Background
Colorectal cancer (CRC) is the third most common malignancy worldwide and the fourth leading cause of cancer-related deaths globally [
1]. Although great advances in chemotherapy and radiotherapy in recent decades have reduced the recurrence and improved survival of CRC, 136 830 new cases were diagnosed and 50 310 deaths attributed to CRC in 2014. Invasion and metastasis are the leading cause for CRC-related mortality. The 5-year survival rate for patients with distant metastasis is only 13% [
2].
Lipocalin2 (LCN2, also known as neutrophil gelatinase-associated lipocalin), is dysregulated in a variety of cancers such as pancreatic cancer [
3], prostate cancer [
4], and hepatocellular carcinoma [
5]. It was initially considered as a defender of innate immunity for its capacity to sequester iron, which blocked bacteria growth [
6]. Recently, it has been reported that LCN2 is involved in cancer progression. However, whether the role of LCN2 is pro- or anti-metastasis remains controversial, and its underlying mechanisms have not been clarified. For example, LCN2 prevented metastasis and epithelial mesenchymal transition (EMT) in hepatocellular carcinoma [
7] and was a novel suppressor of invasion and angiogenesis in pancreatic cancer [
8]. Interestingly, LCN2 has also been correlated with disease advancement [
9] and appears to modulate several pathways such as the ERK/slug pathway in prostate cancer [
10] and the Met/FAK pathway in liver cancer [
11] to contribute to EMT progression.
EMT, a vital process, is regarded as a fundamental program in the metastatic cascade by regulating motility and invasion. During this progress, cells lose their epithelial traits, including cell-cell adhesion proteins, like E-cadherin and Zonula Occludens 1 (ZO-1) and acquire mesenchymal traits such as promigratory cytoskeletal proteins like Vimentin and proteases such as matrix metalloproteinase 9 (MMP9) [
12,
13]. Cells with the EMT phenotype can even invade as single cells [
14,
15]. Thus, exploring the regulatory mechanisms in this progress is essential and crucial to improve patient survival.
Among several critical factors involved in EMT progression, NF-κB was the essential factor that orchestrated the inflammatory process [
16], as well as manipulated the initiation and development of a series of cancers. During the process, NF-ĸB could induce several transcriptional factors which potentiated EMT and metastasis of cancers like snail [
17], twist [
18] and ZEB2 [
19]. We hypothesized that LCN2 may modulate EMT and metastasis that were contributed by NF-κB pathway. Notably, the regulation of the NF-κB pathway by LCN2 has not previously been reported.
Here, we provide the first evidence that LCN2 acted upstream of the NF-κB/snail pathway to prevent cancer progression. The results showed that LCN2 significantly inhibited the NF-κB/snail pathway-induced EMT and metastasis both in vitro and in vivo by attenuating its promoter activity.
Methods
Cell culture and reagents
The human CRC cell lines of SW480, SW620, HT-29, and RKO were from the American Type Culture Collection, and 293FT cells were from Life Science (Norway); they were respectively cultured in RPMI 1640 (Gibco, Grand Island, USA) and Dulbecco’s Modified Eagle’s Medium (DMEM, Boster Biological Technology, Co. Ltd, Wuhan, China) supplemented with 10% fetal bovine serum (FBS, Haoyang Bioscience, Tianjin, China) and grown at 37 °C under 5% CO2. SW620 and RKO cells that stably expressed LCN2 (designated as SW620-LCN2 and RKO-LCN2; vector control-OB), and SW480 and HT29 cells with LCN2 knockdown (designated as SW480-sh-LCN2 and HT29-sh-LCN2; vector control-SHB) were established by lentiviral shuttle vector (Detail of vector was in part of vector construction).
Leptomycin B (LMB, Cat. No.431050) was purchased from Merck Millipore (Darmstadt, Germany), and incubated with SW480 for 20nM (10 h), SW620-LCN2 and RKO-LCN2 for 40nM (6 h). Bay11-7082 (B5556, Cat. No. 196870) and JSH-23 (J4455, Cat. No. 749886) were from Sigma Aldrich and added to the culture medium at 50 μM, for 3 h and 50 μM for 1 h, respectively.
