Murine Sertoli cells promote the development of tolerogenic dendritic cells: a pivotal role of galectin‐1

Summary

Sertoli cells (SCs) possess inherent immunosuppressive properties and are major contributors to the immunoprivileged status of mammalian testis. SCs have been reported to inhibit the activation of B cells, T cells and natural killer cells but not dendritic cells (DCs). Herein, we present evidence that co‐culture with SCs results in a persistent state of DC immaturity characterized by down‐regulation of the surface molecules I‐A/E, CD80, CD83, CD86, CCR7 and CD11c, as well as reduced production of pro‐inflammatory cytokines. SC‐conditioned DCs (SC‐DCs) displayed low immunogenicity and enhanced immunoregulatory functions, including the inhibition of T‐cell proliferation and the promotion of Foxp3⁺ regulatory T‐cell development. Mechanistically, the activation of p38, extracellular signal‐regulated kinase 1/2, and signal transducer and activator of transcription 3 was suppressed in SC‐DCs. More importantly, we demonstrate that galectin‐1 secreted by SCs plays a pivotal role in the differentiation of functionally tolerogenic SC‐DCs. These findings further support the role of SCs in maintaining the immunoprivileged environment of the testis and provide a novel approach to derive tolerogenic DCs, which may lead to alternative therapeutic strategies for the treatment of immunopathogenic diseases.

Abbreviations

ctr‐DCs

control dendritic cells

DCs

dendritic cells

ERK1/2

extracellular signal‐regulated kinase 1/2

GM‐CSF

granulocyte–macrophage colony‐stimulating factor

IDO

indoleamine 2,3‐dioxygenase

IFN‐γ

interferon‐γ

IL‐12

interleukin‐12

imDCs

immature dendritic cells

iNOS

inducible nitric oxide synthase

LPS

lipopolysaccharide

MAPK

mitogen‐activated protein kinase

SCs

Sertoli cells

SCCM

SC‐conditioned medium

SC‐DCs

SC‐conditioned DCs

STAT3

signal transducer and activator of transcription 3

TGF‐β₁

transforming growth factor‐β ₁

TNF‐α

tumour necrosis factor‐α

tolDCs

tolerogenic dendritic cells

Treg

regulatory T

Introduction

The mammalian testis is regarded as an immunologically privileged organ because it protects both auto‐antigenic germ cells and allografts from immunological rejection.1, 2 The blood–testis barrier formed of two adjacent Sertoli cells (SCs) plays a vital role in maintaining the immunoprivileged state of the testis.3 This immunoprotection is attributed to the intercellular tight junctions between SCs, as well as the inherent immunosuppressive properties of SCs.4 The co‐transplantation of SCs has successfully prolonged the survival of allografts or xenografts in rodent models5, 6 and in clinical trials.7, 8 Studies to date on the exact mechanism underlying the immunosuppressive effects of SCs have focused mainly on the release of immunosuppressive factors,4 Fas ligand‐induced apoptosis of activated T cells,5, 9 and the induction of functional Foxp3⁺ regulatory T (Treg) cells.10 However, whether SCs can modulate the differentiation of dendritic cells (DCs) is still unclear.

DCs are the most potent antigen‐presenting cells.11 Upon inflammation or infection, DCs acquire a mature status and migrate to secondary lymphoid organs, where they prime T cells to induce adaptive immune responses, alloantigen elimination and transplant rejection.12, 13 Numerous studies have reported that DCs can maintain an immature state under specific conditions, giving rise to so‐called regulatory or tolerogenic DCs (tolDCs).14, 15 Characterized by an immature/semi‐mature phenotype, tolDCs can efficiently promote the development of anergic T cells and Treg cells.16 Hence, the enrichment of tolerogenic DCs may provide us with an immunotherapeutic approach for the treatment of immune disorders, such as autoimmune disease and transplant rejection.17, 18, 19

The DCs isolated from normal rat testes20 and testicular draining lymph nodes21 have been found to be phenotypically immature and functionally tolerogenic, suggesting that the testicular microenvironment may promote DC differentiation into tolDCs. It is unknown, however, whether SCs are responsible for this outcome. Therefore, we tested the hypothesis that SCs mediate the immunosuppressive activity of DCs.

In the present study, we report for the first time that in vitro exposure to SCs down‐regulated the expression levels of the surface molecules I‐A/E, CD80, CD83, CD86, CCR7 and CD11c on DCs. SC exposure reduced the production of interleukin‐12p70 (IL‐12p70) and tumour necrosis factor‐α (TNF‐α) in DCs under lipopolysaccharide (LPS) stimulation. Moreover, the co‐cultured DCs (SC‐DCs) showed decreased T‐cell priming but improved suppression of T‐cell proliferation. We further demonstrate that galectin‐1 is essential for the suppressive effects of SCs on DC maturation and function. This work provides evidence that murine SCs can promote the development of tolDCs, and this effect depends mainly on the secretion of galectin‐1.

Materials and methods

Mice

BALB/c (H‐2K^d) and C57BL/6 (H‐2K^b) mice were obtained from the Experimental Animal Centre of the Fourth Military Medical University and maintained under specific pathogen‐free conditions. All animal experiments were approved by the Animal Experiment Administration Committee of the Fourth Military Medical University.

