Iron cellular uptake from almond and coconut beverages using an in vitro intestinal cell model

Plant-based beverages could contain iron but also phenolic compounds (PC) and ascorbic acid, which are considered modulators of iron uptake. We investigated whether iron from almond and coconut beverages can be taken up by Caco-2 cells. The iron content in almond and coconut beverages was 19.20 and 19.34 mg kg−1 dry weight (dw), respectively; PC were 727.71 and 673.11 mg gallic acid equivalents kg−1, respectively; and ascorbic acid 10.96 mg kg−1 and 24.44 mg∙kg−1, respectively. We observed an increase in iron uptake by induction of cellular ferritin at concentrations of 15 and 30 g L−1 (almond) and 30 g L−1 (coconut). Both beverages significantly increased ferritin induction when iron sulfate was added. Thus, these beverages represent iron sources which can either contribute directly to iron supply or indirectly by enhancing absorption of exogenous iron and contribute to decrease the impact of diseases considered public health problems.


Introduction
The consumption of plant-based beverages is steadily growing due to multiple reasons. These include, being: (i) dairy milk alternatives, (ii) lactose free products, (iii) a healthy food option, and also (iv) for certain lifestyle choices [1,2]. Plant based beverages can be obtained from a diversity of plants such as soy, oat, cashew nuts, hazelnuts, hemp, rice, peanuts, coconut, quinoa and almonds [3,4].
Among these, almond and coconut beverages are present in the markets of many countries [2]. The varied composition in vitamins, minerals, fibers, antioxidants, unsaturated fatty acids, bioactive compounds and proteins make these beverages very attractive for the consumers, as well as for the food industry [5][6][7][8]. Research developed in Europe determined that plant-based beverages are the most consumed plant-based products in many countries, with high popularity amongst vegetarians, vegans, and flexitarians (people with reduced consumption of animal products) [9].
Some studies reported that almond and coconut beverages contain naturally important minerals, among them iron [10]. Iron is an essential mineral to humans; it participates in many important biochemical processes and its deficiency is considered a public health problem [11,12]. It is estimated that 2 billion people worldwide still suffer from iron deficiency [13]. Its deficiency is endemic in developed and developing countries and impacts social and economic development [11,14,15].
Plant-based beverages present a variable concentration of iron according to the raw material used to produce the beverage, if they are fortified or not and the impact of other components used during the production process [2,10]. Despite that, the iron from plant sources (non-heme iron) has generally less bioavailability than iron from animal sources (heme iron) [16,17]. In this context, the evaluation of iron uptake from these plant-based beverages is important, since they are used as a substitute for other beverages [10]. The iron concentration in almond and coconut beverages has been reported to be between 0-3.9 mg 100 g −1 and 0-0.13 mg 100 g −1 , respectively [6,7,18]. However, up to now, iron uptake from these beverages has not been evaluated.
Iron bioavailability can be affected by food constituents, such as ascorbic acid or phenolic compounds, which are widely distributed in plant-based foods. A number of phenolic compounds could interfere with iron absorption by binding to iron in the gastrointestinal tract, and therefore decreasing or preventing uptake [13,17].
Caco-2 (Cancer coli-2) cell line has been used as a model of the intestinal epithelial barrier and was originally obtained from a colon carcinoma. One of the advantages of these cells are their ability to differentiate into a monolayer of cells with similar properties typical of absorptive enterocytes as found in the small intestine [19]. And they have been used to evaluate iron uptake in foods and food additives [20][21][22].
Despite the health benefits almond and coconut beverages can provide, studies on their contribution to the dietary intake of essential trace elements, such as iron, are scarce. The aim of this research was to investigate whether iron from almond and coconut beverages can be taken up by Caco-2 cells (used as a model for intestinal epithelial cells); and evaluate if phenolic compounds interfere in iron uptake.

Materials
Commercially available almond and coconut beverages were obtained from local supermarkets in Campinas, Sao Paulo, Brazil. The almond concentration in the beverage was of 8%, for the coconut beverage the coconut concentrations was not declared on the label. However, we believe that the coconut beverage was obtained by the commercial scale that employs the screw press or hydraulic to extract the milk [8]. Prior to sample lyophilization, samples were placed in plates and frozen at -40 °C. After at least 24 h freezing, samples were freeze dried for 48 h up to a pressure of < 200 μHg. Then, samples were stored at -20 °C in vacuum-sealed plastic bags.

