Assessing Immunomodulatory Effects of Penicillium Mycotoxins Using Bovine Macrophages Cell Line

Initially, concentration-response curves for each of these PMs were established based on cell proliferation and viability. The potency of these PMs based on their IC50s (concentration that inhibits 50% cell proliferation) from highest to lowest was: 0.56 µM (PAT) > 12.88 µM (OTA) >

29.85 µM (PA) > 91.20 µM (CIT) > not determined (MPA). LC50s (concentrations that kill 50% of cells) for PAT and PA were determined to be 4.46 μM and 175.79 μM, respectively. In addition to this, binary mixtures of some PMs at their respective IC25 and lower were shown to have significant interactions on cell proliferation. CIT+OTA had significant synergism, while

PAT+MPA, CIT+MPA, CIT+PA and MPA+PA exhibited significant antagonism. PAT+PA showed significant antagonism at their respective IC25s, but were antagonistic at lower concentrations. The PMs at their sub-lethal levels of IC25s also differentially altered the gene expression of the following cytokines, including IL-1α, IL-6, IL-10, IL-12, IL-23 and TGF-.

Reactive oxygen species (ROS) production and phagocytosis of Mycobacterium avium subspecies paratuberculosis (MAP) was affected at higher sub-lethal concentrations. OTA in combination with CIT or PA synergistically suppressed the gene expression of the following epigenetic enzymes: DNA methyltransferases (DNMT-3s), histone demethylase (JMJD-3) and histone deacetylase (HDAC-3). When the efficacy of a mycotoxin-binding Mycosorb A+ (MA+), a mycotoxin binder, was assessed based on BoMac proliferation, 0.2% MA+ showed the highest efficacy in preventing OTA toxicity at pH 3.0, while its beneficial effect in preventing CIT toxicity was also observed with a longer incubation time than 6 hours and higher inclusion level of MA+ than 0.5%. The results from this thesis indicate that the PMs at their sub-lethal concentrations could potentially modulate the macrophage functions. These adverse effects also appear to be enhanced by PM interactions. This study also showed beneficial effect of MA+ in preventing the toxicity of PMs.

ACKNOWLEDGEMENTS First of all, I would like to express my sincere appreciation and thanks to my advisor Dr. Niel

Karrow for his support throughout my graduate studies. I have known you for 9 years starting out as a second year undergrad in your class; who would have thought that I’d end up doing a Masters and continuing on with a PhD in your lab? It has been great spending time with you during conferences, camping, movie watching and other fun things! I believe that you did your best in looking for a girlfriend(s) for me, and I really appreciate your help on that. You have shown the attitude and the substance of a genius, as well as continually and persuasively conveying a spirit of research. I feel a deep gratitude for the advice you have given me as a mentor and a lifelong friend.

I also would like to thank my co-advisor, Dr. Herman Boermans. You tried the hardest to teach me how to ride a bike, and then ditched me while I almost got myself killed, but in the end I survived! Thank you for the advice on my project and all the fun adventures on conference trips.

I also would like to thank my advisory committee members, Dr. Trevor Smith and Dr. Swamy

Haladi for all their help on various aspects of my project. It was pleasure to have an opportunity to work on a project with you. I would also like to thank Dr. Nina Cedergreen from University of

Copenhagen for her collaboration on the mycotoxin mixture study. Special thanks goes to Dr.

Margaret Quinton for all her help with statistical analysis.

Most importantly, I would like to express my gratitude to my parents, my father, Dae-Woong

Oh, and my mother, Jung-Ae Park, for all your support throughout my academic and non-academic life. There have been many turning points in my life, but you have supported me both mentally and financially in whichever path I chose. I could not have reached this point today without your support, and for that I am deeply grateful. To my brother Min Oh, thank you for being a great big

v brother. We have shared lots of laughs and supported each other through many aspects of life.

Let’s find a girlfriend and get married within next 10 years before it is too late.

I would like to thank my dear friend Dr. Rebecca Fisher for providing me with advice and support during the writing process even during your busiest time of year, (especially with the first

Chapter of this thesis). Happy marriage to you! To Dr. Bhawani Sharma, thank you for your help during the first half of my graduate studies. You have shared lots of advice and so many other interesting stories. I wish you well and hope you have a full recovery soon. And thanks to Philip

Mead for designing the mCherry-bug and the phagocytosis assay for our lab. This will be used greatly in the future. Special thanks to Qiumei as she taught me all the lab techniques and gave me hand whenever needed. And also thanks to Caroline Balch, and Rachael Cliff for helping me with the gene expression study during the epigenetic trials. I would also like to acknowledge Dr. Bruce

Wilkie, Dr. Julie Schmied and Dr. Prithy Babu for my previous lab experience before starting my grad work. You helped me establish my very first step into the research field. And a special thank- you goes to a good friend of mine Julie Kim, who has helped me greatly since my freshmen year of undergrad. Also special thanks to Robert Quehl for proofreading my long thesis for me.

I would lastly but certainly not least, like to thank all my friends and lab mates: Sanjay, Ziwei,

Nancy, Laura, Judy, Heba, Sam, Chris, J.J., Frank, Yube, Duc, Li, Eddie, Trans, and all other friends who I forgot to mention here. It has been a great time with you ladies and gentlemen. For the last five years from the beginning of my grad work, many things have happened, and I learned a lot and met many people and friends. Although this is the end of my grad work, this is just a beginning of new chapter of my life. Thank you all my families and friends for leading me to where I am standing today. “Thank you for the good times, the days you filled with pleasure.

Thank you for fond memories, and for feelings I will always treasure.”- Karl Fuchs

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... iv

TABLE OF CONTENTS ...... vi

LIST OF TABLES ...... x

LIST OF FIGURES ...... xi

LIST OF ABBREVIATION...... xii

Chapter 1: Literature Review ...... 1

1.1. Introduction ...... 1

1.1.1. General Overview ...... 1

1.1.2. The general occurrence and toxicity of PMs ...... 3

1.1.3. Effect of PMs on physical barriers ...... 7

1.2. Penicillium myotoxins (PMs): immunomodulation and immunotoxicity ...... 10

1.2.1. Ochratoxin A (OTA) ...... 11

1.2.2. Citrinin (CIT) ...... 15

1.2.3. Patulin (PAT) ...... 18

1.2.4. Penicillic acid (PA) ...... 22

1.2.5. Mycophenolic acid (MPA) ...... 24

1.2.6. Roquefortine C (RQC) ...... 30

1.3. Epigenetic changes and its relation to inflammatory response ...... 31

1.4. Summary and Conclusion ...... 34

Chapter 2: Experimental Rationale, Hypothesis, and Objectives ...... 41

2.1. Experimental Rationale ...... 41

2.2. Experimental Objectives ...... 44

2.3. Experimental Hypotheses ...... 45 Chapter 3: Immunotoxicity of Penicillium Mycotoxins on Viability and Proliferation of Bovine Macrophage Cell Line (BoMacs) ...... 46

3.1. Abstract ...... 46

vii

3.2. Introduction ...... 47

3.3. Material and methods ...... 48

3.3.1. Cell Preparation and Mycotoxin Exposure ...... 48

3.3.2. Assessing Cytotoxicity of CIT, OTA, PAT, MPA, and PA to BoMacs ...... 49 3.3.3. Assessing the Inhibitory Effect of CIT, OTA, PAT, MPA, and PA on BoMac Proliferation ...... 50

3.3.4. Assessing the Combined Effects of Penicillium Mycotoxin on BoMac proliferation 50

3.3.5. Data Analysis ...... 50

3.4. Result and Discussion ...... 51 3.4.1. Cytotoxicity and Inhibition of the Cell Proliferation due to Individual Mycotoxin Exposure ...... 51

3.4.2. Combined Effect of Penicillium Mycotoxins on the Proliferation of BoMacs ...... 53

3.5. Summary and Conclusion ...... 54 Chapter 4: Exposure to Penicillium Mycotoxins Alters Gene Expression of Enzymes Involved in the Epigenetic Regulation of Bovine Macrophages (BoMacs) ...... 60

4.1. Abstract ...... 60

4.2. Introduction ...... 61

4.3. Material and methods ...... 64

4.3.1. Mycotoxins ...... 64

4.3.2. BoMac culture ...... 64

4.3.3. Cell preparation and mycotoxin exposure ...... 64

4.3.4. Total RNA isolation and cDNA synthesis ...... 65

4.3.5. Real-time PCR quantification ...... 65

4.3.6. qPCR amplication curve and statistical analyses ...... 66

4.4. Result ...... 67

4.5. Discussion ...... 68

4.6. Summary and Conclusion ...... 74

viii

Chapter 5: Effect of Penicillium Mycotoxins on Bovine Macrophage (BoMacs) Function ...... 79

5.1. Abstract ...... 79

5.2. Introduction ...... 80

5.3. Materials and Methods ...... 82

5.3.1. Cell culture ...... 82

5.3.2. Cytokine gene expression ...... 82

5.3.3. Measuring reactive oxygen species (ROS) and phagocytosis of MAP ...... 83

5.3.4. Statistical analysis...... 85

5.4. Results ...... 86

5.4.1. Effect of PMs on BoMacs cytokine gene expression ...... 86

5.4.2. Effects of PMs on BoMacs ROS production and phagocytosis of MAP ...... 87

5.5. Discussion ...... 87

5.6. Summary and Conclusion ...... 92 Chapter 6: In vitro Exposure of Penicillium Mycotoxins (PMs) with or without Mycosorb A+ (MA+) on Bovine Macrophages (BoMacs) ...... 99

6.1. Abstract ...... 99

6.2. Introduction ...... 100

6.3. Methods ...... 101

6.3.1. Preparation of cells ...... 101

6.3.2. Incubating PMs with MA+ and preparation of exposure media: ...... 101

6.3.3. BoMacs proliferation ...... 103

6.3.4. Data Analysis ...... 103

6.4. Results ...... 104

6.4.1. MA+ effectiveness at different pH levels ...... 104

6.4.2. MA+ effectiveness over time ...... 104

6.4.3. Determining optimal effective percentage of MA+ ...... 104

6.4.4. MA+ efficacy at different PM concentrations ...... 105

6.5. Discussion ...... 105

6.6. Summary and Conclusion ...... 110 Chapter 7: Assessing Interactions of Binary Mixtures of Penicillium Mycotoxins (PMs) by using Bovine Macrophage Cell Line (BoMacs) with the Assumption of Additivity Model ...... 116

7.1. Abstract ...... 116

7.2. Introduction ...... 117

7.3. Methods ...... 120

7.3.1. Data Analysis and Plotting of the Previous PM Toxicity Data ...... 120 7.3.2. Assessing the interactions of PMs using binary combinations over the dilutions of IC25s ...... 121

7.4. Results ...... 123

7.5. Discussion ...... 124

7.6. Summary and Conclusion ...... 131

Chapter 8: General Discussion and Conclusion ...... 135

References ...... 142

Appendix 1: Schematic diagram of how Penicillium mycotoxins alter the cell activities ...... 171 Appendix 2: Preliminary study on the effect of 6 h in vitro PM exposure at different concentrations on BoMac ROS production...... 172 Appendix 3: Intracellular ROS production of BoMacs with different stimulators by using Carboxy-H2DFFDA ...... 173

Appendix 4: Effect of PAT on the viability of BoMacs ...... 174

Appendix 5: Effect of MA+ on BoMac proliferation ...... 175

LIST OF TABLES Table 1.1: The occurrence of Penicillium mycotoxins (PMs) in animal feed and food products...... 366 Table 1.2: Brief toxicological information of PMs other than immunotoxicity ...... 377 Table 1.3: Summary of immunomodulatory effect of PMs. (all references in the context; NA-not available) ...... 378

Table 3.1: Mycotoxin concentrations used for BoMac exposure study (M) ...... 555

Table 3.2. The LC50, IC50, and IC25 of various Penicillium mycotoxins (M)...... 566 Table 3.3: Comparison of Mycotoxin IC50s for cell proliferation of porcine, bovine lymphocytes, and BoMacs (M)...... 577 Table 4.1: Summary of the designed oligonucleotide primer sequences (F forward primer, R reverse primer) ...... 766 Table 5.1: Exposure concentrations of PMs used in the cytokine gene expression, ROS and phagocytosis assay...... 944 Table 5.2: Summary of designed oligonucleotide primer sequences for cytokine gene expression study ...... 955

Table 6.1: Summary of experimental variables and constant factors for each Figure ...... 1111

LIST OF FIGURES Figure 1.1: Simple schematic diagram of how the antimicrobial properties of PMs could disrupt the homeostasis of intestinal immune system...... 398 Figure 1.2: The Schematic diagram of how PMs alter inflammatory response of macrophages by altering HDAC-3...... 39 Figure 3.1: A. Cytotoxicity and B. Proliferation of BoMacs following 48 h of Mycotoxin Exposure...... 57

Figure 3.2: The combined Effects of Mycotoxins on the Proliferation of BoMacs...... 58 Figure 4.1: (A-F) The effect of in vitro PM exposure on the expression of BoMac candidate methylating and demethylating genes relative to the GAPDH house-keeping gene...... 76 Figure 4.2: (A-D) The Effect of in vitro PM treatment on the expression of BoMac candidate acetylating and deacetylating genes relative to the GAPDH...... 787

Figure 5.1 (A-G): Effect of in vitro PM exposure on BoMac cytokine gene expression...... 965

Figure 5.2 (A-E): Effect of 6 h in vitro PM exposure on BoMac ROS production...... 976 Figure 5.3 (A-E): Effect of 6 h in vitro PM exposure on BoMacs phagocytosis of mCherry- MAP...... 987 Figure 6.1 (A-E): The effect of pH (3, 5 and 7) on the efficacy of MA+ in preventing the PM toxicity on BoMac proliferation...... 1121 Figure 6.2 (A-E): The effect of incubation time (1, 2, 4 and 6 h) on the efficacy of MA+ in preventing the PM toxicity on BoMac proliferation ...... 1132 Figure 6.3 (A-E): The effect of percent amount of MA+ (0.05, 0.1, 0.2, 0.5 and 1 %) on the efficacy of MA+ in preventing the PM toxicity on BoMac proliferation ...... 1143

Figure 6.4 (A-E): The effect of MA+ at different PM concentrations...... 1154 Figure 7.1: The comparison between ‘observed’ and ‘estimated’ of mixtures of PMs at their respective IC25s by using the IA model...... 1321

Figure 7.2 (A-J): The predicted curve of PM mixtures toxicity by using the CA model ...... 1332

Figure 7.3 (A-J): The comparison between ‘observed’ and ‘estimated’ of PM mixtures ...... 1343

xii

LIST OF ABBREVIATION

0.25IC25 Second serial dilution of DMSO Dimethyl sulfoxide IC25 DNA Deoxy 0.5IC25 First serial dilution of IC25 DNMT DNA methyltransferases A. Aspergillius DON Deoxynivalenol AbMIR Antibody-mediated immune response E. coli Escherichia coli ADH Yeast alcohol EC Effect concentrations dehydrogenase ELISA Enzyme-linked BALF Bronchioaleolar lavage fluid immunosorbent assay B-cell B-lymphocytes ERK Extracellular signal- regulated kinases Bmi-1 Polycomb ring finger oncogene FAO Food and Agriculture Organization BoMacs Bovine macrophage cell line FBS Fetal bovine serm CA Concentration addition Foxp3 Forkhead Box P3 Caco-2 Human epithelial colorectal adenocarcinoma cells GALT Gut-associated lymphoid tissues CCR Chemokine receptors GAPDH Glyceraldehyde 3-phosphate CD Cluster of Differentiation dehydrogenase CFU Colony forming units GDP Guanosine Diphosphate CI Confidence interval GMA Glucamannan mycotoxin adsorbent CIT Citrinin GSH Glutathione CMIR Cell-mediated immune response GSSG GSH reductase

CO2 Carbon Dioxide GTP Guanosine triphosphate Con A Concanavalin A H2DFFDA 5-(and-6)-carboxy-2',7'- difluorodihydrofluorescein COX-2 Cyclooxygenase-2 diacetate

Ct Cycle threshold H2O2 Hydrogen Peroxide CTLs Cytotoxic T-cells H3K27 Histone H3 lysine 27

xiii

H3K4 Histone H3 lysine 4 JAK Janus kinase HDAC Histone deacetylases JMJD3 Jumonji domain-containing protein 3 HDACi Histone deacetylase inhibitors JNK Jun N-terminal kinase HEPES 4-(2-hydroxyethyl)-1- LC Liquid Chromatography piperazineethanesulfonic acid LC50 Lethal concentration that kills 50% of cells or animals HPLC High performance Lipid Chromatography LD50 Lethal dose that kills 50% of cells or animals hsp72 Heat shock protein 72 LDH Lactic dehydrogenase Iβ- Inhibitor of NF-β LPS Lipopolysaccharide IA Independent action MAP Mycobacterium avium IC25 Concentration that inhibits subspecies paratuberculosis 25% of cell proliferation MAPK Mitogen-activated-protein IC25 Concentration that inhibits kinase 25% of cell proliferation mDC Monocyte-derived dendritic IC50 Concentration that inhibits cells 50% of cell proliferation MIP Macrophage inflammatory IC50 Concentration that inhibits protein 50% of cell proliferation MMF Mycophenolate mofetil IFN Interferon MOI Multiplicity of infection IgA Immunoglobulin A Mono-Mac-6 Human monocytic cell line IgG Immunogloublin G MPA Mycophenolic acid IgM Immunoglobulin M MS Mass spectrometry IL Interleukin Mycosorb A+ MA+ IMPDH Inosine monophosphate dehydrogenase NF-B Nuclear factor-B iNOS Inducible nitric oxide NK cell Natural Killer cell synthase NO Nitric oxide Intravenous ip Oral po Intraperitoneal ip OD Optical Density

xiv

OTA Ochratoxin A RPMI 1640 Roswell Park Memorial Institute 1640 media OVA Ovalbumin RQC Roquefortine C PA or B Percent proliferation from the individual mycotoxins Spp Species PA Penicillic acid sRBCs Sheep red blood cells PAM-3 Pathogen associated STAT Sgnal transducers and molecule Pam3CSK4 activators of transcription PAMP Pathogen-associated T-2 toxin Trichothecene mycotoxin molecular patterns T-cell T-lymphocytes PAT Patulin TER Transepithelial resistance PBL Peripheral blood lymphocytes TGF-β Transforming growth factor PBMC Peripheral blood Th T helper cell type mononuclear cell TLC Thin Layer Chromatography PC plasma cell TNFR-1 TNF-α receptors-1 PCR Polymerase chain reaction TNF-α Tumor necrosis factor-α

PEM Estimated percent TRADD TNF receptor-1-associated proliferation from the death domain mixture T-reg cell Regulatory T-cells PHA Phytohaemagglutinin UTX Tetratricopeptide-repeat X PM Penicillium Mycotoxin VCAM-1 Vascular cell adhesion POM Observed percent molecule 1 proliferation from the mixture PRRs Pattern recognition receptors

PWM Pokeweed mitogen qPCR Quantitative PCR

RBCs Red blood cells RNA Ribodeoxynucleic acid ROS Reactive oxygen species

Chapter 1: Literature Review 1.1. Introduction 1.1.1. General Overview Mycotoxins are biologically active secondary metabolites of filamentous fungi that commonly contaminate a wide variety of cereal and livestock feed crops worldwide. The Food and

Agriculture Organization (FAO) estimates that approximately 25% of food and food stuffs around the world are contaminated by fungal mycotoxins (Boutrif and Canet, 1998). Increasing humidity and temperature accompanying global climate change will likely favour more fungal contamination of feedstuffs, affecting the safety of both human and animal feeds (Miraglia et al.,

2009). Economic losses from mycotoxin contamination occur due to reduced crop yield and value, as well as decreased livestock productivity and health from the consumption of contaminated feed

(Schmale III and Munkvold, 2009).

Penicillium molds are one of the common fungal species found in crops, fruits and animal feeds of Northern Europe, North and South America as well as East Asia (Almeida et al., 2000;

Creppy, 2002; Duong, 1996; Xiao et al., 2011). Poor storage conditions for most grains and food products often provide favourable environmental conditions, such as moderate temperature with high humidity, for mycotoxin production from Penicillium mold species (Dalcero et al., 1998;

Mansfield et al., 2008). Due to their dramatic increase during the ensilaging of animal feed, the

Penicillium mycotoxins (PMs) are also refereed as ‘silage mycotoxins.’

A number of PMs can be detected in stored feed or food such as ochratoxin (OTA), citrinin

(CIT), patulin (PAT), penicillic acid (PA), mycophilic acid (MPA), and roquefortine C (RQC)

(Lennox et al., 2003; Mateo et al., 2007). These PMs are known to cause a variety of toxic effects in animals. For example, OTA, CIT and PA are nephrotoxic (Braunberg et al., 1992; Sansing et

al., 1976) and hepatotoxic (Dickens and Jones, 1965) in monogastric species, while PAT is a potent hepatotoxicant in both monogastric and ruminant species (Hayes et al., 1979; Sommer et al., 1974;

Ukai et al., 1954). These PMs also affect the host immune response, which could predispose animals to secondary diseases (Table 1.1 and 1.2; Al-Anati and Petzinger, 2006; Ferrante et al.,

2008; Herzog-Soares and Freire, 2004).

Regardless of their abundance in animal feed and their potential to affect the immune system, the agricultural and economic importance of PMs are often under-estimated because it is generally assumed that PMs are readily metabolized by gut microbiota and hepatic enzymes (Abrunhosa et al., 2010; Fuchs et al., 2008). However, considering that most of these PMs have anti-microbial properties (Kavanagh, 1947) and are found to be hepatotoxic to animals (Dickens and Jones, 1965;

Hayes et al., 1979; Sansing et al., 1976), it is possible that detoxification of the PMs is attenuated by these properties. RQC for example, was found to inhibit cytochrome-p450 activity by reducing nicotinamide adenine dinucleotide phosphate (NADPH) consumption (Aninat et al., 2001), and p450 is an important enzyme for detoxifying mycotoxins (Smith and Harran, 1993). Furthermore, mycotoxins can also enter the circulation and target tissues without passing though the digestive tract. For example, it has been reported that some Fusarium and/or Penicillium mycotoxins, such as OTA, can enter the circulation during inhalation (Halstensen et al., 2004; Skaug et al., 2001) and dermal exposure (Panda et al., 1997). Since entry via these routes bypasses initial detoxification provided by gut microbiota and the liver, exposure via inhalation and dermal contact could be a health risk for farmers and livestock (EFSA, 2012; Reijula and Tuomi, 2003; Smith et al., 1995).

1.1.2. The general occurrence and toxicity of PMs 1.1.2.1 Ochratoxin A (OTA) OTA is a common PM found in stored cereal grains, nuts as well as some brewed liquors, because temperature and humidity during storage and brewing often favour OTA production

(Mateo et al., 2007; Visconti et al., 1999). OTA was initially isolated from Aspergillius ochraeus in 1965, but was later also isolated from P. citrinin and P. verrucosum (Pitt, 1987). The lethal dose that kills 50% (LD50) of mice via intraperitoneal and oral exposure was estimated to be 29.4 mg/kg and 46.0 mg/kg, respectively (Moroi et al., 1985), which are much higher than typical OTA concentrations found in silage.

Nephrotoxicity is the most common pathology associated with OTA exposure at a concentration greater than 0.05 mg/kg in monogastric species, such as rats and pigs (Baudrimont et al., 1994; Milićević et al., 2008). However, a high dose of 13 mg/kg, or cumulative exposure to a concentration as low as 3 mg/kg was also reported to cause teratogenicity to embryos of rats and chicks (Brown et al., 1976; O’Brien et al., 2005; Stoev, 2010) as well as goats (Ribelin et al.,

1978). A few studies have also shown that OTA significantly reduced the overall performance of both laying hens and broiler chickens (Niemiec and Borzemska, 1994; Prior et al., 1980). OTA was also found to have synergistic toxicity in poultry species when mixed with either CIT or PA

(Stoev et al., 2004; Veselá et al., 1983).

1.1.2.2. Citrinin (CIT) CIT is a common mycotoxin isolated along with the natural yellow biopigment produced by

Monascus spp. (Carvalho et al., 2005). CIT was first isolated from P. citrinin (Hetherington and

Raistrick, 1931), and was subsequently detected from P. verrucosum and P. expansum that

frequently contaminate cereal grains, rotten fruits, and juice (Martins et al., 2002; Scott et al.,

1972).

Several animal studies have reported significant toxicity mediated by CIT, and the oral LD50 to the rat was estimated to be 50 mg/kg (Leatherhead RI, 2013). The LD50s in response to intraperitoneal exposure have been estimated to be 35-58 mg/kg for the mouse and 19 mg/kg for the rabbit (Leatherhead RI, 2013). CIT appears to be a nephrotoxin based on its toxicity to porcine kidney cells (Braunberg et al., 1992), while teratogenicity was also observed in rats that were exposed to 35 mg/kg during gestation and in chicken embryos exposed to CIT concentrations between 1.0 to 10.0 μg (Reddy et al., 1982; Veselá et al., 1983). The toxicity of CIT was shown to be synergistically enhanced in poultry species when co-exposure occurred with OTA (Veselá et al., 1983).

1.1.2.3. Patulin (PAT) PAT is one of the most potent mycotoxins found in food (Puel et al., 2010). P. expansum is the most common mold species that produces PAT (Sommer et al., 1974); however, other

Penicillium and Aspergillius mold species including P. urticae, P. melinii, P. claviforme, A. clavatusand, A. terreus can also produce PAT (Ciegler, 1977). PAT can be found in fruits, flour, cereal grains, and meat (Ciegler, 1977; Martins et al., 2002); however, it is most often detected in the fruit-based products, such as apple and grape juices (Kataoka et al., 2009; Scott and Somers,

1968).

The LD50 of acute PAT exposure for mice, neonatal rats, and weaning rats was determined to be 17, 6.7 and 108-118 mg/kg BW, respectively, with the following pathological signs: focal hepatic necrosis, atelectasis, alveolar septal congestion and intra-alveolar haemorrhaging (Hayes et al., 1979). In 1954, it was suggested that PAT caused haemorrhagic syndrome and the death of

over 100 Japanese dairy cattle that had been fed malt contaminated with PAT produced from P. urticae (Ukai et al., 1954). A similar incident occurred in Germany and France as a result of PAT produced from A. clavatus contamination of wheat (Ciegler, 1977). Chronic exposure to 0.1 mg/kg of PAT in rat feed resulted in various pathological signs, including haemorrhage, plasma cell hyperplasia as well as cell destruction and abnormalities of the nucleus and organelles of rat liver and kidneys (Arzu Koçkaya et al., 2009). No evidence of either teratogenicity or mutagenicity has been observed in mice (Reddy et al., 1978).

1.1.2.4. Penicillic acid (PA) PA is often detected in various grains, tobacco, and has even been found in moldy sausages

(Ciegler, 1977; Ciegler and Kurtzman, 1970; Müller and Amend, 1997; Snow et al., 1972). Its potential toxicity has been known since 1896; however, it was first isolated from P. puberulum in

1913 (Ciegler et al., 1971). Several different Penicillium spp. including P. puberulum, P.martensii,

P. palitans, and P. cyclopium are the major producers of PA (Alsberg and Black, 1913; Ciegler and Kurtzman, 1970), but it is also produced by a few Aspergillius spp. such as A. ochraceus

(Ciegler et al., 1972).

The LD50s for PA for rats via subcutaneous, intravenous and oral exposure were determined to be 110, 250 and 600 mg/kg, respectively (Murnaghan, 1946). PA has been shown to be hepatotoxic and hemorrhagic in dogs (Hayes et al., 1977), and potential carcinogenicity was reported in early rat and mice PA studies (Dickens and Jones, 1965). Similar to CIT, PA has been reported to be nephrotoxic and hepatotoxic to mice, and its toxic effects can be synergistically enhanced when mixed with OTA (Sansing et al., 1976).

1.1.2.5. Mycophenolic acid (MPA) MPA is a common contaminant found in ensilaged and grain feeds (Burkin and Kononenko,

2010), as well as some of fermented dairy products (Usleber et al., 2008). MPA is produced by P. stoloniferum, P. brevicompactum, P.roqueforti and P. carneurn (Bartman et al., 1981; Boysen et al., 1996; Lafont et al., 1979).

LD50s for MPA by oral exposure were determined to be 352 mg/kg for the rat, 1000 mg/kg for the mouse, and over 6000 mg/kg for the rabbit (Menon et al., 2011), which indicates that MPA has relatively low toxicity in animals compared to other PMs. Although low in toxicity, one of the most distinctive characteristics of MPA is its anti-proliferative effect on specific cell types. MPA has been shown to inhibit the proliferation of smooth muscle cells (Ahn et al., 2007), endothelial cells (Mohacsi et al., 1997), as well as some immune cells, such as lymphocytes (Keblys et al.,

2004; Oh et al., 2012). Ironically, MPA’s low toxicity and immunomodulatory properties have resulted in many human pharmaceutical studies, but few animal studies have been carried out despite the fact that it is frequently found at high concentrations in ensiled animal feed (Mansfield et al., 2008). MPA has been used therapeutically for the treatment of arteriosclerosis and rheumatoid arthritis (Cohn et al., 1999; Mohacsi et al., 1997), and to prevent organ rejection following kidney and heart transplantation (Weber et al., 1998).

1.1.2.6. Roquefortine C (RQC) RQC is a major mycotoxin produced from P.Roqueforti, which is used to manufacture

‘Roquefort’ or ‘blue’ cheese. Other Penicillium spp. found in blue cheese or milk, including P commune, P crustosum, P. chrysogenum, P. cyclopium and P. glabrum are also capable of producing RQC (Häggblom, 1990; Rand et al., 2005; Wagener et al., 1980). The presence of RQC is very common in blue cheese products, yet it is considered to be safe for human consumption

due to the low exposure range (0.05 - 10 mg/kg food) and its low toxicity (Finoli et al., 2001). In contrast, however, RQC is found at relatively higher concentrations (up to 17 mg/kg) in moldy animal feed grains and silage, especially in the maize-based silage (Auerbach et al., 1998;

Häggblom, 1990a; Ohmomo et al., 1994).

Despite its relatively low toxicity, it has been reported that natural RQC exposure concentrations may be high enough to cause potential toxicity to both monogastrics (Puls and

Ladyman, 1988) and ruminants (Häggblom et al., 1990b). The LD50 for RQC in the mouse by intraperitoneal exposure was previously estimated to be 20 mg/kg (Ohmomo et al., 1975), but

Tiwary et al. (2009) later re-determined the LD50 via intraperitoneal exposure to be between 169-

184 mg/kg for mice. The pathology of RQC toxicity includes paralysis, trembling and lethargy in canine and murine species (Rand et al., 2005; Scott and Kennedy, 1976; Wagener et al., 1980).

There are no reports of lethality in ruminant species, but there is a study reporting various clinical signs, including paresis, indigestion, diarrhea as well as increased rate of mastitis and abortion from the dairy cows fed 25 mg/kg of RQC (Häggblom et al., 1990b).

1.1.3. Effect of PMs on physical barriers Epithelial cells and commensal microbes create an important physical barrier at mucosal surfaces that protect the host against pathogenic microbes and toxins. Epithelial cells form a mechanical and bioactive barrier providing attachment surfaces for commensal microbes, while the commensal microbes maintain a symbiotic relationship with the host to construct a protective barrier between the outside gut environment and host immune system (Yan and Polk, 2004; Yu et al., 2012). PMs could potentially disrupt these physical barriers by altering commensal microbial populations and destabilizing the epithelial cell barrier.

1.1.3.1. Disruption of commensal microbial populations Hundreds of different microbial species reside within the GIT of monogastrics and ruminants where they maintain a symbiotic relationship with the host. These commensal microbes form a barrier on epithelial surfaces preventing attachment and colonization of pathogenic microbes

(Croswell et al., 2009). Likewise, the biodiversity of these commensal populations is crucial for maintaining gut homeostasis, integrity and normal function of immune system (Magalhaes et al.,

2007; Tappenden and Deutsch, 2007) (Figure 1.1A). For example, colonization of the gut lumen with Escherichia coli, Klebsiella pnemoniae, and Streptococcus viridans increases paracellular permeability, whereas, Lactobacillus brevis has the opposite effect (García-Lafuente et al., 2001).

It is hypothesized that when commensal microbial populations are altered, protective and homeostatic interaction between host and commensals are impaired, and this can lead to disease, such as inflammatory bowel disease (Schütze et al., 2010; Wagner et al., 2002).

The biodiversity of commensal microbiota is especially important for the development and maturation of neonatal immune system. Rapid colonization of the neonatal gut by commensal bacteria for example, leads to hypertrophy of Peyer’s patches, and attenuates the responsiveness of B cells to produce secretory IgA against commensal microbes (Shroff et al., 1995). Furthermore, specific commensal bacterial species promote the development of mucous-associated lymphoid tissues (Rhee et al., 2004), as well as immune system maturation to facilitate immunological tolerance to mucosal commensals and effector function against pathogenic microbes (Tourneur and Chassin, 2013). Interestingly, both delayed and inappropriate colonization of mucosal surfaces appears to be a risk factor for neonatal disease (Tourneur and Chassin, 2013).

Several PMs, such as CIT, PAT, MPA, PA and RQC, possess antimicrobial properties (Aninat et al., 2001; Kavanagh, 1947), which could disrupt commensal biodiversity and population numbers (Figure 1.1B). For example, CIT has an antimicrobial effect against several pathogenic

strains of bacteria and fungi (Devi et al., 2009), while PA has antimicrobial activities against Gram- positive bacterial species (Kavanagh, 1947). The antimicrobial properties of PMs are probably an evolutionary strategy by which Penicillium molds protect themselves against other microbes and reduce competition with other molds within their surrounding environment. Although it is generally assumed that many PMs are rapidly metabolized by commensal microbes (Abrunhosa et al., 2010; Fuchs et al., 2008; Kiessling et al., 1984), their antimicrobial properties could possibly attenuate microbial detoxifying mechanisms. Furthermore, PMs may also disrupt the normal colonization or maintenance of commensal microbes, which could compromise host protection against pathogenic microbes (Croswell et al., 2009; Stecher et al., 2007) and toxins (Pamer, 2007;

Tesh et al., 1994). In support of this hypothesis, enteropathogenic Salmonella Typhimurium was shown to colonize the murine intestine more efficiently when it was not competing with commensal microbiota (Stecher et al., 2007). Additionally, when commensal microbes are disrupted, bacterial toxins from E. coli O157:H7 (Tesh et al., 1994) and endotoxins from

Clostridium difficile (Pamer, 2007) can cause systemic pathology.

1.1.3.2. Physical disruption of epithelial surfaces PMs can also directly disrupt the integrity of epithelial surfaces increasing risk of secondary disease. For example, Maresca et al. (2001) reported a significant reduction in the transepithelial resistance (TER) of HT-29-D4 and Caco-2 cells, which is an indicator of epithelial barrier damage, at OTA concentrations greater than 1 μM. Another study also observed that PAT destabilized epithelial cell membrane cysteine thiol and sulfhydryl groups that compromised HT-29-D4 and

Caco-2 cell integrity and reduced TER (Mahfoud et al., 2002). This epithelial barrier damage increases potential uptake of foreign substances such as PMs and pathogenic microbes into the host systemic circulation, which could predispose animals to disease (Maresca et al., 2008).

1.2. Penicillium myotoxins (PMs): immunomodulation and immunotoxicity In general, the paradigm in immunology categorises the immune system into two branches: innate and adaptive immunity for ease of understanding (Murphy et al., 2008), even though some common cells and molecules are involved in both types of immunity. Innate immunity is characterized as the first line of defence during microbial infection after physical barriers have been penetrated. It generally involves immune surveillance, inflammation and immune regulation, all of which are considered non-specific immune responses. Innate immunity is triggered by the recognition of pathogens by humoral and cellular pattern recognition receptors (PRRs) that recognize pathogen-associated molecular patterns (PAMP) with limited specificity and affinity.

Adaptive immunity, on the other hand, recognises and targets pathogens with higher specificity and affinity. In this case, pathogen antigen recognition is mediated by either T-cell receptors on T-lymphocytes, or membrane-bound immunoglobulins on B-lymphocytes and antibodies secreted by plasma cells. Adaptive immunity is most often distinguishable from innate immunity by its ability to induce a memory response that is characterized as having a more rapid response with higher specificity and affinity during subsequent pathogen exposure.

PMs can potentially modulate molecules and cells involved in both innate and adaptive immune responses in many different ways. For example, most of the PMs that will be discussed in this review affect leukocyte cell numbers, which could be due to cell death, and changes in cell proliferation, differentiation and trafficking throughout the body. Furthermore, some of these PMs may also alter the function and expression of cytokines, inflammatory molecules, as well as the expression of immunoglobulins, which will be described in more detail in the following sections.

PMs have also been reported to alter the morphological structure and function of lymphoid organs such as thymus, spleen and lymphnodes, which are important for the development, and maintenance, and activation of adaptive immune system.

1.2.1. Ochratoxin A (OTA) 1.2.1.1. Leukocyte number OTA is a potent immunomodulatory mycotoxin that can reduce circulating numbers of leukocytes leading to leukopenia (Chang et al., 1979; Stoev, 2010). For example, OTA decreased the total number of blood monocytes and lymphocytes in a dose-dependent manner in broiler chickens (Chang et al., 1979), and a significant decrease in the number of blood neutrophils, basophils, and monocytes was also observed in mice fed 5 mg/kg of OTA in feed (Gupta et al.,

1979). This decrease in the blood leukocyte number was partially due to the reduced production of leukocytes in lymphoid organs, as OTA targets the primary lymphoid organ, such as bone marrow, thymus and bursa of Fabricius, as well as secondary lymphoid organs, including spleen and lymph node of various species (Boorman et al., 1984; Stoev, 2010; Stoev et al., 2004; Xue et al., 2010). For instance, OTA decreased the number of bone marrow pluripotent stem cells that are required for generating leukocytes such as granulocyte macrophage progenitors, as well as both

CD4+ and CD8+ T-cell subpopulations in spleen of mice (Alvarez-Erviti et al., 2005; Boorman et al., 1984). OTA reduces proliferation and viability of these immune cells, leading to the overall decrease in the number of blood and tissue leukocytes (Charoenpornsook et al., 1998). In support of this, OTA prevented the lymphoblastogenesis of porcine and human lymphocytes (Lea et al.

1989; Harvey et al. 1992). Charoenpornsook et al. (1998) also reported that 1.5 μl/ml of OTA inhibited concanavalin A (Con A)-, phytohaemagglutinin (PHA)- and pokeweed mitogen (PWM)- induced bovine peripheral blood mononuclear cell (PBMC) proliferation by 50% (IC50), and similar IC50s have been reported for bovine lymphocytes (Keblys et al., 2004; Stec et al., 2008).

Lea et al. (1989) looked more specifically at the effect of OTA on B cell proliferation, which was induced with polyclonal rabbit anti-human IgM antibodies in the presence of bovine B-cell growth factor (BCGF-1); they reported a significant inhibitory effect of OTA (50 μM) on B-cell

proliferation. We have also observed an inhibitory effect of OTA on the proliferation of a bovine macrophage cell line (BoMacs) after the concentration of 5.2 μM (Oh et al., 2012). In contrast,

OTA concentration between 0.5 and 20 μM failed to inhibit the proliferation of murine lymphocytes induced by either Con A or lipopolysaccharide (LPS) (Alvarez-Erviti et al., 2005).

These contradictory results may be due to the species variation.

1.2.1.2. Inflammatory response and cell death Several studies have demonstrated the OTA as a potential inducer of inflammatory molecule production by macrophages. Ferrante et al. (2006, 2008) for example, observed a significant increase of pro-inflammatory molecules such as cyclooxygenase-2 (COX-2), inducible nitric oxide synthase (iNOS) and heat shock protein 72 (hsp72) from murine macrophages and the macrophage cell line J774A. These inflammatory molecules, especially COX-2, stimulate macrophage tumor necrosis factor (TNF)-α production, which is a potent pro-inflammatory cytokine (Al-Anati et al.,

2005). These studies have also found a strong positive association between the OTA-induced pro- inflammatory mediators and the activity of nuclear factor (NF)-B, a transcription factor that stimulate cellular pro-inflammatory response (Al-Anati et al., 2005; Ferrante et al., 2008;

Lawrence, 2009). For example, OTA induced p65-NF-β protein expression in the macrophage cell line (Al-Anati et al., 2005), and these authors also showed that inhibition of NF-β blocked

TNF-α release that was induced by OTA from perfused rat liver. Ferrante et al. (2008) later demonstrated that OTA induced the degradation of cytosolic inhibitor of NF-β (Iβ-), which allows NF-β for translocation into nucleus and increased expression of pro-inflammatory molecules and cytokines (Figure 1.2). In addition to this, TNF-α also activates the Jun N-terminal kinase (JNK)-p38 mitogen-activated-protein kinase (MAPK) signaling pathway (Deng et al.,

2003), which further enhances the pro-inflammatory response (Elsea et al., 2008), and this was

also previously observed with both canine renal epithelial cells and rat proximal tubular cells

(Gekle et al., 2000; Sauvant et al., 2005).

Increased TNF-α production by OTA not only stimulates pro-inflammatory response, but it also affects cell viability. Interaction between TNF-α and membrane-bound TNF-α receptors-1

(TNFR-1) for example, activates TNF receptor-1-associated death domain (TRADD), which then triggers caspase-dependent apoptosis as indicated by a decrease in Bcl:Bax family ratio, the activation of caspases, and the release of mitochondrial protein cytochrome C (Aggarwal et al.,

2000; George et al., 2010). It has been proposed that OTA reduces the expression of anti-apoptotic protein Bcl-xL, which results in the release of cytochrome C as well as the activation of caspase-3 and -9. These activated caspases trigger a loss in mitochondrial trans-membrane potential leading to apoptotic cell death (Assaf et al., 2004; Chao et al., 1995; Chen et al., 2007). Activation of the

JNK-p38-MAPK pathway by TNF-α further enhances the cell death process (Deng et al., 2003;

Gekle et al., 2000). Lastly, OTA was also shown to induce apoptosis, independent of interaction between TNF-α and TNFR-1 in hepatocytes, which appears to be due to intracellular oxidative stress as indicated by increased production of H2O2, which then triggers apoptosis (Buttke and

Sandstrom, 1994; Essid et al., 2012).

1.2.1.3. Effect on adaptive immunity In addition to stimulating the pro-inflammatory response and death of leukocytes, OTA also modulates other immune cell functions, specifically down-regulating the cell-mediated immune response (CMIR). Harvey et al. (1992) reported that feeding 2.5 mg of OTA to growing gilts during a one-month period decreased the CMIR, including macrophage phagocytic activity, cutaneous basophil hypersensitivity, as well as delay-type hypersensitivity response.

