Anti-Hypercholesterolemia Effects of Edible Seaweed Extracts and Metabolomic Changes in Hep-G2 and Caco-2 Cell Lines

Hypercholesterolemia is a major risk for the development of cardiovascular diseases (CVDs), the main cause of mortality worldwide, and it is characterized by high levels of circulating cholesterol. The drugs currently available for hypercholesterolemia control have several side effects, so it is necessary to develop new effective and safer therapies. Seaweeds serve as sources of several bioactive compounds with claimed beneficial effects. Eisenia bicyclis (Aramé) and Porphyra tenera (Nori) are edible seaweeds that were previously recognized as rich in bioactive compounds. In the present study, we aim to evaluate the anti-hypercholesterolemia effect of these two seaweed extracts and their health potential. Both extracts, but more efficiently Aramé extract, have liver 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGR) inhibitory activity as well as the capability to reduce approximately 30% of cholesterol permeation through human Caco-2 cells by simulating the intestinal lining, which is a target for hypercholesterolemia treatments. An untargeted metabolomic assay on human intestinal Caco-2 and liver Hep-G2 cell lines exposed to Aramé and Nori extracts revealed changes in the cells’ metabolism, indicating the extracts’ health beneficial effects. The metabolic pathways affected by exposure to both extracts were associated with lipid metabolism, such as phospholipids, and fatty acid metabolism, amino acid pathways, cofactors, vitamins, and cellular respiration metabolism. The effects were more profound in Aramé-treated cells, but they were also observed in Nori-exposed cells. The metabolite modifications were associated with the protection against CVDs and other diseases and to the improvement of the cells’ oxidative stress tolerance. The results obtained for the anti-hypercholesterolemia properties, in addition to the revelation of the positive impact on cell metabolism, offer an important contribution for further evaluation of these seaweed extracts as functional foods or for CVD prevention.


Introduction
According to the World Health Organization (WHO), cardiovascular diseases (CVDs) are the leading cause of mortality, both in industrialized and developing countries [1]. In 2019, around 17.9 million people died from CVDs, representing 32% of deaths world-wide. By 2030, it is estimated that at least 24 million people may die mainly from heart attack or stroke [1,2].
Hypercholesterolemia is a key factor for the development of CVDs; therefore, it is a major target in risk reduction programs [3,4]. It is characterized by high levels of low-density lipoprotein cholesterol (LDL) and very low-density lipoprotein (VLDL) or triglycerides, as well as by the decrease of high-density lipoprotein (HDL) [5,6]. Cholesterol is a lipid that plays a crucial role in many biochemical and biophysical processes; it has an irreplaceable role in human health. High levels of circulating cholesterol in the blood is a metabolic Chlorophyta) [29]. The colors of the seaweeds are associated with the pigments such as chlorophyll for green, phycobilin for red, and fucoxanthin for brown seaweeds [33]. Eisenia bicyclis, traditionally known as Aramé [34], is a perennial brown seaweed that belongs to the phylum Ochrophyta, class Phaephyceae, order Laminariales [29], and family Laminariaceae, and is distributed along the mid-Pacific coastlines of Korea and Japan [34]. Aramé seaweed is popular in Asian cuisine for soups and salads [35,36]. Porphyra tenera, traditionally known as Nori, is a red seaweed that belongs to the phylum Rhodophyta, class Rhodophyceae, order Bangiales, and family Bangiaceae [37], and is considered to be one of the most popular edible seaweeds since it is used in sushi and soups [38].
In this work, extracts from these seaweeds were evaluated for their potential against hypercholesterolemia, namely the capacity to inhibit HMG-CoA reductase and cholesterol permeation using Caco-2 cells as an in vitro model of the intestinal lining. In order to better understand their health potential, the study was complemented with a metabolomic analysis of the extract's effects on liver Hep-G2 and Caco-2 cell lines. In particular, liver and intestinal cell line metabolites were seen to be affected due the cells' exposure to the complex mixture of bioactive compounds in the extracts. Understanding the activity of these seaweed extracts in the biological systems of these cell lines is very important in order to explain the induced effects that reveal their bioactive mechanisms of action and how these modulated cell metabolic pathways are mainly associated with hypercholesterolemia and CVDs.
To the best of our knowledge, there are no studies that have evaluated the Eisenia bicyclis extract (Aramé) and Porphyra tenera extract (Nori) in greater detail to explain their mechanisms of action regarding their effect on hypercholesterolemia and on intracellular metabolites.

Preparation of Aramé and Nori Seaweed Extract
The dried and milled seaweed biomass was used to obtain the aqueous extracts. The procedure for obtaining a dried mass of Nori extract (50% g/g yield) through decoction in water of 20 g Porphyra tenera biomass/L was similar to a procedure already reported by our group [19] for obtaining the dried mass of Aramé extract through decoction of 33 g Eisenia bicyclis biomass/L of water (66% g/g yield). The seaweed biomass was mixed with water and the suspension was autoclaved at 121 • C for 15 min. Then, the suspension was filtered, frozen, and freeze-dried. In a recent report from the authors of [25], the aqueous extracts from these seaweeds were characterized according to their composition in total phenolic, protein, and polysaccharide content and the mixture of the compounds present was identified through liquid chromatography coupled with high resolution mass spectrometry. The identification of the bioactive compounds in these extracts showed that these contained different mixtures of phlorotannins, i.e., phloroglucinol polymers [25]. Aramé extract was seen to have a higher content in total phenols and to be a mixture of eckol and fucol derivatives, which are a type of phlorotannin derivative. As for Nori extract, it was seen to contain a higher amount of fucol type phlorotannins.

