326 Biochem. J. (1963) 87, 326

Studies in Carotenogenesis 30. THE PROBLEM OF LYCOPERSENE FORMATION IN NEUROSPORA CRASSA

BY B. H. DAVIES, D. JONES AND T. W. GOODWIN Department of Agricultural Biochemistry, University College of Wales, Aberystwyth, Wales (Received 1 October 1962) There is evidence that the first tetraterpene Institut fur organische Chemie der Universitat, Bern, formed in carotenoid biosynthesis is not lyco- Switzerland. persene, the C40 analogue of squalene, but phytoene Cultural conditions. The fungus was maintained on Oxoid (15,15'-dehydrolycopersene). Davies, Goodwin & Czapek-Dox agar slopes in the light at room temperature. Inoculation of liquid cultures was carried out with a spore Mercer (1961) failed to find lycopersene in any caro- suspension prepared from such a slope. tenogenic system they examined, and it was sug- The organism was grown for 6 days in shake cultures in gested that lycopersene plays no part in carotenoid the light (4 x 100 w tungsten lamps, 2 ft. away) at 37°. The biosynthesis (Davies, 1961, 1962; Goodwin, 1961). formation of large aggregates was prevented by the addi- Mercer, Davies & Goodwin (1963) studied the tion to the culture medium of Tween 80, which also pro- incorporation of several labelled substrates into motes carotenogenesis (Krzeminski & Quackenbush, higher plant terpenes, but found no evidence for 1960a, b). The medium used was similar to that of Mitchell lycopersene being a carotenoid precursor. Ander- & Houlahan (1946), as adapted for maximal carotene son & Porter (1962) found that labelled terpenyl production by Krzeminski & Quackenbush (1960a, b). Its composition, per 1., was: sucrose, 20 g.; ammonium pyrophosphates are incorporated into phytoene by tartrate, 5 g.; NH4NO3, 5-5 g.; KH2PO4, 1 g.; MgSO4,7H20, tomato and carrot plastids, but no radioactive 0.5 g.; CaCl2, 0.1 g.; FeCl3, 5 mg.; ZnSO4,7H20, 2 mg.; lycopersene was detected. biotin, 50,ug.; Tween 80, 8 g. The final pH of the culture In fungi, however, although lycopersene is absent medium was 5 0. Production of fully unsaturated caro- from Phycomyces blakesleeanus (B. H. Davies, un- tenoids was inhibited by the addition of 5 ml. of ethanolic published work), Grob, Kirschner & Lynen (1961) 0-25% diphenylamine/l. of medium. reported that a system from Neurospora crassa Grob & Boschetti (1962) cultured N. crassa statically in could synthesize radioactive lycopersene from the dark at 290 for 6 days. Their culture medium was [14C]geranylgeranyl pyrophosphate. Grob & Bos- essentially the same except that the detergent was omitted, and glucose replaced sucrose as the carbon source (E. C. chetti (1962) reported that the two major com- Grob & A. Boschetti, personal communication). ponents of a phytoene-free fraction of the total Materials. The source of materials and the purification of lipid from N. crassa, which had been cultured in the solvents are described in the preceding paper (Mercer et al. presence of diphenylamine, were squalene and lyco- 1963). persene. The apparent difference between this and Extraction of the unsaponifiable material. The lightly other carotenogenic organisms led to the present pigmented cultures of N. crassa were filtered through investigation into the alleged formation of lyco- cheese-cloth under suction and washed several times with persene in N. crassa. water to remove traces of diphenylamine and detergent. Carotenoid biosynthesis is partially inhibited by The mycelia were then ground under redistilled acetone in N. with washed silver sand until all the pigment had been diphenylamine crassa, as in many other caro- extracted. The' solid residue was filtered off and the lipids tenogenic micro-organisms (see Goodwin, 1959). were transferred from acetone to diethyl ether by the addi- This inhibition results in the accumulation of the tion of an equal volume of diethyl ether and then water, more saturated C40 polyenes, phytoene, phytofluene, dropwise, until two layers formed. Any emulsions formed 6-carotene and neurosporene (Turian & Haxo, by traces of the detergent were broken by the addition of 1952; Turian, 1957; Zalokar, 1957). If