Background
Colon cancer is the third most common cancer worldwide, and the second most common cause of cancer-associated death [
1,
2]. Colon cancer arises from chronically inflamed tissues under the immune surveillance of tumor-infiltrating immune cells. Tumor associated macrophages (TAMs) affect many aspects of colon cancer, such as tumor angiogenesis and metastasis. TAMs display two main phenotypes, namely M1 and M2, which usually have contrasting effects on tumor progression [
3]. M1 macrophages are the classically activated macrophages, which are polarized by lipopolysaccharide (LPS) and interferon-γ (IFN-γ). M1 macrophages express interleukin -1β (IL-1β), IL-12 and cytotoxic substances such as inducible nitric oxide synthase (iNOS) [
4]. M2 macrophages are alternatively activated macrophages, which are polarized in the presence of IL-4, IL-10 or IL-13. M2 macrophages express IL-10 and IL-6, and angiogenic factors such as vascular endothelial growth factor (VEGF) [
5]. High levels of M2 macrophages infiltration are associated with poor prognosis of colon cancer patients [
6,
7]. Differentiation of TAMs to M1 or M2 phenotypes is regulated by the tumor microenvironment, including tumor cells [
8‐
11]. Macrophages polarization is regulated by various microenvironmental signals derived from tumor cells. Tumor cells also secretes significant amounts of cytokines to induce the polarization of TAMs. Protein kinase N (PKN) represent a subfamily of protein kinase C (PKC). As one of the three PKN family members, protein kinase N2 (PKN2) was first described by Parker PJ et al. at 1994 [
12]. PKN2 is a PKC-related serine/threonine-protein kinase and functions as effectors of Rho GTPases in diverse cellular pathways. PKN2 is required for cell cycle progression, cell migration, cell adhesion and transcription activation signaling processes [
13,
14] and it plays important roles in tumor cell migration, invasion and apoptosis [
15,
16]. In HeLa cells, PKN2 has been reported to regulate mitotic entry and cytokinesis [
17]. PKN2 regulates epithelial bladder cells speed and directmovement during cell migration and tumor cell invasion [
18]. In human prostate cancer cells, PKN2 contributes to motility pathways and influences differentiation during prostate cancer progression [
19]. PKN2 is highly expressed in triple-negative breast cancer cells and is required to support the growth of cancer [
20]. However, in the intestine, the role of PKN2 in the regulation of tumor proliferation has never been reported, and the immunomodulatory effects of PKN2 have not been discussed.
In the present study, we found that PKN2 expression in colon cancer cells inhibited tumor growth by inhibiting TAM polarization to M2 like phenotype.
Discussion and conclusions
PKN2 has been recognized as a regulator of multiple aspects of cellular events, such as cell cycle progression, cell migration, and cell adhesion. Recently, PKN2 has emerged as a regulator of cancer growth, invasion and metastasis [
19,
20]. However, the role of PKN2 in colon cancer has never been explored. In the present study, we provided the first evidence that low PKN2 expression is strongly correlated with advanced colon cancer and poor prognosis. These results support the notion of PKN2 as a potential tumor suppressor in colon cancer. A previous study reported that PKN2 could promote cell proliferation [
17]. The present study showed that the overexpression or activation of PKN2 had no influence on the proliferation of colon cancer cells in vitro. Interestingly, in a mice xenograft model, PKN2 significantly inhibited tumor growth. Based on these observations, we hypothesized that PKN2 signaling inhibits the proliferation of tumor cells by modulating the reconstitution of the tumor microenvironment rather than acting directly on tumor cells.
Emerging evidence has revealed that TAMs are associated with tumor growth, invasion and metastasis [
24]. M2- polarized TAMs promote tumor growth and invasion, while M1-like polarized TAMs act as tumor suppressers. In several human cancers, a higher density of the M2 macrophages and lower density of the M1 macrophages is associated with worse clinical outcomes. Consistent with the previous analysis, the present study demonstrated that M2 macrophages promote while M1 macrophages inhibit proliferation of the colon cancer cells. Moreover, we demonstrated that macrophages cocultured with low level PKN2 cancer cells promote, while macrophages polarized by high level PKN2 cancer cells inhibit colon cancer cell proliferation.
