J. Ieuan Harris

Glyceraldehyde 3-phosphate dehydrogenases: I. The protein chains in glyceraldehyde 3-phosphate dehydrogenase from pig muscle

Glyceraldehyde 3-phosphate dehydrogenases have been studied by methods of structural protein chemistry. Derivatives of the pig muscle enzyme prepared by oxidation of the native enzyme with performic acid, and by carboxymethylation with iodo[1- 14 C]acetic acid in 8 M -urea, have been shown to contain four unique residues of cysteic acid, and of S-[1- 14 C]carboxymethylcysteine, respectively, indicating that the structural monomer in the enzyme contains four unique cysteines and that disulphide bridges do not contribute to its molecular structure. Amino-acid and end-group analysis, and a study of tryptic digests by peptide mapping techniques have shown further that the protein monomer in the pig muscle enzyme consists of a single polypeptide chain containing approximately 330 amino acid residues, corresponding to a molecular weight of 36,000 ± 1000. It is proposed that the active enzyme comprises four such identical protein chains, each of which contains one reactive cysteine and combines with one molecule of coenzyme (NAD), to form four structurally independent catalytic sites within the quaternary structure of the active tetramer.

Amino-acid Sequence of Glyceraldehyde 3-Phosphate Dehydrogenase from Lobster Muscle

The protein sub-unit of glyceraldehyde 3-phosphate dehydrogenase consists of a single chain of 333 amino-acid residues, corresponding to a molecular weight of 36,000. Amino-acid sequence studies show that the active enzyme-NAD complex with a molecular weight of 146,000 is composed of four protein chains of identical sequence.

Glyceraldehyde 3-phosphate dehydrogenase: Amino acid sequence of enzyme from baker's yeast

Glyceraldehyde 3-phosphate dehydrogenase* (GPDH) from the skeletal muscle of a wide range of organisms possesses very similar enzymic and molecular properties [ 1,2] . Within mammals the primary structure is very highly conserved [3] and the amino acid sequence around the essential thiol group is identical in all the species that have been studied [3,4] . More striking still is the considerable identity of the complete primary structure of enzyme from the muscle of such distantly related organisms as pig [S] and lobster [6]. Enzyme from baker’s yeast on the other hand shows significant differences in properties, notably in the mode of binding of NAD [7-91, and in the reactivity of the essential thiol group towards alkylating agents [ 10, Ill. The latter observation is the more remarkable since the amino acid sequence of a 17-residue tryptic peptide containing the essential thiol group is identical in the yeast and mammalian enzymes [lo] . As a further step towards the identification of the structural elements responsible for the differences in properties between the yeast and muscle enzymes, we report here a nearly complete amino acid sequence for the protein sub-unit of yeast GPDH. urea [ 131. Peptide fragments that accounted for the entire protein chain of 331 residues were obtained from: a) total tryptic digests of S-[2-i4C] carboxymethylated protein; b) similar tryptic digests of nonradioactive S-carboxymethylated protein in which methionine residues had been reacted with [ 1 -14C] iodoacetamide in 8 M urea at pH 3.0; c) tryptic digests of lysine-blocked (N-trifluoroacetyl-[ 151 and N-maleyl-[16]) S-[2-14C] carboxymethylated protein; d) cyanogen bromide cleavage (in 70% formic acid) of S-[2-14C] carboxymethylated protein, by suitable combinations of gel-filtration on Sephadex columns (G-50, G-75 and G-100) and paper electrophoresis/ chromatography (cf. [ 141). Peptide sequences were established by the dansylEdman method [ 171, and overlaps of many of the primary tryptic peptides were obtained from chymotrypsin and subtilisin digests of lysine-blocked fragments [ 181, and from pepsin digests of cyanogen bromide fragments. Overlaps that were not experimentally established have been assigned provisionally on the basis of apparent sequence homology with the pig and lobster muscle enzymes (cf. [5]). A detailed account of these procedures and results is available [19]. -

