Protein stability and molecular adaptation to extreme conditions

Glutamate dehydrogenase from the thermoacidophilic archaebacterium Sulfolobus solfataricus: studies on thermal and guanidine-dependent inactivation

The hexameric NAD(P)-dependent glutamate dehydrogenase isolated from the thermoacidophilic archaebacterium Sulfolobus solfataricus shows a remarkable thermal stability which is strictly dependent on protein concentration (half-life at 95 degrees C is 0.25 h and 0.5 h at 0.4 and 0.8 mg/ml, respectively). Temperature-dependent inactivation of the enzyme is apparently irreversible; this process is accompanied by a progressive increase in hydrophobic surface area which leads to protein precipitation. 3 M GdnHCl increases the half-life of the enzyme at 90 degrees C and 0.2 mg/ml 6-fold. The hexamer is the only soluble molecular species revealed by glutaraldehyde fixation after thermal inactivation. Lyotropic salts strongly affect the enzyme thermal stability: the half-life at 90 degrees C and 0.2 mg/ml protein concentration increases more than 6-fold in the presence of 0.4 M Na2SO4 and decreases 4-fold in the presence of 0.4 M NaSCN. The maximum protein thermal stability is observed around the isoelectric pH, between pH 5.2 and pH 6.8. Guanidine-dependent inactivation of the enzyme at 20 degrees C is irreversible above 1.5 M GdnHCl. The decline in percentage of reactivation closely parallels the structural changes detected by fluorescence and the loss of hexameric structure accompanied by the dissociation to monomers, as indicated by glutaraldehyde fixation.

Characterization of the type I dehydroquinase from Salmonella typhi

The type I dehydroquinase from the human pathogen Salmonella typhi was overexpressed in an Escherichia coli host and purified to homogeneity. The S. typhi enzyme was characterized in terms of its kinetic parameters, important active-site residues, thermal stability and c.d. and fluorescence properties. In all important respects, the enzyme from S. typhi behaves in a very similar fashion to the well-characterized enzyme from E. coli, including the remarkable conformational stabilization observed on reduction of the substrate/product mixture by NaBH4. This gives confidence that the information from X-ray studies on the S. typhi enzyme [Boys, Fawcett, Sawyer, Moore, Charles, Hawkins, Deka, Kleanthous and Coggins (1992) J. Mol. Biol. 227, 352-355] can be applied to other type I dehydroquinases. Studies of the quenching of fluorescence of the S. typhi enzyme by succinimide show that NaBH4 reduction of the substrate/product imine complex involves a dramatic decrease in the flexibility of the enzyme, with only very minor changes in the overall secondary and tertiary structure.

Stabilization of creatinase from Pseudomonas putida by random mutagenesis

Creatinase (creatine amidinohydrolase, EC 3.5.3.3) from Pseudomonas putida is a homodimer of 45 kDa subunit molecular mass, the three-dimensional structure of which is known at 1.9 A resolution. Three point mutants, A109V, V355M, and V182I, as well as one double mutant combining A109V and V355M, and the triple mutant with all three replacements, were compared with wild-type creatinase regarding their physical and enzymological properties. High-resolution crystal data for wild-type creatinase and the first two mutants suggest isomorphism at least for these three proteins (R. Huber, pers. comm.). Physicochemical measurements confirm this prediction, showing that the mutations have no effect either on the quaternary structure and gross conformation or the catalytic properties as compared to wild-type creatinase. The replacement of V182 (at the solvent-exposed end of the first helix of the C-terminal domain) does not cause significant differences in comparison with the wild-type enzyme. The other point mutations stabilize the first step in the biphasic denaturation transition without affecting the second one. In sum, the enhanced stability seems to reflect slight improvements in the local packing without creating new well-defined bonds. The increase in hydrophobicity generated by the introduction of additional methyl groups (A109V, V182I) must be compensated by minor readjustments of the global structure. Secondary or quaternary interactions are not affected. In going from single to double and triple mutants, to a first approximation, the increments of stabilization are additive.

