Pleiotropic Effects of Tocotrienols and Quercetin on Cellular Senescence: Introducing the Perspective of Senolytic Effects of Phytochemicals

See discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/281588707

Pleiotropic effects of tocotrienols and quercetin on cellular senescence: introducing the perspective of senolytic effects of phytochemicals

ARTICLE in CURRENT DRUG TARGETS · SEPTEMBER 2015 Impact Factor: 3.02

CITATION READS 1 75

8 AUTHORS, INCLUDING:

Marco Malavolta Elisa Pierpaoli INRCA Istituto Nazionale di Ricovero e Cura p… INRCA Istituto Nazionale di Ricovero e Cura p…

118 PUBLICATIONS 2,156 CITATIONS 20 PUBLICATIONS 150 CITATIONS

SEE PROFILE SEE PROFILE

Laura Costarelli Francesco Piacenza INRCA Istituto Nazionale di Ricovero e Cura p… INRCA Istituto Nazionale di Ricovero e Cura p…

50 PUBLICATIONS 854 CITATIONS 34 PUBLICATIONS 344 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Marco Malavolta Retrieved on: 26 October 2015 Pleiotropic effects of tocotrienols and quercetin on cellular senescence: introducing the perspective of senolytic effects of phytochemicals

Marco Malavolta1*, Elisa Pierpaoli2, Robertina Giacconi1, Laura Costarelli1, Francesco Piacenza1, Andrea Basso1, Maurizio Cardelli2, Mauro Provinciali2

1 Nutrition and aging Centre, Scientific and Technological Pole, Italian National Institute of Health and Science on Aging (INRCA)

2 Advanced Technology Center for Aging Research, Scientific Technological Area, Italian National Institute of Health and Science on Aging (INRCA)

*Corresponding Author: Nutrition and aging Centre, Scientific and Technological Pole, Italian National Institute of Health and Science on Aging (INRCA), via Birarelli 8, 60121, Ancona, Italy; Phone: +39-0718004113; Fax: +39- 071206791; email: [email protected]

Abstract

The possibility to target cellular senescence with natural bioactive substances open interesting therapeutic perspective in cancer and aging. Engaging senescence response is suggested as a key component for therapeutic intervention in the eradication of cancer. At the same time, delaying senescence or even promote death of accumulating apoptosis-resistant senescent cells is proposed as a strategy to prevent age related diseases. Although these two desired outcome present an intrinsic dichotomy, there are examples of promising natural compounds that appear to satisfy all the requirements to develop senescence-targeted health promoting nutraceuticals. Tocotrienols (T3s) and quercetin (QUE), albeit belonging to different phytochemical classes, display similar and promising effects “in vitro” when tested in normal and cancer cells. Both compounds have been shown to induce senescence and promote apoptosis in a multitude of cancer lines. Conversely, they display senescence delaying activity in primary cells and rejuvenating effects in senescent cells. More recently, QUE has been shown to display senolytic effects in some primary senescent cells, likely as a consequence of its inhibitory effects on specific anti-apoptotic genes (i.e. PI3K and other kinases). Senolytic activity has not been tested for T3s but part of metabolic and apoptotic pathways affected by these compounds in cancer cells overlap with those of QUE. This suggests that the rejuvenating effects of T3s and QUE on pre-senescent and senescent primary cells might be the net results of a senolytic activity on senescent cells and a selective survival of a sub-population of non-senescent cells in the culture. The meaning of this hypothesis in the context of adjuvant therapy of cancer and preventive anti-aging strategies with QUE or T3s is discussed.

Keywords: Cellular Senescence, Cancer, Nrf2, Tocotrienols, Quercetin, endoplasmic reticulum stress, senolytic compounds

Introduction

Cellular senescence (CS) is a biological response to a variety of stresses that results in persistent growth arrest with a distinct morphological and biochemical phenotype [1]. It is currently considered a “barrier” to prevent malignant transformation and, consequently, a potent anti-cancer mechanism. Exploration of CS as a target to drive towards antitumor adjuvant therapies or age-related diseases preventive strategies by natural substances is currently gaining increasing interest. Cancer cells can be forced to undergo senescence by natural substances, with effects somewhat comparable to those obtained by genetic and epigenetic manipulations, anticancer drugs, and irradiation [1]. These effects have been shown after sustained exposure “in vitro” to a wide range of different substances that are paradoxically used also to obtain cytoprotective and chemopreventive adaptive responses in normal cells [2]. Interestingly, a recent revision of literature on modulators of cellular senescence “in vitro” [2] identifies quercetin (QUE) and tocotrienols (T3s) as the only two phytochemicals (excluding carnosine which is not a phytochemical) that display all the 3 senescence- targeted outcomes that are commonly tested in vitro: 1) induction of a senescent-like phenotype in cancer cells; 2) delay of senescence markers onset in normal primary cells; and 3) rejuvenation of senescent and pre- senescent cells. This similarity could appear as a paradox If we consider the structural and physical differences of T3s (belonging to the methyl-tocols class of phytochemicals) and QUE (belonging to the flavonols class of phytochemicals. However, there are multiple area of overlap in their biological activity that might help to explain their similar outcome in primary senescent and cancer cells. In the first part of this review, we describe in brief the phenomenon of cellular senescence and the similitude between cancer cells and senescent cells. We then describe the effects of QUE and T3s in cancer and cellular senescence and integrate this knowledge with a meta-analysis to propose a unique mechanism of action that might be used to form a rationale to design new interventions in cancer and aging.

Cellular Senescence in Brief

CS is usually defined as a status of growth arrest mediated by insensibility to mitogen stimuli, chromatin and secretome changes, and upregulation of particular tumor suppressor pathways [3-4]. Currently, the causes and mechanisms involved in the phenomenon of CS represent an intense area of investigation and most aspects are still unclear. It is known that a variety of cell-intrinsic and -extrinsic stresses can activate the CS program. These include DNA damage, oxidative stress, critical telomere shortening and damage, chronic mitogenic signaling, oncogene activation and inactivation, loss of tumor suppressors, nucleolar stress, epigenetic changes and others [5]. The pathways and hallmarks shared by these processes have multiple areas of overlap that usually involve p53-p21 and p16-Rb pathways. Moreover, a persistent DNA damage response (DDR) stimulated by unrepaired DNA lesions appears in several experimental models of CS [6]. In replicative senescence, critical shortened telomeres are recognized as DNA breaks with subsequent stabilization of p53 and induction of senescence [7]. Stress and oncogene induced senescence work mainly through the activation of p16, but an interplay between p53 and p16 pathways has been frequently reported [8], so that p16 is currently considered a biomarker of aging [9] and one of the most prominent indicators of the presence of senescent cells in aged tissues [10]. One of the most common mediators of CS is the inhibition of Rb phosphorylation, which results in the inactivation of the E2F transcription factor and its target genes involved in cell cycle progression [11]. The activation of growth pathways, via the mammalian Target of Rapamycin (mTOR), and the autophagic response appear additional important players in establishing CS [12]. One of the major problems in research around CS is the characterization of senescent cells and the absence of a universal biomarker. Therefore, the best way to characterize senescent cells appears to consist in the use a pool of different biomarkers. In addition to activation of p16 and p53 pathways, other hallmarks that can be used to identify senescent cells include an altered morphology (i.e. increased size), activation of senescence-associated β-galactosidase (SA-β-Gal), emergence of chromatin aggregates (heterochromatin foci), markers for activated DNA damage and production of the SASP that include several proteins involved in the inflammation process [13]. However, recent observations support the hypothesis that senescence can be a highly dynamic, multi-step process, during which the properties of senescent cells continuously evolve and diversify, much like tumorigenesis but without cell proliferation as a driver [3]. According to this point of view, senescent cells are roughly subdivided into two main classes (acute senescent cells and chronic or late senescent cells) based on their kinetic of induction and functionality. Acute senescent cells are formed from specific type of cells as a consequence of a biologic program useful for wound healing, tissue repair and embryonic development. Late senescent cells are formed from almost all type of cells (including post-mitotic cells such as neurons and cardiomyocytes) and seem to be not part of a biological program. This last example of CS develops after periods of progressive cellular stress or damage, such as the aging process.

Studies of human tissues and cancer-prone mice argue strongly that CS is one of the most important processes to suppress cancer in vivo [4]. Nevertheless, the altered functional profile of senescent cells might alter tissue microenvironments in ways that can promote cancer in the surrounding microenvironment.

This is most likely the consequence of the SASP produced by senescent cells. Pro-inflammatory cytokines, metalloproteases and chemokines are among the most conserved SASP components [14]. These factors are important to promote clearance of senescent cells by the immune system. However, the pro-inflammatory products of the SASP can induce deleterious effects in the microenvironment by damaging neighboring cells, thus facilitating tumor development and aggressiveness [15], mediating paracrine transmission of CS [16] and promoting age-related dysfunctions. Most importantly, age-related immunodeficiency or the production of a less pro-inflammatory SASP in late senescent cells could hamper the process of immune mediated clearance of senescent cells [3]. These age-related microenvironmental changes and reduced clearance of senescent cells could in part explain why cancer rate markedly increases beyond middle age [4].

