Background
Interleukin-6 (IL-6) is a multifunctional cytokine that normally modulates a variety of physiological events including cell survival and apoptosis [
1], but its dis-regulation has been implicated in various diseases including cancer [
2‐
4] for which it has been associated with tumor progression, drug resistance and poor prognosis [
5‐
7]. IL-6 signaling is triggered by the binding of IL-6 to its specific ligand-binding subunit of the receptor (gp80) to induce phosphorylation and homodimerization of the common signaling-subunit of the receptor (gp130). Three major downstream signaling cascades are then activated: MEK/extracellular signal-related kinase (Erk), phosphatidylinositol 3-kinase (PI3-K)/Akt and Janus kinase (Jak) 2/signal transducer and activator of transcription 3 (Stat3) [
8]. These cascades, the most well-known being Jak2/Stat3 cascade, are responsible for IL-6 mediated cellular responses for both the physiological and pathological events [
9].
Like all members of the Stat family proteins, Stat3 is a latent cytoplasmic transcription factor activated in response to growth factors and cytokines through the phosphorylation of a single tyrosine residue [
9]. This phoshorylation is usually an indicator that Stat3 has been activated. Activated Stat3 forms a dimer and translocates to the nucleus where it binds to DNA in the promoter region of target genes to regulate gene transcription. It has been previously found that the functioning of endogenous Stat3 was inhibited when cells were transfected with S3F (a dominant-negative Stat3 mutant that cannot be tyrosine phosphorylated) or S3D (a dominant-negative Stat3 mutant that cannot bind to DNA) while an additional functioning of exogenous Stat3 was supplied when cells are transfected with S3C (a constitutively-active Stat3 mutant forced to form a dimer constitutively without stimulation) [
2,
10]. The ability of these mutants to affect the functioning of Stat3 makes it possible to study the effect of Stat3 on gene regulation.
IL-6 is induced by a variety of stimuli that mostly achieve this through their activation of NF-κB, C/EBP, CREB and AP-1, which are transcription factors known to bind to IL-6 promoter [
11‐
13]. IL-6 is also known to be auto-regulated in many types of cells [
14,
15]. For example, MEK/Erk and PI3-K/Akt, which are, as mentioned above, downstream pathways triggered by IL-6, also work upstream to regulate the expression of IL-6. PI3-K/Akt does this by activating IKK-α which in turn activates AP-1 and NF-κB to induce the expression of IL-6 [
16,
17], and MEK/Erk kinase does this by activating NF-κB [
18]. Recently, NF-κB, long known to be an important upstream regulator of IL-6 expression [
12,
13], has been found to be activated downstream by IL-6 [
19]. However, the role of the most well-known IL-6 downstream signal, Jak2/Stat3 pathway, remains controversial.
Some studies have suggested that Jak2/Stat3 pathway may also be involved upstream in the regulation of IL-6, but other studies disagree. Studies not directly investigating the role of Stat3 on the expression of IL-6 in cancer cells have found some evidence suggesting Stat3 may increase IL-6 expression. IL-6 mRNA was found to be elevated in tumor tissue in gp130 mutant mice with abnormally activated Stat3 [
20]. IL-6 mRNA was found to be up-regulated in alveolar type II epithelial cells of transgenic mice over-expressing S3C in a tissue-specific manner [
21]. In more recent studies of the role of Stat3 in immune responses in macrophages and fibroblasts, Ogura
et al. reported that IL-6 as well as other cytokines could be decreased by inhibiting Stat3 [
22‐
24]. Another study investigating the role of Stat3 in immune evasion in human melanoma cells, has reported that Stat3 siRNA decreased the mRNA expression of IL-6, IL-10 and VEGF [
25]. Gao
et al. showed that mutant EGFR could activate the gp130/Jak/Stat3 pathway to increase tumorigenesis by up-regulation of IL-6 but the authors did not specifically knock-down Stat3 to show the increase of IL-6 secretion by mutant EGFR is mediated by Stat3 activation in their study [
26]. However, two immunological studies investigating the effect of over-expression of S3C on the production of cytokines found that transfection of S3C suppressed the expression of IL-6 in macrophages [
27,
28]. Another important study investigating the possible effect of Stat3 on immune suppression of cancer cells found that the inhibition of Stat3 with antisense oligonucleotide and with dominant-negative form of Stat3 (Stat3β) resulted in an increase in IL-6 in mouse cancer cells [
29]. Because these investigations were not designed specifically to study or to provide direct evidence of the role of Stat3 on the expression IL-6 in cancer cells, we performed biochemical and genetic studies of manipulating the Stat3 function to clarify its role on the autocrine production of IL-6 in various cancer cell lines and human tumor samples.