lycopersene solid NaCl. The diethyl ether was removed under reduced were concerned in carotenoid biosynthesis, it would pressure and the lipid residue was then saponified by boiling probably accumulate along with phytoene, etc. in for 10 min. with ethanolic KOH. The solution was allowed diphenylamine-inhibited cultures. to cool, diluited with 2 vol. of water and extracted three times with diethyl ether. The ethereal extracts were bulked and washed with water until the washings were no longer EXPERIMENTAL alkaline to phenolphthalein. The pale-yellow solution was dried over anhydrous Na2SO4 and the solvent evaporated Organism. Two strains of Neurospora crassa were used in under N2. The unsaponifiable material was a yellow oil. these investigations. One was isolated from bread and pro- Chromatography of the unsaponifiable material. Samples vided by Dr I. M. Wilson, Department of Botany, Uni- of the unsaponifiable material (usually about 100 mg.) were versity College of Wales, Aberystwyth; the other was a dissolved in light petroleum (40-60°) and chromatographed wild type (A 10336) provided by Professor E. C. Grob, on a 20 g. alumina column. The acid-washed alumina Vol. 87 VolI87LYCOPERSENE IN-N.. CRASSA 327 (Woelm) was deactivated with water to Brockmann grade Spectro.scopic examination ofpolyenes. Absorption spectra II, made up in a slurry with light petroleum and added in were recorded with a Unicam SP. 500 spectrophotometer. this form to a 1-5 cm. diam. glass column. After the Samples were dissolved in redistilled light petroleum alumina particles had settled, the column was washed (b.p. 40 60°). with 100 ml. of light petroleum to remove hydrocarbon impurities. The sample of the unsaponifiable material was RESULTS added to the column in solution in light petroleum, and Impurities in solvents. Light petroleum (Ana- the same solvent was used to develop the chromato- laR) must be purified by redistillation, because the gram. Fractions (5 ml.) were collected and were evapor- residue left after distillation contains components ated to dryness under a stream of N2 before subsequent analysis. that run with Rp values 0-7, 0-4, 0-3, 0-15 and 0-0 Thin-layer chromatography of terpenoid hydrocarbons. on the thin layer of silica gel 'G' and stain with Small samples of the unsaponifiable material and hydro- iodine. Synthetic lycopersene also runs with carbon fractions from the alumina column were chromato- Rp 0-3, but it can be distinguished from the graphed on thin layers of silica gel G (Merck) supported on petroleum impurity by examination under ultra- glass plates (20 cm. x 20 cm. or 50 cm. x 20 cm.); light violet light, when the impurity fluoresces; lyco- petroleum (b.p. 40-60°) was used to develop the chromato- persene does not. While lycopersene gives a pink grams. There was a slight variation in Rp values, although a colour in its antimony trichloride reaction, the saturation chamber was employed (Stahl, 1959; Davies, petroleum impurity stains grey-brown. Similar 1963), but authentic samples of phytoene, lycopersene and squalene were always chromatographed on the same plate. impurities could be detected in the AnalaR diethyl After the solvent had run 10 cm., the positions of the ether and acetone, so these, too, were redistilled hydrocarbons were revealed. before use. Two methods of staining were employed: either iodine Before the unsaponifiable material was applied to vapour (Davies et al. 1961) or SbCl3 (Grob & Boschetti, the alumina column, the column was pre-flushed 1962). When a plate was stained with iodine, it was kept in with 100 ml. oflight petroleum to remove a material an airtight chamber in which a few crystals of iodine were that runs with the same R. as squalene on a silica- dropped on to hot sand in a crucible. The terpenoid (un- gel-G plate. Another impurity had to be removed saturated) hydrocarbons were revealed as brown spots from the sodium chloride and anhydrous sodium against a white background; the use of too much iodine resulted in the overall staining of the plate. The hydro- sulphate before these could be used in the analytical carbons can also be revealed