Sensitivity and variability of the Bradford protein assay in the presence of detergents

The effects of Triton X-100, sodium dodecyl sulfate (SDS), and urea on the response of Coomassie blue G to 16 different proteins and peptides of Mr 1140 to 146,000 were studied to assess the significance of protein conformation and of ionic and nonionic interactions for the dye response to individual proteins. Triton X-100 at a final concentration of 0.008% (v/v) increased the sensitivity of the Bradford assay toward all proteins of Mr 5700 or higher by an average 33%. Increases ranged from +11% with myelin basic protein to +128% with aprotinin. The relative range of absorbance of proteins and deviations from bovine serum albumin decreased by approximately 25%. Triton X-100 appears to facilitate nonionic interactions of the dye with proteins of limited capacity for ionic binding. Conformation of proteins also seemed to be of some significance because the chaotropic agent urea (0.16 M final concentration) increased sensitivity of the assay by 14%. Sensitivity of the assay was lowered by SDS (0.004% final concentration, w/v) by an average 75% from that of the control assay. The results indicate that the incorporation of low concentrations of a nonionic detergent may be useful in improving sensitivity and variability of the Bradford assay.     

A novel alkaline protease from wild edible mushroom Termitomyces albuminosus

A protease with a molecular mass of 30 kDa and the N-terminal sequence of GLQTNAPWGLARSS, was isolated from fresh fruiting bodies of the wild edible mushroom Termitomyces albuminosus. The purification protocol included ion exchange chromatography on DEAE-cellulose, Q-Sepharose, SP-Sepharose and FPLC-gel filtration on Superdex 75. The protein was unadsorbed on DEAE-cellulose and Q-Sepharose, but adsorbed on SP-Sepharose. The optimal pH and temperature of the purified enzyme were 10.6 and 60 °C, respectively. The enzyme was stable in the presence of 2 % (v/v) Tween 80 and 4 M urea. More than 80 % of the enzyme activity was retained in 2 % (v/v) Triton X 100, 54 % in 10 mM EDTA and 31 % in 2 % (w/v) SDS. The enzyme was strongly inhibited by phenylmethylsulfonyl fluoride (PMSF), but not inhibited by dithiothreitol (DTT), pepstatin or lima bean trypsin inhibitor suggesting that it was a serine protease but not a trypsin-like one. The protease was inhibited by Hg(2+), Cu(2+), and Fe(3+) ions. The K(m) and V(max) values of the purified enzyme for casein were 8.26 mg ? ml(-1) and 0.668 mg ? ml(-1) ? min(-1), respectively.     

Dithiothreitol activation of the insulin receptor/kinase does not involve subunit dissociation of the native alpha 2 beta 2 insulin receptor subunit complex

The subunit composition of the dithiothreitol- (DTT) activated insulin receptor/kinase was examined by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and gel filtration chromatography under denaturing (0.1% SDS) or nondenaturing (0.1% Triton X-100) conditions. Pretreatment of 32P-labeled insulin receptors with 50 mM DTT followed by gel filtration chromatography in 0.1% SDS demonstrated the dissociation of the alpha 2 beta 2 insulin receptor complex (Mr 400,000) into the monomeric 95,000 beta subunit. In contrast, pretreatment of the insulin receptors with 1-50 mM DTT followed by gel filtration chromatography in 0.1% Triton X-100 resulted in no apparent alteration in mobility compared to the untreated insulin receptors. Resolution of this complex by nonreducing SDS-polyacrylamide gel electrophoresis and autoradiography demonstrated the existence of the alpha 2 beta 2 heterotetrameric complex with essentially no alpha beta heterodimeric or free monomeric beta subunit species present. This suggests that the insulin receptor can reoxidize into the Mr 400,000 complex after the removal of DTT by gel filtration chromatography. Surprisingly, these apparently reoxidized insulin receptors were also observed to be functional with respect to insulin binding, albeit with a 50% decrease in affinity for insulin and insulin stimulation of the beta subunit autophosphorylation. To prevent reoxidation, the insulin receptors were pretreated with 50 mM DTT followed by incubation with excess N-ethylmaleimide prior to gel filtration chromatography in 0.1% Triton X-100. Under these conditions the insulin receptors migrated as the Mr 400,000 alpha 2 beta 2 complex.(ABSTRACT TRUNCATED AT 250 WORDS)     

