Biochemical Characteristics and Substrate Degradation Pattern of a Novel Exo-Type β-Agarase from the Polysaccharide-Degrading Marine Bacterium Flammeovirga sp. Strain MY04

ABSTRACT

Exo-type agarases release disaccharide units (3,6-anhydro-l-galactopyranose-α-1,3-d-galactose) from the agarose chain and, in combination with endo-type agarases, play important roles in the processive degradation of agarose. Several exo-agarases have been identified. However, their substrate-degrading patterns and corresponding mechanisms are still unclear because of a lack of proper technologies for sugar chain analysis. Herein, we report the novel properties of AgaO, a disaccharide-producing agarase identified from the genus Flammeovirga. AgaO is a 705-amino-acid protein that is unique to strain MY04. It shares sequence identities of less than 40% with reported GH50 β-agarases. Recombinant AgaO (rAgaO) yields disaccharides as the sole final product when degrading agarose and associated oligosaccharides. Its smallest substrate is a neoagarotetraose, and its disaccharide/agarose conversion ratio is 0.5. Using fluorescence labeling and two-stage mass spectrometry analysis, we demonstrate that the disaccharide products are neoagarobiose products instead of agarobiose products, as verified by ¹³C nuclear magnetic resonance spectrum analysis. Therefore, we provide a useful oligosaccharide sequencing method to determine the patterns of enzyme cleavage of glycosidic bonds. Moreover, AgaO produces neoagarobiose products by gradually cleaving the units from the nonreducing end of fluorescently labeled sugar chains, and so our method represents a novel biochemical visualization of the exolytic pattern of an agarase. Various truncated AgaO proteins lost their disaccharide-producing capabilities, indicating a strict structure-function relationship for the whole enzyme. This study provides insights into the novel catalytic mechanism and enzymatic properties of an exo-type β-agarase for the benefit of potential future applications.

IMPORTANCE Exo-type agarases can degrade agarose to yield disaccharides almost exclusively, and therefore, they are important tools for disaccharide preparation. However, their enzymatic mechanisms and agarose degradation patterns are still unclear due to the lack of proper technologies for sugar chain analysis. In this study, AgaO was identified as an exo-type agarase of agarose-degrading Flammeovirga bacteria, representing a novel branch of glycoside hydrolase family 50. Using fluorescence labeling, high-performance liquid chromatography, and mass spectrum analysis technologies, we provide a useful oligosaccharide sequencing method to determine the patterns of enzyme cleavage of glycosidic bonds. We also demonstrate that AgaO produces neoagarobiose by gradually cleaving disaccharides from the nonreducing end of fluorescently labeled sugars. This study will benefit future enzyme applications and oligosaccharide studies.

INTRODUCTION

Agarose is a complex polysaccharide that contains repeated disaccharide units of 3,6-anhydro-l-galactopyranose-α-1,3-d-galactose (1). Agarose has been identified to be one of the polysaccharide components of the cell wall and intercellular substances in red algae (Rhodophyta), such as Gracilaria and Porphyra (2). Because of their strong gel-forming ability and high chemical stability, agarose and its derivatives have been widely used in various applications (3–5).

Agarose can be hydrolyzed by α-agarase (EC 3.2.1.158) to produce agarooligosaccharides (AOs) with 3,6-anhydro-l-galactopyranose (A) as the reducing end (6) or cleaved by β-agarase (EC 3.2.1.81) to yield neoagarooligosaccharides (NAOs) with d-galactose (G) as the reducing end (7). Agarases have been successfully applied in many biotechnological applications, such as DNA gel recovery (8) and algal protoplast preparation (9, 10). Agarose-derived oligosaccharides have shown antioxidation and anti-inflammation activities (11, 12) and probiotic and whitening effects (13, 14), suggesting their potential usage in the cosmetic, food, and medical industries. Therefore, agarases are important tools for oligosaccharide preparation.

During the last decade, numerous agarases have been identified from marine and terrestrial bacteria (2, 15–17). Bacterial agarases have diverse protein sequences, molecular masses, and agarose degradation patterns. Two enzymes have been reported to be α-agarases belonging to glycoside hydrolase (GH) family 96 (GH96) (18, 19), while others are β-agarases of the GH16, GH50, GH86, and GH118 families (20, 21). Most agarases are endo-type enzymes that randomly and internally degrade agarose, producing a series of even-numbered oligosaccharides as products (16), and a few, mainly GH50 β-agarases, are exo-type enzymes that yield neoagarobiose (NA2) as the sole final agarose degradation product (22, 23). In bacterial agarose degradation, endo-type agarases have an important role in initially depolymerizing agarose into oligomers, while exo-type agarases are essential for releasing the disaccharide units to facilitate subsequent sugar metabolism in bacteria (24, 25).

Among the exo-type agarases, the structure and enzyme characteristics of Aga50D of Saccharophagus degradans strain 2-40 have been well characterized (22, 26), which has provided insights into the properties of GH50 family enzymes. The Aga50D protein has an N-terminal carbohydrate-binding module (CBM) and a C-terminal catalytic module. Crystal structures of the oligosaccharides in complex with Aga50D have shown that two glutamine residues (Gln⁵³⁴ and Gln⁶⁹⁵) contributed to the catalytic active site, which is conserved among GH50 family members. Sugar-binding-site residues are distributed in both the catalytic groove and the carbohydrate-binding domain, which facilitate activity on the double helix of agarose (26). However, the biochemical mechanisms by which agarases degrade substrates and release products in an exolytic pattern are still unclear.

