(en)The present invention relates to a method for designing and preparing mutant transglutaminases on the basis of the three-dimensional structure of MTG derived from Streptoverticillium mobaraense (MTG), and the mutant MTG thus prepared. The present invention provides a method for modifying MTG on the basis of the three-dimensional structure, and transglutaminase having reactivity on the substrate improved by the method. In the present invention, the binding site of MTG for the substrate is extrapolated based on the three-dimensional structure obtained by X-ray crystal structure analysis of MTG crystals, and the mutant transglutaminases are designed and produced by replacing, inserting or deleting amino acid residues positioned at the substrate-binding site of the transglutaminase.

1.ApplicationNumber: US-50543806-A
1.PublishNumber: US-2006275872-A1
2.Date Publish: 20061207
3.Inventor: KASHIWAGI TATSUKI
SHIMBA NOBUHISA
ISHIKAWA KOHKI
SUZUKI EI-ICHIRO
YOKOYAMA KEIICHI
HIRAYAMA KAZUO

4.Inventor Harmonized: KASHIWAGI TATSUKI(JP)
SHIMBA NOBUHISA(JP)
ISHIKAWA KOHKI(JP)
SUZUKI EI-ICHIRO(JP)
YOKOYAMA KEIICHI(JP)
HIRAYAMA KAZUO(JP)

5.Country: US
6.Claims:
(en)The present invention relates to a method for designing and preparing mutant transglutaminases on the basis of the three-dimensional structure of MTG derived from Streptoverticillium mobaraense (MTG), and the mutant MTG thus prepared. The present invention provides a method for modifying MTG on the basis of the three-dimensional structure, and transglutaminase having reactivity on the substrate improved by the method. In the present invention, the binding site of MTG for the substrate is extrapolated based on the three-dimensional structure obtained by X-ray crystal structure analysis of MTG crystals, and the mutant transglutaminases are designed and produced by replacing, inserting or deleting amino acid residues positioned at the substrate-binding site of the transglutaminase.

7.Description:
(en)CROSS-REFERENCE TO RELATED APPLICATIONS
The present application is divisional application of U.S. Ser. No. 10/365,434, filed on Feb. 13, 2003, which is a continuation PCT/JP01/07038, filed on Aug. 15, 2001, which claims priority to JP 2000-247664, filed on Aug. 17, 2000, and JP 2000-396695-61793, filed on Dec. 27, 2000.


BACKGROUND OF THE INVENTION
The present invention relates to a method for designing and preparing mutant transglutaminase on the basis of the three-dimensional structure of transglutaminase derived from Streptoverticillium mobaraense (hereinafter referred to as “MTG”) determined by X-ray crystal structure analysis techniques, and the mutant MTGs thus prepared. MTG is widely used for processing foods or the like because it forms a gelatinus substance by forming crosslinking between proteins. The mutant MTG improved in the transglutaminase activity and heat stability contributes to reduce the required amount thereof, and also the mutant MTG having modified substrate specificity and optimum pH will allow the application of the enzyme to a new field.
Transglutaminase is an enzyme which catalyzes the acyl transfer reaction of γ-carboxamide group existing in a peptide chain of a protein. By reacting a protein with this enzyme, ε-(γ-Glu)-Lys crosslinking reaction and replacement reaction of Gln with Glu by deamidation may occur.
Transglutaminases from animals and those from microorganisms are so far known. The former is a Ca 2+ -dependent enzyme and it is widely distributed in organs, skin and blood of animals. This includes, for example, guinea pig liver transglutaminase [K. Ikura et al., Biochemistry, Vol. 27, p. 2898 (1988)], human epidermal keratin cell transglutaminase [M. A. Phillips et al., Proc. Natl. Acad. Sci., USA, Vol. 87, p. 9333 (1990)] and human blood coagulation factor XIII [A. Ichinose et al., Biochemistry, Vol. 25, p. 6900 (1990)]. As for the latter, the Ca 2+ -independent ones were found in Streptoverticillium bacteria such as Streptoverticillium griseocarneum IFO 12776, Streptoverticillium cinnamoneum sub sp. cinnamoneum IFO 12852 and Streptoverticillium mobaraense IFO 13819. Among these enzymes, the transglutaminase found in a culture supernatant of a variant of Streptoverticillium mobaraense is referred to as MTG (Microbial Transglutaminase). Also, a Ca 2+ -independent transglutaminase was found in Streptomyces lydicus NRRL B-3446 (JP-Kokai No. 10-504721).
MTG is a monomeric protein composed of 331 amino acids and having a molecular weight of about 38,000 [Journal of Biological Chemistry, vol. 268, p. 11565 (1993)]. A method for producing active MTG by secretion expression using E. coli or yeast was reported (JP-Kokai No. Hei 5-199883). There has also been reported a method for producing the same wherein MTG is expressed as a protein inclusion body in E. coli , then the inclusion body is solubilized with a protein denaturing agent and it is then regenerate through removing the denaturing agent to obtain active MTG (JP-Kokai No. 6-30771).
Unlike transglutaminases from animals, the transglutaminases from microorganisms, such as MTG, are Ca 2+ -independent. They are therefore utilized for the production of gelled foods such as jellies, as well as yogurts, cheeses or cosmetics in gel form, and also for the improvement of the quality of meats (JP-Kokai No. 64-27471). They are highly useful in the industry in that they are used as materials for heat stable microcapsules or as carriers for immobilized enzymes.
As for the conditions of the enzymatic reaction, for example, gelatinized foods cannot be gelled when the enzymatic reaction time is short, on the contrary, they become too hard to be the commercial products when the enzymatic reaction time is excessively long. Therefore, when MTG is used for producing gelatinized foods such as jellies; yogurts; cheeses; or cosmetics in gel form; or for improving the quality of meats, the concentration of the substrate and enzyme, reaction temperature and reaction time are controlled suitably for the intended product. However, as the foods and reagents produced by using MTG became various, it occurred that the intended products could not be produced in some cases by only controlling the concentration, temperature, time, etc. Thus, the modification of the enzymatic activity of MTG has become necesssary.
For modifying the enzymatic activity of MTG, it is required to produce a MTG mutant and to evaluate the activity, substrate specificity, etc. of the mutant to find out a good mutant. For producing mutants, it is necessary to manipulate wild type genes and, accordingly, it is prerequisite that a recombinant protein can be prepared. For MTG, a large-scale expression system using E. coli has been established (JP-Kokai No. 6-30771). However, MTG thus expressed in a large amount in E. coli is accumulated as insoluble protein inclusion bodies in the microbial cells. The denatured MTG in the protein inclusion body must be activated by refolding it after solubilizing it. Thus, it takes 2 to 3 weeks to prepare one mutant MTG. It has therefore been required to develop a method for reasonably modify MTG using some technique other than the techniques of randomly introducing the mutations through the entire molecule.
SUMMARY OF THE INVENTION
The object of the present invention is to provide a method for improving MTG.
Another object of the present invention is to provide a transglutaminase having an improved reactivity on substrates.
After intensive investigations made for the purpose of solving the above-described problems, the inventors have found that a transglutaminase improved in the reactivity on a substrate can be prepared on the basis of the three-dimensional structure. The present invention has been completed on the basis of this finding.
Namely, the summary of the present invention is as follows:
(1) A method for designing and preparing a mutant transglutaminase by extrapolating the binding site of MTG for a substrate on the basis of the three-dimensional structure obtained by X-ray crystal structure analysis and replacing, inserting or deleting amino acid residues positioned at the extrapolated substrate-binding site of the transglutaminase.
(2) A mutant MTG having modified substrate specificity, in which at least one of the following positions (residues positioned within 20 Å from active residue Cys 64 and exposed on the surface of the presumed binding site for acyl acceptor) in MTG amino acid sequences indicated as SEQ ID NO: 2 is modified: 1-5, 26, 28, 58-59, 62, 69, 74-75, 77, 79, 235-236, 238-244, 248-250, 252-254, 277-278, 282-287, 289, 291, 296,-297 and 300-304.