The decreased CMIR may be partially attributed to the decreased total number of T-cells since

OTA inhibits T-cell proliferation and viability (Alvarez-Erviti et al., 2005; Chang et al., 1979;

Keblys et al., 2004). However, several other studies also discussed that the selective suppression of the CMIR by OTA exposure might have also occurred via the inhibition of both type-I and type-

II interferon (IFN) production (Luster et al., 1987; Xue et al., 2010). The type-I IFNs, IFN-α and

IFN-β, are potent activators of CMIR, and their suppression decreases NK cell anti-viral and antitumor activity (Swann et al., 2007). Type-2 IFN-γ is a crucial cytokine for promoting cytotoxic T- cells (CTLs) and macrophage activation (Schroder et al., 2004; Wigginton et al., 2001). Luster et al. (1987) reported that oral exposure of mice to OTA at the doses between 3.4 to 13.4 mg/kg decreased the serum concentration of total IFNs in a dose-dependent manner. They also observed selective inhibition of natural killer (NK) cell anti-tumor activity against murine mastocytoma

P815 tumor cells, but not anti-tumor activity of either blood CTLs or macrophages. This finding indicates that OTA may specifically inhibit the production of type-I IFNs in mice. However, a much lower concentration of 0.25 mg/kg OTA, when co-exposed with T-2 toxin, led to a significant decrease in serum IFN-γ concentration as well as a slight reduction in the splenic gene expression of IFN-γ in broiler chickens (Xue et al., 2010). Furthermore, in vitro exposure of OTA significantly inhibited not only the cytotoxic activity of NK cells but also the cytotoxic activity of

CTLs against mouse lymphoma Yac-1 target cells (Alvarez-Erviti et al., 2005). Oral administration of 0.05-0.45 mg/kg of OTA to male rats also decreased the cytotoxic activity of CTL and splenocytes against sheep red blood cells (sRBC) (Alvarez et al., 2004). Based on these studies, it can be suggested that OTA suppresses not only type-I IFN, but may also suppress the IFN-γ that is important for stimulating the cytotoxic activity of CTL (Wigginton et al., 2001).

Some studies have reported that OTA also modulates the antibody-mediated immune response

(AbMIR). For example, intraperitoneal injection of 1 μg/kg of OTA significantly decreased sRBC- specific IgG and IgM responses by 90% in mice (Creppy et al., 1983). This decrease may be attributed to a decrease in the total number of B-cells, since OTA was shown to decrease lymphocyte numbers by inhibiting cell proliferation as discussed earlier (Lea et al., 1989). In contrast to the previous study, the oral administration of 2.5 mg/kg OTA to pigs did not affect total immunoglobulin concentrations, or the specific AbMIR to chicken RBCs (Harvey et al., 1992). It is possible that intraperitoneal injection of OTA, used in the Creppy et al. (1983) murine study, might have bypassed microbial and hepatic metabolism, which were previously shown to detoxify mycotoxins (Reddy et al., 1982; Thust and Kneist, 1979)

1.2.2. Citrinin (CIT) 1.2.2.1. Leukocyte number Similar to OTA, CIT has significant effects on leukocyte numbers. For example, CIT exposure reduced the proliferation of porcine and bovine T-lymphocytes, in vitro (Keblys et al., 2004; Stec et al., 2008). The CIT doses of 5 or 10 mg/kg BW via oral gavage also decreased the number of murine splenic macrophages, CD8+ CTLs and CD4+ T-helper cells, while the total number of B cells was reduced in intestinal intra-epithelium, lamina propria and Peyer’s patches of BALB/c mice (Islam et al., 2012). In contrast, these authors also observed that the CIT exposure increased the number of CTLs, T-helper cells, and CD4+CD25+Foxp3+ regulatory T-cells in mesenteric lymph nodes, as well as intestinal CTLs in the intra-epithelium, lamina propria and Peyer’s patches of mice. Another in vivo study reported similar results, where they observed T-cell blastogenesis was stimulated in the spleens of the mice that received CIT at the doses higher than 0.12 mg/kg of ip injection (Reddy et al., 1988). Islam et al. (2012) suggested that the increase in T-cell population

may be due to CIT acting as a T-cell-specific mitogen, activating the NF-β or MAPK. However, they also stated that CIT could also stimulate the migration of T-cells from spleen to other immune tissues, which would explain the increased T-cell numbers that they observed in most secondary lymphoid organs, and reduced numbers of T-cells in the spleen. This seems to be a suitable explanation for the in vivo studies; however, it does not explain the suppressed T-cell proliferation observed in vitro, since CIT should have induced the T-cell proliferation if CIT was indeed a T- cell mitogen. Therefore, there seems to be more complex interaction among T-cells and other cell types when animals are exposed to CIT as compared to in vitro exposure. In support of this,

Dziarski (1985) previously reported a higher murine lymphocyte mitogenic response in vivo than in vitro following exposure to various types of mitogens. Furthermore, the effect observed in the in vivo studies may not be due to CIT in its original form, but rather the effect of its metabolites.

CIT is metabolized in the liver and possibly other organs, and these metabolites may have a different potency and toxic effects in animals (Reddy et al., 1982; Thust and Kneist, 1979). It is also possible that the mitogenic or inhibitory response is dose-dependent, where a low concentration of CIT could have resulted in hormetic effect in vivo. Hormesis is a biphasic concentration-response relationship manifesting as a stimulating response at low concentrations, followed by an inhibitory effect at high concentrations (Calabrese and Baldwin, 2003).

1.2.2.2. Oxidative stress and cell death Regardless of potential mitogenic activity of CIT on lymphocyte proliferation, CIT can also decrease total leukocyte numbers by inducing cell death. Similar to OTA, CIT triggers the apoptosis of leukocytes by inducing caspase pathways (Kumar et al., 2011), and this is probably initiated by increased intracellular oxidative stress.

CIT induces oxidative stress by decreasing the level of intracellular anti-oxidant, glutathione

(GSH) (Kumar et al., 2011), and this in turn reduces other GSH-dependent anti-oxidant enzymes such as GSH reductase (GSSG) and trans-hydrogenases (Ribeiro et al., 1997). Intracellular glutathione peroxidase (GSH-Px) functions as an anti-oxidant enzyme that inhibits the production of reactive oxygen species (ROS) molecules, such as superoxide anion and hydrogen peroxide

(Papp et al., 2012). Oxidative stress subsequently stimulates the production of superoxide anion and ROS via the mitochondrial respiratory chain pathway (Chagas et al., 1992), while at the same time arrests the G0 to G1 phase of cell cycle, reduces Bcl-2/Bax activity ratio, and induces cytochrome C release from mitochondria; all of which eventually stimulate and sensitize cells to the p53-mediated caspase-3 and -9 apoptotic death signal (Basu and Haldar, 1998; Islam et al.,

2012; Kumar et al., 2011). CIT has also been shown to reduce the gene and protein expression of the major inflammatory cytokines interleukin (IL)-1β, IL-6 and TNF-α in human Mono-Mac-6 cells (Johannessen et al., 2005) and murine macrophages (Islam et al., 2012). In contrast, Liu et al.

(2010) observed CIT inhibited the Janus kinase (JAK)/signal transducers and activators of transcription (STAT)-1α and NF-κB signaling pathways, leading to the inhibition of iNOS and nitric oxide (NO) production in a glomerular mesangial cell line in response to LPS and IFN-γ exposure. Therefore, CIT appears to induce leukocyte oxidative stress (Kumar et al., 2011; Ribeiro et al., 1997), independent of NO production, which in turn activates the intrinsic apoptosis pathway that involves stimulation of p53-mediated caspase-3 and caspase-9 (Chan, 2007; Islam et al., 2012;

Kumar et al., 2011).

1.2.2.3. Effect on adaptive immunity CIT appears to affect the balance of T helper cell type (Th)1 and Th2 cytokines that support

CMIR and AbMIR, respectively. An in vitro exposure study for example, reported that CIT

suppressed both IFN-γ and IL-4 production from CD3- and CD28-stimulated human blood mononuclear cells in a dose-dependent manner (Wichmann et al., 2002); the IC50 for cytokine production was 21.6 μg/ml for IL-4, however, it was only 8.3 μg/ml for IFN-γ production. This suggests CIT more severely affects IFN-γ production than IL-4 production thereby favouring

AbMIR over CMIR at low exposure concentrations. Taken together, the suppression of iNOS and

IFN-γ production by CIT has the potential to compromise host defence against intracellular pathogens (Liu et al., 2010). In support of this, Herzog-Soares and Freire (2004) demonstrated that when murine macrophages were pre-exposed to 10 µg/ml of CIT, the infectivity of the cells by

Toxoplasma gondii tachyzoites (T. gondii) increased by 18%, and also enhanced the replication of the tachyzoites within the cells.

At higher concentrations, CIT also appears to suppress AbMIR, since CIT was shown to decrease total B-cell number as well as IL-4 production at higher concentrations (Islam et al., 2012;

Wichmann et al., 2002). Furthermore, 2.5 mg/kg of CIT ip injection decreased both sRBC-specific antibody and complement activity by 75-87.5% and 87.5-93.8%, respectively, in albino mice

(Carvalho et al., 2005). Similar results were also observed in Wistar rats that were fed 10 mg/kg of CIT when sRBC- and ovalbumin-specific antibody responses were measured (Singh et al.,

2011).

1.2.3. Patulin (PAT) 1.2.3.1. Leukocyte number PAT is a potent immunotoxin that can severely alter the number of leukocytes in blood and tissues, and modulate host immune function. In vitro studies have demonstrated that PAT reduces the proliferation and viability of leukocytes, specifically T-cells and macrophages (Keblys et al.,

2004; Oh et al., 2012; Sorenson et al., 1985). For example, Keblys et al. (2004) observed dose-

dependent decrease in the proliferation of blood lymphocytes when porcine T-cells were exposed to PAT in vitro.

In vivo studies have also demonstrated similar effects on the number of blood leukocytes from mice exposed orally to PAT, by both acute and chronic exposure. Escoula et al. (1988) for example, reported a significant decrease in total blood lymphocyte numbers in Swiss mice fed 10 mg/kg of

PAT for 5 to 10 days. A similar observation was also reported with B6C3F mice that consumed

PAT at concentrations greater than 1.28 mg/kg for 28 days, where PAT decreased the number of total blood leukocytes by 30% (Llewellyn et al., 1998). These authors also reported that PAT increased the number of certain subsets of immune cells at a predicted human exposure level of

0.08 – 2.56mg/kg. For example, PAT after a 28 day exposure period, increased the number of splenic CTLs, NK cells and monocytes by 50%, 30% and 24%, respectively, while the number of

CD3+T-cells, CD4+T-cells, and MAC-3+ monocytes remained unchanged (Escoula et al., 1988;

Llewellyn et al., 1998). Similar to the 28-day chronic exposure study, acute 4-day exposure of mice to 10 mg/kg PAT via stomach tubing, and intraperitoneal injection of 2.5 mg/kg PAT in rabbit did not affect the number of either splenic neutrophils or CTLs, but increased the number of splenic T-cells in both mice and rabbit even though the mitogenic response of T-cells and the number of splenic B cells were suppressed (Escoula et al., 1988). Furthermore, another in vivo study reported that 0.1 mg/kg of PAT for 90 days significantly increased apoptosis of rat thymic dendritic cells (DCs) that are required for normal T-cell development (Ozsoy et al., 2008).

1.2.3.2. Oxidative stress and cell death In addition to the above reported cytotoxic effects of PAT on rat thymic DCs (Ozsoy et al.,

2008) , in vitro studies have also shown PAT-induced cytotoxicity to rat alveolar macrophages

(Sorenson et al., 1985), and a bovine macrophage cell line (BoMacs) (Oh et al., 2012). Similar to

OTA and CIT, PAT appears to induce p53-mediated apoptotic cell death triggered by caspase-3 and -9 pathways, which was observed in both murine skin cells (Saxena et al., 2009), and human leukemia HL60 cells (Wu et al., 2008). PAT acted as a strong inhibitor of inflammatory cytokines, suppressing the production of IL-1β, IL-6 as well as TNF- and TNF-β from human mononuclear cells and T-cells (Luft et al., 2008). However, PAT also induced ROS production leading to intracellular oxidative stress, which then stimulated the caspase pathways to induce cell apoptosis

(Liu et al., 2007; Papp et al., 2012).

The mechanism by which PAT induces oxidative stress appears to resemble that of CIT, where both mycotoxins have been shown to inhibit anti-oxidant enzymes, specifically targeting the intracellular GSH-dependent enzymes, such as GSSG (Luft et al., 2008; Papp et al., 2012; Ribeiro et al., 1997). Although it is still unclear for CIT, it appears that PAT destabilizes the disulfide bonding of cysteine within GSH (Papp et al., 2012; Schütze et al., 2010). PAT-induced ROS production damages DNA arresting the cell cycle, and this would eventually initiate caspase apoptotic death pathway by inducing Bax pro-apoptotic protein expression as well as p21 and p53

(Saxena et al., 2009; Wu et al., 2008), subsequently stimulating the apoptosis of leukocytes.

1.2.3.3. Effect on adaptive immunity Intracellular GSH levels have also been explored as a marker to determine the polarization of

DCs, with reduced GSH levels being a marker of DCs of the Th2 phenotype that support AbMIR

(Peterson et al., 1998). Studies have reported that PAT decreased intracellular GSH levels in human PBMCs, T-cells as well as murine DCs (Luft et al., 2008; Schütze et al., 2010). These cells, at the same time, expressed a Th2 dominant phenotype in terms of their cytokine production. For example, studies have shown that PAT decreased the production of both IFN-γ and IL-12. These cytokines are crucial for promoting a Th1 response, and stimulate the CMIR. Impaired production

of these cytokines could increase host susceptibility to intracellular pathogens, such as T. gondii

(Hunter et al., 1994) or Mycobacterium (Wagner et al., 2002). In support of this, PAT at a concentration greater than 0.5 g/ml in media was shown to severely suppress the phagocytic and microbicidal activity of murine peritoneal macrophages to yeast Saccharomyces cerevisiae

(Bourdiol et al., 1990).

In contrast to its effect on the CMIR, PAT had less effect on the expression of Th2 cytokines, such as IL-5 and IL-13, from murine splenocytes and mediastinal lymph node cells (Schütze et al.,

2010). The production of IL-4 by human PBMCs on the other hand, was inhibited at 0.050 g/ml of PAT (Luft et al., 2008), but this study also observed greater suppression of IFN-γ production as compared to that of IL-4. In fact, a low concentration of 0.003 g/ml PAT, rather increased human

PBMC IL-4 production, but suppressed IFN-γ production suggesting a possible quadratic effect of

PAT on IL-4 production (Wichmann et al., 2002). PAT also suppressed the production of other cytokines, including IL-2 and IL-10, all of which are important for the survival and regulation of leukocyte function (Colic et al., 2003; Dooms et al., 2004).

Similar to CIT, PAT at high doses may also suppress the AbMIR by down-regulating plasma cell (PC) antibody production. For example, intake of 10 mg/kg of PAT by oral gavage and intraperitoneal administration of 2.5 mg/kg of PAT dramatically decreased total serum levels of IgG,

IgA and IgM without altering levels of serum albumin and complement component C3 (Escoula et al., 1988b). The reduction of these immunoglobulins may be attributed to B-cell apoptosis. Other studies using a lower dose of PAT exposure reported that PAT did not necessarily decrease immunoglobulin production. In fact, 28 days oral exposure to 0.08 – 2.56 mg/kg of PAT, unlike

CIT, did not alter the serum IgM response to sRBC in B6C3F1 (Carvalho et al., 2005; Llewellyn et al., 1998). Furthermore, the oral and nasal exposure to 5 – 20 g of PAT also increased

ovalbumin (OVA)-specific IgE and IgG1 in BALF/C mice without altering IgG2a levels (Schütze et al., 2010). The increased IgG1/IgG2a is also an indication of a Th2 response, suggesting that

PAT shifts the immune response toward Th2 (Schütze et al., 2010). T-cells on the other hand, appear to remain viable during PAT exposure, even though the capacity to produce certain cytokines, especially those that are important for CMIR, is severely compromised (Colic et al.,

2003; Dooms et al., 2004; Wichmann et al., 2002).

1.2.4. Penicillic acid (PA) 1.2.4.1. Leukocyte number There are only few studies available on the subject of PA immunotoxicity. PA may not necessarily alter immune function, yet it was shown to reduce the proliferation and viability of immune cells, decreasing the overall number of immune cells in vivo. For example, feeding a diet containing 2 mg/kg of PA to broiler chicks led to a severe decrease in the weight of both primary

(thymus and bursa of Fabricus) and secondary lymphoid organs (spleen), and this effect worsened during co-exposure with OTA (Stoev et al., 2004). The reduction in lymphoid organ size may be due to suppressed proliferation of immune cells such as macrophages and lymphocytes rather than the reduced viability, because a much higher PA concentration is required to induce cell death

(Keblys et al., 2004; Oh et al., 2012; Stoev et al., 2004). Stoev et al. (2004) also observed a dose- dependent decrease in the proliferation of poultry PBMCs, but they were unable to reproduce these results using human PBMCs in a later study (Stoev et al., 2009). The IC50s for PA based on cell proliferation have been determined to be 18.0 and 30.0 μM for porcine T-cells and bovine macrophages, respectively (Keblys et al., 2004; Oh et al., 2012).

1.2.4.2. Cell death via necrosis PA can also decrease the viability of immune cells, yet this occurs at much higher concentrations. Sorenson and Simpson (1986) for example, reported that concentrations of PA greater than 1.0 mM caused significant cytotoxicity to rat alveolar macrophages after 2 hours of exposure. PA concentrations greater than 160 μM also significantly decreased the viability of bovine macrophages when exposed for 24 hours, and the LC50 was determined to be 176.0 μM

(Oh et al., 2012).

In contrast to the PMs discussed earlier, the cytotoxicity induced by PA does not occur via apoptosis but rather necrosis, whereby PA reduces cell membrane integrity, leading to cytolysis of immune cells (Bando et al., 2003). Actually, concentrations of PA between 100 and 200 M effectively blocked the activity and self-activation of caspase-3, -8, and -9 in a lymphoma cell line, and the inhibition of these caspase pathways resulted in the decreased DNA fragmentation and apoptotic cell bodies. However, these concentrations at the same time caused unfavourable cell necrosis that reduced the overall cell viability (Bando et al., 2003).

PA could also affect other parameters of immune system, even though this has not widely been studied. Sorenson and Simpson (1986) reported the reduced phagocytic capacity along with the suppressed RNA and protein synthesis in rat alveolar macrophages that were treated with 0.3 mM PA for 2 hours. The metabolic rate of the human PBMC, on the other hand, was induced by

PA (Stoev et al., 2009). Changes in RNA and protein synthesis, as well as metabolic rate, could potentially affect the production of anti-microbial molecules produced by immune cells, however, we are unaware of any studies that have been carried out to assess the potential effect of PA related to these functional aspects.

1.2.5. Mycophenolic acid (MPA) MPA is a well-known potent immunosuppressive PM that has been effectively used to prevent organ rejection following transplantation (Borrows et al., 2006), as well as treat inflammatory disorders such as arteriosclerosis and rheumatoid arthritis (Cohn et al., 1999; Mohacsi et al., 1997).

For these reasons, MPA and its derivative mycophenolate mofetil (MMF) are by far the most widely studied PMs in terms of their immunomodulatory properties. Both functional and transcriptomic studies have revealed that MPA most frequently mediates alterations in cell proliferation, viability as well as the expression of co-receptors required for migration and activation of immune cells (Dubus et al., 2002; Dun et al., 2013; Hauser et al., 1997; Mezger et al.,

2008).

1.2.5.1. Innate immune cell proliferation The effect of MPA on cell proliferation and viability has been thoroughly studied in human medicine (Borrows et al., 2006; Cohn et al., 1999). A study performed by Borrow and colleagues

(2006) reported that when a high oral dose of MPA, up to 1.95 g/day, was given to patients following renal transplantation, there was a significant decrease in the number of blood leukocytes compared to the patients who received a low dose of MPA post transplantation. MPA and MMF likely cause a decrease in the proliferation of leukocytes resulting in reduced leukocyte numbers

(Dubus et al., 2002; Mezger et al., 2008; Ohata et al., 2011). For, example, MPA and MMF, at clinically effective concentrations between 0.6 and 10 μM in media, have been shown to inhibit the proliferation of rat and human mesangial cells without inducing cellular necrosis or apoptosis

(Dubus et al., 2002; Hauser et al., 1997). Other studies have demonstrated that MPA also inhibited the proliferation of different immune cell types including bovine macrophages, human monocyte- derived dendritic cells (mDC), polymorphonuclear cells and NK cells without causing apoptosis

or necrosis (Mezger et al., 2008; Oh et al., 2012; Ohata et al., 2011). The decreased number of these innate immune cells may partially explain why there was reduced rejection in transplant recipients following treatment with MPA.

1.2.5.2. Inhibition of NK cell activation NK cells are important mediators of the immune response and that limit the growth of tumors and spread of microbial infection (Ohata et al., 2011), and also contribute to allograft rejection during the first month of post-transplantation (Eissens et al., 2010; Kroemer et al., 2008). For these reasons, NK cells are one of the innate immune cell groups that have most widely been studied in the context of MPA immunosuppression. Many studies have shown that MPA decreases the activity of NK cells by inhibiting their activation (Ohata et al., 2011; Weigel et al., 2002). For example, when human NK cells were incubated with leukemic cells and exposed to 10 µg/ml of

MPA in vitro for 7 days, there was a significant decrease in their activation markers, NKG2D,

NKp30, NKp44 and NKp46 (Ohata et al., 2011). These NK cells also had reduced cytotoxic activity to leukemic cells, K562 and Daudi (Eissens et al., 2010; Ohata et al., 2011) Eissen et al.

2010 also reported that MPA decreased the shift in NK cell receptor repertoire, such as NKG2A, which might have partially contributed in the decreased cytotoxic activity of NK cells against

K562 (Eissens et al., 2010). The decrease in the activation markers from the studies indicate that

MPA has inhibitory effects on both the maturation and activation of NK cells, thereby reducing their cytotoxic activity. This is one of reasons why MPA has been used as an effective post- transplantation drug.

1.2.5.3. Changes in the cellular surface protein expression on dendritic cells (DCs) DCs are important regulators of the immune system as they are considered to be key mediators of naïve lymphocyte activation (Novak et al., 2010). These cells are highly susceptible to

immunomodulation by MPA. It is thought that MPA affects the expression of membrane proteins on DCs that are required for the activation and function of the host immune response (Cicinnati et al., 2009; Hauser et al., 1997). For example, one study demonstrated that in vivo exposure of MPA reduced the trafficking of human blood myeloid DCs to lymph nodes by down-regulating chemokine receptor (CCR)-7, but increased the migration of these cells to sites of inflammation by up-regulating CCR-1 expression (Cicinnati et al., 2009). MPA also simultaneously induced the expression of endothelial VCAM-1 and E-selectin, both of which are required for the trans- migration of leukocytes across endothelial cells during inflammation (Hauser et al., 1997).

Therefore, these studies suggest that MPA stimulates the recruitment of mDCs towards sites of inflammation.

MPA also inhibits the maturation of DCs thereby limiting their ability to stimulate other immune mediators, such as T and B cells (Cicinnati et al., 2009; Hauser et al., 1997). The surface proteins on DCs play an essential role in lymphocyte activation by acting as co-receptors during antigen presentation to T and B cells (Novak et al., 2010). Therefore, the down-regulation of the co-receptors would lead to incomplete activation of lymphocytes, and reduce the function of lymphocytes (Daniel et al., 1997; Vogel et al., 2013). For example, the study performed by

Cicinnati and colleagues (2009), demonstrated that MPA inhibited the maturation of human mDCs by impairing TLR-3 ligation to polyinosinic-polycytidylic acid, a synthetic double stranded RNA used as a viral antigen to activate innate cells. Other in vitro studies also reported the down- regulation of human mDC maturation markers, such as CD80 and CD86, as well as other costimulatory and adhesion molecules, CD1a, CD40, CD54, CD83 and CD 205 (Colic et al., 2003;

Mezger et al., 2008; Wadia et al., 2009). The in vitro exposure of MMF also down-regulated the expression of co-receptors, CD40, CD80/86, and ICAM-1, on murine DCs (Mehling et al., 2000).

DCs lacking in the above co-receptors would not be able to efficiently activate lymphocytes, and therefore, would affect the maturation and function of both B- and T-cells (Daniel et al., 1997;

Vogel et al., 2013; Wykes and Macpherson, 2000). The absence of CD40 signal from DCs for example, impairs the proliferation as well as maturation of B-cells (Wadia et al., 2009; Wykes and

Macpherson, 2000). Additionally, incomplete activation of T-cells either stimulates apoptosis of the T-cells, or favors their differentiation into T-reg cells (Daniel et al., 1997; Vogel et al., 2013).

In support of this, human DCs, when pre-treated with MPA, stimulated the differentiation of naïve

T-cells to T-reg cells with increased production of IL-5 and anti-inflammatory cytokines, IL-10 and TGF-β (Colic et al., 2003; Lagaraine et al., 2008). IL-10 and TGF-β are strong immunosuppressive cytokines that are required for the differentiation of T-reg cells (Taylor et al.,

2006). Interestingly, direct exposure of naive and memory human T-cells with 1 µM MPA also induced the differentiation to T-reg cells, as indicated by the increased expression of the T-reg cell markers, Foxp3, CTLA4 and PD-1 (Deng et al., 2012; He et al., 2011). Based on these studies, it is speculated that MPA increases the T-reg cell population, via DC dependent and independent signals. Since T-reg cells have strong immunosuppressive phenotypes (Sakaguchi, 2004), an increase in their number likely leads to immunosuppression, and make MPA an effective compound for preventing allograft rejection (Borrows et al., 2006; Mohacsi et al., 1997).

1.2.5.4. Proliferation and viability of T-cells It appears that the immunosuppressive effect of MPA may occur not only through the up- regulation of T-reg cell differentiation but also by reducing the number of T-cells (Chapuis et al.,

2000; Keblys et al., 2004; Quéméneur et al., 2002). For example, a patient treated with 2 g/day of

MPA for a year following organ transplantation had reduced numbers of total T-cells in their

peripheral blood (Weigel et al., 2002). This decrease in lymphocyte number is partially due to

MPA’s ability to inhibit cell proliferation as well as its cytotoxicity to T-cells (Chapuis et al., 2000).

It has been proposed that the selective effect of MPA on proliferation and viability of activated

T-cells is closely linked to down-regulation of type II inosine monophosphate dehydrogenase

(IMPDH-II) activity, which inhibits de-novo purine biosynthesis (Dun et al., 2013; Hauser et al.,

1997; Nagai et al., 1992). There are two isotypes of IMPDH, type I and II. Both isotypes have similar enzymatic function for purine biosynthesis, yet they are differentially expressed in different cell types. For example, Nagai and colleagues (1992) demonstrated that activated T-cells and

HL60 leukemic cells predominantly express type II IMPDH, while type I IMPDH was more constitutively conserved across most other cell types. Regardless of their similar functionality, the activity of type I IMPDH cannot be substituted for type II IMPDH (Gu et al., 2000; Jain et al.,

2004). Activated T-cells and leukemic cells rely heavily on type II IMPDH for GTP synthesis, while naïve T-cells and other cell types, such as macrophages or epithelial cells, actively utilize type I IMPDH (Karnell et al., 2011; Nagai et al., 1992). MPA was shown to specifically down- regulate type II IMPDH and had no effect on Type I IMPDH (Hauser et al., 1997; Nagai et al.,

1992), and this would allow MPA to selectively target the proliferation and viability of activated lymphocytes or other leukemic cells. Inhibited purine biosynthesis compromises the cell’s ability to synthesize GTP/GDP, which is essential for normal immune cell functions such as differentiation, proliferation and apoptosis (Nagai et al., 1992; Yalowitz and Jayaram, 2000). The decreased expression of GTP following MPA treatment partially led to the reduction of the ERK

1/2 and MAPK pathways, which could reduce the all viability by activating capase apoptotic death signals (Allan et al., 2003; Ha et al., 2006). For example, MPA induced the cell death of the

MOLT-4 T-lymphocytic cell line, as well as PHA-activated CD4+ T-cells, but not resting naïve T-

cells (Chapuis et al., 2000; Cohn et al., 1999). Other studies have reported that MPA induced apoptosis in the Jurkat human T-cell line and gastric cancer cell line, as indicated by increased caspase-3 and -9 activity (Choi et al., 2013; Dun et al., 2013).

The depletion of cellular GTP/GDP not only trigger apoptosis, but also causes necrotic cell death in some immune cell types (Chaigne-Delalande et al., 2008; Guidicelli et al., 2009). Chaigne-

Delalande and colleagues (2008), demonstrated that MPA caused cytotoxicity to B- and T- lymphoma cell lines as well as mitogen-activated human PBLs by necrosis or caspase-dependent apoptotic pathways. MPA did not necessarily induce either apoptosis or necrosis to naïve PBLs, but rather cause them to become sensitized to MPA-mediated necrosis upon activation (Chaigne-

Delalande et al., 2008). Necrotic cell death was also demonstrated using the JurkatDoxR and

CEMDoxR lymphocytic leukemia cell lines, which are resistant to doxorubicin-induced apoptosis

(Guidicelli et al., 2009). Therefore, MPA specifically targets activated lymphocytes that heavily rely on type II IMPDH activity for GTP/GDP synthesis, causing either apoptosis or necrosis.

1.2.5.5. Effect on B-cell activation Some studies have also investigated the potential effect of MPA and MMF on the activities of

B-cells (Eickenberg et al., 2012; Karnell et al., 2011). Weigel and colleagues, 2002, reported that

MMF treatment resulted in almost a 50% reduction in B-cell numbers in patients who received

MMF for a year (Weigel et al., 2002). The study also reported that these B-cells from MMF-treated patients were lacking the B-cell activation marker CD69. Karnell et al. (2011) also noted that in vitro MPA exposure to concentrations between 1.0 and 3.0 μM selectively impaired the early activation of human B-cells. Similarly, another study reported significant decreases in the activation as well as differentiation of human B-cells when these cells were exposed to 5 μM MPA in vitro (Eickenberg et al., 2012).

The decreased activation and differentiation of B-cells would decrease the number of plasma cells (PCs), terminally differentiated effector B cells that have a high capacity for producing antibodies (Murphy et al., 2008), subsequently decreasing total antibody levels. In support of this,

Eickenberg et al., 2012, reported that the patients who received an oral dose of MMF between 0.3 and 0.5 µM had significantly lower serum IgG concentrations compared to those from other treatment groups. In addition to this, Karnell et al. also observed that in vitro exposure to MMF significantly decreased the production of IgG, IgM and IgA from PCs.

Regardless of the significant effect of MPA on B-cell activation and differentiation

(Eickenberg et al., 2012; Karnell et al., 2011; Weigel et al., 2002), MPA does not seem to affect cell viability and/or the responsiveness of terminally differentiated human plasma cells (PCs).

Karnell et al. 2011 discussed that the selective effect of MPA on B-cells but not PCs is probably due to the differential utilization of IMPDH enzymes. B cells, like activated T-cells, are predominantly dependent on type II IMPDH activity for GTP/GDP, while terminally differentiated

PCs use type I IMPDH (Karnell et al., 2011).

1.2.6. Roquefortine C (RQC) 1.2.6.1. Immunomodulatory effect of RQC RQC is a common metabolite of P. roqueforti, which is required in the making of blue cheese for human consumption (Auerbach et al., 1998). In terms of the toxicity, RQC is probably the least toxic PM discussed in this review. However, studies have reported various toxicological signs such as paralysis, trembling and diarrhea are associated with RQC exposure to both monogastric (Rand et al., 2005; Scott and Kennedy, 1976; Wagener et al., 1980) and ruminant species (Häggblom et al., 1990), yet very few studies are available on its potential immunotoxicity (Keblys et al., 2004;

Rand et al., 2005).

Similar to the other PMs, RQC can also inhibit the proliferation of lymphocytes. For example, in vitro RQC exposure suppressed the proliferation of porcine lymphocytes, but to a much lesser extent when compared to other PMs, such as OTA, CIT or PAT (Keblys et al., 2004). In addition to this, Rand et al. 2005 observed that the intra-tracheal instillation of RQC that was dissolved in saline solution resulted in a significant increase in the numbers of macrophages and neutrophils in in bronchoalveolar lavage fluid (BALF) of mice after 6 hours of the RQC exposure (Rand et al.,

2005). There was also a significant increase in BALF inflammatory molecules including MIP-2 and TNF-α. Rand and colleagues in their later study suggest that fungal compounds, such as RQC, with low molecular weight may elevate inflammatory gene transcription in alveolar macrophages

(Rand et al., 2011), inducing pro-inflammatory molecules, such as MIP-2 and TNF-α (Rand et al.,

2005). Based from these studies, it can be speculated that RQC irritated the lung tissues to induce the infiltration of immune complexes, such as neutrophils and macrophages, and the elevation of other inflammatory mediators, such as TNF-α, in the mice BALF (Anderson et al., 2012; Rand et al., 2005). This however, would require more evidence to support the speculation at this point.

Considering that high concentrations of RQC (25 mg/kg) was shown to cause various toxicological signs in ruminant species (Häggblom et al., 1990) and the concentration of RQC in animal silage reported ranged from 0.4 to 17 mg/kg (Auerbach et al., 1998; Mansfield et al., 2008), it would be important to investigate the immunotoxicity of RQC.

1.3. Epigenetic changes and its relation to inflammatory response Epigenetics are heritable changes in gene expression caused by mechanisms other than changes in the underlying DNA sequence and these tend to follow environmental cues (Delcuve et al., 2009). These changes are often regulated through DNA and/or histone covalent modifications, such as methylation or acetylation (Fuks et al., 2003; Rice and Allis, 2001), and are

crucial for optimal cellular activity. The regulation of epigenetic modifications is important for the proper functioning of immune cells (Bowdridge and Gause, 2010). For example, M2-macrophage specific genes were regulated by reciprocal changes in the methylation status of lysine 4 and 27 on histone H3 (H3K4 and H3K27), both of which are regulated by epigenetic enzymes (Biel et al.,

2005).

As previously discussed, the exposure of Penicillium mycotoxins could lead to cellular and functional changes of the immune response (Oh et al., 2012; Sorenson and Simpson, 1986;

Sorenson et al., 1985). It has been proposed that some of these functional changes may be influenced by alterations in epigenetic patterns and the activities of the enzymes regulating them

(Marin-Kuan et al., 2007, 2008). Given this, it seems reasonable that exposure to mycotoxins may also alter the functioning of immune cells through epigenetic modifications (Bowdridge and Gause,

2010; Marin-Kuan et al., 2008). In support of this, Marin-Kuan et al. (2007) reported that chronic

OTA exposure resulted in an increased concentration of histone deacetylase (HDAC)-3 protein in mice kidneys. HDAC-3 is a class-1 histone deacetylase (HDAC) that regulates cell functions by deacetylating lysine residues in histone 3 and/or 4. The class-1 HDACs stimulate inflammatory activities of cells by interacting with signalling molecules involved in NF-β transcriptional pathway (Figure 1.2; Ashburner et al., 2001; Chen et al., 2001; Rahman et al., 2004). The inhibition of HDACs suppresses proteasome activity, and reduces the degradation iβ- (Al-Anati et al.,

2005; Place et al., 2005), nuclear translocation and overall activity of NF-β (Ferrante et al., 2008).

OTA was shown to reduce the level of cytosolic iβ- expression, and this in turn, elevated of nuclear p65- NF-β expression in rat liver (Al-Anati et al., 2005). Additionally, decreases in the activity of HDACs demonstrated impairment of NF-β signalling pathways and decreased inflammatory response in human myeloid dendritic cells (Song et al., 2011; Zhu et al., 2010). All

these studies indicate that OTA induces pro-inflammatory response by up-regulating the HDAC-

3 activity, which activates NF-β inflammatory pathways (Al-Anati et al., 2005; Ferrante et al.,

2008; Lawrence, 2009; Marin-Kuan et al., 2007).

Recently, many studies have shown that HDAC-3 can alter the stability of the STAT-3 dimer

(Sica and Bronte, 2007; Yuan et al., 2005). The STAT-3 dimer functions as a transcription molecule for differentiating M2 macrophages that exhibit anti-inflammatory phenotypes (Mullican et al., 2011; Stout et al., 2005). Upon the stimulation of macrophages with anti-inflammatory cytokines such as IL-4, the cytoplasmic portion of the IL-4 receptor acts as a kinase to phosphorylate cytoplasmic STAT-3 (Choi et al., 2013). The phosphorylated STAT-3 subsequently translocates into the nucleus with the help of importin- 3 (Liu et al., 2005). These phosphorylated

STATs then spontaneously bind to each other to form STAT-3 dimers (Liu et al., 2005). However,

Type-1 HDACs such as HDAC-3, deacetylate lysine685 of STAT-3 to destabilize its dimerization, which prevents M2 macrophage differentiation (Yuan et al., 2005). In support of this, Mullican et al. (2011) reported that the macrophages lacking HDAC-3 are hyper-responsive to IL-4 stimulation, resulting in IL-4-induced alternative activation of macrophages in both in vivo and in vitro. From these studies, it is speculated that the increased HDAC-3 activity by OTA could prevent the differentiation of macrophages into M2 macrophages (Stout et al., 2005).

Other epigenetic enzymes, such as DNA methyltransferases, histone demethylase and histone acetylase (Bmi-1), also play an important role in regulating cellular activities (Fatemi et al., 2002;

Subkhankulova et al., 2010). However, there is not enough information available on these enzymes in relation to Penicillium mycotoxicoses. Therefore, it would be interesting to look at the effect of

PMs on these enzymes, because changes in cell viability and proliferation are one of the most

common mechanisms of Penicillium mycotoxicoses as was previously discussed in this review

(Keblys et al., 2004; Oh et al., 2012; Stec et al., 2009).

1.4. Summary and Conclusion Overall, this review provides insight into the potential immunomodulatory effects of PMs through alterations in the viability, proliferation, inflammatory response, cell activation as well as both cytokine and antibody production of immune cells (Keblys et al., 2004; Oh et al., 2012; Xue et al., 2010). PMs differentially affect these parameters of immune response by directly changing cellular activities, such as caspase apoptotic pathways and/or NF-β transcriptional pathways

(Essid et al., 2012; Ferrante et al., 2006; Kumar et al., 2011). Some researchers have also proposed that PMs affect the cellular activity of immune cells indirectly as a result of alteration in the enzymes regulating epigenetic modifications (Marin-Kuan et al., 2007, 2008). Many in vitro studies and also some in vivo studies have provided the evidence of their immunomodulatory effects in poultry, rodent and human clinical models (Borrows et al., 2006; Dziarski, 1985; Keblys et al., 2004; Ozsoy et al., 2008; Stoev et al., 2004). Regardless of the fact that ruminants are constantly exposed to PMs from the consumption of ensiled feed (Mansfield et al., 2008), the potential toxic effect of PMs, including immunomodulation, have been disregarded under a general assumption that PMs are rapidly degraded by commensal microbes in the gut (Fuchs et al., 2008;

Kiessling et al., 1984). However, there are previously reported cases of various toxicological signs associated with the consumption of PM-contaminated silage in ruminant species (Ciegler, 1977;

Häggblom, 1990; Ukai et al., 1954). This suggests that PMs could potentially break through the protection from commensal microbes, potentially leading to immunomodulation, and therefore, possibly leading to secondary diseases, such as Johne’s disease, in ruminant species. In addition

to this, most studies discussed here focussed on the risk observed from individual PM exposure, which may not necessarily reflect the true risk of PMs due to their potential interactions among other PMs (Oh et al., 2012; Stoev et al., 2009; Stork et al., 2007; Veselá et al., 1983), because most

PMs are often found in combinations of two or more in animal feed (Mansfield et al., 2008).

Therefore, in order to gain a greater perspective on how to detect and prevent mycotoxins in the agricultural industry, combinational mycotoxicoses in ruminant species should be considered in future studies.

Table 1.1: The occurrence of Penicillium mycotoxins (PMs) in animal feed and food products.

location or source Converted mean Mycotoxins natural condition (nc), sample material method used concentrations references concentrations artificial condition (ac) (mg/kg)

Mexico, conditions created (nc) corn silage ELISA 5.1 ppb 0.0051 (Reyes-Velázquez et al., 2008) OTA swine feed 11.0 ug/kg 0.011 All regions from Portugal (nc) HPLC (Martins et al., 2012) laying hen feed 4.0 ug/kg 0.004 Calvados, France (nc) 15.0 ppb 0.015 (Garon et al., 2006) corn silage HPLC-MS CIT Normandy, France (nc) 32.0 ug/kg 0.032 (Richard et al., 2007) Local store, Portugal (nc) rotten apples TLC 0.55 mg/kg 0.550 (Martins et al., 2002) Across US states (nc) HPLC-MS 0.08 ug/g 0.080 (Mansfield et al., 2008) maize silage Germany (ac) HPLC; 2D-TLC 15.10 mg/kg 15.100 (Müller and Amend, 1997) PAT Local store, Protugal (nc) rotten apples TLC 20.45 mg/kg 20.450 (Martins et al., 2002) Local store, Japan (nc) apple juice LC-MS 4.74 ng/g 0.005 (Kataoka et al., 2009) PA Germany (ac) maize silage HPLC; 2D-TLC 3.06 mg/kg 3.060 (Müller and Amend, 1997) Across US states (nc) HPLC-MS 0.16 ug/g 0.160 (Mansfield et al., 2008) maize silage 690 ug/kg 0.690 MPA Bavaria, Germany (nc) LC-MS (Schneweis et al., 2000) grass silage 2200 ug/kg 2.200 Germany (ac) maize silage HPLC; 2D-TLC 3.56 mg/kg 3.560 (Müller and Amend, 1997) Across US states (nc) HPLC-MS 0.38 ug/g 0.380 (Mansfield et al., 2008) northern Germany (nc) maize silage HPLC 17.0 mg/kg 17.000 (Auerbach et al., 1998) RQC Tochigi prefecture, Japan (ac) TLC 1.66 mg/kg 1.660 (Ohmomo et al., 1994) Local store, Germany (nc) Blue cheese HPLC 1.47 mg/kg 1.470 (Finoli et al., 2001)

Table 1.2: Brief toxicological information of PMs other than immunotoxicity (all references in the context).

