Synthesis and Bioactivity of Luffarin I

The first synthesis of Luffarin I, sesterterpenolide isolated from sponge Luffariella geometrica, has been accomplished from commercially available sclareol. The key strategy involved in this synthesis is the diastereoselective reduction of an intermediate ketone. Luffarin I against human solid tumor cell lines showed antiproliferative activities (GI50) in the range 12–17 μM.


Introduction
During the last few years, there has been intensive research for new natural pharmacologically active compounds. In general, the chemistry of marine organisms and of sponges in particular has led to the OPEN ACCESS discovery of a great number of novel and interesting metabolites [1]. Many marine-living organisms have developed toxic secondary metabolites to defend themselves against predators [2].
There is a group of pentaprenyl terpenoids whose structures are derivable from geranylfarnesyl diphosphate, known as sestertepenoids, of frequent occurrence in marine sponges. The diverse bioactivity of sesterterpenoids has made them attractive targets for both biomedical and synthetic purposes [3].
The luffarins (A-N), 1-14, Figure 1, were isolated by Butler and Capon from an Australian marine sponge, Luffariella geometrica [11]. Luffarins are sesterterpenolides that have in common with the labdane skeleton the decaline moiety, showing the same stereochemistry for A and B rings. In particular, luffarins bear an eleven carbon atom side chain attached to C-9 with either a butenolide, a hydroxybutenolide or butanolide group, Figure 2. All of them have the same fragment R ( Figure 1) and the skeleton can be defined as luffarane.  Luffarin I, 9, can be proposed as a key intermediate for the synthesis of some luffarins. Herein we report the first synthesis and biological evaluation of luffarin I. Another sesterterpenolide with luffarane skeleton, luffalactone, has been synthesized previously [12].
The 3-bromofurane lithium derivative, achieved by metallation of 3-bromofurane with n-BuLi, was added to aldehyde 19 giving a 1:1 mixture of epimers at C-16 (20a/20b). The oxidation of the mixture 20a/20b with tetrapropylammonium perruthenate (TPAP) in presence of 4-methylmorpholine N-oxide (NMO) led only to decomposition products. Thus, before oxidation of the C-16 hydroxyl group, the double bond was deactivated by conjugation with a carboxylic acid (Scheme 3). Acetoxylation of the secondary alcohol led to the acetoxy derivatives 21a/21b, which by deprotection of the THP group, gave the hydroxyderivatives 22a/22b. The oxidation of the later compounds to the required acids was achieved in two steps; oxidation of the alcohols to aldehydes 23a/23b was carried out with DMP and finally oxidation to the desired acids mixture 24a/24b by oxidation of the aldehydes with sodium chlorite (Scheme 3). The resulting α,β-unsaturated acids 24a/24b were transformed into ketone 26 by hydrolysis of the acetoxy group and oxidation of the resulting alcohol (25a/25b) with DMP [25][26][27] at low temperature and subsequent esterification of the carboxylic group with TMSCHN2 (Scheme 4). The Corey-Bakshi-Shibata [28][29][30][31][32][33] reduction of ketone 26 allows obtaining diastereoselectively only one of the C-16 hydroxyl derivatives (Scheme 4). The generated stereogenic center at C-16 has R configuration as expected, confirmed the application of Mosher methodology [34][35][36][37][38][39], see Supplementary Information. Reduction of either 27 or 28 with diisobutylaluminium hydride (DIBAL-H) afforded diol 29, in good yield in both cases. Conversion of the furane ring of 29, into the γ-hydroxybutenolide was carried out following Faulkner's methodology [40]. Photochemical oxidation of 29 with 1 O2 in the presence of Rose Bengal irradiating with a 200W lamp for 10 min gave quantitatively the hydroxybutenolide 30. Reduction of 30 with NaBH4 [41] transformed the γ-hydroxybutenolide ring into the required γ-butenolide present in luffarin I, 9 (Scheme 5).
The spectroscopic data of 9, as well as its optical rotation [α] D 20 = +69.0 (c 0.51, CHCl3) comply with those corresponding to the natural product described by Butler and Capon as luffarin I [α] D 20 = +64.3 (c 1.4, CHCl3) [11]. It can be concluded that luffarin I has been obtained from methylketone 15, in 15 steps.

