Translation is a non-uniform process. Effect of tRNA availability on the rate of elongation of nascent polypeptide chains.

We reported elsewhere (Varenne et al., 1982) that, during synthesis of a number of colicins in Escherichia coli, intermediate nascent chains of discrete sizes accumulated, suggesting a variable rate of translation. In this paper, a detailed analysis provides arguments that this phenomenon, at least for the proteins under study, is not related to aspects of messenger RNA such as secondary structure. It is linked to the difference in transfer RNA availability for the various codons. Experimental analysis of translation of other proteins in E. coli confirms that the main origin for the discontinuous translation in the polypeptide elongation cycle is the following. For a given codon, the stochastic search of the cognate ternary complex (aminoacyl-tRNA-EF-Tu-GTP) is the rate-limiting step in the elongation cycle: transpeptidation and translocation steps are much faster. The degree of slackening in ribosome movement is almost proportional to the inverse of tRNA concentrations. The verification of this model and its possible physiological significance are discussed.

[Carbonic anhydrases of bovine erythrocytes: comparative study of the anhydrase and esterase activity of different forms]. / Anhydrases carboniques érythrocytaires bovines: étude comparée des activités anhydrasiques et estérasiques des différentes formes.

Comparative study of esterase activities (p- and o-nitrophenylacetate) allowed to characterize three groups of bovine erythrocyte carbonic anhydrases:--the first one includes CI, CII (isozyme of CI) and CIr ("artificial" product of CI).--the second one includes native CIv1 and "artificial" CIv1, first conformational variants of CI,--finally CIv2, second "artificial" conformational variant of CI. Possible modifications of the enzyme site between the first and the other enzyme groups are discussed. Except CIv2 of lower activity, all the products have identical carbonic anhydrase activity. The catalytic constants Km ap and kcat ap for hydrolysis of p-nitrophenylacetate have been determined for all enzymes; this study confirms the lower activity of CIv2.

Lipolysis and heterogeneous catalysis. A new concept for expressing the substrate concentration.

A new concept is proposed for quantifying the substrate concentration during heterogeneous catalysis of the kind which occurs during lipolysis. The number of molecules of protein (enzyme) adsorbable to the lipid substrate interface per unit of volume was evaluated and defined as a volumetric concentration of protein (enzyme) binding site (PEBS). Using porcine pancreatic lipase (EC 3.1.1.3) as a model enzyme, the maximal PEBS concentration was measured under various assay conditions by determining the saturation of the lipid substrate with the enzyme. Abacuses correlating the lipid substrate concentration (M) with the PEBS concentration (M) under each experimental conditions were used to express the kinetic data in terms of a volumetric concentration of PEBS. Comparisons could thus be made between data obtained with various enzymes and lipid interfaces because they were expressed with the same unit. In the case of pancreatic lipase, using triolein and tributyrylglycerol as substrates, Km values of 2.7 and 7.5 nM PEBS were obtained, respectively, and KD values ranging around 9 nM PEBS were also obtained from Scatchard plots. In addition, the average superficial density of PEBS was found to be 10 x 10(11) molecules.cm-2, which is a value commonly obtained with structural proteins and enzymes adsorbed to an acylglyceride-water interface, this finding supports the idea that the PEBS concept represents the room in which the protein molecule adsorbs at the lipidic interface.

Enzymes as biosensors. 1. Enzyme memory and sensing chemical signals.

When a free enzyme exists under different conformations that 'slowly' isomerize during the conversion of a substrate into a product, the corresponding 'slow' relaxation component may interfere with the steady-state component. The apparent steady-state rate that may be measured under these conditions is called the meta-steady-state rate for it refers to the existence of metastable states of the enzyme during the reaction. By contrast to the real steady-state rate, the meta-steady-state rate is dependent upon the initial state of the enzyme, that is on the respective concentrations of the free enzyme forms. The simplest model that may display this type of behaviour is the mnemonical model. For a fixed concentration of the last product of the reaction sequence the meta-steady state is different depending on that concentration being reached by an increase or a decrease of a previous concentration. This means that the meta-steady-state rate describes a hysteresis loop as the product concentration is increased and decreased. Owing to the existence of metastable states, the enzyme system behaves as a biosensor that is able to detect both a concentration and the direction of a concentration change. The existence of the hysteresis loop of the meta-steady-state rate implies that the two free enzyme forms display hysteresis as well. A chemical potential, called the sensing potential, is specifically associated with the 'perception' of the direction of the thermodynamic force generated by the decrease or the increase of the concentration of the ligand that binds to one of the enzyme conformations. The sensing potential of the enzyme conformer that does not bind the product increases and reaches a plateau as the chemical potential of that product is raised. Alternatively the sensing potential of the other conformer vanishes at low and high chemical potentials of the product and is significant for intermediate chemical potentials. Enzymes that display very slow conformation changes may thus be viewed as elementary sensor devices.

