- Email: [email protected]
Selective loss of mitochondrial genome can be caused by certain unsaturated fatty acids
ARCHIVES
OF BIOCHEMISTRY
AND BIOPHYSICS
Vol. 224, No. 1, July 1, pp. 342-350, 1983
Selective Loss of Mitochondrial Genome Can Be Caused by Certain Unsaturated Fatty Acids GUSTAV Department
GRAFF,’ of Biological
ROGER W. SACKS, AND WILLIAM Chemistry,
Received April
University
of Illinois
E. M. LANDS
at Chicago, Chicago, Illinois
27, 1982, and in revised form February
60612
14, 1983
Various unsaturated fatty acids had different effectiveness for maintaining the continued replication of functional mitochondria in an unsaturated fatty acid auxotroph of Saccharomyces cerevisiae (KD115). Certain isomers of octadecenoic acid (i.e., cis-9) and eicosatrienoic acid (i.e.,&-8,11,14) permitted continued replication of mitochondria and provided cultures that contained only 4 to 5% cells that formed petite colonies. On the other hand, cultures grown with cis-12- or cis-l&octadecenoic acid or cis-11,14,17eicosatrienoic acid, produced a 12- to 16-fold greater frequency of petite mutants (5060%) after 8 to 10 generations of growth. The production of the petite mutants occurred despite adequate incorporation of these unsaturated fatty acids into cellular phospholipids and an apparently normal ability to undergo the initial steps in the induction of cellular respiration. The evidence suggests that some cellular processes necessary for continued mitochondrial replication depend on the structural features of the fatty acyl chains as well as the overall content of unsaturated fatty acids in membrane phospholipids. Impairment of that process by certain inadequate fatty acids or by an inadequate supply of a suitable fatty acid leads to a permanent loss of the mitochondrial genome from the cells of subsequent generations.
During studies of the effects of unsaturated fatty acids on Saccharornyces cerevisiae (KD115) (1, 2), we noted an occasional appearance of petite colonies from cultures that had been exposed to different fatty acids. This phenomenon appeared to be similar to that observed under conditims of unsaturated fatty acid deficiency (3). Nevertheless, we had provided similar amounts of the various unsaturated fatty acids to all of the cultures in our experiments and had not expected to see a selective loss of mitochondrial function. Apparently some fatty acid isomers did not meet the requirements of the cells that were adequately met by other acids. Therefore, we examined further the concept that selective structural features of the acyl chains may limit cellular physi‘Author dressed.
to whom correspondence
0003-9861/83 $3.00 Copyright All rights
0 1983 by Academic Press, Inc. of reproduction in any form reserved.
ologic processes. This report describes a highly selective effect of certain acids in causing a loss of the mitochondrial genome in situations where the overall supply of unsaturated acid seemed adequate. MATERIAL
AND
METHODS
Reagents. Yeast extract, yeast nitrogen base, Bactopeptone, and agar were obtained from Difco Company; reagent-grade dextrose was a product of Matheson, Coleman, and Bell. Glycerol Reagent grade, and Tween-80 were obtained from Fisher Scientific Company; Tween-40 was a product of Sigma Chemical Company. Fatty acids and methyl esters of fatty acids were obtained from either the Hormel Institute or from Nucheck Prep Laboratory at the highest purity available. Boron trifluoride (in methanol) was obtained from Applied Science Laboratories. Organisms and media. The mutant used in this study, Saccharomyces cerevisiae KD115 (ole l), was a generous gift of Dr. Alec Keith, Department of Biophysics, Pennsylvania State University. The cells are unable to desaturate fatty acids due to a deficiency
should be ad-
342
SELECTIVE
LOSS OF MITOCHONDRIAL
in the A9 desaturase (4) and they require exogenous unsaturated fatty acid for growth. The cultures were grown in synthetic liquid medium consisting of 1.34% (w/v) yeast nitrogen base and 5% (w/v) dextrose (YNBD)’ (1). Liquid medium containing glycerol as the principal carbon source consisted of 1.34% (w/v) yeast nitrogen base, 0.2% (w/v) yeast extract, 0.3% (w/v) dextrose, and 2.0% (w/v) glycerol (YNBG). The solid medium on which the strain was maintained was 1% (w/v) yeast extract, 2% (w/v) Bacto-peptone, 2% (w/v) dextrose, 1.5% (w/v) agar, and 1% (v/v) Tween-80 (1). Screening for revertants and the petite phenotype was accomplished by replicating colonies grown on Tween-80/dextrose agar plates onto 0.2% (v/v) Tween-401dextrose plates and 1% (w/v) Tween80 agar plates containing 3% (w/v) glycerol and 0.25% dextrose, respectively. Growth rneaswement and umditirms for petite phenotype mutation. Validated grande phenotype colonies of KD115 were transferred into liquid medium (YNBG with only 0.1% dextrose) containing 100 PM ammonium oleate and grown for 17 h to mid-log phase. Ceils were collected by centrifugation, washed with YNBG medium, and resuspended in fresh YNBG medium. Aliquots of this suspension were transferred into tubes (10.0 ml YNBG) containing ammonium salts of unsaturated fatty acids as described in Fig. 1. The ammonium salts used were prepared as described elsewhere (5). The initial cell density of these cuitures was 5-10 X lo6 cells/ml. About 70% of the cells in these inoculations were capable of forming colonies. After 12 h of growth, cells were isolated from each tube by centrifugation and resuspended with fresh YNBG medium. An aliquot of each suspension was transferred into separate fresh culture medium (10 ml YNBG) containing the same unsaturated fatty acid used in the initial culture tube and grown for another period of 12 h. Continued growth is evident from the number of generations noted on the abscissa. The process described above was repeated twice more, giving a total exposure of cells to the unsaturated fatty acid of 48 h. At every transfer of cells into fresh culture medium, aiiquots were removed for
* Abbreviations used: YNBD, yeast nitrogen basedextrose;YNBG, yeast nitrogen base-yeastextractdextrose-glycerol; &s-9-16:1, cis-9-hexadecenoic acid (palmitoleic acid); c&-9-18:1, cis-9-octadecenoic acid (oleic acid); ci.s-12-181, cis-12-octadecenoic acid; c&--13-18:1, cis-13-octadecenoic acid, cis-9,12-18:2, ci.s-9,12-octadecadienoic acid, cis-11,14-20:2, c&11,14eicosadienoic acid; ci.s-8,11,14-20~3, cis-8,11,14-eicosatrienoic acid; tis-11,14,1’7-20:3, cis-11,14,17-eicosatrienoic acid; c&-13-22:1, ci.s-13-docosenoic acid; cis13,16,19-223, &-13,16,19-docosatrienoic acid; cis4,7,10,13,16,19-22:6, cis-4,7,10,13,16,19-docosahexaenoic acid.
GENOME
343
the determination of ceil viability and petite phenotype. The residual portion of the cell suspension obtained from each cell transfer was used for analysis of cell lipid. Experimental growth cultures were shaken at 30°C at 280 rpm on a rotary shaker in tubes (18 X 150 mm) inclined at 50”. Cell densities of these cultures were determined both by the absorbancies at 660 nm and by microscopic cell counts. Conversion of Aw to cells/ ml was achieved by the computer-fitted polynomial equation of the values obtained: million cells/ml = -0.015 + 36.60 (A& - 13.42 (AM)’ + 39.62 (Aw)3 giving an average of 63 f 1 million cells per Asso. Quantitation ofpetite colony fming cells. Cell cultures grown in liquid medium containing glycerol as carbon source (YNBG) were diluted in 0.15 M sodium chloride. Aliquots of this solution were plated onto Tween-80/oleate/dextrose agar plates yielding upon incubation at 30°C for 2 to 3 days between 50 to 200 colonies per plate. Evaluation of colonies with petite phenotype was initially done by visual inspection. Petite colonies were considerably smaller (0.10 mm) than the grand phenotype colonies (0.25 mm). Confirmation of the grande and petite phenotypes was obtained by replica plating the colonies grown on Tween-801dextrose agar plates onto Tween-80/ glycerol agar plates. Petite phenotype colonies failed to grow on Tween-BO/glyceroi agar plates, whereas growth of the respiratory-competent grande colonies was supported by this medium. Induction of respiration Validated grande colonies were transferred into liquid medium (YNBD) conand grown at taining 50 pM oleate or palmitoleate 30°C to mid-log phase (approximately 10 h). Cells were isolated by centrifugation, washed, and resuspended in fresh derepression medium (YNBG containing only 0.1% glucose). Aliquots of the cell suspension were transferred into derepression medium supplemented with unsaturated fatty acid as described in Fig. 2. The initial cell densities of these experimental cultures were 5-10 X lo6 cells/ml. At various time points of ceil growth, aliquots (0.5 to 1.0 ml) of the culture were removed and diluted to 3.0 ml with derepression medium. Cellular respiratory activity was measured with a YSI Model 53 biological oxygen monitor (Yellow Springs Instrument Co., Ohio). Extraction and analysis of ccU lipid Cell suspensions of KD115 were transferred into screw-cap tubes, centrifuged, and the residual medium removed with a Pasteur pipet. The cell pellet was washed twice with 2.0 ml of 0.15 M sodium chloride and the lipids extracted as described by Graff and Lands (5), using tricosanoic acid as internal standard to monitor extraction efficiency. The total lipid extract was evaporated to dryness in a stream of nitrogen, redissolved in 1.0 ml of chloroform/methanol (l:l), and filtered over a glass-wool plug. The glass wool was washed
344
GRAFF,
80.
