Quaternary structure of higher plant glyceraldehyde-3-phosphate dehydrogenases.

1. NAD(P)+-induced changes in the aggregational state of prepurified NADP-linked glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.13) were used to isolate the enzyme from Spinacia oleracea, Pisum sativaum and Hordeum vulgare. Each of the three plant species contains two separate isoenzymes. Isoenzyme 1 (fast moving during conventional electrophoresis) precipitates with the ammonium sulfate fraction 55--70% saturation. It shows two separate subunits in dodecylsulfate gels, which are probably arranged as A2B2 in the native enzyme molecule. Isoenzyme 2 (slow moving during conventional electrophoresis) precipitates with the ammonium sulfate fraction 70--95%. It contains a sigle subunit of the same Mr as subunit A in isoenzyme 1 and is apparently a tetramer (A4). The molecular weights of subunits A/B for spinach, peas and barley were determined as 38,000/40,000, 38,000/42,000 and 36,000/39,000 respectively. 2. The NAD-specific glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.12) was purified from Spinacia oleracea and Pisum sativum by affinity chromatography on blue Sepharose CL-6B. The enzyme from both plant species is shown to be a tetramer of subunits with Mr 39,000. 3. The present findings contrast with heterogeneous results obtained previously by other authors. These results suggested that there are considerable interspecific differences in the quaternary structure of glyceraldehyde-3-phosphate dehydrogenases from higher plants.     

The evolutionary origin of red algae as deduced from the nuclear genes encoding cytosolic and chloroplast glyceraldehyde-3-phosphate dehydrogenases from Chondrus crispus

Algae are a heterogeneous group of photosynthetic eukaryotes traditionally separated into three major subdivisions: rhodophytes, chlorophytes, and chromophytes. The evolutionary origin of rhodophytes or red algae and their links to other photosynthetic and nonphotosynthetic eukaryotes have been a matter of much controversy and speculation. Here we present the first cDNAs of nuclear protein genes from red algae: Those encoding cytosolic and chloroplast glyceraldehyde-3-phosphate dehydrogenases (GAPDH) from Chondrus crispus. A phylogenetic analysis including GAPDH gene sequences from a number of eukaryotic taxa, cyanobacteria, and purple bacteria suggests that chloroplasts and rhodoplasts together form a monophyletic group of cyanobacterial descent and that rhodophytes separated from chlorophytes at about the same time as animals and fungi. The composite GAPDH tree further demonstrates that chloroplast and cytosolic GAPDH genes are closely related to their homologs in cyanobacteria and purple bacteria, respectively, the presumptive ancestors of chloroplasts and mitochondria, thereby firmly establishing the endosymbiotic origin of these nuclear genes and their fixation in eukaryotic cells before the rhodophyte/chlorophyte separation. The present data are in conflict with phylogenetic inferences based on plastid-encoded rbcL sequences supporting a polyphyletic origin of rhodoplasts and chloroplasts. Comparison of rbcL to GAPDH phylogenies suggests that rbcL trees may be misleading because they are composed of branches representing ancient duplicated (paralogous) genes. Correspondence to: R. Cerff     

Nucleotide sequence of complementary DNA and derived amino acid sequence of murine complement protein C3

The nucleotide sequences coding for murine complement component C3 have been determined from a cloned genomic DNA fragment and several overlapping cloned complementary DNA fragments. The amino acid sequence of the protein was deduced. The mature beta and alpha subunits contain 642 and 993 amino acids respectively. Including a 24 amino acid signal peptide and four arginines in the beta-alpha transition region, which are probably not contained in the mature protein, the unglycosylated single chain precursor protein preproC3 would have a molecular mass of 186 484 Da and consist of 1663 amino acid residues. The C3 messenger RNA would be composed of a 56 +/- 2 nucleotide long 5' non-translated region, 4992 nucleotides of coding sequence, and a 3' non-translated region of 39 nucleotides, excluding the poly A tail. The beta chain contains only three cysteine residues, the alpha chain 24, ten of which are clustered in the carboxy terminal stretch of 175 amino acids. Two potential carbohydrate attachment sites are predicted for the alpha chain, none for the beta chain. From a comparison with human C3 cDNA sequence (of which over 80% has been determined) an extensive overall sequence homology was observed. Human and murine preproC3 would be of very similar length and share several noteworthy properties: the same order of the subunits in the precursor, the same basic residue multiplet in the beta-alpha transition region, and a glutamine residue in the thioester region. The equivalent position of the known factor I cleavage sites in human C3 alpha could be located in the murine C3 alpha chain and the size and sequence of the resulting peptide were deduced. A comparison of the amino acid sequences of murine C3 and human alpha 2-macroglobulin is given. Several areas of strong sequence homology are observed, and we conclude that the two genes must have evolved from a common ancestor.     

Nucleotide sequence and derived amino acid sequence of a cDNA encoding human muscle carbonic anhydrase

We report the nucleotide (nt) sequence of a full length cDNA clone, pCA15, which encodes the human muscle-specific carbonic anhydrase, CAIII. pCA15 identifies a 1.7-kb mRNA, which is present at high levels in skeletal muscle, at much lower levels in cardiac and smooth muscle and which appears to be developmentally regulated. The CAIII mRNA is distinguished by a 887-nt long 3'-untranslated region, containing two AAUAAA signal sequences and is longer than either of the mRNAs encoding the erythrocyte CAs, CAI and CAII, which each have relatively shorter 3'-untranslated regions, 360 and 670 nt long, respectively. The derived amino acid (aa) sequence for human CAIII shows 85% homology with ox CAIII, 62% homology with human CAII and 54% with human CAI when simple pairwise aa comparisons are made. We describe an allelic variation at a TaqI restriction site for CAIII which occurs at high frequency in the European population.     

