Development of new "green" methodology for the synthesis of substituted phenylacetic acid derivatives as precursors to isoflavonoids and related compounds

Loading...
Thumbnail Image
Date
2013-01
Authors
Pieterse, Tanya
Journal Title
Journal ISSN
Volume Title
Publisher
University of the Free State
Abstract
English: Flavonoids and isoflavonoids are known to exhibit many important physiological properties and are especially promising candidates for cancer chemoprevention. Similar to most natural products, studies directed at the synthesis of flavonoids have, therefore, emerged from the search for new compounds with beneficial biological properties. Metabolic studies related to flavonoids are, however, frequently hampered by the inaccessibility of a variety of optically active compounds. While a single method for the synthesis of enantiomerically enriched isoflavonoids has been published, this process utilizes phenylacetic acid derivatives which are not always readily available in all naturally occurring substitution patterns. Even though the synthesis of phenylacetic acids are possible via a number of routes, these are based on ancient low yielding chemical processes utilizing harsh reaction conditions, stoichiometric quantities of reagents and in many cases, poisonous heavy metals like lead and thallium. In order to address the availability of phenylacetic acid derivatives of variable substitution patterns, the current study was aimed at the development of methodology for the synthesis of phenylacetic acid derivatives that would be high yielding, environmentally benign, have a limited number of process steps, and are applicable to all naturally occurring flavonoid substitution patterns. In this regard it was envisaged that ozonolysis of substituted allylbenzenes would comply with all of the stated criteria and was therefore investigated as methodology for the synthesis of phenylacetic acid derivatives that could serve as building blocks during isoflavonoid preparations. Since substituted allylbenzenes of all oxygenation patterns are not available commercially, the allylic moiety was introduced into the required phenols by means of a allyl phenyl ether intermediate, through utilization of Williamson ether synthesis (allyl bromide; K2CO3, refluxing CH3CN) followed by Claisen rearrangement of the neat allyl phenyl ethers, allyl 3- methoxyphenyl ether and allyl 3,5-dimethoxyphenyl ether, under microwave irradiation (at 200 ºC in 15 min. intervals and 0-200 W variable power) to obtain the desired allylphenols, 1-allyl-2-hydroxy-4-methoxybenzene and 1-allyl-2-hydroxy-4,6-dimethoxy-benzene, in 44 and 88 % yield, respectively. Apart from the desired allylphenol, Claisen rearrangement of allyl 3-methoxyphenyl ether, however, also led to the formation of 1-allyl-2-hydroxy-6- methoxybenzene in 45% yield, indicating a lack of selectivity towards the formation of the sterically less hindered product under the prevailing reaction conditions. Since free phenolic substituents on the aromatic rings of the envisaged substrates might have a negative effect during ozonolysis reactions, the commercially available allylphenols as well as the two substrates prepared by allylation and Claisen rearrangement (vide supra) were subjected to methylation (MeI; K2CO3; refluxing acetone or acetonitrile) and the respective fully methylated analogues, 1-allyl-3,4-dimethoxybenzene, 1-allyl-2,4-dimethoxybenzene, 1-allyl- 2,4,6-trimethoxy-benzene, and 1-allyl-3,4,5-trimethoxybenzene, obtained in 79, 96, 80, and 77% yield, respectively. Ozonolysis [O3 (6-8 min.), DCM, 0 °C] with reductive work-up [N-methylmorpholine-Noxide (NMMO)] of 1-allyl-2-methoxybenzene, 1-allyl-4-methoxybenzene, and 1-allyl-3,4- dimethoxybenzene afforded the corresponding phenylacetaldehydes in 58, 88 and 15% yield, respectively. Ozonolysis of the highly oxygenated substrates, 