Biofortification potential in common bean (Phaseolus vulgaris L.): bioactive compounds, antioxidant capacity and physicochemical properties of 155 varieties grown in México

Common bean has been classified as an almost perfect food due to its high content of fiber, protein and bioactive compounds and its high antioxidant capacity, which has been attributed to having prevent diabetes, oxidative stress, bowel inflammation and cardiovascular diseases. The objective of this research work was to characterize the physicochemical properties and bioactive compounds of 155 varieties of common bean (Phaseolus vulgaris L.) produced in Mexico with potential to be biofortified. The contents of protein, fiber, fat, carbohydrates and energy were analyzed by the methods established by the AOAC, in addition the antioxidant capacity (DPPH), the content of total phenols, flavonoids, anthocyanins and phytic acid were obtained. Cluster, correlation, and principal component analysis were conducted. A total of 14 outstanding varieties was identified, where four varieties stood out about phenol and flavonoid content. A second group comprised by 10 varieties had high protein and anthocyanin levels and showed the second highest content of phenols and flavonoids. The diversity of beans obtained in outstanding varieties provides options for biofortification programs based on the consumption or place of origin of each variety.


Introduction
At present, changes in demand of food and trends to consume products containing beneficial compounds for health, in addition to a promotion towards plant based products, has forced the industry to find new dietary attributes in its products, so they may have the capacity of attracting new market preferences without neglecting the nutritional quality of the crops, being one of the most important parameters that must 2 be taken into account when talking about food, since they determine their functionality in the diet (Sida-Arreola et al., 2015;Gutiérrez, 2012).
Legumes, such as beans, lentils or chickpeas, have been considered as staple foods, due to their nutritional quality and because they have shown the ability to improve human health. Within this group, beans are the most consumed legume in the world, having an average per capita of 6.6 Kg. Around 150 species of beans are known, of which approximately 70 varieties have their center of origin in Mexico with a great variety of sizes and colors (FIRA, 2016;Silva-Cristobal et al., 2010).
Among the nutritional properties of beans, one of the most notable is its high protein content and to a lesser extent its contribution of carbohydrates, vitamins and a high contribution of fibers and minerals such as calcium, iron and zinc, beans are also rich in antioxidants due to its high phenol content (Korus et al., 2007;Ulloa et al., 2011). The protein content in beans ranges between 15% and 30%, a 90 g portion of beans provides from 7 to 8 g of protein, which constitutes 15% of the recommended daily intake, they also have enough content to meet the needs daily basic amino acids lysine (6.4-7.6 g. 100 g -1 of protein) and phenylalanine plus tyrosine (5.3-8.2 g. 100 g -1 of protein) (Mederos, 2006;Suárez-Martínez et al., 2016). Beans are also a good source of fiber with values that vary from 14 g to 19 g per 100 g of raw food, of which it can reach 50% soluble fiber, which has been related to a reduction in cholesterol levels (Messina, 2014).
The common bean has been classified as an almost perfect food, due to its high content of fiber, protein, its high antioxidant capacity and the large number of bioactive compounds it has, to which qualities have been attributed in the prevention of diseases such as diabetes, oxidative stress, intestinal inflammation and cardiovascular diseases (Xu and Chang, 2009;Câmara et al., 2013). These effects are related to the high antioxidant capacity, granted by high concentrations of phenols, flavonoids and to a lesser extent anthocyanin and phytic acid, which in turn generate antimutagenic and antigenotoxic effects (Akond et al., 2011 1 ). However, Suárez-Martínez et al. (2016), suggest that it is necessary to make efforts to increase bean consumption, making the population aware of the benefits attributed to it. Based on the above, the objective of this research work was to characterize the physicochemical properties and bioactive compounds of 155 varieties of beans produced in Mexico with potential to be biofortified.

Materials and Methods
Selection of material One hundred and fifty-three varieties of beans (Phaseolus vulgaris L.) were used, representative of all Mexico, based on the consumption preference of the States of Chihuahua, Durango, Zacatecas, Sinaloa, Tabasco, Yucatán, Oaxaca, Veracruz, Puebla, Jalisco, Morelos, Guanajuato, Michoacán and State of Mexico, in addition to two more samples from Cuba selected for their nutritional quality.

