Accumulation of Neutral Lipids in Saccharomyces carlsbergensis by myo-Inositol Deficiency and Its Mechanism RECIPROCAL REGULATION OF YEAST ACETYL-CoA CARBOXYLASE BY FRUCTOSE BISPHOSPHATE

The abnormal accumulation of lipids due to myo-inositol deficiency in and the mechanism involved was investigated. The deficient cells contained much more neutral lipids with a greater ratio of unsaturated fatty acids compared to the supplemented cells, whereas there was no significant change in their phospholipid contents. The biosynthesis of fatty acids and sterols from acetate, and of triacylglycerols and sterol esters from palm&ate was markedly augmented in the deficient cells. Acetyl-CoA carboxylase activity of the deficient supernatant was 2- to 5-fold higher than that of the supplemented. However, the activity from both sources was not significantly different after Sephadex G-25 gel filtration of the supernatant, suggesting the presence of low molecular effector in the deficient supernatant. There was a great increase in acid-soluble glycogen, trehalose, and fructose-1,6-P,, as well as a drastic decrease in citrate in the deficient cells. Their intracellular


From the Shizuoka
College of Pharmaceutical Sciences, 2-2-l Oshika, Shizuoka, Japan (422) The abnormal accumulation of lipids due to myo-inositol deficiency in Saccharomyces carlsbergensis, and the mechanism involved was investigated. The deficient cells contained much more neutral lipids with a greater ratio of unsaturated fatty acids compared to the supplemented cells, whereas there was no significant change in their phospholipid contents.
The biosynthesis of fatty acids and sterols from acetate, and of triacylglycerols and sterol esters from palm&ate was markedly augmented in the deficient cells. Acetyl-CoA carboxylase activity of the deficient supernatant was 2-to 5-fold higher than that of the supplemented. However, the activity from both sources was not significantly different after Sephadex G-25 gel filtration of the supernatant, suggesting the presence of low molecular effector in the deficient supernatant.
There was a great increase in acid-soluble glycogen, trehalose, and fructose-1,6-P,, as well as a drastic decrease in citrate in the deficient cells. Their intracellular levels were calculated so that their effects on acetyl-CoA carboxylase was examined over the range of physiological concentration. Citrate strongly inhibited the enzyme activity of the supernatant, but it had no effect on the preparation after gel filtration. On the other hand, fructose-1,6-Pz stimulated the enzyme activity both before and after gel filtration. The acetyl-CoA carboxylase activity in the gel filtrate was measured as a function of citrate concentration at several fixed concentrations of fructose-1,6-P*. Citrate counteracted the activation by fructose-1,6-P, in a dose-dedpendent manner. Citrate lacked the inhibitory effect in the absence of fructose-1,6-P*. It was concluded from these results that neutral lipid accumulation in the deficient cells reflected an increase in the synthesis of fatty acids, at least partly based on an enhancement of acetyl-CoA carboxylase activity, and that the operation of a reciprocal regulation of the enzyme by fructose-1,6-P, and citrate caused a marked elevation of the enzyme activity in the deficient cells with a high fructose-1,6-P, level and a low citrate level.
Although myo-inositol and its phosphatides have been condition. This caused fatty livers (lo), and we found that the shown to be involved in several important physiological deposition of triacylglycerols in the livers of the deficient rats phenomena, myo-inositol is one of the few vitamins whose resulted from stimulated lipolysis in the adipose tissues (11 Michell (4). The role of inositol as a lipotropic factor was first found in 1941 by Gavin and McHenry (5), and a great deal of work has been published since then (6)(7)(8)(9). A problem in these studies has been obtaining an inositol-deficient state in animals. Fortunately, we have succeeded in producing an inositol deficiency in rats in a short period of time under a balanced dietary grown in the absence of inositol (12). Lewin found a similar phenomenon in Saccharomyces carlsbergensis (13), and he reported that the deposit of lipids mainly consisted of triacylglycerols (14). Also, Paltauf and Johnston have examined the effects of various carbon sources on the lipid accumulation in inositol-deficient S. carlsbergensis (15). animals in the phenomenon of lipid accumulation due to inositol deficiency.
