Detection of Proteins in Starch Granule Channels Xian-Zhong Han,1 Mustapha Benmoussa,1 Jonathan A. Gray,1 James N. BeMiller,1 and Bruce R. Hamaker1,2 ABSTRACT
Cereal Chem. 82(4):351–355
Proteins were detected in channels of commercial starches of normal maize, waxy maize, sorghum, and wheat through labeling with a proteinspecific dye and examination using confocal laser scanning microscopy (CLSM). The dye, specifically 3-(4-carboxybenzoyl)quinoline-2-carboxaldehyde (CBQCA), fluoresces only after it reacts with primary amines in proteins, and CLSM detects fluorescence-labeled protein distribution in an optical section of a starch granule while it is still in an intact state. Starch granules in thin sections of maize kernels also had channel proteins, indicating that proteins are native to the channels and not artifacts of isolation. Incubation of maize starch with protease (thermolysin) re-
moved channel proteins, showing that channels are open to the external environment. SDS-PAGE analysis of total protein from gelatinized commercial waxy maize starch revealed two major proteins of about Mr 38,000 and 40,000, both of which disappeared after thermolysin digestion of raw starch. Commercial waxy maize starch granule surface and channel proteins were extracted by SDS-PAGE sample buffer without gelatinization of the granules. The major Mr 40,000 band was identified by MALDI-TOF-MS and N-terminal sequence analysis as brittle-1 (bt1) protein.
Pores in starch granule surfaces were first reported by Hall and Sayre (1970), who concluded that they were artifacts. Fannon et al (1992) first presented evidence that pores in starch granules are real structural features and not artifacts of processing or preparation of specimens for scanning electron microscopy (SEM). (For the history of the discovery and investigations of starch granule channels to date, see Fannon et al 2003). Huber and BeMiller (1997) provided unequivocal evidence that channels of maize and sorghum starch granules connect the granules’ central cavity to their external environment. Preliminary, but not unequivocal, evidence was presented that channels are present in maize starch granules in very early stages of granule development (Huber and BeMiller 2000). The presence of pores and channels in starch granules could have an effect on granule accessibility to reagents and, hence, reactivity of starch granules when starch is chemically modified to meet specific usage in industry (BeMiller 1997; Huber and BeMiller 1997). The pattern of digestion of native starch granules by α-amylase is also greatly influenced by the existence of pores and channels. Maize starch, which contains pores and channels, is more susceptible to enzymatic digestion than is potato starch (Leach and Schoch 1961; Gallant et al 1973; Fuwa et al 1977; Kanenaga et al 1990), which does not have pores (Fannon et al 1992). It has been reported that enzymatic digestion of maize starch initiates at the hilum (Leach and Schoch 1961; Nikuni 1978; Hood and Liboff 1983), and it has been recognized that enzymes might enter the granule interior through pores and channels (Hall and Sayre 1970; Fannon et al 1992). Gunawan (2002) found that, in most cases, the rate of digestion by glucoamylase is proportional to the average degree of channelization of a maize starch cultivar. Heretofore, starch granule proteins have been classified as granule surface and granule-bound proteins (Gillian et al 1981; Schofield and Greenwell 1987). Only recently have the precise locations of granule internal proteins been shown (Han and Hamaker 2002a). Using the dye 3-(4-carboxybenzoyl)quinoline2-carboxaldehyde (CBQCA) and confocal laser scanning microscopy (CLSM), Han and Hamaker (2002a) found that proteins were located on the granule surface and in certain concentric spheres and the central cavity, and appeared to line maize channels. Granule-bound starch synthase (GBSS), in particular, existed in certain, but not all, concentric spheres in potato and maize starches
and in indistinct concentric spheres in wheat starch granules. Gray and BeMiller (2004) showed a lessening of silver staining of channels following thermolysin digestion that they suggested could be due to proteins. Our aim in this study was to verify and study the presence of starch granule channel proteins. Knowledge of specific channel proteins and their role in creating channels may lead to manipulation of the degree of channelization for potential improvements in digestibility and chemical modification.
1 Whistler
Center for Carbohydrate Research and Department of Food Science, 745 Agriculture Mall Drive, Purdue University, West Lafayette, IN 47907-2009. 2 Corresponding author. Fax: 765-494-7953. E-mail:
[email protected] DOI: 10.1094 / CC-82-0351 © 2005 AACC International, Inc.
