Previous Article | Next Article ![]()
Applied and Environmental Microbiology, May 2006, p. 3515-3523, Vol. 72, No. 5
0099-2240/06/$08.00+0 doi:10.1128/AEM.72.5.3515-3523.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Food Biotechnology Division, National Food Research Institute, 2-1-12 Kannondai, Tsukuba, Ibaraki 305-8642, Japan,1 Research Center for Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 6, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8566, Japan,2 Faculty of Science, Saitama University, 255 Shimo-okubo, Saitama 338-8570, Japan,3 Faculty of Bioresources, Mie University, 1515 Kamihama, Tsu 514-8507, Japan4
Received 16 August 2005/ Accepted 11 March 2006
|
|
|---|
-L-arabinofuranosidase-treated AGP. |
|
|---|
Arabinogalactan proteins (AGPs) are an abundant and heterogeneous class of highly glycosylated hydroxyproline (Hyp)-rich glycoproteins found in the extracellular matrix associated with the higher plant plasma membrane and cell wall (7, 11, 13, 36). Recent characterization of genes encoding different AGPs from several species has given rise to the current distinction between classical and nonclassical AGPs (5, 9, 29). Classical AGPs contain a domain responsible for attaching the protein to a glycosylphosphatidylinositol membrane anchor (3, 34, 38, 45). In contrast, nonclassical AGPs lack this glycosylphosphatidylinositol anchor signal, making them soluble components of the extracellular matrix, and often contain Asn- or Hyp-rich domains (13, 27, 35). The carbohydrate moieties of AGPs have a common structure consisting of ß-1,3-galactosyl backbones (sugars in the present study are D series unless otherwise designated) that is substituted at O-6 by side chains of ß-1,6-galactosyl residues. L-Arabinose and lesser amounts of other auxiliary sugars, such as glucuronic acid, L-rhamnose, and L-fucose, are attached to the side chains, usually at nonreducing termini (21).
Molecular and biochemical evidence has indicated that AGPs have specific functions during root formation (41, 43), promotion of somatic embryogenesis (23, 40, 42), and attraction of pollen tubes in the style (6, 33, 44). However, since a large number of putative protein cores exist and because of the complex structures of carbohydrate moieties, it has been difficult to differentiate one species of AGP from another in plant tissues. This, in turn, has made it difficult to assign specific roles to the individual AGPs. In spite of the significant physiological interest in AGPs, there has been very little research done on glycoside hydrolases that cleave their sugar moieties. We believe that it is important to study such enzymes, as hydrolytic enzymes specific to particular sugar residues and type of glycosidic linkage will provide useful tools for the structural analysis of the sugar moieties of AGPs.
Glycoside hydrolases have been classified in the CAZy database (http://www.cazy.org/CAZY/) (14). The classification system of the CAZy database is based on amino acid sequence similarities and reflects the structural features of the enzymes. Since there is a direct relationship between sequence and folding similarities, the members of one family most likely share the same architecture. Glycoside hydrolysis occurs via two major mechanisms, giving rise to either an overall retention or an inversion of anomeric configuration; enzymes in the same family catalyze by the same hydrolysis mechanism (8). The sequence-based classification of families of enzymes provides useful clues to their function but not to their substrate specificity.
The galactanases which can hydrolyze ß-1,3- or ß-1,6-galactosyl linkages are useful tools, as the enzymes hydrolyze the carbohydrate moieties of AGPs. Unfortunately, due to the lack of research, not many such enzymes are known. Recently for the first time, we cloned two such enzymes, the endo-ß-1,6-galactanase belonging to glycoside hydrolase family 5 (GH5) from Trichoderma viride (no EC number) and the exo-ß-1,3-galactanase belonging to GH43 from Phanerochaete chrysosporium (EC 3.2.1.145) (18, 22). The exo-ß-1,3-galactanase is an especially powerful tool to analyze AGP since it is able to accommodate branched side chains when the enzyme hydrolyzes main-chain ß-1,3-linkage of the carbohydrate moiety of AGPs.
