STAT3 signaling in human hepatocellular carcinoma

Cancer Letters 375 (2016) 390–399

Contents lists available at ScienceDirect

Cancer Letters j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / c a n l e t

Original Articles

Non-CSCs nourish CSCs through interleukin-17E-mediated activation of NF-κB and JAK/STAT3 signaling in human hepatocellular carcinoma Yongli Luo a,1, Zhi Yang a,1, Li Su a, Juanjuan Shan a, Huailong Xu a, Yanmin Xu a, Limei Liu a, Wei Zhu a, Xuejiao Chen a, Chungang Liu a, Jun Chen a, Chao Yao a, Feifei Cheng b, Chengcheng Zhang a, Qinghua Ma a, Junjie Shen a,*, Cheng Qian a,** a b

Institute of Pathology and Southwest Cancer Center, Southwest Hospital, Third Military Medical University, Chongqing, 400038, China School of Life Science, Zhejiang Sci-Tech University, Hangzhou, 310018, China

A R T I C L E

I N F O

Article history: Received 30 December 2015 Received in revised form 7 March 2016 Accepted 7 March 2016 Keywords: Cancer stem cell Non-cancer stem cell IL-17E IL-17RB Human hepatocellular carcinoma

A B S T R A C T

Within the cancer stem cell (CSC) niche, non-CSCs play an indispensable role in facilitating a microenvironment capable of maintaining CSC properties. Non-CSCs contribute to not only the structure and topology of the tumor microenvironment but also the maintenance of the dynamic state of CSCs. Interleukin17E (IL-17E/IL-25) is important in allergic inflammation and protection against parasitic infection. Moreover, it has also been demonstrated that IL-17E takes part in different cancers recently. Here, for the first time we demonstrate that discrepant expression of IL-17E and the IL-17 receptor B (IL-17RB) exists in Nanog positive (NanogPos) CSCs and Nanog negative (NanogNeg) non-CSCs in hepatocellular carcinoma (HCC). Moreover, we further demonstrate that IL-17E binding to IL-17RB activates NF-κB and JAK/Stat3 pathways to promote proliferation and sustain self-renewal of CSCs in HCC. Meanwhile, the beneficial effect of IL17E on NanogPos CSCs could be blocked by specific inhibitors of JAK and NF-κB signaling. All the findings indicated that non-CSC-derived secreted IL-17E binds IL-17RB on CSCs to signal via JAK/Stat3 and NFκB pathways to mediate crosstalk between CSCs and non-CSCs. Therefore, IL-17E/IL-17RB signaling represents a potential therapeutic target for treatment of HCC. © 2016 Elsevier Ireland Ltd. All rights reserved.

Introduction Hepatocellular carcinoma (HCC) remains a major cause of cancerrelated mortality globally. Like other solid tumors, HCC is thought to be hierarchically organized with growth driven by small subpopulation of undifferentiated cancer stem cells (CSCs) or termed tumor-initiating cells (TICs) [1,2]. CSCs are characterized by their self-renewal capacity, multi-lineage differentiation properties and highly oncogenic potential [3,4]. While CSCs reside in cancer stem cell niche (CSC-niche) and its fate is largely determined by extrinsic signals stemming from the tumor-microenvironment by means

Abbreviations: CSCs, cancer stem cells; non-CSCs, non- cancer stem cells; IL, interleukin; GFP, green fluorescent protein; HCC, hepatocellular carcinoma; JAK, Janus kinases; STAT, signal transducer and activator of transcription; NF-kB, nuclear factork-gene binding; shRNA, short hairpin RNA; TICs, tumor-initiating cells; RT-PCR, reversetranscriptase polymerase chain reaction; SCID, severe combined immunodeficiency; DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal bovine serum. * Corresponding author. Tel.: 86 23 68765461; fax: 86 23 68752247. E-mail address: [email protected] (J. Shen). ** Corresponding author. Tel.: 86 23 68765957; fax: 86 23 68752247. E-mail address: [email protected] (C. Qian). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.canlet.2016.03.012 0304-3835/© 2016 Elsevier Ireland Ltd. All rights reserved.

of cell–cell contact or secreted factors [5,6]. The concept of CSCs entitles cancer stem cells to the core position for maintaining tumor growth. However, the role of Non-CSCs in sustaining the tumor fate, especially for the CSCs fate, remains largely unexplored. Recent studies have revealed that interactions between different subclones of tumor cells help maintain the tumor heterogeneity [7]. Although some studies have explored the supporting roles of Non-CSCs for CSCs [8], the details of interactions between CSCs and Non-CSCs, especially in HCC, have not been well described. Our previously-established Nanog promoter fluorescence reporter strategy can easily isolate and trace both Nanog positive (NanogPos) CSCs and Nanog negative (NanogNeg) non-CSCs, which facilitates study of the relationship between CSCs and non-CSCs in HCC [9]. Here we describe the crosstalk between NanogPos CSCs and NanogNeg non-CSCs in HCC. Moreover, the interleukin-17 (IL-17) family of cytokines has emerged as a critical factor in inflammatory diseases and some cancers. In this study, we demonstrated that NanogNeg non-CSCs maintain the properties of NanogPos CSCs by promoting Interleukin-17E (IL-17E)/Interleukin-17 receptor B (IL17RB) signaling through both the NF-κB and JAK/Stat3 pathways. Our data underscore the non-CSCs back-feeding role in tumorigenicity and provide new prospect on the effects and underlying mechanisms of IL-17E on the hepatocellular carcinoma CSCs.

