Roles of Hydrogen Bonding Residues in the Interaction between the alpha  and beta  Subunits in the Tryptophan Synthase Complex
Asn-104 OF THE alpha  SUBUNIT IS ESPECIALLY IMPORTANT*

(Received for publication, September 9, 1996, and in revised form, December 6, 1996)

Kaori Hiraga Dagger and Katsuhide Yutani §

From the Institute for Protein Research, Osaka University, 3-2 Yamadaoka, Suita, Osaka 565, Japan

ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
Acknowledgment
REFERENCES


ABSTRACT

The interaction of the alpha  subunit with the beta 2 subunit of tryptophan synthase is known to be necessary for the activation of each subunit and for the catalytic efficiency of the alpha 2beta 2 complex. To elucidate the roles of hydrogen bonds in the interaction site between the alpha  and beta  subunits for subunit association, eight mutant alpha  subunits at five hydrogen bonding residues (N104D, N104A, N108D, N108A, E134A, E135A, N157D, and N157A) were constructed, and the thermodynamic parameters of association with the beta  subunit were obtained using a titration calorimeter. The N104D and N104A mutations remarkably decreased the stimulation activities, the association constants, and the association enthalpies. Although the association constant and the stimulation activities of E134A were reduced in the absence of salt, the change in the association enthalpy was relatively small, and the addition of salt could repair its defects. The substitutions at positions 135 and 157 did not affect the stimulation activity and decreased the Gibbs energy of association corresponding to the defect in 1 mol of hydrogen bond. The present results suggest that the alpha  subunit which has a mutation at position 104 cannot fold into an intact conformation upon complex formation, resulting in reduced stimulation activities. The hydrogen bond with Asn-104, which is a conserved residue among 16 microorganisms, was especially important for alpha /beta interaction and mutual activation.


INTRODUCTION

Protein-protein interactions play a central role in many physiologically important reactions. The tryptophan synthase complex is an excellent model enzyme for exploring the molecular recognition mechanism in protein-protein interaction (1, 2). Bacterial tryptophan synthase is an alpha 2beta 2 complex. The separated alpha  and beta subunits catalyze inherent reactions termed the alpha  and beta  reactions, respectively (3-5). When the alpha  and beta  subunits combine to form the alpha 2beta 2 complex, the enzymatic activity of each subunit is stimulated by 1 to 2 orders of magnitude. It has been reported that this activation is due to conformational changes in both subunits upon alpha /beta subunit interaction (reviewed in Ref. 1).

Some intermolecular electrostatic interactions (hydrogen bonds and salt bridges) and hydrophobic (packing) interactions were observed in many strongly bound complexes (6-8). In particular, electrostatic interaction is thought to play a critical role in protein-protein interaction (8-11). We focused on hydrogen bonds in the interaction site between the alpha  and beta  subunits of tryptophan synthase, termed herein "alpha /beta subunit hydrogen bonds." A thermodynamic study of the association of proteins having mutations at interface residues is important to elucidate the mechanisms of the subunit interaction in the tryptophan synthase complex. Also, it should provide useful information for understanding the energetic contribution of the hydrogen bonds that stabilize the protein complex. The amino acid sequences of the alpha  and beta  subunits from Escherichia coli are highly similar to those from Salmonella typhimurium (12); therefore, the three-dimensional structures have been considered to be almost identical (13). Residues that form hydrogen bonds in the interaction site between the alpha  and beta  subunits were identified from the x-ray structure of the alpha 2beta 2 complex from S. typhimurium (14). Table I identifies the hydrogen bonding partner in the beta  subunit for the alpha  subunit residues and the percent conservation of each residue (12, 15-26). In this paper, eight mutant alpha  subunits at five positions were constructed except for position 133, which forms a hydrogen bond at an atom in the main chain (Table I). The roles of hydrogen bonding residues in the subunit interaction were investigated by activity measurements and titration calorimetry of the association. We will discuss the contribution of electrostatic interaction and the roles of hydrogen bonding residues in the subunit association in the complex.

Table I.

The intersubunit hydrogen bonding residues in the alpha  subunit

All the data were obtained from the x-ray structure (Protein Data Bank code 1wsy) of the alpha 2beta 2 complex from S. typhimurium (13).
Structurea Residue  alpha beta ASAb (%)  alpha ASA (%)  Delta ASAb (%) Hydrogen-bonding part Conservationc Partner in beta  subunit (distance Conservationc

