If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
2 These authors contributed equally to this work. 3 Current address: Department of Veterinary Science, Faculty of Veterinary Medicine, Nippon Veterinary and Life Science University, Tokyo, Japan.
Psoriasis vulgaris is an inflammatory skin disease caused by hyperactivated T cells regulated by positive and negative mechanisms; although the former have been much studied, the latter have not. We studied the regulatory mechanism mediated by myeloid-derived suppressor cells (MDSCs) and showed that MDSCs expanded in melanoma patients express dendritic cell-associated heparan sulfate proteoglycan-dependent integrin ligand, a critical mediator of T-cell suppressor function. We examined expansion of DC-HIL+ MDSCs in psoriasis and characterized their functional properties. Frequency of DC-HIL+ monocytic MDSCs (CD14+HLA-DRno/low) in blood and skin was markedly increased in psoriatic patients versus healthy control subjects, but there was no statistically significant relationship with disease severity (based on Psoriasis Area and Severity Index score). Blood DC-HIL+ MDSC levels in untreated patients were significantly higher than in treated patients. Compared with melanoma-derived MDSCs, psoriatic MDSCs exhibited significantly reduced suppressor function and were less dependent on DC-HIL, but they were capable of inhibiting proliferation and IFN-γ and IL-17 responses of autologous T cells. Psoriatic MDSCs were functionally diverse among patients in their ability to suppress allogeneic T cells and in the use of either IL-17/arginase I or IFN-γ/inducible nitric oxide synthase axis as suppressor mechanisms. Thus, DC-HIL+ MDSCs are expanded in psoriasis patients, and their mechanistic heterogeneity and relative functional deficiency may contribute to the development of psoriasis.
Psoriasis is a common immune-mediated, chronic inflammatory skin disease characterized by hyperproliferative keratinocytes (epidermal hyperplasia or acanthosis) and extensive infiltration of various leukocytes, including T cells, dendritic cells, macrophages, and neutrophils (
). Among these leukocytes, T cells play a central role in development of these characteristic clinical features. In particular, hyperactivated T helper (Th) 1 and Th17 responses are frequently observed in the blood and skin of psoriasis patients and have been considered to be responsible for psoriatic dermatitis (
). However, the pathogenesis of psoriasis still remains ambiguous, particularly regarding mechanisms leading to persistence of T-cell hyperactivation.
T-cell activation is regulated by competing positive and negative co-regulatory signals delivered through interaction of co-regulatory receptors (expressed on T cells) and their ligands (on antigen-presenting cells and nonlymphoid cells) (
). In psoriatic patients, expression of co-stimulators is elevated significantly in hyperactivated T cells and other leukocytes compared with healthy control subjects (
). Treatment of psoriatic patients or psoriatic skin grafts in severe combined immunodeficiency mice with co-stimulator–specific inhibitors (antibodies or chemicals) reduces acanthosis and lymphocyte skin infiltrates (
), indicating that the co-stimulators are critically involved in the development of psoriatic skin. Although T-cell hyperactivation is considered to be also due to dysregulated expression of or deficiency in the function of co-inhibitors, little is known about their contribution to pathogenesis.
Myeloid-derived suppressor cells (MDSCs) were originally identified by the CD11b+Gr1+ phenotype in tumor-bearing mice (
Identification of a new subset of myeloid suppressor cells in peripheral blood of melanoma patients with modulation by a granulocyte-macrophage colony-stimulation factor-based antitumor vaccine.
), because of a lack of the Gr1 gene homologue. In healthy individuals, MDSCs consist of myeloid progenitors that differentiate into dendritic cells, granulocytes, and macrophages, so that MDSCs are a critical component in replenishing potent immune systems (
). In healthy individuals, MDSCs are weakly immunosuppressive. By contrast, in cancer patients, MDSCs fail to differentiate, thereby proliferating and mobilizing from bone marrow to other organs, where they exert potent T-cell suppression (
). Recently, it was reported that expanded MDSC populations with suppressor function are also associated with inflammatory disorders, including alopecia areata (
Increased frequency of myeloid-derived suppressor cells during Active tuberculosis and after recent mycobacterium tuberculosis infection suppresses T-cell function.