Bacteria of the Flammeovirga genus have recently been identified from the surface of algae (27), deep-sea and coastal sediments (28–30), and marine animal innards (15). All reported Flammeovirga strains are efficient in the enzymatic degradation and bacterial utilization of multiple polysaccharides, such as agarose, alginate, and starch. Notably, Flammeovirga sp. strain MY04 could liquefy and grow on agarose as the sole carbon source, indicating that MY04 can produce efficient agarose-degrading enzymes (29). To date, one GH86 β-agarase (AgaP4383 of Flammeovirga pacifica strain WPAGA1) (31) and two GH16 β-agarases (AgaG4 of Flammeovirga sp. strain MY04 and AgaYT of Flammeovirga yaeyamensis strain YT) (27, 32) have been identified from Flammeovirga strains. The AgaYT protein shares a sequence identity of 98.5% with the AgaG4 protein, and Yang et al. determined that neoagarotetraose (NA4) and NA2 were the final oligosaccharide products of recombinant AgaYT (rAgaYT) (27); however, the molecular masses of the final products were not determined. Moreover, the two final oligosaccharide products appeared to be neoagarohexaose (NA6) and NA4 rather than NA4 and NA2 because the R_f value of d-galactose was smaller than that of NA2 but larger than that of NA4 under the designated thin-layer chromatography (TLC) system (22, 23, 33, 34). Therefore, all the three studied agarases of Flammeovirga strains produce NA4 as the smallest oligosaccharide product (31, 32). However, the tetraoligosaccharide cannot be utilized directly and efficiently by bacteria until it is further enzymatically digested. Thus, there should be essential enzymes responsible for the generation of disaccharides from agarose. Herein, we report the biochemical characteristics and novel enzymatic properties of the AgaO protein from Flammeovirga sp. strain MY04 and identify it as a disaccharide-producing β-agarase from polysaccharide-degrading bacteria of the genus Flammeovirga.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

Unless otherwise noted, Escherichia coli strains were cultured at 37°C in Luria-Bertani (LB) medium, supplemented with antibiotics, e.g., ampicillin (100 μg/ml), if necessary. The marine bacterium Flammeovirga sp. strain MY04 (CGMCC 2777) was cultured at 30°C in medium (pH 7.0) composed of 0.40% (wt/vol) tryptone, 0.25% yeast extract, and 3.0% NaCl. Agar powder was added at a concentration of 1.5% (wt/vol) to prepare solid media. Agar, agarose, alginate, carrageenan, ι-carrageenan, λ-carrageenan, cellulose, and xylan were purchased from Sigma-Aldrich (USA).

Sequence analysis of the genes and proteins.

DNA sequences were translated into protein sequences using the software BioEdit, version 7.2.1 (35), and the percent GC contents were calculated. For functional prediction, searches of the similarity of the protein sequences to those of known sequences were performed using the BLAST algorithm from the National Center for Biotechnology Information (NCBI) server (http://www.ncbi.nlm.nih.gov). Signal peptides and their types were analyzed using the SignalP (version 4.1) and the LipoP (version 1.0) online servers (http://www.cbs.dtu.dk/services/), respectively. The molecular weights of the proteins were estimated using the peptide mass tool on the ExPASy server of the Swiss Institute of Bioinformatics (http://swissmodel.expasy.org/). The modules and domains of the proteins were identified using the Simple Modular Architecture Research Tool (https://en.wikipedia.org/wiki/Simple_Modular_Architecture_Research_Tool), the Pfam database (http://pfam.xfam.org), and the Carbohydrate-Active Enzyme database (http://www.cazy.org). Multiple-sequence alignments and phylogenetic analysis were performed using MEGA software, version 6.06 (36).

Construction of expression vectors.

Total genomic DNA of Flammeovirga sp. strain MY04 was extracted using a universal DNA purification kit (TianGen Co., Ltd., Beijing, China). To express the AgaO protein, the full-length gene was amplified using the primers BAgaO-F and BAgaO-R (Table 1) and the high-fidelity DNA polymerase PrimeSTARHS (TaKaRa, Dalian, China). Primer pairs with restriction enzyme sites for XhoI and XbaI (underlined in Table 1) were designed to generate a 6×His tag at the C terminus of the recombinant protein (rAgaO). The DNA products were gel purified and then cloned into the expression vector pBAD/gIII A (Invitrogen, USA). The plasmid carrying the recombinant (pBAgaO) was transformed into Escherichia coli TOP10 cells. The integrity of the nucleotide sequences was confirmed by sequencing the expression plasmids thrice.

TABLE 1.

