Background
Sulindac is a non-steroidal anti-inflammatory drug (NSAID) that has shown to be effective in preventing intestinal tumors in familial adenomatosis polyposis patients that inherit a mutant allele of the adenomatosis polyposis coli (
APC) gene [
1,
2], in inhibiting tumor formation in mouse genetic models (ApcMin and Apc1638N+) in which an allele of the homologous mouse
Apc gene is inactivated by a mutation [
3‐
5], and in inhibiting carcinogen-induced colon tumor formation in rats [
6]. Our previously studies have demonstrated that loss of
p21WAF1/Cip 1 gene could enhance intestinal tumorigenesis in the Apc mouse model (i.e.
Apc/p21 mice) [
7]. Most interestingly, dietary supplemental sulindac was able to inhibit tumor formation in the
Apc+/−/p21+/+ mice, but not in the mice in which even a single p21 allele was inactivated (i.e.
Apc+/−/p21+/−) [
3], because the remaining wild-type p21 allele is inactivated by hypermethylation of a CpG cluster in its promoter in both the intestinal mucosa as well as in the tumors that form [
8]. To determine whether the hypermethylated mouse p21 promoter could be demethylated, and then the response to sulindac on intestinal tumor formation could be resumed, we fed the
Apc/p21 mice with diets supplemented with selenium or combination of selenium and sulindac in present study to determine the intestinal tumor inhibition.
Selenium, an important trace element, is essentially involved in different physiological functions in mammalian and human body. Selenium has significant activity as a chemoprevention agent for cancer. Epidemiological and experimental studies have suggested that intake of dietary selenium is inversely related to overall cancer risk. The effect was most pronounced in gastrointestinal and prostate cancer [
9‐
11]. In addition,
in vivo studies have demonstrated that dietary selenium supplementation can reduce cancer incidence in animal models of melanoma and cancers of colon, breast, liver, esophagus, head and neck, kidney and lung [
10,
11]. The anti-cancer effects of selenium have been postulated to link to inhibition of cell proliferation and induction of apoptosis through different signaling pathways, particularly the anti-oxidative and anti-inflammatory effects mediated through the activity of selenoenzymes [
12]. While the targets and underlying mechanisms of anti-cancer action by selenium are largely unknown. Recently, our group found that sodium selenite inhibits intestinal carcinogenesis
in vivo and in vitro through a novel anti-cancer mechanism - activating JNK1 and suppressing β-catenin signaling [
13], in addition to the action of selenium of impacting methylation by inhibiting DNA methyltransferase [
14,
15].
The unique Apc/p21 mouse model of intestinal tumor was applied in the current student. We found that selenium was synergistic with sulindac and exerted maximum tumor inhibition efficacy through inhibiting p21 promoter methylation, inducing p53, p27 and phosphorylation of JNK1, and suppressing Wnt/β-catenin signaling, although selenium alone showed slight inhibitory effect in the Apc/p21 mice.
Discussion
Epigenetic alterations, particularly hypermethylation of CpG islands located in the promoter regions of tumor suppressor genes, contribute to tumor formation. Therefore, developing chemopreventive strategies using the agents targeting methylation will have great translational potential. Using the unique mouse model of intestinal tumors in this study, incorporating with our earlier results [
3,
4,
8], we found that trace element selenium and sulindac synergistically inhibited intestinal tumor formation.
Several lines of evidences have proven that selenium is effective in preventing tumorigenesis. For instance, several randomized controlled trials have demonstrated significant cancer prevention of selenium in colorectal cancer [
16‐
19] and in
Muc2/p21 mouse model of intestinal cancer reported recently by us [
13]. However, the selenium and vitamin E clinical trial (SELECT) on prostate cancer prevention was failed and had to be terminated earlier [
20], although several factors should be considered. In the current Apc/p21 mouse model, selenium alone showed slight effect on tumor inhibition. It has been demonstrated that anti-cancer action of selenium is through multiple mechanisms. One of the novel mechanisms is to impact methylation. Unlike previous reports, we observed slight tumor inhibitory effect of selenium, and almost no effect on methylation. Sulindac alone has not been reported any demethylation or inhibitory effect on DNA transmethylferase. However, the combination of selenium and sulindac dramatically inhibited tumor formation and significantly affected mouse p21 promoter methylation level. Therefore, this combination of selenium and sulindac could pave a new path for cancer prevention, and had promising translational significance.
