Introduction

In the year 2013, Mexico consumed 6.6 million Mg of bread wheat (Triticum aestivum L.) and durum wheat (T. durum L.), out of which 4.6 million Mg were imported, mainly from the USA and Canada. The total of wheat imports were bread wheat, out of which 63 % were wheats classified as: hard red winter, soft red winter, and hard red spring wheats (^{CANIMOLT, 2014}). The industries are in favor of imports, since the national market does not have the industrial quality required. Thus, rainfed wheat frequently faces marketing problems due to its heterogeneous industrial quality because it is grown under variable environmental conditions (^{Hortelano et al., 2013}).

An option to increase the national wheat production and reduce our food dependence is sowing under rainfed conditions, which contributed 5 % to the national production in 2014 (^{SIAP, 2015}), in areas where maize production suffers initial droughts or early frost, and wheat would be an alternative due to its shorter growth cycle. Tlaxcala and Estado de Mexico provided over 50 % of the rainfed production in 2014 (^{SIAP, 2015}), and there is potential to grow 200,000 to 300,000 ha of wheat, respectively, with yields of 2.1 to 4.8 Mg ha^-1, depending on the location and the variety used (^{Villaseñor and Espitia, 2000}; ^{Hortelano et al. , 2013}). In those states wheat was sown on 33 806 and 9073 ha in 2013 (^{SIAP, 2014}). An advantage of these states is their proximity to the main milling and consumption centers, Mexico City and the Estado de Mexico, where demand is over 60 % of milled grain (^{CANIMOLT, 2014}).

Thus, Mexican wheat must compete in price and industrial quality with imported wheat. Some characteristics of the industrial quality of hard red wheat are: protein 12.6 %, hectoliter weight 79.4 kg hL^-1, firmness of dough 350 x 10^-4J, dough firmness/ extensibility ratio 1.2, and bread volume 842 mL. The hard red spring wheat has 14.6 % protein, hectoliter weight 79.2 kg hL^-1, firmness of dough 500 x 10^-4 J, dough firmness/extensibility ratio 0.9, and bread volume 946 cc (^{Maghirang et al., 2006}). In Canada, specific actions were carried out for improving of the industrial quality, and varieties were developed with 15.2 % protein, related with extra firm doughs (^{McCallum and DePauw, 2008}). ^{Espitia et al. (2003)} pointed out the need to improve protein content and bread volume of the Mexican varieties produced under rainfed conditions. Therefore, it is necessary to identify genetic sources with high protein content (>12.5 % in refined flour) and good baking quality (>800 mL) that can be incorporated into improvement programs.

The aim of this research was to evaluate the industrial quality of genotypes of bread wheat from the USA and Canada grown under rainfed conditions in the Highland valleys of Central Mexico, and identify the desirable traits and incorporating them into the plant improvement program of the Valle de México Experimental Field, of the National Forestry, Agriculture, and Livestock Research Institute (CEVAMEX-INIFAP).

Materials and Methods

Out of the 16 genotypes used, there were 12 introduced, Náhuatl F2000 and Kronstad F2004 as controls, and the advanced lines GAVIA/ROM/3/PIRUL/GUI//TEMP/AGR/4/ JUCH (line 1) and PAMDOLY-PABG-Tardía-C4 (line 2), of the INIFAP rainfed wheat improvement program. The genotypes Kulm, Keene, Waldron, SD3249, SD3195 and SD3236 are from USA; and Bluesky, AC Vista, BW725, HY437, HY439, HY632 are from Canada.

The 16 wheat genotypes were sown in eight sites of the Estado de Mexico, without irrigation, in the spring-summer cycle during 2009 and 2010. The locations were Santa Lucia de Prias, Juchitepec, Coatepec and Chapingo. Sowing was carried out in the first week of June: Coatepec in 2009, Juchitepec in 2009; Coatepec in 2010 and Juchitepec in 2010 during the third week in June; in Chapingo, it was established in the second date in 2009, 20 d after the first sowing season. Chapingo is located at 19° 13' N and 98° 51' W, at 2250 masl, the temperature is 15.9 °C and rainfall is 620 mm year^-1, the climate is temperate subhumid (typeC (Wo) (W) b (y) g); Santa Lucía de Prias has a similar climate and is located at 19° 44' N, 98° 87' W and 2260 masl, its rainfall is 636 mm and 16.1 °C is the average annual temperature; Juchitepec is between 19° 06' N, and 98°53' W a 2571 masl, rainfall is 807 mm year^-1 and the average annual temperature is 15.5 °C, its climate is humid; Coatepec has a humid climate, is located at 2320 masl and the rainfall and average annual temperature is 660 mm and 15.1 °C (^{García, 1981}). The soil in Chapingo is fluvisol, in Santa Lucía, it is epipedon, in Juchitepec it is regosol and in Coatepec it is phaeozem (^{FAO, 1998}).

