THE JOURNAL OP BIOLOGKAL CHEMISTRY

Vol. 249, No. 13, Issue of July 10, PP. 4261-4274, 1974

Printed in U.S.A.

Plant Sterol Metabolism *

ENZYMATIC CLEAVAGE OF THE 9/?,19/%CYCLOPROPAKE RIXG OF CYCLOPROPYL STEROLS IT\;

BRAMBLE TISSUE CULTURES

(Received for publication, August 13, 1973)

ROLAND HEINTZ~ AND PIERRE BENVENISTE

From the Laboratoire de Biochimie V&g&ale, Universitt? Louis Pasteur, Institut de Botanique, 67083 Strasbourg-

Cedex, France

SUMMARY

A cell-free enzyme preparation obtained from bramble tissues (Rubus fruticosus) grown in vitro was found to be capable of opening the cyclopropane ring of cycloeucalenol, producing obtusifoliol. The identification of this substance was based on the use of radiochromatography, formation of functional derivatives, and crystallization to constant specific radioactivity. A boiled preparation failed to mediate the reaction, proving that the opening of the 9/3,19/kyclopropane ring is enzymatic. Enzymatic activity was bound to the microsomal fraction and was not dependent upon the addition of exogenous ATP or NADH.

Microsomes from other higher plant species (germinating peas and tobacco tissue cultures) were found to be active in transforming cycloeucalenol into obtusifoliol, but microsomes from rabbit liver were unable to perform this reaction.

24-Methylene cycloartanol, cycloartenol, and cyclo- eucalenol, which are ubiquitous constituents of higher plants, were tested as substrates for the enzymatic reaction. The 4,4-dimethyl sterols are very poor substrates, whereas cycloeucalenol is efficiently converted. In a second group of experiments, three substrates differing only by the number of methyl substituents in position 4 were incubated with the microsomal fraction. Cycloeucalenol (a 4ar-methyl sterol) and 24-methylene pollinastanol (a 4-desmethyl sterol) were converted to obtusifoliol and 4-desmethyl obtusifoliol, respectively. But 24-methylene cycloartanol (a 4,4-di- methyl sterol) was not transformed, which suggests that the 4/3-methyl group hinders this reaction.

On the basis of the results obtained, a tentative mecha- nism for the opening of the cyclopropane ring is proposed.

The analysis and identification of plant sterols (l-4) raise new problems in the understanding of phytosterol biosynthesis: one such problem concerns the nature of the product of 2(3)-oxido-

* This investigation was supported by Grant 71-7-3044 from the Delegation Generale 6 la Recherche Scientifiaue et Techniaue.

$ l%esent address, Abteilung fur Experimentale Medizin, F. Hoffmann-La Roche, Grenzacherstrasse 124, CH-4002 Basel, Switzerland.

squalene cyclization. Following the establishment of 1anosteroP as the precursor of sterols in animals and fungi (5), it was as- sumed that lanosterol would also act as the precursor of higher plant sterols. However, lanosterol has never been identified in any higher plant tissue. Large quantities of lanosterol have been found in the latex of euphorbia species but sterols are not synthesized in the latex (6-8).

By contrast, cycloartenol, 24.methylene cycloartanol, and cy- cloeucalenol are ubiquitously present in photosynthetic plant tissues and this has led various authors to suggest that cycloar- tenol may replace lanosterol in phystosterol biosynthesis and to propose the intermediacy of 9/3,19&cyclopropyl sterols in the biosynthetic pathway leading to phytosterols (9-l 1) (Scheme

1 (11)).

2(3)-oxidosqualene + cycloartenol + 24-methylene cycloartanol + cycloeucalenol + obtusifoliol + 24.methylene lophenol

----+ phytosterols

SCHEME 1

This hypothesis was based on the following experimental re- sults. (a) Cyclopropyl sterols, especially cycloartenol, became rapidly labeled following incubation of plant tissues with [lJ*C]- acetate or [2-Wlmevalonate (12-14) and their specific radioac- tivities, when compared to those of phytosterols, were in agree-

1 The trivial names used are: cycloartenol, 4,4,14~trimethyl- 98,19fl-cycle-5a-cholest-24-en-3fl-oi; lanosterol, 4,4,14a-trimethyl- 5cu-cholesta-8,24-dien-3P-01; parkeol, 4,4,14a-trimethyl-501.cho- lesta-9(11),24-dien-38.01: 24.methvlene cycloartanol, 4,4,14cu- trimethyl-98,19P-cyclo-50(-ergost-2~(28)-en-~~-ol; 24:methylene lanostenol, 4,4,14a-trimethyl-5a-ergosta-8,24(28)-dien-3~-ol; 24- keto cycloartanol, 4,4,14a-trimethy-24-oxo-9/3,19p-cyclo-5~-cho- lestan3&ol; 24-keto lanostenol, 4,4,14~trimethyl-24.oxo-5a- cholest-8-en-30.01; cycloeucalenol, 4~, 14a-dimethyl-9fl,19&cyclo- 5a-ergost-24(28)-en-38-01; obtusifoliol, 4a,l4a-dimethyl-5~1-er- gosta-8,24(28)-dien-3&01; 31-nor-cycloartenol, 4a, 14rr-dimethyl- 9~,19~-cyclo-5a-cholest-24-en-3~-ol~ 31.nor-lanosterol, 4a,14cu- dimethvl-5~-cholesta-8.24-dien-3B-o1: 24-methvlene lonhenol, 4a- methylrsa-ergosta-7,24(28)-dien-&oi; 24.ethylidene- lophenol, 4a-methyl-5~-stigmasta-7,2-24(28)-dien-3p-o1; 24.methylene pol- linastanol, 14a-methyl-9~,19p-cyclo-5a-ergost-24(28)-en-3~-ol; 4- desmethyl obtusifoliol, 14a-methyl-5a-ergosta-8,24(28)-dien-3B- 01; poriferasterol, 24(8)-24-ethyl-cholesta-5,E-22-dien-3P-o1; iso- fucosterol, stigmasta-5,Z-24(28)-dien-3p-01; campesterol, 24(R)- 24.methyl-cholest-5-en-3@-ol; sitosterol, 24(R)-24-ethyl-cholest-5- en-36-01.

