Purification of phosphate-dependent glutaminase from isolated mitochondria of Ehrlich ascites-tumour cells.

Phosphate-dependent glutaminase was purified to homogeneity from isolated mitochondria of Ehrlich ascites-tumour cells. The enzyme had an Mr of 135,000 as judged by chromatography on Sephacryl S-300. SDS/polyacrylamide-gel electrophoresis displayed two protein bands, with Mr values of 64,000 and 56,000. Two major immunoreactive peptides of Mr values of 65,000 and 57,000 were found by immunoblot analysis using anti-(rat kidney glutaminase) antibodies. The concentration-dependences for both glutamine and phosphate were sigmoidal, with S0.5 values of 7.6 mM and 48 mM, and Hill coefficients of 1.5 and 1.6, respectively. The glutaminase pH optimum was 9. The activation energy of the enzymic reaction was 58 kJ/mol. The enzyme showed a high specificity towards glutamine. A possible explanation for the different kinetic behaviour found for purified enzyme and for isolated mitochondria [Kovacević (1974) Cancer Res. 34, 3403-3407] should be that a conformational change occurs when the enzyme is extracted from the mitochondrial inner membrane.     

ADP-ribosylation in isolated nuclei of Physarum polycephalum.

ADP-ribosylation of histones and non-histone nuclear proteins was studied in isolated nuclei during the naturally synchronous cell cycle of Physarum polycephalum. Aside from ADP-ribosyltransferase (ADPRT) itself, histones and high mobility group-like proteins are the main acceptors for ADP-ribose. The majority of these ADP-ribose residues is NH2OH-labile. ADP-ribosylation of the nuclear proteins is periodic during the cell cycle with maximum incorporation in early to mid G2-phase. In activity gels two enzyme forms with Mr of 115,000 and 75,000 can be identified. Both enzyme forms are present at a constant ratio of 3:1 during the cell cycle. The higher molecular mass form cannot be converted in vitro to the low molecular mass form, excluding an artificial degradation during isolation of nuclei. The ADPRT forms were purified and separated by h.p.l.c. The low molecular mass form is inhibited by different ADPRT inhibitors to a stronger extent and is the main acceptor for auto-ADP-ribosylation. The high molecular mass form is only moderately auto-ADP-ribosylated.     

25-Hydroxylation of vitamin D3 by a cytochrome P-450 from rabbit liver mitochondria.

A cytochrome P-450 catalysing 25-hydroxylation of vitamin D3 was purified from liver mitochondria of untreated rabbits. The enzyme fraction contained 9 nmol of cytochrome P-450/mg of protein and showed only one protein band with an apparent Mr of 52,000 upon SDS/polyacrylamide-gel electrophoresis. The preparation showed a single protein spot with an apparent isoelectric point of 7.8 and an Mr of approx. 52,000 upon two-dimensional isoelectric-focusing-polyacrylamide-gel electrophoresis. The purified cytochrome P-450 catalysed 25-hydroxylation of vitamin D3 up to 5000 times more efficiently than did the mitochondria. The cytochrome P-450 required both ferredoxin and ferredoxin reductase for catalytic activity. Microsomal NADPH-cytochrome P-450 reductase could not replace ferredoxin and ferredoxin reductase. The cytochrome P-450 catalysed, in addition to 25-hydroxylation of vitamin D3, the 25-hydroxylation of 1 alpha-hydroxyvitamin D3 and the 26-hydroxylation of 5 beta-cholestane-3 alpha, 7 alpha, 12 alpha-triol. The enzyme did not catalyse side-chain cleavage of cholesterol, 11 beta-hydroxylation of deoxycorticosterone, 1 alpha-hydroxylation of 25-hydroxyvitamin D3, hydroxylations of lauric acid and testosterone or demethylation of benzphetamine. The results raise the possibility that the 25-hydroxylation of vitamin D3 and the 26-hydroxylation of C27 steroids are catalysed by the same species of cytochrome P-450 in liver mitochondria. The possible role of the liver mitochondrial cytochrome P-450 in the metabolism of vitamin D3 is discussed.     

Selective channelling of arachidonic and linoleic acids into glycerolipids of rat hepatocytes in primary culture.

