The Effect of Different Doses of Zearalenone in Feed on the Bioavailability of Zearalenone and Alpha-Zearalenol, and the Concentrations of Estradiol and Testosterone in the Peripheral Blood of Pre-Pubertal Gilts

The objective of this study was to determine the effect of long-term (48 days), per os administration of specific zearalenone (ZEN) doses (20 and 40 μg ZEN/kg BW in experimental groups EI and EII, which were equivalent to 200% and 400% of the upper range limit of the no-observed-adverse-effect-level (NOAEL), respectively) on the bioavailability of ZEN and the rate of changes in estradiol and testosterone concentrations in the peripheral blood of pre-pubertal gilts. ZEN and α-ZEL levels were similar until day 28. After day 28, α-ZEL concentrations increased significantly in group EI, whereas a significant rise in ZEN levels was noted in group EII. The presence of estradiol in peripheral blood plasma was not observed until day 20 of the experiment. Spontaneous secretion of estradiol was minimal, and it was determined at very low levels of up to 10 pg/mL in EI and EII groups. Testosterone concentrations ranged from 4 to 9 ng/mL in all groups. A decrease in the concentrations of both analyzed hormones was reported in the last stage of the experiment. The results of the experiment indicate that: (i) The bioavailability of ZEN in peripheral blood has low diagnostic value, (ii) exposure to low doses of ZEN induces minor changes in the concentrations of the analyzed hormones, which could lead to situational supraphysiological hormone levels and changes in endogenous hormonal balance.


Introduction
The majority of environmental estrogens (not all of them have toxic or contaminating effects) are endocrine disrupters (EDs) [1] which are commonly encountered in soil, air, water, food products, and animal feeds [2,3]. Phytoestrogens (e.g., genistein and coumestrol) and mycoestrogens (including ZEN produced by moulds) [4,5] are natural EDs found in the environment. Arithmetic means (x) and standard deviation (SD) for the bioavailability of zearalenone (ZEN) and α-zearalenol (α-ZEL) (ng/mL) in the peripheral blood of pre-pubertal gilts on different analytical days and in the experimental groups (group EI 20 µg ZEN/kg BW; group EII 40 µg ZEN/kg BW). Limits of detection (LOD) > values below the limit of detection were regarded as equal to 0. Statistically significant differences were determined at * p ≤ 0.05 and ** p ≤ 0.01.
Significant differences (* p < 0.05) were observed between groups EII and EI: (i) In the bioavailability of ZEN on days 2, 16, 30, 36, and 38 and in their mean values for the entire experimental period (EI < EII on all sampling dates, excluding day 16); (ii) in the bioavailability of α-ZEL on days 4, 24, 28, and 38 (EI < EII on the first two sampling dates, and EI > EII on the remaining days); (iii) in the bioavailability of ZON (see Figure 2) on days 22, 30, 32, and 40, and in their mean values during the entire experiment (EI < EII on all sampling dates). Highly significant differences (** p < 0.01) were noted between groups EI and EII: (i) In the bioavailability of ZEN on days 28 (EI > EII) and 32 (EI < EII) and in ZON concentrations on days 2 (EI < EII) and 28 (EI > EII).

Figure 2.
Arithmetic means (x) and standard deviation (SD) for the bioavailability of ZON (ZEN + α-ZEL) (ng/mL) in the peripheral blood of pre-pubertal gilts on different sampling dates and in the experimental groups (group EI 20 µg ZEN/kg BW; group EII 40 µg ZEN/kg BW). Limits of detection (LOD) > values below the limit of detection were regarded as equal to 0. Statistically significant differences were determined at * p ≤ 0.05 and ** p ≤ 0.01.

