• Open Access

Periodontal Ligament Stem Cells Regulate B Lymphocyte Function via Programmed Cell Death Protein 1§

Authors

  • Ousheng Liu,

    1. Salivary Gland Disease Center and Beijing Key Laboratory of Tooth Regeneration and Function Reconstruction, Capital Medical University School of Stomatology, Beijing, China
    2. Department of Oral and Maxillofacial Surgery, Xiangya Hospital, Central South University, Changsha, China
    Search for more papers by this author
  • Junji Xu,

    1. Salivary Gland Disease Center and Beijing Key Laboratory of Tooth Regeneration and Function Reconstruction, Capital Medical University School of Stomatology, Beijing, China
    Search for more papers by this author
  • Gang Ding,

    1. Salivary Gland Disease Center and Beijing Key Laboratory of Tooth Regeneration and Function Reconstruction, Capital Medical University School of Stomatology, Beijing, China
    Search for more papers by this author
  • Dayong Liu,

    1. Salivary Gland Disease Center and Beijing Key Laboratory of Tooth Regeneration and Function Reconstruction, Capital Medical University School of Stomatology, Beijing, China
    Search for more papers by this author
  • Zhipeng Fan,

    1. Salivary Gland Disease Center and Beijing Key Laboratory of Tooth Regeneration and Function Reconstruction, Capital Medical University School of Stomatology, Beijing, China
    Search for more papers by this author
  • Chunmei Zhang,

    1. Salivary Gland Disease Center and Beijing Key Laboratory of Tooth Regeneration and Function Reconstruction, Capital Medical University School of Stomatology, Beijing, China
    Search for more papers by this author
  • Wanjun Chen,

    1. National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, Maryland, USA
    Search for more papers by this author
  • Yaozhong Ding,

    1. Department of Immunology, Capital Medical University, Beijing, China
    Search for more papers by this author
  • Zhangui Tang,

    1. Department of Oral and Maxillofacial Surgery, Xiangya Hospital, Central South University, Changsha, China
    Search for more papers by this author
  • Songlin Wang

    Corresponding author
    1. Salivary Gland Disease Center and Beijing Key Laboratory of Tooth Regeneration and Function Reconstruction, Capital Medical University School of Stomatology, Beijing, China
    2. Department of Biochemistry and Molecular Biology, Capital Medical University School of Basic Medical Sciences, Beijing, China
    • Capital Medical University School of Stomatology, Tian Tan Xi Li No. 4, Beijing 100050, China
    Search for more papers by this author
    • Telephone: 86-10-83911708; Fax: 86-10-67062012


  • Author contributions: O.L., J.X., and S.W.: designed the study and wrote the manuscript; O.L., J.X., G.D., D.L., and S.W.: performed research and analyzed the data; Z.F.; C.Z., Y.D., and Z.T.: analyzed the data and technical support; W.C.: provided critical input on the immunological studies; S.W.: got funding and approved the manuscript. O.L. and J.X. contributed equally to this article.

  • Disclosure of potential conflicts of interest is found at the end of this article.

  • §

    first published online in STEM CELLS EXPRESS April 3, 2013.

Abstract

Periodontal ligament stem cells (PDLSCs) have provided novel cell sources for tooth and periodontal tissue regeneration. Allogeneic PDLSCs can reconstruct periodontal ligament tissue that has been damaged by periodontal diseases and regulate T-cell immunity. However, the effect of PDLSCs on B cells remains unknown. Here, we treated periodontitis in a miniature pig model using allogeneic PDLSCs and showed a reduction in humoral immunity in the animals. When cocultured with normal B cells, human PDLSCs (hPDLSCs) had similar effects as bone marrow mesenchymal stem cells in suppressing B cell proliferation, differentiation, and migration, while intriguingly, hPDLSCs increased B cell viability by secreting interleukin-6. Mechanistically, hPDLSCs suppressed B cell activation through cell-to-cell contact mostly mediated by programmed cell death protein 1 and programmed cell death 1 ligand 1. Our data revealed a previously unrecognized function of PDLSCs in regulating humoral immune responses, which may represent a novel therapeutic strategy for immune-related disorders. STEM Cells2013;31:1371–1382

INTRODUCTION

Adult stem cells exist in many tissues, including bone marrow, skin, adipose tissue, tendon, lung, heart, liver, placenta, and umbilical cord blood [1–9]. Due to their low immunogenicity and immunoregulatory function, mesenchymal stem cells (MSCs) play a key role in tissue regeneration and have been used in clinical trials in therapy for severe refractory diseases [10–12].

Periodontitis is one of the most widespread chronic infectious diseases in humans, which is the most common cause for tooth loss in adults. Several factors are known to be involved in the occurrence of periodontitis [13, 14]. Our previous study showed that periodontal ligament stem cells (PDLSCs) from periodontitis tissue had impaired immunomodulatory function, which may lead to an imbalanced immune response and the acceleration of osteoclastogenesis and inflammation related bone loss [15]. PDLSCs and other tooth-related stem cells have provided new prospects and potential therapeutic cell sources for tooth regeneration and the reconstruction of periodontal ligament tissue damaged by periodontal diseases [16–18]. However, limited cell sources of autologous dental stem cell populations impede the potential of clinical application.

Allogeneic dental stem cells significantly enlarged the source of seed cells for teeth and periodontal tissue regeneration and reconstruction development. We demonstrated previously that allogeneic PDLSCs exhibited immunosuppressive activities on activated T-cells in vitro [18]. In vivo allogeneic PDLSCs can regenerate and reconstruct periodontal ligament damaged by periodontal diseases and inactivate T-cell immunity. However, the effects of PDLSCs on B cells are unknown. Here, we investigated whether allogeneic PDLSCs could affect B-cell immune responses in vitro and in vivo. We also studied human PDLSC (hPDLSC)-mediated regulation of human B cells. We showed that allogeneic PDLSCs failed to activate humoral immunity in vivo in miniature pigs. We also demonstrated that hPDLSCs inhibit human B cell proliferation, differentiation, and chemotactic behavior. hPDLSCs secreted interleukin (IL)-6 and enhanced B-cell survival. The immunoregulatory function of hPDLSCs in human B cells was achieved by programmed death-1 (PD-1) and its ligand (PD-L1) interaction in cell-to-cell contact dependent manner.

