aminoacidsallostericallyregulatethethiamine …decarboxylation of benzoylformate to benzaldehyde,...

Amino Acids Allosterically Regulate the ThiamineDiphosphate-dependent �-Keto Acid Decarboxylase fromMycobacterium tuberculosis*□S

Received for publication, August 8, 2007, and in revised form, December 13, 2007 Published, JBC Papers in Press, December 17, 2007, DOI 10.1074/jbc.M707983200

Tobias Werther1,2, Michael Spinka1, Kai Tittmann, Anja Schutz3, Ralph Golbik, Carmen Mrestani-Klaus,Gerhard Hubner, and Stephan Konig4

From the Institute of Biochemistry and Biotechnology, Faculty for Biological Sciences, Martin-Luther-University Halle-Wittenberg,06120 Halle (Saale), Germany

The gene rv0853c from Mycobacterium tuberculosis strainH37Rv codes for a thiamine diphosphate-dependent �-ketoacid decarboxylase (MtKDC), an enzyme involved in theamino acid degradation via the Ehrlich pathway. Steady statekinetic experiments were performed to determine the sub-strate specificity of MtKDC. The mycobacterial enzyme wasfound to convert a broad spectrum of branched-chain andaromatic �-keto acids. Stopped-flow kinetics showed thatMtKDC is allosterically activated by �-keto acids. Even more,we demonstrate that also amino acids are potent activators ofthis thiamine diphosphate-dependent enzyme. Thus, meta-bolic flow through the Ehrlich pathway can be directly regu-lated at the decarboxylation step. The influence of aminoacids on MtKDC catalysis was investigated, and implicationsfor other thiamine diphosphate-dependent enzymes arediscussed.

Over the past years numerous thiamine diphosphate-dependent (ThDP)5 �-keto acid decarboxylases have beenidentified. These enzymes catalyze the non-oxidative decar-boxylation of �-keto acids to aldehydes. The elementary stepsof the catalytic cycle of these enzymes are identical. Initially, thecofactor is activated by deprotonation of its C2 atom. Subse-quently, the generated cofactor ylide attacks the �-carbonylatom of the substrate, yielding a tetrahedral pre-decarboxyla-tion intermediate. This intermediate is decarboxylated, result-ing in the second resonance-stabilized carbanion/enamineintermediate. Finally, the enamine intermediate is protonated,and the reaction product (aldehyde) is released (1).

Based on their substrate specificity, non-oxidative ThDP-de-pendent �-keto acid decarboxylases yielding aldehydes can besubdivided into various groups. As a rule, substrate specificityreflects the biological function of the enzyme. Pyruvate decar-boxylases (PDCs) function in the alcoholic fermentation andconvert pyruvate to acetaldehyde (2). Benzoylformate decar-boxylases are found in many microorganisms and catalyze thedecarboxylation of benzoylformate to benzaldehyde, the thirdreaction in themandelate pathway (3, 4). Indolepyruvate decar-boxylases (IPDCs) and phenylpyruvate decarboxylases are keyenzymes in the biosynthesis of the plant hormones indoleaceticacid and phenylacetic acid, which are derived from the aro-matic amino acids tryptophan and phenylalanine, respectively(5, 6). The conversion of a broad spectrum of �-keto acids ischaracteristic for the branched-chain �-keto acid decarboxyl-ases. In addition to �-keto acids derived from branched-chainamino acids (leucine, isoleucine, and valine), aromatic �-ketoacids are used as substrates, too (7, 8). The branched-chain�-keto acid decarboxylases are important enzymes in the Ehr-lich pathway (9, 10). The first step of the Ehrlich pathway is thetransamination of amino acids to their corresponding �-ketoacids. Subsequently, the �-keto acids are decarboxylated toaldehydes. This quasi-irreversible reaction is catalyzed by aThDP-dependent �-keto acid decarboxylase, which converts abroad spectrum of substrates (8, 9). Depending on the redoxstate of the cell, aldehydes are either oxidized to fusel acids orreduced to fusel alcohols (Fig. 1). The Ehrlich pathway, alsodenoted as amino acid fermentation, is best investigated inyeast (9–11). Vuralhan et al. (9) showed that Saccharomycescerevisiae transcriptionally up-regulates aro10, the gene encod-ing the corresponding�-keto acid decarboxylase in yeast, whengrown on amino acids phenylalanine, leucine, or methionine asnitrogen source, whereas the mRNA amount of other ThDP-dependent decarboxylases (PDC1, PDC5, PDC6, THI3)was notaffected. However, a poor correlation between the transcrip-tion level and enzyme activity in cell free extracts indicated thatthe ARO10 activity is not solely regulated by the amount ofenzyme, and an additional posttranscriptional regulation of thedecarboxylase activity was proposed (9, 11).Transcriptional and posttranscriptional regulation mecha-

nisms are also described for some pyruvate decarboxylases,which are key enzymes in the anaerobic fermentation of glu-cose, yielding ethanol and carbon dioxide. It was shown thatexpression of PDC from Kluyveromyces lactis (KlPDC) is

* The costs of publication of this article were defrayed in part by the paymentof page charges. This article must therefore be hereby marked “advertise-ment” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Figs. S1–S5 and Table S1.

1 Both authors have contributed equally to the article.2 Supported by the Graduiertenkolleg of Sachsen-Anhalt.3 Present address: Max-Delbruck Centre for Molecular Medicine, Protein Sam-

ple Production Facility, 13092 Berlin, Germany.4 To whom correspondence should be addressed. Tel.: 49-345-5524829; Fax:

49-345-5527014; E-mail: [email protected] The abbreviations used are: ThDP, thiamine diphosphate; MtKDC, ThDP-de-

pendent �-keto acid decarboxylase from M. tuberculosis; PDC, pyruvatedecarboxylase; IPDC, indole PDC (EC 4.1.1.74); EcIPDC, IPDC from Enter-obacter cloacae; ScPDC, PDC from S. cerevisiae (EC 4.1.1.1); KlPDC, PDC fromK. lactis; ARO10, ThDP-dependent �-keto acid decarboxylase from S. cer-evisiae; Mes, 4-morpholineethanesulfonic acid.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 283, NO. 9, pp. 5344 –5354, February 29, 2008© 2008 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

5344 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283 • NUMBER 9 • FEBRUARY 29, 2008

by guest on September 1, 2020

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/cgi/content/full/M707983200/DC1

http://www.jbc.org/

induced by glucose and low oxygen concentrations at the tran-scriptional level (12, 13). Additionally, an allosteric activationby the substrate pyruvate was described for KlPDC (14), as it isknown for many other PDCs (15–20). Whereas transcriptionalregulation requires time and a high energy input, allosteric reg-ulation is a short-term and quick response to fluctuating con-ditions in the cell. The molecular basis of substrate activationhas long been studied for PDCs from yeast. One of the basickinetic models of substrate activation of PDCs consists of twosteps and describes the rapid binding of an activator moleculeto the regulatory site, which triggers the slow isomerization ofan inactive enzyme to a catalytically active enzyme (15, 21).This model is supported by crystallographic studies (22–26). Aseries of kinetic investigations on ScPDC variants were con-ducted to establish a signal transfer pathway from the substratebound at the regulatory site to the cofactor ThDP at the activesite (27–31). Another kinetic model, based on extensive studies

on the pH dependence of activation, favors the existence ofvarious activated enzyme states (32, 33). For an excellent reviewon PDC activation, see Schowen (34).In this study we provide kinetic evidence that a ThDP-de-

pendent �-keto acid decarboxylase, which participates inamino acid degradation via the Ehrlich pathway, is subject toallosteric activation. We demonstrate that �-keto acids and,most importantly, also the corresponding amino acids are allo-steric activators of enzyme activity. Binding of amino acids atthe regulatory site and binding of substrates at the active siteshow positive cooperativity.To determine which elementary steps in MtKDC catalysis

are affected by amino acids, the cofactor activation and distri-bution of reaction intermediates was analyzed. Eventually,implications for other non-oxidative ThDP-dependent �-ketoacid decarboxylases are discussed.

