Method for cell-free protein synthesis involved with pH control with amino acid decarboxylase

Abstract

A method for cell-free protein synthesis is characterized in that pH is controlled by using an enzyme. For example, by using an amino acid decarboxylase, the pH is controlled according to removal of hydrogen ions that are produced during regeneration of ATP. The method for cell-free protein synthesis of the present invention has an advantage that not only the expression amount of protein is enhanced but also the expressed protein can be directly used for activity analysis without undergoing any separation or purification.

Claims

1. A method for cell-free protein synthesis, comprising: preparing a reaction solution comprising a cell extract, and an energy source comprising adenosine triphosphate, a buffer solution, an amino acid, a gene, and a carbon source; and adding amino acid decarboxylase to the reaction solution to control pH of the reaction solution by removing hydrogen ions that are produced during regeneration of the adenosine triphosphate from the carbon source, wherein the amino acid decarboxylase is glutamic acid decarboxylase having the amino acid sequence of SEQ ID NO: 1, and wherein the glutamic acid at the 89th position of the amino acid sequence of SEQ ID NO: 1 is substituted with glutamine and the amino acids at the 465th position and the 466th position of the amino acid sequence of SEQ ID NO: 1 are deleted.

2. The method for cell-free protein synthesis according to claim 1, wherein the pH is controlled to 6.5 to 8.0.

3. The method for cell-free protein synthesis according to claim 1, wherein the carbon source is at least one selected from the group consisting of monosaccharides, disaccharides, polysaccharides, polyhydric alcohols, and organic acids.

4. The method for cell-free protein synthesis according to claim 1, wherein the carbon source is glucose.

5. A method for cell-free protein synthesis, comprising: preparing a reaction solution comprising a cell extract, a gene, and an energy source comprising adenosine triphosphate, a buffer solution, an amino acid, and a carbon source; and adding amino acid decarboxylase to the reaction solution to remove hydrogen ions that are produced during regeneration of the adenosine triphosphate from the carbon source, wherein the amino acid decarboxylase is at least one selected from the group consisting of glutamate decarboxylase EC 4.1.1.15, arginine decarboxylase EC 4.1.1.19, lysine decarboxylase EC 4.1.1.18, aspartate 4-decarboxylase EC 4.1.1.12, valine decarboxylase EC 4.1.1.14, histidine decarboxylase EC 4.1.1.22, tyrosine decarboxylase EC 4.1.1.25, aromatic-L-amino acid decarboxylase EC 4.1.1.28, phenylalanine decarboxylase EC 4.1.1.53, and methionine decarboxylase EC 4.1.1.57.

6. A method for cell-free protein synthesis, comprising: preparing a reaction solution comprising a cell extract, a gene, and an energy source comprising adenosine triphosphate, a buffer solution, having a concentration of not more than 240 mM, an amino acid, and a carbon source; and adding glutamic acid decarboxylase to the reaction solution to control pH of the reaction solution to 6.5 or higher by removing hydrogen ions that are produced during regeneration of the adenosine triphosphate from the carbon source, wherein the glutamic acid decarboxylase has the amino acid sequence of SEQ ID NO: 1, and wherein the glutamic acid at the 89th position of SEQ ID NO: 1 is substituted with glutamine, and the amino acids at the 465th position and the 466th position of SEQ ID NO: 1 are deleted.

Description

BRIEF DESCRIPTION OF THE DRAWING

(1) FIGS. 1A and 1B graphs for determining the concentration of synthesized sfGFP (g/ml) (FIG. 1A) and degree of pH decrease depending on the concentration of a chemical buffer solution (HEPES buffer) (FIG. 1B) (.diamond-solid. 57 mM, .square-solid. 118 mM, .box-tangle-solidup. 179 mM, and .circle-solid. 240 mM) in cell-free protein synthesis.

(2) FIG. 2 is a graph illustrating the average expression amount of sialyl transferase (ST) which has been obtained by cell-free protein synthesis at condition of 240 mM buffer solution (HEPES buffer).

