jie bluee - iupac.org fl9't2-4jtp68 twa9'lt9'ple wtcjcueaa aiw tjjtcj1 i eoogeq w2...

tQ 3vvc j\—c -

9'0JT4 4JJ6 09'L046Uoqe p1 L0868'o; .pjie 00TT894T01T 946b TI' we 09'L046U0qe ug TI' 4J36 I9'4 26040U e2 90We4J3TJJ4JJ6 9'peO]-iT.çe 0011;nL9'.çou2' 4}J6LT 0 I140 4JJ6 bLOpTeWB o; jJe 846L60GJJ6W8çL2TOLTb wGII W611fl011 9' ;GA CJ.T6Wf 09']' L9'u8;oLnJ9'fl-ou9' 46L T8 I TIT q801T98I AT]']' abe9'Jc TL24 9'p0n.ç 116A C9'L0.çeU0q8 9TJ( AT]']' 0 pço W6TL 019'L9'C.çeL99'ç—

9'JB0 40 IJC]'JTq6 I'GA68.ç 9'IIg iI1biTp]''8JJeg W9'46LT9'J'bLUee 30A6A6L o peii 9'PT6 40 pru 2oc 80we.çJfl-u W0L6 C0IIJbL6JJ6U8T6 9'IJq69'0JJ 2e9'L oog bLOLe8a L6b0L42 9'bbe9'L TI' we 2b60T9'TT84 6b0L.ç8 I JJ9'e weLL0L6LT2 8b69'Jqu we C9'LO46I'6 L6869'L0J16L8 9'L6 p1 jie J39'bb2 L08T4T0I1 AJJ6L6We T60.pIL6 1.61)014 TA6U 1 p2 1)L06880L 2' rT9'9'6u-'leuaeu p1 peue (e;i)0TI1 40 p6TI1 AT4�JJ e J3T2.POLTC9']' ThctO8 W6I140I'Gg pf14 AT]']' ;oo 011 ;omo; we 09'L0.ç6u0g2 4JJ6I1 fl13qeLAeu4 H0A6A6L ui2 bL626114� LL0L688 1.61)014 T I10.ç2J1L6 qou1.ç ueeq 40 46fl 9'IY2 o; 70iT ATff4 9' L9'bq qGAe]'obmeILp 4J6 9'U9'T28TabLoclILeg 0136 o; we JL8c fl6CJCW9'13 DII 20011 98 A9'8 b022TpTe ;o PTI I 9'W]'9'B46 9'0iT4 9' A667c' j? .69'0JJeL 9'.c 4JJ6 4U6 1)L0,6280L 1)9'JT]' R9'LL6L J0J1JJ6116 9b6C.çLflBJ — PTIT AT.JJ bJ040—b]'9'4e2 — TI' 09" ratr JJ6 AJ10]'6 bL0oeqfme8J30A2 fl9'T2-4JTp68 TW A9'LT9'ple WTcJcueaa AIW tJJTcJ1 I eoogeq w2 ;TLac c9'Lo.ç-OuT2 9' ;SA 9'W0I'24 2orr AT]']' L6C0I'T86 AJJ9'4 T qebC.p6g 011 4J12 bJ304oL9'bJs: T4

LJ30ç0L9'bJJ j

1 I1ODflCLIOI

— C9'L046U0ga TI' L086 ;]'oAeLePT0I0T09'T C20IT89'4T011 846b'9'I' 9' CJJ6WT84I8 9'bbLo9'cJJ .ço geebeL mJqeLaç9'11qTu o; we—2.ceLeoCJJewTacL2: 136A 9'88]'I1W6I'4a 0, 9'pBO]'rr.pe 00I1;TJTL9Tou'—InthOI49'11c CJJ6WTCWT çL9'1Ja;oLw9'.çoua—j6A 84LJ1C.pJTL68

p e ;on-oATu 9'L69'8 o; c9'Lo4euoq CJJ6WTacL7:ftp II'46L1J9'4T0IJ9'T 2711th08T!TLU 013 Q9'L04GI'0qab GLI'6 A46L]'9'I'gbYP8LG - L6A6A T bL686134�6q 011 bL0L622 W9'ge 8p106 4J36

}ggwa.çL9'a2e CH—9001 4TLTC}J 2AT46L]'9'I'qI'38fl'4i146 o; oL9'uTc G36WT84L2 III'TA6L8T47 o; J1LTCJ3

COUL9' fl9'I12 EI1846L-

GHVVGEI2VJIO CItEWI2&1A VI4D 2&ESEOQHENI2rdsA OL CV1OLE}1OID2

rsLwou LLG rq fl3c3• EJIJ1Gq tIJ CLG BLTrIJ -?1

464 CONRAD HANS EUGSTER

2. NEW CAROTENOIDS

The graph which Professor Liaaen—Jensen had shown in Berne in 1975 (Ref. 1) has,at the time of report, still scarcely flattened. Numerous new carotenoids haveagain been discovered, however, no more fundamentally new structures. It mustbe clearly stated and contrasted with the speedy advances in other terpenoidclasses, e.g. with the sesqui— or diterpenes. Here, it is indeed still possib-le for a small and active group to isolate and clarify one hundred or more newsubstances in a period of five to ten years Perhaps the transition of mainactivity from the plant world to that of the insects or marine animals willagain bring something new. Examples are found on:

0

6

Scheme 1

Characterization, chemistry and stereochemistry of carotenoids 465

HydrocarbonsThe 3 y,i-.carotenes , , have been isolated by Britton and Goodwin fromCoccinefla septempunctata (Ladybird beetles) (Ref. 2). The constitution prop—osals are based on UV/VIS— and mass spectra. The absolute configuration remainsopen, presumably until again such a sensational mass occurrence of ladybirdbeetles arises as we had in Europe in 1976. Britton and Goodwin postulate thatthe cyclisation step possibly takes place in the animal. Because of the pron—ounced rarity of the y—terminal group one should, in this connection, bereminded of the occurrence of y,y—carotene in aphids (Ref. 3). From thechemical point of view, the y-types should actually occur much more frequently— I will return to this point later. A poly—(Z—)—,t—carotene has been ascer—tamed in fruits of tangerine tomatoes (Ref. 4). It may be closely related tothe "pro—,—carotene" isolated and crystallised by Zechmeister and Schroeder(Ref. 5) from fruits of Butia capitata and Pyracantha angustifolia, whosestructure is still open. For both (Z)—isomers, stronger absorption on Ca(OH)2and !4g0 than. the all—(E)—,J,-carotene is quoted. Both furnish on 12 isomerisat—ion a little all—(E)—,J,—carotene.

I am convinced that the time has now arrived, for these and other (Z)—isomericcarotenes to be isolated and structurally elucidated by modern methods. Theyare, probably, more widespread than one has hitherto assumed. The (Z,E)—isomerism is really one of the most inherent phenomena of the carotenoids. Itstands — at least on this scale — almost alone in the chemistry of naturalsubstances. I shall return later to one special aspect of this characteristic.

A new carotene is tethyatene () (3,4—didehydro—,Z—carotene), which has beenisolated from the orange—red sea—sponge Thethya amamensis (Ref. 6).

Methvlethers and EioxidesI will now mention a few newly discovered carotene methylethers and epoxides.The methoxylated ,—carotene (1 ' 3' ,4 '—didehydro—l' , 2 '—dihydro— ,y—carotene) was isolated by Britton et al. (Ref. 7) as trace carotenoid fromcultures of Rhodomicrobium vannielii. The new substance is interpreted as beinga link in the biogenesis pathways to p—carotene on the one hand, and spirillo—xanthin on the other. The same authors, in their search for new caroteneepoxides (Ref. 8) in extracts from the tomato mutant "6", then came acrosstraces of (l',2'—epoxy—l',2'—dihydro—,i—carotene) and (l',2'—epoxy—l',2'—dihydro—s,iJ—carotene). The derivation of constitution is essentially based oncomparative UV/VIS- and mass spectra and comparative chromatograms. The eluci-dation of the chirality at 0(6) in 1 should no longer present problems today; itis more difficult to prove that of the epoxide group. I must state that,hitherto, it is still not known for any carotenoid with a terminally—placedepoxide group. Should these epoxides play a role in the ring—closing reactions,then the elucidation of the chirality of this terminal group is, of course, ofimportance.

New hydroxycarotenesAgelaxanthin A () ((3R)—,çp—caroten—3—o1)was isolated from the sea—spongeAgelas schmidtii (Ref. 9). I will return to its absolute configuration insection 4. For parasiloxanthin and dihydroparasiloxanthin, two carotenoidsisolated from skin and fins of Parasilurus asotus (Ref. 10), the constitutionsand Q are given: (7' ,8' —dihydro—3,3 '—dihydroxy— ,—carotene and 7,8,7' ,8'—

tetrahydro—3, 3' —dihydroxy— ,n—carotene, respectively). Argumentations forthese constitutions are based essentially on comparison with the propertiesof 3—hydroxyzeacarotene; unfortunately there are no chiroptical data quotedin the work. This shortcoming befalls not only this but also numerous otherworks. Particularly for such rare and difficultly accessible carotenoids,that is very regrettable. In section 4, I will go into more detail aboutchiriquixanthin A () and B (a), isolated from the skin of a frog. Asderivatives of 6,—carotene they both offer particular interest. The possibleidentity with the so—called tunaxanthin (from fish) is still not definitelyclarified.

A very significant and in many ways also instructive structure revision has ari-sen for the hydroxycarotenoids and . I have put the old suggestions inbrackets next to the new ones. We are dealing with carotenoid from thealgae Anacystis nidulans, which on recent investigation (Ref. 11), on the 'basis


1

I0

( )

<3,

<3,-I

<3,-I, 5'

0I

I?'<3,-

<3,r(3,

/

0IScheme 2

I0

r(3,

-K(3,

'4' \0I

I0

/ I

"4<,,

<3,-

<3,

<3,-

<3,

/

0I

I0 0

/<3,-

K)'

<3,—

<3,

/

0I


of UV/VIS-. and mass spectra immediately turned out to be tn— and tetrahydroxycarotenoids, respectively, of the ,—type. The constitutions proposed on thestrength of insufficient microtests (Ref. 12) were thus immediately dropped.Primary hydroxyl groups are absent on the grounds of the 'H—NNR—spectrum,secondary, allylic as well as tertiary —OH's are also lacking, which followsfrom the stability of the compound in HC1/chloroform. Since dioxolane format-ion occurs with acetone, vicinal hydroxyls exist on 0(2)10(3), C(2')/C(3')respectively. From 3J2,3 = 10.5 Hz, a quasi—equatorial position of the OH—groups in the —ning was deduced and, from the obvious 02—symmetry, also ofthose in the '—ring. The OH—group on 0(3') of is certain on the basis of1H41R comparisons. Nostoxanthin is completely different from cruetaxanthinin the IR— and IH_N1VJR_spectra. For derivation of the absolute configurationsee section 4. The transformations demonstrated thin—layer chromatographicallyby Stransky and Hager (Ref. 12):

nostoxanthin caloxanthin () + zeaxanthin

zeaxanthin H202 caloxanthin + nostoxanthin

thus necessitate verification. In this connection, the "Pigment IV" and"Pigment II" (Ref. 13) isolated in very small amounts, from cultures of Rhizo—bium lupini should also be mentioned, for which analogous constitutions toand (that is, wIthout stereochemistry) are given.