SACKS,
AND
LANDS
8
A
l
60.
a l
l m J,, 4
8
12
4 8 12 Number of Generations
4
8
12
FIG. 1. Petite formation following exposure to various unsaturated fatty acids. Cultures were grown at 30°C in YNBG medium plus 0.3% glucose as described under Materials and Methods. The extent of growth is indicated by the number of generations obtained. The cultures were supplemented with: (A) 5 @M (0), 10 @d (O), 40 pM (V), or 60 MM c&r-9-18:1 (v); (B) 40 PM &s-12-18:1 only (0) or plus 5 PM c&9-18:1 (0); (C) 40 I.IM cis-13-l&1 plus 3 pM &s-9,12-18:2 (0) or plus 5 pM c&9-18:1 0; (D) 40 PM cti-11,14-202; (E) 40 ard eis-11,14,17-20~3; (F) 40 PM cis-8,11,14-20:3; (G) 40 PM &s-13-2011 plus 5 MM c&9-18:1; (H) 40 UM c&13,16,19-22:3 plus 5 /.tM c&9-18:1; (I) 40 @d cis4,7,10,13,16,19-22:6.
with 2.0 ml of chloroform/methanol (l/l). After concentration, the lipids were applied onto a silica gel H plate (0.25-mm thickness). The thin-layer plate was developed with petroleum etherjdiethyl ether/acetic acid [50/50/l] (5), and the phospholipid fraction scraped into a screw-cap tube. Boron trifluoride (3% in methanol, 2 ml) was added, followed by a known quantity of pentadecanoic acid. The mixture was heated for 45 min at 70°C cooled, methyl heptadecanoate was added to monitor the extraction yield, and the methyl esters were extracted with petroleum ether. Prior to evaporation of the petroleum ether extract to dryness in a stream of nitrogen, a known quantity of methyl undecanoate and either methyl heneicosanoate or methyl nonadecancoate were added to monitor the loss of short-chain methyl esters upon evaporation and the overall recovery. The methyl ester residue was then dissolved in approximately 25 pl of distilled CSe, and an aliquot analyzed by gas-liquid
chromatography. The methyl esters were separated on a Varian Aerograph Model 2700 gas-liquid chromatograph equipped with a 6-ft X 0.2-mm glass column packed with 10% diethylene glycol succinate. The column was maintained at 18O”C, the injection port at 23O”C, and the detector at 250°C. The carrier gas was nitrogen and the flow rate 25 ml/mm. The column effluent was monitored with a flame ionization detector. Quantitation of the fatty acids contained in each sample was achieved by comparison of the peak areas relative to an internal standard of methyl pentadecanoate with the aid of a Hewlett-Packard Model 3371-B digital integrator. The fatty acid methyl esters were identified by comparison of their equivalent chain lengths with those of standards or published values (6). Isotope pulse labeling. KD115 (grande) cells were grown in repression medium and subsequently trans-
SELECTIVE
LOSS OF MITOCHONDRIAL
8
16
8
16
345
GENOME
8
16
HOURS
FIG. 2. Effect of various unsaturated fatty acids on the induction of cell respiration. Cultures were preinoculated in YNBD-medium supplemented with 56 PM c&s-9-hexadecenoic acid and grown as described under Materials and Methods. Aliquots of washed cell suspensions were transferred into YNBG medium (see Methods) containing (a) no fatty acid (0) or eis-9octadecenoic acid (a), (b) eis-9,12-octadecadienoic acid, (c) cis-11,14,17-eicosatrienoic acid (Cl), or or&-8,11,14-eicosatrienoic acid (O), (d) c&+11,14-eicosadienoic acid, (e) cis-13,16,19-docosatrienoic acid, and (f) 100 prd eis4,‘7,10,13,16,19-docosahexaenoic acid. At various times aliquots were removed from cultures and cellular respiratory activity determined as described under Materials and Methods. Respiratory activities of cell cultures containing no fatty acid are shown in (a) (0) or are indicated by a dashed line (b-f).
ferred into derepression medium containing o&9,1218:2, cis-8,11,14-20~3, or cis-11,14,17-20:3 as fatty acid supplements. At the time of transfer or after 4 or 9 h of growth, additions of [l-i4C]18:2 (sp act 2500 cpm/ nmol) were made and the incubation continued for 40 min. After the IO-min period, the pulse-labeled cells were isolated, washed, and lipids extracted as described above. Aliquots of the total lipid extract were applied to silica gel H plates (thickness 0.25 mm) that had been slurried in 0.25% EDTA (pH 7.0). The thin-layer plate was developed initially with petroleum ether/diethyl ether/acetic acid [15/85/l] to a height of 19 cm. The solvent was then allowed to evaporate from the plate by placing it into an oven at 50-60°C for about 5 min. The thin-layer plate was then scored with a line 13 cm above the point of application and subsequently developed with chloroform/acetone/methanol/acetic acid/water (50/15/30/ 10/S). The approximate R, values were triglyceride (0.95), free fatty acid (0X9), diglyceride and monoglyceride (0.81), phosphatidic acid and cardiolipin (0.66), phosphatidylethanolamine (0.53), phosphatidylserine (0.46), phosphatidylinositol (0.40), and phosphatidylcholine (0.35). The chromatogram was scanned with a Berthold
radioisotope TLC scanner and the complete chromatogram was divided into appropriate fractions containing the major lipids and lipid-free bands. The fractions were scraped into liquid scintillation vials and counted as described previously (5). RESULTS
A balance was selected in the culture conditions employed to accommodate two opposing purposes: (1) to maintain cultures in a state where mitochondrial functions would be derepressed (when mitochondrial transcription, translation, and oxidative phosphorylation is fully functional) by using glycerol as the major carbon energy source, and (2) to minimize the selective advantage of the grande forms over the petite mutants which cannot utilize glycerol as energy source by providing some glucose in the medium. Without some glucose present, there would be an undesirable selection advantage for the grande
346
GRAFF,
SACKS,
cells and a relative suppression of the petites that form in the culture. Cell cultures were transferred into fresh medium every 12 h to provide adequate unsaturated fatty acid and to aid further the survival of the petite mutants formed. When yeast cultures were grown using a limiting concentration (5 PM) of oleate (c&-9-18:1) (Fig. lA), a progressive accumulation of petite mutants was observed (25% after approximately six generations of growth). This result agrees with the earlier findings of Marzuki et al. (3), who emphasized the need for sufficient amounts of unsaturated acids to continue DNA replication. When the cis9-18:l concentration in the medium was increased to 10 and 40 pM, the accumulation of petite mutants in these cultures after 12 to 14 generations of growth decreased to 20 and 7%, respectively. With 40 or 60 PM concentrations of &-g-18:1, the level of petites never exceeded 7% of the total cell population regardless of the number of generations examined. The appearance of cells that form petite colonies during the growth of cultures exposed to two positional isomers of octadecenoic acid (cis-12 and cis-13) is shown in Figs. 1B and C. Supplementation of cell cultures with cis-12-18:l resulted in a continuous increase in petite colony mutants, reaching a level of about 50% after 10 generations of growth even though a 40 j&M concentration of the unsaturated fatty acid was provided. Inclusion of 5 PM c&9-18:1 to a similar experiment appeared not to influence the frequency of petite colonies observed. Because it had been reported earlier that &s-13-18:1 alone failed to support any appreciable cell growth (l), an acid that supported growth (either 3 PM c&9,12-18:2 or 5 PM c&9-18:1) was included when cis-13-18:l was tested for its ability to transform cells into the petite mutant. These mixtures of fatty acids allowed growth for 8-11 generations, and the cultures continuously accumulated petite mutants to a level of approximately 60% of the total cell population after 8 generations of growth. Some 20-carbon unsaturated fatty acids had a similar effect. Growth of cell cultures with 40 PM &s-11,1420:2 or cis-11,14,17-20:3 resulted in similar
