Identification of Lacto-N-Biose I Phosphorylase from Vibrio vulnificus CMCP6

Abstract

A β-1,3-galactosyl-N-acetylhexosamine phosphorylase (GalGlyNAcP) homolog gene was cloned from Vibrio vulnificus CMCP6. In synthetic reactions, the recombinant enzyme acted only with GlcNAc and GalNAc as acceptors in the presence of α-d-galactose-1-phosphate as a donor to form lacto-N-biose I (LNB) (Galβ1 → 3GlcNAc) and galacto-N-biose (GNB) (Galβ1 → 3GalNAc), respectively. GlcNAc was a much better acceptor than GalNAc. The enzyme also phosphorolysed LNB faster than it phosphorolysed GNB, and the k_cat/K_m for LNB was approximately 60 times higher than the k_cat/K_m for GNB. This result indicated that the enzyme was remarkably different from GalGlyNAcP from Bifidobacterium longum, which has similar activities with LNB and GNB, and GalGlyNAcP from Clostridium perfringens, which is a GNB-specific enzyme. The enzyme is the first LNB-specific enzyme that has been found and was designated lacto-N-biose I phosphorylase. The discovery of an LNB-specific GalGlyNAcP resulted in recategorization of bifidobacterial GalGlyNAcPs as galacto-N-biose/lacto-N-biose I phosphorylases.

β-1,3-Galactosyl-N-acetylhexosamine phosphorylase (GalGlyNAcP) (EC 2.4.1.211) is an enzyme which phosphorolyses lacto-N-biose I (LNB) and galacto-N-biose (GNB) reversibly to produce α-d-galactose 1-phosphate (Gal 1-P) with GlcNAc and GalNAc. This enzyme was first found in a cell extract of Bifidobacterium bifidum by Derensy-Dron et al. (5). We have cloned genes from Bifidobacterium longum JCM1217 (lnpA) (19) and B. bifidum JCM1254 (lnpA1 and lnpA2) (31) that encode GalGlyNAcPs, one of which was found to be identical to the enzyme isolated from a cell extract of B. bifidum (31). Although the bifidobacterial GalGlyNAcPs showed similar activities with GNB and LNB (5, 19, 31), we gave the short name lacto-N-biose I phosphorylase (LNBP) to the enzyme (19). This enzyme plays a role in the growth of bifidobacteria in the intestine by utilizing GNB liberated from mucin sugars and LNB processed from human milk oligosaccharides (19, 32).

The amino acid sequences of GalGlyNAcPs did not exhibit significant identity to any proteins with known functions. Several homologous genes have been found in the genomes of bacteria which are pathogens or commensals in humans (19). Therefore, the gene products may play roles similar to the roles of the enzymes in bifidobacteria, helping the bacteria grow by metabolizing sugars. Recently, these enzymes have been classified in a new family, glycoside hydrolase family 112, in the Carbohydrate Active Enzymes database (http://www.cazy.org/) (4, 9). Although bifidobacteria are considered favorable for human health, the other kinds of bacteria are not. Thus, it is important to characterize these enzymes in the pathogenic bacteria in order to understand their catalytic activities because the enzymes might support the growth of the bacteria in human tissue. Recently, we expressed the homologous gene from Clostridium perfringens, some strains of which are known to cause food poisoning, gas gangrene, and septicemia, in Escherichia coli and found that the gene product (GalGlyNAcP_Cp) was a GalGlyNAcP that preferred GNB much more than LNB (27). We named the enzyme galacto-N-biose phosphorylase (GNBP) due to its specificity.

Vibrio vulnificus is a gram-negative gammaproteobacterium that is often detected in seawater, fish, plankton, and shellfish. Because V. vulnificus does not grow at lower temperatures in the sea and grows well at higher temperatures, infection increases in the summer (14, 26). Infection occurs through consumption of contaminated seafood and through exposure of wounds to seawater or seafood products. Regardless of the infection route, the infection causes severe symptoms and high mortality (10). Patients with cirrhosis, diabetes, or immunodeficiency syndrome are especially susceptible to infection by V. vulnificus (10). The genome of V. vulnificus CPCM6 was recently sequenced (16), as was that of V. vulnificus YJ016 (2). Genes homologous to GalGlyNAcP genes are present in these genomes. In this study, we expressed the homologous gene (vv2_1091 gene) from V. vulnificus CMCP6 in E. coli and characterized the gene product (GalGlyNAcP_Vv) in order to understand its roles in V. vulnificus.