Rat hepatocytes in primary culture were incubated with a mixture of linoleic and arachidonic acid at various total fatty acid/serum albumin molar ratios. Mixed fatty acids were taken up at the same rate and distributed with the same pattern as fatty acids added separately. The rates of total uptake, incorporation into hepatocyte and secreted triacylglycerols and beta-oxidation were linearly related to the fatty acid/albumin ratios, whereas the rate of incorporation into phospholipids was saturable. Neither the uptake rate nor the distribution of both fatty acids considered together varied with the arachidonic acid/linoleic acid molar ratio. Changes in this ratio and in the uptake rate led to significant variations in the respective fate of the fatty acids. The preferential channelling of arachidonic acid versus linoleic acid into beta-oxidation and phosphatidylinositol was greatest at a low uptake rate and then decreased as the uptake rose. Conversely, the preferential channelling of arachidonic acid versus linoleic acid into phosphatidylcholine, but not phosphatidylethanolamine, increased with the uptake rate. Moreover, both arachidonic acid and linoleic acid were preferentially incorporated into the 1-palmitoyl molecular species of phosphatidylcholine and phosphatidylethanolamine at a low uptake rate, and of phosphatidylcholine at a high uptake rate. This could be related to the synthesis of biliary phosphatidylcholine, of which 1-palmitoyl-2-linoleoyl and 1-palmitoyl-2-arachidonoyl are the main molecular species. Linoleic and arachidonic acid were selectively distributed into distinct metabolic pools of triacylglycerol, the intrahepatocyte pool which preferentially incorporated linoleic acid at a low uptake rate and the secreted pool in which the relative enrichment of arachidonic acid increased with the uptake rate. This strengthens the central role of hepatic secretion in the supply of arachidonic acid to peripheral tissues.     

Activation and inhibition of microsomal glutathione transferase from mouse liver.

Mouse liver microsomal glutathione transferase was purified in an N-ethylmaleimide-activated as well as an unactivated form. The enzyme had a molecular mass of 17 kDa and a pI of 8.8. It showed cross-reactivity with antibodies raised against rat liver microsomal glutathione transferase, but not with any of the available antisera raised against cytosolic glutathione transferases. The fully N-ethylmaleimide-activated enzyme could be further activated 1.5-fold by inclusion of 1 microM-bromosulphophthalein in the assay system. The latter effect was reversible, which was not the case for the N-ethylmaleimide activation. At 20 microM-bromosulphophthalein the activated microsomal glutathione transferase was strongly inhibited, while the unactivated form was activated 2.5-fold. Inhibitors of the microsomal glutathione transferase from mouse liver showed either about the same I50 values for the activated and the unactivated form of the enzyme, or significantly lower I50 values for the activated form compared with the unactivated form. The low I50 values and the steep slope of the activity-versus-inhibitor-concentration curves for the latter group of inhibitors tested on the activated enzyme indicate a co-operative effect involving conversion of activated enzyme into the unactivated form, as well as conventional inhibition of the enzyme.     

Cholesterol transfer between lipid vesicles. Effect of phospholipids and gangliosides.

The effect of lipid composition on the rate of cholesterol movement between cellular membranes is investigated using lipid vesicles. The separation of donor and acceptor vesicles required for rate measurement is achieved by differential centrifugation so that the lipid effect can be quantified in the absence of a charged lipid generally used for ion-exchange-based separation. The rate of cholesterol transfer from small unilamellar vesicles (SUVs) containing 50 mol% cholesterol to a common large unilamellar vesicle (LUV) acceptor containing 20 mol% cholesterol decreases with increasing mol% of sphingomyelin in the SUVs, while phosphatidylethanolamine and phosphatidylserine have no appreciable effect at physiologically relevant levels. There is a large decrease in rate when phosphatidylethanolamine constitutes 50 mol% of donor phospholipids. Interestingly, gangliosides which have the same hydrocarbon moiety as sphingomyelin exert an opposite effect. The effect of spingomyelin seems to be mediated by its ability to decrease the fluidity of the lipid matrix, while that of gangliosides may arise from a weakening of phosphatidylcholine-cholesterol interactions or from a more favourable (less polar) microenvironment for the desorption of cholesterol provided by the head-group interactions involving sugar residues. If the effect of asymmetric transbilayer distribution of lipids is taken into consideration, the observed composition-dependent rate changes could partly account for the large difference in the rates of cholesterol desorption from the inner and outer layers of plasma membrane. Such rate differences may be responsible for an unequal steady-state distribution of cholesterol among various cellular membranes and lipoproteins.     

Assay and properties of 25-hydroxyvitamin D3 23-hydroxylase. Evidence that 23,25-dihydroxyvitamin D3 is a major metabolite in 1,25-dihydroxyvitamin D3-treated or fasted guinea pigs.

