Seasonally Feed-Related Aflatoxins B1 and M1 Spread in Semiarid Industrial Dairy Herd and Its Deteriorating Impacts on Food and Immunity

1Department of Food Hygiene and Aquaculture, Faculty of Veterinary Medicine, Ferdowsi University of Mashhad, Azadi Square, Mashhad 1793, Iran 2Department of Pathobiology, Faculty of Veterinary Medicine and Institute of Biotechnology, Ferdowsi University of Mashhad, Mashhad, Iran 3Section of Immunology, Department of Microbiology and Immunology, Faculty of Veterinary Medicine, University of Tehran, Tehran, Iran

It is worrying reality that AFs levels in feedstuffs can be far higher than what we apparently measure. Inextricable link of various cancers to environmentally relevant levels of AFs, existence of masked AFs, and limited analytical capabilities for detection of AFs in agricultural commodities, sera, and dairy products have rendered the aspects of AFs' detection and impact the focus of most concern [1][2][3]5].
With EU's tolerable levels for AFB 1 in feed (2-8 g/kg) and AFM 1 in milk (0.05 g/kg) (the EU, 2010) and Iran's permissible AFB 1 in feed (5-20 g/kg) and foods (5-15 g/kg) and AFM 1 in milk [(0.5 g/kg, kg is equal to liter in case 2 Journal of Food Quality Table 1: Cluster random sampling and widespread contamination of aflatoxin B 1 in feedstuffs and aflatoxin M 1 in bulk milk during spring, summer, autumn, and winter in thestudy area (see also Figure 1). for milk) (Anonymous, 2002) (due to contaminated milk, the risk for lactating mammals is much higher compared to nonlactating ones) and globalization, routine monitoring of AFs is more urgent [16]]. Due to highly carcinogenic and immunosuppressive nature of AFs [even permissible level of AFB 1 is hazardous for immune cells/molecules in humans and animals [2][3][4]] and the lack of information on the immunotoxic influence of mixed AFB 1 and AFM 1 we aimed to (1) determine the seasonal levels of AFB 1 in various feedstuffs and of AFM 1 in bulk milk obtained from industrial cattle farms in agroecologically and geopolitically important semiarid northeastern Iran and (2) examine the in vitro effects of environmentally relevant levels of mix of AFB 1 and AFM 1 on different functions (free radicals or ROS production, phagocytic and killing capacity, and necrosis) of blood-and-udder's key innate immune cells, neutrophils (PMN), using luminometry, flow cytometry, and bactericidal assays for two major vertebrates' superbugs (i.e., Staphylococcus (S.) aureus and Escherichia (E.) coli).

