Báo cáo khoa học: Digestive a-amylases of the flour moth Ephestia kuehniella – adaptation to alkaline environment and plant inhibitors

Digestive a-amylases of the flour moth Ephestia kuehniella – adaptation to alkaline environment and plant inhibitors Jana Pytelkova1, Jan Hubert2, Martin Lepsık1, Jan Sobotnık1, Radek Sindelka3, Iva Krızkova2, Martin Horn1 and Michael Mares1 1 Institute of Organic Chemistry and Biochemistry AS CR, v.v.i., Praha, Czech Republic 2 Research Institute of Crop Production, v.v.i., Praha, Czech Republic 3 Institute of Biotechnology AS CR, v.v.i., Praha, Czech Republic Keywords alkaline adaptation; a-amylase; a-amylase inhibitor; Ephestia kuehniella; plant–insect interaction Correspondence M. Mares, Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, Flemingovo nam. 2, 166 10, Praha 6, Czech Republic Fax: +420 220183578 Tel: +420 220183358 E-mail: mares@uochb.cas.cz (Received 12 January 2009, revised 18 March 2009, accepted 24 April 2009) doi:10.1111/j.1742-4658.2009.07074.x The digestive tract of lepidopteran insects is extremely alkaline. In the pres-ent work, molecular adaptation of amylolytic enzymes to this environment was investigated in the flour moth Ephestia kuehniella, an important stored-product pest. Three digestive a-amylases [Ephestia kuehniella a-amy-lase isoenzymes 1–3 (EkAmy1–3)] with an alkaline pH optimum were puri-fied from larvae and biochemically characterized. These isoenzymes differ significantly in their sensitivity to a-amylase inhibitors of plant origin that are directed against herbivores as antifeedants. Such functional variability renders the amylolytic system less vulnerable to suppression by plant defen-sive molecules. Moreover, we found that expression of a-amylases is upreg-ulated in larvae feeding on a diet enriched with an a-amylase inhibitor. The a-amylases are secreted into the larval midgut by an exocytotic mecha-nism, as revealed by immunogold microscopy. The cDNA sequence of EkAmy3 was determined, and a homology model of EkAmy3 was built in order to analyze the structural features responsible for adaptation to alka-line pH. First, the overall fold was found to be stabilized by remodeling of ion pairs. Second, molecular simulations supported by activity measure-ments showed that EkAmy3 does not bind a Cl–, owing to an Arg-to-Gln mutation in a conserved binding site. The Cl–-binding residues are in con-tact with the catalytic residues, and this change might help to fine-tune the catalytic pKa values to an alkaline pH optimum. We conclude that lepidop-teran a-amylases are evolutionarily adapted in terms of structure and expression dynamics for effective functioning in the digestive system. a-Amylases (a-1,4-glucan-4-glucanohydrolases) are a only in the 1990s, as they became targets for regulation group of glycoside hydrolases that are widely distrib-uted in bacteria, fungi, plants, and animal tissues [1,2]. They catalyze the hydrolysis of the a-(1,4) glycosidic of important physiological processes. A promising field of current research involves suppression of the devel-opment of insect pests via impairment of their amylo- linkage found in starch components and other related lytic digestion by naturally occurring a-amylase polysaccharides. a-Amylases are among the oldest inhibitors. The proteinaceous a-amylase inhibitors are known enzymes, but detailed information about their produced in plant tissues, in which they act as defen-structure and inhibition started to become available sive proteins directed against exogenous digestive Abbreviations EkAmy1–3, Ephestia kuehniella a-amylase isoenzymes 1–3; HPA, human pancreatic a-amylase; PPA, porcine pancreatic a-amylase; qPCR, quantitative real-time RT-PCR; TMA, Tenebrio molitor a-amylase. FEBS Journal 276 (2009) 3531–3546 ª 2009 The Authors Journal compilation ª 2009 FEBS 3531 Alkaline digestive a-amylases of Lepidoptera a-amylases of insect herbivores [3]. These inhibitors are effective against stored-product pests that rely on J. Pytelkova et al. Results starch as a major food source. Transgenic crops pro-ducing high levels of a-amylase inhibitors have been successfully tested against bruchid beetles (Coleoptera) in field conditions [4]. The digestive a-amylases of coleopteran pests have been characterized in detail at the biochemical and structural levels [5,6]. On the other hand, much less is known about the digestive amylolytic system of the Lepidoptera, a second major group of granivorous insect pests. In this work, we analyze the digestive a-amylases of the Mediterranean flour moth, Ephestia (Anagasta) kuehniella Zeller, which we chose as a model lepidopteran species. This serious pest is found worldwide; its larvae attack wheat flour and cereal commodities and damage large quantities of food in