High-activity Form Research Articles

Irradiation methods for production of high specific activity radionuclides in no carrier added form

High Specific Activity Radio-Nuclides (HSARN) are a powerful tool to label a large range of chemical species at very low concentration levels. In order to obtain these radiotracers in a very high specific activity form, it is necessary to optimize the production methods, to separate and purify them from the irradiated target without the addition of inactive carrier, to carry out a series of analytical and radioanalytical tests to determine their “true” specific activity and to verify a series of purity parameters. A review of irradiation methods and nuclear parameters adopted in our laboratories for the preparation of some tens of radiotracers are presented.

Metabolic activators of spinach leaf nitrate reductase: Effects on enzymatic activity and dephosphorylation by endogenous phosphatases

Nitrate reductase (NR; EC 1.6.6.1) in spinach (Spinacia oleracea L.) leaves was inactivated in the dark and reactivated by light in vivo. When extracted from dark leaves, NR activity was lower and more strongly inhibited by Mg2+ relative to the enzyme extracted from leaves harvested in the light. When dark extracts were desalted at pH 6.5 and preincubated at 25° C prior to assay, enzyme activity (assayed either in the presence or absence of Mg2+) remained essentially constant, i.e. there was no spontaneous reactivation in vitro. However, addition of certain metabolites resulted in a time- and concentration-dependent activation of NR in vitro. Effective activators included inorganic phosphate (Pi), 5′-AMP, and certain of its derivatives such as FAD and pyridine nucleotides (both oxidized and reduced forms). All of the activators increased NR activity as assayed in the absence of Mg2+, whereas some activators (e.g. Pi, 5′-AMP and FAD) also reduced Mg2+ inhibition. The reduction of Mg2+ inhibition was also time-dependent and was almost completely prevented by a combination of okadaic acid plus KF, suggesting the involvement of dephosphorylation catalyzed by endogenous phosphatase(s). In contrast, the activation of NR (assayed minus Mg2+) was relatively insensitive to phosphatase inhibitors, indicating a different mechanism was involved. Compounds that were not effective activators of NR included sulfate, ribose-5-phosphate, adenosine 5′-monosulfate, coenzyme A, ADP and ATP. We postulate that NR can exist in at least two states that differ in enzymatic activity. The activators appear to interact with the NR molecule at a site distinct from the NADH active site, and induce a slow conformational change (hysteresis) that increases NR activity (assayed in the absence of Mg2+). Possibly as a result of the conformational change caused by certain activators, the regulatory phospho-seryl groups are more readily dephosphorylated by endogenous phosphatases, thereby reducing sensitivity to Mg2+ inhibition. Preliminary results suggest that light/dark transitions in vivo may alter the distribution of NR molecules between the low- and high-activity forms.

Preparation and Purification of High Specific Activity Radionuclides for Environmental Studies

High specific activity radionuclides are a powerful tool for labeling a large range of inorganic and metallorganic species that are present in the environment at very low concentrations. In order to study the environmental impact of trace elements (TE) on the humans, it is of fundamental importance to carry out both in vitro and in vivo experiments on cell cultures and laboratory animals, with concentrations of these TE representative of the actual long-term and low-level exposures. In order to obtain these radiotracers in a very high specific activity form, it is necessary: (a) to optimize the production method by either charged particle accelerator (normally cyclotron) or nuclear reactor irradiation, (b) to separate and purify them from the irradiated target by chemical or physicochemical methods, without the addition of inactive carriers (i.e., the so-called no carrier added form), (c) to carry out a series of analytical and radioanalytical tests, in order to verify the chemical, radiochemical, and radionuclide purities of the tracer or labeled compound under investigation. Some production, separation, purification, and quality control methods developed for the two cyclotron-produced radionuclides 52gMn and 48V will be presented and discussed.

Glutamic acid 86 is important for positioning the 80's loop and arginine 54 at the active site of Escherichia coli aspartate transcarbamoylase and for the structural stabilization of the C1-C2 interface.

