Background

Oat (Avena sativa) kernels contain proteins with good nutritional quality, high bioavailability, and high proportion of globular proteins and have been the target of an increasingly intense breeding campaign [1]. Oat lines with increased protein content would therefore be a very good complement to existing animal and plant protein sources. Soil fertility, genotype, geographical origin, and climatic and environmental factors such as altitude, humidity, moisture, temperature, photoperiod, and incident radiations strongly influence oat protein concentration [1–5]. Protein percent can vary by as much as 3–4 % within the cultivar when this is grown in different geographical regions [5]. Application of nitrogen fertilizers to soil increases protein concentration by up to 2 % [6–9]. Interaction of genetic and non-genetic factors with environmental factors, differences in breeding methods, and additive natural genes due to the double dose of maternal genes of seed endosperm can change the protein concentration in cultivars grown under the same agronomic conditions. This, in turn, can affect nutritional quality [10–13].

High protein content in cereals is a desirable character and goal of many cultivars. Various methods are employed to achieve this, including conventional and mutational breeding, genetic engineering, marker-assisted selection, and genomic analysis. When developing cultivars with improved nutritional value, the improvements must be preserved and the cultivar must be as stable as possible [14]. It is therefore important to ensure that the protein quality is maintained and improvement attained in different environments and under different conditions.

Avenin is the collective name for alcohol-soluble prolamin proteins of oat that are highly heterogeneous in nature because of several polymorphic components [12, 15, 16]. Due to its polymorphic nature and environmental stability, compared with other cereal grain proteins, avenin is a potential candidate for inheritance studies, for identification of landraces, and for identification of genotype and germplasm [3, 17–20]. These identifications are very important for efficient handling of the breeding stocks for maintaining varietal purity and discriminatory markers for protected varieties [21]. In addition to their polymorphic nature, avenin proteins have charge and size heterogeneity that is absent in two other storage protein fractions, globulin and albumin [22]. Consequently, avenin fractions are more suitable for cultivar identification in cereal grains than globulin and albumin fractions [22, 23]. Separation of avenin using SDS-PAGE is recommended by the International Seed Testing Association (ISTA) for verifying species and cultivars [24].

To the best of our knowledge, no study has investigated the stability of avenin in a single oat cultivar exposed to chemicals or grown under different geographical conditions. Thus, the purpose of this study was to evaluate the effect of nitrogen fertilizer, mutation, greenhouse, and geographical location on the stability of avenin protein fractions of Belinda oat cultivars (BOCs) so that the cultivar with stably inherited avenin could be selected to assure the original gene pool, resulting in increased protein content and in improved cultivars.

Methods

Grain material

Grains of 20 oat cultivars and 5 mutated lines were evaluated in this study and were divided into five groups (Table 1). The parental Belinda cultivar was obtained from the SW collection (Svalöf Weibull AB, Landskrona, Sweden). The high-protein oat lines were selected after screening 1050 different mutated lines from a tilling population constructed from the spring oat cultivar SW Belinda (Sunil Kumar et al. in preparation). Control plants and plants treated with nitrogen at different concentration [60–240 N (kg/ha)] were grown in a greenhouse under halogen lamps, giving a photon flux density of 240 μmol/m2/s with a photoperiod of 18 h. Day and night temperatures were 25 and 16 °C, respectively. Plants were grown in 5-L pots in standard soil for oat growth. The treatments were arranged in completely randomized pots with three replications. The amount of soil in each pot was quantified to optimize fertilizer amount to mimic or reflect the actual field conditions. Split application of nitrogen fertilizer was applied for two members in group V in which the first fertilization was given at sowing and the second was at the tillering stage. Fertilizer was sprayed in the growing crop. Seeds were harvested for one season and analyzed.

Table 1 Sampling location, year of sampling, mutated oat lines, and fertilizer application rates of Belinda oat cultivars
Full size table

Chemicals

PageRuler™ Unstained Protein Ladder (Thermo Scientific), sodium dodecyl sulfate (SDS), and acetone were purchased from Sigma Aldrich. Nacl (1 mol/L), 2-chloroethanol, tris(hydroxymethyl)aminomethane, and glycine were obtained from Merck. All chemicals were of analytical grade.

Stock solutions

Five stock solutions were made, using 25 % 2-chloroethanol and Milli-Q water. All stock solutions were prepared at room temperature.

