512 J.C.S. Perkin IPyrroles and Related Compounds. Part XXVIII.l p-Keto-esters in thePorphyrin SeriesBy Michael T. Cox, T. Trefor Howarth, Anthony H. Jackson,t and George W. Kenner,' The RobertRobinson Laboratories, University of Liverpool, P.O. Box 147, Liverpool L69 3BXPorphyrin carboxylic acids are converted into the corresponding P-keto-esters by reaction between the acidchlorides and t-butyl methyl malonate, followed by treatment with trifluoroacetic acid. Iodine in methanol reactswith the magnesium chelate of the keto-ester anion derived from rhodoporphyrin-XV dimethyl ester, giving thechelate of 1 O-methoxyphaeoporphyrin-as dimethyl ester. The significance of these results in relation to thebiosynthesis of chlorophyll is discussed.GRANICK'S classical investigations established thatprotoporphyrin-IX (1) is a biogenetic precursor of chloro-phyll-a (2).2 Perhaps the most intriguing questionabout the complicated transformation of (1) into (2) ishow the carbocyclic (frequently termed isocyclic) ring isformed.In Part 11,3 a speculation about formation ofa carbon-carbon bond by a Michael-type transfer ofelectrons from an enolate of a p-keto-carboxylic side-chain at position 6 to the y-position of the macrocyclewas advanced. A different, but not dissimilar specula-tion about intramolecular reaction between an enolisedP-keto-ester side-chain and the nucleus, activated bydiprotonation, was put forward by W~odward.~ Thep 2 CH MeE tMe( 1 1 ( 2 1purpose of the present work was to prepare porphyrinP-keto-esters and to explore their beliaviour underconditions supposedly f avourable to cyclisation.Thiswork paved the way for investigations more closely con-nected with the biosynthesis of the carbocyclic ring (seefollowing paper).Studies in the synthesis of monopyrrolic p-keto-esters had shown that acylation of dialkyl malonates isa serviceable route in that series. It seemed likely to besuitable also in the porphyrin area because the requisitecarboxylic acids are available. Looking forward tosynthesis of compounds bearing easily reduced vinylsubstituents, we chose t-butyl (rather than benzyl) asthe temporary malonate substituent. Initially, theroute was tested with a porphyrin having only onecarboxy-group and an arbitrary array of methyl andPresent address: Department of Chemistry, UniversityCollege, Cardiff.Part XXVII, T.T. Howarth, A. H. Jackson, and G. W.Kenner, preceding paper.S. Granick, J . Biol. Chem., 1948, 172, 717; The HarveyLectures, 1950, 44, 220.A. C . Jain and G. TV. Kenner, J . Chem. SOC., 1969, 185.R. B. Woodward, I n d . chim. belga, 1962, 27, 1293.ethyl groups at the other seven peripheral positions,l andour scheme was as follows:CO,But I RC02H + RCOCl __t RCO*CH*CO,Me _t(3) (4) (5) RCO*CH,CO,Me(6)R = 2,6,7-triethyl-l,3,5,8-tet rameth ylporphinylThe carboxylic acid (3) was obtained from the methylester; its visible absorption spectrum showed theexpected change from rhodo-type6 (in pyridine) toaetio-type (in methanolic 0-lM-sodium methoxide),corresponding to destruction of the conjugation betweenthe carbonyl group and the macrocyclic x-electrons ;this diagnostic test was important in our later work.The acid chloride (4) was treated directly with a largeexcess of methyl t-butyl sodiomalonate, giving a 71yield of the keto-malonate (5), without any diacylationbeing detected.The structure (5) was not in doubt,but the highest-mass peak in the mass spectrum corre-sponded to complete loss of the keto-malonate side-chain,except for one hydrogen atom. The keto-malonate (5)was considerably enolised (lH n.m.r. spectrum in CDCl,),unlike the monopyrrolic analogue^.