Most TAMs exhibit an M2- phenotype [
25]. Macrophage polarization within tumor tissues is regulated by various microenvironmental signals derived from tumor cells [
26,
27]. We provided the first evidence that the number of M2-type TAMs in clinical colon cancer tumor tissues was negatively correlated with PKN2 expression in tumor cells. This finding prompted us to investigate whether the expression of PKN2 in colon cancer cells can mediate macrophage polarization. We demonstrated that PKN2 expression in colon cancer cells inhibits M2-like polarization of TAMs both in vitro and in vivo. Wild-type PKN2 overexpression in colon cancer cells reduced the M2 polarization of both human CD14
+ PBMCs and mouse TAMs. However, the overexpression of PKN2 with the K686R mutant, which abolished the ATP binding and reduced the catalytic activity of the protein [
23,
28], significantly promoted the M2 polarization both in vitro and in vivo. These findings promoted the inhibition effect of PKN2 on the M2 polarization of TAMs.
Tumor cells secrete significant amounts of cytokines, such as IL4 and IL10, to promote M2 phenotype polarization in tumor microenvironment [
8]. IL4 has been recognized as a potential tumor activator. Studies have shown that IL4, secreted by follicular helper T cells, downregulates antitumor immunity [
29]. IL4 is also involved in radiation-induced aggressive tumor behavior in human cancer cells [
30]. Clinical-based studies suggest that polymorphisms of IL4 and IL4R can affect susceptibility to gastrointestinal cancer [
31]. Moreover, IL4 promotes M2 macrophage activation, which further induces cancer metastasis [
32]. IL10 is an immunosuppressive cytokine that may facilitate carcinogenesis by down-regulating interferon-gamma production and supporting tumor escape from the immune response. IL10 is significantly elevated in the serum and tumor microcirculation of patients with advanced stages of several cancers [
32,
33]. In colorectal cancer, IL10 is mainly secreted by cancer cells, and polarizes TAMs to the M2 phenotype, which in turn promotes cancer cell migration and metastasis [
6]. In the present study, we investigated the role of PKN2 in cytokines production in human colon cancer cells. We found that IL10 and IL4 were declined in PKN2-WT overexpressed cancer cells but elevated in PKN2-DN overexpressed cells. PKN2 inhibited the expression of IL10 and IL4 via regulating the transcription of the two cytokines. Furthermore, blocking IL10 and IL4 attenuated the upregulated M2 macrophages upon PKN2-depletion. Therefore, we concluded that PKN2 decreases the expression and secretion of IL10 and IL4 in colon cancer cells and eventually inhibits M2 macrophage polarization.
Mitogen-activated protein kinases (MAPKs) are a widely conserved family of serine/threonine protein kinases involved in many cellular programs, such as cell proliferation, differentiation, motility and death [
34]. This family primarily comprises three kinase groups: the stress-activated protein kinases/c-Jun NH2-terminal kinases (Erk1/2), a second stress-activated MAPK group (p38 MAPKs) and a third class of stress-activated MAPK (Erk5) [
35]. The present analysis of the KEGG pathway for gene expression profiling revealed a relation between PKN2 and MAPK pathways. Further study showed that PKN2 negatively regulates the phosphorylation of Erk1/2 but has no effect on p38 or Erk5. We further demonstrated that the MEK inhibitors partly blocked the elevated IL4 and IL10 in response to PKN2 knockdown. MEK inhibitors also reduced the M2 polarization of monocytes cocultured with PKN2 knockdown colon cancer cells. These results suggested that PKN2 contributed to the polarization of TAMs via regulating the Erk1/2-IL4/IL10 pathway.
Several TFs located downstream of Erk1/2, including Elk-1, FoxO3, CREB, Pax6 and STAT1/3 [
36‐
39]. Among these TFs, Elk-1 and CREB was regulated by PKN2-Erk1/2 pathway, as assessed by our TFs activity assay. CREB is a phosphorylation dependent TF that stimulates transcription by binding to the DNA cAMP response element (CRE). CRE contains the highly conserved nucleotide sequence 5’-TGACGTCA-3′, and CRE sites are typically found within the promoter or enhancer regions of genes. CREB is activated by Erk1/2 and induces several biological processes [
39‐
41]. Elk-1 is a TF that binds to purine-rich DNA sequences. Elk-1 forms a ternary complex with SRF and the ETS and SRF motifs of the serum response element on the promoter region of genes. Elk-1 is a typical target of Erk1/2, and the Erk1/2-Elk-1 pathway participates in the expression of many genes, such as IL-1β, collagen and TNF-α [
42‐
44]. In the present study, we demonstrated that Elk-1 and CREB could be phosphorylated by Erk1/2, which mediated the expression of IL4 and IL10 by directly binding to their promoters. This process was suppressed by PKN2, explaining the mechanism by which PKN2 inhibited the expression of IL4 and IL10.
