Effect of Pentacyclic Guanidine Alkaloids from the Sponge Monanchora pulchra on Activity of α-Glycosidases from Marine Bacteria

The effect of monanchomycalin B, monanhocicidin A, and normonanhocidin A isolated from the Northwest Pacific sample of the sponge Monanchora pulchra was investigated on the activity of α-galactosidase from the marine γ-proteobacterium Pseudoalteromonas sp. KMM 701 (α-PsGal), and α-N-acetylgalactosaminidase from the marine bacterium Arenibacter latericius KMM 426T (α-NaGa). All compounds are slow-binding irreversible inhibitors of α-PsGal, but have no effect on α-NaGa. A competitive inhibitor d-galactose protects α-PsGal against the inactivation. The inactivation rate (kinact) and equilibrium inhibition (Ki) constants of monanchomycalin B, monanchocidin A, and normonanchocidin A were 0.166 ± 0.029 min−1 and 7.70 ± 0.62 μM, 0.08 ± 0.003 min−1 and 15.08 ± 1.60 μM, 0.026 ± 0.000 min−1, and 4.15 ± 0.01 μM, respectively. The 2D-diagrams of α-PsGal complexes with the guanidine alkaloids were constructed with “vessel” and “anchor” parts of the compounds. Two alkaloid binding sites on the molecule of α-PsGal are shown. Carboxyl groups of the catalytic residues Asp451 and Asp516 of the α-PsGal active site interact with amino groups of “anchor” parts of the guanidine alkaloid molecules.


Introduction
O-glycoside hydrolases are involved in the degradation of various poly-and oligosaccharides that serve as a source of carbon and energy for organism's growth, as well as performing various functions in organisms. Modification or blocking of these functions by powerful selective inhibitors underlies the treatment of a number of infectious diseases, malignant tumors and genetic disorders [1]. Inhibitors of enzymes are molecules that reduce or completely block the catalytic activity of an enzyme, causing either complete death of a cell or modification in the metabolic pathways. The marine sponges are important sources of enzyme inhibitors [2,3].

Effect of Monanchomycalin B, Monanchocidin A, and Normonanchocidin A on Activity of Two Glycosidases
The results of the pretreatment of two marine bacterial glycosidases with pentacyclic guanidine alkaloids within 30 min (Table 1) showed that all three compounds inhibited the activity of recombinant GH36 α-PsGal and had no effect on the recombinant GH109 α-NaGa. It was previously shown that all three compounds significantly activated SpsLamIV endo-β(1→ 3)-D-glucanase of the mollusk Spisula sachalinensis and completely inhibited ChinLam exo-β(1→ 3)-D-glucanase of the marine fungus Chaetomium indicum [23].
We have shown with the example of monanchomycalin B that pentacyclic guanidine alkaloids irreversibly inactivate the α-PsGal. The activity of the enzyme did not recover after dialysis against the buffer solution for 72 h ( Table 2). The decrease of free enzyme activity by 2.6 times was observed, probably, due to the enzyme α-PsGal thermolability [24] and instability at the low concentrations (data not shown).

Figure 1.
Structural formulas of pentacyclic guanidine alkaloids. "Vessel" part is on the left, and the "anchor" part is on the right of the molecule formula.