by guest on May 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

4268

ment with precursor to product relationships (15). (b) Labeled cvcloartenol has been isolated after anaerobic incubations of different cell-free extracts with labeled 2(3)-oxidosqualene (16-181; in no case was lanosterol detected. (c) When the incu- bations contained also S-adenosylmethionine, labeled 24.meth- ylene cycloartanol was isolated (19). Furthermore, it has been demonstrated with a cell-free extract of peas that cycloartenol could be a substrate for a C-24 transmethylation reaction leading to 24.methylene cycloartanol (20). (d) Finally, labeled cyclo- artenol (21)) 24.methylene cycloartanol, and cycloeucalenol (22) are converted in vivo to phytosterols by tissue cultures of Nico- tiana tabacum (21) and to poriferasterol by the alga Ochromonas

malhamensis (22). If 9/?, 19@-cyclopropyl sterols such as cycloartenol, 24-meth-

ylene cycloartanol, or cycloeucalenol are intermediates in the biosynthesis of plant sterols, then an enzyme capable of opening the cyclopropane ring should be present in plants which would distinguish plants from animals and fungi. Furthermore, the absence of lanosterol and the ubiquitous presence of obtusifoliol together with cycloeucalenol (1) in higher plants would suggest that this enzyme probably acts on cycloeucalenol (23).

We therefore investigated the ability of a cell-free ext’ract from bramble tissues grown in vitro to open the cyclopropane ring of cycloeucalenol and other related substrates. Short preliminary reports on part of this work have been published (24, 25).

EXPERIMENTAL PROCEDURE

Materials

Tissue

Bramble (Rubus fruticosus) tissue cultures were obtained in 1948 by Dr. J. Morel (Versailles, France) from stems and have been described elsewhere (26). These tissues were grown under con- tinuous white light at 25” on a synthetic medium having the follow- ing composition : mineral solution of Heller (26) (100 ml), thiamine (1 mg), glucose (50 g), Bacto-agar (Difco, 12.5 g), and twice dis- tilled water (900 ml). They consist mainly of undifferentiated and heterotrophic cells, capable of growing in the absence of an exogenous source of indoleacetic acid.

Analytical Materials

Conventional techniques for thin layer chromatography were used throughout this study for separation and identification of the substances. Precoated thin layer plates (Merck F 254, 0.25 mm) were generally used. For argentation chromatography, thin layer plates were immersed for 1 min in a solution of 20% AgN03 in ethanol-water (3: 1, v/v) ; after which they were allowed to dry at room temperature for 1 hour; finally they were activated at 110” for 30 min. Thin layer plates were sprayed with ethanolic ber- berin hydrochloride (O.ly,), and the compounds were visualized under ultraviolet light. Radiochromatograms were scanned on a Berthold radiochromatoaram scanner LB 2722. Radioactivitv was determined by liquid scintillation counting with an Inter- technique ABAC, SL-40 counter. Basic alumina (Woehl, grade III) was used for column chromatography.

Gas-liquid chromatography was carried out on a Varian Aero- graph model 1740 gas chromatograph equipped with a flame ioniza- tion detector. Coiled columns (1.50 m X 3 mm) were packed with either 1% OV-17, or 1% SE-30 on 80 to 100 mesh hexamethyldi- silasane-treated Chromosorb W. Mass spectrometry was carried out on a Thomson-Houston THN-208 mass spectrometer at 170” with a direct inlet.

Authentic Materials

Cycloeucalenol was extracted with light petroleum in a Soxhlet apparatus from a piece of eucalyptus wood (Eucalyptus robusta) kindlv suunlied bv Dr. E. B. Huddlerton (Svdnev. Australia), and was purit%d by column chromatography.” The ‘alcohol was re- crystallized from methanol giving needles, m.p. 138-140” (litera- ture m.p. (27) 140”). Gas-liquid chromatography analysis showed

a single peak which had the same retention time as authentic cycloeucalenol. The mass spectrum of the acetate showed a frag- mentation pattern consistent with published data (28) : m/e (rela- tive intensity) 468(10) M+, 408(100) [M - CH,COOH]+, 393(67) [M - CH,COOH - 15]+, 300(13) [M - CH&OOH - CsHiz]+, 283(18) [M - CH,COOH - side chain]+, and 175(29) [M - CH&OOH - C~HU - side chain]+.

Obtusifoliol was extracted with light petroleum in a Soxhlet ap- paratus from the latex of Euphorbia obtusifolia which was kindly supplied by Prof. Gonzales Gonzales (Las Palmas, Las Canarias, Spain). The extracted product was purified by column chroma- tography and was then recrystallized from methanol, m.p. 129- 130” (literature m.p. (29) 144°).2 Gas-liquid chromatography analysis showed a single peak which had the same retention time as an authentic sample of obtusifoliol. The structure was es- tablished by nuclear magnetic resonance spectroscopy.

Cycloartenol and 24.methylene cycloartanol were kindly sup-

plied by Prof. G. Ourisson, Strasbourg, France. Lanosterol was purified from a commercial source by recrystalli-

zation of the acetate from methanol. 24-Methylene lanostenol was synthesized from the 24.enoxide of

lanosteryl acetate. This latter was transformed to 24.ketone by treatment with SnC14 in benzene (30). The 24.keto lanostenvl acetate was then converted into 24-methylene lanostenyl acetate by the method of Bertini et al. (31).

24.Methylene pollinastanol was obtained as described by Knapp et al. (32). Banana peels (Musa sapientum) (10 kg fresh weight) were lyophilized for 1 week. The dried peels were then extracted 3 times with chloroform-methanol (l:l, v/v). After evaporation of the organic solvent, water was added, and the aqueous phase was extracted 3 times with light petroleum. The crude extract was chromatographed first on an alumina column. Elution with 5% diethyl ether in light petroleum gave a sterol ester fraction. After saponification, the unsaponifiable matter was extracted and then chromatographed on an alumina column. The order of elution was: 4,4-dimethyl sterols, 4cu-methyl sterols, and 4-desmethyl sterols. The 4.desmethyl sterol fraction was acetylated and the acetates were separated by argentation chromatography as de- scribed before. One of the two products obtained had the same RF as a 24-methylene derivative. The mass spectrum showed a fragmentation pattern consistent with published data (28).

