| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online November 2, 2004 as doi:10.1096/fj.04-2111fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Institute of Cancer Research, Medical University Vienna,
Research Institute for Pharmacology and Toxicology, Veterinary University of Vienna, Austria; and
Faculty of Biotechnology, Jagiellonian University, Krakow, Poland
1Correspondence: Institute of Cancer Research, Medical University Vienna, Borschkegasse 8a, Vienna 1090, Austria. E-mail: Brigitte.marian{at}meduniwien.ac.at
SPECIFIC AIMS
Colorectal tumors have become the second most frequent cause of cancer death in developed countries. Western-style diet has been shown to be the major risk factor for colorectal carcinogenesis; several dietary components have been implicated, including the amount and composition of fat. High-fat diets were found to enhance tumor development in animal models as well as in humans. Dietary fats may exert procarcinogenic effects on several levels, including provision of energy, enhanced secretion of bile acids, and supply of substrates for the synthesis of prostaglandins. In addition, food processing could chemically alter dietary lipids and thereby influence tumorigenic effects of the diet. Products of lipid processing as potential procarcinogenic constituents in the diet have found little attention.
Oxidants from endogenous sources are considered procarcinogenic agents. These include lipid peroxides formed from polyunsaturated fatty acids (PUFA) during (chronic) inflammation. Similar oxidation products may be formed exogenously from PUFA in dietary fats and oils. Oxidation of dietary lipids is greatly enhanced at higher temperatures during cooking. Consequently, fatty acid hydroperoxides are normal contaminants of our diet. As much as 25% of oxidized polar compounds consisting of hydroperoxides and their split products have been found in oils repeatedly used for frying. Using lipid membranes and tissue culture models, we have shown that such hydroperoxides interact with the lipid bilayer of cell membranes. They induce the formation of lipid radicals that enter the cells and cause toxic damage and disturb regulatory networks.
In the present study we have investigated the possibility that exogenous fatty acid hydroperoxides contribute to tumor growth and/or progression. Colorectal carcinomas develop in a long-term process involving aberrant crypts and microadenomas that slowly grow into larger adenomatous polyps from which most carcinomas arise. A necessary prerequisite of polyp growth is the formation of new blood vessels, which is mediated by up-regulation of cyclooxygenase 2 (COX-2). The resulting prostaglandins (PG) induce expression of vascular endothelial growth factor (VEGF), which in turn stimulates vascularization. Thus, induction of VEGF secretion is a key event in colorectal cancer development.
Here we show that the model compound linoleic acid hydroperoxide (LOOH) stimulates expression of COX-2 and VEGF synthesis in LT97 adenoma and SW480 carcinoma cells from the human colon. The results suggest that dietary lipid hydroperoxides may induce blood vessel formation in colon polyps and thereby accelerate one of the rate-limiting steps in colorectal carcinogenesis.
PRINCIPAL FINDINGS
1. Formation of intracellular H2O2 and cytotoxicity of LOOH
LOOH added to cultures of SW480 colorectal carcinoma cells caused formation of intracellular H2O2 that could be detected by reduction of 2’,7’-dichlorofluorescin diacetate to fluorescein. The mean fluorescence as determined by FACS analysis increased from 114.2 ± 18.4 AU in SW480 control cells to 460.2 ± 2.2 AU after exposure to 50 µM LOOH. Cell loss was induced in a dose-dependent manner: 20% loss at 50 µM LOOH and 70% loss at 100 µM after 24 h. Unoxidized linoleic acid (LH) was not cytotoxic. By comparison, LT97 adenoma cells had a 5-fold higher baseline production of intracellular H2O2; LOOH induced a slight increase that did not become significant. Cytotoxic effects of LOOH on LT97 adenoma cells were similar to those observed in SW480 cells: 10% cell loss at 30 µM and 75% cell loss at 50 µM LOOH after 24 h.
2. Stimulation of VEGF production
Exposure to LOOH for 24 h stimulated VEGF secretion into the culture supernatant of SW480 and LT97 cells as determined by ELISA (Fig. 1
a). The increase was 47% and 37%, respectively. VEGF gene expression was determined by real-time PCR and also found to be stimulated 2.4-fold (P=0.007) by LOOH in SW480 cells, whereas LH was without effect (Fig. 1b
). Standard RT-PCR using primers that span most of the coding region of the VEGF gene detected two main isoforms, the VEGF121 and VEG165 variants that are the most common in colorectal cancer cells. In the LOOH group both isoforms were increased compared with control and LH and no shifts in the relative amounts of the isoforms were observed (Fig. 1c
).
|
In LT97, VEGF mRNA was not increased after 24 h LOOH exposure (n.s. at P=0.177). Analysis by standard RT-PCR at earlier time points demonstrated a much faster induction: 3 h after addition of LOOH to the medium, VEGF mRNA levels were almost doubled (184% of controls, P=0.018). VEGF expression in LT97 cells incubated with LH was increased above control (162%, P=0.031 Fig. 1d
). Quantitative analysis of mRNA after 3 h exposure was repeated by real-time PCR, which confirmed these results: VEGF mRNA was increased in both the LOOH (2-fold, P=0.002) and the LH groups (1.3-fold, P=0.034, Fig. 1e
).
