Search

Glucose-6-phosphate dehydrogenase activity has been localized ultrastructurally in fixed tissues. Activity was found in particular in association with ribosomes of granular endoplasmatic reticulum. Biochemical studies indicated that glucose-6-phosphate dehydrogenase activity is also present in the cytoplasm and in peroxisomes. Fixation may be held responsible for selective inactivation of part of glucose-6-phosphate dehydrogenase activity. In the present study, we applied the ferricyanide method for the demonstration of glucose-6-phosphate dehydrogenase activity in unfixed cryostat sections of rat liver in combination with the semipermeable membrane technique and in isolated rat liver parenchymal cells. Isolated liver parenchymal cells were permeabilized with 0.025% glutaraldehyde after NADP+ protection of the active site of glucose-6-phosphate dehydrogenase. This treatment resulted in only slight inactivation of glucose-6-phosphate dehydrogenase activity. The composition of the incubation medium was optimized on the basis of rapid light microscopical analysis of the formation of reddish-brown final reaction product in sections. With the optimized method, electron dense reaction product was observed in cryostat sections on granular endoplasmic reticulum, in mitochondria and at the cell border. However, the ultrastructural morphology was rather poor. In contrast, the morphology of incubated isolated cells was preserved much better. Electron dense precipitate was found on ribosomes of the granular endoplasmic reticulum, in peroxisomes and the cytoplasm, particularly at the periphery of cells. In conclusion, our ultrastructural study clearly demonstrates that it is essential to use mildly-fixed cells to allow detection of glucose-6-phosphate dehydrogenase activity in all cellular compartments where activity is present.

Treatment with peroxisome proliferators induces increased numbers and alterations in the shape of peroxisomes in liver. It ultimately leads to hepatocellular carcinomas induced by the persistent production of high amounts of H2O2 as a result of a dramatical increase in acyl-CoA oxidase activity. The effects of peroxisome proliferators on other peroxisomal oxidase activities are less well documented. In the present study, the distribution patterns of the activity of SdD-amino acid oxidase, SlD-alpha-hydroxy acid oxidase, polyamine oxidase, urate oxidase and catalase activities were investigated in unfixed cryostat sections of liver, kidney and duodenum of rats treated with either clofibrate or bis(2-ethylhexyl)phthalate. The activities of xanthine oxidoreductase, which produces urate, a potent anti-oxidant, and xanthine oxidase, which produces oxygen radicals, were studied as well. The liver was the only organ that was affected by treatment. The number of peroxisomes increased considerably. SdD-Amino acid oxidase and polyamine oxidase activities were completely abolished by the treatment, whereas SlD-alpha-hydroxy acid oxidase activity decreased and urate oxidase activity increased periportally and decreased pericentrally. Total catalase activity increased because of the larger numbers of peroxisomes, but it decreased per individual peroxisome. Xanthine oxidoreductase activity decreased, whereas the percentage of xanthine oxidase remained constant. We conclude that oxidases in rat liver are affected differentially, indicating that the expression of activity of each oxidase is regulated individually. © 1998 Chapman & Hall

Summary 3,39-Diaminobenzidine, in the presence of manganese and cobalt ions, was applied for the detection of superoxide anions in unfixed cryostat sections of rat oesophagus, trachea, skin and intact mesenterium. In all connnective tissues, a blue final reaction product was found in a granular form in mast cells. The amount of final reaction product formed after incubation with diaminobenzidine and cobalt ions was increased by the addition of manganese ions. Electron microscopical analysis revealed that the electron-dense final reaction product was exclusively present in the granules of mast cells and on elastin fibres. It was found that the constitutive spontaneous formation of final reaction product in mast cells was enzymatic and dependent on the presence of oxygen in the medium. Of all the enzyme inhibitors and free radical scavengers tested, only azide strongly reduced the amount of final reaction product. It was concluded that the reaction was partly caused by peroxidase activity, but that superoxide anions are also constitutively and spontaneously produced in mast cell granules. The exact enzymatic source could not be established. Whether this property of mast cell granules plays an antimicrobial role in connective tissues can only be speculated.

