Lung cancer (LC) is the major cause of death by cancer and the number of LC patients is increasing worldwide. This study investigated the therapeutic potential of gene delivery using suppressor of cytokine signaling 1 (SOCS-1), an endogenous inhibitor of intracellular signaling pathways, for the treatment of LC. To examine the antitumor effect of SOCS-1 overexpression on non-small-cell lung cancer (NSCLC) cells, NSCLC cells (A549, LU65, and PC9) were infected with adenovirus-expressing SOCS-1 vector. The cell proliferation assay showed that A549 and LU65, but not PC9, were sensitive to SOCS-1 gene-mediated suppression of cell growth. Although JAK inhibitor I could also inhibit proliferation of A549 and LU65 cells, SOCS-1 gene delivery appeared to be more potent as SOCS-1 could suppress focal adhesion kinase and epidermal growth factor receptor, as well as the JAK/STAT3 signaling pathway. Enhanced phosphorylation of the p53 protein was detected by means of phospho-kinase array in SOCS-1 overexpressed A549 cells compared with control cells, whereas no phosphorylation of p53 was observed when JAK inhibitor I was used. Furthermore, treatment with adenoviral vector AdSOCS-1 in vivo significantly suppressed NSCLC proliferation in a xenograft model. These results suggest that the overexpression of SOCS-1 gene is effective for antitumor therapy by suppressing the JAK/STAT, focal adhesion kinase, and epidermal growth factor receptor signaling pathways and enhancing p53-mediated antitumor activity in NSCLC.
Lung cancer is the leading cause of cancer death in Japan and is a growing health epidemic worldwide. Moreover, therapies that can cure metastatic LC have not been yet established,[2, 3] so there is an urgent need for the development of novel interventions to cure LC.
One of the potential therapeutic targets of NSCLC is STAT3. Constitutively activated STAT3 has been shown to promote tumor cell growth, survival, and tumor angiogenesis, and persistently activated STAT3 has been found in 50% of lung adenocarcinomas. It is thought that STAT3 is activated by JAK, EGFR, or Src family kinases. Among these TYKs, JAK family kinases play an important role in the phosphorylation of STAT3 in NSCLC.[6, 7] Dysregulated activation of the JAK/STAT3 signaling pathway, the major downstream pathway of cytokines such as interleukin-6, has been detected in various cancers including NSCLC. Moreover, it has been recently reported that ruxolitinib, which is a potent and selective JAK1 and JAK2 inhibitor, is associated with marked and durable clinical benefits for patients with myelofibrosis, suggesting that JAK kinases are promising therapeutic targets for cancer.
Cytokine signaling pathways are tightly controlled by negative regulatory mechanisms under homeostatic conditions. Suppressors of cytokine signaling family proteins play a role in the negative regulation of cytokine responses by terminating the activation of the JAK/STAT and other signaling pathways.[10-12]
The SOCS family, characterized by a central src homology 2 domain and a conserved C-terminus SOCS box, is composed of eight structurally related proteins. Of these, SOCS-1 is known as the most potent negative regulator of pro-inflammatory cytokine signaling. It interacts with phosphotyrosine residues on proteins such as JAK kinases to interfere with the activation of STAT proteins or other signaling intermediates.[15, 16] Also, SOCS-1 recruits the Elongin BC-containing E3 ubiquitin-ligase complex through the conserved SOCS box to promote the degradation of target proteins. Studies of SOCS-1 deficient mice have indicated that SOCS-1 is essential for the inhibition of excessive immune responses and is also involved in the suppression of tumor development.[18, 19]
Although it is still not clear whether SOCS shows therapeutic benefit for NSCLC, preclinical analyses of SOCS in therapies of several types of cancers have been carried out worldwide. Previously reported studies by us showed that SOCS-1 or SOCS-3 were effective when used for therapies of malignant pleural mesothelioma or gastric cancer.[21-23] In addition, it has been reported that overexpression of the SOCS-3 gene showed antitumor effects in NSCLC.[24-26]
In the study presented here, we used the NSCLC cell lines A549, LU65, and PC9 to investigate the possibility of the application of SOCS-1 gene transduction to NSCLC therapies and the mechanisms of antitumor effects by SOCS-1.