The cultivation of S. carlsbergensis in a selected medium, which had been found to produce the most severe biochemical lesions during inositol deprivation (16,17), resulted in a greater lipid deposition containing sterol esters and acylglycerols than previously reported (14,15). The accumulation of neutral lipids by inositol deficiency was found to result from an increased synthesis of fatty acids and sterols, with a subsequent accelerated formation of acylglycerols and sterol esters.
The biosynthesis of long chain fatty acids in yeast (18)(19)(20)(21) is catalyzed by three enzymatic systems, also present in mammalian cells: acetyl-CoA synthetase, acetyl-CoA carboxylase, and fatty acid synthetase. Although the activity of all these enzymes are now known to be varied by nutritional manipulation, acetyl-CoA carboxylase has been the first to be claimed as a rate-limiting enzyme in fatty acid synthesis, and has been most extensively studied in relation to its regulation. Accordingly, our attention has been focused on the change of acetyl-CoA carboxylase activity due to the inositol deficiency. The enzyme activity in the 100,000 x g supernatant fraction from the deficient cell homogenate has been found to be 2-to &fold higher than that from the supplemented.
Changes in enzyme activities are caused by alteration in catalytic efficiency by allosteric effecters or by the synthesis or degradation of the enzyme. The results obtained have suggested the existence of low molecular effecters on acetyl-CoA carboxylase in the deficient supernatant fraction rather than an increase in the enzyme protein.
The fluctuation in glycolytic'and Krebs' cycle intermediates due to the inositol deficiency was investigated in reference to their possible roles in the regulation of the enzyme. The two intermediates, fructose-l&P, and citrate found to be markedly increased and decreased, respectively, in the deficient cells, have been shown to exert dual regulatory effects on acetyl-CoA carboxylase over the physiological range of their concentrations. These results will have important implications on the clarification of the mechanisms underlying the neutral lipid accumulation in the inositol-deficient cells. for 20 h in the complete medium and 1 ml of suspension of the well-washed cells (absorbance at 620 nm = 1.0) was added to inoculate 200 ml of medium. The cells were cultivated aerobically in 500-ml Myer flasks containing 200 ml of medium with and without the addition of inositol at 30" for the indicated periods. The cells were collected by refrigerated centrifugation and were quickly washed with cold distilled water three times.
Extraction of Lipids and Its Analysis-The total lipids were extracted as follows (34): packed cells were mixed with 0.5 volume of water and 3.75 volumes of a mixture of chloroform/methanol (12) and occasionally shaken for 2 h at room temperature. Then, 1.25 volumes of water and chloroform were added and the solution was mixed well. The mixture was centrifuged and the upper layer was removed. The lower layer containing the total lipids was filtered through silicone-treated phase-separating paper (Whatman 1~s). This extraction procedure for cell lipids gave the most reproducible result among the several procedures tested. The total lipids were separated into lipid classes by thin layer chromatography on Kiesel Gel H chromatoplate, using petroleum ether/ether/acetic acid (70/30/l) as a developing solvent. Total lipids were measured by the method described by Amenta (35), free fatty acids by Duncombe (36), sterols by Kenny (37), acylglycerols by Block and Jarret (38), and phospholipids by Harris and Popat (39). The fatty acid composition of each lipid class was analyzed by gas-liquid chromatography as previously described (10).