Cell Line Cultures
Hepatocellular carcinoma cell line (Hep-G2) and colorectal adenocarcinoma cell line (Caco-2) were cultured in DMEM and RPMI-1640 medium, respectively, supplemented with 10% (DMEM) and 20% (RPMI-1640) of inactivated FBS, 5 mL of antimycotic and 5 mL of L-glutamine at 37 • C in an atmosphere with 5% CO 2 . The medium was replaced every 48-72 h, and cells were trypsinized before reaching confluence using trypsin 1x.

Inhibition of Enzyme HMG-CoA Reductase (HMGR)
To determine HMGR inhibition, the procedure described in the assay kit (CS1090) was followed. In this assay, the decrease in NADPH (nicotinamide adenine dinucleotide phosphate reduced) absorbance was measured in the presence of HMG-CoA substrate. The initial rate of the enzymatic reaction was quantified by measuring the absorbance at 340 nm for 3 min (V[compound]). A control reaction was carried out using water instead of the extract solution, and this initial rate which was considered 100% of the enzymatic activity (Vcontrol). The percentage of HMGR inhibition (I) for the extracts was determined as the ratio of V[compound] and Vcontrol. All of the assays were carried out in triplicate. The concentration of the extracts used in the reaction mixture was 0.25 mg/mL.

In Vitro Studies in Caco-2 Cell Monolayers Simulating the Intestinal Lining
For these assays, Caco-2 cells were seeded in 12-well Transwell microplate inserts (BD FalconTM) at a density of 5 × 10 4 cells/cm 2 , as described in [39]. The monolayers were formed and differentiated in approximately 23 days. The medium was replaced every 48 h. The monolayer formation was confirmed when it reached a trans-epithelial electrical resistance higher than 250 Ω/cm 2 , measured using a Millicell ERS-2 V Ohm (Millipore, Burlington, MA, USA).

Cholesterol Permeation Assay
To start the assay, the Caco-2 cell monolayers were washed with Hank's Balanced Salt Solution (HBSS). Then, the cell monolayers were incubated for 6 h with 5 mM cholesterol, and either 100 mM ezetimibe or 0.3 mg dried mass extract/mL HBSS or both were added to the samples. These assays were performed three times and compared with control cells that were only incubated with HBSS solution. For both extracts, the concentration used in contact with the cells was selected based on a previous cytotoxicity assay in Caco-2 cells [25].
Following 6 h of incubation, the solutions from both the apical side and basolateral chamber were collected and 75 µL aliquots were analyzed using RP-HPLC-DAD (VWR HITACHI, with an L-2455 detector, L-2300 column oven, L-2200 autosampler, and L-2130 pump). The separation was carried out in an isocratic mode using 50% methanol plus acetonitrile for 15 min with a flow of 1 mL/min. The percentage of cholesterol permeation was determined assuming that 100% cholesterol permeation occurred when cells were incubated for 6 h only in the presence of cholesterol.

Seaweed Compound Permeation Assay
Caco-2 cell monolayers were incubated for 6 h with 0.3 mg dried mass extract/mL HBSS, and once the solutions from both the apical side and basolateral chamber were collected, 25 µL was further analyzed using RP-HPLC-DAD. The separation was carried out using a gradient of trifluoroacetic acid (TFA) (0.05%) and acetonitrile (ACN) for 55 min with a flow of 0.8 mL/min as follows: 0 min, 100% TFA; 30 min, 70% ACN, 30% TFA; 40 min, 20% TFA, 80% ACN; 50 min 70% ACN, 30% TFA; A; 55 min 100% TFA. The percentage of seaweed compound permeation was calculated based on the height of the most intense seaweed peak, assuming that 100% corresponded to the 0.3 mg/mL extract solution in HBSS. These assays for performed three times and compared with the control cells incubated only with HBSS solution.

