The Effect of Genotype, Environment, and Genotype-by-Environment Interaction on Oat Processing and End Product Quality Characteristics | ARDI | Agri-Food Research & Development Initiative | Manitoba Agriculture, Food and Rural Initiatives

Oat production in Canada has a substantial economic influence on the agricultural industry. The estimated area seeded to oats was 1.82 million hectares in the year 2000, making it the fourth most seeded crop in Canada. Currently, the majority of Canadian oat production occurs in eastern Saskatchewan and Manitoba. Manitoba produced 1 million tonnes of oats in 2000 and the predicted seeded acreage was similar for 2001 at 374,300 hectares. Canadian producers are continuing to grow more oats, which is helping to satisfy demands from the largest importer of oats in the world, the United States. While oats have traditionally been used for animal feed, human consumption of oats has doubled to twenty-four percent of total consumption since 1960. Between 60 and 70% of Canadian oats exported to the United States are for milling. Canadian millers also export groats, flakes, and meal. Manitoba exports of these products were valued at approximately $36 million in 1999. As innovative uses for oats are created and human consumption continues to increase, the demand for high quality oats will strengthen. Successful competition in domestic and international markets requires continued improvement of Canadian oat cultivars to meet the changing needs of the agricultural and food industries.

Oat breeding programs are in place in Canada to ensure the availability of cultivars that possess the characteristics desired by producers, millers, food manufacturers, and consumers. Successful introduction of novel or improved traits into adapted cultivars requires a good understanding of the factors that control the expression of the trait. The majority of published research investigating genotypic and environmental effects on oat quality has focused on agronomic traits such as yield and test weight. Variation in oat protein, oil, and beta-glucan content has also been well documented, but not for cultivars commonly grown in Manitoba. There is a lack of information available describing factors that affect oat characteristics important to processing and end-product quality for genotypes grown in western Canadian environments.

AC Assiniboia, AC Medallion, CDC Boyer, Triple Crown, and OT288 were chosen for this study because they are either commonly grown in Manitoba or are important in the current breeding program. Four field replicates were grown at each of six diverse sites in Manitoba (Glenlea, Morden, Silverton in 1998 and Winnipeg, Carman, Silverton in 1999) under four nitrogen fertilization regimes (0, 40, 80, 120 kg/ha). The samples receiving no added fertilizer (0 kg/ha) were used in the first portion of this genotype-by-environment study. A Codema Dehuller was used to obtain the percent hull content of the whole oats, and the amount of groat breakage incurred during dehulling was determined by hand sorting. Groats were ground to wholemeal using a Retsch Mill and tested for protein (Leco Combustion), oil (NIRS), β-glucan (AACC Method #32-23), and total starch (AACC Method #76-13) content. Starch was extracted and analyzed for amylose content (Potentiometric Titration), swelling volume (Crosbie, 1991), thermal properties of gelatinization (DSC), and pasting properties (RVA). The 10% starch paste recovered from RVA analysis was cooled for 24 hours and the strength of the gel tested using the TA-XT2 Texture Analyzer. In order to investigate variation in end-product quality, a small scale oat conditioning and flaking method was devised using a bench-top flaker developed at CRC. Oat flakes were assessed for granulation (Ro-Tap) and water hydration capacity (AACC Method #88-10). Thirty grams of flakes were cooked with water in a microwave and the resulting hot oatmeal was evaluated for textural characteristics using a TA-XT2i Texture Analyzer.

The second portion of this study focused on the analysis of the samples grown under varied nitrogen fertilizer rates. A subset of quality characteristics was looked at including hull content, breakage, protein, oil, beta-glucan, and wholemeal pasting properties (RVA). In addition, the colour of whole oats, groats, and wholemeal was measured with a Minolta Chroma Meter.

Results and Discussion:

Physical Quality: Oats with high proportions of hull can be bulky, thereby increasing storage space and transportation requirements. Hull content is also negatively correlated with test weight, which is an important grading factor. Oats receiving a low grade at the point of sale earn a reduced price for producers and are generally used for animal feed rather than entering the food market. The first step of the oat milling process is removal of the hull from the groat. Whole oats that have a minimum proportion of hull to groat content provide millers with a greater recovery of usable product; oat groats that are susceptible to breakage also reduce the usable portion of the oats for the miller. Due to the economic importance of these characteristics, reducing the hull content and breakage of registered oat cultivars is a major goal of Canadian oat breeding programs.

Genotype and environment significantly affected hull content. Genotype means calculated across replicates and environments ranged from 28.69% (CDC Boyer) to 33.43% (Triple Crown). Overall environment means showed a similar range, but environment contributed slightly more to total variation in hull content (30.72%) than genotype (24.28%). A significant genotype-by-environment interaction was also found; genotypes AC Assiniboia and AC Medallion did not always rank the same relative to OT288 across environments. This effect may have been influenced by the fact that OT288 is a semi-dwarf variety. This significant interaction effect indicated that despite strong genotypic effects, breeding for low hull content requires multiple testing sites.