granaries, flour mills, and households. The digestive tract of lepidopterans is unique in its extremely alkaline pH. The pH values measured in par-ticular compartments of the larval digestive tract span a range between 9.8 and 11.2 [7,8]. In contrast, the envi-ronment of the digestive tract in other insect orders is generally mildly acidic, and often contains a gradient from acidic to neutral pH along the alimentary canal [8]. It can be demonstrated that the pH optima of insect digestive a-amylases are close to the pH of the milieu in which they perform their function [8]. Accordingly, the amylolytic activity analyzed in various lepidopteran species was found to be optimal in the pH range of 9.0– 9.6 [8,9]. This raises the following questions: what is the molecular basis of alkaline adaptation of lepidopteran a-amylases, and what is its evolutionary context? Here, we have attempted to address these issues with the help of a molecular model built for an E. kuehniella a-amy-lase and molecular phylogenetic analysis. Molecular mechanisms of the adaptation of insects to plant defensive a-amylase inhibitors are still poorly understood, in contrast to those of plant defensive pro-tease inhibitors, which have been intensively studied during the last decade. Our aim in the present work was to fill this gap in our knowledge about the coun-ter-defense capability of the insect amylolytic digestive system. Towards this end, we analyzed the inhibitor The digestive tract of E. kuehniella contains alkaline a-amylases a-Amylase activity was detected in the whole body homogenate from E. kuehniella larvae using Remazol Brilliant Blue dyed starch (RBB-starch) as a substrate. The pH profile in Fig. 1 shows that activity was observed in the alkaline range, with a maximum between pH 9 and pH 10. The alkaline a-amylase activity was specifically localized in the larval digestive tract, and was secreted into the gut lumen. This was demonstrated by detection of the activity solely in the gut homogenate and larval excrement (specific activity 20.5 ± 1.2 and 23.4 ± 0.9 UÆmg)1 protein, respec-tively) and not in the homogenate from the degutted body. Isolation of a-amylase isoenzymes The digestive a-amylases were purified from E. kueh-niella larvae using a two-step procedure. In the first step, the crude fraction of a-amylases was obtained by affinity precipitation with glycogen (see Experimental procedures). Subsequently, this material was separated by ion exchange chromatography on a Mono Q col-umn, which yielded three peaks with alkaline a-amy-lase activity (Fig. 2). These peaks correspond to a-amylase isoenzymes of 56 kDa, denoted E. kuehni-ella a-amylase isoenzymes 1–3 (EkAmy1–3). N-termi-nal sequencing revealed that the purified a-amylases have distinct but homologous sequences, and demon-strated that three a-amylase genes are expressed in the sensitivity of the a-amylase isoenzyme arsenal in E. kuehniella, as well as the expression dynamics of a-amylase genes in response to an ingested a-amylase inhibitor. This leads us to the surprising conclusion that the amylolytic and proteolytic digestive systems of insects share common strategies to overcome plant defense, which increases the overall complexity of molecular interactions between insect herbivores and plants. pH Fig. 1. pH profile of a-amylase activity in E. kuehniella larvae. Whole body homogenate was assayed with RBB-starch as a substrate at various pH values. Mean values ± standard errors are normalized to the maximum value. 3532 FEBS Journal 276 (2009) 3531–3546 ª 2009 The Authors Journal compilation ª 2009 FEBS J. Pytelkova et al. Alkaline digestive a-amylases of Lepidoptera Table 2. Interaction of E. kuehniella a-amylases with a-amylase inhibitors, including the saccharide inhibitor acarbose and proteina-ceous inhibitors from wheat (WI-1 and WI-3) and bean (aAI-1). The inhibition was determined with purified isoenzymes at their pH optima and RBB-starch as a substrate. Mean values ± standard errors are given. NI, no inhibition found for 10 lM inhibitor. Inhibitor IC50 (lM) Isoenzyme EkAmy1 EkAmy2 EkAmy3 acarbose 13.62 ± 0.27 5.60 ± 0.22 2.21 ± 0.13 WI-1 NI 5.79 ± 0.23 0.0080 ± 0.0008 WI-3 aAI-1 NI NI NI NI 1.03 ± 0.04 NI lar weight inhibitors of a-amylases (Table 2). The Fig. 2. FPLC separation of alkaline a-amylases EkAmy1–3 from E. kuehniella larvae. The Mono Q column was eluted with an NaCl gradient (dotted line). The fractions with alkaline a-amylase activity (bars) were assayed with RBB-starch as a substrate and analyzed by SDS⁄PAGE with silver staining of proteins (inset: the arrow marks the 56 kDa band corresponding to the a-amylases; molecular mass standards are indicated). Table 1. Biochemical characterization of alkaline a-amylases puri-fied from E. kuehniella larvae. The N-terminal