Effects of the juçara fruit (Euterpe edulis Martius) pulp and lyophilized extract on NRF2, KEAP1, SOD1, and GPX2 expression in human colorectal cancer cell lines

We investigated the effects of the juçara fruit (Euterpe edulis Martius) pulp and lyophilized extract on the expression of cytoprotective genes nuclear factor erythroid 2 (NF-E2)-related factor 2 (NRF2), kelch-like ECH-associated protein 1 (KEAP1), superoxide dismutase (SOD1), and glutathione peroxidase (GPX2) in human colorectal cancer cell lines (HT-29 and Caco-2). Cells were cultured for 24 h in Dulbecco's Modified Eagle's Medium containing juçara fruit pulp (5, 10, or 50 mg/mL) or lyophilized extract (0.05, 0.1, or 0.5 mg/mL), and gene expression was quantified using real-time quantitative reverse transcription polymerase chain reaction. All studied genes showed significant variation in gene expression among different concentrations of pulp or lyophilized extract. Overall, the expression of the selected genes decreased in both cell lines following exposure to the pulp or lyophilized extract in a dose-dependent manner for most of the concentrations studied. In summary, our study showed that the compounds in juçara fruit inhibited the expression of cytoprotective genes associated with the antioxidant response and that, although not cytotoxic at the concentrations studied, they could potentially block the activation of the NRF2/KEAP1 pathway.


Introduction
Colorectal cancer (CRC) is a chronic disease with a high rate of morbidity and mortality worldwide, and therefore a serious public health problem. According to the World Health Organization (WHO), CRC is the third most common and the second most lethal cancer in the world, with 2 million cases and 1 million deaths per year (1).
Etiological and epidemiological studies have attributed the increased incidence of CRC to several environmental factors, including consumption of red meat, smoking, and exposure of the intestinal tissue to carcinogens such as N-nitrous compounds, heterocyclic amines, and infectious agents (2).
The intestinal mucosa is naturally exposed to oxidizing and carcinogenic substances that induce the production of free radicals. The constant production of free radicals results in tissue and cellular damage and, consequently, chronic inflammation. The hydroperoxides contribute with the inflammatory microenvironment and severe oxidative stress can further damage normal cells. Overall, these events can stimulate carcinogenesis (3). The antioxidant defense system of the body consists of several molecules that antagonize free radicals to reduce oxidative stress and protect cells from damage, including the enzymes superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione reductase, and catalase. Some vitamins, such as vitamin C and vitamin E, as well as phenolic compounds found in food also have antioxidant activity (4,5).
The phenolic compounds present in seeds, cereals, herbs, spices, roots, leaves, vegetables, and fruits represent a broad class of antioxidants. These antioxidants can inactivate reactive oxygen species and thereby affect several metabolic and molecular processes associated with neoplastic development (6). Thus, polyphenols act as protective agents by reducing oxidative damage, and are associated with the prevention of chronic diseases such as cancer (7).
The jucara fruit (Euterpe edulis Martius) stands out as a rich source of antioxidant compounds like anthocyanins, which can regulate the expression of cytoprotective genes such as SOD1 and GPx (8). Therefore, we examined the effects of the jucara fruit pulp and lyophilized extract on the expression of cytoprotective genes NRF2, KEAP1, SOD1, and GPX2 in human CRC cell lines.

Material and Methods
Preparation of the jucara fruit pulp and extraction of phenolic compounds For this experiment, a single batch of jucara fruit pulp was obtained from a commercial supplier, Comércio e Beneficiamento de Polpas de Acaí EIRELI ME, Brazil). The pulp was free of dyes and preservatives and hermetically stored at -20°C for protection against light and atmospheric oxygen.
The extraction of phenolic compounds and anthocyanin pigments from the jucara fruit pulp was carried out using a previously published method (9). Briefly, 20 g of jucara fruit pulp was diluted in 200 mL of 70% cereal alcohol and the pH was adjusted to 2.5 with citric acid. The extract was then stored for 2 h at 4°C and protected from light and atmospheric oxygen. After that, the extract was filtered, concentrated in a rotary evaporator, and lyophilized. Lyophilized samples were hermetically stored at 4°C and protected from light until use.

