Linear Arrhenius Plot Research Articles

Effects of the microenvironment on the photophysical properties of hematoporphyrin and its efficacy as a photosensitizer

Percoll had a pronounced influence on the Q bands of the absorption spectrum of hematoporphyrin 1 and, to a lesser extent, of uroporphyrin. As the Q hands are very sensitive to the microenvironment, this indicates that Percoll perturbs this microenvironment. The hematoporphyrin-sensitized photooxidation of histidine and tyrosine exhibited in buffer solutions a linear Arrhenius plot in the temperature range 6–37 °C. Addition of Percoll or Sephadex LH20 to the solution resulted in the appearance of a breaking point around 15 °C in this plot. As these agents have in common that they induce the formation of vicinal water structures and because these structures undergo an abrupt structural transition. amongst others, around 15 °C, these results confirm the effect of the miccro-environment on the photophysical properties of hematoporphyrin. Percoll also induced a breaking point at ≈ 15°C in the Arrhenius plot of uroporphyrin, but not of rose bengal-and aluminum phthalocyanine tetrasulfonate-mediated photooxidation of histidine, indicating that the photophysical properties of these two sensitizers are insensitive to the water structure. Photodynamic inactivation of glyceraldehyde-3-phosphate dehydrogenase dissolved in buffer, with hematoporphyrin, rose bengal or aluminum phthalocyanine tetrasulfonate as sensitizer revealed a linear relationship between log v and 1/ T over the temperature range 6–37 °C. In contrast, with yeast hexokinase, the Arrhenius plot exhibited a sharp breaking point around 30 °C with all three sensitizers. This breaking point should also be ascribed to the effect of the vicinal water layer, in this case on the conformation of the enzyme molecule. Apparently the water structure can effect photochemical reactions by at least two different mechanisms: i. by affecting the photophysical properties of some sensitizers, like porphyrins, and ii. by affecting the conformation of some macromolecular targets, a mechanism not dependent on the type of sensitizer.

Indium diffusion study in α-titanium

The diffusion of implanted In α-Ti has been studied in the 823–1073 K temperature range by using the Rutherford backscattering spectrometry (RBS) technique. The measurements show that the diffusion coefficients follow a linear Arrhenius plot: D( T) = (2.0 ± 1.3) × 10 −6exp[−(260 ± 40) kJ/mol/ RT]m 2 s −1. The diffusion parameters D 0 and Q are typical of a normal substitutional behavior. Comparison of the present and previous published results of impurity diffusion in α-Ti does not show evidence for mass or size effects in the diffusion mechanism.

Oxygen diffusion in GdBa2 Cu3O(6+y) thin films

The oxygenation process of the GdBa2 Cu3O(6+y) thin films was investigated by resistivity measurements in the temperature interval 650–1050 K and oxygen pressure 1 mbar. An iteration procedure is proposed for calculating the diffusion relaxation times from the curves of resistance decay with time. All samples showed a linear Arrhenius plot with an activation energy E0 close to 1.5 eV. The diffusion resulted concentration independent for y≈0.4–0.9.

Diffusion study of Sn implanted in α-Ti

Abstract The diffusion of implanted Sn in α-Ti has been studied in the 873–1073 K temperature range using the Rutherford back-scattering spectrometry technique. For this purpose we have implanted Sn in α-Ti samples with two different impurity contents. The measurements in those with a lower level of impurities show that the diffusion coefficients follow a linear Arrhenius plot with the following parameters: The measurements in the samples with higher Fe content indicate that the diffusion mechanism is sensitive to the impurity concentration.

Sugar transport heterogeneity in the kidney: two independent transporters or different transport modes through an oligomeric Protein? 1. Glucose transport studies.

