Glycobiology Advance Access originally published online on June 13, 2006
Glycobiology 2006 16(10):891-901; doi:10.1093/glycob/cwl016
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Novel carbohydrate-binding activity of bovine liver ß-glucuronidase toward lactose/N-acetyllactosamine sequences
3 Course of Advanced Biosciences, Graduate School of Humanities and Sciences, Tokyo, Japan; 4 Department of Food Science, Otsuma University, Tokyo, Japan; and 5 The Glycoscience Institute, Ochanomizu University, Tokyo, Japan
2 To whom correspondence should be addressed; e-mail: hogawa{at}cc.ocha.ac.jp
Received on August 31, 2005; revised on May 18, 2006; accepted on June 11, 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionConflict of interest statementReferences |
|---|
ß-Glucuronidase is a lysosomal enzyme that plays an essential role in normal turnover of glycosaminoglycans and remodeling of the extracellular matrix components in both physiological and inflammatory states. The regulation mechanisms of enzyme activity and protein targeting of ß-glucuronidase have implications for the development of a variety of therapeutics. In this study, the effectiveness of various carbohydrate-immobilized adsorbents for the isolation of bovine liver ß-glucuronidase (BLG) from other glycosidases was tested. ß-Glucuronidase and contaminating glycosidases in commercial BLG preparations bound to and were coeluted from adsorbents immobilized with the substrate or an inhibitor of ß-glucuronidase, whereas ß-glucuronidase was found to bind exclusively with lactamyl–Sepharose among the adsorbents tested and to be effectively separated from other enzymes. Binding and elution studies demonstrated that the interaction of ß-glucuronidase with lactamyl–Sepharose is pH dependent and carbohydrate specific. BLG was purified to homogeneity by lactamyl affinity chromatography and subsequent anion-exchange high-performance liquid chromatography (HPLC). Lactose was found to activate ß-glucuronidase noncompetitively, indicating that the lactose-binding site is different from the substrate-binding site. Binding studies with biotinyl glycoproteins, lipids, and synthetic sugar probes revealed that ß-glucuronidase binds to N-acetyllactosamine/lactose-containing glycoconjugates at neutral pH. The results indicated the presence of N-acetyllactosamine/lactose-binding activity in BLG and provided an effective purification method utilizing the novel carbohydrate binding activity. The biological significance of the carbohydrate–specific interaction of ß-glucuronidase, which is different from the substrate recognition, is discussed.
Key words: affinity purification / glycoprobe / lactose and N-acetyllactosamine-binding / lysosomal enzyme / zymography / ß-glucuronidase
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConflict of interest statementReferences |
|---|
ß-Glucuronidase is an exoglycosidase that cleaves ß-glucuronic acid linkages from the nonreducing termini of glycosaminoglycans such as chondroitin sulfate, heparan sulfate, and hyaluronic acid. It is present in animals, plants, and bacteria as an essential enzyme for the normal degradation and turnover of components of the extracellular matrix (Paigen, 1989
). In animals, ß-glucuronidase is widely distributed in tissues, and high activity is detected in spleen, liver, and kidney; it also serves in the incorporation of ß-glucuronides of steroid hormones into tissues. Bovine liver ß-glucuronidase (BLG) [EC 3. 2. 1. 31] has been reported to be a homotetramer of 290 kDa (Himeno et al., 1974) and to have N-linked oligosaccharides with mannose 6-phosphate (Kaplan and Achord, 1977) (Natwicz et al., 1982), which is a targeting signal to lysosomes recognized by specific Golgi receptors to segregate BLG into transport vesicles via a glycan-specific sorting mechanism (Kornfeld, 1987).
ß-Glucuronidase from bovine liver is commercially available from several manufacturers. However, most ß-glucuronidase preparations are contaminated with many other proteins including several glycosidases, such as N-acetyl-ß-hexosaminidases and ß-galactosidase. It is difficult to separate ß-glucuronidase from the lysosomal protein mixtures, and multiple purification steps have been required for ß-glucuronidase preparation, that is, a combination of several methods using serial fractionations with ammonium sulfate, ethanol, and organic solvents, gel filtration chromatography, ion-exchange chromatography, and isoelectric chromatofocusing (Himeno et al., 1974; Ho, 1991). In this study, we were trying to isolate BLG from commercial preparations, and while searching for a suitable affinity adsorbent among those immobilized with various carbohydrates including its inhibitor and substrates, we unexpectedly discovered that BLG has binding activity toward lactamyl–Sepharose. Interaction of BLG with glycoconjugates and the effect of carbohydrates on the enzyme activity provide new insights into the biological functions of the carbohydrate binding of ß-glucuronidase.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConflict of interest statementReferences |
|---|
Materials