A comparative study of the effect of various detergents on the structure and function of sarcoplasmic reticulum vesicles

Summary The effects produced by the detergents Triton X-100, sodium dodecylsulphate and sodium cholate on sarcoplasmic reticulum vesicles have been comparatively studied. In all cases, maximal effects are found 5 min after detergent addition. Triton X-100 and SDS are approximately ten times more effective than cholate in protein and phospholipid solubilization. Both Triton X-100 and SDS maintain Ca⁺⁺ accumulation in SR vesicles at detergent concentrations below 10^–3 M; higher concentrations cause a strong inhibition. On the other hand, cholate produces a gradual inhibition of Ca⁺⁺ accumulation in the concentration range between 10^–4 M and 2.5 × 10^–2 M. Triton X-100 and SDS produce a gradual solubilization of the specific Ca⁺⁺-ATPase activity up to a 10^–3 M detergent concentration, above which a strong inactivation occurs, while the enzyme solubilization increases with the presence of cholate in the whole concentration range under study. The different behaviour of sodium cholate, when compared to SDS or Triton X-100, is discussed in relation to the surfactant molecular structures. The possibility of membrane lysis and reassembly in the presence of some detergents is also considered.Abbreviations SR sarcoplasmic reticulum - SDS sodium dodecylsulphate - DTT dithiothreitol - EGTA ethyleneglycoltetraacetate - PEP phosphoenolpyruvate     

Effect of ethanol and methanol on the motility of Saccostrea commercialis sperm and sperm models

The organic solvents methanol and ethanol at concentrations of 2.5% and 5% (v/v), respectively, were found to significantly (P < 0.001) decrease the radius of curvature and track velocity of S. commercialis sperm. To observe the effects of the solvent directly on the axoneme, S. commercialis sperm models were prepared by extraction with Triton X-100 and reactivation with ATP in media containing acetate anions, DTT, magnesium, and cAMP. Concentrations of 0.1% Triton X-100 demembranated sperm while 0.01% and 0.05% Triton X-100 permeabilized sperm. Sperm models were successfully produced after reactivation with 1 mM ATP. At pH 8.25, 1% (v/v) ethanol or methanol was observed to increase waveform asymmetry and significantly (P < 0.001) decrease track velocity of 0.1% Triton X-100 demembranated sperm models. Similarly 1% (v/v) ethanol increased tailwave asymmetry and decreased track velocity of 0.01% and 0.05% Triton X-100 permeabilized sperm models. Reactivated motility of 0.05% Triton X-100 permeabilized sperm models prepared at pH 7.8 were poor and improved after treatment with 7% (v/v) ethanol, which increased waveform asymmetry and doubled the track velocity of sperm. This stimulatory effect of ethanol was unchanged in the presence of the alcohol dehydrogenase inhibitor pyrazole. Concerning the precise mechanism of action of ethanol on the axoneme, we conclude that a stimulatory or inhibitory effect of ethanol is dependent on the pH of the sperm model system used.     

Efficacy and compatibility with mass spectrometry of methods for elution of proteins from sodium dodecyl sulfate-polyacrylamide gels and polyvinyldifluoride membranes

The resolving power of sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) combined with isoelectric focusing in two-dimensional gel electrophoresis has made it one of the most important techniques for resolving complex mixtures, and it is of great importance for proteome mapping projects. As a result of this, methods for postelectrophoretic protein characterization are of great interest as exemplified by in situ protease digestion combined with mass spectrometry (MS), which is the method of choice for identification of proteins. In this study we have developed and compared methods for recovering intact proteins from polyacrylamide gels and electroblotting membranes to define efficient methods compatible with MS. These methods complement in situ digestion protocols and allow determination of the molecular mass of whole proteins separated by SDS-PAGE. Passive elution of proteins from SDS-PAGE gels was efficient only in the presence of SDS, whereas electroelution was achieved using buffers without SDS. Surface-enhanced laser desorption/ionization MS (SELDI-MS) analysis of proteins eluted in the presence of SDS was possible using ion exchange ProteinChip arrays for concentration of sample and removal of SDS. Comparison of different electroblotting methods verified that the different membranes and buffers were equally efficient for transfer of proteins in the range 20-100 kDa. Elution from polyvinyldifluoride membranes was most efficient using either concentrated solutions of trifluoroacetic acid (TFA) or combinations of 8M urea and 1% Triton X-100, 1% Tween 20, or 40% isopropanol. The same result was obtained using nitrocellulose membranes, except that these were incompatible with organic solvent and TFA. Elution by TFA was compatible with matrix-assisted laser desorption/ionization MS (MALDI-MS) but was complicated by a high degree of trifluoroacetylation of the proteins. Alternatively, elution by 8M urea+1% Triton X-100, 1% Tween 20, or 40% isopropanol was compatible with both SELDI-MS and MALDI-MS. Eluted proteins were identified in MS experiments by intact mass determination, by peptide mapping, and by MS/MS analysis.     