Taking into account that some senescent cells can display long term survival and resistance to apoptosis [17], it is likely that this represents one of the most intriguing examples of antagonistic pleiotropy in the context of aging. A direct demonstration that senescent cells can drive age-related pathologies has been recently provided with the development of a transgenic mouse model, in which p16-expressing cells can be speciﬁcally eliminated upon drug treatment, with consequent prevention, delay, or attenuation of some age-related disorders [18]. This is currently raising lot of research aimed to find compounds, even from food or natural sources that are able not only to promote senescence in cancer cells, but also to promote clearance of senescent cells.

The surprising similitude between cancer and senescent cells Cellular senescence, as opposite to the uncontrolled proliferative state of cancer cells, consists essentially of an anti-proliferative program. However, excluding the barriers to the progression of cell cycle, there are several intriguing similitude between cancer and senescence. First of all the phenotypic diversification of senescent cells suggests a state of genomic instability that is also an intrinsic feature of cancer cells. Cellular senescence and cancer can be both related by the involvement of the DDR pathways and by deep chromatin rearrangements [6]. The recent observation that senescent cells can harbor an epigenome more similar to cancer cells then normal cells is indicative of the similitude between these two cellular fates [19]. In particular, replicative senescent human cells (IMR90 cells) exhibit widespread DNA hypomethylation and focal hypermethylation qualitatively similar to those observed in cancer. This ‘reprogrammed’ epigenome is also retained when cells bypass senescence (by infection with a lentivirus encoding simian virus 40). Moreover, while transcriptional activation of endogenous retroelements (ERs) is well established in cancer [20,21], various evidences recently show that expression and retrotransposition are enhanced also in senescent cells [22,23]. Hence, it is not surprising that some critical pathways may act similarly in both cellular states. In particular, pathways leading to resistance to apoptosis and upregulation of pro-survival genes seem to be a common feature of cancer and of various senescent cells and are likely to appear as an attractive therapeutic target. Indeed, resistance to undergo apoptosis is not only a feature of aggressive cancers, but is also among the mechanisms proposed to explain why senescent cells accumulate in aging. Anti-cancer drugs that down- regulate PI3/Akt and Ephrin-B (EFNB) signaling, STAT3 survival proteins and other pathways that repress apoptosis are effective in a multitude of cancers. The hypothesis that senescent cells, like cancer cells, are dependent on these anti-apoptotic pathways to ensure their survival during stress and damage has been recently tested. The study has led to the discovery that Desatinib, a Src tyrosine kinase inhibitor that interferes with EFNB-mediated repression of apoptosis, is a selective ablator of senescent cells, in particular against senescent pre-adipocytes [24]. Interestingly, also QUE was identified as another senolytic compound, more specific for senescent endothelial cells, likely by a mechanism of action involving PI3/Akt and STAT3 pathways. In the same study, it has been also proposed that anticancer drugs targeting p21, BCL-xL, and related genes might also have senolytic effects. This suggests also that phytochemicals targeting these pro-survival pathways or that display similar mechanisms and effects of QUE might display senolytic activity. The striking similarity of effects displayed by QUE and T3s “in vitro” open the way to studies addressing the involvement of anti-survival and senolytic activity of these compounds in various experimental settings.

In vitro effects of Tocotrienols in cellular senescence and cancer

Vitamin E is a generic term used indiscriminately to refer to eight different isomers that belong to two classes: α-, β-, γ- and δ-tocopherols and α-, β-, γ- and δ-T3s. For decades, this term has been used as a synonymous for α-tocopherol, the first identified and the most abundant isoform in nature. T3s are considerably less widespread than tocopherols in the plant kingdom [25]. Palm oil is one of the most abundant natural sources of T3s, with crude palm oil (also referred to as the TRF, Tocotrienol-rich fraction) containing up to 70% of T3s. Nevertheless, more T3s concentrated preparations have been described, the lipid fraction of annatto (Bixa orellana L.) seeds resulting tocopherol free with 100% of T3s content (90% δ- and 10% γ-T3s) [26]. In the last years, T3s have been of increasing interest due to the discovery of interesting biological properties, including rejuvenating-like effects on senescent human fibroblasts (HFs) in culture [27]. In particular, incubation of senescent HFs with TRF decreased SA-β-gal staining and markers of DNA damage, elongated telomeres and reversed the morphology of the cells to resemble that of young cells. In addition to this rejuvenating ability, T3s display strong anticancer effects, not generally evident with tocopherol-rich vitamin E preparations [28]. Among the four isoforms of T3s, γ and δ are those which have proven greater effectiveness in countering the proliferation of tumor cells. Cell culture studies suggest that T3s affect numerous pathways linked with tumor survival, including the apoptotic pathways related to BCL-xL [29,30], the PI3K/Akt signaling pathways [31] as well as the nuclear factor erythroid 2-related factor 2 (Nrf2) and Kelch-like ECH-associated protein 1 (Keap1) pathway [32]. Interestingly, breast cancer cells were found to be significantly more sensitive to the inhibitory effects of γ-T3s on PI3K/Akt signaling than normal cells [31], thus suggesting the potential use of these phytochemicals not only to target cancer but also to target senescent cells. T3s are well known to induce apoptosis through activation of intrinsic mitochondria-mediated [33] and/or extrinsic death receptor-mediated [34] mechanisms, with a different sensitivity for T3s-induced apoptosis among types of cancer. Gamma- and δ-T3s can also induce paraptosis-like death (a programmed cell death that does not involve nuclear fragmentation, chromatin condensation, or formation of apoptotic bodies) in human colon cancer cells [35]. Further anti-cancer mechanisms of T3s possibly associated with cell toxicity include the down-regulation of various mitogenic signal/survival factors [33,36] and the inhibition of angiogenesis [37]. A lesser investigated molecular mechanism that also contributes to extend the spectrum of antitumor action of T3s concerns the ability to induce cell cycle arrest and senescence-like phenotype in various cancer cells “in vitro”. Genes directly or indirectly involved in cell cycle control, such as p21, p27 and p53, may represent the downstream effectors in the anti-cancer signaling of T3s, influencing the balance between signals that drive the cell into senescence pathways and cell cycle arrest or alternatively toward apoptotic cell death. In malignant mouse +SA mammary epithelial cells, 4 μM γ-tocotrienol significantly inhibited cell proliferation with reduction in cell cycle progression from G1 to S, as evidenced by increased p27 level, and a corresponding decrease in cyclin D1, CDK2, CDK4, CDK6 and phospho-Rb levels [38]. Similar results, including increased p16 protein levels, were also observed with a combined treatment of γ-T3s with the natural polyphenolic compound sesamine in neoplastic mouse (+ SA) and human (MCF-7 and MDA-MB-231) mammary cancer cells [39]. These results have been replicated even with mixtures of γ- and δ-T3s, as shown by the upregulation of p53, p21 and p16 in HER-2 overexpressing cell lines [40]. Interestingly, these pro-senescent effects were also associated with increased apoptosis or general cytotoxic effects that are observed also in studies “in vivo”. Oral administration of 100 mg/kg annatto-T3 delayed the spontaneous onset of mammary tumor and reduced tumor number and size through enhancing in situ both apoptosis and senescence markers in a HER2/neu breast cancer mouse model [40]. In this mouse model, T3s have been shown to specifically accumulate in cancer tissues of HER2/neu mice at a very high rate than observed in normal tissues. Hence, the specific intracellular accumulation of these compounds in cancer cells might be responsible for the pro-senescence and pro-apoptotic effects shown by T3s in particular experimental models. Another upstream target of T3s that could mediate senescent-like response or apoptosis in breast cancer cells are estrogen receptors (ER) [41]. T3s display high affinity for ERβ and increase its translocation into the nucleus which, in turn, activates the expression of estrogen-responsive genes (MIC-1, EGR-1 and Cathepsin D) involved in growth arrest, altered morphology and apoptosis of ERβ expressing breast cancer cells (MDA-MB-231 and MCF-7) [42]. Hence, the idea that these compounds might promote senescence and apoptosis in cancer cells while displaying anti-senescence effects in normal cells sounds very promising in view of its potential clinical applications.