Assessed variables
Once the bean materials were obtained, they were cleaned, selected and milled to perform the analysis of physicochemical properties and bioactive compounds, which were determined in accordance with the 3 Association of Official Analytical Chemists (AOAC 2000) and in accordance with the provisions of the Official Standards Current Mexicans. The parameters evaluated were the following: Determination of physicochemical properties Protein Protein determination was performed using the Kjeldahl method of the Association of Official Analytical Chemists (AOAC, 2000). 1 g of homogenized sample was weighed into a Kjeldahl digestion flask, 3 boiling beads, 10 g of potassium sulfate and 20 ml of concentrated sulfuric acid were added for digestion. When the mixture became clear, it was allowed to cool, and 200 ml of tri-distilled water was added. It was connected to the still while adding 100 ml of 30% sodium hydroxide, distilled and at least 150 ml was collected in a flask containing 50 ml of 3% boric acid and titrated with 0.1 N hydrochloric acid. The results obtained were multiplied by the factor 6.25 and expressed as a percentage.

Total fat
The determination of the fat of the bean samples was carried out using the Goldfish method (AOAC, 2000). Goldfish flasks were prepared by oven drying to constant weight. The LABCONCO equipment was assembled and the sample was placed inside filter papers and covered with cotton and introduced into the equipment. Solvent (petroleum ether) was added and refluxed for 2.5 hours. After the extraction was complete, the solvent was recovered by distillation, leaving only the fat in the flask. Finally, the flask with the residue was weighed and the fat percentage was determined according to the Mexican Standard NMX-F-427-1982 published in the official gazette of the federation by the government secretary.
Crude fiber Crude fiber was determined by the Mexican Standard NMX-F-90-S-1978 published in the official gazette of the federation by the government secretary. From the sample that was previously degreased. The weight of the sample was recorded and transferred to the fiber cups, then 200 ml of 1.25% sulfuric acid with 1 ml of isoamyl alcohol as defoamer was added to each beaker. The mixture was left to boil for 30 minutes, at the end, rinses were carried out to eliminate the residues of the reagents and at the same time neutralize the mixture, subsequently, 200 ml of 1.25% sodium hydroxide was added to the sample and it was allowed to boil for another 30 minutes, at the end, fiberglass rinses were performed until neutrality was reached. Subsequently, the fiberglass with the sample was placed in the capsule and they were introduced into the oven, it was left to dry for 12 hours to ensure that the sample was perfectly dry. After drying, the capsule was weighed with the glass fiber and the sample. By difference in weight, the percentage of fiber contained in each of the samples was determined.
Total carbohydrates Determination of total carbohydrates was carried out by difference of the protein, fiber, fat, moisture and ash content and was reported in percentage.

Metabolizable energy
Energy contained in each of the samples was measured by the sum of the calories contained in carbohydrates, fat and protein, calculated using the Atwater system, reported by Acheson et al. (1980) Energy was expressed in Kcal.
Extraction of bioactive compounds 1 g of raw beans was macerated with 10 ml of 80% methanol and centrifuged at 2560 RCF for 10 minutes at 4 °C, then the aliquot was extracted to measure the following four parameters.

Antioxidant capacity
For the determination of the antioxidant activity, the DPPH method described by Brand-Williams et al. (1996) was used, for which 500 μl of each sample was taken and 2.5 ml of DPPH 0.1 mM prepared in 80% methanol was added to each one. All the reactions were carried out in one hour and at room temperature and in the absence of light; after the time the absorbance was measured at 517 nm. Antioxidant activity is expressed as percentage of inhibition corresponding to the amount of DPPH radical neutralized by the extract at a certain concentration according to the following equation.

%I =
A − A A * 100 Where: A is the absorbance of the blank, A1 is the absorbance of the sample and %I is the percentage of inhibition Total phenols The total phenols were determined by colorimetry against a gallic acid standard curve using the Folin-Ciocalteau method (Peyrat-Maillard et al., 2000); 250 µl of sample added with 750 µl of sodium carbonate 2 %; subsequently, 250 µl of reagent Folin-Ciocalteau 50% were added and filled to the mark with distilled water to make up the volume of 2.625 ml, finally, it was vortexed and incubated at room temperature for 60 min; absorbance was measured at 725 nm. Results were expressed in milligrams of gallic acid per gram (mg GA g -1 ).

