Terminal differentiation is an important event for maintaining normal homeostasis in the colorectal epithelium, and the loss of apoptosis is an important mechanism underlying colorectal tumorigenesis. The very limited current data on the role of lipoxygenase (LOX) metabolism in tumorigenesis suggests that the oxidative metabolism of linoleic and arachidonic acid possibly shifts from producing antitumorigenic 15-LOX-1 and 15-LOX-2 products to producing protumorigenic 5-LOX and 12-LOX products. We examined whether this shift occurs in vitro in the human colon cancer cell line Caco-2 in association with the loss of terminal differentiation and apoptosis, or in vivo during the formation of colorectal adenomas in patients with familial adenomatous polyposis (FAP). Restoring terminal differentiation and apoptosis of Caco-2 cells increased the mRNA levels of 5-LOX, 15-LOX-2, and 15-LOX-1, but the only significant increases in protein expression and enzymatic activity were of 15-LOX-1. In FAP patients, 15-LOX-1 expression and activity were significantly down-regulated in adenomas (compared with paired nonneoplastic epithelial mucosa), whereas 5-LOX and 15-LOX-2 protein expressions and enzymatic activities were not. We conducted a validation study with immunohistochemical testing in a second group of FAP patients; 15-LOX-1 expression was down-regulated in colorectal adenomas (compared with nonneoplastic epithelial mucosa) in 87% (13 of 15) of this group. We confirmed the mechanistic relevance of these findings by demonstrating that ectopically restoring 15-LOX-1 expression reestablished apoptosis in Caco-2 cells. Therefore, 15-LOX-1 down-regulation rather than a shift in the balance of LOXs is likely the dominant alteration in LOX metabolism which contributes to colorectal tumorigenesis by repressing apoptosis. (Cancer Res 2005; 65(24): 11486-92)
The oxidative metabolism of linoleic and arachidonic acid influences human tumorigenesis, especially of the colon (1). Research in this area has thus far focused mainly on cyclooxygenase-2 (COX-2). Lipoxygenases (LOX), however, also influence the oxidative metabolism of linoleic and arachidonic acid in human cells (1). Apoptosis is commonly suppressed in human cancers (2), and reestablishing apoptosis is an important mechanism for treating human tumorigenesis (3). The very limited prior data on the contribution of LOXs to apoptosis and tumorigenesis suggests diverse and even opposing roles for the various LOX pathways in these processes (1). Therefore, we hypothesized that the balance struck by linoleic and arachidonic acid metabolisms in the LOX pathway activity shifts from the antitumorigenic 15-LOX-1 and 15-LOX-2 pathways to the protumorigenic 5-LOX and 12-LOX pathways during tumorigenesis (1). Because the prior relevant data are limited mainly to in vitro systems and never come from concurrent, more comprehensive in vitro and in vivo experimental testing, we tested our hypothesis concurrently in in vitro and clinical in vivo models of colorectal tumorigenesis.
Colorectal cancer cells escape terminal cell differentiation and apoptosis. The Caco-2 colon cancer cell line, however, can be treated with sodium butyrate to restore apoptosis and produce terminally differentiated cells that are very similar to normally differentiated colonic epithelial cells at the metabolic and ultrastructural levels (4). Therefore, we employed this widely used in vitro system to compare profiles of LOX expressions and activities before and after sodium butyrate–induced terminal cell differentiation and apoptosis (5). We evaluated the clinical implications of our in vitro results by studying LOX expressions and activities in vivo in paired colorectal adenomas and nonneoplastic epithelial mucosa from patients with familial adenomatous polyposis (FAP) syndrome. We also conducted in vitro gene expression studies to examine the mechanisms underlying LOX pathway effects on colorectal tumorigenesis.
Materials and Methods
Acquisition of clinical samples. We obtained tissue samples from two groups of FAP patients. Samples from the first group (five patients) were obtained through the University of Texas M.D. Anderson Cancer Center Tissue Procurement and Banking Facility from colectomy specimens. For each patient, samples were procured from five separate colorectal adenomas and from nonneoplastic colonic mucosa (total of 30 specimens). Samples from the second group (15 patients) were routine formalin-fixed, paraffin-embedded biopsy specimens of paired colorectal nonneoplastic mucosa and adenomas collected prior to celecoxib treatment in these patients, as described previously (6). The institutional review board approved this study.