(3) FIGS. 3A and 3B are graphs for determining the effect of controlling pH by glutamic acid decarboxylase (FIG. 3A) and the expression amount of sfGFP protein at each condition when the cell-free protein synthesis has been performed at condition of 57 mM buffer solution (HEPES buffer) (FIG. 3B). Control is a control group in which glutamic acid decarboxylase is not added; .box-tangle-solidup. H465A represents addition of mutant glutamic acid decarboxylase of which histidine at the 465.sup.th position is substituted with alanine; .circle-solid. 465-466 represents addition of mutant glutamic acid decarboxylase of which histidine at the 465.sup.th position and threonine at the 466.sup.th position are deleted; and .square-solid. Glu89Gln/465-466 represents addition of mutant glutamic acid decarboxylase of which glutamic acid at the 89.sup.th position is substituted with glutamine and histidine at the 465.sup.th position and threonine at the 466.sup.th position are deleted.

(4) FIG. 4 is a graph illustrating the expression amount of sfGFP protein depending on the concentration of a mutant glutamic acid decarboxylase of which glutamic acid at the 89.sup.th position is substituted with glutamine and histidine at the 465.sup.th position and threonine at the 466.sup.th position are deleted

(5) FIG. 5 is a graph for comparing the change in pH and protein expression amount over time in cell-free protein synthesis in which 1.6 mg/ml GAD (Glu89Gln/465-466) and 240 mM buffer solution (HEPES-KOH) are used.

(6) FIG. 6 is a graph for determining the time for glutamic acid depletion at each initial concentration of glutamic acid, in which the depletion is caused by cell-free protein synthesis.

(7) FIG. 7 is a drawing illustrating the assay process including establishment of the library of an embodiment of the present invention, cell-free protein synthesis using an amino acid decarboxylase, and analysis of the synthesized enzyme.

(8) FIG. 8 is a drawing illustrating the process of cell-free protein synthesis according to one embodiment of the present invention in which glutamic acid decarboxylase and glutamic acid are used.

(9) FIG. 9 is a graph illustrating the amount of randomly-selected mutant sialyl transferase which has been synthesized by cell-free protein synthesis according to one embodiment of the present invention. WT represents a wild type sialyl transferase and NoDNA represents a control group in which cell-free protein synthesis is carried out with exclusion of DNA.

(10) FIG. 10 is a diagram illustrating the conditions for PCR which is used for establishing the gene library of sialyl transferase.

(11) FIG. 11 is a graph illustrating the result of screening the active sialyl transferase, in which the cell-free protein synthesis system is used for the screening.

DETAILED DESCRIPTION

(12) To achieve one or more of the aspects of the present invention in relation to the method for cell-free protein synthesis, an embodiment of the present invention provides a method for cell-free protein synthesis which is characterized in that pH is controlled by adding an enzyme to a reaction solution for cell-free protein synthesis.

(13) The reaction solution for cell-free protein synthesis comprises a cell extract, a gene, an energy source, a buffer solution, an amino acid, or the like, and by further comprising an amino acid such as glutamic acid, arginine, or lysine, it is possible to use those amine acids as an enzyme substrate.

(14) The aforementioned enzyme indicates an amino acid decarboxylase, which is a hydrolyzing enzyme to produce amines by acting on various amino acids and removing a CC bond to generate carbon dioxide and amines which correspond to each amino acid.