The L-lactate dehydrogenase gene of the hyperthermophilic bacterium Thermotoga maritima cloned by complementation in Escherichia coli

The gene for a L(+)-lactate dehydrogenase from the hyperthermophilic bacterium Thermotoga maritima was cloned by complementation of an Escherichia coli pfl. Idh mutant. The gene is part of a 4.5 kb SauIIIA fragment obtained by partial digestion of the Thermotoga genome. The DNA fragment was physically mapped and the putative Shine-Dalgarno sequence within the non-coding region determined. The gene contains 960 bp, including the stop codon, corresponding to 319 amino acids/subunit of the homotetrameric enzyme. Part of the amino acid sequence was confirmed by Edman degradation of peptides obtained from nanomolar quantities of the purified enzyme by tryptic digestion. A comparison of the amino acid sequence with those of known prokaryotic L-lactate dehydrogenases reveals a high similarity, especially with the enzyme from thermophilic sources, where up to 48% identity is found. The gene was expressed as an active enzyme in a heterologous host.

Destabilization of pea lectin by substitution of a single amino acid in a surface loop

Legume lectins are considered to be antinutritional factors (ANF) in the animal feeding industry. Inactivation of ANF is an important element in processing of food. In our study on the stability of Pisum sativum L. lectin (PSL), a conserved hydrophobic amino acid (Val103) in a surface loop was replaced with alanine. The mutant lectin, PSL V103A, showed a decrease in unfolding temperature (Tm) by some 10 degrees C in comparison with wild-type (wt) PSL, and the denaturation energy (delta H) is only about 55% of that of wt PSL. Replacement of an adjacent amino acid (Phe104) with alanine did not result in a significant difference in stability in comparison with wt PSL. Both mutations did not change the sugar-binding properties of the lectin, as compared with wt PSL and with PSL from pea seeds, at ambient temperatures. The double mutant, PSL V103A/F104A, was produced in Escherichia coli, but could not be isolated in an active (i.e. sugar-binding) form. Interestingly, the mutation in PSL V103A reversibly affected sugar-binding at 37 degrees C, as judged from haemagglutination assays. These results open the possibility of production of lectins that are active in planta at ambient temperatures, but are inactive and possibly non-toxic at 37 degrees C in the intestines of mammals.

Protein engineering for unusual environments

The ability to use proteins in unusual or non-natural environments greatly expands their potential applications in biotechnology. Because natural selection has neither maximized the stability of proteins nor optimized them to function under unusual conditions, there is considerable room for their improvement by protein engineering. Significant advances reported within the past year include a dramatic demonstration of a protein's ability to tolerate changes in its amino acid sequence, large increases in protein stability, and the use of random mutagenesis to obtain novel enzymatic properties. Approaches using random or site-directed mutagenesis have been successful in generating proteins able to function in an extended range of environments.

Functional expression of D-glyceraldehyde-3-phosphate dehydrogenase from the hyperthermophilic eubacterium Thermotoga maritima in Escherichia coli. Authenticity and kinetic properties of the recombinant enzyme

The gene coding for D-glyceraldehyde-3-phosphate dehydrogenase from the hyperthermophilic eubacterium Thermotoga maritima has been cloned and functionally expressed in Escherichia coli. Some 90% of the coding gene was amplified by the polymerase chain reaction. The amplified gene segment was in full agreement with the previously determined amino acid sequence [Schultes, V., Deutzmann, R., Jaenicke, R. (1990) Eur. J. Biochem. 192, 25-31] and was completed by the insertion of synthetic linkers using site-directed mutagenesis. The resulting semisynthetic gene was expressed in high yield in the cytoplasm of E. coli and the recombinant enzyme was purified to homogeneity. It was shown to be identical with the enzyme isolated directly from T. maritima in all enzymatic and physicochemical properties investigated. The enzyme is allosterically inhibited by both D- and L-glyceraldehyde 3-phosphate at concentrations above 1 mM, and by arsenate at concentrations above 10 mM. The expressed protein restores the natural E. coli phenotype in a gap- strain, thus providing evidence that the hyperthermophilic protein can fold and associate to its native, functional state in its mesophilic host.

Subunit interactions provide a significant contribution to the stability of the dimeric four-alpha-helical-bundle protein ROP