sequence was deter-mined by Edman sequencing. Enzymatic activity was assayed with RBB-starch as a substrate at the respective pH optima. Mean values ± standard errors are given. Specific wheat a-amylase inhibitors are ecologically relevant for E. kuehniella, and were previously reported to be effec-tive against lepidopteran a-amylases [9,10]. EkAmy3 was most sensitive to inhibition by two types of inhibi-tors from wheat, WI-1 (tetrameric form) and WI-3 (monomeric form), with IC50 values in the nanomolar and micromolar ranges, respectively. EkAmy2 was inhibited only by WI-1, and EkAmy1 was classified as an ‘insensitive’ isoenzyme. The aAI-1 inhibitor from bean was inefficient against all isoenzymes; it is a potent inhibitor of a-amylases from the insect orders Coleoptera, Hymenoptera, and Diptera, in which a mildly acidic pH is required for its action [11]. Acar-bose, an oligosaccharide inhibitor from Streptomyces, Isoenzyme EkAmy1 EkAmy2 EkAmy3 N-terminus YYNPHYAAGRSTMVH YFNPHYAAGRSTMVM YKNPYYASGRSVNVH pH optimum 9.5 9.5 8.0 activity (UÆmg)1) 275 ± 21 532 ± 26 115 ± 12 was able to interact with all of the isoenzymes (IC50 values in the micromolar range). This low selectivity can be explained by the substrate-mimicking structure of acarbose, which allows it to bind to a-amylases and also to a-glucosidases [12]. larval midgut (Table 1). The N-terminal sequences of EkAmy1 and EkAmy2 are more related to each other (86% identity) than to EkAmy3 (66% and 60% iden-tity, respectively). The relative protein yield of isola-tion of EkAmy1–3 was about 1.5 : 2 : 3. Functional characterization of a-amylase isoenzymes The purified a-amylase isoenzymes EkAmy1–3 were characterized with regard to their interaction with sub-strate and inhibitors. The specific activity and individ-ual pH optima measured with RBB-starch as a model Sequence analysis of EkAmy3 The N-terminal sequences of EkAmy isoenzymes show significant homology with the sequences of lepidop-teran a-amylases (represented by Diatraea saccharalis a-amylase in Fig. 3). The conserved regions in lepidop-teran a-amylase sequences were used to design PCR primers for the determination of an internal part of the cDNA sequence encoding an E. kuehniella a-amy-lase (Fig. 4). These primers were employed for screen-ing of a cDNA library prepared from total RNA isolated from the gut tissue of E. kuehniella larvae. The full-length cDNA sequence was then completed by the determination of the 3¢-ends and 5¢-ends of the substrate varied modestly among isoenzymes (Table 1). gene (see Experimental procedures). The sequence In contrast, we found a striking difference in the inhib-itor specificity of individual isoenzymes, which was probed using a set of macromolecular and low molecu- contains an ORF of 1503 nucleotides that encodes a protein of 501 amino acids (deposited in GenBank under accession number FJ489868). The nucleotide FEBS Journal 276 (2009) 3531–3546 ª 2009 The Authors Journal compilation ª 2009 FEBS 3533 Alkaline digestive a-amylases of Lepidoptera J. Pytelkova et al. Fig. 3. Multiple sequence alignment of E. kuehniella a-amylase EkAmy3 with representative homologous a-amylases from Lepidoptera (D. saccharalis, AAP97393), Coleoptera (T. molitor, P56634), and mammals (Homo sapiens pancreatic, AAH07060). The sequence similarities to EkAmy3 are 82%, 61%, and 51%, respectively. Full-length sequences of mature proteins are aligned, and partial N-terminal protein sequences of EkAmy1 and EkAmy2 are included. Amino acids identical with those of EkAmy3 are shaded. Positions of catalytic residues (@) and residues binding the Cl) (#) and the Ca2+ (*) are indicated. Note the mutation Arg331Gln located in the Cl)-binding site of lepidopteran a-amylases. sequence corresponds to EkAmy3, the most abundant The coding sequence of EkAmy3 contains the isoenzyme, based on a match to the N-terminal sequence for a predicted signal peptide followed by one sequence of purified EkAmy3 (Fig. 4). 3534 for the mature protein with a theoretical mass of FEBS Journal 276 (2009) 3531–3546 ª 2009 The Authors Journal compilation ª 2009 FEBS J. Pytelkova et al. Alkaline digestive a-amylases of Lepidoptera Fig. 4. Nucleotide and deduced amino acid sequence of E. kuehniella a-amylase EkAmy3. The signal peptide (italics), N-terminal Edman sequencing of purified EkAmy3 (underlined), potential N-glycosylation site (double under-lined) and catalytic residues Asp193, Glu230 and Asp295 (boxed) are indicated. The amino acid numbering is according to the mature protein (starting residue is marked with a triangle). The PCR primers used are indicated by lines above them (for coding, see Experi-mental procedures). FEBS Journal 276 (2009) 3531–3546 ª 2009 The Authors Journal compilation ª 2009 FEBS 3535 ... - tailieumienphi.vn