Methods
Materials
The AG490, LY294002, U0126, BAY11-7082 and PD98059 inhibitors were purchased from Biomol (Plymouth, PA, USA). The chemotherapeutic agents, paclitaxel, camptothecin, vincristine and etoposide were purchased from Sigma (St Louis, MO, USA). Epirubicin was purchased from Merck (Darmstadt, Germany).
Cell culture
For this study, we used one human lung adenocarcinoma cell line PC14PE6/AS2 (AS2) to study the effect of IL-6 downstream pathways on IL-6 autocrine production and drug resistance. We had previously established this cell line and found it to produce autocrine IL-6 which activated Stat3 and subsequently promoted tumor progression [
2]. In addition, we used a series of AS2-derived cell lines: one vector cell line (AS2/Vec-11) and 6 mutant cell lines expressing plasmids containing constitutively-active (AS2/S3C cells: AS2/S3C-A and AS2/S3C-C) or dominant-negative Stat3 (AS2/S3D cells: AS2/S3D-8, AS2/S3D-9, and AS2/S3F cells: AS2/S3F-3, and AS2/S3F-7).
We used 3 other cancer cell lines, MCF-7, KB and A549 (American Type Culture Collection) and their derived drug resistant cell lines. The MCF-7 derived drug resistant cell line MCF-7/ADR was kindly provided by Dr. Chih-Hsin Yang (National Taiwan University, Taipei, Taiwan). This cell line was maintained with 1 μM epirubicin to ensure it retained its drug resistance [
30]. We used 5 other drug resistant cell lines that we had previously established from KB and A549 cells: KB-CPT100 maintained with 100 nM camptothecin; KB-TAX50 maintained with 50 nM paclitaxel; KB-VIN10 maintained with 10 nM vincristine; KB-7D maintained with 1 μM etoposide; and A549-T12 maintained with 12 nM paclitaxel [
31,
32].
AS2- and MCF-7 parental and derived cells were maintained in MEM-α and DMEM medium (Invitrogen, Carlsbad, CA, USA), respectively, with 10% fetal calf serum (FCS; Invitrogen), and KB and A549 parental and derived cells were maintained in RPMI 1640 (Invitrogen) with 5% FCS.
Patient and sample processing
Lung cancer cells were collected from the lung cancer associated malignant pleural effusion (MPE) of twenty patients treated at National Cheng Kung University Hospital. Each patient provided written informed consent. Each sample was verified to be positive by cytological analysis of MPE or pathological proof based on a pleural biopsy. MPE samples were collected and centrifuged immediately. Tumor cells were separated from MPE-associated lymphocytes by serial gradient centrifugation with Histopaque1077 and Percoll (Sigma) as previously described [
33]. The purity of tumor cells was determined by cytological analysis to be between 70% and 90%. Frozen samples were cryopreserved in 90% FCS/10% DMSO. Freshly isolated or defrosted cells were suspended in RPMI 1640 medium with 10% FCS and allowed to rest at 37°C for 1 hour before treatment with signal pathway inhibitors. The protocol for this study was approved the institutional review board at National Cheng Kung University Hospital.
Enzyme-linked immunosorbent assay (ELISA) for IL-6
Attached cells were plated at concentrations of 0.5 × 105 - 3 × 105 cells/ml/well in 12-well plates. The suspended cancer cells obtained from MPE were grown in sterile tubes to a concentration of 2.5 × 105 cells/ml/tube. After treatment, the conditioned media were collected at indicated time points and stored at -20°C until further use. The collected samples were assayed using a commercially available ELISA kit (Invitrogen).
Cell lysis and Western blot analysis
For cell lysis, the harvested cells were incubated on ice in whole-cell-extract lysis buffer for 30 min, lysates were centrifuged at 14000 rpm for 10 min, and protein concentration measured by Bradford assay (Bio-Rad, Richmond, CA, USA). For Western blot analysis, lysates were then boiled for 5 min with sample buffer before being separated on SDS-polyacrylamide gels. Proteins were transferred to polyvinylidene difluoride membranes (Millipore, Billerica, MA, USA) and blocked with 5% nonfat milk/TBST buffer. Using an electrochemiluminescence kit (Amersham Pharmacia, Biotech Inc., Piscataway, NJ, USA), we detected binding of eight specific antibodies: (1) anti-phospho-Stat3 (Tyr705) (Cell Signaling, Danvers, MA, USA), (2) anti-Stat3 (BD Biosciences, San Jose, CA, USA), and (3) anti-actin (Millipore) (4) anti-phospho-Akt (Ser473) (Cell Signaling), (5) anti-Akt (Cell Signaling), (6) anti-Akt1 (Cell Signaling), (7) anti-phospho-Erk (R & D, Minneapolis, MN, USA), and (8) anti-Erk (Santa Cruz Biotechnology, Santa Cruz, CA, USA).