Effect of Monanchomycalin B, Monanchocidin A, and Normonanchocidin A on Activity of Two Glycosidases
The results of the pretreatment of two marine bacterial glycosidases with pentacyclic guanidine alkaloids within 30 min (Table 1) showed that all three compounds inhibited the activity of recombinant GH36 α-PsGal and had no effect on the recombinant GH109 α-NaGa. It was previously shown that all three compounds significantly activated SpsLamIV endo-β(1→3)-D-glucanase of the mollusk Spisula sachalinensis and completely inhibited ChinLam exo-β(1→3)-D-glucanase of the marine fungus Chaetomium indicum [23].
We have shown with the example of monanchomycalin B that pentacyclic guanidine alkaloids irreversibly inactivate the α-PsGal. The activity of the enzyme did not recover after dialysis against the buffer solution for 72 h ( Table 2). The decrease of free enzyme activity by 2.6 times was observed, probably, due to the enzyme α-PsGal thermolability [24] and instability at the low concentrations (data not shown).
The study of the inhibitory effect of monanchomycalin B, monanchocidin A, and normonanhocidin A at different concentrations and incubation times showed that the IC 50 values of compounds decreased with increasing of the incubation time of α-PsGal with inhibitors (data not shown). The results of kinetic studies on the α-PsGal inactivation by pentacyclic guanidine alkaloids are shown on Figure 2. The curves of the dependences of the residual activity v/v 0 on the time in semilogarithmic coordinates are shown in Figure 2a,c,e.  The study of the inhibitory effect of monanchomycalin B, monanchocidin A, and normonanhocidin A at different concentrations and incubation times showed that the IC50 values of compounds decreased with increasing of the incubation time of α-PsGal with inhibitors (data not shown). The results of kinetic studies on the α-PsGal inactivation by pentacyclic guanidine alkaloids are shown on Figure 2. The curves of the dependences of the residual activity v/vo on the time in semilogarithmic coordinates are shown in Figure 2a,c,e.  The α-PsGal inactivation developed relatively slowly, within a few minutes under these experimental conditions. In this case, the inhibitory activity of the compounds can be more accurately described by the inactivation rate constant (k inact , min -1 ) and equilibrium inhibition constant K i [25]. The values of k obs increased together with the compound concentrations. Sigmoid curves of k obs dependences on concentration of the inhibitors (Figure 2b,d,f) mean that the process of the enzyme (E) inactivation by slowly-binding irreversible inhibitors (I) has a cooperative character, and occurs in two stages: (i) the formation of a reversible enzyme-inhibitor complex [E I n ] and (ii) irreversible inactivation of the enzyme in the E-I n complex. The kinetic Equation (1) describes the irreversible slow inhibition of α-PsGal under the action of the pentacyclic guanidine alkaloids: where n is coefficient of cooperativity, which is interpreted as the number of identical binding sites; K i is an equilibrium constant of inhibition (µM). The experimental dependences of k obs on the concentration of the compounds (I) (Figure 2b,d,g) are approximated by the Hill's Equation (2).
The results of the experimental data fitting with theoretical curves are shown in Table 3. According to the values of K i and k inact , the alkaloids can be arranged in descending of binding-affinity as normonanchocidin A > monanchomycalin B > monanchocidin A, but in increasing inactivation rate in the following order: monanchomycalin B > monanchocidin A> normonanchocidin A.
Thus, based on the results of the kinetic studies, we have suggested that the pentacyclic guanidine alkaloids are slow-binding inhibitors for α-PsGal similarly to the ChinLam glucanase [23]. It is accepted that slow-binding inhibition is observed whenever an enzyme-inhibitor complex forms or undergoes further conversion, at a slower rate relative to the overall reaction rate [26]. Inhibitors of peptidases [27], monoamine oxidases, and acetylcholinesterases [28,29] are examples of the slow-binding. Previously, the property of monanchocidin A as a slow-acting biologically active compound was shown for cancer cells [30]. For α-PsGal, the chlorine and bromine echinochrome derivatives from a sea urchin have been previously shown to be slow-binding inactivators as well [31]. Moreover, we have found that the inhibition rate increases with the binding of at least four molecules of the compounds. D-galactose being a competitive inhibitor for α-galactosidases of GH36 family [32,33] decreased the activity of α-PsGal on 50% at 0.7 mM. The active-site-directed nature of the inactivation was proven by demonstration of the enzyme's protection against inactivation by D-galactose (Figure 3). From the Figure 3, it is evident that this monosaccharide significantly protects α-PsGal from the inactivation.
Regardless of the inhibitor concentration, k obs Gal decreased on average by 50% in the presence of the reaction product D-galactose ( Table 4). The monosaccharide partially protects the enzyme from inactivation. This suggests that the inhibitor interacts with the enzyme molecule in the region of the active center.  From the Figure 3, it is evident that this monosaccharide significantly protects α-PsGal from the inactivation. Regardless of the inhibitor concentration, kobs Gal decreased on average by 50% in the presence of the reaction product D-galactose ( Table 4). The monosaccharide partially protects the enzyme from inactivation. This suggests that the inhibitor interacts with the enzyme molecule in the region of the active center.
Taking into account that the active center of the enzyme and the "vessel" part of the molecules of the compounds are identical in all the experiments, their inhibitory properties towards α-PsGal are determined by the structure of the "anchor" part. In this case, the diaminopropane residue has the greatest affinity, but more slowly penetrates to the active center of the GH36 α-PsGal from the marine bacterium. The monosubstituted diaminopropane has been shown to be also the best inhibitor for the ChinLam exo-(1→3)β-D-glucanase from a marine fungus as well [23]. However, these compounds did not show inhibitor properties towards the GH109 α-NaGa from the marine bacterium A. latericius as well as the GH16 endo-(1→3)β-D-glucanase from the marine bivalve mollusk S. sachalinensis [23].