AND
LANDS
accumulations of petite mutants, reaching a level of 50-55% after 12 generations (Figs. 1D and E). In contrast, the cis8,11,14-20:3 isomer produced only a low level of petite mutants (about 4%) during 15 generations of growth (Fig. 1F). Cultures failed to produce any appreciable cell growth with 22-carbon unsaturated fatty acids on glycerol medium without oleate added (unpublished data). Therefore, cultures grown with c&13-22:1 or c&-13,16,19-22:3 were supplemented with 5 PM c&-9-18:1. Under these conditions, both long-chain acids induced similar levels of petite formation, reaching 30% after 8 generations of growth (Figs. lG, H, and I). Docosahexaenoic acid (cis-4, 7,10,13,16,19-22:6) appears to have an even greater tendency to cause petite cells, reaching a value of approximately 50% petite mutants after 10 generations, (Fig. 11). We considered that petite generation may have occurred because the nutrient acid was inadequately esterified into cellular membrane lipids. A comparison between the mole percentage unsaturated fatty acid content in the total membrane phospholipids and the degree of petite formation by cells exposed to the various unsaturated fatty acids (or mixtures of acids) is presented in Table I. After 7.5 generations of growth, cis-8,11,14-20:3, was esterified into phospholipids at a level of 22 mol% and the culture contained a similar frequency of petite mutation (5%) as did the culture with &s-9-18:1, which was present at 39 mol% of the fatty acids in the phospholipid. In contrast, cell cultures exposed to cis-12-18:1, c&13-18:1, &s-11,1420:2, and cis-11,14,17-20:3 contained as much or more unsaturated fatty acid in the membrane phospholipids (73, 34, 30, and 36 mol%, respectively) and they had developed 34 to 43% petites. Thus, the data indicate that there is no simple relationship between the extent of incorporation of an unsaturated fatty acid in cellular phospholipids and the formation of petite mutants. Extensive incorporation of the nutrient unsaturated fatty acids into cellular phospholipids seemed not to prevent petite formation. There was a similar content of &s-9-18:1 in the cellular phospho-
SELECTIVE
LOSS OF MITOCHONDRIAL TABLE
347
GENOME
I
EFFECTS OF DIFFERENT NUTRIENT FATTY ACIDS UPON PETITE FORMATION AND CELLULAR PHOSPHOLIPIDS Nutrient fatty acid in phospholipid
Nutrient
fatty acid
A 18:l c9 (40 PM) B 18:l cl2 (40 PM) c 18:1 C13 (40 jrM) 18:2 c9,12 (3 phf) D 20:2 c11,14 (40 /.iM) E 20:3 c11,14,17 (40 PM) F 20:3 c8,11,14 (40 PM) G 22:l cl3 (40 PM) 18:l C!9(5 PM) H 22:3 c13,16,19 (40 PM) 18:l c9 (5 pw) I 226 c4,7,10,13,16,19 (40 /JM)
Number of generations
% Petite of total viable cells
7.5 6.9 4.8
5 34 43
6.3 6.9 7.5 7.1
37 43 5 30
4.4
27
4.9
28
mol %
amol/ cell
39 73 29 4.8 30 36 22 0.3 13.7 0.3 13.4 6
199 357 463 76 122 132 116 4 181 3 148 37
nmol/ml culture 16 6 16 2.6 4 7 11 0.1 3.7 0.1 2.9 0.8
PL Total amol/cell 508 491 1500 403 370 541 1320 1102 587
Note. Cell cultures were grown at 30°C in YNBG medium (plus 0.3% dextrose) as described under Materials and Methods and supplemented with unsaturated fatty acid as indicated. Cells were isolated after 24 (A, CF, H, I) or 36 h (B and G), washed, and the cellular lipids extracted and analyzed.
lipids of cells treated simultaneously with 5 PM cis-g-18:1 and either &s-13-22:1 or cis13,16,19-22:3 (181 and 148 amol/cell, respectively; see Table I, experiments G and H). Although this content was similar to that in cells exposed to 40 PM &s-9-18:1 alone (199 amol/cell), it was apparently unable to prevent formation of petite cells when the 22-carbon acid was also present. Cells exposed to the 22-carbon unsaturated fatty acids had greater amounts of total phospholipid per cell (1320 and 1102 amol/cell) and showed a very low percentage content of 22-carbon unsaturated fatty acids in phospholipid (0.3 to 6 mol%). Measurements of the induction of cell respiration of cells transferred from repression medium into derepression medium (Fig. 2) indicated that cells supplemented with &s-11,14-20:2, cis-11,14,1720:3, and cis-12-18:l developed increased respiratory activities, from 0.3 to about 1.0 nmol Oz/min/106 cells, that were similar to those observed for cells supplemented with cis-g-18:1 and cis-8,11,14-20:3. In contrast, cells that were supplemented with &s-13,16,19-22:3 and cis-4,7,10,13,16,19-22:6 appeared unable to permit the full induction of cellular respiration. This lower
ability to support the induction of respiration was observed long before the increased appearance of petite mutants, and was correlated with the lower mole percentage of the nutrient acid incorporated into the cellular lipids (see Table I; 22:3<22:6<20:2). Upon derepression, no differences could be observed in the esterification of added [l-14C]18:2 into the triglycerides and various phospholipid fractions (Table II) of cells supplemented with two different isomers of 20:3. The pattern of linoleate incorporation among the lipids resembled that for cells grown only with linoleate and did not indicate any alteration in lipid formation. The cis-11,14,17-isomer of 20:3 caused many petite cells to form (Fig. l), although it allowed incorporation of [l14C118.2 . into cell lipids to a level similar to that observed for cells supplemented with cis-8,11,14-20:3 which caused no increase in petites above control values. DISCUSSION
Mitochondrial morphogenesis requires protein synthesis on both mitochondrial and cytoplasmic ribosomes. Although mi-
348
GRAFF,
SACKS,
AND
TABLE INCORPORATION
Nutrient [l-‘%]18:2
fatty
acid
OF [l-%]18:2
LANDS
II
INTO CELLULAR LIPID OF S. cerewisiae KD115 (GRANDE) AFTER TRANSFER INTO DEPRESSION MEDIUM
&s-9,12-18:2
eis-8,11,14-2Oz3 (40 @hi)
(20 /.lM)
AT VARIOUS
TIMES
cis-11,14,17-2Oz3 (40 CM)
pulse c0nc
[PMI Time of pulse addition Million cells/ml culture
22.8 0 + 0:40 8.4
11.0 4 - 4:40 18
11.1 9 --+ 940
23.1 0 -+ 0:40
42.8
8.5
[l-“C]18:2
nmol/ml
11.6 4 -4:40 14.1
11.6 9 --+ 940
22.4 0 -0:40
11.2 4 - 4:40
11.2 9 - 940
36.2
8.5
15.9
40.1
culture
TG FA DGiMG
3.8 1.3 0.2
2.6 0.3 0.5
0.6 0.4 0.5
0.5 0.7 0.2
1.5 0.5 0.2
0.6 0.4 0.3
0.8 0.6 0.3
0.9 0.3 0.1
0.9 0.3 0.2
PA + CL PE PS
0.1 0.6 0.3
0.5 0.9 0.5
0.2 1.0 0.8
0.1 0.3 0.2
0.1 0.8 0.5
0.2 1.2 1.3
0.1 0.4 0.3
0.1 0.8 0.6
0.3 1.4 1.3
Pl PC Origin
0.8 3.3 0.6
0.5 4.2 0.3
1.8 5.3 0.3
0.6 1.4 0.1
0.8 2.0 0.1
2.4 3.7 0.2
0.5 1.3 0.1
0.9 1.8 0.1
1.8 3.9 0.2
Note. Cells grown in repression medium were transferred into derepression medium (10.0 ml) containing the fatty acids indicated and grown at 30°C. Additions of [l-“C]18:2 were made at time points indicated. After 40 min, the lipids were extracted from washed cells as described under Materials and Methods.