MTT assay
Cells were seeded at concentrations of 5×103 - 7.5×103 cells/200 μl/well in 96-well plates. After treatment, one-tenth of the original culture volume of MTT (Sigma) stock solution was added to the wells and incubated for 4 hours at 37°C. After removing the supernatant by centrifugation, DMSO was added to release MTT.
Luciferase reporter assays
The p1168huIL6P-luc+, a pGL3 based IL-6 promoter luciferase reporter plasmid containing 1168 bp of the human IL-6 promoter, was kindly provided by Dr. Hsiao-Sheng Liu (National Cheng Kung University, Tainan, Taiwan) [
34], the mammalian expression plasmid for the dominant-negative mutant of Stat3 (S3D) by Dr T Hirano [
35], and the active-form Stat3 (SC) plasmid by Dr James Darnell Jr [
36]. The p1168huIL6P-luc+ plasmid, the control phRL-TK plasmid (for normalization), and either MQ water (mock) or control vector or S3C plasmid or S3D plasmid were co-transfected into AS2 cells using MicroPorator MP-100 (NanoEnTek, Seoul, South Korea). Firefly and Renilla luciferase activities were then measured in cell extracts using the Dual-Luciferase Reporter Assay System (Promega, MI, USA). Data were presented as the ratio of Firefly luciferase activity to Renilla luciferase activity, and normalized with the control group.
RNA extraction and semiquantitative RT-PCR
Total RNA was extracted using the single-step TRIzol method (Invitrogen). For RT-PCR, the first-strand cDNA was synthesized from 0.2 μg of total RNA with oligo-dT primer and the AMV reverse transcriptase (Promega, Madison, WI, USA). The sequences of PCR primers were: IL-6 sense, 5'-CTTTTGGAGTTTGAGGTAGTATACCTA-3'; IL-6 antisense, 5'-GCTGCGCAGAATGAGATGAGTTGTC-3'; β-actin sense, 5'-AGCGGGAAATCG TGCGTG-3' and β-actin antisense, 5'-CAGGGTACATGGTGGTGGTGCC-3'. PCR was performed as follows: after incubation at 94°C for 5 min, IL-6 underwent 30 cycles and β-actin 17 cycles of reaction (94°C for 30 sec, 52°C for 30 sec and 72°C for 1 min). After cycling, the samples were incubated at 72°C for 10 min.
siRNA, shRNA and transfection
To knock-down Stat3, Akt1, Erk1 and Erk2 we used synthetic siRNAs with different targeting sequences: Stat3#1, Stat3#2, Akt1, Erk1 and Erk2 (Ambion, Austin, TX, USA). A scramble siRNA was used as a negative control (Invitrogen). Cells were transfected with siRNA to a final concentration of 50 or 100 nM with MicroPorator MP-100 (NanoEnTek). For long-term suppression of Stat3 expression, Stat3#1 sequence was cloned into the pSUPER vector, kindly provided by Dr R. Agami, The Netherlands Cancer Institute, Amsterdam, Netherlands, as previously described [
37,
38]. Cells were transfected with shRNA using MicroPorator MP-100. After transfection, we treated the cells with Hygromycin-B (Invitrogen) for more than 3 weeks to select stable cell lines containing Stat3 shRNA or control plasmid. The stable cell lines were maintained in media containing 300 μM Hygromycin-B and passaged once in the absence of Hygromycin-B before treatment.
Statistical analysis
Results were expressed as the mean ± standard error of the mean. Statistical significance was set at P < 0.05. Differences between two independent groups were determined using the Student t-test. Differences between two paired groups were determined using paired t-test. All statistical operations were performed using Prism4 (GraphPad Software, San Diego, CA, USA).
