Improving ethanol tolerance of ethyl carbamate hydrolase by diphasic high pressure molecular dynamic simulations

EC hydrolase mutants screening based on dHP-MD

In this study, we combined the enzyme modification strategy based on isothermal compression micro-perturbation developed in our previous research (Zheng et al. 2022) with diphasic molecular dynamic simulation, named diphasic high pressure molecular dynamic simulations (dHP-MD), aimed at targeting enzyme mutants with improved activity and stability, especially organic solution resistance. A total of 1350 ns MD was performed (5 parallels × 3 pressures × 3 concentrations × 30 ns) under gradient pressures (1, 500, 1000 bar) and ethanol concentrations (10%, 50%, 100% v/v) was performed for exploration of the influence of ethanol molecules in structure of EC hydrolase and selection of ethanol-sensitive sites for further experiments (Fig. 1A). The distribution of root mean square deviations (RMSD) indicated the changes in flexibility of the protein during the simulation (Fig. 1B), which showed the greatest changes revealed under the condition of 50% (v/v) ethanol. In contrast, the changes in protein structure reduced at 100% (v/v) ethanol, probably due to the invasion of the protein structure by ethanol. The results indicated that appropriate ethanol concentration and pressure could expose flexible regions of the protein, thus making it easy to identify and screen the unstable regions. The root mean square fluctuation (RMSF) of protein structure under these conditions was also analyzed (Additional file 1: Fig. S1–S3), and the overall trend was similar to RMSD. The secondary structure α helix, β sheet and loop, the most basic units of enzyme spatial structures, was selected to classify the structure of EC hydrolase. Among the three secondary structures α helix, β sheet and loop, the conformational changes of loop were more intense than those of α helix and β sheet. Thus, we choose to screen out the region with large fluctuations from loop region for calculation, where the constant temperature compression coefficient βT was introduced to evaluate the conformation response of the certain loops to ethanol. Six appropriate loop structures were selected for the calculation of βT, namely amino acids loop67-76, loop82-107, loop176-181, loop254-261, loop320-333 and loop381-392. Moreover, the regions loop67-76, loop176-181 and loop320-333 of EC hydrolase showed the highest βT fluctuations that were regarded as “sensitive fluctuation” regions for further analysis (Fig. 1C). The amino acid residues in these three structures were subjected to virtual saturation mutation, and FoldX and I-mutant 2.0 were introduced to evaluate the changes in their stability after mutation. According to the thermodynamic hypothesis, the initial conformation of enzyme is the conformation with the lowest free energy (Zheng et al. 2020). The mutants of ΔΔG (ΔG mutant-ΔG WT)  < 0 obtained by FoldX based on the empirical effective energy function are the positive mutants with increased stability. In order to prevent more negative and neutral structures from being obtained by a single prediction program (Wijma et al. 2014), another prediction program, I-mutant 2.0, was also used to calculate the change of free energy, in which ΔΔG > 0 represents the stability of the mutant is increased. The final calculation results are shown in Fig. 1D, where the upper left corner of each cell represents the result of FoldX, the lower right corner represents the result of I-mutant 2.0. Finally, the mutants that showed increased stability in both predictions were A321I, S325F, S325Y, S325N, Q332E, Q332G, H68A, H68L, H68M, H68K, H68Y, K70R and A178L, and were used for further experimental validation.

Fig. 1figure 1

Mutant screening process. A Experimental procedure for screening mutants by modelling and simulating protein sequences and validating them in experiments. B RMSD of proteins at different ethanol concentrations and different pressures (10%, 50%, 100% ethanol concentrations, 1, 500, 1000 bar). C Compression coefficients of individual protein structures at different ethanol concentrations and different pressures. Each group represented a different loop structure, each line in the group represented a different pressure in the simulated environment, and each point on the line represented βT at different ethanol concentrations. D Heat map of free energy changes at mutant sites obtained from multi-scale free energy calculations, the red box in the figure is the final selected results of the research