Commercial BLG was purchased from Worthington Biochemical Co. (Lakewood, NJ) in most experiments as a crude enzyme preparation and from P-L Biochemicals (present Pharmacia P-L Biochemicals Inc., Milwaukee, WI) or from Sigma-Aldrich Co. (St. Louis, MO) for comparison. p-Nitrophenyl-ß-D-glucuronide, 5-bromo-4-chloro-3-indolyl (X-) ß-D-glucuronide, ceramide (bovine brain), and galactosylceramide (bovine spinal cord) were purchased from Wako Pure Chemicals (Osaka, Japan). p-Nitrophenyl-ß-D-galactoside was purchased from Nakalai Tesque Inc. (Kyoto, Japan). p-Nitrophenyl-ß-D-galactoside, transferrin, fetuin, ovalbumin, bovine submaxillary gland mucin (BSM), bovine lactosylceramide, human spleen glucosylceramide, galactosylceramide, ceramide, bovine brain sulfatide, and streptavidin-biotinylated horseradish peroxidase complex (ABC-HRP) were purchased from Sigma-Aldrich Co. Saccharo-1,4-lactone and ß-galactosidase (from jack bean) were purchased from Seikagaku Kogyo (Tokyo, Japan). All biotinylated glycoprotein probes and their deglycosylated derivatives were prepared in our laboratory. Biotinylation was performed using EZ-linkTM sulfo-NHS-biotin (Pierce, Rockford, IL) according to the instruction manual. Asialoglycoproteins, asialo-agalactoglycoproteins, and ahexosamino-asialoagalactoglycoproteins were prepared from biotinylated glycoproteins by sequential glycosidase treatments with Vibrio cholerae neuraminidase (Roche Diagnostics, Basel, Switzerland) (0.1 units/mg glycoprotein) in 20 mM sodium acetate-buffered saline (pH 5.5), jack bean ß-galactosidase (Seikagaku Kogyo) (0.14 units/mg glycoprotein) in 50 mM sodium citrate buffer (pH 3.5) and then ß-N-acetylhexosaminidase (1.43 units/mg glycoprotein) in 50 mM sodium citrate buffer (pH 5.0), each at 37°C overnight. Biotinyl polymer (BP) sugar probes were purchased from GlycoTech Co. (Gaithersburg, MD).
Preparation of affinity adsorbents
Affinity adsorbents containing various carbohydrates as ligands were prepared using Sepharose 4B gel (Pharmacia, Uppsala, Sweden) in our laboratory. Lactose, maltose, melibiose, glucose, galactose, N-acetylchitotriose and N-acetylchondrosine were immobilized to amino Sepharose by reductive amination (Matsumoto et al., 1981) Saccharo-1,4-lactone was immobilized to amino Sepharose with the aid of N-ethyl-N'-(3-dimethyl aminopropyl)-carbodiimide hydrochloride (Harris et al., 1973). p-Aminophenyl ß-glucuronide was immobilized to formyl Sepharose by reductive amination (Ito et al., 1985). The immobilized carbohydrate concentrations of the adsorbents (µmol per g wet gel) were 45 for saccharo-1,4-lactone-, 35 for N-acetylchondrosine-, and 30–35 for other neutral saccharides coupled by reductive amination; lactamyl, maltamyl, melibiamyl, glucamyl, galactamyl and N-acetylchitotriamyl Sepharose. Those of p-aminophenylglucuronide- and N-acetylchitotriose-Sepharose were not determined.
Preparation of 1-deoxy-4-O-ß-D-galactopyranosyl-1-([2-hydroxyethyl]amino)-D-glucitol and analysis by nuclear magnetic resonance spectroscopy
1-Deoxy-4-o-ß-D-galactopyranosyl-1-([2-hydroxyethyl]amino)-D-glucitol (GHAG) was prepared by coupling 2-aminoethanol (366 µL) with lactose (0.3 M solution, 7.5 mL) by reductive amination with the pH adjusted to 7.5 with acetic acid. After preincubation for 2 h at 40°C, 300 mg of NaBH3CN was added and incubated for 2 h at 90°C; progress of the coupling reaction was confirmed by TLC (Holmes and O’Brien, 1979). The reaction mixture was diluted to 100 mL with water and applied to a charcoal column (2 x 35 cm), followed by washing with water and elution with a 0–20% gradient of ethanol. For determination of the structure, 1H- and 13C-nuclear magnetic resonance (NMR) spectra were obtained by using an ECA-800 spectrometer (JEOL, Tokyo) with a probe temperature of 60°C. The purified product was dried over P2O5 under vacuum and dissolved in (CD3)2SO (Me2SO-d6). Chemical shifts were referenced to tetramethylsilane. Spectral parameters of double quantum-filtered correlated spectroscopy (DQF-COSY), total correlated spectroscopy (TOCSY), heteronuclear multiple-quantum coherence (HMQC), and heteronuclear multiple-bond correlation (HMBC) spectra were as described previously (Iida-Tanaka et al., 2002).
Binding of ß-glucuronidase with various carbohydrate-immobilized Sepharose gels
Interactions between various adsorbents and BLG were studied by affinity chromatography on a small column (0.75 x 4.5 cm, Vt=2 mL). Commercial BLG (6 mg from Pharmacia P-L Biochemicals) was dissolved in 1 mL of 2 mM sodium acetate buffer–5 mM NaCl (pH 7.0) (buffer A) and applied to each column. The column was washed with buffer A and then sequentially eluted with 5 mM sodium acetate buffer–5 mM NaCl (pH 5.0) (buffer B), 10 mM sodium acetate buffer–5 mM NaCl (pH 5.0) (buffer C), 25 mM sodium acetate buffer–5 mM NaCl (pH 5.0) (buffer D), 50 mM sodium acetate buffer–5 mM NaCl (pH 5.0) (buffer E), 50 mM sodium acetate buffer–50 mM NaCl (pH 5.0) (buffer F), and 0.2 M sodium acetate buffer–0.2 M NaCl (pH 5.0) (buffer G). The eluted fractions (2 mL/fraction) were measured for enzyme activities, and proteins were detected by absorbance at 280 nm.