tochondria contain an independent transcription and protein-synthesizing system, the cytoplasmic protein synthetic system provides components essential for mitochondrial morphogenesis in a highly coordinated and mutually interdependent process (7). For example, the m&DNA-dependent RNA polymerase (8), M-DNA polymerase (9), mitochondrial ribosomal proteins, and initiation and elongation factors are coded for by nuclear genes (10, 11) and they are synthesized on cytoplasmic ribosomes (11, 12, 13). Walenga and Lands (2) provided evidence that at least two molecular events during morphogenesis of mitochondria require unsaturated fatty acids for cells to gain respiratory competence. The early event, associated with a priming process, required a high preexisting content of unsaturated fatty acid in phospholipid to permit the synthesis of components on cytoplasmic ribosomes, but did not require a concurrent supply of unsaturated acids. The second event, promoting the induction of respiration, required the concurrent presence of unsaturated fatty acids. cis-13-
Docosenoic acid failed to support an adequate priming process. Furthermore, it did not support the induction of respiration in cells with a high preexisting unsaturated fatty acid content in membrane phospholipids. The present experiments examine the ability of different unsaturated fatty acids to maintain adequate replication and retention of the mitochondrial genome during repeated cell divisions. Failure to maintain an adequate amount of unsaturated fatty acid in the membranes of 5’. cere&siae has been reported to cause uncoupling of oxidative phosphorylation (2, 14, 15), impairment of mitochondrial ATP translocation (16), lowering of the adenylate energy charge (1’7), as well as depletion of 15 and 21 S mitochondrial ribosomal RNA (18, 19) and lack of formation of mitochondrial cytochromes (2,18,20). The observations were insufficient to indicate whether the minimal content of unsaturated fatty acid in phospholipid required for the maintenance of adequate mitochondrial energy transduction corresponds to that needed to maintain general cellular protein syn-
SELECTIVE
LOSS OF MITOCHONDRIAL
thesis. Conceivably there could be a threshold value for unsaturated acids to support energy transduction that differs from that needed for protein synthesis. We visualize yet another threshold that may represent the amount needed to prevent irreversible loss of m&DNA concurrent with the generation of petite cells (3). Some acids which caused a high percentage of petite cells in the present study still permit,ted the nonpetite KD115 cells to develop cellular respiration during derepression with no detectable impairment (Fig. 2) and also allowed unimpaired esterification of linoleate into membrane phospholipids (Table II). This result is consistent with the high content of the nutrient unsaturated fatty acid that occurred in the membrane phospholipids, more than that required for successful priming as reported by Walenga and Lands (2). Although the fatty acids examined showed some differences in their ability to support the subsequent induction of respiration, the process of induction of respiration appeared not to trigger an immediate discontinuation of m&DNA replication and onset of petite generation as long as an adequate level of unsaturated membrane lipids was present. The data shown in Fig. 2 suggest that most nutrient fatty acids were adequate to support the metabolic processes required for the induction of cell respiration and that the transcription and translation of the mitochondrial genome was still adequate during the first 10 h of induction. The loss of the mitochondrial genome in our experiments may have occurred by one of two ways: inadequate unsaturation or selective acyl chain action. Cells that had initially a high content of adequate fatty acyl chains in their membrane lipids may have (a) di.luted these acyl chains with newly syntlhesized saturated fatty acids, or (b) developed insufficient amounts of unsaturated acyl chains due to a reduced rate of esterification which is regulated by the activities and specificities of cellular acyltransferases. Mitochondrial proteins synthesized on cytoplasmic ribosomes have been suggested to require phospholipids for insertion into the mitochondrial mem-
GENOME
349
brane (1, 21, 22). Perhaps a reduced rate of synthesis of lipids containing unsaturated fatty acid may not provide adequate lipids for complexing with cytoplasmically synthesized proteins. The inability to insert proteins bound to appropriately unsaturated lipids into the mitochondrial membrane may lead to the loss of &-DNA replication and eventual loss of the cellular respiratory competence (3). Our results confirm that such conditions may develop when cells are not provided with sufficient quantities of nutrient unsaturated acids and also occur when the acids provided are incorporated poorly into the needed lipids (1, 23). Alternatively, as exemplified with two octadecenoic acid isomers (cis-9 and c&12-18:1) or with the two isomers of eicosatrienoic acid, some cellular processes required for continued mitochondrial morphogenesis may somehow recognize specific structural features of acyl chains in membrane phospholipids. These specific recognition processes appear to be independent of the mole percentage content of unsaturated fatty acid in membrane phospholipids. In each example noted above, the acid causing petite formation was present at a higher mole percentage content in membrane than was the “normal” fatty acid. Our data suggest that some selective effect of the fatty acid can occur beyond a simple inability to be esterified into cellular phospholipids. Precedence for a hypothesis of selective cellular response to acyl chain structure has been obtained in other investigations from this laboratory (24-26) with a mutant of Esherichia coli that requires unsaturated fatty acids (27). A number of positional and geometric isomers showed poor growth-supporting properties for this organism even though their incorporation into cellular lipids was similar to that of effective nutrient acids. The inability to support growth seemed not readily explained by differences in melting points of these fatty acids or their inability to provide adequate fluidities. These positional isomers became as effective as others to support growth upon addition of CAMP to the medium or by conducting growth with glycerol as the carbon source (25, 26). In
350
GRAFF,
SACKS,
the presence of sufficient CAMP, growth with all positional isomers between 5 and 14 reflected a predictable response related to fluidity. In a similar manner, selected trans-fatty acid isomers which are much less effective in supporting growth of E. coli than their cis-isomeric counterparts (26) readily supported cell growth when CAMP was included in the medium or when glycerol was the carbon source (28). Although added CAMP did not alter responses of 5’. cerevisiae, the results in this report indicate that there are cellular regulatory events which can respond to detailed structural differences in the unsaturated fatty acyl chains. The fact that some mutant fatty acids caused large increases in total phospholipid per cell (see also, Refs. (5,29)) confirms the concept that an important regulatory process in S. cerevisiae may be altered by specific fatty acids. Since loss of mitochondria occurred also in cells not showing increased phospholipid contents, we expect to find a variety of pleiotropic events associated with the influence of acyl chain structure upon cellular regulatory controls. ACKNOWLEDGMENT Research for this report was supported by a grant from the United States Public Health Service (AM05310). REFERENCES 1. WALENGA, R. W., AND LANDS, W. E. M. (1975) J. Biol. Chem 250,9121-9129. 2. WALENGA, R. W., AND LANDS, W. E. M. (1975) J. Biol Chem. 250,9130-9136. 3. MARZUKI, S., HALL, R. M., AND LINNANE, A. W. (1974) Biochem Biophys. Res. Cornmun 57.372378. 4. KEITH, A. D., RESNICK, M. R., AND HALEY, A. B. (1969) J. BacterioL 98.415-420. 5. GRAFF, G., AND LANDS, W. E. M. (1976) Chem. Phys. Lipids 17.301-314. 6. HOFSTETTER, H. H., SEN, H., AND HOLMAN, R. T. (1965) J. Amer. oil Chem. Sot 42, 537-540. 7. SCHATZ, G., AND MASON, T. L. (1974) Annu. Rev. Biochem. 43. 51-87.
AND
LANDS
8. WINTERSBERGER, E. (1970) Biochem Biophys Rex Cemmun. 40,1179-1184. 9. CHIN, J. J., AND KALF, G. F. (1969) Arch. Biochem Biophys. 133, 38-45. 10. RICHTER, D. (1971) Biochemistry 10, 4422-4425. 11. PERISI, B., AND CELLA, R. (1971) FEBS. Lett. 14, 209-213. 12. DAVAY, P. J., Yu, R., AND LINNANE, A. W. (1969) B&hem. Biophys. Res. Commun 36,30-34. 13. KUNTZEL, H. (1969) Nature (London) 222,142-146. 14. HASLAM, J. M., PROUDLOCK, J. W., AND LINNANE, A. W. (1971) J. Bioenwget. 2, 351-370. 15. MARZUKI, S., COBON, G. S., HASLAM, J. M., AND LINNANE, A. W. (1975) Arch B&hem. Biophys. 169, 577-590. 16. HASLAM, J. M., AND FELLOWS, N. F. (1975) Biochem. Sot. Trans. 3, 772-775. 17. HASLAM, J. M., AND FELLOWS, N. F. (1977) Biochem J. 166, 565-570. 18. MARZUKI, S., COBON, G. S., CROWFOOT,P. D., AND LINNANE, A. W. (1975) Arch. Biochem. Biophys. 169, 591-600. 19. GORDON, P. H., AND STEWART, P. R. (1972) J. Gem. Micro. 72, 231-242. 20. WATSON, K., HOUGHTON, R. L., BERTOLI, E., AND GRIFFITHS, D. E. (1975) B&hem. J. 146, 409416. 21. KADENBACH, B. (1967) B&him. Biophys Acta 134, 430-442. 22. KADENBACH, B., in Biochemical Aspects of Biogenesis of Mitochondria (Slater, E. C., Tager, J. M., Papa, S., and Quagliariello, E., eds.). Adriatica Editoice, Bari, Italy. 23. GURBA, P. E. (1979) Ph.D. thesis, University of Michigan. 24. LANDS, W. E. M., OHLROGGE, J. B., ROBINSON, J. R., SACKS, R. W., BARVE, J. A., AND GUNSTONE, F. D. (1977) B&him Biophys. Ada 486, 451-461. 25. OHLROGGE, J. B., BARBER, E. D., LANDS, W. E. M., GUNSTONE, F. D., AND JSMAIL, I. A. (1976) Canad J. Biochem 54,736-745. 26. LANDS, W. E. M. (1979) Guidelines to Metabolic Therapy (Guidelines to Medical Science Series Publ. Upjohn Company), Vol. 8, pp. l-3. 27. LINDEN, C. D., WRIGHT, K. L., MCCONNELL, H. M., AND Fox, C. F. (1973) Proc Nat. Ad Sk USA 70, 2271-2275. 28. TSAO, Y. K., AND LANDS, W. E. M. (1980) Science 207, 777-779. 29. VANDENHOFF, G., GUNSTONE, F. D., BARNE, J. AND LANDS, W. E. M. (1975) J. Bid Chem 250,87208727.
