Correspondence to M. Killenberg-Jabs, Martin-Luther-Universität Halle-Wittenberg, Kurt-Mothes-Str. 3, 06120 Halle, Germany. Fax: + 345 5527011, Tel.: +345 5524887, E-mail: firstname.lastname@example.org
Homomeric pyruvate decarboxylase (E.C 18.104.22.168) from yeast consists of dimers and tetramers under physiological conditions, a Kd value of 8.1 µm was determined by analytical ultracentrifugation. Dimers and monomers of the enzyme could be populated by equilibrium denaturation using urea as denaturant at defined concentrations and monitored by a combination of optical (fluorescence and circular dichroism) and hydrodynamic methods (analytical ultracentrifugation). Dimers occur after treatment with 0.5 m urea, monomers with 2.0 m urea independent of the protein concentration. The structured monomers are catalytically inactive. At even higher denaturant concentrations (6 m urea) the monomers unfold. The contact sites of two monomers in forming a dimer as the smallest enzymatically active unit are mainly determined by aromatic amino acids. Their interactions have been quantified both by structure-theoretical calculations on the basis of the X-ray crystallography structure, and experimentally by binding of the fluorescent dye bis-ANS. The contact sites of two dimers in tetramer formation, however, are mainly determined by electrostatic interactions. Homomeric pyruvate decarboxylase (PDC) is activated by its substrate pyruvate. There was no difference in the steady-state activity (specific activity) between dimers and tetramers. The activation kinetics of the two oligomeric states, however, revealed differences in the dissociation constant of the regulatory substrate (Ka) by one order of magnitude. The tetramer formation is related to structural consequences of the interaction transfer in the activation process causing an improved substrate utilization.
4,4′-dianilino-1,1′-binaphthyl-5,5′-disulfonic acid, dipotassium salt
substrate concentration at half maximum rate of the enzymatic activity
Pyruvate decarboxylase (PDC, E.C. 22.214.171.124) is a thiamin diphosphate (ThDP) dependent enzyme within the glycolytic pathway in fermenting cells. It catalyzes the nonoxidative conversion of pyruvate to acetaldehyde and carbon dioxide. The yeast pyruvate decarboxylase is an allosterically regulated enzyme that is activated by its substrate pyruvate. This process is characterized by a sigmoidal v/S plot [1,2]. Using rapid kinetic methods an activation phase is observed and the rate constants of the allosteric activation have been determined .
The structure of yeast PDC was resolved to 2.3 Å resolution . The enzyme was described as a tetramer with a molecular mass of 240 000 Da [5,6]. The tetrameric molecule is a dimer of dimers that binds four molecules ThDP and four magnesium ions . The cofactor ThDP is noncovalently bound in the ‘V’ conformation  at the interface between two monomers involving the α and γ domains . Under the chosen experimental conditions (pH 6.0) the cofactors are tightly bound to the protein. At pH-values above pH 7.0 the cofactors dissociate reversibly . By X-ray small angle scattering a pH-dependent dissociation/association behaviour of the protein could be observed. So far, the catalytic activity of the enzyme is related to the tetrameric species only [9,10].
In this work we address the question to the population and characterization of different oligomeric states of pyruvate decarboxylase under native and equilibrium denaturation conditions by combining experimental studies with structure theoretical methods to identify, whether the enzymatic activity is related to the fraction of tetramers only, or the dimers are the smallest catalytically active unit.