Materials. We obtained the rabbit polyclonal antiserum to recombinant human 15-LOX-1 and 15-LOX-1 expression vector as described previously (7), 5-LOX monoclonal antibody from BD Bioscience (San Jose, CA), 15-LOX-2 polyclonal antibody from Cayman Chemical Co. (Ann Arbor, MI), and polyclonal antibody against poly-(ADP-ribose)-polymerase from Roche Applied Bioscience (Indianapolis, IN). The Caco-2 human colon cancer cell line was obtained from the American Type Culture Collection (Manassas, VA). 13-S-Hydroxyoctadecadienoic acid (13-S-HODE), leukotriene B4 (LTB4), 12-S-hydroxyeicosatetraenoic acid (12-S-HETE), and 15-S-HETE enzyme immunoassay (EIA) kits were purchased from Assay Design Inc. (Ann Arbor, MI). 5-, 12-, and 15-HETEs and 13-S-HODE and their relevant deuterated standards were purchased from Cayman Chemical. Butylated hydroxytoluene (BHT) was obtained from Sigma Chemical Co. (St. Louis, MO). Human 15-LOX-1 cDNA probe was a 2,670 bp full-length cDNA obtained as previously described (7). We obtained the human 5-LOX cDNA probe from Oxford Biomedical Research, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probe from Ambion, Inc. (Austin, TX), and human 12-LOX cDNA probe from Cayman Chemical. Other reagents, molecular-grade solvents, and chemicals were obtained from commercial manufacturers or as specified.
Cell cultures. Caco-2 cells were grown in EMEM that contained 15% fetal bovine serum and were supplemented with 1% penicillin/streptomycin.
Cell differentiation assays. Caco-2 cells were cultured for the indicated times, harvested, and sonicated for 20 seconds at an output setting of 15 using a Sonifier (Sonic Dismembrator 60 Model, Fisher Scientific Co., Fair Lawn, NJ). Total cell lysates were assayed for alkaline phosphatase activity using an alkaline phosphatase kit following the manufacturer's instructions (Pointe Scientific Inc., Canton, MI).
Quantitative real-time reverse-transcription PCR. Total RNA was extracted from cells using TRI reagent (Molecular Research Center, Inc., Cincinnati, OH). The integrity of total RNA was verified on an Agilent 2100 Bioanalyzer using the RNA 6000 Nano LabChip kit (Agilent Technologies, Palo Alto, CA). Extracted RNA was quantified using an RNA quantitation kit (RiboGreen; Molecular Probes, Inc., Eugene, OR). A 500 ng RNA from each sample was reverse transcribed in a 20 μL reaction using the iScript cDNA synthesis kit (Bio-Rad Laboratories, Hercules, CA). The real-time PCR was carried out in 25 μL of a reaction mixture containing 1 μL of cDNA (25 ng/μL), 12.5 μL of 2× TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA), 10.25 μL diethyl pyrocarbonate water and 1.25 μL of a primer and probe mixture (Applied Biosystems). Real-time PCR assays were done in triplicate using a 7300 real-time PCR system (Applied Biosystems) with the following program: 50°C for 2 minutes, 95°C for 10 minutes, 40 cycles at 95°C for 15 seconds, and at 60°C for 1 minute. A sequence detection program calculated the threshold cycle number (CT) at which the probe cleavage–generated fluorescence exceeded the background signal (8).
Measurement of relative RNA expression levels. We calculated the relative RNA expression levels using a comparative CT method (8). The sets of gene primer and probe for the target genes (15-LOX-1, 5-LOX, 12-LOX, and 15-LOX-2) were confirmed to have amplification efficiency equal to that of the internal reference gene (HPRT1). The relative expression level of an individual target gene was normalized to that reference gene and to a calibrator sample that was run on the same plate. The normalized relative expression level of a target gene in an individual sample was calculated using the following formula:
in which the real-time PCR efficiency of the target gene transcript is denoted by Etarget and that of the reference gene transcript, by Ereference (8). Therefore, the relative RNA expression level of a gene is a unitless number relative to that of the calibrator sample (9). Calculations of relative RNA expression were done using commercial software (SDS V1.2; Applied Biosystems).
