Short-term consumption of the mycotoxin zearalenone by pubertal gilts causes persistent changes in the histoarchitecture of reproductive tissues

Abstract

Consumption of zearalenone (ZEN) detrimentally affects tissues and systems throughout the body, and these deleterious effects are especially pronounced in swine. The objectives of this project were to determine the effects of short-term consumption of ZEN (at concentrations that could be found on-farm) on growth, carcass weight, liver weight, and reproductive tissues of pubertal gilts, and to determine if the effects are transient or persistent. Cross-bred gilts (107.25 ± 2.69 kg) were randomly assigned to one of three feed treatments: 1) solvent only for 21 d (CON; n = 10), 2) ZEN for 7 d followed by 14 d of solvent (ZEN-7; 6 mg/d; n = 10), and 3) ZEN for 21 d (ZEN-21; 6 mg/d; n = 10). Body weights were collected at the beginning and end of the experiment (189.1 ± 0.8 and 211.1 ± 0.8 d of age, respectively). Carcass weights and tissues were collected at harvest. There were no treatment-based differences in growth, carcass, liver, or reproductive tissue weights. Histological analyses revealed differences based on treatment and the interaction between treatment and luteal status. The thickness of the ampullary muscularis declined with ZEN exposure (P < 0.05), while the isthmic epithelial cell height (P < 0.01) and uterine endometrial thickness (P < 0.02) increased. Interestingly, the thickness of the isthmic muscularis, uterine myometrium, and epithelial cell height only differed in the presence of a corpus luteum. Uterine epithelial cell height in the luteal phase was lowest in ZEN-7 pigs (P < 0.01). The isthmic muscularis in the luteal phase was thinner in pigs from both ZEN treatments (P < 0.01). Conversely, the luteal-stage myometrium was thicker in pigs from both ZEN treatments (P < 0.01). The discovery of these tissue-based differences during the luteal phase is particularly concerning since this corresponds with the time when embryos would be affected by the functional competency of the oviduct and uterus. The results of this work demonstrate that short-term consumption of ZEN produces microscopic, but not macroscopic alterations in reproductive organs which are likely to have negative effects on their subsequent function and that these differences persist even after ZEN consumption ceases. Taken together, these results indicate that it is insufficient to rely solely on outwardly visible symptoms as indicators of zearalenone exposure, as detrimental effects on reproductive tissues were found in the absence of phenotypic and morphologic changes.

Gilts consuming zearalenone-treated feed exhibited histological changes in their reproductive tissues, even in the absence of gross morphological differences. This observation suggests that reproductive performance is affected by zearalenone consumption at concentrations that do not elicit outwardly visible symptoms.

Introduction

The mycotoxin zearalenone (ZEN) is produced by fungi in the genus Fusarium (Gajęcki et al., 2009), and can develop during any stage of livestock feed production or storage (Kabak et al., 2006; Bryden, 2012; Leslie and Logrieco, 2014). While the FDA has not established a limit for ZEN contamination of livestock feed components (Liu and Applegate, 2020), the Joint FAO/WHO Expert Committee on Food Additives used a safety factor of approximately 100 to establish a limit for provisional maximum tolerable intake of 0.5 µg/kg BW. This value was based upon a 40 µg/kg BW no-effect level (15-d study in pigs) and 200 µg/kg BW lowest-observed-effect level (Joint FAO/WHO Expert Committee on Food Additives, 2000; Gajecki, 2002).

A recent survey of swine feeds and feedstuffs in the United States found that nearly half of the analyzed samples contained ZEN (Pack et al., 2021). Zearalenone is problematic for a number of reasons, but is probably best known as a potent phytoestrogen (Chang et al., 1979; Etienne and Jemmali, 1982; Farnworth and Trenholm, 1983; Gajęcki et al., 2009; Kanora and Maes, 2010). Its structural similarity to estrogen allows it to competitively bind to estrogen receptors, thus disrupting reproductive function (Gajęcki et al., 2009). As a result, consumption of ZEN has been associated with altered estrous cyclicity, reduced embryonic and fetal development, pseudopregnancy, and pelvic organ prolapse (Etienne and Jemmali, 1982; Edwards et al., 1987a, 1987b; Gajęcka et al., 2011). These consequences of ZEN consumption contribute to reproductive failure, which is a significant source of profit loss on swine farms (Arango et al., 2005), thereby impacting the productivity and profitability of swine operations.

