Anti-mycotoxin feed additives: effects on metabolism, mycotoxin excretion, performance, and total tract digestibility of dairy cows fed artificially multi-mycotoxin-contaminated diets

The aim of this study was to evaluate the effects of different anti-mycotoxin feed additives on the concentration of mycotoxins in milk, urine, and blood plasma of dairy cows fed artificially multi-mycotoxin-contaminat-ed diets. Secondarily, performance, total-tract apparent digestibility of nutrients, and blood parameters were evaluated. Twelve multiparous cows (165 ± 45 d in milk, 557 ± 49 kg body weight, and 32.1 ± 4.57 kg/d milk yield at the start of the experiment) were blocked according to parity, milk yield, and days in milk and used in a 4 × 4 Latin square design experiment with 21-d periods, where the last 7 d were used for sampling and data analysis. Treatments were: 1) Mycotoxin group ( MTX ), basal diet ( BD ) without anti-mycotoxin feed additives; 2) Hydrated sodium calcium aluminosilicate ( HSCA ), HSCA added to the BD at 25g/cow/d; 3) Mycotoxin deactivator 15 ( MD15 ), MD (Mycofix® Plus, dsm-firmenich) added to the BD at 15 g/cow/d; and 4) Mycotoxin deactivator 30 (MD30), MD added to the BD at 30 g/cow/d. Cows from all treatments were challenged with a blend of mycotoxins containing 404 μg aflatoxins B1 ( AFB1 ), 5,025 μg deoxynivalenol ( DON ), 8,046 μg fumonisins ( FUM ), 195 μg T2 toxin ( T2 ), and 2,034 μg of zeara - lenone ( ZEN ) added daily to the BD during the last 7 d of each period. Neither performance (milk yield and composition) nor nutrient digestibility was affected by treatments. All additives reduced aflatoxin M1 ( AFM1 ) concentration in milk, whereas MD15 and MD30 group had lower excretion of AFM1 in milk than HSCA. DON, FUM, T2, or ZEN were not detected in milk of MD15 and MD30. Concentrations in milk of DON, FUM, T2, and ZEN were similar between MTX and HSCA. Except for AFM1, none of the analyzed mycotoxins were detected in urine of MD30 group. Comparing HSCA to MD treatments, the concentration of AFM1 was greater for HSCA, whereas MD30 was more efficient at reducing AFM1 in urine than MD15. AFM1, DON, FUM, and ZEN were not detected in the plasma of cows fed MD30, and DON was also not detected in MD15 group. Plasma concentration of FUM was lower for MD15, similar plasma FUM concentration was reported for HSCA and MTX. Plasma concentration of ZEN was lower for MD15 than MTX and HSCA. Serum concentrations of haptoglobin and hepatic enzymes were not affected by treatments. Blood concentration of sodium was lower in HSCA compared with MD15 and MD30 groups. In conclusion, the mycotoxin deactivator proved to be effective in reducing the secretion of mycotoxins in milk, urine, and blood plasma, regardless of the dosage. This reduction was achieved without adverse effects on milk production or total-tract digestibility in cows fed multi-mycotoxin-contaminated diets over a short-term period. Greater reductions in mycotoxin secretion were observed with full dose of MD.


