Article Text
Abstract
Objective Tumour-associated macrophages (TAMs) and their alternative activation contribute greatly to the development of hepatocellular carcinoma (HCC). Tim-3 is highly expressed on macrophages and regulates macrophage functions in several conditions. However, whether Tim-3 is involved in the activation and the function of TAMs has not been reported.
Design Tim-3 expression in HCC samples was evaluated by flow cytometry, immunohistochemistry and confocal analysis. We analysed the effects of Tim-3 knockdown on macrophages in growth of H22 tumour homografts in BALB/c mice. Tim-3 interference was performed by neutralising antibody, small interfering RNA or short hairpin RNA-expressing lentivirus. Cytokine production was evaluated by reverse transcription PCR, ELISA or Cytometric Bead Array. The effects of Tim-3 interference in macrophages were examined with regard to alternative activation of macrophages and proliferation and migration of Hepa1-6 cells. Cell growth curve, colony formation and transwell assays were involved to estimate cell proliferation and migration.
Results Tim-3 expression was significantly increased in both peripheral blood monocytes and TAMs in patients with HCC. The Tim-3 expression in monocytes/TAMs strongly correlated with higher tumour grades and the poor survival of patients with HCC. Consistently, HCC conditioned medium or transforming growth factor-β fostered Tim-3 expression and the alternative activation of macrophages. Moreover, Tim-3 interference in macrophages significantly inhibited the alternative activation of macrophages and suppressed HCC cell growth both in vitro and in vivo. Blocking interleukin 6 reversed the Tim-3-mediated effects on HCC cell growth in vitro.
Conclusions Tim-3 displays critical roles in microenvironment-induced activation and protumoral effects of TAMs in HCC. Interference of Tim-3 might be great potential in HCC therapy.
- CANCER IMMUNOBIOLOGY
- KUPFFER CELL
- HEPATOMA
- MOLECULAR IMMUNOLOGY
- TGF-BETA
Statistics from Altmetric.com
Significance of this study
What is already known on this subject?
Tumour-associated macrophages (TAMs) acquire M2-like phenotype in the tumour microenvironment in hepatocellular carcinoma (HCC) and promote the tumour growth, metastasis, angiogenesis and immunosuppression.
Transforming growth factor (TGF)-β is upregulated in patients with HCC and plays a tumour-promoting role.
Tim-3, a well-described immunosuppressive molecule for T cells, is abundantly expressed on macrophages.
What are the new findings?
Tim-3 expression is significantly increased in macrophages from patients with HCC and correlates with the poor survival of patients with HCC.
TGF-β in the tumour microenvironment promotes the Tim-3 expression on macrophages and facilitates the alternative activation of macrophages via Tim-3.
Tim-3 expressed in TAMs promotes the tumour growth via nuclear factor-κB–interleukin 6 axis.
How might it impact on clinical practice in the foreseeable future?
Tim-3 might be a great potential target in immune therapy of HCC.
Introduction
Hepatocellular carcinoma (HCC) is one of the most common cancers worldwide that accounts for half a million deaths per year.1 Recent studies highlighted the important role of tumour microenvironment in the progression, invasion and metastasis of HCC.2 Immune cells, including tumour infiltrating lymphocytes (TILs) and tumour-associated macrophages (TAMs), are important stromal components of liver tumour microenvironment. Clinical results with agents bolstering T cell responses showed enhanced overall survival; however, only a small proportion of treated patients achieved durable benefits.3 ,4 This deficiency may at least partially be due to the potent immune suppressive effects produced by other infiltrating immune cells.
TAMs, the most abundant cell component of the tumour microenvironment, play leading roles in immune suppression and contribute greatly to the HCC growth and metastasis.5 Previous studies suggested that TAMs are mainly polarised to M2-like phenotype or alternatively activated in response to distinct tumour-derived signals, such as interleukin 10 (IL-10) and transforming growth factor (TGF)-β.6 ,7 In fact, TAMs were reported to express high level of IL-10, CD206, arginase (Arg)-1 and low level of proinflammatory molecules IL-12, inducible nitric oxide synthase (iNOS) and tumour necrosis factor-α (TNF-α).8 ,9 TAMs play critical roles in tumour immune suppression by interacting with stromal cells and tumour cells.10 Therefore, targeting key regulators for TAMs alternative activation presents a potential strategy favouring T cell-promoting immunotherapy.