by spraying the plate with a procedure. This material ran with R. about 0-2 and saturated solution of SbCl3 in CHC13 and heating the plate 'stained blue-green with iodine- vapour. at 1100 for 15-20 min. Staining with iodine is the more Terpenoid- hydrocarbons in diphenylamine-in- sensitive method, since as little as 0-05 Lg. of squalene, hibited cultures of N. crassa. Cultures of N. crassa lycopersene or phytoene can be detected (Davies et al. were grown in the presence of diphenylamine 1961), but all the hydrocarbons stain yellow or brown, (12-5 mg./l.) in the light at 370. The unsaponifiable according to the amount present. Antimony trichloride, material was isolated from' 100 g. of mycelia (fresh though less sensitive, has the advantage that different and on a 20 g. alumina colour reactions are given with different hydrocarbons (see wt.) chromatographed Table 1). The sensitivity of this method, however, can be column (Brockmann grade II), with redistilled light increased by viewing the plate after staining under long- petroleum (b.p. 40-600) as the developing solvent. wave (366 m,) u.v. light. The stained hydrocarbons are Twenty fractions, each of 5 ml., were collected and detected by their orange fluorescence. As little as 1 ,tg. of evaporated'to dryness, and each was chromato- squalene, lycopersene or phytoene can be detected in this graphed on a thin layer of silica gel G. No lyco- way. persene could be detected (Fig. 1), even by viewing A fuller account of some of these methods has been given the antimony trichloride-sprayed chromatogram by Mercer et al. (1963). under long-wave u.v. radiation. The concentration Determination of hydrocarbons. The very small quantities of the squalene in the eluate was maximal after of hydrocarbons can only be measured by putting the above-mentioned colour reactions with iodine and SbCl3 on 30 ml. of light petroleum had run through the a quantitative basis. Since fading occurs, both on the plate column, and the phytoene concentration was and in solution,'elution of the coloured spots followed by maximal after 70ml. A control analysis of a spectrophotometric determination has proved inaccurate mixture of squalene, lycopersene and phytoene (E. I. Mercer & K. J. Treharne, unpublished work). -A showed that, if lycopersene were present, its concen- greater accuracy (within 10 %) has been achieved by direct tration in the eluate should reach a peak value after visual comparison ofthe stain intensities of spots with those 50 ml. of light petroleum had run through the of standard quantities run on the same plate. column. Determination of radioactivity. Measurements of 14C The more sensitive iodine-staining method failed activity were made with a Nuclear Measurements Corp. (Chicago, Ill., U.S.A.) PCC-1OA proportional counter 'to reveal lycopersene, even when the unsaponifiable (efficiency 35%) coupled to a Panax (Redhill, Surrey) material from larger quantities (200-300 g.) of lOOC E.H.T. and fast scaler unit. All counts were made at fresh mycelium was analysed. Several compounds infinite thinness. other than those already described, however, could 328 B. H. DAVIES, D. JONES AND T. W. GOODWIN 1963 1-1 Table 1. Thin-layer chromatography of the un8aponi- from Neurospora crassa 1-0 fiable fraction Materials present in the fraction of unsaponifiable 0-9 material of N. crassa were eluted with light petroleum (200 ml.) from a 20 g. Brockmann grade II alumina column, 08 chromatographed on a thin layer of silica gel G with light petroleum and stained with antimony trichloride. (Syn- 0-7 thetic lycopersene stains pink and has R. 0-3.) Colour with SbCl3 Nature (if known)

- Yellow Waxy material 1-0 Purple 0-8 BZ Grey-blue 0-75 Pink Squalene 0-4 Brown Phytoene 0-2 Dark grey-brown Phytofluene 0-1 Dark grey 8-Carotene 0