MATERIALS AND METHODS

PDLSC-Mediated Treatment for Swine Periodontitis

Bama inbred miniature pigs (6–8 months old, weighing 10–19 kg) were obtained from the Institute of Animal Science of the Dong Bei Agriculture University, China, http://www.neau.edu.cn/english. Miniature pigs were kept under conventional conditions with free access to water and a regular supply of soft food. We generated periodontitis lesions in 12 female miniature pigs as reported previously [17], for a total of 24 defects. For the treatment study, defects were randomly assigned to four groups: control, hydroxyapatite/tricalcium phosphate (HA/TCP, 40 mg) alone, HA/TCP scaffolds plus autologous pPDLSCs (4.0 × 106 cells), and HA/TCP scaffolds plus allogeneic pig PDLSCs (pPDLSCs, 4.0 × 106 cells). All protocols for swine were approved by the Animal Care and Use Committee of Capital Medical University. Details are described in “Supporting Information Methods”.

Clinical Assessment of Periodontal Tissue Regeneration

Clinical assessments, including probing depth, gingival recession, and attachment loss, were measured on all experimental teeth before the generation of the periodontitis models (week −4), before the treatment (week 0), and 12 weeks post-transplantation. Routine blood tests (white blood cells, red blood cells, platelets, and hemoglobin concentration), routine biochemical tests (aspartate aminotransferase, alanine aminotransferase, total protein, blood urea nitrogen, and creatinine), and immunoglobulins (IgG, IgA, and IgM) in blood and gingival crevicular fluid were examined at week −4; at 1, 3, 5, and 7 days post-transplantation; and at 2, 4, 8, and 12 weeks post-transplantation. B cell-related markers (percentage of CD20+ cells, CD25+ cells, and B220+ cells) were examined at 3 and 7 days post-transplantation and at 2, 4, 8, and 12 weeks post-transplantation. A calibrated volumetric micropipette of 5 μL capacity was introduced into the periodontal pocket of the selected site for collection of gingival crevicular fluid by Brill technique. The sample was collected for 20 minutes. The collected sample was then transferred to a sterilized plastic vial with the help of air-spray. The samples of GCF were stored in plastic vials at −70°C until analyzed for antibodies level [19, 20].

Periodontal tissues were taken from autologous and allogeneic groups. In these regenerated tissues, IgG, IgA, and IgM were examined by enzyme-linked immunosorbent assay (ELISA) at 3 days and 7 days post-transplantation. In addition, real-time polymerase chain reaction assays for IL-6, PD-1, PD-L1, and programmed cell death 1 ligand 2 (PD-L2) were evaluated at 1 and 2 weeks post-transplantation, as described in “Supporting Information Methods”.

Imaging and Histological Assessments

Bone regeneration was examined by computed tomography (Siemens, Germany, http://www.usa.siemens.com/answers/en) scanning on transplantation (week 0) and at 12 weeks post-transplantation. At 12 weeks post-transplantation, all animals were killed; samples from the experimental areas were harvested, fixed with 4% formaldehyde, decalcified with 50% formic acid, and embedded in paraffin. Sections (5 μm) were deparaffinized and stained with hematoxylin and eosin. For immunohistological costaining for PDLSCs and B cells, anti-CD146 (Abcam, U.K., http://www.abcam.com) and anti-CD19 (R&D, Minneapolis, MN, US, http://www.rndsystems.com) antibodies were used.

Flow Cytometry

hPDLSCs were harvested, and aliquots (1.0 × 106 cells) were incubated for 40 minutes with specific monoclonal antibodies against STRO-1 (R&D, Minneapolis, MN, US, http://www.rndsystems.com), CD146 (Abcam, U.K., http://www.abcam.com), CD90 (Abcam, U.K., http://www.abcam.com), CD73 (BD, Biosciences, San Jose, CA, US, http://www.bdbiosciences.com), or SSEA-4 (R&D, Minneapolis, MN, US, http://www.rndsystems.com). Purified B cells were cocultured without or with hPDLSC suspensions at the indicated ratio in the presence of CpG oligodeoxynucleotide (ODN), rCD40L, anti-immunoglobulin (anti-Ig), IL-2, and IL-4 for 72 hours. Human bone mesenchymal stem cells (hBMSCs) were used as a positive control. For intracellular staining experiments, B cells were cocultured for 6 hours with or without hPDLSC suspensions at the indicated ratios in the presence of CpG ODN, rCD40L, anti-Ig, IL-2, IL-4, ionomycin (250 ng/ml), and GolgiPlug protein transport inhibitor (1 μl/ml). See details in “Supporting Information Methods”.

Human and Miniature Pig PDLSC and BMSC Cultures

All protocols for the handling of human tissue were approved by the Research Ethical Committee of Capital Medical University, China. Isolation and culture of hPDLSCs and miniature pig PDLSCs (pPDLSCs) are performed as reported previously [16, 17]. All cells used in this study were at 3–4 passages. For each experiment, we used hBMSCs (purchased from American Type Culture Collection, Manassas, VA, US, http://www.atcc.org) and pBMSCs at the same number of passages.

Differentiation of Multipotent PDLSCs

Multilineage differentiation assays toward osteogenic and adipogenic pathways were performed as reported previously [21, 22]. To detect osteogenic differentiation, calcification of the extracellular matrix was checked on day 5 by alkaline phosphatase (ALP) staining and on day 14 by von Kossa staining. Oil red O staining was used to identify lipid-laden fat cells.

B Lymphocyte Isolation and Culture

Peripheral blood mononuclear cells were isolated by Ficoll-Hypaque density gradient (Sigma, St. Louis, MO, US, http://www.sigmaaldrich.com) from the peripheral blood of six healthy donors. Briefly, cell suspensions were indirectly magnetically labeled using a cocktail of biotin-conjugated antibodies against CD2, CD14, CD16, CD36, CD43, and CD235a with anti-biotin microbeads (Miltenyi Biotech, Germany, http://www.miltenyibiotec.com). Negatively selected cells were assessed by flow cytometric analysis with a CD19 monoclonal antibody. All cell cultures were performed in RPMI 1,640 medium supplemented with 15% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 μg/ml streptomycin. Cultures were incubated at 37°C in a 5% CO2 atmosphere, in the absence or presence of the following stimuli: the CpG synthetic oligonucleotide (2.5 μg/ml; Oligonucleotide, Sangon Biotech, Shanghai, China, http://www.sangon.com), recombinant CD40L (rCD40L; 100 ng/ml; R&D, Minneapolis, MN, US, http://www.rndsystems.com), goat anti-human immunoglobulin antibodies (2 μg/ml; Immunotech, Jackson, West Grove, PA, USA, http://www.jacksonimmuno.com/home/disteuro.asp), IL-2 (50 U/ml; R&D, Minneapolis, MN, US, http://www.rndsystems.com), IL-4 (10 ng/ml, R&D), and IL-10 (10 ng/ml; R&D, Minneapolis, MN, US, http://www.rndsystems.com).

