SEARCH

SEARCH BY CITATION

Keywords:

  • esophageal cancer;
  • chemoprevention;
  • vitamin E;
  • selenium;
  • N-nitrosomethylbenzylamine

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Previous human intervention trial demonstrated that vitamin E (Ve) and selenium (Se) supplementation decreased esophageal cancer deaths among younger participants, but may have no effect or produce an opposite effect among older ones. In our study, we intended to mimic this human nutritional trial to determine the chemopreventive effects of Ve/Se supplementation at the early or late stage of esophageal carcinogenesis in rats. Esophageal squamous cell carcinoma (ESCC) was induced in Fischer 344 rats with N-nitrosomethylbenzylamine (NMBzA, 0.35 mg/kg BW, s.c., three times per week for 5 weeks). The rats were maintained on a modified AIN-93M diet with low levels of Ve/Se or supplemented with high levels of Ve/Se at different stages. At Week 25, the number and volume of visible tumors, the numbers of dysplasia and ESCC were significantly lower in rats of supplementation during the early stage (Group C) or during the entire experimental period (Group E), but not during the late stage (Group D). Ve/Se supplementation at the early stage also significantly decreased cell proliferation, nuclear factor kappaB (NFκB) activation, protein and mRNA expression of cyclooxygenase 2 and 5-lipoxygenase and biosynthesis of prostaglandin E2 and leukotriene B4 during the carcinogenesis of rat esophagus. Our results demonstrated that the chemopreventive efficacy of Ve/Se supplementation on NMBzA-induced esophageal cancer is time selective and that supplementation during the early stage is clearly effective but probably ineffective during the late stage of carcinogenesis. NFκB signaling pathway activation and aberrant arachidonic acid metabolism might be the underlying mechanism.

Esophageal cancer is one of the most common malignancies with esophageal squamous cell carcinoma (ESCC) as the major histological type worldwide.1, 2 The incidence of esophageal cancer is a multistage process, which is composed of initiation, promotion and progression. Chemoprevention agents that target different stages of esophageal carcinogenesis are classified as blocking (anti-initiation) agents or suppressing (antipromotion/progression) agents.3 Based on oxidative stress as one critical mechanism of carcinogenesis, vitamin E (Ve) and selenium (Se) were recognized as possible blocking agents and suppressing agents for esophageal cancer according to their antioxidative functions of scavenging electrophiles and inhibiting oxidative DNA damage.4 The General Population Nutrition Intervention Trial conducted from 1985 to 1991 in Linxian, a county in northern China which has the highest incidence rate worldwide and has low nutrient intakes for Ve (78% recommended daily dietary allowance (RDA)) and Se (66% RDA),5 demonstrated that supplementation with Ve (30 mg of α-tocopherol per day) and Se (200 μg of selenomethionine per day) significantly decreased the risk of esophageal and gastric cardia cancer.6–8 A follow-up study indicated that the beneficial effects of the supplementation on mortality were still evident 10 years after the cessation of supplementation. However, the preventive effect of supplementation was observed only in subjects who entered in the trial at ages younger than 55 years, but not in older subjects.9 In addition, a parallel trial with antioxidants and other nutrients, started in 1985 in Linxian, failed to demonstrate a protective effect against esophageal and gastric cardia cancer among adults with esophageal dysplasia.10 These results suggest that supplementation with antioxidant nutrients was effective among younger subjects who were less likely to have severe dysplastic lesions, but ineffective in older subjects who were more likely to have more severe lesions. Consistent with this concept is the result of a subsequent intervention study with selenomethionine, which showed beneficial effects in patients with mild esophageal dysplasia, but not in those with severe esophageal dysplasia.11