Groat breakage was also significantly affected by genotype, environment, and genotype-by-environment interactions (Figure 1). In this case, the genotypes responded differently to the environments but maintained a consistent rank order, indicating that breeders could select for low breakage at any environment. Triple Crown was the least susceptible to breakage (mean 4.31%) whereas AC Medallion had the worst incidence of breakage (mean 13.14%). Environment contributed the most to total variation in groat breakage with some environments producing oats with up to 20% groat breakage.

Figure 1. Effect of genotype and environment on the amount of broken groats after dehulling.

Oat Composition: Now that more oats are being milled for human consumption, it is necessary to develop cultivars that contain nutrients in proportions that are conducive with the low fat, high fibre diets that consumers strive for. For example, it is essential from a competitive standpoint that newly registered Canadian oat cultivars need to meet industry specifications for high beta-glucan and low oil in order to qualify for food labeling health claims in the United States. In addition, high protein content is desirable in oats for high nutritional quality.

Genotypes differed significantly for all three components measured (Table 1). Variation in beta-glucan content was the most dependant on genotype (78.46% of total variation was due to genotype). In contrast, protein and oil contents were significantly affected by environment and, in both cases, this influenced total variation more than genotype. A likely explanation for the large environmental variation observed for protein is the difference in soil nitrogen levels at the six sites. Morden, 1998 Silverton, and 1999 Silverton all had soil nitrogen levels greater than 144 kg/ha, producing oats with mean protein contents of 16.12, 17.83, and 14.20%, respectively. In contrast, the other three sites, which had soil nitrogen levels of 20 to 35 hg/ha, produced oats with lower mean protein contents (12.81 to 13.45%). None of the three components were significantly affected by genotype-by-environment interactions, indicating that breeder selection for high beta-glucan and protein, and low oil are likely to be successful.

Table 1. Genotype means for oat composition.

Cultivar	Protein (%)	Oil (%)	Beta-Glucan (%)
AC Assiniboia	14.39	4.37	4.38
CDC Boyer	14.32	4.51	5.11
AC Medallion	14.10	4.57	4.77
OT288	14.85	4.66	4.33
Triple Crown	15.04	4.22	5.69

Starch Characteristics and Functionality: Starch is the most abundant component in oat groats and thus has a great potential to affect the quality of oat products. Heating starch in the presence of water during the production and preparation of oat products brings about pasting and gelatinization. Pasting is characterized by the swelling of starch granules and disruption of their crystalline structure as they take up water. Thickening of the paste occurs as amylose and amylopectin are co-leached from the granule, followed by gelatinization, which is marked by the irreversible disruption of the granule structure. It is necessary to determine what factors affect variation among oat starches if breeders are to select for cultivars that will respond to processing in specific ways and help food manufacturers produce oat products that meet consumer acceptance.

All starch characteristics measured were significantly affected by both genotype and environment except starch gel firmness for which the environment effect was not significant. Genotype means for total starch content ranged from 62.95% (Triple Crown) to 64.95% (AC Assiniboia). Triple Crown also had the highest starch swelling volume (5.92 cc verses

CDC Boyer, which was lowest at 5.54 cc) and the lowest gelatinization temperature (58.36°C). Starch from AC Assiniboia and AC Medallion exhibited the greatest decrease in viscosity upon stirring at high temperature and also made the least firm and most adhesive gels. The gelatinization temperatures of AC Assiniboia and AC Medallion starches also tended to be high (59.73 and 59.40°C, respectively), but AC Medallion was not significantly different from OT288, which had medium gel firmness and adhesiveness properties. AC Assiniboia, AC Medallion, and OT288 tended to have lower amylose contents as well as high ΔH values for the amylose-lipid complex enthalpy. There is evidence of some chemical and/or physical difference in the AC Assiniboia and AC Medallion starches that cause them to be weaker during hot stirring and as cooled gel, but not necessarily affect their ability to reach a high paste viscosity at 95 or 50°C.

Gel firmness and some of the starch RVA parameters (hot paste, breakdown, shear-thinning) were involved in significant genotype-by-environment interactions that resulted in changes in the ranking of genotypes across growing sites. In most cases, the interaction effect contributed less to total variation than the main effects. Therefore, breeder selection would still be possible, but would require multiple growing sites to ensure accurate selection. For most starch characteristics, environment contributed more to total variation than genotype, except for amylose content, starch RVA breakdown viscosity, and gel firmness. An example of the environment effect of starch RVA pasting characteristics for one genotype is shown in Figure 2.

Figure 2. Effect of environment on starch RVA pasting characteristics.

