Introduction

Triticale (X. Triticosecale Wittmack), a cereal of wheat and rye hybridization, is mainly used as animal feed, in recent years the interest for its use in food production has increased (^{Zhu, 2018}), due to its genetic variability and grain nutritional composition, which has allowed the development of food and beverage products, including bakery products such as biscuits, pasta, breads and malts. Little research has been done on the nutritional composition and various food uses of triticale due to the wide variation in the chemical composition of the grain, suggesting that it has potential as a cereal alternative for various food applications (^{Zhu, 2018}).

Grain and flour are a source of vitamins and minerals (^{Pruska et al., 2017}), it also has more lysine and protein than wheat varieties, which is valuable because it is an essential amino acid that the body does not synthesize, therefore, it is a product of interest for human diet, although its characteristics, such as texture and grain filling, are unfavorable during commercialization, since its flour yield is lower than wheat. Flour extraction from triticale grain can average 65% (^{Riasat et al., 2019}), although it is highly variable depending on the variety, management and harvest conditions.

Triticale can be considered for human diet, but the yield and quality of flours should be increased (^{Oliete et al., 2010}). In Mexico, the selection of genotypes for milling is scarce, because studies have focused on the production of forage and grain; however, ^{Ammar et al. (2004}) point out that yield, adaptation and industrial quality must be increased.

From the predictive point of view of selection, it is important to know if selecting a characteristic modifies another that is correlated with it since especially negative ones can neutralize the efforts of the breeder (^{Pattison et al., 2014}). This reflects the importance of the crop in the food industry and the need for varieties of greater and better benefits that meet the needs and requirements of the industry. Therefore, the objective of this research was to identify triticale lines based on the physical and chemical properties of the grain with value for the food industry.

Materials and methods

Results and discussion

The results of Anova (Table 2), show differences (p< 0.01) between lines for all variables, the effects that contributed to the total variation in HW were 61.29%, WTG (60.05%), because the expression of weight was mainly attributed to the genetic variation of the evaluated lines, as confirmed by the results of ^{Jonnala et al. (2010}), who found differences in WTG and HW in advanced triticale lines, which indicates that the genotypes provide different quality in the grain due to their genetic constitution; other effects that occurred in fiber, proteins and fat showed greater contributions with 77.05, 62.83 and 52.73%, respectively, which shows that the proportion of quality in the grain is influenced by genetic factors of the lines. These results are consistent with those obtained by ^{Negash et al. (2013}), who assessed quality in wheat grains and found that the most important effects were protein, carbohydrates and fiber.

Lines L-18, L-17, L-14, L-19 and L-20 showed numerically the highest averages in WTG and HW (Table 3); however, in WTG, L-18 only exceeded L-3 and L-5 and, in HW, L-20, L-17, L-1 only exceeded L-1 and L-5, so the grain weight and hectoliter weight are considered indicators of the grain yield potential of varieties or lines, in this case L-14, L-17 and L-20 also presented the lowest contents of moisture in grain; that is, a greater accumulation of dry matter and HW.

^{Aisawi et al. (2015}) mention that grain morphology is negatively altered by late sowings, nitrogen deficiency, deficiency in water supply and grain filling due to high or low temperatures. Therefore, the results obtained are based on the genetic differences of the lines, as well as the favorable environmental conditions in which the lines were sown in the field (December 18, 2017).

The results indicate a good grain filling during the cycle, which provides good flour yield during the milling process in the food industry. These results are consistent with what was reported by ^{Pomortsev et al. (2019}); ^{Giunta et al. (2017}) reported that higher grain weight is strongly affected by genetic and environmental factors; the results showed that, within the total variation of the values of these variables, it was due to the genotype-by-environment (GxE) interaction, which shows predicting the behavior of the lines when evaluated.

The moisture content is of great relevance because moisture contents greater than 13-14% favor the damage caused by the presence of microorganisms during temporary storage, consequently it affects the quality of flours in the production of food products. The moisture in lines L-20, L-17, L-14 and L-14 was satisfactory and reliable for processing in the production of biscuits because they were less than 13%, the content of minerals (ash) in the grain is important because, if their concentration is high, especially in grains with low hectoliter weight, it may not be favorable in the food and animal industry (^{Riasat et al., 2019}).

Low ash levels are favorable because they can provide more minerals for the diet, on the other hand, when there are high levels of ash, they are particularly undesirable because they darken semolina and, to a greater degree, food pasta (^{Pattison et al., 2014}). The concentration levels of ash in grain that are considered desirable should be less than 2% (^{Peña et al., 2007}), the results showed that lines L-17, L-18, L-19 and L-14 presented values less than 2 in ashes, which makes them important for the contribution of minerals.

