Materials

The grain of five varieties of beans, two types Flor of Mayo, two bayo and one peanut, all released by the National Institute of Agricultural and Livestock Forestry Research (INIFAP). The multiplication of the genotypes was carried out in Santa Lucia of Prias, Texcoco, Mexico, between June and September 2015. The varieties were: Flor of Mayo M-38 and Flor of Durazno, both of the Flor of Mayo type; Bayo Mecentral, and Bayomex, of the Bayo type, Cacahuate-72 of the peanut type, all of them Phaseolus vulgaris L.

Methods

Determination of moisture content. The Steinlite equipment was used, using samples of 100 g of beans. The humidity of the samples was 12% at the beginning of the experiment.

Accelerated aging test. Samples of 100 grams of seed were placed in containers at 75% RH, containing a saturated solution of sodium chloride (NaCl). The containers were housed in an incubator at 40 °C. After 28 days the samples were removed and left at room temperature to decrease the moisture level to its initial value of 12%. These conditions allow to observe changes in the grain that contrast the response of the genotypes (^{Jacinto et al., 2006}). A batch of each variety, without storage, was kept as a control in polyethylene bags at 5 ºC, until the moment of analysis

Likewise, samples of the same varieties were obtained, coming from a cellar at a temperature of 24 ±4°C and a controlled cycle of 12 h light and 12 darkness, which is located in Texcoco, Mexico. The varieties were stored for two and four years respectively and quality data were available in their fresh state; that is, freshly harvested.

In the genotypes subjected to the accelerated aging test and the prolonged storage in the cellar, the quality parameters of the grain were determined.

The weight and volume of 100 grains, as well as the CAA and the cooking time were determined according to the methodology described by ^{Guzmán et al. (1995)}.

The color determination of the testa of the genotypes was carried out with a Konica-Minolta reflectance spectrophotometer model CM-5. The color readings were made in two repetitions per genotype, both in control samples and in the seeds obtained from the accelerated aging test and in the two and four years of prolonged storage. The values L, a and b were obtained.

Where: L^*= represents the relative clarity of colors on a scale of 0 to 100; a^*= comprises a scale of -100 to 100, where the negative numbers indicate green tones and the positive ones red; b^*= extends on a scale of -100 to 100, where negative values indicate blue and positive yellow tones.

With the results, an analysis of variance was made based on a factorial design of treatments, whose factors were 5 varieties and three storage conditions. For comparison of means, a Tukey test (p< 0.05) was applied and a matrix of correlations was also generated by the SAS statistical package (^{SAS, 2001}).

The grain of the newly harvested genotypes showed varietal differences in quality

The five recently harvested varieties of beans showed significant differences in the variables of color L^*, a^* and b^*, weight and volume of the grain, CAA, the cooking time, the number of open grains and the amount of solids in the broth of cooking (p> 0.01). These differences in quality between varieties are frequent because genotype-environment interactions are generally present (^{Kigel, 1999}).

The color of the testa and the hardening of the grain were the main changes in the genotypes due to accelerated aging

Accelerated aging induced changes (p< 0.01) in the clarity of the testa (L^*), the red tones (a^*) and the yellow tones (b^*), as well as in the CAA and in the cooking time. There were also highly significant differences (p< 0.01) in the amount of solids in the cooking broth, due to the interaction of variety*storage condition. The solids content is an indicator of the density of the broth, which is associated with the palatability of the broth.

In the aged genotypes the darkening of the testa was observed, which was identified with the decrease of the L^* value in comparison with its corresponding recently harvested control. The decrease was 17.4 units in Cacahuate-72 and 10 units in Bayo Mecentral. In contrast M-38 and Peach Blossom only decreased 5.6 units and 5.8 respectively.

The value a^* was increased between 1.5 and 6.3 units; the minimum was observed in Flor of Mayo M-38, and the maximum in Bayomex. While the yellow tones (b^*), they increased from 2.4 in Bayomex, to 6.2 in Flor of Durazno. It is evident that changes in the color (L^* a^* b^*) of the testa are associated with the aging process of the genotypes (Figure 1). The darkening of the bay types, in addition to the decrease in the L^* value, was accompanied by an increase in the red tones (a^* higher in aged than in freshly harvested ones). Unlike the behavior of the Flor of Mayo, in which the yellow tones increase. Apparently, an increase in yellow tones obscures the testa as it is the Flor of Durazno case in contrast to Flor of Mayo M-38. The latter showed lower values of a^* and of b^* and, therefore, greater stability in the color of the seed coat compared with Peach Blossom, which agrees with that reported by ^{Jacinto-Hernández et al. (2007)}regarding the stability in the color of Flor of Mayo M-38. Also in Figure 1 a correlation (r= 0.8) is observed between the increase in red tones with the darkening of the testa (L^*).

Figure 1 Changes in head clarity (L^*) and red tones (a^*) due to accelerated aging. 

The cooking time of the control samples of the genotypes was 50 to 70 min. The change produced by aging was highly significant (p< 0.01).

The aging of the grain increased the cooking time a minimum of 1.7 times its original value, in the case of Flor of Durazno.While Flor of Mayo M-38 and Cacahuate-72 showed the maximum hardening; both varieties increased their cooking time 2.4 times (Figure 2). This increase in cooking time occurred without there being limited capacity to absorb water. To this condition in which the grain is able to absorb water, but does not soften during soaking and cooking, it is called “cooking hardness” (^{Kigel, 1999}).

Figure 2 Changes in cooking time due to accelerated aging. 

It has been reported that the “hardness to cooking” developed during aging, occurs by an association between the denaturation of the reserve proteins and limited hydration of the starch during cooking (^{Liu et al., 1992}). In addition, the cell wall is not softened due to a decrease in the solubilization of pectin (Martínez and ^{Njoroge et al., 2015}).