Flavonoids
Flavonoid analysis was performed following the method of Zhishen et al. (1999), 250 µl of the sample was taken and 75 µl of sodium nitrite 5% was added, it was vortexed, and the reaction was allowed to settle. for 5 minutes, once the time had elapsed, 150 µL of 10% aluminum chloride were added together with 500 µL of sodium hydroxide, finally, it was added with distilled water to make a volume of 3 ml. The samples were incubated for 45 minutes and then absorbance at 510 nm was measured; flavonoids were quantified based on a standard catechin curve and the results were expressed in mg of catechin per gram of dry weight (MgCE.g -1 ).

Monomeric anthocyanins
To determine the content of monomeric anthocyanins, the differential method by pH described by Wrolstad (1976) was used. The reaction was divided in two phases.
The first consisted of taking 0.5 ml of sample and adding 2 ml of 0.25 M potassium chloride adjusted to a pH of 1.0, then it was vortexed and the absorbance was measured at 460 nm. For the second phase, 0.5 ml of sample was taken and 2 ml of 0.4 M sodium acetate adjusted to a pH of 4.5 were added, and the same process was carried out as in phase one, with the difference that it was measured at a wavelength of 710 nm. The monomeric anthocyanin content was calculated using the following formula: A − B * 449.2 * 0.2 * 1000 26900 Where A-B is the difference between absorbances between phase 1 and phase 2, 449.2 is the molecular weight of cynadine-3-glucoside, 0.2 the dilution factor, 1000 is a constant and 26900 is the coefficient of molecular extinction of cynadine-3-glucoside.
The results were reported as mg of cynadin-3 glucoside per gram of sample (mg C3G g -1 ).
Phytic acid For the determination of phytic acid, the method proposed by McKie and McCleary (2016) was used, for which the phosphorous total was obtained by the ammonium metavanadate method in an absorption range of 430 nm against a calibration curve of potassium phosphate. Once the results were obtained, the following formula was applied: 5 Phytic Acid = Total phosphorous 0.282 Where 0.282 is a conversion factor from total phosphorous to phytic acid. Results were expressed in grams of phytic acid per each 100 g of dry weight of sample.

Statistical analysis
A Pearson`s correlation analysis, as well as a cluster analysis of K means and a principal component analysis using the SPSS statistical package (SPSS, 2011).

Physicochemical properties
The selected bean varieties showed a large diversity size, color and probably due to the differences in geographical origin (Table 1). Mean protein content was 23.16%, which is over the one reported by the U.S. Department of Agriculture (21-22%) and with 11.61% of the varieties evaluated over the range (15-26.35%) found for Mexican varieties of common beans (Chávez-Mendoza and Sánchez, 2017;USDA, 2018). In addition, very similar data were recorded for raw bean flour, to those reported in previous works for the variables of metabolizable energy, fat and carbohydrates, while for the fiber values the range found (1.22-7.82%) is very much below the general fiber average for bean varieties worldwide (Campos-Vega et al., 2009;Suárez-Martínez et al., 2016).  (2006) for 62 varieties of common bean. On the contrary, the content of phytic acid obtained an average of 1.45 mg.g -1 , which places it 0.08 mg.g -1 below the average published by Iniestra-González et al. (2005) in 16 bean varieties of different seed coat colors.

Correlation analysis
In Pearson's correlation analysis, protein and fiber content parameters showed a negative correlation with carbohydrate content and with metabolizable energy (Table 2), which could be explained by the carbohydrate calculation method, which in turn is the parameter that most contributes to the calculation of metabolizable energy. A positive correlation was also found between said parameters with antioxidant capacity, while for bioactive compounds a high correlation was found between the content of phenols with flavonoids and anthocyanins, in addition to both being correlated with antioxidant capacity. In the case of beans, this results from the fact that antioxidant activity is related to the phenolic compound content they possess (Huber et al., 2016;Gan et al., 2017;García-Díaz et al., 2018). Finally, phytic acid showed a negative correlation with carbohydrates. **. The correlation is significant at the 0.01 level (bilateral). *. The correlation is significant at the 0.05 level (bilateral).