Northern blot analysis of lipoxygenase mRNA expressions. Total RNA was isolated by TRI reagent (Sigma Chemical), separated by gel electrophoresis, and transferred to membranes that were probed with radiolabeled (32P) cDNA probes as described previously (10). After hybridization with either 5-LOX, 12-LOX, 15-LOX-1, or GAPDH (32P) cDNA probes and washes, the blots were autoradiographed by exposure to hyperfilm-MP films (Amersham Biosciences, Corp., Piscataway, NJ).
Western blot analysis. For Western blotting, protein samples were prepared and subjected to SDS-PAGE under reducing conditions, as described previously (11). After transfer, the blots were probed with a solution of rabbit polyclonal antibody to human 15-LOX-1 (1:2,000), 15-LOX-2 (1:250), 5-LOX (1:250), and PPARP (1:2,000) and then analyzed using the enhanced chemiluminescence method.
Enzyme immunoassay measurements. Organic phase extractions of cell lysates and cell culture media were prepared in a fashion similar to that described previously (12). 13-S-HODE, LTB4, 12-S-HETE, and 15-S-HETE were measured using commercially available EIAs (Assay Designs).
Extraction of hydroxyeicosatetraenoic acids and 13-hydroxyoctadecadienoic acid from Caco-2 cells for measurement by liquid chromatography/tandem mass spectrometry. The culture medium was collected at the end of the experiment, and the cells were harvested by trypsinization. Intracellular HETEs and 13-HODE were extracted by using the modified method of Kempen et al. (13). The upper organic phases were pooled and evaporated to dryness under a stream of nitrogen at room temperature. All extraction procedures were done under conditions of minimal light. Samples were then reconstituted in a 200 μL methanol/10 mmol/L ammonium acetate buffer (70:30, v/v; pH 8.5) before analysis by liquid chromatography/tandem mass spectrometry (LC/MS/MS). HETEs and 13-S-HODE in the cell culture medium were extracted using a solid phase method. An aliquot of 10 μL of 10% BHT was added to 1 mL of the cell culture medium. The solution was applied to a Sep-Pak C18 cartridge (Waters Corp., Milford, MA) that had been preconditioned with methanol and water. HETEs were eluted with 1 mL of methanol. The eluate was evaporated under a stream of nitrogen and the residue dissolved in a 100 μL methanol/10 mmol/L ammonium acetate buffer (70:30, v/v; pH 8.5).
Extraction of hydroxyeicosatetraenoic acids and 13-hydroxyoctadecadienoic acid from tissue samples. Frozen tissue was pulverized to a fine powder using a liquid-nitrogen cooled mortar (Fisher Scientific). The samples were then transferred to microcentrifuge tubes, and 500 μL of ice-cold PBS buffer (containing 0.1% BHT and 1 mmol/L EDTA) were added. After mixing and sonication, an aliquot (50 μL) of homogenate was removed for protein measurements. The protein concentration was determined by the method of Bradford (Bio-Rad). Double-distilled water was added to a total volume of 1,000 μL, and the resulting solution was acidified to a pH of 3.2 to 3.5 with 0.5 mol/L HCl solution. The organic phase of the solution was extracted thrice using water-saturated ethyl acetate. Pooled organic phase solutions were dried under a nitrogen stream. Residues were reconstituted in 100 μL of ethanol and 500 μL of dilution buffer (provided with the EIA kits) for the EIAs and in 100 μL methanol/10 mmol/L ammonium acetate buffer, pH 8.5 (70:30, v/v) for LC/MS/MS analyses. LC/MS/MS was done under the same conditions described below.