Theoretical Model of the Guanidine Alkaloids Complexes with α-Galactosidase
The enzyme α-PsGal is a typical O-glycoside hydrolase of the GH36 family. It was previously shown that its molecule consists of two identical subunits [9,10]. One subunit is a three-domain protein. The active center is located in the central (β/α)8 domain. Asp 451 and Asp 516 are catalytic residues [24]. Figure 4 shows 2D-diagrams of the α-PsGal complexes with the guanidine compounds. The "vessel" part identical for the all compounds (Figure 4a), and the "anchor" parts of monanchomycalin B (Figure 4b   Taking into account that the active center of the enzyme and the "vessel" part of the molecules of the compounds are identical in all the experiments, their inhibitory properties towards α-PsGal are determined by the structure of the "anchor" part. In this case, the diaminopropane residue has the greatest affinity, but more slowly penetrates to the active center of the GH36 α-PsGal from the marine bacterium. The monosubstituted diaminopropane has been shown to be also the best inhibitor for the ChinLam exo-(1→3)β-D-glucanase from a marine fungus as well [23]. However, these compounds did not show inhibitor properties towards the GH109 α-NaGa from the marine bacterium A. latericius as well as the GH16 endo-(1→3)β-D-glucanase from the marine bivalve mollusk S. sachalinensis [23].

Theoretical Model of the Guanidine Alkaloids Complexes with α-Galactosidase
The enzyme α-PsGal is a typical O-glycoside hydrolase of the GH36 family. It was previously shown that its molecule consists of two identical subunits [9,10]. One subunit is a three-domain protein. The active center is located in the central (β/α) 8 domain. Asp 451 and Asp 516 are catalytic residues [24]. Figure 4 shows 2D-diagrams of the α-PsGal complexes with the guanidine compounds. The "vessel" part identical for the all compounds (Figure 4a), and the "anchor" parts of monanchomycalin B (Figure 4b   The "anchor" parts of the compounds are the spermidine residue in monachomycalin B, tetra-substituted morpholinone derivative in monanchocidine A, and monosubstituted diaminopropane in normonanchocidine A.
Two different binding sites for the "vessel" and "anchor" parts of alkaloids in the molecule of α-PsGal were found. The carboxyl groups of the catalytic residues Asp451 and Asp516 in the active site of α-PsGal take part directly in the interaction with amino groups of "anchor" parts of the compounds.
The molecules of the test compounds consist of two polar nitrogen-containing residues connected by hydrophobic polymethylene chains. In this case, the "anchor" part of the molecule is very mobile. Based on the simulation results, the "vessel" part of the molecule binds near the crater of the active center and does not influence the activity of the enzyme, but directs and promotes an increase in the affinity of the "anchor" part; thus, the binding of the latter occurs more slowly and leads to the loss of enzyme activity. D-galactose located in the active center prevents the spermidine residue of monachomycalin B from entering to the catalytic site what can slow down the inactivation of the enzyme ( Figure S1). In accordance with the results of a 3D-superposition of the α-PsGal active site with D-galactose and anchor parts of the compounds, the monosubstituted diaminopropane of normonanchocidine A penetrates most deeply into the pocket of the active center of α-PsGal ( Figure  S1(c)).  The "anchor" parts of the compounds are the spermidine residue in monachomycalin B, tetra-substituted morpholinone derivative in monanchocidine A, and monosubstituted diaminopropane in normonanchocidine A.

