| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 42, 40455-40463, October 17, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Department of Biological Chemistry, the University of Michigan Medical School, Ann Arbor, Michigan 48109-0606
Received for publication, June 26, 2003 , and in revised form, August 4, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
35 kDa associated by non-covalent bonds. From a cDNA library, three similar full-length cDNAs, termed LSLa, LSLb, and LSLc, were generated, each of which had an open reading frame of 945 bp encoding 315 amino acid residues. These proteins share 80–90% sequence identity and showed structural similarity to bacterial toxins: mosquitocidal toxin (MTX2) from Bacillus sphaericus and 
toxin from Clostridium septicum. Native and recombinant forms of LSL showed hemagglutination and hemolytic activity and both activities were inhibited by N-acetyllactosamine, whereas a C-terminal deletion mutant of LSLa (LSLa-D1) retained hemagglutination, but not hemolytic activity, indicating the N-terminal domain is a carbohydrate recognition domain and the C-terminal domain functions as an oligomerization domain. The LSL-mediated hemolysis was protected osmotically by polyethylene glycol 4000 and maltohexaose. Inhibition studies showed that lacto-N-neotetraose (Gal
1–4GlcNAc1–3Gal1–4Glc) was the best inhibitor for LSL. These results indicate that LSL is a novel pore-forming lectin homologous to bacterial toxins. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
-linked galactose at the cell surface. This binding is a prerequisite for translocation of the A-chain into the cytosol. The properties of a hemolytic lectin from marine invertebrate Cucumaria echinata (CEL-III) have also been well characterized (4, 5). CEL-III, a 47-kDa polypeptide, is also composed of two tandemly repeated (QXW)3 domains similar to ricin B-chain and a putative oligomerization domain. CEL-III is a pore-forming protein, which binds to carbohydrates on the cell membrane, forms oligomers, and creates ion-permeable pores. The isolation and partial characterization of an N-acetyllactosamine-specific hemolytic lectin from the mushroom Laetiporus sulfureus were first reported by Konska et al. (6). The hemolytic lectin was isolated by affinity chromatography on Sepharose and was stated to be a heterotetrameric protein of 190 kDa, with subunits of 36 and 60 kDa.
In this paper, we have undertaken the detailed sugar binding specificity, cDNA cloning, expression, and characterization of this unique hemolytic lectin and closely related lectins from L. sulfureus. Binding studies by the techniques of hemagglutination, hemolysis inhibition, and quantitative precipitation showed that the lectin possesses an extended binding site, which recognizes Gal1–4GlcNAc1–3Gal1–4Glc (lacto-N-neotetraose). cDNA sequencing revealed that the L. sulfureus lectins contain a (QXW)3-like motif at their N-termini and show sequence homology to bacterial toxins. The three lectin genes and a deletion mutant were expressed in Escherichia coli and their physicochemical characterization is investigated to shed light on the mechanism of hemolysis mediated by their novel-type of structures.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
1-acid glycoproteins, bovine submaxillary mucin, etc.) were purchased from Sigma. Except for asialofetuin, asialoglycoproteins were prepared by heating the corresponding native glycoproteins in 0.1 M hydrochloric acid at 80 °C for 1 h, followed by dialysis and lyophilization; the removal of sialic acid was confirmed by the thiobarbituric assay (7). Purification of the Lectin—L. sulfureus mushrooms were harvested in October 2002 in Ann Arbor, MI. Fresh mushrooms were cleaned of debris and chopped into small pieces. The chopped tissue (40 g) was homogenized in a Waring Blendor at 4 °C in 200 ml of PBS1 (10 mM sodium phosphate, 0.15 M NaCl, 0.04% sodium azide, pH 7.2) or PBSE (PBS containing 1 mM EDTA) containing 1 ml/liter protein inhibitor mixture (product P8215; Sigma). The homogenate was stirred for 3 h at 4 °C, squeezed through four layers of cheesecloth, and centrifuged at 12,000 x g for 15 min. The supernatant solution, filtered through glass wool to remove a small amount of floating debris was applied onto an affinity column (2.5 x 15 cm) of Sepharose 4B, which had been equilibrated in PBSE. The column was washed with PBSE until the absorbance of the effluent at 280 nm decreased to a minimum value. The affinity adsorbed lectin was eluted with 0.1 M lactose in PBSE and dialyzed against PBSE. Approximately 2 mg of purified lectin were obtained from 40 g of fresh mushroom.
Protein and Carbohydrate Estimations—Protein was determined by a method of Lowry et al. (8) using bovine serum albumin as a standard. Total neutral sugars were determined by the phenol/sulfuric acid method (9).
