Localization of glyceraldehyde-3-phosphate dehydrogenase in intact human erythrocytes. Evaluation of membrane adherence is autoradiographs at low grain density

The complex of band 3 protein of the human erythrocyte membrane and glyceraldehyde-3-phosphate dehydrogenase: stoichiometry and competition by aldolase

The cytoplasmic domain of band 3, the main intrinsic protein of the erythrocyte membrane, possesses binding sites for a variety of other proteins of the membrane and the cytoplasm, including the glycolytic enzymes glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and aldolase. We have studied the stoichiometry of the complexes of human band 3 protein and GAPDH and the competition by aldolase for the binding sites. In addition, we have tried to verify the existence of mixed band 3/GAPDH/aldolase complexes, which could represent the nucleus of a putative glycolytic multienzyme complex on the erythrocyte membrane. The technique applied was analytical ultracentrifugation, in particular sedimentation equilibrium analysis, on mixtures of detergent-solubilized band 3 and dye-labelled GAPDH, in part of the experiments supplemented by aldolase. The results obtained were analogous to those reported for the binding of hemoglobin, aldolase and band 4.1 to band 3: (1) the predominant or even sole band 3 oligomer forming the binding site is the tetramer. (2) The band 3 tetramer can bind up to four tetramers of GAPDH. (3) The band 3/GAPDH complexes are unstable. (4) Artificially stabilized band 3 dimers also represent GAPDH binding sites. In addition it was found that aldolase competes with GAPDH for binding to the band 3 tetramer, and that ternary complexes of band 3 tetramers, GAPDH and aldolase do exist.

Structure within eukaryotic cytoplasm and its relationship to glycolytic metabolism

Taken together, the results reviewed here indicate that both structural proteins and enzymes exist in a relatively mobile, uncomplexed form and in a relatively immobile form, complexed with the matrix. The relative amounts of free and complexed forms of each protein are dependent upon the local concentrations of both small molecules and other macromolecules and hence may vary in time and space throughout the cell. Free and cytomatrix-bound enzymes exchange rapidly, while free and cytomatrix-bound structural proteins exchange more slowly. These two distinct time scales suggest that the slowly exchanging structural proteins form the core of fibrous structural elements--having many stabilizing intermolecular contacts with near neighbours--whereas the more rapidly exchanging enzymes adsorb to the surface of the structural elements and have fewer near neighbour contacts. The hierarchical nature of these associations is depicted schematically in Figure 3. Metabolism is proposed to proceed primarily via transport of small metabolites rather than by transport of enzymes, which may be organized in functional clusters to facilitate, metabolic regulation.

Rotational diffusion of band 3 in erythrocyte membranes. 2. Binding of cytoplasmic enzymes

Time-resolved phosphorescence anisotropy has been used to study the rotational diffusion of eosin-labeled human erythrocyte band 3 in the presence of an enzyme bound at its cytoplasmic pole. With increasing amounts of G3PD (glyceraldehyde-3-phosphate dehydrogenase) added to ghosts, the infinite time anisotropy (r infinity) increases, and at saturating concentrations, very little decay of the anisotropy r(t) occurs at all. These phenomena are reversed by elution of the enzyme with 150 mM NaCl. Prior proteolytic removal of the N-terminal 41-kDa cytoplasmic fragment of band 3 also disenables the G3PD effect. When ghosts are stripped of their residually bound G3PD, a small reduction in the fraction of immobile band 3 is observed. Finally, titration of band 3 sites with aldolase shows effects on the r(t) qualitatively similar to those observed with G3PD. On the basis of our interpretation of the heterogenous anisotropy decay of eosin-labeled band 3 [Matayoshi, E. D., & Jovin, T. M. (1991) Biochemistry (preceding paper in this issue)], we conclude that the binding of G3PD and aldolase results in the immobilization of band 3 oligomers.

Oxidation of articular cartilage glyceraldehyde-3-phosphate dehydrogenase (G3PDH) occurs in vivo during carrageenin-induced arthritis

Articular cartilage proteoglycan biosynthesis was substantially inhibited by the competitive glycolytic inhibitor 2-deoxyglucose (approximately 65% at 100 mM), but to a much lesser degree (approximately 10%) by the oxidative phosphorylation uncoupler, 2,4-dinitrophenol. These results confirm that articular cartilage proteoglycan synthesis mostly utilises ATP which is generated by glycolysis. In addition, we have utilised the loss of the relatively specific labelling of glyceraldehyde-3-phosphate dehydrogenase (G3PDH) by [3H]-iodoacetic acid to show that rabbit articular G3PDH is oxidised in vivo during the animal model of acute arthritis, carrageenin-induced arthritis, in the same way as we have previously shown that cartilage G3PDH is oxidised after in vitro exposure to sublethal doses of H2O2. The oxidation of rabbit G3PDH in vivo (18 hr post-injection) corresponds with the maximal influx of PMNL cells into the arthritic synovial fluid and with substantial inhibition of proteoglycan core protein synthesis. We propose that H2O2 released from "activated" PMNLs and macrophages is responsible for the "down-regulation" of biosynthetic processes found in cartilage during acute inflammation.

