ARTICLE pubs.acs.org/JPCB
Isothermal Titration Calorimetric and Computational Studies on the Binding of Chitooligosaccharides to Pumpkin (Cucurbita maxima) Phloem Exudate Lectin Akkaladevi Narahari,† Hitesh Singla,‡ Pavan Kumar Nareddy,† Gopalakrishnan Bulusu,‡,§ Avadhesha Surolia,|| and Musti J. Swamy†,* †
School of Chemistry, University of Hyderabad, Hyderabad 500 046, India Centre for Computational Natural Sciences and Bioinformatics, International Institute of Information Technology, Hyderabad 500032, India § TCS Innovation Labs, Tata Consultancy Services, Hyderabad 500081, India National Institute of Immunology, Aruna Asaf Ali Marg, New Delhi 110067, India
)
‡
bS Supporting Information ABSTRACT: The interaction of chitooligosaccharides [(Glc NAc)2-6] with pumpkin phloem exudate lectin (PPL) was investigated by isothermal titration calorimetry and computational methods. The dimeric PPL binds to (GlcNAc)3-5 with binding constants of 1.26-1.53 105 M-1 at 25 °C, whereas chitobiose exhibits approximately 66-fold lower affinity. Interestingly, chitohexaose shows nearly 40-fold higher affinity than chitopentaose with a binding constant of 6.16 106 M-1. The binding stoichiometry decreases with an increase in the oligosaccharide size from 2.26 for chitobiose to 1.08 for chitohexaose. The binding reaction was essentially enthalpy driven with negative entropic contribution, suggesting that hydrogen bonds and van der Waals’ interactions are the main factors that stabilize PPL-saccharide association. The three-dimensional structure of PPL was predicted by homology modeling, and binding of chitooligosaccharides was investigated by molecular docking and molecular dynamics simulations, which showed that the protein binding pocket can accommodate up to three GlcNAc residues, whereas additional residues in chitotetraose and chitopentaose did not exhibit any interactions with the binding pocket. Docking studies with chitohexaose indicated that the two triose units of the molecule could interact with different protein binding sites, suggesting formation of higher order complexes by the higher oligomers of GlcNAc by their simultaneous interaction with two protein molecules.
’ INTRODUCTION Lectins are carbohydrate-specific proteins that are ubiquitous in living organisms. They exhibit a characteristic ability to bind carbohydrates in a specific manner but in general do not have catalytic activity.1 Lectins are useful as molecular tools for detection and separation of glycoconjugates, typing blood, cell fractionation, mitogenic stimulation of lymphocytes, preferential agglutination of tumor cells, and bone marrow transplantation.2,3 All these biological activities of lectins are manifested through their ability to specifically and reversibly bind particular carbohydrate structures, although several lectins also interact with noncarbohydrate ligands.4 It is therefore important to investigate lectin-carbohydrate interactions in detail and characterize the associated thermodynamic and kinetic parameters in order to understand their biological functions. A number of techniques such as fluorescence spectroscopy, equilibrium dialysis, surface plasmon resonance, etc., have been employed to characterize the thermodynamic parameters governing lectin-carbohydrate interactions. However, the limitation with all these techniques is that they yield thermodynamic data associated with the binding r 2011 American Chemical Society
process by indirect approaches. Isothermal titration calorimetry (ITC), on the other hand, directly measures energetics associated with the binding reaction and yields accurate information on various thermodynamic parameters associated with the binding process, viz., binding constant, stoichiometry, enthalpy, and entropy from a single experiment.5 In addition, varying the temperature of the experiment allows the determination of heat capacity (ΔCp) for the reaction. Cucurbitaceae phloem lectins are a family of chitooligosaccharide-specific lectins found in the phloem exudate of the fruits/ sieve of various species of this family. They are not related to other Cucurbitaceae lectins such as those present in the seeds and do not have a hevein domain.6 The occurrence of high heamagglutinating activity in the phloem exudate of several cucurbit species has been reported, and lectins from some of them have been well characterized.7-9 Some of these proteins Received: November 2, 2010 Revised: February 17, 2011 Published: March 15, 2011 4110
dx.doi.org/10.1021/jp110468n | J. Phys. Chem. B 2011, 115, 4110–4117
