INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Agar, a polysaccharide present in the cell walls of some red algae, can be degraded by several bacterial strains from marine environments and other sources. Some of the bacterial isolates have been assigned to the genera Alteromonas (1, 2, 21, 27, 33), Cytophaga (43), Streptomyces (36), Vibrio (3, 39), and Pseudomonas (22).

Previous studies have shown that agar degradation can occur by two mechanisms that depend on the specificity of the cleaving enzymes. The first pathway for agar breakdown comes from studies on Pseudoalteromonas atlantica ATCC 19292 (29, 30) and relies on extracellular -agarases. In this bacterium, an endo -agarase I cleaves the -(1,4) linkages of large agar polymers to a mixture of oligosaccharides with neoagarotetraose as the final product. These oligosaccharides are then hydrolyzed by the cell-bound exo -agarase II, yielding neoagarobiose. Finally, neoagarobiose is hydrolyzed to 3,6-anhydro-L-galactose and galactose in the cell cytoplasm by neoagarobiose hydrolase (15). The second lytic mechanism involves the cleavage of -(1,3) linkages on agarose by extracellular -agarases (33, 46, 47), yielding oligosaccharides from the agarobiose series, which contain D-galactose at the nonreducing end. The agarolytic system of Alteromonas agarolyticus strain GJIB consists of two enzymes: an -agarase that cleaves the -(1,3) linkages and a -galactosidase specific for the presence of the 3,6-anhydro-L-galactose units at the reducing end (33). Agarotriose was the smallest product detected in this system.

Biochemical and genetic studies on extracellular -agarases from several bacterial species have revealed a high degree of heterogeneity in terms of their molecular weights, specificities, and catalytic properties (10, 16, 27, 29, 39, 41, 42). The existence of regions of similarity between the amino acid sequences of the -agarases from Streptomyces coelicolor and Alteromonas atlantica was first suggested by Belas (9). Multiple sequence alignments of the amino acid sequences of -carrageenase from A. carrageenovora with 16 glycosyl hydrolases, including -agarase from S. coelicolor, have revealed the presence of invariant aspartic and glutamic residues (5). The sequence EIDXXE, which corresponds to the residues 155 to 160 of -agarase from S. coelicolor, was highly conserved. In addition, two short regions of homology between the amino acid sequences of -agarases from Vibrio sp. strain JT0107, A. atlantica strain T6c and S. coelicolor have also been observed (40). Further studies on the characterization of new agarases and their coding genes will be required to determine the significance of these conserved regions.

In our laboratory, we have isolated a few agar-softening and agar-liquefying bacterial strains from the southern Chilean coast to characterize their extracellular agarases in an attempt to contribute to our understanding of the basis of agar hydrolysis. Previous results on the purification and characterization of an extracellular agarase from the agar-liquefying strain Alteromonas sp. strain C-1 have been reported (27). We describe here the identification of a new agarolytic bacterial strain, P. antarctica strain N-1, and the characterization of an extracellular -agarase.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Strain N-1 was isolated from decomposing algae in Niebla (Valdivia, Chile). The screening was carried out on agar plates in a medium containing 0.25% casein hydrolyzate, 0.05% yeast extract, 0.5% proteose peptone, 3% NaCl, 0.06% NaH₂PO₄, 0.5% MgSO₄, 0.002% FeSO₄ · 7H₂O, 0.01% CaCl₂, and 1.5% agar (medium A). The plates were incubated at 25°C for 48 h. Colonies that formed pits or clearing zones on agar were picked up and purified further by the same plating method. For liquid cultures, agar (0.2%) was added before sterilization. Sugars were sterilized by filtration through 0.2-µm (pore size) membranes.

P. atlantica ATCC 19292 and Shewanella putrefaciens strain 8071 were obtained from the American Type Culture Collection, P. antarctica type strain was available from J. Guinea (University of Barcelona, Barcelona, Spain) (11).

