Enzymes that detoxify marine toxins

Potent microbial toxins found in shellfish are possible starting points for drug discovery, but analogues are needed for biological testing. Toxin-modification enzymes now suggest a new approach for producing these analogues.
Monica E. McCallum is in the Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, USA.

Search for this author in:

Emily P. Balskus is in the Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, USA.

Search for this author in:

There is an acute need for new medications to treat pain. Important sources of therapeutics for pain management and other human conditions are natural products — complex, biologically active small molecules made by living organisms. But compounds isolated directly from natural sources often do not have the optimal properties to be drugs. Therefore, a major challenge faced by those using natural products as leads for drug discovery is how to access a diverse range of closely related molecular structures for biological testing. This can be accomplished using chemical synthesis, but the complex structures of natural products often make that approach challenging.

The difficulties of accessing structural analogues have hampered efforts to investigate a family of natural products called paralytic shellfish toxins (PSTs) as candidate therapeutics for pain1. Many PSTs are highly potent (they elicit a strong response from their molecular biological targets) and are therefore highly toxic, which has hindered their development as drugs and has generated interest in accessing less potent analogues. Writing in ACS Chemical Biology, Lukowski et al.2 report the biosynthetic pathway that generates PSTs to which sulfo groups (SO3) have been added, which are less toxic members of this family of compounds. The sulfotransferase enzymes characterized in the study modify extremely complex substrate molecules, and therefore might facilitate access to other less toxic analogues of PSTs for drug development.

PSTs are produced by marine microorganisms, including cyanobacteria and dinoflagellates3,4. They are responsible for the numbness, tingling and more-severe symptoms of paralytic shellfish poisoning (caused by eating shellfish contaminated with these toxins), and interfere with the voltage-gated sodium channels that are responsible for transmitting signals in the nervous system. Previous efforts to isolate PSTs revealed that microbes often make analogues that bear one or more sulfo groups, leading to the discovery that this chemical modification reduces the potency and toxicity of these natural products5,6.

The biosynthetic pathways and enzymes involved in the installation of these sulfo groups were not understood until a few years ago. The first insights were obtained from assays that used poorly characterized enzyme preparations isolated from dinoflagellates7,8. These studies suggested that the sulfo groups were probably added to PSTs at a late stage of the biosynthetic pathway. More recently, the identification of the cyanobacterial genes encoding the biosynthetic machinery that produces saxitoxin, a highly potent PST, have enabled a molecular understanding of PST assembly9.

Saxitoxin is assembled through transformations that convert the amino acid l-arginine into a series of increasingly elaborate structures. Previous work10 had identified two putative sulfotransferase enzymes (SxtN and SxtSUL) encoded by saxitoxin’s biosynthetic gene clusters, and had found that SxtN can attach a sulfo group to a particular nitrogen atom in saxitoxin to generate an analogue called gonyautoxin 5 (Fig. 1). However, the position on saxitoxin at which the second sulfotransferase (SxtSUL) installs a sulfate (SO4), and the order in which the enzymes are used in nature, were not determined.

Figure 1 | Biosynthesis of sulfated paralytic shellfish toxins (PSTs). PSTs, including saxitoxin, are potentially fatal to humans, but less toxic analogues are potential leads in the search for new painkillers. Lukowski et al.2 have worked out the biosynthetic pathway that converts saxitoxin into less toxic sulfated analogues in microbes. They find that SxtN, a sulfotransferase enzyme, attaches a sulfo group (SO3)specifically to a nitrogen atom in saxitoxin, forming the compound gonyautoxin 5. The GxtA enzyme then selectively adds a hydroxyl (OH) group to the other end of the molecule, forming M1β, and a second sulfotransferase, SxtSUL, converts the hydroxyl group into a sulfate group (SO4), forming the C2 analogue. The work might allow less toxic sulfated PST analogues to be prepared using a combination of conventional chemical synthesis and enzymatic chemistry. PAPS and NADH are enzyme cofactors; VanB is a partner enzyme of GxtA.

