Targeted blockade of TGF-β and IL-6/JAK2/STAT3 pathways inhibits lung cancer growth promoted by bone marrow-derived myofibroblasts

To investigate the role of TGF-β and IL-6 in myofibroblasts (MFs) — lung cancer cell interactions, lung cancer cells (Lewis and CTM-167 cell lines) were stimulated by IL-6, MF-conditioned medium (MF-CM) or MFs, with or without TGF-β signaling inhibitor — SB431542 and/or JAK2/STAT3 inhibitor — JSI-124. MFs were stimulated by TGF-β, cancer cell-CM or cancer cells, with or without SB431542 and JSI-124. Cell proliferation, the levels of cytokines, expression of mRNA and protein were determined. Mice bearing xenograft tumors were intraperitoneally treated with SB431542 or JSI-124 and monitored for up to 45 days. In co-culture systems, MFs secreted high levels of IL-6, while cancer cells produced high levels of TGF-β. Recombinant IL-6 and MF-CM activated STAT3 and upregulated TGF-β in cancer cells. In contrast, cancer cell-CM or TGF-β stimulated MFs to produce IL-6. Blockade of JAK2/STAT3 and TGF-β signaling by specific inhibitors significantly inhibited cell proliferation in vitro and tumor growth in vivo of lung cancer cells. Our study demontrated that the TGF-β and IL-6/JAK2/STAT3 signaling pathways form a positive feedback signaling loop that mediated the interactions between MFs and lung cancer cells. Targeted inhibiton of this signaling loop could be a new approach for lung cancer prevention and therapy.

intracellular signaling through JAK tyrosine kinases and is amplified by other downstream signaling effectors including PI3K, MAPKs, and STATs 7 . Changes in signaling via another pro-inflammatory cytokine, transforming growth factor-β (TGF-β), are also closely linked to various activities related to cancer onset and migration [8][9][10] . Signaling pathways regulated by TGF-β in lung cancer cells include Wnt/β-catenin, MAPK, and JAK/STAT3 signaling 11 .
Recent studies indicate that patients with diagnosed lung cancer have elevated serum IL-6 (compared with normal subjects) and it is correlated with poor prognosis 12,13 . Given the stimulatory role of IL-6 on JAK/STAT signaling, IL-6/JAK/STAT3 signaling may be involved in lung cancer progression. Epithelial-mesenchymal transition occurs when epithelial cells take on mesenchymal properties and is important for progression and metastasis of cancer. As such, pro-inflammatory cytokine signaling through JAK/STAT3 could be critical for the interactions between lung cancer cells and stromal cells in the transformation of lung cancer cells to take on stromal cell properties, which promote progression of lung cancer and result in poor prognosis 14 .
Gaining a better understanding of the inflammatory signaling cascades in the interaction between lung cancer cells and MFs would aid in the development of approaches for inhibiting cancer progression. The present study aims to investigate the roles that TGF-β and IL-6/JAK2/STAT3 signaling play in cancer cell-MF interaction and how this interaction could influence cancer cell proliferation and disease progression in both in vitro and in vivo systems.

Results
Lung cancer cell-produced TGF-β induces MF proliferation and cytokine secretion. To measure MF proliferation, we cultured the cells with different treatments, and CCK-8 was assayed at 0 h, 24 h and 48 h. Although normal lung cancer cell culture medium stimulated MF proliferation over time, culture in Mouse non-small lung cancer cell line-conditioned medium (CMT-167-CM) and Lewis lung cancer cell line-conditioned medium (LLC-CM) further enhanced MF cell proliferation as determined through CCK-8 release ( Fig. 1A and B). At both 24 h and 48 h, in addition of mrTGF-β (mouse recombinant TGF-β), co-cultureing with CMT-167 cells, and culturing with CMT-167-CM increased all mIL-6 protein levels ( Fig. 1C and E) and at 48 h, In a second set of experiments, MFs were cultured alone, with mrTGF-β (2 ng/mL), with caner condition-medium or co-cultured with CMT/LLC cells. mIL-6 concentration in supernatant was measured by ELISA (C-F) and mIL-6 protein in MFs was assessed by Western blots (G,H). mrTGF-β, lung cancer cells and cancer-CM enhanced expression of mIL-6 in MFs. n = 6 wells/group; *Compared to MF group, p < 0.05; # Compared to NM group, p < 0.05.
SCiENTiFiC RepoRTs | 7: 8660 | DOI:10.1038/s41598-017-09020-8 co-cultureing with LLC cells, and culturing with LLC-CM ( Fig. 1D and F) both elevated IL-6 production by MFs in comparison to MFs culture alone (p < 0.05). mRNA levels in MFs under these different culture conditions were similar at both 24 h and 48 h (Supplemental Fig. 1A,B). Western blotting results for mIL-6 protein were consistent with the ELISA data showing that mIL-6 expression was increased by all treatments over normal culture conditions, especially mrTGF-β ( Fig. 1G and H).
MFs promote lung cancer cell proliferation and cytokine production via IL-6. Cancer cell lines (CMT/LLC) were cultured in normal medium or MF-CM and their proliferation was assessed by CCK-8 at 0 h, 24 h and 48 h ( Fig. 2A and B). The results showed that treatment with MF-CM promoted the proliferation of cancer cells.
To determine if IL-6 was promoting this proliferation and TGF-β expression by cancer cells, CMT/LLC cells were cultured in 6-well plates by themselves in normal medium or treated with mrIL-6 (mouse recombinant IL-6), or co-cultured with MFs in a transwell system. mTGF-β concentrations in supernatant were measured by ELISA (Fig. 2C-F). Related mTGF-β protein levels and mRNA levels by CMT/LLC cells were measured by Western blots (Fig. 2G and H) and by RT-PCR (Supplemental Fig. 2A,B). Our results showed that mrIL-6, MFs and MF-CM significantly enhanced the expression of mTGF-β in these lung cancer cells (*p < 0.05 compared with the MF alone group).