We discovered a co-inhibitory pathway comprising the DC-HIL receptor on inflammatory antigen-presenting cells and its paired receptor syndecan-4 (SD-4) on effector/memory T cells (
). Using mouse models, we showed the DC-HIL/SD-4 pathway to be a potent regulator of T cell-mediated immune responses in contact hypersensitivity, graft-versus-host disease, experimental autoimmune encephalomyelitis, and melanoma (
). Recently, we found that an expanded MDSC population in melanoma patients expressed DC-HIL, and this expression positively correlated with melanoma stage progression; DC-HIL is a critical mediator of MDSC suppressor function (
). Thus, DC-HIL may serve as a marker of the immunosuppressive capacity of MDSCs.
To explore the possible contributions of MDSC-mediated suppression to psoriasis development, we examined the expansion of DC-HIL+ MDSCs in psoriatic patients, its correlation with disease severity, and its functional properties. Data were interpreted compared with those of MDSCs from melanoma patients and healthy control subjects. Although exhibiting some similarities with their melanoma counterparts, psoriatic MDSCs had a reduced ability to suppress activation of autologous T cells and exhibited heterogeneous suppressive mechanisms among patients. Their functional deficiency and mechanistic diversity indicate an array of impaired immunosuppression that may contribute to the autoreactivity in chronic psoriasis patients.
Results
Psoriasis induces expansion of DC-HIL+ monocytic MDSCs
Because some inflammatory diseases can induce MDSC expansion as strongly as in cancer patients and because immunosuppressive MDSCs induced in melanoma patients express DC-HIL, we examined expansion of MDSCs and their DC-HIL expression in the blood of psoriatic patients versus that of healthy control subjects. We recruited patients (n = 49) and age-/sex-matched healthy donors (n = 21); demographics are summarized in Table 1. At the time of blood draw, 19 patients were untreated, and 30 were treated with various regimens (topical steroids, topical vitamin D analogues, topical calcineurin inhibitors, and/or narrow-band UVB). Blood samples were assayed by flow cytometry for frequency (%) of CD14+HLA-DRno/low monocytic MDSCs, total peripheral blood mononuclear cells (PBMCs), and DC-HIL expression on the MDSCs (Figure 1a). These MDSCs were also CD15negCD33+CD11b+ (see Supplementary Figure S1 online). MDSC levels were significantly elevated in psoriatic patients, albeit with considerably high variation, compared with heathy control subjects (median = 6.1%, range = 0.2–11.7% vs. median = 0.4%, range = 0.1–2.7%, respectively; P < 0.0001) (Figure 1b). Although MDSCs in healthy control subjects expressed DC-HIL at 8.8% (range = 1–26%, mean fluorescence intensity = 5.7 ± 5.38), most psoriatic MDSCs expressed DC-HIL (median = 81.5%, range = 2.8–100%, mean fluorescence intensity = 81 ± 93; P < 0.0001) (Figure 1c). Using these data, we calculated the percentage of DC-HIL+ MDSCs in total PBMCs and found the population to be expanded significantly in psoriatic blood (median = 4.2%, range = 0–11% vs. median = 0.1%, range = 0.01–0.27%, respectively; P < 0.0001, Figure 1d). This level was higher than in metastatic melanoma patients (mean ± standard deviation = 2.6% ± 0.6%) (
). The T-cell ligand SD-4 of DC-HIL was also up-regulated in CD4+ and CD8+ T cells of psoriatic patients (see Supplementary Figure S2 online). Thus, DC-HIL+ monocytic MDSCs were markedly proliferated in the blood of psoriatic patients.
Table 1Demographics of psoriasis patients and healthy control subjects
Figure 1Expansion of DC-HIL–expressing monocytic MDSCs in blood of psoriasis patients. PBMCs freshly isolated from blood samples of healthy donors (HD) or psoriasis patients (PSO) were analyzed for expression of HLA-DR and CD14, and CD14+HLA-DRno/low cells were gated (shown by a small window) and examined for expression of DC-HIL versus CD14. (a) Flow cytometric data shown are representative of each cohort, calculated for frequency (%) of (b) MDSCs among total PBMCs, (c) DC-HIL+ cells among MDSCs, and (d) DC-HIL+ MDSCs among PBMCs (median and range) and summarized in scatter graphs. HLA-DR, HLA-antigen D related; MDSC, myeloid-derived suppressor cell; PBMC, peripheral blood mononuclear cell.