Since the change in the color of the testa, and that of the cooking time are carried out in the same storage condition, there could be chemical reactions that influence the physicochemical properties of both the testa and the cotyledon. The increase in red tones was correlated (r= 0.75^**) with an increase in the cooking time.

The peanut-72 variety, which showed the highest values of a^* (15.6) also showed the extended cooking time (127 min), while the Bayo Mecentral with a value of a^* of 12.2 showed the short cooking time (113 min). The greater the increase of the red tones in the aged samples, the smaller amount of solids was quantified in the cooking broth (r= -0.78^**).

Regarding the weight and volume of 100 grains, no effect was observed due to accelerated aging. Regarding the CAA, Flor of Durazno decreased by 8%, however, this change was not enough for the aged grain to present “hard test”. The other genotypes increased their CAA between 5 and 20%. Even when the water absorption capacity of the aged samples increased, as reported by ^{Hinks et al. (1987)}, this could be due to the fact that part of the water is not absorbed, but remains between the testa and the cotyledon. There was no correlation between the CAA with the cooking time.

The prolonged storage caused darkening of the head and scarce hardening of the cotyledon

After two years of storage there were highly significant differences (p< 0.01) in L^* in a^*, as well as in weight and volume of 100 grains, cooking time, both between varieties, as a result of the storage condition and in the interaction variety*condition. While in the yellow tones (b^* value), no differences were observed due to the effect of the storage condition, but due to the variety and the interaction variety*storage condition (p< 0.01), likewise in the CAA (p< 0.05). There were highly significant differences in the amount of solids in the cooking broth, due to the effect of the variety (p< 0.01), while the differences due to the effect of the storage condition during two years and the variety interaction storage condition were significant (p< 0.05). After four years of storage, the differences in these variables were highly significant (p< 0.01) for all sourcesof variation.

The darkening of the testa was also associated, although in a lesser proportion with an increase in red tones both after two (r= -0.62^**) and four years (r= 0.67^**).

As the storage time was prolonged, the darkening of the grain was greater. After four years of storage the darkening of the testa was in a range between 8.7 and 13.8 units of L^*. The testa of Cacahuate-72 was the one that showed the greatest darkening. No significant differences were detected in the a^* value; however, it was observed that while bay types increased between 4.1 and 4.9 units, Flor of Mayo increased between 1.5 and 2.2 units. While in the yellow tones (b^*) these same varieties increased between 0.55 and 1.5, in the Bayo varieties no significant differences were detected.

The Flor of Mayo varieties after four years of storage showed a decrease between 20 and 55% in the CAA, which is identified as a “hard test” problem. ^{Kigel (1999)}, describes that this behavior can be due to a low permeability of the testa, which according to ^{Agbo et al. (1987)} may be related to the size of the micropyle, in addition to other differences in the microstructure, which are related to changes in the permeability of the testa. The CAA of the varieties was correlated with the cooking time (r= -0.65).

Flor of Mayo M-38 and Bayomex were the genotypes that showed the greatest increase in cooking time (14%) after four years of storage. Bayo Mecentral, although it did not show significant difference in the cooking time compared to the recently harvested beans, was the variety that showed the greatest decrease in the solids in the cooking broth. While Flor of Durazno was the one that showed the least change. This confirms that the culinary quality and its speed of deterioration during storage are characteristics related to the cultivars (^{Proctor and Watts 1987}; ^{Kigel, 1999}). It is then possible to take advantage of genetic variability in the propensity to deteriorate quality during storage between genotypes to obtain relatively stable varieties.

Comparison of changes in induced by accelerated aging and those occurring during prolonged storage

Similar changes occurred in the color of the seed coat when the genotypes were subjected to accelerated aging conditions and when they remained for a long time in storage. A correlation was detected (r= -0.75^*) between grain darkening (decrease L^*) due to the effect of accelerated aging and the increase of red tones after four years of storage (Figure 3).

Figure 3 Association of the decrease in the L^* value due to accelerated aging with the increase in red tones (value a^*) during prolonged storage. 

The value of b^*, which was increased by the effect of accelerated aging, was correlated with the value of b* after two (r= 0.95^**) and four years (r= 0.87^**) of storage (Figure 4). It was also observed that the darkening of the testa value of L^* after two years of storage (r= 0.8^**).

Figure 4 Changes in yellow tones (b*) promoted by accelerated aging that are also observed during prolonged storage. 

In accelerated aging, there was no significant change in the CAA of the varieties, nor after two years of storage. Only Bayo Mecentral showed a decrease in its capacity to absorb water during the soak, after two years of storage it absorbed 31% less than the control and also due to the effect of accelerated aging, it decreased its CAA by 29%. After four years, highly significant differences (p< 0.01) were observed in the CAA of the varieties.

During accelerated aging the cooking time increased on average by 114% while after storage for four years the average increase was only 14%. Prolonged storage, under conditions of high T and high HR, promoted the cooking hardness defect, which may affect the loss of nutrients due to the excessive heating required to complete the cooking. While during prolonged storage was not observed.

The cooking time after accelerated aging correlated with a higher level of red tones (r= 0.66^*) after four years, and was also associated with a higher solids content in the broth. The solids content after aging treatment was correlated (r = 0.84^**) with that detected after four years of storage. As expected, there were correlations between the quality characteristics after two and four years of storage. As more was darkened in grain after two years (decrease L^*), the cooking time was longer after four years (r= -0.92^**); When the value a^* (red tones) was higher after two years of storage, lower CAA was observed in the genotypes after four years (r= -0.82^**).