Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.enzyme.chem.msu.ru/~tishkov/Publications/IAN_2010_59_01_01.pdf Äàòà èçìåíåíèÿ: Sun Mar 7 19:25:55 2010 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 23:12:11 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: ice cap

Russian Chemical Bulletin, International Edition, Vol. 59, No. 1, pp. 1--, January, 2010

1

The role of residues Arg169 and Arg220 in intersubunit interactions of yeast D amino acid oxidase
N. V. Cherskova,
a a,b

S. V. Khoronenkova,

a,c

and V. I. Tishkov

a,b

Department of Chemistry, M. V. Lomonosov Moscow State University, 1 Leninskie Gory, 119992 Moscow, Russian Federation. Fax: +7 (495) 939 2742. E mail: vitishkov@gmail.com bA. N. Bach Institute of Biochemistry, Russian Academy of Sciences, 33 2 Leninsky prosp., 117234 Moscow, Russian Federation. Fax: +7 (495) 954 2732 cInnovations and High Technologies MSU Ltd., office 96, 16 ul. Tsimlyanskaya, 109559 Moscow, Russian Federation. Fax: +7 (495) 939 3208

D Amino acid oxidase from the yeast Trigonopsis variabilis (EC 1.4.3.3, TvDAAO) exists as a dimer consisting of two identical subunits. The dimeric structure of the enzyme is stabilized by 12 (six pairs) hydrogen bonds, the residues Arg169 and Arg220 of each subunit being involved in eight hydrogen bonds. The Arg169Glu and Arg(169,220)Ala mutants of TvDAAO were pre pared. Both mutant enzymes were expressed in E. coli cells as insoluble but catalytically active inclusion bodies. The introduction of amino acid substitutions at the intersubunit interface resulted in a change in the substrate specificity profile and a strong decrease in thermal stability.

Key words: D amino acid oxidase, site directed mutagenesis, quaternary structure, sub strate specificity, thermal stability.

Amino acid oxidase (EC 1.4.3.3, DAAO) is the FAD containing enzyme that catalyzes the oxidative deamination of D amino acids to form keto acids, hy drogen peroxide, and an ammonium ion. In spite of the fact that the enzyme plays an essential role in the regula tion of a number of very important processes in animal organisms (nervous activity, hormone secretion, etc.)1,2 and is widely used in practice,3--5 the structure--function relationship of DAAOs from different sources is little known. By 2008, the structures of only three D amino acid oxidases were solved, viz., those from porcine kidney (pkDAAO),6,7 human kidney (hDAAO),8,9 and the yeast Rhodotorula gracilis (RgDAAO).10,11 The enzyme from the yeast Trigonopsis variabilis (TvDAAO) is the fourth best characterized DAAO. The latter enzyme was obtained12 in a highly purified state in 1976 and is most widely used in practice.3--5 One can assume that, in view of the practical importance of this enzyme, numerous attempts were made in the last 30 years to crystallize TvDAAO. However, no structural data on this enzyme are available in the Protein Data Bank (PDB) and there are no publications concern ing this problem, which suggests that all the attempts were unsuccessful. We have also carried out a large scale screen ing of the crystallization conditions for wild type TvDAAO but failed to obtain crystals. Only after obtaining the
D

Cys108Phe mutant of TvDAAO,13 we succeeded in grow ing crystals of this enzyme and then solved its structure first at 2.8 å resolution14 and then at 1.8 å resolution.2 Porcine kidney D amino acid oxidase in vivo is a mono mer, whereas the enzymes from other sources exist in na ture as dimers. In the crystal, pkDAAO also exists as a dimer, in which two subunits are linked in a head to head fashion.6,7 For human D amino acid oxidase, two crystal modifications were grown. In one modification, like in the dimer of pkDAAO, the subunits are arranged in a head to head fashion,9 whereas in another modification the subunits are arranged in a cross like fashion.8 The dimers of TvDAAO C108F and RgDAAO are arranged in a head to tail fashion. However, unlike the latter enzyme, the subunits of the TvDAAO C108F molecule do not lie in one plane and are shifted with respect to one another. The enzymes also differ in the mode of formation of the dimer interface. In the enzyme RgDAAO, the loops consisting of 14 amino acid residues located in the lower part of the FAD binding domain play a key role in the formation of the dimer. These loops provide the necessary orientation and the efficient interaction between the positively charged residues Arg305 and Arg314 of this loop from one subunit and the negatively charged residues Asp269, Glu275, and Glu276 of the helices I3´ and I3 from another sub

Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 1, pp. 262--268, January, 2010. 1066 5285/10/5901 01 © 2010 Springer Science+Business Media, Inc.