Beyond its effects on reproduction, ZEN detrimentally affects a multitude of other organs and systems throughout the body. Zearalenone consumption induces stress and causes inflammation and/or deterioration of the liver and kidneys (Jiang et al., 2010). Hepatic lesions, structural abnormalities, and cellular damage are commonly reported (Skiepko et al., 2020; Dolenšek et al., 2021; Papatsiros et al., 2021; Zhang et al., 2021). It is also known to affect hematocrit, white blood cell count, hemoglobin, and platelet count (Maaroufi et al., 1996; Jiang et al., 2010). Within the cell, ZEN exposure adversely affects mitochondria, lysosomes, endoplasmic reticula, Golgi bodies, DNA synthesis, protein synthesis, and protein degradation (Kouadio et al., 2005; Wang et al., 2022). Furthermore, in long-term studies involving the consumption of high doses of ZEN, rodents developed hepatic and pituitary adenomas and carcinomas (National Toxicology Program, 1982). These are just some of the numerous whole body and tissue-specific effects of ZEN. In addition to the consequences reported in the literature, there are likely a myriad of other effects that have yet to be elucidated.

Many of the known detrimental actions of ZEN exposure are microscopic or molecular in nature, and therefore, not immediately apparent in production settings. For this reason, it is important to learn as much as possible about the effects of ZEN in controlled research studies using concentrations that could be experienced on-farm. In order to broaden the current understanding of the detrimental effects of ZEN, the objectives of this study were 2-fold: 1) to monitor whole body responses to ZEN consumption, with an emphasis on the morphology and histology of gilt reproductive tissues following short-term exposure and 2) to determine whether the effects of ZEN are transient or persistent following cessation of consumption.

Materials and Methods

Animals and treatments

All animal procedures were approved by the Virginia Tech Institutional Care and Use Committee (IACUC). Thirty cross-bred pubertal gilts (Yorkshire × Landrace cross; 107.25 ± 2.69 kg; 189.1 ± 0.8 d of age) were selected for two replicate experiments, during which they were housed individually at the Virginia Tech Swine Center. Animals were randomly assigned to have one of three treatments added to their daily feed ration: 1) solvent only for 21 d (CON; n = 10), 2) ZEN for 7 d followed by 14 d of solvent (ZEN-7; n = 10), and 3) ZEN for 21 d (ZEN-21; n = 10). See Table 1 for BW and ages of the gilts at the beginning of the experiment. Commercially available swine grower feed (Big Spring Mill, Elliston, VA) was the base feed to which all treatments were added.

Table 1.

Measurements of body, carcass, and liver characteristics of gilts consuming zearalenone (ZEN)

Component	Control1	ZEN-71	ZEN-211	P-value
Initial BW, kg	104.29 ± 5.33	109.73 ± 5.44	107.77 ± 4.78	0.78
Initial age, d	189.0 ± 1.4	190.8 ± 1.5	187.9 ± 1.3	0.38
Final BW, kg	115.29 ± 5.70	112.66 ± 5.33	115.56 ± 6.74	0.92
Change in BW, kg	7.98 ± 2.18	5.37 ± 2.07	10.33 ± 2.14	0.31
Carcass wt., kg	86.69 ± 3.45	83.53 ± 3.69	85.91 ± 3.27	0.81
Dressing, %	73.88 ± 0.76	73.86 ± 0.69	72.88 ± 0.87	0.63
Liver wt., kg	1.72 ± 0.08	1.54 ± 0.08	1.73 ± 0.07	0.20
Liver % of BW	1.62 ± 0.08	1.40 ± 0.07	1.45 ± 0.10	0.13
Liver % of carcass	2.09 ± 0.10	1.86 ± 0.11	2.01 ± 0.09	0.31

¹Control, solvent only for 21 d (CON; n = 9); ZEN for 7 d followed by 14 d of solvent (ZEN-7; n = 8); ZEN for 21 d (ZEN-21; n = 10).

Description of treatment and feed preparation is provided in detail in previous work (Pack et al., 2020a, 2020b). Briefly, the ZEN solution was prepared by dissolving crystalline ZEN (≥98% purity, J&K Scientific, Beijing, China) in acetonitrile (Thermo-Fisher, Waltham, MA, USA) to produce a working solution of 10 mg/mL. Individual doses were prepared by adding the working solution to small quantities of feed (600 μL working solution to 20 g of feed) and placing it under a fume hood overnight to allow the solvent to evaporate. Solvent-only feed was prepared in a similar manner using equivalent amounts of acetonitrile. In total, gilts receiving ZEN treatment were fed 6 mg of ZEN per d. This dose was selected based on literature that reported estrogenic symptoms in pubertal gilts similar to what might be observed on-farm (3 to 10 μg/g ZEN; Ensley and Radke, 2019). Zearalenone concentrations in the solvent-only feed (CON) were below the limits of detection (<0.01 µg ZEN; Pack et al., 2020a).

In order to ensure consumption of the entire treatment (solvent or solvent + ZEN), the daily ration was offered in a stepwise manner. At feeding, gilts were initially offered 227 g of feed containing their assigned treatment (solvent or solvent + ZEN). After the animals had consumed the treated feed, the remainder of their daily ration was offered (2.04 kg). The gilts were observed throughout the feeding process to ensure the entirety of the treatment and daily ration were consumed. Animals were examined daily for symptoms typically associated with ZEN exposure including decreased feed intake, vulva swelling, and vulva reddening.

Tissue collection

At the end of the 21-d experiment, the gilts were harvested at the Virginia Tech Meat Center. Reproductive tracts and livers were collected. Precise stage of the estrous cycle at the time of harvest was not known, but ovaries were examined to determine whether gilts were luteal or non-luteal at the time of sacrifice. Liver weight, overall tract weight, and individual tissue weights, lengths, and widths were measured and recorded immediately after harvest.

Histology and morphometry

After the uterine and ovarian tissues were weighed and measured, samples from the left and right uterine horns, as well as the left and right oviduct were dissected and fixed for 24 h in 10% formalin. The preserved tissues were then trimmed, transferred to 70% ethanol, and shipped to Histo-Scientific Research Laboratories, Inc. (Mount Jackson, VA) for paraffin embedding, sectioning (5µm sections), and staining with hematoxylin and eosin (H&E). Ultimately, two slides were made for each reproductive structure of each pig, one from the left and one from the right (e.g., one from left uterine horn and one from right uterine horn). An EVOS xl Core Imaging System (Life Technologies Corporation, Carlsbad, CA) was used to capture images from multiple fields of view on each slide of each tissue from each pig. Each image represented one field of view. The number of images collected varied based on the size of the tissue or the specific structure being evaluated. Tissue-specific elements were measured using ImageJ Software (National Institutes of Health, Bethesda, MD) and representative images are presented in Figure 1. For uterine tissue, epithelial cell height (6 images × 10 measurements per image; 400×), endometrial thickness (3 images × 10 measurements per image; 400×), myometrial thickness (3 images × 10 measurements per image; 400×), and uterine gland density (uterine glands per image, 6 images; 200×) were measured and recorded. Measurements of the oviductal tissue included epithelial cell height (6 images × 10 measurements per image; 400×), submucosal thickness (1 image × 15 measurements per image; 100× for isthmus; 200× for ampulla), and thickness of the muscularis (1 image × 15 measurements per image; 100× for ampulla; 200× for isthmus).

Figure 1.

Examples of histological measurements conducted in oviductal and uterine tissues of pubertal gilts. Cross-sections were H&E stained, and measurements were taken with ImageJ software. Oviduct epithelial cell height (A), isthmus submucosal thickness (B), ampulla submucosal thickness (C), isthmus muscularis thickness (D), and ampulla muscularis thickness (E) measurements were recorded. Uterine epithelial cell height (F), gland density number (G), endometrial thickness (H), and myometrial thickness (I) measurements were recorded. Lines or dots indicate points of measurement or count. Scale bar = 1,000 µm (C, D, H, I), 200 µm (B, E, G), or 100 µm (A, F).

Statistical analyses

Three gilts were removed from the experiment for reasons unrelated to the treatments (n = 1 from CON and n = 2 from ZEN-7). Data were analyzed using the MIXED procedure of SAS (SAS Institute, Inc, Cary, NC). Independent variables were treatment (control, ZEN-7 or ZEN-21), luteal status (luteal or non-luteal), and their interaction. Experimental replicate and harvest groups were included in the model as covariates. When replicate, group, and/or luteal status were not significant, they were removed from the statistical model. The change in body weight was calculated by subtracting the initial body weight from the final body weight. Dressing percent was calculated by dividing carcass weight by the final body weight. Organ weights were analyzed in their raw form (absolute weight) as well as calculated as a percent of body weight and percent of carcass weight. Pig was included as the repeated variable. For each analysis, eight covariance structures were tested and the most appropriate was selected based on Akaike’s information criterion, Akaike’s information criterion with correction and Bayesian information criterion values. Results are reported as least squares means ± standard errors of the means. Statistical significance was declared at P ≤ 0.05 and tendencies to differ at 0.05 ≤ P ≤ 0.10.

Results and Discussion

While numerous studies involving ZEN consumption in swine have been published, many use doses much greater than would be typically found on-farm. Furthermore, few have endeavored to measure recovery following cessation of consumption. We hypothesized that even short-term exposure to moderate concentrations of ZEN would detrimentally affect the liver and female reproductive tract, but that these effects would be transient and resolve in 14 d or less.

In this study, the pigs assigned to ZEN treatment received 6 mg of ZEN per d in 2.27 kg of feed, thus equating to 2.64 ppm (2.64 mg/kg of feed; Pack et al., 2020a). While low in comparison to concentrations administered in some published studies (Farnworth and Trenholm, 1981; Etienne and Jemmali, 1982; Gao et al., 2022), this amount is at the high end of concentrations typically found in ingredients used for livestock feeds (Khatibi et al., 2014; Pack et al., 2021). Major disruptions in reproductive function such as abnormal estrous cycles and pelvic organ prolapse have been observed when gilts consumed feed containing as little as 1 ppm of ZEN (Ensley and Radke, 2019), while molecular and histological changes in the reproductive tract have been demonstrated at even lower doses (Song et al., 2021; Wan et al., 2022). In the current study, animals were examined each day throughout the experiment for symptoms associated with ZEN consumption, but no gross morphological changes such as vulva swelling or reddening were observed. Furthermore, no clinical abnormalities of tissues were found at the time of harvest and tissue collection.

Whole body and liver weights

At the beginning of the study, gilts were randomly assigned to treatments based on body weight, which ultimately did not differ between treatment groups (Table 1). Age of the gilts at treatment initiation also did not differ (Table 1). Final body weight, change in body weight over the 21 d of the experiment, carcass weight, and dressing percentage were also similar between treatment groups (Table 1). These results are consistent with the results of many previous studies. Although sometimes associated with reduced feed intake and average daily gain (Diekman and Green, 1992), consumption of ZEN-contaminated diets often has no impact on growth and body weight (Young and King, 1986; Jiang et al., 2009).

Following consumption, ZEN can be found in liver tissue (Pack et al., 2020b) where it affects liver structure and functional competence (Tiemann et al., 2006). Zearalenone alone, or in combination with other fusariotoxins induces hepatic cell apoptosis and necrosis, increases inflammatory infiltrates, causes dilation of hepatic sinusoids, and alters the amount of interlobular connective tissue (Smith et al., 2017; Skiepko et al., 2020; Dolenšek et al., 2021). In light of these previously demonstrated effects of ZEN consumption on liver structure and function, liver weights were collected at harvest in the current study and analyzed as absolute weight as well as a percentage of body weight and a percentage of carcass weight. Consistent with some previous studies (Wang et al., 2012; Denli et al., 2015), no treatment-based differences existed in any of the liver weight parameters (Table 1). The lack of differences in liver weights is not necessarily indicative of an absence of effects of ZEN in hepatic tissues, however. Liver weight is a gross measurement where changes induced by ZEN consumption could be masked by the balance of consequences that would increase liver weight (such as inflammation) vs. those consequences that would reduce liver weight (such as apoptosis and necrosis). Thus, future studies examining the consequence of ZEN consumption would benefit from additional measurements of liver structure and function.

Gross measurements of the reproductive organs

The combined and individual weights of the reproductive tissues as well as the length of the uterine horns are presented in Table 2. Similar to body and liver weight analyses, no significant differences were observed in the weight or length measurements of the reproductive tissues across treatment groups. While ZEN consumption does not always cause enlargement of the reproductive tract (Etienne and Jemmali, 1982), the results of this study conflict with studies showing increased edema, protein synthesis and cell proliferation in the tubal portions of the reproductive tract (Obremski et al., 2003; Zhou et al., 2018a). Furthermore, previous analyses of the tissues collected from these gilts found that ZEN and/or its metabolite, α-zearalenol, was present in measurable quantities within many of the reproductive tissues (Pack et al., 2020b). In previous studies where differences in reproductive tract or organ weight were reported, the direction of the change was toward heavier tissue weights in those pigs consuming ZEN (Wang et al., 2012; Denli et al., 2015; Song et al., 2021), consistent with the phytoestrogenic properties of ZEN.

Table 2.

Morphological characteristics of reproductive tissues of gilts consuming zearalenone (ZEN)

Component	Control1	ZEN-71	ZEN-211	P-value
Total tract wt., g	471.1 ± 80.5	598.8 ± 85.3	668.0 ± 76.3	0.22
Tract % of BW	0.36 ± 0.08	0.51 ± 0.07	0.58 ± 0.09	0.16
Tract % of carcass	0.54 ± 0.08	0.70 ± 0.09	0.76 ± 0.08	0.16
Ovary wt., g	13.3 ± 5.3	13.8 ± 5.6	21.0 ± 5.0	0.51
Ovary % of BW	0.01 ± 0.01	0.01 ± 0.01	0.02 ± 0.01	0.28
Ovary % of carcass	0.01 ± 0.01	0.02 ± 0.01	0.02 ± 0.01	0.41
Total UT2 wt., g	367.8 ± 69.1	472.5 ± 73.3	466.0 ± 65.5	0.50
UT % of BW	0.27 ± 0.06	0.41 ± 0.06	0.43 ± 0.07	0.18
UT % of carcass	0.42 ± 0.07	0.55 ± 0.08	0.52 ± 0.07	0.41
Cervix wt., g	57.8 ± 10.1	61.3 ± 10.7	82.0 ± 9.5	0.19
Cervix % of BW	0.05 ± 0.01	0.05 ± 0.01	0.08 ± 0.01	0.25
Cervix % of carcass	0.07 ± 0.01	0.07 ± 0.01	0.10 ± 0.01	0.17
UT horn length, cm	211.1 ± 33.3	213.4 ± 35.6	191.0 ± 31.5	0.87

¹Control, solvent only for 21 d (CON; n = 9); ZEN for 7 d followed by 14 d of solvent (ZEN-7; n = 8); ZEN for 21 d (ZEN-21; n = 10).

²UT, uterine.

Histology of the reproductive organs

While there were no gross differences in the reproductive tissues, the histological analyses revealed differences that were not initially apparent. Treatment-based differences were observed in both the oviduct and uterus. In some cases, those differences were independent of luteal status, while for others there was an interaction between the treatments and luteal status.

In the ampulla, the thicknesses of the epithelial and submucosal layers were similar between treatments (P = 0.94 and P = 0.49, respectively), while the thickness of the muscularis decreased with ZEN exposure. The ampullary muscularis was thickest in CON pigs, intermediate in ZEN-7 pigs, and lowest in ZEN-21 pigs (P < 0.05; Figure 2). Intermediate thickness in the ZEN-7 treatment group could indicate less severe consequences of the shorter duration of ZEN exposure or partial recovery since tissues were collected 14 d after the cessation of ZEN treatment. Within the isthmus of the oviduct, epithelial cell height was low in CON pigs and higher in ZEN-7 and ZEN-21 pigs (P < 0.05; Figure 2). The similarity in epithelial cell height between the ZEN-7 and ZEN-21 treatments suggests that the effects of ZEN consumption on this layer of the isthmus are long-lasting, as they persisted for 14 d after the termination of ZEN treatment. The submucosa of the isthmus did not differ between treatments (P = 0.52). The muscularis of the isthmus differed by treatment as it was greater in the CON pigs (1,094.2 ± 34.4) than ZEN-21 pigs (920.7 ± 32.8; P < 0.01), and also differed based on the interaction between treatment and luteal status. In the absence of corpora lutea, the thickness of the isthmic muscularis was similar between all treatments. It was only during the luteal phase that differences emerged. In the luteal phase, pigs that had consumed ZEN (ZEN-7 and ZEN-21) exhibited thinner muscularis layers in the isthmus than CON pigs (P < 0.05; Figure 3A). These luteal-phase-specific findings are noteworthy as there are fewer numbers of estrogen receptors in porcine oviductal cells in the presence of progesterone (Chen et al., 2013). Consequently, gilts have comparatively limited ability to respond to phytoestrogens during the luteal phase, yet this was the phase during which differences were detected, suggesting robust activation of the receptors that are present.

Figure 2.

Thickness of the ampullary muscularis, isthmic epithelium, and uterine endometrium of gilts assigned to control (CON), ZEN for 7 d followed by 14 d of solvent (ZEN-7), or ZEN for 21 d (ZEN-21). ^a,bP < 0.05.

Figure 3.

Thickness of the isthmic muscularis (A), uterine epithelium (B), and uterine myometrium (C) during the non-luteal and luteal phase of gilts assigned to control (CON; n = 4 non-luteal, n = 5 luteal), ZEN for 7 d followed by 14 d of solvent (ZEN-7; n = 2 non-luteal, n = 6 luteal), or ZEN for 21 d (ZEN-21; n = 3 non-luteal, n = 7 luteal). ^a,b,cP < 0.05.

All layers of the uterus were affected by the treatment and/or the interaction between the treatment and luteal status. Uterine gland density, however, was similar across all analyses (P = 0.14), which agrees with previously published work (Döll et al., 2004). The uterine endometrium was thicker in those pigs that had consumed ZEN (Figure 2), but was not affected by luteal status or the interaction. The ZEN-induced increase in thickness of the endometrium is consistent with previous reports (Gajęcka et al., 2012; Zhou et al., 2018b, 2019). There was no main effect of treatment on uterine epithelial cell height (P = 0.15). Uterine epithelial cell height following ZEN exposure has been previously measured in pregnant or prepubertal females with some reporting an increase in cell height (Wu et al., 2020) and others reporting no change (Long et al., 1992; Döll et al., 2004) as was demonstrated in the current study by the lack of main effect of treatment. Despite the absence of main effect of treatment, an interaction between treatment and luteal status was evident. For CON and ZEN-7 pigs, the uterine epithelial cell height decreased from non-luteal to luteal status. This difference is consistent with previously reported cycle-based fluctuations in uterine epithelial cell height (Walter and Bavdek, 1997; Kangawa et al., 2017). In contrast, the uterine epithelial cell height in ZEN-21 pigs was nearly identical for non-luteal and luteal stages (Figure 3B). The uterine myometrium differed in an inverse manner. The thickness of the myometrium was similar for non-luteal and luteal pigs in the CON and ZEN-7 treatment groups. Pigs consuming ZEN for 21 d had a thicker myometrium during the luteal phase (Figure 3C). Similar findings have been reported in pre-pubertal gilts exposed to less than half of the ZEN dose used in the current study (Zhou et al., 2019). However, much like the previously described findings in the isthmic muscularis, these changes were unexpected as they occurred in the presence of progesterone, when sensitivity to phytoestrogen exposure would likely be limited by estrogen receptor populations (Koziorowski et al., 1984). In uterine myometrium, there was also a tendency for a main effect of treatment (P = 0.06), with myometrial thickness being greater in ZEN-21 pigs than in CON pigs. This increase with ZEN exposure is similar to previous reports (López et al., 1988; Gajęcka et al., 2012; Zhou et al., 2018b, 2019).

