Nutritional composition and minerals bioaccessibility of commercial fruit flours

Commercial fruit flours were evaluated concerning: (i) the nutritional composition (proximate composition, total phenolic content, and minerals content), (ii) their contribution to estimated mineral daily intake, (iii) the bioaccessibility of essential minerals using the in vitro INFOGEST digestion method and (iv) the influence of their chemical composition on minerals bioaccessibility. The 20 samples analysed presented high variability concerning the content of dietary fibre (7.5 to 69.7 g/100 g), carbohydrates (4.1 to 74.9 g/100 g), protein (2.9 to 12.9 g/100 g), ash (1.0 to 7.0 g/100 g), lipids (1.0 to 8.1 g/100 g) and total phenolic content (2.9 to 41.0 mg GAE/g. The mineral content of fruit flours provides a great contribution to the daily mineral requirements (especially Mg, Fe, Mn and Cu) with a daily intake of 30 g and very low contribution to the daily requirements of Na (0–3%). Low bioaccessibility was observed for Ca (18.0%) and Fe (28.9%), while Mg was the most bioaccessible mineral (81.5%). Though, the bioaccessible fraction of Mg showed negative correlation with total dietary fibre content (r = − 0.77) and lipids (r = − 0.46).


Introduction
Fruit flours are functional products become increasingly popular over the years. Green banana and passion fruit flours are some of the most common to be found, but there are many types of fruit flour that are commercially available, like orange flour, grape flour, apple flour, among others.
These products are mainly used as dietary supplements [1,2] or food ingredients due to their high fibre content [3,4]. They can be obtained from the peel [5,6], seeds [7], pulp [8], pulp and seeds [2] or pulp and peel [9], which will depend on the characteristics of the fruit and the raw material available. These flours are often produced from by-products of the manufacture of juices and other fruit products. Thus, the production of fruit flours represents an alternative for the use of waste generated in fruit processing [2,6,7,[9][10][11][12]. Considering the growing interest in fruit flours as functional foods, it is particularly important to know their nutritional composition and nutrients release in the gastrointestinal tract. Compounds like phytates, dietary fibres, polyphenols, oxalates and phosphates can inhibit the absorption of minerals, while other compounds like organic acids (ascorbic, citric, lactic) are known to enhance their absorption [13].
The bioaccessibility of nutrients in foods has been estimated mainly by in vitro digestion methods, as they are simple, reproducible, and relatively inexpensive when compared to in vivo procedures. The standardized INFOGEST method was proposed for simulating digestion of foods and comprises three steps, corresponding to the oral, gastric, and 1 3 intestinal phases of the human digestion procedure [14]. Some authors have reported the mineral bioaccessibility in fruit by-products, like residues of citrus [15]. Moreover, the bioaccessibility of Mg, Ca, Zn, Mn, Cu and Fe was assessed in three commercial brands of green banana [16]. The authors observed that the bioaccessibility of the elements was negatively influenced by the presence of proteins and phytic acid. However, studies related to the chemical composition and mineral bioaccessibility in commercial fruit flours are still scarce. Thus, the aims of this study were to evaluate (i) the nutritional composition (proximate composition, total phenolic and mineral contents), (ii) the estimated daily intake (EDI) of minerals with a daily consumption of 30 g, (iii) the bioaccessibility of essential minerals and (iv) the influence of chemical composition on minerals bioaccessibility in different commercial fruit flours.

Proximate composition pH, acidity and total phenolic content (TPC)
The proximate composition of the fruit flour samples was analysed according to AOAC official methods [17]. Total dietary fibre content was determined by the enzymatic-gravimetric method using commercially available kits (K-TDFR, Megazyme, Cork, Ireland), based on AACC method 32-05.01 and AOAC method 985.29. Carbohydrate content was calculated through the equation: 100 − (sum of moisture, protein, fat, ash, and total dietary fibre). Energy values were estimated using the Atwater factors. For pH measurements, about 2.5 g of fruit flour was weighed and added to 100 mL of water. The acidity measurements were performed by titration with a 0.1 M NaOH solution to pH 8. 2-8.4. The acidity values were expressed as mg of organic acid per g of sample, considering the most abundant organic acid (malic acid for green banana and apple flours, citric acid for passion fruit and orange flours and tartaric acid for grape, blackberry, açaí and plum flours). The total phenolic content was determined by the Folin-Ciocalteu method and the results were expressed as mg of gallic acid equivalent (GAE) per g of sample.