% %
t N104 1.06 60.72 59.66 Side chain ND2 100 292G; main chain O 100
(2.72 Å)
t N108 15.94 66.29 50.35 Side chain ND2 25 290Ad; main chain O A:13
(2.85 Å) E:44
r V133 3.77 6.60 2.83 Main chain N 25 19Q; side chain OE1 19
(2.78 Å)
r E134 40.73 73.90 33.17 Side chain OE1 75 19Q; side chain NE2 19
(2.70 Å)
r E135 13.81 66.05 52.24 Side chain OE1 81 15M; main chain N 13
(2.69 Å)
r N157 20.40 91.23 70.83 Side chain ND2 31 181Y; side chain OH 19
(2.76 Å)

a  Secondary structure locating the residue of the alpha  subunit. t and r represent a turn structure and a random loop, respectively.
b  The values of alpha beta ASA and alpha ASA mean the accessible surface area (ASA) calculated for the residues in the complex form and in the alpha  subunit monomer without the beta  subunit, respectively. The values of Delta ASA are the difference between them, indicating the area that is buried in the protein interior due to complex formation.
c  The percentage of conservation at each position among the alpha  or beta  subunits from 16 organisms: E. coli (12), S. typhimurium (12), Klebsiella aerogenes (12), Vibrio Parahaemolyticus (15), Pseudomonas putida (16), Pseudomonas aeruginosa (17), Caulobacter crescentus (18), Brevibacterium lactofermentum (19), Bacillus subtilis (20), Lactobacillus casei (21), Saccharomyces cerevisiae (22), Thermus thermophilus (23), Bacillus stearothermophilus (24), Haloferax volcanii (25), Methanococcus voltae (26), and Bacillus coagulans.
d  The hydrogen bonding partner of Asn-108 is Ala-290 in the complex from S. typhimurium, whereas it is Glu-290 in the complex from E. coli.


EXPERIMENTAL PROCEDURES

Materials

Mutant alpha  Subunits from E. coli

Five alpha /beta hydrogen bonding residues of the alpha  subunit from E. coli were replaced by means of site-directed mutagenesis using synthetic oligonucleotides. The sequences of the template DNAs and the oligonucleotides are shown in Table II. Glu-134 and Glu-135 could be directly substituted with Ala by a single code mutagenesis (E134A and E135A). Asn residues at positions 104, 108, and 157 were substituted with Asp (N104D, N108D, and N157D) or Ala (N104A, N108A, and N157A). The eight mutant alpha  subunits from E. coli were purified as described (27).

Table II.

The sequences of the synthetic oligonucleotide and template M13-trpA DNAs used for the site-directed mutagenesis of the alpha subunit of tryptophan synthase from E. coli


mutant Desired mutation DNA sequencea Template M13-trpAb

-5'-TGTATGCC<UNL>AAT</UNL>CTGGTGT-3'-
N104D Asnright-arrowAsp    TGTATGCC<UNL><B>G</B>AT</UNL>CTGGTG Wild type
N104A Aspright-arrowAla     GTATGCC<UNL>G<B>C</B>T</UNL>CTGGTGT N104D
-5'-TGGTGTTT<UNL>AAC</UNL>AAAGGCA-3'-
N108D Asnright-arrowAsp    TGGTGTTT<UNL><B>G</B>AC</UNL>AAAGGC Wild type
N108A Aspright-arrowAla      GGTGTTT<UNL>G<B>C</B>C</UNL>AAAGGCA N108D
-5'-GCCAGTT<UNL>GAA</UNL>GAGTCCG-3'-
E134A Gluright-arrowAla    GCCAGTT<UNL>G<B>C</B>A</UNL>GAGTCCG Wild type
-5'-CCAGTTGAA<UNL>GAG</UNL>TCCGCGCCC-3'-
E135A Gluright-arrowAla    CCAGTTGAA<UNL>G<B>C</B>G</UNL>TCCGCGCCC Wild type
-5'-GCCCGCCA<UNL>AAT</UNL>GCCGATG-3'-
N157D Asnright-arrowAsp    GCCCGCCA<UNL><B>G</B>AT</UNL>GCCGAT Wild type
N157A Aspright-arrowAla      CCCGCCA<UNL>G<B>C</B>T</UNL>GCCGATG N157D

a  The DNA sequence coding a residue to be substituted and the mismatch nucleotide are represented by the underlined and bold letters, respectively.
b  The templates M13-trpA coding the wild-type alpha  subunit were used for the mutations of Asn with Asp and of Val or Glu with Ala, and the templates M13-trpA' coding the Asp at each position were used for the mutation of Asp with Ala.

beta Subunit from E. coli

The beta  subunits of tryptophan synthase from E. coli can be obtained as a dimer form (beta 2). The E. coli strain SP974 transformed with plasmid pUC9BG-2 (a gift from Dr. Somerville, Purdue University, West Lafayette, IN) coding the wild-type beta  subunit of tryptophan synthase from E. coli was grown (28), and the protein was purified as described (29).

Methods

Protein Concentrations

Protein concentrations of the wild-type and mutant alpha  subunits were estimated from the absorbance at 278.5 nm, assuming E1%cm = 4.4 (30, 31). The beta  subunit was estimated from the absorbance at 280 nm, assuming E1%cm = 6.5 (32).