In the present study, the endogenous formation of reactive oxygen species was localized in rat liver and small intestine. The 3,3′-diaminobenzidine (DAB)-Mn2+ technique in which cobalt ions were included in the incubation medium was applied to unfixed cryostat sections of intact tissues. Addition of manganese ions to the DAB-Co2+- containing medium greatly increased the amounts of final reaction product formed compared with incubations with only DAB and cobalt ions. In liver, a blue final reaction product was deposited, particularly in hepatocytes surrounding portal tracts. In the small intestine, the DAB--cobalt complex was mainly found at the basal side of enterocytes. Goblet cells remained unstained. Electron microscopical images revealed that an electron-dense reaction product was exclusively present at both inner and outer membranes and at the intermembrane space in mitochondria of liver parenchymal cells and duodenal enterocytes. It was shown that the spontaneous formation of final reaction product was enzymatic and dependent on the presence of oxygen in the medium. Sulphide decreased the reaction, which may indicate that cytochrome c oxidase was partially involved. Benzoquinone and histidine, which are scavengers of superoxide anions and singlet oxygen respectively, reduced the amount of final reaction product considerably. Furthermore, the formation of final reaction product was sensitive to specific inhibitors of NADH:coenzyme Q reductase and aldehydeoxidase, indicating that these enzymes were at least partly responsible for the generation of superoxide anions and singlet oxygen and for the formation of the DAB--cobalt complex.

In the present study a technique was developed to demonstrate 5′-nucleotidase activity in unfixed cryostat sections of rat liver at the light- and electron-microscope level using a semipermeable membrane. In order to retain the ultrastructure of the unfixed material as much as possible, incubations were also performed at 4°C rather than at 37°C. The optimized incubation medium contained 300 mm Tris-maleate buffer, pH 7.2, 5 mm adenosine monophosphate as substrate, 30 mm cerium chloride as capturing agent for liberated phosphate, 10 mm magnesium chloride as activator and 1.5% agar. At the light-microscope level, similar localizations of 5′-nucleotidase activity were obtained when incubations were performed at 37°C and 4°C. Enzyme activity was present mainly at bile canalicular membranes and at sinusoidal membranes of hepatocytes; total activity was higher in pericentral than in periportal areas. Cytophotometric analyses revealed that specific formation of final reaction product (FRP) (test minus control reaction) at 37°C followed a hyperbolic curve with time. A linear relationship was found between specific amounts of FRP and section thickness up to 8μm. 5′-Nucleotidase activity was about three-fold higher after incubation for 30 min at 37°C than at 4°C. At the electron-microscope level, it was demonstrated that the ultrastructure of rat liver was rather well-preserved after incubating unfixed cryostat sections attached to a semipermeable membrane and electron-dense FRP was found at bile canalicular and sinusoidal plasma membrane of hepatocytes. The most distinct changes in ultrastructure after incubation at 37°C, in comparison with that at 4°C, were the appearance of multi-lamellar structures at bile canaliculi at 37°C. We conclude that the present method is valid for the demonstration of 5′-nucleotidase activity in unfixed cryostat sections of rat liver at both the light- and electron-microscope levels and that hypothermic incubations improve ultrastructural morphology substantially.

Precise localization of xanthine oxidase activity might elucidate physiological functions of the enzyme, which have not been established so far. Because xanthine oxidase is sensitive to chemical (aldehyde) fixation, we have localized its activity in unfixed cryostat sections of rat duodenum, oesophagus and tongue mounted on a semipermeable membrane. Previous studies had shown that this procedure enables the exact localization of activities of peroxisomal oxidases with maintenance of acceptable ultrastructure. Moreover, leakage and/or diffusion of enzyme molecules was prevented with this method. The incubation medium to detect xanthine oxidase activity contained hypoxanthine as substrate and cerium ions as capturing agent for hydrogen peroxide. After incubation, reaction product in the sections was either visualized for light microscopy or sections were fixed immediately and processed for electron microscopy. At the ultrastructural level, crystalline reaction product specifically formed by xanthine oxidase activity was found to be present in the cytoplasmic matrix of enterocytes and goblet cells and in mucus of duodenum. Moderate activity was found in the cytoplasm of apical cell layers of epithelia of oesophagus and tongue, with highest activity in the cornified layer. Moreover, large amounts of reaction product were found to surround bacteria present between cell remnants of the cornified layer of the oesophagus. Many bacteria surrounded by the enzyme showed signs of destruction and/or cell death. The intracellular localization of xanthine oxidase activity in the cytoplasm of epithelial cells as well as the extracellular localization suggest that the enzyme plays a role in the lumen of the digestive tract, for instance in the defence against microorganisms.