Proliferation Assay

Purified B cells were labeled with 2 μM carboxyfluorescein succinimidyl ester (CFSE; Invitrogen, Carlsbad, CA, US, http://www.invitrogen.com). Then, B cells were cocultured in flat-bottomed 6-well plates (Corning, Costar, MA, US, http://www.corning.com/lifesciences) with hPDLSC suspensions from allogeneic donors at ratios of 5:1, 1:1, and 1:5 in the presence of CpG ODN, rCD40L, anti-Ig, IL-2, and IL-4 in a total volume of 3 ml RPMI 1,640 medium per well in triplicate. hBMSCs were used as a positive control. After 3 days, B cells were collected and analyzed using flow cytometry (BD Biosciences, San Jose, CA, US, http://www.bdbiosciences.com). In another assay, B cells labeled with CFSE were centrifuged, resuspended in supernatants from hPDLSCs or hBMSCs grown to confluence, and cultured for 3 days as described.

Apoptosis Assay

The percentage of apoptotic B cells after1 and 3 days of culture with or without hPDLSCs or hBMSCs, in the presence of CpG ODN, rCD40L, anti-Ig, IL-2, or IL-4, was evaluated using the FITC Annexin V Apoptosis Detection kit I (BD Pharmingen, San Jose, CA, US, http://www.bdbiosciences.com) according to the manufacturer's instructions.

Transwell Cultures

Transwell chambers with a 0.4-μm pore size membrane (Corning, Costar, MA, US, http://www.corning.com/lifesciences) were used to physically separate the B lymphocytes from the hPDLSCs. Purified B cells (2.0 × 105 cells) were labeled with 2 μM CFSE and seeded in the upper chamber, and hPDLSCs (2.0 × 105 cells) were placed in the bottom chamber, in the presence of CpG ODN, rCD40L, anti-Ig, IL-2, and IL-4 in a total volume of 2 ml RPMI 1,640 medium per well in triplicate. hBMSCs were used as a positive control. After 3 days, B cells were collected and analyzed using flow cytometry (BD Biosciences, San Jose, CA, US, http://www.bdbiosciences.com).

Migration Assay

B-cell chemotaxis was investigated using 24-transwell plates with a 4-μm pore-size polycarbonate membrane (Corning, Costar, MA, US, http://www.corning.com/lifesciences). The migration assay was performed as reported previously [23]. Briefly, purified B cells were cocultured for 24 hours without or with hPDLSC suspensions at a ratio of 1:1 in the presence of CpG ODN, rCD40L, anti-Ig, IL-2, and IL-4. hBMSCs were used as a positive control. Next, 2.0 × 105 B cells were dispensed in the upper transwell chamber, while chemokines or medium alone were added to the lower chamber. CXCL12, CXCL13, and CCL19 were tested at 300 ng/ml, which is a concentration chosen on the grounds of dose–response experiments performed with normal tonsil B cells [23, 24]. Plates were incubated for 2 hours at 37°C. Cells that migrated into the lower chambers were harvested, stained with CD19 mAb, counted, and calculated as the percent ratio between the number of B cells dispensed in the upper chamber and the number of B cells recovered from the lower chamber after migration. The difference between input obtained following chemokine stimulation and that obtained after incubation with medium alone was statistically analyzed as net percent input.

ELISA and Cytometric Bead Array (CBA) for Antibody Detection

Purified B cells were cocultured using 12-transwell plates in a total volume of 3 ml RPMI 1,640 medium per well in triplicate, without or with hPDLSC suspensions at the indicated ratios in the presence of CpG ODN, rCD40L, anti-Ig, IL-2, IL-4, and IL-10. hBMSCs were used as a positive control. After 1 week, supernatants were collected and tested by ELISA and CBA assay (BD Biosciences, San Jose, CA, US, http://www.bdbiosciences.com). Details are described in “Supporting Information Methods”.

Measurement of Soluble Factors and Cytokines in Suspension

Purified B cells alone, hPDLSCs or hBMSCs cocultured with B cells, or hPDLSCs or hBMSCs alone were cultured in the presence of CpG ODN, rCD40L, anti-Ig, IL-2, or IL-4. After 1, 2, and 3 days, the respective supernatants were collected. ELISAs were used to quantify prostaglandin E2 (PGE2), and CBA assays were used to detect IL-2, IL-4, IL-6, IL-10, IL-12p70, tumor necrosis factor alpha (TNFα), interferon gamma (IFNγ), and tumor growth factor beta (TGFβ) with a human Th1/Th2 cytokine kit, human IL-12p70 flex set, and human TGF-β kit (BD Biosciences, San Jose, CA, US, http://www.bdbiosciences.com) according to the manufacturer's instructions.

IL-6 for Apoptosis and Proliferation

Purified B cells (2 × 105 cells/well) were cocultured with or without hPDLSCs or hBMSCs and with or without CpG ODN, rCD40L, anti-Ig, IL-2, or IL-4by adding purified NA/LE rat anti-human IL-6 antibody (BD PharMingen, San Jose, CA, US, http://www.bdbiosciences.com) to the cultures at concentrations of 0, 50, 100, or 200 ng/ml. After 3 days, B cells were collected and the percentage of apoptotic B cells was evaluated using the FITC Annexin V Apoptosis Detection kit I according to the manufacturer's instructions. For proliferation assay, CFSE-labeled B cells were collected after 3 days of culture and counted using flow cytometry.