INTRODUCTION
Mycotoxins are low molecular weight secondary metabolites produced by filamentous fungi (Jarvis and Miller, 2005), which are widespread toxic contaminants Anti-mycotoxin feed additives: effects on metabolism, mycotoxin excretion, performance, and total tract digestibility of dairy cows fed artificially multi-mycotoxin-contaminated diets and endanger both animal and human health (Bryden, 2012).The paramount fungal genera responsible for mycotoxin production include Aspergillus, Penicillium, and Fusarium.Within these genera, prominent classes of mycotoxins are aflatoxins (AF), ochratoxin A (OTA), fumonisins (FUM), deoxynivalenol (DON), and zearalenone (ZEN) (Coppa et al., 2019).The production of mycotoxins is linked to environmental conditions, plant stress, and damage caused by rodents and pests to grains, as well as abiotic factors such as food pH and moisture content (Bhat et al., 2010;Magan et al., 2011;Makau et al., 2016).Because mycotoxins exhibit remarkable stability due to their physical and chemical properties, they persist over extended periods throughout the grain harvesting, transportation, and storage processes (Qin et al., 2020).
Upon consuming feed tainted with mycotoxins, animals may manifest an array of symptoms, including gastrointestinal dysregulation, diarrhea, soft stools, immunosuppression, and a general decline in performance (EFSA, 2004;Pestka et al., 2004).Although the precise biological mechanisms driving these responses remain have not been fully elucidated, hallmark indicators often include ruminal or gut dysbiosis, heightened permeability of the rumen or gut epithelia, and damage of gut epithelium (Antonissen et al., 2014).In addition, when dairy cows ingest mycotoxins, their metabolites or unmetabolized compounds may be transferred to milk providing an additional source of dietary mycotoxins exposure for humans (Campagnollo et al., 2016).
Among the post-harvest mitigation strategies to mycotoxin-contaminated diets, feeding hydrated sodium calcium aluminosilicate (HSCA) -a clay-based sequestering agent -has been effectively used to reduce gastrointestinal absorption of AF and its secretion in milk (Jiang et al., 2021).However, clay-based sequestering agents have shown low efficacy against other type of mycotoxins that impairs animal performance, especially against Fusarium toxins (Döll et al., 2005;Van Le Thanh et al., 2015;Liu et al., 2022).As an alternative to clay-based sequestering agents, studies have investigated the effects of mycotoxin deactivating products (MDP) composed of a blend of inorganic components, biological components, enzymes, and phycophytic compounds in diets of dairy cows.The effectiveness of MDP was supported by positive responses in performance, immunity, and metabolism parameters in animals exposed to different types of mycotoxins (Pietri et al., 2009a;Kiyothong et al., 2012;Gallo et al., 2020).Despite studies revealing positive effects of MDP on several health and performance attributes, they have failed to report the effects of MDP on toxins excretion in milk.
The aim of this study was to evaluate the effects of different anti-mycotoxin feed additives on the concentration of mycotoxin in milk, urine, and blood plasma of dairy cows fed artificially multi-mycotoxin-contaminated diets over a short-term period.We hypothesized that the contaminated diet, even provided for a short-term period, would reduce total-tract digestibility, hence reduce the performance of cows, as well as the MDP in the diets would inhibit these effects and reduce the concentration of mycotoxins in the blood plasma, urine, and milk.

MATERIALS AND METHODS
This study was carried out between April and July 2021 at the Laboratory on Dairy Cattle Research (Laboratório de Pesquisa em Bovinos de Leite, Pirassununga, Brazil) under the approval of the Ethics Committee on Animal Use from the School of Veterinary Medicine and Animal Sciences, University of Sao Paulo (protocol 6324160123).

Treatments and Experimental Design
Twelve Holstein cows (165 ± 45 DIM, 557 ± 49 kg BW, and 32.1 ± 4.57 kg/d milk yield at the start of the experiment) blocked according to parity, milk yield, and DIM were enrolled into a replicated 4 × 4 Latin square design experiment and cows within each block (Latin square) were randomly assigned to one of the treatment sequences.Experimental periods lasted 21 d of which the first 14 d were allowed for treatment adaptation and 7 d were used to for mycotoxin challenge and data collection.Treatments were: 1) Mycotoxin group (MTX), basal diet (BD) without anti-mycotoxin feed additives; 2) Hydrated sodium calcium aluminosilicate (HSCA), HSCA added to the BD at 25g/cow/d; 3) Mycotoxin deactivator 15 (MD15), MD (Mycofix Plus, dsm-firmenich) added to the BD at 15 g/cow/d; and 4) Mycotoxin deactivator 30 (MD30), MD added to the BD at 30 g/cow/d (Table 1).The doses used were those recommended by the product manufacturers.According to the manufacturer, the MD mechanism of action is associated with 3 fundamental properties (Pietri et al., 2009a;Grenier et al., 2013;Murugesan et al., 2015): (1) An inorganic components, such as bentonite, can adsorb polar and planar mycotoxins like aflatoxin B1 (AFB1); (2) The Genus nov.species nov. of the family Coriobacteriaceae strain Biomin BBSH797 deactivates trichothecenes, and a biological component derived from inactivated yeast which deactivates ZEN, as well as a purified enzyme called FUMzyme (dsmfirmenich) at a concentration of 30,000 U/kg of Mycofix transforms FUM into non-toxic metabolites; and (3) phycophytic substances, typically a blend of plant and algae extracts, offer "bio-protection" that protects vulnerable organs such as the liver and strengthen the immune system of animals.Mycotoxins culture material pool (AFB1, FUM, ZEN, DON, and T2-toxin) and the HCSA and MDP were administered on TMR top-dressed and hand mixed into the first third top of TMR to guarantee total intake by animals.HSCA and MDP were administered during 21 d of each experimental period once a day at 0700 h while the mycotoxins pool was administered during the last 7 d of each experimental period once a day at 0700 h.Cows received the multi-mycotoxin culture material with 404 µg of AFB1 + 5,025 µg of DON + 2,034 µg of ZEN + 8,046 µg of FUM + 195 µg of T2 toxin (T2) per day during the 7 d of contamination period.
Cows were housed in a barn with individual pens (17.5 m 2 of area), with concrete floor, sanded beds, individual feed bunks, fans, and free access to water.Diets were provided twice a day (0700 and 1300 h).The BD (Table 2) was formulated according to the NRC (2001) nutrient requirements estimates of cows with 600 kg BW, 32.0 kg/d milk yield, and 3.5% fat.Diets had a forage concentrate ratio of 48:52.Refusals of each cow were weighted daily to maintain orts between 5 and 10% (on an as-fed basis) of feed supplied on the previous day.

