Tài liệu Báo cáo khoa học: Site-directed mutagenesis of a loop at the active site of E1 (a2b2) of the pyruvate dehydrogenase complex A possible common sequence motif docx

Site-directed mutagenesis of a loop at the active site of E1 (a 2 b 2 ) of the pyruvate dehydrogenase complex A possible common sequence motif Markus Fries*, Hitesh J. Chauhan † , Gonzalo J. Domingo ‡ , Hyo-Il Jung § and Richard N. Perham Cambridge Centre for Molecular Recognition, Department of Biochemistry, University of Cambridge, UK Limited proteolysis of the pyruvate decarboxylase (E1, a 2 b 2 ) component of the pyruvate dehydrogenase (PDH) multi- enzyme complex of Bacillus stearothermophilus has indicated the importance for catalysis of a site (Tyr281-Arg282) in the E1a subunit (Chauhan, H.J., Domingo, G.J., Jung, H I. & Perham, R.N. (2000) Eur. J. Biochem. 267, 7158–7169). This site appears to be conserved in the a-subunit of hetero- tetrameric E1s and multiple sequence alignments suggest that there are additional conserved amino-acid residues in this region, part of a common pattern with the consensus sequence -YR-H-D-YR-DE This region lies about 50 amino acids on the C-terminal side of a 30-residue motif previously recognized as involved in binding thiamin diphosphate (ThDP) in all ThDP-dependent enzymes. The role of individual residues in this set of conserved amino acids in the E1a chain was investigated by means of site- directed mutagenesis. We propose that particular residues are involved in: (a) binding the 2-oxo acid substrate, (b) decarboxylation of the 2-oxo acid and reductive acety- lation of the tethered lipoyl domain in the PDH complex, (c) an Ôopen–closeÕ mechanism of the active site, and (d) phosphorylation by the E1-specific kinase (in eukaryotic PDH and branched chain 2-oxo acid dehydrogenase com- plexes). Keywords: pyruvate dehydrogenase; multienzyme complex; thiamin diphosphate; limited proteolysis; enzyme mecha- nism. The family of 2-oxo acid dehydrogenase (2-OADH) multi- enzyme complexes contains three members: the pyruvate dehydrogenase (PDH), the 2-oxoglutarate dehydrogenase (OGDH) and the branched-chain 2-oxo acid dehydrogenase (BCDH) complexes. These are responsible for the oxidative decarboxylation of pyruvate, 2-oxoglutarate and branched- chain 2-oxo acids, respectively, in each case generating the corresponding acyl-CoA and NADH. The complexes all occupy important positions in the metabolism of the cell and generally comprise three component enzymes: a 2-oxo acid decarboxylase (E1, EC 1.2.4), a dihydrolipoyl acyl- transferase (E2, EC 2.3.1), and dihydrolipoyl dehydroge- nase (E3, EC 1.8.1.4). E1 catalyses the first and irreversible step of the overall reaction, the thiamin-diphosphate (ThDP)-dependent oxidative decarboxylation of the 2-oxo acid, followed by the reductive acylation of a lipoyl prosthetic group covalently bound to a lysine residue in the lipoyl domain of the E2 chain. The reaction catalysed by E1 is rate-limiting for the overall activity of the complex [1,2], probably at the reductive acylation step [2,3]. The E2 component catalyses the transfer of the acyl group from the lipoyl-lysine group to CoA, and the cycle is completed by reoxidation of the resulting dihydrolipoyl group catalysed by E3, a flavoprotein, generating NADH and H + from NAD + . For recent reviews, see de Kok et al.andPerham [4,5]. In eukaryotic PDH and BCDH complexes, modulation of catalytic activity is achieved by phosphorylation- dephosphorylation of E1; an E1-specific kinase inactivates E1 by phosphorylating certain serine residues in the E1a component and activity is restored by the action of a specific phosphatase [6–8]. Depending on the organism and the type of 2-OADH complex, the E1 component exists either as a heterotetramer (a 2 b 2 ) or a homodimer (a 2 ). In the PDH complexes of Gram-negative bacteria and in all OGDH complexes, E1 is a homodimer; in PDH complexes of Correspondence to R. N. Perham, Department of Biochemistry, University of Cambridge, Sanger Building, Old Addenbrooke’s Site, 80 Tennis Court Road, Cambridge CB2 1GA, UK. Fax: + 44 1223 333667, Tel.: + 44 1223 333663, E-mail: r.n.perham@bioc.cam.ac.uk Abbreviations: BCDH, branched-chain 2-oxo acid dehydrogenase; DCPIP, 2,6-dichlorophenolindophenol; E1, 2-oxo acid decarboxylase; E1a,alphasubunitofE1;E1b, beta subunit of E1; E1p, pyruvate decarboxylase of PDH complex; E2p, dihydrolipoyl acetyltransferase; E3, dihydrolipoyl dehydrogenase; IPTG, isopropyl thio-b- D -galacto- side; OGDH, 2-oxoglutarate dehydrogenase; PSBD, peripheral subunit-binding domain; SPR, surface plasmon resonance; ThDP, thiamin diphosphate. Enzymes: pyruvate decarboxylase (EC 1.2.4.1); dihydrolipoyl acetyl- transferase (EC 2.3.1.12); dihydrolipoyl dehydrogenase (EC 1.8.1.4). *Present address: Cambridge Institute for Medical Research, Wellcome Trust/MRC Building, Box139 Addenbrooke’s Hospital, Hills Road, Cambridge CB2 2XY, UK. Present address: Adprotech Ltd, Chesterford Research Park, Little Chesterford, Saffron Walden, Essex CB10 1XL, UK. àPresent address: Seattle Biomedical Research Institute, 4 Nickerson Street, Seattle, WA 98109, USA. §Present address: Wolfson Institute for Biomedical Research, The Cruciform Building, Gower Street, London WC1E 6BT, UK. (Received 19 September 2002, revised 5 December 2002, accepted 20 December 2002) Eur. J. Biochem. 