Bit~chlmlca et Biophysica Acre. 1081 (19911279 284 © 1991 ElsevierSciencePublishers B,V.(BiomedicalDivisionl0005-2760/91/$03.50 ADONIS 0005276091000883

279

Effect of sterol side-chain structure on the feed-back control

of sterol biosynthesis in yeast W a r r e n M . C a s e y ~, J a s o n P. B u r g e s s ~- a n d L e o W . P a r k s i I Department oi"Microbiology. North Carolina State Unirervt(r~R~leigh, NC ( U S A I and : Department of Chelnlstry. North Co~hna State Univers#rv.1~aleiglL ~VC( [LS A I

IReceived 11 May 1990) ( Revised manu.~eript11 October 19901 Key words: Oxysterol; Sterol synthesis: Sterol: (Yeast) W e nkoasured the incorporation of radiulabeled methionine and acetate into the sterul componem ol G204, a Saceharomyce~ cercvisiae mutant strain which is partially heine competent. By comparing the amount of label incorporated into the sterul pool o1[ a control culture, to which no exogenous sterol was added, with a culture which had various steruls added to the 8 r o w ~ medium, we were able to determine the specific structural f e a ~ of ergosterol which facilitate its ability to restrict the sterol biosynthetic pathway. These experiments demonstrate that sterols which contain both a C22 unsaluration and a C24 methyl group are capable of reducing stemt b i ~ n t h e s i s by approx. 50%, regardless of B-riag stmetexe. We examined the regulatory properties of various oxysterols; 24,25-epoxylanosterel reduced endogenousbi~ynthesis by 49%, whereas all cholesterol derivatives tested, including 25-hydroxycholesterol, had little effect. A new procedure for the synthesis of ergosterol peroxides is also described.

Introduction Ergosterol is the principal product of the sterol biosynthetic pathway in yeast. Various reports indicate that ergosterol is capable of regulating the isoprenoid pathway by some type of feedback control mechanism [1-4]. In this study we have examined the specific structural features of ergosterol which precipitate this regulatory role. In addition, we examined the possible regulatory effects of lanosterol, as well as various oxysterols, on sterol biosynthesis in yeast. Saccharomyces cereoisiae strain G204 possesses a leaky mutation in H E M I , the structural gene for ,$aminolevulinic acid (ALA) synthase. This enzyme catalyzes the conversion of succinyl coenzyme A and glycine to ALA, a precursor of heme. As a result of this block in the heme pathway, G204 is capable of taking up exogenously supplied sterol under aerobic conditions. Conversely, uptake decreases as the cells become more

Abbreviations: ALA, 8-amino!cvulini¢ acid; ¢p0xylanosterol. 24,25(R.$)-epoxylanosterol; ergosterol peroxide, ergosta-5.8-end~ peroxy-6,22-dien-3B-ol~ HPLC. Irish-performance liquid chromatography: RRT. relative retention time. Correspondence: L.W. Parks. Dept. Microbiology, North Carolina State Urb,e~ay. Box 7615, Raleigh.NC, 27695-7615,U.S.A.

heme competent through the addition of ALA [5,6]. Heine products are also required in the sterel biosynthetic pathway for the activity of CI4 demethylase [7] and C22 desaturase [:8]. Consequently, O204 accumulates principally lanosterol and ergosta-5,7-dienol, with only sparking amounts [9,10] of ergosterol being produced. As mentioned above, heine c~mpounds have both regulatory and catalytic functions in sterol biosynthesis. Thus, a totally heine-deficient mutant cannot synfll~ize sterols past lanosterol. Complete heme competency, however, precludes cells from taking exogenous sterols from the growth medium. Because we wanted to measure the effects of exogenously supplied sterol on cultures which were as close to wild-type as possible, we fed our strain enough ALA to partially restore heine competency while still allowing the uptake of sterols (one-half the uptake of non ALA supplemented cells). In the experiments described here we tested the effect of exogenously supplied sterols on endogenous sterol production. The intracellular sterol level was determined by following the incorporation of radiolabeled acetate or methionine into the sterol pool. Our results indicate that sterols that possess a C22 unsaturation in conjunction with C24 methyl group, regardless of B ring unsaturatioa, are most effective in reducing endogenous sterol synthesis. Additionally. the only oxyslerol that