DUSP6 is a cytoplasmic MAP kinase phosphatase and expressed in a variety of tissues [
45]. DUSP6 comprises a conserved C-terminal catalytic domain and an N-terminal regulatory non-catalytic domain connected by a linker region. The linker region of DUSP6 contains important putative regulatory elements, including several residues subjected to phosphorylation, an active nuclear export sequence (NES) and a KIM-like motif [
46‐
49]. Additionally, the linker region plays an important regulatory role [
50,
51]. DUSP6 is an Erk1/2 specific phosphatase that specifically binds to and inactivates the Erk1/2 MAP kinases in mammalian cells [
22]. We demonstrated that wild-type PKN2 elevated the phosphatase activity of DUSP6. Moreover, we uncovered the inner mechanism that PKN2 directly binds to the linker region of DUSP6 and promotes DUSP6 activity in Erk1/2 dephosphorylation.
The present study reports a new biological role for PKN2 in promoting the alternative activation of TAMs and inhibiting colon tumor growth. PKN2 reduced the polarization of TAMs towards the M2 phenotype by inhibiting Il10 and IL4 expression in colon cancer cells. PKN2 directly binds to DUSP6 to inactivate Erk1/2 and further suppress the transcriptional activities of CREB and Elk-1 by reducing their phosphorylation. Overall, we uncovered a novel role for PKN2 in the regulation of macrophage polarization and tumor growth in colon cancer (Fig.
7i). These findings suggest that targeting the PKN2 signaling pathway may be a potential therapeutic strategy for the treatment of colon cancer.
Methods
Collection of human colon cancer samples
Ninety samples from colon cancer patients who underwent surgery in Nanfang Hospital between June 2007 and April 2010 were obtained during operations. Seven polyp samples, 14 adenoma samples, 14 metastasis adenocarcinoma samples and ten normal colon tissue samples were collected. The diagnosis of cancer was confirmed by pathology. Patients with at least 5-year follow-up were included in this study. All specimens were acquired after signed informed consent using procedures approved by the Ethics Committee of Nanfang Hospital. This research was approved by the Ethics Committee of Nanfang Hospital. Tumor staging was determined according to AJCC Cancer Staging Manual. Patients’ characteristics and histological data are shown in Table
2. A tissue microarray was constructed.
Table 2
Characteristics of patients with colon cancer
Gender |
Male | 46 | 51.11 |
Female | 44 | 48.89 |
Age |
≤ 55 | 12 | 13.33 |
> 55 | 78 | 86.67 |
Tumor location |
Right Hemicolon | 40 | 44.44 |
Left Hemicolon | 50 | 55.56 |
Pathological type |
Adenocarcinoma | 88 | 97.78 |
Signet ring cell carcinoma | 2 | 2.22 |
Undifferentiated carcinoma | 0 | 0.00 |
Cell culture and transient transfection
The human colon cancer cell line HCT116, RKO, SW480, HCT8 and HT-29 were obtained from Cell Bank of Typical Culture Preservation Commission, Chinese academy of Sciences. CD14
+ monocytes were isolated from whole blood collected from healthy donors. HCT116 and HCT8 cells were cultured in RPMI1640 medium (Invitrogen), SW480 and HT-29 cells were cultured in DMEM (Invitrogen). Monocytes and macrophages were cultured in IMEM (Invitrogen). All medium was supplemented with 10% fetal bovine serum (Invitrogen) and 1% penicillin/streptomycin (Invitrogen). For transient transfection, cells were transfected with plasmid or siRNA using Lipofectamine 2000 and Opti-MEM (Invitrogen), according to the manufacturer’s instructions. Sequences of the siRNAs are summarized in Additional file
1: Table S2.
CD14+ peripheral blood mononuclear cells isolation
The whole blood was collected from healthy donors. PBMCs were isolated by density gradient centrifugation using Ficoll-Paque™ Plus (Amersham Pharmacia Biotech) according to the manufacturer’s instructions. The diluted cellular fraction was overlaid onto the Ficoll-Paque Plus and subjected centrifugation at 900×g for 30 min. PBMCs were collected and washed twice with MACS® Buffer (Miltenyi Biotec GmbH) by centrifugation at 450×g for 10 min. Immediately after collection, CD14+ monocytes were isolated from PBMCs using a positive magnetic bead-assisted sorting assay (MC CD14 Monocyte Cocktail, human, Miltenyi Biotec GmbH), according to the manufacturer’s protocol. CD14+ monocytes purity was always above 95% as assessed by flow cytometry.