Engineering the allosteric properties of archaeal non-phosphorylating glyceraldehyde-3-phosphate dehydrogenases

The archaeal non-phosphorylating glyceraldehyde-3-phosphate dehydrogenase (GAPN, EC 1.2.1.9) is a highly allosteric enzyme activated by glucose 1-phosphate (Glc1P). Recent kinetic analyses of two GAPN homologs from Sulfolobales show different allosteric behaviors toward the substrate glyceraldehyde-3-phosphate (GAP) and the allosteric effector Glc1P. In GAPN from Sulfolobus tokodaii (Sto-GAPN), Glc1P-induced activation follows an increase in affinity for GAP rather than an increase in maximum velocity, whereas in GAPN from Sulfolobus solfataricus (Sso-GAPN), Glc1P-induced activation follows an increase in maximum velocity rather than in affinity for GAP. To explore the molecular basis of this difference between Sto-GAPN and Sso-GAPN, we generated 14 mutants and 2 chimeras. The analyses of chimeric GAPNs generated from regions of Sto-GAPN and Sso-GAPN indicated that a 57-residue module located in the subunit interface was clearly involved in their allosteric behavior. Among the point mutations in this modular region, the Y139R variant of Sto-GAPN no longer displayed a sigmoidal K-type-like allostery, but instead had apparent V-type allostery similar to that of Sso-GAPN, suggesting that the residue located in the center of the homotetramer critically contributes to the allosteric behavior.     

Structural basis for the thermal stability of glyceraldehyde-3-phosphate dehydrogenases

The amino acid sequences of two thermophilic and five mesophilic glyceraldehyde-3-phosphate dehydrogenases have been compared with the known three-dimensional structure of this enzyme to determine the factors responsible for thermal stability. The changes are greatest in the S-loop regions at the center of the tetramer, which show a quantitative increase in hydrophobicity and polarity that can strengthen subunit interactions in a complementary manner. The S-loops also show increases in residue volume and bulk that may indicate a tighter packing at the molecular center. In addition, there are changes in the secondary structural parameters indicating that the helices, in particular, may be more stable in the thermophilic proteins. Increases in the hydrophobicity of domain and subunit contacts for the Thermus aquaticus glyceraldehyde-3-phosphate dehydrogenase may explain why it is the most thermostable protein in this series.     

Selectivity of 3-bromo-isoxazoline inhibitors between human and Plasmodium falciparum glyceraldehyde-3-phosphate dehydrogenases

Compounds based on the 3-Br-isoxazoline scaffold fully inhibit glyceraldehyde 3-phosphate dehydrogenase from Plasmodium falciparum by selectively alkylating all four catalytic cysteines of the tetramer. Here, we show that, under the same experimental conditions that led to a fast and complete inhibition of the protozoan enzyme, the human ortholog was only 25% inhibited, with the alkylation of a single catalytic cysteine within the tetramer. The partial alkylation seems to produce a slow conformational rearrangement that severely limits the accessibility of the remaining active sites to bulky 3-Br-isoxazoline derivatives, but not to the substrate or smaller alkylating agents.     

Nucleotide sequence of genes and complete amino acid sequence of tick-borne encephalitis virus strain 205

The 10466 nucleotide long sequence of the cDNA copy of the tick-borne encephalitis strain 205 viral genome has been determined. It includes the 5'-nontranslating region, the genes for structural as well as nonstructural proteins and the first 93 nucleotides of 3'-nontranslating region. The difference in amino acid sequences of structural and nonstructural proteins of strains 205. Sofjin and Neudoerfl of the tick-borne encephalitis virus and the nucleotide changes in 5'- and 3'-nontranslating of these strains are discussed.     

Probing the coenzyme specificity of glyceraldehyde-3-phosphate dehydrogenases by site-directed mutagenesis

By combining our knowledge of the crystal structure of the glycolytic NAD-dependent glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and the sequence of the photosynthetic NADP-dependent GAPDH of the chloroplast, two particular amino acid residues were predicted as the principal determinants of differing coenzyme specificity. By use of site-directed mutagenesis, the amino acids Leu 187 and Pro 188 of GAPDH from Bacillus stearothermophilus have been replaced with Ala 187 and Ser 188, which occur in the sequence from the chloroplast enzyme. The resulting mutant was shown to be catalytically active not only with its natural coenzyme NAD but also with NADP, thus confirming the initial hypothesis. This approach has not only enabled us to alter the coenzyme specificity by minimal amino acid changes but also revealed factors that control the relative affinity of the enzyme for NAD and NADP.     

Nucleotide sequence analysis of the gene encoding the Deinococcus radiodurans surface protein, derived amino acid sequence, and complementary protein chemical studies.

The complete nucleotide sequence of the gene encoding the surface (hexagonally packed intermediate [HPI])-layer polypeptide of Deinococcus radiodurans Sark was determined and found to encode a polypeptide of 1,036 amino acids. Amino acid sequence analysis of about 30% of the residues revealed that the mature polypeptide consists of at least 978 amino acids. The N terminus was blocked to Edman degradation. The results of proteolytic modification of the HPI layer in situ and Mr estimations of the HPI polypeptide expressed in Escherichia coli indicated that there is a leader sequence. The N-terminal region contained a very high percentage (29%) of threonine and serine, including a cluster of nine consecutive serine or threonine residues, whereas a stretch near the C terminus was extremely rich in aromatic amino acids (29%). The protein contained at least two disulfide bridges, as well as tightly bound reducing sugars and fatty acids.