The kinetics of Na+/d-glucose cotransport (SGLT) were reevaluated in rabbit renal brush border membrane vesicles isolated from the whole kidney cortex using a fast-sampling, rapid-filtration apparatus (FSRFA, US patent #5,330,717) for uptake measurements. Our results confirm SGLT heterogeneity in this preparation, and both high (HAG) and low (LAG) affinity glucose transport pathways can be separated over the 15-30 degrees C range of temperatures. It is further shown that: (i) Na+ is an essential activator of both HAG and LAG; (ii) similar energies of activation can be estimated from the linear Arrhenius plots constructed from the Vmax data of HAG and LAG, thus suggesting that the lipid composition and/or the physical state of the membrane do not affect much the functioning of SGLT; (iii) similar Vmax values are observed for glucose and galactose transport through HAG and LAG, thus demonstrating that the two substrates share the same carrier agencies; and (iv) phlorizin inhibits both HAG and LAG competitively and with equal potency (Ki = 15 microM). Individually, these data do not allow us to resolve conclusively whether the kinetic heterogeneity of SGLT results from the expression in the proximal tubule of either two independent transporters (rSGLT1 and rSGLT2) or from a unique transporter (rSGLT1) showing allosteric kinetics. Altogether and compared to the kinetic characteristics of the cloned SGLT1 and SGLT2 systems, they do point to a number of inconsistencies that lead us to conclude the latter possibility, although it is recognized that the two alternatives are not mutually exclusive. It is further suggested, from the differences in the Km values of HAG transport in the kidney as compared to the small intestine and SGLT1 cRNA-injected oocytes, that renal SGLT1 activity is somehow modulated, maybe through heteroassociation with (a) regulatory subunit(s) that might also contribute quite significantly to sugar transport heterogeneity in the kidney proximal tubule.

Anomalous depth profiles and diffusional behavior of noble gases implanted into photoresist films

In the present work are reported results of thermal diffusion of Kr implanted into AZ1350 positive photoresist films and Xe implanted into “waycoat” negative photoresist films. Kr and Xe ions were implanted at 80 K and the depth profiles determined “in situ” by using the Rutherford backscattering technique in a 80–273 K temperature range. The extracted Kr diffusion coefficients follow a linear Arrhenius plot over the studied temperature range with a characteristic activation energy of E b = 100 meV. The Xe diffusion into the “waycoat” photoresist develops in a very narrow temperature range (210–230 K) which only allows determination of the diffusion coefficient at 230 K. Comparison between the present and previous published results show that Kr and Xe display very similar diffusional characteristics when implanted in the AZ1350 photoresist. On the other side, the diffusion into the “waycoat” photoresist seems to be much faster than in the AZ1350 film.

Diffusion study of Kr, Rb and Xe implanted into positive and negative photoresist films

In the present work results are reported of thermal diffusion of Kr and Rb implanted into AZ1350 positive photoresist films and Xe implanted into “waycoat” negative photoresist films. Kr and Xe ions were implanted at 80 K and the depth profiles determined in situ by using the Rutherford backscattering technique in a 80–473 K and 80–273 K temperature range respectively. The extracted Kr diffusion coefficients follow a linear Arrhenius plot over the studied temperature range with a characteristic activation energy of E b = 100 meV. The Xe diffusion into the “waycoat” photoresist develops in a very narrow temperature range (210–230 K) which only allows determination of the diffusion coefficient at 230 K. The Rb diffusional behavior was studied in a 273–473 K temperature range and it is characterized by an activation energy of E b = 300 meV. Comparison between the present and previous published results show that when implanted into the same polymer Kr and Xe display very similar diffusional characteristics. On the other side, the diffusion into the “waycoat” photoresist seems to be much faster than in the AZ1350 film. Finally a comparison between Rb and Kr results indicates that the Rb-photoresist bonds play an important role in the thermal diffusion process.

Radiation chemistry of aqueous solutions of hydrazine at elevated temperatures. Part 1.—Oxygen-free solutions

Rate constants for the reactions of eaq– and ˙OH with N2H4 and N2H5+ in liquid water up to 200 °C have been measured by pulse radiolysis. Linear Arrhenius plots for the reactions of eaq– gave k(20 °C)= 106 and 1.6 × 108 dm3 mol–1 s–1, and Ea= 13.5 and 18.2 kJ mol–1, respectively. H is the product of the reaction with N2H5+. Non-linear Arrhenius behaviour was observed for the reactions of ˙OH with k(20 °C)= 4.5 × 109 and 8.2 × 107 dm3 mol–1 s–1, respectively. The pKa of N2H5+ decreases linearly with temperature from 8.1 at 20 °C to 4.2 at 200 °C. The products of the ˙OH reactions are ˙N2H3 and ˙N2H4+, respectively, and the pKa of ˙N2H4+ also decreases with increasing temperature. The self-reaction of ˙N2H3 shows the same temperature dependence as that of ˙OH with k(20 °C)= 2 × 109 dm3 mol–1 s–1. The product of this reaction is tetrazene. Up to 200 °C the data are consistent with successive eliminations of NH3 to form triazene and then N2. The pH-dependent kinetics of these processes indicate that the decomposition of N3H3 is acid- and base-catalysed over the whole temperature range.