Binding and elution of BLG to lactamyl Sepharose
Lactamyl Sepharose 4B (0.1 g) and 1 mg of BLG (from Worthington) were incubated at 4°C for 4 h in a 1.5-mL tube with 0.3 mL of various buffers: 2 mM sodium acetate buffer (pH 5–6), ammonium acetate (pH 6), or Tris–HCl (pH 6–9), each containing 5 mM NaCl. After centrifugation at 900 g for 5 min, the supernatant was removed, and the gel was washed three times with the same buffer; then the bound proteins were eluted with buffer G for 4 h at 4°C, and the enzyme activity of the eluted fraction was measured. To study the elution conditions from lactamyl Sepharose gel, we incubated BLG with lactamyl Sepharose gel in 0.2 mL of 2 mM sodium acetate buffer–5 mM NaCl (pH 6.0) (buffer A'). After centrifugation, the supernatant was removed, the gel was washed three times with buffer A', then incubated in each solution for 4 h at 4°C, and the enzyme activity of the eluted fraction was measured.
Purification of BLG on a lactamyl–Sepharose column and anion-exchange high-performance liquid chromatography
For large-scale purification, the binding of BLG to lactamyl–Sepharose was performed by a batch-wise method to improve the binding capacity. BLG (Worthington or Sigma, 15 mg) was mixed with 13 mL of lactamyl–Sepharose 4B in buffer A' and incubated for 5 h at 4°C with gentle shaking, and then, the gel was poured into a column (1.5 x 8 cm). After extensive washing with buffer A', BLG was successively eluted with buffer C, 25 mM sodium acetate buffer–25 mM NaCl (pH 5.0) (buffer E'), and buffer G. Eluted fractions were monitored by absorbance at 280 nm and measured for enzyme activities. Alternative elution was performed with 0.1 M GHAG after 1 mg of BLG was applied onto a lactamyl–Sepharose column (0.75 x 4 cm). The peak fractions were pooled and concentrated with a Microcon YM-10 Filter Unit (Millipore, Billerica, MA).
For further purification, the buffer C-eluted fraction from the lactamyl column was applied to ion-exchange high-performance liquid chromatography (HPLC) on a DEAE-5PW column (21.5 x 150 mm, Tosoh Corp., Tokyo, Japan). Elution was performed at a flow rate of 1.0 mL/min at room temperature using two solvents, 10 mM Tris–HCl (pH 7.5) (TBS) containing 1 M NaCl. After injection of the sample, the NaCl concentration was increased linearly from 0 to 0.4 M, and 0.4–1.0 M in 100 min. Each peak was collected and concentrated to measure enzyme activities and subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and binding studies.
Polyacrylamide gel electrophoresis and zymography
Polyacrylamide gel electrophoresis in the presence of 0.1% SDS was carried out according to the method of Laemmli (1970) under nonreducing conditions using a 9.5 or 7.5% acrylamide separation gel, and native PAGE was performed in the absence of SDS (Davis, 1964). Protein bands were detected with Coomassie Brilliant Blue (CBB) or silver staining (Atto Corp., Tokyo, Japan). For zymography, the gel was incubated in 2.5% Triton X-100 for 1 h after SDS–PAGE to remove SDS or used directly after native-PAGE. The gels were washed and incubated in each substrate solution for several hours at 37°C with gentle shaking. Substrate solutions contained 1 mM X-ß-D-glucuronide in 0.2 M sodium acetate buffer (pH 5.0)-50 mM NaCl for BLG.
Solid phase binding assay
Purified BLG fractions eluted from ion-exchange HPLC (each 100 µL) were serially diluted with TBS, placed in the wells of an Immulon I plate (Dynatech Laboratories, Chantilly, VA), and immobilized for 4 h at room temperature. The binding with biotinyl glycoproteins or BP sugar probes (10 µg/mL in TBS) was demonstrated by ELISA using ABC-HRP as described previously (Ueda et al., 1999). Various concentrations of lipids in MeOH (100 µL/well) were dried at 37°C. After the wells were blocked with 3% bovine serum albumin (BSA) in TBS, 100 µL of biotinylated BLG (10 µg/mL) was added to each well, followed by incubation for 2 h. The wells were washed and measured by ELISA as described above.
