstra, R., Taylor, P.A., Panoskaltsis-Mortari, A., Serody, J.S., Munn, D.H., Tolar, J., Ochoa, A.C., Blazar, B.R., 2010. Bone marrow myeloid-derived suppressor cells (MDSCs) inhibit graft-versushost disease (GVHD) via an arginase-1-dependent mechanism that is up-regulated by interleukin-13. Blood 116, 5738-5747. Huang, B., Pan, P.Y., Li, Q., Sato, A.I., Levy, D.E., Bromberg, J., Divino, C.M., Chen, S.H., 2006. Gr-1+CD115+ immature myeloid suppressor cells mediate the development of tumor-induced T regulatory cells and T-cell anergy in tumor-bearing host. Cancer Res 66, 1123-1131. Ioannou, M., Alissafi, T., Lazaridis, I., Deraos, G., Matsoukas, J., Gravanis, A., Mastorodemos, V., Plaitakis, A., Sharpe, A., Boumpas, D., Verginis, P., 2012. Crucial role of granulocytic myeloid-derived suppressor cells in the regulation of central nervous system autoimmune disease. J Immunol 188, 1136-1146.

32

ACCEPTED MANUSCRIPT

King, I.L., Dickendesher, T.L., Segal, B.M., 2009. Circulating Ly-6C+ myeloid precursors migrate to the CNS and play a pathogenic role during autoimmune demyelinating disease. Blood 113, 3190-3197.

RI

PT

Lange, C., Doster, H., Steinbach, K., Kalbacher, H., Scholl, M., Melms, A., Bischof, F., 2009. Differential modulation of CNS-specific effector and regulatory T cells during tolerance induction by recombinant invariant chains in vivo. Brain Behav Immunity 23, 861867.

SC

Matsushita, T., Yanaba, K., Bouaziz, J.D., Fujimoto, M., Tedder, T.F., 2008. Regulatory B cells inhibit EAE initiation in mice while other B cells promote disease progression. J Clin Invest 118, 3420-3430.

MA

NU

Moline-Velazquez, V., Cuervo, H., Vila-Del Sol, V., Ortega, M.C., Clemente, D., de Castro, F., 2011. Myeloid-derived suppressor cells limit the inflammation by promoting T lymphocyte apoptosis in the spinal cord of a murine model of multiple sclerosis. Brain Pathol 21, 678-691.

TE

D

Movahedi, K., Guilliams, M., Van den Bossche, J., Van den Bergh, R., Gysemans, C., Beschin, A., De Baetselier, P., Van Ginderachter, J.A., 2008. Identification of discrete tumor-induced myeloid-derived suppressor cell subpopulations with distinct T cellsuppressive activity. Blood 111, 4233-4244.

AC CE P

Nagaraj, S., Nelson, A., Youn, J.I., Cheng, P., Quiceno, D., Gabrilovich, D.I., 2012. Antigen-Specific CD4+ T Cells Regulate Function of Myeloid-Derived Suppressor Cells in Cancer via Retrograde MHC Class II Signaling. Cancer Res 72, 928-938. Nagaraj, S., Youn, J.I., Gabrilovich, D.I., 2013. Reciprocal relationship between myeloidderived suppressor cells and T cells. J Immunol 191, 17-23. Qin, A., Cai, W., Pan, T., Wu, K., Yang, Q., Wang, N., Liu, Y., Yan, D., Hu, F., Guo, P., Chen, X., Chen, L., Zhang, H., Tang, X., Zhou, J., 2013. Expansion of Monocytic MyeloidDerived Suppressor Cells Dampens T Cell Function in HIV-1-Seropositive Individuals. J Virol 87, 1477-1490. Serafini, P., 2013. Myeloid derived suppressor cells in physiological and pathological conditions: the good, the bad, and the ugly. Immunol Res 57, 172-184. Serafini, P., Borrello, I., Bronte, V., 2006. Myeloid suppressor cells in cancer: recruitment, phenotype, properties, and mechanisms of immune suppression. Semin Cancer Biol 16, 53-65. Serafini, P., Mgebroff, S., Noonan, K., Borrello, I., 2008. Myeloid-derived suppressor cells promote cross-tolerance in B-cell lymphoma by expanding regulatory T cells. Cancer Res 68, 5439-5449. Solito, S., Bronte, V., Mandruzzato, S., 2011. Antigen specificity of immune suppression by myeloid-derived suppressor cells. J Leukoc Biol 90, 31-36.

33

ACCEPTED MANUSCRIPT

Sunshine, J., Taube, J.M., 2015. PD-1/PD-L1 inhibitors. Curr Op Pharmacol 23, 32-38.

PT

Tedder, T.F., 2015. B10 cells: a functionally defined regulatory B cell subset. J Immunol 194, 1395-1401.

RI

Topalian, S.L., Drake, C.G., Pardoll, D.M., 2015. Immune checkpoint blockade: a common denominator approach to cancer therapy. Cancer Cell 27, 450-461.

SC

Yi, H., Guo, C., Yu, X., Zuo, D., Wang, X.Y., 2012. Mouse CD11b+Gr-1+ myeloid cells can promote Th17 cell differentiation and experimental autoimmune encephalomyelitis. J Immunol 189, 4295-4304.

NU

Yin, B., Ma, G., Yen, C.Y., Zhou, Z., Wang, G.X., Divino, C.M., Casares, S., Chen, S.H., Yang, W.C., Pan, P.Y., 2010. Myeloid-derived suppressor cells prevent type 1 diabetes in murine models. J Immunol 185, 5828-5834.

MA

Yoshizaki, A., Tedder, T.F., 2015. [IL-21 induces regulatory B cell differentiation and immunosuppressive effect through cognate interaction with T cells]. Nihon Rinsho Meneki Gakkai Kaishi 38, 57-64.

TE

D

Zhu, B., Bando, Y., Xiao, S., Yang, K., Anderson, A.C., Kuchroo, V.K., Khoury, S.J., 2007. CD11b+Ly-6C(hi) suppressive monocytes in experimental autoimmune encephalomyelitis. J Immunol 179, 5228-5237.

AC CE P

Zhu, B., Kennedy, J.K., Wang, Y., Sandoval-Garcia, C., Cao, L., Xiao, S., Wu, C., Elyaman, W., Khoury, S.J., 2011. Plasticity of Ly-6C(hi) myeloid cells in T cell regulation. J Immunol 187, 2418-2432.

34

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 1

35

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 2

36

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 3

37

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 4

38

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 5

39

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 6

40