^ Its visible spec-trum was of the oxorhodo-type (in chloroform) ratherthan of the rhodo-type, but on ionisation it changed tothe aetio-type (in methanolic sodium methoxide).Thet-butyloxycarbonyl group was removed from (5) by coldtrifluoroacetic acid, and the resulting keto-ester (6)(obtained in 77 yield) possessed the expected spectralproperties, except that the highest-mass ion correspondedto acetylporphyrin rather than the molecular ion. Lossof 58 mass units proved to be the rule in the mass spectraof keto-esters throughout this series of investigations,but more recently, true molecular ions have been ob-served (MS12 or Varian CH 6D spectrometer instead ofMS9).Encouraged by these results, we turned to synthesis ofthe P-keto-ester (7d), which is closely related to thesupposed intermediates in chlorophyll biosynthesis.The synthesis involved differentiation of the two esterfunctions in rhodoporphyrin-XV dimethyl ester (7a).This was achieved by alkaline total hydrolysis and acid-catalysed partial re-esterification. It was predicted that5 T.T. Howarth, A. H. Jackson, J. Judge, G. W. Kenner, andD. J. Newman, J.C.S. Perkin I , 1974, 490.6 J. E. Falk, ' Porphyrins and Metalloporphyrins,' Elsevier,Amsterdam, 1964, p. 731974 513protonation of the nuclear carboxy-group would beinhibited by the double positive charge on the macro-cycle, whereas the side-chain carboxy-group would re-tain almost aliphatic disposition to esterification. A( 7 )C02Mea; R = CO2Meb; R = C 0 2 HC; R = C O , E td ; R = CO.CH2.CO2MeM e n NH N E ta i R = Hb; R = O M emonoester was indeed obtained without difficulty, andits structure (7b) was confirmed by (i) the change in itsvisible absorption spectrum from rhodo-type (in chloro-form) to aetio-type (in methanolic sodium methoxide) ,(ii) the lack of the low-field methoxy-resonance (ringcurrent effect) in the lH n.m.r.spectrum (in trifluoro-acetic acid) compared with that of the diester (7a),rsquo; and(iii) the conversion into the acid chloride and thenethanolysis, to give an ethyl methyl diester which stillshowed the loss of 73 mass units in the mass spectrum,corresponding to fragmentation of the propionate methylester side-chaim8 The last criterion also affordedevidence for homogeneity of the acid chloride, which wastreated with methyl t-butyl sodiomalonate as in thecase of (4).In this instance, the keto-malonate couldnot be crystallised, but it yielded the crystalline keto-ester (7d) in an overall yield of Soy0 from the carboxylicacid (7b)l having the expected spectral properties, in-cluding a rhodo-type visible absorption spectrum.With the two porphyrin fi-keto-esters (6) and (7d) inhand, we were able to attempt non-enzymic closure of thecarbocyclic ring. Woodward made the appealingproposal that discharge of a porphyrin dication by anenolised p-keto-ester system would lead to a phlorinsalt containing the carbocyclic ring, and that the reactioncould be carried forward essentially irreversibly in thebiosynthetic direction through oxidation of the phlorinnucleus back to the porphyrin level. Both keto-estersare appreciably enolised in deuteriochloroform (n.m.r.spectroscopy) and they were readily converted into thegreen diprotonated species, but evidence for cyclisationwas not obtained from experiments with addition ofhydrogen chloride, sulphuric acid, or boron trifluoride-ether complex to solutions in methylene chloride, chloro-form, methanol, benzene, ether, or tetrahydrofuran.Our own proposal was less specific. Activation of theketo-ester, by either ionisation or enolisation, wasenvisaged, and it was considered that the electrons dis-rsquo; P.A. Burbidge, G. L. Collier, A. H. Jackson, and G. W.Kenner, J . Chem. SOC. (B), 1967, 930.placed from the y-position during ring closure wouldhave to be accepted by either an oxidising agent or aconjugated double bond in the magnesium chelate whichwas