(15) Preferred examples of the amino acid decarboxylase include at least one selected from glutamate decarboxylase EC 4.1.1.15, arginine decarboxylase EC 4.1.1.19, lysine decarboxylase EC 4.1.1.18, aspartate 4-decarboxylse EC 4.1.1.12, valine decarboxylase EC 4.1.1.14, histidine decarboxylase EC 4.1.1.22, tyrosine decarboxylase EC 4.1.1.25, aromatic-L-amino acid decarboxylase EC 4.1.1.28 acting on phenyl alanine, tryptophane, tyrosine or the like, phenylalanine decarboxylase EC 4.1.1.53, and methionine decarboxylase EC 4.1.1.57. More preferably, it is at least one selected from glutamate decarboxylase EC 4.1.1.15, arginine decarboxylase EC 4.1.1.19, and lysine decarboxylase EC 4.1.1.18. Even more preferably, it is glutamic acid decarboxylase having an amino acid sequence of SEQ ID NO: 1 of which the glutamic acid at the 89.sup.th position is substituted with glutamine and the amino acids at the 465.sup.th position and the 466.sup.th position are deleted, but not limited thereto. It is possible that part of the amino acids of the amino acid decarboxylase may be deleted, substituted, or inserted, and amino acids in modified form like phosphorylation or methylation form may be also included.

(16) As for the aforementioned pH, it is preferable that the pH is not decreased to 6.5 or lower. It is more preferable that the pH is controlled to 6.5 or 8.0.

(17) The aforementioned amino acid decarboxylase can control pH by removing the hydrogen ions that are produced during regeneration of ATP from a carbon source. It is preferable that the carbon source is at least one selected from monosaccharides, disaccharides, polysaccharides, polyhydric alcohols, and organic acids. More preferably, it is glucose, but not limited thereto.

(18) Also, the method for controlling pH using enzyme is not limited to cell-free protein synthesis. It may also be used for controlling pH during enzymatic reactions inside or outside cells.

(19) Also provided by the present invention is a library of proteins that are synthesized by the above method for cell-free protein synthesis. The library of proteins is preferably a library of sialyl transferase proteins, but not limited thereto. The sialyl transferase is an enzyme which plays a role of attaching sialic acid to a terminal of a sugar chain of a glycoprotein, and sialyl lactose and sialyl oligosaccharides derived from the enzyme are characterized in that they can be also a major source for supplying sialic acid to a body tissue, a skin, a brain, or the like.

(20) Still also provided by the present invention is a method for screening an active type protein which is characterized in that an active type protein is screened by having, as a subject, a library of proteins that are synthesized by the method for cell-free protein synthesis of an embodiment of the present invention. The protein library is characterized in that it allows screening without having any separation or purification of proteins.

(21) Herein below, the present invention is explained in greater detail in view of the Examples. However, it would be evident for a person having common knowledge in the pertinent art that the following Examples are given only for specific exemplification of the present invention and by no means the scope of the present invention is limited to those examples.

EXAMPLES

(22) [Materials and Experimental Methods]

(23) 1. Preparation of Cell Extract (S12 Extract)

(24) 5 ml of BL21-Star (DE3) derived from E. coli was inoculated to a LB medium and incubated for 12 hours at 37 C. After that, it was subjected to overnight subculture at 37 C. in 40 ml of 2YTPG medium, and then inoculated to a fermentation tank containing 4 L of 2YTPG followed by culture at the same temperature. When the absorbance (OD.sub.600) becomes 0.6, 1 mM isopropylthio--D-galactoside (IPTG) was finally added to the fermentation tank to express T7 RNA polymerase. The cell culture was terminated when the absorbance (OD.sub.600) becomes 4.5, and according to centrifuge (4,500 rpm, 15 minutes, 4 C.), only the E. coli cell pellets were collected from the medium.

(25) The collected E. coli cells were subjected 3 times to a process of thorough washing of the cells by adding 20 mM buffer solution A [10 mM Tris-acetate buffer (pH 8.2), 14 mM magnesium acetate, 60 mM potassium glutamate, 1 mM dithiothreitol (DTT), 0.05% (v/v) 2-mercaptoethanol (2-ME)] per gram of the cells.