MATERIALS AND METHODS

Cloning, expression, and purification.

pUC57 plasmid DNA with the synthesized vv2_1091 gene (GenBank accession number AAO07997) inserted was commercially prepared by GenScript Corporation (Piscataway, NJ). The gene was amplified by PCR using KOD plus DNA polymerase (Toyobo, Osaka, Japan), the custom plasmid pUC57 as the template, and a pair of primers (forward primer 5′-GGCCAGTGAACATATGACCCAATTTGCCTCTCGTAAAGGTCACATGACCC-3′ and reverse primer 5′-CCCGGGACTCGAGGTTATCGAACCAAC-3′ containing NdeI and XhoI sites [underlined], respectively). A silent mutation (bold type) was introduced to eliminate the original NdeI site in the gene. The amplified product was digested with NdeI and XhoI and was inserted into the corresponding sites of pET30a(+) (Novagen, Madison, WI) so that it contained a His tag at the C terminus. E. coli BL21(DE3) cells containing the constructed plasmid were grown in LB medium containing 100 μg/ml of kanamycin until the absorbance at 660 nm at 37°C was 0.8. The gene was expressed by adding 0.1 mM isopropyl-β-d-thiogalactopyranoside, followed by incubation for 3 h at 30°C. After the cells were collected by centrifugation at 5,000 × g, they were suspended in 20 mM sodium phosphate (pH 7.3) containing 500 mM NaCl (buffer A). The suspended cells were sonicated and centrifuged at 15,000 × g for 15 min to extract the proteins. The supernatant was loaded onto an Ni-nitrilotriacetic acid agarose (Qiagen, Hilden, Germany) column equilibrated with buffer A. The column was washed with buffer A containing 20 mM imidazole, and the recombinant protein was eluted with a linear 20 to 250 mM imidazole gradient. The fractions containing the recombinant protein were collected and dialyzed against 20 mM morpholinepropanesulfonic acid (MOPS) (pH 7.0) buffer. The protein was concentrated using an Amicon Ultra 10,000-molecular-weight-cutoff membrane (Millipore, Billerica, MA).

Measurement of the phosphorolytic activity.

Phosphorolytic activity was determined by quantifying the Gal 1-P produced from LNB or GNB by the method of Nihira et al. (30). The reaction mixtures (500 μl) containing 50 mM Tris-HCl (pH 7.5), 10 mM sodium-potassium phosphate buffer, 10 mM LNB or GNB, and enzyme were incubated at 30°C. At 1-min intervals, 70 μl of a reaction mixture was incubated at 75°C for 10 min to stop the enzymatic reaction. Samples were cooled on ice and mixed with an equal volume of a coloring reagent containing 100 mM Tris-HCl (pH 7.5), 10 mM MgCl₂, 10 μM glucose 1,6-bisphosphate, 1 mM UDP-Glc (Wako, Osaka, Japan), 0.5 mM thio-NAD⁺ (Oriental Yeast, Tokyo, Japan), 0.05 U/ml GalT from B. longum, 1 U/ml phosphoglucomutase (Oriental Yeast), and 1 U/ml glucose-6-phosphate dehydrogenase (Oriental Yeast) (30), and this was followed by incubation at 37°C for 30 min. The Gal 1-P produced was quantified by measuring the absorbance at 400 nm derived from thio-NADH (ɛ₃₉₈ = 11,700 cm⁻¹ M⁻¹) (30). One unit of activity was defined as the amount of enzyme that generated 1 μmol Gal 1-P under the conditions described above.

Measurement of the synthetic activity.

Synthetic activity was determined by measuring the increase in the phosphate content of a reaction mixture containing 5 mM Gal 1-P and 10 mM acceptor in 100 mM MOPS buffer (pH 7.0) at 30°C by the method of Lowry and Lopez (23), as described below. The substrate solution (142.5 μl; 5.26 mM Gal 1-P and 10.53 mM acceptor in 105 mM MOPS buffer [pH 7.0]) was preincubated at 30°C for 10 min and then mixed with 7.5 μl of an enzyme solution kept at room temperature (25°C) to start the reaction. Aliquots (12.5 μl) were taken at 3-min intervals and added to 100 μl of 0.2 M sodium acetate (pH 4.0) to stop the enzymatic reaction. Then 12.5 μl of 1% ammonium molybdate containing 25 mM sulfuric acid and 12.5 μl of 1% ascorbic acid containing 0.05% potassium bisulfate were mixed with the samples. The mixtures were incubated at 37°C for 1 h, and the absorbance at 700 nm was determined.