T cell immunoglobulin and mucin-domain containing protein-3 (Tim-3) has been well recognised as a key negative regulator of T cell-mediated responses.11–13 Blockage of Tim-3 rescued the exhausted T cells in chronic virus infection.14–16 A strong correlation between Tim-3 expression and immunosuppression in TILs has been recently confirmed in tumours.17 ,18 Blockade of the Tim-3 pathway enhanced T cell-mediated antitumour immunity in HCC.19 Abundant Tim-3 expression has been demonstrated in macrophage. Previous studies suggested that Tim-3 regulates macrophages in different ways under distinct stimulations. Tim-3 negatively regulated macrophage activation in response to toll-like receptor-4 signalling,20 ,21 while on the other hand, Tim-3 induced activation of macrophages and facilitated the production of H2O2, NO for the clearance of invaded pathogens during pregnancy.22 ,23 However, the role of Tim-3 in TAMs remains completely unknown.
In the present study, we investigated the expression pattern and the role of Tim-3 in TAMs in patients with HCC. Our results showed the augmented Tim-3 expression on TAMs promoted by HCC conditioned medium or TGF-β in HCC. The increased Tim-3 expression negatively correlated with patients’ survival. Interference of Tim-3 expression in macrophages suppressed HCC growth both in vitro and in vivo, which might be because of the reduced production of IL-6 in macrophages. Our data here mark Tim-3 as the potential target in HCC immunotherapy which might both booster T cell-mediated antitumour immunity and revise macrophage-induced immunosuppression in tumour microenvironment.
Materials and methods
Human subjects and mononuclear cells isolation
Peripheral blood (group 1) or tumour (group 2) samples obtained from patients with pathologically confirmed HCC at Shandong Provincial Hospital, Shandong University, were involved for the isolation of peripheral or infiltrating mononuclear cells (see online supplementary material). Twelve of HCC blood samples (group 3) were used for the serum isolation. None of the individuals were positive for HCV or HIV, consumed excessive alcohol or received chemotherapy prior to sampling. Informed consent was obtained from all patients before the study was initiated. Tissue microarrays (group 4) purchased from KangChen Bio-tech, Shanghai, were involved in immunohistochemical staining, confocal analysis and overall survival analysis. All the data of the human subjects are summarised in online supplementary table S1.
Cell lines
Mouse HCC cell lines H22 and Hepa1-6, the mouse macrophage cell line RAW264.7 and human HCC cell line Huh-7 were maintained in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS). Human monocyte cell line THP-1 cells were cultured in 1640 with 10% fetal calf serum.
Establishment of orthotopic-transplanted mice model of liver tumour
A total of 1×106 H22 cells were injected subcutaneously into male BALB/c mice. When tumours reached approximately 1 cm3 in size 8 days later, the subcutaneous tumours were dissected and cut into pieces of around 1 mm3 and transplanted into liver parcel of other BALB/c mice (n=8). The mice were sacrificed 10 days after inoculation. The liver of these mice was harvested and pictured. The tumours were separated from the livers (approximately 2 mm distance from tumour edges). The mononuclear cells isolated from liver tumours and non-tumour liver tissues were included for flow cytometry (FCM).
Macrophage preparation and polarisation
Peritoneal macrophages (PMs) and bone marrow-derived macrophages (BMDMs) were enrolled to be polarised to M1 or M2. PMs were harvested after intraperitoneal injection with 1 mL sterile 6% starch solution for 72 h. PMs were stimulated by lipopolysaccharides (LPS) (100 ng/mL, 24 h) for M1 or IL-4 (10 ng/mL, 24 h) for M2, phosphate buffered saline as a control. For BMDMs, bone marrow cells were flushed from femurs and tibias and seeded. Then BMDMs were cultured with 20 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF), M1 or 100 ng/mL macrophage colony-stimulating factor (M-CSF), M2 for 5 days, respectively, and stimulated with LPS (100 ng/mL; 24 h for M1) or IL-4 (10 ng/mL; 24 h for M2).
