Targeting Myeloid-Derived Suppressor Cells Is a Novel Strategy for Anti-Psoriasis Therapy

Psoriasis is a common immune-mediated, chronic inflammatory genetic-related disease that affects patients' quality of life. Myeloid-derived suppressor cells (MDSCs) are a heterogeneous population of progenitor and immature myeloid cells which are expanded in psoriatic skin lesions and peripheral blood. However, the role of MDSCs in the pathogenesis of psoriasis remains unclear. Here, we confirmed that the accumulation of human MDSCs is remarkably increased in skin lesions of psoriasis patients by flow cytometry. Depleting MDSCs by Gemcitabine significantly suppresses IMQ-induced psoriatic inflammation and epidermal thickening as well as Th17 and Treg cell accumulation. Moreover, through the RNA-Seq technique, we validated some differentially expressed genes on CD4+ T-cells of IMQ-induced-MDSC-depleted mice such as IL-21 and Timd2, which are involved in Th17-cell differentiation or T-cell activation. Interestingly, neutralizing IL-21R by antibody reduces IMQ-induced epidermal thickening through downregulating the infiltration of MDSCs and Th17 cells. Our data suggest that targeting myeloid-derived suppressor cells is a novel strategy for antipsoriasis therapy. IL-21 may be a potential therapeutic target in psoriasis.


Introduction
Psoriasis is a common immune-mediated, chronic inflammatory skin disease, which has been characterized by epidermal acanthosis, hyperkeratosis, parakeratosis, and extensive inflammatory cell infiltration including T-lymphocytes, macrophages, mast cells, and neutrophils [1]. Accumulating evidence showed that the psoriatic keratinocytes (KCs) not only have been shown uncontrolled proliferation but also respond to cytokines such as IL-22 or IL-17A/IL-17F released from Th17 or Th22 cells, which facilitate the secretion of proinflammatory factors such as AMP activating dendritic cells to initiate specific T-cell-related immune responses [1,2]. More importantly, psoriatic KCs recruit immune cells into psoriatic skin lesions through the production of chemokines or cytokines including myeloid-derived suppressor cells (MDSCs) [3][4][5][6].
In normal conditions, MDSCs have differentiated into mature granulocytes, macrophages, or dendritic cells (DCs) in bone marrow [9]. However, under pathological conditions such as cancer, chronic inflammatory diseases, and immune diseases, those undifferentiated immature myeloid cells have been recruited and infiltrated into the specific organ from bone marrow [7]. Although MDSCs have been shown a remarkable ability to suppress T-cell responses in cancer, it becomes more heterogeneous and complicated in autoimmune diseases. Recent studies revealed that expanded MDSCs induce immune responses in systemic lupus erythematosus (SLE), autoimmune arthritis (RA), and autoimmune encephalomyelitis [12][13][14][15]. Interestingly, studies showed that the population of MDSCs has been expanded in psoriasis patients, which produce cytokines including IL-23, IL-1β, and CCL4 [16][17][18]. Moreover, MDSCs isolated from psoriasis patients fail to suppress T-cell activation and express reduced programmed cell death protein-1 (PD-1), as a consequence of losing the ability to induce regulatory T-cell conversion compared with those cells from healthy controls or melanoma patients [16,19], indicating MDSCs showed great heterogeneity under different pathogenesis.
In this study, we aim to investigate the proinflammatory roles of MDSCs in the pathogenesis of psoriasis. We found it is a novel strategy to target myeloid-derived suppressor cells for antipsoriasis therapy.

Materials and Methods
2.1. Human Skin Samples. This study was reviewed and approved by the local ethics Institutional Review Board (IRB) (Xiangya Hospital, Central South University, IRB-201512526). All experiments were conducted in accordance with the Declaration of Helsinki Principles. We performed a cross-sectional study of 27 patients with psoriasis and 17 healthy control subjects without inflammatory skin disease. 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 skin samples. Exclusion criteria included patients on subcutaneous and intravenous systemic immunosuppressant medications. Patients were clinically evaluated for psoriasis subtype and PASI score.