Isothermal Titration Calorimetry (ITC)1 Measurements of the Association between the alpha  and beta  Subunits

The apparent association constant (Ka) and association enthalpy (Delta Ha) of the alpha  subunit with the beta  subunit were calorimetrically determined at 40 °C using an OMEGA titration calorimeter from MicroCal, Inc., as already described (29, 33). Prior to the experiments, the sample proteins were dialyzed against a 50 mM potassium phosphate buffer, pH 7.0, containing 0.1 mM dithiothreitol, 5 mM EDTA, and 0.2 mM pyridoxal 5'-phosphate. For the experiments in the presence of additional salt, 300 mM KCl was added to the buffer. The solution of the beta  subunits in a cell (0.015-0.04 mM, 1.3115 ml) was titrated with the solution of the alpha  subunits (0.1-0.3 mM) using a 250-µl long needle injection syringe. The Delta Ha and Ka values were calculated using the ORIGIN computer program (MicroCal, Inc.). The experimental data were analyzed using a single set of identical sites model, where two alpha  subunit binding sites per beta 2 subunit were identical and were independent of each other. The association Gibbs energy (Delta Ga) and entropy (Delta Sa) can be calculated from experimental Ka and Delta Ha values using the following equations,
&Dgr;G<SUB><UP>a</UP></SUB>=−RT <UP>ln</UP>K<SUB><UP>a</UP></SUB>
&Dgr;S<SUB><UP>a</UP></SUB>=(&Dgr;H<SUB><UP>a</UP></SUB>−&Dgr;G<SUB><UP>a</UP></SUB>)<UP>/T</UP>

Enzyme Assay

The forward alpha  reaction, which is the conversion of indole-3-glycerol phosphate to indole, was measured by changes in the absorption at 340 nm of NADH produced in the reaction coupled with glyceraldehyde-3-phosphate dehydrogenase (34). The beta  reaction, which is the formation of L-tryptophan from indole and L-serine, was assayed utilizing the difference in absorption between indole and L-tryptophan at 290 nm (35). 1 unit of activity in each reaction is the conversion of 0.1 µM substrate to product in 20 min at 37 °C.

The inherent activities of the alpha  subunits in the forward alpha  reaction were determined in the absence of the beta 2 subunit in 0.1 M Tris-HCl buffer, pH 8.0, at 37 °C. The stimulation activities in the forward alpha  and beta  reactions were determined in the presence of a constant amount of the wild-type or mutant alpha  subunits and various amounts of the beta  subunit. The assay conditions for the stimulation activities were 50 mM potassium phosphate buffer, pH 7.0, at 37 °C, the same conditions as used for the calorimetry, in the presence and absence of 300 mM KCl. For the stimulation activities in the beta  reaction, the inherent activities of the beta subunit at various concentrations were measured and subtracted.


RESULTS

Stimulation Activities of the Mutant Tryptophan Synthase

To determine the effect of the deletion of the hydrogen bonds on the stimulation activities, the five mutant alpha  subunits having Ala at the intersubunit hydrogen bonding site (N104A, N108A, E134A, E135A, and N157A) were investigated. We also investigated three Asp mutant alpha  subunits (N104D, N108D, and N157A) to determine the effect of the introduction of a charged residue at these positions. All the mutant enzymes should lack one hydrogen bond between the alpha  and beta  subunits. These mutant alpha  subunits retained the inherent activities in the forward alpha  reaction (about 0.1 units/mg), indicating that the substitutions did not affect the monomer structure of the protein.

Fig. 1 shows the stimulation activities in the beta  reaction of several alpha  subunits (wild type, N104D, N104A, N108D, N108A, and E134A) as a function of concentration of the beta  subunit. The other substitutions (E135A, N157D, and N157A) had little effect on the stimulation activities (data not shown). In the absence and presence of 300 mM KCl, the titration curves of N104D, N104A, and N108D alpha  subunits were more gradual than that of the wild-type alpha  subunit, indicating weaker association with the beta  subunit. The curve of N108A alpha  subunit was sharp and similar to that of the wild type, suggesting that the weakened association of N108D alpha  subunit is not due to the lack of a hydrogen bond. In the absence of KCl, the titration curve of E134A alpha  subunit was more gradual than that of the wild type, but the weakened association could be repaired by the addition of 300 mM KCl. In the forward alpha  reaction, similar titration curves were found (data not shown). These results indicate that the hydrogen bonds, except for position 104, are not required for the mutual activation.


Fig. 1. The activation abilities of mutant alpha  subunits in the beta  reaction in the presence (a) and the absence (b) of 300 mM KCl. The stimulation activity was measured in the presence of a constant small amount of the alpha  subunit (2.5 µg) and various amounts of the beta  subunit. (×, wild type alpha  subunit; bullet , N104D; open circle , N104A; black-square, N108D; square , N108A; triangle , E134A) The inherent activity of the beta  subunit at each concentration was subtracted from the data.
[View Larger Version of this Image (19K GIF file)]


Titration Calorimetry of the Interaction of Mutant alpha  Subunits with the beta 2 Subunit