OF BIOCHEMISTRY
AND BIOPHYSICS
Vol. 224, No. 1, July 1, pp. 342-350, 1983
Selective Loss of Mitochondrial Genome Can Be Caused by Certain Unsaturated Fatty Acids GUSTAV Department
GRAFF,’ of Biological
ROGER W. SACKS, AND WILLIAM Chemistry,
Received April
University
of Illinois
E. M. LANDS
at Chicago, Chicago, Illinois
27, 1982, and in revised form February
60612
14, 1983
Various unsaturated fatty acids had different effectiveness for maintaining the continued replication of functional mitochondria in an unsaturated fatty acid auxotroph of Saccharomyces cerevisiae (KD115). Certain isomers of octadecenoic acid (i.e., cis-9) and eicosatrienoic acid (i.e.,&-8,11,14) permitted continued replication of mitochondria and provided cultures that contained only 4 to 5% cells that formed petite colonies. On the other hand, cultures grown with cis-12- or cis-l&octadecenoic acid or cis-11,14,17eicosatrienoic acid, produced a 12- to 16-fold greater frequency of petite mutants (5060%) after 8 to 10 generations of growth. The production of the petite mutants occurred despite adequate incorporation of these unsaturated fatty acids into cellular phospholipids and an apparently normal ability to undergo the initial steps in the induction of cellular respiration. The evidence suggests that some cellular processes necessary for continued mitochondrial replication depend on the structural features of the fatty acyl chains as well as the overall content of unsaturated fatty acids in membrane phospholipids. Impairment of that process by certain inadequate fatty acids or by an inadequate supply of a suitable fatty acid leads to a permanent loss of the mitochondrial genome from the cells of subsequent generations.
During studies of the effects of unsaturated fatty acids on Saccharornyces cerevisiae (KD115) (1, 2), we noted an occasional appearance of petite colonies from cultures that had been exposed to different fatty acids. This phenomenon appeared to be similar to that observed under conditims of unsaturated fatty acid deficiency (3). Nevertheless, we had provided similar amounts of the various unsaturated fatty acids to all of the cultures in our experiments and had not expected to see a selective loss of mitochondrial function. Apparently some fatty acid isomers did not meet the requirements of the cells that were adequately met by other acids. Therefore, we examined further the concept that selective structural features of the acyl chains may limit cellular physi‘Author dressed.
to whom correspondence
0003-9861/83 $3.00 Copyright All rights
0 1983 by Academic Press, Inc. of reproduction in any form reserved.
ologic processes. This report describes a highly selective effect of certain acids in causing a loss of the mitochondrial genome in situations where the overall supply of unsaturated acid seemed adequate. MATERIAL
AND
METHODS
Reagents. Yeast extract, yeast nitrogen base, Bactopeptone, and agar were obtained from Difco Company; reagent-grade dextrose was a product of Matheson, Coleman, and Bell. Glycerol Reagent grade, and Tween-80 were obtained from Fisher Scientific Company; Tween-40 was a product of Sigma Chemical Company. Fatty acids and methyl esters of fatty acids were obtained from either the Hormel Institute or from Nucheck Prep Laboratory at the highest purity available. Boron trifluoride (in methanol) was obtained from Applied Science Laboratories. Organisms and media. The mutant used in this study, Saccharomyces cerevisiae KD115 (ole l), was a generous gift of Dr. Alec Keith, Department of Biophysics, Pennsylvania State University. The cells are unable to desaturate fatty acids due to a deficiency
should be ad-
342
SELECTIVE
LOSS OF MITOCHONDRIAL
in the A9 desaturase (4) and they require exogenous unsaturated fatty acid for growth. The cultures were grown in synthetic liquid medium consisting of 1.34% (w/v) yeast nitrogen base and 5% (w/v) dextrose (YNBD)’ (1). Liquid medium containing glycerol as the principal carbon source consisted of 1.34% (w/v) yeast nitrogen base, 0.2% (w/v) yeast extract, 0.3% (w/v) dextrose, and 2.0% (w/v) glycerol (YNBG). The solid medium on which the strain was maintained was 1% (w/v) yeast extract, 2% (w/v) Bacto-peptone, 2% (w/v) dextrose, 1.5% (w/v) agar, and 1% (v/v) Tween-80 (1). Screening for revertants and the petite phenotype was accomplished by replicating colonies grown on Tween-80/dextrose agar plates onto 0.2% (v/v) Tween-401dextrose plates and 1% (w/v) Tween80 agar plates containing 3% (w/v) glycerol and 0.25% dextrose, respectively. Growth rneaswement and umditirms for petite phenotype mutation. Validated grande phenotype colonies of KD115 were transferred into liquid medium (YNBG with only 0.1% dextrose) containing 100 PM ammonium oleate and grown for 17 h to mid-log phase. Ceils were collected by centrifugation, washed with YNBG medium, and resuspended in fresh YNBG medium. Aliquots of this suspension were transferred into tubes (10.0 ml YNBG) containing ammonium salts of unsaturated fatty acids as described in Fig. 1. The ammonium salts used were prepared as described elsewhere (5). The initial cell density of these cuitures was 5-10 X lo6 cells/ml. About 70% of the cells in these inoculations were capable of forming colonies. After 12 h of growth, cells were isolated from each tube by centrifugation and resuspended with fresh YNBG medium. An aliquot of each suspension was transferred into separate fresh culture medium (10 ml YNBG) containing the same unsaturated fatty acid used in the initial culture tube and grown for another period of 12 h. Continued growth is evident from the number of generations noted on the abscissa. The process described above was repeated twice more, giving a total exposure of cells to the unsaturated fatty acid of 48 h. At every transfer of cells into fresh culture medium, aiiquots were removed for
* Abbreviations used: YNBD, yeast nitrogen basedextrose;YNBG, yeast nitrogen base-yeastextractdextrose-glycerol; &s-9-16:1, cis-9-hexadecenoic acid (palmitoleic acid); c&-9-18:1, cis-9-octadecenoic acid (oleic acid); ci.s-12-181, cis-12-octadecenoic acid; c&--13-18:1, cis-13-octadecenoic acid, cis-9,12-18:2, ci.s-9,12-octadecadienoic acid, cis-11,14-20:2, c&11,14eicosadienoic acid; ci.s-8,11,14-20~3, cis-8,11,14-eicosatrienoic acid; tis-11,14,1’7-20:3, cis-11,14,17-eicosatrienoic acid; c&-13-22:1, ci.s-13-docosenoic acid; cis13,16,19-223, &-13,16,19-docosatrienoic acid; cis4,7,10,13,16,19-22:6, cis-4,7,10,13,16,19-docosahexaenoic acid.
GENOME
343
the determination of ceil viability and petite phenotype. The residual portion of the cell suspension obtained from each cell transfer was used for analysis of cell lipid. Experimental growth cultures were shaken at 30°C at 280 rpm on a rotary shaker in tubes (18 X 150 mm) inclined at 50”. Cell densities of these cultures were determined both by the absorbancies at 660 nm and by microscopic cell counts. Conversion of Aw to cells/ ml was achieved by the computer-fitted polynomial equation of the values obtained: million cells/ml = -0.015 + 36.60 (A& - 13.42 (AM)’ + 39.62 (Aw)3 giving an average of 63 f 1 million cells per Asso. Quantitation ofpetite colony fming cells. Cell cultures grown in liquid medium containing glycerol as carbon source (YNBG) were diluted in 0.15 M sodium chloride. Aliquots of this solution were plated onto Tween-80/oleate/dextrose agar plates yielding upon incubation at 30°C for 2 to 3 days between 50 to 200 colonies per plate. Evaluation of colonies with petite phenotype was initially done by visual inspection. Petite colonies were considerably smaller (0.10 mm) than the grand phenotype colonies (0.25 mm). Confirmation of the grande and petite phenotypes was obtained by replica plating the colonies grown on Tween-801dextrose agar plates onto Tween-80/ glycerol agar plates. Petite phenotype colonies failed to grow on Tween-BO/glyceroi agar plates, whereas growth of the respiratory-competent grande colonies was supported by this medium. Induction of respiration Validated grande colonies were transferred into liquid medium (YNBD) conand grown at taining 50 pM oleate or palmitoleate 30°C to mid-log phase (approximately 10 h). Cells were isolated by centrifugation, washed, and resuspended in fresh derepression medium (YNBG containing only 0.1% glucose). Aliquots of the cell suspension were transferred into derepression medium supplemented with unsaturated fatty acid as described in Fig. 2. The initial cell densities of these experimental cultures were 5-10 X lo6 cells/ml. At various time points of ceil growth, aliquots (0.5 to 1.0 ml) of the culture were removed and diluted to 3.0 ml with derepression medium. Cellular respiratory activity was measured with a YSI Model 53 biological oxygen monitor (Yellow Springs Instrument Co., Ohio). Extraction and analysis of ccU lipid Cell suspensions of KD115 were transferred into screw-cap tubes, centrifuged, and the residual medium removed with a Pasteur pipet. The cell pellet was washed twice with 2.0 ml of 0.15 M sodium chloride and the lipids extracted as described by Graff and Lands (5), using tricosanoic acid as internal standard to monitor extraction efficiency. The total lipid extract was evaporated to dryness in a stream of nitrogen, redissolved in 1.0 ml of chloroform/methanol (l:l), and filtered over a glass-wool plug. The glass wool was washed
344
GRAFF,
80.
