German researchers have shown that RNA-sequencing can be a potent tool to identify gene expression signatures that can distinguish livestock animals illegally treated with anabolic steroids.
The team, from three German institutions, published a report in Analytical Chemistry this month evaluating RNA-seq as a tool for finding gene expression changes associated with physiological effects of growth-promoting agents.
Using RNA-seq on just a few samples from a breed of heifer, and then profiling additional samples for the most promising candidate biomarkers using RT-qPCR, the researchers were able to precisely distinguish animals treated with anabolic steroids from controls, they reported.
The team said that they were also able to verify the candidate genes in both boars and calves, demonstrating the potential of the approach to find predictive signatures that could be useful for drug testing across livestock species and, potentially, also in humans.
Abuse of all anabolic substances in animal husbandry is forbidden within the EU, though allowed through specific licenses in some cases in the US and other countries, according to the German team.
Irmgard Riedmaier, a research associate at the Technical University of Munich and the study's first author, told In Sequence in an e-mail that the team's experiments clearly showed that RNA-seq could detect genes whose expression is changed due to anabolic steroid treatment. However, because the analysis of those genes using RT-qPCR is much cheaper and faster, any eventual test based on the biomarkers the group found in their initial study, or in future work, would most likely be done using PCR.
"The idea of the commercial use of such a method is present, but the way toward that is still long and [requires] validation of biomarker candidates in many different samples from animals of different breed, age or sex," she wrote.
In the team's initial report, Riedmaier and her colleagues offered a proof of principle that RNA-seq could be used to identify predictive gene expression markers indicative of biological changes induced by anabolic steroid treatments. Riedmaier said that RNA-seq, compared to other high-throughput methods like microarrays, is a "much more sensitive method with lower background levels and also [offers] other possibilities, like de novo detection of transcripts."
RNA-seq "detects the physiological effects on the level of the transcriptome of steroid treatment, which should always be present regardless of which steroid you are using or in which concentration," Riedmaier said. "Even if the concentration is below the detection limit of the applied detection methods, physiological effects are present and should be measurable."
In the study, she and her colleagues analyzed samples from 18 "Nguni heifers" to establish a gene expression signature that could distinguish between steroid-treated and non-treated animals. The group performed RNA-seq on samples from the livers of three heifers treated with a combination of trenbolone acetate and estradiol, as well as from five untreated animals, picking a set of the 40 best candidates to validate in all 18 animals using RT-qPCR.
Of these 40 genes, the group found 20 to be significantly different in treated animals vs. controls by RT-qPCR. Nine genes were significantly down-regulated and 11 were significantly up-regulated, the group reported. In all cases except one, the direction of regulation, up or down, matched what the group measured in its RNA-seq results.
Further analysis showed that the 20-gene signature could define a "clear separation" between the two treatment groups.
The researchers then tested the signature in two other animal groups — boars and calves. They could successfully quantify fourteen of the genes in the boar samples and were able to distinguish clearly between treated and untreated animals using these.
In the calf experiment, the team analyzed gene expression in three groups of calves: those treated once with a "pour-on" hormone treatment; those treated three times with the same treatment; and a control group. The researchers found that genes from the initial RNA-seq discovery could clearly distinguish the thrice-treated calves from the other two groups, but yielded an "incomplete separation" between the once-treated and control-group animals.
This was likely because the single pour-on treatment did not have a large enough physiological effect to alter gene expression perceptibly, the authors wrote.
Overall, the authors concluded that RNA-seq could identify more potentially predictive genes than a targeted PCR approach they previously used on the same heifer liver samples. "This technology enables quantification of changes in the expression of non expected genes or new splice variants and the influence of treatment with anabolic agents on the pathways that are not known to be influenced yet can also be detected," the group wrote.
According to the researchers, the 20 genes identified through RNA-seq and RT-qPCR validation in additional heifer samples could serve as initial biomarkers for developing a test to identify cattle treated with anabolic agents.
Additionally, by analyzing several different species, the team showed that the signature they identified may be independent of species, breed, sex, and age.
To verify the initial signature as a method of detecting misuse of anabolic steroids in farm animals, the group wrote, it will be necessary to perform more trials with other anabolic substances and different species.
Riedmaier said the team is also planning to develop the approach to work on non-invasive blood samples. "Currently we are analyzing samples obtained in a new animal study, especially blood samples, and hope that we can also find promising mRNA biomarkers," she wrote.
"The results obtained in cattle could be verified in pigs, which are genetically related to humans. So we hope that results obtained in animals could also be a hint for potential biomarkers [that] can be used in humans, too," she added.