In vitro effects of Quercetin in cellular senescence and cancer

Quercetin (QUE) (3,3′,4′,5,7-pentahydroxyl-flavone) is a ubiquitous molecule present in most plants, fruits and vegetables. Apples and onions are the most common food sources of QUE. QUE is also famous its anti- senescence activity in normal cells. Senescent fibroblasts treated with 6-7 μM of QUE for 5 consecutive days were shown to restart proliferation compared to the control cultures [43]. QUE has been shown also to modulate signal transduction pathways involving Nrf2/keap1, which are associated with the processes of inflammation, stress response and carcinogenesis thus suggesting a potential role in cancer prevention [44]. However, there are lots of studies that propose the use of QUE to induce apoptotic and non-apoptotic forms of cell death in cancer cells. Anti-cancer activity of QUE at doses from few to above 50 μM has been demonstrated in lots of cancer cellular models including glioma [45], osteosarcoma [46], cervical cancer [47], prostate cancer [48], breast cancer [49,50], colorectal cancer [51], myeloid leukemia [52] and oral cavity cancer [53]. QUE has been shown to display various activities that are important for cancer survival and progression such as regulation of cell cycle, interaction with type II estrogen binding sites and tyrosine kinase inhibition [54,44]. The complex interactions and targets of QUE that affect molecular phosphorylation states and gene expression thus appear to either inhibit or strengthen the survival signals.

It is surprising the finding that in certain circumstances it is also possible to use QUE to induce senescence in cancer cells. Chronic administration of 25 µM QUE with resveratrol (10 µM) was shown to induce a senescent- like growth arrest in human glioma cells [55]. However, the pro-senescence-like activity of this combination of compounds was also associated with cytotoxic activity. The inhibition of histone deacetylases (HDAC) in the glioma cellular models has been proposed as a mechanism to explain the cell cycle arrest [56]. Interesting, this inhibitory activity on HDAC was not observed in normal astrocytes. QUE was also shown to activate and stabilize p53 by inhibiting its RNA degradation and protein ubiquitination in HepG2 cells, thus promoting p21 expression and cyclin D1 suppression in favour of cell cycle arrest and promotion of apoptosis [57]. Hence, circumstances where p53 is not stabilized or where HDAC is over-activated pave the way to a potential use of QUE to induce a senescent-like cell cycle arrest and apoptosis in cancer.

Common cellular targets of quercetin and tocotrienols

QUE and T3s are structurally different but their original role in nature is to protect the plant from the aggression of pathogens. In particular, QUE and T3s are chemical weapon built for the defense of the fruit and the seeds, respectively. First of all, both compounds have been claimed of antioxidant activity which, in biological systems can be mediated by Nrf2-Keap1 stress responsive signaling [58,59]. Although the effects could depend on the cell nature [60] and time of exposure [61], several cell culture studies suggest that NRF2- Keap1 pathway can be targeted with both T3s [32] and QUE [44,61,62]. Therefore, their ability to activate the antioxidant response elements (AREs), via Nrf2-Keap1, may represent a common mechanism of action at least in certain conditions or cellular models. Another interesting similarity regards the mechanisms of interference a variety of growth factor receptor kinases. Epidermal growth factor receptor (EGFR) is an important player in the onset and progression of a various cancers, including breast cancer. A crucial signalling pathway downstream of EGFR is the PI3K/AKt pathway, which regulates cellular growth, survival, proliferation and migration. QUE is known since long time as an inhibitor PI3K [63] and this activity has been proposed as a mechanisms to explain its anti-proliferative and pro-apoptotic effects in cancer cells [64-66]. Similarly, T3s have been shown to inhibit PI3K/Akt signaling and to exert anti-proliferative and pro-apoptotic effects in neoplastic mammary epithelial cells [67,68]. While the inhibitory activity of QUE appear to be mediated by the CB1 receptor [69], the effects of T3s are likely mediated by the suppression of ErbB-receptor tyrosine phosphorylation [36,70], but the downstream effects appears to be similar. Another crucial activator of pro- proliferative and pro-survival genes (i.e. c-myc and cyclinD1) as well as anti-apoptotic genes (i.e. bcl2, bclxl, or mcl1) in certain cancers is STAT3 signalling. QUE shows STAT3 signalling inhibitory activity in a multitude of cancer models [71-73] while T3s, in particular γ-T3s, have been identified as potent inhibitors of STAT3 pathways in a model of hepatocellular carcinoma [74] as well as in multiple myeloma cell lines [75]. Interestingly, also endoplasmic reticulum stress has been frequently associated with the pro-apoptotic effects of QUE [76,77] or T3s [34,78] in various cancer cells.

We have also attempted a meta-analysis of gene expression data of T3s and QUE on cancer cells. We have searched for published data using 10-40 µM QUE (a widely used range of concentration used in vitro) and similar concentrations of T3s on cell lines that were obtained with the same platform using the same probesets. We identified four studies performed in four different cell lines that were compatible with our research criteria. Transcriptomic data have been retrieved from these experiments that have been deposited in GEO. An integrated meta-analysis using the original *.CEL files was performed with the Partek Genomic Suite Software (Partek). The data have been deposited as four GEO DataSets: 1) GSE21946 [78], consisting of 4 replicates of controls and treatments (Affymetrix U133A 2.0 Array) using γ-T3s (40 µM for 24 h) in MCF-7 cells; 2) GSE7259 [79], consisting of 4 replicates of controls and treatments (Affymetrix U133A 2.0 Array) using ascorbate-stabilized QUE (40 µM for 24 h) in Caco-2 cells; 3) GSE59368 [80], from where we extracted the data of the 6 replicates of controls and treatments (Affymetrix HT HG-U133 Array) using QUE (40 µM for 24 h) in HT-1080 cells; 4) GSE48668 [81], from where we extracted only the data of the 2 replicates for controls and the 3 replicates for treatments (NuGO array manufactured by Affymetrix) using γ-T3s (10 µM for 24 h) in HeLa cells. Albeit the experiments were performed in different cellular models, the meta-analysis could be useful as preliminary analysis to confirm or identify common pathways.

When we searched for common probesets significantly upregulated (p < 0.05) and downregulated (p < 0.05) in all datasets we identified 8 common downregulated genes (Fig1A) and 21 common upregulated genes (Fig 1B). Additional 10 common upregulated genes and 5 downregulated genes where identified (p < 0.05 for at least one probeset in all studies) when we searched for common “gene symbols”, thus suggesting that these data refers to genes with multiple probesets not identically modulated in the different studies.

Many of these genes (Table 1) are involved in endoplasmic reticulum stress (C19orf10, CEBPB, GFPT1, KDELR3, PPP1R15A, SDF2L1, SEC24A, SEC61A1), apoptosis (PMAIP1, TNFRSF21, PHLDA1, NFKBIB) and metabolism of protein, lipids or carbohydrate (CEBPB, GFPT1, KLF4, FTH1, MED8, NAMPT1, ADH5, ECHS1, ECI2, SRR, IVD, DBI). Additional pathways and molecular function of interest for the activity of QUE and T3s that were identified in this meta-analysis include the regulation of cytosolic calcium ion concentration (CD55, GEM), negative regulation of MAPK/ERK (DUSP4), trace elements traffic (FTH1 and SLC31A1), NAD metabolism (NAMPT), cell cycle and DNA replication (RNASEH1, GORASP2). We also performed gene set analysis by GeneAnalytics (Gene Set Analysis, https://ga.genecards.org ) and identified additional genes involved in the activity of QUE and T3s by string interaction network. If we repeated the analysis in GeneAnalytics including these genes, QUE and T3s emerged as compounds that matched the gene set with a good score 27.49 and 15.40, respectively. The genes that characterized the activity of QUE were BAX, BCL2, BCL2L1, CFTR, CHUK, HSPA4, MAPK1, MAPK14, MAPK3, MAPK8, NFE2L2 (alias Nrf2), SP1 and TGFA, while these characterizing the activity of T3s were BAX, BCL2L1 and MYC. Also in this case QUE and T3s appears to involve a common network of genes involved in apoptosis and mediated by BAX and BCL2L1.

The results of this meta-analysis are also consistent with the finding of the original datasets submissions.

The finding that T3s activity can be ascribable to the induction of a cellular stress at the level of the endoplasmic reticulum and that this event can lead to induction of apopotosis in cancer cells was already observed in the publications originated from GSE48668 [82] and GSE21946 [78], as well as also in other experimental settings [83]. Regarding the findings around QUE, GSE7259 focused specifically on the p53- mediated DNA damage response [80]. However, among the p53 regulated genes that were identified in this work, ARFGAP3, C19orf10 and MAFF were also identified in our list and are also known to be involved in the processes of endocytosis, endoplasmic reticulum stress and Nrf2 transcriptional response, respectively. GSE7259 used an ascorbate stabilized form of QUE and the results were mostly contrary to what is expected for a cancer preventive agent. The changes mediated by ascorbate-stabilized QUE were concordant with those occurring in human colorectal carcinogenesis, but were opposite to those previously described for Caco-2 cells exposed to QUE in the absence of ascorbate. However, our study has identified common targets of QUE that are likely to be independently modulated by the production of free radicals (which are likely to be suppressed by ascorbate). The involvement of endoplasmic reticulum stress in the action of QUE in cancer cells was recently confirmed by other independent studies [76,77].

These area of overlap between QUE and T3s could explain the similitude in the cellular outcomes shown by both compounds and might propose a rationale to explain the pro-senescence and rejuvenating effects observed in cancer and senescent cells, respectively.