Tumor xenograft study
All protocols for animal research were approved by the Animal Care Committee of the Southern Medical University. Female nude BALB/c mice were raised in specific pathogen-free conditions with a 12-h light/dark schedule at 25 °C. To generate subcutaneous mice colon cancer xenografts, 2.5 × 106 PKN2-WT/ PKN2-K686R/ wild type control (WT) HCT116 cells were injected subcutaneously, respectively (n = 10/group). Tumor size was measured once every other day using vernier caliper. Tumor volume was calculated based on two perpendicular measurements and using the formula: volume = (length × width2)/2. Fifteen days after tumor cell inoculation, all mice were sacrificed, and the tumors were removed for further FACS, magnetic bead-assisted sorting assay and IHC analysis.
Isolation of macrophages and tumor cells from mice tumor tissue
Single cell suspensions were prepared from fresh tumors using Tumor Dissociation Kit (Miltenyi Biotec GmbH). Cells were then immediately separated using a negative magnetic bead-assisted sorting assay (Mouse Cell Depletion Kit). TAMs were separated from the positive cell suspensions using CD11b+ magnetic beads. All operations were performed according to the manufacturer’s protocol.
Flow cytometry
The cell suspension was collected and washed twice with MACS® Buffer, and blocked by FcR Blocking Reagent (human/mouse) (Miltenyi Biotec GmbH). Cells were then stained with CD206-FITC (human), CD206-PE (mouse), CD86-PE (human), CD16/32-APC.Cy7 (mouse), CD14-APC (human) or CD11b-PE (mouse) antibodies (BD Biosciences, La Jolla, CA, USA), respectively. Separated mouse macrophages were permeated and fixed using Cytofix/CytoPerm Plus™ kit (BD Biosciences) following the instruction, then stained with CD206 and CD16/32 antibodies. Cells were examined with a BD Accuri C6 flow cytometer (BD Biosciences) and all the tests were controlled by the homologous isotype control antibodies.
Immunohistochemistry
The immunohistochemistry (IHC) staining was performed as described [
52]. Microarray chips and mice xenografts were stained with anti-PKN2, anti-CD68, anti-CD206, anti-CD86, anti-CD163, anti-CD16/32, anti-DUSP6, anti-Ki67 (Abcam), anti-iNOS (Santa Cruz Biotechnology), and anti-phospho-Erk1/2 (Thr202/Tyr204), anti-Erk1/2 (Cell Signaling) antibodies. Control staining with only secondary antibodies was included to ensure specificity. Mouse IgG monoclonal-Isotype control and rabbit IgG polyclonal-Isotype control (Abcam, Cell Signaling & Santa Cruz) were used as negative controls. Staining was independently assessed by two experienced pathologists (Wanfu Xu and Junhong Zhao) blinded to the clinical characteristics of the patients. The score for PKN2 and DUSP6 staining was based on the integrated staining intensity and the proportion of positive cells. Staining intensity was scored as follows: 0 = no color; 1 = yellow; 2 = light brown; and 3 = dark brown. The proportion of immune-positive tumor cells (number of positively labeled tumor cells / number of total tumor cells) was scored as follows: 0, positive cells <10%; 1, 10%- 40% positive cells; 2, 40%- 70% positive cells; and 3, positive cells ≥ 70%. The final score was determined by adding the staining intensity score and average proportion of positive cells score and expressed as follows: 0, negative staining, marked -; 0-2, weak expression, marked +; 3-4, moderate expression, marked ++; and 5-6, strong expression, marked +++. IHC staining of CD68/iNOS/CD206 was calculated by the positive cell numbers in the stroma per high field. IHC staining of p-Erk1/2 was calculated by the positive nuclear tumor cell numbers per high field. All the percentages/numbers of positive cells were expressed as the average of six randomly selected microscopic fields.
Luciferase assay
Cells of 80% confluence were transfected using Lipofectamine 2000. Luciferase reporter gene plasmid and pRL-TK Renilla luciferase plasmid were co-transfected per well of a 12-well plate. Cell extracts were prepared at 22 h after transfection. The luciferase activity was measured with a Dual Luciferase Reporter Assay System (Promega).