IMQ-Induced
Psoriasis-Like Skin Inflammation. Six-to eight-week-old mice were treated with daily topical doses of 62.5 mg of IMQ cream (5%, 3.125 mg of the active compound; Aldara, 3M Pharmaceuticals), which was applied to their shaved backs for 6 consecutive days. A scoring system based on the clinical Psoriasis Area and Severity Index (PASI) was used to evaluate the skin inflammation on the skin lesions of mice. Briefly, erythema, scale, and infiltration were graded on a scale from 0 to 4 as follows: 0, none; 1, slight; 2, moderate; 3, marked; and 4, very marked. The level of erythema was scored using a table with red taints. The cumulative score served as a measure of inflammation severity (scale: 0-12) [20]. The animal study protocol was approved by the Ethics Committee of Xiangya Hospital (Central South University, China, #2015110134).

Tissue
Processing. Skin lesions of psoriasis patient or mice were cut into small pieces and digested in 5 ml PBS containing 2 mg/ml collagenase type IV and 1 mg/ml dispase II (both Sigma-Aldrich, USA) while shaking at 37°C for 150 minutes. Enzyme activity was stopped using 10% FBS medium. The tissue was further homogenized with a syringe and filtered through a 40 μm cell strainer. The cell strainer was washed with 20 ml PBS followed by centrifugation (500 x g at 4°C for 10 min). Single-cell suspensions from the spleens were obtained by mashing the spleens through 40 μm cell strainers. The cell strainer was washed with 20 ml PBS followed by centrifugation (500 x g at 4°C for 5 min) and then split red blood cells by means of lysing solution (BD Pharm Lyse™, USA). Single cells were then stained with fluorescence antibodies for flow cytometry.
2.5. Flow Cytometry. All utilized antibodies are summarized in Supplementary Table S2. Firstly, Zombie Aqua™ Fixable Viability Dye was used for selecting living cells. Then, TruStain fcX anti-mouse CD16/32 was used to block Fc receptor on the immune cells of mice. For surface staining, single cells isolated from the skin or the spleens were incubated with antibodies at 4°C for 30 min, followed by washing and centrifugation (500 x g at 4°C for 5 min). For intracellular cytokine staining (Th17), cells were restimulated in 100 μl RPMI supplemented with GolgiPlug (1 : 1000, BD), PMA (50 ng/ml, AppliChem), and ionomycin (750 ng/ml, Invitrogen) for 4 to 6 hours at 37°C. After surface staining, cells were permeabilized and fixed in 250 μl BD Cytofix/Cytoperm™ according to the manufacturer's instructions. Then, the cells were washed with permeabilization buffer and stained intracellularly at 4°C for 30 min in the permeabilization buffer. For intranuclear staining (Tregs), after surface staining, cells were fixed and permeabilized using the eBioscience Foxp3/transcription factor fixation/permeabilization concentrate and diluent 2 Mediators of Inflammation from ThermoFisher followed by incubation with antimouse/rat Foxp3 antibodies at room temperature for 40 min according to the manufacturer's instructions. To better distinguish the border between positive and negative subsets, we set FMO-controls for markers including IL-17A, IFN-γ, CD25, and Foxp3. The acquisition was performed with FACS Canto II (BD Biosciences). Flow cytometric analysis on live, single cells was performed using FlowJo (Tree Star) software.

Quantitative RT-PCR (qRT-PCR).
Total RNA was extracted with Trizol (Invitrogen), and cDNA was synthesized via reverse transcription using a HiScript Q RT Kit (Vazyme) (R123-01). qRT-PCR was performed using an UltraSYBR Mixture with ROX (CWBio, Beijing, China) according to the manufacturer's instructions on a Quant-Studio 3 RT-PCR instrument (ThermoFisher, USA). The reaction mixture contained 0.5 ml of forward and reverse mouse primers, as described in Supplementary Table S1. Values were normalized to Gapdh. All reactions were conducted in triplicate across. Relative quantification was performed using the ΔΔCT method, and the results were expressed in a linear form using the formula 2 −ΔΔCT .

Cell
Sorting for RNA Sequencing. Splenic cells were isolated from the freshly obtained spleen of mice. CD4 + T-cells were positively selected from splenic cells using magnetic CD4 microbeads (Miltenyi Biotech, San Diego, CA) with a magnet according to the manufacturer's instructions. The purity of the CD4 + T-cells after sorting was >95%. The cDNA library construction, library purification, and transcriptome sequencing were implemented according to the Shanghai Genergy Biotechnology Sequencing Company's instructions.