Cell culture
The human colon adenocarcinoma cell line Caco-2 was obtained from the German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany). Cells were routinely cultivated in 75 cm 2 cell culture flasks from Corning (Corning, USA) in EMEM containing 10% (v/v) FCS, 1% NEAA, 1% glutamine, 50 units mL −1 penicillin G/50 µg mL −1 streptomycin. Cells were maintained in a humidified atmosphere with 5% CO 2 at 37 °C. Cell culture medium was replaced three times a week.

Determination of ferritin (iron uptake) in Caco-2 cells
The cellular uptake of iron was determined by the induction of the iron storage protein ferritin (nanograms of cell ferritin per milligram of cell protein), that is considered an indicator of cell iron uptake [24]. Ferritin measurement is a widely used method to evaluate iron uptake in cells and this method is also widely accepted in the literature [21,[25][26][27].
Caco-2 cells were seeded into 6 well cell culture plates (10 5 cells well −1 ) and grown in EMEM; with FCS at 37 °C in an incubator with 5% CO 2 until they reached confluence. Cells were used for experiments 7 days after seeding. The cell morphology and the level of confluence were monitored by microscopy daily. Only cells at the same level of confluence and with the same morphology were used for the experiments. Each experiment included a control treatment with 30 µM FeSO 4 .
The following samples were used for the iron uptake study in Caco-2 cells: 1.5 g L −1 ; 15 g L −1 ; and 30 g L −1 of the lyophilized almond and coconut beverages; 3 µM and 30 µM iron sulfate, and 30 g L −1 of the lyophilized almond and coconut beverage enriched with 30 µM iron sulfate in EMEM with FCS. The cells were incubated with 2 mL of each sample preparation for 24 h. The cells viability on each beverage concentration were evaluated at the microscope. The maximal concentration of 30 g lyophilized beverage per liter cell culture medium was chosen to be used. This concentration corresponds to 3% (w/v) of almond and coconut beverage concentration. Considered the dilution of beverages (e.g., 8% for almond beverage) by gastrointestinal juices in the gastrointestinal tract. Used concentrations of beverages are close to the physiological conditions. After the treatments, the cells were washed twice with PBS and thereafter lysed with Triton X-100 reduced form [1% in KH 2 PO 4 buffer (0.1 M)]. After cell lysis, ferritin was quantified using an ELISA kit (APLCO (25-FERHU-E01) according to the supplier instructions. Cell protein content was determined according to Lowry et al. [28] using the DC Protein Assay kit II (Bio Rad, Heidelberg, Germany).

Iron content determination by flame atomic absorption spectrometry (FAAS)
Iron contents of the beverages were performed according to Silva et al. [10]. 4 mL of the liquid beverages were added to 6 mL of nitric acid and 2 mL of hydrogen peroxide (30%) and the mixtures were incubated in a digester block (TE4025 model, Tecnal, Basil) for 4 h at 130 °C. After digestion and cooling, the samples were solubilized in an ultrasound bath and filtered through an ash-free paper filter (Nalgon, 9 cm diameter) and the volume made to 25 mL with ultrapure water. The samples were analyzed by flame atomic absorption spectrometer (FAAS), model AAnalyst 200, with a deuterium lamp for correction of background radiation and hollow cathode lamps for determination of iron at 248.3 nm (PerkinElmer). The samples were placed into a nebulizer and mixed with air-acetylene flame (2.5/10 L h −1 ) at approximately 2000 °C.
The moisture content of the samples was determined according to the method described by the Association of Official Analytical Chemists (AOAC) [29]. Ascorbic acid was quantified in samples by titration and reduction of 2,6-dichlorophenol indophenol [29,30].