Cell culture
Two human CRC cell lines, HT-29 and Caco-2 (Cell Bank, Brazil), were used. The cells were cultivated in Dulbecco's Modified Eagle's Medium (DMEM, Gibco, Germany). For the Caco-2 cell line, the medium was supplemented with 20% fetal bovine serum and 0.1% antibiotic-antimycotic solution. For HT-29 cells, the medium was supplemented with 10% fetal bovine serum and 1% antibiotic-antimycotic solution. Cell growth was monitored and replaced with fresh medium every two days until the cells reached at least 80% confluence.
Cell viability and proliferation following exposure to jucara fruit pulp and lyophilized extract For this analysis, 8 Â 10 3 HT-29 and Caco-2 cells were seeded onto 96-well plates and cultured for 48 h before exposure to the jucara fruit product. After that, medium containing different concentrations of the jucara fruit pulp (5, 10, and 50 mg/mL) and lyophilized extract (0.05, 0.1, and 0.5 mg/mL) was added to the culture, and cytostatic or cytotoxic effects were determined after 24 h. Control cells, unexposed to the jucara fruit product, were cultivated under the same conditions. The doses used were calculated such that the same amount of cyanidin-3-glucoside was present in both the pulp and extract.
Cell viability and proliferation were measured using the CellTiter 96 s AQueous One Solution Cell Proliferation Assay (Promega, USA). Briefly, exposed and unexposed cells were added to a 96-well plate, and 20 mL of the MTS substrate was added for each 100 mL of medium followed by incubation for 4 h at 37°C. Subsequently, the absorbance of each well was measured at 490 nm with a microplate reader (ELx800 TM , BioTek Instruments, USA). All analyses were performed in triplicate, and the mean values were used for comparisons.

RT-qPCR assay
For this assay, 1 Â10 5 HT-29 and Caco-2 cells were seeded onto 6-well plates and cultured until they reached the ideal confluence. The medium containing different concentrations of the jucara fruit pulp (5, 10, and 50 mg/mL) and lyophilized extract (0.05, 0.1, and 0.5 mg/mL) was added, and total mRNA was extracted after 24 h.
Briefly, samples were incubated with 1.5 mL of Trizol (Invitrogen, USA) and gene expression was measured using GoTaq s 1-Step RT-qPCR System (Promega A6020 kit). The analysis was performed on a 7500 Fast Real-Time PCR System (Applied Biosystems Inc., USA). All analyses were performed in triplicate. The relative expression of each gene was normalized to that of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) using the 2 -DDCT method. All real-time (RT) primers are listed in Table 1.

Statistical analysis
GraphPad s v.7.00 software (GraphPad Software, USA) was used for statistical analysis. Differences were considered significant at Po0.05. Means±SD were calculated for replicates. The cytostatic or cytotoxic effects were evaluated using two-way ANOVA followed by post hoc multiple comparisons t-tests. Relative expression levels of NRF2, KEAP1, SOD1, and GPX2 were analyzed by one-way ANOVA followed by Tukey's test. The comparison of relative expression between cell lines at the different concentrations tested was assessed by multiple t-tests.

Cell viability
Juc¸ara fruit pulp did not show any significant effect on cell viability (P=0.8734) or proliferation at the concentrations studied. Cell proliferation was also not affected and was similar in HT-29 and Caco-2 cell lines (P=0.5603) in the presence of the pulp ( Figure 1A). Furthermore, the lyophilized extract of jucara fruit did not have any effect on cell viability (P=0.3596) at the concentrations studied. However, cell proliferation was significantly different between Caco-2 and HT-29 cells, particularly in the presence of 0.5 mg/mL extract (P=0.0204) ( Figure 1B).
Juc¸ara fruit pulp altered GPX2 expression in HT-29 cells (Po0.0001). At all concentrations tested, GPX2 was under-expressed in relation to unexposed cells, but cells exposed to 5 mg/mL pulp showed higher expression compared to cells exposed to 10 mg/mL (P=0.0008) and 50 mg/mL (P=0.0012). In Caco-2 cells, GPX2 expression also differed after jucara fruit pulp exposure (P=0.0003), but the expression was significantly lower in cells exposed to 10 mg/mL (P=0.0015) and 50 mg/mL (P=0.0016).