Characterization of the single-point mutations

All mutants were purified and assayed for specific enzyme activity (Fig. 2A). More than half of the mutants showed higher specific enzyme activity than WT. S325N, H68A, K70R and A178L showed significant increases, whose specific enzyme activity were 4.18 U/mg, 5.26 U/mg, 3.38 U/mg and 5.36 U/mg, respectively, compared to 1.89 U/mg in WT, which was an increase of about 2.21-fold, 2.78-fold, 1.79-fold and 2.84-fold. The specific enzyme activity of A321I, S325F, S325Y, H68L decreased compared to WT. Such as the specific enzyme activity of S325Y was only 0.43 U/mg, which was 0.23-fold than that of WT.

Fig. 2figure 2

Enzymatic properties of single mutant proteins. A Specific enzymatic activity of single mutant proteins. B Optimal temperature of single mutant proteins. C Ethanol tolerance of single mutant proteins. D Variation in Tm values of single mutant proteins

The optimum temperatures of most mutants were at 30 °C, which was lower than that of WT (optimum temperature at 37 °C) (Fig. 2B). But as the temperature rose above 40 °C, most mutants showed a slow trend in decreasing of enzyme activity and a higher relative activity than WT. The tolerance to temperature were improved in H68A and A178L. The relative enzyme activity of H68A at 50 °C and that of A178L at 45 °C were 26.90%, 48.31%, which were both higher than WT. In contrast, Q332E had the greatest decrease which was only 19.24% relative enzyme activity remained after reaction at 40 °C.

Most of the single point mutants showed an enhanced ethanol tolerance compared to WT (Fig. 2C). Among them, S325N, K70R and A178L showed an obvious improvement in relative enzyme activity, which were 2.62-fold, 1.80-fold and 1.78-fold higher than that of WT at 10% (v/v), respectively. Both mutant S325F and S325Y were less tolerant to ethanol than WT, with only 1.93% and 3.87% relative enzyme activities remaining after the reaction at 15%(v/v) ethanol, which was lower than WT (8.12%). The relative enzyme activities of H68L were higher than those of WT under 5% (v/v) and 10% (v/v) ethanol, but less than that of WT under 15% (v/v) ethanol.

A178L had improved the most, which elevated from 43.7 °C in WT to 55.9 °C (Fig. 2D). Furthermore, another mutant S325N also had a higher Tm value of 50.5 °C. The values of Km and Kcat/Km for WT were 37.67 mM and 491.42 s−1·M−1 (Table 1), respectively, while the values of Km and Kcat/Km for H68A and K70R were 31.38, 33.38 mM and 1589.33, 944.16 s−1·M−1, respectively, indicating that the affinity and catalytic efficiency of these two single point mutated proteins were improved. The kinetic constants of the rest of the mutants are listed in Additional file 1: Table S2.

Table 1 Enzymatic kinetic parameters of the three mutant proteinsEnzymatic properties of combinatorial variants

H68A, K70R, A178L and S325N were selected as the candidates applied in subsequent combinatorial mutation validation as well as the mutant Q328C which improved ethanol tolerance and thermal stability previously reported (Liu et al. 2016). Generally, 10 double mutants and 10 triple mutants were constructed to measure enzymatic performance.

The specific enzyme activity of double mutants and triple mutants were general higher than that of WT (Fig. 3A), H68A/Q328C, K70R/A178L/Q328C had no enzyme activity. The results of some mutants were slightly less than that of WT, but the thermal stability and ethanol tolerance of mutants were increased greatly. In double mutants, the specific enzyme activity of K70R/A178L, K70R/S325N had a definite improved, which was 4.35 U/mg, 4.86 U/mg and was 2.30-fold, 2.57-fold than WT. Among triple mutants, a significant increase was shown in H68A/K70R/S325N, S325N/Q328C/A178L and H70A/A178L/S325N which was 6.46 U/mg, 5.71 U/mg, 4.74 U/mg, an increase of approximately 3.42-fold, 3.02-dold, 2.51-fold. The Tm value of H68A/K70R/S325N was also higher than that of double mutants (Fig. 3A), which was increased from 43.7 °C to 54.8 °C.