The experimental design was complete randomized blocks with two repetitions in each location and the experimental unit consisted of four furrows, 3 m long and separated by 30 cm. The planting density was 120 kg ha^-1, we fertilized using the formula 40-20-0, and all the N and P₂O₅ were applied during sowing. As a source of fertilization we used urea [(CO (NH₂)₂] with 46 % N and triple calcium superphosphate [Ca (H₂PO₄)₂] with 46 % P₂O₅. Narrow leaf weeds were controlled with 900 mL ha^-1 of Topik 24EC^® 30 d after planting (das), and wide leaf weeds were controlled with 1 L ha^-1 of Esteron 47^® 35 das. During booting stage 500 mL ha^-1 of Folicur^® were applied to control diseases. Harvesting was carried out when the humidity of the grain was lower than 14%, using a mini-combined machine.

The industrial quality analyses were carried out in the Wheat Quality Laboratory of the CEVAMEX-INIFAP. The hectoliter weight (kg hL^-1) was determined in a 500 g grain sample using a volumetric scale (Seedburo Equipment CO., Chicago, IL.). Grain hardness was measured using the pearling index on 20 g of grain (Strong Scott pearler-USA), indicating the ease for partially eliminating the external layers of the grain. The standard procedure included abrasion for 1 min, sieving with a 1.25 μ mesh and evaluation of the grain weight. Refined flour was obtained from the grain, using a Brabender mill (Quadrumat Senior, C.W. Brabender OHG, Germany) and sieved through a 129 μm mesh. Protein content in the grain (%) was measured with a NIR infralyzer 300 (method 39-10; ^{AACC, 2005}). The volume of sedimentation (mL) was determined in a 3.2 g sample of refined flour with the presence of lactic acid and isopropyl alcohol (Zeleny, 1946), which indicates the hydration and expansion capacity of the protein; in this test, greater volume indicates more strength. In the Chopin Alveograph (Tripette & Renaud, France) we obtained the alveogram, using the method 54-30A of the ^{AACC (2005)}, and 60 g of refined flour; the alveogram helped calculate the firmness (W) and the tenacity: extensibility ratio (PL) of the dough. Values W and PL are used to classify the dough. Values for W greater than 300x10^-4 J correspond to strong dough, 200x10^-4 J to 300x10^-4 J, to medium strong dough, and below 200x10^-4 J, weak dough. The PL values classify dough as balanced (PL = 1.1), extensible (PL<1) and strong (PL>1.2) (^{Martínez-Cruz et al., 2014}). The volume of bread (mL) was obtained from bread baked using the procedure of direct dough (method 10-09, ^{AACC, 2005}), from 100 g refined flour, 3 g powdered milk, 3 g vegetable oil, 25 mL of a yeast solution at 24 % and 25 mL of a sugar-salt solution at 20 and 4 %, respectively, fermentation for 3 h and 25 min in 90 % relative humidity and 35 °C. It was baked in a Despatech Oven Co. (Minneapolis USA) brand oven at 220 °C, for 25 min; the volume was measured in a volutometer with the displacement of Brassica campestris L. seeds.

Results and Discussion

Highly significant differences were found in all variables studied between locations and genotypes, and the genotype x environment was also significant (Table 1). This agrees with observations by ^{Williams et al. (2008)}, ^{Zecevic et al. (2013)} and ^{Hasniza et al. (2014)}, who reported the effect of these variables in the industrial quality of bread wheat.

Table 1 Significance of the average squares of the analysis of variance for variables of quality of the genotypes of bread wheat evaluated in rainfed, for the spring-summer cycles of 2009 and 2010. 

*p≤0.05; **p≤0.01; DF: degrees of freedom; ^†error of the general analysis; SV: source of variation; HLW: hectoliter weight; PG: grain protein; FG: grain hardness; VS: sedimentation volume; W: dough strength; PL: tenacity/extensibility; VB: bread volume.

The environmental variation was the most important for the hectoliter weight, volume of sedimentation, dough strength, and bread volume; the variation due to genotype was for the content of protein and of grain hardness; the variation by the genotype x environment interaction was the most important for the tenacity: extensibility ratio. In this way, more then 30 % of the total variation was explained by the locations for the variables of hectoliter weight, volume of sedimentation, firmness of dough and volume of bread, respectively (Figure 1). These results are similar to those obtained by ^{Aucamp et al. (2006)}, ^{Dencic et al. (2011)}, ^{Surma et al. (2012)} and ^{Rozbickia et al. (2015)}, who indicated that the environments were the main source of variation in these variables of industrial quality. For firmness of the dough, the participation of environmental and genotypic variances were similar (Figure 1) with regard to the total variation. These results are similar to those reported by ^{Uthayakumaran et al. (2012)}.

