wet weight) was placed in tepid water briefly to loosen the packed cell mass which was then shaken into a large mortar (18 cm) containing

Transcription

1 LIPID SYNTHESIS IN LOW PROTEIN HOMOGENATES OF YEAST EFFECTS OF MITOCHONDRIA AND ETHYLENEDIAMINETETRAACETATE' SEYMOUR GREENFIELD2 AND HAROLD P. KLEIN Biology Department, Brandeis University, Waltham, Massachusetts Received for publication September 25, 1959 Previous reports on lipid synthesis in yea st homogenates have emphasized that the removal of the mitochondrial fraction does not diminislh the activity of these extracts (Klein and Booher, 1955; Klein, 1957). A fraction sedimenting in 3 min between 25, X G and 6, X G, wheni added to the supernatant obtained by centrifugation at 1, X G for 3 min, was found to be highly active in lipogenesis. On the other hand, Corwin et al. (1956), and Schuytema and Lata (1958) presented evidence f or the apparenit involvement of mitochondria. in fat formation in yeast extracts. These conflicting views may now be reconciled by findings to be detailed below. When yeast cells are subjected to mild conditions of rupture, the resulting (low protein) extracts are indeed inactive in the absence of the mitochondrial fraction. However, even in these preparations it can be shown that the enzymatic equipment involved in lipid synthesis is extramitochondrial, since low concentrations of ethylenediaminetetraacetate (EDTA) can substitute for the mitochondrial fraction. MATERIALS AND METHODS Organism and conditions for cultivation. The organism used was Saccharomyces cerevisiae strain LK2G12. Details concerning media and methods of cultivation have been given (Klein, 1957). Preparation of low protein cell-free extracts. Cells from 1.7 L of medium contained in a 2-L flask were washed twice in.1 M phosphate buffer (ph 7.8), resuspended in 75 ml of.1 M phosphate buffer (ph 7.), containing 1 per cent glucose, and then aerated for 21 hr on a magnetic stirrer. After this, they were centrifuged again, l Supported in part by a grant from the U. S. Public Health Service (H-2421 C2). 2 Present address: Department of Microbiology, School of Medicine, Western Reserve University, Cleveland 6, Ohio. washed twice with cold distilled water, and immediately frozen at -2 C. As a routine procedure, cells kept frozen for more than 1 week were not used in any experiments. The centrifuge tube containing the frozen cells (approximately 35 g wet weight) was placed in tepid water briefly to loosen the packed cell mass which was then shaken into a large mortar (18 cm) containing 4 ml of finely powdered dry ice and 3 g of washed glass beads (Superbrite type 11, Minnesota Mining Company). The cells were ground with a pestle for about 25 min, 2 to 3 ml of.1 M phosphate buffer (ph 7.) were added, and the mixture was allowed to thaw undisturbed. Differential centrifugation, conducted according to the procedure of Klein (1957), yielded crude homogenates containing approximaxtely 1 mg protein per ml.3 Incubation procedure. Samples (1. ml) of the crude homogenate, or fractions thereof, were incubated in air or under 1 per cent CO2 at 3 C in Warburg vessels containing 3.3,umoles of 1-C'4-acetate (approximately 5 X 1' cpm), 1.3,umoles adenosine triphosphate, plus various additions according to the particular experiment. The total volume varied between 1.1 and 1.5 ml. After 4 hr, metabolic activity was stopped by the addition of.5 ml of saturated KOH, and the lipids were separated according to procedures given earlier (Klein 1955, 1957). Radioactivity determinations. The lipid extracts, dissolved in chloroform, were plated directly on stainless steel planchets as infinitely thin suspensions and, after drying, were counted with a Packard-Wood end window gas flow tube operating in the proportional range. The scaler used was manufactured by Atomic Instruments, Inc. (model No. 19). All experiments were 3By contrast, extracts (high protein), obtained by the original method (Klein, 1957), contain 2 to 3 mg protein per ml. 691