270, 861–870 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03444.x Gram-positive bacteria and eukaryotes and in BCDH complexes, E1 is a heterotetramer [9]. In all cases, E1 acts as a functional dimer with two active sites of equal catalytic efficiency [10,11] and two ThDP per E1, albeit bound with different affinities [11–13]. There is some evidence that the two active sites of pigeon breast muscle E1p work in an Ôalternating site mechanismÕ and that intersubunit inter- actions may play an essential part in the catalysis by, and regulation of, the enzyme [11]. This is consistent with the finding that phosphorylation of only one of two active sites in mammalian E1p leads to complete inactivation [7]. Crystal structures of the heterotetrameric E1s from the Pseudomonas putida [14] and human [15] BCDH complexes, both E1(a 2 b 2 ) heterotetramers, are available (Fig. 1). The E1a chain houses a conserved ThDP-binding site [16], the phosphate groups of ThDP being anchored by residues of this sequence motif through a bound Mg 2+ ion, and the aminopyrimidine end bound by the b-subunit [17]. Thus, both subunits are needed for a catalytically active enzyme. Limited proteolysis of E1 (a 2 b 2 ) from the PDH com- plexes of Bacillus subtilis [18] and pig heart [19] leads to cleavage of the E1a subunit, whereas the E1b subunit remains intact. A detailed study of the limited proteolysis of the E1 component of B. stearothermophilus PDH complex [20] identified Tyr281 (for chymotrypsin) and Arg282 (for trypsin) of the E1a subunit, two conserved residues in a loop at the entrance to the active site (Fig. 1), as the main cleavage sites. Cleavage of the E1a-subunit activated the enzyme, as determined by a decarboxylation assay of the free E1 in the presence of an artificial electron acceptor, but the catalytic activity of a reconstructed PDH complex was found to be substantially lowered. Multiple sequence alignments reveal additional conserved residues in the region of Tyr281 and Arg282. We describe here a study of the role of these conserved residues and compare the results with those of limited proteolysis in ascribing possible functions to them. Materials and methods Materials Restriction endonucleases were obtained from Pharmacia Biotech (St Albans, UK) or New England Biolabs (Hitchin, UK), Pfu DNA polymerase was from Stratagene (Cam- bridge, UK) and T4 DNA ligase from Promega Fig. 1. Crystal structure of heterotetrameric (a 2 b 2 )E1.(A) P. putida E1b [14]. (B) Human E1b [15]. The two a-subunits are shown in orange and red, the two b-subunits in light blue and dark blue. The cofactor ThDP- Mg 2+ , located at the interface between an a- and b-subunit, is indicated in spacefill. Residues 301–314 are not included in the human E1b structure, so that not all residues of the sequence motif could be highlighted. (C) Conserved active site motif derived from the crystal structure of P. putida E1b. (D) Conserved active site motif derived from the crystal structure of human E1b. 862 M. Fries et al. (Eur. J. Biochem. 270) Ó FEBS 2003 (Southampton, UK). Isopropyl thio-b- D -galactoside (IPTG) and phenylmethanesulfonyl fluoride were purchased from Melford Laboratories (Chelsworth, UK), bacterio- logical media from Beta Laboratory (West Molesely, UK) and Duchefa (Haarlem, the Netherlands), and ampicillin from Beecham Research Laboratories (Brentford, UK). Pyruvate, NAD + , ThDP, 2,6-dichlorophenolindophenol (DCPIP), coenzyme A and FAD were from Sigma (Poole, UK). Solid [2- 14 C]pyruvic acid, sodium salt (specific activity, 15.9 mCiÆmmol )1 ) was from New England Nuclear (Bos- ton, MA, USA). Centrifugal filter devices were purchased from Millipore. Bacterial strains and plasmids E. coli strain TG1recO and plasmids pKBstE1a and pKBstE1b, expressing genes encoding the B. stearother- mophilus E1a and E1b subunits, respectively, have been described previously [21]. E. coli strain BL21 (DE3) [B, F – , ompT, hsdS B (r B -,m B -), gal, dcm (DE3)] from Novagen (Madison, USA) was employed to express the B. stearo- thermophilus E3 gene (from pBSTNAV/E3) and E2p gene (from pETBstE2) [21] and a subgene encoding a di-domain that comprises the lipoyl domain, the peripheral subunit- binding domain and the linker between them (from pET11d2D and pET11ThDD). The subgene in plasmid pET11d2D encodes residues 1–171 of the wild-type B. ste- arothermophilus E2p chain [22]; plasmid pET11ThDD encodes the same di-domain but with a thrombin-cleavage site inserted in the linker region between the lipoyl and binding domains [23]. Lipoate protein ligase A was purified from E. coli BL21 (DE3) cells transformed with the plasmid pTM202 [24]. Recombinant DNA techniques and mutagenesis Recombinant DNA techniques were carried out as described elsewhere [25]. DNA fragments were isolated from gels using the QIAquick gel extraction kit, and plasmids were prepared by means of the Qiagen plasmid kit, both from Qiagen (Hilden, Germany). Site-specific mutations were introduced into the plasmid pKBstE1a using splicing-by-overlap exten- sion PCR [26]. The fidelity of the amplified DNA fragments was established by DNA sequence analysis after subcloning into the vector. All the sequences differed slightly from the original pKBstE1a plasmid [21] in the noncoding region following the stop codon. The differences (four C instead of three C at positions 1405–1407, and two additional G after position 1435) were the same in all instances, suggesting that there may have been errors in this part of the original pKBstE1a sequence. Protein purification Purification of wild-type and mutant E1s was carried out as described previously [21,27]. The E1aS283C mutant was purified with 1 m M dithiothreitol added to all buffers to prevent formation of disulfide bonds. E2 and E3 were purified as described elsewhere [28]. The E2p di-domain [29] and the thrombin-cleavable di- domain and the lipoyl domain subsequently released by thrombin treatment [23] were purified essentially as described elsewhere. The apo-forms of the lipoyl domain, di-domain and the E2 component were lipoylated by means of lipoate protein ligase A, also essentially as described elsewhere [22,30]. Enzyme assays The E1 component was assayed for catalytic activity by means of three different assays. DCPIP assay. This measures the rate of reduction of the artificial electron acceptor, DCPIP, by the E1 component [31]. The decrease in A 600 was monitored at a temperature of 30 °C. The assay mixture contained 0.2 m M ThDP, 2 m M MgCl 2 ,50l M DCPIP, 100 m M potassium phosphate, pH 7.0 and 20 lg of E1. The reaction was started by adding pyruvate (final concentration 400 l M ) after the assay mix had been incubated for 10 min at 30 °C. Reductive acetylation assay. This measures the reductive acetylation of the free lipoyl domain with [2- 14 C]pyruvate at 25 °C [30,32]. The assay mixture (350 lL) contained 0.2 m M ThDP, 1 m M MgCl 2 ,0.3mgÆmL )1 BSA, 20 m M potassium phosphate, pH 7.0, plus 3 nmol lipoyl domain and 5 lg E1. The reaction was started by addition of 14 C- labelled pyruvate (1 lCi, about 60 nmol) after the assay mix had been incubated for 10 min at 25 °C. PDH assay. This measures the rate of formation of NADH at 340 nm and 30 °C after the complex has been reconstituted from its individual components [1,33,34]. The assay contained 0.2 m M ThDP, 1 m M MgCl 2 ,2.6m M cysteine HCl, 2 m M pyruvate, 0.13 m M CoA, 50 m M potassium phosphate, pH 7.0. E1, E2 and E3 were added in molar ratios of E1(a 2 b 2 )/E2(a)/E3(a 2 )of3:1:3;the amount of E2 used was 0.5 lg, 1.0 lg, 1.5 lgand2.0lg. The reaction was started by adding the pyruvate and CoA after the PDH complex had been allowed to assemble in the assay mixture for at least 3 min at 30 °C. Specific activities for the reconstituted PDH complex are expressed as U per mg E2. All assays were normally carried out at least three times and the results are expressed as the mean ± SD. Determination of kinetic parameters K m values for pyruvate and ThDP were determined using the DCPIP assay by varying the concentrations of pyruvate from 0.2 l M to 4000 l M and the ThDP concentrations from 0.04 l M to 200 l M . When determining the kinetic param- eters for ThDP, the pyruvate concentration was kept at 1m M . Data were analysed and fitted to a Michaelis– Menten curve using the SIGMA PLOT software. Temperature-dependence of catalytic activity The dependence of the catalytic activity of E1 on tempera- ture was investigated using the DCPIP assay over a temperature range of 15 °Cto85°C. Assay mixtures were incubated for exactly 10 min at the relevant temperature before the reaction was started by the addition of pyruvate and then monitoring the decrease in A 600 . Ó FEBS 2003 Active site of E1 component of PDH complex (Eur. J. Biochem. 270) 863 Interaction of E1 with the peripheral subunit-binding domain Mixtures of E1 and di-domain were submitted to non- denaturing PAGE using the Pharmacia Phast System. The interaction of di-domain with E1 was also investigated using surface plasmon resonance detection (BIAcore, Pharmacia Biosensor AB), immobilizing the lipoyl domain by means of its lipoyl group. Both sorts of experiment were carried out as described in detail elsewhere [35]. Results Multiple sequence alignment of E1 components of 2-OADH complexes No obvious sequence homology can be detected between homodimeric forms of E1p and E1o. Likewise, amino-acid sequences of heterotetrameric (a 2 b 2 ) E1s do not readily align with sequences of homodimeric E1s. However, a sequence motif of about 30 amino acids, beginning with a highly conserved amino-acid triplet of GDG and ending withNN,hasbeenidentifiedascommontoallThDP- dependent enzymes [16], and confirmed as a ThDP-binding motif in the crystal structures of several ThDP-utilizing enzymes [17]. Aside from this motif, no active site residues other than those directly in contact with ThDP have been reported as conserved [36]. However, inspection of the available E1 sequences suggests that there may be another set of amino-acid residues common to the E1 (a 2 b 2 ) components of 2-OADH complexes. This consensus set, -Y/F/WR-H-D-YR-D/EE- in the E1a chains (Fig. 2), includes the Y281R282 site of limited proteolysis [20] and lies about 50 amino-acid residues to the C-terminal side of the ThDP-binding motif. The crystal structures of P. putida [14] and human [15] E1b show them as located on a loop region in Ela close in space to the cofactor ThDP-Mg 2+ (Fig. 1). Also included in this region are two serine residues in the Ela chain of mamma- lian PDH and BCDH complexes that serve as phosphory- lation sites 1 and 2, responsible for inhibiting E1 when they become phosphorylated. Other consistently conserved resi- dues in the aligned sequences appear to mark the beginning and end of the loop region: notably -PXXXE- (where X is generally a large aliphatic residue) at the N-terminal end, and -DP- (or -DHP-) at the C-terminal end (Fig. 2). Choice of mutations The importance of the loop region in the a-subunit of the B. stearothermophilus E1p component was highlighted by its susceptibility to limited proteolysis with trypsin and chymo- trypsin and the effects of the cleavages on catalytic activity [20]. In the present study, residues Phe266, Arg267 and Asp276 in the loop region of B. stearothermophilus E1a were replaced with alanine,as were Tyr281 and Arg282, the sites of cleavage with chymotrypsin and trypsin, respectively. Like- wise the neighbouring residue, Ser283, corresponding to a phosphorylation site in eukaryotic BCDH complexes, was also replaced with alanine. The Y281S/R282S mutation was constructed to test the effect of preserving a hydrophilic but uncharged character in the loop region, and the S283C mutant was a replacement of the serine hydroxyl group with the more nucleophilic thiol group (also potentially capable of subsequent chemical modification, if required). Effects of mutations on catalytic activity The catalytic activity of E1 mutants was investigated using three different assays: the DCPIP assay, which measures E1 activity bymonitoring the reduction ofDCPIP as an artificial electron acceptor instead of lipoamide; the reductive acety- lation assay, which measures the rate of reductive acetylation of the lipoyl group on a free lipoyl domain; and the PDH assay, which measures the overall activity of a PDH complex reconstituted from E2, E3 and the relevant mutant E1. In all assays, the E1aS283C and E1aF266A mutants behaved essentially the same as wild-type E1. In the DCPIP assay, the E1aY281A and E1aR282A mutants displayed a catalytic activity about twice that of wild-type E1. The single mutants E1aR267A and E1aD276A and the multiple mutants E1aY281A/R282A/S283A and E1aY281S/R282S displayed catalytic activities 2.5 and 3.5 times, respectively, that of wild-type E1. In contrast, the reductive acetylation assay showed no significant changes for the mutants compared with wild-type E1. However, in the PDH assay, the