Relationship of DC-HIL+ MDSC blood levels with psoriatic severity and treatment
Given that the percentage of DC-HIL+ MDSCs/PBMCs positively correlated with melanoma cancer stages, we examined the relationship of percentage of DC-HIL+ MDSCs/PBMCs with disease severity, assessed by Psoriasis Area and Severity Index (PASI) scores (
). Statistical analysis that included all patients showed a very weak correlation between percentage of DC-HIL+ MDSCs/PBMCs or percentage of MDSCs/PBMCs with PASI scores (Spearman r = 0.13 and 0.11, respectively; P = 0.35 and 0.31, respectively) (Figure 2a). We then sorted patients into untreated and treated groups, with variation in MDSC expansion analyzed (Figure 2b). Blood DC-HIL+ MDSC (or total MDSC) levels in the untreated group were higher than in the treated group (median = 8.2% vs. 3.3% for % MDSCs/PBMCs, P = 0.0076; median = 5.9% vs. 1.9% for % DC-HIL+ MDSCs/PBMCs, P = 0.0053) but fluctuated within a considerable range. There was no significant difference in PASI score between these two groups (P = 0.87). Thus, there was no statistical correlation of DC-HIL+ MDSC expansion with disease activity, whereas treated psoriatic patients had decreased levels of DC-HIL+ MDSCs.
Figure 2Expansion of DC-HIL+ MDSCs in untreated versus treated patients. (a) Correlations between percentage of DC-HIL+ MDSCs/PBMCs (or percentage of MDSC/PBMC) and percentage PASI score in all patients are analyzed using Spearman’s coefficient r. (b) Percentage MDSCs in PBMCs, percentage DC-HIL+ MDSCs in PBMCs, or % PASI score is compared between untreated (open circles) and treated patients (closed circles); median, range, and P-values are shown. MDSC, myeloid-derived suppressor cell; PASI, Psoriasis Area and Severity Index; PBMC, peripheral blood mononuclear cell.
Psoriatic DC-HIL+ MDSCs are T-cell suppressors but are less potent than melanoma MDSCs
We examined the ability of psoriatic MDSCs to suppress activation of autologous T cells (Figure 3a). MDSCs and T cells were isolated from PBMCs of the same patient or healthy donor. As a positive control, MDSCs from patients with stage III melanoma were also examined. T cells were co-cultured with increasing doses of purified MDSCs. Their activation was triggered by anti-CD2/CD3/CD28 co-stimulation, and IFN-γ response was measured. Whereas MDSCs from healthy donors exhibited a marginal effect on IFN-γ response, psoriatic MDSCs inhibited T-cell activation in a dose-dependent manner, with a 70–80% decline at the highest dose (1:1 cell ratio of MDSC:T cell). However, this inhibition was weaker than that seen for MDSCs from melanoma patients, which suppressed T-cell activation more strongly than psoriatic MDSCs at every dose point in co-cultures. This outcome was also true for two other pairs of psoriasis and melanoma patients. Psoriatic MDSCs also suppressed proliferation and IL-17 response of autologous T cells in a dose-dependent manner (Figure 3b and c).
Figure 3Functional characterization of psoriatic MDSCs. (a) CD14+HLA-DRneg MDSCs isolated from psoriasis patients (PSO), healthy donors (HD), or melanoma patients (Mel) were co-cultured with autologous T cells at varying cell ratios, with anti-CD2/CD3/CD28 antibody. IFN-γ secretion was measured (mean ± standard deviation, n = 3). Cultures of T cells alone measured 98 ± 7 ng/ml for HD, 45 ± 2.7 for PSO, and 58 ± 4.3 for Mel. Data shown are representative of three experiments using different patients. (b) In the same assays but with two different patients, IFN-γ and IL-17A secretion in co-cultures were measured. (c) Proliferation in cultures of T cells alone and in 1:1 co-cultures with MDSCs was analyzed by CFSE dilution assay. (d) MDSCs from three different patients were co-cultured with autologous or allogeneic T cells isolated from HD or PSO, at varying cell ratios, and IFN-γ response was measured. ∗P < 0.001. CFSE, carboxyfluorescein succinimidyl ester; MDSC, myeloid-derived suppressor cell.