Glycogen phosphorylase activity has been demonstrated at the ultrastructural level in liver and heart tissue of fasted rats. Unfixed cryostat sections were incubated by mounting them on a semipermeable membrane stretched over a gelled incubation medium. The medium contained a high concentration of glucose 1-phosphate which enables indirect detection of glycogen phosphorylase activity on the basis of the synthesis of glycogen. Tissue fixation, dehydration and embedding for electron microscopical study were performed after the incubation had been completed. The ultrastructure of both liver and heart tissue was rather well preserved. Glycogen granules resulting from glycogen phosphorylase activity were found in the cytoplasmic matrix of both hepatocytes and cardiomyocytes; no relationship with membranous structures could be detected. It is concluded that the semipermeable membrane method is well suited for localizing cytosolic enzyme activities at the ultrastructural level without prior tissue fixation; this opens further perspectives for correlations between histochemical and biochemical data.

d-Amino acid oxidase activity was demonstrated in peroxisomes of rat liver using unfixed cryostat sections and a histochemical technique using cerium ions as capture reagent for hydrogen peroxide and diaminobenzidine, cobalt ions and exogenous hydrogen peroxide to visualize the final reaction product for light microscopical analysis. Cytophotometric analysis of liver sections revealed similar zero-order reaction velocities of d-amino acid oxidase with activity twice as high in periportal areas as in pericentral areas of liver lobuli when using either d-proline or d,l-thiazolidine-2-carboxylic acid as substrates. On the other hand, a 4–5 times higher KM value was found for d-proline than for d,l-thiazolidine-2-carboxylic acid. The KM values in periportal and pericentral areas were similar for each substrate. These findings support the suggestion that the physiological substrate for d-amino acid oxidase may be d,l-thiazolidine-2-carboxylic acid, the adduct of cysteamine and glyoxylic acid. d-Amino acid oxidase may play a role in vivo in the production of oxalate which may participate in metabolic control processes as an intracellular messenger molecule.

NADH oxidase activity has been detected at the ultrastructural level using cerium ions to trap H2O2 generated by the enzyme (via intermediate reactive oxygen species). In an attempt to localize NADH oxidase activity at the light microscope level using the cerium-diaminobenzidine (DAB)-nickel-H2O2, the cerium-DAB-cobalt-H2O2 or the cerium-alkaline lead procedures, the distribution patterns of the revealed enzyme were found to be identical to those for non-specific alkaline phosphatase and especially 5'-nucleotidase activity. With the cerium-DAB-cobalt-H2O2 visualization procedure, the distribution pattern of the final reaction product was similar to that obtained with the other two techniques but much less final reaction product was formed. Incubations for NADH oxidase activity performed in the presence of exogenous catalase or in the absence of catalase or peroxidase inhibitors did not affect the staining intensity, whereas inhibitors of 5'-nucleotidase (EDTA) and non-specific alkaline phosphatase (levamisole) always did. Therefore, phosphatases contribute to the formation of the final reaction product. Since NADH initially cannot be hydrolysed by either of these two phosphatases, then presumably nucleotide pyrophosphatase (E.C.3.6.1.9) cleaves NADH into 5'-AMP and nicotinamide mononucleotide in a first step. Both nucleotides can be hydrolysed further by the two monophosphatases. These then generate cerium phosphate which is detected by the DAB-nickel-H2O2, DAB-cobalt-H2O2 or lead visualization methods.