Glu-86, which interacts with the side chain of Arg-54 across the C1-C2 interface of Escherichia coli aspartate transcarbamoylase, tethers the end of the flexible 80's loop, which moves into the active site during the T to R transition. In order to determine whether this interaction is important for the correct positioning of the 80's loop and Arg-54 at the active site and also for the structural stabilization of the enzyme, a mutant version was created in which Glu-86 was replaced by Gln (Glu-86-->Gln). Although the mutant holoenzyme exhibits almost normal homotropic cooperativity, both the holoenzyme and catalytic subunit exhibit substantial reductions in activity and affinity for aspartate and carbamyl phosphate. Furthermore, the mutant holoenzyme shows a marked decrease in the activation by ATP and by the bisubstrate analog N-(phosphonoacetyl)-L-aspartate, reduced inhibition by CTP, as well as reduced affinities for these ligands. Results from molecular dynamics simulations of the Glu-86-->Gln and Glu-86-->Ala enzymes suggest that the positions of the 80's loop and Arg-54 are significantly perturbed by the introduction of these mutations. Taken together, these results indicate that the interaction between Glu-86 and Arg-54 is important for the formation of the high affinity, high activity form of the enzyme by stabilizing the correct position of the 80's loop and Arg-54 at the active site. Heat inactivation experiments also demonstrated that Glu-86 plays a significant role in the structural stabilization of the C1-C2 interface, since the temperature required for loss of half of the activity of the Glu-86-->Gln catalytic subunit is reduced by 5 degrees C relative to the wild-type catalytic subunit.

Light regulation of nitrate reductase in green leaves of higher plants

Light stimulates de nova synthesis, as well as activation of higher plant nitrate reductase at the protein level. In green leaves, the light effects are mediated through photosynthetically active light by products of the Calvin cycle. A daily, light‐induced increase in nitrate reduclase (NR) mRNA. protein, and activity is observed in various higher plants. Glucose or sucrose can replace light in eliciting the increase in nitrate reductase mRNA accumulation, and sonic metabolite from nitrate assimilation exerts a negative feedback on transcription of the nitrate reductase genes. The positive feedforward induced by sugars, and the negative feedback from nitrogen compound(s) apparently result in the circadian rhythms of NR mRNA observed in various plants. Nitrate reductase shows hysteretic behaviour, converting between a low and a high activity form in response to NAD(P)H. The low activity form is inhibited by calcium/ magnesium, as opposed to the high activity form that is not inhibited. Examination of nitrate reductase activity in crude extracts made from plants before and after a light‐dark transition suggests the existence of two different forms of nitrate reductase: a‘light form’with a pH optimum of 7.8 in potassium phosphate buffer that is not inhibited by calcium or magnesium, and a “dark form’with li pH optimum of 7.5 that is strongly inhibited by calcium or magnesium. It is postulated that a transient lack of NADH as a result of decreased illumination can convert nitrate reductase into the less active form in situ.

Alcohol and aldehyde dehydrogenase polymorphisms and alcoholism

The alcohol-flush reaction occurs in Asians who inherit the mutant ALDH2*2 allele that produces an inactive aldehyde dehydrogenase enzyme. In these individuals, high blood acetaldehyde levels are believed to be the cause of the unpleasant symptoms that follow drinking. We measured the alcohol elimination rates and intensity of flushing in Chinese subjects in whom the alcohol dehydrogenase ADH2 and ALDH2 genotypes were determined. We also correlated ADH2, ADH3, and ALDH2 genotypes with drinking behavior in 100 Chinese men. We discovered that ADH2*2 and ADH3*1, alleles that encode the high activity forms of alcohol dehydrogenase, as well as the mutant ALDH2*2 allele were less frequent in alcoholics than in controls. The presence of ALDH2*2 was associated with slower alcohol metabolism and the most intense flushing. In those homozygous for ALDH2*1, the presence of two ADH2*2 alleles correlated with slightly faster alcohol metabolism and more intense flushing, although a great deal of variability in the latter was noted.

Biosynthesis of platelet-activating factor (PAF) induced by chemotactic peptide is modulated at the lyso-PAF:acetyl-CoA acetyltransferase level by calcium transient and phosphatidic acid.