Variable-Temperature Studies of Laser-Initiated 5T2 → 1A1 Intersystem Crossing in Spin-Crossover Complexes: Empirical Correlations between Activation Parameters and Ligand Structure in a Series of Polypyridyl Ferrous Complexes

Results are presented from a variable-temperature solution-phase laser photolysis study of the 5T2 → 1A1 intersystem crossing in a series of related complexes: [Fe(tpen)](ClO4)2, [Fe(tppn)](ClO4)2, [Fe(tptn)](ClO4)2, [Fe(t-tpchxn)](ClO4)2, and [Fe(dpa)2](ClO4)2. The hexadentate ligands are formed with four 2-pyridylmethyl arms attached to ethylenediamine (tpen), 1,2-diaminopropylene (tppn), 1,3-diaminopropylene (tptn), or trans-1,2-diaminocyclohexane (t-tpchxn). The dpa ligand is a tridentate analogue of these ligands, namely, bis(2-pyridylmethyl)amine. The complex [Fe(mtpen)](ClO4)2·2/3H2O, where mtpen is the same as tpen except one of the pyridyl rings has a 6-methyl substituent, crystallizes in the space group C2/c, which at 173 K has a unit cell with a = 41.390(13) Å, b = 9.5239(16) Å, c = 24.016(6) Å, β = 108.24(3)°, and Z = 12. Refinement with 2844 observed [F > 5.0σ(F)] reflections gave R = 0.075 and Rw = 0.076. The complex [Fe(tppn)](ClO4)2·2/3H2O crystallizes in the space group P21/n, which at 296 K has a unit cell with a = 12.979(4) Å, b = 12.624(4) Å, c = 19.475(6) Å, β = 108.17(2)°, and Z = 4. Refinement with 2357 observed [F > 5.0σ(F)] reflections gave R = 0.1198 and Rw = 0.1141. The mtpen complex is a high-spin FeII complex at all temperatures (4.2−400 K). The hydrated tpen complex is a spin-crossover complex with the temperature where there are 50% high-spin complexes (T1/2) is ≃ 385 K, the hydrated tppn complex is also spin-crossover with a higher T1/2 value, and the hydrated tptn complex is low spin up to 400 K. The present crystallographic results, together with previously reported structural results for the tpen complex at two temperatures, are used to show that the conversion from low spin to high spin leads to an increase in the trigonal twist of these distorted octahedral complexes. The influence of this variation in trigonal twist on the rate of 5T2 → 1A1 intersystem crossing is examined with variable-temperature laser-flash photolysis. Data collected for the tpen complex in MeOH in the 190−294 K range give a linear Arrhenius plot with an activation energy of Ea = 767 ± 22 cm-1 and a preexponential term of A = (1.35 ± 0.2) × 109 s-1. The tppn complex gives similar results of Ea = 771 ± 17 cm-1 and A = (1.45 ± 0.2) × 109 s-1. At 294 K the rate (k-1) for the 5T2 → 1A1 intersystem crossing is 2.87 × 107 s-1 for the tpen complex, and 3.21 × 107 s-1 for the tppn complex. On the other hand the tptn complex has k-1 = 6.25 × 108 s-1 at 295 K as measured with a picosecond spectrometer, and together with nanosecond data measured in the 186−210 K range gives an Arrhenius activation energy of Ea = 777 ± 50 cm-1 with A = (2.6 ± 0.8) × 1010 s-1. The bis(tridentate) complex [Fe(dpa)2](ClO4)2 in MeOH is found to give k-1 = 4.59 × 107 s-1 at 282 K and with the 191−282 K data gives Arrhenius values of Ea = 339 ± 13 cm-1 and A = (2.5 ± 0.25) × 108 s-1. The terpyridine complex [Fe(terpy)2](ClO4)2 is found to have k-1 = 1.0 × 108 s-1 in MeOH at 239 K, and the analysis of 190−239 K data gives Ea = 532 ± 36 cm-1 and A = (2.4 ± 0.4) × 109 s-1. Previous studies have shown that the greater the trigonal twist, the lower in energy is the 3T1 state which facilitates the spin−orbit interaction between the 5T2 high-spin and 1A1 low-spin states. It is suggested that the trigonal twist is a vibrational coordinate strongly coupled to the 5T2 → 1A1 intersystem crossing. It is additionally shown that the data are consistent with a model wherein the “intrinsic” rate of 5T2 → 1A1 intersystem crossing, as gauged by the preexponential term, is a function of how far along the reaction coordinate a complex proceeds.

