Arsenic trioxide is a documented environmental toxicant and a potent chemotherapeutic agent, which has been used therapeutically for decades[
1]. Besides being an anticancer drug against acute promyelocytic leukemia (APL), arsenic trioxide has also been proved as an effective compound that can inhibit the growth of many solid tumors both
in vitro and
in vivo[
1,
2]. It has been reported that liver is the most important site of arsenic biotransformation by alternating reduction of pentavalent arsenic to trivalent and addition of a methyl group from
S-adenosylmethionine[
3]. This evidence suggests that the hepatocellular carcinoma (HCC) cells could be more likely to be killed as a result of the aggregation of intracellular arsenicals after arsenic trioxide treatment. However, a recent phase II trial showed that single agent arsenic trioxide was poorly effective against advanced liver cancer with failure to increase the five-year survival rates[
4], which was inconsistent with the
in vitro reports. Although lots of factors could explain the inefficacy of arsenic trioxide in liver cancer patients, anti-cancer drug resistance might be the most important reason for this problem[
5].
The increased drug efflux is defined as a characteristic of the multidrug resistant phenotype. Overexpression of transporters from ATP-binding cassette (ABC) superfamily is one of the most common reasons contributed to drug resistance. It was reported that the As-GSH conjugates is substrates of some ABC transporter proteins and could be pumped out by the ABC superfamily members[
6,
7]. In our previous study, we found that arsenic trioxide resistant HCC cells overexpressed p-glycoprotein (p-gp), which could decrease the intra-cellular arsenicals[
5]. The resistant cells also overexpressed MDM2, which could inactivate p53 or p73, leading to the defence of apoptosis induced by arsenic trioxide. Interestingly, the expression of p53 was increased in arsenic trioxide resistant cells, suggesting there might be p53 mutations, which could lead to the stabilization of p53. In the current study, we hypothesized that long-term exposure of cells to arsenic trioxide in the stepwise selection of arsenic trioxide resistant HCC cells induced p53 mutations, which can result in arsenic trioxide resistance. Fortunately, unlike p53, another member of the p53 family, p73, is rarely mutated in cancers[
8]. In addition, a few stimuli, including arsenic trioxide, have been identified to induce p73 and subsequent apoptosis in cancer cells[
8,
9]. However, although arsenic trioxide could induce p73, some negative moderators of p73, such as mutant p53 (mutp53) and MDM2 can suppress the apoptotic function of p73[
10]. As reported, the most common p53 mutation is single amino acid substitutions in the DNA binding domain of the p53 protein. In addition to the loss of tumor suppressive activities of wild-type p53, many tumor-associated mutp53 proteins gain new oncogenic functions, defined as gain-of-function (GOF), which enable them to promote tumorigenesis, metastasis and chemoresistance[
11,
12]. Therefore, we hypothesized that p53 mutation could be an ideal target to restore the sensitivity of HCC resistant cells to arsenic trioxide and inhibit HCC tumor metastasis.
Nutlin-3, a novel MDM2 inhibitor, has been shown to inhibit the p53-MDM2 or p73-MDM2 interaction, leading to the stabilization of p53 or p73 protein[
10,
13]. Furthermore, Nutlin-3 has also been reported to interfere with p-gp-mediated drug efflux for acting as a transporter substrate[
14]. It revealed a potential therapeutic way for HCC resistant cells, especially in combination with arsenic trioxide. We designed this study to investigate the underlying mechanism of arsenic trioxide resistance and to evaluate whether Nutlin-3 could reverse the resistance.