Revising the pro-senescence and rejuvenating effects of tocotrienol and quercetin in cellular senescence

The present revision of literature around selected effects of T3s and QUE point out the involvement of endoplasmic reticulum stress, Nrf2, EGFR, PI3K/Akt, STAT3 pathways, as well as apoptotic pathways mediated by BAX and BCL2L1. Although neither BAX and BCL2L1 were identified directly in our meta-analysis, they are included in the network of genes regulating apoptosis that we have identified by string interaction network in GeneAnalytics. It seems important thus to focus on the relationship between this apoptotic pathway and ER stress. ER stress can be regarded as a common mediator of the pro-apoptotic and pro-senescent effects of these compounds. Indeed, endoplasmic reticulum (ER) stress may promote a senescent-like response [84], a SASP-like inflammatory response [85] and, ultimately, cell death [86]. This sequence of events is the result of the stress signals activated by unfolded proteins, which involves attenuation of protein synthesis, higher capacity of ER protein traffic, folding and transport, increased pathways for protein degradation and autophagy [87]. These adaptive mechanisms aims at promoting survival both in cancer and normal cells. However, if the cell cannot resolve the protein-folding defect, or if one of these mechanisms is defective, cells enter apoptosis. It has been recently found that BCL2L1 can protect from a C/EBP homologous protein (CHOP)-dependent apoptosis in the presence of unfolded protein response [88]. Hence, in presence of ER stress induced by QUE and T3s, it could be hypothesized that a particular cancer epigenome (in contrast to normal cells) can favour to the dissociation of BCL2L1 from Bax. Bax can subsequently translocate and multimerize to the mitochondrial membrane with the release of cytochrome c and activation of the apoptotic cascade. Taking into account the similitude in the epigenome of cancer and senescent cells it is not excluded that this hypothesis can be valid also in the case of senescent cells.

Other defects of cancer cells can also determinate cell death in response to ER stress. In cancer cells, autophagy has a dual role, acting as a tumor suppressor or as a mechanism of cell survival. However, defects in autophagy mechanisms as well as in other players of ER pathways have been frequently detected, especially in human breast cancers that overexpress ERBB2 [89]. Also the role of autophagy in senescence is still unclear. Autophagy impairment can induce senescence of primary human fibroblasts [90] and defects in autophagy have been detected in various models of cellular senescence [91]. But, more recently, it has been documented that autophagy is required in senescent cells to mitigate the proteotoxic stress induced by the high protein synthesis rate involved in the SASP and in energy production [92]. Hence, T3s and QUE, might potentially be able to induce cell death in cells with a defective mechanism of autophagy. However, their pro-apoptotic activity can potentially be expanded to cells with non-defective autophagy via their potential interference with the EGFR, PI3K/Akt and STAT3 pathways. Unfortunately, these effects are likely to be dependent on cell type and experimental conditions.

Anyway, if the most likely outcome of treatments of cancer and senescent cells with QUE and T3s is cell death, how can we justify the rejuvenating effects of T3s and QUE in pre-senescent and senescent human cells? Here we propose a new perspective to approach the problem that starts from the consideration of the heterogeneity of cell cultures. For instance, when cells approach senescence (and maybe even later) not all cells are marked by senescent markers. Hence, even if the large majority of the cells in culture consists of senescent cells, it still remains a population of “younger” cells. In this condition, the SASP produced by the senescent cells in culture should in brief promote senescence or inhibit proliferation in the whole cell culture, including the sub-population of non-senescent cells. However, in the presence of a tool that selectively kills senescent cells, the non-senescent sub-population could be free to proliferate and expand again due to the diminished SASP inhibitory effect. The overall macroscopic effect is that the pre-senescent cell population restart to growth. Staining and measurement of senescence markers should also be diminished thus resembling a real rejuvenating effect. If this hypothesis will be experimentally confirmed, the rejuvenating effects of Que and T3s in pre-senescent culture could be the consequence of a selective removal of senescent cells and a selective survival advantage for sub-population of non-senescent cells. This concept could also be extended to culture of normal cells were the Nrf2 mediated response might favour the selection of “younger” cells resistant to the treatment with T3s and QUE at the expenses of damaged and pre-senescent cells.

Clinical perspectives Despite the number of clinical trials conducted to examine the multi-faceted health benefits of T3s [93], very little is known about the efficacy of T3s as adjuvant supplements in cancer therapy. A pilot clinical trial on the synergistic effect of T3s and tamoxifen in women failed to show a significant impact on the survival of patients compared to tamoxifen alone. However, measurements of T3s in malign and benign adipose breast tissues of a Malaysian population found that total T3s levels were lower in the malignant tissues compared to the benign [94]. These data reinforce the idea that T3s may provide some kind of protection against breast cancer but the circumstances and modality of intervention would require further studies.

The trials performed with Que also deserve appropriate consideration for their interestingly results [95,96], especially in decreasing cancer related biomarkers in various kind of cancer [96]. These data might be consistent with the interpretation that Que might have selectively removed senescent cancer cells previously induced by therapy. This consideration might be important in the case cancer cells may use senescence as an escape strategy from therapy. If this hypothesis will be confirmed, new therapeutic schemes using QUE or T3s after chemo- or radiotherapy can be proposed without the needs for escalation doses and continuous treatments. Up to now there is no evidence of an approach using for example a single “senolytic” dose of such bioactive natural compound after drug or radiation treatments in cancer.

Conclusions

QUE and T3s are among the most promising phytochemicals currently used or proposed as complementary and alternative medications for the prevention and treatment of cancers [97]. The use of these substances as antitumor agents is an attractive idea because they are readily available and likely exhibit little or no toxicity. Moreover, these compounds may act both in the prevention and therapy of cancer, inducing apoptosis in damaged and eventually senescent cells. Emerging evidence has demonstrated that therapy-induced senescence is a critical mechanism through which many anticancer agents inhibit the growth of tumor cells [98]. Interestingly, it has been shown that therapy-induced senescence can be achieved at much lower doses of chemotherapy with the potential to significantly reduce the side effects of anticancer therapy and thus improve the quality of life for cancer patients. Several results obtained “in vitro” and, in some cases, in experimental animal models, have provided evidence that lots of natural bioactive compounds are able to promote senescence in cancer cells, but the meaning and the potential of these results as adjuvant therapy in humans remain unclear. Moreover, induction of CS does not ensure that these cells can be cleared off by the immune system and they might eventually use senescence as a system to escape death with the potential to revert their phenotype later in time. This problem might be potentially avoided with the use of single doses of QUE and T3s as adjuvant in the cancer therapy. Clearing off from the organism accumulating senescent cells appears also a promising strategy to prevent or to diminish the burden of age-related diseases. However, appropriate experiments in vitro and in animal models need to be conducted in order to verify these hypothesis and open the perspective to a new therapeutic scheme in adjuvant treatments for cancer and in preventive medicine.

Acknowledgments

The study has been partially funded through the revenues of the “5 per Mille” donations received by INRCA through the Italian Ministry of Health. The authors declare that there are no conflicts of interest.

References

1. Provinciali M, Cardelli M, Marchegiani F, Pierpaoli E. Impact of cellular senescence in aging and cancer. Curr Pharm Des. 2013;19:1699-709.

2. Malavolta M, Costarelli L, Giacconi R, Piacenza F, Basso A, Pierpaoli E, Marchegiani F, cardelli M, Provinciali M, Mocchegiani E. Modulators of cellular senescence: mechanisms, promises, and challenges from in vitro studies with dietary bioactive compounds. Nutr Res. 2014; 34: 1017-35.

3. van Deursen JM. The role of senescent cells in ageing. Nature. 2014;509:439-46.

4. Campisi J. Aging, cellular senescence,and cancer. Annu. Rev.Physiol. 2013;75:685–705

5. López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013;153:1194- 217

6. Sulli G, Di Micco R, d'Adda di Fagagna F. Crosstalk between chromatin state and DNA damage response in cellular senescence and cancer. Nat Rev Cancer 2012;12:709–20

7. Thanasoula M, Escandell JM, Martinez P, Badie S, Muñoz P, Blasco MA, Tarsounas M. p53 prevents entry into mitosis with uncapped telomeres. Curr Biol. 2010;20:521-6

8. Kuilman T, Michaloglou C, Mooi WJ, Peeper DS. The essence of senescence. Genes Dev 2010;24:2463–79

9. Krishnamurthy J, Torrice C, Ramsey MR, Kovalev GI, Al-Regaiey K, Su L, Sharpless NE. Ink4a/Arf expression is a biomarker of aging. J Clin Invest. 2004;114:1299–307

10. Naylor RM, Baker DJ, van Deursen JM. Senescent cells: a novel therapeutic target for aging and age-related diseases. Clin Pharmacol Ther. 2013;93:105-16

11. Collado M, Blasco MA, Serrano M. Cellular senescence in cancer and aging. Cell 2007;130:223–33

12. Blagosklonny MV. Hypoxia, MTOR and autophagy: converging on senescence or quiescence. Autophagy. 2013;9:260-2.

13. de Jesus BB, Blasco MA. Assessing cell and organ senescence biomarkers. Circ Res. 2012;111:97-109.