280 caused a reduction in sterol biosynthesis (aside from the ergosterol peroxides, which have an ergosterol-like side-chain) was 24.25-epoxylanosterol. Materials and Methods Ye¢t~t strain Saccharomyces cerevisiae stre, in G204 (heml his4) [11] was used in all experiment~. Chemica[~ Ergosterol, cholesterol, 7-dehydrocholesteroL 25-hydroxycholesterol, 5,6-epoxycholesterol, lanosterol and hematoporphyrin were purchased from Sigma Chemical (St. Louis, MO). 22-Dehydrocholesterol, 22-hydroxycholesterol and 24r-methyl-cholesterol were purchased from Research Plus (Bayonne, NJ). Ergostanol was a gift from the late Dr. Henry Kirchner. 24,25(R, S)Epoxylanosterol (epoxylanosterol) was prepared by the method of Panini et al. [12]. Its structure was confirmed by GC-MS analysis as well as co-elution on both GC and HPLC with an authentic standard that was provided by Dr. Thomas Spencer (Dartmouth College, NH). Ergosta-5, 8-endoperoxy-6, 22-dien-3fl-ol (ergosterol peroxide) and ergosta-5, 8-endoperoxy-6, 9(ll), 22-trien-3~-ol (9(ll)-dehydroergosterol peroxide) were prepared as described below. Ergosterol, cholesterol, cholestanol, ergosta-5.7-dienol, lanosterol, epoxylanosterol and the ergosterol peroxides were all purified by HPLC immediately prior to use [131. L-[Me~4C]Methionine and [3H]CH3COONa were purchased from New England Nuclear (Wilmington, DE). 8Aminolevulinic acid (ALA) was purchased from Sigma. Organic solvents were from Fishc," (Fair Lawn, N J). Media and growth conditions All cultures were grown in defined medium that consisted of 2% glucose, 0.67% yeast nitrogen base (Difcc, Laboratories, Detroit, MI), 1% casamino acids (Difco). 30 # g / m l adenine, and 20 # g / m l of each methionine and uracil. The medium was buffered with 50 mM succinic acid and adjusted to pH 5.5 with KOH. The cells were supplemented with an unsaturated fatty acid mixture consisting of oleic and palmitoleic acids (4:1, v : v ) in tergitol/ethanol ( 1 : 1 , v:v), to a final concentration of 50 ~g/ml. Sterols were added to a concentration of 10/~g/ml from stocks of 10 m g / m l in tergitol:ethanol. ALA was added to the medium to a concentration of 10 g g / m l in order to partially restore heine competency. Ceils were grown at 30°C with rotary shaking in the presence of either [t4C]methionine 10.005 /~Ci/ml) or [~Hlacetate (0.020 #Ci/ml). Cell growth was monitored at 600 nm using a Varian (Sugarland, TX) double beam spectrophotometer.