Mechanisms of hPDLSC- and hBMSC-Mediated Immunosuppression

hPDLSCs or hBMSCs (2 × 105 cells/well) were cocultured with or without purified B cells at a ratio of 1:1 in the presence of CpG, ODN, rCD40L, anti-Ig, IL-2, and IL-4. After 3 days, hPDLSCs or hBMSCs were collected and subjected to cytofluorometric analysis using monoclonal antibodies against PD-1, PD-L1, and PD-L2 (eBioscience, San Diego, CA, US, http://www.ebioscience.com). Blocking experiments were performed by adding anti-PD-1, anti-PD-L1, or anti-PD-L2 monoclonal antibodies (eBioscience, San Diego, CA, US, http://www.ebioscience.com) at concentrations of 0, 0.12, 0.25, or 0.5 μg/ml; or anti-PD-L1 and anti-PD-L2 monoclonal antibodies (eBioscience, San Diego, CA, US, http://www.ebioscience.com) at concentrations of 0.5 μg/ml to cultures in which purified B cells had been labeled with 2 μM CFSE. After 3 days, B cells were collected and analyzed with flow cytometry (BD Biosciences, San Jose, CA, US, http://www.bdbiosciences.com).

Statistical Analysis

Statistical significance was assessed using a two-tailed Student's t test or analysis of variance; p < .05 was considered statistically significant.

RESULTS

Allogeneic pPDLSC Transplantation Suppresses Periodontitis and Humoral Immune Responses in Miniature Pig

At 12 weeks after transplantation, periodontal tissue was only partially regenerated in the control (Fig. 1A, 1E, 1I, 1M) and HA/TCP groups (Fig. 1B, 1F, 1J, 1P). Both autologous and allogeneic pPDLSC treatments significantly improved periodontal tissue regeneration compared with the HA/TCP group and control group (Fig. 1). The autologous and allogeneic pPDLSC groups did not differ significantly regarding probing depth, attachment loss, or gingival recession (Fig. 1Yi–iii). To test the immunological response induced by allogeneic pPDLSC transplantation, we analyzed B cell-associated markers (Fig. 2A–2D), immunoglobulins (Fig. 2E–2G), and routine blood and biochemical tests in whole blood (Supporting Information Table S1A–S1B), as well as immunoglobulins in gingival crevicular fluids (Fig. 2H–2J). Allogeneic and autologous PDLSC-treated groups did not differ significantly regarding the aforementioned parameters. The immunoglobulin levels in regenerated periodontal tissues (Fig. 2K, 2L) did not differ significantly between allogeneic and autologous PDLSC-treated pigs. The data suggest there were no immunological rejections in the animals that received allogeneic pPDLSC transplantation.

Figure 1.

Allogeneic miniature pig periodontal ligament stem cells (PDLSCs) regenerate periodontitis-related defects. (AD): Intraoral photographs showed that, 12 weeks post-transplantation, only limited periodontal tissues were regenerated in the control group (A) and hydroxyapatite/tricalcium phosphate (HA/βTCP) group (B). Autologous (C) and allogeneic (D) pPDLSC-mediated periodontal tissue regeneration were nearly normal. (E–H): Computed tomography images revealed limited bone formation in the control group (E) and the HA/βTCP group (F) 12 weeks after pPDLSC transplantation. Autologous pPDLSCs (G) and allogeneic pPDLSCs (H) mediated nearly complete alveolar bone regeneration. (I–L): X-Ray imaging showed that, 12 weeks after pPDLSC transplantation, autologous (K) and allogeneic (L) pPDLSCs can regenerate more alveolar bone than the control group (I) or the HA/βTCP group (J). (M–X): Hematoxylin/eosin staining in autologous (S) and allogeneic (V) pPDLSC groups revealed new periodontal tissue regeneration (T, U, W, X) in the periodontal defect area. In the control (M) and HA/βTCP groups (P), deep periodontal pockets (N, O, Q, R) and few new bone and periodontal ligament fibers remained. (Yi–iii): Clinical assessment of periodontal tissue regeneration. At week −4 and week 0, there was no significant difference among the four groups. However, 12 weeks post-transplantation, the probing depth (PD), gingival recession (GR), and attachment loss (AL) were significantly improved in autologous and allogeneic pPDLSC-treatment groups compared with the control group and HA/βTCP group. There was also a significant difference between the control group and the HA/βTCP group. *, p < .05. (M, P, S, V) Scale bar = 1 mm; (N, O, Q, R, T, U, W, X) Scale bar = 150 μm. Abbreviations: B, bone; C, cementum; CEJ, cemeto-enamel junction; D, dentin; HA/βTCP, hydroxyapatite/β tricalciumphosphate (M–V); HAB, height of alveolar bone; PDL, periodontal ligament; SE, sulcular epithelium.

Figure 2.

Dynamic evaluation of humoral immunological parameters. (A–D): There were no significant differences in the indicated time points regarding the percentage of CD20+ or CD25+ B cells in the allogeneic pig periodontal ligament stem cells (PDLSC) transplantation group at different time points. (E–G): Immunoglobulin (IgA, IgG, and IgM) in whole blood was nearly identical among the four test groups at week −4; at 1, 3, 5, 7 days, 2, 4, 8, and 12 weeks post-transplantation. (H–J): Immunoglobulin in gingival crevicular fluid was almost identical among the four test groups at week −4; at 1, 3, 5, 7 days, 2, 4, 8, and 12 weeks post-transplantation. (K–L) Immunoglobulin expression in regenerated periodontal tissues did not differ significantly between allogeneic and autologous PDLSC groups after 1 week (K) and 2 weeks (L) post-transplantation. (M): The expression of interleukin (IL)-6, programmed death (PD)-1 and PD-L1 was significantly higher in regenerated periodontal tissues of the allogeneic PDLSCs group than in the autologous PDLSCs group at 1 week and 2 weeks after transplantation, whereas PD-L2 level did not change. *, p < .05, #*, p < .01. Abbreviations: HA, hydroxyapatite; TCP, tricalcium phosphate.

Effects of hPDLSCs on B-Cell Proliferation

A cell population from the periodontal ligament of the human third molar was isolated. These cells were positive for MSC markers, including STRO-1, CD146, CD90, SSEA-4, OCT-4, and CD73, thus identifying them as hPDLSCs (Supporting Information Fig. 1A) [22]. After osteogenesis-induced culturing for 5 and 14 days, hPDLSCs were detected by ALP staining and vonKossa staining, respectively (Supporting Information Fig. 1B). hPDLSCs also possessed the potential to develop into oil red O-positive lipid-laden fat cells after 4 weeks of culture with an adipogenesis-inducing medium (Supporting Information Fig. 1B).