Milling and defatting of oat flour

Grain samples were ground using a benchtop ball miller (Minimill Pulversett 23, Fritsch, Germany) at 50 Hz for 5 min one day before analysis and kept in a desiccator to equalize the water content in all samples. The milled oat was defatted with acetone by dispersing 10 wt% of oat flour in acetone, and the mixture was stirred using a magnetic stirrer for 1 h at room temperature. The mixture was centrifuged at 14,000 rpm for 20 min at 20° C using a Beckman Optima LE-80K Ultracentrifuge (Palo Alto, California). The precipitate was collected and kept open overnight at room temperature to remove the remaining acetone present in the precipitate.

Analysis of total protein content

Total and soluble protein contents of the ground oat seeds were determined by measuring nitrogen content with elemental Flash EA 1112 N (Thermo Fisher Scientific, Delft, Netherlands) and using a modified Dumas method [25]. Total protein percentage was calculated by measuring the total nitrogen content in seed flour using a thermal conductivity detector. Aspartic acid was used as the standard. A nitrogen-to-protein conversion factor of 6.25 was used in the calculations. A total protein content of 11.7 % for non-mutated parental oat cultivar Belinda was used as control.

Extraction of avenins

Albumins and globulins were extracted with 200 µL of 1 mol/L Nacl in 0.1 mol/L sodium phosphate buffer, pH 7 and temperature 4 °C. The crude extract was centrifuged (15 min, 6000 rpm). After centrifugation, 120 µL of supernatant was pipetted into another microcentrifuge tube and lyophilized for 6 h. After separation of albumins and globulins, the lyophilized material was extracted with 25 % 2-chloroethanol (200 µL per grain) for 4 h at the room temperature. The samples were then centrifuged (15 min, 6000 rpm). After centrifugation, 120 µL of supernatant was pipetted into a fresh microcentrifuge tube and lyophilized for 6 h. This has been shown to be the optimum concentration of 2-chloroethanol for extraction of avenin. After centrifugation, the clear supernatant was diluted with two volumes of pre-cooled 1.5 % NaCl solution and the precipitate separated by centrifugation at 6000 rpm for 30 min. The sample was then dried under vacuum in a desiccator.

Preparation of samples for SDS-PAGE electrophoresis

The avenin protein fractions in 25 samples under study were separated by SDS-PAGE under reducing conditions using 8–25 % gradient Phast® gels (GE HealthCare) and a GE PhastGel system (GE Healthcare, Uppsala, Sweden). The SDS-PAGE conditions were 250 V, 10.0 mA, 3.0 W, 15 °C, and 65 Vh. The gels were developed at room temperature using PlusOne Silver Staining Kit (GE Healthcare, Uppsala, Sweden) and according to the provided instructions. After staining, gels were routinely air-dried at room temperature between two sheets of wetted, untreated cellophane and clamped in an acrylic plastic frame.

Statistics

GraphPad Prism Program (GraphPad Softwares, SanDiego, CA) was applied for the analysis of unpaired t test. Mean and standard error values were calculated using Excel software [26, 27].

Results

Protein content in the wild BOCs ranged from 11.6 ± 0.01 to 12.9 ± 0.07 and was similar when the BOCs were grown in the field (in Sweden 2009–2013) or in green house (2011–2015) (Table 2). The same oat cultivar grown in China exhibited significantly higher protein levels, 12.01 ± 0.1 to 14.1 ± 0.05, compared to those grown in Sweden, thereby showing the effect of geographical location on protein content. The variation in protein content in the BOCs grown in greenhouses, where growing parameters are controlled, was less than field-grown BOCs. Mutated BOLs, or those cultivars treated with N2, had significantly higher protein levels, clearly showing the effect of treatment on protein levels (Sunilkumar B et al., manuscript under preparation). In the mutated lines, the protein content ranged from 17.3 to 20.2 %. Similarly, BOLs treated with N2 fertilizers showed higher protein levels and highest variation in protein content, ranging from 11.7 to 24 %.

Table 2 Total protein percentage of all Belinda oat lines (BOLs) analyzed using particle elemental analyzer
Full size table

A total of 5–8 protein bands were resolved among the five groups numbered I, II, III, IV, and V (Table 2) in the SDS-PAGE gels stained with silver staining procedure, as shown in Fig. 1a. Only one group of cultivars, obtained from the Swedish field, possessed all eight bands, which is shown in the diagrammatic representation of respective alleles based on SDS-PAGE gel, shown in Fig. 1b. Cultivars from the greenhouse, mutated as well as cultivars grown under excess N2, had four bands. The cultivar obtained from China had five bands. As shown in Fig. 1b, four distinct loci were present in all groups and four additional loci (one dense and three mild bands) in group I, which represent the cultivars grown in Swedish fields. Molecular weight of all major four polypeptides ranged from 15 to 33 kDa (numbered D to I in Fig. 1b). The molecular weight in the three additional bands in the cultivar from the Swedish fields ranged from 40 to 66 kDa (Fig. 1a, group I). Three major bands and one minor band consistently stained heavily (mild line region), appearing in all oat lines examined. The cultivars grown in China showed an additional minor band with a molecular weight of 25 kDa, shown by an arrow in the gel picture (Fig. 1b, lane 3).