The mutual activation between the alpha  and beta  subunits is induced by the formation of the alpha 2beta 2 complex. To determine the effect of deletion of a hydrogen bond on the alpha /beta subunit interaction, the thermodynamic parameters of association of mutant alpha  subunits with the beta  subunit were examined using ITC. Fig. 2 shows the calorimetric titration curves of the beta  subunit with the wild-type and mutant alpha  subunits at 40 °C, pH 7.0. The titration curves of four mutant alpha  subunits (N104D, N104A, N108D, and E134A) had a more gradual transition than that of the wild-type alpha  subunit, indicating the decreases in the association constants (Ka) of these mutant alpha  subunits. The thermodynamic parameters of association were calculated by the computer program ORIGIN (29, 33), assuming that the two alpha  subunit binding sites of the beta 2 subunit were identical and independent of each other (i.e. single set of identical sites model), and are shown in Table III. The independent or interactive two sites model was also tried, but these models failed to fit the titration curves. All titration curves of the mutant enzymes demonstrate a similar stoichiometry of about 0.7 alpha  chain/beta chain. The departure from the expected 1:1 stoichiometry probably results from the presence of a fraction of inactive beta 2 subunit, which cannot bind the alpha  subunit (29). The thermodynamic parameters of association are independent of the binding number because they are estimated by the molar concentration of the alpha  subunit.


Fig. 2. Titration calorimetry of the association of the wild-type and mutant alpha  subunits with the beta  subunit at 40 °C in 50 mM potassium phosphate buffer, pH 7. The plots are experimental and the solid lines correspond to the best fit curve obtained by least squares methods. a, bullet , N104D and open circle , N104A; b, bullet , N108D and open circle , N108A; c, triangle , E134A and square , E135A; and d, bullet , N157D and open circle , N157A; and in each figure × represents the wild-type alpha  subunit.
[View Larger Version of this Image (31K GIF file)]


Table III.

The thermodynamic parameters of association of the wild and mutant alpha  subunits with the beta  subunit at 40°C in the absence and the presence of 300 mM KCl

All the molar concentrations for each parameter represent the molar concentration of the alpha  chain. Experimental errors for Ka and Delta Ha of the wild type are ± 0.93 × 106 M-1 and ± 8.0 kJ mol-1, respectively, which result in errors in Delta Ga of ± 0.5 kJ mol-1 and -TDelta Sa of ± 7.9 kJ mol-1K-1.
 alpha subunit Ka  Delta Ha  Delta Delta HaM-W  Delta Ga  Delta Delta GaM-W  -TDelta Sa  -TDelta Delta SaM-W

106/M kJ/mol
 -KCla
  Wild type 5.32  -147.3  -40.3 107.0
  N140D 0.30  -97.1 50.2  -32.8 7.5 64.3  -42.7
  N104A 0.35  -119.2 28.1  -33.2 7.1 86.0  -21.0
  N108D 0.29  -105.9 41.4  -32.8 7.5 73.1  -33.9
  N108A 4.40  -139.3 8.0  -39.8 .5 99.5  -7.5
  E134A 0.16  -134.7 12.6  -31.2 9.1 103.5  -3.5
  E135A 1.20  -151.9  -4.6  -36.4 3.9 115.5 8.5
  N157D 1.25  -150.2  -2.9  -36.6 3.7 113.6 6.6
  N157A 2.45  -152.7  -5.4  -38.3 2.0 114.4 7.4
+KCla
  Wild type 11.05  -118.8  -42.2 76.6
  N104D 0.19  -62.3 56.5  -31.6 10.6 30.7  -45.9
  N104A 0.22  -96.2 22.6  -32.0 10.2 64.2  -12.4
  E134A 2.50  -135.1  -16.3  -38.4 3.8 96.7 20.1

a  Thermodynamic parameters were measured in 50 mM potassium phosphate buffer, pH 7.0, in the absence (-KCl) or presence (+KCl) of 300 mM KCl as described under "Experimental Procedures."

The negative Delta Ga values of almost all mutant alpha  subunits were decreased compared with that of the wild-type alpha  subunit. The changes in the Delta Ga varied with the substituted positions. The change in the Delta Ga of the N108A alpha  subunit was the smallest among the mutant enzymes. However, there was a subtle change in the Delta Ha, and it was compensated by a change in the -TDelta Sa. The Delta Ga values for E135A, N157D, and N157A alpha  subunits were affected by 2.0-3.9 kJ mol-1, and these defects resulted from an unfavorable increase in the -TDelta Sa values. Remarkable changes were observed for N104D, N104A, N108D, and E134A alpha  subunits. Their Ka values were decreased by 1 order of magnitude, and the decreases in the negative Delta Ga values were 7.1-9.1 kJ mol-1 compared with that of the wild-type alpha  subunit. The negative Delta Ha values of N104D, N104A, and N108D alpha  subunits were unfavorably decreased compared with that of the wild-type alpha  subunit. The decrease in Delta Ha of the E134A alpha  subunit was relatively smaller than that of the N104D, N104A, or N108D alpha  subunits although the Ka value of the E134A alpha  subunit was similar to those of the others.

Effect of Salt on alpha /beta Subunit Interaction

Electrostatic interactions have been reported to contribute to many intermolecular associations based on x-ray crystallographic analysis (8, 9), thermodynamic analysis (10), and computer simulation (11). To investigate the electrostatic contribution to the interaction between the alpha  and beta  subunits, the effect of salt was examined by ITC. Fig. 3a shows the titration curve for the wild-type alpha  subunit in the absence and presence of 300 mM KCl. The thermodynamic parameters of association are shown in Table III. The Ka values showed that the complex formation of the wild-type tryptophan synthase was increased by the addition of KCl. The negative Delta Ha of the wild-type alpha  subunit in the presence of KCl was lower than that in the absence of additional salt (Table III). However, the favorable increase in the Delta Sa overcame the unfavorable Delta Ha and resulted in a favorable decrease in Delta Ga.


