Activated monocytes in peritumoral stroma of hepatocellular carcinoma promote expansion of memory T helper 17 cells

Authors

  • Dong-Ming Kuang,

    1. State Key Laboratory of Oncology in Southern China, Cancer Center, Guangzhou, P.R. China
    2. State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-Sen University, Guangzhou, P.R. China
    Search for more papers by this author
  • Chen Peng,

    1. State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-Sen University, Guangzhou, P.R. China
    Search for more papers by this author
  • Qiyi Zhao,

    1. State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-Sen University, Guangzhou, P.R. China
    Search for more papers by this author
  • Yan Wu,

    1. State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-Sen University, Guangzhou, P.R. China
    Search for more papers by this author
  • Min-Shan Chen,

    1. State Key Laboratory of Oncology in Southern China, Cancer Center, Guangzhou, P.R. China
    Search for more papers by this author
  • Limin Zheng

    Corresponding author
    1. State Key Laboratory of Oncology in Southern China, Cancer Center, Guangzhou, P.R. China
    2. State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-Sen University, Guangzhou, P.R. China
    • School of Life Sciences, Sun Yat-Sen University, Guangzhou 510 275, P.R. China
    Search for more papers by this author
    • fax: 86-20-84112169


  • Potential conflict of interest: Nothing to report.

Abstract

Although cancer patients exhibit a generalized immunosuppressive status, substantial evidence indicates that the inflammatory reaction at a tumor site can promote tumor growth and progression. Hepatocellular carcinoma (HCC) is usually derived from inflamed cirrhotic liver with extensive leukocyte infiltration. We recently found that proinflammatory T helper (Th)17 cells are accumulated in HCC tissue, where they promote disease progression by fostering angiogenesis. Here we show that interleukin (IL)-17-producing cells were enriched predominantly in peritumoral stroma of HCC tissues, and their levels were well correlated with monocyte/macrophage density in the same area. Most peritumoral CD68+ cells exhibited an activated phenotype. Accordingly, tumor-activated monocytes were significantly superior to the suppressive tumor macrophages in inducing expansion of Th17 cells from circulating memory T cells in vitro with phenotypic features similar to those isolated from HCCs. Moreover, we found that tumor-activated monocytes secreted a set of key proinflammatory cytokines that triggered proliferation of functional Th17 cells. Inhibition of monocytes/macrophages inflammation in liver markedly reduced the level of tumor-infiltrating Th17 cells and tumor growth in vivo. Conclusion: The proinflammatory Th17 cells are generated and regulated by a fine-tuned collaborative action between different types of immune cells in distinct HCC microenvironments, and allows the inflammatory response of activated monocytes to be rerouted in a tumor-promoting direction. Selectively modulating the “context” of inflammatory response in tumors might provide a novel strategy for anticancer therapy. (HEPATOLOGY 2009.)

Hepatocellular carcinoma (HCC) is characterized by progressive development, high postsurgical recurrence, and extremely poor prognosis. The dismal outcome has been attributed to the highly vascular nature of HCC, which increases the propensity to spread and invade into neighboring or distant sites.1

Tumor progression is now recognized as the product of evolving crosstalk between different cell types within tumors.2, 3 HCC is usually present in inflamed fibrotic and/or cirrhotic liver with extensive leukocyte infiltration. Thus, the immune status at a tumor site can largely influence the biologic behavior of HCC.1, 4 Recent studies have shown that high infiltration of intratumoral regulatory T cells is associated with reduced survival and increased invasiveness in HCC.5 These findings are in accordance with the general view that the tumor microenvironment induces tolerance. However, there is substantial evidence that the inflammatory response associated with cancers can also promote tumor progression by stimulating angiogenesis and tissue remodeling.4, 6

Macrophages (Mψ) constitute a major component of the leukocyte infiltrate in tumors. These cells are derived from circulating monocytes, and, in response to environmental signals, they acquire special phenotypic characteristics with diverse functions.7–9 In a previous study, we found that tumor environments can alter the normal development of Mψ that is intended to trigger transient early activation of monocytes in the peritumoral region, which in turn induces formation of suppressive Mψ in cancer nests.8 Notably, the density of these activated monocytes is selectively associated with vascular invasion and poor prognosis in HCC patients.10 These results strongly indicate that, besides inducing immune tolerance, tumors may also reroute the proinflammatory immune response into a tumor-promoting direction, although the relative mechanism remains largely unknown.