Co-immunoprecipitation (co-IP) assay
Pretreated cells grown in 6 cm dishes were rinsed twice with precooled PBS, and lysed with 400 μl of ice-cold lysis buffer on ice for 30 min, followed by centrifugation at 12000×g for 10 min. Supernatants were incubated with anti-DUSP6, anti-PKN2 (Abcam), anti-HA, anti-flag (Cell Signaling) or control IgG at 4 °C overnight on a rotator, followed by addition of 30 μl prewashed protein A/G agarose beads for another 2 h. After extensive washing with a diluted lysis buffer, the lysate was used for western blot analysis.
Western blot analysis
The cells were lysed, and proteins were extracted through standard protocols. The proteins were separated by SDS-polyacrylamide gel electrophoresis and subjected to western blot analyses. Protein bands were detected by the chemi-luminescence method. Specific primary antibodies against PKN2, GFP, DUSP6, p-Elk-1 (Ser383), Elk-1 (abcam), p-Erk1/2 (Thr202/Tyr204), Erk1/2, HA, flag, p-CREB (Ser133), CREB, Erk5, JNK1, JNK2, JNK3, p38, p-Erk5 (Thr218/Tyr220), p-JNK (Thr183/Tyr185), p-p38 (Thr180/Tyr182) (Cell Signaling) were used. GAPDH (Cell Signaling) was used as a loading control.
RNA isolation, reverse transcription (RT) and real-time PCR
Total RNA from tissues and cultured cell lines was isolated using the Trizol reagent (Invitrogen) according to the manufacturer’s instruction. Primers for real-time RT-PCR were designed using Primer Express v2.0 software (Applied BioSystems). Sequences of the primers are summarized in Additional file
1: Table S3. RT was carried out with the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen) according to the manufacturer’s protocol. Real-time PCR was carried out using SYBR Green I (Applied BioSystems). The data were normalized to the geometric mean of housekeeping gene GAPDH and calculated as 2
−ΔΔCT.
Cell counting kit-8 (CCK8) assay
The pretreated cells were seeded into a 96-well plate. The cells were incubated with CCK8 reagent (DingGuo Bio) at 37 °C for 2 h and absorbance at 450 nm were measured using a microplate reader (BioTek).
Cell cycle analysis
Cells were detached using trypsinization, washed twice with precooled PBS, and fixed in 70% ethanol at −20 °C overnight. The fixed cells were suspended in 100 μg/ml of RNaseA (KeyGen BioTECH) and 50 μg/ml of propidium iodide (PI) (KeyGen BioTECH) and incubated at room temperature for 40 min in the dark. After filtration, the cell cycle was examined by flow cytometry.
Apoptosis assessment
Following treatment, cells were washed with PBS and then stained using the Annexin V-FITC Apoptosis Detection Kit (Affymetrix eBioscience) according to the according the instruction. Cells were analyzed with a FACS flow cytometer (BD Biosciences).
Gene expression profiles analysis
Total RNA was isolated from pretreated cells with Trizol. RNA of HCT116 cells were labeled and hybridized to Human OneArray v7 (Phalanx Biotech Group) which contains 29,204 DNA oligonuceotide probes, and each probe is a 60-mer designed in the sense direction. Probes correspond to the annotated genes in RefSeq v42 and Ensembl v59 database. Signal intensity was normalized for each microarray and genes with a signal below 100 were ignored. Genes with a fold change equal to or higher than 2 compared with the control were picked out as potential targets.
Plasmids and siRNA
HA tagged plasmid PKN2-WT was constructed by ligating full-length open reading frame (ORF) of wild type PKN2 (1-936aa, Homo sapiens) and cloned into a expression vector pCMV-N-HA (Beyotime Biotech). PKN2-K686R plasmid was generated by PKN2-WT plasmid with a K686R point mutation at the ATP binding site. The expression vector pCMV-N-HA was used as control. The expression vector pCMV-N-flag (Beyotime Biotech) was used to construct plasmids that encode wild type DUSP6 with a N-terminal flag tag. pCMV-DUSP6-WT expression plasmid was generated by ligating full-length open reading frame of DUSP6 (1-381aa, Homo sapiens) into pCMV-N-flag. The truncation mutant plasmids of DUSP6 were generated by ligating part of DUSP6’s ORF into pCMV-N-flag (150-205aa; 1-205aa; 1-150, 205-381aa). The flag tagged plasmids containing full-length ORF of Elk-1 and CREB were purchased from Vigene Biosciences. The luciferase reporter plasmids were generated by ligating -1500 bp~0 bp of the promoter sequences of IL4 and IL10 into the pGL3-ENHANCER plasmid (Promega Corp.). pRL-TK Vector was purchased from Promega.