Antifungal effect of Hevea brasiliensis latex with various fungi. Its synergistic action with amphotericin B against Candida albicans.

Antifungal activity of latex from Hevea brasiliensis was observed with various fungi in macrobroth dilution assays. The strongest antifungal effect was obtained with Trichosporon cutaneum (MIC 80% = 40.615 microg protein ml(-1), Kaff = 0.075 microg(-1) protein ml) and Cryptococcus neoformans (MIC 80% = 56.078 microg protein ml(-1), Kaff = 0.059 microg(-1) protein ml). Amphotericin B was synergized with all H. brasiliensis latex concentrations tested. The rates of synergy were about 50, 44 and 55% with 15, 30 and 60 microg protein ml(-1) latex, respectively. The putative role of glycosidase activities measured in crude latex, especially alpha-d-mannosidase and N-acetyl-beta-d-glucosaminidase activities which are potentially able to split off intraparietal linkages between glycosidic residues, is discussed. The possible role of another antifungal factor associated with rubber particles of latex is suggested.

Mathematical modelling of antifungal action.

In this paper a simplified modelling approach indicated that yeast growth was inhibited by an antifungal drug according to an exponential function. In addition, the corresponding inhibition rate followed a hyperbolic function the parameters of which permit us to determine the percentage of maximum inhibition and the minimum inhibitory concentration for 80%. From the equation of a hyperbola it was also possible to calculate an affinity constant Kaff corresponding to the inverse of the concentration of antifungal drugs giving half the maximal inhibition. The affinity constant was demonstrated to be characteristic of the yeast strain and of the antifungal drug employed. Simulation of the mathematical modelling enabled determination of a theoretical inhibition level corresponding to strong concentrations of antifungal drugs which cannot be carried out for technical reasons (precipitates, opacity etc.). The interest of this mathematical modelling of growth and inhibition to predict the doses of antifungals which can act synergistically is discussed.

Equilibrium binding of thioredoxin fB to chloroplastic fructose bisphosphatase. Evidence for a thioredoxin site distinct from the active site.

Thioredoxin fB, the protein activator of chloroplastic fructose 1,6-bisphosphatase, strongly binds its target enzyme with a stoichiometry of one protein dimer per enzyme tetramer. The thioredoxin binding site is distinct from the active site and the dissociation constant of the protein-enzyme complex has the extremely small value of 769 nM at pH 7.5. This interaction involves both ionic and hydrophobic contributions and is enhanced by a pH increase from 7 to 8. These results suggest that the above molecular properties may be involved in the light activation of chloroplastic fructose bisphosphatase.

Thermodynamics of information transfer between subunits in oligomeric enzymes and kinetic cooperativity. 1. Thermodynamics of subunit interactions, partition functions and enzyme reaction rate.

The strength of quaternary constraints between two subunits of a polymeric enzyme depends upon the number of neighboring subunits and upon whether these subunits are liganded or not. These quaternary constraints between two subunits of a complex polymeric enzyme may be expressed, however, in terms of quaternary constraints that exist within ideal dimers. The influence of quaternary constraints on the reaction rate of a complex polymeric enzyme may thus be expressed in terms of the intersubunit strain that exists within dimers. This conclusion, that was far from evident, appears to be the consequence of the postulates of structural kinetics, and derive as well from usual thermodynamic principles. The structural steady-state equations may be expressed in terms of partition and sub-partition functions. As applied to structural kinetic models, a partition function expresses how, during the steady state, the energy of a population of enzyme molecules is distributed over n states. Similarly a sub-partition function describes how, during the steady state, the energy of these enzyme molecules is partitioned among only n-k of these states. Although the concept of partition function was initially formulated for equilibrium processes, it may be extended without any loss of generality to non-equilibrium processes. Moreover it is reminiscent of the concept of binding polynomial presented some years ago by Wyman for the equilibrium binding of a ligand to a protein. With this formalism derived from statistical mechanics, a structural rate equation may be derived from the ratio of a sub-partition function of degree n-1 and of a partition function of degree n. Again these properties are the consequence of the postulates of structural kinetics associated with simple ideas derived from statistical thermodynamics.