MATERIALS AND METHODS Materials Commercial normal and waxy maize starches were provided by Tate and Lyle North America, Decatur, IL. Commercial wheat starch was provided by Midwest Grain Products, Atchison, KS. A sorghum cultivar (high protein digestibility/high lysine mutant) was obtained from the Purdue University Agronomy Farm. Maize and sorghum starches were laboratory-isolated using the toluene method (Banks and Greenwood 1975). Coomassie Brilliant Blue R-250 was purchased from Sigma Chemical Co. (St. Louis, MO). CBQCA protein quantitation kit (C-6667) containing 3-(4-carboxybenzoyl)quinoline-2-carboxyaldehyde was purchased from Molecular Probes (Eugene, OR). Protein Labeling and Confocal Laser Scanning Microscopy Starch samples (7 mg) were suspended in 135 µL of 0.1M sodium borate buffer (pH 9.3), combined with 5 µL of KCN (20 mM), and mixed well by vortexing. CBQCA (40 mM) stock solution was prepared using CBQCA dissolved in dimethylsulfoxide. A working solution of CBQCA (5 mM), made by diluting the stock solution with 0.1M sodium borate buffer (pH 9.3), was used immediately. The CBQCA working solution (10 µL) was added to the starch suspension. After incubation for 3–5 hr, each sample was loaded onto a slide and transferred to the confocal microscope stage. The locations of fluorescent-labeled protein associated with starch granules were observed using CLSM (BioRad MRC-1024, Hercules, CA). A krypton-argon laser at an excitation wavelength of 488 nm was used. CLSM digital images were acquired using the BioRad LaserSharp program. Digestion of Channel Proteins by Thermolysin The method for thermolysin digestion of channel proteins was based on that of Mu-Foster et al (1996). Starch (50 mg) was digested using 0.2% thermolysin (Sigma Co., St. Louis, MO) (starch basis) in 1 mL of 5 mM CaCl2 solution. After 30 min, the digestion was terminated by addition of EDTA (20 mM). Starch was then dried in a forced air oven (37°C) for 12 hr. Vol. 82, No. 4, 2005
351
Protein Extraction, SDS-PAGE, and Protein N-Terminal Sequencing For extraction of surface protein from raw starch, starch (5 g) was added to 15 mL of SDS-PAGE buffer containing 2% 2mercaptoethanol, 2% SDS, 10% glycerol (v/v), 66 mM Tris (pH 6.8), and 0.2 mL of 0.2% bromophenol blue, and shaken for 48 hr. The starch suspension was then centrifuged at 10,000 × g for 15 min. The supernatant was retained. This procedure was repeated twice more by adding 5 g of starch to the supernatant, shaking the mixture for 48 hr, and centrifuging to obtain the protein extract. The protein extract was dialyzed in a pleated dialysis tube with molecular weight cutoff of 10,000 (product #68100, Pierce, Rockford, IL). The extract was then freeze-dried and redissolved in the same sample buffer (0.3 mL). Protein samples (30 µL) were fractionated using a 12% (w/v) slab gel and electrophoresed in a minigel apparatus (BioRad Corp., Van Nuys, CA). Gels were stained with Coomassie Brilliant Blue R-250. For N-terminal sequencing, the extraction method was modified. Starch (5 g) was added to 20 mL of extraction buffer containing 2% 2-mercaptoethanol, 2% SDS, 66 mM Tris (pH 6.8), and mixed using a shaker for 6 hr. Extract was centrifuged at 27,000 × g for 15 min at 4°C and supernatant was concentrated in Centricon centrifugal filter devices (YM-10, Amicon, Bedford, MA) by centrifugation at 5000 × g. Samples were boiled for 3 min in the sample buffer (0.0625M Tris-HCl, pH 6.8; 2.3% SDS, 10% [v/v] glycerol; 0.05% [w/v] bromophenol blue; and 5.0% [v/v] mercaptoethanol). The resulting solution was loaded on 1810 gradient gels. Gels were stained with Coomassie Brilliant Blue R-250. The molecular weights of proteins were determined using protein molecular weight standards from Bio-Rad Laboratories. The resolved polypeptides were electroblotted onto Trans-Blot
MALDI-TOF-MS Characterization of the 40-kDa Protein Proteins were identified by peptide mass mapping with matrixassisted laser-desorption-ionization, time-of-flight mass spectrometry (MALDI-TOF-MS). After destaining the SDS-PAGE gel in 30% methanol, the band of interest was washed 5× with 50% acetonitrile, 25 mM ammonium bicarbonate, pH 8.0. The solution was discarded after each wash. The gel slice was incubated in 50 µL of 50 mM dithiothreitol (DTT) in 100 mM ammonium bicarbonate at 56°C for 30 min. The DTT solution was removed and replaced with 50 µL of 100 mM iodoacetamide, and the mixture was incubated for 30 min at 45°C. The solvent was discarded, and the gel slice was washed with 500 µL of 50% acetonitrile/100 mM ammonium bicarbonate, pH 8.9, for 1 hr.