(3) A mutant transglutaminase having transglutaminase activity and modified substrate specificity, wherein when the alignment of the sequence of said mutant transglutaminase with the MTG amino acid sequence according to SEQ ID NO: 2 is carried out, at least one of the positions corresponding to the following positions (residues positioned within 20 Å from active residue Cys 64 and exposed on the surface of the presumed binding site for acyl acceptor) in the MTG amino acid sequence according to SEQ ID NO: 2 is modified: 1-5, 26, 28, 58-59, 62, 69, 74-75, 77, 79, 235-236, 238-244, 248-250, 252-254, 277-278, 282-287, 289, 291, 296,-297 and 300-304.
(4) A mutant transglutaminase having the transglutaminase activity and a modified substrate specificity, wherein when the alignment of the sequence of said mutant transglutaminase with the MTG amino acid sequence according to SEQ ID NO: 2 is carried out by treading method, at least one of the positions corresponding to the following positions (residues within 20 Å from active residue Cys 64 and exposed on the surface of the extrapolated binding site for acyl acceptor) in the MTG amino acid sequence according to SEQ ID NO: 2 is modified: 1-5, 26, 28, 58-59, 62, 69, 74-75, 77, 79, 235-236, 238-244, 248-250, 252-254, 277-278, 282-287, 289, 291, 296,-297 and 300-304.
(5) A mutant MTG or a mutant transglutaminase having changed substrate specificities through reducing the negative charge by modifying acidic amino acid residues located at the positions defined in any one of the above items (2) to (4) or amino acid residues adjacent to said acidic amino acid residues.
(6) A gene encoding the mutant MTG or the mutant transglutaminase defined in any one of the above items (2) to (5).
(7) A recombinant DNA containing the gene defined in above item (6).
(8) A microorganism harboring the recombinant DNA defined in above item (7).
(9) A method for producing a mutant MTG or a mutant transglutaminase, which comprises culturing the microorganisms defined in item (8) and collecting the mutant MTG or the mutant transglutaminase.
(10) A monoclinic MTG crystal belonging to P2 1 space group.
Namely, the present invention provides a method for rationally designing and preparing a mutant MTG on the basis of the three-dimensional structure of MTG, and the mutant MTG thus prepared.


BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a ribbon model showing the crystal structure of MTG.
FIG. 2 shows the surface roughness of the crystal structure of MTG observed in the same direction as that in FIG. 1 .
FIG. 3 shows the alignments of amino acid sequences of MTG and TG from Streptoverticillium cinnamoneum. The top sequence is from Streptoverticillium mobaraense (SEQ ID NO:2) and the bottom sequence is from Streptoverticillium cinnamoneum (SEQ ID NO:6).
FIG. 4 shows the alignments of amino acid sequences of MTG and TG from Streptoverticillium lydicus . The top sequence is from Streptoverticillium mobaraense (SEQ ID NO:2) and the bottom sequence is from Streptoverticillium cinnamoneum (SEQ ID NO:8).
FIG. 5 shows the primer sets used for the site-directed mutagenesis of S2Y•S2R•S2D mutants. The sequences appearing in this figure are represented in the Sequence Listing as follows: S2Y(s)=SEQ ID NO: 13, S2Y(as)=SEQ ID NO: 14, S2R(s)=SEQ ID NO: 15, S2R(as)=SEQ ID NO: 16, S2D(s)=SEQ ID NO: 17, S2D(as)=SEQ ID NO: 18, translated amino acid sequence for S2Y=SEQ ID NO: 30, translated amino acid sequence for S2R=SEQ ID NO: 31, and translated amino acid sequence for S2D=SEQ ID NO: 32.
FIG. 6 shows the primer sets used for site-directed mutagenesis of del 1-2 and del 1-3 mutants. The sequences appearing in this figure are represented in the Sequence Listing as follows: del 1-2(s)=SEQ ID NO: 19, del 1-2(as)=SEQ ID NO: 20, del 1-3(s)=SEQ ID NO: 21, del 1-3(as)=SEQ ID NO: 22, translated amino acid sequence for del 1-2=SEQ ID NO: 33, and translated amino acid sequence for del 1-3=SEQ ID NO: 34.
FIG. 7 shows 1H-15N HSQC spectrum of ovalbumin reacted with the wild type in the presence of 15 NH 4 Cl.
FIG. 8 shows the peak intensity of signal “a” in FIG. 7 in the glutamine residues of ovalbumin reacted with the wild type, Ser type, S2R, del 1-2 or del 1-3 in the presence of 15 NH 4 Cl, plotted vs. the reaction time.
FIG. 9 shows the peak intensity of signal “a” in FIG. 7 in the glutamine residues of ovalbumin reacted with the wild type, S2Y, S2R or S2D mutant in the presence of 15 NH 4 Cl, plotted vs. the reaction time.
FIG. 10 shows 1 H- 15 N HSQC spectrum of ovalbumin reacted with the wild type, Sg4 or Sg7 in the presence of 15 NH 4 Cl.
FIG. 11-1 to 11 - 50 show the atomic coordinates of MTG (SEQ ID NO: 2).


DESCRIPTION OF THE PREFERRED EMBODIMENTS
The present invention will be concretely described thereinafter.
Transglutaminase is widely used for the production of foods such as gelatins, cheeses, yogurts, tofu, boiled fish pastes, hams, sausages and noodles and also for improving the quality of meats (JP-Kokai No. 64-27471). Transglutaminase is also used for various industrial purposes, for example, in the production of materials for heat stable microcapsules and the production of carriers for immobilized enzymes. Transglutaminase catalyzes the acyl transfer reaction of γ-carboxamide group of glutamine residue present in the peptide chain of a protein molecule. When ε-amino group of lysine residue in a protein molecule acts as an acyl acceptor, ε-(γ-Glu)-Lys bond is formed in the protein molecule and between the molecules.
The inventors found that transglutaminase does not react with all glutamine residues, namely, the possibility of the reaction, i.e. reactivity of transglutaminase varies depending on the kinds and positions of the residues surrounding the glutamine residues. Therefore, modification of the reactivity of transglutaminase on the glutamine residues under specified environment lead to the modification of the gelation rate and, as a result, the range of the use of transglutaminase may be further widened.
In the present invention, for modifying the substrate specificity of MTG, the three-dimensional structure thereof is determined by X-ray crystal structure analysis and the manner of binding to the substrate is predicted and the mutations suitable for the purpose are designed and introduced. The term “substrate specificity” herein used indicates the reactivity with each glutamine residue in the acyl acceptor.
The determination of the three-dimensional structure of a protein by X-ray crystal structure analysis will be carried out by the following procedure:
(1) The protein is crystallized. The crystallization is indispensable for determining the three-dimensional structure. In addition, the crystallization is industrially useful for the purification of a protein in high purity and also for a method of stable storage of a protein in high density, which method is highly resistant against proteases.
(2) The crystals thus obtained are irradiated with X-ray to collect the diffraction data. In many cases, protein crystals are damaged by the X-ray irradiation which reduces the diffraction capacity thereof. In such cases, low temperature measurement technique has recently been spread wherein the crystals are rapidly cooled to about −173° C. and the diffraction data are collected in this state. Finally, the synchrotron radiation with high intensity is utilized for collecting the high-resolution data used in the determination of the structure.
(3) For the analysis of the crystal structure, the phase information is required in addition to the diffraction data. The MTG structure cannot be determined by molecular replacement method because the crystal structures of analogous proteins are yet unknown. Thus, the phase problem must be solved by heavy atom isomorphous replacement method. This method comprises introducing a metal atom having a large atomic number such as mercury or platinum into the crystal and obtaining the phase information taking advantage of the contribution of the large X-ray scattering capacity of the metal atom to the X-ray diffraction data. The phase thus determined can be improved by smoothing the electron density in the solvent region in the crystal. Since the electron density of water molecules in the solvent region is scarcely observed because of their have high fluctuation, the electron density very close to the true electron density can be obtained by approximating the electron density in this region to 0, which improves the phase. When plural molecules are contained in an asymmetric unit, the phase will be further remarkably improved by averaging the electron density of these molecules. A protein model is fit to the electron density map calculated using thus improved phase. This process is carried out using a program such as QUANTA (MSI Co., U.S.A.) by computer graphics. Then the structure is refined using a program such as X-PLOR (MSI Co., U.S.A.) to complete the structure analysis.