SDS-PAGE—SDS-PAGE was carried out on 0.75-mm slab gels in an alkaline buffer system (Tris glycine, pH 8.3) (10), using a mini-Protean II apparatus (Bio-Rad). BenchMark prestained protein molecular mass standards used in SDS-PAGE were from Invitrogen.
Subunit Structure—The subunit structure of purified lectins was determined by gel filtration through a G2000-SWXL Progel-TSK column (30 x 0.78 cm; Supelco, Bellefonte, PA) using a Beckman System Gold high-performance liquid chromatography system as described previously (11) and by SDS-PAGE performed on samples with and without heating for 5 min in boiling water, and in the presence or absence of 2-mercaptoethanol.
Hemolytic Assay and Hapten Inhibition Assay—Hemolytic activity of the lectins was determined by absorbance at 540 nm because of hemoglobin release. A lectin solution (50 µl) and PBSE (50 µl) as a control were mixed with 50 µl of a 5% suspension of human type A erythrocytes, and incubated for 30 min at room temperature. After centrifugation at 1,500 x g for 5 min, the absorbance of the supernatant solution was measured at 540 nm against a control. Hapten inhibition was assayed in the same system. Increasing amounts of saccharides in 25 µl were incubated with 25 µl of purified lectin (1 µg) for 30 min at room temperature followed by addition of 50 µl of 5% cell suspension after 30 min. After incubation at 37 °C for 30 min followed by centrifugation to remove membranes and intact cells, the absorbance of the supernatant solution was recorded at 540 nm.
Hemagglutination Assay and Hapten Inhibition Assay—The hemagglutinating activity of the lectin was determined by a 2-fold serial dilution procedure using formaldehyde-treated (12) human and rabbit erythrocytes as described previously (13). The sample (10 µl) (2-fold serial dilutions in PBSE) was combined with 10 µl of a 2% cell suspension in v-shaped microtiter plates (96-well) and hemagglutination was observed after incubation for 30 min at room temperature. The titer was defined as the reciprocal value of the end point dilution causing hemagglutination. For hapten inhibition of hemagglutination, 10 µl of the serially diluted saccharide solutions were incubated with 10 µl of 8 agglutinin units of the lectin in microtiter plates followed by addition after 30 min of an erythrocyte suspension (20 µl). The lowest concentration of saccharide that visibly decreased the extent of agglutination was defined as the minimum inhibitory concentration.
Quantitative Precipitation Assay—Assays were performed by a microprecipitation procedure (14). Varying amounts of glycoproteins or polysaccharides ranging from 0 to 100 µg were added to 20 µg of the purified lectin in a total volume of 120 µl of PBSE. After incubation at 37 °C for 1 h, the reaction mixtures were stored at 4 °C for 48 h. The precipitates formed were washed twice with 500 µl of ice-cold PBS, dissolved in 0.05 M NaOH, and assayed for protein by Lowry's method (8) using bovine serum albumin as standard.
Peptide Sequencing and Analysis—Peptide sequences were determined by the Protein Structure Facility at the University of Michigan Core Facilities. Briefly, purified protein was reduced, S-carboxamidomethylated with monoiodoacetamide, and then digested with trypsin. The derived peptides were separated by reversed-phase high-performance liquid chromatography and the amino acid sequences of the isolated peptides were analyzed by a gas-phase protein sequencer.
RNA Isolation and cDNA Cloning—Fresh tissue of the mushroom was ground to a powder with a pestle under liquid nitrogen. Total cellular RNA was isolated with Concert Plant RNA reagent (Invitrogen, CA) and subsequently poly(A)+ RNA was isolated with Micro-Fast-Track 2.0 kit (Invitrogen). Using this protocol, 5 µg of poly(A)+ RNA/1 g of mushroom was isolated.
Adapter-ligated cDNA library was constructed with the Marathon cDNA amplification kit (Clontech). Two degenerate forward primers (LSLF1, GCNGTNWSNACNGTNGARWSNGGNATHATHAA; LSLF2, GTNGARWSNGGNATHATHAAYGTNCCNTTYAC) were designed from the amino acid sequence AVSTVDSGIINVPFT of a trypsin-digested fragment of the purified lectin for rapid amplification of cDNA ends (RACE). 3'-RACE was conducted with a combination of primers, adapter primer 1 (Invitrogen) and LSLF1, and Platinum Pfx DNA polymerase (Invitrogen) as follows. DNA was denatured at 94 °C for 3 min, followed by three-step cycles (40 cycles); 92 °C for 0.5 min, 55 °C for 0.5 min, and 68 °C for 1 min, and further extended at 68 °C for 5 min. The amplified DNA fragment was subsequently amplified with LSLF2 and adapter primer 2 (Invitrogen). The amplified 1-kilobase pair fragment was cloned using the Zero Blunt TOPO PCR cloning kit (Invitrogen). Inserted DNA was sequenced with T7 and SP6 primers by the DNA Sequencing Core facility of the University of Michigan, and three similar but different genes (termed LSLa, LSLb, and LSLc) including poly(A)+ were obtained. A gene-specific reverse primer (LSLR1, ATCATTGACTTTCGACACGCAACATTGATAGAGC) was designed and 5'-RACE was conducted with adapter primer 1 and LSLR1. Finally, a gene-specific forward primer (LSLF3, CCTAACGTACACGTTACGCTCCCATTCACC) was designed and 3'-RACE was conducted with adapter primer 1 and LSLF3 to confirm full-length cDNA sequences.