Cluster analysis
The cluster analysis (Table 3) was performed based on the physicochemical properties and the content of bioactive compounds, obtaining as a result five groupings based on their similarities, from which clusters 2 and 3 were selected because the varieties that the conglomerates presented the highest values for protein, fibre and anthocyanins (cluster 2) as well as the highest phenol and flavonoid values (cluster 3). Phytic acid content was not taken into account when selecting the outstanding groups, because its capacity as a beneficial or antinutritional compound does not depend on the content or the source that comes from Rather, it depends 7 on the diet followed by the consumer, since in a diet based on grains and cereals with a low mineral protein intake, phytic acid can act negatively, on the contrary, in a balanced diet with micronutrients availability, the phytic acid expresses its value as a bioactive compound (Díaz-Batalla et al., 2006). We obtained a total of 14 outstanding varieties (Table 4), where 4 varieties stood out in terms of their content of phenols and flavonoids, a second group composed of 10 varieties with high levels of protein and anthocyanins. Moreover, this group showed a low percentage of carbohydrates and a high content of fibre and fat, which makes these varieties future foods with a low glycaemic index (American Diabetes Association, 2015). A great diversity can be observed within the outstanding group, where there are varieties from different regions of the country and different shapes, sizes and colours of the bean grain. The outstanding varieties, according to the cluster analysis, showed great diversity, 'FIRA (2016)', indicate that beans are located in three large consumption groups: highly preferred ('Azufrado', 'Mayocoba', 'Negro Jamapa', 'Peruano', 'Flor de Mayo' and 'Flor de Junio'), preferred ('Garbancillo', 'Manzano', 'Negro' and 'Pinto Americano') and non-preferred ('Alubia', 'Bayo', 'Negro Zacatecas', 'Ojo de Cabra' and 'Pinto Mexicano'). Therefore, within the results obtained, there is at least one variety of the most consumed beans within the national territory. Also, five varieties of black and pinto beans were obtained, which together contribute more than 60% of the national production (Lara-Flores, 2015;FIRA, 2016).

Main components analysis
On the other hand, the principal component analysis had information on 6 variables that included bioactive compounds, protein content and carbohydrate content ( Table 5). The first three components explained 74.75% of the total variation: first component 31.09%, second component 24.48% and third component 19.18%. In the first component, the most significant variables were the content carbohydrate, phenols and flavonoids. In addition, there was a contrast in the protein content with carbohydrates. In the second component, there was a contrast between the bioactive compounds (phenols, flavonoids, anthocyanins and phytic acid) and the protein content with the carbohydrate content. While in the third component, he contrasted the concentration of phytic acid with that of anthocyanins. The spatial distribution of varieties was made based on the first 2 components (Figure 1), it can be observed in quadrant I (upper-right) a clustering trend based on the high concentrations of the bioactive compounds evaluated, in turn quadrant IV (upper-left) groups varieties with high concentrations of bioactive compounds and also adds varieties with high levels of protein. Finally, in quadrants II and III (Lower-right and lower-left) those varieties were found with a high content of carbohydrates, but which did not stand out in the levels of the other variables evaluated.  10 Conclusions Fourteen outstanding varieties were obtained, based on the variables evaluated, where 4 stood out for their high levels of phenols and flavonoids; On the other hand, 10 more varieties stood out for a high level of protein, in addition to acceptable levels of anthocyanins, phytic acid and a low content of carbohydrates. Varieties such as 'Frijolón 1', 'Frijolón Amarillo', 'Frijolón negro' and 'Sangre de toro' were remarkable due to their high content of bioactive compounds, so it is recommended to incorporate these varieties to a biofortification program to improve the nutritional quality of beans. The other 10 remaining varieties, due to their high protein content, are recommended to be used as an alternative of vegetal protein with a good glycemic index. The diversity of beans obtained in the outstanding varieties generates options for biofortification programs based on the consumption of each variety or its place of origin.