Measurement of enzyme activity
Enzyme activities were measured in a test tube using a 50 µL aliquot of the sample. For BLG activity, 0.7 mL of 7 x 10–5 M p-nitrophenyl-ß-D-glucuronide in 0.2 M sodium acetate buffer (pH 5.0)–50 mM NaCl was added to the sample according to the method previously described (Harris et al., 1973). After incubation at 37°C for 1 h, 0.25 mL of 2 M glycine-NaOH (pH 10.4) was added to stop the reaction, and the liberated chromogen, p-nitrophenol, was measured at 400 nm. For N-acetyl ß-galactosaminidase and N-acetyl-ß-glucosaminidase, p-nitrophenyl N-acetyl-ß-galactosaminide and p-nitrophenyl-N-acetyl-ß-glucosaminide, respectively, were used as substrates according to the method described previously (Kawai and Anno, 1971). After adding 0.2 mL of 0.7 mM substrate dissolved in 0.1 M sodium acetate buffer (pH 5.0), the test tube was incubated at 37°C for 15 min. Then the reaction was stopped with 1 mL of 0.2 M Na2CO3. For ß-galactosidase, 0.3 mL of 2 mM p-nitrophenyl ß-galactoside in 0.05 M sodium citrate buffer (pH 4.0) was added to 30 µL of the sample and incubated at 37°C for 30 min (Li and Li, 1972). The reaction was stopped with 0.9 mL of 0.2 M Na2B4O7 (pH 9.8). Arylsulfatase activity was measured under the same conditions as BLG but using potassium p-nitrophenylsulfate as a substrate. The concentrations of p-nitrophenol were calculated from the absorbance at 400 nm by using free p-nitrophenol as a standard (
= 17,600 mol–1). The specific activity was defined as the amount in µmol of p-nitrophenol liberated per minute per mg of protein at 37°C. The protein concentration was determined using a DC Protein Assay Kit (Bio-Rad Laboratories, Hercules, CA) with BSA as the standard.
Kinetic studies were carried out by a microscale assay. The affinity-purified sample (12.5 µL, 10 µg/mL as protein) was added to 175 µL of 0.12 mM to 1.4 mM p-nitrophenyl-ß-D-glucuronide dissolved in 200 mM sodium acetate buffer–50 mM NaCl (pH 5.0), in the presence or absence of various saccharides (1 µM to 130 mM), in the wells of a plastic microtiter plate at 4°C and mixed well. After incubation for 1 h at 37°C, 62.5 µL of 2 M glycine-NaOH (pH 10.0) was added to each well and mixed, and the absorbance at 405 nm was measured using a microplate reader.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConflict of interest statementReferences |
|---|
Affinity chromatography of BLG on various carbohydrate-immobilized Sepharose gels
Binding of BLG with various carbohydrate-immobilized gels was studied by affinity chromatography using short columns of various adsorbents, and the activities of the eluted BLG and other enzymes were measured. As shown in Figure 1A, when BLG was bound to an immobilized saccharo-1,4-lactone column, a competitive inhibitor (Kurtin and Schwesinger, 1985
), 85% of the activity was eluted with buffer C, and the rest was eluted with buffer D. Surprisingly, N-acetyl-ß-D-glucosaminidase (ß-D-GlcNAcase) and N-acetyl ß-D-galactosaminidase (ß-D-GalNAcase) also bound to and were eluted from the saccharo-1,4-lactone column together with BLG by buffers C and D (Figure 1A). Similar chromatographic behavior of the enzymes was observed on a p-aminophenyl ß-glucuronide-immobilized column, a substrate for BLG, as shown in Figure 1B, and more than 60% of the activity was eluted with buffer C, together with the other two enzymes. The specific activity and increase in BLG activity and recovery of activities of each enzyme in the fractions eluted from each adsorbent are summarized in Table I. Because of the unexpected chromatographic behavior of the enzymes on these adsorbents, seven other kinds of carbohydrate adsorbents were compared for the ability to separate BLG.
|
), N-acetyl-ß-D-glucosaminidase (ß-D-GlcNAcase) (
), and N-acetyl ß-D-galactosaminidase (ß-D-GalNAcase) (
).
|
As shown in Figure 1C, lactamyl–Sepharose gave the best separation of BLG from ß-D-GlcNAcase and ß-D-GalNAcase. BLG bound to the lactamyl–Sepharose column, and 70% of the bound BLG was eluted by the buffer D. In the fractions eluted with buffer D (fraction D), activities of ß-D-GlcNAcase and ß-D-GalNAcase drastically decreased, and activities of ß-D-galactosidase and arylsulfatase were not detected (data not shown). Specific activity of BLG in fraction D increased by 20-fold (Table I). From the other adsorbents summarized in Table I, most BLG activity was eluted by buffer C together with ß-D-GlcNAcase and ß-D-GalNAcase presenting to various extents. Comparison of the ligand structures of lactamyl– and melibiamyl–Sepharose indicated that ß-linked-D-galactopyranosyl residue may contribute to the strong binding and separation of BLG from other enzymes. Based on these observations, the lactamyl–Sepharose column was used for further examination.
Binding characteristics of BLG to lactamyl-Sepharose
The characteristics of BLG binding to lactamyl–Sepharose were studied by microtube assays. As shown in Figure 2A, binding of BLG to lactamyl–Sepharose was maximum at pH 6 and decreased at acidic and basic pH. The amount of bound BLG at pH6 was twice as that at pH 7 in sodium acetate buffer. Based on this result, the binding procedure was hereafter performed at pH 6 (buffer A') instead of pH 7 (buffer A) to increase the binding capacity.
|
The abilities of various reagents to elute the bound BLG from lactamyl–Sepharose are summarized in Figure 2B. Buffer G and 5 mM EDTA eluted BLG almost quantitatively, indicating that electrostatic interaction contributes to the binding. Despite the effect of EDTA, addition of Ca2+ to the incubation buffer inhibited the binding (data not shown), suggesting that the binding is not Ca2+-dependent and that elution with EDTA may be because of removal of other metal cations or an electrostatic effect. Various sugars including 0.1 M lactose and galactose did not elute more than 20% of the bound BLG from lactamyl–Sepharose (Figure 2B). To examine whether the binding of BLG to lactamyl–Sepharose is a ligand-specific interaction, GHAG, a lactose derivative that is analogous to the ligand structure of lactamyl–Sepharose, was synthesized and studied for ability to elute BLG.