Scheme 3

17


The constitution is given for a tetrahydroxy carotene from the juice ofValencia oranges (Ref. 14) . The arguments are : 474 , 445 , 424 (solvent ?).The glycol with analogous chromophoric system shows X 476, 448, 425(ethanol); see section 4; formation of a diacetate and conversion into muta—toxanthin by sufficiently long treatment with HOl. Chiroptical data areunfortunately not communicated. The formation of from antheraxanthin islikely. The stereochemistry of the epoxide hydrolysis is of interest ip viewof the analogous cases mentioned in sections 3 and 4.

Prom an older sample of " taraxanthi&' (see Note a) , which had been isolated byKarrer et al. from Ranunculus acer, a glycol was also isolated; which accordingto Ref. 16 is assigned structure (argumentation for the absolute configurat—ion see section 4).

Carotenoids with carbonygroupsA new lycopenal () ; 13—(Z)—3' ,4 ' , 2 '—dihydro—ip,ip—caroten—20—al;possibly also 3,4—didehydro—l,2—dihydro—structure) has been found in additionto neurosporene , lycopene, 1,2—dihydroneurosporene , 1,2—dihydrolycopene , 1,2—dihydro—3,4—dehydrolycopene and 13—(Z)—ip,ip—caroten—20—al in cultures of Rhodo—pseudomonas viridis (Ref. 18).

,ip—Caroten—4—one was isolated as the main pigment from an Arthrobacter sp.(Ref. 19). A whole series of 2—oxo—, 2,2'—dioxo—, 2—oxo—2'—hydroxy—,—carotenes(2) including some with 3,4—didehydro structures are reported to occur in theinsect Oarausius morosus (Ref. 20).

Two Japanese teams, evidently independently of each other and in mutual unaware-ness of the results, have been working on the carotenoids from the sea—spongeTedania digitata. According to Okukado (Ref. 21), the main pigment is namedtedanin and on the basis of extensive investigations assigned structure ?.Q

(2,3—didehydro—3—hydroxy— ,X—caroten—3—one) (see Scheme 4). Since Okukado' sisolation included a saponification step, it is possible that the known dehydro—genation of the original keto to the enolised diketone has occurred. From thesame species, Tanaka et al. (Ref. 22) obtained an also crystalline carotenoid,which they called tedaniaxanthin and assigned structure . The strikingend group was not commented upon in the work of the authors. A carotenoidnamed clathriacine, isolated by Tanaka and Katayama from the sea—spongeClathria frondifera, was assigned structure Q (Ref. 23). The esters of Q arenamed "Clathriaxanthin" by these authors. Thus, the tedanin from Okukado andthe clathriacine from Tanaka and Katayama are synonymous.

The compounds 22 and are 4—oxo—,—carotene derivatives, which have beendiscovered in cultures of Rhizobiurn lupini (Ref. 13). The trans—configurationfor the glycol grouping was deduced on the basis of the 1H—NR—spectra. Theabsolute configuration is still unknown.

Fritechiellaxanthin is a new carotinoid with a 4—oxo function (Q); see sect-ion 4 (Ref. 24).

The constitution was proposed for agelaxanthin C from the previously mention-ed sea—sponge Agelas schmidtii, whilst agelaxanthin B (j) possibly representsthe 19—0—methylether of agelaxanthin C (Ref. 9); for the absolute configurationof the x—terminal group see the comments in section 4. Besides , j4 and,this sponge also contains a—carotene, isorenieratene, zeaxanthin and triken—triorhodin.

From the red "fermenting yeast" Phaffia rhodozyma, 3—hydroxy—4—oxo—torulene(; 3—hydroxy—3',4'—didehydro—,ip—caroten—4—one) was isolated in slightamounts (Ref. 25). The 7,8—didehydroastaxanthin ester (Q) from starfisch(Ref. 26) may well be identical with asterinic acid (see section 4).

Note a. The "taraxanthin" from R. Kuhn and B. Lederer (Ref. 15), a nicelycrystalline substance, is not, as is maintained in several newerworks, identical with lutein epoxide but is a mixture of luteinepoxide/flavoxanthin/chrysanthemaxanthin (73:13:14 ex Taraxacumofficinale) (Refs. 16 & 17).


Finally, the carotene j, named papilioerythrinone isolated from integumentsof pupae of Papilio xuthus (swallow-4ail) and carapaces of crab (Paralithodesbrevipes) should be mentioned (Ref. 27). It occurs besides papilioerythrin(3'—OH instead of 00), canthaxanthin, astaxanthin, etc. (swa1low.tail).Chiroptical data are missing in this work. Therefore the biogenetic sequenceproposed by the authors must be regarded with caution.

V I I

24R=CH3 25R=HHOScheme 4

AocarotenoidsOne knows today, that carotenoids are transformed in plant organs into verymany fragments, which scarcely show any further connection with carotenoids.I draw your attention here to flavours from rum, whisky, wine, tea, tobaccoetc. This is not the point here, I will rather limit myself to substances,which still show a more distinct connection with carotenoids.

0 20

0 23


HOJ'CH2OH

30 C5H,—CO

COOH

31 OH 32

OHCOOH

OH

Scheme 5

—Citraurinine () and -.citraurol () were isolated from citrus fruits(oranges, mandarines) (Ref. 28). The hydrocarbon is very unstable in thecrystalline state arid quickly discolourises. No reproducible 1H—MR—spectracould be obtained. is also unstable. To date, 7 naturally occurring 8'-.apocarotenoids are known, of which 5 originate from citrus fruits. For thisreason, the authors take one particular biogenetic pathway into consideration.19—Hexanoyloxy--paracentrone (Q) was isolated, in addition to l9'-.hexanoyloxy—fucoxanthin, from Coccolithus huxleyi (Ref. 29).

In well documented works, Kulkarni et al. (Ref. 30) tiave isolated and clarif—led a whole range of apocarotenic acids from the parasitic Aeginetia indica(Orobanchaceae). Compound B (aeginetic acid) has constitution, accordingto the IH...NMR... and mass spectra. A trans.- (diaxial) position of the vicinalOH—groups was concluded from the negative periodate cleavage, as well as frompyridine-induced-NMR—shifts in conjunction with a conformational argument.In my opinion, however, the lastly—mentioned conclusions are not conclusive,because not only the axial but also the equatorial CH3 at 0(1), together withCH3—C( 5), experience practically the same large paramagnet Ic shifts • However,the derivation of the configuration of the glycol grouping can be upheld byother means: comparison with the chemical shifts on natural azafrin ester(Ref. 31) and "cis.-OH-.azafrin ester" (Ref. 32) (not carried out by the Indianauthors) shows that only a trans—diaxial configuration comes into considerat-ion. Also, the compound B has a different melting point to the racemiccompound prepared by Tamura et al. (Ref. 33).

Compounds D and E () appear to be dimorphous. They both exhibit very similar1H—NNR—spectra to compound B, with regard to the ring protons, and show in themass spectrum the fragmentation behaviour described by Enzell (Ref. 34) onazafrin. Compound F is an azafrin type, possibly a stereoisomer of azafrin.Besides these compounds, the lactone (aeginetolide) was also isolated.

I

'II /

/

Characterization, chemistry and stereochemistry of carotenoids

0-

471

-D0CUU

I

I///

0) 0C') 0.

/

II/

COcv)

COCv)

÷

Scheme 6


) LC)//I0

—

r1co

/'I I

n +

Scheme 7


3. SELECTED CHEMICAL REACTIONS ON CAROTENES AND CAROTENOIDS

At the time of report, new experiments on the functionalisation of easilyobtainable carotenes, such as p—carotene, canthaxanthin etc. have been made,which could gain significance for future synthetic work. They are mentioned inthe form of an extract in this section. In addition, interesting reactionmodes have been discovered during structure elucidation.

Reduct ionsElectrochemical reduction under acetylation conditions on canthaxanthin ()gave the retro-.'compound (R = Ac) in addition to traces of retro-'monoacetate(, R = H) and . From , after hydrolysis, the compounds and (5,5'-.dihydrocanthaxanthin and 7 , 7 ' -dihydrocanthaxanthin, respectively) were formedconcurrently (Ref. 35). In this connection, the analogous electrochemicalreduction of u—carotene should be mentioned (Ref. 36), in which 65% 7,7'—dihyd-.ro—-carotene was formed. These reactions therefore proceed, very similarlyto the older chemical reductions of polyenes and polyene-.ketones with the metal!acid combination; e.g. with sodium amalgam/ether/H20 (Refs. 37, 38, 39) orzinc/pyridine/glacial acetic acid (Ref. 40).

PinacolisationWork on these reactions has repeatedly been done in recent times. Thus vitamin A-.aldehyde () could be converted into 15 , 15 ' ...dihydro—15, 15 ' ..dihydroxy—, 3—carot-ene () in a yield of 10% by controlled potential electrolysis at a mercurycathode (Ref. 41). Substantially better yields are obtainable by chemicalmeans according to Ref. 42 (zinc amalgam, zinc—copper couple in pyridine).Principally it concerns the same methods which had been applied by Kuhn andWinterstein (Ref. 43) for the first time in the polyene series. We have like-'wise obtained good yields of glycol by this method (Ref. 46). The lowvalency complexes of Ti (prepared from TiCl4/LiA1H4 or TiCl4/Zn according toRef. 44, or TiCl4/LiA1H4 (Ref. 45)), which were used by these authors for a newsynthesis of p-carotene in yields of 85—90%, also give the pinacol undersuitable conditions (—70°) (Ref. 46). The main product has C2—symmetryacøording to the 1H—NMR-spectrwn and is thus the racemic form. From , theepoxide jQ can be prepared (Ref. 46) with the sulfurane according to Ref. 82.

Epoxidation reactionsI would now like to mention a few epoxdiation reactions, which have recentlybeen carried out. The team in Cluj (Rumania) has made extensive investigationson in—chain epoxides of canthaxanthin (Ref. 47). The following epoxides werecharacterised: mono-.9,lO, di—9 ,lO,9' ,lO', mono 13,14, mono—li, 12 • The rearran-gement to furanoid epoxides and their behaviour towards BH4e and LiA1H4 wasalso examined. Researchers in Pcs (Hungary) have published similar reactionson capsanthin (Ref. 48). The isolated epoxides (capsanthin epoxide A (4.; 35,55, 6R) and B (4,; 3S,5R,6S)) were formed in the ratio 3:1. The two epimerscould be well separated on CaCO3 with benzene; whereby 4.1 adhered better thanj. The almost enantiomeric CD—curves of the two compounds in the region of200—370 nm is noteworthy. Their acid—catalysed rearrangement resulted in thefuranoid capsochromes A + A' and B + B' respectively. Somewhat different ratioswere found (Ref. .49) on epoxidation of ,e—caroten—2—ol (4). The epoxides 4.and 4 were formed in the ratio 3 : 2. With this pair, the trans—epoxide(OH... epoxy—group) adhered better. For lutein (4.), the situation is moreextreme: The usual epoxidation methods always lead to almost pure cis-.epoxide

(4.) (ratio j:4. = ca. 19:1) (Refs. 50 & 51).