In relation to the fat content, variation was found between the lines, within the group with the best response, lines L-17, L-18 and L-20 showed higher values and were higher than L-1, L-3, L-7, L-8, L-13, L-14 and L-19; in general, cereals have low amounts of lipid compounds, triticale is between 1.5 fats (^{Peña et al., 2007}), which are mainly present in the germ and aleurone layer of the grain. Therefore, only the lines of triticale L-3, L-8, L-13 and L-19 presented reduced values and the rest are within the range known for the fat content in the cereal, in turn ^{Pruska et al. (2017}) mentioned that those cereals with higher fat content should be considered to extend their use in human diet, at the same time increase crop yields and the quality of their flours, since fats represent the main source of energy from foods, as is the case of the concentration that some cereals, such as wheat and oats, have in the grain.

In the case of fiber, the highest numerical values were shown by lines L-18, L-20 and L-17, but the most important statistical group was also made up of L-15, L-13 and L7, the range of this variable is 3.1 in triticale (^{Peña et al., 2007}), so the 20 lines represent a good source of fiber for humans, who are the ones that need the most for digestion (^{Moiraghi et al., 2011}). The results showed values from 3.3 to 3.94, higher than the range established in triticale (3.1).

Lines L-17, L-18, L-20 and L-2 produced numerically the highest concentrations of proteins, but in the same group were L-12, L-10, L-15, L-5, L-7, L-8, L-13 and L-3; in this regard, ^{Peña et al. (2007}) report that triticale usually has a protein content of 14.3%, so only lines L-17, L-18, L-20 and L-2 are considered with desirable protein contents to make pasta and cookies due to their strong gluten, since they also exceeded by 4.61 the protein content of wheats (12%), ^{Gulmezoglu et al. (2010}) reported that triticale contains a good source of proteins, carbohydrates and fats that can be used in the animal diet and in processes for the food industry for the processing of pasta, biscuits or breads with own flours or mixtures with flours of other cereals.

In relation to the carbohydrate content in the grain, it was observed that in lines L-17, L-18, L-14, they presented higher numerical values of carbohydrates, but when considering the values obtained by ^{Jonnala et al. (2010}), lines L-9, L-8, L-4, L-13, L-16, L-13, L-6, L-11, L-1 and L-20, lines L-10, L-15, L-5, L-19, L-7, L-12 and L-2 were also higher; nevertheless, as they are classified within the group with the highest values, they are not considered desirable because they do not meet the necessary requirements. ^{Pattison et al. (2014}) establish that foods rich in carbohydrates are present in seeds, pasta, breads, cookies and tubers, among others, they are important since they represent 55% of the total foods in the diet of developed countries and more than 80% in underdeveloped countries, so the lines with higher percentages of proteins, carbohydrates and fibers have the potential to be used in the production of food formulas for humans and animals (Jonnala et al., 2010).

The cluster analysis classified the 20 lines of triticale into four groups with similarity based on the physicochemical characteristics of the grain (Figure 1), the first group was formed by lines L-17 and L-18, they provided higher WTG and HW and exceeded the general mean; however, for ashes, proteins, fiber and fat, their values were higher than what was reported by ^{Peña et al. (2007}), being acceptable for the milling industry, especially for the production of pasta, biscuits, macaroni and soups, ^{Pattison et al. (2014}) mentioned that a good quality in triticale grains should be when they present values higher than 14.3% in protein, less than 2% of ashes and greater than 1.5% of fat, the results of this research significantly exceeded that reported by ^{Peña et al. (2007)}.

Figure 1 Dendrogram generated based on the contents of physicochemical parameters of 20 elite lines of spring triticale, Celaya, Guanajuato, Mexico. 

The second group was formed by 40% of the lines, integrating L-10, L-7, L-20, L-5, L-19, L-15, L-12 and L-2, this group showed values lower than the general mean of WTG and HW, these parameters indicate the density and degree of filling of the grain, mainly for the morphology of the grain characteristic of the variety, normally, when the grain is not completely full, the hectoliter weight values are low, in this research the values presented in HW and WTG were 5.14 kg hl^-1 higher than what is reported in wheat (70 kg hl^-1), which ^{Peña et al. (2006)} report.

It should be noted that the field experiment was carried out from December 2017 to April 2018, which favored grain filling, the third group contributed 25%, formed by lines L-13, L-8, L-16, L-9 and L-4, their characteristics are that they showed 0.26 kg hl^-1 in HW, lower than the general average (75.14 kg hl^-1) and 2.6 g in WTG, these results are not significant for them to be used in the milling industry since they exceeded that reported in wheat by 4.8 kg hl^-1.

However, the average values of moisture, ashes, fat, and fiber were higher than those reported by ^{Pattison et al. (2014}) in wheat, with 13, 2, 2.62, and 3.58%, respectively. ^{Castaño et al. (2015}) report that the milling industry produces three grades of flour quality, related to their physicochemical properties: common or standard, fine and extrafine, which constitute the basis for the preparation of breads, tortillas, cookies and cakes, where semolinas may vary slightly in their grade of fineness due to the quality of flours of the genetic material to be used in the preparation of pasta (spaghetti, macaroni, soups), the results in this research are indicators that the lines in this group can be used in the preparation of cookies, pasta, cakes and macaroni due to the gluten, with the exception of the content of proteins and carbohydrates, which, despite having low values, are still favorable to consider them in the baking industry (^{Zhu, 2018}).

