A ruthenium(II)-curcumin compound modulates NRF2 expression balancing the cancer cell death/survival outcome according to p53 status

The oncosuppressor p53 plays a key role in cell growth and apoptosis in response to various stress signals [1]. Given its central role in maintaining genomic stability and preventing oncogenesis, p53 is the most inactivated oncosuppressor in human tumors by gene mutations or by protein deregulation [2]. Mutant (mut) p53 proteins may acquire a misfolded hyperstable conformation [3] that may be achieved by binding heat shock proteins (HSP) such as HSP90, a cellular chaperone that is crucial for the stability of many client proteins including mutp53 [4, 5]. Besides loss of function and dominant-negative effect on the wild-type (wt) p53 activity, the hotspot p53 mutants may also acquire new oncogenic functions, contributing to cancer progression, invasion and resistance to therapies [6]. Thus, targeting mutp53 is a challenging strategy to halt cancer growth [7]. In this regard, several different approaches have been taken in the last years developing small molecule or using phytochemicals from nature to induce mutp53 degradation or conformational changes, providing new insight on mutp53 reactivation [8, 9], as also demonstrated by our studies [10‐13]. Autophagy has been shown to be involved in mutp53 degradation [14‐23], suggesting the use of autophagy stimulators to counteract mutp53 oncogenic activity. Thus, mutp53 has been shown to counteract autophagy mechanism to likely halt its own degradation [24]. Finally, mutp53 degradation by autophagy has been shown to increase the cytotoxic effects of chemotherapeutic drugs [17]. Mutp53 oncogenic activities ma also depend by modifications of the tumor microenvironment altering the secretion of inflammatory cytokines that affect the crosstalk between cancer and stromal cells [25, 26] or by interaction with other transcription factors such as NRF2 (nuclear factor erythroid 2-related factor 2, encoded by NFE2L2 gene) or HIF-1 (hypoxia-inducible factor 1) to support tumor progression and resistance to therapies [27]. Therefore, understanding the interplay between these oncogenic pathways may have an impact on the development of more efficient targeted anticancer therapies.

NRF2 is the main regulator of cellular antioxidant response [28] and is activated in response to oxidative and/or electrophilic stress, the so-called canonical conditions. Following activation, NRF2 detaches from its negative regulator KEAP1 (Kelch-like ECH-associated protein 1), stabilizes and moves to the nucleus where it binds to sequence-specific responsive elements of anti-oxidant target genes promoters. Among these genes there are catalase, superoxide dismutase (SOD), HO-1 (heme-oxygenase 1), NAD(P)H quinone oxidoreductase 1 (NQO1), and glutathione (GSH), that help to restore the cellular redox homeostasis [29]. Constitutive activation of NRF2 is found in several different tumors also by gain-of-function mutations of the NFE2L2 gene or by inactivating mutations of the KEAP1 gene. These mutations are considered drivers of cancer progression, metastasis, and resistance to therapies [30]. NRF2 noncanonical activation may depend by p62/SQSTM1-mediated KEAP1 degradation [31], or by p21Cip1/WAF1 (target of p53) that binds to KEAP1 to interrupt the KEAP1/NRF2 complex [32]. NRF2 may have both tumor suppressive and tumor-promoting actions and is therefore considered a “double face” molecule [33]. Thus, while NRF2 transient activation is certainly considered cytoprotective, its continual activation may support tumor progression and tumor resistance to therapies. Therefore, NRF2 overexpression in cancer cells may be considered a marker of chemoresistance [34].

Curcumin is considered a chemopreventive molecule with anti-oxidant, anti-apoptotic and anti-inflammatory properties and with an excellent safety profile although it presents low solubility and bioavalibility [35]. Curcumin has been shown to activate the NRF2 pathway triggering cellular protection against oxidative injury that, in advanced stage cancers, may induce chemoresistance [36]. On the other hand, curcumin may induce mutp53 degradation through autophagy or convert mutp53 protein to transcriptionally active wtp53, reversing the therapeutic resistance and improving cancer cell death, as also demonstrated by our studies [16‐18, 37, 38].

In this study we aimed at evaluating the effect of a novel curcumin compound in several cancer cell lines carrying different endogenous p53 status. We used a water-soluble ruthenium(II)-curcumin (RuCUR) compound that presents high solubility and cytotoxic effect ]compared to other similar curcumin compounds [39‐46]. Moreover, the compound show selectivity toward cancer cells over nontumorous cells, suggesting an in vivo use [42]. We found that RuCUR induced cancer cell death in all tested cell lines and independently of the p53 status. At molecular level we found NRF2 activation, mutp53 degradation and/or wtp53 activation. Pharmacologic or genetic NRF2 inhibition further increased the RuCUR-induced cell death in both mutp53- and wtp53-carrying cancer cell lines while did not increase cell death in p53 null cells. In the present study, we found that activation of the NRF2 antioxidant pathway contributed to resistance to RuCUR treatment, despite mutp53 degradation and activation of wtp53. Furthermore, inhibition of NRF2 could overcome this resistance in particular in cancer cells carrying p53 either wild-type or mutant, indicating the crucial role for a functional p53 pathway to efficiently induce cell death in response to treatments.