Positively selected B cell populations contained, on average, 93.28% B cells, as assessed by flow cytometric analysis with a CD19 monoclonal antibody (Supporting Information Fig. 1C). To examine whether hPDLSCs affect the B-cell proliferation, the hPDLSCs were cocultured with blood B cells (93.3% CD19+, Supporting Information Fig. 1C) prelabeled with CFSE and stimulated with the CpG ODN, rCD40L, anti-Ig, IL-2, and IL-4. hPDLSCs inhibited B-cell proliferation in a cell-number-dependent manner (Fig. 3A, 3B). At the highest ratio of B cells to hPDLSCs (1:5), hPDLSCs almost completely suppressed B-cell proliferation. As expected, hBMSCs also inhibited B-cell proliferation (Fig. 3A, 3B).

Figure 3.

Human periodontal ligament stem cells (hPDLSCs) inhibit B-cell proliferation and apoptosis. (A): Flow cytometric analysis confirmed that purified carboxyfluorescein succinimidyl ester (CFSE)-labeled B cells (from peripheral blood from healthy donors) were cocultured for 3 days with allogeneic hPDLSC suspensions at 5:1, 1:1, 1:1 (Transwell), and 1:5 ratios with or without CpG oligodeoxynucleotide (CpG ODN), recombinant CD40L (rCD40L), anti-Ig antibodies, interleukin (IL)-2, and IL-4. Cell proliferation was assessed by flow cytometric analysis. (B): Statistical analysis revealed that hPDLSCs inhibited dose-dependent B cell proliferation. *, p < .05, #*, p < .01, n = 5. (C): hPDLSCs and B cells were seeded 1:1 in 12-well plates on the opposite sides of a 4-μm pore-size polycarbonate membrane (Transwell) and cultured for 3 days. Aliquots of the same hPDLSC and B-cell suspensions were cocultured 1:1 for 3 days. Control B cells were cultured alone. Stimuli were CpG ODN, rCD40L, anti-Ig, IL-2, and IL-4. B-cell proliferation was assessed by CFSE. Results showed that hPDLSCs, similar to human bone mesenchymal stem cells, inhibited B cell proliferation under both cell-cell contact and transwell cultures, although the suppression was more severe under cell-to-cell contact conditions than in the transwell system. *, p < .05, #*, p < .01, n = 5. (D): B-cells cultured for 24 hours and 72 hours with or without hPDLSCs (1:1) in the presence of stimuli were double stained with propidium iodide and Annexin V and analyzed by flow cytometry. (E): Statistical analysis showed that hPDLSCs inhibited B cell apoptosis. *, p < .05, n = 5. Abbreviations: AV FITC, FITC conjugated Annexin V; CFSE, carboxyfluorescein succinimidyl ester; hBMSCs, human bone mesenchymal stem cells; hPDLSCs, human periodontal ligament stem cell; Sti, Stimulation.

To investigate the potential involvement of apoptosis in hPDLSC- or hBMSC-mediated inhibition of B-cell proliferation [18, 25], activated B cells that had been cocultured without or with hPDLSCs or hBMSCs at a 1:1 ratio for 1 and 3 days were stained with Annexin V and propidium iodide. Frequencies of Annexin V+ apoptotic B cells detected in cultures performed with hPDLSCs or hBMSCs were less than those without hPDLSCs or hBMSCs (Fig. 3D, 3E). Moreover, a decrease in the B cell number in periodontal tissues after PDLSC treatment was observed (Supporting Information Fig. 2). These findings indicate that hPDLSC-mediated inhibition of B-cell proliferation was not caused by B cell death.

hPDLSCs Inhibit B-Cell Proliferation Through Cell-to-Cell Contact and Soluble Factors

CFSE-labeled B cells were plated in the lower chamber of a transwell system, physically separated from hPDLSCs or hBMSCs (at a 1:1 ratio, in the upper chamber). For cell–cell contact experiments, B cells were cocultured with hPDLSCs or hBMSCs at a 1:1 ratio. We showed that hPDLSCs, similar to hBMSCs, inhibited B-cell proliferation under both cell–cell contact and transwell cultures (Fig. 3A, 3C), although the suppression was more severe under cell-to-cell contact conditions than in the transwell system. There were no significant changes in PGE2, IL-2, IL-4, IL-10, IL-12p70, TNFα, IFNγ, or TGFβ production observed in the supernatants of transwell system (Supporting Information Fig. 3A–3H). Furthermore, there was no difference of B cell proliferation index between cocultured with autologous PDLSCs group and allogeneic PDLSCs group (Supporting Information Fig. 3I). The data suggest that direct cell-to-cell contact contributed in large part to the mechanisms of hPDLSC-mediated suppression of B-cell proliferation.

hPDLSCs Suppress Immunoglobulins in B Cells

To investigate whether hPDLSCs and hBMSCs could inhibit B cell differentiation, B cells were cultured with or without hPDLSCs or hBMSCs at a 1:1 ratio in the presence of CpG ODN, rCD40L, anti-Ig, IL-2, IL-4, and IL-10. After 3 days, nonadherent cells were collected and stained with anti-CD138 mAb, which is an optimal marker for plasma cell detection. A significant decrease in CD138 expression was observed in B cells cocultured with hPDLSCs or hBMSCs (*, p = .003 and 0.002, respectively; Fig. 4A). After 7 days, immunoglobulins in the supernatants were determined by ELISA (Fig. 4B) and CBA assays (not shown). hPDLSCs or hBMSCs inhibited the production of IgM, IgG, and IgA at a B cell/mesenchymal cell ratio of 1:1. In contrast to their regulation of B-cell proliferation and differentiation, hPDLSCs and hBMSCs failed to affect the expression of costimulatory molecules CD40, HLA-DR, CD86, and CD80 on B cells (Fig. 4C). These results indicate that hPDLSCs or hBMSCs suppress both B cell differentiation and immunoglobulin production.

Figure 4.