Liquid chromatography/tandem mass spectrometry measurements of hydroxyeicosatetraenoic acids and 13-hydroxyoctadecadienoic acid. LC/MS/MS was done using a Quattro Ultima tandem mass spectrometer (Micromass, Beverly, MA) equipped with an Agilent HP 1100 binary pump high-performance liquid chromatography inlet. HETEs and 13-HODE were separated using Luna 3μ phenyl-hexyl 2 × 150 mm (Phenomenex, Torrance, CA). The mobile phase consisted of 10 mmol/L ammonium acetate (pH 8.5; phase A) and methanol (phase B). The flow rate was 250 μL/min, with a column temperature maintained at 50°C. The sample injection volume was 25 μL. Samples were kept at 4°C during the analysis. 13-HODE were detected using electrospray-negative ionization and MRM to monitor the transitions at m/z 295.3 → 277.3. Fragmentation for the HETEs and HODE was done using argon as the collision gas at a collision cell pressure of 2.10 × 10−3 torr. For cell culture experiments, the results were expressed as nanograms of measured product per 106 cells. The cells were counted with an electronic particle counter (Coulter, Hialeah, FL). The final concentration of eicosanoids in the tissues was expressed as nanograms per microgram of protein.
Assessments of apoptosis. Inverted light (phase contrast) microscopy was used to assess gross evidence of growth inhibition and apoptosis. Floating and attached cells were harvested as specified for each experiment, and the DNA fragmentation assay was done as described previously (7). The PARP-cleavage assay was done as described above (Western blot analysis).
15-Lipoxygenase-1 immunohistochemical staining. Paraffin-embedded tissue blocks were cut into 5-mm-thick sections and deparaffinized in xylene. Sections were stained for 15-LOX-1, as previously described (12) with the following modifications: (a) a 15 LOX-1 primary antibody was used at a 1:400 concentration and with overnight incubation; (b) the staining intensity was rated on a scale of 0 to 3 by an experienced pathologist.
Statistical analyses. The data were log10 transformed in order to accommodate the normality and homoscedasticity assumptions implicit to the statistical procedures used. We used one-way ANOVA to compare various quantifiable outcome measures (e.g., 13-S-HODE and LTB4 levels) in different cell line experimental conditions (e.g., sodium butyrate−treated and controls). Paired t tests were used to test the statistical significance of the differences between the levels of linoleic and arachidonic acid metabolite (13-S-HODE, LTB4, 5-HETE, 12-S-HETE, 15-S-HETE) in the paired colorectal adenomas and nonneoplastic epithelial mucosa. Data were analyzed using SAS software (SAS Institute, Cary, NC). 15-LOX-1 immunohistochemistry staining differences between colorectal adenomas and nonneoplastic epithelial mucosa were tested using a nonparametric test (Sign test). Reported P values are two-sided and were determined to be significant at the 0.05 level, with Tukey adjustments used for all multiple comparisons in the ANOVA analyses.
Lipoxygenase expression and enzymatic activity during colon cell differentiation. Treatment with sodium butyrate significantly increased alkaline phosphatase in Caco-2 cells (data not shown), which confirmed the induction of cell differentiation. This differentiation led to a 5.6-fold up-regulation of 15-LOX-1 mRNA expression levels at 4 hours (compared with levels in untreated control cells; P < 0.0001). Levels peaked at 24 hours with a 41.2-fold increase and continued to exceed those of the untreated cells at 48 and 72 hours (Fig. 1A; P < 0.0001). Induced cell differentiation also up-regulated 15-LOX-2 (Fig. 1B) and 5-LOX (Fig. 1C) mRNA levels, starting at 24 hours and increasing through 72 hours (P < 0.0001 for both LOXs at both time points). 12-LOX mRNA expression was not detected in undifferentiated or differentiated Caco-2 cells. Northern blot analyses showed that induced cell differentiation (a) increased 15-LOX-1 expression starting at 6 hours, peaking at 24 hours, and continuing for 48 and 72 hours; (b) increased 5-LOX expression starting at 48 hours; and (c) had no effect on 12-LOX expression, which was absent throughout (data not shown). 15-LOX-1 or 15-LOX-2 protein expression was not detected in undifferentiated Caco-2 cells (Fig. 1D). 15-LOX-1 expression was induced with terminal cell differentiation starting at 24 hours, whereas 15-LOX-2 protein expression remained undetectable (Fig. 1D). 5-LOX protein expression was detected in undifferentiated Caco-2 cells and seemed to remain unchanged during the induction of cell differentiation (Fig. 1D).
