Investigation on the microscopic mechanism of curdlan with different strengths formed by heating

Abstract

In this study, temperature-dependent changes in the molecular conformation of curdlan were investigated using differential scanning calorimetry (DSC), microscopic observation, particle size analysis and molecular dynamics (MD) simulation. The DSC results showed that the gel formation temperature of curdlan is related to the glass transition temperature (Tg). The results of microscopic observation and particle size analysis showed that the curdlan particles swelled with increasing temperature, and the original rough and irregular surface morphology gradually became relatively smooth. The MD results show that the weak gel is the result of hydrogen bonds and hydrophobic interaction, whereas the strong gel is less affected by hydrogen bonds, and hydrophobic interaction is the main driving force.

Graphical Abstract

Open in new tabDownload slide

Introduction

Curdlan is an extracellular polysaccharide without a branched chain structure that is produced by fermentation by Bacillus faecalis and other microorganisms (Zhang et al., 2018; Mohsin et al., 2019). Curdlan has several types of molecular conformations, including triple-stranded helices, double-stranded helices, single-stranded helices and irregular curls, which can coexist and transform under certain conditions; therefore, curdlan has unique physical and chemical properties (Zhang et al., 2003; Feng et al., 2020). The triple-stranded helical conformation is not only the natural conformation of curdlan but also the most valuable conformation. It is generally believed that this conformation is related to the biological activity of curdlan (Chen & Wang, 2020; Meng et al., 2020). Curdlan and its derivatives have important applications in food additives (Lee & Chin, 2019; Weng et al., 2024), antitumour therapy (Rui et al., 2016; Bao et al., 2021), anti-infection (Chen & Liang, 2017; Lin et al., 2021) and anticoagulation (Lee et al., 2001).

Curdlan is almost insoluble in cold water, but its aqueous suspension forms gels of varying strength when heated. At a heating temperature of 55 °C–65 °C, curdlan water suspensions form a reversible weak gel on cooling, similar to agar. When the heating temperature exceeds 80 °C, the irreversible strong gel is formed after cooling (Zhang et al., 2000; Gagnon & Lafleur, 2011). There are two main mechanisms by which curdlan forms this unique gel: gel formation is related to the association of curdlan, and the strength of the gel depends on the amount of bound water in the association region (Hatakeyama et al., 2016). The strength of the gel is related to the re-engagement between the molecular chains caused by the unwinding of the triple-stranded helix conformation (Funami et al., 1999).

Although there have been several studies on the mechanism of curdlan gel, there are few reports on its microscopic mechanism and a lack of direct molecular evidence. Similar to macromolecules, including proteins and nucleic acids, the maintenance of the higher conformation of polysaccharides depends on secondary interactions, such as hydrogen bonding and hydrophobic interactions. However, due to the inherent complexity of curdlan, it is difficult to evaluate the microscopic mechanism of gel formation using conventional experimental methods, and the conformation and conformational changes of curdlan in solution remain elusive in most cases (Fittolani et al., 2020; Guo et al., 2021).

MD, based on theoretical calculations, can not only evaluate the interaction between curdlan and solution through energy differences but also visually study the dynamic changes of the atoms that make up curdlan. Wang et al. (2022) used MD to demonstrate that soybean hull polysaccharides can stably adsorb bile acids. Meng et al. (2018) found by MD that the triple-stranded helical conformation of Auricularia auricula polysaccharides has a lower scattering area and energy state than the single-stranded helical conformation in water, making it the most stable molecular conformation. Feng et al. (2021) verified by MD that the right-handed triple helix conformation is the most stable conformation of curdlan, and that conformational maintenance depends on intramolecular hydrogen bonds formed by the C₂ hydroxyl groups in the helical cavity. In this study, the microscopic mechanisms of curdlan-forming gels with different strengths were investigated using macroscopic experiments and MD, which helps understand the bioactivity of curdlan at the molecular level.