Phenotypic analysis of the strain. Strain N-1 was identified by using Bergey's Manual of Systematic Bacteriology and The Prokaryotes as previously described (7, 18). Staining, morphology, and motility were determined as described by Cowan (14). Oxidation and fermentation tests were done in MOF medium as recommended by Leifson (26), but without agar. Anaerobic conditions were obtained by using Anaerocult A (Merck, Darmstad, Germany). The type of flagellum was determined by negative staining with uranyl acetate and electron microscopy as described by Cole and Popkin (13). Other biochemical and physiological tests were carried out essentially as described by Stolp and Gadkari (38) and Stanier et al. (37). Genomic DNA was prepared by the procedure of Ausubel et al. (4), and the G+C content was determined by high-performance liquid chromatography (HPLC) by the method of Kumura et al. (23).

PCR amplification of the 16S RNA gene. Amplification of the 16S ribosomal DNA (rDNA) was carried out as described by Ruimy et al. (35). First, 10 to 20 ng of purified genomic DNA was amplified in 50 µl of a reaction mixture consisting of 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 0.12 mM deoxynucleoside triphosphates, and 2.5 U of Taq DNA polymerase with primers 5'-AAGTCGTAACAAGGTAAC-3' and 5'-CTGAGCCATCAAACTCT-3' (7 µM concentrations of each). The initial denaturation step was 4 min at 95°C; this was followed by an annealing step at 52°C for 80 s and an extension step at 72°C for 90 s. The thermal profile then consisted of 25 cycles of annealing at 52°C for 80 s, extension at 72°C for 90 s, and denaturation at 94°C for 45 s. A final extension step was carried out at 72°C for 5 min. The single DNA band of approximately 1.5 kb as detected by agarose gel electrophoresis was purified by using the DNA extraction kit Wizard (Promega, Madison, Wis.). The DNA sequence was determined by direct sequencing of the PCR product on an Applied Biosystems sequencer (Ana-Gen Technologies, Inc., Palo Alto, Calif.).

Phylogenetic analysis and alignment. The sequence of the 16S rDNA of strain N-1 was aligned with the sequences of a number of Pseudoalteromonas strains available and was analyzed essentially as described by Gauthier et al. (20).

Cell growth and activity measurements. An overnight culture of isolated colonies was prepared in a medium of the same composition as that of medium A, except that the agar concentration was lowered to 0.2%, and was used to inoculate 100 ml of fresh medium. The cells were grown in an orbital shaker at 140 rpm and 25°C to the stationary phase. Phenylmethylsulfonylfluoride (PMSF) was added to a final concentration of 0.1 mM and centrifuged at 8,000 × g. Agarase activity was determined by the method of Dygert et al. (17) in the conditions described by Leon et al. (27).

Purification of agarase N-1. Unless specified otherwise, all operations were done at 4°C. An overnight culture of isolated colonies of strain N-1 was prepared in the medium described above and used to inoculate 2 liters of fresh medium containing 0.15% agar. The cells were grown in a Lab-line orbital shaker at 140 rpm and 25°C to the stationary phase (30 h). PMSF was added to a final concentration of 0.1 mM, and the cells were centrifuged at 6,000 × g for 25 min. The supernatant was brought to 75% saturation with solid ammonium sulfate over 3 h and centrifuged at 6,000 × g for 25 min. The pellet was resuspended in 22 ml of 20 mM Tris-HCl (pH 7.1), 0.1 mM EDTA, and 0.1 mM PMSF (buffer A) at 0°C and dialyzed three times against the same buffer at 4°C. The dialyzate was loaded onto a DEAE-cellulose column (10 by 1.5 cm) equilibrated with buffer A. The protein was eluted batchwise with 90 ml of 1.5 M NaCl in buffer A and concentrated by precipitation with ammonium sulfate (75% saturation), dissolved in 2 ml of buffer A, and loaded on a Sephadex G75 column (60 by 2.5 cm) equilibrated with buffer A. Fractions (3 ml) were collected, pooled on the basis of activity, and then loaded onto the DEAE-cellulose column (10 by 1.5 cm). Under these conditions more than 85% of the enzyme eluted in the flowthrough. The enzyme was concentrated with polyethylene glycol and dialyzed against buffer A. The enzyme was stored at 20°C and was stable for more than 6 months.

