Comparative spectroscopic and rheological studies on crude and purified soluble barley and oat β-glucan preparations

Abstract

The rheology of crude and purified barley (BBG) and oat (OBG) β-glucan samples were characterized. Sample content and major impurities was characterized by Fourier-transform near infrared Raman and infrared (FT-IR) spectroscopy revealing substantial differences between the β-glucan samples. The purification procedure increased the β-glucan content from 66.7 to 82.4% and from 30.1 to 68.4% for BBG and OBG, respectively. Proton nuclear magnetic resonance (¹H NMR) analysis was applied to estimate the β-(1 → 3) to β-(1 → 4) linkage ratio of the β-glucans. The molar mass of BBG and OBG was determined by high performance size-exclusion chromatography (HPSEC) using β-glucan standards and was found to be 126 and 355 kDa, respectively.

The viscosity of crude and purified β-glucans was studied at various concentrations (2.5, 5% w/v), temperatures (10–80 °C) and shear rates (1–100 s^− 1). BBG was characterized as a low-viscosity β-glucan with Newtonian flow behavior while OBG was characterized as a high-viscosity β-glucan with shear thinning flow behavior. At equivalent β-glucan concentration in solutions the viscosity for OBG was found to be ~ 100 fold higher than for BBG. A direct viscosity dependence on exact β-glucan content regardless of amount and composition of α-glucan impurities was found for both OBG and BBG. This study suggests that the structural characteristics of the β-glucan polymers such as molar mass are of greater functional importance than the presence of lager amounts of starch/α-dextrins as long as the β-glucan samples are compared at equivalent β-glucan doses.

Introduction

Health benefits and hydrocolloid functionality of mixed linkage (1 → 3, 1 → 4)-β-D-glucans from cereals make them interesting food constituents and they have been suggested as ingredients in health-promoting functional foods (Brennan and Cleary, 2005, Ebringerova et al., 2005, Inglett et al., 2004, Lazaridou and Biliaderis, 2007, Lee et al., 2005, Temelli et al., 2004, Volikakis et al., 2004). However, β-glucan sample composition and β-glucan structure are crucial characteristics affecting the functionality in food systems and human health. In complex food systems β-glucan functionality may be affected by interactions or incompatibilities between β-glucan and other food components (Lee et al., 1995, Satrapai and Suphantharika, 2007) and detailed knowledge about structure/function relationships of β-glucan containing systems is required in order to exploit the full health beneficial properties of this important dietary fiber.

Traditionally plant cell wall structure has been analyzed using chemical agents that gently extract and purify specific components. The cell walls are typically analyzed for sugars and sugar linkages using chemical (methylation analysis) and physical techniques (¹H and ¹³C NMR). Additional information on the repetitive units in the polymers can be gained from the use of glycanases that cleave specific glycosidic bonds and subsequent analysis of the oligosaccharides by chromatographic and/or mass analysis (Guillon, Saulnier, Robert, Thibault, & Champ, 2007). Cereal β-glucan polymers generally contain ~ 70% (1 → 4)-β-glucosyl residues and ~ 30% (1 → 3)-β-glucosyl residues. The (1 → 3)-linkages occur singly, linking together three (cellotriosyl unit) or four (cellotetraosyl unit) (1 → 4)-β-linked sequences and to a lesser extend longer cellulose-like fragments (Ebringerova et al., 2005). Structurally related β-glucans differ in the ratio of tri- and tetraosyl units which is 2.8–3.3 for β-glucan of barley and 2.1–2.4 for oat (Lazaridou et al., 2004, Wood et al., 1994).

Nuclear magnetic resonance (NMR) spectroscopy has been widely used for the overall structure and linkage sequence analysis of cereal β-glucans (Cui et al., 2000, Johansson et al., 2004, Lazaridou et al., 2004, Morgan et al., 1999, Petersen et al., 2000, Roubroeks et al., 2000, Seefeldt et al., 2008, Wood et al., 1994). Liquid state ¹H NMR allows estimation of the ratio of β-(1 → 3) to β-(1 → 4) linkages in the β-glucan polymers as well as the residual constituents present in the β-glucan extract. Fourier-transform infrared (FT-IR) spectroscopy has also been successfully used for the analysis of plant cell wall polysaccharides including β-glucans (Johansson et al., 2004, Kacurakova and Wilson, 2001, Seefeldt et al., 2009). Raman spectroscopy has a great potential in analyzing plant cell wall polysaccharides (Engelsen and Norgaard, 1996, Goral and Zichy, 1990, Mohacek-Grosev et al., 2001, Salomonsen et al., 2008, Sene et al., 1994) but to the best of our knowledge, structural and compositional cereal β-glucan analysis by Raman spectroscopy has not been reported before. Both IR and Raman spectroscopic methods are rapid, sensitive and able to distinguish the α- and β-conformations of saccharides (Cael et al., 1974, Engelsen and Norgaard, 1996, Kacurakova et al., 2000). In contrast to chemical, chromatographic, or mass spectrometric methods, spectroscopic methods require no chemical extraction as the intact sample matrix is examined which is an advantage when analyzing delicate and complex biological samples (Seefeldt et al., 2008, Seefeldt et al., 2009).