In spite of their interest, exo-ß-1,3-galactanases have been previously identified only in fungi (18, 32, 39). When the amino acid sequence of exo-ß-1,3-galactanase from P. chrysosporium (Pc1,3Gal43A) was compared with sequences in the protein and nucleic acid database BLAST (on the website of the National Center for Biotechnology Information, http://www.ncbi.nlm.nih.gov/BLAST/), Pc1,3Gal43A showed 34% similarity with the hypothetical protein of Clostridium thermocellum ATCC 27405 (GenBank accession number ZP_00312459.1). To demonstrate that the exo-ß-1,3-galactanases are not exclusive to fungi, we cloned the gene of the hypothetical protein of C. thermocellum and investigated the properties of the recombinant protein.
Here, for the first time, we report an exo-ß-1,3-galactanase from a bacterium. In this paper, we demonstrate that exo-ß-1,3-galactanases are one of the components of cellulosomes and that the enzymes are widely distributed in nature.
|
|
|---|
-Galp, PNP-ß-Glcp, PNP-ß-L-Arap, PNP-ß-Xylp, PNP-ß-Manp, and PNP-ß-Fucp), larch arabinogalactan, laminarin, birchwood xylan, oat spelt xylan, ß-1,3-galactosyl L-arabinofuranoside, and ß-1,3-galactosyl glucosaminide were purchased from Sigma Chemical Company (St. Louis, MO). Debranched arabinan and pectic galactan from lupine were from Megazyme International, Ltd. (Wicklow, Ireland). ß-1,3-Galactosyl galactosaminide was from BioCarb (Lund, Sweden). Gum arabic was from Nacalai Tesque, Inc. (Kyoto, Japan). Carboxymethyl curdlan (CM-curdlan) was from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Arabinan, ß-1,3-galactan, and ß-1,3-ß-1,6-galactan from Prototheca zopfii, native and
-L-arabinofuranosidase-treated AGPs from radish, and ß-1,3-, ß-1,4-, and ß-1,6-galacto-oligosaccharides, 4-O-MeGlcA-ß-1,6-Gal [4MeGlcA(ß1-6)Gal], and 4-O-MeGlcA-ß-1,6-Gal-ß-1,6-Gal [4MeGlcA(ß1-6)Gal2] were prepared as previously described (18). Methyl ß-glycoside of ß-1,3-galactotetraose and -pentaose were kind gifts from P. Ková
of the National Institutes of Health, NIDDKD (National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD). ß-1,3-Xylan was a kind gift from Y. Kitamura of the National Food Research Institute (Tsukuba, Japan).
Expression of recombinant Ct1,3Gal43A.
For the expression of the full-length enzyme (Ct1,3Gal43A-CBM13-DocI) (where Ct1,3Gal43A is the C. thermocellum exo-ß-1,3-galactanase and CBM13 is the family 13 CBM of Ct1,3Gal43A), the catalytic module (Ct1,3Gal43A), or the CBM (CBM13-DocI), the pET expression system (Novagen, Madison, WI) was used. The PCRs were performed by using the following sets of primers: F1 (forward primer, 5'-CAT ATG GCA GAA GGG GTT ATA GTC AAC-3') and R1 (reverse primer, 5'-CTC GAG TTA CAA ATC CAC TTC CAT AAG-3') for Ct1,3Gal43A-CBM13-DocI; F1 and R2 (reverse primer, 5'-CTC GAG AGT GTC AGG TAT GTA TTC AGA-3') for Ct1,3Gal43A; and F2 (forward primer, 5'-AAG CTT ACA CGC TAC AAA CTG GTA AAC-3') and R1 for CBM13-DocI (Fig. 1). The amplified fragment was subcloned into the pGEM-T Easy vector (Promega, Madison, WI), and the complete nucleotide sequences of the genes were determined with an ABI PRISM 310 genetic analyzer (Applied Biosystems, Foster City, CA). Each gene was individually inserted into the pET30 or pET27 vector with a restriction enzyme site of NdeI-XhoI or HindIII-XhoI (underlined).
|
View larger version (10K): [in a new window] |
FIG. 1. Domain structure of Ct1,3Gal43A. GH43, glycoside hydrolase family 43; CBM13, carbohydrate-binding module family 13; DocI, type I dockerin domain.
|
Characterization of recombinant Ct1,3Gal43A.
