A novel lectin from Sarcophaga. Its purification, characterization, and cDNA cloning.

A novel C-type lectin that agglutinates rabbit red cells was purified from NIH-Sape-4 cells derived from the flesh fly (Sarcophaga peregrina), and its cDNA was isolated. This lectin, named granulocytin, appeared to be a trimer of a 20-kDa subunit consisting of 151 amino acid residues. The gene for granulocytin was activated in third instar larvae, and its expression was enhanced when the larval body wall was injured. In third instar larvae, granulocytin was found to be synthesized by hemocytes and secreted into the hemolymph. The molecular mass and gene expression patterns of granulocytin were very similar to those of Drosophila lectin that we reported previously (Haq, S., Kubo, T., Kurata, S., Kobayashi, A., and Natori, S. (1996) J. Biol. Chem. 271, 20213-20218). However, these two lectins showed amino acid identities of 20% at most, and no significant hapten sugar for granulocytin was identified.

Insects respond to microbial infection through cellular and humoral defense mechanisms and are known to secrete various antimicrobia proteins and lectins into their hemolymph in response to bacterial challenges and body injury (1)(2)(3)(4)(5)(6)(7)(8). Most of these humoral defense proteins are synthesized by the fat body (9 -12) and some by hemocytes (13). Although several antimicrobial proteins have been isolated from various insects, sequenced, and characterized (14), the primary sequences of very few insect lectins have been determined.
In a previous study, we purified a galactose-binding C-type lectin from the hemolymph of immunized larvae of Sarcophaga peregrina (flesh fly) and characterized it (15)(16)(17). This lectin (Sarcophaga lectin) was a large molecule with a molecular mass of 190 kDa, consisting of 32-and 30-kDa subunits in a molar ratio of 2:1. These two subunits were essentially the same protein derived from a single gene, and their difference in size was due to glycosylation. Sarcophaga lectin was found to be needed for the elimination of sheep red cells introduced into the larval body cavity (18). It was also found to play roles in imaginal disc differentiation, suggesting that Sarcophaga lectin functions in both defense and development of this insect (19 -21).
To study the biological role of Sarcophaga lectin in the development of this insect, we identified its specific binding pro-tein with a molecular mass of 10 kDa (22). This binding protein was isolated from the membrane fraction of NIH-Sape-4 cells, an embryonic cell line of Sarcophaga, and was shown to be distributed heterogeneously on the surface of imaginal discs (22). Incidentally, we found that a protein that reacted with a monoclonal antibody raised against the Sarcophaga lectinbinding protein was a novel C-type lectin of this insect. In this paper, we report the purification, characterization, and cDNA cloning of this new lectin named granulocytin.

MATERIALS AND METHODS
Animals and Collection of Hemocytes, Hemolymph, and Fat Bodies-Sarcophaga peregrina was reared by the methods of Ohtaki (23). Hemolymph was collected by cutting off the anterior tip of a third instar larva with fine scissors and collecting the drop of hemolymph that exuded into 5 ml of insect saline (130 mM NaCl, 5 mM KCl, 1 mM CaCl 2 ). Hemocytes were collected by centrifuging the hemolymph solution for 3 min at 100 ϫ g at 4°C, and the fat body was excised from each larva under a binocular microscope.
Purification of a 20-kDa Protein That Reacts with SLB1 from NIH-Sape-4 Cells-NIH-Sape-4 cells (20 g) were suspended in 400 ml of 10 mM Tris-HCl buffer (pH 7.4) containing 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, 0.1 g/ml pepstatin, and 100 mM galactose and homogenized in a glass homogenizer with a Teflon pestle. The homogenate was centrifuged at 100 ϫ g for 1 min to remove the cell debris, and the supernatant was centrifuged at 100,000 ϫ g for 30 min. The resulting precipitate was washed twice by repeating centrifugation, suspended in 200 ml of 10 mM Tris-HCl buffer (pH 7.4) containing 150 mM NaCl and 0.5% (v/v) Nonidet P-40, kept for 1 h at 4°C, and centrifuged at 100,000 ϫ g for 30 min. The resulting supernatant was used as the solubilized protein fraction.