A subset of interleukin (IL)-17-producing CD4+ T helper 17 (Th17) cells with potent proinflammatory properties has recently been detected in human tumors.11, 12 Studies in other systems have found that several key cytokines, including IL-1β, IL-6, IL-23, tumor necrosis factor alpha (TNF-α), and transforming growth factor beta (TGF-β), can create a cytokine milieu that regulates the expansion of human Th17 cells.13–17 These cytokines are often present in environments that have the potential to promote the incidence and growth of tumors.18–20 Despite the implications of Th17 cells indicated by the mentioned findings, very little is known about the nature and regulation of Th17 in human tumors.

In a recent study,21 we observed that Th17 cells were highly enriched in HCCs and their levels were positively correlated with microvessel density in tissues and poor survival in HCC patients. In contrast to the classical Th17 cells that hardly express interferon (IFN)-γ, almost half of the IL-17-producing CD4+ T cells we isolated from HCC tissues were able to simultaneously produce IFN-γ, suggesting that the tumor microenvironment can profoundly determine the phenotype of such cells. Inasmuch as monocytes/Mψ represent an abundant population of antigen-presenting cells (APCs) in solid tumors and their density is inversely associated with the prognosis in HCC,8, 10 we investigated whether monocytes/Mψ can regulate Th17 and, if so, how they exert that influence, paying particular attention to the tissue microlocalization and phenotype of these cells in HCC.

Abbreviations

Ab, antibody; APCs, antigen-presenting cells; CCM, conditioned medium from control (untreated) monocytes; HCC, hepatocellular carcinoma; IFN, interferon; IL, interleukin; Mψ, macrophage(s); TAM, tumor-associated Mψ; TCM, conditioned medium from TSN-exposed monocytes; TGF, transforming growth factor; Th, T helper; TIL, tumor-infiltrating leukocyte; TNF, tumor necrosis factor; TSN, tumor culture supernatant.

Patients and Methods

Patients and Specimens.

Detailed information about the patients and specimens is described in the Supporting Materials and Methods and Supporting Table 1.

Isolation of Mononuclear Cells from Peripheral Blood and Tissues.

Peripheral leukocytes were isolated by Ficoll density gradient centrifugation.8, 22 Isolation of tumor-infiltrating leukocytes (TILs) are detailed in the Supporting Materials and Methods.

Cell Lines and Preparation of Tumor Culture Supernatant (TSN).

Detailed information about cell lines and preparation of TSNs is described in the Supporting Materials and Methods.

Isolation of Monocytes and Preparation of Conditioned Media.

Monocytes were selected from peripheral blood mononuclear cells using anti-CD14 magnetic beads (Miltenyi Biotec). To generate conditioned media, monocytes were left untreated or cultured for 1 hour with 20% TSN from HepG2 cells and then washed and cultured in RPMI 1640 containing 10% human AB serum for 16 hours. Thereafter, the supernatants were harvested, centrifuged, and stored in aliquots at −80°C.

Enzyme-Linked Immunosorbent Assay (ELISA).

Concentrations of cytokines were determined using ELISA kits (eBioscience, San Diego, CA).

In Vitro T Cell Culture System.

In 2-day incubation, purified CD3 T cells, naive T cells, and memory T cells (Miltenyi Biotec) were left untreated or were cocultured with autologous monocytes or Mψ or were exposed to 50% conditioned medium or medium supplemented with recombinant IL-6, IL-23, and/or IL-1β (Peprotech) in the presence of 2 μg/mL anti-CD3 and 1 μg/mL anti-CD28. Thereafter, cells were washed and maintained in RPMI medium supplemented with 20 IU/mL IL-2 for indicated times with conditioned medium or different cytokines. In some experiments, cells were labeled with carboxyfluorescein succinimidyl ester (CFSE) or pretreated with a neutralizing Ab against IL-6, IL-23, or IL-1β, or a control immunoglobulin G (IgG) (R&D Systems), and subsequently exposed to conditioned media.

Flow Cytometry.

Detailed information about flow cytometry is described in the Supporting Materials and Methods and Supporting Table 2.

Immunohistochemistry and Immunofluorescence.

Paraffin-embedded or frozen HCC samples were processed for immunohistochemistry or immunofluorescence, respectively, as described in the Supporting Materials and Methods.

Evaluation of Immunohistochemical Variables.

The evaluation of immunohistochemical variables is detailed in the Supporting Materials and Methods.