Potential Penicillium spp. Previous reported toxicological signs LD50s intraperitoneal (ip), intravenous Mycotoxins antimicrobial Producing mycotoxins other than im munotoxicity (iv), subcutaeous (sc), oral (po) property P. citrinin nephrotoxic (rats, pigs, chicken) ip 29.4 mg/kg (mice) OTA P. verrucosum No hepatotoxic (goat, pigs, chicken) po 46.0 mg/kg (mice) teratogenic (rats) P. citrinin nephrotoxic (pigs) po 50 mg/kg (mice) CIT P. verrucosum Yes teratogenic (rats) ip 35-58 mg/kg (mice) P. expansum genotoxic (rats) ip 19 mg/kg (rabbits) P. expansum haemorrhagic syndrome (rats, cattles) po 17 mg/kg (mice) PAT P. urticae teratogenicity/mutagenicity (mice) po 108-118 mg/kg (rats) Yes P. melinii hepatotoxic (rats) P. claviforme destruction of various cell types (rats) P. puberulum hepatotoxic (dogs, rats, mice) sc 110 mg/kg (rats) PA P. martensii nephorotoxic (rats, mice) iv 250 mg/kg (rats) Yes P. palitans haemorrhagic syndrome (dogs) po 600 mg/kg (rats) P.cyclopium carcinogenic (rats, mice) P. stoloniferm po 352 mg/kg (rats) P. brevicompactum No report of other significant toxic po 1000 mg/kg (mice) Yes MPA P. roqueforti effects than immunotoxicity po 6000 mg/kg (rabbits) P. carneurn P. roqueforti ip 20 mg/kg, 169-184 mg/kg (mice)

P commune

RQC P crustosum Yes mastitis, abortion, diarrhea (cattles) P chrysogenum paralysis, trembling, fatigue (dogs, mice) P cyclopium P glabrum

Table 1.3: Summary of immunomodulatory effect of PMs. (all references in the context; NA-not available) OTA CIT PAT PA MPA RQC Target Neutrophils Macrophages Monocytes Macrophages Macrophages Macrophages leukocytes Basophils Lymphocytes Macrophages Lymphocytes DCs Neutrophils Monocytes DCs NK cells Macrophages CTLs activated T-cells NK cells naïve B-cells Lymphocytes Proliferation ↓ (in vitro and in vivo) ↓ (in vitro) ↓ (in vitro and in vivo) ↓ (in vitro and in vivo) ↓ (in vitro and in vivo) ↓ (in vitro) ↑ (in vivo; mitogenic activity) Cell death ↑ caspase 3/9 ↑ caspase 3/9 apoptotic pathways ↓ caspase 3/8/9 Severely affect the cells NA apoptotic pathways: ↑ necrosis mediated by oxidative stress due to mediated apoptotic (activated T-cells & activated through ↓GSH-based antioxidants pathways naïve B-cells) utilizing TRADD ↑ necrosis due to type II IMPDH for GTP ↑ necrosis due to oxidative stress at synthesis oxidative stress concentration higher ↑ caspase 3/9 apoptotic than 100 μM (in vitro) pathways ↑ necrosis due to ↓ GTP/GDP synthesis Inflammation ↑ NF-β and MAPK ↓ NF-β and MAPK ↓ IL-1β, IL-6, TNF- α, NA ↓ MAPK pathways May act as an Pathways Pathways –β, IL-10, ROS ↓ ROS production irritant ↑ COX-2, iNOS, ↓ IL-1β, IL-6, TNF-α, production ↑ trafficking of DCs to ↑ MIP-2, TNF- hsp72, TNF-α iNOS, NO inflammatory sites by ↑ α in BALF CCR-1 ↑ IL-10, TGF-β Cell-mediated ↓ IFN-α, -β, -γ ↓ IFN-γ ↓ IFN-γ, IL-12, IL-2 NA ↓ activation of NK NA immune ↓ phagocytosis, anti- ↑ infectivity of ↓ phagocytosis and cells, DCs, T-cells response tumor activity of NK intracellular microbicidal activity ↑ T-reg cell number cells, basophil pathogens (T. gondii) to S. cerevisiae ↑ intracellular infection hypersensitivity, DTH response Antibody- ↓ IgM and IgG ↓ IL-4 (less sensitive ↓ IL-4 (less sensitive NA ↓ activation B-cells NA mediated (due to ↓ cell number) compared to IFN-γ) compared to IFN-γ) ↓ IgM, IgA, IgG immune ↓ antibody response to ↓ IgM, IgA, IgG ↑ IL-5 response antigens (sRBC, ↑ IgG1/IgG2 ratio No effect of PCs ovalbumin)

B. In the absence of PMs

A. In the presence of PMs

Figure 1.1: Simple schematic diagram of how the antimicrobial properties of PMs could disrupt the homeostasis of intestinal immune system. A. In the absence of PMs versus B. In the presence of PMs.

Figure 1.2: The Schematic diagram of how PMs alter inflammatory response of macrophages by altering HDAC-3 (Adapted from Rahman et al., 2004). Green arrow represents ‘stimulation’, and red indicates ‘inhibition.

Chapter 2: Experimental Rationale, Hypothesis, and Objectives 2.1. Experimental Rationale Penicillium molds are one of common fungal species that contaminate a wide variety of cereal and livestock feed crops worldwide (Almeida et al., 2000; Creppy, 2002; Duong, 1996). Many of these mold species can produce biologically active secondary metabolites, commonly known as

'mycotoxins.' Poor storage conditions for most grains and food products such as moderate temperature with high humidity, often provide favourable conditions for the production of

Penicillium mycotoxins (PMs) (Dalcero et al., 1998; Mansfield et al., 2008), adversely affect the safety of both human and animal feeds (Miraglia et al., 2009). Due to their dramatic increase during ensilaging of animal feed, PMs are also refereed as ‘silage mycotoxins.’

There are several PMs found in stored feed or food, such as citrinin (CIT), ochratoxin (OTA), patulin (PAT), mycophilic acid (MPA) and penicillic acid (PA). Studies have found that these PMs have a variety of toxicological effect in animals (Dickens and Jones, 1965; Ferrante et al., 2008;

Sansing et al., 1976). For example, some mycotoxins, such as OTA, CIT and PA, are nephrotoxic

(Braunberg et al., 1992; Sansing et al., 1976) and hepatotoxic (Dickens and Jones, 1965) in monogastric species, while PAT is a potent hepatotoxicant in both monogastric and ruminant species (Hayes et al., 1979; Sommer et al., 1974; Ukai et al., 1954). Some studies have also reported that these mycotoxins are potential immunomodulators, which could predispose animals to secondary diseases (Al-Anati and Petzinger, 2006b; Ferrante et al., 2008; Herzog-Soares and

Freire, 2004).

In terms of their immunomodulatory activity, PMs range from being inducers of immune cell apoptosis or necrosis, and altering activities of local or systemic immune responses. Most of these

PMs have been shown to reduce the proliferation and viability of immune cells under similar or

different mechanisms of action. CIT, OTA and PAT for example, induced cellular oxidative stress and ROS production, which can trigger the apoptosis or necrosis of leukocytes, and therefore, decreasing the cell number (Essid et al., 2012; Kumar et al., 2011; Wu et al., 2008). PA, on the other hand, has been reported to prevent apoptosis, but also decreased cell viability by necrosis of lymphoma cell line (Bando et al., 2003). In addition to this, these PMs have also been shown to inhibit the proliferation of bovine as well as porcine lymphocytes (Keblys et al., 2004; Stec et al.,

2008).

Other studies have also reported that these PMs at their sub-lethal levels can modulate the function of immune cells. Ferrante et al. (2006, 2008) for example, reported that OTA was a potential inducer of inflammatory molecules, such as cyclooxygenase-2 (COX-2), inducible nitric oxide synthase (iNOS) and heat shock protein 72 (hsp72), in murine macrophages. In contrast to

OTA, MPA is a well-known immunosuppressive drug that has been used for treating inflammatory disorders such as arteriosclerosis and rheumatoid arthritis (Cohn et al., 1999; Mohacsi et al., 1997), as well as preventing organ rejection after transplantation (Weber et al., 1998). In addition to this,

OTA when combined with either CIT or PA was previously shown to synergistically enhance the toxicity of these mycotoxins to liver and kidney, as well as macrophage (Sansing et al., 1976;

Veselá et al., 1983; Oh et al. 2012).

There are hundreds of mycotoxins, produced from many different mold species, and only a few of them, specifically those from either Fusarium or Aspergillius species, have been thoroughly studied and are known to pose a human and animal health risk. Potential risk of mycotoxicoses caused by PMs, on the other hand, has been disregarded regardless of the high prevalence of PMs in stored fruits and crops for human and animal consumption, and their potential to impact health

(Häggblom et al., 1990; Kulik and Holaday, 1966; Pitt, 1987). Therefore, it is important that we

understand the risk posed by exposure to these bioactive compounds; to date relatively little is known about the risk of exposure to PMs. Regardless of their abundance in animal feed and their potential to affect the immune system, the agricultural and economic importance of PMs are often under-estimated because it is generally assumed that PMs are readily metabolized by gut microbiota and hepatic enzymes (Abrunhosa et al., 2010; Fuchs et al., 2008). However, considering that most of these PMs have anti-microbial properties (Kavanagh, 1947) and are found to be hepatotoxic to animals (Dickens and Jones, 1965; Hayes et al., 1979; Sansing et al., 1976), it is possible that detoxification of the PMs is attenuated because of these properties. Furthermore,

PMs may also enter the circulation system and target tissues without passing through the digestive tract. For example, it has been reported that mycotoxins can enter circulation during inhalation

(Halstensen et al., 2004; Skaug et al., 2001) and dermal exposure (Panda et al., 1997). Since entry via these routes bypasses initial detoxification provided by gut microbiota and the liver, exposure via inhalation and dermal contact could be a serious health risk for farmers as well as their livestock

(EFSA, 2012; Reijula and Tuomi, 2003; Smith et al., 1995).

In this thesis, we have used a bovine macrophage cell line (BoMacs; provided by Stabel and

Stabel, 1995) as an in vitro cell culture model to test the immunomodulatory effect of PMs on macrophages. Macrophages have been widely used as an in vitro model to assess the potential effect of mycotoxins on immune responses (Ferrante et al., 2008; Islam et al., 2012; Sorenson et al., 1986; Wigginton et al., 2001) because they are one of the key immune mediators that bridge between innate and adaptive immune systems (Mosser, 2003). Disruption of the normal function of macrophages can alter the activities of other immune cells and mediators, potentially leading to abnormal immune response and therefore, progression of secondary diseases, such as rumatoid arthritis, tumor development or other secondary infections (Ma and Pope, 2005; Mosser, 2003;

Pollard, 2004). Therefore, looking at the immunomodulatory effect of PMs on BoMacs, which are anatomically relevant because they are derived fromperitoneal macrophages (Stabel and Stabel,

1995), would provide better understanding of the potential risk of PM exposure to macrophages functions, and therefore their potential impact on intestinal immune health in bovine species.

2.2. Experimental Objectives 1) To determine the toxic ranges of individual PMs using BoMac proliferation and viability and the potential interaction of binary mixtures of two. Published in The Open Mycology Journal 2012;

6(1): 11-16.

2) To determine if in vitro exposure to individual and combined PMs alters the expression of genes involved in epigenetic regulation, including DNA methyltransferases (DNMT1, 2, 3a, and 3b), histone demethylases (JMJD3 and UTX) as well as a histone acetylase (Bmi-1) and histone deacetylases (HDACs) in BoMacs. Published in Mycotoxin Research 2013 Nov; 29(4): 235-43.

With contribution of two undergraduate research project students, Caroline G. Balch and Rachael

L. Cliff.

3) To investigate the potential effect of PMs on BoMacs function including cytokine gene expression, ROS production and phagocytosis of Mycobacterium avium subspecies paratuberculosis (MAP).

4) To assess the efficacy of Mycosorb A+ (MA+) for preventing PMs toxicity using BoMac proliferation as a common endpoint.

5) Re-examination of the interactions of the binary mixtures of mycotoxins by applying an independent action (IA) and a concentration addition (CA) model.

2.3. Experimental Hypotheses 1) PMs alter the proliferation and cell viability of BoMacs in a dose-dependent manner, and certain combinations of these mycotoxins interact to further inhibit the cell proliferation.

2) The process of PM-induced changes in gene expression is attributed, in part, to changes in the expression of genes that encode enzymes involved in epigenetic modifications such as DNA methylation as well as histone demethylation, acetylation and deacetylation.

3) PMs differentially alter the gene expression of BoMacs cytokines, and this in turn, is reflected by changes in the activities of the macrophages, including ROS production and phagocytosis of

MAP.

4) Different factors such as pH, incubation time and percent concentration of MA+ would differentially affect the efficacy of MA+ for preventing PM toxicity.

5) The IA or CA model can be applied to predict the interactions of certain binary mixtures of PMs, including synergism or antagonism, and these models may provide better estimation of the true risks associated with PM mixture toxicity.

Chapter 3: Immunotoxicity of Penicillium Mycotoxins on Viability and Proliferation of

Bovine Macrophage Cell Line (BoMacs)

Published: Oh S-Y, Boermans HJ, Swamy HVLN, Sharma BS and Karrow NA. The Open

Mycology Journal 2012; 6(1): 11-16. doi: 10.2174/1874437001206010011 (Figures, Tables,

References formatted to fit for the thesis)

3.1. Abstract Penicillium mycotoxins are natural contaminants found in grains, crops, fruits, and fermented products, especially during postharvest as well as storage periods. Contamination by individual and combinations of these toxins is likely to compromise food quality and safety. In this study, the potential immunotoxicity of citrinin (CIT), ochratoxin A (OTA), patulin (PAT), mycophenolic acid (MPA) and penicillic acid (PA) was evaluated using a bovine macrophage cell line (BoMacs) by assessing their potential cytotoxicity and then their effects on cell proliferation. The BoMacs were exposed to a range of mycotoxin concentrations, and then to different mycotoxin combinations for 48 hrs. Some cytotoxicity was evident at concentrations greater than 2.4 µM for

PAT, and 160 µM for PA, however, at the IC50 (concentration that inhibits 50% cell proliferation), no cytotoxicity was observed for either of these mycotoxins. The mycotoxin IC50s from most potent to least potent were 0.56 µM (PAT), 12.88 µM (OTA), 29.85 µM (PA), and 91.20 µM

(CIT). Concentrations of MPA greater than 80 µM did not inhibit cell proliferation enough to calculate an IC50. Significant higher inhibition of cell proliferation was observed from the combinations of CIT+OTA, OTA+PAT, and OTA+PA compared to the effects of individual mycotoxins suggesting additive and in some cases synergistic activity between these paired mycotoxins.

3.2. Introduction Mycotoxins are metabolites of filamentous fungi that naturally contaminate a wide variety of crops, and therefore cereals and livestock feeds, worldwide (Placinta et al., 1999; Tanaka et al.,

1988). The Food and Agriculture Organization of the United Nations (FAO) estimated that approximately 25% of the food and food stuffs in the world are compromised by mycotoxin contamination (Park et al., 2009). Economic losses from mycotoxin contamination occur due to reduced crop yield and value, and reduced livestock productivity and animal health from the consumption of contaminated feed (Schmale III and Munkvold, 2009). Environmental conditions associated with high humidity and temperature favor fungal contamination of feed stuffs, and this can significantly affect the safety of both human and animal feed (Miraglia et al., 2009). The effects of individual mycotoxins, such as carcinogenicity, teratogenicity, genotoxicity, nephrotoxicity as well as immunotoxicity have been studied, but their combinational effects have not been assessed (Ciegler et al., 1977; Keblys et al., 2004; Liu et al., 2003). One of the difficulties associated with characterizing the toxicity of mycotoxins is that contaminated feedstuffs may contain various combinations of mycotoxins produced by different fungal species (Mansfield et al., 2008). For example, Penicillium expansum produces citrinin (CIT), ochratoxin A (OTA), and patulin (PAT), while Aspergillus ochraceus produces OTA and PA (Andersen et al., 2004; Martins et al., 2002; Sansing et al., 1976). Additionally, Fusarium mycotoxins are typically more abundant in corn silage immediately post-harvest, whereas Penicillium mycotoxin contamination increases during storage (Haladi, personal communication).

Among many fungal species, the Penicillium, Aspergillus, and Fusarium fungi are the most predominant species found in livestock feed such as corn silage (Amiri and Bompeix, 2005;

Dalcero et al., 1997; Pereyra et al., 2008; Pozzi et al., 1995). Many studies have characterized the

impact of oral exposure to Fusarium and Aspergillus mycotoxins on livestock species and examples of genotoxicity and immunotoxicity have been reported (Amiri and Bompeix, 2005;

Visconti et al., 1991). However, little is known about the impact of exposure to Penicillium mycotoxins, despite their potential to modulate immune function (Keblys et al., 2004; Richetti et al., 2003; Stec et al., 2008). It is assumed that mycotoxins in general are detoxified by rumen microflora (Kiessling et al., 1984), since commensal microbes within ruminants have capacity to neutralize mycotoxins produced from Fusarium or Aspergillus mold before they enter the systemic circulation via the hepatic portal vein (Swanson et al., 1987). Many Penicillium mycotoxins on the other hand, are known to have anti-microbial properties, which could disrupt the normal function of these microbes, including their detoxification of Penicillium mycotoxins (Bennett and Klich,

2003; Dzidic et al., 2010). Therefore, Penicillium mycotoxin exposure may be especially relevant to the health of ruminant species, especially since they are frequently found in contaminated silage.

In this study, we evaluated the cytotoxicity and combined effect of exposure to CIT, OTA,

PAT, mycophenolic acid (MPA), and penicillic acid (PA) on the proliferation of a bovine macrophage cell line (BoMacs). We hypothesized that Penicillium mycotoxins will alter the proliferation of BoMacs in a dose-dependent manner, and that certain combinations of these mycotoxins will additively and/or synergistically inhibit the cell proliferation.

3.3. Material and methods 3.3.1. Cell Preparation and Mycotoxin Exposure

The BoMacs (provided by Stabel and Stabel, 1995) were cultured in RMPI 1640 supplemented with 2.0 mM L-glutamine, 10% heat inactivated fetal bovine serum (FBS), 100 unit/ml of penicillin, 100 µg/ml of streptomycin, 0.25 µg/ml of amphotericin B, and 25 mM

HEPES buffer (Invitrogen, ON, Canada). After reaching confluence, the BoMacs were dislodged

with 0.05% trypsin-EDTA (Invitrogen, ON, Canada), seeded into 96-well flat bottom plates

(10,000 cells per well), and incubated at 37°C with 5% CO2 for 1 h. The BoMacs were then exposed to a range of concentrations of CIT, OTA, PAT, MPA, and PA (Sigma, ON, Canada,

Table 3.1) dissolved in dimethylsulfoxide (DMSO) for 47 h at 37°C with 5% CO2. The ranges of exposure concentration were selected based on cited IC50s determined for bovine and porcine lymphocytes (Keblys et al., 2004; Stec et al., 2008), and several preliminary trials using BoMacs that were used to optimize the toxic range of concentrations for each mycotoxin.

3.3.2. Assessing Cytotoxicity of CIT, OTA, PAT, MPA, and PA to BoMacs

After 47 h of mycotoxin exposure, another batch of BoMacs was seeded into 96-well plates

(50,000, 37,500, 25,000, 12,500, 0 cells per well) and incubated for 1 h at 37°C with 5% CO2- these cells were used as viable standards for the cytotoxicity assay. A parallel standard for dead cells was also prepared by incubating BoMacs in 70% methanol at room temperature for 1 h- these cells were stained with 0.1% trypan blue and viewed under a light microscope to confirm cell death.

The standard for dead cells was prepared by seeding the dead cells into the 96-well plates already containing the standard for live cells such that the live cell proportion was equivalent to 100, 75,

50, 25, and 0%. At 48 h post mycotoxin exposure, all plates including the standards were washed with PBS, and cytotoxicity was assessed using a commercially available cytotoxicity kit

(Invitrogen, ON, Canada). The number of live cells was estimated using a 1420 Victor-2 Multilabel

Counter (Beckman Coulter, Inc. California) by measuring the fluorescence of calcein AM

(excitation 494/emission 517 nm), once it was hydrolyzed by intracellular esterases, and the number of dead cells was estimated by measuring the fluorescence of ethidium homodimer-1

(excitation 528/emission 617 nm) bound to nucleic acids. When possible, the LC50 (concentration

that kills 50% of cells) was calculated for each mycotoxin using the Graphpad Prism software (La

Jolla, California, U.S.A.).

3.3.3. Assessing the Inhibitory Effect of CIT, OTA, PAT, MPA, and PA on BoMac Proliferation

The BoMacs were exposed to the mycotoxins as described above, and after 47 h of exposure, a standard for the proliferation assay was prepared by seeding another batch of BoMacs (50,000,

25,000, 12,500, 6,250, 3,125, 1,563, 782, 0 per well) into the 96-well plates for 1 h of incubation at 37°C with 5% CO2. The media was then removed from all plates by blotting, and the plates were frozen for 24 h at -80°C prior to the proliferation assay. The number of cells was estimated using a commercially available proliferation kit (Invitrogen, ON, Canada) that uses CyQUANT® GR dye (excitation 494/emission 517 nm) to label nucleic acids. The fluorescence intensity was measured with a 1420 Victor-2 Multilabel Counter. The IC50 (concentration that inhibits 50% cell proliferation) for each mycotoxin was calculated using the Graphpad Prism software.

3.3.4. Assessing the Combined Effects of Penicillium Mycotoxin on BoMac proliferation

The IC25 for each mycotoxin was calculated using the Graphpad Prism software and is shown in the toxicity curves in Figure 3.1. The BoMacs were then subjected to exposure by various combinations of CIT, OTA, PAT, MPA, and PA at their respective IC25s and to the individual mycotoxins (Table 3.2); cell proliferation was then assessed as described above.

3.3.5. Data Analysis

All data are presented as percent cytotoxicity and percent cell proliferation, based on the combined results of three independent experiments. For the dose-response data, an ANOVA followed by Dunnett’s test was used to test for statistical significance between the solvent controls

and treatments (b), and a T-test was used to make comparisons between the control and solvent control (a). For the combined mycotoxins, an ANOVA followed by Tukey’s test for comparisons was used to compare treatment groups (c) (Figure 3.2). The concentration data were log transformed for the analysis, and significance was determined at a p-value < 0.05 using the

Graphpad Prism software.

3.4. Result and Discussion 3.4.1. Cytotoxicity and Inhibition of the Cell Proliferation due to Individual Mycotoxin Exposure

Cytotoxicity was evident at concentrations greater than 2.4 µM for PAT, and 160 µM for PA; the LC50s for PAT and PA were estimated to be 4.46 µM and 175.79 µM, respectively (Table 3.2).

The cytotoxicity of PAT and PA may be due to their genotoxic effect on BoMacs. In support of this, Liu et al. (2003) concluded PAT as a potent clastogen with the ability to cause oxidative damage to DNA in hamster ovary cells and human peripheral blood lymphocytes, while CIT did not exhibit any significant evidence of such genotoxicity (Liu et al., 2003). In the present study,

CIT, OTA, and MPA did not induce any cytotoxicity within the concentration ranges tested (Figure

3.1A). It is possible that higher concentrations of these mycotoxins would have induced sufficient cytotoxicity to estimate their respective LC50s, however, the concentration of DMSO at these higher concentrations would have also inhibit the proliferation of BoMacs, biasing the results of the proliferation assay (Data not shown).

In the context of cell proliferation, the mycotoxin IC50s calculated in this study occurred at concentrations where no overt cytotoxicity occurred, and their order of potency from highest to lowest was: 0.56 µM (PAT) > 12.88 µM (OTA) > 29.85 µM (PA) > 91.20 µM (CIT) (Table 3.2).

When these IC50s were compared to IC50s determined from other studies, it appears that BoMacs are less sensitive to CIT, OTA, and PA, but are more sensitive than porcine and bovine lymphocytes to PAT (Table 3.3). The exposure concentrations of some of the mycotoxins investigated and IC50s determined in this study, such as OTA, were higher than circulating concentrations reported from in vivo studies (Blank et al., 2003; Höhler et al., 1999). However, there is the potential for their bioaccumulation in certain target tissues, such as kidney, intestine, and cutaenous fat (Galtier et al., 1981; Prelusky et al., 1986), as well as their synergistic and additive interaction among other mycotoxins within these tissues (Sansing et al., 1976).

Based on the observations of this study, MPA was found to have little effect on BoMacs.

Although MPA concentrations of 32 µM and 320 µM did inhibit cell proliferation, the response was not significantly different between these concentrations, and was not substantial enough to estimate an IC50 (Figure 3.1B). The inhibitory mechanism of MPA on cell proliferation has been well studied. Mycophenolic acid and its commercial analogue, mycophenolate mofetil (MMF), block the de-novo purine biosynthesis pathway, which is required for lymphocyte proliferation, by reducing cellular NF-kB levels (Hauser et al., 1997). While the proliferation of B- and T- lymphocytes relies heavily on the de novo pathway, BoMacs may equally utilize both de-novo and salvage pathways for their proliferation (Duncan and Wilkes, 2005). Furthermore, a more gradual reduction in cell proliferation was observed at the lower MPA concentrations when compared to the other mycotoxins (Figure 3.1B). These distinct characteristics may suggest why MPA and

MMF have been successfully used for treatment of allogenic transplantation as well as other immune-mediated diseases (Chaudhry et al., 2001; Pérez-Simón et al., 2008).

3.4.2. Combined Effect of Penicillium Mycotoxins on the Proliferation of BoMacs

When the ten different combinations of mycotoxins were evaluated in the present study, three combinations significantly inhibited the proliferation of BoMacs when compared to the individual mycotoxins. These combinations included; CIT+OTA, OTA+PAT, and OTA+PA, (Figure 3.2).

Combinations of CIT, OTA, and PA were of interest because these mycotoxins can be produced by Penicillium viridicatum, whereas Aspergillus ochraceus produces OTA and PA (Bacon et al.,

1973; Sansing et al., 1976). Penicillium Expansum, which is commonly found in rotten apples, is also capable of producing CIT, PAT, and OTA (Andersen et al., 2004; Martins et al., 2002). The combination of CIT+OTA was previously demonstrated to increase chicken embryonic toxicity compared to CIT or OTA, alone (Veselá et al., 1983). Additionally, Sansing et al. (1976) reported that CIT+OTA and OTA+PA inhibited orotic acid incorporation into liver and kidney tissues, while this inhibitory effect was not observed from any of the individual mycotoxins (Sansing et al., 1976). Furthermore, the synergistic and additive interaction between CIT and OTA as well as

OTA and PA on nephrotoxicity, hepatotoxicity and teratogenicity has been well addressed in other studies (Shepherd et al., 1981; Speijers and Speijers, 2004). In the present study, the proliferation of BoMacs was shown to be inhibited in an additive manner with the combinations of CIT+OTA and OTA+PAT, while the synergistic activity was observed from the combination of OTA+PA

(Figure 3.2).

3.5. Summary and Conclusion

The inhibitory effect of the following Penicillium mycotoxins, CIT, OTA, PAT, MPA, and

PA, as well as their combinations, on the cytotoxicity and proliferation of BoMacs was evaluated in this study. All mycotoxins except MPA inhibited cell proliferation by at least 50% with the evidence of cytotoxicity from PAT and PA for the concentration ranges used in this study. Most of these toxic concentrations determined in the study are higher than circulating concentrations from in vivo studies, yet the continuous consumption of feed contaminated with these toxins may lead to their accumulation in host tissues potentially increasing their bioavailability to the host immune cells. Of greater concern, however, was the additive or synergistic effects of various combinations of mycotoxins including; CIT+OTA, OTA+PAT, and OTA+PA. Since these mycotoxins can be naturally found in these combinations, their effects on the immune system warrant further investigation.

Table 3.1: Mycotoxin concentrations used for BoMac exposure study (M)

Mycotoxins Concentrations (M)

CIT 2.5, 5, 10, 20, 30, 40, 60, 80, 160, 320

OTA 0.30,0.60, 1.2, 2.4, 4.8, 7.0, 9.6, 14, 19.2, 38.4

0.0038, 0.0075, 0.015, 0.038, 0.075, 0.15, 0.2, 0.3, PAT 0.4, 0.5, 0.6, 1.2, 2.4, 4.8

MPA 0.05, 0.10, 0.2, 0.4, 0.6, 0.8, 1.6, 3.2, 32, 160, 320

PA 2.5, 4, 5, 10, 15, 20, 40, 80, 160, 320

Table 3.2. The LC50, IC50, and IC25 of various Penicillium mycotoxins (M). LC50s estimated from Figure 3.1A; IC 50s and IC25s estimated from Figure 3.2B. *R-square: the fitness of curve with the data points from Figure 3.2B.

CIT OTA PAT MPA PA

LC50 N/A N/A 4.46 N.A 175.79

IC50 91.20 12.88 0.56 N/A 29.85

IC25 52.72 8.91 0.32 0.50 13.90

R-Square* 0.945 0.981 0.841 0.871 0.987

Table 3.3: Comparison of Mycotoxin IC50s for cell proliferation of porcine, bovine lymphocytes, and BoMacs (M).

Porcine Bovine

Lymphocyte Lymphocyte BoMacs (Keblys et al., 2004) (Stec et al., 2008)

CIT 38.00 44.27 91.20

OTA 1.30 4.41 12.88

PAT 1.20 3.63 0.56

PA 18.00 N/A 29.85

A. % Cell Death of BOMACs upon Mycotoxin Exposures

CIT OTA PAT MPA PA 110 100 90 80 70 60 LC50 40

% Cell Death Cell % 30 20 10 0 -3 -2 -1 0 PAT 1 2 PA 3 Exposure Concentration, log( )

B. % Proliferation of BOMACs upon Mycotoxin Exposures

110 100 90 80 IC25 70 60 IC50 40 30

% Proliferation % 20 10 0 -3 -2 -1 PAT 0 OTA1 PA CIT2 3 Exposure Concentration, log( )

Figure 3.1: A. Cytotoxicity and B. Proliferation of BoMacs following 48 h of Mycotoxin Exposure.

CIT(□), OTA(△), PAT(▽), and MPA(◇), PA(○). LC50s (the concentraion that kills 50% of cells) are indicated in A; IC25s (The concentrations that inhibits 25% cell proliferation) and IC50s (The concentrations that inhibits 50% cell proliferation) are indicated in B with dotted lines.

(a) (c) (b) 120 *** P < 0.05 110

100

90 *** *** *** 80 IC25 *** 70 *** *** 60

IC50

% proliferation *** 40 ***

30 *** 20

CIT PA OTA PAT MPA CIT+PA control CIT+OTACIT+PATCIT+MPA OTA+PA PAT+PAMPA+PA OTA+PATOTA+MPA PAT+MPA Treatment SolventSolvent control Control 1 2

Figure 3.2: The combined Effects of Mycotoxins on the Proliferation of BoMacs. Solvent control

1 is only for the combinations of CIT+OTA. Solvent control 2 is for all other combinations and individual mycotoxins. (a) control vs solvent controls; (b) the solvent controls vs individual treatments; (c) combined vs individual treatments.

Chapter 4: Exposure to Penicillium Mycotoxins Alters Gene Expression of Enzymes

Involved in the Epigenetic Regulation of Bovine Macrophages (BoMacs)

Published: Oh S-Y, Balch CG, Cliff RL, Sharma BS, Boermans HJ, Swamy HVLN, Quinton VM,

Karrow NA. Mycotoxin Research 2013 Nov; 29(4): 235-43. doi: 10.1007/s12550-013-0174-y

(Figures, Tables, References formatted to fit for the thesis)

4.1. Abstract In this study, the modulation of key enzymes involved in epigenetic regulation was assessed in immortalized bovine macrophages (BoMacs) following in vitro exposure to the following

Penicillium mycotoxins (PMs): citrinin (CIT), ochratoxin A (OTA), patulin (PAT), mycophenolic acid (MPA), penicillic acid (PA), or a combination of one of the above with OTA at the concentration that inhibits BoMac proliferation by 25% (IC25). Real-time PCR analysis of the genes coding DNA methyltransferases (DNMTs), histone demethylases (JMJD-3 and UTX), as well as the class-1 histone deacetylases (HDAC -1, -2, and -3) and histone acetylase (Bmi-1) was assessed following 6 h of mycotoxin exposure. A change in the expression of JMJD-3 as well as

HDAC-3, MPA (p=0.1) and PA (p=0.08), by at least one of the treatments was observed at their respective IC25. The expression of JMJD-3 was significantly induced by PA, but synergistically suppressed by CIT+OTA. The combination of CIT+OTA also synergistically suppressed the expression of DNMT-3a and DNMT-3b. The combination of PAT+OTA reduced DNMT-3a expression, while PA+OTA reduced DNMT-3b expression. Lastly, MPA and PA slightly reduced

HDAC-3 expression, while OTA in combination with CIT, PAT, MPA or PA synergistically suppressed HDAC-3 expression. The results of this study demonstrate that PM exposure,

specifically OTA and other mycotoxin combinations, can alter the expression of BoMac enzymes that are involved in epigenetic regulation. These findings suggest a potential novel regulatory mechanism by which mycotoxins can modulate macrophage function.

4.2. Introduction Mycotoxins are secondary metabolites produced by microfungi that are capable of causing toxic effects in animals (Bennett and Klich, 2003). In the context of livestock production, sub- acute or chronic mycotoxicosis decreases food efficiency, fertility or overall health status

(Diekman and Green, 1992), which could potentially lead to substantial economic losses (CAST,

2003).

The following Penicillium mycotoxins (PMs): citrinin (CIT), ochratoxin A (OTA), patulin

(PAT), mycophenolic acid (MPA), and penicillic acids (PA) frequently contaminate crops, grains, feeds and fruits of Northern Europe and North and South America as well as East Asia (FAO,

2003; Flajs and Peraica, 2009; Kang et al., 2007; Mansfield et al., 2008; Tangni and Pussemier,

2006; Yue et al., 2011). For example, P. verrucosum is a major producer of CIT and OTA in stored grains (Frisvad et al., 2005), while CIT and PAT are produced by P. expansum, which can be found in fruits and cereal grains(Ciegler, 1977; Martins et al., 2002). Some of Aspergillus species are also capable of producing these mycotoxins. However, Frisvad et al. 2004 suggested that the combined effect of P. verrucosum and P. viridicata is a more common source of co-contamination of OTA and PA in cereal grains rather than Aspergillus ochraceus (Frisvad et al., 2004, 2005).

Furthermore, the concentration of these mycotoxins from Penicillium spp. tends to dramatically increase during the post harvest, especially during the storage periods (Mansfield et al., 2008;

Sweeney and Dobson, 1998).

Some of these PMs, including MPA, PAT, and PA have immunomodulatory properties in addition to their antimicrobial properties (Bennett and Klich, 2003; Dzidic et al., 2010). For example, in vitro exposure of rat alveolar macrophages to PA or PAT resulted in numerous toxic effects including inhibited protein and RNA synthesis, which is likely to interfere with macrophage function (Sorenson et al., 1986). MPA, on the other hand, selectively suppresses cell proliferation and the differentiation of immune cells, such as T- and B-lymphocytes (Dzidic et al., 2006). OTA modulates the production of specific cytokines such as interleukin-1 (IL-1) and tumor necrosis factor-alpha (TNF-) (Dhuley, 1997), and also induces immune cell death (Al-Anati and Petzinger,

2006b). CIT is often found in combination with OTA and acts synergistically with OTA to inhibit

RNA synthesis of specific proteins that alter cellular function (Sansing et al., 1976).

Each of the mycotoxins discussed herein have shown evidence of causing cellular disruption by altering of RNA and protein synthesis. We hypothesize that the process of mycotoxin-induced changes in gene expression is attributed, in part, to changes in the expression of genes that encode enzymes involved in epigenetic modifications such as DNA methylation as well as histone demethylation, acetylation and deacetylation.

Epigenetics refers to heritable changes in gene expression caused by mechanisms other than changes in the underlying DNA sequence. This change is accomplished by modification of chromatin either by DNA methylation or covalent modification of histones (e.g. acetylation, deacetylation, or methylation). Changes in epigenetic patterns can occur throughout an individual’s life in response to both internal and external stimuli (Delcuve et al., 2009).

The process of DNA methylation is catalyzed by a family of enzymes called DNA methyltransferases (DNMTs): DNMT1, DNMT2, DMNT3a, and DNMT3b (Miranda and Jones,

2007). Changes in the methylation pattern can also occur on histone proteins that help to package

the DNA by demethylases, such as Jumonji domain-containing protein 3 (JMJD3) and tetratricopeptide-repeat X (UTX). Histone deacetylases (HDACs), such as HDAC-1, HDAC-2, and HDAC-3, are enzymes that remove acetyl groups (O=C-CH3) from N-terminal of internal lysine on histone protein 3 and/or 4 (Kuo and Allis, 1998). Polycomb ring finger oncogene (bmi-

1) is part of polycomb repressor complex (PRC1) that has ubiquitin E3 ligase as well as histone acetylase activities (Sparmann and van Lohuizen, 2006). The abnormal activity of these enzymes is closely linked to various diseases, such as neurological diseases (Fischer et al., 2010), cancer

(Qin et al., 2009), and immune disorders (Royce and Karagiannis, 2012).

Many macrophage cell functions are under the control of epigenetic modifications and this regulation is crucial for a successful cell activity and inflammatory response (Bowdridge and

Gause, 2010). The characteristic marker genes of anti-inflammatory M2 macrophages for example, are epigenetically regulated (Ishii et al., 2009). It is possible that mycotoxins induce changes in gene expression by influencing epigenetic mechanisms as opposed to acting directly on target gene transcription. For example, Marin-Kuan et al. (2008) hypothesized that the carcinogenicity of OTA is likely due to the disruption of epigenetic mechanisms as opposed to a direct action of OTA on the DNA. Several studies have shown that PM exposure disrupts normal activities of immune cells, such as proliferation and inflammation (Dhuley, 1997; Dzidic et al., 2010; Oh et al., 2012;

Sorenson et al., 1986), and this may be the result of changes in epigenetic enzyme expression. The present study was designed to investigate the effects of PM exposure on the gene expression of enzymes responsible for epigenetic modifications using bovine macrophage cells (BoMac).

Specifically, the objective of the present study was to determine if in vitro exposure to individual and combined mycotoxins alters the expression of genes coding DNA methyltransferases

(DNMT1, 2, 3a, and 3b), histone demethylases (JMJD3 and UTX) as well as a histone acetylase

(Bmi-1) and histone deacetylases (HDACs) in BoMacs. Changes in expression of these genes may indirectly reflect global changes in DNA methylation and histone methylation.

4.3. Material and methods 4.3.1. Mycotoxins The following PMs were obtained from Sigma (Ontario, Canada) and dissolved in dimethyl sulfoxide (DMSO) to a stock concentrations: citrinin (CIT; 20.0 mM), ochratoxin A (OTA; 2.5 mM), patulin (PAT; 32.4 mM), mycophenolic acid (MPA; 156.1 mM), and penicillic acid (PA;

58.8 mM).

4.3.2. BoMac culture The BoMacs (provided by Stabel and Stabel, 1995) were grown in Roswell Park Memorial

Institute (RPMI) 1640 medium, supplemented with 2.0 mM L-glutamine, 10% heat inactivated fetal bovine serum (FBS), 100 unit/ml of penicillin, 100 g/ml of streptomycin, 0.25 g/ml of

2 amphotericin B, and 25 mM HEPES buffer. Cells were incubated at 37C with 5% CO2 in 75 cm flask prior to use.

4.3.3. Cell preparation and mycotoxin exposure BoMacs (1 x 106) were seeded into six-well flat-bottom plates (Corning Inc., U.S.A), centrifuged for 2 min and incubated at 37 C with 5% CO2 for 60 min. After incubation, the cells were stimulated with 1 g/ul of Escherichia coli lipopolysaccharide (LPS) (Escherichia coli

O111:B4, Sigma, Oakville, Canada) and exposed immediately to the PMs at their respective IC25, the concentration that inhibits cell proliferation by 25%. The IC25s were determined from a previous study with BoMacs and cell viability was not adversely affected at these concentrations

(Oh et al., 2012). 52.72 M for CIT, 8.91 M for OTA, 0.32 M for PAT, 0.50 M for MPA, and

13.90 M for PA. Following combinations of mycotoxins, CIT+OTA, PAT+OTA, MPA+OTA and PA+OTA, were also selected as these combinations were previously shown to synergistically suppress the proliferation of BoMacs without causing overt cytotoxicity (Oh et al., 2012). Three controls were included in the study; ‘DMSO control’ (containing both lipopolysaccharide (LPS) and DMSO equivalent to the highest treatment concentration), ‘control’ (containing only LPS), and ‘No LPS’ control (not containing neither LPS nor DMSO). These IC25s were chosen so the mycotoxins would inhibit cell proliferation without inducing cell death (Oh et al., 2012). For all treatments, cells were incubated for 6 h following mycotoxin exposure. This procedure was repeated three times for a total of three independent trials.

4.3.4. Total RNA isolation and cDNA synthesis Total RNA extraction and isolation from the BoMacs was carried out using TRIzol reagent

(Invitrogen, Canada) according to the manufacturer’s protocol. Total RNA isolated and RNA integrity number (RIN) was assessed with an Agilent 2100a Bioanalyzer (Agilent Tech Inc.,

U.S.A). Samples with a RIN value higher than 5 were used to synthesize cDNA. Reverse transcription of total RNA into cDNA was completed using Superscript III Reverse Transcriptase

(Invitrogen, Canada) according to the manufacturer’s instructions. Each reaction used 500 ng of oligio (dT)12-18 primers (Invitrogen, Canada) and 1 g of total RNA. Since the trials were repeated at least three different times, the extraction was performed separately for each trial.

4.3.5. Real-time PCR quantification Target gene expression was evaluated with quantitative PCR (qPCR) by standard curve analysis (Rutledge and Côté, 2003) using the ABI Prism 7000 Sequence Detection System and

ABI Prism 7000 SDS Software (Applied Biosystems, Canada). The qPCR was performed in a 25

L reaction using Platinum SYBR Green qPCR SuperMix-UDG with ROX (Invitrogen, Canada).

The qPCR protocol was 50 C for 2 min, 95 C for 10 min, 40 cycles of denaturation at 95 C for

15s, annealing at primer-specific temperature for 30 seconds, and extension at 72 C for 30s. A dissociation curve was run following each qPCR to ensure the presence of a uniform and single

PCR product. Each cDNA sample from each trial was run in triplicate for each gene, and these triplicates were averaged to give mean of cDNA sample for each trial or block for the statistical analysis. Primer sequences, product size, and annealing temperatures are shown in Table 4.1.

4.3.6. qPCR amplication curve and statistical analyses

Cycle threshold (Ct) values for each sample were determined using the auto Ct function of the ABI

Prism 7000 SDS Software; the Ct values were averaged between the three sample replicates. To derive input values, gene-specific standard curves were developed using pooled cDNA. The constitutively expressed housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used to standardize the expression of the target genes and determine their relative expression.

Data from the experiment was analyzed as a randomized complete block design, in which the trials represented the blocks, and the treatments consisted of two factors, OTA (with versus without) and other treatments (DMSO control, CIT, MPA, PA, PAT) in a factorial arrangement. The analysis was performed using the Mixed procedure from SAS 9.3. Preliminary residual plots indicated the standard deviation of gene expression increased with the mean for all the genes, and therefore all data was log-transformed prior to analysis in order to stabilize variances. Differences between treatment means were assessed using the Dunnett test, comparing the mean of the treatment with and without OTA to all other means for genes with a significant interaction, and

DMSO control to each of CIT, PAT, MPA, PA, for genes without a significant interaction. A separate F-test was done for the comparison between ‘control’ and ‘No LPS’ to determine the

effect of LPS on the gene expression. (*) and (**) were denoted as significant differences at p- value < 0.05 and 0.01 relative to the control, respectively, and trends  0.10 were also noted.

4.4. Result

The effect of in vitro PM treatment on the relative expression of genes responsible for epigenetic modifications was evaluated in BoMacs using GAPDH as reference gene (Figure 4.1).

LPS had no effect on the gene expression of the methylating and demethylating enzymes. LPS slightly induced the expression of Bmi-1 as well as HDAC-1, -2 and -3; however, the increase was not statistically significant (data not shown).

The expression of JMJD-3 was significantly induced by PA (p < 0.05; Figure 4.1E), and the expression of HDAC-3 was slightly suppressed by MPA (p = 0.1) and PA (p=0.08) (Figure 4.2C).

The individual mycotoxin treatments appeared to alter HDAC-2 expression, yet the statistical test failed to detect the significance among the treatments due to high variance (Figure 4.2B). The expression of DNMT-1, DNMT-2, UTX (Figure 4.1A, B, E) as well as HDAC-1, HDAC-2 and Bmi-

1 (Figure 4.2A, B and D) were not significantly affected by any of individual mycotoxin treatments at their respective IC25s.