Psoriatic MDSCs are heterogeneous in their suppressive mechanisms
We next characterized T cell-suppressor mechanisms of psoriatic MDSCs. To examine the influence of T-cells on MDSC effects, psoriatic MDSCs were allowed to suppress autologous or allogeneic T-cells isolated from a healthy donor. They suppressed IFN-γ response of autologous T cells but had no effect on allogeneic T cells (n = 3) (Figure 3d). We then questioned whether psoriatic T cells are susceptible to allogeneic MDSC suppressor activity. In two patients, purified MDSCs and T cells were switched and co-cultured. MDSCs from both patients were potent suppressors of autologous T cells, whereas these two patients’ MDSCs differed in their effects on allogeneic T-cells: MDSCs from patient #509 were equally potent for allogeneic/psoriatic T cells, whereas those from patient #510 were not. We performed the same crisscross experiments using four additional pairs, and data showed clear-cut results that 10 patients (in total) were clearly sorted into the two types; MDSCs from 30% of patients were insensitive to psoriatic allo-T cells, whereas those from 70% of patients were sensitive to allo-T cells (see Supplementary Figure S3 online).
To examine whether DC-HIL mediates the suppressor function of psoriatic MDSCs, anti-DC-HIL mAb or control IgG was added to co-cultures of MDSCs/T cells (Figure 4a). In a dose-dependent manner, anti-DC-HIL mAb (but not control IgG) incompletely restored T-cell IFN-γ response (up to 63%) that was suppressed by MDSCs. This moderate effect by the mAb was not consistent with data of similar assays using melanoma-derived MDSCs, in which the same mAb reversed the suppressor activity almost completely (
). DC-HIL–triggered up-regulation of nitric oxide (NO) in psoriatic MDSCs showed DC-HIL to be functional (Figure 4b): MDSCs of healthy control subjects did not produce NO (≤0.003 μmol/L), even after stimulation with anti-DC-HIL mAb. To probe mechanisms other than DC-HIL, we tested the effect of inhibitors to varying mediators on MDSC function (Figure 4c). The suppressor function of MDSCs from two patients (#507 and #510) was reversed markedly by an arginase I inhibitor and anti-IL-17A at 20 μg/ml but not by an inducible nitric oxide synthase inhibitor, anti-IL-10, or anti-IFN-γ. By contrast, MDSCs from two other patients (#509 and #261325) had a completely opposite pattern: reversal by an inducible nitric oxide synthase inhibitor and by anti-IFN-γ Ab at 20 μg/ml but no effects by others. Our finding of CCL4, IL-1β, and IL-23 gene expression in psoriatic MDSCs (Figure 4d) suggests involvement of Th17 in MDSC function. Thus, psoriatic MDSCs use heterogeneous mechanisms in exerting their suppressor function.
Figure 4Mechanisms for the T-cell suppressor function of psoriatic MDSCs. (a) Increasing doses of anti-DC-HIL mAb or control IgG were added to the co-culture of MDSC/T cell (1:1 cell ratio). IFN-γ amount is expressed as percentage relative to the T cell-alone culture (48 ng/ml). (b) Psoriatic MDSCs were cultured with immobilized anti-DC-HIL mAb or control IgG. At indicated days after culturing, cells were measured for the intracellular nitric oxide levels (mean ± standard deviation, n = 3). A second patient showed similar results. (c) The co-cultures of four different patients were added with an inhibitor to inducible nitric oxide synthase or arginase I, or anti-cytokine antibody at 2 or 20 μg/ml. (d) MDSCs isolated from PSO or HD (a pool of three patients) were assayed for messenger RNA expression of CCL4, IL-1β, and IL-23. ∗P < 0.001. Arg, arginase I; iNOS, inducible nitric oxide synthase; MDSC, myeloid-derived suppressor cell; mRNA, messenger RNA; PBS, phosphate buffered saline.