Protein determination. The amount of protein in the column fractions was determined by measuring the A₂₈₀. The amounts of protein in the pooled fractions were estimated by the method of Bradford (12) with fructose-1,6-bisphosphatase as the standard.

SDS-PAGE. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions was performed by the procedure of Laemmli (24) with 12% acrylamide gels. Proteins were stained with Coomassie brilliant blue R-250. Lysozyme (14.4 kDa), trypsin inhibitor (21 kDa), carbonic anhydrase (31 kDa), ovalbumin (45 kDa), and bovine serum albumin (66 kDa) were used as markers.

Hydrolysis product analysis. To characterize the hydrolysis products of agar with the purified enzyme, a solution of 0.2% agar (50 ml) was digested with 5 U of agarase in 1% ammonium carbonate (pH 7.0) for 48 h at 30°C. A 1.5-ml aliquot was concentrated in a Speedvac to 0.1 ml and analyzed by HPLC by using a Polyspher CHNa column (Merck) equilibrated with water at 72°C. Then 20 µl of a 3% (wt/vol) solution was injected. The oligosaccharides were detected by evaluation of the refractive index.

NMR spectroscopy. The undigested polysaccharides from the previous digest (48.5 ml) were precipitated with 50% ethanol, and the soluble material (70% [wt/wt]) was lyophilized prior to nuclear magnetic resonance (NMR) analysis. NMR experiments were performed on a Bruker AC 200 spectrometer at 25°C. ¹³C NMR spectra of 3% (wt/vol) oligosaccharide solutions in D₂O were acquired with composite-pulse decoupling or inverse-gated decoupling. ¹³C chemical shifts were referenced to tetramethylsilane by setting the internal dimethylsulfoxide resonance to 39.6 ppm or the internal acetone resonance to 31.07 ppm.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Strain properties and identification. Strain N-1 colonies softened the agar and produced halos of clearing after 24 to 48 h of incubation at 25°C. At longer incubation times, the colonies produced a red-brown diffusible pigment. The amount of pigment was dependent on the addition of tyrosine to the culture medium, suggesting the presence of a melaninlike pigment (2).

Phylogenetic analysis of 16S rRNA. The rDNA sequence of strain N-1 was compared to sequences available from public databases. Figure 1 shows an unrooted tree of the Pseudoalteromonas species. Strain N-1 and P. antarctica formed a robust clade. Based in these data we propose the assignment of our strain as P. antarctica N-1. However, we must point out that strain N-1 differs from the type strain of this species in some properties.

View larger version (25K): [in this window] [in a new window]   FIG. 1.   Dendrogram of the relatedness of strain N-1 with several Pseudoalteromonas species based on the 16S rDNA sequences. The unrooted tree was constructed by neighbor-joining analysis. +, Branch found by parsimony; *, branch found by maximum likelihood (P < 0.01). Percentages are indicated by bootstraps (500 replicates for neighbor-joining analysis; 100 replicates for parsimony).

Effect of carbon sources on bacterial growth and agarase production. Figure 2 shows the growth curve of P. antarctica N-1 and the production of agarase in the presence of agar. The highest level of agarase was reached during the stationary phase. At longer incubation periods the level of agarase decreases, a trend probably due to the presence of proteases. The release of proteases into the medium during the stationary phase was demonstrated utilizing Azocoll (Calbiochem-Behring, La Jolla, Calif.) as a chromogenic substrate for the proteases. In the presence of agar, glucose or galactose did not affect the production of agarase in this strain (data not shown). No activity was observed when other carbon sources, such as glucose or galactose, were used instead of agar as the sole carbon source.

