Yupingfeng Granule Improves Th2-Biased Immune State in Microenvironment of Hepatocellular Carcinoma through TSLP-DC-OX40L Pathway

The tumor immunological microenvironment in hepatocellular carcinoma (HCC) is the T-helper (Th) 2 dominant inhibition state. Improving the immunosuppressive tumor microenvironment represents an important strategy for HCC treatment. TSLP-OX40L pathway is a target to improve Th2 immunosuppression. Yupingfeng granule (YPF) is clinically used to effectively improve the immune status of HCC. In this study, YPF increased the percentage of mature dendritic cells (DCs) and decreased levels of TSLP, TSLPR, and OX40L in tumor and adjacent tissues of the orthotopic-HCC mice model. This occurs together with the decreased levels of Th2 cytokines and increased levels of Th1 cytokines and Th1/Th2 ratio. In vitro experiment showed that YPF not only increased the percentage of mature DCs and stimulated IL-12 secretion in DCs but also reduced the positive rate of OX40L expression, decreased the proportion of CD4+ IL-13+ T cells, increased levels of Th1 cytokines, and decreased levels of Th2 cytokines from TSLP-treated DCs. In summary, these findings demonstrated that YPF promoted the maturation of DCs, decreased OX40L in TSLP-induced DCs, and improved the immunosuppressive state of Th2 in HCC microenvironment. Our results suggest that the mechanism underlying the improving effect of YPF on the immunosuppression is related to the DC-mediated TSLP-OX40L pathway.


Introduction
Tumor microenvironment in the chronic inflammatory states makes infiltrated immune cells differentiate along the direction of growth, invasion, and metastasis of tumor, thus accelerating the progression of tumor and immune escape processes [1]. e tumor immunological microenvironment in hepatocellular carcinoma (HCC) is the T-helper ( ) 2 dominant inhibition state [2]. Improving the immunosuppressive tumor microenvironment in HCC represents an important strategy for HCC treatment.
ymic stromal lymphopoietin (TSLP) is a pleiotropic cytokine and activates dendritic cells (DCs) by its receptor (TSLPR). TSLP-activated DCs prime for 2 cell differentiation. And a variety of studies demonstrated that TSLP could affect 2 inflammation to participate in the occurrence and development of diseases [3,4]. OX40, a costimulatory receptor, is transiently expressed on activated T cells [5]. OX40L, the ligand of OX40, is directly regulated by TSLP that is generated by epithelial cells, mast cells, and DCs engaged in 2 responses [6]. TSLP-activated DCs have the overexpression of costimulatory molecules (i.e., CD80/ CD86 and MHC class-II), which leads CD4 + T cells to display the phenotypes of 2 differentiated cells (production of IL-4, IL-5, and IL-13) and upregulates the OX40L expression [7,8]. OX40L lost the ability to polarize 2 cells in the presence of IL-12. OX40L functions as the 2-cellpolarizing signal derived from TSLP-activated DCs [9]. TSLP-OX40L signaling pathway mediated by 2 responses is involved in human inflammation and immune system [10]. e effects of TSLP on tumor microenvironments of tumor including lung cancer, breast cancer, and pancreatic cancer had been carried out step by step in previous studies. TSLP overexpression was found in tumor tissues of lung cancer and pancreatic cancer. TSLP is almost limited to express in inflammatory sites and activates DCs to promote 2 immune responses and produces the immunosuppressive factors such as IL-10 and IL-13 [11][12][13]. us, TSLP overexpression and its pathways play an important role in inducing 2-immunosuppressive state in tumor microenvironment. However, whether TSLP expresses in HCC microenvironment is still unknown. Interestingly, our previous study has found that TSLP, TSLPR, and OX40L are highly expressed in HCC microenvironment [14], which is associated with immunosuppressive status. erefore, it is likely that the TSLP-OX40L pathway is a target to improve 2 immunosuppression in HCC microenvironment. Yupingfeng granule (YPF), an ancient traditional Chinese medicine (TCM) formula, is composed of Astragali Radix (AR; Huangqi; the root of Astragalus membranaceus (Fisch.) Bunge), Atractylodis Macrocephalae Rhizoma (AMR; Baizhu; the rhizomes of Atractylodes macrocephala Koidz.), and Saposhnikoviae Radix (SR; Fangfeng; the roots of Saposhnikovia divaricate (Turcz.) Schischk.). YPF is used for preventing cold and treat upper respiratory tract infection in clinical practice and also used for treatment of immunological diseases such as chronic bronchitis, allergic rhinitis, and asthma [15,16]. In clinical use for treatment of HCC, the antineoplastic property of YPF could improve the life quality, reduce the malignancy, and enhance the immunity [17][18][19]. Moreover, a previous study by our group has shown that YPF could improve the immune disorder in mice bearing with HCC by reducing the proportion of CD4 + CD25 + regulatory T cells, promoting the activation of CD8 + T cells, and increasing the proportion of memory CD8 + T cells. YPF exerts a therapeutic effect on HCC by improving the immunosuppressive tumor microenvironment. Surprisingly, we found that YPF could significantly reduce the TSLP expression and restore 1/ 2 balance in tumor and adjacent tissues [16,[20][21][22]. However, whether YPF regulates the TSLP-OX40L pathway to promote normal maturation of DCs, thereby improving the immunosuppressive tumor microenvironment and exerting an antitumor effect on HCC remains unknown. is study demonstrated that YPF increased the expression levels of DC surface markers, decreased OX40L expression in DCs, and reduced the differentiation of T cells into 2 cells. Our results suggest that the mechanism underlying the improving effect of YPF on the immunosuppression of tumor microenvironment in HCC is related to the DC-mediated TSLP-OX40L pathway.