Immunoblotting.

The proteins were extracted as described10 and as detailed in the Supporting Materials and Methods.

In Vivo Mψ Inhibition.

All mouse procedures were performed as described in the Supporting Materials and Methods.

Statistical Analysis.

The statistical analysis is detailed in the Supporting Materials and Methods.

Results

Correlations Between Monocyte/Mψ and IL-17-Producing Cell Density in Peritumoral Stroma of HCC Patients.

APCs are critical for initiating and maintaining tumor-specific T-cell responses. Because Mψ markedly outnumber other APCs in tumors,8, 23 we first investigated the association between monocytes/Mψ and IL-17-producing cells in human HCCs, paying particular attention to the tissue microlocalization of the cells. The presence of IL-17+ cells was visualized by immunohistochemical staining of IL-17 in paraffin-embedded tissues from 106 untreated HCC patients. As shown in Fig. 1A, such cells were present throughout the tissue, but often predominantly in the peritumoral stroma rather than in the cancer nests (43.2 ± 4.9 and 10.5 ± 1.2 cells/field, respectively; n = 106 for both). The numbers of IL-17+ cells in both peritumoral tissue and stroma were significantly increased and correlated with disease progression in HCC patients.

Figure 1.

IL-17-producing cells occurred predominantly in the peritumoral stroma of HCCs and were associated with monocyte/Mψ density. (A) Density of IL-17+ lymphocytes in peritumoral tissues, peritumoral stroma, and intratumoral tissues (n = 73 for stage I and n = 33 for stage II-IV HCC patients). (B) Representative expression of IL-17 in peripheral blood and tumor-infiltrating T cells from HCC patients (n = 55 for blood and n = 30 for tumor samples). (C) Correlations between densities of monocytes/Mψ and IL-17-producing cells in peritumoral stroma of HCC patients (n = 101). Immunohistochemistry was performed using anti-CD68 and anti-IL-17 Abs, and Adobe Photoshop was used to convert the immunohistochemical staining to fluorescence. The micrographs at higher magnification show the stained peritumoral stroma (1) and cancer nest (2). (D) Representative distribution of CD68+ (brown) and HLA-DR+ (red) cells in HCC samples (n = 34). The micrographs at higher magnification show the stained cancer nest (1) and peritumoral stroma (2).

To identify the phenotypic features of tumor IL-17+ cells, we used flow cytometry to analyze leukocytes freshly isolated from tissues obtained from 30 HCC patients undergoing surgery. The results showed that the levels of IL-17+ cells were significantly higher in tumors (7.6% ± 1.6%) than in nontumoral liver tissues (2.8% ± 0.7%) and peripheral blood (0.7% ± 0.1%; n = 55; P < 0.0001 for both). Most tissue of IL-17+ T cells (81.7% ± 8.8%) were CD4+ and appeared to be Th17 cells. Interestingly, over 40% of tumor Th17 cells were able to produce both IL-17 and IFN-γ (Fig. 1B). By contrast, most of the circulating Th17 cells did not express IFN-γ (Fig. 1B). The differences in phenotypes between circulating and tumor Th17 cells indicate that the tumor environment can promote the expansion/differentiation of Th17 cells in situ.

Mψ are also predominantly found in the peritumoral stroma,8, 23 and hence we examined the correlation between densities of Mψ and Th17 cells in serial sections of HCCs stained for CD68 (marker for monocytes/Mψ) and IL-17. In peritumoral stroma we found a significant correlation between the levels of CD68+ cells and IL-17+ lymphocytes (r = 0.845, P < 0.001). However, there was no such correlation in intratumoral tissue (Fig. 1C and Supporting Fig. 1), suggesting that Mψ in different parts of a tumor play disparate roles in Th17 cell expansion. Furthermore, most of the CD68+ cells in peritumoral stroma had a smaller volume and showed marked expression of human leukocyte antigen DR (HLA-DR) (Fig. 1D), which implies that they were newly recruited and activated monocytes. In contrast, most Mψ in the cancer nests were negative for HLA-DR. Confocal microscopic analysis showed that most CD68+ cells in HCC tumors were also stained positive for CD14, but negative for CD3 (Supporting Fig. 2).

Tumor-Activated Monocytes Promote Expansion of IL-17-Producing Cells from Memory T Cell Population.