Lipoxygenase and cyclooxygenase products during induced differentiation in colon cancer cells. 13-S-HODE levels were markedly higher (ng/μg protein range) than were levels of 15-S-HETE and LTB4 (pg/μg protein range; Fig. 1E-G). 12-S-HETE was undetectable despite methods using sensitive EIA and LC/MS/MS. 5-HETE levels were also undetectable. 15-S-HETE levels remained relatively unchanged between pre- and post-terminal Caco-2 cell differentiation (P = 0.17, 0.63, and 0.196 at 24, 48, and 72 hours, respectively; Fig. 1E). The LTB4 level increased moderately at 72 hours (P = 0.034) but was not statistically significantly at 24 and 48 hours (P = 0.083 and 0.12 at 24 and 48 hours, respectively; Fig. 1F). 13-S-HODE levels increased steadily and significantly during terminal differentiation, as indicated by EIA (P = 0.0019, <0.0001, and <0.0001 at 24, 48 and 72 hours, respectively; Fig. 1G) and LC/MS/MS measurements (data not shown).
Lipoxygenase expression and activity in colorectal nonneoplastic mucosa and adenomas of patients with familial adenomatous polyposis. 15-LOX-1 protein expression was down-regulated in adenomas (compared with paired nonneoplastic epithelial mucosa) from the colons of FAP patients (Fig. 2A,, top row). 5-LOX expression was similar in adenomas and nonneoplastic mucosa (Fig. 2A,, middle row), whereas 15-LOX-2 expression was absent in adenomas and nonneoplastic mucosa (Fig. 2A,, bottom row). EIA and LC/MS/MS indicated that 13-S-HODE was the predominant LOX product (compared with the LOX products LTB4, 5-HETE, 12-S-HETE, and 15-S-HETE) in nonneoplastic mucosa (Fig. 2B; data not shown for EIA assays). The levels of 5-HETE, 12-HETE, or 15-HETE were not significantly different between adenomas and nonneoplastic mucosa (P = 0.74 for 5-HETE, 0.63 for 12-HETE, and 0.09 for 15-HETE; Fig. 2B). 13-HODE levels were significantly higher in nonneoplastic colorectal mucosa than in colorectal adenomas: 63.52 ± 34.46 ng/μg protein (mean ± SD; nonneoplastic mucosa) versus 5.94 ± 3.28 ng/μg protein (adenomas) by LC/MS/MS (P = 0.0021) and 26.24 ± 4.3 ng/μg protein (nonneoplastic mucosa) versus 10. 25 ± 1.16 ng/μg protein (adenomas) by EIA (P < 0.0001). A validation test involving immunohistochemistry staining of tissue samples from a group of 15 other FAP patients showed that 15-LOX-1 expression was down-regulated in colorectal adenomas (versus paired nonneoplastic epithelial mucosa) in 87% (13 of 15) of the patients (P = 0.01; Fig. 2C).
Effects of 15-lipoxygenase-1 expression on apoptosis of colon cancer cells. Transfecting a 15-LOX-1 expression vector induced 15-LOX-1 expression in Caco-2 cells (Fig. 3A). The expressed 15-LOX-1 protein was enzymatically active, as shown by its associated increase in 13-S-HODE levels over 13-S-HODE levels in cells transfected with control (empty) vector or in mock-transfected cells (transfected with transfection medium alone; P = 0.0123; Fig. 3B). 15-LOX-1 expression in Caco-2 cells (a) significantly reduced the number of viable cells (compared with mock transfection; P = 0.0134; Fig. 3C), whereas control vector transfection did not; (b) increased the ratio of floater cells to total cells (compared with mock or control vector transfection; P = 0.0148; Fig. 3D); (c) induced DNA fragmentation in a typical DNA pattern of apoptosis, which was not the case with control vector or mock transfection (Fig. 3E); and (d) induced PARP cleavage, which was not the case with control vector or mock transfection (Fig. 3F).