Preparation of mouse SCs

Testes isolated from 1‐ to 2‐week‐old male C57BL/6 mice were digested using 0·25% trypsin‐EDTA (Gibco, Grand Island, NY) for 15 min at 37° in the first step and transferred into Hanks’ solution containing 2 mg/ml collagenase D (Roche, Mannheim, Germany), 40 U/ml DNase I and 1 mg/ml hyaluronidase (Sigma‐Aldrich, St Louis, MO) for 20 min in the second step. Cells were cultured in complete RPMI‐1640 medium (10% fetal bovine serum) containing 100 μg/ml streptomycin and 100 U/ml penicillin (Gibco). Following 48 hr of culture, cells were subjected to hypotonic shock treatment at room temperature with 20 mm Tris–HCl buffer, pH 7·4 for 150 seconds. When SCs reached 75% confluence, cell culture medium was replaced with RPMI‐1640 medium (0·2% fetal bovine serum). After 24 hr conditioning, the supernatants were collected and centrifuged at 200 g for 15 min at 4° to remove cellular debris and subsequently used as SC‐conditioned medium (SCCM).

Generation of mouse bone marrow‐derived DCs

Bone marrow cells were collected from femurs and tibias of 5‐ to 6‐week‐old male C57BL/6 mice. After lysing contaminating erythrocytes, cells were cultured in complete RPMI‐1640 medium supplemented with 10 ng/ml granulocyte–macrophage colony‐stimulating factor (GM‐CSF) and 10 ng/ml IL‐4. On day 6 of culture, non‐adherent and loosely adherent cells were collected and purified by CD11c immunomagnetic beads (Miltenyi Biotec, Bergisch Gladbach, Germany). The magnetically sorted cells (> 90% pure) were regarded as immature DCs (imDCs). Recombinant mouse GM‐CSF and IL‐4 were purchased from PeproTech (Rocky Hill, NJ), and other recombinant proteins used for DC culture were purchased from R&D Systems (Minneapolis, MN).

Co‐culture experiments of SCs and DCs

When SCs in six‐well plates reached 75% confluence, imDCs were seeded onto the SC monolayers (2 × 10⁶ cells/well) and continued to grow for another 5 days in complete RPMI‐1640 medium supplemented with 10 ng/ml GM‐CSF and 10 ng/ml IL‐4. Then, DCs cultured on the SC monolayer were gently washed off and used as SC‐conditioned DCs (SC‐DCs). In parallel, imDCs cultured for another 5 days without SCs were used as control DCs (ctr‐DCs). In the following experiments, part of the SC‐DCs and ctr‐DCs were further stimulated with 1 μg/ml LPS (Sigma‐Aldrich) for 24 hr (named as LPS‐SC‐DC and LPS‐ctr‐DC, respectively). For the transwell system, equivalent amounts of SCs were plated in six‐well plates, and DCs were seeded in transwell chambers (Millicell, 0·4‐μm pore size; Millipore, Bedford, MA). SCCM‐conditioned DCs (SCCM‐DCs) were prepared by adding SC supernatants (half volume) to the culture medium of the imDCs.

Flow cytometric analysis

Dendritic cell phenotypic profiles were analysed by staining cells with the following antibodies: FITC‐anti‐CD11c (clone HL3), FITC‐anti‐CD86 (clone GL1), phycoerythrin (PE)‐anti‐CD83 (clone Michel‐19), PE‐anti‐I‐A/E (clone M5/114.15.2), peridinin chlorophyll protein (PerCP)‐Cy5.5 (clone 16‐10A1), PE‐anti‐CD11b (M1/70) and allophycocyanin (APC)‐anti‐CCR7 (clone 4B12; BioLegend, San Diego, CA). For the analysis of T‐cell proliferation and differentiation, cells were harvested and stained with PerCP‐Cy5.5‐anti‐CD3ε (clone 145‐2C11), PE‐anti‐CD4 (clone GK1.5), APC‐anti‐CD25 (clone PC61), AF488‐anti‐Foxp3 (clone MF‐14; BioLegend), and PE‐anti‐CD69 (clone H1.2F3; BioLegend). Cells were first gated using forward scatter and side scatter to remove debris. Ten thousand events were acquired for each sample. Anti‐mouse CD16/32 was routinely used before staining to block Fcγ receptors. Appropriate species and immunoglobulin isotype controls were used for all staining. All antibodies were from BD Pharmingen (San Diego, CA) unless otherwise specified. Flow cytometric analysis was performed on a FACSAria (BD Biosciences), and data were analysed with flowjo software (Treestar, Ashland, OR).