Evolutionary relationships in superoxide dismutase

Two distinct types of superoxide dismutase have been isolated from aerobic organisms. A blue-green Cu-Zn protein with a mol. wt of 32 000 and comprising two identical subunits has been obtained from a wide range of eukaryotes (e.g. [3,4]) while a pinkish-purple Mncontaining enzyme, also composed of identical subunits, has been isolated from eukaryotic mitochondria [S] and from prokaryotes [6,7]. In addition, an Fecontaining superoxide dismutase, similar to its Mncontaining counterpart, has been obtained from,!?. coli [8]. The prokaryotic Mn and Fe-containing enzymes are dimers with mol. wts of 40 000 and the mitochondrial Mn-enzyme, with a mol. wt of 80 000, is a tetramer. The subunits of the Mn and Fe-containing dismutases are of the same chain length, comprising approx. 185 to 190 residues [6-81. In addition they possess similar amino acid compositions and N-terminal sequences [9]. The Cu-Zn enzyme from bovine erythrocytes on the other band possesses a smaller subunit (of 150 residues [ 1 O]), with a different amino acid composition [ 111 and N-terminal sequence [9]. These results led Steinman and Hill [9] to suggest that the two classes of dismutase are probably composed of entirely different proteins of independent evolutionary origin. Insofar as the sequence comparisons of the respective protein chains were confined to the N-terminal segments of 25 to 30 residues, the possibility

Glyceraldehyde-3-phosphate dehydrogenase from Bacillus stearothermophilus

Glyceraldehyde-3-phosphate dehydrogenase has been isolated in pure crystalline form from a number of different sources including muscle, yeast and bacteria (for references see [l] ). Unique among NADlinked enzymes, crystalline muscle glyceraldehyde-3phosphate dehydrogenases contain up to four equivalents of firmly bound NAD and active enzyme-NAD complex is thus directly amenable to study by X-ray crystallographic methods. The apoenzyme, which is obtained by treating the holoenzyme with charcoal, is less stable and does not readily crystallise in the absence of added NAD [2] . The enzyme from a given species is composed of four polypeptide chains of essentially identical primary structure [ 1] and X-ray crystallographic studies with holoenzyme from lobster muscle [3,4] have shown that the individual polypeptide chains could be related in pairs within the tetramer. The holoenzyme crystals are orthorhombic (space group P212121) and the unit cell was found to consist of four tetrameric molecules. Subsequently, apolactic dehydrogenase was shown to possess a unit cell consisting of only one tetrameric molecule (with the monomer forming the crystallographic asymmetric unit) [5] , and it therefore seemed desirable to attempt to prepare crystals of NAD-free glyceraldehyde-3-phosphate dehydrogenase so that the crystallographic properties of apoenzyme crystals could also be investigated. Numerous attempts to crystallise apoenzyme from pig and lobster muscle have, however, been unsuccessful due, at least in part, to the instability of the apoenzyme in solution even at 4’ (cf. [6]). We have therefore sought to prepare suitably stable crystals of apoenzyme from a thermo-

Purification by affinity chromatography, properties and crystallisation of phosphofructokinase from thermophilic micro-organisms

Phosphofructokinase (PFK) catalyses the interconversion of fructose 6-phosphate (F-6P) and fructose 1,6-diphosphate (FDP) in a reaction involving ATP that represents the first unique step in glycolysis (for review see [1] ). The activity of PFK is regulated by a number of intracellular metabolites, and a study of its mode of action, and of the mechanisms through which it exercises control over the rate of glycolysis in the cell, is therefore of particular biochemical interest. PFK has been isolated in pure form from a number of different sources [1]. Enzymes from muscle and from yeast possess complex oligomeric structures (with large subunits with mol. wts of 85 000 [1] and 130 000 [2] respectively) that have hitherto proved difficult to study by methods of protein chemistry and X-ray crystallography. PFKs of microbial origin have been reported [1] to possess lower molecular weights and less complex subunit structures. More specifically earlier work from this laboratory (reference [3] and unpublished results of E. Kolb and J. I. Harris) has shown that PFK from B. stearothermophilus with a mol. wt of 130 000 is a thermostable tetrameric enzyme with four identical protein chains each comprising 300 amino acid residues. The enhanced stability of enzymes from thermophilic micro-organisms can be utilised to advantage in enzyme purification and in studies of enzyme structure and mechanism (cf [4, 5] ). For example, in an exploratory study of glycolytie enzymes from the extreme thermophile, T. aquaticus (grows optimally at 7 0 75°C), Hocking and Harris [6] showed that pure glyceraldehyde 3-phosphate dehydrogenase (GPDH) could be obtained in high yield from partially purified cell extracts by affinity chromatography on NADSepharose. Partial purification of T. aquaticus PFK was also achieved. Other examples of the use of Sepharose-linked nucleotides in the purification of dehydrogenases and kinases have since been reported (eg. [7, 8] ) and we now describe a method that allows pure GPDH and PFK to be obtained from cell extracts of T. aquaticus and of B. stearothermophilus by the combined application of NAD-Sepharose and AMPSepharose. Some molecular and kinetic properties of the purified PFKs are also given together with a method for crystallising the B. stearothermophilus enzyme for X-ray crystallographic analysis.