Analyses
Other than Lipids-DNA in the cells was extracted by a modified method of Ogur and Rosen (40), and determined according to the method by Ceriotti (41) was consecutively centrifuged at 800 x g for 10 min, at 10,000 x g for 20 min. and at 100,000 x g for 60 min. The 100,000 x g supernatant was passed through a column of Sephadex G-25 equilibrated with the same buffer, or dialyzed against 0.1 M phosphate buffer (pH 7.0) at 0" overnight. Either the 10,000 x g supernatant fraction, the 100,000 x g supernatant fraction, the Sephadex G-25 eluate, or the dialysate was used as the enzyme source for acetyl-CoA carboxylase. Standard Assay of Acetyl-CoA

Carboxylase
Actioity-Acetyl-CoA carboxylase was assayed by measuring the recovery of acid-stable radioactivity after incubation with '"CO,, essentially according to the method of Nakanishi and Numa (43). The standard assay medium (total 0.8 ml) contained 50 mM Tris/HCl (pH 7.5), 10 mg of MgCl,, 3.75 rnM glutathione reduced, 3.75 mM ATP, 0.125 mM acetyl-CoA, 0.75 mg/ml of bovine serum albumin, 12.5 mM of NaHCOs containing 2 &i of "COZ and 2 to 5 mg of protein of enzyme preparation.
The reaction was initiated by adding the enzyme preparation (ordinarily, 0.3 ml) to 0.5 ml of substrate and cofactor mixture in scintillation vials. After incubation at 30" for 10 min, the reaction was terminated by an addition of 0.2 ml of 5 N HCl and the mixture was taken to dryness at 80" for 40 to 60 min. The residue was dissolved in 0.5 ml of water and 10 ml of Bray's scintillation cocktail (60 g of naphthalene, 4 g of 2,5-diphenyloxazolyl (PPO), 0.2 g of 1,4-bis [2-(5-phenyloxazolyl) Ibenzene (POPOP), 100 ml of methanol, 20 ml of ethylene glycol in liter dioxane) and the radioactivity was counted. The reaction was linear up to at least 10 mg of protein in 100,000 x g supernatant fraction. Extmction of Intermediates-The cells harvested at the specified time, were quickly washed by ice-cold distilled water and 1.5 ml of 10% perchloric acid were immediately added to 1.5 ml of cell suspension in water containing approximately 1.35 mg of DNA cell. The mixture was kept for 30 min at room temperature and centrifuged at 20,000 x g for 10 min. An aliquot of the supernatant fluid was neutralized with a buffered potassium hydroxide mixture (44), the resulting precipitate by guest on March 23, 2020 http://www.jbc.org/ Downloaded from being removed by centrifugation. This acid extract was used for the analyses of intermediates, adenosine nucleotides, and oxidized pyridine nucleotides.
Reduced pyridine nucleotides were extracted by heating the cell pellet (approximately 1.35 mg of DNA) with 2.5 ml of 0.5 N KOH at 70" for 1 min with constant agitation and then cooling in ice. After adjusting the pH to 8.5 with 1 M triethanolamine hydrochloride, the solution was centrifuged at 25,000 x g for 10 min (45). Trehalose, glycogen, glucan, and mannan were extracted as described by Trevelyan and Harrison (46 at 800 x g for 15 min. This procedure was repeated until most of the cells were broken. The cell wall and membrane fraction, and the particulated fraction were separated by consecutively centrifuging the homogenate at 800 x g for 10 min, and 100,000 x g for 60 min. The respective fractions were resuspended in the same buffer, centrifuged at 100,000 x g for 60 min, and their volumes of the precipitates were measured. The extracellular volume was measured as follows; 2 ml of cell suspension containing approximately 2 mg of DNA was mixed with 1 ml of 3% dextran and the mixture was centrifuged at 3,000 rpm for 10 min.  (04) and without (04) inositol (25 rg/ml). (b), the time course of the lipid accumulation in the inositol-supplemented and -deficient yeast during cultivation. Aliquots of the cultures were withdrawn for the measurement of the total lipids at the specified time. conditions, the total lipids in the deficient cells were found to be approximately 10 times that in the supplemented cells. The contents of triacylglycerols, diacylglycerols, free fatty acids, sterol esters, and free sterols in the deficient cells were respectively 12-, 3-, 8-, 13-, and g-fold higher than those in the supplemented cells, whereas there was no significant change in their phospholipid content. It should be noted that such a marked increase in the contents of sterol esters and free sterols comparable to the increase in triacylglycerols, has never been observed by the previous workers (14,15), and furthermore, the deposition of neutral lipids was greater than the values already reported. This may result from the usage of medium found to produce severe metabolic lesions on inositol deprivation (16,17). The contents of triacylglycerols and sterol esters in the deficient cells gradually increased in a similar manner to the total lipids ( Fig. l(b)) as the cultivation proceeded up to 96 h while that of phospholipids was always within the range of the supplemented cells.