Fig. 3. Titration curves of the association in the presence and absence of additional 300 mM KCl. In each figure, the open and closed symbols represent the experimental values in the absence and presence of KCl, respectively. a, the wild type; b, E134A; c, N104D; and d, N104A.
[View Larger Version of this Image (26K GIF file)]


The effects of the substitutions at the interface on the electrostatic properties of association were then examined by ITC in the presence of 300 mM KCl for the mutant alpha  subunits at positions 104 (N104D and N104A) and 134 (E134A), which have decreased Ka values in the absence of salt (Fig. 3 and Table III). The negative Delta Ha values of N104D and N104A alpha  subunits were decreased by the addition of KCl in a way similar to that of the wild type, whereas the Ka values were slightly decreased. In the case of E134A alpha  subunit, the Ka value was greatly increased by the addition of KCl. The negative Delta Ha value of E134A alpha  subunit in the presence of KCl was larger than that of the wild type, in contrast with that in the absence of KCl.


DISCUSSION

The Electrostatic Repulsion between the alpha  and beta  Subunits

Recently, it has been reported that monovalent cations change the conformation of the tryptophan synthase complex and activate it (36-41). The preferential binding of a cation (such as Na+ or K+) to a specific site in the beta  subunit has high affinity, and the site is saturated at 20 mM cation concentration (38). In this study, because we treated the sample in 50 mM potassium phosphate buffer, the concentration of K+ might be enough to saturate the cation binding sites of the tryptophan synthase complex even in the absence of additional KCl. Therefore, the increase in the Ka value of the wild type on the addition of salt observed here (Table III) should not result from a conformational change due to the effect of the cation. Alternatively, the present results suggest the presence of electrostatic repulsion between the alpha  and beta  subunits. The addition of salt might diminish this repulsion and cause an increase in the affinity. Similar stabilizing effects of salts on the oligomer structure have been reported for some proteins (42-47).

We examined whether there is an electrostatic repulsion between the alpha  and beta  subunits due to the complex structure (14). The charges on the subunit interface were calculated using the Insight II-Delphi program (48), and it was found that both subunits were relatively negatively charged at the subunit interface. A qualitative estimation using the Insight II-Docking program showed that the electrostatic effect acts unfavorably on the association between the alpha  and beta  subunits, whereas the van der Waals interaction is favorable.

In the cases of the mutant alpha  subunits at position 104, the presence of salt to shield the electrostatic repulsion did not improve the affinity for the beta  subunit (Table III). In contrast, the weakened affinity of E134A alpha  subunit was repaired by the presence of salt and was close to that of the wild type (Table III).

The Role of the Hydrogen Bond between the alpha  and beta  Subunits in the Association

Shirley et al. (49) have reported that the contribution of a hydrogen bond to the Gibbs energy of unfolding is 5.4 ± 2.5 kJ mol-1 using mutant proteins. Fersht (50) has also evaluated it to be 2.1-7.5 kJ mol-1. In the present study, the average Delta Delta Ga of eight mutant proteins was 5.2 kJ mol-1 (Table III), and it fell within the ranges reported previously. However, the other thermodynamic parameters, Delta Ha and Delta Sa, varied with the substituted positions (Table III). This indicates that hydrogen bonds play different roles in the association depending on their positions (see below).

The Effects of the Deletions of the Hydrogen Bonds at Positions 135 and 157 Are Localized in Substituted Positions

The effects on the thermodynamic parameters of association of the mutant alpha  subunits at positions 135 and 157 (E135A, N157D, and N157A) could be considered to be due to the deletion of only 1 mol of hydrogen bond (Table III). In each case, the reasonable decrease in the negative Delta Ga value corresponds to the deletion of one hydrogen bond as reported previously (49, 50). Although the changes in the Delta Ha and -TDelta Sa of these mutant alpha  subunits were mostly within experimental error, there is a tendency for the unfavorable Delta Ga to be accompanied by an unfavorable change in the positive -TDelta Sa and partially compensated by a favorable change in the negative Delta Ha. These results agree with the conclusion of Privalov and Makhataze (51) that the hydrogen bonds between the polar groups are stabilized entropically. Connelly et al. (52) have studied the effect of the substitution of a hydrogen bonding residue on the protein-ligand binding using ITC and have found similar results. Because the changes observed for the substitutions at positions 135 and 157 correspond to the deletion of one hydrogen bond as described above, the effects due to the mutations must be localized in the substituted region.

The substitutions at positions 135 and 157 (E135A, N157D, and N157A) had no effect on the stimulation activity, and the changes in the thermodynamic parameters are in the ranges of the contribution from one hydrogen bond as described above (Table III), suggesting that these substitutions do not result in a defect in the intact conformation of the complex. These residues are located on the edge of the interface with the beta  subunit, far from the intersubunit tunnel, and the partner residues of the beta  subunit (Met-15 and Tyr-181 of the beta  subunit) are not highly conserved. Therefore, it can be concluded that these hydrogen bonds are not very important for the mutual activation.