Y. Luo et al./Cancer Letters 375 (2016) 390–399

Materials and methods Cell culture and cell lines All cell lines were cultured in Dulbecco’s modified Eagle’s medium (DMEM; GIBCO-BRL, Grand Island, NY) supplemented with 10% heat-inactivated fetal bovine serum (FBS; GIBCO-BRL) at 37 °C under a 5% CO2 atmosphere. Human HCC cell lines Huh7 and PLC/PRF/5 were purchased from the Shanghai Cell Collection (Shanghai, China). Patient-derived primary HCC cells T1115 and T1224 were obtained from patient tumor specimens with informed consent according to protocols approved by the Institutional Review Board of the Southwest Hospital, Third Military University (Chongqing, China), as previously reported [9]. Cytokines array analysis Approximately 5 × 106 isolated NanogPos or NanogNeg cells derived from the HCC cell line Huh7 were seeded into a 10 cm dish and incubated for 4 h with DMEM + 2% FBS, followed by a change to 5 mL serum-free DMEM and further incubation for 24 h. Supernatant was collected for a cytokines array test performance (RayBiotech, Inc.Cat#:AAH-BLM-1-2(2-2)). Clone formation assay For clone formation efficiency assay, 10 cells were sorted by FACS and seeded per well in 96-well plates. After 14 days culture, clones were fixed by methanol and dye with Giemsa (Sigma-Aldrich) and clone (>50 cells) numbers were assessed microscopically. Fresh media was added every 3 days. Sphere formation assay For sphere formation efficiency assay, single NanogPos and NanogNeg cell were sorted and plated into ultra-low attachment 96-well plates (Costar, Corning Inc., Corning, NY). Each well was seeded with 10 cells. Cells were cultured in DMEM/ F12 media (Sigma-Aldrich, St. Louis, MO) with B27 supplement (GIBCO), antibiotics, 20 ng/mL epidermal growth factor (PeproTech, Rocky Hill, NJ), 20 ng/mL basic fibroblast growth factor (PeproTech), and 10 ng/mL of hepatocyte growth factor (PeproTech). Next, 1% methylcellulose (Sigma-Aldrich) was added to prevent cell aggregation, and individual spheres derived from a single cell were confirmed. After 4–5 days, equal fresh media was added. Cells were incubated for 2 weeks, and spheres of diameter >75 μm were counted. Cell proliferation assays Cells were harvested and suspended for cell counting using TC10™ Trypan Blue Dye, and a Cell Titer-Blue Cell proliferation assay was performed according to the manufacturer’s instructions (Promega). Briefly, 96-well plates were seeded with 1000 or 3000 cells/well. Every 24 h–48 h, reagent was added to each well and absorbance at 450 nm measured. Tumor formation assay Male non-obese diabetic mice with severe combined immunodeficiency (NODSCID) mice at 4–6 weeks of age were maintained in pathogen-free conditions at the animal facility of the Third Military Medical University and received humane care according to the criteria outlined in the “Guide for the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences. NanogPos and NanogNeg cells were sorted and counted by FACS. Approximately 1 × 104 cells were then resuspended in serum-free medium and mixed with Matrigel at ratio of 1:1. Cells were injected subcutaneously into NOD-SCID mice. Tumor formation was evaluated regularly after injection by palpation of injection sites. Statistical analysis All data are presented as mean ± standard deviation. When two groups were compared, Student’s t-test was used. A P-value < 0.05 was considered significant statistically and is marked with an asterisk. A P-value < 0.01 was considered highly significant statistically and is marked with a double asterisk.

Results Crosstalk between CSCs and non-CSCs promotes and maintains tumor malignancy To investigate whether an interaction between CSCs and nonCSCs exists in HCC, a Nanog promoter-driven green fluorescent protein (GFP) reporter system was used [9]. Firstly, according to the GFP expression intensity, NanogPos CSCs and NanogNeg NonCSCs were sorted from Huh7 cells by fluorescence activated cell

391

sorting (FACS). Then, the cells were divided into serial combinations: NanogPos, NanogNeg and their mixture with different proportions (NanogNeg : NanogPos = 1:49, 1:19, 1:9 and 1:1). All groups were cultured for 9 days in vitro to analyze their proliferation rate. Comparing with the NanogPos alone group, co-cultured NanogNeg non-CSCs and NanogPos CSCs groups increased cell proliferation rates by 0.7%, 13.9%, 43.5% and 26.4% at ratio of 1:49, 1:19, 1:9 and 1:1, respectively (Fig. 1A). Interestingly, we found that mixing NanogPos cells with a small amount of NanogNeg cells could significantly promote the global cells growth rate, although non-CSCs generally proliferate significantly slower than CSCs counterparts do, including these Nanog Neg non-CSCs and Nanog Pos CSCs cells [10]. Importantly, the cell proportion of 1:9 (NanogNeg :NanogPos = 1:9) displayed the greatest difference among all these groups (P < 0.01). We also noticed that adding more NanogNeg cells into NanogPos cells as the ratio of 1:1(NanogNeg/Pos=1:1) decreased the global cells proliferation compared with NanogNeg/Pos=1:9 group. That might be caused by increasing proportion of slow-growth NanogNeg cells reduced the whole group proliferation. Additionally, adding NanogPos cells into NanogNeg cells group as the ratio of 1:9 (NanogPos/Neg=1:9) also raised the global cells proliferation (data not shown). All these findings suggested that the cell proportion ratio of 1:9 represent a reasonable rate within all these tested groups, which is used for the rest of this study. Then, the mixture of NanogNeg/Pos=1:9 and NanogPos/Neg=1:9 as well as their respective counterparts NanogPos and NanogNeg cells were cultured in vitro for 11 days in Huh7 and PLC/PRF/5 cells. The data showed that co-culture of NanogPos and NanogNeg cells grow much faster than the corresponding control cells. Moreover, the proliferation rates of co-cultured groups were significant higher than the theoretical value. This value was calculated by the ratio of NanogPos and NanogNeg multiplied respective proliferation rate (Fig. 1B and Supplementary Fig. S1). Subsequently, 1 × 104 cells of these groups were subcutaneously implanted into non-obese diabetic with severe combined immunodeficiency (NOD-SCID) mice to investigate their tumor-initiating capacity in vivo. Tumor volume and weight were measured regularly and animals were sacrificed after 36 days. As shown in Fig. 1C, tumors derived from NanogNeg/Pos=1:9 or NanogPos/ Pos or Neg=1:9 groups were significantly larger and heavier than Nanog NanogNeg groups, respectively (P < 0.01; Fig. 1C). Noticeably, the tumor derived from NanogNeg/Pos=1:9 group was initiated the earliest (17 days after injection) among these groups. Taken together, our data suggest that, indeed, Nanog Neg non-CSCs and Nanog Pos CSCs interact, which results in more aggressive tumor behavior both in vitro and in vivo. IL-17E is a potential candidate for mediating the crosstalk between NanogPos CSCs and NanogNeg non-CSCs in HCC To investigate whether the interaction between NanogNeg nonCSCs and NanogPos CSCs was mediated by means of a cell–cell direct or an indirect manner. NanogPos cells were indirectly co-cultured with NanogNeg cell, separated by Millicell™ Culture Plate Inserts for 7 days (Fig. 2A), or cultured with supplied of NanogNeg cells condition medium (CM) for 9 days (Fig. 2B). Then the proliferation rates were measured. We interestingly found that under both conditions, NanogPos CSCs and NanogNeg non-CSCs cultured within mixed supernatant significantly improved the proliferation compared with both the control groups and the theoretical value (Indicated by “△”). These findings suggested that the indirect communication plays a crucial role in mediating the crosstalk between NanogPos and NanogNeg cells. We next searched for the mediator of the NanogPos CSC–NanogNeg non-CSC interaction by a cytokines-array analysis on supernatant from a 24-h culture of NanogPos CSCs and NanogNeg non-CSCs. As a result, TGF-β2, IL-17E (also known as IL-25) and MMP-13

