Nutritional properties of various oat and naked oat cultivars

Two husked oat (Avena sativa L.) cultivars (Effektiv and Max) and ten naked oat (full botanical name Avena sativa L. ssp. nuda Gillet & Magne) cultivars (Klimt, Tatran, Hronec, Vazec, Samuel, Oliver, Otakar, Kamil, Tibor and Saul), were grown on the experimental fields in Austria by Saatzucht Edelhof (48°36'29.5"N 15°13'29.2"E) in the years 2014 and 2015 on loamy sand soil (pH 5.8–6.3). Crops were cultivated organically, neither fertilizer nor pesticides were used. Sowing density was 400 seeds/m²for husked and 500 seeds/m²for naked oat. In 2014, the preceding crop was winter triticale, sowing date was 28^th of March, harvest date was 22^nd of August. Precipitation from March to August 2014 was 434 mm and the average temperature was 13.2°C. In 2015, the preceding crop was winter rye, sowing date was 23^rd of March and harvest date 4^th of August. Precipitation from March to August 2015 was 276 mm and the average temperature was 14.0°C.

All cultivars are registered in the EU (see Supplemental Table 1 for maintainers). The de-hulled grains of cultivars Effektiv and Max and the hull-less grains of the naked oat cultivars were ground with the ultra-centrifugal mill Retsch ZM 200 (Retsch, Haan, Germany). The oat flour was stored at -4°C in sealed plastic containers. Moisture content of the oat flour was determined by measuring the weight loss after drying for 3 h at 105°C (ICC standard method no. 110/1). All analytical results were expressed per dry matter.

Fat content was determined according to the ICC standard method no. 136 (Soxhlet method), without acid hydrolysis prior to the extraction. Crude protein was determined from the nitrogen content according to the standard method no. 105-2 (Kjeldahl method). A factor of 5.34 was used for the conversion from nitrogen to protein (Mariotti et al., 2008). Ash content was determined according to the ICC standard method no. 104-1. The content of β-glucan was measured using the ICC standard method no. 166.

For the extraction of the Osborne fractions, 4 ml of distilled water was added to 250 mg of the ground oat samples, mixed on a vortex mixer, extracted for 10 min on an overhead shaker and then centrifuged for 20 min at 3220 g. The supernatant containing the albumin fraction was then collected and the pellet was once again extracted with water. The supernatant of both extractions was then combined, and water was added to a total volume of 10 ml. The pellet was then further extracted in the same way with 1 M aqueous sodium chloride solution to obtain the globulin fraction, and then with 60% ethanol for the prolamin fraction.

Albumin and globulin concentrations were determined photometrically via the Bradford assay using Roti-Quant solution, as per the manufacturer’s instructions (Carl Roth, Germany). Prolamin concentration was measured via RP-HPLC on a Shimadzu HPLC system and a diode array detector (Shimadzu Cooperation, Kyoto, Japan). Water and acetonitrile with 0.05% trifluoroacetic acid were used as mobile phase, applying a gradient of 25 to 90% acetonitrile for 60 min. LabSolutions Software (Shimadzu Cooperation, Kyoto, Japan) was used to quantify the peaks relative to those obtained with the PWG gliadin standard.

Tocopherols and tocotrienols were extracted from the oat samples using the method of Panfili et al. (2003). The concentration was determined by NP-HPLC on a Shimadzu HPLC system with a fluorescence detector. The mobile phase was a mixture of 70% n-hexane und 30% ethyl acetate. Samples were quantified relative to the tocopherol and tocotrienol standards (Supelco/Sigma-Aldrich, USA). For the extraction of phenolic compounds, a protocol optimized for maximum avenanthramide extraction was used (Maliarova et al., 2015). Extracts were stored at -30°C. Avenanthramide concentration was measured via RP-HPLC on a Shimadzu HPLC system equipped with a diode array detector (Shimadzu Cooperation, Kyoto, Japan). Water and acetonitrile with 0.25% trifluoroacetic acid were used as mobile phase with a gradient of 12 to 95% acetonitrile for 50 min. LabSolutions Software (Shimadzu Cooperation, Kyoto, Japan) was used to quantify the peaks relative to the avenanthramide standards A, B and C (Sigma-Aldrich, USA). Total phenolic content (TPC) was determined using the Folin-Ciocalteu method by using a modified protocol given by Singleton et al. (1999). Radical scavenging capacity was determined with ABTS^•+-radical assay (Re et al., 1999), expressed as μmol Trolox-equivalents (TE)/g sample. For further analysis of antioxidant capacity, a ferric ions (Fe³⁺) reducing antioxidant power assay (FRAP) according to (Benzie and Strain, 1999) was performed.