2

Russ.Chem.Bull., Int.Ed., Vol. 59, No. 1, January, 2010

Cherskova et al.

unit.15 Mutants of the enzyme with deletions in this loop were prepared.15--17 It appeared that five deletions in this loop are sufficient for RgDAAO to lose the ability to form dimers. The monomeric form of mutant RgDAAO has a much lower thermal stability compared to the wild type enzyme, and the FAD binding constant of the mutant is 5 times lower.15,17 These results indicate that the above mentioned loop plays a key role in the existence of RgDAAO as a dimer in vivo and provides strong binding of the FAD molecule in the active site. An analysis of the structure of the Cys108Phe TvDAAO mutant showed that the dimeric structure of the enzyme is stabilized by six pairs of hydrogen bonds between residues from different subunits, two residues, viz., Arg169 and Arg220, forming two hydrogen bonds each (Fig. 1). The residue Arg169 forms two hydrogen bonds through the NH(1) and NH(2) atoms of the guanidinium group with the carbonyl oxygen atom of the residue Gly250 from the second subunit (the distances are 3.10 and 3.11 å, respec tively). The residue Arg220 forms hydrogen bonds through

the NH(E) and NH(2) atoms of the guanidinium group with the carbonyl oxygen atom of the residue Glu32 (the distances are 2.86 and 3.06 å, respectively) (see Fig. 1, b). Therefore, two pairs of the residues Arg169 and Arg220 of each subunit are responsible for the formation of eight of the twelve (67%) hydrogen bonds. The aim of the present study was to reveal the role of the residues Arg169 and Arg220 in the structure and sta bility of TvDAAO, primarily, to investigate the influence of these residues on the oligomeric structure of the en zyme. For this purpose, we prepared two mutants of the enzyme, Arg220Glu TvDAAO and Arg(169,220)Ala TvDAAO. In the former mutant, the substitution of the negatively charged residue Glu for the positively charged residue Arg should lead to the repulsion from the residue Glu32 of the adjacent subunit. The substitution of the residues Ala for two residues Arg enables the elimination of hydrogen bonds. Experimental
All "Molecular Biology Grade" reagents were used in the gene engineering. Microbiological experiments were performed with the use of the following reagents: Bacto tryptone, the yeast extract and agar (Difco, USA), glycerol (99.9%, ultra pure), potassium hydrogenphosphate and sodium chloride (ultra pure), lysozyme (Fluka/BioChemika, Switzerland), potassium chloride and magnesium chloride (ultra pure, Merck, Switzerland), iso propyl D thiogalactoside (IPTG), kanamycin and chloram phenicol (Sigma, USA), sodium hydrogenphosphate (analytical grade), and glucose (ReaKhim, Russia). Restriction endonu cleases, bacteriophage T4 DNA ligase, and Taq DNA poly merase (Fermentas, Lithuania) were used for the DNA frag ment cloning and site directed mutagenesis. Oligonucleotides for the polymerase chain reaction and sequencing were synthe sized by Sintol (Russia). The experiments were performed with MilliQ water (Millipore, USA). Introduction of mutations into the tvdaao gene. Single muta tions were introduced by the polymerase chain reaction (PCR) according to a procedure described previously.14 The pDAAO2B plasmid was used as the matrix, in which the daao gene was under control of the strong promoter of the bacteriophage T7 RNA polymerase.18 The mutations were introduced with the use of the forward (DAOFor1) and reverse (DAORev5) primers for the beginning and the end of the gene, respectively (Fig. 2, a), as well as of the forward (Pr1) and reverse (Pr2) primers carrying the required substitution in the tvdaao gene (Fig. 2, b). The reaction mixture for the PCR reaction contained 2.5 µL of a 10â SE buffer for Taq DNA polymerase (60 mM Tris HCl, pH 8.5, at 25 °C), 1.5 mM MgCl2, 25 mM KCl, 10 mM 2 mer captoethanol, 0.1% Triton X 100), 2.5 µL of a mixture of dNTP (dATP, dGTP, dTTP, dCTP; the concentration of each compo nent was 2.5 µmol L­1), 1 µL of the DNA matrix (10 ng µL­1), 2 µL of each primer (5 nmol mL­1), 1 µL of Taq DNA poly merase (5 U µL­1), and deionized water (added to the total volume of the mixture of 25 µL). The PCR reaction was carried out in a 0.5 mL thin wall plastic tube (Eppendorf, Germany). To prevent the evaporation