392

Y. Luo et al./Cancer Letters 375 (2016) 390–399

Fig. 1. Crosstalk between NanogPos CSCs and NanogNeg non-CSCs promotes and maintains tumor malignancy. (A) Flow cytometry-sorted NanogPos CSCs and NanogNeg nonCSCs from the HCC cell line Huh7 according to the GFP expression intensity. NanogPos CSCs were co-cultured with NanogNeg non-CSCs at ratios of 1:49, 1:19, 1:9, and 1:1, with NanogPos CSCs alone as control. (B) The ratio of 1:9 was used to co-culture NanogPos CSCs mixed with NanogNeg non-CSCs (Left, NanogNeg/Pos = 1:9) and co-culture NanogNeg non-CSCs mixed with NanogPos CSCs (Right, NanogPos/Neg = 1:9). NanogPos CSCs and NanogNeg non-CSCs only were cultured as control. (C) Xenograft tumorigenesis assay using NOD/SCID mice injected with 1 × 104 Huh7 cells, NanogNeg non-CSCs, NanogPos CSCs, and NanogNeg/Pos = 1:9 or NanogPos/Neg = 1:9 cells. Theoretic value was calculated by the value of 10% NanogNeg adding 90% NanogPos, the value of 10% NanogPos adding 90% NanogNeg, correspondingly. All experiments were performed in triplicate. Results are presented as means ± standard deviation. Statistical significance was determined using the Student’s t-test. *P < 0.0 5, **P < 0.01 and ***P < 0.001.

expressed higher levels in NanogNeg non-CSC supernatant, while IGFBP-7, GDF-15, WIF-1, uPAR, VEGFR2, CXCL16, and TSP expressed more highly in NanogPos CSC supernatant (Fig. 2C and Supplementary Fig. S2A). Quantitative real-time PCR (RT-PCR) further verified the results in Huh7 and PLC/PRF/5 cells and the data showed that among these cytokines, IL-17E and TSP were significantly highly expressed in NanogNeg non-CSCs and NanogPos CSCs, respectively, in both of cells (P < 0.01 Fig. 2D). Considering our previous data that a small amount of NanogNeg non-CSC cells or their supernatant could have a significant impact on the proliferation of whole population (Fig. 1A and B) or NanogPos CSC cells (P < 0.05) (Fig. 2A and B). Thus, IL-17E, the cytokine primarily secreted by NanogNeg cells, might be participated in the interaction. Interlaced high expression of IL-17E and IL-17RB in NanogNeg nonCSCs and NanogPos CSCs respectively in HCC An ELISA analysis of supernatant from 24-h cultured Huh7 and PLC/PRF/5 HCC cells further confirmed that IL-17E was expressed at a significantly higher level in NanogNeg non-CSCs (P < 0.05 Fig. 3A). Moreover, currently studies have shown that Interleukin-17 (IL17A) could promote self-renewal of CD133 + CSCs [10], and Interleukin-17 receptor B (IL-17RB) could also improve tumorigenicity in breast cancer [11]. Based on our findings and the previous studies, we proposed that IL-17E is probably a candidate for meditating NanogPos CSC–NanogNeg non-CSC interaction.

IL-17E has been demonstrated to be important in allergic inflammation and protection against parasitic infection by binding with heterodimer receptor IL-17RA and IL-17RB [12,13]. Thus, we conducted RT-PCR to investigate mRNA level of the IL-17RA and IL17RB in NanogPos and NanogNeg cells. The data showed that IL17RB was significantly highly expressed in NanogPos cell compared with NanogNeg cells (P < 0.05) (Fig. 3B), while IL-17RA showed no significant difference between them (Supplementary Fig. S2B). As reported that IL-17E might be able to conduct signals by conjugating to IL-17RB in the absence of IL-17RA [14,15], thus ligand– receptor interaction of IL17E/IL-17RB more probably play a major role in intermediating the communication between NanogPos and NanogNeg cells. Increased expression of IL-17RB in NanogPos cells was further confirmed by Western Blotting tests in Huh7, PLC/PRF/5, and patient-derived primary HCC cells T1115 and T1224 (Fig. 3C). Similarly, an immunofluorescence assay demonstrated that IL-17RB was predominately localized to NanogPos CSCs (Fig. 3D). Furthermore, immunohistochemical analysis of engrafted tumors derived from NanogPos CSCs and NanogNeg non-CSCs yielded similar results (Fig. 3E). We then performed immunohistochemical analysis to assess IL17RB expression in HCC patient specimens and adjacent nontumor tissues. The data displayed that the majority of IL-17RB staining was confined to carcinoma tissues (Fig. 3F). Western blotting next confirmed that IL-17RB is more highly expressed in HCC tissue compared with adjacent normal tissue (Fig. 3G). Importantly, consistent with previous cell line results (Fig. 3B and C), the