On the other hand, results from other chemoprevention trials with Ve and Se remain inconsistent. In the Alpha-Tocopherol Beta-Carotene Cancer Prevention Study, Ve and β-carotene failed to prevent upper aerodigestive tract cancers.12 The Selenium and Vitamin E Cancer Prevention Trial (SELECT) showed no benefit of Ve, Se or their combination at the tested doses and formulations for prostate cancer risk reduction after nearly 7 years intervention13, 14 and recently updated the finding with increased prostate cancer risk with Ve after the extended 3 years follow-up.15 Results from some other clinical trials even raise concern that certain micronutrients could promote growth of preexisting tumors or precancerous lesions.16, 17 For example, supplementation with folic acid could possibly promote colorectal cancer.18 An animal study also suggested that folate supplementation at the early stage (prior to the existence of preneoplastic lesions) could inhibit colorectal cancer formation; however, supplementation at the late stage could promote carcinogenesis.18, 19 It was proposed that the efficacy of cancer chemoprevention by nutrients may be time selective during the multistages of carcinogenesis.20 However, it is unclear whether the “timing” of the intervention period can account for the inconsistent results from these chemoprevention trials with nutrients. Therefore, it is important to understand how the timing of Ve/Se supplementation affects esophageal carcinogenesis, as well as the underlying molecular mechanism.

N-Nitrosomethylbenzylamine (NMBzA)-induced esophageal carcinogenesis in the Fischer 344 (F344) rat has proven to be a valuable animal model for chemopreventive studies.21 Our previous work demonstrated that dietary Ve/Se deficiency promoted NMBzA-induced esophageal carcinogenesis in rats and that Ve/Se supplementation at normal levels in diet was preventive.22 In our study, we use this model to mimic the Ve/Se nutritional intervention in the Linxian trial. We aim to test our hypothesis that Ve/Se supplementation at the early stage of esophageal carcinogenesis is chemopreventive for cancer through suppressing nuclear factor kappaB (NFκB) signaling pathway activation and blocking aberrant arachidonic acid (AA) metabolism while late supplementation has no efficacy.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Animals, diets and treatment

A total of 249-week-old male F344 rats were purchased from Vitalriver (Beijing, China) and were randomly divided into six experimental groups (Groups A–F), 40 rats each group. Rats were housed in a controlled environment with a 12-hr light/dark cycle and given water ad libitum. Three different diets were prepared by Keaoxieli Diet (Beijing, China), low Ve/Se diet, normal diet and high Ve/Se diet. The normal diet was produced according to AIN-93M formula, which contained 80 IU/kg of α-tocopherol and 0.15 mg/kg of Se.23 The low Ve/Se diet (46 IU/kg of α-tocopherol and 0.05 of mg/kg Se) mimicked the Ve/Se-deficient nutritional status in the human populations such as that in Linxian. The high Ve/Se diet (162 IU/kg of α-tocopherol and 0.30 mg/kg of Se) mimicked the Ve/Se supplementation in General Population Nutrition Intervention Trial.6

After acclimatization for 2 weeks, rats in Groups A, B, C, D and E were treated with NMBzA (Ash Stevens, Detroit, MI), solubilized in 20% dimethyl sulfoxide (DMSO)/water, at 0.35 mg/kg body weight, s.c., three times per week for 5 weeks. Group F was given the vehicle as negative control. Our preliminary study Additional Supporting Information may be found in the online version of this article. On the time course of NMBzA-induced tumorigenesis indicated that hyperplasia was the most prevalent lesion observed in the esophagus (70% of the rats) 10 weeks after NMBzA treatment. At this time point, 20% of the rats had papillomas, but no carcinoma was observed. At Week 15, however, 10% of the rats developed carcinoma. Therefore, we set Week 10 as the dividing point of the early and late stages to conduct this supplementation study. As shown in Figure 1a, deficiency group (Group A) was maintained on low Ve/Se diet throughout the entire experiment (Weeks 0–25); normal control group (Group B) and negative control group (Group F) were maintained on normal diet during Week 0–25; early supplementation group (Group C) was put on high Ve/Se diet at the early stage (Week 0–10) and low Ve/Se diet at the late stage (Weeks 11–25); late supplementation group (Group D) was put on low Ve/Se diet at the early stage and high Ve/Se diet at the late stage and continuous supplementation group (Group E) was maintained on high Ve/Se diet during the whole stage (Weeks 0–25). Body weight and food intake were measured once a week during the experiment. The care of laboratory animal and animal experimental operation have been performed conforming to the Beijing administration rule of laboratory animal and approved by the Animal Care and Welfare Committee of Institute of Nutrition and Food Safety.

thumbnail image

Figure 1. Experimental design and nutritional and general observation of rats. (a) Experimental design, (b) plasma levels of α-tocopherol (error bars = SD; n = 10 per group), (c) plasma levels of Se (error bars = SD; n = 10 per group), (d) body weight and (e) diet consumption.