When BoMacs were co-exposed to OTA and CIT, PAT, MPA, or PA, the relative expression of DNMT-3a, DNMT-3b, JMJD-3, and HDAC-3 was significantly affected by at least one of the

PM treatments (Figure 4.1C, D and E). Although PA alone did not significantly affect the expression of DNMT-3b (p = 0.2) and HDAC-3 (p = 0.08), PA when combined with OTA showed a significant synergistic suppressive effect on the expression of both DNMT-3b (p < 0.05; Figure

4.1D) and HDAC-3 was observed (p < 0.01; Figure 4.2C). Likewise, the combination of CIT and

OTA synergistically suppressed the gene expression of DNMT-3a (p < 0.01), DNMT-3b (p < 0.01),

JMJD-3 (p < 0.01) as well as HDAC-3 (p < 0.05). When PAT was combined with OTA, it also synergistically down-regulated the expression of both DNMT-3a (p < 0.05; Figure 4.1C) and

HDAC-3 (p < 0.01; Figure 4.2C). Lastly, the combination of MPA and OTA synergistically suppressed HDAC-3 expression (p < 0.05; Figure 4.2C). Combinations of mycotoxins appeared to alter HDAC-2 expression, but the difference was not statistically significant among the treatments due to high variance (Figure 4.2B). Lastly, the expression of DNMT-1, DNMT-2, UTX (Figure

4.1A, B, E) as well as HDAC-1, HDAC-2 and Bmi-1 (Figure 4.2A, B and D) were not significantly affected by OTA combined mycotoxin treatments.

4.5. Discussion

Previous studies have determined cellular and functional changes in macrophages following exposure to PMs (Oh et al., 2012; Sorenson et al., 1985, 1986), and some of these mycotoxins, such as OTA, have been proposed to alter various cell activities through epigenetic mechanisms

(Marin-Kuan et al., 2008). The normal functions of macrophages, such as proliferation and inflammatory response, could indirectly be modulated via altered epigenetic patterns following mycotoxin exposure (Bowdridge and Gause, 2010; Ishii et al., 2009; Marin-Kuan et al., 2007).

In a previous study, dose-response relationships were determined for the following PMs: CIT,

OTA, PAT, MPA, and PA using BoMac proliferation and cell death as endpoints (Oh et al., 2012).

Based on this previous study, IC25s were selected for the current study to standardize mycotoxin exposure concentrations to a constant physiological response (e.g. cell proliferation) without

causing overt cytotoxicity. Environmental concentrations of these mycotoxins could potentially exceed the IC25s used in this study when the favorable conditions are provided to mold for mycotoxin synthesis. For example, both OTA and PA have been found at higher concentrations in maize silage (Müller and Amend, 1997; Reyes-Velázquez et al., 2008). Furthermore, PAT and

MPA are sometimes found at similar or higher concentrations in silage (Mansfield et al., 2008;

Schneweis et al., 2000). Although the IC25s of the mycotoxins used in the study, such as OTA, were higher than circulating concentrations reported from in vivo studies (Blank et al., 2003;

Höhler et al., 1999), there is the potential for their bioaccumulation in certain target tissues, such as kidney, intestine, and cutaenous fat (Galtier et al., 1981; Prelusky et al., 1986), as well as their synergistic and additive interaction among other mycotoxins within these tissues (Sansing et al.,

1976).

In the present study, alterations in the gene expression of key enzymes responsible for epigenetic regulation, such as DNMTs, JMJD-3, UTX, Bmi-1 and HDACs were measured in

BoMacs exposed to the PMs. Lipopolysaccharide (LPS) was used to stimulate BoMacs prior to mycotoxin exposure to induce the enzymes, specifically the expression of HDAC-3 (Zhu et al.,

2010), and to determine how the mycotoxins alter the LPS-induced gene expression of the enzymes.

In this study, however, LPS did not significantly alter the expression of methylation and demethylation enzymes, though this dose has been previously shown to stimulate macrophage gene expression (Martinet et al., 1988). Lipopolysaccharide slightly induced the expression of

Bmi-1 as well as all HDACs, but the increase was not significant (data not shown). Despite this, we were still able to assess the effects of mycotoxins on the basal expression of genes used this study, and expression of genes coding some of the epigenetic enzymes was determined to be significantly altered by PM exposure, especially by combinations of mycotoxins with OTA.

PA significantly increased the gene expression of the histone demethylase JMJD-3, which is involved in decreasing histone methylation that leads to up-regulation of certain target genes. For example, JMJD-3 demethylates lysine 27 in histone 3 (H3K27) at the p16INK4A/p19ARF locus, which inhibits cell-cycle progression of human diploid fibroblasts (Agger et al., 2009). Both p16 and p19 are tumor suppressor genes that can induce p53 activity triggering the caspase-apoptotic pathway

(Bardeesy et al., 2002; Schuler et al., 2000). p53 can also induce cell-cycle inhibitory protein p21 activity, which functions to suppress cell proliferation (Wilson et al., 2006; Wong et al., 2011).

Bando et al., 2003 in contrast, reported that PA blocks Fas-induced apoptosis of Burkitt's lymphoma Raji cells, maintaining cell viability. PA concentrations greater than 100 M, however, significantly increased the cell cytotoxicity in both lymphoma Raji cells (Bando et al., 2003) and

BoMacs (Oh et al., 2012). In present study, however, IC25s were chosen so that they would not cause overt cytotoxicity; the IC25 of 13.90 M did not cause any cytotoxicity to the cells (Oh et al., 2012). This may be why the expression of DNMT-1 and DNMT-3s was not significantly suppressed by PA (Figure 4.1D), since the activity of these two enzymes is crucial for cell survival

(Beaulieu et al., 2002; Fatemi et al., 2002; Kurita et al., 2010).

PA also slightly inhibited the expression of HDAC-3, and so did MPA exposure (Figure 4.2C).

HDAC-3 activity has been shown to be positively associated with the proliferation of both human colorectal adenocarcinoma cells and colorectal cancer cells (SW480); where inhibition of HDAC-

3 activity increased cell-cycle inhibitory protein p21 in SW480, which inhibited cell proliferation

(Spurling et al., 2008). Given this, it is possible that the inhibition of BoMac proliferation by either

PA or MPA that was observed previously (Oh et al., 2012), could be partially due to the suppressed

HDAC-3 expression.

OTA is frequently detected in combination with either CIT or PA (Frisvad et al., 2004;

Vrabcheva et al., 2000), and the combination of these mycotoxins has synergistic effects in various tissues, especially the liver and kidney (Sansing et al., 1976; Shepherd et al., 1981). PAT, on the other hand, is less commonly found with OTA as compared to CIT or PA, but both PAT and OTA are the common contaminants found in corn/maize silage (Mansfield et al., 2008; Reyes-

Velázquez et al., 2008).

It is interesting to note that although CIT and OTA alone had no significant effect on gene expression, when combined, the expression of DNMT-3a, DNMT-3b, JMJD-3, and HDAC-3 was significantly reduced. Likewise, the combination of PA and OTA significantly suppressed the expression of DNMT-3b and HDAC-3. The other two PM combinations also showed synergistic suppression; PAT+OTA synergistically suppressed the expression of DNMT-3a and HDAC-3, while MPA+OTA synergistically down-regulated the expression of HDAC-3. These findings suggest that OTA when combined with any of other four mycotoxins, including CIT, PAT, MPA or PA, acts synergistically to alter the gene expression of these enzymes, especially HDAC-3 and this may lead to potential toxic effects that we reported previously with combinations of these PMs

(Oh et al., 2012). These enzymes regulate de novo cytosine methylation patterns, specifically on

H3K27 (Agger et al., 2009), which alters cell activities such as viability and proliferation.

The effects of mycotoxin exposure on DNMT expression could be important since DNMT-3a and DNMT-3b are known to methylate different sites within the genome (Hsieh, 1999; Okano et al., 1999). For example, they can both suppress p53 activity (Kurita et al., 2010; Wang et al., 2005), preventing activation of caspase apoptotic pathways (Schuler et al., 2000). In this study, CIT+OTA synergistically suppressed the expression of both DNMT-3a and DNMT-3b. PA+OTA suppressed the expression of DNMT-3b, while the combination of PAT+OTA inhibited the expression of

DNMT-3a. Inhibition of DNMT-3s may potentially induce p53-mediated programmed cell death

(Kurita et al., 2010; Schuler et al., 2000; Wang et al., 2005). These combinations, however, failed to induce BoMac cell death in the previous cytotoxicity study (Oh et al., 2012). Kurita et al. 2010 reported that DNMT-1 and DNMT-3b functionally co-operate together to maintain cell viability, and silencing of both genes is required to induce programmed cell death (Kurita et al., 2010).

Similar to DNMT-3b, DNMT-3a also functionally co-operates with DNMT-1 during de novo methylation (Fatemi et al., 2002). Therefore, it is possible that even though we observed suppressed expression of both DNMT-3s by CIT+OTA, DNMT3a by PAT+OTA, and DNMT-3b by PA+OTA in the present study, the suppression of DNMT-3s alone might not be sufficient to induce programmed cell death and/or inhibit cell proliferation (Wilson et al., 2006; Wong et al.,

2011) as long as the expression of DNMT-1 is maintained.

The combination of PA and OTA significantly inhibited JMJD-3 expression that was originally induced by PA exposure alone. Inhibited JMJD-3 expression would likely induce cell proliferation

(Wilson et al., 2006; Wong et al., 2011), yet our previous study determined that PA+OTA synergistically inhibited the proliferation of BoMacs (Oh et al., 2012). It is possible that other mechanisms or molecules that have similar function as JMJD-3 may help to maintain the activity that is typically carried out by JMJD-3 after its suppression by PA+OTA. For example, UTX is a paralog of JMJD-3 that is constitutively expressed, and has a similar function as JMJD-3 (Agger et al., 2007).

Co-exposure of OTA with either CIT, PAT, MPA, or PA synergistically decreased HDAC-3 expression. Given that the activity of HDAC-3 is positively correlated with cell proliferation

(Spurling et al., 2008), the synergistic suppression in HDAC-3 expression by these mycotoxin combinations may lead to inhibited cell proliferation. In support of this, we previously observed

the significant synergistic suppression of BoMacs proliferation from most of these combinations, except for MPA+OTA (Oh et al., 2012), with the greatest synergistic suppression in both HDAC-

3 expression and BoMacs proliferation being observed with the combination of PA and OTA. This suggests that the reduced expression of HDAC-3 by these mycotoxin combinations may potentially lead to the down-regulated HDAC-3 protein expression, and this may have contributed to the inhibited proliferation of BoMacs that was previously observed (Oh et al., 2012). HDAC-3 also regulates mitochondrial function and lipid metabolism (Sun et al., 2011), which are critical for the survival of kidney or liver cells (Begriche et al., 2011). Therefore, the synergistic inhibition of

HDAC-3 expression by these mycotoxin combinations may also potentially contribute to the synergistic nephrotoxicity (Shepherd et al., 1981), and hepatotoxicity (Sansing et al., 1976) reported from the mice studies.

Although we did not find statistically significant changes in Bmi-1 expression by the mycotoxin treatments, a trend of suppression in Bmi-1 expression was observed with co-exposure of PA and OTA (Figure 4.2D), where the combination also synergistically suppressed HDAC-3 expression. Since Bmi-1 expression can be suppressed by histone deacetylase inhibitors (HDACi) in immortalized breast epithelial (MCF10A) cells (Bommi et al., 2010), these two enzymes could be involved with or regulated by the same molecular pathways. Similar to HDAC-3, Bmi-1 is a crucial epigenetic regulator of cell proliferation (Lessard and Sauvageau, 2003; Pietersen et al.,

2008); Bmi-1 regulates cell-cycle progression, is important for DNA repair and the renewal of hematopoietic stem cells (Godlewski et al., 2008), and its expression is positively correlated with the growth of cancer cells (Qin et al., 2009). Bmi-1 was also reported to suppress p21 activity in mice cerebellar granule cell precursors (Subkhankulova et al., 2010). Both Bmi-1 and HDAC-3 regulate p21 activity that can alter cell proliferation. It is possible that the exposure duration was

not long enough to significantly affect Bmi-1 expression, since cell proliferation would not have actively occurred within the exposure period. Furthermore, it is possible that higher concentrations of mycotoxins could have significantly altered the expression of Bmi-1 and HDAC-3, even though this needs confirmation.

Regardless of the similar level of stress that BoMacs were subjected to by each mycotoxin, these mycotoxins at their IC25s differently modulated the relative gene expression of several epigenetic enzymes. It is possible that PMs inhibit cell proliferation via different pathways. For example, it has been proposed that MPA inhibits cell proliferation by inhibiting inosine monophosphate dehydrogenase (IMPDH) leading to the inhibition of de-novo purine biosynthesis

(Wadia et al., 2009b). Since the present study only investigated the expression of mRNA, which may not necessarily reflect actual translation product and activity of epigenetic enzymes, it will be important in the future to measure protein expression as well as the activity of these enzymes in both in vitro and in vivo studies.

4.6. Summary and Conclusion In summary, findings of the present study help to support the hypothesis that the consumption of food products contaminated with PMs, especially when combined, may affect the expression of epigenetic regulatory enzymes and this outcome could be a novel mechanism contributing to mycotoxicoses (Appendix 1). In future studies, it will be beneficial to study the effects of mycotoxin exposure on the expression of these genes using different physiological endpoints that are under the influence of epigenetic regulation, such as differentiation, apoptosis, or inflammatory response of macrophages. Additionally, it would be useful to explore the expression of genes coding other epigenetic enzymes such as UTX, as well as carry out global gene expression

profiling to identify candidate genes and gene pathways that may be influenced by these enzymes during in vivo mycotoxin exposure.

Table 4.1: Summary of the designed oligonucleotide primer sequences (F forward primer, R reverse primer)

PCR Annealing Gene Primer Sequence (5’-3’) Product Temperature size (bp) (C)

F TCAGCGTGTACTGTAAACGCG DNMT-1 102 59 R CATATATCGGCTTTGCTGAACC

F ACCACCAAAGTCATTACTGCGA DNMT-2 109 56 R TGTCCCTTCTATGTAGCGTCCA

F ACCTTCCTGGTATGAACAGGCC DNMT-3a 104 60 R CCTCACTTTGCTGAACTTGGCT

F AGTGGTTTGGTGATGGCAAGTT DNMT-3b 103 58 R ACCAGCTTATTGAAGGTCGCC

F TCTCGCCGGTGCTGTTACAGTT JMJD-3 102 63 R TGGAGAAGAGGCCCAAGTTGAG

F TGATCCCAGCTTTTGTCGAG UTX 139 56 R AGCATTGGACAAAGTGCAGG

F GAAATCTAAGGAGGAGGTGAATGA Bmi-1 233 59 R AGGTCGAACTCTGTATTTCAAAGG

F ACTCTTGAACCCTAATGGGTACTG HDAC-1 180 59 R AATGGCTGATGTTACCTTAAGAGC

F TGGAGGATTACATCATGCTAACAA HDAC-2 233 59 R TATCCCTTAAGTCTCCTGTTCCAG

F CAAGACTGTGGCTTATTTCTACGA HDAC-3 198 59 R TCAATGTAGTCCTCAGAGTGGAAG

F TAACTTCTGTGCTGTGCCAGCC GAPDH 103 61 R TTAAAAGCAGCCCTGGTGACC

DNMT-1 DNMT-2

2 w/o OTA with OTA 5 2 w/o OTA with OTA 5 1 4 1 4 0 3 0 3 -1 2 -1 2 -2 1 -2 1 -3 0 -3 0 DMSO CIT PAT MPA PA DMSO CIT PAT MPA PA

Expression) Log (Relative LSMsof A Treatments Expression) Log (Relative LSMsof B Treatments

DNMT-3a DNMT-3b w/o OTA with OTA w/o OTA with OTA 2 5 2 5 1 4 1 4 0 * * 3 0 3 * -1 * 2 -1 2 -2 1 -2 * 1 * -3 0 -3 0 DMSO CIT PAT MPA PA DMSO CIT PAT MPA PA LSMs of Log (Relative Expression) Log (Relative LSMsof C Treatments Expression) Log (Relative LSMsof D Treatments

JMJD-3 UTX

2 w/o OTA with OTA 5 2 w/o OTA with OTA 5 1 * 4 1 4 0 3 0 3 * -1 * 2 -1 2 -2 1 -2 1 -3 0 -3 0 DMSO CIT PAT MPA PA DMSO CIT PAT MPA PA

Expression) Log (Relative LSMsof E Treatments Expression) Log (Relative LSMsof F Treatments

Figure 4.1: (A-F) The effect of in vitro PM exposure on the expression of BoMac candidate methylating and demethylating genes relative to the GAPDH house-keeping gene. Relative expression is based on least square mean values for three independent trials. (*) and (**) denote significant differences at p<0.05 and 0.01 relative to the control, respectively.

HDAC-1 HDAC-2

2 w/o OTA with OTA 5 2 w/o OTA with OTA 5 1 4 1 4 0 3 0 3 -1 2 -1 2 -2 1 -2 1 -3 0 -3 0 DMSO CIT PAT MPA PA DMSO CIT PAT MPA PA

LSMs of Log (Relative Expression) Log LSMsof (Relative Treatments Expression) Log LSMsof (Relative Treatments A B

HDAC-3 Bmi-1

2 w/o OTA with OTA 5 2 w/o OTA with OTA 5 1 4 1 4 0 * * * 3 0 3 * * -1 * 2 -1 2 -2 1 -2 1 -3 0 -3 0 DMSO CIT PAT MPA PA DMSO CIT PAT MPA PA

Expression) Log LSMsof (Relative C Treatments Expression) Log LSMsof (Relative D Treatments

Figure 4.2: (A-D) The Effect of in vitro PM treatment on the expression of BoMac candidate acetylating and deacetylating genes relative to the GAPDH. Relative expression is based on least square mean values for three independent trials. (*) and (**) denote significant differences at p<0.05 and 0.01 relative to the control, respectively.

Chapter 5: Effect of Penicillium Mycotoxins on Cytokine Gene Expression, Reactive Oxygen

Species (ROS) production and Phagocytosis of Bovine Macrophage (BoMacs) Function

Submitted for publication: Oh S-Y, Mead PJ, Sharma BS, Quinton VM, Boermans HJ, Swamy

HVLN, Smith TK, Karrow NA. Toxicology, submitted 2014 Nov 14th, 2014

5.1. Abstract Penicillium mycotoxins (PMs) are natural contaminants that are commonly found in improperly stored animal feeds. Although exposure to certain PMs has been reported to affect immune function, little data are available for ruminant species. Therefore, in this study bovine macrophages (BoMacs) were exposed to the following PMs: citrinin (CIT), ochratoxin A (OTA), patulin (PAT), mycophenolic acid (MPA) and penicillic acid (PA), and macrophage function was assessed by measuring cytokine gene expression, the production of reactive oxygen species (ROS), as well as phagocytosis of Mycobacterium avium ssp. Paratuberculosis (MAP), the causative agent of Johne’s disease. Real-time PCR analysis of pro-inflammatory cytokine genes interleukin

(IL)-1α and IL-6, anti-inflammatory cytokine genes IL-10 and transforming growth factor (TGF)-

, as well as the neutrophil stimulating cytokine genes IL-12 and IL-23 was assessed following 6 and 24 h of PM exposure at the concentration that inhibit BoMac proliferation by 25% (IC25).

Mycotoxin treatments altered the expression of cytokine genes at 24 h. OTA significantly induced

IL-1α expression (p < 0.05), while the expression of IL-6 was suppressed (p < 0.01). MPA significantly induced the expression of IL-1α (p < 0.05) and reduced the expression of IL-12α (p

< 0.01) and IL-10 (p < 0.01). PAT significantly suppressed the expression of IL-23 (p < 0.01), IL-

10 (p < 0.05) and TGF- (p < 0.05). Neither CIT, nor PA affected the expression of these genes.

PMs did not affect BoMac intracellular ROS production and phagocytosis of MAP at their respective IC25, however, effects were observed when ROS and phagocytosis were assessed at higher sub-lethal concentrations. Pretreatment with PAT and PA for example, significantly decreased the percent phagocytosis of MAP at 5.0 M (p < 0.01) and 15.6 M (p < 0.01), respectively, but only PA significantly suppressed PAM-3 stimulated ROS production at 31.3 (p

< 0.05) and 250.0 M (p < 0.01). OTA did not affect ROS production, but it significantly increased the percent phagocytosis of MAP at 6.25 (p < 0.05) and 12.5 (p < 0.01). These findings suggest that exposure to sub-lethal concentrations of PMs can affect macrophage function, which could affect immunoregulation and innate disease resistance to pathogens.

5.2. Introduction Penicillium molds are one of the most common mycotoxin-producing fungal species found in silage for animal feed (Almeida et al., 2000; Duong, 1996). Poor storage conditions for grains and silages often provide favorable environmental conditions, such as moderate temperature with high humidity, for mycotoxin production from these mold species (Mansfield et al., 2008). The

Penicillium mycotoxins (PMs), such as citrinin (CIT), ochratoxin A (OTA), patulin (PAT), mycophenolic acid (MPA) and penicillic acids (PA), can frequently be found in animal feeds

(Mansfield et al., 2008; Martins et al., 2002; Reyes-Velázquez et al., 2008), and are reported to have various toxicological effects when consumed by animals (Charoenpornsook et al., 1998;

Dickens and Jones, 1965; Sansing et al., 1976). Some studies have also characterized PMs as immunomodulators that could predispose animals to secondary diseases (Ferrante et al., 2006;

Islam et al., 2012).

A number of studies have used macrophages as an in vitro cell model to assess the potential effect of mycotoxins on immune system, (Ferrante et al., 2008; Islam et al., 2012; Sorenson et al.,

1986; Wigginton et al., 2001) because they are one of the important innate immune cells that contribute to the host defense against pathogens (Wynn et al., 2013). Macrophages can function as host sentinel cells that recognize pathogens and initiate and regulate the host inflammatory response (Forman and Torres, 2002; Isomaki et al., 1996; Wynn et al., 2013). They also act as effector and regulatory cells that control pathogen invasion and the severity of the immune response by releasing various types of cytokines such as interleukin (IL)-1 and -6, as well as effector molecules such as reactive oxygen species (ROS) (Isomaki et al., 1996; Wynn et al., 2013).

Macrophages can also regulate the subsequent acquired immune response that provides long-term protection against invading pathogens (Wynn et al., 2013; Xing et al., 2000). For example, macrophages secrete IL-12 during early Mycobacterium bovis infection (Xing et al., 2000), which favors interferon (IFN)-γ production that subsequently enhances macrophage effector function

(Flynn et al., 1995). Therefore, an appropriate macrophage response is crucial for host survival

(Segal, 2007), and failure to respond could potentially increase susceptibility to secondary infection (Ishikawa et al. 1998) and/or lead to autoimmune disease (Cavin et al., 2009)

Several studies have reported that exposure to PMs could potentially alter macrophage function, leading to an inappropriate innate and acquired immune response during infection

(Ferrante et al., 2008; Islam et al., 2012; Johannessen et al., 2005). Most of these studies, however, have used primary cells or human or mice cell lines, even though livestock species, especially the ruminants, are often exposed to PMs via consumption of poorly stored silage (Mansfield et al.,

2008). It is generally assumed that PMs are rapidly degraded by commensal microbes in the gut

(Fuchs et al., 2008; Kiessling et al., 1984), however, their antimicrobial properties could attenuate microbial detoxification of these toxins.

Since ruminants are constantly exposed to a wide variety of pathogens such as Mycobacterium avium subspecies paratuberculosis (MAP), the causative agent of Johne’s disease (Harris and

Barletta, 2001; Momotani, 2012), it is reasonable to investigate the potential effect of PMs on bovine macrophage function and their interaction with MAP infection. Therefore, in the present study, the potential immunomodulatory effect of PMs on macrophage functions such as the expression of cytokine genes, ROS production, and phagocytosis of MAP was evaluated by using a bovine macrophage cell line (BoMacs) as an in vitro cell culture model.

5.3. Materials and Methods 5.3.1. Cell culture BoMacs (Stabel and Stabel, 1995) were cultured in Roswell Park Memorial Institute (RPMI)

1640 medium, supplemented with 2.0 mM L-glutamine, 10% heat inactivated fetal bovine serum

(FBS), 100 unit/ml of penicillin, 100 g/ml of streptomycin, 0.25 g/ml of amphotericin B, and

2 25 mM HEPES buffer. Cells were incubated at 37C with 5% CO2 in 75 cm flasks, and reached at least 80% confluence prior to use for the study.

5.3.2. Cytokine gene expression BoMacs (1.0 x 106) were seeded into six-well flat-bottom plates (Corning Inc., U.S.A), centrifuged for 2 min at 100 g, and incubated at 37C with 5% CO2. After 1 h of incubation, the cells were stimulated with 1 g/l of Escherichia coli lipopolysaccharide (LPS; E.coli O111:B4,

Sigma, Oakville, Canada) and exposed immediately to PM at their respective IC25 (the concentration that inhibit 25% of cell proliferation), which was determined from a previous study with BoMacs (Table 5.1; Oh et al., 2012). A total of five PM treatments (CIT, OTA, PAT, MPA,

PA) were used, as well as three controls: +LPS/+dimethyl sulfoxide carrier solvent (DMSO

equivalent to the highest treatment concentration), +LPS/-DMSO and -LPS/-DMSO for each time point. All mycotoxins were ordered from Sigma-Aldrich, U.S.A. After 6 and 24 h of PM exposure, total RNA was extracted using TRIzol reagent (Invitrogen, Canada) according to the manufacturer’s protocol. Messenger RNA was then converted to cDNA, and the target gene expression was evaluated by semi-quantitative PCR (qPCR) using standard curve analysis and the

PerfeCTa SYBR Green SuperMix (Quanta, BioScience Inc. U.S.A), and an ABI PRISM® 7000

(ABI applied Biosystem, U.S.A) (Table 5.2; Rutledge and Côté, 2003). The qPCR steps included: activation (50C for 3 min), initial denaturing (95C for 10 min), followed by a maximum of 40 cycles of denaturing (95C for 30 s), annealing for 30 s (Temperatures in Table 5.2) and elongation

(72C for 30 s). Each trial was repeated 3 times to account for variability.

Three reference genes were tested, including -actin, -tublin and glyceraldehyde 3- phosphate dehydrogenase (GAPDH). The expression of GAPDH was found to be the most stable across samples and treatments, therefore, GAPDH was used as a reference gene to standardize the expression of the following target genes and determine their relative expression (IL-1α, IL-6, IL-

10, IL-12α, IL-12, IL-23 and transforming growth factor (TGF)-).

5.3.3. Measuring reactive oxygen species (ROS) and phagocytosis of MAP BoMacs (1.0 x 105) were seeded into 96-black well flat-bottom plates, centrifuged for 2 min and incubated at 37C with 5% CO2. After 2 h of incubation, the cells were pre-exposed to series concentrations of PMs for 6 h (Table 5.1). For the ROS and phagocytosis studies, ethanol was used as solvent to dissolve all PMs, because DMSO was found to affect ROS production. Exposure media containing PMs was removed and cells were washed with incomplete media prior to either

ROS or phagocytosis assay.

For ROS production, BoMacs were reconstituted in RPMI media containing 10.0 l/ml

Pam3CSK4 (PAM-3; ROS stimulator; InvivoGen, U.S.A) and 50 mM of Carboxy-H2DFFDA (5-

(and-6)-carboxy-2',7'-difluorodihydrofluorescein diacetate (ROS indicator; Invitrogen, Canada).

After 6 h of incubation at 37C with 5% CO2, the media was removed, cells were washed with

PBS, and intracellular ROS production was immediately assessed by measuring intracellular fluorescing Carboxy-H2DFFDA residue (excitation 475 nm/ emission 510 nm) with a 1420 Victor

2 Multi-label Counter (PerkinElmer, U.S.A).

For phagocytosis of MAP, BoMacs were reconstituted in RPMI media containing mCherry- labeled MAP at multiplicity of infection (MOI) of 1:10, approximately 100,000 colony forming units (CFU) were used per well. The CFU of MAP was estimated by plotting the CFU versus optical density (OD) and OD versus mCherry fluorescence intensity, all of which were constructed previously in our lab (Qiumei, 2014). The plates containing both the cells and mCherry-MAP were centrifuged at 100 g for 5 min, and incubated for 6 h at 37C with 5% CO2. The media was then removed, cells were washed with RPMI media without FBS, and the amount of phagocytized mCherry-MAP was assessed by measuring mCherry fluorescence (excitation 587/ emission 610 nm) with a 1420 Victor 2 Multi-label Counter.

Cell viability was also measured after assessing both ROS production and phagocytosis of

MAP by adding 20 µM Calcein AM (Invitrogen, Canada) to each well and incubating for 30 min at room temperature. Cell viability was assessed by measuring the fluorescence of intracellular

Calcein AM (excitation 475 nm/ emission 510nm) using a 1420 Victor 2 Multi-label Counter.

Although the fluorescence of Calcein AM overlaps with the Carboxy-H2DFFDA, the fluorescence of Carboxy-H2DFFDA contributed to less than 1% of Calcein AM fluorescence.

5.3.4. Statistical analysis The gene expression data were analyzed as a randomized complete block design, in which the trials represented the blocks, and the treatments consisted of two factors, treatments (DMSO control, no LPS, control, CIT, OTA, PAT, MPA, PA) and time (6 and 24 h). The Mixed procedure

(SAS 9.3) was used for statistical analysis. Preliminary residual plots indicated the standard deviation of gene expression increased with the mean for all the genes, and therefore all data were log-transformed prior to analysis in order to stabilize variances. Differences between treatment means were assessed using Dunnett’s test, comparing the mean of the treatment between 6 and 24 h to all other means for genes with a significant interaction, and DMSO control to each of CIT,

PAT, MPA, PA, for genes without a significant interaction. (*) and (**) were denoted as significant differences at p-value < 0.05 and 0.01, respectively, relative to the DMSO control.

The ROS and phagocytosis of mCherry-MAP data were converted to percent comparison to the control using the following equation: (T-NC)/(PC-NC), where T represents sample fluorescence reading, NC is the fluorescence reading of the negative control without PAM-3 stimulation or mCherry-MAP exposure, and PC is fluorescence reading of the positive control with PAM-3 stimulation or mCherry-MAP exposure but not the mycotoxins and ethanol. All the data were log transformed prior to statistical analysis. Brown and Forsythe tests of the transformed

ROS production and phagocytosis data indicated variance differences among the toxins, so separate residual variances for each toxin were incorporated into the model prior to analysis. The

Mixed procedure (SAS 9.3) was used for all analyses. The following data were analyzed as randomized complete block design, in which the trials represented the blocks. Cell viability, indicated as dotted line in Figures 5.2 and 5.3, and the interaction of cell viability and toxins was used as a covariate for all analysis. Dunnett’s test was used in all the analyses to compare each PM

concentration with its’ respective ethanol control on the same plate. (*) and (**) were denoted as significant differences at p-value < 0.05 and 0.01 relative to the ethanol control, respectively, and trends  0.10 were also noted.

5.4. Results 5.4.1. Effect of PMs on BoMacs cytokine gene expression None of the PMs significantly affected cytokine gene expression at 6 h. Significant changes in cyokine gene expression were, however, evident within the treatments exposed to mycotoxins for 24 h. There was significant time (6 versus 24 h) and treatment (mycotoxins) interaction for the expression of both IL-1α and IL-6 (Figure 5.1A and B), but not for the expression of IL-10, IL-

12α, IL-12, IL-23, and TGF- (Figure 5.1C-G). LPS significantly stimulated the gene expression of IL-6 at 24 h, but not at 6 h (p < 0.01; Figure 5.1B), and it also stimulated the expression of IL-

12α (p < 0.01; Figure 5.1E). The increased expression of IL-12α was significantly attenuated by the DMSO carrier solvent (p < 0.01; Figure 5.1E), but DMSO had no effect on the gene expression of the other cytokines (Figure 5.1).

OTA at 24 h significantly induced IL-1α expression (p < 0.05; Figure 5.1A), and suppressed the expression of IL-6 induced by LPS (p < 0.01; Figure 5.1B). MPA also significantly induced the expression of IL-1α (p < 0.05; Figure 5.1A), but had no significant effect on IL-6 (Figure 5.1B).

MPA also significantly reduced the expression of IL-10 (p < 0.01; Figure 5.1C) and IL-12α (p <

0.01; Figure 5.1E). The expression of IL-12 was slightly suppressed by MPA (p = 0.06; Figure

5.1F). PAT significantly suppressed the expression of both anti-inflammatory cytokines, IL-10 (p

< 0.05; Figure 5.1C) and TGF- (p < 0.05; Figure 5.1D) as well as IL-23 expression (p < 0.01;

Figure 5.1G). CIT and PA did not significantly alter the gene expression of any cytokines, even though PA seemed to slightly reduce the expression of IL-23 (p = 0.07; Figure 5.1G).

5.4.2. Effects of PMs on BoMacs ROS production and phagocytosis of MAP The potential effect of PMs on PAM3-stimulated ROS production as well as phagocytosis of mCherry-MAP was also investigated. Initial experiments indicated that the PMs at their respective

IC25 did not affect these cell functions (Appendix 2). Therefore, a range of PM concentrations was tested to determine the effective of PM concentration after 6 h of exposure (Table 5.1).

The effect of these PMs on intracellular ROS production and percent phagocytosis of MAP was found to occur at much higher concentrations (Figure 5.2 and 5.3). OTA did not affect ROS production (Figure 5.2B), yet OTA at 6.3 M (p < 0.05) and 12.5 M (p < 0.01) significantly increased the percent phagocytosis of MAP (Figure 5.3B). PA significantly decreased the PAM-3 stimulated ROS production at the concentrations of 62.5 M (p < 0.05) and 250.0 M (p < 0.01;

Figure 5.2E), and the percent phagocytosis of MAP was reduced at 15.6 M (p < 0.01; Figure

5.3E). PAT did not affect ROS production, but showed decrease in the trend of ROS production throughout the concentrations tested (Figure 5.2C). PAT at concentrations greater than 2.5 M also showed an inhibitory trend on the phagocytosis of MAP, and statistically inhibited phagocytosis at 5.0 M (p < 0.01; Figure 5.3C). Neither CIT nor MPA significantly altered ROS production and percent phagocytosis of MAP (Figure 5.2A, D and 5.3A, D). All the significant effects observed from both ROS and phagocytosis studies occurred at the sub-lethal PM concentrations except for PA at 250.0 M as shown with dotted line in Figures 5.2E and 5.3E.

5.5. Discussion PMs are common contaminants found in food, especially in storage feed for animal consumption (Mansfield et al., 2008). Several studies have determined that PMs have immunomodulatory effects, altering the production of inflammatory cytokines from human and murine macrophages (Ferrante et al., 2006; Islam et al., 2012; Luft et al., 2008). The following

study also demonstrated that the PMs differently altered bovine macrophage cytokine gene expression as well as ROS production and phagocytosis, all of which are important for regulating inflammatory response and host resistance against bacterial infection.

Lipopolysaccharide (LPS) was used in the present gene expression study to stimulate BoMacs cytokine gene expression. LPS significantly increased the expression of IL-6, yet it did not significantly affect the expression of any other cytokines tested in the present study. Along with stimulated IL-6 expression, one would expect to also see an increase in IL-1α expression in response to LPS stimulation, a well-known potent inducer of inflammation (Pang et al., 1994). A slight increase in IL-1α expression was observed, but it was not statistically significant. This may be due to low TLR-4 expression associated with BoMacs, which is one of the main pattern- recognition receptors for LPS (Nomura et al., 2000). In a preliminary study, it was also noted that

LPS failed to induce ROS production (Appendix 3), which was why PAM-3 was used as an alternate for LPS to stimulate ROS production.

DMSO, the solvent control, also appeared to significantly suppress the expression of IL-12α; however, it did not alter the expression of other cytokines. Since DMSO also affected both ROS production in preliminary study (data not shown), ethanol was used as a solvent to dissolve all

PMs for the ROS production and phagocytosis assays.

The present study showed that OTA at its respective IC25 significantly induced IL-1α expression after 24 h of exposure. IL-1α is a potent pro-inflammatory cytokine that can induce the expression of other pro-inflammatory cytokines and molecules such as IL-6, hepatic acute-phase proteins, and the production of ROS (Dinarello, 1994). Interestingly, OTA did not significantly induce BoMacs IL-6 gene expression, but rather decreased it. Although IL-1α and IL-6 are known to be potent inflammatory cytokines that stimulate macrophage ROS production (Radeke et al.,

1990; Sung et al., 2000), IL-6 also possesses anti-inflammatory and regenerative capabilities that depend on its interaction with IL-6 receptors. Binding of IL-6 to soluble IL-6 receptors for example, induces ROS (Kojima et al., 2012), whereas when IL-6 interacts with membrane-bound IL-6 receptors, macrophages have an anti-inflammatory phenotype (Komohara et al., 2008). Therefore, the suppressed IL-6 expression observed in this study by OTA may not necessarily reflect increased and/or decreased inflammatory response of macrophages, and this is supported by the fact that ROS production was not induced even at higher OTA concentrations. Furthermore, these cytokines are stored as inactive pro-IL-1 and pro-IL-6 molecules in macrophage granules and become activated by caspase-1 (Maelfait et al., 2008). Therefore, it is likely that the expression of both IL-1 α and IL-6 in the present study does not reflect what is happening at the level of protein expression.

Higher concentrations of OTA (12.5 M and 25.0 M) in the present study also significantly increased BoMacs phagocytosis of mCherry-MAP, which from a host’s perspective could be good or bad, depending on the state of macrophage activation (Flynn et al., 1995; O’Leary et al., 2011).

Interestingly, an acute exposure study carried out by Hibi et al. (2013) recently demonstrated significant induction of CD44 mRNA expression in rodent proximal tubule cells following OTA exposure; CD44 is also expressed on macrophages, and has been reported a major attachment site for M. tuberculosis (Leemans et al., 2003; Palaniappan et al., 2012). It would be interesting in future studies to assess if CD44 expression is altered on BoMacs by OTA exposure, and if this mediates phagocytosis of MAP.

Although no studies have previously investigated the effect of OTA on macrophage phagocytosis of MAP, previous in vivo studies using sheep red blood cells (sRBCs) as a target reported that OTA suppressed macrophage phagocytosis (Harvey et al., 1992; Hassan et al., 2012).

These contrasting results may be due to differences in study conditions (in vitro versus in vivo), the expression of receptors mediating phagocytosis (Harvey et al., 1992; Restrepo et al., 2014), and to the phagocytosis target, which in our case was an intracellular pathogen that relies on host phagocytosis for its survival and replication.

MPA was another PM that induced IL-1α expression. This was not anticipated, however, because MPA is a potent immunosuppressant that is used to prevent tissue rejection after organ transplantation and treating inflammatory diseases (Cohn et al., 1999; Weber et al., 1998). The results from this study, including up-regulated IL-1α expression and down-regulated IL-10 expression, suggest that MPA induces an inflammatory response from BoMacs. This has also been previously observed from a couple of in vivo studies, where MPA significantly elevated the protein expression of IL-6, TNF-α and MIP-2 (Dzidic et al., 2010; Rand et al., 2005). It has also been reported that MPA suppressed the cell-mediated immune response by down-regulating the gene expression of IL-12 and IL-23, all of which are important for cell-mediated immune response (Park and Scott, 2001; Uhlig et al., 2006). Regardless of its significant effect on cytokine gene expression,

MPA, even at higher concentrations, did not affect ROS production or macrophage phagocytosis of MAP.

In contrast to these findings, a few studies have demonstrated that MPA is a potent suppressor of ROS production (Park et al., 2004, 2008). Park et al. for example, (2004; 2008) reported that

MPA significantly inhibited ROS production by down-regulating the extracellular regulated kinase

(ERK)1/2-mitogen-activated protein kinase (MAPK) mediated pathway (Svegliati et al., 2005; Yu et al., 2011). Since PAM-3 in the present study activated macrophage ROS production through

TLR 1/2, which leads to nuclear factor (NF)-β signaling in addition to ERK1/2-MAPK signaling

(Maelfait et al., 2008), it is possible that macrophages were able to maintain ROS production

regardless of MPA exposure concentrations. This is just a speculation, but warrants investigation in the future studies.

PAT did not alter the expression of IL-1α and IL-6, yet it significantly decreased the expression of IL-10 and TGF- in this study. Since IL-10 and TGF- are considered anti- inflammatory cytokines (Isomaki et al., 1996; Paglinawan et al., 2003) and are important for steering the differentiation of regulatory T-cells (Taylor et al., 2006), their inhibition could favor cell-mediated immunity and excessive inflammation (Jarry et al., 2011).

PAT also significantly reduced the expression of IL-23, which is an important cytokine for activating neutrophils (Mus et al., 2010; Pelletier et al., 2010), and promotes intestinal inflammation (Uhlig et al., 2006). Indramohan et al., (2012) also showed that IL-23 promotes the trafficking of monocytes to Listeria monocytogenes infection sites, and regulates macrophage activity in an autocrine manner (Indramohan et al., 2014). Therefore, the suppression of IL-23 expression by PAT could potentially reduce the functional activity of cell-mediate immune cells, such as the generation of ROS and macrophage phagocytosis (Mikawa et al., 1994). In the present study, PAT did not significantly affect ROS production, however, a trend of decreasing ROS production was noted at concentrations between 1.4 and 10.0 μM. Bourdiol et al. (1990) observed that 13 μM of PAT pre-incubation for 2 h significantly reduced superoxide anion production from murine peritoneal macrophages without affecting the cell viability. They also observed significantly inhibited ROS production at 26 μM, though this was probably due to decreased macrophage viability, because we observed that PAT dramatically decreased macrophage viability above 20 μM.

The present study also showed that PAT significantly decreased the phagocytosis of mCherry-MAP at 5.0 μM. In agreement with this study, Borudiol et al. (1990) reported that PAT

concentrations higher than 3.2 μM significantly decreased the phagocytosis of Saccharomyces cerevisiae. Additionally, their study found that the phagosome-lysosome fusion decreased above

0.65 μM, and lysosomal enzyme and microbicidal activity, and intracellular killing of yeast decreased at PAT concentrations above 3.2 μM (Bourdiol et al., 1990). Since the fusion of phagosome and lysosome is crucial for clearing MAP (Woo and Czuprynski, 2008), it is possible that PAT exposure could increase MAP survival during macrophage infection, and this warrants future investigation.

Similar to PAT, PA slightly reduced the expression of IL-23 (p = 0.07), and significantly suppressed ROS production at the concentrations between 31.3 μM and 125.0 μM. PA also significantly inhibited the macrophage phagocytosis of mCherry-MAP at 15.6 M. We are unaware of any previous studies that have investigated the effect of PA on macrophage function.

However, it appears that PA may share a similar toxic effect with PAT, based on the results from this study. Interestingly, Indramohan et al. (2014) reported a reduction of both ROS production and phagocytosis of L. monocytogenes in IL-23 knockdown mice as compared to normal C57BL/6 mice (Indramohan et al., 2014), This suggests that deceased IL-23 expression in the present study may have contributed to the observed decreases in BoMacs ROS production and phagocytic activity.

5.6. Summary and Conclusion Changes in cytokine gene expression regulating the inflammatory and cell-mediated immune response were measured in BoMacs that were exposed to CIT, OTA, PAT, MPA or PA. OTA,

PAT and PA also significantly affected macrophage ROS production and phagocytosis of the intracellular pathogen, MAP, at much higher sub-lethal concentrations. These results suggest that

sub-lethal concentrations of PMs can affect macrophage functions, which could in turn affect host resistance to intracellular pathogens.