Exposure to 5 mg/mL showed GPX2 expression similar to that in unexposed cells (P=0.8434). Further, while comparing cell lines, GPX2 expression was higher (P=0.0068) only in the HT-29 cells exposed to 10 mg/mL fruit pulp ( Figure 2D).
Effect of lyophilized jucara fruit extract on NRF2, KEAP1, SOD1, and GPX2 expression in HT-29 and Caco-2 cell lines In HT-29 cells, NRF2 expression differed significantly in different concentrations of jucara fruit lyophilized extract (Po0.0001), with the highest expression in cells exposed to 0.1 mg/mL (P=0.0172) and was under-expressed in cells exposed to 0.05 mg/mL (Po0.0001) and 0.5 mg/mL (Po0.0001) compared to unexposed cells. In the Caco-2 cell line, NRF2 expression differed significantly in the different concentrations of extract (Po0.0001). NRF2 under-expression was observed in all groups exposed to lyophilized extract; however, cells exposed to 0.05 mg/mL had a greater expression compared to other concentrations. When comparing NRF2 expression between the two cell lines, Caco-2 cells demonstrated higher expression at 0.05 mg/mL and 0.5 mg/mL concentrations (P=0.0072 and P=0.0265, respectively), whereas at 0.1 mg/mL, HT-29 cells showed higher expression (Po0.0001) ( Figure 3A).
The effect of the jucara fruit lyophilized extract on SOD1 expression in HT-29 cells showed that different concentrations promoted under-expression compared to the unexposed cells (Po0.0001). In the Caco-2 cell line, there was also a significant decrease in SOD1 expression at all concentrations used compared to the unexposed cells (Po0.0001). When comparing the expression of SOD1 between cell lines, HT-29 cells presented higher expression at 0.1 mg/mL (Po0.0001), while Caco-2 cells showed higher expression at 0.05 mg/mL (P=0.0035) and 0.5 mg/mL (P=0.0013) ( Figure 3C).

Discussion
In this study, both the pulp as well as the lyophilized extract that are rich in phenolic compounds, mainly anthocyanins, could not inhibit cell proliferation in the two cell lines studied. The anthocyanin (cyanidin-3-glucoside) did not affect HT-29 cell viability in another study as well. Anthocyanins, including cyanidin-3-glucoside, from grape skins did not inhibit the proliferation of a human colon cancer cell line (HCT-116) at a concentration of 200 mg/mL (11).
It is important to note that anthocyanin extracts from different sources vary in their structure and characteristics, such as glycosylation, and therefore have distinct anti-cancer functions (12). For instance, in contrast to nonacylated anthocyanin compounds from blueberry (Vaccinium myrtillus L.), anthocyanin-rich extracts containing acylated monoglycosides from grapes (Vitis vinifera) inhibited the proliferation of HT-29 cells (13). Moreover, another study showed that anthocyanins present in the acidified extract from blue corn (Zea mays L.) have antiproliferative activity in Caco-2 cell lines (14). Thus, antiproliferative effects are apparently dependent on the specific chemical characteristics of the anthocyanin present in the tested compound.
In addition to glycosylation, antiproliferative effects of anthocyanins are also dependent on variables such as period and type of storage (15). Further, the duration of exposure and anthocyanin concentrations also affect growth inhibition in HT-29 and Caco-2 cells (13,14). The concentrations and time of exposure to anthocyanins in jucara fruit products used in this study did not inhibit the proliferation of Caco-2 and HT-29 cell lines. However, the expression of cytoprotective genes responsible for redox state homeostasis and cell survival were modulated by the different concentrations of jucara fruit product studied.
Interestingly, NRF2 expression was modulated in a dose-dependent manner in both human CRC cell lines used in this study. In addition, the NRF2 expression in Caco-2 cells was lower than the expression in HT-29 cells following exposure to the jucara fruit pulp. NRF2 and CRC studies have shown that this protein is overexpressed in these neoplasms (16). Activation of the NRF2 pathway is a mechanism of the cellular antioxidant defense system to overcome redox imbalance and promote intestinal homeostasis (17). Thus, the NRF2 under-expression observed in this study could make cancer cells more responsive to antineoplastic agents, considering that high NRF2 expression is associated with HT-29 cells resistant to chemotherapeutic agents routinely used in CRC treatment (18).
