Currently, there is a growing interest in mycotoxin biomarkers in the farming community based on the increasing number of farmers enquiring about this topic. The term ‘mycotoxin biomarker’ in that context is being associated with the so-called exposure-related indicators (mycotoxins and their metabolites in body fluids, faeces or organs). This type of biomarkers is seen by researchers as a tool to measure dietary exposure of human and animals to mycotoxins. In contrast, industry looks at it from another perspective; as a cheaper, faster and a more reliable test of the effectiveness of an anti-mycotoxin additive compared to ‘true’ endpoints such as performance parameter.
This article will therefore summarise information about biomarkers for the most important mycotoxins in different animal species. It will also highlight critical points of field evaluation of mycotoxin sequestering agents’ efficacy by measuring direct biomarkers in different physiological substrates.
Types of biomarkers
Many authors define two types of biomarkers: direct and indirect. Direct or exposure-based biomarkers are specific, while indirect or biomarkers of effect are generally nonspecific and represent structural or functional alterations produced in the body under exposure to certain drugs or toxins. However, these alterations in some cases may serve as biomarkers of exposure when both processes are directly linked (Groopman and Kensler, 1999; Perera and Weinstein, 2000; Silins and Högberg, 2011). However, the suggestions would be to use three subtypes for mycotoxin biomarkers (figure 1): exposure-based (direct), mechanism-based (indirect) and effect-based (indirect) biomarkers.
Exposure-based biomarkers characterise the presence of mycotoxins or their metabolites (e.g. glutathione or glucuronide conjugates). These compounds can be detected in easily-accessed biological fluids or tissues (Baldwin et al., 2011). In contrast, mechanism-based biomarkers assess changes specific to the mycotoxin in specific proteins, cellular metabolites, or gene expression. Effect-based biomarkers express the typical consequence of the mycotoxin on the performance and other health parameters such as gut integrity, antibody titre, typical lesions, level of serum liver enzymes or even weight gain and feed conversion. However, they are less specific to the mycotoxin when compared to mechanism-based biomarkers and more so when compared to exposure-based biomarkers.
Figure 1. Suggested classification of mycotoxin biomarkers
Physiological substrates of biomarkers
Specific biomarkers (exposure- and mechanism-based) are measured in body fluids or tissues, as the molecule itself, a metabolite(s), or a product of a reaction with a biological molecule. The most common parameters in quantifying exposure to mycotoxins are measured in urine, serum and milk. However, there are other biological matrices such as faeces and hair that can provide important information.
Blood. Many mycotoxins (aflatoxins, ochratoxin A, deoxynivalenol (DON), zearalenone) are rapidly absorbed in the upper gut resulting in a sharp peak in blood within 2 hours of oral ingestion (figure 2) in most animal species. In contrast, fumonisins have a limited availability and their levels in blood are insignificant. Consequently, exposure-based biomarkers of mycotoxins with high bioavailability can only be detected in serum and plasma in high concentrations shortly after oral ingestion. They are also rapidly cleared from the bloodstream. This is crucial as the time of blood collection is key during in vivo studies of blood biomarkers. The mechanism-based biomarkers such as protein adducts have longer half-lives in blood. Among all biomarkers, they provide the most information on their cumulative effects despite the fact that they may be less specific to a mycotoxin.
Figure 2. Plasma concentration-time profile of DON after oral administration of DON to 6 piglets (Nutriad research based on method of Devreese et al., 2012)
Bile: The portal vein conveys blood with mycotoxins absorbed in the stomach, pancreas, and intestines to the liver. Biliary excretion is a major route of elimination of mycotoxins regardless of theirhigh or low absorption rates. In fact, in some cases, in which mycotoxins are fed at low concentrations, the mycotoxin can only be detected in bile (Armorini et al., 2015). As sampling bile requires animal euthanasia, bile biomarker substrates are unfortunately, only practical for field studies in poultry.
Urine: The main fraction of the rapidly absorbed mycotoxins, passes through the liver and briefly circulates through the blood before it is excreted as its metabolites or unchanged in the urine. However, correlation between ingested mycotoxin and the level of mycotoxin in urine is lower when compared to blood due to food-related variations of urine amount. Several authors recommend using creatinine as an indicator of the amount of urine produced (Kraft and Dürr, 2005) as the excretion of creatinine is not food-related. The ratio between toxin concentration and creatinine content of urine is recommended for further consideration. This physiological substrate is not a suitable matrix for field studies in poultry.