1-allyl-2-hydroxy-4- methoxybenzene, 1-allyl-2,4,6-trimethoxybenzene, and 1-allyl-3,4,5-trimethoxybenzene, however, only led to cleavage of the aromatic ring and formation of unidentifiable product mixtures. Cleavage of the aromatic rings of these substrates was confirmed by 1H NMR analysis of the reaction mixture [O3 (6-8 min.), CDCl3, -78 °C] where the formation of the 1,2,4-trioxolane intermediate could be detected for the substrates that gave the desired phenylacetaldehydes but not for the highly oxygenated analogues. In order to reduce electron density on the aromatic ring of the highly oxygenated substrates and prevent ring ozonolysis in this way, the free hydroxy function on each substrate was changed into a trifluoromethanesulfonyl ester and the subtrates, 1-allyl-2- trifluoromethanesulfonyloxy-4-methoxybenzene, 1-allyl-4-trifluoromethanesulfonyloxy-3- methoxybenzene, 1-allyl-4-trifluoromethanesulfonyloxy-3,5-dimethoxybenzene, and 1-allyl- 2-trifluoromethanesulfonyloxy-4,6-dimethoxybenzene submitted to ozonation with NMMO work-up again. While the phloroglucinol based sulfonyl ester gave the desired phenylacetaldehyde in 71% yield, the other three substrates furnished NMMO induced double bond migration with the subsequent formation of the benzaldehyde equivalent products. When the ozonation reaction was repeated on the resorcinol, catechol and pyrogallol trifluoromethanesulphonyl esters, with replacement of the NMMO with dimethyl sulphide (DMS) as reductant, the desired phenylacetaldehydes or trioxolanes were, however, obtained in 63, 32 and 31% yield, respectively. Ozonolysis with oxidative work-up [(i) O3/MeOH; (ii) Ac2O-Et3N] applied to the monomethoxy substrates, 1-allyl-2-methoxybenzene and 1-allyl-4-methoxybenzene, afforded the desired methyl phenylacetates in 91 and 32% yield, respectively. Similar to what was found for the reductive work-up procedure, the higher oxygenated substrates had to be converted to their respective triflates before ozonolysis of the allylic double bond could be effected successfully and the phenylacetic acid esters, methyl 4-trifluoromethanesulfonyloxy- 3-methoxyphenyl acetate, methyl 4-trifluoromethanesulfonyloxy-3,5- dimethoxyphenyl acetate, and methyl 2-trifluoromethanesulfonyloxy-4,6-dimethoxyphenyl acetate obtained in 9, 17 and 65% yield, respectively. Since the published process for the stereoselective synthesis of isoflavonoids would require the phenyl acetates prepared through ozonolysis to be transformed into the corresponding amides, the possibility of direct formation of the nitrogen derivatives, like anilides, during the ozonolysis reaction was subsequently investigated. While first attempts at having the aniline present during the ozonolysis reaction only led to nitrogen oxidation, the process was amended to addition of the nitrogen nucleophile after formation of the 1,2,4-trioxolane, which resulted in the desired 2-methoxyphenylacetanilide being formed in 37% yield. The scope of this novel reaction was subsequently extended to the reaction of 1-allyl-4- methoxybenzene with acetamide leading to the product being obtained in 39% yield. While this reaction gave indications that deactivated nitrogen nucleophiles could also be used in this process, the reaction with 2-imidazolidinone, a secondary amide, did not succeed indicating that the new reaction still needs to be optimized to be useful in the enantioselective synthesis of isoflavonoids. Finally, it was shown during the current study that the phenylacetic acid derivatives prepared via ozonolysis could be transformed into deoxybenzoins, another isoflavonoid precursor, through formation of the acid chloride followed by reaction with a phenyl Grignard reagent. Thus phenylacetyl chloride could be reacted successfully with phenylmagnesiumbromide at - 78 oC in diethyl ether to give the deoxybenzoin in almost 60% yield.