Immunofluorescence assay

Sertoli cells were fixed for 30 min in 4% paraformaldehyde before being permeabilized (PBS, 0·3% Triton X) for 10 min at room temperature. Preparations were blocked with 1% BSA (Millipore) in PBS for 30 min and then exposed to anti‐Müllerian hormone primary antibody (Santa Cruz Biotechnology, Santa Cruz, CA; goat anti‐mouse, monoclonal, 1 : 50 in PBS) or Sox9 primary antibody (Abcam, Cambridge, UK; goat anti‐rabbit, monoclonal, 1 : 250 in PBS) at 4° overnight. Cells were then washed with PBS, followed by incubation with secondary antibodies. Normal goat serum was substituted for the primary antibody in negative controls. AF488‐conjugated goat anti‐mouse IgG and Cy3‐conjugated goat anti‐rabbit IgG were purchased from Boster (Wuhan, China). Fluorescence micrographs were obtained using an Olympus FSX100 microscope.

Cytokine assays

Cell culture supernatants of DCs and T cells were harvested for the cytokine assays. The concentrations of IL‐10, IL‐12p70, IL‐2, interferon‐γ (IFN‐γ), transforming growth factor‐β ₁ (TGF‐β ₁) and TNF‐α were determined using ELISA kits (ELISA MAX Deluxe set; BioLegend). The cytokine profile of SCs was determined using the Mouse Cytokine Array C6 kit (Raybiotech, Redwood City, CA) according to the manufacturer's instructions. The galectin‐1 concentration in SCCM was determined by ELISA (Raybiotech).

T‐cell proliferation assays

Alloantigen‐pulsed DCs were generated by incubation of DCs (H‐2K^b) with BALB/c (H‐2K^d) splenocyte lysates (prepared by subjecting cells to three rapid freeze–thaw cycles) overnight before stimulation with or without LPS. Naive CD4⁺ T cells from C57 (H‐2K^b) spleens were sorted by negative selection using an EasySep mouse naive CD4⁺ T‐cell isolation kit (StemCell Technologies, Vancouver, Canada). After labelling with carboxyfluorescein diacetate succinimidyl ester (CFSE; BioLegend), naive CD4⁺ T cells (5 × 10⁵ cells/well) were co‐cultured with alloantigen‐pulsed DCs (1 × 10⁵ cells/well) in U‐bottom 96‐well plates in the presence or absence of plate‐bound anti‐CD3ε and soluble anti‐CD28 antibodies (BioLegend). The mixed cells were cultured for 3 days before analysis by FACS.

Mixed lymphocyte reaction

The CFSE‐labelled BALB/c (H‐2K^d) splenocytes (5 × 10⁵ cells/well) were cultured together with irradiated (20 Gy of γ‐ray) C57 (H‐2K^b) splenocytes (5 × 10⁴ cells/well) in the presence or absence of H‐2K^b‐DCs (5 × 10⁴ cells/well) in a total volume of 200 μl medium for 72 hr before analysis by FACS.

Evaluation of DC phagocytic ability

Dendritic cells in each group were incubated at 37° or at 4° as a negative control for 30 min in RPMI‐1640 medium containing 1 mg/ml FITC‐dextran (FD40S; Sigma‐Aldrich), washed twice with cold PBS, and then analysed by FACS.

Cell‐cycle analysis

Dendritic cells were collected, washed and fixed with 75% ethanol overnight at 4°. The fixed cells were incubated with PBS‐containing propidium iodide (25 μg/ml) and RNase A (50 μg/ml) for 30 min at 37°. Analysis of cells in different phases of the cell cycle was performed on a FACSAria (BD Biosciences).

Knockdown of galectin‐1

Knockdown of galectin‐1 in SCs was performed using small interfering RNAs (siRNAs) as well as a scrambled control siRNA (Santa Cruz Biotechnology) via Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to a standard procedure. After 24 hr, the transfection mixture was removed and replaced with fresh RPMI‐1640 containing 0·2% fetal bovine serum for another 24 hr conditioning.

Quantitative PCR

Total RNA was extracted from DC lysates using TRIzol reagent (TaKaRa, Shiga, Japan) and then reverse‐transcribed into cDNA using the PrimeScript RT Master Mix kit (TaKaRa). Quantitative PCR was performed in an IQ5 Real‐Time system (Bio‐Rad, Hercules, CA). Data were normalized to the reference gene GAPDH. The primers used are listed in the Supplementary material (Table S1).

Western blotting

Dendritic cells in each group were lysed in RIPA buffer following LPS stimulation at different time‐points (0, 30, 60 and 90 min). Total cell protein was separated in 10% SDS–PAGE gels and transferred onto PVDF membranes. Membranes were incubated overnight with phospho‐specific or pan antibodies against β‐actin, extracellular signal‐regulated kinase 1/2 (ERK1/2), p38, and signal transducer and activator of transcription 3 (STAT3). All primary monoclonal antibodies were purchased from Cell Signaling Technologies (Beverly, MA) and diluted at 1 : 1000. After washing with TBS‐T, membranes were incubated with horseradish peroxidase‐conjugated secondary antibodies (1 : 3000) (Boster) at 37° for 1 hr, and blots were developed with the FluorChem FC system (Alpha Innotech, San Leandro, CA).

Statistical analysis

Comparison between two groups was performed using an unpaired two‐tailed Student's t‐test. Statistical analysis among multiple groups was determined using one‐way analysis of variance and Tukey post‐tests with graphpad prism software (GraphPad, San Diego, CA). Differences with P < 0·05 were considered significant.