Materials and methods

Materials

Commercial curdlan was purchased from Jiangsu Yiming Biological Technology Co. Ltd (Jiangsu, China). The molecular weight, polydispersity and average degree of polymerisation of curdlan in pure dimethyl sulphoxide are listed in Table S1. Crystal violet staining solution was purchased from Guangdong Hengjian Pharmaceutical Co., Ltd (Guangdong, China).

Differential scanning calorimetry

The 3.0–5.0 mg curdlan powder was encapsulated in an aluminium crucible and compared with an empty crucible treated by the same method. Thermal analysis was carried out using a differential scanning calorimeter (NETZSCH, Bavaria, Germany). In the thermal analysis experiment, the heating rate of 10.0 °C min⁻¹ is heated from 25.0 ± 0.1 °C to 150.0 ± 0.1 °C under nitrogen protection.

Curdlan gel preparation

A Curdlan water suspension (CUR-water, 0.1% w/v) was accurately prepared and then stirred at a magnetic force of 50 rpm for 12 h to completely dilute it. The inflated CUR-water was divided into two batches, one batch was stirred and heated in a constant temperature water bath at 30.0 °C, 40.0 °C, 50.0 °C, 60.0 °C, 70.0 °C, 80.0 °C and 90.0 °C for 1 h and cooled naturally to room temperature, and the other batch underwent the same heating and cooling treatment without stirring. A comparison of the refractive index of stirred heated CUR-water centrifuged supernatant with deionised water is shown in Figure S1.

Microscopic observation

After heating for 1 h without stirring, the CUR-water was dripped into a circular mound on the microscope slide and dried in an oven at the appropriate temperature to avoid dust contamination during the drying process. After drying, 100 μL of crystal violet staining solution diluted 10-fold with anhydrous ethanol was used for staining. After evaporation of the anhydrous ethanol, the specimen was observed and photographed using a 100× light microscope.

Particle size

CUR-water treated by stirring and heating at different temperatures was dropped into the blood counting chamber and then observed and photographed under a 400× light microscope. The actual particle size of free curdlan in CUR-water was measured and converted using a calliper gauge with the edge length of the blood counting chamber plate (approximately 50 μm) as a ruler.

Molecular dynamics simulation

The conformational changes of curdlan in water were simulated using Materials Studio 2019. A molecular model of water was constructed using this software. The single-stranded helical (s-CUR), double-stranded helical (d-CUR) and triple-stranded helical (t-CUR) conformations were downloaded from PolySac3DB (Sarkar & Pérez, 2012). The water molecule and the three conformations of the curdlan molecular model are shown in Fig. 1. The accuracy of the MD was medium, the algorithm used was SMART, the simulation force field selected was COMPASS II (Bunte & Sun, 2000), and the other options were default unless otherwise specified.

Determination of the initial molecular dynamics structure

The water molecule and conformational models of curdlan were manually charged and optimised using the Geometry Optimisation option of the Forcite module. Three types of CUR-water simulation systems were created using the Amorphous Cell module: curdlan triple helix system (water + t-CUR), curdlan triple helix semi-unwinding system (water + h-CUR) and curdlan triple helix fully unwinding system (water + s-CUR). Detailed parameters and construction diagrams of the three simulation systems are presented in Table 1 and Fig. 1, respectively. The initial density of the system was 1.0 g cm⁻³ and the box was a cube with a side length of 52.1 Å. When constructing the box, the geometry was optimised to minimise the energy and the number of frames in the box was 30. To achieve a reasonable density, the frame with the lowest energy from the 30-frame box was selected for the annealing simulation. Geometry optimisation was performed several times before the annealing simulation to optimise the overall box structure. The temperature range of the annealing simulation is 200.0–400.0 K, the pressure was applied to standard atmospheric pressure and the pressure control method was Berendsen. Ten annealing cycles were performed and the frame with the lowest energy was selected as the initial MD structure.

The specific process of molecular dynamics

MD was carried out using the dynamics option in the Forcite module. The NVT simulation was carried out. The temperature gradient was set to 303.0, 313.0, 323.0, 333.0, 343.0, 353.0 and 363.0 K, the temperature control method was Nose, the simulation step was 100 000 steps, the step size was 1.0 fs and the simulation time was 100.0 ps. At the end of the NVT simulation, the NPT simulation was performed, the pressure was set to standard atmospheric pressure, the Berendsen pressure control method was used, and the other settings were the same as for the NVT.