Lentivirus infection and stable clone selection
The human shRNA sequences specifically targeting PKN2 (PKN2 shRNA#1: 5′- CCGGTACTTTGGAAGTTCGTCTTATCTCGAGATAAGACGAACTTCCAGTATTTTTG-3′; PKN2 shRNA#2: 5′-CCGGGCAGGAATTAAATGCACATATCTCGA.
GATATGTGCATTTAATTCCTGCTTTTT -3′) were cloned into pGLVH1/ GFP + Puro vector (Genepharma). The expression construct of PKN2-WT (human) was generated by ligating full-length ORF of wild type PKN2 (1-936aa, Homo sapiens) and cloned into pGLV3/H1/GFP + Puro vector (Genepharma). PKN2-K686R mutant (human) was created with a dominant negative (DN)(K686R) point mutation at the ATP binding site. Lentivirus was produced and collected after plasmid transfection of 293 T cells. HT-29 and SW480 cells were transduced with PKN2 shRNA or scramble shRNA (shCTL) lentivirus expressing GFP. SW480 and HCT116 cells were infected with PKN2-WT (human), PKN2-K686R or control(Vector) lentivirus. Stable cell lines were selected by puromycin treatment (2 μg/ml) for 2 weeks. Knockdown or overexpression of PKN2 was confirmed by Western blotting.
Soft agar assay
1 × 10
6 monocyte-derived macrophages were collected and seeded into the upper chamber of 24-well plates (0.4 μm pore size). Colon cancer cells were used for soft agar assay as previously described [
53]. 1 × 10
4 HCT116 cells were cultured for 14 days. The medium was changed every 48 h. Viable colonies larger than 50 μm were counted.
Macrophage and colon cancer cell co-culture
After infected with shCTL/shPKN2-1/shPKN2-2, or vector/ PKN2-WT/ PKN2-K686R lentivirus, and selected the stably expression clones, the colon cancer cells were seeded into the upper chamber of 24-well plates (0.4 μm pore size) (1~9 × 105cells/well) (Corning Corp., NY, USA). CD14+ monocytes were added in the lower chamber of the transwell apparatus according to the E:T ratio. Cells were co-cultured for indicated time and then harvested for subsequent experiment.
Transcription factor activity array
Nuclear extracts were prepared using Nuclear Extract Kit (Panomics, CA, USA) according to the manufacturer’s instructions. The concentration of nuclear protein was determined using the bicinchoninic acid protein assay reagent kit (Pierce, Rockford, IL) to normalize for the amounts of protein within each experiment. TF array analysis (Panomics) was used to profile activities of 345 TFs. Any spots with a two-fold increase or decrease are considered significant.
Chromatin immunoprecipitation (ChIP) assay
ChIP analysis was performed on colon cancer cells transfected with pSuper, PKN2, or shPKN2 using the Pierce™ Magnetic ChIP Kit (Thermo Fisher) according to the manufacturer’s instructions. PCR analysis were performed on immunoprecipitated DNA. After amplification, PCR products were separated on 1% agarose gels and visualized by ethidium bromide. Sequences of primers for promoter region used in this study are showed in Additional file
1: Table S4.
ELISA
ELISA was conducted according to the instructions. Concentrations of IL-4 (human), and IL-10 (human) in the culture supernatant of treated cells were measured with the use of a commercially available kit (CUSABIO).
Kinase activity assay
Pretreated cells were dealed with immunoprecipitation assay as described above, and pulled down by anti-DUSP6, anti-HA or anti-flag antibodies, respectively. PKN2 activity was determined with Universal Kinase Activity Kit (R&D Systems) following the manufacturer’s instruction. DUSP6 activity was determined with Phosphatase Assay kit (Sangon Biotech).
Statistical analysis
Data are presented as means ± SEM from at least three experiments. All statistical analyses were performed using SPSS 13.0 (SPSS Inc.). Student’s t test was used to compare control and treatment groups. The Kaplan-Meier estimation method was used for overall survival analysis, and a log-rank test was used to compare differences. P < 0.05 was considered to be significant.