Fatty Acid Profile of Deposited
Lipids-The profiles of the fatty acid distribution in each lipid class were compared between the supplemented and deficient cells (Fig. 2). Generally, the deficient cells of 48-h growth contained more unsaturated fatty acids such as palmitoleic and oleic acid, with fewer saturated fatty acids such as palmitic and stearic acid. This tendency was more remarkable in free fatty acid, triacylglycerol, and sterol ester fractions than phospholipid fraction. Inositol deficiency also increased the ratio of Cls fatty acids to Cl6 fatty acids, especially in free fatty acid fractions. Although    (100) alteration of a property in the cells. Accordingly, the activity of acetyl-CoA carboxylase which is considered to be a rate-limiting enzyme in fatty acid synthesis in animal (51) was compared in the two cells.
The assay of the enzyme activity was performed essentially by the method of Nakanishi and Numa (43), but the addition of citrate into the assay medium and preincubation with it was omitted because unlike the procedure for the animal enzyme it appeared to be unnecessary.  10,000 x g supernatant fraction for 5 to 20 min at 45" was greater with the deficient preparation than with the supplemented. This may result from the protection of acetyl-CoA carboxylase molecule by low molecular effector( Changes in Levels of Metabolites due to Inositol Deficiency-In an attempt to explore the low molecular effecters on acetyl-CoA carboxylase suggested to be in the supernatant fraction from the deficient cell homogenate, the levels of intermediates in carbohydrate metabolism were comparatively examined, because of a known close relation between the activity of lipogenic enzymes and carbohydrate metabolism, and of the occurrence of some altered metabolites in the inositol-deficient culture (13, 16,17,[53][54][55]. To analyze the intermediates, the fermentation of yeast was immediately stopped by adding 10% perchloric acid (final concentration 5%) into the culture or into the packed cells collected at 36 h after inoculation.
Around this time of cultivation, there was a sufficient increase in acetyl-CoA carboxylase activity in the deficient cell preparation. In Table  V, the intermediates expressed per mg of DNA from the supplemented and deficient cells are compared. Among the intermediates measured, there is a great increase in the content of trehalose and fructose-1,6-P2, whereas a drastic decrease in citrate results in the deficient cells compared to the supplemented cells. Trehalose and fructose-1,6-P* in the deficient cells were 7 and 4 times that of the supplemented cells, respectively, while only one-third of citrate in the supple-mented cells was found in the deficient cells. There was also an increase in acid-soluble glycogen and glucan, and a slight but significant increase in P-enolpyruvate and pyruvate in the deficient cells. The cofactors, ATP, ADP, NAD+, and NADH, were all significantly higher in the deficient cells than in the supplemented cells. The content of AMP was unchanged. The reduction states (NAD+/NADH) in both types of cells were in the same range (1.09 for the supplemented and 1.04 for the deficient).

Measurement of Intracellular
Volume in Cells-The extracellular water volume in packed cells was measured by using dextran, and the intracellular volume were obtained by the conventional method as described under "Experimental Procedure." As shown in Table VI, the whole cell volume expressed relative to DNA of the deficient cells was approximately 40% greater than that of the supplemented, and this increase was exclusively ascribable to an increase in the volume of the cell wall fraction of the deficient cells. No significant change was observed in the volumes of the particulated component and the cytoplasmic fraction of two types of cells. The increase in the volume of cell walls in the deficient cells probably reflects an accumulation of glucan due to inositol deficiency as Ghosh and co-workers (55) Table V. It should be noted that there was a big fluctuation between the supplemented and deficient cells in levels of fructose-1,6-Pz and citrate which have been known to have regulatory functions on a number of enzymes in yeast and mammals.