Materials
Two different binding sites for the "vessel" and "anchor" parts of alkaloids in the molecule of α-PsGal were found. The carboxyl groups of the catalytic residues Asp451 and Asp516 in the active site of α-PsGal take part directly in the interaction with amino groups of "anchor" parts of the compounds.
The molecules of the test compounds consist of two polar nitrogen-containing residues connected by hydrophobic polymethylene chains. In this case, the "anchor" part of the molecule is very mobile. Based on the simulation results, the "vessel" part of the molecule binds near the crater of the active center and does not influence the activity of the enzyme, but directs and promotes an increase in the affinity of the "anchor" part; thus, the binding of the latter occurs more slowly and leads to the loss of enzyme activity. D-galactose located in the active center prevents the spermidine residue of monachomycalin B from entering to the catalytic site what can slow down the inactivation of the enzyme ( Figure S1). In accordance with the results of a 3D-superposition of the α-PsGal active site with D-galactose and anchor parts of the compounds, the monosubstituted diaminopropane of normonanchocidine A penetrates most deeply into the pocket of the active center of α-PsGal ( Figure S1c).

Production and Purification of Recombinant α-D-galactosidase
The recombinant wild-type α-D-galactosidase α-PsGal was produced as described earlier [35]. The plasmid DNA pET-40b(+) containing insertion of the gene from the marine bacterium Pseudoslteromonas sp. KMM 701 encoding α-PsGal was transformed in the Escherichia coli strain Rosetta (DE3). Heterological expression was carried out at optimal conditions as described previously [36]. Purification of the recombinant α-PaGal was performed according to the procedures described in the reference [35].

Production and Purification of Recombinant α-Nacetylgalactosaminidase
The recombinant wild-type α-Nacetylgalactosaminidase α-NaGa was produced as described earlier [35]. The plasmid DNA pET-40b(+) containing insertion of the gene from the marine bacterium Arenibacter latericius KMM 426 T encoding α-NaGa was transformed in the Escherichia coli strain Rosetta (DE3). Heterological expression was carried out at optimal conditions as described previously [37].
The new purification procedure was modified and carried out at 4 • C. The cleared supernatant containing α-NaGa and 20% glycerol was loaded directly onto a Ni-sepharose column (5 cm × 36 cm) equilibrated with the buffer A (10 mM NaH 2 PO 4 , 10 mM Na 2 HPO 4 , 0.5 M NaCl, 5 mM imidazole, 20% glycerol, pH 8.0). The recombinant protein was eluted with the 5-500-mM linear imidazole gradient. The eluted fractions were analyzed, collected and dialyzed against the buffer B (10 mM NaH 2 PO 4 , 10 mM Na 2 HPO 4 , 50% glycerol, pH 8.0). Then, the protein solution was loaded onto a column (2 cm × 15 cm) with an ion-exchange resin Source 15Q equilibrated with the buffer B. The recombinant protein was eluted with the 0-1.5 M linear NaCl gradient. The fractions exhibiting the activity of α-NaGa were collected and examined using sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE).