Epoxide hydrolysesIn connection with the elucidation of the configuration of natural carotenoid-'5,6—glycols, the question of their origination from 5,6—epoxides and formationtherefrom in vitro has, in recent times, received increased consideration.Karrer and Stt&rzinger (Ref. 52) converted —ionone—5,6—epoxide (4.) (carotenoidnuxnbering) into the crystalline glycol Q for the first time. Its configurat-ion, plausible for mechanistic reasons, was, however, proved by X—ray of therelated (racemic) compound (Ref. 53). This method of hydrolysis is like-wise successful with typical carotenoid epoxides, provided that one works witaqueous mineral acids with the admixture of dioxane, tetrahydrofurane, etc.

c(C

H2O

H

HO

' R'

÷ H

Oh1

bH

OH

H

O*O

R2

+ H

O13

R2

56

57

HO

u1T

' R

3, 3—carotene

(1. T1(OAc)3/benzene/aceticacid

2. Saponi-fication)

Scheme 9

Characterization, chemistry and stereochemistry of carotenoids

Didehydro—13, 3—carotene, m.p. 138—139

475

Zeaxanthindiacetate

Luteindiacetate

67

HO13, 68

Scheme 10 69


The main hydrolysis products in the reactions hitherto known are, however,always furanoid epoxides, which in non—aqueous medium are known to constitutethe only products. The following glycols were prepared: from diadinoxanthin(.a; absolute configuration see section 4) 2 tetraols arose, the more polar ofwhich proved to be identical with heteroxanthin (Ref. 54) . Since the twoepimers strongly differ in the IR-.spectrum, structure was assigned to thatpossessing the stronger intramolecular H...bonding (heteroxanthin) , and structure

5.4 to the O(5)-epimer. It is noteworthy that, under the conditions stated, nofurane oxide is formed from the tetraols and (compare also the reactionon ) . Thus an inversion at C(6) is improbable.

Finally, two very polar pentaols could be isolated by chromatography from neo—xanthin samples originating from Trollius europaeus ("Trollixanthin"). Thesecould also be synthesised from neoxanthin with aqueous acid. They show thesame chemical and spectral differences as and , and thus have structures

and (Ref. 54).

Oxidations with KNnO and Nn02These oxidations belong, as you know, to the classical method by which thestructure elucidation of ,—carotene was carried out. Recently, researchersfrom Pécs (Ref. 55) were able to show that, under suitable conditions, thisdegradation of caroten—5,6— and 5,8—epoxides leads to 8'— or lO'—apocarotenals.Thus, 5, 6—epoxy—apo—lO' ——carotenal was formed from 5, 6—epoxy—3 , p—carotene,5,6—epoxy—a, s-carotene respectively, and 5,6,5' ,6'—diepoxy-. ,3—carotene. Fromaurochrome, the 10'— and from mutatochrome the 8'—apocarotenal were synthesised.Interesting is the observation by Coman et al. (Ref. 56) that diosphenols of theastacéne type () can be converted into 2—nor compounds of the (blue) viol—erythrin type () with Nn02 in acetone.

Oxidations with 02 in alkaline solutionK—t butylate/02 permits the conversion of 4—oxocarotenes into diosphenols ofthe astacene type. If the diosphenol system is first reduced with borohydrideto the glycol and then selectively oxidised at the allylic hydroxyl group withMn02 (or by Oppenauer), a synthesis of (±)—astaxanthin can be realised (Ref.57). The same authors prepared analogously: ()-3—hydroxy-canthaxanthin, (±)-3—hydroxyechinenone and the corresponding (±)—l5,l5'—didehydro analogues. Inthis connection attention is drawn to the evidently unexpected easily occurringair oxidation of n—carotene, which was observed on microcell C chromatograms(Ref. 58).

Oxidations with Tl(AA study on the reaction of selected carotenes and carotenoids (Ref. 59) result-ed in a number of reaction products of which the following have hitherto beencristallised and identified: from ,—carotene — a didehydro—f3,—carotene (60)of still unknown structure; (±)—mutatochrome (a), (±)—isocryptoxanthin (6237(±)—4 '—hydroxyxnutatochrome (a); (±)—isozeaxanthin (a); a (±)—hydroxyisozea—xanthin (a). Zeaxanthin gave mutatoxanthin () and 4-hydroxymutatoxaxthin ()and from lutein the cis—epoxide () was formed as well as a 4—hydroxylutein(a).

The results show that the expected allylic oxidation predominantly occurs. Theepoxidee may have been formed from an unstable thallium compound (Q) on sapon-ification of the reaction products, e.g.

AcO OAc

R ®oH +R

OAc

70

Scheme 11


Halogenations of the side chain with NBSSince in recent times various carotenoids with oxygenated methyl groups havecome to light as natural products, new procedures for the rational ayntbesis ofsuch compounds must be developed. A first access was found by the Norwegianteam (Ref. 60) in the radical bromination of crocetindial with NBS, in whichboth in—chain—methyls are preferentially attacked. Mono— and di—substitutedproducts arose, which after hydrolysis gave the corresponding alcohols. Promthis ip,p—caroten—2O—ol, rhodopin—20—ol and rhodopin—20'—ol. have now beenprepared according to the usual procedures (Ref. 61).

HC1 reactions on carotenoidsThe reaction of zeaxanthin, isozeaxanthin and lutein with chloroform/HCl wasinvestigated by Pfander & Leuenberger (Ref. 62) with regard to the products.It led to the foreseeable substitution and elimination products. More inter-esting is the corresponding reaction of allenyl alcohols of the neoxanthin type

L

Scheme 12

A careful study (Ref. 63) now confirmed an earlier finding (Ref. 64) accordingto which, in this reaction, elimination occurs with the formation of the C(7)/0(8)—acetylene bond giving , whereby though considerable amounts of 7—chloro—substituted carotenoide (J,) are also formed.

4. STEREOCHEMISTRY: ABSOLUTE CONFIGURATIONS

Since the carotenoid symposium in Berne (1975), at which Professor Liaaen—Jen—sen summarised the situation on the configurational findings of carotenoids atthat time (Ref. 1), a considerable number of absolute configurations have againbeen established. The methods developed earlier have proved very fruitful andthe basis for numerous correlations by means of chiroptical methods has turnedout to be quite reliable. The contributions again came almost exclusively fromthe laboratories of Liaaen—Jensen, Weedon and Eugster. It should be rememberedthat the carotenoids represent one of the last great groups of natural productswhere this work was still to be achieved: with amino—acids, sugars, steroids,terpenoids, etc. this work had, essentially, already been accomplished earlier.The specific difficulties in the carotenoid chemistry are due to the fact thatthe relative configurations were also still unknown, so that the determinationof a single centre could not almost automatically lead to the absolute confi—guration of the whole molecule.

I would like to classify the new chirality determinations, taking the hydro-carbons as a basis, and discuss them in this sequence.

72

71


Acyclic carotenoidsIn contrast to an earlier publication (Ref. 66), Johansen and Liaaen—Jensen(Ref. 65) have guaranteed the (2S,2'S)—configuration of bacterioruberin ()and bisanhydrobacterioruberin (fl) by synthesis of tetraanhydrobacterioruberin() using (-)—(R)—lavandulol () as starting material. j can be prepared byelimination reactions from and fl. Remarkable is that comparison of theCD—curves of and with those of had given a different result, sincethey are almost enantiomeric. Thus it follows anew that such chiropticalcomparisons only allow conclusions on the absolute configurations when themost similarly built molecules are compared with the most similar functionalgroups. Evidently, the conformational equilibria in the case of , , fl and

are partially different. The absolute configuration at C(2) in and flcorresponds to that in decaprenoxanthin, sarcinaxanthin and C.p. 450.

I I I I N1

CHOH

OH 76

H

OH 77

H

78

Scheme 13

Monoçclic carotenoidsThe (2'R)—chirality for aleuriaxanthin () (Ref. 68) followed on the basis ofHoreau experiments with the gas—chroniatographic modification according to Ref.67 and on the assumption that 2—propenyl >CH2. At RT, the Cotton effects areextremely weak in f. For the synthesis of (5S,6R)—5,6—epoxy—5,6—d1hydro—,ip—carotene (, = y—carotene epoxide), which is so far unknown as a naturalsubstance, Eschenmoser and Eugster (Ref. 69) used azafrinal (80) as condensa-tion partner for the Wittig salt on ip—ionol. The crystalline5R, 6R)—5, 6—dihyd-ro—,ip—caroten—5,6—dio1 () was hereupon converted into the crystalline epoxidf() with sulfurane according to Ref. 82. The glycol shows very weak Cottoneffects, with the epoxide they are substantially more pronounced.


79

+

I

0:81

Scheme 14

HOV*L\c=O,23

""'r"

z\ 8=0,22

82

86

Scheme 15PAAC 51/3—o


Bicyclic carotenoidsThe ,y—carotene § isolated from Caloscypha fulgens has (6'S)—chirality (Ref.70), which follows from a comparison of the CD-.curves with the synthetic

product ent , prepared from enriched (R)—y--ionone (see Note b). The agela—xanthin A already mentioned in section 2 has structure §4, as a result of theagreement of the CD—spectrum with (—)—zeaxanthin (3R,3'R); consequently it is(3R)—,g—caroten—3-ol (Ref. 9). There have now been numerous arguments publish—ed (Ref. 75) which clarify the absolute configuration of sarcinaxanthin ()and its mono——D—glucoside: CD— and 1H—NNR—comparisons with the (2R,6S,2'R,6'S)—2,2—dimethyl—y,y—carotene () synthesised from cis—y—irone 86 suggestthat 0(6) and C(6') have (R)—chirality in § and that the C(2)—substituent iscis to the side—chain, provided the substituent at C(2) has no essential in—fluence on the CD effect. However, doubts about this line of reasoning arestill possible because the published CD—curves of and 87 as well as of 2'—methyl—,y—carotene deviate quite clearly from that of the unsubstituted f3,y—carotene (Ref. 70).

From this must be concluded that the substituent at C(2) exerts a considerableinfluence on the conformation of the y—ring and thus on the CD—effect. Thecis—position of the C(2)—C(6) substituents derived from the 1H—NIMR—spectrumshould, moreover, be confirmed by synthesis of the analogous trans—compound andcomparison with its lHN1V1Rspectra.

In section 2 I have already mentioned the two ,s—carotendiols chiriquixanthinA (88) and B () (Ref. 76). According to the 1H—NNR—spectra B has C2-symmetryand one ring in A shows analogous chemical shifts to lutein and is consequentlytrans—3,6—disubstituted (compare Ref. 77), whereas the other must therefore becis—3,6—disubstituted. The same was also concluded from the 1H—NNR—spectra forboth rings in B. The absolute configuration at C( 6) and C( 6 ' ) was derived fromcomparison of the CD—curves with configurationally guaranteed compounds of thee—series. In the last analysis, it is based on that of ,s—carotene itself(Refs. 71 & 72).

The absolute configuration of decaprenoxanthin () was previously derived inBerne (Ref. 1). Since then, the full publication has appeared (Ref. 78) andthe following arguments must be appended: the absolute configuration, as givenin Q, is based on the one hand on chiroptical comparisons with the syntheticcia— and trans—dimethyl—6,6-carotenes and , which were synthesised fromthe optically active —irones (absolute configuration see Ref. 79), and on theother hand on 1H—NMR-data. Since and respectively show opposite and thesame CD spectra compared with the chiroptical standard ((65,6'S)—6,6—carotene(Refs. 71 & 72)), it follows that the 0(2) substituent in this case makes noessential contribution to the Cotton effect. From this, the chirality of Qwith respect to C( 6) and C( 6') ensues • For the determination of chirality at0(2) and C (2') shift arguments in the 1HNMRspectra served, see Scheme 16.The not obvious equalisation of the influence of hydroxy—isopentenyl and methylon the chemical shift of the geminal methyl groups was ultimately confirmed bystereospecific synthesis of (±)—2,6—trans—decaprenoxanthin, which was carriedout by Chopra et al. (Ref. 80).