Total phenolic compound determination by UV-vis Spectrophotometry
Total phenolic compounds were extracted from the almond and coconut beverages (the initial beverages before freezedrying) according to Varga et al. [31], with modifications, and quantified according to Singleton and Rossi [32]. The extraction was performed through the addition of 4 mL of methanol solution (80%) to 400 μL of sample, followed by shaking and extraction for 15 min and centrifugation for 5 min at 6000 rpm (Refrigerated Centrifuge SL-706, SOLAB).
For phenolic compounds determination, 85 µL of extract were pipetted into a 96-well microplate and 43 μL of Folin-Ciocalteau reagent (1 N) and 212 μL of Na 2 CO 3 (75 g L −1 ) were added. After 30 min at 45 °C, absorbance of the mixtures was determined at 750 nm with a multi-mode microplate reader (BMG Labtech, Germany, model Fluostar Omega). The system was calibrated with gallic acid (0.5-10 mg L −1 ). The results were expressed as mg of gallic acid equivalents (GAEs) per kilogram of sample (mg GAE kg −1 ) on a dry basis.

Statistical analysis
Statistical analysis of the iron uptake (ng ferritin/mg protein) was performed using analysis of variance (ANOVA) followed by Tukey post-hoc test method with 95% confidence for contents of iron, phenolic compound and ascorbic acid; and Dunn's post-hoc test method with 95% confidence for iron uptake evaluation using the Sigma Plot software version (Systat Software Inc., San Jose, CA, USA). If necessary, the data were log-transformed to achieve normality of Vol:.(1234567890)

Results
The moisture content of the almond and coconut beverages was 88.9% and 74.6%, respectively. The iron concentrations in the freeze-dried almond and coconut beverages showed similar values (Table 1), after moisture correction. The concentration of total phenolic compounds in the almond beverage was higher than that in the coconut beverage. The ascorbic acid concentration was higher in coconut samples than in almond beverages. The cellular uptake of iron was measured by induction of the iron storage protein ferritin. As negative control, cells were incubated with cell culture medium (EMEM containing 10% FCS) alone. Control cells showed the intracellular ferritin concentration of 18.00 ± 6.75 ng ferritin/mg protein ( Table 2). It has been shown that 10% FCS can contain iron with a maximal concentration of 96 µg L −1 [23], which is about 5 times lower than those in the both beverage samples at the concentration of 30 mg beverage/L EMEM medium containing 10% FCS. The ferritin concentration in the negative control (EMEM + FCS) significantly differed from the positive controls of 3 µM FeSO 4 (9.8-fold increase) and 30 µM FeSO 4 (33.9-fold increase) ( Table 2). No significant difference in ferritin induction between the negative control was observed for the almond and coconut beverages in a concentration of 1.5 g L −1 of each. However, a statistically significant increase in ferritin induction was found in the almond beverage at the 15 g L −1 (about 1.5-fold) and 30 g L −1 (1.7-fold) concentrations and for the coconut beverage in the 30 g L −1 (1.5-fold) concentration. The almond and coconut beverages showed comparable ferritin levels. However, it seems that iron from the almond beverage could be more effectively induce ferritin in cells than that from the coconut beverage (Table 2).
It is also of interest whether almond and coconut beverages are capable of modulating the cellular uptake of exogenous iron when added to the beverages. Figure 1a shows that addition of 30 µM FeSO 4 to an almond beverage with a concentration of 15 g L −1 significantly increased the cellular iron uptake compared to the EMEM-control (183.4-fold) of the almond beverage with a concentration of 15 g L −1 (1.7-fold) and 30 µM FeSO 4 solution (33.9-fold). The observed effect seemed to be synergistic and not additive. A similar effect on iron uptake, but not as pronounced as for almond beverage, was observed for the coconut beverage (Fig. 1b). The addition of 30 µM FeSO 4 to a coconut beverage with a concentration of 15 g L −1 increased the cellular iron uptake significantly compared to the EMEM-control (126.1-fold) and the coconut beverage with a concentration of 15 g L −1 (1.5-fold).