Material and Methods
Both A549 and LU65 cell lines were obtained from the Japanese Collection of Research Bioresources (Osaka, Japan). The PC9 cell line was kindly provided by Prof. Nishio of Kinki University of Medicine, Department of Genome Biology (Osaka, Japan). Details are described in Data S1.
PD153035 and JAK inhibitor I were purchased from EMD Millipore (Billerica, MA, USA) and Calbiochem (La Jolla, CA, USA), respectively.
Preparation of adenoviruses
The replication-defective recombinant adenoviral vector expressing the mouse SOCS-1 gene was provided by Dr. Hiroyuki Mizuguchi (Osaka University, Osaka, Japan), which was constructed with an improved in vitro ligation method, as described previously.[27, 28] An adenoviral vector expressing the LacZ gene was constructed using a similar method and expression of these genes was regulated by means of a CMV promoter/enhancer and intron A. Details are described in Data S1.
Expression of phosphorylated proteins was detected with the Proteome Profiler Human Phospho-Kinase Array kit (R&D Systems, Minneapolis, MN, USA). Details are described in Data S1.
Cell viability assay
The NSCLC cell lines were plated in 96-well plates at a density of 1 × 103 cells per well and incubated in RPMI-1640 medium containing 10% FCS. Details are described in Data S1.
SDS-PAGE and Western blot analysis
Whole cell protein extracts were prepared from NSCLC cells and tumor tissue in RIPA buffer containing phosphatase inhibitor cocktail and a protease inhibitor cocktail (both from Nacalai Tesque, Kyoto, Japan and both at a concentration of 1×) followed by centrifugation (16 100g, 4°C, 15 min). Details are described in Data S1.
Small interfering RNA transfection
Commercial FAK siRNA was obtained from Qiagen (Hilden, Germany). Details are described in Data S1.
Mouse xenograft model
All animal experiments were carried out according to the institutional ethical guidelines for animal experimentation of the National Institute of Biomedical Innovation (Osaka, Japan). Details are described in Data S1.
Subcutaneously implanted tumors were harvested and paraffin embedded for immunohistochemical analysis using anti-SOCS-1 antibody (Abcam, Cambridge, MA, USA) and anti-Ki-67 antibody (Novocastra Laboratories, Newcastle, UK). A TUNEL assay (with DAPI nuclear counterstaining) for apoptosis was carried out using the ApopTag Fluorescein In Situ Apoptosis Detection Kit (Chemicon International, Temecula, CA, USA) according to the manufacturer's instructions.
Data are shown the mean ± SD (unless stated otherwise) for the number of experiments indicated. Details are described in Data S1.
SOCS-1 gene delivery shows marked antiproliferative effects in A549 and LU65 cells
We used A549, LU65, and PC9 cell lines, which have been used as NSCLC cell lines in many other experiments. Because the JAK/STAT3 pathway has recently received attention as a novel target for treatment of NSCLC, we investigated the expression levels of JAK family kinases and of STAT3, as well as tyrosine phosphorylation of STAT3 in NSCLC cells. The four known mammalian members of the JAK family are TYK2, JAK1, JAK2, and JAK3. These kinases are widely expressed in a variety of different cell types, with the exception of JAK3, which is selectively expressed in cells of hematopoietic origin. The expression of each JAK was at a similar level in the three cell lines (Fig. 1a). Although phosphorylated STAT3 was detectable in all cell lines, it was markedly more elevated in A549 and LU65 (Fig. 1b).
We next investigated whether SOCS-1 could suppress proliferation of A549, LU65, and PC9. Following SOCS-1 gene delivery, overexpression of SOCS-1 was detected in all cell lines by Western blot analysis (Fig. 1c), but overexpression of SOCS-1 had antiproliferative effects on A549 and LU65 but not on PC9 (Fig. 1d).