Mineral analysis
For the total mineral content determination, microwaveassisted acid digestion of the fruit flours was performed using an MLS 1200 Mega Milestone (Sorisole, Italy) microwave digestion system equipped with an HPR-1000/10 S rotor. Approximately 250 mg of homogenized sample were directly weighed into a microwave oven PTFE vessel and 3.0 mL of nitric acid (HNO 3 69% w/v, TraceSELECT®, Fluka, France) and 1.0 mL of hydrogen peroxide (H 2 O 2 30% w/v, TraceSELECT®, Fluka, Germany) were added. The vessels were closed, and a microwave oven program was performed as follows: 1 min at 250 W, 2 min at 0 W, 5 min at 250 W, 5 min at 400 W and 5 min at 600 W. After cooling, the sample solutions were diluted to 25 mL with ultrapure water (> 18.2 MΩ.cm at 25 °C) in a volumetric flask. Each sample was digested in triplicate. The same procedure of acid digestion was applied to the certified reference material (CRM) BCR 679 to confirm the accuracy of the procedure (Supplementary Table S1).
The simulated in vitro digestion was performed according to INFOGEST standardized method [18]. The assays were performed in triplicate for each sample. Ten blanks were also performed to correct samples results. After the simulated gastrointestinal digestion, the samples were cooled by immersion in an ice bath and centrifuged at 10,000 rpm for 10 min at 4 °C to separate the soluble bioaccessible fraction from the residual fraction. The supernatants were filtered using 0.45 µm nylon membrane syringe filters and frozen at − 20 °C until further analysis. SSF, SGF and SIF were prepared as described by Minekus et al. [19]. The resultant solutions were submitted to mineral analysis and the bioaccessibility rates were calculated using the equation: The determination of Ca, Mg, Na and K was performed by flame atomic absorption spectrometry (FAAS) using an AAnalyst 200 instrument (Perkin-Elmer, Uberlingen, Germany) operated as per manufacturer recommendations. The other minerals (Fe, Mn, Zn, Cu and P) were determined by inductively coupled plasma mass spectrometry (ICP-MS) using an iCAP™ Q instrument (Thermo Fisher Scientific, Bremen, Germany). The following elemental isotopes (m/z Bioaccessibility(%) = (mineral content in bioaccessible fraction∕total mineral content) × 100 ratios) were monitored for analytical determinations: 57 Fe, 55 Mn, 66 Zn, 65 Cu, 31 P. The elemental isotopes 89 Y, 103 Rh, 159 Tb and 195 Pt were used as internal standards.
Calibration standards for FAAS analysis were prepared from QC Std 3 (100 mg/L) multi-element standard solution (SCP Science, Baie-d'Urfé, Quebec, Canada). Calibration standards for ICP-MS analysis were prepared from a 10 mg/L multi-element standard solution (PlasmaCAL SCP-33-MS, SCP Science). The internal standard solution was prepared by preparing a multi-element solution (10 µg/L of Y, Rh, Tb and Pt in 2% HNO 3 ) by appropriate dilution of single element solutions of Y, Rh, Tb and Pt (1000 mg/L, Sigma-Aldrich, Switzerland, Buchs). All solutions were prepared using ultrapure water. Calibration curve parameters are summarized in supplementary Table S2.

Estimated daily intake of minerals
The Estimated Daily Intake (EDI) of essential minerals was calculated based on the total concentration of each element in the fruit flours (C element ; mg per 100 g) and an average per capita daily consumption of 30 g, using the equation: EDI = (C element × 30)/RDI. EDI was expressed as the percentage of Reference Daily Intake (RDI) for adults [20]. The daily consumption of 30 g of fruit flour was established based on its recommendation as dietary supplement [1].