Previous studies have shown that NRF2 underexpression is helpful in CRC treatment as this protein provides cytoprotection against xenobiotics (19). Yokoo et al. (20) demonstrated that NRF2 silencing prevented the increase in aberrant crypt foci and subsequent evolution into CRC by inhibition of the COX2 gene. Notably, NRF2 expression has been reported to offer resistance to CRC development (19). Thus, it can be hypothesized that underexpression of NRF2 in cancer cells by the bioactive compound found in jucara fruit could cause cell death by lowering cytoprotection. In addition, a study that inhibited the expression of NRF2 genes in HT-29 cells increased the cytotoxic activity of 5-fluorouracil, demonstrating that NRF2 is a potential therapeutic target against CRC (21).
External factors, such as oxidative stress, directly induce NRF2 expression. However, NRF2 expression can also be induced indirectly by other mechanisms, such as signaling through the KEAP1 repressor (22). According to Zimta et al. (23), miRNAs may play a role in NRF2 regulation as miRNAs can inhibit the expression of cytoprotective genes including KEAP1 and NRF2.
Moreover, anthocyanins possess inactive free radicals that can potentially affect the external factors involved in NRF2 expression. The chemical structures of anthocyanins make them amenable to hydroxylation, such as the hydroxylation in the 3' position to form cyanidin-3-glucoside, thereby increasing their antioxidant capacity (24). The oxidation of anthocyanins helps in reducing reactive oxygen species (ROS)-induced cellular damage and oxidative stress (25). Neoplastic development is associated with ROS production and proliferation of intestinal stem cells (26). However, an exacerbated production of ROS can cause cell cycle arrest and apoptosis inhibition (27), this mechanism being used by cancer cells to obtain redox homeostasis (28).
NRF2/KEAP1 pathway inducers, such as ROS, interact with KEAP1 residues leading to a structural alteration of KEAP1 and a simultaneous release of NRF2, which enters the nucleus and induces the transcription of antioxidant enzymes (29).
Our results were similar to those of Kropat et al. (29) wherein anthocyanins from Vaccinium myrtillus L. and cyanidin-3-glucoside reduced the expression of NRF2 and other cytoprotective genes in HT-29 cells. Moreover, Yan et al. (30) found that the anthocyanins present in blackberries decreased the expression of NRF2 in liver cells.
Thus, anthocyanins present in the jucara fruit confer antioxidant protection and thereby reduce the expression of the genes studied in HT-29 and Caco-2 cells.
Overall, our analysis showed that NRF2 expression is critical for cytoprotection and that modulation of its regulation by endogenous or exogenous agents could be used to make cancer cells more susceptible to treatment. Further studies are needed to understand the exact molecular mechanisms by which compounds in the jucara fruit regulate NRF2 expression in CRC cell lines.
The expression of another important gene involved in cytoprotection, KEAP1, was also modulated in a dosedependent manner by the jucara fruit product. Moreover, the pattern of KEAP1 expression in HT-29 and Caco-2 cell lines was similar to that of NRF2. Shi et al. (31) observed that KEAP1 under-expression in hepatocellular carcinoma is associated with overexpression of miR-141 and resistance of cancer cells to antineoplastic agents. Thus, it could be hypothesized that the compounds present in the jucara fruit decrease KEAP1 and NRF2 expression and, consequently, the expression of other antioxidant enzymes studied via mechanisms involving miRNAs.
The NRF2/KEAP1 signaling pathway is an important component of the antioxidant defense system. Skrzycki et al. (32) observed that in hypoxia, SOD1 expression was reduced in human CRC cells, thus indicating low levels of ROS production and the absence of oxidative stress. In this study, SOD1 expression was lower in HT-29 and Caco-2 cells exposed to the jucara fruit pulp and lyophilized extract than in control cell lines. In addition, SOD1 expression levels were higher in Caco-2 than in HT-29 cells following exposure to the jucara fruit but not to the lyophilized extract. Our observations suggested that the compounds present in the jucara fruit could have neutralized ROS, thereby decreasing the oxidative stress and consequently reducing the expression of SOD1.