Afrikaans: Flavonoïede en isoflavonoïede beskik oor ‘n wye verskeidenheid belangrike biologiese eienskappe waaronder belowende eienskappe t.o.v. kanker voorkoming. Soortgelyk aan die studie van die meeste ander natuurprodukte, het flavonoïedchemie sy oorsprong te danke aan die soeke na nuwe verbindings met voordelige biologiese eienskappe. Metabolise studies met flavonoïede word egter dikwels belemmer deur die ontoegangklikheid van substrate met verskillende substitusiepatrone in opties aktiewe vorm en slegs ‘n enkele proses vir die sintese van enantiomeries verrykte isoflavonoïede het tot dusver die lig gesien. Aangesien hierdie proses op fenielasynsuurderivate berus, is dit dus noodsaaklik dat hiedie tipe verbindings met alle moontlike natuurlike substitusie patrone beskikbaar moet wees. Hoewel sommige fenielasynsure en derivate kommersieël beskikbaar is, is dit nie die geval vir verbindings van alle substitusiepatrone nie en moet verskeie van hierdie verbindings, wanneer benodig, berei word. Alhoewel metodes vir die sintese van fenielasynsure wel bestaan, is baie hiervan op antieke chemiese prosesse waarin drastiese reaksie kondisies, stoigiometriese hoeveelhede reagense en giftige swaarmetale gebruik word, gebaseer en is baie metodes nie geskik vir die bereiding van substrate met hoë vlakke van oksigenering nie. Ten einde die beskikbaarheid van fenielasynsure van alle moontlike substitusie patrone aan te spreek en ‘n omgewingsvriendelike proses met beperkte aantal prosesstappe vir die sintese van hierdie groep verbindings daar te stel, is ‘n ondersoek na die osonolise van gesubstitueerde allielbensene as metodiek vir die sintese van fenielasynsuurderivate aangepak. Aangesien allielbensene met alle moontlike substitusie patrone nie kommersieël beskikbaar is nie, is besluit om die allielgroep aan die fenielring te heg d.m.v. Williamson etersintese (allielbromied, K 2 CO 3 , kokende CH 3 CN) gevolg deur Claisen-herrangskikking van die gevormde allielfenieleters. Die allielfenieleters, alliel-3-metoksifenieleter en alliel-3,5- dimetoksifenieleter, is dus onder oplosmiddelvrye toestande aan mikrogolf bestraling blootgestel (200 o C, 15 min. intervalle, 0 ௅ 200 W veranderlike krag) om die allielbensene, 1- alliel-2-hidroksi-4-metoksibenseen en 1-alliel-2-hidroksi-4,6-dimetkosibenseen in 44 en 88% opbrengs onderskeidelik, op te lewer. Benewens die verlangde allielfenol, het die Claisen- herrangskikking van alliel-3-metoksifenieleter ook tot die vorming van die ongunstige isomeer, 1-alliel-2-hidroksi-6-metkosibenseen, in 45% opbrengs gelei, sodat afgelei kon word dat geen regioselektiwiteit onder die toestande waarby die reaksie uitgevoer is, moontlik is v nie. Weens die feit dat vry-fenoliese hidroksigroepe tydens die osonolise proses tot ongewenste newereaksies kon lei, is die hidroksifunksies van die bereide allielbensene asook dié van die kommersieël beskikbare analoë deur metilering (MeI, K 2 CO 3 , kokende asetoon of asetonitriel) beskerm en is die volledig gemetileerde verbindings, 1-alliel-3,4- dimetoksibenseen, 1-alliel-2,4-dimetoksibenseen, 1-alliel-2,4,6-trimetoksibenseen en 1-alliel- 3,4,5-trimetoksibenseen in onderskeidelik 79, 96, 80 en 77% opbrengs, daargestel. Osonolise [O 3 (6–8 min), dichlorometaan, 0 o C] met daaropvolgende reduktiewe opwerk [ N - metiel-morfolien- N -oksied (NMMO)] van 1-alliel-2-metoksibenseen, 1-alliel-4- metoksibenseen en 1-alliel-3,4-dimetoksibenseen het dan ook die ooreenstemmende feniel- asetaldehiede, in onderskeidelik 58, 88 en 15% opbrengs, gelewer. Reaksie van die hoogs geoksigeneerde substrate, 1-alliel-2-hidroksi-4-dimetoksibenseen, 1-alliel-2,4,6-trimetoksi- benseen en 1-alliel-3,4,5-trimetoksibenseen, het egter as gevolg van osonolise van die aromatiese ring se dubbelbindings slegs tot ‘n mengsel van onidentifiseerbare produkte gelei. Hierdie waarneming is m.b.v. 1 H KMR bevestig waar die vorming van die 1,2,4-trioksolaan tussenproduk waargeneem kon word tydens osonolise [O 3 (6–8 min), CHCl 3 , -78 o C] van die