Radiochemicals

Sodium [3H]borohydride (5 Ci per mmole) and tritiated water (5 Ci per ml) were supplied by the Commissariat a 1’Energie Ato- mique (Gif-sur-Yvette, France).

[d-3H]Cycloeucalenol&Cycloeucalenol (175 mg) was dissolved in 20 ml of acetone-benzene (2:1, v/v) and 0.6 ml of Jones’ reagent was added dropwise while stirring at O-5” under Nz. After 10 min, the reaction mixture was diluted with water and extracted with diethyl ether; the organic phase was dried over NatSO and the solvent was evaporated. The ketone obtained was chromato- graphed on an alumina column, pure cycloeucalenone was eluted with benzene-hexane (1: 1, v/v), and finally recrvstallized from acetone giving needles, m.p. 84-8S”. Cycloeucal”enone (40 mg) was reduced with NaBaH (20 pmoles) in anhydrous diethyl ether (previously distilled over CaH2) under Nz for 1 hour. The result- ing cycloeucalenol was extracted, purified by thin layer chroma- tography, and recrystallized from methanol. The specific radio- activity of the product was adjusted to about 100&i per pmole and the labeled product was recrystallized again twice from methanol; the specific radioactivities of crystals from both first and second recrystallization and from the corresponding mother liquors were shown to be 100 PCi per pmole. The purity of the labeled CYC~O-

eucalenol was greater than 99% on the basis of gas-liquid chroma- tography.

[9-3H]CycZoartenoZ-The same procedure as for cycloeucalenol was used. The specific radioactivity thus obtained was about 65 j&i per Mmole.

d4-Methylene-[2,4-aH]pollinastanol-24-Methylene pollinastanol was oxidized to 24-methylene pollinastanone by action of the Jones reagent as described before. The ketone was first purified by thin

2 The discrepancy between the meeting point reported in the literature and ours for obtusifoliol could be explained by the pres- ence of a slight amount (less than 370 in our case, as shown by gas- liquid chromatography) of euphorbol (27).

by guest on May 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

4269

layer chromatography (cyclohexane-ethyl acetate (90:10, v/v)) and then recrystallized twice from methanol. The purified ketone was tritiated by the method of Klein and Knight (33). The ketone (2 mg) was dissolved in a small volume of benzene-hexane (1: 1, v/v), applied to a column of basic alumina previously deactivated with tritiated water (5 Ci per ml), and left on the alumina for 90 min; the tritiated ketone was then eluted with benzene-hexane (1: 1, v/v) and reduced by LiAlH, in anhydrous diethyl ether giv- ing a mixture of 36.OH and 3m-OH isomers. The 3fl-OH isomer (RF 0.40) and the 3ol-OH isomer (RF 0.48) were easily separated by thin layer chromatography (cyclohexane-ethyl acetate (90: 10, v/v), two migrations). The 24-methylene-[2,4-3H]pollinastanol (3p-OH) thus obtained was more than 98% pure by gas-liquid chromatography analysis. The specific radioactivity thus ob- tained was 39 &i per pmole.

[2V4C]24-Methylene cycloartanol was kindly prepared by Dr. R. Labriola. Buenos Aires. Argentina. The 24.euoxide of cvclo- artenyl aceiate was treated wirh SnC14 in dry beizene (30). “The 24.ketocycloartanyl acetate obtained was then treated with “C- labeled Wittig reagent, as described elsewhere (34), giving the expected 24.methylene cycloartanol with a specific radioactivity of 25 PCi per rmole.

Methods

Preparation of Subcellular Fractions

Bramble tissues (50 g fresh weight) were homogenized in a loosely fitting Elvehjem-Potter Teflon glass homogenizer in the presence of twice their weight of a buffer containing: 0.25 M sac- charose, 0.004 M MgC12, 0.005 M cysteine, and 0.1 M Tris-HCl (pH 7.6). The extract was filtered through four layers of muslin, and the filtrate was passed through nylon (Blutex, Tripette et Renaud, 50 Mm). An initial centrifugatibn at 750 X.g for 10 min gave a pellet of crude nuclei and olastids (Cl) and a suoernatant. This aas centrifuged at 6,000 X g for 10 min giving a crude mitochon- drial pellet (Cc) and a second supernatant which was further cen- trifuged at 120,000 X g for 90 min to give the microsomal pellet (C120) and a supernatant (S,s,,). This was used for the preparation of a soluble supernatant fraction. The pellets (Cl, Cg, and Cl,,,) were resuspended in the following medium: 2 mM Mg2+, 2 mM mercaptoethanol, 20 mM Tris-HCl (pH 7.7) (Medium A).

Preparation of Soluble Supernatant Fraction

Protein of the supernatant S 120 was fractionated with solid am- monium sulfate. The fraction which precipitates between 45 and 80% of saturation was dissolved in a minimal volume of Medium A and dialyzed against Medium A for 12 hours. Glycerol (20yo, v/v) was finally added to the dialyzed enzymes. The mixture was stored at -20”.

Activity Assays

Assay mixtures contained 20 mM Tris-HCI (pH 7.6)) 6 mM Mg2+, 6 mM K+, particles, supernatant fraction, and finally the substrate was dissolved in 50 ~1 of a solution of 0.5oj, Tween-80 in acetone. Incubations were carried out for 5 hours at 31” after addition of the labeled substrates and terminated by addition of 10% KOH in ethanol. In each experiment, a duplicate consisting of particles boiled for 30 min was used.

Analytical Procedures

After incubation, the reaction mixture was extracted by ultra- sonication for 15 s (Ultrasons, Annemasse, France; Y = 26 kHz, power = 400 watts) in the presence of 3 volumes of light petroleum. The organic phase was then decanted; this procedure was repeated three times. The three extracts were combined, washed with ethanol-water (1: 1, v/v), dried over NatSOd, and evaporated under reduced pressure. The residue was subjected to thin layer chro- matography (ethyl acetate-cyclohexane (10:90, v/v), two migra- tions). Five bands were seen under ultraviolet light and identi- fied in order of increasing polarity as: squalene and esters, 2(3)- oxidosqualene, 4,4-dimethyl sterols, 4ol-methyl sterols, and 4-des- methyl sterols. Each band was scraped off and the product eluted from the silica with either diethyl ether or methylene chloride. After solvent evaporation, an aliquot was transferred to a scintillation vial and counted.

