Substantial evidence indicates that inflammation is a critical component of tumor progression. The proinflammatory IL-17-producing cells have recently been detected in tumors, but the effect of IL-17 on antigen-presenting cells in tumors is presently unknown. We recently found that B7-H1+ macrophages (Mφs) were enriched predominantly in the peritumoral stroma of hepatocellular carcinomas (HCCs). Here, we found a positive correlation between IL-17-producing cells and B7-H1-expressing Mφs in the same area. The B7-H1+ monocytes/Mφs from HCC tissues expressed significantly more HLA-DR, CD80, and CD86 than B7-H1– cells. Accordingly, IL-17 could activate monocytes to express B7-H1 in a dose-dependent manner. Although culture supernatants derived from hepatoma cells also induced B7-H1 expression on monocytes, IL-17 additionally increased hepatoma-mediated B7-H1 expression. Autocrine inflammatory cytokines released from IL-17-activated monocytes stimulated B7-H1 expression. Moreover, these IL-17-exposed monocytes effectively suppressed cytotoxic T-cell immunity in vitro; the effect could be reversed by blocking B7-H1 on those monocytes. Consistent with this, cytotoxic T cells from HCC tissues expressed significant B7-H1 receptor programmed death 1 (PD-1) and exhibited an exhausted phenotype. These data reveal a fine-tuned collaborative action between different stromal cells to counteract T-cell responses in tumors. Such IL-17-mediated immune tolerance should be considered for the rational design of effective immune-based anti-cancer therapies.
The tumor microenvironment, composed of non-cancer cells and their stroma, is now recognized as a major factor influencing the progression of cancer. Normal stroma is non-permissive for neoplastic growth, but cancer cells can modulate adjacent stroma to generate a supportive microenvironment. Increasing evidence suggests that the proinflammatory response at the tumor stroma can also be rerouted into a tumor-promoting direction 1, 2. We recently observed that IL-17-producing cells were highly enriched in the peritumoral stroma of hepatocellular carcinoma (HCC), and their levels were positively correlated with disease progression in patients 3, 4. At present, little is known about the potential regulatory function of IL-17 in tumor immunopathology.
Macrophages (Mφs) markedly outnumber other antigen-presenting cells (APCs) and represent an abundant population of stroma cells in tumors 5–8. Tumor-associated Mφs are derived almost entirely from circulating blood monocytes 9–11. Although the precise underlying mechanisms are not yet clear, it is generally assumed that the tumor microenvironment is a critical determinant of the phenotype of Mφs in situ. Both clinical and experimental studies in autoimmune diseases have demonstrated that the cytokine IL-17 can potentiate the local inflammatory responses by altering the functional characteristics of APCs 12–18. Thus, immune functional data of monocytes/Mφs in response to proinflammatory IL-17 are essential for understanding their roles during neoplastic progression.
B7-H1 (also termed PD-L1 and CD274) is expressed on resting T cells, B cells, dendritic cells, and monocytes/Mφs and is further up-regulated on activation. B7-H1, which acts as an antiapoptotic receptor, is also expressed on some tumor cells 19–21. In one of our recent studies, we observed that B7-H1+ monocytes/Mφs were enriched predominantly in the peritumoral stroma of HCC tissues, and the expression of B7-H1 proteins was regulated by autocrine inflammatory cytokines from tumor-activated monocytes 10. The present study shows that the levels of B7-H1+ monocytes/Mφs correlate well with IL-17-producing cell density in the peritumoral stroma. The proinflammatory cytokine IL-17 could induce expression of B7-H1 on monocytes in a dose-dependent manner. Moreover, we also found that IL-17-treated monocytes suppressed the cytotoxic T-cell immunity through B7-H1/programmed death 1 (PD-1) interactions. Therefore, B7-H1 expression on IL-17-exposed monocytes may represent a novel mechanism by which the proinflammatory response is linked to immune tolerance in the tumor milieu.
Correlations between IL-17+ cell and B7-H1+ monocyte/Mφ density in the peritumoral stroma of HCC
We have recently observed that both B7-H1-expressing monocytes/Mφs and IL-17-producing cells accumulate predominantly in the peritumoral stroma of human HCC; high infiltration of these two cell types is well correlated with transient activation of monocytes/Mφs in the same areas 3, 10. These findings suggest a possible interaction between IL-17-producing cells and B7-H1-expressing cells in the designated areas of human HCC tissues. To test this assumption, the presence of these two cell types was visualized by dual immunohistochemical staining with antibodies (Abs) against IL-17 and B7-H1 in paraffin-embedded tissues from 80 untreated HCC patients (Table 1). B7-H1-expressing cells and IL-17-producing cells accumulated in the peritumoral stroma of tissues (148.6±6.9 and 51.0±4.8 cells/field respectively); and a positive correlation was found between the densities of these two cell types in the same area (r=0.741, p<0.001; Fig. 1A and B). Moreover, we confirmed that the B7-H1 protein was expressed mainly on monocytes/Mφs (n=10) and was only weakly expressed on other stromal and hepatoma cells by dual staining of B7-H1 and CD68 (Fig. 1C). Of note, we only detected marginal levels of B7-H1-expressing cells and IL-17-producing cells in the cancer nests 3, 4, 10.