2 692 GREENFIELD AND KLEIN [VOtL. 79 carried out in duplicate and the radioactivity values given are the result of averaging each set of experimental determinations. Protein determination. Proteins were determined by the biuret method according to Gornall et al. (1949). Chemicals. Sodium-acetate (1_C14) (New England Nuclear Corporation) and adenosine triphosphate (ATP) (disodium salt, Nutritional Biochemicals Corporation) were used. Other chemicals were C.P. unless specified otherwise. RESULTS Effect of EDTA and of the mitochondrial fraction on acetate incorporation. Table 1 illustrates the behavior of the low protein extracts. Removal of the mitochondrial fraction decreased acetate incorporation into lipids to less than 5 per cent of the value obtained with the crude homogenate. However, addition of EDTA at a final concentration of.2 M restored much of the original activity to the mitochondria-free suspension. These observations suggested that the mitochondrial fraction and EDTA might have a common site of action. If this were the case, EDTA should have no stimulatory effect on a system saturated by mitochondria. Similarly, addition of the mitochondrial fraction to a suspension containing optimal EDTA concentrations should cause no increase in acetate incorporation. The data, presented in figures 1 and 2, show that this is not the case. When increasing amounts of EDTA were added separately to crude homogenate and to the mitochondria-free extract TABLE 1 Effect of EDTA on lipogenesis in low protein honogenates Fraction and Additions Acetate NSFa Incorporation FAFa cpim efm Supernatant Ilb ,23 58,66 Supernatant I1b ,38 Supernatant III + EDTA,.2 M... 7,28 23,64 a NSF = nonsaponifiable lipid fraction; FAF = fatty acid fraction. bsupernatant II = crude homogenate; supernatant III = crude homogenate after removal of mitochondrial fraction. Supernatant II in this experiment contained 1.2 mg protein per ml. v' 1 o I t,/supernatant Em a.s tl.. o 4 NONSAPONIFIABLE LIPID _ SUPERNATAN 3 SUERATN 2 l o SUPERNATAN T m 1 2 xio 5xlO 14 2x 1 5xlO 2XO EDTA CONCENTRATION (MOLES/LITER) Figure 1. Effect of EDTA on supernatant II and supernatant III. Samples (1 ml) of supernatant II or supernatant III were incubated in air for 4 hr with 3.3 ismoles of 1-CI4-acetate (5 X 11 cpm), 1.3 samoles ATP, and EDTA in the concentrations indicated. (figure 1), the former was as sensitive to EDTAstimulation as the latter.4 Furthermore, when graded amounts of the mitochondrial fraction were added to supernatant III, the presence of.1 M EDTA resulted in increased incorporation at all levels of the mitochondrial fraction tested (figure 2). Effects under anaerobiosis. Since mitochondria are thought to be primarily involved in oxidative phosphorylation, it seemed possible that this fraction might be involved indirectly in lipogenesis in the generation of energy. Accordingly, incorporation experiments were carried out under anaerobic5 conditions in order to preclude oxidative activity. In figures 3 and 4 are given the results of 4EDTA inhibits incorporation of acetate in these preparations at concentrations greater than.1 M. Such inhibitory effects have been reported by Klein (1958) using high protein homogenates. I As incorporation into the nonsaponifiable lipid fraction in the absence of air is virtually absent (Klein, 1957), only the fatty acid fractions were assayed in these experiments.