Overexpression of SOCS-1 shows stronger antiproliferative effect than JAK inhibitor in NSCLC cells
Immunoblotting analysis showed that the expression levels of some of the JAK family kinases in all cell lines were reduced by overexpression of SOCS-1 (Fig. 2a). Next, phosphorylation levels of STAT3 (Tyr705), a downstream molecule of the JAK kinase family, were analyzed by Western blot analysis. Phosphorylation levels of STAT3 (Tyr705) were decreased in response to overexpression of the SOCS-1 gene in all cell lines (Fig. 2b). These results agree with previously reported findings that SOCS-1 can act directly on JAK family kinases to suppress their kinase activity as well as to accelerate their degradation by the recruitment of an E3 ubiquitin ligase.
Although the JAK/STAT3 pathway was downregulated by SOCS-1 in all cell lines, overexpression of the SOCS-1 gene showed an antiproliferative effect only on A549 and LU65 cells. Accordingly, we next used JAK inhibitor I, a blocker of JAK1, JAK2, and JAK3, instead of AdSOCS-1 to examine the effects of JAK inhibition on all cell lines. Proliferation of A549 and LU65 cells was significantly suppressed by treatment with JAK inhibitor I, however, proliferation of PC9 was not significantly suppressed (Fig. 2c). Immunoblotting analysis showed that tyrosine phosphorylation of STAT3 was downregulated in all cell lines after introduction of JAK inhibitor I (Fig. 2d). These findings suggest that JAK-mediated signals are critical for the proliferation of A549 and LU65 cells, but are not needed for that of PC9 cells. In addition, because JAK1 and JAK2 are overlapping target molecules between SOCS-1 and JAK inhibitor I, the primary target molecules of SOCS-1 in NSCLC cells are likely to be JAK1 and/or JAK2. Therefore, we assumed that the antiproliferative effect of SOCS-1 was essentially determined by the JAK-dependence of NSCLC cells.
Immunoblotting analysis showed that introduction of both the SOCS-1 gene (at 40 MOI) and of JAK inhibitor I (5 μM) could effectively suppress STAT3 phosphorylation (Fig. 2b,d). However, SOCS-1 was more effective than JAK inhibitor I for inhibiting proliferation of A549 and LU65 cells (Fig. 2e). Because SOCS-1 is known as an adaptor of several molecules other than JAK family kinases, it was expected that AdSOCS-1 could also exert JAK-independent action in A549 and LU65. To investigate in detail the mechanisms of antitumor effects induced by overexpression of SOCS-1, we used A549 cells for further analyses.
Focal adhesion kinase is downregulated by overexpression of SOCS-1 in A549
Focal adhesion kinase is known as an adhesion molecule and previous investigations have indicated that FAK contributes to tumor cell proliferation, survival, and metastasis. SOCS-1 was found to inhibit FAK-dependent signaling events by suppressing FAK-associated kinase activity and tyrosine phosphorylation of FAK, and also by promoting polyubiquitination and degradation of FAK in a SOCS box-dependent manner. For this reason, we next focused on the FAK pathway and examined the levels of FAK expression and of FAK tyrosine phosphorylation after overexpression of the SOCS-1 gene.
Focal adhesion kinase expression was upregulated by introduction of the LacZ gene in A549 cells (Fig. 3a), probably reflecting the previously reported non-specific stimulation by adenovirus vectors. Compared to the cells introduced by LacZ, FAK phosphorylation (Tyr397) was downregulated after the introduction of the SOCS-1 gene in A549 (Fig. 3a). We also investigated whether JAK inhibitor I, instead of SOCS-1, can suppress activation of FAK in A549 cells, and found that JAK inhibitor could not (Fig. 3b).