Sterol extraction and quantification Cells were harvested by centrifugation at mid-log (A~0o ~ 0.3 0.5, approximately 12-15 h). The cell pellet was washed once with 0.5/% tergitol and twice with distilled water before being frozen to - 7 0 " C and lyophilized for 24 h. The pellet was then weighed and divided into at least two, usually three, samples. The sterol extraction procedure was basically as described previously [14]. Briefly, dimethylsulfoxide (0.5 ml) was added and the samples were steamed for 1 h, after which time a ml of 2 M KCl/methanol (4: 1) was added. The samples were then extracted twice with 5 ml hexane and once with 2 ml ether. The organic layers from each sample were pooled and dried under reduced pressure. The samples were resuspended in ether (3 × 150 /tl) and spotted on analytical TLC plates (Kiesegel 60 F254, 20 × 20 cm, 0.025 mm thickness, Merck). The plates were developed in the solvent system described by Skipski et al. [151. Radioactivity on the plates was detected using a BioScan System 200 Imaging Scanner (BioSean, Washington, DC). Radioactive bands from each sample corresponding to the free and esterifled sterol fractions were scraped, and the amount of radioactivity was determined using a Beckman LS 5801 series scintillation system. When the cells were labelled with [3H]acetate, the ester fraction was saponified [14] and re-run on the TLC system before being counted. All samples were done in duplicate or triplicate, and each experiment was repeated a minimum of one time. in our experiments we assumed that the sterols tested were taken up from the medium at least as well as ergosterol [4,16-18]. The growth rate of the cultures was essentially identical on all the stesols tested, although cells grown in the presence of exogenous sterol had a somewhat reduced lag time (generally 2 - 3 h shorter than the control). All cell pellet dry weights were were within 20% of the control. Synthesis of ergosterol peroxides Ergosterol (500 rag) was dissolved in 100 ml ehloroform/metlianol ( 2 : 1 , v/v) and hematoporphyrin was added to a concentration of 1 ?zM. The solution was continuously aerated while being exposed to the light emitted from four G E 120 W floodlamps which were 30 cm from the 250 ml flask. The solution was continuously stirred for 30 min, after which time 20 g of silica gel G (EM laboratories, EImsford, NY) was added to the flask. The solvents were then evaporated under reduced pressure. The dried residue was scraped from the flask, extracted with diethyl ether and the silica removed by filtration. The ether was evaporated and the products were spotted on preparative thin layer chromatography plates (1 mm Silica Gel (3, 20 × 20 cm, Analtech, Newark, DE) which were developed in toluene/ethyl acetate (2 : 1, v : v). The peroxides, which

281 appeared as a s;cgle band ( R F - 0.45). were scraped and eluted off the silica with chloroform, which was evaporated under reduced pressure and the products resuspended in methanol. The peroxides were then purified by semi-preparative H P L C as described above. Using ergosterol as the reference compound, ergosterol peroxide (I) had a relative retention time ( R R T ) of 0 4 4 9(ll)dehydroergosterol (11) had a R R T of 0,35 and a third, unidentified endoperoxide (111), had a R R T of 0.30. This procedure typically yielded 80 90 m g L 35 40 m g n and less than 1 rag IlL

Mass spectroscopy Low resolution mass spectroscopy ( L R M S ) was performed via chemical ionization on a H P 5985B G C MS. using methane as the ionizable gas. Samples were delivered by direct insertional probe after purification by H P L C . High resolution mass spectroscopy ( H R M S ) was performed on I and II by electron i m p a c t / p e a k matching with a J E O L H X I I O H F double-focussing mass spectrometer equipped with a high field (2,3 Tesla) homogeneous magnet and D E C 1 1 / 7 3 computer. H R M S analysis of I demonstrated a molecular weight of 428.32905 (428.32906 calculated) while II had a measured molecular weight of 426.31345 (426.31341 calculated). The L R M S d a t a for each of the three products is given: I: m / z = 429 (relative intensity 35.71, 411 (100), 393 (59.21, 377 (20.2), 303 (12,11, 125 (73.9); I1: m / : = 427 (52.31, 409 (10el), 391 (31,3). 375 (17.21, 301 (14.51. 125 (62.4): Ill; ra/z ~ 427 (10.5), 409 (10.01, 391 (24.0), 375 (70.4), 301 (4.3), 125 (100). Combustion analysis Products I and U were analyzed for carbon and hydrogen b y Atlantic Microlah (Norcross, GA). All measurements from the combustion analysis were within 0 . 3 0 ~ of calculated values for the p r o p o ~ d structures. Nuclear magnetic resonance N M R analysis was performed on a G E G N 3 0 0 N M R spectrometer operating a 300.521 M H z for protons and 75.574 H M z for carbons. A pulse delay of 1 s was used for protons and 3 s for carbons. All samples were dissolved in C D C I ~ and chemical shifts are given as p p m d o w n f i d d of tetramethylsilane which was added as an internal standard. W e were only able to perform 13C-analysis on I and II due to the low yield of III in our method of synthesis. The 1"~C-NMR spectra demonstrated 28 distinct signals in both I and I! as well as ergosterol (Table I). Signal assignments were made by using the carbon spectrum of each molecule in conjunction with an Attached Proton Test (APT) and previously reported values for the starting material [19]. The I H - N M R d a t a for all three molecules is given: I ~i = 0.80 (M, 9H), 8 - 0.90 (M, 6H). 8 = 1 . 0 0 (d, 3H. J - 8 . 2 H z ) . 8 = 1 . 2 0 2.15