Activity of exo-ß-1,3-galactanase was determined by the Somogyi-Nelson method (37). The reaction mixture consisted of 20 µl of McIlvaine buffer (0.2 M Na2HPO4, 0.1 M citric acid), pH 6.0, and 25 ml of 1% (wt/vol) ß-1,3-galactan. After 5 min of preincubation at 37°C, 5 µl of enzyme was added to the solution and the reaction mixture was incubated at 37°C for 10 min. The reaction was terminated by heating the solution at 100°C for 5 min (standard condition). One unit of enzyme activity is defined as the amount of enzyme that released 1 µmol of galactose/min. Enzyme concentration was determined by using a bicinchoninic acid protein assay reagent kit (Pierce Biotechnology, Inc., Rockford, IL) with bovine serum albumin as a standard. SDS-PAGE was carried out using a 12% polyacrylamide gel, and the protein was stained with Coomassie brilliant blue R-250. The effects of pH on the activity and stability of Ct1,3Gal43A were investigated according to a method previously described (18). The N-terminal amino acid sequences of recombinant enzymes were determined with a model HP G105A protein sequence system. Affinity electrophoresis was performed as previously described (18). Matrix-assisted laser desorption ionization-time-of-flight mass spectrometry was performed with a KOMPACT MALDI IV tDE system (Shimadzu, Kyoto, Japan). Recombinant Ct1,3Gal43A-CBM13-DocI dissolved in 0.5 µl of water was crystallized by adding 0.5 µl of matrix solution (containing 1% [wt/vol] sinapinic acid and 0.1% [vol/vol] trifluoroacetic acid) and 0.5 µl of 1% (wt/vol) NaCl, which was then allowed to dry. For the mass calibration, bovine serum albumin (molecular mass, 66.431 kDa) was used.
Substrate specificity.
The substrate specificity of Ct1,3Gal43A was determined using various PNP glycosides as substrates. The enzyme was incubated with 1 mM PNP glycoside in McIlvaine buffer, pH 6.0, at 37°C for 10 min, and the reactions were terminated by adding Na2CO3 to the final concentration of 0.1 M. The amount of PNP released was then determined by measuring absorbance at 400 nm with an extinction coefficient of 19,608 M1 cm1.
The substrate specificity toward polysaccharides and oligosaccharides was determined at 37°C in McIlvaine buffer, pH 6.0, with 0.5% (wt/vol) polysaccharide or 5 mM of oligosaccharide as the substrate and 0.16 µM of enzyme. After incubation for the appropriate reaction time, the amount of hydrolysis was quantified by the Somogyi-Nelson method (37).
The catalytic efficiency of the enzyme against ß-1,3-galacto-oligosaccharides was investigated as previously reported (18). Briefly, the enzyme (1.8 nM) was incubated with substrate (10 µM) in McIlvaine buffer, pH 6.0, at 37°C. At regular time intervals, 100-µl aliquots were taken and heated at 100°C for 5 min to terminate the reaction. The substrate and the primary reaction products were then quantified by high-performance anion-exchange chromatography with a pulsed amperometric detection system (HPAEC-PAD) as previously reported (18, 28).
The hydrolysis products of ß-1,3-galactan or
-L-arabinofuranosidase-treated AGP from radish root were determined by HPAEC-PAD. The enzyme reactions were performed at 37°C in McIlvaine buffer, pH 6.0, containing 0.5% (wt/vol) substrate and 0.2 µM of enzyme and terminated by heating the solution at 100°C for 5 min. The products were analyzed by comparing their retention time by HPAEC to that of a standard sample.
FAC.
The affinity of CBM13 of Ct1,3Gal43A-CBM13 for a range of oligosaccharides was determined by frontal affinity chromatography (FAC) using Ct1,3Gal43A as a control, according to a method described previously (16-18, 31). Pyridylaminated (PA) oligosaccharides, containing a terminal galactose residue, used as substrates are shown in Fig. 4B. PA-Lac (Galß1-4Glc), PA-Gb3 (Gal
1-4Galß1-4Glc), PA-GA1 (Galß1-3GalNAcß1-4Galß1-4Glc), PA-LNT (Galß1-3GlcNAcß1-3Galß1-4Glc), PA-LNnT (Galß1-4GlcNAcß1-3Galß1-4Glc), PA-Galili penta (Gal
1-3Galß1-4GlcNAcß1-3Galß1-4Glc), and PA-rhamnose were purchased from TaKaRa (Kyoto, Japan). Nonlabeled oligolactosamines LN1 (Galß1-4GlcNAc), LN2 (Galß1-4GlcNAcß1-3Galß1-4GlcNAc), and LN3 (Galß1-4GlcNAcß1-3Galß1-4GlcNAcß1-3Galß1-4GlcNAc) and milk oligosaccharides ßGalLac (Galß1-3Galß1-4Glc) and ßGal2Lac (Galß1-3Galß1-3Galß1-4Glc) were gifts from K.-I. Yoshida (Seikagaku Co., Japan) and T. Urashima (Obihiro University of Agriculture and Veterinary Medicine, Japan), respectively. The nonlabeled glycans were pyridylaminated with GlycoTAG (TaKaRa Bio, Inc., Otsu, Japan) prior to use. PNP-Lac or PNP-LacNAc (Galß1-4GlcNAc), purchased from Sigma, was used for determination of the total amount of immobilized recombinant enzyme in the column (see Fig. 4A), according to a method described previously (16-18, 31).