The alcohol fractions were acetylated at room temperature for

16 hours by 50 ~1 of pyridine and 200 ~1 of acetic anhydride. The reaction mixture was then lyophilized and the acetates were chro- matographed first on thin layer plates (ethyl acetate-cyclohexane (5:95, v/v)) after addition of pure unlabeled products: cyclo- artenyl acetate (0.5 mg), lanosteryl acetate (0.5 mg), and 24. methylene lanostenyl acetate (0.5 mg) were added to the acetates of the labeled 4,4-dimethyl sterols; cycloeucalenyl acetate (0.5 mg) and obtusifoliyl acetate (0.5 mg) were added to the acetates of the labeled 4ol-methyl sterols; finally 24.methylene pollinastanyl ace- tate (0.5 mg) and obtusifoliyl acetate (0.5 mg) were added to the acetates of the labeled 4.desmethyl sterols. The 8,24(28)-diepox- ide of obtusifoliyl acetate has, under our conditions, the same chromatographic properties as the diepoxide of 4-desmethyl obtu- sifoliyl acetate.

4,4-Dimethyl Sterols-The mixture of acetates was chromato- graphed a second time by argentation chromatography as de- scribed under “Materials” (benzene-cyclohexane (40:60, v/v), continuous migration for 17 to 41 hours). Three fractions were obtained corresponding to cycloartenyl + lanosteryl acetates (fastest migrating fraction), parkeyl acetate, and 24.methylene cycloartanyl + 24-methylene lanostenyl acetates (slowest migrat- ing fraction). These fractions were analyzed by gas-liquid chro- matography and by mass spectrometry as previously described (12). The partially purified acetates were epoxidized with p-nitroperbenzoic acid (20 mg) in anhydrous diethyl ether (0.5 ml) as described earlier (12). The reaction was carried out at room temperature overnight and was stopped by addition of an aaueous solution of Na&O. (5% w/v). Enoxides of 4.4.dimethvl

- “~I ” I

steryl acetates were extracted 3 times with’ diethyl ether and were chromatographed on thin layer plates (ethyl acetate, cyclohexane (10:90, v/v), two migrations). In these conditions, the 24(28)- epoxide of 24.methylene cycloartanyl acetate could be easily separated from the 8,24(28)-diepoxide of 24-methylene lanostenyl acetate, and the 24.epoxide of cycloartenyl acetate could be separated from the 8,24-diepoxide of lanosteryl acetate.

&-Methyl Sterols-These were treated as the 4,4-dimethyl sterols. After argentation chromatography, three fractions were obtained corresponding to 24-ethylidene lophenyl acetate3 (fastest migrating fraction), cycloeucalenyl + obtusifoliyl acetate, and 24-methylene lophenyl acetate 3 (slowest migrating fraction). After epoxidation and chromatography of the epoxides of 4a- methyl steryl acetates, the 24(28)-epoxide of cycloeucalenyl acetate could be separated from the 8,24(28)-diepoxide of obtusi- foliyl acetate.

.&Desmethyl Sterols-After argentation chromatography, three fractions were obtained, corresponding to sitosteryl + campes- teryl acetates3 (fastest migrating fraction), isofucosteryl acetate3 and 24-methylene pollinastanyl + 4-desmethyl obtusifoliyl ace- tate (slowest migrating fraction). Using the same techniques as described before, the 24(28)-epoxide of 24-methylene pollinastanyl acetate could be easily separated from the 8,24(28)-diepoxide of 4-desmethyl obtusifoliyl acetate.

In most cases, the analytical procedure was completed by crys- tallization of the purified epoxides of steryl acetates to a constant specific radioactivity.

RESULTS

Evidence for Enzymatic Opening of Cyclopropane Ring of Cycloeucalenol

A supernatant (S1) and a control consisting of an identical

enzyme preparation inactivated by heating at 100” for 30 min were incubated with [3H]cycloeucalenol (Table I). After thin

layer chromatography of the nonsaponifiable fraction, which separates 4,4-dimethyl sterols (RF 0.45) from 4cu-methyl sterols (RF 0.40) and from 4-desmethyl sterols (RF 0.25), radioscanning showed the presence of only one peak, with an RF value corre- sponding to that of 4cu-methyl sterols. After acetylations, the 4cu-methylsteryl acetates were separated by argentation chro- matography into 24-methylene lophenyl acetatea (RF 0.35), cy-

3 24.Methylene lophenol, 24.ethylidene lophenol, sitosterol, campesterol, and isofucosterol have been shown to be present in bramble microsomes (M. A. Hartmann, unpublished results).

by guest on May 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

4270

TABLE I

Enzymatic opening of 9p, 19@-cyclopropane ring of cycloeucalenol

A crude supernatant Si (10 ml) of bramble homogenate was used in the experiment; the assay mixture contained 0.25 M sucrose, 4 mM Mg2+, 5 mM cysteine, 0.1 M Tris-HCl (pH 7.6), and 5 pM [3H]- cycloeucalenol.

TABLE III

Intracellular localization of enzymatic activity

Assays were performed under standard conditions; particles (2 ml = 3.4 mg of proteins) were incubated in the presence of 10 pM

[3H]cycloeucalenol and, where indicated, soluble supernatant fraction (2 ml = 2 mg of proteins).

Experiment (Si). 10 X

Control (boiled S,)...... 11 X

TABLE II

Variation of specific radioactivity of diepoxide of obtusifoliyl acetate during recrystallization

Carrier diepoxide of obtusifoliyl acetate (11.75 mg) was added to the labeled diepoxide of obtusifoliyl acetate (8.5 X lo5 dpm). The specific radioactivity of the crude mixture before recrystal- lization was 72,000 f 4,000 dpm per mg.

