Understanding and manipulating alkaloid biosynthesis

Abstract

Humans have exploited plant alkaloids as medicines since at least the Neolithic Era. Today, alkaloids such as vinblastine (isolated from Catharanthus roseus) and morphine (isolated from Papaver somniferum) are prescribed to treat various cancers and relieve pain, respectively. Despite this storied use and palpable presence in the current pharmacopeia, relatively little is known about the biosynthesis, regulation and transport of these molecules. For example, monoterpene indole alkaloid (MIA) biosynthesis, a set of metabolic pathways that produces hundreds of bioactive natural products, has not been fully elucidated in any organism. Here we examine the biosynthesis of secologanin, which contributes the monoterpene moiety to all MIAs. Specifically, we excavate C. roseus transcriptomic datasets to identify 1 0-hydroxygeraniol oxidoreductase, a missing step in secologanin biosynthesis. 10-hydroxygeraniol oxidoreductase catalyzes the oxidation of both hydroxyl moieties of 10-hydroxygeraniol to form 10-oxogeranial, which is the substrate for iridoid synthase, the reductive cyclase that assembles the characteristic iridoid scaffold. Despite having an incomplete understanding of MIA biosynthesis, several engineering strategies have been successfully deployed to incorporate halogenation into the MIA machinery and yield halogenated alkaloids. Although alkaloids and plant natural products have been used to treat various diseases, these compounds have not evolved specifically to do so. Therefore, these compounds frequently require editing to effectively tune their biological and pharmacological activities. We also describe efforts to reengineer tryptophan halogenase RebH to preferentially install chlorine onto tryptamine, the direct indole precursor for the MIAs. After reengineering RebH, we then overexpressed the tryptamine-specific mutant RebH Y455W and flavin reductase RebF in C. roseus and observed the de novo biosynthesis of a chlorinated unnatural natural product 12-chloro- 19,20-dihydroakuammicine. Lastly, we describe the serendipitous discovery of a P. somniferum codeine-Odemethylase mutant that selectively demethylates codeine, a benzylisoquinoline alkaloid involved in morphine biosynthesis, instead of both codeine and thebaine. This mutant may selectively disable a redundant route in the biosynthesis of morphine that has been associated with poor seed and licit opium quality.