Mycotoxin Production and Evaluation in the Diets
The implemented doses of mycotoxins in this study were based on an initial survey that aimed to determine the types and levels of mycotoxins occurring in the dairy farms from Brazil (Gruber et al., 2019).The production of aflatoxins (B1, B2, G1, and G2), fumonisins (B1 and B2), T2 toxin, deoxynivalenol, and zearalenone was carried out at the University of São Paulo following the protocol described by Müller et al. (2017).The concentration of each mycotoxin in the diet was standardized throughout the entire experimental period.After weighing, these mycotoxins were added and mixed into the diets in the last 7 d with each experimental period.A mixture of 400 µg of AFB, 5000 µg of DON, 2000 µg of ZEN, 8000 µg of FUM, and 200 µg of T2 was administered daily for 7 d during the contamination period.
Once a week, 3 samples of TMR were collected and analyzed for mycotoxin evaluation, the samples were immediately sent to the Laboratory of Food Microbiology and Mycotoxicology at the University of São Paulo.The quantification of mycotoxins (AFB1, AFB2, AFG1, AFG2, DON, FUM1, FUM2, T2, and ZEN) was made by ultra-performance liquid chromatography electrospray ionization tandem MS (UPLC-ESI-MS/MS, Waters Acquity; Waters Corp.), involving the isotopic dilution step and a data normalization process, as described by Franco et al. (2019).

Multi-Mycotoxins Evaluation in Milk, Urine, and Plasma samples
Multi-mycotoxins were evaluated in milk, urine, and blood plasma samples from the dairy cows.Before determining multi-mycotoxins in milk, milk production was  measured daily, and the data from the last 7 d in each experimental period were recorded.For the evaluation of mycotoxins in milk and urine, samples were collected on the day before supplying a pool of the mycotoxins and on the last day of the contamination period.Blood was collected on the final day of each multi-mycotoxin contamination period without the addition of preservatives and was stored under refrigeration (−80°C) before the analytical testing.The analysis of multiple mycotoxins in milk, urine, and blood plasma was performed as the methodology proposed by Solfrizzo et al. (2011) andFlores et al. (2017), with minor modifications.Before the extraction of mycotoxins in urine and blood plasma, samples were centrifuged to remove particulate matter and supernatants.After this procedure, glucuronidase/sulfatase was added to the sample for the enzymatic deconjugation of mycotoxins.Then, samples were incubated under static conditions at 37°C overnight.The determination of multi-mycotoxins milk, urine, and blood plasma samples were conducted using a Waters Acquity I-Class ultra-performance liquid chromatography (UPLC) system (Waters Corp.) equipped with a BEH Column C18 (2.1 × 50 mm, 1.7 μm) and coupled to a Xevo TQ-S mass spectrometer (Waters Corp.).Mass spectrometry (MS) analyses were carried out in a multiple reaction monitoring (MRM) mode by using electrospray ionization in a positive ion mode.The chromatographic procedure, MS parameters, and the MRM transitions (Supplemental Tables S1, S2, and S3; https: / / doi .org/ 10 .6084/m9 .figshare.25782945)were the same adopted by Franco et al. (2019) and Frey et al. (2021).