The Accumulation of Human MDSCs Is Remarkably
Increased in Skin Lesions of Psoriasis Patients. Recently, the accumulation of MDSCs has been observed in the peripheral blood or spleen of murine models in autoimmune disorders such as SLE and RA, which are positively related to disease severity [12,13,15] and the number of MDSCs has been found expanded in psoriasis patients [16,19,21]. To study the relationship between psoriasis and MDSCs, we analyzed the population of MDSCs in skin lesions of psoriasis patients    Mediators of Inflammation by flow cytometry. The human MDSCs have been identified with CD11b + CD33 + HLA-DR - [7,10]. The details of patients for subjects participating in this study are shown in Table 1. We found that the accumulation of human MDSCs (CD11b + CD33 + HLA-DR -) is remarkably increased in psoriatic skin lesions compared with healthy controls (Figure 1), indicating there is a correlation between psoriasis and the accumulation of MDSCs, to some extent.

MDSC Inhibitor (Gemcitabine) Significantly
Attenuates IMQ-Induced Psoriasis-Like Skin Inflammation through Downregulating Th17 and Treg Cells. Although the number of MDSCs has been found elevated in both skin lesions and peripheral blood, the effect of MDSCs on the pathogenesis of psoriasis remains to be elucidated. Gemcitabine (GEM) is well known to be an inhibitor of MDSCs, which reduces the accumulation of MDSCs with no significant influence on other immune cells such as T, B cells, NK cells, and macrophages [22]. Therefore, we treated mice with GEM to study the relationship between MDSCs and psoriasis. The specific drug use scheme is shown in Figure 2(a). The murine MDSCs have been characterized by CD11b + and Gr-1 + [7,10], and we found that IMQ treatment significantly induces psoriasis-like skin inflammation as well as the accumulation of MDSCs in spleen and skin lesions (Figures 2(b) and  3(a)). As expected, GEM treatment significantly reduces IMQ-induced accumulation of MDSCs in skin lesions and spleen (Figure 3(a)), therefore alleviating the phenotype of IMQ-induced psoriasis-like skin inflammation (Figure 2(b)) based on the Psoriasis Area and Severity Index (PASI) score (Figure 2(c)). In addition, GEM treatment markedly decreases IMQ-induced epidermal thickening and inhibited splenomegaly compared with the vehicle on day 6 after IMQ application for 5 consecutive days topically (Figures 2(b) and 2(d)). Moreover, GEM treatment remarkably decreases IMQ-mediated infiltration of Th17 and Treg cells in the spleen (Figures 3(b) and 3(c)), indicating depletion of MDSCs by GEM abrogates IMQ-induced psoriasislike skin inflammation such as erythema, skin thickening, scaling, and the infiltration of Th17 and Treg cells.

Neutralizing IL-21R In Vivo Inhibits IMQ-Induced
Epidermal Thickening, Cutaneous MDSC Infiltration, and Splenic Th17 Infiltration. Evidence revealed that IL-21 is   Statistical analysis data is shown in (B). * P < 0:05, * * P < 0:01, * * * P < 0:001, and * * * * P < 0:0001; ns: not significant. One-way ANOVA with Dunnett's post hoc test was used.  7 Mediators of Inflammation highly expressed in the psoriatic skin lesions, which stimulates the proliferation of keratinocytes [31]. Moreover, IL-21 is well known to be related to immune diseases and regulates the differentiation of CD4 + T-cells [32]. IL-21R, a receptor for IL-21, is a class I cytokine heterodimeric receptor, which mainly expressed on lymphoid cells such as circulating T-cells, B cells, NK cells, and nonlymphocytic cells and tissues including keratinocytes [31]. To verify the role of IL-21 in the progression of psoriasis, we administrated the anti-IL-21R antibody to neutralize the IL-21 signaling pathway through IMQ-induced psoriasis-like BALB/c mouse models. The experimental design scheme is shown in Figure 5   . The results were normalized to Gapdh. * P < 0:05, * * P < 0:01, * * * P < 0:001, and * * * * P < 0:0001; ns, not significant. One-way ANOVA with Dunnett's post hoc test was used. 8 Mediators of Inflammation    Figure 5(b)). Moreover, neutralizing IL-21R with anti-mouse IL-21R antibody significantly reduces IMQ-mediated accumulation of MDSCs in skin lesions and splenic Th17 cells ( Figure 5(c)), indicating targeting IL-21 is a therapeutic approach for psoriasis.