a

b
Glu32A Gly250B Arg169A NH(1) O NH(2) NH(2) NE O

Arg220B

Fig. 1. Hydrogen bonds in the structure of TvDAAO formed by the residues Arg169 and Arg220: a, the positions of these re sidues in the protein globule (enclosed in the black square); b, an enlarged fragment.


Modification of intersubunit interactions in DAAO

Russ.Chem.Bull., Int.Ed., Vol. 59, No. 1, January, 2010

3

a
DAOFor1 DAORev5

5´-ATA TAC CAT GGC TAA AAT CGT TGT TAT TGG TGC-3´ 5´-GTT TGG ACG AGT AAG AGC TCT TTC GAC­3 ´, b

Substitution Arg220Glu Pr1 Pr2

5´-GTC GTC CTT GTT GAA AAC TCT CTT CCT TTT ATG GCC TC-3´ 5´-CAT AAA AGG AAG AGA GTT TTC AAC AAG GAC GAC TTG TCC TCG-3 ´ 5´-GTC GTC CTT GTT GCA AAC TCT CTT CCT TTT ATG GCC TC-3´ 5´-CAT AAA AGG AAG AGA GTT TGC AAC AAG GAC GAC TTG TCC TCG-3 ´ 5´-ACG GTG GTT AAA GCT CGA GTG AAC CAT ATC AAG GAT GC-3´ 5´-TT CAC TCG AGC TTT AAC CAC CGT GGC GCC G-3´

Substitution Arg220Ala Pr1 Pr2

Substitution Arg169Ala Pr1 Pr2

Fig. 2. Forward (DAOFor1) and reverse (DAORev5) primers for the beginning and the end of the daao gene, respectively (a) and the forward (Pr1) and reverse (Pr2) primers carrying the required substitution in the tvdaao gene (b). The nucleotide substitutions providing the mutation are in the boldface type. of the reaction mixture and its condensation on the cap, 30 µL of mineral oil was added to the tube. The tube was heated at 95 °C for 5 min, and then the PCR reaction was carried out using the following program: in the first step, 95 °C, 30 s; in the second step, 56--60 °C, 60 s; in the third step, 72 °C, 2 min; a total of 25--35 cycles. Then the reaction mixture was kept at 72 °C for 10 min. The temperature in the second step was 3--5 °C lower than the melting point of the duplexes (Tm) formed by the prim ers. The melting points Tm were determined by the empiri cal formula: Tm = 2(nA + nT) + 4(nG + nC), where nX is the number of nucleotides X (X = A, T, C, or G) in the primer. To obtain fragments containing the required mutation, we carried out two PCR reactions using pairs of the primers DAOFor1/Pr2 (fragment 1) and Pr1/DAORev5 (fragment 2). The PCR products, fragment 1 and fragment 2, were purified using 1% agarose gel. In the third step, PCR was conducted with the primers DAOFor1 and DAORev5 and the previously pre pared fragments 1 and 2 used as the DNA matrix. The product of the third PCR reaction was purified analogously to fragments 1 and 2 and treated with Bsp119 I and EcoICR I restriction en donucleases. Then the DNA fragments were purified and ligated with the plasmid pDAAO2, which was treated with the same restriction endonucleases. After the ligation, E. coli TG1 cells were transformed with the reaction mixture, seeded into Petri dishes with an agar medium containing kanamycin (30 µg mL­1), and kept for 16 h. For each mutant, six colonies were taken from the dishes, and plasmids were isolated from the colonies. The introduction of the requires mutations was monitored by sequencing the plasmid DNA. Expression of TvDAAO and its mutants in E. coli cells. The expression of TvDAAO and its mutants was carried out in E. coli BL21 (DE3) pLysS Codon Plus cells. The producer strain of E. coli BL21 (DE3) pLysS Codon Plus cells was transformed with the corresponding plasmid and seeded into Petri dishes