HCC conditioned medium and TGF-β interference
H22 or Hepa1-6 cells were cultured in serum-free DMEM for 24 h. The supernatants were collected and used as HCC conditioned medium (HCM). For HCM treatment, freshly isolated PMs were cultured with different amount of HCM or exposed to 300 μL HCM for different time points. Different amount of HCM (100, 200 and 300 µL) was added into complete medium to reach the final volume 1300 μL. The rest of volume was made up by FBS-free DMEM. So, the percentage of the 0, 100, 200 and 300 μL HCM were 0%, 7.69%, 15.38% and 23.08%, respectively.
To exploit the role of TGF-β produced by tumour cells, TGF-β expression was knockdown in H22 cells by transfection with TGF-β small interfering RNA (siRNA; see online supplementary table S3). HCM derived from TGF-β-silenced H22 cells or control group was included for the stimulation of macrophages. For TGF-β inhibitory assay, PMs were preincubated with SB431542 (Sigma) or dimethyl sulfoxide for 24 h before the HCM stimulation.
Tim-3 interference
Tim-3 blocking antibody, Tim-3 siRNA or Tim-3 shRNA-expressing lentivirus (shTim-3-Lv) was involved to interfere Tim-3 pathway (see online supplementary material).
Analysis of tumour cell proliferation, migration and in vivo tumour growth
Hepa1-6 cells were cultured in the presence of supernatant from Tim-3-silenced M2 cells or controlled M2 cells. Cell Counting Kit-8 (CCK-8, Beyotime, China), colony formation assays and standard transwell assays were used to measure cell proliferation and migration (see online supplementary material).
Murine hepatoma homografts were prepared by subcutaneous injection of H22 cells together with homologous BMDMs infected with shTim-3-Lv at ration of 4 : 1. The tumour growth assay was performed as described in online supplementary material.
NF-κB inhibitory assay
shTim-3-Lv-infected macrophages were preincubated with 5 μg/mL NF-κB inhibitory ligand (Sigma) before further TGF-β or IL-4 stimulation.
IL-6, IL-10 and GM-CSF neutralisation assay
shTim-3-Lv or Ctr-Lv-infected macrophages were stimulated by IL-4 for 24 h and the supernatants were collected. The supernatants of macrophages were preincubated with immunoglobulin G or indicating cytokine neutralisation antibody: anti-IL-6 (abcam), anti-IL-10 (R&D), anti-GM-CSF (R&D) before added to the Hepa1-6 cells. The cell growth was analysed 3 days later by CCK-8.
Flow cytometry, ELISA and western blot
Tim-3 and surface markers of macrophages were evaluated by FCM. Cytokine production was analysed by Cytometric Bead Array (CBA) or ELISA (ebioscience). Western blots were used to assay the effects of Tim-3 on signalling pathway. Details are described in the online supplementary material.
Immunohistochemistry, immunofluorescence and confocal analysis
Immunohistochemical staining of Tim-3, CD68 and Ki-67 was performed in HCC tissue sections or H22 homografts. Immunofluorescence and confocal analysis of Tim-3 and CD68 were complete in HCC tissue microarray (see online supplementary material).