EXPERIMENTAL PROCEDURES

Reagents—All substrates (except �-ketovalerate and �-keto-caproate), ThDP, D-amino acids, and alcohol dehydrogenasefrom yeast and horse liver were obtained from Sigma Aldrichand Fluka (Seelze, Germany). L-Amino acids and NADH wereobtained from AppliChem (Darmstadt). Synthesis of �-ke-tovalerate and �-ketocaproate were performed according toFischer (35).Cloning—Genomic DNA fromM. tuberculosis strain H37Rv

was used as the PCR template. The following gene-specificprimers containing 5�-NdeI and 3�-HindIII restriction sites(underlined in the primer sequence) were designed. An addi-tional GCT codon was introduced in the forward primer 5�-AAAACATATGGCTGTGACACCCCAGAAGAGCGATG-CCTGCAG-3� to increase the expression efficiency. Thereverse primer 5�-AAAAAAGCTTTTATCACTGCGGCGC-CATGGATCCCACGAGTTGG-3� contains one extra stopcodon to suppress the production of the C-terminal His6 tagincluded in the cloning vector. The amplified rv0853c gene wascloned into the pCR-BluntII TOPO vector and transformedinto One Shot TOP10 cells (Invitrogen). Plasmid DNAwas iso-lated, and the identity of the cloned fragment was confirmed bysequencing.MtKDC coding plasmid DNA was finally digestedand ligated into theNdeI/HindIII restriction sites of expressionvector pET22b(�) (Novagen, Darmstadt).Protein Expression and Inclusion Body Isolation—For gene

expression, Escherichia coli Rosetta 2 (DE3) (Novagen) trans-formants were grown at 30 °C in 2� YTmedium containing 50�g/ml ampicillin. Gene expression was induced at an A600 of0.8–1.0 by the addition of 0.5mM isopropyl-�-D-thiogalactopy-ranoside. Cells were harvested after 8 h, flash-frozen, andstored at �80 °C. Although various growth conditions havebeen tested (varying temperature and isopropyl-�-D-thiogalacto-pyranoside concentration using variousE. coli expression strains),formation of inclusion bodies could not be prevented. Inclusionbodies were isolated as described by Rudolph et al. (36).Inclusion Body Solubilization, Refolding, and Purification—

Inclusion bodies were suspended in 8 M urea, 100 mM Tris, pH8.0, 100mMdithioerythritol, and 1mMEDTAand stirred for 2 hat room temperature. The remaining insoluble material wasremoved by centrifugation (25 min, 70,000 � g). Denatured

FIGURE 1. Ehrlich pathway. 1, transamination; 2, decarboxylation; 3, reduc-tion; 4, oxidation.

Amino Acids as Allosteric Regulators of Enzyme Activity

FEBRUARY 29, 2008 • VOLUME 283 • NUMBER 9 JOURNAL OF BIOLOGICAL CHEMISTRY 5345


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

protein was dialyzed twice against 6 M urea, 100 mM Tris/HClpH8.0, and 1mMEDTA. Insolublematerialwas again separatedby centrifugation. The supernatant was diluted to 2mg/mlwith6 M urea, 100 mM Tris/HCl, pH 8.0, and 1 mM EDTA andrefolded while dialyzing against 50 mM Tris/HCl pH 7.5, 1 mMEDTA, and 1 mM dithioerythritol. After refolding, 200 mMammonium sulfate was added, and the solution was incubatedat 30 °C for 20 min. The precipitated protein was separated bycentrifugation, and the supernatant was dialyzed against 20mMTris/HCl pH 7.5, 1 mM EDTA, and 1 mM dithioerythritol. Therefolded protein was then applied to an ion exchange column(Fractogel TMAE, column 26 � 200 mm). Elution was per-formedusing a linear gradient of 180ml of 0–50%500mMNaClin 20 mM Tris, pH 7.5. TheMtKDC-containing fractions, elut-ing at 100–200 mM NaCl, were pooled and concentrated.MtKDC was further purified to �95% homogeneity by gel fil-tration on Superdex 200 (GE Healthcare, column 16 � 600mm), equilibratedwith 50mMMes/NaOH, pH 6.5, and 150mMammonium sulfate. The identity of the protein was confirmedby combination of tryptic digestion and matrix-assisted laserdesorption ionization time-of-flight mass spectrometry.Characterization of Substrate Derivatives—Keto-enol tau-

tomerism of �-ketoisocaproate and �-ketoisovalerate, dis-solved in 100 mM acetate, pH 6.0, and 10% (v/v) deuteriumoxide, was investigated by 1HNMR (H,H COSY) and 13CNMR(13C,1H HSQC (heteronuclear single quantum correlation),Attached Proton Test). The amount of keto form, enol form, andhydrate form was deduced from the peak areas of the 1H NMRspectra.All experimentswereperformedonaBrukerAvanceARX400 NMR spectrometer (supplemental Figs. 1 and 2).Protein Concentration—Samples containing ThDP were

determinedwith Bradford assay using bovine serum albumin asstandard (37). Otherwise, the protein concentration was calcu-lated from UV spectra using the calculated molar extinctioncoefficient of 55,350 M�1�cm�1 at 280 nm for theMtKDCmon-omer (see ExPASy).Kinetic Measurements—Measurements including 1H NMR

experiments were performed at 30 °C in 100 mM Mes/NaOH,pH 6.5. The decarboxylation of pyruvate,�-ketobutyrate,�-ke-tovalerate, and �-ketocaproate was monitored using a coupledenzymatic assay with alcohol dehydrogenase from yeast as theauxiliary enzyme (4, 39). For the aromatic �-keto acids indo-lepyruvate, phenylpyruvate, and 4-hydroxyphenylpyruvate,alcohol dehydrogenase from horse liver was used as auxiliaryenzyme (40). The conversion ofNADHwas followed at 340 and366 nm. The decarboxylation of the branched-chain �-ketoacids (�-ketovalerate, �-keto-�-methylvalerate, and �-ketoiso-caproate) was measured by using a direct assay, monitoring then3�* transition of the substrate carbonyl group. For the cal-culation of specific activities, the following molar extinctioncoefficients were determined under standard measurementconditions: �-ketoisovalerate 32.5 M�1�cm�1 at 314 nm, �-keto-�-methylvalerate 33.0 M�1�cm�1 at 316 nm, �-ketoisocaproate25.4 M�1�cm�1 at 317 nm. Catalytic constants of indolepyruvateconversion were calculated, taking the slow keto-enol equilib-rium into account. The effective fraction of the reactive ketoform under measurement conditions was 85% for indolepyru-vate (40). For 4-hydroxyphenylpyruvate and phenylpyruvate,

87.5 and 98% keto form are found in equilibrium. However,because the equilibria establish very rapidly for these two sub-strates, the nominal molar concentrations have been used fordata analysis. For branched-chain�-keto acids,most of the sub-strate exists in the keto form (�-ketoisovalerate, 95–96% ketoform and 4–5% hydrated form; �-ketoisocaproate, 98% ketoform and 2% hydrated form; see supplemental Figs. 1 and 2). Oneunit of catalytic activity is defined as the amount of enzyme con-verting 1�mol of�-keto acid/min at 30 °C. kcat values were deter-mined per monomer using amolecular mass of 59,783 Da (calcu-lated from the deduced amino acid sequence).Stopped-Flow Measurements—Substrate activation and

amino acid activationwere analyzed using a stopped-flow spec-trophotometer (SX18MV, Applied Photophysics, Surrey) andthe standard buffers described above. A solution containingsubstrate (for substrate activation analysis) or substrate andamino acid (for amino acid activation studies) was mixed in a1:1 ratio with a solution containing the auxiliary enzyme alco-hol dehydrogenase (0.5–0.9 units/ml), NADH (0.7–1.9 mM),andMtKDC (0.27–4.9�M). Progress curves were fitted accord-ing to Equation 1 (A is absorbance at time t, and A0 is initialabsorbance), thereby obtaining the observed first order rateconstant for the substrate activation process (kobs), the initialvelocity (v0), and the steady state velocity (vSS).