Metabolomic Analysis of Caco-2 and Hep-G2 Cells Treated with Seaweed Extract
Hep-G2 cells were grown in T25 flasks, as previously described, to reach confluence. Caco-2 cells were also cultivated in T25 flasks, as described previously; they were allowed to reach confluence and afterwards differentiated.
Both cells were incubated with 0.3 mg seaweed extract dried mass/mL in growth medium for 24 h at 37 • C in an atmosphere with 5% CO 2 . The concentration used was selected based on previous reported cytotoxicity assays in both cell lines [25]. Control assays with cells incubated only in growth medium were additionally performed. Twentyfour hours later, the cells were washed with PBS and the cells were collected for metabolite extraction using an adaptation of a procedure described in [40].
The extracted metabolites were analyzed using liquid chromatography (LC) in an UHPLC ELUTE autosampler coupled with high resolution mass spectrometry (HRMS) in an Impact II QToF (quadruple time of flight) mass spectrometer equipped with an electrospray source (ESI) from Bruker (Bruker DaltoniK GmbH, Bremen, Germany). Extracted metabolites were injected (5 µL) in an UHPLC chromatographic column Intensity Solo 2 RP-18, 100 × 2.1 mm, 2.0 µm column (Bruker, Bremen, Germany). The column was kept at 35 • C and the samples at 10 • C. A gradient of water with 0.1% formic acid (eluent A) and acetonitrile with 0.1% formic acid (eluent B) was used at a flow rate of 0.250 mL/min as follows: 0 min-95% A; 1.5 min-95% A; 13.5 min-25% A; 18.5 min-0% A; 21.5 min-0% A; 23.5 min-95% A; 30 min-95% A. High resolution mass spectra were acquired in ESI positive and negative modes. Mass spectra were acquired in the range of 120-1000 m/z and the mass spectrometer parameters were adjusted to optimize the signal-to-noise (S/N) ratio for the ions of interest. Briefly, −3.5 kV and +4.0 kV; end plate offset, 500 V, nebulizer gas (N 2 ) 2.0 bars; dry gas (N 2 ), 8 Lmin −1 ; dry heater, 200 • C; collision cell energy was set to 5.0 eV. The internal calibration was performed with 250 mL H 2 O, 50 mL iPrOH, 750 µL acetic acid, 250 µL formic acid, and 0.5 mL 1N NaOH solution in HPC mode. The acquired data were processed and compared using MetaboScape ® 4.0 software (Bruker Daltonik GmbH, Bremen, Germany). Four replicates were made for each condition.
Metabolites from the cells incubated with the extracts were listed by their charge/mass (m/z) ratio, retention time, and abundance, either in terms of signal intensity or area. The metabolites (retention time-m/z pairs) from cells exposed to each seaweed extract were statistically compared with listed metabolites (retention time-m/z pairs) from the control cells. MetaboScape program generates a statistical analysis and the metabolites considered as having a significant variation between the control and the cells treated with the extract were those presenting a significance level at p-value < 0.05 and fold change below 0.5 and above 1.7.
Multivariate statistics of the metabolites, such as a principal component analysis (PCA) [41], were performed to evaluate the degree of metabolic differences between groups, i.e., control cells versus cells exposed to the extract, but also the similarity between the assayed replicates (n = 4). Another statistical analysis was applied to assess whether the detected differences in the metabolites were statistically significant between the compared groups. Volcano plots were represented as the log 2 mean fold changes in the abundance of the metabolites between the control cells and the extract-exposed cells plotted against −log 10 (p-value). The metabolites showing a p-value lower than 0.05 and a fold change between 0.5 and 1.7 were considered significant. Projection on latent structure (PLS) regressions were additionally applied using MetaboScape to identify similarities and differences in metabolites and/or intensities of the cellular metabolites in the control cells relative to the extract exposed cells, assigning the importance of the altered metabolites depending on the variation found. The variable importance is a measure of the contribution of each individual metabolite for the separation between groups [41]. The cell metabolites from both groups were tentatively identified using databases such as PubChem, HMDB, and Metlin. Additionally, an analysis of the metabolic pathways associated with the changes identified in the metabolites was performed using MetaboAnalyst 5.0.

Data Analysis
The software used for the treatment was Microsoft ® Excel (Microsoft Office 365) and the results were expressed as average ± standard deviation. An additional analysis of variance was carried out using a one-way ANOVA for value comparisons, and the difference between mean values were considered significant when p-value < 0.05.

Results
3.1. Seaweed Extracts' Anti-Hypercholesterolemia Effects 3.1.1. Inhibitory Effect on HMG-CoA Reductase (HMGR) As previously mentioned, the enzyme HMGR is the key enzyme in the cholesterol de novo biosynthesis and is a target for the treatment of hypercholesterolemia [11]. Synthetic drugs, such as statins, are commonly prescribed, but due to their side effects, novel drugs are needed. In addition to lovastatin, which was the first drug to hit the market, there are six other statins available to date: simvastatin, pravastatin, fluvastatin, atorvastatin, rosuvastatin, and pitavastatin [42].
The potential of marine origin products for the treatment of hypercholesterolemia is still largely unexplored, though there are several that show or are claimed to have lipid lowering activities [43]. Porphyra tenera (Nori) and Eisenia bicyclis (Aramé) extracts may affect the cholesterol de novo biosynthesis. HMGR inhibitory activity of these extracts was evaluated and compared with pravastatin. Though there were previously reported results by our group for Aramé extract [19], to the best of our knowledge this is the first report for Nori extract. The results for 0.25 mg/mL solution of Nori dried extract, Aramé extract at the same concentration [19], and pravastatin are shown in Table 1. The Nori extract exhibits a lower inhibition capacity than the Aramé extract with 41% and 79% HMGR inhibition, respectively, and, the positive control pravastatin shows 95% enzyme inhibition. The high value of HMGR inhibition of Aramé extract, as a natural product extract, may reflect the strong potential of edible Aramé seaweed for the reduction of cholesterol levels as a functional food targeting hypercholesterolemia. As for the concentration of the Aramé extract for the achieved 50% HMGR inhibitory activity (IC 50 ), a value of 0.16 mg/mL was obtained, and an IC 50 of 0.30 mg/mL was obtained for the Nori extract. Table 1. HMG-CoA reductase (HMGR) inhibitory activity of 0.25 mg/mL Eisenia bicyclis extract (Aramé), 0.25 mg/mL Porphyra tenera extract (Nori), and pravastatin. Results presented correspond to the mean and standard deviation of three independent assays.
Regarding the differences in the bioactive compounds present in these extracts, the Aramé extract has, as previously seen, a higher content in phenolic compounds and eckol type phlorotannin than the Nori extract [25]. A previous study using brown seaweeds, such as Ecklonia stolonifera, also reported the effect of this eckol and dieckol containing extracts as anti-hyperlipidemic agents in rats [44].