Structural analysis of GHAG and its BLG elution activity
The structure of GHAG was confirmed by NMR. As summarized in Table II, 1H- and 13C-chemical shifts and the coupling constants (data not shown) of the galactosyl residue were in good agreement with typical ones obtained in Me2SO-d6 (Iida-Tanaka and Ishizuka, 2000). The chemical shifts of the C-1 carbon in the glucosyl residue, however, did not resonate in the anomeric region, but at 52.7 p.p.m. in the region of the carbon bound by an amino group, -NH-. In addition to the observation of geminal H1 protons, H1a and b, the above result suggests that the aldehyde group in the glucosyl residue was reductively aminated to produce the structure of -CH2-NH-. Long-range connectivity of ethanol-C1/Glc-H1s and Glc-C1/ethanol-H1s was observed in the HMBC spectrum.
|
As shown in Figure 2B, 77% of the BLG activity was eluted from lactamyl–Sepharose with 25 mM GHAG, with small amounts of ß-galactosidase and ß-GlcNAcase, indicating that most of the BLG bound to lactamyl–Sepharose via the ligand-specific interaction and that GHAG eluted it more effectively than lactose and galactose owing to its structural similarity to the ligand of lactamyl–Sepharose.
Large-scale purification by affinity chromatography on a lactamyl–Sepharose column
The binding of BLG and subsequent washing of the column were performed with buffer A' at pH 6.0, and BLG was eluted by buffers at pH 5.0. As shown in Figure 3A, all the applied BLG bound to the column at pH 6.0, and a major part of the bound BLG activity was eluted with buffer C (10 mM sodium acetate–5 mM NaCl [pH 5.0]) and buffer E' (25 mM sodium acetate–25 mM NaCl [pH 5.0]), at slightly lower salt concentrations than those required for elution when the binding was performed at pH 7.0 (cf. Figure 1C).The specific activity of the buffer C-eluted fraction (fraction C) showed a 20 times increase, and the contaminating ß-galactosidase and ß-D-GalNAcase markedly decreased. Increasing the ion concentration to 200 mM sodium acetate buffer–200 mM NaCl (pH5.0) (buffer G) did not elute any proteins (data not shown). As shown in Figure 3B, GHAG eluted BLG as a single peak from a lactamyl–Sepharose affinity chromatography column.
|
The buffer C- and GHAG-eluted fractions showed similar protein patterns on SDS–PAGE (Figure 3C). Zymography of both the fractions indicated the presence of an active form of BLG at the migration position corresponding to 160 kDa, whereas the 78-kDa subunit monomer was shown to be inactive (Figure 3C, lanes e–f). The reported molecular size of the active BLG in solution, 290 kDa, on sucrose density gradient centrifugation was considered to be a tetramer of the 78-kDa subunit (Himeno et al., 1974
). The reason for the discrepancy between the reported and observed molecular mass is yet unclear, but the possibility that the 160-kDa band on SDS–PAGE is a tetramer cannot be denied. Another possibility is that the 160-kDa band corresponded to a dimer whose activity was recovered after removing SDS. In this case, whether active BLG forms a tetramer in gel or remains a dimer is unknown. As shown in Figure 3C, the buffer C eluted fraction more of the active form of BLG than of the monomer, whereas the monomer is predominant in the GHAG-eluted fraction. The BLG activity in fraction E' in Figure 3A was lost after PAGE (data not shown) possibly because a factor that stabilizes the active form may have been removed in fraction E' and the activity could not be restored after PAGE. Based on these observations, fraction C was subjected to further purification because it still contained a few other protein bands. BLG purchased from Sigma gave essentially the same elution profile on lactamyl–Sepharose column and protein bands on SDS–PAGE as those of BLG obtained from Worthington (data not shown).
Ion exchange HPLC of BLG on DEAE-5PW
Fraction C was separated into five fractions by NaCl gradient on ion-exchange HPLC on a DEAE-5PW column, as shown in Figure 4A. Peak b exhibited ß-glucuronidase activity with trace activity of other enzymes (Figure 4B), and a single protein band corresponding to the monomeric BLG on SDS–PAGE under reducing conditions (Figure 4C). ß-Galactosidase activity was detected mainly in peak c. In the course of the study, gel filtration HPLC was also tried for further purification of fraction C. In that case, ß-galactosidase was separated from the BLG peak, but ß-glucosaminidase and ß-galactosaminidase activities remained in the BLG peak (data not shown).
|
), ß-D-galactosidase (
), or N-acetyl-ß-D- glucosaminidase (ß-D-GlcNAcase) activities (
). (C) Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) of peak b under reducing conditions.