![]() ![]() ![]() View larger version (82K): [in a new window] |
FIG.4. FAC analysis of CtCBM13 moiety. (A) Elution profiles used to determine the affinity of CtCBM13 moiety for (a) Lac or (b) LacNAc (top panels) and Eadie-Hofstee plots of the FAC data (bottom panels). (B) Structures of oligosaccharides used for FAC. (C) Elution profiles of PA oligosaccharides after application to immobilized CtCBM13 moiety.
|
|
|
|---|
Further analysis revealed that residues 77 to 267 of Ct1,3Gal43A showed sequence similarity to GH43 enzymes, while residues 357 to 490 showed similarity with CBM13 (Fig. 2A and B). The sequence at the N terminus is a signal sequence, and the C-terminal sequence is a dockerin domain, a cohesin binding region which normally mediates association with the cellulosome (Fig. 2C). The latter suggests that Ct1,3Gal43A is a cellulosome component (1, 46). Since AGPs might be the adhesive matrix between cellulose microfibrils and/or other matrix polysaccharides, the ß-1,3-galactanases could be involved in exposing cellulose from the intricacy of the plant cell wall matrix for the parent cellulolytic bacterium.
![]() ![]() ![]() View larger version (139K): [in a new window] |
FIG.2. Partial sequence alignment of Ct1,3Gal43A. (A) Sequence alignment of catalytic amino acids of P. chrysosporium Pc1,3Gal43A (GenBank accession number AB200390), C. thermocellum Ct1,3Gal43A (ZP_00312459.1), Arabidopsis thaliana At1 (CAB66926.1), and A. thaliana At2 (BAB08462.1). The alignment was performed with ClustalW. Identical amino acid residues are enclosed in black boxes. Conserved and semiconserved amino acid residues are in gray boxes and underlined, respectively. (B) Sequence alignment of selected CBM13s. CtCBM13, carbohydrate-binding module from exo-ß-1,3-galactanase from C. thermocellum; XBD, xylan-binding domain from xylanase 10A from Streptomyces olivaceoviridis (Protein Data Bank entry 1XYF_A); RTB1 and RTB2, domains 1 and 2 of ricin toxin B chain from Ricinus communis (Protein Data Bank entry 2AAI_B); EW29, galactose-binding lectin from the earthworm Lumbricus terrestris (GenBank accession number BAA36394) (15). indicates residues involved in substrate binding in RTB. The amino acid residues corresponding to the forming hydrophobic core of RTB are enclosed in black boxes. (C) Sequence alignment of selected type I dockerins. CtDocI is the type I dockerin module from exo-ß-1,3-galactanase from C. thermocellum. Others are PDB entries 1CLC (20), 1DAV_A (26), 1DAQ_A (26), and 1OHZ_B (4).
|
When Ct1,3Gal43A was incubated with ß-1,3-galactan, exo-ß-1,3-galactanase activity was detected and the specific activity was 21.3 U/mg. The enzyme achieved maximal activity at pH 6.0, and the activity on this substrate was stable between pH 3.0 and 10.0. The enzyme achieved maximal activity at 50°C, and more than 80% activity was retained after incubation at 55°C and pH 6.0 for 1 h.
Substrate specificity of Ct1,3Gal43A.