probably involved in the cyclisation.A magnesiumchelate would be instantly decomposed by acids, andtherefore base-catalysis was indicated. All attempts,however, to initiate cyclisation by simple base-catalysedreactions of the keto-esters or their metal chelates (zincor magnesium) were unsuccessful. At this point, ourthoughts were decisively influenced by a communicationwhich described the ready oxidation of porphyrinmagnesium chelates to x-cation radicals. It stimulatedus to explore the oxidative mechanism for acceptanceof the electrons displaced from the y-position.Fuhrhop and Mauzerall9 showed that removal of oneelectron from the magnesium chelate of octaethylpor-phyrin was accompanied by a spectral change from404, 542, and 578 to 398 and 683 nm.The mag-nesium chelate prepared from (7d) by the action ofn-propoxymagnesium bromide and transesterificationwith methanol was, however, unaffected by iodine inmethanol, a reagent which was effective with magnesiumoctaethylporphyrin. This result was not surprising, inview of the electron-withdrawing power of the group atposition 6, whether in the keto or the enol form. Bro-mine or iron(II1) chloride in methanol did produce thedesired absorption at 680 nm, but there was no furtherspectral change indicative of cyclisation, during severalhours. On the other hand, addition of a little aqueous10 sodium carbonate to the magnesium chelate movedthe absorption maxima from 418, 554, and 602 to410,544, and 580 nm, presumably by proton abstractionE t M e E t M eEtM e( 1 0 Iyielding the sodium enolate (9). Addition of iodine tothis solution caused a rapid change to maxima at 550 and620 nm (diffuse).Acidic removal of the magnesiumand subsequent chromatography yielded two porphyrinicproducts. The major product (7 yield) was shownconclusively by mass spectroscopy and comparison withan authentic sample, prepared from methyl phaeophor-bide-a, to be 10-methoxyphaeoporphyrin-a, diinethylester (8b). The minor component was apparently (massspectrum) an oxophlorin derived from (8b). If trans-esterification with sodium methoxide in methanol wasnot performed after magnesium insertion, the 7-n-propyl ester, homologous to (8b), was produced (mass13 A.H. Jackson, G. W. Kenner, K. M. Smith, R. T. Aplin,H. Budzikiewicz, and C. Djerassi, Tetrahedron, 1966, 21, 2913.@ J.-H. Fuhrhop and D. Illauzerall, J . Amer. Chem. SOC., 1968,90, 3875514 J.C.S. Perkin Ispectral molecular ion at 664 and loss of C2H,02 fromthe 10-position, instead of C4H,0, from a 10-propoxy-group). The 10-methoxyphaeoporphyrins gave, as ex-pected, a negative reaction in the ' phase test ' for anenolisable hydrogen atom in the isocyclic ring.The mechanism of this cyclisation deserves somecomment although any such considerations are neces-sarily speculative. Our preliminary announcement l ostated that ' formation of the isocyclic ring may beenvisaged as reaction between the radical cation derivedfrom the magnesium-macrocycle complex and the radicalderived from the enolate '.We had in mind, in additionto the work of Furhrhop and Ma~zerall,~*ll the oxidativecoupling of enolates.12 Dolphin and his co-workers l3have specifically preferred to regard our cyclisation asaddition of the unoxidised enolate to the magnesiumporphyrin x-dication,14 generated by removal of twoelectrons from the macrocycle. At first sight there seemslittle ground for such a supposition, because iodine is notsufficiently powerful an oxidising agent to convertmagnesium octaethylporphyrin into the dication. Indefence of Dolphin's suggestion, however, we point outthat the enolate ion may be supposed to lower the oxida-tion potential of the macrocycle conjugated to it, sothat double electron abstraction by very weak oxidisingagents becomes feasible.Until physical methods are re-fined to the point where they can differentiate betweencation radical-enolate radical and dication-enolateanion situations in conjugated systems, further discus-sion seems pointless. In any case, removal of two elec-trons from the enolate of the magnesium chelate (9) andformation of a carbon-carbon bond leads to a new chelate(10) of an i~oporphyrin.