The chemotactic peptide fMLP (N-formyl-methionyl-leucyl-phenylalanine) induced the production of platelet-activating factor (PAF) by human polymorphonuclear leukocytes (PMN) incubated with cytochalasin B (CB). CoA-independent transacylase showed similar activity in both resting and stimulated PMN, and PAF production only occurred when lyso-PAF:acetyl-CoA acetyltransferase had been converted into the high activity form. PAF formation was coincidental with an increase of the concentration of cytosolic Ca2+ ([Ca2+]i), and with an enhanced formation of 1-O-[3H]alkyl-2-acyl-sn-glycerol. Both fMLP-induced PAF production and the activation of lyso-PAF:acetyl-CoA acetyltransferase were diminished by propranolol. Since several molecular species of phosphatidic acid (PA) produced an inhibition of both PAF production and acetyltransferase activation on intact cells, a portion of the inhibitory effect of propranolol was related to the accumulation of PA. Furthermore, whereas CB increased both the extent and the duration of the fMLP-induced [Ca2+]i transient, propranolol was found to inhibit the CB-induced increase of the [Ca2+]i transient. These data indicate that both the attenuation of [Ca2+]i transient and the accumulation of PA may operate as termination signals for PAF production by actin on lyso-PAF:acetyl-CoA acetyltransferase.

Electrophoretic analysis of low and high activity forms of catechol-O-methyltransferase in human erythrocytes

Analysis of the catechol-O-methyltransferase (COMT) enzyme in human RBC lysates from 15 samples exhibiting inherited variations in level of activity and thermal stability was performed. Electrophoretic blotting and immune fixation was carried out following sodium dodecyl sulfate polyacrylamide gel electrophoresis or isoelectric focusing of lysate protein. These techniques did not reveal a major structural alteration of the protein that could account for the observed variation in enzyme activity or thermal stability. Future studies utilizing molecular genetic techniques should make it possible to determine the basis for inherited variations in human RBC COMT activity and thermal stability.

Low- and high-activity forms of glutamine synthetase from Rhodospirillum rubrum: sensitivity to feed-back effectors and activation of the low-activity form

Glutamine synthetase from Rhodospirillum rubrum can be isolated in two forms, with low and high activity, respectively, depending on the concentration of combined nitrogen in the medium before harvest. The two forms have been studied with respect to their dependence on Mn2+ and Mg2+ in both the transferase and the biosynthetic assay. There is no difference in pH optimum between the forms in the biosynthetic assay. In addition the pH-optima for the two cations studied are very close, 7.4 (Mg2+) and 7.2 (Mn2+). It also shows that the activity of the low-activity form is higher than that of the high-activity form in the Mn(2+)-dependent biosynthetic assay. The two forms of Rsp. rubrum glutamine synthetase have also been studied with respect to their sensitivity towards feed-back effectors. In the transferase assay both forms are inhibited to essentially the same degree by alanine, glycine, histidine, AMP, CTP and UTP, CTP being the most effective of the nucleotides and of the amino acids alanine causes the highest inhibition. In the biosynthetic assay these effectors show different degrees of inhibition on the two different forms; the high-activity form being the most sensitive. The results are discussed in relation to properties of glutamine synthetase from Escherichia coli and other phototropic bacteria in which regulation of glutamine synthetase is known to be due to adenylylation. It is also shown that the low-activity form of Rsp. rubrum glutamine synthetase can be activated in crude extracts in a reaction that is inhibited by glutamine.

Glucose utilization by tumor cells: A post-translational modification of mitochondrial hexokinase may play a regulatory role

In Northern blot analysis of a series of tumor cell lines a single hexokinase mRNA species of 4.3 Kb was detected. Detailed examination of one such line, the rat AS-30D hepatoma, revealed that two mitochondrial species of hexokinase are present with a molecular mass of 115 and 107 KDa. The smaller of the two species is 4-fold more active than the larger. Only the larger, less active species is detected in the well differentiated H-35 rat hepatoma cell line which exibits a lower glucose catabolic rate. These results suggest that a post-translational proteolytic event may play a central role in regulating the glucose utilization capacity of tumor cells by modulating the relative levels of high and low activity forms of hexokinase.

A pH-dependent activation-inactivation equilibrium in glutamate dehydrogenase of Clostridium symbiosum

