Enhancement Of Activity And Enantioselectivity Of Bacillus Subtilis Lipases By Directed Evolution And Site-directed Mutagenesis

Bacillus subtilis is a Gram-positive, aerobic and endosporeforming bacterium commonly found in soil, aquatic habitats and associated with plants. Its genome, proteome and secretome have been studied in great detail. Lipolytic enzymes produced and secreted by B.subtilis include a phospholipase, a lipase LipA and an esterase LipB. Lipases can catalyze the hydrolysis, (trans)esterification, and synthesis of a board range of esters. Their application for the preparation of chiral building blocks, especially by kinetic resolution of racemic mixtures, is of particular interest for the pharmaceutical, agrochemical and food industries.However, the enantioselectivity of most natural lipases is often not high enough for a desired reaction, which is a mainly obstacle for their applications. It is very important to improve the enantioselectivity of lipases.We had cloned, expressed and characterized the Bacillus subtilis lipase A (LipA). The enantioselectivity of LipA was enhanced by site-directed and site-saturation mutagenesis combined with high throughput screening. Furthermore, specific activity of a novel Bacillus subtilis lipase (BSL2) was increased by directed evolution.1. Cloning, expression, purification and characterization of LipAThe genome of Bacillus subtilis has been determined, and it is the first Gram-positive bacterium whose genome had been determined. We designed a pair of primers to amplify the LipA gene by PCR. The gene was cloned into plasmid pET28a(+) and the combined plasmid pET28a-lipA was transformed into E.coli BL21(DE3) competent cells. LipA consists of 181 amino acid residues and is particularly interesting, since its molecular mass (19.3 kDa) is much smaller than that of the lipases from other organisms. LipA represents one of the few examples of a lipase that does show interfacial activation in the presence of oil-water interfaces. LipA can hydrolyze triacylglyceride and p-nitrophenyl-ester substrates with varying sizes of the fatty acid site chains. Triacylglycerides with short chain fatty acids (< 8C) are preferentially hydrolyzed with maximum activity towards the C8 substrate tricaprylin.2. Study on the Kinetic parameters of hydrolysis of (R, S)-ethyl mandelate of LipA and A75GLipA has a consensus sequence Ala-X-Ser-X-Gly, where an Ala replaces the Gly found in most of the bacterial lipases. We constructed the LipA variant A75G and investigated the enantioselectivities of both lipases for the hydrolysis of R, S-ethyl mandelate. We found the E value had increased from 2.64 to 7.66 when the Ala was replaced by Gly. To explore the reason, kinetic parameters Km and kcat were detemined. We could find that the Km values toward R-enantiomer and S-enantiomer did not have obvious alteration. Furthermore, the Km values of A75G were generally slight higher than those of LipA. However, the kcat values toward R- and S- enantiomers showed significant difference. The kcat values for R-enantiomer of all lipases were higher than those for S-enantiomer. Moreover, the kcat values of A75G for both enantiomers were decreased compared to those of LipA, but for S-enantiomer the kcat values decreased more rapidly, which leaded to the increase of enantioselectivity. Kinetic properties of lipases for the hydrolysis of p-nitrophenyl caprylate were similar to these results. The enzymatic hydrolysis followed the Michaelis-Menten mechanism, wherein the kcat indicates the converted rate from the substrate to product and the Km stands for the binding affinity between the substrate and the enzyme. The results showed that the Km values of both enantiomers were almost the same while the kcat values of the R-enantiomer were higher than those of S-enantiomer, which suggested the forming of transition state rather than the binding complex was the determined step of the enzymatic hydrolysis. The results indicated that the mutate positions located in the catalytic site rather than in the binding site.The effect of temperature, pH and organic solvent on the hydrolysis of ethyl mandelate of LipA and A75G was examined. The activity increased with the temperature from 25℃to 37℃, the maximal enzyme activity was observed at 37℃. Further increasing of temperature resulted in the decrease of enzyme activity. However, the enantioselectivity of both lipases increased when the temperature was above 37℃. LipA and A75G had the highest E value at 42℃and 45℃, respectively. The E values of both lipases increased in the pH range of 6.5-9.5, and the highest enantioselectivity was at pH 9.5 (E=2.5 and 6.0, respectively), then decreased from pH 9.5 to 11.5. Both lipases showed optimum activity at pH 8.5 and the activity decreased strongly above pH 10.5 or below pH 7.5. At lower pH (pH 6.5-8.5), both lipases had similar activity. However, when the pH was above 8.5, LipA had higher activity than that of A75G. The enantioselectivity was not significantly affected by the organic solvents, but the enzymatic activity had obvious difference in