Bacterial strains, plasmids, and primers used in the present study

Strain, plasmid, or primer	Descriptiona or sequenceb	Source or reference
Strains
Flammeovirga sp. strain MY04	A polysaccharide-degrading marine bacterium (patented as CGMCC 2777)	29
E. coli BL21(DE3)	F− ompT hsdSB (rB− mB−) dcm gal λ(DE3)pLysS Cmr	Novagen
E. coli TOP10	F− mcrA Δ(mrr-hsdRMS-mcrBC) ϕ80dlacZΔM15 ΔlacX74 deoR recA1 araD139 Δ(araA-leu)7697 galU galK rpsL endA1 nupG	Invitrogen
Plasmids
pBAD/gIII A	Expression vector; Apr	Invitrogen
pCold TF	Expression vector; Apr	TaKaRa
pBAgaO	pBAD/gIII A carrying an amplified NcoI-XbaI fragment encoding the recombinant protein rAgaO	This study
pCTF-AgaO	pCold TF carrying an amplified NcoI-XhoI fragment encoding the recombinant protein rTFAgaO	This study
Primers
BAgaO-F	5′-CTCGAGTTTTTTTTAGCATTCAATCCGCC-3′	This study
BAgaO-R	5′-TCTAGAAAGTTATTCACATTACCCAAACGG-3′	This study
TC-D344-R	5′-TCTAGAAAATCAAATTGTTGATAGAATCCTAATGGCTTCC-3′	This study
TC-A530-R	5′-TCTAGAAAAGCATATACATCTTCATTGGCACCATCACC-3′	This study
TC-D626-R	5′-TCTAGAAAGTCTTTAGCTACTTTTACCCCTGGGTGG-3′	This study
TN-F164-F	5′-CTCGAGTTTGGGATGTTGAAACGTGATGTGACC-3′	This study
TN-N350-F	5′-CTCGAGAATTTCTATCAAGCCAACTTATATCG-3′	This study
TF-AgaO-R	5′-TCTAGATTAGTTATTCACATTACCCAAACGG-3′	This study

^aCm^r, chloramphenicol resistant; Ap^r, ampicillin resistant.

^bXhoI and XbaI restriction enzyme sites are underlined.

Moreover, the full length of the agaO gene was amplified using the primer pair BAgaO-F and TF-AgaO-R (Table 1) with restriction enzyme sites for XhoI and XbaI, to generate a recombinant protein (rTFAgaO) fused with a cold shock protein (TF) at the N terminus and a 6×His tag at the C terminus. The DNA products were finally cloned into the expression vector pCold TF (TaKaRa, Dalian, China). The resulting plasmid, pCTF-AgaO, was transformed into E. coli BL21(DE3) cells.

Heterologous expression and purification of the recombinant proteins.

Unless otherwise noted, protein analyses were performed at 4°C (32). Briefly, E. coli TOP10 cells harboring the plasmid pBAgaO were initially cultured in LB broth. When the cell density reached an A₆₀₀ of 0.6 to 0.8, l-arabinose was added at final concentrations ranging from 0.001 to 1.0 mM. After continued cultivation for an additional 24 h at 16°C, the cells were harvested by centrifugation, washed twice with ice-cold buffer A (50 mM Tris, 150 mM NaCl, pH 8.0), resuspended, and disrupted by sonication. After centrifugation, the soluble fraction was collected and loaded onto a Ni-nitrilotriacetic acid (Ni-NTA) agarose column (Novagen, USA), and the column was then eluted using imidazole at increasing gradient concentrations (0, 10, 50, 100, and 250 mM). Purified protein samples were diluted and dialyzed against buffer B (25 mM Tris, 5% [vol/vol] glycerol, pH 8.0) (1:50, vol/vol). SDS-PAGE was performed using 12% (wt/vol) polyacrylamide gels as described by Sambrook and Russell (37). Proteins were detected by staining the gel with Coomassie brilliant blue R-250. Protein concentrations were determined by the Folin-Lowry method using the Folin-Ciocalteu phenol reagent (Sigma-Aldrich, USA) and bovine serum albumin as a standard.

E. coli BL21(DE3) cells harboring the plasmid pCTF-AgaO were used to express the recombinant fusion protein rTFAgaO, and IPTG (isopropyl-β-d-thiogalactopyranoside) was added at final concentrations ranging from 0.001 to 1.0 mM. The recombinant protein rTFAgaO was expressed and purified using a strategy similar to that described above for rAgaO.

Enzyme activity assay.

The polysaccharide-degrading activity of rAgaO was determined by quantifying reducing sugars using the 3,5-dinitrosalicylic acid (DNS) assay (38). The absorbance of the reducing sugar product was measured at 540 nm, with d-galactose used as a standard. One unit of enzyme was defined as the amount of enzyme that produced 1 μmol reducing sugars per minute.

To determine the optimal substrate of rAgaO, various polysaccharides (e.g., agar, agarose, alginate, carrageenan, ι-carrageenan, λ-carrageenan, cellulose, and xylan) were individually dissolved in deionized water to prepare stock solutions (1.0 mg/ml). Each stock solution (500 μl) was mixed with the appropriately diluted enzyme in buffer B at an equal volume and then incubated for 12 h at different temperatures ranging from 0 to 70°C. Enzyme-treated samples were initially heated in boiling water for 10 min and then cooled on ice. After centrifugation at 15,000 × g for 15 min, the supernatant was collected and analyzed for the enzymatic products.

Biochemical characterization of rAgaO.

To determine the optimal conditions for rAgaO activity, stock solutions of agarose (1.0 mg/ml) were prepared using buffers with different pH values, including acetate buffer (50 mM, pH 4.0 to 6.5), NaH₂PO₄-Na₂HPO₄ buffer (50 mM, pH 6.0 to 8.0), and Tris-HCl buffer (50 mM, pH 7.0 to 10). The optimal temperature of rAgaO was assayed by monitoring the agarase activity at temperatures ranging from 0 to 70°C at pH 7.0 for 1 h. The pH dependence of rAgaO was tested at 45°C for 1 h. The thermostability of rAgaO was evaluated by measuring the residual activity after incubating the enzyme at various temperatures for 2 h. The effects of pH on rAgaO stability were determined by measuring the residual activities after incubating the enzyme in various pH environments (4.0 to 10) at 4°C for 2 h. The effects of metal ions and chelating agents on the activity of rAgaO were examined by determining its activity in the presence of 1 mM and 10 mM various chemicals, respectively.