catalytic activity fell to about 50% of the wild-type value for the E1aD276A and E1aR282A mutants and to about 25% for the E1aY281A mutant. The double mutants showed even lower (< 20%) activity, but the most Fig. 2. Active site sequence alignment of the E1a chains of 2-OADH complexes. The numbering refers to the sequence of the E1a chain from B. stearothermophilus (ODPA_BACST). The sequences were taken from the SwissProt and NCBI databases and aligned using CLUSTALW 1.8. (*), identical residues in sequences in the alignment by CLUSTALW ; (:), conserved substitutions; (.), semiconserved substitu- tions. Residues conserved in all aligned sequences are printed in bold, the conserved sequence motif at the active site is underlined. P1 and P2 mark phosphorylation sites 1 and 2 (serine residues) in the E1a chains of eukaryotic PDH and BCDH complexes. ODPA and ODPT, E1a chain of PDH complexes with heterotetrameric E1(a 2 b 2 ); ODBA, E1a chain of heterotetrameric E1(a 2 b 2 )ofBCDHcomplexes. 864 M. Fries et al. (Eur. J. Biochem. 270) Ó FEBS 2003 significant drop in PDH complex activity (to < 10%) was experienced by the E1aR267A mutant (Table 1). Determination of kinetic parameters Kinetic parameters for the mutant E1s were determined for the substrate pyruvate and the cofactor ThDP using the DCPIP assay (Table 2). All activities recorded were good fits to Michaelis–Menten kinetics. A big increase in the K m for pyruvate was observed for the E1aR267A mutant, to nearly 300 l M compared with about 1 l M for wild-type E1. The K m for pyruvate was also found to be substantially increased to more than 50 l M for E1s carrying mutations at E1aY281. This was reflected in a major drop in the value of k cat /K m , in spite of the k cat being more than twice that for wild-type E1. In contrast, the E1aD276A and E1aR282A mutants more closely resembled wild-type E1, with a K m for pyruvate only about 10 times higher. The E1aF266A and E1aS283C mutants were almost identical to wild-type E1 (K m for pyruvate 1.2 l M and 2.1 l M , respectively). Wild-type E1was found to have an apparent K m for ThDP of 23 l M . Interestingly, the mutants with a significantly increased K m for pyruvate displayed markedly (5- to 20-fold) lower apparent K m s for ThDP than did wild-type E1. The E1aR267A mutant was found to have the lowest apparent K m for ThDP (< 1 l M ), followed by the E1aY281S/R282S, E1aY281A/R282A/S283A and E1aY281A/R282A mutants ( 3.0 l M ). The single mutations E1aR282A (K m ¼ 5.2 l M ), E1aY281A (K m ¼ 5.5 l M )andE1aS283C (K m ¼ 13 l M ) were less severe in their effects. Replacement of E1aF266 and E1aD276 with alanine had no significant influence on the K m for ThDP (K m ¼ 21 l M ). Dependence of E1 activity on temperature The temperature-dependence of the catalytic activity of E1 was measured using the DCPIP assay over a range of 25 to 85 °C (Fig. 3). The reaction mixture with E1 was incubated for exactly 10 min at the relevant temperature and the catalytic activity at that temperature was then determined. The inactivation temperatures for all the E1 mutants were found to be 5–10 °C below that of wild-type E1; mutants with a replacement of E1aY281 tended to lose catalytic activity at a slightly lower temperature than mutants retaining that residue. The E1aR267A mutant displayed a distinctly lower specific activity at higher temperatures than the other E1a mutants or than wild-type E1. The specific activity of the E1aD276A mutant increased steadily to reach twice that of wild-type E1 at 65 °C, after which the enzyme was inactivated. Binding of E1 to the peripheral subunit-binding domain of E2 The E1 component of the PDH complex of B. stearother- mophilus is bound to the E2 core mainly by interaction of its Table 2. Kinetic parameters of wild-type and mutant E1s determined by the DCPIP assay. Pyruvate ThDP k cat (s )1 ) K m (l M ) k cat /K m ( M )1 Æs )1 · 10 3 ) k cat (s )1 ) K m (l M ) k cat /K m ( M )1 Æs )1 · 10 3 ) Wild-type 0.48 ± 0.01 1.1 ± 0.1 427 0.46 ± 0.01 23 ± 2 20 E1aF266A 0.30 ± 0.01 1.2 ± 0.1 246 0.35 ± 0.01 21 ± 2 17 E1aR267A 1.62 ± 0.02 291 ± 14 6 1.25 ± 0.01 0.56 ± 0.04 2232 E1aD276A 0.71 ± 0.01 11.1 ± 0.5 64 0.91 ± 0.02 21 ± 2 42 E1aY281A 0.82 ± 0.01 63 ± 3 13 0.97 ± 0.01 5.5 ± 0.3 177 E1aR282A 0.55 ± 0.01 8.8 ± 0.6 63 0.62 ± 0.01 5.2 ± 0.4 121 E1aS283C 0.38 ± 0.01 2.1 ± 0.2 186 0.45 ± 0.01 13 ± 1 33 E1aY281A/R282A/S283A 1.23 ± 0.02 49 ± 4 25 1.39 ± 0.02 2.9 ± 0.2 481 E1aY281A/R282A 0.96 ± 0.01 56 ± 3 17 0.98 ± 0.01 3.0 ± 0.1 326 E1aY281S/R282S 1.21 ± 0.02 66 ± 4 19 1.30 ± 0.01 2.7 ± 0.1 475 Table 1. Specific activities of wild-type and mutant E1s in the various assays. DCPIP assay Reductive acetylation assay PDH assay UÆmg )1 %10 )3 UÆmg )1 %UÆmg )1 E2 % Wild-type 0.128 ± 0.004 100 2.6 ± 0.3 100 12.4 ± 0.3 100 E1aF266A 0.133 ± 0.001 104 – 9.6 ± 0.2 77 E1aR267A 0.325 ± 0.002 254 – 0.93 ± 0.03 8 E1aD276A 0.319 ± 0.005 249 – 5.67 ± 0.03 46 E1aY281A 0.268 ± 0.003 210 2.8 ± 0.3 108 3.30 ± 0.02 27 E1aR282A 0.230 ± 0.007 180 2.6 ± 0.3 100 6.10 ± 0.07 49 E1aS283C 0.156 ± 0.004 122 2.3 ± 0.1 90 10.5 ± 0.3 85 E1aY281A/R282A/S283A 0.452 ± 0.003 354 3.0 ± 0.1 115 1.75 ± 0.06 14 E1aY281A/R282A 0.279 ± 0.004 218 2.2 ± 0.1 85 1.74 ± 0.04 14 E1aY281S/R282S 0.41 ± 0.01 326 2.6 ± 0.1 100 1.83 ± 0.05 15 Ó FEBS 2003 Active site of E1 component of PDH complex (Eur. J. Biochem. 