SACKS,
AND
LANDS
8
A
l
60.
a l
l m J,, 4
8
12
4 8 12 Number of Generations
4
8
12
FIG. 1. Petite formation following exposure to various unsaturated fatty acids. Cultures were grown at 30°C in YNBG medium plus 0.3% glucose as described under Materials and Methods. The extent of growth is indicated by the number of generations obtained. The cultures were supplemented with: (A) 5 @M (0), 10 @d (O), 40 pM (V), or 60 MM c&r-9-18:1 (v); (B) 40 PM &s-12-18:1 only (0) or plus 5 PM c&9-18:1 (0); (C) 40 I.IM cis-13-l&1 plus 3 pM &s-9,12-18:2 (0) or plus 5 pM c&9-18:1 0; (D) 40 PM cti-11,14-202; (E) 40 ard eis-11,14,17-20~3; (F) 40 PM cis-8,11,14-20:3; (G) 40 PM &s-13-2011 plus 5 MM c&9-18:1; (H) 40 UM c&13,16,19-22:3 plus 5 /.tM c&9-18:1; (I) 40 @d cis4,7,10,13,16,19-22:6.
with 2.0 ml of chloroform/methanol (l/l). After concentration, the lipids were applied onto a silica gel H plate (0.25-mm thickness). The thin-layer plate was developed with petroleum etherjdiethyl ether/acetic acid [50/50/l] (5), and the phospholipid fraction scraped into a screw-cap tube. Boron trifluoride (3% in methanol, 2 ml) was added, followed by a known quantity of pentadecanoic acid. The mixture was heated for 45 min at 70°C cooled, methyl heptadecanoate was added to monitor the extraction yield, and the methyl esters were extracted with petroleum ether. Prior to evaporation of the petroleum ether extract to dryness in a stream of nitrogen, a known quantity of methyl undecanoate and either methyl heneicosanoate or methyl nonadecancoate were added to monitor the loss of short-chain methyl esters upon evaporation and the overall recovery. The methyl ester residue was then dissolved in approximately 25 pl of distilled CSe, and an aliquot analyzed by gas-liquid
chromatography. The methyl esters were separated on a Varian Aerograph Model 2700 gas-liquid chromatograph equipped with a 6-ft X 0.2-mm glass column packed with 10% diethylene glycol succinate. The column was maintained at 18O”C, the injection port at 23O”C, and the detector at 250°C. The carrier gas was nitrogen and the flow rate 25 ml/mm. The column effluent was monitored with a flame ionization detector. Quantitation of the fatty acids contained in each sample was achieved by comparison of the peak areas relative to an internal standard of methyl pentadecanoate with the aid of a Hewlett-Packard Model 3371-B digital integrator. The fatty acid methyl esters were identified by comparison of their equivalent chain lengths with those of standards or published values (6). Isotope pulse labeling. KD115 (grande) cells were grown in repression medium and subsequently trans-
SELECTIVE
LOSS OF MITOCHONDRIAL
8
16
8
16
345
GENOME
8
16
HOURS
FIG. 2. Effect of various unsaturated fatty acids on the induction of cell respiration. Cultures were preinoculated in YNBD-medium supplemented with 56 PM c&s-9-hexadecenoic acid and grown as described under Materials and Methods. Aliquots of washed cell suspensions were transferred into YNBG medium (see Methods) containing (a) no fatty acid (0) or eis-9octadecenoic acid (a), (b) eis-9,12-octadecadienoic acid, (c) cis-11,14,17-eicosatrienoic acid (Cl), or or&-8,11,14-eicosatrienoic acid (O), (d) c&+11,14-eicosadienoic acid, (e) cis-13,16,19-docosatrienoic acid, and (f) 100 prd eis4,‘7,10,13,16,19-docosahexaenoic acid. At various times aliquots were removed from cultures and cellular respiratory activity determined as described under Materials and Methods. Respiratory activities of cell cultures containing no fatty acid are shown in (a) (0) or are indicated by a dashed line (b-f).
ferred into derepression medium containing o&9,1218:2, cis-8,11,14-20~3, or cis-11,14,17-20:3 as fatty acid supplements. At the time of transfer or after 4 or 9 h of growth, additions of [l-i4C]18:2 (sp act 2500 cpm/ nmol) were made and the incubation continued for 40 min. After the IO-min period, the pulse-labeled cells were isolated, washed, and lipids extracted as described above. Aliquots of the total lipid extract were applied to silica gel H plates (thickness 0.25 mm) that had been slurried in 0.25% EDTA (pH 7.0). The thin-layer plate was developed initially with petroleum ether/diethyl ether/acetic acid [15/85/l] to a height of 19 cm. The solvent was then allowed to evaporate from the plate by placing it into an oven at 50-60°C for about 5 min. The thin-layer plate was then scored with a line 13 cm above the point of application and subsequently developed with chloroform/acetone/methanol/acetic acid/water (50/15/30/ 10/S). The approximate R, values were triglyceride (0.95), free fatty acid (0X9), diglyceride and monoglyceride (0.81), phosphatidic acid and cardiolipin (0.66), phosphatidylethanolamine (0.53), phosphatidylserine (0.46), phosphatidylinositol (0.40), and phosphatidylcholine (0.35). The chromatogram was scanned with a Berthold
radioisotope TLC scanner and the complete chromatogram was divided into appropriate fractions containing the major lipids and lipid-free bands. The fractions were scraped into liquid scintillation vials and counted as described previously (5). RESULTS
A balance was selected in the culture conditions employed to accommodate two opposing purposes: (1) to maintain cultures in a state where mitochondrial functions would be derepressed (when mitochondrial transcription, translation, and oxidative phosphorylation is fully functional) by using glycerol as the major carbon energy source, and (2) to minimize the selective advantage of the grande forms over the petite mutants which cannot utilize glycerol as energy source by providing some glucose in the medium. Without some glucose present, there would be an undesirable selection advantage for the grande
346
GRAFF,
SACKS,
cells and a relative suppression of the petites that form in the culture. Cell cultures were transferred into fresh medium every 12 h to provide adequate unsaturated fatty acid and to aid further the survival of the petite mutants formed. When yeast cultures were grown using a limiting concentration (5 PM) of oleate (c&-9-18:1) (Fig. lA), a progressive accumulation of petite mutants was observed (25% after approximately six generations of growth). This result agrees with the earlier findings of Marzuki et al. (3), who emphasized the need for sufficient amounts of unsaturated acids to continue DNA replication. When the cis9-18:l concentration in the medium was increased to 10 and 40 pM, the accumulation of petite mutants in these cultures after 12 to 14 generations of growth decreased to 20 and 7%, respectively. With 40 or 60 PM concentrations of &-g-18:1, the level of petites never exceeded 7% of the total cell population regardless of the number of generations examined. The appearance of cells that form petite colonies during the growth of cultures exposed to two positional isomers of octadecenoic acid (cis-12 and cis-13) is shown in Figs. 1B and C. Supplementation of cell cultures with cis-12-18:l resulted in a continuous increase in petite colony mutants, reaching a level of about 50% after 10 generations of growth even though a 40 j&M concentration of the unsaturated fatty acid was provided. Inclusion of 5 PM c&9-18:1 to a similar experiment appeared not to influence the frequency of petite colonies observed. Because it had been reported earlier that &s-13-18:1 alone failed to support any appreciable cell growth (l), an acid that supported growth (either 3 PM c&9,12-18:2 or 5 PM c&9-18:1) was included when cis-13-18:l was tested for its ability to transform cells into the petite mutant. These mixtures of fatty acids allowed growth for 8-11 generations, and the cultures continuously accumulated petite mutants to a level of approximately 60% of the total cell population after 8 generations of growth. Some 20-carbon unsaturated fatty acids had a similar effect. Growth of cell cultures with 40 PM &s-11,1420:2 or cis-11,14,17-20:3 resulted in similar