different solvents. The highest activity was obtained by using DMSO as solvent. When acetone was used as solvent, both lipases had highest E values, but the enzymatic activity was lower compared with other solvents. The relative activity in acetone was about 42 % of it in DMSO.3. Enhancement the enantioselectivity of A75G by site-saturation mutagenesisAlthough the A75G showed a higher enantioselectivity than that of LipA, the E value was still lower for a real application. It had been reported that Asn 18 residue was important for the enantioselectivity of LipA. To enhance the enantioselectivity of A75G, site-saturation mutagenesis at this position and high throughput screening based on GC were used in this study. (S)-2-octyl-(R,S)-mandelate was used as the screening substrate. There were 200 colonies in the site-saturation mutagenesis library. Twenty variants with increasing enantioselectivities for (S)-2-octyl-(R)-mandelate were obtained. We selected one of the best from mutant library, it was found to have an Asn to Ser displacement. The E value of the variant A75G/N18S was beyond 19. Single mutant N18S was constructed and detemined with A75G to find the contribution to enantioselectivity of the two mutant residues.We found an interesting result that the enantioselectivity of A75G/N18S for the resolution of (R,S)-2-octanol was enhanced simultaneously. In this study, R, S-2-octyl-R-mandelate was used as substrate and the hydrolysis was detected. To investigate whether the increase of enantioselectivity was caused by the mandelic acid, (R,S)-2-octyl-(S)-mandelate was also used to determine the enantioselectivity. The E value of A75G/N18S was 3-fold that of LipA in favor of (S)-2-octyl-R-mandelate. For the (R, S)-2-octyl-(S)-mandelate, A75G/N18S also displayed a higher enantioselectivity than LipA. The enhancement of enantioselectivity of variant A75G/N18S might because the (S)-2-octyl-mandelate was used in screening during the site-saturation mutagenesis. It seemed to be a new strategy to enhance enantioselectivity for two type enantiomers simultaneously by using the combined substrate in high throughput screening. For example, A represented the rac-acids or rac-alcohols, B represented the corresponding rac-alcohols or rac-acids. The ester R-A-S-B (if the R-A and S-B were the rapid reaction enantiomers, respectively) was used as the substrate in high throughput screening. The mutant enzymes might have good enantioselectiviy for both A and B if the S-B or R-A as the chiral reagents in the resolution reactions.We determined the characteristics of LipA and mutant lipases. A75G and A75G/N18S had lower specific activity than that of LipA and N18S, which meant that the Ala to Gly displacement decreased the activity of LipA. The effect of temperature on the lipases was determined. The optimum temperature of LipA and A75G was 42℃, while the optimum temperature of N18S and A75G/N18S was 45℃. After incubating at temperature range of 20-60℃for 1 h, the residual activities of lipases were measured. LipA and N18S retained above 90 % activity at temperatures below 40℃and about 70 % activity at 45℃, whereas A75G and N18S/A75G showed 40 % activity at 45℃. A75G and N18S/A75G displayed marked sensitivity when the temperatures were over 45℃. The optimum pH for lipases was determined from pH 6.5 to pH 11.5 in various buffers. The optimum pH of LipA and A75G was 8.5, and the optimum pH of N18S and A75G/N18S was 8.5. All lipases were stable at an alkaline pH range of 8.5-10.5 where they showed 85-95 % residual activity after 24 h incubation at room temperature. At lower pH values (pH 5.5-7.5) the residual activity of LipA was below 20 %, and the N18S was slightly higher than LipA. However, the residual activities of A75G and A75G/N18S were beyond 40 %.4. Increasing activity of a novel Bacillus subtilis lipase (BSL2) by directed evolutionWe have cloned, expressed and characterized a novel Bacillus subtilis lipase (BSL2). BSL2 has 90 % homology with LipA, but the special activity of BSL2 was lower than that of LipA. The specific activity of BSL2 was increased by directed evolution. Through two cycles of error prone PCR, coupled with a sensitive high-throughput screening method, a mutant named 3-1B2 was obtained. Its catalytic activity was 4.5-fold compared to that of wild lipase BSL2. DNA sequencing revealed that two amino acids were changed. Further experiments showed that the thermostability and pH stability of 3-1B2 were slightly increased than those of BSL2, and the optimum temperature and pH of the mutant lipase were similar to those of wild lipase. The molecular structures of BSL2 and 3-1B2 were homologously modeled based on the crystal structure of BSLA and docked with the substrate. The results showed that the binding energy of 3-1B2 with substrate was 1.29 kcal/mol, lower than that of BSL2, and the distance between catalytic residue Ser77 of 3-1B2 and substrate was 0.278 nm,lower than that of BSL2 (0.319 nm). This implied that small distance between Ser77 and substrate might increase reaction rate and then cause an increase in activity.