2.1.5. Iron Deposits

Prussian blue (PB) staining was used for the visualization of iron deposits. In control pigs, small amounts of iron deposits were observed in hepatocytes and Browicz–Kupffer cells in the form of very fine (pollen) granules. Iron-loaded cells are either isolated or grouped together without any lobular systematization. In connective tissues, positive staining for iron corresponds to deposition within fibrocytes or macrophages. In experimental animals, increased amounts of iron deposits were observed both in the liver lobules and perilobular connective tissues, especially after 6 weeks of ZEN and DON + ZEN administration (Figure 6, Table 4).


**Table 4.** Semiquantitative analysis of iron deposits.

For a description of the score, see Table 2.

**Figure 6.** (**A**,**B**) Iron deposits in the liver of pigs receiving ZEN + DON for six weeks. (**A**) Iron deposits in hepatocytes in the form of fine granules. (**B**) Iron deposits with variable sizes in perilobular connective tissue. Prussian blue staining.

Physiologically, iron deposits are present within hepatocytes as fine granules at the biliary pole of cells and are distributed throughout the lobule according to a decreasing gradient from periportal to centrilobular areas. Iron excess is associated with chronic liver diseases of various causes [46–48]. It is known that in states of persistent iron overload, the liver could be seriously affected [49], mainly because of the extremely reactive hydroxyl radical formed during Fe2<sup>+</sup> metabolism (by Fenton reaction) in lysosomes [50].

Previous studies on the effect of mycotoxins on pig liver did not consider the effect of ZEN on iron storage [7]. Based on the results obtained in our study, it can be stated that ZEN induced duration-dependent iron accumulation in hepatocytes, Browicz–Kupffer cells, and connective tissue cells. This indicated that disruption of iron metabolism in the liver worsened along with the duration of ZEN administration. One possible mechanism is the effect of ZEN on hepcidin, a liver-derived peptide hormone, which is a master regulator of iron metabolism [50]. Our results show a similar effect of food contaminated with DON + ZEN on iron accumulation in the liver, as in the case of ZEN, which suggested that DON had no effect on iron accumulation.

### *2.2. Ultrastructural Study*

### 2.2.1. Hepatocytes

The hepatocytes of control pigs (Figure 7A) with polygonal or rectangular cross sections were arranged in regular rows. Their nuclei were distinctly round and centrally located, with a predominance of euchromatin. Hepatocytes had abundant amounts of both rough and smooth endoplasmic reticulum (ER). Medium-length cisterns of rough ER were located close to the nucleus and around the mitochondria (Figure 7B). The smooth ER was in the form of branched tubules and vesicles (Figure 7B). Mitochondria with numerous cristae and an electron-dense matrix were distributed in the cytoplasm in clusters or individually (Figure 7A). Lysosomes were localized mainly in the bile pole of hepatocytes. The content of glycogen deposits varied significantly between individual hepatocytes. After the first week of experiment, glycogen deposits were rather sparse; however, they were more numerous after three and six weeks. Lipid droplets were few and randomly distributed in the cytoplasm of hepatocytes. The vascular domain surface of hepatocytes was covered by numerous microvilli.

The differences in hepatocyte ultrastructure in experimental pigs compared with control animals were evident after 1, 3, and 6 weeks of treatment. The hepatocytes of ZEN-treated animals were characterized by a specific ER (Figure 7C,D). The smooth ER consisted of a network of closely located very narrow tubules and small vesicles, and occupied most of the cross sections of the hepatocytes. The poorly developed rough ER created several clusters of cisterns arranged in parallel. The mitochondria were located near these cisterns. Changes in the ER were observed from the first week of experiment; however, the intensity increased markedly after 3 and 6 weeks of treatment with ZEN. The ultrastructure of the mitochondria and lysosomes in ZEN-treated pigs did not differ significantly from that in the control animals. However, occasionally, some damaged mitochondria were observed. The amount of glycogen deposits in hepatocytes was usually low, except in the samples taken after the first week of the experiment. Some necrotic hepatocytes were noted, especially after 6 weeks of treatment.

In hepatocytes of DON-treated pigs, both types of ER were formed by abundant short, dilated cisterns and vesicles (Figure 7E,F), starting from the first week of the experiment. Both types of ER were quite evenly distributed in the cytoplasm of hepatocytes. In some cells, very dilated vacuoles of rough ER filled with protein micelles were observed. Large autophagosomes were present in the cytoplasm of numerous hepatocytes. Necrotic cells were occasionally observed.

The ultrastructure of hepatocytes in DON + ZEN-treated pigs did not differ from that in DON-treated pigs, except that necrotic hepatocytes were more frequently observed.