Human periodontal ligament stem cells (hPDLSCs) affect B-cell chemokine receptor expression and function, immunoglobulin production, and the expression of costimulatory molecules. (A): B cells cocultured with or without hPDLSCs at a 1:1 ratio in the presence of stimuli and stained with anti-CD138 antibody. Control B cells were cultured alone without stimuli. Results are shown as a percentage of positive cells. The differences are similarly related to changes in the expression of CD138+ cells (plasmocytes) caused by coculturing with hPDLSCs or human bone mesenchymal stem cells. n = 5, *, p = .003 and 0.002, respectively. (B): B cell expression of costimulatory molecules is unaffected by incubation with hPDLSCs. Purified B cells were cultured for 72 hours with or without hPDLSCs at ratios of 1:1 and 1:5, in the presence of CpG oligodeoxynucleotide (CpG ODN), recombinant CD40L (rCD40L), anti-Ig antibodies, interleukin (IL)-2, and IL-4. After culturing, B cells were stained with anti-HLA-DR, anti-CD40, anti-CD86, and anti-CD80 antibodies in combination with CD19 antibody and analyzed by flow cytometry. Results are expressed as median percent positive cells, maximum value, and minimum value obtained from five different experiments. (C): hPDLSCs inhibit immunoglobulin production. B cells were incubated for 7 days without or with hPDLSCs at a 1:1 ratio in the presence of CpG ODN, rCD40L, anti-Ig antibodies, IL-2, IL-4, and IL-10. Quantitation of IgM, IgG, and IgA in culture supernatants was evaluated by ELISA. n = 5. *, p = 0 .036 for IgA, p = .0001 for IgG, and p = .013 for IgM. (D): Purified B cells were cultured 24 hours without or with hPDLSCs at 1:1 with CpG ODN, rCD40L, anti-Ig antibodies, IL-2, and IL-4. After culture, B cells were stained with anti-CXCR4, anti-CXCR5, or anti-CCR7 antibodies in combination with CD19 antibody, and analyzed by flow cytometry. Results are expressed as median percent positive cells obtained from five different experiments. *, p = .036 for CXCR4 and 0.035 for CXCR5 at a B-cell/hPDLSC 1:1 ratio; p = .002 for CCR7 at a 1:1 ratio. (E): Chemotaxis of activated B cells cultured with or without hPDLSCs at a 1:1 ratio in response to CXCL12, CXCL13, CCL19 (the ligands of CXCR4, CXCR5, and CCR7, respectively), or medium alone. Migrated B cells were identified by CD19 staining. n = 5. *, p = .021 for CXCL12 at the B-cell/hPDLSC 1:1 ratios, and *, p = .032 for CXCL13 at a 1:1 ratio. Abbreviations: hBMSCs, human bone mesenchymal stem cells; HLA-DR, human leukocyte antigen - DR; hPDLSCs, human periodontal ligament stem cells; Sti, Stimulation.

Effects of hPDLSCs on B Cell Chemokine Receptor Expression and Function

CXCR4, CXCR5, and CCR7, which function in the homing of B cells to secondary lymphoid organs, are important chemokine receptors expressed constitutively by B cells [26, 27]. The expression of CXCR4, CXCR5, and CCR7 in B cells significantly decreased after coculture with hPDLSCs or hBMSCs at a 1:1 ratio for 3 days in the presence of CpG ODN, rCD40L, anti-Ig, IL-2, and IL-4 (CXCR4: *, p = .036 for hPDLSC coculture, *, p = .023 for hBMSC coculture; CXCR5: *, p = .035 for hPDLSC coculture, *, p = .021 for hBMSC coculture; CCR7: *, p = .002 for hPDLSC coculture, *, p = .0023 for hBMSC coculture) (Fig. 4D). Expression of these chemokine receptors did not differ between B cells cocultured with hPDLSCs and hBMSCs (Fig. 4D).

To investigate whether these chemokine receptors could affect the chemotaxis of activated B cells in response to their ligands (CXCL12, CXCL13, and CCL19, respectively), we dispensed purified B cells that had been cocultured with or without hPDLSCs or hBMSCs for 24 hours in the upper transwell chamber, with CXCL12, CXCL13, CCL19 or medium alone added to the lower chamber. The results showed that CXCL12-, CXCL13-, and CCL19-driven B cell chemotaxis were significantly inhibited by culture with hPDLSCs or hBMSCs (Fig. 4E). These results suggest that hPDLSCs or hBMSCs could inhibit CXCR4-, CXCR5-, and CCR7-mediated B-cell chemotaxis.

B Cells Promote IL-6 Production in hPDLSCs

We next investigated whether hPDLSCs affected cytokine release in B cells. The expression of PGE2, IL-2, IL-4, IL-6, IL-10, IL-12p70, TNFα, IFNγ, and TGFβ was determined in cell supernatants in cultures containing hPDLSCs or hBMSCs alone, B cells alone, or B cells plus hPDLSCs or hBMSCs for 3 days in the presence of CpG ODN, rCD40L, anti-Ig antibodies, IL-2, and IL-4. No significant differences in PGE2, IL-2, IL-4, IL-10, IL-12p70, TNFα, IFNγ, or TGFβ production were observed in the supernatants of any cultures (Supporting Information Fig. 3A–3H). Interestingly, we observed a significant increase in IL-6 in the supernatant of B cells cocultured with hPDLSCs or hBMSCs (Fig. 5A). In addition, hPDLSCs and hBMSCs each secreted more IL-6 than B cells when they were cultured alone (Fig. 5A). Furthermore, in vivo we also detected that the expression of IL-6 mRNA was significantly higher in regenerated periodontal tissues from the allogeneic PDLSC group than the autologous PDLSC group 1 and 2 weeks post-transplantation (Fig. 2M).

Figure 5.