The solubilized protein was diluted 4-fold with 10 mM Tris-HCl buffer (pH 7.4) containing 150 mM NaCl and applied to an SLB1-Separose column (1 ϫ 5 cm) equilibrated with 10 mM Tris-HCl buffer (pH 7.4) containing 150 mM NaCl and 0.125% Nonidet P-40. The column was washed well, and the adsorbed material was eluted with the above buffer containing 8 M urea. This fraction was dialyzed against 10 mM Tris-HCl buffer (pH 7.4) containing 40 mM octylglucoside and 150 mM NaCl, concentrated to about 500 ml, and then subjected to fast protein liquid chromatography on a Mono Q column (Amersham Pharmacia Biotech). A protein that reacted with SLB1 was recovered in the flowthrough fraction. At this stage, the protein was pure enough to yield a single 20-kDa band after SDS-polyacrylamide gel electrophoresis. The protein concentration was measured by the method of Lowry et al. (26) using bovine serum albumin as the standard.
Cloning and Sequencing of the 20-kDa Protein cDNA-An antibody * This work was supported by a grant-in-aid for scientific research from the Ministry of Education, Science, Sports, and Culture of Japan and by Core Research for Evolutional Science and Technology (CREST) of Japan Science and Technology Corporation (JST). The costs of publication of this article were defrayed in part by the payment of page charges. This 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 /EBI Data Bank with accession number(s) AB006076.
was raised against the 20-kDa protein and affinity-purified using the method described previously (27,28). About 10,000 clones of an NIH-Sape-4 cDNA library (29,30) were screened with this affinity-purified antibody and the alkaline phosphatase-conjugated secondary antibody (Bio-Rad). For nucleotide sequencing of the cDNA, the insert DNA was subcloned into pBluescript (Stratagene) and sequenced by the dideoxy chain termination method using a Taq dye terminator sequencing kit (Applied Biosystems). The nucleotide sequences of both strands were determined.
Electrophoresis and Immunoblotting-Electrophoresis on SDS-polyacrylamide slab gel was carried out by the method of Laemmli (31). Samples were denatured by boiling them for 5 min in 1% SDS containing 2% (v/v) 2-mercaptoethanol, and after electrophoresis, the gel was stained using the method of Fairbanks et al. (32). For immunoblotting, the proteins separated by electrophoresis were transferred electrophoretically from the gel onto filters (Immobilon-P, Millipore), which were immersed in 20 mM Tris-HCl buffer (pH 7.9) containing 5% (w/v) skim milk for 1 h. After washing with a rinsing solution (10 mM Tris-HCl buffer, pH 7.9, containing 150 mM NaCl, 1 mM EDTA, 0.1% Triton X-100, 0.01% (w/v) sodium azide, and 0.25% skim milk), they were dipped in rinsing solution containing the affinity-purified antibody against the 20-kDa protein and kept for 1 h. Then, they were washed well with rinsing solution, transferred to 5 ml rinsing solution containing radioiodinated anti-rabbit IgG (18.5 kBq), and kept at room temperature for 1 h. Finally, they were washed well with rinsing solution, dried, and subjected to autoradiography using Kodak XAR film.
Assay of Hemagglutinating Activity-Rabbit red blood cells (RRBC) 1 were washed well and suspended in buffered insect saline. To measure hemagglutinating activity, 25 l of serially diluted sample solution was mixed with 25 l of 2.5% (w/v) RRBC suspension in a well of a Vbottomed microtiter plate, and then incubated at 37°C for 1 h. Agglutinated RRBC formed a diffuse mat, whereas unagglutinated RRBC formed a clear dot on the well bottom.
Immunofluorescence Study-Hemocytes were suspended in buffered insect saline at a density of approximately 10 6 cells/ml, and 15 l of suspension was placed in each well of a 12-well multitest slide. The hemocytes were allowed to settle for 15 min at room temperature, fixed for 10 min by adding 15 l of 4% paraformaldehyde, 0.02% glutaraldehyde, phosphate-buffered saline, the slides were rinsed with phosphatebuffered saline and blocked with 100 mM glycine, 1% bovine serum albumin, phosphate-buffered saline for 1 h at room temperature. The blocking reagent was replaced with 0.1% bovine serum albumin/phosphate-buffered saline containing 0.02% NaN 3 and the slides were stored at 4°C. 20 l of affinity-purified antibody solution (20 mg/ml) was added to each well and incubated for 2 h at room temperature. Then, the antibody solution was removed, 20 l of fluorescein isothiocyanatelabeled secondary antibody solution was added, the slides were incubated for 2 h at room temperature, washed, mounted, and observed using an Olympus BH-2 fluorescence microscope.