270) 865 E1b chains with the PSBD of the E2 chain. To investigate whether this interaction was impaired by the mutations in the E1a subunit, non-denaturing PAGE and surface plasmon resonance (SPR) detection were employed. To assess the binding, mutant E1s were mixed with a molar excess of di-domain (lipoyl domain and PSBD, joined by the natural linker region) and submitted to non-dena- turing PAGE. Binding of E1 to the di-domain will create an E1-di-domain band shifted towards a higher molecular mass than that of free E1. No difference in the ability of the mutant E1s to bind to the PSBD was observed (Fig. 4A), though it should be noted that small differences in affinity would not be detected by this Ôband-shiftÕ assay. SPR detection was used to determine the kinetic parameters of the interaction of the mutant E1s with the PSBD (Fig. 4B). For this purpose, lipoylated di-domain was immobilized on a BIAcore sensor chip by means of the lipoyl group on the lipoyl domain (see Materials and methods). E1 was then injected and allowed to interact with the exposed PSBD of the di-domain retained on the sensor surface. The k on (association rate constant), k off (dissociation rate constant) and K d (dissociation constant) of the E1–PSBD interaction were determined. The kinetic parameters for the interaction of wild-type E1 with PSBD reported earlier [35] are: k on , 3.27 · 10 )6 M )1 Æs )1 ; k off ,1.06· 10 )3 s )1 ; K d ,3.24· 10 )10 M )1 . The kinetic constants and K d values for mutant and wild-type E1s were found to be essentially identical (normally within ± 10%). Discussion Function of the mutated amino acids The E1aY281 and E1aR282 mutants of B. stearothermo- philus E1 show a higher V max in the DCPIP assay, no change in activity in the reductive acetylation assay and a significant drop in overall activity in the PDH assay when compared with wild-type E1. This is similar to the effects on E1 catalysis of limited proteolysis at Tyr281 and Arg282 [20]. Therefore, Fig. 4. Gels and SPR. (A) Non-denaturing polyacrylamide gel elec- trophoresis of wild-type and mutant E1s with di-domain at a 16-fold molarexcessofdi-domainoverE1.Lane1:E1wild-type;lane2:E1 wild-type + di-domain; lane 3: E1aF266A mutant; lane 4: E1aF266A mutant + di-domain; lane 5: E1aR267A mutant; lane 6: E1aR267A mutant + di-domain; lane 7: E1aD276A mutant; lane 8: E1aD276A mutant + di-domain. The gels for the other mutants were virtually identical (data not shown). (B) SPR sensorgrams for the interaction of wild-type and mutant E1s with the PSBD. Shown are the sensorgrams for wild-type E1, the E1aF266A, E1aR267A and E1aD276A mutants. The sensorgrams of the other mutants were essentially identical (data not shown). Fig. 3. Temperature-dependence of the activity in the DCPIP assay of wild-type and mutant E1s. 866 M. Fries et al. (Eur. J. Biochem. 270) Ó FEBS 2003 it is clear that those effects were not due simply to cleavage of the Ela chain backbone at these positions. In comparing the results from the different assays, one should be aware of the differences in reaction conditions. Thus, the reductive acetylation assay is performed with the lipoyl domain at a concentration of 8.5 l M and E1 at 95 n M , whereas in the PDH assay the local concentrations of E1 and lipoyl domains in the assembled complex are in the millimolar range and closer to equimolar. This may be of particular importance as the interaction between E1 and the lipoyl domain for the E. coli PDH complex is estimated to have a dissociation constant in the millimolar range or higher despite the fact that the K m is c. 20 l M [37,38]. Another difference in the assays is the concentration of the substrate, pyruvate. The different types of assay are reflected in the different rates; the PDH assay has a turnover number 100-fold higher than the DCPIP assay and 4800-fold higher than the reductive acetylation assay (referring to wild-type E1, Table 1). Thus, subtle changes in the kinetics of the E1 reaction may not be detected by the reductive acetylation assay, but will be detected by the PDH assay. The interaction of E1 (a 2 b 2 )withthePSBDwas unaffected by any of the mutations in E1a (Fig. 4). Therefore, any effects of the mutations on the catalytic activities of the PDH complex must be due to direct effects on the reactions catalysed by E1. The falls in PDH complex activity were most marked for the R267A mutant and all the Y281A mutants (Table 1). Given that there were no detectable effects on the catalytic activities in the reductive acetylationassaymeasuredwiththefreelipoyldomainas substrate (Table 1) and that the PDH complex activity was measured with pyruvate at a concentration of 2 m M ,well above the highest K m (300 l M ) recorded for any mutant E1 (R267A), the mutations appear to be affecting the ability of E1 to catalyse the reductive acetylation of the tethered lipoyl domain in the assembled PDH complex. It should be noted that Tyr281 and Arg282 are located at the mouth of the funnel-shaped active site of E1, at the bottom of which lies the ThDP cofactor some 20–25 A ˚ from the protein ÔsurfaceÕ [14]. Thus it may be that these mutations are affecting the efficacy of the recognition of the tethered lipoyl domain by E1, an essential prelude to reductive acetylation of the pendant lipoyl group [5,9,39]. The damaging effects of mutations in the b-turn region of the B. stearothermophilus lipoyl domain on its reductive acetylation by native E1 have been reported earlier [23,40]. The E1aR267A mutant of E1 exhibited a vastly increased K m for pyruvate (300 l M ) compared with wild- type E1 (1 l M ). Interaction of the positively charged side chain of Arg267 with the negatively charged carboxyl group of the 2-oxo acid substrate thus appears to be very likely. The crystal structures of E1 show the corresponding arginine residues in P. putida [14] and human [15] E1s to be in a suitable position to participate in such an electrostatic interaction (Fig. 1). In contrast, given the K m (9 l M )for pyruvate shown by the E1aR282A mutant and its other modest differences from wild-type E1 (Tables 1 and 2), a direct role for Arg282 in the E1 reaction appears improb- able. This too is consistent with the E1 crystal structures in that the arginine residues corresponding to E1aR282 of the B. stearothermophilus E1 are not pointing towards the active site, identified by the C2-carbon of the cofactor ThDP. Although Arg282 is even more strictly conserved in sequence alignments than Tyr281, the latter is more important for the reaction catalysed by B. stearothermophi- lus E1p. Apart from the K m for pyruvate being markedly increased for all the Tyr281 mutants (Table 2), these enzymes started to lose catalytic activity at lower temper- atures than the other mutants or wild-type E1 (Fig. 3). This suggests that Tyr281 has some part to play in conferring thermal stability. In the crystal structure