Pearling Barley and Rye to Produce Phytosterol-Rich Fractions Anna-Maija Lampia,*, Robert A. Moreaub, Vieno Piironena, and Kevin B. Hicksb aDepartment of Applied Chemistry and Microbiology, University of Helsinki, Finland, and bUSDA, ARS, Eastern Regional Research Center, Wyndmoor, Pennsylvania 19038, ABSTRACT: Because of the positive health effects of phyto- in the kernels and are more concentrated in the outer layers sterols, phytosterol-enriched foods and foods containing than in the starch-rich endosperm (6,7). During the milling of elevated levels of natural phytosterols are being developed. some grains, pearling is a traditional way of gradually remov- Phytosterol contents in cereals are moderate, whereas their lev- ing the hull, pericarp, and other outer layers of the kernels and els in the outer layers of the kernels are higher. The phytosterols germ as pearling fines to produce pearled grains. It is the most in cereals are currently underutilized; thus, there is a need to common technique used to fractionate barley (8). The abra- create or identify processing fractions that are enriched in sion of rye and barley to produce high-starch pearled grains phytosterols. In this study, pearling of hulless barley and rye was investigated as a potential process to make fractions with higher also has been used to improve fuel ethanol production (9,10). levels of phytosterols. The grains were pearled with a labora- There is a need to find new food uses for the pearling fines tory-scale pearler to produce pearling fines and pearled grains. and other possible low-starch by-products remaining after Lipids were extracted by accelerated solvent extraction, and separation of the high-starch pearled grain. The objective of nonpolar lipids were analyzed by normal-phase HPLC with this study was to evaluate pearling as a potential process to ELSD and UV detection. Total sterol analyses were performed make fractions with higher levels of cereal phytosterols. The by GC. After a 90-s pearling, the amounts of pearling fines from cereals studied were hulless barley and rye. hulless barley and rye were 14.6 and 20.1%, respectively, of the original kernel weights. During pearling, higher levels of phytosterols and other lipids were fractionated into the fines. MATERIALS AND METHODS The contents of free sterols and sterols esterified with FA in the The grains used for pearling were a new winter hulless barley fines were at least double those in the whole grains. Pearling fines of hulless barley and rye contained >2 mg/g phytosterol variety, Doyce, released in 2003 by the Virginia Polytechnic compounds, which makes them a good source of phytosterols Institute and State University and grown in Virginia. The rye and thus valuable raw materials for health-promoting foods. variety, Flesynt, was from the North Florida Research and Ed- Paper no. L9521 in Lipids 39, 783–787 (August 2004). ucation Center, Institute of Food and Agricultural Services, University of Florida (Quincy, FL). Rye grains were also hul- less. Both grains were harvested in the 2003 season, and the Phytosterols are being studied extensively because of their moisture content of the grains ranged between 11 and 14%. positive health effects. Research projects have focused on Grains were pearled with a laboratory-scale barley pearler their biological functions, safety, and chemical and physical (30 grit carborundum stone, no. 7 mesh screen; Seedburo properties, as well as on attempts to develop new phytosterol- Equipment Co., Chicago, IL). For each pearling experiment, enriched foods (1,2). As the result of a recent workshop on 50 g of grains were pearled, producing a fraction of pearling sterols and stanols with 26 leading researchers as participants fines and a mixture of broken kernels and pearled grains. (i.e., the Stresa Workshop), an extensive review was pub- The grains were first pearled sequentially to learn how phyto- lished on the effects of using phytosterols and stanols to con- sterols were localized in the kernels and to determine a relevant trol serum cholesterol and the safety of phytosterol and stanol time for the production of pearling fractions. Sequential pearling enrichments in foods (3). The authors concluded that daily consisted of five steps of 30 s each. After each pearling, fines consumption of 2 g of sterols or stanols decreases serum LDL were separated from the broken kernels by passing them cholesterol levels by approximately 10%. through an 18-mesh sieve and collected as the product. Pearled Cereals are considered to be a good source of dietary grains and broken kernels were combined and pearled again. To phytosterols. Although their levels in whole grains are mod- obtain enough material, each pearling sequence was repeated erate (0.4–1.2 mg/g) (4,5), the total