Enzyme and Protein Assays
The activity of α-PsGal and α-NaGa were determined by increasing the amount of p-nithrophenol (pNP). The mixtures containing 50 µL of an enzyme solution and 100 µL of a substrate solution (1 mg/mL) in 0.05 M sodium phosphate buffer (pH 7.0) were incubated at 20 • C during 5 min for α-PsGal and 30 min for α-NaGa. The reactions were stopped by the addition of 150 µL of 1 M Na 2 CO 3 . One unit of the activity (U) was determined as the amount of an enzyme that releases 1 µmol of pNP per 1 min at 20 • C. The amount of the released pNP was determined spectrophotometrically (ε 400 = 18,300 M −1 cm −1 ). The specific activity was calculated as U/mg of protein. The protein concentration was determined by the Bradford method calibrated with BSA as a standard [38]. Buffer solutions of α-PsGal (0.1 U/mL) and α-NaGa (0.05 U/mL) were used in the further experiments.

The Effects of Monanchomycalin B, Monanchocidin A, and Normonanchocidin A on Glyctosidases
To study the effect of pentacyclic guanidine alkaloids on α-PsGal and α-NaGa, 25 µL of an aqueous solution of monanchomycalin B, monanchocidin A, or normonanchocidin A (1 mg/mL) was mixed with 50 µL of the enzyme solutions in wells of the 96-cell plates and incubated for 30 min. Reactions were initiated by the addition of 75 µL of substrate solutions (p-nithrophenil galactopyranoside (pNP-α-Gal) for α-PsGal and p-nithrophenil N-acetylgalactosaminide (pNP-α-NAcGal) for α-NaGa) in 0.05 M sodium phosphate buffer (pH 7.0). The reaction mixture was incubated at 20 • C for 2-30 min in the final volume 150 µL, and then 150 µL of 1M Na 2 CO 3 solution was added to the incubation mixture to stop the reaction. Each reaction mixture was prepared in duplicate. The absorbance was measured at 400 nm. Results were read with a Gen5 and treated with ExCel software. The activity of α-PsGal or α-NaGa was determined as described above. The residual activity was calculated as the ratio v/v 0 (%), where v is the enzyme activity in the presence of an inhibitor, and v 0 is the enzyme activity in the absence of an inhibitor. The v 0 was taken for 100%.

The Irreversibility of Monanchomycalin B Inhibition
To determine the reversibility of the inhibition of the α-PsGal activity, 40 µL (18 µM, H 2 O) of the monanchomycalin B solution was added to 60 µL of the enzyme solution; the mixture was incubated for 60 min. Two volumes of 20 µL were taken from the reaction mixture, 380 µL of a pNP-α-Gal solution (3.32 mM, in probe was~5 K m ) was added to each mixture, and then the reaction was stopped by addition of 0.6 mL of 1M Na 2 CO 3 after 30 min of incubation. The value of optical density at the wavelength 400 (OD 400 ) was measured in the 1-cm cuvette. The activity of α-PsGal was determined as described above. The remaining 60 µL of the reaction mixture was dialyzed against 1 L of 0.02 M sodium phosphate buffer (pH 7.0) for 72 h at 4 • C. To estimate the dilution, the volume of the reaction mixture after dialysis was measured. The enzyme activity was determined as described above and recalculated taking the dilution into account (1.7 times). A sample of α-PsGal untreated by monanchomycalin B (60 µL of the enzyme solution and 40 µL of H 2 O) was used as a control. The experiment was carried out in two replicates. The residual activity was calculated as described above. were added to stop the reaction, and OD 400 for the reaction mixtures were immediately measured by a microplate spectrophotometer. The time of each reaction was strictly monitored by stopwatch. The standard and residual activity v/v 0 were calculated as described above.