Estradiol Concentrations
On the first ten sampling dates, the concentrations of E 2 were below the limit of sensitivity in all experimental groups (see Figure 3). This was also noted in group EI on day 20. Significant differences at p < 0.05 were noted between groups EI and C on days 24 and 36 (EI > C), and between groups EII and C on day 26 (EII > C). Highly significant differences at p < 0.01 were observed between groups EI and C on day 38 (EI > C). . Arithmetic means (x) and standard deviation (SD) for the concentrations of estradiol (pg/mL) in the peripheral blood of pre-pubertal gilts on different sampling dates and in the experimental groups (group EI 20 µg ZEN/kg BW; group EII 40 µg ZEN/kg BW). Limits of detection (LOD) > values below the limit of detection were regarded as equal to 0. Statistically significant differences were determined at * p ≤ 0.05 and ** p ≤ 0.01.
Between experimental days 36 and 42, a clear increase in E 2 levels was noted in both E groups in comparison with group C (see Figure 3). The concentrations of E 2 in the analyzed pre-pubertal gilts ranged from 3.0 to 10.5 pg/mL during the entire experiment. The mean increase in the values of E 2 was similar during the experiment, but considerably higher in both E groups until day 42. The concentrations of E 2 in experimental groups decreased towards the end of the experiment in comparison with control. The mean values of E 2 in peripheral blood were determined at 3.46 pg/mL in group C, 3.90 pg/mL in group EI and 3.83 pg/mL in group EII on 24 sampling dates, and at 5.77 pg/mL, 6.33 pg/mL, and 6.22 pg/mL, respectively, on the last 15 sampling dates (see Figure 3).

Testosterone Concentrations
Testosterone concentrations in blood plasma samples from pre-pubertal gilts are presented in Figure 4. During the entire experiment, T concentrations in the analyzed pre-pubertal gilts ranged from 4.0 to 9.0 ng/mL. Testosterone concentrations were higher in group C on 15 out of 25 sampling dates (60%). The changes in T concentrations were highly similar during the experiment. The mean concentration of T in peripheral blood during the entire experiment (25 sampling dates) was determined at 6.18 ng/mL in group C, 5.98 ng/mL in group EI, and 6.35 ng/mL in group EII. . Arithmetic means (x) and standard deviation (SD) for the concentrations of testosterone (ng/mL) in the peripheral blood of pre-pubertal gilts on different sampling dates and in the experimental groups (group EI 20 µg ZEN/kg BW; group EII 40 µg ZEN/kg BW). Statistically significant differences were determined at * orˆp ≤ 0.05 and ** p ≤ 0.01; *-EI or EII compared with C;ˆ-EII compared with EI. Statistically significant differences at p < 0.05 were reported between groups EI and C on days 14 and 30 (C > EI), between groups EII and C on days 6 (C < EII) and 30 (C > EII), and between groups EII and EI on days 2 (EI > EII), 20 (EI < EII), and 32 (EI < EII). Significant differences at p < 0.01 were observed between groups EII and C on day 34 (C < EII). The differences were statistically significant on days 6 and 34. A decrease in T concentrations in the experimental groups, relative to the control group, or no changes in T levels were observed more frequently.

Discussion
According to Shephard [25], animals are constantly exposed to mycotoxins which are secondary products of mould metabolism. Their elimination depends on the animal's age and the bioavailability of toxins in the blood plasma. This suggests that mycotoxins are always present in the natural environment, but at LOD doses. The bioavailability of ZEN and α-ZEL in the blood plasma is determined by the rate and the form of their biotransformation to water-soluble substances which are broken down by enzymes in the liver or other tissues as part of the detoxification process. According to some authors [26,27], biotransformation can generate products which are more toxic (bioactivation) than the parent mycotoxin. One of such compounds is ZEN which is transformed to α-ZEL in pre-pubertal gilts, and the consequences of this process are visible in tissues sensitive to that metabolite [13]. Tiemann et al. [28] observed that a decrease in steroid synthesis in cultures of porcine ovarian follicular granulosa cells resulting from the inhibition of P450scc activity is inversely proportional to the α-ZEL dose.