We recently observed that tumor environments can alter the normal development of Mψ that is intended to trigger transient early activation of monocytes in the peritumoral region.8 To investigate whether such a mechanism is also responsible for the selective accumulation of Th17 in peritumoral stroma, we purified circulating monocytes (CD14high cells), nontumor-and tumor-infiltrating monocytes from the same HCC patient, and then cultured those cells with purified autologous T cells. The result showed that tumor monocytes expressed higher levels of HLA-DR, secreted larger amounts of inflammatory cytokines IL-1β, IL-6, and IL-23 (Fig. 2A,B), and were more potent in promoting the expansion of Th17 cells with phenotypic features similar to those isolated from HCCs (Fig. 2C).

Figure 2.

Activated monocytes from HCC tissues induced expansion of IL-17-producing cells. Circulating monocytes (Blood), nontumor-infiltrating monocytes (Liver), and tumor-infiltrating monocytes (Tumor) were purified from the same individual with HCC, and then cultured with purified autologous circulating T cells at a ratio of 1:5 in the presence of anti-CD3/CD28 Abs as described in Patients and Methods. (A,B) Expression of surface HLA-DR and production of cytokines by CD14+ cells were determined by flow cytometry and ELISA, respectively. (C) Expression of IL-17 and IFN-γ in CD4+ T cells were measured by flow cytometry. The flow cytometry data are representative of three separate experiments, and the data on cytokine production represent the mean ± standard error of the mean (SEM) of three experiments; **P < 0.01 compared with nontumor-infiltrating monocytes.

To further elucidate the effect of tumor monocytes/Mψ on Th17 expansion, we incubated monocytes with TSN from hepatoma cells to generate tumor-activated monocytes or immunosuppressive tumor-associated Mψ (TAMs) (Fig. 3A),8 and then cultured those cells with purified autologous T cells. The results showed that tumor-activated monocytes were significantly superior to TAMs in inducing Th17 expansion, whereas the effects of control monocytes or Mψ were negligible. Notably, almost half of the Th17 cells generated from tumor-activated monocytes were able to produce both IL-17 and IFN-γ (Th17/Th1), whereas most of the TAM-induced Th17 cells were negative for IFN-γ (Fig. 3B).

Figure 3.

TSN-activated monocytes induced expansion of IL-17-producing cells. (A-C) Purified monocytes were left untreated (black line) or incubated with 20% TSN (gray line) from HepG2 cells for 1 hour or 6 days to generate different monocytes/Mψ (A), and then cultured for 9 days with autologous T cells (B), or purified autologous naive or memory T cells (C) as described in Patients and Methods. Expression of surface markers in monocytes/Mψ or expression of IL-17 in T cells was determined by flow cytometry. Med, medium; MO, monocytes; TSN-MO, tumor-activated monocytes; Mψ, macrophages; TAM, tumor-associated macrophages. (D) Monocytes were left untreated or were stimulated with 20% TSN for 1 hour or 6 days, and then washed and cultured for an additional 16 hours. Production of cytokines was determined by ELISA. (E) Monocytes were stimulated with 20% TSN for 1 hour in the presence or absence Pep-1 (200 μg/mL) or control peptide (Cpep), and then washed and cultured alone for an additional 16 hours or cocultured for 9 days with autologous T cells as described in Patients and Methods. In parallel, some untreated monocytes were incubated for 1 hour with TSN from mock or SiHAS2 cells. Production of cytokines at 16 hours in monocytes and expression of IL-17 in T cells on day 9 were determined by ELISA and flow cytometry, respectively. The flow cytometry data are representative of at least four separate experiments, and the data on cytokine production represent the mean ± SEM of six experiments; **P < 0.01 compared with normal monocytes or Mψ.

We used commercial kits to further purify naive and memory T cells. Tumor-activated monocytes effectively promoted the expansion of Th17 cells (45.4% ± 7.2%, n = 4) from memory CD4+ T cells, and most of these memory IL-17+ cells (65.7% ± 11%, n = 4) were also able to produce IFN-γ (Fig. 3C). These results indicate that, compared to immunosuppressive TAMs, monocytes activated by a short exposure to a tumor environment play a more prominent role in the development of memory T cells into IL-17-producing T cells that have phenotypic features similar to those of tumor-infiltrating Th17 cells.