MATERIALS and METHODS
Preparation and purification of homomeric pyruvate decarboxylase
The production of the enzyme and its purification was performed as described  with slight modifications. The clear supernatant after the cell rupture, ammonium sulfate precipitation and dialysis was supplied to ionic exchange chromatography on Fractogel® EMD TMAE-650 S (Merck) equilibrated with 20 mm bis-Tris buffer, pH 6.0, 1 mm dithioerythritol, 0.1 mm ThDP. The protein was eluted by a discontinous ammonium sulfate gradient. Active fractions were collected, concentrated and supplied on a gel filtration column (Superdex™ 200, Pharmacia) equilibrated with 50 mm sodium phosphate buffer, pH 6.0. Active fractions were concentrated to final concentrations above 5 mg·mL−1 on Filtron 10 kDa, aliquoted and stored at −80 °C. The final yield was about 58% (average from 15 preparations) with a specific activity of 74 U·mg−1. In the course of the preparation pyruvate decarboxylase was checked by SDS/PAGE and the enzymatic activity determined according to Holzer et al. . All concentrations of the enzyme are related to the molecular mass of the monomer (60 000 Da). The preparation of the apoenzyme was carried out according to Killenberg-Jabs et al. .
The enzymatic activity was determined by a coupled optical test using alcohol dehydrogenase and NADH .
The allosteric substrate activation kinetics of homomeric pyruvate decarboxylase were determined on an Applied Photophysics BioSequential DX.18MV stopped-flow spectrometer (Leatherhead, UK) at protein concentrations between 1 µg·mL−1 and 150 µg·mL−1. Data evaluation was performed according to Hübner et al.  using Eqn 1 derived from a minimal model describing the substrate activation in a two step mechanism, in which an initial, reversible complex SE (with a dissociation constant Ka) isomerizes in a rate limiting step to an active complex (SEact) formed by the rate constants k+ and k− according to the following model (Scheme 1).
The pseudo-first-order rate constant of activation by pyruvate kact is defined as:
Absorbance spectra of the protein and activity determinations were carried out on an UV/VIS spectrophotometer DU 70 (Beckman) or 941 (Kontron).
The intrinsic fluorescence of PDC (70 µg·mL−1; ≈ 90% dimer) was recorded from 295 nm to 450 nm on excitation at 280 nm at different denaturant concentrations and at 20 °C using a F-4500 Hitachi fluorimeter. The bandpass for both the excitation and the emission monochromators was set at 5 nm, the scan speed at 240 nm·min−1 and the response time automatically adapted by the device. All fluorescence spectra were corrected according to the suppliers recommendation using Rhodamine B as standard. The relative fluorescence intensity at 320 nm was used for monitoring urea unfolding.
The dipotassium salt of the fluorescent dye bis-ANS was dissolved in water giving a stock concentration of 1.48 mm. Fluorescence titration was performed with 1.3 µm pyruvate decarboxylase in 10 mm sodium phosphate buffer, pH 6.0 at 20 °C. The same protein concentration and experimental conditions were used for monitoring the bis-ANS fluorescence at different urea concentrations. The fluorescent dye bis-ANS was excited at 382 nm for direct excitation. In energy transfer experiments, the sensitized fluorescence was recorded from 295 nm to 540 nm on excitation at 280 nm. The fluorescence intensity of bis-ANS in aqueous solution is very low at 550 nm, but increases remarkably and shifts to 500 nm on binding both to the holoenzyme and apoenzyme of homomeric pyruvate decarboxylase, respectively.
Far UV circular dichroism spectra of the protein at a concentration of 70 µg·mL−1 were recorded on a Jasco J710 spectropolarimeter using cuvettes of 0.1 cm optical pathlength. Spectra were acquired at a scan speed of 20 or 50 nm·min−1, a slit width of 1 or 2 nm and a response time of 1 or 4 s. The ellipticity at 222 nm was used for monitoring urea unfolding.
Analytical ultracentrifugation measurements were performed on a XL-A (Beckman, Palo Alto, CA, USA) ultracentrifuge using an AN 50Ti rotor and double sector cells. Sedimentation equilibrium (10 000 r.p.m.) and velocity experiments (30 000 r.p.m.) were recorded using the scanning absorbance optical system at 280 nm (aromatic chromophores). Samples were run at 20 °C at different protein and denaturant concentrations. The apparent values of the molecular mass and the sedimentation velocity were calculated using the software provided by Beckman Instruments. For correction of the nonideality of the protein solution at high protein concentrations the dependence of the apparent s-value on the protein concentration of PDC from Zymomonas mobilis, which is tetrameric over the whole concentration range, was measured. From the corrected s-values the dissociation constant of the tetramer was calculated as described [13,14].