Table 1. Clinical characteristics of the 91 HCC patients
To further elucidate the functional states of monocytes/Mφs in HCCs, we next examined the phenotypic features of B7-H1+ and B7-H1− monocytes/Mφs isolated from tumor tissues, and the results showed that the B7-H1+ monocytes/Mφs from HCC tissues expressed significantly more HLA-DR, CD80, and CD86 than B7-H1– monocytes/Mφs (Fig. 1D). Measuring mRNAs of various cytokines expressed in non-tumor- and tumor-infiltrating monocytes/Mφs (over 50% of which were B7-H1+) using real-time PCR revealed up-regulation of IL-1β, IL-6, IL-10, IL-23, and TNF-α in monocytes/Mφs from the tumor tissues (Fig. 1E). Collectively, these data show that activated B7-H1+ monocyte/Mφ density increases with IL-17+ cell infiltration in human HCC patients.
Proinflammatory IL-17 induces expression of B7-H1 on monocytes/Mφs
Recent studies have shown that IL-17 can facilitate the proinflammatory responses of monocytes/Mφs 13. To investigate whether such mechanisms are also responsible for the selective expression of B7-H1 on monocytes/Mφs in the peritumoral stroma, we incubated healthy blood monocytes with recombinant human (rh) IL-17 at different concentrations for 48 h. As expected, the rhIL-17 induced expression of B7-H1 on monocytes in a dose-dependent manner (Fig. 2A and B). Notably, 24-h exposure of monocytes to rhIL-17, at a concentration up to 100 ng/mL, only induced marginal expression of B7-H1 (Supporting Information Fig. S1). In a recent study, we observed that hepatoma cell culture supernatants can stimulate monocytes to express B7-H1 in vitro 10. Although rhIL-17 could not elicit marked expression of B7-H1 at a level comparable to that exhibited by monocytes in response to hepatoma cell culture supernatants, it did additionally increase the hepatoma-mediated B7-H1 expression on monocytes (Fig. 2A and B). This suggests a contributing role for IL-17 in the induction of B7-H1 expression on monocyte in the tumor milieu.
APCs are more susceptible to express B7-H1 when they are activated 10, 19; hence, we examined the activation status of IL-17-treated monocytes. Consistent with the in vivo observations, exposure to rhIL-17 for 24 h resulted in significant up-regulation of HLA-DR, CD80, and CD86 on monocytes, albeit not as prominent as that induced by hepatoma cell culture supernatants (Fig. 2C). Similar patterns of cytokine productions were obtained in the IL-17-exposed monocytes, including the accumulation of IL-1β, IL-6, IL-10, IL-23, and TNF-α in the culture media (Fig. 2D). Furthermore, comparison of the kinetics of B7-H1 expression and the cytokine production in IL-17-treated monocytes revealed that cytokine profiles preceded B7-H1 expression (Fig. 2A and D; and data not shown). These findings suggest that IL-17-induced activation of monocytes led to expression of B7-H1.
Autocrine cytokines from IL-17-activated monocytes induce expression of B7-H1
We recently observed that inflammatory cytokines derived from hepatoma-activated monocytes/Mφs can trigger expression of B7-H1 in human tumors 10. To ascertain whether autocrine cytokines from IL-17-activated monocytes contribute to the expression of B7-H1, we used specific neutralizing Abs to abolish the effects of IL-1β, IL-6, IL-10, IL-23, or TNF-α (Table 2). Supporting our hypothesis, blocking IL-10 effectively inhibited the up-regulation of B7-H1 protein on monocytes (Fig. 3A). Interestingly, monocyte B7-H1 expression was also partially attenuated by Ab against IL-1β, IL-23, or TNF-α, whereas it was not affected by treatment with a concentration of anti-IL-6 Ab that effectively neutralized IL-6 in the culture system (Fig. 3A and Table 2). To gain further support for the role of autocrine cytokines in induction of B7-H1 on monocytes, we tested rh cytokines in the culture system. As shown in Fig. 3B, although rhIL-1β, rhIL-10, rhIL-23, or rhTNF-α alone, at concentrations similar to their levels in the culture supernatant from IL-17-treated monocytes, only had a marginal effect on expression of monocyte B7-H1, a combination of the four cytokines did induce significant expression of B7-H1. These findings, together with results showing that IL-17-producing cells were enriched predominantly in the peritumoral stroma (Fig. 1A), suggest that autocrine inflammatory cytokines released by IL-17-treated monocytes lead to expression of B7-H1 in the same area.