MDSCs are present among leukocytic infiltrates in lesional psoriatic skin
Because of technical difficulties in identifying CD14+HLA-DRno/low MDSCs in immunofluorescent staining of skin specimens, we chose an explant culture, in which skin punch biopsy samples from nonlesional or lesional skin of patients (who were treated with topical corticosteroids) were cultured with GM-CSF and IL-4, which protect myeloid cells from apoptotic death in culture (
) (Figure 5a). Two weeks after culture, emigrant cells from skin were harvested and examined for expression of CD14, HLA-antigen D related (HLA-DR), and DC-HIL. Most cells from nonlesional or lesional skin were CD14+ (∼80%; detailed data are summarized in Supplementary Table S1 online). Cells from nonlesional skin co-expressed HLA-DRhi and DC-HILlow (80–90%, which are likely antigen-presenting cells), whereas cells from lesional skin were HLA-DRneg and DC-HILhi. These data suggest that monocytic MDSCs are recruited into lesional (but not nonlesional) skin. Skin infiltration was supported by high expression of the skin-homing receptor cutaneous lymphocyte antigen (
Figure 5DC-HIL–expressing MDSCs infiltrate in lesional skin of psoriatic patients. (a) Punch biopsy samples from the nonlesional or lesional skin of same psoriasis patient (PSO #113-1 or PSO #114) were explant-cultured with granulocyte macrophage colony-stimulating factor GM-CSF and IL-4. Skin-emigrated cells were fluorescently stained for expression of CD14 (shown in green), HLA-DR (blue), and DC-HIL (red), and examined under confocal microscope: original magnification ×100, scale bar = 1 μm. (b) CD14+HLA-DRno/low MDSCs in PBMCs of a psoriasis patient were gated in the dot-plot of HLA-DR versus CD14 and examined by flow cytometry for expression of CD14 versus DC-HIL or CLA. Data shown are representative of four patients. CLA, cutaneous lymphocyte antigen; HLA-DR, HLA-antigen D related; MDSC, myeloid-derived suppressor cell; PBMC, peripheral blood mononuclear cell.
Co-stimulatory signals are required for the full activation of T cells and maintenance of effector T cells; they also trigger expression of co-inhibitory receptors on T cells and their ligands on antigen-presenting cells and keratinocytes. Thus, the proinflammatory environment in psoriatic skin likely induces and activates co-inhibitory pathways. Persistent hyperactivation of T cells in psoriasis could be due to repetitive and prolonged co-stimulation that overwhelms the inhibitory effects of co-inhibitors working correctly or due to a functional deficiency of co-inhibitors that permit T cells to continue as effector cells. We addressed this issue through studying the highly potent suppressor function of MDSCs. We found that MDSCs proliferated in blood of psoriatic patients to a degree greater than in melanoma patients, and that these MDSCs infiltrated psoriatic lesional skin. Most of the expanded MDSCs expressed DC-HIL and were T-cell suppressors. However, psoriatic MDSCs differ from melanoma MDSCs in 4 aspects: first, psoriatic DC-HIL+ MDSCs do not further proliferate with increasing disease severity, unlike those in melanoma, which increase in proportion to cancer stages (
); second, psoriatic DC-HIL+ MDSCs are less potent in suppressing T-cell activation compared with melanoma MDSCs; third, their suppressor function is less dependent on DC-HIL; and fourth, they are divergent in the use of inhibitory mediators among patients.
There are a few reports showing involvement of other co-inhibitors in the pathogenesis of psoriasis. Blood-circulating T cells expressing cytotoxic T-lymphocyte-associated antigen-4 (CTLA-4) or programmed cell death protein-1 (PD-1) in psoriatic patients were increased significantly compared with in control subjects, but the levels were not as high as those in T cells activated in vitro from healthy control subjects (
). Unfortunately, the functional properties were not studied. Th17 and Th1 cells in psoriasis did not efficiently induce the co-inhibitor Tim-3 upon in vitro activation, compared with those from healthy donors (
). Lastly, significant associations of polymorphisms in the CTLA4 gene with susceptibility to psoriasis were found in Japanese and Polish white patients (
Distribution of the CTLA-4 single nucleotide polymorphisms CT60G>A and +49A>G in psoriasis vulgaris patients and control individuals from a Polish Caucasian population.