Structure theoretical studies
The X-ray crystallography structure of pyruvate decarboxylase from yeast (PDB accession no. 1PVD) at 2.3 Å resolution was the basis for studying the contact region between monomers in forming dimers and between dimers in forming tetramers .
The amino acid distribution in the contact region was calculated using our own program pdbsnoop. This search comprises van der Waals interactions of residues on different subunits within distances less than 4.5 Å between any two atoms of these residues. Bridging water molecules forming a hydrogen bond to one subunit and an additional hydrogen bond to the other subunit were taken into account.
The residue accessible surface area of the aromatic amino acids in the respective oligomeric states was determined using the program stride. The percentages of accessibility of tryptophans, tyrosines and phenylalanines were calculated according to Chothia .
RESULTS and DISCUSSION
Association state of PDC
Equilibrium denaturation. The crystal structure of PDC reveals its tetrameric association state. However, the oligomerization state of the homomeric enzyme is unknown under conditions used for the enzymatic characterization, especially at low protein concentrations. An equilibrium denaturation by urea was used to populate different oligomeric states of the protein. As shown in Fig. 1, well separated transitions can be measured by intrinsic fluorescence, far UV circular dichroism and by determination of the enzymatic activity. First, a transition between 0.5 m and 1.5 m urea is observed. The intensity of the intrinsic fluorescence increases and its maximum shifts to 342 nm indicative for significant changes in the structure or association state of homomeric PDC. However, the maximum position of the intrinsic fluorescence and the far UV circular dichroism spectrum clearly prove that PDC is still a highly structured protein even at 2–3 m urea. In the second transition from 3 to 6 m urea the intensity of the intrinsic fluorescence decreases (Fig. 1A) and the maximum shifts to 350 nm (Fig. 1C). The transition has a midpoint of about 4 m urea. The far UV circular dichroism spectrum approaches to random coil with a decrease in the mean residual ellipticity (Fig. 1D).
In order to ensure a saturation of the holoenzyme the cofactor experiments were performed in the presence of 100 µm ThDP. A higher concentration of the cofactor is not recommended because of its optical properties. The transitions show the same picture as measured for the holoenzyme without additional ThDP (data not shown).
Analytical ultracentrifugation. In order to analyze whether one of the transitions observed by spectroscopic techniques is caused by a dissociation of tetramers or dimers, we used analytical ultracentrifugation. Surprisingly, already in the absence of urea a tetramer/dimer equilibrium has been observed. Quantitative analyses of the dependence of the sedimentation velocity of PDC on the protein concentration (Fig. 2) yielded a dissociation constant of Kd = 8.1 µm of the tetramer. Consequently, at protein concentrations from 50 µg·mL−1 to 70 µg·mL−1 used to monitor the equilibrium transitions, about 90% of PDC is dimeric, resulting in an apparent molecular mass slightly higher than calculated for the dimer (Table 1). Consistently, the apparent molecular mass increases with increasing protein concentration, e.g. at 0.2 mg·mL−1≈ 25% of PDC is associated to tetramers. In the presence of 0.5 m urea, a value directly before the first transition of the equilibrium denaturation curve, PDC proved to be a dimer up to concentrations of about 1 mg·mL−1 (Table 1). However, in the presence of 3.5 m urea, which is in the plateau region after the first transition of the equlibrium denaturation curve, PDC dissociated to folded monomers. Thus, the first transition is related to the dissociation of the dimers to monomers. The increase in the intrinsic fluorescence is caused by a dequenching of the aromatic amino acids in the contact region of the subunits and the accompanying loss of the cofactor [17,18]. Finally, the second transition reveals the unfolding of the monomers.