Glass transition temperature

The density of the Curdlan triple-stranded helix system with temperature was analysed using the analysis option in the Forcite module, and the glass transition temperature (Tg) of the Curdlan molecule under the simulated conditions was obtained according to the relationship between the density of the simulation system and temperature (Shi et al., 2023).

Binding energy

The energy of the MD results was calculated using the energy option in the Forcite module, and the total energy E_solvent+CUR of the CUR-water simulation system, the energy E_solvent of water and the energy E_CUR of curdlan were calculated. The formula for calculating the binding energy E_bind, E_bind between curdlan and water is as follows (Bello, 2014; Biedermann & Schneider, 2016; Luckhaus et al., 2017):

Potential energy

The Energy option of the Forcite module was used to analyse the potential energy of the curdlan molecular conformation, which mainly consisted of bond, angle and torsional energy.

Hydrogen bonds clustering

The radial distribution function (RDF) of the oxygen atoms and all surrounding hydrogen atoms in the curdlan was obtained using the analysis option in the Forcite module. The truncation distances of hydrogen bonds of different strengths were determined from the RDF curves (Ahsan et al., 2020; Kacar & De With, 2020). The hydrogen bond between curdlan (HBCC) and the hydrogen bond between curdlan and water molecules (HBCS) were determined using the Hydrogen Bonds option in the software's Build menu bar. The hydrogen bond statistics were calculated using Materials Studio 2019's script and the truncation angle was set to 120.0°.

Solvent accessible surface area (SASA)

A water molecule was used as a probe and its radius was set to 1.4 Å. The SASA of the curdlan conformation was determined using the Atom Volumes & Surfaces option in the tools menu bar of the software.

Statistical analysis

All experiments were carried out in triplicate. Results are expressed as mean ± standard deviation. Statistical analysis was performed using Origin 2021 (Origin Lab, Northampton, MA, USA).

Results and discussion

Differential scanning calorimetry analysis

Phase transition information, including the Tg and viscous flow temperature of curdlan, can be obtained through differential scanning calorimetry (DSC) analysis (Drzeżdżon et al., 2019). The DSC curve of the curdlan powder is presented in Fig. 2. According to the theory of amorphous polymers, the regions a, b, c, d and e represent the glassy state, glass transition region, highly elastic state, viscoelastic transition region and viscous flow state of curdlan, respectively. In the glassy state, the molecular segments of curdlan are insensitive to temperature, and the molecular conformation of curdlan is rigid. The molecular chain segments of curdlan in the glass transition region began to move, indicating a decrease in the rigidity of the molecular conformation. The Tg of curdlan used in this study was approximately 63.5 °C (Parodi et al., 2017), which is consistent with the previously reported temperature range of weak gel formation of curdlan. Therefore, it can be hypothesised that the weak gel formation of curdlan may be related to its glass transition. The molecular segments of the highly elastic curdlan exhibited movement, yet there was no relative slip between the molecular chains, which demonstrated a degree of recovery comparable to that of rubber. The molecular conformation remained semi-rigid. In the viscoelastic transition region, the curdlan not only exhibited high elasticity but also began to exhibit fluidity. The molecular chain centre of gravity began to shift relatively, and the molecular conformation began to become compliant. The shift of the molecular chain centre of gravity of curdlan in the viscous flow state was comparable to that observed in small molecule liquids, with the molecular conformation exhibiting a predominantly compliant nature.