Milk: Metabolites of aflatoxins, DON, zearalenone and ochratoxin A, can be detected in various concentrations in the milk of non-ruminant mammals. In ruminants, milk is an easy and widely used matrix, only for estimation of aflatoxin exposure. There is currently no correlation between the levels of toxins in milk compared to serum, suggesting that the transfer from blood to milk is not yet a fully understood process (Biasucci et al., 2010).
Faeces: The gastric absorption rate of some mycotoxins (e.g. fumonisins) is very low and a large part is eliminated in faeces. Mycotoxins with a high bioavailability are excreted in faeces at a very low rate (Turner et al., 2010). Quantification of biomarkers in faeces as an estimation of the mycotoxin binder has potential; however, this approach has many limits including the possible metabolisation of bound mycotoxin by microflora in the large intestine. Another worrying observation is that faecal concentrations of mycotoxins can decrease with an increasing dose of adsorbent in the diet. It is also suggested that clay-bound toxins may be undetectable in faeces when using analytical procedures standard for other physiological substrates or for feed (Sulzberger et al., 2017).
Organs: Analysis of mycotoxin biomarkers in organs requires an invasive procedure such as a biopsy or a post-mortem study. The accumulation of several mycotoxins in tissues happens primarily in the liver and kidneys. Using organs as a biological matrix has several benefits such as the accumulation effect of the mycotoxin, and the lower variation linked to the feeding time and sample collection. Some mycotoxins even when fed at low concentrations are retained in the liver and kidneys in an unmetabolised (reviewed in Voss et al., 2007) or a metabolised form. Mycotoxins persist in kidneys much longer than in plasma or the liver, and the levels can be 10 times the amount in the liver (Martinez-Larranaga et al., 1999; Riley and Voss, 2006).
Hair: This is the easiest biological sample to collect in biomarker assay. Sewram et al. (2003) showed for the first time that fumonisins can accumulate in the hair of humans and primates. Unfortunately, there is no reliable information about the accumulation of fumonisin or other mycotoxins in hairs of domestic animals.
Species specific differences
In order to select a consistent physiological substrate in the biomarker field assay, the absorption, distribution, metabolism and excretion processes of each mycotoxin must be understood. This enables the most appropriate biomarkers for each mycotoxin to be identified for a particular group of animals.
Mycotoxins have diverse metabolic pathways in vivo that can be very specific to animal species. Mycotoxin uptake and subsequent tissue distribution is governed by gastrointestinal absorption that is different for each mycotoxin and for each animal species (Grenier and Applegate, 2013). Mycotoxin metabolism in all animals occurs in both the digestive tract and the extraintestinal tissues and organs. The setup of the digestive system (absence or presence of pre-stomachs) and the different intestinal cell metabolism in the gut epithelium or the species-specific gut microbiota may significantly influence the formation of biomarkers of different types. The fate of a mycotoxin in the body, namely its possible transformation in pre-stomachs prior absorption, or its absorption speed/rate, is important in choosing its representative substrate (biomarker) in each animal species.
Poultry: Due to the relatively fast absorption in poultry the serum levels of exposure-based blood biomarkers for aflatoxins and ochratoxin A will show a sharp peak within 1-2 hours of ingestion of contaminated feed. Hence, the study of direct biomarkers in the blood matrix, is very sensitive to the time of blood collection. The analysis of exposure-based biomarkers of DON, fumonisins and T-2 toxins in poultry blood has a lower practical value in field evaluation because of the low absorption levels of these mycotoxins. Additionally, in poultry, blood sampling may result in sudden death of birds. Therefore, this method is only recommended for post mortem studies in poultry. There is a high biliary excretion of both high and weak absorption mycotoxins (Cavret and Lecoeur, 2006), thus, bile can be a reliable substrate for exposure-based biomarkers for most mycotoxins in poultry. In conclusion, the most suitable matrix for the field evaluation of exposure-based biomarkers in poultry are limited to bile, faeces and possibly the kidneys. For fumonisins due to their low bioavailability the exposure-based biomarkers assay in blood is not suitable, hence, the mechanism-based biomarkers could be the lone option. The only non-invasive matrix for exposure- and mechanism-based biomarkers in poultry is faeces. In general, the use of some effect-based biomarkers is the easiest for biomarker field studies in poultry with a post mortem study of organs and blood.
Pigs: The higher cost of a pig compared to a chicken requires low- or non-invasive sampling methods during field evaluation of mycotoxin biomarkers. The rapid and moderate-to-high absorption of DON, aflatoxins and ochratoxin A in pigs makes the field study of exposure-based biomarkers in blood very dependent on the sampling time since they peak in blood within 15–30 min of ingestion (Prelusky et al., 1988). For rapidly absorbed mycotoxins, urine might be a more reliable matrix for exposure-based and mechanism-based biomarkers in swine. Relatively reliable and easy evaluation of the effect-based biomarkers are available in pigs for high levels of zearalenone and DON (e.g. reproductive performance and feed intake).
