Isolation and Chemical Modification of Clerodane Diterpenoids from Salvia Species as Potential Agonists at the k-Opioid Receptor by Yiqiang Li a ), Stephen M. Husbands a ), Mary F. Mahon b ), John R. Traynor c ),and Michael G. Rowan* a ) a )DepartmentofPharmacyandPharmacology,UniversityofBath,ClavertonDown,Bath,BA27AY, UK(phone:01225-386-789;fax:01225-386-114;e-mail:prsmgr@bath.ac.uk) b )DepartmentofChemistry,UniversityofBath,ClavertonDown,Bath,BA27AY,UK c )Department of Pharmacology, University of Michigan, Ann Arbor, Michigan, MI 48109, U.S.A. The clerodane diterpenoid salvinorin A (1), the main active component of the psychotropic herb Salviadivinorum,hasbeenreportedtobeapotentagonistatthe k-opioidreceptor.Computermodeling suggested that splendidin (2) from S. splendens , as well as related compounds, might possess similar activities.Inthepresentstudy,thishypothesiswastestedbydeterminationofthebindingpropertiesofa series of structural congeners, compounds 2 – 8,atthe m-, d-,and k-opioid receptors. However, none of thesecompoundsshowedsignificantbindingtoanyoftheopioid-receptorsubtypes,thusdisprovingthe abovehypothesis. The novel compounds 7 and 8 were obtained semi-synthetically by selective modification of salvifarin(5),isolatedfrom Salvia farinacea,uponepoxide-ringopeningwithAcOHinthepresenceof indium(III)triflate.Also,theX-raycrystalstructureofsalvifaricin(6 ; Fig. ),obtainedfrom S.farinacea, wasdeterminedforthefirsttimeandused,incombinationwithin-depthNMRexperiments,toelucidate the absolute configurations of the new products. Our experiments demonstrate that the relatively well- accessible diterpenoid 6 couldbeusedasstartingmaterialforfuturestudiesintothestructure–activity relationship at the k-opioid receptor. Introduction. – Salviadivinorum,alsoreferredtoas(magicmint)or(holymint),isa psychotropicplantfirstdescribedinthe1960s[1].Anumberofclerodanediterpenoids havebeenisolatedfromthisspecies[2],ofwhichsalvinorinA(1)wasidentifiedasthe major active constituent [3][4]. This compound was subsequently shown to act as an agonist at the k-opioid receptor, the first non-nitrogenous compound to do so [5][6]. Modeling studies with 1 led to the hypothesis that splendidin (2), another clerodane diterpenoid from Salvia splendens , might also possess k-opioid activity [5]. However, the study of structure–activity relationships (SAR) by a number of groups using modified derivatives of 1 have shown particular structural requirements lacking in 2, includinganacetyl(Ac)grouporanothersmallsubstituentattheC(2)atom[7],aswell as trans-fused B/C rings,witha b-orientatedH-atomatC(8)[8]. The purpose of the present study was to test whether or not splendidin (2) is, indeed,a k-opioidagonist,andtocompareitsbiologicalpropertieswiththoseofother diterpenoids from S. splendens , including salviarin ( 3) and splenolide B (4).Wealso isolated the clerodanes salvifarin ( 5) and salvifaricin (6)from S. farinacea,andtested them together with the semi-synthetic derivatives 7 and 8. CHEMISTRY&BIODIVERSITY–Vol.4(2007) 1586 # 2007 Verlag Helvetica Chimica Acta AG, Zürich