Discussion
IL-6 has been found to induce its own self-synthesis in many types of cells through transcriptional mechanisms [
14,
15]. Through this self-synthesis, the secreted IL-6 may induce further IL-6 production in cancer cells in which IL-6 is commonly produced. The IL-6 downstream signaling pathways MEK/Erk, PI3-K/Akt and NF-κB have been also found to be important regulators of IL-6 expression [
12,
16,
18]. Several studies have noted an association between the most well-known IL-6 downstream pathway Jak2/Stat3 and expression of IL-6 as well [
20,
21], but direct proof has been lacking. Some studies, not specifically designed to study this relationship, have found some indication that there may be such a relationship, though some have not. Stat3 decoy oligonucleotide inhibited the expression of IL-6 and IL-10 mRNA [
40] and Stat3 siRNA decreased the expression of IL-6, IL-10 and VEGF in melanoma cells [
25], while the introduction of Stat3 siRNA did not inhibit Cox-2-induced IL-6 expression in the lung cancer cell A549 [
41] and inhibition of Stat3 using antisense oligonucleotide and dominant-negative form of Stat3 in mouse cancer cells increased the expression of IL-6 [
29]. Thus, we designed a series of biochemical and genetic studies of various established cancer cell lines and clinically isolated cancer cells to directly investigate the regulatory role of Stat3 on IL-6.
We found that blocking Jak2/Stat3 pathway as well as blocking the well-known PI3-K/Akt, MEK/Erk, and NF-κB pathways decreased IL-6 autocrine production in AS2 cells. We found that there was a clear association between Stat3 activation status and IL-6 expression pattern as well as paclitaxel resistance in AS2-derived cells and that knocked-down Stat3 by siRNA or shRNA decreased IL-6 expression in AS2 cells. Moreover, we also found that Stat3 also contributed to the elevation of IL-6 in drug resistant cell lines (KB-CPT100 and MCF-7/ADR) and that Jak2/Stat3 pathway cooperated with other IL-6 downstream pathways to regulate the expression of IL-6 in various drug resistant cancer cell lines and in clinically isolated lung cancer cells. Therefore, we clearly proved that Jak2/Stat3 together with the well characterized IL-6 downstream MEK/Erk, PI3-K/Akt and NF-κB pathways, jointly and differentially, regulates the autocrine production of IL-6 in a broad spectrum of established cancer cell lines as well as in clinical lung cancer samples.
Jak2/Stat3, as well as PI3-K/Akt, MEK/Erk, and NF-κB, are key signal pathways involved in cell survival. The blockage of these pathways by inhibitors or siRNAs may reduce cell survival. Thus, the reduction on IL-6 production by inhibitors or siRNAs might be indirectly caused by a reduced cell survival. We therefore investigated the effect of inhibitors and siRNAs on cell survival at the same treatment doses and periods as that used in the ELISA assay in all the tested cell lines. The siRNA transfection did not affect cell viability in any of the tested cells (Figure
4D and Additional file
3, Figure S3A-S3C) and the majority of inhibitors had only limited suppressive effect on cell viability (below 25%) (Additional file
1, Figure S1A-S1H) except that the PI3-K/Akt pathway inhibitor LY294002 had more suppressive activity on the cellular viability by 30 to 50% (Additional file
1, Figure S1C-S1F). However, LY294002 induced much greater (75 to 90%) decrease of IL-6 in these cells (Figure
7A to
7D). There is only one exception that the AG490-induced reductions of cell survival and IL-6 secretion were both about 30% in KB-7D cells (Additional file
1, Figure S1F and Figure
7D). Therefore, the reduction of cell survival might have major contribution to the suppression of IL-6 secretion by AG490 in this cell line. Taken together, we concluded that the reduction of IL-6 by pharmacological inhibitors and siRNAs used in this study are mainly caused by their specific effects on the targets rather than from the suppression of cellular viability.
In addition to Jak2/Stat3 pathway, PI3-K/Akt and MEK/Erk could also contribute to the regulation of IL-6 autocrine production in cancer cells. Thus, the three major down-stream pathways of IL-6 might crosstalk in the regulation of IL-6 autocrine production in cancer cells. In the experiment to test this possibility, we found that these three major IL-6 down-stream pathways were activated by the stimulation of IL-6 with different activating kinetics. There is no significant relationship between each other was found (Additional file
4, Figure S4A and S4B). Though the three pharmacological inhibitors could effectively inhibit both the basal and IL-6 induced phosphorylation of their targeting signal pathway, respectively, in AS2 cells (Additional file
4, Figure S4C and S4D), there was no off-target inhibition effect except that AG490 slightly increased the phosphorylation of Erk. This result is consistent with previous observation of the other studies [
42]. However, AG490 still effectively decreased IL-6 expression in AS2 cells in spite of the slight increase of Erk phosphorylation. Therefore, Jak2/Stat3, MEK/Erk and PI3-K/Akt pathways individually contribute to the regulation of IL-6 autocrine production in cancer cells.