Table 5.1: Exposure concentrations of PMs used in the cytokine gene expression, ROS and phagocytosis assay. IC25 values, determined from a previous study (Oh et al., 2012), were used for the cytokine gene expression study.

Concentrations (M) Mycotoxins IC25s Concentrations used for the PAM-3 stimulated Concentrations used for the phagocytosis of for gene expression study ROS production assay mCherry-MAP assay (Oh et al., 2012) CIT 52.7 18.8, 37.5, 75.0, 150.0, 300.0, 600.0 9.4, 18.8, 37.5, 75.0, 150.0, 300.0

OTA 8.9 1.6, 3.1, 6.3, 12.5, 25.0, 50.0 0.8, 1.6, 3.1, 6.3, 12.5, 25.0

PAT 0.3 1.3, 2.5, 5.0, 10.0, 20.0, 40.0 0.3, 0.6, 1.3, 2.5, 5.0, 10.0

MPA 0.5 28.1, 56.3, 112.5, 225.0, 450.0, 900.0 14.1, 28.1, 56.3, 112.5, 225.0, 450.0

PA 13.9 31.3, 62.5, 125.0, 250.0, 500.0, 1000.0 3.9, 7.8, 15.6, 31.3, 62.5, 125.0

Table 5.2: Summary of designed oligonucleotide primer sequences for cytokine gene expression study (F forward primer, R reverse primer).

Annealing PCR Product Gene Primer Sequence (5’-3’) Temperature size (bp) (C)

F TCCATACCTGACGGCTACTACAT IL-1α 138 62

R TGAGCACTCACAAACAGTTGAGT

F CAGAGAAAACCGAAGCTCTCAT IL-6 141 62

R CCTCCATTTTTGGAAGATTCAG

F AAAGCCATGAGTGAGTTTGACA IL-10 155 62 R TGGATTGGATTTCAGAGGTCTT

F CTGTGTTCGTCAGCTCTACATTG TGF- 129 61 R TACTGTGTATCCAGGCTCCAGAT

F TGATGGATCCTAAGAGGCAAAT IL-12α 159 60 R GAAGGATGCAGAGCTTGACTTT

F CTCTGAGTCGTTCACTCCTCCT IL-12 127 62 R CCAGAATAATCCTTTGCCTCAC F AGCAACTCTGAGCCCCTAAAG IL-23 196 61 R TCAGCCTCTCCTAGGTCAACA F GATGAGATTGGCATGGCTTTA -actin 140 60 R AGAACTTTGGGAATGCTCGAT

FATCATGAACACCTTTAGCGTCA -tubulin 114 59 RGTAGGTCTCATCCGTGTTCTCC

F ACGTGTCTGTTGTGGATCTGAC GAPDH 112 61 RTAGCCTAGAATGCCCTTGAGAG

IL-1 α IL-6 6hrs 24hrs 6hrs 24hrs 2 5 2 5 * * 1 4 1 ** 4 0 3 0 ** 3 -1 2 -1 2 -2 1 -2 1 -3 0 -3 0

CIT PA CIT PA -LPS OTA PAT MPA -LPS OTA PAT MPA

LSMs Log of (Relative Expression)

+LPS/-DMSO+LPS/+DMSO +LPS/-DMSO+LPS/+DMSO A Treatments B Treatments

IL-10 TFG - β

2 5 2 5 1 4 1 4 * 0 * ** 3 0 3 -1 2 -1 2 -2 1 -2 1 -3 0 -3 0

CIT PA CIT PA -LPS OTA PAT MPA -LPS OTA PAT MPA

LSMs Log of (Relative Expression)

+LPS/-DMSO+LPS/+DMSO +LPS/-DMSO+LPS/+DMSO C Treatments D Treatments

IL-12 α IL-12 β

2 5 2 5 1 ** 4 1 4 0 ** 3 0 3 -1 2 -1 2 -2 1 -2 1 -3 0 -3 0

CIT PA CIT PA -LPS OTA PAT MPA -LPS OTA PAT MPA

LSMs Log of (Relative Expression)

+LPS/-DMSO+LPS/+DMSO +LPS/-DMSO+LPS/+DMSO E Treatments F Treatments

IL-23

2 5 1 4 0 ** 3 -1 2 -2 1 -3 0

CIT PA -LPS OTA PAT MPA

LSMs Log of (Relative Expression)

+LPS/-DMSO+LPS/+DMSO G Treatments

Figure 5.1 (A-G): Effect of in vitro PM exposure on BoMac cytokine gene expression.

PAM-3 induced % intracellular ROS PAM-3 induced % intracellular ROS after 6 h of CIT Exposure after 6 h of OTA Exposure ROS production ROS production 200 200 cell viability cell viability 150 150

contrl contrl

50 50

0 0

% % relaitve to control 1.0 1.5 2.0 2.5 3.0 % relaitve to control 0.0 0.5 1.0 1.5 2.0 A log (exposure concentration) B log (exposure concentration)

PAM-3 induced % intracellular ROS PAM-3 induced % intracellular ROS after 6 h of PAT Exposure after 6 h of MPA Exposure ROS production ROS production 200 200 cell viability cell viability 150 150

contrl contrl

50 50

0 0

% % relaitve to control 0.0 0.5 1.0 1.5 2.0 % relaitve to control 1.0 1.5 2.0 2.5 3.0 C log (exposure concentration) D log (exposure concentration)

PAM-3 induced % intracellular ROS after 6 h of PA Exposure ROS production 200 cell viability 150 * contrl **

% % relaitve to control 1.0 1.5 2.0 2.5 3.0 E log (exposure concentration)

Figure 5.2 (A-E): Effect of 6 h in vitro PM exposure on BoMac ROS production. Exposure concentrations are listed in Table 1. Cell viability (represented as dotted line) was used as covariance for statistical analysis. 'contrl' on y-axis designated as a dotted line indicates the percent solvent (ethanol) control relative to controls without ethanol exposure. (*) and (**) denote significant differences at p<0.05 and 0.01 relative to the ethanol control, respectively.

% Phagocytosis of MAP after 6 h of CIT Exposure % Phagocytosis of MAP after 6 h of OTA Exposure % Phagocytosis of mCherry-MAP % Phagocytosis of mCherry-MAP 300 300 ** cell viability cell viability * 250 250 200 200 150 150 contrl contrl 50 50 0 0

% % relaitve to control 0.5 1.0 1.5 2.0 2.5 3.0 % relaitve to control -0.5 0.0 0.5 1.0 1.5 2.0 A log (exposure concentration) B log (exposure concentration)

% Phagocytosis of MAP after 6 h of PAT Exposure % Phagocytosis of MAP after 6 h of MPA Exposure % Phagocytosis of mCherry-MAP % Phagocytosis of mCherry-MAP 300 300 cell viability cell viability 250 250 200 200 150 ** 150 contrl contrl 50 50 0 0

% % relaitve to control -1.0 -0.5 0.0 0.5 1.0 1.5 % relaitve to control 0.5 1.0 1.5 2.0 2.5 3.0 C log (exposure concentration) D log (exposure concentration)

% Phagocytosis of MAP after 6 h of PA Exposure % Phagocytosis of mCherry-MAP 300 cell viability 250 200 150 ** contrl 50 0

% % relaitve to control 0.0 0.5 1.0 1.5 2.0 2.5 E log (exposure concentration)

Figure 5.3 (A-E): Effect of 6 h in vitro PM exposure on BoMacs phagocytosis of mCherry- MAP.

Exposure concentrations are listed in Table 1. Cell viability (represented as dotted line) was used as covariance for statistical analysis. 'contrl' on y-axis designated as a dotted line indicates the percent solvent (ethanol) control relative to controls without ethanol exposure. (*) and (**) denote significant differences at p<0.05 and 0.01 relative to the ethanol control, respectively.

Chapter 6: In vitro Exposure of Penicillium Mycotoxins (PMs) with or without Mycosorb A+

(MA+) on Bovine Macrophage proliferation (BoMacs)

6.1. Abstract Penicillium mycotoxins (PMs) are contaminants that are frequently found in grain or crop- based silage for animal feed. Previously, we have characterized the potential immunotoxicity of the following PMs: citrinin (CIT), ochratoxin A (OTA), patulin (PAT), mycophenolic acid (MPA) and penicillic acid (PA) by using a bovine macrophage cell line (BoMacs). In the present study, cell proliferation was used as a bioassay endpoint to evaluate the efficacy of Mycosorb A+ (MA+) which contains an esterified-glucomannan derived from a yeast cell wall component and blue- green algae, for preventing PM toxicity under various in vitro conditions such as: pH (3, 5, 7), incubation time (1, 2, 4, 6 h), percentage of MA+ (0.05, 0.1, 0.2, 0.5, 1.0%) and PM concentration.

MA+ was most effective in preventing the toxicity of 12.88 µM and 25.8 µM OTA at pH 3.0 (p <

0.0001), regardless of incubation time (p < 0.0001) and the percentage of MA+ (p < 0.0001). An incubation time of 6 h (p < 0.05), or 0.5 and 1.0 percent MA+ (p < 0.0001) significantly improved the efficacy of MA+ for preventing CIT toxicity. In contrast, 0.5 and 1.0% of MA+ appeared to exacerbate the PAT toxicity (p < 0. 0001). This exacerbated effect on PAT toxicity was constantly observed with higher PAT concentrations, and it reached significance at a concentration of 0.70

µM (p < 0.0001). MA+ had no effect on PA toxicity. These results suggest that MA+ may reduce

OTA toxicity and to some extent CIT toxicity at pH 3.0. Although PAT toxicity was increased by

MA+ treatment, PAT is readily degraded during heat treatment, and may therefore be dealt with using other preventative measures.

6.2. Introduction Mycotoxins are natural toxicants produced by fungi that are commonly found in crops and stored grain-based animal feed (Boutrif, 1995). Previous studies have shown potential toxicity of mycotoxins when consumed by animals (Braunberg et al., 1992; Sansing et al., 1976). The consumption of animal feed heavily contaminated with mycotoxins can decrease livestock productivity and health, and this contributes to economic losses for producers (Schmale III and

Munkvold, 2009). Due to their significant impact on agricultural commodities, many efforts have been made to reduce mycotoxin exposure via animal feed.

Although there are approximately 300 compounds that have been identified as mycotoxins

(Bennett and Klich, 2003), studies have focused on only a few mycotoxins that have a history of outbreak, such as aflatoxins, T-2 toxins and deoxynivalenol (Forgacs, 1962; Marasas et al., 1987;

Ramakrishna et al., 1989). In comparison to these Fusarium or Aspergillius mycotoxins, there are considerably fewer studies available related to mycotoxins produced from Penicillium spp., even though they are regarded as the most common mycotoxins found in silage for animal feed

(Mansfield et al., 2008). Therefore, there is a need for more studies to examine their potential toxicity, as well as effective preventive methods to deal with Penicillium mycotoxins (PMs).

In previous studies, we used the bovine macrophage cell line (BoMacs) as an in vitro model to assess the potential toxic effects of the following PMs: citrinin (CIT), ochratoxin A (OTA), patulin (PAT), mycophenolic acid (MPA) and penicillic acid (PA). Endpoints including cell proliferation, viability, cytokine gene expression, reactive oxygen production and phagocytosis of

Mycobacterium avium spp. Paratuberculosis (MAP) were used to assess PM immunotoxicity and/or immunomodulation (Oh et al., 2012, 2013). These studies demonstrated that these PMs

100

differentially modulate macrophage functions, which in turn could predispose animals to secondary disease.

The present study was designed to test the efficacy of the Mycosorb A+ (MA+; Alltech Inc.

Kentucky) derived from the cell wall of Saccharomyces cerevisiae and culture extract of blue green algae (Spirulina spp.) for preventing the toxic effects of Penicillium mycotoxins using an in vitro BoMac proliferation bioassay. The objective of this study was to determine the optimal in vitro PM binding conditions, such as pH, incubation time and percentage of MA+ and PM concentration.

6.3. Methods 6.3.1. Preparation of cells The BoMacs (provided by Stabel and Stabel, 1995) were cultured in Roswell Park Memorial

Institute (RPMI) in 1640 medium, supplemented with 2.0 mM L-glutamine, 10% heat inactivated fetal bovine serum (FBS), 100 unit/ml of penicillin, 100 µg/ml of streptomycin, 0.25µg/ml of amphotericin B, and 25 mM HEPES buffer. All cell culture products were purchased from

2 Invitrogen, Canada. Cells were incubated at 37°C with 5% CO2 in a 75 cm flask, and were grown to above 80% confluence prior to use for the study.

6.3.2. Incubating PMs with MA+ and preparation of exposure media: All mycotoxins were dissolved in ethanol for the study. The IC50s (The concentrations of

PMs that inhibit 50% of BoMac proliferation) were selected as concentrations for testing the efficacy of MA+ based on a previous study (Oh et al., 2012). In this previous study, the IC50 for

MPA could not be determined even at a concentration of 320 µM, where the solvent DMSO began

101

to inhibit cell proliferation. Therefore, 32 µM was used instead, because it was the highest MPA concentration previously tested that was not adversely affected by DMSO,. The IC50 of PAT was readjusted from 0.56 µM to 0.70 µM due to the decreased potency of PAT, probably due to the use of a different solvent. For these preliminary studies, one variable was addressed at a time, and a total of 4 different variables including pH, incubation time, percentage MA+, and various PM concentrations were considered. Mycosorb A+ was donated by Alltech Inc. for the purpose of this study.

To test the effect of pH, each PM was incubated with 0.2% MA+ in 500 ml of Milli-Q water at pH: 3, 5 and 7 for 2 h. The pH was adjusted with 1M of hydrogen choloride (HCl) and Sodium hydroxide (NaOH) after MA+ was added to Milli-Q water. To see the effect of incubation time, each mycotoxin was incubated with 0.2% MA+ in 500 ml of Milli-Q water at pH 3.0 for 1, 2, 4 or

6 h. To determine the most effective percentage of MA+ that inhibited mycotoxin toxicity, various percentages (0.05, 0.1, 0.2, 0.5, 1%) of MA+ were incubated with each mycotoxin in 500 ml of

Milli-Q water for 2 h. Lastly, 0.2% MA+ was added to 5 different PM concentrations (Table 6.1) in 500 ml of Milli-Q water at pH 3 for 2 h to observe MA+ effectiveness at different PM concentrations. All incubations were carried out in 1.5 ml tubes that were placed horizontally at room temperature (22-25°C) with horizontal shaking at 30 g. After incubation, the MA+ was removed by centrifugation at 10,000 g for 10 min before the cell culture procedure. Supernatants containing the PM residues were then added to 96 well PCR plates for drying by desiccation for

24 h using silica gels & molecular sieves; Sigma-Aldrich, U.S.A) in a vacuum-sealed desiccator.

After complete drying, residual PMs were reconstituted with RPMI media containing HEPES buffer (pH 7.2-7.4) prior to BoMacs culture.

102

6.3.3. BoMacs proliferation BoMacs (1.0x104) were seeded into 96-well flat-bottom plates, centrifuged at 100 g for 2 min and incubated at 37C with 5% CO2 for 2 h until the cells adhered completely to the bottom of the plates. The media was removed and 100 µl of the prepared exposure media containing the PM residues was added to each well. After 47 h of PM exposure, a cell proliferation standard was prepared by seeding another batch of BoMacs in each plate (1.0x105, 7.5x104, 5.0x104, 2.5x104 and 1.0x103 cells per well), followed by centrifugation at 100 g for 2 min speed, then incubation for 1 h at 37°C with 5% CO2.

The proliferation assay was carried out as described from a previous study (Oh et al. 2012).

The media was removed from all wells by blotting, and the plates were placed in a -80°C freezer for at least 24 h to lyse the cells. The number of cells was estimated using a commercially available cell proliferation kit (Invitrogen, ON, Canada) that uses CyQUANT® GR dye (excitation

494/emission 517 nm) to label nucleic acids. The fluorescence intensity was measured with a 1420

Victor2 Multi-label Counter (PerkinElmer, U.S.A).

6.3.4. Data Analysis All raw data were normalized to the number of cells based on a standard curve for each plate, and presented as percent cell proliferation compared to a negative control that contained MA+ but no PM. The percentage values for each treatment were log-transformed prior to statistical analysis in order to stabilize variances. The Tukey-Kramer test was used to compare the means of all the treatments that contained MA+ with the means of positive controls for each PM sample that contained mycotoxin but no MA+. For Figures 6.1 and 6.2, the positive control was also compared to another control that did not undergo air-dry to determine potential PM degradation during drying.

For Figure 6.4, the dotted lines represent the positive control (PM without MA+), and solid lines

103

represent PM with MA+. (*), (**) and (***) denote significant differences at p < 0.05, 0.01 and

0.0001 relative to the positive control, respectively.

6.4. Results The present study used the proliferation of BoMacs as a bioassay endpoint to determine the potential efficacy of MA+ for preventing PM toxicity. The drying procedure significantly reduced the toxicity of OTA (p < 0.05 in Figure 6.1B; p < 0.0001 in Figure 6.2B), but the efficacy of MA+ for OTA was still observed. MA+ was the most effective in preventing OTA toxicity and to some extent CIT toxicity among five mycotoxins tested in this study (Figure 6.1-6.4).

6.4.1. MA+ effectiveness at different pH levels 0.2% MA+ incubation for 2 h was effective in preventing OTA (p < 0.0001) and MPA (p <

0.0001) toxicity at pH 3.0 (Figure 6.1B and D). MA+ did not have a significant protective effect at pH 5.0 or 7.0.

6.4.2. MA+ effectiveness over time A longer incubation time did not necessarily improve the efficacy of MA+ in preventing PM toxicity (Figure 6.2). MA+, however, showed a significant preventive effect on CIT toxicity (p <

0.01) by 9-33% after incubation for 6 h (Figure 6.2A). OTA toxicity was also significantly reduced by incubation with MA+ by 26-37% (p < 0.0001), but the incubation time did not necessarily improve its efficacy (Figure 6.2B).

6.4.3. Determining optimal effective percentage of MA+ As the percentage of MA+ increased, MA+ showed higher efficacy in preventing the toxicity of both CIT and OTA, but rather exacerbated PAT toxicity (Figure 6.3). The most effective MA+

104

percentage (1.0%; p < 0.0001) significantly protected against CIT-inhibited cell proliferation by

55-58% (Figure 6.3A). Furthermore, 1.0% MA+ protected against OTA-inhibited cell proliferation by 34-43% (p < 0.0001; Figure 6.3B). For PAT, MA+ did not reduce toxicity, but rather increased the toxicity at two highest percentages 0.5% and 1.0%, MA+ significantly increased PAT toxicity by 38-57% (p < 0.0001) and 45-57% (p < 0.0001), respectively. A separate viability test confirmed that this was due to increased cell death at these MA+ percentages (Appendix 4), even though

MA+ alone had no effect on cell viability when used at 1.0% or below (Appendix 5). MA+ did not have a protective effect on MPA or PA toxicity regardless of the incubation time (Figure 6.2D and

E) and the percentage of MA+ (Figure 6.3D and E). Although it was previously observed from the pH trials that 0.2% MA+ appear to have some degree of protective effect for MPA, this effect was not observed throughout other trials (Figure 6.2-6.4D).

6.4.4. MA+ efficacy at different PM concentrations For the last study, five different concentrations of each PM were incubated with 0.2% of MA+ in Milli-Q water at pH 3.0 for 2 h (Figure 6.4). The study showed that the preventive efficacy of

MA+ on PM toxicity largely dependent on PM concentration. MA+ significantly prevented OTA- inhibited cell proliferation with the most significant protection occurring at 12.88 µM by 38-45%

(Figure 6.4B). In contrast, MA+ significantly increased MPA toxicity at 0.3 µM by 20-21% (p <

0.001; Figure 6.4D). MA+ also significantly exacerbated PAT toxicity at 0.7 µM by 36-71% (p <

0.0001; Figure 6.4C). In this study, MA+ did not significant affect PA toxicity (Figure 6.1-6.4E).

6.5. Discussion In previous studies, we characterized the immunomodulatory effect of the following PMs,

CIT, OTA, PAT, MPA and PA on macrophages using biological endpoints including proliferation,

105

cytotoxicity, ROS production and phagocytosis (Oh et al., 2012, 2013). Since BoMac proliferation was the only endpoint to be commonly affected by these PMs, it was chosen to test the efficacy of

MA+.

The MA+ used in this study is a new variant of glucomannan mycotoxin adsorbent (GMA) sold commercially under the name of Mycosorb A+ by Alltech Inc. (Kentucky), and was derived from combinations of blue-green algae and a glucomannan-enriched cell wall component of

Saccharomyces cerevisiae. . Glucomannan reduces the bioavailability of mycotoxins by binding to the mycotoxins via hydrogen bonding and van der Waals force (Yiannikouris et al., 2004), and dietary inclusion of glucomannan added mycotoxin adsorbent product has become a popular postharvest strategy to prevent mycotoxicoses in livestock (Kabak et al., 2006). Studies have shown glucomannan reduces toxic effects caused by aflatoxins and some Fusarium mycotoxins (Dvorska et al., 2007; Karaman et al., 2005; Korosteleva et al., 2007; Raju and Devegowda, 2000). Blue- green algae, such as Spriulina spps., has been tested not only as a potential binder for aflatoxins,

T-2 toxins and OTA but also as a potential anti-oxidant counteraction the oxidative stress from aflatoxins (Fanelli et al., 1985; Manafi et al., 2009, 2012). In comparison to these mycotoxins, there are only a few studies that have investigated the effectiveness of MA+ on PMs, and these have been limited to OTA (Evans and Dawson, 2000; Raju and Devegowda, 2002, 2000).

In the present study, MA+ was most effective at preventing the toxic effects of CIT and OTA among the five tested PMs. This finding is in agreement with studies done by Raju and Devegowda

(2000, 2002), where they observed significant in vitro binding as well as the preventive effect of esterified-glucomannan on OTA toxicity to broiler chicks. Raju and Devegowda (2002), however, reported that pH did not alter the binding efficacy of the glucomannan at 2 mg/kg of OTA, where they observed 25% binding of OTA to the glucomannan at pH 4.5 and 6.5. In contrast, we observed

106

a huge effect of pH on MA+ effectiveness for preventing OTA toxicity, in which MA+ significantly reduced the toxicity of OTA up to 42% at pH 3.0, but not at either pH 5.0 or pH 7.0.

Fruhauf et al., (2012) reported that mineral content, such as smecitite and calcite, is one of important factors that affected the binding of mycotoxins to yeast cell wall components. Since

Raju and Devegowda (2002) used sodium phosphate buffer (PBS) when the esterified- glucomannan was incubated with OTA in vitro, whereas Milli-Q water, was used in the present study, this may explain the enhanced binding of OTA to the glucomannan in their study regardless of the pH difference.

The present study also showed that pH is a major factor that contributes to the efficacy of

MA+ for preventing OTA toxicity. el Khoury and Atoui (2010) postulated that pH can alter the solubility of OTA. For example, OTA is soluble in polar organic solvents at acidic and neutral pH, whereas it is less soluble at high pH. Therefore, in the present study, OTA at pH 3.0 might have undergone a conformational change, which could have increased the strength of hydrogen bonding and van der Waals force between OTA and glucomannan component of MA+ in Milli-Q water during the incubation resulting in higher binding capacity (el Khoury and Atoui, 2010;

Yiannikouris et al., 2004).

Although MA+ also significantly decreased the toxicity of MPA at pH 3 (Figure 6.1D), this was not reproducible in the other trials (Figure 6.2-6.4). This suggests that the combined effect of pH and drying might have affected the potency of MPA during the pH testing trials.

Incubation time was another factor that affected MA+ binding to PMs. For example, 0.2%

MA+ protected the BoMacs from CIT toxicity only after 6 h of the incubation. In contrast, incubation of 0.2% MA+ at pH 3.0 in water effectively reduced OTA toxicity independent of the

107

incubation time (Figure 6.2B), whereas, it did not affect the toxicity of PAT, MPA and PA (Figure

6.2 C, D and E).

The recommended levels of GMA for commercial inclusion in animal feeds with a heavy mycotoxin contamination range from 0.05 % to 0.2%, which has been reported to have beneficial effects in reducing the signs of mycotoxicoses in livestock (Bursian et al., 2004; Evans and

Dawson, 2000; Raju and Devegowda, 2002). For this study, different percentages of MA+, ranging from 0.05% to 1.0%, were tested to determine if different inclusion levels of MA+ would affect

PM toxicity. This study showed that higher percentages of MA+ were more effective at preventing

CIT and OTA toxicity than lower percentages. Similarly, Evans and Dawson (2000) also observed much higher binding of ergotamine with 0.2% Mycosorb, another variant of GMA without blue- green algae from Alltech, as compared to that of 0.01% Mycosorb. This suggests that the percent level of glucomannan product in animal diets may contribute as more important than incubation time, at least for CIT and OTA, for reducing toxicity in vitro. This, however, would require confirmation through in vivo trials.

It should be noted that some of the PM concentrations used in this study were higher than those found in natural conditions (Garon et al., 2006; Müller and Amend, 1997; Reyes-Velázquez et al., 2008). Additionally, it is possible that high concentrations of PMs might have saturated

MA+ binding sites. Therefore, different ranges of PM concentrations were also tested to determine if MA+ efficacy at 0.2% was dependent on PM concentrations. As expected, the efficacy of 0.2%

MA+ for decreasing OTA toxicity was significant at 12.88 and 25.78 µM.

It was interesting to note that at low concentrations of CIT (11.40 and 22.80 µM) and MPA

(0.03 and 0.32 µM), the incorporation of MA+ appeared to slightly increase the toxicity of these

108

two mycotoxins but not at higher concentrations. There is no clear explanation for this phenomena.

In contrast to MPA and PA, CIT toxicity might have effectively been prevented if a longer incubation time, a higher amount of MA+, or a combination of the two were used together.

Regardless of its potential efficacy in preventing CIT and OTA toxicity, MA+ appeared to exacerbate PAT toxicity. For example, 0.1% MA+ appeared to slightly reduce PAT toxicity, but its beneficial effect was negated at 0.2% MA+ and PAT toxicity increased when incubated with

0.5% and 1.0% MA+ for 2 h. A similar observation was made when the various PAT concentrations were incubated with 0.2% MA+, and this reached statistical significant at 0.70 µM.

Previously, Yue et al. (2011) tested 10 different strains of inactive yeast powder, and observed that at least 8 strains significantly reduced the concentration of PAT more than 50% in apple juice after

24 h of incubation at 25°C. Based on this previous binding study, one would expect a decrease in

PAT toxicity to BoMacs. It is possible that soluble components of MA+,which could not be eliminated from the centrifugation method of this study, might have bound to PAT and protects it from degradation during the drying process. In fact, both Figures 6.1C and 6.2C indicates that the drying process slightly decreased the potency of PAT. Nevertheless, further study would be required to provide with more concrete explanations for this.

This is probably the first time a bioassay was used to test the efficacy of mycotoxin adsorbent for preventing toxic effects of PMs. The method allows us to look at the efficacy of MA+ in preventing the toxicity of PMs rather than looking at the binding efficacy of MA+. Therefore, it seems that this method reflects more of the bioavailability of mycotoxins as compared to other binding tests using HPLC (Evans and Dawson, 2000; Yue et al., 2011). However, the method cannot determine whether MA+ binds to PMs or not. Therefore, it would be important to do a follow-up study with HPLC-MS to understand how MA+ or other mycotoxin reducers work in

109

reducing the mycotoxicoses, and therefore, to design a better tool for reducing the bioavailability of mycotoxins to animals.

6.6. Summary and Conclusion In the present study, we have shown that incubation time and pH as well as the amount of

MA+ and the mycotoxin concentrations are the factors that can alter the efficacy of MA+ in preventing the toxicity of some PMs. As for OTA, pH was the major factor, while incubation time and amount of MA+ were the most important for preventing CIT toxicity. MA+ did not appear to effectively prevent the toxicity of PAT, MPA and PA in the present study, but rather increased the toxicity of PAT. Therefore, more research would be required to understand of why MA+ increased

PAT toxicity in the present study.

110

Table 6.1: Summary of experimental variables and constant factors for each Figure.

Constant Factors Experimental (Variable) Factors

Factors tested Incubation Temperature Percentage of pH Time (h) Mycotoxin concentrations (µM) Media (°C ) MA+ (%)

3 Effect of pH 5 2 (Figure 1) 7 CIT: 91.20 0.2 1 Effect of incubation 2 time OTA: 12.88 4 (Figure 2) 6 PAT: 0.70

MPA: 3.2 0.05 Effect of MA+ 22-25°C with 0.1 PA: 29.85 percentage Milli-Q 150 rpm 0.2 (Figure 3) Water horizontal 0.5 shaking 1.0 3

CIT: 11.40, 22.80, 45.60, 91.20, 182.40 2

OTA: 1.61, 3.22, 6.44, 12.88, 25.76 Different PM concentrations 0.2 PAT: 0.09. 0.18, 0.35, 0.70, 1.40 (Figure 4) MPA: 0.03, 0.32, 3.2, 32.0, 64.0

PA: 3.73, 7.45, 14.93, 29.85, 59.70

111

CIT OTA PAT A B C 150 150 150 *** control control *** control * 50 50 50

%proliferation

%proliferation 0 0 0 pos water 1 2 4 6 pos water 1 2 4 6 pos water 1 2 4 6 Incubation time (h) 0 0 1 2 4 6 Incubation time (h) 0 0 1 2 4 6 Incubation time (h) 0 0 1 2 4 6 CIT + + + + + + OTA + + + + + + PAT + + + + + + MA+ - - + + + + MA+ - - + + + + MA+ - - + + + + Dry - + + + + + Dry - + + + + + Dry - + + + + +

MPA PA D E 150 150

control control

50 50

%proliferation

% proliferation % 0 0 pos water 1 2 4 6 Incubation time (h) pos0 water0 1 2 4 6 Incubation time (h) 0 0 1 2 4 6 MPA + + + + + + PA + + + + + + MA+ - - + + + + MA+ - - + + + + Dry - + + + + + Dry - + + + + +

Figure 6.1 (A-E): The effect of pH (3, 5 and 7) on the efficacy of 0.2% MA+ in preventing the PM toxicity on BoMac proliferation.

The dotted line with 'control' label indicates negative control (control with MA+ incubation but without mycotoxin exposure) at 100% proliferation. (*), (**) and (***) denote significant differences at p < 0.05, 0.01 and 0.0001 relative to the control, respectively.

112

CIT OTA PAT A B C 150 150 150 *** control control *** control * 50 50 50

%proliferation

MPA PA D E 150 150

control control

50 50

%proliferation

Figure 6.2 (A-E): The effect of incubation time (1, 2, 4 and 6 h) on the efficacy of 0.2% MA+ in preventing the PM toxicity on BoMac proliferation. The dotted line with 'control' label indicates negative control (control with MA+ incubation but without mycotoxin exposure) at 100% proliferation. (*), (**) and (***) denote significant differences at p < 0.05, 0.01 and 0.0001 relative to the control, respectively.

113

CIT OTA PAT A B C 150 150 150 *** control *** control control

* *** 50 50 50

% proliferation %

% proliferation % 0 0 0 0.00% 0.05% 0.10% 0.20% 0.50% 1.00% 0.00% 0.05% 0.10% 0.20% 0.50% 1.00% 0.00% 0.05% 0.10% 0.20% 0.50% 1.00% % MA+ 0 0.05 0.1 0.2 0.5 1.0 % MA+ 0 0.05 0.1 0.2 0.5 1.0 % MA+ 0 0.05 0.1 0.2 0.5 1.0

MPA PA D E 150 150

control control

50 50

% proliferation %

% proliferation % 0 0 0.00% 0.05% 0.10% 0.20% 0.50% 1.00% 0.00% 0.05% 0.10% 0.20% 0.50% 1.00% % MA+ 0 0.05 0.1 0.2 0.5 1.0 % MA+ 0 0.05 0.1 0.2 0.5 1.0

Figure 6.3 (A-E): The effect of percent amount of MA+ (0.05, 0.1, 0.2, 0.5 and 1 %) on the efficacy of MA+ in preventing the PM toxicity on BoMac proliferation. The dotted line with 'control' label indicates negative control (control with MA+ incubation but without mycotoxin exposure) at 100% proliferation. (*), (**) and (***) denote significant differences at p < 0.05, 0.01 and 0.0001 relative to the control, respectively.

114

Figure 6.4 (A-E): The effect of 0.2 % MA+ at different PM concentrations. The dotted plot line indicates the positive control of PMs without MA+ during incubation, and the solid plot line indicates the treatments that contained both PMs and MA+ during the 2 hrs of incubation. (*), (**) and (***) denote significant differences at p < 0.05, 0.01 and 0.0001 relative to the control, respectively.

115

Chapter 7: Assessing Interactions of Binary Mixtures of Penicillium Mycotoxins (PMs) by using Bovine Macrophage Cell Line (BoMacs) with the Assumption of Additivity Model

7.1. Abstract Penicillium mycotoxins (PMs) are common contaminants found in stored silage for animal feed. Several studies have reported the potential toxicity of single PMs. However, this may not reflect the true risks associated with PMs, because PMs are often found in combinations of two or more, and therefore there are possible interactions, such as synergism or antagonism, among these

PMs. We have previously demonstrated interactions between PMs in their ability to inhibit the proliferation of a bovine macrophage cell line (BoMacs). The study however, was lacking in both statistical and experimental methods for determining true interactions among the binary mixtures of PMs. Therefore, in the present study, we have re-examined the interactions of binary mixtures of PMs by applying the conceptual models of additivity, including independent action (IA) or concentration addition (CA), which are generally used to test mixture toxicity. When the previous

IC25 data (concentration that inhibited the proliferation of BoMacs by 25%) were reanalyzed, the

IA model determined that OTA combined with either CIT, PAT or PA demonstrated synergism (p

< 0.05), while PA combined with either PAT or MPA showed antagonism (p < 0.05). The CA model determined that OTA+PA showed synergism (p < 0.05), while PAT+PA showed antagonism (p < 0.05). However, when a more complex experimental design was carried out using binary combinations of three different dilution levels (i.e. IC25, 0.5IC25, 0.25IC25) under the assumption of the IA model, CIT+OTA and PAT+PA showed synergism at the 0.5IC25s (p < 0.05) with a significant quadratic relationship occurring between the observed and expected effect (p <

0.0001). PAT+PA at their IC25s exhibited antagonism (p < 0.05), and PAT+MPA showed

116

antagonistic interactions across all exposure levels. Other binary mixtures of PMs, including

CIT+MPA, CIT+PA and MPA+PA, also showed antagonism at their IC25s and 0.25IC25s (p <

0.05) with a significant linear effect (p < 0.0001). There was no significant binary interaction observed at the 0.25IL25. The present study demonstrated that binary interactions among PMs are largely dependent on concentrations of the PMs. Furthermre, the results varied depending on which conceptual model was applied to the study. The present study provided an informative and fast initial screening to determine interactions of many binary PM mixtures, taking account of the variability of interactions of PM mixtures. Therefore, the study may be useful as an initial test in narrowing down some important mixtures of PMs with interactions prior to use of more complex experimental models, such as isobolograms or ray ratio design.

7.2. Introduction The term 'mycotoxin' was first coined in 1962 following an outbreak on a turkey farm in

London, England. Since then, mycotoxins have garnered much attention from the public and researchers for their potential to reduce both human and animal food safety (Bennett and Klich,

2003; Forgacs, 1962). Penicillium molds are one of the most common species of mycotoxin- producing fungi, and frequently contaminate crops and stored grains (Lennox et al., 2003).

Penicillium mycotoxins (PMs) can lead to substantial economic loss to the animal feed industry by decreasing crop yields and value, as well as compromising livestock productivity and health

(Schmale III and Munkvold, 2009).

PMs have been shown to cause a wide range of toxic effects in animals. Nephrotoxicity and hepatotoxicity are probably the most common clinical signs reported in animals fed PM contaminated feed (Braunberg et al., 1992; Dickens and Jones, 1965; Sansing et al., 1976). Some

117

studies have also reported their potential modulatory effect on the host’s immune response

(Ferrante et al., 2006; Keblys et al., 2004; Stec et al., 2009). For example, Keblys et al., (2004) reported the dose-dependent inhibition of bovine lymphocyte proliferation by PMs, and a similar observation was made with porcine lymphocytes (Stec et al., 2009). We have previously reported significant inhibitory effects of the following PMs: citrinin (CIT), ochratoxin A (OTA), patulin

(PAT), mycophenolic acid (MPA) and penicillic acid (PA), on the proliferation of a bovine macrophage cell line (BoMacs) (Oh et al., 2012).

Most toxicity studies involving mycotoxins have focused on the toxicity of individual mycotoxins. However, contaminated animal feeds in reality do not usually contain a single mycotoxin, but are rather contaminated with various combinations of mycotoxins. Mansfield et al. for example, reported that more than 50% of maize silage collected from 30 different Pennsylvania dairies between 2001-2002 contained more than one mycotoxin (Mansfield et al., 2008). Therefore, the potential risk of exposure to mycotoxins may be underestimated due to synergistic interactions, or overestimated due to antagonistic interactions among mycotoxins (Sansing et al., 1976;

Shepherd et al., 1981; Veselá et al., 1983).

In the context of PMs, various interactions have also been reported. Specifically, CIT and PA toxicity was increased synergisitically when animal co-exposure occurred with OTA (Sansing et al., 1976; Veselá et al., 1983). We have also determined from a previous study that binary PM mixtures at their respective IC25s (the concentration of PM that inhibited BoMac proliferation by

25%) also synergistically inhibited cell proliferation without causing overt cytotoxicity (Oh et al.,

2012). This previous proliferation study, however, was lacking in both experimental and statistical methods for more precisely determining interactions between PMs. Specifically, conclusions were

118

solely based on single concentration data, and therefore did not take into account binary interactions at different PM concentrations (Meadows et al., 2002; Oh et al., 2012).

There are several different ways of determining the interaction of compounds in a mixture, and one common method is to use the assumption of additivity, whereby the compounds in the mixture are expected to have neutral risk when acting together (Berenbaum, 1981; Cedergreen et al., 2008; Stork et al., 2007). There are two basic concepts for predicting the combined effects of mixtures from the assumption of additivity: independent action (IA) and concentration addition

(CA) (Cedergreen et al., 2008; Kjaerstad et al., 2010; Kortenkamp and Altenburger, 1998). The

IA model assumes that compounds in a mixture have a completely independent mode of action that affects the same endpoint, while the CA model assumes that the compounds have a similar mode of action. In the absence of the mode of action of compounds in mixture, the IA model is often used as a default in assessing the mixture’s toxicity on human or animals, while the CA model is used as a default in environmental risk assessments (European Commission and

Directorate General for Health & Consumers, 2012).

In the present study, the conceptual models were applied to re-examine the potential interaction of binary mixtures of PMs. The mode of action of these PMs on the proliferation of

BoMacs is unknown at this point. Considering that the purpose of this study is to assess the toxicity of PMs on animals at dilutions of certain effect levels such as the mixture of IC25s, rather than an environmental assessment, the IA model was chosen over the CA model to estimate the additivity of PMs in the mixtures in the present study. In addition to this, the IA and CA models were also applied to the proliferation data from the previous study, and re-analyzed in order to compare the results from the present study.

119

7.3. Methods 7.3.1. Data Analysis and Plotting of the Previous PM Toxicity Data The previous proliferation data (Oh et al., 2012) were analyzed and plotted to determine the interaction of the binary mixtures by using independent action (IA) or concentration addition (CA) models.

Based on the assumptions of the IA model, the effect of binary mycotoxin mixture was estimated from the observed effect for the individual mycotoxin exposure using the following equation (Jørgensen, 2013): PEM = PA * PB, where PEM represents the estimated percent proliferation from the mixture, and PA and PB represents the percent proliferation from the individual mycotoxins. The observed percent proliferation from the mixture (POM) and PEM were plotted in Figure 7.1 to determine the interactions among the mycotoxins. For example, if the estimated proliferation fell within the 95% confidence limits (CI) of the observed proliferation, then the assumption of additivity was accepted, and mycotoxins within the mixture were said to have an additive effect. In contrast, when the mean of PEM was outside of the 95% CI of POM, then significant interaction between two mycotoxins occurred (p < 0.05) as either synergism (PEM being located above 95% CI of POM), or antagonism (PEM being located below 95% CI of POM ).

Significant differences at a p-value < 0.05 were denoted as (*) in Figure 7.1.

For the CA model, whole dose-response mixture prediction curves for the binary mixtures were constructed from the proliferation data of individual mycotoxin. The effect concentrations

(ECs) for each proliferation response (eg. the concentrations that inhibited BoMacs proliferation by 99%...1%) were calculated from the curves giving the best fit, including biphasic curves

(SigmaPlot 13). To calculate a mixture CA curve from several curves with different slopes, several

120

EC concentrations are calculated from fixed effect values. The mixture concentrations eliciting a certain effect are then calculated by 1/(pA/ECxA+pB/ ECxB...pn/ECxn), where A and B represent different mycotoxins with p being the proportion of the individual mycotoxin in the mixture

(sum(pA....pn)=1 (Ohlson et al., 2010). From the data, a theoretical (or prediction) curve for each

PM mixture was constructed to compare with the observed proliferation. The error-bars are 95%

CI of POM (observed). For both the IA and CA model, if the prediction curve falls out of 95% CI of the ‘observed’ at a specific concentration, it indicates significant interaction between two PMs, and therefore, is defined as either synergism (PEM being located above 95% CI of POM) or antagonism (PEM being located below 95% CI of POM ). (*) was used to denote significant differences at p-value < 0.05 in Figure 7.2.

7.3.2. Assessing the interactions of PMs using binary combinations over the dilutions of IC25s 7.3.2.1. Cell Preparation The BoMacs (provided by Stabel and Stabel, 1995) were cultured in Roswell Park Memorial

Institute (RPMI) in 1640 medium, supplemented with 2.0 mM L-glutamine, 10% heat inactivated fetal bovine serum (FBS), 100 unit/ml of penicillin, 100 µg/ml of streptomycin, 0.25 µg/ml of amphotericin B, and 25 mM HEPES buffer. All cell culture products were purchased from

2 Invitrogen, Canada. Cells were incubated at 37°C with 5% CO2 in a 75 cm flask, and were grown to above 80% confluence prior to use for the study.

7.3.2.2. BoMacs Exposure to Mycotoxins BoMacs (1.0 x 104) were seeded into 96-well flat-bottom plates, centrifuged at 100 g for 2

o min and incubated at 37 C with 5% CO2 for 2 h until the cells completely adhered to the bottom of the plates. The BoMacs were then exposed to three dilutions (0, 0.5, 0.25) of the mycotoxin

121

concentration that inhibited BoMacs proliferation by 25% (IC25s) in either combinations or individual concentrations of CIT (52.72 μM), OTA (8.91 μM), PAT (0.32 μM), MPA (0.50 μM) and PA (13.90 μM) (Oh et al., 2012). All PMs were ordered from Sigma, Aldrich, U.S.A, and dissolved in ethanol prior to use. After 47 h of mycotoxin exposure, a cell proliferation standard was prepared by seeding another batch of BoMacs (1.0x105, 7.5x104, 5.0x104, 2.5x104 and 1.0x103 cells per well) in each plate, centrifuging at 100 g for 2 min, then incubating for 1 h at 37°C with

5% CO2.