A � A0 � �SS � t ��SS � �0

kobs� �1 � e�kobs � t� (Eq. 1)

1H NMR Experiments—To investigate the influence of acti-vators on the deprotonation rate constant at the C2 atom ofenzyme-bound ThDP, H/D exchange kinetics were monitoredin presence and absence of an amino acid at 10 mg ofMtKDC/ml, 30 °C, and pH 6.5 (41). Additionally, distributions ofenzyme-bound covalent ThDP intermediates were determinedin the absence and presence of an amino acid according to themethod described in Tittmann et al. (42) at 15 mg of enzyme/ml. In both experiments L-leucine was chosen as the aminoacid, because it acts as a strong activator in conventional kineticexperiments. For each reaction run, the enzyme was preincu-bated with 5 mM L-leucine. For reaction intermediate analyses,measurements were started by the addition of 60 mM pyruvateand stopped by acid quench after 5 s. In the absence ofL-leucine, the reaction does not reach the (final) steady statewithin this reaction time. Because the enzyme converts the sub-strate very rapidly at the required high protein concentration,the reaction time cannot be significantly expanded beyond 5 s.This restriction prevented the calculation of microscopic rateconstants for the catalytic cycle. However, semiquantitativeanalyses have been carried out.The Kinetic Model—A two-site model has been developed to

fit the kinetic data (Fig. 2). Formally it is composed of two sec-tions. In the absence of the effectorX (amino acids in our study)only the species of the upper branch (ElS, El, SEl, SEa, SEaS) arepopulated. Substrate molecules may either bind to the activesite (formation of ElS) or to the regulatory site (formation ofSEl) of the initial enzyme state. Binding at the regulatory sitetriggers a conformational switch to more active enzyme forms(SEa, SEaS). This isomerization step is rate-limiting for the acti-




http://ww

w.jbc.org/

Dow

nloaded from



http://www.jbc.org/

vation in absence of additional effectors. Hence, progresscurves display lag phases whose time constants report on theisomerization step. Thus, themodel proposed closely conformsto the model of Schellenberger, Hubner, and Schowen (15, 21).However, a certain activity is already inherent to the initialenzyme state El. The steady state for the low activity species israpidly established. Therefore, the observed lag phases reflectthe transition from one (initial) steady state to another (final)steady state of product formation. As shown in detail in thesupplemental information, the following equations apply in theabsence of further effectors.

� � �0 � ��SS � �0� � �1 � e�kobs � t� (Eq. 2)

kobs �k�iso � KM

S � KM�

kiso

�1 � KA/K1�� S

KA � K1

K1 � KA� S

�k�iso � KM

S � KM�

k�iso � S

Ka � S(Eq. 3)

with kiso� � kiso /(1 � KA/K1) and Ka � KA�K1/(K1 � Ka)

�0 �

kcat0 � KA

K1 � KA

K1 � KA

K1 � KA� S

�kcat

0� � S

KM0 � S

(Eq. 4)

�SS �Vmax�C�S � S2�

A � B � S � S2 (Eq. 5)

with A � KA�Kiso�KM, B � KM�(1 � Kiso�(1 � KA/K1)), C �kcat0 �A/kcat�K1, and Vmax � kcat�E0.

Obviously it holds kcat0� /KM0 � kcat0 /K1. The complex dependence

of kobs on S after Equation 3 can run throughminimaormaximabut may also account for quasihyperbolic kobs/S-plots (14).In the presence of the effector X the entire scheme of Fig. 2

theoretically becomes relevant. In this case, S and X competefor the regulatory site, whereas X is not supposed to bind to theactive site. Because X actuates conformation changes (steps k4and k6) leading to the activated species XEa and XEaS, it acts asa heterotropic activator. Because of the two independent acti-vation pathways (substrate driven and amino acid driven activa-tion), progress curves should display two lag phases. Empirically,this is not the case. This leads to the assumption that in the pres-ence of amino acids and moderate substrate concentrations, theamino acid-driven pathway dominates, and hence, the sequenceSEl^ SEa^ SEaSmight be neglected.The intrinsic mathematical properties of the model are out-

lined in the section below. The following assumptions aremade. (i) The substrate binding steps are fast compared withthe isomerization steps SEl/SEa, XEl/XEa, and X EtS /XEaS. (ii)In the presence of amino acid the transition of SEl to SEa issuppressed. All relevant differential equations can be added togive

dSEl

dt�

dEl

dt�

dElS

dt�

dXEl

dt�

dXElS

dt� �k4 � XEl � k6 � XElS

� k�4 � XEa � k�6 � XEaS (Eq. 6)

The mass balance equation of all enzyme species reads as E0 �SEl � El � ElS � XEl � XElS � XEa � XEaS

E0 � E l � �1 �S

KA�

S

K1�

X

K�

�X � S

K� � K2� � XEa � �1 �

S

K3� (Eq. 7)

Insertion of Equation 7 into Equation 6 yields

dE l

dt� �kobs � El �

� � E0

� � (Eq. 8)

with

kobs �

�k4 � k6 �S

K2� �

X

K�

�1 �S

KA�

S

K1�

X

K�

�S � X

K2 � K�� k�4 � K3 � k�6 � S

K3 � S � (Eq. 9)

v � k�4 � k�6�S/K3, � 1 � S/K3, and � � 1 � S/KA � S/K1 �X/K� � S�X/K2�K�.As a solution to Equation 8 one obtains

E l � �Ela � Eeq� � e�kobs � t � Eeq (Eq. 10)

with Ela � E0/� and Eeq � v�E0/��kobs . Ela is the concentra-tion of the enzyme species El immediately after the rapid equil-ibration between the speciesEl,ElS,XEl, andXElS, which occursafter mixing the enzyme with S and X. Because the isomeriza-tion steps (k4, k-4, k6, k-6) are considered to be slow, virtually noXEa or XEaS has been formed at this stage. The index “eq”relates to the final equilibrium of all enzyme species. Equation10 can be rearranged to give

FIGURE 2. Proposed kinetic model of MtKDC. This model describes theamino acid and substrate activation behavior and the observed initial activi-ties of MtKDC as well as cooperativity between amino acid and substratebinding. Enzyme species are designated with E, whereas the index l repre-sents enzyme with low activity, and the index a represents the fully activatedenzyme. Ligands are designated with X (amino acid) and S (substrate).Ligands written on the left of the enzyme species are bound to the regulatorysite of the enzyme. Substrates written right from the enzyme are bound to theactive site. The indicated K values represent the corresponding dissociationconstants. The steps leading to fully activated enzymes species are triggeredby binding of ligands to regulatory sites. The isomerization steps are slowcompared with the binding steps.