Construction, Expression, and Purification of Recombinant L. sulfureus lectin—The protein coding region of LSL amplified by PCR using synthetic oligonucleotides incorporating NdeI and BamHI, or NdeI and XhoI restriction enzyme sites were used for cloning purposes. The approximately 1-kilobase pair PCR products cloned into expression vector pET-43a (Novagen), yielding pET-LSLa, pET-LSLb, and pET-LSLc.
A deletion mutant (LSLa-D1) was generated by PCR using the GeneTailor site-directed mutagenesis system (Invitrogen), pET-LSLa as a template, and mutagenesis primers (LSLaD1F, CATCATCCCAGACCCAGGAGTAGTCATTTAAT; LSLaD1R, CTCCTGGGTCTGGGATGATGTATTCTCGAG) following by the manufacturer's protocol. Underlined nucleotides of LSLaD1F (see above) were altered to introduce a stop codon to generate a deletion mutant of LSLa (Met1-Glu187) and the nucleotide sequence of mutant clones was verified by DNA sequencing.
Nova Blue (DE3) strain of E. coli harboring expression vector was pre-cultured in 5 ml of Luria broth (LB) medium containing 50 µg/ml ampicillin at 37 °C for 3 h and was added to 1 liter of medium. After the optical density at 600 nm reached 0.4–0.6, 1 ml of 1 M isopropyl-1-thio--D-galactoside was added to the medium, and the cells were further cultured at 25 °C overnight. The induced cells were collected by centrifugation, resuspended in a lysis buffer (10 mM sodium phosphate, 0.15 M NaCl, 1 mM EDTA, 0.04% sodium azide, pH 7.2 (PBS), containing 1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 1 mM 2-mercaptoethanol, and proteinase inhibitor mixture (Roche Diagnostics)), and sonicated at 0 °C. The insoluble fraction was removed by centrifugation at 10,000 x g for 30 min at 0 °C. Recombinant lectins were purified from the soluble fraction by absorption on a Sepharose 4B column and elution by 0.2 M lactose.
Circular Dichroism Spectroscopy—Circular dichroism (CD) spectra were measured on a Jasco J-715 spectropolarimeter (Japan Spectroscopic Co.) using a protein concentration of 0.3 mg/ml in 10 mM phosphate buffer (pH 7.2) and an optical path length of 1 mm. Buffer baselines were measured under the same conditions and subtracted from the corresponding spectrum. The data were transformed to mean residue molar ellipticity and smoothed using the standard analysis program. Secondary structure was calculated based on the method of Yang et al. (15).
Sequence Data Processing—Multiple sequence alignment was performed by the Clustal W program (16). Homologous sequences were searched for by the FASTA program. Internal repeats were searched for by the SMART program (17, 18). Hydropathy profile analysis was performed on the basis of the primary structure by the procedure of Kyte and Doolittle (19).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
To prevent any metalloproteinase degradation of the lectin, metal-free buffer containing EDTA and proteinase inhibitor mixture was used throughout subsequent purification. It should be noted that hemagglutination activity of the lectin was unchanged by extensive dialysis of the lectin in metal-free buffer containing EDTA, indicating the absence of a divalent metal ion requirement (data not shown). As shown in Fig. 1, upon SDS-PAGE at pH 8.3 with unheated samples, the purified lectin (nLSL-p2) showed a single band at approximately 170 kDa. When the lectin was boiled in SDS with 2-mercaptoethanol, a single band of 35 kDa was observed, suggesting that the native structure is an oligomer of this monomer. This preparation was used for further investigation.
Subunit Structure—The molecular mass was also estimated by size exclusion chromatography on a silica-based matrix. The purified lectin migrated as a single, nearly symmetrical band of approximately 140 kDa, based on standardization with known proteins (data not shown). Together with the SDS-PAGE analysis, these results indicate that the lectin exists as a tetramer of subunits of approximately 35 kDa that requires boiling in SDS to dissociate completely.