Results

Sertoli cells inhibit maturation of bone‐marrow‐derived DCs in vitro

The SCs derived from testes were > 98% pure as indicated by anti‐Müllerian hormone and Sox9 monoclonal antibody staining, and contained less than 1% monocytes/macrophages as assessed by FACS analysis (see Supplementary material, Fig. S1). Additionally, SC‐DCs isolated from SC monolayers were highly pure (> 95%) and consisted of less than 1% contaminating SCs as assessed by FACS analysis (see Supplementary material, Fig. S2).

The SC‐DCs showed lower expression levels of MHC class II molecules, co‐stimulatory molecules CD80, CD83 and CD86, and the integrin CD11c compared with ctr‐DCs. This profile was more noticeable after LPS stimulation (Fig. 1a). The molecule CCR7, which regulates DC migration, was significantly up‐regulated in ctr‐DCs but not in SC‐DCs when under LPS stimulation.

Figure 1.

Sertoli cells (SCs) inhibit lipopolysaccharide (LPS)‐induced dendritic cell (DC) maturation. (a) Surface phenotypes of control DCs (ctr‐DCs) and SC‐conditioned DCs (SC‐DCs) with or without LPS stimulation were analysed by FACS. Dotted lines indicate isotype controls. Numbers indicate the mean fluorescence intensity (MFI) of each DC population. The results are representative of five independent experiments. (b) Production of interleukin‐10 (IL‐10), transforming growth factor‐β ₁ (TGF‐β ₁), interleukin‐12‐p70 (IL‐12p70) and tumour necrosis factor‐α (TNF‐α) by DCs. Supernatants were collected from ctr‐DCs and SC‐DCs stimulated with or without LPS and analysed by ELISA. Data are indicated as the mean ± SEM from three independent experiments (Student's t‐test; *P < 0·05, **P < 0·01; ND, not detectable). (c) Phagocytic ability was assessed by analysing the cellular uptake of FITC‐dextran using FACS. Data in the left panel are representative of three independent experiments. Dotted lines indicate negative controls treated with FITC‐dextran at 4° for 30 min. Numbers indicate the mean fluorescence of each DC population incubated with FITC‐dextran at 37° for 30 min. Data in the right panel are expressed as the fold change of MFI ± SEM relative to the values observed for ctr‐DCs (LPS–), and the significance was calculated with one‐way analysis of variance and Tukey post‐tests (n = 3; *P < 0·05, **P < 0·01).

To further characterize SC‐DCs, we measured their cytokine levels by ELISA. In contrast to ctr‐DCs, SC‐DCs spontaneously secreted more TGF‐β ₁ but less TNF‐α. Following LPS stimulation, SC‐DC secretion of IL‐10 and TGF‐β ₁ increased by approximately twofold, but secretion of IL‐12p70 and TNF‐α was significantly lower compared with controls (Fig. 1b). This cytokine profile was further confirmed by real‐time PCR (see Supplementary material, Fig. S3). The mRNA levels of IL‐12p35, IL‐12p40, TNF‐α and IL‐6 were decreased, whereas those of IL‐10 and TGF‐β ₁ were significantly increased in SC‐DCs. Compared with controls, LPS‐SC‐DCs showed reduced mRNA levels of inducible NO synthase (iNOS) and cyclooxygenase‐2 (COX‐2), both of which are reportedly up‐regulated during LPS‐induced DC maturation.22 The gene expression of indoleamine 2,3‐dioxygenase (IDO), a key enzyme involved in nuclear factor‐κB activation and DC maturation,23 was also decreased in SC‐DCs.

As a characteristic of DC immaturity, the phagocytic ability of DCs was determined by analysing the cellular uptake of FITC‐dextran. Stimulation with LPS strongly inhibited the dextran uptake capability of both ctr‐DCs and SC‐DCs. Notably, SC‐DCs exhibited a significant increase in dextran uptake compared with ctr‐DCs both before and after LPS stimulation (Fig. 1c).

Furthermore, cell‐cycle analysis showed that the percentage of SC‐DCs in the S/G2 phases was significantly higher than that of ctr‐DCs (see Supplementary material, Fig. S4), indicating that SCs may also promote DC proliferation.

Sertoli cells suppress the T‐cell priming function of DCs

To investigate the effects of SCs on DC immunostimulatory function, different groups of DCs were pulsed with alloantigen and co‐cultured with CFSE‐labelled naive CD4⁺ T cells (ratio 1 : 5). The results revealed that naive CD4⁺ T cells could be activated by alloantigen‐pulsed ctr‐DCs (both with and without LPS stimulation), although with only minimal proliferation (Fig. 2a). Much less proliferation was observed when naive CD4⁺ T cells were stimulated by an equal amount of alloantigen‐pulsed SC‐DCs (Fig. 2a). Simultaneously, SC‐DC‐stimulated CD4⁺ T cells displayed significantly lower expression of the T‐cell activation marker CD69 (Fig. 2a,b) and an impairment in IL‐2 and IFN‐γ production compared with ctr‐DC‐stimulated CD4⁺ T cells (Fig. 2c). Unexpectedly, LPS stimulation had no significant influence on the T‐cell priming activity of both ctr‐DCs and SC‐DCs (Fig. 2a,b).