Y. Luo et al./Cancer Letters 375 (2016) 390–399

393

Fig. 2. IL-17E is a potential candidate for mediating the crosstalk between NanogPos CSCs and NanogNeg non-CSCs in HCC. (A) Huh7 cells were cultured with Millicell™ Culture Plate Inserts for 7 days. (B) Huh7 cells were cultured with conditioned medium (CM) which was collected from the supernatant of NanogNeg non-CSCs or NanogPos CSCs cultured for 48 hours in Huh7 cells and was centrifuged before used. (C) Cytokines-array analysis of supernatant from cultured NanogPos CSCs and NanogNeg non-CSCs of Huh7 origin. (D) RT-PCR analysis of cytokine mRNA expression in Huh7 and PLC/PRF/5 cells. “△” Represent theoretical value in this Fig. The experiments were performed in triplicate, except for the cytokines-array analysis. Results are presented as means ± standard deviation. Statistical significance was determined using the Student’s t-test. *P < 0.0 5, **P < 0.01 and ***P < 0.001.

expression of IL-17RB was positively associated with Nanog in HCC patient samples as well (R = 0.92, P < 0.0001 Fig. 3H). Collectively, these findings demonstrated that IL-17E and IL-17RB is more highly expressed in NanogNeg Non-CSCs and NanogPos CSCs, respectively, also the interlaced expression of such signaling factor and its receptor may contribute to building the connection channel between NanogPos CSCs and NanogNeg Non-CSCs.

IL-17E promotes proliferation and anti-apoptotic effects in HCC via the NF-κB pathway To investigate the function of IL-17E/IL-17RB interaction, we firstly treated Huh7 and PLC/PRF/5 HCC cells with 100 or 200 ng/mL exogenous recombination human IL-17E (rhIL17E), followed by counting of cell numbers after 7 days. Exogenous IL-17E could

394

Y. Luo et al./Cancer Letters 375 (2016) 390–399

Fig. 3. Interlaced high expression of IL-17E and IL-17RB in NanogNeg non-CSCs and NanogPos CSCs respectively in HCC. (A) ELISA analysis of IL-17E levels in NanogPos CSCs and NanogNeg non-CSCs from Huh7 and PLC/PRF/5 cells. (B) IL-17RB mRNA expression in Huh7 and PLC/PRF/5 cells as determined by RT-PCR. (C) Western blotting analysis of IL-17RB protein levels in Huh7, PLC/PRF/5, and patient-derived primary HCC cells T1115 and T1224. (D) Immunofluorescence detection of IL-17RB in Huh7 and PLC/ PRF/5 cells, scale bars = 20 μm. (E) Immunohistochemical detection of IL-17RB in tumors derived from NanogPos CSCs and NanogNeg non-CSCs from Huh7 cells, scale bars = 50 μm. (F) Immunohistochemical detection of IL-17RB in tumor and adjacent non-tumor patient tissues, scale bars = 50 μm. (G) Western blotting analysis of both IL-17RB and Nanog protein levels in patient-tissues. (H) Correlation analysis showed correlative expression between IL-17RB and Nanog. Experiments were performed three times and data are shown as means ± standard deviation. *P < 0.05.

Fig. 4. IL-17E promotes proliferation and anti-apoptotic effects in HCC via the NF-κB pathway. (A) NanogPos CSCs were treated with 100 or 200 ng/mL recombinant human IL-17E (rhIL-17E) in Huh7 and PLC/PRF/5 cells. (B) Huh7 cell-growth of NanogNeg non-CSCs following knocking-down of IL-17E and NanogPos CSCs knocking-down of IL17RB. Both knocked down by lentiviral vectors expressing a short hairpin RNA (shRNA) and rhIL-17E was used to rescue. (C, D) Flow cytometry analysis of apoptosis following treatment of NanogPos CSC with 200 ng/mL rhIL-17E and CM of NanogNeg non-CSCs for 48 h by a Sorafenib-induced apoptosis model. FITC-Annexin V/Propidium Iodide dual staining was applied. Cells in the upper right (UR) portion indicate late apoptotic cells, whereas cells in the lower right (LR) portion indicate early apoptotic cells. (E) Western blotting of phosphorylated protein levels of NF-kB (p65) represented by p-NF-kB-(p65) and IκB represented by p-IkB in Huh7 and PLC/PRF/5 cells following IL-17RB silencing. Scramble was used as control. (F) Western blotting of phosphorylated protein levels of NF-kB (p65) and IκB, following treatment with NF-κB activator TNF-α and IL17E (200 ng/mL). (G) Western blotting of phosphorylated protein levels of NF-kB (p65) and IκB, following treatment with NF-κB inhibitor PDTC and IL-17E. Sorafenib 5 μM/ mL; TNF-α 100 ng/mL; IL-17E 200 ng/mL; PDTC 5 μM/mL were used. Experiments were performed three times and schematic one was shown. Data are shown as means ± standard deviation. *P < 0.05; **P < 0.01.

