HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ochiai, M. Articles by Ashida, M. Search for Related Content PubMed PubMed Citation Articles by Ochiai, M. Articles by Ashida, M. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 7, 4995-5002, February 18, 2000

A Pattern-recognition Protein for -1,3-Glucan
THE BINDING DOMAIN AND THE cDNA CLONING OF -1,3-GLUCAN RECOGNITION PROTEIN FROM THE SILKWORM, BOMBYX MORI^*

Masanori Ochiai and Masaaki Ashida

From the Institute of Low Temperature Science, Hokkaido University, Sapporo 060-0819, Japan

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The -1,3-glucan recognition protein (GRP) has strong specific affinity for -1,3-glucan, a component of the fungal cell wall. Its interaction with -1,3-glucan initiates the activation of the prophenoloxidase cascade, which is an important defense system in invertebrates of many species. We cloned the cDNA of the GRP of the silkworm Bombyx mori. The GRP mRNA transcript was constitutively expressed in the hemocytes, fat body, and epithelial cells of the naive silkworm. At the same time, a bacterial or yeast challenge was indicated to intensify the transcription. Comparison of the deduced amino acid sequence with known sequences revealed that the GRP contained a region (Thr²⁶⁴ to Pro³⁸⁶) displaying significant similarity to the catalytic regions of bacterial -1,3-glucanases and much higher similarity to the glucanase-like regions of Gram-negative bacteria-binding proteins found in the silkworm B. mori and the mosquito Anopheles gambiae. The region (Thr²⁶⁴ to Pro³⁸⁶) of the GRP, however, was demonstrated not to have appreciable affinity for -1,3-glucan. A recombinant peptide corresponding to an N-terminal region (Tyr¹ to Ala¹⁰²) of the GRP bound strongly to -1,3-glucan. These results indicate that the binding domain of the GRP for -1,3-glucan is located in the N-terminal region. Glucanases and the current pattern-recognition proteins that contain a glucanase-like region seem to have a common origin in their molecular evolution.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The innate immune system is recognized to be important for self-defense against invading micro-organisms, especially in invertebrates, because they lack the adaptive immunity employing clonally selected molecules for the recognition of non-selves. In mammals, the importance of the innate immune system is now appreciated even more than before because recent findings have revealed that it is essential to make the adaptive immune responses active (1). Over the past several years, proteins belonging to the Rel family have been demonstrated to work in intracellular signaling for the activation of acute-phase protein genes in both mammals and insects (2). In the extracellular signaling of the recognition of microbes as foreign, insects and mammals have also been shown to employ homologues. Toll, a receptor on the fat body (an organ equivalent to the vertebrate liver) of Drosophila (3), and Toll-like receptor 2 on peripheral blood leukocytes of humans (4) have been indicated to work in the transduction of extracellular signals. The peptidoglycan recognition protein (PGRP)¹ is another example where homologues have been demonstrated or suggested to take part in the extracellular recognition of foreignness in the innate immunity of both mammals and insects (5, 6). These common employments of homologues in the intracellular and extracellular signaling mechanisms indicate close links between innate immunity of mammals and insects.

A variety of bacterial and fungal cell wall components including lipopolysaccharide, peptidoglycan, and -1,3-glucan are collectively referred to as pathogen-associated molecular patterns (PAMPs) (7). They are biologically active and elicit various innate immune reactions such as septic shock, cytokine synthesis, and acute-phase protein synthesis in both insects and mammals. In the innate immune mechanism, PAMPs are recognized by particular pattern-recognition proteins, which could be present on effector cell surfaces as receptors or as free-floating molecules in the plasma fraction. In insects, any receptor for PAMP has not been identified yet, but several of the free-floating pattern-recognition proteins have been identified. They are the -1,3-glucan recognition protein (GRP) (8), PGRP (9), lipopolysaccharide-binding protein (10, 11), Gram-negative bacteria-binding protein (GNBP) (12), hemolin (13), and lectins with varied specificities for saccharides (14-17). Physiological functions of some pattern-recognition proteins are known. However, none of the mechanisms used for binding to particular patterns have been elucidated.

The molecular mechanisms of the recognition of -1,3-glucan as foreign have not been studied in detail at present. Although the receptors and recognition proteins have been indicated to be present in mammals (18), they have not been isolated, and their molecular identities remain to be studied. In invertebrates, no receptor for -1,3-glucan has been demonstrated either, but proteins with affinity for -1,3-glucan have been isolated from the silkworm (8), crayfish (19), earthworm (20), and horseshoe crab (21), and some of their properties have been reported. All of these proteins have been reported to participate in triggering the proteolytic cascade for melanin formation or blood coagulation. However, their properties at the molecular level are quite different, and the binding domains for -1,3-glucan have not been identified.

Insect GRP was originally identified as a component of the prophenoloxidase cascade of the silkworm Bombyx mori (8). The GRP has been shown to have a strong specific affinity for -1,3-glucan. Binding of the GRP with -1,3-glucan initiates the activation of the prophenoloxidase cascade. The binding of PGRP with peptidoglycan has also been shown to initiate the activation of the cascade (9). Besides PGRP and GRP, the cascade is composed of serine protease zymogens, prophenoloxidase, and components yet to be discovered (22). The cascade has often been thought to be an effector mechanism only for the production of melanin around foreign objects such as bacteria, fungi, and non-habitual endoparasites. However, it is now speculated to be a signaling mechanism by which recognition signals for fungi (-1,3-glucan) and bacteria (peptidoglycan and lipopolysaccharide) are relayed to effector cells.

Realizing the paucity of our understanding of the recognition mechanisms of -1,3-glucan as foreign in innate immunity of invertebrates and mammals, we conducted experiments on the silkworm GRP with the following aims: to establish the identity of the silkworm GRP and to identify the binding domain of GRP for -1,3-glucan.