Carotene epoxides are very widespread in nature, however still nothing isknown about their absolute configuration. In connection with the determinationof the absolute configuration of azafrin and azafrinal (Q; see below) Eachen—moser and Eugster then first synthesised (5R,6R)—5,6—dihydroxy—,—caroten—5,6—diol () and (5R, 6R, 6 'R)—5, 6—dihydroxy—, 6 —caroten—5 ,6—diol (fl), respective-ly (Refe. 69 & 81) with the aid of the corresponding Wittig salts 9 and .These were then converted into the epoxides with the sulfurane reagent of

Note b. Regarding the determination of chirality of (—)—y—ionone by correla-tion with (—)—(S)—a—ionone see Refs. 71, 72 & 73. In the recentlypublished correction by Ohloff & Vial (Ref. 74) of their rotationvalue for dihydro—y—ionone (not of the chirality), the derived abso-lute configuration of y—ionone in the joint publications (Refs. 71 &72) is not dealt with: (—)—(S>-y—ionone is connected with (—)—(S)—a—ionone and (—)—(S)—dihydro—a—ionone; the oata1yt.c hydrogenation of(—)—(S)—y—ionone yields (-i-)—(S)—dihydro—y—ionone and not (—)—(S)—dihydro—y—ionone as previously stated.


HO

&=O,15

L5 0,19

L\0,01

Scheme 16

Martin (Ref. 82). The glycols 95 and are still unknown as natural products.However, they show such unexpected properties with regard to their solubilityand chromatographic behaviour that they have possibly been overlooked up to now.Their properties lie much closer to the hydrocarbons than to the usual xantho—phylls in spite of the diol grouping present. One must attribute this to theditertiary nature of the trans-glycol grouping, which is in addition stronglyhindered sterically by the neighbouring methyl groups. The CD-curve ofshows only weakly pronounced maxima, as expected; that of is determined bythe centre of chirality at C(6'). The Cotton effects of the two epoxides jand 9 are very much more strongly pronounced. With these data, the chiralitiesof the natural products can now be clarified. Strangely enough, apparentlystill no chiroptical properties of 13— and a—carotene epoxides, respectively,have been published.

In connection with the discussion on compounds 81, , 97 a few further glycolsshould be named, whose absolute configurations have been published at the timeof reporting (Refs. 83 & 54). It deals with heteroxanthin () and.the tetra—

and pentahydroxy compounds , ,j, , and (see sections 2 and 3). Onthe one hand, diadinoxanthin () could be converted into diatoxanthin () ofknown configuration (Ref. 50) by lithium aluminium hydride reduction of thesilyl ether, whereby the configuration at 0(3) and 0(3') was determined (theconfiguration at the epoxide grouping resulted from IH_N1VLR_spectra of diadino—chrome and comparison with data from 3,5—cis— and 3,5—trans—isomeric furaneoxides (Ref. 84). On the other hand, epoxide hydrolysis in a protic solvent(see section 3) gave the glycols and j, of which proved identical withheteroxanthin. Consequently, diadinoxanthin () has the (3S,5R,6S,3'R)— andheteroxanthin the (3S,5S,65,3'R)—configuration. (The cis— and trans—configurat-ion, respectively of 0(3)/C(S) was established by IR—spectroscopy; see section3.)

482 CONRAD BANS EUGSTEF

'(7/

co0)

-

/0)

(3'

>0s;

I I

/}

Scheme 17

00)0

(1

0

Lt)0) N

0)


I IHOI.cJII 99

55

48 — 46 ' 47

R

+HO,117

R HOI' R + HO1' R

100 101 102 103/HO#11 HO)II'IJ

104 105

Scheme 18

The glyools and were also prepared by protic hydrolysis, this time ofneoxanthin (see section 3). The absolute configuration of the tetraol fromRanunoulus acer is not entirely certain, it is based on the assumption that weare dealing with a protic hydrolysis product of lutein epoxide (j). Withregard to the absolute configuration of 4, the arguments of Professor Weedon'steam and those from Zurich are now in agreement (Refs. 51 & 84). However, forthe furarioid epoxides flavoxanthin and chrysanthemaxanthin derived therefrom,they differ with respect to the configuration at 0(8) (Refs. 17 & 84) and alsoconcerning the assignment of the 1H—NMR—signals for H(7) and H(8) and whetherthe 3J(7,8)—coupling of 2 Hz is due to the cis— or trans—compound (QQ, 101respectively). In our opinion it is due to the cis—compound )9ç. This is,however, chrysanthemaxanthin and not flavoxanthin (Qi). The matter has acertain significance in carotenoid chemistry, since furanoid epoxides arewidespread and flavoxanthin/chrysanthemaxanthin can represent the basis forfurther chiroptical correlation. The two epimers can easily be differentiatedin the CD-spectrum, see Fig. 1. For our work on flavoxanthin/chrysanthema-xanthin, my co—workers have collected, extracted and chromatographically workedup about 140 kg of dandelion flowerheads.

On rearranging synthetic lutein epoxide il (cis) with acid under aproticconditions, one obtains the 0(5)—epimeric furane oxides. We have also clarif-ied their absolute configurations and assigned structure to 5—epi—flavo—xanthin and structure to 5—epi—chrysanthemaxanthin (Ref. 85). The nickelperoxide degradation of lOO/9j leads to (—)—loliolide, and ofQ/Q to(÷)—isololiolide.

56 57


':__3—

o_

-5—

—

,'z:

.\

fl'? .

Fig. 1

The carotenoids caloxanthin () and nostaxanthin () have already been men-tioned in section 2. The absolute configurations given there (Ref. 11) resultfrom application of Mills' rule: The preferred conformation of the p—ring can beinferred from the CD and this itself is determined by the substituent at C(3).Furthermore, the 'H—NNR-.spectra reveal a trans—configuration for the glycolgroup. Thus caloxanthin is (2R,3R,3'R)—13,—caroten—2,3,3'—triol () andnostoxanthin is (2R,2'R,3R,3'R)—,—caroten—2,2' ,3,3'—tetraol (a).

In a publication dating back relatively far, Entschel and Karrer (Ref. 86) habeen able to connect (—)—zeaxanthin with (+)—escholtzxanthin (106) as follows:

(—)-zeaxanthin physalien1. NBS/chlorof. /-18°

2. N-ethylmorpholin

di—0—palmitoyleschschóltzxanthin (+)—eschscholtzxanthin (Q)

Scheme 19

106

With the clarification of the absolute configuration of (—)—zeaxanthin (Ref.50), that of eschscholtzxanthin thus also followed — which has been overlookedin several newer reviews.

KetocarotenoidsAll preparations of astaxanthin (Hommarus, Halocynthia, Haematocoocus, Sohizo—nobia (Ref. 87), goldfish (Ref. 24), Fritsohiella (Ref. 24)) isolated up tillrecently have (3S,3'S)—chirality (Q) with the striking exception of a

20OH

- HOLI'

tLE

Flavoxanthin

j Chrysanthemaxanthin

200 250 300 350 400

A. (nm)—..-

450 500 550


OH

I

111RCOO f0

HOLJ112

Scheme 20

carotenoid recently isolated from Phaffia rhodozyma (Ref. 88), for which (3R,3'R)—chirality was derived on the basis of CD—spectra. In this connection,attention should also be drawn to the syntheses of all—(E)—(3S,3'S)— and 15,15 '—(Z)—(3S, 3 'S)—astaxanthin by Englert et al. (Ref. 89), where it was shownthat alteration in the geometry of the C(l5)—C(l5')—double bond leads to analmost complete reversal of the Cotton effects.

The so—called a—doradexanthin, which has been ascertained in fish and insects,is interpreted as being an intermediate in the biogenetic pathway to asta—xanthin. Its absolute configuration was hitherto unknown. Dr. Buchecker willfurnish evidence for it in his lecture in terms of 108 (Refs. 24 & 90). It isthus 4—keto—3 '—epilutein ((38,6 'R,3 'S)—3,3 '—dihydroxy— ,e—caroten—4—one).On the other hand, the so—called x—doradexanthin which has been isolated fromthe terrestrial green—algae Fritechiefla tuberosa (Ref. 91) is epimeric atC( 3') and accordingly a new carotenoid. We name it fritschiellaxanthin (Ref.24) (Q; (38,6'R,3'R)—3,3'-.dihydroxy—Ø,s—caroten—4—one). In connection withthese works, the "3 '—ketolutein" ((38,6 'R)—3—hydroxy— , S —caroten—3—one) hasalso been prepared in larger amounts and precisely characterised. These data(Ref. 24) may be valuable for the structure elucidation of philøsamiaxanthinand similar ketocarotenoids.

107

110


113 "OAcHO

AcO116 "OH

115'

C5H11—CO HO

0

C5H11

Scheme 21

According to Berger et al. (Ref. 92), asterinic acid (110 and 111) has (33,3'S)—chirality — as does astaxanthin. This conclusion was derived from chiropticalcomparisons of the NaBH4—reduction products with the corresponding tetraolsfrom Q1: C(4)— and 0(4' )—OH--groups provide no essential contribution to theCotton effect. Moreover the CD—curves of the configurationally—certaincarotenoids diatoxanthin () and alloxanthin (7',8'—didehydro—) werecompared and almost identical curves ascertained; the acetylene bonds in thepolyene chain thus only led to a flattening out of the CD—maxima.

Chopra et al. (Ref. 93) have prepared trikentriorhodine (112) from (+)—camphor.by a synthesis based on the Claisen reaction (compare Ref. 94). The productobtained, however, only showed weak Cotton effects. Comparison with thetrikentriorhodine isolated by Buchecker et al. (Ref. 9) from Agelas schmidtiishows that the synthetic product is largely racemised or epimerised:Ref. 93: L.c 262 (—0,05), 303 (+0,06), 370 (—0,06) (solvent unknown)Ref. 9: Lc 264 (—1,1), 301 (+1,3), 360 (—0,7) (in ether/hexane l:J)

Allenic carotenoidsThe (5R,65)—chirality in fucoxanthin () was hitherto acknowledged as beingprovisional. Bernhard et al. (Ref. 95) have now proved it by conversion ofinto The 3,5—trans position of the 0—functions can easily be recognised inthe 1H—NNR—spectra.

1)2) H3O

H

114

117


1L-L 118AcO

OH

3'.pbromobenzoate)

CHO

Br__i_COO)-,L

120

j03

Fucoxanthin-p-bromo— Violaxanthin—di-p.bromo.benzoate benzoate

Scheme 22

l9'—Hexanoylfucoxantbin, whose constitution had been communicated in 1975 (Ref.I), has now been completely characterised (Ref. 97). Since it only shows weakCotton effects which allow no conclusion whatever as to the absolute configur-ation (its saponification and reduction products also show the same behaviour),the compound was degraded by ozonolysis to fl. With the exception of thesignals due to the hexanoyl residue, this presented identical 1 1iLR and CD—spectra to the corresponding degradation product from fucoxanthin. And sothe structure of the allene side was clarified. Identical 1H.NMR_signals in

and with regard to CH3(l6),(l7),(l8) and CH2(7) point to a similarrelative configuration at the epoxide ring. The absolute configurationexpressed at this ring in structure is, however, not proven.