Fig. 1
figure 1

a Avenin polyacrylamide gel electrophoresis pattern of (I). Belinda oat cultivars from Swedish field from the year 2009–2013 (II). Five different Belinda cultivars grown in green house (III). Five different Belinda cultivars grown in different location from China (IV). Mutated Belinda oat lines (V). Five Belinda plants treated with nitrogen fertilizers. An arrow showed in group III third lines is the nonallelic loci. b Diagrammatic representations of SDS-PAGE banding patterns of the respective allelic blocks which indicated (by arrows) in the figure of SDS-PAGE gel image. The lines numbered 1, 2, 3, 4, and 5 represent groups I, II, III, IV, and V, respectively, shown in (a). Each protein bands were numbered from A to I. Intensely stained bands are numbered using alphabet with high font size and mildly stained bands with alphabet having small font size

Full size image

Discussion

Although the cultivars and mutated lines originated from a Swedish spring oat Belinda cultivar, it is not surprising to find considerable similarities in electrophoretic pattern as well as total protein percentage. However, total protein determination showed that the cultivars obtained from Sweden in different years (group I), greenhouse (group II) and from different provinces of China (group III) showed change in protein percentage that was either higher or lower than the 12 % seen in the original SW Belinda. However, the SDS-PAGE electrophoresis of avenin protein fraction showed eight allelic loci for cultivar grown in Sweden compared with five loci for groups II and III, so the Swedish cultivars were markedly different from the other groups. A small increase in protein percent was observed in the Chinese cultivars, but the variation between lines was not great (12–14 %) compared with cultivars from the greenhouse. Both have the same avenin loci, despite a minor band being found in one of the Chinese lines obtained from the province of Shanghai. According to the previous reports, some of the avenin alleles are associated with geographical origin of the cultivars, and SDS-PAGE showed alleles with alternative protein bands with different mobility [3]; such minor bands may occur in cultivars from Sweden as well as from China.

A significant increase in protein percentage was found in cultivars grown under different treatments. However, both showed nearly identical nature of electrophoretic patterns for avenin loci, indicating no such differences in avenin protein fraction obtained after mutation and fertilizer application. Evidence from the previous experiments showed that fertilization with high amounts of N2 increased globulin relative to other fractions of oat protein [28]. Consequently, an increase in protein content here may be due to the capacity for maximum absorption of nitrogen in the container-grown plants in the greenhouse due to less percolation compared with soil in the field, which allows maximum permeation or percolation [29]. This could probably increase the globulin fractions that eventually gave a high protein percentage. Similarly, in mutated BOLs, the possibility of alteration in protein encoding genes that modulate certain biosynthetic pathways could overproduce protein [30]. These results indicated that the effect of mutation and nitrogen fertilizers on protein content was more pronounced than the effect from geographical factors. However, another discriminatory method must be developed to resolve avenin gene families for variety identification in oats.

Production of genetically pure seed is an important criterion for any seed certification program. Identification of oat cultivars using electrophoresis offers many advantages over some of the current field testing procedures, which are purely based on morphological characteristics such as color and size of the plant, and the presence or absence of awns [31]. Even so, ISTA 1999 and other authors [22, 32, 33] reported that SDS-PAGE patterns can be useful to differentiate oat species, only requiring simple laboratory equipment and minimum laboratory supplies. Development of such tests can prove beneficial for developing and protecting high-quality cereal grains, especially for self-fertilizing species [34]. Developing an appropriate oat cultivar and its further acceptability in the market depends solely on the yield, maturity, disease resistance, and nutritional value for food, feed, and functionality [35]. Our results from this study will help us develop new high-protein oat lines that are stable and environmentally sound, with reasonable grain quality, by using those genotypes showing consistency in protein content with distinct banding pattern through a crossing program.

Conclusions

Five different groups studied here showed a distinct banding pattern for avenin. The difference in cultivars obtained from the field related to other four groups showed the clear indication of adaptive strategy and appeared to be evidence of an ecological pattern of variation in avenin. This needs to be considered when different breeding strategies are applied to produce cultivars adapted to one or more locations for food and feed.