Effect of Citrate on Acetyl-CoA Carboxylase Activity-In contrast to an augmentation in acetyl-CoA carboxylase activity in the 100,000 x g supernatant fraction from the deficient cells, the level of citrate which is known to activate acetyl-CoA carboxylase from several sources was markedly reduced in the deficient cells. Thus, the effect of citrate on the enzyme preparation from the yeast was examined. The enzyme activity in the 100,000 x g supernatant fraction from the deficient cells, was always 2-to 5-fold higher than that from the supplemented cells depending on the preparations. An addition of 5 to 50 mM citrate into the assay medium strongly inhibited the activity from both sources in a dose-dependent manner as shown in Fig.  3(a), where inhibition is expressed as the percentage of activity relative to that of samples incubated in the absence of citrate. Preincubation with citrate was not needed to cause the inhibitory effect. Furthermore, the magnitude of the inhibition by citrate was much greater with the deficient cells than with the supplemented cells. Fifty per cent inhibition was observed at 25 and 10 mM citrate with the supplemented and deficient enzyme preparations, respectively. Fig. 3(b) represents the effect of citrate on the acetyl-CoA carboxylase activity after Sephadex G-25 chromatography of the 100,000 x g supernatant fractions. By gel filtration, the inhibitory effect of citrate completely diminished up to 20 mM. Citrate neither simulated nor inhibited the acetyl-CoA carboxylase activity in the filtrate, coinciding with the results in brewers' yeast by Matsuhashi and his co-workers (18). A slight inhibition of the activity was observed at 50 mM of citrate. In this experiment, the enzyme activities in the absence of citrate before and after the gel filtration were 5.02 x 10m3 and 1.60 x 10m3 dpm/mg of protein from the deficient uersus 1.36 x 10m3 and 1.22 x 10m3 dpm/mg of protein from the supplemented.
Acetyl-CoA carboxylase' from rat liver was prepared by overnight dialysis of the 100,000 x g supernatant of the homogenate in 0.25 M sucrose against 0.1 M potassium phosphate (pH 7.0); its responsiveness to citrate was tested in the same assay medium used for yeast acetyl-CoA carboxylase without preincubation with citrate. As numerous reports have The suspension of 36-h growth cells (approximately 2 mg of DNA) was centrifuged at 3,000 rpm for 10 min and the whole cell volume was determined using dextran as described in the text. The cell wall and particulated components were separated by centrifuging the cell homogenate at 800 x g for 10 min and 100,000 x g for 60 min, respectively. The volume of the cytoplasm was calculated by subtraction.  (22), rat liver acetyl-CoA carboxylase was markedly stimulated by the presence of citrate. Effect of Fructose-1,6-P2 on Acetyl-CoA Carboxylase Actiuity-Fructose-1,6-P* is another intermediate whose level has changed over a wide range due to inositol deficiency. Fig.  4(a) shows the acetyl-CoA carboxylase activity in the Sephadex G-25 gel filtration of the 100,000 x g supernatant from both types of cells as a function of fructose-1,6-P* concentration. The addition of fructose-1,6-Pz (0 to 5 mM) within the physiological range of concentration to the assay medium stimulated the acetyl-CoA carboxylase activity in a sigmoidal curve. Stimulation by fructose-1,6-P* was observed at as low as 0.5 mM and reached a maximum around 5 mM. The Hill coefficient calculated from a Hill plot of the data (abscissa; log [fructose-1,6-Pz], ordinate; log [V/(V,,, -V)]) was 3.2 for the supplemented and 3.5 for the deficient. The responsiveness of the enzyme to stimulation by 2 mM fructose-1,6-Pz was compared in the various enzyme preparations before and after the gel filtration. The magnitude of the activation by 2 mM fructose-1,6-Pz was much greater in the supplemented 100,000 x g supernatant fraction (7-fold) than in the deficient (2-fold), and furthermore, the magnitude was citrate @In) citrate w4> FIG. 3. Acetyl-CoA carboxylase activity as a function of citrate concentration. Acetyl-CoA carboxylase activity in the 10,000 x g supernatant (a) and in the Sephadex G-25 gel filtrate of the 100,000 x g supernatant (b) from the homogenate of 48-h growth cells was measured in the standard assay medium in the presence of various concentrations of citrate. 