The crystal and molecular structure of yeast L-lactate dehydrogenase (cytochrome b2). An electron microscopy study by negative staining and freeze-etching techniques

Bromopyruvate as an affinity label for baker's yeast flavocytochrome b2. Kinetic study of the inactivation reaction

Bromopyruvate was shown to completely inactivate cytochrome b2 in a reaction that obeyed the kinetic criteria required for affinity labels: it inactivated flavocytochrome b2 according to saturation kinetics, and the inactivation reaction was competitively inhibited by the substrate or competitive inhibitors. Inactivation was irreversible. The behaviour of both forms of flavocytochrome b2 (lintact and proteolytically cleaved) was examined. It was found that the reduced cleaved enzyme was not inactivated by bromopyruvate; this phenomenon can probably be ascribed to a structural change undergone upon reduction. The value of the lactate dissociation constant of intact cytochrome b2 cytochrome b2 was determined in competition experiments with bromopyruvate. By comparison with the divergent published values for the Ks of the cleaved from, it appears that only those that differ from the Km by a factor of two or three are reasonable. This study opens the way for the identification of an active site residue and localization in the peptide chain of the bifunctional enzyme.

Controlled proteolysis of flavocytochrome b2. Characterization of a 15000-dalton heme-binding core and comparison with detergent solubilized cytochrome b5

It is known that each subunit of the tetrameric flavocytochrome b2 can be cleaved by yeast proteases to fragments of molecular weight 33-36000 and 21 000, with some modification of catalytic properties, but without destruction of the oligomeric state of the protein. We report here experimental conditions which enabled us to simulate this specific cleavage in a controlled fashion with chymotrypsin and subtilisin. With trypsin and papain, on the other hand, it was not found possible to stop the digestion in such a way as to obtain a homogeneous still active product. A characterization of the enzymatic forms obtained by digestion with chymotrypsin and subtilisin at 0 degrees C shows that modification of enzymatic and solubility properties occurs in a stepwise fashion. It is also ccluded that cleavage by yeast proteases is accompanied by loss of 10 to 25 residues. At 37 degrees C, chymotrypsin digestion yields a heme-binding core of molecular weight 15 000, larger than the already characterized tryptic heme-binding core by about 40 residues. Although the latter is known to be very similar to trypsin-solubilized cytochrome b5, the lack of aggregation of the former in aqueous solution, its amino acid composition and circular dichroism spectra do not point to a similarity of its additional peptide segment with the hydrophobic tail of detergent-solubilized cytochrome b5.

Complete amino acid sequence of the heme-binding core in bakers' yeast cytochrome b2 (L-(+)-lactate dehydrogenase)

We are reporting here an analysis of the chymotryptic peptides obtained from the tryptic heme-binding fragment of flavocytochrome b2 (cytochrome b2 core). These results completely establish the sequence of the 96 residue-long fragment, for which preliminary evidence has been published before [24]. We also report full experimental details concerning the automatic degradation, the specific cleavages at the unique arginine and methionine residues, and the analysis of the tryptic peptides. In addition, it is shown that the main heme-binding fragment resulting from cytochrome b2 proteolysis by yeast proteases has an additional glutamic acid residue at the C-terminal end relative to the main tryptic heme-binding fragment. The slight sequence modifications presented here (amide groups and insertion of a lysine residue after position 71) do not substantially modify the comparison with liver microsomal cytochrome b5. A new sequence alignment is proposed for the two proteins, and a few structural considerations are presented, based on the inspection of calf liver cytochrome b5 three-dimensional model [50,51].

Flavocytochrome b2: simulation studies of the electron-transfer reactions among the prosthetic groups

Simulation studies by digital computer were undertaken in order to test and clarify the interpretations deduced from experimental data concerning the electron transfer mechanism from L-lactate to flavocytochrome b2, which were presented in a preceding paper in this journal. The reaction scheme proposed as the "best" one is composed of 7 steps. It allows the best fitting of the time courses established for the oxidized flavin (Flox), the flavin semiquinone (Flsq), the fully reduced flavin (Flred), and the reduced haem (Hred); it can be extended to 1 s. This scheme also allows a good simulation of the general shape of preequilibrium titration curves obtained at a 200-ms reaction time for Hred and Flsq, and a valuable simulation of the reduction electron paramagnetic resonance time course established for Hred and Flsq at low lactate concentration. The agreement between experimental and simulated curves led to an estimation of some rate constants experimentally unknown, relative (in particular) to the electron exchange between flavin and haem and between couples of flavins. Another interest of these stimulation studies was to point out the obligatory involvement of a slow final step to perform the flavocytochrome b2 full reduction; this step could be controlled by some conformational change of the protein.