(26) 12.7 ml of the buffer solution B (buffer solution A from which 2-ME is removed) was added per 10 grams of the E. coli cells which have been washed as described above. After dispersing them homogeneously, the cells were disrupted using a press (French Pressure Cell Press, Thermo Scientific) at constant pressure (12,000 psi). The cell lysate was subjected to centrifuge (30,000 rcf, 30 minutes, 4 C.) to obtain a supernatant, which was then incubated at 37 C. for 30 minutes. After aliquoting the supernatant in a small amount, it was stored in an ultra-low temperature freezer (80 C.) until use.

(27) 2. Cell-Free Protein Synthesis and Determination

(28) The cell-free protein synthesis reaction was carried out according to addition to a small tube of 1.75 ml and reaction in an incubator at 30 C.

(29) 57 mM Buffer solution (HEPES-KOH, pH 8.2), 1.4 mM ATP, 1.0 mM each of CTP, GTP, and UTP, 1.8 mM DTT, 90 mM potassium glutamate, 80 mM ammonium acetate, 8 mM magnesium acetate, 20 mM potassium phosphate, 34 g/mL 1-5-formyl-5,6,7,8-tetrahydrofolic acid (folinic acid), 3.2 mM each of 20 amino acids, 2% polyethylene glycol (PEG) 8000, 80 mM D-glucose, 13.3 g/mL plasmid, and 27% (v/v) S12 extract were used for the reaction solution for cell-free protein synthesis.

(30) Concentration of the above buffer solution (HEPES-KOH, pH 8.2), amino acids, glucose, and L-glutamate was adjusted as necessary. Total amount of the produced proteins was determined based on the radioactivity of a radioisotope after TCA-precipitation.

(31) Quantification of active type sfGFP was performed by measuring the fluorescence intensity using VICTOR X3 multilabel plate reader (Perkin-Elmer, Waltham, Mass.). Size of the produced sfGFP was determined based on Coomassie blue-stained Tricine-SDS-PAGE gels.

(32) The change in pH was recorded by measuring the pH every hour using 90 l of a reaction solution and a micro pH electrode (i.e., micro-combination pH-electrode; InLab 423, Mettler-Toledo GmbH, Switzerland).

(33) 3. Cell-Free Protein Synthesis using Glutamic Acid Decarboxylase (GAD) and Purification

(34) Each gene sequence of the wild type GAD enzyme and the mutant GAD (Glu89Gln/465-466) was cloned in the pET28b vector, and used for transformation of E. coli BL21 (DE3). Each bacterial cell was cultured at a temperature of 37 C. in 500 ml LB medium in which kanamycin is included. For the expression of GAD, 1 mM IPTG was added when OD.sub.600 value reaches 2.5. After further culturing the cells for 2 hours at the same conditions, the cells were collected and washed with the buffer solution A. The cell aggregate was then disintegrated by using the buffer solution B, and disrupted at a pressure of 12,000 psi by using a press (French Pressure Cell Press). According to centrifuge at 30,000 rcf and 4 C. for 30 minutes, a supernatant was collected.

(35) The collected supernatant was separated and purified by using Ni-NTA agarose beads (Qiagen), and then concentrated with the buffer solution B by using VIVA-spin 20 (MWCO 10 kDa, Sartorius). Thereafter, the protein concentration was measured by Bradford method, and the concentrate was stored at 80 C. until use.

Example 1. Determination of Amount of Synthesized Protein and Change in pH Depending on Concentration of Buffer Solution of Cell-Free Protein Synthesis

(36) To determine a pH change depending on the concentration of a buffer solution of cell-free protein synthesis, cell-free protein synthesis of sfGFP protein was carried out by using a buffer solution of 57 mM, 118 mM, 179 mM, or 240 mM.

(37) As shown in FIGS. 1A and 1B, it was found that the decrease amount of pH is reduced and the protein synthesis amount is increased as the concentration of the chemical buffer solution (I.e., HEPES buffer) is increased. At that time, in order to maintain the pH level of 6.5 which is the minimum condition required for the reaction using an excess amount of glucose (80 mM), concentration of the buffer solution (I.e., HEPES buffer) was optimized, i.e., 240 mM or less. As a result of carrying out the cell-free protein synthesis at a condition including 240 mM buffer solution, it was confirmed that sfGFP protein is obtained at a level of 2 mg/ml, approximately.