1. On transferring Clostridium symbiosum glutamate dehydrogenase from pH 7 to assay mixtures at pH 8.8, reaction time courses showed a marked deceleration that was not attributable to the approach to equilibrium of the catalysed reaction. The rate became approximately constant after declining to 4-5% of the initial value. Enzyme, stored at pH 8.8 and assayed in the same mixture, gave an accelerating time course with the same final linear rate. The enzyme appears to be reversibly converted from a high-activity form at low pH to a low-activity form at high pH. 2. Re-activation at 31 degrees C upon dilution from pH 8.8 to pH 7 was followed by periodic assay of the diluted enzyme solution. At low ionic strength (5 mM-Tris/HCl), no re-activation occurred, but various salts promoted re-activation to a limiting rate, with full re-activation in 40 min. 3. Re-activation was very temperature-dependent and extremely slow at 4 degrees C, suggesting a large activation energy. 4. 2-Oxoglutarate, glutarate or succinate (10 mM) accelerated re-activation; L-glutamate and L-aspartate were much less effective. 5. The monocarboxylic amino acids alanine and norvaline appear to stabilize the inactive enzyme: 60 mM-alanine does not promote re-activation, and, as substrates at pH 8.8 for enzyme stored at pH 7, alanine and norvaline give progress curves showing rapid complete inactivation. 6. Mono- and di-nucleotides (AMP, ADP, ATP, NAD+, NADH, NADP+, CoA, acetyl-CoA) at low concentrations (10(-4)-10(-3) M) enhance re-activation at pH 7 and also retard inactivation at pH 8.8. 7. The re-activation rate is independent of enzyme concentration: ultracentrifuge experiments show no changes in molecular mass with or without substrates. 8. The activation-inactivation appears to be due to a slow pH-dependent conformational change that is sensitively responsive to the reactants and their analogues.

Nitrous oxide reductase from Pseudomonas stutzeri

The oxidation-reduction and spectroscopic properties of various forms of nitrous oxide reductase from Pseudomonas stutzeri were investigated. The high-activity form I of the enzyme (purple, 8 Cu, Mr 140,000) was reduced by a large variety of cationic, anionic and photochemically generated agents. The blue form III was the only product found in these experiments under anaerobic conditions. Reductive (dithionite) and oxidative (ferricyanide) titrations showed that the conversion of the purple form I to the blue species III was fully reversible in the absence of dioxygen. Two kinetically different phases of the reaction of form I with a stoichiometric amount of dithionite (1e- -equivalent/Cu) were detected: in the fast phase (seconds), the purple chromophore with lamba max at 540 nm disappeared almost completely, whereas in the slower phase (minutes) the blue species with lambda max around 650 nm was generated. Irrespective of the nature of the reductant the blue species did not react even at large excess of reductant. It was reoxidized by ferricyanide, hydrogen peroxide and nitric oxide. A new, catalytically inactive derivative of nitrous oxide reductase (form V, 2 Cu, Mr 140,000) was isolated from a transposon Tn5-induced mutant with defective chromophore biosynthesis. The pink color of the mutant protein faded almost completely after addition of 0.5e- -equivalent/Cu. In this case no blue species was found, similar to earlier observations for the regenerated, catalytically inactive protein. Varying with the sample and the pH, 50-80% of the total copper of form I was in an electron-paramagnetic-resonance-(EPR)-silent state as compared to 47% in the mutant protein. The broad, featureless EPR signal recorded at 9.32 GHz for the blue, reduced form III of nitrous oxide reductase represented approximately 20% of the total copper. For the blue species no resolution enhancement was achieved at 34 GHz. At this frequency both forms I and V showed similar EPR signals with apparent g-values at 2.16 and 1.99. At 9.32 GHz, form V had an EPR signal with gII at 2.18, AII = 3.55 mT (4 or 5 lines, in contrast to form I) and gI at 2.03. Above 100 K the splitting of the gII region into seven equidistant lines in the EPR signal of the high-activity form I and the hyperfine structure of the perpendicular transition disappeared. Carbon monoxide and nitric oxide, but not nitrous oxide, had marked effects on the spectroscopic properties of the purple form I. Marked effects were also obtained for the exogenous ligands nitrite, azide, cyanate and thiocyanate.(ABSTRACT TRUNCATED AT 400 WORDS)

Activation of a low specific activity form of DNA polymerase α by inositol-1,4-bisphosphate

A low activity form of DNA polymerase alpha immunoaffinity-purified from adult-derived human fibroblasts was activated by interaction with phosphatidylinositol-4-monophosphate, while a high activity form of the enzyme did not interact with phosphatidylinositol-4-monophosphate or its derivatives. Phosphatidylinositol-4-monophosphate was apparently hydrolyzed in the presence of a highly purified low activity form of DNA polymerase alpha, effecting the release of diacylglycerol and the retention of inositol-1,4-bisphosphate by the enzyme complex. The resulting inositol-1,4-bisphosphate/protein complex exhibited increased affinity of binding to DNA template/primer and increased deoxynucleotidyltransferase activity. These data indicate that inositol-1,4-bisphosphate may function as an effector molecule in the activation of a low activity form of human DNA polymerase alpha and suggest that it may function as a second messenger during the initiation of mitosis in stimulated cells.