Treatment with SB-431542 and JSI-124 reduces cytokine expression in mouse tumor tissues.
The protein expression of IL-6, TGF-β and α-SMA in tumor tissues from BALB/c athymic nude mouse was determined by Western Blots (Fig. 7A). α-SMA expression in tumor tissues was also analyzed by immunohistochemistry (Fig. 7B). Based on the blots, it appeared that each inhibitor and the combination of inhibitors reduced expression of IL-6, TGF-β and α-SMA protein in tumor tissues isolated from the mice injected with CMT-167, MF and CMT + MF. Likewise, the immunohistochemical staining of α-SMA was reduced in histological sections of tumor tissue in each of the inhibitor-treated mouse group.

Discussion
In the present study, we aimed to elucidate the mechanisms by which MFs and cancer cell interact to produce cytokines and growth factors in vitro and in vivo. We hope these studies will shed light on mechanisms that may be targeted to inhibit tumorigenesis, tumor progrsssion and metastasis of lung cancers in a clinical setting. These studies build upon important previous findings that highlight key signaling pathways, such as JAK/STAT and TGF-β in the interaction between lung cancer cells and MFs.
Tumor tissues are not simply made up of cancer cells, but also other cells that comprise vasculature, inflammatory cells, and fibroblasts. The interaction between cancer cells and the stromal microenvironment is crucial for tumor progression and metastasis. Of the stromal tissue involved in this process, MFs are well documented to promote cancer develoment and progression [15][16][17] and are considered the primary stromal cell type within human tumors 18 . MFs also contribute to migratory action of cancer cells and tumor angiogensis 19,20 . The interplay between MFs and cancer cells is mediated by contact, as well as by paracrine-mediated signaling between the cell types. It is generally accepted that MFs are in contact with cancer cells, and express α-SMA 21 . The process through which stromal fibroblasts differentiate into MFs and contribute to cancer progression is of great interest, and this process may be targeted for cancer theraputics.
TGF-β is thought to be a key paracrine factor that triggers differentiation of MFs by cancer cells 22,23 . Similarly, TGF-β is also likely to trigger a paracrine feedback loop by which differentiated stromal-origin MFs promote cancer-supporting processes and events 22 . Our findings support the role of TGF-β in enhancing proliferation of MFs. When MFs were cultured with cancer cell-CM, proliferation of MFs increased. Importantly, by inhibiting TGF-β signaling, we were able to prevent the effect of cancer-CM on proliferation of MFs. We also found that cancer cells and cancer-CM promoted IL-6 secretion by MFs.
IL-6 has been previously shown as a growth-promoting factor for cancer cells 24 . We showed that MFs secrete IL-6 in response to factors secreted by cancer cells. MFs and MF-CM induced cancer cell proliferation, similar to the action of IL-6. Therefore, TGF-β and IL-6 form an important paracrine signaling cycle between MFs and cancer cells, and this signal cycle contributes to cancer progression.
Our studies with inhibitors of TGF-β and JAK/STAT3 signaling implicate these pathways as a key mechanism by which TGF-β influences growth and cytokine expression between these two cell types. Blockade of TGF-β as well as JAK2/STAT3 signaling reduced the proliferation and cytokine production of both cancer cells and MFs in culture (Figs 3 and 5). This is consistent with previous observations that IL-6 activates cancer cell growth 25 and prevents cancer cell death via STAT3 activation 26,27 . IL-6-mediated STAT3 activation has been shown to promote drug resistance in cancer cells 27 , underscoring the importance of such cytokine signaling between stromal MFs and cancer cells as a possible target for cancer therapeutic development.
To further support our in vitro result, we inoculated athymic nude mice with MFs and cancer cells, and found that suppression of TGF-β and JAK2/STAT3 signaling by inhibitors reduced tumor size and minimized tumor histopathology in vivo (Fig. 6). We also observed that inhibiting these signaling pathways reduced α-SMA expression in tumor tissues. There is evidence that α-SMA-negative MFs do not promote cancer cell-associated behaviors that are found in α-SMA-positive MFs 20,22 . As such, our study extends findings concerning mechanisms of paracrine-mediated interactions between MFs and cancer cells in an animal model, and provides support for the use of these cytokines as targets for preventing cancer progression.
Future studies should explore the role of MF differentiation and its involvement in tumor formation and progression. Immune cell infiltration, such as by CD 8 + T-lymphocytes, is an important immune response to cancer cells and positively correlated with good prognosis in various cancers [28][29][30][31] . Some observations suggest that MFs may prevent interaction between such infiltrating immune cells with cancer cells, thus preventing immune-mediated anti-cancer responses 32 . This mechanism, coupled with the influence of cancer-associated fibroblasts with mechanical and matrix remodel influences on the tumor microenvironment 33 , could be important topics for future studies.
In conclusion, our study expands upon prior research in providing evidence for an important paracrine-mediated interaction between cancer cells and stromal-origin MFs in tumor formation and progression. Our findings highlight mechanistic importance of TGF-β and JAK2/STAT3 signaling in this positive feedback loop in vitro and in vivo. Future studies will further investigate specific mechanisms and inhibitors for the interactions between these two cell types to aid in the development of effective cancer therapeutics.