A quantitative histochemical procedure was developed for the demonstration of purine nucleoside phosphorylase in rat liver using unfixed cryostat sections and the auxiliary enzyme xanthine oxidase. The optimum incubation medium contained 18% (w/v) poly(vinyl alcohol), 100 mM phosphate buffer, pH 8.0, 0.5 mm inosine, 0.47 mm methoxyphenazine methosulphate and 1 mm Tetranitro BT. An enzyme film consisting of xanthine oxidase was brought onto the object slides before the section was allowed to adhere. The specificity of the reaction was proven by the low amount of final reaction product generated when incubating in the absence of inosine. Moreover, 1 mmp-chloromercuribenzoic acid, a non-specific inhibitor of purine nucleoside phosphorylase, inhibited the specific reaction by 90%. The specific reaction defined as the test reaction, in the presence of substrate, minus the control reaction, in the absence of substrate was linear with incubation time at least up to 30 min as measured cytophotometrically. A high activity was observed in endothelial cells and Kupffer cells of rat liver and a lower activity in liver parenchymal cells. Pericentral hepatocytes showed an activity higher than that of periportal hepatocytes. In human liver, purine nucleoside phosphorylase activity was also high in endothelial cells and Kupffer cells, but the activity in liver parenchymal cells was only slightly lower than it was in non-parenchymal cells. The localization of the enzyme is in agreement with earlier ultrastructural findings using fixed liver tissue and the lead salt procedure.

The 'nothing dehyrogenase' reaction is defined as the reduction of tetrazolium salts in media lacking specific substrates for dehydrogenases. In this investigation, the kinetics of the 'nothing dehydrogenase' reaction were studied in cryostat sections of rat heart and liver with the use of various polyvinyl alcohol-containing incubation media. Formazan production was measured at 585 nm with a cytophotometer. The 'nothing dehydrogenase' reaction was substantially lower in the heart than in the liver which was due to low levels of endogenous lactate and the absence of proteins containing thiol groups, such as albumin, in the heart. In vitro ischaemia resulted in a reduced 'nothing dehydrogenase' reaction due to loss of NAD+, possibly as a consequence of its breakdown by glycohydrolase activity. One hour reperfusion following one hour ischaemia caused a decreased 'nothing dehydrogenase' reaction in certain areas of the liver. This reduction was a result of leakage of lactate dehydrogenase and thiol-containing molecules. It appeared at the ultrastructural level that parenchymal and endothelial cells were heavily damaged in the areas containing a low 'nothing dehydrogenase' activity. In conclusion, early ischaemic damage in liver can be detected with the 'nothing dehydrogenase' reaction.

The tetrazolium method for the histochemical detection of monoamine oxidase (MAO) activity in rat liver cryostat sections has been tested for its specificity and its possible use in quantification. The tetrazolium salt tetranitro blue tetrazolium is recommended for the localization of MAO activity, rather than nitro blue tetrazolium or BPST [2-(2-benzothiazolyl)-3-(4-phthalhydrazidyl)-5-styryl-tetrazolium] . Hardly any formazan was produced in the absence of the substrate tryptamine and Marsilid, a specific inhibitor of MAO activity, prevented formazan production almost completely. A linear relationship between the integrated absorbance measured with a microdensitometer and either the incubation period or section thickness was obtained. We conclude that the method described in this paper can be used for the quantitative analysis of MAO activity in tissue sections of rat liver. MAO activity was found to be 20-25% higher in the periportal zone of rat liver than in the perivenous zone.

Acid phosphatase activity in isolated rat liver parenchymal cells has been investigated with quantitative histochemical means during short-term starvation, which leads to a considerable loss in protein mass in the parenchyma. Animals trained to a meal-feeding regime in which food was available during 1 h only per 24 h (using an automatic food dispensing machine), were sacrificed 6, 12, 18, 24 and 36 h after food had been withheld at the time point (23.00 h) of meal feeding. Acid phosphatase activity was analysed cytophotometrically in isolated hepatocytes incorporated into polyacrylamide gels before the enzyme reaction technique with the post-azo coupling was carried out. No indication could be found for any significant changes in the amount of acid phosphatase activity per individual hepatocyte during the entire period of fasting, as compared with two time points (11.00 and 23.00 h) before the theoretical onset of fasting. It is concluded that the considerable enhancement of protein degradation in the lysosomal apparatus during fasting is not reflected by changes in the cellular acid phosphatase activity.