~,~~ Loss of a proton, as cx-pected, would then give the magnesium chelate of (Sa).The 10-methoxy-substituent in (8b) is an unwelcomeexcrescence. It is easy to see how it arises from theanion of the chelate of (sa), a reaction which is wellknown with the magnesium-free porphyrin.15 In prin-ciple, it should be possible to achieve closure of thecarboxyclic ring without this unwelcome subsequentreaction.We made some experiments along these lineswith the magnesium chelate of (7d) without any success,but this work was suspended when we discovered a veryefficient cyclisation of (7d) and its 2-vinyl congener to thethallium(II1) chelates of (Sa) and the corresponding vinylphaeoporphyrin respectively.16 Nevertheless, the pre-sent work is, in our dpinion, biogenetically significant inindicating the probable viability of a biological pathwayfrom the series of magnesium chelates of porphyrinsderived from protoporphyrin-IX (1) to the chlorophyllseries through electron abstraction from enolates ofmagnesium chelates of porphyrin P-keto-esters.M.T. Cox, T. T. Howarth, A. H. Jackson, and G. W. Kenner,J.-H. Fuhrhop and D. Mauzerall, J . Awzev. CJzem. Soc., 1969,l2 0. H. Matteson and C. A. Wachtmeister, Tetrahedron Letters,l3 D. Dolphin, R. H. Felton, D. C . Borg, and J. Fajer, J . Ameu.J . Amer. Chew. SOC., 1969, 91, 1232.91, 4174.1967, 1855.Chem. SOC., 1970, 92, 743.EXPERIMENTALh1.p.s were measured on a microscope hot-stageapparatus. Neutral alumina (Woelin ; Brockmann gradeV) was used for chromatographic separations. Visibleabsorption spectra were measured with a Unicam SP 800spectrophotometer, lH n.m.r. spectra with a Varian A-60or HA-100 spectrometer, and mass spectra with an A.E.I.MS9 instrument (at 50 PA and 70 eV, with direct inlet andsource temperature between 200 and 220").2,6,7-Triethyl- 1,3,5,8-tetranzeth~l~orphin-4-carboxylic Acid(3) .- 2,6,7-Triethyl-4-rnethoxycarbonyl- 1,3,5,8-tetrameth-ylporphin (150 mg) in pyridine (150 ml) and 10 w/vpotassium hydroxide-methanol ( 150 ml) was heated underreflux during 3 h, during which time the initial rhodo-typevisible absorption spectrum (A,,,.5 10, 548, 575, and 633 nm)changed to the aetio-type (A, 500, 535, 568, and 620 nm).The solution was poured into water (500 ml) and neutralisedwith 2~-hydrochloric acid. The precipitated porphyrin wasextracted with methylene chloride-pyridine (ca. 10 : l ) , theextract was washed with water (2 x 100 ml), dried (MgSO,),and evaporated to leave the porphyrin, which crystallisedfrom pyridine-methanol as purple needles (125 mg, 86),map.320" (Found: C, 75.3; H, 7.1; N, 11.3. C,1Hamp;4Ogrequires C, 75.3; H, 6.9: N, 11-3), -r (CF,*CO,H) -1.80,-1.13, -0.92, and -0.86 (4 wzeso-H), 5-81 (9) and 8.19 (t)(3Et), 5.89 (3-Me), 6*28(3Me), and 13-8br (NH), k, (pyridine)4.07 (c 186,000), 507 (11,000), 544 (13,800), 571 (SOOO),and 628 nm (1890), A,, (CHC1,HCI) 423 (amp; 212,000), 559(13,700), and 607 nm (10,500), Lx. (O*lM-NaOMe-MeOH)396 (c 173,000), 498 (13,100), 534 (10,200), 568 (6360), and620 nm (3540).Chloride (4) .-The foregoing porphyrincarboxylic acid (50mg) was dissolved in thionyl chloride ( 1 ml) and kept a troom temperature for 1 h. Evaporation in vacuo gave agreen residue which was used immediately, without purifica-tion, in subsequent reactions.A portion of the acidchloride was treated with methanol; t.1.c. showed almostquantitative formation of the methyl ester.2,6,7- Triethyl-4- (wzethoxycavbonyl) -t-butoxycarbonylacetyll-1,3,5,8-tetramethylporphin (5) .