Effects on Cholesterol Permeation In Vitro through Human Caco-2 Cells
Human epithelial colorectal adenocarcinoma cells, Caco-2, are well characterized in an intestinal in vitro model with morphologic resemblance to intestinal epithelia [45]. When cultured under appropriate conditions, this cell model is accepted as a surrogate for human intestinal permeability measurements by the regulatory agencies such as the FDA and EMA, and also as a screening tool for intestinal absorption, transport, and metabolism in support of drug discovery [46]. In order to compare the impact of the two seaweed extracts in the intestinal cholesterol absorption, Caco-2 cells were cultured as a monolayer and allowed to differentiate for 21 days, simulating the human intestinal epithelial membrane. The differentiated cell permeability to cholesterol was assessed in the presence of the extracts and both in the absence or presence of ezetimibe (Ezet.), the drug prescribed for reducing diet cholesterol permeation. Ezetimibe was also used alone as a positive control. Figure 1 shows the results after 6 h of incubation with cholesterol of the percentage of reduction in cholesterol permeation at the different conditions tested. The permeability of cholesterol alone through the differentiated cells was considered 100% permeation. Ezet. Reduction in cholesterol permeation (%) Figure 1. Reduction of cholesterol permeation through Caco-2 cell monolayer with extracts of Aramé and Nori. Ezetimibe (Ezet.) was used as a positive control. Caco-2 cells were differentiated for 21 days and incubated with 5 mM cholesterol for 6 h. Cholesterol was quantified using RP-HPLC-DAD, as mentioned in Section 2.5. a,b correspond to values that are statistically different between the samples under study (p < 0.05). * Results in [19]. Results presented correspond to the mean and standard deviation of three independent assays. This is the first report demonstrating Nori extract's ability to inhibit cholesterol permeation in Caco-2 cells, a model of the intestinal lining. However, it can be seen in Figure 1 that in the presence of ezetimibe plus Nori (Nori + Ezet.), a lower percentage of reduction in cholesterol permeation was reached (24%) than using ezetimibe alone (52%), anticipating that the Nori extract interferes with ezetimibe's effect. One may speculate that Nori extract's major bioactive compounds, tentatively identified previously as smallersized fucol type phloroglucinol derivatives [25], may interfere with ezetimibe's effect on the intestinal lining, which is associated with NPC1L1 protein endocytosis inhibition [48]. The awareness of interactions of seaweed compounds with drugs is limited to a few studies, mostly with anticancer drugs [49,50]; however, this is an important issue that needs to be carefully addressed considering that edible seaweeds, such as Nori, are often used in the diet.

Seaweed Extract Permeation In Vitro through Human Caco-2 Cells
A permeability assay was also conducted to evaluate whether the extract compounds permeated the gastrointestinal barrier. The results are shown in Figure 2. Overall, the results revealed that the major compounds of the analyzed seaweed extracts permeated Caco-2 cells, simulating the human intestinal epithelial membrane by less than 30%. Considering the nature of the bioactive compounds in the extracts, which are mostly lower molecular weight phlorotannin derivatives [25], it was expected that these would permeate the intestinal barrier [51,52]. ) was used as a positive control. Caco-2 cells were differentiated for 21 days and incubated with 5 mM cholesterol for 6 h. Cholesterol was quantified using RP-HPLC-DAD, as mentioned in Section 2.5. a,b correspond to values that are statistically different between the samples under study (p < 0.05). * Results in [19]. Results presented correspond to the mean and standard deviation of three independent assays.
It can be seen in Figure 1 that the ezetimibe (Ezet.) reduced the cholesterol permeation of the Caco-2 cells by 52%, and a similar effect was obtained when cells were in contact with ezetimibe in addition to the Aramé extract (Aramé + Ezet.). The incubation of the cells solely with any of the extracts, Aramé or Nori, only reduced cholesterol permeation by approximately 30%. Though these values are below the effect of ezetimibe, these results are considered promising as the extracts are a mixture of several compounds. Therefore, the compounds in the extracts have a promising potential for lowering cholesterol permeation in the intestinal lining, contributing to control of dietary cholesterol absorption. Considering previous results on other brown seaweeds, such as Fucus vesiculosos [47], and the composition of Aramé and Nori extract [25], it can be suggested that the extract compounds could interfere with the expression of cholesterol transporters in the intestinal lining NPC1L1 and ABC transporters. This is the first report demonstrating Nori extract's ability to inhibit cholesterol permeation in Caco-2 cells, a model of the intestinal lining. However, it can be seen in Figure 1 that in the presence of ezetimibe plus Nori (Nori + Ezet.), a lower percentage of reduction in cholesterol permeation was reached (24%) than using ezetimibe alone (52%), anticipating that the Nori extract interferes with ezetimibe's effect. One may speculate that Nori extract's major bioactive compounds, tentatively identified previously as smaller-sized fucol type phloroglucinol derivatives [25], may interfere with ezetimibe's effect on the intestinal lining, which is associated with NPC1L1 protein endocytosis inhibition [48]. The awareness of interactions of seaweed compounds with drugs is limited to a few studies, mostly with anticancer drugs [49,50]; however, this is an important issue that needs to be carefully addressed considering that edible seaweeds, such as Nori, are often used in the diet.