Recent studies have shown that proinflammatory cytokines released by activated APC can facilitate the differentiation and expansion of Th17.13–15, 24 Therefore, we examined cytokine profiles of TSN-exposed monocytes/Mψ at an early and a late stage of differentiation. Consistent with the results from tumor-infiltrating monocytes (Fig. 2B), TSN-exposed monocytes secreted significant amounts of TNF-α, IL-1β, IL-6, IL-12, IL-23, and IL-10, and that pattern was dramatically reduced in Mψ that had been incubated in TSN for 7 days, with the exception of IL-6 and IL-10 (Fig. 3D). In contrast, supernatant from normal liver cells (L02) did not affect the cytokine production in monocytes (data not shown). Furthermore, our previous study showed that hyaluronan fragments constitute a common factor produced by several types of human tumors, including hepatoma, to stimulate the activation of monocytes.8 Here we found that the production of proinflammatory cytokines by monocytes and the expansion-promoting effect of monocytes on Th17 cells was significantly impaired when monocyte activation was attenuated either by adding a hyaluronan-specific blocking peptide (Pep-1) or by silencing hyaluronan synthase 2 in tumor cells to reduce hyaluronan levels in TSN (Fig. 3E).8, 22

Conditioned Medium from Tumor-Activated Monocytes Triggered Proliferation of Functional Th17 and Th17/Th1 Cells and Induced Expression of Retinoic Acid-Related Orphan Receptor gammat (RORγt) and T Box Transcription Factor Expressed in T Cells (T-bet).

Our next endeavor was to determine whether soluble factors secreted from TSN-activated monocytes could suffice to induce the expansion of Th17 and Th17/Th1 cells. Purified T cells were cultured in conditioned medium from control monocytes (CCM) or in conditioned medium from TSN-activated monocytes (TCM). We found that TCM, but not CCM, effectively induced the development of both Th17 and Th17/Th1 cells in a time-dependent manner that reached a maximum or a plateau within 9 days (Fig. 4A). Such generation of Th17 cells was associated with a parallel reduction in Th2 cells (Fig. 4B). To determine the proliferation of Th17 cells, we labeled T cells with CFSE and then cultured them in one of the conditioned media. As the T cells proliferated, the frequency of Th17 cells exposed to TCM gradually increased and reached a maximum on day 9; in contrast, only a small percentage (4.1% ± 0.6%, n = 4) of the cells treated with CCM were Th17 cells on day 9 (Fig. 4C). In agreement with that, we found that TCM elicited robust production of IL-17 and IFN-γ by T cells (Fig. 4D). Taken together, these results suggest that activated monocytes play a critical role in maintaining functional Th17 and Th17/Th1 pools in tumor environments in humans.

Figure 4.

Conditioned medium from TSN-activated monocytes triggered proliferation of functional Th17 and Th17/Th1 cells, activation of STAT1 and STAT3, and induction of RORγt and T-bet expression. Purified T cells were left untreated or labeled with CSFE, and they were subsequently cultured for the indicated times with conditioned medium from autologous monocytes (CCM) or corresponding medium from TSN-activated monocytes (TCM), as described in Patients and Methods. (A-C) Intracellular cytokines of CD4 cells were detected by flow cytometry. (D) Cytokines present in culture supernatants before cocktail stimulation were determined by ELISA. (E) The indicated proteins were measured by immunoblotting. The flow cytometry and immunoblotting data are representative of at least fpir separate experiments, and the data on cytokine production represent mean ± SEM of six experiments.

In both mice and humans, phosphorylation of signal transducer and activator of transcription 3 (STAT3) and induction of RORγt expression are essential for Th17 development, whereas STAT1 and T-bet are selective for Th1.13, 15, 25 Accordingly, we performed immunoblotting to determine whether those molecules were also involved in the Th17 and Th17/Th1 expansions observed in our study. Although activation of STAT1 and expression of T-bet gradually increased over time in both the CCM and TCM culture systems, expression of RORγt and phosphorylation of STAT3 were markedly up-regulated in T cells exposed to TCM (Fig. 4E). The coexistence of RORγt and T-bet proteins in T cells in situ was further confirmed by confocal microscopic analysis of frozen tumor tissues (Supporting Fig. 3) and TCM-cultured T cells (data not shown).

Distinct Roles of IL-6, IL-23, and IL-1β in Expansion of Th17 and Th17/Th1 Cells.

