Bovine Peripheral Blood Mononuclear Cells Are More Sensitive to Deoxynivalenol Than Those Derived from Poultry and Swine

Deoxynivalenol (DON) is one of the most prevalent mycotoxins, contaminating cereals and cereal-derived products. Its derivative deepoxy-deoxynivalenol (DOM-1) is produced by certain bacteria, which either occur naturally or are supplemented in feed additive. DON-induced impairments in protein synthesis are particularly problematic for highly proliferating immune cells. This study provides the first comparison of the effects of DON and DOM-1 on the concanavalin A-induced proliferation of porcine, chicken, and bovine peripheral blood mononuclear cells (PBMCs). Therefore, isolated PBMCs were treated with DON (0.01–3.37 µM) and DOM-1 (1.39–357 µM) separately, and proliferation was measured using a bromodeoxyuridine (BrdU) assay. Although pigs are considered highly sensitive to DON, the present study revealed a substantially higher sensitivity of bovine (IC50 = 0.314 µM) PBMCs compared to chicken (IC50 = 0.691 µM) and porcine (IC50 = 0.693 µM) PBMCs. Analyses on the proliferation of bovine T-cell subsets showed that all major subsets, namely, CD4+, CD8β+, and γδ T cells, were affected to a similar extent. In contrast, DOM-1 did not affect bovine PBMCs, but reduced the proliferation of chicken and porcine PBMCs at the highest tested concentration (357 µM). Results confirm the necessity of feed additives containing DON-to-DOM-1-transforming bacteria and highlights species-specific differences in the DON sensitivity of immune cells.


Introduction
Mycotoxins are fungal secondary metabolites that can cause disease and ultimately death in both humans and animals. According to mycotoxin surveys, up to 72% of the world's agricultural commodities are contaminated by mycotoxins [1]. The trichothecene deoxynivalenol (DON), mainly produced by Fusarium graminearum and Fusarium culmorum, is one of the most prevalent

Chicken PBMCs
We next assessed the effect of DON and its metabolite DOM-1 on the proliferation of chicken PBMCs ( Figure 2). DON had no effect on proliferation at concentrations between 0.01 and 0.21 µM. A significant reduction in proliferation was observed between 0.42 (−28%, p = 0.007) and 3.37 µM (−83%, p = 0.000) DON. An IC50 of 0.691 µM was calculated for DON. While no effect was observed with 1.39 to 178 µM DOM-1, 357 µM significantly decreased the proliferation of chicken PBMCs by approximately 40% (p = 0.001). in the presence of ConA (2.5 µg/mL) ( ) for 28 h. Proliferation was measured via BrdU proliferation assays and calculated relative to the ConA control, which was set as 100%. Data from experiments with PBMCs isolated from eight (DON) and seven (DOM-1) different animals (mean + SD) are shown. Asterisks indicate significant differences compared to control (* p < 0.05, ** p < 0.01, *** p < 0.001).

Bovine PBMCs
As a next step, the proliferation of bovine PBMCs treated with DON or DOM-1 was assessed. The exposure of bovine PBMCs to DON led to a significant dose-dependent decrease in proliferation at 0.21-3.37 µM (Figure 3). At 3.37 µM DON, proliferation was reduced by a maximum of 86% (p < 0.001). An IC50 value of 0.314 µM was calculated for DON. DOM-1 did not significantly influence the proliferation of bovine PBMCs at 1.39-357 µM. in the presence of ConA (2.5 µg/mL) ( ) for 72 h. Proliferation was measured via BrdU proliferation assays and calculated relative to the ConA control, which was set as 100%. Data from experiments with PBMCs isolated from six different animals (mean + SD) are shown. Asterisks indicate significant differences compared to control (* p < 0.05, ** p < 0.01, *** p < 0.001).