Gel volume analysis

Given that the Tg of the polymer is influenced by factors such as concentration, solvent environment, etc. (Hosseininejad et al., 2024), it can be expected that the Tg of CUR-water will be lower than the 63.5 °C measured by DSC. To account for the sedimentation of curdlan particles during heating, two batches of gels were prepared by stirring and not stirring the heated CUR-water. Figure 3 presents digital images of the two batches of gels. A comparison of the images revealed that no gel was formed in the two batches of CUR-water at 30.0 °C, 40.0 °C and 50.0 °C, with minimal change observed compared with the unheated CUR-water. The molecular conformation of curdlan was found to be rigid. Weak gels were formed in 60.0 °C and 70.0 °C CUR-water, but the strength of the weak gels formed by heating without stirring was relatively high. The gel formation temperature was close to the Tg obtained from the DSC curve, indicating that the formation of the weak Curdlan gel was related to Tg. The strong gel was formed by non-stirring heating of CUR-water at 80.0 °C and 90.0 °C, whereas the weak gel was heated with stirring, as this prevented the mutual aggregation of curdlan particles. It was therefore hypothesised that the formation of the strong curdlan gel may be related to the hydrophobic interaction between curdlan molecules.

Analysis of the results of microscopic observation

The morphology, aggregation state and gelation of curdlan in CUR-water treated at different temperatures were studied by observing unstirred heated CUR-water under a 100× light microscope. A digital image of this is presented in Fig. 4. The comparison of microscopic images of stirred and non-stirred curdlan is shown in Figure S2. At temperatures below 50.0 °C, the curdlan particles were predominantly irregular, and no obvious swelling phenomenon was observed. The distance between the particles diminishes with rising temperature, yet no gelation is observed. The curdlan particles treated at 60.0 °C exhibited a full, predominantly spherical morphology, with only a minor presence of irregular morphology. The degree of aggregation between the particles increased, accompanied by the emergence of local gelling behaviour. Curdlan particles treated at 70.0 °C and above exhibited a fuller morphology with increasing temperature, with the prevalence of irregular morphology diminishing and the degree of gelation markedly increasing with rising temperature.

Particle size analysis

To ascertain the contribution of individual curdlan particles to the gel strength, the particle size statistics of free curdlan particles in stirred, heated CUR-water were calculated. The resulting statistical histogram of curdlan particle size is presented in Fig. 5. As the temperature of the heating process increased, the average particle size of the curdlan particles increased overall. As the temperature increased, the average particle size became more pronounced, and the gel strength exhibited a positive correlation with the average curdlan particle size. The average particle sizes of curdlan treated at 30.0 °C, 40.0 °C and 50.0 °C were 25.07 ± 3.87, 28.23 ± 3.31 and 28.58 ± 4.86 μm, respectively. The molecular conformation of curdlan was observed to be rigid within the specified temperature range, and thus the effect of heating on the average curdlan particle size was not found to be significant. The mean particle size of curdlan treated at 60.0 °C and 70.0 °C was found to be 33.62 ± 7.88 and 36.45 ± 7.62 μm, respectively. The molecular conformation of curdlan in this temperature range exhibited a gradual transition from rigid to semi-rigid, accompanied by an increase in the average particle size compared with the rigid conformation. The mean particle size of curdlan treated at 80.0 °C and 90.0 °C was found to be 39.10 ± 7.19 and 41.97 ± 7.78 μm, respectively. The average particle size of curdlan exhibited a continued increase in line with the observed rise in its molecular conformational flexibility.

Molecular dynamics analysis

Glass transition temperature analysis

Due to the limitations of the current computing resources, the molecular weight of curdlan in the simulation system is relatively low. There are only ten glucose units that make up a single chain of curdlan, and therefore the molecular weight cannot be compared with hundreds of glucose units in the real environment. Furthermore, the impact of the polydispersity of curdlan on its molecular conformation could not be replicated in the simulation, resulting in discrepancies between the simulation and macroscopic experimental results. The Tg of the curdlan in the simulation system can be obtained by simulating the relationship between system density and temperature. The relationship between density and temperature of the simulated system is illustrated in Fig. 6. The temperature at the intersection of the red and blue fitted curves was approximately 320 K (47 °C), indicating that the Tg of curdlan in the simulated system was approximately 47 °C. The Tg of curdlan in the simulated system is lower than that in the actual environment (63.5 °C) due to the lower molecular weight of curdlan in the simulated system.