Zearalenone and Its Metabolite
In this study, the bioavailability of ZEN and α-ZEL in the blood plasma indicates that ZEN levels remained low in the peripheral blood of the experimental pre-pubertal gilts (groups EI and EII) in the first 28 days of the experiment. This can probably be attributed to enhanced biotransformation of ZEN and the distribution (carryover) of ZEN and α-ZEL to estrogen-sensitive tissues such as enterocytes, in particular in the duodenum and jejunum [19] in the initial period of exposure [29] and, subsequently, their bioavailability in peripheral blood.
The decrease in the values of ZEN in peripheral blood (see Figure 5) in both groups in the first two weeks of exposure provides evidence for the above, and it could indicate that: (i) All forms of mycoestrogen are intercepted by enterocytes (at absorptive level) or (ii) estrogen-sensitive tissues have a high demand for estrogen (at post-absorptive level) in pre-pubertal animals. In this experiment, the bioavailability of ZEN in peripheral blood 30 min after per os administration (maximum ZEN concentration in the blood [11]), was very low in comparison with the values reported by Dänicke and Winkler [30] and the values noted in intestinal tissues by Zielonka et al. [19] and Zheng et al. [31], which makes the resulting data difficult to interpret. It should also be noted that those values are inversely proportional to the ZEN dose (see Figure 5). Therefore, exposure to a lower ZEN dose leads to a hormetic dose response [32] by influencing the concentrations of the parent compound and its metabolite in peripheral blood, as well as the synthesis and secretion of sex steroid hormones (see Figures 3 and 4) [31,33,34]. The above state limits bioavailability. The analyzed values reflect on liver metabolism, i.e., the unmetabolised portion of the active substance (e.g., ZEN) and the produced and unutilised α-ZEL metabolite. The resulting "free ZEN" can be utilized for specific purposes [17], for example, for competing with endogenous E 2 or coregulating its supply. The above is illustrated by the fact that both doses lead to a significant increase in the concentrations that induce hyperestrogenism although rather supraphysiological hormone levels (increase in the concentrations of E 2 in both E groups and increase in T levels in group EII relative to group C) [34,35]. This observation points to a counter response to the presence of specific ZEN doses in pre-pubertal gilts. The presented results can be extrapolated [19,33,34] to suggest that "free ZEN" is captured by estrogen receptors (ERs) in the digestive tract [36], and that it stimulates qualitative changes (activation?) of ERs. The analyzed scenario testifies to the multidirectional effects of ZEN and its metabolite [33]. Therefore, it could be postulated that a dose of 20 µg ZEN/kg is more effectively utilized by pre-pubertal gilts. The above implies that biotransformation processes proceed in an identical manner, but the bioavailability of ZEN is higher in group EII and the bioavailability of α-ZEL is higher in group EI, whereas the parent compound (ZEN) and its metabolite, α-ZEL, are utilized more effectively in group EI (due to lower supply).
The values of ZON (see Figure 2), i.e., the total concentration of the parent compound and its metabolite, confirm the above suggestions. The values of ZON were higher in group EII on 15 sampling dates (60%). These findings indicate that the bioavailability of ZEN was higher in group EII, but they do not prove that ZEN was more effectively utilized for physiological processes. Low values of ZON at the beginning of the experiment can probably be attributed to the "saturation" of digestive tract tissues, in particular in regions with a higher concentration of ERs [19,36], and the low concentrations of endogenous hormones in animals during sexual maturation [5,8,34,37]. In other words, the involvement of the analyzed hormones in the stimulation of steroidogenesis and the conversion of T to E 2 ultimately led to supraphysiological hormone levels, in particular in group EII.

Concentrations of Estradiol and Testosterone
Zearalenone is an ED [22], but its presence in the diet of experimental animals did not affect the production of endogenous hormones such as E 2 and T relative to the control group [38]. The presence of E 2 (see Figure 3) was not observed on the first nine sampling dates (19 days) of the experiment. On successive sampling dates (15 dates, 53.33%), E 2 values were higher in both E groups than in group C. The concentrations of T (see Figure 4) were lower on 60% sampling dates in both experimental groups than in the control group, which provides indirect evidence for the acceleration of metabolic processes or the conversion of T to E 2 [39] under the suppressive influence of ZEN [34,[40][41][42]. The concentrations of E 2 after 20 days of exposure and serum T levels throughout the experiment were directly proportional to the ZEN dose. According to Vandenberg et al. [22], ZEN and/or α-ZEL exert a "complementary" effect on the concentrations of endogenous hormones (see Figures 3 and 4). The above is reflected by an increase in E 2 levels from day 20 to day 42 of the experiment and the decreasing trend in the concentrations of T between day 2 and the end of the experiment, which could alleviate disruptions in hormonal homeostasis [40,41]. The observed T values could also be attributed to the fact that T regulates sexual differentiation, muscle and bone mass, and erythropoietic and metabolic processes which play a very important role in the development of pre-pubertal gilts. These findings suggest that the physiological demands of young gilts considerably exceed the supply of endogenous E 2 and T. "Free ZEN" compensates for (or relieves) the resulting endocrinological dysfunction, e.g., by changing the expression of HSD enzymes [34].