Nutrient intake, Feed Sorting Index, and Apparent Digestibility
Feed offered and refusals were recorded daily to determine feed intake.Samples of silage and refusals were collected daily during the last 7 d of each experimental period, the concentrate ingredients were sampled during the concentrate manufacturing before starting the collection period.Samples of daily refusals represented approximately 10% of total refusals.The refusal and feed samples were analyzed for contents of DM (method 930.15;AOAC International, 2000), ash (method 942.05;AOAC International, 2000), CP (N × 6.25;Kjeldahl method 984.13;AOAC International, 2000), and ether extract (EE;method 920.39;AOAC International, 2000).Neutral detergent fiber (Van Soest et al., 1991) was analyzed using α-amylase (TE-149 fiber analyzer; Tecnal Equipamentos para Laboratório Inc.), and acid detergent fiber (ADF) and lignin (method 973.18) were analyzed according to AOAC (2000).Feed ingredients were analyzed for contents of starch using an enzymatic degradation method (Amyloglicosidase, Novozymes Latin America Ltda.) and absorbances measured on a spectrophotometer (SBA-200, Celm) according to Hendrix (1993).The feed sorting index was calculated based on the predicted intake of particle size distribution of TMR and refusals as described by Silveira et al. (2007).
Undigested NDF (uNDF) content in feeds, refusals, and feces were used to estimate fecal excretion of DM.Fecal samples (n = 8) were collected directly from the rectum of cows every 9 h during 3 consecutive days (d 15 -0600, 1500, and 0000 h; d 16 -0900 and 1800 h; d 17 -0300, 1200, and 2100 h) and pooled for further analyses.For the uNDF analysis, ground samples (2mm) of feeds, refusals, and feces were placed in nonwoven fabric bags (12 µm pore size, 5 × 5 cm at 20 mg DM/cm 2 ) and incubated in the rumen of 2 cannulated dry cows during 288 h (Huhtanen et al., 1994;Casali et al., 2008).After removal from the rumen, bags were washed in running tap water, dried, and NDF content of residue was determined.Digestibility of DM and nutrients were calculated using following equations: .

Blood metabolites, BW, and BCS
Blood samples were collected from the coccygeal vessels in 10 mL vacuum tubes without clot activator (BD Vacutainer, Becton Dickinson) and lithium heparin tubes 4 h after the morning feeding (1100 h) on d 21 in each experimental period.After clotting, blood samples from tubes without clot activator were centrifuged (2,000 × g for 15 min at room temperature), the serum was harvested and stored at −20°C for analysis of hepatic enzymes aspartate aminotransferase (AST), alanine transaminase (ALT), and gamma-glutamyl transpeptidase (GGT).Enzymes were analyzed by commercial colorimetric kits (Bioclin) and absorbances were measured on a semiautomatic biochemical analyzer (SBA-200, Celm).Serum haptoglobin concentration was analyzed in a commercial laboratory (VidaVet) using the method described by Cooke and Arthington (2013).
Whole blood samples from lithium heparin vacuum tubes were analyzed in portable biochemical blood analyzers VetScan VS2 (Zoetis Services LLC) for determination of albumin, amylase, blood urea nitrogen, calcium, creatinine, globulin, glucose, potassium, phosphorus, and total protein.The analysis of whole blood samples for hemoglobin, hematocrit, chloride, bicarbonate, pH, anion gap, base excess in the extracellular fluid compartment, partial pressure of carbon dioxide, and total carbon dioxide content were performed using the I-STAT EC8+ CARTRIDGE equipment (Abbot Point Care Inc.).The analysis of vitamin E and β-carotene were performed using the iCheck Vitamin E equipment (BioAnalyt) with the same whole blood samples.
Body weight was measured weekly before the morning feeding and after milking, using an electronic scale for large animals.Body condition score was assessed on the last day of each experimental period using a 5-point system (1 = emaciated to 5 = obese) according to Wildman et al. (1982).

Statistical Analysis
Data from the last 7 d of milk yield and DMI were averaged and used for statistical analysis.Milk composition data from each period were averaged and used for calculating yields of fat, protein, lactose, and FCM used in statistical analysis.Data were submitted to ANOVA using the PROC MIXED of SAS 9.4 (Statistical Analyses for Windows -SAS Institute Inc.) according to the following model: ; 0 2 σ ; where: n = Gaussian distribution; σ c 2 = estimated variance associated to cows; and σ e 2 = estimated residual variance, Y ijkl = observation on animal l, given treatment i, at period j, in square k; μ = overall mean, A i = fixed effect of the ith treatment (i = 1 to 4); P j = fixed effect of the jth period (j = 1 to 4); S k = fixed effect of the kth Latin square (k = 1 to 3); TP ik = interaction fixed effect between treatment and period; AS ik = interaction fixed effect between treatment and Latin square; α kl = random effect of animal within square (l = 1 to 12); ε ijkl = random error associated with each observation.Means were adjusted by LSMEANS and degrees of freedom were calculated using the Kenward and Roger (1997) method.LSMEANS were computed for each treatment level to estimate the mean response variable.Differences in response variables were evaluated by Tukey's test.The significance level was set at P ≤ 0.05.