Crystallization

dPm/mg

Specific radioactivity in the diepoxide of obtu- sifoliyl acetate. 66,000 63,000 60,000 61,000

f 4,000 f 4,000 f 3,500 f 3,500 Specific radioactivity in

crystals from the mother liquor. 75,000 64,000 61,500

f 4,500 f 4,000 f 3,500

cloeucalenyl + obtusifoliyl acetates (RF 0.40) and 24.ethylidene lophenyl acetate3 (RF 0.50). Radioscanning showed activit,y in only one peak which corresponded to cycloeucalenyl + obtusi- foliyl acetates. No radioactivity was found in the areas of 24. methylene or 24.ethylidene lophenyl acetate. After epoxidation of the mixture of cycloeucalenyl + obtusifoliyl acetates and sub- sequent chromatography, the epoxides were easily resolved into the 24(28)-epoxide of cycloeucalenyl acetate (RF 0.50) and the 8,24(28)-diepoxide of obtusifoliyl acetate (RF 0.35). Radio- scanning of the epoxides from the active enzyme preparation (Si) showed two peaks of radioactivity corresponding to the mono- epoxide of cycloeucalenyl acetate and the diepoxide of obtusi- foliyl acetate, whereas radioscanning of the control acetate epoxides showed only one peak of radioactivity with the same RF as the monoepoxide of cycloeucalenyl acetate. Quantitative data shown in Table I clearly demonstrate that an important transformation of cycloeucalenol into obtusifoliol has taken place with the active enzyme preparation, the diepoxide of obtusifoliyl acetate containing 16% of the total radioactivity. Compared to this, the control experiments with an inactivated enzyme prep- aration contained less than 0.02% of the total radioactivity at the RF of diepoxide of obtusifoliyl acetate, making the recrystalli- zation of diepoxide of obtusifoliyl acetate from the control un- necessary. The diepoxide of obtusifoliyl acetate from the active enzyme preparation was further purified by recrystallization from methanol after addition of unlabeled carrier; the results are given in Table II. The radioactivity of crystals of diepoxide of obtusifoliyl acetate was constant during the last three recrystalli- zations and equal to that found in the mother liquor.

Fraction Diepoxide of obtusifoliyl acetate

Cl CS C 120

C~ZO + soluble supernatant fraction Soluble supernatant fraction alone

nmoles/ng of membrane pO&?U

0.22 0.30 2.75 2.85 0.20

Intracellular Localization of Enzymatic Activity

Subcellular fractions were characterized as described else- where (35). Electron microscopy of sectioned embedded pellets showed that Fraction Ci contained a mixture of nuclei, chromo- plasts, and unidentified vesicles, Fraction Ca consisted essen- tially of mitochondria, and Fraction C&O (microsomes) contained free ribosomes and smooth and rough vesicles coming from endo- plasmic reticulum. These different fractions were incubated under the same conditions with [3H]cycloeucalenol. The results (Table III) show that the highest enzymatic activity was associ- ated with microsomes and that the soluble supernatant fraction showed only a very low activity and was unable to stimulate significantly the activity of the microsomal fraction.

As microsomes were by far the most active fraction, they were used as a source of enzyme. It should be noted that the enzyme was tightly bound to the microsomal fraction. Cycloeucalenol and related substrates are also completely insoluble in water. These two facts could explain the relatively high standard errors. For this reason only differences greater than 30 ‘% have been con- sidered as significant in the work described below.

Cofactors Required for Enzymatic Cleavage of Cyclopropane Ring of Cycloeucalenol

Microsomes (2 ml) corresponding to 3.5 mg of proteins were incubated under standard conditions and in the presence of ATP (4 mM) and NADH (0.5 mM). No significant difference (+25%) in obtusifoliol formation could be observed whether ATP and NADH were present or not. Thus ATP and NADH do not seem to play any role in the cyclopropane ring-opening reaction.

Substrate Specijicity of Enzyme

Behavior of Cyclopropyl Sterols Present Ubiquitously in Higher Plants Toward Enzyme-In most plant tissues, cycloartenol and 24-methylene cycloartanol are present together with cycloeucal- enol. A biogenetic scheme leading to phytosterols (15) postu- lated that these three compounds could be substrates of a cyclo- propane ring opening enzyme. With this in mind, we have com- pared the ability of these compounds to act as substrates for the present enzyme.

Table IV shows the results obtained after an incubation carried out with cycloartenol, 24.methylene cycloartanol and cycloeucal- enol. The radioactivities associated with the diepoxide of lanosteryl acetate and the diepoxide of 24.methylene lanostenyl acetate which could be expected to be formed by opening the cyclopropane ring of cycloartenol and 24-methylene cycloar- tanol, respectively, were very low compared to the radioactivities

by guest on May 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

4271

TABLE IV

Comparative incubations of cycloartenol, 24.methylene cycloartanol and cycloeucalenol with bramble microsomes

Microsomes (3.6 ml = 6 mg of proteins) were incubated under standard conditions.

Incubation

Radioactivity in the substrate (dpm). ...................... Substrate concentration (PM). ............................... Radioactivity in diepoxides (dpm). ....................... Amount of carrier product added to labeled diepoxides (mg) Specific radioactivity in crystals during recrystallization (dpm/

w) l........................................................ 2 ........................................................ 3 ... .... ............................ .................. 4 ................................ .............

Total radioactivity (dpm). ................................ Amount of substrate converted (nmoles). ....................

a Radioactivity in diepoxide of lanosteryl acetate. b Radioactivity in diepoxide of 24.methylene lanosteryl acetate. c Radioactivity in diepoxide of obtusifoliyl acetate. d Diepoxide of lanosteryl acetate e Diepoxide of 24-methylene lanostenyl acetate. 1 Diepoxide of obtusifoliyl acetate.