The inventors crystallized purified natural MTG from Streptoverticillium mobaraense and recombinant MTG (Met type) with additional Met originating from initiation codon at the N-terminal, but could not obtained a crystal suitable to be analyzed. However, because better crystals were obtained with natural MTG rather than with Met-type MTG, the inventors prepared Asp-type MTG (Example 1) after having considered that the recombinant MTG (Asp type MTG) starting with Asp1, like the natural one, obtained by removing Met at the N-terminal, might have the possibility of resulting in the crystals tolerable to the analysis. As a result, Asp-type MTG yielded a good crystal (Example 2), and the three-dimensional structure was successfully obtained (Example 3). The atomic coordinates in the MTG three-dimensional structure are shown in FIGS. 11 to 60 .
The reaction mechanism of transglutaminase is considered to be as follows: An acyl donor containing Gln reacts with a transglutaminase to form a reaction intermediate. Then an acyl acceptor containing Lys attacks the intermediate to form ε-(γ-Glu)-Lys bond between the acyl donor and the acyl acceptor. For using a protein as the acyl donor, it is not sufficient that it contains Gin, but it is required that the region around Gln can be bound appropriately to the substrate-binding site of MTG. On the contrary, it is enough in most cases for the protein to contain a primary amine to serve as the acyl acceptor. Therefore, modifying the substrate specificity of MTG means modifying the structure of the binding sites for the acyl donor. If MTG can be modified so as to reduce the substrate specificity thereof or, in other words, so as to accept a larger amount of the acyl donor, the cases increases wherein the joining of proteins may be possible using a smaller amount of MTG, and therefore, the reduction of the amount used is expected. Further, if the joining of proteins in which the cross-linking is presently impossible become possible, the application of MTG in a new field will become possible. The three-dimensional structure of MTG is indispensable for extrapolating the binding manner between MTG and the acyl donor.
MTG molecule elucidated in the present invention is in the shape of discoid form having a size of 65×59×41 Å (see FIGS. 1 and 2 ). Namely, in FIG. 1 , the crystal structure of MTG is shown by a ribbon model. α-helices and β-sheets are shown by spirals and arrows, respectively. An active residue, Cys 64 , is shown by boll and stick model (upper central part).
FIG. 2 shows the surface concavoconvex of the crystal structure of MTG in the same direction as in FIG. 1 . The regions having the electric charge are indicated in dark color and those free of the electric charge are indicated in light color. Observing the disk from upper side, it has a wedge-shaped cleft. In the inner part of the cleft, active residue Cys 64 is located. The fact that Asp 255 is positioned in the neighborhood of the active residue Cys 64 suggests that SH group of Cys 64 , which may be dehydrogenated by Asp 255 , causes the reaction. The side chain of Cys 64 faces the solution side and it can approach the acyl donor and acyl acceptor. In Ca 2+ -dependent human blood coagulation factor XIII, the active residue Cys is not exposed to the solvent. Therefore, for binding with the acyl donor, human blood coagulation factor XIII must undergo a significant structure change for the active residue to bind to the acyl donor. On the other hand, MTG is considered to be able to bind to the acyl donor without changing its structure. Therefore, it is supposed that MTG can use more Gln residues as the substrates than human blood coagulation factor XIII uses. Further, MTG has a merit in that a mutant having further reduced substrate specificity can be easily designed, because it is not required to take the structural change associated with the binding to the acyl donor into consideration. Furthermore, the molecular weight of MTG is as low as 38,000, while that of human blood coagulation factor XIII is 166,000, which make the determination of the binding site of MTG for the acyl donor easy. By observing the structure of MTG, it would be understood that the substrate is very likely to bind in the wedge-shaped cleft leading from the solvent to Cys 64 .
Many hydrophobic amino acid residues and aromatic amino acid residues are positioned in the cleft leading to active residue Cys 64 , and thus the mutual action thereof with hydrophobic residues or aromatic residues in the acyl donor is considered to be one of the keys for the determination of the substrate specificity. The polar resides including Asp are also scattered on the cleft surface, which suggests that they play an important role in the recognition of the acyl donor. A mutant transglutaminase having a modified substrate specificity can be obtained by the replacement, insertion or deletion of amino acid residues located in the cleft which is supposed to be the acyl donor binding region, preferably by the replacement of the polar residues (Asp, Glu, Lys and Arg) or aromatic residues (Phe, Tyr and Trp) with other amino acids, the replacement thereof with polar residues or aromatic residues, the deletion of the polar residues or aromatic residues or insertion of polar residues or aromatic residues. It is to be noted, however, since it is supposed that the activity will be lost by modifying Cys 64 and Asp 255 which play an important role for the activity, these residues should be excluded in the production of the mutants.
As shown in Examples given below, acidic amino acids present adjacent to the reactive site in the substrate proteins. Glutamine residues reactive with MTG are positioned on the surface of the substrate protein in many cases and it is considered that acidic amino acids which may easily exposed to the solvent are likely to present around them. When acidic residues are present at the interaction site of the substrate protein of MTG, the negative charge in MTG and that in the substrate protein repel each other to reduce the enzymatic reaction velocity. The negative charge can be removed by modifying the acidic amino acid residues exposed on the surface of the cleft of MTG containing Cys 64 which is supposed to be bound by the acyl donor or, alternatively, the negative charge can be relaxed by converting amino acid residues sequentially or spatially adjacent to the acidic amino acids into basic amino acids to obtain the transglutaminase having a modified substrate specificity. Such acidic amino acids include the residues Nos. 1, 3, 4, 28, 58, 249, 300 and 304 of MTG described in SEQ ID NO: 2.
The amino acids adjacent to the acidic amino acids are the amino acid residues within 5 Å from carboxyl groups of the acidic amino acids such as aspartic acid and glutamic acid, and are preferably amino acid residues having the degree of exposition to the solvent of at least 10%. They include residues Nos. 2, 5, 59, 248-250, 252, 278, 283-285, 289, 291, 296-297 and 302-303 of MTG described in SEQ ID NO: 2.
The amino acid sequence of MTG is shown in SEQ ID NO: 2, and the nucleotide sequence of the gene encoding MTG is shown in SEQ ID NO. 1.
As shown in Example 5, it was confirmed that the reaction rate on a particular glutamine residue in a protein molecule could be elevated by the deletion of Asp 1 or Asp 3 or by the replacement of Ser2. Because the distance from active residue Cys 64 to Asp 1 at the N-terminal is 20 Å, it is suggested that there is a possibility that the substrate specificity can be modified by modifying the amino acid residues located within 20 Å from Cys 64 . In addition, because it was confirmed that at least the loop regions (233-253 and 276-288) defined the substrate specificity, it was considered that the substrate specificity could be modified by modifying these regions. The possibility of changing the substrate specificity is further increased by modifying the residues exposed on the surface of the cleft having Cys 64 supposed to be bound by the acyl donor. To know whether the amino acid residues are exposed to the solvent or not, the solvent accessibility may be calculated using a program such as QUANTA. The solvent accessibility is the value resulting from dividing the exposure area to the solvent of a residue in the protein by the exposure area to the solvent of the amino acid in its free form, which is indicated as a percentage. The possibility of changing the substrate specificity of MTG may be high when the residues existing within 20 Å from Cys 64 and having a solvent accessibility of at least 10% are modified. Such residues correspond to residue Nos. 1-5, 26, 28, 58-59, 62, 69, 74-75, 77, 79, 235-236, 238-244, 248-250, 252-254, 277-278, 282-287, 289, 291, 296-297 and 300-304 of MTG described in SEQ ID NO. 2.
In designing and producing a mutant transglutaminase, the codon in the gene encoding the transglutaminase corresponding to the amino acid residues at the above-described substrate-binding site may be modified by site-directed mutagenesis and the obtained gene may be introduced into a suitable vector and transferred into a host, and then the transformant may be cultured. The intended mutant transglutaminase may be recovered from the transformed cultured microbial cells of and evaluated.