Figure 2.

Sertoli cells (SCs) exert a suppressive effect on the T‐cell priming function of dendritic cells (DCs). (a) Gating on CD3⁺ CD4⁺ cells, the proliferation rate and CD69 expression of CD4⁺ T cells were analysed by FACS. Unpulsed DC represent syngenic DCs that were not pulsed with alloantigen; dotted lines indicate naive CD4⁺ T cells cultured alone. Numbers in the upper panels represent the percentages of proliferating CD4⁺ T cells; numbers in the lower panels indicate the percentages of CD69⁺ cells. (b) Graphs compiling the frequencies of proliferating CD4⁺ T cells (left panel) and CD3⁺ CD69⁺ T cells (right panel). Data are expressed as the percentage of indicated cells ± SEM from three independent experiments (*P < 0·05, ***P < 0·001; NS, no significance). (c) The concentrations of interleukin‐2 (IL‐2) and interferon‐γ (IFN‐γ) were analysed by ELISA. The supernatants from the DC‐T‐cell co‐culture system were collected on day 3. Data are indicated as the mean ± SEM from three independent experiments (*P < 0·05, **P < 0·01, ***P < 0·001; NS, no significance).

SC‐conditioned DCs display enhanced immunosuppressive properties

To further investigate the effects of SC‐DCs on T‐cell proliferation and differentiation, we co‐cultured different groups of DCs with anti‐CD3/28 mAb‐activated CD4⁺ T cells. The activated CD4⁺ T cells cultured alone showed considerable proliferation (Fig. 3a). Interestingly, all groups of DCs in our experiments suppressed CD4⁺ T‐cell proliferation and promoted T‐cell differentiation into CD4⁺ CD25⁺ Foxp3⁺ Treg cells (Fig. 3a,b). Notably, compared with ctr‐DCs, particularly those stimulated by LPS, SC‐DCs were more potent at suppressing T‐cell proliferation and inducing Treg development (Fig. 3a,b).

Figure 3.

Sertoli cell‐conditioned dendritic cells (SC‐DCs) exhibit tolerogenic functions. (a) The proliferation and differentiation of anti‐CD3/28‐activated T cells were analysed by FACS. CFSE‐labelled CD4⁺ T cells were co‐cultured with DCs in the presence of anti‐CD3/28 antibodies. Dotted lines in the upper panels indicate naive T cells cultured alone; numbers represent the percentages of proliferated CD4⁺ T cells. Unlabeled CD4⁺ T cells under the same condition were stained with anti‐CD25 and anti‐Foxp3 antibodies and gated on CD3⁺ CD4⁺ T cells, as shown in the lower panels. Data are representative of four independent experiments. (b) Graphs compiling the frequencies of proliferating CD4⁺ T cells (left panel), the frequencies (middle panel) and absolute numbers (right panel) of CD25⁺ Foxp3⁺ T cells (right panel). Absolute numbers of CD25⁺ Foxp3⁺ T cells among all groups were calculated by multiplying the total viable leucocyte numbers per well by the frequencies of the CD25⁺ Foxp3⁺ population. Viable cell numbers were obtained using an automated cell counter (Invitrogen) with trypan blue staining for dead cells. Data are indicated as the mean ± SEM from four independent experiments (*P < 0·05, **P < 0·01, ***P < 0·001; NS, no significance). (c) Proliferation of responder cells in mixed leucocyte reaction experiments in the presence of different groups of DCs. Splenocytes of BALB/c (H‐2K^d) mice were labelled with CFSE and cultured as responder cells, while lethally irradiated C57BL/6 (H‐2K^b) splenocytes were stimulator cells (R : S = 10 : 1). Cells were gated based on the forward scatter and side scatter (lymphocyte gate). Numbers indicate the percentages of proliferated responder cells. The representative dot plots of three independent experiments are shown.

Similar trends were observed in the mixed lymphocyte reaction assay. Following stimulation of irradiated allogeneic splenocytes (H‐2K^b), responder splenocytes (H‐2K^d) showed considerable proliferation (Fig. 3c). The presence of SC‐DCs (with or without LPS stimulation) resulted in a significant inhibition of responder cell proliferation. Surprisingly, lymphocyte proliferation was also decreased when mixed splenocytes were cultured with ctr‐DCs, even those stimulated by LPS (P < 0·05, Fig. 3c).

Sertoli cells suppress the activation of p38, ERK1/2, and STAT3 in DCs through a cell‐contact‐independent mechanism

Transwell experiments were carried out to investigate whether cell contact is essential for the induction of SC‐DC immaturity. Notably, imDCs cultured with SCs in the separated transwell chambers (trans‐DCs) showed down‐regulation of CD80, CD83 and CD86, and the efficacy was comparable to that observed in the direct cell–cell contact group (SC‐DCs) (Fig. 4a), suggesting that SCs may inhibit DC maturation by releasing soluble factor(s). To elucidate the potential mechanistic basis of the inhibitory effects on DC maturation, the activities of mitogen‐activated protein kinase (MAPK) and STAT signalling pathways in SC‐DCs were examined. Phosphorylation of ERK1/2, p38 and STAT3 was significantly reduced in both SC‐DCs and trans‐DCs, before and after LPS stimulation for 30 and 60 min (Fig. 4b). Moreover, these molecules were equivalently expressed in SC‐DCs and trans‐DCs (Fig. 4b).