The proliferation assay was carried out as described previously (Oh et al. 2012). The media was removed from all wells by blotting, and the plates were frozen in a -80°C freezer for at least

24 h to lyse cells. The number of cells was estimated using a commercially available cell proliferation kit (Invitrogen, ON, Canada) that uses CyQUANT® GR dye (excitation 494/emission

517 nm) to label nucleic acids. The fluorescence intensity was measured with a 1420 Victor2 Multi- label Counter (PerkinElmer, U.S.A).

7.3.2.3. Data and Statistical Analyses All raw data were normalized to cell number based on the standard curve of each plate, and presented as percent cell proliferation compared to the control. Since the mode of action by which

PMs inhibit BoMac proliferation is unknown, the independent action (IA) model was applied to estimate the additive effect of PMs on cell proliferation (European Commission and Directorate

General for Health & Consumers, 2012).

In this study, two factors were statistically analysed. To test the linear or quadratic trends over dilution levels (IC25, 0.5IC25, 0.25IC25), linear and quadratic orthogonal polynomial contrasts were used between the plots of POM and PEM (SAS 9.3; Kuehl, 2000). Similar to the previous IA model analysis, the difference between POM (indicated as ‘observed’) and PEM

122

(indicated as ‘estimated’) for each respective dilution was evaluated using the 95% confidence interval (CI) of POM in comparison to the mean of PEM as indicated in Figure 7.3.

7.4. Results The proliferation data from the previous studies were re-analyzed and plotted by using a proper IA or CA model to determine the interactions among the mixtures of PMs. From the IA model, the mixtures of OTA with CIT, PAT, or PA showed significant synergism, while the mixtures of PA with PAT, or MPA showed significant antagonistic interactions on BoMac proliferation (p < 0.05; Figure 7.1). From the CA model, OTA+PA was the only mixture that showed a synergistic interaction (p < 0.05; Figure 7.2D), while PAT+PA showed an antagonistic interaction (p < 0.05; Figure 7.2I).

In the follow-up study, mixtures of three different dilutions were tested for each binary mixture of PMs. The mixture of IC25s, the first dilution of IC25 mixtures and the second dilution of IC25s are denoted as ‘IC25,’ ‘0.5IC25’ and ‘0.25IC25 in Figure 7.3, respectively. The study determined that some binary mixtures of PMs had significant interactions between POM (observed) and PEM (estimated) with a presence of a significant linear or quadratic relationship. For example,

CIT+OTA showed a significant difference in the quadratic relationship (p < 0.01), where the

0.5IC25 mixture of POM exhibited significant synergistic inhibition of BoMac proliferation in comparison to that of PEM (p < 0.05), but not at the combinations of IC25 or 0.25IC25 (Figure

7.3A). PAT+PA also showed significant synergism at the mixture of 0.5IC25 (p < 0.05; Figure

7.3I). However, the mixture also exhibited significant antagonism at the mixture of their IC25s (p

< 0.05) with a significant quadratic relationship between POM and PEM (p < 0.05). Similar to

PAT+PA, the mixture of OTA and PAT also showed a similar pattern with a slight quadratic effect

123

(p = 0.10), but did not exhibit either significant synergism or antagonism at any dilution levels

(Figure 7.3B).

The mixture of PAT and MPA showed the strongest antagonistic inhibitory effect on the proliferation of BoMacs among the ten binary mixtures of PMs tested (Figure 7.3H). PAT+MPA showed antagonism across all dilutions (p < 0.05) without the presence of either linear or quadratic relationships. CIT+MPA, CIT+PA and MPA+PA showed significant antagonism at the mixtures of IC25 and 0.5IC25 (p < 0.05; Figure 7.3F, G and J). MPA+PA exhibited a linear relationship between POM and PEM (p < 0.05). CIT+MPA also showed a slight linear relationship (p = 0.09), while CIT+PA did not.

7.5. Discussion Penicillium mycotoxins (PMs) are common toxic compounds found in stored silage for animal feed (Mansfield et al., 2008), and several studies have demonstrated inhibitory effects of individual

PMs on the proliferation of leukocytes (Keblys et al., 2004; Stec et al., 2009). However, the risk observed from the individual PM exposure may not reflect the true risk of PMs due to their potential interactions (Stork et al., 2007), because most PMs are often found in combinations of more than one in animal feed (Mansfield et al., 2008). Previously, we examined the potential interaction among the binary mixtures of PMs, including CIT, OTA, PAT, MPA and PA, at their respective IC25s by using BoMac proliferation (Oh et al., 2012). From the study, we reached a conclusion that the combinations of OTA with CIT, PAT or PA showed synergistic and additive interaction, while the mixture of OTA and MPA showed antagonistic effects. The study, however, was lacking in both experimental and statistical methods; a simple Tukey’s test comparing the percent proliferation of individual PMs to that of PM mixtures may not be sufficient for

124

determining interactions of PMs. Therefore, we have re-examined the previous data under a proper assumption of IA or CA models.

By using either of two models, the effect of mixtures can be estimated from the effect of individual compounds under the assumption of additivity. After that, the predicted effect can be compared to the observed effect of the mixtures, and therefore, can determine a potential interaction among the compounds by rejecting the assumption of additivity (Kjaerstad et al., 2010;

Stork et al., 2007).

Independent action (IA) model assumes that compounds of a mixture have completely independent modes of action affecting the same end point, such as proliferation (Cedergreen et al.,

2013; Plackett and Hewlett, 1963). Under the assumption of the IA model, a similar conclusion as the previous paper was reached. For example, OTA in combination with either CIT, OAT or PA had synergistic effects on the proliferation of BoMacs. However, the combinations of PA with either PAT or MPA were determined to have antagonistic effects, but not the mixture of OTA and

MPA.

In contrast to the IA model, the CA model assumes that the compounds of a mixture behave as simple dilutions of one another (Cedergreen et al., 2013; Kortenkamp and Altenburger, 1998).

While IA uses single substance effect to estimate a mixture effect, the CA model uses effect concentrations to estimate the same effect concentration of the mixture (Cedergreen et al., 2013).

In other words, the CA model takes account of the proportion of the concentrations at various effect levels. For example, if the IC25s of CIT and OTA were determined to be 52.71 μM and 8.91

μM (Oh et al., 2012), then IA model takes the ratio of their effect, which is their IC25 in this case; therefore, the ratio between CIT and OTA is 1:1 in IA model. On the contrary, the CA model takes

125

into account the ratio of concentrations, not the effect, giving a ratio of 6:1 (52.71 μM/8.91 μM) instead of 1:1 from the IA model. Interestingly, under the assumption of the CA model, only

OTA+PA showed a synergistic interaction, while PAT+PA was antagonistic to cell proliferation.

So far, the results of the previous proliferation data that were reanalyzed by using the IA or

CA model were discussed. Clearly, there are differences in the results between these two models, yet both models seem to be in the agreement that OTA+PA mixture exhibited synergism, while

PAT+PA showed antagonism on the proliferation of BoMacs. The synergism of OTA and PA has been also reported in a number of studies. Sansing et al., (1976) described the synergistic toxicity of the mixture on the liver and kidney tissues of mice. Another in vivo feeding study also reported a similar result, in which they observed significant synergistic nephrotoxicity along with the increased mortality rate of mice when these mice consumed both OTA and PA compared to OTA or PA alone (Shepherd et al., 1981). Considering that some common mold species in grain, such as Penicillium viridicatum and Aspergillus ochraceus, can produce both OTA and PA (Bacon et al., 1973; Sansing et al., 1976), assessing the risk of these two mycotoxins also needs to take into account their potential synergism.

PAT plus PA was another mixture that was identified to have an interaction of antagonism from both models. The interaction between PAT and PA has not been largely studied as compared to that of OTA+PA, but there is one study reporting their potential antagonism. Ashoor and Chu

(1973) examined the inhibitory effect of PAT and PA on the in vitro production of yeast alcohol dehydrogenase (ADH) and rabbit-muscle lactic dehydrogenase (LDH). They reported that PAT and PA were non-competitive inhibitors of ADH, but showed a competitive inhibitory effect on

LDH production. This suggests that PAT and PA may exhibit antagonism on LDH as they compete with each other to target one specific mode of action to affect LDH production in this case.

126

However, the study also reported that the mixture also showed an additive effect on ADH production (Ashoor and Chu, 1973). This suggests that the antagonism of PAT and PA may be dependent on different cell types, and this may also apply to other PM mixtures.

The application of IA or CA model provides the threshold point or line of expected additivity effect from the individual exposure data (Cedergreen et al., 2008). This provides a statistical means of distinguishing interactions of synergism, additivity and antagonism between PM mixtures.

However, since the mode of action of these PMs on the proliferation is not known, it is unclear which model is suitable to test interactions for the toxicity of PM mixtures. As it was discussed previously, results can differ depending on which model is applied to a study with the same data set available. In the absence of a mode of action, this would depend on which end point a study is focusing on according to the guidelines of the European Commission and Directorate General for

Health & Consumers, 2012. For example, if a study focuses on the human or animal health perspective, where it examines an effective level, such as certain signs or symptoms of a compound, then IA model is probably a better model for an initial test. If a study is more interested in looking at the concentrations that cause these signs and symptoms, then CA model is probably the best choice (Cedergreen et al., 2013; European Commission and Directorate General for Health &

Consumers, 2012).

When it comes to toxicological testing for mycotoxins, both effect levels and concentrations are important. However, since determining the changes in the effect levels due to the interaction of PMs in mixtures are the primary propose of this study, the IA model is probably a better model to apply for the purpose of this study (Cedergreen et al., 2013). Nevertheless, neither would be any better without the proper information available in determining interactions of mixtures. Previous studies have assessed the observed effect of PM mixtures only at their IC25s (Figure 1 and 2; Oh

127

et al., 2012). This may describe the interaction or the PM mixtures at their IC25s, but this would not reflect interactions of the mixtures at other concentrations or effect levels due to the lack of experimental data. Therefore, in this study we also tested three IC25 dilutions of PM mixtures to see if their interactions changed over the different effective levels by using the IA model.

The results from the present study showed completely different results as compared to the previous proliferation data. For example, the estimated effect of OTA+PA overlapped with the observed effect of OTA+PA across all dilutions of the mixture, indicating that OTA and PA showed additivity, not synergism, regardless of which dilutions were used. The synergism was rather observed from the mixtures of CIT+OTA and PAT+PA at their 0.5IC25s. The trials were repeated in the same way as the previous proliferation trials except for the use of more treatments and different solvents, where ethanol was used to replace DMSO in this study. It is unclear how exactly the solvent affected the potency and interaction of these PMs, but the solvents, DMSO or ethanol, might have altered the synergistic interactions of these mycotoxins (Crankshaw and Raper,

1971).

The synergism between CIT and OTA was observed at IC25s from previous proliferation data with the IA model. The synergism of CIT+OTA was also observed in the present study, but it did not occur at IC25s but rather occurred at the mixture of 0.5IC25s in the present study. There is no clear reason why the synergism occurred at different dilution levels, but we do know that these two mycotoxins exhibited synergism in inhibiting the proliferation of BoMacs. The synergism between CIT and OTA was also previously reported from other studies. For example, the mixtures of CIT and OTA were previously shown to increase chicken embryonic toxicity compared to CIT or OTA alone (Veselá et al., 1983). Other studies also reported synergistic toxicity of CIT and

OTA mixtures on liver and kidney tissues (Sansing et al., 1976; Speijers and Speijers, 2004). Since

128

the results from the present study indicate that the interaction of CIT and OTA only occurred at the 0.5IC25s, their synergistic toxicity in animals may or may not be observable depending on their biological concentrations in animals.

Similar to the previous proliferation data with the IA or CA model, the present study also determined that the mixture of PAT and PA showed antagonism on the proliferation of BoMacs at

IC25s. However, PAT and PA also exhibited a quadratic relationship, where PAT and PA at the combination of 0.5IC25s showed synergistic effects, and additive effects at 0.25IC25s. This suggests that the interaction of PAT and PA is also largely dependent on the concentration, similar to CIT+OTA. The only difference is that PAT+PA showed synergism, antagonism and additive effects depending on their dilution level. Since there is a previous report indicating a possibility that PAT and PA may have the same mode of action for altering the hydrogenase production in yeast and rabbit muscle tissues (Ashoor and Chu, 1973), the CA model would probably provide better information in determining the interaction for this specific mixture.

In comparison to the previous proliferation study, the present study revealed more PM mixtures with antagonistic interactions. The result from this study clearly indicated that PAT and

MPA demonstrate antagonism on the proliferation of BoMacs regardless of their dilution levels.

Antagonism was also observed from the other PM mixtures, including CIT+MPA, CIT+PA and

MPA+PA. This antagonism was largely dependent on the dilution levels, where their significant interaction was most observable at the mixtures of IC25 and 0.5IC25, but not at 0.25IC25. For example, both MPA+PA and CIT+MPA showed significant antagonism with a linear relationship between the observed and estimated proliferation. This indicates that the antagonisms observed from these combinations will likely increase as the concentration of the mixtures increases. For most compounds, interactions, such as antagonism or synergism, generally occur at relatively

129

medium or high concentrations, but they are not likely to occur at low concentrations (European

Commission and Directorate General for Health & Consumers, 2012), and this seems to be the case in most mixtures in this study, where their significant interaction generally diminished as the overall concentration of the compound in the mixtures decreased.

In addition to this, there is a possibility that the compounds in the mixtures that had a linear relationship, including CIT, MPA and PA, may share the same mode of action, targeting the same molecules to inhibit the proliferation of BoMacs when mixed together. The concept is somewhat similar to that of the saturation model used in the kinetics of enzymes (Khera et al., 2014). These mycotoxins at low concentrations would show non-competitive inhibition, and therefore, an additive effect on the proliferation of the BoMacs was observed. When they reach a certain concentration, they compete for the same target rather than going after other targets. Although

CIT+PA did not show any linear effect, the mixture still showed significant antagonism at the mixtures of IC25s and 0.5IC25s, similar to that of MPA+PA and CIT+MPA. This is probably what

Ashoor and Chu (1973) observed in the mixture of PAT and PA, showing competitive inhibition on ADH production but not on LDH production. The yeast and rabbit tissue may share the similar mode of action in producing dehydrogenases, but the amount of the specific molecules for dehydrogenase production may differ depending on the cell types (Ashoor and Chu, 1973). Since

CIT, MPA and PA appear to share the same mode of action for inhibiting the proliferation of

BoMacs, the use of CA might have been a better model to test for their interaction (Cedergreen et al., 2008). Interestingly, the combinations of CIT+MPA were also determined to be antagonistic from the IA analysis of the previous proliferation data, but the mixture was determined to have an additive effect when the CA model was applied. Therefore, it is possible that the following

130

mixtures, MPA+PA, CIT+MPA and CIT+PA, may not show antagonism under the use of the CA model.

7.6. Summary and Conclusion In the present study, the additivity models were applied to determine the potential interactions among the binary mixtures of PMs. First, the previous proliferation data were reanalyzed by using the IA or CA model. Although the two models came up with different results, both models agreed that the mixture of OTA and PA produced synergism on the proliferation of BoMacs, while the mixture of PAT and PA exhibited antagonism. When the trials were repeated with more concentrations of the binary mixtures by diluting the mixtures of IC25s, CIT+OTA and PAT+PA were shown to have synergism at the combination of their 0.5IC25s. Interestingly, PAT+PA also produced antagonism at the mixture of IC25s as it was previously observed. This indicates that the interaction of PMs was largely dependent on their concentrations in the mixture. Other PM mixtures, including PAT+MPA, CIT+MPA, CIT+PA and MPA+PA, showed the antagonism on the proliferation of BoMacs. However, only PAT+MPA appears to have clear antagonism under the IA model regardless of dilution levels, while the antagonisms of other mixtures was only observable at the combination of IC25s and 0.5IC25s but not at 0.25IC25s. The present study provided an informative initial screening to determine interactions of ten binary PM mixtures. For future study, it would be interesting to apply other complex models, such as ray designs, to establish a full interaction curve for those mixtures that showed significant interaction in this study, such as OTA+CIA, PAT+MPA or PAT+PA (Cedergreen et al., 2013; Meadows et al., 2002; Stork et al., 2007).

131

100 Observed Estimated * 80 * * 60 * * 40

%proliferation 0

CIT+PACIT+PACIT+PA PAT+PA CIT+OTACIT+OTACIT+PATCIT+PATCIT+PATCIT+MPACIT+MPA OTA+PAOTA+PAOTA+PA PAT+PAPAT+PAMPA+PAMPA+PAMPA+PA CIT+OTA CIT+MPA OTA+PATOTA+PATOTA+PATOTA+MPAOTA+MPAOTA+MPA PAT+MPAPAT+MPAPAT+MPA Treatments

Figure 7.1: The comparison between ‘observed’ and ‘estimated’ of mixtures of PMs at their respective IC25s by using the IA model. Clear squares indicate the mean of ‘estimated’ proliferation from the PM mixtures. Black squares represent the mean of ‘observed’ proliferation, and the error-bars are 95% CI. If the mean of estimated proliferation overlaps with 95% CI of observed proliferation, then it indicates there is no significant interaction between two PMs. If the means of estimated proliferation falls above or below the 95% CI of observed proliferation, then it indicates a significant synergism or antagonism, respectively (p < 0.05). (*) denoted significant differences at p-value < 0.05.

132

CIT+OTA OTA+PAT 120 120 Observed Estimated Observed Estimated 100 100 80 80 60 60 40 40

% proliferation % 20 proliferation % 20 0 0 0 20 40 60 80 0 5 10 15 20 A Concentrations (μ M) B Concentrations (μ M)

OTA+MPA OTA+PA 120 120 Observed Estimated Observed Estimated 100 100 80 80 60 60 * 40 40

% proliferation % 20 proliferation % 20 0 0 0 5 10 15 0 10 20 30 C Concentrations (μ M) D Concentrations (μ M)

CIT+PAT CIT+MPA 120 120 Observed Estimated Observed Estimated 100 100 80 80 60 60 40 40

% proliferation % 20 proliferation % 20 0 0 0 20 40 60 80 0 20 40 60 80 E Concentrations (μ M) F Concentrations (μ M)

CIT+PA PAT+MPA 120 120 Observed Estimated Observed Estimated 100 100 80 80 60 60 40 40

% proliferation % 20 proliferation % 20 0 0 0 25 50 75 100 0.0 0.5 1.0 1.5 G Concentrations (μ M) H Concentrations (μ M)

PAT+PA MPA+PA 120 120 Observed Estimated Observed Estimated 100 100 80 * 80 60 60 40 40

% proliferation % 20 proliferation % 20 0 0 0 10 20 30 0 10 20 30 I Concentrations (μ M) J Concentrations (μ M)

Figure 7.2 (A-J): The predicted curve of PM mixtures toxicity by using the CA model.

133

CIT+OTA OTA+PAT 125 125 Observed Estimated Observed Estimated 100 100 * 75 75

50 50

25 25

% proliferation % proliferation % 0 0 0.25IC25 0.50IC25 IC25 0.25IC25 0.50IC25 IC25 A Mixture Dilutions B Mixture Dilutions OTA+MPA OTA+PA 125 125 Observed Estimated Observed Estimated 100 100 75 75 50 50 25 25

% proliferation %

% proliferation % 0 0 0.25IC25 0.50IC25 IC25 0.25IC25 0.50IC25 IC25 C Mixture Dilutions D Mixture Dilutions CIT+PAT CIT+MPA 125 125 Observed Estimated Observed Estimated 100 100 * 75 75 * 50 50 25 25

% proliferation % proliferation % 0 0 0.25IC25 0.50IC25 IC25 0.25IC25 0.50IC25 IC25 E F Mixture Dilutions Mixture Dilutions CIT+PA PAT+MPA 125 125 Observed Estimated Observed Estimated 100 100 * * * 75 75 * * 50 50

25 25

% proliferation % proliferation % 0 0 0.25IC25 0.50IC25 IC25 0.25IC25 0.50IC25 IC25 G Mixture Dilutions H Mixture Dilutions

PAT+PA MPA+PA 125 125 Observed Estimated Observed Estimated 100 100 * * 75 75 * * 50 50

25 25

% proliferation %

% proliferation % 0 0 0.25IC25 0.50IC25 IC25 0.25IC25 0.50IC25 IC25 I Mixture Dilutions J Mixture Dilutions

Figure 7.3 (A-J): The comparison between ‘observed’ and ‘estimated’ of PM mixtures.

134

Chapter 8: General Discussion and Conclusion

Since the first reported outbreak on a turkey farm in London, England, mycotoxins have generated much attention from the public and researchers as a potential threat, endangering both human and animal food safety (Bennett and Klich, 2003; Forgacs, 1962). Up until now, over 300 biologically active secondary metabolites of filamentous fungi have been identified as mycotoxins

(Bennett and Klich, 2003), but relatively few of them, specifically those from Fusarium or

Aspergillius species, have been thoroughly studied regarding their potential toxicity due to their history of outbreak (Forgacs, 1962; Marasas et al., 1987; Ramakrishna et al., 1989). Potential mycotoxicoses of Penicillium mycotoxins (PMs), on the other hand, has been widely disregarded despite their high prevalence in crops and grain-based feed for livestock animals, especially ruminant species (Lennox et al., 2003; Mansfield et al., 2008; Pitt, 1987). Therefore, it is important to better understand the risks posed by exposure to these bioactive compounds. To date relatively little is known about the exposure risk to PMs.

A number of PMs, citrinin (CIT), ochratoxin (OTA), patulin (PAT), mycophilic acid (MPA) and penicillic acid (PA) can be found in stored animal feed or food (Lennox et al., 2003; Mansfield et al., 2008). These mycotoxins have been shown to modulate the immune response, which could predispose animals to secondary diseases (Al-Anati and Petzinger, 2006; Ferrante et al., 2008;

Herzog-Soares and Freire, 2004). For example, the consumption of these PMs was shown to reduce the number of leukocytes in the blood as well as in primary (thymus and bursa of Fabricus) and secondary lymphoid organs (spleens and lymph nodes) in poultry or rodent species (Chang et al.,

1979; Gupta et al., 1979; Reddy et al., 1988; Stoev et al., 2004). More recent in vitro studies demonstrated that the PMs decrease leukocyte viability and proliferation, and this likely explains the dramatic reduction of leukocyte numbers reported from in vivo studies (Keblys et al., 2004;

135

Lea et al., 1989; Sorenson et al., 1985; Stec et al., 2008). In agreement with these in vitro studies,

Chapter 3 of this thesis evaluated the potential effect of the following PMs, including CIT, OTA,

PAT, MPA and PA, on cell viability and proliferation of a bovine macrophage cell line (BoMacs)

(Oh et al., 2012). These mycotoxins differentially affected the viability of BoMacs; only PAT and

PA showed evidence of cytotoxicity, where LC50s (concentration that kills 50% of cells) were determined to be 4.46 μM and 175.79 μM, respectively. The potency of these PMs based on their

IC50s (concentration that inhibits 50% cell proliferation) from highest to lowest was: 0.56 µM

(PAT) > 12.88 µM (OTA) > 29.85 µM (PA) > 91.20 µM (CIT) > N/A (MPA). The findings in this study support the results from other in vitro exposure studies in which the authors listed the same order of potency for these PMs on the proliferation of porcine and bovine lymphocytes (Keblys et al., 2004; Stec et al., 2008). The concentrations determined in the study, such as that of OTA, were higher than circulating concentrations observed from in vivo studies (Blank et al., 2003; Höhler et al., 1999). However, there is the potential for their bioaccumulation in certain target tissues, such as kidney, intestine, and cutaneous fat, as well as their interaction with other mycotoxins within these tissues (Sansing et al., 1976).

Besides cell proliferation and viability, some studies have also demonstrated that PMs affect the other functions of macrophages, such as cytokine production, the inflammatory response and phagocytosis (Ferrante et al., 2008; Herzog-Soares and Freire, 2004; Islam et al., 2012;

Johannessen et al., 2005; Liu et al., 2003). These functions of macrophages are important for an adequate immune response, and an inappropriate response could potentially increase susceptibility to secondary infection (Ishikawa et al., 1998) and/or lead to auto-immune disease (Cavin et al.,

2009). Chapter 5 of this thesis also determined that sub-lethal concentrations of PMs differentially altered the gene expression of inflammatory cytokines in BoMacs. The study also demonstrated

136

the potential effect of PMs, specifically OTA, PAT and PA, on ROS production and phagocytosis of Mycobacterium avium spp. paratuberculosis (MAP), an intracellular pathogen that causes

Johne’s disease in cattle. In the study, the cytokine gene expression did not necessarily reflect the endpoints measured, such as PAM-3 stimulated ROS production and phagocytosis of MAP. For example, OTA and MPA significantly changed the gene expression of both IL-1α and IL-6, but these mycotoxins had no effect on ROS production even at higher concentrations. Some cytokines, such as IL-1α and IL-6 are stored in an inactive form in macrophage vacuoles, and become activated and released in response to inflammatory signals, such as LPS stimulation (Maelfait et al., 2008). Therefore, gene expression may not necessarily reflect the actual amount of active cytokine that is released by the macrophage. Given this, future studies may want to explore the effect of PMs on cytokine proteins that are secreted by macrophages.

The study, however, observed some interesting aspects, including that both PAT and PA decreased not only the gene expression of IL-23 but also decreased the intracellular ROS and phagocytosis. A similar phenomenon was also reported from a recent study done by Indramohan and colleagues (2014). These authors observed an equivalent reduction in both ROS production and phagocytosis of Listeria monocytogenes in IL-23 knockdown mice as compared to normal

C57BL/6 mice (Indramohan et al., 2014). This indicates a possible association among the gene expression of IL-23, intracellular ROS production and phagocytosis of intracellular pathogens.

Therefore, it would be interesting to do further correlation studies to determine the relationships among these parameters and to investigate their potential functional interactions in macrophages.

While several studies that have investigated the toxic effect of individual PMs (Keblys et al.,

2004; Stec et al., 2008), only a few have assessed the potential interactions among these PMs. PMs in reality do not occur individually, but most often are found in combinations of more than one

137

mycotoxin (Mansfield et al., 2008; Martins et al., 2002). Furthermore, some of these PMs can exhibit either synergistic or antagonistic interactions, leading to either underestimation or overestimation of the risk associated with mixed compounds (Stork et al., 2007). Chapter 3 of this thesis showed that OTA when combined with CIT, PAT and PA showed potential synergistic inhibitory effects on BoMac proliferation. However, the study was lacking in both experimental and statistical methods for thoroughly investigating PM interactions. Our conclusion was solely based on single concentrations of binary mixtures, and there was no clear line of distinction between additivity, synergism and antagonism.

Therefore, the interactions of binary PM mixtures were re-examined more thoroughly in

Chapter 7 under the assumption of additivity using the independent action (IA) and the concentration addition (CA) models. When previous proliferation data were reanalyzed, both the

IA and CA models determined synergistic relationships for OTA and PA, and antagonistic relationships for PAT and PA. However, when three different dilutions of IC25 mixtures were subsequently retested using the IA model, the result was completely different from that of the reanalyzed proliferation data from Chapter 3. In Chapter 7, CIT+OTA and PAT+PA showed synergism at medium tested concentrations. The observed proliferation curve of these mixtures also had a quadratic effect in comparison to the estimated proliferation curve. This quadratic effect indicates that the interactions are largely dependent on the effect levels of the mycotoxins in the mixtures. For example, PAT+PA exhibited additivity at low, antagonism at medium and synergism at the high effect level of PM mixtures. The antagonistic effects were also determined in other binary mixtures of PMs, including PAT+MPA, CIT+MPA, CIT+PA and MPA+PA. However, the observed proliferation curve of the following PM mixtures, CIT+MPA and MPA+PA, had a significant linear effect in comparison to the estimated proliferation curve. The linear relationship

138

between the observed and estimated proliferation curve suggests that the antagonism is dependent on the effect level of PMs, where their antagonistic effects will likely increase as the effect level of the mixtures increase. This study provided an informative initial screening to determine interactions of many binary PM mixtures. However, more complex experimental designs, isobologram and ray ratio designs for example, could be explored in future studies to better approximate PM interactions within mixtures (Meadows et al., 2002; Stork et al., 2007). These models would require much more experimental effort to establish full additivity curves compared to 3-point curves that were used in our study.

In the previous paragraphs, we have summarized that PMs had significant effects on the activities of bovine macrophages (Oh et al., 2012; Sorenson et al., 1985, 1986). These macrophage functions are under the control of epigenetic regulation (Bowdridge and Gause, 2010; Ishii et al.,

2009). Therefore, there is a possibility that the effect of PMs observed may be attributed to the changes in epigenetic patterns, such as DNA or histone acetylation or methylation patterns, and the activities of the enzymes regulating them (Marin-Kuan et al., 2007, 2008). Chapter 4 of this thesis demonstrated that individual PMs at their respective IC25s did not affect the gene expression of epigenetic enzymes. However, when OTA was co-exposed with either CIT or PA, a strong synergistic effect on the gene expression of the epigenetic enzymes methyltransferases (DNMT)-

3s, histone demethylase (JMJD-3) and histone deacetylase (HDAC)-3 was observed. Studies have suggested that the activities of these enzymes are closely linked to cell proliferation, viability, and inflammatory response (Begriche et al., 2011; Kurita et al., 2010; Sun et al., 2011; Wilson et al.,

2006). Therefore, it is possible that the synergistic suppression in BoMacs’ proliferation that was observed in Chapter 3 might be due to the changes in the expression of the epigenetic enzymes observed in Chapter 4 of this thesis. Unfortunately due to time constraints, we did not evaluate the

139

potential synergistic effect of these combinations on the gene expression of cytokines, ROS production and phagocytosis. Therefore, it will be important to address the potential synergistic effect of PMs on these parameters to determine the possible link between the epigenetic changes and the macrophage functions.

So far, it was discussed in this thesis that PMs have potential toxic effects on BoMacs, specifically, that these mycotoxins affected the cell proliferation, viability, cytokine gene expression, ROS production and phagocytosis of MAP with respect to the bovine macrophages

(Oh et al., 2012, 2013). PMs differentially modulated the macrophage functions and viability, and this in turn, could predispose animals to secondary diseases, such as Johne’s disease. The next important step after determining their potential immunomodulatory effect would be to find possible solutions to resolve these issues. Therefore, Chapter 6 of this thesis looked into the efficacy of MA+ for preventing the potential toxicity of PMs using BoMacs proliferation as a bioassay endpoint. From this study, MA+ showed the highest efficacy preventing OTA toxicity at pH 3.0 but not at pH 5 or 7. Raju and Devegowda, (2002, 2000) also demonstrated a beneficial effect of MA+ for preventing OTA toxicity using both a in vitro binding test and in vivo studies with broiler chicks. MA+ also showed a preventive effect on CIT toxicity, but this required a longer incubation time than 6 h and a higher percent inclusion of MA+ than 0.5%. MA+ was not effective for preventing the toxicity of the other mycotoxins, PAT, MPA and PA, tested in the trial.

There are, however, other factors to consider, such as temperature and the mineral content in the incubation media. These factors could also largely contribute to the efficacy of MA+ as well as the toxic potency of mycotoxins (Evans and Dawson, 2000; Fruhauf et al., 2012).

Overall, this study provides insight into the potential immunomodulatory effect of PMs by using the BoMacs in vitro model. The study demonstrated that PMs differentially affected cell

140

viability, proliferation, ROS production, phagocytosis as well as the gene expression of the cytokines and epigenetic enzymes that regulate cellular activities. However, all the trials were carried out as in vitro studies using a bovine macrophage cell line (BoMacs). In comparison to in vivo studies, in vitro studies provide more control of variables, and allow researchers to investigate specific effects of mycotoxins (Emami, 2006). At the same time, it is a reality that most in vitro studies, including the trials in this thesis, do not consider all other biological and environmental factors that could alter the effect of mycotoxins or MA+ efficacy (Emami, 2006; Raju and

Devegowda, 2002, 2000). Therefore, it would also be important to continue on with some follow- up studies to examine the immunomodulation of PMs by using in vivo ruminant models in relation to the prevalence of common infectious diseases, such as Johne’s disease.

In the end, the most important question to ask would be, “Can these Penicillium mycotoxicoses be prevented?” in terms of the animal industry’s view. Proper feed management is the most effective way to reduce the risk of mycotoxin occurrence in livestock feed post-harvest.

However, if a high prevalence of mycotoxins is observed, there are several methods, such as inclusion of clay or other mycotoxin binders, such as GMA or MA+, available to reduce the risk of mycotoxicoses in animals (reviewed by Kabak et al. 2006). This, however, would depend on the types of PMs, as the efficacy of mycotoxin binders largely varies with different PMs (Fruhauf et al., 2012; Kabak et al., 2006). In addition to this, in reality, mycotoxins do not occur individually but most often are found in combinations in ensiled grain- and/or crop-based livestock feed, making it difficult to truly assess their risk and toxicity (Mansfield et al., 2008; Oh et al., 2012;

Stoev et al., 2004; Veselá et al., 1983). Therefore, in order to gain a greater perspective on how to detect and prevent mycotoxins in the agricultural industry, combinational mycotoxicoses should be considered in future studies.

141

References Abrunhosa, L., Paterson, R.R.M., and Venâncio, A. (2010). Biodegradation of Ochratoxin A for Food and Feed Decontamination. Toxins 2, 1078–1099.

Aggarwal, S., Gollapudi, S., Yel, L., Gupta, A.S., and Gupta, S. (2000). TNF-alpha-induced apoptosis in neonatal lymphocytes: TNFRp55 expression and downstream pathways of apoptosis. Genes Immun. 1, 271–279.

Agger, K., Cloos, P.A.C., Christensen, J., Pasini, D., Rose, S., Rappsilber, J., Issaeva, I., Canaani, E., Salcini, A.E., and Helin, K. (2007). UTX and JMJD3 are histone H3K27 demethylases involved in HOX gene regulation and development. Nature 449, 731–734.

Agger, K., Cloos, P.A.C., Rudkjaer, L., Williams, K., Andersen, G., Christensen, J., and Helin, K. (2009). The H3K27me3 demethylase JMJD3 contributes to the activation of the INK4A-ARF locus in response to oncogene- and stress-induced senescence. Genes Dev. 23, 1171–1176.

Ahn, H.J., Park, J., Song, J.S., Ju, M.K., Kim, M.S., Ha, H., Song, K.H., and Kim, Y.S. (2007). Mycophenolic acid inhibits oleic acid-induced vascular smooth muscle cell activation by inhibiting cellular reactive oxygen species. Transplantation 84, 634–638.

Allan, L.A., Morrice, N., Brady, S., Magee, G., Pathak, S., and Clarke, P.R. (2003). Inhibition of caspase-9 through phosphorylation at Thr 125 by ERK MAPK. Nat. Cell Biol. 5, 647–654.

Almeida, A.P., Corrêa, B., Mallozzi, M.A.B., Sawazaki, E., and Soares, L.M.V. (2000). Mycoflora and aflatoxin/fumonisin production by fungal isolates from freshly harvested corn hybrids. Braz. J. Microbiol. 31.

Alsberg, C., and Black, O. (1913). Contributions to the study of maize deterioration; biochemical and toxicological investigations of Penicillium puberulum and Penicillium stoloniferum (1913) US Dept Agr Bur Plant Ind Bull. Dept Agr Bur Plant Ind Bull 270, 1–47.

Alvarez, L., Gil, A.G., Ezpeleta, O., García-Jalón, J.A., and López de Cerain, A. (2004). Immunotoxic effects of Ochratoxin A in Wistar rats after oral administration. Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 42, 825–834.

Alvarez-Erviti, L., Leache, C., González-Peñas, E., and de Cerain, A.L. (2005). Alterations induced in vitro by ochratoxin A in rat lymphoid cells. Hum. Exp. Toxicol. 24, 459–466.

Amiri, A., and Bompeix, G. (2005). Diversity and population dynamics of Penicillium spp. on apples in pre- and postharvest environments: consequences for decay development. Plant Pathol. 54, 74–81.

Al-Anati, L., and Petzinger, E. (2006). Immunotoxic activity of ochratoxin A. J. Vet. Pharmacol. Ther. 29, 79–90.

Al-Anati, L., Katz, N., and Petzinger, E. (2005). Interference of arachidonic acid and its metabolites with TNF-alpha release by ochratoxin A from rat liver. Toxicology 208, 335–346.

142

Andersen, B., Smedsgaard, J., and Frisvad, J.C. (2004). Penicillium expansum: consistent production of patulin, chaetoglobosins, and other secondary metabolites in culture and their natural occurrence in fruit products. J. Agric. Food Chem. 52, 2421–2428.

Anderson, S.E., Franko, J., Jackson, L.G., Wells, J.R., Ham, J.E., and Meade, B.J. (2012). Irritancy and Allergic Responses Induced by Exposure to the Indoor Air Chemical 4-Oxopentanal. Toxicol. Sci. 127, 371–381.

Aninat, C., Hayashi, Y., André, F., and Delaforge, M. (2001). Molecular requirements for inhibition of cytochrome p450 activities by roquefortine. Chem. Res. Toxicol. 14, 1259–1265.

Arzu Koçkaya, E., Selmanoğlu, G., Ozsoy, N., and Gül, N. (2009). Evaluation of patulin toxicity in the thymus of growing male rats. Arh. Hig. Rada Toksikol. 60, 411–418.

Ashburner, B.P., Westerheide, S.D., and Baldwin, A.S. (2001). The p65 (RelA) subunit of NF- kappaB interacts with the histone deacetylase (HDAC) corepressors HDAC1 and HDAC2 to negatively regulate gene expression. Mol. Cell. Biol. 21, 7065–7077.

Ashoor, S.H., and Chu, F.S. (1973). Inhibition of alcohol and lactic dehydrogenases by patulin and penicillic acid in vitro. Food Cosmet. Toxicol. 11, 617–624.

Assaf, H., Azouri, H., and Pallardy, M. (2004). Ochratoxin A induces apoptosis in human lymphocytes through down regulation of Bcl-xL. Toxicol. Sci. Off. J. Soc. Toxicol. 79, 335–344.

Auerbach, H., Oldenburg, E., and Weissbach, F. (1998). Incidence ofPenicillium roqueforti and roquefortine C in silages. J. Sci. Food Agric. 76, 565–572.

Bacon, C.W., Sweeney, J.G., Robbins, J.D., and Burdick, D. (1973). Production of penicillic acid and ochratoxin A on poultry feed by Aspergillus ochraceus: temperature and moisture requirements. Appl. Microbiol. 26, 155–160.

Bando, M., Hasegawa, M., Tsuboi, Y., Miyake, Y., Shiina, M., Ito, M., Handa, H., Nagai, K., and Kataoka, T. (2003). The mycotoxin penicillic acid inhibits Fas ligand-induced apoptosis by blocking self-processing of caspase-8 in death-inducing signaling complex. J. Biol. Chem. 278, 5786–5793.

Bardeesy, N., Morgan, J., Sinha, M., Signoretti, S., Srivastava, S., Loda, M., Merlino, G., and DePinho, R.A. (2002). Obligate roles for p16(Ink4a) and p19(Arf)-p53 in the suppression of murine pancreatic neoplasia. Mol. Cell. Biol. 22, 635–643.

Barradas, M., Anderton, E., Acosta, J.C., Li, S., Banito, A., Rodriguez-Niedenführ, M., Maertens, G., Banck, M., Zhou, M.-M., Walsh, M.J., et al. (2009). Histone demethylase JMJD3 contributes to epigenetic control of INK4a/ARF by oncogenic RAS. Genes Dev. 23, 1177–1182.

Bartman, C.D., Doerfler, D.L., Bird, B.A., Remaley, A.T., Peace, J.N., and Campbell, I.M. (1981). Mycophenolic Acid Production by Penicillium brevicompactum on Solid Media. Appl. Environ. Microbiol. 41, 729–736.

143

Basu, A., and Haldar, S. (1998). The relationship between BcI2, Bax and p53: consequences for cell cycle progression and cell death. Mol. Hum. Reprod. 4, 1099–1109.

Baudrimont, I., Betbeder, A.M., Gharbi, A., Pfohl-Leszkowicz, A., Dirheimer, G., and Creppy, E.E. (1994). Effect of superoxide dismutase and catalase on the nephrotoxicity induced by subchronical administration of ochratoxin A in rats. Toxicology 89, 101–111.

Beaulieu, N., Morin, S., Chute, I.C., Robert, M.-F., Nguyen, H., and MacLeod, A.R. (2002). An essential role for DNA methyltransferase DNMT3B in cancer cell survival. J. Biol. Chem. 277, 28176–28181.

Begriche, K., Massart, J., Robin, M.-A., Borgne-Sanchez, A., and Fromenty, B. (2011). Drug- induced toxicity on mitochondria and lipid metabolism: mechanistic diversity and deleterious consequences for the liver. J. Hepatol. 54, 773–794.

Bennett, J.W., and Klich, M. (2003). Mycotoxins. Clin. Microbiol. Rev. 16, 497–516.

Berenbaum, M.C. (1981). Criteria for analyzing interactions between biologically active agents. Adv. Cancer Res. 35, 269–335.

Biel, M., Wascholowski, V., and Giannis, A. (2005). Epigenetics--an epicenter of gene regulation: histones and histone-modifying enzymes. Angew. Chem. Int. Ed Engl. 44, 3186–3216.

Blank, R., Rolfs, J.-P., Südekum, K.-H., Frohlich, A.A., Marquardt, R.R., and Wolffram, S. (2003). Effects of chronic ingestion of ochratoxin a on blood levels and excretion of the mycotoxin in sheep. J. Agric. Food Chem. 51, 6899–6905.

Bommi, P.V., Dimri, M., Sahasrabuddhe, A.A., Khandekar, J., and Dimri, G.P. (2010). The polycomb group protein BMI1 is a transcriptional target of HDAC inhibitors. Cell Cycle Georget. Tex 9, 2663–2673.

Boorman, G.A., Hong, H.L., Dieter, M.P., Hayes, H.T., Pohland, A.E., Stack, M., and Luster, M.I. (1984). Myelotoxicity and macrophage alteration in mice exposed to ochratoxin A. Toxicol. Appl. Pharmacol. 72, 304–312.

Borrows, R., Chusney, G., Loucaidou, M., James, A., Lee, J., Tromp, J.V., Owen, J., Cairns, T., Griffith, M., Hakim, N., et al. (2006). Mycophenolic acid 12-h trough level monitoring in renal transplantation: association with acute rejection and toxicity. Am. J. Transplant. Off. J. Am. Soc. Transplant. Am. Soc. Transpl. Surg. 6, 121–128.

Bourdiol, D., Escoula, L., and Salvayre, R. (1990). Effect of patulin on microbicidal activity of mouse peritoneal macrophages. Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 28, 29–33.

Boutrif, E. (1995). FAO programmes for prevention, regulation, and control of mycotoxins in food. Nat. Toxins 3, 322–326; discussion 341.

144

Boutrif, E., and Canet, C. (1998). Mycotoxin prevention and control: FAO programmes. Rev. Med Vet 149, 681–694.

Bowdridge, S., and Gause, W.C. (2010). Regulation of alternative macrophage activation by chromatin remodeling. Nat. Immunol. 11, 879–881.

Boysen, M., Skouboe, P., Frisvad, J., and Rossen, L. (1996). Reclassification of the Penicillium roqueforti group into three species on the basis of molecular genetic and biochemical profiles. Microbiol. Read. Engl. 142 (Pt 3), 541–549.