The site-directed mutagenesis methods for introducing intended mutation at the intended sites of DNA include, for example, a PCR method shown in Examples [Higuchi, R., 61, in PCR technology; Stockton Press edited by Erlich, H. A. (1989); Carter P., Meth. In Enzymol., Vol. 154, p. 3382 (1987)]; and a method wherein a phage is used [Kramer, W. and Frits, H. J., Methods in Enzymology, Vol. 54, p. 350 (1987), Kunkel, T. A. et al., Methods in Enzymology, Vol. 154, p. 367 (1987)]. The vectors for introducing the genes encoding mutant transglutaminase are not particularly limited so far as they can be replicated in hosts. When Escherichia coli is used as a host, plasmids capable of autonomously replicating in this bacterium can be used. For example, pUC19, pET or pGEMEX can be used.
Preferred hosts are, for example, strains of Escherichia coli , however, any bacterium can be used as the host so far as the replication origin of the constructed recombinant DNA and the mutant transglutaminase genes can function, the mutant DNA can be replicated and the expression of the mutant transglutaminase genes can be expressed in the bacterium. The representative preferred host is Escherichia coli BL21(DE3)pLysS when T7 promoter is used or Escherichia coli JM109 in other cases.
An ordinary method can be employed, without any particular limitation, for introducing the recombinant DNA obtained by ligating the vector with the DNA fragment containing the gene encoding the mutant transglutaminase into the host. When Escherichia coli is used as the host, there can be employed calcium chloride method [J. Mol. Biol., Vol. 53, p. 159 (1970)], Hanahan method [J. Mol. Biol., Vol. 166, p. 557 (1983)], SEM method [Gene, Vol. 96, p. 23 (1990)] and Method of Chung et al. [Proceedings of the National Academy of Sciences of the USA, Vol. 86, p. 2172 (1989)].
MTG may be expressed in the bacterial cell at a high level by culturing the transformant harboring the recombinant DNA containing the gene encoding the mutant transglutaminase in a suitable medium containing a carbon source, a nitrogen source, an inorganic ion and, if necessary, further containing an organic nutrient source. For extracting the mutant transglutaminase from the cultured cells, the cells are collected after culturing them, suspended in a buffer solution, and subjected to the lysozyme treatment, freezing-and-thawing treatment, ultrasonic disruption, etc., and then, they are disrupted and centrifuged to separate the suspension into the supernatant and the precipitant.
Since the mutant transglutaminase is produced in the form of a protein inclusion body and separated as the precipitant, it may be solubilized with a denaturing agent or the like and then the denaturing agent may be removed to isolate and purify the protein. The denaturing agents for solubilizing the produced protein inclusion body may be, for example, urea (for example, 8M) and guanidine hydrochloride (for example, 6 M). By reducing the concentration of the denaturing agent by dilution or the like, the protein having transglutaminase activity is regenerated. The solutions used for the dilution may include phosphate buffers and Tris buffers.
After regenerating the activity, the active protein can be isolated and purified by a suitable combination of known isolation method and purification method. Examples of these methods are salting out, dialysis, ultrafiltration, gel filtration, ion exchange chromatography, affinity chromatography and reversed-phase high-performance chromatography.
The term “substrate specificity” in the context of the present invention indicates the reactivity of each glutamine residue in the substrate protein molecule to a transglutaminase. The change in the substrate specificity of a transglutaminase means the change in the reactivity of at least one glutamine residue. When the mutant transglutaminase having the changed substrate specificity is reacted on a protein, the degree of crosslinking of the protein, namely, the degree of the gelation, is also changed compared with the results obtained by reacting the original transglutaminase.
Methods for comparing the substrate specificities of transglutaminases include, for example, the method wherein the gelation time of casein is determined, the method wherein the cross-linked polymer obtained after the transglutaminase reaction is analyzed by electrophoresis, and the method wherein the reaction product is fragmented and then analyzed. It is also possible that the substrate protein is reacted with 15 N-labeled ammonium ion by using the transglutaminase and then the labeled glutamine residues are detected by NMR (JP-Application No. 2000-141152).
The analysis of substrate specificity by the latter method is the method of comparing the reactivity of individual glutamine residue in the protein, taking advantage of the fact that the carboxamide nitrogen in the glutamine residue, which functions as a substrate, is 15 N-labeled by a transglutaminase. The substrate specificities of the transglutaminases can be compared with each other, taking advantage of that glutamine residue which acts as a substrate for transglutaminase is specifically 15 N-labeled and that the labeling rate varies depending on the reactivity of the transglutaminase to the glutamine residue.
The unit activity of transglutaminase is determined and defined as follows: The reaction is carried out using benzyloxycarbonyl-L-glutaminylglycine and hydroxylamine as the substrates, the resulting hydroxamic acid is converted into an iron complex hereof in the presence of trichloroacetic acid and then the amount thereof is determined as an absorbance at 525 nm. Thus the calibration curve is prepared based on the amount of hydroxamic acid. The amount of the enzyme which produces 1 μmole of hydroxamate per one minute is defined to be the unit of transglutaminase activity, one unit. The details of this determination method were previously reported (for example, JP-Kokai No. 64-27471).
Besides the transglutaminase from Streptoverticillium mobaraence (MTG), it is also possible to modify an enzyme which has the transglutaminase activity and amino acid sequence homologous to MTG, or an enzyme which has the transglutaminase activity and which is supposed to have the three-dimensional structure similar to that of MTG, on the basis of the three-dimensional structure of MTG. The amino acid replacement effective for the modification of the substrate specificity or the like of MTG is also effective for analogous enzymes such as those from Streptoverticillium cinnamoneum and Streptoverticillium lydicus (JP-Kokai No. 10-504721). The amino acid sequence of the transglutaminase from Streptoverticillium cinnamoneum is shown in SEQ ID NO: 6, the nucleotide sequence of the gene encoding the amino acid sequence is shown in SEQ ID NO: 5, the amino acid sequence of the transglutaminase from Streptomyces lydicus is shown in SEQ ID NO: 8, and the nucleotide sequence of the gene encoding the amino acid sequence is shown in SEQ ID NO: 7. In this connection, the homology of the amino acid sequence between MTG and transglutaminase from Streptoverticillium cinnamoneum is 78% ( FIG. 3 ), and that of the amino acid sequence between MTG and transglutaminase from Streptomyces lydicus is 79% ( FIG. 4 ). Namely, FIG. 3 shows the alignment of amino acid sequences of MTG and transglutaminase from Streptoverticillium cinnamoneum . Conserved amino acid residues are shown by “*”. FIG. 4 shows the alignment of amino acid sequences of MTG and the transglutaminase from Streptoverticillium lydicus . Conserved amino acid residues are shown by “*”.
The coordinate assignment of amino acid residues of two different proteins may be revealed from the alignment of the amino acid sequences (Sequence Alignment) when the homology between the amino acid sequences of them is above about 20%, and it may be revealed from the three-dimensional structures and the alignment of the amino acid sequences (Threading) when the homology is below about 20%. The former can be carried out using a program such as BLAST and the latter can be carried out using a program such as INSIGHT II. BLAST can be obtained by using ftp as the file adapted to the computer to be used among the files existing in “/blast/executable” of ncbi.nlm.nih.gov. The operating instructions are described in detail in http://genome.nhgri.nih.gov/blastall/blast install. INSIGHT II is commercially available (MSI Co.).
A mutant transglutaminase having modified substrate specificity is usable for the preparation of gelled foods such as jellies, as well as yogurts, cheeses and gelled cosmetics; for the improvement in the quality of meats; for the production of materials of thermally stable microcapsules and carriers for immobilized enzymes; and for the enzymatic labeling of proteins. When a protein is 15 N-labeled, the structure analysis by NMR is made possible (JP-Application No. 2000-141151), and when it is labeled with a primary amine, the solubility and stability thereof can be improved (JP-Application No. 2000-141152). Proteins usable as the substrates for the transglutaminase may vary. Thy include those constituting muscle such as actin and myosin; human plasma components such as albumin, immunoglobulin and blood coagulation factors; enzymes such as protease and transferase; hormones such as growth hormone and erythropoietin; cell growth factors for proliferating or inhibiting cells; immune reaction modulators for cell differentiation, induction and stimulation; and biologically active proteins produced by cells, such as monokine, cytokine and lymphokine. The origins of those proteins are not particularly limited and the proteins may be derived from animals, vegetables and microorganisms. Further, the proteins may be those expressed in Escherichia coli , yeast and animal cells by introducing the genes of these proteins into them or those expressed by using a cell-free protein synthesis system.