Seaweed Extract Permeation In Vitro through Human Caco-2 Cells
A permeability assay was also conducted to evaluate whether the extract compounds permeated the gastrointestinal barrier. The results are shown in Figure 2. Overall, the results revealed that the major compounds of the analyzed seaweed extracts permeated Caco-2 cells, simulating the human intestinal epithelial membrane by less than 30%. Considering the nature of the bioactive compounds in the extracts, which are mostly lower molecular weight phlorotannin derivatives [25], it was expected that these would permeate the intestinal barrier [51,52]. These findings are significant and suggest that the extract compounds in vivo have the potential for permeating the gastrointestinal barrier and then moving towards other body cells, tissues, or organs, ultimately successfully reaching the liver where HMGR may be inhibited, as described in Section 3.1.1. Due to the ability of the extract compounds to permeate barriers, it seemed appropriate to further explore the impact of these seaweed extracts on the metabolism using both Caco-2 cells and human liver hepatoblastomaderived cell line Hep-G2 cells.

Metabolomic Effect of Seaweed Extract on Caco-2 and Hep-G2
To evaluate the mode of action and potential health effects of edible seaweed extract intake, an untargeted liquid chromatography associated with high-resolution mass spectrometry (LC-QTOF-MS) metabolomics analysis was performed. The aim of this approach was to identify key metabolites and metabolic pathways modified in liver Hep-G2 cells and intestinal Caco-2 cells in response to exposure to the extracts.

•
Aramé extract Caco-2 cell metabolites (Control) were compared with the metabolites from Caco-2 cells incubated in the presence of the Aramé extract. Following the statistical analysis of the list of metabolites collected through mass spectrometry, either in the MS positive or negative mode, (Supplementary Material, Figure S1) the statistically different putative metabolites were recognized and several were tentatively identified using databases such as PubChem, HMDB, and Metlin. In the heatmaps shown in Figures 3A and 4A for the MS positive and negative modes, respectively, the most abundant identified metabolites are represented in green and the least abundant metabolites are shown in red. Figures 3B  and 4B show the relative abundance (log2 (fold change)) for the putative identified Caco-2 cells were differentiated for 21 days and incubated with the extract for 6 h; after 6 h, the extracts were quantified using RP-HPLC-DAD, as mentioned in Section 2.5. Letter a corresponds to values that are not statistically different between the samples under study (p < 0.05). Results presented correspond to the mean and standard deviation of three independent assays. These findings are significant and suggest that the extract compounds in vivo have the potential for permeating the gastrointestinal barrier and then moving towards other body cells, tissues, or organs, ultimately successfully reaching the liver where HMGR may be inhibited, as described in Section 3.1.1. Due to the ability of the extract compounds to permeate barriers, it seemed appropriate to further explore the impact of these seaweed extracts on the metabolism using both Caco-2 cells and human liver hepatoblastomaderived cell line Hep-G2 cells.

Metabolomic Effect of Seaweed Extract on Caco-2 and Hep-G2
To evaluate the mode of action and potential health effects of edible seaweed extract intake, an untargeted liquid chromatography associated with high-resolution mass spectrometry (LC-QTOF-MS) metabolomics analysis was performed. The aim of this approach was to identify key metabolites and metabolic pathways modified in liver Hep-G2 cells and intestinal Caco-2 cells in response to exposure to the extracts.  Figure S1) the statistically different putative metabolites were recognized and several were tentatively identified using databases such as PubChem, HMDB, and Metlin. In the heatmaps shown in Figures 3A and 4A for the MS positive and negative modes, respectively, the most abundant identified metabolites are represented in green and the least abundant metabolites are shown in red. Figures 3B and 4B show the relative abundance (log 2 (fold change)) for the putative identified metabolites between the control cells (blue) and the Caco-2 cells in contact with Aramé extract (orange).

Caco-2 metabolites MS positive mode Control Aramé
180,135   The same type of analysis and data treatment were carried out for liver Hep-G2 cells exposed to the Aramé extract. The cell-modified metabolites were extracted, MS was analyzed in both positive and negative modes, and the results were compared with the metabolites from non-exposed control cells. Following the statistical analysis   The same type of analysis and data treatment were carried out for liver Hep-G2 cells exposed to the Aramé extract. The cell-modified metabolites were extracted, MS was analyzed in both positive and negative modes, and the results were compared with the metabolites from non-exposed control cells. Following the statistical analysis (Supplementary Materials, Figure S2), the heatmaps of the statistically significantly different abundant metabolites between the control cells and Aramé-exposed cells were tentatively identified, and the differences can be seen in Figures 5A and 6A for the MS positive and negative modes, respectively. Figures 5B and 6B show the differences in relative abundance (log2 (fold change)) for the putative identified metabolites between Hep-G2 control cells (blue) and Hep-G2 cells in contact with Aramé extract (orange).  Figure S2), the heatmaps of the statistically significantly different abundant metabolites between the control cells and Aramé-exposed cells were tentatively identified, and the differences can be seen in Figures 5A and 6A for the MS positive and negative modes, respectively. Figures 5B and 6B show the differences in relative abundance (log2 (fold change)) for the putative identified metabolites between Hep-G2 control cells (blue) and Hep-G2 cells in contact with Aramé extract (orange).
(A) (B)  Figure S2), the heatmaps of the statistically significantly different abundant metabolites between the control cells and Aramé-exposed cells were tentatively identified, and the differences can be seen in Figures 5A and 6A for the MS positive and negative modes, respectively. Figures 5B and 6B show the differences in relative abundance (log2 (fold change)) for the putative identified metabolites between Hep-G2 control cells (blue) and Hep-G2 cells in contact with Aramé extract (orange).
(A) (B) It was also noted through a PLS plots analysis (Figures S1 and S2c) that in both types of cells, the most important metabolites altered due to cell exposure to Aramé were mostly vitamins, phospholipids, or related fatty acid metabolites, as well as amino acids or peptides. Other types of metabolites were seen to be decreasing in exposed cells, such as reduced glutathione (GSH), which is one of the most important scavengers of reactive oxygen species (ROS) [53]. On the contrary, nucleotides, such as as uridine (U) and hypoxanthine (HPX), which is an important metabolite from the purine metabolism often associated with metabolic disorders [54], were seen to increase in the extract-treated cells.
A pathway enrichment analysis of the obtained data was performed using Metabo-Analyst 5.0 tool and Figure 7 presents the metabolic pathways estimated to be affected by exposure to the Aramé extract in Caco-2 and Hep-G2 cells. It was also noted through a PLS plots analysis (Figures S1 and S2c) that in both types of cells, the most important metabolites altered due to cell exposure to Aramé were mostly vitamins, phospholipids, or related fatty acid metabolites, as well as amino acids or peptides. Other types of metabolites were seen to be decreasing in exposed cells, such as reduced glutathione (GSH), which is one of the most important scavengers of reactive oxygen species (ROS) [53]. On the contrary, nucleotides, such as as uridine (U) and hypoxanthine (HPX), which is an important metabolite from the purine metabolism often associated with metabolic disorders [54], were seen to increase in the extract-treated cells.
A pathway enrichment analysis of the obtained data was performed using MetaboAnalyst 5.0 tool and Figure 7 presents the metabolic pathways estimated to be affected by exposure to the Aramé extract in Caco-2 and Hep-G2 cells.