We report here the cloning of GRP cDNA from the cDNA library of the hemocytes of the silkworm B. mori. The sequence analysis showed that GRP is a protein containing a glucanase-like region in the central part and a glycosylphosphatidylinositol anchor attachment site in the hydrophobic C-terminal portion. The glucanase-like region appears not to have affinity for -1,3-glucan, but the N-terminal fragment with 102 amino acids was shown to bind strongly to the glucan.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Insects and Microorganisms-- Silkworms, B. mori (strain Kinshu × Showa), were reared on an artificial diet as described previously (23). In the experiments for immune challenge of the insect, Enterobacter cloacae (JCM1232) and Candida albicans (JCM2078) were used. C. albicans was grown in a Potato-Dextrose broth medium (Difco) at 25 °C. At the late logarithmic phase of the growth, the cells were collected and suspended in 10 mM bis-tris propane buffer, pH 6.5, containing 150 mM NaCl. The larvae on day 5 of the fifth instar were injected with 10 µl of the suspension at A₆₀₀ = 0.1. The bacterial challenge was carried out as described previously (6).

Preparation of 20- and 43-kDa Fragments Derived from GRP-- The GRP was highly purified from the hemolymph of silkworm larvae on day 5 of the fifth instar as described previously (8). 400 µg of the protein was incubated with 1.0 µg of -chymotrypsin (Sigma, TLCK-treated type VII from bovine pancreas) in 1.4 ml of 0.1 M Tris-HCl buffer, pH 8.5, at 37 °C for 30 min. This treatment caused a cleavage of GRP into two fragments, which migrated to positions of 20- and 43-kDa proteins in SDS-polyacrylamide gel electrophoresis (SDS-PAGE). After addition of 40 µl of 0.1 M phenylmethanesulfonyl fluoride in ethanol, the reaction mixture was diluted 6 times with 20 mM bis-tris propane buffer, pH 6.5, and immediately applied to a Mono Q column HR5/5 (Amersham Pharmacia Biotech). The adsorbed GRP-derived 20- and 43-kDa fragments were separated at a flow rate of 1 ml/min with a liner gradient of NaCl from 50 to 250 mM in a total volume of 20 ml of 10 mM bis-tris propane buffer, pH 6.5. The 20- and 43-kDa fragments were eluted at 200 and 120 mM NaCl, respectively. Both the fragment solutions were individually dialyzed against 10 mM Tris maleate buffer, pH 6.5, containing 150 mM NaCl. The dialyzed solutions were centrifuged at 16,000 × g for 10 min, and the supernatants were used as the 20- and 43-kDa fragment solutions.

Protein Analysis-- SDS-PAGE was performed according to Laemmli's method (24) in 12.5% separating gel under reducing conditions. For the determination of the relative molecular mass in the SDS-PAGE, marker proteins (Bio-Rad) were run with samples. Protein concentrations were determined by the method of Lowry et al. (25) with bovine serum albumin as a standard.

Preparation of Curdlan (Insoluble -1,3-Glucan) Beads and Curdlan Binding Assay-- Curdlan has a nature to form gel by neutralization of its alkaline solution (26). The curdlan beads were prepared as follows: 60 ml of 3% (w/v) curdlan in 0.1 N NaOH solution was dropped at a flow rate of 60 ml/h into 240 ml of 1-butyl alcohol containing 0.5% (w/v) Tween 20 with vigorous stirring to disperse the curdlan solution. The dispersed solution was dropped to a mixture of 1-butyl alcohol (260 ml) and acetic acid (20 ml) with stirring to neutralization. After gelation of the curdlan, the beads were washed 5 times by decantation in 2 liters of distilled deionized water. About 15 ml of the aqueous slurry of beads was obtained, and the shape of the beads was spherical in diameters of 100-300 µm.

The curdlan binding assay was carried out by essentially the same method as used before (8). Briefly, 50 µl of the curdlan beads suspension was incubated with a sample solution (50-200 µl) for 10 min at room temperature. After the supernatant was removed by centrifugation at 800 × g for 1 min, the beads were suspended in 0.5 ml of 10 mM bis-tris propane buffer, pH 6.5, containing 150 mM NaCl and centrifuged. The suspension and centrifugation was repeated three times. Protein adsorbed to the beads was eluted with 20 µl of SDS-PAGE solubilizing buffer (82 mM Tris-HCl buffer, pH 8.8, containing 1% SDS, 1% -mercaptoethanol, 30% glycerol, and 0.01% bromphenol blue) at boiling temperature for 5 min. The beads were removed by centrifugation, and the supernatant was subjected to SDS-PAGE under reducing conditions. Proteins were stained with Coomassie Brilliant Blue and quantified densitometrically using the Analytical Imaging Station software (Imaging Research Inc., Canada). It was confirmed that at least 280 µg of the purified GRP could bind the beads, and the amount of the bound protein was quantitatively determined under the experimental conditions.

Amino Acid Sequence Analysis and Mass Number Analysis-- For the determination of N-terminal amino acid sequences of the GRP, 20- and 43-kDa fragments, the preparations were desalted on a Wakopak C₄ column (4.6 × 150 mm, Wako Pure Chemical Industries) in high performance liquid chromatography. The isolated samples were analyzed by automated Edman degradation using a protein sequencer PPSQ-10 (Shimadzu Corp.). For obtaining the internal sequences of the 20- and 43-kDa fragments, the S-pyridylethylated fragments were digested separately with lysylendopeptidase (Wako Pure Chemical Industries) at a molar ratio of enzyme to substrate of 1:100. Each digest was fractionated by reversed-phase high performance liquid chromatography on a column of STR ODS-II PEEK (4.6 × 150 mm, Shimadzu Corp.). The peptides eluted as separated peaks were subjected to sequence analysis.