The fourth group represented 25% and was made up of L-14, L-11, L-6, L-3 and L-1, whose WTG and HW values were 1.83 and 0.43 kg hl^-1 higher than the average, in terms of moisture their values were less than 13.5%, in ashes only L-14 and L-3 showed values lower than 2%, so they are suitable for milling, in fat and fiber they had higher values, proteins and carbohydrates obtained slightly lower values than those reported by ^{Peña et al. (2007}), of the total food raw material generated by the milling industry, approximately 62% is destined to baking, 26% to biscuits, tortillas and others and 12% to the production of food pasta, these lines are promising for the production of the latter product, these results coincide with what was reported by ^{Jing et al. (2016}), who reported that the functional properties and structure in flours of cereals such as wheat and triticale are the fundamental basis for obtaining good milling quality.

The AMMI analysis (Figure 2), allowed dispersing the 20 lines in their physicochemical properties with greater and lesser association, forming three groups between the two components; the first contributed 60%, it was integrated by L-19, L-11, L-6, L-3, L-12, L-2, L-16, L-9, L-8, L-4 and L-13, these lines showed greater association with proteins, fats and fiber (Figure 2), L-20, L-2 and L-12 showed 1.2 to 1.5% more protein, in fiber and fat the 12 lines obtained on average between 1 and 3%, respectively, than reported by ^{Peña et al. (2007}).

Figure 2 Biplot of the 20 lines of spring triticale with physicochemical variables, evaluated in Celaya, Guanajuato. 

These lines have good availability of nitrogen in the grain and the levels of this component are acceptable for flours intended for baking, the production of tortillas, cookies and pasta, this is relevant because the gluten content is the most important factor when defining the quality of both cooking pasta (spaghetti and soups) and baking (^{Pattison et al., 2014}). Therefore, the contents of these parameters could be an attributable factor of the climatic conditions in which the field grains were collected, the percentages of these parameters were between 1 to 3% higher than what was reported by ^{Ferreira et al. (2015}), who mentioned that the quality of protein present in grains such as wheat and triticale is decisive for use in the food industry; the second group contributed 25%, with lines L-7, L-18, L-17, L-14 and L-10 being associated with WTG, HW and carbohydrates, these parameters are important in the grain content, it is defined mainly by its morphology in each variety, this can be negatively altered by late sowings, nitrogen deficiency, deficiency in water supply, and grain filling due to high or low temperatures.

The hectoliter weight is considered an indicator of the yield potential in flours that a variety or line has during milling, in such a way that varieties with low hectoliter weight (less than 70 kg hl^-1) usually show low yields of flour, for this reason, during this research higher values were found in HW, the lowest value was in L-7 (76.10 kg hl^-1) and the highest in L-17 (78.07 kg hl^-1), these results indicate that the HW in this group is a decisive factor in determining the quality in flours in obtaining better yields (^{Jing et al., 2016}).

The third group, formed by L-1, L-5 and L-15, showed a negative relationship with the content of minerals (ashes) and moisture in the grain, this result is important because, if their concentrations of ashes and moisture were greater than 2 and 14%, respectively, especially in grains with low hectoliter weight, it can significantly contaminate the semolina and flour from milling, the highest levels of ash contamination are particularly undesirable in hard or crystalline grains, because the particles darken the semolina and to a greater degree, food pasta, the range of ashes in triticale is 2%, for moisture the contents in the grain greater than 13-14% favor the damage, causing the presence of microorganisms (^{Fan, 2018}).

The concentration levels of ashes in grain that are considered desirable should be less than 2% (^{Peña, 2007}). The results indicate that this group, especially lines L-5 and L-15, showed ash values less than 2%, and moisture only line L-15 presented 12.43%, these results agree with those obtained by ^{Pattison et al. (2014}), who evaluated the quality of gluten in the grain of triticale varieties and found that values above 14% in moisture and above 2.1% in ashes contribute more to the quality in flours, therefore, the negative results in this research may present effects in having a quality of flours not suitable for the food industry.

Conclusions

The results obtained show that it is possible to use triticale as an alternative in the food industry for its nutritional quality of grain, suitable for processes of making pasta, breads, tortillas, macaroni, cakes and soups. It is necessary to modernize the diversity of triticale varieties that are grown in Mexico for their protein quality; it is recommended to use lines L-18, L-17, and L-14, in addition to L-20, L-10 and L-19 because they met the industrial quality ranges in this research, such as higher WTG and HW, higher percentages of fat, fiber, protein and carbohydrates, as well as lower values in moisture and ashes.