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

XEaS � XEaSeq � �1 � e�kobs � t� (Eq. 11)

which in turn can easily be converted to

� � �0 � ��SS � �0� � �1 � e�kobs � t� (Eq. 12)

with

�SS �Vmax

app � S

KMapp � S

(Eq. 13)

Vmaxapp �

kcat0 � E0�1 �

X

K�

�kcat

X

kcat0

X

K� � K6�

1 �K1

KA� X � � 1

K�

�1

K� � K6� (Eq. 14)

and

KMapp �

K1�1 �X

K�

� �1 �1

K4��

1 �K1

KA� X � � 1

K�

�1

K� � K6� (Eq. 15)

Notably, the analytical expressions for v0 and vSS in Equation 12differ from those in Equation 2. Most importantly, the aminoacid drivenmechanism predictsMichaelis-Menten behavior asonly one substrate molecule is bound at the working enzyme.Integration of Equation 12 reproduces Equation 1. Equation 9defines a surface function, which is spanned over the XS plane.As seen from the slightly rearranged formulation in Equation16, kobs depends always hyperbolically on X.

kobs � kb �kmax

f � X

KXobs � X

(Eq. 16)

with

kb �k�4 � K3 � k�6 � S

K3 � S(Eq. 17)

kmaxf �

k4 � K2 � k6 � S

K2 � S(Eq. 18)

and

KXobs � K� � K2 � �1 � S/K1

K2 � S � (Eq. 19)

kmaxf is the rate constant for the activation in forward directionat a given substrate concentration under amino acid saturation.kb relates to the reverse reaction. KX

obs represents an apparentdissociation constant of the effector X for the initial enzymestate, again relevant for a defined substrate concentration.Technically, KX

obs is the half-saturation constant of the kobs-Xdependence at a given substrate concentration. The depend-ence of kobs on S is potentiallymore complex. Depending on theindividual values of themicroscopic constants, kobsmay displayupward or downward curvatures over the substrate domain. Intheory, Equations 16–19 offer a straightforward method fordata evaluation. A comprehensive kinetic analysis wouldrequire the determination of kb, kmax

f , and KXobs for several suf-

ficiently varying substrate concentrations. Subsequently, thethree apparent constants kb, kmax

f , and KXobs should be replotted

as functions of S. Ideally, k4, k6, and K2 could be extracted froma plot of kmax

f versus S, whereas analysis of KXobs would allow the

estimation ofK� andK2/K1. Plotting of kb versus Swould finallyyield k�4, k�6, and K3. In the current study this approachremained restricted to the S/X couple benzoylformate/L-leucine. In all other cases we merely report the net rate con-stants kb, kmax

f , andKXobs at a constant substrate concentration of

5 mM instead.For the comparison of the activation potential of�-keto acids

and amino acids, it is expedient to introduce specificity con-stants for activation, defined in analogy to catalytic efficiencies(kcat/S0.5). At low concentrations the activation potential ofamino acids can be characterized by the second-order rate con-stant kmax

f /KXobs. The activation potential of substrates is given

by the ratio of kiso�/Ka, which is numerically equal to kiso/KA(see Equation 3).

RESULTS AND DISCUSSION

Cloning, Expression, and Purification—The mycobacterialgene rv0853c encoding a putative PDC or IPDC was amplifiedfrom genomic DNA. The MtKDC was expressed as inclusionbodies, refolded, and purified to homogeneity using anionexchange chromatography and gel filtration. Approximately 50mg of protein were obtained from 6 liters of cell culture.Substrate Specificity—As based on amino acid sequence

comparison, MtKDC was originally proposed to be a PDC orIPDC (UCLA-DOE Institute for Genomics and Proteomics).To classify purified MtKDC on empirical grounds, we deter-mined the catalytic constants for the conversion of 11 aliphatic,aromatic, and branched-chain �-keto acids. Table 1 summa-rizes the calculated values for S0.5 and kcat. Pyruvate was con-verted with the lowest catalytic efficiency (kcat/S0.5) and dis-played a weak substrate inhibition (Fig. 3C). The highestcatalytic efficiencies were found for indolepyruvate and �-ke-toisocaproate. In general, the order of the S0.5 values point topreferred binding of hydrophobic substrates, with the possibleexception of benzoylformate. The highest kcat values wereindeed observed for branched-chain �-keto acids. Thus,MtKDC appears to be optimized for the conversion of aromaticand branched-chain �-keto acids, which contrasts strikingly toPDCs that prefer pyruvate as substrate (44). A similar substratespectrum was found for other ThDP-dependent enzymes par-taking in amino acid degradation, e.g. for the branched-chain�-keto acid decarboxylase from Lactococcus lactis (KdcA) (7, 8)and for the�-keto acid decarboxylaseARO10 from S. cerevisiae(9). KdcA has the highest catalytic specificities for phenylpyru-vate and �-ketoisovalerate but the lowest for pyruvate (7).ARO10 was shown to convert phenylpyruvate, �-ketoisovaler-ate, �-keto-�-methylvalerate, and �-ketoisocaproate (9). Addi-tionally, we show also that EcIPDC decarboxylates phenylpyru-vate as well as branched-chain and extended aliphatic �-ketoacids (supplemental Table 1). Taken together, ThDP-depend-ent enzymes involved in amino acid degradation via the Ehrlichpathway display a broad substrate spectrum, some preferencefor hydrophobic substrates, and a low specificity for pyruvate.




http://ww

w.jbc.org/

Dow

nloaded from


http://www.jbc.org/

Interestingly, MtKDC exhibits a sigmoid dependence of itscatalytic activity on substrate concentration. Whereas the sig-moidicity is distinctly pronounced for benzoylformate, �-keto-caproate, 4-hydroxyphenylpyruvate, and phenylpyruvate, it ismuch weaker in case of the other �-keto acids analyzed (seeTable 1 and supplemental Fig. 3). Sigmoid dependence of thecatalytic activity on the substrate concentration generally indi-cates substrate activation behavior.Substrate Activation Behavior—Progress curves displayed

distinct lag phases, pointing to an allosteric substrate activation

mechanism (Fig. 3A). Because theobserved kcat values are at least oneorder of magnitude higher than thekobs values associated with the lagphases, catalytic steps cannot berate-limiting for these transients.Such behavior was described for allPDCs investigated so far (14–16),except for the enzyme fromZymomonas mobilis (45). A basickinetic model for this phenomenonwas originally developed by Hubneret al. (15) and extended later byAlvarez et al. (21). An alternativemodel with various activatedenzyme states was advanced by Ser-gienko and Jordan (32, 33).Stopped flow transients pos-

sessed significant initial velocities(v0) (Fig. 3, A and B). A similar phe-nomenon was found for PDC fromPisum sativum (16). In contrast, ini-tial velocities close to zero areobserved for other substrate-acti-vated PDCs such as ScPDC andKlPDC. These enzymes are, thus,potentially inactive in the absence ofsubstrate (14, 15, 46). The observedinitial velocities ofMtKDCand theirhyperbolic dependence on substrateconcentration (Fig. 3C, Table 1, seealso supplemental Fig. 3) cannot be

explained by the kinetic model of Alvarez et al. (21). Therefore,we postulate an initial enzyme state that is partially active in theabsence of the substrate (Fig. 2). The addition of substratestriggers the slow transition of the enzymewith a lowbasic activ-ity (SEl, El) into a fully activated enzyme species (SEa, SEaS).Alternatively, initial activities could be described by an equilib-riumbetween inactive and active enzyme species, which under-goes a shift toward the activated state upon the addition ofsubstrate. So far, this is not substantiated by our data but cannotbe completely ruled out at the moment. Table 1 includes a sur-

FIGURE 3. Kinetics of MtKDC activation in the absence of amino acids. Shown are original (A) and differen-tiated (B) progress curves of catalysis at different pyruvate concentrations (1, 10 mM; 2, 15 mM; 3, 25 mM; 4,40 mM). C, plot of the steady state activity (closed circles) and initial activity (open circles) versus pyruvateconcentration. Experimental data of steady state activity and initial activity was fitted according to equationvss � Vmax�S

2/A � B�S � (1 � S/Ki)�S2) and v0 � Vmax

0 �S/(K1�(1 � S/KA) � S), respectively. D, semilogarithmic plotof the dependence of the activation rate constant (kobs) on substrate concentration for different �-keto acids(circles, pyruvate; cross-hairs, �-ketobutyrate; squares, �-keto-�-methylvalerate; triangles, phenylpyruvate; dia-monds, indole pyruvate). Experimental data were fitted according to Equation 3. The semilogarithmic scale wasused to simplify the comparison of various �-keto acids, as their kiso

� /Ka values differ by orders of magnitude.