3 1961 LIPID SYNTHESIS IN YEAST 693 addition of the mitochondrial fraction to the latter increases acetate incorporation (figure 4). As a consequence of these observations, therefore, oxidative phosphorylation cannot be invoked to explain the stimulatory effects recorded. These experiments also demonstrate the stimu- SUPERNATANT m 4 NONSAPONIFIABLE LIPID SUPERNATANT Il + SUPERNATANT m O MILLIGRAMS PROTEIN OF MITOCHONDRIAL Figure 2. Effect of adding EDTA and the mitochondrial fraction to supernatant III. Samples (1 ml) of supernatant III were incubated in air for 4 hr with 3.3 zmoles of 1-C14-acetate (5 X 15 cpm), 1.3,umoles ATP, and the mitochondrial fraction at the concentrations indicated. Open circles: in presence of 1-4 M EDTA; closed circles: without EDTA. 4 SUPERNATANT I[ to 3 FATTY ACID X 1* SUPERNATANT i m \ ~ ~3 3 5xlO- 1 2xl1 5x; 2x1- EDTA CONCENTRATION(MOLES/LITER) Figure S. Effect of EDTA on supernatant II and supernatant III under anaerobic conditions. Conditions were the same as in figure 1, except that the extracts were incubated under an atmosphere of C2. experiments in which incubation was under an atmosphere of CO2. Here again, it is clear that the crude homogenate is more active than the mitochondria-free extract (figure 3) and that 8C,, 6 E 4 a.2 C.) 2~ O MILLIGRAMS PROTEIN OF MITOCHONDRIAL Figure 4. Effect of EDTA and the mitochondrial fraction on supernatant III under anaerobic conditions. Conditions were the same as in figure 2, except that the extracts were incubated under C2, and the concentration of EDTA was increased to 2 X 1-4 M. TABLE 2 Effect of various chelating agents on lipogenesis in yeast homogenates Fraction and Additions FATTY ACID SUPERNATANT m + 2x 1-4M EDTA SUPERNATANT 3m Acetate Incorporation NSFa FAFa cpm cpm Supernatant 11b... 2,41 8,43 Supernatant III EDTA,.2 M... 7,7 13,78 + EDTA,.1 M... 4,8 13,93 + Salicylaldoxime,.2 M 84 + Salicyladoxime,.1 M 1,2 + 8-Hydroxyquinolin,. 2 M Hydroxyquinolin,. 1 M Kojic acid,. 2 M Kojic acid,.1 M... 1,9 + Citric acid,.2 M... + Citric acid,.1m 17 + o-phenanthroline,.2m... + o-phenanthroline,. 1 M For these designations, see table 1. bprotein content = 7.9 mg per ml.

4 694 GREENFIELD AND KLEIN [vol. 79 latory effects of EDTA in both crude and mitochondria-free preparations. Effect of other agents. Structural considerations (Martell and Calvin, 1952) suggest that EDTA TABLE 3 Effect of various reducing agents on lipogenesis in yeast homogenates Fraction and Additions Acetate Incorporation NSFa FAFG cpim cpm Supernatant III Supernatant III EDTA,.2M Wc + Glutathione,.2 M... 32c Mercaptoethanol,.2 M c + Cysteine,.2 M Ascorbic acid,.2 M c a For these designations, see table' 1. I Protein content = 9. mg per ml. c Single determination. acts by chelating with a multivalent metal. Therefore, other metal chelators were tested for their ability to restore lipid synthesis after removal of the mitochondrial fraction (table 2). Certain reducing compounds also are knowin to act as metal chelators. Four such compounds were tested with supernatant III for their effect on lipogenesis. The data (table 3) show that of these cysteine was somewhat effective in increasing incorporation above the level of the controls. Attempts to fractionate mitochondria. Thus far, attempts to fractionate the stimulatory system of the mitochondrial fraction have proved unsuccessful. Heating of this fraction at 1 C for 5 min destroyed its activating ability. Sonic disintegration in a Raytheon apparatus (1 kc) for 3 min disrupted this fraction, but also resulted in complete loss of activity. With shorter periods of sonic treatment, the stimulatory activity remained with the material sedimentable at 19, X G (i. e. the intact mitochondrial fraction). Similarly, when this fraction was homogenized by hand in a Ten Brock tissue grinder, and subsequently subjected to differ- TABLE 4 Interchanging of fractions of high protein and low protein homogenates Acetate Incorporation Fraction and Additions LPa extract HPa extract NSF_ FAFb NSF FAF cpm cpm cpm cfn Supernatant Ic 6,76 2,43 14,99 36,33 Supernatant III 11,25 34,8 Supernatant III + 2 X 1-4 M EDTA 2,66 3,71 -d Supernatant IV 3,83 4,24 Supernatant IV + LP small particle fraction (.56 mg 5,49 4,1 protein) Supernatant IV + LP small particle fraction (.56 mg pro- 1,75 2,99 tein) + 2 X 1-4 M EDTA Supernatant IV + LP small particle fraction (1.12 mg pro- - 3,73 2,68 tein) Supernatant IV + HP small particle fraction (1.36 mg pro- 2,53 3,26 - tein) Supernatant IV + HP small particle fraction (2.73 mg pro- 4,94 7,91 12,87 12,97 tein) a LP = low protein; HP = high protein. b For these designations see table 1. Protein content: LP = 1.5 mg per ml; HP = 21.2 mg per ml. d Not done.