Thermodynamics of information transfer between subunits in oligomeric enzymes and kinetic cooperativity. 2. Thermodynamics of kinetic cooperativity.

The principles of structural kinetics allow one to define the thermodynamic conditions that are sufficient to generate a certain type of kinetic behavior. If subunits are loosely coupled, that is if no quaternary constraint exists between them, the kinetic behavior of the polymeric enzyme is qualitatively defined by the behavior of an ideal dimer. The nature and the extent of the kinetic cooperativity are defined by the energy of interaction, delta G rho, between two subunits. This energy of interaction is that of an ideal dimer relative to that of the A2 and B2 states. This thermodynamic formulation of a given type of cooperativity holds whatever the degree of polymerization of the enzyme. Under these conditions of loose coupling between subunits, positive kinetic cooperativity cannot be associated with any sigmoidicity of the rate curve. The range of energy coupling where positive kinetic cooperativity must, of necessity, be observed becomes more and more narrow as the number of subunit interactions is increased. This range, however, is independent of the number of subunits. The same situation is not observed for negative cooperativity which appears to be independent of both the number of subunits and the number of subunit interactions. If the subunits are tightly coupled, that is if quaternary constraints exist between them, three thermodynamic parameters, delta G' rho, delta G lambda, delta G mu, are required to define the nature of kinetic cooperativity. delta G' rho is the free energy of an ideal strained dimer relative to that of strained A2 and B2 states. delta G lambda and delta G mu represent the difference of strain energies between conformations A and B and B and B relative to that existing between conformations A and A. One may determine in the parametric space (delta G' rho, delta G lambda, delta G mu) the boundaries between the sufficient conditions that generate a certain type of cooperativity and the lack of these conditions. The kinetic parameters of the rate equation are not all independent. A number of constraint conditions exist between them which depend upon the subunit design of the polymeric enzyme. The existence of these constraint conditions may be diagnostic of a certain type of subunit interactions.

Kinetic implications of the occurrence of several relaxations in the conformational transition of mnemonical enzymes.

If the conformational transition involved in enzyme memory occurs in several elementary steps, the time constant of the overall 'slow' relaxation is mostly determined by the individual values of the rate constants pertaining to the overall transconformation. The extent of kinetic co-operativity of the enzyme reaction, however, is mostly controlled by the degree of reversibility of the elementary steps of the conformational transition. There is then no simple relation between the time scale of the 'slow' transition and the extent of kinetic co-operativity of the enzyme reaction. A slow transition of about 10(-3) s-1 is therefore perfectly compatible with a strong positive or negative co-operativity and in particular with the negative co-operativity observed with wheat germ hexokinase LI. The relationship that has been established recently [Pettersson, G. (1986) Eur. J. Biochem. 154, 167-170] between the 'slow' enzyme relaxation and the extent of kinetic co-operativity holds only in the specific case where the transconformation occurs in one step. Owing to the possible occurrence of a multistep conformation change, the lack of this relationship means nothing as to the validity, or the invalidity, of the concept of mnemonical transition. More informative than the time scale of the 'slow' transition is its dependence with respect to glucose and glucose 6-phosphate, which both react with the enzyme. The effect of reaction products on the modulation of kinetic co-operativity is also of cardinal importance in the diagnosis of enzyme memory. Since an alternative model has been recently proposed by Pettersson (cited above) to explain the mechanistic origin of kinetic co-operativity of monomeric enzymes, the effect of products on the kinetic co-operativity predicted by this alternative model has been studied theoretically, in order to determine whether it is consistent with the experimental results obtained with wheat germ hexokinase LI. This analysis shows that the predictions of this model are in total disagreement with both the predictions of the mnemonical model and the experimental results obtained with wheat germ hexokinase LI, as well as with other enzymes. This alternative model cannot therefore be considered as a sensible explanation of the mechanistic origin of co-operativity of monomeric enzymes. It is therefore concluded that the mnemonical model which rests on numerous experimental results, obtained by different research groups, on different enzymes is the simplest and most likely explanation of the kinetic subtleties displayed by some monomeric enzymes, and in particular wheat germ hexokinase LI.