TSN-activated monocytes secreted several key cytokines, including IL-1β, IL-6, IL-23, and TNF-α, which have been shown to regulate the development of Th17 cells.12–17 To evaluate the contribution of those molecules to the expansion of human Th17 and Th17/Th1 cells, we initially used a specific neutralizing Ab that effectively abolished the role of IL-1β, IL-6, IL-23, or TNF-α in our conditioned medium coculture systems (Supporting Table 3). As expected, Ab blockade of IL-1β, IL-6, or IL-23 efficiently inhibited the generation of IL-17+ CD4 cells by 30% to 70%. Furthermore, a combination of the three mAbs almost completely abrogated the induction of such cells (Fig. 5A and Supporting Fig. 4), whereas the anti-TNF-α mAb had no effect (data not shown). Phenotypic analysis revealed a distinct role of these cytokines in inducing Th17 and Th17/Th1 subsets. Blockade of IL-1β had the most potent inhibitory impact on induction of both IFN-γ-negative and IFN-γ-positive IL-17-producing T cells, whereas abolishment of IL-23 mainly affected the IFN-γ-negative Th17 cells (Fig. 5A). Those findings concur with measurements of cytokines in the culture systems, which revealed that IL-1β was necessary for the induction of both IFN-γ and IL-17, whereas IL-23 influenced the production of IL-17 but not IFN-γ (Fig. 5B and Supporting Fig. 4).

Figure 5.

Distinct effects of IL-6, IL-23, and IL-1β on induction of Th17 and Th17/Th1 cells. (A,B) Purified T cells were left untreated or were cultured for 9 days in the presence of conditioned medium from TSN-activated monocytes and an mAb against IL-6 (40 μg/mL), IL-23 (10 μg/mL), or IL-1β (10 μg/mL), or all three of those mAbs combined, or a control Ab (IgG1, 40 μg/mL), as described in Patients and Methods. (C,D) Purified T cells were left untreated or were cultured for 9 days with IL-6 (10 ng/mL), IL-23 (1 ng/mL), and/or IL-1β (10 ng/mL) as described in Patients and Methods. The cytokines within CD4 cells (A,C) and those in culture supernatant before cocktail stimulation (B,D) were determined by flow cytometry and ELISA, respectively. The illustrated results represent the mean ± SEM of four separate experiments; *P < 0.05 and **P < 0.01 compared with control Ab-treated T cells (A,B) and untreated T cells (C,D).

To further elucidate the roles of IL-6, IL-23, and IL-1β in Th17 and Th17/Th1 inductions, we added recombinant human cytokines to the culture systems at concentrations similar to their levels in TCM. In support of the above-mentioned findings, IL-1β effectively increased the frequency of both Th17 and Th17/Th1 subsets, and IL-23 selectively induced the expansion of IFN-γ-negative Th17 cells (Fig. 5C). Similar results were also obtained when we measured the production of IFN-γ and IL-17 in the cultures (Fig. 5D). Notably, the combination of IL-1β, IL-6, and IL-23 elicited marked production of IL-17 at a level comparable to that exhibited by T cells in response to TCM (Fig. 5B,D).

Inhibiton of Monocyte/Mψ Inflammation Decreased Tumor IL-17+ CD4 Cells in Mice.

To test the role of monocyte/Mψ in generating IL-17+ CD4 cells in vivo, C57BL/6 mice-derived hepatoma (Hepa1-6) tissue was inoculated under the liver envelope for 5 days, after which the mice were left untreated or were injected with GdCl3 to inhibit the monocytes/Mψ inflammation.26–27 As shown in Fig. 6A, the percentage of Th17 cells in T-cell populations was significantly higher in hepatoma tissues (9.3% ± 2.5%, n = 8) than in normal liver tissues (0.7% ± 0.1%; n = 8). Inhibition of monocytes/Mψ inflammation in hepatoma-bearing mice reduced the number of tumor Th17 cells by about 80%, and it also caused a marked reduction of tumor growth (Fig. 6B). In contrast, treatment with GdCl3 did not affect the proportion of Th17 cells in peripheral blood or liver tissues from control mice (Fig. 6A). Of note, using GdCl3in vitro had no direct effect on tumor cell function or cytokine-mediated Th17 expansion (data not shown). Therefore, these findings suggest that tumor-infiltrating monocytes/Mψ might regulate the accumulation of IL-17+ cells and the progression of cancer in the tumor-bearing host.

Figure 6.