The findings of the present study revealed that a novel water-soluble ruthenium(II)-curcumin compound (RuCUR) [42] was able to induce cancer cell death that correlated with mutp53 downregulation and with activation of wtp53; they also revealed a resistance mechanism via the NRF2-induced antioxidant system likely enhanced by ROS reduction. In addition, our findings suggest that NRF2 inhibitors can overcome the RuCUR resistance via inhibition of the antioxidant system (Fig. 8a, b). However, the lack of p53 did not contribute to counteract the death resistance even following NRF2 inhibition, underlying the important role of p53 (re)activation for cancer cells demise.

Mutp53 proteins may promote tumor invasion, metastasis and chemoresistance, disclosing mutp53 as a guardian of the cancer cells [59]. Restoration of wild-type p53 function prompts the rapid elimination of human cancers carrying a functional loss of p53, often by gene mutation [7‐9]. Many p53-reactivating compound have been shown to induce anti-tumor effects in several tumor types [60]. A number of evidences are now supporting the role of autophagy in mutp53 degradation. Autophagy is a catabolic process that helps to eliminate unfolded proteins or damaged organelles, promoting cell survival [61, 62] even if, in some instances, it may induce cell death [63]. Besides mutp53, other oncogenic proteins are degraded via autophagy including BCR-ABL, PML-RAR, Ret, KIT and Myc, consistent with the tumor-suppressive activity of autophagy [64, 65]. Therefore, inducing autophagy may contribute to cancer cell death and to disable oncogenes but it can also have possible disadvantages due to autophagy crosstalk with the immune response which is fundamental for the success of anticancer therapies [65, 66].. Interestingly, a mutual interplay between autophagy and mutp53 may occur: while autophagy degrades mutp53, mutp53 inhibits autophagy counteracting its own elimination as a self-protecting mechanism, thus promoting chemoresistance [67]. Curcumin has been shown to induce autophagy [38] and to contribute to mutp53 degradation [16‐18, 37]. In line with this findings, here we found that RuCUR triggered mutp53 downregulation likely through autophagy; in addition, it also induced HSP90 reduction, a molecular chaperone critical for maintaining mutp53 stability [4] protecting mutp53 from MDM2-induced degradation [68]. Curcumin has been shown to have an indirect effect on HSP90 downregulation [54], although the exact mechanism of HSP90 reduction in our setting needs to be further explored. Interestingly, HSP90 chaperoning activity can target also wtp53, promoting its transcriptional activity [53], underscoring the two-faced role of molecular chaperones in p53 activity [69]. Thus, here we found that RuCUR increased HSP90 levels in wtp53 cells that correlated with p53 activation, although the mechanism of HSP90/wtp53 regulation has not been clearly elucidated yet.

Curcumin may induce NRF2 pathway activation [36]. NRF2 induces cytoprotective genes that, on one hand, protect cells from the oxidative stress [28] and, on the other hand, induce cancer cells proliferation, resistance to drugs, and apoptosis inhibition [70]. NRF2 may also interact with mutp53 [56, 58] suggesting a criminal alliance to sustain cancer cells [71]. Here, we found that RuCUR induced NRF2 stabilization in all cell lines, regardless of the p53 status. Several mechanisms can correlate with NRF2 stabilization such as ROS inhibition, increased p62/SQSTM1 [31] or p21^Cip1/WAF1 levels [32]. We found that RuCUR reduced ROS generation in both wtp53 and mutp53 cells, it also induced p62 in mutp53 cells and p21 in wtp53 cells, although the exact mechanisms of NRF2 stabilization in our setting need to be further explored. Interestingly, NRF2 inhibition increased cell death in both wtp53 and mutp53 cells, underscoring the resistance effect of the antioxidant system. At molecular level, NRF2 inhibition further decreased mutp53 levels, underscoring the link between NRF2 and mutp53 to sustain cancer cell survival (Fig. 8). NRF2 inhibition in wtp53 cells impaired the RuCUR-induced p21 and DRAM expression, while increased Puma expression, suggesting that inhibiting NRF2 could reestablish p53 apoptotic activity [48, 52]. The mechanism of NRF2 inhibition of the p53 apoptotic function is an interesting filed of study and we hypothesize that NRF2 and the oxidative system may impair the function of the p53 apoptotic activator homeodomain-interacting protein kinase 2 (HIPK2) [72‐75], although further studies are needed to demonstrate this hypothesis. It is worth to note that p53 apoptotic activation in turn inhibits NRF2 cytoprotective function that may hamper the p53-induced apoptosis [76], in a complex regulatory loop between NRF2 and p53 (Fig. 8). Finally, reduction of the antioxidant system increased cell death only in mutp53 and wtp53 cells while it did not have this effect in p53 null cells. These findings underscore the critical role of p53 for efficient cancer cell death, although the exact mechanisms need to be further explored.

This study suggests that the overexpression of the antioxidant system is involved in the mechanism of cancer cell resistance to RuCUR treatment. The elimination of mutp53 and the activation of wtp53 induced cell death although only after inhibition of the antioxidant system cell death greatly improved. Interestingly, p53 null cells did not undergo increased cell death following inhibition of the NRF2 pathway, underscoring the important role of p53 in cancer cell death. These findings also suggest that the use of phitochemicals that can increase the antioxidant system and, for this reason, being useful antiaging agents, may not be beneficial for the demise of tumors, despite autophagy induction targeting oncogenes such as mutp53, underlining the complex relationship between autophagy, antioxidant system and tumor cell death. In conclusion, these findings may represent a paradigm for better understanding the complex interplay between NRF2 and p53 molecular pathways, in order to design more efficient anticancer therapies. Further preclinical and clinical investigations of RuCUR /NRF2 inhibition combination should be performed in order to explore the anticancer activities.