Thursday, January 27, 2011

Natural products

Do a Google search and you’ll find all sorts of stuff claimed to be “natural products” — amazingly, some of them even are “chemical-free”! Now seriously. To quote IUPAC’s 1999 recommendations [1],
The nomenclature of natural products has suffered from much confusion.
That does not surprise me. What is surprising, however, that neither these nor the previous recommendations [2] tell us what the “natural products” are. Ditto the Gold Book. It may define terpenoids as “natural products and related compounds formally derived from isoprene units” but the natural products go without explanation. (Nor is it clear what “related compounds” are.) The very same Gold Book says that natural graphite is “a mineral found in nature”. Therefore, “natural” means “found in nature”. Right? Right. Is natural graphite a natural product? I am not sure.

Let us look in Webster then:
A chemical substance produced by a living organism; — a term used commonly in reference to chemical substances found in nature that have distinctive pharmacological effects. Such a substance is considered a natural product even if it can be prepared by total synthesis.
Is that any better? Both dioxygen and nitric oxide are produced by living organisms and have rather distinctive pharmacological effects, yet most chemists would hesitate to call them natural products. In The Concise Oxford Dictionary, the first definition of “natural” is
existing in or caused by nature; not artificial.
Of course human beings are parts of nature, but maybe this negative definition, “not artificial”, is indeed most useful?

In ChEBI, natural product (no definition so far) is an organic molecular entity (so no O2 or NO here) and includes the following classes:
On the first glance, nothing looks particularly disturbing here. But I see a bit of a problem with ontology. First, all is a children of natural product have to be natural products. What if we have, say, (artificially) fluorinated carbohydrates? Are they still carbohydrates? If no, then the True path rule is broken. If yes, then some “unnatural” compounds will be considered natural products. I don’t mind that — perhaps that will cover “related compounds” (whatever they are) nicely.

Second, CHEBI:33243 belongs to chemical entity ontology, which is (or at least it should be) purely structure-based. The origin does not enter here. There is no such thing as intrinsic “naturalness” in a natural product molecule: natural product remains a natural product even if (artificially) prepared by total synthesis.

Natural products often are equated with secondary metabolites. This does not seem right. In ChEBI, secondary metabolite (“A metabolite that is not directly involved in the normal growth, development or reproduction of an organism” — another negative definition?) belongs to role ontology. (The role ontology sounded such a good idea at the time... no, don’t get me started.) At best, one can say some natural products have role “secondary metabolite”. Yuck.

To summarise: “natural product” appears to be a rather useless top-level term. Let us look at the sources of natural products: plants, fungi, bacteria, animals. What if, instead of saying “fungal natural product”, we say “fungi-specific compound”? In this case, we discard primary metabolites, other simple compounds found just about everywhere in the universe and are left with exactly what we want: molecules isolated from and specific for fungi.

Or are we? Antibiotics, naturally synthesised by fungi, are not naturally found in humans. But when we take them, they are naturally metabolised in our liver and eventually excreted with urine. Are these metabolites the natural products? If yes, are they fungal or animal or neither?
  1. Revised Section F: Natural products and related compounds (IUPAC Recommendations 1999). Pure Appl. Chem. 71, 587—643 (1999).
  2. Nomenclature of Organic Chemistry. Section F: Natural Products and Related Compounds. Recommendations 1976. Eur. J. Biochem. 86, 1—8 (1978).