Caco-2
Hep-G2 As can be seen from Figure 7, the metabolic pathways most affected in both types of cells when exposed to Aramé extract are those associated with lipid metabolism such as phospholipids and fatty acid metabolism. Furthermore, the pathways of amino acid metabolism, cofactors, and vitamins, as well as the energy metabolism of cellular As can be seen from Figure 7, the metabolic pathways most affected in both types of cells when exposed to Aramé extract are those associated with lipid metabolism such as phospholipids and fatty acid metabolism. Furthermore, the pathways of amino acid metabolism, cofactors, and vitamins, as well as the energy metabolism of cellular respiration, namely catabolism or oxidation, were identified to be altered when the cells are exposed to the Aramé extract.
Modifications in several of these pathways have been associated with CVDs besides hypercholesterolemia and oxidative stress. Sulfur amino acid pathways were reported as being associated with the risk of CVDs [8,55]. Sulfur amino acid pathways, such as homocysteine degradation pathway, methionine, and cysteine metabolism, are related pathways. Within the body tissues, the metabolism of methionine and cysteine determines the concentrations of several metabolites including coenzyme A, glutathione, and taurine [56][57][58][59]. In the cells treated with Aramé extract, a decrease in the level of GSH was detected relative to the control cells, which can be associated with a higher utilization rate of this metabolite inside the treated cells, a mechanism important to prevent oxidative stress. A modification in glutathione metabolism was seen in Hep-G2 cells exposed to a brown seaweed Fucus vesiculosus extract [60].
The metabolism of sulfur amino acids also has an impact on the synthesis of fatty acids and phospholipids [55]. These metabolites are important for cell membrane, especially for shielding the cells from oxidative stress, preventing several disorders such as CVDs, metabolic conditions, and neurodegenerative disorders [61]. It was seen that there were differences in the level and type of phospholipids and fatty acids between the control and Aramé-exposed cells. Phosphatidylcholine (PC) and phosphatidylethanolamine (PE) are major phospholipids in the mammalian membrane and their modifications are reported to modify the cell membrane behavior [62], which may restrict cell membrane fluidity, thus interfering with permeation processes and/or protecting from damage. Furthermore, PC is the major phospholipid component of lipoproteins and PE is its precursor. A reduction in PC/PE level is reported to significantly lower the levels of very low-density circulating lipoproteins (VLDLs) [63], and this reduction was observed in Aramé-treated cells.
The spermidine and spermine biosynthesis are also often associated with the ability to protect against CVDs [64], as these polyamines can act as anti-inflammatory, antioxidants, and free radical scavengers [65], and are also associated with the glutathione metabolism. In Aramé-exposed cells, a higher level of spermine was detected compared with the control cells.
Purine metabolism was also seen to be elevated in Aramé Hep-G2 and Caco-2-treated cells with accumulation of hypoxanthine (HPX), inosine (I), and deoxyinosine (D), which are products of ATP catabolism. Modifications in the purine pools in the cells are often exploited as a therapeutic target against several diseases such as cancer [66] and gout [67] and also central nervous system disorders [68].
In Hep-G2 cells exposed to Aramé extract, the thiamine metabolism was seen to be affected with a lower level of thiamine (vitamin B1) in treated cells relative to the control cells. It may be suggested that in extract-treated Hep-G2 cells, thiamine pool can be depleted due to a high metabolic rate, which increased the need of this vitamin, due to the thiamine diphosphate coenzyme role in the energy metabolism of carbohydrates, lipids, and amino acids [69]. It is known that under certain stress conditions, the intracellular thiamine pool is depleted, while increasing oxidative stress tolerance [70].