Hemagglutination and Hemolysis—The hemagglutination and hemolytic activity of nLSL was assayed toward erythrocytes from several mammalian species. The lectin agglutinated formaldehyde-treated rabbit erythrocytes at a minimum concentration of approximately 1 ng/ml, whereas formaldehyde-treated sheep, dog, and human erythrocytes of any blood type required 80 or 160 ng/ml for agglutination (Table I). As shown in Fig. 2A, the lectin hemolyzed rabbit, sheep, and human erythrocytes of any blood type in a dose-dependent manner. The concentrations of the lectin required to induce 50% hemolysis were determined to be 0.9, 14, 17, 14, and 17 µg/ml for rabbit, human A, human B, human O, and sheep erythrocytes, respectively. In contrast, dog erythrocytes were only slightly hemolyzed by the lectin (Fig. 2A).
|
|
Amino Acid Sequencing—N-terminal amino acid sequencing of the electroblotted protein yielded a very weak signal, indicating that the protein is blocked at the N terminal. Despite the weak signal, the sequence XDIYIPPEGL was determined. Enzymatic digestion with trypsin, purification of peptide fragments, and Edman degradation of the tryptic peptides yielded three peptide sequences: AVSTVDSGIINVPFT, STGFEVTTEGI, and GVSSWDLR.
Carbohydrate Analysis—Neutral sugar was not detected in the purified lectin by the phenol/sulfuric acid assay.
Inhibition of Hemagglutination—The sugar binding specificity of the lectin was initially investigated by hemagglutination inhibition assay (Table II). Of the mono- and disaccharides tested, Me-N-acetyllactosaminide was the best inhibitor.
|
Inhibition of Hemolysis—The detailed sugar binding properties of the lectin were further elucidated by hemolysis inhibition assay (Fig. 3) and the concentration of saccharides required for 50% inhibition is shown in Table III. Gal1,4Glc (lactose) was about 4–10 times more active than lactose-related saccharides such as Gal1,4Man, Gal1,4Fru (lactulose), and Gal1,4Glucitol (lactitol). Gal1,4Gal (galactobiose) showed only 26% inhibition at 12.5 mM. Gal1,4GlcNAc (N-acetyllactosamine) was approximately twice as active as Gal1,4Glc, about 8 times more active than Gal1,3GlcNAc (lacto-N-biose) and Gal1,3Ara, indicating that the lectin has a preference for the 1,4-linked isomers. The Me- and p-nitrophenyl glycosides of Gal1,4GlcNAc were the best inhibitors among disaccharides tested, and were 3.8 times more active than Gal1,4Glc and twice more active than Gal1,4GlcNAc. Of the oligosaccharides tested, Gal1,4GlcNAc1,3Gal1,4Glc (lacto-N-neo- tetraose) was the best inhibitor, which was about 27 times more active than Gal1,4Glc, and 4 times more active than Gal1,3GlcNAc1,3Gal1,4Glc and (Gal1,4GlcNAc)21,6 Gal1,4Glc (lacto-N-neohexaose).
|
|
Quantitative Precipitation—The ability of various glycoproteins and polysaccharides to precipitate the purified nLSL was investigated by the quantitative precipitation assay; among those examined were: both native and desialylated fetuin, 1-acid glycoproteins, thyroglobulin, transferrin, ovine submaxillary mucin, bovine submaxillary mucin, glycophorin, -tetragalactosylglucose, and pneumococcus type 14 polysaccharide. Although the hemagglutination and hemolytic activity of nLSL was specifically inhibited by lactose and N-acetyllactosamine, asialo 1-acid glycoproteins and asialofetuin, which contain mixtures of N-acetyllactosamine-type N-linked glycans, failed to precipitate the lectin. Of many glycoproteins and polysaccharides examined, only pneumococcus type 14 polysaccharide gave a precipitation curve with nLSL (Fig. 4).
|
Molecular Cloning of L. sulfureus Lectin—3'-RACE with the adapter primers and the degenerate primers that were designed from the trypsin-digested fragment yielded a 1-kilobase pair product. Cloning and sequencing of 5'-RACE products generated three different full-length nucleotide sequences including polyadenylation (termed LSLa, LSLb, and LSLc). Of 17 clones, 9 clones appeared to contain LSLa, whereas 6 clones contained LSLb and 2 clones contained LSLc. LSLa, LSLb, and LSLc contain a 57-bp 5'-untranslated region, followed by a 945-bp open reading frame encoding 315 amino acid residues, and the 218, 122, and 124 bp 3'-untranslated regions, respectively (Fig. 5). None of them contain an adenylation signal sequence. The calculated molecular masses of LSLa (34,964 Da), LSLb (35,150 Da), and LSLc (35,101 Da) without N-terminal Met are in good agreement with the molecular mass of native lectin (35 kDa) estimated by SDS-PAGE.