5 196] LIPID SYNTHESIS IN YEAST 695 ential centrifugation, once more only the pellet sedimenting at 19, X G showed any stimulatory activity when added to supernatant III. As none of the procedures detailed above yielded pertinent information on the possible locus in the mitochondrial fraction of the stimulatory effect, we turned to a direct comparison of low protein (LP) and high protein (HP) preparations, since it seemed reasonable to assume that the latter contained this material. Cells from 5 flasks of media were harvested and aerated as described, divided into two equal parts, and frozen overnight. One part was ground as described in the Materials and Methods section to produce LP homogenates. The other part was ground according to Klein (1957) yielding HP homogenates. Twenty-two ml of.1 M phosphate buffer (ph 7.5) were added to each homogenate. From each, supernatants II and III were obtained, after which a portion of each supernatant III was centrifuged for 3 min at 1, X G to obtain the small particle fraction. Samples of HP particle-free supernatant (supernatant IV) were combined with.1 ml of the HP small particle fraction, or with.1 and.2 ml of the LP small particle fraction. Samples of LP supernatant IV were combined with.1 ml of the LP small particle fraction, or with.5 and.1 ml of the HP small particle fraction. Subsequent procedures were performed as described before. The results are recorded in table 4. As was to be expected, LP supernatant III was inactive whereas HP supernatant III was active in incorporating acetate into lipids. The low activity observed with HP supernatant IV may be attributed to contamination with the small particle fraction, as even a slight trace of small particles permits some lipid synthesis to occur (Klein, unpublished results). The combination of LP supernatant IV and LP small particle fraction was inactive unless EDTA was added. However, when the HP small particle fraction was combined with this same LP supernatant IV, synthesis occurred. On the other hand, the LP small particle fraction was inert when added to HP supernatant IV. Thus the particle-free supernatants of both preparations can be stimulated by particles from HP extracts. The two particle preparations differ, however, and it may tentatively be assumed that this difference is due to the presence of particle-bound material from the mitochondrial fraction in the HP small particle fraction. DISCUSSION The contribution of the mitochondria to lipid synthesis in cell-free systems is somewhat obscure, although the majority of evidence (PopjAk, 1958) suggests that these particles are not directly involved in this process. In those instances where the mitochondria have been implicated, it was demonstrated that they could be dispensed with by the addition of electrolytes to the homogenization medium (Dituri et al., 1957). Presumably, some essential factor was extracted from the mitochondria by this procedure. In the present experiments we have described a lipid-synthesizing system from yeast in which the mitochondrial fraction appears to be necessarv for synthesis. However, closer observation reveals that the mitochondrial fraction is not obligatory for lipogenesis since low concentrations of EDTA can substitute for this fraction. Since EDTA presumably functions by chelating multivalent metals, it is possible that the mitochondrial fraction is effecting lipid synthesis by a similar process. However, Friess (1954), studying the effects of chelation on myosin adenosine triphosphatase, presented evidence which suggested that EDTA acted not by chelation but that it had a direct effect on this enzyme. This possibility is not ruled out here since other chelators (except possibly cysteine) failed to substitute for EDTA in mitochondriafree extracts. Experiments have shown that two types of cell-free homogenates can be prepared from our strain of S. cerevisiae, depending on the method of grinding: high protein extracts, which can synthesize lipids in the absence of the mitochondrial fraction, and low protein extracts, which require this fraction or EDTA. The difference between these homogenates appears to reside in their respective small particle fractions. The fact that high protein homogenates are active in the absence of added mitochondria whereas low protein extracts are not suggests that the former preparations already contain mitochondrial stimulating materials. Furthermore, since the small particle fraction from such extracts also stimulates low protein preparations, it is probable that the more rigorous procedures used to obtain high protein extracts damage the mitochondria. This damage presumably releases the activating material in a particle-bound condition.