Vector construction and small interfering RNA knockdown (siRNA)
The lentiviral shuttle vector pLVX-IRES-ZsGreen1-LCN2 was constructed to stably overexpress LCN2. The EGFP-C2 vector containing green fluorescence protein (GFP) was constructed to stably overexpress snail, snail (55KD) was a fusion protein (snail-30KD and GFP). The 597 bp coding sequence of human LCN2 (NM_005564) and the 794 bp coding sequence of human snail (NM_005985) was amplified from the cDNA templates of human SW480 and subcloned into the EcoRI and BamHI sites of the pLVX-IRES-ZsGreen1 vector (Clontech Laboratories) and EGFP-C2 vector. The lentiviral shuttle vector pLVX-shRNA2 (Clontech) was used to express shRNA from the U6 promoter. In addition to expressing shRNA, pLVX-shRNA2 also expresses the green fluorescent protein ZsGreen1. The target sequence was: shLCN2-1, 5′-GAGTTCACGCTGGGCAACATTAAGA-3′.
For small interfering RNA (siRNA) knockdown of p65, cells were transfected with either p65 or negative siRNA (GenePharma, Shanghai, China) using the PowerFect siRNA Transfection Reagent (SignaGen Laboratories, Gaithersburg, MD) following the manufacturer’s instruction. The siRNAs of p65 sequences are listed in Additional file
1.
Western blot assay
Cell culture medium was concentrated by centrifugation at 4000 rpm with Amicon Ultra-15 (Millipore, Billerica, MA) for 25 min. Cells were scraped, washed 3 times with PBS, and then lysed in buffer consisting of 7 M urea, 2 M thiourea, 4% CHAPS, 65 mM DTT, and 0.2% Bio-lyte (pH3–10, Bio-Rad, Hercules, California, USA) by sonication on ice. The lysates were centrifuged at 12000 rpm for 1 h at 4 °C. Lysate from fresh xenograft tissues was prepared with a DNA/RNA/Protein Isolation kit (R6734-02, Omega) and the nuclear protein was extracted using a Nucleoprotein and Cytoplasm Protein Extract Kit (Kaigen Bioscience, Nanjing, China) following with manufacturer’s instructions.
Subsequently, the protein concentrations were determined by the Bradford method, and aliquots of the protein samples were stored at −80 °C.
Aliquots of protein extracts (15–50 μg) were separated on 10–12% SDS-polyacrylamide gels according to the protein molecular weights. Then, the proteins were electrophoretically transferred onto NC membranes (Bio-Rad, Laboratories, Hercules, California, USA). After blocking with TBS-Tween 20 (TBST) containing 5% low-fat milk, the membranes were incubated with primary antibody overnight at 4 °C. The primary antibodies are listed in Additional file
2. The membranes were then incubated with secondary antibodies (Odyssey, Li-COR, Bioscience, Lincoln, USA) for 1 h at room temperature. Finally, the membranes were developed with the Odyssey system. Loading differences were normalized to a monoclonal GAPDH antibody.
Reverse transcription-polymerase chain reaction (PCR) and real-time PCR
Total RNA was extracted from cells using TRIzol (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol, and its concentration was measured using a spectrophotometer (Eppendorf, Hamburg, Germany). Total RNA (4 μg) was reverse-transcribed using the PrimeScript TM RT Reagent Kit (Takara, Bio, Inc, Otsu, Shiga, Japan). Quantitative real-time PCR was performed using the SYBR Premix Ex TaqTM (Takara, Dalian, China) on a LightCycler480 II (Roche, Diagnostics, Switzerland). The relative expression level was calculated using the 2 − ΔΔCt method and normalized by GAPDH expression. The sequences of the specific primers are listed in Additional file
3.
Enzyme-linked immunosorbent assay
Quantification of serum LCN2 concentration was performed using a commercially available anti-human LCN2 ELISA kit according to the manufacturer’s instructions (DLCN20, R&D system, Minnesota, USA). The absorbance was measured by a microplate reader (Bio-TEK, ELx 800, Saxony, USA) at 450 nm.