Contamination and mycotoxin intake
In this longitudinal assessment of the diets used for feeding cows, several mycotoxins were detected in the corn silage and in the diet before the artificial contamination (between d 1-14 of each period; Table 3).The daily intake of mycotoxins was slightly greater (1.0, 1.22, and 0.136% for AF, FUM, and ZEN, respectively) than planned due to natural mycotoxin contamination present in the feed (corn silage and concentrate) and the relatively high DMI by the cows.However, no differences were observed in the daily mycotoxin intake (between d 15-21 of each period) between treatments as proposed in the study (Table 4).

Concentrations of mycotoxin in milk, urine, and blood
Milk concentration of AFM1 was lower (P < 0.001) when cows were fed additives compared with MTX.MD15 and MD30 group had lower excretion of AFM1 in the milk than HSCA (P < 0.001) and were similar to each other (Table 5).DON, FUM, T2, or ZEN were not detected in the milk of cows fed MD15 and MD30.Concentrations of DON, FUM, T2, and ZEN were similar between MTX and HSCA groups.
Urine concentration of AFM1 was higher (P < 0.001) for MTX treatment than other treatments.Comparing HSCA to MD treatments, the concentration of AFM1 was greater for HSCA, and MD30 was more efficient at reducing AFM1 in urine than MD15.Urine concentrations of DON was greater (P < 0.001) for HSCA than MTX, whereas DON was not detected in urine of cows in MD15 and MD30 groups.FUM was not detected in urine of cows fed MD treatments, and similar concentrations of FUM were observed in MTX and HSCA groups.α-Zearalenol (α-Zel), or β-zearalenol (β-Zel) were not detected in urine of cows fed MD30.Urine concentration of α-Zel was lower (P < 0.001) in MD15 group in comparison with HSCA and MTX, whereas α-Zel concentration was the greatest in MTX group.Urine concentration of β-Zel was lower for MD15 whereas urine β-Zel concentration was similar between MTX and HSCA groups.
AFM1, DON, FUM, and ZEN were not detected in plasma of cows fed MD30, and DON was also not detected in MD15 group.T2 toxin was not detected in the plasma of any of the cows.Blood concentration of AFM1 was reduced (P < 0.001) with additives whereas MD15 was more efficient in reducing AFM1 in blood than HSCA.Plasma concentration of DON was similar between MTX and HSCA groups.Plasma concentration of FUM was lower (P < 0.001) in cows fed MD15, whereas similar plasma FUM concentration was observed between HSCA and MTX groups.Plasma concentration of ZEN was lower (P < 0.001) for MD15 in comparison with MTX and HSCA.

Performance, total-tract digestibility, and blood metabolites
Dry matter intake, or nutrient intake and digestibility were not affected by treatments (Table 6).Feed sorting index, BW, and BCS were similar among treatments.Milk yield, FCM yield, or milk composition were not affected by treatments (Table 7).
Treatment comparisons did not reveal differences in serum concentrations of haptoglobin and hepatic enzymes (ALT, GGT, AST, and ALP; Table 8).Blood creatinine concentration was greater (P = 0.009) in MD15 group in comparison with other treatments, which exhibited similar blood creatinine concentration to each other (Table 9).Blood concentrations of albumin, globulin, hemoglobin, total protein, urea-N, glucose, amylase, and bicarbonate were similar among treatments.Similar blood hematocrit, total carbon dioxide, partial pressure of carbon dioxide, pH and base excess in the extracellular fluid compartment were observed among treatment groups.
Blood concentration of sodium was lower (P ≤ 0.012) in HSCA compared with MD15 and MD30 groups (Table 10).Blood concentrations of chlorine, calcium, phosphorus, potassium, vitamin E and β-carotene were similar among treatments.An interaction effect between treatment and Latin square was observed (P = 0.022) for blood vitamin E concentration where cows in the Latin square with the lowest initial milk yield and DMI tended to exhibit different (P = 0.065) concentrations for HSCA and MD15 (6.07 and 7.65 mg/L, respectively).