The Journal of Physical Chemistry B have been shown to bind to a variety of RNAs also. While the carbohydrate binding activity may be involved in defense mechanisms of the plant against pests and pathogens,6,10 the RNA binding activity appears to play a role in long distance movement of plant RNAs and viroids.11-13 The present study is focused on delineating the thermodynamic forces that characterize the binding of chitooligosaccharides to a representative phloem exudate lectin from the Cucurbitaceae, namely, the pumpkin (Cucurbita maxima) lectin. Thermodynamic studies employing ITC were reported on the interaction of chitooligosaccharides with Urtica dioica agglutinin (UDA) and wheat germ agglutinin (WGA).14-16 ITC studies have also been carried out on the binding of chitooligosaccharides [(GlcNAc)3-6] to the AVR4 elicitor of Cladosporium fulvum and on the association of chitooligosaccharides with hevein.17-19 These studies showed that the smallest ligand recognized by WGA and hevein is the monomer (GlcNAc) and that UDA shows detectable binding only with a disaccharide [(GlcNAc)2], whereas the AVR4 elicitor was reported to bind chitotriose repeats, i.e., it requires at least a trisaccharide to exhibit appreciable binding. However, the interaction of chitooligosaccharides with Cucurbitaceae phloem exudate lectins has not been investigated so far by this technique. Pumpkin phloem exudate lectin (PPL) is a chitooligosaccharide-specific, homodimeric protein with a mass of 48 kDa and a subunit mass of 26 kDa.20 The two subunits are covalently linked by buried disulfide bonds, since under nondenaturing conditions reducing agents could not dissociate the dimer.20 Analysis of its cDNA sequence has shown that PPL is a polypeptide of 218 amino acids.21,22 Recently, we reported a rapid affinity method for the purification of this protein and characterized its secondary structure by CD spectroscopy, which indicated that PPL is a predominantly β-sheet protein.23 Fluorescence quenching and timeresolved fluorescence studies indicated that tryptophan residues in PPL are significantly exposed to the aqueous environment.24 In the present study thermodynamics of the interaction of chitooligosaccharides with PPL has been investigated by ITC and computational approaches. The results obtained revealed novel features wherein the lectin exhibited slightly higher affinity toward chitotriose as compared with chitotetraose and chitopentaose, whereas chitohexaose was recognized with ∼40-fold higher affinity than chitotriose with a binding stoichiometry that is approximately one-half of that observed for the latter, which is consistent with two triose repeats from one hexaose molecule binding to two different molecules of the lectin. Molecular modeling, ligand docking, and molecular dynamics (MD) studies yielded results that are in very good agreement with the calorimetric data, especially with regard to binding stoichiometry.
ARTICLE
Figure 1. Calorimetric titration of PPL with (A) chitotriose and (B) chitopentaose. Upper panels show the raw ITC data obtained from 30 automatic injections of 7 μL aliquots of 2 mM chitotriose and chitopentaose into 60 μM of PPL. Lower panels show the integrated heats of binding obtained from raw data shown in the upper panels. T = 298 K.
presence of β-mercaptoethanol, corresponding to the monomer.20 Lectin concentration was estimated according to Petersen.25 Isothermal Titration Calorimetry. Calorimetric titrations were performed on a VP-ITC isothermal titration calorimeter from MicroCal (Northampton, MA), essentially as described earlier.26 Briefly, 7-10 μL aliquots of a 2.0-5.5 mM chitooligosaccharide solution were added at 4 min intervals via a rotating stirrer syringe to a 60-100 μM lectin solution contained in a 1.445 mL sample cell. Samples were dialyzed extensively against PBS-βME and degassed prior to loading into the cell. Since the first injection was often found to be inaccurate, a 2 μL injection was added first and the resultant point was deleted before the remaining data were analyzed using the ‘one set of sites’ model in the MicroCal Origin ITC analysis software as described earlier.26,27 The analysis yielded values of the following parameters: number of binding sites (n), binding constant for the interaction (Kb), and enthalpy of binding (ΔHb). From these values, free energy of binding (ΔGob) and entropy of binding (ΔSb) were calculated according to the following basic thermodynamic equations ΔGo b ¼ - RT ln Kb ð1Þ