TABLE 1Catalytic constants for the decarboxylation of different �-keto acids by MtKDCThe values S0.5 and kcat were derived from the fit of the v/S-plot according to the equation vss � Vmax�S2/(A � B�S � (1 � S/Ki)�S2). Hill coefficients n, calculated accordingto the equation n � (2�A � B�S0.5)/(A � B�S0.5) (32), were included as a phenomenological measure of sigmoidicity. Numbers are given as the average S.D. of twoindependent experiments.

Substrate S0.5 kcat kcat/S0.5 n kcat0� Km0

mM s�1 mM�1 s�1 s�1 mM

Indole pyruvate 0.26 0.12 2.98 0.11 11.46 1.22 0.40 0.09�-Ketoisocaproate 2.90 31.5 10.86 1.30Phenyl pyruvate 0.83 0.08 7.50 0.10 9.03 1.41 1.47 1.17�-Keto caproate 0.92 0.22 7.06 0.01 7.67 1.61 1.25 0.83�-Keto valerate 1.31 0.24 6.95 0.04 5.30 1.33 1.13 1.214-Hydroxyphenylpyruvate 0.93 0.18 3.89 0.04 4.18 1.52�-Keto-�-methyl valerate 6.68 0.97 19.75 1.21 2.96 1.32 10.05 20.25Benzoylformate 8.25 0.78 18.57 0.47 2.25 1.96 7.05 26.11�-Keto isovalerate 24.47 3.06 28.35 2.19 1.16 1.21�-Keto butyrate 14.17 3.95 5.98 0.33 0.42 1.04 3.68 138.30Pyruvatea 98.89 37.48 2.14 0.21 0.02 1.02 0.40 0.40

a Weak substrate inhibition (Ki � 124 mM) was detected only for pyruvate.




http://ww

w.jbc.org/

Dow

nloaded from



http://www.jbc.org/

vey of empirical kcat values of the initial enzyme state (kcat0� ) forseveral substrates aswell as their respectiveKM

0 values. It shouldbe noted that the given kcat0� are drawn from fittings of v0 versusS according to Equation 4, therefore not being identical to themicroscopic kcat0 values, which will be somewhat higher. Thestrongest activation effects in terms of kcat/kcat0� have beenobserved for indolepyruvate (7.45), �-ketovalerate (6.15), and�-ketocaproate (5.65).

Because of limited solubility of the substrates, it was impos-sible to reach the saturation ranges of the kobs/S-plots for any ofthe substrates, apart from pyruvate (Fig. 3D). However, amarked curvature indicating substrate saturation behavior isseen for most substrates (see supplemental Fig. 4). Thus, theassociated fitting errors of the maximum activation rate con-stant (kiso� ) and the half-saturation concentration of the activa-tion rate (Ka) are relatively high. In some cases these parameterscould not have been determined at all (see Table 2 and supple-mental Fig. 4). Nevertheless, in each case the ratio of kiso� /Kacould be calculated with sufficient accuracy from the initialslope of the concentration dependence of kobs (Table 2). Thesevalues represent a measure of the activation potentials of sub-strates. Table 2 demonstrates that aromatic�-keto acids exert aconsiderable activating effect onMtKDC, with indolepyruvatebeing the most efficient activator. In contrast, pyruvate is theweakest �-keto acid activator for MtKDC. Under equivalentconditions, the activation potential (kiso� /Ka � kiso/KA) of pyru-vate is significantly higher for substrate-activated PDCs(KlPDC, 0.015 mM�1�s�1; ScPDC, 0.070 mM�1�s�1; PsPDC,0.375 mM�1�s�1) (14). As evidenced in a series of publications,Cys221 of ScPDC is the starting point for the information trans-fer in the process of substrate activation (47–49). Interestingly,this amino acid is not conserved at the equivalent position ofMtKDC (supplemental Fig. 5). As a consequence, themolecularactivation mechanism of MtKDC should differ from that ofScPDC.Amino Acid Activation—It is well known that many bac-

teria catabolize aromatic and branched-chain amino acidsvia the Ehrlich pathway and that the rate-limiting step of thisroute is catalyzed by a ThDP-dependent decarboxylase (8,50). Thus, we investigated whether precursors (amino acids)and end products (fusel alcohols and fusel acids) of the Ehr-lich pathway affect MtKDC activity. Whereas tryptophol(fusel alcohol), phenylacetic acid, and indoleacetic acid(fusel acids) had no effect, we could clearly demonstrate that

amino acids are potent activators of MtKDC (Fig. 4A). Wetested the effect of L-leucine, L-valine, D-valine, L-isoleucine,L-phenylalanine, D-phenylalanine, L-tyrosine, L-tryptophan,and L-alanine. All these amino acids except L-tryptophanproduce a distinct shortening of the lag phases of the pro-gress curves. A hyperbolic dependence of the activation rateconstant kobs on the amino acid concentration was generallyfound (Fig. 4, B and C), except for L-tyrosine and D-pheny-lalanine due to their insufficient solubility. These findingsrelate to the amino acid section of the kinetic model ofMtKDC in Fig. 2, postulating that the binding of amino acidat the regulatory site triggers a slow isomerization of enzymespecies with low activity (XEl, XElS) to completely activatedenzyme species (XEa, XEaS). The kinetic constants for aminoacid activation were calculated according to Equation 16 andare summarized in Table 3. A comparison of the activationpotential of amino acids in terms of kmax

f /KXobs revealed that

L-leucine and L-isoleucine are the strongest activators forMtKDC. A significant lower affinity of MtKDC was shownfor D-valine as compared with L-valine (Fig. 4B, Table 3),indicating that the configuration at the C�-atom is impor-tant for the binding of ligands to the regulatory site, too.As anticipated for allosteric activators, the presence of amino

acids at saturation concentration shifts the sigmoid character ofthe v/S-plots towardMichaelis-Menten behavior. Fig. 4D illus-trates the dependence of the reaction rate on the �-keto-isovalerate concentration in the absence and the presence oftwo different L-valine concentrations. The transition from asigmoid curve to a hyperbolic one is accompanied by decreas-ing S0.5 values and increasing kcat values. In absence of L-valine,kcat/S0.5 is 1.16 mM�1�s�1, and in the presence of L-valine (50�M and 1 mM), kcat/S0.5 values are 2.02 and 8.3 mM�1�s�1,respectively. Consequently, amino acids substantially increase thecatalytic efficiency of MtKDC for its substrates. Because theempiricalkcat valuesof aminoacid activatedMtKDC(kcatX ) areonlymoderately elevatedcomparedwith thatof the substrate-activatedenzyme (kcat), themajor part of this increase of catalytic efficiencyrests on the lowered S0.5 values.Interdependence of Substrate and Amino Acid Binding—To

obtain further information on substrate activation and aminoacid activation as well as on their interplay, we investigated thedependence of the activation rate on the amino acid concentra-tion at different substrate concentrations. An increase of thesubstrate concentration resulted in an increase of the amino

TABLE 2Microscopic constants for the activation of MtKDC by �-keto acidsValues for k�iso, k iso

� , andKawere determined by fitting the experimental data to Equation 3 (see also supplemental Fig. 4), whereas values present the fitting errors. k iso� /Ka

is ameasure of the activation efficiency of the�-keto acid and is numerically equal to kiso/KA. For phenylpyruvate, ketocaproate, and benzoylformate this ratiowas calculatedfrom the initial slope of the concentration dependence of kobs.