Biological Studies
From the set of synthesized analogues, a total of four compounds were submitted to biological assays. The in vitro activity was assessed in A549, HBL-100, HeLa, SW1573, T-47D and WiDr human solid tumor cells. The results expressed as GI50 were obtained using the SRB assay [42], and the results are given in Table 1. The standard anticancer drugs cisplatin and etoposide were used as positive controls. Overall, the data on antiproliferative activity show that all tested compounds exhibited growth inhibition in at least four of the cell lines of the panel. The natural compound 9 is the most active of the series with GI50 values in the range 12-17 µM. Table 1. The antiproliferative activity is comparable to the reference drugs in the most resistant cell lines T-47D and WiDr. Although the set of compounds in this study is small, the presence of the butenolide fragment can explain the enhanced activity of 9 when compared to analogues 29 and 30.

Experimental Section
16-Hydroxy-14,15-dinor-labd-8-en-13-one (16): To a stirred solution of 15 (60 mg, 0.23 mmol) in MeOH (1.1 mL) was added slowly a solution of KOH (76 mg, 1.15 mmol) in MeOH (1.75 mL) and the mixture was reacted at 0 °C for 10 min. Afterwards, (diacetoxyiodo)benzene (DIB) (146 mg, 0.46 mmol) was added, and the mixture was stirred at 0 °C following the reaction evolution by TLC. When the reaction had finished, a 5% aqueous solution of H2SO4 (1.5 mL) was added and the mixture was reacted at 0 °C for 90 min. It was quenched with water and the product was extracted with DCM. The combined organic layers were washed with brine, dried (Na2SO4), filtered, and concentrated in vacuo. The resulting crude residue was purified by flash CC (hexane-AcOEt, 98:2) to obtain 16 (33 mg, 50%). 16-(R,S)-Acetoxy-19,20-epoxy-luffara-8,13Z,17 (20),18-tetraen-21-oic acid (24a/24b): To a solution of 23a/23b (8 mg, 0.019 mmol) in t-BuOH (0.25 mL) and 2-methyl-2-butene (51 μL), a solution of monobasic sodium phosphate (NaH2PO4, 10 mg) in water (0.1 mL) and 5% aqueous solution of NaClO2 (48 μL) were added. The reaction mixture was stirred at rt for 30 min. Then, water and 2 M aqueous solution of HCl were added. It was extracted with AcOEt and the combined organic layers were washed with water until neutral pH was reached, dried (Na2SO4), filtered, and concentrated in vacuo to afford 24a/24b (8 mg, 99%). and borane dimethylsulfide (1.0 M in toluene; 0.19 mL, 0.19 mmol) was added. The reaction mixture was stirred at −30 °C for 20 h. It was quenched with MeOH (2 mL) and it was allowed to warm to room temperature. Then it was added water and Et2O and extracted with Et2O. The combined organic layers were washed with water and brine, dried (Na2SO4), filtered, and concentrated in vacuo. The resulting crude residue was purified with a column with Amberlyst 15 (NH4 + ) and after that, with a flash CC (hexane-AcOEt, 85:15) to obtain 27 (19 mg, 52%) and 28 (  was added dropwise. The mixture was reacted at rt for 90 min and then, AcOEt was added. It was quenched with a saturated aqueous solution of potassium sodium tartrate (1 mL), and it was stirred for 15 min. After that it was extracted with AcOEt and the combined organic layers were washed with 6% aqueous solution of NaHCO3, water and brine, dried (Na2SO4), filtered, and concentrated in vacuo. The resulting crude residue was purified by CC (hexane-AcOEt 9:1) to obtain 29 (11 mg, 85%).
To a solution of 28 (3 mg, 0.008 mmol) in DCM (0.22 mL) under argon atmosphere, DIBAL-H (1.0 M in hexane; 48 μL, 0.05 mmol) was added dropwise. The mixture was reacted at rt for 2.5 h and then, AcOEt was added. It was quenched with a saturated aqueous solution of potassium sodium tartrate (1 mL), and it was stirred for 15 min. After that it was extracted with AcOEt and the combined organic layers were washed with 6% aqueous solution of NaHCO3, water and brine, dried (Na2SO4), filtered, and concentrated in vacuo. The resulting crude residue was purified by CC (hexane-AcOEt 9:1) to obtain 29 (3 mg, 95%).

Conclusions
The first synthesis of Luffarin I has been achieved from (−)-sclareol confirming its structure and absolute configuration as 5S, 10S, 16R. This methodology opens the way for the synthesis of other marine natural compounds of this class. The study of the antiproliferative activity of Luffarin I showed remarkable biological activity towards human cancer cell lines. A more detailed structure-activity relationship study may be necessary in order to establish the scope and limitations of the new scaffold. Experiments needed to validate the usefulness of this compound as potential anticancer drug are in progress and will be reported elsewhere.