In the pharmacological experiments, Akt and Erk inhibitors significantly decreased IL-6 production in various cancer cells. To confirm these findings, we used siRNA against Akt1, Erk1 and Erk2 in AS2 cells. All of these siRNAs could effectively knock-down the expression of their targets (Additional file
5, Figure S5A and S5B) without affecting cell survival (Additional file
3, Figure S3D and S3E). Knocking-down Akt1 significantly decreased IL-6 secretion in AS2 cells (Additional file
5, Figure S5C). Knocking-down Erk1 significantly decreased IL-6 secretion but knocking-down Erk2 increased IL-6 secretion. The combinational knocking-down of Erk1 and Erk2 resulted in a limited reduction of IL-6 secretion only, compared to the mock and scramble siRNA control groups (Additional file
5, Figure S5D). We observed events of compensation that knocking-down of Erk1 induces an increase of phosphorylation on Erk2 and knocking-down of Erk2 induces an increase of phosphorylation on Erk1 (Additional file
5, Figure S5B). The limited reduction of IL-6 secretion by the combinational treatment may be caused by the compensation effect. Similarly, Lefloch
et al. had also reported the compensational induction of Erk phosphorylation caused by siRNA knocking-down [
43], which supports our speculation. Because, in our study, the siRNA approach is not an idea method to suppress Erk phosphorylation, we used another MEK/Erk inhibitor PD98059 to exclude the possible non-specific activity from U0126. The PD98059 effectively inhibited the phosphorylation of Erk1 and Erk2 (Additional file
5, Figure S5E) and decreased IL-6 secretion dose-dependently in AS2 cells (Additional file
5, Figure S5F). The treatment did not compromise cell survival (Additional file
1, Figure S1I). Collectively, these results confirm that both PI3-K/Akt and MEK/Erk pathways contribute to the regulation of IL-6 autocrine production in cancer cells.
Most studies investigating the regulation of IL-6 expression were performed in cell lines or animal models. In the present study, we took cancer cells from MPE of lung cancer patients and found that IL-6 regulation in human lung cancer samples to be similar to that in cancer cell lines. We found that the NF-κB pathway was the most important, but not an essential, regulator of IL-6 secretion in the tested cancer samples and that Jak2/Stat3 pathway contributed substantially to the regulation of IL-6 secretion in many cancer samples. Different cancer cells utilize different combinations of signals to orchestrate IL-6 autocrine production (Figure
8A and
8B). None of the tested signal pathways was found to be responsible for the regulation of IL-6 autocrine production alone. Instead, the IL-6 downstream signal pathways, including Jak2/Stat3, co-cooperated to control the IL-6 autocrine production in the cancer cells we tested.
In the literature, Stat3 siRNA did not affect COX-2-induced IL-6 expression in A549 cells [
41]. In our study, however, knocking-down Stat3 with Stat3 siRNAs resulted in a decrease in IL-6 expression in AS2 cells and two drug resistant cancer cell lines (KB-CPT100 and MCF-7/ADR). To further evaluate this difference in findings, we also studied the effect of Stat3 on IL-6 expression in A549 cells. We found that Stat3 siRNA (Stat3#1) effectively knocked-down the expression of total amount of Stat3 protein and Stat3 phosphorylation (Additional file
6, Figure S6A) without affecting cell survival (Additional file
3, Figure S3F) but it did not decrease the secretion of IL-6 in A549 (Additional file
6, Figure S6B). Consistently, our biochemical studies, which showed limited side effects on cell survival (Additional file
1, Figure S1J), also demonstrated that inhibition of Jak2/Stat3 pathway did not reduce the secretion of IL-6 in A549 cells, but inhibition of NF-κB and PI3-K/Akt pathways did (Additional file
6, Figure S6C). Our knock-down studies of AS2, MCF-7/ADR, and KC-CPT100 cells and our pharmacological inhibition experiments with seven established cell lines and 20 clinical samples revealed that Stat3 did in fact affect expression of IL-6 in most of the cancer cells we tested.