Animal
Model and Treatment. C57/BL6 mice were used for the study after three days of adaptive feeding. e orthotopic transplanted model of murine HCC was established according to our previous study [23]. e HCC murine model was induced by intrahepatic implantation of 1 × 10 6 Hepa1-6 cells into the left liver lobes of mice, and the animals were maintained in laminar-flow cabinets under pathogen-free conditions. After ten days, the mice were at the initial stage of the tumor and were randomly distributed into five groups (eight mice in each group). Mice in YPFtreated groups were treated with 10 g·kg −1 , 15 g·kg −1 , and 20 g·kg −1 YPF in 0.2 ml (dissolved in saline) via intragastric administration. Saline served as the model control group, and cisplatin (10 mg/bottle, Qilu Pharmaceutical Co. Ltd., Shandong, China) served as the chemotherapy control once every other day for 2 weeks. Meanwhile, eight C57BL/6 mice without tumor cell inoculation were treated with saline as the normal control group.
Twenty-four hours after the last administration at day 14, the mice were sacrificed. e tumor and adjacent tissues were weighed and then collected. Tumor inhibition rate (%) � (mean tumor weight of the model control group-− mean tumor weight of the experiment group)/mean tumor weight of the model control group × 100%.

Preparation of the Homogenate in Tumor and Adjacent
Tissues. Tumor and adjacent liver tissue specimens of HCCbearing mice treated with saline or YPF were homogenized on ice with phosphate buffer (PBS, 50 mg tissue and 500 μl PBS). Homogenates were centrifuged at 3,000 ×g for 10 min at 4°C, and the supernatants (100 μl) were used for analysis.

Cytometric Bead Array (CBA) and ELISA Assays of TSLP/ TSLPR/OX40L Expression in the Homogenate.
To measure cytokine levels, the mouse CBA flex kits were used. IL-2, IL-4, IL-5, IL-10, IL-12p70, IL-13, TNF-α, and IFN-c (BD Biosciences, San Jose, California, USA) cytokines were detected in a single sample. e assays were performed according to the manufacturer's instructions. Samples and standards were acquired on the BD Accuri C6 flow cytometer (BD Biosciences, San Jose, California, USA), and the generated FSC files were analyzed using FCAP Array version 3.0 software (BD Biosciences, San Jose, California, USA). e concentrations of TSLP/TSLPR/OX40L (R&D Systems, Minneapolis, MN, USA) in the tissue homogenate were quantitatively measured using the commercially available ELISA kits according to manufacturer's protocols.