0, activity from the supplemented cells; 0, activity from the deficient cells. The enzyme activity from the supplemented and deficient cells in the absence of citrate was 3.92 x lOa dpm/mg of protein and 7.44 x lOa dpm/mg of protein respectively, in (a), and 1.22 x 10' dpm/mg of protein and 1.60 x 10' dpm/mg of protein, respectively, (b). Each point represents the mean of the FIG. 4. Acetyl-CoA carboxylase activity as a function of fructose-1,6-P, (a) and glucose-6-P (b) concentration. The 100,000 x g supernatant of the 42-h growth cell homogenate was passed through a column of Sephadex G-25, and acetyl-CoA carboxylase activity in the filtrate was measured in a standard assay medium in the presence of various concentrations of fructose-1,6-P, or glucose-6-P. 0 indicates the activity from the supplemented cells and 0 that from the deficient cells. Each point represents the mean of the triplicates. unchanged with the supplemented enzyme preparation before and after gel filtration, whereas the degree of the activation was greatly increased with the deficient enzyme preparation after gel filtration to that with the supplemented preparation. Klein and co-workers have tested various glycolytic intermediates for a stimulatory effect on acetyl-CoA carboxylase (31) and fatty acid synthesis (32, 33) in Saccharomyces cereuisiae, and have observed that fructose-l&P2 has a stimulatory effect on acetyl-CoA carboxylase. But the effect he found was much less than that we found, even above the physiological concentration of fructose-1,6-P,. Effects of Glucose-&P, GlyceroUP, Inositol, and Cyclic AMP- Fig.  4(b) represents the acetyl-CoA carboxylase activity as a function of glucose-6-P concentration. Glucose-6-P (0 to 5 mM) slightly stimulated the enzyme activity, and the responsiveness to it was similar in both enzyme preparations. Glucose-6-P is unlikely to be involved in the augmentation of acetyl-CoA carboxylase activity in the deficient yeast. The reasons for this are (a) no change occurred in the levels of glucose-6-P between the supplemented and deficient cells, (b) the concentration required for the activation is much higher than the intracellular concentration, and (c) the degree of activation by glucose-6-P is much less than that by fructose-1,6-Pz.
Glycerol-3-P (0 to 5 mM), inositol, and cyclic AMP in the presence of 1 mM theophylline failed to affect the acetyl-CoA carboxylase activity from both types of cells. Glycerol-3-P was reported to stimulate acetyl-CoA carboxylase from S. cereuisiae at the low concentration of 2 mM (31,32). Cyclic AMP has been shown to inhibit lipogenesis in uiuo in rat liver (56), and to inhibit directly acetyl-CoA carboxylase from the same tissue (57,58), but the effect of cyclic AMP on this enzyme is still controversial. Counteraction by Citrate of Fructose-l&P, Activation of Acetyl-CoA Carboxylase-The acetyl-CoA carboxylase activity in the supernatant fraction from the deficient cells was inhibited by citrate to a greater degree ( Fig.  3(a)) and stimulated by fructose-1,6-Pz to a lesser degree than that from the supplemented cells, whereas the enzyme preparation after Sephadex G-25 chromatography was not responsive to citrate inhibition ( Fig.  3(b)), but was still very sensitive to fructose-1,6-P* stimulation (Fig. 4). These results were interpreted to mean that the enzyme in the deficient cells was activated by the presence of some effector one of which may be fructose-1,6-P*, and that exogenously added citrate reversed the activation by effector( Accordingly, the following experiment was designed to test the hypothesis for a reciprocal regulation of the acetyl-CoA carboxylase by fructose-1,6-P, and citrate in uiuo. Fig. 5   However, except for glycerol-3-P, their effects were tested at concentrations not within the physiological range.