Y. Luo et al./Cancer Letters 375 (2016) 390–399

395

396

Y. Luo et al./Cancer Letters 375 (2016) 390–399

significantly promote CSC cells proliferation both in the 100 ng/ mL groups (P < 0.05) and 200 ng/mL groups (P < 0.01) (Fig. 4A). Next, IL-17RB and IL-17E were knocked down by lentiviral vectors expressing a short hairpin RNA (shRNA). After confirming the knocking down effects of these vectors (Supplementary Fig. S3A–D), our data showed that knocking down of either IL-17RB or IL-17E could significantly reduce cell proliferation. Noticeably, exogenous supplementation of rhIL17E could rescue the impact of silenced IL17E, but could not reverse the effects of IL-17RB knocking-down. This further demonstrated that IL-17RB is dominating receptor for transferring signal of IL-17E in HCC (P < 0.05 Fig. 4B). Previous studies have demonstrated that IL-17E binding with receptor IL-17RB could promote NanogPos cells proliferation (Fig. 4A). However, the profound mechanism of proliferation-promote effects performed by changing cell-cycle or preventing apoptosis need to be further explored. Firstly, NanogPos cells were treated with 200 ng/ mL IL-17E for 24 or 48 h, following detection with flow cytometry. No significant differences were observed in cell-cycle distribution, which consisted with the results treated NanogPos cells with CM from NanogNeg cells for 12 or 24 h (Supplementary Fig. S4A and B). Therefore, by using a Sorafenib-induced apoptosis model, we moved forward to treat cells with or without rhIL-17E for 48 h and examined the ability of NanogPos cells to resist apoptosis. NanogPos cells cultured in the presence of rhIL-17E had a lower apoptotic rate (8.78%) compared with untreated cells (20.735%; Fig. 4C). Consistently, treatment of NanogPos cells with CM from NanogNeg cells resulted in reducing the apoptosis rate from 16.08% to 6.915% in Huh7 cells and from 12.645% to 7.285% in PLC/PRF/5 cells (Fig. 4D). These anti-apoptosis effects were mirrored through a Cisplatin-induced apoptosis model (Supplementary Fig. S4C). All the data demonstrated that NanogNeg secreted IL-17E may predominantly bind to the IL-17RB of NanogPos cells, so as to promote cells proliferation. NF-κB signaling pathway has been demonstrated to modulate anti-apoptosis effect via IL-17RB in breast cancer [11,16]. We then detected NF-κB (p65) and IκB activation status after manipulating IL-17E and IL-17RB expression. Silencing IL-17RB decreased phosphorylation of both NF-κB and IκB (Fig. 4E). Consistently, exposure to IL-17E up-regulated phosphorylation of NF-κB (p65) and IκB, which reproduced the stimulate effect of TNF-α, an activator of NFkB (p65) (Fig. 4F). In addition, 5 μM of the NF-κB inhibitor pyrrolidine dithiocarbamate (PDTC) blocked these effects of IL-17E in NanogPos cells (Fig. 4G and Supplementary Fig. S5C). Together, these findings indicated that IL-17E enhances proliferation of NanogPos cells by promoting NF-κB-mediated anti-apoptosis effects.

cells or their control cells were subcutaneously implanted into NODSCID mice. Our data exhibited that either knocking down of IL17E in NanogNeg cells or IL-17RB in NanogPos cells could significantly impair the tumor development, which was represented by reducing both of tumor volume and weight (P < 0.05) (Fig. 5E). All these data suggested that IL-17E/IL-17RB signaling not only promotes proliferation by inhibiting apoptosis but also maintains the self-renewal properties of NanogPos CSCs in HCC. JAK-Stat3 signaling is involved in IL-17E-mediated stemness maintaining of NanogPos CSCs in HCC It is well known that Stat3 pathway plays a vital role in maintaining stemness of CSCs in liver cancer [17], while IL-17 family can activate Stat3 through promoting autocrine IL-6 [18]. Therefore, we proposed that IL-17E might maintain the self-renewal properties of CSCs through Stat3 pathway. In order to test this hypothesis, we conducted Western Bolt analysis and indeed found that knockingdown of IL-17RB remarkably down-regulated phosphorylated Stat3 (p-Stat3) in both Huh7 and PLC/PRF/5 cell (Fig. 6A). Consistently, supplementation with rhIL-17E could significantly up-regulate p-Stat3 level. This result was similar to the effect of the treatment of Nanog Pos cells with IL-6, the activator of Stat3 (Fig. 6B and Supplementary Fig. S5A and B). Moreover, we found that Stat3 inhibitor S31-201, Jak pan-inhibitor INCB, and Jak2 inhibitor AG490 could all effectively block Stat3 phosphorylation following IL-17E treatment (Fig. 6C). Sphere formation efficiency and clone formation efficiency analysis showed that cells treated with S31201, INCB, or AG490 exhibited similar effects to IL-17RB depletion (Fig. 6D and E). These findings indicated that IL-17E/IL-17RB could activate JAK/Stat3 pathway to maintain the stemness of NanogPos cells in HCC. To further investigate whether IL-17E-mediated Stat3 pathway through producing IL-6 [19], we assessed levels of IL-6 produced by Huh7 and PLC/PRF/5 cells in the presence or absence of IL-17E. Medium from Mesenchymal stem cells (MSCs), which is known to autocrine IL-6, served as a positive control [20,21]. However, production of IL-6 was undetectable under both conditions (Supplementary Fig. S6 and Supplementary Table S1). This result indicated that IL-17E/IL-17RB activation of the Stat3 pathway might be independent of IL-6. Taken together, we postulate that the crosstalk between NanogPos CSCs and NanogNeg Non-CSCs would be modulated by IL-17Emediated NF-kB and Stat3 pathways. Discussion

NanogNeg non-CSCs secrete IL-17E to maintain the stemness of NanogPos CSCs both in vitro and in vivo The above data have testified to the role of IL-17E/IL-17RB in fascinating the proliferation of NanogPos cells. We next questioned that whether IL-17E/IL-17RB pathway could also take part in maintaining CSCs’ properties, which generally evaluated by the capabilities of sphere and clone formation in vitro and the tumorigenicity in vivo. Our data displayed that IL-17E could significantly elevate both of sphere formation efficiency and clone formation efficiency of NanogPos cells (P < 0.05) in Huh7 and PLC/PRF/5 cells (Fig. 5A and B). Moreover, these effects were blocked by neutralizing antibody against for IL-17E (Fig. 5A and B) or by knocking down of IL-17RB (Fig. 5C and D). To investigate whether the great tumorigenic enhancement effect of NanogPos cells and NanogNeg cells mixing (Fig. 1C) is IL-17E/IL17RB involved, we knocked down both of IL-17E in NanogNeg and IL-17RB in NanogPos, and mixed them with 9-fold number of NanogPos and NanogNeg cells, respectively. Afterward, 1 × 104 of these mixture

Research targeted at the interaction between CSCs and Non-CSCs have developed a great interest and provided the potential promise to cure tumors [8,22,23]. In this study, we investigated the interplay between NanogPos CSCs and NanogNeg Non-CSCs in HCC through NanogGFP promoter fluorescence reporter system. Interestingly, we found that mixing NanogPos and NanogNeg group cells behaved more malignantly than those of NanogPos or NanogNeg alone group. Particularly, mixing a small amount of NanogNeg Non-CSCs (10%) into NanogPos CSCs (90%) could significantly enhance cells proliferation capability and tumorigenicity, in spite of the fact that NanogPos CSCs displayed more competence to proliferate in vitro and tumor develop in vivo compared with the NanogNeg non-CSCs. This “1 plus 1 is greater than 2” gainful effect drew enough attention to further explore the interaction between NanogPos CSCs and NanogNeg Non-CSCs. Continuing studies elucidate that CSCs recruit macrophages, inflammatory cells, and MSCs to construct the favorable microenvironment by means of secreting chemokines, cytokines and inflammatory factors [24]. On the other hand, the topology and