TABLE I ~CNMR data The chemical shifts for carbons in ergosterol (Ergl were based on previously rcporled dala [19I. Carbon assignments for the peroxides were based on values for ergosterol in coniur.ctlon with APT experiments, as well as chemical ghif! rules estabbshed for sterov3s 130]. Carbon No I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

6 Ippm) Erg 38.3 31.7 70.3 40.6 139.6 llg5 116 1 141.2 46.o 37.0 19.4 39.2 42.6 54.4

22.8 28.1 SS.6 12.1 16,1 40.3 19.8 135.5 131.8 42.6 3Z9 20.9 19.7 17.2

I

n

36.6 30.0 660 39.0 79.1 135.0 i304 82.0 50 g 34.4 20.5 36.6 44.5 51.4 231

35.8 30.4 66.2 37.8 7g.3 135.0 1306 82.6 47.q 32,0 20.5 41.4 434

28.4

5sg 12.7 18,0

39.5 19.5 135.2 132.0 42.5 32.8 20,6 19.7 17.4

142.4

119.6 28.4 55,6 12,7 25.4 39.8 19.5 135.3 132.3 42.5 33.0 20.7 19.7 17.6

(complex mult, 22H), 8 = 3.97 (m, IH), 8 = 5.21 (m, 2 H k 5 = 6.24 (d, 1H, J = 8.11, 8 = 6.52 (d, 1H, J = 8.11; II ~ = 0.75 (S, 3H). 8 = 0.82 (m, 6H). 8 = 0.90 (d. 3H, J = 8 . 1 H z ) , 8 = 1 . 0 0 (d, 3H, J = 8 . 1 H z ) , 8 ~ 1 . 1 0 (S, 311) 8 1.20-2.35 (complex mult, 19H), 8 = 4.01 (m, I H L 8 - 5.22 (m, 2H). 8 = 5.42 (dd. I H . J = 2.2. 5.81, 8 = 6 . 3 1 Id, I H , J - 8.81, 8 ~ 6 . 6 2 (d, 1H, J ~ 8,8); III - 0 . 8 - 2 . 2 5 (broad muir, 37H), 8 = 4 . 0 1 (m, 1HI, 8 5.22 (m, 2HI, 8 ~ 5.56 (d, I H , . / = 9.5), 8 = 5.91 (d, I H , J = 9,5k 8 - 5,95 (dd, 1HI.

Metabolism of sterols [14C]Ergosterol peroxides (13.42 / z C i / # m o l ) were synthesized and purified as described above using [14C]ergosterol [20] as a starting material. The radiolabeled sterol was then supplied in the medium as described above. T h e cells were harvested at mid log, frozen and lyophylized. In order to avoid the potential conversion of the sterols during the somewhat harsh conditions described in the sterol extraction procedure above, an alternate method of sterol extraction was