Effect of Roasting Time and Temperature on the Generation of Nonvolatile (Polyhydroxyalkyl)pyrazine Compounds in Peanuts, As Determined by High-Performance Liquid Chromatography

A high-performance liquid chromatographic (HPLC) method was used for the determination of (polyhydroxyalkyl)pyrazine (PHAP) compounds in peanuts roasted to different levels as part of a 3×3 time/temperature study. The content of volatile alkylpyrazine compounds generated at each point was determined by headspace gas chromatography−mass spectrometry (GC−MS) and compared to the PHAP content. In general, results showed that the levels of both PHAP and alkylpyrazine compounds increased with both time and temperature, reaching a maximum at the longest roast time and highest temperature. Formation rates for both types of compounds showed a strong dependence on temperature, giving linear Arrhenius plots with similar slopes. There was no indication that the PHAP compounds formed during roasting decomposed to form alkylpyrazines. Keywords: Roast peanuts; (polyhydroxyalkyl)pyrazines; alkylpyrazines; LC-MS; flavor generation; headspace GC−MS

Rates of Uncatalyzed Peptide Bond Hydrolysis in Neutral Solution and the Transition State Affinities of Proteases

To assess the relative proficiencies of enzymes that catalyze the hydrolysis of internal and C-terminal peptide bonds, the rates of the corresponding nonenzymatic reactions were examined at elevated temperatures in sealed quartz tubes, yielding linear Arrhenius plots. The results indicate that in neutral solution at 25 °C, peptide bonds are hydrolyzed with half-times of approximately 500 years for the C-terminal bond of acetylglycylglycine, 600 years for the internal peptide bond of acetylglycylglycine N-methylamide, and 350 years for the dipeptide glycylglycine. These reactions, insensitive to changing pH or ionic strength, appear to represent uncatalyzed attack by water on the peptide bond. Comparison of rate constants indicates very strong binding of the altered substrate in the transition states for the corresponding enzyme reactions, Ktx attaining a value of less than 10-17 M in carboxypeptidase B. The half-life of the N-terminal peptide bond in glycylglycine N-methylamide, whose hydrolysis might have provided a reference for assessing the catalytic proficiency of an aminopeptidase, could not be determined because this compound undergoes relatively rapid intramolecular displacement to form diketopiperazine (t1/2 ∼ 35 days at pH 7 and 37 °C). The speed of this latter process suggests an evolutionary rationale for posttranslational N-acetylation of proteins in higher organisms, as a protection against rapid degradation.

Evidence for the reducibility of multifragment emission to an elementary binary emission in Xe-induced reactions

Multifragmentation for 129Xe-induced reactions on several targets ( natCu, 89Y, 165Ho, 197Au) has been studied at E A = 40, 50 and 60 MeV . The probability of emitting n intermediate mass fragments is shown to be binomial at each transversal energy and reducible to an elementary binary probability p. For each target and at each bombarding energy, p shows a thermal behavior as demonstrated by linear Arrhenius plots. A nearly universal linear Arrhenius plot is observed at each bombarding energy, independent of target. The sensitivity of p to the Z threshold points to its physical significance and rules out auto-correlation effects.

Dynamic spin-crossover in [Fe II(TRIM) 2]Cl 2 investigated by Mössbauer spectroscopy and magnetic measurements

The magnetic and Mössbauer data of the new coordination complex [Fe II(TRIM) 2]CI 2, where TRIM = 4′-(4-methylimidazole-2′- (2″-methylimidazole)-imidazole) afford evidence for a complete and progressive spin conversion between the 1A 1 and 5T 2 spin states. The interstate conversion frequencies measured in the 220–300 K temperature range follow an Arrhenius behavior. The thermal variation of the equilibrium constant over the whole temperature range approximately follows a linear Arrhenius plot, thus indicating a very low degree of cooperativity in this material; a small distortion from the linear plot suggests a vibrational effect.

Rapid long-range proton diffusion along the surface of the purple membrane and delayed proton transfer into the bulk.