6 696 GREENFIELD AND KLEIN [VOL. 79 When the mitochondrial fraction and EDTA are used together, an additive stimulation occurs in low protein extracts. The respective effects of these agents, then, appear to be independent and it would be reasonable to assume that they act at different points in the biosynthesis of lipids. However, it should be emphasized that either agent alone is sufficient to allow synthesis to occur. It seems reasonable to conclude from these findings that both affect the production or maintenance of the same specific component in the biosynthetic sequence at a rate-limiting point, but by different mechanisms. The fact that EDTA and the mitochondrial fraction were effective under anaerobic conditions excludes the possibility of their functioning in energy production via oxidative phosphorylation. On the basis of these studies, the following hypothesis is proposed. We suggest that mitochondria-free preparations of low protein extracts contain some substance(s), possibly involved in electron transfer, which is broken down or inactivated by a metal-catalyzed system also present in these preparations. The addition of EDTA to such extracts prevents the loss of this substance by chelating the metal catalyst. On the other hand, addition of the mitochondrial fraction or of the subfraction found in HP small particle fractions, introduces an enzyme system which is capable of adding or generating enough of this substance to enable lipid synthesis to proceed. SUMMARY A cell-free system from Saccharomyces cerevisiae capable of incorporating acetate into fatty acids and nonsaponifiable lipids has been described. These extracts (low protein) differ from those previously reported with this organism (high protein) by being inactive after removal of the mitochondrial fraction. Low concentrations of ethylenediaminetetraacetate and to a lesser extent cysteine, but not salicylaldoxime, o-phenanthroline, 8-hydroxyquinoline, kojic acid, citric acid, glutathione, ascorbate, 2-mercaptoethanol, also stimulate lipid synthesis. Ethylenediaminetetraacetate exerts its effect even in the presence of the mitochondrial fraction to give an additive effect. Direct attempts to extract the activating system from the mitochondrial fraction were unsuccessful. However, when low protein and high protein homogenates were prepared simultaneously and various fractions interchanged, the difference between these preparations appeared to reside in their respective particle fractions. It is suggested that the high protein small particle fraction contains the activating system of the mitochondrial fraction. REFERENCES CORWIN, L. M., SCHROEDER, L. J., AND MCCULLOUGH, W. G Studies of lipid synthesis in cell-free yeast extracts. Arch. Biochem. Biophys., 72, DITURI, F., SHAW, W. N., WARMS, J. V. B., ANID GURIN, S Lipogenesis in particle-free extracts of rat liver. I. Substrates and cofactor requirements. J. Biol. Chem., 226, FRIESS, E. T The effect of a chelating agent on myosin ATPASE. Arch. Biochem. Biophys., 51, GORNALL, A. G., BARDWELL, C. J., AND DAVID, M. M Determination of serum protein by means of the biuret reaction. J. Biol. Chem., 177, KLEIN, H. P Synthesis of lipids in resting cells of Saccharomyces cerevisiae. J. Bacteriol., 69, KLEIN, H. P Some observations on a cellfree lipid synthesizing system from Saccharomyces cerevisiae. J. Bacteriol., 73, KLEIN, H. P Cobalt activation of fattyacid synthesis in yeast homogenates. Science, 128, KLEIN, H. P. AND BOOHER, Z. K Synthesis of lipids in cell-free extracts of yeast. Bacteriol. Proc., 1955, 136. MARTELL, A. M. AND CALVIN, M Chemistry of the metal chelate compounds. Prentice- Hall, Inc., New York. POPJAK, G Biosynthesis of cholesterol and related substances. Ann. Rev. Biochem., 21, SCHUYTEMA, C. G. AND LATA, G Ergosterol synthesis and storage in the yeast Torulopsis lipofera. Arch. Biochem. Biophys., 75, 4-45.