AND
LANDS
accumulations of petite mutants, reaching a level of 50-55% after 12 generations (Figs. 1D and E). In contrast, the cis8,11,14-20:3 isomer produced only a low level of petite mutants (about 4%) during 15 generations of growth (Fig. 1F). Cultures failed to produce any appreciable cell growth with 22-carbon unsaturated fatty acids on glycerol medium without oleate added (unpublished data). Therefore, cultures grown with c&13-22:1 or c&-13,16,19-22:3 were supplemented with 5 PM c&-9-18:1. Under these conditions, both long-chain acids induced similar levels of petite formation, reaching 30% after 8 generations of growth (Figs. lG, H, and I). Docosahexaenoic acid (cis-4, 7,10,13,16,19-22:6) appears to have an even greater tendency to cause petite cells, reaching a value of approximately 50% petite mutants after 10 generations, (Fig. 11). We considered that petite generation may have occurred because the nutrient acid was inadequately esterified into cellular membrane lipids. A comparison between the mole percentage unsaturated fatty acid content in the total membrane phospholipids and the degree of petite formation by cells exposed to the various unsaturated fatty acids (or mixtures of acids) is presented in Table I. After 7.5 generations of growth, cis-8,11,14-20:3, was esterified into phospholipids at a level of 22 mol% and the culture contained a similar frequency of petite mutation (5%) as did the culture with &s-9-18:1, which was present at 39 mol% of the fatty acids in the phospholipid. In contrast, cell cultures exposed to cis-12-18:1, c&13-18:1, &s-11,1420:2, and cis-11,14,17-20:3 contained as much or more unsaturated fatty acid in the membrane phospholipids (73, 34, 30, and 36 mol%, respectively) and they had developed 34 to 43% petites. Thus, the data indicate that there is no simple relationship between the extent of incorporation of an unsaturated fatty acid in cellular phospholipids and the formation of petite mutants. Extensive incorporation of the nutrient unsaturated fatty acids into cellular phospholipids seemed not to prevent petite formation. There was a similar content of &s-9-18:1 in the cellular phospho-
SELECTIVE
LOSS OF MITOCHONDRIAL TABLE
347
GENOME
I
EFFECTS OF DIFFERENT NUTRIENT FATTY ACIDS UPON PETITE FORMATION AND CELLULAR PHOSPHOLIPIDS Nutrient fatty acid in phospholipid
Nutrient
fatty acid
A 18:l c9 (40 PM) B 18:l cl2 (40 PM) c 18:1 C13 (40 jrM) 18:2 c9,12 (3 phf) D 20:2 c11,14 (40 /.iM) E 20:3 c11,14,17 (40 PM) F 20:3 c8,11,14 (40 PM) G 22:l cl3 (40 PM) 18:l C!9(5 PM) H 22:3 c13,16,19 (40 PM) 18:l c9 (5 pw) I 226 c4,7,10,13,16,19 (40 /JM)
Number of generations
% Petite of total viable cells
7.5 6.9 4.8
5 34 43
6.3 6.9 7.5 7.1
37 43 5 30
4.4
27
4.9
28
mol %
amol/ cell
39 73 29 4.8 30 36 22 0.3 13.7 0.3 13.4 6
199 357 463 76 122 132 116 4 181 3 148 37
nmol/ml culture 16 6 16 2.6 4 7 11 0.1 3.7 0.1 2.9 0.8
PL Total amol/cell 508 491 1500 403 370 541 1320 1102 587
Note. Cell cultures were grown at 30°C in YNBG medium (plus 0.3% dextrose) as described under Materials and Methods and supplemented with unsaturated fatty acid as indicated. Cells were isolated after 24 (A, CF, H, I) or 36 h (B and G), washed, and the cellular lipids extracted and analyzed.
lipids of cells treated simultaneously with 5 PM cis-g-18:1 and either &s-13-22:1 or cis13,16,19-22:3 (181 and 148 amol/cell, respectively; see Table I, experiments G and H). Although this content was similar to that in cells exposed to 40 PM &s-9-18:1 alone (199 amol/cell), it was apparently unable to prevent formation of petite cells when the 22-carbon acid was also present. Cells exposed to the 22-carbon unsaturated fatty acids had greater amounts of total phospholipid per cell (1320 and 1102 amol/cell) and showed a very low percentage content of 22-carbon unsaturated fatty acids in phospholipid (0.3 to 6 mol%). Measurements of the induction of cell respiration of cells transferred from repression medium into derepression medium (Fig. 2) indicated that cells supplemented with &s-11,14-20:2, cis-11,14,1720:3, and cis-12-18:l developed increased respiratory activities, from 0.3 to about 1.0 nmol Oz/min/106 cells, that were similar to those observed for cells supplemented with cis-g-18:1 and cis-8,11,14-20:3. In contrast, cells that were supplemented with &s-13,16,19-22:3 and cis-4,7,10,13,16,19-22:6 appeared unable to permit the full induction of cellular respiration. This lower
ability to support the induction of respiration was observed long before the increased appearance of petite mutants, and was correlated with the lower mole percentage of the nutrient acid incorporated into the cellular lipids (see Table I; 22:3<22:6<20:2). Upon derepression, no differences could be observed in the esterification of added [l-14C]18:2 into the triglycerides and various phospholipid fractions (Table II) of cells supplemented with two different isomers of 20:3. The pattern of linoleate incorporation among the lipids resembled that for cells grown only with linoleate and did not indicate any alteration in lipid formation. The cis-11,14,17-isomer of 20:3 caused many petite cells to form (Fig. l), although it allowed incorporation of [l14C118.2 . into cell lipids to a level similar to that observed for cells supplemented with cis-8,11,14-20:3 which caused no increase in petites above control values. DISCUSSION
Mitochondrial morphogenesis requires protein synthesis on both mitochondrial and cytoplasmic ribosomes. Although mi-
348
GRAFF,
SACKS,
AND
TABLE INCORPORATION
Nutrient [l-‘%]18:2
fatty
acid
OF [l-%]18:2
LANDS
II
INTO CELLULAR LIPID OF S. cerewisiae KD115 (GRANDE) AFTER TRANSFER INTO DEPRESSION MEDIUM
&s-9,12-18:2
eis-8,11,14-2Oz3 (40 @hi)
(20 /.lM)
AT VARIOUS
TIMES
cis-11,14,17-2Oz3 (40 CM)
pulse c0nc
[PMI Time of pulse addition Million cells/ml culture
22.8 0 + 0:40 8.4
11.0 4 - 4:40 18
11.1 9 --+ 940
23.1 0 -+ 0:40
42.8
8.5
[l-“C]18:2
nmol/ml
11.6 4 -4:40 14.1
11.6 9 --+ 940
22.4 0 -0:40
11.2 4 - 4:40
11.2 9 - 940
36.2
8.5
15.9
40.1
culture
TG FA DGiMG
3.8 1.3 0.2
2.6 0.3 0.5
0.6 0.4 0.5
0.5 0.7 0.2
1.5 0.5 0.2
0.6 0.4 0.3
0.8 0.6 0.3
0.9 0.3 0.1
0.9 0.3 0.2
PA + CL PE PS
0.1 0.6 0.3
0.5 0.9 0.5
0.2 1.0 0.8
0.1 0.3 0.2
0.1 0.8 0.5
0.2 1.2 1.3
0.1 0.4 0.3
0.1 0.8 0.6
0.3 1.4 1.3
Pl PC Origin
0.8 3.3 0.6
0.5 4.2 0.3
1.8 5.3 0.3
0.6 1.4 0.1
0.8 2.0 0.1
2.4 3.7 0.2
0.5 1.3 0.1
0.9 1.8 0.1
1.8 3.9 0.2
Note. Cells grown in repression medium were transferred into derepression medium (10.0 ml) containing the fatty acids indicated and grown at 30°C. Additions of [l-“C]18:2 were made at time points indicated. After 40 min, the lipids were extracted from washed cells as described under Materials and Methods.