A method for direct application of human plasmin on a dithiothreitol-containing agarose stacking gel system

A new simplified procedure for identifying human plasmin was developed using a DTT copolymerized agarose stacking gel (ASG) system. Agarose (1 %) was used for the stacking gel because DTT inhibits the polymerization of acrylamide. Human plasmin showed the lowest activity at pH 9.0. There was a similar catalytically active pattern observed under acidic conditions (pH 3.0) to that observed under alkaline conditions (pH 10.0 or 11.0). Using the ASG system, the primary structure of the heavy chain could be established at pH 3.0. This protein was found to consist of three fragments, 45 kDa, 23 kDa, and 13 kDa. These results showed that the heavy chain has a similar structure to the autolysed plasmin (Wu et al., 1987b) but there is a different start amino acid sequence of the N-termini.     

Stability of the larvicidal activity of Bacillus thuringiensis subsp. israelensis: amino acid modification and denaturants.

The Bacillus thuringiensis subsp. israelensis mosquito larvicidal toxin is not a sulfhydryl-activated toxin. The protein disulfide bonds were cleaved and blocked without loss of toxicity. In contrast, modification of the lysine side chains eliminated toxicity. Additionally, the toxin was resistant to high concentrations of salt (8 M NaBr), organic solvents (40% methanol), denaturants (4 M urea), and neutral detergents (10% Triton X-100). However, it was inactivated by both positively and negatively charged detergents and by guanidine hydrochloride.     

Stability of the larvicidal activity of Bacillus thuringiensis subsp. israelensis: amino acid modification and denaturants

The Bacillus thuringiensis subsp. israelensis mosquito larvicidal toxin is not a sulfhydryl-activated toxin. The protein disulfide bonds were cleaved and blocked without loss of toxicity. In contrast, modification of the lysine side chains eliminated toxicity. Additionally, the toxin was resistant to high concentrations of salt (8 M NaBr), organic solvents (40% methanol), denaturants (4 M urea), and neutral detergents (10% Triton X-100). However, it was inactivated by both positively and negatively charged detergents and by guanidine hydrochloride.     

Quantitation of human immunoglobulin G and albumin in electroimmunodiffusion gels containing ionic and nonionic detergents

Quantitation of human immunoglobulin G (IgG) and albumin by agarose electroimmunodiffusion is influenced by the incorporation of ionic and nonionic detergents in the gel. The highest concentrations of each detergent at which human IgG and albumin determinations could be performed without perturbing the quantitations were 4% Triton X-100, 4% Tween 80, 1% NP-40, 0.5% sodium deoxycholate (SDOC), 0.5% Zwittergent, and 0.1% sodium dodecyl sulfate (SDS), and mixtures of Triton X-100, SDOC, and SDS. These detergent combinations all resulted in greater perturbations of albumin quantitation than of IgG. Immunoprecipitation of human IgG was quantitated in the absence and presence of Triton X-100, Zwittergent, and SDS. SDS was shown to cause nonspecific precipitation, whereas below 1% Triton X-100 or 0.5% Zwittergent no effects upon the immunoprecipitations were observed.     