Interactions of glycolytic enzymes with erythrocyte membranes

Partition equilibrium experiments have been used to characterize the interactions of erythrocyte ghosts with four glycolytic enzymes, namely aldolase, glyceraldehyde-3-phosphate dehydrogenase, phosphofructokinase and lactate dehydrogenase, in 5 mM sodium phosphate buffer (pH 7.4). For each of these tetrameric enzymes a single intrinsic association constant sufficed to describe its interaction with erythrocyte matrix sites, the membrane capacity for the first three enzymes coinciding with the band 3 protein content. For lactate dehydrogenase the erythrocyte membrane capacity was twice as great. The membrane interactions of aldolase and glyceraldehyde-3-phosphate dehydrogenase were mutually inhibitory, as were those involving either of these enzymes and lactate dehydrogenase. Although the binding of phosphofructokinase to erythrocyte membranes was inhibited by aldolase, there was a transient concentration range of aldolase for which its interaction with matrix sites was enhanced by the presence of phosphofructokinase. In the presence of a moderate concentration of bovine serum albumin (15 mg/ml) the binding of aldolase to erythrocyte ghosts was enhanced in accordance with the prediction of thermodynamic nonideality based on excluded volume. At higher concentrations of albumin, however, the measured association constant decreased due to very weak binding of the space-filling protein to either the enzyme or the erythrocyte membrane. The implications of these findings are discussed in relation to the likely subcellular distribution of glycolytic enzymes in the red blood cell.

Genomic organization of the glyceraldehyde-3-phosphate dehydrogenase gene family of Caenorhabditis elegans

Glyceraldehyde-3-phosphate dehydrogenase (GAPDHase) is encoded by four genes designated gpd-1 through gpd-4 in the nematode Caenorhabditis elegans. gpd-1 has been isolated and sequenced, and is shown here to have a nearly identical copy (gpd-4) with respect to coding and regulatory flanking sequence information as well as to the placement of its two introns. Both genes, which are separated by 250,000 to 300,000 base-pairs were assigned to chromosome II by in situ hybridization and physically linked to a DNA polymorphism located near unc-4 on the genetic map. The genes gpd-2 and gpd-3 are also nearly identical with each other but differ from the gpd-1 and gpd-4 pair with respect to the positions of their two introns and a cluster of amino acid changes within the amino-terminal region of the enzyme. Furthermore, one gene from each pair (gpd-4 and gpd-2) exhibits a single amino acid substitution at positions heretofore known to be conserved in all other systems so far examined including the extreme thermophiles. gpd-2 and gpd-3 are organized as a direct tandem repeat separated by only 244 base-pairs. They have been assigned to an 85,200 base-pair contig that maps to the left end of the X chromosome. The absence of gpd-3 from C. elegans var. Bergerac was used as a marker to map the gpd-2,3 gene pair near unc-20. Northern analyses have shown that gpd-1 and gpd-4 are preferentially expressed in embryos, while the expression of gpd-2 and gpd-3 increases during postembryonic development. These analyses indicate that the gpd-1,4 gene pair encodes the minor isoenzyme, GAPDHase-1, present in all cells of the nematode while the other gene pair (gpd-2,3) encodes the major isoenzyme, GAPDHase-2, preferentially expressed in the bodywall muscle. The G + T-rich and T-rich regions essential for vertebrate beta-globin polyadenylation were also observed for gpd-3.

A reappraisal of the binding of cytosolic enzymes to erythrocyte membranes

Several cytosolic proteins have been shown to be associated with hypotonic erythrocyte ghosts via electrostatic interactions with the anion transport band 3 protein. This article considers the problems of demonstrating binding under physiological conditions and reviews the evidence for the relevance of enzyme binding to the membrane for the regulation of glycolysis. The hypotheses for the existence of topological and sequential multienzyme complexes of the glycolytic enzymes in erythrocytes are also discussed.