Materials and Methods
Cluster random 72 feedstuffs (silage, concentrate, and totally mixed ration, TMR) and bulk milk ( = 200) samples were simultaneously seasonally obtained from industrial cattle farms (see Table 1 and Figure 1; = 25 farms) in sterile cold condition accordingly (ISO 6497 2002 [10]). Much lower detected levels of AFB 1 and AFM 1 were eventually used for in vitro effects of mix of AFB 1 and AFM 1 on isolated bovine blood and milk PMN.
Indirect ELISA were performed according to the test kits for feedstuffs (AFB 1 ELISA, EuroProxima, Beijerinckweg, The Netherlands) and bulk milk (Ridascreen5 AFM 1 , R-Biopharm, Germany), by measuring absorbance at 450 nm with ELISA reader (ELx 800, BioTek Instruments, USA). The lower detection limit for AFB 1 and AFM 1 was 0.05-0.1 g/kg.
For in vitro effects of AFs on blood and milk PMN a group of 12 physiologically healthy dairy cows were used for PMN isolation and analyses, accordingly [2,17]. AFB 1 and AFM 1 were obtained from Sigma and prepared according to Mehrzad et al. [2], with some modification (i.e., AFM 1 was directly dissolved in DPBS); they were then separately but simultaneously added, at a final concentration of 0 (control) and 20 ng/ml (10 ng/ml AFM 1 plus 10 ng/ml AFB 1 ) to the PMN cultured in complete RPMI 1640 medium (12 h, 37 ∘ C, 95% humidity, 5% CO 2 ). PMN were then washed and chemiluminescence (CL), flow cytometry-based neutrophils phagocytosis, and necrosis were measured according to the reported procedure [2,3].
To evaluate the effect of AFs on superoxide anion (O 2 − ) production and myeloperoxidase (MPO) activity of blood PMN, O 2 − production was measured in post-AFs treated PMN by converting to nM of cytochrome reduced using the extinction coefficient 550 nm = 2.1 × 10 4 M −1 cm −1 as described [18]. Post-AFs treated PMN MPO activity was measured based on the oxidation of ortho-dianisidine (0.8 mmol/l) of supernatant of sonicated PMN extract containing 0.1 mmol/l of added H 2 O 2 using microtiter plate spectrophotometer at 450 nm (Multiskan Plus Type 314, Labsystems, Helsinki, Finland). The analyses of phagocytosis and killing of E. coli and S. aureus were done accordingly [2,17].    Journal of Food Quality Statistical analysis was performed using SPSS version 19 and SAS Version 9.1 (SAS Institute Inc., Cary, NC) software. The Chi-square and Fisher exact tests were used to assess the possible differences in AFs spread in feeds and milk. To compare the measured parameters of two groups, analysis of variance was used; all data were presented as means ± SEM and hypothesis testing was done at the 5% significance level.
The PMN phagocytosis-(in)dependent CL/(non)particlestimulated luminol-enhanced CL (Figures 2(a1) and 2(a2)) and the flow cytometry-based phagocytosis assay of blood (Figure 2(c1)) and milk (Figure 2(c2)) PMN consistently revealed a significant decrease in phagocytic activity and killing capacity by AFs-exposed PMN. Among PMN stimulated with PMA, latex beads, or Pansorbin5, the AUC for the mix of AFs-exposed blood and milk PMN were, respectively, 16, 38, and 30% and 18, 37, and 31% lower than control PMN. Further, max in the AFs treated blood PMN stimulated with PMA, latex, and Pansorbin was slightly decreased (data not shown). Interestingly, the luminol-dependent CL arising from added hypochlorite (HClO) was significantly higher in mix of AFB 1 -and AFM 1 -treated group (Figure 2(b), upper panel insert). Furthermore, while low levels of mix of AFB 1 and AFM 1 showed little effects on blood PMN necrosis (Figure 2(c)), it decreased both the production of O 2 − ( < 0.01) by PMN (Figure 2(d)) and the MPO activity ( < 0.05) of PMN (Figure 2(e)). Unlike post-AFs treated necrotic PMN, neutrophils exposed to mix of AFB 1 and AFM 1 were less efficient to phagocytose and kill S. aureus and E. coli (see Figures 2(a) and 2(b)).