Download figure to PowerPoint

Nutritional analysis

At Weeks 0, 5, 15 and 25, 0.5 ml blood samples of ten rats in each group were collected from the retroorbital vein after anesthetizing with ether. Plasma was stored at −80°C until analysis. α-Tocopherol levels were determined by high Performance liquid chromatography (HPLC). In brief, lipids were extracted with hexane from plasma and separated in an Elipse XDB C18 column in an Agilent 1200 HPLC system (Agilent, Santa Clara, CA). The column was equilibrated in mobile phase of a mixture of methanol:water (98:2) and eluted at a flow rate of 1.0 ml/min. α-Tocopherol was detected with a UV detector (Waters 2487 Dual Absorbance Detector) at 292 nm and quantified by comparing with internal standards. Se in plasma was determined by atomic fluorescence spectrometry. Plasma was digested with mixed acids of HNO3:HClO4 (5:1) in a TMW-200 Dielectric-Assisted Microwave Digestion System (Titan, Beijing, China). Se was measured with an AFS-933 atomic fluorophotometer (Titan) at 196 nm.

Histopathological analysis

At the end of Week 25, rats were sacrificed 2 hr after injection of 5-bromo-2-deoxyuridine (BrdU, 50 mg/kg, i.p.), and the esophagi were excised and opened longitudinally. Tumors ≥1 mm in diameter were counted. The esophagus was cut longitudinally, with one-half fixed in 10% neutral buffered formalin for pathological analysis. The other half of the esophagus was stripped of muscle, quickly frozen in liquid nitrogen and then stored at −80°C for PCR and enzyme immunoassay (EIA). Serial 5-μm esophageal sections were cut for staining. Hematoxylin and eosin (H&E)-stained sections were used for histological diagnosis, normal, hyperplasia, dysplasia, papilloma and ESCC. Diagnostic criteria were the same as those of Pozharisski.24

Immunohistochemical staining

BrdU immunohistochemistry was performed using an in situ detection kit (BD Pharmingen, San Jose, CA) according to the manufacturer's instruction. In brief, antigens were unmasked in citrate buffer for 10 min at 89°C. Endogenous peroxidase was quenched by a 10-min incubation in 3% hydrogen peroxide in phosphate buffered solution (PBS). Sections were incubated with primary biotinylated antibody for 1 hr in a humidified chamber. Negative controls were processed in the absence of the primary antibody. Streptavidin-horseradish peroxidase (HRP) and diaminobenzidine (DAB) were successively applied to the sections. Six noncontiguous, randomly selected fields in each lesion type were photographed under 400× magnification. The proliferation index was calculated by dividing the number of positively stained cells by the total number of epithelial cells.

Immunohistochemistry (IHC) of cyclooxygenase 2 (COX2), 5-lipoxygenase (5LOX) and NFκB was similarly performed using rabbit primary antibodies (Abcam, Cambridge, MA). HRP-conjugated secondary antibody (Zhongshanjinqiao, Beijing, China) and DAB were applied for staining. Immunoreactive scores were calculated by multiplying the percentage of positive cells by the grade of staining intensity.25

Western blot of NFκB

Frozen esophagus tissue was homogenized and lysed in radio immunoprecipitation assay (RAPI) buffer, and proteins supernatant was collected by centrifugation at 12,000 rpm at 4°C for 20 min. Electrophoresis on 8% sodium dodecyl sulfate (SDS)–polyacrylamide gel was performed using Mini-protean 3 systems (Bio-Rad, Hercules, CA). Proteins were transferred to a nitrocellulose membrane and incubated with either a rabbit NFκB p65 antibody (Abcam, Hong Kong) at 1:500 overnight at 4°C and then with a HRP-labeled secondary antibody (Santa Cruz, Santa Cruz, CA) at 1:20,000 for 2 hr at room temperature. β-Actin was detected in the same samples as the loading control. Densitometric analyses of the immunoblots were performed using the Gel Imaging System and the Quantity One Software version 4.0 (Bio-Rad).