Discussion
In the present study, the iron content of almond and coconut beverages was determined. Both beverages were able to induce an increase in cellular ferritin concentration in intestinal Caco-2 cells indicating the uptake of iron by epithelial cells. The concentration of iron in both beverages is comparable. However, the cellular ferritin concentration induced by iron uptake from the almond beverage is slightly higher than that caused by coconut beverage. Additionally, the almond beverage had improved ferritin induction with exogenously added iron when compared to the coconut beverage (183fold vs. 126.1-fold compared to EMEM-Control). Compounds forming non-absorbable or non-soluble complexes with iron have been reported to decrease iron absorption [33]. There are a number of phenolic compounds present in plant foods chelating iron and therefore reducing its cellular uptake [20]. Complex formations between iron and phenolic compounds could vary according to the phenolic compound chemical structure; with some being stronger chelators than others [13,34]. Several studies evaluated the capacity of phenolic compounds to bind iron; e.g. protocatechuic acid was considered to be a weak iron chelator, followed by hydroxytyrosol, gallic acid and caffeic acid; while chlorogenic acid was considered a strong chelator [34]. Weak chelators, such as protocatechuic acid, have been identified in almond skins [35,36], whereas stronger chelators such as gallic acid, caffeic acid and chlorogenic acid have been reported in coconuts [37]. Thus, despite the fact that the coconut beverage contained a lower concentration of phenolic compounds compared to the almond beverage, the presence of stronger iron chelators in the coconut beverage might explain the observed higher iron uptake from the almond beverage. Cellular iron uptake can be increased by reducing compounds present in the beverages. Those compounds might be capable of reducing ferric (iron (III)) to ferrous (iron (II)) ions, the molecular form of iron transported across the cell membrane [38]. Reducing constituents of foods may also prevent iron (II) oxidation and thus the formation of insoluble, non-absorbable iron (III) hydroxide at a neutral pH [21,39].
Ascorbic acid or Vitamin C is known as an enhancer of iron availability, preventing its oxidation, being considered a potent antioxidant [40]. It participates in the absorption of non-heme iron in digestion processes, participating in the reduction of Fe (III) to Fe (II), improving its solubility and moreover increasing their possibilities to be transported into by cells. However, the presence of other compounds which could bind iron could be more important during the cellular uptake, since some of them might bind with soluble iron and prevent its absorption [40].
In this study we observed that almond presented a lower content of ascorbic acid when compared to coconut beverage. Despite the potential of ascorbic acid to improve the iron uptake measured by induction of the iron storage protein ferritin. its concentration was not high enough to significantly increase the iron uptake of coconut beverage in comparison with almond beverage. Probably because its concentration was too low to affect the cellular uptake of iron when compared to the other compounds of the food matrix. In this context, we speculate that compounds other than ascorbic acid or phenolics, estimated by the Folin-Ciocalteu method, might have more impact on iron uptake.
In addition, possibly the individual composition of phenolic compounds could have a decisive impact on iron uptake, which should be explored in further studies involving plant-based beverages. Moreover, a separation and identification of phenolic compounds from each matrix by a chromatographic method associated to a mass spectrometry could clarify the compounds that could influence (positively and/or negatively) the iron uptake.
Future research involving the gastrointestinal digestion of plant-based beverages, as almond and coconut beverages, and further application on Caco-2 cells could be performed to evaluate the impact of the digestion process on the beverages and on iron uptake by cells. The performance of the digestive enzymes, residence time, pH and temperature on the macromolecules from matrix could assess whether they could affect the ferritin induction [41,42]. Furthermore, an in vitro model of Caco-2 cells cultured on semi-permeable inserts could allow to study iron transcellular transport from the apical to the basolateral compartment.
The present study shows that the almond and coconut beverages could increase the ferritin induction of iron from an iron salt; indicating that, fortified beverages have a good potential to increase iron absorption according to the beverage composition.

Conclusions
In our study, iron uptake was found to be slightly higher in almond (15 and 30 g L −1 ) and coconut (30 g L −1 ) beverages compared to the negative control. Furthermore, both beverages significantly increased uptake of exogenously added iron (FeSO 4 ). Thus, almond and coconut beverages represent iron sources, either contributing directly or indirectly to iron supply, and are made even stronger by enhancing absorption with exogenous iron.
Iron from these plant sources could contribute to provide an amount of the daily iron needs, minimizing nutritional deficiencies and contributing to improve the health and nutrition of individuals. These findings are of interest, once the milk alternative beverages consumption is raising worldwide and a questioning about the nutritional quality of these products in offering mineral nutrients with or without supplementation has grown among consumers and scholars.