View larger version (20K): [in this window] [in a new window]   FIG. 2.   Growth and agarase activity of P. antarctica N-1.

Purification of agarase N-1. Strain N-1 was cultured in liquid medium containing 0.15% agar at 25°C. Purification was attempted after 30 h of incubation. The enzyme was purified by taking advantage of its high binding affinity to DEAE-cellulose when loaded at low salt concentrations at cruder stages. The enzyme was slowly released from the DEAE-cellulose by a washing with 1.5 M NaCl. Additional purification of the enzyme was achieved by gel filtration on Sephadex G75 (Fig. 3), at which point a large amount of material absorbing at 280 nm and polysaccharides eluted ahead of the enzyme. Further purification of agarase N-1 was achieved by rechromatography on a DEAE-cellulose column. At this step the enzyme eluted in the flowthrough, indicating that the strong binding seen at the beginning of the purification could be mediated by an unidentified extracellular component of this strain or by an agar-derived product that is separated during the gel filtration step.

View larger version (22K): [in this window] [in a new window]   FIG. 3.   Chromatography of agarase from P. antarctica N-1 on Sephadex G75.

View larger version (11K): [in this window] [in a new window]   FIG. 4.   SDS-PAGE of purified agarase from P. antarctica N-1. Lane 1, molecular mass standards; lane 2, purified agarase (ca. 10 µg).

Molecular mass. Agarase N-1 had a molecular mass of 33 kDa, as determined by a comparison with the mobility of protein standards (Fig. 4). This value is close to those reported for -agarase from P. atlantica ATCC 19292 (32 kDa) (28), Pseudomonas sp. strain PT-5 (31 kDa) (45), and S. coelicolor (10). The molecular mass of the enzyme was estimated by gel filtration by using Sephadex G25 and Superdex 75 columns. In both cases the enzyme showed a molecular mass of 16 kDa, indicating an interaction with these resins.

Effects of pH and temperature on enzyme activity. The pH profile of agarase from strain N-1 was bell shaped, with a maximum at pH 7.0 (Fig. 5A). The enzyme was stable under the conditions of this assay as determined by measuring the residual activity at pH 7.0 after a 30-min incubation at the different pH values (data not shown). Similar results were observed for -agarase I from P. atlantica ATCC 19292 (not shown).

View larger version (16K): [in this window] [in a new window]   FIG. 5.   (A) Effect of pH on the activity of the purified agarase. The activity was determined at a pH between 3.6 and 10.0 using the following buffers: 100 mM sodium acetate (pH 3.6 to 5.0 []), 20 mM morpholineethanesulfonic acid (pH 5.0 to 6.0 []), 20 mM sodium phosphate (pH 6.0 to 7.6 []), Tris-HCl (7.5 to 9.0 []), and 50 mM glycine-NaOH (pH 9.0 to 10 []). (B) Effect of temperature on the stability of agarase from P. antarctica N-1. The enzyme was incubated in 50 mM sodium phosphate at pH 6.5, and the residual activity was determined at 30°C.

Effect of salt concentration on enzyme activity. When enzyme activity was measured in the presence of NaCl in concentrations of up to 0.5 M, no significant changes were observed.

Kinetic properties. The Michaelis constants of agarase from strain N-1 and -agarase I from P. atlantica X82134 were also determined. The assays were carried out in 50 mM phosphate (pH 7.0). K_m values of 0.077 and 0.044 mg/ml, respectively, were obtained from the double reciprocal plots (not shown).

Agar hydrolysis pattern. As shown by the HPLC profile, the purified enzyme from strain N-1 hydrolyzed agar to give two main oligosaccharide products (Fig. 6). The ¹³C NMR spectrum of this oligosaccharide mixture (Fig. 7) showed the typical patterns for a mixture of neoagarotetraose and neoagarohexaose (30). The identities of these oligosaccharides were confirmed by thin-layer chromatographic analysis on silica gel plates (not shown). The neoagarooligosaccharide series is typically produced by the cleavage of -(1,4) linkages by -agarase. Resonances at about 97 and 93 ppm are characteristic for the  and  anomeric forms, respectively, of galactose residues at the reducing end of the neoagarooligosaccharides (25). There was no evidence of a signal at 90.72 ppm, a finding which could be attributed to hydrolyzed -(1,3) linkages (3, 4). It can be concluded that agarase from strain N-1 is a -agarase.