with an agar medium containing the required antibiotics. To prepare the inoculate, a single colony was taken from the Petri dish and cultured for 16 h on a 2YT medium (16 g L­1 Bacto tryptone, 10 g L­1 yeast extract, 1.5 g L­1 sodium dihydrogen phosphate, 1 g L­1 sodium chloride, 1 g L­1 potassium hydro genphosphate, pH 7.5) in the presence of 30 µg mL­1 kanamy cin and 25 µg mL­1 chloramphenicol. After overnight incu bation, the cells were transferred into a fresh medium (the dilu tion was 1 : 1000) and cultured at 30 °C until the absorbance A600 0.6--0.8. The inoculate was placed into flasks for cultiva tion in an amount of 10% of the total volume of the medium (the modified LB medium containing 10 g L­1 yeast extract, 5 g L­1 Bacto tryptone, 10 g L­1 glucose, 1.5 g L­1 sodium dihydrophos phate, and 1 g L­1 potassium hydrogenphosphate, pH 7.5) con taining 30 µg mL­1 kanamycin. The cultivation was carried out in 1 L conical flasks equipped with paddle impellers (the volume of the medium was at most 10--15% of the flask volume). The temperature of cultivation varied in the range of 18--27 °C; the shaker speed was 120--160 rpm. After the achievement of the absorbance A600 0.7--0.8, the expression of the enzyme was induced by adding IPTG to the medium. After the induction, the cells were cultured for 24 h and then precipitated on a Beck mann J 21 centrifuge (8000 rpm, a J 20 rotor, 30 min, 4 °C). The precipitate was resuspended in a 0.02 M Tris HCl buffer (pH 7.5, 8.0, or pH 8.5) in a ratio of 1 : 4 (w/w) and subjected to sonication. The suspension was kept at ­20 °C. Determination of activity and kinetic parameters of DAAO. The activity of D amino acid oxidase was determined with the use of the DAAO--horseradish peroxidase two enzyme system. D Methionine and 2,2´ azinobis(3 ethylbenzothiazoline 6 sul fonic acid) (ABTS) were used as substrates for these two en zymes, respectively. The activity was determined at 30 °C from the accumulation of the oxidation product of ABTS (the absor bance at 414 nm, 414 = 36000 L mol­1 cm­1) using a Shimadzu UV 1601PC spectrophotometer (Germany). A measurement cell of the spectrophotometer (working volume was 1 mL, optical path was 1 cm) was filled with 870 µL of an air saturated 50 mM potassium phosphate buffer (PPB), pH 8.0, 100 µL of a 100 mM


4

Russ.Chem.Bull., Int.Ed., Vol. 59, No. 1, January, 2010

Cherskova et al.

Met solution in 50 mM PPB, 20 µL of an ABTS solution in water (16 mg mL­1), and 10 µL of a solution of peroxidase in 50 mM PPB (5 mg mL­1). The cell was stored under tempera ture controlled conditions for 10 min at 30 °C and then a sample of wild type TvDAAO or the corresponding mutant was added (30--40 µL). To determine the maximum reaction rate and the Michaelis constant, the concentration of the corresponding D amino acid was varied from 0.5 to 5 KM. The approximate value of KM was determined in a separate experiment by measuring the reaction rate at concentrations of the corresponding D amino acid of 0.1, 0.5, 1.0, 5.0, 10, and 50 mmol L­1. The kinetic data were analyzed using the OriginPro 7.0 SR4 program by linear and non linear regression methods. Study of stability at different temperatures. The thermal sta bility of the mutants and wild type TvDAAO was studied in a 0.1 M potassium phosphate buffer, pH 8.0. For each experi ment, 0.5 mL plastic tubes containing 100 µL of the enzyme solution were placed into a water thermostat preheated to the required temperature (the accuracy of the temperature setting was ±0.1 °C). At certain time intervals, the tubes were with drawn and cooled in ice for 1--2 min, and then the activity was measured as described above. The sampling interval was chosen so that the activity of the enzyme in the course of the experiment decreased to 10--15% of the initial value. The rate constants for inactiviation were determined by plotting the residual activity versus time. The plots were processed by the non linear regres sion method using the OriginPro 7.0 SR4 program (OriginLab).
D