Cycloartenol

3 x 106 5.5

0.015 x 1060 14.5d

580 f 30 320 f 20 250 f 20 260 f 20

3,625 0.025

24.Methylene cycloartanol

29.1 x 106 140

0.033 x 10Q 14e

1,200 f 50 500 f 25 185 f 15 50 f 5 < 750

<0.013

Cycloeucalenol

6 x 106 7.5

0.48 x 106~ 9f

51,500 f 2,500 50,000 f 2,500 51,000 f 2,500 51,060 f 2,500

450,000 2.08

associated with the diepoxide of obtusifoliyl acetate. The specific radioactivities associated with the crystals of either the diepoxide of lanosteryl acet,ate or the diepoxide of 24.methylene lanostenyl acetate dropped rapidly to a very low value after the fourth recrystallization, whereas that of the diepoxide of ob- tusifoliyl acetate showed only a relatively small decrease before reaching a constant value. Taking the value after the fourth crystallization, the quantity of radioactivity in the crystals, cor- responding to the labeled product formed as a result of the en- zymatic reaction, could be calculated. In the case of the di- epoxide of lanosteryl acetate, this value (3,625 dpm) represented a conversion of 0.025 nmoles of the added substrate; in the case of the diepoxide of 24.methylene lanostenyl acetate, this value (750 dpm) represented a very low conversion of 0.013 nmole of the added substrate. In contrast, in the case of the diepoxide of obtusifoliyl acetate, this value (450,000 dpm) represented a rela- tively important conversion of 2.1 nmoles. The results demon- strate that of the three cyclopropyl sterols ubiquitously present in higher plants, only cycloeucalenol was efficiently attacked.

Injluence oj Number and Position of Methyl Groups on Carbon &The results presented in the preceding paragraph suggested that the number of methyl groups on carbon 4 would have an important effect on the rate of the enzymatic reaction. One of the main structural difference between cycloeucalenol, on one hand, and 24.methylene cycloartanol and cycloartenol, on the other, is the absence in cycloeucalenol of the 4cy-methyl group. With this in mind, a second set of experiments was performed with the aim of comparing the rate of conversion of three cyclo- propyl sterols differing only by the number and the position of methyl groups on carbon 4; i.e. 24-methylene cycloartanol (two methyl groups on carbon 4), cycloeucalenol (one 4cu-methyl group) and 24.methylene pollinastanol (no methyl group on carbon 4). Results presented in Table V show clearly that while 24.methylene cycloartanol was not converted to 24-methylene lanostenol, cycloeucalenol and 24-methylene pollinastanol were converted at about the same rate to obtusifoliol and a product with the chromatographic behavior of 4-desmethyl obtusifoliol, respectively. Complete identification of the latter was not possible since authentic 4-desmethyl obtusifoliol was not availa-

TABLE V

Influence of number of methyl groups in four on enzymatic opening of 98,19@-cyclopropane ring; comparative incubations of

24.methylene cycloartanol, cycloeucalenol, and 2Q-methylene pollinastanol

Microsomes (0.6 ml) corresponding to 1 mg of proteins were

Substrate

incubated under standard conditions.

Radioac- Substrate tivity of concen-

substrate tration

24.Methylene cyclo- artanol.

Cycloeucalenol 24.Methylene pollinas-

tan01

dgnt PJf

2.4 x lo6 75 4.0 x 106 30

2.1 x 106 40 -

- Radioactivity

in 8,24(28)- diepxides

&‘m

103a 239 x lo@

88 x 103’

Amount of substrate converted

nmoles/mg of proteins

0.005 1.10

1.05

a Radioactivity in 8,24(28)-diepoxide of 24.methylene lanos- tenyl acetate.

b Radioactivity in 8,24(28)-diepoxide of obtusifoliyl acetate. c Radioactivity in 8,24(28)-diepoxide of 4-desmethyl obtu-

sifoliyl acetate.

ble in amounts sufficient for recrystallizations to constant specific radioactivity.

Presence of Cyclopropane Ring-opening Enzyme in Other Higher Plant Species

Comparison with Cell-free Extract of Liver-A comparative study of the enzymatic opening of the cyclopropane ring of cyclo- eucalenol was made by the use of microsomes prepared from bramble tissues grown in vitro, tobacco tissues grown in vitro,

germinating pea seeds, and rabbit liver. Table VI shows that all the cell-free enzyme preparations of higher plants were able to transform cycloeucalenol to obtusifoliol, whereas rabbit mi- crosomes were unable to perform this reaction: in this latter case no difference in the rate of conversion of cycloeucalenol into obtusifoliol could be detected between active microsomes (exper- imental) and boiled microsomes (control).

by guest on May 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

4272

TABLE VI

Comparison of cycloeucalenol-obtusifoliol isomerase activity in microsomes obtained from different organisms

Organism

Bramble tissue cul- tures.

Germinating pea seeds.............

Tobacco tissue cul- tures.

Rabbit liver.

dpm PM

9.4 x 106 4.7

13 x 106 13

6.4 X lo6 8.2 4.9 x lo6 11.6

Radioactivity Amount of n diepoxide of obtusifoliyl substrate

acetate converted

dgm

,120 x 103 1.5

180 x 103 0.023

160 X lo3 0.12 <103 <O.OOl

DISCUSSION

Evidence has been given for the presence of an enzyme capable of opening the cyclopropane ring of cycloeucalenol in a cell-free extract of bramble tissue cultures. This enzyme transforms cycloeucalenol into obtusifoliol. Demonstration of this enzy- matic activity in cell-free extracts prepared from plant tissues belonging to three different families (rosaceae, solanaceae, and legumes), support the generality of this reaction in plants.4

The enzymatic activity is tightly bound to microsomal mem- branes, and soluble supernatant enzymes are unable to stimulate significantly the activity of microsomes. Thus it could be con- cluded that a soluble protein factor such as that involved in rat liver squalene epoxidase (37)) or such as the sterol carrier protein described in many other steps of cholesterol biosynthesis (38), does not seem to be involved here.

Studies of the enzyme specificity toward natural cyclopropyl sterols possessing a different number of methyl groups on carbon 4 have surprisingly shown that 4,4-dimethyl cyclopropyl sterols (24-methylene cycloartanol and cycloartenol) are very poor sub- strates, whereas cycloeucalenol which is a 4a-methyl sterol is a good substrate as is 24-methylene pollinastanol, a 4-desmethyl cyclopropyl sterol. This suggests that the presence in 4,4-d- methyl cyclopropyl sterols of a 4@methyl group hinders the action of the enzyme in some way, either by interfering with the binding of enzyme and substrate or by hindering the approach of a chemical group essential for the opening of the cyclopropane ring.