DISCUSSION
It was postulated that feeding MDP treatments would reduce the absorption of mycotoxin and consequently the concentration in the blood, thereby reducing the secretion of mycotoxins in the milk and urine of dairy cows challenged with a mycotoxin blend in the diet.Indeed, cows fed with both doses of MD had lower concentrations of mycotoxins in milk, urine, and blood in comparison with MTX and HSCA treatments.In addition, it was expected that mycotoxins would have detrimental effects in the gastrointestinal tract that could impair digestibility (Liew and Mohd-Redzwan, 2018) and consequently milk production.However, no effects on animal performance and total and total apparent digestibility were observed, possibly due to the contamination levels not being sufficiently high and the short-term period of exposure (7 d).It is important to highlight that this study has no genuine control group (i.e., a group with cows not exposed to mycotoxin challenge) and the results should be interpreted with caution.Concentrations of mycotoxins initially proposed in this study are very close to what was observed in the mycotoxicological analysis (Gruber et al., 2019).Treatments containing MDP reduced mycotoxins in milk of dairy cows.Studies have shown reduced AF in milk when antimycotoxin additives were fed (Kutz et al., 2009;Pietri et al., 2009), but further studies are warranted to determine the effectiveness of these additives in reducing the secretion of other mycotoxin in milk.As expected, HSCA was effective in reducing concentration of AFM1 in blood, milk, and urine.It is well known that HSCA is effective in binding AF and preventing its absorption by the intestine and entering the bloodstream (Kubena et al., 1993;Neeff et al., 2013;Awuchi et al., 2021).Although HSCA is an effective adsorbent for AF, it does not minimize the toxic effects of other mycotoxins such as DON and ZEN (Döll et al., 2005;Van Le Thanh et al., 2015b;Liu et al., 2022).In this study, feeding MD15 and MD30 has been shown to be more effective in reducing AFM1 in milk, urine, and blood in comparison to feeding HSCA.A previous study showed that an earlier MDP was able to reduce the AF in milk by 31% and 41% when fed at 20 g/ cow/d and 50 g/cow/d, respectively (Pietri et al., 2009).Several reviews have also described the effectiveness of MDP in counteracting the negative effects of mycotoxins in poultry, swine, and dairy cows (Cheng et al., 2006;Gallo et al., 2020;Kehinde et al., 2020).
The US Food and Drug Administration has set an action level for AFM1 of 0.50 μg/kg in liquid milk, total AF of 20 μg/kg in feed ingredients offered to dairy and breeding cattle (FDA, 2019).The European Commission set up an action level for AFM1 of 0.05 μg/kg in liquid milk, aflatoxin B1 of 20 μg/kg in all feedstuffs, 10 μg/kg in complete feeds for cattle, sheep, and goats, and 5 μg/ kg in complete feeds for dairy cattle (EC, 2003;2006).In this study, the level of AFM1 found in milk was 1.31 μg/ kg, 0.79 μg/kg 0.08 μg/kg, and 0.03 μg/kg for the MTX, HSCA, MD15, and MD30 respectively.It is worth noting that the cows in the MTX group were also contaminated with mycotoxins, which is the reason why they exhibited such a high level of mycotoxin in their milk.However,  3 P-values for effects of treatment (Trt), period (Per), Latin square (LS), interaction between treatment and period (Trt*Per), interaction between treatment and Latin square (Trt*LS).
only MD30 has been able to reduce AFM1 in milk below the limit allowed by the European Commission.
Regarding the other mycotoxins, HSCA presented no difference compared with the MTX, meaning it did not reduce the concentration of FUM, T2, or ZEN in the blood, milk, and urine due to the varying affinities that different mycotoxins have for adsorbents.HSCA is an inorganic adsorbent that shows a remarkable ability to bind with nonpolar mycotoxins, such as AF (Vekiru et al., 2007), being inefficient to bind other kinds of mycotoxins.However, treatment MD15 was able to reduce other mycotoxin (DON, FUM, T2, and ZEN) concentrations in  2 P-values for effects of treatment (Trt), period (Per), Latin square (LS), interaction between treatment and period (Trt*Per), interaction between treatment and Latin square (Trt*LS).
3 No sorting: 1, values 1 indicates sorting for particles on the particular particle size range; sorting index was calculated according to Silveira et al. (2007). 2 P-values for effects of treatment (Trt), period (Per), Latin square (LS), interaction between treatment and period (Trt*Per), interaction between treatment and Latin square (Trt*LS).
milk to levels below the detection limit.The same occurred to DON, FUM, and T2 in the urine.α-Zel and β-Zel are metabolites of ZEN that was measured in the urine, whereas the MD15 has caused 97.5% and 92.4% reduction of α-Zel and β-Zel, respectively.The results of the mycotoxins in the blood resemble those found in milk and urine.Plasma levels of DON and T2 were reduced below the LOD for MD15 group, while FUM and ZEN decreased by a remarkable 96.3% and 97.7% for MD15 and MD30, respectively.MD30 treatment was the most effective in reducing the levels of mycotoxins; except for AFM1 in milk and urine, mycotoxins analyzed in this  2 P-values for effects of treatment (Trt), period (Per), Latin square (LS), interaction between treatment and period (Trt*Per), interaction between treatment and Latin square (Trt*LS).
