An amphitropic cAMP-binding protein in yeast mitochondria. 2. Phospholipid nature of the membrane anchor

We describe the first example of a mitochondrial protein with a covalently attached phosphatidylinositol moiety acting as a membrane anchor. The protein can be metabolically labeled with both stearic acid and inositol. The stearic acid label is removed by phospholipase D whereupon the protein with the retained inositol label is released from the membrane. This protein is a cAMP receptor of the yeast Saccharomyces cerevisiae and tightly associated with the inner mitochondrial membrane. However, it is converted into a soluble form during incubation of isolated mitochondria with Ca2+ and phospholipid (or lipid derivatives). This transition requires the action of a proteinaceous, N-ethylmaleimide-sensitive component of the intermembrane space and is accompanied by a decrease in the lipophilicity of the cAMP receptor. We propose that the component of the intermembrane space triggers the amphitropic behavior of the mitochondrial lipid-modified cAMP-binding protein through a phospholipase activity.     

An amphitropic cAMP-binding protein in yeast mitochondria. 3. Membrane release requires both Ca2(+)-dependent phosphorylation of the cAMP-binding protein and a phospholipid-activated mitochondrial phospholipase

The amphitropic cAMP-binding protein in mitochondria of the yeast Saccharomyces cerevisiae is released from the inner membrane into the intermembrane space by the degradation of its lipid membrane anchor consisting of or containing phosphatidylinositol. The releasing reaction depends on the presence of an N-ethylmaleimide-sensitive protein (releasing factor) in the intermembrane space and is controlled by Ca2+ and phospholipid (or lipid derivatives). Here we demonstrate that these two effector molecules act on different activation steps within a complex releasing pathway involving both the cAMP receptor and the releasing factor: Ca2(+)-dependent phosphorylation of the receptor protein seems to be prerequisite for its subsequent lipolytic liberation from the inner membrane. In the presence of phospholipid (or lipid derivatives) the previously soluble releasing factor, which may be identical with a soluble diacylglycerol-binding protein in the mitochondrial intermembrane space, associates with the inner membrane. This change in the intramitochondrial location of the releasing factor, which thus exhibits amphitropic behavior itself, may be required for (direct or indirect) activation of the mitochondrial phospholipase which then releases the cAMP receptor from the inner membrane in a form liable to dissociation from the C subunit by cAMP.     

Interactions of calcium with yeast mitochondria.

The interactions of Ca2+ with mitochondria from Saccharomyces cerevisiae were explored. Mitochondria were loaded with the metallochromic dye Fluo-3 to measure the concentration of free calcium in the matrix. Addition of EGTA or Ca2+ led to fluctuations in mitochondrial free calcium between 120 and 400 nM. Ca2+ variations were slower at 4 degrees C than at 25 degrees C or in the presence of phosphate instead of acetate. The net uptake of 45Ca2+ was higher with phosphate than with acetate. The optimum pH for Ca2+ uptake was 6.8. Ruthenium red did not affect the uptake of Ca2+. Addition of antimycin-A or uncouplers led to a small and transient release of Ca2+. Addition of EGTA or the monovalent cations Na+ or K+ resulted in higher release of Ca2+. Site I but not site II dependent O2 consumption was partially inhibited by EGTA. The effect of Ca2+ on NADH oxidation is similar to results reported with enzymes from mammalian sources which use NADH, such as the pyruvate, isocitrate and oxoglutarate dehydrogenases.     

A cAMP-binding ectoprotein in the yeast Saccharomyces cerevisiae.

Purified plasma membranes from the yeast Saccharomyces cerevisiae bind about 1.2 pmol of cAMP/mg of protein with high affinity (Kd = 6 nM). By using photoaffinity labeling with 8-N3-[32P]cAMP, we have identified in plasma membrane vesicles a cAMP-binding protein (Mr = 54,000) that is present also in bcy1 disruption mutants, lacking the cytoplasmic R subunit of protein kinase A (PKA). This argues that it is genetically unrelated to PKA. Neither high salt, nor alkaline carbonate, nor cAMP extract the protein from the membrane, suggesting that it is not peripherally bound. The observation that (glycosyl)phosphatidylinositol-specific phospholipases (or nitrous acid) release the amphiphilic protein from the membrane, thereby converting it to a hydrophilic form, indicates anchorage by a glycolipidic membrane anchor. Treatment with N-glycanase reduces the Mr to 44,000-46,000 indicative of a modification by N-linked carbohydrate side chain(s). In addition to the action of a phospholipase, the efficient release from the membrane requires the removal of the carbohydrate side chain(s) or the presence of high salt or methyl alpha-mannopyranoside, suggesting complex interactions with the membrane involving not only the glycolipidic anchor but also the glycan side chain(s). Topological studies show that the protein is exposed to the periplasmic space, raising intriguing questions for the function of this protein.     