Proliferation of Bovine T-Cell Subsets
Having identified the lowest IC50 value for DON in bovine PBMCs, we next investigated whether the predominant T-cell subsets in cattle are particularly sensitive to this mycotoxin. Hence, bovine PBMCs were labeled with a proliferation dye, and major T-cell subsets were identified by surface staining of CD4, CD8β, and T-cell receptor (TCR) γδ ( Figure 4). In these experiments, PBMCs were cultivated either in the presence of concanavalin A (ConA) alone or in combination with different DON concentrations (0.1, 0.2, 0.4, 0.8, and 1.6 µM). Lymphocyte blast cells were identified by light scatter properties and analyzed for proliferation (Figure 4a, top row), and further subgated for the expression of CD8β, CD4, and TCR-γδ. These subpopulations were again analyzed for proliferation ( Figure 4a

Discussion
The present in vitro study clearly demonstrates that porcine, chicken, and bovine PBMCs possess different sensitivities to deoxynivalenol (DON). Although cattle are considered particularly resistant to mycotoxins, bovine PBMCs were far more sensitive to the inhibitory effects of DON on their proliferation than chicken or porcine PBMCs. Furthermore, in agreement with several other studies [23][24][25][26][27][28], the toxic effect of DON was strongly attenuated through de-epoxidation to its metabolite deepoxy-deoxynivalenol (DOM-1).
We examined the in vitro effects of DON and DOM-1 on the proliferation of ConA-stimulated porcine, chicken, and bovine PBMCs. Plant-derived mitogens such as ConA, phytohemagglutinin (PHA), and pokeweed mitogen act differently on lymphocytes; for example, ConA and PHA stimulate T cells, while others, such as pokeweed, activate B and T cells [29,30]. We recorded changes in the proliferative capacity of cells by measuring DNA synthesis via the incorporation of the thymidine analog bromodeoxyuridine (BrdU) into newly synthesized DNA. PBMC proliferation is frequently quantified via the convenient MTT and lactate dehydrogenase (LDH) assays, though these provide nonspecific measurements of cell proliferation. The MTT response, for example, may vary considerably in viable cells depending on their metabolic state [31]. As demonstrated by Goyarts et al. [32] and Charoenpornsook et al. [33], DNA synthesis-based methods, such as BrdU or [ 3 H]-thymidine incorporation, provide a far more sensitive measurement of lymphocyte proliferation than LDH, MTT, or trypan blue exclusion methods.
Interestingly, the measurement of the highly sensitive BrdU incorporation showed that among porcine, chicken, and bovine PBMCs, bovine cells were the most sensitive to DON in terms of the inhibition of lymphocyte proliferation. DON significantly decreased bovine lymphocyte proliferation at 0.21 µM with an IC50 of 0.314 µM. In comparison, the proliferation of chicken-and porcine-derived PBMCs was significantly reduced at 0.42 µM (IC50 = 0.691 µM) and 0.84 µM (IC50 = 0.693 µM), respectively. These findings are in agreement with Charoenpornsook et al. [33], who reported a 50% inhibition of bovine PBMC proliferation at 0.24 µM, using the [ 3 H]-thymidine incorporation method. Dänicke et al. [22] investigated the proliferation of bovine PBMCs in response to DON via the MTT assay and reported an IC50 of 0.5 µM DON. According to the authors, the age of the cows had an influence on their susceptibility: PBMCs isolated from calves were more sensitive than those isolated from adult cows. Similarly, Wada et al. [34] reported a 25% proliferation reduction in PHA-stimulated adult cow PBMCs, compared to a 55% reduction in calf PBMCs, following treatment with 3.37 µM DON for 72 h. Therefore, in our study, PBMCs of fetal and young animals were excluded to avoid possible effects of age-derived sensitivities.
The proliferation dye carboxyfluorescein diacetate succinimidyl ester (CFDA-SE) and its derivatives such as the violet proliferation dye allow a simultaneous analysis of proliferation and cell phenotype [35], a method that is also well established for bovine T cells [36]. We made use of this approach to investigate potential differences in DON sensitivity between CD4 + , CD8β + , and γδ T cells in bovine-derived PBMCs. The results indicated that all three major T-cell subsets showed a similar impairment in proliferation when stimulated by ConA. These findings suggest that the inhibitory effect of DON on ConA-induced T-cell proliferation in all three T-cell subsets is based on the same mechanism.
The IC50 DON values obtained for bovine total lymphocytes and the derived T-cell subsets were somewhat higher than those obtained in BrdU assays (0.314 µM for BrdU and 0.5 to 0.6 µM for total lymphocytes and T-cell subsets). This discrepancy is probably attributable to the different methodologies: BrdU analysis addresses DNA synthesis, whereas proliferation dyes assess a loss of fluorescence intensity, resulting from the development of daughter cells that bear a reduced number of proliferation dye-labeled proteins in their cytoplasm. Notably, the IC50 values obtained in the violet proliferation assays for bovine PBMCs were still below the BrdU IC50 values for chicken and porcine lymphocytes.
The high DON sensitivity of bovine PBMCs is relevant [37] with respect to several aspects that are often insufficiently considered. It is frequently suggested that ruminants are resistant to mycotoxins due to the composition of their intestinal microbiota, which is capable of converting DON to its metabolite DOM-1 before it reaches the bloodstream. However, it is often left unconsidered that some mycotoxins found in silage and stored feedstuffs possess antimicrobial characteristics, through which they can modify the ruminal microbiota, thereby reducing its detoxification capacity. Consequently, contaminated digesta can reach the duodenum, where contaminants are absorbed into the bloodstream [38]. Moreover, high-concentrate diets, frequently used in high-performance cows, enhance the development of rumen acidosis. The latter can damage the barrier function of the rumen wall [13,39], a phenomenon associated with the translocation of rumen bacteria (e.g., Fusobacterium necrophorum) to the liver via the portal vein as well as the transport of harmful substances into the bloodstream [22,40,41]. Furthermore, He et al. [42] demonstrated that DON transformation is inhibited by pH values below 5.2 due to the inactivation of microorganisms in acidic conditions or by specific inhibitory effects on the de-epoxidation process. Consequently, not only DOM-1, which we report to have no effect on bovine PBMCs up to a concentration of 357 µM, but also intact DON as well as its acetylated derivatives may enter systemic circulation. The entrance of DON into the bloodstream can trigger nonspecific clinical symptoms such as metabolic and hormonal imbalances, inflammation, and immune responses. In serum, maximal values of 0.03 µM DON [22] and 0.06 µM DON [43] were found in cattle fed DON-contaminated feed. Interestingly, in one study assessing the serum of dairy cows, concentrations up to~0.2 µM DON were found along with an increased incidence of mastitis [44]; thus reflecting our in vitro values. In fact, due to a wide range of mycotoxin-producing fungi, their varied distributions in feed, the multidirectional effects on the host, and the high cost of laboratory analysis, mycotoxicosis is difficult to diagnose. Moreover, the detection of feed contaminated with mycotoxins is challenging, and mycotoxicosis may be suspected in all disease cases that display nonspecific symptoms and high resistance to conventional treatment [45].
Although less susceptible when compared to bovine PBMCs, the proliferative capacity of both porcine and chicken lymphocytes was strongly affected by DON. Despite several studies characterizing the effect of trichothecenes such as DON on immune parameters, we are the first to show the effect of DON and DOM-1 on chicken lymphocyte proliferation in vitro. Generally, the high bacterial content in poultry, which converts toxic DON to nontoxic DOM-1 before the small intestine, strongly decreases the amount of DON that reaches the small intestine, making the animal more robust against oral intoxication [4,16]. Reports additionally suggest that in poultry, only a small amount of ingested DON (~19%) actually reaches the small intestine as a native toxin [46]. It was recently also reported that DON-3α-sulfate is a major detoxification metabolite in chickens, accounting for approximately 88.6% of the administered doses of DON. Accordingly, the ability of chickens to biotransform DON to DON-sulfates could explain the reduced DON susceptibility of poultry [47]. Nevertheless, in accordance with our findings, literature reports indicate that DON impairs immune function in poultry. Exposure to a DON-contaminated diet, for example, reduced serum antibody titers against Newcastle disease virus in laying hens and broilers [48,49]. Diets containing high levels of Fusarium-toxin-contaminated grain fed to broilers have been shown to decrease the peripheral blood monocyte and B cell counts without altering serum immunoglobulin concentrations [50]. Other studies suggest that Fusarium mycotoxins in feed can decrease the level of blood leukocytes, B lymphocytes, and CD4 + and CD8 + T cells [51]. In a broiler study, Ghareeb et al. [52] reported a decrease in the plasma levels of tumor necrosis factor α as well as a downregulation of interleukin 1β, transforming growth factor β receptor 1, and interferon-γ gene expression in response to DON. The authors concluded that DON provokes and modulates immunological reactions in broilers, which could increase disease susceptibility.
With respect to the DON-induced inhibition of porcine PBMC proliferation, our findings confirm those of Taranu et al. [7], who reported that while DON exerted stimulating effects on porcine PBMC proliferation at concentrations up to 0.33 µM, further increases in the concentrations of the mycotoxin dose-dependently reduced proliferation. The authors reported considerable inhibition of cell proliferation at concentrations between 3.37 and 33.7 µM DON. Furthermore, Goyarts et al. [53] found significant reductions in porcine PBMC proliferation using 0.94 and 1.89 µM DON, with IC50 values of 1.04 µM (MTT assay) and 0.67 µM (BrdU assay). According to a study by Dänicke et al. [27], even 0.7 µM DON significantly reduced the viability of ConA-stimulated porcine PBMCs. Maximum proliferation decreases of approximately 80% were observed at 2.7 µM. In accordance with our findings, the same authors reported DOM-1 to have no effect on porcine proliferation up to a concentration of 22.9 µM.

Conclusions
In our study, DON inhibited lymphocyte proliferation in all three tested species, where the greatest decrease was seen in bovine PBMCs, followed by chicken and porcine PBMCs. This decrease can most likely be attributed to the capacity of the mycotoxin to cause ribotoxic stress and inhibit protein synthesis. We herein hypothesize that DON also might pose a risk factor to the immune system of animals that are frequently considered to possess a low DON susceptibility, such as ruminants and poultry. This effect could indicate that under certain conditions, for instance inflammation or gastric diseases, low dietary DON concentrations might impair animal health by acting as an immunosuppressant.

Cell Isolation
To isolate peripheral blood mononuclear cells (PBMCs), fresh porcine, chicken, and bovine whole blood was obtained from an abattoir in EDTA-containing (120 mg/mL) (Sigma Aldrich, St. Louis, MO, USA) centrifuge tubes and diluted 1:2 with phosphate buffered saline (PBS) (Gibco, Life Technologies, Carlsbad, CA, USA). Information on sex, breed, and history was not available. Pigs were slaughtered at an age of approximately 6.5 months, chicken at an average age of 70 days, and cattle at an age of approximately >18 months. Exact age from individual animals is unknown, but all used study animals had reached at least late adolescence. Of this mixture, 35 mL was gently layered onto 15 mL Ficoll-Paque TM Plus (GE Healthcare, Little Chalfont, UK) in 50 mL centrifuge tubes and centrifuged at 300× g without brake for 30 min. PBMCs, located in the middle of the obtained layers, were carefully aspirated and transferred into a fresh 50 mL centrifuge tube. Cells were washed three times with 40 mL PBS and cell numbers were determined via Tuerk solution (Sigma Aldrich, St. Louis, MO, USA) staining. Cells were used for assays immediately after isolation (no prior cryopreservation).  [54].

Deoxynivalenol (DON) and
Cells were immediately treated with DON (0.01-3.37 µM) or DOM-1 (1.39-357 µM) in the presence of 1.25 µg/mL (porcine PBMCs) or 2.5 µg/mL (chicken and bovine PBMCs) ConA (Sigma-Aldrich, St. Louis, MO, USA). Cells were incubated at 39 • C for 28 h (chicken PBMCs) or at 37 • C for 72 h (porcine and bovine PBMCs). The optimal seeding density, incubation temperature, and incubation time have been assessed in pilot experiments to assure the highest proliferation rates (data not shown). Control cells were treated with fully supplemented culture medium with (positive control) and without (negative control) ConA at each concentration.

Cell Proliferation ELISA, BrdU Assay
Relative proliferation was determined via the cell proliferation ELISA, BrdU assay (Roche, Rotkreuz, Switzerland). The assay was performed according to the manufacturer's instructions. Briefly, cells were labeled with BrdU labeling solution for 4 h, and the microplate was subsequently centrifuged at 300× g for 10 min. The labeling medium was removed, and cells were completely dried and fixed with the provided fixing solution for 30 min at room temperature. Subsequently, anti-BrdU-peroxidase solution was added to each well for 90 min at room temperature. Then, the antibody conjugates were removed and cells were washed three times with the provided washing solution. After treatment with the substrate solution, the reaction was stopped with 1 M sulfuric acid, and optical density (OD) was measured at 450 nm and 690 nm (reference wavelength) by a microplate reader (Biotek Instruments, Inc., Winooski, VT, USA). The highest measured OD values for ConA-stimulated PBMCs were around 1.1 (mostly, OD values were between 0.6-0.9). For the cell control (PBMCs only), OD values were around 0.04-0.09. The proliferation response is expressed as the stimulation index (SI), calculated as described in Equation 1. According to literature, a stimulation index of >2 was considered as positive [55]. Therefore, experiments where ConA-stimulated PBMCs did not fulfill this performance standard were excluded.
Stimulation index (SI) = OD (mean of treated cells) OD (mean of negative control) Equation (1): Calculation of the stimulation index (SI).

