Article pubs.acs.org/JAFC
Immunological Characterization of the Gluten Fractions and Their Hydrolysates from Wheat, Rye and Barley Prasad Rallabhandi,*,† Girdhari M. Sharma,† Marion Pereira, and Kristina M. Williams Immunobiology Branch, Office of Applied Research and Safety Assessment, Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 8301 Muirkirk Road, Laurel, Maryland 20708, United States ABSTRACT: Gluten proteins in wheat, rye and barley cause celiac disease, an autoimmune disorder of the small intestine, which affects approximately 1% of the world population. Gluten is comprised of prolamin and glutelin. Since avoidance of dietary gluten is the only option for celiac patients, a sensitive gluten detection and quantitation method is warranted. Most regulatory agencies have set a threshold of 20 ppm gluten in foods labeled gluten-free, based on the currently available ELISA methods. However, these methods may exhibit differences in gluten quantitation from different gluten-containing grains. In this study, prolamin and glutelin fractions were isolated from wheat, rye, barley, oats and corn. Intact and pepsin-trypsin (PT)-digested prolamin and glutelin fractions were used to assess their immunoreactivity and gluten recovery by three sandwich and two competitive ELISA kits. The Western blots revealed varied affinity of ELISA antibodies to gluten-containing grain proteins and no reactivity to oat and corn proteins. ELISA results showed considerable variation in gluten recoveries from both intact and PT-digested gluten fractions among different kits. Prolamin fractions showed higher gluten recovery compared to their respective glutelin fractions. Among prolamins, barley exhibited higher recovery compared to wheat and rye with most of the ELISA kits used. Hydrolysis resulted in reduced gluten recovery of most gluten fractions. These results suggest that the suitability of ELISA for accurate gluten quantitation is dependent upon various factors, such as grain source, antibody specificity, gluten proteins and the level of their hydrolysis in foods. KEYWORDS: gluten, ELISA, celiac disease, gluten-free, wheat, allergens
■
INTRODUCTION
Cereal proteins can be broadly classified based on their solubility.16 Gluten is comprised of two major protein fractions, prolamins (alcohol-soluble) and glutelins (acid/alkali-soluble). Gluten proteins are classified based on sulfur-content (S-rich or S-poor), molecular weight and protein subunits (monomeric or polymeric).17,18 While prolamins are monomeric with intramolecular disulfide bonds, glutelins are multimeric with both intra- and inter-molecular disulfide bonds. Both prolamins and glutelins are proline and glutamine rich, which render gluten highly resistant to proteolytic degradation.19 Prolamins are termed gliadin in wheat, secalin in rye, hordein in barley, and avenin in oat. While prolamins from wheat, rye and barley have been attributed as the causative proteins of CD, there is emerging evidence that glutelins, which share some amino acid sequences with prolamins,2 can also contribute to the manifestation of CD in susceptible individuals.20−23 Immunoassays are routinely used to detect and quantify gluten in foods, and enzyme-linked immunosorbent assay (ELISA) is the most commonly used method for gluten quantitation. Several ELISA kits, based on different monoclonal and polyclonal antibodies in both sandwich and competitive formats, are commercially available. But, the differences among these ELISA kits (for example, sample extraction buffer, extraction conditions, calibration standards, and antibody specificity) may contribute to a substantial variation resulting
