Mechanistic Basis for Understanding the Dual Activities of the Bifunctional Azotobacter vinelandii Mannuronan C-5-Epimerase and Alginate Lyase AlgE7

Abstract

The structure and functional properties of alginates are dictated by the monomer composition and molecular weight distribution. Mannuronan C-5-epimerases determine the monomer composition by catalyzing the epimerization of β-d-mannuronic acid (M) residues into α-l-guluronic acid (G) residues. The molecular weight is affected by alginate lyases, which catalyze a β-elimination mechanism that cleaves alginate chains. The reaction mechanisms for the epimerization and lyase reactions are similar, and some enzymes can perform both reactions. These dualistic enzymes share high sequence identity with mannuronan C-5-epimerases without lyase activity. The mechanism behind their activity and the amino acid residues responsible for it are still unknown. We investigate mechanistic determinants involved in the bifunctional epimerase and lyase activity of AlgE7 from Azotobacter vinelandii. Based on sequence analyses, a range of AlgE7 variants were constructed and subjected to activity assays and product characterization by nuclear magnetic resonance (NMR) spectroscopy. Our results show that calcium promotes lyase activity, whereas NaCl reduces the lyase activity of AlgE7. By using defined polymannuronan (polyM) and polyalternating alginate (polyMG) substrates, the preferred cleavage sites of AlgE7 were found to be M|XM and G|XM, where X can be either M or G. From the study of AlgE7 mutants, R148 was identified as an important residue for the lyase activity, and the point mutant R148G resulted in an enzyme with only epimerase activity. Based on the results obtained in the present study, we suggest a unified catalytic reaction mechanism for both epimerase and lyase activities where H154 functions as the catalytic base and Y149 functions as the catalytic acid. IMPORTANCE Postharvest valorization and upgrading of algal constituents are promising strategies in the development of a sustainable bioeconomy based on algal biomass. In this respect, alginate epimerases and lyases are valuable enzymes for tailoring the functional properties of alginate, a polysaccharide extracted from brown seaweed with numerous applications in food, medicine, and material industries. By providing a better understanding of the catalytic mechanism and of how the two enzyme actions can be altered by changes in reaction conditions, this study opens further applications of bacterial epimerases and lyases in the enzymatic tailoring of alginate polymers.

Keywords: alginate; alginate C-5-epimerase; alginate lyase; enzyme mechanism; multifunctional enzyme; nuclear magnetic resonance (NMR); site‐directed mutagenesis; time-resolved NMR.

FIG 1

Proposed alginate lyase and epimerase reaction mechanisms. Both mechanisms start with proton abstraction from C-5 by AA2 (amino acid 2), requiring a charge neutralization of the carboxylate by AA1. In the lyase reaction, a proton is donated by AA3 to the leaving group of the glycosidic bond, whereas in the epimerase reaction, a proton is donated to the sugar ring to create the epimer. Only the M-lyase reaction is shown.

FIG 2

NMR spectra showing AlgE7 acting on polyM and polyMG. Underlined monomer residues give rise to the signals, and they give distinct peaks based on their nearest neighbor residues. (A) ¹H spectra after epimerization for 60 h. The mixture contained 5 mM HEPES, 75 mM NaCl, and 2.5 mM CaCl₂ (pH 7.0). A total of 2.5 mg/mL of the substrate and 92 nM enzyme were used in the reaction (1:300 [wt/wt] enzyme-to-substrate ratio). Fractions of residues calculated from the integration of the ¹H spectra at different time points are plotted to the right and shown in Table S1 in the supplemental material (calculation of sequential parameters is explained in Fig. S3). (B) Time-resolved ¹³C spectra of the reaction from 0 to 1,000 min (16.7 h), recorded at 25°C, with a mixture containing 10 mM MOPS, 75 mM NaCl, and 2.5 mM CaCl₂ at pH 6.9. A total of 11 mg/mL of the substrate and 2.9 μM enzyme were used (1:42 [wt/wt] enzyme-to-substrate ratio).