Mass number analyses were performed using a Kompact matrix-assisted laser desorption and ionization (MALDI) mass spectrometer (model IV, Shimadzu Corp.) in linear time of flight mode. A saturated sinapinic acid in 0.1% (v/v) trifluoroacetic acid was used as a matrix. The MALDI mass spectrometer was calibrated for the mass/H⁺ value using horse heart myoglobin (M_r 16,950.9) and bovine serum albumin (M_r 66,525) as standards before use.

Effect of 20- and 43-kDa Fragments on the Activation of the Prophenoloxidase Cascade by -1,3-Glucan-- The plasma fraction of hemolymph was prepared according to the method of Ashida (27). Mixtures of 5 µl of 100 µg/ml zymosan (a mixture of -1,3-glucan and mannan) and 10 µl of 20- or 43-kDa fragment solution at various concentrations were preincubated at 25 °C for 30 min. Each of the mixtures was added to a mixture of 83 µl of plasma and 2 µl of 250 mM CaCl₂, followed by incubation at 25 °C. At intervals, phenoloxidase activity of the mixtures at 25 °C was assayed spectrophotometrically as was described previously (23). In a control experiment, 100 µg/ml peptidoglycan from Micrococcus luteus was used instead of zymosan.

cDNA Cloning and Sequencing of GRP-- Poly(A)⁺ RNA was extracted from the hemocytes of silkworm larvae on day 5 of the fifth instar using a Quick Prep Micro mRNA purification kit (Amersham Pharmacia Biotech). A gt10 cDNA library was constructed from the poly(A)⁺ RNA using a cDNA synthesis system and rapid cloning module (Amersham Pharmacia Biotech). The library was used as a template in the polymerase chain reaction (PCR). A pair of degenerate oligonucleotide primers was synthesized on the basis of the internal sequences, NEEMEG and WFPTWD, respectively, of 20- and 43-kDa fragments: 5'-AAGAATTCAA(C/T)GA(A/G)GA(A/G)ATGGA(A/G)GG-3' and 5'- AAGAATTCTC(A/G)TCCCA(A/C/G/T)GT(A/C/G/T)GG(A/G)AACCA-3'. The PCR was repeated for 40 cycles with denaturation at 94 °C for 1 min, annealing at 60 °C for 2 min, and extension at 72 °C for 3 min. The product was digested with EcoRI, purified with agarose gel electrophoresis, and labeled with [-³²P]dCTP using a BcaBEST labeling kit (Takara Shuzo Co.) to prepare a probe for GRP cDNA. In screening the gt10 hemocyte cDNA library, hybridization was carried out at 65 °C for 16 h in 6× SSC (0.9 M NaCl, 90 mM sodium citrate, pH 7.2), 5× Denhardt's solution (0.1% polyvinylpyrrolidone, 0.1% bovine serum albumin and 0.1% Ficoll 400), 0.5% (w/v) SDS, and 100 µg/ml denatured salmon sperm DNA. The membrane was washed twice in 2× SSC containing 0.5% (w/v) SDS at 65 °C for 15 min and subjected to autoradiography. The insert DNA of the positive clones was subcloned into a BamHI site in the pBluescript II SK vector (Stratagene) and sequenced by a DNA sequencer (model 377, PE Applied Biosystems). The longest clone was used as the GRP cDNA.

Determination of -1,3-Glucanase Activity of GRP-- A mixture (pH 6.5) of 50 µl of 2 mg/ml laminarin, 40 µl of 1 mg/ml GRP, and 10 µl of 50 mM CaCl₂ was incubated at 25 °C for 24 h. After the incubation, 5 µl of the mixture was analyzed by thin layer chromatography on a silica gel TLC aluminum sheet (Merck) using a mixture of n-propyl alcohol/acetic acid/H₂O (3:3:2, v/v) as a developing solvent. The detection spray solution specific for reducing sugars was 10 mg/ml orcinol in 50% H₂SO₄. In a control experiment, Zymolyase-100T (Seikagaku Kogyo Co. LTD), -1,3-glucan laminaripentaohydrolase from Arthrobacter luteus, was used as an enzyme with -1,3-glucanase activity.

Expression of the N-terminal Region of GRP in Escherichia coli-- To prepare various deletions in the N-terminal region of GRP, cDNAs encoding the regions were amplified by PCR using the GRP cDNA as a template. To add EcoRI and SalI sites to the 5' and 3' ends, respectively, of the PCR products, the primers were designed as follows: 5'-CGGAATTCTACGAGGCACCACCG-3' (G1-F); 5'-CGGAATTCGCAATACACCCTAAAGG-3' (G10-F); 5'-CGGAATTCGTTTCTGTTCCTGA-3' (G18-F); 5'-CGGAATTCAACGAGGAAATGGAAGG-3' (G35-F); 5'-CGGAATTCGGAGATAAGATTTACT-3' (G70-F); 5'-ACGGTCGACTCAGAGCTTACCGTGAAACG-3' (G34-R); 5'-ACGGTCGACTCAGATTTTCAGCGCTGCAT-3' (G69-R); 5'-ACGGTCGACTCAGTATCCTAAGCCGTCCT-3' (G86-R); 5'-ACGGTCGACTCATGTCCACTCCCCGTTAT-3' (G94-R); 5'-ACGGTCGACTCATTCAACTGTCCACTCCCCGTTAT-3' (G96-R); 5'-ACGGTCGACTCATACGAAACCTTCAACTG-3' (G99-R); ACGGTCGACTCAGGCTTCATCTACGAAAC-3' (G102-R); and ACGGTCGACTCATTCTACTCCTGGTGTTA-3' (G119-R). The PCR products were cleaved with EcoRI and SalI and cloned into the pET32a(+) vector (Novagen) which contains a sequence encoding the thioredoxin (Trx), a His tag, and an enterokinase cleavage site. After transformation of E. coli (strain, AD494(DE3) pLysS) by the vectors, the cultures were grown at 37 °C. When A₆₀₀ of the cultures reached 0.4, the expressions of the proteins coded by the vectors were induced by the addition of isopropyl--D-thiogalactopyranoside to the final concentration of 1.0 mM. The cells were incubated at 30 °C with shaking for 3 h and collected by centrifugation. The cell pellets were suspended in 1/10 culture volume of 10 mM Tris-HCl buffer, pH 7.5, containing 150 mM NaCl, lysed by freezing and thawing, and centrifuged at 16,000 × g for 10 min. The supernatants were recovered for the curdlan binding assay and protein purification. The fusion proteins of Trx with the N-terminal region of GRP were purified with a His-Bind resin according to the manufacturer's recommendations (Novagen). When the peptide containing the Trx and spacer was removed from the fusion proteins, the proteins were treated with enterokinase (Novagen) and purified by Mono Q column chromatography.

