The battle over whether tramadol is a natural product or not has been a heated one. Over the last couple of years, it was reported to be produced by a west African shrub, then this was reported to be an artifact of feeding the drug to cattle, and then this hypothesis was challenged by work on an actual biosynthetic pathway. The latest cannonball to be fired might end the debate, although I’m not sure I’d bet on that just yet. It’s from the Dortmund group that reported the tramadol contamination problem, and they’ve returned to the area of North Cameroon (with a colleage from the University of Mourou there) to analyze the material using accelerator mass spectrometry (AMS).
The idea is to measure the amount of carbon-14 in the tramadol itself. Natural products get their carbon, fundamentally, from carbon dioxide in the environment, fixed by photosynthetic organisms into material for the food chain. As such, there should always be some radioactive carbon in there, since it’s being continually produced in the upper atmosphere from nitrogen atoms interacting with cosmic rays. This is the whole idea behind carbon dating – a continual source of 14C. Synthetic organic compounds, though, tend to be made from precursors that are ultimately derived from petroleum/natural gas feedstocks. These have been sitting underground for an extended period, not exchanging very much carbon with the surrounding environment, and are thus depleted in carbon-14. You need some pretty hard-core analysis to detect this sort of thing (thus AMS), but the differences are large once you can see them. (As an aside, the various people who are convinced that a “natural” compound is just somehow different than the identical molecule prepared by synthetic routes may not realize that yes, in some cases the all-natural one is slightly more radioactive).
The authors grew the plant in question (Sarcocephalus latifolius) from seed and confirmed that (1) it had no tramadol in its leaves or roots as grown, (2) that when fed with labeled phenylalanine (the starting material for the previously proposed biosynthetic pathway), that this was taken up and converted to various intermediates, but not to any detectable tramadol, and (3) that when plants were grown in tramadol-laced soil, that they did indeed take up the compound into various tissues.
There are many field observations in the paper as well. Tramadol, as purchased from a street vendor in Cameroon, has (as expected) very low 14C content, consistent with it being a compound derived (ultimately) from petroleum sources. Various locations around northern Cameroon were found to have tramadol in the soil, rivers, and drinking water (!), and all of these samples also showed the same low levels of 14C. Sampling the plants in these regions gave highly variable results. Many of them had no detectable tramadol at all, but the amounts found in them (and in soil samples) seemed to correlate well with collection during the rainy season, which also suggests anthropogenic contamination. In short, the whole area is laced with synthetic tramadol in various concentrations, some of them rather alarmingly high – the drug is clearly a persistent contaminant in the environment, and this really does seem to be the source of it in plant samples.
So unless someone can demonstrate tramadol with higher radiocarbon content, or (especially) show that it is produced in plants grown under controlled conditions, this case would seem to be closed. We’ll see if it really is, though. . .