The Hydrogen Bond at alpha  Subunit Asn-104 Is Especially Important for Association With the beta  Subunit

Due to the substitutions at position 104 (N104D and N104A), the stimulation activities were decreased (Fig. 1), and the decreased Ka values of these mutant alpha  subunits (Table III) might be related to the reduced stimulation activities. We have reported that the complex formation couples with the folding and the rearrangement events in the alpha  or/and beta  subunits, and the folding might occur not only at the subunit contact surface but also at other parts in the molecules (53). The unfavorable decreased Delta Ha values of the mutant enzymes at position 104, compared with that of the wild type (Table III), might suggest that the mutant proteins cannot fold into an intact conformation upon complex formation. Therefore, the stimulation activities of their resulting complexes were lower than that of the wild-type complex. Furthermore, the affinities of N104D and N104A with the beta  subunit could not be improved by the addition of salt (to seal electorostatic repulsion) (Table III), suggesting that these mutant alpha  subunits could not associate with the beta  subunit to fold the similar conformation to the alpha 2beta 2 complex of the wild-type protein.

Position 104 is apparently located far from the intersubunit tunnel. However, in the alpha 2beta 2 complex form, the Asn residue at position 104 of the alpha  subunit contacts the residue at position 292 of the beta  subunit in a sharp turn in the trypsin-sensitive "hinge" region (Table I). This hinge region is included in the long stretch from 260-310 (with a loop conformation), which forms one side of the intersubunit tunnel of the beta  subunit (14). The substitution (28, 54) or proteolysis (55-57) in the "hinge" region of the beta  subunit reduces the stimulation activity, the substrate affinity, and the association between the alpha  and beta  subunits. Thus, this region is believed to play a critical role in the conformational change upon the subunit association (1). The present results revealed the special role of 104; the hydrogen bonding at position 104 is necessary to maintain the hinge of the beta  subunit in an intact form, and a substitution at this position causes a significant defect in the stimulation activity. Regarding other mutations at position 104 of the alpha  subunit (58), it has also been reported that the N104S mutant complex decreases the activity in the beta and alpha beta reactions. Asn-104 and its hydrogen bonding partner Gly-292 of the beta  subunit are completely conserved residues among the tryptophan synthase from 16 different organisms (Table I). It can be concluded that the hydrogen bond at Asn-104 of the alpha  subunit with Gly-292 of the beta  subunit is critical to induce a perfect complement association.

The Other Hydrogen Bonds at Positions 108 and 134

Substitutions at positions 108 and 134 might also involve conformational changes because the changes in the thermodynamic parameters could not be interpreted based on the deletion of only one hydrogen bond as described above.

At position 108, Asp mutation caused decreases in the Ka value by 1 order of magnitude and in a negative Delta Ha (Table III). Asn-108 of the alpha  subunit contacts the residue at position 290, which is Ala in S. typhimurium and is replaced by Glu in E. coli in the hinge region of the beta  subunit. The extreme substitution with a bulkier and more polar residue at position 108 (N108D) might prevent intact conformation of the complex, but hydrogen bonding at position 108 is not required for complex formation as shown in results of N108A. Another mutation at this position (N108S) also does not affect the activity of the complex (58).

The Delta Ga value of E134A alpha  subunit in the absence of KCl was remarkably affected by the substitution, although the change in the Delta Ha was relatively small (Table III). In the case of E134A alpha  subunit, the conformational change due to the complex formation might correspond to that of the wild type, despite its low Ka value. This is also supported by the results in which the Delta Ga was improved by the addition of the KCl, and the negative Delta Delta HaM-W and positive -TDelta Delta SaM-W values could be explained by the deletion of one hydrogen bond as described for the substitutions at positions 135 and 157. Glu-134 is 40% exposed to the solvent even in the complex form (Table I). The percentage of conservation of Glu-134 is 75%, but its hydrogen bonding partner Gln-19 of the beta  subunit is not highly conserved (Table I). Present results indicate that this residue is not required to induce complement association although the substitution of Glu-134 affected the affinity for the beta  subunit.

Conclusions

The hydrogen bonding residues located in the alpha /beta subunit interface play various roles in the association and mutual activation, depending on their positions. The deletion of the hydrogen bond at Asn-104 affected the thermodynamic parameters of association with the beta  subunit and remarkably reduced the stimulation activity in the beta  reaction. It can be concluded that the hydrogen bond at Asn-104 of the alpha  subunit is especially important for intact association with the beta  subunit and mutual activation of the complex.


FOOTNOTES

*   This work was supported in part by fellowships from the Japan Society for the Promotion of Science for Japanese Junior Scientists (to K. H.) and by a grant-in-aid for special project research from the Ministry of Education, Science, and Culture of Japan (to K. Y.). The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Dagger    Present address: Tokyo University of Pharmacy and Life Science, Dept. of Molecular Biology, 1432 Horinouchi, Hachioji, Tokyo 192-03, Japan.
§   To whom correspondence should be addressed: Institute for Protein Research, Osaka University, 3-2 Yamadaoka, Suita, Osaka 565, Japan. Tel.: 81-6-879-8615; Fax: 81-6-879-8616; E-mail: yutani{at}protein.osaka-u.ac.jp.
1    The abbreviation used is: ITC, isothermal titration calorimetry.