Detailed thermodynamic and spectroscopic studies were carried out on the ColE1-ROP protein in order to establish a quantitative basis for the contribution of noncovalent interactions to the stability of four-helix-bundle proteins. The energetics of both heat- and GdnHCl-induced denaturation were measured by differential scanning microcalorimetry (DSC) and/or by following the change in circular dichroism in the far-UV range. Sedimentation equilibrium analyses were performed to characterize the state of aggregation of the protein. No intermediate species could be detected during thermal unfolding of the dimer in the absence of GdnHCl. Under these conditions ROP unfolding exhibits a strict two-state behavior. The thermodynamic parameters for the reaction N2<->2D are delta HD = 580 +/- 20 kJ.(mol of dimer)-1, delta Cp = 10.3 +/- 1.3 kJ.(mol of dimer)-1.K-1, and Tm = 71.0 +/- 0.5 degrees C. The corresponding Gibbs energy change of unfolding is delta GD degree = 71.7 kJ.(mol of dimer)-1 at 25 degrees C and pH 6. In the presence of 2.5 M GdnHCl, however, ROP dissociates into monomers at elevated temperatures, as the loss of the concentration dependence of Tm and the decreased molecular weight demonstrate. The corresponding transition parameters are delta HD (2.5 M GdnHCl) = 130 +/- 10 kJ.(mol of monomer)-1 and Tm = 51.6 +/- 0.3 degrees C. Isothermal unfolding studies at 19 degrees C using GdnHCl as denaturant yielded a Gibbs energy change of unfolding of 22.4 kJ.(mol of monomer)-1. This extrapolated value is 38% lower than the corresponding delta GD degree value of 35.85 kJ.(mol of monomer)-1 calculated from thermal unfolding for the monomer in the absence of GdnHCl, where the protein is known to be a dimer. These results suggest that subunit interactions are an important source of stabilization of the native four-helix-bundle structure of ROP.

What does protein refolding in vitro tell us about protein folding in the cell?

The classical in vitro denaturation-renaturation studies by Anson, Anfinsen, Neurath, Pauling and others clearly suggested that the primary structure of proteins determines all higher levels of protein structure. Protein folding in the cell is inaccessible to a detailed analysis of its kinetic mechanism. There are obvious differences: nascent proteins acquire their native structure co- and post-translationally, with half-times in the minutes range, whereas refolding starts from the complete polypeptide chain, with rates varying from seconds to days. In the cell, accessory proteins are involved in regulating the rate of folding and association. Their role can be analysed both in vivo, by mutant studies, or by coexpression together with recombinant model proteins, and in vitro, by folding experiments in the absence and in the presence of 'foldases' and molecular chaperones, with the following general results: (i) folding is a sequential process involving native-like structural elements and a 'collapsed state' as early intermediates; (ii) the major side-reaction is caused by 'kinetic partitioning' between correct folding and wrong aggregation; (iii) rate-determining steps may be assisted by protein disulphide isomerase, peptidyl prolyl-cis-transisomerase, and molecular chaperones; and (iv) extrinsic factors, not encoded in the amino acid sequence, may be of crucial importance.

Substoichiometric amounts of the molecular chaperones GroEL and GroES prevent thermal denaturation and aggregation of mammalian mitochondrial malate dehydrogenase in vitro

The molecular chaperones GroEL and GroES were produced at very high levels in Escherichia coli, purified, and shown to protect pig mitochondrial malate dehydrogenase (MDH) against thermal inactivation in vitro. The apparent rate of MDH inactivation at 37 degrees C was reduced by a factor of at least 5 in a process which required only GroEL, GroES, and ATP. GroEL alone did not protect MDH against thermal inactivation but kept the denatured protein soluble and thereby prevented its aggregation. Reactivation of this soluble and inactive form of MDH could be achieved by addition of GroES even after 120 days of storage at -20 degrees C. Protection could be extended for more than 24 hr at 37 degrees C and was observed at molar ratios of chaperones to MDH as low as 1:4, suggesting that GroEL and GroES perform multiple turnovers in the absence of auxiliary chaperones. The availability of these chaperones in large quantities combined with the apparent promiscuity of GroEL binding shows great potential for stabilization of many proteins for which thermostable variants are not available. We speculate that GroEL and GroES perform similar protective roles in vivo and thereby increase the half-life of proteins which otherwise might aggregate under physiological conditions.