Enzyme memory. Effect of glucose 6-phosphate and temperature on the molecular transition of wheat-germ hexokinase LI.

A theory is presented that associates burst (orlag) kinetics with the respective concentrations of enzyme initial states X1 and X6 and with the cooperation of a mnemonical enzyme. The theory predicts that for an enzyme with a negative cooperation, decreasing the initial concentration of X1 (or increasing that of X6) tends to increase the induction time. This increase may correspond to a reversal of a burst in a lag. Similarly, if the enzyme has a positive cooperation, decreasing the initial concentration of X6 (or increasing that of X1) increases the induction time. The first case above is expected to apply to wheat germ hexokinase LI, X1 being the form that binds glucose preferentially, and X6 the one that binds glucose 6-phosphate. By changing solely the respective concentrations of the two initial forms, one may expect to modify the pre-steady-state phase but not the steady-state kinetics of the reaction. By jumping the temperature of the enzyme solution from 4 degrees C to 30 degrees C and letting the transconformation ewuilibrium relax for various periods of time before mixing enzyme with the substrates, one can analyse the effect of the relative concentrations of X1 and X6 on the induction time. One can estimate in that way one of the rate constants of the transconformation between the two free enzyme forms. The shorter the incubation time at 30 degrees C then the smaller is the negative induction time (in absolute values). Another possibility of controlling the ratio between the two initial concentrations of the enzyme, is to pre-mix hexokinase with glucose 6-phosphate and to arrange that glucose-6-phosphate concentration, after mixing enzyme and substrates, is held constant whatever the pre-mixing conditions. When wheat germ hexokinase LI is pre-mixed 30 min at 30 degrees C with glucose 6-phosphate before the reaction starts, the burst does not disappear. If, on the other hand, pre-mixing is effected at 4 degrees C the burst is reversed into lag. This result is taken to mean that the equilibrium constant between the two free enzyme forms (the 'circle' and the 'rhombus') is strongly dependent on temperature. A direct study of the effect of glucose 6-phosphate on the conformational equilibrium of wheat germ hexokinase, gives support to this interpretation. If hexokinase is mixed at 4 degrees C with glucose 6-phosphate a slow increase in fluorescence of tryptophanyl residues is observed, which indicates that the 'rhombus' conformation accumulates under these conditions. On the other hand, at 30 degrees C, glucose 6-phosphate does not produce any significant change in the fluorescence of the protein. As expected, these results imply that the equilibrium between the two free enzymes species is freely reversible a 4 degrees C and nearly irreversible at 30 degrees C. The equations derived from the mnemonical model allow fitting or simulation of the experimental results.

Investigation on the kinetic mechanism of octopine dehydrogenase. A regulatory behavior.

The kinetic scheme of octopine dehydrogenase of Pecten maximus L., a monomeric enzyme obeying a bi-ter sequential mechanism, was completed, essentially in the forward reaction, by steady-state studies over a wide range of substrate concentration at pH 7.0. Deviation from the Michaelis-Menten behavior with respect to NAD+ and other significant kinetic data led us to ascribe for octopine dehydrogenase mechanism the mnemonical enzyme concept. In addition, another regulatory behavior can be envisaged involving the formation of two dead-end complexes enzyme.NADH.D-octopine and enzyme.NAD+.pyruvate.L-arginine.

The 'slow' pH-induced conformational transition of chloroplast fructose 1,6-bisphosphatase and the control of the Calvin cycle.

A slow conformation change of chloroplastic reduced fructose bisphosphatase is detected upon raising the pH from 7 to 8 or upon lowering the pH from 8 to 7. These conformation changes are fully reversible. In the time scale investigated these processes occur in one step. Their time constants and their amplitudes have been determined and analyzed as a function of proton concentration. The results obtained are consistent with the view that upon ionization or protonation of a strategic ionizable group the protein undergoes a 'slow' conformational transition that may be followed by conventional fluorescence techniques. Since illumination brings about a pH rise of chloroplastic stroma from 7 to 8, the above results suggest that light activation of fructose bisphosphatase is at least in part due to a slow conformation change of this enzyme.