’ EXPERIMENTAL METHODS Materials. Pumpkin fruits were obtained from local vendors. 2-Mercaptoethanol, chitin (from crab shells), and all the chitooligosaccharides were obtained from Sigma Chemical Co. (St. Louis, MO). All other chemicals and reagents were obtained from local suppliers and of the highest purity available. Pumpkin Phloem Exudate Lectin. PPL was purified by affinity chromatography on chitin as described previously.22 The affinity-eluted lectin was dialyzed thoroughly against 20 mM sodium phosphate buffer, pH 7.4, containing 150 mM sodium chloride and 10 mM β-mercaptoethanol (PBS-βME). The lectin thus obtained gave a single band at ∼24 kDa in SDS-PAGE in the
ΔGo b ¼ ΔHb - TΔSb
ð2Þ
Molecular Modeling. The three-dimensional structure of PPL was predicted by homology modeling, and binding of chitooligosaccharides was investigated by molecular docking and molecular dynamics simulations. The procedures followed included the following steps: homology modeling of the monomer, refinement of the modeled structure and its evaluation, generation of the dimer, prediction of the ligand binding pocket, and molecular dynamics simulations. Detailed information of the procedures followed are provided in the Supporting Information. 4111
dx.doi.org/10.1021/jp110468n |J. Phys. Chem. B 2011, 115, 4110–4117
The Journal of Physical Chemistry B
ARTICLE
Table 1. Binding Constants (Kb) and Thermodynamic Parameters for the Association of Chitooligosaccharides with PPL -ΔH° (kcal mol-1)
-ΔS° (cal mol-1 K-1)
(GlcNAc)2
288
2.0a
0.047
4.84
11.65
23.6
293
2.0a
0.033
4.72
11.53
23.3
298
2.0a
0.023
4.58
11.63
23.3
288
2.22
3.24
7.27
15.68
29.2
293
2.17 ((0.0)
2.41 ((0.04)
7.21
16.17 ((0.03)
30.55 ((0.15)
298
1.96 ((0.07)
1.53 ((0.19)
7.06
16.53 ((0.33)
31.73 ((1.35)
288
1.84
1.26
6.73
20.69
48.5
293 298
1.85 1.82
1.10 1.26
6.77 6.84
21.10 21.80
48.9 50.2
288
1.59
1.65
6.88
21.63
51.2
293
1.6 ((0.07)
1.67 ((0.27)
7.12
22.3 ((0.77)
52.15 ((2.95)
298
1.57 ((0.07)
1.43 ((0.2)
7.03
23.73 ((0.78)
56.0 ((2.9)
288
1.06 ((0.02)
90.1 ((4.9)
9.18
28.73 ((0.28)
67.9 ((0.9)
293
1.24 ((0.08)
79.4 ((12.8)
9.25
29.8 ((0.35)
70.0 ((0.9)
298
1.08 ((0.09)
61.6 ((2.2)
9.26
30.7 ((0.3)
71.9 ((1.1)
(GlcNAc)4
(GlcNAc)5
(GlcNAc)6
a
-ΔG° (kcal mol-1)
T (K)
(GlcNAc)3
N
Kb 10-5 (M-1)
sugar
Stoichiometry fixed during the fit.
’ RESULTS Thermodynamics of Carbohydrate Binding. In the present study, thermodynamic parameters associated with the binding of chitooligosaccharides to the pumpkin phloem exudate lectin have been characterized by ITC at different temperatures between 288 and 298 K. Results of representative calorimetric titrations obtained for the binding of chitotriose and chitopentaose to PPL at 25 °C are shown in Figure 1A and 1B, respectively. In these experiments 7 μL aliquots of a 2 mM solution of the appropriate chitooliogosaccharide (chitotriose to chitohexaose) were added to a 1.445 mL sample of 60 μM PPL in the calorimeter cell at intervals of 4 min. In the case of chitobiose, however, 10 μL aliquots from a 5 mM ligand solution were added to 90-100 μM protein, in view of the ∼66-fold lower affinity of the protein for this ligand as compared to chitotriose. In Figure 1A and 1B, the upper panels show the exothermic heat released upon binding at each injection, which decreases monotonically with successive injections until saturation is achieved. A plot of the incremental heat released as a function of chitotriose and chitopentaose to PPL monomer ratio is shown in the corresponding lower panels, together with nonlinear leastsquares fits of the data to ‘one set of sites’ model (shown as solid lines). The high quality of the fits in these figures indicates that the experimental data could be described well by this model. The fit in Figure 1A (lower panel) yielded the values of various parameters associated with the binding of chitotriose to PPL: number of binding sites, n = 2.06 ((0.012); binding constant, Kb = 1.81 ((0.1) 105 M-1; enthalpy of binding, ΔHb = -16.95 ((0.129) kcal mol-1; entropy of binding, ΔSb = -33.3 cal mol-1 K-1. Three independent titrations yielded the average values of these parameters as follows: n = 1.96 ((0.07); binding constant, Kb = 1.53 ((0.19) 105 M-1; enthalpy of binding, ΔHb = -16.53 ((0.33) kcal mol-1; entropy of binding, ΔSb = 31.73 ((1.35) cal mol-1 K-1. From the fit shown in Figure 1B the corresponding parameters for the binding of chitopentaose to PPL were obtained as follows: n = 1.5 ((0.034), Kb = 1.23 ((0.18) 105 M-1, ΔHb = -24.5 ((0.075) kcal mol-1, ΔSb = -58.9 cal mol-1 K-1. Two independent titrations yielded the following average values: n = 1.57 ((0.07), Kb = 1.43 ((0.2)
Figure 2. Dependence of binding stoichiometry on the size of oligosaccharide for the association of chitooligosaccharides with PPL. Stoichiometry (n) corresponds to the number of ligand molecules that bind to each molecule of PPL dimer. T = 298 K.