Substrate k�iso k iso� Ka kiso� /Ka

s�1 s�1 mM mM�1�s�1

Indolepyruvate 0.031 0.035 0.741 0.222 2.06 1.11 0.3594-Hydroxyphenylpyruvate 0.027 0.048 0.327 2.04 3.57 18.10 0.092Phenylpyruvate 0.028�-Ketocaproate 0.026�-Ketovalerate 0.080 0.245 0.378 0.039 21.1 5.35 0.018Benzoylformate 0.015�-Keto-�-methylvalerate 0.054 0.125 0.246 0.026 84.83 16.35 0.0029�-Ketobutyrate 0.068 0.066 0.121 0.013 132.0 30.60 0.00092Pyruvate 0.080 0.032 0.017 0.002 29.9 27.20 0.00057




http://ww

w.jbc.org/

Dow

nloaded from






http://www.jbc.org/

acid affinity toMtKDC (Fig. 5, A and B). This positive cooper-ativity between the binding of the substrate and the amino acidindicates the presence of a regulatory site spatially separatedfrom the active site. In accordance with the measured kineticdata, we propose the existence of a common regulatory site forthe keto acids and the amino acids. Furthermore, an increase ofthe substrate concentration resulted in a decrease of the maxi-mum observed activation rate constant under amino acid satu-ration (Fig. 5,C andD). Fitting our empirical kobs values accord-ing to Equation 16–19, the following constants for the

activation ofMtKDCby L-leucine asactivator and benzoylformate assubstrate can be estimated: K1 �126.3 mM, K� � 14.5 mM (Equation19, Fig. 5B), K� � 0.103 mM (Fig.5B), K2 � 0.89 mM (Equation 18,Fig. 5C), k4 � 0.40 s�1 (Equation18, Fig. 5C), and k6 � 0.11 s�1

(Equation 18).1H NMR Experiments—To eluci-

date which reaction steps of the cat-alytic cycle are specifically affectedby allosteric activation of aminoacids, we analyzed the H/Dexchange at the C2 atomof ThDP aswell as the distribution of reactionintermediates formed during thecatalysis in absence and presence ofamino acids. H/D-exchange experi-ments clearly show that the C2 dep-rotonation of the cofactor ThDP isnot rate-limiting for catalysis.Moreover, the presence of L-leucinedid not substantially increase therate constant for the H/D exchange(kobs � 122 13.7 s�1 withoutL-leucine and 157 73.9 s�1 withL-leucine, as compared with thehighest kcat value of 31.5 s�1, seeTable 1). This points again to a fun-damental difference between theactivation mechanism of PDCs andMtKDC. In contrast, in ScPDC C2deprotonation of the cofactor isconsiderably accelerated in thepresence of the artificial activatorpyruvamide (41).Finally, we investigated the influ-

ence of the strongly activatingamino acid L-leucine on the distri-bution of intermediates, formed inthe presence of the poorly activatingsubstrate pyruvate. Our experi-ments revealed that in the absenceof the amino acid, 32% of theenzyme-bound was found as thereaction intermediate 2-hydroxy-ethyl-ThDP, leaving most of the

enzyme-bound ThDP unreacted. In the presence of L-leucine,however, 76%of the enzyme-bound cofactor existed in the formof reaction intermediates 2-hydroxyethyl-ThDP (67.5%) and2-lactyl-ThDP (8.5%) (Fig. 6A). Hence, the catalytic stepaffected by amino acids precedes or is identical with the forma-tion of the first covalently bound reaction intermediate 2-lac-tyl-ThDP. Because the formation of the ylide is not affected bythe amino acid, the carbonyl addition of the substrate seems tobe the target of amino acid activation (Fig. 6B). Furthermore, inthe presence of L-leucine the concentration ratio of 2-lactyl-

FIGURE 4. Kinetics of MtKDC activation in the presence of amino acids. A, progress curves of benzoyl-formate catalysis (5 mM) at different concentrations of L-leucine (1, 0 mM; 2, 0.01 mM; 3, 0.05 mM; 4, 0.2 mM; 5, 2mM). B and C, dependence of the activation rate constant (kobs) on amino acid concentration. Experimental datawere fitted according to Equation 16. All measurements were performed at 5 mM benzoylformate as substrate.B: closed hexagons, L-leucine; open squares, L-valine; cross-hairs, D-valine. C: closed diamonds, L-tryptophan;closed circles, L-alanine; filled triangles; L-phenylalanine; filled inverted triangles, D-phenylalanine; closed hexa-gons, L-leucine. For better visualization and comparison of various amino acids, a semilogarithmic scale wasused. D, influence of L-valine on the steady state rates of �-ketoisovalerate decarboxylation. MtKDC was pre-incubated with 0.05 mM (open circles) and 1 mM L-valine (closed diamonds), respectively; closed circles, withoutpreincubation. The conversion of �-ketoisovalerate was monitored directly at 314 nm. Experimental data werefitted according to equation vss � Vmax�S

2/(A � B�S � S2). E, comparison of kobs values for amino acids and theircorresponding �-keto acids (closed squares, L-isoleucine; open squares, �-keto-�-methylvalerate; closed circles,L-alanine; open circles, pyruvate). Experimental data were fitted according to Equation 3 for �-keto acid and toEquation 16 for amino acid dependences.




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

ThDP and 2-hydroxyethyl-ThDP is altered, implicating thatalso the rate ratio of decarboxylation and aldehyde release isslightly modified by this activator.Correlation between the Kinetic Model and the Experimental

Data—In the absence of amino acids the model concomitantlyexplains (i) the hyperbolic v/S-plots of the initial velocities (v0,Equation 4), (ii) the sigmoid v/S-plots of the velocities in the

final steady state (vSS, Equation 5), and (iii) the complexdependence of kobs on the substrate concentration in keepingwith Equation 3 (supplemental Fig. 4). It is, however, difficult todetermine statistically reliable values for A, B, and C of Equa-tion 5. Therefore, we preliminarily restricted our analysis to S0.5values and Hill coefficients (Table 1). This circumstance pre-vents the calculation of elementary parameters (kiso, KA, K1).

In the presence of amino acids the following findings arerelevant to our analysis. (i) kobs depends always hyperbolicallyon X. This is in keeping with Equation 16. (ii) kobs decreaseswith S at high amino acid concentration. This result is easilyaccommodated by the model and requires k4 � k6. (iii) Anincrease of substrate (benzoylformate) concentration leads todecreasingKX

obs values for L-leucine,which phenomenologicallyillustrates the positive cooperativity between substrate andamino acid if bound to the initial enzyme state. As is obvious fromEquation 19,KX

obs will decrease with increasing S ifK1/K2 � 1. Onthe grounds ofmicroscopic reversibility (imposingK1K� �K�K2)this is equivalent toK�/K� � 1.Hence, the affinity of the substrateis higher forXEl than for El (K1 � K2). Accordingly, the affinity ofamino acids is higher for enzyme species in which the substrate isalready bound at the active site (K� � K�).In summary, the model qualitatively conforms to the kinetic

data and, moreover, provides asound frame for quantitative dataanalysis. Nevertheless, it is a phe-nomenological one. Themodel pos-tulates the existence of a regulatorysite spatially separated from theactive site. This implies an informa-tion transfer from the former to thelatter, which remains to beexplored. In terms of formal kinet-ics, the proposed model is largely alogical extension of the activationscheme of Hubner, Schellenberger,and Schowen (15, 21). Whetherthere might be a kinetically relevantcommunication between the activesites as shown in the case of pyru-vate dehydrogenase complex (51)and discussed in case of ScPDC (52)cannot be decided at the currentstage. Apart from detailed kineticinvestigations, the next step to ver-ify the validity of the model wouldinvolve the crystallographic analysisof MtKDC in the presence andabsence of activators. A similarstudy has recently been publishedon the related enzyme phenylpyru-vate decarboxylase (53, 54), provid-ing evidence that the regulatorybinding site for the substrate differsfrom that in ScPDC.The discovery ofmultiphasic pro-

gress curves for ScPDC, particularlyat pH values below 6.0, made the

FIGURE 5. Dependence of the observed rate constant of activation kobs on the amino acid concentrationand substrate concentration, respectively. A, dependence of kobs on the amino acid concentration(L-leucine) for two substrate concentrations (closed inverted triangles, 3 mM benzoylformate; closed triangles, 40mM benzoylformate); fits are according to Equation 16. B, dependence of the dissociation constant KX

obs ofL-leucine on benzoylformate concentration. Experimental data were fitted according to Equation 19.C, dependence of kobs on the substrate concentration of benzoylformate in the absence (open circles) and thepresence of 40 mM L-leucine (open squares). Experimental data in the absence and the presence of L-leucinewere fitted according to Equation 3 and Equation 16, respectively. D, three-dimensional plot of the depend-ence of the activation rate constant kobs on substrate and amino acid concentration, given as superposition ofA and C.