Table 2. The use of specific Abs to block the effects of IL-1β, IL-6, IL-10, IL-23, or TNF-α during exposure of monocytes to IL-17
Purified monocytes were stimulated for 24 h by 100 ng/mL of rhIL-17 in the presence of an antibody specific for IL-1β, IL-6, IL-10, IL-23, or TNF-α, after which the concentration of corresponding cytokine (pg/mL) in the culture supernatants were determined by ELISA. Each value represents the mean±SD of the results of three separate experiments.
IL-17-treated monocytes suppress T-cell immunity through PD-1/B7-H1 interactions
APCs are critical for initiating and maintaining tumor-specific T-cell responses; and B7-H1 expressed by APCs has previously been found to inhibit T-cell immunity in humans 10, 22–25. To identify whether IL-17-treated monocytes suppress T-cell function, these monocytes were generated by exposure to rhIL-17 for 48 h and then cultured for 6 days with autologous blood CD8+ T cells in the presence of polyclonal stimulation. The results showed that exposure to IL-17-treated monocytes induced dysfunctional CD8+ T cells that had low cytotoxicity and attenuated production of IFN-γ (Fig. 4A). As expected, such monocyte-mediated dysfunction of cells happened only when CD8+ T cells expressed the B7-H1 receptor PD-1 (Fig. 4A). Furthermore, we analyzed the cytotoxic capacity of CD8+ T cells that have been exposed to untreated or IL-17-treated monocytes. The CD8+ T cells cultured with untreated monocytes could effectively inhibit the proliferation of HepG2 and exert their cytotoxicity against those cells (Fig. 4C and D). However, exposure to IL-17-treated monocytes for 6 days induced dysfunctional CD8+ T cells that had impaired cytotoxicity against HepG2 cells. Consistent with our hypothesis, blockade of B7-H1 significantly attenuated such T-cell suppression mediated by IL-17-treated monocytes (Fig. 4B–D). These findings show that B7-H1 contributes to IL-17-treated monocyte-mediated suppression in vitro.
The results described above suggest that cytotoxic CD8+ T cells are educated by tumor environments to adopt an exhausted phenotype. To test this, we examined the functional states of CD8+ T cells freshly isolated from tumoral and non-tumoral liver tissues of patients with HCC. A significantly larger portion of the tumor-infiltrating cytotoxic T cells was found to express PD-1 (68.7±8.5%; n=11; Fig. 5A). In support of data from in vitro culture, only PD-1+ cytotoxic T cells isolated from tumor tissues exhibited attenuated production of IFN-γ, IL-2, and TNF-α as well as low cytotoxic potential (Fig. 5B and C).
Despite the generally immunosuppressed status of cancer patients, substantial evidence shows that inflammatory reactions at a tumor site can promote disease progression 26, 27. We recently found that proinflammatory IL-17-producing cells are enriched predominantly in the peritumoral stroma of HCC tissues, where they promote tumor progression by fostering angiogenesis 3, 4, 28. The present study demonstrated that IL-17 induced the expression of B7-H1 on monocytes/Mφs in the peritumoral stroma of HCC, and that subsequently, the IL-17-exposed monocytes/Mφs fostered immune privilege through B7-H1/PD-1 interactions. These findings reveal a fine-tuned collaborative action between different types of immune cells in distinct tumor microenvironments, which reroutes the inflammatory response into immunosuppression.
In contrast to the immunosuppressive micromilieu in most intratumoral areas, the peritumoral stroma contains a significant number of infiltrated leukocytes with potent proinflammatory properties 26, 27. We have recently shown that most monocytes/Mφs in the peritumoral stroma exhibit an activated phenotype that favors the generation of IL-17-producing T cells in the same area, and their levels were correlated with disease progression in HCC patients 3, 4, 28. In the current study, we observed a positive association between B7-H1+ monocyte/Mφ and IL-17-producing cell density in the peritumoral stroma of HCC patients. Data from recent studies showed that functional IL-17-producing cells in the peritumoral stroma stimulated monocytes/Mφs to secret inflammatory cytokines that up-regulated B7-H1 expression. Thereafter, the B7-H1+ monocytes/Mφs in the peritumoral stroma induced dysfunctional cytotoxic T cells via PD-1 signaling. These data therefore provide direct evidence that monocytes/Mφs play an important role in human HCC progression by serving as a link between proinflammatory IL-17 and T-cell dysfunction in the tumor milieu. This notion is supported by our recent findings that the density of monocytes/Mφs in the peritumoral stroma correlated with advanced disease stages and could serve as an independent predictor of poor survival in HCC patients 10.