The reaction of such a protein in the presence of the transglutaminase is carried out under conditions where the transglutaminase can function. For example, the substrate protein and the transglutaminase are kept at a pH in the range of about 5.0 to 9.0, preferably abut 6.0 to 8.0 and at a temperature in the range of about 4 to 55° C., preferably about 25 to 40° C. in an aqueous solvent. In this reaction, the concentration of the substrate protein is preferably in the range of 1 μM to 1 M, and the amount of the transglutaminase is desirably in the range of about 10 nM to 100 μM which corresponds to about 0.01 to 20 units per mmol of the protein. However, the reaction conditions are not limited to those described above.
The range of the industrial utilization of MTG can be further widened when the heat stability and optimum pH can be modified in addition to the modification of the substrate specificity. If the heat stability is improved, the gelation of protein at a higher temperature may become possible. In the production of tofu (bean curd) filled into a pack, which is commonly sold in supermarkets, MTG is added to soybean milk and then the obtained mixture is tightly closed in the pack and then heated. Therefore, the temperature easily elevates to a point at which MTG is denatured. If the heat stability of MTG can be improved, the amount of protein inactivated by denature can be reduced, whereby the cost is lowered. The thermal stability can be improved by, for example, the introduction of proline residue, the replacement of the residues forming the counterclockwise helix structure with glycine residues [Protein Engng, Vol. 6, pp. 85-91 (1993)], or filling of the cavities in the protein [Biochemistry, Vol. 32, pp. 6171 to 6178 (1993)]. Further, if the optimum pH can be shifted to the alkaline region, the gelation efficiency of a substance which reacts under an alkaline reaction condition can be elevated such as the case of udon which are prepared by joining gluten. The modification of optimum pH can be achieved by modifying pKa of the active residues [Protein Engineering, Vol. 11, pp. 383-388 (1998)].
EXAMPLES
The following Examples will further illustrate the present invention, which by no means limit the invention.
Example 1
Preparation of Recombinant Native Type (Asp Type) MTG
MTG expression plasmid pETMTGXa-01 which will be described below was constructed to produce recombinant MTG having the same N-terminal amino acid sequence as that of native type in Escherichia coli . pETMTGXa-01 is a plasmid capable of expressing a fused protein having the recognition sequence (IEGR) of factor Xa between the amino terminal region of T7 gene 10 protein and MTG. The procedures for constructing pETMTGXa-01 are described hereinafter. Firstly, primer pGEXMTGF01 (SEQ ID NO: 9 in the sequence listing) containing the sequence encoding BamHI cleavage site and Factor Xa recognition site at the upstream side of MTG gene and also primer pGEXMTGR01 containing Sac I cleavage site (SEQ ID NO: 10 in the sequence listing) were synthesized. PCR was carried out using these primers and MTG genes in MTG expression plasmid pUCTRPMTG-02(+) (JP Kokai No. 11-75876) as a template, and the amplified fragment was cloned in pGEM-T Easy Vector (Promega). PCR were carried out by 25 cycles of “96° C. for 30 seconds, 50° C. for 15 seconds and 60° C. for 1 minute”. The cloning of the amplified fragment was carried out according to the instructions of supplier. Clone in which the amplified fragments were inserted in an opposite direction to lacZ gene were selected. The nucleotide sequences of the plasmids in the clones were analyzed and the plasmid having the intended correct sequence was named “pGEMMTGXa”. Then pET5a (Takara Shuzo Co., Ltd.) vector was cleaved by EcoRI, bunt-ended, and cleaved by BamHI. pGEMMTGXa was cleaved by SalI, blunt-ended, cleaved by BamHI, and the obtained fragment containing MTG gene was subcloned in the above-described enzymatically digested pET5a to construct pETMTGXa-01.
Escherichia coli BL21(DE3)pLysS (Promega) having pETMTGXa-01 introduced therein was inoculated in 50 ml of L medium (10 g/L of Tryptone, 5 g/L of yeast extract, 5 g/L of NaCl and 1 g/L of glucose; pH: 7.2) containing 200 μg/ml of ampicillin and pre-cultured at 37° C. for 6 hours. Then, 8 ml of the pre-culture was subcultured for 4 hours at 37° C. in M9 casamino acid medium (8 g/L of casamino acids, 5 g/L of ammonium chloride, 0.2 g/L of yeast extract, 2 mg/L of vitamin B 1 hydrochloride, 14.5 mg/L of calcium chloride dihydrate, 15.1 g/L of disodium hydrogen phosphate 12 hydrate, 3 g/L of potassium dihydrogen phosphate, 0.5 g/L of magnesium sulfate heptahydrate and 5 g/L of glucose; wherein magnesium sulfate heptahydrate and glucose were separately sterilized) containing 200 μg/ml of ampicillin. When OD 660 reached to about 2.0, 0.8 ml of 1M IPTG was added to the culture, and the culture was further continued at 37° C. for 14 hours.
The cells were collected from the broth by centrifugation and suspended in 200 ml of a mixture of 20 mM Tris-HCl, 30 mM of NaCl and 5 mM of EDTA, pH 7.5. 1 mg/ml of lysozyme was added to the obtained suspension. After leaving the resultant mixture at 4° C. for 1 hour, the cells were crushed by ultrasonication. Protein inclusion bodies containing denatured MTG were collected by centrifugation. The protein inclusion bodies were thoroughly suspended in a small amount of water. Reagents and water were added to the obtained suspension so that the suspension contains 8 M of urea, 20 mM of sodium phosphate, 1 mM of EDTA and 20 mM of DTT/30 ml to dissolve the protein inclusion bodies. After incubating at 37° C. for 2 hours, pH was lowered to 4 by adding dropwise 1 M of hydrochloric acid, and the precipitant was removed after centrifugation. The concentration of the solubilized MTG was determined, which was found to be about 20 mg/ml. 20 ml of the above solubilized MTG solution were slowly added dropwise to 1000 ml of 20 mM sodium acetate, pH 4.0, containing 2 mM DTT, and the mixture was left to stand for 2 hours. After 2 hours, 4 M of sodium hydroxide was added dropwise to the mixture to elevate pH to 6 to complete the refolding. The concentration of MTG was determined to be about 0.14 mg/ml. The refolding procedure was totally carried out in a cold room at 4° C.
The total solution thus obtained was concentrated to a volume of 200 ml by ultrafiltration (Hydrosart, 10 kDa Molecular cut off; Sartocon Slice Casette, Sartorius). and the solvent was replaced with 20 mM Tris-HCl (pH 7.5) by using Sephadex G25 (M) (Pharmacia Aktiebolag). 5 mg of bovine factor Xa (Hematologic Technology Inc.) was added to 250 ml of the obtained solution and then the mixture was left to stand overnight at 5° C. The concentration of MTG was determined to be about 0.36 mg/ml. The N-terminal amino acid sequence of MTG cleaved by factor Xa was analyzed with a protein sequencer (Shimadzu Corporation) to confirm that the sequence attached to the N-terminal had been completely cleaved by factor Xa and also that there was only the recombinant MTG having the mature type N-terminal amino acid sequence.
After the cleavage with factor Xa, pH of the solution was adjusted to 5.8 with 1 M hydrochloric acid and then the solution was diluted 5-fold with a buffer solution (20 mM sodium acetate, pH 5.8) used for equilibration of cation exchange chromatography. The obtained solution was applied to the cation column (CM Sepharose FF, 2.6 φ×10 cm; Amersham Pharmacia Biotech) sufficiently equilibrated with the buffer solution. After the re-equilibration with the same buffer solution, a protein fraction (46 ml) eluted according to the linear concentration gradient of 0-0.4 M NaCl was recovered using UV absorption at a wavelength of 280 nm as an index. The recovered fraction was diluted 10-fold with a buffer solution (20 mM sodium acetate, pH 5.5) used for the equilibration of the cation exchange chromatography. ⅓ of the diluted solution was applied to the cation column (6 ml of Resource S; Amersham Pharmacia Biotech) sufficiently equilibrated with the same buffer solution. After the re-equilibration with the same buffer solution, a protein fraction eluted according to the linear concentration gradient of 0-0.5 M NaCl was fractionated using UV absorption at a wavelength of 280 nm as an index. A fraction (2 ml) in which no impurities were detected by reversed-phase HPLC and which was free from low pl form of MTG was recovered from the front side of the peak top. The similar purification was carried out twice, and the side fraction of the recovered fraction was re-chromatographed to finally recover the fraction of 8 ml (40 mg). Again, the solvent was replaced with 20 mM phosphoric acid, pH 6.0 by using Sephadex G25(M). The recovered fraction was analyzed by isoelectric focusing to confirm that it was the recombinant MTG of a high purity substantially free from analogues having different pl. All of the chromatography procedures were carried out at room temperature.