Figure 1 Figure 1. Proportion of the variation due to the sources of variation of end-use quality traits of bread wheat genotypes. HLW: hectoliter weight; PG: protein in grain; FG: grain hardness; VS: sedimentation volume; W: dough strength; PL: tenacity: extensibility; VB: bread volume. 

The genotypic variation provided more than 50 % of the total variation to the firmness and protein content in the grain (Figure 1). This agrees with observations by ^{Yong et al. (2004)} and ^{Surma et al. (2012)} for grain firmness. However, for protein content in grains, it is the opposite to the conclusions reached by ^{Vázquez et al. (2012)}, ^{Surma et al. (2012)} and ^{Finlay et al. (2007)}, who pointed out that the environment provides the greatest variation.

The participation of the total variation due to the genotype x environment interaction did dot surpass 20 %; the highest were for hectoliter weight and the ratio tenacity: extensibility. For dough strength, grain protein, and bread volume, the contribution to the variation of interaction did not exceed 10 % (Figure 1).

The bread volume varied between 786 and 919 mL. Genotypes Keene, SD3249, HY437, BW725, Bluesky and Kulm stood out for their values of over 870 mL. This is due to the fact that these genotypes combined over 13.9 %, protein in grains, and surpassed the best control, Náhuatl F2000. Besides, they displayed dough strength (higher than 300x10^-4 J) as well as extensible, and a ratio tenacity: extensibility lower than 1.1 or balanced by its PL=1 (Table 2). Genotypes HY439 and HY632 were related to of intermediate bread volume, near to 850 mL, although they displayed the highest values of grain protein content, which is a variable that favors the bread end-use and nutritional quality. This agrees with reports by ^{Clarke et al. (1994)}; ^{Maghirang et al. (2006)}; ^{McCallum and DePauw (2008)} and ^{Humphreys et al. (2010)}, who indicated that a high strength of the dough, related to low PL values and a high protein content, favors the quality of bread. Besides their good quality for bread, genotypes BW725 and Bluesky were classified as having hard grains due to their values below 47 %, which is unfavorable for germination in the spike in rains during physiological maturity, which reduces the quality (^{Dencic et al., 2013}).

Table 2 Averages of variables of quality of bread wheat genotypes evaluated under rainfed conditions. Spring-Summer cycles 2009 and 2010. 

^†Average values with different letters in a column are statistically different (p≤0.05). HLW: hectoliter weight; PG: protein in grain; FG: grain hardness; VS: sedimentation volume; W: general dough firmness; PL: tenacity: extensibility; VB: bread volume; CV : coefficient of variation.

The intervals of the averages between locations, regarding genotypes, were greater for hectoliter weight, volume of sedimentation, firmness of dough, and volume of bread (Table 2 and Table 3). This confirmed its importance in the total variation of variables related to the industrial quality, according to ^{Vázquez et al. (2012)} who described the effect of the environment on industrial quality. Santa Lucia 2009 gave the highest volume of bread due to the combination of the highest values for protein content in grains and firmness of dough, related to good extensibility. An opposite behavior was displayed by wheat obtained in Coatepec in 2010 and Juchitepec in 2010, which decreased these variables, and therefore the volume of bread (Table 3).

Table 3 Averages of variables of quality of locations of bread wheat genotypes evaluated under rainfed conditions. SpringSummer cycles 2009 and 2010 

^† Average values with different letters in a column are statistically different (p≤0.05). 1F: first day of plantation; 2F: second day of plantation; HLW: hectoliter weight; PG: grain protein content; FG: grain hardness; VS: sedimentation volume; W: dough strength; PL: tenacity: extensibility; VB: bread volume; CV : coefficient of variation.

Conclusions

The variables of the industrial quality of bread wheat protein and firmness of grain were mostly controlled by the genotype, and for firmness of the dough the influence of the environment and the genotype was similar. Hectoliter weight, sedimentation volume, and bread volume of bread mainly affected by the environment; and the tenacity: extensibility ratio was due mostly to the genotype x environment interaction.

Keene, SD3249, HY437, BW725, Bluesky and Kulm must be used as parents within the plan of the rainfed bread wheat breeding program, to improve the bread volume and grain protein content; and HY439 and HY632 for a high protein content.