Figure 4.

Sertoli cells (SCs) inhibit the activation of mitogen‐activated protein kinase and signal transducer and activator of transcription 3 (STAT3) signalling pathways in dendritic cells (DCs) through soluble factor(s). (a) The surface phenotype of DCs was analysed by FACS. DCs were co‐cultured with SCs by direct contact (SC‐DCs) or in the separated transwell chambers (trans‐DCs), with or without lipopolysaccharide (LPS) stimulation. Data are expressed as the fold change of MFI ± SEM relative to the values observed for control DCs (ctr‐DCs; LPS–) (n = 3; *P < 0·05, **P < 0·01; NS, no significance). (b) Following LPS stimulation at different times (0, 30, 60 and 90 min), DCs were lysed and analysed by immunoblotting with phospho‐specific or pan antibodies against β‐actin, extracellular signal‐regulated kinase 1/2 (ERK1/2), p38 and STAT3. The lower panels show the ratios between phosphorylated target proteins and the total ones, and data are shown as the fold change relative to values obtained for ctr‐DCs (LPS–) (n = 3; *P < 0·05; NS, no significance).

Sertoli cell‐secreted galectin‐1 plays a central role in mediating SC‐DC differentiation

To investigate the mediators responsible for SC‐DC differentiation, we determined the SC cytokine profiles by performing a mouse cytokine array. The expression of multiple proteins was observed in SCCM, as shown in Fig. 5(a). We further examined the biological effects of decorin, galectin‐1, growth arrest specific gene 6 (GAS6), insulin‐like growth factor binding protein 6 (IGFBP‐6), and TNF‐related weak inducer of apoptosis receptor (TWEAK R) on the phenotype of imDCs by applying recombinant proteins. Only DCs cultured with mouse recombinant galectin‐1 showed significant down‐regulation of CD83 compared with controls (Fig. 5b). Furthermore, galectin‐1 enhanced the secretion of IL‐10 and TGF‐β ₁ but reduced the secretion of TNF‐α by imDCs in a dose‐dependent manner (Fig. 5c).

Figure 5.

Sertoli cell (SC)‐derived galectin‐1 regulates the phenotype of dendritic cells (DCs). (a) Comparison of the cytokine profiles from the SC‐conditioned medium (SCCM) and control medium using a mouse cytokine array. RPMI‐1640 medium (0·2% fetal bovine serum) was used as the control medium (upper panel). Each spot signal was corrected for the adjacent background intensity and normalized to the positive control on the membranes. The proteins which showed > 20‐fold increase in SCCM are labelled in the panels. (b) CD83 expression on imDCs after exposure to different recombinant proteins. Mouse recombinant GAS6, IGFBP‐6, decorin, TWEAK R and galectin‐1 (100 ng/ml for all) were added to the culture medium of immature DCs. Following 3 days of culture, cells were collected and analysed by FACS. Dotted lines indicate isotype controls; numbers indicate the mean fluorescence of each DC population. The results are representative of three independent experiments. (c) ELISA of cytokines in supernatants of normal DCs or galectin‐1‐treated DCs. Supernatants were collected from immature DCs that were cultured with 10 ng/ml, 100 ng/ml or 1000 ng/ml or without galectin‐1 for another 3 days. Data are indicated as the mean ± SEM from three independent experiments (*P < 0·05, **P < 0·01; NS, no significance).

To determine whether galectin‐1 plays a role during the differentiation of SC‐DCs, the galectin‐1 expression in SCs was effectively knocked down using siRNA, resulting in a significant decrease of galectin‐1 secretion by SCs (Fig. 6a). SCCM collected from control siRNA‐treated SCs maintained the ability to suppress CD83 expression on DCs, whereas the treatment with galectin‐1 siRNA effectively abrogated this effect (Fig. 6b). Knockdown of galectin‐1 also inhibited the suppressive effects of SCCM on LPS‐induced IL‐12p35 and TNF‐α expression in DCs (Fig. 6c). Most importantly, DCs cultured in conditioned medium of galectin‐1‐knockdown SCs showed impaired ability in suppressing anti‐CD3/28‐activated T‐cell proliferation (Fig. 6d).

Figure 6.