This enzyme operates without addition of exogenous ATP or NADH, in agreement with a hypothetical reaction mechanism equivalent to an acid-catalyzed isomerization like that involved in the cyclization of 2(3)-oxidosqualene. Biosyn- thesis of cycloartenol involves formation of the 9p,l9&cyclopro- pane ring; cleavage of this same ring occurs during obtusifoliol formation. The similarities existing between cyclopropane ring formation and cleavage leads to the suggestion that the two reac- tions would involve the closely related intermediates (111 and VII) as shown in Scheme 2. Cyclization of 2(3)-oxidosqualene into cation II is initiated by a proton or its biochemical equiva- lent. Cation II is then stabilized by methyl and hydride back- bone 1,2-rearrangements terminated by neutralization of the in- termediary cation at C-9 by a suitable nucleophilic Group B of

4 It has been shown (36) that a latex from Euphorbia latyris can transform cycloartenol into lanosterol. This result is consistent with the presence of both cycloartenol and lanosterol in this sys- tem; it would imply the involvement of a different reaction mecha- nism from that proposed in this article to explain the opening of the cyclopropane ring.

the enzyme, giving Compound III as proposed previously by Rees et al. (13). Specific attack and elimination of a proton at C-19 by a basic radical of the enzyme leads to the formation of the cyclopropane ring and to the subsequent tram-elimination of radical B, giving cycloartenol IV (in that case, truns-elimination of proton 8 and radical B which give lanosterol would be forbid- den). After two more enzymatic transformations (24-methyla- tion and 4-demethylation), cycloartenol gives cycloeucalenol VI which is a 4a-methyl compound. By a “push-pull” mechanism involving perhaps the same radicals A and B of the enzymatic system, cycloeucalenol would be transformed into intermediate VII which could be identical with III but without a 4&methyl. In this last intermediate, truns-elimination of proton 8 and radical B would be possible and would be performed by another radical C, leading to obtusifoliol VIII (23).

The opening of the cyclopropane ring of cycloartenol involves unfavorable 4j%methyl, lo-methyl interactions, i.e. high transi- tion state energy, in intermediate III which do not exist in the case of intermediate VII derived from cycloeucalenol. There- fore, from a thermodynamic and kinetic point of view, one would expect the opening of the cyclopropane ring to be more favorable in the case of cycloeucalenol compared to cycloartenol. Another critical step in this hypothetical scheme would be the proton elimination from the intermediates III and VII. This operation could be induced by a specific active group of the enzymatic pro- tein and would also be controlled by specific structural charac- teristics of the substrate. Possibilities are the number of methyl groups at C-4 and their stereochemistry. Our laboratory is in- vestigating the influence of the stereochemistry at the C-4 posi- tion by using the 4@-methyl isomer of cycloeucalenol.

Microsomes prepared from rabbit liver were unable to open the cyclopropane ring of cycloeucalenol and to transform it into obtusifoliol. As cycloeucalenol was shown to be a suitable sub- strate for testing the presence of a cyclopropane ring-opening enzyme, our results suggest that this enzyme may be restricted to higher plants or perhaps more generally to photosynthetic eu- karyot,es.5 Gibbons et al. (39) reached the same conclusion in showing that a cell-free extract from rat liver did not transform cycloartenol to lanosterol.

Finally, as neither cycloartenol nor 24.methylene cycloartanol is a substrate for the enzyme, it is possible to rule out lanosterol and 24.methylene lanostenol from the major phytosterol path- way. The results of the present study support a modification of a previous scheme (15) in which it was proposed that lanosterol and 24.methylene lanostenol were intermediates. It is suggested (Scheme 3) that a branch point exists at the level of cycloartanol, which may be methylated to give 24.methylene cycloartanol following the pathway postulated by other workers (ll), or demethylated to give 31.nor-cycloartenol. This latter compound may be methylated to give cycloeucalenol or may also be a sub- strate for the enzyme discussed here leading to 31-nor-lanosterol.6 The passage from the pathway represented by broken lines to the pathway represented by full lines involves the reaction of C-24

methylation which has been demonstrated in vitro with a micro- somal system from bramble tissue cultures (19) and with a cell- free extract of pea seedlings (20). Finally the two parallel routes converge at obtusifoliol. This scheme is in agreement with the kinetic results (15)) with the absence of lanosterol and the occur-

5 It has been recently shown in our laboratory that a yeast cell- free extract cannot open the cyclopropane ring of any 4,4-di- methyl, 4a-methyl, and 4-desmethyl cyclopropyl sterol (C. And- ing, L. Parks, and G. Ourisson, unpublished results).

6 R. Heintz, unpublished results.

by guest on May 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

4273

1ANOSlEROL

c

OH

R

M.

(I)

Z(3)-OXIDOSQUALENE

24-METHYLENE CYCLOARTANOL~ CYCLOARTENOL

(V) +24M* (IV)\

1 ” ) WI R _ H Ii* R&A.

CYCLOEUCALENOl

m + 19H

WI I I, I Ii R2* M*

OBTUSIFOLIOL

WIllI

SCHEME 2

: 31-NOR CYCLOAATENOL H-NdRLANOSTEROL

SCHEME 3

PHYTOSTEROLS

by guest on May 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

4274

rence of cycloeucalenol and obtusifoliol together in higher plants 18. and with the identification in some species of 31.nor-cycloartenol (40) and 31.nor-lanosterol (41); relevant to this last point, the

19.

identification of cholesterol in all plant tissues studied up to now 20. supports the idea that 31.nor-cycloartenol and 31-nor-lanost,erol are ubiquitously distributed in higher plant kingdom. 21.

22. Acknowledgmenfs-We are indebted to Mrs. P. Schmitt and

Mrs. P. Geoffroy for their expert technical assistance. We thank Dr F. Schuber‘for helpful &scussions.

REFERENCES

1

2.

3

4.