Y. Luo et al./Cancer Letters 375 (2016) 390–399

397

Fig. 5. NanogNeg non-CSCs secrete IL-17E to maintain the stemness of NanogPos CSCs both in vitro and in vivo. (A) Sphere formation efficiency analysis of NanogPos CSCs in Huh7 and PLC/PRF/5 cells following treatment with 200 ng/mL rhIL-17E and IL-17E neutralizing antibody. IgG was used as control. (B) Clone formation efficiency analysis of NanogPos CSCs in Huh7 and PLC/PRF/5 cells following treatment with 200 ng/mL rhIL-17E and IL-17E neutralizing antibody. (C, D) Sphere formation efficiency (C) and clone formation efficiency (D) analysis of NanogPos CSCs in Huh7 and PLC/PRF/5 cells after silencing of IL-17RB.A scrambled shRNA sequence served as control and rhIL-17E used to rescue. (E) Tumor growth analysis of 1 × 104 Huh7 cells and NanogPos CSCs or NanogNeg non-CSCs treated with shRNA targeting IL-17E or IL-17RB as indicated. Experiments were performed three times and data are shown as means ± standard deviation. *P < 0.05; **P < 0.01.

dynamic behavior of CSCs are sculpted by tumor microenvironment factors as well [7,25]. Moreover, CSCs can directly transform into other cells to maintain their survival [26]. Here, we firstly reveal that CSCs-differentiated NanogNeg Non-CSCs could interplay with and back-feed to NanogPos CSCs in HCC [27]. CSCs are generally determined to differentiate both in vitro and in vivo and their stemness maintaining rely on harboring with some specific growth factors or certain tumor environment [9,28]. Our findings enlighten that the differentiation event of CSCs is probably a necessary initial step to build the CSCs-niche. Consistently, adding minor NanogNeg NonCSCs to NanogPos CSCs in advance may contribute to establish the niche early. In this article, we firstly identified that the inflammatory cytokine IL-17E perform a considerable role in mediating the crosstalk between NanogPos CSCs and NanogNeg Non-CSCs in HCC. It has been reported that IL-17E was produced traditionally by CD4+ cell, CD8+ T cell, mast cell, eosinophil, epithelial cell and endothelial cells [13].

However, possible factors and mechanism for the elevated levels of IL-17E from non-CSCs in liver cancer are not clear and remain to be explored. Although IL-17E has been previously reported as an antitumor factor by recruiting B cell and eosinophil to initial immune killing effects or promoting caspase-mediated apoptosis in breast cancer [29–31]. IL-17E, like other inflammatory cytokines, also associates with tumors development and prognosis [32,33]. Thus IL17E might be a component of the CSC-niche to create the proinflammatory milieu in cancers. In addition, we investigated that IL-17E binding with IL-17RB activate NF-kB pathway so as to inhibit apoptosis and promote proliferation, although the profound mechanism is unknown. The possible explanation of the dual facets of IL-17E could largely attribute to its target cells. Moreover, the discovery a special implication of IL-17E in HCC is that it takes part in mediating CSCs self-renewal, which is by far the most significant in this study. Mountains of research have demonstrated that CSCs tend to resist to the conventional chemotherapies or radio-

398

Y. Luo et al./Cancer Letters 375 (2016) 390–399

Fig. 6. JAK-Stat3 signaling is involved in IL-17E-mediated stemness maintaining of NanogPos CSCs in HCC. (A-C) Western blot determination of Stat3 protein levels in NanogPos CSCs following: IL-17RB silencing in Huh7 and PLC/PRF/5 cells (A); treatment with Stat3 activator IL-6 and rhIL-17E (B); or treatment with Stat3 inhibitor S31-21, JAK paninhibitor INCB, or JAK2 inhibitor AG490 following treated with IL-17E(C). (D) Sphere formation efficiency analysis of NanogPos CSCs in Huh7 and PLC/PRF/5 cells after depletion of IL-17RB, treatment with AG490, INCB, or S31-21 as indicated. (E) Clone formation efficiency analysis of NanogPos CSCs in Huh7 and PLC/PRF/5 cells after depletion of IL-17RB, treatment with AG490, INCB, or S31-21 as indicated. Experiments were performed three times, and data are shown as means ± standard deviation. *P < 0.05; **P < 0.01.

therapies [34,35]. However, it might be taken into consideration whether pro-inflammatory circumstance may promote the residual CSCs to restart tumor development [32]. Thus, it might be a risky trial to use IL-17E as an antitumor agent and its oncogenic roles remain to be detailed illustrated. Evidence has demonstrated that IL-17 receptor family contains the conserved domain termed similar expression to fibroblast growth factor genes and IL-17R SEF/IL-17R (SEFIR) [36]. The SEFIR domain contains Box 1 and Box 2 motifs might interact with JAK [37]. Cytokine binding with its receptor leads to JAK and the receptor itself activates via the Box 1 and Box 2 motifs. Phosphorylation of JAK in turn initiates recruitment and activation of the downstream target STATs [37–39]. Computer simulation of the crystal structure of IL17RB has shown that IL-17RB [14,40], as other IL-17Rs family member, also contains the Box 1 and Box 2 regions (Supplementary Fig. S7A and B) [41], which indicates that IL-17E might activate Stat3 pathway through binding to its upstream activator JAKs directly (Fig. 7). However, further studies are needed to determine whether mutating the residue of Box 1 and/or Box 2 of IL-17RB will block the JAK/Stat3 signaling pathway in NanogPos liver CSCs. Notably, both the NF-KB and Stat3 pathways are persistently activated in liver cancer and these pathways could be the causes and consequences for each other [35,42,43]. Our study provides further evidence that the NF-κB and Stat3 pathways are both closely involved in HCC progression. In summary, our study showed that there is cross talk between NanogNeg Non-CSCs and NanogPos CSCs in HCC. As the component of CSC niche, NanogNeg non-CSCs nourish and support NanogPos CSCs through secreting IL-17E to activate NF-kB and JAK/Stat3 pathways so as to prevent apoptosis and maintain stemness, respectively (Fig. 7). Although NanogPos CSCs also displayed profitable effects on NanogNeg non-CSCs (Fig. 1B), we did not demonstrate the potential mechanism. Moreover, another limitation of this study is that we illustrated the indirect interaction between CSCs and non-CSCs, but whether existence of any cell–cell direct interactions needs to be further investigated. Together, we provided novel insights into a supporting role of non-CSCs to CSCs, which enriches the concept of CSCniche as well as proposes the potential therapeutic strategy for targeting of the IL-17E/IL-17RB interaction in HCC.