Statistical analysis
Statistical analyses were performed using Statistica 8.0 software (Statsoft, Tulsa, OK, USA). The descriptive analysis of the data was performed considering the mean, standard deviation, coefficient of variation, minimum and maximum from each variable. Cochran's C test was applied to check the homogeneity of variances and Shapiro-Wilk's test was performed to assess the normality of data. Significant differences (p < 0.05) were analysed using one-way ANOVA and Tukey's post hoc tests when variables followed normal distribution; whereas, non-normal variables were analysed using Kruskal-Wallis test and Dunn's post hoc test. Correlations between the bioaccessibility of minerals and samples composition were analysed using Spearman's correlation coefficients. The statistical significance of the data was assessed at 95% confidence level (p < 0.05). Graphs representing the correlations between variables were obtained by software Past 3.26.

Results and discussion
Chemical composition of the fruit flours Table 1 summarizes the proximate composition, pH, acidity, and total phenolic contents of the studied fruit flours.
Blackberry flour showed the lowest pH (3.2) and the highest acidity (8.1 g/100 g), while plum flour showed the highest pH (6.2) and lowest acidity (0.6 g/100 g). Great variability (P < 0.05) was observed in the contents of total dietary fibre (TDF) (7.5 to 69.7 g/100 g), carbohydrates (4.1 to 74.9 g/100 g), proteins (2.9 to 12.9 g/100 g), lipids (1.0 to 8.1 g/100 g), and ash (1.0 to 7.0 g/100 g), because these flours were obtained from different fruits and from different raw materials (pulp, peel, seeds, or mixtures of these) [11,12]. Moreover, the soil composition and the fruit ripening can also interfere in the flour's composition, even in case of flours obtained from the same fruit type [9].
The total phenolic content (TPC) in the fruit flours samples ranged from 2.9 to 41.0 mg GAE/g. Green banana, passion fruit and açaí flours showed the lowest levels (4.5, 6.3 and 5.4 mg GAE/g, respectively), while blackberry flour showed the highest level (41.0 mg GAE/g), followed by grape and orange flours (20.3 and 17.6 mg GAE/g).

Mineral composition of the fruit flours
The fruit flours showed variable mineral contents, with high concentrations of K, P, Mg and Ca, as presented in Table 2. Carli et al. [19] observed similar levels of these elements in 10 different commercial fruit flours. An Estimation of the Daily Intake (EDI) of essential elements as % of RDI for adults [20] was calculated considering a daily consumption of 30 g of fruit flour, as presented in Table 3. For all fruit flours, this dose does not exceed the daily value for fibre of 25 g per day on a 2000-cal diet for adults, neither the Tolerable Upper Intake Levels of each mineral. Overall, the fruit flours showed a great contribution to the daily mineral requirements, especially Fe, Mn, and Cu. Some of these flours contribute more than 100% of the daily requirements for Cu and Mn and therefore they could be considered an important source of these minerals. Grape flours stood out for their remarkably high Cu content. Considering a daily consumption of 30 g of flour, grape flours would provide 3.9-5.0 mg of Cu, corresponding to about 4-5 times the daily requirement of this mineral, but below the Tolerable Upper Intake Levels (ULs). Regarding the Na content, the fruit flours showed very low contribution to the daily requirements of this element (0-3%). Despite it being an essential mineral, there is a great concern about the excessive consumption of Na due to the risk of increasing blood pressure levels.  The minerals' bioaccessibility in the five samples of passion fruit flours analysed showed coefficients of variation ranging between 5 and 45%. This variation may be related to the differences observed in the proximate composition of these flours in terms of carbohydrate and dietary fibre contents. Green banana flours stood out for low bioaccessibility of Ca (the bioaccessible fraction of this element was below the LOQ). Mg presented the highest bioaccessibility for most samples, which ranged from 62.4 to 107.4%. Ferreira and Tarley [16] also observed high Mg bioaccessibility in green banana flours, which ranged from 84 to 101% for the gastrointestinal phase. Nevertheless, Silva et al. [15] observed lower bioaccessibility of Mg in citrus fruit residues, ranging from 30.0 to 85.4%. Calcium and Fe were the elements with the lowest average bioaccessibility (18.0 and 28.9%, respectively). Bioaccessibilty of Ca ranged from 0 to 57.5%, with passion fruit flours presenting the lowest bioaccessibility and plum flour presenting the highest value. Furthermore, grape flours showed the lowest bioaccessibility of Fe (2.4-2.5%) and the highest value was observed for açaí flour (60.1%). Ferreira and Tarley [16] observed bioaccessibilities ranging from 24 to 34% for Ca and from 21 to 35% for Fe in green banana flours.