End-Product Quality: To ensure the success of a cultivar in the industry, it is essential to test end-product quality. For example, in the wheat breeding program, potential lines are baked into bread which is evaluated for high loaf volume. Currently, there is no equivalent end-product test for oats. This is, in part, due to the lack of laboratory scale oat processing methodology. In response to this need, a small scale method for conditioning oats was developed to mimic the heat/moisture treatments used in the industry to inactivate enzymes detrimental to quality. The conditioned groats were then flaked using a bench top flaking machine developed at CRC. Hot oatmeal was chosen to test for differences in end-product quality among genotypes and environments.

Flake granulation varied significantly (P<0.01) with G and E. Genotype response to growing site varied, but only one genotype pair was involved in a significant change in rank order for the proportion of largest flakes. The majority of total variation in the size of flakes was due to genotype. Water absorption capacity of oat flakes was also significantly (P<0.01) affected by genotype, environment, and genotype-by-environment interactions. The only significant change in rank order occurred between CDC Boyer and Triple Crown, but both were low absorption types.

Variation was observed among genotypes for hot oatmeal texture. The force required for the probe to descend into the oatmeal as well as the amount of sample that adhered to the probe was significantly (P<0.01) affected by genotype, but not environment. Genotypes with high positive force and area values (Triple Crown and CDC Boyer) appeared to be more fluid with two distinct phases: whole flakes and paste. The texture curves also peaked more rapidly. These observations likely corresponded to the relative ease of the probe to travel through the relatively weak paste followed by a rapid increase in force as the probe sensed flakes that had settled to the bottom of the canister. Alternatively, oatmeals made with relatively low positive force and area values (AC Assiniboia, OT288, AC Medallion) appeared thicker, with flakes more uniformly dispersed throughout the samples. These texture curves had a more gradual slope approaching the peak. Oatmeal made from Triple Crown stuck to the probe the least and AC Assiniboia tended to stick the most.

Stringiness, which is a measure of the length of time “ropes” of oatmeal extend from the sample surface to the ascending probe, varied with genotype and environment. The most stringy oatmeals were made from Triple Crown grown at all environments. Genotype-by-environment interactions were only significant at a 5% probability level, but involved several cross over effects between OT288 with CDC Boyer and AC Medallion.

Physical Quality and Composition: Nitrogen fertilization rate significantly affected hull, protein, oil, and beta-glucan content. Environment-by-nitrogen interaction effects also significantly influenced these parameters. This means that the response to nitrogen fertilization depended on the growing site. For example, Figure 3 shows that increasing the nitrogen rate from 0 to 120 kg/ha resulted in an increase in beta-glucan by as much as 1% at the Winnipeg site. This response was less apparent at environments with high initial soil nitrogen, indicating that there is likely an optimum level of nitrogen availability above which fertilization does not increase beta-glucan. A similar trend was observed for protein. Nitrogen fertilization had a slight decreasing effect on hull and oil contents at environments with low initial soil nitrogen levels. Nitrogen had a greater influence on the total variation in oat composition compared to hull content, but either genotype or environment still had the most effect in all cases. Breakage was not significantly affected by nitrogen fertilization.

Wholemeal Pasting Properties: Wholemeal pasting properties, as measured by the RVA, were significantly affected by nitrogen fertilization except setback viscosity. Component of variation analysis indicated that either environment or genotype played a greater role in the variation of wholemeal pasting.

Colour: Nitrogen significantly affected whole oat, groat, and wholemeal L* (brightness), a* (red to green scale), and b* (blue to yellow scale) values. However, component of variation analysis revealed that nitrogen fertilization caused negligible variation in colour compared to the effects of genotype and environment. Genotype was the major contributor to total variation in hull and groat L* and a*, and environment played an increasing role in the variation of wholemeal values. The lightest coloured wholemeals were produced at the Silverton and Morden sites in 1998 and Triple Crown had the lightest wholemeal at all sites.

Conclusion:

The results from this study provide producers, millers, oat processors, and breeders with information regarding the relative effects of genotype, environment, and nitrogen fertilization on the quality of oats destined for the food industry. This study reiterated the importance of soil fertility testing, as the availability of sufficient levels of nitrogen will help ensure high beta-glucan and protein as well as low hull and oil contents, all of which are important for food quality oats. On the other hand, nitrogen fertilization above a critical level will not likely affect these traits. Soil fertility also has the potential to influence oat wholemeal pasting properties, which becomes important when the oats are processed into food products. Controlling variation in groat breakage or colour is not possible through the fertilization treatments used in this study. Improving the milling properties, composition, and starch functionality of oats is best done by breeding due to large genotypic effects in addition to careful consideration to environment. Further research is needed to identify the specific environmental factors that affect oat quality. Utilizing lab scale oat processing methodology, breeding programs could select for oats with specific end-product texture. Genotype-by-environment interactions indicate that multiple growing sites may be required to select for flake granulation and water absorption capacity. Future research should investigate links between instrumental texture analysis and sensory evaluation of oatmeal to determine which measurements are most important to the food industry.

We would like to acknowledge the financial support from the Agri-Food Research and Development Initiative (ARDI) for this project.