Origin of viscosity effects in carbonic anhydrase catalysis. Kinetic studies with bulky buffers at limiting concentrations.

In our earlier paper we showed that the rates of CO2 hydration and HCO3- dehydration catalyzed by the high-activity form of mammalian erythrocyte carbonic anhydrase (CA II) were dependent on solution viscosity increase and that the effect was linked to some kind of proton-transfer-related event [Pocker, Y., & Janjić, N. (1987) Biochemistry 26, 2597-2606]. In order to further elucidate the source of the observed viscosity effect, the dependence of kcat and Km for CA II catalyzed HCO3- dehydration at pH 5.90 on sucrose-induced viscosity increase was investigated at several concentrations of 2-(N-morpholino)ethanesulfonic acid (MES) buffer, including the very low buffer concentration region (less than 10 mM) where the proton transfer between the shuttle group on the enzyme and buffer becomes rate limiting. In all examined cases, kcat steadily decreased with added sucrose while Km remained independent of the viscosity increase. The extent to which this reaction was dependent on viscosity was found to be constant, within experimental error, over the entire range of MES buffer concentrations studied (1-20 mM). Furthermore, the viscosity effect was qualitatively and quantitatively the same when an exceptionally large buffer (i.e., bovine serum albumin) was used instead of the more commonly used biological buffer (i.e., MES).(ABSTRACT TRUNCATED AT 250 WORDS)

The Lipoxygenases in Developing Soybean Seeds, Their Characterization and Synthesis in Vitro

A number of lipoxygenase isoenzymes were identified in developing soybean (Glycine max L. Merrill cv Provar) seeds and two have been partially characterized. In a study of lipoxygenase level in developing soybean seeds, the enzyme content increased markedly during development. Comparisons of the lipoxygenases from mature soybean seeds and immature seeds by isoelectric focusing, chromatofocusing, sodium dodecyl sulfate polyacrylamide gel electrophoresis and peptide mapping identified two categories of isoenzyme. The isoenzymes from immature seeds were found by electron paramagnetic resonance spectroscopy to be isolated at least in part as the high spin iron(III) or active form of the enzyme in contrast to lipoxygenases from mature seeds which were isolated as electron paramagnetic resonance silent, high spin iron(II) species. The discovery of increased levels of lipoxygenases during seed development and their isolation in an active form suggests that the enzyme may play a physiological role during the maturation process. The incorporation of iron-59 from the nutrient medium into lipoxygenase during culture of immature seeds was indicative of de novo synthesis of the enzyme. The efficiency of the iron uptake was high, as indicated by the level of radioactivity found in the enzyme (one gram atom of iron per mole of lipoxygenase).

Interconversion of high and low adenosinetriphosphatase activity forms of Escherichia coli F1 by the detergent lauryldimethylamine oxide.

The amphipathic detergent lauryldimethylamine oxide (LDAO) stimulated ATP hydrolytic activity of Escherichia coli membranes and isolated ECF1 and ECF1-F0 ATPase complexes in a concentration-dependent manner. The enzyme was maximally activated 3-fold in membranes and 5-6-fold for isolated ECF1 or the ECF1-F0 complex. The maximal specific activity of activated ECF1 was 140-160 mumol of ATP hydrolyzed min-1 mg-1. The activation by LDAO was reversible. LDAO specifically released subunit delta from ECF1, generating a four subunit enzyme (alpha, beta, gamma, and epsilon subunits). The removal of subunit delta was not responsible for the stimulation of ATPase activity as evidenced by the full activation of the four subunit enzyme by LDAO. Treatment of ECF1 with 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide generated a beta-epsilon cross-link in high yield [Lötscher, H.R., DeJong, C., & Capaldi, R. A. (1984) Biochemistry (accompanying paper in this issue)]. The formation of this cross-link was greatly reduced in the presence of LDAO, indicating that the detergent perturbated the interaction between epsilon and beta subunits although epsilon was not removed from the ECF1 complex. The results suggest that the interconversion of ECF1 from a low to a high ATPase activity form by LDAO is in major part due to a release of the inhibitory action of subunit epsilon on subunit beta.

