(+)- Citreoviral 2, a highly substituted tetrahydrofuran natural product, is a precursor to a potent anti- HIV agent (-)- Citreoviridin 1. Herein, a formal synthesis of 2 has been developed starting from chiral benzoate 59 obtained from desymmetrization of 2-methylglycerol. The skeleton and chiral centers of 2 were installed by Wittig olefination and modified Sharpless asymmetric dihydroxylation. The iodocyclization of the triol 70 rendered the tetrahydrofuran core 73 stereoselectively. The configuration of the resulting tetrahydrofuran was confirmed by nOe experiment. Deiodination of 73 and further lactonization followed by Wittig olefination gave the conjugated ester 78 which is known as an advanced intermediate to (+)-citreoviral. The key features in our synthetic approach are the desymmetrization of the 2-methylglycerol and the efficiently diastereoselective formation of the tetrahydrofuran moiety.
Arenaric acid 80, a metabolite of an actinomycete strain, was isolated from the Streptomycessp since 1999. Its family, for example K41-A 82 and oxolonomycin 81, possesses antibacterial and potent antimalarial activity. The polycyclic back-bond of these compounds which confers the lipophilic character and a terminal carboxyl group plays an important role in the formation of an oxygen-rich internal cavity which is capable of binding to metal ions.
The Arenaric Acid 80 was disconnected at C11-C12 bond to give two fragments 84 and 85, each of them contains one tertiary chiral center which was installed by our desymmetrization method at the starting point.
In the synthesis of the eastern part 85, several strategies have been studied. The first attempts were to construct the cis-THF ring prior to the trans-THF ring. The trials via iodocyclization failed to give the desired cis- stereoisomer with different substrates (87, 145, 150, 153, 154) and various reaction conditions. In the next attempts, the formation of cis-THF was efficiently accomplished by using oxidative cyclization of trihydroxy alkene 105, 113 and 128. However, the formation of the trans-THF from these tetrahydrofuran derivatives via iodocyclization (104) or epoxide opening strategies (112, 121, 124, 133, 143) failed to give the satisfactory results.
By changing the formation order, first the trans- THF ring and then the cis-THF ring, we succeeded in constructing the di-THF core of the fragment 85. Iodocyclization of the γ-hydroxy alkene 169 rendered the trans-THF derivative 165 in excellent stereoselectivity. Intramolecular substitution finally generated the cis-THF moiety. However, functionalization of the side chain of the resulting cis,trans-di-THF 164 was unsuccessful due to steric hindrance. This problem was solved by introduction of the side chain before the formation of the di-THF core.
The hydroxyl alkene 211 was assembled from phosphonium salt 195 and alcohol 196 which were prepared from known starting materials 93 and 197. Iodocyclization of 211 rendered the desired trans-THF ring in 12:1 diastereomer ratio. The subsequent intramolecular substitution generated the cis-THF ring exclusively without elimination. Finally, the western part 192 was successfully prepared and converted to the silyl enol ether 218.
The synthesis of the western part was started with the same process as the eastern part to provide the aldol adduct 229 which is an enantiomer of 206. The dimethyl ether on A ring was installed by lactonization and methylation. The model aldol coupling of silyl enol ether 218 with aldehyde 246, which was derived from lactone 232, gave a mixture of two diastereomers (248, 247) in 90% total yield, in which the desired isomer 247 was obtained as minor product. These two compounds were separately converted to the corresponding cyclic carbonate to confirm the stereostructure. Interestingly, the undesired 248 was successfully converted to the desired isomer 247 which promised the total synthesis of Arenaric Acid.
The methylated lactone 221 was opened and converted to the aldol adduct 219 which was subjected to oxidation-cyclization to afford the hemiketal 236 and 237. In addition, two other analogues 242 and 244 were also prepared to investigate the coupling reaction with the western part. Unfortunately, the Mukaiyama aldol coupling of 240 and 244 with 218 gave an inseparable mixture of isomers which is proved to be not useful. Meanwhile, 242 reacted with 218 to give a separable mixture of two isomers 264 and 265. The methylation and cyclization of the desired isomer 264 failed to give the spiroketal moiety. Changing protecting group from TBS ether to TES ether and using less acidic environment provided the spiroketal 277 in high conversion. However, attempts to cleave the oxazolidinone moiety were not successful. In addition, conversion of the undesired isomers 265 and 274 to the desired one failed due to elimination of oxazolidinone moiety under basic conditions.
Therefore, the desired lactone spiroketal 257 was converted to the spiroketal 287 via methylation followed by reduction or reduction and subsequent methylation. The later route proved to be more efficient. From 287, a fully assembled skeleton of Arenaric acid 298 was efficiently contructed. However, the final hydrolysis of oxazolidinone did not proceed to give the PMB protected Arenaric acid. This failure suggested that for the total synthesis of Arenaric acid, the chiral auxiliary should be removed in before the formation of the hemiketal (A) moiety.