•
Nori extract The same type of evaluation was performed for the metabolites from Caco-2 cells after 24 h exposure to Nori extract compared with the metabolites from non-exposed Caco-2 cells (Control). The statistical analysis is shown in Supplementary Materials Figure S3, both for the MS negative and positive modes. In the heatmaps, Figures 8A and 9A show the statistically different metabolites tentatively identified using the previous mentioned databases and are colored from green to red to visualize the differences in the abundance between the control and extract-treated cells. The log 2 (Fold change) plots in Figures 8B and 9B, represent the ratio between the control cell metabolite abundance and Nori-exposed cells. The most abundant metabolites in the control cells are represented by the positive blue bars and the metabolites most abundant in Nori-exposed cells are represented by orange bars.
positive blue bars and the metabolites most abundant in Nori-exposed cells are represented by orange bars. The same type of data analysis and evaluation were carried out for the comparison of the metabolites in Hep-G2 cells exposed to Nori extract with the metabolites from nonexposed Hep-G2 control cells. The statistical analysis is shown in Supplementary Materials Figure S4 The same type of data analysis and evaluation were carried out for the comparison of the metabolites in Hep-G2 cells exposed to Nori extract with the metabolites from nonexposed Hep-G2 control cells. The statistical analysis is shown in Supplementary Materials Figure S4 The same type of data analysis and evaluation were carried out for the comparison of the metabolites in Hep-G2 cells exposed to Nori extract with the metabolites from nonexposed Hep-G2 control cells. The statistical analysis is shown in Supplementary Materials Figure S4, both for the MS negative and positive modes. The heatmaps obtained after tentative identification of significantly variant metabolites are shown in Figures 10A and 11A for the MS positive and negative modes, and the log 2 (Fold change) plots are shown in Figures 10B and 11B. Using the PLS plots (( Figures S3 and S4c)) and analysis of Figures 8 to 11, it can be seen that overall, for both type of cells, the most important metabolites that changed between the control cells and Nori-exposed cells were mostly phospholipids or related fatty acids, nucleotides, or related purines and pyrimidines, as well as vitamins and cofactors. In the case of Hep-G2 cells exposed to Nori extract, it was additionally seen that reduced glutathione (GSH) decreased where hypoxanthine (HPX) increased, and thiamine (B1) levels decreased similarly to that previously seen for Aramé-treated Hep-G2 and Caco-2 cells. However, in the case of cells treated with Nori, the variation was less intense, as lower values of log2(FoldChange) were obtained relative to the cells treated with Aramé.

Hep-G2 metabolites MS positive mode Control Nori
Benzylpenicilloyl-Cysteine (BPO-cys) 49  Using the PLS plots (( Figures S3 and S4c)) and analysis of Figures 8 to 11, it can be seen that overall, for both type of cells, the most important metabolites that changed between the control cells and Nori-exposed cells were mostly phospholipids or related fatty acids, nucleotides, or related purines and pyrimidines, as well as vitamins and cofactors. In the case of Hep-G2 cells exposed to Nori extract, it was additionally seen that reduced glutathione (GSH) decreased where hypoxanthine (HPX) increased, and thiamine (B1) levels decreased similarly to that previously seen for Aramé-treated Hep-G2 and Caco-2 cells. However, in the case of cells treated with Nori, the variation was less intense, as lower values of log2(FoldChange) were obtained relative to the cells treated with Aramé.

Hep-G2 metabolites MS positive mode Control Nori
Benzylpenicilloyl-Cysteine (BPO-cys) 49  Using the PLS plots (( Figures S3 and S4c)) and analysis of Figures 8-11, it can be seen that overall, for both type of cells, the most important metabolites that changed between the control cells and Nori-exposed cells were mostly phospholipids or related fatty acids, nucleotides, or related purines and pyrimidines, as well as vitamins and cofactors. In the case of Hep-G2 cells exposed to Nori extract, it was additionally seen that reduced glutathione (GSH) decreased where hypoxanthine (HPX) increased, and thiamine (B1) levels decreased similarly to that previously seen for Aramé-treated Hep-G2 and Caco-2 cells. However, in the case of cells treated with Nori, the variation was less intense, as lower values of log 2 (FoldChange) were obtained relative to the cells treated with Aramé.
As a pathway analysis can give a more intuitive interpretation of the modifications, the metabolites were mapped to the metabolic pathways using Metaboanalyst 5.0 pathway enrichment analysis, and Figure 12 was obtained.
As a pathway analysis can give a more intuitive interpretation of the modifications, the metabolites were mapped to the metabolic pathways using Metaboanalyst 5.0 pathway enrichment analysis, and Figure 12 was obtained.