(38) After that, for screening the activity of a sialyl transferase, by using 96 sialyl transferase gene libraries, cell-free protein synthesis was carried out at a condition including 240 mM buffer solution which has been optimized as described above.

(39) As shown in FIG. 2, when a library of 96 types of sialyl transferase is used for cell-free protein synthesis in a 96-well plate, it was found that the sialyl transferase is expressed at a level of 630101 g/ml. However, according to the reaction for measuring the activity in which a pH indicator and a substrate are used, no color change was observed. It is believed that no color change of a pH indicator was based on the buffer activity of the buffer solution at high concentration (i.e., HEPES buffer, 240 mM) against the change in hydrogen ion concentration which occurred even in a solution for measuring the activity.

(40) As such, after separating and purifying the sialyl transferase, it was determined whether or not the activity analysis can be made with the enzyme. As a result, in case of the separated and purified enzyme, color change was observed depending on the activity of the protein.

(41) It was confirmed based on Example 1 that, for cell-free protein synthesis, a buffer mechanism exhibiting no interference in activity analysis is required under pH control.

Example 2. Determination of Effect of Controlling pH Change by Glutamic Acid Decarboxylase in Cell-Free Protein Synthesis and Expression Amount of sfGFP Protein

(42) As a buffer mechanism which exhibits no interference in activity analysis under pH control of cell-free protein synthesis, an amino acid decarboxylase was used.

(43) Numerous microorganisms have an intrinsic buffer system for their survival at acidic conditions. For example, in case of E. coli, several kinds of an amino acid decarboxylase (i.e., arginine decarboxylase, glutamic acid decarboxylase, and lysine decarboxylase) are induced and expressed at acidic conditions. Those enzymes are known to remove hydrogen ions while dissociating the carboxyl group from each amino acid and have a function of maintaining intracellular pH of E. coli even at acidic conditions. By taking advantage of such characteristics, the anti-acidic mechanism of those microorganisms was applied to control the pH change in a cell-free protein synthesis system.

(44) In this Example 2, to remove the hydrogen ions that are generated according to a glycolysis process in a cell extract, glutamic acid decarboxylase (GAD) was used. One molecule of glutamic acid plays the role of removing the same mole number of hydrogen ions while the acid is converted into one molecule of derivatized amino acid (-aminobutyric acid) by the above enzyme. As such, when GAD and glutamic acid are added to a solution for cell-free protein synthesis, the hydrogen ions that are generated during the ATP regeneration process are removed so that the pH can be stably maintained.

(45) However, the system for controlling pH by using GAD shows disappearance of the buffer effect as soon as the glutamic acid is depleted. Furthermore, as the wild type GAD activity of pH-dependent E. coli exhibits the enzyme activity and protein cooperativity at pH of from 3.8 to 4.6, the GAD activity is not shown at pH of from 6.5 to 8.0, which is the optimum pH of a cell-free protein synthesis system.

(46) Recently however, it is luckily known that the pH dependency of the GAD derived from E. coli is decided by several specific amino acid residues. Namely, it is noted that the pH dependency of the GAD enzyme can be modified based on engineering of the enzyme. As an example, according to the studies carried out by Ho, et. al., it was shown that, when glutamic acid as the 89.sup.th amino acid of E. coli GAD is substituted with glutamine, the activity can be exhibited even at a pH in a neutral range. Furthermore, when the pH dependent cooperativity is removed by deleting the histidine and threonine, which are the 465.sup.th and the 466.sup.th amino acid, respectively, it was shown that the pH range in which the activity is shown can be broadened while the pH dependent cooperativity is removed. According to application of this effect based on mutation, it was able to obtain a mutant GAD (Glu89Gln/465-466) which exhibits the activity at a pH close to neutral pH and also in a wide pH range.