282 employed. The ceils were resuspended in 50 mM tris buffer and an equal volume of glass beads was added. The cell walls were disrupted by vortexing vigoursly for five 1 rain intervals, with the cells being cooled on ice for one minute in between. The sterols were extracted with chloroform/methanol (4:1, v/v). Samples were run and eluted off TLC plates as described above. Sterols were then analyzed by analytical HPLC using a Beckman 171 Radioisotope Detector irt-line with a Beckman uttrasphere a D S column. A similar feeding and extraction proceedure was used for non radiolabeled epoxylanosterol, with the sterols being detected at 210 nm for the H P L C analysis. Results G204 was grown on [14C]methionine in the presence of sterols with various structures, harvested at mid-log and the amount of radioactivity in the total sterol pool was measured. When these samples were compared to a control in which no sterol was added, we were able to determine the decrease in the amount of radioactivity incorporated into the sterol component. The sterols tested, as well as the results, are given in Table II. These data indicate that ergosterol and the ergosterol peroxides were the most efficient in reducing sterol biosynthesis (47-5670 of control levels). Both 24fl-methylcholesterol and ergostanol were moderately effective (73¢Z and 76% of control levels, respectively), while cholesterol and various cholesterol derivatives basically had no effect on sterol production (90-102% of control).

TABLEn [ 14C]Methi~me labeling of endo$enot~ly produced sterols Cells were grown in the presence of Ihe sterols listed below and labeled with [HClmeddonineas described in the text. Approx. 20~oof the ladiolaheled methionine in the medium was incorporaled into Ihe cell pellet.Sterols were extracted and the amount of radiolabel in the sterol fraction was determined. Results are given as the percentage of the dpm per nag dry weightcompared to a ~ntrol (no sterol added). Tile standard deviation fax) and the number of times the experiment was performed (hi are given. Sterol added Ergosterol Ergoster~aperoxide 9(11)Dehydroergosterol peroxide 24-Methylcholesterol Ergostanol 22-Dehydrocholesterol 7-Dehydr~holesterol 5.6-Epoxycholesterol 25-Hydroxycbolesterel 22-Hydroxycholestero[ Cholesterol

%Control (dpm/mg dw) 47.0 54.6 55.6 72.7 76.2 89.8 92.5 93.5 97.5 101.4 1021

Sx

n

6.3 8,5 ILl 4.5 2.4 4.3 2.1 2.6 0.8 2.5 4.1

5 3 3 2 2 2 3 3 3 3 6

TABLElit [ ~H]Acetate labeling of endogenol~ly produced sterols Cells were grown in the presence of the sterols listed helow and labeled with [SH]acelateas described in the re:d. Approx. 45% of the radiolabeled acetate from the medium was in~rporated into the cell pellet, aterols ~ere extracted and the amount of radiolabel in the slerol fraction was determined. Resultsare given as the percentage of the dpm per mg dry weightcompared to a control (no sterol added). The standard deviaaon (Sx) and the number of times the experiment was p~formed (n) are given Slerol added Ergosterol Ergosterol peroxide Cholesterol 25-Hydroxycbelesterol Lanosterol 24.25-Epoxylanoslerol

%Control (dpm/mg dw) 47.7 56.8 1043 98.6 68.6 51.2

Sx

n

3.4 4.2 6.6 s.s 1.5 3.3

3 2 3 2 2 4

Because m~hionine labeling only gives an indication o1' the amount of C24 methylated sterol present and because G204 has a heine mutation that causes an accumulation a non-C24 methylated sterol (lanosterol), we sought to determine what effect certain sterols would have on the total sterol pool as determined by acetate labeling (Table lll). This experiment demonstrated that epoxylanosterol decreased sterol biosynthesis by 49%, while lanosterol only caused a 31% decrease. Ergosterol, ergosterol peroxide, cholesterol and 25-hydroxycholesterol all demonstrated basically the same effects seen in the methionine labeling experiment. In order to address the possibility that ergosterol might undergo autooxidation during the experiments, a control was performed in which vitamin E (atocopherol) was added to the medium at 5 g g / m l [21]. The addition of this antioxidant had no effect on the ability of ergosterol to repress sterol biosynthesis (data not shown). Of the sterols which precipitated a substantial decrease in sterol biosynthesis, ergosterol was found to he taken up from the medium least effecively (1.82:1:0.53 ~,g/mg dry weight). The ergosterol peroxides, epoxyla~osterol and cholesterol were found to be taken up equally well (approx. 4 p g / m g dw). In addition, these sterols were found to be highly esterified, whereas ergosterol was found predominantly in the free fraction. We were unable to detect any metabolism of ergosterol or the ergosterol pero~des in cells harvested at mid log, our limits of detection being 10 ng sterol/mg dry weight. However, epoxylanosterol appears to lie metabolized to several different products. The identification of these products and their possible physiological effects are currently under investigation.