14. Coppé JP, Patil CK, Rodier F, Krtolica A, Beauséjour CM, Parrinello S, Hodgson JG, Chin K, Desprez PY, Campisi J. A human-like senescence-associated secretory phenotype is conserved in mouse cells dependent on physiological oxygen. PLoS One. 2010;5:e9188

15. Coppé JP, Patil CK, Rodier F, Sun Y, Muñoz DP, Goldstein J, et al. Senescence-associated secretory phenotypes reveal cell-nonautonomous functions of oncogenic RAS and the p53 tumor suppressor. PLoS Biol 2008;6:2853–68

16. Acosta JC, Banito A, Wuestefeld T, Georgilis A, Janich P, Morton JP, et al. A complex secretory program orchestrated by the inflammasome controls paracrine senescence. Nat Cell Biol 2013;15:978–90 17. Hampel B, Wagner M, Teis D, Zwerschke W, Huber LA, Jansen-Dürr P. Apoptosis resistance of senescent human fibroblasts is correlated with the absence of nuclear IGFBP-3. Aging Cell 2005;4:325–30

18. Baker DJ, Wijshake T, Tchkonia T, LeBrasseur NK, Childs BG, van de Sluis B, et al. Clearance of p16Ink4a- positive senescent cells delays ageing-associated disorders. Nature 2011;479:232–6

19. Cruickshanks HA, McBryan T, Nelson DM, Vanderkraats ND, Shah PP, van Tuyn J, Singh Rai T, Brock C, Donahue G, Dunican DS, Drotar ME, Meehan RR, Edwards JR, Berger SL, Adams PD. Senescent cells harbour features of the cancer epigenome. Nat Cell Biol. 2013;15:1495-506

20. Cardelli M, Marchegiani F. Good, bad, mobile elements: genome's most successful "parasites" as emerging players in cell and organismal aging. Curr Pharm Des 2013;19:1739-52

21. Ade C, Roy-Engel AM, Deininger PL. Alu elements: an intrinsic source of human genome instability. Curr Opin Virol 2013;3:639-45

22. Sedivy JM, Kreiling JA, Neretti N, De Cecco M, Criscione SW, Hofmann JW, Zhao X, Ito T, Peterson AL. Death by transposition - the enemy within? Bioessays. 2013;35:1035-43

23. De Cecco M, Criscione SW, Peckham EJ, et al. Genomes of replicatively senescent cells undergo global epigenetic changes leading to gene silencing and activation of transposable elements. Aging Cell 2013;12:247-56

24. Zhu Y, Tchkonia T, Pirtskhalava T, Gower AC, Ding H, Giorgadze N, Palmer AK, Ikeno Y, Hubbard GB, Lenburg M, O'Hara SP, LaRusso NF, Miller JD, Roos CM, Verzosa GC, LeBrasseur NK, Wren JD, Farr JN, Khosla S, Stout MB, McGowan SJ, Fuhrmann-Stroissnigg H, Gurkar AU, Zhao J, Colangelo D, Dorronsoro A, Ling YY, Barghouthy AS, Navarro DC, Sano T, Robbins PD, Niedernhofer LJ, Kirkland JL. The Achilles' heel of senescent cells: from transcriptome to senolytic drugs. Aging Cell. 2015 Mar 9. doi: 10.1111/acel.12344

25. Horvath G, Wessjohann L, Bigirimana J, Jansen M, Guisez Y, Caubergs R, Horemans N. Differential distribution of tocopherols and tocotrienols in photosynthetic and non-photosynthetic tissues. Phytochemistry. 2006;67:1185-95.

26. Frega N, Mozzon M, Bocci F. Identification and Estimation of Tocotrienols in the Annatto Lipid Fraction by Gas Chromatography–Mass Spectrometry. JAOCS 1998;75:1723–7

27. Makpol S, Durani LW, Chua KH, Mohd Yusof YA, Ngah WZ. Tocotrienol-rich fraction prevents cell cycle arrest and elongates telomere length in senescent human diploid fibroblasts. J Biomed Biotechnol. 2011;2011:506171

28. McIntyre BS, Briski KP, Gapor A, Sylvester PW. Antiproliferative and apoptotic effects of tocopherols and tocotrienols on preneoplastic and neoplastic mouse mammary epithelial cells. Proc Soc Exp Biol Med. 2000;224:292-301.

29. Wilankar C, Khan NM, Checker R, Sharma D, Patwardhan R, Gota V, Sandur SK, Devasagayam TP. γ- Tocotrienol induces apoptosis in human T cell lymphoma through activation of both intrinsic and extrinsic pathways. Curr Pharm Des. 2011;17(21):2176-89; 30. Ahn KS, Sethi G, Krishnan K, Aggarwal BB. Gamma-tocotrienol inhibits nuclear factor-kappaB signaling pathway through inhibition of receptor-interacting protein and TAK1 leading to suppression of antiapoptotic gene products and potentiation of apoptosis. J Biol Chem. 2007;282:809-20

31. Sylvester PW, Ayoub NM. Tocotrienols target PI3K/Akt signaling in anti-breast cancer therapy. Anticancer Agents Med Chem. 2013;13:1039-47

32. Kannappan R, Gupta SC, Kim JH, Aggarwal BB. Tocotrienols fight cancer by targeting multiple cell signaling pathways. Genes Nutr. 2012;7:43-52

33. Pierpaoli E, Viola V, Pilolli F, Piroddi M, Galli F, Provinciali M. Gamma- and delta-tocotrienols exert a more potent anticancer effect than alpha-tocopheryl succinate on breast cancer cell lines irrespective of HER- 2/neu expression. Life Sci. 2010;86:668-75.

34. Park SK, Sanders BG, Kline K. Tocotrienols induce apoptosis in breast cancer cell lines via an endoplasmic reticulum stress-dependent increase in extrinsic death receptor signaling. Breast Cancer Res Treat. 2010;124:361-75

35. Zhang JS, Li DM, Ma Y, He N, Gu Q, Wang FS, Jiang SQ, Chen BQ, Liu JR. γ-Tocotrienol induces paraptosis-like cell death in human colon carcinoma SW620 cells. PLoS One. 2013;8:e57779.

36. Samant GV, Sylvester PW. gamma-Tocotrienol inhibits ErbB3-dependent PI3K/Akt mitogenic signalling in neoplastic mammary epithelial cells. Cell Prolif. 2006;39:563-74.

37. Selvaduray KR, Radhakrishnan AK, Kutty MK, Nesaretnam K. Palm tocotrienols decrease levels of pro- angiogenic markers in human umbilical vein endothelial cells (HUVEC) and murine mammary cancer cells. Genes Nutr. 2012;7:53-61.

38. Samant GV, Wali VB, Sylvester PW. Anti-proliferative effects of gamma-tocotrienol on mammary tumour cells are associated with suppression of cell cycle progression. Cell Prolif. 2010;43:77-83.

39. Akl MR, Ayoub NM, Abuasal BS, Kaddoumi A, Sylvester PW. Sesamin synergistically potentiates the anticancer effects of γ-tocotrienol in mammary cancer cell lines. Fitoterapia. 2013;84:347-59.

40. Pierpaoli E, Viola V, Barucca A, Orlando F, Galli F, Provinciali M. Effect of annatto-tocotrienols supplementation on the development of mammary tumors in HER-2/neu transgenic mice. Carcinogenesis. 2013;34:1352-60

41. Berger C, Qian Y, Chen X. The p53-estrogen receptor loop in cancer. Curr Mol Med. 2013;13:1229-40

42. Nesaretnam K, Meganathan P, Veerasenan SD, Selvaduray KR. Tocotrienols and breast cancer: the evidence to date. Genes Nutr. 2012;7:3-9

43. Chondrogianni N, Kapeta S, Chinou I, Vassilatou K, Papassideri I, Gonos ES. Anti-ageing and rejuvenating effects of quercetin. Exp Gerontol. 2010;45:763-71

44. Murakami A, Ashida H, Terao J. Multitargeted cancer prevention by quercetin. Cancer Lett. 2008;269:315- 25 45. Braganhol E, Zamin LL, Canedo AD, Horn F, Tamajusuku AS, Wink MR, Salbego C, Battastini AM. Antiproliferative effect of quercetin in the human U138MG glioma cell line. Anticancer Drugs. 2006;17:663-71

46. Xie X, Yin J, Jia Q, Wang J, Zou C, Brewer KJ, Colombo C, Wang Y, Huang G, Shen J. Quercetin induces apoptosis in the methotrexate-resistant osteosarcoma cell line U2-OS/MTX300 via mitochondrial dysfunction and dephosphorylation of Akt. Oncol Rep. 2011;26:687-93