Ruminants: In ruminants, the only way to evaluate mycotoxin biomarkers in the field is the use of low- or non-invasive sampling methods. Milk is the substrate most preferred by dairy producers in monitoring [exposure-based] biomarker, aflatoxin M1. This is a common practice especially in countries with a higher risk of aflatoxins in ruminant animal feed. However, this matrix is less suitable for the detection of exposure-based biomarkers of other mycotoxins due to the significantly lower feed-to-food transfer. Some DON, ochratoxin A, zearalenone or their metabolites can be detected in the blood of ruminants, however, urine is the preferred non-invasive matrix.
Field versus Scientific evaluation
The most common reason of quantifying direct mycotoxin biomarkers in scientific studies is the estimation of the dietary exposure to the mycotoxins. In the animal industry, besides exposure assessment, estimation of the potential elimination of the effects of mycotoxins in animals are also of critical importance.
Mycotoxin sequestrating agents (alternative names: deactivator, detoxifier, binder) are feed additives which aim to reduce mycotoxin toxicity by means of binding the toxin (mycotoxin binders), alterating the chemical structure of the mycotoxin in the gastro-intestinal tract to non-toxic metabolites (mycotoxin transforming agents) or diminishing the negative secondary effects of the mycotoxin exposure to animals (mycotoxin deactivators). While farmers seem excited about using exposure-based biomarkers as a practical tool to check the performance of mycotoxin sequestrants, it is crucial to consider the several important limitations:
- bioavailability of mycotoxins: Some of the mycotoxins like fumonisins are poorly absorbed, therefore, can hardly be detected in blood or urine. Others, such as DON, have species specific kinetics and are well absorbed in pigs but poorly absorbed in poultry and adult ruminants. Therefore, the choice of a physiological substrate is specific to each mycotoxin and to each animal species.
- kinetics of absorbed mycotoxins: Most mycotoxins peak in blood within 1-2 hours of ingestion. Choosing blood as the substrate of exposure-based biomarker for field test requires a very controlled feeding and blood collection schedule. Additionally, parent mycotoxins can be rapidly metabolised in the liver to another toxic or non-toxic compound. Consequently, it is important to be able to detect several mycotoxin metabolites that can also be species specific.
- biological effect of each mycotoxin: For instance, the main negative effects of fumonisins and trichothecenes are related to the adverse effect on the gastrointestinal tract. The use of certain binders can result in a reduction of DON levels in blood and urine and this can lead to a wrong conclusion being drawn regarding “a positive” effect of an additive. According to some researchers, such binders can give rise to higher levels of DON in the distal parts of the gut that can negatively affect gut integrity (Osselaere et al., 2013). This can increase the incidence and severity of bacterial and coccidial outbreaks.
- high variations between individual animals is important to consider when measuring biomarkers levels.
- analytical procedure: Blood, urine, faeces and organs are very complex matrices and only a few laboratories in the world have validated methods for the quantitative evaluation of exposure-based biomarkers of concern. It is especially more of a problem for metabolites, because they require other, very specific, expensive and hardly available standards.
- the primary mode of action of the product: The assay of mycotoxins levels and their metabolites in body fluids and tissues can only represent either the adsorption mode of action of the product or its biotransforming action. Measuring exposure-based biomarkers cannot demonstrate the effect of mycotoxin deactivators with other modes of action against mycotoxins or their negative consequences. The most representative example is fumonisins, where the mode of action other than binding is responsible for the positive effects of complex mycotoxin deactivators on control of Sa/So levels.
- control group is needed to draw the correct conclusion about the binding ability of the product in vivo.
It could seem easier to evaluate the efficacy of a product by measuring the level of mycotoxin or its metabolite in physiological fluids than to rely on animal performance data obtained in the long term. However, this approach depends a lot on several factors, including the specific toxin being measured, the availability of certain tissues or liquids, the specific purpose of the study and mode of action of the product. The analytical method available is certainly a bottleneck, with factors such as specifity, sensitivity, cost and throughput being important. Most methods for specific biomarkers still need to be optimised and validated (Vineis and Garte, 2008).
Therefore, there is no practical method based on direct (exposure- and mechanism-based) biomarkers that can draw accurate conclusions about the efficacy of the mycotoxin sequestrant that can be recommended for farmers. Currently, combining information about clinical indicators (indirect effect-based biomarkers), suspected or confirmed exposure (analysis of feed) and research in vivo data of the producer of the anti-mycotoxin additive is the best and the most economical solution for the evaluation of the product.