105 M-1, ΔHb = -23.73 ((0.78) kcal mol-1, ΔSb = -56.0 ((2.9) cal mol-1 K-1. These values as well as the corresponding values obtained for the calorimetric titrations performed with the other chitooligosaccharides, namely, chitobiose, chitotetraose, and chitohexaose, at various temperatures are listed in Table 1. In addition, values of Gibbs free energy (ΔGob), calculated from the Kb values using eq 1, are also listed in Table 1. From the results presented in this table it is interesting to note that the binding stoichiometry obtained for different oligosaccharides decreases with increasing ligand size. When the observed stoichiometry was plotted as a function of the number of GlcNAc units in the oligosaccharide, a linear dependence was observed up to chitotetraose, whereas the stoichiometry decreased more steeply for chitopentaose and chitohexaose (Figure 2). The association constants determined at 25 °C exhibit interesting trends. The Kb value increases from 2.3 103 M-1 for chitobiose to 1.53 105 M-1 for chitotriose, i.e., a 66-fold increase. However, the Kb value then decreases marginally for chitotetraose and chitopentaose, whereas a 43-fold 4112
dx.doi.org/10.1021/jp110468n |J. Phys. Chem. B 2011, 115, 4110–4117
The Journal of Physical Chemistry B
ARTICLE
Table 2. Changes in Enthalpy, Entropy, and Free Energy per Additional Saccharide for the Binding of Chitooligosaccharides to PPL at 298 K sugara
-ΔΔH° (kcal mol-1)
-ΔΔS° (cal mol-1 K-1)
-ΔΔG° (kcal mol-1)
(GlcNAc)2
11.63
23.30
4.58
(GlcNAc)3 (GlcNAc)4
4.90 5.63
8.43 18.47
2.48 -0.22
(GlcNAc)5
1.93
5.80
0.19
(GlcNAc)6
6.97
15.90
2.23
Values for (GlcNAc)2 are in comparison for PPL alone. Hence, the -ΔΔH°, ΔΔS°, and -ΔΔG° values are for the two GlcNAc residues of the disaccharide. a
increase is seen for the binding of chitohexaose as compared to chitopentaose. The data presented in Table 1 also indicates that the enthalpy of binding and entropy of binding increase with increasing size of the chitooligosaccharides. The increase in enthalpy with increasing number of GlcNAc residues in the ligand indicates that the lectin combining site contains several subsites which interact with the individual monosaccharide units of the oligosaccharide. In order to investigate this further the contributions of the different monosaccharide units of the chitooligosaccharides to the binding enthalpy, entropy, and free energy have been calculated from the thermodynamic data presented in Table 1 by subtracting the values corresponding to the oligosaccharide containing ‘(n - 1)’ monosaccharide units from the values corresponding to the oligosaccharide containing ‘n’ GlcNAc residues. The results obtained are presented in Table 2. From the data presented in Table 2 it can be seen that addition of the third, fourth, fifth, and sixth N-acetylglucosamine residues to chitobiose increases the enthalpy of binding by 4.9, 5.63, 1.93, and 6.97 kcal/mol, respectively. This shows that while contributions from the third and fourth GlcNAc residues toward the overall binding enthalpy are nearly the same, a significant decrease is seen for the contributions of the fifth GlcNAc residue. However, due to a larger negative contribution from the entropy of binding from the fourth GlcNAc residue, the Kb value decreases for chitotetraose as compared to chitotriose. The incremental enthalpy due to the fifth GlcNAc residue is almost nearly compensated by the incremental entropy due to it, resulting in only a marginal increase in the Kb value for chitopentaose as compared to chitotetraose. Addition of the sixth GlcNAc residue, however, leads to a significant increase in the enthalpy of binding, whereas the corresponding increase in the entropy of binding is relatively smaller, which leads to an approximately 43-fold increase in the association constant for chitohexaose over chitopentaose. Effect of Temperature on the Thermodynamic Parameters. In order to investigate the effect of temperature on the thermodynamic parameters, ITC measurements were carried out at three different temperatures for all the chitooligosaccharides investigated. The data presented in Table 1 show that binding enthalpies for all the saccharides except chitobiose increase slightly with temperature. The lack of change in the enthalpy for the binding of chitobiose may be ascribed to the weak affinity of PPL for this ligand, due to which the enthalpy estimate is not very reliable. Plots of ΔHb versus temperature (Figure 3) indicate that ΔHb values for all the saccharides exhibit linear dependence, indicating that ΔCp is temperature independent in the range studied. From the slopes of the linear fits shown in Figure 3 the ΔCp values for chitotriose, chitotetraose, chitopentaose, and chitohexaose were estimated as -85 ((8), -111
Figure 3. Temperature dependence of enthalpy of binding for the association of chitooligosaccharides with PPL. The ligands investigated are chitobiose (4), chitotriose (b), chitotetraose ([), chitopentaose (0), and chitohexaose (2). From the slope of linear least-squares fits the ΔCp value is obtained for each saccharide.