Example 2
Crystallization
The crystallization was carried out using the hanging drop mode of vapor diffusion method. Recombinant Asp type MTG solution (concentration: 15 mg/ml) and 83 mM buffer (pH 5.0) of sodium cacodylate containing 25% polyethylene glycol 1000 and 25 mM calcium chloride were dropped in the same amounts (2 μl) on a siliconized cover glass and mixed together. The cover glass was placed on the buffer in such a manner that the drops of the mixed liquid were hanging over the well filled with 500 μl of the buffer, and they were left to stand at 20° C. Crystals were formed one week after. 10 days after, the crystals grew into plate-like crystals of a measurable size (about 0.5×0.3×0.1 mm). The MTG crystal was analyzed at a low temperature, because MTG crystal would be damaged by X-ray which leads to the gradual reduction of resolution when analyzed at ambient temperature. The crystal was transferred into 117 mM buffer (pH 5.0) of sodium cacodylate containing 35% of polyethylene glycol 1000 and 35 mM of calcium chloride and rapidly cooled by blowing nitrogen gas at −173° C.
X-ray diffraction data of 2.7 Å resolution were obtained with a X ray diffractometer, R-AXIS IIc (Rigaku Co., Ltd.) and the crystallographic parameters were determined. The space group was P2 1 , lattice constants were as follows: a=78.4 Å, b=117.1 Å, c=85.7 Å and β=112.9°. Assuming that the asymmetric unit contains 4 molecules (molecular weight: 38,000), water content of the crystals was calculated at 50%.
Example 3
Determination of Three-Dimensional Structure
The crystals were immersed in a solution of heavy metal salts and the heavy atom derivatives were screened. The diffraction data of the heavy atom derivative crystals were obtained by using SMART6000 (Bruker) and the synchrotron radiation facilities BL-6B of the National Laboratory for High Energy Physics. From difference Patterson map compared with the native data, it was shown that EMTS, K 2 OSCl 6 and K 2 IrCl 6 provided good heavy atom isomorphous-replacement crystals. Major heavy atom sites were determined from the differential Patterson maps of the respective heavy atom isomorphous replacement crystals, and then, other minor heavy atom sites were determined using the difference Fourier maps. The coordinates of these heavy atom sites were refined by the phase determination program “MLPHARE” and the phase was calculated. The phase was improved by solvent flattering and histogram matching using the program DM and then an electron density map was prepared. In the electron density map, 4 MTG molecules were refined in an asymmetric unit. The electron density corresponding to the 4 molecules was averaged using the program DM to remarkably improve the quality of the electron density. The electron density map prepared with 2.7 Å resolution was very good and all the amino acid residues from N terminal to C terminal could be fitted into the electron density.
The first model was constructed on computer graphics using QUANTA program, and the structure was refined using X-PLOR program. The final models ( FIGS. 1 and 2 ) refined with X-ray diffraction data of a 2.4 Å resolution at the synchrotron radiation facilities BL-6B (National Laboratory for High Energy Physics, Tsukuba,) contained all of the 331 amino acid residues and 400 water molecules. The crystallographic reliability factor (R factor) determined by using the reflection at 10 to 2.4 Å resolution was 19.6%. In a Ramachandran plot prepared with the program PROCHECK, it was shown that 82% of residues other than glycine located in the most desirable region and 15% thereof located in the second most desirable region.
Example 4
Preparation of Site-Directed Mutant
Because N-terminal Asp 1 is positioned at the entrance of the cleft, it is supposed that it has some responsibility on the recognition of the substrate ( FIGS. 1 and 2 ). The side chain of Asp 1 has no interaction with other residues constituting the enzyme and, therefore, it is considered that even when the residue is modified, there is no risk of distorting the three-dimensional structure of the protein. The effects of Asp 1 on the substrate specificity can be examined by preparing a mutant in which this residue has been removed (hereinafter referred to as “Ser type mutant” because it starts with Ser 2 ) and then comparing the substrate specificity thereof with that of the wild type. Because Escherichia coli obtained by transforming plasmid containing a gene encoding Ser-type mutant had been already constructed (JP-Kokai No. 11-075876), we designed to prepare the Ser-type mutant using this plasmid.
It was also supposed that the substrate specificity could be modified also by replacing Ser 2 located at the entrance of the channel with another amino acid. we designed to prepare three mutants, i.e. S2Y obtained by replacing Ser 2 with Tyr for the purpose of enhancing the hydrophobic mutual action and S2R and S2D obtained by replacing Ser 2 with Arg and Asp, respectively, for the purpose of enhancing the electrostatic interaction. These mutations were introduced into MTG having additional Met corresponding to the initiation codon at the N-terminal (Met type).
Additionally, insulin A chain (Sigma) wherein cysteine residue had been oxidized was bought as the substrate protein, and the glutamine residues which could be the substrates of MTG were examined. Insulin A chain is a protein (SEQ ID NO: 23) consisting of 21 amino acid residues, which has glutamine residues at the 5 th position and the 15 th position from the N-terminal. The reasons why insulin A chain was selected for examining the structural characteristics of the substrate protein are that it contains plural glutamine residues and the positions of the glutamine residues reacted with MTG can be easily determined because it has a low molecular weight.
200 μl of 1 M 15 NH 4 Cl and 2 μl of 1 mM MTG were added to 1 mg of insulin A chain wherein the cysteine residues were oxidized and then they were stirred and subjected to mass spectrometry. The mass spectrometry was carried out during fragmenting individual amino acid residues to identify the 15 N-labeled residue which was proved to be the 15 th glutamine residue from the N-terminal. As a result, it was found that in the two glutamine residues in insulin A chain, the 15 th glutamine residue was the substrate of MTG.
Glutamic acid is present at position 17 in insulin A chain. This glutamic acid residue is located adjacent to the 15 th glutamine residue with the insertion of one residue. Since amino acid side chains on a polypeptide chain incline to face the same direction when they are located with the insertion of one residue, the glutamine residue at position 15 and the glutamic acid at position 17 are therefore considered to be not only close to each other in the sequence but also to be spatially close to each other. Because acidic amino acid is present near the 15 th glutamine residue, it is not preferred that an amino acid having a negative electric charge is present at the substrate binding site of MTG. In most cases, the glutamine residues reactive with MTG are positioned on the surface of the substrate protein. It is considered that acidic amino acids which may easily exposed to the solvent may present surrounding the glutamine residue. This is also applicable to the substrate proteins other than insulin.
Accordingly, it is supposed that the substrate specificity can be improved when the negative charge is reduced by modifying the acidic amino acids or by converting the surrounding amino acids into basic amino acids to obtain a mutant transglutaminase having a modified substrate specificity. Ser-type MTG was the one in which the aspartic acid in SEQ ID No: 2 had been removed, and S2R mutant was the one in which basic amino acids were arranged near the acidic amino acids to reduce the negative charge, all of which meet the above criterion. They satisfy the above-described conditions. Additionally, del 1-2 was prepared by removing residue Nos. 1 and 2 and del 1-3 was prepared by removing residue Nos. 1 to 3 to remove the negative charge of Asp 1 and Asp 3 , respectively.
Because the two loops (233-253 and 276-288) which locate to surround the active residue Cys 64 are positioned as if the substrate is sandwiched by them, they are likely to define the substrate specificity. However, the loops cannot be blindly removed because they also play an important role in the folding of MTG. Therefore, the sites satisfying the following conditions (1) and (2) were searched based on the three-dimensional structure of MTG: (1) the site hardly interact with the sites other than the loops and it does not affect the folding and (2) the distance between N-terminal and C-terminal of the deleted site (residue) is not longer than 7 Å and it can be replaced with 3 glycine residues. The 3 glycine residues are mentioned only considering that the turn structure can be formed with them, and they are not restrictively mentioned. For example, because residues Nos. 241 to 252 have substantially no interaction with other sites and since the distance between N-terminal of residue No. 241 and C-terminal of residue No. 252 is about 6 Å, residue Nos. 241-252 can be replaced by 3 glycine residues. It was attempted to produce Sg4 by replacing residue Nos. 241 to 252 with 3 glycine residues and Sg7 by replacing residue Nos. 278 to 287 with 3 glycine residues as the mutants satisfying such conditions in which the loops were removed to elucidate the contribution of the two loops surrounding Cys 64 active residue.