Acknowledgments This work was supported by funds from National Natural Sciences Foundation of China (81520108025, 81330048 to C. Qian, 81472292 to J. Shen, 81301959 to Z. Yang), National Basic Research Program of China (973 Program, No 2010CB529406 to C. Qian).

Fig. 7. Schematic shows the IL-17E/IL-17RB signaling in liver tumor cells. IL-17E secreted by NanogNeg non-CSCs binding to IL-17RB on NanogPos CSC to active IL-17E/ IL-17RB signaling. The Box 1 and Box 2 domain of IL-17RB may bind with JAK. Then JAK phosphorylate and in turn initiate recruitment of the Stat3 and subsequent phosphorylation of Stat3 proteins. The phosphorylated Stat3 translocates to the nucleus where they bind to specific DNA binding sites regulating gene transcription to maintain CSC properties of NanogPos CSCs in HCC. IL-17E binding with IL-17RB may also activate NF-κB signaling to promote Nanog Pos CSCs proliferation by an antiapoptosis effect.

Y. Luo et al./Cancer Letters 375 (2016) 390–399

Conflict of interest No potential conflicts of interest were disclosed. Appendix: Supplementary material Supplementary data to this article can be found online at doi:10.1016/j.canlet.2016.03.012. References [1] M. Shackleton, E. Quintana, E.R. Fearon, S.J. Morrison, Heterogeneity in cancer: cancer stem cells versus clonal evolution, Cell 138 (2009) 822–829. [2] A. Marusyk, K. Polyak, Tumor heterogeneity: causes and consequences, Biochim. Biophys. Acta 1805 (2010) 105–117. [3] J.E. Visvader, G.J. Lindeman, Cancer stem cells in solid tumours: accumulating evidence and unresolved questions, Nat. Rev. Cancer 8 (2008) 755–768. [4] T. Reya, S.J. Morrison, M.F. Clarke, I.L. Weissman, Stem cells, cancer, and cancer stem cells, Nature 414 (2001) 105–111. [5] J. Ye, D. Wu, P. Wu, Z. Chen, J. Huang, The cancer stem cell niche: cross talk between cancer stem cells and their microenvironment, Tumour Biol. 35 (2014) 3945–3951. [6] V. Plaks, N. Kong, Z. Werb, The cancer stem cell niche: how essential is the niche in regulating stemness of tumor cells?, Cell Stem. Cell 16 (2015) 225–238. [7] A. Marusyk, D.P. Tabassum, P.M. Altrock, V. Almendro, F. Michor, K. Polyak, Non-cell-autonomous driving of tumour growth supports sub-clonal heterogeneity, Nature 514 (2014) 54–58. [8] S. Fatrai, S.J. van Schelven, I. Ubink, K.M. Govaert, D. Raats, J. Koster, et al., Maintenance of clonogenic KIT(+) human colon tumor cells requires secretion of stem cell factor by differentiated tumor cells, Gastroenterology 149 (2015) 692–704. [9] J. Shan, J. Shen, L. Liu, F. Xia, C. Xu, G. Duan, et al., Nanog regulates self-renewal of cancer stem cells through the insulin-like growth factor pathway in human hepatocellular carcinoma, Hepatology 56 (2012) 1004–1014. [10] T. Xiang, H. Long, L. He, X. Han, K. Lin, Z. Liang, et al., Interleukin-17 produced by tumor microenvironment promotes self-renewal of CD133+ cancer stem-like cells in ovarian cancer, Oncogene 34 (2015) 165–176. [11] C.K. Huang, C.Y. Yang, Y.M. Jeng, C.L. Chen, H.H. Wu, Y.C. Chang, et al., Autocrine/ paracrine mechanism of interleukin-17B receptor promotes breast tumorigenesis through NF-kappaB-mediated antiapoptotic pathway, Oncogene 33 (2014) 2968–2977. [12] J. Lee, W.H. Ho, M. Maruoka, R.T. Corpuz, D.T. Baldwin, J.S. Foster, et al., IL-17E, a novel proinflammatory ligand for the IL-17 receptor homolog IL-17Rh1, J. Biol. Chem. 276 (2001) 1660–1664. [13] Y. Iwakura, H. Ishigame, S. Saijo, S. Nakae, Functional specialization of interleukin-17 family members, Immunity 34 (2011) 149–162. [14] E.A. Rickel, L.A. Siegel, B.R. Yoon, J.B. Rottman, D.G. Kugler, D.A. Swart, et al., Identification of functional roles for both IL-17RB and IL-17RA in mediating IL-25-induced activities, J. Immunol. 181 (2008) 4299–4310. [15] J.M. Reynolds, Y.H. Lee, Y. Shi, X. Wang, P. Angkasekwinai, K.C. Nallaparaju, et al., Interleukin-17B Antagonizes Interleukin-25-Mediated Mucosal Inflammation, Immunity 42 (2015) 692–703. [16] B. Hoesel, J.A. Schmid, The complexity of NF-kappaB signaling in inflammation and cancer, Mol. Cancer 12 (2013) 86. [17] S.J. Myung, J.H. Yoon, S.J. Yu, STAT3 & Cytochrome P450 2C9: a novel signaling pathway in liver cancer stem cells, Biomed. Pharmacother. 66 (2012) 612–616. [18] F.M. Gu, Q.L. Li, Q. Gao, J.H. Jiang, K. Zhu, X.Y. Huang, et al., IL-17 induces AKT-dependent IL-6/JAK2/STAT3 activation and tumor progression in hepatocellular carcinoma, Mol. Cancer 10 (2011) 150. [19] L. Wang, T. Yi, M. Kortylewski, D.M. Pardoll, D. Zeng, H. Yu, IL-17 can promote tumor growth through an IL-6-Stat3 signaling pathway, J. Exp. Med. 206 (2009) 1457–1464.