Braunberg, R.C., Gantt, O., Barton, C., and Friedman, L. (1992). In vitro effects of the nephrotoxins ochratoxin A and citrinin upon biochemical function of porcine kidney. Arch. Environ. Contam. Toxicol. 22, 464–470.

Bringold, F., and Serrano, M. (2000). Tumor suppressors and oncogenes in cellular senescence☆. Exp. Gerontol. 35, 317–329.

Brown, M.H., Szczech, G.M., and Purmalis, B.P. (1976). Teratogenic and toxic effects of ochratoxin A in rats. Toxicol. Appl. Pharmacol. 37, 331–338.

Burkin, A.A., and Kononenko, G.P. (2010). Producers of mycophenolic acid in ensiled and grain feeds. Appl. Biochem. Microbiol. 46, 545–550.

Bursian, S.J., Mitchell, R.R., Yamini, B., Fitzgerald, S.D., Murphy, P.A., Fernandez, G., Rottinghaus, G.E., Moran, L., Leefers, K., Choi, I., et al. (2004). Efficacy of a commercial mycotoxin binder in alleviating effects of ochratoxin A, fumonisin B1, moniliformin and zearalenone in adult mink. Vet. Hum. Toxicol. 46, 122–129.

Buttke, T.M., and Sandstrom, P.A. (1994). Oxidative stress as a mediator of apoptosis. Immunol. Today 15, 7–10.

Calabrese, E.J., and Baldwin, L.A. (2003). H ORMESIS : The Dose-Response Revolution. Annu. Rev. Pharmacol. Toxicol. 43, 175–197.

Carvalho, J., Oishi, B., Pandey, A., and Soccol, C. (2005). Biopigments from Monascus: strains selection, citrinin production and color stability. Braz Arch Biol Technol 48, 885–894.

CAST (2003). Mycotoxins — risks in plant, animal and human systems, Task Force Report, No. 139. (Ames, Iowa), pp. 1–191.

Cavin, C., Delatour, T., Marin-Kuan, M., Fenaille, F., Holzhäuser, D., Guignard, G., Bezençon, C., Piguet, D., Parisod, V., Richoz-Payot, J., et al. (2009). Ochratoxin A-mediated DNA and protein damage: roles of nitrosative and oxidative stresses. Toxicol. Sci. Off. J. Soc. Toxicol. 110, 84–94.

Cedergreen, N., Christensen, A.M., Kamper, A., Kudsk, P., Mathiassen, S.K., Streibig, J.C., and Sørensen, H. (2008). A review of independent action compared to concentration addition as

145

reference models for mixtures of compounds with different molecular target sites. Environ. Toxicol. Chem. SETAC 27, 1621–1632.

Cedergreen, N., Svendsen, C., and Backhaus, T. (2013). Toxicity Prediction of Chemical Mixtures. In Encyclopedia of Environmental Management., (New York: Taylor and Francis), pp. 2572–2581.

Chagas, G.M., Campello, A.P., and Klüppel, M.L.W. (1992). Mechanism of citrinin-induced dysfunction of mitochondria. I. Effects on respiration, enzyme activities and membrane potential of renal cortical mitochondria. J. Appl. Toxicol. 12, 123–129.

Chaigne-Delalande, B., Guidicelli, G., Couzi, L., Merville, P., Mahfouf, W., Bouchet, S., Molimard, M., Pinson, B., Moreau, J.-F., and Legembre, P. (2008). The immunosuppressor mycophenolic acid kills activated lymphocytes by inducing a nonclassical actin-dependent necrotic signal. J. Immunol. Baltim. Md 1950 181, 7630–7638.

Chan, W.-H. (2007). Citrinin induces apoptosis via a mitochondria-dependent pathway and inhibition of survival signals in embryonic stem cells, and causes developmental injury in blastocysts. Biochem. J. 404, 317–326.

Chang, C.F., Huff, W.E., and Hamilton, P.B. (1979). A leucocytopenia induced in chickens by dietary ochratoxin A. Poult. Sci. 58, 555–558.

Chao, D.T., Linette, G.P., Boise, L.H., White, L.S., Thompson, C.B., and Korsmeyer, S.J. (1995). Bcl-XL and Bcl-2 repress a common pathway of cell death. J. Exp. Med. 182, 821–828.

Chapuis, A.G., Paolo Rizzardi, G., D’Agostino, C., Attinger, A., Knabenhans, C., Fleury, S., Acha- Orbea, H., and Pantaleo, G. (2000). Effects of mycophenolic acid on human immunodeficiency virus infection in vitro and in vivo. Nat. Med. 6, 762–768.

Charoenpornsook, K., Fitzpatrick, J.L., and Smith, J.E. (1998). The effects of four mycotoxins on the mitogen stimulated proliferation of bovine peripheral blood mononuclear cells in vitro. Mycopathologia 143, 105–111.

Chaudhry, V., Cornblath, D.R., Griffin, J.W., O’Brien, R., and Drachman, D.B. (2001). Mycophenolate mofetil: a safe and promising immunosuppressant in neuromuscular diseases. Neurology 56, 94–96.

Chen, L., Fischle, W., Verdin, E., and Greene, W.C. (2001). Duration of nuclear NF-kappaB action regulated by reversible acetylation. Science 293, 1653–1657.

Chen, M., Guerrero, A.D., Huang, L., Shabier, Z., Pan, M., Tan, T.-H., and Wang, J. (2007). Caspase-9-induced mitochondrial disruption through cleavage of anti-apoptotic BCL-2 family members. J. Biol. Chem. 282, 33888–33895.

Choi, H., Choi, H., Han, J., Jin, S.H., Park, J.-Y., Shin, D.W., Lee, T.R., Kim, K., Lee, A.-Y., and Noh, M. (2013). IL-4 Inhibits the Melanogenesis of Normal Human Melanocytes through the JAK2–STAT6 Signaling Pathway. J. Invest. Dermatol. 133, 528–536.

146

Cicinnati, V.R., Hou, J., Lindemann, M., Horn, P.A., Sotiropoulos, G.C., Paul, A., Gerken, G., and Beckebaum, S. (2009). Mycophenolic acid impedes the antigen presenting and lymph node homing capacities of human blood myeloid dendritic cells. Transplantation 88, 504–513.

Ciegler, A. (1977). Patulin. In Mycotoxins in Human and Animal Health., (Park Forest South, Ill.: Pathotox Publishers Inc), pp. 609–624.

Ciegler, A., and Kurtzman, C.P. (1970). Penicillic acid production by blue-eye fungi on various agricultural commodities. Appl. Microbiol. 20, 761–764.

Ciegler, A., Detroy, R.W., and Lillehoj, E.B. (1971). Patulin, penicillic acid and other carcinogenic lactones. Microb. Toxin Ch6, 409–434.

Ciegler, A., Mintzlaff, H.J., Weisleder, D., and Leistner, L. (1972). Potential production and detoxification of penicillic acid in mold-fermented sausage (salami). Appl. Microbiol. 24, 114– 119.

Ciegler, A., Vesonder, R., and Jackson, L. (1977). Production and Biological Activity of Patulin and Citrininfrom Penicillium expansum. Appl Environ Microbiol. 33, 1004–1006.

Cohn, R.G., Mirkovich, A., Dunlap, B., Burton, P., Chiu, S.H., Eugui, E., and Caulfield, J.P. (1999). Mycophenolic acid increases apoptosis, lysosomes and lipid droplets in human lymphoid and monocytic cell lines. Transplantation 68, 411–418.

Colic, M., Stojic-Vukanic, Z., Pavlovic, B., Jandric, D., and Stefanoska, I. (2003). Mycophenolate mofetil inhibits differentiation, maturation and allostimulatory function of human monocyte- derived dendritic cells. Clin. Exp. Immunol. 134, 63–69.

Crankshaw, D.P., and Raper, C. (1971). The effect of solvents on the potency of chlordiazepoxide, diazepam, medazepam and nitrazepam. J. Pharm. Pharmacol. 23, 313–321.

Creppy, E.E. (2002). Update of survey, regulation and toxic effects of mycotoxins in Europe. Toxicol. Lett. 127, 19–28.

Creppy, E.E., Størmer, F.C., Röschenthaler, R., and Dirheimer, G. (1983). Effects of two metabolites of ochratoxin A, (4R)-4-hydroxyochratoxin A and ochratoxin alpha, on immune response in mice. Infect. Immun. 39, 1015–1018.

Croswell, A., Amir, E., Teggatz, P., Barman, M., and Salzman, N.H. (2009). Prolonged impact of antibiotics on intestinal microbial ecology and susceptibility to enteric Salmonella infection. Infect. Immun. 77, 2741–2753.

Dalcero, A., Magnoli, C., Chiacchiera, S., Palacios, G., and Reynoso, M. (1997). Mycoflora and incidence of aflatoxin B1, zearalenone and deoxynivalenol in poultry feeds in Argentina. Mycopathologia 137, 179–184.

147

Dalcero, A., Magnoli, C., Luna, M., Ancasi, G., Reynoso, M.M., Chiacchiera, S., Miazzo, R., and Palacio, G. (1998). Mycoflora and naturally occurring mycotoxins in poultry feeds in Argentina. Mycopathologia 141, 37–43.

Daniel, P.T., Kroidl, A., Cayeux, S., Bargou, R., Blankenstein, T., and Dörken, B. (1997). Costimulatory signals through B7.1/CD28 prevent T cell apoptosis during target cell lysis. J. Immunol. Baltim. Md 1950 159, 3808–3815.

Delcuve, G.P., Rastegar, M., and Davie, J.R. (2009). Epigenetic control. J. Cell. Physiol. 219, 243– 250.

Deng, G., Xiao, Y., Zhou, Z., Nagai, Y., Zhang, H., Li, B., and Greene, M.I. (2012). Molecular and biological role of the FOXP3 N-terminal domain in immune regulation by T regulatory/suppressor cells. Exp. Mol. Pathol. 93, 334–338.

Deng, Y., Ren, X., Yang, L., Lin, Y., and Wu, X. (2003). A JNK-dependent pathway is required for TNFalpha-induced apoptosis. Cell 115, 61–70.

Devi, P., D’Souza, L., Kamat, T., Rodrigues, C., and Naik, C. (2009). Batch culture fermentation of Penicillium chrysogenum and a report on the isolation, purification, identification and antibiotic activity of citrinin. IJMS 38, 38–44.

Dhuley, J.N. (1997). Effect of some Indian herbs on macrophage functions in ochratoxin A treated mice. J. Ethnopharmacol. 58, 15–20.

Dickens, F., and Jones, H.E.H. (1965). Further Studies on the Carcinogenic Action of Certain Lactones and Related Substances in the Rat and Mouse. Br. J. Cancer 19, 392–403.

Diekman, M.A., and Green, M.L. (1992). Mycotoxins and reproduction in domestic livestock. J. Anim. Sci. 70, 1615–1627.

Dinarello, C.A. (1994). The interleukin-1 family: 10 years of discovery. FASEB J. Off. Publ. Fed. Am. Soc. Exp. Biol. 8, 1314–1325.

Dooms, H., Kahn, E., Knoechel, B., and Abbas, A.K. (2004). IL-2 induces a competitive survival advantage in T lymphocytes. J. Immunol. Baltim. Md 1950 172, 5973–5979.

Dubus, I., Vendrely, B., Christophe, I., Labouyrie, J.-P., Delmas, Y., Bonnet, J., and Combe, C. (2002). Mycophenolic acid antagonizes the activation of cultured human mesangial cells. Kidney Int. 62, 857–867.

Dun, B., Sharma, A., Xu, H., Liu, H., Bai, S., Zeng, L., and She, J.-X. (2013). Transcriptomic changes induced by mycophenolic acid in gastric cancer cells. Am. J. Transl. Res. 6, 28–42.

Duncan, M.D., and Wilkes, D.S. (2005). Transplant-related immunosuppression: a review of immunosuppression and pulmonary infections. Proc. Am. Thorac. Soc. 2, 449–455.

148

Duong, T.A. (1996). Infection due to Penicillium marneffei, an emerging pathogen: review of 155 reported cases. Clin. Infect. Dis. Off. Publ. Infect. Dis. Soc. Am. 23, 125–130.

Dvorska, J.E., Pappas, A.C., Karadas, F., Speake, B.K., and Surai, P.F. (2007). Protective effect of modified glucomannans and organic selenium against antioxidant depletion in the chicken liver due to T-2 toxin-contaminated feed consumption. Comp. Biochem. Physiol. Toxicol. Pharmacol. CBP 145, 582–587.

Dziarski, R. (1985). Comparison of in vitro and in vivo mitogenic and polyclonal antibody and autoantibody responses to peptidoglycan, LPS, protein A, PWM, PHA and Con A in normal and autoimmune mice. J. Clin. Lab. Immunol. 16, 93–109.

Dzidic, A., Prgomet, C., Mohr, A., Meyer, K., Bauer, J., Meyer, H.H.D., and Pfaffl, M.W. (2006). Effects of mycophenolic acid on inosine monophosphate dehydrogenase I and II mRNA expression in white blood cells and various tissues in sheep. J. Vet. Med. A Physiol. Pathol. Clin. Med. 53, 163–169.

Dzidic, A., Meyer, H.H.D., Bauer, J., and Pfaffl, M.W. (2010). Long-term effects of mycophenolic acid on the immunoglobulin and inflammatory marker-gene expression in sheep white blood cells. Mycotoxin Res. 26, 235–240.

EFSA (2012). Scientific Opinion on the risks for public and animal health related to the presence of citrinin in food and feed. EFSA J.10 p. 2605.

Eickenberg, S., Mickholz, E., Jung, E., Nofer, J.-R., Pavenstadt, H.J., and Jacobi, A.M. (2012). Mycophenolic acid counteracts B cell proliferation and plasmablast formation in patients with systemic lupus erythematosus. Arthritis Res. Ther. 14, R110.

Eissens, D.N., Van Der Meer, A., Van Cranenbroek, B., Preijers, F.W.M.B., and Joosten, I. (2010). Rapamycin and MPA, but not CsA, impair human NK cell cytotoxicity due to differential effects on NK cell phenotype. Am. J. Transplant. Off. J. Am. Soc. Transplant. Am. Soc. Transpl. Surg. 10, 1981–1990.

Elsea, C.R., Roberts, D.A., Druker, B.J., and Wood, L.J. (2008). Inhibition of p38 MAPK suppresses inflammatory cytokine induction by etoposide, 5-fluorouracil, and doxorubicin without affecting tumoricidal activity. PloS One 3, e2355.

Emami, J. (2006). In vitro - in vivo correlation: from theory to applications. J. Pharm. Pharm. Sci. Publ. Can. Soc. Pharm. Sci. Société Can. Sci. Pharm. 9, 169–189.

Leatherhead Food RI (2013) Citrinin fact sheet. Retrieved from http://services.leatherheadfood.com/eman/FactSheet.aspx?ID=14

Escoula, L., Bourdiol, D., Linas, M.D., Recco, P., and Seguela, J.P. (1988). Enhancing resistance and modulation of humoral immune response to experimental Candida albicans infection by patulin. Mycopathologia 103, 153–156.

149

Essid, E., Dernawi, Y., and Petzinger, E. (2012). Apoptosis induction by OTA and TNF-α in cultured primary rat hepatocytes and prevention by silibinin. Toxins 4, 1139–1156.

European Commission, and Directorate General for Health & Consumers (2012). Toxicity and assessment of chemical mixtures. (Brussels: European Commission).

Evans, J., and Dawson, K. (2000). The ability of mycosorb to bind toxins present in endophyte- infected tall fescue. T. Lyons, and K. Jacques, eds. (Nottingham: Nottingham University Press), pp. 409–422.

Fanelli, C., Fabbri, A.A., Pieretti, S., Finotti, E., and Passi, S. (1985). Effect of different antioxidants and free radical scavengers on aflatoxin production. Mycotoxin Res. 1, 65–69.

FAO (2003). Worldwide regulations for mycotoxins in food and feed in 2003.

Fatemi, M., Hermann, A., Gowher, H., and Jeltsch, A. (2002). Dnmt3a and Dnmt1 functionally cooperate during de novo methylation of DNA. Eur. J. Biochem. FEBS 269, 4981–4984.

Ferrante, M.C., Bilancione, M., Raso, G.M., Esposito, E., Iacono, A., Zaccaroni, A., and Meli, R. (2006). Expression of COX-2 and hsp72 in peritoneal macrophages after an acute ochratoxin A treatment in mice. Life Sci. 79, 1242–1247.

Ferrante, M.C., Raso, G.M., Bilancione, M., Esposito, E., Iacono, A., and Meli, R. (2008). Differential modification of inflammatory enzymes in J774A.1 macrophages by ochratoxin A alone or in combination with lipopolysaccharide. Toxicol. Lett. 181, 40–46.

Finoli, C., Vecchio, A., Galli, A., and Dragoni, I. (2001). Roquefortine C occurrence in blue cheese. J. Food Prot. 64, 246–251.

Fischer, A., Sananbenesi, F., Mungenast, A., and Tsai, L.-H. (2010). Targeting the correct HDAC(s) to treat cognitive disorders. Trends Pharmacol. Sci. 31, 605–617.

Flajs, D., and Peraica, M. (2009). Toxicological properties of citrinin. Arh. Hig. Rada Toksikol. 60, 457–464.

Flynn, J.L., Goldstein, M.M., Triebold, K.J., Sypek, J., Wolf, S., and Bloom, B.R. (1995). IL-12 increases resistance of BALB/c mice to Mycobacterium tuberculosis infection. J. Immunol. Baltim. Md 1950 155, 2515–2524.

Forgacs, J. (1962). Mycotoxicoses—the neglected diseases. Feedstuffs 34, 124–134.

Forman, H.J., and Torres, M. (2002). Reactive oxygen species and cell signaling: respiratory burst in macrophage signaling. Am. J. Respir. Crit. Care Med. 166, S4–S8.

Frisvad, J.C., Smedsgaard, J., Larsen, T.O., and Samson, R.A. (2004). Mycotoxins, drugs and other extrolites produced by species in Penicillium subgenus Penicillium. Studies in Mycology 49, 201– 241.

150

Frisvad, J.C., Lund, F., and Elmholt, S. (2005). Ochratoxin A producing Penicillium verrucosum isolates from cereals reveal large AFLP fingerprinting variability. J. Appl. Microbiol. 98, 684–692.

Fruhauf, S., Schwartz, H., Ottner, F., Krska, R., and Vekiru, E. (2012). Yeast cell based feed additives: studies on aflatoxin B₁ and zearalenone. Food Addit. Contam. Part Chem. Anal. Control Expo. Risk Assess. 29, 217–231.

Fuchs, S., Sontag, G., Stidl, R., Ehrlich, V., Kundi, M., and Knasmüller, S. (2008). Detoxification of patulin and ochratoxin A, two abundant mycotoxins, by lactic acid bacteria. Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 46, 1398–1407.

Fuks, F., Hurd, P.J., Wolf, D., Nan, X., Bird, A.P., and Kouzarides, T. (2003). The methyl-CpG- binding protein MeCP2 links DNA methylation to histone methylation. J. Biol. Chem. 278, 4035– 4040.

Galtier, P., Alvinerie, M., and Charpenteau, J.L. (1981). The pharmacokinetic profiles of ochratoxin A in pigs, rabbits and chickens. Food Cosmet. Toxicol. 19, 735–738.

García-Lafuente, A., Antolín, M., Guarner, F., Crespo, E., and Malagelada, J.R. (2001). Modulation of colonic barrier function by the composition of the commensal flora in the rat. Gut 48, 503–507.

Garon, D., Richard, E., Sage, L., Bouchart, V., Pottier, D., and Lebailly, P. (2006). Mycoflora and multimycotoxin detection in corn silage: experimental study. J. Agric. Food Chem. 54, 3479–3484.

Gekle, M., Schwerdt, G., Freudinger, R., Mildenberger, S., Wilflingseder, D., Pollack, V., Dander, M., and Schramek, H. (2000). Ochratoxin A induces JNK activation and apoptosis in MDCK-C7 cells at nanomolar concentrations. J. Pharmacol. Exp. Ther. 293, 837–844.

George, J., Banik, N.L., and Ray, S.K. (2010). Genistein induces receptor and mitochondrial pathways and increases apoptosis during BCL-2 knockdown in human malignant neuroblastoma SK-N-DZ cells. J. Neurosci. Res. 88, 877–886.

Godlewski, J., Nowicki, M.O., Bronisz, A., Williams, S., Otsuki, A., Nuovo, G., Raychaudhury, A., Newton, H.B., Chiocca, E.A., and Lawler, S. (2008). Targeting of the Bmi-1 oncogene/stem cell renewal factor by microRNA-128 inhibits glioma proliferation and self-renewal. Cancer Res. 68, 9125–9130.

Gu, J.J., Stegmann, S., Gathy, K., Murray, R., Laliberte, J., Ayscue, L., and Mitchell, B.S. (2000). Inhibition of T lymphocyte activation in mice heterozygous for loss of the IMPDH II gene. J. Clin. Invest. 106, 599–606.

Guidicelli, G., Chaigne-Delalande, B., Dilhuydy, M.-S., Pinson, B., Mahfouf, W., Pasquet, J.-M., Mahon, F.-X., Pourquier, P., Moreau, J.-F., and Legembre, P. (2009). The necrotic signal induced by mycophenolic acid overcomes apoptosis-resistance in tumor cells. PloS One 4, e5493.

Gupta, M., Bandopadhyay, S., Paul, B., and Majumder, S.K. (1979). Hematological changes produced in mice by ochratoxin A. Toxicology 14, 95–98.

151

Ha, H., Kim, M.S., Park, J., Huh, J.Y., Huh, K.H., Ahn, H.J., and Kim, Y.S. (2006). Mycophenolic acid inhibits mesangial cell activation through p38 MAPK inhibition. Life Sci. 79, 1561–1567.

Häggblom, P. (1990). Isolation of roquefortine C from feed grain. Appl. Environ. Microbiol. 56, 2924–2926.

Häggblom, P., Persson, I., and Holmberg, T. (1990). A mycotoxin with neurological effects isolated from grain used for feed. Sven. Veterinärtidning 42, 619–620.

Halstensen, A.S., Nordby, K.-C., Elen, O., and Eduard, W. (2004). Ochratoxin A in grain dust-- estimated exposure and relations to agricultural practices in grain production. Ann. Agric. Environ. Med. AAEM 11, 245–254.

Harris, N.B., and Barletta, R.G. (2001). Mycobacterium avium subsp. paratuberculosis in Veterinary Medicine. Clin. Microbiol. Rev. 14, 489–512.

Harvey, R.B., Elissalde, M.H., Kubena, L.F., Weaver, E.A., Corrier, D.E., and Clement, B.A. (1992). Immunotoxicity of ochratoxin A to growing gilts. Am. J. Vet. Res. 53, 1966–1970.

Hassan, Z.U., Khan, M.Z., Saleemi, M.K., Khan, A., Javed, I., and Noreen, M. (2012). Immunological responses of male White Leghorn chicks kept on ochratoxin A (OTA)- contaminated feed. J. Immunotoxicol. 9, 56–63.

Hauser, I.A., Johnson, D.R., Thévenod, F., and Goppelt-Strübe, M. (1997). Effect of mycophenolic acid on TNF alpha-induced expression of cell adhesion molecules in human venous endothelial cells in vitro. Br. J. Pharmacol. 122, 1315–1322.

Hayes, A.W., Under, P.D., and Williams, W.L. (1977). Acute toxicity of penicillic acid and rubratoxin B in dogs. Ann. Nutr. Aliment. 31, 711–721.

Hayes, A.W., Phillips, T.D., Williams, W.L., and Ciegler, A. (1979). Acute toxicity of patulin in mice and rats. Toxicology 13, 91–100.

He, X., Smeets, R.L., Koenen, H.J.P.M., Vink, P.M., Wagenaars, J., Boots, A.M.H., and Joosten, I. (2011). Mycophenolic acid-mediated suppression of human CD4+ T cells: more than mere guanine nucleotide deprivation. Am. J. Transplant. Off. J. Am. Soc. Transplant. Am. Soc. Transpl. Surg. 11, 439–449.

Herzog-Soares, J.D.A., and Freire, R.B. (2004). Effect of citrinin and in association with aflatoxin B(1) on the infectivity and proliferation of Toxoplasma gondii in vitro. Braz. J. Infect. Dis. Off. Publ. Braz. Soc. Infect. Dis. 8, 101–108.

Hetherington, A., and Raistrick, H. (1931). Studies in the biochemistry of micro-organisms. XI. On the production and chemical constitution of a new yellow colouring matter, citrinin, produced from glucose by Penicillium citrinum Thom. Philos Trans R Soc Lond. Ser B220, 269–297.

Hibi, D., Kijima, A., Kuroda, K., Suzuki, Y., Ishii, Y., Jin, M., Nakajima, M., Sugita-Konishi, Y., Yanai, T., Nohmi, T., et al. (2013). Molecular mechanisms underlying ochratoxin A-induced

152

genotoxicity: global gene expression analysis suggests induction of DNA double-strand breaks and cell cycle progression. J. Toxicol. Sci. 38, 57–69.

Höhler, D., Südekum, K.H., Wolffram, S., Frohlich, A.A., and Marquardt, R.R. (1999). Metabolism and excretion of ochratoxin A fed to sheep. J. Anim. Sci. 77, 1217–1223.

Hsieh, C.L. (1999). In vivo activity of murine de novo methyltransferases, Dnmt3a and Dnmt3b. Mol. Cell. Biol. 19, 8211–8218.

Hunter, C.A., Subauste, C.S., Van Cleave, V.H., and Remington, J.S. (1994). Production of gamma interferon by natural killer cells from Toxoplasma gondii-infected SCID mice: regulation by interleukin-10, interleukin-12, and tumor necrosis factor alpha. Infect. Immun. 62, 2818–2824.

Indramohan, M., Sieve, A.N., Break, T.J., and Berg, R.E. (2012). Inflammatory Monocyte Recruitment Is Regulated by Interleukin-23 during Systemic Bacterial Infection. Infect. Immun. 80, 4099–4105.

Indramohan, M., Break, T., Witter, A., and Berg, R. (2014). The role of IL-23 in regulating the function of phagocytic cells during Listeria monocytogenes infection (MPF3P.822). J Immunol 192, (1 supplement) 132.22.

Ishii, M., Wen, H., Corsa, C.A.S., Liu, T., Coelho, A.L., Allen, R.M., Carson, W.F., Cavassani, K.A., Li, X., Lukacs, N.W., et al. (2009). Epigenetic regulation of the alternatively activated macrophage phenotype. Blood 114, 3244–3254.

Ishikawa, H., Claudio, E., Dambach, D., Raventós-Suárez, C., Ryan, C., and Bravo, R. (1998). Chronic inflammation and susceptibility to bacterial infections in mice lacking the polypeptide (p)105 precursor (NF-kappaB1) but expressing p50. J. Exp. Med. 187, 985–996.

Islam, M.R., Roh, Y.-S., Cho, A., Kim, J., Kim, J.-H., Eo, S., Lim, C.-W., and Kim, B. (2012). Immune modulatory effects of the foodborne contaminant citrinin in mice. Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 50, 3537–3547.

Isomaki, P., Luukkainen, R., Saario, R., Toivanen, P., and Punnonen, J. (1996). Interleukin-10 functions as an antiinflammatory cytokine in rheumatoid synovium. Arthritis Rheum. 39, 386–395.

Jain, J., Almquist, S.J., Ford, P.J., Shlyakhter, D., Wang, Y., Nimmesgern, E., and Germann, U.A. (2004). Regulation of inosine monophosphate dehydrogenase type I and type II isoforms in human lymphocytes. Biochem. Pharmacol. 67, 767–776.

Jarry, A., Bossard, C., Sarrabayrouse, G., Mosnier, J.-F., and Laboisse, C.L. (2011). Loss of interleukin-10 or transforming growth factor β signaling in the human colon initiates a T-helper 1 response via distinct pathways. Gastroenterology 141, 1887–1896.e1–e2.

Johannessen, L.N., Nilsen, A.M., and Løvik, M. (2005). The mycotoxins citrinin and gliotoxin differentially affect production of the pro-inflammatory cytokines tumour necrosis factor-alpha and interleukin-6, and the anti-inflammatory cytokine interleukin-10. Clin. Exp. Allergy J. Br. Soc. Allergy Clin. Immunol. 35, 782–789.

153

Jørgensen, S.E. (2013). Encyclopedia of environmental management.

Kabak, B., Dobson, A.D.W., and Var, I. (2006). Strategies to prevent mycotoxin contamination of food and animal feed: a review. Crit. Rev. Food Sci. Nutr. 46, 593–619.

Kang, S.W., Park, C.H., Hong, S.I., and Kim, S.W. (2007). Production of penicillic acid by Aspergillus sclerotiorum CGF. Bioresour. Technol. 98, 191–197.

Karaman, M., Basmacioglu, H., Ortatatli, M., and Oguz, H. (2005). Evaluation of the detoxifying effect of yeast glucomannan on aflatoxicosis in broilers as assessed by gross examination and histopathology. Br. Poult. Sci. 46, 394–400.

Karnell, J.L., Karnell, F.G., Stephens, G.L., Rajan, B., Morehouse, C., Li, Y., Swerdlow, B., Wilson, M., Goldbach-Mansky, R., Groves, C., et al. (2011). Mycophenolic acid differentially impacts B cell function depending on the stage of differentiation. J. Immunol. Baltim. Md 1950 187, 3603–3612.

Kataoka, H., Itano, M., Ishizaki, A., and Saito, K. (2009). Determination of patulin in fruit juice and dried fruit samples by in-tube solid-phase microextraction coupled with liquid chromatography-mass spectrometry. J. Chromatogr. A 1216, 3746–3750.

Kavanagh, F. (1947). Activities of Twenty-two Antibacterial Substances against Nine Species of Bacteria. J. Bacteriol. 54, 761–766.

Keblys, M., Bernhoft, A., Höfer, C.C., Morrison, E., Larsen, H.J.S., and Flåøyen, A. (2004). The effects of the Penicillium mycotoxins citrinin, cyclopiazonic acid, ochratoxin A, patulin, penicillic acid, and roquefortine C on in vitro proliferation of porcine lymphocytes. Mycopathologia 158, 317–324.

Khera, M., Crawford, D., Morales, A., Salonia, A., and Morgentaler, A. (2014). A new era of testosterone and prostate cancer: from physiology to clinical implications. Eur. Urol. 65, 115–123.

El Khoury, A., and Atoui, A. (2010). Ochratoxin A: General Overview and Actual Molecular Status. Toxins 2, 461–493.

Kiessling, K.H., Pettersson, H., Sandholm, K., and Olsen, M. (1984). Metabolism of aflatoxin, ochratoxin, zearalenone, and three trichothecenes by intact rumen fluid, rumen protozoa, and rumen bacteria. Appl. Environ. Microbiol. 47, 1070–1073.

Kjaerstad, M.B., Taxvig, C., Andersen, H.R., and Nellemann, C. (2010). Mixture effects of endocrine disrupting compounds in vitro. Int. J. Androl. 33, 425–433.

Kojima, H., Kunimoto, H., Inoue, T., and Nakajima, K. (2012). The STAT3-IGFBP5 axis is critical for IL-6/gp130-induced premature senescence in human fibroblasts. Cell Cycle Georget. Tex 11, 730–739.

Komohara, Y., Ohnishi, K., Kuratsu, J., and Takeya, M. (2008). Possible involvement of the M2 anti-inflammatory macrophage phenotype in growth of human gliomas. J. Pathol. 216, 15–24.

154

Korosteleva, S.N., Smith, T.K., and Boermans, H.J. (2007). Effects of feedborne Fusarium mycotoxins on the performance, metabolism, and immunity of dairy cows. J. Dairy Sci. 90, 3867– 3873.

Kortenkamp, A., and Altenburger, R. (1998). Synergisms with mixtures of xenoestrogens: a reevaluation using the method of isoboles. Sci. Total Environ. 221, 59–73.

Kroemer, A., Xiao, X., Degauque, N., Edtinger, K., Wei, H., Demirci, G., and Li, X.C. (2008). The Innate NK Cells, Allograft Rejection, and a Key Role for IL-15. J. Immunol. 180, 7818–7826.

Kuehl, R.O. (2000). Design of experiments: statistical principles of research design and analysis (Pacific Grove, CA: Duxbury/Thomson Learning).

Kulik, M.M., and Holaday, C.E. (1966). Aflatoxin: a metabolic product of several fungi. Mycopathol. Mycol. Appl. 30, 137–140.

Kumar, R., Dwivedi, P.D., Dhawan, A., Das, M., and Ansari, K.M. (2011). Citrinin-generated reactive oxygen species cause cell cycle arrest leading to apoptosis via the intrinsic mitochondrial pathway in mouse skin. Toxicol. Sci. Off. J. Soc. Toxicol. 122, 557–566.

Kuo, M.H., and Allis, C.D. (1998). Roles of histone acetyltransferases and deacetylases in gene regulation. BioEssays News Rev. Mol. Cell. Dev. Biol. 20, 615–626.

Kurita, S., Higuchi, H., Saito, Y., Nakamoto, N., Takaishi, H., Tada, S., Saito, H., Gores, G.J., and Hibi, T. (2010). DNMT1 and DNMT3b silencing sensitizes human hepatoma cells to TRAIL- mediated apoptosis via up-regulation of TRAIL-R2/DR5 and caspase-8. Cancer Sci. 101, 1431– 1439.

Lafont, P., Debeaupuis, J., Gaillardin, M., and Payen, J. (1979). Production of mycophenolic acid by Penicillium roqueforti strains. Appl Envrion Microbiol 37, 365–368.

Lagaraine, C., Lemoine, R., Baron, C., Nivet, H., Velge-Roussel, F., and Lebranchu, Y. (2008). Induction of human CD4+ regulatory T cells by mycophenolic acid-treated dendritic cells. J. Leukoc. Biol. 84, 1057–1064.

Lawrence, T. (2009). The Nuclear Factor NF- B Pathway in Inflammation. Cold Spring Harb. Perspect. Biol. 1, a001651–a001651.

Lea, T., Steien, K., and Størmer, F.C. (1989). Mechanism of ochratoxin A-induced immunosuppression. Mycopathologia 107, 153–159.

Leemans, J.C., Florquin, S., Heikens, M., Pals, S.T., van der Neut, R., and Van Der Poll, T. (2003). CD44 is a macrophage binding site for Mycobacterium tuberculosis that mediates macrophage recruitment and protective immunity against tuberculosis. J. Clin. Invest. 111, 681–689.

Lennox, C.L., Spotts, R.A., and Cervantes, L.A. (2003). Populations of Botrytis cinerea and Penicillium spp. on Pear Fruit, and in Orchards and Packinghouses, and Their Relationship to Postharvest Decay. Plant Dis. 87, 639–644.

155

Lessard, J., and Sauvageau, G. (2003). Bmi-1 determines the proliferative capacity of normal and leukaemic stem cells. Nature 423, 255–260.

Liu, B.-H., Yu, F.-Y., Wu, T.-S., Li, S.-Y., Su, M.-C., Wang, M.-C., and Shih, S.-M. (2003). Evaluation of genotoxic risk and oxidative DNA damage in mammalian cells exposed to mycotoxins, patulin and citrinin. Toxicol. Appl. Pharmacol. 191, 255–263.

Liu, B.-H., Wu, T.-S., Yu, F.-Y., and Su, C.-C. (2007). Induction of oxidative stress response by the mycotoxin patulin in mammalian cells. Toxicol. Sci. Off. J. Soc. Toxicol. 95, 340–347.

Liu, B.-H., Chi, J.-Y., Hsiao, Y.-W., Tsai, K.-D., Lee, Y.-J., Lin, C.-C., Hsu, S.-C., Yang, S.-M., and Lin, T.-H. (2010). The fungal metabolite, citrinin, inhibits lipopolysaccharide/interferon-γ- induced nitric oxide production in glomerular mesangial cells. Int. Immunopharmacol. 10, 1608– 1615.

Liu, L., McBride, K.M., and Reich, N.C. (2005). STAT3 nuclear import is independent of tyrosine phosphorylation and mediated by importin- 3. Proc. Natl. Acad. Sci. 102, 8150–8155.

Llewellyn, G.C., McCay, J.A., Brown, R.D., Musgrove, D.L., Butterworth, L.F., Munson, A.E., and White, K.L. (1998). Immunological evaluation of the mycotoxin patulin in female B6C3F1 mice. Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 36, 1107–1115.

Luft, P., Oostingh, G.J., Gruijthuijsen, Y., Horejs-Hoeck, J., Lehmann, I., and Duschl, A. (2008). Patulin influences the expression of Th1/Th2 cytokines by activated peripheral blood mononuclear cells and T cells through depletion of intracellular glutathione. Environ. Toxicol. 23, 84–95.

Luster, M.I., Germolec, D.R., Burleson, G.R., Jameson, C.W., Ackermann, M.F., Lamm, K.R., and Hayes, H.T. (1987). Selective immunosuppression in mice of natural killer cell activity by ochratoxin A. Cancer Res. 47, 2259–2263.

Ma, Y., and Pope, R.M. (2005). The role of macrophages in rheumatoid arthritis. Curr. Pharm. Des. 11, 569–580.

Maelfait, J., Vercammen, E., Janssens, S., Schotte, P., Haegman, M., Magez, S., and Beyaert, R. (2008). Stimulation of Toll-like receptor 3 and 4 induces interleukin-1beta maturation by caspase- 8. J. Exp. Med. 205, 1967–1973.

Magalhaes, J.G., Tattoli, I., and Girardin, S.E. (2007). The intestinal epithelial barrier: how to distinguish between the microbial flora and pathogens. Semin. Immunol. 19, 106–115.

Mahfoud, R., Maresca, M., Garmy, N., and Fantini, J. (2002). The mycotoxin patulin alters the barrier function of the intestinal epithelium: mechanism of action of the toxin and protective effects of glutathione. Toxicol. Appl. Pharmacol. 181, 209–218.

Manafi, M., Swamy, H.N., and Pirany, N. (2009). In vitro binidng avility of mycotoxin binder in commercial boriler feed. Afr. J. Agric. Res. 4, 141–143.

156

Manafi, M., Murthy, H., Pirany, N., and Swamy, H.N. (2012). Comparative Study of Several Mycotoxin Binders During Aflatoxicosis in Body Weight, Feed Consumption, Feed Efficiency and Egg Production Parameters of Broiler Breeders. Global Veterinaria 8, 484–490.

Mansfield, M.A., Jones, A.D., and Kuldau, G.A. (2008). Contamination of fresh and ensiled maize by multiple penicillium mycotoxins. Phytopathology 98, 330–336.

Marasas, W.F., Yagen, B., Sydenham, E., Combrinck, S., and Thiel, P.G. (1987). Comparative yields of T-2 toxin and related trichothecenes from five toxicologically important strains of Fusarium sporotrichioides. Appl. Environ. Microbiol. 53, 693–696.

Maresca, M., Mahfoud, R., Pfohl-Leszkowicz, A., and Fantini, J. (2001). The mycotoxin ochratoxin A alters intestinal barrier and absorption functions but has no effect on chloride secretion. Toxicol. Appl. Pharmacol. 176, 54–63.

Maresca, M., Yahi, N., Younès-Sakr, L., Boyron, M., Caporiccio, B., and Fantini, J. (2008). Both direct and indirect effects account for the pro-inflammatory activity of enteropathogenic mycotoxins on the human intestinal epithelium: stimulation of interleukin-8 secretion, potentiation of interleukin-1beta effect and increase in the transepithelial passage of commensal bacteria. Toxicol. Appl. Pharmacol. 228, 84–92.

Marin-Kuan, M., Nestler, S., Verguet, C., Bezençon, C., Piguet, D., Delatour, T., Mantle, P., Cavin, C., and Schilter, B. (2007). MAPK-ERK activation in kidney of male rats chronically fed ochratoxin A at a dose causing a significant incidence of renal carcinoma. Toxicol. Appl. Pharmacol. 224, 174–181.

Marin-Kuan, M., Cavin, C., Delatour, T., and Schilter, B. (2008). Ochratoxin A carcinogenicity involves a complex network of epigenetic mechanisms. Toxicon Off. J. Int. Soc. Toxinology 52, 195–202.

Martinet, Y., Yamauchi, K., and Crystal, R.G. (1988). Differential expression of the tumor necrosis factor/cachectin gene by blood and lung mononuclear phagocytes. Am. Rev. Respir. Dis. 138, 659–665.

Martins, H.M., Almeida, I., Camacho, C., Costa, J.M., and Bernardo, F. (2012). A survey on the occurrence of ochratoxin A in feeds for swine and laying hens. Mycotoxin Res. 28, 107–110.

Martins, M.L., Gimeno, A., Martins, H.M., and Bernardo, F. (2002). Co-occurrence of patulin and citrinin in Portuguese apples with rotten spots. Food Addit. Contam. 19, 568–574.

Mateo, R., Medina, A., Mateo, E.M., Mateo, F., and Jiménez, M. (2007). An overview of ochratoxin A in beer and wine. Int. J. Food Microbiol. 119, 79–83.

Meadows, S.L., Gennings, C., Carter, W.H., and Bae, D.-S. (2002). Experimental designs for mixtures of chemicals along fixed ratio rays. Environ. Health Perspect. 110 Suppl 6, 979–983.

157

Mehling, A., Grabbe, S., Voskort, M., Schwarz, T., Luger, T.A., and Beissert, S. (2000). Mycophenolate mofetil impairs the maturation and function of murine dendritic cells. J. Immunol. Baltim. Md 1950 165, 2374–2381.

Menon, S.K., Modi, N.R., Mistry, B., and Joshi, K. (2011). Improvement of some pharmaceutical properties of mycophenolate mofetil (MMF) by para sulphonatocalix[4]resorcinarene inclusion complex. J. Incl. Phenom. Macrocycl. Chem. 70, 121–128.

Mezger, M., Wozniok, I., Blockhaus, C., Kurzai, O., Hebart, H., Einsele, H., and Loeffler, J. (2008). Impact of mycophenolic acid on the functionality of human polymorphonuclear neutrophils and dendritic cells during interaction with Aspergillus fumigatus. Antimicrob. Agents Chemother. 52, 2644–2646.

Mikawa, K., Akamatsu, H., Maekawa, N., Nishina, K., Obara, H., and Niwa, Y. (1994). Inhibitory effect of gabexate mesilate on human neutrophil function. J. Int. Med. Res. 22, 245–254.

Milićević, D., Jurić, V., Stefanović, S., Jovanović, M., and Janković, S. (2008). Survey of slaughtered pigs for occurrence of ochratoxin A and porcine nephropathy in Serbia. Int. J. Mol. Sci. 9, 2169–2183.

Miraglia, M., Marvin, H.J.P., Kleter, G.A., Battilani, P., Brera, C., Coni, E., Cubadda, F., Croci, L., De Santis, B., Dekkers, S., et al. (2009). Climate change and food safety: an emerging issue with special focus on Europe. Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 47, 1009–1021.

Miranda, T.B., and Jones, P.A. (2007). DNA methylation: the nuts and bolts of repression. J. Cell. Physiol. 213, 384–390.

Mohacsi, P.J., Tüller, D., Hulliger, B., and Wijngaard, P.L. (1997). Different inhibitory effects of immunosuppressive drugs on human and rat aortic smooth muscle and endothelial cell proliferation stimulated by platelet-derived growth factor or endothelial cell growth factor. J. Heart Lung Transplant. Off. Publ. Int. Soc. Heart Transplant. 16, 484–492.