Statistical analysis
Data were reported as mean values±SEM. Cell experiments were performed in triplicate and a minimum of three independent experiments were evaluated. Differences were assessed for statistical significance by GraphPad Prism. The statistical significance of differences between groups was determined by the Student's t test and the differences between groups were estimated by the log-rank test. The p values <0.05 were considered to be significant
Results
Increased Tim-3 expression on TAMs significantly correlates with the poor survival of patients with HCC
We first analysed the expression of Tim-3 on peripheral CD14+ monocytes from patients with HCC and healthy donors. As shown in figure 1A, B, both the percentage of Tim-3+ CD14+ monocytes and the mean fluorescence intensity of Tim-3 expression on CD14+ cells from patients with HCC were significantly higher than that from healthy controls. Further analysis showed that Tim-3 expression on CD14+ monocytes correlated with tumour grades (figure 1C). CD14+ monocytes are greatly heterogeneous cell lineage and are recently classified into different subgroups. Therefore, we further compared Tim-3 expression on distinct groups of CD14+ monocytes. As shown in figure 1D, E, Tim-3 could be detected on different CD14+ monocytes, including CD14+HLA-DRlow/− myeloid-derived suppressor cells (MDSC), CD14+HLA-DR+, CD14+CD206− and CD14+CD206+ cells. Compared with CD14+HLA-DRlow/− MDSCs, regular CD14+HLA-DR+ monocytes demonstrated much higher Tim-3 expression (figure 1D). However, Tim-3 expression on CD206+CD14+ monocytes was significantly higher than that on CD206−CD14+ monocytes (figure 1E), indicating that Tim-3 mainly expressed in M2-like monocytes from patients with HCC. In addition, both Tim-3 expression on CD4+ T and CD8+ T cells from patients with HCC was much higher than that from healthy donors (see online supplementary figure S1A), which is consistent with previous report.19
To investigate the role of Tim-3 in TAMs, we subsequently compared Tim-3 expression on infiltrated macrophages freshly isolated from tumour and adjacent non-tumour tissues of patients with HCC. FCM analysis showed that Tim-3 on macrophages from tumour tissues was significantly higher than that from the paired surrounding non-cancerous tissues (figure 2A). Immunofluorescent and double immunohistochemical staining confirmed the colocalisation of Tim-3 and CD68 in HCC tissues (figure 2B; see online supplementary figure S1B). In accordance with the results of FCM, confocal analysis also demonstrated significantly increased Tim-3+CD68+ cells in tumour sections compared with paratumour sections (figure 2B, C). More interestingly, the Tim-3 expression on CD68+ macrophages was significantly correlated with the poor survival of patients with HCC (figure 2D, see supplementary table S2). All these suggested that Tim-3 expression is selectively upregulated on TAMs in HCC that might indicate poor prognosis.
Tim-3 interference in macrophages inhibits the growth of H22 homografts in BALB/c mice
To detect Tim-3 expression on TAMs in mice, we established mice orthotopic-transplanted liver tumour model (figure 3A, left). Consistent with results from clinical studies, percentage of CD11b+Tim-3+ cells isolated from H22 orthotopic homograft was significantly higher than that from non-tumour liver tissues (figure 3A). Similar results were obtained from subcutaneous H22 homografts in BALB/c mice (see online supplementary figure S1C). To further evaluate the role of augmented Tim-3 in TAMs in the HCC progression, Tim-3-knockdown macrophages together with H22 cells were subcutaneously injected into BALB/c mice as described in the Materials and methods section. As shown in figure 3B, compared with control macrophages, shTim-3-Lv-infected macrophages significantly inhibited the growth of H22 homograft. Consistently, the weight of the tumours mixed with Tim-3-knockdown TAMs was less than that with control macrophages (figure 3C). Also, Ki67 immunochemical staining in the shTim-3-Lv-treated tumour section was significantly lesser than that from control mice (figure 3D). Taken together, our data suggest that Tim-3 expression in TAMs promotes HCC growth in vivo.
TGF-β in tumour microenvironment promotes Tim-3 expression on macrophages and facilitates M2-like activation
Accumulated evidence has shown that cancer cells ‘educate’ macrophages in the microenvironment.6 ,24 To investigate whether HCC tumour microenvironment induces Tim-3 expression on TAMs, HCM was involved. As shown in figure 4A and online supplementary figure S2A and S2B, HCM increased Tim-3 expression on PMs in a dose-dependent and time-dependent manner. HCM-stimulated macrophages also displayed enhanced expression of M2 markers CD206 and Arg-1 (figure 4B, C; see online supplementary figure S2B) with decreased expression of M1 marker, TNF-α (figure 4C; see online supplementary figure S2B). Similar results were obtained from HCM derived from Hepa1-6 (see online supplementary figure S2C) and tumour lysates from H22 homografts (see online supplementary figure S2D). More importantly, compared with healthy serum, serum isolated from patients with HCC greatly enhanced Tim-3 and CD206 expression on human monocyte cell line THP-1 cells (figure 4D). All these data suggest that tumour microenvironment promotes Tim-3 expression on macrophages and facilitates the M2-like phenotype.