283 Discussion The results of our labeling experiments with both acetate and methionine indicate that those sterols that have art unsaturation at C22 and a methyl group at C24 and those that are "lanosterol-like' (sterols that contain a 14a-methyl and a 4.4-dimethyl group), have the capability to inhibit sterol biosynthesis in yeast. For example, by comparing the structures of ergosterol, ergosterol peroxide and 7-dehydrocholesterol, we can see that ergosterol and 7-dehydr0cholesterol have an identical B-ring structure (C5 and C7 unsaturations), while ergosterol peroxide possesss a radically different structure in the B-ring (Fig. 1). Ergosterol and ergosterol peroxide have identical sidechains, while 7-dehydrocholesterol lacks the C22 unsaturation and the methyl group at C24. The data in Table II indicate that. of these three sterols, only ergosterol and ergosterol peroxide are effective in reducing sterol biosynthesis. Thus, it would appear that the structure of the side-chain, not the B-ring, is important to the inhibition process. Ergostanol and 24fl-methylcholesterol, are moderately effective in restricting sterol biosynthesis. This could be due solely to the methyl at C24. or because the yeast is introducing an unsaturation at C22 to yield a repressive sterol structure as described above. In fight of the fact that oxysterols have been demonstrated to regulate cholesterugeoesis in mammalian systems [12,22-24], we wanted to determine what influence they had on sterol biosyntheis in yeast. We tested the effects of various oxysterols and found that the

cholesterol derivatives, including 25-hydroxycholesteroL were basically ineffective. The introduction of an epoxide group to the side-chain of lanosterol dramatically increased its ability to reduce sterol biosynthesis. Lanosterol is a native sterol of yeast and has previously been shown to inhibit squalene epoxidase [18]. 24,25Epoxylanosterol has been isolated from yeast that were exposed to an inhibitor of 2,3-oxidosqualene cyclase [25], but it has not been detected in a wild-type strain. Ergosterol peroxide, which reduced biosynthesis by approx. 50%, has been isolat~:d from yeast and other fungi [26 28] and has been proposed to be a natural metabolite of ergosterol [28], Some researchers have proposed that only those sterols which have an unsaturation a the C5 position are effective inhibitors of sterol biosynthesis [3,4]. Our present study asserts that, under the described conditions, only those sterols which contain an unsaturation at C22 and a methyl at C24 are effective inhibitors (with the exception of 'lanosterol-like" compounds)+ and that C5 uosaturation is inconsequential. Our findings do not necessarily contradict previous work done in this area. One possible explanation for the different results could be that them are two levels of regulation, each with different structural requirements. In this scenario, our data would reflect only one level of regulation, with the other level(s) being masked by endogenously produced ergosta-5,7-dienol. This would be consistent with both a proposed dual pathway hypothesis [3] and recent findings that demonstrate the two structural genes for 3-hydroxy-3-methylglutaryl coenzyme A reductase are regulated by different mechanisms [29]. Our results do not premit us to define the specific physiological interactions attendant to the depression of sterol biosynthesis we have observed. However. we have shown that the structure of the sterol side-chain is one critical element in this regulatory process. Acknowledgements

Derox,~e

Ergostanol was a gift from the late Dr. Henry Kirchner. Authentic standards of 24,25(R)-epoxytanosteroi and 24.25(S'pepoxylanosterol were kindly provided by Dr. Thomas Spencer. We would like to thank Dr. Carl Bumgardner for helpful discussions involving the N M R data. This research was supported by grants from the National Science Foundation (DCB8814387), the National Institutes of Health ( D K ~7222) and by the North Carolina Agricultural Research Service. References

Fig~ I. Structures of ergosterol, ergosterol peroxide and 7-dehydrecholesterol.