Real-time PCR of 5LOX and COX2

Total RNA was extracted from frozen esophagus tissue by the RNA isolation kit (Research Bio-Lab, Beijing, China) and chloroform. Reverse transcription of RNA to cDNA was carried out using cDNA synthesis kit (Research Bio-Lab) in a Peltier thermal cycler (LongGene, Hangzhou, China) at 65°C for 5 min. PCR primers were designed with a primer analysis software (Oligo, CO) as follows: 5LOX (GeneBank accession number J03960), 5′-CGGGCACCGACGACTACATT-3′ (sense) and 5′-GGGGCAGATCCTTGTGGCAT-3′ (antisense); COX2, 5′-ACACCTGAGCGGTTACCAC-3′ (sense) and 5′-AGCGGAT GCCAGTGATAGAG-3′ (antisense) and β-actin, 5′-GGGAAA TCGTGCGTGACATT-3′ (sense) and 5′-GCGGCAGTGGC CATCTC-3′ (antisense). Total gene specificity of the nucleotide sequences chosen for primers was confirmed by results of BLAST searches (GenBank database sequences) and then synthesized by Sangon Biotech (Shanghai, China). The real-time PCR was performed in a SYBR Green Supermix (Research Bio-Lab) with 10 μl of reaction volume using a MyiQ Real-Time PCR Detection system (Bio-Rad). The PCR reaction started at a temperature of 94°C for 30 sec and followed by 45 PCR cycles: denaturation at 94°C for 30 sec, annealing at 60°C for 30 sec, extension at 72°C for 30 sec and finally extended at 72°C for 5 min. The amount of target cDNA in each sample was established by the fractional PCR threshold cycle number (Ct). The gene expressions of 5LOX and COX2 were normalized versus the housekeeping gene β-actin.

EIA of prostaglandin E2 and leukotriene B4

To determine prostaglandin E2 (PGE2) and leukotriene B4 (LTB4) in esophageal tissues, frozen samples of each group were homogenized in 0.9% saline on ice with a tissue homogenizer. PGE2 and LTB4 in supernatant were determined with EIA kits (RapidBio Lab, West Hills, CA) according to the manufacturer's instructions. The absorbance was measured with a Multi-Mode Microplate Reader (Bio-Tek, Gene, VT), and PGE2 and LTB4 levels were quantified with the standard curves.

Statistical analysis

Incidence rates of visible tumors, papilloma and carcinoma of the treatment groups were compared using χ2 test and Fisher's exact test. Tumor multiplicity, number of microscopic lesions and blood sample markers were expressed as mean ± standard deviation and were evaluated using one-way analysis of variance. Comparisons between groups were made using Tukey-Kramer test. All the analyses were performed using the SAS statistical computer program (SAS Institute, Cary, NC).

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

General observations and nutritional status of NMBzA-treated rats

The general appearance and activity level of animals were not affected by treatments: continuous low Ve/Se diet (Group A), normal diet (Group B), early-stage supplementation (by switching from the low Ve/Se diet to high Ve/Se diet; Group C), late-stage supplementation (Group D) and continuous supplementation (Group E). From Weeks 20 to 25, food consumption in NMBzA-treated groups gradually reduced and remained lower than that of negative control group (Group F, p < 0.05; Fig. 1e). However, body weights did not show significant difference among the six groups (Fig. 1d).

As shown in Figures 1b and 1c, the carcinogen treatment had a profound effect in lowering the plasma α-tocopherol and Se levels in the rats on the low Ve/Se diet (Group A). When compared with Groups A and B (which consumed normal diet), Ve/Se supplementation at the early stage (Groups C and E) resulted in significantly higher plasma levels of α-tocopherol and Se at Week 10 (p < 0.05). Since Week 10, switching to low Ve/Se diet decreased α-tocopherol and Se levels in Group C. In contrast, α-tocopherol and Se levels in late supplementation group (Group D) were elevated and became higher than those of Group C at Week 25 (p < 0.05).