Flavocytochrome b2: kinetic studies by absorbance and electron-paramagnetic-resonance spectroscopy of electron distribution among prosthetic groups

The reduction by L-lactate of the prosthetic groups of flavocytochrome b2 (L-lactate cytochrome c oxidoreductase from aerobic yeast, a tetrameric molecule containing one haem and one flavin mononucleotide per protomer) was reinvestigated. It was confirmed that the enzyme ultimately takes up 3 electrons per protomer from this 2-electron donor. Stopped-flow absorbance data at an haem isosbestic point to follow the oxidized flavin and in a haem band indicate that, under the conditions used, haem and flavin reduction time courses are indistinguishable, both being biphasic (phases I and II). Comparison with electron paramagnetic resonance data (Fe3+ haem and flavosemiquinone signals) led to a complete description at 24 degrees C of the time courses of the various reduction states of the prosthetic groups. It has been previously demonstrated (Morton and Sturtevant, 1964) that, after the formation of the enzyme-substrate complex, the electron transfer to the enzyme takes place as the first and rate-limiting step of the turnover. In the present study, an initial burst of fully reduced flavin, of small amplitude, is detected at the very beginning of phase I (before 6 ms). The redox forms which accumulate thereafter till the end of phase I (30-35 ms) are the reduced haem (up to 80%), the flavin semiquinone (up to 50%) and the fully reduced flavin (from 25% up to 35%); the total of electrons distributed at the end of phase I is about 2 per protomer meaning that, in this phase, each enzyme site acts as a 2-electron and not a 3-electron acceptor. A 2-electron flow as the limiting step during phase I with the rate constant kI accounts for the steady-state electron flow during catalysis. Phase I is followed by the much slower phase II which corresponds to the entry of the third electron and cannot be involved in the turnover. The interpretation of the results are given as a scheme, with the proper rate constants, allowing a satisfactory fitting of experimental data by simulation. Among the elementary steps required are a rapid distribution of one electron from reduced flavin to the haem, a rapid interprotomers dismutation between couples of flavin semiquinone regenerating two oxidized flavin per tetramer. The very low reactivity of the latter for the entry of the third electron per protomer is tentatively explained by the occurrence of a slow additional step limiting the final reduction reaction. It was observed that, over phase I and the beginning of phase II, from 15 to 200 ms, all the redox species remain apparently under equilibrium conditions. Parallel studies (titrations of flavocytochrome b2 by L-lactate) showed that the set of equilibrium parameters relative to haem and flavin species is significantly different in the "final" equilibrium (after 30 s) from that in the time interval 15-200 ms. Such an anomaly suggests a conformation change takes place very slowly in the molecule after the acceptance of the first two electrons per protomer.

Freeze-fracture of microtubules and bridges in motile axostyles

A freeze-fracture study of the motile axostyles of the flagellate protozoa Saccinobaculus and Pyrsonympha has been undertaken in order to obtain a view of the relationships of microtubules and their cross bridges not dependent on conventional preparative procedures. Reactivation studies using isolated axostyles prepared for freeze-fracture and then thawed demonstrate that we are observing the structure of a potentially functional axostyle. Cross fractures through the axostyle demonstrate more extensive interrow bridging than expected on the basis of observations of thin-sectioned material. Each microtubule has approximately sixfold bridge-binding sites with connections to as many as four interrow bridges. Measurements of microtubule diameter and spacing are significantly larger than those made from sectioned material and may indicate that conventional processing for electron microscopy results in the loss of structurally important water within the microtubule in addition to loss of intertubule material. Longitudinal fractures through the axostyle at various orientations demonstrate a minimum longitudinal periodicity of 160 A for both the spacing of the globular subunits within the microtubule wall and the spacing of the intrarow bridges. While intrarow bridges are strictly periodic and always oriented in parallel, interrow bridges are not strictly periodic and can be oriented at varying angles to the microtubule axis.

Homology between bakers' yeast cytochrome b2 and liver microsomal cytochrome b5

The amino-acid sequence of the hemebinding region of bakers' yeast cytochrome b(2) [L-(+)-lactate dehydrogenase, EC 1.1.2.3] has been determined. It shows a strong similarity with the sequence of microsomal cytochrome b(5), and appears to be compatible with the same kind of peptide-chain folding, in agreement with data obtained previously by various physiochemical methods. The comparison shows that the fifth and sixth heme ligands must be histidine residues, thus substantiating previous conclusions drawn in particular from photooxidation experiments and nuclear magnetic resonance studies. The data reported in this paper suggest a common origin for the two proteins. Implications for their biochemical evolution are presented.