Basic properties.

Temperature stability and pH stability were determined by measuring residual activity after incubation of the enzyme (36 μg/ml) in 100 mM MOPS (pH 7.0) at various temperatures for 30 min and in various 100 mM buffers at 30°C for 30 min, respectively. The residual activities were determined by measuring the synthetic activities using GlcNAc as the acceptor. The activities at various pHs were measured by determining the synthetic activities with GlcNAc as the acceptor when the MOPS buffer was replaced by various buffers. The activities at various temperatures were measured by performing the synthetic reaction with GlcNAc as the acceptor using different preincubation and reaction temperatures. In each case, an aliquot was taken every 1 min for 5 min.

Kinetic analysis.

A kinetic analysis of the phosphorolytic reaction was performed by using the continuous Gal 1-P assay (30). Reaction mixtures (200 μl) were prepared in wells of a 96-well microtiter plate by adding various concentrations of LNB or GNB, various concentrations of phosphate, and 80 ng/ml (for LNB) or 2.4 μg/ml (for GNB) of GalGlyNAcP_Vv to the half-diluted coloring reagent for the Gal 1-P assay with 100 mM Tris HCl buffer (pH 7.5). The reaction was carried out in a temperature-controlled microplate reader (Sunrise Rainbow Thermo; Tecan, Männedorf, Switzerland) at 30°C, and the absorbance at 400 nm was monitored at 1-min intervals. The kinetic parameters were calculated by curve fitting the experimental data with the theoretical equation using Grafit version 4 (Erithacus Software, Middlesex, United Kingdom).

RESULTS

Sequence analysis.

The deduced amino acid sequence of GalGlyNAcP_Vv (728 amino acids, 83,582 Da) has no N-terminal signal sequence according to SignalP 3.0 Server (http://www.cbs.dtu.dk/services/SignalP) (1, 29). The amino acid sequence of GalGlyNAcP_Vv exhibits 38% and 39% identity to the sequences of GalGlyNAcP from B. longum JCM1217 (GalGlyNAcP_Bl) and GalGlyNAcP from C. perfringens ATCC 13124 (GalGlyNAcP_Cp), respectively. The level of amino acid sequence identity between GalGlyNAcP_Vv and the homologous protein from another strain of V. vulnificus (strain YJ016) was 98.8%.

Basic properties of recombinant GalGlyNAcP_Vv.

Purification of GalGlyNAcP_Vv expressed in E. coli was carried out by using Ni-nitrilotriacetic acid agarose affinity chromatography to obtain approximately 1.3 mg of protein from 150 ml of a culture. A single band was detected at 80 kDa on a sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel, which corresponded to the calculated molecular mass of GalGlyNAcP_Vv with a six-His tag. Of the sugars tested, GalGlyNAcP_Vv acted only with GlcNAc and GalNAc as acceptors with Gal 1-P as the donor. GlcNAc was seven times more effective than GalNAc as an acceptor at a concentration of 10 mM (Table 1). GalGlyNAcP_Vv also phosphorolysed LNB much faster than it phosphorolysed GNB (Table 2). GalGlyNAcP_Vv was stable at pH 6.5 to 9.0, and the optimum pH was 6.5 to 7.5 (Fig. 1A). The neutral optimum pH was similar to the values determined for GalGlyNAcP_Bl and GalGlyNAcP_Cp, both of which are cytosolic enzymes. It was stable at temperatures up to 30°C during incubation for 30 min (Fig. 1B). The reactions proceeded linearly for 5 min at temperatures up to 37°C but did not proceed linearly at 45°C (Fig. 1C). No reaction was observed at 55°C. The time course of the reaction at 45°C was curve fitted with the following equation, in which thermal inactivation of the enzyme was considered: [P_i] = A₀·[1 − exp(−k_D·t)]/k_D, where t is the reaction time, A₀ is the initial activity, and k_D is the thermal inactivation rate.

TABLE 1.

Substrate specificity of GalGlyNAcP_Vv

Substratea	v/E0 (s−1)
d-Glucose	NDb
d-Galactose	ND
N-Acetyl-d-glucosamine	35
N-Acetyl-d-galactosamine	5.1
d-Glucosamine	ND
d-Fructose	ND
2-Deoxy-d-glucose	ND
d-Mannose	ND
Maltose	ND
Sucrose	ND

^aFor the reaction, 5 mM Gal 1-P and 10 mM substrate were used.