Inhibition of monocytes/Mψ inflammation in hepatoma-bearing mice reduced levels of tumor-infiltrating IL-17+ CD4 cells and tumor growth. Hepa1-6 hepatoma tissue was inoculated under the liver envelope of C57BL/6 mice for 5 days. Thereafter, the mice were left untreated (Mψ+) or were injected with GdCl3 (Mψ) every 3 days to inhibit the Mψ inflammation. After 15 days the percentages of circulating and tissue Th17 cells from normal or tumor-bearing mice were ascertained by flow cytometry (A), and tumor volume was measured using calipers (B). The results are representative of at least eight different mice; **P < 0.01 compared with untreated tumor-bearing mice.

Phenotypic Characteristics of Human Th17 and Th17/Th1.

Because the tumor-activated monocytes and control cells differed with regard to their capacity to induce expansion of Th17 and Th17/Th1, we next examined the phenotypic features of these two subsets. In general, Th17 and Th17/Th1 shared similar phenotypic features, except for slightly higher expression of chemokine receptor 4 (CCR4) and CCR6 in the former and higher TNF-α in the latter (Fig. 7 and Supporting Fig. 5). Most of the cells exhibited a CD45RO+CD62LCCR7 effector memory phenotype with substantial expression of CCR4 and CCR6, which is consistent with the general view about Th17. Analysis of immune modulatory molecules on Th17 and Th17/Th1 cells revealed that most of the cells showed extensive expression of the activation markers HLA-DR and CD25, as well as several molecules such as PD-1, CTLA-4, and GITR, which are known to be expressed on activated T cells to suppress the antitumor T cell immunity (Fig. 7 and Supporting Fig. 5). Moreover, a remarkable portion of these cells expressed the proinflammatory cytokines IL-22 and TNF-α, but not the antiinflammatory IL-4 or IL-10, which supports the proinflammatory properties of IL-17-producing cells.13, 28, 29 Similar phenotypic features were also found in Th17 and Th17/Th1 cells isolated from HCC tissues (Ref.21 and data not shown), which indicates that both these T-cell subsets are permanent residents in such tissue and that they undergo full activation and express molecules to suppress antitumor T cell-responses.

Figure 7.

Phenotypic features of human Th17 and Th17/Th1 cells. Purified T cells were cultured for 9 days in conditioned medium from either autologous monocytes (A) or TSN-activated monocytes (B), as described in Patients and Methods. The phenotypic characteristics of Th1 (IFN-γ+IL-17), Th17 (IFN-γIL-17+), and Th17/Th1 (IFN-γ+IL-17+) cells were determined by flow cytometry. The results represent the mean ± SEM of four separate experiments; *P < 0.05 compared with Th1 cells.

Discussion

Although cancer patients exhibit a generalized immunosuppressive status, there is substantial evidence that the inflammatory reaction at a tumor site can foster growth and progression of the tumor.4, 18, 19 In the present study we observed that IL-17-producing cells were enriched predominantly in peritumoral stroma, and their levels were well correlated with the density of monocytes/Mψ in the same area. Most of these CD68+ cells exhibited an activated phenotype, and, accordingly, tumor-stimulated monocytes effectively promoted in vitro expansion of Th17 cells displaying phenotypic features similar to those seen in such cells isolated from HCCs. These findings suggest an intricate mechanism in which Th17 cells in humans are generated and regulated by a fine-tuned collaborative action between different types of immune cells in distinct tumor microenvironments.

Human tumor tissues can be classified anatomically into areas of intratumoral and peritumoral stroma, each with distinct compositions and functional properties.4, 8, 30 Intratumoral environments usually contain abundant immunosuppressive molecules and cells to evade immune recognition.31 In contrast, peritumoral stroma contains a significant number of infiltrated leukocytes, which are thereby situated close to the advancing edge of a tumor.8, 9, 22 In the current study we observed that Th17 cells were present primarily in the peritumoral stroma, and they were colocalized with monocytes/Mψ that exhibited an activated phenotype. However, the number of these peritumoral Th17 cells was correlated with disease progression, and we found that these tumor-activated monocytes could not stimulate antitumor T-cell responses in HCC patients.32 These observations suggest that the proinflammatory responses in peritumoral stroma may not represent the host reaction to the malignancy, but that they instead constitute effects that are rerouted in a tumor-promoting direction to induce tissue remodeling and angiogenesis. This notion is supported by our recent investigations in which we found that the frequency of tissue Th17 cells was positively correlated with microvessel density in tumors, and high numbers of monocytes in peritumoral stroma were selectively associated with vascular invasion and poor prognosis in HCC patients.10, 21 Consistent with our observations, recent studies have shown that IL-17 could recruit neutrophils, which in turn stimulate angiogenesis and tissue remodeling.33–34