((17), -211 ((43), and -197 ((10) cal mol-1 K-1, respectively. It is interesting to note that the ΔCp values for chitotriose and chitotetraose are relatively closer, whereas the corresponding values for chitopentaose and chitohexaose are significantly larger. In the absence of any protonation/deprotonation occurring during the binding process or major conformational changes resulting from ligand binding, the ΔCp values are attributed to hydration and dehydration of apolar and polar groups, which have been shown to correlate well with changes in the solventaccessible surface areas.26,28,29 This suggests that the amount of water displaced from the surface of the protein and ligand due the binding of these two ligands is comparable, which is consistent with a rather shallow binding site on PPL. On the other hand, the ΔCp value for the binding of chitopentaose and chitohexaose, which are expected to bind to two different protein molecules simultaneously, is significantly larger. This is consistent with displacement of a larger amount of water due to removal of water molecules from the binding sites on the surface of two different protein molecules. Identification of the ligand binding site formed by several loops on the protein surface by the computational studies reported in this study is consistent with the above observations. Homology Modeling of PPL. In order to understand whether larger chitooligosaccharides such as chitopentaose and chitohexaose can bind to two molecules of PPL at the same time, the three-dimensional structure of PPL was predicted by homology modeling and molecular docking and molecular dynamics 4113
dx.doi.org/10.1021/jp110468n |J. Phys. Chem. B 2011, 115, 4110–4117
The Journal of Physical Chemistry B
Figure 4. (A) Homology model of PPL generated by I-TASSER. (B) Chitobiose docked at the putative binding site of PPL. In the PPL model, helices are shown in violet, sheets are shown in pink, and loops are depicted in yellow. In chitobiose, shown as a stick plot, the carbon skeleton is shown in green, oxygens are in red, and hydrogens connected to oxygen are shown in white. In the protein side chains interacting with the ligand, carbon atoms are shown in dark gray and nitrogens are in blue. Water molecules that are involved in the interactions between PPL and chitobiose are labeled Wat1 through Wat5. All interactions, including those mediated via water molecules, are indicated by dashed blue lines.
studies were performed using the predicted model. The best predicted model is shown in Figure 4A. Ramachandran plot (Figure S1, Supporting Information) for this model shows 84.3% residues in favored regions, 10.2% in allowed regions, and 5.6% in outlier regions. Further refinement of the structure improved the stereochemical properties (82.3% in favored region, 14.4% in allowed region, and 3.3% (including 2 GLY residues) in outlier region). Secondary structural data of the homology model (11.9% R-helix, 32.1% extended-strand, 32.1% β-turn, and 23.9% random coil) is consistent with experimental data derived from circular dichroism spectroscopy (9.7% R-helix, 35.8% β sheet, 22.5% turns, and 32.3% unordered structure).23 Since experimental studies have shown that PPL is a dimeric protein,20 the homology model was dimerized using SymmDock server,30,31 and the structure obtained is shown in Figure S2, Supporting Information. Molecular Dynamics Simulations. Chitooligosaccharides were docked into the protein binding pocket, as described in the Methods section, with flexible ligand and rigid protein. The binding pocket consists of at least three subsites based on the juxtaposition of the oligosaccharides with respect to the protein surface, as seen from the docking results (Figure 4B and Figure S4, Supporting Information). The docked chitooligosaccharides showed major interactions with Leu-100 and Glu-102 (subsite 1), Ile-101, Ser-104, and Trp-199 (subsite 2), and Trp-199 and
ARTICLE
Lys-200 (subsite 3). These residues were found to be largely conserved in several other phloem proteins (see Supporting Information, Figure S3). On the basis of the docking results, MD simulation of selected complexes of PPL with different chitooligosaccharides was performed using GROMACS with explicit solvent, and the results obtained are shown in Figure S4, Supporting Information, and Figure 5. Simulation results suggest that the protein binding pocket can accommodate up to three GlcNAc residues irrespective of the saccharide length. These GlcNAc residues interact through water-mediated and van der Waals’ interactions. In the case of chitotetraose, chitopentaose, and chitohexaose also only a triose unit was seen to interact with the protein, while the additional GlcNAc residues of the oligosaccharide were found to hang outside the binding pocket without any interactions with the protein. Interestingly, when another PPL monomer is placed near the three noninteracting GlcNAc residues of chitohexaosePPL complex, several interactions were observed with the second PPL molecule (Figure 5), indicating that the two triose units of a single chitohexaose molecule can interact with the sugar binding sites on two different molecules of PPL (see Supporting Information, jp110468n_si_002.mpg and jp110468n_si_003.mpg). Barring a few, the interactions were mostly indirect, i.e., water mediated (Figure S4, Supporting Information). Chitotriose and chitopentaose appear not to have any direct hydrogen-bonding interactions in our limited analysis. However, there were interesting interactions between the indole ring of Trp-199 and the N-acetyl side chain of one of the GlcNAc residues. For example, the third N-acetyl moiety of chitotriose is involved in a CH 3 3 3 π interaction with the indole ring of Trp-199. Similarly, an N-H 3 3 3 π interaction is observed between the third N-acetyl moiety of chitotetraose and the indole side chain of Trp-199.