Fig. 3figure 3

Diagram of the enzymatic properties of the three mutant proteins. (A) Specific enzymatic activity and Tm values of double and triple mutant proteins. B Optimal temperature of the triple mutant proteins. C Temperature stability of the triple mutant proteins. D Ethanol tolerance of the triple mutant proteins. E Ethanol stability of the triple mutant proteins

Most of the double mutants had a higher optimal temperature than WT (Additional file 1: Fig. S4). The optimal temperature of K70R/Q328C, A178L/S325N and A178L/Q328C was increased from 37 °C to 45 °C, and was the highest temperature among these variants. On the other side, when the temperature was 60 °C, H68A/K70R had a higher relative enzyme activity than other mutants, which was 12.03% and was 9.11-fold than WT.

H68A/K70R, K70R/A178L were sensitive to ethanol, they performed lower residual activity when ethanol existed than WT (Additional file 1: Fig. S5). On the contrast, H68A/S325N, K70R/S325N and A178L/S325N were the top three with the highest residual activity under 20% (v/v) ethanol, which were 32.42%, 27.73%, 25.74% and was 3.95-fold, 3.38-fold, 3.14-fold than that of WT. H68A/S325N showed higher affinity for the substrate than WT, meanwhile K70R/S325N showed higher catalytic efficiency than WT (Table 1).

Furthermore, triple mutants were constructed and analyzed to obtain mutants with higher stability and greater ethanol tolerance. In order to analyze the enzymatic properties more comprehensively, the determination of thermal stability and ethanol stability was added to the experiment (Fig. 3B–E).

Similar trends were shown in the optimum temperature and thermal stability with enzyme activity mostly decreasing above 40 °C. A slow decline was shown in the enzyme activity of A178L/S325N/Q328C with temperature increasing, both in optimum temperature and thermal stability assay. High ethanol tolerance and stability was shown in mutant H68A/K70R/S325N, whose relative enzyme activity was 41.16% after 15 min of reaction in 20% (v/v) ethanol, 5.02-fold higher than WT relative enzyme activity of 8.20%. Its relative enzyme activity was still 15.87%, in 20% (v/v) ethanol for 1 h.

The affinity for the substrate and catalytic efficiency of H68A/K70R/S325N was increased (Table 1). The overall performance of triple mutant H68A/K70R/S325N has improved compared to WT, which proved the success of mutant screening. A subsequent investigation, structural and molecular dynamics simulation analysis, was carried out to investigate why the ethanol tolerance and stability of H68A/K70R/S325N was improved.

Structural analysis and molecular dynamics simulation of EC hydrolase variants

In order to investigate the mechanism for improved ethanol tolerance of H68A/K70R/S325N, we focused on the internal molecular interaction changes between WT and the triple variant. The mutant not only retained the previous hydrogen bond between Ala68 and Glu63, but formed a new hydrogen bond between Ala68 and Arg70 (Fig. 4A, B). We also analyzed the mutant site Asn325 in the result of RMSF (Additional file 1: Fig. S6), in this site the mutant showed a less volatility than WT, which meant Asn325 had a greater stability than Ser325.