47. Vidya Priyadarsini R, Senthil Murugan R, Maitreyi S, Ramalingam K, Karunagaran D, Nagini S. The flavonoid quercetin induces cell cycle arrest and mitochondria-mediated apoptosis in human cervical cancer (HeLa) cells through p53 induction and NF-κB inhibition. Eur J Pharmacol. 2010;649:84-91

48. Hsieh TC, Wu JM. Targeting CWR22Rv1 prostate cancer cell proliferation and gene expression by combinations of the phytochemicals EGCG, genistein and quercetin. Anticancer Res. 2009;29:4025-32

49. Duo J, Ying GG, Wang GW, Zhang L. Quercetin inhibits human breast cancer cell proliferation and induces apoptosis via Bcl-2 and Bax regulation. Mol Med Rep. 2012;5:1453-6

50. Choi EJ, Bae SM, Ahn WS. Antiproliferative effects of quercetin through cell cycle arrest and apoptosis in human breast cancer MDA-MB-453 cells. Arch Pharm Res. 2008;31:1281-5

51. Priego S, Feddi F, Ferrer P, Mena S, Benlloch M, Ortega A, Carretero J, Obrador E, Asensi M, Estrela JM. Natural polyphenols facilitate elimination of HT-29 colorectal cancer xenografts by chemoradiotherapy: a Bcl-2- and superoxide dismutase 2-dependent mechanism. Mol Cancer Ther. 2008;7:3330-42

52. Duraj J, Zazrivcova K, Bodo J, Sulikova M, Sedlak J. Flavonoid quercetin, but not apigenin or luteolin, induced apoptosis in human myeloid leukemia cells and their resistant variants. Neoplasma. 2005;52:273- 9

53. Kang JW, Kim JH, Song K, Kim SH, Yoon JH, Kim KS. Kaempferol and quercetin, components of Ginkgo biloba extract (EGb 761), induce caspase-3-dependent apoptosis in oral cavity cancer cells. Phytother Res. 2010;24 Suppl 1:S77-82

54. Lamson DW, Brignall MS. Antioxidants and cancer, part 3: Quercetin. Altern Med Rev. 2000;5:196–208

55. Zamin LL, Filippi-Chiela EC, Dillenburg-Pilla P, Horn F, Salbego C, Lenz G. Resveratrol and quercetin cooperate to induce senescence-like growth arrest in C6 rat glioma cells. Cancer Sci. 2009;100:1655-62

56. Vargas JE, Filippi-Chiela EC, Suhre T, Kipper FC, Bonatto D, Lenz G. Inhibition of HDAC increases the senescence induced by natural polyphenols in glioma cells. Biochem Cell Biol. 2014;92:297-304

57. Tanigawa S, Fujii M, Hou DX. Stabilization of p53 is involved in quercetin-induced cell cycle arrest and apoptosis in HepG2 cells. Biosci Biotechnol Biochem. 2008;72:797-804

58. Giudice A, Arra C, Turco MC. Review of molecular mechanisms involved in the activation of the Nrf2-ARE signaling pathway by chemopreventive agents. Methods Mol Biol. 2010;647:37-74 59. Costa G, Francisco V, Lopes MC, Cruz MT, Batista MT. Intracellular signaling pathways modulated by phenolic compounds: application for new anti-inflammatory drugs discovery. Curr Med Chem. 2012;19:2876-900

60. Hsieh TC, Elangovan S, Wu JM. Differential suppression of proliferation in MCF-7 and MDA-MB-231 breast cancer cells exposed to alpha-, gamma- and delta-tocotrienols is accompanied by altered expression of oxidative stress modulatory enzymes. Anticancer Res. 2010;30:4169-76

61. Granado-Serrano AB, Martín MA, Bravo L, Goya L, Ramos S. Quercetin modulates Nrf2 and glutathione- related defenses in HepG2 cells: Involvement of p38. Chem Biol Interact. 2012;195:154-64

62. Lee YJ, Lee DM, Lee SH. Nrf2 Expression and Apoptosis in Quercetin-treated Malignant Mesothelioma Cells. Mol Cells. 2015;38:416-25.

63. Walker EH, Pacold ME, Perisic O, Stephens L, Hawkins PT, Wymann MP, Williams RL. Structural determinants of phosphoinositide 3-kinase inhibition by wortmannin, LY294002, quercetin, myricetin, and staurosporine. Mol Cell. 2000;6:909-19

64. Gulati N, Laudet B, Zohrabian VM, Murali R, Jhanwar-Uniyal M. The antiproliferative effect of Quercetin in cancer cells is mediated via inhibition of the PI3K-Akt/PKB pathway. Anticancer Res. 2006;26:1177-81

65. Yuan Z, Long C, Junming T, Qihuan L, Youshun Z, Chan Z. Quercetin-induced apoptosis of HL-60 cells by reducing PI3K/Akt. Mol Biol Rep. 2012;39(7):7785-93

66. Sun ZJ, Chen G, Hu X, Zhang W, Liu Y, Zhu LX, Zhou Q, Zhao YF. Activation of PI3K/Akt/IKK-alpha/NF-kappaB signaling pathway is required for the apoptosis-evasion in human salivary adenoid cystic carcinoma: its inhibition by quercetin. Apoptosis. 2010;15:850-63

67. Sylvester PW, Shah SJ, Samant GV. Intracellular signaling mechanisms mediating the antiproliferative and apoptotic effects of gamma-tocotrienol in neoplastic mammary epithelial cells. J Plant Physiol. 2005;162:803–10

68. Shah SJ, Sylvester PW. Gamma-tocotrienol inhibits neoplastic mammary epithelial cell proliferation by decreasing Akt and nuclear factor kappaB activity. Exp Biol Med (Maywood) 2005;230:235–41

69. Refolo MG, D'Alessandro R, Malerba N, Laezza C, Bifulco M, Messa C, Caruso MG, Notarnicola M, Tutino V. Anti Proliferative and Pro Apoptotic Effects of Flavonoid Quercetin Are Mediated by CB1 Receptor in Human Colon Cancer Cell Lines. J Cell Physiol. 2015;doi: 10.1002/jcp.25026 [Epub ahead of print]

70. Comitato R, Leoni G, Canali R, Ambra R, Nesaretnam K, Virgili F. Tocotrienols activity in MCF-7 breast cancer cells: involvement of ERbeta signal transduction. Mol Nutr Food Res. 2010;54:669-78

71. Mukherjee A, Khuda-Bukhsh AR. Quercetin Down-regulates IL-6/STAT-3 Signals to Induce Mitochondrial- mediated Apoptosis in a Nonsmall- cell Lung-cancer Cell Line, A549. J Pharmacopuncture. 2015;18:19-26

72. Cao HH, Tse AK, Kwan HY, Yu H, Cheng CY, Su T, Fong WF, Yu ZL. Quercetin exerts anti-melanoma activities and inhibits STAT3 signaling. Biochem Pharmacol. 2014;87:424-34 73. Michaud-Levesque J, Bousquet-Gagnon N, Béliveau R. Quercetin abrogates IL-6/STAT3 signaling and inhibits glioblastoma cell line growth and migration. Exp Cell Res. 2012;318:925-35

74. Rajendran P, Li F, Manu KA, Shanmugam MK, Loo SY, Kumar AP, Sethi G. γ-Tocotrienol is a novel inhibitor of constitutive and inducible STAT3 signalling pathway in human hepatocellular carcinoma: potential role as an antiproliferative, pro-apoptotic and chemosensitizing agent. Br J Pharmacol. 2011;163:283-98

75. Kannappan R, Yadav VR, Aggarwal BB. {Gamma}-tocotrienol but not {gamma}-tocopherol blocks STAT3 cell signaling pathway through induction of protein-tyrosine phosphatase SHP-1 and sensitizes tumor cells to chemotherapeutic agents. J Biol Chem. 2010;285:33520–8

76. Yang Z, Liu Y, Liao J, Gong C, Sun C, Zhou X, Wei X, Zhang T, Gao Q, Ma D, Chen G. Quercetin induces endoplasmic reticulum stress to enhance cDDP cytotoxicity in ovarian cancer: involvement of STAT3 signaling. FEBS J. 2015;282:1111-25

77. Liu KC, Yen CY, Wu RS, Yang JS, Lu HF, Lu KW, Lo C, Chen HY, Tang NY, Wu CC, Chung JG. The roles of endoplasmic reticulum stress and mitochondrial apoptotic signaling pathway in quercetin-mediated cell death of human prostate cancer PC-3 cells. Environ Toxicol. 2014;29:428-39

78. Patacsil D, Tran AT, Cho YS, Suy S, Saenz F, Malyukova I, Ressom H, Collins SP, Clarke R, Kumar D. Gamma- tocotrienol induced apoptosis is associated with unfolded protein response in human breast cancer cells. J Nutr Biochem. 2012;23:93-100