TABLE 3Microscopic constants for the activation of MtKDC by amino acidsThe values kb, kmax

f , and KXobs were determined in the presence of 5 mM benzoyl-

formate as substrate and derived from the fit of the plot of the activation rateconstant (kobs) versus amino acid concentration by fitting according to Equation 16,whereas values are the fitting errors.

Amino acid kb kmaxf KX

obs kmaxf /KX

obs

s�1 s�1 mM s�1�mM�1

L-Leu 0.021 0.012 0.225 0.006 1.53 0.28 0.147L-Ile 0.022 0.013 0.258 0.010 1.74 0.34 0.148L-Val 0.018 0.016 0.258 0.021 4.34 1.28 0.059L-Phe 0.014 0.007 0.262 0.056 3.80 0.45 0.069D-Val 0.016 0.002 0.205 0.008 119.3 11.8 0.002D-Phea 0.019 0.001 0.001L-Ala 0.013 0.002 0.124 0.007 157.8 16.8 0.0008

a Because of insufficient solubility, no saturation for the activation rate constantcould be reached. Therefore, the kmax

f and KXobs values could not be determined.

For L-tryptophan, only kb was determined (0.017 0.0004 s�1). The activationpotential of amino acids is given by the ratio of kmax

f /KXobs.




http://ww

w.jbc.org/

Dow

nloaded from


http://www.jbc.org/

development of complex activation models necessary (32, 33).In the case of MtKDC, all transients have been found to beperfectly single exponential under all conditions applied. Thishas been checked by comprehensive analyses of residual plots(data not shown). However, we cannot exclude that hiddenkinetic complexity might surface under different experimentalconditions, e.g. at other pH values.Conclusions—Differences in the allosteric regulationmecha-

nism ofMtKDC and substrate-activated PDCs described abovemay be attributed to the fact that those enzymes are importantin different metabolic pathways. The substrate spectrum ofMtKDC presumably requires an enhanced degree of the activecenter hydrophobicity as compared with that of the active cen-ters of PDCs. Strong hydrophobicity and/or low polarity of thecofactor environment is paramount to fastH/D exchange at thecofactor C2 atom (55–57). Therefore, in a strongly hydropho-bic active center deprotonation at C2 is rapid even before acti-vation. Conclusively, another step has to take the role of thetarget of activation (i.e. C-C-bond formation in case ofMtKDC). Whereas PDCs are key enzymes in the alcoholic fer-mentation, ARO10, KdcA, andMtKDC play major roles in theamino acid degradation. Thus, it is not surprising that the sub-strate-activated ScPDC is not modulated by amino acids (ana-lyzed for concentrations up to 750 mM L-alanine; data notshown).For the ThDP-dependent decarboxylases participating in

either the alcoholic fermentation or the amino acid degrada-tion, enzymes were described that either display or lack allo-steric regulation. ScPDC andKlPDC are both allosterically reg-ulated by their substrate pyruvate. Examples for non-regulatedThDP-dependent enzymes involved in amino acid degradationare EcIPDC (40) and KdcA (7). The mycobacterial enzymeinvestigated in this study and the phenylpyruvate decarboxyl-ase fromAzospirillum brasilense (53) are examples for alloster-ically regulated enzymes involved in amino acid catabolism.

Here, we present for the first timedirect kinetic evidence that aminoacids are potent activators of aThDP-dependent �-keto aciddecarboxylase. Furthermore, wedemonstrate that amino acids andsubstrates modulate the catalyticefficiency of MtKDC in a coopera-tive manner. Metabolic flowthrough the Ehrlich pathway can,thus, be specifically regulated at thedecarboxylation step. The effect ofamino acids on the key enzyme ofthe Ehrlich pathway was previouslyinvestigated on the level of tran-scription for the MtKDC homo-logue enzyme ARO10 from S. cer-evisiae (9, 11). The authorspostulated an additional posttran-scriptional regulation of the enzymeactivity. We assume that ARO10 iskinetically regulated by amino acidsin the same way as found for

MtKDC. For further studies on the mechanism of amino acidactivation, kinetic and structural investigations of ARO10 andphenylpyruvate decarboxylase fromA. brasilense (53) might beinteresting. Particularly, structural data are necessary to obtainmore information on activator binding. Small angle x-ray scat-tering experiments withMtKDCdemonstrated structural rear-rangements upon the addition of amino acids (43). Hitherto,attempts to crystallizeMtKDC have been unsuccessful.Transposon site hybridization experiments showed that

MtKDC expression is not essential for the optimal growth ofMycobacterium tuberculosis (38). Thus,MtKDC is not a poten-tial target for novel anti-tuberculosis drugs. However, our find-ings might be of common interest for studies in the field ofmetabolic pathway regulation, enzyme catalysis, andmore spe-cifically, regulation of ThDP-dependent enzymes.

Acknowledgments—We thank Angelika Schierhorn for performingmass spectroscopy measurements and Manfred S. Weiss for materialsupport for MtKDC cloning.

REFERENCES1. Schellenberger, A. (1998) Biochim. Biophys. Acta 1385, 177–1862. Neuberg, C., and Karczag, L. (1911) Biochem. Z. 36, 68–753. Hegeman, G. D. (1966) J. Bacteriol. 91, 1140–11544. Weiss, P. M., Garcia, G. A., Kenyon, G. L., Cleland,W.W., and Cook, P. F.

(1988) Biochemistry 27, 2197–22055. Koga, J., Adachi, T., and Hidaka, H. (1992) J. Biol. Chem. 267,

15823–158286. Somers, E., Ptacek, D., Gysegom, P., Srinivasan, M., and Vanderleyden, J.

(2005) Appl. Environ. Microbiol. 71, 1803–18107. Yep, A., Kenyon, G. L., and McLeish, M. J. (2006) Bioorg. Chem. 34,

325–3368. Smit, B. A., van Hylckama Vlieg, J. E. T., Engels,W. J., Meijer, L.,Wouters,

J. T., and Smit, G. (2005) Appl. Environ. Microbiol. 71, 303–3119. Vuralhan, Z., Luttik,M. A., Tai, S. L., Boer, V.M.,Morais,M. A., Schipper,

D., Almering,M. J., Kotter, P., Dickinson, J. R., Daran, J.M., and Pronk, J. T.(2005) Appl. Environ. Microbiol. 71, 3276–3284

FIGURE 6. A, intermediate distribution in MtKDC catalysis of pyruvate in the presence and absence of L-leucine.The reaction was started by the addition of 60 mM pyruvate in the presence and absence of 5 mM L-leucine andquenched with trichloroacetic acid/hydrochloric acid after 5 s. HEThDP, 2-hydroxyethyl-ThDP LThDP, 2-lactyl-ThDP. B, catalytic cycle of MtKDC with intermediates of the non-oxidative decarboxylation of pyruvate. 1,cofactor activation (ylide formation); 2, formation of the Michaelis complex; 3, covalent binding of the substrateat the C2-atom of ThDP; 4, decarboxylation; 5, aldehyde release after protonation of the �-carbanion.