(47) For the pH control by an enzyme, the inventors of the present invention decided to use the mutant GAD (Glu89Gln/465-466) as described above, and it was expected that the pH decrease is reduced in a cell-free protein synthesis reaction in which glucose is used.

(48) Accordingly, in this Example 2, the effect of controlling the pH change by glutamic acid decarboxylase (GAD) in cell-free protein synthesis and the expression amount of sfGFP protein were determined.

(49) According to the result, it was found that the amount of the protein obtained by a cell-free protein synthesis reaction is significantly increased while the time during which neutral pH is maintained has also increased. In a case in which 57 mM HEPES-KOH buffer solution is used as a control group, pH of the reaction solution decreases to 6.5 or less within 80 minutes. However, when the GAD protein at 0.4 mg/ml is added to the same reaction solution, the time till to have pH of 6.5 has increased to about 130 minutes (FIG. 3A).

(50) As shown in the above, according to the reaction in which the pH decrease rate is lowered, the amount of the synthesized protein has increased simultaneously, showing the productivity of about 2 times or so. On the other hand, in case of adding the wild type GAD, the effect of controlling pH was hardly shown, and the protein productivity was almost the same as the reaction of a related art (FIG. 3B). Such result indicates that, in terms of the pH decrease of a reaction solution, the mutant GAD (Glu89Gln/465-466) is more effective than the wild type GAD.

(51) Furthermore, when the concentration of GAD (Glu89Gln/465-466) in a reaction solution for cell-free protein synthesis is increased to 1.6 mg/ml or more, even higher anti-oxidizing activity is obtained so that the effect of having higher protein productivity in a reaction solution for cell-free protein synthesis was shown (FIG. 4).

(52) Furthermore, the pH change and the amount of synthesized protein over time were compared when each of 1.6 mg/ml GAD (Glu89Gln/465-466) and 240 mM HEPES-KOH is used for the reaction solution for cell-free protein synthesis. As a result, it was found as shown in FIG. 5 that, even if the HEPES-KOH as a buffer solution is completely removed from the reaction solution, the pH can be fully maintained if GAD is present at constant concentration or higher.

(53) Furthermore, as the initial concentration (166 mM) of glutamic acid included in the reaction solution completely disappears according to the reaction for 6 hours, a buffer-free condition is yielded (FIG. 6). Thus, it was found that the time for cell-free protein synthesis is preferably as long as 6 hours if glutamic acid is used at initial concentration of 166 mM. In this regard, further addition of glutamic acid after 6 hours to have a progress of the reaction is not excluded, and it is expected that the time for cell-free protein synthesis reaction can be extended by suitably regulating the glutamic acid concentration.

Example 3. Establishment, Expression, and Activity Analysis of Sialyl Transferase Library

(54) In this Example 3, establishment of a sialyl transferase library, cell-free protein synthesis using an amino acid decarboxylase, and an assay for analyzing the activity of synthesized enzyme were performed according to the process shown in FIG. 7.

(55) In Example 3, according to a method for cell-free protein synthesis which is characterized in that pH is controlled by using glutamic acid decarboxylase (GAD) and glutamic acid, about 10,000 gene libraries were expressed and activity of the expressed proteins was screened (FIG. 8).

(56) As illustrated in FIG. 9, it was confirmed that the mutant sialyl transferases, which have been obtained by cell-free protein synthesis by using the GAD system in which pH is controlled using glutamic acid decarboxylase (GAD) and glutamic acid, are synthesized at the almost same level. Accordingly, it was found that the result of the activity analysis which is obtained by collecting a constant amount of the reaction solution is not based on a difference in the expression amount.

(57) It was also confirmed that the result obtained by applying the sialyl transferase, which has been produced by cell-free protein synthesis using GAD and used without undergoing a separation or purification process, shows the same profile as the result of the activity analysis which has been obtained with the purified sialyl transferase.