Table 2. Labelling of components of the unsaponifi- 0-1 able fraction from Neurospora crassa Radioactivity was measured in components of the un- saponifiable fraction from N. crassa grown in the presence 0 10 20 30 40 50 60 70 80 90 100 of diphenylamine and DL-[2-14C]mevalonic acid (17 ,umoles, Vol. of light petroleum (ml.) 20 ,uc) Total activity Fig. 1. Squalene (0) and phytoene (@) obtained from the Fraction (counts/min.) unsaponifiable fraction of 100 g. of N. cra8sa grown in the Total unsaponifiable fraction 3-5 x 105 presence of diphenylamine, separated on an alumina Light-petroleum eluate from alumina 1-8 x 104 column (20 g., Brockmann grade II) with light petroleum. column Fractions from thin-layer chromatography: be detected in the unsaponifiable material by thin- (1) Solvent-front fraction (waxy 90 material) layer chromatography followed by staining with (2) (Stains purple with SbCl3) 150 iodine or antimony trichloride. The R. values of (3) (Stains grey with SbCl3) 100 these materials on a fresh silica-gel-G plate are (4) Squalene 5000 shown in Table 1. (5) Synthetic carrier lycopersene 5 (6) Phytoene 5000 These results are representative not only of (7) Phytofluene 650 analyses carried out on cultures grown under the (8) 8-Carotene fraction (at origin) 6000 conditions described above, but also of those of shake or static cultures grown in the light or in also present, as indicated by its fluorescence and its darkness over the temperature range 29-37o. The absorption maxima at 331, 348 and 367 m,u. only variations are in the amounts of 0-carotene, Carrier synthetic lycopersene was added to this phytofluene and phytoene, which are slightly fraction, and the mixture chromatographed on a increased in cultures grown in the light. There was thin layer of silica gel G (20 cm. x 20 cm.). A no detectable difference in the hydrocarbon con- vertical strip, 1 cm. wide, was removed from the tents of the two strains of N. crassa that were layer with transparent adhesive tape, and this was examined. treated with iodine vapour to reveal the positions Incorporation of DL-[2-14C]mevalonate into ter- of the components. When this was replaced on the penoid hydrocarbons by diphenylamine-inhibited plate, the appropriate sections of the adsorbent N. crassa. N. crassa was grown in the presence of could be scraped off and extracted with ethanol. diphenylamine and 20 pc of [2-14C]mevalonic acid The residue from each zone was extracted three in the light, and the unsaponifiable material was times with 5 ml. quantities of ethanol; the three isolated. The 14C activity of this was assayed and extracts were pooled and evaporated to dryness. then the fraction was chromatographed on an The 14C activity of each of the hydrocarbons was alumina column, as previously described, and the then determined. The results are shown in Table 2. first 200 ml. of the light-petroleum eluate was collected. This was evaporated to dryness, leaving DISCUSSION a yellow, oily residue. The visible absorption maxima in light petroleum indicated that the An analysis of the less polar constituents of the yellow colour was due to the presence of 0-carotene unsaponifiable fraction of two strains of N. crassa, (357, 375, 396-5 and 420 m,). Phytofluene was cultured under conditions of diphenylamine inhi- Vol. 87 LYCOPERSENE IN N. CRASSA 329 bition, failed to reveal the presence of lycopersene such as higher plants and photosynthetic bacteria in this organism, although squalene, phytoene, (Davies et al. 1961; Mercer et al. 1963). In all cases, phytofluene and 0-carotene, and other unidentified phytoene appears to be the first C40 compound hydrocarbons less polar than squalene, are present formed in carotenoid biosynthesis. in appreciable amounts. The incorporation of [2-_4C]mevalonate into squalene and phytoene was SUMMARY demonstrated, and although there was slight in- corporation of radioactivity into certain other 1. A detailed analysis of the less polar hydro- hydrocarbons, no radioactive lycopersene was carbons of the unsaponifiable fraction of Neuro- detected. It seems clear, from these results there- spora crassa, cultured in the presence of diphenyl- fore that lycopersene plays no obvious part in amine, has revealed the presence of squalene, carotenoid biosynthesis in N. crassa. phytoene, phytofluene and 0-carotene. Several This conclusion is at variance with that of Grob unidentified hydrocarbons, less polar than squalene, & Boschetti (1962), who obtained