Phosphorylation of the yeast phospholipid synthesis regulatory protein Opi1p by protein kinase C

Opi1p is a negative regulator of expression of phospholipid-synthesizing enzymes in the yeast Saccharomyces cerevisiae. In this work, we examined the phosphorylation of Opi1p by protein kinase C. Using a purified maltose-binding protein-Opi1p fusion protein as a substrate, protein kinase C activity was time- and dose-dependent, and dependent on the concentrations of Opi1p and ATP. Protein kinase C phosphorylated Opi1p on a serine residue. The Opi1p synthetic peptide GVLKQSCRQK, which contained a protein kinase C sequence motif at Ser(26), was a substrate for protein kinase C. Phosphorylation of a purified S26A mutant maltose-binding protein-Opi1p fusion protein by the kinase was reduced when compared with the wild-type protein. A major phosphopeptide present in purified wild-type Opi1p was absent from the purified S26A mutant protein. In vivo labeling experiments showed that the phosphorylation of Opi1p was physiologically relevant, and that the extent of phosphorylation of the S26A mutant protein was reduced by 50% when compared with the wild-type protein. The physiological consequence of the phosphorylation of Opi1p at Ser(26) was examined by measuring the effect of the S26A mutation on the expression of the phospholipid synthesis gene INO1. The beta-galactosidase activity driven by an INO1-CYC-lacI'Z reporter gene in opi1Delta mutant cells expressing the S26A mutant Opi1p was about 50% lower than that of cells expressing the wild-type Opi1p protein. These data supported the conclusion that phosphorylation of Opi1p at Ser(26) mediated the attenuation of the negative regulatory function of Opi1p on the expression of the INO1 gene.     

Synergistic effects of ethanol and temperature on yeast mitochondria

At temperatures lower than 37°C, the ethanol inhibition constant (Ki) for growth or fermentation inrho ⁺ cells of theSaccharomyces cerevisiae strain S288C was always higher (1.1M) than inrho ^– mutants (0.7M). At 37°C these differences disappeared, and both strains were equally inhibited by ethanol (Ki=0.7m). Mitochondrial activity can be inhibited by high ethanol concentration and temperature. In fact, the stronger inhibition by ethanol of therho ⁺ strain at 37°C was due to the fact that, under these conditions, this strain loses the advantage conferred by mitochondrial activity since the induction ofrho ^– cells in the population is very high. This does not result in an increase in the frequency ofrho ^– mutants because of the poor viability of these mutants in conditions of high temperature and ethanol. In consequence, S288C strain becomes as strongly inhibited by ethanol as therho ^– mutant strains. Differences in viability were not related to the fatty acids and ergosterol composition of the strain. In the presence of ethanol, bothrho ⁺ andrho ^– strains modified their lipids in the same way, but these changes did not improve their ethanol tolerance. They were not due to differences in adaptation to ethanol either, since after successive transfers in ethanol, growth () and fermentation () rates in therho ^– mutants were increasingly inhibited with time, whereas in the S288C strain inhibition of and by ethanol remained unaltered. Rather,rho ^– mutants are less viable thanrho ⁺ cells because of the inability of the former to respire. At 37°C the Ki increased to 0.9M ethanol either when mitochondrial from highly ethanol-tolerant wine yeasts were transferred torho ^– mutants of the strain S288C or when the mitochondria of strain S288C were preadapted by growing the strain in glycerol instead of glucose before it was cultivated in ethanol.     

Inner membrane protease I, an enzyme mediating intramitochondrial protein sorting in yeast.

Several precursors transported from the cytoplasm to the intermembrane space of yeast mitochondria are first cleaved by the MAS-encoded protease in the matrix space and then by additional proteases that have not been characterized. We have now developed a specific assay for one of these other proteases. The enzyme is an integral protein of the inner membrane; it requires divalent cations and acidic phospholipid for activity, and is defective in yeast mutant pet ts2858 which accumulates an incompletely processed cytochrome b2 precursor. The protease contains a 21.4 kd subunit whose C-terminal part is exposed on the outer face of the inner membrane. An antibody against this polypeptide inhibits the activity of the protease. As overproduction of the polypeptide does not increase the activity of the protease in mitochondria, the enzyme may be a hetero-oligomer. This 'inner membrane protease I' shares several key features with the leader peptidase of Escherichia coli and the signal peptidase of the endoplasmic reticulum.     