Acknowledgment

DNA sequence analysis was carried out by an automated DNA sequencer (Applied Biosystems, Inc.) at the Research Center for Protein Engineering, Institute for Protein Research, Osaka University.


REFERENCES

  1. Miles, E. W. (1995) in Subcellular Biochemistry, Proteins: Structure, Function and Protein Engineering (Biswas, B. B., and Roy, S., eds), pp. 207-254, Plenum Press, New York
  2. Swift, S., and Stewart, G. S. (1991) Biotechnol. Genet. Eng. Rev. 9, 229-294 [Medline] [Order article via Infotrieve]
  3. Miles, E. W. (1991) Adv. Enzymol. Relat. Areas Mol. Biol. 64, 93-172 [Medline] [Order article via Infotrieve]
  4. Miles, E. W., Bauerle, R., and Ahmed, S. A. (1987) Methods Enzymol. 142, 398-414 [Medline] [Order article via Infotrieve]
  5. Yanofsky, C., and Crawford, I. P. (1972) in The Enzymes (Boyer, P. D., ed), 3rd Ed., pp. 1-31, Academic Press, New York
  6. Argos, P. (1988) Protein Eng. 2, 101-113 [Abstract]
  7. Janin, J., Miller, S., and Chothia, C. (1988) J. Mol. Biol. 204, 155-164 [Medline] [Order article via Infotrieve]
  8. Miller, S. (1989) Protein Eng. 3, 77-83 [Abstract]
  9. Janin, J., and Chothia, C. (1990) J. Biol. Chem. 265, 16027-16030 [Free Full Text]
  10. Ross, P. D., and Subramanian, S. (1981) Biochemistry 20, 3096-3102 [Medline] [Order article via Infotrieve]
  11. Warwicker, J. (1989) J. Mol. Biol. 206, 381-395 [Medline] [Order article via Infotrieve]
  12. Nichols, B. P., Blumenberg, M., and Yanofsky, C. (1981) Nucleic Acids Res. 9, 1743-1755 [Abstract]
  13. Miles, E. W. (1979) Adv. Enzymol. Relat. Areas Mol. Biol. 49, 127-186 [Medline] [Order article via Infotrieve]
  14. Hyde, C. C., Ahmed, S. A., Padlan, E. A., Miles, E. W., and Davies, D. R. (1988) J. Biol. Chem. 263, 17857-17871 [Abstract/Free Full Text]
  15. Crawford, I. P., Han, C. Y., and Silverman, M. (1991) DNA Sequence 1, 189-196 [Medline] [Order article via Infotrieve]
  16. Crawford, I. P., and Eberly, L. (1989) Biochimie (Paris) 71, 521-531 [Medline] [Order article via Infotrieve]
  17. Hadero, A., and Crawford, I. P. (1986) Mol. Biol. Evol. 3, 191-204 [Abstract]
  18. Ross, C. M., and Winkler, M. E. (1988) J. Bacteriol. 170, 753-768
  19. Matsui, K., Sano, K., and Ohtsubo, E. (1986) Nucleic Acids Res. 145, 10113-10114
  20. Henner, D. J., Band, L., and Shimotsu, H. (1985) Gene (Amst.) 4, 169-177
  21. Natori, Y., Kano, Y., and Imamoto, F. (1990) J. Biochem. (Tokyo) 107, 248-255 [Abstract]
  22. Zalkin, H., and Yanofsky, C. (1982) J. Biol. Chem. 257, 1491-1500 [Abstract/Free Full Text]
  23. Koyama, Y., and Furukawa, K. (1990) J. Bacteriol. 172, 3490-3495 [Medline] [Order article via Infotrieve]
  24. Ishiwata, K., Yoshino, S., Iwamori, S., Suzuki, T., and Makiguchi, N. (1989) Agric. Biol. Chem. 53, 2941-2948
  25. Lam, W. L., Cohen, A., Tsouluhas, D., and Doolittle, W. F. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 6614-6618 [Abstract]
  26. Sibold, L., and Henriquet, M. (1988) Mol. & Gen. Genet. 214, 439-450 [Medline] [Order article via Infotrieve]
  27. Yutani, K., Ogasahara, K., Tsujita, T., and Sugino, Y. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 4441-4444 [Abstract]
  28. Zhao, G.-P., and Somerville, R. L. (1992) J. Biol. Chem. 267, 526-541 [Abstract/Free Full Text]
  29. Ogasahara, K., Hiraga, K., Ito, W., Miles, E. W., and Yutani, K. (1992) J. Biol. Chem. 267, 5222-5228 [Abstract/Free Full Text]
  30. Adachi, O., Kohn, L. D., and Miles, E. W. (1974) J. Biol. Chem. 249, 7756-7763 [Abstract/Free Full Text]
