The authors have declared that no competing interests exist.
Conceived and designed the experiments: DPO JHW TGC HPR. Performed the experiments: DPO LB JFW DRG. Analyzed the data: DPO JHW TGC HPR. Contributed reagents/materials/analysis tools: TGC JHW HPR. Wrote the paper: DPO.
Surgery induced inflammation is a potent promoter of tumour recurrence and metastasis in colorectal cancer. The recently discovered family of Nox enzymes represent a major source of endogenous reactive oxygen species (ROS) and are now heavily implicated in tumour cell metastasis. Interestingly, Nox enzymes can be ‘purposefully’ activated by inflammatory cytokines and growth factors which are present in abundance in the peri-operative window. As colon cancer cells express Nox enzymes and Toll-like receptor 4 (TLR-4), we hypothesised that LPS may potentiate the ability of colon cancer cells to metastasise via Nox enzyme mediated redox signalling. In support of this hypothesis, this paper demonstrates that LPS induces a significant, transient increase of endogenous ROS in SW480, SW620 and CT-26 colon cancer cells. This increase in LPS-induced ROS activity is completely abrogated by a Nox inhibitor, diphenyleneiodonium (DPI), Nox1 siRNA and an NF-κB inhibitor, Dihydrochloride. A significant increase in Nox1 and Nox2 protein expression occurs following LPS treatment. Inhibition of NF-κB also attenuates the increase of Nox1 and Nox2 protein expression. The sub-cellular location of LPS-induced ROS generation lies mainly in the endoplasmic reticulum. LPS activates the PI3K/Akt pathway via Nox generated ROS and this signal is inhibited by DPI. This LPS activated Nox mechanism facilitates a significant increase in SW480 colon cancer cell adhesion to collagen I, which is inhibited by DPI, Nox1 siRNA and a PI3K inhibitor. Altogether, these data suggest that the LPS-Nox1 redox signalling axis plays a crucial role in facilitation of colon cancer cell adhesion, thus increasing the metastatic potential of colon cancer cells. Nox1 may represent a valuable target in which to prevent colon cancer metastasis.
Colorectal cancer is the second most common cause of cancer related mortality in the western world
Tumour cell adherence is an essential step of the metastatic cascade. Recent evidence has demonstrated how exogenous surgery-induced reactive oxygen species (ROS) enhance the ability of circulating tumour cells to adhere to the endothelial lining by creating intercellular gaps, allowing tumour cells to adhere preferentially to the exposed extra-cellular matrix. These destructive, cytotoxic effects of ROS occur at high levels. However, at low levels, endogenous ROS can promote cell survival through regulation of redox sensitive survival pathways such as PI3K/Akt, which has been heavily implicated in facilitating tumour cell metastasis.
Nox enzymes are a major source of endogenous ROS generation in response to inflammatory mediators such as cytokines, growth factors and hypoxic conditions, all of which are elevated in response to surgical trauma
Lipopolysaccharide (LPS) or endotoxin is a potent trigger of host inflammatory responses in the peri-operative window. LPS is a gram negative bacterial antigen that translocates across the bowel wall following major surgery or during a septic episode, resulting in an endotoxaemia
As recent evidence suggests, successful tumour cell metastasis is promoted by the destructive effects of exogenous ROS. We hypothesised that endogenous non-toxic levels of ROS can also play a major role in orchestrating tumour cell metastasis. Herein, we demonstrate how an LPS-Nox1 signalling axis gives rise to a significant increase in the adhesive ability of colon cancer cells. LPS activation of Nox activity occurs in a NF-κB dependent manner which results in a transient increase of intracellular ROS. This transient rise of intracellular ROS causes phosphorylation of redox sensitive Akt. Altogether, these data suggest that the LPS-Nox1 redox signalling axis plays a crucial role in facilitation of colon cancer cell adhesion, thus increasing the metastatic potential of colon cancer cells.