View larger version (15K): [in this window] [in a new window]   FIG. 6.   Hydrolysis products of agar by agarase from P. antarctica N-1. Digested agar (20 µl, 3% [wt/vol]) was injected into a Polyspher CH-Na column (Merck) equilibrated with deionized water at 0.3 ml/min as described in Materials and Methods. The oligosaccharides were detected by determining the refractive index with a detector (Gilson, Middleton, Wis.). The positions of the neoagarohexaose (NH), neoagarotetraose (NT), neoagarobiose (NA), and galactose (G) standards are indicated.

View larger version (22K): [in this window] [in a new window]   FIG. 7.   (A) 13C NMR spectrum of the hydrolysis products of unsubstituted agar by agarase from P. antarctica N-1. (B) Oligosaccharides released by agarase.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

-Agarases can be divided into several groups according to their sizes, as shown in Table 3. Agar clearing, softening, and depressions around the colonies is characteristic for bacteria in groups 1 and 2. This effect would be related to the production of low-molecular-weight agarases that can diffuse though the gel pores. Cleavage of the polysaccharide chains causes agar softening and allows faster evaporation of water, leading to the formation of depressions. The exception is Alteromonas sp. strain C-1 (27), for which agar liquefaction appears to be dependent on the production of high concentrations of agarase.

We describe here the characterization of a new agarolytic bacterium isolated from the southern Chilean coast. This strain was identified as P. antarctica N-1 by phylogenetic studies based on analysis of the 16S rDNA gene sequence. An extracellular agarase was purified to homogeneity in high yield by gel filtration and two steps of ion-exchange chromatography on DEAE-cellulose. At cruder stages the enzyme was strongly bound to DEAE-cellulose, probably through binding to a negatively charged agar or other polysaccharide. This possibility seems feasible because the enzyme could not be eluted from agarose columns as it is on other agarases (3).

The purified enzyme had a molecular mass of 33 kDa, as indicated by SDS-PAGE under reducing conditions. The low molecular mass estimated by gel filtration indicates that the enzyme interacts with the resins, and we cannot establish whether or not it is a monomer. A molecular mass of 20 kDa was estimated for a -agarase from Vibrio sp. strain AP-2 by gel filtration on TSK-Fractogel HW-55 by Aoki et al. (3); however, the molecular mass as determined by SDS-PAGE was not reported. This behavior was not observed for -agarase I from P. atlantica.

HPLC analysis of the hydrolysis products of unsubstituted agar generated by agarase from P. antarctica N-1 showed the presence of neoagarotetraose and neoagarohexaose as the main products. These products were further analyzed by NMR to determine the specificity of the cleavage. The ¹³C NMR spectrum showed resonances at 97 and 93 ppm, which are typical for the  and  anomeric forms of D-galactose at the reducing end (30), indicating that the cleavage occurs at the -(1,4) linkages. Cleavage at the -(1,3) linkages leaves 3,6-anhydro-L-galactose at the reducing end. The C-1 signal in this case is observed at 90.72 ppm (33). Furthermore, the reducing power of the agarooligosaccharides is greatly decreased by the presence of 3,6-anhydro-L-galactose at the reducing end (33). In contrast, the reducing powers of the products generated by agarase from P. antarctica N-1 and -agarase I were similar under identical assay conditions.

-Agarase from P. antarctica N-1 and -agarase I from P. atlantica share several properties. However, some differences in their molecular masses and mainly in their stabilities at temperatures over 30°C were observed. Further characterization of the encoding genes of the -agarases from related species will be required to provide insight into the existence of regions involved in substrate binding or catalysis.