Affinity Purification of 20-kDa Protein from NIH-Sape-4
Cells-In a previous study, we isolated a Sarcophaga lectinbinding protein from NIH-Sape-4 cells and raised a monoclonal antibody, named SLB1, against it as described before (22). To purify the Sarcophaga lectin-binding protein in quantity from the membrane fraction of NIH-Sape-4 cells, we performed affinity chromatography using SLB1-conjugated Sepharose CL-4B followed by fast protein liquid chromatography on a Mono Q column. The protein thus obtained yield a broad band with a molecular mass of 20 kDa after SDS-polyacrylamide gel electrophoresis (Fig. 1). Unlike the 10-kDa Sarcophaga lectinbinding protein, this 20-kDa protein showed essentially no Sarcophaga lectin binding activity, indicating that it is not the Sarcophaga lectin-binding protein we characterized before (22). The molecular mass of the intact protein, estimated from the elution volume from a molecular sieve column of Suparose 12, was 60 kDa, suggesting that the intact protein is a trimer of the 20-kDa subunit. About 100 g of 20-kDa protein was obtained from 4 ϫ 10 10 NIH-Sape-4 cells.
To characterize this 20-kDa protein further, we raised a polyclonal antibody against the purified protein, affinity-purified it and used it for immunoblotting and cDNA cloning. Immunoblotting revealed that not only NIH-Sape-4 cells, but also their culture medium contained this 20-kDa protein (Fig. 2A). The immunoreactive proteins in the culture medium that migrated faster than the 20-kDa protein were probably its degradation products. These results suggest that the 20-kDa protein is not a membrane protein, but a secretory protein, which was confirmed by detecting the 20-kDa protein in hemolymph. As shown in Fig. 2B, this protein was detected in hemocytes, hemolymph, and fat bodies from third instar larvae, suggesting that this protein is a hemolymph protein, produced possibly by 1 The abbreviations used are: RRBC, rabbit red blood cells; CRD, carbohydrate recognition domain. Isolation of cDNA for the 20-kDa Protein-To determine the amino acid sequence of the 20-kDa protein, we tried to isolate its cDNA by screening an NIH-Sape-4 cDNA library with the affinity-purified antibody. Three positive clones had identical nucleotide sequences, and the protein encoded in this cDNA contained five peptides obtained by digesting purified 20-kDa protein with trypsin: WFK, LITLK, TYREDELLSQYLK, QTY-QPLPEFWLR, and ANDNIDLSDINK. The 13 amino-terminal residues, APLTKWFKTDNNT, which we determined separately, were also present in this protein. Therefore, we concluded that this is cDNA for the 20-kDa protein.
The nucleotide and deduced amino acid sequences encoded in this cDNA are shown in Fig. 3. The amino-terminal residue of the 20-kDa protein was assigned as Ala at position 23. Therefore, this cDNA was shown to encode the 20-kDa protein consisting of 151 residues and its leader peptide consisting of 22 residues. Two possible N-glycosylation sites (33), Asn-Asn-Thr and Asn-Ser-Thr, were found in the 20-kDa protein. Therefore, we examined whether the purified 20-kDa protein was deglycosylated by treating it with N-glycanase and subjecting the product to SDS-polyacrylamide gel electrophoresis to determine its molecular mass. The broad band of intact 20-kDa protein shifted to the 14-kDa position and became a sharp band, suggesting that the 20-kDa protein is a glycoprotein (Fig. 4).
As shown in Fig. 5, the 20-kDa protein was found to contain a carbohydrate recognition domain(CRD)-like sequence with significant similarity (20 -27%) to the CRDs of Drosophila (34) and Sarcophaga lectin (35). This sequence was found to contain four Cys residues that are conserved in the CRDs of many C-type lectins (36), suggesting that the 20-kDa protein is a C-type lectin, but it lacked two of three residues, Glu and Asp, which are reported to be responsible for sugar and calcium binding (37,38). The molecular mass of the 20-kDa protein was about the same as that of Drosophila lectin.