The human colon cancer cell lines, SW480, SW-620 and CT-26 were obtained from the American Type Culture Collection (Manassas, VA). Cells were maintained in a sub-confluent state using RPMI (Roswell Park Memorial Institute) culture medium supplemented with 10% fetal calf serum, 1% penicillin/streptomycin, and 4 mmol/L of L-Glutamine all from Sigma Aldrich, Dublin, Ireland. Cells were incubated at 37°C in a humidified incubator with 5% CO2. Cells were plated overnight prior to LPS treatment to allow attachment.
LPS derived
5×104 cells per well were plated overnight in a 24 well plate. Cells were treated with LPS at a concentration of 1 ug/ml for 0,10,20,30,40,50,60 minutes and 24 hours. ROS production following LPS treatment was monitored using 2′,7′ dichlorodihydrofluorescin diacetate (DCF) (Molecular Probes, Leidin, Netherlands). Cells were trypsinised, centrifuged for 5 minutes at 1,000 rpm and then collected. DCF was then added at a final concentration of 50 µM and samples were incubated for 15 minutes at 37°C before analysis on a FACScan flow cytometer (Becton Dickinson, Oxford, UK). Dichlorofluorescin (DCF) fluorescence, is generated when DCF interacts with ROS. ROS was thus measured at fluorescence channel 1 (FL-1) (530 nm) with excitation at 488 nm. CellQuest software (Becton Dickinson) was used for data analysis.
Western Blotting was performed to measure protein content of cells treated with 1 ug/ml LPS for 0,10,20,30,40,50,60 minutes and 24 hours. Cells were then trysinised and centrifuged for collection. Whole cell extracts were then obtained. Cell pellets were washed with ice cold PBS and then resuspended in cell lysis buffer 50 mM Tris–HCl pH 7.4, 150 mMNaCl, 1 mMNa3VO4, 1 mMNaF, 1 mMEGTA,1% Nonidet P-40, 0.25% sodium deoxycholate containing protease inhibitors (Roche Diagnostics, Lewes, UK) and 0.2 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride] and incubated on ice for 20 min. All debris was removed by centrifugation at 4°C and protein concentration was quantified using the Bio-Rad protein assay (Bio-Rad, Hemel Hempstead, UK) using bovine serum albumin as a standard. Equivalent amounts of protein were resolved using denaturing sodium dodecyl sulphate–polyacrylamide gel electrophoresis, followed by transfer to nitrocellulose membranes (Schleicher & Schuell, Whatman, Dassel, Germany). Membranes were blocked with 5% (w/v) non-fat dry milk in Tris-buffered saline/0.1% Tween-20 for 1 h at room temperature. They were incubated at 4°C overnight with the appropriate dilution of primary antibody (1∶1000 unless otherwise stated). After 4×5-minute washes with Tris-buffered saline/0.1% Tween-20, blots were incubated with the corresponding peroxidase conjugated secondary antibody (dilution 1∶1000) for 1 hr. They were then washed again and developed with enhanced chemiluminescence reagent (Amersham Biosciences, Buckinghamshire, UK). Probing for GAPDH (1∶5000) was used to determine equal loading of protein.
In order to localise ROS generated on treatment with LPS in SW480 cells, cells were cultured for 48 hours before the experiment in glass bottomed dishes (35 mm Petri-dishes with 14 mm microwells; MatTek Corporation, Ashland, MA, USA). Cells used for live imaging were incubated in 50 µM DCF and 1 µM ER tracker dye (Molecular probes, Leiden, the Netherlands) for 1 hour at 37°C. One micro-molar LPS was added to the cells containing the dyes 30 minutes before imaging. Where stated, cells were treated with 10 µM DPI with DCF and ER tracker dyes for 1 hour at 37°C and 1 µM LPS was added to the cells 30 minutes before imaging. After this incubation with the dyes, cells were rinsed and imaged in either growth medium containing LPS or LPS and DPI or growth medium alone. Concentrations for DPI and LPS were maintained in the growth medium wash and while cell imaging. SW480 cells were also stained with 8 µM Menadione for 1 hr together with 10 µM MitoPY and 1 µM mitotracker deep red (Invitrogen). Menadione was used as a positive control. Additionally, cells stained with MitoPY were treated with LPS 1 µg/ml prior to imaging. SW480 cells were imaged with a multiphoton laser scanning microscope Flouview1000 MPE (Mason Technology, Dublin, Ireland) with an Infrared red Ti:Sapphire Laser that is mode locked. Images were acquired and visualised using an XLPLN 25×WMP water immersion objective (1.05 numerical aperture: Olympus Optical GmbH, Hamburg, Germany) and images were stored with an Olympus flouview1000 software (Mason Technology, Dublin, Ireland). During acquisitions, detection settings were kept constant for DCF to allow direct comparison of ROS levels, where cells treated with LPS were compared to untreated or LPS and DPI treated cells. Images have been represented as a single slice from a Z-stack projection.
