Barley is an important cereal crop, globally grown for animal feed use and as a source of fermentable sugars in brewing and distilled beverages production [1
]. Barley was one of the first cereals recognised as a food that provides vital nutrients and energy to maintain body functions and health. Cultivation of barley for food purposes represents only 2% of current global production, but in some countries with extreme climates, in Asia, Himalayan nations, and North Africa, it remains a staple cereal food source [2
Barley is naturally high in dietary fibre, especially β-glucan and phenolic compounds, which have the potential to lower cholesterol and blood glucose levels and helps gut microbial balance [1
]. Mixed linked β-D-glucans, together with arabinoxylans, are prevailing fibre constituents of the barley endosperm and aleurone cell walls [5
]. It is a characteristic of barley that its β-glucan is distributed throughout the aleurone and endosperm rather than being confined to outer branny layers [6
]. The β-glucan content of barley grain is mainly determined by genotype and less by environmental conditions during the grain filling period, but a significant influence of both genetic and location factors on barley β-glucan content was also noticed [2
]. Significant differences in β-glucan content among barley types with various starch amylose contents were observed. The high-amylose starch type barleys contained higher amounts of β-glucan than the waxy, zero, and normal amylose endosperm types [7
]. It was reported that hull-less atypical amylose varieties had a significantly higher concentration of β-glucan than the normal amylose ones. Further, the difference in β-glucan content between the hull-less and hulled varieties of normal starch type was not significant, but there was an observed tendency towards a higher β-glucan content in hull-less barley samples [5
In addition to providing high amounts of dietary fibre, especially soluble β-glucan, barley grains are rich in protein, complex carbohydrates, minerals, and fat-soluble vitamin E (tocols) [1
]. Due to the high content of proteins, the second major component of dry grain, barley is an affordable source of food proteins. According to their extractability characteristics (classification method developed by Osborne), barley grain proteins can be classified into four groups, i.e., albumins, globulins, hordeins, and glutelins. Albumins and globulins are mainly located in the bran and embryo, thus, called cytoplasmic proteins. They comprise 3–4% and 10–20% of the total seed proteins, respectively. Hordeins belong to prolamin protein fraction, an alcohol-extractable protein fraction of barley grain that accounts, on average, up to for up to 60% of the total grain nitrogen, which makes them the main barley endosperm storage proteins. In accordance with electrophoretic mobility and amino acid composition, the hordeins are classified into five groups of polypeptides, called A-hordein, B-hordein (70–90%), C-hordein (10–20%), D- and γ-hordein (less than 5%). Hordeins and glutelins are mainly found in the starchy endosperm, together contributing to about 70–90% of the total dry seed proteins [2
Barley also contains significant content of phenolic compounds, naturally occurring antioxidants with antiradical and antiproliferative potentials [13
]. Phenolic compounds constitute a large heterogeneous group of compounds that can be classified in different ways, based on their chemical structure, occurrence, and distribution in nature, and according to their location in the plant tissue [4
]. Phenolic acids are the most abundant phenolic group of phytochemicals in barley, existing dominantly in the form bound to compounds of the cell wall, followed by conjugated and free forms. Free forms of phenolic acids in barley account for a small portion of the total phenolic acid concentration, and they are mainly ferulic acid, vanillic acid, syringic acid, and p-coumaric acid [13
]. It was reported that phenolic acid concentration on dry basis in barley approximately ranges between 4.6 and 23 mg/g for the free form, between 86 and 198 mg/g for the conjugated form, and between 133 and 523 mg/g for the bound form, whereas the total phenolic acid concentration ranges between 604 and 1346 mg/g [18
]. The most present phenolic acid found in barley grain is ferulic acid, accounting approximately 68% of total phenolic acids in barley, and its concentration in barley grains ranges between 149 and 413 mg/g [20
]. Researchers have linked a diet high in polyphenols and their metabolites to the maintenance of the gut microbial balance and a reduced risk of developing some chronic diseases [4
Cereal grains have a long history of usage for both human and animal consumption and as a raw material for industrial purposes. For consumption food, milling of grains and producing of the flours is often the first step in preparing daily meals in many parts of the world. Different milling processes applied are associated with nation consuming customs and characteristics of the cereals used to produce various mill streams. The milling process can basically be described as a sequence of procedures involving grinding, sifting, and separation and regrinding in order to separate particular parts of the grain [22