The Kinetic Parameters of Inactivation
The equilibrium inhibition constants (K i ) and kinetic inactivation constants (k inact ) were determined by the classical methods [39]. The inactivation of α-PsGal by the different concentrations of inhibitors (1.3-50 µM) was performed in 0.05 M sodium phosphate buffer (pH 7.0) at a temperature 20 • C. An aqueous solution (25 µL) of the compound at the different concentrations was added to 50 µL of the α-PsGal solution (0.2 U/mL), held for 5, 10, 15, 20, and 25 min at 20 • C, then 100 µL of the pNP-α-Gal solution was added and incubated for 2-15 min at 20 • C. The same conditions were used in the control reaction, but the inhibitor was replaced with distilled water. The reactions were stopped by the addition of 1 M Na 2 CO 3 (150 µL); the amount of pNP formed in 1 min was determined as described above. The residual activity v/v 0 was presented as a function of time. The pseudo-first-order rate constant of inactivation (k obs ) was determined for each inactivator concentration as the slope of the v/v 0 dependence on the incubation time in semilogarithmic coordinates. The ExCel software was used for these calculations. The second order rate constants for the inactivation process were determined by fitting the dependences of the k obs values on the concentration of the inactivators to the Hill's equations. An analysis of the curves and the choice of models for calculation of K i (µM) and k inact (min -1 ) were performed with the Origin 8.1 software (OriginLab, Northampton, MA, USA).

Protection of α-PsGal Inactivation by D-galactose
The active-site-directed nature of the inactivation was confirmed by demonstrating protection against the inactivation by competitive inhibitor D-galactose. Inactivation mixtures (75 µL) containing 50 µL of the enzyme solution and 10 µL of D-galactose (0.7 mM in mixture) were preincubated for 15 min, then 15 µL of the monanchomycalin B solution (11.4 µM and 14.2 µM in mixture) were added and incubated at various time intervals as described above. The residual activity of the enzyme was assayed as described above.

Theoretical Models of α-PsGal Complexes with Guanidine Alkaloids
The target-template alignment customization of the modeling process and 3D model building of α-PsGalA (GenBank: ABF72189.2) were carried out using the Molecular Operating Environment version 2018.01 [37] package using the forcefield Amber12: EHT. The α-D-galactosidase from Lactobacillus acidophilus NCFM (Protein data bank (PDB) code: 2XN2) with a high-resolution crystal structure was used as a template. The evaluation of structural parameters, contact structure analysis, physical-chemical properties, molecular docking, and visualization of the results were carried out with the Ligand interaction and Dock modules in the MOE 2018.01 program. The results were obtained using the equipment of Shared Resource Center Far Eastern Computing Resource of Institute of Automation and Control Processes Far Eastern Branch of the Russian Academy of Sciences (IACP FEB RAS) [40].

Conclusions
For study the effect of the marine sponge metabolites with a therapeutic potential, we used two well-characterized α-glycosidases for justifying a possible mechanism of their inhibitor action. Monanchomycalin B, normonanchocidin A, monanchocidin A have been shown to be irreversible slow-binding inhibitors of the GH36 family α-galactosidase α-PsGal from the marine bacterium Pseudoalteromonas sp. KMM 701, but have no effect on the activity of the GH109 family α-NaGa from the marine bacterium Arenibacter latericius KMM 426 T . The inhibitory ability of the alkaloids depends on the chemical structure of the anchor parts of their molecules. The alkaloids can be arranged in the descending order of the binding-affinity: normonanchocidin A > monanchomycalin B > monanchocidin A, and in the decreasing order of the inactivation rate: monanchomycalin B > monanchocidin A > normonanchocidin A. These highly active marine compounds selectively acted on the enzymes from the different structural GH families, binding to the electronegative areas of the protein surfaces formed mainly by carboxylic acid side groups in the active-site-directed manner. The well-characterized α-glycosidases of marine bacteria have been proved to be suitable models for characterizing the novel properties of the alkaloids.
Supplementary Materials: The following are available online at http://www.mdpi.com/1660-3397/17/1/22/s1, Figure S1. 3D-superposition of the α-PsGal active site with D-galactose and spermidine residue of monachomycalin B (a), tetra-substituted morpholinone derivative of monanchocidine A (b) and monosubstituted diaminopropane of normonanchocidine A (c). Parts of guanidine alkaloids are shown as "ball and stick" with grey color, galactose shown as "stick" with yellow color. The molecular surface closed to the ligands is shown in pink (H-bonding), green (hydrophobic) and blue (mild polar).