(1) Preparation of Ser-type MTG
Escherichia coli JM109 harboring plasmid pUCTRPMTG-02(+) containing Ser-type MTG gene (JP-Kokai No. 11-075876) having Ser-type MTG gene was inoculated into 50 ml of L medium containing 200 μl/ml of ampicillin, and pre-cultured at 30° C. for 7 hours. 16 ml of the pre-culture was transferred into 800 ml of M9 medium, and the culture was further continued at 37° C. for 24 hours. The cells were collected from the broth by centrifugation and suspended in 12.5 ml of 20 mM Tris and 30 mM of NaCl. 1 mg/ml of lysozyme and 0.5 MEDTA were added to the obtained suspension, and they were left to stand at 4° C. for 1 hour. Then the cells were crushed by ultrasonication for 20 minutes. The protein inclusion bodies containing denatured MTG were collected by centrifugation. The protein inclusion bodies were thoroughly suspended in a small amount of water, and then, the reagents and water were added to the obtained suspension to a final concentration of 8 M urea·20 mM sodium phosphate·1 mM MEDTA·20 mM DTT/5 ml to dissolve the protein inclusion bodies.
After incubating at 37° C. for 2 hours, pH was lowered to 4 and the mixture was centrifuged to remove the precipitant. The solution of solubilized MTG was slowly dropped into 250 ml of 20 mM sodium acetate (pH 4.0) 2 mM DTT, and the mixture was left to stand for 2 hours. After 2 hours, pH was elevated to 5, and the insoluble matter was removed by centrifugation. pH was elevated to 6 to complete the refolding. The whole refolding procedure was carried out in a cold room at 4° C.
After replacing 250 ml of the solution with 20 mM sodium acetate (pH 5.8) with Sephadex G25(M) (Pharmacia), a protein fraction (55 ml) was recovered, which was eluted from an ion exchange resin (CM Sepharose FF) (Pharmacia) by the linear concentration gradient of 0 to 0.4 M of sodium chloride in 20 mM sodium acetate (pH 5.8). Again the solvent was replaced with 20 mM sodium phosphate (pH 6.0) with G25(M).
When the subsequent amino acid of N-terminal Met is Ser, since N-terminal Met will be removed by methionine aminopeptidase of Escherichia coli , the codon corresponding to Ser may be positioned after the initiation codon to prepare the Ser-type mutant having Ser at the N-terminal.
(2) Preparation of Mutant by Replacing Ser 2
For producing three mutant MTGs (S2Y, S2R and S2D) by Escherichia coli , MTG expression plasmid pGEMMTG3 (encoding the amino acid sequence in SEQ ID No: 4 and containing the base sequence of SEQ ID No: 3) was constructed and used as a template for the site-specific mutagenesis using PCR. pGEMMTG3 is a plasmid highly expressing MTG by T7 promoter inducibly, and which was constructed by removing T7 gene 10 protein gene from plasmid pGEM15TG(Xa) expressing the fusion protein of amino-terminal side region of T7 gene 10 protein and MTG [Biosci. Biotech. Biochem. Vol. 61 (5), pp. 830-835 (1997)] and introducing a synthetic gene designed so that a high expression is possible in Escherichia coli as described in JP-Kokai No. 11-75876.
The procedures for constructing pGEMMTG3 are described below. Firstly, primer pGEMTGF01 (SEQ ID No: 11) which was designed so that 42 residues from T7 gene 10 protein were deleted and which had PstI cleavage site at its 5′ terminal and primer pGEMTGR01 (SEQ ID No: 12) having EcoRI cleavage site in MTG gene were synthesized. PCR was carried out by using these primers and pGEM15TG(Xa) as a template, and the amplified fragments were cloned into pUC19 (Takara Shuzo Co., Ltd.). 25 cycles of PCR were carried out with the conditions of 96° C. for 30 sec, 50° C. for 15 sec and 60° C. for 1 minute. The nucleotide sequence of the insert-containing plasmid was analyzed and the plasmid having the intended correct sequence was named “pUCTGN”. Because pGEM15TG(Xa) had two NdeI cleavage sites, NdeI/EcoRI fragment of pUCTGN was replaced by NdeI/EcoRI fragment of pGEM15TG(Xa) through multiple steps as described below. Namely, the small fragment obtained by cleaving pGEM15TG(Xa) with SmaI and EcoRI was subcloned into pBluescript (Toyobo Co., Ltd.) vector which had been cleaved by SmaI and EcoRI to obtain “pBS15TG(SE)”. Then the fragment containing TG gene obtained by cleaving pUCTGN with NdeI and EcoRI was ligated to the segment obtained by cleaving pBS15TG(SE) with NdeI and EcoRI, which did not contain TG gene, to obtain “pBSTG(SE)”. Further, the large fragment obtained by cleaving pGEM15TG(Xa) with SmaI and EcoRI was ligated to the fragment containing TG gene obtained by cleaving pBSTG(SE) with NdeI and EcoRI to obtain “pGEMTG”. Finally, the MTG gene segment (PvuII-HindIII cleavage fragment) of pGEMTG was replaced with MTG gene (PvuII-HindIII cleavage fragment) of pUCTRPMTG-02(+) (described in JP-Kokai No. 11-75876) to construct pGEMMTG3.
The mutation was introduced using quick exchange (registered trade mark) site-directed mutagenesis kit (Stratagene) with a primer ( FIG. 5 , SEQ ID NOS: 13-18) corresponding to each mutant enzyme according to the manufacturer's protocol. Namely, FIG. 5 shows the primer sets used in the site-directed mutagenesis for S2Y·S2R·S2D mutant. Escherichia coli JM109 (Takara Shuzo Co., Ltd.) was transformed by using the PCR product. The transformed cells were plated on L agar medium plates containing 100 μl/ml of ampicillin and incubated at 37° C. for 16 hours. Colonies thus formed were taken and cultured in L-medium containing 100 μl/ml of ampicillin by shaking overnight. The cells were recovered from the broth by centrifugation, and the plasmid was extracted by using Automatic DNA Isolation System PI-50 (KURABO) according to the manufacturer's protocol. The nucleotide sequence encoding each mutant MTG was confirmed by DNA sequencing. The plasmid was transformed into Escherichia coli BL21 (Promega).
Escherichia coli BL21(DE3) into which the mutant MTG gene had been introduced was inoculated into 50 ml of L medium (10 g/L of Tryptone, 5 g/L of yeast extract, 15 g/L of NaCl and 1 g/L of glucose, pH 7.2) containing 200 μg/ml of ampicillin, and pre-cultured at 37° C. for 6 hours. 8 ml of the pre-culture was transferred into M9 Casamino acid medium (8 g/L of casamino acid, 5 g/L of ammonium chloride, 0.2 g/L of yeast extract, 2 mg/L of vitamin B 1 hydrochloride, 14.5 mg/L of calcium chloride dihydrate, 15.1 g/L of disodium hydrogen phosphate 12 hydrate, 3 g/L of potassium dihydrogen phosphate, 0.5 g/L of magnesium sulfate heptahydrate and 5 g/L of glucose, pH 7.0) containing 200 μg/ml of ampicillin, and further cultured at 37° C. for 4 hours. When OD 660 reached to about 0.6, 0.8 ml of 1 M IPTG was added to the culture, and the culture was further continued for 14 hours at 37° C. The treatment of the cells after the culture, the solubilization of the protein inclusion body, the protein refolding and the purification of the protein were carried out as described in (1).