Preparation of Single-Cell Suspension in Tumor and Adjacent Tissues and Flow Cytometry Detection.
e tumor and adjacent tissues were collected and then were cut into small pieces. ese pieces were digested at room temperature for 2 h with Hanks solution containing 1 mg·ml −1 type-II collagenase. e cells were filtered with the 200 μm pore size microporous membrane and then were washed with PBS twice. DCs were sorted and purified by anti-CD11c microbead isolation kits (Miltenyi Biotec, Auburn, CA, USA) according to the manufacturer's protocol. After cell sorting, DCs were detected by flow cytometry for the markers CD80, CD86, and MHC-II (eBioscience, Grand Island, NY, USA).

Culture and Maturation of Bone Marrow-Derived DCs.
e method used for the culture of mouse DCs was reported in a previous study [24]. Bone marrow cells from femur and tibia specimens were flushed out with 10 ml of RPMI-1640 (Gibco, Grand Island, NY, USA) using a syringe and filtrated with a 200 mesh filter. After lysis of red blood cells, the remaining cells were washed with PBS twice. Bone marrow cells were cultured in dishes containing the RPMI-1640 medium with 10% FBS, 20 ng·ml −1 recombinant murine GM-CSF (Peprotech, Rocky Hill, NJ, USA), and 20 ng·ml −1 recombinant murine IL-4 (Peprotech, Rocky Hill, NJ, USA) at 37 C in a humidified atmosphere containing 5% CO 2 . Every other day, half of the culture medium was replaced with the fresh medium containing GM-CSF and IL-4. On

Mix Lymphocyte Reaction.
is assay evaluated the stimulating effect of DCs on T cells. DCs cocultured with TSLP 24 h were used as stimulator cells. T lymphocytes were obtained from C57/BL6 mice using the lymphocyte separation medium (Ficoll-Paque PLUS, GE, MA, USA). e stimulated DCs and T cells were cocultured in 6-well flatbottom plates at a ratio of 1 : 10 in the RPMI-1640 medium with 10% FBS as described previously [25]. e different concentrations of YPF were added in the mixed lymphocyte for 48 h. e collected cells were detected by flow cytometry for CD4 and IL-13 (eBioscience, Grand Island, NY, USA). e collected culture supernatants were used for cytokine level assessment with CBA flex kits (BD Biosciences, San Jose, California, USA).

Statistical Analysis.
All data represent at least three independent experiments, and results of the experimental studies are expressed as mean ± standard deviation (SD). Statistical significance of differences was analyzed by the Student's t-test or one-way analysis of variance followed by the Bonferroni or Dunnett's post hoc tests (Graph Pad Prism Software, San Diego, CA, USA). e values of p < 0.05 were considered statistically significant.

YPF Inhibited the Tumor Growth and Improved 2-Biased Immune State of Tumor Microenvironment in Mice
Bearing with HCC. After the last administration (food deprivation for 12 h), the mice were weighed and sacrificed. Tumor and adjacent tissues were isolated from tumorbearing mice, and homogenate was prepared. Cytokine contents were detected by flow cytometry with the CBA kit. Compared with the model control group, YPF increased the expression of IL-12, TNF-α, and IFN-c in tumor and adjacent tissues (p < 0.05 and p < 0.01) but did not affect IL-2 level (p > 0.05). Compared with the normal control group, the levels of 1 cytokines (IL-2, IL-12, TNF-α, and IFN-c) in the adjacent tissues of the model control group were on a downward trend, but there was no statistical significance (p > 0.05). YPF (20 g·kg −1 ) increased the expression of TNFα and IFN-c in adjacent tissues (p < 0.05, p < 0.01; Figure 1(b)). It was suggested that YPF could improve the overall level of 1 cytokines in tumor and adjacent tissues of tumor-bearing mice. Compared with the model control group, YPF decreased the expression of IL-4 and IL-10 in tumor tissues and adjacent tissues (p < 0.05 and p < 0.01) and IL-5 in adjacent tissues (p < 0.05 and p < 0.01) but had no effect on IL-13 (p > 0.05). Compared with the normal control group, the levels of 2 cytokines (IL-4, IL-5, IL-10, and IL-13) in the adjacent tissues of the model control group showed no significant change (p > 0.05). But, YPF significantly decreased the expression of IL-5 and IL-10 in adjacent tissues (p < 0.05, p < 0.01; Figure 1(c)). It was suggested that YPF could reduce the overall level of 2 cytokines in tumor and adjacent tissues of tumor-bearing mice.
Based on the study of the changes of 1 and 2 cytokines, the 1/ 2 ratio was used to reflect the immune status of tumors and adjacent tissues. IFN-c/IL-4 value often represents the 1/ 2 balance. Compared with the model control group, the ratio of 1/ 2 (IFN-c/IL-4) in tumors and adjacent tissues was significantly increased in YPF groups (15 and 20 g kg −1 ) (p < 0.05, p < 0.01; Figure 1(d)). It was suggested that YPF could adjust the 2 dominant status in tumors and adjacent tissues of tumorbearing mice.  Evidence-Based Complementary and Alternative Medicine e results also showed that YPF could inhibit the growth of HCC in a dose-dependent manner. Compared with the model control group, YPF (10, 15, and 20 g·kg −1 ) treatment significantly suppressed tumor growth (p < 0.01). Tumor inhibitory rates for YPF (10,15, and 20 g kg −1 ) were 25.48%, 34.92%, and 46.76% (Figure 1(a)).
ese results mainly indicated that YPF may inhibit the growth of tumors by adjusting 2-biased in the state of tumors.