Results and Discussion Preparation of Arg220Glu and Arg(169,220)Ala TvDAAO mutants. Nucleotide substitutions providing the

required mutations were introduced by the polymerase chain reaction. The Arg(169,220)Ala TvDAAO double mutant was prepared in two steps. First, the Arg169Ala TvDAAO mutant was prepared, and then the Arg220Ala substitution was made. For each of the three mutants, six plasmids were selected for sequencing. The results of se quencing showed that only required mutations were present in all plasmids in the tvdaao gene, whereas other nucle otide substitutions were absent. Plasmids containing the Arg220Glu and Arg(169,220)Ala substitutions in the tvdaao gene were transformed into the BL21 (DE3) Codon Plus/pLysS strain. The recombinant strains thus prepared were cultured for 24 h according to the procedure described in the Experimental section. The cultivation showed that the Arg220Glu and Arg(169,220)Ala mutants of TvDAAO were synthesized completely in the insoluble form as inclusion bodies (Fig. 3) with retention of the enzymatic activity. It is known that a decrease in the concentration of the inductor and the temperature of cultivation allows the preparation of the target recombinant protein in the solu ble form. In the present study, we carried out experiments on the optimization of cultivation by varying the concen tration of the inductor IPTG from 0.005 to 0.1 mmol L­1, and reduced the temperature of cultivation from 25 to 18 °C. A decrease in the concentration of IPTG from 0.1 to 0.05 mmol L­1 was shown to increase the yield of the enzyme by a factor of 1.5--2.5. However, both mutants were still expressed only in the insoluble form. We also made efforts to solubilize the Arg220Glu and Arg(169,220)Ala TvDAAO mutants. For this purpose, the

1

2

3

4

5

6

Fig. 3. Protein electrophoresis under denaturing condition of the Arg(169,220)Ala TvDAAO mutant (lanes 1 and 2), wild type TvDAAO (lanes 3 and 4), and the Arg220Glu TvDAAO mutant (lanes 5 and 6). The soluble and insoluble fractions are labeled by odd and even numbers, respectively. The positions of the mutant TvDAAO and the wild type enzyme on the gel are indicated by arrows.


Modification of intersubunit interactions in DAAO

Russ.Chem.Bull., Int.Ed., Vol. 59, No. 1, January, 2010

5

precipitate of the cell suspension, which was prepared by sonication followed by removal of the supernatant, was treated with non ionogenic surfactants (Twin 20, Twin 40, Twin 80, Twin 100, Triton X 100, and Triton X 350) and with low concentrations of urea. However, in these cases we also failed to obtain the enzyme in the soluble form. Since the precipitate of the cell suspension exhibited activity, we decided to measure the kinetic parameters of the membrane bound enzyme. Determination of the kinetic parameters of the Arg220Glu and Arg(169,220)Ala TvDAAO mutants. The results of determination of the kinetic properties of the Arg220Glu and Arg(169,220)Ala mutants of TvDAAO are given in Table 1. It was impossible to determine the abso lute values of the catalytic constant kcat, because there are no methods that can be used to determine the concentra tion of the active sites in the membrane bound mutants of TvDAAO. Hence, the activities for different D amino ac ids are given relative to the activity with respect to D methionine, which is the best substrate for wild type TvDAAO, as well as for its mutants. As can be seen from Table 1, the introduction of the Arg220Glu and Arg(169,220)Ala mutations into the TvDAAO molecule leads to a sharp change in the sub strate specificity profile. In particular, the enzyme lost the activity with respect to D Leu, D Thr, D Tyr, and D Ala. In addition, the Arg220Glu TvDAAO mutant does not exhibit activity with respect to D Ser and D Phe. The larg est changes in the kinetic parameters were observed for the Arg220Glu TvDAAO mutant. The Michaelis constants of the latter mutant with respect to most of the D amino acids under study are substantially higher. For some of the substrates, the almost linear dependence of the reaction rate on the concentration of the D amino acid was ob

