Biosynthetic pathways harbor diverse enzyme functions, and identifying those that catalyze unusual or synthetically challenging transformations offers new routes for biocatalytic development.
In addition to their biological activity, the identification of new natural products indicates the presence of another potentially valuable activity: that of the enzymes involved in their biosynthetic pathways. In some cases, the reactions performed by these enzymes, such as the rearrangement of epoxides (Nat. Chem. Biol. 13, 325–332, 2017) or the incorporation of halogens (Nat. Chem. Biol. 13, 537–543, 2017), challenge synthetic chemists, suggesting that the enzymes may be useful as biocatalytic tools. Identifying the individual transformations that comprise the biosynthetic pathway of a given natural product can at times be just as challenging as finding and structurally characterizing the compound itself, but those pathways with enigmatic transformations also hold the greatest potential for uncovering new biological chemistry. In this issue, three papers exemplify the ongoing quest to solve biosynthetic mysteries and, in the process, reveal the unique and potentially useful capabilities of natural catalysts.
Synthetic organic chemists have many tricks for forming carbon–carbon bonds, but it was unclear how organisms do so for the cylindrocyclophanes, a family of cyanobacterial natural products with an all-carbon [7.7]paracyclophane scaffold. Now, it has been revealed that the reaction is achieved in two discontinuous steps by a pair of enzymes (Article, p. 916). In the first of these steps, a halogenase (CylC) selectively chlorinates one carbon atom of decanoic acid, after which the resulting alkyl chloride is further extended and processed by a polyketide synthase (PKS). The alkylating enzyme CylK then utilizes the chlorinated centers to create the eponymous macrocycle via dimerization involving the formation of new carbon–carbon bonds. While it was previously understood that a PKS was involved in cylindrocyclophane biosynthesis, it was not clear from the structure of the natural product that a chlorinated intermediate was involved or that the alkyl chloride was installed prior to extension by the PKS. Elucidation of this pathway adds a new enzymatic approach to the available collection of carbon–carbon bond–forming reactions.
Although the assortment of known enzymes that form carbon–carbon bonds is diverse and growing, enzymes known to catalyze the formation of nitrogen–nitrogen bonds are much rarer. Piperazate is a nonproteinogenic amino acid containing such a bond, but, despite its presence in a variety of nonribosomal peptide natural products, the enzyme responsible for formation of the nitrogen–nitrogen bond remained elusive. It turns out that KtzT was hiding in plain sight, originally annotated as a putative transcriptional regulator with a flavin-binding domain when in fact it is a heme-binding protein that catalyzes the formation of nitrogen–nitrogen bonds (Brief Communication, p. 836). With the function and substrate selectivity of KtzT now properly understood, the identification of homologs could guide the search for novel piperazate-containing metabolites, as well as add an entirely new reaction to the biocatalytic toolbox.
Biocatalytically useful enzymes must not only catalyze desirable chemical reactions, but also do so in a well-defined manner on accessible substrates. The N-methyltransferase involved in biosynthesis of the omphalotins, a group of ribosomally synthesized and post-translationally modified peptide (RiPP) natural products, is notable not for the chemistry that it catalyzes—many N-methyltransferases are known throughout biology—but for its unusual substrate and enzymatic architecture (News and Views, p. 821; Brief Communication, p. 833). The N-methyltransferase OphA installs methyl groups on select backbone amide nitrogen atoms of a precursor peptide, in a reaction that is without precedent in RiPP biosynthesis. In an additional twist distinguishing OphA from other RiPP tailoring enzymes, the OphA methyltransferase domain is fused to the precursor peptide, meaning that the enzyme catalyzes N-methylation on its own C-terminal region. With such autocatalytic activity, it is possible to replace the C-terminal precursor peptide with other sequences of interest to be backbone N-methylated, which the authors validate with the sequences of cyclosporin A and dictyonamide A. This engineering of OphA to methylate desired peptide sequences demonstrates the potential of natural product biosynthetic enzymes to be useful biotechnological tools for more than just their native chemistry.
CylK, KtzT, and OphA are just a few examples of the diverse enzymes lurking in the biosynthetic pathways of natural products, and undoubtedly countless others are yet to be revealed. The discovery of new biosynthetic chemistry is even more assured as entirely new natural product scaffolds are uncovered, as these scaffolds will likely be created by unprecedented enzymatic activity. However, the identification of new natural product classes is much more difficult than finding new products within a known class—the curse of “you get what you know to look for.” One potential approach to finding new scaffolds is integrated screening, such as the combination of metabolomics scoring with heterologous expression, also described in this issue (Article, p. 895). Such high-throughput approaches, which also help to circumvent the hurdles associated with silent gene clusters, are becoming ever more necessary to keep pace with the rapidly growing availability of genome sequences. Other promising approaches involve using mass spectra to perform rapid dereplication of known compounds in silico to focus on those that are potentially novel (Nat. Chem. Biol. 13, 30–37, 2017), or correlating structures with sequence data to predict the transformations comprising their biosynthetic pathways (Nat. Chem. Biol. 12, 1007–1014, 2016). With the latter approach, perhaps the transformations that are more difficult to predict bioinformatically are those that are more likely to result from novel chemistry, and thus of particular interest for detailed follow-up. In the future, bioinformatics approaches such as these could have a dual focus: to find both new natural products and the novel chemical transformations that comprise their biosynthesis.
As chemical biologists expand understanding of the functions of these enzymes and contribute to the identification of biosynthetic gene clusters, they advance a robust cycle that continues with new natural products leading to the discovery of novel enzymatic transformations and potential biocatalytic approaches. These efforts, combined with a keen interest in applying new chemical insights, demonstrate the capabilities of chemical biologists to reveal and expand on the diverse hidden chemistry of cells.
About this article
Cite this article
Hints of hidden chemistry. Nat Chem Biol 13, 815 (2017). https://doi.org/10.1038/nchembio.2449