Gluten, which gives viscoelasticity to dough, is a heterogeneous complex of proteins present in the endosperm of wheat, rye, barley and their hybrids.1,2 These cereal grains are widely used in many processed foods. Gluten has been identified as a causative agent of celiac disease (CD), an autoimmune disorder. CD leads to villous atrophy of the small intestine, resulting in various gastrointestinal and systemic complications in celiac patients.3,4 CD affects approximately 1% of the general population.5,6 Furthermore, wheat gluten and other wheat proteins, such as albumins and globulins, have been shown to cause an IgE-mediated allergic response in susceptible individuals.7,8 Since there is no cure or treatment for CD, strict avoidance of dietary gluten is the only option for celiac/ gluten-sensitive patients.9 In order to protect the majority of patients, many regulatory agencies have set a gluten threshold of 20 ppm for foods labeled “gluten-free”. Pure oats are generally believed to be tolerated by a majority of celiac patients,10,11 even though oat proteins from specific cultivars have been shown to be toxic in a fraction of susceptible individuals.12−14 Undeclared gluten can be inadvertently present in foods through cross-contact and mislabeling. Recently, Gendel and Zhu (2013) reported that wheat ranks second among food allergen recalls, following milk, even though these recalls are not gluten-specific and do not include rye and barley related recalls.15 Sensitive and accurate analytical methods for gluten quantitation in foods are extremely important for “gluten-free” labeling compliance in order to ensure consumer safety. This article not subject to U.S. Copyright. Published 2015 by the American Chemical Society
Received: Revised: Accepted: Published: 1825
November 25, 2014 January 20, 2015 January 25, 2015 January 26, 2015 DOI: 10.1021/jf505716p J. Agric. Food Chem. 2015, 63, 1825−1832
Journal of Agricultural and Food Chemistry
Article
in over- or underestimation of gluten.24 The gluten quantitation by ELISA can be further affected by thermal25−27 and nonthermal28−30 food processing conditions. Since most ELISA methods use gliadin as a calibration standard, the calculated gluten content would be derived from the measured gliadin.31,32 However, such quantitation may not give an accurate measure for other prolamins (secalin and hordein) and glutelins. Prolamins and glutelins differ in their proportions and molecular properties in wheat, rye and barley.33,34 This will influence the availability of gluten epitopes, affecting antigen−antibody interaction in immunoassays and causing variability in gluten quantitation. Moreover, hydrolysis of gluten proteins may reduce the number of available epitopes leading to underestimation of gluten content. Hence, it is important to assess the contribution of prolamin and glutelin fractions and their hydrolysates to facilitate accurate gluten quantitation. So far, there is limited information available on the analysis of prolamins from rye and barley.30 Furthermore, the quantitation of glutelin fractions from wheat, rye and barley is not well studied. The aim of this study was to assess the relative recovery of prolamin and glutelin fractions and their pepsin-trypsin (PT)-digests based on the gluten quantitation by various commercial ELISA kits.
■
changes of 0.1 M acetic acid. All the fractions were lyophilized and stored at −20 °C until further use. The protein content of all lyophilized fractions was analyzed by micro-Kjeldahl method (MicroAnalysis, Inc. Wilmington, DE) using the conversion factor of 5.83. Pepsin-Trypsin (PT)-Digestion. Approximately 250 mg protein was dissolved in 2.5 mL pepsin solution (2 mg/mL prepared in 0.2 N HCl), followed by addition of 2.5 mL 0.2 N HCl. The enzyme to protein ratio was 1:50 (w/w). The tubes were incubated at 37 °C for 2 h in shaking water bath. The peptic digest was then adjusted to pH 7.4 with 2 M NaOH, followed by addition of 50 μL trypsin solution (100 mg/mL prepared in distilled water) to have an enzyme to protein ratio of 1:50 (w/w). The tubes were mixed and incubated at 37 °C for 4 h in shaking water bath. Thereafter, the tubes were transferred to a boiling water bath and incubated for 30 min to inactivate the enzymes. Appropriate controls (S1, Sigma gliadin; S2, Sigma gluten; S3, Sigma casein) were also used. Ten milligram aliquots of PT-digest were lyophilized and stored at −20 °C until further use. Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) and Western Blotting. Protein samples were boiled in 1x SDS-PAGE Laemmli sample buffer (Biorad, Hercules, CA) for 10 min, and suitable aliquots were loaded on large 1.5 mm thick 10−20% SDS-PAGE Tris-Glycine gel (Jule, Inc., Milford, CT). The gels were run at 50 mA current per gel for 1.5 h, and then reduced to 30 mA per gel until the dye reached the bottom edge of the gel. The gels were stained overnight with 0.25% w/v Coomassie Brilliant Blue R containing 50% v/v methanol and 10% v/v acetic acid. The gels were destained once with 50% v/v methanol containing 10% v/v acetic acid (destain buffer), followed by destaining buffer diluted with distilled water (∼1:1 v/v) until the background was clear. For the Western blot analysis, the polypeptides were separated on 10−20% SDS-PAGE Tris-Glycine gels (Invitrogen, Grand Island, NY) by running at 100 V until the dye reached the bottom edge. The separated polypeptides were transferred to a polyvinylidene fluoride (PVDF) membrane at 0.4A current for 1.5−2 h at 8 °C. The transferred proteins were visualized by Ponceau S stain. The membrane was washed with Tris buffered saline containing 0.05% Tween 20 (TBS-T) to remove the stain and then blocked for 1 h at RT with 5% nonfat dried milk in TBS-T. The membrane was briefly washed with TBS-T and incubated with anti-gluten antibodies conjugated to horseradish peroxidase (HRP) overnight at 4 °C. The membrane was then washed thrice for 15 min each with TBS-T. The reactive bands on the membrane were developed on X-ray film using ECL substrate (Pierce). Estimation of Gluten by Enzyme-Linked Immunosorbent Assay (ELISA). Both undigested and PT-digested prolamins and glutelins were quantified using sandwich and competitive ELISA kits as per the manufacturers’ instructions. Briefly, lyophilized gluten fractions were dissolved in the extraction buffers of respective kits, and were used as stock solutions to prepare dilutions of appropriate protein content (10−100 ng/mL), based on calculated protein amounts analyzed by micro-Kjeldhal method. Appropriate controls (Sigma gliadin, S1; Sigma gluten, S2; Sigma casein, S3) were also used. The ELISA plates were read using a SpectraMax M5 microplate reader (Molecular Devices, Sunnyvale, CA). A four parameter standard curve was plotted using the SoftMax Pro 5.4 software, and was used to calculate the measured gluten/wheat protein. Further, the recovery was represented as a percent ratio of the measured gluten to the actual protein assayed. For all ELISAs, two samples of each protein fraction were analyzed in duplicate (n = 4), and the data were presented as mean ± standard error (SE).
MATERIALS AND METHODS
Materials. Pure grains of wheat, rye, barley, oats and corn were obtained from the Grain Inspection, Packers & Stockyards Administration (GIPSA), U.S. Department of Agriculture (USDA). Commercial gliadin (S1), wheat gluten (S2), casein (S3), pepsin and trypsin were from Sigma (St. Louis, MO). The sandwich ELISA kits used were RIDASCREEN Gliadin (R7001; R-Biopharm AG, Darmstadt, Germany), Wheat protein ELISA kit (gliadin) (181 GD; Morinaga Institute of Biological Science, Inc., Yokohama, Japan) and AgraQuant Gluten G12 (COKAL0200; Romer Labs UK Ltd., Cheshire, UK), whereas the competitive ELISA kits included RIDASCREEN Gliadin competitive (R7021; R-Biopharm AG, Darmstadt, Germany) and GlutenTox ELISA Competitive (KT4758; Biomedal, Spain). BenchMark prestained protein ladder was purchased from Invitrogen (Grand Island, NY). Methods. Isolation of Gluten. The grains of wheat, rye, barley, oats and corn were ground to flour and then independently subjected to Osborne fractionation as described by Tatham et al. (2000)17 with minor modifications. Briefly, 100 g flour was suspended in 1000 mL 0.5 M NaCl and magnetically stirred for 1 h at room temperature (RT), followed by centrifugation at 5,000g for 10 min at 20 °C. The extraction was repeated again with the pellet by suspending in 0.5 M NaCl and the two supernatants (containing albumins and globulins) were pooled before dialyzing against distilled water for a total of 48 h with at least 6 water changes using an 8 kDa cutoff dialysis membrane. The globulins were separated by centrifuging at 10,000g for 15 min at 4 °C. Thereafter, all centrifugation steps were carried out at 5,000g for 10 min at 20 °C. The residual flour from NaCl extraction was washed with water for 10 min at RT followed by centrifugation. The washed residual flour was then resuspended twice in 1000 mL 70% ethanol (for oats, 50% ethanol) and magnetically