|
The deduced amino acid sequences revealed that LSLa contains precisely the same sequence as the sequenced peptides of purified lectin, whereas the two other lectins contain slightly different sequences (Fig. 5). No signal sequence could be discerned in the deduced amino acid sequences of the LSLs, indicating their synthesis on free polysomes. LSLa and LSLc have two, and LSLb has one potential N-linked glycosylation site(s) (Asn-Xaa-Ser/Thr) in their sequences (Fig. 5). However, as indicated above, no carbohydrate was detected in the native lectin. These data suggested that LSLa corresponds to the same protein as nLSL.
Osmotic Protection of Erythrocytes—Because L. sulfureus lectin showed sequence identity with several pore-forming toxins, it was postulated that the hemolytic activity of the lectin was because of the formation of ion-permeable pores in the membrane. Therefore, osmotic protection experiments were performed in the presence of human type A erythrocytes. If the hydrodynamic diameter of the protectant is larger than that of the membrane pore formed by LSL, it would be expected to suppress the LSL-induced lysis by counterbalancing the osmotic pressure arising from intracellular solutes. As shown in Fig. 6, when erythrocytes were incubated with nLSL in the presence of neutral sugars and polyethylene glycols of different molecular diameters, lysis was inhibited increasingly as the size of the molecules increased. The LSL-induced hemolysis was entirely suppressed by polyethylene glycol 4000 and maltoheptaose. The hydrodynamic diameters of polyethylene glycol 4000 has been previously estimated to be 3.8 nm (20). When erythrocytes incubated with LSL and polyethylene glycol 4000 were subsequently washed with PBSE and resuspended in PBSE, immediate lysis was observed. These results indicated that LSL formed ion-permeable pores with a functional diameter smaller than 3.8 nm in the cell membranes of human erythrocytes.
|
Construction, Isolation, and Characterization of Recombinant L. sulfureus Lectins—To express recombinant lectins, the entire open reading frames of LSLa, LSLb, and LSLc were cloned into expression vector pET-43a and introduced into E. coli strain Nova Blue (DE3) cells. Active recombinant LSLa and LSLb were isolated from E. coli extracts by a single chromatographic step on a Sepharose 4B column. The yields of rLSLa and rLSLb, each purified by affinity chromatography, were approximately 4 and 1 mg, respectively, from 1 liter of culture. However, rLSLc accumulated as aggregates in inclusion bodies and the active form was not obtained.
SDS-PAGE analysis of the subunit structures of recombinant lectins was conducted (Fig. 1). Although rLSLa and rLSLb were presumably initially expressed in the crude lysates as intact forms (35 kDa) in E. coli, the recombinant lectins were largely cleaved by bacterial proteases, producing truncated forms of 21 and 22 kDa. All attempts to prevent the proteolytic digestion of LSLa and LSLb failed; this included the use of sonication buffer containing mixtures of several protease inhibitors, detergent (Nonidet P-40), and reducing agent (2-mercaptoethanol) at low temperature (0 °C). Interestingly, not only the intact recombinant LSLa and LSLb, but also the truncated forms bound to Sepharose were eluted by lactose. Samples prepared without heating showed a band at 170 kDa and a lesser amount of a 35-kDa band, suggesting that 35-kDa subunits are associated into an oligomer (Fig. 1). This analysis indicates that recombinant L. sulfureus lectins are composed of an intact form of 35-kDa subunits and proteolytically digested forms of 21- and 22-kDa subunits, which retain binding affinity to Sepharose. The molecular mass of the recombinant lectins was also estimated by size exclusion chromatography. rLSLa migrated as a major band of approximately 20 kDa and a minor band of 140 kDa, whereas rLSLb migrated as a major band of approximately 22 kDa and a minor band of 140 kDa.
The N-terminal amino acid sequences of intact and truncated forms of recombinant lectins were determined: the 35-kDa subunit of rLSLa was H2N-TDIYIPPEGL and the 21-kDa subunit was H2N-TDIYIPPEGLYFRLLGFASRQVIFA. These results indicate that the N-terminal methionine of these proteins was removed, presumably by a bacterial methionyl aminopeptidase. The truncated 21-kDa subunits of rLSLa and rLSLb represent the N-terminal domains of the intact 35-kDa subunits and those of the C-terminal domains (14 kDa) were presumably digested by bacterial proteases.
rLSLa and rLSLb showed both hemagglutination and hemolytic activity (Table I and Fig. 2), indicating that the cloned genes encode hemolytic lectins. Minimum concentrations required for hemagglutination activity of rLSLa and rLSLb against formaldehyde-treated rabbit erythrocytes were estimated to be 40 and 80 ng/ml, respectively, which made them much less active than nLSL (1 ng/ml), whereas the hemolytic activity of rLSLa was virtually the same as nLSL and that of rLSLb was somewhat more active than nLSL (Fig. 2 and Table I).