-Methyl t-butyl malonate(0-5 g, 22 mol. equiv.) in dry tetrahydrofuran (40 ml) wastreated with sodium hydride (63 mg, 20 niol. equiv.).After the initial brisk effervescence had subsided, thesuspension was left a t room temperature during 30 min,then added to the foregoing acid chloride (from 70 mg ofcarboxylic acid) and the mixture was shaken vigorouslyuntil all the porphyrinic material had dissolved. Aftera further 15 min a t room temperature, the solvent wasevaporated off and the residue was partitioned betweenmethylene chloride and 2~-hydrochloric acid (150 ml; 2 : 1).The organic layer was washed with water (3 x 60 ml),dried (RIgSO,), and evaporated to dryness; the residue wastreated with an excess of diazomethane in ether.After30 min a t 0" the solution was washed with dilute aceticacid, aqueous 10 sodium hydrogen carbonate (50 ml), andwater, dried (MgSO,), and evaporated. The oily residuewas chromatographed on grade V neutral alumina (elutionwith methylene chloride and then chloroform) to give anamorphous red solid from the chloroform eluates. Re-I4 J. Fajer, D. C. Borg, ,4. Forman, D. Dolphin, and R. H.Felton, J . Amer. Chem. Soc., 1970, 92, 3451.l5 H. Fischer and A. Stern, ' Die Chemie des Pyrrols,' Aka-demischc Veilag, Leipzig, 1940, p.176.l6 G. W. Kenner, S. W. McCombie, and K. amp;I. Smith, J.C.S.Chem. Comrn., 1972, 844; .J.C.S. Perkin I , 1974, 627.2 6,7-TrietJayl- 1,3,5,8-tetrarnethyl~or~hin-4-carbony1974 515crystallisation from chloroform-methanol gave the Product(65 mg, 71) as minute purple needles. An accurate m.p.could not be determined owing to thermal decomposition,but an estimate, based on observation when the sample wasplaced on the preheated hot plate, is 210-215" (Found: C,71.8; euro;3, 7-1; pu', 8.8. C3,H,,N40, requires C, 72.0; H,7.1 ; N, 8-6). The n.m.r. spectrum (CDC1,) was extremelycomplex, owing to a high level of enolisation. Resonancesconclusively assigned were: z 5-90 (3-Me), 8-55 (But), and8-26 (t, 3CH,*CH,); v,, (CHCl,) 1625 and 1715 cm-l,amp; (CHCl,) 407 (E 173,000), 507 (8870), 543 (11,700),568 (9400), and 620 nm (1770), A,, (CHC1,-HCl) 429(E 218,000), 570 (11,000), and 618 nm (7100), amp; ( 0 .1 ~ -NaOMe-MeOH) 396 (E 173,000), 497 (13,200), 534 (10,600),566 (6340), and 619 nm (4150).2,6,7- Triethyl- 4-met hoxycarbonylacetyl- 1,3,5,8-tetra,methyl-povphin (6).-The foregoing porphyrin (25 mg) in trifluoro-acetic acid (0.63 ml) was kept a t room temperature during10 niin. The solvent was evaporated off and a solution ofthe residue in methylene chloride (30 ml) was washed withaqueous 10 sodium hydrogen carbonate (20 ml) and water(2 x 20 ml), dried (MgSO,), and evaporated. The residuewas chromatographed on grade V alumina, with methylenechloride as eluant, to give a red solid which crystallised fromether.Recrystallisation from methylene chloride-ineth-anol gave the Product (16.5 m g 77) as minute purpleprisms, which decomposed above 260" (Found : C, 74-3 ; H,7-1; OMe, 5.7. C3amp;3,hT,O, requires C, 74.2: H. 7.0; OMe,5.6), T (CDCl,) -0.45, 0.43, 0.65, and 0.81 (4 meso-H),5.55 (2H) and 6.48 (3H) (CO-CH,CO,Me), 6.13 (3-Me), 6.65,6.71 and 6-83 (3Me), and 8.33 (m, SCH,*CH,), vmx. (CHCl,)1645 and 1725 cm-l, Lx (CHCl,) 410 (E 163,000), 510 (8680),547 (12,400), 571 (8260), and 628 nm (1450), h,, (CHCI,HCl) 423 (E 178,000), 560 (12,000), and 605 nm (SZSO),A,, (0.h-NaOMe-MeOH) 397 (E 155,000), 499 (11,600),534 (8990), 568 (5870), and 619 nm (3040), wz/e 492 (63)and 450 (1OOyh).2,4-DiethyZ- 7- (2-methoxycarbonylethyZ) - 1,3,5,8-tetramethyl-~orphin-6-cnrbo.~ylic Acid (7b) .