All results are given as mean values of triplicates, unless otherwise specified. All data was tested for outliers (Grubb’s test). One-way ANOVA (analysis of variance) was used for finding differences between cultivars and years at a significance level of p < 0.05. The post hoc test Fisher’s least significant difference test (LSD) was used to visualize the differences (shown by different superscript letters in results and discussion). Multi-factor ANOVA was used to study the effects of genotype and year on the parameters, and to investigate the genotype-year interactions. To identify correlations between two factors, the Pearson product-moment correlation coefficient was determined. All statistical analyses were conducted using Statgraphics Centurion XVI, Statpoint Technologies Inc., The Plains, VA, USA.

Dry matter contents ranged between 91.5 and 96.5%, with higher values in samples from 2015. As dry matter content is dependent on storage conditions, these differences in dry matter content are within the expected range. Results of the analysis of ash, fat, protein and β-glucan are summarized in Table 1. No significant difference between the values of ash, fat and β-glucan content in the two different harvest seasons were found. Fat contents ranged between 5.3 and 7.4%, which was in accordance with the literature values (Zhou et al., 1999). With Soxhlet method, polar lipids like phospholipids, which can make up to 20% of oat lipids, as well as lipids bound to proteins and carbohydrates, are not fully extracted; therefore, the actual fat content might be higher (Zhou et al., 1999). Effektiv, Max and Saul cultivars were lowest in fat, Klimt, Tatran and Vazec were the cultivars highest in fat. High fat contents in oat indicate a higher risk of a development of rancid off-flavors, and risk of clumping during milling. As the fatty acid composition in oat is highly beneficial for human consumption, a high fat content is nevertheless seen as a sign of quality in oat (Decker et al., 2014).

Ash contents varied in a relatively narrow range between 1.9% (Hronec) and 2.5% (Klimt). Oat is high in potassium, calcium and magnesium; so, oat cultivars with high ash content and therefore high mineral content can pose a beneficial addition to human diet (Arendt and Zannini, 2013).

None of the cultivars showed high β-glucan contents, with values ranging between 2 and 3.5%. In a two-year study with US oat cultivars, β-glucan contents between 3.9 and 6.4% were found (Lim et al., 1992) and according to Arendt and Zannini (2013), contents of up to 8% are possible. Oliver and Otakar were especially low in β-glucan, while Max, Kamil, Tibor and Saul showed the highest β-glucan values of the analyzed cultivars.

Protein contents in 2014 were significantly higher than in 2015 (Table 1). Values between 11.4 and 16.0% (2014) and 10.1 and 12.7% (2015) were observed. Klimt, Tatran and Samuel exhibited high protein contents, while cultivars Effektiv, Max, Hronec, Oliver and Tibor showed low protein contents in both years. These values are consistent with protein contents found in previous studies, which reported protein contents between 12 and 20% of dry matter (Arendt and Zannini, 2013).

The results of the analyses of tocopherol and tocotrienol content are provided in Table 2. Tocopherol and tocotrienol analyses of the cultivars showed that only the α-forms of both substances were present. In 2015, the tocopherol and tocotrienol content was significantly higher than in 2014. The lowest value of total tocopherols and tocotrienols in 2014 was 2.3 mg/100 g sample (cultivar Tatran), highest was 3.2 mg/100 g (Hronec). In 2015, 2.8 mg/100 g was the lowest content of total tocopherols and tocotrienols (Tatran), and 6.5 mg/ 100 g the highest (Otakar). Cultivars Effektiv, Klimt, Tatran, Vazec and Samuel showed lower tocopherol and tocotrienol contents while Oliver, Otakar and Saul contained very large amounts. The α-tocotrienol content was higher than the α-tocopherol content in both years and all cultivars except for Vazec. While α-tocotrienol has a higher antioxidant capacity than α-tocopherol (Serbinova et al., 1991) and is therefore of higher benefit for protection of oat lipids against oxidation, α-tocopherol has a higher bioavailability and, in addition to its antioxidant properties, also positive effects on cell division and thrombocyte aggregation (Sen et al., 2006). Other studies have reported oat vitamin E contents to be around 1.6 mg/100 mg (Arendt and Zannini, 2013) or even lower (Berga and Zute, 2012); so the values found in the present study were significantly higher. Tocol levels in cereals are highest in lipid rich milling fractions like germ and bran, so a correlation between total fat content and tocopherol and tocotrienol content was to be expected (Ko et al., 2003). But interestingly, no statistically significant correlation could be found between fat content and the concentration of tocopherol and tocotrienol.