Analysis of the agarose-degrading pattern of rAgaO.

Agarose (1.0 mg/ml) was digested by rAgaO (0.01 U/ml) at 45°C for over 72 h. Aliquots of the digestions were collected for time course analysis by thin-layer chromatography (TLC) using silica gel 60 plates (F254; Merck, Germany) (32). The R_f value of each oligosaccharide fraction was calculated according to the spots visualized on the TLC plates. The concentrations of oligosaccharides from the enzymatic digestions were determined using the DNS reducing sugar assay (38).

Characterization of the final oligosaccharide products.

To obtain size-defined oligosaccharides, 100 mg agarose was degraded by excess rAgaO at 45°C for 72 h. The final enzymatic digests were loaded onto a Superdex peptide 10/300 GL column (GE Healthcare, USA) and monitored with a refractive index detector (Shimadzu, Kyoto, Japan). The mobile phase was a NH₄HCO₃ solution (0.20 M), which was used at a flow rate of 0.4 ml/min. Data analysis was performed using the software LCsolution, version 1.25. The size-defined oligosaccharide fractions were collected and repeatedly freeze-dried to remove the NH₄HCO₃. The rate of agarose conversion was calculated as the mass ratio of the obtained oligosaccharide products to the initial polysaccharide substrates.

To determine the molecular masses, approximately 0.1 to 1.0 μg purified oligosaccharide products was assayed using matrix-assisted laser desorption ionization–time of flight mass spectrometry (MS) (Axima-CFR plus; Shimadzu, Japan). For nuclear magnetic resonance (NMR) spectroscopy, purified oligosaccharide samples (∼2 mg) were dissolved in 0.5 ml of D₂O in 5-mm NMR tubes and repeatedly freeze-dried. The spectra were recorded on a JNM-ECP600 (JEOL, Japan) apparatus set at 600 MHz. Data analyses were performed using MestReNova (version 6.1.0-6224) software.

Analysis of the oligosaccharide degradation pattern of rAgaO.

To assay the smallest substrate, various oligosaccharides were reacted with the enzyme rAgaO, and the digests were analyzed by TLC as described above. d-Glucose and d-galactose (1.0 mg/ml) were used as standard markers to assay the lactose-degrading activity as described previously (39).

To determine the molar ratios of the final disaccharide products to the starting oligosaccharide substrate, each substrate (∼1 μg) and its enzymatic digests were individually labeled using 2-aminobenzamide (2-AB) (40) and were then analyzed using the Superdex 10/300 GL column, monitored with excitation and emission wavelengths of 330 and 420 nm, respectively. A mixture composed of 2-AB-labeled oligosaccharides, e.g., d-galactose, NA2, NA4, NA6, and neoagarooctaose (NA8), was used as a standard marker. The molar ratios were calculated according to the corresponding peak area shown in high-performance liquid chromatography (HPLC) analysis.

To further examine the oligosaccharide degradation pattern of rAgaO, each size-defined oligosaccharide fraction (∼1 μg) was labeled with excess 2-AB, purified with gel filtration using the Superdex 10/300 GL column, and repeatedly freeze-dried to eliminate NH₄HCO₃. Each 2-AB-labeled oligosaccharide substrate (∼0.1 μg each) was mixed with rAgaO (0.01 U) in a 30-μl volume, incubated at 45°C, and traced over 12 h. The final enzymatic digests were analyzed using the HPLC system with fluorescence detection.

Saccharide sequencing of oligosaccharides.

To determine the sequences of the oligosaccharides produced by rAgaO, the enzymatic digests (∼1 μg) of the oligosaccharides or agarose were fluorescently labeled using excess 2-AB. Oligosaccharide samples and 2-AB-labeled oligosaccharide mixtures were initially analyzed by mass spectrometry (MS) and subsequently by two-stage MS (MS/MS) analyses. Data analysis was performed using the software LCMSsolution, version 3.80.410.

Gene truncation, protein expression, and enzyme characterization.

The full-length gene of AgaO was PCR amplified using the primers listed in Table 1 to generate truncated gene fragments, as indicated in Fig. 1A. The DNA products were individually gel purified and cloned into the plasmid pBAD/gIII A. The resulting plasmids were each expressed using the same method described for pBAgaO. Soluble bacterial fractions containing the truncated proteins were each used as an enzyme preparation for further activity tests and product analyses.

FIG 1.

Sequence properties of the agarase AgaO from Flammeovirga sp. strain MY04. (A) The gene encoding AgaO and the module organization of AgaO. The numbers indicate nucleotides. The protein contains an N-terminal signal peptide (Met¹ to Phy²⁵, in black on the left) and a hypothetical catalytic module (Ser³⁹⁵ to Ile⁶⁰², in black on the right). The primer pairs indicated at the top and bottom were used to generate the full-length gene and various truncated fragments. (B) Phylogenetic analysis based on the protein sequence alignment. The neighbor-joining tree was obtained using MEGA (version 5.05) software. The numbers on the branches indicate the bootstrap confidence values from 1,000 replicates. The bar is equal to the distance corresponding to 1 amino acid substitution per 10 amino acid residues.

Accession number(s).

The nucleotide sequence of agaO in strain MY04 has been submitted to the GenBank database under accession number KU524066.