Caco-2
Hep-G2 Figure 12. Metabolic pathways of Caco-2 and Hep-G2 cells positively affected by the proposed compounds in the presence or absence of Nori. Obtained using MetaboAnalyst 5.0 pathway enrichment analysis of significantly different metabolites between groups (p < 0.05). The enrichment rate is calculated as the number of hits between a particular metabolic pathway divided by the expected number of hits.
In Figure 12, the pathway enrichment analysis identified that the most affected metabolic pathways in both cell types due to contact with Nori extract were lipid metabolism pathways, amino acid, cofactor and vitamin, and energy metabolism pathways, as seen also for cells of exposed to the Aramé extract.
As it was mentioned for the Aramé-treated cells, variations in cell membrane lipids seen also in Nori-treated cells, are reported to be associated with modifications in cell membrane fluidity and permeation behavior [62,63], as well as on the possible effect on decreasing lipoprotein levels, which may have beneficial effects on plasma circulating cholesterol [64]. However, in the case of cells treated with Nori, the variation was less intense, as lower values of log2(FoldChange) were obtained relative to the cells treated with Aramé.
As it was mentioned for the Aramé-treated cells, variations in cell membrane lipids seen also in Nori-treated cells, are reported to be associated with modifications in cell membrane fluidity and permeation behavior [62,63], as well as on the possible effect on decreasing lipoprotein levels, which may have beneficial effects on plasma circulating cholesterol [64]. However, in the case of cells treated with Nori, the variation was less intense, as lower values of log 2 (FoldChange) were obtained relative to the cells treated with Aramé.
Furthermore, in the case of the amino acid tryptophan metabolism, in Caco-2 cells treated with Nori extract, an increase in the levels of quinone-derived metabolites often associated with this pathway was observed; again, this was also seen in Hep-G2 and Caco-2 cells treated with Aramé, but more significantly. The formation of these types of metabolites after cell exposure to the extract may be associated with cell protection, as the quinone moiety is highly reactive, thus prescribing its use against several diseases [71].

Discussion
The potential of marine origin products, including seaweeds, for prevention or treatment of diseases is still largely unexplored. As some seaweeds and derived products are already available as nutraceuticals or claimed as functional foods, it is urgent to demonstrate and further clarify their effects on consumers' health.
The results here reported have demonstrated that the often-consumed brown seaweed Aramé and red seaweed Nori extracts may have an effective potential against hypercholesterolemia, one of the main factors associated with the risk of CVDs. Both extracts have demonstrated in vitro the capacity to reduce the cholesterol de novo synthesis by inhibiting the HMGR enzyme, a target for hypercholesterolemia treatment. Aramé extract showed a better performance than Nori extract for the same concentration. Given the differences in the composition of these extracts, Aramé extract showed higher levels of phenolic compounds and the eckol type phlorotannin, which are absent in the Nori extract [25]. It could be suggested that the inhibition is associated with the latter. Additionally, both extracts showed the capacity to reduce cholesterol permeation in the in vitro gastrointestinal barrier, also a target of the hypercholesterolemia drug ezetimibe. Nevertheless, it was demonstrated for the first time that the Nori extract may interfere with the ezetimibe effect on the intestinal lining. It was also demonstrated that the compounds of Aramé and Nori extracts could permeate the gastrointestinal barrier in vitro, and have the potential to move towards other body cells, tissues, or organs, and hypothetically reach the liver. These results are in accordance with our previous results that demonstrated in vivo that Aramé extract reduced cholesterol levels in Wistar rats [19].
To complement these results, an untargeted metabolomic analysis was carried out in intestinal Caco-2 and liver Hep-G2 cell lines to elucidate the mechanism of action associated with the consumption of both seaweeds. Our findings demonstrated variations occurring in Aramé and Nori extract-exposed cells, and showed that these could be associated with health benefits, particularly towards metabolic and other disorders, such as CVDs and oxidative-related diseases.
The cell metabolome in contact with the extracts showed an increase or decrease in several metabolites, and these were related to alterations in the cells' major pathways such as lipid, amino acid, vitamin, and nucleotide pathways. Using either of the extracts, the analyzed cell lines showed a decrease in GSH relative to the control cells, possibly due to a higher utilization rate of this metabolite inside the treated cells, suggesting the enhancement of cells' antioxidant mechanisms.
Modifications in lipids could be overall associated with changes in cell membranes, but particularly a decrease in PC and PE levels, suggesting that this could contribute to reducing the formation of circulating VLDLs by these cells. Other types of alterations in key metabolites, such as spermine, quinone derivatives, and vitamin B1, with antioxidant and anti-inflammatory potential, could be associated with protection against CVDs and other diseases, as well as with the improvement of cell oxidative stress tolerance. Both extracts have been already reported as showing antioxidant activity, with Aramé extract showing a better performance than Nori extract [25]. Herein it was also noticed, that though some modifications were seen in cells exposed to either extract, these were more robust in Aramé-treated cells than in Nori-exposed cells. Furthermore, for the Aramé extract, the obtained metabolomic changes are in agreement with recent results obtained in an in vivo study of the effect of the extract in the serum of Wistar rats, showing a decrease in blood cholesterol relative to control rats [19].

Conclusions
We have presented scientific evidence about the potential of brown seaweed Aramé and red seaweed Nori extracts for managing hypercholesterolemia. Both extracts demonstrated an efficient reduction of cholesterol in vitro, targeting cholesterol de novo biosynthesis and additionally the permeation in intestinal lining, showing a similar effect as the drugs often prescribed. In order to fully explore the potential of these seaweed extracts for hypercholesterolemia treatment, further research is necessary to understand the pharmacokinetic and pharmacodynamic properties to establish the adequate dosage for therapeutic products. However, it is noteworthy that the incorporation of Aramé seaweed in a healthy and balanced diet stands as a beneficial way to prevent hypercholesterolemia and to promote health. These applications, were likewise, herein scientifically demonstrated by our metabolomic findings in vitro, which showed that these seaweed extracts impact in the cells' pathways is possibly advantageous for cell protection against several disorders, including the prevention of CVDs.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/life13061325/s1. Figure S1-Analysis of the compounds of Caco-2 cells that vary in the presence or absence of Aramé in positive (1) and negative (2)