Interlocked molecules: One-pot pentaknot.

Knots are commonly encountered in everyday life, but their significance extends far beyond their obvious practical utility of being able to bind objects together. As well as serving as decorative motifs — most notably in some illuminated manuscripts such as the Book of Kells — the symbolism often associated with some knots’ intricate interwoven structures has seen them used to represent countries, religious ideals and concepts such as unity. Knots have also long fascinated mathematicians with an interest in topology. Mathematically speaking, however, a knot is only considered to be a true knot if it cannot be undone without breaking it. Chemists have long sought to tie molecules into knots1. This interest in making knots partly stems from the inherent beauty of such structures and the considerable synthetic challenge they represent. The appeal of knots goes well beyond the aesthetic, however, because functional biomolecules including DNA and some proteins are known to form knotted structures. Both natural and synthetic polymers may form knots and this has a marked effect on their physical properties, with a knot weakening a polymer strand and making it more likely to break2. Most examples of molecular knots made by chemists have been the simplest of the true knots, the trefoil knot. Now, writing in Nature Chemistry, David Leigh and his team report the synthesis of the first non-DNA molecular knot of higher order, namely the pentafoil or cinquefoil knot3. Their rational strategy for making a molecular pentafoil knot combines many aspects of modern supramolecular chemistry, including metaldirected assembly, anion templation and reversible covalent bond formation. Using metal cations to direct how organic ligands come together has had a rich recent history in creating topologically complex chemical architectures1. Such strategies require crossover points, either between different circuits, such as in a chain-like link (a catenane) or in the Borromean rings, or within a single circuit, such as in a knot. Researchers led by Jean-Pierre Sauvage demonstrated in the late 1980s that the three crossover points required to ultimately generate a trefoil knot can be introduced through a helicate4. Helicates consist of multiple ligand strands that wrap around metal cations in a helical fashion. Sauvage recognized that a twostranded helicate with two metal centres will contain the three crossover points required for a trefoil knot. If the ends of the ligands in the helicate are linked together in the appropriate fashion a knot will ensue (Fig. 1a). Although extending the length of the ligand strands and increasing the number of metal cations should, in theory, INTERLOCKED MOLECULES