FIG 3

(A) Alignment of AlgE7 from A. vinelandii with two consensus sequences. ConE is the consensus sequence of the A-modules displaying only epimerase activity from A. vinelandii and A. chroococcum (AlgE1A1, AlgE1A2, AlgE2A, AlgE3A1, AlgE3A2, AlgE4, AlgE5, AlgE6, and AcAlgE1). ConL is the consensus of the A-modules that have both lyase and epimerase activities (AlgE7, AcAlgE2A, and AcAlgE3A). Residues that are not conserved in the consensus sequence are represented in the alignment as “X,” whereas dots denote residues that are almost or completely conserved. Highlighted letters indicate amino acids replaced in this study: purple indicates the catalytic amino acids, blue indicates the four residues differing between ConE and ConL, and green indicates other studied residues. AlgE7 has 82% identity with ConL and 63% identity with ConE. The alignment and consensus sequences are made from global protein alignments using the BLOSUM 62 scoring matrix. (B) The residues highlighted in panel A, shown in sticks in the crystal structure of AlgE4 (right) (PDB accession number 2PYH), and a homology model of AlgE7 (left), with the same color code as the highlights in panel A. Dashes are shown between residues that potentially form hydrogen bonds. The calcium ion needed for structural stability is visualized as an orange sphere. The model of the AlgE7 A-module was built by SWISS-MODEL (30), using the structure of the AlgE4 A-module (PDB accession number 2PYH), and then energy minimized using the YASARA server (31).

FIG 4

Product profiles from endpoint ¹H NMR of wild-type (wt) and mutant AlgE7 proteins acting on 1 mg/mL polyM, with 2.5 mM CaCl₂ added to the reaction mixtures, incubated at 25°C for 24 h. A total of 40 μL of enzyme extracts was added to 500 μL of the reaction mixture. Each column represents one experiment, and duplicates are positioned together. Green, internal G residues; orange, Δ residues; dark green, internal M residues; dark purple, M residues at reducing ends; light purple, G residues at reducing ends. The sum of these fractions amounts to 1.

FIG 5

NMR spectra showing R148G activity on polyM and polyMG. Underlined monomer residues give rise to the signals, and they give distinct peaks based on their nearest neighbor residues. (A) ¹H spectra after epimerization for 60 h, in a mixture containing 5 mM HEPES, 75 mM NaCl, and 2.5 mM CaCl₂ (pH 7.0). A total of 2.5 mg/mL of the substrate and 0.092 μM enzyme were added (1:300 [wt/wt] enzyme-to-substrate ratio). Plots show fractions calculated from the integration of spectra recorded at different time points of the reactions. (B) Time-resolved ¹³C spectra from 0 to 1,000 min (16.7 h), recorded at 25°C with a mixture containing 10 mM MOPS, 75 mM NaCl, and 2.5 mM CaCl₂ at pH 7.0. A total of 11 mg/mL of the substrate and 3.0 μM enzyme were used (1:42 [wt/wt] enzyme-to-substrate ratio).

FIG 6

Salt dependency experiments of wild-type AlgE7. (Top) Slopes from a factorial spectrophotometric lyase activity assay at two different substrate concentrations [S], three different pHs, three different calcium concentrations, and four different NaCl concentrations. A total of 0.3 μM enzyme was added to each reaction mixture. The slope is given in units per hour, where U is the absorbance units at 230 nm. Points are averages from two repetitions, and error bars show the ranges of values for the repetitions. (Bottom) Integrated time points from ¹³C NMR with a constant substrate concentration (9 mg/mL) and constant pH (6.9) at four different calcium and salt concentrations. A total of 3 μM enzyme was added to each reaction mixture. y axes show the relative integrals of peaks representing the given residues.

FIG 7

Illustration of the possible cleavage sites from the observed product patterns of AlgE7 on polyM (left) and polyMG (right). In the middle, AlgE7 is shown as it binds to the substrate that it is presented with initially, but this will change during the reaction course. Except for M↓MM, all cleavage sites are in accordance with cleavage after a processive binding event or due to a preferred attack of the end of MG and GG sequences. For polyM, we observed the creation of M_red, G_red, and ΔM, which can be caused by cleavage at the four cleavage sites illustrated in the left panels. On polyMG, mainly G_red and ΔM are observed, corresponding to cleavage site G↓GM. However, we also observe small amounts of M_red and ΔG, and four cleavage sites shown in the right panels are possible on polyMG. Sugar residues are colored according to the symbol nomenclature for graphical representations of glycans (49). The purple arrows indicate the proposed direction of movement for the processive AlgE4 (28). AlgE7 consists of one A-module and three R-modules, and the outline of all four modules is shown. The spatial arrangement of these modules is suggested to be elongated, but this has not been studied experimentally.

FIG 8

Proposed mechanisms of the epimerization and lyase reaction of AlgE7 on M residues (XMMX substrates).