The mechanism of human periodontal ligament stem cells (hPDLSCs') effect on B-cells' proliferation and apoptosis. (A): The interleukin (IL)-6 concentration significantly increased with or without hPDLSCs; with or without CpG oligodeoxynucleotide (CpG ODN), recombinant CD40L (rCD40L), anti-Ig antibodies, IL-2, and IL-4; and without B cells in the Transwell culture system. However, it was less than the IL-6 concentration in cell-to-cell contact culture. n = 5, *, p < .05. (B): B cells were cocultured with or without hPDLSCs and with or without CpG ODN, rCD40L, anti-Ig antibodies, IL-2, and IL-4 in the presence of Gorky protein inhibitors. B cells were fixed, permeabilized, and stained with CD19 and anti-IL-6mAb. Results show that IL-6 was secreted by hPDLSCs (n = 5, *, p < .05). (C): The neutralizing monoclonal antibody against IL-6 failed to restore B-cell proliferation. (D): The percentage of apoptotic B cells increased in a concentration-dependent manner after adding anti-IL-6 monoclonal antibodies to the cultures at 0, 50, 100, or 200 ng/ml. (E): hPDLSCs were cocultured for 3 days without or with B cells at 1:1 without or with CpG ODN, rCD40L, anti-Ig antibodies, IL-2, and IL-4. Post-culture, hPDLSCs were stained with monoclonal antibodies against programmed death (PD)-1, PD-L1, and PD-L2. Expression of PD-1, PD-L1, and PD-L2 by human bone mesenchymal stem cells (hBMSCs) under different culture conditions was analyzed by flow cytometry. PD-1, PD-L1, and PD-L2 expression were significantly increased in hPDLSCs and hBMSCs in the coculture system (hPDLSCs: *, p = .009, 0.0003, and 0.012, respectively; hBMSCs: *, p = .0002, 0.017, and 0.048, respectively). (F): We performed blocking experiments using monoclonal antibodies against PD-1, PD-L1, and PD-L2. Anti-PD-1 antibody (0.50 μg/ml) restored approximately 90% of B-cell proliferation in coculture with hPDLSCs, and cell proliferation was rescued by anti-PD-1 in a dose-dependent fashion. The blockade of PD-L2 did not produce any significant effect at the concentration of 0.12, 0.25, and 0.5 μg/ml. There was only a little significant effect by adding anti-PD-L1 antibody at the concentration of 0.5 μg/ml on the hPDLSC-mediated inhibition of cell proliferation. Simultaneous blockade of PD-L1 and PD-L2 did not produce any significant effect at 0.12 or 0.25 μg/ml. There was a small but significant effect on the hPDLSC-mediated inhibition of cell proliferation at 0.5 μg/ml. In coculture with hBMSCs, B-cell proliferation was restored to approximately 60% of B-cell proliferation. B-cell proliferation was rescued by anti-PD-1, although not in a dose-dependent fashion; maximum rescue of B-cell proliferation was observed at 0.5 μg/ml. Although the simultaneous blockade of PD-L1 and PD-L2 partly rescued B-cell proliferation, it did not significantly affect the BMSC-mediated inhibition of cell proliferation. Abbreviations: hBMSCs, human bone mesenchymal stem cells; hPDLSCs, human Periodontal ligament stem cells; IL-6, interleukin-6; PD, programmed death; Sti, Stimulation.

To define the cellular source of IL-6, ionomycin (250 ng/ml) and GolgiPlug protein transport inhibitor (1 μl/ml) were added to the culture. We analyzed intracellular IL-6 protein with flow cytometry in hPDLSCs or hBMSCs and B cells after 6 hours of coculture. We found that the frequency of IL-6+ cells in hPDLSCs or hBMSCs cocultured with B cells was much higher than that in hPDLSCs or hBMSCs cultured alone (Fig. 5B). There were no significant differences between IL-6+ B cells cocultured with hPDLSCs or hBMSCs and B cells cultured alone (Fig. 5B). The data revealed that B cells promote IL-6 production in hPDLSCs and hBMSCs.

hPDLSC-Derived IL-6 Protects B Cells from Apoptosis

Previous studies have shown that IL-6 is a pleiotropic cytokine produced by lymphoid and nonlymphoid cells in response to several stimuli [28, 29]. The biological functions of IL-6 include the regulation of proliferation, differentiation, and activity of a wide variety of cell types [30] and participation in neuroendocrine and immune system homeostasis. In this study, we showed that the concentration of IL-6 in the supernatants of hPDLSCs or hBMSCs cocultured with B cells was more than in hPDLSCs, hBMSCs, or B cells cultured alone. Therefore, next we investigated the function of IL-6 in culture.

Neutralization of IL-6 with anti-IL-6 antibody had no significant effect on B-cell proliferation at 0, 50, 100, or 200 ng/ml (Fig. 5C). However, the percentage of apoptotic B cells increased substantially in a concentration-dependent manner after adding anti-IL-6 monoclonal antibodies (Fig. 5D). These results indicate that hPDLSCs inhibit B-cell apoptosis via IL-6.

Involvement of PD-1 in hPDLSC-Mediated Immunosuppression

PD-1, an inhibitory costimulatory molecule found on activated T-cells, is involved in the regulation of immune responses and peripheral tolerance [31–33]. We studied whether PD-1 is involved in the hPDLSC-mediated immunoregulation of B cell function. First, we detected that the expression of PD-1 and PD-L1 mRNA, but not PDL-2 mRNA, were significantly higher in regenerated periodontal tissues from the allogeneic PDLSC group than from the autologous PDLSC group 1 and 2 weeks after transplantation in vivo in miniature pigs (Fig. 2M). Next, we cultured B cells with hPDLSCs or hBMSCs together with CpG ODN, rCD40L, anti-Ig antibodies, IL-2, and IL-4. PD-1, PD-L1, and PD-L2 were significantly increased in hPDLSCs or hBMSCs cocultured with B cells at a 1:1 ratio (hPDLSCs: *, p = .009, 0.0003, and 0.012, respectively; hBMSCs: *, p = .0002, 0.017, and 0.048, respectively) (Fig. 5E).

To study the role of the PD-1 inhibitory pathway in mediating the suppression of cell proliferation by hPDLSCs or hBMSCs, we performed blocking experiments using monoclonal antibodies against PD-1, PD-L1, and PD-L2. The presence of 0.5 μg/ml anti-PD-1 antibody restored approximately 90% of B-cell proliferation in coculture with hPDLSCs. Antibodies had a dose-dependent effect on proliferation. The blockade of PD-L2 did not produce any significant effect at 0.12, 0.25, or 0.5 μg/ml. Adding anti-PD-L1 antibody (0.5 μg/ml) had only a minimal effect on the hPDLSC-mediated inhibition of B-cell proliferation. The simultaneous blockade of PD-L1 and PD-L2 (0.5 μg/ml) produced no significant effect on the hPDLSC-mediated inhibition of cell proliferation. Similar results were obtained when B cells were cocultured with hBMSCs (Fig. 5F). Altogether, these results reveal a role for the PD-1 pathway in hPDLSC-mediated immunoregulation.

DISCUSSION

In this study, we provided novel evidence that allogeneic PDLSCs failed to activate humoral immunity in vivo in miniature pigs. We also showed that hPDLSCs inhibited human B-cell proliferation, apoptosis, differentiation into antibody secreting cells, and chemotaxis in vitro.