Incubation of 25-hydroxyvitamin D3 with kidney cortex mitochondria from 1,25-dihydroxyvitamin D3-treated guinea pigs resulted in the formation of 23,25-dihydroxyvitamin D3 as the major product. The identity of the product was verified by g.c.-m.s. and quantification was performed by h.p.l.c. The rates of the reaction were in the range 1.0-1.8 pmol/min per mg of mitochondrial protein (at 37 degrees C), which were 5-10 times the rates of formation of 24,25-dihydroxyvitamin D3. In mitochondrial preparations from untreated guinea pigs, the rate of 23-hydroxylation was below detection limit (0.02 pmol/min per mg of mitochondrial protein). Fasting the animals for 24 h induced the 23-hydroxylase almost as efficiently as treatment with 1,25-dihydroxyvitamin D3, with a concomitant depression of the 1 alpha-hydroxylase. The 23-hydroxylase reaction required oxidizable substrate, was decreased by low O2 partial pressures and inhibited by CO or the uncoupler carbonyl cyanide p-trifluoromethoxyphenylhydrazone. It was stimulated by the respiratory-chain inhibitors rotenone, antimycin A and KCN. These results indicate that the guinea-pig renal mitochondrial 23-hydroxylase is a cytochrome P-450 and that the reducing equivalents are primarily supplied by NADPH via the energy-dependent transhydrogenase.     

Studies on protein kinases involved in regulation of nucleocytoplasmic mRNA transport.

The rate of energy-dependent nucleoside triphosphatase (NTPase)-mediated nucleocytoplasmic translocation of poly(A)-containing mRNA [poly(A)+mRNA] across the nuclear envelope is thought to be regulated by poly(A)-sensitive phosphorylation and dephosphorylation of nuclear-envelope protein. Studying the phosphorylation-related inhibition of the NTPase, we found that phosphorylation of one polypeptide of rat liver envelopes by endogenous NI- and NII-like protein kinase was particularly sensitive to poly(A). This polypeptide (106 kDa) was also phosphorylated by nuclear-envelope-bound Ca2+-activated and phospholipid-dependent protein kinase (protein kinase C). Activation of kinase C by tumour-promoting phorbol esters resulted in inhibition of nuclear-envelope NTPase activity and in a concomitant decrease of mRNA (actin) efflux rate from isolated rat liver nuclei. Protein kinase C, but not nuclear envelope NI-like or NII-like protein kinase, was found to be solubilized from the envelope by Triton X-100, whereas the presumable poly(A)-binding site [the 106 kDa polypeptide, representing the putative carrier for poly(A)+mRNA transport] remained bound to this structure. RNA efflux from detergent-treated nuclei lost its susceptibility to phorbol esters. Addition of purified protein kinase C to these nuclei restored the effect of the tumour promoters. Protein kinase C was found to bind also to isolated rat liver nuclear matrices in the absence but not in the presence of ATP. The NII-like nuclear-envelope protein kinase co-purified together with the 106 kDa polypeptide which specifically binds to poly(A) in an ATP-labile linkage.     

Cartilage proteoglycans. Assembly with hyaluronate and link protein as studied by electron microscopy.

Aggregates formed by the interaction of cartilage proteoglycan monomers and fragments thereof with hyaluronate were studied by electron microscopy by use of rotary shadowing [Wiedemann, Paulsson, Timpl, Engel & Heinegård (1984) Biochem. J. 224, 331-333]. The differences in shape and packing of the proteins bound along the hyaluronate strand in aggregates formed in the presence and in the absence of link protein were examined in detail. The high resolution of the method allowed examination of the involvement in hyaluronate binding of the globular core-protein domains G1, G2 and G3 [Wiedemann, Paulsson, Timpl, Engel & Heinegård (1984) Biochem. J. 224, 331-333; Paulsson, Mörgelin, Wiedemann, Beardmore-Gray, Dunham, Hardingham, Heinegård, Timpl & Engel (1987) Biochem. J. 245, 763-772]. Fragments comprising the globular hyaluronate-binding region G1 form complexes with hyaluronate with an appearance of necklace-like structures, statistically interspaced by free hyaluronate strands. The closest centre-to-centre distance found between adjacent G1 domains was 12 nm. Another fragment comprising the binding region G1 and the adjacent second globular domain G2 attaches to hyaluronate only by one globule. Also, the core protein obtained by chondroitinase digestion of proteoglycan monomer binds only by domain G1, with domain G3 furthest removed from the hyaluronate. Globule G1 shows a statistical distribution along the hyaluronate strands. In contrast, when link protein is added, binding is no longer random, but instead uninterrupted densely packed aggregates are formed.     

Site-directed mutagenesis to fine-tune enzyme specificity

We have used a combination of a genetic selection and oligonucleotide-directed mutagenesis to introduce a series of amino acid replacements for a single residue into Escherichia coli glutaminyl-tRNA synthetase. The mutant enzymes mischarge supF tRNA(Tyr), with glutamine, to varying degrees depending on the polarity of the side chain introduced but apparently not depending on the size or shape of the side chain. These results indicate that repulsive charge-charge interactions may be important for specific recognition of nucleic acids by proteins and illustrate how a mutant, derived from genetic selection, may be further modified in activity by oligonucleotide-directed mutagenesis.