Despite recent advances in understanding the differentiation of Th17 cells in humans,15–19 little is known about the mechanisms underlying the regulation of Th17 cells in tumors. The present investigation provides evidence that proinflammatory cytokines released by tumor-activated monocytes/Mψ play a dominant role in the development of Th17 cells in HCCs, as indicated by the results of four sets of experiments. First, we observed that the level of Th17 cells was about 4 times higher in peritumoral stroma than in cancer nests, and there were significant correlations between the densities of Th17 and HLA-DRhighCD68+ cells in peritumoral stroma, which was not the case in cancer nests, where most of the CD68+ cells were negative for HLA-DR. Second, tumor-activated monocytes were significantly superior to the suppressive TAMs in inducing expansion of Th17 cells exhibiting phenotypic features more similar to those of tumor-infiltrating Th17 cells (e.g., a remarkable proportion of Th17/Th1). Third, blocking a set of cytokines released from tumor-activated monocytes clearly inhibited the generation of Th17 cells, and relatively low concentrations of the recombinant cytokines could mimic the stimulatory effect of TCM culturing in this regard. Fourth, inhibition of monocytes/Mψ inflammation in hepatoma-bearing mice markedly reduced the number of tumor Th17 cells and tumor growth. Therefore, activation of monocytes in tumors may represent a novel route to promote Th17 expansion in human cancer. This concept is supported by studies showing that activated APCs are involved in the differentiation and expansion of Th17 cells and thereby also in Th17-mediated chronic inflammation.16, 35, 36 It should be noted that, in addition to the local expansion of Th17 cells, migration from blood is also a potential source for the increased Th17 cells in tumors. In this context, we have recently found that CCR6 is expressed in the majority of Th17 cells and that CCL20, the ligand for CCR6, is significantly increased in HCCs.21

In one of our latest studies21 we observed that most of the Th17 isolated from HCCs exhibited a CD45RO+CD62LCCR7 effector memory phenotype. Consistent with that finding, in the present experiments we found that tumor-activated monocytes promoted the development of IL-17-producing cells from memory T cells. Moreover, we identified several key cytokines secreted by tumor-activated monocytes that appeared to be responsible for the expansion of Th17. One of those molecules, IL-1β, played a dominant role in the induction of both Th17 and Th17/Th1, whereas IL-23 stimulated essentially only Th17. IL-6 was also involved in this process, although to a lesser extent. These results agree with the general view that a proinflammatory cytokine milieu facilitates the development of Th17.12–17, 37 Interestingly, we noted that comparatively low concentrations of these proinflammatory cytokines (i.e., about 1/5 to 1/10 of the levels commonly used by other researchers) effectively induced the generation of both Th17 and Th17/Th1. Other studies have also demonstrated a critical role of TGF-β in the development of human Th17,14, 15 but we did not find such a correlation in HCC tissues (data not shown). In one of the mentioned studies, it was observed that TGF-β up-regulated RORγt expression but simultaneously inhibited the ability of that molecule to induce IL-17 expression, and that such inhibition can be relieved by proinflammatory cytokines.15 Thus, the proinflammatory cytokines might be the critical determinant in triggering Th17 expansion in HCC tissues, which usually contain TGF-β produced by both tumor and stroma cells.

There is substantial evidence that it is not inflammation per se, but rather the inflammatory “context” that determines the ability of proinflammatory factors to facilitate or prevent tumor growth.4, 18, 19 Our results provide important new insights into the role of monocytes/Mψ in human tumor progression. Soluble factors derived from cancer cells can trigger transient activation of newly recruited monocytes in the peritumoral stroma8 and thereby induce the monocytes to produce a significant amount of cytokines. The IL-1β, IL-6, and IL-23 promote Th17-mediated inflammation in peritumoral stroma, whereas TNF-α and IL-10 released from tumor-activated monocytes up-regulate PD-L1 on the surface of those cells to inhibit tumor-specific T-cell immunity.32 In that way, these activated monocytes repurpose the inflammatory response away from antitumor immunity (the sword) and toward tissue remodeling and proangiogenic pathways (a plowshare). Therefore, studying the mechanisms that can selectively modulate the functional activities of monocytes/Mψ might provide a novel strategy for anticancer therapy.

Ancillary

Advertisement