N5,N10-methenyltetrahydromethanopterin cyclohydrolase from the extremely thermophilic sulfate reducing Archaeoglobus fulgidus: comparison of its properties with those of the cyclohydrolase from the extremely thermophilic Methanopyrus kandleri

Archaeoglobus fulgidus and Methanopyrus kandleri are both extremely thermophilic Archaea with a growth temperature optimum at 83 degrees C and 98 degrees C, respectively. Both Archaea contain an active N5,N10-methenyltetrahydromethanopterin cyclohydrolase. The enzyme from M. kandleri has recently been characterized. We describe here the purification and properties of the enzyme from A. fulgidus. The cyclohydrolase from A. fulgidus was purified 180-fold to apparent homogeneity and its properties were compared with those recently published for the cyclohydrolase from M. kandleri. The two cytoplasmic enzymes were found to have very similar molecular and catalytic properties. They differed, however, significantly with respect of the effect of K2HPO4 and of other salts on the activity and the stability. The cyclohydrolase from A. fulgidus required relatively high concentrations of K2HPO4 (1 M) for optimal thermostability at 90 degrees C but did not require salts for activity. Vice versa, the enzyme from M. kandleri was dependent on high K2HPO4 concentrations (1.5 M) for optimal activity but not for thermostability. Thus the activity and structural stability of the two thermophilic enzymes depend in a completely different way on the concentration of inorganic salts. The molecular basis for these differences are discussed.

Comparison of stability properties of lactate dehydrogenase B4/epsilon-crystallin from different species

epsilon-Crystallin occurs as an abundant lens protein in many birds and in crocodiles and has been identified as heart-type lactate dehydrogenase (LDH-B4). Lens proteins have, due to their longevity and environmental conditions, extraordinary requirements for structural stability. To study lens-protein stability, we compared various parameters of LDH-B4/epsilon-crystallin from lens and/or heart of duck, which has abundant amounts of this enzyme in its lenses, and of chicken and pig, which have no epsilon-crystallin. Measuring the thermostability of LDH-B4 from the different sources, the t50 values (temperature at which 50% of the enzyme activity remains after a 20-min period) for LDH-B4 from duck heart, duck lens and chicken heart were all found to be around 76 degrees C, whereas pig heart LDH-B4 was less thermostable, having a t50 value of 62.5 degrees C. A similar tendency was found with urea inactivation studies. Plotting the first-order rate constants obtained from inactivation kinetic plots against urea concentration, it was clear that LDH-B4 from pig heart was less stable in urea than the homologous enzymes from duck heart, chicken heart and duck lens. The duck and chicken enzymes were also much more resistant against proteolysis than the porcine enzyme. Therefore, it is concluded that avian LDH-B4 is structurally more stable than the homologous enzyme in mammals. This greater stability might make it suitable to function as a crystallin, as in duck, but is not necessarily associated with high lens expression, as in chicken.

Salt dependence, kinetic properties and catalytic mechanism of N-formylmethanofuran:tetrahydromethanopterin formyltransferase from the extreme thermophile Methanopyrus kandleri

N-Formylmethanofuran(CHO-MFR):tetrahydromethanopterin(H4MPT) formyltransferase (formyltransferase) from the extremely thermophilic Methanopyrus kandleri was purified over 100-fold to apparent homogeneity with a 54% yield. The monomeric enzyme had an apparent molecular mass of 35 kDa. The N-terminal amino acid sequence of the polypeptide was determined. The formyltransferase was found to be absolutely dependent on the presence of phosphate or sulfate salts for activity. The ability of salts to activate the enzyme decreased in the order K2HPO4 > (NH4)2SO4 > K2SO4 > Na2SO4 > Na2HPO4. The salts KCl, NaCl and NH4Cl did not activate the enzyme. The dependence of activity on salt concentration showed a sigmoidal curve. For half-maximal activity, 1 M K2HPO4 and 1.2 M (NH4)2SO4 were required. A detailed kinetic analysis revealed that phosphates and sulfates both affected the Vmax rather than the Km for CHO-MFR and H4MPT. At the optimal salt concentration and at 65 degrees C, the Vmax was 2700 U/mg (1 U = 1 mumol/min), the Km for CHO-MFR was 50 microM and the Km for H4MPT was 100 microM. At 90 degrees C, the temperature optimum of the enzyme, the Vmax was about 2.5-fold higher than at 65 degrees C. Thermostability as well as activity of formyltransferase was dramatically increased in the presence of salts, 1.5 M being required for optimal stabilization. The efficiency of salts in protecting formyltransferase from heat inactivation at 90 degrees C decreased in the order K2HPO4 = (NH4)2SO4 >> KCl = NH4Cl = NaCl >> Na2SO4 > Na2HPO4. The catalytic mechanism of formyltransferase was determined to be of the ternary-complex type. The properties of the enzyme from M. kandleri are compared with those of formyltransferase from Methanobacterium thermoautotrophicum, Methanosarcina barkeri and Archaeoglobus fulgidus.