^bND, not detected.

TABLE 2.

Kinetic parameters for reactions of GalGlyNAcP_Vv with LNB and GNB^a

Substrate	kcat (s−1)	Km(A) (mM)	Km(B) (mM)	Ki(A) (mM)	kcat/Km(A) (mM−1 s−1)
LNB	45 ± 1.6	3.6 ± 0.34	1.3 ± 0.24	22 ± 4.3	13
GNB	2.2 ± 0.14	10 ± 1.2	0.064 ± 0.019	1,400 ± 420	0.22

^aThe values were obtained by regressing the data with the following equation using Grafit version 4.0.10: v = k_cat[E₀][A][B]/{K_i(A)K_m(B) + K_m(A)[B] + K_m(B)[A] + [A][B]}, where A is GNB or LNB and B is P_i.

FIG. 1.

Basic properties of GalGlyNAcP_Vv. (A) Stability (open symbols and dashed lines) and activity (filled symbols and solid lines) at different pHs. The buffers used were sodium acetate (circles), morpholineethanesulfonic acid (MES)-NaOH (triangles), MOPS-NaOH (squares), and Tris-HCl (diamonds). The activity in MOPS buffer (pH 7.0) without pretreatment was defined as 100%. (B) Thermal stability. The activity without pretreatment was defined as 100%. (C) Time course of reactions at various temperatures, including 4°C (filled triangles), 10°C (open triangles), 20°C (filled squares), 30°C (open squares), 37°C (filled circles), 45°C (open circles), and 55°C (filled diamonds).

The theoretical curve agreed well with the experimental data. The k_D was calculated to be 0.25 min⁻¹, meaning that 29% of the activity remained after incubation for 5 min (Fig. 1C). However, the percentage of activity remaining after incubation at 45°C for 5 min in the absence of substrates was found to be less than 5%, suggesting that substrates stabilized the enzyme. Next the activity remaining after incubation at 45°C for 5 min in the presence of each substrate was examined. No stabilization effect was observed in the presence of 10 mM GlcNAc or 5 mM LNB, but 27% residual activity was observed in the presence of 5 mM Gal 1-P, suggesting that the enzyme was stabilized in the presence of Gal 1-P.

Kinetic analysis of the phosphorolytic reactions.

Both double-reciprocal plots of the initial velocity versus the initial concentration of LNB or GNB contained a series of lines intersecting at a point in the second quadrant (Fig. 2). These results indicate that the phosphorolytic reaction of GalGlyNAcP_Vv followed the sequential bi-bi mechanism, like the reactions of inverting phosphorylases, such as GalGlyNAcP from B. bifidum (5), GalGlyNAcP_Bl (27), GalGlyNAcP_Cp (27), maltose phosphorylase (39), cellobiose phosphorylase (15, 17, 28, 34), chitobiose phosphorylase (12), and laminaribiose phosphorylase (18).

FIG. 2.

Double-reciprocal plot of the phosphorolytic reaction with different concentrations of inorganic phosphate. (A) LNB. (B) GNB. Symbols: open circles, 1 mM P_i; filled circles, 2 mM P_i; open triangles, 4 mM P_i; filled triangles, 6 mM P_i; open squares, 10 mM P_i.

The kinetic parameters of the phosphorolytic reaction are shown in Table 2. The k_cat for the phosphorolysis of LNB was approximately 20 times higher than the k_cat for the phosphorolysis of GNB. The k_cat/K_m for LNB was approximately 60 times higher than the k_cat/K_m for GNB. These parameters clearly indicate that GalGlyNAcP_Vv is specific for LNB.

DISCUSSION

Role of GalGlyNAcP_Vv in V. vulnificus.

Generally, Vibrio species have two circular chromosomes. A large chromosome contains the genes essential for survival, pathogenicity, and proliferation, and the smaller chromosome is assumed to play a role in responses to environmental change and to provide the bacterium with an evolutionary advantage (24). While the genes around the vv2_1091 gene constitute a gene cluster (vv2_1090 to vv2_1096) in the small chromosome of V. vulnificus (Fig. 3A), this gene cluster is not present in the genomes of other pathogenic and nonpathogenic Vibrio strains whose genomic sequences are available, including strains of Vibrio cholerae (7), Vibrio parahaemolyticus (24), and Vibrio fischeri (36). V. cholerae and V. parahaemolyticus grow in the human intestinal tract and cause food poisoning (6, 25). On the other hand, V. vulnificus causes not only food poisoning but also septicemia by invading and growing in blood vessels. Therefore, the gene cluster may be characteristic of V. vulnificus.