Sertoli cell (SC)‐derived galectin‐1 is the key mediator for the differentiation of co‐cultured dendritic cells (SC‐DCs). (a) Galectin‐1 concentration in the supernatants of control small interfering RNA (siRNA)‐treated SCs and galectin‐1 siRNA‐treated SCs was measured by ELISA. The supernatants were collected from transfected SCs after 24 hr of conditioning. The collected supernatants were used as control siRNA‐ SC‐conditioned medium (SCCM) and galectin‐1 siRNA‐SCCM. Data are indicated as the mean ± SEM from three independent experiments (Student's t‐test; ***P < 0·001). (b) CD83 expression on immature DCs after exposure to control siRNA‐SCCM and galectin 1‐siRNA‐SCCM. SCMM or control medium (half volume) was added to the culture medium of immatureDCs. Cells were collected and analysed by FACS following another 3 days of culture. Dotted lines indicate isotype controls; numbers indicate the mean fluorescence of each DC population. The results are representative of three independent experiments. (c) Gene expression of interleukin‐12p35 (IL‐12p35) and tumour necrosis factor‐α (TNF‐α) in DCs treated by control siRNA‐SCCM and galectin‐1 siRNA‐SCCM. The immature DCs exposed to SCCM and control medium were stimulated with 1 μg/ml lipopolysaccharide (LPS) for another 24 hr, followed by the RNA extraction and real‐time PCR. Data are expressed as the fold change of gene expression ± SEM relative to control medium‐DC group (n = 3; *P < 0·05, **P < 0·01, ***P < 0·001; NS, no significance). (d) Proliferation of anti‐CD3/28‐activated CD4⁺ T cells cultured with DCs. DCs co‐cultured in different media were collected and co‐cultured with CFSE‐labelled, anti‐CD3/28‐activated T cells (DC : T = 1 : 5). Gating on CD3⁺ CD4⁺ cells, T‐cell proliferation was measured on day 3 by FACS. Dotted lines indicate naive T cells cultured alone; numbers represent the percentages of proliferating T cells. The results are representative of three independent experiments. Data in the right panel are expressed as the percentage of proliferating T cells ± SEM (n = 3; *P < 0·05, **P < 0·01; NS, no significance). (e) DCs cultured in different media were lysed and analysed by immunoblotting with antibodies against β‐actin, and phospho‐specific antibodies against extracellular signal‐regulated kinase 1/2 (ERK1/2), p38 and signal transducer and activator of transcription 3 (STAT3). The right panel shows the ratios between phosphorylated target proteins and β‐actin relative to the values observed for control medium‐DCs (n = 3; *P < 0·05, **P < 0·01; NS, no significance).

To determine whether galectin‐1 in SC supernatants is responsible for the down‐regulation of MAPK and STAT3 signalling in SC‐DCs, the expression levels of phosphorylated ERK1/2, p38 and STAT3 in DCs cultured in different media were compared. Knockdown of galectin‐1 partially rescued the diminished phosphorylation levels of p38 and ERK1/2 but not STAT3 (Fig. 6e).

Discussion

Numerous factors, such as TGF‐β ₁, activin A, programmed death ligand‐1 (PDL‐1) and testosterone, have been shown to contribute to the immunosuppressive effects of SCs by either direct or indirect inhibition of immune response.6 SCs reportedly inhibit the activation of B cells, T cells and natural killer cells4, 24 and induce the differentiation of Foxp3⁺ Treg cells,5, 10 but it remains unclear whether SCs regulate the differentiation of antigen‐presenting cells. As an extension of a previous study showing that DCs isolated from normal testes display tolerogenic characteristics,20 we tested the hypothesis that SCs regulate the differentiation and functionality of DCs.

In the current study, we demonstrated that in vitro exposure to SCs decreased the levels of cell surface molecules, including I‐A/E, CD80, CD83, CD86, CCR7 and CD11c, on DCs. Along with a reduced production of pro‐inflammatory cytokines and increased phagocytic ability, the phenotype of SC‐DCs indicates that exposure to SCs leads to a persistent state of DC immaturity, which is a prerequisite for immunosuppressive functionality.25 Additionally, SC‐DCs revealed enhanced production of TGF‐β ₁ and IL‐10, both of which reportedly play a crucial role in immune tolerance induction.26, 27 These profiles are consistent with those of reported tolDCs induced by various agents,16 suggesting that SCs may induce the tolerogenic phenotype of DCs.

However, unlike many types of reported tolDCs, SC‐DCs showed significant down‐regulation of IDO and iNOS, both of which reportedly serve as regulatory molecules by inhibiting T‐cell proliferation and altering their differentiation.28, 29 In contrast, others reported that LPS‐induced maturation leads to up‐regulation of IDO, iNOS and COX‐2 in DCs,22 which confirms the inhibitory effect of SCs in LPS‐induced DC maturation and further suggests that DCs may differentiate into a new population different from the current reported ones in the presence of SCs. However, the mechanism by which these genes are regulated still remains unclear.

Tolerogenicity of DCs is characterized by the loss of the ability to prime naive T cells, the inhibition of T‐cell proliferation, and the promotion of Treg cell development.14 Indeed, SC‐DCs revealed almost all of these characteristics in our study. Naive T cells co‐cultured with alloantigen‐pulsed SC‐DCs displayed decreased proliferation, lower expression of CD69, and reduced IL‐2 and IFN‐γ production compared with control groups. SC‐DCs suppressed the proliferation of anti‐CD3/28 monoclonal antibody‐activated T cells and promoted their differentiation into CD4⁺ CD25⁺ Foxp3⁺ Treg cells to a larger extent than ctr‐DCs. The phenotypically immature DCs differentiated in the presence of SCs are evidently functionally tolerogenic.