3 Total carbon dioxide. 4Partial pressure of carbon dioxide. 5Base excess in the extracellular fluid compartment.study were below the detection limit for milk, urine, and blood in cows fed MD30.MD has 3 mechanisms of action that supports its antimycotoxin effect.Inorganic constituents, specifically a blend of bentonites and diatomaceous earths present in MD, serve to adsorb polar mycotoxins such as AF (Vekiru et al., 2007).Non-adsorbable mycotoxins (e.g., trichothecenes, ZEN) are biotransformed by Genus nov.species nov. of the family Coriobacteriaceae strain (BBSH7) and a yeast strain affiliated to the Trichosporon genus (Trichosporon mycotoxinivorans MTV; Schatzmayr et al., 2006).Ultimately, phycophytic compounds sourced from a marine algae (Ascophyllum nodosum) and extracts from plants (Silybum marianum) function to protect vulnerable organs such as the liver and strengthen the immune system (Murugesan et al., 2015).The reduction of AF in the blood was observed for HSCA and MD15 (35% and 91%, respectively), although MD30 decreased AF concentration in plasma below LOD (0.033 μg/L).
Hepatic enzymes (GGT, AST, ALT, and ALP) are indicators of liver function (Santos;Fink-Gremmels, 2014;Xiong et al., 2015).There were no significant differences in most serum biochemical parameters among the treatments.Similar results have been observed in dairy cows fed with a basal diet contaminated with AF (0 μg/kg, 20 μg/kg and 40 μg/kg of AF), and no changes were found in hepatic enzymes (Wang et al., 2019).Data on levels of hepatic enzymes in healthy bovine is still controversial in the literature, probably because AST, ALT, and GGT are not constant during the lactation and pregnancy phases (Tainturier et al., 1984;Stojević et al., 2005) and it depends on breed, sex, age, feeding, etc. (Zaitsev et al., 2020).Kaneko et al. (2008) suggest adequate blood concentrations of 78 -132 U/L for AST, 6.1 -17.4 U/L for GGT, and 11 -40 U/L for ALT, however the cattle breed data is not available.Stojevic et al. (2005) studied the hepatic enzymes levels of 120 clinically healthy cows at different stages of lactation and found 44.9 ± 6.9 U/L, 20.08 ± 3.7 U/L, and 14.72 ± 3.7 U/L for AST, ALT, and GGT respectively for cows after the peak of lactation.In the current study, serum concentrations of AST, GGT, and ALT were 69.7 U/L, 26.9 U/L and 49.8 U/L, respectively, slightly higher than reported in the literature (Stojevic et al., 2005;Kaneko et al., 2008), as diets from all treatments were contaminated with mycotoxins.An increase in these enzymes after the mycotoxin challenge was expected, at least in MTX and HSCA groups, as they can cause damage photosensitization in cattle (Casteel et al., 1995).Serum amylase is considered a biomarker of acute pancreatitis (Salt and Schenker, 1976;Furey et al., 2020).Creatinine serves as an indicator of glomerular filtration rate, which is considered the most reliable measure of kidney function (Kdigo, 2013).In this trial, serum creatinine concentration was greater in MD15 compared with other treatment groups.However, serum creatinine remained within the range of normal concentration (Titgemeyer and Loest, 2001).The reasons for different serum creatinine concentrations between MD15 and other groups are unclear.
Mycotoxins adsorbents can bind to vitamins and minerals (Ward et al., 1991;Huwig et al., 2001;Liu et al., 2022), but no treatment differences were found in blood concentrations of β-carotene, Ca, or P. Despite reports of mycotoxin adsorbents binding to vitamins and minerals, the results of our study did not show differences in plasma concentrations of vitamin E, β-carotene, Ca or P. MD has shown not to bind to vitamins in minerals in other studies also (Gallo et al., 2020).Blood sodium  2 P-values for effects of treatment (Trt), period (Per), Latin square (LS), interaction between treatment and period (Trt*Per), interaction between treatment and Latin square (Trt*LS).
concentration was lower in the HSCA group compared with MD15 and MD30.Despite HSCA has the capacity to absorb vitamins and minerals as well (Moshtaghian et al., 1991;Huwig et al., 2001;Liu et al., 2022), we are not able to attribute this effect to HSCA binding properties as blood levels of sodium in healthy animals on the same diet but without mycotoxins were not evaluated.However, cows with the highest mycotoxin challenge relative to DMI (i.e., cows with the lowest DMI) showed marginal differences in blood vitamin E concentration, whereas cows fed MD15 tended to present greater concentrations than those fed HSCA.
No differences were found in mycotoxin intake because the amounts of mycotoxins offered to the cows were carefully controlled, and the supply was administered through top dressing.Dry matter intake and milk yield were not affected by treatments.Similar results on these aspects are found in the literature (Kutz et al., 2009;Pietri et al., 2009).In a previous study evaluating the effects of a long period of exposure to mycotoxins (54 d) in a naturally contaminated diet, no differences were found in performance of dairy cows (n = 30) either among control group, diet with low contamination, and diet with high contamination and 100g/cow/d of antimycotoxin feed additive (Catellani et al., 2023).These results can be explained by the fact that ruminants are less sensitive to mycotoxins, they are generally protected against toxins by the rumen microbiota (Fink-Gremmels, 2008).However, Santos and Fink-Gremmels (2014) affirm that the exposure of dairy cows for a long period of the time (2 mo) can have negative effects on milk production.Therefore, the short-term exposure period of this trial can explain the lack of effects on performance and digestibility, but it reflects what can happen in dairy farms.