  31. Ogasahara, K., Yutani, K., Suzuki, M., Sugino, Y., Nakanishi, M., and Tsuboi, M. (1980) J. Biochem. (Tokyo) 88, 1733-1738 [Abstract]
  32. Hathaway, G. M., and Crawford, I. P. (1970) Biochemistry 9, 1801-1808 [Medline] [Order article via Infotrieve]
  33. Wiseman, T., Williston, S., Brandts, J. F., and Lin, L.-N. (1989) Anal. Biochem. 179, 131-137 [Medline] [Order article via Infotrieve]
  34. Creighton, T. E. (1970) Eur. J. Biochem. 13, 1-10 [Medline] [Order article via Infotrieve]
  35. Higgins, W., Fairwell, T., and Miles, E. W. (1979) Biochemistry 18, 4827-4835 [Medline] [Order article via Infotrieve]
  36. Dunn, M. F., Brzovic', P. S., Leja, C. A., Pan, P., and Woehl, E. U. (1994) in Biochemistry of Vitamin B6 and PQQ (Marino, G., Sannia, G., and Bossa, F., eds), pp. 119-124, Birkhauser Verlag, Basel, Switzerland
  37. Peracchi, A., Mozzarelli, A., and Rossi, G. L. (1994) in Biochemistry of Vitamin B6 and PQQ (Marino, G., Sannia, G., and Bossa, F., eds), pp. 125-129, Birkhauser Verlag, Basel
  38. Peracchi, A., Mozzarelli, A., and Rossi, G. L. (1995) Biochemistry 34, 9459-9465 [Medline] [Order article via Infotrieve]
  39. Ruvinov, S. B., Ahmed, S. A., McPhie, P., and Miles, E. W. (1995) J. Biol. Chem. 270, 17333-17338 [Abstract/Free Full Text]
  40. Woehl, E. U., and Dunn, M. F. (1995) Biochemistry 34, 9466-9476 [Medline] [Order article via Infotrieve]
  41. Rhee, S., Parris, K. D., Ahmed, S. A., Miles, E. W., and Davies, D. R. (1996) Biochemistry 35, 4211-4222 [CrossRef][Medline] [Order article via Infotrieve]
  42. Bonafe, C. F. S., Villas-Boas, M., Suarez, M. C., and Silva, J. L. (1991) J. Biol. Chem. 266, 13210-13216 [Abstract/Free Full Text]
  43. Gueguen, J., Chevalier, M., Barbot, J., and Schaeffer, F. (1988) J. Sci. Food Agric. 44, 167-182
  44. Koshiyama, I. (1971) Agric. Biol. Chem. 35, 385-392
  45. Pedrosa, C., and Ferreira, S. T. (1994) Biochemistry 33, 4046-4055 [Medline] [Order article via Infotrieve]
  46. Ruan, K., and Weber, G. (1988) Biochemistry 27, 3295-3301 [Medline] [Order article via Infotrieve]
  47. Schwenke, K. D., and Schultzmand, L. K. J. (1975) Die Nabrung 19, 425-432
  48. Honig, B., Sharp, K., and Yang, A.-S. (1993) J. Phys. Chem. 97, 1101 -
  49. Shirley, B. A., Stanssens, P., Hahn, U., and Pace, C. N. (1992) Biochemistry 31, 725-732 [Medline] [Order article via Infotrieve]
  50. Fresht, A. R. (1987) Trends Biochem. Sci. 12, 301-304 [CrossRef]
  51. Privalov, P. L., and Makhatadze, G. I. (1993) J. Mol. Biol. 232, 660-679 [CrossRef][Medline] [Order article via Infotrieve]
  52. Connelly, P. R., Aldape, R. A., Bruzzese, F. J., Chambers, S. P., Fitzgibbon, M. J., Fleming, M. A., Itoh, S., Livingston, D. J., Navia, M. A., Thomson, J. A., and Wilson, K. P. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 1964-1968 [Abstract]
  53. Hiraga, K., and Yutani, K. (1996) Eur. J. Biochem. 240, 63-70 [Abstract]
  54. Zhao, G.-P., and Somerville, R. L. (1993) J. Biol. Chem. 268, 14921-14931 [Abstract/Free Full Text]
  55. Högberg-Raibaud, A., and Goldberg, M. E. (1977) Biochemistry 16, 4014-4020 [Medline] [Order article via Infotrieve]
  56. Kaufmann, M., Schwartz, T., Jaenicke, R., Schnackerz, K. D., Mdyer, H. E., and Bartholmes, P. (1991) Biochemistry 30, 4173-4179 [Medline] [Order article via Infotrieve]
  57. Linkens, H. J., Djavadi-Ohaniance, L., and Goldberg, M. E. (1993) FEBS Lett. 320, 224-228 [CrossRef][Medline] [Order article via Infotrieve]
  58. Lim, W. K., Sarkar, S. K., and Hardman, J. K. (1991) J. Biol. Chem. 266, 20205-20212 [Abstract/Free Full Text]

©1997 by The American Society for Biochemistry and Molecular Biology, Inc.