Bioaccessibility of minerals in fruit flours
The low bioaccessibility of some minerals in plant foods may be related to the presence of compounds considered anti-nutrients, such as tannins, phytic acid and oxalic acid, which form insoluble compounds with these elements. Lakshmi and Kaul [16] observed significant correlations between the contents of these compounds and the bioaccessibility of Ca, Fe and Zn in watermelon seeds. The relationship between the chemical composition of fruit flours and mineral bioaccessibility was assessed using Spearman's correlation. The bioaccessible fraction of Mg (MgBio) showed a strong negative correlation with total dietary fibre (TDF) contents (r = − 0.77). This may be related to the presence of negatively charged polysaccharides, which can form complexes with some minerals by electrostatic interaction [15].
The MgBio also showed negative correlation with lipids content (r = − 0.46) and acidity (r = − 0.48). According to Lopez et al. [21], some bivalent metals can bind to molecules of unsaturated fatty acids forming insoluble salts, which decreases the bioaccessibility of these elements. Etcheverry et al. [22] reported that the bioaccessibility of Mg can be influenced by fibre, protein, and phosphorus contents. In this study, no significant correlations were observed between mineral bioaccessibility and protein contents.
The bioaccessible fraction of Ca (CaBio) showed a negative correlation with ash contents (r = − 0.65). This fact may be related to a possible inverse correlation between the CaBio and the levels of K and Mg since these elements are the main ash components in the evaluated samples. According to several authors, CaBio in foods is influenced mainly by the contents of phytate, besides dietary fiber, tannin and oxalate contents [16].
The results also indicated an inverse correlation between the bioaccessible fraction of Cu (CuBio) and the total phenolic contents (TPC). As well as phytates, phenolic compounds tend to form insoluble complexes with metallic cations, especially Fe, Zn and Cu, which decrease the bioaccessibility of these elements [22].
In this work, we observed a negative correlation between MgBio and Mn and the Fe contents (r = − 0.68 and r = − 0.60, respectively). The CaBio showed a strong negative correlation with Mg and K contents (r = − 0.80) and weak negative correlation with Mn contents (r = − 0.49). The bioaccessibility of Fe (FeBio) also showed a weak negative correlation with Mn contents. Besides that, negative correlations between ZnBio and FeBio and the Cu contents were observed (r = − 0.72 and r = − 0.53, respectively). However, only the CuBio showed a weak negative correlation with P content (r = − 0.46), which can be explained by the formation of insoluble phosphates with this element [23]. This correlation could also be related to the presence of phytates since most of total phosphorus in vegetables is present as phytic acid [23]. However, the inhibitory effect of phytic acid on the bioaccessibility of Cu has been questioned by some authors, since these compounds tend to bind more easily to the elements Ca, Fe and Zn [21,23]. Nevertheless, Ferreira and Tarley [16] observed significant negative

Conclusion
The commercial fruit flours evaluated in this work presented high variability on nutrients composition, namely, dietary fibres, carbohydrates, phenolic compounds, and minerals. All these flours are good sources of fibres and phenolic compounds, which makes their use as functional ingredients relevant. Moreover, an average per capita daily consumption of 30 g of fruit flour provides a relevant contribution to the daily mineral requirements, especially Mg, Fe, Mn, and Cu. However, some of these elements showed low bioaccessible fractions (mainly Ca and Fe), depending on the type of fruit flour and its proximate composition, which have a significant impact on minerals bioavailability. Even so, these flours proved to be important sources of fibre and minerals, specially, Mn and Cu, whose concentrations in some flours contribute more than 100% of the reference daily intake. Funding Open access funding provided by FCT|FCCN (b-on). This research was supported by AgriFood XXI I&D&I project (NORTE-01-0145-FEDER-000041) cofinanced by European Regional Development Fund (ERDF), through the NORTE 2020 (Programa Operacional Regional do Norte 2014/2020).
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.