SOD1 overexpression is a characteristic of CRC, and increased SOD1 and GPX2 activity, primarily due to induced gene expression, was observed in several patients with advanced stages of the disease (33). In addition, the increased activity of SOD1 results in a higher transformation rate of the superoxide radical, thus promoting the production of hydrogen peroxide, which is also associated with some types of human cancers (34). Further, free radicals induce the production of angiogenic factors that favor tumor growth and metastasis (35). Dos Reis et al. (8) showed that including the jucara fruit in the diet of carcinogen-induced mice decreased the number of aberrant crypts in the colon and rectum and increased SOD1 protein expression, thereby conferring protection to the colonic mucosa, which demonstrated the ability of phenolic compounds to reduce oxidative stress. Thus, the under-expression of SOD1 observed in this study could be potentially due to the ability of compounds present in the jucara fruit to control oxidative stress without the need for induction of antioxidant enzymes, as shown by the low expression of NRF2. Additional studies are needed to understand the exact role of jucara fruit compounds in chemoprevention.
The expression of GPX2 was similar to SOD1, and it was negatively modulated in a dose-dependent manner in the two human CRC cell lines used in this study. GPX2 mRNA decreased as the concentration of the jucara fruit pulp in the culture medium increased. However, the same pattern was not observed in cells exposed to the lyophilized extract. Increased GPX2 mRNA was observed at the median concentration (0.1 mg/mL) in HT-29 cells and at a 0.5 mg/mL in Caco-2 cells. In all the other studied concentrations, Caco-2 cells had higher levels of GPX2 mRNA than HT-29 cells in the presence of jucara fruit compounds.
GPX2 is involved in Wnt signaling and NRF2 signaling and contains promoter binding sites for the binding of associated transcription factors, such as b catenin/TCF and NRF2. Studies have shown that natural substances rich in flavonoids, such as sulfurane and curcumin, can activate the GPX2 transition via the NRF2 pathway (36). Emmink et al. (28) demonstrated that inhibition of GPX2 by the NRF2 pathway in human CRC cells makes these cells more susceptible to cell death and the action of antineoplastic agents, such as cisplatin, resulting in high levels of intracellular ROS and consequently apoptosis of these cells. In addition, they showed that GPX2 plays an important role in tumor development, such as the survival of neoplastic cells against ROS, favoring the growth of established tumors and promoting the development of neoplastic masses.
Thus, the GPX2 under-expression observed in this study could be the consequence of reduced NRF2 expression. Thus, the under-expression of GPX2 following exposure to jucara fruit compounds may be useful in cancer therapy, based on the reduced antioxidant defense potential shown in HT-29 and Caco-2 cells. Notably, high expression of GPX2 is useful in controlling inflammation by inhibiting COX2; however, in CRC, it increases the rate of cell proliferation and decreases apoptosis in cancer cells (36).
The differences in expression between the HT-29 and Caco-2 cell lines are explained by the fact that they are different neoplastic types. The HT-29 cell line is derived from a colorectal adenocarcinoma in a Caucasian adult female (44 years old) whereas the Caco-2 cell line is derived from a colorectal adenocarcinoma in a Caucasian adult male (77 years old) (37). The differences in gene expression could also be due to the cells being in different stages of neoplastic development and therefore respond differently to changes in the microenvironment and metabolism. Although differences in gene expression between the studied cell lines have been highlighted throughout the discussion, the overall trends were similar in that the expression of the studied genes decreased as the dose of the jucara fruit product increased.
In summary, the jucara fruit pulp or extract did not interfere in the proliferation of HT-29 and Caco-2, but modulated the expression of cytoprotective genes NRF2, KEAP1, SOD1, and GPX2 in a dose-dependent manner, with a tendency to an under-expression of the NRF2/ KEAP1 pathway at higher concentrations. Thus, the pulp and extract of the jucara fruit can potentially reduce the cytoprotection conferred by the genes associated with the antioxidant defense system in human CRC cells. The study also highlighted the pathway as a potential therapeutic target. Further studies are needed to elucidate the exact mechanisms underlying the activity of these products against neoplastic cells.