Ve/Se time-selectively suppressed esophageal carcinogenesis of NMBzA-treated rats

The incidence, average number and volume of visible tumors are summarized in Table 1. In comparison to Group A that had a tumor incidence of 100%, early, late and continuous high Ve/Se supplementation (Groups C–E) significantly reduced tumor incidence to 64.1, 86.5 and 65.8%, respectively (p < 0.05). In addition, tumor multiplicity and volume were also both significantly decreased by early supplementation (Group C) and continuous supplementation (Group E, p < 0.05) when comparing with Group A, but not by late supplementation (Group D, p > 0.05). When comparing with normal diet (Group B), early and continuous supplementation (Groups C and E) appeared to be slightly more effective in suppressing tumorigenesis (p > 0.05); however, late supplementation (Group D) significantly promoted tumor incidence and growth (p < 0.05). Taken together, the early-stage supplementation efficiently prevented visible tumor formation in rat esophagus, but the late-stage supplementation did not. In this experiment, 15 rats (six in Group A, three in Group B, one in Group C, three in Group D and two in Group E) died prematurely from Weeks 22 to 24, possibly due to the obstruction of the esophagus by big papillomas. All these 15 rats had esophageal tumors. However, the esophagus samples could not be used for histological analysis, and therefore, they were not included in Table 1. If these dead rats (all with esophageal tumors) were counted, the tumor incidence for Group A would be 100%, Group B would be 72.5%, Group C would be 65%, Group D would be 87.5% and Group E would be 67.5%; the results of the statistical analysis still remained the same.

Table 1. Time-selective prevention of NMBzA-induced esophageal carcinogenesis in rats by Ve/Se supplementation
inline image

As for microscopic changes, when comparing with Group A, the incidence and multiplicity of ESCC were both significantly reduced by early supplementation (Group C) and continuous supplementation (Group E, p < 0.05), but not by late supplementation (Group D, p > 0.05; Table 1). When comparing with Group B, early supplementation (Group C) also resulted in less ESCC, although the differences were not statistically significant (p > 0.05); however, late supplementation (Group D) significantly increased the numbers of ESCC and dysplasia as well (p < 0.05). Furthermore, both early supplementation (Group C) and continuous supplementation (Group E) produced significantly less ESCC and dysplasias than late supplementation (Group D, p < 0.05). The incidences of ESCC and papilloma in Group C also seemed markedly lower than those in Group D; however, no significant difference was found (p > 0.05).

Ve/Se time-selectively inhibited cell proliferation in the esophagi of NMBzA-treated rats

As shown in Figure 2a, proliferating cell population in epithelium of esophagus apparently increased as the tumorigenesis progressed over time. The proportions of BrdU-labeled cells in each histopathologic lesion especially papilloma were distinct among different groups. Quantitative analysis on proliferation index is shown in Figure 2b. When comparing with Group A (low Ve/Se), the esophagi of Group C (early supplementation) and Group E (continuous supplementation) showed significantly less proliferative cells in various lesions. However, late supplementation (Group D) did not inhibit cell proliferation in lesions of hyperplasia, dysplasia, papilloma and carcinoma when compared with Group A. In addition, proliferation indices of early supplementation group (Group C) were significantly less than those of late supplementation group (Group D) in each histopathologic lesion (p < 0.05). In lesions of carcinoma and papilloma, early and continuous supplementation groups (Groups C and E) appeared to be more effective than normal diet (Group B) on inhibiting cell proliferation, but without statistical significance (p > 0.05).

thumbnail image

Figure 2. Ve/Se time-selectively inhibits cell proliferation during esophageal carcinogenesis in rats. (a) H&E and BrdU immunostaining in different pathologic lesions of rat esophagus. (b) Proliferation index of each lesion was calculated as the number of BrdU-positive cells divided by the total number of epithelial cells. Esophageal tissues were randomly selected from six rats in each group and used for analysis (error bars = SD; n = 6 per group). Six noncontiguous, randomly selected fields were photographed in each lesion type. Values with different superscripts in each column were significantly different based on ANOVA test followed by Tukey-Kramer test (p < 0.05). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Ve/Se time-selectively altered expression of inflammatory markers in carcinogenesis