]. The initial studies on roller milling of hull-less barley have often been designed to follow the already-installed procedure of wheat milling, focused on flour yield and flour composition [23
]. More recently, several studies were dealing with hull-less barley milling in order to produce products other than standard flour which are enriched in valuable dietary fibre originating from endosperm cell walls and other bioactive phytochemicals. It has been brought to researchers’ attention that the coarse material coming out from the final reduction passage, appointed as “shorts” in wheat milling, originates mainly from endosperm cell walls in barley milling, and is enriched in β-glucans and other fibre constituents. To reflect its composition, and to distinguish it from bran fraction, this coarse fraction resulting from hull-less barley roller milling procedure was designated as a fibre-rich fraction [25
]. Nowadays, consumers’ awareness of the need for healthier life habits is growing, and reformulation of a daily diet in terms of nutritional quality can certainly contribute. Hull-less barley has been increasingly researched because of its beneficial composition, which can be utilised in developing of functional food products. It is a cereal that could be easily milled to produce flours of different compositional characteristics using common milling systems. When compared to wheat, the milling processing of barley is still relatively uncommon and less developed to be widely recognised [27
]. Supplementing wheat flour with hull-less barley, either as a whole grain or as a flour addition, especially when enriched in dietary fibre, phenolic compounds, and protein content, could help enhance the value of the basic raw material for traditional bakery recipes or diverse pasta products.
The present study was concerned with the milling of hull-less barley and the analytical characterisation of hull-less barley milling streams and flours. In addition to conventional flour and millstream characterisation mainly focusing on complete chemical analysis and health-promoting fibres (β-glucans), in this study, the distribution of important functional phytonutrients, such as phenolic acids and specific proteins, were determined.
2. Materials and Methods
Two hull-less barley varieties, Mandatar (recognised 2017) and Osvit (recognised 2014), created at the Agricultural Institute Osijek, were included in this study. They were sown as winter barley in an area of the Slavonia region (45.5550° N, 18.6955° E), which belongs to the eastern part of Croatia, characterised by a moderate continental climate. The field trials with hull-less barley were installed in fall of 2017, and grains were harvested in the early summer of 2018. During harvesting, the hull was easily separated from the grain. The naked grains were collected and stored until milling analysis. Physical characteristics of the grains were estimated. Samples were analysed for thousand kernel weight and hectolitre weight, accordingly.
2.2. Milling Procedure
The grain samples of hull-less barleys were subjected to the roller milling process using laboratory milling equipment. The flour research laboratory at the Faculty of Food Technology Osijek (Osijek, Croatia) uses Bühler MLU 202 laboratory mill (Bühler AG, Uzwill, Switzerland), which is mostly used for wheat samples testing. Samples for roller milling were prepared by weighing 3 kg of each hull-less barley. Prior to milling, the barley grains were cleaned to ensure the samples were free from any impurities and conditioned to 14.0%, w
moisture content. Conditioning was performed in two steps. The first step was conditioning to 13.5%, w
moisture content, by adding a calculated amount of water to the grain, thoroughly mixing to evenly distribute moisture throughout the grain, and storing it for 24 h in a sealed container at the room temperature. Leaving it overnight allows water to penetrate the kernel, and in order to adjust the intended moisture content of 14% (w
), the remaining amount of water was added half an hour before milling. The mill uses a six-roll system to make three breaks (B1–B3), three reductions (C1–C3) of refined flours, bran, and shorts [30
]. The flour fractions were separated from bran and shorts on sieves with following mesh openings: 710 (30), 600 (36W), 530 (40W), 180 (8XXX), and 150 µm (9XXX) and using preset feed rate and roll gap settings. Samples were prepared and milled as duplicates.
2.3. Composition of Hull-Less Barley Fractions
Ash content was determined by incineration in a muffle furnace at 550 °C for 3 h [31
]. Protein content was determined according to the Kjeldahl method [32
]; Kjeltec 2300 Analyser Unit, Foss Analytical AB, Höganäs, Sweden). Fat content was analysed by the Soxhlet extraction method. Starch content was determined according to the Ewers polarimetric method [33
2.4. β-Glucan Content
The total β-glucan content was determined according to the method reported by Mc Cleary and Glennie [34
] using a Megazyme mixed-linkage β-glucan assay kit (Megazyme Ltd., Bray, Ireland).