Catalytic properties of the yeast phospholipid transfer protein

The properties of the phosphatidylcholine (PC) transfer reaction catalyzed by the yeast phospholipid transfer protein (TP-I) were examined in vitro. Donor and acceptor membranes consisted of unilamellar (ULV) and multilamellar (MLV) vesicles, respectively. The phospholipid composition of the membranes participating in the transfer reaction, and in particular that of the MLV acceptors, have a tremendous effect upon the rate of PC-catalyzed transfer. Phosphatidylethanolamine (PE) is an essential component of the acceptor membrane, but it alone is not sufficient to sustain appreciable transfer rates. If combined in an equimolar ratio with PC, there is only a modest increase in transfer rates. On the other hand, when combined with alternate substrates such as phosphatidylinositol (PI) or phosphatidylserine (PS), very high rates of PC transfer occur. The measurement of transfer rates is not affected by the molecular species of PC used as the radioactive tracer. Evidence is also presented to indicate that the two forms of the transfer protein (TP-I and TP-II) are not identical in terms of their interactions with a membrane surface: differences occur in the levels of transfer of PC, PE, PI, and PS at equilibrium. Finally, by kinetic analysis, the mechanism of the protein-catalyzed transfer of PC is shown to conform to a ping-pong bibi model with excess substrate inhibition, analogous to ordinary two-substrate enzyme-catalyzed reactions. Both the rates of desorption and adsorption of the protein from the surface of the ULV are much greater than those describing the similar interactions of the protein with MLV.(ABSTRACT TRUNCATED AT 250 WORDS)     

Reconstitution of ATP-32-Pi exchange by phospholipid addition to the purified oligomycin-sensitive ATPase from yeast mitochondria.

A purified preparation of the oligomycin-sensitive ATPase from yeast mitochondria has been shown to elicit ATP-³²P_i exchange when combined with phospholipids. The reconstitution was normally carried out by dialysis of an ATPase-phospholipid-bile detergent mixture, but could also be achieved by direct addition of the lipid. Vesicle structures with diameters between 200 and 1500 Å were seen by electron microscopy.The ATP-³²P_i exchange was independent of electron transport but sensitive to uncouplers and energy-transfer inhibitors. As in mitochondria, ATPase activity in the reconstituted system was stimulated by a range of uncouplers which inhibited ATP-³²P_i exchange. Taken together, the results raise the possibility that the terminal coupling mechanism might still be intact within the ATPase complex.     

Direct control by calcium of 25-hydroxycholecalciferol-1-hydroxylase activity in chick kidney mitochondria

Kidney mitochondria were isolated from rachitic chicks and their activity in the metabolism of 25-OH-D₃ was studied in relation to the amount of calcium added $\text{in vitro}$. The addition of 0.050.2 mM calcium to a mitochondrial suspension caused a marked and dose-related stimulation of 1-hydroxylation. A sharp decline in the activity was induced by higher concentrations (0.3-0.5 mM) of calcium. The rate of 24-hydroxylation was not influenced by calcium. In these effects, calcium was relatively specific among various divalent cations. These data strongly suggest that calcium is directly involved in the regulation of the vitamin D activation in kidney mitochondria.     

A novel protein, Mpm1, of the mitochondria of the yeast Saccharomyces cerevisiae.

A previously uncharacterized yeast protein, YJL066c, was discovered in the membrane fraction although it has no hydrophobic stretch. The protein was partly solubilized by Triton X-100 in an oligomeric form, while it was insoluble in alkali or salt. By immunofluorescent microscopy, its localization coincided with the mitochondria. We therefore propose it should be named Mpm1 (mitochondrial peculiar membrane protein 1).     

Assembly of the iron-sulfur protein into the cytochrome b-c1, complex of yeast mitochondria.

The assembly of the iron-sulfur protein into the cytochrome bc1 complex after import and processing of the precursor form into mitochondria in vitro was investigated by immunoprecipitation of the radiolabeled iron-sulfur protein from detergent-solubilized mitochondria with specific antisera. After import in vitro, the labeled mature form of the iron-sulfur protein was immunoprecipitated by antisera against both the iron-sulfur protein and the entire bc1 complex from mitochondria solubilized with either Triton X-100 or dodecyl maltoside. After sodium dodecyl sulfate solubilization of mitochondria, however, the antiserum against the iron-sulfur protein, but not that against the bc1 complex, immunoprecipitated the radiolabeled iron-sulfur protein. These results suggest that in mitochondria the mature form of the iron-sulfur protein is assembled with other subunits of the bc1 complex that are recognized by the antiserum against the bc1 complex. By contrast, the intermediate and precursor forms of the iron-sulfur protein that accumulated in the matrix when proteolytic processing was blocked with EDTA and o-phenanthroline were not efficiently assembled into the bc1 complex. The import and processing of the iron-sulfur protein also occurred in mitochondria lacking either cytochrome b (W-267) or the iron-sulfur protein (JPJ1). The mature form of the iron-sulfur protein was immunoprecipitated by antisera against the bc1 complex or core protein I after import in vitro into these mitochondria, suggesting that the mature form is associated with other subunits of the bc1 complex in these strains.     