Association between T-lymphocyte regulatory gene CTLA4 single nucleotide polymorphism at position 49 in exon 1 and HLA-DRB1*08 in Japanese patients with psoriasis vulgaris.
). Consistent with our findings, these reports support the possibility that deficient co-inhibitory function contributes to the etiology of psoriasis.
Like MDSCs, regulatory T cells (Tregs) are also highly potent suppressors of T-cell function and are critically involved in the development of autoimmune diseases and cancers. The Foxp3+ Treg population was increased in the circulation and lesional skin of psoriatic patients and positively correlates with disease severity (PASI score) (
). Surprisingly, psoriatic Tregs were incapable of suppressing IL-17 secretion by CD4+ T cells, although these cells inhibit proliferation and IFN-γ secretion by CD4+ T cells (
). By contrast, other reports showed no statistical difference in circulating or skin-infiltrating Tregs between psoriasis patients and healthy control subjects (
). Unlike Tregs, DC-HIL+ MDSCs were markedly increased in psoriatic patients (but showed very low levels in healthy control subjects). We also found that psoriatic MDSCs were potent suppressors of both IFN-γ and IL-17 responses. Because MDSCs can expand Tregs (
), MDSCs may be more important than Tregs in regulating psoriatic inflammation.
It is well established that cancer-induced MDSCs play a pathogenic role by blunting T-cell function. What about the role of expanded/activated MDSCs in psoriatic inflammation? In an experimental autoimmune encephalomyelitis mouse model, we showed that DC-HIL-gene–disrupted mice developed hyperactivated Th1 and Th17 responses and exacerbated autoimmune disease after immunization with myelin oligodendrocyte glycoprotein peptide (
). Such worsened disease was alleviated by adoptive transfer of DC-HIL+ MDSCs isolated from experimental autoimmune encephalomyelitis-affected wild-type mice. Similar results were found in a psoriatic mouse model (unpublished data), in which psoriasis-like acanthosis was induced by intradermal injection of IL-23 (
). These studies indicate that DC-HIL+ MDSCs protect hosts from these forms of inflammation. In contrast to the mouse model, there was no significant association of DC-HIL+ MDSC levels with disease severity (PASI score), nor with serum IL-17A and IFN-γ levels (see Supplementary Table S2 online), cytokines that positively correlate with severity (
). On the other hand, we found that psoriatic DC-HIL+ MDSCs exhibit significantly reduced suppressor capacity compared with their melanoma counterparts. However, interpretation of these data may be limited because melanoma MDSCs may be an inappropriate reference. Further studies are required to define the role of expanded MDSCs in psoriasis.
We found that psoriatic MDSCs are heterogeneous in their suppressor mechanisms in at least two ways: First, regarding suppression of allogeneic T-cell response, all psoriatic MDSCs tested (∼30 patients) exhibited suppressor effects to autologous T cells, whereas they are unable to suppress allogeneic T cells from healthy donors (n = 3). Moreover, MDSCs from some patients suppressed allogeneic/psoriatic T-cell response, but others did not. This difference may be ascribed to T cells that may have been primed for susceptibility to MDSC suppression. Second, in four patients, we found heterogeneity: MDSCs of some patients used IL-17 and arginase I. In these cases, IL-17 may up-regulate arginase I expression in MDSCs (
Interleukin-17A promotes arginase-1 production and 2,4-dinitrochlorobenzene-induced acute hyperinflammation in human papillomavirus E7 oncoprotein-expressing skin.
Role of interferon regulatory factor-1 and mitogen-activated protein kinase pathways in the induction of nitric oxide synthase-2 in retinal pigmented epithelial cells.
). It is unlikely that IL-17 is involved in selecting between the two mechanisms, because all patients produced similar levels of serum IL-17A (11.6–25.3 pg/ml). Rather, the differences may be unique to each individual patient. It would be good to know which cytokines (or soluble factors) govern the selection of critical mediators and whether the precise mediators relate to psoriasis severity. Lack of commonly shared inhibitory mediators among psoriatic patients suggests diversity in the pathogenesis of this inflammatory disorder.