(58) As such, it is expected that pH-based enzyme screening which uses a library of proteins synthesized by the cell-free protein synthesis method of an embodiment of the present invention in which pH is controlled by using an amino acid decarboxylase can be suitably carried out.

(59) For establishing a sialyl transferase gene library, random mutation of the codons of threonine (Thr) at the 265.sup.th amino acid, arginine (Arg) at the 313.sup.th amino acid, and threonine (Thr) at the 357.sup.th amino acid of SEQ ID No. 2 was carried out by a PCR method, and the primers used for PCR are shown in Table 1. Among the sites that are expected to be an active site of the sialyl transferase, amino acids at three positions were selected and a randomly mutated gene library was established followed by expression to select a mutant gene of which enzyme activity is enhanced compared to the wild type.

(60) TABLE-US-00001 TABLE1 Primersforestablishingsialyltransferasegenelibrary Primername Startpoint Nucleotidesequence(5->3) T7promoter Forward TCGATCCCGCGAAATTAATACGACTCACTATA GG(SEQIDNO:3) 1.sup.stfragment Backward AAAGATAAATTTAGCTTGTTGCACTTC(SEQID NO:4) 2.sup.ndfragment Forward GAAGTGCAACAAGCTAAATTTATCTTTNNSGG CACG(SEQIDNO:5) 2.sup.ndfragment Backward AGGATGCCCTTTAAAGTAGATTTT (SEQIDNO:6) 3.sup.rdfragment Forward AAAATCTACTTTAAAGGGCATCCTNNSGGTGG TGAAATTAATGACTACATTCTGA (SEQIDNO:7) 3.sup.rdfragment Backward TGAACTTGCAACACCACCCAC(SEQIDNO:8) 4.sup.thfragment Forward GTGGGTGGTGTTGCAAGTTCANNSTATTTC (SEQIDNO:9) T7terminator Backward CAAAAAACCCCTCAAGACCCGTTTA(SEQID NO:10)

(61) The conditions for carrying out the PCR like temperature and time are shown in FIG. 10. The genes obtained by the PCR were cloned in pIVEX2.3d vector. After obtaining a colony in which each of the above threonine (Thr) at the 265.sup.th amino acid, arginine (Arg) at the 313.sup.th amino acid, and threonine (Thr) at the 357.sup.th amino acid is substituted with a different gene sequence, a gene library was established. By using a colony collector (K3, K biosystems), 10,000 colonies were collected from the library and then inoculated to a 96 well plate in which 200 l of LB medium is contained. Then, the resulting 10.sup.4 plates of a 96 well plate were subjected to culture by shaking for 15 hours at 37 C.

(62) The cultured cells were transferred, each in an amount of 1.5 l, to 26 plates of a 384 well plate in which 23.5 l of the PCR reaction solution is contained. Thereafter, each gene amplified by the PCR was transferred, each in an amount of 3.0 l, to a new 384 well plate in which 19.5 l of the reaction solution for cell-free protein synthesis is contained. The plate was added to a shaker at 30 C. with high humidity and subjected to cell-free protein synthesis reaction for 3 hours.

(63) Immediately after the completion of the cell-free protein synthesis, the reaction solution contained in the sialyl transferase library which has been established in Example 3 was diluted 3 times with 5 mM Tris-Cl (pH 8.5), and then 1.8 l of the diluted solution was transferred to a 384 well plate, in which 58.5 l of activity screening solution (5 mM Tris-Cl pH 8.5, 4 mM CMP-NeuAc, 0.4 mM cresol-red, and 4 mM lactose) is added, and incubated for 15 minutes at room temperature. OD.sub.600 value of the library was measured using a plate reader (Victor 3, PerkinElmer), and a sample showing the highest absorbance was selected. For the transferring step of each of the above processes, an automated liquid handling system (JANUS Automated Workstation, PerkinElmer) was used.

(64) As shown in FIG. 11, it was able to identify a sialyl transferase with significantly increased activity based on the expression of a sialyl transferase from 10,000 genes and screening of the active type proteins.