To our knowledge, this is the first report of reproductive tract histology during both the non-luteal and luteal phase following exposure to a phytoestrogenic mycotoxin. In general, it appeared that ZEN consumption interrupted expected fluctuations in the respective tissues; appearing to either cause or block changes compared to what was observed in CON pigs. It is notable that where interactions between treatment and luteal status were detected, the differences were always found in tissues from the luteal phase. As previously mentioned, during the luteal phase, estrogen receptor populations are typically low in oviductal (Chen et al., 2013) and uterine cells (Koziorowski et al., 1984), thereby limiting the ability of these reproductive tissues to respond to the phytoestrogenic properties of ZEN. Regardless of tissue sensitivity, ZEN ­consumption caused changes in the histoarchitecture of the oviduct and uterus in the current study. These effects were likely a consequence of untimely receptor activation. Normally, the reduction in receptor numbers during the luteal phase would coincide with relatively low levels of estrogen production. In the presence of ZEN, however, estrogen receptors are aberrantly activated in the luteal phase, resulting in the histological differences in reproductive tissues observed herein.

Unfortunately, elucidation of the molecular mechanisms responsible for the ZEN-induced changes in the oviduct and uterus was beyond the scope of this study. Irrespective of the ontogenesis, histomorphological differences in reproductive tissues are problematic as variations are associated with subsequent fertility (Małopolska et al., 2021). The thinned muscularis of the isthmus during the luteal phase as well as the overall thinner muscularis of the ampulla is concerning ­considering the importance of these structures for luteal-phase events such as sperm and oocyte transport leading up to fertilization and subsequent transport of embryos toward the uterus. Likewise, the observed differences in the uterine layers (most of which were specific to the luteal phase) indicate detrimental changes in uterine competency and function that are essential to early embryo development and implantation (Wu et al., 2020). While ZEN-induced changes in the oviductal and uterine histoarchitecture were evident in the present study, additional work is needed to confirm the impact of ZEN on the functional capacity of these tissues.

Conclusions

The amount of ZEN consumed by gilts in the present study did not produce any phenotypic or morphologic changes in their external genitalia, reproductive organs, liver or overall growth. Closer examination of the reproductive tissues revealed changes in the histoarchitecture of portions of the tract in both groups of pigs consuming ZEN. Many of these changes were evident after only 7 d of ZEN exposure and then persisted through the 14-d recovery period, indicating that the effects of ZEN exposure can develop quickly and be long-lasting. Taken together, the results of this experiment are proof that ZEN consumption, even at levels below the threshold for phenotypic and morphologic symptoms, can affect reproductive tissues long-term. Ultimately, these findings underscore the insufficiency of visible symptoms as the sole indicators of harmful levels of ZEN contamination in swine feed.

Glossary

Abbreviations

CON

control

ZEN

zearalenone

ZEN-7

zearalenone for 7 d followed by 14 d of solvent

ZEN-21

zearalenone for 21 d

Conflict of Interest Statement

The authors declare no conflicts of interest associated with this research.