Ct1,3Gal43A was tested for the activities reported for the GH43 family, using the following substrates: ß-1,3-galactan, gum arabic, larch arabinogalactan, debranched arabinan, arabinan, soluble oat spelt xylan, soluble birchwood xylan, and PNP glycosides. The enzyme showed significant activity only toward ß-1,3-galactan and did not show any additional activities previously reported for GH43 (data not shown). Subsequently, activity of the enzyme towards other polysaccharides which have structures similar to that of ß-1,3-galactan was tested (Table 1). The enzyme did not hydrolyze ß-1,4-linked galactose-containing polysaccharides, such as pectic galactan and
-L-arabinofuranosidase-treated pectic galactan from lupine, and also did not hydrolyze laminarin, CM-curdlan, or ß-1,3-xylan, suggesting that the enzyme acts only on a ß-1,3-linked galactan chain. This is also apparent from the substrate specificities for oligosaccharides. The enzyme hydrolyzed ß-1,3-linked galacto-oligosaccharides but did not hydrolyze ß-1,4-linked and ß-1,6-linked galacto-oligosaccharide (Table 2). Ct1,3Gal43A also could not hydrolyze the ß-1,3-galactosyl linkage of ß-1,3-galactosyl galactosaminide, ß-1,3-galactosyl glucosaminide, or ß-1,3-galactosyl L-arabinofuranoside (Table 2). These data suggest that Ct1,3Gal43A cleaves only ß-1,3-linkage between two galactosyl residues, concluding that Ct1,3Gal43A is an exo-ß-1,3-galactanase.
|
View this table: [in a new window] |
TABLE 1. Substrate specificity of Ct1,3Gal43A toward polysaccharidesa
|
|
View this table: [in a new window] |
TABLE 2. Substrate specificity of Ct1,3Gal43A toward galacto-oligosaccharidesa
|
|
View this table: [in a new window] |
TABLE 3. Rates of ß-1,3-linked galacto-oligosaccharide hydrolysis by Ct1,3Gal43Aa
|
-L-arabinofuranosidase-treated AGP, the activity was lower than that for ß-1,3-galactan. However, peaks corresponding to ß-1,6-galactobiose, ß-1,6-linked galactotriose, 4MeGlcA(ß1-6)Gal, and 4MeGlcA(ß1-6)Gal2 were detected in addition to galactose (Fig. 3C). The data suggest that the enzyme is able to accommodate the ß-1,6-linked galactosyl side chains in AGPs, a property similar to that of Pc1,3Gal43A. Therefore, we concluded that Ct1,3Gal43A is an exo-ß-1,3-galactanase.
![]() View larger version (13K): [in a new window] |
FIG. 3. HPAEC-PAD analysis of hydrolysis products of ß-1,3-galactan or radish AGP generated by Ct1,3Gal43A. Ct1,3Gal43A (0.2 µM) was incubated with 0.5% (wt/vol) substrate in McIlvaine buffer, pH 6.0, at 37°C. At regular intervals, reaction products were analyzed by means of HPAEC-PAD as described in Materials and Methods. (A) Standards. (B) ß-1,3-Galactan as a substrate. (C) -L-Arabinofuranosidase-treated AGP from radish root as a substrate.
|
, ß, and
, and these three units assemble around the pseudo threefold axis, forming a globular structure. This combination of three subdomains results in a fold similar to the ß-trefoil fold. This fold is shared by plant galactose-binding lectins, ricin B chain, and abrin B chain, in which two sets of domains with a triple repeated sequence are arranged in tandem (12). Comparison with the ricin/lactose complex showed that most of the residues involved in lactose binding in ricin are strictly conserved among all three subdomains of CtCBM13, indicating that CtCBM13 could bind to galactose. However, there are no reports of CBM13 binding to ß-1,3-galactan. Thus far, several studies of the characteristics of CtCBM13 have been conducted. To clarify the binding specificity of CtCBM13, Ct1,3Gal43A-CBM13 was analyzed by affinity gel electrophoresis and FAC by using Ct1,3Gal43A as a control. When Ct1,3Gal43A or Ct1,3Gal43A-CBM13 was applied on gels which contained polysaccharides, such as hydroxyethylcellulose, soluble birchwood xylan, laminarin, gum arabic, gum arabic after twice-repeated Smith degradation, gum arabic after thrice-repeated Smith degradation (ß-1,3-galactan), larch arabinogalactan, potato galactan, or ß-1,3-xylan, no significant affinity for these substrates was observed (data not shown). However, the result of FAC using Ct1,3Gal43A-CBM13 indicated that CtCBM13 binds to several oligosaccharides, including ßGalLac and ßGal2Lac, which contained ß-1,3-linked galactosyl residues at the nonreducing end, and Lac, LNnT, LN1, LN2, and LN3, which possessed ß-1,4-linked galactosyl residues at the nonreducing end (Fig. 4), whereas Ct1,3Gal43A did not show affinity for any substrate (data not shown). The affinity of CtCBM13 for ßGalLac, which has an additional ß-1,3-Gal at the nonreducing end of lactose, was significantly higher than that for lactose (Fig. 4C). The Kd values for ßGalLac and ßGal2Lac (both containing ß-1,3-linked galactosyl residues) were 15.2 µM and 16.9 µM, respectively. The Kd values for LN3, LN2, and LNnT (substrates containing ß-1,4-linked galactosyl residues) were 36 µM, 21.8 µM, and 15.0 µM, respectively. CtCBM13 showed higher affinity for the oligosaccharide consisting of two LacNAc units (LN2) than for LacNAc. In contrast, the affinity of CtCBM13 for three LacNAc units (LN3) was lower than that for LacNAc. Moreover, CtCBM13 bound more strongly to LNnT than to LN2.