It has been well known that TGF-β is upregulated in patients with HCC and plays a tumour-promoting role.25 In accordance, plenty of TGF-β was detected by ELISA assay in H22 HCM (see online supplementary figure S3A; 816.0±19.68 pg/mL). Also, TGF-β was significantly higher in serum from patients with HCC than that from healthy donors (see online supplementary figure S3B). To test the potential role of TGF-β in Tim-3 upregulation in TAMs, TGF-β receptor inhibitor (SB431542) or exogenous TGF-β were involved. As shown in figure 5A, B, SB431542 significantly decreased the Tim-3 expression promoted by HCM, whereas exogenous TGF-β significantly increased Tim-3 expression in cultured PMs. To further confirm the role of TGF-β produced by liver tumour cells, TGF-β expression was silenced by siRNA in H22 cells and the Tim-3 expression was compared in macrophages stimulated with HCM derived from TGF-β-silenced H22 and that from control H22. As shown in figure 5C and online supplementary figure S3C, TGF-β silence significantly decreased HCM-induced Tim-3 expression on macrophages. Consistently, luciferase reporter assay showed that TGF-β enhanced the activity of Tim-3 promoter both in Huh7 and in RAW264.7 cells in a dose-dependent manner (figure 5D). In line with the decreased expression of Tim-3, phenotypical analysis showed that SB431542 significantly repressed HCM-induced M2 activation (downregulated CD206, Arg-1, IL-10 production but upregulated IL-12, TNF-α expression; figure 5E, F). Collectively, these results suggest that TGF-β in the HCM promotes the Tim-3 expression and M2-like polarisation of TAMs. Supportingly, preincubation of THP-1 cells with SB431542 greatly inhibited Tim-3 and CD206 expression promoted by serum from patients with HCC (figure 5G).
Interference of Tim-3 reverses the M2-like phenotypes of TAMs
Since the high consistence of Tim-3 expression with M2 phenotypes of macrophages, we subsequently explored the role of Tim-3 in macrophage alternative activation. As expected, Tim-3 expression in M2 was significantly higher than that in M1 although both M1 and M2 macrophages displayed relatively Tim-3 expression (figure 6A). Polarisation of M1 and M2 from BMDMs and PMs was further confirmed by phenotypical analysis including CD206 (figure 6B), IL-10, IL-12, Arg-1, iNOS and TNF-α (see online supplementary figure S4).
To evaluate the role of Tim-3 in polarisation of macrophages, three different approaches were involved to interfere Tim-3 pathway in macrophages, including Tim-3 blocking antibody (anti-Tim-3), Tim-3 siRNA (siTim-3) and Tim-3 shRNA-expressing lentivirus (shTim-3-Lv). As shown in figure 6C and online supplementary figure S5A, Tim-3 interference by different agents significantly weakened IL-10 secretion but enhanced IL-12 in either HCM or TGF-β-stimulated macrophages, indicating the important role of Tim-3 in TGF-β-dependant M2 polarisation. The knockdown effect of Tim-3 siRNA and shTim-3-Lv was confirmed by western blot and FCM (see online supplementary figure S5B, S5C). Similar results were obtained from reverse transcription PCR analysis. As expected, HCM upregulated Tim-3 expression and Arg-1 expression but decreased TNF-α production in macrophages in a dose dependant manner, indicating the M2 polarisation. Knockdown of Tim-3 greatly weakened both the decrease tendency of TNF-α and the increase tendency of Arg-1 promoted by HCM (figure 6D). In accordance, Tim-3 knockdown increased the activation of signal transducer and activator of transcription-1 (STAT1), which is crucial for M1 activation, but decreased the activation of STAT6, one of the critical transcription factors for M2 activation.26 ,27 (figure 6E). These data support the idea that Tim-3 is crucial for tumour-induced macrophage alternative activation.
Tumour promotes M2 polarisation of macrophages in different pathways. To address the role of Tim-3 in TGF-β-independent M2 polarisation, Tim-3 knockdown assay was introduced in LPS or IL-4-induced M1 or M2 polarisation in BMDMs. IL-10 and IL-12 in the supernatants were detected by ELISA. As expected, Tim-3 knockdown significantly increased IL-12 but decreased IL-10 secretion (see online supplementary figure S5D). These results further support the idea that Tim-3 promoted polarisation towards M2.