We assumed that downregulation of FAK by SOCS-1 could contribute to the inhibition of A549 cell proliferation. Therefore, examined whether FAK siRNA indeed had an antiproliferative effect on A549, and found that proliferation of A549 cells was suppressed by FAK knockdown, indicating that these cells are dependent on FAK (Fig. 3c,d). We also investigated the combined effect of JAK inhibitor I and FAK siRNA on A549 cell proliferation. The results showed that the combined effect was stronger than that of JAK inhibitor I alone (Fig. 3e). These findings suggest that FAK has an important role in A549 cell proliferation independent of JAK and that SOCS-1-mediated FAK inhibition might contribute to the suppression of proliferation of A549 cells treated with AdSOCS-1.
Epidermal growth factor receptor is downregulated by overexpression of SOCS-1 in A549
Expression of SOCS-1 reportedly interacts with the cytoplasmic domain of EGFR and is likely to induce ubiquitination and degradation of ligand-bound EGFR. This notion was confirmed by means of immunoblotting analysis in our study, in that EGFR expression was downregulated 48 h after introduction of the SOCS-1 gene in A549 cells. The EGFR phosphorylation (Tyr1068) was also downregulated after overexpression of the SOCS-1 gene had occurred in A549 cells (Fig. 4a). In contrast, JAK inhibitor I did not suppress EGFR activation in A549 cells (Fig. 4b).
We next examined the antiproliferative effects of the EGFR inhibitor PD153035 on A549 cells. Expression of EGFR in A549 cells was much lower than that in PC9 cells (data not shown), which harbor a deletion of an EGFR mutation. Nevertheless, our findings show that proliferation of A549 cells with wild-type EGFR were also somewhat dependent on EGFR (Fig. 4c,d). We also investigated the combined effect of JAK inhibitor I and PD153035 on proliferation of A549 cells and found that they have an additive effect (Fig. 4e). This finding suggests that EGFR is also involved in A549 cell proliferation independent of JAK and that its downregulation by SOCS-1 may have an inhibitory effect on proliferation of A549 cells.
Accordingly, JAK1, JAK2, FAK, and EGFR should be considered critical for the proliferation of A549 cells, so that simultaneous inhibition of these molecules by SOCS-1 may have a potent antiproliferative effect on A549.
Upregulation of p53 by AdSOCS-1 in A549 cells
We also used a phospho-kinase array to determine the expression profile of phosphorylated proteins in A549 in order to identify other target molecules for SOCS-1 in addition to JAK, FAK, and EGFR (Fig. 5a). Phosphorylation of p53 was upregulated in A549 cells after the introduction of the SOCS-1 gene. This was confirmed by immunoblotting analysis, which showed that p53 phosphorylation was upregulated in A549 cells by overexpression of the SOCS-1 gene (Fig. 5b). We also investigated whether JAK inhibitor I, instead of SOCS-1, could activate p53 in A549 cells, but found that it did not (Fig. 5b). Given the well-established antitumor effect of p53 phoshorylation,[35-38] SOCS-1-induced p53 activation may thus also contribute to the suppression of A549 cell proliferation.
Antitumor activity of SOCS-1 in a lung cancer xenograft model
We also evaluated the therapeutic effect of AdSOCS-1 injection on the growth of NSCLC cells in vivo. We established a xenograft model of ICR nu/nu mice in which A549 cells were s.c. implanted. Injection of AdSOCS-1 vector (4 × 108 pfu) intratumorally twice per week significantly suppressed tumor growth compared to control AdLacZ injection (Fig. 6a). AdSOCS-1 in vivo could modulate intracellular signaling in NSCLC cells as in vitro, as Western blot analysis showed that phosphorylation levels of STAT3 were decreased in the A549 tissues from AdSOCS-1 injected animals (Fig. 6b). Furthermore, few Ki-67-positive nuclei were detected by immunohistochemical analysis in AdSOCS-1 infected tissues compared to AdLacZ, indicating that proliferating cells are decreased by overexpression of SOCS-1 (Fig. 6c). Additionally, induction of apoptosis was detected in AdSOCS-1 infected A549 tissue compared to AdLacZ by TUNEL analysis (Fig. 6c).
