Here, we describe the synthesis of the genetically predicted biosynthetic intermediates of the neurotoxin saxitoxin (STX) (1), 2, 6 and 7, and identification of 2 and 6 in toxin-producing microorganisms. This is the first chemical evidence supporting the genetically predicted biosynthetic route toward 1.

Saxitoxin (STX) and its analogues are potent voltage-gated sodium channel blockers biosynthesized by freshwater cyanobacteria and marine dinoflagellates. We previously identified genetically predicted biosynthetic intermediates of STX at early stages, Int-A' and Int-C'2, in these microorganisms. However, the mechanism to form the tricyclic skeleton of STX was unknown. To solve this problem, we screened for unidentified intermediates by analyzing the results from previous incorporation experiments with N-labeled Int-C'2. The presence of monohydroxy-Int-C'2 and possibly Int-E' was suggested, and 11-hydroxy-Int-C'2 and Int-E' were identified from synthesized standards and LC-MS. Furthermore, we observed that the hydroxy group at C11 of 11-hydroxy-Int-C'2 was slowly replaced by CD O in CD OD. Based on this characteristic reactivity, we propose a possible mechanism to form the tricyclic skeleton of STX via a bicyclic intermediate from 11-hydroxy-Int-C'2.

Saxitoxin, the most potent voltage-gated sodium channel blocker, is one of the paralytic shellfish toxins (PSTs) produced by cyanobacteria and dinoflagellates. Recently, putative biosynthetic genes of PSTs were reported in these microorganisms. We previously synthesized genetically predicted biosynthetic intermediates, Int-A’ and Int-C’2, and also Cyclic-C’ which was not predicted based on gene, and identified them all in the toxin-producing cyanobacterium Anabaena circinalis (TA04) and the dinoflagellate Alexandrium tamarense (Axat-2). This study examined the incorporation of 15N-labeled intermediates into PSTs (C1 and C2) in A. circinalis (TA04). Conversions from Int-A’ to Int-C’2, from Int-C’2 to Cyclic-C’, and from Int-A’ and Int-C’2 to C1 and C2 were indicated using high resolution-LC/MS. However, Cyclic-C’ was not converted to C1 and C2 and was detected primarily in the extracellular medium. These results suggest that Int-A’ and Int-C’2 are genuine precursors of PSTs, but Int-C’2 converts partially to Cyclic-C’ which is a shunt product excreted to outside the cells. This paper provides the first direct demonstration of the biosynthetic route towards saxitoxin and a shunt pathway.

We recently reported the chemical synthesis and identification of the genetically predicted biosynthetic intermediates of saxitoxin (STX), including a 2-aminoimidazole-bearing monoguanidine compound (Int-C'2) in two paralytic shellfish toxin (PST)-producing microorganisms. In this study, we achieved the direct conversion of Int-C'2 into a tricyclic bisguanidine compound (called Cyclic-C'), which is structurally related to STX, through oxidative intramolecular guanidine transfer to 2-aminoimidazole catalyzed by Pd/C under basic conditions in air. By using HPLC-MS analysis, Cyclic-C' was also identified in the PST-producing microorganisms, suggesting that Cyclic-C' is either another biosynthetic intermediate or a shunt product of PSTs. In addition, a weak inhibitory activity of Cyclic-C' to the voltage-gated sodium channels was detected by using a cell-based assay.

No abstract