Despite elevated expression of DC-HIL and the DC-HIL ligand SD-4 on MDSCs and T cells, respectively, DC-HIL was not critical to the inhibitory mechanisms. This less-critical role of MDSC in psoriasis (vs. in melanoma) may be explained by psoriatic T cells expressing high levels of IL-17 and/or IFN-γ, which are cytokines that can induce NO production (
Role of interferon regulatory factor-1 and mitogen-activated protein kinase pathways in the induction of nitric oxide synthase-2 in retinal pigmented epithelial cells.
). Such induction may compromise the critical role of the DC-HIL pathway, because we assume that its T-cell–inhibitory effect is achieved primarily through NO and reactive oxygen species production and is more potent than SD-4 signaling (
In sum, we found that blood DC-HIL+ monocytic MDSCs were expanded in psoriasis and that they infiltrated lesional skin. Although DC-HIL expression did not correlate significantly with disease severity, MDSC-mediated immunosuppression contributed to regulating hyperactivated T cells in psoriatic skin.
Materials and Methods
Subjects
We performed a cross-sectional study of 49 patients with psoriasis and 21 healthy control subjects without inflammatory skin disease. Study protocols were approved by the institutional review boards of University of Texas Southwestern Medical Center and Parkland Hospital, Dallas, Texas. Inclusion criteria included psoriasis patients or healthy control subjects older than 18 years of age, able to give written informed consent, and able to give blood and/or skin samples. Exclusion criteria included patients on subcutaneous and intravenous systemic immunosuppressant medications. Written, informed consent was obtained from all subjects. Patients were clinically evaluated for psoriasis subtype (plaque, majority of lesions plaque; guttate, majority of lesions raindrops; inverse, only involving intertriginous areas; pustular, majority of lesions erythematous patches with pustules; palmo-plantar, majority of lesions on palms and soles) and PASI score, and they completed a clinical questionnaire about demographics and current and past psoriasis treatments (Table 1). From two patients, two 4-mm punch skin biopsy samples (from lesional and nonlesional skin) were procured. Melanoma patients (n = 3) were diagnosed with cancer stage III.
Flow cytometry
Blood was collected into tubes with sodium citrate (BD Vacutainer SST, BD Biosciences, San Jose, CA). PBMCs (5 × 105 cells) isolated from the blood samples were treated with human IgG (1 μg/ml for blocking) and incubated with 10 μg/ml 3D5 mouse anti-human DC-HIL mAb (
) or isotypic control mouse IgG and 1 μg/ml phycoerythrin-anti-mouse IgG [F(ab′)2 fragment]. After washing, cells were further stained with allophycocyanin-conjugated anti-HLA-DR and FITC-anti-CD14 antibody (each 2.5 μg/ml) and analyzed by flow cytometry. Antibodies were purchased from eBiosciences (San Diego, CA) or Jackson ImmunoResearch (West Grove, PA).
T-cell suppression assays
CD14+HLA-DRneg and T cells were freshly isolated from blood samples (20–30 ml) of the same donor. For crisscross experiments to switch MDSC and T cells, two untreated patients were recruited on the same day. PBMCs were depleted of HLA-DR+ cells using biotinylated anti-HLA-DR antibody and anti-biotin microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany). The pass-through fraction was sorted into the CD14+ and CD14neg subfractions using anti-CD14 Ab-beads (Miltenyi); the former was used as MDSCs (containing >95% CD14+HLA-DRneg cells, see Supplementary Figure S1) and the latter as T-cell preparations (∼90% CD3+ cells). Autologous or allogeneic T cells (5 × 104 cells/well) were co-cultured with MDSCs at varying cell ratios with anti-CD2/CD3/CD28 beads (Miltenyi) (1.5 beads per T cell) in microculture wells (triplicate) for 5 days. This assay was repeated twice with different individuals. (All patients had >80% DC-HIL+ cells in total MDSCs and all healthy donors had ∼5%). T-cell proliferation was measured by carboxyfluorescein succinimidyl ester dilution assay. For inhibition studies, varying inhibitors were added to the same culture (1:1 cell ratio), including anti-DC-HIL mAb, anti-cytokine antibody, 0.5 mmol/L of N6-(1-iminoethyl)-L-lysine (to inducible nitric oxide synthase), and 1 mmol/L N-hydroxyl-nor-arginine (arginase I). IFN-γ and/or IL-17A amount in the culture was determined by ELISA, and suppressive activity was assessed by cytokine amount (%) relative to culture without suppressor cells. Most data shown were performed with blood of untreated patients.