Sugar binding specificity of recombinant L. sulfureus lectins was determined by hemagglutination and hemolysis inhibition assays (Tables II and III). Little or no differences were observed between nLSL and rLSLa, indicating that rLSLa encodes the native lectin, whereas the sugar binding specificity of rLSLb was different from nLSL and rLSLa. For rLSLb, Gal1,4GlcNAc (N-acetyllactosamine) had the same inhibitory activity as lactose and (Gal1,4GlcNAc)21,6Gal1,4Glc (lacto-N-neohexaose). Gal1,3GlcNAc1,3Gal1,4Glc (lacto-N-tetraose) and Gal1,4GlcNAc1,3Gal1,4Glc (lacto-N-neotetraose) were only 1.5 to 2.3 times more active than Gal1,4Glc (lactose), indicating that rLSLb might have a reduced capacity to accommodate tetra- and hexasaccharide. rLSLa and rLSLb both gave reduced precipitin reactions with pneumococcus type 14 polysaccharide (Fig. 4).
Construction, Isolation, and Characterization of a Deletion Mutant of rLSLa—A C-terminal deletion mutant of rLSLa (Met1-Glu187, calculated molecular mass: 21272.41), termed LSLa-D1, was constructed by site-directed mutagenesis and expressed in E. coli. LSLa-D1 bound to a Sepharose column and was eluted by lactose. The yield of LSLa-D1 (approximately 60 mg from a 1-liter culture) was 15 times higher than that of rLSLa. Upon SDS-PAGE, LSLa-D1 gave a major band at approximately 22 kDa, in the presence or absence of 2-mercaptoethanol in SDS sample buffer (Fig. 1). Upon size exclusion chromatography, intact LSLa-D1 in solution showed a major peak at approximately 22 kDa, indicating that LSLa-D1 is a monomer of 22 kDa.
The N-terminal amino acid sequence of LSLa-D1 was H2N-TDIYIPPEGLY identical to LSLa. LSLa-D1 agglutinated rabbit, human, sheep, and dog erythrocytes. Minimum concentrations required for hemagglutination activity of LSLa-D1 against formaldehyde-treated rabbit erythrocytes was estimated to be 0.2 µg/ml, making it much less active than nLSL (1 ng/ml) (Table I). LSLa-D1 did not show hemolytic activity to any mammalian erythrocytes used in this study (Table I and Fig. 2). Hemagglutination inhibition assay showed that the sugar binding specificity of LSLa-D1 was essentially the same as the native and recombinant lectins (Table II).
CD Spectra of LSL—The spectrum of nLSL showed a maximum molar ellipticity at 198 nm (Fig. 8). The content of -helix, -structure, -turn, and random coil structures were calculated to be 3.8, 89.4, 0, and 6.8%, respectively, based on the procedure of Yang et al. (15). The CD spectra of rLSLa, rLSLb, and LSLa-D1 were essentially similar to that of nLSL, although the positive peaks of their spectra shifted to 197 nm. rLSLa, rLSLb, and LSLa-D1 were also mainly composed of -structure (86.5–87.9%). The influence of N-acetyllactosamine on the CD spectrum of LSL was examined and observed to have little effect (Fig. 7).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
1,4Glc/GlcNAc (lactose/N-acetyllactosamine)-related sequences are potent inhibitors of LSL as shown by Konska et al. (6). A notable observation was the high affinity of LSL for Gal1,4GlcNAc1,3Gal1,4Glc (lacto-N-neotetraose). Pneumococcus type 14 polysaccharide, which is composed of repeating lacto-Nneotetraose-like structures (Fig. 4B), was the only polysaccharide tested to give a precipitin curve with LSL, although Gal1,4GlcNAc containing N-linked glycoproteins (asialofetuin and asialo 1-acid glycoproteins) did not precipitate LSL. These results suggested that the binding site of LSL might be a cavity large enough to accommodate a tetrasaccharide of -anomers of galactose, and is most complementary to lacto-N-neotetraose and Gal1,4Glc/GlcNAc-related sequences.