The pH-indicator dye fluorescein was covalently bound to the surface of the purple membrane at position 72 on the extracellular side of bacteriorhopsin and at positions 101, 105, 160, or 231 on the cytoplasmic side by reacting bromomethylfluorescein with the sulfhydryl groups of cysteines introduced by site-directed mutagenesis. At position 72, on the extracellular surface, the light-induced proton release was detected 71 +/- 4 microseconds after the flash (conditions: pH 7.3, 22 degrees C, and 150 mM KCl). On the cytoplasmic side with the dye at positions 101, 105, and 160, the corresponding values were 77, 76, and 74 +/- 5 microseconds, respectively. Under the same conditions, the proton release time in the bulk medium as detected by pyranine was around 880 microseconds--i.e., slower by a factor of more than 10. The fact that the proton that is released on the extracellular side is detected much faster on the cytoplasmic surface than in the aqueous bulk phase demonstrates that it is retained on the surface and migrates along the purple membrane to the other side. These findings have interesting implications for bioenergetics and support models of local proton coupling. From the small difference between the proton detection times by labels on opposite sides of the membrane, we estimate that at 22 degrees C the proton surface diffusion constant is greater than 3 x 10(-5) cm2/s. At 5 degrees C, the proton release detection time at position 72 equals the faster of the two main rise times of the M intermediate (deprotonation of the Schiff base). At higher temperatures this correlation is gradually lost, but the curved Arrhenius plot for the proton release time is tangential to the linear Arrhenius plot for the rise of M at low temperatures. These observations are compatible with kinetic coupling between Schiff base deprotonation and proton release.

Adenylate cyclase activity from Hirudo medicinalis segmental ganglia: modulation by temperature and Mg 2+ ions

The basal, serotonin- and fluoride-stimulated activities of homogenate preparations from Hirudo medicinalis segmental ganglia increase with incubation temperature, reaching a maximum at 35°C. Both basal and fluoride-stimulated activities yield linear Arrhenius plots, whereas serotonin-enhanced activity shows a break at 25°C. Increasing Mg 2+ concentrations stimulate the adenylate cyclase both in the absence and in the presence of serotonin. The affinity of the enzymatic system for the cation is unaffected by the amine.

Catalysis of the water gas shift reaction by polymer-immobilized rhodium complexes

Catalysts for the water gas shift prepared from RhCl 3·3H 2O immobilized on several porous aminated polymers in neutral aqueous and aqueous organic solvent media are described. Those using poly(4-vinylpyridine) as the support proved to be the most active among those surveyed, and the effects of varying different experimental parameters were investigated. The turnover frequencies for the production of H 2 displayed a nonlinear dependence on the partial pressure of CO in the 0–1.5 atm range but were independent of the polymer loading in the range 0.5 to 1.5 × 10 −4 mol RHCl 3H 2O/ g of polymer. An activation energy of 18.4 ± 0.9 kcal/mol was obtained from a linear Arrhenius plot based on H 2 production over the temperature range 80–110°C. The activity of the poly(4-vinylpyridine)-anchored catalyst was found to be insensitive to pH in the 3–5 zone but to fall off sharply outside this range. A catalysis mechanism consistent with the quantitative behavior is proposed. This species also catalyzes the Reppe-type transformations (hydrogenation and hydroformylation) of alkenes as well as the reduction of nitrobenzene to aniline.

Electrochemical measurements of the lateral diffusion of electroactive amphiphiles in supported phospholipid monolayers

Chronocoulometry was used to characterize the fluidity and lateral diffusion coefficient of supported phospholipid bilayer assemblies. The bilayers were formed on the inner surfaces of the microporous template films of aluminum oxide on gold electrodes. The lipid monolayers were formed by adsorption and fusion of phospholipid vesicles on alkylated oxide surfaces. Octadecyltrichlorosilane (OTS) was used in the initial alkylation step. The surface concentration of the lipids in monolayer assemblies was measured by a radioactive assay method. Surface densities corresponding to 48 +/- 10 A2/molecule (DPPC) and 56 +/- 11 A2/molecule (DMPC) were obtained (for exposure times > 120 min) independent of the temperature of the vesicle's fusion (below or above chain-melting transition). Octadecylviologen (C18MV2+) was used as an electroactive probe species. Its limiting lateral diffusion coefficient in DMPC monolayers was 5 x 10(-8) cm2/s, measured as C18MV2+ mole fraction extrapolated to 0 decreasing linearly from 20 to below 1 mol%. Linear Arrhenius plots for C18MV2+ diffusion in DMPC monolayers were obtained with slopes of approximately 40 kJ/mol between 18 and 45 degrees C, demonstrating homogeneity and fluidity of the lipid monolayers. Chronocoulometry was also used to obtain lateral diffusion coefficient of ubiquinone in DMPC/OTS bilayers. A value of 1.9 x 10(-8) cm2/s at 30 degrees C was obtained.