tochondria contain an independent transcription and protein-synthesizing system, the cytoplasmic protein synthetic system provides components essential for mitochondrial morphogenesis in a highly coordinated and mutually interdependent process (7). For example, the m&DNA-dependent RNA polymerase (8), M-DNA polymerase (9), mitochondrial ribosomal proteins, and initiation and elongation factors are coded for by nuclear genes (10, 11) and they are synthesized on cytoplasmic ribosomes (11, 12, 13). Walenga and Lands (2) provided evidence that at least two molecular events during morphogenesis of mitochondria require unsaturated fatty acids for cells to gain respiratory competence. The early event, associated with a priming process, required a high preexisting content of unsaturated fatty acid in phospholipid to permit the synthesis of components on cytoplasmic ribosomes, but did not require a concurrent supply of unsaturated acids. The second event, promoting the induction of respiration, required the concurrent presence of unsaturated fatty acids. cis-13-
Docosenoic acid failed to support an adequate priming process. Furthermore, it did not support the induction of respiration in cells with a high preexisting unsaturated fatty acid content in membrane phospholipids. The present experiments examine the ability of different unsaturated fatty acids to maintain adequate replication and retention of the mitochondrial genome during repeated cell divisions. Failure to maintain an adequate amount of unsaturated fatty acid in the membranes of 5’. cere&siae has been reported to cause uncoupling of oxidative phosphorylation (2, 14, 15), impairment of mitochondrial ATP translocation (16), lowering of the adenylate energy charge (1’7), as well as depletion of 15 and 21 S mitochondrial ribosomal RNA (18, 19) and lack of formation of mitochondrial cytochromes (2,18,20). The observations were insufficient to indicate whether the minimal content of unsaturated fatty acid in phospholipid required for the maintenance of adequate mitochondrial energy transduction corresponds to that needed to maintain general cellular protein syn-
SELECTIVE
LOSS OF MITOCHONDRIAL
thesis. Conceivably there could be a threshold value for unsaturated acids to support energy transduction that differs from that needed for protein synthesis. We visualize yet another threshold that may represent the amount needed to prevent irreversible loss of m&DNA concurrent with the generation of petite cells (3). Some acids which caused a high percentage of petite cells in the present study still permit,ted the nonpetite KD115 cells to develop cellular respiration during derepression with no detectable impairment (Fig. 2) and also allowed unimpaired esterification of linoleate into membrane phospholipids (Table II). This result is consistent with the high content of the nutrient unsaturated fatty acid that occurred in the membrane phospholipids, more than that required for successful priming as reported by Walenga and Lands (2). Although the fatty acids examined showed some differences in their ability to support the subsequent induction of respiration, the process of induction of respiration appeared not to trigger an immediate discontinuation of m&DNA replication and onset of petite generation as long as an adequate level of unsaturated membrane lipids was present. The data shown in Fig. 2 suggest that most nutrient fatty acids were adequate to support the metabolic processes required for the induction of cell respiration and that the transcription and translation of the mitochondrial genome was still adequate during the first 10 h of induction. The loss of the mitochondrial genome in our experiments may have occurred by one of two ways: inadequate unsaturation or selective acyl chain action. Cells that had initially a high content of adequate fatty acyl chains in their membrane lipids may have (a) di.luted these acyl chains with newly syntlhesized saturated fatty acids, or (b) developed insufficient amounts of unsaturated acyl chains due to a reduced rate of esterification which is regulated by the activities and specificities of cellular acyltransferases. Mitochondrial proteins synthesized on cytoplasmic ribosomes have been suggested to require phospholipids for insertion into the mitochondrial mem-
GENOME
349
brane (1, 21, 22). Perhaps a reduced rate of synthesis of lipids containing unsaturated fatty acid may not provide adequate lipids for complexing with cytoplasmically synthesized proteins. The inability to insert proteins bound to appropriately unsaturated lipids into the mitochondrial membrane may lead to the loss of &-DNA replication and eventual loss of the cellular respiratory competence (3). Our results confirm that such conditions may develop when cells are not provided with sufficient quantities of nutrient unsaturated acids and also occur when the acids provided are incorporated poorly into the needed lipids (1, 23). Alternatively, as exemplified with two octadecenoic acid isomers (cis-9 and c&12-18:1) or with the two isomers of eicosatrienoic acid, some cellular processes required for continued mitochondrial morphogenesis may somehow recognize specific structural features of acyl chains in membrane phospholipids. These specific recognition processes appear to be independent of the mole percentage content of unsaturated fatty acid in membrane phospholipids. In each example noted above, the acid causing petite formation was present at a higher mole percentage content in membrane than was the “normal” fatty acid. Our data suggest that some selective effect of the fatty acid can occur beyond a simple inability to be esterified into cellular phospholipids. Precedence for a hypothesis of selective cellular response to acyl chain structure has been obtained in other investigations from this laboratory (24-26) with a mutant of Esherichia coli that requires unsaturated fatty acids (27). A number of positional and geometric isomers showed poor growth-supporting properties for this organism even though their incorporation into cellular lipids was similar to that of effective nutrient acids. The inability to support growth seemed not readily explained by differences in melting points of these fatty acids or their inability to provide adequate fluidities. These positional isomers became as effective as others to support growth upon addition of CAMP to the medium or by conducting growth with glycerol as the carbon source (25, 26). In
350
GRAFF,
SACKS,
the presence of sufficient CAMP, growth with all positional isomers between 5 and 14 reflected a predictable response related to fluidity. In a similar manner, selected trans-fatty acid isomers which are much less effective in supporting growth of E. coli than their cis-isomeric counterparts (26) readily supported cell growth when CAMP was included in the medium or when glycerol was the carbon source (28). Although added CAMP did not alter responses of 5’. cerevisiae, the results in this report indicate that there are cellular regulatory events which can respond to detailed structural differences in the unsaturated fatty acyl chains. The fact that some mutant fatty acids caused large increases in total phospholipid per cell (see also, Refs. (5,29)) confirms the concept that an important regulatory process in S. cerevisiae may be altered by specific fatty acids. Since loss of mitochondria occurred also in cells not showing increased phospholipid contents, we expect to find a variety of pleiotropic events associated with the influence of acyl chain structure upon cellular regulatory controls. ACKNOWLEDGMENT Research for this report was supported by a grant from the United States Public Health Service (AM05310). REFERENCES 1. WALENGA, R. W., AND LANDS, W. E. M. (1975) J. Biol. Chem 250,9121-9129. 2. WALENGA, R. W., AND LANDS, W. E. M. (1975) J. Biol Chem. 250,9130-9136. 3. MARZUKI, S., HALL, R. M., AND LINNANE, A. W. (1974) Biochem Biophys. Res. Cornmun 57.372378. 4. KEITH, A. D., RESNICK, M. R., AND HALEY, A. B. (1969) J. BacterioL 98.415-420. 5. GRAFF, G., AND LANDS, W. E. M. (1976) Chem. Phys. Lipids 17.301-314. 6. HOFSTETTER, H. H., SEN, H., AND HOLMAN, R. T. (1965) J. Amer. oil Chem. Sot 42, 537-540. 7. SCHATZ, G., AND MASON, T. L. (1974) Annu. Rev. Biochem. 43. 51-87.
AND
LANDS
8. WINTERSBERGER, E. (1970) Biochem Biophys Rex Cemmun. 40,1179-1184. 9. CHIN, J. J., AND KALF, G. F. (1969) Arch. Biochem Biophys. 133, 38-45. 10. RICHTER, D. (1971) Biochemistry 10, 4422-4425. 11. PERISI, B., AND CELLA, R. (1971) FEBS. Lett. 14, 209-213. 12. DAVAY, P. J., Yu, R., AND LINNANE, A. W. (1969) B&hem. Biophys. Res. Commun 36,30-34. 13. KUNTZEL, H. (1969) Nature (London) 222,142-146. 14. HASLAM, J. M., PROUDLOCK, J. W., AND LINNANE, A. W. (1971) J. Bioenwget. 2, 351-370. 15. MARZUKI, S., COBON, G. S., HASLAM, J. M., AND LINNANE, A. W. (1975) Arch B&hem. Biophys. 169, 577-590. 16. HASLAM, J. M., AND FELLOWS, N. F. (1975) Biochem. Sot. Trans. 3, 772-775. 17. HASLAM, J. M., AND FELLOWS, N. F. (1977) Biochem J. 166, 565-570. 18. MARZUKI, S., COBON, G. S., CROWFOOT,P. D., AND LINNANE, A. W. (1975) Arch. Biochem. Biophys. 169, 591-600. 19. GORDON, P. H., AND STEWART, P. R. (1972) J. Gem. Micro. 72, 231-242. 20. WATSON, K., HOUGHTON, R. L., BERTOLI, E., AND GRIFFITHS, D. E. (1975) B&hem. J. 146, 409416. 21. KADENBACH, B. (1967) B&him. Biophys Acta 134, 430-442. 22. KADENBACH, B., in Biochemical Aspects of Biogenesis of Mitochondria (Slater, E. C., Tager, J. M., Papa, S., and Quagliariello, E., eds.). Adriatica Editoice, Bari, Italy. 23. GURBA, P. E. (1979) Ph.D. thesis, University of Michigan. 24. LANDS, W. E. M., OHLROGGE, J. B., ROBINSON, J. R., SACKS, R. W., BARVE, J. A., AND GUNSTONE, F. D. (1977) B&him Biophys. Ada 486, 451-461. 25. OHLROGGE, J. B., BARBER, E. D., LANDS, W. E. M., GUNSTONE, F. D., AND JSMAIL, I. A. (1976) Canad J. Biochem 54,736-745. 26. LANDS, W. E. M. (1979) Guidelines to Metabolic Therapy (Guidelines to Medical Science Series Publ. Upjohn Company), Vol. 8, pp. l-3. 27. LINDEN, C. D., WRIGHT, K. L., MCCONNELL, H. M., AND Fox, C. F. (1973) Proc Nat. Ad Sk USA 70, 2271-2275. 28. TSAO, Y. K., AND LANDS, W. E. M. (1980) Science 207, 777-779. 29. VANDENHOFF, G., GUNSTONE, F. D., BARNE, J. AND LANDS, W. E. M. (1975) J. Bid Chem 250,87208727.