Northern Blot Analysis-- Poly(A)⁺ RNA preparations from hemocytes, fat body, epidermal cells, midgut, silk gland, and Malpighian tubules of naive silkworms and total RNA from the fat body of the bacteria- or yeast-challenged silkworms were prepared as described previously (6). 2 µg of poly(A)⁺ RNA preparations were run on 1% agarose gels, transferred to nylon membranes (Hybond-N+, Amersham Pharmacia Biotech), and hybridized with a ³²P-labeled GRP cDNA probe at 65 °C for 16 h in the hybridization solution as described above. The membranes were washed twice with 2× SSC containing 0.5% SDS at 65 °C and subjected to autoradiography. In the bacteria- or yeast-challenged experiments, 30-µg aliquots of the total RNA preparations were subjected to Northern blot analyses using cDNA probes of GRP, cecropin B (0.4 kbp), and -tubulin (1.3 kbp) as described previously (6).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

20- and 43-kDa Fragments Derived from GRP-- The silkworm GRP is sensitive to -chymotrypsin. Digestion by the protease gave two major fragments. The mobilities of the fragments corresponded to the positions of 20- and 43-kDa polypeptides in SDS-PAGE under reducing conditions (Fig. 1A). These fragments were purified on Mono Q column chromatography and subjected to MALDI spectrometric analysis. The analysis revealed that the mass numbers of GRP, the smaller, and larger polypeptides were 54,594, 13,338, and 41,313 Da, respectively (data not shown). The N-terminal sequences of these fragments and GRP were analyzed as follows: GRP, YEAPPATLEA; 20-kDa fragment, YEAPPATLEA; and 43-kDa fragment, TSTSLNPESP. These results indicate that -chymotrypsin digested GRP at a specific site under the experimental conditions and that the 20-kDa fragment originated from the N-terminal of GRP.

View larger version (84K): [in this window] [in a new window]   Fig. 1.   SDS-PAGE (A) and curdlan binding assay (B) of the 20- and 43-kDa fragments derived from the silkworm GRP. Procedures for the isolation of fragments and the binding assay are described under "Experimental Procedures." Proteins were separated in 12.5% polyacrylamide gel and stained with Coomassie Brilliant Blue. Samples applied to each lane are as follows (the amounts of protein subjected to SDS-PAGE (A) and the curdlan binding assay (B) are indicated in parentheses): lane 1, purified GRP (A, 3 µg; B, 6 µg); lane 2, digestion of the GRP by -chymotrypsin (A, 3 µg; B, 6 µg); lane 3, purified 43-kDa fragment (A, 2 µg; B, 4 µg); lane 4, purified 20-kDa fragment (A, 1 µg; B, 2 µg); lane M, marker proteins.

For preliminary experiments to clarify the binding domain of GRP to -1,3-glucan, the digestions of GRP with -chymotrypsin were subjected to the curdlan (insoluble -1,3-glucan) binding assay. The 20-kDa fragment bound strongly to curdlan, but binding of the 43-kDa fragment to curdlan could not be detected (Fig. 1B). The purified 20-kDa fragment also bound to curdlan. Any appreciable difference in the affinity for curdlan was not detectable between the 20-kDa fragment and the purified GRP. These results indicate that the 20-kDa fragment originated from the N terminus of the GRP and has the binding domain for -1,3-glucan.

Effects of 20- or 43-kDa fragments on the activation of the prophenoloxidase cascade by -1,3-glucan were investigated (Fig. 2). When -1,3-glucan preincubated with the purified 20-kDa fragment was added to the plasma fraction of the silkworm, the -1,3-glucan could not activate the cascade until at least 100 min. On the other hand, peptidoglycan, another elicitor for the cascade, triggered the activation of the cascade even if it had been preincubated with the 20-kDa fragment. -1,3-Glucan and peptidoglycan preincubated with the 43-kDa fragment retained their ability as elicitors for the cascade. These results suggest that the 20-kDa fragment bound to -1,3-glucan prevented GRP in the plasma from interacting with the glucan, resulting in the deprivation of elicitor activity of the preincubated -1,3-glucan.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Activation of the prophenoloxidase cascade by zymosan or peptidoglycan preincubated with the 20-kDa fragment (A) and the 43-kDa fragment (B). 5 µl of 100 µg/ml of zymosan (), 100 µg/ml of peptidoglycan (), or H2O () as a control was preincubated with 10 µl of 75 µg/ml 20-kDa fragment or 150 µg/ml 43-kDa fragment at 25 °C for 10 min. Each of the preincubated solutions was added to a mixture of 2 µl of 250 mM CaCl2 and 83 µl of silkworm plasma. The reaction mixtures were incubated at 25 °C, and at intervals a 5-µl aliquot was assayed for phenoloxidase activity to monitor the activation of the prophenoloxidase cascade. 20- and 43-kDa fragments, respectively, were used for the preincubation in A and B.