A FASTA search revealed that LSL shows sequence homologies (22–25%) with mosquitocidal toxins (MTX2) from Grampositive bacterium Bacillus sphaericus and toxin from Grampositive bacterium Clostridium septicum. MTX2, a 31.8-kDa polypeptide, is lethal to mosquito larvae, but its toxic mechanism has not been clarified (21, 22). The toxin is a poreforming protein homologous to aerolysin from Aeromonas hydrophila, which is probably the best characterized watersoluble toxin (23, 24). The toxin and aerolysin recognize glycosylphosphatidylinositol-anchored protein receptors on the cell surface and the glycosyl portion of the receptors may be the major determinant for the activity of aerolysin (25, 26). It is probable that the carbohydrate binding activity makes toxins more specific to their target organisms.
A SMART search revealed that LSLa, LSLb, and LSLc contain three tandemly repeated subdomains at their N termini (Fig. 8). Four and nine amino acid residues are invariant and conserved among the subdomains of LSLa and LSLb, respectively, which showed both hemagglutination and hemolytic activity. LSLc, not recovered in active form from E. coli, lacks some of these residues, indicating that LSLc might not be a hemolytic lectin. Closer analysis of the three subdomains of LSL indicates that they all contain the conserved Gln/Asn-Xaa-Phe motif, which resembles the QXW motif present in the ricin B-chain (Fig. 8). In ricin, the conserved Trp is necessary for hydrophobic packing of the core structure, whereas Gln interacts with the conserved Asp that hydrogen bonds with the O-3 and O-4 of galactose. The corresponding Gln/Asn and Phe in LSL could function in a similar fashion. The structural properties of LSL, which are composed of (QXW)3-like domains at its N-terminal is also the same as that found in the mushroom lectin from Marasmius oreades (27). M. oreades lectin is a blood group B-specific lectin composed of 293 amino acid residues that also contains a (QXW)3 domain at its N terminus. However, the sequence homology between LSL and M. oreades lectin is very low (10%).
The expression system for rLSLa and rLSLb in E. coli was constructed and the products obtained were characterized with regard to their hemagglutination, hemolysis, and sugar binding properties. Although the C-terminal truncated forms of rLSLa and rLSLb lost the ability to form oligomeric structures as shown by the results of SDS-PAGE (Fig. 1) and size exclusion chromatography (data not shown), the recombinant lectins showed hemagglutination and hemolytic activity toward several mammalian erythrocytes (Table I). Inhibition data indicated that rLSLa and rLSLb recognize Gal1, 4GlcNAc1,3Gal1,4Glc (lacto-N-neotetraose) and Gal1,4Glc/GlcNAc-related sequences, thus resembling the native L. sulfureus lectin in terms of their carbohydrate binding specificity (Tables II and III). The N-terminal amino acid sequences of the truncated 21-kDa subunits of rLSLa and rLSLb corresponded to those of their corresponding intact forms, indicating that the truncated forms represent the N-terminal domains of the intact 35-kDa proteins, and those of the C-terminal domains (14 kDa) were cleaved. Interestingly, although rLSLa and rLSLb were largely digested by bacterial proteinases into their C-terminal truncated forms, their hemolytic activity was indistinguishable from that of the native lectin, whereas the hemagglutination activity of these recombinant lectins was reduced. From these results, it was suggested that the N-terminal sequences composed of a (QXW)3-like domain is probably the carbohydrate-binding domain, whereas the C-terminal domain might function as an oligomerization domain. To clarify this possibility, we expressed a C-terminal deletion mutant of rLSLa, termed LSLa-D1 (Figs. 1, 5, and 8). LSLa-D1 bound to a Sepharose column and was eluted by lactose. SDS-PAGE and size exclusion chromatography of LSLa-D1 showed it to be a monomer consisting of a 21-kDa subunit. On SDS-PAGE, it migrated with similar mobility with a truncated 21-kDa subunit of rLSLa (Fig. 1). Although LSLa-D1 lost its ability to form an oligomeric structure, it agglutinated mammalian erythrocytes (Table I). Hemagglutination inhibition assay showed that LSLa-D1 displayed the same sugar binding specificity as rLSLa. In contrast, LSLa-D1 did not hemolyze any mammalian erythrocytes. These results indicated that the N-terminal domain (187 amino acids, 21 kDa) of rLSLa consists of a carbohydrate recognition (QXW)3-like domain containing at least two binding sites, whereas the C-terminal domain (128 amino acids, 14 kDa) is an oligomerization domain (Fig. 8). The fact that the monomeric C-terminal deletion mutant (LSLa-D1) agglutinated erythrocytes and binds onto an affinity column is a strong indication that it is multivalent containing two or more binding sites. Oligomer formation might be required for LSL-mediated hemolysis similar to bacterial toxins (20, 21). A hydropathy profile of LSLa showed that the profile of the N-terminal domain (1–187 amino acids) differs from that of the C-terminal domain (188–315 amino acids). It also showed that the C-terminal domain contains two hydrophobic clusters, one cluster between amino acids 200 and 230, and a second between amino acids 250 and 280. These hydrophobic clusters might be related to the formation of a tetrameric structure for LSL.