As }Iertzberg et al. (Ref. 96) demonstrate, dinoxanthin has the structure 116and is 3—0—acetylneoxanthin. Saponification, acid rearrangement to furane-oxide and chromatographic separation of the C (8' )—epimers gave products whicheach show well—concurring CD—spectra with the corresponding epimers of neo—chrome. Prom the available data, probably similar absolute configurations ofdinoxanthin and neoxanthin may be concluded.

ApocarotenoidsOn the basis of 1H—NMR—comparisons of the signals due to CH3(l), (5), (9),H-C(8) and the CH3CO—group of peridinin (118) with fucoxanthin and neoxanthin,the same relative configuration at the allene end group in the two compoundscan be concluded (Ref. 97). The ozonolysis furnished clear evidence for bothends: degradation ketone with the allene grouping proved to be identical inIHVff.. and CD-spectra with the known degradation product from fucoxanthin(Ref. 98); the same is true for the epoxide end 120, which for comparison wassynthesised from the di—p—bromobenzoate of violaxanthin.

The arguments communicated at the 4th International Carotenoid Symposium forthe absolute configuration of azafrin ester in terms of structure (Ref. 99)have now been published (Refe. 31, 81, 100). The decisive experiment was theuse of complementary— and regioselective—acting reagents, which allowed thepreparation of the enantiomeric pairs 3/ and respectively, from a

119

CO

OC

H3

121

[ 12

2 12

3

3)j

2)J

CO

OC

H3

126

127

1) sulfurane a

ccor

ding

to Ref. 82;

2) Ti014/benzene/ether; 3) H

3O;

4) Ni02


pure stereoleomer. From this, (+)—(S)--dihydroactinidiolide (126) was synthesis—ed on the one hand, and (—)—(R)—dihydroactinidiolide () on the other in verygood yield. . On the plausible assumption, verified in many examples, that TiOl4attacks the allylic hydroxyl and that sulfurane attacks the sterically lesshindered o( 5 )—OH group , the rearrangment and degradation products obtained caneasily be construed.

Azafrin shows only weakly pronounced Cotton effects, as do all other carotenoid

glycols (: ' .1); however when the epoxides, e.g. are atissue they are much stronger.

E)isomeric carotenoidsThe 13C—NMR—method has proved to be the most important modern technique fordetermining the position of (Z)—double bonds in polyenes (Refs. 101 & 102).Attention is drawn equally here to the extensive syntheses of phytoenes with(E,E,E)—, (E,Z,E)—, (E,E,Z)— and (E,Z,Z)—configurations in the triene part andof model substances by Than et al. (Ref. 103). Accordingly, the main phytoeneisomer isolated from carrot oil, commercial tomatoes, Phycomyces blakesleanusand Chlorella vulgaris has been shown to have the (B,Z,B)—configuration, where-as phytoene from diphenylamine—irihibited cultures of Flavobacterium dehydro—genans consists principally of the (B,E,E)—isomer. In addition to these twoisomers, a third isomer was found in Rhodospirillum rubrum inhibited by 2—hydroxybiphenyl and in Nucor hiemalis inhibited by fluoren—9-one to whichGranger et al. (Ref. 104) attribute the (E,E,Z)—configuration. The stereo-chemistry of this new isomer has been reexamined by Barlow et al. (Ref. 105)by using comparative 13C—NNR—data. It is suggested that it could also be(Z,E,Z)—configurated instead of (E,E,Z).

Neocapsorubin A and neozeaxanthin A are the (Z)—13—, neocapsorubin B and neozea—xanthin B are the (Z)—9—isomers (Refs. 102 & 106). Violeoxanthin is (Z)—l3—violaxanthin (Refs. 102 & 107). Isomerisation of violeoxanthin to violaxanthingave a product with identical CD—spectra. From this as well as from 1H—1INR-spectra of the furane oxides, analogous absolute configuration of the stereo—isomers is indicated. According to Ref. 107, tareoxanthin from dandelionflowers (Ref. 108) is a mixture of neolutein epoxide A + neolutein epoxide B+ a little cis—violaxanthin.

Fucoxanthin stereoisomers: when all—(E)—fucoxanthin is isomerised with 12/hi',4 new isomers arise which were chromatographically separated (Ref. 109). Allthe isomers showed CD—spectra different from that of the all—(E)—form. On thebasis of IR—, UV/VIS— and 1H-.MIR—spectra, the (Z)—l3— and (Z)—l3'—geometries,respectively, are assumed for the isomer pairs I and II. Isomer III showed nocis—peak, but a hypsochromic shift by 2 nm of the longwave band. Significantshifts of CH3(l6'), (17') as well as of H—C(8') in the 1H—NNR—spectrum suggeststereoisomerism of the allene bond. This isomer was ascertained earlier in upto 6% in crude fucoxanthin from Fucus serratus. This finding supports thesuggestion by Isoe (Ref. 110) for the formation of the allene grouping; seealso Ref. 111. The following reaction sequence is postulated:

02/hv R

Fucoxanthin()Scheme 24

Isomer—Ill(R = acyl or H)

(unstable)


As a very important outcome of newer CD—measurements on (Z,E)—isomeric caroten—oids, the fact should be mentioned that the geometry of the double bonds canexert an unexpectedly large influence on the course of the curve; in isolatedcases the curves run almost enantiomerically The following examples are worthyof mention:

all—(Z)—astaxanthin and (Z)—15,l5 '—astaxanthin (Ref. 89):

fucoxanthin

CD—curves qualitatively enantiomeric

227 (+0,44),(—0,75)(—0,65)(—1,70)

265 (—0,55), 331(+1,50)(+0,15)(+1,60)

(+0,29)(—1,00)(—0,50)(—0,45)

CD—curve qualitatively enantiomeric inthe region 200—350 nm; in VIS thesame ; 01W up to 350 nm enantiomeric.

a few samples of different origin showdifferent CD—curves; including a pairof almost enantiomeric ORD—curves inthe region 200—350 nm.

almost all maxima enantiomeric tothose of the all—(E)—compounds.

These few examples show that systematic investigations are essential on opti-cally active polyenes with regard to the correlation between geometricalisomerism and CD. The danger exists that wrong conclusions are made from theCD-curves as to the underlying configuration.

CYCLISATION REACTIONI will now come to the problem of the stereochemical course of the biologicalcarotene cyclisation. A few years ago, Goodwin (Ref. 113), Britton (Ref. 114),Davies (Ref. 115) and Porter (Ref. 116) adequately summarised the general stateof knowledge. The first model conceptions (e.g. in Ref. 117) were naturallystill very tentative. Since then, they have been continually modified andrefined. More recently, important impetuses have come from the determinationof the absolute configuration of natural (+)—cz-carotene (Refs. 71 & 72) andtrisporic acids (Ref. 118). From this and also from extensive incorporationstudies by different teams, above all by Goodwin et al., the opinion issteadily gaining ground that the C(l) methyl groups of lycopene and neurosporeneretain their identity during the cyclisation; i.e. according to the conforrnat—ional folding, the (E)—methyl group becomes the pro-(R)— or pro—(S)-methyl groupon the cyclohexene and vice-versa. Unequivocal proof for this, however, is hardlyyet at hand. If one studies the schemata proposed so far for the cyclisationstep, one notices that they are still incomplete and have hitherto disregardedone important aspect of carotenoid chemistry. I am thinking here of the reallyfacile (E,Z)—isomeriation of conjugated double bonds which occurs undercompletely different influences. If we examine the following factors for thecyclisation of a l,5—diene system, namely

1) Chair or boat folding;2) Geometry of the C(l)—C(2), C(5)—C(6)

double bond resp. — also in view of theindividuality of the C(l)—methyl groups;

3) Pro—chirality of the —ene surfaces.

mono(Z)-lutein, mono—(Z)-zeaxanthinand mono (Z)—diatoxanthin (Ref. 1):

8 possibilities follow schematically:

(all—(E)—8'—(R)) (Ref. 109):(l3—(Z))(l3'—(Z))(8'—(S))

violaxanthin/violeoxanthin (Ref. 107):

all—(E)—neoxanthin and 9-( Z)—neoxan—thin (Refs. 50 & 112):•

C—CH2/

re/si II - — IIC-- C,\si/rere/si


Enantiomeric chair foldings:

c1(E.E) E9( (Enzyme)

p

(1R,2s,6R) (1R,2s) (Scheme 25)

In vitro, the cyclisation is an overall trans—addition. In vivo, however, onefinds only the products of an elimination reaction. The products formed aregiven.

Nu (Enzyme)

Si(

E H

E/C

Nu (Enzyme)

EH

E ,

(1R,2S,6S)

c1(E,Z1

(1R,2S,6R) (1R,2S,6S) (1R,2S) (Scheme 26)

In this second case, we change the configuration of the C(5)—C(6)the resulting products are given. I would like to point out thatmeric cations don't necessarily have to lead to the same products

02(E,E)

double bond:the diastereo—

y or

H

E%J E H E3P(1S,2R,6R) (1S,2R,6S) (1S,2R) (Scheme 27)


c2(E,z)p

gJ4Hre(2)—si(l)-re(6)

Nu (Enzyme)

E5P(1S,2R,6S) (1S,2R,6R) (1S,2R) (Scheme 28)

Now the electrophile assumes a n—position, the (E)—methyl group, however, an a—position and the configuration of the C(5)—0(6) double bond again determinesthe configuration at C( 5)

enantiomeric boat foldings:

B1(E,E) H

E

"... HE"H E...,,,,

(1R, 2S,6R) (1R, 2S,6S) (1R,2S)

(1R,2S,6S) (1R,2S,6R) (1R,2S) (Scheme 30)

si—re—si

B1 (B,z)p

(Scheme 29)

Nu (Enzyme)

si—re—re

E EE ..,,,,, P

EH

Summary of the 8 possibilities

c1(E,E) _____B1(E,Z)

03(E,E)

B2(E,Z)

C1(E,z)

B1 (E,E)

02(E,z)

B2(E,E)

I

I

Ep

EH

+ E,,4

+H

+ E%H

÷

+

+ E4.)%P

+

÷ E4.%P

(Scheme 33)


Both foldings result in products, which have already been found with the chairfoldings; the same also applies to the enantiomeric B2 foldings.

B2(E,E) Nu (Enzyme)

E4.J(1S,2R,6S)

E3P

B2(E,Z)

(1S,2R,6R) (1S,2R) (Scheme 31)

Nu (Enzyme)

E%(iS, 2R,6R)

E%JP(iS, 2R,6S)

E\JP(iS, 2R) (Scheme 32)


From this comparison, several conclusions can be drawn. I will discuss a fewof them. To begin with, can we conclude anything from the hitherto known absol-ute configurations with regard to the stereochemistry of the cyclisation step ?

F(+)—cis—a—irone (Ref.79) from C2(E,E), B2(E,Z)

(— ) —trans—a—irone (79) c2(B,z), B2(E,B)

TIO0 (+)——irone (79) c2(E,z), B2(E,E)c2(B,E), B2(E,Z)

a—carotene, X=Ha—zeacarotene, X=H8—carotenes—carotene, X=Hsemi—a—carotenone ,X=Hlutein, X=a—OHfritschiellaxanthin,X=a-OH

chiriquixanthin A,X=a-OH

chiriquixanthin B,X=-0H

a—doradexanthin,X=-0H

(118) a1(E,E), B1(B,E)c1(E,z), B1(E,Z)

HOCH1L"4,,p

de caprenoxanthin (78) c1(B,E), B1(B,Z)

y—carotene (70) 01(E,E), B1(E,Z)c2(E,z), B2(E,E)

sarcinaxanthin (75) c1(E,E), B1(E,Z)

2—2H—zeaxanthin

(cont.)