’ DISCUSSION Although the pumpkin phloem lectin was purified to homogeneity many years ago and its carbohydrate specificity was shown to be directed toward β (1-4) linked oligomers of N-acetylglucosamine,20 detailed thermodynamic studies on the interaction of carbohydrate ligands with this protein have not been reported so far. In this study, the binding of chitooligosaccharides to PPL has been investigated by ITC and molecular modeling and the association constants as well as the thermodynamic parameters associated with the binding process have been determined. The results obtained are discussed here. Analysis of thermodynamic data associated with the binding of chitooligosaccharides of different chain lengths, presented in Tables 1 and 2, shows that the carbohydrate binding site of PPL is an extended one, with several subsites, each of which can accommodate a sugar residue of the oligosaccharide that is bound. Molecular docking and MD simulation studies are also in agreement with this interpretation and indicate that the binding site contains most likely three subsites. On the basis of thermodynamic analysis of carbohydrate binding, two other phloem exudate lectins, namely, Luffa acutangula agglutinin (LAA) and Coccinia indica agglutinin (CIA), which specifically recognize chitooligosaccharides, were shown earlier to contain extended binding sites, made up of a number of subsites.8,9 ITC studies on the interaction of UDA and WGA have shown that these two lectins also contain extended binding sites that are complementary to a trisaccharide and a tetrasaccharide, respectively.14-16 ITC studies on the binding of chitooligosaccharides to hevein 4114
dx.doi.org/10.1021/jp110468n |J. Phys. Chem. B 2011, 115, 4110–4117
The Journal of Physical Chemistry B
ARTICLE
Figure 5. Complex of chitohexaose with PPL dimer obtained from molecular docking studies and optimized by molecular dynamics simulations. Color code is the same as that for Figure 4.
Figure 6. Enthalpy-entropy compensation plot for the binding of chitooligosaccharides to PPL. The oligosaccharides investigated are chitobiose (O), chitotriose (b), chitotetraose (2), chitopentaose (0), and chitohexaose (9). The solid line indicates a linear least-squares fit to all the data points.
also demonstrated the presence of an extended binding site in the protein.18 The association constants determined for PPL-chitooligosaccharide interaction (Table 1) are about an order of magnitude higher than the corresponding values obtained with UDA and WGA but are in the same range as the values obtained with LAA and CIA. The results of calorimetric titrations presented in Table 1 clearly show that the dimeric PPL has two identical binding sites for chitobiose and chitotriose. However, the binding stoichiometry decreases substantially with an increase in the size of the chitooligosaccharide, i.e., the number of GlcNAc residues in the oligosaccharide. This is quite clear from Figure 2, which shows that the stoichiometry of binding decreases when the number of GlcNAc residues in the chitooligosaccharide is increased, reaching a value of 1.08 ligand (chitohexaose) molecules per protein dimer. This decrease in stoichiometry suggests the formation of higher order structures, wherein some of the ligand molecules simultaneously interact with two binding sites, most likely located on two independent protein molecules. Thus, while chitotriose interacts with only one lectin combining site, a majority of the chitohexaose molecules most likely interacts with
the combining sites from two different lectin molecules. Similar observations have been made regarding the binding of AVR4 elicitor of Cladosporium fulvum to chitotriose.17 The binding reactions for the association of [(GlcNAc)2-6] with PPL investigated here are essentially enthalpy driven with the binding enthalpy (ΔHb) at 298 K for the different chitooligosaccharides ranging between -11.6 and -30.7 kcal mol-1, whereas the entropic contribution to the binding reaction is negative, with the value of binding entropy (ΔSb) being in the range from -23.3 to -71.9 cal mol-1 K-1 (Table 1). The enthalpic nature of binding reactions suggests that the main factors that stabilize the interaction of saccharides with PPL are hydrogen-bonding and van der Waals’ interactions. This interpretation is supported by molecular docking studies, which indicated that the interaction