) using primers for chemokine (C-C motif) ligand-4 (CCL4), 5′-TCCTCGCAACTTTGTGGTAG-3′ and 5′-TTCAGTTCCAGGTCATACACG-3′; for IL-1β, 5′-TCTACACCAATGCCCAACTC-3′ and 5′-AAGTGAGTAGGAGAGGTGAGAG-3′; and for IL-23, 5′-ATGTTCCCCATATCCAGTGTG-3′ and 5′-GCTCCCCTGTGAAAATATCCG-3′. Messenger RNA expression was shown as the expression level relative to the GAPDH gene.
NO assay
Purified MDSCs (5 × 106) were cultured in microculture wells precoated with 3D5 mAb or control IgG (10 μg/ml). After 1 or 2 days of culture, cells were measured for NO production using the Griess method (
) by replacing cytokines for T cells with those for monocytes. Lesional and nonlesional skin punch biopsy samples from same patient were ex vivo cultured with granulocyte macrophage colony-stimulating factor and IL-4 (each 10 ng/ml) for 10–14 days with refeeding cytokines every 4 days, and skin-emigrated cells were harvested and cytospun onto slide glasses, followed by immunofluorescent staining with anti-DC-HIL mAb plus Alexa546-anti-mouse IgG, Alexa350-anti-HLA-DR, and Alexa488-anti-CD14 Ab (each 10 μg/ml). Staining was analyzed under a TCS-SP1 laser scanning confocal microscope (Leica Micro-systems, Bannockburn, IL).
Statistical analysis
Statistical analyses were performed using Mann-Whitney U test or Spearman correlation and Student t test for evaluation of in vitro assays.
Conflict of Interest
The authors state no conflict of interest.
Acknowledgments
We thank Irene Dougherty and Lin-Chiang Tseng for technical assistance and Therona Ramos for administrative assistance. This research was supported by a National Institutes of Health grant (AI064927-05).
Increased frequency of myeloid-derived suppressor cells during Active tuberculosis and after recent mycobacterium tuberculosis infection suppresses T-cell function.
Role of interferon regulatory factor-1 and mitogen-activated protein kinase pathways in the induction of nitric oxide synthase-2 in retinal pigmented epithelial cells.
Identification of a new subset of myeloid suppressor cells in peripheral blood of melanoma patients with modulation by a granulocyte-macrophage colony-stimulation factor-based antitumor vaccine.
Distribution of the CTLA-4 single nucleotide polymorphisms CT60G>A and +49A>G in psoriasis vulgaris patients and control individuals from a Polish Caucasian population.
Association between T-lymphocyte regulatory gene CTLA4 single nucleotide polymorphism at position 49 in exon 1 and HLA-DRB1*08 in Japanese patients with psoriasis vulgaris.
Interleukin-17A promotes arginase-1 production and 2,4-dinitrochlorobenzene-induced acute hyperinflammation in human papillomavirus E7 oncoprotein-expressing skin.
Traditionally, myeloid-derived suppressor cells (MDSC) have been studied in regard to their increased numbers of circulating cells in cancer patients. Recent research efforts have also increased awareness of MDSC in non-malignant inflammatory diseases, including asthma, inflammatory bowel disease, and arthritis. Psoriasis can now be added to the growing list of inflammatory disorders with an MDSC component. Cao et al. report increased numbers of monocytic myeloid-derived suppressor cells (Mo-MDSC) in psoriasis patients and examine the implication of dysregulated Mo-MDSC function.
31258-1&id=gr1.jpg){kind=link}
31258-1&id=gr2.jpg){kind=link}
31258-1&id=gr3.jpg){kind=link}
31258-1&id=gr4.jpg){kind=link}
31258-1&id=gr5.jpg){kind=link}