Results of our present study provide further information to improve the understanding of the roles of 15-LOX-1, 15-LOX-2, and 5-LOX in colorectal carcinogenesis. RNA expression levels of 5-LOX, 15-LOX-1, and 15-LOX-2 increased in Caco-2 cells during sodium butyrate–induced terminal cell differentiation (compared with undifferentiated Caco-2 cells), but only the protein expression or enzymatic activity of 15-LOX-1 increased significantly. Only 15-LOX-1 expression and enzymatic activity were down-regulated in vivo in adenomas (compared with nonneoplastic colorectal mucosa) of FAP patients. Ectopic expression of 15-LOX-1 restored apoptosis in Caco-2 cells. These findings, we believe, along with others outlined below, support the conclusion that 15-LOX-1 is an important factor in the regulation of colorectal epithelial cell terminal differentiation and apoptosis.
Although restoring terminal differentiation in Caco-2 cells increased the RNA expression of 15-LOX-1, 15-LOX-2, and 5-LOX, only the protein expression and product (13-S-HODE) of 15-LOX-1 also significantly increased (compared with levels in nondifferentiation Caco-2 cells). Therefore, decreased 15-LOX-1 and 13-S-HODE were the predominant alterations involving LOXs in the oxidative metabolism of linoleic and arachidonic acid in undifferentiated cells (compared with metabolism in terminally differentiated cells), and induction of terminal differentiation reversed these alterations.
Other groups also have reported that induced terminal differentiation of Caco-2 cells restores 15-LOX-1 expression (14, 15) but proposed that 15-LOX-1 expression inhibited cell differentiation (14, 15). This theory was based on studies using nonspecific LOX inhibitors such as nordihydroguaiaretic acid in concentrations that can affect a variety of molecular targets (14, 15). Our current report provides the first demonstration that restoring 15-LOX-1 via ectopic expression in undifferentiated Caco-2 cells (not treated with sodium butyrate) inhibited their proliferation and, more important, induced apoptosis, which is the expected outcome of the terminal differentiation process. These findings indicate that 15-LOX-1 expression during sodium butyrate–induced cell differentiation in Caco-2 cells promotes apoptosis.
The oxidative metabolism of linoleic and arachidonic acid involves several LOX pathways that can have diverse biological functions, especially in relation to tumorigenesis (1). Technical limitations such as semiquantitative Northern blot assays and incubation assays that may not detect endogenous products have narrowed the scope of the relevant prior studies, which have focused on only one or a few of these metabolic pathways mainly in in vitro models without concomitant comparative in vivo studies to assess the clinical relevance of the in vitro findings. For example, one study of LOXs during Caco-2 terminal differentiation did not detect 5-LOX or 12-LOX RNA expression with Northern blotting (15), and another such study showed that 5-LOX RNA and protein expression increased, but 5-LOX enzymatic activity did not (16). Our current study provides the first comprehensive information on both the expression of various LOXs (from sensitive quantitative real-time PCR) and their endogenous products (from EIA and LC/MS/MS measurements). The sensitivity of our assays and ability to assess LOXs in vitro and in vivo led to our novel finding that 15-LOX-1 up-regulation, which is lost in undifferentiated cells, is the dominant LOX alteration involved in colorectal epithelial cell terminal differentiation. We detected no significant alteration in the protein expression or enzymatic activity of 5-LOX, 12-LOX, or 15-LOX-2. These data support our conclusion that 15-LOX-1 is the major LOX contributor to terminal cell differentiation and apoptosis.
