A Critical Assessment of the Quantification of Wheat Grain Arabinoxylans Using a Phloroglucinol Colorimetric Assay

Cereal Chem. 89(3):143–150 Arabinoxylans (AX) of wheat (Triticum aestivum L.) play a critical role in processing, end-use quality, and human health and nutrition. Consequently, an efficient, accurate method of AX quantification is desirable. The objective of this work was to evaluate a standard phloroglucinol colorimetric method for quantification of wheat AX. The method is based on the formation and spectrophotometric quantification of a phloroglucide product that results from the reaction of furfural produced during the condensation of pentose sugars with phloroglucinol. Method parameters, including reaction reagents and reaction times, were varied to identify areas for improved accuracy and consistency. Phloroglucide formation at three xylose concentrations was examined over time. The optimal reaction reagents and reaction times were determined based upon improved consistency in xylose quantification. The optimized method was used on xylose and arabinose standards and on whole meal wheat samples for total and water-extractable AX content. Glucose was shown to be unnecessary in the reaction and was eliminated. A second-order polynomial equation provided a slightly better fit to the nearly linear standard xylose curve. A reduced concentration of phloroglucinol of 10% was found to give equivalent results to the standard 20%. Optimum reaction time was 25 min, and it required the inclusion of all reagents. The phloroglucide product decreased in absorbance over time such that, within the range of xylose concentration examined, about 40–50% of the colored product was lost over 100 min; however, the rate of loss was linear over time. Four operators performed the optimized method on whole wheat meal samples for total and water-extractable AX. Interand intraoperator variation was identified as an area requiring further study and improvement. However, all operators tended to rank the samples in a consistent manner. Compared with a gas chromatography–flame ionization detection method, the phloroglucinol method underestimated total AX by about 2.3% and water-extractable AX by about 0.08%. The nonstarch carbohydrates of cereal grains are important for food processing, end-product quality, and human health and nutrition. In wheat (Triticum aestivum L.) grain, nonstarch carbohydrates are primarily associated with cell walls. Cell walls are composed of up to 75% nonstarch carbohydrates, which are mostly (85%) substituted pentose polymers referred to as arabinoxylans (AX) (Mares and Stone 1973). Other nonstarch carbohydrates include cellulose, lignin, glucomannans, and β-glucans (Lineback and Rasper 1988). AX are composed of a β-1,4 linked D-xylopyranosyl backbone with substituted monomeric α-L-arabinofuranoside at the second and/or third carbon positions. The arabinose moiety can possess ferulic acid at the fifth carbon position (Courtin and Delcour 2002). The three-dimensional structure of AX is mainly determined by the length of the xylan backbone, the ratio of arabinose to xylose, the substitution pattern on the backbone, and the ferulic acid coupling to other AX molecules or the cell wall (Courtin and Delcour 2002). AX have a somewhat flexible structure (Dervilly et al 2000). These structural characteristics and the large molecular weight (65,000) (Andrewartha et al 1979) contribute to empirically derived subfractions based on extractability (generally in water at room temperature), namely, water-extractable (WEAX) and water-unextractable AX (Izydorczyk and Biliaderis 1992; Courtin and Delcour 2002). AX also have the capacity to form oxidative cross-links and gel networks (Izydorczyk and Biliaderis 1992) and affect end-product quality (Bettge and Morris 2007; Ramseyer et al 2011a, 2011b), thus increasing their overall molecular weight and exerting different effects on food processing and human physiology. AX exhibit nutritive properties as well as influencing end-use quality (Lu et al 2000). The postprandial glucose concentration was decreased after healthy individuals ate bread products with AX as an additive (Lu et al 2000). AX added to bread products also lowered the glucose and insulin levels of the blood 2 hr after consumption in individuals with type 2 diabetes (Lu et al 2004). There are several techniques to quantify AX, including colorimetric assays (Wheeler and Tollens 1889; Dische and Borenfreund 1957; Douglas 1981; Ford 1981; Bell 1985) and gas chromatography–flame ionization detection (GC-FID) (Gebruers et al 2009). Some structural characteristics of AX can be determined by GC-FID (Gebruers et al 2009), enzyme mapping (Saulnier and Quemener 2009), molecular weight distribution (Andersson et al 2009), Fourier transform infrared spectroscopy (Toole et al 2009), Raman spectroscopy (Toole et al 2009), fluorescence microscopy (Toole et al 2009), and carbohydrate-binding molecules (Toole et al 2009). For quantification, phloroglucinol has been the most commonly used colorimetric assay since its first description in 1889 (Wheeler and Tollens 1889), as it can measure five-carbon monosaccharides. The method was later extended to quantify hydrolyzed pentosans (Kröber 1901; Kröber et al 1902). This general method was used with phloroglucinol as the reactant and with a distillation step to capture furfural until 1957 (Dische and Borenfreund 1957), when the distillation was eliminated, and the reagent concentrations and timing evolved into the widely used Douglas (1981) method. This method is among the quickest and most efficient methods for the quantification of total AX (TAX) and WEAX. The biggest disadvantage observed in this method, however, was the loss of colored product over time, which decreased the accuracy of the test by roughly 20% in 60 min (Douglas 1981). Nevertheless, the method of Douglas (1981) has been particularly popular among cereal chemists. According to information provided by ISI Web of Science (Thomson Reuters, New York, NY), this method has been cited approximately 110 times. Rouau and Surget (1994) developed a semiautomated version of the assay. The phloroglucinol method of Douglas (1981) has been used to determine the AX content of wheat flour (Hashimoto et al 1987; 1 Department of Crop and Soil Sciences, Washington State University, Pullman, WA 99164-6376; affiliated with the USDA-ARS Western Wheat Quality Laboratory. 2 Laboratory of Food Chemistry and Biochemistry, University of Leuven, Kasteelpark Arenberg 20, B-3001 Leuven, Belgium. 3 U.S. Department of Agriculture (USDA), Agricultural Research Service, Western Wheat Quality Laboratory, E-202 Food Quality Bldg., Washington State University, P.O. Box 646394, Pullman, WA 99164-6394. Names are necessary to report factually on available data; however, the USDA neither guarantees nor warrants the standard of the product, and the use of the name by the USDA implies no approval of the product to the exclusion of others that may also be suitable. 4 Corresponding author. Phone: (509) 335-4062. Fax: (509) 335-8573. E-mail: