Malekinejad et al. [9] demonstrated that in a hepatic biotrasformation pigs convert ZEA predominantly into α-ZOL. An extrahepatic biotransformation of ZEA into α-ZOL was reported in porcine granulosa cells by means of 3 α-HSD [10]. In the sow, it has been reported that the intestinal mucosa has a very active glucuronide conjugation (about 30 times) of ZEA compared with the levels of reduction. It seems likely that ZEA is conjugated before it is reduced to zearalenol [6]. The glucuronide of ZEA is excreted in bile to be re-absorbed and metabolized further by intestinal mucosal cells (manly in α-ZOL), ultimately entering the liver and the systemic circulation via the portal blood supply [8]. The entero-hepatic cycle has the effect of prolonging the retention of ZEA and its derivatives in the circulatory system, retarding elimination and enhancing the duration of adverse effects [8]. After a low-dosage feeding (192 μg/Kg b.w./day for 4 day) of prepubertal gilt, α-ZOL was detected in plasma one hour later at concentration twice (2 ng/mL) as high as ZEA and rose to 3-4 times (8 ng/mL) during the day of ZEA feeding [11]. The maximum blood levels of both mycotoxins (33,000 pg/L) are 500 times higher than the concentration of 17β-oestradiol (17βE₂) at oestrous but they are bound to glucuronic acid, which inactivates the compound and facilitates the excretion [11]. The urinary concentrations of ZEA and α-ZOL reach maximum level (158.9 ng/mL and 170.8 ng/mL, respectively) in the end of ZEA feeding and both compounds could be traced 4 days after [11]. Higher blood levels in gilt and sows (from 3 to 100 ng/mL), in wild boars (from 2.2 to 43 ng/mL) and foxes (from 5 to 145 ng/mL) were found in Poland by Gajecki [7] respect to concentrations (up to 0.96 ng/mL) in Romanian slaughtered pigs [12].

The ovine metabolism of ZEA includes the synthesis of five metabolites, such as ZAN, α- and β-ZOL, α- and β-ZAL and high levels of some of these forms may be excreted in the urine as glucuronides by grazing sheep [8]. As reported by Malekinejad et al. [9], the sheep liver post-mitochondrial fraction converts ZEA mainly into α-ZOL. The concentration of α- + β-ZOL, found by [13] in urine collected from pasture-fed animals, resulted higher (up to 86 ng/mL) than that of α- + β-ZAL (up to 2.1 ng/mL).

It has been reported that bovine hepatocytes and granulosa cells (GCs) produce predominantly β-ZOL [6, 9, 10] that was found as the predominant metabolite in urine and faeces from a cow fed with zearalenone. As reported by Erasmuson et al. [13], high amounts of α- + β-ZAL (up to 12.3 ng/mL) and α- + β-ZOL (up to 163 ng/mL) were found in urine samples (68%) of pasture-fed cattle.

The main excretion of ZEA in the horse come through faeces, followed by urine [14]. The main metabolites detected in the equine faeces are ZEA (73 ng/mL), α-ZOL (50 ng/mL) and β-ZOL (45 ng/mL). α-ZAL (27 ng/mL), β-ZAL (41 ng/mL) and ZAN (29 ng/mL) were found at low levels. Zearalenone urinary levels of 3 ± 2 (mean ± standard deviation) and 43 ± 57 ng/mL were found in Italian and Romanian horses, respectively, as natural exposure to ZEA [15]. Erasmuson et al.[13] reported very high levels of α - + β-ZOL (up to 19 ng/mL) and α -+ β-ZAL (2157 ng/mL) in urine samples probably related to the grazing practices for those species.

Zearalenone and its derivatives compete effectively with 17 β-E₂ for the specific binding sites of the oestrogen receptors (ERs) occurring in different organs. Two subtypes of ER exist, ER- alpha and ER-beta that are differently distributed in the body. Several investigations have demonstrated that binding of ZEA and its derivatives initiate a sequence of events known to follow estrogen stimulation of target organs [16]. In the uterus, the cytoplasmic receptor complex is translocated to the nucleus (with a time retention longer that E₂) and the following events, typifying early estrogen response, occur: an increase in uterine RNA synthesis as well as an increase in RNA polymerase activity, synthesis of uterine estrogen-induced protein [16]. The binding of ZEA to ER in target tissues is < 1–10% than that of E₂, whereas α-ZOL shows somewhat stronger binding and β-ZOL much less binding [17]. The relative binding affinities to the rat uterine cytoplasmic receptor for ZEA and derivatives are α-ZAL>α-ZOL> β-ZAL>ZEA>β-ZOL [17]. The competition of binding to ERs is demonstrated with in vitro systems. The proliferation of ER-positive (MCF-7) and ER-negative (MDA-MB-231) human breast cancer cells lines, which respond to physiological concentrations of E₂ (1nM) was used to characterize the oestrogenic activity of ZEA and derivatives through oestrogenic parameters such as proliferative effect (PE), relative proliferative effect (RPE) and relative proliferative potency (RPP) [4]. On MCF-7, α-ZOL induced a higher PE and RPP, probably ER α–mediated, (because of partial antagonism of mycotoxin-related proliferation by tamoxifen). The potency of ZEA and derivatives were lower (from 10⁻² to 10⁻³) than E₂ [4]. β-ZOL showed higher EC₅₀ value (5.2 × 10⁻³ μM) than those of other ZEA-related compounds [4]. ZEA and its derivatives rapidly enhance phasic spontaneous smooth contraction in myometrial strips of prepubertal lamb at nanomolar concentration like E₂. Both intracellular ER and plasma membrane receptors could be involved in this effect [18]. In fact, referring to myometrium, the existence of novel E₂ membrane binding proteins, structurally related to the intracellular ER, has been previously described in rabbit [19]. The involvement of non-genomic actions is observed especially for ZEA, which increases the uterine activity similarly to E₂. An antiestrogenic effect is found for α-ZOL, which can be based on the fact that α-ZOL blocks ER on cell membrane following no biochemical response [18].

The structure of the ZEA resembles many characteristics of steroids and binds to ER as an agonist. The ZEA conversion to α-ZOL and β-ZOL shows similarities to processes in steroid metabolism where HSDs play a pivotal role in synthesis of various steroid hormones including E₂ and testosterone. Consequently, ZEA is also a substrate for 3α-HSD and 3β-HSD present in many steroidogenic tissues, such as liver, kidney, testis, prostate, hypothalamus, pituitary, ovary, intestine [6]. As observed in Figure 2, the enzymes involved in the biosynthesis of the gonadal steroid hormones (such as progesterone, E₂ and testosterone) fall into two major classes of proteins: the cytochrome p450 heme-contaning proteins and the hydroxysteroid dehydrogenases (HSD) [20,21,22]. The first and rate limiting step in the synthesis of steroid hormones is the conversion of cholesterol to pregnenolone, catalysed by the enzyme CYP11A (cytochrome p450_scc) found in the ovary (theca interna and GCs) and testis (Leydig cells). The CYP17 is expressed exclusively in the Leydig and in thecal cells where there is the site of androgen production. Cytochrome p450 aromatase catalyses the conversion of androgen to estrogens in two different major pathways: first, the aromatase pathway (transformation of testosterone into E₂ and of androstenedione into estrone) and secondly, the 5α-reductase pathway (transformation of testosterone into dihydrotestosterone). The major sites in human and rats are in the preovulatory follicle, the corpus luteum, placenta and Leydig cells. 3α-HSD shows high affinity for 5α-dihydrotestosterone. The seven human isoenzymes of 17β-HSD play an important role in the biosynthesis of mainly androgens and oestrogens, testosterone and estradiol and their weaker precursor androstenedione and estrone which are expressed in GCs of developing follicles and in testes (Leydig cells). The 3 β-HSD catalyses the conversion of pregnenolone into progesterone. The isoforms of 3 β-HSD are expressed in a cell- and tissue-specific manner, particularly in ovary (thecal and GCs of growing and preovulatory follicles) and testis (Leydig cells).

The conversions of testosterone and progesterone by these tissues are important components of the mechanisms by which these hormones exert their effect on both gonadotropin regulation and sexual behavior. By competitively inhibiting the 3α-reduction of these hormones, ZEA could cause an accumulation of the active components, because 3α-HSD is an important factor in ovarian follicular development [6]. Zearalenone interference with HSDs may affect the reproduction, in addition to the effect via ER.

In vitro studies on mouse Leydig cell function demonstrated that ZEA and α-ZOL exposure (from 10⁻⁸ to 10⁻⁴ M) can interfere with the process of spermatogenesis reducing the hCG-stimulated testosterone synthesis owing to the down-regulation of P450scc, 3β-HSD-1 and steroidogenic acute regulatory protein (StAR) transcription [23].