COX2 and 5LOX are both important markers for inflammation-associated cancers including ESCC.26, 27 As shown by the immunostaining in Figure 3a, COX2 overexpressed at epithelium cytoplasm in premalignant lesions and even at nuclear and more significant in carcinoma. Ve/Se supplementation at the early stage of carcinogenesis (Group C) significantly reduced COX2 expression in lesions of hyperplasia, dysplasia and carcinoma when compared with low Ve/Se (Group A, p < 0.05); however, the late-stage supplementation (Group D) did not (Fig. 3b). Similarly, strong nuclear immunostaining revealed 5LOX overexpression in various pathologic lesions. Moreover, Ve/Se also time-selectively lowered 5LOX level in the dysplastic lesion by the early-stage supplementation (Group C, p < 0.05; Fig. 3c).

thumbnail image

Figure 3. Ve/Se time-selectively alters expression of inflammatory markers during esophageal carcinogenesis in rats. (a) Representative H&E and immunostaining of COX2, 5LOX and NFκB in different pathologic lesions of rat esophagus. (b) COX2 immunoreactive score. (c) 5LOX immunoreactive score. (d) NFκB immunoreactive score. Immunoreactive scores were calculated by multiplying the percentage of positive cells by the grade of staining intensity. Esophageal tissues were randomly selected from six rats in each group and used for analysis (error bars = SD; n = 6 per group). Six noncontiguous, randomly selected fields were photographed in each lesion type. Values with different superscripts in each column were significantly different based on ANOVA test followed by Tukey-Kramer test (p < 0.05). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Immunostaining of NFκB p65 clearly showed NFκB activation and translocation into nuclear of epithelium in carcinogenesis (Fig. 3a). When compared with Group A, Ve/Se supplementation during the early stage (Group C) significantly prohibited NFκB activation in dysplastic lesions of rat esophagus, but the late-stage supplementation (Group D) did not (Fig. 3d). However, no significant difference of NFκB overexpression was observed among various groups in lesions of papilloma and carcinoma (p > 0.05), which probably indicated that Ve/Se supplementation effectively prevented the activation of inflammatory pathway only at the early stage of carcinogenesis.

Ve/Se time-selectively affects AA metabolism through modulating NFκB pathway

Western blots in Figure 4a demonstrated that NFκB expression in esophagus of control group was very low; in contrast, NFκB clearly overexpressed in NMBzA-treated groups (Groups A–E). In comparison with low Ve/Se group (Group A), the level of NFκB was significantly reduced by early supplementation (Group C) and continuous supplementation (Group E, p < 0.05), but not by late supplementation (Group D). This was consistent with the immunohistochemistry result on NFκB p65 protein. Densitometric analysis also showed that early supplementation (Group C) significantly decreased NFκB expression when compared with normal diet (Group B, p < 0.05).

thumbnail image

Figure 4. Ve/Se time-selectively affects arachidonic acid metabolism in rat esophageal epithelium by modulating NFκB pathway. (a) NFκB expression in the esophagi of NMBzA-treated rats. Expression of NFκB p65 in rat esophageal epithelium was determined by Western blotting and was semiquantified by the band intensity relative to that of β-actin. Six pooled esophagus samples (three in one) in each group were applied to Western blotting, and three independent experiments were performed. Values with different superscripts in each column were significantly different based on ANOVA test followed by Tukey-Kramer test (p < 0.05). (b) COX2 mRNA levels. (c) 5LOX mRNA levels. Transcriptional expression of COX2 and 5LOX at mRNA levels in rat esophageal epithelium was determined by real-time PCR, and the amount of target cDNA in each sample was normalized versus β-actin. Six esophagus samples in each group were applied to real-time PCR, and three independent experiments were performed. (d) PGE2 levels. (e) LTB4 levels. Production of PGE2 and LTB4 was determined by EIA assay. Six pooled esophagus samples (three in one) in each group were applied to EIA analysis, and three independent experiments were performed. Each sample was assayed in triplicate in EIA test and expressed as mean; bars, ±SD. Values with different superscripts in each column were significantly different based on ANOVA test followed by Tukey-Kramer test (p < 0.05).