Fig. 1. Confocal laser scanning micrographs of optical sections of commercial normal (A) and waxy (B) maize starch granules. Proteins in channels are shown by fluorescence (bright white dots [large arrows] and white lines [small arrows]) produced by reaction of the protein (primary amine) with CBQCA, which produces a fluorescent compound.
Fig. 3. Confocal laser scanning micrographs of optical sections of sorghum (A) and commercial wheat (B) starches. Proteins in channels are shown by fluorescence (bright white dots and lines) through reaction with the protein (primary amine)-specific CBQCA dye.
Fig. 2. Confocal laser scanning micrographs of optical sections of commercial normal (A) and waxy (B) maize starch granules at higher magnification. Proteins in channels and central areas are shown by fluorescence (bright white regions) through reaction with the protein (primary amine)-specific CBQCA dye.
Fig. 4. Confocal laser scanning micrographs of optical sections of thin slices of optical sections of commercial normal (A) and waxy (B) maize kernels. Proteins in channels are shown by fluorescence (bright white regions of protein surrounding and within granules) through reaction with the protein (primary amine)-specific CBQCA dye.
352
CEREAL CHEMISTRY
polyvinylidene difluoride membranes (Bio-Rad Laboratories) and visualized by Coomassie Brilliant Blue R-250 staining. N-Terminal sequences were obtained by automated Edman degradation on an Applied Biosystems Precise 492 (ABI). Sequences were determined by automatic alignment to known protein sequences in a protein sequence database (NCBI-Blast search system, http:// www.ncbi.nih.gov/BLAST/ ). For extraction of total protein from gelatinized starch granules, 15 mg of starch was added to 0.6 mL of sample buffer containing 2% 2-mercaptoethanol, 2% SDS, 10% glycerol (v/v), 66 mM Tris (pH 6.8), and 0.2 mL of 0.2% bromophenol blue. Suspensions were boiled for 20 min with intermittent stirring and 20 µL was loaded into each well of a polyacrylamide slab minigel consisting of 4% stacking gels and 12% separating gels. The gel was run for 30 min at a constant voltage of 150V. Gels were stained using Coomassie Brilliant Blue R-250 or silver stain reagent and photographed.
The solvent was discarded, and the gel was cut into 2–3 pieces and transferred to a 200-µL microcentrifuge tube. Acetonitrile (50 µL) was added to shrink the gel pieces. After 10–15 min, the solvent was discarded, and the gel slices were dried in a rotatory evaporator. The gel pieces were reswollen with 10 µL of 25 mM ammonium bicarbonate containing 0.15 µL of trypsin (Promega Corp., Madison, WI). After 10–15 min, additional buffer (10–20 µL) was added to cover the gel pieces, and the gel slices were incubated overnight at 37°C. The peptides were extracted from the gel by twice adding 50 µL of 60% acetonitrile/0.1%TFA/water and sonicating for 15 min. Supernatants were pooled and concentrated to 20 µL. The samples were desalted using C18 Zip-Tip columns (Millipore, Framingham, MA), and the purified peptides were spotted on MALDI target plates with α-cyano-4-hydroxycinnamic acid. Mass spectra were acquired on a Reflex III MALDITOF mass spectrometer (Bruker Daltonics, Billerica, MA). Protein identification was performed by searching the ExPASy (http://us.expasy.org) protein sequence database. The following parameters were used for database searches: maize, molecular weight from 36–54 kDa, at least four peptides required to match, cysteine treated with iodoacetamide to form carbamidomethylcysteine, and maximum missed cleavages of 2.
To identify the principal proteins in channels, raw waxy maize starch granules were first treated with thermolysin to remove channel proteins. Untreated and protease-treated granules were then gelatinized, protein was extracted, and electrophoretic banding patterns were compared. The resulting silver-stained SDS-PAGE gel revealed that two protein bands of about Mr 38,000 and 40,000 (Fig. 6, lane 1) were completely digested by thermolysin (Fig. 6, lane 2). The low molecular weight proteins (