Hemagglutinating Activity of the 20-kDa Protein-As the 20-kDa protein was found to be structurally related to Sarcophaga and Drosophila C-type lectins, we examined its hemagglutinating activity using various red cells in the presence of 1 mM CaCl 2 . No agglutination of sheep, pig, chicken, horse, guinea pig, bovine, or human red cells was observed, but significant agglutination of RRBC did occur. Calcium was essential for this hemagglutinating activity, indicating that the 20-kDa protein is a C-type lectin. Next we examined the effects of various carbohydrates on the hemagglutinating activity of the 20-kDa protein. No specific hapten sugar has been identified in the monosaccharides tested so far. Glucose, galactose, fucose, xylose, rhamnose, N-acetylmuramic acid, and N-acetylneuraminic acid (all at 100 mM) inhibited the hemagglutinating activity by 50%. Bovine submaxillary gland mucin (10 g/ml) inhibited the hemagglutinating activity, but porcine stomach mucin had no inhibitory effect. Lipopolysaccharide from Escherichia coli 0111,B4 was a weak inhibitor (100 g/ml) but the other lipopolysaccharides tested showed no inhibitory effects. This carbohydrate specificity was clearly different from those of Sarcophaga and Drosophila lectins, the hapten sugars of which were identified as galactose (15,34).
Immunofluorescence Study of the 20-kDa Protein-As the 20-kDa protein was found to be a novel C-type lectin of Sarcophaga synthesized by hemocytes, we investigated which cells produce this protein by carrying out an immunofluorescence study. As shown in Fig. 6, fluorescence was detected only in large granulocytes not in other cells such as a spindly cell indicated by the arrow, which is probably a plasmatocyte. These results indicate that this protein is produced by large granulocytes. Furthermore, the fluorescence in these cells was distributed heterogeneously and seemed to be concentrated in the granules, suggesting that the 20-kDa protein is stored in the granules and eventually secreted into the hemolymph. To differentiate this novel lectin from Sarcophaga lectin, we named it granulocytin.
Expression of the Granulocytin Gene-Immunoblotting detected granulocytin both in hemocytes and fat bodies of third instar larvae. However, as shown in Fig. 7A, Northern blotting detected 0.9 kilobase mRNA for granulocytin exclusively in  4. Digestion of the purified 20-kDa protein with N-gly-canase. The purified 20-kDa protein (5 mg) was incubated with or without 1 unit of N-glycanase for 18 h at 37°C, and the mixture was subjected to SDS-polyacrylamide gel electrophoresis. Lanes: 1, with N-glycanase; 2, without N-glycanase; 3, purified 20-kDa protein. The gel was calibrated as described in the legend to Fig. 1, and the asterisk indicates the position of N-glycanase.
hemocytes, not in the fat body, indicating that this lectin was synthesized by the hemocytes.
Expression of the granulocytin gene was examined at various developmental stages of Sarcophaga. No significant expression was detected throughout the embryonic stage to the second instar larval stage, but this gene was expressed in third instar larvae and the expression continued until the early pupal stage. Then, this gene was turned off during the rest of the pupal stage and was activated again in the newly emerged adult flies (Fig. 7B). We examined the post-injury acute-phase expression of the granulocytin gene in hemocytes from third instar larvae. As shown in Fig. 7C, expression of this gene was enhanced 6 h after pricking the larval body wall with a thin needle and declined slightly 18 h later. These expression patterns were very similar to those of the Drosophila lectin gene (34). DISCUSSION In this study, we identified a novel C-type lectin of Sarcophaga and purified it from the membrane fraction of NIH-Sape-4 cells by SLB1-Sepharose chromatography. SLB1 is a monoclonal antibody against Sarcophaga lectin-binding protein with a molecular mass of 10 kDa (22). As this novel lectin (granulocytin) clearly differed from the Sarcophaga lectinbinding protein, SLB1 should have affinity for both the Sarcophaga lectin-binding protein and granulocytin. The epitope for SLB1 in the Sarcophaga lectin-binding protein may be a specific carbohydrate chain, which granulocytin may also contain. In which case, why was only granulocytin recovered from the SLB1-Sepharose column? Granulocytin content of NIH-Sape-4 cells is probably an order of magnitude higher than that of the Sarcophaga lectin-binding protein and, therefore, the former protein was trapped selectively by the column.