The sequence alignment of selected CBM13s with CtCBM13 is shown in Fig. 2B. The aromatic residues involved in substrate binding are conserved in all three subdomains of CtCBM13, suggesting that CtCBM13 has three binding pockets. Interestingly, some residues in the xylan-binding domain (XBD) (12, 24), the earthworm Lumbricus terrestris 29-kDa lectin (EW29) (15), and the ricin toxin B chain (RTB) (30), which were shown to form hydrogen bonds with the substrate, were not conserved in CtCBM13. The variation in these residues may cause the difference in substrate specificity in CBM13. The affinity of CtCBM13 for lactose (25.8 µM) was much higher than that of the xylan-binding module from Streptomyces lividans (SlCBM13, 1.0 mM), and it was the same order as for the other lectins (2). The results of FAC analysis indicated that CtCBM13 binds to oligosaccharides containing ß-1,3-linked or ß-1,4-linked galactosyl residues. The affinity of CtCBM13 for ßGalLac was almost the same as that for ßGal2Lac (Fig. 4C), suggesting that CtCBM13 recognizes ß-1,3-galactobiose (not triose) at the nonreducing end.
When the exo-1,3-galactanase activities of the two variants with or without CtCBM13 were compared, CtCBM13 appeared to interfere with the exo-ß-1,3-galactanase activity on ß-1,3-galactan and
-L-arabinofuranosidase-treated AGP (Fig. 5). This could be explained by the fact that CtCBM13 recognizes the ß-1,3-galactobiose at the nonreducing end and thereby blocks the enzyme from accessing its substrate. However, further analysis to examine the competition between the CtCBM13 and the catalytic domain could not be performed because it was not possible to express CtCBM13 only.
![]() View larger version (7K): [in a new window] |
FIG. 5. Effect of CtCBM13 on the hydrolysis rate of Ct1,3Gal43A. Enzyme (0.2 µM) was incubated with 0.5% (wt/vol) substrate in McIlvaine buffer, pH 6.0, at 37°C. At regular intervals, the amounts of hydrolysis products were determined by the Somogyi-Nelson method. Symbols: , Ct1,3Gal43A; , Ct1,3Gal43A-CBM13. (A) ß-1,3-Galactan. (B) -L-Arabinofuranosidase-treated AGP from radish root.
|
(National Institutes of Health, NIDDKD) for supplying the oligosaccharide substrates. We are also grateful to Y. Kitamura (National Food Research Institute) for supplying ß-1,3-xylan. We wish to thank S. Nakamura and R. Yabe for FAC analyses and S. Numao and L. Lo Leggio for critically reading the manuscript. |
|
|---|
6)-galactanase gene. Biochem. J. 377:749-755.[Medline]
3)-ß-D-galactanase from Aspergillus niger. Carbohydr. Res. 264:281-291.[Medline]
. 1990. Purification of an exo-ß-(1
3)-D-galactanase of Irpex lacteus (Polyporus tulipiferae) and its action on arabinogalactan-proteins. J. Biol. Chem. 265:7207-7215.This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»