Immunofluorescence staining
The established cell lines (SW480-SHB, SW480-sh-LCN2, SW620-OB and SW620-LCN2) were seeded on slides, after 24 h they were fixed in 4% paraformaldehyde for 15 min at room temperature and washed three times with PBS, then permeabilized with ice-cold 0.5% Triton ×-100 for 20 min at room temperature, followed by three washes with PBS. The slides were blocked in PBS containing 10% bovine serum albumin for 30 min at room temperature, and then were incubated with antibodies of E-cadherin, vimentin and snail (Additional file
2) overnight at 4 °C. After washing three times with PBS, cells were incubated with Alexa Fluor 546 donkey anti-rabbit IgG (Invitrogen, Eugene, OR) for 1 h at room temperature. Finally, cells were counterstained with 50 mg/ml 4′,6-diamidino-2-phenylindole for 20 min before capturing images with a confocal microscope (LSM-510 Meta, Zeiss, Germany). Assays were performed in triplicate.
Luciferase reporter assay
Luciferase activity was measured as described previously [
20]. 450 ng of PGL3.0-NF-ĸB-luc and 50 ng of Renilla luciferase plasmids or 450 ng of PGL3.0-luc and 50 ng of Renilla luciferase plasmids were co-transfected into 293FT-LCN2-overexpression, 293FT-LCN2-knockdown and corresponding control cells. After 48 h of treatment, luciferase activity was assayed using the Dual Luciferase Assay System (Promega, Madison, WI) according to the manufacturer’s protocol on a Luminometer (Berthold Technologies, Bad Wildbad, Germany). Each firefly luciferase value was corrected for its co-transfected Renilla luciferase value.
Proliferation assay
For the proliferation assays, 5 × 103 cells were seeded into 96-well plate with 100 μl RPMI 1640 supplemented with 10% FBS, using the Cell Counting Kit (CCK8, Boster Biological Technology, Co. Ltd, Wuhan, China) following the manufacturer’s instructions and detected by a microplate reader (Bio-TEK, ELx 800, Saxony, USA).
A cell suspension (2 × 103 cells) in 2 ml RPMI 1640 supplemented with 10% FBS and 0.3% agar (Affymetrix, Inc, California, USA) were layered onto 6-cm plates (Corning, Inc, Corning, New York, USA) containing 2 ml RPMI 1640 with 10% FBS and 0.6% agar. Plating was carried out in triplicate and repeated at least three times. After 21 days of growth, colonies were photographed.
SW480-SHB, SW480-sh-LCN2, SW620-OB, and SW620-LCN2 cells were separately seeded into six-well plates at 6000 cells/well. The media were changed three times a week, and after 20 days, the cells were fixed in 4% paraformaldehyde for 30 min, stained with 0.5% crystal violet (Beyotime, Biotechnology, Shanghai, China) for 1 h, rinsed three times with PBS to remove excess dye, then photographed and counted.
In vivo mouse tumorigenicity and metastasis assays
SW620-OB (5 × 106), SW620-LCN2 (5 × 106), SW480-SHB (1 × 107), and SW480-sh-LCN2 cells (1 × 107) in 100 μl PBS were inoculated subcutaneously into the left shoulder of 4 to 5-week-old BALB/c nude mice. Tumor length and width were measured with calipers, and tumor volumes were calculated according to the equation: V (mm3) = ½ [width2 (mm2) × length (mm)]. Growth curves were plotted based on mean tumor volume within each experimental group at the indicated time points. Tumor growth was observed for 27 days (for SW620-OB and SW620-LCN2 cells) and 16 days (for SW480-SHB and SW480-sh-LCN2 cells), respectively.
For metastasis assay, 4.7 × 106 cells for SW620-OB or SW620-LCN2 cells suspended in 100 μl PBS were injected into mice through the tail vein. Mice were assessed for lung metastasis. All the mice were sacrificed at 10 weeks and the number of metastatic lung nodules was counted.
In vivo tumorigenicity or metastasis experiments were performed using five mice per treatment group respectively.
In Vitro transwell cell migration/invasion assay
Cell migration and invasion assays were performed in 24-well transwell chambers with 8 μm-pore polycarbonate filter inserts (Corning, Inc, Corning, New York, USA). SW480, SW620 were seeded at a density of 2.5 × 105 each well, and HT29, RKO were seeded at a density of 1.5 × 105 each well in uncoated or Matrigel-coated (BD Biosciences, Franklin Lakes, New Jersey, USA) inserts in 200 μl of serum-free RPMI 1640, respectively. The lower chambers were filled with 600 μl of 10% FBS-supplemented RPMI 1640 as chemoattractant. After 48 h and 72 h, cells on the upper side of the filter were removed and those on the lower surface of the insert were fixed in 4% paraformaldehyde (Sinopharm chemical reagent, Co. Ltd, China) for 15 min and then stained with crystal violet (Beyotime, Biotechnology, Shanghai, China) for 10 min. Migrated cells were then digested in 33% acetic acid and quantified by measuring absorbance at 570 nm with a 96-well plate on a microplate reader (Bio-TEK, ELx 800, Saxony, USA). Three independent experiments were performed.