YPF Affected the TSLP-OX40L Pathway and DC Maturation on Tumor Microenvironment in Mice Bearing with
HCC. TSLP is a key factor mediating the 2 reaction at the interface between the body and the environment (skin, intestine, respiratory tract, etc.). Activating DCs through TSLPR can upregulate the expression of the OX40 ligand (OX40L) in DCs [26,27], thereby inducing 2 inflammatory response [28]. erefore, we further observed the expression of TSLP, TSLPR, and OX40L in tumor and adjacent tissue after administration of YPF. e results showed that the levels of TSLP, TSLPR, and OX40L in the adjacent tissues of the model control group were higher than those of the normal control group. Among these variations, a statistically significant difference in TSLP was found (p < 0.05). Compared with the model control group, YPF decreased the levels of TSLP, TSLPR, and OX40L in tumor and adjacent tissues (p < 0.05, p < 0.01; Figure 2(a)). It was suggested that YPF could decrease the levels of TSLP, TSLPR, and OX40L of tumor microenvironment in mice bearing with HCC.
At the same time, we also observed the maturation of DC cells in tumor and adjacent tissues. Single-cell suspensions of DCs were stained for the markers CD80, CD86, and MHC-II. Compared with the model control group, YPF increased the expression of CD80, CD86, and MHC-II of DCs in tumors and adjacent tissues (p < 0.05, p < 0.01; Figure 2(b)). It showed that YPF could increase the percentage of mature DCs and promote the maturation of DCs in tumor microenvironment in mice bearing with HCC, thereby adjusting the immunosuppressive state of the 2 type.

YPF Effect on Maturation and IL-12 Secretion of DCs In
Vitro. Further in vitro experiments were carried out. e effects of YPF on the differentiation and maturation of DCs derived from the bone marrow of mice (CD80, CD86, and MCH-II expression) and the secretion level of IL-12 were observed. e results showed that immature DCs were cultured and differentiated for 6 days. LPS (1 μg·mL −1 ) was added for 24 h to stimulate the maturation of immature DCs. Compared with the control group, LPS increased the expression levels of CD80, CD86, and MHC-II (p < 0.05, p < 0.01; Figure 3(a)). YPF increased the expression levels of CD80, CD86, and MHC-II (p < 0.05, p < 0.01; Figure 3(b)) and stimulated IL-12 secretion (p < 0.01; Figure 3(c)) in DCs. It was further confirmed that YPF could stimulate the maturation of DCs in vitro.