Primary Structure of GRP-- A pair of oligonucleotide primers corresponding to the amino acid sequences of peptides derived from each of the 20- and 43-kDa fragments was used in the PCR. The reaction yielded a 1.3-kbp cDNA fragment that gave 1.0- and 0.3-kbp fragments in the digestion with EcoRI. By using the 1.0-kbp cDNA fragment as a probe for screening the cDNA library, 22 positive plaques were obtained. Among the cDNA clones in the plaques, the longest three clones were sequenced. The complete sequence of GRP cDNA was determined and is shown in Fig. 3 with the deduced amino acid sequence. This cDNA is composed of 1769 bp. An open reading frame of 1575 bp, nucleotides 123-1608, encoded a polypeptide consisting of 495 amino acid residues. The first N-terminal amino acid of the purified GRP corresponded to the 17th amino acid from the beginning of the deduced sequence. Thus, the mature protein consists of 479 residues, which give rise to a polypeptide with a calculated molecular mass of 53,859.7 Da. The amino acid sequence contains a putative N-glycosylation site at the 362nd residue in the mature protein and a potential sequence of glycosylphosphatidylinositol anchor attachment site (Ala⁴⁵⁵- Arg ⁴⁵⁶-Ser⁴⁵⁷) (28) in the C-terminal hydrophobic region. The predicted amino acid sequence established here agrees well with the amino acid composition of the purified GRP reported in a previous report (8). The cleavage site by -chymotrypsin was identified to be Phe¹¹⁹-Thr¹²⁰.

View larger version (70K): [in this window] [in a new window]   Fig. 3.   Nucleotide sequence of cDNA encoding GRP from B. mori and the deduced amino acid sequence. The nucleotide sequence is numbered at left (upper) and the amino acid sequence at right (lower), beginning at the N terminus of the mature protein. The underlined amino acid residues were confirmed by sequencing of the peptides isolated after lysylendopeptidase treatment of the S-pyridylethylated GRP. The cleavage site of the mature protein by -chymotrypsin and the putative N-linked glycosylation site are indicated by a closed and an opened triangle, respectively. The potential glycosylphosphatidylinositol anchor sequence is boxed.

In the data base search, GRP was revealed to show the highest sequence similarity to the GNBP of the silkworm B. mori (12). Both of the proteins have a glucanase-like domain of which the sequence is very similar to the catalytic domain of bacterial -1,3-glucanases. Proteins with the glucanase-like domain were extracted from a data base. The sequences of the glucanase-like domains are aligned in Fig. 4 together with the putative catalytic domains of -1,3-glucanases of bacteria (29-31) and the sea urchin Strongylocentrotus purpuratus (32). Fig. 4 also includes the glucanase-like sequences from Drosophila melanogaster found in the data base of the expressed sequence tag. All glucanase-like domains in Fig. 4 have been located in the central portion of the polypeptides except the one found in subunit  of factor G from the horseshoe crab Tachypleus tridentatus (33). The glucanase-like domain of the subunit  has been reported to be in the N-terminal portion. The silkworm GRP did not exhibit any appreciable glucanase activity under the conditions described under "Experimental Procedures."

View larger version (66K): [in this window] [in a new window]   Fig. 4.   Sequence alignment of the -1,3-glucanase-like region of the silkworm GRP with the homologues. The sequence of the silkworm B. mori GRP is from the present study. Other sequences were found in the GenBankTM data base (release 110.0 1999) by BLASTA. The aligned sequences are from proteins as follows (abbreviations used in this figure and accession numbers are indicated in parentheses): silkworm B. mori GNBP (L38591) (12); mosquito Anopheles gambiae GNBP (AJ001042) (42); fruit fly D. melanogaster expression sequence tag (EST, AI109637); earthworm E. foetida CCF-1 (AF030028) (20); horseshoe crab T. tridentatus factor G subunit  (FG-, D16622) (33); sea urchin S. purpuratus -1,3-glucanase (S.p. -Gase, U49711) (32); B. circulans -1,3-glucanase A1 (B.c. -Gase, M34503) (29); Clostridium thermocellum -1,3(4)-glucanase (C.t. -Gase, X89732) (30); and Thermotoga neapolitana -glucosidase (T.n. -Gase, Z47974) (31). The alignments were done using a CLUSTAL V program. Gaps indicated by hyphens are introduced to optimized sequence alignment. Numbers on the left and right indicate the residue numbers of the amino acid sequence of each protein. Amino acid residues identical to silkworm GRP are indicated by shaded boxes. The catalytically active residues of B. licheniformis -1,3-1,4-glucanase (40) are indicated by closed triangles.

Expression of GRP mRNA-- Northern blot analysis of poly(A)⁺ RNA preparations from different tissues was performed by using a GRP cDNA probe. A 1.8-kb GRP mRNA transcript was detected in hemocytes, fat body, and epidermal cells but not in Malpighian tubules, silk gland, and midgut of naive silkworm (Fig. 5A). This result indicates that the GRP mRNA transcript is constitutively expressed in the fat body and epidermal cells as major sites and in hemocytes as a minor site. The inducibility of the GRP mRNA synthesis by the infection of microorganisms was investigated by Northern blot analysis of total RNA from the fat body of bacteria-injected and yeast-injected silkworms using the GRP probe. The amount of the GRP mRNA transcript was shown to increase and to remain at a high level from 6 to 24 h after injection of bacteria (E. cloacae) and from 12 to 36 h after injection of yeast (C. albicans) (Fig. 4B). The same blot was dehybridized and re-hybridized with a cDNA probe of cecropin B, an inducible anti-bacterial peptide, or -tubulin, a constitutively expressed protein. The cecropin B mRNA transcripts were confirmed to increase after injection of bacteria or yeast in a kinetics paralleled to that observed in the expression of the GRP mRNA transcript. These results indicated that the expression of GRP gene is inducible not only by a bacterial challenge but also by a yeast challenge and that the GRP is an acute-phase protein.