2.5. Extraction of Samples
Extracts for the determination of total phenolic content and antioxidant activity were prepared by weighing 2 g of hull-less barley flour fraction and mixed with 5 mL of acidified methanol (HCl/methanol, 1:100, v/v). The mixture was homogenised by vortex for 2 min, placed afterwards in a Sonorex Digitec ultrasonic bath (Model RK510H; Bandelin, Germany) at room temperature and sonicated for 1 h. It was then centrifuged at 9000 rpm for 5 min at 4 °C (Universal 320R; Hettich, Germany). After centrifugation, the supernatant was removed, and the extraction procedure of the residue was repeated twice, firstly with 5 mL of acidified methanol and then with 2 mL with a time of sonication of 30 min. The supernatants were pooled. To avoid oxidation, extracts (12 mL) were stored in the dark at −20 °C, and analyses were performed within the next 24 h. All samples were prepared and analysed as triplicates.
2.6. Total Phenolic Content (TPC)
The TPC in hull-less barley flour extracts was determined with Folin-Ciocalteu reagent using the method described by Singleton and Rossi [35
], with some modifications. To 0.1 mL of the extract obtained above, 5 mL of ten-fold diluted Folin-Ciocalteu’s phenol reagent and 0.9 mL of distilled water were added. The mixture was shaken, mixed with 4 mL of 7.5% Na2
, and left for incubation for 2 h at room temperature in the dark. The absorbance was read at 765 nm. The absorbance value was measured with a UV/VIS-Spekol spectrophotometer (Analytik Jena AG, Jena, Germany). Total phenolic content was expressed as milligrams of gallic acid equivalents per gram of dry weight (mg GAE/gdw
) using a calibration curve constructed with gallic acid (50–1000 µg/mL). All samples were analysed as triplicates.
2.7. Antioxidant Activity (DPPH Radical Scavenging Activity)
Antioxidant activity was measured using a modified version of the method explained by Brand-Williams, Cuvelier, and Berset [36
]. This involved the use of free radical 2,2-diphenyl-1-picrylhydrazyl (DPPH) solution in the methanol. Every sample extract (0.2 mL) was reacted with 1 mL of a 0.5 mmol/L methanol solution of DPPH and 2 mL of methanol. The reaction mixture was shaken and incubated in the dark. The absorbance (A) of the solution was measured against a methanol blank at 517 nm after 30 min. Stable DPPH radical reaches the absorbance maximum at 517 nm, and its colour is purple. The change of this colour into yellow is a result of the pairing of an unpaired electron of a DPPH radical with the hydrogen of the antioxidant, thus generating reduced DPPH-H. Adding an antioxidant results in the decrease of absorbance, that is proportional to the concentration and antioxidant activity of the compound. Lower absorbance of the reaction mixture indicates higher free radical scavenging activity. Antioxidant activity was calculated as inhibition of free radical DPPH in per cent (%)
by using the following equation:
% Antioxidant activity = (1 − (A of samplet=30/A of controlt=0)) × 100
All samples were analysed as triplicates.
2.8. Extraction of Albumins/Globulins and Hordeins
Barley protein extractions were done according to the method of Celus et al. [37
], with some modifications: for extraction of albumins and globulins, 50 mg of the sample with 1 mL of 4 M sodium chloride were vortexed for 5 min at 3500 rpm (VibroMix 204 EV, Tehtnica, Slovenia), incubated at 25 °C for 30 min at 500 rpm (Thermomixer 5436, Eppendorf, Germany) and centrifuged (Centrifuge 5415 C, Eppendorf, Germany) for 10 min at 14000 rpm. The supernatant was decanted, filtered in glass vials through 0.45 μm polyvinylidene fluoride (PVDF) membrane filter (Ahlstrom GmbH, Germany) and stored at −20 °C until analysis. Hordeins were extracted from the pellet residue by adding 1.0 mL of 50% 1-propanol + 1% dithiothreitol, vortexed for 5 min, incubated at 60 °C for 60 min, and centrifuged for 15 min. The supernatant was decanted, filtered in glass vials through 0.45 μm PVDF membrane filter and stored at −20 °C until analysis. All extractions were done as duplicates.