CONCLUSIONS
Feeding a mycotoxin-deactivator over a short-term contamination period effectively reduces AFM1, DON, FUM, T2, and ZEN in blood, milk, and urine of dairy cows in both doses.MD30 presented the greatest reduction of mycotoxins in blood, urine, and milk, followed by MD15 and HSCA.MD30 reduced all other mycotoxins below the limit of detection in blood, urine, and milk, except for AFM1 in milk and urine.The lower dose of mycotoxin deactivator (MD15) presented a significant reduction in all mycotoxins, while HSCA presented small reduction in AF, and it was not able to reduce other mycotoxins secretions in blood, urine, and milk.This study did not detect differences in performance and digestibility of dairy cows.However, even during a short-term exposure to contaminated diets, dairy cows are capable of absorbing and secreting mycotoxins in milk, which are harmful to human health.This study demonstrated the effectiveness of the mycotoxin-deactivator in reducing absorption by the cow and producing milk that is safer for consumers.
Vieira et al.: ANTI-MYCOTOXIN FEED ADDITIVES TO MID-LACTATION DAIRY COWS Vieira et al.: ANTI-MYCOTOXIN FEED ADDITIVES TO MID-LACTATION DAIRY COWS Vieira et al.: ANTI-MYCOTOXIN FEED ADDITIVES TO MID-LACTATION DAIRY COWS Vieira et al.: ANTI-MYCOTOXIN FEED ADDITIVES TO MID-LACTATION DAIRY COWS Vieira et al.: ANTI-MYCOTOXIN FEED ADDITIVES TO MID-LACTATION DAIRY COWS Vieira et al.: ANTI-MYCOTOXIN FEED ADDITIVES TO MID-LACTATION DAIRY COWS Vieira et al.: ANTI-MYCOTOXIN FEED ADDITIVES TO MID-LACTATION DAIRY COWS Vieira et al.: ANTI-MYCOTOXIN FEED ADDITIVES TO MID-LACTATION DAIRY COWS

Table 1 .
Treatments description

Table 2 .
Ingredients and chemical composition of the experimental diet

Table 3 .
Concentrations of mycotoxins in feeds and mycotoxin pool

Table 8 .
Concentration of haptoglobin and hepatic enzymes in the blood of dairy cows submitted to a multi-mycotoxin challenge and fed different anti-mycotoxin feed additives

Table 9 .
Blood parameters of dairy cows submitted to a multi-mycotoxin challenge and fed different anti-mycotoxin feed additives