-2,4-Diethyl-6-methoxy-carbonyl-7- (2-niethoxycarbonylethyl) - 1,3,6,8-tetramethyl-porphin (165 mg) in pyridine (30 ml) was treated withpotassium hydroxide (3 g) in methanol (30 ml).Thesolution was heated under reflux during 30 min, then leftovernight a t room temperature, poured into 2~-hydrochloricacid, and neutralised with dilute ammonium hydroxide.The solid was extracted into methylene chloride-pyridine(ca. 5 : 1) and the organic layer was washed with water,dried (MgSO,) , and evaporated. The residue was dissolvedin 5 w/v sulphuric acid-methanol (250 ml) and set asidein the dark for 16 h. The solution was poured into dilutcammonium hydroxide and neutralised with 2~-hydrochloricacid.The insoluble solid was extracted into methylenechloride (300 ml); the extract was washed with water,dried (MgSO,), and evaporated. The resulting solid crystal-lised from tetrahydrofuran-benzene to give the product (1 10mg, 68) as purple needles, m.p. 310deg; (Found: C, 71-6;H, 6-7; N, 10.3. C,,H,,N,O, requires C, 71.7; H, 6.6; N,10.1), s (CF,*CO,H) -1.84, -1.14, -0.94, and -0.88(4 meso-H), 5.80 (9) and 8.18 (t) (2Et), 5-88 (5-Me), 5.41 (t),6.63 (t), and 6-23 (s) (CH,-CH,*CO,Me), and 6.26 (3H) and6-28 (6H) (3Me), Amax. (CHCl,) 406 (E 187,000), 509 (10,200),548 (13,600), 575 (SOOO), and 631 nm (1970), A,, (CHC1,-HC1) 425 (E 222,000), 563 (13,600), and 610 nm (9550),A,, (0-lM-NaOMe-MeOH) 397 (E 219,000), 497 (17,000),533 (12,100), 569 (8110), and 621 nm (4910).2,4-~iet?zyl-7-(8-methox3.lcnrbo~z~~~et~z~1)-1,3,5,~-tetramethyZ-fiovphin-6-carbonyZ ChZoride.-The foregoing acid (1 10 mg)in thionyl chloride (6 ml) was kept at room temperatureduring 1 h, then evaporated to give the acid chloride as agreen residue which was used immediately, without purifica-tion.Treatment of this acid chloride (from 5 ing of the corre-sponding acid) with dry ethanol gave a solid which was re-crystallised from chloroform-methanol to give lustrouspurple prisms (3.7 mg, 73) of 6-ethoxycarbonyl-2,4-di-ethyl-7- (2-methoxycarbonylethyl) - 1,3,5,8-tetramethylpor-phin (7c), identified by its mass spectrum m/e 580 (amp;?+)I.2,4-Diethyl-6- (met hoxycarbonylacetyl) - 7- (2-methoxycarb-onylethyl) - 1,3,5,8-tetramethyZpor~h~n (7d) .-To methyl t-butyl malonate (3-8 g) in dry tetrahydrofuran was addedsodium hydride (0.52 g), and after the brisk effervescencehad subsided the suspension was left a t room temperatureduring 30 min.The foregoing acid chloride (from 110 mgof the carboxylic acid) in methylene chloride (10 ml) wasadded and the stoppered flask was shaken during 30 min a troom temp. The solvent was evaporated off and the residuewas partitioned between methylene chloride (100 ml) and2~-hydrochloric acid (50 ml) . The organic layer was washedwell with water, dried (MgSO,), and evaporated to dryness,and the residue was chromatographed on grade V alumina(elution first with methylene chloride and then with chloro-form).Concentration of the chloroforin fraction gave a redgum (the keto-diester) which could not be crystallised andwas therefore taken into trifluoroacetic acid (3 ml) and lefta t room temperature during 20 min. Methylene chloride(50 ml) was added and the solution was washed with aqueous10 sodium hydrogen carbonate (2 x 60 ml) and water(2 x 50 ml), and then dried (MgSO,). Evaporation gaveminute purple prisms which were purified by chromato-graphy on grade V alumina (elution with methylenechloride). The resulting purple solid crystallised fromchloroform-petroleum (b.p. 60-80") to give the firoduct(55 mg, 46 based on porphyrin acid) as purple needles,m.p. 282-284" (decomp.) . Further recrystallisation gavecrystals with m.p. 292-294" (decomp.).These m.p.s weredetermined by placing the sample on the hot-stage at ca.250" with subsequent rapid heating. If the sample wasplaced on the block a t ca. 216" and heated rapidly, it did notmelt below 310". If the sample was placed on the block a t260' i t melted immediately (Found: C, 71.0; H, 6.6; N,9.3. C3,H4,N4O5 requires: c , 71.0; H, 6.6; N, 9*2),T ( 0 . 