served up to the maximum solubility of the latter. Hence, only the ratio Vmax/Km but not the individual parameters can be determined. For the Arg(169,220)Ala TvDAAO double mutant, the Michaelis constants with respect to some D amino acids, in particular, D valine, D asparagine, and D lysine, in creased several fold. Thermal stability of Arg220Glu and Arg(169,220)Ala TvDAAO. The Arg220Glu and Arg(169,220)Ala mutants of TvDAAO are much less stable than the wild type en zyme. For wild type TvDAAO, TM (the temperature at which the initial activity decreased by 50% in 30 min) is 54 °C, whereas TM for the Arg220Glu and Arg(169,220)Ala mutants are 28 and 20 °C, respectively, i.e., by 26 and 34 degrees lower than that for the wild type enzyme. Hence, the kinetics of thermal inactivation of the Arg220Glu and Arg(169,220)Ala mutants of TvDAAO was measured in the temperature range of 32--40 and 20--28 °C, respectively. Previously, the investigation of the thermal denatur ation of wild type TvDAAO showed that the time depen dence of the residual activity of the enzyme is described by the sum of two exponentials, and the inactivation occurs at least in two steps.14 The reversible dissociation of the dimer into monomers as the first step is followed by the irreversible inactivation. The time dependences of the re sidual activity (A/A0) of the Arg220Glu TvDAAO mutant for the lower and upper limits in the temperature range of measurements (32 and 40 °C, respectively) are shown in Fig. 4. The inactivation curve at 32 °C is well described only by the sum of two exponentials. At 40 °C, the two exponential character of the inactivation curve is less evident. Similar results were obtained also for the Arg(169,220)Ala TvDAAO mutant (not shown in Fig. 4). Hence, Fisher's criterion was used for the quantitative es

Table 1. Kinetic parameters of wild type TvDAAO and its Arg169Ala/Arg220Ala and Arg220Glu mutants Amino acid TvDAAO Arg169Ala/Arg220Ala K /mmol L
­1

TvDAAO Arg220Glu K /mmol L
­1

Wild type TvDAAO K /mmol L­1

(%)
D D D D D D D D D D D

(%) 100 2 48 8 42 3 * * * * * 0.78±0.5 * >100 * 30±4 >100 * * * * *

(%) 100 * *** * 74 ** * * * * *

(%) 100 * 17.7 * 2 49.4 * * * * * 0.24 28.7 17.7 0.40 15.8 49.4 0.78 10 0.73 0.3 0.78

(%) 100 15 89 40 59 39 34 5 38 30 34

(%) 100 0.1 1.2 24 0.9 0.2 10 0.1 13 24 10

Met Ser Val Phe Asn Lys Leu Thr Tyr Ala Leu

3.0±0.9 105±33 4.3±0.9** 38±5 5.6±0.7 27 ±7 * * * * *

100 62 69 100 85 25 * * * * *

* The reaction does not proceed. ** The kinetic parameters of the mutant enzyme that exceed those of wild type TvDAAO are in the boldface type. *** Data are absent.


6

Russ.Chem.Bull., Int.Ed., Vol. 59, No. 1, January, 2010

Cherskova et al.

A/A 1.0 0.8 0.6 0.4 0.2

0

a

A/A 1.0 0.8 0.6 0.4 0.2

0

b

0 A/A 1.2 1.0 0.8
0

20

40

60

80 t/min

0 A/A 1.0 0.8 0.6
0

20

40

60

80 t/min

c

d

0.6 0.4 0.4 0.2 0 5 10 15 20 25 t/min 0.2

0

5

10

15

20

25 t/min

Fig. 4. The inactivation curves of the Arg220Glu TvDAAO mutant at 32 (a, b) and 40 °C (c, d) described by one exponential (a, c) and by the sum of two exponentials (b, d) (0.1 M phosphate buffer, pH 8.0).