Antioxidant capacity (measured in ABTS^•+-radical assay and FRAP assay) and total phenolic content in all cultivars were higher in 2014 than 2015 (Table 3). Cultivars Effektiv, Max and Hronec exhibited low antioxidant capacity and total phenolic content, while Klimt, Tatran, Samuel and Oliver showed high antioxidant capacity. Klimt, Vazec and Kamil exhibited a high total phenolic content. Few studies have been conducted on the factors that influence oat antioxidant activities. One Swedish study (Dimberg et al., 2005) investigated the concentration of antioxidative substances in three organically and conventionally grown oat cultivars from three consecutive harvests. There, the factor year was also found to have the greatest influence on antioxidant concentration, with concentrations in samples from a warm and dry year about 25% lower than from years with average precipitation.

As expected, strong correlations between antioxidant capacity measured in ABTS^•+-radical assay and FRAP assay (r = 0.96, p < 0.001) could be observed. Antioxidant capacity and total phenolic content also correlated strongly, with results from ABTS^•+-radical assay and FRAP assay correlating equally strongly with total phenolic content (r in both correlations = 0.83, p < 0.001). High phenolic content and antioxidant capacity are desirable traits in oat, as antioxidative oat extracts have been shown to reduce intracellular inflammatory activities and lessen the symptoms of colits in mice (Hasnat et al., 2015), and prevent ROS-induced stress in the livers and kidneys of diabetic mice (Marmouzi et al., 2017).

Avenanthramide analysis showed that contents were much higher in 2014 than in 2015 (Table 4). In 2014, total content of avenanthramides 2p, 2f and 2c ranged between 12.1 and 36.9 mg/100 g; while in 2015, contents between 1.7 and 6.2 mg/ 100 g were found. Other studies reported avenanthramide concentrations between 0.2 und 80 mg/100 g (Dimberg et al., 2005; Tong et al., 2014; Maliarova et al., 2015). Effektiv, Max and Hronec had the lowest, Tatran and Otakar the highest avenanthramide contents. Of the three analyzed avenanthramides, avenanthramide 2f was the most abundant, and avenanthramide 2c the least abundant in all cultivars (Table 4). In-vitro studies have shown avenanthramide 2c to have the highest, 2p the lowest antioxidant activity, a high proportion of avenanthramide 2c is therefore beneficial (Fagerlund et al., 2009). Cultivars Effektiv, Max and Hronec have low overall avenanthramide concentrations and an especially low content of avenanthramide 2c. Tatran and Otakar had the highest avenanthramide concentrations of the analyzed cultivars, and also a more beneficial avenanthramide profile. Avenanthramide concentration correlated strongly with antioxidant capacity measured with ABTS^•+-radical assay and with FRAP assay (r = 0.95 and 0.94, respectively, p < 0.001). As was expected, avenanthramide content also correlated strongly with total phenolic content (r = 0.84, p < 0.001).

Aside from the three quantified peaks, six more peaks could be observed in the chromatogram (Figure 1). These are most probably other, less common avenanthramides for which no commercial HPLC standards are currently available.

Figure 1

While albumin concentrations were higher in 2015 than in 2014, the opposite trend could be observed in globulins. No difference in prolamin concentrations could be observed between the harvest years (Table 5). Albumin contents ranged between 0.95 and 1.9% of dry matter, or between 7.8 und 17.1% of total protein content. Globulins made up between 3.6 and 6.8% of dry mass or 34.7 and 42.7% of total protein. Prolamins constituted between 0.5 and 1.4 of dry matter or 3.7 and 10.9% of total protein, with cultivars Effektiv, Max, Samuel, Oliver, Kamil and Tibor showing relatively low levels of prolamins. Other studies have found albumins to make up 1–20%, globulins 70–80% and prolamins 4–14% of total protein (Klose et al., 2009; Runyon et al., 2013; Rasane et al., 2015), which indicates that the globulin fraction in this study was not fully extracted. Higher volumes of solvent and longer extraction times seem to be needed. As discussed by Sunilkumar and Tareke (2017), numerous different extraction methods for the Osborne fractionization in oat have been used, so results vary widely. Further research to identify the optimal extraction parameters for a standardized Osborne fractionization of oat protein is needed.

The RP-HPLC spectra of the prolamin fraction showed major differences in chromatogram patterns between the cultivars (Figure 2). While all cultivars showed clusters of peaks between minutes 21 and 23 and minutes 26 and 29, the number, shape, and distribution of peaks were characteristic for each cultivar. Chromatograms could therefore be used for oat cultivar identification, as is already a standard method for wheat cultivars (Belitz et al., 2008).

Figure 2