of chitooligosaccharides involves several hydrogen bonds, including some water-mediated ones, besides van der Waals’ interactions between the functional groups on the ligand molecules and the protein (see Figures S4 and S5, Supporting Information). Close examination of the values of ΔHb and ΔSb determined for different oligomers of chitin at different temperatures given in Table 1 indicate that the changes in enthalpy and entropy are compensatory in nature. This is clearly seen in a plot of -ΔHb versus -TΔSb for the chitooligosaccharides at different temperatures (Figure 6). The data plotted here could be fitted to a straight line with a slope of 0.79 and a correlation coefficient (R) of 0.98. This indicates that in the temperature range (288-298 K) in which the measurements have been made, binding of chitooligosaccharide to PPL follows enthalpy-entropy compensation. Enthalpy-entropy compensation has been observed earlier for the interaction of a number of lectins with carbohydrates.26,32-36 A widely accepted model for enthalpy-entropy compensation in protein-ligand interaction is related to the reorganization of water structure around the binding site of the protein and the carbohydrate ligand.37,38 Since a number of studies indicate that water molecules play a crucial role in carbohydrate binding by lectins and since ITC studies provided direct evidence for the involvement of water molecules in the interaction of mannooligosaccharides by Con A,39 the enthalpyentropy compensation observed here can be explained in terms of changes in water structure associated with carbohydrate binding. Consistent with this interpretation, several watermediated hydrogen bonds have been observed between PPL 4115
dx.doi.org/10.1021/jp110468n |J. Phys. Chem. B 2011, 115, 4110–4117
The Journal of Physical Chemistry B and chitooligosaccharides in the molecular docking and molecular dynamics studies presented above. From the data presented in Table 2, it can be seen that the change in ΔG (ΔΔG) resulting from the addition of one more GlcNAc residue to chitobiose is negative, which increases the binding strength. Interestingly, addition of another GlcNAc residue to chitotriose resulted in a slight positive change in the ΔΔG value, leading to a marginal decrease in the binding affinity. This indicates that binding of chitotriose is thermodynamically more favorable than that of chitotetraose and clearly shows that the lectin combining site can accommodate only a trisaccharide or smaller structures. Interestingly, addition of fifth and sixth GlcNAc residues to chitotetraose resulted in negative changes in the ΔG values, indicating that these two chitooligosaccharides bind to PPL more strongly than chitotetraose. This, when considered together with the decrease in the binding stoichiometry, suggests that the higher oligomers of GlcNAc such as chitopentaose and chitohexaose most likely interact simultaneously with two molecules of PPL. Molecular docking and MD studies are in good agreement with this interpretation. In summary, the ITC, docking, and MD simulation studies reported here indicate that the pumpkin phloem exudate lectin can accommodate a trisaccharide (chitotriose) unit in its combining site. Most likely higher order structures are formed between PPL and larger chitooligosaccharides such as chitopentaose and chitohexaose. The chitooligosaccharide-PPL interaction is governed by enthalpic forces with a negative contribution from the binding entropy. Enthalpy-entropy compensation has been observed in the PPL-chitooligosaccharide interaction, underscoring the role of the water structure in the binding process. This is well supported by results of computational studies, which indicate that water-mediated interactions play a significant role in the interactions between amino acid residues in the carbohydrate binding site of the lectin and the GlcNAc residues of the ligand.
’ ASSOCIATED CONTENT
bS
Supporting Information. Methods for molecular modeling, molecular docking, and molecular dynamics simulations, five figures (S1-S5), and two video files (jp110468n_si_002.mpg and jp110468n_si_003.mpg). This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*Phone: þ91-40-2313-4807. Fax: þ91-40-2301-2460. E-mail:
[email protected]. Website: http://chemistry.uohyd.ernet. in/∼mjs/.