a positive have also been detected, but lycopersene, the C40 identification of squalene and lycopersene as analogue of squalene, was absent. components of a lipid fraction of diphenylamine- 2. [2-_4C]Mevalonic acid was significantly in- inhibited N. crassa. It is difficult to see where this corporated into squalene and phytoene and other discrepancy has arisen. The possibility of a strain tetraterpenes more unsaturated than phytoene by difference has been eliminated since the results N. crassa grown in the presence of diphenylamine. presented above refer not only to our own strain, Labelled lycopersene could not be detected. but also to the wild type (A 10336) that was used 3. These results indicate that in N. crassa, as in by Grob & Boschetti (1962). In their method for other carotenogenic organisms, phytoene, and not detecting lycopersene, these workers did not lycopersene, is the first C40 compound formed in saponify their lipid extract, so the possibility existed carotenoid biosynthesis. that a saponifiable compound which ran with the Thanks are due to Dr 0. Isler (F. Hoffman-La Roche and same RF, or was broken down to a compound which Co. Ltd., Basle, Switzerland) for a sample of synthetic ran with the same RF as lycopersene, was present lycopersene, to Professor B. C. L. Weedon for a sample of in this total lipid fraction. The application of our phytoene, to Dr. I. M. Wilson for a culture of her strain of method of analysis to the total lipid, however, has Neurospora crassa, and to Professor E. C. Grob for kindly failed to reveal any material with RF similar to providing a culture of N. crassa (wild type, A 10336) and that of lycopersene. We had previously shown that for supplying details of his experimental methods before lycopersene is not destroyed by our saponification publication. techniques (Mercer et at. 1963). REFERENCES Anderson & Porter (1962) and Chichester and his co-workers (C. 0. Chichester, personal communi- Anderson, D. G. & Porter, J. W. (1962). Arch. Biochem. Biophys. 97, 509. cation) have failed to find any evidence for the Davies, B. H. (1961). Ph.D. Thesis: University of Wales. participation of lycopersene in carotenoid bio- Davies, B. H. (1962). Biochem. J. 85, 2P. synthesis in tomato plastids and preparations from Davies, B. H. (1963). J. Chromat. (in the Press). P. blakesleeanu8 respectively. However, Grob et al. Davies, B. H., Goodwin, T. W. & Mercer, E. I. (1961). (1961) reported the synthesis of lycopersene in vitro Biochem. J. 81, 40P. from [14C]geranylgeranyl pyrophosphate in a cell- Goodman, DeW. S. & Popjak, G. (1960). J. Lipid Res. 1, free system from N. crassa in the presence of 286. NADPH. Although NADPH is the hydrogen donor Goodwin, T. W. (1959). Advanc. Enzymnol. 21, 295. concerned with the synthesis of squalene from Goodwin, T. W. (1961). Symp. der deutschen Gesellechaft fur Ernihrung8forschung, Mainz. farnesyl pyrophosphate (Goodman & Popjak, Grob, E. C. & Boschetti, A. (1962). Chimia, 16, 15. 1960), NADP+ is the cofactor concerned in the Grob, E. C., Kirschner, K. & Lynen, F. (1961). Chimia, 15, synthesis of phytoene, and NADPH is inhibitory 308. (Anderson & Porter, 1962). If lycopersene is Krzeminski, L. F. & Quackenbush, F. W. (1960a). Arch. indeed formed in the N. crassa system, then its Biochem. Biophys. 88, 64. formation would appear to take place under some- Krzeminski, L. F. & Quackenbush, F. W. (1960b). Arch. what artificial conditions. The separation of Biochem. Biophys. 88, 288. squalene and lycopersene by Grob et al. (1961) was Mercer, E. I., Davies, B. H. & Goodwin, T. W. (1963). carried out with a reversed-phase paper-chromato- Biochem. J. 87, 317. Mitchell, K. H. & Houlahan, M. B. (1946). Fed. Proc. 5, graphy system; this is much less efficient than the 370. thin-layer method now in use. Stahl, E. (1959). Arch. Pharm., Berl., 292, 411. The results described above indicate that, in Turian, G. (1957). Physiol. Plant. 10, 667. tetraterpene formation, there is no difference Turian, G. & Haxo, F. (1952). J. Bact. 63, 690. between fungi and other carotenogenic organisms Zalokar, M. (1957). Arch. Biochem. Biophys. 70, 561.