Molecular cloning of a thermo-alkaliphilic lipase from Bacillus subtilis DR8806: Expression and biochemical characterization

A thermo-alkaliphilic lipase from Bacillus subtilis DR8806 was functionally expressed as an N-terminal 6xHis-tagged recombinant enzyme in Escherichia coli BL21 using pET-28a(+) expression vector. Sequence analysis revealed an open reading frame of 639 bp encoding a 212-amino acid protein containing the well-conserved Ala-His-Ser-Met-Gly motif. One-step purification of the His-tagged recombinant lipase was achieved using Ni-NTA affinity chromatography with a specific activity of 1364 U/mg. The purified enzyme with an apparent molecular mass of 26.8 kDa demonstrated the maximum activity at 70 °C and pH 8.0 for hydrolysis of p-nitrophenylbutyrate as substrate. The enzyme activity was strongly inhibited by divalent ions of heavy metals such as Hg²⁺ and Cu²⁺, while retained over 90% of the original activity in the presence of several reagents including DTNB (5,5′-dithiobis-(2-nitrobenzoic acid)), SDS (sodium dodecyl sulfate), urea, DMF (dimethylformamide), DTT (dithiothreitol), glycerol and Triton X-100. While being considerably stable in organic solvents, imidazolium-based ionic liquids (ILs) had stimulatory effects on the activity of purified lipase. Remarkable stabilization of enzyme at alkaline pH and in ionic liquids as well as its thermostability/thermoactivity are among the most fundamental characteristics which offer great potential for various biotechnological applications including detergent formulation, bioremediation processes and biotransformation in non-aqueous media.     

Isolation of Casparian Strips from Pea Roots

Casparian strips were isolated from roots of pea seedlings.The plasma membrane was firmly attached to the wall of the individualcells in the isolated Casparian strip. The outer electrondenselamella of the plasma membrane, which is characteristic of themembrane of the Casparian strip, was not removed from the isolatedCasparian strip by treatment with either 1 M NaCl or 2 M sodiumthiocyanate. Although treatment with 1% Triton X-100 disruptedthe lamellar structure of the membrane and removed most of theelectron-dense material of the outer lamella, no detectablepolypeptides were extracted by treatment of isolated Casparianstrips with 1% Triton X-100. Small amounts of electron-densematerial remained after treatment with Triton X-100 and thismaterial seemed to be removed almost entirely by subsequenttreatment of the same sample with 2% SDS. Since polypeptidesof 46, 30 and 20 kDa, which were detected in the extract obtainedby treatment with SDS, were not detected in an extract of totalcell walls of pea roots, it seems possible that these polypeptidesare specific to the Casparian strip and, therefore, that theymay be responsible singly or together for the adhesion of theplasma membrane to the cell wall at the Casparian strip. (Received January 30, 1992; Accepted April 23, 1992)     

Photodynamic effect in lysozyme: a kinetic study in different micellar media.

The influence of medium heterogeneity on the kinetics of the photodynamic effect on native protein lysozyme (Lyso), as well as the interaction of protein and the medium, anionic (SDS) micelles, neutral (Triton X-100) micelles and reversed micelles of AOT, were investigated at pH 8. The interaction between Lyso, Triton X-100 and SDS micelles was quantified by determining the respective associations constant (K(Lyso)). Values were 37 M(-1) for Triton X-100 and 514 M(-1) for SDS, indicating that the Lyso molecule binds Triton X-100 micelles effectively and SDS micelles even more strongly. Time-resolved phosphorescence detection (TRPD) indicates that the protein interacts with O2 (1deltag), with overall rate constants of the order of 10(8) M(-1)/S in direct micelles and 10(7) M(-1)/S in reverse micelles. Apparent reactive rate constants for eosin-sensitized photo-oxidation (singlet molecular oxygen [O2 (1deltag)]-mediated) of the protein were determined through oxygen uptake experiments for the direct micelles, while the fade in the protein fluorescence spectrum upon sensitized irradiation was used in AOT. The results indicate that the O2 (1deltag) attack on the interior of Lyso on amino acid residues, was more effective in leading to a photo-oxidative reaction in SDS and in Triton X-100 at surfactant concentrations < 1 x 10(-2) M than in a homogeneous solution. However, Lyso reactivity reached a maximum when the concentration of micelles was approximately 1 x 10(-5), the same as the protein concentration In AOT reverse micelles, the quenching rate constants decreased > 75% with respect to water. This effect can be attributed to the decrease in accessibility of the amino acid residues to O2 (1deltag).     

Preparation and properties of soluble, immunoreactive apoLDL.