Kinetics of Human Apohemoglobin Dimer Dissociation

Heme chain exchange was employed to investigate the dimer dissociation reaction of human apohemoglobin in 0.1 M potassium phosphate buffer, pH 7.0, at 20°C. Incubation of apohemoglobin (α 0β 0) with either α h (or β h) allowed the monitoring of the formation of a semihemoglobin α hβ 0 (or α 0β h) species with time. Analysis revealed that the rate of formation of both semihemoglobins was essentially identical and coincided with the disappearance of heme chain. Time courses were exponential and followed first order kinetics yielding a dimer dissociation rate constant of 0.54 (±0.07) h −1. A study over the pH range from 6.5 to 8.0 revealed that this dissociation rate exhibited a maximum at pH 7.0 (implicating a histidyl residue). The effect of temperature (6-37°C) on this dimer dissociation rate yielded a linear Arrhenius Plot and an energy of activation of 7.2 kcal/mol. These results are consistent with αG-βG helical pairing being a major contributor to apohemoglobin dimer integrity.

Possible kinetic mechanism of human placental alkaline phosphatase in vivo as implemented in reverse micelles.

Human placental alkaline phosphatase is an integral membrane protein. It catalyzes the hydrolysis of phosphoester linkage of a broad-range substrate. We have embedded the enzyme in a reverse micellar system prepared by dissolving the surfactant sodium bis(2-ethylhexyl)sulfosuccinate (AOT) in isooctane. Linear Lineweaver-Burk and Eadie-Hofstee plots for the substrate and linear Arrhenius plot for the temperature-dependent enzyme reaction were obtained in reverse micelles suggesting that the substrate diffusion limitation was not a rate-limiting step in the system and exchange of materials between reverse micelles was very rapid. The catalytic constant (kcat) of the enzyme was decreased, and the Km for the substrate was increased in reverse micelles, both in an exponential way with the [H2O]/ AOT] ratio (omega O). The enzyme was more stable in reverse micelles than in aqueous solution at 30 degrees C but was unstable at higher temperature (65 degrees C). The activation energy of the enzyme in reverse micelles was 46.5 +/- 2.6 kJ/mol, which was about 20 kJ/mol higher than that in aqueous solution and reflected in the lower Kcat value at low omega O. The binding affinity between the substrate 4-nitrophenyl phosphate and the enzyme in reverse micelles was decreased as implemented by the higher Km and higher Ki for phosphate values. In aqueous solution, the log kcat/pH plot suggested that amino acid residues with pKa values of 9.03 +/- 0.03 and 11.37 +/- 0.10 are involved in catalysis. The former should be deprotonated and the latter should be protonated for the reaction to proceed. In reverse micelles, both the above-mentioned pKa values were detected. However, both groups have to deprotonated to give the optimum catalytic function. In aqueous solution, the enzyme's specificity was highly dependent on pH and buffer. The pKa value of the amino acid residues involved in substrate binding was found to be 8.69 +/- 0.07 in carbonate buffer, but changed to 9.80 +/- 0.06 in Tris buffer. Our results suggested that the rate-limiting step of the enzyme-catalyzed reaction may be changed from phosphate releasing in aqueous solution to another critical step in the reverse micelles.

Is stability prediction possible for protein drugs? Denaturation kinetics of beta-galactosidase in solution.

Denaturation and aggregation kinetics of Aspergillus oryzae beta-galactosidase in solution were studied in order to determine whether the stability of protein drugs can be predicted. Denaturation of beta-galactosidase, monitored by measuring enzyme activity, conformed to first-order kinetics, whereas aggregation of the denatured form, monitored by high performance size exclusion chromatography, showed a reaction order higher than 1. Denaturation of beta-galactosidase was irreversible and exhibited a biphasic kinetic pattern which could be explained by assuming that two isoenzymes denatured irreversibly at different rates. Linear Arrhenius plots were obtained for the estimated rate constants, and delta H not equal to and delta S++ to were estimated according to the Eyring equation. The estimated delta H++ was much larger than delta H++ observed in usual chemical reactions. The present study suggests that the denaturation of protein drugs can be analyzed by the Eyring equation in the same manner as chemical degradation, contradicting the general consensus that accelerated testing can not be used to predict the stability of protein formulations.