(122) c2(E,E), B2(E,E)c2(E,z), B2(B,Z)

\,IH(+) —cis—y—irone

trisporic acid

(79) C2(B,E), B2(E,Z)

c1(B,z), B1(E,E)a2(E,E), B2(E,Z)

(71,72)(119)(120)(120)(121)(77.)

(24)

(76)

(76)

(24)


HO

I2—OH--—carotene (Ref.123) from C2(E,E), B2(E,E)

c2(E,z), B2(E,Z)

HO'p nostoxanthin (11) C1(E,E), B1(E,E)

HOcaloxanthin (ii) 01(E,Z), B1(E,Z)

HOCH2

c.p. 450 (1,78) c1(E,E), B1(E,E)o1(E,z), B1(E,Z)

Scheme 34

As a simplification, it is assumed that each substituent at 0(2), i.e. OH3, OH,isopentenyl, corresponded to the cyclisation inducing electrophile analogous toH+, D+. Even if one leaves the biogenetically not particularly clear ironesout of consideration, one immediately comes to the result that none of thehitherto established absolute configurations suffices to enable a certainconclusion regarding the stereochemistry of the cyclisation step. The nearestone comes to it (the certain conclusion) is in the case of decaprenoxanthinand sarcinaxanthin, where it must concern the C— or B1.-folding. The surveyalso shows quite clearly, however, that there is more than one type of foldingwhich is made use of by the different organisms. But there is still no recog-nisable connection with carotene types (3—, a— or y—carotenes, i.e. u—, s— ory—series) or with the organisation level of the organisms (bacteria, fungi,higher plants).

I would now like to make a few speculations about the —, y— and —trichotomy.These series are not, according to what is known (Ref. 113), biologically inter-.convertible. The origin of the 3 different series has hitherto remained un-solved. It follows, however, quite simply — at least in part — if one postul-ates that the configuration at 0(6) of the formed cyclohexane ring decideswhich series arises, and one can furthermore connect this configuration withthe geometry of the C(5)—C(6) double bond and the conformation of the folding.

Scheme 35

Looking at the cations formed from the C1(E,E)— and C1(E,Z)—foldings, it followsthat different situations are at hand for the elimination of a proton: if ananti—H is removed as proton, which is probable for stereo—electronic reasons,then the product from A will be different to the product from B Furtherspeculation is forbidden by the sparse experimental facts avai1b1e to date.However, I would like to emphasise again that I postulate that this problem(i.e. the transition into the s— n—series) is connected with the geometry of

PAAC 51/3—E


the 0(5)-.C(6) double bond in the acyclic precursor as lycopene or neurosporene.Experimental investigations in this direction are still lacking.

Future biochemical investigations with suitably configurated or labelled carot—enes must give an insight into the problems outlined of the stereochemistry ofthe biological ring—closure. Suitable molecules for such investigations wouldhave to possess the following end groups:

16 18L2iP I IDD3C ( ( D3C

D D D P

A B

D3C D3CH(P = lycopene— or neuro—

C D sporene—residue)

Scheme 36

Since the stereochemistry at 0(2) can be clarified by the elegant technique ofBritton et al. (Ref. 122) and the chirality at 0(6) in the s— and y—series iswell known, we have contented ourselves for the time being with synthesisingthe d3—type (C). Thereby, we are of course aiming at the determination ofconfiguration at 0(1) in p.-, s— and y—carotenes.

Synthesis of 17,17,17,17' ,l7' ,l7 '—d—lycopene (Ref. 124)

Deuterated lycopenes have already been synthesised by different teams (Refs.125, 126, 127). For our objective, however, none of the techniques employed inthe works cited came into consideration (e.g. catalytical deuteration of triplebonds, LiA1D4 -reductions of carbonyl groups, base catalysed H = D—exchange ofazide protons, syntheses with d3-vinyl chloride etc.).

For the selective conversion of CH3(l7) into CD3(l7) we employed the followingreaction, namely between an allylic system and selenious acid to give an allylicseleninic acid () (Ref. 128) which is today interpreted as ene—reaction.This very quickly undergoes a [2,3]—sigmatropic rearrangement and is then

O-CH2[2,3]I c -H

RO"\HH

* HCH3

/ CH2

128

OSeOH

ci CH2OH

H (H20),

—Se(OH)2

CH2CH2

129 130

Scheme 37


,JJC7H7 HOCH2'L CH3OCOL."131 132 133

® ®HOCD2 BrCD2 . CD3

134 135 136

1165

0/0/0137 16'

® 5e02/ethanol; 30—60°. ® Nn02/ethyl acetate; Ag20/NaOH; CH2N2;

20%. ® L1A1D4/ether; 90%. ® PBr3/CaH2/hexane. ® L1A1D4/ether;30%, based on ® Li/HN(Et)2, —78; 49—56%. ® PBr3/py;P(C6H5)3; 47%. ® NaOCH3; crocetindial; 44—61%.

Scheme 38

hydrolytically cleaved. The regioselectivity attains more than 90%. Applicat-ion to geranyl—0—benzylether () (Ref. 129) gave first of all the alcohol, which was then converted in several stages into the d3—compowid andthen by conventional steps into d—lycopene The (E)—configuration of the0(1) double bond was preserved in all steps. This was verified by detailed13C1analysis of the ester 133 and d3'-geranylbenzylether (see Note c). Inthis way with the help of decoupling experiments e.g. on , the 3J—oouplingconstants of the carbonyl C with the methoxyl group were determined (3Jred =

3J0=3.7

HII� ,, 0C-,J0=6.3

(CH2OCH2C7H7''CH3

133'=45 ± 0.5 HzHHI/ H

3J0=0(de— ,2c\ 1D 'J0=l26 ± 1 Hz

D3c coupled) HC H#CH2—OC7H7r CH_0C7H7

HL,9 'J0=130 ± 1Hz OCH" H

136 3Jo=ca. 'J0=128 ± 1 Hz3.0 ± 0.5 Hz

Scheme 39

Note c. We are indebted to Dr. Ulrich Vögeli from the group of Prof. W. vonPhilipsborn for extensive measurements and help in interpretation.


2.96 Hz; 3Jo = 3.7 Hz) and those to the methine proton measured (3Jred = 6.08Hz; 3Jo = 6.3 Hz). Since the range of 6.5 — 7.6 Hz is typical for a cis—relat—ionship whilst it amounts to 12.8 — 14.5 Hz for a trans—arrangement (Ref. 130),the stereochemistry at the double—bond is clarified. A similar process wascarried out for the compound Finally, the signal of the methyl groupabsorbing at lowest field is missing in the 1H—NMR—spectrum of the d—1ycopeneobtained.

With sufficient amounts of d6—lycopene fl at hand (we have about 0.5 g), in-corporation studies with carotene cyclases can now be started. I refer torelevant works by Porter, Goodwin and Davies. When ,f3—carotene is formed onecan expect:

3—carotene:

C, B1)or: (1S,l'S) (from 02, B2)

,c—carotene: (p—ring fromC or B1,

E c—ring from01(E,Z) orB1(E,E))

or: (n—ring fromC or B1,c—ring from02(E,E) orB2(E,Z))

or: (p-ring from02 or B2,

E c—ring fromC1(E,Z) orB1 (E,E))

or: H3C CD3 (p—ring fromC2 or B2,

E E c—ring fromI

(lS,6'R,l'R) H3C D3 C2(B,E)or

Scheme 40

The decision between the various possibilities thus requires that the configur-ation of the products formed at 0(1) and 0(1') in ,—carotene, and at 0(1),0(1') and 0(6') in ,c—carotene can be diagnosed. For 0(6') chiropticalmethods are reliable; for 0(1') in ,c—carotene we have considered the 1H—NNR—method, above all because today it is applicable to sub—milligram quantities.The geminal methyl groups in the c-ring do, of course, have different chemicalshifts, yet, still no assignment of originals has hitherto been made.

Synthesis of (—)—(1R465j—lO JO. l0—d3—homo—a—cyclogeranic acid (Ref. 131)We started with Ca. 1 kg of configuratively known abietic acid (EQ) and degra-ded it in a time—consuming but lucid way to the double—bond isomeric d3—homo—cyclogeranic acids Q, and :

(lR, 1 'R)

(presuming analogous cyclisation steps at both ends)

E


®

çjCOOHCD3 COOCH3

149

f'COOCH3 +

150

D3C

cIIIIii1iT

151

+

D3C

152

1. NBS/Ba003; 2. NaOAc/HOAc (Ref. 132). ® L1A1D4/ether; 98%.

(3 C6H5P/NBS/THF; 80-90%. ® 1. Mg; 2. D20 or Li(Et3BD)/THF;

90—95%. &3 Cr03/acetone/H2504 according to Jones; 95%.

®CH30000H/Na2HPO4; 65—70% based on ( CH3OH/HC1. ® 0/CHC13

MeOll; Pd/BaSO4/H2 or KMnO4/MgSO4/acetone; 40%. ® Pb(OAc)4/hv

>300 nm/benzene; separation by HPLC on SiO 2/AgNO3; Q,45—50% from obtained Q : : = 1 : 1.4 — 1.6 : 0.3 —

0.4 (Ga).

Scheme 41

Subsequently we first of all synthesised the important ,s—carotene asfollows (Ref. 131):

Synthesis of (l'R.6'S)—l6' .16' ,l6'—d3—. s—carotene (see Scheme 42)The crystalline d3—,e—carotene obtained has practically the same CD—spect-ra as the non—deuterated compound. In the 1H—NMR—spectrwn, the signal at 0.91ppm for the C( 16' )—CH3 is missing, whilst that at higher field (0.82 ppm) isunchanged. Thus for the first time an assignment of these signals can be made:that at highest field is cis to the side chain. One must not, however, applythis fact to all intermediates, for in one case the signals reverse. Thus— assuming successful incorporation of — the chirality at c( 1') in , s—carotene can now be spectroscopically determined. Corresponding syntheses ofd6—y,y—carotene and d6—,—carotene are in progress but not yet comple-ted.

COOCH3 COOCH3

140 141

cD3O147

142 R=CD2OH

143 R=CD2Br144 RCD2M9Br145 R=CD3

146 148


DC D3C D3C

150 _______ _______ CHO ® ,' COOCH5

153 154 155

LCHO

156 157

®

158 160

162

I®16 17 19 20 +

17

163 16

(0,91) 0,82

LiA1H; 95%. (j Pfitzner—Moffat oxidation with DCCD on Merri-

field—Polymer/DMSO/py/benzene/CF3000H (Ref. 133); 85%. Wittig

with q—P=CZ3Et, benzene; 75—82%. ® LIA1H4; 95%. ® Nn02/ethyl acetate; 95%. ® Wittig with q13_P=CH_cH=C(Ref. 134); 40—45%. ® 160 + BuM + 80% based on

HBr (48%)/0H2c12/2.5 mm _500; ® Lidl/H isomerisat—

ion; cryst.; 37%.

Scheme 42


D3C

151D3C

164

152 D3jScheme 43

165

I believe that with the syntheses just outlined, a contribution from the chemi-cal side has been achieved which, together with biochemists, can successfullyhelp to solve the problem of the stereochemistry of the biological ring—closurein carotenes.