The present investigation provides evidence that proinflammatory IL-17 in the peritumoral stroma plays a contributing role in the induction of immune escape in HCC. The results of four sets of experiments support this conclusion. First, we observed that both IL-17-producing cells and B7-H1-expressing cells accumulated in the peritumoral stroma of HCC, and the densities of these two cell types were correlated in the same area. Second, the cytokine IL-17 alone induced significant expression of B7-H1 on monocytes in vitro, and in parallel, it could synergistically increase the hepatoma-mediated B7-H1 expression. Third, blocking a set of autocrine cytokines, including IL-1β, IL-10, IL-23, and TNF-α released from IL-17-treated monocytes clearly inhibited the expression of B7-H1, and relatively low concentrations of the recombinant cytokines could mimic the stimulatory effect of IL-17 in this regard. Fourth, the IL-17-stimulated monocytes effectively suppressed the cytotoxic T-cell immunity, and that effect was attenuated by blocking B7-H1 on these monocytes. As high infiltration of IL-17-producing cells in the peritumoral stroma was associated with disease progression of HCC 3, 4, IL-17-mediated immunosuppression might contribute to its protumoral effects in the human tumor microenvironment.
Although tumor cells 10 and IL-17-producing cells can both induce suppressive monocytes/Mφs, it is still unclear whether there is a functional difference between tumor-exposed monocytes and IL-17-treated monocytes. In general, we found that these two monocyte subsets had similar phenotypic features, which are characterized by high expression of proinflammatory cytokines and surface markers for antigen presentation during early differentiation stages, and subsequent constitutive expression of B7-H1. This present finding provides evidence for a previously unidentified role of IL-17 in tumor immunopathology: IL-17 can foster immune privilege in tumor microenvironment. The study significantly complements our recent reports 28 that IL-17 plays an important role in human tumor progression by serving as a link between the proinflammatory response and angiogenesis in the tumor milieu.
In addition to triggering tumor progression, IL-17-producing cells are also capable of being antitumorigenic by co-expressing IFN-γ against tumor cells 29–31. Accordingly, IL-17-producing cells, including Th17 and Tc17, have been shown to have both antitumorigenic and protumorigenic functions, owing to their multifaceted biological activities and cellular targets in the tumor microenvironments 30–33. Therefore, a better understanding of the inflammatory context might provide a novel strategy for the rational design of anticancer therapies 34–37.
Materials and methods
Patients and specimens
Tumor samples were obtained from 91 patients with pathologically confirmed HCC at the Cancer Center of Sun Yat-sen University. None of the patients received anti-cancer therapy before sampling. Individuals with concurrence of autoimmune disease, HIV, or syphilis were excluded. Freshly paired tumor and non-tumor (at least 3 cm from the tumor site) tissues from 11 patients who received therapy between 2009 and 2010 were used for the isolation of tumor and non-tumor-infiltrating leukocytes. Samples from another 80 patients who had undergone curative resection between 2003 and 2008 were used in immunohistochemistry assay. The clinical characteristics of all patients are listed in Table 1. All samples were anonymously coded in accordance with the local ethical guidelines (as stipulated by the Declaration of Helsinki). Written informed consent was obtained from the patients and the protocol was approved by the Review Board of Sun Yat-sen University.
Immunohistochemistry and evaluation of immunohistochemical variables
Paraffin-embedded and formalin-fixed samples were cut into 5-μm sections, which were then processed for immunohistochemistry as previously described 10. In short, the specimens were simultaneously incubated with mouse anti-B7-H1 (MIH1; eBioscience) and goat anti-IL-17 (R&D Systems), or with mouse anti-B7-H1 and rabbit anti-CD68 (Santa Cruz Biotechnology), followed by alkaline phosphatase anti-mouse IgG and HRP anti-goat IgG, or by alkaline phosphatase anti-mouse IgG and HRP anti-rabbit IgG, and then they were stained with alkaline phosphatase red and diaminobenzidine.
