Vol.4 No.4 2012

40/62

Research paper : Efficient production of active form of vitamin D3 by microbial conversion (Y. Yasutake et al.)−230−Synthesiology - English edition Vol.4 No.4 (2012) enzyme first hydroxylated the carbon-25 of VD3 and then hydroxylated the carbon-1 of 25(OH)VD3. Also, a small amount of 26(OH)VD3 was detected, and these substances matched the results detected in the cellular VD3 conversion by P. autotrophica. We determined that this P450 was the enzyme actually responsible for microbial conversion, and named it vitamin D3 hydroxylase (Vdh). The P450 will be called Vdh hereafter. 3.3 Intracellular conversion using the recombinant expressionNext we constructed the microbial conversion system that accomplished the VD3 hydroxylation, using the recombinant cell of R. erythropolis. The VD3 hydroxylation activity was very low with a single expression of Vdh, and it was necessary to coexpress some kind of redox partner protein. Therefore, we constructed a thiostrepton-inducible expression vector that contained genes encoding the Vdh and redox partner proteins (ferredoxin and ferredoxin reductase) derived from R. erythropolis. Then, co-expression was conducted in the R. erythropolis cell, and VD3 was added to the culture. As a result, it was confirmed that the active forms of VD3 were produced when the R. erythropolis cell was used. It has been reported that the redox partner that is capable of most efficiently supplying the electrons to P450 is not necessarily the protein with which P450 couples in the cell of the original organism[7]. This is thought to be because the electron transfer efficiency is affected drastically by the slight difference in the intracellular environment or the intracellular expression level of the gene. Therefore, we conducted the conversion tests by inserting various electron transfer protein genes to the aforementioned co-expression vectors, and looked for the redox partner that showed high VD3 hydroxylation activity. As a result, it was found that the proteins called AciB and AciC from Acinetobacter were the most compatible partner to Vdh. 3.4 Two different approaches to enzyme improvementIn general, the enzymes produced by organisms are catalysts that specifically respond to certain substrates and are exclusive to some specific reaction. However, since VD3 is not found in the soil where the P. autotrophica exists, it is thought that Vdh is not an enzyme that evolved for (is specific to) the hydroxylation and metabolism of VD3. In fact, the VD3 hydroxylation activity of the isolated and purified enzyme is fairly lower than the activity of the P450 with the specific function involved in the biosynthesis of some substance. Therefore, the VD3 hydroxylation activity of Vdh is not at all optimized as an enzyme, and we believe it can be improved further.There are two completely different approaches when introducing variations to improve the enzyme. One is the rational design where the 3D structure of the protein is analyzed and the mutation is introduced based on this structural information. While this is a powerful method in the case where the correlation of the structural functions is clear, since the 3D structure of the protein is a complex system composed of multiple parameters, there may be no simple correlation between the amino acid residue and the function. The other approach is the evolutionary engineering (directed evolution) where the genetic variation library to which the random mutation is introduced is created to screen the variants with improved performance. While much effort will be required to create the library and to verify the variations by assay, this may enable extracting the variations that may improve the enzyme function at any part of the sequence. In our research, we conducted improvement of enzymes using both methods, without choosing one of the two approaches (Fig. 2). As a result, the advantages of both the strategy of variation introduction based on structure and the strategy based on evolutionary engineering could be utilized, and we were successful in creating a useful variant. 3.4.1 High activation of the enzymeThe mutant that significantly improved the VD3 hydroxylation activity was constructed by combining mutations generated by directed evolution. The quadruple mutant (Vdh-K1) that had the highest improvement showed about 12 times increase of 25-hydroxylase activity and about 25 times increase of 1-hydroxylase activity compared to the wild type Vdh (Vdh-WT)[6][8]. Interestingly, the four mutations were located far from the active site, and it is difficult to find such mutations by rational design. The discovery of Vdh-K1 was the result of the maximization of the benefits of evolutionary engineering unfettered by structural information. On the other hand, we were able to infer why such variations brought about major activity improvement through the analysis of the 3D structure. Major structural change was observed between the Vdh-WT and Vdh-K1, and three of the four mutations might have induced such structural changes. This suggested that the activity increase of Vdh was not caused by an optimization of the substrate binding pocket, but by the orchestration of the conformational equilibrium between open and closed forms (Fig. 3)[8]. P450 is an enzyme involved in the detoxification and the biosynthesis of the secondary metabolites and various substances in nature, and there are many molecular species of the enzyme with wide substrate specificity. The conformational shift by introduction of a few mutations observed in our research may be a mechanism that allows in nature to adapt P450 to ever-emerging new conditions and substrates. Through the results obtained, the possibility of significantly improving the production efficiency of hydroxylated VD3 was found. However, due to the issues in membrane permeability of the substrate that will be described later, major increase in the production efficiency of hydroxylated VD3 had not been achieved simply by increasing the enzyme performance.

※このページを正しく表示するにはFlashPlayer9以上が必要です