(3) Preparation of Mutants in which the Acidic Amino Acid Residues are Modified
MTG expression plasmid pGEMMTG3 was used as the template for the site-specific mutagenesis and synthetic DNA (SEQ ID NOS: 19-22 in FIG. 6 ) corresponding to each mutant enzyme was used as the primer. The construction of the mutant plasmid, the culture, the treatment of the cells, the solubilization of the protein inclusion body, the protein refolding and the purification of protein were carried out as described in Example 4 (2).
(4) Preparation of the Mutants, Sg4 and Sg7 in which the Loops were Removed
For Sg4 genes, the genes corresponding to residue Nos. 1-240 and 253-331 was excised, and for Sg7 gene, the genes corresponding to residue Nos. 1-277 and 288-331 was excised, and then the two fragments were ligated together, respectively. Methods for constructing the Sg4 expression plasmid and the Sg7 expression plasmid are described below. PCR was carried out by using the primer (SEQ ID NO: 24) having SmaI cleavage site at the 5′-terminal and the primers (SEQ ID NOS: 25 and 26) containing a fragment encoding 3 glycine residues at the 3′-terminal, and pGEMMTG3 as the template. Similarly, PCR was carried out by using the primers (SEQ ID NOS: 27 and 28) containing a fragment encoding 3 glycine residues at the 5′-terminal and the primer (SEQ ID NO: 29) having HindIII cleavage site at the 3′-terminal, and pGEMMTG3 as the template. PCR was then carried out by using the primer (SEQ ID NO: 24) having SmaI cleavage site at the 5′-terminal and the primer (SEQ ID NO: 29) having HindIII cleavage site at the 3′-terminal together with the two amplified DNA fragments as the templates to obtain amplified Sg4 gene and Sg7 gene. For Sg7 gene, the fragment was cleaved by SmaI and HindIII, and cloned in pGEMMTG3 between SmaI site and HindIII site. On the other hand, for Sg7, the fragment was cleaved by EcoRI and HindIII, and cloned in pGEMMTG3 between EcoRI site and HindIII site. Further, the nucleotide sequences of the inserts of the obtained plasmids were confirmed to complete the construction of the Sg4 expression plasmid and the Sg7 expression plasmid.
The methods of the culture, cell treatment, solubilization of the protein inclusion body, protein refolding and protein purification were carried out as described in Example 4 (2).
Example 5
Comparison of Substrate Specificity
For comparing the substrate specificity, ovalbumin was employed as the substrate protein. Ovalbumin has 12 glutamine residues, and the reactivity to each of them was traced by NMR. For utilizing the protein in the fields of foods, medicines, etc., for example, the crosslinking rate of the protein molecules should be increased. Therefore, for some of the mutants, the time required for gelling casein in the respective cases was also compared with each other.
(1) Analysis of Substrate Specificity using NMR
(1-1) Comparison between Wild Type and Ser Type
Ovalbumin, 15 NH 4 Cl and transglutaminase were mixed together so that the final concentration is 100 mg/ml for ovalbumin, 200 mM for 15 NH 4 Cl and 1 μM for transglutaminase. The first 1 H- 15 N HSQC determination was carried out 20 minutes after, and then the 1 H- 15 N HSQC determination was repeated 13 times at an interval of 2 hours and 15 minutes. An example of 1 H- 15 N HSQC obtained after reacting the wild type for 3 hours is shown in FIG. 7 . For both of the wild type and Ser type, the signal peak intensity (longitudinal axis) shown as “a” in FIG. 7 was plotted against the reaction time (hour; horizontal axis) ( FIG. 8 ). As a result, it was found that the glutamine residue generating the signal indicated as “a” in FIG. 7 was more rapidly 15 N-labeled when it was reacted with the Ser-type rather than when reacted with the wild type.
(1-2) Comparison of Wild Type with S2Y·S2R·S2D Mutant
Ovalbumin, 15 NH 4 Cl and transglutaminase were mixed together so that the final concentration is 100 mg/ml for ovalbumin, 200 mM for 15 NH 4 Cl and 1 μM for transglutaminase. 20 minutes after, the first 1 H- 15 N HSQC determination was carried out and then the 1 H- 15 N HSQC determination was repeated 13 times at an interval of 2 hours and 15 minutes. The peak intensity (longitudinal axis) of the spectra obtained by reacting the wild type, S2Y or S2D on the glutamine residue giving the signal “a” in FIG. 7 , was plotted, respectively ( FIG. 9 ). As a result, it was found that as for the glutamine residue generating the signal “a” in FIG. 7 , the order of the 15 N-labeling rate was S2Y<S2D<wild type <S2R.
(1-3) Comparison with Mutants having Modified Acidic Amino Acid Residues
Ovalbumin, 15 NH 4 Cl and transglutaminase were mixed together so that the final concentration is 100 mg/ml for ovalbumin, 200 mM for 15 NH 4 Cl and 1 μM for transglutaminase. 20 minutes after, the first 1 H- 15 N HSQC determination was carried out and then the 1 H- 15 N HSQC determination was repeated 13 times at an interval of 2 hours and 15 minutes. The peak intensity (longitudinal axis) of the spectra obtained by reacting wild type, Ser type, S2R, del 1-2 or del 1-3 to the glutamine residue giving the signal “a” in FIG. 7 was plotted ( FIG. 8 ) respectively in the same manner as that of Example 5. As a result, it was found that the reactivity of each of del 1-2 and del 1-3 was improved as compared with that of the wild type. Furthermore, it can be understood that the reactivity of del 1-3 was even much higher than that of Ser type or S2R.
Thus, the cross-linking of proteins can be carried out by using smaller amount of transglutaminase than the amount used in the prior art by generating a mutant transglutaminase having modified substrate specificity by reducing the negative charge through modifying acidic amino acid residues or converting amino acid residues adjacent to those acidic amino acid residues into basic ones, which leads the remarkable reduction of the costs.
(14) Comparison with Mutants in which the Loop Region is Removed
Ovalbumin, 15 NH 4 Cl and transglutaminase were mixed together so that the final concentration is 100 mg/ml for ovalbumin, 200 mM for 15 NH 4 Cl and 1 μM for transglutaminase. 6 hours after, the 1 H- 15 N HSQC determination was carried out to find that only 3 groups and 2 groups of the signals were found when Sg4 and Sg7 were used, respectively, while 6 groups of the signals were found when the wild type was used ( FIG. 10 ). It was known from the transglutaminase-free experiments that among the signals observed in FIG. 10 , unnumbered ones are not from the glutamine residues labeled by the transglutaminase reaction. As for the spectrum of ovalbumin labeled using Sg7, the figure extracted from a lower level is shown because the reactivity of Sg7 was reduced to lower the labeling efficiency. It is to be noted that, for this reason, unnumbered signals were slightly emphasized.
Only glutamine residues corresponding to (1), (3) and (4) were labeled using Sg4, while those corresponding to (1) to (6) were labeled with the wild type. Therefore, it was shown that when Sg4 was used, the number of glutamine residues which could serve as the substrates decreased, namely its substrate specificity was higher, compared with the case when the wild type was used. It was also shown that with Sg7, glutamine residues only corresponding to (1) and (3) can be the substrate, although the reactivity was reduced. It was considered that the binding area for the substrate was reduced by removing the loop, and that the kinds of the substrates with which the mutant could be reacted were reduced. As the results, it was suggested that the substrate specificity of MTG could be modified by preparing mutants thereof based on the three-dimensional structure. By using the transglutaminase having modified substrate specificity, the application of transglutaminase to subjects different from previously described may be possible, and the new development in industries is expected.
(2) Comparison of the Time of Gelatinization of Casein
57.6 μg of wild, Ser type, S2Y, S2R or S2D mutant was respectively added to 2.5 g of 8% (w/w) casein solution, and the time required for the gelatination of casein in each case was compared with each other. The reaction temperature was 40° C. The time required for the gelation was 90 minutes for wile type, 85 minutes for Ser type, 100 minutes for S2Y mutant, 85 minutes for S2R mutant and 95 minutes for S2D mutant. Thus, it was shown that the gelatinization proceeded in a short period for Ser type or S2R mutant, and the gelatination slowly proceeded for S2Y and S2D.
According to the present invention, the improvement of MTG on the basis of the three-dimensional structure is facilitated. In particular, transglutaminase having an improved reactivity on the substrate can be provided by the present invention. Further, new products and novel techniques can be provided with the transglutaminase having the improved reactivity on the substrate.