View larger version (53K): [in this window] [in a new window]   Fig. 5.   Northern blot analyses of tissue-specific expression (A) and inducibility (B) of GRP mRNA. A, poly(A)+ RNA preparations were obtained from tissues of silkworm larvae on day 5 of the fifth instar. 2 µg of poly(A)+ RNA was loaded on each lane. Sources of the poly(A)+ RNA preparations loaded to the lanes were as follows: F, fat body; E, epidermal cells; H, hemocytes; M, Malpighian tubules; S, silk gland; and G, midgut. B, silkworm larvae at the same developmental stage as in A were challenged by E. cloacae or C. albicans as described under "Experimental Procedures." Total RNA was extracted from the fat body at the indicated times after the challenge. Total RNA (30 µg) was subjected to Northern blot analysis where the blots were hybridized separately with 32P-labeled cDNA probes of silkworm GRP, cecropin B, and -tubulin. The names of the probes and microorganisms for the challenge are indicated at the left and top of the figure, respectively.

The Binding Domain of GRP for -1,3-Glucan-- In the present study, the 20-kDa fragment covering the N-terminal region of GRP was shown to bind to -1,3-glucan (Fig. 1). To characterize more specifically the -1,3-glucan binding domain in this region, we constructed 13 deletions of the N-terminal region of GRP fused with bacterial thioredoxin (Trx-G), as shown in Fig. 6A. All the fusion proteins were expressed as soluble proteins and subjected to a curdlan binding assay. The results showed that Trx-G1-34, 35-69, 70-119, 1-69, 35-119, 10-102, 18-102, 1-86, and 1-94 failed to bind -1,3-glucan. However, the binding abilities of Trx-G1-119, which corresponds to the fusion protein of Trx with the entire amino acid sequence of the 20-kDa fragment, and Trx-G1-102 were not very much different from the ability of the intact GRP under the experimental conditions. Trx-G1-96 and 1-99 appeared to bind to -1,3-glucan with weaker affinities than the GRP (Fig. 6B). To remove the Trx and spacer peptide from the Trx-G1-102, the fusion protein was cleaved by enterokinase. The resulting peptide corresponding to G1-102 was purified on Mono Q column chromatography. The purified G1-102 had affinity for -1,3-glucan. The -1,3-glucan preincubated with the purified recombinant peptide could not trigger the activation of the prophenoloxidase cascade (data not shown) as in the case of the 20-kDa fragment (Fig. 2). These results indicate that the first 102 amino acid residues of GRP constitute the -1,3-glucan binding domain of GRP.

View larger version (38K): [in this window] [in a new window]   Fig. 6.   The curdlan binding assay of fusion proteins of Trx and deletions of GRP. A, schematic diagram and summary of the deletions of GRP for the binding experiments. These deletions were expressed in E. coli as fusion proteins with Trx. A rectangle on top represents GRP and the amino acid residue numbers are given below. The position of the peptide bond cleaved by -chymotrypsin is indicated by a closed triangle. On the left of the lower figure, amino acid residues at the N-terminal and the C-terminal of each deletion are indicated by a set of numbers that refer to amino acid residues of GRP as in Fig. 3. The location of deletions in the primary structure of 20-kDa fragment is indicated by a bar. In the right column, the results of the binding experiments of the fusion proteins are summarized by using the following signs: +, binding; ±, weak binding; , no binding. B, SDS-PAGE of the extracts of E. coli expressing deletions and the results of curdlan binding assay of the deletions. Methods for preparation of the extracts from the transformed E. coli are described under "Experimental Procedures." After the extracts were analyzed by SDS-PAGE (left panel), the expressed fusion proteins in the extracts were quantified densitometrically. The extracts (about 50-200 µl) containing 100 µg of the fusion protein that correspond to about 2.5 nmol of the proteins and 2.5 nmol of the GRP were subjected to curdlan binding assay. One-tenth of proteins eluted from the curdlan beads was subjected to SDS-PAGE (middle panel). The proteins bound to the beads were quantified by densitometry of the stained gel, and their molar amounts were determined from the calibration curves. The binding abilities of the fusion proteins are given in the right panel as (moles of bound fusion protein)/(moles of bound GRP) × 100. Samples applied to lanes 1-5 originated from E. coli expressing fusion proteins as follows: lane 1, Trx-G1-94; lane 2, Trx-G1-96; lane 3, Trx-G1-99; lane 4, Trx-G1-102; lane 5, Trx-G1-119. M, marker proteins.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In the present study, cDNA of the silkworm GRP was cloned. The amino acid sequence that was deduced from the GRP cDNA has a putative signal sequence consisting of 16 amino acids and a mature protein of 479 amino acid residues. The molecular masses calculated from the deduced amino acid sequence and determined by MALDI mass spectrometry of the purified GRP were 53,859.7 and 54,571 Da, respectively. This discrepancy may be due to the post-translational modification of the nascent GRP molecule. In fact, a potential N-linked glycosylation site was identified at Asn³⁶², and the amino acid could not be detected by automated Edman degradation. We previously reported that the GRP migrated to the position corresponding to 62 kDa in SDS-PAGE as shown in Fig. 1 (8). It is probable that the silkworm GRP behaved anomalously in SDS-PAGE and migrated slower than the speed that was expected from its molecular mass determined by MALDI spectrometry. A similar anomalous behavior in SDS-PAGE was experienced with the 20-kDa fragment of the GRP but not with the 43-kDa fragment, suggesting that the anomalous behavior of the GRP may be assigned to the property of the 20-kDa fragment.

