So far, despite the fact that the suspension was thrown out, the strongest evidence claimed against Ryan Braun is that the testosterone he took came from outside his body. For those of us more interested in the question of, “Did he do it?” than, “Did he get caught?” this is the finding that makes us question his innocence. But unlike the T/E ratio, which has been discussed ad infinitum, the actual results of the isotope test that led to this claim have not been discussed to any extent that I could find.
Since I am a scientist and a Brewers fan, and had a bunch of time on my hands, I decided to investigate the details of the technique used. What I found was a relative lack of publicly available literature on the topic, and only a few papers that really got at the heart of the issue.
How does Isotope Ratio Mass Spectrometry (IRMS) analysis work?
At its core, it’s about Newton’s 2nd law, or Force = mass x acceleration. You hit your sample with electrons, which gives your sample a negative charge. Then the sample is passed down a tube at a fixed rate. After a certain length, the sample encounters the positive pole of a magnet. Because of the attraction, the sample will be pulled off of its path and towards the magnet. How far the sample is pulled off of its path will depend on its mass and its charge. The charge on each molecule in the sample is the same, due to the method of ionization. Therefore, the sole determinant of how far the sample deflects from its path is its mass. Collectors are arranged to sense how much material is being deflected at each relative position, and the current produced as a result of detection events can be used to find the ratio of isotopes.
So, how does this measure carbon 12 and carbon 13 in testosterone?
That was my next question. After all, testosterone has more than one carbon in it, and carbon 12 and carbon 13 are not the only two isotopes of carbon. If you were testing the molecule as a whole, you’d have no way of telling the difference between a testosterone that had two carbon 13s and a testosterone that had one carbon 14. The answer, I discovered, is that this isn’t really just a mass spectrometry technique. It would more accurately be called gas chromatography/combustion/isotope ratio mass spectrometry (GC/C/IRMS) . To start from the beginning, the testosterone, epitestosterone, and other sterols are isolated from the urine through chemical extraction. Then those samples are evaporated and separated using gas chromatography, which can separate the sterols into individual components. As each component of the sample emerges from the gas chromatograph, it is passed down a combustion tube. There, it is burned completely until all the carbon has been transformed into carbon dioxide. The carbon dioxide is separated from the other combustion gases using a water trap, then is ionized and passed down the mass spectrometer. So in that last step, you are measuring CO2, which has only one carbon. As a result, you only need to monitor mass at three molecular weights (44, 45, and 46). The ratio of mass 45 to mass 44 gives you your carbon 13 to carbon 12 ratio, and that’s what is compared to the standard.
What’s the standard?
That’s a good question. There are two separate standards, an absolute standard for the test and a relative internal standard. The absolute standard is known as the PDB standard, which was the fossil group first used to study the carbon 13 ratio . These fossils contained a remarkably high amount of carbon 13. The result number, referred to as d13, is the ratio in the sample divided by the ratio in the PDB standard, minus one, multiplied by 1000. This number will be negative, because the PDB standard is so rich in carbon 13. In terms of determining whether the result implies testosterone use, I found methodologies that used two separate standards. One standard was comparing an athlete’s result against a broad population average. The second standard was using a carbon-containing molecule from the sample that wouldn’t be impacted by taking pharmaceutical testosterone. Of those two standards, I think the second one would provide a more airtight argument. However, I don’t know which standard WADA uses, and couldn’t find it published.
So why is this important?
It’s important because groups of plants process the isotopes of carbon differently. There are two types of photosynthesis, C3 and C4, which are relevant here. Plants that use C4 photosynthesis, like corn and sugar cane, tend to have higher carbon 13 content than plants which use C3 photosynthesis, like rice, barley, and soy. The d13 for plants with C4 metabolism tends to be around -14±6, while the d13 for plants with C3 metabolism is -28±6 . When synthetic testosterone is made, it is produced from soy by-products, which give it a d13 around -30. For naturally produced testosterone, the carbon source is the overall diet, which contains plants of both types. As a result, a normal d13 is in the middle. A study of Americans found that the baseline d13 was -23.8±0.93 . A study of five Chinese athletes found a baseline d13 around -27±1 . The primary difference is the relative quantities of various grains in the diet. Because of the precision of the technique, these differences are sufficient for statistically significant detection. That said, dietary variation can significantly alter d13. The baseline level in the Chinese athletes might have been flagged as suspicious based on the population average found using Americans.
Not really, for two reasons. First, the total amounts of carbon 12 and carbon 13 aren’t going to change. Neither one is radioactive or would degrade on the time scale we are dealing with. Second, due to their reduced mass, molecules with lesser mass isotopes tend to react quicker. This is called the chemical kinetic effect. In the case of either thermal degradation or contamination, it would be expected that the testosterone with more carbon 12 in it would degrade first. Any degradation would theoretically enrich the carbon 13, or make the sample look more natural.
So, did he do it or not?
Sadly, the answer is that we don’t have enough information. We know that the WADA is claiming that the d13 is low enough to indicated exogenous testosterone. To really come to a conclusion, we would have to know the d13 in Ryan Braun’s sample as well as the standard being used for comparison. Additionally, if Braun uses a nutritionist, or follows a special diet, it’s possible that this ratio would be influenced to the point of flagging the result.
The best case scenario for Braun supporters is that his d13 is ~-27, he can document a rice-based diet, and a general population standard was applied to evaluate the result.
The worst case scenario would be a d13 ~-31 for his testosterone, and a d13 ~-23 for other organics in the sample.
Thank you for taking the time to read this.
4. Saudan, C., et al., Short-term stability of testosterone and epitestosterone conjugates in urine samples: Quantification by liquid chromatography–linear ion trap mass spectrometry. Journal of Chromatography B, 2006. 844(1): p. 168-174.
5. Ayotte, C., D. Goudreault, and A. Charlebois, Testing for natural and synthetic anabolic agents in human urine. Journal of Chromatography B: Biomedical Sciences and Applications, 1996. 687(1): p. 3-25.