5 CAROTENOIDS IN ROSE FLOWERS

In conclusion, I would briefly like to refer to one aspect of the carotenoidchemistry which has an eminently aesthetic character. By that, I mean the onlymost recently discovered fact that carotenoids are essential flower pigments inyellow—, orange—, salmon— or copper—coloured garden roses. Hitherto it wasassumed that it concerned flavonoids. Up till now the following have come tolight as typical rose pigments:

OH

OHHO ,

%%O—GIu

°GIu

Scheme 44

166

OCH3

HO

OH

Gu

0..Glu

167

When cyanin 166 predominates, then the flowers exhibit a deep red tone (e.g.Rosa gallica, "Papa Meilland", "Arturo To scanini", "Mr. Lincoln", "Reine desViolettes", etc.). With peonin (i) as the main pigment, the flowers are deeppink (e.g. Rosa rugosa,"Conrad Ferdinand Meyer", "Queen Elizabeth", "Roseraiede l'Hay"). When pelargonin () is the predominant pigment, we have themodern geranium—, scarlet— and brick—red tones. This colouration has only be-come prominent since about 1950. We find it for example in "Gloria mundi","Princess van Orange", "Independence", "Baccara", "Tropicana", "Sarabande",

168 169


"Radar". There can be no true blue roses, since the corresponding anthocyanindelphinin () does not occur in roses. Few rose admirers know that yellowgarden roses (teahybrids, polyantha, floribunda) have also only fairly recentlycome into being (see Note d). They can all be traced back with two exceptions(namely "Gottfried Keller" and "Poulsen's Yellow") to "Soleil d'Or", whichoriginated from crossing the violet—red hybrid perpetual "Antoine Ducher" withpollen from the yellow Rosa foetida persiana. We therefore decided to invest—igate this very influential R. foetida more closely. Sacrificing two year'sblossoms from my garden — it flowers only once a year — gave the followingresults (Ref. 135):

4,5%

OH

,çJ#OHHO17,4%

HONote d. Among other rose classes, yellow roses were found above all amongst

tea roses (e.g. "Safrano", "Gloire de Dijon", "Madame Levet") andamongst noisette roses (e.g. "Marechal Niel", "Alister Stella Gray","Chroxnatella"). They owe their emergence to intercrossing of theyellow china rose "Parks Yellow Tea—scented China" (R. x odorata ochro-.leuca) in the first half of the 19th century.

30,8%

21,9%

9,2%

4,1 %

(mutatoxanthin + apocarotenoids 4.1%)

Scheme 45


70% of carotenoids consist of epoxides. Valadon and Mummery (Ref. 136) havealso ascertained a similarly high percentage of epoxides in modern yellow teahybrids ( elSutters Gold" , "Allgold" , "Jan Spek" , "Gold Crown" , "Gold Gleam","Bossa Nova", "Zambra"). Yellow roses thus owe their colouration mainly tocarotenoids, blends, e.g. "Piccadilly", "Masquerade", possess mixtures ofcarotenoids and anthocyanins.

One can conclude, therefore, that by intercrossing Rosa foetida, the ability toproduce and store carotenoids in high concentration in the petals has for thefirst time become a hereditary characteristic. Accordingly, carotenoids occurnot only in (rose—)hips, which was of course known a long time ago, but arealso essential constituents of the yellow—, salmon—, apricot— and coppercolou—red hybrids. This extension of the carotenoid chemistry into horticulturalareas holds an especial fascination for some of us.

Acknowledgement - I thank the Swiss National Thmds for the generoussupport of our work in the carotenoid field. Substantial contribut—ions were made by the following co—workers: Dr. Richard Buchecker,Walter Eschenmoser, Arnold Hofer, Hanspeter Närki, Gerald Werner,Mrs. Edith Märki—Fischer, Nikolas Vergos, Herbert Cadosch, Dr. PeterRtiedi and Dr. Ulrich Vogeli.

REFERENCES

1. S. Liaaen—Jensen, Pure APP1. Chem. 41 129 (1976).2. G. Britton, W. J. S. Lockley, G. A. Harriman and T. W. Goodwin, Nature 266,

49 (1977).3. A. G. Andrewes, H. KjØsen, S. Liaaen-Jensen, K. H. Weisgraber, R. J. J. C.

Lousberg and U. Weiss, Acta Chem. Scand. j, 3878 (1971).4. R.W. Glass and K. L. Simpson, Phytochem. , 1077 (1976).5. L. Zechmeister and W. A. Schroeder, J. Amer. chem. Soc. , 1173 (1942).6. Y. Tanaka and T. Katayama, Nippon Suisan Gakkaishi , 1229 (1977); ref.

Chem. Abstr. 88, 19017 (1978).7. G. Britton, R. K. Singh, T. W. Goodwin and A. Ben-Aziz, Phytochem. , 2427

(1975).8. G. Britton and T. W. Goodwin, Phytochem. j, 2530 (1975).9. R. Buchecker, C. H. Eugster and C. Litchfield, Helv. Chim. Acta 60, 2780

(1977).10. T. Matsuno, S. Nagata and K. Kitamura, Tetrahedron Letters 4601.11. R. Buchecker, S. Liaaen-Jensen, G. Borch and H. W. Siegelman, Phytochem., 1015 (1976); R. L. Smallidge and F. W. Quackenbush, d. 12, 2481

(1973).12. H. Stransky and A. Hager, Arch. Mikrobiôl. Q, 132, 164 (1970); j, 84

(1970).13. H. Kleinig, W. Heumann, W. Meister and G. Englert, Helv. Chim. Acta 60,

254 (1977).14. J. Gross, M. Carmon, A. Lifshitz and B. Sklarz, Phytochem. , 249 (1975).15. R. Kuhn and E. Lederer, Z. physiol. Chem. (HoppeSeyler) 200, 108 (1931).16. R. Buchecker, S. Liaaen-Jensen and C. H. Eugeter, Helv. Chim. Acta ,

1360 (1976).17. H. Cadosch, U. Vögeli, P. Rtiedi and C. H. Eugster, Helv. Chim. Acta ,

783 (1978).18. G. Britton, H. C. Maihotra, R. K. Singh, T. W. Goodwin and A. Ben-Aziz,

yochem. , 1749 (1976).19. J.-L. Fiasson, N. Arpin and J. Perrier, Canad. J. Biochem. J, 1016 (1976).20. H. Kayser, Z. Naturforech. C, 327 (1977).21. N. Okukado, Bull. chem. Soc. Jap. j, 1061 (1975).22. Y. Tanaka, Y. Fujita and T. Katayama, Bull, of Jap. Soc. Scientific

eries j, 761 (1977).23. Y. Tanaka and T. Katayama, Nippon Suisan Gakkaishi 4, 801 (1976); ref.

Chem. Abstr. , 119902 (1976).24. R. Buchecker, C. H. Eugster and A. Weber, Helv. Chim. Acta , in press.25. A. G. Andrewes, H. J • Phaff and M. P. Starr, Phytochem. , 1003 (1976).26. Y. Tanaka and T. Katayama, Nippon Suisan Gakkaishi j, 807 (1976); ref.

Chem. Abstr. , 119903 (1976).27. K. Harashima, J .—I. Nakahara and G. Kato, Agr. Biol. Chem. , 711 (1976).


28. U. Leuenberger and I. Stewart, Phyto. , 227 (1976); iidem., J. Org.Chem. 4;L, 891 (1976).

29. N. Arpin, W. A. Svec and S. Liaaen—Jensen, Phytochem. , 529 (1976).30. S. S. Dighe and A. B. Kulkarni, Indian J. Chem. 11, 404 (1973); 12, 413,

414 (1974); 3. S. Dighe, S. V. Nanerikar and A. B. Kulkarni, ibid. ,546, 550 (1977).

31. W. Esehenmoser and C. H. Eugster, Helv. Chim. 5, 1722 (1975).32. W. Vetter, G. Englert, N. Rigassi and U. Sehwieter in "CarotenOids" (0.

Isler, H. Gutmann and U. Soims, ed.), p. 230, Birkhuser, Basel (1971).33 . S • Tamura and N. Nagao , biol. Chem. 1393 (1970).34. C. R. Enzell, Pure Appl. Chem. Q, 497 (1969).35. E• A. H. Hall, G. P. Noss, J. H. P. Utley and B. C. L. Weedon, Chem.

2:9J' 586.36. V. G. Mairanovskii, L.—A. Vakulova, N. I. loffe, A. A. Bngovatov, B. I.

Korsunova, T. I. Maksakova and G. I. Samokhvalov, Khim.—Farin. Zh U,111 (1977); ref. Chem. Abst. , 37999 (1978).

37. R. Kuhn and A. Winterstein, Helv. Chim. Acta , 123 (1928).38. P. Karrer and R. Norf, Helv. Chim. Acta , 833 (1931).39. P. Karrer and A. Riiegger, Helv.Chixn. Acta , 955 (1940).40. P. Karrer and C. H. Eugster, Helv. Chirn. Acta , 1934 (1949).41. R. B. Sioda, B. Terem, J. H. P. Utley and B. C. L. Weedon, J. chem. Soc.

Perkin I, ;1l 561.42. Eastman Kodak Co., British Pat l'O97497, 3.1.1968.43. R. Kuhn and A. Winterstein, Helv._Chim. Acta U, 87 (1928).44. A. Ishida and T. Mukaiyama, Chem. Letters 1127.45. J . B. McMurry and M. P. Fleming, J. Amer. chem. 9, 4708 (1974).46. N. Vergos, W. Esehenmoser and C. H. Eugster, unpublished experiments(1978).47. D. OQianu, E. Nicoara and C. Bodea, Rev. Roum. ChAin. , 1085 (1977); see

a'so i;:.t. 2;' 1203 (1974); 765, 925 (1971); , 1965 (1970).48. N. Baranyal, J. Kajtr, Gy. Bujtás and J. Szabolcs, Acta Chim. Acad. Sci.

3;?& 67 (1977).49. G. Nybraaten and S. Liaaen—Jensen, Chem. Scand. B , 483, 485 (1974).50. L. Bartlett, W. Klyne, W. P. Nose, P. N. Scopes, G. Galasko, A. K. Mallams,

B. C. L. Weedon, J. Szabolcs and G. T6th, J. chem. Soc. (C) 2527.51. H. Cadosch and C. H. Eugster, Helv. hui , 1466 (1974).52. P. Karrer and H. StUrzinger, Helv. Chim. Acta , 1829 (1946).53. U. Schwieter, W. Arnold, W. E. Oberhänsli, N. Rigassi and W. Vetter, Helv.

Chim. Acta , 2447 (1971).54. R. Buchecker and S. Liaaen—Jensen, yochem. , 729 (1977).55. P. Molnr and .J. Szabolcs, Acta Chim. Acad. Sci. Hung. , 465 (1973).56. R. Coman, A. P. Leftwick and B. C. L. Weedon, J. Chem. Soc. Perkin I

2140.57. R. D. G. Cooper, .1. B. Davies, A. P. Leftwick, C. Price and B. C. L. Wee—

don, J. Chem. Soc. Perkin I 2195.58. D. B. Rodriguez, Y. Tanaka, T. Katayama, K. L. Simpson, T. —C. Lee and C.