FIG. 3.

Role of GalGlyNAcP_Vv in V. vulnificus. (A) Gene cluster containing the vv_1091 gene in the genomic sequence of V. vulnificus CMCP6. The loci involved in LNB metabolism are indicated by bold type. (B) Predicted metabolic pathway for LNB in V. vulnificus. Proteins encoded by the genes in the gene cluster shown in panel A are indicated by bold type. One-step and multistep reactions in the metabolism are indicated by solid and dotted arrows, respectively. The question mark indicates a step not predicted by the annotations.

A homology search of the VV2_1090 and VV2_1092 proteins revealed that they belong to polysaccharide lyase family 8 (PL8) (4) and glycoside hydrolase family 88 (GH88) (8), respectively. PL8 includes enzymes that are involved in cleavage of acidic polysaccharides to form 4,5-unsaturated β-glucuronide at the nonreducing ends of the products (22). GH88 contains 4,5-unsaturated β-glucuronyl hydrolase, which releases 4,5-unsaturated β-glucuronic acid from the products of polysaccharide lyase enzymes (13). The catalytic residues of PL8 and GH88 enzymes are conserved in the VV2_1090 and VV2_1092 proteins, respectively (13, 22). The VV2_1093 protein was annotated as 2-deoxy-gluconate-3-dehydrogenase (EC 1.1.1.125), which plays a role in the metabolism of 4,5-unsaturated glucuronic acid (3). The VV2_1094 and VV2_1095 proteins are annotated as UDP-glucose-galactose-1-phosphate uridylyltransferase (GalT; EC 2.7.7.12) and UDP-glucose 4-epimerase (GalE; EC 5.1.3.2), both of which are involved in the Leloir pathway metabolizing Gal 1-P (21). These proteins have 66% amino acid sequence identity with E. coli GalT (GenBank accession number AAC73845) (42) and 54% amino acid sequence identity with B. longum JCM1217 GalE (GenBank accession number AB303839) (32). The VV2_1096 protein is annotated as a sulfate permease. Based on the annotations, the gene cluster described above is likely to contain genes encoding enzymes that metabolize LNB (vv2_1091, vv2_1094, and vv2_1095) and some acidic polysaccharides, such as hyaluronan and chondroitin (vv2_1090, vv2_1092, and vv2_1093). Because the gene cluster is predicted to be regulated through the metabolites of unsaturated glucuronate based on the existence of the VV2_0915 protein homologous to KdgR from Erwinia chrysanthemi, a transcriptional regulator (35), GalGlyNAcP_Vv seems to be related to the metabolism of some acidic polysaccharides.

The gene cluster explains the metabolism of the galactosyl residue in LNB. GalGlyNAcP_Vv phosphorolyses LNB to form Gal 1-P and GlcNAc. Then Gal 1-P is converted into Glc 1-P by the concerted action of GalT and GalE through part of the Leloir pathway to enter the glycolysis pathway, as shown in Fig. 3. The metabolism of Gal 1-P is likely to be identical to GNB/LNB metabolism with GalGlyNAcP_Bl in B. longum (19, 33).

Although the gene cluster of V. vulnificus containing the GalGlyNAcP gene encodes a set of enzymes for Gal 1-P metabolism similar to the set in B. longum (19, 33), the most significant difference is the absence of the gene encoding N-acetylhexosamine 1-kinase (NahK), which phosphorylates both GlcNAc and GalNAc at their α-anomeric hydroxyl groups (30), in the whole-genome sequence of V. vulnificus.

In the bifidobacterial pathway, phosphorolysis of LNB and GNB by GalGlyNAcP produces GlcNAc and GalNAc, respectively. GalNAc is converted into N-acetyl-α-d-glucosamine 1-phosphate (GlcNAc 1-P) by the concerted actions of NahK, GalT_Bl, and GalE_Bl via N-acetyl-α-d-galactosamine 1-phosphate, UDP-GalNAc, and UDP-GlcNAc. On the other hand, GlcNAc is directly converted to GlcNAc 1-P by NahK. Then GlcNAc 1-P enters the amino sugar metabolic pathway (30) after conversion into N-acetylglucosamine 6-phosphate by N-acetylglucosamine phosphomutase (EC 5.4.2.3). In C. perfringens, whose GalGlyNAcP is GNB specific, GalNAc is predicted to be metabolized through phosphorylation of GalNAc by an NahK homolog (the CPF1410 protein) (27).