Binding energy analysis

The binding energy is the energy released by the combination of components from the free state into a whole. It can be regarded as the sum of the energies of the interactions between the components. The greater the binding energy, the more tightly the components are bonded together. By comparing the binding energies between curdlan and water molecules in three types of CUR-water simulation systems, the molecular conformation of curdlan at the corresponding temperature was determined. The binding energies of the three types of CUR-water systems simulated at different temperatures are presented in Fig. 7.

In the simulated system, the Tg of curdlan was approximately 47 °C. The molecular conformation of curdlan below Tg was rigid, and the interaction between curdlan and water molecules was minimal. Consequently, the molecular conformation of curdlan in the 30.0 °C and 40.0 °C simulated system was the triple-stranded helix conformation with the lowest binding energy. When the temperature of the simulation was above the Tg, the rigidity of the molecular conformation of curdlan decreased, and the interaction between curdlan and water molecules increased. By the tenet that higher binding energy signifies greater system stability, the molecular conformation of curdlan in the 50.0 °C, 60.0 °C and 70.0 °C simulation systems is that of a semi-unwound triple helix. With an increase in temperature, the degree of unwinding also rises. The molecular conformation of curdlan in the 70.0 °C simulation system exhibited a high degree of unwinding. Upon reaching temperatures of 80.0 °C and 90.0 °C in the simulation, the binding energy of water + s-CUR surpassed that of water + h-CUR, indicating that the molecular conformation of curdlan had undergone a complete unwinding.

In summary, the molecular conformations of curdlan in the CUR-water simulation system treated at different temperatures are as follows (Fig. 8). At temperatures between 30.0 °C and 40.0 °C, the molecule assumes a triple helix conformation. Between 50.0 °C and 70.0 °C, it exhibits a semi-unwinding triple helix conformation. Finally, at temperatures between 80.0 °C and 90.0 °C, the molecule assumes a single helix conformation. The effect of temperature on the molecular conformation of curdlan was found to be similar to that observed by atomic force microscopy (Xiao et al., 2017), which elucidates the changes in curdlan particle morphology and average particle size with temperature in macroscopic experiments.

Potential energy analysis

The potential energy may be considered to represent the cohesion energy of the curdlan molecular chain. A reduction in potential energy results in an increase in structural stability. The potential energies of curdlan at different temperatures are presented in Table 2. As the temperature rises, the potential energy of curdlan increases continuously. It can be observed that the bond and angle energies contribute more to the potential energy than the torsional energy. The energy transition of the bond energy was observed to occur between 40.0 °C and 50.0 °C, which is within the temperature range of Tg (approximately 47 °C). This indicates that the glass transition primarily influenced the bond energy. Both the bond and angle energies exhibited energy transitions between 60.0 °C and 70.0 °C, indicating that the molecular chain of curdlan became more extended, which is conducive to the unwinding of the three-chain spiral conformation of curdlan.

Hydrogen bonds cluster analysis

Curdlan contains a significant number of hydroxyl groups, which can form complex hydrogen bonding networks. Hydrogen bonding plays a pivotal role in the maintenance of the curdlan gel network. Consequently, it is of paramount importance to analyse the hydrogen bonding network of the simulation system in great detail. The O-H bond length and truncation distance of hydrogen bonds of varying strengths can be characterised by the RDF curves. Figure 9 presents the RDF curves of the oxygen atoms in curdlan and the surrounding hydrogen atoms. The peak at 0.975 Å represents the O-H bond length of the hydroxyl group of curdlan. The regions 1.385–1.875, 1.875–2.235 and 2.235–2.995 Å represent strong, moderate and weak hydrogen bonds, respectively. Peaks occurring after 2.995 Å indicate a reduction in the strength of the hydrogen bonds and a correspondingly minor impact on the simulation system.