Hsp60-independent protein folding in the matrix of yeast mitochondria.

Proteins that are imported from the cytosol into mitochondria cross the mitochondrial membranes in an unfolded conformation and then fold in the matrix. Some of these proteins require the chaperonin hsp60 for folding. To test whether hsp60 is required for the folding of all imported matrix proteins, we monitored the folding of four monomeric proteins after import into mitochondria from wild-type yeast or from a mutant strain in which hsp60 had been inactivated. The four precursors included two authentic matrix proteins (rhodanese and the mitochondrial cyclophilin Cpr3p) and two artificial precursors (matrix-targeted variants of dihydrofolate reductase and barnase). Only rhodanese formed a tight complex with hsp60 and required hsp60 for folding. The three other proteins folded efficiently without, and showed no detectable binding to, hsp60. Thus, the mitochondrial chaperonin system is not essential for the folding of all matrix proteins. These data agree well with earlier in vitro studies, which had demonstrated that only a subset of proteins require chaperones for efficient folding.     

Differential regulation of basic protein phosphorylation by calcium phospholipid and cyclic-AMP-dependent protein kinases

Myelin basic protein, an 80-kilodalton (kDa) protein in rat oligodendrocytes, and an 80-kDa basic protein in neuroblastoma x neonatal Chinese hamster brain explant hybrids were phosphorylated extensively when the cells were treated with either phorbol esters (TPA) or diacylglycerols (e.g., oleyoyl-acetylglycerol). TPA-stimulated phosphorylation was inhibited by pre-incubation with 50 microM psychosine (galactosyl-sphingosine), confirming that it is mediated through the phospholipid-dependent protein kinase C (PK-C). Surprisingly, phosphorylation of these proteins was inhibited by incubation of cells with agents which result in activation of cyclic-AMP-dependent protein kinase (dibutyryl cyclic AMP or forskolin). In contrast, phosphorylation of other nonbasic proteins, for example, the oligodendrocyte-specific 2',3'-cyclic nucleotide phosphohydrolase, was stimulated under these conditions (Vartanian et al.: Proceedings of the National Academy of Sciences of the United States of America 85:939, 1988). The possible role of cyclic AMP in activating specific phosphatases or restricting the availability of diacylglycerol for PK-C activation is discussed.     

Stability control of MTL1 mRNA by the RNA-binding protein Khd1p in yeast

Khd1p (KH-domain protein 1) is a yeast RNA-binding protein highly homologous to mammalian hnRNP K. Khd1p associates with hundreds of potential mRNA targets including a bud-localized ASH1 mRNA and mRNAs encoding membrane-associated proteins such as Mid2p and Mtl1p. While Khd1p negatively regulates gene expression of Ash1p by translational repression, Khd1p positively regulates gene expression of Mtl1p by mRNA stabilization. To investigate how Khd1p regulates the stability of MTL1 mRNA, we searched for cis-acting elements and trans-acting factors controlling MTL1 mRNA stability. Regional analysis revealed that partial deletion of the coding sequences of MTL1 mRNA restored the decreased MTL1 mRNA and protein levels in khd1Δ mutants. This region, encompassing nucleotides 532 to 1032 of the Mtl1p coding sequence, contains CNN repeats that direct Khd1p-binding. Insertion of this sequence into other mRNAs conferred mRNA instability in khd1Δ mutants. We further searched for factors involved in the destabilization of MTL1 mRNA. Mutations in CCR4 and CAF1/POP2, encoding major cytoplasmic deadenylases, or of SKI genes, which code for components of a complex involved in 3' to 5' degradation, did not restore the decreased MTL1 mRNA levels caused by khd1Δ mutation. However, mutations in DCP1 and DCP2, encoding a decapping enzyme complex, and XRN1, encoding a 5'-3' exonuclease, restored the decreased MTL1 mRNA levels. Furthermore, Khd1p colocalized with Dcp1p in processing bodies, cytoplasmic sites for mRNA degradation. Our results suggest that MTL1 mRNA bears a cis-acting element involved in destabilization by the decapping enzyme and the 5'-3' exonuclease, and Khd1p stabilizes MTL1 mRNA through binding to this element.