0. Chichester, J. Agric. Food Chern. , 819 (1976).59. B. Märki-Fischer, R. Buchecker and C. H. Eugster, unpublished exp. (1978).60. H. Kj$sea and S. Liaaen—Jensen, Acta Chem. Scand. , 1500 (1971); 0. Pun-.'

tervold and S. Liaaen—Jensen, ibid. , 1096 (1974); B. C. L. Weedon,Pure Appi, Chem. 4, 165 (l976T'

61. 1. B. Johansen and S. Liaaen—Jensen, Acta Chem. Scand. , 315 (1975).62, H. Pfander and U. Leuenberger, Chimia Q, 71 (1976).63. R. Buchecker and S. Liaaen—Jensen, Chim. Acta , 89 (1975).64. B. Johansen & S. Liaaen—Jensen, Acta Chem. Scand. B 28, 949 (1974).65. J. B. Johansen and S. Liaaen—Jensen, Tetrahedron Letters 955.66. J. B. Johansen and S. Liaaen—Jensen, Proc. 4th mt. Symposium on Caroten—

oids, Berne (1975).67. C. J. W. Brooks and J. D. Gilbert, Chem. Commun. flfl, 194.68. R. Buchecker, N. Arpin and S. Liaaen—Jensen, Phytochem. , 1013 (1976).69. W. Eschenmoser and C. H. Eugster, Proc.h mt. Symposium on Carotenoids,

Madison (1978).70. N. Hallenstvet, R. Buchecker, G. Borch and S. Liaaen—Jensen, Phytochem. 16,

583 (1977).71. C. H. Bugster, R. Buchecker, 0. Tscharner, G. Uhde and G. Ohloff, Helv.

Chim. Acta , 1729 (1969).72. R. Buchecker, R. Egli, H. Regel—Wild, C. Tscharner, C.H. Bugster, G. Uhde

and G. Ohloff, Helv. Chim. Acta , 2548 (1973).73. G. Ohloff, B. Otto, V. Rautenstrauch and G. Snatzke, Helv. Chim. Acta ,

1874 (1973); ibid. , 352 (1976).

'(ILGt) t°a 'xenwi.rca 'LLG •d '('Pa 'SttflOS •flptZ8 UtV8U4UC H 'ieII o) spçoueo.x trç uptpoo

'T 5 •oo weto •f 'uopeej ' .0 a pu E a 'eureirer, 'o3T8-r8e o 't9a; o 'sooqz •f 'TBtI9H V •3I 'A3OtflOt •'I •tI

•gj7g :rr i u-p.xea °°s 'mato •' 'UOp88M "I •O •o: P' aqno 3 ' 'Tt88STLI 'P g "8IOH f 'eonqn f •rti:

__________________ 68o1

'T7:T 'uB)I L PU uOa&j ff 'wrnmex s 'eosi ou 668c 'vz:

-eq uozpet.iaj 'uopee rj •ç • ptr8 T9J 'A 'so c3 0 'P1tT19 )j 601 (9c61) U0I2UT

•-t!8BM 'o6iOu qn€j en'pstx] aceu.reo '1eVtciotux ;a'i 'uiz H H 901

______________ (L6T) io: 'T t1lH

•pov •tUTTW °V 'uopee •r[ '3 • ptI8 •A:o 'otoqz 'r 'sso L01 . (9L61)

L99 't uozpeia 'TPH 'rI PUB t9J •A:o 'sooqz r 'iraxa w 901 6?oT '•rr ' up.ze oo •wei r 'uapue4 ptr oix 'i O1

_____________ (cL6t) c17 'oi:

uose •Ti: •f 'XTaXflOTTTA 2 P 'j '8Ufl '.Ieu'ezo 'd 'iOT

___________ Lt'T 'IT 'i upxa

•oo met 'p 'uopae rj • • ptr eqnoj 'j 'xeqaor] 'imtrj

(9L61) Lf7T '7V weto iiuy a.xtu 'aooqz r ot '(9L61) L6 'rv 'weo TdV eznj 'soj f

_______________ (GL6I)

' °V WTtO ATGH 'JTt H j pU zesowuetosa M 'TTcA fl 001

__________________________ ____ (L6T) euzea

'SpTOUeOX3 UO urnçSoduL& UI ''OO.I3 'ze.sn 'H 'ü esoweoa 'i '66 iL 'trniinuO

'uopeej r1 j ' pus essn 'p ' 'esnotsnH ' 'w 'errcAea 'a '96 '(LL6i) 'ox

AN 'ssez wnue 'd 'TOTUaI 'H 'M pure IetnrLj 'r 'a) 'Axspaeu43 s onpoxit neii eucxw1 'zesuer—ueevI ' PU uesutor 'a 'r 'LG

'(LL6T) L9 'T 'weoo 'ties -uefueeçrJ ' pure urew1ee-ç$ '1k 'H 'tO.IOa '0 'tIeStIeIOW 'j 'ieqzzej ' '96

________ •Grt 'T se uozpete 'uzopeej r1 ' ' pure t9 'JVf) 'ssow 'a 'o 'Pzrnuzea 'x 6 '(9L61) 691 'IV 'weuW [d e.zria 'uopee "I '0 'a '6

'L9i 'TTT 'unwtaoo 'wetj 'uopaeM "I ' ' pure seow ' 'xdot) ' 'V 'c6

— ___________ '(LL6I)

çj7 'ic 'uw ov 'uesuer_ueecrr pure uo.zoa ' "'°a 'r '6 '(L61) o1 '7 '1oTstrc1uxezuxua 'z

'zeqe ' pure ureTuzonew 'w t(L61) 'ocqo.nççj 'u.[o.x'y 'xeqe 'V '16

__________________________________ '(8L61) uzosp

'spTouzeoreo wruçsoaw.& u] w 'oo.xa 'iesna 'H ') pure xeoeuona 'i '06 '(LL6T) 6o1 ' ov 'urcu.To 'ATOH 'eoN ' pum eizuei ' '.xe1ua ' '6 '(9L61) 6001 ' 'wet(ooIAta 'a pure seAe.xpuy ' 'v '

________________ '(L61) GG ' 'pureo$ 'weu 'uesuef.ueBe-çr ' pure uesxepe 'j 'ioxo 'o 'treuLiee 'V 'L9

'(L61) 6091 ' qov 'mco 'reH 'xe.xzt 'a pure teuos.ua 'I 99

_______________ '(9L61) 9L

' °V UITTW 'TeH 'H '0 pure pej 'a 'iiep. 'fi 't(O9OQ 'H '9 '6c 'TT s,xe4err uo.'peue.x.

— 'uopee "i 'o 'a pure u.ci 'AVO 'so1oqz$ ' 'ssoj, a 'o 'Aoe;poo 'a 'i '(L61) eu.xeaspTouxeoreo uzo wnçsodwA 'u uq4 'oo.ia '.zeoe1ona 'i 'c

_________________ '(iL6T)

09 ' '3O 'WetO 'iewy 'f r pui zuraxa " 'f 'wçqrejg '0 'f '?8 (8L61) 8 'T' ov 'wTuw 'ATeH '.xesna 'H '0 pus zesowueTosa '1k '19

L 'ITT 'tmwzio 'weu 'uxop

-eeM 'i 'o 'a PUB esow 'a 'o 'tzeppB 'V 'qursu 'a 'a 'idotj 'i 'V '09 '(1L61) 9LLI 'F OV 'flITtED 'AISH 'JO1tE0 '0 PUB t.onxsue).nB3j A '6L

_______________ '(9L61) tI2

'o a 'pumo$ 'wet 'uesueruxee-crr ' pure taxoa 'e 'seAezpuxy 'o V '8L '(i7L6I) 1c9 '7 OV 'flITtED 'AISH (L6T) i'T

' CtL61) 61 ' TWTtED 'zesrta H '0 UB WWBH 'a 'Ie)ToetEona 'H 'LL 96 'iIT tmwuxo 'weu. 'seAalpuV '0 'V pure .xetEsow H 'uT8tEUTa 'V '9L

'(LL61j3 'T 'T 'uxesuefueBeçrf '9 pure IeqzIe '9 (LL61) 1? 'C pureo 'uIeu3 B0V 'uesuefueBwçrf ' pus axoa '0 'seAezpuv '0 •V 'L

'(LL6I) L9L? ' BOV TtED 'AIeH 'IBTA '0 pure J°1tE0 '0 'iL

coc spiouoi jo isimaqooais pu tsiuiip 'uotzT1aiqD


114. G. Britton, Pure Appl. Chem. 4j, 223 (1976); "Chem. Biochem. Plant Pip—ments" (T. W. Goodwin, ed.), Vol. 1, p. 262, Academic Press, London(1976).

115. B. H. Davies, Pure Appl. Chem. , 1 (1973).116. J • W. Porter, Pure_Appl. Chem. 20, 449 (1969).117. T. W. Goodwin, Pure Appl. Chem. 20, 483 (1969).118. J. D. Bu'Lock, D. J. Austin, G. Snatzke and L. Hruban, Chem. Commun.

255; D. J. Austin, J. D. Bu'Lock and D. Drake, Experientia 26, 348(1970); J. D. Bu'Lock, B. B. Jones and N. Winskill, Pure Appl. Chem., 191 (1976).

119. R. Buchecker and C. H. Bugster, Chim. Acta , 1124 (1973).120. R. Buchecker and C. H. Eugster, Chim. Acta , 327 (1971).121. R. Buchecker, H. Yokoyarna and C. H. Eugster, Helv. Chim. Acta , 1210

(1970).122. G. Britton, W. J. S. Lockley, N. J. Patel, T. W. Goodwin and G. Englert,

Chem. Commun. flfl, 655.123. R. Buchecker, C. H. Eugster, H. Kj$sen and S. Liaaen—Jensen, Helv. Chim.

Acta , 2899 (1973); iidem, Acta Chem. Scand. B 28, 449 (1974).124. A. Hofer and C. H. Eugster, unpublished experiments TT978).125. U. Scbwieter, G. Englert, N. Rigassi and W. Vetter, Pure Appl. Chem. 20,

414 (1969).126. k. Eidem and S. Liaaen—Jensen, Acta emScand. B 28, 273 (1974); J. E.

Johansen and S. Liaaen-.Jensen, , 301, 349 (1974).127. H. Brzezinka, B. Johannes and H. Budzikiewicz, Z. Naturfrsch. b, 429

(1974).128. U. T. Bhalerao and H. Rapoport, J. Am. chem. Soc. , 4835 (1971); D.

Arigoni, A. Vasella, K. B. Sharpless and H. P. Jensen, d. , 7917(1973); N. Rabjohn, Organic Reactions , 261 (1976).

129. L. J. Altman, L. Ash and S. Marson, hesis 1914 129.130 U. Vögeli and W. v.Philipsborn, Org. Nagn. Res. 7, 617 (1975).131. H. P. Mrki, G. Werner and C. H. Eugster, unpublished experiments (1978).132. 0. Jeger, 0. Dtirst and G. Biichi, Chim. Acta Q, 1853 (1947).133. M. Weinshenker, C. Id!. Shen and J. Y. Wong, 0rg Synth. , 95 (1977).134. 0. Isler, L. H. Chopard—dit-Jean, M. Nontavon, R. Rüegg and P. Zeller,

Helv. Chim. Acta Q, 1256 (1957).135. R. Buchecker and C. H. Eugster, Helv. Chim. Acta Q, 1754 (1977).136. L. R. G. Valadon and R. S. Mummery, Phyton , 151 (1968).

jie bluee - iupac.org fl9't2-4jtp68 twa9'lt9'ple wtcjcueaa aiw tjjtcj1 i eoogeq w2...

Documents