The molar mass of a β-glucan polysaccharide is an important characteristic as it determines the physicochemical properties such as viscosity (Ebringerova et al., 2005). Molar mass values for oat and barley β-glucans have been reported to range between 44–3000 and 126–2500 kDa, respectively (Beer et al., 1997, Ebringerova et al., 2005, Lazaridou et al., 2004, Papageorgiou et al., 2005, Vaikousi et al., 2004, Wood et al., 1991). The values strongly depend on the method of extraction and analysis which makes comparisons difficult. Increasing the extraction temperature leads to an increase in molar mass of the extracted β-glucan (Izydorczyk et al., 1998, Zhang et al., 1998; Temelli, 1997). Zhang et al. (1998) extracted oat β-glucan at 40, 65 and 100 °C and reported molar mass values of 118–1024 kDa at 40 °C, 985–1919 kDa at 65 °C and 2300 kDa at 100 °C. Size-exclusion chromatography (SEC) is often used in β-glucan size determination (Johansson et al., 2004, Morgan and Ofman, 1998). SEC require calibration standards for the estimation of the molar masses unless it use a molar mass sensitive detection system such as LALLS, MALLS or RALLS (Christensen et al., 2001). Various studies have discussed the use of β-glucan, pullulan or dextran standards as references in molar mass determination of β-glucans (Beer et al., 1997, Christensen et al., 2001, Roubroeks et al., 2000, Varum et al., 1991, Zhang et al., 1998). It is well-known that calibration with pullulan standards leads to overestimation of the molar mass, however, pullulan standards offer a greater range of molar masses compared to β-glucan standards, especially for large molecules (Christensen et al., 2001, Varum et al., 1991).

Rheological methods such as viscometry have been widely used in studies of β-glucan flow behavior (Burkus and Temelli, 2005, Dawkins and Nnanna, 1995, Doublier and Wood, 1995, Lazaridou et al., 2003). Increasing the concentration or the molar mass of a β-glucan polymer generally results in increased viscosity (Anttila et al., 2004, Wood, 2004). Often, β-glucan from oat exhibits somewhat higher viscosity compared to barley β-glucan due to longer polymer chains (Beer et al., 1997). Johansson, Karesoja, Ekholma, Virkki, and Tenhu (2008) compared viscosities of equal size oat and barley β-glucans and the higher viscosity of the oat β-glucan was suggested to stem from differences in the fine structure i.e. the ratio of tri- and tetraosyl units. Shear thinning flow behavior of β-glucan solutions is an established fact (Burkus & Temelli, 2005). In addition, low-viscosity β-glucan extracts and high-viscosity β-glucan extracts at low concentrations often show Newtonian flow behavior (Autio et al., 1987, Burkus and Temelli, 2005, Doublier and Wood, 1995). Since starch is one of the major components of foods, understanding the mechanism of interaction of β-glucan with native starch and its hydrolytic products and its implication for rheological properties is of interest (Faraj et al., 2006, Grimm et al., 1995, Lee et al., 1995, Satrapai and Suphantharika, 2007). Grimm et al. (1995) studied the aggregation of β-glucan molecules in aqueous maltose solutions and found a minimum of aggregation near 5% maltose. This was attributed to a preferential binding of maltose to β-glucan, which partly breaks up the aggregated β-glucan clusters. Faraj et al. (2006) investigated the influence of hydrolysed starch fractions (low, medium and high molar mass) on the solution viscosities of low and high purity barley β-glucans. None of the hydrolysates affected the high purity β-glucan viscosity whereas the viscosity of the low purity β-glucan increased in the presence of the medium molar mass starch fractions. It was concluded that some non-β-glucan components in the low purity β-glucan may influence the solution viscosity of β-glucan-hydrolysed starch blends.

The aim of the present study was to conduct comparative spectroscopic, chromatographic and rheological studies on β-glucan concentrate samples. Special interest in these samples was the detection and characterization of α-glucans and to study the effect of α-glucans on the viscous properties of the sample solutions. Two of the β-glucan samples were recently investigated in a dialysis study targeted at studying the β-glucan affinity with small molecules (Simonsen et al., 2009).