Thermodynamics of information transfer between subunits in oligomeric enzymes and kinetic cooperativity. 3. Information transfer between the subunits of chloroplast fructose bisphosphatase.

The theory and the methods that have been described in the two preceding papers in this journal have been used to analyze the kinetic properties of chloroplast fructose bisphosphatase. The enzyme is a tetramer made up of apparently identical subunits and displays a sigmoidal kinetics with respect to its substrate, fructose bisphosphate. The free ionic species, magnesium and fructose bisphosphate bind to the enzyme and the chelate fructose-bisphosphate-magnesium does not affect the sigmoidicity of the rate curves. The Hill coefficient with respect to free fructose bisphosphate is equal to 2.3, which is indeed incompatible with the view that the enzyme behaves as a dimer of dimers. This conclusion is confirmed by direct analysis of the rate curve. On the basis of the sum of the residuals, their sum of squares, the standard error of the kinetic parameters of the equation, the kinetic scheme associated with a dimer of dimers may be ruled out. On the basis of the same criteria, the fit of an Adair equation to the rate data cannot be retained as satisfactory. This is a direct proof that neither the Monod nor the Koshland model can correctly fit these kinetic data. In fact the model that fits these data best is a structural kinetic scheme where information transfer occurs between each subunit and its three neighbors ('tetrahedral' mode of information transfer). The fit of these models to a large number of kinetic data allows one to compute the free energy profile during the successive binding processes of the four substrate molecules to the enzyme. Whereas the first two steps are associated with an increase of free energy, all the other subsequent steps are associated with a decrease of free energy.

Kinetics of the modulation of chloroplastic fructose-1,6-bisphosphatase activity by thioredoxin fb.

The inactivation of reduced chloroplast fructose-bisphosphatase by oxidized thioredoxin fb has been studied during the enzyme reaction along the principle of Tian and Tsou [Biochemistry (1982) 21, 1028-1032]. A minimum model for this process is presented and its kinetic and equilibrium parameters have been determined. Thioredoxin fb binding to the enzyme is fast relative to catalysis and product desorption. Under quasi-equilibrium conditions oxidized thioredoxin is a non-competitive inhibitor of the enzyme reaction and must bind to a regulatory 'thioredoxin site'. The slow deactivation is thermodynamically favoured, and as expected from binding data, slowed down by the presence of substrate, fructose bisphosphate. The desorption of thioredoxin fb from the enzyme is extremely slow and this small protein may be regarded as a 'regulatory' subunit of fructose-bisphosphatase.

Kinetics of membrane-bound nitrate reductase A from Escherichia coli with analogues of physiological electron donors--different reaction sites for menadiol and duroquinol.

We have compared the steady-state kinetics of wild-type nitrate reductase A and two mutant forms with altered beta subunits. To mimic conditions in vivo as closely as possible, we used analogues of the physiological quinols as electron donors and membranes with overexpressed nitrate reductase A in preference to a purified alpha beta gamma complex. With the wild-type enzyme both menadiol and duroquinol supply their electrons for the reduction of nitrate at rates that depend on the square of the quinol concentration, menadiol having the higher catalytic constant. The results as a whole are consistent with a substituted-enzyme mechanism for the reduction of nitrate by the quinols. Kinetic experiments suggest that duroquinol and menadiol deliver their electrons at different sites on nitrate reductase, with cross-inhibition. Menadiol inhibits the duroquinol reaction strongly, suggesting that menaquinol may be the preferred substrate in vivo. To examine whether electron transfer from menadiol and duroquinol for nitrate reduction requires the presence of all of the Fe-S centres, we have studied the steady-state kinetics of mutants with beta subunits that lack an Fe-S centre. The loss of the highest-potential Fe-S centre results in an enzyme without menadiol activity, but retaining duroquinol activity; the kinetic parameters are within a factor of two of those of the wild-type enzyme, indicating that this centre is not required for the duroquinol activity. The loss of a low-potential Fe-S centre affects the activity with both quinols: the enzyme is still active but the catalytic constants for both quinols are decreased by about 75%, indicating that this centre is important but not essential for the activity. The existence of a specific site of reaction on nitrate reductase for each quinol, together with the differences in the effects on the two quinols produced by the loss of the Fe-S centre of +80 mV, suggests that the pathways for transfer of electrons from duroquinol and menadiol are not identical.

