The New York Times broke the story today that the testosterone found in Tour de France champion Floyd Landis's blood was not from a natural source. Just how do they know that, and how reliable is the test?
The first thing an anti-doping lab looks for in such a case is the ratio of testosterone to the isomeric epitestosterone - too high an imbalance is physiologically unlikely and arouses suspicion. Landis already is in trouble from that reading, but the subject of the Times scoop is the isotopic ratio of the testosterone itself. And that one is going to be hard to get away from, if it's true.
Update: people are asking me why athletes don't just take extra epistestosterone to even things out. That they do - that's the most basic form of masking, and if Landis's ratio was as far off as is being reported, it's one of the odd features about this case. But the isotope test will spot either one, if it's not the kind your body produces itself - read on.
Steroids, by weight, are mostly carbon atoms. Most of the carbon in the world is the C-12 isotope, six protons and six neutrons, but around one per cent of it has an extra neutron to make it C-13. Those are the only stable isotopes of carbon. You can find tiny bits of radioactive C-14, though, and you can also get C-11 if you have access to a particle accelerator. Work fast, though, because it's hot as a pistol.
So, testosterone has 19 carbon atoms, and if on average every one out of a hundred carbon atoms is a C-13, you can calculate the spread of molecular weights you could expect, and their relative abundance. One out of every ten thousand molecules would have two C-13 atoms in there somewhere, one out of every million or so would have three, and so on. A good mass spectrometer will lay this data out for you like a deck of cards.
But here's the kicker: those isotopic forms of the elements behave a bit differently in chemical reactions. The heavier ones do the same things as their lighter cousins, but if they're involved in or near key bond-breaking or bond-making steps, they do them more slowly. It's like having a heavier ball attached to the other end of a spring. This is called a kinetic isotope effect, and chemists have found all sorts of weird and ingenious ways to expoit it. But it's been showing up for a lot longer than we've been around.
The enzymatic reactions that plants and bacteria use when they take up or form carbon dioxide have been slowly and relentlessly messing with the isotope ratios of carbon for hundreds of millions of years. And since decayed plants are food for other plants, and the living plants are food for animals, which are food for other animals and fertilizer for still more plants. . .over all this time, biological systems have become enriched in the lighter, faster-reacting C-12 isotope, while the rest of the nonliving world has become a bit heavier in C-13. You can sample the air next to a bunch of plants and watch as they switch from daytime photosynthesis to nighttime respiration, just based on the carbon isotope ratios. Ridiculously tiny variations in these things can now be observed, which have led to all sorts of unlikely applications, from determining where particular batches of cocaine came from to figuring out the dietary preferences of extinct herbivores.
So, if your body is just naturally cranking out the testosterone, it's going to have a particular isotopic signature. But if you're taking the synthetic stuff, which has been
partly worked on with abiotic forms of carbon derived from a different source (see below), the fingerprints will show. (Update: yes, this means that the difference between commercial testosterone and the body's own supply isn't as large as it would be otherwise, since the commercial synthesis generally starts from plant-derived steroid backbones. But it's still nothing that a good mass spec lab would miss). If the news reports are right, that's what Landis's blood samples have shown. And if they have, there seems only one unfortunate conclusion to be drawn.
Chem-Geek Supplemental Update: for the folks who have been wondering where exactly the isotopic difference comes in, here's the story: synthetic testosterone is made from phytosterol percursors, typically derived from wild yams or soy. Those are both warm-climate C3 plants, which take up atmospheric carbon dioxide by a different route than temperate-zone C4 plants, leading to noticeably different isotope ratios. That's where all the isotope-driven studies of diet start from. The typical Western industrial-country diet is derived from a mixture of C3 and C4 stocks, so the appearance of testosterone with a C3-plant isotopic profile is diagnostic.