Summary: | Neolaulimalide (1) and laulimalide (2) are structurally related cytotoxic marine polyketides belonging to the microtubule-stabilising family of anti-cancer agents. Our approach toward the laulimalides is based on the highly convergent synthesis of a common advanced intermediate 3 which, it was imagined, could be used to obtain both natural products. It was envisaged that the synthesis of 3 would feature a similar strategy to that of Paterson’s second generation route to laulimalide. The key fragment coupling would be mediated by an asymmetric reagent-controlled boron aldol reaction between methyl ketone 4 and aldehyde 5, followed by a macrolactonisation under Mitsunobu conditions to construct the macrocycle. The central dihydropyran moiety of the C<sub>1</sub>-C<sub>14</sub> methyl ketone 4 was constructed using an asymmetric Jacobsen’s Hetero-Diels-Alder (HDA) reaction between aldehyde 6 and Danishefsky’s diene (7). The C<sub>13</sub>-C<sub>28</sub> aldehyde was assembled from an aldol addition/elimination reaction between pyran aldehyde 8, itself constructed using a ring-closing metathesis approach from (<i>R</i>)-glycidol, and methyl ketone 10 (derived from (<i>S</i>)-malic acid). Difficulties were encountered when performing the key Mitsunobu macrolactonisation reaction upon advanced <i>seco</i>-acid substrates, in which the sterically-hindered allylic alcohol underwent exclusive S<i><sub>N</sub></i>2’-type displacement to deliver the corresponding rearrangement products. The prepared laulimalide analogues 12, 13 and 14 were tested for biological activity, displaying reduced microtubule-stabilising activity and cytotoxicity when compared with the natural product data. a revised strategy towards the laulimalides was therefore envisaged that hinged on a Yamaguchi macrolactonisation protocol in which the unwanted S<i><sub>N</sub></i>2’-type side-reaction would be avoided.
|