The above truncation distances were used to evaluate HBCC and HBCS using cluster analysis. The data about the strong, medium, and weak hydrogen bonds are presented in Table 3. Given that the contribution of strong hydrogen bonds to the hydrogen bond network was considerably greater than that of medium and weak hydrogen bonds, the change in the number of strong hydrogen bonds was analysed initially, with the difference between the latter two considered when the number of strong hydrogen bonds was similar. When the temperature was below the Tg (approximately 47 °C), the number of HBCC exhibited a slight decrease with increasing temperature, whereas the number of HBCS exhibited a slight increase with increasing temperature. This was attributed to the slight swelling of the triple-stranded helix conformation. In the temperature range of 50.0 °C–70.0 °C, the number of HBCC increased with increasing temperature, whereas the change in HBCS was the opposite. This temperature range corresponds to the formation of weak Curdlan gels in macroscopic experiments, indicating that hydrogen bonding plays a significant role in the formation of weak Curdlan gels. At temperatures of 80.0 °C and 90.0 °C, the formation of strong gels in macroscopic experiments occurs. In the simulation system, the number of HBCC and HBCS decreases, indicating that the higher temperature is not conducive to the existence of hydrogen bonds and that hydrogen bonding is not the main factor driving the formation of strong gels.

Solvent accessible surface area analysis

The effect of different heating methods on curdlan gel volume suggests that hydrophobic interactions may be another important factor driving gel formation. Although the COMPASS II force field does not explicitly account for hydrophobic interactions in terms of energy, it can indirectly reflect hydrophobic interactions between curdlan (Chen & Panagiotopoulos, 2019). The schematic and comparison diagrams of the curdlan SASA in the CUR-water simulation system are shown in Figs. 10 and 11, respectively. In these diagrams, AB represents the SASA of the curdlan aggregates, whereas A + B represents the sum of the SASA of each independent part of the curdlan aggregates.

The numerical values of A + B and AB in the 30.0 °C and 40.0 °C simulation systems were found to be similar, indicating that the two triple helix conformations are separated from each other and that there is minimal hydrophobic interaction between the molecules. In the 50.0 °C, 60.0 °C and 70.0 °C simulation systems, the value of A + B was larger than that of AB. Furthermore, the difference between them was positively correlated with temperature, indicating that the unwinding of the triple helix conformation is beneficial to the enhancement of the hydrophobic interaction between curdlan. The values of A + B in the 80.0 °C and 90.0 °C simulation systems were considerably larger than those of AB, which can be attributed to the complete unwinding of the triple helix conformation, which further enhances the hydrophobic interaction between curdlan. Consequently, hydrophobic interactions are of significant importance in the formation of both weak and strong curdlan gels.

Conclusion

This study demonstrated that the temperature of curdlan gel formation was correlated with Tg, as evidenced by the integration of macroscopic experiments and MD simulations. Weak curdlan gel formation was found to be facilitated by an increase in hydrogen bonding and hydrophobic interactions. The formation of a strong gel was found to be related to an increase in the hydrophobic interactions of curdlan but was weakly related to hydrogen bonding. Due to the constraints of computational resources, the experimental outcomes were subject to certain limitations. It is anticipated that with the continued advancement of computer power and simulation methods, the gel mechanism of curdlan will be more clearly elucidated, thereby providing a theoretical foundation for the study of its biological activity.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Grant number: 31760469), the Scientific Research and Technology Development Program of Guangxi (AA22117014-1 and AA22117002-5), the Central Guided Local Science and Technology Local Development Funding Program (2023ZYZX3016), and the Agriculture Research System of China (CARS-17-0502).

Author contributions

Huaiguang Wang: Methodology; software; investigation; formal analysis; writing – original draft. Zhikun Lv: Investigation; writing – review and editing. Yuhao Hao: Writing – review and editing. Mengling Lu: Writing – review and editing. Zhi Huang: Supervision; writing – review and editing. Jianbin Li: Funding acquisition; project administration; resources.

Conflict of interest statement

The authors declare that there are no conflicts of interest related to the publication of this article.

Ethical guidelines

Ethics approval was not required for this research.

Peer review

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/ijfs.17436.

Data availability statement

The data used to support the findings of this study can be made available by the corresponding author upon request.

References