Table 1. Analytical ultracentrifugation sedimentation equilibrium measurements of recombinant pyruvate decarboxylase in 20 mm sodium phosphate buffer, pH 6.0 at 20 °C.
Protein concn (mg·mL−1)
Urea concn (m)
Apparent molecular mass (kDa)
Calculated molecular mass (kDa)
a Experiments were performed in the presence of 200 mm MgSO4. b Data were evaluated considering the density of the solvent.
Dimeric and tetrameric contact sites. The change in the intrinsic fluorescence of PDC upon dissociation of the dimer has already indicated the involvement of aromatic residues in the dimerization process. Thereby, the intrinsic fluorescence represents the optical signal with the biggest change. Hence, it was of interest to us to qualify the interactions in the contact region of the monomers and the dimers. The relation of the different transitions described above to changes of the quaternary structure of PDC was confirmed by structure theoretical studies on the basis of the X-ray crystallography data. The amino acid distributions in the contact region of the monomers in forming dimers and in that of the dimers in forming the tetrameric structure were analyzed. The contacts of the dimers are mostly determined by interactions between polar amino acids (Glu, Asn), that of the monomers by interactions, preferably between aromatic (Phe, Tyr, Trp) and hydrophobic (Leu) amino acids (Fig. 3). The strong interactions of the aromatic amino acids in the monomer contact region involve three tryptophans, four tyrosines and two phenylalanines. The biggest change, i.e. from a more exposed to a buried state, is calculated in the dimer formation (Table 2).
Table 2. Correlation between the accessibility of aromatic residues calculated from the crystal structure and measured by dye bound fluorescence studies. Aromatic amino-acid residues were selected on the basis of the highest contribution in the interaction differences of the oligomeric states. Energy transfer measurements were performed on excitation of the intrinsic fluorophores at 280 nm as described in Materials and methods. Integral intrinsic fluorescence was calculated from the emission spectrum recorded between 320 nm and 380 nm. Integral bis-ANS fluorescence was calculated from the emission spectrum recorded between 320 nm and 540 nm.
Difference in solvent area, Δ(Å2)
Total surface area
Relative change of
Relative change of
Total surface area
integral intrinsic fluorescence
integral bis-ANS fluorescence
The change of the solvent accessible surface area of the aromatic amino acids in the course of the assembly process from monomer via dimer to tetramer derived from structure theoretical analyses was experimentally confirmed by binding studies of the fluorescent dye bis-ANS to the respective populated oligomeric species. The amino acid interactions between the monomers in forming dimers were studied by bis-ANS binding to the respective oligomeric state of PDC populated by urea. The fluorescent dye changes its fluorescence properties on binding to hydrophobic surfaces and can be excited either directly or indirectly via energy transfer from the intrinsic fluorophores . Most of the aromatic amino acids of PDC are located in the contact area between the monomers being suitable as spectroscopic probes of the association/dissociation process. Two transitions could be detected in the course of the urea denaturation of the dye-labeled protein (data not shown). The first transition corresponds to the formation of the monomers and an increase in the fluorescence intensity of the bound dye is monitored on direct excitation. The second transition corresponds to the unfolding of the monomers resulting in a decrease in the dye fluorescence. A similar pattern could be obtained by measuring the denaturant dependence of the energy transfer from the excited intrinsic fluorophores of the protein to the noncovalently bound bis-ANS (Fig. 4). The higher the transfer efficiency, the higher the sensitized fluorescence of the bound dye, and the lower the residual intrinsic fluorescence. The highest energy transfer efficiency is measured at about 2 m urea, a denaturant concentration where monomers are populated. These experimental findings correlate with the increase in the solvent accessible surface area calculated for those aromatic amino acids, which are involved in the intermolecular interactions of the monomers in the dimer formation. A very good correlation was found between the calculated solvent accessible surface area and the fluorescence change in bis-ANS binding studies on the different oligomeric species populated at the respective denaturant concentrations (Table 2).