timation of the validity of the two exponential model for the description of the inactivation curves of both mutants throughout the temperature range of measurements. Ac cording to this criterion, the model proposed for the ap proximation of the experimental data is adequate if calcu lated Fisher's coefficient (Fexp) is smaller than the theo retical value (Fcalc). Table 2 exemplifies the results of cal culations of Fexp for one and two exponential models for the description of the inactivation curves of the Arg220Glu and Arg(169,220)Ala mutants of TvDAAO at 40 and 28 °C, respectively, at which the two exponential model is less evident. The theoretical coefficient Fexp was taken from Fisher's table for the confidence probability of 0.95. As can be seen from Table 2, in the case of the simple expo nential model, calculated Fisher's coefficient is much larg er than the theoretical value. This means that this model is not adequate for the description of experimental data. In the case of the approximation by the sum of two exponen tial functions, calculated Fisher's criterion indicates that the model is adequate. Hence, only the approximation of the experimental data on the inactivation of both the Arg220Glu and Arg(169,220)Ala mutants of TvDAAO by the sum of two exponential functions is adequate and sta

tistically significant. Therefore, the results of calculations indicate that, as in the case of the wild type enzyme, the inactivation of both mutant TvDAAO occurs at least in two steps. Hence, it can be concluded that both mutant enzymes also exist as dimers. Table 3 gives the numerical values of the calculated effective rate constants for inactivation for the first and second steps in the temperature range under study. As is

Table 2. Estimation of the validity of the one (I) and two expo nential (II) models for the description of the kinetics of inactiva tion of the Arg220Glu and Arg(169,220)Ala TvDAAO mutants at 40 and 28 °C, respectively Mutant enzyme Parameter I TvDAAO Arg(169,220)Ala TvDAAO Arg220Glu * I: y = A·exp
­kx

Model* II 2.24 2.62 1.96 2.43

Fexp Fcalc Fexp Fcalc
­k1·x

21.15 2.45 134.45 2.32 + B·exp
­k2·x

, II: y = A·exp


Modification of intersubunit interactions in DAAO

Russ.Chem.Bull., Int.Ed., Vol. 59, No. 1, January, 2010

7

Table 3. Effective rate constants for inactivation of R220E TvDAAO and R169A/R220A TvDAAO at different temperatures Mutant enzyme / ° k1 min TvDAAO R220E 32 34 36 38 40 TvDAAO R169A/R220A 20 22 24 26 28 0.184±0.023 0.253±0.08 0.490±0.11 0.730±0.18 1.000±0.26 0.250±0.004 0.480±0.10 0.510±0.12 0.770±0.15 0.960±0.17
­1

