Nature exploits the unique chemistry of molybdenum in many reactions. Structures of the enzyme Cnx1 reveal unexpected mechanisms for slotting the metal ion into its reactive position in the cofactor Moco.
From bacteria to mammals, molybdenum (Mo) is crucial for survival. Enzymes that take advantage of the distinctive redox chemistry of this metal are involved in numerous metabolic reactions in the carbon, nitrogen and sulphur cycles1,2. To be available for these reactions, however, water-soluble molybdate anions from the environment must be bound, transported into cells and manipulated to place them in the correct context to exploit their chemical properties. This complex manoeuvring of Mo involves an ancient and highly conserved biosynthetic pathway, and a key enzyme of the pathway in plants is called Cnx1. On page 803 of this issue, Kuper et al.3 report two crystal structures of the Mo-binding domain of Cnx1. Unexpectedly, the structures suggest a new step in the pathway, involving an adenosine metabolite, and they implicate copper ions in the process as well — two fortuitous and intriguing observations.
The vast majority of enzymes that make use of molybdenum (termed molybdoenzymes) do so through the Mo cofactor — Moco for short. This cofactor contains a single Mo coordinated to an organic molecule called molybdopterin. The biosynthesis of Moco is highly conserved from bacteria through to mammals and involves four stages (Fig. 1): first, conversion of a guanine nucleotide into a derivative termed precursor Z; second, conversion of precursor Z into molybdopterin; third, binding of Mo by molybdopterin, producing Moco; and fourth, attachment of a nucleotide moiety to Moco, forming molybdopterin guanine dinucleotide. In eukaryotes, Moco is considered the active form of the cofactor, but most bacterial enzymes require the molybdopterin guanine dinucleotide to make them functionally active.
How Moco is then inserted into molybdoenzymes is not yet understood because of the intrinsic instability of the chemical species concerned, but the intermediates and cofactor probably remain bound to other proteins throughout the biosynthetic process until the final incorporation of Moco to fully constitute the molybdoenzymes. The crystal structures of several proteins involved in Moco biosynthesis have been solved, and they mostly form oligomers4. This, together with biochemical evidence showing that many of these proteins are involved in the formation of a heterogeneous complex, indicates that specific protein–protein interactions are crucial in the early stages of Moco biosynthesis5.
The new crystal structures3 are of Cnx1, which is homologous to gephyrin (in mammals) and MogA and MoeA (in bacteria). Once the molybdopterin moiety is formed at the end of the second stage, the metal ion has to be prepared and inserted into it — and this is where Cnx1 comes into play. Cnx1 contains two domains, termed G and E, that catalyse the transfer and insertion of Mo into the molybdopterin6. Kuper et al.3 characterized domain G of Cnx1 structurally and biochemically under anaerobic conditions (because oxygen would cause some degradation). They report surprising results.
First, analysis of the structure of Cnx1 with a mutation in its active site revealed an adenosine moiety (AMP) covalently linked to molybdopterin. This chemical species has not been seen before, and the authors go on to show that it is Cnx1G itself that attaches the AMP to molybdopterin. This finding introduces a new step to Moco biosynthesis and is consistent with previous work5. Kuper et al.3 postulate that the product of Cnx1G, the molybdopterin–AMP, is cleaved by domain E of Cnx1, which also transfers the Mo to the molybdopterin sulphur groups to produce Moco. Understanding domain E and its interplay with domain G will now be crucial in completing the picture of this biosynthetic pathway.
Second — and perhaps most important — is the observation of copper binding to the two sulphur groups of the molybdopterin (Fig. 1). These sulphur atoms are potentially reactive and some mechanism is required to protect them until they actually coordinate Mo. The copper ion was not added to the protein during the experiment, but rather seems to have been picked up from the bacterial expression system used to produce the protein, which suggests that the levels of copper present in culture are sufficient for the purpose. The implication is that in vivo Cnx1 binds copper, probably for a combination of reasons: a metal ion complexed to the reactive sulphur groups would not only protect them, but would also provide a mechanism to assist in the insertion of Mo into the cofactor as the metals are exchanged. The suggestion of a direct link between Cu and Mo biology is highly intriguing and will spur further investigations to delineate the relationship between these essential metal ions.