Enzymatic activity of dimers and tetramers. Certain conditions were defined by investigating the association/dissociation behaviour of PDC to populate the dimeric and tetrameric state of this protein, respectively. Thus, the influence of the oligomerization state on the enzymatic activity could be analyzed. Both dimers and tetramers are enzymatically active, whereas monomers are not.
However, steady-state measurements indicated differences between the two enzymatically active oligomeric states (Table 3, Fig. 5). An increased S0.5-value is observed in the presence of 0.5 m urea or at protein concentrations of less than 70 µg·mL−1, where dimers are populated. Dimers and tetramers exhibit the same specific enzymatic activity.
Table 3. Properties of the active species (dimer and tetramer) of recombinant pyruvate decarboxylase. All activity measurements were performed by the coupled optical test as described in Materials and methods using 0.1 m sodium citrate buffer, pH 6.0. The kinetic constants were determined according to . The relationship between the kinetic parameters Ka, k1, k1 and S0.5 is described in [22,23]. Tetramer measurements were performed directly after dilution of the protein stock solution (0.1 mm) to 1.7 µm final concentration and several seconds incubation time (dilution 1 : 11). Dimer protein (12 nm) was incubated at 4 °C for 2 h (dilution 1 : 2). Dimer by urea protein (0.3 µm) was incubated in 0.5 m urea at 4 °C for 4 h (dilution 1 :11) The error of the determined kinetic parameters is about 10%. ND, not determined.·
Constants of substrate activation
Dimer by urea
A substrate activation according to Scheme 1 has been found for the PDC, which can be detected by observing typical lag-phases in the progress curves of the product formation (stopped-flow kinetics) . The dependence of the observed pseudo-first-order rate constant of the allosteric substrate activation on the substrate concentration differs significantly for the dimers and tetramers (Fig. 5). Evaluation of the kinetic traces shows that mainly the dissociation constant of the binding of pyruvate to the regulatory site is influenced by the oligomerization (Table 3).
The activation kinetics of the dimers populated either at 0.5 m urea or at a defined protein concentration are the same, which means that these proteins are functionally equivalent. The small increase in the specific activity at 0.5 m urea may be related to a higher flexibility of the native protein. The decrease in the activity at urea concentrations higher than 0.5 m correlates with the formation of monomers, which are enzymatically inactive. The activation kinetics of the tetramers populated at high protein concentration yielded the rate constants k+ and k− as determined for the dimer, but the Ka is smaller by about one order of magnitude (Table 3). This kinetic parameter of the activation process is indicative of the equilibrium between the regulatory substrate molecule(s) and the regulatory site(s) of the respective oligomer. On the other hand, the steady-state activity was found to be the same for both the dimer and the tetramer, representing the same catalytic power at the active site being not influenced by the (initial) oligomeric state. The experimental results are related to the specific properties of the contact sites of the monomeric unit within the respective oligomeric states. Structure-theoretical studies revealed mainly interactions of electrostatic nature between dimers in the tetramer formation of PDC. As a consequence, the regulatory site may undergo a conformational change influencing the information transfer to the active site during the activation process. Hence, the unique orientation of the α, β and γ domain within the two subunits realized in the dimer is required not only for the formation of the active site, but also for the substrate activation of PDC [20,21].
Taken together, we could demonstrate an association/dissociation equilibrium of homomeric PDC with functional importance for the enzyme. Small angle X-ray scattering data  and the crystal structure  of pyruvate decarboxylase revealed the enzyme as a tetrameric protein. However, these two methods require high protein concentrations. In the concentration range relevant for a functional characterization, the dimeric state predominates. The association/dissociation characteristics are not specific for the homomeric PDC (α4) used in this work, but could also be observed for the heteromeric enzyme (α2β2; unpublished data).
The work was supported by grants from the Deutsche Forschungsgemeinschaft and the Fond der Chemischen Industrie.
Enzyme: pyruvate decarboxylase (E.C 126.96.36.199).
Note: this paper is dedicated to Prof. Dr R. Jaenicke on the occasion of his 70th birthday.