Contrary to our expectation, ctr‐DCs also displayed suppressor functions even after LPS stimulation, which appears to contradict the consensus that mature DCs are highly immunogenic and promote T‐cell immunity.30 In fact, numerous studies have challenged the misconception that DCs expressing maturation markers always promote T‐cell immunity.31 Zhou et al. reported that LPS‐stimulated bone marrow‐derived DCs could inhibit the development of experimental autoimmune encephalomyelitis owing to immune tolerance mediated by LPS‐induced apoptotic DCs.32 Others reported that repetitive transfer of DCs matured with TNF‐α blocks the development of experimental autoimmune encephalomyelitis through the induction of antigen‐specific immune tolerance.33 Consistent with these findings, our data also indicate that mature DCs can be immunosuppressive under specific conditions. Factors such as LPS/TNF‐α‐induced DC apoptosis,32 the extent of DC maturity34 and the presence of TGF‐β ₁ 26 have been found to switch DC immunogenicity to tolerogenicity. Evidently, the relationship between DC maturity and function is extremely complex, with many details that remain to be elucidated.

Galectin‐1 is a highly conserved pro‐apoptotic β‐galactoside‐binding protein that exerts immunomodulatory effects on experimental models of autoimmunity.35 Galectin‐1 can induce activated T‐cell apoptosis36 and promote the development of tolDCs.37 A recent study has shown that galectin‐1 produced by Sertoli, Leydig and germ cells is a critical regulator of the testicular immune microenvironment.38 Consistent with these findings, we observed the production of galectin‐1 by SCs cultured in vitro.

Galectin‐1‐exposed imDCs displayed less CD83 surface expression and secreted more TGF‐β ₁ but less TNF‐α, a profile resembling that of SC‐exposed imDCs. Inhibition of SC galectin‐1 secretion by siRNA knockdown blocked SCCM‐induced development of immunosuppressive DCs. Thus, galectin‐1 may play a key role in the differentiation of functionally tolerogenic SC‐DCs.

As reported, galectin‐1 can endow DCs with tolerogenic potential involving the activation of STAT3 and increased secretion of immunosuppressive IL‐10.39, 40 STAT3 negatively regulates DC maturation and function, and its expression is elevated during the differentiation of several types of tolDCs.41, 42 Interleukin‐10 is an anti‐inflammatory cytokine that suppresses effector T‐cell function and promotes the development of Treg cells.43, 44

In our study, exposure of imDCs to recombinant galectin‐1 (1 μg/ml) caused an increase in IL‐10 secretion. However, exposure to SCs failed to promote IL‐10 secretion by imDCs, although IL‐10 mRNA did increase compared to ctr‐DCs. SC exposure also resulted in lower phosphorylation of STAT3 instead of its activation in DCs. These apparent discrepancies might be explained by the differences in the culture condition, as well as the source and amount of galectin‐1. In addition, SC‐secreted factors affecting DC differentiation are more complex than anticipated. Knockdown of galectin‐1 did not reverse SCCM‐induced suppression of STAT3 phosphorylation in DCs, indicating that the down‐regulation of STAT3 is not solely mediated by galectin‐1. There may be other mediators that regulate DC differentiation in SC supernatants, which remain to be identified.

The activation of the MAPK signalling pathway reportedly promotes the maturation and immunostimulatory abilities of DCs.45 The impairment of DC phenotype and function is accompanied by down‐regulation of MAPKs in several studies.46, 47 SC‐conditioned DCs showed decreased phosphorylation of p38 and ERK1/2, which was partially dependent on galectin‐1. Others, however, have demonstrated that galectin‐1 treatment leads to the activation of ERK1/2 in monocytes.48, 49 This discrepancy might have to do with experimental variation, such as differences in cell types, the timing of galectin‐1 treatment, or the timing of analysis.

Considering their tolerogenicity, galectin‐1‐induced tolDCs may possess therapeutic potential to ameliorate immune disorders, such as autoimmune disease and transplant rejection. However, before their applications for cell therapy, the functionality and stability of galectin‐1‐induced tolDCs in vivo need to be validated. Galectin‐1 may also possess therapeutic potential to treat immunopathogenic diseases, considering its immunosuppressive effects in vitro 39 and in vivo.50, 51 Galectin‐1 unfortunately also plays a critical role in tumour immune escape, leading to tumour progression and poor outcome.52, 53 Therefore, further study is needed to demonstrate the efficacy and safety of galectin‐1 application.

Our research reveals the ability of SCs to alter the development of DCs towards a phenotypically immature subset with tolerogenic functions via secretion of galectin‐1. This study contributes to a better understanding of the mechanisms underlying SC immunosuppressive effects and provides a novel approach for the generation of tolDCs, which may shed more light on the discovery of new therapeutic strategies for the cell‐based treatment of immune disorders.

Disclosures

The authors declare no conflict of interest.

Supporting information

Figure S1. Pure cultures and characterization of murine Sertoli cells.

Figure S2. The purity of Sertoli cell‐conditioned dendritic cells (SC‐DCs).

Figure S3. The transcription levels of the indicated genes.

Figure S4. Cell cycle analysis.

Table S1. Primer pairs used for quantitative PCR in this study.