Our mechanistic study provided further insight of the combination of selenium and sulindac in cancer prevention, for example, selenium was synergistic with sulindac to suppress Wnt/β-catenin signaling and induce Wnt/β-catenin inhibitors. It is worth to point out that, the combination of selenium and sulindac significantly induced the expression of p27 and p53 and phosphorylation of JNK1. It is well known that abnormally low levels of p27 protein are frequently found in human carcinomas, and these low levels are directly correlated with histological aggressiveness, lymph node metastasis and poor prognosis of esophagus, gastric, breast and colorectal carcinomas [
21,
22]. Overexpression of p27 and p53 could downregulate β-catenin and its downstream targets, inhibit cell proliferation and promote cell differentiation and apoptosis [
23‐
25]. Most importantly, phosphorylated JNK1 plays a critical role in intestinal cell maturation, e.g. inducing apoptosis and differentiation and inhibiting proliferation [
26] through negatively interacting with β-catenin and facilitating β-catenin degradation
in vivo and in vitro[
27,
28]. There are a lot of JNK1 stimulating agents, among them, both selenium and sulindac are most effective in activating JNK1 [
13,
29]. Thus, the combination of selenium and sulindac could exhibit maximum tumor inhibitory effects in current study. These results were different with our previous study in another mouse model of intestinal cancer –
Muc2/p21 mice [
13], in which we found that selenium supplemented in the Western-style high risk diet was able to inhibit intestinal tumorigenesis and the selenium alone was likely to induce JNK1 phosphorylation and to inhibit β-catenin and Cox-2. The causes of this discrepancy could be resulted from the differences of mouse model backgrounds and diets. In that study [
13], the
Muc2/p21 model was generated from mating
p21−/− mice with
Muc2 knockout mice, in which
Muc2 gene plays a critical role in causing tumor formation and β-catenin is inactivated [
30]; in the current study, the
Apc/p21 mice were generated from mating of
p21−/− mice with
Apc1638+/− mice, in which
Apc mutation plays a causing role on tumor formation because β-catenin is activated [
31]. In addition, in the previous study, sodium selenite was added into the high-risk Western-style diet with high-fat and low-vitamin D; in the current study, the selenium was added into a defined AIN-76A control diet.
With novel signaling pathways involving in carcinogenesis and angiogenesis have been discovered, whether preventive or therapeutic agents can target to these pathways are large unknown. For example, the polo-like kinase 3 (Plk3) has been recently found to phosphorylate and destabilize hypoxia-inducible factor 1α (HIF-1α). Plk3 can also phosphorylate and stabilize PTEN phosphatase, a known regulator of HIF-1α and tumor angiogenesis [
32]. Whether the combination of selenium and sulindac can target on Plk3-HIF-1α-PTEN signaling, for colorectal cancer prevention and therapy, is under investigation. Recent reports have shown that gastrointestinal stromal tumors (GIST) account for approximately 1% to 3% of all malignant GI tumors [
33], but the treatment for GIST is still a big challenge. Whether the combination of selenium and sulindac has similar effects on GIST prevention is unknown and needs to be investigated.
In conclusion, selenium and sulindac are synergistic to inhibit intestinal tumorigenesis in the Apc/p21+/− mice through impacting methylation, inducing the expression of p27, p53 and phosphorylation of JNK1, and suppressing β-catenin signaling. Current finding provides an important chemopreventive strategy using combination of anti-cancer agents instead of a single one. Therefore, the finding from this study has a great impact on cancer prevention and has a promising translational potential.
Acknowledgements
We would like to thank Drs.Wenfeng Fang and Anjia Han (University of Illinois at Chicago, Chicago, IL) and the Research Resources Center (University of Illinois at Chicago, Chicago, IL) for technique support, and thank Dr. Leonard Augenlicht (Albert Einstein Cancer Center, Bronx, New York) for the Apc1638N+ mice and Dr. Philip Leder (Harvard Medical School, Boston, MA) for the p21−/− mice.
Grants support
This work was supported in part by National Institutes of Health grant, USA (R01 CA112081 to W. Yang), the grants from the Nature Science Foundation of China (Grants 81272251 and 91229115 to W.Yang, Grants 81001003 and 81272333 to X. Bi) and the Program of Liaoning Excellent Talents in University (LETU #LJQ2011002 to X. Bi).
Competing interests
All authors have no competing interests to declare.
Authors’ contributions
YW designed and guided the experiments, BX, PN and DH conducted the experiments, BX and YW analyzed the data and wrote the manuscript. All authors read and approved the final manuscript.