amount of phytosterols is five times, and the five fractions of pearling fines were extracted significant because of the large amounts of cereals consumed. and analyzed separately. In the second experiment, the grains Phytosterols, like many other bioactive compounds (e.g., to- were pearled for 90 s. Pearling fines and pearled grains from copherols, tocotrienols, and folates), are unevenly distributed three separate pearlings were combined and subjected to further analysis. Each pearling experiment was performed in duplicate. *To whom correspondence should be addressed at Dept. of Applied Chem- istry and Microbiology, Latokartanonkaari 11, P.O. Box 27, FIN-00014 Uni- The grains and pearled grains were ground to 20 mesh in a versity of Helsinki, Finland. E-mail: anna-maija.lampi@helsinki.fi Wiley mill (Thomas Scientific, Swedesboro, NJ) before lipid Copyright © 2004 by AOCS Press 783 Lipids, Vol. 39, no. 8 (2004) 784 A.-M. LAMPI ET AL. extraction. The pearling fines were extracted without addi- the study and were 1.8 min for sterols esterified with FA (St- tional grinding. Lipids were extracted using a Dionex ASE FA), 4.5 min for TAG, 9.2–10.6 min for FFA, 21.2 min for 200 accelerated solvent extractor (ASE) (Sunnyvale, CA) free stanols, 22.0 min for free sterols, and 26.6 min for sterols with 2.0-g sample sizes and 11-mL extraction vessels as esterified with ferulic acid (St-Fer). All three lipid extracts for reported earlier (11). In this study, the extraction mixture each sample were analyzed for nonpolar lipids. consisted of hexane and isopropanol (3:2, vol/vol) (12). The Total phytosterols of the grains and pearling fines were an- extracts were used for gravimetric measurements of lipid ex- alyzed by GC using FID after acid and alkaline hydrolysis (5). tracts and nonpolar lipid analyses by HPLC (13). The extracts Pearling fractions were also analyzed for moisture and ash were dried under nitrogen at ≤40°C and weighed for total ex- using AACC official methods AACC 44-15A and AACC tractable lipids. The extracts were redissolved in hexane for 08-01 (14), respectively. Nonpolar lipid and total phytosterol nonpolar lipid analysis and filtered through 0.2-µm polyvinyli- results are presented as means and SD derived from three sub- dene fluoride filters (Acrodisc LC 13; Pall Gelman Labora- samples, and moisture and ash are given as means derived tory, Ann Arbor, MI) when needed. Each grain sample was from two subsamples. extracted in triplicate. In each extraction batch, whole-grain rye flour (Hodgson Mill Inc., Effingham, IL) was included as RESULTS AND DISCUSSION an in-house reference sample to monitor the extraction proce- dure. Extraction efficacy was also verified by spiking the Evaluation of the lipid analysis method. Reproducibility of same flour with stigmasterol (95%; Sigma Chemical Co., St. the lipid extraction method was examined by analyzing non- Louis, MO) and calculating its recovery. polar lipids of whole-grain rye flour at least once in each ex- Nonpolar lipids were analyzed by normal-phase HPLC traction batch during the study (N = 23). The contents of with LiChrosorb DIOL (5 µm, 3 × 100 mm) columns TAG, St-FA, stanols, sterols, and St-Fer were 3.65 ± 0.18, (Chrompack, Raritan, NJ) using instruments and running 0.74 ± 0.03, 0.06 ± 0.01, 0.20 ± 0.01, and 0.06 ± 0.01 mg/g, conditions as presented by Moreau et al. (13) except that the respectively. Recovery of spiked stigmasterol from the flour gradient was slightly modified. The linear gradient elution was 98% (N = 6). Detector responses of the nonpolar lipid consisted of three steps followed by a 20-min stabilization pe- analysis remained stable; the contents of TAG, St-FA, and St- riod. In step 1 (0–8 min), the eluent consisted of 100% sol- Fer of corn fiber oil were 656 ± 22 mg/g (N = 29), 44.7 ± 2.9 vent A (hexane with 0.1% acetic acid); in step 2 (8–10 min), mg/g (N = 59), and 44.9 ± 2.0 mg/g (N = 59), respectively. the proportion of solvent B (hexane with 1% isopropanol) Sequential pearling. When hulless barley and rye were was increased to 25%; and in step 3 (10–40 min), the eluent pearled sequentially, the pearling fine yields were approxi- consisted of 75% solvent A and 25% solvent B. The flow rate mately 7 and 5%, respectively, from each 30-s pearling step. was 0.5 mL/min. All lipids were detected using ELSD (All- Barley grains contained free sterols and St-FA (Fig. 1). The tech-Varex Mark III; Alltech Assoc., Deerfield, IL) except for total amount of sterol compounds in the whole barley grains phytosterols esterified with ferulic acid, which were detected was 0.70 mg/g. The highest level, 2.8 mg/g, was obtained in using UV at 280 nm. An external standard method was used the fines from the first pearling step. The total sterol content of for quantification (13).