We solubilized granulocytin from the membrane fraction of NIH-Sape-4 cells with octylglucoside as described before (22). As granulocytin turned out to be a soluble hemolymph protein, hemolymph would probably have been a better source for purification. We might have purified granulocytin contaminating the membrane fraction or the granules that store granulocytin Hemocytes from naive third instar larvae were treated with an affinity-purified antibody followed by fluorescein isothiocyanate-conjugated secondary IgG, and the nuclei were localized by 4Ј,6-diamidino-2-phenylindole staining. A, immunofluorescence of the 20-kDa protein; B, localization of the nuclei (the same field); C, bright field photograph (the same field). The arrow in each panel indicates a spindly cell that does not express the 20-kDa protein.
might have been recovered in the membrane fraction we prepared. Indeed, we found that the latter is the case. The purified 20-kDa protein yielded a broad band after SDS-polyacrylamide gel electrophoresis and sometimes yielded a doublet band. As the amino-terminal sequences of the proteins from these two bands were identical, their different electrophoretic mobilities were probably due to differences in their carbohydrate chains.
Of the seven types of red cells tested, only those from rabbits were agglutinated by this lectin possibly because this lectin recognizes a specific sequence of carbohydrates present only in RRBC. Granulocytin appears to be the Sarcophaga counterpart of Drosophila lectin because their molecular masses and gene expression patterns during development and expression during the acute-phase response to injury are very similar (34). However, the carbohydrate specificities of these two lectins are different. At present, the biological role of granulocytin is unknown. However, as it is synthesized prior to pupation, production of this lectin is probably prerequisite for metamorphosis. This lectin may play a role in self/nonself recognition by hemocytes during the initial stage of metamorphosis when larval tissue ingestion by hemocytes starts. Hemocytes have to recognize larval tissues as nonself to ingest them, and lectin is a good candidate for a modulator of self/nonself recognition by hemocytes.
The Northern blotting analysis findings suggest that expression of the granulocytin gene is involved in the acute-phase response to injury, as is Drosophila lectin gene expression (34). These lectins may be involved in the defense systems that eliminate invading parasites as well as in the process that disintegrates unnecessary larval tissues during metamorphosis. Our results demonstrated clearly that granulocytin is synthesized exclusively by hemocytes. We detected a positive signal of granulocytin in the fat body by immunoblotting. This signal may have been due to hemocytes contaminating the fat body preparation.
Five of the seven residues commonly found in the CRDs from various C-type lectins are conserved in the structure of the putative CRD of granulocytin. Of these seven residues, Glu, Asp, and Asn (see Fig. 5) were recently reported to be responsible for sugar and calcium binding (37,38). Although the CRD of Drosophila lectin contains these three residues (34), that of granulocytin contains only Asn. The hapten sugar of Drosophila lectin was identified as galactose, but we could not identify that of granulocytin because many sugars at relatively high concentrations inhibited the hemagglutinating activity of granulocytin. The deficiency of these two residues might explain the broad and weak carbohydrate specificity of granulocytin. FIG. 7. Northern blot analysis of hemocytin gene expression. Total RNAs extracted from various sources were subjected to Northern blot hybridization using granulocytin cDNA as a probe. The recovery and integrity of each RNA were assessed from the 18 S ribosomal RNA pattern. A, comparison of hemocytes and fat body. Lanes: 1, RNA from hemocytes (5 g); 2, RNA from fat body (25 g). B, expression of the granulocytin gene during Sarcophaga development. Each lane contained 25 g total RNA. Sarcophaga is an ovoviviparous insect, so embryos were recovered from the female's uterus. RNA was extracted from lanes: 1, unfertilized eggs; 2-5, embryos from 5-, 7-, 9-, and 11-day-old flies; 6, first instar larvae; 7, second instar larvae; 8, early third instar larvae; 9, late third instar larvae; 10 -13, 1-, 3-, 5-, and 7-day-old pupae; 14, newly emerged adults. C, changes in hemocytin gene expression after injury to the larval body wall. Total RNA was extracted from hemocytes from naive and injured (by pricking with a thin needle) larvae. Each lane contained 5 g of RNA extracted from hemocytes from lanes: 1, naive third instar larvae; 2, injured larvae (6 h after pricking); 3, injured larvae (24 h after pricking). The arrows indicate the position of granulocytin mRNA.