79. Dihal AA, Tilburgs C, van Erk MJ, Rietjens IM, Woutersen RA, Stierum RH. Pathway and single gene analyses of inhibited Caco-2 differentiation by ascorbate-stabilized quercetin suggest enhancement of cellular processes associated with development of colon cancer. Mol Nutr Food Res. 2007;51:1031-45

80. Clewell RA, Sun B, Adeleye Y, Carmichael P, Efremenko A, McMullen PD, Pendse S, Trask OJ, White A, Andersen ME. Profiling dose-dependent activation of p53-mediated signaling pathways by chemicals with distinct mechanisms of DNA damage. Toxicol Sci. 2014;142:56-73

81. Comitato R, Leoni G, Virgili F. Expression data from HeLa cells treated with tocotrienols. http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE48668

82. Comitato R; Guantario B; Leoni G; Canali R; Virgili F. Endoplasmic reticulum stress contributes to tocotrienol induces opoptosis in HeLa cell. 11th NuGOweek Nutricenomics of foods 8-11 sept. 2014 The Vesuvian Insitute Castellammare di Stabia, Italy, Ed. NuGo Nutrigenomics Organisation

83. Wali VB, Bachawal SV, Sylvester PW. Endoplasmic reticulum stress mediates gamma-tocotrienol-induced apoptosis in mammary tumor cells. Apoptosis. 2009;14:1366–77

84. Bourougaa K, Naski N, Boularan C, Mlynarczyk C, Candeias MM, Marullo S, Fåhraeus R. Endoplasmic reticulum stress induces G2 cell-cycle arrest via mRNA translation of the p53 isoform p53/47. Mol Cell. 2010;38:78-88

85. Schröder M. Endoplasmic reticulum stress responses. Cell Mol Life Sci. 2008;65:862-94 86. Han J, Back SH, Hur J, Lin YH, Gildersleeve R, Shan J, Yuan CL, Krokowski D, Wang S, Hatzoglou M, Kilberg MS, Sartor MA, Kaufman RJ. ER-stress-induced transcriptional regulation increases protein synthesis leading to cell death. Nat Cell Biol. 2013;15:481-90

87. Wang M, Kaufman RJ. The impact of the endoplasmic reticulum protein-folding environment on cancer development. Nat Rev Cancer. 2014;14:581-97

88. Gaudette BT, Iwakoshi NN, Boise LH. Bcl-xL protein protects from C/EBP homologous protein (CHOP)- dependent apoptosis during plasma cell differentiation. J Biol Chem. 2014;289:23629-40

89. Lozy F, Cai-McRae X, Teplova I, Price S, Reddy A, Bhanot G, Ganesan S, Vazquez A, Karantza V. ERBB2 overexpression suppresses stress-induced autophagy and renders ERBB2-induced mammary tumorigenesis independent of monoallelic Becn1 loss. Autophagy. 2014;10:662-76

90. Kang HT, Lee KB, Kim SY, Choi HR, Park SC. Autophagy impairment induces premature senescence in primary human fibroblasts. PLoS One. 2011;6:e23367

91. Cuervo AM, Dice JF. Age-related decline in chaperone-mediated autophagy. J Biol Chem. 2000; 275:31505- 13

92. Dörr JR, Yu Y, Milanovic M, Beuster G, Zasada C, Däbritz JH, Lisec J, Lenze D, Gerhardt A, Schleicher K, Kratzat S, Purfürst B, Walenta S, Mueller-Klieser W, Gräler M, Hummel M, Keller U, Buck AK, Dörken B, Willmitzer L, Reimann M, Kempa S, Lee S, Schmitt CA. Synthetic lethal metabolic targeting of cellular senescence in cancer therapy. Nature. 2013;501:421-5

93. Fu JY, Che HL, Tan DM, Teng KT. Bioavailability of tocotrienols: evidence in human studies. Nutr Metab (Lond). 2014;11:5

94. Nesaretnam K, Gomez PA, Selvaduray KR, Razak GA. Tocotrienol levels in adipose tissue of benign and malignant breast lumps in patients in Malaysia. Asia Pac J Clin Nutr. 2007;16:498-504

95. Cruz-Correa M, Shoskes DA, Sanchez P, Zhao R, Hylind LM, Wexner SD, Giardiello FM. Combination treatment with curcumin and quercetin of adenomas in familial adenomatous polyposis. Clin Gastroenterol Hepatol. 2006;4:1035-8

96. Ferry DR, Smith A, Malkhandi J, Fyfe DW, deTakats PG, Anderson D, Baker J, Kerr DJ. Phase I clinical trial of the flavonoid quercetin: pharmacokinetics and evidence for in vivo tyrosine kinase inhibition. Clin Cancer Res. 1996;2:659-68

97. Gullett NP, Ruhul Amin AR, Bayraktar S, Pezzuto JM, Shin DM, et al. Cancer prevention with natural compounds. Semin Oncol. 2010;37:258–81

98. Gewirtz DA, Holt SE, Elmore LW. Accelerated senescence: an emerging role in tumor cell response to chemotherapy and radiation. Biochem Pharmacol. 2008;76:947-57

Table 1 Gene String interaction GSE72 GSE219 Gene Title Pathways GO- Biological Process GSE48668 GSE59368 Symbol network 59 46 Upregulated in all treatments identified by (Highest score for each gene) FC (p) FC (p) FC (p) FC (p) probeset ID COPB2, COPG, COPB1, ADP-ribosylation factor GTPase activating 1.17 2.15 1.27 1.42 ARFGAP3 PAK pathway Intracellular protein transport COPA, POLR3B, protein 3 (0.007) (0.001) (0.011) (0.002) ARFGAP3 Activation of signalling protein activity in UBC, UBQLN4, EIF6, 1.16 1.20 1.05 1.40 C19orf10 chromosome 19 open reading frame 10 Unfolded protein response unfolded protein response; Endoplasmic ATF2 (0.024) (0.010) (0.002) (0.000) reticulum unfolded protein response CCAAT/enhancer binding protein (C/EBP), Glucose / Energy SMAD4, DDIT3, ESR1, 1.14 1.88 1.15 1.61 CEBPB regulation of transcription beta Metabolism SMARCA4, SMARCA2 (0.038) (0.002) (0.030) (0.002) GTP binding protein overexpressed in skeletal cAMP / Ca2+ Signaling TRIM23, GMIP, PDLIM7, 1.67 3.78 1.39 1.32 GEM cell surface receptor signaling pathway muscle Pathway YWHAB, CACNB1 (0.028) (0.000) (0.002) (0.006) Endoplasmic Reticulum Unfolded Protein glutamine--fructose-6-phosphate Glucose/Energy Response, Activation of Signaling Protein PRKAA2, PRKAB2, 1.28 1.81 1.35 1.53 GFPT1 transaminase 1 metabolism Activity Involved in Unfolded Protein PRKCA, UBC, EPRS (0.003) (0.001) (0.017) (0.003) Response MAPK1, CFTR, BLZF1, 1.36 1.26 1.16 1.39 GORASP2 golgi reassembly stacking protein 2, 55kDa Cell cycle Mitotic cell cycle GOLGA2, TGFA (0.000) (0.016) (0.006) (0.006) Toll Comparative Pathway, STAT1, JAK2, IFNG, 1.42 1.61 1.13 1.72 IFNGR1 interferon gamma receptor 1 Interferon-gamma receptor activity IL-2 Pathway IGHA1, TRBC1 (0.000) (0.002) (0.016) (0.003) NOTCH3, NOTCH2, 1.81 1.42 1.30 1.51 JAG1 jagged 1 Notch signaling pathway Regulation of Cell Proliferation NOTCH1, UBC, NEURL (0.002) (0.038) (0.036) (0.000) Activation of signalling protein activity in KDEL (Lys-Asp-Glu-Leu) endoplasmic Unfolded Protein TUSC3, SSR2, STX5, 1.15 1.80 1.27 1.98 KDELR3 unfolded protein response; Endoplasmic reticulum protein retention receptor 3 Response DPM1, UQCR10 (0.026) (0.000) (0.004) (0.000) reticulum unfolded protein response Fatty Acid, Triacylglycerol, Regulation of Cell Proliferation; Negative SP1, CREBBP, EP300, 2.02 1.82 1.05 2.09 KLF4 Kruppel-like factor 4 (gut) and Ketone Body Regulation of Chemokine (C-X-C Motif) HDAC2, HDAC5 (0.011) (0.001) (0.033) (0.001) Metabolism Ligand 2 Production Tacrolimus/Cyclosporine v-maf avian musculoaponeurotic Regulation of Epidermal Cell NRF1, NFE2L2, NFE2, 2.24 5.36 1.56 2.03 MAFF Pathway, fibrosarcoma oncogene homolog F Differentiation NFE2L1, HOXD12 (0.003) (0.001) (0.022) (0.001) Pharmacodynamics Tacrolimus/Cyclosporine Regulation of Epidermal Cell v-maf avian musculoaponeurotic MAFF, NFE2L2, CREBBP, 1.16 1.37 1.15 1.50 MAFG Pathway, Differentiation; Regulation of Cell fibrosarcoma oncogene homolog G NFE2, NFE2L1 (0.028) (0.014) (0.011) (0.002) Pharmacodynamics Proliferation phorbol-12-myristate-13-acetate-induced Negative Regulation of Fibroblast BCL2L1, BAX, BCL2, 2.26 1.64 1.17 1.83 PMAIP1 Apoptosis protein 1 Proliferation; MCL1, HUWE1 (0.024) (0.011) (0.026) (0.002) EIF2S1, PPP1CC, protein phosphatase 1, regulatory subunit Protein Processing in Endoplasmic Reticulum Unfolded Protein 1.59 2.75 1.25 2.51 PPP1R15A PPP1CA, SMARCB1, 15A Endoplasmic Reticulum Response (0.042) (0.004) (0.034) (0.001) PPP1CB Development ERBB-family PLXNB1, CIT, ARHGAP5, 1.36 1.73 1.09 2.80 RND3 Rho family GTPase 3 GTP catabolic process Signaling, IL-2 Pathway UBXN11, ROCK1 (0.022) (0.001) (0.034) (0.000) RRP1B, UBC, MYC, 1.42 1.63 1.15 1.34 SDF2L1 stromal cell-derived factor 2-like 1 - KNCN, TERF2 (0.001) (0.012) (0.008) (0.036) RPN1, MCM3AP, Protein Processing in Membrane Organization; COPII Vesicle 1.11 1.33 1.15 1.17 SEC13 SEC13 homolog (S. cerevisiae) NUP107, SEH1L, SEC31A, Endoplasmic Reticulum Coating (0.002) (0.005) (0.026) (0.020) SEH1L SEC24A SEC24 family member A Protein Processing in Membrane Organization; COPII Vesicle TMED2, SEC23A, SEC13, 1.25 1.96 1.27 2.06 Endoplasmic Reticulum Coating SEC22B, STX5 (0.008) (0.001) (0.027) (0.001) Protein Processing in Antigen processing and presentation of APOB, SEC61B, SERP1, 1.14 1.56 1.16 1.48 SEC61A1 Sec61 alpha 1 subunit (S. cerevisiae) Endoplasmic Reticulum peptide antigen via MHC class I USP11, USP19 (0.012) (0.002) (0.001) (0.000) transmembrane emp24 protein transport SURF4, TMED10, 1.16 1.15 1.18 1.30 TMED9 Golgi organization COPI Coating of Golgi Vesicle domain containing 9 TMED7, TMED2, ATL1 (0.003) (0.003) (0.001) (0.001) Oligodendrocyte Apoptotic Process; tumor necrosis factor receptor superfamily, TRADD, APP, WASH1, 1.75 1.49 1.11 1.28 TNFRSF21 Apoptosis Negative Regulation of Interleukin-13 member 21 MATR3, STX11 (0.000) (0.005) (0.039) (0.000) Secretion Additional genes upregulated in all

