In addition to its use in the treatment of tuberculosis, rifamycin is an example of a microbial natural product whose commercial use predated by many years the elaboration of the intermediates, enzymes, and encoding genes involved in its biosynthesis. A key feature in the biosynthesis of rifamycin is its 3-amino-5-hydroxybenzoic acid (AHBA) core (Scheme 1). This biosynthetic feature is shared with other medicinally important natural products such as the mitomycins and ansamitocins, which are anticancer agents.

Scheme 1

An important step in the delineation of the biosynthesis of AHBA was the formulation of the aminoshikimate pathway (Scheme 2). In 1992, Floss and coworkers synthesized 4-amino-3, 4-dideoxy-D-arabino-heptulosonic acid 7-phosphate (aminoDAHP) and demonstrated its conversion to AHBA.1 The aminoshikimate pathway (Scheme 2) is, in essence, a variation of the shikimate pathway, which leads to the biosynthesis of aromatic amino acids. Condensation of 1-deoxy-1-imino-D-erythrose 4-phosphate (iminoE4P) with phosphoenolpyruvic acid (PEP) catalyzed by aminoDAHP synthase was the suggested first step in the aminoshikimate pathway (Scheme 2). However, aminoDAHP synthase activity could not be detected in cell-free lysate of rifamycin-synthesizing Amycolatopsis mediterranei.

Scheme 2

(a) aminoDAHP synthase (rifH); (b) aminoDHQ synthase (rifG); (c) aminoDHQ dehydratase (rifJ); (d) AHBA synthase (rifK).

Scheme 3a

a enzymes (encoding genes): (a) UDP-d-glucose dehydrogenase (rifL); (b) UDP-3-keto-D-glucose transaminase/AHBA synthase (rifK); (c) UDP-kanosamine phosphatase (rifM); (e) glucokinase; (f) transketolase (orf15); aminoDAHP synthase (rifH).

Our proposal (Scheme 3) for the genesis of iminoE4P involved transketolase-catalyzed fragmentation of 3-amino-3-deoxy-d-fructose 6-phosphate (aminoF6P). To test this hypothesis, aminoF6P was synthesized by a multistep chemoenzymatic synthesis.2 Addition of synthetic aminoF6P to A. mediterranei cell-free lysate led to the formation of aminoDAHP as well as AHBA.2 After establishing the critical precursor required for aminoDAHP synthase activitity, we worked our way backward in the biosynthesis of iminoE4P to 3-amino-3-deoxy-D-glucose, which is a natural product better known as kanosamine (Scheme 3).3

Establishing kanosamine as an intermediate enabled us to piece together the entire biosynthetic pathway extending from UDP-glucose to aminoDAHP (Scheme 3). Verification of each step has followed from chemical synthesis of putative intermediates (Scheme 3) such as UDP-3-keto-D-glucose, UDP-kanosamine, kanosamine, and kanosamine 6-phosphate (aminoG6P) followed by identification of the products formed when each of these chemically synthesized biosynthetic intermediates was added to A. mediterranei cell-free lysate. The rifamycin biosynthetic gene cluster has also been cloned and sequenced by Floss and Hutchinson.4 As a consequence of this confluence of molecular biological, biochemical and chemical approaches, each of the intermediates, enzymes, and encoding genes responsible for iminoE4P and AHBA biosynthesis has now been identified (Scheme 3). These results illuminate the functional intertwining of kanaosamine biosynthesis with the aminoshikimate pathway. In its role of providing the iminoE4P required for AHBA biosynthesis, kanosamine biosynthesis also emerges as a more widespread biosynthetic pathway than previously thought.