Immunoreactive apo-low density lipoprotein (LDL), soluble in mildly alkaline buffers of low ionic strength, was prepared by attaching LDL to a DEAE-Sepharose column and eluting the lipids with a 0--2% (w/v) gradient of nonionic detergents. Brij-36T, Nonidet P-40, and Triton X-100 gave similar results. After washing the detergent from the column, the bound apoLDL was eluted with 1 M NaCl, pH 7.4, with recoveries up to 85%. This apoLDL could be dialyzed extensively against low ionic strength solutions, and remained soluble as long as the pH was above 7. Spectrophotometric analysis showed that less than 0.1% %w/v) of cholesterol or phospholipids and less than 1% (w/v) of detergent remained associated with the protein. The apoLDL cross-reacted with LDL against antisera prepared vs. intact LDL. Pore-gradient polyacrylamide gel electrophoresis, with SDS and urea, showed that this preparation was less aggregated than organic solvent extracted apolLDL and appeared to be made of oligomers of two monomeric subunits, one with molecular weight around 22,700 and a smaller one of approximately 8000. Isoelectric focusing showed that there also was charge heterogeneity in the soluble apoLDL.     

Purification and chemical modification of a phosphatidylinositol kinase from sheep brain.

A membrane-bound phosphatidylinositol (PtdIns) kinase has been purified approximately 9500-fold to apparent homogeneity from sheep brains. The purification procedure involves: solubilisation of the membrane fraction with Triton X-100, ammonium sulphate fractionation and a number of ion-exchange and gel-filtration chromatography steps. The purified enzyme exhibited a final specific activity of 1149 nmol.min-1.mg-1. The molecular mass of the enzyme was estimated to be 55 kDa by SDS/PAGE and 150 +/- 10 kDa by HPLC gel filtration in the presence of Triton X-100. Kinetic measurements have shown that the apparent Km value of PtdIns kinase for the utilisation of PtdIns is 22 microM and for ATP 67 microM. Mg2+ was the most effective divalent cation activator of PtdIns kinase, with maximal enzymatic activity reached at a concentration of 10 mM Mg2+. In addition to adenosine and ADP, the 2'(3')-O-(2,4,6-trinitrophenyl) derivative of ATP was found to be a strong competitive inhibitor of the enzyme, with a Ki of 32 microM. Enzymatic activity was found to be stimulated by Triton X-100 but inhibited by deoxycholate.     

Stabilizing hyperactivated lecitase structures through physical treatment with ionic polymers

Lecitase Ultra has been covalently immobilized on cyanogen bromide cross-linked 4% agarose (CNBr) beads, maintaining 70% of the initial activity. The activity of the immobilized enzyme was improved in the presence of Triton X-100, sodium dodecyl sulfate (SDS), and cetyltrimethyl ammonium bromide (CTAB) (e.g., up to 800% when using CTAB). However, CTAB and Triton X-100 presented a negative effect on enzyme stability even at low concentrations, and SDS cannot be used for a long time at 1% concentration. To maintain the hyperactivated conformation of the enzyme in the absence of detergent, ionic polymers were added during incubation of the immobilized enzyme in the presence of detergents. Coating the immobilized enzyme with polyethylenimine in aqueous buffer (PEI) produced a 3-fold increase in enzyme activity. However, in the presence of 0.1% SDS (v/v), this coating produced a 50-fold increase in enzyme activity. Using PEI and 0.01% (v/v) CTAB, the Lecitase activity decreased to 10%. Using irreversible inhibitors, it could be shown that the PEI/SDS-CNBr-Lecitase preparation allowed its catalytic Ser to be more accessible to the reaction medium than the unmodified CNBr-Lecitase.     

Effects of ionic and nonionic detergents on antigen-antibody reactions.

Using highly sensitive and quantitative radioimmunoassay procedures we have measured the effects of different concentrations of three commonly used detergents, SDS, DOC, and Triton X-100, on antibody-antigen reactions. Triton X-100, had a relatively mild effect on primary antigen-antibody bindings, the precipitin reaction, and a double antibody RIA as evidenced by only an 8 to 10% inhibition of binding or precipitation. These results were not detergent concentration dependent, as Triton concentrations ranging from 5 to 0.1% had virtually no differential effects. Sodium deoxycholate (DOC) had a more profound effect on both primary antigen-antibody binding and the precipitin reaction than did Triton X-100, and its effects, unlike those of Triton X-100, were concentration dependent. There was a direct relationship between concentration of DOC and degree of inhibition of both primary binding and immune precepitation especially in antigen excess. Sodium dodecylsulfate (SDS), at concentrations 10- to 100-fold less than either Triton X-100 or DOC, had profound inhibitory effects on primary antigen-antibody binding, the precipitin reaction, and a double antibody radioimmunoassay. Generally, at concentrations greater that 0.01% SDS, almost all immunochemical reactivity is destroyed.     