5. 6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

GOAD, L. J., AND GOODWIN, T. W. (1972) in Progress in Phyto- chemistry (RHEINHOLD, L., AND LIWSCHITZ, Y., eds) Vol. III, p. 113, Interscience, London

BENVENISTE, P., HIRTH, L., AND OURISSON, G. (1964) C. R. Hebd. Seances Acad. Sci. Ser. D Sci. Nat. (Paris) 269, 2284-2287

SCHREIBER, K., AND OSSKE, G. (1962) Kulturpjfanze 10, 372- 382

KNAPP, F. F., AND NICHOLAS, H. J. (1969) Phytochemistry 8, 207-214

CLAYTON, R. B. (1965) Quat. Rev. 19, 168-200 GONZALEZ. A. G.. AND MORA. M. C. G. (1952) An. Real. SOC.

23.

24.

25.

26.

27.

28.

29.

Esp. Fis. Quik. 48B, 475 ’ ~

PONSINET, G., AND OURISSON, G. (1967) Phytochemistry 6, 1235-1242

PONSINET, G., AND OURISSON, G. (1968) Phytochemistry 7, 89-98

BENVENISTE, P., HIRTH, L., AND OURISSON, G. (1965) Bull. Sot. FT. Physiol. V6g. 11, 252-268

VON ARDENNE, M., OSSKE, G., SCHREIBER, K., STEINFELDER, K.. AND TUMMLER. R. (19651 Die Kulturvflanze 13. 115-129

GOA;), L. J., AND GOODWIN, ‘I!. W. (1966) kochem. j. 99, 735- 746

BENVENISTE, P., HIRTH, L., AND OURISSON, G. (1966) Phyto- chemistry 6, 45-58

REES, H. H., GOAD, L. J., AND GOODWIN, T. W. (1968) Bio- them. J. 107, 417-426

KNAPP, F. F., AND NICHOLAS, H. J. (1971) Phytochemistry 10, 85-95

BENVENISTE, P., HEWLINS, M. J. E., AND FRITIG, B. (1969) Eur. J. Biochem. 9, 526-533

REES, H. H., GOAD, L. J., AND GOODWIN, T. W. (1968) Tetra- hedron Lett. 723-725

HEINTZ, R., AND BENVENISTE, P. (1970) Phytochemistry 9, 1499-1503

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

REES, H. H., GOAD, L. J., AND GOODWIN, T. W. (1969) Bio- chim. Biophys. Acta 176, 892-894

HEINTZ, R., AND BENVENISTE, P. (1972) C. R. Hebd. Seances Acad. Sci. Ser. D Sci. Nat. (Paris) 274, 947-950

RUSSELL, P. T., VAN ALLER, R. T., AND NES, W. R. (1967) J. Biol. Chem. 242, 5802-5806

HEWLINS, M. J. E., EHRHARDT, J. D., HIRTH, L., AND OURIS- SON, G. (1969) Eur. J. Biochem. 8, 184-188

LENTON, J. R., HALL, J., SMITH, A. R. H., GHISALBERTI, E. L., REES, H. H., GOAD, L. J., AND GOODWIN, T. W. (1971) Arch. Biochem. Biophys. 143, 664-674

GOAD, L. J., WILLIAMS, B. L., AND GOODWIN, T. W. (1967) Eur. J. Biochem. 3, 232-236

HEINTZ, R., BENVENISTE, P., AND BIMPSON, T. (1972) Bio- them. Biophys. Res. Commun. 46, 766-772

HEINTZ, R., BIMPSON, T., AND BENVENISTE, P. (1972) Biochem. Biophys. Res. Corn&n. 49, 820-826

GAUTHERF,T. R. J. (1959) La culture des Tissus VBabtaux. D. 52, Masson, Paris. ’

I A

OURISSON, G., CRABBE, P., AND RODIG, 0. It. (1964) Tetra- cyclic Triterpenes, Hermann, Leeds

AUDIER. H. E.. BEUGELMANS. It.. AND DAS, B. C. (1966) Tetra- hedron Lett. 4341-4347 ’

BERMEJO BARRERA. J.. BRF,TON. J. L.. DELGADO MARTIN. J.. AND GONZALES, A. G. (1967) An. Real Sot. Espan. Fis. Quim. 63, 191

LABRIOLA, R., AND OURISSON, G. (1971) Tetrahedron 27, 1901- 1908

BERTINI, F., GRASSELLI, P., AND ZUBUNI, G. (1970) Tetra- hedron 26, 1281-1290

KNAPP, F. F., PHILLIPS, D. O., GOAD, L. J., AND GOODWIN, T. W. (1972) Phytochemistry 11, 3497-3501

KLEIN, P. D., AND KNIGHT, J. C. (1965) J. Amer. Chem. Sot. 87, 2657

WRZECIONO. U., MURPHY, C. F., OURISSON, G., LHOMMF,, M. F., EHRH~RDT, J. fi., COR~ANO, S., A&D TELLER, G: (1970) Bull. Sot. Chim. 966-974

HARTMANN, M. A., FERNE, M., GIGOT, C., BRANDT, R., AND BENVENISTE, P. (1973) Physiol. Vbg. 11, 209-230

PONSINET, G., AND OURISSON, G. (1968) Phytochemistry 7, 757-764

TAI, H.-H., AND BLOCH, K. (1972) J. Biol. Chem. 247, 3767- 3773

SCALLEN, T. J., SCHUSTER, M. W., AND DH~R, A. K. (1971) J. Biol. Chem. 246, 224-230

GIBBONS, G. F., GOAD, L. J., GOODWIN, T. W., AND NES, W. 11. (1971) J. Biol. Chem. 246, 3967-3976

DF,VYS, M., ALCAIDE, A., PINTE, F., AND BARBIER, M. (1970) Tetrahedron Lett. 63, 4621-4622

REES, H. H., GOAI), L. J., AND GOODWIN, T. W. (1968) Phyto- chemistry 7, 1875-1877

by guest on May 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Roland Heintz and Pierre BenvenisteTISSUE CULTURES

-CYCLOPROPANE RING OF CYCLOPROPYL STEROLS IN BRAMBLEβ,19βPlant Sterol Metabolism: ENZYMATIC CLEAVAGE OF THE 9

1974, 249:4267-4274.J. Biol. Chem. 

  http://www.jbc.org/content/249/13/4267Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/249/13/4267.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on May 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from