Total Synthesis of Solanoeclepin A
Admittedly I don't check Nature Chemistry as often as I should, so I only noticed this truly epic synthesis of solanoeclepin A a few days ago. I remember being shocked by my first sight of the structure during a talk by Prof. Henk Hiemstra a couple of years back, especially that improbable looking DEF ring system. This synthesis is obviously a phenomenal technical achievement, and it must have been an incredibly demanding task, but at first glance there aren't too many sexy steps. The abstract mentions 'addressing one of the critical food issues of the twenty-first century' and solving natural supply problems, goals towards which this synthesis could be the first step.
I'd imagine that when attempting a synthesis of this complexity you'd want to stick as closely as you could to well established chemistry that can be performed on large scale, so I was surprised to see this nice little cross coupling around step 30:
I had no idea you could do this (although the methodology is older than I am), and even if I did, I probably wouldn't plan it in the middle of a 50+ step total synthesis! The method was originally described with aryl bromides and a range of silyl enol ethers, and although even the yields in the seminal paper are a bit so-so (50-65%), any alternative sequence I can think of is quite a bit longer. The α,β-unsaturated ketone is then immediately used as the dienophile for the intramolecular Diels-Alder reaction which forges the A,B and C rings. Wow. Just wow.
 A great organic chemist once told me that every synthesis should have at least one 'sexy step'.
 By my calculations every 10g of that not-so-readily available starting material gives 41 mg of the natural product. A little goes a long way, though, as it's active at down to 10^-10 g/mL in water.