Effect of free saccharides on BLG activity
BLG activity was measured using fraction C from lactamyl affinity chromatography as BLG in the presence or absence of various saccharides (1 µM to 130 mM) at 0.12 mM to 1.4 mM substrate concentrations. The results are summarized in Table III, and examples of the double reciprocal of Lineweaver–Burk plots are shown in Figure 5. According to the Lineweaver–Burk plot analyses, BLG is dependent on the lactose concentration, as shown in Figure 5A. At 1 mM lactose, the interception of the x-axis (1/Km) of the plot was unchanged from that of the control, indicating that lactose binds to both the free enzyme and the enzyme–substrate complex. The result suggests that the lactose-binding site of BLG is different from the catalytic site. On the contrary, at lactose concentrations higher than 13 mM, the plot showed a tendency toward competitive inhibition with the interception with the y-axis (1/Vmax) unchanged. In contrast, saccharo-1,4-lactone (1 µM) and D-glucuronic acid (1 mM) showed a typical competitive inhibition, as shown in Figure 5B. Cellobiose showed a very weak tendency toward noncompetitive activation. On the contrary, other tested saccharides, maltose (Figure 5C), D-glucose and D-galactose mannose-6-phosphate, N-acetylglucosamine, and N-acetylneuraminic acid (1 mM), did not have a significant or integrative effect on the enzyme activity, although BLG bound to the affinity adsorbents of some of these saccharides.
|
|
) 1 mM, (
) 13 mM, and (
) 130 mM; . (B) (
) 1 µM Saccharo-1,4-lactone, or (
) 1 mM D-glucuronic acid; (C) D-maltose at (
) 10 mM, (*) 100 mM, and (
Binding studies of BLG with glycoproteins and lipids
The binding activities of purified BLG to biotinyl glycoprobes were studied at pH 7.5 and pH 5.0 by solid-phase binding assay. At pH 5, BLG did not bind to any of tested glycoproteins and lipids (data not shown). As shown in Figure 6A, immobilized BLG bound to asialofetuin better than fetuin and to asialoagalactofetuin but not to transferrin, asialotransferrin, ovalbumin, or BSM at pH7.5. It suggested that BLG recognized the exposed galactose residues of triantennary complex-type N-glycans but not the biantennary complex-type of asialotransferrin or the high Man or hybrid type of ovalbumin. BLG bound to lactosyl ceramide but not to Glc-cer, Gal-cer, ceramide, or sulfatide (Figure 6B). The direct binding of BLG to the sugar residues was demonstrated using sugar-BP probes. As shown in Figure 6C), BLG bound to N-acetyllactosamine-BP better than to the Lac-BP probe. Taken together, BLG was shown to bind to the N-acetyllactosamine or lactosyl sequence of glycoconjugates at neutral pH but not at pH 5.
|
), asialofetuin (
), asialogalactofetuin (
), transferrin (
), asialotransferrin (
), ovalbumin (
) and BSM (
). (B) Binding assay of biotinylated BLG to immobilized glycolipids. Lactosyl-ceramide (
), glucosyl-ceramide ( |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConflict of interest statementReferences |
|---|
In this study, BLG was demonstrated for the first time to have binding activity toward lactose and N-acetyllactosamine sequences. BLG was effectively separated from contaminating glycosidases by affinity chromatography on a lactamyl–Sepharose column, and the specific activity was increased by 20-fold during one-step affinity chromatography (Table I). Non- or uncompetitive regulation of BLG with lactose indicated that the lactose-binding site is different from the substrate-binding site (Table III). Purified BLG bound best to the glycoconjugates possessing a nonreducing terminal N-acetyllactosamine/lactose such as asialofetuin and lactosylceramide (Figure 6), which is attributable to the carbohydrate-binding activity of BLG toward lactose/lactosamine structures.
Preparation of BLG has required multiple purification steps including heat denaturation of proteins (Ho, 1991), or fractionation with organic solvents (Himeno et al., 1974), in combination with several chromatography steps to dissociate complex of BLG with other lysosomal proteins. This study provided convenient protocol to isolate BLG from other contaminating enzymes under mild conditions and furthermore opens new insights into the biological functions of the carbohydrate-specific interaction of ß-glucuronidase. The observations that BLG was eluted more effectively with GHAG than lactose and galactose from lactamyl–Sepharose column (Figure 2B), and BLG bound to N-acetyllactosamine-BP better than to the Lac-BP probe (Figure 6C) suggest the significance of N-acetyllactosamine structure present in the glycoconjugates as the biological ligand for BLG.
Interaction of BLG with carbohydrate ligands in lysosome
Although the lactamyl-binding activity of BLG was maximal at pH 6–7 and weakened to one-third at pH 5, the physiological pH of lysosomes (Figure 2), lactose noncompetitively activated BLG at 1 mM at pH 5, indicating that the carbohydrate-binding activity is exhibited even at pH 5. Therefore, the lactose binding may contribute to regulation of the enzyme activity in lysosomes. As a candidate for lysosomal ligands other than free lactose, BLG is supposed to interact with lysosome-associated membrane glycoproteins 1 and 2 (Lamp1 and 2), major carriers for poly-N-acetyllactosamines (Laferte and Dennis, 1989), because the repeating N-acetyllactosamine sequence may enhance the affinity for BLG by a multivalent effect. However, we did not detect the binding of BLG to polylactosaminoglycans of human erythrocyte Band 3 glycoprotein (Fukuda et al., 1984) at pH 5 (data not shown) by solid phase assay, while BLG did bind to it at pH 7 (Figure 6 and our unpublished data). Whether interaction between BLG and the glycoconjugates in lysosomes is possible is unknown at this point.