Changes in hepatocyte ultrastructure accompany numerous subclinical and clinical intoxications, both acute and chronic. The nature of these changes depends on the chemical properties of the toxin, dose, and duration of exposure. The most common changes are smooth and rough ER swelling, damage of mitochondria, and modification in the amount and distribution of lipid drops and glycogen deposits [51]. To date, data on the effect of mycotoxins on the ultrastructure of hepatocytes in pigs are scarce [7].

Our results show that the mycotoxins used in the experiment affected the ultrastructure of hepatocytes, especially the ER. ZEN administration resulted in conspicuous smooth ER proliferation and rough ER marginalization. Smooth ER is involved in the metabolism of various chemicals, and cells exposed to these chemicals showed hypertrophy of the smooth ER as an adaptive response. For this reason, smooth ER hypertrophy is considered a sensitive toxicological parameter [52].

The administration of DON and DON + ZEN resulted in the dilatation of ER cisterns and no clear difference between the two types of reticulum. The dilatation of ER cisternae, which indicates a loss of ER homeostasis, is known as ER stress [53]. ER stress is associated with liver injury and fibrosis. The hepatic factors that regulate ER stress remain unknown [54]. The fact that DON and DON + ZEN induced ER dilatation suggested that chronic ingestion of low doses of these mycotoxins caused injury in pig hepatocytes.

**Figure 7.** (**A**–**F**) Ultrastructure of hepatocytes in control pigs, 6th week of the experiment (**A**,**B**), pigs treated with ZEN for 6 weeks (**C**,**D**) and pigs treated with DON for 6 weeks (**E**,**F**). Note the specific organization of endoplasmic reticulum with the predominance of smooth reticulum over rough reticulum in pigs treated with ZEN. In DON-treated animals, both types of endoplasmic reticulum were formed by abundant short, dilated cisterns and vesicles. SD, space of Disse; BC, bile canaliculus; N, nucleus; M, mitochondria; RER, rough endoplasmic reticulum; SER, smooth endoplasmic reticulum; G, glycogen particles.

### 2.2.2. Sinusoids and Perivascular Species

Hepatic sinusoids were formed by flat endothelial cells with prominent fenestrations, usually lacking the basal lamina. Browicz–Kupffer cells with electron-lucent cytoplasm and numerous granules with variable appearance were located inside the sinusoids. Ito cells situated outside the vessels were numerous and usually contained one lipid droplet of moderate size. Pit cells situated inside the vessels were sporadically observed.

There were no differences in the ultrastructure of endothelial cells between control pigs and animals receiving mycotoxins. Browicz–Kupffer cells appeared to be more numerous in pigs receiving ZEN for 1 week as well as in pigs treated with DON and DON + ZEN for 1, 3, or 6 weeks than those in the control animals (Figure 8A–C). In these animals, Browicz–Kupffer cells showed prominent cell processes and occupied a large part of the vessel lumen. Their cytoplasm contained numerous phagosomes and rest bodies. Pit cells were more frequently observed in pigs treated with DON and DON + ZEN for 3 and 6 weeks than in the control animals (Figure 8C). In pigs treated with DON and DON + ZEN for 1, 3, or 6 weeks Ito cells were characterized by the presence of prominent cisterns of the rough ER (Figure 8D).

The focal accumulation of collagen fibers was observed in the perivascular spaces of pigs receiving DON and DON + ZEN for 3 and 6 weeks (Figure 9A). Their number increased with the duration of treatment. The penetration of collagen fibers between hepatocytes and foci of fibrosis was noted in pigs receiving DON + ZEN for 3 and 6 weeks (Figure 9B).

Our data shows, for the first time, that the administration of mycotoxins, especially DON, induced changes in the population of Browicz–Kupffer cells, which suggested their activation. These cells play an important role in the clearance of toxins from the portal blood [55]. Several cytokines, chemokines, and reactive nitrogen and oxygen species are released by activated Browicz–Kupffer cells, allowing these cells to modulate microvascular responses and the functions of hepatocytes and Ito cells [55,56]. Browicz–Kupffer cells show large plasticity, adopting changes in local metabolic and immune environment. They can play a protective role through their tolerogenic phenotype in toxin-induced liver injury, but can also shift to a pathologically activated state and contribute to liver inflammation. It could be considered that the changes in sinus diameters observed in our histological studies were due to the activation of Browicz–Kupffer cells. Because of their location and shape, these cells can interact with blood flow. Activation of Browicz–Kupffer cells is probably responsible for the more frequent occurrence of pit cells in liver sinuses [55,57].

Ito cells are "quiescent" in the normal liver, having received no stimuli to transform into a myofibroblastic state [56]. In our study, treatment with DON and DON + ZEN for 1, 3, or 6 weeks resulted in the presence of Ito cells with prominent cisterns of rough ER. The changes in Ito cells were probably related to their activation and transformation. They correlated with the occurrence of collagen fibers in the spaces of Disse.

The obtained data show that the administration of low doses of DON and ZEN resulted in prominent changes in the ultrastructure and histology of pig liver. In view of our results, even low levels of these mycotoxins, being below or close to No Observed Adverse Effect Level (NOAEL) values, should be considered as affecting the pig liver. The published data on the effects of such doses of DON and ZEN on the liver biochemistry are limited; however, they showed rather weak effects of these toxins on the serum activity of hepatic enzymes [58–60]. These findings agree with our morphological data showing that the responses of hepatocytes to intoxication with ZEN and DON seem to be adaptive. However, the changes in ER caused by ZEN and DON can affect the response of hepatocytes to other stressors or toxins [7,51]. In our opinion, special attention should be paid to the influence of ZEN and DON of the perilobular connective tissue, Browicz–Kupffer cells and Ito cells, which may lead to chronic damage of the liver. These effects cannot be detected with routine laboratory diagnostic methods.

**Figure 8.** (**A**,**B**) Browicz–Kupffer cells (BK) in the sinusoid of pigs treated with DON for 3 weeks (**A**) and 6 weeks (**B**). Note the presence of numerous processes in Figure (**A**) and lysosomes and rest bodies in Figure (**B**,**C**) BK and pit (P) cells in the sinusoid of a pig treated with DON + ZEN for 6 weeks. (**D**) Ito cell (It) in the perivascular space of pigs treated with DON + ZEN for 6 weeks. Note the cisterns of the rough endoplasmic reticulum.

**Figure 9.** (**A**) Accumulation of collagen fiber in the space of Disse in a pig receiving DON for 6 weeks. (**B**) Penetration of collagen fibers between hepatocytes in a pig treated with DON + ZEN for 6 weeks.

### **3. Conclusions**

Our current data provided strong evidence that the administration of low doses of DON and ZEN affected the ultrastructure and histology of pig liver. The changes caused by mycotoxins varied depending on the toxin and duration of intoxication. Both mycotoxins induced prominent modifications in the hepatocyte ER. The hepatocytes of ZEN-treated animals were characterized by extremely well-developed smooth ER, which comprised a dense network of narrow tubules. In the hepatocytes of DON-treated pigs, both rough and smooth ER were formed by an abundance of very short, dilated cisterns and vesicles. The effect of DON was greater than that of ZEN, as the ultrastructure of hepatocytes in pigs treated with DON + ZEN did not differ from that in DON-treated pigs. The total HAI scores in pigs receiving DON and DON + ZEN were significantly higher than that in the control animals, mainly because of the greater degree of hepatocyte necrosis and focal infiltration of inflammatory cells. Treatment with DON significantly increased the thickness of the perilobular connective tissue. The focal accumulation of collagen fibers was also observed in the perivascular spaces around the sinusoids, as shown by electron microscopy. These changes pointed to fibrosis as a potential effect of DON intoxication. Browicz–Kupffer cells were more frequently found and formed more prominent processes in pigs receiving mycotoxins, especially DON, and this phenomenon could be responsible for the dilation of sinusoids. Pit cells were more frequently observed in pigs treated with DON and DON + ZEN than in control animals. Examinations of glycogen and deposits of iron show the minor effect of the examined mycotoxins on these features. Our data demonstrate that DON and ZEN act both on hepatocytes and on liver immune/connective tissue cells. The response of hepatocytes to low doses of these toxins seems to be adaptive, and it is not related to serious cell damage. The influence of toxins on immune processes in the liver requires special attention because it can be destructive to the liver. Further molecular and biochemical studies are necessary to clarify the mechanisms of ZEN and DON toxicity in the liver.

### **4. Material and Methods**

### *4.1. Animals, Toxins, and Experimental Design*

The study was performed on 36 clinically healthy gilts of mixed breed (White Polish Big x Polish White Earhanging), with body weights of 25 ± 2 kg at the beginning of the experiment. The animals were purchased from a farm where they received feed without detectable amounts of ZEN, DON, α-zearalenol, aflatoxin, and ochratoxin. The pigs were fed twice daily and had free access to water.

The animals were divided into three experimental groups (D, Z, and M; *n* = 9 in each group) and a control group (C; *n* = 9). Group D received DON at a dose of 12 μg/kg BW per day, group Z received ZEN at a dose of 40 μg/kg BW per day, and group M received a mixture of DON + ZEN (ZEN 40 μg/kg BW + DON 12 μg/kg BW per day). The mycotoxins were synthesized and standardized at the Department of Chemistry, Faculty of Wood Technology, Pozna ´n University of Life Sciences, Poland. The mycotoxins were administered orally during morning feeding in water-soluble capsules containing oat bran as a vehicle. The gilts were weighed every week to establish the amount of DON and ZEN administered to each animal. The animals in group C received capsules without mycotoxins.

The animals from the control and experimental groups were killed by intravenous administration of sodium pentobarbital (Vetbutal, Biowet, Poland) at a dose of 140–150 mg/kg and by exsanguination after 1, 3, and 6 weeks of experiment. Tissue samples were taken no longer than 3 min after cardiac arrest.

All procedures were carried out in compliance with Polish legal regulations for the determination of the terms and methods for performing experiments on animals and with the European Community Directive for the ethical use of experimental animals. The protocol was approved by the Local Ethical Council in Olsztyn (opinion No. 88/N of 16 December, 2009).

### *4.2. Histological Examination*

Tissue samples (approximately 1 × 0.5 cm) were cut from the middle part of the liver. They were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 48 h, dehydrated in ethanol (TP 1020; Leica, Wetzlar, Germany), and embedded in paraffin (EG1150; Wetzlar, Leica). Next, 4-μm-thick sections were prepared using an HM 340E microtome (Microm, Lugo, Spain) and stained with hematoxylin and eosin (HE), Mallory's trichrome, PAS, and PB to detect iron using an automated multistainer ST 5020 (Leica, Wetzlar, Germany). For histological evaluation, the sections were scanned using a Mirax Desk scanner (Carl Zeiss, Oberkochen, Germany). The slides were signed in a way that prevented the people involved in the microscopic analysis from knowing the kind and duration of the animal treatment.

We evaluated the architectural changes in the liver, noting the organization of hepatic cords, the appearance of perilobular connective tissue, and the penetration of collagen into the parenchyma on Mallory- and HE-stained specimens.

Histological changes in the liver parenchyma were analyzed on HE-stained slides. The following parameters were evaluated:


For morphometrical evaluations, the following parameters were determined:


Measurements were performed on HE-stained sections using the Pannoramic Viewer 1.15 software (3D-Histech, Budapest, Hungary).

Glycogen content and iron accumulation in hepatocytes was evaluated in PAS- and PB-stained sections, respectively.

### *4.3. Ultrastructural Examination*

Samples of liver tissue were collected from sites adjacent to the sampling sites for histological examination, and then immersion-fixed in a mixture of 1% paraformaldehyde and 2.5% glutaraldehyde in 0.2 M phosphate buffer (pH 7.4) for 2 h at 4 ◦C. Next, they were washed and post-fixed in 2% OsO4 in 0.2 M phosphate buffer (pH 7.4) for 2 h. After dehydration, the samples were embedded in Epon 812. Semi-thin sections were cut from each block of tissue, stained with 1% toluidine blue, and examined under a light microscope to choose the sites for preparing ultrathin sections. Ultrathin sections were cut using a Leica Ultracut III ultramicrotome (Leica, Wetzlar, Germany) and contrasted with uranyl acetate and lead citrate. They were examined with a Tecnai 12 Spirit G2 BioTwin transmission electron microscope (FEI, Hillsboro, OR, USA) equipped with two digital cameras: Veleta (Olympus, Tokyo, Japan) and Eage 4k (FEI, Hillsboro, OR, USA).

### *4.4. Statistical Analysis*

The data from morphometric investigations were analyzed using one-way ANOVA with a Duncan test as a post-hoc procedure, and the data from semiquantitative analyses were analyzed using Kruskal–Wallis non-parametric ANOVA. Statistical analyses were performed using Statistica 10.0 software (StatSoft Polska, Cracow, Poland).

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2072-6651/12/7/463/s1. Figure S1. Microphotographs of the histopathological parameters taken into account in HAI A. Portal inflammation. B. Periportal inflammation C. Acinar inflammation. D. Confluent necrosis. E. Focal lytic necrosis F. Hemorrhage, Figure S2. Dilatation of hepatic sinusoids in zone III (periportal) of acinus in a pig receiving DON + ZEN for 3 weeks. HE staining.

**Author Contributions:** Conceptualization, M.G. (Maciej Gaj ˛ecki) and B.L.; methodology, M.G. (Maciej Gaj ˛ecki) and B.P.-G.; investigation, M.G. (Magdalena Gaj ˛ecka), B.P.-G., N.S., and B.L.; writing—original draft preparation, N.S., B.P-G.; writing—review and editing, B.L.; visualization, N.S.; supervision, B.L.; project administration, M.G. (Maciej Gaj ˛ecki) and B.L.; funding acquisition, M.G. (Maciej Gaj ˛ecki) All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by research grant No NR12-0080-10 from the Polish National Centre for Research and Development. Publication costs were covered by the Ministry of Science and Higher Education under the program entitled "Regional Initiative of Excellence for the years 2019-2022", Project No. 010/RID/2018/19; the amount of funding was 12.000.000 PLN.

**Acknowledgments:** The authors would like to thank Jacek Sztorc and Krystyna Targo ´nska for their skillful technical assistance.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