Quantitative PCR was performed on oligo-dT generated cDNA using the MJ Research Opticon 2 detection system in combination with the Quantitect SYBR Green PCR Master Mix (Qiagen, Crawley, UK). The primers for Nox1, Nox2 and β-Actin were purchased as Quantitect Primer Assays (Qiagen). The following PCR parameters were used for each primer set: denaturing at 95°C for 15 min, followed by 45 cycles of 94°C for 15 seconds, annealing temperature of 56°C for 30 seconds and extension at 72°C for 30 seconds. RNA samples were analyzed in triplicate, and Nox1 or Nox2 expression relative to β-Actin was determined via the 2−ΔΔCt method.
96 well plates were coated with 150 µl/well of 0.01% collagen I overnight at 4°C and then blocked with 1% BSA for 1 hour at 37°C before seeding cells. 5×104 cells were resuspended in 100 µl of culture medium and seeded in each well. Cells were incubated at 37°C for 1 hour. Each well was then washed with PBS and then 1% crystal violet was used to stain cells for 20 minutes. Plate was washed 3 times and then left overnight to dry at room tempeture. Crystal violet staining was then dissolved in 10% acetic acid and concentration was determined by measuring absorbance at 570 nm using a spectrophotometric microplate reader.
Statistical analysis was performed using SPSS version 18.1 for Windows (SPSS, Dublin, Ireland). Data are given as mean± SD. Statistical significance was evaluated by Student's t-test for comparisons between groups, and analysis of variance (ANOVA). Densitometry on western blots was performed using the programme ImageJ (from
SW480, Sw620 and CT-26 cells express TLR-4 and were thus chosen to examine the effect of LPS induced ROS production
(a) LPS induced a transient, significant rise in ROS activity in SW480, SW620, Ct-26 colon cancer cells. SW480, SW620 and CT-26 cells were treated with LPS (1 µg/ml) at twenty, forty and sixty minutes. DCF fluorescence was measured using FACScan flow cytometer and CellQuest software. These results are compiled in the above histograms. The y-axis represents cell counts and the x-axis measures the fluorescence level. A shift to the right along the x-axis represents a higher level of DCF fluorescence and thus ROS production. All histograms show a control (solid black line) with a colour line representing an increase in DCF fluorescence. The effect of LPS on ROS activity in SW480, SW620 and CT-26 colon cancer cells was quantified using CellQuest to measure the geometric means of the curves and compared to untreated controls tested on the same day, and compared in a bar chart. (b) LPS induces a dose dependent ROS burst at 40 minutes. A flow cytometry histogram demonstrates a dose dependent shift in ROS activity. (Solid black line = untreated, green = 0.1 µg/ml, red = 1 µg/ml, blue = 10 µg/ml. The dose dependent effect was quantified and compared in a bar chart. * P<0.05. These data are representative of 3 independent experiments.
We next sought to examine if this LPS-induced increase in endogenous ROS was responsive to variations in the dose of LPS. We show that as the dose of LPS increases, the ROS response increases (
Given that TLR-4 signalling induces a significant increase in endogenous ROS levels, we next sought to establish the intracellular source of LPS-induced ROS. LPS-induced ROS activity is known to involve TLR-4 mediated interaction with Nox4 in embryonic kidney cells
(a) A significant attenuation of fluorescence was seen in samples treated with the DPI (2 µM). DCF fluorescence was measured using FACScan flow cytometer and CellQuest software. Solid black line (control). Red line (LPS treated). Green line (LPS+DPI treated). (b,c) Rotenone and diclofenac failed to inhibit LPS (1 µg/ml) induced ROS activity at 40 minutes. * P<0.05. These data are representative of 3 independent experiments.