Antifungal action of Hevea brasiliensis latex. Its effect in combination with fluconazole on Candida albicans growth.

Latex from Hevea brasiliensis and its subcellular fractions (L-serum and C-serum) were tested for antifungal activity alone or in combination with fluconazole. Candida albicans growth was inhibited with the same efficacy when yeasts were inoculated into culture medium supplemented over the total growth phase with latex as when latex was added during the exponential phase only: the minimum inhibitory concentration (MIC 80%) of H. brasiliensis latex was 123 micrograms protein ml-1. By means of a non-linear regression analysis of the experimental data, two distinct fixation sites for fluconazole (FCZ) could be determined: one of strong affinity (Kaff = 0.0162 microgram-1 protein ml) and another of low affinity (Kaff = 0.0071 microgram-1 protein ml). After addition of a mixture of FCZ and latex during the exponential phase, the affinity constant of yeasts for FCZ was calculated: when latex was in a final concentration of 21 micrograms protein ml-1 (Kaff = 1 microgram-1 protein ml) or 42 micrograms protein ml-1 (Kaff = 0.277 microgram-1 protein ml) and without latex (Kaff = 0.0502 microgram-1 protein ml). In two cases a synergistic effect between latex and FCZ was obtained. The highest efficacy was obtained with a latex concentration of 21 micrograms protein ml-1. The addition of subcellular fractions of latex, L-serum and C-serum, did not cause an antifungal effect. The indispensable role of rubber particles for raising an antifungal effect is demonstrated. Electron microscopy observations indicated a limited cell wall degradation and a high percentage of coagulated yeasts.

On the activation of fructose-1,6-bisphosphatase of spinach chloroplasts and the regulation of the Calvin cycle.

Activation by light of spinach fructose-1,6-bisphosphatase is mimicked by dithiothreitol. This process of activation by dithiothreitol implies the specific reduction of two disulfide bridges and a conformation change of the enzyme that makes eight sulfhydryl groups available. The activated enzyme has an apparent allosteric kinetic behavior with respect to both fructose 1,6-bisphosphate and magnesium.

Substrate-binding isotherms of spinach chloroplastic fructose-1,6-bisphosphatase and the photoregulation of the Calvin cycle.

Fluorescence titration experiments of chloroplastic fructose-1,6-bisphosphatase by fructose bisphosphate and magnesium have been effected using the inactive dimeric, the inactive tetrameric and the active tetrameric enzyme forms. Magnesium binding to the inactive dimeric enzyme exhibits a positive cooperativity whereas fructose 1,6-bisphosphate exhibits no cooperativity at all. The binding of either magnesium or fructose bisphosphate to the inactive oxidized tetramer exhibits a succession of negative and positive cooperativities (mixed cooperativity). Upon reduction of the inactive tetramer by dithiothreitol and activation, the ligand binding properties of the enzyme are changed. Magnesium and fructose bisphosphate are bound to the active enzyme with a positive cooperativity. Non-linear least-square fitting allows one to estimate thea binding constants, and therefore the free energy of binding of either magnesium or fructose bisphosphate to the various forms of the enzyme. Whatever the state of fructose-1,6-bisphosphatase, one ligand is bound per subunit. The affinity constants of magnesium for the active or inactive enzyme forms are much greater than those of fructose bisphosphate. This suggests that the metal ion gives to the enzyme the right conformation to bind the sugar phosphate.

Affinity chromatography, on fructose-bisphosphatase-Sepharose, of two chloroplastic thioredoxins F. Purification and comparative molecular properties.

A new method of purification of chloroplastic thioredoxins has been presented. This method is based on affinity chromatography on fructose-bisphosphatase--Sepharose columns. Two thioredoxin, fA and fB, may be extracted and purified to homogeneity from the same leaf extract. Whereas fA is monomeric and has an Mr of 11 400 +/- 500, fB is dimeric with an Mr of 18 000 +/- 600. The dimer dissociates in two halves in the ultracentrifuge under the effect of high pressures. Raising the ionic strength results in the same effect. Thioredoxins fA and fB activate to similar extents chloroplastic fructose bisphosphatase and NADP--malate dehydrogenase. Chloroplastic sedoheptulose bisphosphatase is activated by thioredoxin fB but not by thioredoxin fA.