RESULTS

AgaO gene and protein sequences.

In the genomic sequence of Flammeovirga sp. strain MY04, the agaO open reading frame is 2,118 bp in length (GenBank accession no. KU524066) and has a GC content of 35.5%. The predicted protein, AgaO, has a molecular mass of ∼81.3 kDa, and its calculated isoelectric point is 7.07. SignalP (version 4.1) and LipoP (version 1.0) analyses indicated that the type II signal peptide of AgaO was composed of 25 amino acid residues (Met¹ to Phy²⁵) (Fig. 1A).

AgaO, which has no homologous gene in the MY04 genome, is a unique protein. According to BLASTp searches, the mature protein of AgaO (Asn²⁶ to Asn⁷⁰⁵) shared sequence identities of less than 40% with characterized β-agarases of the GH50 family. The main fragment of AgaO (Met¹⁶³ to Asn⁷⁰²) showed 35% identity with the exo-type β-agarase Aga50D (Met¹⁵⁵ to Ser⁷⁴⁹; PDB accession no. 4BQ4, chain A) of S. degradans strain 2-40, a type member of the GH50 family (22, 26). Protein sequence alignment showed that AgaO shares homologous regions with GH50 β-agarases, mostly in the C-terminal region (see Fig. S1 in the supplemental material), which was identified to be the catalytic module in Aga50D (26). Furthermore, within these partially conserved regions, AgaO contained two glutamine residues (Gln⁴⁵⁸ and Gln⁶⁰⁶) that were strictly conserved in the catalytic motif of GH50 β-agarases (see Fig. S1 in the supplemental material), Gln⁵³⁴ and Gln⁶⁹⁵ in Aga50D, respectively, while it did not contain the same N-terminal CBM-like module, as previously reported for Aga50D (26). Analysis using the Carbohydrate-Active Enzyme database and the Simple Modular Architecture Research Tool suggested that AgaO contained only one putative catalytic module (Ser³⁹⁵ to Ile⁶⁰²). Phylogenetic analysis showed that AgaO is far distant from other characterized GH50 β-agarases (Fig. 1B). The results indicate that AgaO is a novel agarase obtained from bacteria of the genus Flammeovirga and that it belongs to a novel branch of the GH50 family to be defined.

Heterologous expression of AgaO in E. coli.

The full-length agaO gene was amplified from the genomic DNA of Flammeovirga sp. strain MY04. The 2.1-kb DNA product was gel purified and cloned into the vector pBAD/gIII A downstream of a P_BAD promoter. A secretion peptide from the gIII phage and a 6×His tag, respectively, were individually added to the N and C termini of the protein product (rAgaO) in the expression vector (pBAgaO). SDS-PAGE analysis indicated that TOP10 cells harboring the plasmid pBAgaO could produce soluble recombinant proteins (∼120 mg/liter) with an appropriate molecular mass (i.e., 85 kDa) (see Fig. S2 in the supplemental material). After sonication and centrifugation, soluble crude enzyme was extracted from the E. coli cultures. Protein fractions of rAgaO could be eluted from the Ni-NTA column using imidazole at concentrations higher than 50 mM. The rAgaO protein was further purified through gel filtration chromatography, if needed. SDS-PAGE analysis indicated that the purified soluble protein rAgaO had a purity of >99%, a gel recovery of ∼60%, and an initial concentration of ∼5 mg/ml (see Fig. S2 in the supplemental material).

To obtain high protein yields, the full-length agaO gene was also cloned into the vector pCold TF. BL21(DE3) cells harboring the resulting plasmid, pCTF-AgaO, could produce soluble rTFAgaO at a yield of ∼1.4 g/liter, nearly 12-fold greater than that of rAgaO.

Enzymatic characteristics of AgaO.

The recombinant proteins rAgaO and rTFAgaO did not digest alginate, carrageenan, ι-carrageenan, λ-carrageenan, cellulose, or xylan but could efficiently hydrolyze agar and agarose to produce oligosaccharide products. The enzymatic digests of agar and agarose exhibited a strong absorbance at 540 nm in the DNS assay, suggesting that AgaO is a hydrolase of agarose.

The full-length enzyme rAgaO demonstrated its highest activity at 45°C when agar or agarose was used as the substrate (Fig. 2A). The rAgaO enzyme retained more than 80% of its residual activity after incubation for 2 h at temperatures ranging from 0 to 40°C (Fig. 2A). The optimal pH, determined at 45°C in various buffers, was 7.0 (Fig. 2B). The enzyme retained more than 80% of its residual activity after incubation at 4°C for 2 h in environments with pHs ranging from 6 to 8 (Fig. 2B).

FIG 2.

Biochemical characteristics of the agarase rAgaO. (A) Thermostability of rAgaO and effects of temperature on its enzyme activities. (B) The pH stability of rAgaO and effects of pH on its enzyme activities. (C) Effects of various compounds on enzyme activities. PMSF, phenylmethylsulfonyl fluoride; 2-ME, 2-mercaptoethanol; DTT, dithiothreitol. (D) Effects of NaCl on enzyme activities.