Wound healing assay
5 × 105 cells in serum-free RPMI 1640 on six-well plates were rinsed three times with phosphate-buffered saline (PBS) and then a sterile 200 μl pipette tip was used to create three separate, parallel wounds. The wells were then washed several times with PBS to remove floating cells. Representative images of cells migrating into the wounds were captured under a microscope immediately and at 48 h after scratch. Three independent experiments were performed.
Immunohistochemical staining and scoring
Immunohistochemical staining by Envision method was performed on formalin-fixed paraffin-embedded slides, which had been dewaxed and rehydrated before antigen retrieval step [
20]. The antibodies and dilutions are listed in Additional file
2. Each specimen was scored according to the proportion of positive tumor cells: 0, negative or <5%; 1, 5–25%; 2, 26–50%; 3, 51–75%; 4, >75%. Score of NF-κB(p65 in nucleus,
n = 126), E-cadherin (cytomembrane,
n = 108), β-catenin (nucleus,
n = 122), Ki67 (nucleus,
n = 126) and snail (nucleus,
n = 124) immunohistochemical staining were from the CRC database in our laboratory. Subsequently, NF-κB, E-cadherin, β-catenin, Ki67 and snail were defined as negative (score 0) and positive (scores 1–4). LCN2 was defined as low (score ≤2) and high (score: 3–4) in survival analysis and comparison of E-cadherin, but was defined as low (score ≤3) and high (score = 4) when compared with β-catenin and Ki67.
Clinical specimens
All 400 patients (219 males, 181 females) were enrolled with informed consent in accordance with a protocol approved by the Ethics Committee of Zhejiang University. The age at diagnosis ranged from 24 to 91 years (median, 61). There were 207 colon specimens and 193 rectum specimens. Of all cases, 307 were tubular adenocarcinoma and 93 were other types of adenocarcinoma. According to TNM staging system, 79 cases were stage I, 120 were stage II, 160 were stage III, and 41 were stage IV. Follow-up data were obtained from the Xiaoshan tumor registry system database. The follow-up period ranged from 1 to 200 months (median, 52).
Statistical analysis
The statistical software SPSS (version 19.0; IBM, New York, NY) was applied. For results from cell lines, all data are reported as mean ± S.D. Unpaired Student’s t tests were used for normally distributed data and non-parametric Mann-Whitney U-tests were used for non-normally distributed data to compare central tendencies. For results in CRC tissues, comparisons of clinicopathological parameters and EMT markers in the LCN2-low and LCN2-high groups were done by the χ
2 test or Fisher’s exact test. Correlation coefficients were calculated by the Spearman rank correlation test. The 5-year survival rate was calculated using the Life-table method, and survival curves were drawn using the Kaplan-Meier method and compared with the log-rank test. Significance was set at P < .05.
Discussion
Here, we provide new evidence directly showing that LCN2 prevents CRC cells from undergoing EMT and metastasis/invasion by targeting the promoter activity of NF-κB, which inhibits the NF-κB/snail signaling pathway. These findings identify the LCN2/NF-κB/snail signaling pathway as highly suitable candidates for therapeutic agents for CRC patients.
It has been reported that LCN2 inhibits tumor metastasis/invasion in diverse cancers [
5,
8,
26]. Here, we identified two aspects of the clinical importance of LCN2 in CRC. First, patients with higher LCN2 and negative NF-κB had a better prognosis. Second, immunohistochemical analysis of CRC tissues also showed that LCN2 expression was positively correlated with E-cadherin expression and negatively associated with nuclear-β-catenin. Thus, we conclude that LCN2 is an important modulator in the progression of CRC.