Cell Proliferation and Phenotyping of Bovine PBMCs by Violet Proliferation Assays
The isolation protocol for bovine lymphocytes used in the violet proliferation assays was the same as described above, with the exception that the centrifugation on the Ficoll-Paque gradient was performed at 920× g for 30 min. Following isolation, PBMCs used in violet proliferation assays underwent a freezing/thawing procedure as described by Leitner et al. [56].
Following thawing, bovine PBMCs were washed in PBS and counted in trypan blue. Cell viability was between 77 and 90%. The PBMCs were labeled with a CellTrace TM Violet Cell Proliferation Kit (Thermo Fisher Scientific, Waltham, MA, USA) as follows. The cell number was adjusted to 2 × 10 7 in 1 mL of PBS, and 1 mL of violet dye solution was added to the cells (resulting in a violet dye concentration of 5 µM), followed by incubation at 37 • C for 10 min in a water bath. To stop the violet dye uptake, 2 mL of FBS (PAN-Biotech, Aidenbach, Germany) was added and incubation continued for 15 min at room temperature in the dark. Finally, cells were washed three times in culture medium (RPMI 1640 + 10% heat-inactivated FBS + 100 U/mL penicillin, 100 µg/mL streptomycin, all from PAN-Biotech), followed by counting and plating at 2 × 10 5 cells per well. Cells were cultivated in the presence of ConA (3 µg/mL, Amersham Biosciences, Uppsala, Sweden) with or without different DON concentrations (0.1, 0.2, 0.4, 0.8, and 1.6 µM) in round-bottom 96-well plates at 37 • C for four days.
After the in vitro cultivation, cells were analyzed for proliferation and surface marker expression by four-color staining flow cytometry. The following primary antibodies were used: anti-CD8β (clone CC58, VMRD, Pullman, WA, USA), anti-CD4 (clone CC8, AbD Serotec, Kidlington, UK), and anti-TCR-γδ (clone GB21A, VMRD). These primary antibodies were labeled by the following isotype-specific secondary antibodies: goat anti-mouse IgG1 Alexa647 (Thermo Fisher Scientific, Waltham, MA, USA), goat anti-mouse IgG2a-PE (Southern Biotech, Birmingham, AL, USA), and goat anti-mouse IgG2b-Alexa488 (Thermo Fisher Scientific, Waltham, MA, USA), respectively. Initially, cells were harvested and washed once in PBS + 3% FBS. Cells were stained with primary and secondary antibodies in two consecutive incubations performed at 4 • C for 20 min in 96-well round-bottom plates. Between the first and the second incubation and after the second incubation, cells were washed twice in PBS + 3% FBS. Cells were finally resuspended in the same buffer and analyzed by a FACSCanto II flow cytometer (BD Biosciences, San Jose, CA, USA). Between 5 × 10 4 and 2 × 10 5 PBMCs were acquired per sample. Flow cytometry data were analyzed and processed by FlowJo software version 10.3 (FlowJo LLC, Ashland, OR, USA).

Statistics
Statistical analysis was performed with IBM ® SPSS (Version 19.0, IBM Corp., New York, NY, USA, 2010). Values from independent experiments are expressed as the means of 6-8 replicates ± standard deviation (SD). One replicate means isolating and testing the PBMCs of one individual animal in triplicate. The stimulation index and the relative values were calculated and the relative means of the replicates ± SD were plotted and used for statistical evaluation. All values were analyzed for normality (Shapiro-Wilk) and homogeneity of variance (Levene statistics). Normally distributed homogenous data were analyzed by analysis of variance (ANOVA) and Dunnett's t-tests and compared to control data. If data did not follow a normal distribution, a Kruskall-Wallis test was used. IC50 values were calculated with relative numbers using GraphPad Prism (Version 7.0, GraphPad Software, Inc., La Jolla, CA, USA, 2017). In short, data was normalized and then a four-parameter nonlinear regression curve (log (inhibition) versus response with variable slope (least squares ordinary fit, with the condition that the Hillslope is <0) was applied to calculate the IC50 values.