In Stat3-null mouse embryonic fibroblasts, S3F up-regulated IL-6 mRNA expression suggesting that unphosphorylated Stat3 plays a role in regulating IL-6 expression [
44]. In our study, however, treatment with A490 or over-expression of S3F inhibited Stat3 phosphorylation (Figure
2B and
3A) and reduced IL-6 expression (Figure
2A and
2C, and Figure
3B and
3E) in the Stat3 active AS2 cells. Similarly, AG490 treatment also decreased the IL-6 secretion in various drug resistant cancer cells exhibiting constitutively active Stat3 (Figure
7A to
7E). We hypothesized that unphosphorylated Stat3 may have a basal activity in the regulation of IL-6 expression but tyrosine phosphorylated Stat3 has better activity in the induction of IL-6 expression.
To date, no Stat3 binding site has yet been identified in IL-6 promoter. Using prediction software, we were also unable to find any specific Stat3 binding site 5 kb upstream from the transcriptional start site of IL-6 promoter. However, in the promoter experiments, we showed that a transient transfection of S3C plasmide into AS2 cells increased IL-6 promoter luciferase activity. On the contrary, the transient transfection of S3F plasmid or treatment with AG490 reduced IL-6 promoter luciferase activity in AS2 cells (Figure
2D and
3C). These results suggest that Stat3 might regulate IL-6 transcription at the promoter level. Stat3 has been reported to induce the expression of AP-1 proteins (JunB and c-Fos) [
45,
46] and C/EBPα, β and δ [
47‐
49]. The AP-1 and C/EBP transcriptional factors are major regulators of IL-6 expression [
13]. Therefore, Stat3 may increase the expression of IL-6 indirectly through the regulation of these transcriptional factors. However, it may do so directly by interacting with other transcription factors and co-localizing to IL-6 promoter at non-consensus sites. For example, Stat3 has been shown to interact directly with NF-κB forming a complex that synergistically promotes target genes expression [
50,
51]. Stat3 could also cooperate with C/EBPs [
47], CREB [
52], or AP-1 [
53,
54] to regulate target gene expression by binding to either its consensus sites or the non-consensus regions [
51,
55]. Regardless of how Stat3 contributes to the regulation of IL-6 expression, Stat3 DNA-binding activity is required. Our study demonstrates that over-expression of S3D suppresses IL-6 expression in AS2 cells (Figures
3B and
3D). That S3D is unable to bind to DNA suggests that Stat3 DNA binding activity plays an important role in the regulation of IL-6 expression. Our results will, however, need to be confirmed by further studies that further seek to uncover underlying mechanisms.
Consistent with previous literature [
30,
39], we found that drug resistant cancer cells secreted more IL-6 secretion than the parental cells (Figures
6 and
7), and not only NF-κB, PI3-K/Akt and MEK/Erk but also Jak2/Stat3 pathway contributed to the autocrine production of IL-6 in these cells. In the AS2-derived cells with different Stat3 activation statuses, we found a clear association among Stat3 activation status, IL-6 autocrine production and paclitaxel resistance. Similarly, the AS2 cells stably expressing Stat3 shRNA expressed less IL-6 mRNA, secreted less IL-6 protein, and were more sensitive to paclitaxel than the parental and vector-control cells (Figures
5B, 5C, and
5D). Paclitaxel resistance in these two cells could be modestly restored by adding exogenous IL-6 (Figure
5D), indicating that the IL-6-induced paclitaxel resistance is mediated by both Stat3-dependent and Stat3-independent pathways. By targeting Stat3, we may directly inhibit Stat3-dependent drug resistant mechanisms and inhibit Stat3-independent drug resistant mechanisms indirectly by decreasing IL-6 autocrine production in cancer cells simultaneously.
WLH, Ph.D. (Post-doctoral research fellow, molecular biologist)
HHY, Ph.D. (Post-doctoral research fellow, molecular biologist)
CCL, M.D. (Visiting staff member, pulmonologist)
WWL, M.D. (Associate-principle investigator, thoracic surgeon)
JYC, M.D. (Principle investigator, medical oncologist)
WTC, Ph.D. (Associate-principle investigator, molecular biologist)
WCS, M.D. (Principle investigator, medical oncologist)
Authors' contributions
WLH performed most experiments and wrote initial draft of the paper. HHY established AS2-derived Stat3 mutant cells. CCL and WWL participated in the collection of MPE from lung cancer patients. JYC contributed to the establishment and the assay of the drug resistant cancer cells. WTC contributed to the siRNA and shRNA associated experiments. WCS designed experiments and wrote final version of paper. All authors read and approved the final manuscript.