The deduced GRP amino acid sequence does not have the sequence that has often been observed with a transmembrane domain of receptors, but a glycosylphosphatidylinositol anchor attachment site was detected in the hydrophobic C-terminal region of the GRP. Our previous report on the immunocytochemical localization indicated that the GRP was contained not only in the plasma fraction but also in the granules of granulocytes and in the spherules and on the surfaces of spherulocytes of the silkworm B. mori (34). Western blot analysis of the fat body and integument failed to detect the GRP, but our recent preliminary study on the immunolocalization indicated that the GRP is present also on the surfaces of the fat body cells.² These observations seem to suggest that the GRP is present both as free-floating protein in the plasma and cell-bound form on the surface. The cell-bound GRP might cooperate with other proteins with a transmembrane domain to transduce the signal for the recognition of -1,3-glucan.

The prophenoloxidase cascade has also been reported in several crustaceans and an earthworm species. Crayfish -1,3-glucan-binding protein (GBP) is reported to be a lipoprotein with dual functions working as a recognition protein for -1,3-glucan and a lipid transfer protein (35). The crustacean GBP has been shown to be present in plasma. The complex of the crustacean GBP and -1,3-glucan was demonstrated to trigger a receptor-mediated degranulation of hemocytes (36). The content of the GBP in plasma has been reported to be very high, about 0.1-0.4% of plasma protein of the crayfish Pacifastacus leniusculus (19). Previously, we developed a method to remove specifically GRP from the plasma fraction of the silkworm hemolymph of the fifth instar larvae, and we named the fraction plasma-CPB (37). We have observed that -1,3-glucan could not trigger prophenoloxidase cascade in the plasma-CPB in which lipophorin is present. In insect hemolymph, lipophorin is a plasma lipoprotein and functions as the lipid transfer protein (38), and it is speculated to be an insect coagulogen (39). The lipophorin seems to be the only major lipoprotein present in the hemolymph of fifth instar silkworm larvae. Crayfish GBP and silkworm GRP are reported to be 100 and 62 kDa in SDS-PAGE, respectively, and their other molecular properties are very different. Thus, it appears to be certain that crayfish GBP and silkworm GRP are not homologous proteins and that the silkworm GRP is not a lipophorin or lipoprotein. Coelomic cytolytic factor 1 (CCF-1) of the earthworm, Eisenia foetida, was originally described to be responsible for the cytolytic, opsonizing, and hemolytic properties of the coelomic fluid. The CCF-1 has recently been reported to bind with -1,3- glucan and lipopolysaccharide (20). As the binding of the CCF-1 with -1,3-glucan elicits the activation of the earthworm prophenoloxidase cascade, the CCF-1 is considered to be functionally equivalent to the silkworm GRP in terms of activation of the cascade with -1,3-glucan.

When the silkworm prophenoloxidase cascade is triggered by -1,3-glucan or peptidoglycan, zymogens of serine protease are activated (23). To test the possibility that the GRP is cleaved in the process of the activation of the cascade, the cascade in plasma was triggered with -1,3-glucan and the change of molecular weight of the GRP during the activation of the cascade was investigated by Western blot analysis using a specific antibody against the GRP. A change, however, was not detected.³ The factor G of horseshoe crab initiates the activation of the blood coagulation cascade when it binds to -1,3-glucan. Factor G consists of subunit  with affinity for -1,3-glucan and subunit  with a protease domain. The subunit  is degraded when factor G is incubated with -1,3-glucan, and its subunit  becomes enzymatically active (21). Judging from the deduced amino acid sequence, it is almost clear that silkworm GRP itself is not a protease zymogen. This is consistent with our previous observation that a complex of the purified GRP with -1,3-glucan did not hydrolyze any of the 26 peptidyl-7-amino-4-methylcoumarins, substrates for various proteases (8). Furthermore, we examined the behavior of GRP on gel-permeation chromatography to get information of the possibility that GRP is present in non-covalent association with other proteins with a protease domain in the plasma fraction of the silkworm hemolymph. In chromatography of the plasma, the GRP was eluted in a symmetrical peak at the same retention time as the purified GRP,³ indicating that the GRP is not present as a non-covalently associated form with other macromolecules. These observations indicate that GRP is different from factor G not only structurally but also may be different in the mode of its action. More thorough study on the mechanism by which a complex of GRP and -1,3-glucan activates the immediate downstream component of the prophenoloxidase cascade is necessary. Such study would answer our question: what kind of activity does the GRP have when it binds to -1,3-glucan?