3. Regulation of VEGF expression by COX-2 and c-fos
To assess the possibility that VEGF production is stimulated via up-regulation of COX-2 expression in LT97 cells, as has been shown in vivo, COX-2 mRNA levels were determined by RT-PCR after 1, 3, and 6 h of LOOH exposure. They were increased after 3 h but there was no significant change in the LH group (Fig. 2
a). Results were confirmed using real-time PCR: COX-2 mRNA was increased 
3-fold in the LOOH group (P=0.015) and remained unchanged in the LH group (Fig. 2b
).
|
Besides stimulation of COX-2 expression, LOOH could expand the arachidonic acid pool by activating phospholipase A2; in this way, PG formation from COX-1 and COX-2 would both be increased. Specific inhibitors for COX-1 (SC58560, IC50 COX-1=7 nM) or COX-2 (SC58236, IC50 COX-2=5 nM) were used to block PG production and their effect on VEGF expression was determined. 1 and 5 µM SC58560 by itself induced VEGF expression in a concentration-dependent way. However, in LOOH-stimulated cultures the inhibitor efficiently prevented VEGF induction (Fig. 2c
). SC58236 did not affect basal VEGF expression, but prevented the induction by LOOH (Fig. 2d
).
An alternative pathway controlling VEGF expression is mediated by AP1. In SW480, c-fos expression was stimulated 1 h after LOOH addition to the medium, but not at 0.5 or 3 h. In LT97 cells c-fos expression was increased by LOOH throughout the experiment.
CONCLUSIONS AND SIGNIFICANCE
Dietary fat is known to enhance tumor development in the colon. The mechanistic components of this effect are selective growth stimulation and expression of tumor-related genes in a premalignant cell population. Growth of colorectal adenomas depends on the induction of VEGF and consequential vascularization. This occurs in early adenomas and is stimulated by tumor promoting prostaglandins, especially PGE2. In the min-mouse model of colon carcinogenesis, inhibition of prostaglandin synthesis as well as deletion of COX-2 or of the prostaglandin receptor EP2 prevents VEGF production, vascularization, and polyp growth. Our results show that the model compound LOOH can induce this key step of tumor development. This finding may be of general pathophysiological interest: colorectal cancer is frequently associated with chronic inflammation, which should provide considerable levels of endogenous lipid peroxides. In addition, dietary lipid hydroperoxides may be important exogenous risk factors.
LOOH (20–30 µM) increased the intracellular formation of hydrogen peroxide when added to the culture medium of LT97 adenoma or SW480 carcinoma cells with minimal cell loss. Intracellular reactive oxygen species activate redox-sensitive transcription factors switching on the expression of several genes, including VEGF. In SW480 carcinoma cells VEGF induction was observed only in cultures exposed to LOOH, not in LH exposed controls. In the LT97 adenoma cells the distinction was less clear: both LOOH and LH stimulated VEGF gene expression even though the effect of LH was always weaker. This could be due to the 5-fold higher level of H2O2 in the LT97 adenoma cells causing oxidation of LH to LOOH. As in the adenoma cell line, high levels of oxidative stress markers are found in colonic adenomas in vivo, so that the premalignant lesions might be capable of forming hydroperoxides from dietary PUFA in vivo. However, whereas VEGF mRNA was induced by LH in LT97 adenoma cells, production of VEGF protein was not stimulated, supporting the stronger procarcinogenic activity of LOOH compared with LH.
In LT97 cells COX-2 expression was increased after LOOH addition and COX-activity was found to be essential for induction of VEGF. Again, this reflects the in vivo situation where VEGF expression depends on COX-2 up-regulation and PGE2 production.
Alternatively, VEGF and COX-2 could be up-regulated independently by AP1-mediated gene expression as indicated by up-regulation of c-fos.
In summary, our results indicate that LOOH, considerable quantities of which may occur in typical Western-style diet, can induce expression of both COX-2 and VEGF and may be an important exogenous risk factor of colorectal carcinogenesis. The effect of LOOH can be blocked by inhibitors of PG synthesis.
|
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-2111fje;
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