The enzyme activity of rAgaO was strongly inhibited by most of the tested chemicals, such as the metal ions Ag⁺, Co²⁺, and Fe³⁺, at 10 mM or even 1.0 mM (Fig. 2C). The enzyme activity was inhibited by sodium dodecyl sulfate, phenylmethylsulfonyl fluoride, imidazole, and disodium EDTA. In contrast, the enzyme activity was increased to 124% by MgSO₄ (10 mM) and to 140% by glycerol at 5% (vol/vol) (Fig. 2C). Notably, sodium chloride at concentrations under 0.9 M increased the activity, with optimal activity (134%) being at ∼0.2 M (Fig. 2D). The results indicate that, consistent with its origin from a marine bacterium, the agarase AgaO can be activated by Mg²⁺ and Na⁺, common metal ion components in the sea.

Under optimal conditions (45°C, 50 mM NaH₂PO₄-Na₂HPO₄, 200 mM NaCl, pH 7.0), the specific activity of rAgaO for agarose was measured as described in Materials and Methods and was ∼185 U/mg of protein.

Agarose degradation pattern of AgaO.

To determine the polysaccharide-degrading pattern, the digestion of agarose by rAgaO was monitored at 45°C. The reaction time varied from 0 to 72 h. The digests (∼1 μg) were loaded onto a silica gel plate and subsequently developed. TLC analysis indicated that rAgaO initially produced a series of even-numbered oligosaccharide products which had R_f values similar to those of standard NAO markers (Fig. 3A). Remarkably, rAgaO rapidly converted them into the final product, which had the same R_f value as NA2 (Fig. 3A). The results suggest that AgaO is a neoagarobiose-producing β-agarase.

FIG 3.

Polysaccharide degradation pattern of the agarase rAgaO. (A) TLC analysis of the degradation products of agarose (1.0 mg/ml) by rAgaO (0.1 U/ml) at 45°C. Lane M, standard oligosaccharide markers; lane (−), negative control. (B) HPLC analysis of the final agarose digests by rAgaO at 45°C for 72 h. The oligosaccharide products were purified with a Superdex peptide 10/300 GL column monitored by use of a parallax detector. (C) HPLC analysis with fluorescence detection of the final agarose digests by rAgaO at 45°C for 72 h. The final products were 2-AB labeled, purified by gel filtration, and monitored at an excitation wavelength of 330 nm and a monitoring wavelength of 420 nm. M1, M2, M4, M6, and M8, 2-AB-labeled oligosaccharide markers of d-galactose, NA2, NA4, NA6, and NA8, respectively. The oligosaccharide by-products in the final agarose digestions are indicated by asterisks.

Characterization of the disaccharide product.

To identify the predominant oligosaccharide, 100 mg agarose was degraded using excess rAgaO, and the digests were fractionated by gel filtration chromatography. The main fraction, which had a peak retention time of 44.6 min in the designated HPLC system (Fig. 3B), was recovered as the final oligosaccharide product. A total of ∼54 mg pure oligosaccharides was obtained from the final digests, indicating that AgaO could convert agarose into oligosaccharides at a ratio of nearly 50% (wt/wt).

Approximately 100 ng pure oligosaccharides was used in each MS analysis. The results indicated that the purified oligosaccharide product had a molecular mass of 324. Further MS/MS analysis showed a strong signal of the pseudo ion [G+Na]⁺ at m/z 203 and a weak signal of the pseudo ion [A+Na]⁺ at m/z 185 (Fig. 4A). Thus, the predominant oligosaccharide product yielded by AgaO was identified to be a disaccharide of agarose composed of A and G units. However, direct MS analyses could not determine whether the disaccharide product is an agarobiose or a neoagarobiose.

FIG 4.

Identification of the disaccharide product yielded by rAgaO. (A) Direct MS/MS analysis of the purified disaccharide product (m/z 324). (B) MS/MS analysis of 2-AB-labeled disaccharide product (m/z 444). (C) ¹³C NMR spectrum of the purified neoagarobiose product. Inten, intensity; A, the 4-O-linked 3,6-anhydro-α-l-galactopyranose; G, the 3-O-linked β-d-galactopyranose; r and nr, residues at the reducing and nonreducing ends, respectively; α/β, the respective anomer.

To sequence the disaccharide product, i.e., to determine which sugar unit, A or G, comprises the reducing end, we fluorescently labeled the reducing ends of the purified disaccharide product using excess 2-AB and then tested the resulting derivative (∼50 ng) by MS analyses. Initial MS analysis indicated that the signal of the 2-AB-labeled disaccharide was m/z 444. Further MS/MS analysis showed strong signals from the 2-AB-labeled G units (m/z 323) and associated 2-AB-labeled fragments (m/z 201 and 231) but not from the 2-AB-labeled A unit (Fig. 4B). Thus, the final disaccharide product of AgaO was identified to be a neoagarobiose that had a G unit as the reducing end.

To confirm the results of the above-described fluorescence assays, ∼2 mg of the purified disaccharide products was analyzed to obtain a ¹³C NMR spectrum. The result showed two resonance types at δ92 and δ97 ppm, which indicated the existence of a free anomeric carbon in the G unit at the reducing end (Fig. 4C) (7). Moreover, only signals from the A units at the nonreducing end (δ98 ppm) were found, with no signals from G units at the nonreducing end or A units at the reducing end being found (Fig. 4C). The results demonstrated that AgaO yields neoagarobiose instead of agarobiose as the disaccharide product when degrading agarose. Thus, AgaO is identified to be a neoagarobiose-producing β-agarase from Flammeovirga sp. strain MY04. Moreover, the ¹³C NMR spectrum analysis confirmed that the fluorescence assays are useful in determining the glycoside-type selectivity of a new agarase.