stirred for 1 h at RT, followed by centrifugation. The pooled supernatant containing prolamins was precipitated overnight at 4 °C with 2 volumes of 1.5 M NaCl, centrifuged, resuspended the pellet in 100 mL 70% ethanol (for oats, 50% ethanol) and dialyzed against 0.1 M acetic acid for 48 h with at least 6 changes of 0.1 M acetic acid. Glutelins were finally extracted by resuspending the residual flour from ethanol extraction twice in 1000 mL 50% propan-1-ol containing 1% acetic acid and 2% β- mercaptoethanol for 1 h at RT, followed by centrifugation. The pooled supernatant was precipitated overnight at 4 °C with 2 volumes of 1.5 M NaCl, centrifuged, resuspended the pellet in 100 mL 50% propan-1-ol containing 1% acetic acid and 2% β- mercaptoethanol, followed by dialysis against 0.1 M acetic acid for 48 h with at least 6
■
RESULTS Gluten Fractions and Their PT-Digests. The protein content of lyophilized prolamin and glutelin fractions from wheat, rye and barley were estimated by the micro-Kjeldahl method. Lower protein content was observed in oat (prolamin, 33% and glutelin, 59%) and corn (prolamin, 42% and glutelin, 64%) fractions as compared to wheat (prolamin, 68% and glutelin, 82%), rye (prolamin, 52% and glutelin, 71%) and 1826
DOI: 10.1021/jf505716p J. Agric. Food Chem. 2015, 63, 1825−1832
Journal of Agricultural and Food Chemistry
Article
Figure 1. Coomassie stained SDS-PAGE gels of prolamin and glutelin fractions. Intact (A) and PT-digested (B) protein fractions. Protein load was 20 μg/lane for undigested and 50 μg/lane for PT-digested fractions. P, prolamin; G, glutelin; S1, gliadin; S2, gluten; S3, casein; M, MW marker; (−), pepsin-trypsin only.
Figure 2. Analysis of gluten fractions and their PT-digests using the R-Biopharm sandwich ELISA kit and its antibodies. Recoveries of intact gluten fractions (A) and their PT-digests (B) in ELISA. Western blot analysis of flour proteins (C) and their gluten fractions (D) using the R-Biopharm kit antibodies. For ELISA, 25 ng/mL of sample extract was used, except that 100 ng/mL of PT-digests were used when the measured value was less than LOQ. W, wheat; R, rye; B, barley; O, oat; C, corn; P, prolamin; G, glutelin; PT, pepsin-trypsin.
measure the percent gluten recovery from various gluten fractions and their hydrolysates. Sandwich ELISA. In general, the S1 and S2 controls behaved similar to wheat gluten fractions, as they are isolated from wheat. The average measured value for S3 and all oat and corn protein fractions used to calculate the percent gluten recovery was found to be lower than the limit of quantitation (LOQ) of the respective sandwich ELISA. The LOQs are 5 ppm gluten (R-Biopharm), 0.3 ppm wheat protein (Morinaga) and 4 ppm gluten (Romer Labs). Barley flour protein showed lower immunoreactivity than wheat and rye in Western blot. However, a high gluten recovery was observed with barley prolamin by ELISA. More detailed results on immunoreactivity of different grain protein fractions are described below. With the R-Biopharm kit, the wheat gluten fractions showed a lower gluten recovery as compared to recoveries from rye and barley. The percent gluten recovery of prolamin and glutelin ranged from 65.8 (wheat) to 248.3 (barley) and 25.4 (wheat) to 97.0 (barley), respectively (Figure 2A). However, upon PTdigestion, the recovery was further reduced as compared to the undigested proteins, and it was 46.5 (wheat) to 185.8 (barley)
barley (prolamin, 62% and glutelin, 71%). All experiments in this study were carried out based on the micro-Kjeldahl estimated protein contents. The yield of prolamin and glutelin proteins per 100 g of flour was 1.44 and 2.66 g (wheat), 0.61 and 0.81 g (rye), 1.41 and 1.7 g (barley), 0.79 and 2.78 g (oat), and 0.55 and 1.27 g (corn), respectively. The purity and protein composition of the prolamin and glutelin fractions of different grains were analyzed on a 10−20% SDS-PAGE gel (Figure 1A). Gluten fractions of wheat, rye, barley and oats consisted of polypeptides in the expected molecular weight range in agreement with the previous reports.2,17,35,36 Upon treating with pepsin and trypsin (PT), the gluten fractions from all grains were digested predominantly to low molecular weight polypeptides (