Considering only the metabolism of GlcNAc, the direct conversion of GlcNAc to N-acetylglucosamine 6-phosphate by N-acetylglucosamine kinase (NagK; EC 2.7.1.59) seems to be more advantageous because it requires only a single enzyme (Fig. 3B). V. vulnificus possesses an NagK homolog (locus vv1_2570), which has approximately 54% amino acid sequence identity with E. coli NagK (GenBank accession number AAC74203) (40). The absence of an NahK homolog seems reasonable because GalGlyNAcP_Vv, which was found to be specific for LNB, is not likely to be involved in the metabolism of GNB to form GalNAc. No gene homologous to the endo-α-N-acetylgalactosaminidase gene has been found in the V. vulnificus genome; the product of this gene releases GNB from the core 1 structure of mucin (2, 16), supporting the substrate specificity of GalGlyNAcP_Vv.

In the human body, epitopes of some ABO antigens and Lewis antigens contain an LNB unit (20). Some gangliosides on the cell surface also contain this unit (11). If V. vulnificus possesses an extracellular enzyme to liberate LNB from such sugar chains and a system to transport LNB into the cell, the hypothesis that the gene cluster encoding GalGlyNAcP_Vv aids the growth of V. vulnificus in the human body by allowing the organism to utilize LNB is reasonable.

Lacto-N-biosidase (EC 3.2.1.140), which has been found only in Streptomyces sp. strain 142 (37) and B. bifidum JCM1254 (41), is the only enzyme reported to release LNB from oligosaccharide with a type I chain. Both enzymes belong to glycoside hydrolase family 20. Although V. vulnificus has genes encoding three glycoside hydrolase family 20 homologs (the VV1_0241, VV1_1666, and VV2_0591 proteins) in its genome, the levels of amino acid sequence identity with the lacto-N-biosidases are very low. In addition, no protein highly homologous to the GNB/LNB binding protein, a component of the GNB/LNB transporter identified in B. longum JCM1217 (38), was found in V. vulnificus. Therefore, aspects of the LNB intake system of V. vulnificus are still unclear. A complete understanding of the LNB-metabolizing system should reveal the mechanism of growth of V. vulnificus in human tissue.

Classification of GalGlyNAcP.

We determined that GalGlyNAcP_Vv is a GalGlyNAcP specific for LNB (Table 2). This is the first report concerning an LNB-specific phosphorolytic enzyme. We previously assigned the name lacto-N-biose I phosphorylase to bifidobacterial GalGlyNAcPs which exhibited similar activities with LNB and GNB (19, 31). The existence of the LNB-specific enzyme GalGlyNAcP_Vv clearly suggests that the assigned name of the bifidobacterial enzymes is not appropriate. In this report, we redefine LNBP as a kind of GalGlyNAcP that has an increased preference for LNB rather than GNB. Thus, GalGlyNAcP_Vv is the first LNBP that has been reported. GalGlyNAcP exhibiting similar activities with LNB and GNB is therefore renamed galacto-N-biose/lacto-N-biose I phosphorylase (GLNBP). The bifidobacterial enzymes should be categorized as GLNBPs. Additionally, GalGlyNAcPs should be categorized in three classes, LNBP, GNBP, and GLNBP, based on substrate preference. Table 3 shows the kinetic parameters of the various GalGlyNAcPs with LNB and GNB, as well as their categories.

TABLE 3.

Substrate specificities of GalGlyNAcPs^a

Strain	Enzyme	Sp act (U/mg) with:		Category
		LNB	GNB
B. longum JCM1217	GalGlyNAcPBl	15	27	GLNBP
C. perfringens ATCC 13124	GalGlyNAcPCp	0.28b	3.7b	GNBP
V. vulnificus CMCP6	GalGlyNAcPVv	18	0.53	LNBP

^aThe reactions were carried out in the presence of 50 mM Tris-HCl (pH 7.5) containing 10 mM LNB or GNB and 10 mM P_i at 30°C, unless indicated otherwise.

^bActivities were measured at 37°C.