substrate wat die verlange produkte gelewer het, maar nie tydens osonolise van die hoogs geoksigeneerde uitgangstowwe nie. Ten einde die elektronrykheid van die aromatiese ringe, van die resorsinol, katesjol, floroglusinol en pirogallol substrate, wat moontlik die oorsaak van die mislukking van die osonolise reaksies kon wees, te verminder, is een van die OH-funksies van elk van hierdie uitgangstowwe as trifluorometaansulfonielester beskerm en die reaksies op 1-alliel-2- trifluorometaansulfonieloksi-4-metoksibenseen, 1-alliel-4-trifluorometaansulfonieloksi-3- metoksibenseen, 1-alliel-4-trifluorometaansulfonielloksi-3,5-dimetokibenseen en 1-alliel-2- trifluorometaansulfonieloksi-4,6-dimetoksibenseen met NMMO opwerk, herhaal. Die floroglusinol gebaseerde triflaatester het dan ook die verlangde fenielasetaldehied in 71 % opbrengs gelewer, maar die ander substrate het slegs die bensaldehiedanaloë as produkte, gelewer. Hierdie verskynsel kan waarskynlik aan basis gekataliseerde dubbelbindingsmigrasie, weens die teenwoordigheid van die NMMO, toegeskryf word. Herhaling van die osonolise reaksies met DMS as reduktant het dan ook hierdie feit bevestig, aangesien die gesogte fenielasetaldehiede of trioksolane in 63, 32 en 31% opbrengs uit die reaksies verkry kon word. vi Osonolise gevolg deur oksidatiewe opwerk [(i) O 3 /MeOH; (ii) Ac 2 O-Et 3 N] van die monometoksi-allielbensene, 1-alliel-2-metoksibenseen en 1-alliel-4-metoksibenseen, het die ooreenstemmende metielfenielasetate in 91 en 32 % opbrengs onderskeidelik, gelewer. Soortgelyk aan wat vir die reduktiewe opwerk proses gevind is, moes die hoër geoksigeneerde subtrate ook as die monotriflaatesters beskerm word ten einde suksesvolle osonolise moontlik te maak en kon die verlangde produkte, metiel-4-trifluoro- metaansulfonieloksi-3-metoksifenielasetaat, metiel-4-trifluorometaansulfonieloksi-3,5-di- metoksifenielasetaat en metiel-2-trifluorometaansulfonieloksi-4,6-dimetoksifenielasetaat, op hierdie wyse in 9, 17 en 65% opbrengs onderskeidelik, berei word. Aangesien die gepubliseerde proses vir die enantioselektiewe sintese van isoflavonoïede vereis dat die fenielasetate, wat d.m.v. osonolise verkry word, na die amiede omgeskakel moet word, is besluit om ‘n ondersoek na die moontlikheid om die allielbenseensubstrate tydens osonolise direk na amiede om te skakel, in te stel. Aanvanklike pogings waarin anilien tydens die osonolisereaksie by die substraat en oplosmiddel gevoeg is, het egter slegs tot oksidasie van die anilien gelei, maar nadat die proses gewysig is om na vorming van die 1,2,4-trioksolaan die oortollige osoon te verwyder voordat die anilien bygevoeg is, het dit daartoe gelei dat die verlangde produk, 2-metoksifenielasetanilied in 37% opbrengs verkry kon word. ‘n Uitbreiding van hierdie proses na die reaksie van 1-alliel-4-metoksibenseen met asetamied waartydens die produk in 39% opbrengs verkry is, het voorts getoon dat gedeaktiveerde stikstof nukleofiele ook tydens die nuwe unieke reaksie gebruik kan word. Die benutting van 2-imidasolidinone, ‘n sekondêre stikstof nukleofiel, het egter nie tot die vorming van enige bruikbare produk gelei nie, waardeur aangetoon is dat heelwat navorsingswerk nog benodig word ten einde die volle omvang van hierdie nuwe reaksie te bepaal en dit vir die enantioselektiewe sintese isoflavonoïede bruikbaar te maak. Laastens is dit ook tydens die huidige ondersoek aangetoon dat die fenielasynsuur derivate wat d.m.v. osonolise gevorm word, ook na deoksibenzoïene, ‘n ander belangrike isoflavonoïed-voorloper, omgeskakel kan word. In hierdie verband is ongesubstitueerde deoksibenzoïen in ongeveer 60% opbrengs berei deur die reaksie van fenielasetielchloried met fenielmagnesiumbromied by -78 o C.
Description
Keywords
Dissertation (M.Sc. (Chemistry))--University of the Free State, 2013, Flavonoids, Phenols, Deoxybenzoin, Phenylacetamide, Phenylacetanilide, Methyl phenylacetate, Phenylacetaldehyde, 1,2,4-trioxolane, Ozonolysis, Allylbenzene, Claisen rearrangement, Isoflavonoids
Citation