In our experiments, fructose-1,6-Pz activated the acetyl-CoA carboxylase from both cells in a sigmoidal manner and citrate counteracted the fructose-1,6-Pz activation over the range of their physiological concentrations.
Citrate alone, however, did not affect the activity. Our preliminary experiments showed that fructose-1,6-P* would interact with the enzyme protein at three sites and make a conformational change affecting K, and V max rather than causing a macromolecular change. Fructose-1,6-Pz decreased K, and increased V,,,,, for ATP and acetyl-CoA, while citrate decreased V,,, increased by fructose-1,6-Pi. It was demonstrated by Plate and co-workers (26) that the inhibition of fatty acid synthetase by malonyl-CoA could be reversed by fructose-1,6-P*, lowering the K, for NADPH which had been raised by malonyl-CoA.
In summary, the following conclusions may be drawn for these studies. The accumulation of neutral lipids in the inositol-deficient yeast results, at least partly, from an enhanced acetyl-CoA carboxylase activity caused by low molecular effecters in the cell rather than by an increase in enzyme protein. The greater rate of activation of the enzyme by the higher level of fructose-1,6-P* and the smaller rate of counteraction by the lower level of citrate, namely a dual control by fructose-1,6-Pz and citrate, result in a marked elevation of acetyl-CoA carboxylase activity in the deficient cells.
We can not assess at this time what causes the fluctuation of these intermediates in the deficient cells, that is, what is the primary effect of the inositol deficiency. To approach this problem, the activities of the related enzymes was compared between the supplemented and deficient cells. Our result suggests a significant increase in the activities of phosphofructokinase and citrate lyase, and a decrease in that of aldolase due to the inositol deficiency. This focus on fructose-1,6-Pz seems to be consistent with the observation of Paltauf and Johnston (15) that when ethanol, pyruvate, or lactate was used for the carbon source, no lipid accumulation occurred in the deficient cells. The high activity of phosphofructokinase in the deficient cells may secondarily result from the lower citrate level. On the other hand, the report by Sullivan and Debusk (52) seems to be applicable to the decreased activity of aldolase observed in the deficient cells. They observed that glutamate dehydrogenase (L-glutamate:NADP+ oxidoreductase EC 1.4.1.4) and glucose-6-P dehydrogenase (n-glucose-6-P:NADP+ oxidoreductase EC 1.1.1.49) in the 100,000 x g supernatant fraction from an inositol-requiring mutant strain of N. crassa (89601) grown in an inositol-limited medium were more heatlabile than those from the yeast grown in a complete medium. They ascribed the decreased heat stability in the two enzymes to the leakage of proteolytic enzymes into the cytoplasm from protease particles, whose single layer membranes are much richer in phosphatidylinositol than are other cytoplasmic membranes. According to Matile (72), about 75% of the protease activity which would otherwise be in protease particles, is found in soluble fraction of the cells when grown at a suboptimal concentration of inositol. Assuming that the released protease affects specific enzyme(s) at the earlier stage of the inositol deficiency, there will be a subsequent change in the levels of the metabolites. This concept might be plausible if the short life span and accumulation of lipid granules in S. carlsbergensis grown without inositol, are considered as part of the aging mechanism suggested by Hochschild (73) which involves the leakage of hydrolytic enzymes into the cytoplasm.
Aclznowledgments-We gratefully appreciate the excellent technical assistance of MS K. Kinae, T. Ito, M. Fukui, M. Eto, and N. Shiratori.