Conclusions
The results of endocrinological analyses of selected estrogens (E 2 ) and androgens (T) in pre-pubertal gilts, administered at specific doses (200% and 400% of the highest NOAEL value) of ZEN (20 or 40 µg ZEN/kg BW) for 48 days suggest that: (i) The bioavailability of ZEN in peripheral blood is very low and highly varied throughout exposure, and that its diagnostic values are difficult to determine; (ii) experimentally-induced hyperestrogenism or, in other words, "supraphysiological hormone levels" contributed to a minor increase in total concentrations of E 2 (which could intensify proliferative processes) and influenced the levels of T (T values were lower in the experimental groups on 60% of sampling dates); (iii) the presented results can be extrapolated to suggest that the investigated ZEN doses caused different responses-the lower dose probably generated stimulatory/adaptive effects, whereas the higher dose probably inhibited life processes in the studied animals.
By extrapolation [34], it can be stated that the analyzed ZEN values affect the levels of E 2 and T in pre-pubertal gilts, thus reducing the risk of fighting for establishing the dominance hierarchy in the herd, contributing to changes in animal behaviour, slowing down sexual maturation, and boosting metabolism. Such observations could be important for pig breeders and veterinarians supervising pig farming and commercial feed production.

Materials and Methods
All of the experimental procedures involving animals were carried out in compliance with Polish legal regulations determining the terms and methods for performing experiments on animals (Opinion No. 14/2005/N of the Local Ethics Committee for Animal Experimentation at the University of Warmia and Mazury in Olsztyn, Poland in of 26 January, 2005).

Experimental Animals
This study was conducted at the Department of Veterinary Prevention and Feed Hygiene, Faculty of Veterinary Medicine, University of Warmia and Mazury in Olsztyn, Poland, on 18 clinically healthy, two-month-old pre-pubertal gilts. The mean initial body weight was 30 ± 2 kg. The pre-pubertal gilts were housed in individual cages with ad libitum access to water. Prior to the experiment, a cannula was placed in the cranial vena cava to minimize stress during blood sampling. The animals were administered with standard feed which was tested for the following mycotoxins: Aflatoxin, ochratoxin, ZEN, α-ZEL, and deoxynivalenol. Mycotoxin levels in the feed were estimated by common separation techniques with the use of immunoaffinity columns and HPLC (Hewlett Packard, type 1050 and 1100, Santa Clara, CA, USA) with fluorescent and/or UV detection techniques [34]. The obtained values were below the sensitivity of the tests.

Experimental Design
The animals were divided into two experimental groups and a control group of six individuals each. Group EI gilts were orally administered 20 µg ZEN/kg BW (200% more than the highest NOAEL value) for 48 days, group EII gilts were orally administered 40 µg ZEN/kg BW (400% more than the highest NOAEL value) per os for 48 days, and group C animals were orally administered placebo for 48 days ( Table 1). The placebo was a gelatine capsule filled with a control medium onto which ZEN was applied in experimental groups. Table 1. Daily feed intake in a restricted feeding regimen (kg/day) and mean dietary concentrations of ZEN per kg of feed (µg/kg feed).

Reagents
Sample weight of ZEN (Z-0167, Sigma Chemical Co., Steinheim, Germany) were administered per os daily in gelatine capsules before the morning feeding according to the weight of gilts. ZEN samples were diluted in 300 µL 96% ethyl alcohol (96% ethyl alcohol, SWW 2442-90, Polskie Odczynniki Chemiczne SA, Gliwice, Poland) to produce ZEN doses of 20 and 40 µg/kg BW. The resulting solutions were added to the feed, placed in gelatine capsules, and stored at room temperature for 12 h to evaporate the solvent.