YPF Effect on TSLP-Mediated DCs In Vitro.
As TSLP is a potent activator of DCs [25], we evaluate whether YPF improves the mature states of DCs induced by TSLP. First, cell-surface receptors of DCs are required for T-cell activation [29]. Compared with the control group, both TSLP and YPF increased the expression levels of CD80, CD86, and MHC-II (p < 0.05, p < 0.01; Figure 4(a)). As DCs express OX40L, a 2-polarizing molecule [30], we further examined the ability of TSLP to activate DCs to produce OX40L. Overexpression of CD252 (OX40L) was found in TSLPinduced DCs, and YPF decreased the positive expression rate (Figure 4(a)).
Additionally, TSLP induces the expression of maturation markers on DCs and increases their ability to polarize lymphocytes into the 2 phenotype [31]; we assessed the capability of DCs in different groups to stimulate and polarize T-cell response by coculturing them with naïve T cells, an assay used to evaluate the stimulatory ability of DCs to alter T-cell differentiation and cytokine production. Compared with the mix control group, YPF reduced the proportion of CD4 + IL-13 + T cells (p < 0.05, p < 0.01; Figure 4(b)). Compared with the mix control group, TSLP decreased the expression of 1-type cytokines (IL-12, TNF-α, and IFN-c, p < 0.05, p < 0.01) and increased the expression of 2-type cytokines (IL-4, IL-10, and IL-13, p < 0.05, p < 0.01). DC-T-cell supernatants in the YPF groups showed significantly high levels of IL-12, TNF-α, and IFN-c (p < 0.05, p < 0.01) and lower levels of IL-10 and IL-13 (p < 0.05, p < 0.01), while IL-4 shows a downward trend but has no statistical significance compared with the TSLP group (Figures 4(c) and 4(d)). ese findings demonstrated that YPF promoted the expression levels of TSLP-induced DC surface markers, decreased OX40L in TSLP-induced DCs, changed the differentiation of T cells into 2 cells, and affected the potential of DCs mediated by TSLP to polarize lymphocytes.

Discussion
In this study, in the orthotopic-mice model of HCC, YPF increased the levels of 1 cytokines (IL-2, IL-12, TNF-α, and IFN-c), decreased the levels of 2 cytokines (IL-4, IL-5, IL-10, and IL-13), and increased the 1/ 2 ratio in tumor and adjacent tissues. It suggests that YPF exerts anti-HCC effect by improving the 2-immunosuppressive state. Our data also showed that YPF decreased the levels of TSLP, TSLPR, and OX40L, increased the expression of CD80, CD86, and MHC-II in DCs, and stimulated the maturation of DCs in tumor and adjacent tissues of mice bearing with HCC. Meanwhile, YPF increased the expression of CD80, CD86, and MHC-II and the secretion of IL-12 in DCs in vitro. YPF also reduced the positive expression rate of CD252 (OX40L), decreased the proportion of CD4 + IL-13 + T cells, increased the levels of 1 cytokines (IL-12, TNF-α, and IFN-c), and decreased the levels of 2 cytokines (IL-4, IL-10, and IL-13) from TSLP-induced DCs. ese findings demonstrated that YPF promotes the expression levels of Evidence-Based Complementary and Alternative Medicine DC surface markers, decreases OX40L in DCs, changes the differentiation of T cells into 2 cells, and affects the potential of DCs mediated by TSLP to polarize lymphocytes. Our results suggest that the mechanism underlying the improving effect of YPF on the immunosuppression of tumor microenvironment in HCC is related to the DCmediated TSLP-OX40L pathway.
1 and 2 lymphocytes are the two main subsets of CD4 T-helper cells with different functions and patterns of cytokine secretion, which are identified in both mice and humans. 1 and 2 are in a state of dynamic balance, maintaining the normal immune function of the body. 1/ 2 bias is different with different immune statuses. 1 cells mainly secret cytokines such as IL-2, IL-12, TNF-α, and IFNc, which can enhance the cytotoxicity of immune cells to kill tumor cells and make the body's anticancer immune response stronger [32,33]. 2 cells mainly secrete cytokines such as IL-4, IL-5, IL-10, and IL-13, which are involved in humoral immunity. It is generally believed that 2 cytokines can inhibit the differentiation of CD4 + T cells into 1 cells and weaken the immune response to antitumor effect, thus promoting the development of tumors [34,35].