Biological function of the carbohydrate binding of BLG
Alternatively, the lactose/N-acetyllactosamine-binding activity may play a specific role at neutral pH in the endoplasmic reticulum (ER) and Golgi apparatus during glycoprotein maturation and in extracellular matrices after secretion. One possibility is that the lactose/N-acetyllactosamine-binding activity of BLG may be involved in the formation of the active tetrameric form because BLG contains a considerable amount of complex-type asialoglycans (Himeno et al., 1974), and a wild-type BLG produced in the presence of tunicamycin was inactive (Shipley et al., 1993).
In the ER and Golgi apparatus, various glycoconjugates are involved in the biosynthetic and lysosome-sorting processes of BLG. For example, phosphodiester 
-GlcNAcase, which catalyzes the second step of attachment of the Man6P signal on BLG, is one of the N-glycosylated glycoproteins. The active site and the recognition motif of phosphotransferase have been elucidated on human ß-glucuronidase (Jain et al., 1996), but recognition motif of -GlcNAcase has not yet been clarified. Because only a limited number of the -GlcNAc phosphodiesters that are attached by GlcNAc-1-phosphotransferase are hydrolyzed by -GlcNAcase to generate Man6P monoester (Natowicz et al., 1982), the carbohydrate-binding activity of BLG could play a role in accessing phosphodiester -GlcNAcase.
A carbohydrate-binding activity of a macromolecule-degrading enzyme might help localize the enzyme on an appropriate scaffold to exhibit catalytic action efficiently and stably in vivo. Lysosomal enzymes, including BLG, are released by the fusion of whole lysosomes with the plasma membrane into the synovial fluid in inflammatory joint diseases and in the invasion by metastatic tumor cells to focal dissolution of the extracellular matrix of surrounding tissues or penetration of endothelial membranes. (Sloane et al., 1986; Andrei et al., 2004). The observation that highly metastasizing tumor cells express more poly-N-acetyllactosamine in lysosomes than do their normal and poorly metastasizing counterparts (Dennis et al., 1999, Chakraborty and Pawelek, 2003) supports the idea that the interaction of BLG with poly-N-acetyllactosamines may well be involved in concentrating the hydrolases and catalyzing the substrate hydrolysis efficiently at the cell surface when secreted outside the cell. A benefit for BLG of anchoring to poly-N-acetyllactosamines is cooperation with other hydrolases because the carbohydrate-binding activity is shared among several other lysosomal exoglycosidases, as shown in this study. ß-GalNAcase and ß-GlcNAcase exhibited considerable binding activity to affinity adsorbents immobilized with saccharides other than their substrates, such as galactose and triN-acetylchitotriose (Table I). Several extracellular matrix glycoproteins that have polylactosaminoglycans, such as laminin, integrin, and neuronal glycoproteins, could provide a scaffold for lysosomal glycosidases secreted from tumor cells, so that the enzymes can act on substrates cooperatively. Such a hypothetical tie-up of glycan-degrading exoglycosidases on the poly-N-acetyllactosaminoglycan chain will increase the efficiency of the degradation of their common substrates. Those possibilities are under investigation in our laboratory.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConflict of interest statementReferences |
|---|
None declared.
|
Footnotes |
|---|
1 These two authors contributed equally to this article.
|
Abbreviations |
|---|
BLG, bovine liver ß-glucuronidase; BP, biotinyl polymer; GHAG, 1-deoxy-4-o-ß-D-galactopyranosyl-1-[(2-hydroxyethyl)amino]-D-glucitol; HPLC, high-performance liquid chromatography; Me2SO-d6, 2[CH3]2SO; PAGE, polyacrylamide gel electrophoresis; SDS, sodium dodecyl sulfate; TBS, 10 mM Tris–HCl–150 mM NaCl (pH 7.5); X-, 5-bromo-4-chloro-3-indolyl-; ß-D-GalNAcase, N-acetyl ß-D-galactosaminidase; ß-D-GlcNAcase, N-acetyl-ß-D-glucosaminidase
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConflict of interest statementReferences |
|---|
Andrei, C., Margiocco, P., Poggi, A., Lotti, L.V., Torrisi, M.R., and Rubartelli, A.T. (2004) Phospholipases C and A2 control lysosome-mediated IL-1 beta secretion: Implications for inflammatory processes. Proc. Natl. Acad. Sci. U. S. A., 101, 9745–9750.