Characterization of the stabilizing effect of point mutations of pyruvate oxidase from Lactobacillus plantarum: protection of the native state by modulating coenzyme binding and subunit interaction

Point mutations in the gene of pyruvate oxidase from Lactobacillus plantarum, with proline residue 178 changed to serine, serine 188 to asparagine, and alanine 458 to valine, as well as a combination of the three single point mutations, lead to a significant functional stabilization of the protein. The enzyme is a tetrameric flavoprotein with tightly bound cofactors, FAD, TPP, and divalent metal ions. Thus, stabilization may be achieved either at the level of tertiary or quaternary interactions, or by enhanced cofactor binding. In order to discriminate between these alternatives, unfolding, dissociation, and cofactor binding of the mutant proteins were analyzed. The point mutations do not affect the secondary and tertiary structure, as determined by circular dichroism and protein fluorescence. Similarly, the amino acid substitutions neither modulate the enzymatic properties of the mutant proteins nor do they stabilize the structural stability of the apoenzymes. This holds true for both the local and the global structure with unfolding transitions around 2.5 M and 5 M urea, respectively. On the other hand, deactivation of the holoenzyme (by urea or temperature) is significantly decreased. The most important stabilizing effect is caused by the Ala-Val exchange in the C-terminal domain of the molecule. Its contribution is close to the value observed for the triple mutant, which exhibits maximum stability, with a shift in the thermal transition of ca. 10 degrees C. The effects of the point mutations on FAD binding and subunit association are interconnected. Because FAD binding is linked to oligomerization, the stability of the mutant apoenzyme-FAD complexes is increased. Accordingly, mutants with maximum apparent FAD binding exhibit maximum stability. Analysis of the quaternary structure of the mutant enzymes in the absence and in the presence of coenzymes gives clear evidence that both improved ligand binding and subunit interactions contribute to the observed thermal stabilization.

Identification of the gene encoding transcription factor NusG of Thermus thermophilus

The nusG gene of Thermus thermophilus HB8 was cloned and sequenced. It is located 388 bp downstream from tufB, which is followed by the genes for ribosomal proteins L11 and L1. No equivalent to secE preceding nusG, as in Escherichia coli, could be detected. The nusG gene product was overproduced in E. coli. A rabbit antiserum raised against the purified recombinant NusG reacted exclusively with one protein band of T. thermophilus crude extracts in Western blot (immunoblot) analyses, and no cross-reaction of the antiserum with E. coli NusG was observed. Recombinant NusG and the reacting T. thermophilus wild-type protein had identical sizes on sodium dodecyl sulfate-polyacrylamide gels. T. thermophilus and E. coli NusG have 45% identical and 22.5% similar amino acids, and similarities between the two proteins are most pronounced in carboxy-terminal regions. The T. thermophilus nusG gene could not rescue a nusG-deficient E. coli mutant strain.

Chemical crosslinking and the stabilization of proteins and enzymes

The technique of chemical crosslinking has been used to enhance the stability of proteins and enzymes. In this procedure, the molecule is braced with chemical crosslinks either intramolecularly or intermolecularly to another species to reinforce its active structure. Various chemicals have been used for this purpose. The bifunctional reagents are the most prominent. These compounds are derived from group-specific reagents and may be classified into homobifunctional, heterobifunctional, and zero-length crosslinkers. Different physical and chemical characteristics have been incorporated into these chemicals. Their versatility holds great potential in preparing chemically, thermally, and mechanically stable proteins and enzymes for industrial applications.

Correlation functions as a tool for protein modeling and structure analysis

Proteins present unique folding structures whose conformations are determined primarily by their amino acid sequences. At present, there is no algorithm that would correlate the sequences with the structures determined by X-ray analysis or NMR. Comparative modeling of a new protein sequence based on the known structure of a functionally related protein promises to yield model structures that may provide relevant properties of the protein. To analyze the quality of a model structure, a set of correlation functions was derived from calculations on a subset of proteins from the structure database. Twenty-three highly resolved protein structures with resolutions of at least 1.7 A from various protein families were used as the primary database. The purpose of this initial work was to find highly sensitive functions (including statistical error limits for the parameters) that describe properties of "real" proteins. Each correlation described is characterized by the correlation coefficient, the parameters for linear or nonlinear regression (coefficients of the equation), standard deviation and variance, and the confidence limits describing the statistical probability for values to occur within these limits, e.g., the natural variability of the property under examination. In addition, a method was developed for creating reasonably misfolded proteins. The ability of a correlation function to discriminate between the native structure and the misfolded conformations is expressed by the reliability index, which indicates the sensitivity of a correlation function. The term correlation functions thus summarizes a variety of efforts to find a mathematical description for the properties of protein structures, for their correlation, and for their significance.