of pyruvate decarboxylase (EC 4.1.1.1) from Zymomonas mobilis,a tyrosine sidechain is found in the active site and is suitably oriented to take part in catalysis, as model building with reaction intermediates suggests [41]. It was speculated that the tyrosine might form a hydrogen bond with the carboxy group of 2-(2-hydroxypropionyl)-ThDP and contribute to the stabilization of its negative charge. The conserved Tyr281 in the B. stearothermophilus E1a chain may play a similar role in the E1 component of this and other 2-oxo acid dehydrogenase (Fig. 1) complexes. The E1aD276A mutant displayed moderate changes in its kinetic properties compared with wild-type E1, some- what similar to those observed for the R282A mutant. Adjacent to E1aD276 in the primary structure of the E1a chain from B. stearothermophilus E1p is another aspartate residue, E1aD277. This is replaced by proline in many E1a chains (Fig. 2), and has not been examined here. However, the corresponding residue, E1aD296 of rat E1b, has been reported to be essential [42], so further experiments may be justified. The E1aF266A and E1aS283C mutants behaved almost identically to wild-type E1 in all aspects investigated; a crucial role for Phe266 and Ser283 in the reaction mechanism can thus be excluded. A common sequence motif The high conservation of amino-acid residues in the sequence spanning positions 255–295 discussed above was noted earlier [43], including the phosphorylation sites P1 and P2 (Fig. 2), but no particular role was assigned to them. The importance of three residues in the E1a component of rat E1b, namely, E1aR288, E1aH292 and E1aD296 of rat E1b, was recognized by alanine mutagenesis [42]. These residues were among 10 in the neighbourhood of phos- phorylation site 1 (E1aS293) that were examined, and their replacement with alanine resulted in totally inactive enzymes. They correspond to residues Arg267, His271 and Asp276 in the proposed YR–H–D–YR–DE sequence motif (Fig. 2). Our results indicate that Arg267 of B. stearothermophilus E1p is involved in binding the 2-oxo acid substrate and that E1aTyr281 and, to a lesser extent, E1aAsp276 and E1aArg282, have some effect on the decarboxylation of pyruvate and the reductive acetylation of the tethered lipoyl domain in the active PDH complex (although the replacement of none of these residues caused complete inhibition). Other experiments (M. Fries & R. N. Perham, unpublished work) indicate that E1aHis271 has a crucial part to play in the decarboxylation of pyruvate, probably by serving to stabilize the energetically unfavou- rable dianion formed after nucleophilic attack of ThDP at Ó FEBS 2003 Active site of E1 component of PDH complex (Eur. J. Biochem. 270) 867 the 2-oxo group of the substrate [for a detailed formulation of the mechanism to date, see [44]). The importance of this region is emphasized by the reports of clinically deleterious mutations in the human E1a chain; for example, the replacement of His263 (equivalent to His271 in B. stearothermophilus E1a, Fig. 2) with leucine is associated with a very low PDH complex activity, as is the replacement of Arg273 (equivalent to Arg282) in B. stearo- thermophilus E1a with cysteine [45,46]. It is interesting to note that the R282A mutation in B. stearothermophilus E1a had only a modest effect on the E1 catalytic activity in vitro (Tables 1 and 2), perhaps because the amino-acid replacing the arginine is different. Two phosphorylation sites in eukaryotic E1a chains are located within the region of the sequence motif outlined in Fig. 2. In addition to the E1aR288A mutant of rat E1b being inactive, it was incapable of becoming phosphory- lated, suggesting that this residue may be involved in the interaction with the E1 kinase [42]. Studies on synthetic peptides as substrates for mammalian pyruvate dehydro- genase kinase have indicated that an acidic residue to the C-terminal side of phosphorylation site 1 (Ser293) is also an important specificity determinant for the kinase [47]. This might be true similarly for phosphorylation site 2 (Ser300). There are conserved acidic residues to the C-terminal side of both phosphorylation sites 1 and 2 in the Ela sequence alignment (Fig. 2). Non-equivalence of active sites has been observed in the crystal structure of the thiamin-dependent yeast pyruvate decarboxylase; one active site was found to be in an open conformation, with two loop regions disordered, whereas in the other these loop regions were well-ordered and shielded the active site from the bulk solution [48]. The loop housing the conserved sequence motif in B. stearothermophilus E1p is exposed and flexible, as indicated by limited proteolysis; moreover, it appears to adopt two different conformations, one susceptible to proteolysis and the other not [20]. Thus, this loop in the B. stearothermophilus E1p might take part in a similar Ôopen-closeÕ mechanism. If in the mutant E1s the loop region became more disordered, it might not function properly as a lid for the active site, thereby granting easier access for the cofactor ThDP to its binding site. The lower thermal stability of the mutant E1s and the increased V max in the DCPIP assay, which could be due to easier access of DCPIP to the active site or to a more ready release of product, would be consistent with this idea. We have looked at the recently determined crystal structure of the dimeric E1p from the E. coli PDH complex [49] but have been unable to locate a similar sequence of amino acids in a comparable position with respect to the active site. Some conclusions about the importance of individual residues in the E. coli E1p have been drawn [49], but a detailed comparison of the two types of E1 (dimeric and heterotetrametric) will have to be the subject of further investigation. Acknowledgements This work was supported by a research grant from the Biotechnology and Biological Sciences Research Council (to R. N. P.). M. F. is grateful to the Verband der Chemischen Industrie and the Cusanu- swerk for a studentship and Clare Hall for additional support, H. J. C. thanks the BBSRC for the award of an earmarked Research Studentship, and H I. J. is grateful to the Cambridge Overseas Trust, St John’s College, Cambridge and the Department of Biochemistry, University of Cambridge for financial support. References 1. Danson, M.J., Fersht, A.R. & Perham, R.N. (1978) Rapid intramolecular coupling of active sites in the pyruvate dehydro- genase complex of Escherichia coli: mechanism for rate enhancement in a multimeric structure. Proc. Natl Acad. Sci. USA 75, 5386–5390. 