Download figure to PowerPoint

To test whether NFκB activation could promote transcription of downstream gene related to inflammatory process, we measured mRNA levels of COX2 and 5LOX as well as the contents of their metabolites in esophagus mucosa. As shown in Figures 4b and 4c, there were nearly 14-fold and fivefold increase in the amount of COX2 and 5LOX mRNA, respectively, in the esophageal mucosa of low Ve/Se group (Group A) when compared with the negative control group (Group F). In comparison with Group A, both normal diet and Ve/Se supplementation (early, late and continuous) significantly decreased the levels of COX2 and 5LOX mRNA. Furthermore, the level of COX2 mRNA in early supplementation group (Group C) was significantly lower than that in late supplementation group (Group D, p < 0.05). Similar result was found in the analysis on the levels of 5LOX mRNA; however, there was no significant difference between early and late supplementation (p > 0.05).

PGE2 and LTB4 levels in the esophageal tissues were similarly increased by NMBzA treatment in comparison with the negative control (Group F; Figs. 4d and 4e). When compared with low Ve/Se (Group A), Ve/Se supplementation (Groups C–E) significantly reduced PGE2 and LTB4 production (p < 0.05). In addition, early supplementation (Group C) was significantly more effective (p < 0.05) than late supplementation (Group D) in reducing PGE2/LTB4 production when compared with normal diet (Group B).

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Clinical and experimental outcomes raised a hypothesis that intervention time may be important for cancer chemoprevention by nutrients or drugs during the multistages of carcinogenesis. Our study demonstrated that supplementation of Ve/Se at the early stage (0–10 weeks) of esophageal carcinogenesis significantly reduced visible tumor multiplicity and volume, carcinoma incidence and multiplicity, and the number of dysplasia in the esophagus of NMBzA-treated rats was maintained on low Ve/Se diet. In contrast, supplementation at the late stage (11–25 weeks) did not inhibit carcinogenesis in this model based on these parameters. Early supplementation was more effective in suppressing carcinogenesis than late supplementation, which appeared ineffective in reducing dysplasia and carcinoma. These results are in agreement with those from intervention trials in Linxian, which suggest that the timing of the intervention is a factor in affecting the outcome.6, 8–11

In esophageal cancer model induced by NMBzA, free radicals generated from metabolic activation of NMBzA may elicit oxidative DNA damage and produce extra mutations in addition to DNA methylation. Many antioxidative micronutrients or phytochemicals were found to be protective against NMBzA-induced carcinogenesis.28, 29 Some epidemiological and interventional studies and laboratory experiments have suggested the possible role of Ve/Se in suppressing cancer because of their antioxidative functions.30, 31 Our previous study also clearly showed that deficiency of α-tocopherol and Se increased 8-hydroxy-2′-deoxyguanosine (8OH-dG) in pathological lesions induced by NMBzA in our model.22 It was known that NFκB signaling pathway, which can be activated by reactive oxygen species (ROS) and oxidative stress, regulates the expression of specific genes that promote cell proliferation and inflammatory response.32–34 Numerous studies have indicated that the activation of NFκB pathway is one critical mechanism in a variety of cancer types.35–37 Recent research has clearly demonstrated that aberrant AA metabolism is involved in human carcinogenesis, including esophageal cancer.38 COX2 and 5LOX are two critical enzymes in AA metabolism, and their respective metabolites PGE2 and LTB4 are key mediators that link inflammation to cancer via oxidative stress.39 PGE2 and LTB4 are important inflammatory factors that induce inflammatory response, enhance cell proliferation and stimulate angiogenesis in various carcinogenesis including lung cancer, esophageal cancer, colorectal cancer and prostate cancer.40–42 COX2 is an inducible protein that rarely expresses in normal tissues but overexpresses in cancerous tissue including ESCC and esophageal adenocarcinoma (EAC).40, 43 Previous study has demonstrated that COX2 overexpression can be induced by NFκB activation and thereby promoted angiogenesis in colorectal cancer.44 Our results showed that Ve/Se supplementation at the early stage of esophageal carcinogenesis significantly decreased activated NFκB (p65/RelA) expression in rat esophagus especially dysplastic lesions, but late-stage supplementation did not. In addition, we found that the early-stage supplementation of Ve/Se appeared more effective in suppressing COX2 and 5LOX expression at the level of gene transcription. PGE2 and LTB4, the respective metabolites of COX2 and 5LOX, were also more efficiently reduced by early supplementation, and the results correlated well with COX2 and 5LOX mRNA levels. This pathway was also well supported by our study on cell proliferation using BrdU immunohistochemical staining. The results showed that cell proliferation in low Ve/Se group was significantly inhibited by early supplementation but not by late supplementation. These data suggested that NFκB activation and cyclooxygenase and lipoxygenase pathways of AA metabolism may be important to Ve/Se in suppressing NMBzA-induced carcinogenesis in rat esophagus.