Materials and Methods
Cell culture and reagents. CMT-167 cells and LLC cells were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Bone marrow-derived myofibroblasts (MFs) were isolated in our   laboratory from dysplastic gastric tissues of EGFP + bone marrow-transplanted IL-1β transgenic mice based on previous methods 34 . RPMI 1640 medium containing 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin (Gibco BRL, Grand Island, NY, USA). Cucurbitacin I (JSI-124) and SB 431542 were purchased from Sigma (St. Louis, MO, USA) and were dissolved in dimethyl sulfoxide (DMSO; Sigma, St. Louis, USA) and stored at −20 °C. Mouse recombinant IL-6 (mrIL-6) and mouse recombinant TGF-β (mrTGF-β) were purchased from RD (Minneapolis, MN, USA).
For the production and isolation of MF conditioned medium (MF-CM) and lung cancer cell conditioned medium (CMT-CM and LLC-CM), MFs, CMT and LLC cells were cultured in RPMI 1640 medium containing 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin. Culture medium was changed once cells were 70% confluent in 10 cm dishes. The final medium for experimental use was collected after 24 h and centrifuged at 800 rpm for 5 min. The supernatant was passed through a 0.45 μm filter to remove any remaining debris and cell components and stored at 4 °C. In experiments involving cell transplantation for tumor formation in mice, cells were passed through a 40 μm filter for isolation. An enzyme-linked immunosorbent assay (ELISA) kit was purchased from BD Biosciences (San Jose, CA, USA) for cell and tissue protein quantification. For Western blotting, gels and reagents were purchased from Millipore (Bedford, MA, USA). Absorbance was measured at 450 nm by a Multiscan MC reader (Labsystems Multiskan, MS, Finland). intraperitoneally three times per week starting one day after tumor cell inoculation, and control mice receive the same volume of diluent. All mice were monitored daily and euthanized after 45 days. Tumors were processed for further histology analysis, immunohistochemistry and Western Blot.

Protein isolation and Western blot analysis. Cells from in vitro experiments and tumor tissues from in
vivo experiments were harvested and homogenized in lysis buffer on ice for 30 min. Cell and tissue homogenates were then centrifuged for 15 min at 13,000 RPM at 4 °C. Supernatant was collected with isolated protein, and a Bradford assay was performed to standardize protein concentrations for Western blot and ELISA 35 .
For Western blot analysis, 20 ug protein of each sample was run on a 10% SDS-PAGE gel, transferred to a membrane, and blocked with 5% powdered milk in PBS for 1 h at room temperature. Then, the membranes were incubated with designated antibodies in blocking buffer overnight at 4 °C. The next day, the membranes were washed, appropriate secondary antibodies were applied for 1 h at room temperature, and bands were visualized with an enhanced chemiluminescence kit (Amersham Biosciences, Beijing, China).
Histopathology and immunohistochemistry. To assess tumor histopathology and smooth muscle actin (SMA) expression, tumor tissues were sectioned and stained with hematoxylin/eosin staining and immunohistochemistry (IHC) was performed. For IHC, tissue sections were rinsed in PBS, blocked with 10% nomal goat serum in PBS for 1 h at room temperature, and incubated overnight with rabbit anti-α-SMA antibody at the manufacturer's recommended dilution (Abcam, Cambrige, MA, USA). The next day, the tissue sections were washed and incubated in peroxidase-labeled goat anti-rabbit secondary antibodies (Dako Cytomation, Copenhagen, Denmark). Laser confocal fluorescence microscopy (OLYMPUS A, FV1000) was performed after serial dehydration and mounting of the tissue slides.
Statistical analysis. Data were presented as means ± SEM. Statistical analysis was performed by using SPSS