Blood Sampling to Mycotoxicological Analyses
Blood was sampled on the first day of the experiment after the administration of ZEN, and ZON levels (average total ZEN and α-ZEL concentrations) were estimated. Blood samples for determinations of ZEN and α-ZEL concentrations were collected every 48 h, 1 h after mycotoxin administration. The samples were directly transferred to chilled centrifuge tubes containing heparin and were centrifuged at 3000 rpm for 20 min at the temperature of 4 • C. The resulting plasma was transferred to 3 mL polypropylene Eppendorf tubes 3810X, it was frozen and stored at −20 • C until analysis. Blood loss was compensated through the administration of multi-electrolyte solutions in amounts corresponding to the volume of collected samples.

Analysis of ZEN and Its Metabolite in Blood Plasma
The presence of ZEN (see Figure 1) and α-ZEL (see Figure 1) in the blood plasma was determined by various separation methods with the use of immunoaffinity columns (Zearala-Test TM Zearalenone Testing System, G1012, VICAM, Watertown, MA, USA) and HPLC with fluorescent detection.

HPLC Analysis
The presence of ZEN and α-ZEL in blood plasma was determined using immunoaffinity columns Mycotoxin concentrations were determined according to the external standard and were expressed in ppb (ng/mL). The limits of detection (LODs) for individual mycotoxins were determined as the concentrations at which the signal-to-noise ratio decreased to three. The LOQ was estimated as the triple LOD value. Quantification limits (LOQs) were determined at 0.100 ng/mL for ZEN and α-ZEL. The correlation coefficient (r) for the calibration curve was 0.9996 and 0.9989 for ZEN and α-ZEL, respectively.

Determination of Mycotoxin Bioavailability in Blood Plasma
The data were recorded and integrated with the use of the Computer Integrator POL-LAB application and CHROMAX for Windows v. 2000 software for processing chromatographic data (Pol-Lab Artur Dzieniszewski, Warszawa, Poland). Mycotoxin concentrations were determined according to the external standard method and given in terms of ng/mL.

Determination of Hormone (Estradiol and Testosterone) Concentrations in Blood Plasma
Blood was sampled every 48 h (25 times), 1 h after mycotoxin administration. The samples were transferred to chilled centrifuge tubes and were centrifuged at 3000 rpm for 20 min at the temperature of 4 • C. The resulting plasma was transferred to plastic polypropylene Eppendorf tubes, it was frozen and stored at −18 • C until analysis of selected steroid hormone levels.
Estradiol (see Figure 3) and T (see Figure 4) concentrations in the blood plasma were estimated by RIA procedures described for E 2 by Hotchkiss et al. [43] and for T-by Kotwica and Williams [44]. Estradiol and T antibodies (obtained from the Institute of Animal Physiology, University of Warmia and Mazury in Olsztyn, Poland) were characterized by Szafrańska et al. [45]. The sensitivity of E 2 and T assays was 5 pg and 2.5 ng (per tube), respectively. Intra-assay and inter-assay CVs for E 2 and T were determined at 6.9% and 11.8% and 7.6% and 12.5%, respectively.

Bioavailability Factor
The daily dose of ZEN (20 or 40 µg/kg BW) was administered to each animal individually, which corresponds to exposure to 928.74-1856.8 µg ZEN/kg of a complete diet, depending on daily feed intake ( Table 1).
The bioavailability factor (see Figure 5) was calculated as the ratio of ZEN + α-ZEL = ZOL bioavailability in peripheral blood (expressed as ppb) to ZEN concentration in the diet (expressed as ppb), expressed as the overall dose administered in each week of the experiment, according to a previously described method.

Statistical Analysis
Plasma concentrations of ZEN (see Figure 1), α-ZEL (see Figure 1), and ZON (see Figure 2) as well as E 2 (see Figure 3) and T (see Figure 4) levels in the blood plasma of the examined gilts were presented as arithmetic means (x) and SD (±). Statistical calculations were performed using the Statistica application (Statsoft Inc., Tulsa, OK, USA). Due to the applied ZEN dose and the length of application, arithmetic means were compared by one-way analysis of variance for systems with repeatable measurements. The homogeneity of variances in the compared groups (required in analyses of variance) was verified by the Brown-Forsythe test. Differences between groups were analyzed by Tukey's honestly significant difference test (p < 0.05 or p < 0.01).