Lukowski et al. have now characterized the ability of purified SxtN and SxtSUL to modify saxitoxin and other PSTs. Researchers from the same group had previously shown11 that an oxygenase enzyme called GxtA catalyses the selective addition of a hydroxyl (OH) group to a normally unreactive carbon centre in saxitoxin (Fig. 1). In the current work, the authors combined SxtN and SxtSUL with GxtA, and thereby not only confirmed that SxtN installs a sulfo group on the previously identified nitrogen atom, but also discovered that SxtSUL selectively sulfates the hydroxyl group generated by GxtA.

Unlike the non-enzymatic transformations typically used in the chemical synthesis of natural products, these enzymatic reactions are highly selective for single sites on the PST scaffold, and tolerate the presence of the many densely packed, reactive chemical groups that are embedded in the complex molecular architecture of PSTs. Lukowski et al. were therefore able to produce a variety of sulfated PSTs directly from saxitoxin. When they measured the biological activity of these compounds, the results confirmed that the addition of multiple sulfo groups to PSTs reduces the compounds’ binding affinities to voltage-gated sodium channels. This strongly suggests that sulfo groups reduce PST toxicity, further highlighting their potential for incorporation into PST-based drug candidates.

The use of biosynthetic enzymes to modify PSTs represents a strategy that is distinct from the chemical-synthesis approaches more frequently used to make analogues of these natural products12. Although many of those synthetic efforts have been successful, they often involve long sequences of reactions and deliver low yields of products as a consequence of the challenging architectures of the PSTs — which contain an abundance of reactive oxygen and nitrogen atoms that complicate the use of more-standard chemical reactions. Lukowski and colleagues’ findings now offer researchers the opportunity to combine conventional synthetic chemistry with biocatalysis, using enzymes to further modify PST scaffolds obtained by synthetic routes. This could potentially streamline access to sulfated versions of these natural products. It might eventually even be possible to use this approach to make non-natural PST analogues for evaluation as candidate therapeutics.

However, substantial barriers must be surmounted before these sulfotransferase enzymes can be fully integrated into PST syntheses. Their catalytic efficiency is very low, and they have not yet been used on a large scale — Lukowski and colleagues worked at a sub-milligram scale, but multi-gram quantities of PST analogues would eventually be needed for the preclinical development of drug candidates. Also, the reactivity of the enzymes towards non-natural PST scaffolds, or towards members of related toxin families, has yet to be evaluated. If the reactivity and selectivity of the sulfotransferases can be optimized using enzyme engineering, these biocatalysts will become powerful synthetic tools in the search for new pain therapeutics.

Nature 570, 315-316 (2019)

doi: 10.1038/d41586-019-01742-1


  1. 1.

    Durán-Riveroll, L. M. & Cembella, A. D. Mar. Drugs 15, 303 (2017).

  2. 2.

    Lukowski, A. L. et al. ACS Chem. Biol. (2019).

  3. 3.

    van Apeldoorn, M. E., van Egmond, H. P., Speijers, G. J. A. & Bakker, G. J. I. Mol. Nutr. Food Res. 51, 7–60 (2007).

  4. 4.

    Bustillos-Guzmán, J. J. et al. Food Addit. Contam. A 32, 381–394 (2015).

  5. 5.

    Andrinolo, D., Michea, L. F. & Lagos, N. Toxicon 37, 447–464 (1999).

  6. 6.

    Andrinolo, D., Iglesias, V., Garcia, C. & Lagos, N. Toxicon 40, 699–709 (2002).

  7. 7.

    Sako, Y. et al. J. Phycol. 37, 1044–1051 (2001).

  8. 8.

    Yoshida, T. et al. Fish. Sci. 68, 634–642 (2002).

  9. 9.

    Wang, D.-Z., Zhang, S.-F., Zhang, Y. & Lin, L. J. Proteom. 135, 132–140 (2016)

  10. 10.

    Cullen, A. et al. ACS Chem. Biol. 13, 3107–3114 (2018).

  11. 11.

    Lukowski, A. L. et al. J. Am. Chem. Soc. 140, 11863–11869 (2018).

  12. 12.

    Berlinck, R. G. S., Bertonha, A. F., Takaki, M. & Rodriguez, J. P. G. Nat. Prod. Rep. 34, 1264–1301 (2017).

Download references

Nature Briefing

An essential round-up of science news, opinion and analysis, delivered to your inbox every weekday.