항 HIV에이전트로잠재성을갖고있는시트레오비리딘의시트로비랄의합성은 2-메틸글리세롤의입체선택적인비대칭화반응으로부터얻을수있는키랄벤조에이트59로부터합성이시작되었다. Wittig 올레피네이션과수정된샤플리스입체선택적인비대칭화다이하이드록실레이션으로부터 2번의키랄센터를확보하였다. 트리올70의요오드화고리반응은 THF 코어73을입체선택적으로제공하였다. 73의아이오딘제거반응과락톤화반응에이은 Wittig 올레피네이션으로컨쥬게이트에스터78을만들었다. 78은알려진시트로비랄의중요중간체이다. 본합성의중요한전략은 2메틸글리세롤의입체선택적인비대칭화반응과 THF의효율적인입체선택적형성이다.
방성균계통의대사에작용하는아리나르산80은 1999년Streptomycessp로부터얻어졌다. 같은군으로 K41-A 82와옥소로노마이신81이있는데그것들은안티박테리아와잠재적으로항말라리아효능을가진다. 아리나르산80은두개의파트84와85로나뉘어지는데각각은본연구실에서개발한입체선택적인비대칭화반응으로부터도입할수있는삼급키랄센터를가지고있다.
오른쪽85번파트의합성을위해몇가지방법이연구되었다. 트랜스 THF 고리를형성하기이전에, 다양한기질과반응조건이용하여요오드화고리반응으로시스 THF 고리를만들고자하였으나실패하였다. 두번째로트리하이드록시알켄105, 113, 128의산성고리화반응을이용하여시스 THF 고리를효율적으로형성하였다. 그러나요오드화고리화반응 (104) 또는에폭시열림반응의전략 (112, 121,124, 133, 143) 을통한모노 THF로부터의트렌스 THF 형성은만족스러운결과를가져오지못하고실패하였다.
순서를바꾸어트렌스 THF 고리를먼저형성하기로하였으며85번의두개의 THF 코어형성을성공하였다. 감마하이드록시알켄의요오드화고리반응으로트렌스 THF의유도체165를얻었으며그것은시스 THF 형태를만들기위해분자내치환반응을시쳤다. 그러나시스트렌스두개의 THF 164의사이드채인의기능화는입체장애로인해성공하지못했다. 이문제는사이드체인의도입을먼저한후 THF 고리를형성하는것으로해결하였다.
포스포늄염195와알코올196의결합으로211알켄을얻었고그것은93과197의알려진출발화합물이다. 211의요오드화고리화반응은원하는트렌스 THF 고리를 12대 1의입체선택도로만들어졌다. 그후에분자내치환반응은시스 THF 고리를제거반응이일어나지않는상태로생성했다. 마침내왼쪽파트192가성공적으로준비되었고실릴엔올이서218로전환되었다.
206의이성질체인알돌애덕트209을공급함으로오른쪽파트와같은방법으로왼쪽부분이합성되었다. 디메틸이서가있는 A고리는락톤화반응과메틸화반응으로얻어졌다. 실릴엔올이서의모델알돌커플링은알데히드246과진행되었는데246은락톤232로부터얻을수있고그결과248, 247의혼합물이 90%의수율로얻어졌다. 247이부생성물이였다. 흥미롭게도원하는248은아리나르산전합성을위한중요한스파이로키탈257로성공적으로전환되었다.
218오른쪽파트의알돌커플링을위해몇가지알데히드 (240, 242, 244)가락톤221로부터준비되었다. 불행하게도무카야마알돌반응은분리할수없는이성질체를내놓았다. 그러는동안242은218과반응하여분리할수있는두가지이성질체264와265를만들었다. 264의메틸화반응과고리화반응으로스파이로키탈277을만들고자하였으나실패하였다. 알코올보호기를 TBS에서 TES로바꾸고약산성조건에서스파이로키탈277을높은전환율로합성을하였다. 하지만옥사졸리디논의분해반응은성공적이지못하였다. 원하는물질을얻기위해원치않는이성질체인265와274의전환을염기조건하에서시도하였으나제거반응이일어났다.
그래서원하는스파이로키달257은메틸레이션, 환원반응, 메틸레이션을통해스파이로키탈287로전환되었으며좀더효율적이였다. 287로부터아리나르산의골격이완벽하게완성되었고두끝을관능기로전환하려하였따. 하지만옥사졸린의가수분해가파라메톡시벤조일로보호되있는아리나르산으로진행되지는않았다. 이실패는헤미키탈 (A)형태에서키랄보조체를먼저제거해야지아리나르산의전합성을완성시켜줄것이라고예측되었다.