2. Cate, R.L., Roche, T.E. & Davis, L.C. (1980) Rapid intersite transfer of acetyl groups and movement of pyruvate dehydro- genase component in the kidney pyruvate dehydrogenase complex. J. Biol. Chem. 255, 7556–7562. 3. Berg, A., Westphal, A.H., Bosma, H.J. & de Kok, A. (1998) Kinetics and specificity of reductive acylation of wild-type and mutated lipoyl domains of 2-oxo-acid dehydrogenase complexes from Azotobacter vinelandii. Eur J. Biochem. 252, 45–50. 4. de Kok, A., Hengeveld, A.F., Martin, A. & Westphal, A.H. (1998) The pyruvate dehydrogenase multi-enzyme complex from Gram- negative bacteria. Biochim. Biophys. Acta 1385, 353–366. 5. Perham, R.N. (2000) Swinging arms and swinging domain in multifunctional enzymes: catalytic machines for multistep reactions Annu. Rev. Biochem. 69, 961–1004. 6. Barrera, C.R., Namihira, G., Hamilton, L., Munk, P., Eley, M.H., Linn,T.C.&Reed,L.J.(1972)a-Keto acid dehydrogenase complexes. XVI. Studies on the subunit structure of the pyruvate dehydrogenase complexes from bovine kidney and heart. Arch. Biochem. Biophys. 148, 343–358. 7. Yeaman, S.J., Hutcheson, E.T., Roche, T.E., Pettit, F.H., Brown, J.R., Reed, L.J., Watson, D.C. & Dixon, G.H. (1978) Sites of phosphorylation on pyruvate dehydrogenase from bovine kidney and heart. Biochemistry 17, 2364–2370. 8. Cook, K.G., Bradford, A.P., Yeaman, S.J., Aitken, A., Fearnley, I.M. & Walker, J.E. (1984) Regulation of bovine kidney branched- chain 2-oxoacid dehydrogenase complex by reversible phosphorylation. Eur J. Biochem. 145, 587–591. 9. Perham, R.N. (1991) Domains, motifs, and linkers in 2-oxo acid dehydrogenase multienzyme complexes: a paradigm in the design of a multifunctional protein. Biochemistry 30, 8501–8512. 10. Khailova, L.S. & Korochkina, L.G. (1982) Determination of the number of active-centers in the pyruvate dehydrogenase component of the pyruvate dehydrogenase complex from pigeon breast muscle. Biochem. Int. 5, 525–532. 11. Khailova, L.S., Korochkina, L.G. & Severin, S.E. (1989) Organization and functioning of muscle pyruvate dehydrogenase active centers. Ann. NY Acad. Sci. 573, 36–54. 12. Yi, J., Nemeria, N., McNally, A., Jordan, F., Machado, R.S. & Guest, J.R. (1996) Effect of substitutions in the thiamin diphosphate-magnesium fold on the activation of the pyruvate dehydrogenase complex from Escherichia coli by cofactors and substrate. J. Biol. Chem. 271, 33192–33200. 13. Jordan,F.,Nemeria,N.,Guo,F.,Baburina,I.,Gao,Y.,Kahy- aoglu, A., Li, H., Wang, J., Yi, J., Guest, J.R. & Furey, W. (1998) Regulation of thiamin diphosphate-dependent 2-oxo acid decarboxylases by substrate and thiamin diphosphate. Mg (II) – evidence for tertiary and quaternary interactions. Biochim. Biophys. Acta 1385, 287–306. 14. Aevarsson, A., Seger, K., Turley, S., Sokatch, J.R. & Hol, W.G.J. (1999) Crystal structure of 2-oxoisovalerate and dehydrogenase and the architecture of 2-oxo acid dehydrogenase multienzyme complexes. Nat. Struct. Biol. 6, 785–792. 15. Aevarsson, A., Chuang, J.L., Wynn, R.M., Turley, S., Chuang, D.T. & Hol, W.G.J. (2000) Crystal structure of human 868 M. Fries et al. (Eur. J. Biochem. 270) Ó FEBS 2003 branched-chain alpha-ketoacid dehydrogenase and the molecular basis of multienzyme complex deficiency in maple syrup urine disease. Structure 8, 277–291. 16. Hawkins, C.F., Borges, A. & Perham, R.N. (1989) A common structural motif in thiamin pyrophosphate-binding enzymes. FEBS Lett. 255, 77–82. 17. Muller,Y.A.,Lindqvist,Y.,Furey,W.,Schulz,G.E.,Jordan,F.& Schneider, G. (1993) A thiamin diphosphate binding fold revealed by comparison of the crystal structures of transketolase, pyruvate oxidase and pyruvate decarboxylase. Structure 1, 95–103. 18. Lowe, P.N., Hodgson, J.A. & Perham, R.N. (1985) Limited proteolysis of the pyruvate dehydrogenase multienzyme complex of Bacillus subtilis. Biochem. J. 225, 249–253. 19. Koike, K., Urata, Y. & Goto, S. (1992) Proteinase-catalyzed activation of porcine heart muscle pyruvate dehydrogenase and identification of its cleavage site. Biochim. Biophys. Acta 1118, 223–230. 20. Chauhan,H.J.,Domingo,G.J.,Jung,H.I.&Perham,R.N.(2000) Sites of limited proteolysis in the pyruvate decarboxylase component of the pyruvate dehydrogenase multienzyme complex of Bacillus stearothermophilus and their role in catalysis. Eur J. Biochem. 267, 7158–7169. 21. Lessard, I.A. & Perham, R.N. (1994) Expression in Escherichia coli of genes encoding the E1 alpha and E1 beta subunits of the pyruvate dehydrogenase complex of Bacillus stearothermophilus and assembly of a functional E1 component (alpha 2 beta 2) in vitro. J. Biol. Chem. 269, 10378–10383. 22. Jung, H.I., Bowden, S.J., Cooper, A. & Perham, R.N. (2002) Thermodynamic analysis of the binding of component enzymes in the assembly of the pyruvate dehydrogenase multienzyme complex of Bacillus stearothermophilus. Protein Sci. 11, 1091–1100. 23. Wallis, N.G., Allen, M.D., Broadhurst, R.W., Lessard, I.A. & Perham, R.N. (1996) Recognition of a surface loop of the lipoyl domain underlies substrate channelling in the pyruvate dehydrogenase multienzyme complex. J. Mol Biol. 263, 463–474. 24. Morris, T.W., Reed, K.E. & Cronan, J.E. Jr (1994) Identification of the gene encoding lipoate-protein ligase A of Escherichia coli. Molecular cloning and characterization of the lplA gene and gene product. J. Biol. Chem. 269, 16091–16100. 25. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seid- man, J.G., Smith, J.A. & Struhl, K. (1992) Current Protocols in Molecular Biology. Wiley, J. & Sons, New York, USA. 26. Horton, R.M., Ho, S.N., Pullen, J.K., Hunt, H.D., Cai, Z. & Pease, L.R. (1993) Gene splicing by overlap extension. Methods Enzymol. 