Chakraborty, A.K. and Pawelek, J.M. (2003) GnT-V, macrophage and cancer metastasis: a common link. Clin. Exp. Metastasis, 20, 365–373.[CrossRef][Web of Science][Medline]
Davis, B.J. (1964) Disc electrophoresis. II. Method and application to human serum proteins. Ann. N. Y. Acad. Sci., 121, 404–427.[Web of Science][Medline]
Dennis, J.W., Granovsky, M., and Warren, C.E. (1999) Glycoprotein glycosylation and cancer progression. Biochim. Biophys. Acta, 1473, 21–34.[Medline]
Fukuda, M., Dell, A. and Fukuda, M.N. (1984) Structure of fetal lactosaminoglycan. The carbohydrate moiety of Band 3 isolated from human umbilical cord erythrocytes. J. Biol. Chem. 259, 4782–4791.
Harris, R.G., Rowe, J.J.M., Stewart, P.S., and Williams, D.C. (1973) Affinity chromatography of ß-glucuronidase. FEBS Lett., 29, 189–192.[CrossRef][Web of Science][Medline]
Himeno, M., Hashiguchi, Y., and Kato, K. (1974) ß-Glucuronidase of bovine liver. Purification, properties, carbohydrate composition. J. Biochem., 76, 1243–1252.
Ho, K.J. (1991) A large-scale purification of ß-Glucuronidase from human liver by immunoaffinity chromatography. Biotechnol. Appl. Biochem., 14, 296–305.
Holmes, E.W. and O’Brien, J.S. (1979) Separation of glycoprotein-derived oligosaccharides by thin-layer chromatography. Anal. Biochem., 93, 167–170.[Web of Science][Medline]
Iida-Tanaka, N., Hikita, T., Hakomori, S., and Ishizuka, I. (2002) Conformational studies of a novel cationic glycolipid, glyceroplasmalopsychosine, from bovine brain by NMR spectroscopy. Carbohydr Res., 337, 1775–1779.
Iida-Tanaka, N. and Ishizuka, I. (2000) Complete 1H and 13C NMR assignment of mono-sulfated galactosylceramides with four types of ceramides from human kidney. Carbohydr Res., 324, 218–222.[CrossRef][Web of Science][Medline]
Ito, Y., Seno, N., and Matsumoto, I. (1985) Immobilization of protein ligands on new formyl-spacer-carriers for the preparation of stable and high capacity affinity adsorbents. J. Biochem., 97, 1689–1694.
Jain, S., Drendel, W.B., Chen, Z.-w., Mathews, F.S., Sly, W.S., and Grubb, J.H. (1996) Structure of human ß-glucuronidase reveals candidate lysosomal targeting and active-site motifs. Nat. Struct. Biol., 3, 375–381.[CrossRef][Web of Science][Medline]
Kaplan, A. and Achord, D.T. (1977) Phosphohexosyl components of a lysosomal enzyme are recognized by pinocytosis receptors on human fibroblasts. Proc. Natl. Acad. Sci. U. S. A., 74, 2026–2030.
Kawai, Y. and Anno, K. (1971) Mucopolysaccharide-degrading enzymes from the liver of the squid, Ommastrephes sloani pacificus. I. Hyaluronidase. Biochem. Biophys. Acta, 242, 428–436.[Medline]
Kornfeld, S. (1987) Trafficking of lysosomal enzymes. FASEB J., 1, 462–468.[Abstract]
Kurtin, W.E. and Schwesinger, W.H. (1985) Assay of ß-glucuronidase in bile following ion-pair extraction of pigments and bile acids. Anal. Biochem., 147, 511–516.[CrossRef][Web of Science][Medline]
Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685.[CrossRef][Medline]
Laferte, S. and Dennis, J.W. (1989) Purification of two glycoproteins expressing ß1–6 branched Asn-linked oligosaccharides from metastatic tumour cells. Biochem. J., 259, 569–576.[Web of Science][Medline]
Li, Y.T. and Li, S.C. (1972) -Mannosidase, ß-N-acetylhexosaminidase, and ß-galactosidase from jack bean meal. Meth. Enzymol., 28, 702–713.
Matsumoto, I., Kitagaki, H., Akai, Y., Ito, Y., and Seno, N. (1981) Derivatization of epoxy-activated agrose with various carbohydrates for the preparation of stable and high-capacity affinity adsorbents. Anal. Biochem., 116, 103–110.[CrossRef][Web of Science][Medline]
Natowicz, M., Baenziger, J.U., and Sly, W.S. (1982) Structural studies of the phosphorylated high mannose-type oligosaccharides on human ß-glucuronidase. J. Biol. Chem., 257, 4412–4420.
Paigen, K. (1989) Mammalian ß-glucuronidase: genetics, molecular biology, and cell biology. Progr. Nucl Acid Res. Molec. Biol., 37, 155–205.
Shipley, J.M., Grubb, J.H., and Sly, W.S. (1993) The role of glycosylation and phosphorylation in the expression of active human ß-glucuronidase. J. Biol. Chem., 268, 12193–12198.
Sloane, B.F., Rozhin, J., Johnson, K., Taylor, H., Crissman, J.D., and Honn, K.V. (1986) Cathepsin B: association with plasma membrane in metastatic tumors. Proc. Natl. Acad. Sci. U. S. A., 83, 2483–2487.
Ueda, H., Saitoh, T., Kojima, K., and Ogawa, H. (1999) Multi-specificity of a Psathyrella velutina mushroom lectin: heparin/pectin binding occurs at a site different from the N-acetylglucosamine/N-acetylneuraminic acid-specific site. J. Biochem., 126, 530–537.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||