CD analysis revealed that nLSL is composed mainly of -structure (89%). The CD spectra of recombinant lectins and a deletion mutant were similar to that of nLSL, indicating that these recombinant proteins were properly folded during expression in E. coli.
The identification of a pore-forming lectin from the mushroom L. sulfureus homologous to bacterial toxins is of interest in understanding the importance of lectin-carbohydrate interactions in organisms. The fact that LSL contains a (QXW)3-like domains at its N terminus, similar to that of M. oreades lectin, suggests that LSL and M. oreades lectin might have evolved from the same ancestral protein. Similarly, the fact that LSL is hemolytic, and shows homology to bacterial toxins, could indicate that it might have evolved from the ancestral protein to acquire an oligomerization domain in forming pores at the membrane. The recombinantly expressed lectins and a deletion mutant described in this paper will be useful not only for investigating the mechanism of LSL-mediated hemolysis, but also for numerous applications in biological and medical research because of their unique specificity toward Gal1,4GlcNAc1,3Gal1,4Glc (lacto-N-neotetraose)-related structures.
|
FOOTNOTES |
|---|
* This work was supported in part by National Institutes of Health Grant GM29470. 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. 
Recipient of a Research Fellowship of the Naito Foundation.
To whom correspondence should be addressed: Dept. of Biological Chemistry, University of Michigan Medical School, Ann Arbor, MI 48109-0606. Tel.: 734-763-3511; Fax: 734-763-4581; E-mail: igoldste{at}umich.edu.
1 The abbreviations used are: PBS, phosphate-buffered saline; PBSE, phosphate-buffered saline containing EDTA; LSL, L. sulphureus lectin; nLSL, native LSL; nLSL-p1, nLSL purified with PBS as an extraction buffer; nLSL-p2, native L. sulphureus lectin purified with PBSE containing protease inhibitor mixture as an extraction buffer; rLSL, recombinant LSL; RACE, rapid amplification of cDNA ends; Me, methyl.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Hatakeyama, H. Unno, Y. Kouzuma, T. Uchida, S. Eto, H. Hidemura, N. Kato, M. Yonekura, and M. Kusunoki C-type Lectin-like Carbohydrate Recognition of the Hemolytic Lectin CEL-III Containing Ricin-type -Trefoil Folds J. Biol. Chem., December 28, 2007; 282(52): 37826 - 37835. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Sher, Y. Fishman, M. Zhang, M. Lebendiker, A. Gaathon, J.-M. Mancheno, and E. Zlotkin Hydralysins, a New Category of {beta}-Pore-forming Toxins in Cnidaria J. Biol. Chem., June 17, 2005; 280(24): 22847 - 22855. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Mancheno, H. Tateno, I. J. Goldstein, M. Martinez-Ripoll, and J. A. Hermoso Structural Analysis of the Laetiporus sulphureus Hemolytic Pore-forming Lectin in Complex with Sugars J. Biol. Chem., April 29, 2005; 280(17): 17251 - 17259. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Uchida, T. Yamasaki, S. Eto, H. Sugawara, G. Kurisu, A. Nakagawa, M. Kusunoki, and T. Hatakeyama Crystal Structure of the Hemolytic Lectin CEL-III Isolated from the Marine Invertebrate Cucumaria echinata: IMPLICATIONS OF DOMAIN STRUCTURE FOR ITS MEMBRANE PORE-FORMATION MECHANISM J. Biol. Chem., August 27, 2004; 279(35): 37133 - 37141. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||
, human B; 
, human 
; 
, sheep; x, dog; *, rabbit erythrocytes.
subdomains in the ricin B-chain (A) and sequence alignment of tandemly repeated subdomains (1, 2, and 3) of LSLa and LSLb (B). Amino acid residues involved in galactose binding in ricin B are denoted by asterisks, and amino acid residues required for formation of the hydrophobic core are denoted by plus signs. Invariant amino acid residues among the subdomains of LSL are boxed and shaded. Conserved amino acid residues among the subdomains of LSL are shaded. The underline denotes the conserved (Q/N)XF motif. C, schematic representation of LSL and a deletion mutant of LSLa (LSLa-D1). D, hydrophathy profile of LSLa. Analysis was performed by the procedure of Kyte and Doolittle (