As DPI resulted in complete abrogation of LPS-induced ROS activity, we wished to further investigate the role of Nox enzymes in LPS-induced ROS generation in SW480 cells. Western blotting was used to identify Nox enzyme expression. Nox1 as well as Nox2 were found to be expressed (
(a) Using Western Blotting we show that Nox1 expression in SW480 cells increases in response to LPS (1 ug/ml). (b) Western Blotting also shows that LPS increased expression of Nox2 at 40 minutes. (c) p22phox is shown to be expressed and expression increases earlier than Nox1 and Nox2. (d) p47phox is shown to be expressed but expression is stable over the time points. (e) Quantitative PCR analysis of Nox1 and Nox2 mRNA in SW480 cells treated with LPS (1 µg/ml) over one hour. Data is represented as fold-change relative to control untreated cells hours as determined by the 2−ΔΔCt method. Results are expressed as mean±SD and are representative of three independent experiments. * P<0.05. These data are representative of 3 independent experiments.
Nox enzyme activation is reliant upon assembly of individual sub-units. We therefore also looked at the effect of LPS treatment on protein expression of p22phox and p47phox (
We further investigated the increase of Nox1 and Nox2 protein expression in response to LPS treatment using RT-PCR to quantify the Nox1 and Nox2 mRNA levels in response to LPS (
As there is an established link evident between LPS and NF-κB activation via the MyD88 pathway, we investigated if NF-κB played a role in LPS induced ROS activity. I-Kappa-B kinase (IKK) is an enzyme which is necessary to activate NF-κB. An IKK inhibitor was used to inhibit NF-κB activity (
(a) Quantification of DCF fluorescence using CellQuest software demonstrates a significant reduction of LPS-induced 40 minute ROS activity in the presence of IKK inhibitor (40 µg/ml) following LPS treatment (1 µg/ml). (b) p-IκB increased in expression at 20 minutes following LPS treatment. * P<0.05. These data are representative of 3 independent experiments. (c) LPS (1 µg/ml) induced Nox1 expression increased at 40 minutes following LPS treatment, however pre-treatment with IKK inhibitor for 1 hour reduced the level of expression to untreated levels. (d) LPS induced Nox2 expression at 40 minutes is also reduced by the IKK inhibitor. * P<0.05. These data are representative of 3 independent experiments.
In order to prove that NF-κB activation was required for an increase in Nox1 and Nox2 expression following LPS treatment, an NF-κB inhibitor was again used. LPS-induced Nox1 protein expression at 40 minutes was reduced by the IKK inhibitor (
Recent anti-oxidant therapies are more effective due to greater bioavailability and the ability to target specific ROS generating organelles such as the mitochondria. As there was no decrease in ROS levels with the use of rotenone, we questioned the subcellular compartment that was responsible for ROS generation in response to LPS. We first co-stained SW480 cells with DCF and an ER tracker dye and then examined fluorescence using multiphoton microscopy. DCF displays a distinct peri-nuclear staining pattern in SW480 cells in response to LPS treatment (
Cells were cultured for 48 hours in glass bottomed dishes. (a) Cells were treated with DCF and an ER tracker dye for 1 hour at 37°C with LPS (1 µg/ml) added 30 minutes before imaging with a multiphoton laser scanning microscope. (b) Cells were stained with mitotracker deep red (1 µM) and (10 µM) MitoPY, and treated with Menadione (8 µM) or LPS (1 µg/ml) for 1 hour at 37°C. The scale bar represents 20 µm.