217, 270–279. 27. Lessard, I.A. & Perham, R.N. (1995) Interaction of component enzymes with the peripheral subunit-binding domain of the pyruvate dehydrogenase multienzyme complex of Bacillus stearothermophilus: stoichiometry and specificity in self-assembly. Biochem. J. 306, 727–733. 28. Lessard, I.A., Domingo, G.J., Borges, A. & Perham, R.N. (1998) Expression of genes encoding the E2 and E3 components of the Bacillus stearothermophilus pyruvate dehydrogenase complex and the stoichiometry of subunit interaction in assembly in vitro. Eur J. Biochem. 258, 491–501. 29. Hipps, D.S. & Perham, R.N. (1992) Expression in Escherichia coli of a sub-gene encoding the lipoyl and peripheral subunit-binding domains of the dihydrolipoamide acetyltransferase component of the pyruvate dehydrogenase complex of Bacillus stearothermophilus. Biochem. J. 283, 665–671. 30. Jones, D.D., Horne, H.J., Reche, P.A. & Perham, R.N. (2000) Structural determinants of post-translational modification and catalytic specificity for the lipoyl domains of the pyruvate dehydrogenase multienzyme complex of Escherichia coli. J. Mol Biol. 295, 289–306. 31. Khailova, L.S., Bernkhardt, R. & Khiubner, G. (1977) Study of the kinetic mechanism of the pyruvate-2,6-dichloro- phenolindophenol reductase activity of muscle pyruvate dehydrogenase. Biokhimiia 42, 113–117. 32. Packman, L.C., Perham, R.N. & Roberts, G.C. (1984) Domain structure and 1 H-n.m.r. spectroscopy of the pyruvate dehydro- genase complex of Bacillus stearothermophilus. Biochem. J. 217, 219–227. 33. Henderson, C.E. & Perham, R.N. (1980) Purificaton of the pyruvate dehydrogenase multienzyme complex of Bacillus stearothermophilus and resolution of its four component polypeptides. Biochem. J. 189, 161–172. 34. Domingo, G.J., Chauhan, H.J., Lessard, I.A., Fuller, C. & Per- ham, R.N. (1999) Self-assembly and catalytic activity of the pyruvate dehydrogenase multienzyme complex from Bacillus stearothermophilus. Eur J. Biochem. 266, 1136–1146. 35. Lessard, I.A., Fuller, C. & Perham, R.N. (1996) Competitive interaction of component enzymes with the peripheral subunit- binding domain of the pyruvate dehydrogenase multienzyme complex of Bacillus stearothermophilus: kinetic analysis using surface plasmon resonance detection. Biochemistry 35, 16863– 16870. 36. Hasson, M.S., Muscate, A., McLeish, M.J., Polovnikova, L.S., Gerlt, J.A., Kenyon, G.L., Petsko, G.A. & Ringe, D. (1998) The crystal structure of benzoylformate decarboxylase at 1.6 A resolution: diversity of catalytic residues in thiamin diphosphate- dependent enzymes. Biochemistry 37, 9918–9930. 37. Graham, L.D., Packman, L.C. & Perham, R.N. (1989) Kinetics and specificity of reductive acylation of lipoyl domains from 2-oxo acid dehydrogenase multienzyme complexes. Biochemistry 28, 1574–1581. 38. Graham, L.D. & Perham, R.N. (1990) Interactions of lipoyl domains with the E1p subunits of the pyruvate dehydrogenase multienzyme complex from Escherichia coli. FEBS Lett. 262, 241–244. 39. Jones, D.D., Stott, K.M., Reche, P.A. & Perham, R.N. (2001) Recognition of the lipoyl domain is the ultimate determinant of substrate channelling in the pyruvate dehydrogenase multienzyme complex. J. Mol. Biol. 305, 49–60. 40. Wallis, N.G. & Perham, R.N. (1994) Structural dependence of post-translational modification and reductive acetylation of the lipoyl domain of the pyruvate dehydrogenase multienzyme complex. J. Mol Biol. 236, 209–216. 41. Dobritzsch, D., Konig, S., Schneider, G. & Lu, G. (1998) High resolution crystal structure of pyruvate decarboxylase from Zymomonas mobilis. Implications for substrate activation in pyruvate decarboxylases. J. Biol. Chem. 273, 20196–20204. 42. Hawes, J.W., Schnepf, R.J., Jenkins, A.E., Shimomura, Y., Popov, K.M. & Harris, R.A. (1995) Roles of amino acid residues surrounding phosphorylation site 1 of branched-chain alpha- ketoacid dehydrogenase (BCKDH) in catalysis and phosphorylation site recognition by BCKDH kinase. J. Biol. Chem. 270, 31071–31076. 43. Wexler, I.D., Hemalatha, S.G. & Patel, M.S. (1991) Sequence conservation in the a and b subunits of pyruvate dehydrogenase and its similarity to branched-chain a-keto acid dehydrogenase. FEBS Lett. 282, 209–213. 44. Pan, K. & Jordan, F. (1998) D,L-S-Methyllipoic acid methyl ester, a kinetically viable model for S-protonated lipoic acid as the oxidizing agent in reductive acyl transfers catalyzed by the 2-oxoacid dehydrogenase multienzyme complexes. Biochemistry 37, 1357–1364. 45. Mullinax, T.R., Stepp, L.R., Brown, J.R. & Reed, L.J. (1985) Synthetic peptide substrates for mammalian pyruvate dehydro- genase kinase and pyruvate dehydrogenase phosphatase. Arch. Biochem. Biophys. 243, 655–659. Ó FEBS 2003 Active site of E1 component of PDH complex (Eur. J. Biochem. 270) 869 46. Chun, K., MacKay, N., Petrova-Benedict, R. & Robinson, B.H. (1993) Mutations in the X-linked E1a subunit of pyruvate dehydrogenase leading to deficiency of the pyruvate dehydro- genase complex. Human Mol. Genet. 2, 449–454. 47. Dahl, H.H.M., Hansen, L.L., Brown, R.M., Danks, D.M., Rogers, J.G. & Brown, G.K. (1992) X-linked pyruvate dehydrogenase E1a subunit deficiency in heterozygous females: variable manifestation of the same mutations. J. Inher. Metab. Dis. 15, 835–847. 48. Lu, G., Dobritzsch, D., Baumann, S., Schneider, G. & Konig, S. (2000) The structural basis of substrate activation in yeast pyruvate decarboxylase. A crystallographic and kinetic study. Eur J. Biochem. 267, 861–868. 49. Arjunan, P., Nemeria, N., Brunskill, A., Chandrasekhar, K., Sax, M.,Yan,Y.,Jordan,F.,Guest,J.R.&Furey,W.(2002)Structure of the pyruvate dehydrogenase multienzyme complex E1 component from Escherichia coli at 1.85 angstrom resolution. Biochemistry 41, 5213–5221. 870 M. Fries et al. (Eur. J. Biochem. 270) Ó FEBS 2003 . Site- directed mutagenesis of a loop at the active site of E1 (a 2 b 2 ) of the pyruvate dehydrogenase complex A possible common sequence motif Markus. the 2-oxo acid substrate and that E1aTyr281 and, to a lesser extent, E1aAsp276 and E1aArg282, have some effect on the decarboxylation of pyruvate and the