The glyceraldehyde-3-phosphate dehydrogenase gene family in the nematode, Caenorhabditis elegans: isolation and characterization of one of the genes

The isolation and genomic sequence of one of possibly four glyceraldehyde-3-phosphate dehydrogenase genes in the nematode, Caenorhabditis elegans is presented. The complete nucleotide sequence of the coding as well as the noncoding flanking regions of this gene has been determined. The deduced amino-acid sequence agrees with the sequence of typical glyceraldehyde-3-phosphate dehydrogenase enzymes and its molecular weight of 36,235 agrees with its size determined previously (Yarbrough, P. and Hecht, R. (1984) J. Biol. Chem. 259, 14711-14720). That this isolated gene encodes a nematode glyceraldehyde-3-phosphate dehydrogenase is additionally confirmed by demonstrating its immunoreactivity to an anti-nematode glyceraldehyde-3-phosphate dehydrogenase antibody after its expression as a fusion protein with dihydrofolate reductase. Codon utilization follows a pattern typical of other expressed nematode genes. The gene is split by two introns that are highly conserved in comparison to other introns observed in C. elegans. The placement of one of these introns is conserved with respect to the chicken glyceraldehyde-3-phosphate dehydrogenase gene. Within the 5' flanking sequence homology to actin and the homology 2 block of the major myosin gene (unc-54) is noted. It is of interest that the 3' flanking region contains a CAAAT box, followed by a TATAAT box, before an open reading frame of a closely linked gene that also contains a small AT-rich intron with the nematode consensus splice junction.

Lack of binding of glyceraldehyde-3-phosphate dehydrogenase to erythrocyte membranes under in vivo conditions

A filtration method is described for separating membrane-free cytoplasm from concentrated erythrocyte haemolysates. The method has been used to assess glyceraldehyde-3-phosphate dehydrogenase binding to erythrocyte membranes. The relative amounts of glyceraldehyde-3-phosphate dehydrogenase and lactate dehydrogenase in the cytoplasm (either oxygenated or deoxygenated) indicate there is no detectable binding of glyceraldehyde-3-phosphate dehydrogenase to the membranes under physiological conditions.

Characterization of a photosensitive glucose derivative. A photoaffinity reagent for the erythrocyte hexose transporter

The photosensitive reagent 6-N-(4-azido-2-hydroxy-3,5-diiodobenzoyl)-D-glucosamine has been assessed as a potential photoaffinity label for the hexose transporter. Under zero-trans conditions, transport experiments performed in the dark reveal that the reagent inhibits the uptake of D-glucose in resealed human erythrocyte ghosts. Increasing the concentration of glucose in the transport medium has a protective effect, reducing the inhibition. Kinetic analysis indicates that the probe acts as a competitive inhibitor with high affinity for the erythrocyte hexose transporter (Ki between 0.07 and 0.2 microM). Exposure to a 280 nm filtered high intensity mercury-vapor lamp results in a rapid and efficient photolysis. At low concentrations of the probe, specific labeling of membrane preparations was observed. Autoradiograms of 10% SDS gels revealed the specific labeling of bands 4.51 and 6. This labeling was concentration-dependent and protected by D-glucose (not the L-isomer) and phloretin in the medium. When subjected to multiple exposures of low concentration of the photoaffinity reagent, apparent saturation was achieved.

Quantitative characterization of the interactions of aldolase and glyceraldehyde-3-phosphate dehydrogenase with erythrocyte membranes

Results of studies on the interactions of aldolase and glyceraldehyde-3-phosphate dehydrogenase with erythrocyte ghosts have been reinterpreted by making allowance for possible multivalency of the enzymes in regard to their interactions with matrix sites. It is shown that the curvilinearity of the experimental Scatchard plots may be attributed fully to the formation of enzyme-membrane complexes in which tetravalent enzyme may form crosslinks between several membrane sites. This interpretation of the results is preferable to earlier analyses based on heterogeneity of membrane sites in that (a) it takes into account the tetrameric nature of aldolase and glyceraldehyde-3-phosphate dehydrogenase, and (b) it is consistent with experimental demonstrations that band 3 protein is the sole site for enzyme interaction with the erythrocyte matrix. The dependence on ionic strength of the intrinsic association constant for either interaction is such that the binding of neither aldolase nor glyceraldehyde-3-phosphate dehydrogenase could be detected at ionic strengths in excess of 0.08 I. This finding is discussed in relation to the claims and counterclaims concerning the physiological significance of these interactions between glycolytic enzymes and erythrocyte membranes.

Centrifugal analysis of undiluted packed human erythrocyte lysates. Studies of the association of glyceraldehyde-phosphate dehydrogenase with the membrane fraction

Concentrated human erythrocyte lysates (greater than 99% initial haematocrit) were subjected to high centrifugal fields. This caused the membrane fraction to separate from the cytoplasmic portion, due to the lower density of the former. Enzyme distribution data indicated that glyceraldehyde-phosphate dehydrogenase was predominantly in the cytoplasmic fraction.