We next wanted to examine the mitochondria as an intracellular source of LPS induced ROS. A previous study by Dickenson et al demonstrated that peroxy-yellow (MitoPY), a fluorescent probe, can effectively be used to image H2O2 within the mitochondria of live cells
Regulation of cellular signalling pathways is one of an array of intra-cellular effects derived from TLR-4 signalling. Individual survival pathways including the PI3K/Akt pathway, which is known to facilitate tumour metastasis, can be activated by TLR-4 signalling. The inhibitory proteins of this pathway such as PTEN, are known to be redox sensitive. Having established that SW480 cells generate endogenous ROS from the ER in an NF-kB dependent manner in response to TLR-4 signalling, we thus investigated if LPS induced endogenous ROS could play a role in regulation of these redox sensitive pathways. We see an increase in pAkt at 40 minutes following LPS treatment which corresponds with the increase in LPS-induced ROS activity also seen at 40 minutes (
(a) LPS (1 µg/ml) treatment causes a transient increase in Akt phosphorylation, maximal at 30–40 minutes. (b) pAkt expression is increased in response to LPS treatment at 40 minutes and completely inhibited by DPI(2 µM). *P<0.05. These data are representative of 3 independent experiments.
Recent studies have reported that LPS can increase cancer cells adherence, an essential step for successful metastasis
(a,b) LPS treatment resulted in a significant increase in SW480 cell adhesion to collagen I after 1 hour. This was inhibited using a recognised Nox inhibitor, DPI (2 µM), and a PI3K inhibitor, LY294002(5 µM). (c) Nox1 siRNA abrogates LPS induced ROS and Nox1 knockdown is confirmed by Western Blotting. (d) The increase in SW480 colon cancer cell adhesion was completely inhibited by Nox1 siRNA. * P<0.05. These data are representative of 3 independent experiments.
Nox1 siRNA treated cells were then used to investigate the role of Nox1 in colon cancer cell adherence. Interestingly, a significant reduction in tumour cell adherence was seen in the Nox1 siRNA cells following LPS treatment, indicating that Nox1 derived ROS are essential for potentiation of colon cancer cell adherence in response to LPS (
The link between systemic inflammation and promotion of tumour metastasis has been well established
Surgical inflammation potentiates tumour cell metastasis through PI3K/Akt signalling. Previous studies have shown a significant rise in pulmonary metastasis following surgery and this effect is inhibited by a PI3K inhibitor
This study shows that the potentiating effects of inflammation on tumour cell metastasis are derived from endogenous ROS, predominantly generated from Nox1 in colon cancer cells. This effect is inhibited by Nox1 siRNA which may well represent a useful target to prevent effects of inflammation on promotion of tumour cell metastasis. Nox1 expression in colon cancer has been previously described and activation of Nox1 in colon cancer has been shown to be an important mediator of invadopodia formation which can facilitate cancer cell invasion
Interestingly, the level of Nox1 expression was seen to increase corresponding with ROS production following LPS treatment. Although the levels of Nox enzyme expression was weak in untreated cells, the capacity of the Nox family to generate endogenous ROS in response to an inflammatory stimulus should not be underestimated. TNF-α, a vital cytokine produced by the inflammatory response has previously been reported to increase transcription of Nox1 and increase superoxide production in colon cancer cells after 24 hours
A functional homology has been suggested to exist between Nox1 and Nox2 (gp91phox) in phagocytes
TLR-4 signalling activates NF-kB through the MyD88 pathway and leads to transcription of pro-inflammatory cytokines and many important components of the inflammatory response. Previously, NF-kB has provided a mechanistic link between inflammation and cancer
This study provides further insight into the subcellular location of TLR-4 induced ROS. An association between TLR-4/MD2 complex signalling and the ER has previously been made. Our study provides evidence of the role of the ER in TLR-4 signalling as it was observed that the ER was the subcellular location responsible for LPS-induced ROS generation which regulate redox sensitive signalling pathways. An important therapeutic implication of this finding relates to the potential use of targeted anti-oxidant therapy to counteract the redox signalling effects conferred by TLR-4 activation. Mitochondria have traditionally been attributed as the main intracellular source of ROS and are being targeted with the latest generation of anti-oxidants
In summary, unravelling of the LPS-Nox1 signalling axis reveals potential redox targets that could be used to prevent successful tumour metastasis in response to inflammation.