0 5 ~ in CDCl,) * - 0.46, 0.35, 0.40, and 0.43 (4 meso-H),5-43 (s, disappeared on shaking with NaOD, CO*CH,*CO,Me),5.76 (t), 6.83 (t), and 6-41 (s) (CH,*CH,*CO,Me), cat. 6.2 (m),8.28 (t), and 8.30 (t) (2Et), 6-12 (amp;Me), 6.39 (CO*CH,*CO,-Me), 6-60 (6H) and 6.69 (3H) (3Me), and 14-26br (2NH),Am, (CHC1,) 410 (E 222,000), 509 (12,100), 544 (15,900),673 (10,700), and 630 nm (2070), 1- (CHC1,-HC1) 423(E 267,000), 560 (18,000), and 606 nm (10,900), Lx.( 0 . 1 ~ -NaOMe-MeOH) 390 (E 172,000), 497 (12,900), 533 (9650),568 (6350), and 620 nm (3770), v- (CHCl,) 1620, 1645, and1745 cm-1, m/e 550 (23) and 508 (100).10-Methoxyphaeopo~fihy~i~z-a, Dimethyl Ester (8b) .-(a)Magnesium 2,4-diethy1-6- (2-methoxycarbonylacetyl) -7-(2-methoxycarbonyZethyl)-1,3,5,8-tetra~nethylporphin. The fore-going porphyrin P-keto-ester (27 mg) was treated with anexcess of a suspension of n-propoxymagnesium bromide inn-propanol a t 75-78" (oil-bath) during 4 h under nitrogen.* This spectrum was complex owing to the presence of bothketo and enol forms516 J.C.S. Perkin IThe solution was evaporated iut vacuo and the residue waspartitioned between ether (30 ml) and water (30 ml).Theorganic layer was washed with a solution of disodium hydro-gen phosphate (2.5 g) and ammonium acetate (2.5 g) inwater (25 ml), then with water (2 x 20 ml), dried (MgSO,),and evaporated to dryness. The residue was dissolved inmethanol (100 ml) and a 0-lM-solution of sodium in methanol(15 ml) was added. After stirring during 21 h in the dark,the solution was poured into methylene chloride (100 ml)and water (100 ml) and the organic phase was washed withwater (2 x 100 ml), dried (AllgSO,), and evaporated to dry-ness. The residue crystallised from ether-light petroleum(b.p. 60-80deg;) as a greenish purple powder (20 mg, 71).(b) Side-chain cyclisation. The foregoing magnesiumcomplex (20 mg) was dissolved in methanol (100 ml) (amp;418, 554, and 602 nm) and aqueous 10 sodium carbonate(1.6 ml) was added (spectrum changed to 410, 544, and 580nm), followed by a solution of iodine (50 mg) in methanol(10 ml). After 30 min at room temperature (there was nochange in the visible absorption spectrum, Am= 550 and620 nm, after 5 min) the mixture was poured into methylenechloride (100 ml) and water (100 ml). The organic layerwas washed successively with 2~-hydrochloric acid (100 ml) ,aqueous 4 sodium thiosulphate (100 ml), aqueous 7sodium hydrogen carbonate (50 ml), and finally water (50ml), dried (MgSO,), and evaporated to dryness. The residuewas chromatographed on alumina (grade 111) (10 g) elutionwith methylene chloride-benzene (1 : 1 and then 7 : 3) andfinally methylene chloride alone. The band which waspale green on the column and red-brown in solution wascollected initially, and this was shown to be an oxophlorinwith an isocyclic ring and a 10-methoxy-group (m/e 652)aZ, (CH,Cl,) 418, 607, 544, 573, and 640 nm. The majorfraction appeared dark green on the column and red-greenin solution. Crystallisation from methylene chloride-methanol gave purple needles (1.4 mg, 7), m.p. 255-258", mixed m.p. with authentic 10-methoxyphaeopor-phyrin-a, dimethyl ester 254-257'; mass spectra of (a)authentic material: vnle 637 (lo), 636 (22), 579 (ll),578 (42), 577 (loo), 504 (3), 503 (7), 502 (6), 318 (2), and288.5 (2); (b) synthetic material: me 637 (17), 636 (26),579 (19), 578 (31), 577 (loo), 504 (6), 603 (13), 502 (lo),318 (4), and 288.5 (5) (Found: M+, 636.295; M+ -59,577.281. Required: M+, 636.295; M+ -59, 577.281).3/1970 Received, 25th September, 1973
展开▼