Analysis was performed by two independent observers using a Leica DM IRB inverted microscope. The tissue sections were screened at low power (100×), and the five most representative fields were selected. To evaluate the densities of B7-H1+ and IL-17+ cells, the respective peritumoral stromal areas were measured at 400× magnification. Thereafter, the nucleated cells in each area were counted manually and expressed as number of cells per field.
Isolation and culture of monocytes
PBMC were isolated from buffy coats derived from the blood of healthy donors by Ficoll density gradient as described previously 11. PBMC in DMEM alone were plated at 4×106 per well in 24-well plates for 1.5 h and then washed. Thereafter, the monocytes in DMEM containing AB serum were cultured in the presence of rh cytokines (all from R&D Systems) or 20% culture supernatants from HepG2 cells 10, or in medium alone for indicated time. In some experiments, before exposure to rhIL-17, the monocytes were pretreated with control IgG1 or with specific blocking Ab against IL-1β, IL-6, IL-10, IL-23, or TNF-α at the indicated concentrations.
Monocytes/Mφs were detached with 5 mM EDTA and then washed and resuspended in PBS supplemented with 1% heat-inactivated FBS. Thereafter, the cells were stained with fluorochrome-conjugated Abs against human CD14, B7-H1, CD80, CD86, and HLA-DR (BD Pharmingen or eBioscience) according to the manufacturer's instructions and analyzed by multicolor flow cytometry (Gallios, Beckman Coulter).
Enzyme-linked immunosorbent assay (ELISA)
The culture supernatants were centrifuged to remove particulate debris and then stored in aliquots at −80°C. Concentrations of IL-1β, IL-6, IL-10, IL-23, and TNF-α were determined by using ELISA kits (eBioscience) according to the instructions provided by the manufacturer.
In vitro T-cell immunosuppression assay
Untreated monocytes and IL-17-treated monocytes were pretreated with 10 μg/mL mitomycin C (Sigma-Aldrich) for 30 min, washed, and cocultured with polyclonal-stimulated (2 μg/mL anti-CD3 and 2 μg/mL anti-CD28) autologous blood CD8+ T cells at a ratio of 1:10 in the presence or absence of 5 μg/mL anti-B7-H1 Ab. After 6 days, the CD8+ T cells were left untreated or stimulated at 37°C for 5 h with Leukocyte Activation Cocktail (BD Pharmingen). Thereafter, the expression of PD-1, perforin, and IFN-γ (eBioscience) in CD8+ T cells was analyzed by multicolor FACS.
In vitro T-cell cytotoxicity
HepG2 were cultured with CD8+ T cells exposed to untreated or IL-17-treated monocytes at the 1:50 ratios for 4 h, with EdU (Invitrogen) present during the final 1 h. Thereafter, cells were collected for EdU staining according to the manufacturer's instructions. CFSE-labeled HepG2 cells were cultured with CD8+ T cells exposed to untreated or IL-17-treated monocytes at the 1:50 ratios for 24 h. Thereafter, cells were collected for propidium iodide (PI) staining, and the death of CSFE+ HepG2 cells was determined by flow cytometry. In some experiments, the IL-17-treated monocytes were pretreated with anti-B7-H1 Ab (MIH1) or control Ab before coculture with CD8+ T cells.
Isolation of mononuclear cells from tissues
Fresh non-tumor- or tumor-infiltrating mononuclear cells were obtained as described elsewhere 3, 4. In short, liver biopsy specimens were cut into small pieces and digested in RPMI1640 supplemented with 0.05% collagenase IV (Sigma-Aldrich), 0.002% DNase I (Roche), and 20% FBS (HyClone Laboratories) at 37°C for 20 min. Dissociated cells were filtered through a 150-μm mesh and separated by Ficoll centrifugation, and the mononuclear cells were washed and resuspended in PBS medium supplemented with 1% FBS. Thereafter, the expression of PD-1, IFN-γ, IL-2, TNF-α, perforin, and granzyme B in CD8+ T cells and B7-H1, HLA-DR, CD80, and CD86 in CD14+ monocytes were analyzed by multicolor flow cytometry; and the expression of cytokine mRNAs (Supporting Information Table S1) in non-tumor- and tumor-infiltrating monocytes were detected by real-time PCR 28.
All statistical analyses were performed using the Student's t test on SPSS statistical software (version 13.0), and two-tailed tests were applied to all data unless otherwise specified, considering p<0.05 statistically significant.
The authors thank Weaver R. Q. for linguistic revision of the manuscript. This work was supported by the Fundamental Research Funds for the Central Universities (11lgzd12) and project grants from the National Natural Science Foundation of China (81000915) and the Education Department of Guangdong Province (LYM10008).
Conflict of interest: The authors declare no financial or commercial conflict of interest.