Our data showed that LCN2 was higher in CRC tissues than in normal tissues or metastatic lymph node lesions, consistent with the TCGA database and with a previous gene expression profiling study, which showed that LCN2 is highly upregulated in CRC tissues [
27]. Elevated levels of serum LCN2 have also been observed in our data. Although the regulatory actions of elevated LCN2 in cancer are not yet fully understood, aberrant LCN2 expression has been found under several conditions. Recent studies have reported that LCN2 expression is elevated in both the colorectal adenoma-carcinoma sequence and cancer progression [
28]. Furthermore, the LCN2 promoter non-methylated status (67.2%) is much higher than its methylated status (32.8%) in breast cancer patients [
29]. In bladder cancer, the LCN2 methylation ratio of the CpG site on the promoter is also much lower in the tumor tissue than in the normal tissue [
30].
Moreover, inflammatory cytokines, such as interleukin (IL)-6, induce LCN2 expression depending on activation of the STAT3 signaling pathway, IL-1β and the tumor cell-derived factor IL-10 also promote LCN2 expression in lung [
31] and breast cancer [
32]. Interestingly, upregulated expression of IL-1β, IL-6, and TNF-α, and sustained activation of the STAT1, STAT3, and JNK pathways, both occur in LCN2-knockout mice along with an increase in the susceptibility to infection with
Klebsiella pneumoniae or
Escherichia coli in the liver and in mice with breast cancer [
33,
34]. It has been noted that inflammation of the colon, including inflammatory bowel disease (IBD) or colitis, increases the risk of CRC by causing a cellular immune response and accumulating genetic alterations that might trigger specific oncogenic pathways [
35,
36]. Besides, the finding that the long-term prognosis of CRC is poorer in patients with IBD than in those with sporadic CRC [
37] implies that elevated LCN2 expression plays a prominent role in CRC progression.
LCN2 has been implicated in the regulation of proliferation in terms of its association with a variety of proliferative cells [
38]. On one hand, LCN2 expression provokes tumor growth and progress in breast cancer [
39] and increases the migration/invasion of pancreatic cancer [
3]. On the other hand, LCN2 leads to apoptosis in leukemia cells [
40] and acts as a suppressor of proliferation and metastasis in hepatocellular carcinoma [
5,
7]. In our study, LCN2 inhibited the proliferation and metastasis/invasion of CRC cells in vitro, as well as tumor growth and lung metastasis in vivo. This is consistent with previous reports that LCN2 suppresses invasion and the liver metastasis of highly metastatic CRC KM12SM cells [
41]. Although the involvement of the NF-κB/snail signaling pathway in promoting metastasis/invasion through EMT has been shown in vivo and in vitro [
23,
42], the regulation of LCN2 in the NF-κB/snail signaling pathway-induced metastasis and EMT in cancer had not been reported.
It is widely accepted that NF-κB induces a significant increase in the expression level of snail [
18,
23,
25], which leads to a remarkable decrease of E-cadherin-mediated intracellular adhesion, subsequently inducing EMT and metastasis/invasion in cancer cells. We tested the hypothesis that LCN2 blocks the NF-κB/snail signaling pathway. Indeed, our results revealed that LCN2 significantly decreased the translocation of p65 into the nucleus by suppressing NF-κB promoter activity, leading to inhibition of the snail-dependent EMT and migration in CRC. Although many studies have shown that NF-κB enhances LCN2 expression to promote progression in a series of cancers [
9,
43], we focused on the regulation of LCN2 in NF-κB/snail pathway-induced cancer progression. The results showed that LCN2 may indirectly regulate NF-κB activity, but the specific mechanisms underlying their interactions in different cancers need further exploration. These results provide a new perspective on the possible underlying mechanism by which LCN2 is involved in metastasis and EMT progression in cancer. Moreover, less nuclear staining of snail was found in the patients with higher LCN2 and negative NF-κB. Therefore, expression of LCN2 combined with NF-κB may be a candidate biomarker for CRC patients.
Collectively, these results clearly demonstrate that LCN2 suppresses proliferation, metastasis/invasion, and EMT by attenuating the promoter activity of NF-κB in CRC, which inhibits the NF-κB/snail signaling pathway. Thus, LCN2 may play a protective role against EMT and metastasis in CRC.
Acknowledgements
We are grateful to Professor Ralf Weiskirchen for generous help with the NF-κB promoter luciferase vector.