In clinical research, nonsteroidal anti-inflammatory drugs (NSAIDs) were applied in the chemoprevention of esophageal cancer and colorectal cancer through suppressing cyclooxygenase and lipoxygenase to reduce inflammatory mediators generation.45–47 Furthermore, it has been indicated by in vitro and in vivo studies that NSAIDs could inhibit the initiation and development of cancer by blocking NFκB activation and the transcription of downstream genes.48, 49 However, some studies showed no benefit in ESCC from intervention with NSAIDs. For example, long-term administration of celecoxib in patients with Barrett's esophagus with dysplasia for 48 weeks did not prevent progression to cancer.50 Another randomized controlled trial of selenomethionine and/or celecoxib for 10 months in residents from Linxian, China, showed beneficial effects in patients with mild esophageal dysplasia, but not in those with severe esophageal dysplasia.11 Therefore, the clinical intervention of esophageal cancer may also be “time selective,” and it probably has no benefit in patients with more severe precancerous lesions. These are consistent with our results on esophageal carcinogenesis and NFκB signaling pathway activation. During the early stage of esophageal carcinogenesis, Ve/Se supplementation could effectively scavenge free radicals and block NFκB activating, consequently suppress COX2/5LOX expression and reduce PGE2/LTB4 production and finally inhibit cell proliferation and cancer progression. Once the malignant lesions are established, signaling pathway and aberrant AA metabolism will be fully activated when inflammatory response and oxidative stress promote each other in a vicious circle.39 During this stage, supplementation with Ve/Se could not prevent progression of the lesions and became less effective in cancer chemoprevention. Our results showing that NFκB and 5LOX expression were time-selectively affected by Ve/Se supplementation in only dysplastic lesion but not in papilloma and carcinoma (Fig. 3), from another side, also supported this view.

Results from other chemoprevention trials with Ve and Se remain inconsistent, which could be attributed to various factors. The effects may be closely related to supplementary dose of nutrients; meanwhile, baseline nutrients levels can modify an individual's response in cancer prevention trails. The SELECT is a trial with pharmacological doses of Se (200 g of L-selenomethionine) and Ve (400 IU α-tocopherol acetate), alone and in combination, for the prevention of prostate cancer in healthy men with rather high levels of plasma α-tocopherol and Se (with median levels of 12.5 and 0.135 μg/ml, respectively). Our study and the Linxian trials were studied at the nutritional levels, in which supplementation provided beneficial effects in individuals who were in low Ve and Se nutritional status. On the other side, our results also did not show apparent benefit of continuous Ve/Se supplementation (Group E) for esophageal carcinogenesis prevention in rats with normal baseline Ve/Se levels (Group B). As opposed to synthetic pharmaceuticals, Ve/Se as naturally occurring dietary constituents are part of normal physiology, and a U-shaped-dose response curve may exist in which either deficiency or supraphysiological doses are harmful.15 In our study, however, we could not explore the interaction between Ve and Se, which is another important factor. Therefore, further investigation on dose-dependent effects of each single agent and analysis on the interactions, with the addition of the effect of supplementation duration as indicated by our study, will contribute to the biological explanation of these discrepancies.

In summary, our study demonstrated that Ve/Se supplementation was time selective in the chemoprevention of NMBzA-induced esophageal carcinogenesis. The early-stage supplementation significantly prevented cancer development, whereas the late-stage supplementation did not show clear benefit. Suppression of cell proliferation by blocking NFκB signaling pathway activation and aberrant AA metabolism might be the underlying mechanism. Our data in an animal model provide further experimental support to the hypothesis that the efficacy of cancer chemoprevention by nutrients may be time selective during the multistages of carcinogenesis.

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_27423_sm_SuppTab1.doc43KSupporting Information Table 1. Esophageal carcinogenesis in NMBzA-treated rats

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.