Momotani, E. (2012). Epidemiological situation and control strategies for paratuberculosis in Japan. Jpn. J. Vet. Res. 60 Suppl, S19–S29.

Moroi, K., Suzuki, S., Kuga, T., Yamazaki, M., and Kanisawa, M. (1985). Reduction of ochratoxin A toxicity in mice treated with phenylalanine and phenobarbital. Toxicol. Lett. 25, 1–5.

Mosser, D.M. (2003). The many faces of macrophage activation. J. Leukoc. Biol. 73, 209–212.

Müller, H.M., and Amend, R. (1997). Formation and disappearance of mycophenolic acid, patulin, penicillic acid and PR toxin in maize silage inoculated with Penicillium roqueforti. Arch. Für Tierernähr. 50, 213–225.

Mullican, S.E., Gaddis, C.A., Alenghat, T., Nair, M.G., Giacomin, P.R., Everett, L.J., Feng, D., Steger, D.J., Schug, J., Artis, D., et al. (2011). Histone deacetylase 3 is an epigenomic brake in macrophage alternative activation. Genes Dev. 25, 2480–2488.

158

Murnaghan, M.F. (1946). The pharmacology of penicillic acid. J. Pharmacol. Exp. Ther. 88, 119– 132.

Murphy, K., Paul, T., and Walport, M. (2008). Janeway’s Immunobiology (LLC, UK: Garland Science, Taylor & Francis Group).

Mus, A.M.C., Cornelissen, F., Asmawidjaja, P.S., van Hamburg, J.P., Boon, L., Hendriks, R.W., and Lubberts, E. (2010). Interleukin-23 promotes Th17 differentiation by inhibiting T-bet and FoxP3 and is required for elevation of interleukin-22, but not interleukin-21, in autoimmune experimental arthritis. Arthritis Rheum. 62, 1043–1050.

Nagai, M., Natsumeda, Y., and Weber, G. (1992). Proliferation-linked regulation of type II IMP dehydrogenase gene in human normal lymphocytes and HL-60 leukemic cells. Cancer Res. 52, 258–261.

Niemiec, J., and Borzemska, W. (1994). The effect of Ochratoxin A on egg quality development of embryos and the level of toxin in egg and tissue of hens and chicks. J Anim. Feed Sci 3, 309– 316.

Nomura, F., Akashi, S., Sakao, Y., Sato, S., Kawai, T., Matsumoto, M., Nakanishi, K., Kimoto, M., Miyake, K., Takeda, K., et al. (2000). Cutting edge: endotoxin tolerance in mouse peritoneal macrophages correlates with down-regulation of surface toll-like receptor 4 expression. J. Immunol. Baltim. Md 1950 164, 3476–3479.

Novak, N., Koch, S., Allam, J.-P., and Bieber, T. (2010). Dendritic cells: bridging innate and adaptive immunity in atopic dermatitis. J. Allergy Clin. Immunol. 125, 50–59.

O’Brien, E., Prietz, A., and Dietrich, D.R. (2005). Investigation of the teratogenic potential of ochratoxin A and B using the FETAX system. Birth Defects Res. B. Dev. Reprod. Toxicol. 74, 417–423.

Oh, S.-Y., Boermans, H.J., Swamy, H.V.L.N., Sharma, B.S., and Karrow, N.A. (2012). Immunotoxicity of Penicillium Mycotoxins on Viability and Proliferation of Bovine Macrophage Cell Line (BOMACs). Open Mycol. J. 6, 11–16.

Oh, S.-Y., Balch, C.G., Cliff, R.L., Sharma, B.S., Boermans, H.J., Swamy, H.V.L.N., Quinton, V.M., and Karrow, N.A. (2013). Exposure to Penicillium mycotoxins alters gene expression of enzymes involved in the epigenetic regulation of bovine macrophages (BoMacs). Mycotoxin Res. 29, 235–243.

Ohata, K., Espinoza, J.L., Lu, X., Kondo, Y., and Nakao, S. (2011). Mycophenolic acid inhibits natural killer cell proliferation and cytotoxic function: a possible disadvantage of including mycophenolate mofetil in the graft-versus-host disease prophylaxis regimen. Biol. Blood Marrow Transplant. J. Am. Soc. Blood Marrow Transplant. 17, 205–213.

Ohlson, C.-G., Berg, P., Bryngelsson, I.-L., Elihn, K., Ngo, Y., Westberg, H., and Sjögren, B. (2010). Inflammatory markers and exposure to occupational air pollutants. Inhal. Toxicol. 22, 1083–1090.

159

Ohmomo, S., Sato, T., Utagawa, T., and Abe, M. (1975). Isolation of Festuclavine and Three New Indole Alkaloids, Roquefortine A, B and C from the Cultures of Penicillium roqueforti (Production of Alkaloids and Related Substances by Fungi Part XII). J Antibact Antifung Agents 49, 615–623.

Ohmomo, S., Kitamoto, H.K., and Nakajima, T. (1994). Detection of roquefortines in penicillium roqueforti isolated from moulded maize silage. J. Sci. Food Agric. 64, 211–215.

Okano, M., Bell, D.W., Haber, D.A., and Li, E. (1999). DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99, 247–257.

O’Leary, S., O’Sullivan, M.P., and Keane, J. (2011). IL-10 blocks phagosome maturation in mycobacterium tuberculosis-infected human macrophages. Am. J. Respir. Cell Mol. Biol. 45, 172– 180.

Ozsoy, N., Selmanoğlu, G., Koçkaya, E.A., Gül, N., and Cebesoy, S. (2008). Effect of patulin on the interdigitating dendritic cells (IDCs) of rat thymus. Cell Biochem. Funct. 26, 192–196.

Paglinawan, R., Malipiero, U., Schlapbach, R., Frei, K., Reith, W., and Fontana, A. (2003). TGFbeta directs gene expression of activated microglia to an anti-inflammatory phenotype strongly focusing on chemokine genes and cell migratory genes. Glia 44, 219–231.

Palaniappan, N., Anbalagan, S., and Narayanan, S. (2012). Mitogen-activated protein kinases mediate Mycobacterium tuberculosis-induced CD44 surface expression in monocytes. J. Biosci. 37, 41–54.

Pamer, E.G. (2007). Immune responses to commensal and environmental microbes. Nat. Immunol. 8, 1173–1178.

Pang, G., Couch, L., Batey, R., Clancy, R., and Cripps, A. (1994). GM-CSF, IL-1 alpha, IL-1 beta, IL-6, IL-8, IL-10, ICAM-1 and VCAM-1 gene expression and cytokine production in human duodenal fibroblasts stimulated with lipopolysaccharide, IL-1 alpha and TNF-alpha. Clin. Exp. Immunol. 96, 437–443.

Papp, G., Horváth, E., Mike, N., Gazdag, Z., Belágyi, J., Gyöngyi, Z., Bánfalvi, G., Hornok, L., and Pesti, M. (2012). Regulation of patulin-induced oxidative stress processes in the fission yeast Schizosaccharomyces pombe. Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 50, 3792–3798.

Park, A.Y., and Scott, P. (2001). Il-12: keeping cell-mediated immunity alive. Scand. J. Immunol. 53, 529–532.

Park, D., Njapau, H., and Boutrif, E. (2009). Food, nutrition and agriculture 23 Minimizing risks posed by mycotoxins utilizing the HACCP concept.

Park, J., Ha, H., Seo, J., Kim, M.S., Kim, H.J., Huh, K.H., Park, K., and Kim, Y.S. (2004). Mycophenolic acid inhibits platelet-derived growth factor-induced reactive oxygen species and mitogen-activated protein kinase activation in rat vascular smooth muscle cells. Am. J. Transplant. Off. J. Am. Soc. Transplant. Am. Soc. Transpl. Surg. 4, 1982–1990.

160

Park, J., Chang, H.K., Ha, H., Kim, M.S., Ahn, H.J., and Kim, Y.S. (2008). Mycophenolic Acid Inhibits Cell Proliferation and Extracellular Matrix Synthesis in Rat Vascular Smooth Muscle Cells Through Direct and Indirect Inhibition of Cellular Reactive Oxygen Species. J. Surg. Res. 150, 17–23.

Pelletier, M., Maggi, L., Micheletti, A., Lazzeri, E., Tamassia, N., Costantini, C., Cosmi, L., Lunardi, C., Annunziato, F., Romagnani, S., et al. (2010). Evidence for a cross-talk between human neutrophils and Th17 cells. Blood 115, 335–343.

Pereyra, C.M., Alonso, V.A., Rosa, C.A.R., Chiacchiera, S.M., Dalcero, A.M., and Cavaglieri, L.R. (2008). Gliotoxin natural incidence and toxigenicity of <i>Aspergillus fumigatus</i> isolated from corn silage and ready dairy cattle feed. World Mycotoxin J. 1, 457–462.

Pérez-Simón, J.A., Martino, R., Caballero, D., Valcarcel, D., Rebollo, N., de la Cámara, R., de Oteiza, J.P., Heras, I., Calvo, M.V., Sierra, J., et al. (2008). Reduced-intensity conditioning allogeneic transplantation from unrelated donors: evaluation of mycophenolate mofetil plus cyclosporin A as graft-versus-host disease prophylaxis. Biol. Blood Marrow Transplant. J. Am. Soc. Blood Marrow Transplant. 14, 664–671.

Peterson, J.D., Herzenberg, L.A., Vasquez, K., and Waltenbaugh, C. (1998). Glutathione levels in antigen-presenting cells modulate Th1 versus Th2 response patterns. Proc. Natl. Acad. Sci. U. S. A. 95, 3071–3076.

Pietersen, A.M., Evers, B., Prasad, A.A., Tanger, E., Cornelissen-Steijger, P., Jonkers, J., and van Lohuizen, M. (2008). Bmi1 regulates stem cells and proliferation and differentiation of committed cells in mammary epithelium. Curr. Biol. CB 18, 1094–1099.

Pitt, J.I. (1987). Penicillium viridicatum, Penicillium verrucosum, and production of ochratoxin A. Appl. Environ. Microbiol. 53, 266–269.

Place, R.F., Noonan, E.J., and Giardina, C. (2005). HDAC inhibition prevents NF-kappa B activation by suppressing proteasome activity: down-regulation of proteasome subunit expression stabilizes I kappa B alpha. Biochem. Pharmacol. 70, 394–406.

Placinta, C.., D’Mello, J.P., and Macdonald, A.M. (1999). A review of worldwide contamination of cereal grains and animal feed with Fusarium mycotoxins. Anim. Feed Sci. Technol. 78, 21–37.

Plackett, R.., and Hewlett, P. (1963). A Unified Theory for Quantal Responses to Mixtures of Drugs: The Fitting to Data of Certain Models for Two Non-Interactive Drugs with Complete Positive Correlation of Tolerances. Biometrics 19, 517–531.

Pollard, J.W. (2004). Opinion: Tumour-educated macrophages promote tumour progression and metastasis. Nat. Rev. Cancer 4, 71–78.

Pozzi, C.R., Corrêa, B., Gambale, W., Paula, C.R., Chacon-Reche, N.O., and Meirelles, M.C. (1995). Postharvest and stored corn in Brazil: mycoflora interaction, abiotic factors and mycotoxin occurrence. Food Addit. Contam. 12, 313–319.

161

Prelusky, D.B., Hamilton, R.M., Trenholm, H.L., and Miller, J.D. (1986). Tissue distribution and excretion of radioactivity following administration of 14C-labeled deoxynivalenol to White Leghorn hens. Fundam. Appl. Toxicol. Off. J. Soc. Toxicol. 7, 635–645.

Prior, M.G., O’Neil, J.B., and Sisodia, C.S. (1980). Effects of ochratoxin A on growth response and residues in broilers. Poult. Sci. 59, 1254–1257.

Puel, O., Galtier, P., and Oswald, I.P. (2010). Biosynthesis and toxicological effects of patulin. Toxins 2, 613–631.

Puls, R., and Ladyman, E. (1988). Roquefortine Toxicity in a Dog. Can Vet J 29, 569.

Qin, Z.-K., Yang, J.-A., Ye, Y.-L., Zhang, X., Xu, L.-H., Zhou, F.-J., Han, H., Liu, Z.-W., Song, L.-B., and Zeng, M.-S. (2009). Expression of Bmi-1 is a prognostic marker in bladder cancer. BMC Cancer 9, 61.

Qiumei, Y. (2014). Construction of a REporter Mycobacterium avium subsp. paratuberculosis (Map) Strain and Infection of Monocyte-Derived Macrophages from Cows homozyhous for SNP -298 A>G in the Macrophage Migration Inhibitory Factor (MIF) Gene. In 10th World Congress on Genetics Applied to Livestock Production, (Animal Soceity of Animal Science), p. 87.

Quéméneur, L., Flacher, M., Gerland, L.-M., Ffrench, M., Revillard, J.-P., and Bonnefoy-Berard, N. (2002). Mycophenolic acid inhibits IL-2-dependent T cell proliferation, but not IL-2-dependent survival and sensitization to apoptosis. J. Immunol. Baltim. Md 1950 169, 2747–2755.

Radeke, H.H., Meier, B., Topley, N., Flöge, J., Habermehl, G.G., and Resch, K. (1990). Interleukin 1-alpha and tumor necrosis factor-alpha induce oxygen radical production in mesangial cells. Kidney Int. 37, 767–775.

Rahman, I., Marwick, J., and Kirkham, P. (2004). Redox modulation of chromatin remodeling: impact on histone acetylation and deacetylation, NF-kappaB and pro-inflammatory gene expression. Biochem. Pharmacol. 68, 1255–1267.

Raju, M., and Devegowda, G. (2002). Esterified-Glucomannan in Broiler Chicken Diets- Contaminated with Aflatoxin, Ochratoxin and T-2 Toxin: Evaluation of its Binding Ability (in vitro) and Efficacy as Immunomodulator. AJAS 15, 1051–1056.

Raju, M.V., and Devegowda, G. (2000). Influence of esterified-glucomannan on performance and organ morphology, serum biochemistry and haematology in broilers exposed to individual and combined mycotoxicosis (aflatoxin, ochratoxin and T-2 toxin). Br. Poult. Sci. 41, 640–650.

Ramakrishna, Y., Bhat, R.V., and Ravindranath, V. (1989). Production of deoxynivalenol by Fusarium isolates from samples of wheat associated with a human mycotoxicosis outbreak and from sorghum cultivars. Appl. Environ. Microbiol. 55, 2619–2620.

Rand, T.G., Giles, S., Flemming, J., Miller, J.D., and Puniani, E. (2005). Inflammatory and cytotoxic responses in mouse lungs exposed to purified toxins from building isolated Penicillium

162

brevicompactum Dierckx and P. chrysogenum Thom. Toxicol. Sci. Off. J. Soc. Toxicol. 87, 213– 222.

Rand, T.G., Dipenta, J., Robbins, C., and Miller, J.D. (2011). Effects of low molecular weight fungal compounds on inflammatory gene transcription and expression in mouse alveolar macrophages. Chem. Biol. Interact. 190, 139–147.

Reddy, C.S., Chan, P.K., and Hayes, A.W. (1978). Teratogenic and dominant lethal studies of patulin in mice. Toxicology 11, 219–223.

Reddy, R.V., Mayura, K., Hayes, A.W., and Berndt, W.O. (1982). Embryocidal, teratogenic and fetotoxic effects of citrinin in rats. Toxicology 25, 151–160.

Reddy, R.V., Taylor, M.J., and Sharma, R.P. (1988). Evaluation of citrinin toxicity on the immune function of mice. J Food Prot. 51, 32–36.

Reijula, K., and Tuomi, T. (2003). Mycotoxins of aspergilli: exposure and health effects. Front. Biosci. J. Virtual Libr. 8, s232–s235.

Restrepo, B.I., Twahirwa, M., Rahbar, M.H., and Schlesinger, L.S. (2014). Phagocytosis via Complement or Fc-Gamma Receptors Is Compromised in Monocytes from Type 2 Diabetes Patients with Chronic Hyperglycemia. PLoS ONE 9, e92977.

Reyes-Velázquez, W.P., Isaías Espinoza, V.H., Rojo, F., Jiménez-Plasencia, C., de Lucas Palacios, E., Hernández-Góbora, J., and Ramírez-Alvarez, A. (2008). Occurrence of fungi and mycotoxins in corn silage, Jalisco State, Mexico. Rev. Iberoam. Micol. 25, 182–185.

Rhee, K.-J., Sethupathi, P., Driks, A., Lanning, D.K., and Knight, K.L. (2004). Role of commensal bacteria in development of gut-associated lymphoid tissues and preimmune antibody repertoire. J. Immunol. Baltim. Md 1950 172, 1118–1124.

Ribeiro, S.M., Chagas, G.M., Campello, A.P., and Klüppel, M.L. (1997). Mechanism of citrinin- induced dysfunction of mitochondria. V. Effect on the homeostasis of the reactive oxygen species. Cell Biochem. Funct. 15, 203–209.

Ribelin, W.E., Fukushima, K., and Still, P.E. (1978). The toxicity of ochratoxin to ruminants. Can. J. Comp. Med. Rev. Can. Médecine Comparée 42, 172–176.

Rice, J.C., and Allis, C.D. (2001). Histone methylation versus histone acetylation: new insights into epigenetic regulation. Curr. Opin. Cell Biol. 13, 263–273.

Richard, E., Heutte, N., Sage, L., Pottier, D., Bouchart, V., Lebailly, P., and Garon, D. (2007). Toxigenic fungi and mycotoxins in mature corn silage. Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 45, 2420–2425.

Richetti, A., Cavallaro, A., Ainis, T., and Fimiani, V. (2003). Effect of some mycotoxins on superoxide anion production of isolated human neutrophils and in whole blood. Immunopharmacol. Immunotoxicol. 25, 441–449.

163

Royce, S.G., and Karagiannis, T.C. (2012). Histone deacetylases and their role in asthma. J. Asthma Off. J. Assoc. Care Asthma 49, 121–128.

Rutledge, R.G., and Côté, C. (2003). Mathematics of quantitative kinetic PCR and the application of standard curves. Nucleic Acids Res. 31, e93.

Sakaguchi, S. (2004). Naturally arising CD4+ regulatory t cells for immunologic self-tolerance and negative control of immune responses. Annu. Rev. Immunol. 22, 531–562.

Sansing, G.A., Lillehoj, E.B., Detroy, R.W., and Miller, M.A. (1976). Synergistic toxic effects of citrinin, ochratoxin A and penicillic acid in mice. Toxicon Off. J. Int. Soc. Toxinology 14, 213– 220.

Sauvant, C., Holzinger, H., and Gekle, M. (2005). The nephrotoxin ochratoxin A induces key parameters of chronic interstitial nephropathy in renal proximal tubular cells. Cell. Physiol. Biochem. Int. J. Exp. Cell. Physiol. Biochem. Pharmacol. 15, 125–134.

Saxena, N., Ansari, K.M., Kumar, R., Dhawan, A., Dwivedi, P.D., and Das, M. (2009). Patulin causes DNA damage leading to cell cycle arrest and apoptosis through modulation of Bax, p(53) and p(21/WAF1) proteins in skin of mice. Toxicol. Appl. Pharmacol. 234, 192–201.

Schmale III, D., and Munkvold, G. (2009). Mycotoxins in Crops: A Threat to Human and Domestic Animal Health.

Schneweis, I., Meyer, K., Hörmansdorfer, S., and Bauer, J. (2000). Mycophenolic acid in silage. Appl. Environ. Microbiol. 66, 3639–3641.

Schroder, K., Hertzog, P.J., Ravasi, T., and Hume, D.A. (2004). Interferon-gamma: an overview of signals, mechanisms and functions. J. Leukoc. Biol. 75, 163–189.

Schuler, M., Bossy-Wetzel, E., Goldstein, J.C., Fitzgerald, P., and Green, D.R. (2000). p53 induces apoptosis by caspase activation through mitochondrial cytochrome c release. J. Biol. Chem. 275, 7337–7342.

Schütze, N., Lehmann, I., Bönisch, U., Simon, J.C., and Polte, T. (2010). Exposure to mycotoxins increases the allergic immune response in a murine asthma model. Am. J. Respir. Crit. Care Med. 181, 1188–1199.

Scott, P.M., and Kennedy, P.C. (1976). Analysis of blue cheese for roquefortine and other alkaloids from Penicillium roqueforti. J. Agric. Food Chem. 24, 865–868.

Scott, P.M., and Somers, E. (1968). Stability of patulin and penicillic acid in fruit juices and flour. J. Agric. Food Chem. 16, 483–485.

Scott, P.M., Van Walbeek, W., Kennedy, B., and Anyeti, D. (1972). Mycotoxins (ochratoxin A, citrinin, and sterigmatocystin) and toxigenic fungi in grains and other agricultural products. J. Agric. Food Chem. 20, 1103–1109.

164

Segal, B.H. (2007). Role of macrophages in host defense against aspergillosis and strategies for immune augmentation. The Oncologist 12 Suppl 2, 7–13.

Shepherd, E.C., Phillips, T.D., Joiner, G.N., Kubena, L.F., and Heidelbaugh, N.D. (1981). Ochratoxin A and Penicillic acid interaction in mice. J. Environ. Sci. Health Part B 16, 557–573.

Shroff, K.E., Meslin, K., and Cebra, J.J. (1995). Commensal enteric bacteria engender a self- limiting humoral mucosal immune response while permanently colonizing the gut. Infect. Immun. 63, 3904–3913.

Sica, A., and Bronte, V. (2007). Altered macrophage differentiation and immune dysfunction in tumor development. J. Clin. Invest. 117, 1155–1166.

Singh, N.D., Sharma, A.K., Dwivedi, P.D., Kumar, M., and Patil, R. (2011). Immunosuppressive effect of combined citrinin and endosulfan toxicity in pregnant Wistar rats. Vet Arh. 81, 751–763.

Skaug, M.A., Eduard, W., and Størmer, F.C. (2001). Ochratoxin A in airborne dust and fungal conidia. Mycopathologia 151, 93–98.

Smith, J.E., and Harran, G. (1993). Microbial degradation of mycotoxins. Int. Biodeterior. Biodegrad. 32, 205–211.

Smith, J.E., Solomons, G., Lewis, C., and Anderson, J.G. (1995). Role of mycotoxins in human and animal nutrition and health. Nat. Toxins 3, 187–192; discussion 221.

Snow, J.P., Lucas, G.B., Harvan, D., Pero, R.W., and Owens, R.G. (1972). Analysis of tobacco and smoke condensate for penicillic acid. Appl. Microbiol. 24, 34–36.

Sommer, N.F., Buchanan, J.R., and Fortlage, R.J. (1974). Production of patulin by Penicillium expansum. Appl. Microbiol. 28, 589–593.

Song, W., Tai, Y.-T., Tian, Z., Hideshima, T., Chauhan, D., Nanjappa, P., Exley, M.A., Anderson, K.C., and Munshi, N.C. (2011). HDAC inhibition by LBH589 affects the phenotype and function of human myeloid dendritic cells. Leukemia 25, 161–168.

Sorenson, W.G., and Simpson, J. (1986). Toxicity of penicillic acid for rat alveolar macrophages in vitro. Environ. Res. 41, 505–513.

Sorenson, W.G., Simpson, J., and Castranova, V. (1985). Toxicity of the mycotoxin patulin for rat alveolar macrophages. Environ. Res. 38, 407–416.

Sorenson, W.G., Gerberick, G.F., Lewis, D.M., and Castranova, V. (1986). Toxicity of mycotoxins for the rat pulmonary macrophage in vitro. Environ. Health Perspect. 66, 45–53.

Sparmann, A., and van Lohuizen, M. (2006). Polycomb silencers control cell fate, development and cancer. Nat. Rev. Cancer 6, 846–856.

165

Speijers, G.J.A., and Speijers, M.H.M. (2004). Combined toxic effects of mycotoxins. Toxicol. Lett. 153, 91–98.

Spurling, C.C., Godman, C.A., Noonan, E.J., Rasmussen, T.P., Rosenberg, D.W., and Giardina, C. (2008). HDAC3 overexpression and colon cancer cell proliferation and differentiation. Mol. Carcinog. 47, 137–147.

Stabel, J.R., and Stabel, T.J. (1995). Immortalization and characterization of bovine peritoneal macrophages transfected with SV40 plasmid DNA. Vet. Immunol. Immunopathol. 45, 211–220.

Stec, J., Rachubik, J., Szczotka, M., and Kuzmak, J. (2008). Effects of Penicillium Mycotoxins: Ctrinin, Ochratoxin A, and Patulin on In Vitro Proliferation of Bovine Lymphocytes. Bulletin of the Vet. Insti in Pulaway 52, 163–167.

Stec, J., Żmudzki, J., Szczotka, M., and Rachubik, J. (2009). Effects of aflatoxin B1, ochratoxin A, patulin, citrinin, and zearalenone on the in vitro proliferation of pig blood lymphocytes. Bulletin of the Vet. Insti in Pulaway 53, 129–134.

Stecher, B., Robbiani, R., Walker, A.W., Westendorf, A.M., Barthel, M., Kremer, M., Chaffron, S., Macpherson, A.J., Buer, J., Parkhill, J., et al. (2007). Salmonella enterica serovar typhimurium exploits inflammation to compete with the intestinal microbiota. PLoS Biol. 5, 2177–2189.

Stoev, S.D. (2010). Studies on carcinogenic and toxic effects of ochratoxin A in chicks. Toxins 2, 649–664.

Stoev, S., Denev, S., Dutton, M., and Nkosi, B. (2009). Cytotoxic Effect of Some Mycotoxins and their Combinations on Human Peripheral Blood Mononuclear Cells as Measured by the MTT Assay. Open Toxinology J. 2, 1–8.

Stoev, S.D., Stefanov, M., Denev, S., Radic, B., Domijan, A.M., and Peraica, M. (2004). Experimental mycotoxicosis in chickens induced by ochratoxin A and penicillic acid and intervention with natural plant extracts. Vet. Res. Commun. 28, 727–746.

Stork, L.G., Gennings, C., Carter, W.H., Johnson, R.E., Mays, D.P., Simmons, J.E., Wagner, E.D., and Plewa, M.J. (2007). Testing for additivity in chemical mixtures using a fixed-ratio ray design and statistical equivalence testing methods. J. Agric. Biol. Environ. Stat. 12, 514–533.

Stout, R.D., Jiang, C., Matta, B., Tietzel, I., Watkins, S.K., and Suttles, J. (2005). Macrophages sequentially change their functional phenotype in response to changes in microenvironmental influences. J. Immunol. Baltim. Md 1950 175, 342–349.

Subkhankulova, T., Zhang, X., Leung, C., and Marino, S. (2010). Bmi1 directly represses p21Waf1/Cip1 in Shh-induced proliferation of cerebellar granule cell progenitors. Mol. Cell. Neurosci. 45, 151–162.

Sun, Z., Singh, N., Mullican, S.E., Everett, L.J., Li, L., Yuan, L., Liu, X., Epstein, J.A., and Lazar, M.A. (2011). Diet-induced lethality due to deletion of the Hdac3 gene in heart and skeletal muscle. J. Biol. Chem. 286, 33301–33309.

166

Sung, J.Y., Hong, J.H., Kang, H.S., Choi, I., Lim, S.D., Lee, J.K., Seok, J.H., Lee, J.H., and Hur, G.M. (2000). Methotrexate suppresses the interleukin-6 induced generation of reactive oxygen species in the synoviocytes of rheumatoid arthritis. Immunopharmacology 47, 35–44.

Svegliati, S., Cancello, R., Sambo, P., Luchetti, M., Paroncini, P., Orlandini, G., Discepoli, G., Paterno, R., Santillo, M., Cuozzo, C., et al. (2005). Platelet-derived growth factor and reactive oxygen species (ROS) regulate Ras protein levels in primary human fibroblasts via ERK1/2. Amplification of ROS and Ras in systemic sclerosis fibroblasts. J. Biol. Chem. 280, 36474–36482.

Swann, J.B., Hayakawa, Y., Zerafa, N., Sheehan, K.C.F., Scott, B., Schreiber, R.D., Hertzog, P., and Smyth, M.J. (2007). Type I IFN contributes to NK cell homeostasis, activation, and antitumor function. J. Immunol. Baltim. Md 1950 178, 7540–7549.

Swanson, S.P., Nicoletti, J., Rood, H.D., Jr, Buck, W.B., Cote, L.M., and Yoshizawa, T. (1987). Metabolism of three trichothecene mycotoxins, T-2 toxin, diacetoxyscirpenol and deoxynivalenol, by bovine rumen microorganisms. J. Chromatogr. 414, 335–342.

Sweeney, M.J., and Dobson, A.D. (1998). Mycotoxin production by Aspergillus, Fusarium and Penicillium species. Int. J. Food Microbiol. 43, 141–158.

Tanaka, T., Hasegawa, A., Yamamoto, S., Lee, U.S., Sugiura, Y., and Ueno, Y. (1988). Worldwide contamination of cereals by the Fusarium mycotoxins nivalenol, deoxynivalenol, and zearalenone. 1. Survey of 19 countries. J. Agric. Food Chem. 36, 979–983.

Tangni, E.K., and Pussemier, L. (2006). Ochratoxin A and citrinin loads in stored wheat grains: impact of grain dust and possible prediction using ergosterol measurement. Food Addit. Contam. 23, 181–189.

Tappenden, K.A., and Deutsch, A.S. (2007). The physiological relevance of the intestinal microbiota--contributions to human health. J. Am. Coll. Nutr. 26, 679S – 83S.

Taylor, A., Verhagen, J., Blaser, K., Akdis, M., and Akdis, C.A. (2006). Mechanisms of immune suppression by interleukin-10 and transforming growth factor-beta: the role of T regulatory cells. Immunology 117, 433–442.

Tesh, V.L., Ramegowda, B., and Samuel, J.E. (1994). Purified Shiga-like toxins induce expression of proinflammatory cytokines from murine peritoneal macrophages. Infect. Immun. 62, 5085– 5094.

Thust, R., and Kneist, S. (1979). Activity of citrinin metabolized by rat and human microsome fractions in clastogenicity and SCE assays on Chinese hamster V79-E cells. Mutat. Res. 67, 321– 330.

Tiwary, A.K., Puschner, B., and Poppenga, R.H. (2009). Using roquefortine C as a biomarker for penitrem A intoxication. J. Vet. Diagn. Investig. Off. Publ. Am. Assoc. Vet. Lab. Diagn. Inc 21, 237–239.

167

Tourneur, E., and Chassin, C. (2013). Neonatal immune adaptation of the gut and its role during infections. Clin. Dev. Immunol. 2013, 270301.

Uhlig, H.H., McKenzie, B.S., Hue, S., Thompson, C., Joyce-Shaikh, B., Stepankova, R., Robinson, N., Buonocore, S., Tlaskalova-Hogenova, H., Cua, D.J., et al. (2006). Differential activity of IL- 12 and IL-23 in mucosal and systemic innate immune pathology. Immunity 25, 309–318.

Ukai, T., Yamamoto, Y., and Yamamoto, T. (1954). Substance from a strain of Penicillium. II. Culture method of Hori-Yamamoto strain and chemical structure of its poisonous substance. J Pharm Soc Jpn 74, 450.

Usleber, E., Dade, M., Schneider, E., Dietrich, R., Bauer, J., and Märtlbauer, E. (2008). Enzyme immunoassay for mycophenolic acid in milk and cheese. J. Agric. Food Chem. 56, 6857–6862.

Veselá, D., Veselý, D., and Jelínek, R. (1983). Toxic effects of ochratoxin A and citrinin, alone and in combination, on chicken embryos. Appl. Environ. Microbiol. 45, 91–93.

Visconti, A., Minervini, F., Lucivero, G., and Gambatesa, V. (1991). Cytotoxic and immunotoxic effects of Fusarium mycotoxins using a rapid colorimetric bioassay. Mycopathologia 113, 181– 186.

Visconti, A., Pascale, M., and Centonze, G. (1999). Determination of ochratoxin A in wine by means of immunoaffinity column clean-up and high-performance liquid chromatography. J. Chromatogr. A 864, 89–101.

Vogel, I., Verbinnen, B., Maes, W., Boon, L., Van Gool, S.W., and Ceuppens, J.L. (2013). Foxp3+ regulatory T cells are activated in spite of B7-CD28 and CD40-CD40L blockade. Eur. J. Immunol. 43, 1013–1023.

Vrabcheva, T., Usleber, E., Dietrich, R., and Märtlbauer, E. (2000). Co-occurrence of ochratoxin A and citrinin in cereals from Bulgarian villages with a history of Balkan endemic nephropathy. J. Agric. Food Chem. 48, 2483–2488.

Wadia, P.P., Herrera, N.D., Abecassis, M.M., and Tambur, A.R. (2009). Mycophenolic acid inhibits maturation and function of human dendritic cells and B cells. Hum. Immunol. 70, 692– 700.

Wagener, R.E., Davis, N.D., and Diener, U.L. (1980). Penitrem A and Roquefortine Production by Penicillium commune. Appl. Environ. Microbiol. 39, 882–887.

Wagner, D., Sangari, F.J., Kim, S., Petrofsky, M., and Bermudez, L.E. (2002). Mycobacterium avium infection of macrophages results in progressive suppression of interleukin-12 production in vitro and in vivo. J. Leukoc. Biol. 71, 80–88.

Wang, Y.A., Kamarova, Y., Shen, K.C., Jiang, Z., Hahn, M.-J., Wang, Y., and Brooks, S.C. (2005). DNA methyltransferase-3a interacts with p53 and represses p53-mediated gene expression. Cancer Biol. Ther. 4, 1138–1143.

168

Weber, L.T., Shipkova, M., Lamersdorf, T., Niedmann, P.D., Wiesel, M., Mandelbaum, A., Zimmerhackl, L.B., Schütz, E., Mehls, O., Oellerich, M., et al. (1998). Pharmacokinetics of mycophenolic acid (MPA) and determinants of MPA free fraction in pediatric and adult renal transplant recipients. German Study group on Mycophenolate Mofetil Therapy in Pediatric Renal Transplant Recipients. J. Am. Soc. Nephrol. JASN 9, 1511–1520.

Weigel, G., Griesmacher, A., Karimi, A., Zuckermann, A.O., Grimm, M., and Mueller, M.M. (2002). Effect of mycophenolate mofetil therapy on lymphocyte activation in heart transplant recipients. J. Heart Lung Transplant. Off. Publ. Int. Soc. Heart Transplant. 21, 1074–1079.

Wichmann, G., Herbarth, O., and Lehmann, I. (2002). The mycotoxins citrinin, gliotoxin, and patulin affect interferon-gamma rather than interleukin-4 production in human blood cells. Environ. Toxicol. 17, 211–218.

Wigginton, J.M., Gruys, E., Geiselhart, L., Subleski, J., Komschlies, K.L., Park, J.W., Wiltrout, T.A., Nagashima, K., Back, T.C., and Wiltrout, R.H. (2001). IFN-gamma and Fas/FasL are required for the antitumor and antiangiogenic effects of IL-12/pulse IL-2 therapy. J. Clin. Invest. 108, 51–62.

Wilson, A.J., Byun, D.-S., Popova, N., Murray, L.B., L’Italien, K., Sowa, Y., Arango, D., Velcich, A., Augenlicht, L.H., and Mariadason, J.M. (2006). Histone deacetylase 3 (HDAC3) and other class I HDACs regulate colon cell maturation and p21 expression and are deregulated in human colon cancer. J. Biol. Chem. 281, 13548–13558.

Wong, J.C., Guo, L., Peng, Z., Zhang, W., Zhang, N., Lai, W., Zhang, Z., Zhang, C., Zhang, X., Song, S., et al. (2011). Application of p21 and klf2 reporter gene assays to identify selective histone deacetylase inhibitors for cancer therapy. Bioorg. Med. Chem. Lett. 21, 110–116.

Woo, S.-R., and Czuprynski, C.J. (2008). Tactics of Mycobacterium avium subsp. paratuberculosis for intracellular survival in mononuclear phagocytes. J. Vet. Sci. 9, 1.

Wu, T.-S., Liao, Y.-C., Yu, F.-Y., Chang, C.-H., and Liu, B.-H. (2008). Mechanism of patulin- induced apoptosis in human leukemia cells (HL-60). Toxicol. Lett. 183, 105–111.

Wykes, M., and Macpherson, G. (2000). Dendritic cell-B-cell interaction: dendritic cells provide B cells with CD40-independent proliferation signals and CD40-dependent survival signals. Immunology 100, 1–3.

Wynn, T.A., Chawla, A., and Pollard, J.W. (2013). Macrophage biology in development, homeostasis and disease. Nature 496, 445–455.

Xiao, C.L., Kim, Y.K., and Boal, R.J. (2011). First Report of Occurrence of Pyrimethanil Resistance in Penicillium expansum from Stored Apples in Washington State. Plant Dis. 95, 72– 72.

Xing, Z., Zganiacz, A., and Santosuosso, M. (2000). Role of IL-12 in macrophage activation during intracellular infection: IL-12 and mycobacteria synergistically release TNF-alpha and nitric oxide from macrophages via IFN-gamma induction. J. Leukoc. Biol. 68, 897–902.

169

Xue, C.Y., Wang, G.H., Chen, F., Zhang, X.B., Bi, Y.Z., and Cao, Y.C. (2010). Immunopathological effects of ochratoxin A and T-2 toxin combination on broilers. Poult. Sci. 89, 1162–1166.

Yalowitz, J.A., and Jayaram, H.N. (2000). Molecular targets of guanine nucleotides in differentiation, proliferation and apoptosis. Anticancer Res. 20, 2329–2338.

Yan, F., and Polk, D.B. (2004). Commensal bacteria in the gut: learning who our friends are. Curr. Opin. Gastroenterol. 20, 565–571.

Yiannikouris, A., André, G., Buléon, A., Jeminet, G., Canet, I., François, J., Bertin, G., and Jouany, J.-P. (2004). Comprehensive Conformational Study of Key Interactions Involved in Zearalenone Complexation with β- D -Glucans. Biomacromolecules 5, 2176–2185.

Yu, L.C.-H., Wang, J.-T., Wei, S.-C., and Ni, Y.-H. (2012). Host-microbial interactions and regulation of intestinal epithelial barrier function: From physiology to pathology. World J. Gastrointest. Pathophysiol. 3, 27–43.

Yu, T., Jhun, B.S., and Yoon, Y. (2011). High-glucose stimulation increases reactive oxygen species production through the calcium and mitogen-activated protein kinase-mediated activation of mitochondrial fission. Antioxid. Redox Signal. 14, 425–437.

Yuan, Z.-L., Guan, Y.-J., Chatterjee, D., and Chin, Y.E. (2005). Stat3 dimerization regulated by reversible acetylation of a single lysine residue. Science 307, 269–273.

Yue, T., Dong, Q., Guo, C., and Worobo, R.W. (2011). Reducing patulin contamination in apple juice by using inactive yeast. J. Food Prot. 74, 149–153.

Zhu, H., Shan, L., Schiller, P.W., Mai, A., and Peng, T. (2010). Histone deacetylase-3 activation promotes tumor necrosis factor-alpha (TNF-alpha) expression in cardiomyocytes during lipopolysaccharide stimulation. J. Biol. Chem. 285, 9429–9436.

170

Appendix 1: Schematic diagram of how Penicillium mycotoxins alter the cell activities.

Green arrow represents ‘stimulation’, and red indication represents ‘inhibition.’ MPA or PA inhibits HDAC-3 expression and Bmi-1 increasing p21 activity, which then down-regulate the proliferation (Spurling et al., 2008; Wilson et al., 2006; Wong et al., 2011). The halted proliferation by p21 can also stimulate the differentiation of the cells. PA stimulates JMJD-3 expression, which can also induce p21 activity by inducing p53 (Barradas et al., 2009; Bringold and Serrano, 2000). Although PA inhibits DNMT-3b, it does not alter the p19/ARF and p16/INK4A activities as DNMT-1 is still active (Kurita et al., 2010).

171

Appendix 2: Preliminary study on the effect of 6 h in vitro PM exposure at different concentrations on BoMac ROS production.

CIT (0.1 ~ 800 M) OTA (0.00 ~ 0 M) 250 250 % ROS production % cell viability % ROS production % cell viability 200 200

150 150 contrl contrl 100 100

50 50

% relative to control to relative %

% relative to control to relative % 0 0 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3 log(exposure concentration) log(exposure concentration) PAT (0.0007 ~ 50 M) 250 MPA (0.002 ~ 1,000 M) % ROS production % cell viability 250 200 % ROS production % cell viability 200 150 150 contrl 100 contrl 100 50 50

% relative to control to relative %

0 control to relative % -4 -3 -2 -1 0 1 2 0 -3 -2 -1 0 1 2 3 4 log(exposure concentration) log(exposure concentration)

PA (0.03 ~ 1,000 M) 250 % ROS production % cell viability 200

150 contrl 100

% relative to control to relative % 0 -2 -1 0 1 2 3 4 log(exposure concentration)

‘Blue rectangle’ with blue line indicates the % ROS production in relative to control; ‘black dotted line’ represents the cell viability. ‘Red arrow’ indicates IC25 for each PMs. Prior to selecting the exposure concentrations for ROS production (Table 5.1), wide ranges of PM concentrations were tested to determine approximate concentration range that may have altered the ROS production.

172

Appendix 3: Intracellular ROS production of BoMacs with different stimulators by using Carboxy-H2DFFDA.

Intracellular ROS production with different stimulators 20000

LPS 1 ug/ml 15000 LPS 1 g/ml LPS 0.100 g/ml Smeg 10 g/ml Smeg 1 g/ml 10000 Smeg 0.1 g/ml PAM-3 10 g/ml tBHp 50 M control 5000

0 0 5 10 15 20 25 30

Intracellular Carboxy-H2DFFDA flu reading flu Carboxy-H2DFFDA Intracellular time (h)

Only PAM-3 at 10 μg/ml dramatically induced the ROS production in BoMacs after 3 h of incubation. 24 h of PAM-3 had higher intracellular ROS reading as compared to 6 h of PAM-3. However, the most increase was observed in between 3 to 6 h of incubation, so 6 h incubation with 10 μg/ml PAM-3 was used as optimized ROS assay condition with H2DFFDA dye.

173

Appendix 4: Effect of PAT on the viability of BoMacs.

100 % cell death 80

60 40

% cell d eath 20

CyQUANT®0 GR dye

(Ethidium hom0.00 odim er-1/0.10 0.20 0.50 1.00 % M A +

To determine if PAT incubation with MA+ decrease the cell viability. (See section 6.3 for more detail method). Prior to freezing down the plates for proliferation assay, ethidium homodimer-1 (excitation 528/emission 617 nm) were used to estimate the death cell number from the half of the replicates (triplicates) of treatments. The cell number from proliferation assay indicated the total cell number from the treatment, and the measurement from Eth-1 solution indicated the dead cell number. The percent in the graph was estimated by dividing dead cell number from the total cell number. (This step was necessary, since the total cell number was affected depending on the percent amount of MA+ as it was shown in Figure 3C. The study showed that the cell death increased as the percent MA+ increased.

174

Appendix 5: Effect of MA+ on BoMac proliferation.

The respective percent of MA+ was mixed with water at pH 7.0 and incubated for 2 h. The MA+ was then removed by centrifugation at 10,000 g for 10 min before the cell culture procedure. Calcein AM (excitation 494/emission 517 nm) was used to measure the cell viability (see section 3.3.2. in Chapter 3 for detail).

175