1 Servouse.M. and Karn, F (1986) Biochem.J. 240, 541-547. 2 Bard, M and Downing. J,F, (1981) J. Gem Microbiol. 125, 415420

284 3 Nes, W.R. and Dhanuka, L C (19881 J. Biol Chem. 263 (241. 11844-11850. 4 Pinto. J,P.. Lozano, R, and Nes, W.R. (19951 Blochim, Biophys. Acta 836. 89-95. 5 Shinabarger, D.L. Keesler, G.A. and Parks, L.W. (1999) Steroids 53, 608-623. 6 Lewis. T.A., Taylor, F.R. and Parks L.W, (1985) J. Bacteriol. 163. 199-207. 7 Aoyama, Y. and Yoshida, Y. (1978) Biochem. Biophys. Res. Commun. 85, 28-34. 8 Hata, S., Nishino, T., Kalsuki. H. Aoyama, Y and Yoshida. Y. (1987) Agric. BioL Chem., 51, 1349-1354. 9 Rodriguez, R,J, Taylor, F.R. and Parks. LW. (19821 Biochem. Biophys. Res. Commun. 106. 435-448. I0 Rodriguez. R.J. and Parks, L.W (19R3) Arch. Bi~hgm. Biophys. 225. 861-871 11 Urhan-Grimal. D. and Labbe-Bois, R (1981) MOL Gem Genet. 183.85-92 1] Panini, S.R, Sexton, R.C.. Gupta. A.K,. Parish. EJ,. Chitrakom. S and Rudney. H. (19861 J. Lipid Res. 27, 1190-1204. 13 Rodrigue~ RJ. and Parks. L.W. (1982) Anal. Bi~hem, 119. 209 204. 14 Parks. L.W., Botlema, C D.K., Rodriguez, R.J. and Lewis, T.A. (19851 Methods Enzymol. I l L 333-346, 15 Skipski, V.P. and Barclay, M, (19701 Melhods Enzymol. 14, 530599. 16 Salerno. L.F. and Parks, L.W. (19~i3) Biochim. Biophys. Acta 752, 240 243.

17 Lorenz, R.T., RodrJguez, R.J., Lewis, T.A. and Parks. L,W. 11986) J. Bacteriok 167, 981-985. 18 Parish. EJ.. Nanduri. V,B.B.. Kohl, H FI and Taylor, F R 0986) Lipids 21, 27 3P,. 19 Cushley, R.J. and Filipenko, J.D. 0976) Org, Magn Reson, 8, 308-309. 20 Casey, W.M and Parks. L,W. 0989) Phot~hem. Photobiol 50, 553-556. 1; Tayhlr, F.R. and Kandutsch, A.A. (19851 Methods Enzymoh 110, 9-19. 22 Kandutuh, A,A. and Chert, H.W (19781 Lipids 13, 704 707. 23 Kandutsh, A.A,. Chen, H.W. and Heiniger, H. (19781 Science 201. 498-501. 24 M'Baya, B., Veguer. M., Servouse, M. and Kant. P. (19891 Lipids 24,1020-1023. 25 Field, R.B. and Holmlund. C.E. (19771 Arch. Biochem Biophys. 180, 465-431, 26 Turn~r, W.B. (19711 in Fungal Metabolites. p. 256, Academic Press. New York. 27 Fisch, M.H., Ernst, R., Flick, B.H., Arditli. J., Mangnus, P.D. Menzies, I.D. and Barton, D.R. (19731 J. Chem. Soc. Chem. Comm. 530. 28 Ncs. W.D., Xu. S. and Haddon, W.F, {1989) Slemids 53. 533-558. 29 Thorness. M., Schafer, W., D'Ari. L, and Rin¢, J, (1989) Moh Cell. Bioh 9. 5702-5712. 30 Akihisa, T. (1989) in Analysis of Sterols zad Other Biologically Sisnificant Steroids (Nes, W.D. and Parish, EJ., eds.), pp. 251-265. Academic Press. San Diego.