Tim-3 expression in M2-like TAMs promotes the growth of HCC cell lines by enhancing IL-6 production
It has been well documented that M2-like TAMs promote the HCC progression and metastasis.5 In accordance, Tim-3 knockdown in M2 macrophages significantly reduced M2-promoted cell growth and colony formation of Hepa1-6 (figure 7A, B). Transwell assays further demonstrated that interference of Tim-3 in the alternative activated macrophages greatly inhibited both the migration and the invasion of Hepa1-6 cells (figure 7C). Results of CBA showed that Tim-3 knockdown significantly increased IL-12 but decreased IL-10 production. Moreover, Tim-3 significantly enhanced the production of IL-6 and GM-CSF, which have been reported as the major protumoral cytokines produced by TAMs.5 ,27 However, IL-1β secretion was not affected by Tim-3 interference (figure 7D).
It has been reported that in macrophages from experimental autoimmune encephalomyelitis (EAE) mice Tim-3 activates NF-κB pathway,22 which is the well-known transcription controller of IL-6. We, therefore, investigated the role of NF-κB-IL-6 axis in the protumoral function of Tim-3-expressed M2 cells. As expected, the phosphorylated p65 was downregulated by shTim-3-Lv in macrophages stimulated by IL-4 or TGF-β (see online supplementary figure S6A), whereas suppression of NF-κB restored the inhibitory effects of shTim-3-Lv on IL-6 production (figure 7E), supporting that NF-κB plays a key role in Tim-3-promoted IL-6 production. Significantly, neutralisation of IL-6 greatly suppressed the cell growth in Hepa1-6 cells and weakened the growth difference in Hepa1-6 cocultured with shTim-3-Lv or Control-Lv-infected M2-like macrophage (figure 7F). Although Tim-3 knockdown also decreased the IL-10 and GM-CSF production, neutralisation of IL-10 or GM-CSF showed no effects on siTim-3-promoted HCC cell growth (see online supplementary figure S6B). Taken together, our results suggest that Tim-3 promotes IL-6 production by activating NF-κB in macrophage and therefore enhances liver cancer cell growth.
Discussion
Tim-3 has been recognised as an important immune regulator, which involves in many inflammation-related diseases. However, the roles of Tim-3 in cancer-related inflammation remain largely unknown. Here, for the first time, we demonstrated the enhanced Tim-3 expression in TAMs promoted by HCC microenvironment, which facilitates the alternative activation of macrophages and accelerates the tumour growth. Our data provide a previously unrecognised link between Tim-3 and tumour-related inflammation in HCC, which might suggest Tim-3 as a potential target for liver cancer therapy.
The direct link between Tim-3 and TAM-related protumoral inflammation in HCC is suggested by several lines in our studies. First, Tim-3 expression was significantly increased in both peripheral monocytes and TAMs in HCC and correlated with the disease progression identified by tumour grades and patients’ survival (figures 1 and 2). Second, Tim-3 fostered M2-like activation promoted by tumour micromilieu or TGF-β (figure 6). Moreover, results from Tim-3 knockdown assay in LPS/IL-4-stimulated BMDMs suggested that Tim-3 promoted polarisation towards M2 in either TGF-β-dependent or TGF-β-independent pathways (figure 5; also see online supplementary figure S5B), further supporting the important role of Tim-3 in tumour-promoted M2 polarisation. Third, blockage or knockdown of Tim-3 in macrophages significantly inhibited the secretion of cytokines, including IL-10, IL-6 and GM-CSF (figure 7), all of which has been well recognised as promoters in tumour progression,27 but promoted IL-12 production, which may enhance the antitumour immunity and prevent the cancer development. Finally, knockdown of Tim-3 in macrophages significantly repressed the macrophages-induced tumour growth both in vitro and in vivo (figures 3 and 7). Recently, Tim-3 has been demonstrated as a negative master of TILs leading to immunosuppression in HCC.20 In accordance, we detected increased Tim-3 expression on peripheral CD4+T and CD8+T cells from patients with HCC (see online supplementary figure S1A). Moreover, our data with monocytes/macrophages here point Tim-3 as the crucial factor controlling the formation of TAM-related immunosuppression in HCC. Collectively, our data and previous report strongly suggest that increased Tim-3 as a crucial factor promoting the formation of HCC related immunosuppression. Interference of Tim-3 might be great potential in HCC therapy by activating both adaptive and innate immunity.