Armorini, S., Al-Qudah, K.M., Altafini, A., Zaghini, A. and P. Roncada. 2015. Biliary ochratoxin A as a biomarker of ochratoxin exposure in laying hens: An experimental study after administration of contaminated diets. Res Vet Sci. June 2015; 100:265-70
Baldwin, T.T. Riley, R.T., Zitomer, N.C., Voss, K.A., Coulombe Jr. R.A., Pestka, J.J., Williams, D.E. and A.E. Glenn. 2011. The current state of mycotoxin biomarker development in humans and animals and the potential for application to plant systems. World Mycotoxin Journal, August 2011; 4 (3): 257-270
Biasucci, G., Calabrese, G., Di Giuseppe, R., Carrara G., Colombo, F., Mandelli, B., Maj, M., Bertuzzi, T., Pietri, A. and F.L. Rossi. 2010. The presence of ochratoxin A in cord serum and in human milk and its correspondence with maternal dietary habits. Eur J Nutr. 2011 April; 50(3):211-8
Cavret, S. and S. Lecoeur. 2006. Fusariotoxin transfer in animal. Food Chem Toxicol. 2006, 44, 444–453
Devreese, M., De Baere, S., De Backer, P. and S. Croubels. 2012. Quantitative determination of several toxicological important mycotoxins in pig plasma using multi-mycotoxin and analyte-specific high performance liquid chromatography-tandem mass spectrometric methods. Journal of Chromatography A, September 2012; 28; 1257: 74-80
Grenier, B. and T.J. Applegate. 2013. Modulation of Intestinal Functions Following Mycotoxin Ingestion: Meta-Analysis of Published Experiments in Animals, Toxins, 5(2), 396-430
Groopman, J. D. and T.W. Kensler. 1993. Molecular biomarkers for human chemical carcinogen exposures. Chem. Res. Toxicol. November-December 1993; 6(6):764-70
Kraft, W. and U. Dürr. 2005. Klinische Labordiagnostik in der Tiermedizin. 6. Auflage, Schattauer. Stuttgart, pp. 42-217
Martinez-Larranaga, M.R., Anadon, A., Diaz, M.J., Fernandez-Cruz, M.L., Martinez, M.A., Frejo, M.T., Martinez, M., Fernandez, R., Anton, R.M., Morales, M.E. and M. Tafur .1999. Toxicokinetic and oral bioavailability of fumonisin B1. Vet Human Toxicol, 41, 357–62
Osselaere, A., Santos, R., Hautekiet, V., De Backer, P., Chiers, K., Ducatelle, R. And S. Croubels. 2013. Deoxynivalenol Impairs Hepatic and Intestinal Gene Expression of Selected Oxidative Stress, Tight Junction and Inflammation Proteins in Broiler Chickens, but Addition of an Adsorbing Agent Shifts the Effects to the Distal Parts of the Small Intestine. PLOS One. July 2013; Volume 8, Issue 7, e69014
Perera, F.P. and I.B. Weinstein. 2000. Carcinogenesis. 21(3), 517-524
Prelusky, D.B., Hartin, K.E., Trenholm, H.L., and J.D. Miller. 1988. Pharmacokinetic fate of 14C-labeled deoxynivalenol in swine. Fundam. Appl. Toxicol. 10: 276–286
Riley, R.T. and K.A. Voss. 2006, Differential sensitivity of rat kidney and liver to fumonisin toxicity: organ-specific differences in toxin accumulation and sphingoid base metabolism. Toxicol Sci, 92: 335–45
Sewram, V., Mshicileli, N., Shephard, G., and W. Marasas. 2003. Fumonisin mycotoxins in human hair. Biomarkers, 8:2: 110-118
Sulzberger, S.A., Melnichenko, S. and F.C. Cardoso. 2017. Effects of clay after an aflatoxin challenge on aflatoxin clearance , milk production, and metabolism of Holstein cows. J. Dairy Sci. 100:1856–1869
Turner, P., White, K., Burley, V., Hopton, R., Rajendram, A., Fisher, J., Cade, J.E. and C.P. Wild. 2010. A comparison of deoxynivalenol intake and urinary deoxynivalenol in UK adults. Biomarkers. September 2010; 15(6): 553-562
Voss, K.A., Smith, G.W. and W.M. Haschek. 2007. Fumonisins: toxicokinetics, mechanism of action and toxicity. Anim Feed Sci Technol, 137, 299–325
Vineis, P. and S. Gart. 2008. Molecular Epidemiology of Chronic Diseases, Chapter 6. Biomarker Validation, pp 71-81