Furthermore, 50% (v/v) of ethanol-water diphasic of MD simulations were proceeded to uncover the fluctuation of hydration shell between EC hydrolase variants. We compared the number of ethanol molecules and water molecules near the mutation sites and active center of the triple mutant and WT to investigate the invasion of ethanol (Fig. 4C, D). The number of water molecules near Asn325 and the number of ethanol molecules near Ala68 increased in the triple mutant, while the number these molecules in the mutant at the other mutant sites and active center were less than that of WT. The invasion effect on enzyme activity by ethanol was weakened with reduced number of ethanol and water molecules at active center in triple mutant, which could protect enzyme activity in ethanol system. In addition, hydrophilic and hydrophobic accessible surface areas were also calculated (Additional file 1: Fig. S7), both of them in mutant were decreased within 30 ns under 50% (v/v) ethanol compared to WT. Results above suggested that the mutant enzyme formed a tighter contracted structure overall, thus the hydrated layer of the protein itself was protected and the impact of ethanol intrusion was reduced. These changes made overall structure of the protein more stable and less susceptible to changes in ethanol concentration and temperature, resulting in a higher enzyme activity than WT high ethanol concentrations. The amount of water and ethanol surrounding WT and mutant proteins was plotted by VMD (Fig. 4E). It can be seen that the number of ethanol molecules surrounding the mutant protein was reduced, indicating that the mutant protein had an increased affinity for water, thus protecting the enzyme from the influence of ethanol. The ethanol-sensitive sites in EC hydrolase were selected through dHP-MD, which were modified for mutants with ethanol resistance. The results showed that the triple mutant protected the hydration layer of the protein and enhanced the ethanol resistance. Interestingly, the balance between the activity and the stability of ethanol could be maintained, with improved activity as well as stability.

Fig. 4figure 4

Structural and simulation analysis of the triple mutant protein in the ethanol system. A Hydrogen bonds around WT mutation residue B Hydrogen bonds around mutation residue of the triple mutant C Water molecules near the mutation sites and active center D Ethanol molecules near the mutation sites and active center E Water molecules and ethanol molecules around the wild-type and triple mutant proteins

Application of immobilized triple mutant

Furthermore, immobilized EC hydrolase with chitosan was used for improvement of stability (Fig. 5A). Firstly, the degradation of EC by free and immobilized pellets was verified in simulated wine samples (500 μg/L EC, 15% ethanol, pH 4.5) (Fig. 5B). After the final concentration of 1.5 mg/ml free enzyme was added to the sample and degraded EC at 30 °C for 12 h, 14.81% EC was degraded by WT and 28.62% by free H68A/K70R/S325N. Alternatively, the immobilized WT enzyme degraded 18.01% EC in the simulated wine sample, and the immobilized mutants degraded 32.52% EC under the same conditions for 12 h. The results above showed that the triple mutant had better degradation ability of EC than WT under simulated wine conditions, while the immobilized enzyme performed better than free enzyme.

Fig. 5figure 5

Determination and application of enzymatic properties after immobilization of EC hydrolase. A Enzyme immobilization B Application of WT and mutants in simulated wine samples C Ethanol stability of immobilized enzyme D Ethanol tolerance of immobilized enzymes E Optimum temperature for immobilized enzyme F Reuse of immobilized enzyme

Meanwhile, other enzymatic properties of the immobilized EC hydrolase were determined. The activity of immobilized EC hydrolase decreased significantly after being kept in a certain amount of ethanol (5–20% v/v) for 1 h, (Fig. 5C), and the enzyme activity of H68A/K70R/S325N decreased slower than that of immobilized WT with the increase of ethanol concentration. However, the residual activity of the immobilized mutant was significantly increased in direct reaction in various concentrations of ethanol (Fig. 5D). The relatively residual activity of the immobilized mutant was 76.31% at 20% (v/v) ethanol, about 1.85-fold of that of the free triple mutant. The optimal temperature of the immobilized enzyme remained unchanged (Fig. 5E), but the enzyme activity of the mutant was increased to 44.61% at 50 °C, and the relative enzyme activity of immobilized WT was also increased to 29.63% at 50 °C. Meanwhile, the recovery rate for recycle was also determined (Fig. 5F). The residual enzyme activity of immobilized WT in the fifth experiment was 45.02%, while that of H68A/K70R/S325N was 54.31%. The above results showed that the immobilization successfully improved the stability and ethanol tolerance of the protein, and provided potential for practical application of EC degradation.

留言 (0)

沒有登入
gif