Elevated leukocyte cyclic AMP—phosphodiesterase in atopic disease: a possible mechanism for cyclic AMP—agonist hyporesponsiveness

Previous studies have documented that leukocytes from atopic individuals show reduced cyclic AMP (cAMP) responsiveness to isoproterenol, histamine, and prostaglandin E 1. We questioned whether this blunted response was due to rapid enzymic breakdown of cAMP. We measured cAMP-phosphodiesterase activity in mononuclear leukocytes of patients with atopic dermatitis and allergic rhinitis and noted consistent, significantly elevated levels. Kinetic studies showed differences in two of three leukocyte enzyme forms between normals and patients. Isolated lymphocytes from atopic dermatitis patients lacked a low Km phosphodiesterase form and showed an increase in the high Km, high-activity form seen in normal monocytes. The increased phosphodiesterase activity provides an explanation for the blunted cAMP responsiveness in atopic leukocytes and may reflect a basic biochemical characteristic relevant to abnormal immunocellular regulation in atopic disease.

Immunohistochemical demonstration of extracellular carbonic anhydrase in epiphyseal growth cartilage.

The distribution of carbonic anhydrase isoenzymes C and B in the rate epiphyseal growth cartilage was demonstrated by an immunohistochemical method. The isoenzymes were found in different locations. Isoenzyme C was in the extracellular matrix of the hypertrophic and calcifying cartilage, and no reaction was observed in the chondrocytes. In contrast, the antiserum against isoenzyme B revealed only a weak cellular staining. This supports the hypothesis that carbonic anhydrase isoenzyme C, which is the high-activity form, changes the pH in the extracellular fluid of calcifying cartilage, favoring the deposition of calcium phosphate.

Osmotically induced changes in the ornithine decarboxylase activity of Dictyostelium discoideum.

Myxamoebae of Dictyostelium discoideum from exponentially growing cultures showed altered ornithine decarboxylase activity upon external osmotic perturbation. On transfer to hypotonic NaCl solutions (20 mosmol/kg), cells showed high enzyme activity which was relatively independent of the concentration of the coenzyme pyridoxal phosphate (assay concentrations, 5 and 200 microM). In hypertonic solution (400 mosmol/kg) cells had a reduced level of ornithine decarboxylase activity which was dependent on the coenzyme concentration. The changes in activity were freely reversible in further external osmotic manipulation. The response to osmotic change occurred rapidly, within a few minutes. The changes still occurred at 7 degrees C but 2 mM sodium azide prevented the formation of the high activity form, although this effect was reversed when azide was removed. Cycloheximide had no effect on the osmotically induced changes. Addition of putrescine caused ornithine decarboxylase eventually to the converted to the low-activity form regardless of the osmolality of the solution. The characteristic cofactor concentration dependence of the high- and low-activity form were retained on storage of the cell extracts. No evidence was found for diffusible effectors which stabilized one or the other form of the activity. The enzymes responsible for the two forms were of the same molecular size as judged by gel filtration, and the activities had similar thermostabilities. The results are interpreted in terms of an osmotically induced interconversion of two forms of a single ornithine decarboxylase.

Photolytic studies on the carbon monoxide complex of sulphaemoglobin.

The CO complex of sulphaemoglobin was found to be photolytically dissociable at pH 6.0 and pH 8.5. The recombination kinetics of the CO complex after flash photolysis show two phases, which differ in rate by approximately 20-fold. At both pH 6.0 and pH 8.5 the two phases show an identical linear dependence on CO concentration, with associated second-order rate constants of 6.0 x 10(3) M-1 . s-1 and 1.2 x 10(5) M-1 . s-1 respectively. The percentage contribution of each of the two phases to the total absorbance change is independent of the CO concentration employed, but is a function of the protein concentration used. The percentage of the faster phase increases with protein dilution. These results taken together with computer simulation suggest the existence of appreciable amounts of monomeric species in solution at both pH 6.0 and pH 8.5 at low concentration of liganded protein. Estimates of the dissociation constants for the dimer in equilibrium monomer equilibrium yield values of 8 microM and 3.5 microM at pH 6.0 and pH 8.5 respectively. Stopped-flow delayed-flash-photolysis studies show that the liganded dimer decays to the liganded monomer with a half-life of approx. 0.5 s. Constant-illumination experiments allow an estimation of the half-life for the combination of deoxy monomers to deoxy dimers of 22 s. A model explaining the range of CO recombination kinetics seen for sulphaemoglobin does not require the existence of a photolytically produced high-activity form, even at high pH, in contrast with the case for normal haemoglobin.