5 February 2009

4 minute read

But scientists are working on new methods to catch these crooks. In a recent AC paper (DOI 10.1021/ac801966g), Chris Elliott and colleagues at Queen’s University Belfast and the Agri-Food and Biosciences Institute (both in the U.K.) discuss their efforts to address this problem by using high-throughput instruments that are found in hospitals throughout the world: clinical blood-chemistry analyzers.

Currently, low-throughput LC/MS/MS and GC/MS/MS technologies are used to detect livestock doping, so testing any more than ~0.05 per cent of the total European herd population is a daunting task. A blood-chemistry analyzer, however, offered the researchers the opportunity to run dozens of immunochemical assays for a variety of analytes typically found in bovine (and human) blood. By using a high-throughput method, they increased the number of samples that could be tested and therefore increased the likelihood of identifying doped animals.

“In recent years, a trend of administration of very low dose cocktails of naturally occurring hormones has made physiochemical analysis even more problematic,” Elliott says. “And even if minute traces of steroids can be detected, proving definitive illegal administration under these circumstances is close to impossible.”

Elliott’s group has focused its efforts not on the offending GPAs themselves but rather on the physiological response triggered by the doping agents. “The ability to detect evidence of such administrations using metabolic markers would be a major scientific advance,” Elliott says. “The cost per analysis is much lower, allowing a greater number of samples to be processed. And those found to be suspect can be subjected to a high degree of scrutiny by use of advanced MS and alternative matrices, such as hair.”

In a proof-of-concept study, the researchers compared blood samples taken from steers injected with estrogen and heifers injected with testosterone with samples from corresponding control animals. Using the commercial blood analyzer, they monitored 20 different biomarkers of growth and muscle development, including creatinine, total protein, cholesterol, urea, and enzymes such as alanine transaminase. They then used support vector machines (SVMs) to generate metabolic profiles of the treated and control animals. SVMs are algorithms that “learn” to separate test data into different categories on the basis of how well they fit one or more training sets of known data points. The algorithms refine the training set by iteratively testing each member data point against the rest of the set.

According to Elliott, the markers that exhibited the greatest degree of change in response to the hormone regimes were those typically associated with increased rates of protein accumulation. But no single analyte served as an absolute marker of anabolic steroid abuse over the full study period.When taken in combination, however, the panel of markers was sufficient to clearly distinguish between treated and control animals. The researchers confirmed this by analyzing a herd that had never been exposed to hormones.

These animals tested “clean” when the new method was used.Elliott notes that the use of standardized analyzers also opens the door to the possibility of on-site testing with portable instrumentation. “As with many types of analytical equipment, smaller and more portable devices are becoming available,” he says. “In the case of clinical analyzers, this is certainly the case, and such instruments have been developed for point-of-care testing.”

The feasibility of developing such devices for the detection of hormone doping in cattle, he says, will depend on the ability to reduce the overall number of analytes that require testing without reducing the number of positive hits.

Currently, the latter problem would be a limiting factor in the development of portable testing devices.Elliott acknowledges that the current study had a limited scope. “The study was based on quite a wide population of cattle in terms of breed, sex, and feeding regimes,” Elliott says. “However, it is clear that other factors such as age and the nutritional and health status of the animals will also be important factors.”

Furthermore, he adds, the administration of a single steroid probably meant that a lower metabolic response was generated compared with what was achieved through cocktail hormone treatments. “It will be important to build a library of data from animals treated with a range of anabolic agents,” he says, to get a more complete panel against which to test suspected cattle.

TheCattleSite News Desk