399

[20] K.S. Tsai, S.H. Yang, Y.P. Lei, C.C. Tsai, H.W. Chen, C.Y. Hsu, et al., Mesenchymal stem cells promote formation of colorectal tumors in mice, Gastroenterology 141 (2011) 1046–1056. [21] A. De Luca, L. Lamura, M. Gallo, V. Maffia, N. Normanno, Mesenchymal stem cell-derived interleukin-6 and vascular endothelial growth factor promote breast cancer cell migration, J. Cell. Biochem. 113 (2012) 3363–3370. [22] M. Imhof, I. Karas, I. Gomez, A. Eger, M. Imhof, Interaction of tumor cells with the immune system: implications for dendritic cell therapy and cancer progression, Drug Discov. Today 18 (2013) 35–42. [23] L. Zimmerlin, A.D. Donnenberg, J.P. Rubin, P. Basse, R.J. Landreneau, V.S. Donnenberg, Regenerative therapy and cancer: in vitro and in vivo studies of the interaction between adipose-derived stem cells and breast cancer cells from clinical isolates, Tissue Eng Part A 17 (2011) 93–106. [24] J.D. Lathia, J.M. Heddleston, M. Venere, J.N. Rich, Deadly teamwork: neural cancer stem cells and the tumor microenvironment, Cell Stem Cell 8 (2011) 482–485. [25] R. Axelrod, D.E. Axelrod, K.J. Pienta, Evolution of cooperation among tumor cells, Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 13474–13479. [26] V.L. Bautch, Cancer: tumour stem cells switch sides, Nature 468 (2010) 770– 771. [27] S.C. Bendall, M.H. Stewart, P. Menendez, D. George, K. Vijayaragavan, T. Werbowetski-Ogilvie, et al., IGF and FGF cooperatively establish the regulatory stem cell niche of pluripotent human cells in vitro, Nature 448 (2007) 1015– 1021. [28] Z. Huang, T. Wu, A.Y. Liu, G. Ouyang, Differentiation and transdifferentiation potentials of cancer stem cells, Oncotarget 6 (2015) 39550–39563, doi:10.18632/oncotarget.6098. [29] S. Furuta, Y.M. Jeng, L. Zhou, L. Huang, I. Kuhn, M.J. Bissell, et al., IL-25 causes apoptosis of IL-25R-expressing breast cancer cells without toxicity to nonmalignant cells, Sci. Transl. Med. 3 (2011) 78ra31. [30] T. Benatar, M.Y. Cao, Y. Lee, J. Lightfoot, N. Feng, X. Gu, et al., IL-17E, a proinflammatory cytokine, has antitumor efficacy against several tumor types in vivo, Cancer Immunol. Immunother. 59 (2010) 805–817. [31] T. Benatar, M.Y. Cao, Y. Lee, H. Li, N. Feng, X. Gu, et al., Virulizin induces production of IL-17E to enhance antitumor activity by recruitment of eosinophils into tumors, Cancer Immunol. Immunother. 57 (2008) 1757–1769. [32] H. Korkaya, S. Liu, M.S. Wicha, Regulation of cancer stem cells by cytokine networks: attacking cancer’s inflammatory roots, Clin. Cancer Res. 17 (2011) 6125–6129. [33] J. Bromberg, T.C. Wang, Inflammation and cancer: IL-6 and STAT3 complete the link, Cancer Cell 15 (2009) 79–80. [34] C. Holohan, S. Van Schaeybroeck, D.B. Longley, P.G. Johnston, Cancer drug resistance: an evolving paradigm, Nat. Rev. Cancer 13 (2013) 714–726. [35] B.B. Zhou, H. Zhang, M. Damelin, K.G. Geles, J.C. Grindley, P.B. Dirks, Tumourinitiating cells: challenges and opportunities for anticancer drug discovery, Nat. Rev. Drug Discov. 8 (2009) 806–823. [36] M. Novatchkova, A. Leibbrandt, J. Werzowa, A. Neubuser, F. Eisenhaber, The STIR-domain superfamily in signal transduction, development and immunity, Trends Biochem. Sci. 28 (2003) 226–229. [37] K. Yamaoka, P. Saharinen, M. Pesu, V.E. Holt 3rd, O. Silvennoinen, J.J. O’Shea, The Janus kinases (Jaks), Genome Biol. 5 (2004) 253. [38] J.J. O’Shea, S.M. Holland, L.M. Staudt, JAKs and STATs in immunity, immunodeficiency, and cancer, N. Engl. J. Med. 368 (2013) 161–170. [39] S.G. Rane, E.P. Reddy, Janus kinases: components of multiple signaling pathways, Oncogene 19 (2000) 5662–5679. [40] B. Zhang, C. Liu, W. Qian, Y. Han, X. Li, J. Deng, Crystal structure of IL-17 receptor B SEFIR domain, J. Immunol. 190 (2013) 2320–2326. [41] C. Liu, S. Swaidani, W. Qian, Z. Kang, P. Sun, Y. Han, et al., A CC’ loop decoy peptide blocks the interaction between Act1 and IL-17RA to attenuate IL-17and IL-25-induced inflammation, Sci. Signal. 4 (2011) ra72. [42] H. Lee, A. Herrmann, J.H. Deng, M. Kujawski, G. Niu, Z. Li, et al., Persistently activated Stat3 maintains constitutive NF-kappaB activity in tumors, Cancer Cell 15 (2009) 283–293. [43] S.I. Grivennikov, M. Karin, Dangerous liaisons: STAT3 and NF-kappaB collaboration and crosstalk in cancer, Cytokine Growth Factor Rev. 21 (2010) 11–19.