Candidate by-products in the agarose digestions of rAgaO.

TLC analysis suggested that during the agarose digestion, rAgaO appeared to have yielded a low proportion of by-products, with the R_f values being quite different from those of even-numbered NAOs, e.g., NA2, NA4, and NA6 (Fig. 3A). To identify the proportion of these abnormal by-products, the final agarose digests by rAgaO (∼1.0 μg) were fluorescently labeled using excess 2-AB and subsequently gel filtered using an HPLC system with fluorescence detection. The 2-AB-labeled products showed retention times different from those of the 2-AB-labeled NAO markers (Fig. 3C). According to the peak area, the candidate by-products each accounted for ∼0.01 to 0.10% (molar ratio) of the predominant product, NA2 (Fig. 3C). The results confirmed that AgaO produces NA2 as the sole final product in agarose degradation.

Oligosaccharide degradation patterns of rAgaO.

To assay the smallest substrate, oligosaccharides with different chain lengths (e.g., NA2, NA4, NA6, NA8, and NA10) were individually dissolved in water and then reacted with excess rAgaO at pH 7.0 and 45°C for 24 h. TLC analysis indicated that rAgaO could degrade oligosaccharides larger than NA2 in size and produced NA2 as the final product (Fig. 5A). Moreover, rAgaO could not further degrade NA2 to produce any detectable monosaccharide units (Fig. 5B). Therefore, the smallest neoagarooligosaccharide substrate of AgaO is NA4, and the smallest oligosaccharide product is NA2.

FIG 5.

Pattern of oligosaccharide degradation by the agarase rAgaO. (A) TLC analysis of the final digests of NA2, NA4, NA6, NA8, and NA10. Lane M, standard NAO markers; lanes 1, 3, 5, 7, and 9, control groups; lanes 2, 4, 6, 8, and 10, groups reacting with rAgaO. (B) TLC analysis of the reaction products of lactose with rAgaO. Lane 1, d-galactose; lane 2, d-glucose; lane 3, control group for lactose; lane 4, lactose degradation by rAgaO. (C) HPLC analysis with fluorescence detection of the final digests of lactose and NAOs. Each final product was 2-AB labeled and analyzed by HPLC. (D) HPLC analysis with fluorescence detection of the exo-type degradation pattern of rAgaO. The oligosaccharide NA6 was initially 2-AB labeled and then reacted with rAgaO.

To further determine the molar ratios of the NA2 products to each oligosaccharide substrate, the substrates and the products of their final digestions by rAgaO were initially labeled using excess 2-AB and then analyzed using an HPLC system with fluorescence detection. According to the peak area, the molar ratios of the NA10, NA8, NA6, NA4, and NA2 substrates to their NA2 products were determined to be approximately 5:1, 4:1, 3:1, 2:1, and 1:1, respectively (Fig. 5C). The results demonstrate that AgaO can degrade agarose and associated oligosaccharides completely, yielding neoagarobiose as the final disaccharide product.

To determine the oligosaccharide-degrading pattern of AgaO, the oligosaccharide NA6 was initially labeled using excess 2-AB, subsequently reacted with rAgaO, and monitored over time using HPLC analysis with fluorescence detection. The agarase rAgaO initially degraded 2-AB-labeled NA6 into 2-AB–NA4 and unlabeled NA2 and finally converted 2-AB–NA4 into 2-AB–NA2 and unlabeled NA2, thus yielding NA2 and 2-AB–NA2 as the final products (Fig. 5D). Notably, almost 1% (molar ratio) of the substrate 2-AB–NA4 could not be hydrolyzed completely to release the NA2 and 2-AB–NA2 products, even if the enzyme amount or the reaction time was doubled (Fig. 5D). This suggested that the 2-AB label at the reducing end of NA4 caused a stereospecific blockade and thus weakly inhibited the enzymatic degradation by rAgaO. The results demonstrate an exo-type disaccharide-yielding property of the β-agarase AgaO, which releases neoagarobiose by gradually cleaving the units from the nonreducing end of 2-AB-labeled oligosaccharides.

Because AgaO shares low homology with the β-galactosidase VadG925 of Victivallis vadensis strain ATCC BAA-548, two different disaccharides, lactose and neoagarobiose, were used as the substrates to assay the galactosidase activities. The former disaccharide contained a G unit at its nonreducing end, while the latter contained the G unit as its reducing end. TLC (Fig. 5B) and HPLC analyses with fluorescence detection (Fig. 5C) indicated that, unlike the β-galactosidase VadG925 (39), rAgaO could not degrade the disaccharide lactose to release any monosaccharide units of d-galactose or d-glucose. Thus, AgaO is a neoagarobiose-producing β-agarase without detectable galactosidase activities.

Agarose bioconversion properties of the truncated proteins.

To identify functional modules within the protein AgaO, various gene fragments were generated via PCR, using the full-length AgaO gene as the DNA template. Five truncated proteins were successfully generated, i.e., three proteins (TC-D344, TC-A50, and TC-626) with deletion of the N-terminal noncatalytic region from AgaO and two recombinant proteins (TN-F164 and TN-N350) containing the full-length putative catalytic module (Fig. 1A and Table 1). The crude enzyme preparations containing each truncated protein were initially reacted with agarose at 30°C to 45°C for 12 h, and the digestions were assayed. DNS assays showed that three of the truncated proteins, i.e., TC-626, TN-F164, and TN-N350, weakly produced reducing sugars. However, further TLC analysis indicated that all five protein preparations lost the ability to produce NA2 as products (negative results; data not shown). Therefore, the putative catalytic module is not the sole functional element required for the full-length enzyme AgaO to produce disaccharide products. Moreover, it is suggested that the interaction between the putative modules is essential for the β-agarase AgaO to act in an exo-type pattern, which requires strict structure-function coordination.

