Biomarkers for mycotoxins - Potentials and pitfalls


Mycotoxins are toxic metabolites produced by fillamentous fungi and can be found in almost all types of grains. About 95% of all mycotoxin contamination occur before harvest. Despite the widespread use of preventive measures in good agricultural practice, 81% of more than 4,200 feed samples tested positive for mycotoxins in 2013 (BIOMIN Mycotoxin Survey, 2013). As the consequences and health effects of mycotoxins differ greatly between single animals, scientists, veterinarians and farmers have been on a persistent search for diagnostically conclusive biomarkers.

What are mycotoxin biomarkers?

Biomarker of exposure

It is important to differentiate between biomarkers of exposure and effect. A good example of a biomarker of exposure is aflatoxin M1 (AfM1) in the milk of cows (see Table 1). Biomarkers of exposure measure the mycotoxin or its metabolites in the blood, milk, urine, feces or other physiological samples. To some extent, the mycotoxins can be detected unchanged in physiological samples while the rest are metabolized.

Table 1. Potential biomarkers of exposure and effect for the main mycotoxins used in scientific studies.

Source: BIOMIN, adapted from Baldwin et al., 2011

Depending on the milk production yield among other factors, it is estimated that 1-6% of ingested AfB1 can be found in form of AfM1 (hydroxylated metabolite) in the milk of cows. Roughly calculated, 0.05 ppb of AfM1 (EU maximum level for milk) would correlate to a range of AfB1 contamination from 0.8 to 5 ppb in compound feed (5 ppb is the EU maximum level for compound feed in dairy cattle).

This example shows that conducting mycotoxin analyzes on feed is recommended in order to prevent the economic risk of aflatoxin contaminated milk close to the EU maximum level.

Biomarker of effect

Biomarkers of effect, also called mechanism-based biomarkers, should be directly linked to a specific step in the disruption of metabolic and cellular processes.

For instance, the first step leading towards porcine pulmonary edema in pigs is the disruption of the sphingolipid metabolism by fumonisin B1 (FB1). This compound inhibits the ceramide synthase resulting in an elevated sphinganine-to-sphingosine (Sa/So) ratio. The Sa/So ratio is a scientifically recognized biomarker of effect for fumonisins (FUM) in pigs, but not in humans.

Practical challenges

In the case of FUM, the Sa/So ratio applies to scientific trials but not at the farm level. It is difficult to provide controlled feeding and the lack of non-exposed groups on farms makes it impossible to define the cut-off value.

In addition, for a biomarker to have practical relevance, there must be a linear correlation between the exposure and ingestion of the mycotoxin. In some published scientific trials, a linear relationship could be found for DON and its metabolites measured in the blood or urine of swine; however, there are limitations.

Nonetheless, the deviation of individual amounts of mycotoxins detected in physiological samples does not allow any conclusion to be made on the amount of ingested mycotoxins and their health effects in single animals. These are the reasons for the lack of established guidance levels on critical concentrations of DON or other mycotoxins in the blood or other physiological samples of animals, which renders the interpretation of results impossible.

The situation is further complicated by the need for a precise time when sampling for a representative analysis. This is because of the peak in DON and its metabolites in the blood within two hours after ingestion, followed by a rapid depletion afterwards. ZEN takes longer to deplete due to the enterohepatic circulation (absorption in the blood, excretion via bile and reabsorption in the blood). Farm animals are usually fed ad libitum which makes sampling time unpredictable, thereby yielding results that are not representative.

Another important aspect is the fact that DON, like other mycotoxins, is converted into metabolites such as DON-glucuronide, deepoxy-DON and also unknown metabolites. The proportion depends on the species, life cycle, gut microbiota and the health status of the animal.

Furthermore, the toxicity of DON metabolites may differ from the parental compound; for example, deepoxy-DON is non-toxic. ZEN can be found as alpha- and beta-zearalenol, alpha- and beta-zearalanol and their glucuronated forms in physiological specimens. The transformation of ZEN into alpha-zearalenol increases estrogenicity. As a result, analyzing for only one individual mycotoxin is not enough.

Analyzing biomarkers

A trend in recent years has been the development of liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS)-based methods, which are highly selective and sensitive enough to detect mycotoxins at very low concentrations. LC-MS/MS offers the possibility to quantify several metabolites in parallel.

In contrast, enzyme-linked immunosorbent assay (ELISA) methods can only serve as a rough screening method as the matrix effects caused by body fluids alter the results. Antibodies used in ELISA tests to quantify mycotoxins have a wide cross-reactivity to related metabolites. For example, most ELISA kits for ZEN also detect alpha-zearalenol but cannot differentiate between the metabolites. The cross-reactivity for the different metabolites is often neither evaluated nor specified precisely in the user manual.

While there exist validated methods to analyze mycotoxins in feed, there are hardly any for biomarkers. In contrast to feed, quality control for mycotoxin analyzes of physiological samples has yet to be established for commercial laboratories.

Although biomarkers are valuable tools in scientific studies, more knowledge is needed on the factors influencing the bioavailability, kinetics and metabolic profile of mycotoxins in animals before biomarkers could be used in practice on farms. There is still a lack of linear correlation for biomarkers. The use of control groups and elaborate sampling is indispensable, which makes the procedure very costly.

The analysis of mycotoxins in feed is a well-established, reliable approach to assessing possible risks and is therefore the method of choice.

Why not ELISA?

Although quick and inexpensive, ELISA can only be used in validated raw materials and is not a suitable method for analyzing non-validated physiological specimens.

Serum and milk samples were analyzed for DON in two different laboratories. While the first laboratory detected concentrations in the range of 69.5-117.5 µg/L by ELISA, levels were below the detection limit measured by HPLC in the second lab. Evidently the results by ELISA were false positive, as this method is not appropriate for mycotoxin analysis of complex matrices like feed, milk and blood.

Table 2. Comparison of ELISA and HPLC for physiological samples.
1 Laboratory 1
2 S. Dänicke, Institute of Animal Nutrition, Friedrich-Loeffler-Institute,
Federal Research Institute for Animal Health, Braunschweig, Germany