Periodontitis is characterized by progressive destruction of periodontal supporting tissue that results from microbial plaque accumulation, and it ultimately leads to tooth loss. In this study, we observed that allogeneic PDLSCs, similar to autologous PDLSCs, could effectively repair bone defects caused by periodontitis in the miniature pig, which significantly improved periodontal tissue regeneration compared with the HA/TCP groups. There were no significant alterations of immunoglobulins or B cell-associated activation markers (CD20, CD25) [34] between allogeneic and autologous PDLSC-treated miniature pigs, suggesting lack of humoral immunological rejection in the animals receiving allogeneic PDLSC transplantation. Those results showed that allogeneic PDLSC transplantation will not cause humoral immunological response, which shed a light on further preclinical studies.

Next, we investigated underlying mechanisms of B cell tolerance to allogeneic PDLSC transplantation. We showed that hPDLSCs modulated B-cell functions, demonstrated by decreases in proliferation, apoptosis, differentiation, chemotaxis, and the expression of costimulatory molecules of B cells in vitro. After culture with hPDLSCs, IgM-, IgG-, and IgA-producing B cells were significantly suppressed. Because B cell activation is influenced by T-cells that can be inhibited by hPDLSCs [35] an indirect role through T-cell-mediated suppression of the PDLSC-mediated effect on B cells in vivo cannot be completely excluded.

hBMSCs down-regulated HLA-DR, CD80 and CD86 costimulatory molecules in human mature myeloid dendritic cells [36–38] but not in B cells [25]. In this study, we did not find hPDLSCs affect HLA-DR, CD40, CD80 and CD86 expressions on B cells, suggesting that the antigen-presenting function of B cells was unsuppressed. However, we showed that CXCR4, CXCR5, and CCR7 in B cells were down-regulated after incubation with hPDLSCs, paralleled by decreased B-cell chemotaxis in response to CXCL12, CXCL13, and CCL19, respectively.

In previous studies, hBMSC-mediated immune suppression of activated B cells has been attributed to the secretion of anti-proliferative soluble factors, such as PGE2, CCL2, CCL7, Blimp-1, and PD-1 and its ligands, as well as cell-to-cell contact [38–42], albeit not unanimously [34], and CD73, which is expressed in BMSCs, produces adenosine which can directly inhibit B cells [43, 44]. In this study, we demonstrated that hPDLSCs suppressed B-cell proliferation through PD-1 and PD-L1 interaction in vitro. This mechanism differs from that of hBMSCs, which suppressed B-cell proliferation through the interaction of PD-1 with PD-L1 and PD-L2. Although our findings indicated that hPDLSC-mediated inhibition of B-cell proliferation occurred through cell-to-cell contact, the detailed mechanisms involved require additional investigation. Similarly to observations regarding T-cells [18], B cells failed to proliferate but did not undergo apoptosis with hPDLSC treatment. On the contrary, our data showed that hPDLSCs inhibited B cell apoptosis by secreting IL-6. We also observed that the expression of IL-6, PD-1, and PD-L1 mRNAs (but not PD-L2 mRNA) was significantly higher in regenerated periodontal tissues from the allogeneic PDLSC group than the autologous PDLSC group 1–2 weeks post-transplantation. After transplantation, Th1, Th17 and the high level of IL-17, TNF-α and IFN-γ in the periodontitis tissue can increase the expression of PD-L1 on MSCs, which inhibits the differentiation of B cells via PD-1 /PD-L1 pathway [45]. All MSCs were uniformly positive for HLA-ABC and lacked the expression of HLA-DR and the costimulatory molecules (e.g., CD40, CD80, CD86, CD134, and CD252) required for full T-cell activation [46], and in our previous study we showed that hPDLSCs displayed low immunogenicity [18]. However, after transplantation, a small amount of allogeneic PDLSCs might be presented to B cells as antigen by antigen-presenting cells, and these B cells might be stimulated and express IL-6 and B7, which activated the PD-1/PD-L1 expression of MSCs in periodontal tissue. Previous studies suggest that activated human B cells can circulate throughout the body [47], so the B cells in inflamed periodontal tissue can be existed in situ or migrated from lymph nodes and other inflammatory tissues. After treatment, the allogeneic MSCs migrated followed CXCL12, CXCL13, and CCL19 density [11, 48] toward the site of inflamed area, suppressed the chemotaxis in B cells. Meanwhile, the proliferation of endogenous B cells were also inhibited through cell-to-cell contact.

The therapeutic potential of BMSCs is currently being explored in a number of phase I/II and III clinical trials [48], and a BMSCs product, Prochymal (Osiris Therapeutics Inc.), has now been used for clinical treatment. It has been reported that PDLSCs possessed greater proliferative potential than BMSCs. PDLSCs have also same potential of single-colony-strain generation, multipotential differentiation [49], and exhibit the same immune property to what cells? [18]. Moreover, PDLSCs can regenerate cementum/periodontal-ligament-like tissue [16], thus PDLSCs were more suitable for periodontal tissue regeneration.

SUMMARY

In summary, allogeneic PDLSCs can regenerate periodontal tissues in vivo without evidence of humoral immune responses. hPDLSCs interfere with human B-cell function by acting at multiple levels, including but not limited to proliferation, differentiation into antibody-producing cells, and chemotaxis. Thus, the inhibition of B-cell functions in concert with the suppression of T-cell activation further supports that the use of allogeneic hPDLSCs might be a promising approach for tooth regeneration and periodontal ligament reconstruction in patients with periodontal diseases.

Acknowledgements

This work was supported by the grant from Beijing Municipal Committee for Science and Technology (Z121100005212004; to S.W.); the National Basic Research Program of China (2010CB944801; to S.W.), the Funding Project for Academic Human Resources Development in Institutions of Higher Learning Under the Jurisdiction of Beijing Municipality (PHR20090510; to S.W.), National Natural Science Foundation of China (NSFC NO.81070799; to G.D.), and the Funding Project to Science Facility in Institutions of Higher Learning Under the jurisdiction of Beijing Municipality (PXM 2009-014226-074691; to S.W.). W.C. was supported by the Intramural Research program of the National Institute of Dental and Craniofacial Research, National Institutes of Health.

DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST

The authors indicate no potential conflicts of interest.

Ancillary