Renaturation of recombinant Treponema pallidum rare outer membrane protein 1 into a trimeric, hydrophobic, and porin-active conformation

We have previously observed that while native Treponema pallidum rare outer membrane protein 1 (Tromp1) is hydrophobic and has porin activity, recombinant forms of Tromp1 do not possess these properties. In this study we show that these properties are determined by conformation and can be replicated by proper renaturation of recombinant Tromp1. Native Tromp1, but not the 47-kDa lipoprotein, extracted from whole organisms by using Triton X-114, was found to lose hydrophobicity after treatment in 8 M urea, indicating that Tromp1's hydrophobicity is conformation dependent. Native Tromp1 was purified from 0.1% Triton X-100 extracts of whole organisms by fast-performance liquid chromatography (FPLC) and shown to have porin activity in planar lipid bilayers. Cross-linking studies of purified native Tromp1 with an 11 A cross-linking agent showed oligomeric forms consistent with dimers and trimers. For renaturation studies of recombinant Tromp1 (rTromp1), a 31,109-Da signal-less construct was expressed in Escherichia coli and purified by FPLC. FPLC-purified rTromp1 was denatured in 8 M urea and then renatured in the presence of 0.5% Zwittergent 3,14 during dialysis to remove the urea. Renatured rTromp1 was passed through a Sephacryl S-300 gel exclusion column previously calibrated with known molecular weight standards. While all nonrenatured rTromp1 eluted from the column at approximately the position of the carbonic anhydrase protein standard (29 kDa), all renatured rTromp1 eluted at the position of the phosphorylase b protein standard (97 kDa), suggesting a trimeric conformation. Trimerization was confirmed by using an 11 A cross-linking agent which showed both dimers and trimers similar to that of native Tromp1. Triton X-114 phase separations showed that all of renatured rTromp1, but none of nonrenatured rTromp1, phase separated exclusively into the hydrophobic detergent phase, similar to native Tromp1. Circular dichroism of nonrenatured and renatured rTromp1 showed a marked loss in alpha-helical secondary structure of renatured rTromp1 compared to the nonrenatured form. Finally, renatured rTromp1, but not the nonrenatured form, showed porin activity in planar liquid bilayers. These results demonstrate that proper folding of rTromp1 results in a trimeric, hydrophobic, and porin-active conformation similar to that of the native protein.     

The presumption of proteolytic enzymes related to the formation of intermediates during the terminal differentiation

The [¹⁴C]Gly-labelled keratin polypeptides extracted with 1% SDS and 10 mM DTT were made to undergo changes with an enzyme fraction (ammonium sulfate, 50–75% saturated fraction) prepared from a human epidermis in the presence of 1% Triton X-100. In particular, 69-67 kDa peptides were considerably decreased with the above enzyme fraction in the time course experiments, and the components strongly bound to the cell membrane had little effect on the above reaction. In addition, in the case of the [¹⁴C]Gly-labelled keratin filament assembly, 69 and 62 kDa peptides were decreased and 55, 52 and 50 kDa peptides were increased with the same enzyme fraction in the time course experiments. From these results, we estimated that the proteolytic enzyme(s) may exist in the human epidermis, and may be processed to keratin intermediates from prekeratin during the initial stage of terminal differentiation in the human epidermis.     

Glucose dehydrogenase in beef (Bos taurus) and rainbow trout (Oncorhynchus mykiss) liver: a comparative study

The monomer molecular mass of glucose dehydrogenase (GDH, EC 1.1.1.47) from rainbow trout liver and beef liver were estimated to be 90 kDa for both enzymes, by electrophoresis in the presence of Na-dodecyl-SO₄ (SDS). The 90-kDa proteins were partially degraded to about 60 kDa when purified with a delayed procedure without protease inhibitors. Tryptic cleavage of the 90-kDa proteins gave fragments of about 60 kDa and 30 kDa, being similar for trout and beef GDH. Isoelectric points, kinetic and thermodynamic properties of the two enzymes are markedly different. Triton X-100 stimulated and stabilized the reactions catalysed by the purified enzymes.