2.9. HPLC Analysis of Protein Fractions
Analysis of protein fractions was carried out using a Series 200 HPLC system (Perkin Elmer, Waltham, MA, USA) coupled with Discovery BioWide Pore RP-C18 column (4.6 × 150 mm, 300 Å, 5 μm). The mobile phase consisted of Millipore water with 1% trifluoroacetic acid (v/v) (A) and acetonitrile with 1% trifluoroacetic acid (v/v) (B). The column temperature was 50 °C and injection volume was 20 uL. The gradient elution profile was as follows: from 24% to 58% B in 30 min, isocratic at 90% B for 5 min, returning to the initial conditions in 5 min, and column equilibration in 5 min. The flow rate was controlled at 1.0 mL/min. The peaks were detected at 210 nm with a photodiode array detector. The groups of peaks under chromatograms belonging to D-, B-, C-hordeins and their proportion was expressed as a % of total hordeins area. For the calibration of HPLC absorbance area and the calculation of the albumins/globulins and total hordeins concentration, bovine serum albumin was used as a protein standard, and consequently, a calibration curve of five points (5–120 ug, r = 0.999) was constructed. All samples were analysed as duplicates.
2.10. Extraction of Phenolic Acids
The improved microscale method of Zavala‑Lopez and Garcia‑Lara [38
] was used for phenolic acid extraction. Free phenolic acids were extracted as follows: 150 mg of hull-less barley flour fractions with 2.1 mL of 80% methanol were vortexed for 5 min at 3500 rpm (VibroMix 204 EV, Tehtnica, Slovenia). The sample was incubated at 25 °C for 15 min at 500 rpm (IKA KS 260 basic, Germany) and centrifuged (Universal 320R, Hettich, Germany) for 10 min at 5000 rpm. The supernatant was decanted and stored at −20 °C until analysis. The bound phenolic acids were extracted from the pellet residue as follows: 1.5 mL of 2 M NaOH was added and vortexed for 5 min at 3500 rpm. The alkaline hydrolysis was conducted at 90 °C for 1 h at 250 rpm (shaking water baths GFL 1092, Burgwedel, Germany). After hydrolysis, the sample was acidified with 1.5 mL of 2 M HCl at pH 2 (adjusted with 2 M HCl using pH strips). Lipids were removed by adding 2.4 mL of n-hexane to the samples, vortexed for 5 min at 3500 rpm, incubated at 25 °C for 10 min at 500 rpm, and centrifuged for 10 min at 5000 rpm. The hexane layer (upper phase) from the three-layered system was discarded. This procedure was repeated two more times. The bound phenolic acids were recovered by adding 2.4 mL of ethyl acetate, vortexed for 5 min at 3500 rpm, incubated at 25 °C for 10 min at 500 rpm, and centrifuged for 10 min at 5000 rpm. The ethyl acetate layer (upper phase) from the three-layered system formed was collected. The previous procedure with ethyl acetate washes was repeated two more times. The pooled ethyl acetate was evaporated to dryness (BÜCHI B-720 Vacuum Controller, Flawil, Switzerland), re-suspended in 600 μL of 50% methanol, and stored at −20 °C until analysis.
2.11. HPLC Analysis of Phenolic Acids
Analysis of free and bound phenolic acids was carried out using a Series 200 HPLC system (Perkin Elmer, Waltham, MA, USA) coupled with Kinetex Core-Shell RP-C18 column (150 × 4.6 mm, 100 Å, 5 um) according to Matilla et al., with some modifications [39
]. Prior to analysis, samples were filtered through 0.2 μm nylon filter (Ahlstrom GmbH, Bärenstein, Germany). The mobile phase consisted of an A—Millipore water acidified with 1% trifluoroacetic acid (v
) and B—acetonitrile acidified with 1% trifluoroacetic acid (v
). The column temperature was 30 °C and injection volume was 20 uL. The gradient elution profile was as follows: from 5% to 40% B in 40 min, isocratic at 90% B for 5 min, returning to the initial conditions for 5 min and column equilibration for 5 min. The flow rate was controlled at 1.0 mL/min. The peaks were detected at 275 nm with a photodiode array detector. Peak identification was based on the retention times and spectral data of standards (190–400 nm). Calibration curves of five points were made by diluting stock standards (1 mg/mL) in methanol where all phenolic acids showed a linear response within the studied range (R2
> 0.999). All samples were prepared and analysed as duplicates.
2.12. Statistical Analysis
The experimental data were processed by one-way ANOVA using Statistica ver. 13.0 software (Stat Soft Inc., Tulsa, OK, USA) and PAST: paleontological statistics software package for PCA analysis. Significant differences among means (post-hoc tests) were calculated using Fisher’s least significant difference (LSD) test at the significance level of α = 0.05.