Discussion
The level which is alarmingly higher than the permissible levels of AFs in feedstuffs and milk (5 g/kg for feed AFB 1 and 5 ng/kg for milk AFM 1 ) is largely attributed to climate conditions and inappropriate management of grasses and grains during harvest, transportation, drying and mixing stages [7,19]. For example, improper processing/storage condition of higher dry matter corn silage likely boosts AFB 1 production [6, 7, 10], due mainly to weakening anaerobic bacterial mediated lactic acid production, increasing mi(a)croenvironment pH, thereby boosting aflatoxigenic fungal growth. Owing to being categorized as group-1 carcinogens, many countries and regulatory agencies have harshly imposed tolerable limits on AFs. For example, the European Commission has set tolerable limit for AFB 1 and total AFs (B 1 , B 2 , G 1 , and G 2 ) 2-8 g/kg and 4-15 g/kg, respectively, in crops such as nuts and grains (EU, 2010); furthermore, Iran has set the tolerable limit for AFB 1 for feedstuffs and foods 5-20 g/kg and 5-15 g/kg, respectively (Anonymous, 2002); also, maximal tolerable for milk AFM 1 in EU and Iran is 0.05 and 0.5 g/kg, respectively. As such, routine harsher monitoring of AFs levels in different feedstuffs is urgent in semiarid northeastern Iran. For the detection of AFB 1 and AFM 1 , herein we used ELISA with sensitivity and specificity of above 98% with acceptable recovery rate, LOD/LOQ, and precision; nevertheless it is worth simultaneously confirming their quantification with some more advanced analytical tool like HPLC or LC/MS/MS as done by others [12,13].
During harsh seasons more contaminated moldy feeds are often used in feedlots; analytical incapability to detect very low (invisible) levels of AFs and lack of AFs toxicity data in cows [5] exacerbate the complexity of AFs exposure in food animals and humans.
Among various feedstuffs analyzed in central and northeastern Iran, concentrate was maximally contaminated to both aflatoxigenic molds and AFB 1 [9,10]. Though a tiny amount of feed AFB 1 appears in milk of AFB 1 -consumed dairy cattle, AFB 1 simply transmit into milk as AFM 1 with high cancer risks for consumers, especially neonates and children, who consume far more milk and dairy products. AFM 1 is hardly degradable by pasteurization procedure [5,15]; nonetheless, fermentation could lead to biodegradation of AFs in fermented dairy products and thus less toxicity.
The background of the selected doses examined here was (1) in line with the first part of this study on AFB 1 and AFM 1 quantification, (2) our current understanding on the metabolism and toxicodynamics of AFB 1 , and (3) from researchers [2][3][4][5]20] in bovine, porcine, and human in vitro models. Researchers have even used higher doses of AFs [21]. Although our knowledge on interaction(s) of AFs effects in animals, especially food/dairy-producing ruminants, is   Journal of Food Quality limited, nevertheless it is worth mentioning the point of intake of AFB 1 with the diet and its transfer to milk in high yielding lactating cows. Given that a cow with average daily milk production of 50 litter is consuming ∼40 kg TMR per day and ∼50% of AFB 1 in contaminated TMR goes to the blood with a tenth (mainly converted to AFM 1 ) to milk [20,22], the assumed AFB 1 and AFM 1 in blood/milk can accumulatively reach above our ex vivo tested levels.
The inhibiting effect of AFs on the killing capacity of PMN for E. coli and S. aureus was manifested mainly by reduction of MPO activity and intracellular ROS production; this further confirms the negative effects of mix of AFB 1 and AFM 1 on the innate immunity; that is why AFB 1exposed dairy cows are more susceptible to environmental infections, specially mastitis and metritis (unpublished data). Nevertheless, we know little on how mix of AFB 1 and AFM 1 behaves in the phagosome and lysosome, where extremely large amounts of oxidants and granule constituents, through MPO-H 2 O 2 -HOCl system, are released [2,17,23]. What happened on different stages of PMN necrosis and apoptosis remained further investigation.
In short, a relatively high level of feed and milk AFs with more pronounced AFB 1 spread in cereal crops during winter and considering the economic, health, and cancer importance of AFs, percussion measures, hugely requires boosting food/health quality and lowering the risk of AFs exposure in farm animals and human in the study region.

Additional Points
Practical Applications. Poor harvesting, drying, transportation, and storage of feed/agricultural commodities lead to mold growth and particularly AFs formation in feedstuffs and milk and thus many animal and public health risks. Feedstuffs and bulk milk were sampled from various industrial dairy farms during 4 seasons in semiarid northeastern Iran for AFB 1 and AFM 1 detection and their ex vivo effect on peripheral granulocytes. Alarmingly, maximal immunotoxic levels of AFB 1 and AFM 1 in feedstuffs and milk occurred in winter followed by autumn, spring, and summer. More controls of feed materials would be needed to prevent an undesirable contamination of feed, foods, and milk/dairy products.

Conflicts of Interest
No potential conflicts of interest were reported by the authors.