5'-Nucleotidase (EC 3.1.3.5) activity was demonstrated in cryostat sections of rat liver using the Wachstein-Meisel medium and polyvinyl alcohol as tissue stabilizer. Optimum activity was obtained using an incubation medium containing 5 mM AMP, 10 mM magnesium chloride, 7.2 mM lead nitrate, 0.1 M Tris-maleate buffer, pH 7.2, and 17% (w/v) polyvinyl alcohol (Sigma, type III). The activity was localized at the bile canalicular and sinusoidal side of the plasma membranes of liver parenchymal cells as well as in the plasma membranes of endothelial cells of central veins and in fibroblasts surrounding portal tracts. The reaction was specific for 5'-nucleotidase because it was inhibited by ADP. Alkaline phosphatase did not interfere in the reaction. Cytophotometric analysis revealed a linear relationship between the formation of the final reaction product and incubation times up to 20 min and section thicknesses up to 8 micron. The activity in pericentral zones was 1.35 times the activity in periportal zones. The Michaelis constant for AMP was 1.4 mM in pericentral zones and 0.8 mM in periportal zones, suggesting that the bile canalicular and sinusoidal enzymes differ in their kinetic characteristics.

Glycogen phosphorylase (EC 2.4.1.1) has been demonstrated in sections of liver from rats starved for 24 h. The method is based on the measurement of the amount of glycogen formed after incubation in a gelled medium containing glucose 1-phosphate as substrate, using the semipermeable membrane technique. Glycogen was demonstrated with the periodic acid-Schiff (PAS) reaction. Phosphorylase activity appeared to be highest in periportal areas. The optimum substrate concentration for revealing activity of the enzyme was 60–120mm. After incubation in the absence of substrate, the staining intensity, as measured cytophotometrically as the mean integrated absorbance at 560 nm, was similar to that of an unincubated section.p-Chloromercuribenzoate, a non-specific inhibitor of glycogen phosphorylase activity, reduced the formation of final reaction product attributable to phosphorylase activity completely. The Michaelis constants (K m ) of the enzyme in periportal and pericentral areas differed. This was probably due to the presence of thea form only in periportal areas and of thea andb forms in pericentral areas. The mean integrated absorbances in both the periportal and pericentral areas increased linearly with incubation time (4–16 min). A linear relationship was also found with section thickness (4–10 µm). The total activity of glycogen phosphorylase in the periportal areas was double the pericentral activity. It is concluded that the semipermeable membrane technique, combined with the PAS reaction for glycogen, can be used as a valid method for the demonstration and quantification of glycogen phosphorylase activity in livers from starved rats.

Creatine kinase (EC 2.7.3.2) activity was demonstrated in rat myocardium using a polyvinyl alcohol-containing incubation medium and auxiliary enzymes. The activity was quantified by microdensitometry using both endpoint measurements and kinetic measurements. Control reactions were performed in the absence of creatine phosphate and ADP. The linear regression lines of the absorbances of reduced Nitro BT at the isobestic wavelength (585 nm) on incubation time were highly significant for both endpoint and kinetic measurements. The activity obtained from endpoint measurements was about 40% lower. This was caused by loss of the formazan reaction product from the tissue sections when the incubation medium was removed at the end of the reaction. The relationship between creatine kinase activity (test minus control reaction) and section thickness was not linear for either myocardium or skeletal muscle; control reactions, however, showed linear relationships with section thickness for both tissues. Limited penetration of auxiliary enzymes into the sections may be responsible for this disporportionality. Therefore, care should be taken in the interpretation of quantitative data obtained with different tissues. In conclusion, multi-step enzyme reactions can be used for quantitative histochemical purposes provided it is taken into account that the reactivity is not proportional to section thickness.