’ ACKNOWLEDGMENT This work was supported by research grants from the Department of Biotechnology (India) to M.J.S. and A.S. A.N. was supported by a Senior Research Fellowship from CSIR (India). The MicroCal VP-ITC isothermal titration calorimeter used in this study was obtained from the University with Potential for Excellence (UPE) funds provided by the UGC (India) to the University of Hyderabad.
ARTICLE
’ REFERENCES (1) Lis, H.; Sharon, N. Chem. Rev. 1998, 98, 637–674. (2) Goldstein, I. J.; Poretz, R. D. In The Lectins. Properties, Functions and Applications in Biology and Medicine; Liener, I. E., Sharon, N., Goldstein, I. J., Eds.; Academic Press: Orlando, FL, 1986; pp 35-247. (3) Sharon, N.; Lis, H. Lectins, 2nd ed.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2003; 454 pp. (4) Komath, S. S.; Kavitha, M.; Swamy, M. J. Org. Biomol. Chem. 2006, 4, 973–988. (5) Dam, T. K.; Brewer, C. F. Chem. Rev. 2002, 102, 387–429. (6) Van Damme, E. J. M.; Peumans, W. J.; Barre, A.; Rouge, P. Crit. Rev. Plant Sci. 1998, 17, 575–692. (7) Sabins, D. D.; Hart, J. W. Planta 1978, 142, 97–101. (8) Anantharam, V.; Patanjali, S. R.; Swamy, M. J.; Sanadi, A. R.; Goldstein, I. J.; Surolia, A. J. Biol. Chem. 1986, 261, 14621–14627. (9) Sanadi, A. R.; Surolia, A. J. Biol. Chem. 1994, 269, 5072–5077. (10) Peumans, W. J.; Van Damme, E. J. M. Plant Physiol. 1995, 109, 347–352. (11) Owens, R. A.; Blackburn, M.; Ding, B. Mol. Plant-Microbe Interact. 2001, 14, 905–909. (12) Dinant, S.; Clark, A. M.; Zhu, Y.; Vilaine, F.; Palauqui, J.-C.; Kusiak, C.; Thompson, G. A. Plant Physiol. 2003, 131, 114–128. (13) Aoki, K.; Suzui, N.; Fujimaki, S.; Dohmae, N.; YonekuraSakakibara, K.; Fujiwara, T.; Hayashi, H.; Yamaya, T.; Sakakibara, H. Plant Cell 2005, 17, 1801–1814. (14) Lee, R. T.; Gabius, H. J.; Lee, Y. C. Glycoconjugate J. 1998, 15, 649–655. (15) Katiyar, S.; Van Damme, E. J. M.; Peumans, W. J.; Surolia, A. Biosci. Rep. 1999, 19, 411–419. (16) Bains, G.; Lee, T. R.; Lee, Y. C.; Freire, E. Biochemistry 1992, 31, 12624–12628. (17) Van der Burg, H. A.; Spronk, C. A. E. M.; Boeren, S.; Kennedy, M. A.; Vissers, J. P. C.; Vuister, G. W.; de Wit, P. J. G. M.; Vervoot, J. J. Biol. Chem. 2004, 279, 16786–16796. (18) Asensio, J. L.; Ca~ nada, F. J.; Siebert, H. C.; Laynez, J.; Poveda, A.; Nieto, P. M.; Soedjanaamadja, U. M.; Gabius, H. J.; Jimenez-Barbero, J. Chem. Biol. 2000, 7, 529–543. (19) Garcia-Hernandez, E.; Zubillaga, R. A.; Rojo-Domínguez, A.; Rodriguez-Romero, A.; Hernandez-Arana, A. Proteins 1997, 29, 467–477. (20) Read, S. M.; Northcote, D. H. Eur. J. Biochem. 1983, 134, 561–569. (21) Bostwick, D. E.; Dannenhoffer, J. M.; Skaggs, M. I.; Lister, R. M.; Larkins, B. A.; Thomspon, G. A. Plant Cell 1992, 4, 1539–1548. (22) Bostwick, D. E.; Skaggs, M. I.; Thompson, G. A. Plant Mol. Biol. 1994, 26, 887–897. (23) Narahari, A.; Swamy, M. J. Biosci. Rep 2010, 30, 341–349. (24) Narahari, A.; Swamy, M. J. J. Photochem. Photobiol. B Biol. 2009, 97, 40–47. (25) Peterson, G. L. Anal. Biochem. 1977, 83, 346–356. (26) Sultan, N. A. M.; Swamy, M. J. Arch. Biochem. Biophys. 2005, 437, 115–125. (27) Wiseman, T.; Williston, S.; Brandts, J. F.; Lin, L. N. Anal. Biochem. 1989, 179, 131–137. (28) Gomez, J.; Hilser, V. J.; Xie, D.; Freire, E. Proteins: Struct. Funct. Genet. 1995, 22, 404–412. (29) Baxa, U.; Cooper, A.; Weintraub, A.; Pfeil, W.; Seckler, R. Biochemistry 2001, 40, 5144–5150. (30) Schneidman-Duhovny, D.; Inbar, Y.; Nussinov, R.; Wolfson, H. J. Nucleic Acids Res. 2005, 33, W363–W367. (31) Schneidman-Duhovny, D.; Inbar, Y.; Nussinov, R.; Wolfson, H. J. Proteins 2005, 60, 224–231. (32) Schwarz, F. P.; Puri, K. D.; Bhat, R. G.; Surolia, A. J. Biol. Chem. 1993, 268, 7668–7677. (33) Surolia, A.; Sharon, N.; Schwarz, F. P. J. Biol. Chem. 1996, 271, 17697–17703. 4116
dx.doi.org/10.1021/jp110468n |J. Phys. Chem. B 2011, 115, 4110–4117
The Journal of Physical Chemistry B
ARTICLE
(34) Gupta, D.; Cho, M.; Cummings, R. D.; Brewer, C. F. Biochemistry 1996, 35, 15236–15243. (35) Swaminathan, C. P.; Gupta, D.; Sharma, V.; Surolia, A. Biochemistry 1997, 36, 13428–13434. (36) Komath, S. S.; Kenoth, R.; Swamy, M. J. Eur. J. Biochem. 2001, 268, 111–119. (37) Spohr, V.; Paszkieuicz-Hnatiw, E.; Morishima, N.; Lemieux, R. U. Can. J. Chem. 1992, 70, 254–271. (38) Lemieux, R. U. Acc. Chem. Res. 1996, 29, 373–380. (39) Swaminathan, C. P.; Surolia, N.; Surolia, A. J. Am. Chem. Soc. 1998, 120, 5153–5159.
4117
dx.doi.org/10.1021/jp110468n |J. Phys. Chem. B 2011, 115, 4110–4117