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

10. Dickinson, J. R., Salgado, L. E., andHewlins,M. J. (2003) J. Biol. Chem. 278,8028–8034

11. Vuralhan, Z., Morais, M. A., Tai, S. L., Piper, M. D., and Pronk, J. T. (2003)Appl. Environ. Microbiol. 69, 4534–4541

12. Breunig, K. D., Bolotin-Fukuhara, M., Bianchi, M. M., Bourgarel, D., Fal-cone, C., Ferrero, I. I., Frontali, L., Goffrini, P., Krijger, J. J., Mazzoni, C.,Milkowski, C., Steensma, H. Y., Wesolowski-Louvel, M., and Zeeman,A. M. (2000) Enzyme Microb. Technol. 26, 771–780

13. Kiers, J., Zeeman, A. M., Luttik, M., Thiele, C., Castrillo, J. I., Steensma,H. Y., van Dijken, J. P., and Pronk, J. T. (1998) Yeast 14, 459–469

14. Krieger, F., Spinka, M., Golbik, R., Hubner, G., and Konig, S. (2002) Eur.J. Biochem. 269, 3256–3263

15. Hubner, G., Weidhase, R., and Schellenberger, A. (1978) Eur. J. Biochem.92, 175–181

16. Dietrich, A., and Konig, S. (1997) FEBS Lett. 400, 42–4417. Davies, D. D. (1967) Proc. Biochem. Soc. 104, 50P18. Zehender, H., Trescher, D., and Ullrich, J. (1987) Eur. J. Biochem. 167,

149–15419. Neuser, F., Zorn, H., Richter, U., and Berger, R. G. (2000) Biol. Chem. 381,

349–35320. Acar, S., Yucel, M., and Hamamci, H. (2007) EnzymeMicrob. Technol. 40,

675–68221. Alvarez, F. J., Ermer, J., Hubner, G., Schellenberger, A., and Schowen, R. L.

(1991) J. Am. Chem. Soc. 113, 8402–840922. Dyda, F., Furey, W., Swaminathan, S., Sax, M., Farrenkopf, B., and Jordan,

F. (1993) Biochemistry 32, 6165–617023. Arjunan, P., Umland, T., Dyda, F., Swaminathan, S., Furey, W., Sax, M.,

Farrenkopf, B., Gao, Y., Zhang, D., and Jordan, F. (1996) J. Mol. Biol. 256,590–600

24. Furey, W., Arjunan, P., Chen, L., Sax, M., Guo, F., and Jordan, F. (1998)Biochim. Biophys. Acta 1385, 253–270

25. Lu, G., Dobritzsch, D., Konig, S., and Schneider, G. (1997) FEBS Lett. 403,249–253

26. Lu, G., Dobritzsch, D., Baumann, S., Schneider, G., and Konig, S. (2000)Eur. J. Biochem. 267, 861–868

27. Baburina, I., Dikdan, G., Guo, F., Tous, G. I., Root, B., and Jordan, F. (1998)Biochemistry 37, 1245–1255

28. Baburina, I., Li, H., Bennion, B., Furey,W., and Jordan, F. (1998) Biochem-istry 37, 1235–1244

29. Li, H., and Jordan, F. (1999) Biochemistry 38, 10004–1001230. Li, H., Furey, W., and Jordan, F. (1999) Biochemistry 38, 9992–1000331. Joseph, E., Wei, W., Tittmann, K., and Jordan, F. (2006) Biochemistry 45,

3517–1352732. Sergienko, E. A., and Jordan, F. (2001) Biochemistry 40, 7382–740333. Sergienko, E. A., and Jordan, F. (2002) Biochemistry 41, 3952–396734. Schowen, R. L. (2007) Isotopes Environ. Health Stud. 43, 1–16

35. Fischer, G. (1971) Investigations on Structure and Reactivity of Alpha-ketoAcidsandAlpha-ketoAmides.Doctoraldissertation,Martin-LutherUniver-sity Halle-Wittenberg

36. Rudolph, R., Bohm,G., Lilie, H., and Jaenicke, R. Folding proteins (1997) inProtein Function, a Practical Approach (Creighton, T. E., ed) pp. 55–99,IRL Press at Oxford University Press, Oxford

37. Bradford, M. M. (1976) Anal. Biochem. 72, 248–25438. Sassetti, C. M., Boyd, D. H., and Rubin, E. J. (2003) Mol. Microbiol. 48,

77–8439. Holzer, H., Schultz, G., Villar-Palasi, C., and Juntgen-Sell, J. (1956) Bio-

chem. Z. 327, 331–34440. Schutz, A., Golbik, R., Tittmann, K., Svergun, D. I., Koch,M.H. J., Hubner,

G., and Konig, S. (2003) Eur. J. Biochem. 270, 2322–233141. Kern, D., Kern, G., Neef, H., Tittmann, K., Killenberg-Jabs,M.,Wikner, C.,

Schneider, G., and Hubner, G. (1997) Science 275, 67–7042. Tittmann, K., Golbik, R., Uhlemann, K., Khailova, L., Schneider, G., Patel,

M., Jordan, F., Chipman, D. M., Duggleby, R. G., and Hubner, G. (2003)Biochemistry 42, 7885–7891

43. Werther, T., Konarev, P., Svergun, D. I., and Konig, S. (2006) HasylabAnnual Report, pp. 361–362, Hamburger SynchrotronstrahlungslaborHASYLAB am Deutschen Elektronen-Synchrotron DESY in der Helm-holtz-Gemeinschaft HGF, Hamburg, Germany

44. Lehmann, H., Fischer, G., Hubner, G., Kohnert, K. D., and Schellenberger,A. (1973) Eur. J. Biochem. 32, 83–87

45. Bringer-Meyer, S., Schimz, K. L., and Sahm, H. (1986) Arch Microbiol.146, 105–110

46. Hubner, G., and Schellenberger, A. (1986) Biochem. Int. 13, 767–77247. Baburina, I., Gao, Y., Hu, Z., Jordan, F., Hohmann, S., and Furey,W. (1994)

Biochemistry 33, 5630–563548. Baburina, I., Moore, D. J., Volkov, A., Kahyaoglu, A., Jordan, F., and Men-

delsohn, R. (1996) Biochemistry 35, 10249–1025549. Wang, J., Golbik, R., Seliger, B., Spinka, M., Tittmann, K., Hubner, G., and

Jordan, F. (2001) Biochemistry 40, 1755–176350. Koga, J. (1995) Biochim. Biophys. Acta 1249, 1–1351. Frank, R. A.W., Titman, C.M., Pratap, J. V., Luisi, B. F., and Perham, R. N.

(2004) Science 306, 872–87652. Jordan, F., Nemeria, N. S., and Sergienko, E. (2005) Acc. Chem. Res. 38,

755–76353. Versees, W., Spaepen, S., Vanderleyden, J., and Steyaert, J. (2007) FEBS J.

274, 2363–237554. Versees, W., Spaepen, S., Wood, M. D. H., Leeper, F. J., Vanderleyden, J.,

and Steyaert, J. (2007) J. Biol. Chem. 282, 35269–3527855. Crosby, J., and Lienhard, G. E. (1970) J. Am. Chem. Soc. 92, 5707–571656. Jordan, F., Li, H., and Brown, A. (1999) Biochemistry 38, 6369–637357. Zhang, S., Zhou, L., Nemeria, N., Yan, Y., Zhang, Z., Zou, Y., and Jordan, F.

(2005) Biochemistry 44, 2237–2243




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Mrestani-Klaus, Gerhard Hübner and Stephan KönigTobias Werther, Michael Spinka, Kai Tittmann, Anja Schütz, Ralph Golbik, Carmen

Mycobacterium tuberculosisAcid Decarboxylase from -KetoαAmino Acids Allosterically Regulate the Thiamine Diphosphate-dependent

doi: 10.1074/jbc.M706569200 originally published online December 17, 20072008, 283:5344-5354.J. Biol. Chem.

10.1074/jbc.M706569200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

http://www.jbc.org/content/suppl/2007/12/19/M706569200.DC1

http://www.jbc.org/content/283/9/5344.full.html#ref-list-1

This article cites 54 references, 10 of which can be accessed free at


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M706569200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;283/9/5344&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/283/9/5344

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=283/9/5344&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/283/9/5344

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/suppl/2007/12/19/M706569200.DC1

http://www.jbc.org/content/283/9/5344.full.html#ref-list-1

http://www.jbc.org/

aminoacidsallostericallyregulatethethiamine …decarboxylation of benzoylformate to benzaldehyde,...

Documents