Purification and partial characterization of a thermostable trithionate hydrolase from the acidophilic sulphur oxidizer Thiobacillus acidophilus

Cell-free extracts of Thiobacillus acidophilus catalysed the quantitative conversion of trithionate (S3O6(2-) to thiosulphate and sulphate. A continuous assay for quantification of experimental results was based on the difference in absorbance between trithionate and thiosulphate at 220 nm. Trithionate hydrolase was purified to near homogeneity from cell-free extracts of T. acidophilus. The molecular masses of the native enzyme and the subunit were 99 kDa (gel filtration) and 34 kDa (SDS/PAGE). The purified enzyme has a pH optimum of 3.5-4.5 and a temperature optimum of 70 degrees C. Enzyme activity was stimulated by sulphate. The stimulation of the enzyme activity by sulphate was half maximal at a concentration of 0.23 M. The Km for trithionate is 70 microM at 30 degrees C and 270 microM at 70 degrees C. Enzyme activity was lost after 36 days at 0 degrees C, 27 days at 70 degrees C; but after 97 days at 30 degrees C, 40% of the initial activity was still present: The enzyme activity was inhibited by mercury chloride, N-ethylmaleimide, thiosulphate and tetrathionate. Tetrathionate S4O6(2-) was not hydrolysed by trithionate hydrolase.

Molecular, functional and structural properties of an archaebacterial elongation factor 2

The elongation factor 2 (aEF-2) from the extreme thermo-acidophilic archaebacterium Sulfolobus solfataricus, is the only cytosolic target protein which is ADP-ribosylated by diphtheria toxin in presence of NAD. Once ADP-ribosylated, aEF-2 is no longer able to sustain poly(Phe) synthesis in vitro. aEF-2 displays a great thermoresistance: at the growth temperature of the archaebacterium, 87 degrees C, its half-life is 3 h. The amino acid sequence of the N-terminal region of aEF-2 has been determined up to residue 22. In the first 15 positions such a sequence is identical to that of EF-2 from Sulfolobus acidocaldarius and very similar to that of EF-2 from other archaebacteria or eukaryotes. The same is true for the primary structure of the peptide containing the ADP-ribosylation site. The fact that the primary structure of EF-2 at the ADP-ribosylation site is highly conserved ensures either the correct recognition of the histidine residue by the enzymes involved in its modification to diphthamide, or the proper interaction with the diphtheria toxin.

Evidence for temperature-dependent conformational changes in the L-lactate dehydrogenase from Bacillus stearothermophilus

L-Lactate dehydrogenase from Bacillus stearothermophilus (BSLDH) has been shown to change its conformation in a temperature-dependent manner in the temperature range between 25 and 70 degrees C. To provide a more detailed understanding of this reversible structural reorganization of the tetrameric form of BSLDH, we have determined in the presence of 5 mM fructose, 1,6-bisphosphate (FBP) the effect of temperature on far-UV and near-UV circular dichroism (CD), Nile red-binding to the enzyme surface, NADH binding, fluorescence polarization of fluorescamine-labeled protein, and hydrogen-deuterium exchange. In addition, we have analyzed the temperature dependence of the dimer-tetramer equilibrium of this protein by steady-state enzyme kinetics in the absence of FBP. The results obtained from these measurements at various temperatures can be summarized as follows. No changes in the secondary-structure distribution are detectable from far-UV CD measurements. On the other hand, near-UV CD data reveal that changes in the arrangements of aromatic side chains do occur. With increasing temperature, the asymmetry of the environment around aromatic residues decreases with a small change at 45 degrees C and a more pronounced change at 65 degrees C. Nile red-binding data suggest that the BSLDH surface hydrophobicity changes with temperature. It appears that decreasing the surface hydrophobicity may be a strategy to increase the protein stability of the active enzyme. We have noted significant alterations in the thermodynamic binding parameters of NADH above 45 degrees C, indicating a conformational change in the active site at 45 degrees C. The hydrodynamic volume of BSLDH is also temperature dependent.(ABSTRACT TRUNCATED AT 250 WORDS)