Downregulated
1 and upregulated 2 cytokine production has been found in tumor and adjacent tissues of HCC patients. It suggests that 1/ 2 shift exists in the immunosuppressive state [33]. In the present study, the levels of 1 cytokines (IL-2, IL-12, TNF-α, and IFN-c) were on a downward trend, but the levels of 2 cytokines (IL-4, IL-5, IL-10, and IL-13) were increased. e ratio of 1/ 2 (IFN-c/IL-4) was on a downward trend in the adjacent tissues of mice bearing with HCC. ese results showed that the tumor microenvironment of HCC was 2-dominanted immunosuppression. YPF, a classical TCM formulation, has been traditionally used for treatment of immune systemrelated diseases such as chronic bronchitis, allergic rhinitis, and asthma [36]. A previous study has demonstrated that YPF efficiently inhibited 2 cytokines [15]. e current study showed that YPF could improve the immune suppression caused by the imbalance of 1/ 2 ratio in the tumor microenvironment of HCC. It suggests that YPF exerts anti-HCC effect by improving the 2 immunosuppressive state.
TSLP is a member of the 4-helix bundle cytokine family, and a distant paralog of IL-7 is first identified as an activity in the supernatants of a mouse thymic stromal cell line. It has now become apparent that a major TSLP-responsive cellular subset in both humans and mice is myeloid-derived dendritic cells (mDCs) [37]. Pedroza-Gonzalez showed that inflammatory 2 cells that promoted tumor development were driven by cancer cell-derived TSLP, which induced and maintained OX40L-expressing DCs in the tumor microenvironment. TSLP induces DC maturation through combination with TSLPR on the surface of DCs. e maturing DCs can upregulate the expression of OX40L through TSLPR. And mature DCs then promote 2 differentiation of naïve CD4 T cells and induce CD4 + T cells to secrete type 2 cytokines such as IL-13, contributing to tumor development [28,38]. In our experiment in vitro, a similar phenomenon was observed. TSLP promoted the maturation of DCs, increased the expression of OX40L, differentiated CD4 + IL-13 + T cells, decreased the secretion of 1 cytokines, and increased the secretion of 2 cytokines. Interestingly, YPF can promote the maturation of DCs but reduce the expression of OX40L (CD252) induced by TSLP. However, YPF significantly reduced the proportion of CD4 + IL-13 + Tcells, increased the secretion of 1 cytokines (IL-12, TNFα, and IFN-c), and decreased the secretion of 2 cytokines (IL-4, IL-10, and IL-13) in DCs induced by TSLP. Our data also showed that YPF decreased the levels of TSLPR and OX40L in tumor and adjacent tissues in vivo. We speculated that the decrease of the OX40L expression was related to the decrease of the TSLPR expression. It was further confirmed that the mechanism of YPF in improving tumor microenvironment in mice with HCC was related to the TSLP-OX40L pathway.
Recently, a new and unexpected function for TSLP has been found for both promotion and suppression of solid tumor growth [39]. In mouse models of breast and pancreatic carcinogenesis, it was found that early administration of TSLP blocked cancer development [40]. But, in the middle and the late stage, the level of the TSLP expression was correlated with tumor growth and metastasis [12], and tumor-derived TSLP could act on TSLPR + endothelial cells to promote angiogenesis in cervical cancer [41]. Conversely, Di Piazza and Demehri demonstrated that TSLP exerted a tumor-suppressing role in the murine model of skin cancer [26,40]. Yue et al. also reported the decreased TSLP expression in human colon cancer and TSLP levels was negatively correlated with the clinical staging score of cancer, and TSLP enhanced apoptosis of colon cancer cells * * * Count through the engagement of TSLPR [42]. e present study shows for the first time that the expressions of TSLP, TSLPR, and OX40L, which were the key target proteins of the TSLP-OX40L pathway, were increased in the tumor microenvironment of mice bearing with HCC. e TSLP expression in HCC was positively correlated with its microenvironment immunosuppression. But, the different roles of TSLP in early and advanced HCC are deserved for further study.

Conclusions
We present the first evidence of TSLP protein expression in tumor and adjacent tissues of HCC. Our results suggest that the TSLP-OX40L pathway is involved in the immunosuppression of HCC microenvironment, and YPF is an immunomodulator against the microenvironment immune suppression of HCC in mice. e mechanism underlying the improving effect of YPF in improving the immunosuppression of tumor microenvironment in HCC may be related to the TSLP-OX40L pathway, which provides the experimental and theoretical basis for YPF to prevent and treat HCC. Development of the targeted TSLP pathway to improve the microenvironment immunosuppression of HCC provides a new strategy for prolonging the survival of patients with HCC and new targets for ideal therapeutic drugs.
Data Availability e data and materials supporting the conclusions of this article are included within the article.  Evidence-Based Complementary and Alternative Medicine