One important question is how Tim-3 is upregulated in TAMs. Our data suggest that tumour microenvironment in HCC, especially TGF-β, fosters the Tim-3 expression in macrophages. TGF-β has been reported as a multifunctional cytokine that exerts its biological effects by regulating gene expression through its downstream Smads pathway.28 ,29 By MatInspector, we found a Smad-binding site in the 5′ region of Tim-3 (our unpublished data), indicating that TGF-β might promote Tim-3 transcription by downstream Smads. Supportingly, results of luciferase assays showed that TGF-β enhanced the Tim-3 promoter activities (figure 5D). Considering the active involvement of macrophages in many TGF-β-related diseases,30 ,31 our data here highlight the potential role of Tim-3 in multiple pathological conditions by regulating macrophage functions.
Another issue we addressed is how Tim-3 accelerates TAM-mediated HCC progression. Our data suggest the important role of NF-κB-IL-6 axis in M2 polarisation of TAMs exerting the proneoplasitic effects promoted by Tim-3. IL-6 can be produced by immune cells and tumour cells under the control of several transcription factors including NF-κB and plays an important role in mediating inflammation to liver cancer transformation.32 It activates several different pathways, including PI3K, Jak/Stat, P38, to promote cell proliferation and survival of tumour cells.33 ,34 In accordance, our results showed that IL-6 neutralisation effectively destroyed the cell growth suppression in Hepa1-6 cells promoted by Tim-3 silencing (figure 7F). NF-κB inhibition significantly blocked the decrease of IL-6 in shTim-3-Lv-infected macrophage (figure 7E).
NF-κB is the well-known master regulator of inflammatory and immune responses. Recently, its important role in maintaining the M2 phenotype and protumoral function in TAMs has been highlighted in different cancers, including HCC.35 ,36 Our results showed that Tim-3 silence decreased phosphorylated p65 expression in TGF-β or IL-4-stimulated macrophages (see online supplementary figure S6A), suggesting that Tim-3 mediates the activation of NF-κB pathway. This is consistent with the previous report in EAE mice.22 However, Tim-3 might promote TAM activation by several different mechanisms. In 2007, Weigert et al showed that induction of apoptosis in tumour cells and subsequent recognition of apoptotic debris by macrophages participates in the macrophage phenotype shift.37 Tim-3 has been identified as a phosphatidylserine receptor and involved in phagocytosis of apoptotic cells. Therefore, whether Tim-3 could also foster TAMs activation by mediating recognition of apoptotic tumour cells still needs further studies.
In conclusion, the data presented in this study suggest that TGF-β in liver tumour microenvironment enhances the transcription of Tim-3 in TAMs, which in turn promotes the alternative activation of macrophages and fosters the HCC growth by producing high level of inflammatory cytokines, including IL-6 (figure 8). Our finding here highlights the potential role of Tim-3 in the protumoral functions of TAMs in HCC and provides a new target aiming TAMs in HCC therapy.
Acknowledgments
We thank Jun Liu, Yantian Xu, Jianping Wang and Yang Yang for the collection of liver tissues. We are grateful to Kai Wang for helping us with the blood samples.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online supplement
Footnotes
WY and XL contributed equally.
Contributors WY and XL: Performed experiments, including FCM, western blots, cell proliferation and colony formation assays, IHC, luciferase assays and drafting of the manuscript. HM: Performed some reverse transcription PCR. HZ: Collected the clinical specimen for FCM. XS: Performed some western blot. XL: Performed clinical studies, participated in data interpretation and drafting of the manuscript. LG: Performed statistical analysis and supervised aspects of the study. CM: Primarily responsible for studies described here, including study concept, experimental design and critical revision of the manuscript for intellectual content; performed statistical analysis and obtained funding for studies described here.
Funding This work was supported by grants from the National Science Foundation of China (No. 91129704, 30972753), the National Science Fund for Distinguished Young Scholars (81425012), Research Fund for the Doctoral Program of Higher Education of China (RFDP) (20110131110034 and Independent Innovation Foundation of Shandong University 2014QY004-14).
Competing interests None.
Patient consent Obtained.
Ethics approval Shandong University Medical Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.