k

2

the removal of hydrogen bonds involved in the intersub unit contact leads to the transformation of TvDAAO in the completely insoluble form. We plan to perform amino acid substitutions that would enhance the intersubunit con tact in the TvDAAO molecule and lead to an increase in thermal stability of the enzyme. The present study was supported by the Russian Foun dation for Basic Research (Project No. 08 04 01703) and the Federal Agency for Science and Innovations of the Russian Federation (Contract 02.512.11.2299). References
1. L. Pollegioni, L. Piubelli, S. Sacchi, M. S. Pilone, G. Molla, Cell Mol. Life Sci., 2007, 64, 1373. 2. S. V. Khoronenkova, V. I. Tishkov, Usp. Biol. Khim., 2008, 48, 359 [Biochemistry (Moscow),. Special Issue, Biological Chemistry Reviews (Engl. Transl.), 2008, 73, 1511]. 3. V. I. Tishkov, S. V. Khoronenkova, Biokhimiya, 2005, 70, 51 [Biochemistry (Moscow) (Engl. Transl.), 2005, 70, 40]. 4. L. Pollegioni, G. Molla; S. Sacchi, E. Rosini, R. Verga, M. S. Pilone, Appl. Microbiol. Biotechnol., 2008, 78, 1. 5. V. I. Tishkov, S. S. Savin, S. V. Khoronenkova, Izv. Akad. Nauk, Ser. Khim., 2008, 57, 1014 [Russ. Chem. Bull., Int. Ed., 2008, 57, 1033]. 6. H. Mizutani, I. Miyahara, K. Hirotsu, Y. Nishina, K. Shiga, C. Setoyama, R. Miura, J. Biochem. (Tokyo), 1996, 120, 14. 7. A. Mattevi, M. A. Vanoni, F. Todone, M. Rizzi, A. Teplyak ov, A. Coda, M. Bolognesi, B. Curti, Proc. Natl. Acad. Sci., USA, 1996, 93, 7496. 8. T. Kawazoe, H. Tsuge, M. S. Pilone, K. Fukui, Protein Sci., 2006, 15, 2708. 9. T. Kawazoe, H. K. Park, S. Iwana, H. Tsuge, K. Fukui, Chem.Rec., 2007, 7, 305. 10. S. Umhau, L. Pollegioni, G. Molla, K. Diederichs, W. Welte, M. S. Pilone, S. Ghisla, Proc. Natl. Acad. Sci., USA, 2000, 97, 12463. 11. L. Pollegioni, K. Diederichs, G. Molla, S. Umhau, W. Welte, S. Ghisla, M. S. Pilone, J. Mol. Biol., 2002, 324, 535. 12. C. P. Berg, F. A. Rodden, Anal. Biochem., 1976, 71, 214. 13. V. I. Tishkov, S. V. Khoronenkova, L. I. Savina, RF Pat. 2 362 806, July 23, 2007, Byul. Izobr. Poleznye modeli [Inven tor Bull., Useful Models], 2009, No. 21 (in Russian). 14. S. V. Khoronenkova, V. I. Tishkov, Chin. J. Biotechnol., 2008, 24, 2122. 15. L. Piubelli, G. Molla, L. Caldinelli, M. S. Pilone, L. Polle gioni, Protein Eng., 2003, 16, 1063. 16. L. Piubelli, L. Caldinelli, G. Molla, M. S. Pilone, L. Polle gioni, FEBS Lett., 2002, 526, 43. 17. L. Pollegioni, S. Iametti, D. Fessas, L. Caldinelli, L. Piubel li, A. Barbiroli, M. S. Pilone, F. Bonomi, Protein Sci., 2003, 12, 1018. 18. S. V. Khoronenkova, V. I. Tishkov, Anal. Biochem., 2008, 374, 405.

0.0078±0.0019 0.020±0.005 0.039±0.008 0.047±0.013 0.068±0.015 0.0096±0.0024 0.032±0.005 0.044±0.008 0.061±0.110 0.076±0.016

evident from the above considered data, the effective rate constant k1 characterizing the dissociation of the dimer more slowly increases with temperature than the effective rate constant k2 characterizing the rate of inactivation of the monomer, i.e., the energy of activation for the second step is higher than that for the first step. This is in good agreement with our model of inactivation of TvDAAO, because the energy of activation for the dissociation of the dimer into individual subunits should be lower than that necessary for the unfolding of the protein globule in the step of inactivation of the monomer. The above considered data indicate that the arginine residues in positions 169 and 220 play a considerable role in the formation of the quaternary structure and ensuring the stability of TvDAAO. The substitution of other amino acid residues for these residues results in that the enzyme is synthesized in the active but insoluble form. The intro duction of the Arg220Glu mutation into the TvDAAO molecule substantially impairs the kinetic parameters with different substrates and leads to a decrease in stability of the enzyme. In the case of the double substitution Arg169Ala/Arg220Ala, the higher constants KM with re spect to a series of D amino acids were observed in spite of the fact that the kinetic properties were studied for the cell wall bound enzyme. It should be noted that substantial changes in the substrate specificity profile were caused by the substitutions of the amino acid residues that are not involved in the active site of the enzyme. An analo gous effect was observed in the case of the Cys108Phe, Cys108Ala, and Cys108Ser single mutations.13 The first mutation leads to an increase in thermal stability of TvDAAO, whereas the second and third mutations cause a decrease in thermal stability. The results of our experiments also indicate that the existence of TvDAAO in the dimeric form is more critical for its stability compared to the enzyme from the yeast R.gracilis. In the latter case, the monomeric mutant en zyme was obtained in the active and soluble form, whereas

Received August 26, 2009