The specific manifestations are dependent upon the relative dose of ZEA and the stage of estrous cycle or pregnancy during which ZEA was consumed. The oestrogenic syndrome in swine affects primarily the reproductive tract and mammary gland, and in more sensitive young gilts 1–5 ppm of ZEA induce clinical signs such as hyperemia, edematous swelling of the vulva, sometime vaginal prolapse and even rectal prolapse [24]. This hyperoestrogenic mycotoxicosis primarily affects weaned and prepubertal gilts because the lack of full development of the endogenous endocrine system in the young [24]. In immature gilts, dose of 200 μg/Kg b.w. leads to the development and maturation of the largest follicles through the activation of an apoptotic-like process in the granulosa layer of single mature follicle and ovarian follicle atresia [25, 26]. After recovery of oocytes during ovario-hysterectomy of gilts exposed to ZEA and DON by feeding (from 0.2 and 0.004 to 9.6 and 0.36 mg/Kg feed, respectively), oocytes with compact cumuli showed a significantly reduced proportion with immature chromatin in groups exposed to high mycotoxin levels [27]. The proportion of oocytes having degenerated meiotic chromatin was significantly higher in these groups compared with groups exposed to lower concentrations. The oocyte quality was significantly reduced and was directly related to the reduced proportion of normal germinal vesicle-stage oocytes [27]. The feeding of wheat contaminated naturally with DON and ZEA to gilts did not influence the activity of enzymes (P450scc and 3β-HSD) involved in progesterone synthesis, as observed in vivo-derived porcine granulosa cell cultures [27]. Oocyte degeneration and reduced meiotic competence of compact cumulus oocyte complexes were reported after in vitro maturation [27].

In cyclic animals, nymphomania, pseudopregnancy, ovarian atrophy and changes in the endometrium are reported [24]. Zearalenone causes sterility in sows by inciting a malfunction of the ovary [28]. The oocyte dies in the Graafian follicles and despite signs of oestrus, there is no ovulation. In gilts, after 100 ppm in feedstuff for one week of treatment, ZEA appears to cause maturation by stimulating primary and secondary follicles in the ovary and proliferation of uterine glands [28]. Zearalenone acts similarly to E₂ in inhibiting the release and secretion of follicle stimulating hormone (FSH), thus depressing the maturation of ovarian follicles during the preovulatory stage. The changes induced by ZEA depend on time of administration in relation to oestrous cycle as well as on the dose administered [24]. This indicates that ZEA may exhibit a luteotrophic property which prolongs the life span of the corpus luteum equal to or longer than the normal gestation period [24].

During pregnancy, the timing of the exposure is critical, because at threshold level of ZEA (1 mg/Kg b.w.), administered on days 7 to 10 after mating, does not disrupt pregnancy, when administered on days 2 to lead to degenerative changes in embryos that begin to be well advanced by day 13 after mating [29]. These effects are similar to those of other exogenous estrogens on pregnancy, such as E₂. During pregnancy, ZEA reduces embryonic survival, when administered a threshold level (200 μg/Kg b.w.), and sometimes decreases fetal weight [8]. Placental transfer of ZEA results in teratogenic effects in piglets characterized by various abnormalities of genitalia [8].

The effect of Fusarium cultures per os on male swine showed a 30% of weight reduction of testes, decrease of fertility related to reduction of sperm quality and viability [28]. In boar, ZEA depressed serum testosterone, inducing feminization and suppressing libido [8]. No adverse effect on the reproductive potential of mature boars was described at levels normally found in contaminated feed, but in female swine several severe reproductive problems can occur after feeding with ≤200 ZEA mg/Kg [30]. The α-ZOL may reduce aggressive behavior, testis growth and sexual activity in farmed fallow bucks by ear implants at a dose of 36 mg at 90 day intervals [31].

In cow, infertility, reduced milk production and hyperoestrogenism have been associated with ZEA [8]. Heifers were fed with ZEA containing diet over three estrous cycles, conception rates decline from 87 to 62% [8].

A field outbreak of ZEA mycotoxicosis in horses was associated with corn screenings containing approximately 2.6 mg/Kg of ZEA [34]. The concentration of the toxin recorded in naturally contaminated oats (ZEA and DON at 1 and 12 ppm levels, respectively) does not have a significant effect on the release of reproductive hormones, cycle length or uterine histology in mares [35]. Local ovarian effects of both mycotoxins on follicular growth cannot be excluded because the number of growing follicles during the second half of the cycle were increased [35]. Zearalenone, administered at low dose (7 mg) starting 10 days after ovulation, did not influence the interovulatory interval, the length of luteal and follicular phase, plasma progesterone concentration, the number of large follicles (>2cm) [36]. Outbreak of early fetal losses, abortions, stillbirths and neonatal foal deaths (collectively referred to as the Mare Reproductive Loss Syndrome MRLS) were reported in North-Central Kentucky during the spring of 2001. Several feed-related causes, such as infestation with caterpillar larvae and/or natural toxic compounds (cyanogens, ergot alkaloids and mycotoxins) were hypothesized, but MRLS was not linked to mycotoxins [37].