Discussion
Psoriasis has been documented to be a T-cell-mediated chronic inflammatory disease [2,33]. The IL-23/IL-17A-Th17 axis has a crucial role in the development of psoriasis [2,34,35]. IL-23, secreted by DCs or KCs, facilitates Th17 differentiation which produces proinflammatory cytokines including IL-17A, IL-17F, IL-6, IL-21, and IL-22, resulting in the infiltration of Th17 and high levels of Th17-mediated proinflammatory cytokines in skin lesions and peripheral blood of psoriasis patients [2,36,37]. The Treg cells, constitutively expressing Foxp3 (the master transcriptional factor of Treg cells), are believed to maintain immune homeostasis through suppressing the function of other lymphocytes such as Th1, Th2, and Th17, resulting in inhibition of immune and inflammatory responses [38][39][40].
Although the role of Treg cells in psoriasis has not been fully elucidated, studies showed that numbers of Treg cells are upregulated in psoriatic skin lesions [41][42][43][44] or peripheral blood [39,43,44] of psoriasis patients or murine models [45]. In addition, evidence has indicated that Foxp3 + Treg cells can converse into inflammation-associated Th17 cells under proinflammatory conditions both in psoriasis [18,46,47] and in rheumatoid arthritis (RA) [48]. And there is a positive correlation between Treg cells and Th17 cells in psoriasis [43]. Moreover, accumulating studies demonstrated that the polarization of Th17 cells has been related to the induction of Foxp3 + Treg cells [18,46,49].
MDSCs are known to be a heterogeneous population of progenitor and immature myeloid cells derived from different stages and have essential roles for regulating the function of Th17 and Treg cells. The expanded MDSCs enhance the differentiation of naive CD4 + T-cell precursors into Th17 cells and are positively correlated with disease severity of SLE and RA patients as well as their murine models [12][13][14][15]. Our results showed that GEM, an MDSC inhibitor, inhibits IMQ-induced epidermal thickening and the accumulation of Th17, Treg cells, and MDSCs ( Figure 3). Furthermore, we investigated the effect of depleting MDSCs by GEM treatment on gene expression profiles of CD4 + Tcells and the results exhibited that IMQ-induced IL-21 expression has been dramatically suppressed by GEM treatment (Figures 4(a) and 4(d)). IL-21 is highly expressed in skin lesions and peripheral blood of psoriasis patients, which is required for epidermal hyperplasia and Th17-cell polarization [23,24,31,50]. And it was reported that IL-21 promotes psoriatic inflammation by inducing an imbalance of Th17 and Treg cells [47]. Consistent with those results, neutralizing IL-21R by its antibody abrogates IMQ-induced epidermal thickening, MDSC migration, and Th17 infiltration ( Figure 5), indicating IL-21 may be a potential therapeutic target for psoriasis treatment.
Still, there are limitations in the present study which merit consideration. For example, our intervention to deplete MDSCs by GEM was at the animal level; thus, our hypothesis needs further investigations to verify. In addition, we have noticed the numerous side effects of GEM during application in humans, such as the dose-limiting toxicity (myelosuppression, thrombocytopenia, and anemia) and the minimal nonhematologic toxicity (nausea, shortness of breath, mouth sores, diarrhea, neuropathy, hair loss, etc.) [51], which may limit the chance of GEM being a useful therapy in psoriasis patients. However, we verified the significant anti-inflammatory effects by depleting MDSCs. Despite the severe side effects of Gemcitabine, we still can conclude that targeting MDSCs is a potential strategy for antipsoriasis therapy.
In summary, our study provided evidence that MDSCs play a proinflammatory role in IMQ-induced psoriasis-like skin inflammation and regulating the infiltration of CD4 + T-cells ( Figure 5(d)). Depleting MDSC by its inhibitor (Gemcitabine) significantly suppresses the IMQ-mediated psoriatic phenotype as well as the accumulation of Th17 and Treg cells. Furthermore, we identified and validated the transcriptional expression changes of genes including IL-21 and Timd2 on CD4 + T-cells of GEM-treated mouse models, which are involved in Th17-cell differentiation or T-cell activation. Neutralizing IL-21R by antibody reduces IMQinduced epidermal thickening through downregulating the infiltration of MDSCs and Th17 cells ( Figure 5(d)), suggesting the accumulation of MDSCs exerts important function for the pathogenesis of psoriasis and IL-21 may be a potential therapeutic target in psoriasis.

Conclusions
Targeting myeloid-derived suppressor cells is a novel strategy for antipsoriasis therapy. IL-21 may be a potential therapeutic target in psoriasis.

Data Availability
The RNA-seq data that support the findings of this study have been deposited in the CNSA (https://db.cngb.org/ cnsa/) of CNGBdb with accession number CNP0001133.