The widely used Caco-2 terminal differentiation model for comparing undifferentiated with differentiated colonic epithelial cells has a limited ability to predict in vivo effects. Therefore, we assessed the clinical relevance of our in vitro findings in the human in vivo model system of paired samples of colorectal adenomas and the nonneoplastic epithelial mucosa of FAP patients, who have a very high risk of developing colorectal cancer. Consistent with our in vitro findings, 15-LOX-1 expression was down-regulated in the FAP adenomas (compared with nonneoplastic colorectal epithelial mucosa), whereas the expression of 5-LOX and 15-LOX-2 remained unchanged. The predominant metabolite in nonneoplastic colorectal epithelial mucosa was 13-S-HODE, the primary product of 15-LOX-1, and the only significant change in LOX products was a decrease in 13-S-HODE levels in adenomas (compared with nonneoplastic epithelial mucosa). Our major finding of down-regulated 15-LOX-1 in colorectal adenomas was confirmed in a second group of FAP patients. Therefore, the 15-LOX-1 down-regulation in colon cancers that we reported previously (12) seems to occur at earlier stages and seems to be a predominant event in LOX metabolic changes of colonic tumorigenesis in FAP patients. We observed no significant in vitro or in vivo alterations in other LOXs that would indicate that the down-regulation of 15-LOX-1 expression is due to a shift in the balance of LOX metabolism. As we have reported previously, the suppression of 15-LOX-1 expression during colorectal tumorigenesis also is not likely related to a substrate shift into COX-2 pathways (17). These findings support the previously reported conclusion that the down-regulation of 15-LOX-1 in colorectal cancer cells is regulated transcriptionally (10, 18). Because our current results are limited to colorectal tumorigenesis models, other studies will be needed to determine whether our findings apply to other organ systems, such as the prostate, in which 12-LOX and 15-LOX-2 may play an important role in tumorigenesis (19, 20).
Based on studies using nonspecific LOX inhibitors (14, 15), some groups have proposed that 15-LOX-1 acts to antagonize Caco-2 terminal cell differentiation. Our in vitro and in vivo findings, however, correlated the down-regulation of 15-LOX-1 expression with the loss of terminal cell differentiation in Caco-2 cells and with the formation of colorectal adenomas in FAP patients, suggesting that 15-LOX-1 promotes terminal cell differentiation. The specific role of 15-LOX-1 in apoptosis induction has not previously been examined independently of sodium butyrate treatment. Ectopically expressed 15-LOX-1 inhibited cell proliferation and induced apoptosis in Caco-2 cells in our present study. These findings disagree with a report of Yoshinaga et al. (15) showing that ectopic 15-LOX-1 expression increased proliferation in HCT-116 colon cancer cells (15). Our results agree, however, with a report of Nixon et al. showing that 15-LOX-1 ectopic expression inhibited the growth of HCT-116 xenografts in mice (21). We also found that ectopic expression of 15-LOX-1 induces apoptosis in HCT-116 cells (data not shown). Other results by our group and others supporting the present finding that 15-LOX-1 expression inhibits proliferation and tumorigenesis and restores apoptosis include the following: (a) 13-S-HODE induced apoptosis and cell cycle arrest in transformed colonic cells (12); (b) 15-LOX-1 expression was crucial to apoptosis induced by nonsteroidal antiinflammatory drugs in colonic, esophageal, and gastric cancer cells (7, 11, 22, 23); (c) the transient expression of 15-LOX-1 inhibited the proliferation of human osteosarcoma cells (24); and (d) ectopic expression of 15-LOX-1, but not 5-LOX, promoted apoptosis in virally transformed 293 kidney cells (25).
In conclusion, we have shown for the first time that the loss of 15-LOX-1 expression likely is the dominant change in LOX metabolism during, and contributes to, colorectal tumorigenesis and that this change is unlikely to be secondary to a substrate shift to other LOX pathways. This finding highlights the need to develop strategies targeting the restoration of 15-LOX-1 expression, and thus apoptosis, which are lost during colorectal tumorigenesis, for the treatment and prevention of colorectal cancer.
Grant support: National Cancer Institute, NIH, Department of Health and Human Services K07 grant CA86970, Cancer Center Support Grant P30 CA16672 (to I. Shureiqi); the American Cancer Society Scholar Award RSG-04-020-01-CNE (to I. Shureiqi); the University of Texas M.D. Anderson Cancer Center Institutional Research Grant (to I. Shureiqi), and funding from the National Colorectal Cancer Research Alliance. Dr. Lippman holds the Ellen F. Knisely Distinguished Chair in Colon Cancer Research.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.