Amino acid sequence homology in alcohol dehydrogenase.

Thiol groups have been implicated in the catalytic activity of a number of dehydrogenases (for recent review see [ 11). Moreover, in the case of glyceraldehyde-3-phosphate dehydrogenase there is good evidence to show [2] that one cysteine residue per subunit (Cys-149 in the primary structure [3]) is directly involved in the catalytic reaction. The enzyme is inhibited by reaction of this cysteine with thiol group reagents and the amino acid sequence around the reactive cysteine is virtually identical in enzymes from a number a widely different species [4]. Alcohol dehydrogenase are also inhibited by reagents that react with thiol groups [5-71. In the horse liver enzyme inhibition by iodoacetate has been shown to be due to its selective reaction with one cysteine residue [8] identified as cysteine-46 in the enzyme [9]. Alcohol dehydrogenase from yeast is inhibited in a similar manner and although there is some similarity in the amino acid sequence around the reactive cysteines in the two proteins [6] it is not as extensive as that found between the corresponding glyceraldehyde-3phosphate dehydrogenases [4]. Moreover, differences in sequence close to the reactive cysteine in alcohol dehydrogenase have been shown to occur [ 10, 1 l] even within mammals. Alcohol dehydrogenase from yeast and liver differ in several respects. For example, the latter is a dimer with a subunit consisting of 374 amino acids [8,9] while the former is a tetrameric protein with a smaller subunit consisting of about 330 amino acids [6, 121. Although there is a similarity in the amino acid sequence around the respective reactive cysteines [6] comparison of the compositions of other tryptic peptides

Human placental 17β-oestradiol dehydrogenase. Sequence of a tryptic peptide containing an essential cysteine

Recent studies of the 17P-oestradiol dehydrogenase from human placenta have indicated that the active enzyme is a dimer with a sub-unit molecular weight of 34,000 [l-4]. Inhibition of the enzyme withNethylmaleimide (NEM) or p-chloromercuribenzoate (@MB) has been shown to be due to reaction of these reagents with at least two out of a total of twelve cysteines per mole. These cysteines are not reactive in the presence of coenzyme (NADP) but the substrate (17P-oestradiol) has no such protective effect ]41* In order to obtain more information concerning the subunit structure of the enzyme and the chemical environment of the essential thiol groups a study of the primary structure around these residues has been undertaken. A tryptic peptide containing the chemically modified cysteines essential for the enzymatic activity has been obtained and its amino acid sequence has revealed that the two reactive cysteines occur in a unique sequence in the primary structure of the enzyme.

Superoxide dismutase from Bacillus stearothermophilus: Crystallization and preliminary X-ray diffraction studies

Superoxide dismutase from Bacillus stearothermophilus is a dimeric manganese-containing enzyme with a molecular weight of 40,000. It has been crystallised in the monoclinic space group P21 with unit cell dimensions of a = 50 A, b = 70 A, c = 69 A and β = 111 ° 10′. The asymmetric unit appears to be the dimer.

Glyceraldehyde 3-Phosphate Dehydrogenase from an Extreme Thermophile, Thermus Aquaticus

Most of the concepts concerning the thermal stability, or instability, of biologically active systems are based upon studies carried out with organisms that live exclusively at moderate (<40°C) temperatures. Temperature is an important environmental factor in controlling the evolution and activities of organisms and although the existence of living cells is presumably limited to temperatures at which water exists as a liquid it is nevertheless well established that microbial life does exist at temperatures that approach the boiling point of water (1). As chemists and biochemists we are therefore left to ponder the mechanisms that allow biologically active molecules to be synthesised and to survive in a biologically active form at temperatures that are lethal to their counterparts in mesophiles.