In this study, we investigated the possibility that SOCS-1 could be used in LC therapies. Previous reports showed that PC9 harbors a deletion mutation in EGFR and that A549 and LU65 cells possess wild-type EGFR.[34, 39] Although EGFR mutation in NSCLC was previously reported to activate AKT, MAPK, and STAT3 signaling, our research showed that STAT3 was more strongly expressed in A549 and LU65 than in PC9 cells, and that sensitivity to overexpression of SOCS-1 was also higher in A549 and LU65 cells than in PC9 (Fig. 1). In addition, JAK inhibitor I significantly suppressed proliferation of A549 and LU65 cells, but not of PC9. Therefore, we consider that the marked antiproliferative effect by overexpression of SOCS-1 on A549 and LU65 cells, but not on PC9, was attributable to the inhibition of JAK/STAT3 pathway in vitro. As SOCS-1 also shows an antiproliferative effect in vivo (Fig. 6), AdSOCS1 gene therapy might be effective for patients with NSCLC, in which the JAK/STAT3 signaling pathway is constitutively activated. It has been reported that approximately 50% of NSCLC tumors showed elevated phosphorylation levels of STAT3 (Tyr705) by immunohistochemical analysis. There is a further possibility that LC patients harboring STAT3 dependence, detected by immunostaining analysis of phosphorylation levels of STAT3 (Tyr705) in specimens obtained surgically or bronchoscopically, could be selected for treatment with SOCS-1 overexpression.
Comparative analyses of the antiproliferative effects of SOCS-1 gene introduction and JAK inhibitor I treatment suggest that overexpression of SOCS-1 may have a stronger effect than that of the JAK inhibitor I on A549 and LU65 cells (Fig. 2e). In fact, SOCS-1, but not JAK inhibitor I, inhibited FAK and EGFR, which are important for the survival of A549 cells. In addition, the combined effect of FAK siRNA and JAK inhibitor I, or that of PD153035 and JAK inhibitor I, was superior to the antiproliferative effect of JAK inhibitor alone (Figs 3, 4). These findings suggest that the potent antiproliferative effect of SOCS-1 depends not only on JAK inhibition but also on the suppression of other distinct signal transduction pathways, such as FAK and EGFR. In addition, phosphorylation of p53 at Ser15 was enhanced by the overexpression of the SOCS-1 gene in A549 cells (Fig. 5). Because phosphorylation of p53 at Ser15 contributes to antitumor effects under certain experimental conditions,[35-38] and A549 cells express wild-type p53, activation of p53 by SOCS-1 overexpression seems to be involved in the antitumor effects.
In conclusion, the findings of our study suggest that SOCS-1 gene therapy is potentially effective for at least a subset of NSCLC both in vitro and in vivo. It was shown that SOCS-1 had a potent antiproliferative effect on JAK-dependent NSCLC cells by targeting the JAK/STAT3 pathway. In addition, SOCS-1 successfully targeted many factors such as FAK, EGFR, and p53 in NSCLC cells. It is thus possible that SOCS-1 gene therapy could have a unique advantage over JAK inhibitor for the treatment of NSCLC.
Further studies will be needed to elucidate the mechanism of JAK/STAT3 pathway-dependence in NSCLC, and to validate the benefits that SOCS-1 gene therapy could provide for NSCLC treatment in clinical practice.
This work was supported by a Grant-in-Aid from the Ministry of Health, Labour and Welfare, Japan (T. Naka) and a grant from the Kansai Biomedical Cluster Project in Saito, which is promoted by the Knowledge Cluster Initiative of the Ministry of Education, Culture, Sports, Science and Technology, Japan (T. Naka). The authors are grateful to Ms. M. Urase for experimental assistance and Ms. Y. Kanazawa and Ms. J. Yamagishi for secretarial assistance.