DISCUSSION

Because of the excellent agarose degradation and utilization capabilities of Flammeovirga strains, various agarase preparations, such as crude extracellular enzymes (29) and pure recombinant agarases (27, 31, 32), have been investigated. And all of these enzyme preparations yielded NA4 rather than NA2 as the smallest oligosaccharide product. Recently, we have sequenced and annotated the genome of a polysaccharide-degrading marine bacterium, Flammeovirga sp. strain MY04, to identify the genes involved in diverse polysaccharide metabolic pathways (41). In addition to a previously reported GH16 β-agarase, AgaG4, which yielded NA4 and NA6 as the final oligosaccharide products (32), we have identified another candidate gene (agaO) in the MY04 genome. Interestingly, due to its novel sequence properties and low homologies to previously identified enzymes, AgaO appeared to be an exo-type GH50 β-agarase, while it contains a putative catalytic module (Fig. 1A and B). In this study, we expressed the AgaO protein in E. coli cells to obtain soluble proteins (see Fig. S2 in the supplemental material) for enzyme characterization. The recombinant enzyme rAgaO was identified as a disaccharide-producing β-agarase but not a galactosidase of the Flammeovirga genus.

During the process of agarose degradation by rAgaO, we also found a series of candidate by-products in the final oligosaccharide products by TLC and HPLC analyses with fluorescence detection (Fig. 3A and C). These products had very low molar proportions in the final agarose degradation products (Fig. 3C), even if we increased the amount of enzyme by 2-fold or doubled the reaction time. However, similar products were not found in the final digests of various size-defined NAO substrates (Fig. 5A and C). Because AgaO is an agarase without galactosidase activities, it is suggested that the candidate by-products are not oligosaccharides produced by rAgaO via any glycosyl transferase activity but that they are complex components of agarose, e.g., sulfated oligosaccharide units, and indigestible by rAgaO.

The ¹³C NMR spectrum analysis of oligosaccharide products can easily distinguish AOs from NAOs by analyzing the ¹³C NMR signals of the anomeric carbon, which belongs to the A or G unit at the reducing end and thus determines the α or β hydrolysis pattern of agarases (6, 7). However, the NMR analysis requires oligosaccharide products with a purity of >99% and is efficient at a mass grade of a milligram, which is time-consuming and expensive. MS/MS analysis can easily distinguish unit A from unit G by assaying the molecular masses (Fig. 4A). Moreover, it does not require purification of the target oligosaccharide products, while it is efficient at a lower mass grade of a microgram or even a nanogram. However, direct MS and subsequent MS/MS analyses were insufficient to determine the reducing end of the final disaccharide product (Fig. 4A and B). To solve the problem, we initially labeled the reducing end of the disaccharide product using 2-AB, followed by MS/MS analysis. In the MS/MS spectrum, a main peak of 2-AB–G (m/z 323) without any 2-AB–A peaks was detected (Fig. 4B). Thus, although it has a low molecular mass, the final disaccharide product produced by rAgaO was easily demonstrated to be NA2 with a G unit as the reducing end, which is consistent with the result of the ¹³C NMR spectrum analysis (Fig. 4C). This strategy can also be used to identify larger oligosaccharide products (42) and therefore is very useful for rapidly determining the glycoside cleavage patterns of the corresponding enzymes.

Agarase can degrade agarose by either an endolytic or an exolytic pattern. Numerous disaccharide-producing agarases have been well characterized with regard to their biochemical characteristics and agarose degradation products (22, 23, 26, 33). However, only one NA2-producing β-agarase, Aga50D of S. degradans strain 2-40, which has the exolytic pattern, determined by crystal structure analysis of the protein-oligosaccharide complex, has been reported (26). Therefore, biochemical demonstration of the exolytic pattern of agarases is still needed. In the present study, through the use of fluorescence labeling and further HPLC analyses, we demonstrated that AgaO is an NA2-producing exolytic agarase. First, when degrading agarose or associated oligosaccharides, rAgaO always produces NA2 as the sole final product (Fig. 3C and 5C). Second, when degrading 2-AB-labeled oligosaccharides, rAgaO releases the NA2 product by gradually cleaving the unit from the nonreducing end of oligosaccharide chains (Fig. 5D). To the best of our knowledge, the second finding represents novel biochemical visualization data on the exolytic pattern of an agarase.

Conclusion.

AgaO is a disaccharide-producing β-agarase obtained from agarolytic bacteria of the genus Flammeovirga. It belongs to a novel branch of the GH50 family to be defined. The enzyme can efficiently degrade agarose and associated oligosaccharides into neoagarobiose by sequential digestion from the nonreducing end of sugar chains. Moreover, this study has provided a novel biochemical visualization of the exolytic pattern of an agarase. The fluorescence assays will be helpful for demonstrating the substrate-degrading and product-yielding patterns of various agarases.

Funding Statement

This work was financially supported by the Major State Basic Research Development Program of China (grant no. 2012CB822102), the National Natural Science Foundation of China (grant no. 31300664 and 31570071), and the State Key Laboratory of Microbial Technology of Shandong University (grant no. M2013-11).