treatments identified by “gene symbols” Complement Activation CD4-positive, Alpha-beta T Cell Cytokine CD55 molecule, decay accelerating factor for CD97, GPLD1, LCK, FYN, CD55 Pathways; Signaling by Production; positive regulation of complement (Cromer blood group) CR1 GPCR cytosolic calcium ion concentration Clathrin Derived Vesicle Membrane Organization; COPI Vesicle COPE, COPB1, COPB2, COPA coatomer protein complex, subunit alpha Budding; Delta508-CFTR Coating COPG, COPZ1 traffic/ER to Golgi traffic Development ERBB-family MAP Kinase Tyrosine/serine/threonine MAPK9, MAPK14, DUSP4 dual specificity phosphatase 4 Signaling Phosphatase Activity MAPK1, MAPK8, MAPK3 Clathrin derived vesicle budding; Glucose/Energy Membrane Organization, Negative CSF3R, TFRC, UBC, FTL, FTH1 ferritin, heavy polypeptide 1 metabolism, Mineral Regulation of Fibroblast Proliferation DAXX Absorption Fatty Acid, Triacylglycerol CTDP1, MED1, MED10, MED8 mediator complex subunit 8 and Ketone body protein ubiquitination TM4SF1, MED19 metabolism MT-ND1, IFITM3, FTL, NAMPT nicotinamide phosphoribosyltransferase NAD metabolism Nicotinamide Metabolic Process UBE2L6, ADORA2A nuclear factor of kappa light polypeptide Apoptosis, Development REL, RELA, IKBKG, IKNKB, NFKBIB Signal transduction gene enhancer in B-cells inhibitor, beta ERBB-family Signaling CHUK pleckstrin homology-like domain, family A, Apoptotic process, FasL Biosynthetic HSPA4L, HSPA4, EIF3D, PHLDA1 member 1 Process RPL14, PABPC4 DNA strand elongation, RNASEH1 ribonuclease H1 Telomere C-strand RNA-DNA Hybrid Ribonuclease Activity synthesis solute carrier family 31 (copper transporter), Copper Ion Transmembrane Transporter UPF1, ZDHHC14, SLC31A1 Mineral absorption member 1 Activity ZDHHC9, ZDHHC18 Downregulated in all treatments identified

by probeset ID alcohol dehydrogenase 5 (class III), chi Formaldehyde Catabolic Process, TP53, TP63, TP73, -1.21 -1.16 -1.07 -1.14 ADH5 Fatty Acid Metabolism polypeptide Peptidyl-cysteine S-nitrosylation ALDH3A2, ADHFE1 (0.002) (0.000) (0.010) (0.008) Sterol regulatory element- diazepam binding inhibitor (GABA receptor CRLS1, TAZ, NUP155, -1.15 -1.44 -1.15 -1.40 DBI binding proteins (SREBPs) Triglyceride metabolic process modulator, acyl-CoA binding protein) TSPO, ADSL (0.019) (0.001) (0.004) (0.001) pathway enoyl CoA hydratase, short chain, 1, CAPN3, DDA1, NT5C2, -1.18 -1.10 -1.05 -1.20 ECHS1 Fatty Acid Metabolism Metabolic Process mitochondrial CCDC42, TIMM22 (0.047) (0.003) (0.004) (0.005) Metabolic Process ; Fatty Acid Catabolic USP20, PECI, TYSND1, -1.42 -1.11 -1.07 -1.22 ECI2 enoyl-CoA delta isomerase 2 Fatty Acid Metabolism Process TAZ, ADSL, PEX14 (0.001) (0.049) (0.003) (0.015) high mobility group nucleosomal binding TERF1, TERF2IP, TERF2, -1.14 -1.29 -1.14 -1.62 HMGN3 - Chromatin modification domain 3 UBC, GRB2 (0.013) (0.000) (0.012) (0.001) -1.33 -1.31 -1.24 -1.23 HOXC6 homeobox C6 - Transcription, DNA-templated UBC, HMGB1, PBX3 (0.032) (0.004) (0.027) (0.041) Glycine, Serine and D-serine Biosynthetic Process; Serine HSPD1, GOLGA3, EIF2B2, -1.45 -1.32 -1.23 -1.29 SRR serine racemase Threonine Metabolism Family Amino Acid Metabolic Process UBC, EEF2 (0.017) (0.003) (0.002) (0.027) -1.34 -1.30 -1.11 -1.46 TMEM14A transmembrane protein 14A - - ELAVL1, NUP35 (0.012) (0.001) (0.003) (0.003) Additional genes downregulated in all

treatments identified by “gene symbols” GNB2, ABCE1, UBC, OGT, ATXN10 ataxin 10 Akt Signaling - EGFR Valine, Leucine and Leucine catabolic process; Branched EHHADH, TERF2, IVD isovaleryl-CoA dehydrogenase Isoleucine Degradation chain amino acid catabolic process, BCKDHA, BCKDHB, DBT PDS5, regulator of cohesion maintenance, Mitotic RNGTT, RAD21, WAPAL, PDS5A Negative Regulation of DNA Replication homolog A (S. cerevisiae) Telophase/Cytokinesis SMC3, STAG2 Cytoskeletal anchoring at plasma ITGB2, ITGA7, TES, ABL1, TLN2 talin 2 ERK/MAPK Signaling membrane; Cell-cell junction assembly, PIP5k1C cell adhesion ZP3, BAT2, ATXN1, UBAP2L ubiquitin associated protein 2-like - - D65S1 *The genes included in this table were identified by a Vienn Diagram using “probesets ID” as key column in the Partek Genomic Suite. Additional genes significantly upregulated and downregulated in all treatments were identified by Vienn Diagram using “gene symbols” as key columns (in this case fold changes are not reported as the gene refers to different probesets. Pathways, GO-biological process and String interaction network were inferred from GeneAnalytics (https://ga.genecards.org ).

Legend to Figures

Fig.1 A. Common upregulated genes identified by the same probesets in our meta-analysis

Fig. 1B. Common downregulated genes identified by the same probesets in our meta-analysis