To identify the domain responsible for the binding to -1,3-glucan, we prepared recombinant deletion fragments of the N-terminal region (Tyr¹ to Phe¹¹⁹) of the GRP. By examining the affinity of the deletions for -1,3-glucan by the curdlan binding assay, the sequence consisting of the N-terminal 10 residues and the three residues from Asp¹⁰⁰ to Ala¹⁰² were shown to be important for the deletions to exhibit affinity for -1,3-glucan (Fig. 6). The N-terminal region (20k-Da fragment) does not contain a domain displaying similarity to the putative catalytic domain of -1,3-glucanase of microbial origin. Among the invertebrate proteins having a glucanase-like domain in Fig. 4, the domains of the GRP, GNBP, CCF-1, and factor G have been shown to be inactive catalytically as -1,3-glucanase. The GRP and factor G have specific binding affinity for -1,3-glucan. The GNBP has been reported to bind to Gram-negative bacteria but not to -1,3-glucan, and the CCF-1 binds to both -1,3-glucan and lipopolysaccharide. Glucanase-like domains of the GNBP and CCF-1 were speculated to be binding sites for saccharide. The binding domain of factor G for -1,3-glucan has been suggested to be three tandem repeats of a xylanase A-like domain (33). However, unambiguous identification of the binding domain for -1,3-glucan of the invertebrate proteins with a glucanase-like region was achieved in the present study for the first time. With regard to a bacterial glucanase, -1,3-1,4-glucanase from Bacillus licheniformis, its catalytically active site has been demonstrated to be present in the glucanase-like domain that is homologous to those shown in Fig. 4. The glutamic acids and aspartic acid in the sequence of Glu-Ile-Asp-Ile-Glu, of which location corresponds to Glu-Ile-Asp-Ile-Met-Glu in Fig. 4, were proved to be essential or important for the enzymatic activity (40). With another glucanase, Bacillus circulans -1,3-glucanase A1 containing a glucanase domain, the binding domain for -1,3-glucan has been reported to be the N-terminal region that is different from the glucanase domain (41). These observations indicate that the binding domain and the catalytic domain of bacterial -1,3-glucanases are located separately in their primary structures. Therefore, it seems to be reasonable that the 43-kDa fragment containing a glucanase-like domain of GRP was shown not to display appreciable affinity for -1,3-glucan. The -1,3-glucan binding region (Tyr¹ to Ala¹⁰²) of GRP shows only 26.5% identity and 40% similarity to the corresponding N-terminal region of GNBP, whereas the glucanase-like region (Thr²⁶⁴ -Pro³⁸⁶) of GRP displayed 49.6% identity and 62.6% similarity to the glucanase-like region (Ser²⁴³-Ala³⁷²) of GNBP. In light of this result, it is understandable that Lee et al. (12) could not detect any apparent affinity of GNBP for -1,3-glucan. The -1,3-glucan binding region (Tyr¹ to Ala¹⁰²) of the GRP does not display significant similarity to the xylanase A domain of factor G. Furthermore, in sequences of factor G and CCF-1, we could not find a sequence displaying significant similarity to the -1,3-glucan binding region (Tyr¹ to Ala¹⁰²) of GRP. From these observations, it seems reasonable to speculate that there exist more than two kinds of domains with specific affinity for -1,3-glucan among invertebrate defense molecules that have affinity for -1,3-glucan. Identification of the binding domain of factor G and CCF-1 for -1,3-glucan is awaited. Such knowledge together with our present observations of the binding domain of GRP would contribute to advance our understanding of the phylogeny and functions of protein domains with affinity for the -1,3-glucan.

A feature of GRP is that the synthesis is induced by a bacterial or yeast challenge, although the gene is constitutively expressed. This inducibility suggests that GRP can be classified as an acute-phase protein. We recently reported that the expression of the PGRP gene is also induced by a bacterial or peptidoglycan challenge and that the promoter region contains several cis-regulatory elements such as cAMP response element, NF-B-like element, and GATA motif that have been found in both insect and mammalian acute-phase protein genes. It is possible that the promoter region of the GRP gene has such cis-regulatory elements. We have observed before that the prophenoloxidase cascade could not be triggered by -1,3-glucan or peptidoglycan under the conditions where the recognition protein concentration is low (8, 9). It is probable that the concentration of the recognition proteins decreases in vivo after insects are invaded by bacteria or fungi. Therefore, one possible physiological role of the inducibility of the GRP and PGRP genes is to maintain the concentrations of the recognition proteins higher than a certain level in the hemolymph and to make the prophenoloxidase cascade constantly ready to be triggered by microbes with -1,3-glucan or peptidoglycan.

It is well known that glucanase is distributed in bacteria, fungi, and plants. Among Metazoa, only the sea urchin has been shown to have a gene coding for -1,3-glucanase. Bachman and McClay (32) argued for the occurrence of the extremely ancient divergences of glucanases in the prokaryotic/eukaryotic separation. Although it is not known why the glucanase-like region without the catalytic activity is conserved in some proteins of invertebrates, it seems to be probable that all these proteins work as pattern-recognition proteins in their primary immune responses to microbes and construct a family of glucanase-like proteins. Our search of the current data base showed that a vertebrate protein with sequence similarity to the glucanase-like region of GRP has not been deposited yet. It should be noted, however, that homologues present both in insects and in mammals are currently being found to be employed in their innate immune mechanisms. Considering this and the fact that our studies on the mechanisms for the recognition of -1,3-glucan as foreign have not advanced in either vertebrates or invertebrates, there seems to be a possibility that recognition molecules homologous to the proteins with a glucanase-like region will be found in vertebrates in the future.

	ACKNOWLEDGEMENT

We are grateful to Dr. Y. Hayakawa in our laboratory for technical advice.

	FOOTNOTES

^* This work was supported in part by Research Grants 05740502, 09265201, and 09304075 from the Ministry of Education, Science, Sports and Culture of Japan.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EMBL Data Bank with accession number(s) AB026441.

To whom correspondence and reprint requests should be addressed: The Institute of Low Temperature Science, Hokkaido University, Kita-ku Sapporo 060-0819, Japan. Tel.: 81-11-706-6878; Fax: 81-11-706-7142; E-mail: ochiai@orange.lowtem.hokudai.ac.jp.

² M. Ashida and M. Ochiai, unpublished observations.

³ M. Ochiai, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: PGRP, peptidoglycan recognition protein; PAMP, pathogen-associated molecular pattern; GRP, -1,3-glucan recognition protein; GNBP, Gram-negative bacteria-binding protein; PAGE, polyacrylamide gel electrophoresis; MALDI, matrix-assisted laser desorption and ionization; PCR, polymerase chain reaction; Trx, thioredoxin; Trx-G, fusion protein of Trx and a deletion of the N-terminal region of GRP; GBP, -1,3-glucan binding protein; CCF-1, coelomic cytolytic factor 1; bis-tris, 2-[bis(2-hydroxyethyl)- amino]-2-(hydroxymethyl)-propane-1,3-diol; bp, base pair(s); kbp, kilobase pair(s).

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES