The Consequences of Acrylamide Exposure on the Male Germ Line

117

Belinda Jean Nixon B. Biotech. (Hons) Class I

A thesis submitted to the Faculty of Science and Information Technology, The University of Newcastle, Australia, in fulfilment of the requirements of the degree of Doctor of Philosophy.

December 2013

thesis by publication thesis by publication thesis by publication Acknowledgements

Firstly, I'd like to thank my supervisor, Dr. Shaun Roman, not just because my PhD would have been impossible without him, but for everything he taught me along the way. Thank you for being able to talk sci-fi as much as science, talk music (from punk to DnB) and for teaching me how to put the 'Bam!' into a scientific presentation. I'd particularly like to thank him for the special effort he made to be at my final presentation, despite the difficult circumstances, it really meant a lot. It was never boring, very challenging, hilarious fun and when I look back, I doubt I would have had it any other way.

I also owe a debt of gratitude to my co-supervisor, Prof. Brett Nixon. Your guidance and advice were invaluable, and you provided an all too important third opinion in many discussions! I always knew I could count on your support and often your words were what stayed with me and pulled me through the most difficult times, I'm truly grateful.

I'd like to extend my thanks to everyone in the Reproductive Science group. In particular, I'd like to thank Simone Stanger, for all her help in the lab; Aimee Katen, for being a welcome helping hand, and also Prof. Eileen McLaughlin, for her advice and support. Thanks also to my good friend Amanda Anderson, who's been with me from the beginning, and to the rest of my PhD buddies Jessie Sutherland, Alex Sobinoff, Kate Redgrove and Skye McIver. Special thanks go to my fellow PhD comrade, Andrew Reid, who educated me in all other things important to student life, including beer, pool, post-rock, guitar solos and chicken satay. Dude, thanks for everything, it was a blast and I know I've got a friend for life.

To my good friends, Sarah Bullock, Brett Edman, and my little bro, Gareth, thanks for all the fun times and laughter, you guys know how I feel :) I'm also indebted to my parents, who have given me so much. Thank you for all your love and encouragement over the years (and for not kicking me out of home).

Finally, to Anthony Philippa, you picked me up from my lowest of lows and celebrated my triumphs as if they were your own. I will always be in awe of your unwavering love and support, particularly after all the crazy I threw your way. Thank you for sticking around; words cannot express how much it meant to me.

Manuscripts Included as Part of this Thesis

o Nixon, B. J., Stanger, S., Nixon, B. and Roman, S. (2012). Chronic exposure to acrylamide induces DNA damage in male germ cells of mice. Toxicological Sciences: An official journal of the Society of Toxicology. 129: 135-145.

Permission Regarding Copyright

I, Belinda Jean Nixon, warrant that I have obtained, where necessary, permission from the copyright owners to use my own published journal articles in which the copyright is held by another party (e.g. publisher).

Date: 21st December 2013

iii

List of Additional Publications o Nixon, B. J., Nixon, B. and Roman, S. D. (2012) The Consequences of Acrylamide Exposure on the Male Germ Line. Society for Reproductive Biology, Brisbane, QLD, Australia. Oral presentation. o Nixon, B. J., Nixon, B. and Roman, S. D. (2011) The Consequences of Chronic Exposure to Acrylamide on the Male Germ Line. Biology RHD Conference, University House, University of Newcastle, Australia. Oral presentation. o Nixon, B. J., Nixon, B. and Roman, S. D. (2011) The Consequences of Acrylamide Exposure on the Male Germ Line. Joint World Congress on Reproductive Biology and Society for Reproductive Biology, Cairns, QLD, Australia. Oral presentation. o Nixon, B. J., Nixon, B. and Roman, S. D. (2011) The Response of Early Male germ Cells to Acrylamide Exposure. Joint World Congress on Reproductive Biology and Society for Reproductive Biology, Cairns, QLD, Australia. Poster presentation. o Nixon, B. J., Nixon, B. and Roman, S. D. (2010) The Consequences of Acrylamide Exposure on the Male Germ Line. Society for Reproductive Biology, Sydney Exhibition and Convention Centre, Australia (2010) Oral presentation. o Nixon, B. J., Nixon, B. and Roman, S. D. (2010) The Consequences of Acrylamide Exposure on Early Male Germ Cells. OzBio, Melbourne Convention and Exhibition Centre, Australia. Poster presentation. o Nixon, B. J., Nixon, B. and Roman, S. D. (2009) The Consequences of Chronic Exposure to Acrylamide on the Male Germ Line. ARC Centre of Excellence for Biotechnology and Development Conference, Melbourne, Australia. Oral presentation.

Table of Contents

Declarations ...... 1

Acknowledgements ...... 2

Manuscripts Included as Part of this Thesis ...... 3

Permission Regarding Copyright ...... 3

List of Additional Publications ...... 4

Abstract ...... 1

Chapter 1: Literature Review...... 2

INTRODUCTION ...... 2

ACRYLAMIDE IN THE DIET ...... 4

ABSORPTION, DISTRIBUTION, METABOLISM & EXCRETION OF ACRYLAMIDE ...... 7

ACRYLAMIDE TOXICITY ...... 10 Neurotoxicity ...... 10 Carcinogenicity ...... 11 Reproductive Toxicity ...... 12

SPERMATOGENESIS ...... 15

ACRYLAMIDE GENOTOXICITY IN LATE-STAGE SPERMATOGENESIS ...... 19

DNA DAMAGE IN THE MALE GERM LINE ...... 20

RESEARCH DESIGN ...... 25

Chapter 2 ...... 27

CHAPTER 2: OVERVIEW ...... 28

MOUSE SPERMATOCYTES EXPRESS CYP2E1 AND RESPOND TO ACRYLAMIDE EXPOSURE ...... 29

Chapter 3 ...... 57

CHAPTER 3: OVERVIEW ...... 58

CHRONIC EXPOSURE TO ACRYLAMIDE INDUCES DNA DAMAGE IN MALE GERM CELLS OF MICE ...... 59

Chapter 4 ...... 87

CHAPTER 4: OVERVIEW ...... 88

THE IMPACT OF 12-MONTH CHRONIC ACRYLAMIDE EXPOSURE ON MOUSE TESTICULAR GENE EXPRESSION 89

Chapter 5: Discussion ...... 118

AIM 1: TO EXAMINE THE NATURE OF GENETIC DAMAGE INDUCED BY ACRYLAMIDE IN ISOLATED MALE GERM

CELLS...... 118

AIM 2: TO INVESTIGATE WHETHER CHRONIC ACRYLAMIDE EXPOSURE INDUCES DNA DAMAGE IN MALE GERM

CELLS IN VIVO...... 120

AIM 3: TO ELUCIDATE THE MOLECULAR MECHANISMS BY WHICH DNA DAMAGE IS GENERATED BY ACRYLAMIDE

AND THE RESPONSE OF THE MALE GERM LINE TO CHRONIC TOXIC EXPOSURE...... 121

CONCLUSIONS & CONTRIBUTION TO CURRENT LITERATURE ...... 123

FUTURE DIRECTIONS ...... 124

References ...... 126

Appendices ...... 133

APPENDIX A: CHAPTER 2 SUPPLEMENTARY DATA ...... 134

APPENDIX B: CHAPTER 3 SUPPLEMENTARY DATA ...... 140

APPENDIX C: CHAPTER 4 SUPPLEMENTARY DATA ...... 145

An electronic version of this thesis, together with all associated files and data, are included on disc found on the inside back cover of this thesis.

Abstract

Acrylamide is a commonly used industrial compound; however it is also naturally occurring in cooked foods such as potatoes, breads and coffee. Since the discovery of acrylamide formation in food, significant research has been carried out to determine the consequences of human dietary exposure to acrylamide. The reproductive toxicity of acrylamide was the focus of this thesis; however the compound is also known to cause neurotoxicity, carcinogenicity and genotoxicity. In rodent studies, acrylamide reproductive toxicity is primarily mediated through paternal exposure, and has been observed to induce DNA damage, male infertility, dominant lethality, heritable translocations and embryo resorptions. Thus, prolonged exposure to acrylamide in males not only has implications within the individual, but may have consequences for future offspring.

The aims of this thesis were to examine the nature of DNA damage induced by acrylamide in male germ cells, and to investigate whether this damage could be induced at levels equivalent to human estimates. These aims were addressed using a series of experiments in isolated mouse germ cells, as well as a chronic exposure study in which acrylamide was administered to male mice via the drinking water for one year. Genome-wide microarray analyses were subsequently used to explore the molecular mechanisms that mediate the damage induced in the testis of acrylamide exposed mice.

The results of this study indicated that acrylamide-mediated DNA damage was likely due to the presence of both DNA adducts as well as oxidative damage in mouse male germ cells. Furthermore, chronic exposure to acrylamide in male mice led to significant dose and time- dependent increases in DNA damage in male germ cells. Microarray analyses offered insight into the mechanisms that generate deleterious effects in the testes of mice following acrylamide exposure, and identified several potential biomarkers of exposure. The outcomes of this research will provide better understanding of acrylamide toxicity and shed light on the consequences of xenobiotic exposure in the male.

1 Chapter 1: Literature Review

Introduction

Acrylamide is a common industrial chemical that has been found to form during the cooking of various carbohydrate-rich foods (Tareke et al. 2002). The compound elicits a number of toxic effects such as neurotoxicity, genotoxicity and reproductive toxicity; and has been classified by the International Agency for Research on Cancer as a Group 2A carcinogen that is “probably carcinogenic to humans” (IARC 1994). The major sources of acrylamide are in industry as it is an intermediate monomer of polyacrylamide, which has numerous applications in water treatment, mining, ore processing, soil stabilisation, cosmetics, paper production, fabric manufacturing, and in biomolecular laboratories for use in gel electrophoresis (Parzefall 2008). Cigarettes are another significant source of acrylamide as the compound forms in relatively large quantities in tobacco smoke (Schumacher et al. 1977). However, the presence of acrylamide in commonly consumed foods indicates that exposure is more widespread within the general population. Due to the potential toxic effects of acrylamide, there is concern that chronic dietary exposure to acrylamide may have detrimental effects on human health.

The discovery of acrylamide in food originated from a Swedish study by Bergmark (1997) in which the level of acrylamide exposure was quantified in biomolecular laboratory personnel, one of the largest groups exposed to industrial levels of acrylamide. The level of haemoglobin adducts of acrylamide in the blood was used as a measure of exposure and was compared to the level in smokers and non-smokers. However, an unusually high background level of adducts were identified in the non-smoking control group (see Fig. 1, Pg. 3). This background level was too high to be the result of environmental exposure to tobacco smoke or acrylamide intake through drinking water; thus, alternative origins of exposure were considered, such as food and beverages. This was further investigated in an experiment by Tareke et al. (2000) in which the level of acrylamide haemoglobin adducts were measured in rats fed fried or non-fried animal feed. Rats on the fried diet had a higher level of haemoglobin adducts than rats on the non-fried diet, similar in magnitude to the level observed in non-smoking humans. Two years later, the same authors reported that

2 trace amounts of acrylamide could be detected in commonly consumed, cooked or fried foods such as French fries, potato crisps and breads (Tareke et al. 2002).

Figure 1: Mean (±SD) levels of acrylamide haemoglobin adducts (AAVal) detected in blood samples from non-smokers (Controls), polyacrylamide gel electrophoresis (PAGE) workers and Smokers as reported by Bergmark et al. (1997). PAGE workers and Smokers were found to have significantly increased levels of acrylamide adducts compared to Controls. The authors speculated that the unusually high background level observed in the Controls may be attributable to food or beverages.

In response to this discovery, the Food and Agriculture Organisation and World Health Organisation (FAO/WHO) held an international meeting to address the issue of dietary acrylamide exposure in the general population. Whilst the presence of acrylamide in the diet was of significant concern, it was concluded that a proper risk assessment could not be made until the mechanisms of acrylamide formation and toxicity were better understood. Thus, the committee made a number of recommendations to further research into the formation and toxic effects of acrylamide, improve human exposure estimates, and to develop strategies to reduce levels of the compound in foods (FAO/WHO 2003; WHO 2002; WHO 2011). A number of these recommendations relate to the reproductive toxicity of acrylamide exposure and included the further assessment of:

o The mechanisms of action and dose response characteristics for the effects of acrylamide and glycidamide on germ cell damage,

o The genotoxic effects on germ cells using genome-wide expression profiling, and;

o The dose-response characteristics of acrylamide and glycidamide relative to toxicity, disposition, and binding to DNA and macromolecules.

To date, the reproductive toxicity of acrylamide has only been observed in animal experiments. However, there are concerns that widespread, chronic acrylamide exposure in humans may have downstream consequences for human reproductive health and development of future progeny (Exon 2006). Thus, the research presented in this thesis was aimed at addressing the above recommendations to further aid in risk assessments of acrylamide exposure in humans.

In order to provide sufficient background to this topic of research, the following pages review the current literature regarding human exposure to acrylamide, animal exposure studies and the potential mechanisms involved in male reproductive toxicity.

4 Acrylamide in the Diet

The formation of acrylamide in food occurs through the Maillard browning reactions; complex chemical reactions between amino acids and reducing sugars that are responsible for the development of colour and flavours during cooking. Acrylamide formation during these reactions largely relies on the interaction of the amino acid, asparagine, with the carbonyl group of reducing sugars such as glucose and fructose (Friedman and Levin 2008). Typical foods that contain significant quantities of acrylamide are usually plant products that are low in protein, high in carbohydrate, and cooked at temperatures 120°C and above. Data from 17 different countries from Europe, North America, Asia and the Pacific regions indicate that the major food items that contribute to acrylamide in the diet are potato chips (16-30%), potato crisps (6-46%), coffee (13-39%), pastry and sweet biscuits (10-20%), breads and rolls/toast (10-30%) (WHO/JECFA 2005). Table 1 (Pg. 6) gives an overview of the range of acrylamide levels found in foods, based on data provided from Australia, Norway, Sweden, Switzerland, the United Kingdom and the United States of America. However, it can be difficult to quantify the acrylamide content of specific foods, as its formation is affected by conditions of time, temperature, pH, food constituents and water content. Therefore, levels of the compound can vary depending on different cooking methods, different brands, or even between different batches within the same brand, which complicates estimates of human exposure (Friedman and Levin 2008).

Several studies have estimated the average acrylamide intake in the diet to be approximately 0.5 μg/kg bodyweight/day, though wide variations exist between different populations, genders, and age groups. For comparison, the calculated mean and upper boundary limit for industrial inhalation exposure of acrylamide is 1.4 – 18.6 µg/kg bodyweight/day (NTP 2005). In children, dietary acrylamide intake is generally two to three folds greater than in adults, primarily due to differences in bodyweight. Different eating habits also affect estimates of exposure and values can range from 0.3 to 2.0μg/kg bodyweight/day or 0.6 to 3.5μg/kg bodyweight/day for high consumers (Friedman et al. 2008; Exon 2006). Furthermore, difficulties in estimation of acrylamide intake also arise due to imprecise information given by respondents on their dietary habits (WHO 2011). These estimates however, generally do not account for acrylamide exposure via drinking water or

5 cosmetics. For further details regarding acrylamide exposure sources, the reader is referred to the NTP-CERHR Monograph on the Potential Human Reproductive and Developmental Effects of Acrylamide (NTP 2005).

Table 1: Summary of acrylamide levels in different foods and food product groups provided by several national agencies, including Australia, Norway, Sweden, Switzerland, the United Kingdom, the United States of America and the International Agency for Research on Cancer (IARC). (WHO 2002) . Acrylamide levels (µg/kg)1 Minimum – Number of Food/Product Group Mean2 Median2 Maximum samples Crisps, potato/sweet potato3 1312 1343 170 – 2287 38 Chips, potato4 537 330 <50 – 3500 39 Batter based products 36 36 <30 – 42 2 Bakery products 112 <50 <50 – 450 19 Biscuits, crackers, toast, bread crisps 423 142 <30 – 3200 58 Breakfast cereals 298 150 <30 – 1346 29 Crisps, corn 218 167 34 – 416 7 Bread, soft 50 30 <30 – 162 41 Fish and seafood products, crumbed, battered 35 35 30 – 39 4 Poultry or game, crumbed, battered 52 52 39 – 64 2 Instant malt drinks 50 50 <50 – 70 3 Chocolate powder 75 75 <50 – 100 2 Coffee powder 200 200 170 – 230 3 Beer <30 <30 <30 1

1 The limits of detection and quantification varied among laboratories; values reported as less than a value are below the limit reported by the laboratory. 2 Mean and median were calculated where individual data were available; samples sizes were extremely small particularly for some food categories; where the mean and median are different it reflects the skewed distribution of the underlying data that were collected in different countries and may represent different food items within the larger category. 3 Products that are thinly sliced and fried (Includes foods called potato chips in some regions including North America) 4 Products that are more thickly sliced (Includes foods called French fries in some regions including North America)

This level of exposure is still considerably lower than doses observed to induce acrylamide related toxicity in animal studies. For example, the No Observable Effect Level (NOEL) for acrylamide neurotoxicity and reproductive toxicity is 0.2mg and 2mg/kg bodyweight/day respectively. In addition, acrylamide carcinogenicity in rodents is observed at doses 1000 to 10,000 times greater than estimated human dietary exposure (FAO/WHO 2005; Mucci and Adami 2009). However, studies in animals have predominantly focussed on toxic effects following acute doses, which do not reflect the human situation. It is possible that long-term exposure to acrylamide will have a cumulative effect, leading to development of cancers,

6 neurotoxic effects, germ cell mutations, or adverse effects on reproduction and early development.

Absorption, Distribution, Metabolism & Excretion of Acrylamide

The uptake of acrylamide can occur through inhalation, ingestion or absorption through the skin. Due to its low molecular weight and hydrophilicity, acrylamide has been found to rapidly distribute to tissues in the body in rats, pigs and dogs (NTP 2005). Following oral administration of acrylamide in male and pregnant female Swiss-Webster mice, acrylamide distributed to the gastrointestinal tract, oesophagus, liver, pancreas, brain, gallbladder, bronchi, and foetal tissue (Marlowe et al. 1986). Acrylamide can also transit the blood/testis barrier as the compound was detected in mouse testis following exposure using radioactively labelled [14C] acrylamide, 1 hour after administration (Marlowe et al. 1986). Labelled acrylamide was observed to move progressively from the testis to the epididymis, paralleling the movement of late stage germ cells during their development in spermatogenesis (See Figure 3, Pg 17). In humans, acrylamide has been observed to cross the blood/placenta barrier in an in vitro model, as well as the human blood/breast milk barrier in vivo (Sorgel et al. 2002).

Metabolism of acrylamide generates the more reactive epoxide metabolite, glycidamide (Figure 2). Both acrylamide and glycidamide can interfere with proteins and form adducts with biological macromolecules as well as DNA; though the reactivity of acrylamide with DNA is relatively slow and the presence of acrylamide-DNA adducts have not been observed in experimental animals to date (NTP 2012). The enzyme that mediates this conversion of acrylamide to glycidamide is a cytochrome P450 enzyme, CYP2E1.

Figure 2: Metabolism of Acrylamide. Acrylamide can be oxidised by a cytochrome P450 enzyme, CYP2E1, to produce glycidamide. Glycidamide is classified as an epoxide as it contains an ether ring structure. Acrylamide and glycidamide can form adducts with DNA (see Fig 3); however acrylamide reacts relatively slowly with DNA. Alternatively, both acrylamide and glycidamide can form adducts with the cysteine residues of proteins, or be conjugated with glutathione (GSH). Glutathione conjugates are metabolised to various mercapturic acids that are excreted in the urine. The major urinary metabolites found in humans are acrylamide, N-acetyl-S-(2-carbamoylethyl)cysteine, glycidamide, 2,3- dihydroxypropanamide, N-acetyl-S-(1-carbamoyl-2-hydroxyethyl)cysteine and N-acetyl-S-(2- carbamoyl-2-hydroxyethyl)cysteine. Adapted from (NTP 2012) .

8 Cytochrome P450 enzymes (CYPs) are largely expressed in the liver and have major roles in endogenous metabolism as well as xenobiotic metabolism of drugs, pollutants and environmental chemicals. The oxidative activity of CYPs however, often leads to bioactivation, or generation of a more reactive metabolite than the parent compound (Guengerich 2008). CYP2E1 metabolises a range of exogenous substances, such as ethanol, aromatic hydrocarbons, drugs, solvents, as well as acrylamide. It is one of several CYP enzymes known to cause bioactivation, such as oxidation of acrylamide to glycidamide (Nebert and Russell 2002). Alternative pathways for the generation of glycidamide from acrylamide are not known; the activity of CYP2E1 appears to be the primary mechanism of glycidamide formation in mammals. Experiments in Cyp2e1 knockout mice (lower case when referring to mouse homologues) exposed to 50mg/kg acrylamide, found that levels of glycidamide-DNA adducts were reduced by 52-66 fold in the liver, testes, and lung tissues, compared to wildtype mice (Ghanayem et al. 2005a). CYP2E1 is also the relevant enzyme for acrylamide-glycidamide conversion in humans as supported by a study by Settels et al. (2008) that demonstrated with genetically engineered V79 cells expressing human CYP2E1 and with human liver microsomes, acrylamide was metabolised to glycidamide. In addition, the introduction of specific CYP2E1 inhibitors in these systems inhibited the generation of glycidamide by almost 90%. Thus, CYP2E1 oxidation is likely the specific pathway that leads to acrylamide-glycidamide formation in humans as well.

Competing metabolic pathways for acrylamide and glycidamide also exist and involve glutathione conjugation, leading to detoxification (shown in Figure 2, Pg 8). These detoxifying pathways occur through the activity of Phase II enzymes, glutathione-S- transferases (GSTs). GSTs conjugate glutathione to render compounds water soluble for excretion; although direct involvement of GSTs in acrylamide metabolism has yet to be confirmed (Fennell et al. 2005). The glutathione conjugates of acrylamide and glycidamide are further metabolised in both humans and rodents to produce various mercapturic acids that are excreted in the urine (Doroshyenko et al. 2009; Fuhr et al. 2006). The majority of acrylamide is metabolised via these detoxifying pathways, however the mechanisms thought to elicit acrylamide toxicity relate to either the reactivity of the acrylamide molecule, or its metabolised derivative, glycidamide.

9 Acrylamide Toxicity

As mentioned previously, the major toxicological endpoints of acrylamide exposure include neurotoxicity, carcinogenicity, genotoxicity, and reproductive toxicity. The neurotoxic effects of acrylamide have been reported in humans as well as in animals, though the carcinogenic, genotoxic and reprotoxic effects have only been observed in animal experiments. The reproductive effects of acrylamide primarily affect the male (Tyl et al. 2000b); thus, the toxicity of acrylamide will be discussed in the context of male reproduction. However, acrylamide neurotoxicity and carcinogenicity will also be briefly addressed as the different toxic effects share some overlap between proposed mechanisms of action. For a more in depth review of acrylamide toxicity, several papers are available on the subject (Exon 2006; Shipp et al. 2006) and associated neurotoxic (Lopachin and Gavin 2008), reprotoxic (Tyl and Friedman 2003), carcinogenic and genotoxic effects (Rice 2005; Dearfield et al. 1995).

Neurotoxicity In a study by Hagmar et al. (2001), the effect of acrylamide neurotoxicity in humans was documented in railway tunnel workers that were exposed to acrylamide in a grouting agent, both dermally and via inhalation. The symptoms observed in these individuals included numbness in the hands and feet, ataxia (lack of muscle coordination), progressive muscle weakness, and excessive tiredness. Similar effects have been identified in other mammals such as mice, rats, guinea pigs, cats, dogs and monkeys, with effects most prominent in the hind limbs. Animal experiments in rats demonstrated that acrylamide intoxication at 5 - 50mg/kg bodyweight/day leads to hind-limb foot splay, ataxia, and skeletal muscle weakness, such as reduced fore and hind-leg grip strength (Lopachin and Gavin 2008). Additionally, the neurotoxic effects appear to be cumulative, as low doses over an extended period were found to induce similar effects to that observed from high dose experiments (Lopachin 2004).

Similar to other neurotoxic agents, acrylamide neurotoxicity is associated with nerve damage in the central and peripheral nervous systems; causing retrograde degeneration, or a ‘dying back’ pattern, of distal axonal regions (Miller and Spencer 1985). Hence, neurotoxic

10 effects of acrylamide initially manifest in the extremities and peripheral nervous system. However, the exact molecular mechanisms by which acrylamide produces this axonal degeneration is unclear. One argument involves the formation of adducts with cysteine residues of presynaptic proteins, which impairs the release of neurotransmitters at the nerve terminus (Sickles et al. 2002). Alternatively, acrylamide may inhibit kinesin motor proteins, which are important in fast anterograde transport of signals and nutrients between axons (Friedman et al. 2008). Both mechanisms produce nerve degeneration, which can result in the central and peripheral neuropathy observed in humans and animals following exposure.

Carcinogenicity Acrylamide has been found to act as a multi-site carcinogen in several studies in rats and mice (Reviewed in Rice, 2005 & NTP, 2012). Screening assays in A/J mice, a strain predisposed to lung tumours, have reported dose dependent increases in lung adenomas following acrylamide exposure in the range of 0 - 60mg/kg bodyweight, administered by either oral gavage or intraperitoneal injections (Bull et al. 1984). Two long-term experiments have also been conducted in male and female Fischer 344 rats, in which acrylamide was administered in the drinking water at doses between 0.01 - 3mg/kg bodyweight/day for two years (Friedman et al. 1995; Johnson et al. 1986). At high doses, increased frequencies of mammary gland adenocarcinomas, thyroid follicular cell adenomas and carcinomas, tumours of the central nervous system and malignant mesotheliomas in the epididymis and testicular tunica vaginalis were observed.

The molecular mechanisms of acrylamide carcinogenicity are thought to occur through genotoxic means, as the acrylamide metabolite, glycidamide, is reactive with DNA. Indeed, glycidamide is positive in bacterial Ames tests in Salmonella typhimurium, which are an indicator of mutagenic potential (IARC 1994). Interestingly the reactivity of acrylamide with kinesin related proteins, described earlier as a mechanism for neurotoxicity, could also have a role in carcinogenicity as kinesin proteins are involved in spindle formation and segregation of chromosomes during cell division. It is also interesting to note that in the Fischer 344 rat studies described above, acrylamide exposure produced increased

11 incidences of mesotheliomas in the epididymis and testis (NTP 2012), suggestive of an affinity of acrylamide for hormonally regulated or male reproductive tissues (particularly as the compound also acts as a male reproductive toxicant). Despite this however, there is a lack of evidence to support an endocrine dysregulation mechanism of action (Bowyer et al. 2008; Camacho et al. 2012) and indeed, the mechanisms of acrylamide carcinogenicity in animal experiments have yet to be fully elucidated.

Despite evidence of carcinogenicity in rodent studies, increased cancer risk in humans due to acrylamide exposure has not been observed. A number of epidemiological studies have assessed cancer mortality in occupationally exposed workers and in all cases, no significant correlation between level of exposure and elevated cancer risk was established. Since the discovery of acrylamide in foods, numerous epidemiological studies have investigated the association of dietary acrylamide exposure with increased cancer risk in different populations. Twenty-seven of these studies have been summarised by Mucci and Adami (2009), which collectively cover multiple cancer sites such as bladder, breast, colorectal, endometrial, oesophageal, laryngeal/oropharyngeal, ovarian, pancreatic, prostate, and renal cancer. Convincing evidence of a relationship between dietary acrylamide exposure and incidence of cancer in humans was not found. However, these epidemiological studies are considered to be limited in their statistical power to detect small cancer risks that have a high background incidence (Rice 2005). Numerous variables and confounding factors affect these population-based studies, such as genetic diversity, behaviour, eating habits, imprecise reporting of dietary intake, as well as everyday exposure to a myriad of unrelated substance; thus, these limitations need to be taken into account when considering epidemiological data.

Reproductive Toxicity The reproductive toxicity of acrylamide has been observed in a number of rodent breeding experiments in which acrylamide treatment produced dominant lethality, leading to increases in pre/post implantation losses, embryo resorptions, and reduced litter sizes (Adler et al. 2000; Bishop 1991; Generoso et al. 1996; Shelby et al. 1986; Smith et al. 1986; Tyl et al. 2000b). In a multigenerational study by Tyl et al. (2000a), the F0 and F1

12 generations of male and female rats were exposed to 0.5 to 60 mg/kg bodyweight/day acrylamide for 10 weeks. At exposures 5 mg/kg bodyweight/day and above, significant reductions were found in the number of implantations per dam and survival of pups per litter at birth in both F0 and F1 generations. Similar effects have been observed in matings between Swiss CD-1 mice, in which reductions in live litter size were found following acrylamide exposure at 7.2 mg/kg bodyweight/day in the drinking water (Chapin et al. 1995).

Interestingly, evidence from crossover breeding experiments (treated males mated with untreated females and vice versa) indicated that the reproductive toxicity of acrylamide is dependent on male exposure. Sakamoto and Hashimoto (1986) reported a male-mediated effect of acrylamide in mice exposed to 0 - 19 mg/kg bodyweight/day for eight days. At the highest dose, acrylamide treated males mated with untreated females were found to have low fertility rates and increased incidence of embryo resorptions. These effects were not observed in treated females mated with untreated males. In a similar experiment by Zenick et al. (1986), low pregnancy rates and significant increases in post implantation losses in rats were also only observed following paternal exposure to acrylamide. Other effects on male reproduction in rodents have been reported, including decreased copulatory behaviour, reduced deposition of sperm in the uterus, low sperm counts, decreased sperm motility and abnormal sperm morphology. High doses of acrylamide can also cause reduced testis weight and vacuolations in the seminiferous tubules of the testes (Hashimoto et al. 1981; Yang et al. 2005; Sublet et al. 1989; Wise et al. 1995).

Reproductive toxicity associated with acrylamide exposure has not been observed in humans, however results from rodent studies imply that acrylamide may impact on male fertility and potentially affect future offspring through exposure in the father. Other environmental factors are thought to cause similar effects through paternal exposure, which are reviewed in Savitz and Chen (1990). For example, an increased risk in nervous system tumors in children was reported to correlate with exposure to pesticides and paints in the father. Leukemia was also found to be more prevalent in children whose fathers worked in the woodwork industry (Feychting et al. 2001). In a paper by Regidor et al. (2004), fetal death due to congenital abnormalities was found to be related to paternal pesticide

13 exposure in agriculture. However, the significance of paternal exposure and its impact on offspring is uncertain and the mechanisms involved in the transmission of toxic effects to the next generation are poorly understood.

There has been much speculation into the mechanisms that elicit acrylamide reproductive toxicity. Despite the effect on males, an endocrine disruptor mechanism, which is characteristic of most toxins that are sex specific, is not considered to be involved (Clement et al. 2007). However, acrylamide has an affinity for sulfyhydryl groups, which are necessary for protein function. These sulfyhydryl groups are regenerated by glutathione. Both acrylamide and glycidamide are detoxified by glutathione (depicted in Figure 2, Pg 8) hence, glutathione depletion through acrylamide metabolism may further contribute to reduced protein function. Sulfyhydryl groups are also present in sperm flagellum, thus functionality and motility of spermatozoa may be compromised in males exposed to acrylamide (Tyl and Friedman 2003; Exon 2006).

Reproductive toxicity of acrylamide in rodent studies has also been suggested to be related to neurotoxic effects. For example, acrylamide neurotoxicity manifesting in the hind limbs of male rodents may influence copulatory/mounting behaviour, affecting the deposition of sperm in the uterus and reducing breeding competence (Sublet et al. 1989; Tyl et al. 2000b). On a molecular level, the effect of acrylamide on kinesins, as mentioned previously, may have implications for reproductive parameters. Chromosomal segregation during meiosis may be impaired by acrylamide inhibition of kinesin motors. This may produce abnormal genetic content in developing male germ cells. Few studies have investigated the effects of acrylamide on kinesins specific to spindle formation; although one paper by Sickles et al. (2007) has shown that both acrylamide and glycidamide inhibit the activity of two mitotic/meiotic kinesin motors, KIFC5A (which bundles microtubules for spindle formation) and KRP2 (involved in microtubule depolymerisation). Kinesin proteins are also necessary for the transport of molecules and reorganisation of microtubules during the dramatic remodelling of spermatozoa, thus, acrylamide reactivity with these proteins may impair the unique morphology of sperm that is inherent to sperm function.

14 However, a study by Ghanayem et al. (2005b) demonstrated that acrylamide reproductive toxicity is dependent on its metabolic conversion to glycidamide. Intraperitoneal injections of 0 - 50 mg acrylamide were administered to Cyp2e1-null and wild-type male mice, which were mated to unexposed females. Similar to previous studies, increases in embryo resorptions and reduced fertility rates were observed in females paired with acrylamide treated, wildtype males. Interestingly, these effects were not found in females mated with acrylamide treated Cyp2e1-null mice. Adler et al. (2000), also showed that the dominant lethal effects associated with acrylamide exposure in paternal males were blocked following pre-treatment with 1-aminobenzotriazole, an inhibitor of P450 metabolism. The results from both of these studies indicated that the metabolism of acrylamide to glycidamide by Cyp2e1 is essential in eliciting dominant lethality in mice. Because of the mutagenic potential of glycidamide, a favourable hypothesis was that activation of acrylamide to glycidamide by Cyp2e1 metabolism leads to genetic damage in spermatogenesis, that may have consequences for offspring and future generations. Indeed, DNA damage in sperm has been linked to male infertility and adverse effects on the health of offspring (Olsen et al. 2010). The molecular mechanisms by which acrylamide and/or glycidamide generate DNA damage is discussed later in this review, in section, 'DNA Damage in the Male Germ Line', Pg. 20.

15 Spermatogenesis

To assist in understanding the male reproductive toxicity of acrylamide, a brief revision of spermatogenesis is provided. Spermatogenesis occurs within the seminiferous tubules of the testes where spermatogonial stem cells divide and differentiate to produce spermatozoa. During this development, germ cells move from the basement membrane towards the tubular lumen where they are released as spermatozoa and transported to the epididymis (Figure 3, Pg. 17). This process is protected by the blood/testis barrier, which prevents the passage of cytotoxic substances from the circulating blood stream into the seminiferous tubules. The blood/testis barrier is formed by the tight and gap junctions of the Sertoli cells; which secrete various endocrine and paracrine substances and play a critical role in the nutrition, maintenance and organisation of spermatogenesis (Shalet 2009). The permeability of molecules to cross the blood/testis barrier is dependent on small molecular size and lipid solubility, thus the transport of the majority of foreign molecules and toxicants is restricted (Lee and Dixon 1978). Acrylamide however, has the capacity to cross this barrier and interfere with cells undergoing spermatogenesis (Marlowe et al., 1986).

The blood/testis barrier divides the seminiferous tubule into the basal and adluminal compartments. Spermatogonia reside in the basal compartment of the tubule, and as they differentiate, pass through the barrier into the adluminal compartment. Initially, spermatogonia divide mitotically in order to maintain a self-renewing population of stem cells. The products of these divisions generates Type A spermatogonia, which remain part of the stem cell pool, and Type B spermatogonia, which are committed to differentiation. The diploid Type B spermatogonial cells move into the adluminal compartment and become primary spermatocytes as they commence meiosis.

Figure 3: Spermatogenesis occurs in the seminiferous tubules of the testes. As germ cells develop, they move from the basement membrane to the lumen of the tubule. Spermatogonial cells divide mitotically and undergo meiosis during spermatogenesis to form primary spermatocytes, secondary spermatocytes and haploid spermatids. The flagella and acrosome cap develop and the nuclear chromatin is condensed to produce mature spermatozoa. Spermatozoa are released into the lumen to undergo subsequent maturation in the epididymis and travel through the vas deferens before they are ejaculated. Adapted from Raven et al. (2005).

17 During prophase, chromosomes condense and duplicate, homologous chromosomes are paired, and genetic recombination between sister chromatids takes place. Pairs of homologous chromosomes align along the spindle apparatus and segregate in the first meiotic division, after which cells are classified as secondary spermatocytes. These cells carry out the second meiotic division, in which no DNA replication occurs, to produce haploid spermatids. Upon the completion of meiosis, spermatids undergo spermiogenesis, which involves the formation of the characteristic flagellum, the development of the acrosome cap by the golgi apparatus, and condensation of the nuclear chromatin (Holstein et al. 2003). The majority of the cytoplasm is shed as a cytoplasmic droplet, yielding mature spermatozoa that enter the luminal space and travel to the epididymis (Figure 3, Pg 17). Spermatozoa gain functional competence within the epididymis and travel through the vas deferens, prior to ejaculation (Gilbert 2000). The differentiation and development from a spermatogonial cell to a mature spermatozoon takes around 64 days in humans and 42 days in mice. Note: More details of the localization of testicular cells within the seminiferous tubules at different stages of spermatogenesis are shown in supplementary data, sFig. 1, Pg. 135.

The condensation of chromatin during the spermatid to spermatozoan stages is a significant and complex event, which functions to protect the male genome during its transport to the oocyte. Chromatin condensation in male germ cells occurs through the replacement of histone proteins (responsible for DNA packaging in somatic cells) with protamines. This exchange causes the nuclear material in male germ cells to be condensed to one-tenth the volume of an immature spermatid (Holstein et al. 2003). However, due to this reorganisation of nuclear material, DNA repair mechanisms are progressively halted during late spermatogenesis; therefore, cells in the spermatid to spermatozoan stages of development are considered vulnerable to genetic damage. Indeed, numerous mutagens such as acrylamide have been found to be most effective in producing heritable genomic damage in post-meiotic male germ cells (Shelby 1996; Marchetti and Wyrobek 2005).

18 Acrylamide Genotoxicity in Late-stage Spermatogenesis

The effect of acrylamide at different stages of spermatogenesis can be studied in mouse breeding experiments by altering the interval between exposure and fertilisation (spermatozoa, 1–7 days; spermatids, 8–21 days; spermatocytes, 22–35 days; differentiating spermatogonia, 36–49 days; and spermatogonial stem cells, greater than 49). Several papers have demonstrated that acrylamide acts on post-meiotic male germ cells, and induces heritable damage in specific-locus mutation assays and heritable translocation tests (Reviewed in Favor & Shelby, 2005).

The specific-locus mutation assay involves exposure to a mutagenic agent in a mouse that is homozygous wildtype for a set of marker loci, and mated with an untreated mouse that is homozygous recessive for mutant alleles of the same loci. Obvious phenotypes are utilised as marker loci, such as coat colour and ear size; and if no mutation occurs, the heterozygous F1 generation will express the wildtype phenotype. If mutations are induced, the F1 generation will express the recessive phenotype. Russell et al. (1991) conducted a specific- locus test in male mice subjected to five consecutive daily injections of 50 mg/kg acrylamide, and mated with unexposed females at varying intervals. Significant increases in specific-locus mutation rates were observed at post-stem cell stages of spermatogenesis, with a maximum effect at late spermatid/spermatozoa stages. A similar specific-locus assay by Ehling & Neuhauser-Klaus (1992) also demonstrated that acrylamide exposure was effective in the first two weeks of mating, corresponding to late spermatid/spermatozoa stages.

Heritable translocation tests, which detect chromosomal damage in progeny, also indicate that acrylamide exposure primarily affects late stage spermatogenesis (Generoso et al. 1996; Adler et al. 2004). The basic premise of this test is that mutagenic agents may induce translocations between non-homologous chromosomes in the germ cells of the paternal male. These males are mated with unexposed females, and the resultant progeny are heterozygous carriers of reciprocal translocations. Meiotic segregation in these individuals produces gametes that have either a ‘balanced’ haploid set of chromosomes, or an ‘unbalanced’ set, with extra or missing chromosomal segments. Thus, translocation

19 heterozygotes are sterile (or semi-sterile) as fertilisation with unbalanced gametes leads to embryonic lethality. The heritable translocation test measures the frequency of these heterozygotes in the F1 generation, which is confirmed through cytogenetic assessment of gametes for chromosomal rearrangement (Favor and Shelby 2005). In a heritable translocation test by Shelby et al. (1987), acrylamide treatment of male mice at 50mg/kg bodyweight for five days produced high frequencies of translocation carriers. The increased frequency of heterozygotes was observed in matings 7 - 10 days following the last injection, thus supporting the argument that acrylamide affects post-meiotic germ cells. Interestingly, glycidamide treatment in mice produced similar results in the heritable translocation test, which also primarily affected late stage germ cells (Generoso et al. 1996).

DNA Damage in the Male Germ Line

The molecular mechanisms that generate specific-locus mutations, heritable translocations and dominant lethality in acrylamide exposed male rodents remain uncertain. However, it is clear from numerous studies that germ cells in late spermatid to early spermatozoa stages are most susceptible to acrylamide toxicity.

An alternative argument follows that the dominant lethality associated with acrylamide are the result of damage at the chromosomal level, rather than the mutagenic potential of glycidamide DNA adducts. Both acrylamide and glycidamide can cause chromosomal damage through alkylation of protamines, which are responsible for DNA condensation in spermatids and spermatozoa. This argument fits with the previous assertion that acrylamide affects late-stage spermatogenesis, due to the nature of the chromatin in post-meiotic germ cells. Indeed, significantly high levels of protamine-bound acrylamide have been found in the head of spermatozoa, extracted from male mice injected with 14C-labelled acrylamide at 125 mg/kg bodyweight (Sega et al. 1989). Adducts with protamines may promote physical stress in the chromatin structure of spermatozoa, causing double strand breaks in the paternal genome. Alternatively, protamine adducts have the potential to act as bulky adducts, which cannot be removed upon formation of the pro-nucleus following fertilisation (Marchetti and Wyrobek 2005).

A third possibility is that the metabolism of acrylamide in male germ cells by Cyp2e1 may lead to DNA damage through oxidative damage, resulting in reproductive toxicity. Oxidative damage is known to have vast implications in cells as the free radicals generated by oxidative stress react with many cellular processes. As shown in Figure 2 (Pg. 8), acrylamide and glycidamide detoxification involves conjugation with glutathione. Glutathione has a critical role in the deactivation of reactive-oxygen species, thus depletion of glutathione by acrylamide metabolism may lead to an increase in oxidative stress. Additionally, the metabolic activity of CYP2E1 has been implicated in producing free radicals and reactive oxygen species, particularly in its metabolism of ethanol (Cederbaum et al. 2009). In spermatozoa, there is significant evidence that indicates oxidative damage leads to defects in sperm motility, acrosomal development and fertilising ability (Aitken and Baker 2004). Furthermore, reactive oxygen species can also damage DNA through the formation of oxidative adducts.

Another mechanism by which acrylamide may damage the paternal genome is through an epigenetic effect. Epigenetics refers to the heritable, non-sequence based alterations to DNA that regulate gene expression. Genomic imprinting is one example of an epigenetic mechanism that leads to the variable expression of paternal or maternal genes in the embryo (O'Flynn O'Brien et al. 2010). Imprinting usually involves DNA methylation of cytosine residues at dinucleotide ‘cytosine phosphate guanine’ sites by methyltransferases. The pattern of methylation in sperm DNA undergoes many changes during its transit through the epididymis, although most of the methylation pattern is removed from both the maternal and paternal genome during embryogenesis. However, methylation is maintained on some genes (imprinting) which controls gene expression such that only one allele, inherited from either the mother or father, is expressed. Imprinted genes are usually critical to early development and there is some evidence that disruption of DNA methylation in the paternal gamete can lead to poor embryogenesis or genetic disorders (Emery and Carrell 2006; Trasler 2009). Epigenetic effects have been given little consideration in regards to acrylamide reproductive toxicity, however one paper reported that male rats orally exposed to acrylamide for two weeks produced abnormal genomic imprinting of insulin-like growth factor II (Igf2) in sperm (Wang et al. 2010). Thus, acrylamide exposure may affect embryo

21 development by interfering with the DNA imprinting pattern in male germ cells, prior to fertilisation.

Genetic damage induced in these cells by acrylamide may be related to a direct effect on DNA, potentially through glycidamide formation. Glycidamide forms two major DNA adducts with purine bases, guanine and adenine, known as N7-(2-carbamoyl-2-hydroxyethyl)guanine (N7-GA-Gua) and N3-(2-carbamoyl-2-hydroxyethyl)adenine (N3-GA-Ade) (Figure 4 Pg. 23). The presence of these glycidamide DNA adducts have been identified in various tissues following acrylamide exposure of mice such as in the lung, liver, kidney and testes (Ghanayem et al. 2005a; Gamboa da Costa et al. 2003). Formation of these adducts in male germ cells may produce mutations in the germ line, leading to dominant lethality and embryo resorptions. Genetic damage is likely to be reversible in early male germ cells, as DNA repair mechanisms are still intact; however post-meiotic cells may not have the capacity to repair genetic damage as effectively.

Figure 4: Glycidamide forms several DNA adducts including N7-(2-carbamoyl-2- hydroxyethyl)guanine (N7-GA-Gua; from the depurination of N7-(2-carbamoyl-2-hydroxyethyl)deoxyguanosine), N3-(2-carbamoyl-2-hydroxy-ethyl)adenine (N3-GA-Ade; from the depurination of N3-(2-carbamoyl-2-hydroxyethyl)deoxyadenosine), N1-(2-carboxy-2- hydroxyethyl)deoxyadenosine, N6-(2-carboxy-2-hydroxyethyl)deoxyadenosine, N1,N6-(2- hydroxypropanoyl)-deoxyadenosine, and N3,N4-(2-hydroxypropanoyl)-deoxycytidine. Glycidamide forms adducts with purine bases, guanine and adenine, resulting primarily in the formation of N7-GA-Gua and N3-GA-Ade (highlighted above). N7-GA-Gua is typically formed to a 100-fold greater extent than N3-GA-Ade. Acrylamide reacts slowly with DNA and forms the following adducts (these adducts have not been detected in experimental animals and thus not shown above); N1-(2-carboxyethyl)-deoxyadenosine, N3-(2- carboxyethyl)deoxycytidine, N7-(2-carbamoylethyl)guanine, (from the depurination of N7- (2-carbamoylethyl)deoxyguanosine), N6-(2-carboxyethyl)deoxyadenosine, and N1-(2- carboxy-ethyl)deoxyguanosine. Adapted from NTP (2012) and Klug & Cummings (2003).

23 An array of DNA repair mechanisms exist in cells that correct different types of genetic lesions, such as additions, deletions or base substitutions within the genetic coding. These lesions can arise through adduct formation, alkylation, oxidation and hydrolysis reactions, or through errors during replication. Nucleotide excision repair (NER) is one of the main mechanisms that repair a wide range of DNA helix distorting lesions in somatic cells (Jansen et al. 2001). In NER, the DNA lesion is recognized, incised and the oligonucleotide is replaced via DNA synthesis and ligation (Olsen et al. 2005). Intriguingly however, it has been reported that spermatogenic cells in rats and mice have inefficient or non-functional NER, possibly due to inaccessibility of DNA lesions in the condensed chromatin structure in male germ cells (Jansen et al. 2001; Andreassen et al. 2011). Thus, other repair pathways may be active in rodent spermatogenesis, such as base excision repair (BER), mismatch repair (MMR) or recombination repair (RR). In BER a modified base is first remove via a lesion specific DNA glycosylase, leaving a baseless site. This site is then incised (by the glycosylase, if bifunctional, or by an independent AP endonuclease). DNA polymerase enzymes subsequently repair the missing bases using the homologous strand as a template (Tichy and Stambrook 2008). In MMR, a gap is formed and filled by repair synthesis, followed by ligation; whereas in recombination repair, the DNA-ends are joined and ligated (which can be error-prone unless it is repaired by homologous recombination) (Baarends et al. 2000). Indeed, there is evidence that BER and MMR plays a role in spermatogenesis (Leduc et al. 2008).

One method of detecting DNA repair activity is through measuring unscheduled DNA synthesis, in non-replicating cells. Sega et al. (1990) investigated unscheduled DNA synthesis in the testes of mice injected with acrylamide, and 6 hours post acrylamide treatment, a peak in unscheduled DNA synthesis was detected at the mid-spermatid stage. This lag time between exposure and detection was considered to be the result of the required time to metabolise acrylamide to glycidamide, which would subsequently react with DNA and signal DNA repair.

An important question regarding a genotoxic mechanism of acrylamide/glycidamide is whether metabolic activation of acrylamide occurs in the liver, or whether this occurs in extrahepatic tissue, such as the testis. Circulation of glycidamide in the blood following

24 hepatic activation of acrylamide in the liver is less likely due to the relative instability of the epoxide molecule. Furthermore, radioactively [14C] labelled acrylamide is detected in testis tissue only 1 hour after administration (Marlowe et al. 1986), which suggests that acrylamide reaches testicular tissue in its unmetabolised form. Thus, in situ activation of acrylamide to glycidamide may contribute to DNA damage in the testis. However, the reproductive toxicity of acrylamide most likely affects many cellular pathways in male reproduction. It is hypothesised that DNA damage in late spermatogenesis leads to dominant lethality and decreased reproductive parameters, though several modes of action may be involved in eliciting this damage. Thus, the nature of DNA damage needs to be elucidated in order to determine the consequences of acrylamide exposure on male reproduction in humans.

25 Research Design

There has been much research conducted on the toxicity of acrylamide and the consequences of acrylamide exposure in rodents. However, questions remain regarding the effects of acrylamide on male reproduction; such as the genetic damage induced in the male germ line, and what mechanisms give rise to this damage. Additionally, it is uncertain whether the reproductive effects of acrylamide observed in animal experiments relate to the human situation, as the majority of studies examine acute doses that far exceed the estimated human level of exposure. Indeed, for most compounds, experimental toxicology studies in animals are almost always characterised by the problem of extrapolation from acute, short-term exposures, to realistic human levels. Thus this chronic experiment is a unique aspect of the current thesis. With this in mind, the overall aim of this research project was to provide a mechanistic understanding of acrylamide in an effort to improve risk assessment of acrylamide concentrations in the human diet. This was achieved by addressing the following three aims: 1. To examine the nature of genetic damage induced by acrylamide in isolated male germ cells. 2. To investigate whether chronic acrylamide exposure induces DNA damage in male germ cells in vivo. 3. To elucidate the molecular mechanisms by which this damage is generated and the response of the male germ line to chronic toxic exposure.

The above aims were addressed in the form of three prepared/published manuscripts presented in the following chapters. The first aim involved a series of acrylamide exposure experiments to observe the effect of acrylamide in isolated early male germ cells from mice. The second and third aims were approached using a long-term drinking water study in male mice that focused on the consequences of acrylamide exposure, at doses in range of human exposure levels. DNA damage assessment was conducted on germ cells obtained from these mice and microarray analyses were used to investigate testicular gene expression effects following prolonged exposure to acrylamide. To link the results obtained in these three studies, a brief overview precedes each manuscript. The contribution of this research to the current understanding of acrylamide toxicity is subsequently discussed in the final chapter of this thesis.

Chapter 2 Mouse Spermatocytes Express Cyp2e1 and Respond to Acrylamide Exposure

27 Chapter 2: Overview

The aim of the following manuscript was to examine the nature of genetic damage induced by acrylamide in isolated mouse male germ cells. Firstly however, it was important to determine whether mouse male germ cells express the acrylamide metabolising enzyme, Cyp2e1. Using Q-PCR and immunological techniques, it was established that Cyp2e1 was predominantly expressed at the spermatocyte germ cell stage. Furthermore, induction of Cyp2e1 mRNA expression was found in these cells following acrylamide exposure, an effect which had not previously been examined in male germ cells.

Spermatocytes were thus the focus of this study, and DNA damage induced following exposure to either acrylamide or glycidamide in these cells was examined. Damage was assessed using a version of the Comet assay, modified to detect DNA adducts. The current study builds upon previous work conducted by Hansen et al. (2010), in which the Comet assay was used to examine damage in dissociated mouse testicular cells at much higher doses of acrylamide and glycidamide for shorter incubation periods. Both acrylamide and glycidamide were found to increase levels of DNA damage in spermatocytes dose- dependently; however, a greater response was observed following glycidamide exposure. Further characterisation of this damage indicated that oxidative DNA damage may also partially contribute to the damage induced by glycidamide in isolated male germ cells.

These results demonstrated that male germ cells not only express Cyp2e1, but respond to acrylamide by upregulating Cyp2e1 gene expression. Additionally, in vitro exposure to acrylamide or its metabolite induced DNA damage in spermatocytes, which could be detected using the Comet assay. The results of this study therefore shed further light on acrylamide mediated damage in the male germ line and provided the foundation for a large-scale exposure study in mice.

Note: Supplementary data for the following manuscript is located in Appendix A, Pg. 134.

28 Mouse Spermatocytes Express CYP2E1 and Respond to Acrylamide Exposure

Belinda J. Nixon a, Aimee L. Katen a, Simone J. Stanger a, John E. Schjenken a, Brett Nixon a,b and Shaun D. Roman a,b,c

a Reproductive Science Group, School of Environmental and Life Sciences, University of Newcastle, Callaghan, New South Wales 2308, Australia

b Australian Research Council Centre of Excellence in Biotechnology and Development, School of Environmental and Life Sciences, University of Newcastle, Callaghan, New South Wales 2308, Australia

c Corresponding author: Dr. Shaun Roman Discipline of Biological Sciences ARC Centre of Excellence in Biotechnology & Development School of Environmental & Life Sciences University of Newcastle Callaghan, NSW 2308, Australia Phone: +61 2 4921 6818 Fax: +612 4921 6308. E-mail: [email protected]

29 Abstract

Metabolism of xenobiotics by cytochrome P450s (CYP) often leads to bio-activation, producing reactive metabolites that interfere with cellular processes and cause DNA damage. In the testes, DNA damage induced by xenobiotics has been associated with impaired spermatogenesis and adverse effects on reproductive health. We previously reported that chronic exposure to the reproductive toxicant, acrylamide, produced high levels of DNA damage in spermatocytes of Swiss mice. Thus, to investigate the mechanisms of acrylamide toxicity in mouse male germ cells, we examined the expression of the CYP, CYP2E1, which metabolises acrylamide. Using Q-PCR and immunohistochemistry, we establish that CYP2E1 is expressed in germ cells, predominantly in spermatocytes. Additionally, CYP2E1 gene expression was upregulated in these cells following in vitro acrylamide exposure (1 µM, 18 h). CYP2E1 metabolises acrylamide to glycidamide, which can form adducts with DNA. Isolated spermatocytes were therefore exposed to acrylamide (1 µM, 18 h) or glycidamide (0.5 µM, 18 h) and the presence of DNA-adducts was investigated using the comet assay, modified to detect DNA-adducts. Both compounds produced significant levels of DNA damage in spermatocytes, with a greater response observed following glycidamide exposure. A modified Comet assay indicated that direct adduction of DNA by glycidamide was a major source of DNA damage. Oxidative stress played a small role in eliciting this damage, as a relatively modest effect was found in a Comet assay modified to detect oxidative adducts following glycidamide exposure, and glutathione levels remained unchanged following treatment with either compound. Our results indicate that the male germ line has the capacity to respond to xenobiotic exposure by inducing enzyme expression, and the DNA damage elicited by acrylamide in male germ cells is likely due to the formation of glycidamide adducts.

Keywords acrylamide; glycidamide; cytochrome P450; DNA damage.

30 Introduction

Paternal exposure to environmental toxicants or xenobiotics has been associated with adverse reproductive effects such as impaired fertility, birth defects, miscarriages, and childhood genetic diseases [1]. The potential for xenobiotics to induce genetic damage in male germ cells is thought to be involved in mediating these reproductive effects, as the male germ line has limited capacity for DNA repair [2,3,4]. The genotoxic impact of xenobiotics may also be amplified (bio-activated) following their metabolism via enzymes, such as the CYPs. CYP2E1 is one of several CYPs known to cause bio-activation and metabolises a range of exogenous substances, including acrylamide [5,6]. Acrylamide is of particular interest, as traces of the compound have been detected in numerous carbohydrate-rich foods such as potato chips and breads [7,8]. Additionally, acrylamide is a known neurotoxicant in humans and acts as a carcinogen, genotoxin and reproductive toxicant in rodents [9]. Whilst the reproductive toxicity of acrylamide has not been observed in humans to date, there are concerns that chronic dietary exposure to the compound may have a cumulative effect on human fertility and reproductive health [10].

Interestingly, the reproductive toxicity of acrylamide primarily affects the male. Rodent studies have reported decreases in copulatory behaviour and a loss of spermatogenesis following exposure to acrylamide in males [11]. Multi-generational effects have also been described, such as the loss of post implantation embryos and reduced postnatal survival. In a study by Sakamoto and Hashimoto [12], high doses of acrylamide in the drinking water of male mice lead to decreases in fertility, reduced litter sizes, and increases in embryo resorptions. However, the mechanisms by which acrylamide elicits these reproductive effects are not clear. Based on the timing between exposure and effect (loss of embryos), sperm transit of the epididymis has been the major focus, with failure to fertilise attributed to protamine alkylation in maturing sperm [13,14]. Alternatively, a clastogenic mechanism may be involved, as the formation of kinesin adducts in the meiotic/mitotic spindles may impair chromosomal segregation in germ cells during spermatogenesis [15].

31 The above mechanisms require the interaction of acrylamide with an intermediary protein, as acrylamide does not directly react with DNA and preferentially forms adducts with cysteine residues of proteins [16]. However, the metabolism of acrylamide by Cyp2e1 generates the epoxide metabolite, glycidamide, which has a higher mutagenic potential than acrylamide and directly interacts with DNA, forming adducts [17]. Indeed, the presence of these adducts have been identified in the lung, liver, kidney and testes following acrylamide exposure in mice [18]. Studies by Ghanayem et al. [19,20,21] have also demonstrated the significance of the metabolic conversion of acrylamide to glycidamide in CYP2E1 knockout mice. The multigenerational effects were abrogated in CYP2E1-null males, indicating that the presence of CYP2E1 is essential in mediating acrylamide reproductive toxicity. It is therefore hypothesised that the reproductive effects of acrylamide are related to its conversion to glycidamide, which generates DNA adducts, leading to genetic damage in the male germ line.

We recently reported that chronic exposure of male mice to acrylamide at doses relevant to human exposure lead to significantly increased levels of DNA lesions in spermatocytes [22]. Given that acrylamide is metabolised by CYP2E1, the expression and regulation of this CYP was examined in the present study within specific stages of early male germ cell development. We show that spermatocytes express CYP2E1, indicating that they have the capacity to metabolise acrylamide to glycidamide. Furthermore, the DNA damage induced in spermatocytes by acrylamide or its metabolite, glycidamide, following in vitro exposure was examined using the comet assay, modified to detect DNA adducts. The results of the present study support the notion that DNA damage induced in meiotic germ cells by acrylamide is likely attributable to the formation of DNA adducts generated by glycidamide.

32 Materials and Methods

Animal Ethics Statement

Experiments involving animals were conducted in strict accordance with the policies set out by the Animal Care and Ethics Committee of the University of Newcastle (Ethics Numbers: SR1004 0708, A-2008-145). Swiss mice were housed under conditions of 16 hours light, 8 hours dark, with food and water provided ad libitum. Animals were euthanized by CO2¬ asphyxiation and all efforts were made to minimize suffering.

Chemicals and reagents

All chemicals and reagents including custom designed primers were obtained from Sigma Chemicals (St Louis, MO) unless otherwise stated, and were of molecular biology or research grade. Rabbit polyclonal anti-cytochrome p450 2E1 antibody (anti-CYP2E1, ab28146) was obtained from Abcam (Cambridge, MA). Mouse anti-cAMP dependent Protein Kinase [Catalytic subunit] antibody (anti-PKA[C], #610981) was purchased from BD Transduction Laboratories. Rat anti-germ cell nuclear antigen antibody (anti-GCNA) was a gift from Dr. George Enders [23]. Secondary antibodies, Alexa Fluor 594 goat anti-rabbit immunoglobulin G (IgG) (A11012) was purchased from Invitrogen (Carlsbad, CA). Dulbecco’s Modified Eagle Media (DMEM) and supplements for cell culture were obtained from Sigma and Invitrogen. Acrylamide was obtained from Sigma (≥ 99% purity, A9099) and glycidamide (98% purity, G615250) was obtained from Toronto Research Chemicals (North York, CA). Oligo(dT)15 primer, RNasin, dNTPs, M-MLV-Reverse Transcriptase, RQ1 DNase, GoTaq Flexi, MgCl2 and GoTaq quantitative PCR master mix were obtained from Promega (Madison, WI). DNA repair DNA glycosylases, formamidopyrimidine-DNA glycosylase (FPG) and 8-oxoguanine DNA glycosylase (hOGG1) were purchased from New England Biolabs Inc. (Arundel, Qld).

33 Germ cell isolation

Spermatogonia, spermatocytes and spermatids were enriched from dissected mouse testes using density sedimentation at unit gravity as described previously [24]. Briefly, testes were disassociated and tubules were digested sequentially with 0.5 mg/ml collegenase/DMEM and 0.5% v/v trypsin/EDTA to remove extra-tubular contents and interstitial cells. Remaining cells were loaded onto a 2 - 4% w/v bovine serum albumin (BSA)/DMEM gradient to separate male germ cell types according to density. Germ cell fractions were collected, washed and counted. This method enables the isolation of germ cells with very little to no somatic cell contamination, as extra-tubular cells are digested and removed prior to density sedimentation. Greater than 90% purity can be achieved for spermatogonial isolations, 65 - 70% purity for spermatocyte isolations, and 85 - 95% purity for spermatid isolations [24].

RNA extraction

Total RNA was isolated from male germ cells and whole testis using two rounds of a modified acid guanidinium thiocyanate-phenol chloroform protocol [25], in which cells and tissues were lysed with lysis buffer (4 M guanidinium thiocyanate, 25 mM sodium citrate, 0.5% v/v sarkosyl, 0.72% v/v β-mercaptoethanol). RNA was isolated by phenol/chloroform extraction and isopropanol precipitated.

Quantitative PCR (Q-PCR)

Reverse transcription was performed with 2 μg of isolated RNA, 500 ng oligo(dT)15 primer, 40 U of RNasin, 0.5 mM dNTPs, and 20 U of M-MLV-Reverse Transcriptase. Total RNA was DNase treated prior to reverse transcription to remove genomic DNA contamination. Q-PCR was performed using SYBR Green GoTaq qPCR master mix according to manufacturer's instructions on an MJ Opticon 2 (MJ Research, Reno, NV, USA). The sequences of all primers used in this study and the predicted size of the amplicons are provided in supplementary data (sTable 1). Reactions were performed on cDNA equivalent to 100 ng of total RNA and carried out for 40 amplification cycles. SYBR Green fluorescence was measured after the

34 extension step at the end of each amplification cycle and quantified using Opticon Monitor Analysis software Version 2.02 (MJ Research). Each sample was examined in triplicate and a replicate omitting the reverse transcription step was undertaken as a negative control. Real- time data were normalized to cyclophilin expression using the equation 2e-∆C(t), where C(t) is the cycle at which fluorescence was first detected above background fluorescence [26]. Real-time data were presented as the average of each replicate, normalized to each reference sample (±SEM). Further validation of Q-PCR data was conducted using the geometric mean of cyclophilin and a second reference gene, hypoxanthine-guanine phosphoribosyltransferase (hprt), in accordance with Vandesompele et al. [27] (supplementary data, sFig. 2). The 2-∆∆C(t) transformation [26] was used for comparisons between treated and vehicle control primary cultures. Values for each replicate were averaged, and relative expression levels between treated samples were depicted as percentages of controls (+SEM). Each data set is the average of at least three separate experiments.

Immunohistochemistry & Immunocytochemistry

Mouse testes were fixed in Bouin’s fixative, embedded in paraffin wax and sectioned at 5 μm thickness. Sections were de-paraffinized, rehydrated, and antigen retrieval was performed using Proteinase K (20 µg/ml) for 30 min at room temperature. Sections were blocked in 3% w/v BSA/phosphate buffered saline with 0.05% v/v Tween-20 (PBST) for 1 h at room temperature, after which they were incubated with anti-CYP2E1 (1:50 with 1% w/v BSA/PBST) overnight at 4°C. Sections were then washed and incubated with fluorescent- conjugated secondary antibody, Alexa Fluor 594 goat anti-rabbit IgG (1:200 with 1% w/v BSA/PBST), for 1 h at room temperature. Counterstaining was conducted using 4'-6- diamidino-2-phenylindole (DAPI) for 3 min. Sections were then mounted in Mowiol and observed under fluorescence on an Axio Imager A1 fluorescent microscope (Carl Zeiss MicroImaging, Inc., Thornwood, NY). As a control, parallel testis tissue sections were probed with rabbit serum in the absence of primary antibody, which did not produce a detectable signal (blank images not shown). Images were taken using an Olympus DP70 microscope camera (Olympus America, Centre Valley, PA).

35 Isolated spermatogonia, spermatocytes and spermatids were air dried onto a 12 well slide and blocked in 3% w/v BSA/PBST for 1 h at room temperature. Germ cells were dual-stained with anti-CYP2E1 (1:50 with 1% w/v BSA/PBST) and anti-GCNA (1:20 with 1% w/v BSA/PBST) or anti-PKA[C] (1:25 with 1% w/v BSA/PBST) for 1 h at room temperature. Anti-GCNA labels spermatogonia and spermatocytes whereas anti-PKA[C] is a marker for spermatids [23,28]. Cells were then washed and incubated with appropriate secondary antibodies for 1 h at room temperature. As a control, cells were also probed with rabbit serum in the absence of primary antibody, which did not produce a detectable signal (blank images not shown), and slides were subsequently viewed under fluorescence as previously described above.

Treatment of germ cells with chemicals

Isolated germ cells were suspended in 1 ml sterile DMEM (100 cells/µl), supplemented with 100 µM sodium pyruvate, 200 µM L-glutamate, 100 U/ml penicillin, 10 µg/ml streptomycin, and 5% v/v fetal bovine serum. Cells were treated with acrylamide or glycidamide at a final concentration between 5 nM- 10 µM and all samples were incubated for 18 h at 37°C, 5%

CO2. In a previous study, treatment of dissociated testicular cells with acrylamide or glycidamide was carried out at millimolar concentrations, for relatively short exposure times of 2 h (Hansen et al. 2010). Therefore, due to the length of exposure used in the current study (18 h), the concentration of acrylamide and glycidamide was adjusted to the 5 nM - 10 µM range to ensure cell viability remained largely unaffected (Fig. 3) whilst still eliciting a cellular response to chemical exposure in spermatocytes (Figs. 4 and 5). Negative control samples were treated with vehicle only (distilled water, dH2O). An additional control in which germ cells were treated at room temperature with hydrogen peroxide (H2O2) at a final concentration of 500 µM for 5 min was also included. Following treatment, cell suspensions were washed and collected by centrifugation prior to further analysis.

Trypan Blue exclusion and germ cell staining

Germ cells exposed to acrylamide were assessed for cell viability using trypan blue live/dead staining. Cells were stained with 0.08% v/v trypan blue and more than 200 cells per replicate

36 were scored using a haemocytometer. To observe germ cell morphology following acrylamide treatment, cells were fixed in 4% v/v paraformaldehyde and 1 x 104 cells were air dried onto a 12-well slide. Slides were blocked with 0.2% v/v Triton® X-100/PBS for 10 min at room temperature and cells were incubated with peanut lectin conjugated to fluorescein isothiocyanate (FITC-PNA) for 15 min at room temperature. FITC-PNA fluorescently labels the developing acrosome green and can be used to distinguish different testicular cell types. Cells were counterstained with propidium iodide (PI), mounted in Mowiol, and visualised using fluorescence microscopy as previously described above.

Determination of DNA damage in germ cells using the Comet assay

The extent of germ cell DNA damage elicited by exposure to acrylamide was assessed using an alkaline Comet assay according to methods published by [29] with modifications detailed below. Fully frosted Dakin slides (ProSciTech, Australia) were coated with one layer of 1% w/v normal melting point agarose. Ten μl of germ cells suspended in PBS (1 x 107 cells/ml) were mixed with 70 μl of 0.5% w/v low-melting point agarose (0.4% final agarose concentration), and this suspension was layered onto the pre-coated slides and covered with a coverslip. Once agarose was set, the coverslip was removed and cells were immersed in fresh lysis solution (2.5 M NaCl, 100 mM Na2EDTA, 10 mM Trizma and 1% v/v Triton X- 100 at pH 10) for 1 h at 4°C. Cells were incubated in lysis buffer with dithiothreitol (10 mM final concentration) for a further 30 min at 4°C, after which lithium diiododsalicyclate was added (4 mM final concentration) and cells were incubated for 1.5 h at room temperature (according to the Comet protocol for sperm, Comet Assay Interest Group site, accessed 24/11/2013, http://www.cometassay.com/index_files/Page290.htm). Following lysis, cells were treated with DNA glycoylases that recognise types of modified bases, formamidopyrimidine glycosylase (FPG) or 8-oxoguanine DNA glycosylase (hOGG1) at 1:1000 [0.4 units/gel] and 1:500 [0.16 units/gel] respectively, for 30 min at 37°C (enzyme optimisation data included in supplementary data, sFig. 3). Control cells were treated with enzyme buffer (40 mM HEPES, 0.1 M KCl, 0.5 mM EDTA, 0.2 mg/ml BSA, pH 8.0 with KOH) and all samples were incubated in chilled alkaline electrophoresis buffer (0.3 M NaOH, 1 mM EDTA) for 20 min. Electrophoresis was carried out for 5 min at 0.9 V/cm, 300 mA, after

37 which slides were drained and neutralised (0.4 M Tris, pH 7.5). Slides were stained with SYBR green (Trevigen, Gaithersburg, MD) and visualised under fluorescence. The DNA integrity of 50 - 100 cells per slide was analysed using Comet Assay IV software (Perceptive Instruments, Suffolk, UK). The fluorescence intensity in the comet “tail” was used as measure of DNA damage (Tail DNA %). Highly damaged cells with irregular or blown out nuclei, referred to as 'clouds' or 'hedgehogs', were excluded from analysis (as according to the 5th JaCVAM Initiative Comet Validation Meeting, Comet Assay Atlas, 2009). No statistically significant differences in the frequency of ‘hedgehog’ comets was found across treatments (supplementary data, sFig 4 - 5).

Quantification of cellular glutathione

Levels of glutathione (GSH) in germ cells and P19 embryonal carcinoma cells were quantified using GSH-GloTM Glutathione Assay kit (Promega) according to manufacturers’ instructions. Luminescence was analysed using a luminometer plate reader (FLUOstar Optima, BMG Labtech) and data presented is the average of three replicate experiments.

Statistics

Statistical analyses were performed using JMP software Version 9 (SAS Institute, Cary, NC). Data was tested for normality using the Shapiro-Wilk test. When data was not from the Gaussian distribution, the non-parametrical Kruskal-Wallis test was applied. If a statistically significant difference was found across groups of means, then a post-hoc Steel-Dwass multiple comparisons test was used to examine significant differences between pairs of groups. Differences between control and treated samples were considered to be statistically significant if the probability of the difference being due to chance was less than 5% (p < 0.05) and F statistic and degrees of freedom are indicated in parentheses in Figure captions. All experiments were replicated at least three times with independent samples and data are presented as the mean values +SEM.

38 Results

Expression of CYP2E1 in the male germ line

To characterise the expression of CYP2E1 in the germ line, gene expression was examined in whole mouse testis as a developmental time series, from 2 d after birth to adult (older than 56 d) (Fig. 1). Immature testis from 2 to 6 d exhibited limited expression of Cyp2e1. However, Cyp2e1 gene expression was maintained at relatively high levels between 11 to 18 d after birth, after which the level of Cyp2e1 was approximately halved and remained at this level through to adulthood. Examination of RNA from isolated germ cells by Q-PCR indicated that the highest levels of Cyp2e1 expression were in spermatogonia, compared to spermatocytes and spermatids. Spermatogonia are present in the testis after the migration of gonocytes to the basal membrane (4 - 6 d after birth) [30]. Thus, the high testicular expression of Cyp2e1 from day 11 to 18 (Fig. 1) is likely attributed to the relatively high proportion of spermatogonia that dominate the testicular environment during this early stage of development. In contrast, later stage germ cells including round and elongating spermatids, begin to develop in mouse testis at day 20 [29], and since neither of these cell types express high levels of Cyp2e1, the overall Cyp2e1 expression level diminishes in maturing testis.

Interestingly, Cyp2e1 appeared to be subject to translational repression. Immunohistochemistry revealed that the CYP2E1 protein was predominantly expressed in early pachytene spermatocytes, observed in testis from 14 d after birth (Fig. 2 A). Downregulation of CYP2E1 protein expression was observed in spermiogenic stages in 36 d old and adult testis (a diagram of germ cell stages of spermatogenesis in the testis is included in supplementary data, sFig. 1). Such differences in gene and protein expression are likely the result of translational delay and repression, commonly found during the process of spermatogenesis [31]. Further confirmation of CYP2E1 protein expression in spermatocytes was carried out using immunocytochemistry in isolated male germ cells (Fig. 2 B), suggesting that acrylamide metabolism to glycidamide is likely to occur in this germ cell type. Thus, as translation of CYP2E1 mRNA appeared to be delayed in spermtogonia and spermatids exhibited relatively little CYP2E1 protein expression, investigation into the

39 effects of acrylamide exposure was primarily focussed on spermatocytes in the present study.

Male germ cells respond to acrylamide by increasing CYP gene expression

One feature of cells responding to xenobiotic exposure is the induction of xenobiotic metabolising enzymes. Indeed, upregulation of numerous P450s in response to xenobiotic exposure has previously been observed in liver tissue of mice [38]. CYP genes, Cyp2e1 and Cyp1b1, were therefore examined in isolated spermatocytes following acrylamide exposure by Q-PCR, as preliminary data indicated that these genes were both constitutively expressed in these cells (data not shown). As previously mentioned, CYP2E1 specifically metabolises acrylamide; CYP1B1 however, is known to facilitate the metabolism of polycyclic hydrocarbons as well as endogenous compounds such as estradiol and retinoic acids [6,32,33]. As described in the Materials and Methods section, treatment of germ cells with acrylamide was carried out at 1 µM for 18 h as this level of exposure did not affect cell viability and morphology, which was confirmed by fluorescent cell staining and trypan blue exclusion (Figs. 3 A and B). Q-PCR analysis indicated that spermatocytes responded to acrylamide exposure by significantly increasing Cyp2e1 and Cyp1b1 gene expression by approximately 2.5 fold and 3.8 fold respectively, compared to controls (p < 0.05, Fig. 4).

In vitro exposure to acrylamide or glycidamide induces DNA adducts in male germ cells

The expression of CYP2E1 in spermatocytes suggests that acrylamide is metabolised to glycidamide in these cells and may lead to the generation of DNA adducts. Thus, the DNA integrity of spermatocytes was examined using the alkaline comet assay, following either acrylamide (1 µM for 18 h) or glycidamide treatment (0.5 µM for 18 h). A significant increase in DNA damage (Tail DNA %) was observed following glycidamide exposure (p < 0.05); however, only a modest effect was found in cells exposed to acrylamide (Fig. 5 B). The sensitivity of the comet assay was therefore enhanced by utilising the DNA glycosylase FPG, which recognises and introduces strand breaks at sites of DNA adducts. The inclusion of FPG led to greater detection of DNA damage in both acrylamide and glycidamide treated cells (p

40 < 0.001), with slightly higher levels of damage observed in the glycidamide treated sample (Fig. 5 B).

Acrylamide and glycidamide were also found to elicit a dose-dependent increase in DNA damage in the comet assay, in the presence of FPG (Fig. 5 C, D). Statistically significant increases were observed at doses of 100 nM, 1 µM and 10 µM of acrylamide after 18 hours of exposure in spermatocytes (p < 0.001). Spermatocytes appeared to be more sensitive to glycidamide treatment, with significant increases in damage observed at exposures as low as 5 nM (p < 0.01). H2O2 treatment (500 µM, 5 min) was used as a positive control in these experiments and generated significant increases in damage in both the presence and absence of FPG (Fig. 5 B, C, D).

Oxidative adducts contribute little to the DNA damage induced by acrylamide in male germ cells

Exposure to acrylamide or glycidamide may lead to the generation of oxidative adducts as both compounds can be conjugated with GSH, which plays a critical role in anti-oxidant defence. Indeed, the FPG DNA glycosylase recognises a range of DNA lesions, including oxidative adducts; thus the DNA damage induced by acrylamide or glycidamide in spermatocytes was further characterised using an alternate cleavage enzyme, hOGG1, in the comet assay. The hOGG1 enzyme specifically recognises oxidative DNA adducts, such as 8- oxo-7,8-dihydroguanine (8-oxoGua) which is an adduct induced by reactive oxygen species [34]. The use of hOGG1 in the comet assay did not produce a significant increase in Tail DNA % in spermatocytes treated with acrylamide (1 µM, 18 h); although a modest increase was observed following treatment with glycidamide (0.5 µM, 18 h). These results suggest that oxidative adducts represent a fraction of the damage induced by the acrylamide metabolite, glycidamide; in spermatocytes however, acrylamide does not directly generate oxidative DNA damage in these cells.

Since both acrylamide and its metabolite, glycidamide, can be conjugated with glutathione (GSH) GSH levels were measured in spermatocytes treated with acrylamide using an established GSH assay (Fig. 6 B). Intriguingly, spermatocytes have relatively low basal levels

41 of cellular GSH compared to that of a control cell line (P19 embryonal carcinoma cells). No significant differences in GSH levels were observed after exposure of spermatocytes to acrylamide or glycidamide. These data indicated that antioxidant levels are constitutively low in spermatocytes and that oxidative damage is unlikely to be a major consequence of acrylamide or glycidamide exposure in these cells.

Discussion

While steroidogenic enzymes such as Cyp17a1 have previously been found in the germ line [35], our results are the first to demonstrate that CYPs involved in xenobiotic metabolism are also expressed at specific stages of meiotic male germ cell differentiation. Male germ cells were found to express both CYP2E1 mRNA and protein (Figs. 1 and 2). Previously, CYP2E1 protein was reported to be exclusively expressed in the interstitial cells of the testis [36]. However, in the present study, immunohistochemistry of mouse testis (Fig. 2 A) and isolated populations of germ cells (Fig. 2 B) clearly revealed protein expression of CYP2E1 in spermatocytes. The elevated expression of CYP2E1 in spermatocytes compared to other germ cells types may serve as increased cellular defence against chromosomal damage during this stage of germ cell development, as spermatocytes are entering into meiosis. However, while the presence of CYP enzymes in the male germ line may confer additional protection, it also comes with attendant risks. The generation of reactive metabolites as a product of detoxification processes may induce both cellular and DNA damage. In male germ cells, DNA damage is of particular importance as xenobiotic exposure may contribute to reduced fertility and impact on the health of future progeny [4].

The specificity of CYP2E1 for acrylamide enabled us to examine the role of one particular metabolic enzyme in the male germ line in response to xenobiotic exposure. In the present study, we utilised a relatively low level of acrylamide exposure (1 µM, 18 h), that was not cytotoxic to isolated spermatocytes (Fig. 3), but sufficient to elicit a pronounced cellular response (Fig. 4). Spermatocytes were found to respond to acrylamide exposure by increasing mRNA expression not only of Cyp2e1, but also Cyp1b1. While CYP1B1 is not specifically involved in the metabolism of acrylamide, members of both the CYP1 and CYP2 families of CYPs are primarily involved in metabolism of xenobiotics and are typically

42 induced as part of the toxic response [37]. Thus, it is possible that the toxic response triggered by acrylamide exposure increases expression of multiple genes involved in xenobiotic metabolism in spermatocytes. Indeed, regulation of various CYPs has been observed in the liver of acrylamide exposed mice [38]. However, to our knowledge, the present study is the first to observe such a response in the male germ line.

The quantitative increase in CYP2E1 mRNA, observed in acrylamide treated spermatocytes suggested that CYP2E1 is upregulated in germ cells at the transcriptional level, through either transcriptional activation or via inhibition of mRNA degradation. Interestingly, several studies have demonstrated that the elevation of CYP2E1 protein levels by exogenous substrates in the liver is mediated largely through protein stabilisation, with little to no change in mRNA levels [39]. Conversely, other studies have indicated that induction of CYP2E1 gene expression does occur in extrahepatic tissues, such as kidney, intestine and lung [40,41] which is similar to our observations in early male germ cells (Fig. 4). However, none of the aforementioned studies examined CYP2E1 induction in response to acrylamide, and induction of CYP2E1 in male germ cells has not previously been explored. The results of the current study indicate that male germ cells do respond to xenobiotic exposure and that specific mechanisms of CYP regulation are active in the mouse male germ line. Additionally, our results suggest that spermatocytes have the capacity to upregulate metabolism of acrylamide to glycidamide.

DNA damage in spermatocytes was investigated using the comet assay, which measures DNA strand breaks as opposed to adduct formation. Hence, cleavage enzymes that recognise specific DNA adducts were also utilised. The cleavage enzyme, FPG, recognises sites of 8-oxoguanine, 8-oxoadenine, fapy-guanine, methy-fapy-guanine, fapy-adenine, aflatoxin B1-fapy-guanine, 5-hydroxy-cytosine and 5-hydroxy-uracil. The major DNA adducts that glycidamide forms are N7-(2-carbamoyl-2-hydroxyethyl)guanine (N7-GA-Gua) and N3- (2-carbamoyl-2-hydroxyethyl)adenine (N3-GA-Ade) [19]. Indeed, FPG has been used in previous studies in the comet assay to recognise these glycidamide adducts and introduce strand breaks at these adduct sites [42,43]. In the absence of FPG, a significant increase in DNA damage was induced in spermatocytes treated with glycidamide (p < 0.05, Fig. 5 B). This may be due to depurination of glycidamide adducts, resulting in single or double strand breaks which are detectable by the comet assay [44]. However, upon the addition of FPG,

43 the level of DNA damage detected more than doubled for both acrylamide and glycidamide treated cells (Fig. 5 B) and was found to increase dose-dependently (Fig. 5 C, D). From this, it could be inferred that acrylamide induces DNA damage in spermatocytes via adducts, rather than DNA strand breaks, and may be indicative of the presence of glycidamide-DNA adducts as exposure to glycidamide in these cells produced similar increases in damage.

In contrast, a previous study by Hansen et al. [42] found that two hours of acrylamide exposure at millimolar concentrations did not induce DNA damage in dissociated testicular cells as detected by the FPG modified comet assay. These doses were 100 fold greater than the doses used in our study and thus the differences in exposure times between the two experiments may be indicative of the required time for cells to metabolise acrylamide to glycidamide. As detailed above, germ cells respond by inducing greater levels of CYP2E1 mRNA. Therefore the longer exposure time in our experiments (18 h) may also lead to enhanced levels of CYP2E1, potentially upregulating the metabolic conversion of acrylamide to glycidamide. Indeed, Sega et al. [45] reported that a maximum peak in DNA repair activity in mouse testis was observed six hours post acrylamide treatment, which was considered to be related to the period of time needed for acrylamide metabolism to occur.

As previously mentioned, the damage detected by FPG in the comet assay could be indicative of oxidative adducts as well as glycidamide adducts as FPG has the capacity to recognise several types of DNA lesions, including 8-oxoGua. Exposure to either acrylamide or glycidamide may deplete glutathione levels and subsequently result in increased oxidative stress. Indeed, oxidative stress is known to contribute to poor sperm function and generate DNA damage in sperm [46]. However, cellular GSH levels were unaffected by either acrylamide or glycidamide in spermatocytes (Fig. 6 B). Further characterisation of oxidative DNA damage using hOGG1 failed to identify significant increases in the acrylamide treated spermatocytes (Fig. 6 A), and only a modest increase was found in glycidamide treated cells (22% Tail DNA, p < 0.001). Hence, the DNA damage induced by acrylamide and glycidamide in spermatocytes are likely due to the presence of glycidamide adducts, and a minor contribution of oxidative damage.

DNA damage is likely to be reversible in early stage germ cells, as DNA repair mechanisms are still intact [1]. However, post-meiotic cells may not have the capacity to repair DNA

44 damage as effectively. The blood/testis barrier prevents the passage of most toxic substances through the seminiferous tubule towards the residing post-meiotic germ cells. Acrylamide however has been found to reach the testes in an un-metabolised state and is able to transit the blood/testis barrier, due to its low molecular weight and hydrophilicity [47]. Spermatogonia lie outside the blood/testis barrier (see supplementary data, sFig. 1) and have functional repair mechanisms to deal with DNA damage. In contrast, spermatocytes, which lie within the blood/testis barrier, are in the process of undergoing meiosis and DNA damage generated at this stage would be expected to be of significant consequence. In a study by Olsen et al. [48], spermatocyte exposure to benzopyrene produced DNA damage that was retained throughout spermatogenesis, resulting in adducts present in the spermatozoa. Thus, it is highly possible that the DNA damage induced in spermatocytes by acrylamide will persist through to mature spermatozoa in mice exposed to acrylamide.

Induction of CYP gene expression in response to xenobiotic exposure in male germ cells represents an additional mechanism of cellular defence that has not been extensively examined in the germ line in previous studies. Having shown that spermatocytes are the specific germ cell type that expresses CYP2E1, it is interesting to note that in our previous study [22], chronic exposure to acrylamide in male mice also produced high levels of DNA damage in these cells. In the present study we also show that the presence of DNA adducts in spermatocytes following acrylamide exposure are likely due to glycidamide, rather than oxidative adducts. Indeed, the upregulation of Cyp2e1 expression in spermatocytes exposed to acrylamide may in turn upregulate the metabolism of acrylamide to glycidamide, leading to an increase in glycidamide-DNA adducts. These results provide further evidence of the consequences of acrylamide exposure during spermatogenesis and sheds light on the mechanisms of xenobiotic metabolism present in early male germ cells.

Acknowledgements

We thank Prof. Eileen McLaughlin for her constructive criticism of this manuscript.

45 References

1. Marchetti F, Wyrobek AJ (2005) Mechanisms and consequences of paternally-transmitted chromosomal abnormalities. Birth Defects Res C Embryo Today 75: 112-129.

2. Jansen J, Olsen AK, Wiger R, Naegeli H, de Boer P, et al. (2001) Nucleotide excision repair in rat male germ cells: low level of repair in intact cells contrasts with high dual incision activity in vitro. Nucleic Acids Res 29: 1791-1800.

3. Olsen AK, Bjortuft H, Wiger R, Holme J, Seeberg E, et al. (2001) Highly efficient base excision repair (BER) in human and rat male germ cells. Nucleic Acids Res 29: 1781-1790.

4. Olsen AK, Duale N, Bjoras M, Larsen CT, Wiger R, et al. (2003) Limited repair of 8-hydroxy- 7,8-dihydroguanine residues in human testicular cells. Nucleic Acids Res 31: 1351-1363.

5. Guengerich FP (2008) Cytochrome p450 and chemical toxicology. Chem Res Toxicol 21: 70-83.

6. Nebert DW, Russell DW (2002) Clinical importance of the cytochromes P450. Lancet 360: 1155-1162.

7. Tareke E, Rydberg P, Karlsson P, Eriksson S, Tornqvist M (2002) Analysis of acrylamide, a carcinogen formed in heated foodstuffs. J Agric Food Chem 50: 4998-5006.

8. WHO (2011) Evaluation of certain contaminants in food. World Health Organization technical report series: 1-105, back cover.

9. Shipp A, Lawrence G, Gentry R, McDonald T, Bartow H, et al. (2006) Acrylamide: review of toxicity data and dose-response analyses for cancer and noncancer effects. Crit Rev Toxicol 36: 481-608.

10. Exon JH (2006) A review of the toxicology of acrylamide. J Toxicol Environ Health B Crit Rev 9: 397-412.

11. Tyl RW, Friedman MA (2003) Effects of acrylamide on rodent reproductive performance. Reprod Toxicol 17: 1-13.

12. Sakamoto J, Hashimoto K (1986) Reproductive toxicity of acrylamide and related compounds in mice--effects on fertility and sperm morphology. Arch Toxicol 59: 201-205.

13. Sega GA (1991) Adducts in sperm protamine and DNA vs. mutation frequency. Prog Clin Biol Res 372: 521-530.

14. Sega GA, Alcota RP, Tancongco CP, Brimer PA (1989) Acrylamide binding to the DNA and protamine of spermiogenic stages in the mouse and its relationship to genetic damage. Mutat Res 216: 221-230.

15. Sickles DW, Sperry AO, Testino A, Friedman M (2007) Acrylamide effects on kinesin- related proteins of the mitotic/meiotic spindle. Toxicol Appl Pharmacol 222: 111-121.

46 16. Lopachin RM, Decaprio AP (2005) Protein adduct formation as a molecular mechanism in neurotoxicity. Toxicological sciences : an official journal of the Society of Toxicology 86: 214- 225.

17. Sumner SC, Fennell TR, Moore TA, Chanas B, Gonzalez F, et al. (1999) Role of cytochrome P450 2E1 in the metabolism of acrylamide and acrylonitrile in mice. Chem Res Toxicol 12: 1110-1116.

18. Gamboa da Costa G, Churchwell MI, Hamilton LP, Von Tungeln LS, Beland FA, et al. (2003) DNA adduct formation from acrylamide via conversion to glycidamide in adult and neonatal mice. Chem Res Toxicol 16: 1328-1337.

19. Ghanayem BI, McDaniel LP, Churchwell MI, Twaddle NC, Snyder R, et al. (2005) Role of CYP2E1 in the epoxidation of acrylamide to glycidamide and formation of DNA and hemoglobin adducts. Toxicol Sci 88: 311-318.

20. Ghanayem BI, Witt KL, El-Hadri L, Hoffler U, Kissling GE, et al. (2005) Comparison of germ cell mutagenicity in male CYP2E1-null and wild-type mice treated with acrylamide: evidence supporting a glycidamide-mediated effect. Biol Reprod 72: 157-163.

21. Ghanayem BI, Witt KL, Kissling GE, Tice RR, Recio L (2005) Absence of acrylamide- induced genotoxicity in CYP2E1-null mice: evidence consistent with a glycidamide-mediated effect. Mutat Res 578: 284-297.

22. Nixon BJ, Stanger SJ, Nixon B, Roman SD (2012) Chronic exposure to acrylamide induces DNA damage in male germ cells of mice. Toxicological sciences : an official journal of the Society of Toxicology 129: 135-145.

23. Enders GC, May JJ, 2nd (1994) Developmentally regulated expression of a mouse germ cell nuclear antigen examined from embryonic day 11 to adult in male and female mice. Dev Biol 163: 331-340.

24. Baleato RM, Aitken RJ, Roman SD (2005) Vitamin A regulation of BMP4 expression in the male germ line. Dev Biol 286: 78-90.

25. Chomczynski P, Sacchi N (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156-159.

26. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25: 402-408.

27. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, et al. (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome biology 3: RESEARCH0034.

28. San Agustin JT, Witman GB (2001) Differential expression of the C(s) and Calpha1 isoforms of the catalytic subunit of cyclic 3',5'-adenosine monophosphate-dependent protein kinase testicular cells. Biology of reproduction 65: 151-164.

29. Singh NP, McCoy MT, Tice RR, Schneider EL (1988) A simple technique for quantitation of low levels of DNA damage in individual cells. Exp Cell Res 175: 184-191.

47 30. Russell LD, Ettlin RA, SinaHikim AP, Clegg ED (1990) Histological and Histopathological Evaluation of the Testis. 1st Ed., pp. 1-20 Cache River Press in Clearwater, FI. U.S.A.

31. Kleene KC (1996) Patterns of translational regulation in the mammalian testis. Mol Reprod Dev 43: 268-281.

32. Seliskar M, Rozman D (2007) Mammalian cytochromes P450--importance of tissue specificity. Biochim Biophys Acta 1770: 458-466.

33. Tsuchiya Y, Nakajima M, Kyo S, Kanaya T, Inoue M, et al. (2004) Human CYP1B1 is regulated by estradiol via estrogen receptor. Cancer Res 64: 3119-3125.

34. Smith CC, O'Donovan MR, Martin EA (2006) hOGG1 recognizes oxidative damage using the comet assay with greater specificity than FPG or ENDOIII. Mutagenesis 21: 185-190.

35. Liu Y, Yao ZX, Bendavid C, Borgmeyer C, Han Z, et al. (2005) Haploinsufficiency of cytochrome P450 17alpha-hydroxylase/17,20 lyase (CYP17) causes infertility in male mice. Mol Endocrinol 19: 2380-2389.

36. Healy LN, Pluta LJ, Recio L (1999) Expression and distribution of cytochrome P450 2E1 in B6C3F1 mouse liver and testes. Chemico-biological interactions 121: 199-207.

37. Gonzalez FJ, Kimura S (1999) Role of gene knockout mice in understanding the mechanisms of chemical toxicity and carcinogenesis. Cancer Lett 143: 199-204.

38. Mei N, Guo L, Tseng J, Dial SL, Liao W, et al. (2008) Gene expression changes associated with xenobiotic metabolism pathways in mice exposed to acrylamide. Environ Mol Mutagen.

39. Cederbaum AI, Lu Y, Wu D (2009) Role of oxidative stress in alcohol-induced liver injury. Arch Toxicol 83: 519-548.

40. Amacher DE (2010) The effects of cytochrome P450 induction by xenobiotics on endobiotic metabolism in pre-clinical safety studies. Toxicol Mech Methods 20: 159-166.

41. Pavek P, Dvorak Z (2008) Xenobiotic-induced transcriptional regulation of xenobiotic metabolizing enzymes of the cytochrome P450 superfamily in human extrahepatic tissues. Curr Drug Metab 9: 129-143.

42. Hansen SH, Olsen AK, Soderlund EJ, Brunborg G (2010) In vitro investigations of glycidamide-induced DNA lesions in mouse male germ cells and in mouse and human lymphocytes. Mutat Res 696: 55-61.

43. Thielen S, Baum M, Hoffmann M, Loeppky RN, Eisenbrand G (2006) Genotoxicity of glycidamide in comparison to (+/-)-anti-benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide and alpha-acetoxy-N-nitroso-diethanolamine in human blood and in mammalian V79-cells. Mol Nutr Food Res 50: 430-436.

44. Gates KS, Nooner T, Dutta S (2004) Biologically relevant chemical reactions of N7- alkylguanine residues in DNA. Chem Res Toxicol 17: 839-856.

45. Sega GA, Generoso EE, Brimer PA (1990) Acrylamide exposure induces a delayed unscheduled DNA synthesis in germ cells of male mice that is correlated with the temporal

48 pattern of adduct formation in testis DNA. Environmental and Molecular Mutagenesis 16 137-142.

46. De Iuliis GN, Thomson LK, Mitchell LA, Finnie JM, Koppers AJ, et al. (2009) DNA damage in human spermatozoa is highly correlated with the efficiency of chromatin remodeling and the formation of 8-hydroxy-2'-deoxyguanosine, a marker of oxidative stress. Biology of reproduction 81: 517-524.

47. Marlowe C, Clark MJ, Mast RW, Friedman MA, Waddell WJ (1986) The distribution of [14C]acrylamide in male and pregnant Swiss-Webster mice studied by whole-body autoradiography. Toxicol Appl Pharmacol 86: 457-465.

48. Olsen AK, Andreassen A, Singh R, Wiger R, Duale N, et al. (2010) Environmental exposure of the mouse germ line: DNA adducts in spermatozoa and formation of de novo mutations during spermatogenesis. PloS one 5: e11349.

49 Figures

(Note: Supplementary data for this manuscript is located in Appendix A, Pg. 135)

Figure 1

FIG. 1. Gene expression of CYP2E1 in the male germ line. Q-PCR analysis of CYP2E1 mRNA expression in mouse testis at different developmental stages 2, 6, 11, 14, 18, 22, and 36 d after birth, and adult (older than 56 d). Expression was also examined in isolated male germ cells, spermatogonia, spermatocytes and spermatids. Data are representative of n = 3 experiments and depicted as transformed values, 2e-∆C(t) (Mean +SEM), as described in Materials and Methods. CYP2E1 gene expression was found predominantly in spermatogonia. Statistically significant differences were found in spermatogonia, 11, 14 and

18 d compared to 2 d testis (F10,79 = 4.0,* p < 0.05).

50 Figure 2

FIG.2. Protein expression of CYP2E1 in the male germ line. (A) Immunolocalisation of CYP2E1 (red staining) in testis sections at different developmental stages, 11 d after birth, 22 d after birth, 36 d after birth, and adult (older than 56 d). Sections were sequentially probed with anti-CYP2E1 and appropriate secondary antibody before being counter-stained with DAPI (blue staining). CYP2E1 protein expression was found in spermatocytes at 22 d after birth to adult testis, with weaker staining observed in spermatids. Diagram outlining where different male germ cell types reside in the seminiferous tubule is shown in supplementary data. (B) Immunolocalisation of CYP2E1 in isolated spermatogonia, spermatocytes and spermatids, showing CYP2E1 expression (green staining) in spermatocytes. Germ cells were probed with anti-CYP2E1 and anti-GCNA, which labels spermatogonia and spermatocytes, or anti-PKA[C], which labels spermatids (red staining). Both tissue sections and cells were probed with rabbit serum in the absence of primary antibody as a control, which did not produce a detectable signal (blank images not shown). Scale bars equal to 50 µm.

51 Figure 3

FIG. 3. Acrylamide or glycidamide treatment (1 µM, 18 h) did not impact on spermatocyte morphology or viability. (A) Adult mouse spermatocytes were treated with acrylamide (1µM 18 h) or glycidamide (0.5 µM 18 h) and dual stained with FITC-PNA, which labels the developing acrosome (green), and PI (red) to observe cell morphology. Scale bar is equal to 50 µm. (B) The viability of spermatocytes treated with acrylamide (1µM 18 h) or glycidamide (0.5 µM 18 h) assessed by trypan blue exclusion. Data are representative of n=3 experiments, measured in triplicate (Mean, +SEM), and > 200 cells were scored per replicate. At the doses used in the current study, no differences in cell morphology or viability were observed following treatment with acrylamide or glycidamide.

52 Figure 4

FIG. 4. Acrylamide exposure elicits an increase in CYP gene expression in the male germ line. Gene expression levels were analysed in spermatocytes by Q-PCR following incubation with 1 µM acrylamide for 18 h. CYP2E1 and CYP1B1 gene expression was significantly increased in spermatocytes following acrylamide exposure (F1,16 = 20.1, * p < 0.05). Data are depicted as transformed values (2e-∆∆C(t)) as described in Materials and Methods, and is representative of n = 3 experiments (Mean +SEM).

53 Figure 5

FIG. 5. Acrylamide and glycidamide induces DNA damage in the male germ line. DNA damage was assessed in spermatocytes treated with acrylamide or glycidamide using the Comet assay in the presence or absence of FPG. (A) Representative comet images from control, acrylamide (1µM 18 h) or glycidamide (0.5 µM 18 h) and H2O2 (500 µM, 5 min) treated spermatocytes. (B) The average Tail DNA % was assessed for each sample and in the absence of FPG, a modest increase in Tail DNA % was observed in spermatocytes treated with glycidamide (F7,663 = 61.7, * p < 0.05). In the presence of FPG however, both acrylamide and glycidamide produced significant increases in Tail DNA % (*** p < 0.001) with a greater

54 response observed following glycidamide exposure. Treatment of spermatocytes with H2O2 (500 µM, 5 min) was used as a positive control for damage, and induced significant increases in Tail DNA % in both the presence and absence of FPG. (C) Spermatocytes were assessed for DNA damage using the FPG Comet assay following acrylamide (10 nM to 10 µM for 18 h) or (D) glycidamide (5 nM to 5 µM for 18 h) exposure. Spermatocytes treated with H2O2 (500 µM, 5 min) were used as a positive control for DNA damage. Significant increases in Tail DNA % were observed in spermatocytes following 100 nM acrylamide treatment and above

(F9,856 = 51.8, ** p < 0.01), and 5 nM glycidamide treatment and above (*** p < 0.001).

Significant increases in Tail DNA % were also observed in cells treated with H2O2 (*** p < 0.001). All data are representative of n = 3 experiments (Mean +SEM).

55 Figure 6

FIG. 6. Oxidative stress may play a role in the DNA damage induced by acrylamide in spermatocytes. (A) The Comet assay was conducted on spermatocytes treated with acrylamide (1 µM, 18 h) or glycidamide (0.5 µM, 18 h) in the presence or absence of hOGG1. Significant levels of DNA damage were not detected in the acrylamide treated spermatocytes in either the presence or absence of hOGG1. A modest but significant increase in Tail DNA % was observed in spermatocytes following glycidamide treatment in the absence of hOGG1 (F7,677 = 134.4, * p < 0.001); however, hOGG1 treatment resulted in greater detection of DNA following glycidamide exposure. Treatment with H2O2 (500 µM, 5 min) was used as a positive control and produced significant increases in Tail DNA % in the presence of either enzyme (*** p < 0.001). (B) Glutathione (GSH) levels in spermatocytes were measured using a GSH assay following acrylamide (1 µM, 18 h) or glycidamide (0.5 µM, 18 h). No significant differences in GSH levels in male germ cells were observed following acrylamide or glycidamide treatment. Additionally, relatively low GSH levels were found in spermatocytes (0.19 µM) compared to P19 embryonal carcinoma cells (2.14 µM,

F3,6 = 1676.2, *** p < 0.001). Data are representative of n = 3 experiments (Average +SEM).

Chapter 3 Chronic Exposure to Acrylamide Induces DNA Damage in Male Germ Cells of Mice

57 Chapter 3: Overview

The second major aim of this research thesis was to investigate whether chronic acrylamide exposure could lead to the generation of genetic damage in male germ cells in vivo. To address this, a long-term chronic exposure experiment was conducted over one year. Acrylamide was administered via the drinking water to adult male Swiss mice at concentrations of 0, 0.001, 0.01, 0.1, 1 and 10 µg/ml and time points 1, 3, 6, 9 and 12 months were measured. The dose range used in this experiment were well below exposures that have been used in previous rodent studies and encompassed human dietary exposure estimates of 0.5 µg/kg bodyweight/day (equivalent to the 0.01 µg/ml dose).

The following published manuscript details the experimental design and recorded monitoring data regarding mouse bodyweight, water consumption and the calculated administered dose. However, the major findings reported in this manuscript concerned the DNA damage assessment that was conducted at each time point. In the previous chapter, the FPG modified Comet assay was demonstrated to be useful in detecting acrylamide induced DNA damage in isolated spermatocytes; thus this technique was utilized to assess levels of DNA damage in spermatocytes from mice chronically exposed to acrylamide. Additionally, the presence of double strand breaks was assessed in these cells using a meiotic spread method and gamma H2A.X immunolocalisation in testis sections. Significant increases in DNA damage were observed in male germ cells from acrylamide exposed mice after 6 months of chronic exposure at high doses. This damage increased dose and time dependently; and following 12 months of exposure, increased DNA damage in male germ cells was detectable at doses relevant to human levels of exposure.

This study was the first chronic exposure model to be established in mice and demonstrated that long term exposure to low-doses of acrylamide could indeed induce DNA damage in the male germ line.

Note: Supplementary data is located in Appendix B, Pg. 140.

58 Chronic Exposure to Acrylamide Induces DNA Damage in Male Germ Cells of Mice

Belinda J. Nixon,* Simone J. Stanger,* Brett Nixon,*,† and Shaun D. Roman*,†,1

*Reproductive Science Group, School of Environmental and Life Sciences, University of Newcastle, Callaghan, New South Wales 2308, Australia; and †Australian Research Council Centre of Excellence in Biotechnology and Development, School of Environmental and Life Sciences, University of Newcastle, Callaghan, New South Wales 2308, Australia

1To whom correspondence should be addressed at Discipline of Biological Sciences, Australian Research Council Centre of Excellence in Biotechnology and Development, School of Environmental and Life Sciences, University of Newcastle, Callaghan, New South Wales 2308, Australia. Fax: +612 4921 6308.

E-mail: [email protected].

Toxicological sciences 129(1), 135–145 (2012) doi:10.1093/toxsci/kfs178

Received March 22, 2012; accepted May 8, 2012

59 Abstract

Acrylamide is a reproductive toxicant that has been detected in foods such as potato chips and breads. The consequences of chronic exposure to acrylamide in the human diet are unknown; however, rodent experiments have shown that acute acrylamide exposure in males can lead to decreased fertility and dominant lethality. One of the possible mechanisms by which acrylamide elicits these effects is thought to be related to its metabolic conversion to glycidamide, which can form DNA adducts. To determine whether chronic acrylamide exposure produces genetic damage in male germ cells in vivo, male mice were subjected to acrylamide through their drinking water. Acrylamide was administered at 0.001, 0.01, 0.1, 1, and 10 μg/ml for up to 1 year, which was equivalent to 0.0001–2 mg/kg bodyweight/day. At 1, 3, 6, 9, and 12 months, early male germ cells were assessed for DNA damage using a Comet assay modified to detect adducts and γH2A.X expression, a marker of double-strand breaks. Acrylamide treatment did not significantly affect mouse or testis weight, and no gross morphological effects were observed in the testis. However, a significant dose-dependent increase in DNA damage was observed in germ cells following 6 months of exposure in the two highest dosage groups (1 and 10 μg/ml). After 12 months of exposure, increases in damage were detected at doses as low as 0.01 μg/ml (0.001 mg/kg bodyweight/day). The results of this study are the first to demonstrate that chronic exposure to acrylamide, at doses equivalent to human exposures, generates DNA damage in male germ cells of mice.

Keywords

Acrylamide, glycidamide, chronic, reproductive toxicity, spermatogenesis, DNA damage

60 Introduction

For decades acrylamide has been used for the production of polyacrylamide, which has various industrial applications in water treatment, cosmetics, soil conditioning, and biomolecular laboratories for use in gel electrophoresis (Parzefall, 2008). Exposure to acrylamide was previously thought to occur through industrial activities or through cigarette smoking, as the compound is also found in tobacco smoke. It is now known that acrylamide forms in numerous baked or fried carbohydrate-rich foods, particularly in plant-based products such as potato chips, crisps, breads, and coffee (Tareke et al., 2002). These findings heightened concerns regarding widespread human exposure, as acrylamide is a known neurotoxicant in humans. Additionally, animal studies have demonstrated that acrylamide also acts as a genotoxin, carcinogen, and reproductive toxin; however, to date, epidemiological studies have not established a causative relationship between these effects and acrylamide exposure in humans (FAO/WHO, 2003).

The reproductive effects of acrylamide have been observed in numerous rodent studies, with acute doses resulting in reduced fertility, embryo implantation losses, and reduced postnatal survival (Tyl and Friedman, 2003). However, limited data are available regarding the consequences of long-term, low-level acrylamide exposure on reproductive health at doses relevant to the human situation. According to the 72nd report on Food additives by the Food and Agriculture Organization and World Health Organization, the average dietary intake of acrylamide was estimated to be 1 μg/kg bodyweight/day, or 4 μg/kg bodyweight/day for high consumers (WHO, 2011). Although these estimates generally do not account for acrylamide exposure via drinking water or cosmetics, this level of exposure is considerably lower than doses known to induce acrylamide toxicity in animal studies. For example, the no observable adverse effect level (NOAEL) for acrylamide reproductive toxicity is considered to be 2–5 mg/kg bodyweight/day, depending on the endpoint of fertility or embryonic death (Exon, 2006). Hence, dietary acrylamide exposure is considered unlikely to induce reproductive toxicity in humans.

However, there are concerns that long-term exposure to the compound may have cumulative effects. Indeed the neurotoxicity of acrylamide, characterized by hind-limb foot splay, ataxia, and skeletal muscle weakness, develops progressively. Following long

61 exposure periods (60 days to 2 years), low doses of acrylamide have been documented to produce neurotoxic effects comparable to those observed in animals exposed at acute doses for shorter durations (10–30 days) (Lopachin and Gavin, 2008). Chronic carcinogenic studies have been carried out in rats over 2 years at 0.1–2 mg/kg bodyweight/day, in which high doses resulted in increased tumor incidences in certain tissues (Friedman et al., 1995; Johnson et al., 1986). To our knowledge, however, the reproductive toxicity of acrylamide has not been examined at doses less than 0.5 mg/kg bodyweight/day, nor for more than 6 months of exposure (Erkekoglu and Baydar, 2010; Tyl and Friedman, 2003).

Evidence from previous studies indicates that reproductive toxicity of acrylamide primarily manifests in the male. In a study by Sakamoto and Hashimoto (1986), acrylamide exposure in male mice (19 mg/kg bodyweight/day for 8 days) mated to unexposed females lead to increased incidences of embryo resorptions and reduced fertility rates. Other effects of acrylamide on male reproduction include decreased reproductive behaviour, testicular atrophy, abnormal spermatogenesis, and reduced sperm quality (Hashimoto et al., 1981; Sublet et al., 1989; Wise et al., 1995; Yang et al., 2005). The mechanisms by which acrylamide elicits these effects have not been fully elucidated. However, studies in cyp2e1 knockout mice indicate that the cytochrome P450, Cyp2e1, may be essential in mediating acrylamide reproductive toxicity (Ghanayem et al., 2005b,c). CYP2E1 specifically metabolizes acrylamide to the epoxide, glycidamide, which can form DNA adducts and cause DNA damage. DNA repair mechanisms are limited in male germ cells during late spermatogenesis (Jansen et al., 2001; Olsen et al., 2005); hence, unresolved genetic damage in these cells may impact on downstream fertilization and the health of future progeny.

Although levels of acrylamide in the human diet may not induce overt reproductive toxicity, chronic exposure could generate DNA damage in the male germ line due to the indirect genotoxicity of acrylamide via glycidamide. Thus, we conducted a chronic 12-month exposure experiment in which acrylamide was administered to male mice at a low-level range of 0.1 μg–2 mg/kg bodyweight/day via the drinking water. Genetic lesions in the form of DNA adducts and double-strand breaks (DSBs) were assessed in enriched spermatocytes from exposed animals using a modified version of the Comet assay and a meiotic spread method. In this study, we report an accumulation of genetic damage in male germ cells of mice following 6–12 months of exposure, at doses relevant to human exposure.

62 Materials and Methods

Chemicals and reagents

All chemicals and reagents were obtained from Sigma (St Louis, MO) unless otherwise stated and were of molecular biology or research grade. Dulbecco’s modified Eagle’s media (DMEM), supplemented with 1% v/v L-glutamine 200mM, 1% v/v sodium pyruvate 100mM, 2% v/v HEPES solution 1M, and 1% v/v penicillin streptomycin solution were obtained from Sigma and Invitrogen (Carlsbad, CA). Mini complete protease inhibitor cocktail tablets were purchased from Roche (Mannheim, Germany), and Halt phosphatase inhibitor cocktail was purchased from ThermoFisher Scientific (Rockford). DNA-repair endonuclease, formamidopyrimidine-DNA glycosylase (FPG) was purchased from New England Biolabs Inc. (Arundel, QLD). Primary antibodies, rabbit polyclonal anti-SCP3 (anti-SCP3, ab15093) and mouse monoclonal anti-γ H2A.X (phospho S139) (anti-γ H2A.X, ab18311), were obtained from Abcam (Cambridge, MA). Alexa Fluor 594 goat anti-rabbit (A11012) and Alexa Fluor 488 goat anti-mouse (A11001) immunoglobulin G (IgG) were purchased from Invitrogen.

Animals

Experimental procedures involving animals were conducted in accordance with the policies set out by the Animal Care and Ethics Committee of the University of Newcastle (Ethics Number: SR A-2009-121). Male Swiss mice were 5–6 months old at the beginning of the study and ranged in weight from 31 to 55 g. All animals were housed under controlled temperature and humidity conditions, 16 h light, 8 h dark, with food and water provided ad libitum. Animals were acclimated for at least 1 month prior to treatment.

Experimental design

One hundred and eighty male mice were randomly distributed into 30 different treatment groups (six doses of acrylamide at five different time points). Six males were assigned to each group to ensure three replicates for each experimental endpoint (Fig. 1). Testes from three males were used for germ cell isolation, and the testes from the remaining three males were used for either tissue sectioning (in which testes were fixed in Bouin’s fixative, embedded in paraffin wax, and sectioned at 5 μm thickness) or protein extraction. Up to six individuals were housed in one cage, and all mice were monitored for signs of neurotoxicity

63 or other adverse effects for the duration of treatment. Mouse bodyweight (Table 1) and water consumption (Supplementary table 1) were recorded every second day. It was established after 3 months that treatment did not significantly affect water consumption or bodyweight, and these parameters were monitored on a weekly basis for the remainder of the experiment. These records were used to calculate the average daily acrylamide dose and total cumulative dose administered to each treatment group (Table 2).

Acrylamide treatment

Acrylamide drinking water solutions were prepared using research-grade acrylamide obtained from Sigma (≥ 99% purity, A9099) diluted in filtered, deionized water at concentrations 0.001, 0.01, 0.1, 1, and 10 μg/ml. Drinking water containing acrylamide was prepared fresh on a weekly basis, as the compound is stable in water for at least 1 week (Mei et al., 2008; Wang et al., 2010), and was substituted for normal drinking water at the commencement of treatment. Test animals were exposed to acrylamide via the drinking water for a period of 1 month (4 weeks exactly), 3 months (12 weeks ± 1 week), 6 months (24 weeks ± 1 week), 9 months (36 weeks ± 1 week), or 12 months (48 weeks ± 1 week). Scheduled culling time points were staggered over a 3-week period (hence ± 1 week) for logistical purposes. However, this staggering was deemed inappropriate for the 1-month time point, as this treatment was only a 4-week period. Therefore, the timing of the 1- month time point was co-ordinated specifically to achieve exactly 4 weeks of exposure.

Immediately after treatment, animals were euthanized by CO2 asphyxiation, and tissues were collected.

Early germ cell isolation

Testes collected from all mice were weighed and spermatocytes were extracted using density sedimentation as described previously (Baleato et al., 2005). Briefly, testes were disassociated and seminiferous tubules were digested sequentially with 0.5 mg/ml collagenase/DMEM and 0.5% trypsin/EDTA to remove extratubular contents and interstitial cells. Remaining cells were loaded onto a 2–4% bovine serum albumin (BSA)/DMEM gradient to separate male germ cell types according to density. Germ cell fractions were collected, washed, and counted. This method enables the enrichment of germ cells with very little to no somatic cell contamination, as extratubular cells are digested and removed

64 prior to density sedimentation. A purity of 65–70% can be obtained for pachytene spermatocyte isolations, with contaminating cells consisting of other germ cells, mostly early spermatocytes (Baleato et al., 2005).

Comet assay

DNA damage in isolated spermatocytes was measured using an alkaline Comet assay according to methods published by Singh et al. (1988) with modifications as follows. Isolated germ cells suspended in PBS (1 × 107 cells/ml) were mixed with 0.5% low–melting point agarose and layered onto fully frosted Dakin slides (ProSciTech, Australia), precoated with 1% normal melting point agarose. Agarose was allowed to solidify at 4°C for 30 min and were immersed in fresh lysis solution (2.5M NaCl, 100mM Na2EDTA, 10mM Trizma, and 1% Triton X-100, pH 10) for 1 h at 4°C. Cells were incubated in lysis buffer with dithiothreitol (10mM final concentration) for a further 30 min at 4°C, after which lithium diiododsalicylate was added (4mM final concentration) and incubated for 1.5 h at room temperature. Following lysis, cells were treated with adduct specific cleavage enzyme, FPG, for 30 min at 37°C. All samples were incubated in chilled alkaline electrophoresis buffer (1mM EDTA, 0.3M NaOH) for 20 min. Electrophoresis was carried out for 5 min (25 V, 300 mA), and slides were drained and neutralized (0.4M Tris, pH 7.5). Slides were stained with 20 μg/ml ethidium bromide and visualized under fluorescence. The DNA integrity of 50–100 cells per slide was analyzed using Comet Assay IV software (Perceptive Instruments, Suffolk, U.K.). The fluorescence intensity in the comet “tail” was used as measure of DNA damage (tail DNA%). Mean Tail DNA was expressed as the percentage increase in damage relative to control samples to enable comparison of data between measurements across all time points (Fig. 3). Highly damaged cells with irregular or blown out nuclei, also referred to as “clouds” or “hedgehogs,” were excluded from analysis.

Immunohistochemistry

Mouse testes were fixed in Bouin’s fixative, embedded in paraffin wax, and sectioned at 5- μm thickness. Sections were deparaffinised and rehydrated, and antigen retrieval was performed by microwaving sections for 3 × 3 min in 50mM Tris-HCl (pH 10). Sections were blocked in 3% BSA/PBS with 0.05% Tween-20 (PBST) for 1 h at room temperature, after which they were incubated with anti-γH2A.X (1:500 with 1% BSA/PBST) overnight at 4°C.

65 Sections were washed and incubated with fluorescent conjugated secondary antibody, Alexa Fluor 594 goat anti-rabbit IgG (1:200 with 1% BSA/PBST) for 1 h at room temperature. Counterstaining was conducted using the DNA-specific f -diamidino-2- phenylindole (DAPI) for 3 min. Slides were mounted in Mowiol and observed under fluorescence on an Axio Imager A1 fluorescence microscope (Carl Zeiss MicroImaging Inc., Thornwood, NY). Images were taken using an Olympus DP70 microscope camera (Olympus America, Center Valley, PA).

Meiotic spreads

DNA damage in the form of DSBs was further examined in spermatocytes from mice subjected to 12 months of acrylamide exposure. In order for all samples at the 12-month time point to be run concurrently, isolated spermatocytes were stored at −80°C, in PBS containing phosphatase inhibitor cocktail, protease inhibitor cocktail, and 10% dimethyl sulfoxide. Cells were thawed and resuspended in filtered 0.1M sucrose with fresh protease inhibitor cocktail, and the cell suspension was immediately dropped onto Superfrost microscope slides (ThermoFisher Scientific), precoated with an even layer of 1% paraformaldehyde and 0.1% Triton X-100. Slides were air-dried for 2 h, washed in 0.4% Photo-Flo Solution (Kodak Professional, NY), and air-dried. Immunostaining was carried out by first washing slides with buffer (3% BSA in PBS/0.015% Triton-X 100 [PBS-TX]) followed by blocking in antibody diluent (10% goat serum, 3%BSA, PBS-TX) for 5 min. Slides were incubated with primary antibodies, anti-γH2A.X (1:500) and anti-SCP3 (1:500), for 1 h at room temperature, washed in buffer, and reblocked. Appropriate secondary antibodies were added to slides for 1.5 h at room temperature followed by washing in PBS-TX. Slides were mounted in Mowiol, observed under fluorescence as described above, and the number of γH2A.X foci (green staining; Figs. 5A and B) were counted per cell, for 50–100 cells per slide. Anti-γH2A.X labeling of asynapsed sex chromosomes in spermatocytes was excluded from foci counts.

Statistics

Statistical analyses were performed using JMP software Version 9 (SAS Institute, Cary, NC), and detailed statistical output has been included in supplementary data. All body weight, testis weight, water consumption, DNA damage, and meiotic spread data were tested for

66 normality using the Shapiro-Wilk test. When data were not from the Gaussian distribution, the nonparametrical Kruskal-Wallis test was applied for each independent variable separately (acrylamide dose or period of exposure). If a statistically significant difference was found across groups of means, then a post hoc Steel-Dwass multiple comparisons test was used to examine specific significant differences between pairs of groups. Using a two- way mixed model ANOVA, comparisons between acrylamide dose (factor A), period of exposure (factor B), and the interaction between these factors (AB) were also carried out. Additionally, a power analysis was conducted to evaluate the sensitivity of the study to detect the smallest differences between groups. A 5% rejection index of null hypothesis was applied to all tests performed.

Results

Mouse Body Weight, Water Consumption, and Calculated Acrylamide Dose

The initial goal of this study was to establish a chronic exposure mouse model. As such, it was important to establish the validity of our intervention and to assess the broader consequences for the mice. Individual body weight (g) and water consumption (ml) were monitored for each mouse every second day for the first 3 months of the study. After 3 months of acrylamide exposure, no effects on bodyweight and water consumption were apparent; hence, monitoring of these parameters was carried out on a weekly basis. Table 1 includes the mean bodyweight, the percentage weight difference, and standard deviation (SD) of bodyweight for each treatment group (six mice per group). Significant increases in mouse body weight were found between mice of the 1-month time point compared with mice of other time points (p < 0.05, shown in Supplementary table 3), which was attributable to normal growth of the mice during the study. More importantly, however, the concentration of acrylamide in the drinking water had no significant effect on mouse bodyweight, indicating chronic exposure was not having severe adverse effects on the general well-being of the mice.

Water consumption was monitored throughout the study to ensure the presence of acrylamide in the drinking water did not impact on drinking levels. The mean daily water consumption was recorded per cage and converted to per mouse values. The average water consumption values ranged from 5.4 to 9.7 ml per day (Supplementary table 1), and a

67 modest increase in the levels of daily water consumption was found for the 1- and 3-month time points compared with all other time points (p < 0.05, Supplementary table 4). However, acrylamide exposure through the drinking water did not significantly affect mouse water consumption, as drinking levels of acrylamide-treated mice were not significantly different from control mice.

The average daily dose (mg/kg bodyweight/day) and total cumulative dose (mg/kg bodyweight) of acrylamide administered to mice were calculated based on water consumption and mouse bodyweight values (Table 2). Animals treated with the same concentrations of acrylamide in the drinking water received similar daily doses of acrylamide across all time points. Mice on the 0.01 μg/ml dosing regimen received between 0.001 and 0.002 mg/kg bodyweight/day that is equivalent to the estimated acrylamide intake in humans of 0.001 mg/kg bodyweight/day (WHO, 2011). The 10 μg/ml treatment group at 3 months received the highest daily dose of 2 mg/kg bodyweight/day. This value borders on the NOAEL for acrylamide reproductive toxicity (2–5 mg/kg bodyweight/day). At the two highest doses of 1 and 10 μg/ml, mice received between 0.14 and 2 mg/kg bodyweight/day; and a total cumulative dose of up to 520.1 mg/kg bodyweight. These exposures were in range of the NOAEL and lowest observed adverse effect level for acrylamide neurotoxicity; 0.2–0.5 mg/kg bodyweight/day and 2 mg/kg bodyweight/day, respectively (Lopachin and Gavin, 2008). Although testing of neurotoxic defects was not carried out during this experiment; based on visual inspection, no overt effects of neurotoxicity such as hind-limb foot splay were observed in test animals.

Testis to Bodyweight Ratio and Testis Histology

Chronic acrylamide exposure in male mice did not impact severely on testicular weight and histology. At each time point, animals were euthanized, and testes were recovered and weighed. The testis to bodyweight ratio was determined using final mouse bodyweights, and the average testis/bodyweight ratio of each treatment group (six individuals) was calculated. No significant differences in testis/bodyweight ratios were found across different acrylamide doses or across different time points (Supplementary table 2). Testes were also fixed, sectioned, and hematoxylin and eosin stained in order to examine testis histology (Fig.

68 2). No signs of gross morphological changes were found in the testes, even at the highest 10 μg/ml acrylamide dose after 12 months of exposure.

Chronic Acrylamide Exposure In Vivo Generates DNA Damage in Male Germ Cells

Early male germ cells were isolated from the testes of acrylamide-treated mice and assessed for DNA damage. The comet assay was used with the addition of FPG, a restriction enzyme that has previously been used in the comet assay to detect and cleave sites of glycidamide- DNA adducts (Hansen et al., 2010; Thielen et al., 2006). Significant increases in DNA damage (Tail DNA%) in the FPG-modified comet assay were first observed in mouse spermatocytes following 6 months of chronic acrylamide treatment, at doses of 1 and 10 μg/ml (Fig. 3). Similar increases were also found in subsequent 9 and 12-month time points, with doses as low as 0.01 μg/ml exhibiting significantly increased levels of DNA damage after 12 months (p < 0.05). The statistical relationship between acrylamide exposure and its effect on Tail DNA was verified across all comet data using a two-way mixed model ANOVA (Supplementary table 5). Statistical modeling revealed that the factor of exposure time and the interaction of time and acrylamide dose had a significant, positive relationship on germ cell Tail DNA (p < 0.05). Therefore, the results of the comet assay were indicative of a statistically significant, causal relationship between chronic acrylamide exposure in male mice and early germ cell DNA damage.

Sites of DSBs in meiotic cells are associated with surrounding chromatin modifications, such as phosphorylation of histone H2A.X, referred to as γH2A.X (Mahadevaiah et al., 2001). Investigation of DNA damage was, therefore, carried out using immunohistochemistry on mouse testis sections probed for γH2A.X at the 6-, 9-, and 12-month time points (red staining; Fig. 4). Basal expression of γH2A.X in spermatocytes could be observed in control samples; however, γH2A.X staining was more prominent in testes from mice exposed to acrylamide (0.1, 1, and 10 μg/ml), particularly at the two highest doses tested. In these testis sections, γH2A.X expression was predominantly localized to spermatocytes, with modest staining observed in later stage germ cells.

To confirm the presence of DSBs in early male germ cells, isolated spermatocytes from the 12-month time point were further examined using a meiotic spread method. Spermatocytes were spread and fixed onto a microscope slide and visualized using antibodies: anti-γH2A.X

69 (green staining) and anti-SCP3, which labels condensed chromosomes (red staining) (Figs. 5A and B). DSBs in spermatocytes were quantified by counting the average number of γH2A.X foci per cell (Fig. 5C). It should be noted that anti-γH2A.X also labels the asynapsed sex chromosomes, as γH2A.X modifications persist in the sex chromosome chromatin throughout meiotic prophase (Mahadevaiah et al., 2001). Staining of the sex chromosomes (shown in Figs. 5A and B) was not included in γH2A.X foci counts (Figs. 5C and D).

Male germ cells sustain a number of DSBs during meiosis, due to normal homologous recombination or gene conversion events (Matulis and Handel, 2006). Therefore, early male germ cells typically exhibit some level of γH2A.X staining that could be observed in control samples of Figures 4 and 5. However, significant increases in the average number of γH2A.X foci were found in spermatocytes from mice exposed to acrylamide for 12 months at all doses (Fig. 5C, p < 0.05). Cells that had more than 50 foci (counted separately, Fig. 5D) were also found to be more prevalent in spermatocytes from acrylamide-treated mice although this was not statistically significant. Thus, although a percentage of the observed number of γH2A.X foci are likely representative of normal meiotic recombination, the data presented in Figures 4 and 5 indicated that acrylamide exposure in male mice leads to an increased frequency of DSBs in early male germ cells.

Discussion

The present study is the first to examine the effect of long term acrylamide exposure in the male, at doses in the range of human exposure levels. Due to the length of the experiment, it was important to establish that mouse weight gain and overall health remained unaffected, particularly in the 10 μg/ml treatment group, which was in range of neurotoxic levels (1.4 – 2 mg/kg bodyweight/day; Table 2). Typical signs associated with acrylamide neurotoxicity such as hind-limb foot splay or muscle weakness have been observed in laboratory animals at repeated daily doses of 0.5–50 mg/kg bodyweight/day (Exon, 2006). Overt symptoms of acrylamide neurotoxicity were not observed in mice in the current study, though animals were not examined for underlying effects such as peripheral nerve degeneration, which has been reported in rats at doses as low as 0.67 mg/kg bodyweight/day over 2 years (WHO, 2011).

70 In regards to reproductive toxicity, the dosing regimen used was lower than that previously found to affect male reproduction in rodents (Tyl and Friedman, 2003). Thus, effects on fertility rates and dominant lethality, frequently observed at exposures greater than 5 mg/kg bodyweight/day, would not be induced at these levels. However, a more immediate endpoint of DNA damage in male germ cells was the primary focus of this study, as genetic damage in the male germ line could have adverse consequences for the health and development of offspring. Indeed, paternal exposure to acrylamide at high doses (50 mg/kg bodyweight/day) has been found to produce chromosomally abnormal zygotes and two-cell embryos in mouse breeding studies (Marchetti et al., 2009; Marchetti and Wyrobek, 2005). Levels of DNA damage were found to increase in spermatocytes of acrylamide-exposed animals in both a dose- and time dependent manner, first observed after 6 months of acrylamide exposure in the FPG-modified Comet assay (Tail DNA%; Fig. 3). More importantly, significant increases in DNA damage were apparent by 9 months in the 0.01 μg/ml group that received a daily acrylamide dose equivalent to human dietary estimates (1 μg/kg bodyweight/day).

The mechanisms by which acrylamide exerts genotoxicity and DNA damage have not been fully elucidated, though several means of action have been proposed that mostly relate to the adduct chemistry of acrylamide or its metabolite, glycidamide. Acrylamide is described as a soft electrophile that preferentially forms adducts by Michael additions with soft nucleophiles, typically sites of cysteine thiol groups on proteins and glutathione in biological systems (Lopachin and Decaprio, 2005). Indeed, both acrylamide and glycidamide can cause chromosomal damage through alkylation of protamines, which are necessary proteins in DNA condensation in spermatids and spermatozoa (Sega et al., 1989). Alternatively, the conjugation of glutathione by acrylamide or glycidamide may deplete cellular glutathione levels essential for the deactivation of reactive oxygen species that can damage DNA. Furthermore, the oxidation of acrylamide to glycidamide may also induce oxidative stress, as the activity of CYP2E1 can lead to the production of free radicals and reactive oxygen species (Cederbaum, 2006).

However, the metabolism of acrylamide to glycidamide by CYP2E1 may be the most significant mechanism of action as the epoxide, glycidamide, is a hard electrophile that reacts with hard nucleophilic sites, such as the oxygen atoms of purines and pyrimidines

71 (Lopachin and Decaprio, 2005). Although acrylamide itself reacts slowly with DNA, glycidamide on the other hand directly interacts with DNA and typically produces adducts with guanine, N7-(2-carbamoyl-2-hydroxyethyl)guanine, and to a lesser extent adenine adducts, N3-(2-carbamoyl-2-hydroxyethyl)adenine (Ghanayem et al., 2005a). Indeed, glycidamide-DNA adducts have been detected via liquid chromatography tandem mass spectrometry in various tissues following acrylamide exposure in rodents (Gamboa da Costa et al., 2003; Watzek et al., 2012).

The alkaline Comet assay, which was used in the current study, measures DNA strand breaks and alkali-labile sites, a form of genetic damage transformed into strand breaks under alkaline conditions (Singh, 2000). In vitro studies in mouse testicular cells and V79 cell line cells have demonstrated that the addition of the restriction enzyme, FPG, enhances the sensitivity and specificity of the Comet assay to detect damage following exposure to the acrylamide metabolite, glycidamide (Baum et al., 2008; Hansen et al., 2010; Thielen et al., 2006). Mice were not exposed to glycidamide in this experiment, as the aim of the study was to simulate chronic acrylamide exposure in humans. However, glycidamide adducts, generated from endogenous metabolism of acrylamide to glycidamide, would be detected using FPG in conjunction with the Comet assay. As stated earlier, acrylamide is unlikely to directly react with DNA, thus it could be said that the DNA damage observed in spermatocytes in Figure 3 is representative of glycidamide-related DNA lesions induced at doses of acrylamide comparable to human exposure levels.

Expression of γH2A.X was also investigated in testis and spermatocytes of acrylamide- exposed mice. The assembly of γH2A.X is thought to have an essential role in the recruitment of DNA-repair proteins and has been used as a marker of DSBs, as it spreads over several megabases from the strand break site (Matulis and Handel, 2006). Immunohistochemistry of testis sections revealed that overall γH2A.X expression was elevated in acrylamide-treated animals at the 6-, 9-, and 12-month time point (Fig. 4). Interestingly, this effect appeared to be more pronounced at the higher dose treatments of 1 and 10 μg/ml, which implied a dose-dependent effect. The presence of DSBs was further explored in spermatocytes using meiotic spreads at the 12-month time point. Meiotic spreads have previously been used to study DSBs following γ irradiation in isolated mouse spermatocytes by preparing cell spreads and immunostaining for γH2A.X (Matulis and

72 Handel, 2006). This was a novel method for examining acrylamide toxicity in male germ cells; and spermatocytes from mice exposed to acrylamide for 12 months were found to have significantly increased frequencies of γH2A.X foci at all doses tested (Fig. 5).

The data presented herein support the notion that chronic acrylamide exposure leads to increased levels of genetic damage in male germ cells. However, unlike the results of the FPG-modified Comet assay, γH2A.X staining was indicative only of DSBs as opposed to adduct-related DNA lesions. DNA adducts may be resolved into DSB via the activity of repair mechanisms. Therefore, the DSBs observed in meiotic spreads of spermatocytes likely represent a fraction of the range of lesions detected by the Comet assay, suggesting that not all of the genetic damage induced by acrylamide is repaired. This is likely as during spermatogenesis, male germ cells are deficient in certain repair proteins (Olsen et al., 2005). Spermatogonia carry out numerous mitotic divisions and undergo meiosis to form primary spermatocytes, secondary spermatocytes, and haploid spermatids. Spermatids then undergo spermiogenesis, involving cellular differentiation and extensive chromatin condensation to give rise to spermatozoa. Consequently, male germ cells at different developmental stages of spermatogenesis have varying degrees of susceptibility to DNA damaging agents (Olsen et al., 2005). The postmeiotic phase is considered to be the most vulnerable to environmental genotoxins, and several studies have demonstrated that the dominant lethal effects of acrylamide act on postmeiotic male germ cells (Favor and Shelby, 2005; Ghanayem et al., 2005b).

The typical cellular response to irreparable DNA damage is apoptosis, which functions to remove damaged cells from the pool of developing male germ cells (Print and Loveland, 2000). Acrylamide exposure in rodents has previously been reported to result in germ cell loss, vacuolations in the testis, atrophied seminiferous tubules, and multinucleated germ cells, following oral doses between 20 and 60 mg/kg bodyweight/day (Burek et al., 1980; Hashimoto et al., 1981; Yang et al., 2005). However, at the chronic exposures used in this study, no gross morphological effects or signs of germ cell apoptosis were observed in testis histology of mice (Fig. 2) despite elevated levels of genetic damage in spermatocytes (Figs. 3 and 5). Therefore, the question remains of whether genetic lesions induced by chronic acrylamide exposure in early male germ cells of mice persist until the point of sperm maturation and downstream fertilization. Indeed, DNA adducts induced by other well-

73 characterized environmental toxicants, such as benzo[a]pyrene, have been demonstrated to accumulate in spermatogenic cells without eliciting apoptosis, and these lesions are not necessarily repaired in the fertilized oocyte (Olsen et al., 2005). Additionally, several studies have demonstrated that acrylamide exposure at high doses results in heritable damage in specific-locus mutation assays and heritable translocation tests (Favor and Shelby, 2005). However, it is unknown whether chronic acrylamide exposure at levels of human exposure results in heritable DNA damage, and based on the results of the current study, the answer to the above question remains to be explored.

It was also interesting to note that DNA damage in spermatocytes of acrylamide-exposed mice appeared to accumulate over time (Figs. 3–5). However, male germ cells are continuously generated through the process of spermatogenesis. Spermatocytes examined at the 12-month time point, e.g., would have been generated in the testis (and directly exposed to acrylamide) for a period of only 21 days prior to measurement, as it takes around 42 days for spermatogonia to develop into spermatozoa (Gilbert, 2000). Several mechanisms may account for this buildup of damage in a self-renewing cell population. Genetic damage induced by acrylamide exposure may accumulate in replicating spermatogonial stem cells, which are responsible for the regeneration of all subsequent germ cell types. Furthermore, long-term, chronic acrylamide exposure may allow for more efficient conversion of acrylamide to glycidamide, subsequently producing more genetic damage over time. Certainly, the internal dose of glycidamide in mice has been found to be greater using lower dosing rates of acrylamide (Doerge et al., 2005). The factor of age may also contribute to accumulation of DNA damage, as mice were around 18 months old by the end of the study. Indeed, there is strong evidence that common repair mechanisms are diminished in older individuals (Xu et al., 2008), and thus, chronic acrylamide exposure may be of more consequence to the reproductive health of older males, particularly considering current trends in advanced parental age.

To summarize, we demonstrated that acrylamide exposure via the drinking water, at a range of low-level exposures encompassing human dietary estimates, induces genetic damage in male germ cells of mice. This damage was generated in the absence of major effects on mouse health or defects in spermatogenesis. Additionally, DNA damage appeared to be both dose and time dependent and attributable to a range of genetic lesions arising

74 from either the generation of free radicals via acrylamide metabolism or the formation of DNA adducts by the acrylamide metabolite, glycidamide. Our data indicate that not all of these lesions may be resolved by DNA-repair mechanisms in spermatocytes, and this damage may be exacerbated in older animals, possibly due to reduced efficiency of DNA- repair mechanisms. In conclusion, the results we present here indicate that acrylamide exposure generates male germ line DNA damage at levels equivalent to human dietary intake. As genetic damage in the male germ line may be transmitted to future offspring, we suggest that the genotoxicity of acrylamide be taken into further consideration regarding male reproductive health in humans.

Supplementary Data

Supplementary data are available online at http://toxsci.oxfordjournals.org/.

Funding

Hunter Medical Research Institute; The Australian Research Council Centre of Excellence in Biotechnology and Development and the Priority Research Centre for Chemical Biology, University of Newcastle. B.J.N. is the recipient of an Australian Postgraduate Award Scholarship from the Commonwealth of Australia.

Acknowledgements

We thank Prof. Frank Grützner and Aaron Casey from the University of Adelaide for introducing us to the technique of meiotic spreads.

References

Baleato, R. M., Aitken, R. J., and Roman, S. D. (2005). Vitamin A regulation of BMP4 expression in the male germ line. Dev. Biol. 286, 78–90.

Baum, M., Loeppky, R. N., Thielen, S., and Eisenbrand, G. (2008). Genotoxicity of glycidamide in comparison to 3-N-nitroso-oxazolidin-2-one. J. Agric. Food Chem. 56, 5989– 5993.

75 Burek, J. D., Albee, R. R., Beyer, J. E., Bell, T. J., Carreon, R. M., Morden, D. C., Wade, C. E., Hermann, E. A., and Gorzinski, S. J. (1980). Subchronic toxicity of acrylamide administered to rats in the drinking water followed by up to 144 days of recovery. J. Environ. Pathol. Toxicol. 4, 157–182.

Cederbaum, A. I. (2006). CYP2E1—Biochemical and toxicological aspects and role in alcohol- induced liver injury. Mt Sinai J. Med. 73, 657–672.

Doerge, D. R., Young, J. F., McDaniel, L. P., Twaddle, N. C., and Churchwell, M. I. (2005). Toxicokinetics of acrylamide and glycidamide in B6C3F1 mice. Toxicol. Appl. Pharmacol. 202, 258–267.

Erkekoglu, P., and Baydar, T. (2010). Toxicity of acrylamide and evaluation of its exposure in baby foods. Nutr. Res. Rev. 23, 323–333.

Exon, J. H. (2006). A review of the toxicology of acrylamide. J. Toxicol. Environ. Health B Crit. Rev. 9, 397–412.

FAO/WHO. (2003). Food and Agriculture Organization and World Health Organization. Joint FAO/WHO Food Standards Programme Codex Committee on Food Additives and Contaminants. Discussion paper on acrylamide. Rotterdam, The Netherlands, March.

Favor, J., and Shelby, M. D. (2005). Transmitted mutational events induced in mouse germ cells following acrylamide or glycidamide exposure. Mutat. Res. 580, 21–30.

Friedman, M. A., Dulak, L. H., and Stedham, M. A. (1995). A lifetime oncogenicity study in rats with acrylamide. Fundam. Appl. Toxicol. 27, 95–105.

Gamboa da Costa, G., Churchwell, M. I., Hamilton, L. P., Von Tungeln, L. S., Beland, F. A., Marques, M. M., and Doerge, D. R. (2003). DNA adduct formation from acrylamide via conversion to glycidamide in adult and neonatal mice. Chem. Res. Toxicol. 16, 1328–1337.

Ghanayem, B. I., McDaniel, L. P., Churchwell, M. I., Twaddle, N. C., Snyder, R., Fennell, T. R., and Doerge, D. R. (2005a). Role of CYP2E1 in the epoxidation of acrylamide to glycidamide and formation of DNA and haemoglobin adducts. Toxicol. Sci. 88, 311–318.

Ghanayem, B. I., Witt, K. L., El-Hadri, L., Hoffler, U., Kissling, G. E., Shelby, M. D., and Bishop, J. B. (2005b). Comparison of germ cell mutagenicity in male CYP2E1-null and wild-type mice treated with acrylamide: Evidence supporting a glycidamide-mediated effect. Biol. Reprod. 72, 157–163.

Ghanayem, B. I., Witt, K. L., Kissling, G. E., Tice, R. R., and Recio, L. (2005c). Absence of acrylamide-induced genotoxicity in CYP2E1-null mice: Evidence consistent with a glycidamide-mediated effect. Mutat. Res. 578, 284–297.

Gilbert, S. F. (2000). The saga of the germ line. In Developmental Biology, 6th ed., pp. 599- 601. Sinauer Assoc. Inc, Sunderland, MA.

Hansen, S. H., Olsen, A. K., Soderlund, E. J., and Brunborg, G. (2010). In vitro investigations of glycidamide-induced DNA lesions in mouse male germ cells and in mouse and human lymphocytes. Mutat. Res. 696, 55–61.

76 Hashimoto, K., Sakamoto, J., and Tanii, H. (1981). Neurotoxicity of acrylamide and related compounds and their effects on male gonads in mice. Arch. Toxicol. 47, 179–189.

Jansen, J., Olsen, A. K., Wiger, R., Naegeli, H., de Boer, P., van Der Hoeven, F., Holme, J. A., Brunborg, G., and Mullenders, L. (2001). Nucleotide excision repair in rat male germ cells: Low level of repair in intact cells contrasts with high dual incision activity in vitro. Nucleic Acids Res. 29, 1791–1800.

Johnson, K. A., Gorzinski, S. J., Bodner, K. M., Campbell, R. A., Wolf, C. H., Friedman, M. A., and Mast, R. W. (1986). Chronic toxicity and oncogenicity study on acrylamide incorporated in the drinking water of Fischer 344 rats. Toxicol. Appl. Pharmacol. 85, 154–168.

LoPachin, R. M., and DeCaprio, A. P. (2005). Protein adduct formation as a molecular mechanism in neurotoxicity. Toxicol. Sci. 86, 214–225.

LoPachin, R. M., and Gavin, T. (2008). Acrylamide-induced nerve terminal damage: Relevance to neurotoxic and neurodegenerative mechanisms. J. Agric. Food Chem. 56, 5994–6003.

Mahadevaiah, S. K., Turner, J. M., Baudat, F., Rogakou, E. P., de Boer, P., Blanco-Rodriguez, J., Jasin, M., Keeney, S., Bonner, W. M., and Burgoyne, P. S. (2001). Recombinational DNA double-strand breaks in mice precede synapsis. Nat. Genet. 27, 271–276.

Marchetti, F., Bishop, J., Lowe, X., and Wyrobek, A. J. (2009). Chromosomal mosaicism in mouse two-cell embryos after paternal exposure to acrylamide. Toxicol. Sci. 107, 194–205.

Marchetti, F., and Wyrobek, A. J. (2005). Mechanisms and consequences of paternally- transmitted chromosomal abnormalities. Birth Defects Res. C Embryo Today 75, 112–129.

Matulis, S., and Handel, M. A. (2006). Spermatocyte responses in vitro to induced DNA damage. Mol. Reprod. Dev. 73, 1061–1072.

Mei, N., Hu, J., Churchwell, M. I., Guo, L., Moore, M. M., Doerge, D. R., and Chen, T. (2008). Genotoxic effects of acrylamide and glycidamide in mouse lymphoma cells. Food Chem. Toxicol. 46, 628–636.

Olsen, A. K., Lindeman, B., Wiger, R., Duale, N., and Brunborg, G. (2005). How do male germ cells handle DNA damage? Toxicol. Appl. Pharmacol. 207(Suppl. 2), 521–531.

Parzefall, W. (2008). Minireview on the toxicity of dietary acrylamide. Food Chem. Toxicol. 46, 1360–1364.

Print, C. G., and Loveland, K. L. (2000). Germ cell suicide: New insights into apoptosis during spermatogenesis. BioEssays: News and Reviews in Molecular, Cellular and Developmental Biology 22, 423–430.

Sakamoto, J., and Hashimoto, K. (1986). Reproductive toxicity of acrylamide and related compounds in mice--effects on fertility and sperm morphology. Arch. Toxicol. 59, 201–205.

Sega, G. A., Alcota, R. P., Tancongco, C. P., and Brimer, P. A. (1989). Acrylamide binding to the DNA and protamine of spermiogenic stages in the mouse and its relationship to genetic damage. Mutat. Res. 216, 221–230.

77 Singh, N. P. (2000). Microgels for estimation of DNA strand breaks, DNA protein crosslinks and apoptosis. Mutat. Res. 455, 111–127.

Singh, N. P., McCoy, M. T., Tice, R. R., and Schneider, E. L. (1988). A simple technique for quantitation of low levels of DNA damage in individual cells. Exp. Cell Res. 175, 184–191.

Sublet, V. H., Zenick, H., and Smith, M. K. (1989). Factors associated with reduced fertility and implantation rates in females mated to acrylamide-treated rats. Toxicology 55, 53–67.

Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S., and Tornqvist, M. (2002). Analysis of acrylamide, a carcinogen formed in heated foodstuffs. J. Agric. Food Chem. 50, 4998–5006.

Thielen, S., Baum, M., Hoffmann, M., Loeppky, R. N., and Eisenbrand, G. (2006). Genotoxicity of glycidamide in comparison to (+/-)-anti-benzo[a] pyrene-7,8-dihydrodiol- 9,10-epoxide and alpha-acetoxy-N-nitroso-diethanolamine in human blood and in mammalian V79-cells. Mol. Nutr. Food Res. 50, 430–436.

Tyl, R. W., and Friedman, M. A. (2003). Effects of acrylamide on rodent reproductive performance. Reprod. Toxicol. 17, 1–13.

Wang, R. S., McDaniel, L. P., Manjanatha, M. G., Shelton, S. D., Doerge, D. R., and Mei, N. (2010). Mutagenicity of acrylamide and glycidamide in the testes of Big Blue mice. Toxicol. Sci. 117, 72–80.

Watzek, N., Bohm, N., Feld, J., Scherbl, D., Berger, F., Merz, K. H., Lampen, A., Reemtsma, T., Tannenbaum, S. R., Skipper, P. L., et al. (2012). N7-glycidamide-guanine DNA adduct formation by orally ingested acrylamide in rats: A dose-response study encompassing human diet-related exposure levels. Chem. Res. Toxicol. 25, 381–390.

WHO. (2011). Evaluation of certain contaminants in food. World Health Organization Technical Report Series 959. 1–105.

Wise, L. D., Gordon, L. R., Soper, K. A., Duchai, D. M., and Morrissey, R. E. (1995). Developmental neurotoxicity evaluation of acrylamide in Sprague Dawley rats. Neurotoxicol. Teratol. 17, 189–198.

Xu, G., Herzig, M., Rotrekl, V., and Walter, C. A. (2008). Base excision repair, aging and health span. Mech. Ageing Dev. 129, 366–382.

Yang, H. J., Lee, S. H., Jin, Y., Choi, J. H., Han, D. U., Chae, C., Lee, M. H., and Han, C. H. (2005). Toxicological effects of acrylamide on rat testicular gene expression profile. Reprod. Toxicol. 19, 527–534.

78 Table 1

Mean body weight and percentage weight difference of male mice following 12 months of acrylamide exposure a

1 month* 3 months 6 months 9 months 12 months Exposure (µg/ml) Mean (g)b (∆%)c SD Mean (g) (∆%) SD Mean (g) (∆%) SD Mean (g) (∆%) SD Mean (g) (∆%) SD

Control 37.0 (-1) 2.8 40.4 (0) 1.6 42.6 (5) 2.4 47.4 (6) 2.0 41.4 (8) 3.2 0.001 40.2 (3) 1.5 41.2 (3) 1.9 46.0 (5) 2.4 42.3 (15) 3.1 49.8 (17) 4.4 0.01 40.7 (6) 2.5 46.2 (5) 4.5 47.9 (2) 4.8 48.7 (15) 5.5 48.6 (18) 5.5 0.1 38.3 (4) 5.2 42.4 (13) 4.4 43.8 (5) 3.7 43.2 (8) 3.9 44.5 (7) 3.2 1 40.2 (1) 2.0 48.4 (4) 5.8 45.3 (4) 2.3 45.7 (8) 3.5 51.0 (-4) 1.9 10 41.6 (6) 2.9 44.5 (9) 2.7 43.5 (6) 3.1 49.0 (19) 6.3 45.9 (5) 2.8

a There were 6 mice in each treatment group. b The mean body weight (g) immediately after treatment. c (∆%) Difference in weight immediately after treatment minus intial weight at start of treatment, expressed as percentage of initial weight. *One month bodyweight data was significantly lower compared to all other time points due to normal mouse growth (p < 0.001), however acrylamide treatment did not significantly affect mouse bodyweight. Detailed statistical analyses of mouse bodyweight shown in supplementary data (sTable 1).

79 Table 2

Calculated daily dose (mg/kg bodyweight/day) and total cumulative dose of acrylamide in male mice during 12 months of exposure a

Mean Daily Acrylamide Dose (mg/ kg bodyweight/ day)b

1 month 3 months 6 months 9 months 12 months Exposure (µg/ml) Mean SD Mean SD Mean SD Mean SD Mean SD

0 (Control) 0 0 0 0 0 0 0 0 0 0 0.001 0.0002 0.00 0.0002 0.00 0.0001 0.00 0.0002 0.00 0.0001 0.00 0.01 d 0.0022 0.00 0.0019 0.00 0.0015 0.00 0.0014 0.00 0.0011 0.00 0.1 0.0173 0.01 0.0207 0.00 0.0147 0.00 0.0145 0.00 0.0173 0.00 1 0.1775 0.01 0.1977 0.05 0.1501 0.01 0.1424 0.01 0.1677 0.01 10 1.7964 0.18 2.0221 0.59 1.4064 0.09 1.5722 0.22 1.5314 0.06

Mean Total Cumulative Acrylamide Dose (mg/ kg bodyweight)c

0 (Control) 0 0 0 0 0 0 0 0 0 0 0.001 0.006 0.0 0.017 0.0 0.023 0.0 0.042 0.0 0.041 0.0 0.01 0.063 0.0 0.157 0.1 0.255 0.0 0.345 0.1 0.374 0.0 0.1 0.484 0.2 1.731 0.2 2.450 0.2 3.654 0.3 5.817 0.7 1 4.970 0.4 16.629 4.8 25.084 1.3 35.743 2.5 56.935 3.4 10 50.300 4.9 169.726 56.4 235.489 16.4 394.203 49.1 520.142 22.5 a There were six mice in each treatment group. b Daily acrylamide dose was calculated using total water consumption for each mouse, multiplied by acrylamide concentration, divided by mouse weight immediately after treatment and divided by the number of days exposed. Values presented as mean daily dose (mg/kg bodyweight day). c Cumulative dose was calculated using total water consumption for each mouse, multiplied by acrylamide concentration, and divided by mouse weight immediately after treatment. Values presented as mean cumulative dose (mg/kg bodyweight). d Mice in the 0.01 µg/ml treatment group received a daily acrylamide dose equivalent to human exposure levels (0.001 mg/kg bodyweight/day, WHO 2011).

80 Figure 1

Acrylamide administered via the drinking water for 12 months

Monitored for bodyweight and water consumption

Testes extracted Testes extracted

Isolated early Fixed and germ cells sectioned

Comet assay/ Histology and Meiotic spreads Immunohistochemistry

FIG. 1. Experimental design. Acrylamide was administered to male mice through the drinking water at concentrations of 0.001, 0.01, 0.1, 1, or 10 μg/ml, and animals were exposed for a period of 1, 3, 6, 9, or 12 months (i.e., 30 treatment groups). Six individuals were allocated to each treatment group, totaling 180 mice. Mouse bodyweight and water consumption were monitored throughout the duration of the study. The testes from three males were used for early germ cell isolation (see Materials and Methods section). Testes from the remaining three males were fixed and sectioned for histological analysis and immunohistochemistry.

81 Figure 2

FIG. 2. Histology of testis from control and acrylamide-exposed mice at 10 μg/ml for 12 months. Acrylamide exposure did not induce gross morphological effects in mouse testis after 12 months of acrylamide exposure (0–10 μg/ml) through the drinking water. Scale bars equal to 100 μm. Note. See online version for color version.

82 Figure 3 A

250 Control 0.001μg/ml ****** 0.01μg/ml *** *** 200 0.1μg/ml *** 1μg/ml *** 150 10μg/ml *** *** ****** ***

100 Mean Tail DNA % DNA Tail Mean

50 (% increase relative to control) increaserelativeto (%

0 1 month 3 months 6 months 9 months 12 months

FIG. 3. Adduct-related DNA damage in spermatocytes from mice exposed to acrylamide in the drinking water (0–10 μg/ml) over the course of 12 months, as measured by a Comet assay in the presence of FPG. (A) Mean Tail DNA%, expressed as % increase relative to control to enable comparison of data between measurements at different time points. Significant increases in DNA damage were first detected after 6 months at the two highest doses (1 and 10 μg/ml). After 9 and 12 months, increases in DNA damage were observed in doses as low as 0.01 μg/ml compared with respective control data (***p <0.001). Detailed statistical analyses are shown in Supplementary table 5. (B) Representative comet images from control samples, 0.01 μg/ml treatment and the highest 10 μg/ml treatment at the 6- month and 12-month time point. See online version for color version.

83 Figure 4

84 FIG. 4. Immunolocalization of γH2A.X, a marker of DSBs, in testis sections from mice exposed to acrylamide in vivo (Control, 0.1, 1, and 10 μg/ml) following 6, 9, and 12 months of exposure. Sections were sequentially probed with anti-cyp2e1 and appropriate secondary antibody before being counterstained with DAPI (shown for control sections). Overall expression of γH2A.X appeared to be upregulated in testis from acrylamide-treated mice compared with control testis sections. Expression of γH2A.X was localized to spermatocytes, with some staining in later stage germ cells, round spermatids. Scale bars equal to 100 μm. See online version for color version.

85 Figure 5

C 16 14 *** 12 *** 10 *** 8 *** *

H2AX Foci/cellH2AX 6 γ 4 2 Mean 0

18 16 14 12 10 8 6

4 H2A.X foci >50 fociH2A.X γ 2 0

withcellsnumberof Mean

FIG. 5. DSBs observed in spermatocytes from mice exposed to acrylamide in vivo, measured by meiotic spreads and γH2A.X foci counts. (A, B) Representative images of spermatocyte meiotic spreads. SCP3 labels condensed chromosomes; γH2A.X labels both the asynapsed sex chromosomes (arrows) and sites of DSBs (γH2A.X foci, indicated by arrow heads). (A) Spermatocyte with relatively few γH2A.X foci. (B) Spermatocyte exhibiting numerous γH2A.X foci. (C) Mean counts of γH2A.X foci per cell at each dose after 12 months of exposure. Statistically significant increases in DSBs were observed in spermatocytes in all doses compared with control cells (*p < 0.05, ***p < 0.001). Cells with greater than 50 γH2A.X foci were counted separately in (D), which shows mean numbers of cells (> 50 foci) for each dose at the 12-month time point; however, these were not statistically significant. See online version for color version.

Chapter 4 The Impact of 12-Month Chronic Acrylamide Exposure on Mouse Testicular Gene Expression

87 Chapter 4: Overview

The third aim of this thesis was to elucidate the mechanisms by which chronic acrylamide exposure leads to genetic damage in male germ cells. The scale of the chronic exposure model described in Chapter 3 was designed to generate a plethora of tissue samples that could be stored for future analysis. Thus, in the following manuscript, microarray analysis was used to explore the effect of acrylamide on gene expression in mouse testis RNA obtained from the chronic exposure study.

Microarrays were conducted on testis RNA samples from all doses, at the 1, 6 and 12 month time points, and differential gene expression was examined. Out of the total probes on the Illumina Beadchip, 4% were identified as differentially expressed (1,052 unique genes); and surprisingly, hierarchical clustering of these genes indicated that a large majority of these genes exhibited dose-dependent changes in expression. Microarray data were therefore screened using automated dose-response modelling, and 16% of the total probes were identified as dose-dependently regulated (3,704 unique genes). These genes were subsequently entered into Ingenuity Pathway Analysis (IPA), which revealed that the major pathways regulated in response to acrylamide were associated with spermatogenesis, signalling, epoxide metabolism, oxidative stress, cancer signalling, tight and gap junction signalling, and DNA repair.

The results of this study indicated that low doses of acrylamide can induce widespread transcriptional changes in the testis. Additionally, the expression of numerous genes was found to be regulated dose-dependently, at time points previous to the induction of DNA damage observed in the previous chapter. Thus, microarray data from this study may also provide important data on potential transcriptional biomarkers of acrylamide. Furthermore, these results shed light on the mechanisms and pathways affected, or regulated in response to, chronic acrylamide exposure in the male.

Note: Supplementary figures are located in Appendix C, Pg. 145. Supplementary tables are located on disc found on the inside back cover of this thesis.

88 The Impact of 12-Month Chronic Acrylamide Exposure on Mouse Testicular Gene Expression

Belinda J. Nixon*, Aimee L. Katen*, Simone J. Stanger*, Brett Nixon *†, and Shaun D. Roman*†1

*Reproductive Science Group, School of Environmental and Life Sciences, University of Newcastle, Callaghan, New South Wales 2308, Australia

†The Australian Research Council Centre of Excellence in Biotechnology and Development and the Priority Research Centre for Chemical Biology, University of Newcastle

1To whom correspondence should be addressed: Dr. Shaun Roman, Biology, School of Environmental and Life Sciences, University of Newcastle, University Drive, Callaghan 2308, Australia, Phone: +61 2 4921 6818, Fax: +612 4921 6308. E-mail: [email protected]

89 Abstract

Acrylamide is a known male reproductive toxicant in rodents at acute doses; however, the impact of long term, chronic exposure to acrylamide on male reproductive health is largely unknown. In this study, the transcriptional effect of acrylamide was investigated in mouse testis following chronic doses in the range of human exposures (0 - 10 µg/ml), administered via the drinking water for 1, 6 or 12 months. Microarray analyses identified a total of 1052 genes that were significantly, differentially expressed in the testis of acrylamide treated and control mice. Interestingly, the expression profiles of many of these genes exhibited clear dose-dependent effects. Thus, automated dose-response modelling was conducted to identify all genes with sigmoidal dose-response curves and EC50 values ranging from 0 – 10 µg/ml of acrylamide. This approach revealed a total of 3,704 unique genes that were regulated in a dose-dependent manner. Of the differentially expressed genes, 32% satisfied the criteria for dose-dependency. Several of these genes, such as Gsg1, Spata16, Igf2 and Ephx1, exhibited dose-dependent regulation prior to the detection of DNA damage. These genes therefore hold considerable promise as transcriptional biomarkers of chronic acrylamide exposure. Analysis of all genes regulated in a dose-dependent manner indicated that the major pathways affected by acrylamide were associated with spermatogenesis, signalling, epoxide metabolism, oxidative stress, cancer signalling, tight and gap junction signalling, and DNA repair. In conclusion, our data indicated that a suite of testicular genes respond dose-dependently to low-doses of acrylamide and provides novel insight into the identification of potential biomarkers and the mechanisms associated with chronic acrylamide toxicity in the male.

Keywords

Acrylamide, Chronic, Testis, Microarray, Biomarker, Gene Expression

90 Introduction

Acrylamide is a synthetic compound primarily used for the production of polymers in numerous industries, e.g. wastewater treatment, ore processing, cosmetics and in laboratory gels (Exon 2006). More recently, acrylamide has been found to form during the cooking of carbohydrate-rich foods, particularly in various potato products and breads (Tareke et al. 2000; Tareke et al. 2002). Occupational exposure to acrylamide at acute doses generates neurotoxicity in humans, such as numbness in the hands and feet, ataxia and progressive muscle weakness (Hagmar et al. 2001). In rodent studies, acrylamide is genotoxic and carcinogenic in various organs; although the majority of epidemiological studies report a lack of an association between dietary acrylamide intake and increased incidence of cancer in humans (Allred et al. 2005; Ehlers et al. 2012). Acrylamide also acts as a male reproductive toxicant in rodents, generating reduced fertility, dominant lethality, increased embryo resorptions and reduced litter sizes; particularly following paternal exposure (Tyl et al. 2000; Generoso et al. 1996; Bishop 1991; Sublet et al. 1989; Sakamoto and Hashimoto 1986). The No Observable Adverse Effect Level for acrylamide reproductive toxicity in rodents is between 2 – 5 mg/kg bodyweight/day, which is more than 2000 fold greater than the estimated human dietary consumption (0.4 – 1 µg/kg bodyweight/day) (Exon 2006). However, the impact of long term, chronic exposure to acrylamide on male reproductive health is largely unknown.

The toxic effects of acrylamide on male reproductive parameters have been well documented (Tyl and Friedman 2003), yet the mechanisms by which acrylamide elicits these effects have not been fully delineated. Acrylamide is a soft electrophilic alkene that reacts and forms adducts with soft nucleophiles, such as the thiol groups on proteins and glutathione (Lopachin and Decaprio 2005). For example, acrylamide has been found to interfere with kinesin motor proteins, which may contribute to chromosomal aberrations and cell division defects during male germ cell development (Sickles et al. 2007). However, metabolism of acrylamide by the Cytochrome P450, Cyp2e1, yields glycidamide; an epoxide metabolite that typically forms adducts with hard nucleophillic sites on adenine and guanine of DNA (Lopachin and Decaprio 2005). Indeed, the significance of Cyp2e1 in acrylamide toxicity in males was demonstrated by the elegant studies of Ghanayem et al. (2005a; 2005b; 2005c); in which acrylamide induced genotoxicity was shown to be abrogated in

91 Cyp2e1-null mice. DNA repair mechanisms are limited in late stage germ cells (Olsen et al. 2005), thus a genotoxic mode of action is considered likely responsible for mediating the reproductive toxicity of acrylamide; either through a clastogenic mechanism of acrylamide or via the mutagenic activity of its metabolite, glycidamide.

The discovery of acrylamide in foods prompted several recommendations by the joint Food and Agriculture Organization/World Health Organization (WHO 2002), including further research of acrylamide genotoxicity in somatic and germ cells using genome-wide expression profiling. In the last decade, a handful of studies that examine the effect of acrylamide on gene expression have been published. In the rat, acrylamide exposure at 60 mg/kg bodyweight/day altered testis gene expression patterns, as measured by microarray analysis (Yang et al. 2005). Genes relating to testicular function, apoptosis, cellular redox, cell cycle and nucleic acid binding were significantly regulated in response to acrylamide exposure. Despite the affinity of acrylamide for male reproductive tissues, microarray and Q-PCR studies in rats report a lack of evidence for hormonal dysregulation in the hypothalymus-pituitary-thyroid axis (Bowyer et al. 2008; Camacho et al. 2012). Similar microarray experiments in mice have focussed on hepatic gene expression, following high acrylamide or glycidamide exposures via the drinking water (500 mg/L) (Mathur and D'Cruz 2011; Lee et al. 2012). In these studies, the hepatic gene functions most affected by exposure included xenobiotic, glutathione and steroid metabolism, mitochondrial function and oxidative phosphorylation. In addition, cancer development pathways and diseases in neurological and reproductive systems were identified.

However, the above studies were conducted using short term, acute doses of acrylamide, which far exceed estimates of human dietary exposure to acrylamide. We previously described a 12-month chronic acrylamide exposure study in mice; in which dose-dependent increases in DNA damage were observed in male germ cells at doses that encompass human exposure levels (0.0001 – 2 mg/kg bodyweight/day) (Nixon et al. 2012). In the current study, microarray analysis was conducted on testis tissue harvested from the same set of mice to examine gene expression regulation in response to acrylamide exposure via the drinking water. To our knowledge, this is the first study to profile testicular gene expression patterns in mice exposed to chronic levels of acrylamide.

92 Materials & Methods

Animals

All experimental procedures involving animals were conducted in accordance with the policies set out by the Animal Care and Ethics Committee of the University of Newcastle (Ethics Number: SR A-2009-121). Animals were housed under controlled temperature and humidity conditions, 16 h light, 8 h dark, with food and water provided ad libitum.

Experimental Design

The study design, acrylamide dosage, animal weights and water consumption have previously been described (Nixon et al. 2012) and are summarised in supplementary data, sFig. 1. Briefly, 180 male Swiss mice were divided into 30 treatment groups (six mice per group) and subjected to 6 doses of acrylamide (A9099, Sigma, St Louis, MO) via the drinking water at 0, 0.001, 0.01, 0.1, 1 or 10 µg/ml (equivalent to 0.0001, 0.002, 0.017, 0.167 or 1.666 mg/kg bodyweight/day) provided ad libitum, for 1, 3, 6, 9, or 12 months. At each time point, animals were euthanized using CO2 and male germ cells were extracted from testes of the first set of triplicate mice for DNA damage analysis (Nixon et al. 2012, see supplementary data, sFig. 1). One testis from the second set of triplicate mice was also used for DNA damage analysis; hence the remaining testes from these males were used in the present study for microarray analyses. Microarrays were conducted on samples from all 6 doses of the 1, 6 and 12 month time points only. This was done in order to gain expression data across the entire time of exposure. Each treatment group of mice was setup in triplicate (54 samples in total). However, due to injuries received from inter-mouse fighting, several replicate animals were removed from the 12 month time point in accordance with animal ethics regulations. Thus, 46 samples remained for microarray analysis with 2 – 3 replicate animals remaining per treatment group.

93 RNA Extraction & Microarray

RNA was extracted from individual testes of acrylamide exposed mice using RNeasy Mini Kit (Qiagen, Chadstone Centre, VIC) according to manufacturers’ instructions. Quantification and purity of RNA samples was measured using the NanoDrop spectrophotometer (Thermo Scientific). Testis RNA quality was confirmed using the Agilent Bioanalyser 2100, and 3 µg of RNA was labelled and prepared for hybridisation to an Illumina Sentrix Mouse ref8v2 Beadchip. Array scanning was carried out according to the Illumina manual on an Illumina BeadArray Reader and scanner operating software, BeadStudio, converted the array signal into a text file for analysis. Labelling, hybridising, washing and array scanning was conducted by the Australian Genome Research Facility (AGRF) Melbourne, Australia. All data is MIAME compliant and has been deposited in a MIAME compliant database, accessible via http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=nzmjxqykgoiaknw&acc=GSE47574, accession number GSE47574.

Statistics & Analysis of Microarray Data

Microarray data were normalised by quantile normalisation method using Bioconductor lumi package software and imported into Partek Genomics Suite (Partek) for analysis. One sample was identified in the principle component analysis as an outlier and was omitted from analysis (lowest dose at the earliest time point: 0.001 µg/ml, 1 month, supplementary data, sFig. 2). A flow diagram of the subsequent data analysis approach utilised in this study was presented in Fig. 1. Comparisons between control and acrylamide treated samples were conducted using pair-wise analysis of variance (ANOVA) and statistically significant, differentially expressed genes were defined by a fold difference of 1.5 and p < 0.05. The differentially expressed genes common to more than one time point were identified and hierarchical clustering analysis was conducted on all differentially expressed genes at each time point using Cluster 3.0 software (freeware provided by Stanford University, written by Michiel de Hoon, Siya Imato and Saturo Miyano based on the original script by Michael Eisen). Images of Cluster data were generated using open source software Java TreeView v1.1.6r2 (created by Alok).

94 Dose-Response Modelling Analysis

Gene expression dose-response changes were examined using ToxResponse Modeler (Burgoon and Zacharewski 2008). For each gene, ToxResponse Modeler determines the best fit model from five different mathematical models (linear, exponential, Gaussian, sigmoidal and quadratic) and calculates the half maximal effective concentration (EC50) value. As indicated in the analysis flowchart (Fig. 1C) microarray data were first screened for probes with expression > 8.0, considered as background. All Expressed Probes were entered into ToxResponse Modeler. Probes with a best fit to the sigmoidal model were selected and the half maximal effective concentration (EC50) values were calculated. Probes were then filtered for EC50 values that were within the experimental dose range of 0.001 - 10 µg/ml

(reasonable EC50, Table 1). The 'dose-dependent' genes that were common to more than one time point were identified, and hierarchical clustering analysis was carried out (Fig. 4), as previously described above for differentially expressed genes. The dose-dependent genes identified by ToxResponse Modeler were subsequently analysed using Ingenuity Pathways Analysis (Ingenuity systems, Redwood City, CA).

Results

Differential Gene Expression following Acrylamide Exposure

Microarray analysis of testis gene expression in acrylamide exposed mice identified 1,052 differentially expressed genes that had a fold expression difference of 1.5 and p < 0.05, as determined by pair-wise ANOVA (Fig. 1B). Interestingly, the majority of the differentially expressed genes were found to be regulated at 6 months (865 in total) compared to the 1 and 12 month time points (248 and 180 in total respectively) as depicted in Venn diagram, Fig. 2A. The total gene numbers and complete list of differentially expressed genes are included in supplementary data, sTable 1 & 2, and represent 9.3% of the total number of transcripts on the Illumina Beadchip.

Hierarchical clustering analysis was conducted for all acrylamide doses (Control, 0.001, 0.01, 0.1, 1 and 10 µg/ml) within each time point (Fig. 2B). It was immediately evident that a large number of differentially expressed genes exhibited dose-dependent expression profiles,

95 particularly following 6 months of acrylamide exposure. Dose-dependent expression could also be observed following 1 month of exposure, though expression patterns were more variable at the 12 month time point. In Fig. 2B, approximately 60% of the differentially expressed genes at 1 month, and around 75% of the differentially expressed genes at 6 months appeared to be dose-dependently downregulated (red to green), with the remaining genes showing a general increase in expression with increasing acrylamide dose (green to red).

The Venn diagram in Fig. 2A also indicated that several differentially expressed genes were regulated at more than 1 time point, with 18 genes found to be common to all time points measured. These 18 common differentially expressed genes are listed in Fig. 3A and hierarchical clustering analysis separated these genes into two distinct branches. Ten genes were identified that displayed a tendency to have increasing expression with dose, along with decreasing expression with time (Fig. 3B). Regulation of germ cell-specific gene 1 (Gsg1) and spermatogenesis associated 16 (Spata16) in this first branch (highlighted in black, Fig. 3A) was of particular interest due to their involvement in spermatogenesis; though their precise functions have not been fully elucidated. GSG1 is known to interact with testis-specific poly(A) polymerase, which is involved in the nuclear export, translation, and stability of mRNA in the testis (Choi et al. 2008). Indeed, this form of translational control is particularly important as numerous genes are subject to transcriptional delay during male germ cell differentiation (Kleene 1996).

SPATA16 is thought to have a role in acrosome formation, and mutations of this gene in humans have been implicated in globozoospermia, a feature of teratozoospermia characterised by round-headed spermatozoa that lack an acrosome (Dam et al. 2007). A third gene of interest was insulin-like growth factor 2 (Igf2), highlighted in dark grey in Fig. 3A. Igf2 is a well-characterised, paternally imprinted gene and epimutations of Igf2 have been associated with poor sperm parameters and male infertility (Rajender et al. 2011). Igf2 was grouped in the second clustering branch, which had a tendency to increase in expression over time, although an overall effect of dose was less clear (Fig. 3B). The functions of the remaining genes are included in supplementary data, sTable 3, and were primarily involved in signal transduction, metabolism and cell maintenance.

96 Quantitative Dose-Response Modelling

The evident dose-response patterns of gene expression displayed in the heatmap of differentially expressed genes (Fig. 2B) prompted a more in-depth analysis of the testicular genes that responded to, or were affected by, acrylamide treatment. Thus, as outlined in Fig. 1C, microarray data were screened for All Expressed Genes (probes with background expression > 8) and dose-response modelling was carried out. The dose-response modelling program, ToxResponse Modeler (Burgoon & Zacharewski 2008) was utilised to model 10,237 probes, representing 7,943 unique genes. At the 1, 6 and 12 month time points respectively, 15%, 26% and 18% of the expressed probes exhibited sigmoidal dose-response curves (see Table 1). The highest number of sigmoidal probes was observed at the 6 month time point (2,620); supporting the clear dose dependency observed at 6 months in the heatmap of differentially expressed genes (Fig. 2B).

ToxResponse Modeler also provided relative chemical potency data as the EC50 value is calculated for each gene. At the 1, 6 and 12 month time points respectively, 0.8%, 24% and

16% of the expressed probes were found to have an EC50 value within the experimental dose range of 0.001 – 10 µg/ml (Table 1, Reasonable EC50). Low EC50 values were indicative of sensitive regulation of gene expression in response to exposure. Interestingly, when categorised by increasing dose, a general increase in the number of probes with reasonable

EC50 values was also observed (Table 1, EC50 Range). The 3,704 total unique genes that satisfied the dose-dependent criteria will be referred to as ‘dose-dependent genes’ and are listed in supplementary data, sTable 4.

Similar to the previous analysis on differentially expressed genes, dose-dependent genes were depicted in a Venn diagram (Fig. 4A) and hierarchical clustering analysis was conducted for each time point (Fig. 4B). The data in the Venn diagram identified 67 dose- dependent genes that were common to all three time points (Fig. 4A) and included a number of interesting targets. These 67 genes are listed in supplementary data, sTable 5, with corresponding EC50 values and clustering trend. However, the major finding of the dose-response modelling was that acrylamide not only affected the expression of significantly regulated genes (differentially expressed genes), but a suite of testicular genes responded dose-dependently to acrylamide exposure (Fig. 4B). In contrast to the heatmaps

97 of differentially expressed genes, the majority of these genes were dose-dependently upregulated (green to red), with approximately two thirds at 1 month and 6 months exhibiting upregulated expression profiles. However, as previously observed for differentially expressed genes, dose-dependent expression profiles at the 12 month time point were less clear.

Of the total differentially expressed genes, 32% were found to be dose-dependently regulated at one or more time points (listed in supplementary data, sTable 6) and these data were compiled into a third Venn diagram (Fig. 5). In regard to the 18 common differentially expressed genes (Fig. 3), none were found to be dose-dependently regulated at every time point (Fig. 5). However, four were dose-dependently regulated at more than one time point (Map2K7, Serpinf1, Gsg1 and Igf2), five were regulated at 6 months (Sh3gl3, Rffl, Spata16,

Hmgcs2 and Acsbg1) and one was regulated at 12 months (Glt8d1). The EC50 values of genes, Gsg1, Spata16 and Igf2, ranged from 0.04 – 6.11 µg/ml. Interestingly, Gsg1 and Igf2 appeared to be sensitively regulated in response to acrylamide exposure, as these genes were first identified as dose-dependent at the earliest time point with relatively low EC50 values (0.04 & 0.12 µg/ml respectively, supplementary data, sTable 6).

Highest Ranking Pathways Regulated by Acrylamide Exposure

The dose-dependent genes were subsequently analysed using IPA to examine pathways and molecular functions affected by chronic acrylamide exposure. The top 10 canonical pathways were identified for each time point (Fig. 6) and the molecules involved in the highest ranking pathways for each of these time points were outlined in Fig.s 7 - 9.

The highest scoring pathway identified at the 1 month time point was RAN signalling (Fig. 6), which is essential for selective nuclear transport through nuclear pore complexes (Holt et al. 2007). An example of the regulation observed in the RAN signalling pathway is depicted in Fig. 7, in which the majority of molecules were upregulated (indicated in red). Upregulation was mostly observed in the carrier proteins, Importin α (KPNA1) and Importin β (KPNB1), that mediate recognition and passage of cargo proteins into the nucleus. Indeed, the majority of the highest ranking pathways identified at the 1 month time point were involved

98 in signalling (3-phosphoinositide biosynthesis, neurotrophin/TRK signalling, hepatocyte growth factor signalling, superpathway of inositol phosphate compounds & tight junction signalling). Interestingly, two pathways were involved in cancer signalling (pancreatic adenocarcinoma signalling, acute myeloid leukemia signalling), and the remaining pathways were related to metabolism and the immune response (dolichyl-diphosphooligosaccharide biosynthesis & histamine degradation).

Following 6 months of chronic acrylamide exposure, the top regulated pathway was JAK/STAT signalling (Fig. 6), a signal transduction pathway that activates transcription in a number of cellular processes (Sutherland et al. 2012). Several molecules were regulated in this pathway, (represented in Fig. 8) with downregulation observed for the Signal Transducer and Activator of Transcription (STAT) proteins (indicated in green). Other pathways regulated at 6 months that were relevant to spermatogenesis included; mitochondrial dysfunction, protein ubiquitination pathway, cleavage and polyadenylation of pre-mRNA pathway and hormone signalling pathways (IGF-1 signalling & aldosterone signalling in epithelial cells). The remaining highest ranking pathways involved cancer signalling (prostate cancer signalling & PI3K/AKT signalling) and one neurological disease pathway (Huntington's disease signalling).

Whilst the previous two time points exhibited regulation in various signalling pathways, it was intriguing to note that more than 50% of the top 10 pathways at the 12 month time point were relevant to junction signalling (tight junction signalling, gap junction signalling, remodelling of epithelial adherens junctions, germ cell-Sertoli cell junction signalling & Sertoli cell-Sertoli cell junction signalling) (Fig. 6). Two hormone signalling pathways were also identified (androgen signalling & glucocorticoid receptor signalling), as well as one immune response pathway (mechanisms of viral exit from host cells). The highest ranking pathway regulated at 12 months was the nucleotide excision repair (NER) pathway, which was of particular interest given the DNA damaging potential of the acrylamide metabolite, glycidamide. Downregulation of NER molecules were observed in global genomic repair, transcription coupled repair and in repair synthesis (indicated in green, Fig. 9).

99 Discussion

Considering the relatively low dose range of acrylamide administered to mice in the current study (0.0001 – 2 mg/kg bodyweight/day), it was surprising that a large number of the differentially expressed genes exhibited dose-dependent gene expression profiles (Fig. 2B). Furthermore, widespread gene expression effects were observed after only 1 month of chronic acrylamide exposure, whereas in our previous study (Nixon et al. 2012), DNA damage was not detected in male germ cells prior to the 6 month time point; and then only at the highest doses of administration (i.e. 2 mg/kg bodyweight/day). Thus, gene expression changes may represent a more sensitive measure of acrylamide exposure than DNA damage. Indeed, as altered gene expression occurred prior to the induction of DNA damage, the results of this study not only shed light on the molecular mechanisms and functions affected by acrylamide, but enable the identification of transcriptional biomarkers of acrylamide exposure in the testis.

Analysis of testis microarray data identified a number of genes that may be useful markers of acrylamide exposure. Of the 1,052 differentially expressed genes that were significantly expressed, 18 genes were regulated throughout the 12 month exposure period (Fig. 3), 10 of which were found to have dose-dependent expression profiles at one or more time points (Fig. 5). Given the toxic effects of acrylamide on male reproductive parameters, genes of particular interest were associated with spermatogenesis (Gsg1, Spata16) or development (Igf2), and interestingly, these were amongst the genes that were also dose- dependently regulated (Fig. 5). As previously mentioned, GSG1 may have a role in mRNA export and translation, whereas SPATA16 is thought to be involved in acrosome formation (Choi et al. 2008; Dam et al. 2007). Igf2 on the other hand, is a well-characterised, paternally imprinted gene known to be involved in foetal and postnatal growth (Rajender et al. 2011). Epimutations of Igf2 have been associated with abnormal sperm parameters and male infertility and indeed, acrylamide exposure has been reported to generate methylation defects of Igf2 in early male germ cells of rats (Wang et al. 2010). Gsg1, Spata16 and Igf2 may also represent suitable acrylamide biomarkers, as dose-dependent expression of these genes was identified early in the exposure regime (Gsg1 & Igf2 at 1 month, Spata16 at 6 months). The calculated EC50 values for these genes were relatively low, particularly for Gsg1 and Igf2 (0.04 & 0.12 µg/ml respectively, supplementary data, sTable 6), which

100 indicated that the regulation of these genes was sensitive to low-level acrylamide exposure. This highlights the advantage of dose-response modelling in microarray analyses of toxicology studies, as it facilitates the identification of potential biomarkers that are both dose-dependently regulated and highly sensitive.

Dose-response modelling identified 67 dose-dependent genes that were common to all three time points (Fig. 4), several of which could be considered as potential acrylamide biomarkers (supplementary data, sTable 5). One interesting example was microsomal epoxide hydrolase 1 (Ephx1), which functions in the activation and detoxification of epoxides (Clement et al. 2007). Thus, the upregulation of Ephx1 in the testis of acrylamide exposed mice may be indicative of the generation of the epoxide metabolite, glycidamide. Indeed, EPHX1 has previously been suggested as a sensitive transcriptional marker of glycidamide exposure in human epithelial cells (Clement et al. 2007). Other potentially important biomarker genes included genes associated with spermatogenesis & development (Spata7, Foxj3), gap junctions (Gja1), or transcription (Brdt, Cxxc1) (supplementary data, sTable 5).

The regulation of the above genes at all three time points was indicative that chronic acrylamide exposure in the testis affects molecular functions relevant to male germ cell development. Certainly, IPA analysis of the dose-dependent genes identified a number of functions relating to male reproduction and spermatogenesis (supplementary data, sTable 7). Interestingly, IPA analysis identified RAN signalling and JAK/STAT signalling as the highest ranking pathways at the 1 and 6 month time points respectively (Fig. 7 & 8). The highly organised, RAN signalling pathway mediates nuclear transport of proteins and is thought to be involved in directing male germ cell differentiation and development (Holt et al. 2007). Indeed, interference in RAN signalling leads to improper signalling and cell cycle defects in mouse fibroblast cells (Battistoni et al. 1997). Several molecules of the RAN signalling pathway were identified as upregulated in the current study (Fig. 7), suggestive of increased signalling in response to acrylamide; particularly as the majority of the highest ranked pathways at the 1 month time point were also involved in signalling (Fig. 6). Following 6 months of chronic acrylamide exposure, both up and downregulation of molecules were observed in molecules of the JAK/STAT signalling pathway (Fig. 8). JAK/STAT signalling is not known to have a role in mammalian spermatogenesis, though it has been described in stem

101 cell maintenance of Drosophila testis and primordial follicle activation in mammalian oogenesis (Hombria and Brown 2002; Sutherland et al. 2012).

The second and third highest ranking pathways regulated at 6 months were mitochondrial dysfunction and the protein ubiquitination pathway, both of which have been implicated in impaired spermatogenesis and male infertility (Lu et al. 2008; Baarends et al. 2000). Mitochondrial dysfunction was of particular interest, as mitochondria are a major source of reactive oxygen species and oxidative stress has been associated with acrylamide toxicity (Lee et al. 2012). The toxicological responses reported by IPA (supplementary data, sTable 8) also identified mitochondrial dysfunction at the 6 and 12 month time point as well as glutathione depletion and NRF2-mediated oxidative stress response, providing further evidence of an oxidative mechanism of acrylamide toxicity in the testis. Recently, NRF2 was identified as part of the ROS mediated response to acrylamide in neuronal cells (Caito et al. 2013; Jin et al. 2013). Given the known carcinogenicity of acrylamide in rodents following acute acrylamide doses (Rice 2005), it was not unexpected that several cancer signalling pathways were also regulated at these earlier time points at chronic exposure levels (Fig. 6).

An intriguing observation at the final 12 month time point was regulation of a number of pathways associated with junction signalling and the blood/testis barrier (Fig. 6). For example, several members of the tight junction, gap junction and germ cell-Sertoli cell signalling pathways were primarily downregulated (indicated in green, supplementary data sFig. 3A, B, C). The protective blood/testis barrier is formed by the tight junctions and gap junctions between the supporting Sertoli cells of the seminiferous tubule and is not only crucial for male germ cell development and differentiation, but also inhibits the passage of toxins and foreign compounds (Cheng et al. 2011). Thus, dysregulation and breakdown of the blood/testis barrier may enable greater concentrations of acrylamide to pervade the seminiferous tubule. Furthermore, the top regulated pathway at the 12 month time point was NER, which functions to remove bulky adducts from DNA (Gu et al. 2010). Intriguingly, whilst NER is one of the primary mechanisms that repairs DNA damage in somatic cells, alternative repair mechanisms are thought to be active during spermatogenesis, such as base excision repair or mismatch repair, as NER is known to be non-functional in male germ cells (Jansen et al. 2001). Thus the downregulation of molecules associated with DNA damage recognition and repair initiation in NER (indicated in green, Fig. 9) may be due to

102 expression changes in support cells in the testis, rather than germ cells. Indeed, reduced repair in supporting testicular cells may also affect blood-testis barrier integrity in the testis.

Taken together, these data suggest a possible breakdown of the blood/testis barrier and junction signalling, concomitant with an increased susceptibility to DNA damage following long-term acrylamide exposure in mouse testis. This mechanism may be responsible for the accumulation of DNA damage detected over time in male germ cells in our previous study (Nixon et al. 2012). Additionally, this may also explain the diminished dose-dependent expression profiles observed in the hierarchical clustering of the 12 month time point (Fig. 2 & 4). Signalling pathways were initially upregulated following short-term exposure; however, continued exposure may overwhelm the sensitivity of these responses, culminating in breakdown of signalling and increased testicular toxicity. Indeed, many EC50 values of dose-dependent genes commonly regulated at all time points were observed to increase over time (supplementary data, sTable 5), suggesting a temporal reduction in sensitivity of these genes to acrylamide exposure.

In summary, chronic exposure to acrylamide induces significant transcriptional changes in the testis, the large majority of which are dose-dependent. The pathways most represented by these genes were relevant to spermatogenesis, signalling, epoxide metabolism, oxidative stress, cancer signalling, tight and gap junction signalling, and DNA repair. The usefulness of dose-response modelling in identifying potential biomarkers was also demonstrated; though, further research is needed to validate these potential transcriptional markers of exposure. Certainly, several molecules associated with the above pathways may represent biomarkers of chronic acrylamide exposure, as their regulation was identified prior to the detection of DNA damage in male germ cells of these mice (Nixon et al. 2012). Thus, the results of this study strongly indicate that chronic exposure to acrylamide in mice, at human exposure levels, does indeed have a widespread effect in the testis. Our results also offer insight into a potential mechanism of chronic acrylamide toxicity, by which overwhelmed signalling leads to deleterious consequences in the male.

103 Funding

This work was supported by funding from Hunter Medical Research Institute, The Australian Research Council Centre of Excellence in Biotechnology and Development and the Priority Research Centre for Chemical Biology, University of Newcastle. B.J.N. is the recipient of an Australian Postgraduate Award Scholarship from the Commonwealth of Australia. A.L.K. is the recipient of a Faculty of Science and Information Technology Research Training Scholarship.

Acknowledgements

The authors would like to thank Dr. Rust Turakulov (Australian Genome Research Facility Ltd) for his assistance in data analysis, as well as our colleague, Dr. Natalie Beveridge, for her helpful advice in the preparation of this manuscript. Additionally, the authors also thank Rance Nault and Prof. Timothy Zacharewski for their assistance in the use of ToxResponse Modeler, as well as Richard Dear (University of Newcastle Computing Officer) and Nic Croce (University of Newcastle Statistical Consultant) for their assistance with data handling.

References

Allred, C. D., Twaddle, N. C., Allred, K. F., Goeppinger, T. S., Churchwell, M. I., Ju, Y. H., Helferich, W. G. and Doerge, D. R. (2005). Soy processing affects metabolism and disposition of dietary isoflavones in ovariectomized BALB/c mice. Journal of agricultural and food chemistry 53(22), 8542-8550.

Baarends, W. M., van der Laan, R. and Grootegoed, J. A. (2000). Specific aspects of the ubiquitin system in spermatogenesis. Journal of endocrinological investigation 23(9), 597- 604.

Battistoni, A., Guarguaglini, G., Degrassi, F., Pittoggi, C., Palena, A., Di Matteo, G., Pisano, C., Cundari, E. and Lavia, P. (1997). Deregulated expression of the RanBP1 gene alters cell cycle progression in murine fibroblasts. Journal of cell science 110 ( Pt 19), 2345-2357.

Bishop, J., Chapin, R., Fail, P., George, J., Grizzle, T., Sadler, B., Heindel, J. (1991). Acrylamide induced dominant lethality in mice following low dose chronic administration in drinking water. . Environ. Mol. Mutagen. (abstract) 17(suppl. 9):A25.

Bowyer, J. F., Latendresse, J. R., Delongchamp, R. R., Muskhelishvili, L., Warbritton, A. R., Thomas, M., Tareke, E., McDaniel, L. P. and Doerge, D. R. (2008). The effects of subchronic acrylamide exposure on gene expression, neurochemistry, hormones, and histopathology in the hypothalamus-pituitary-thyroid axis of male Fischer 344 rats. Toxicology and applied pharmacology 230(2), 208-215.

104 Burgoon, L. D. and Zacharewski, T. R. (2008). Automated quantitative dose-response modeling and point of departure determination for large toxicogenomic and high- throughput screening data sets. Toxicological sciences : an official journal of the Society of Toxicology 104(2), 412-418.

Caito, S., Yu, Y. and Aschner, M. (2013). Differential response to acrylonitrile toxicity in rat primary astrocytes and microglia. Neurotoxicology 37C, 93-99.

Camacho, L., Latendresse, J. R., Muskhelishvili, L., Patton, R., Bowyer, J. F., Thomas, M. and Doerge, D. R. (2012). Effects of acrylamide exposure on serum hormones, gene expression, cell proliferation, and histopathology in male reproductive tissues of Fischer 344 rats. Toxicology letters 211(2), 135-143.

Cheng, C. Y., Wong, E. W., Lie, P. P., Li, M. W., Mruk, D. D., Yan, H. H., Mok, K. W., Mannu, J., Mathur, P. P., Lui, W. Y., Lee, W. M., Bonanomi, M. and Silvestrini, B. (2011). Regulation of blood-testis barrier dynamics by desmosome, gap junction, hemidesmosome and polarity proteins: An unexpected turn of events. Spermatogenesis 1(2), 105-115.

Choi, H. S., Lee, S. H., Kim, H. and Lee, Y. (2008). Germ cell-specific gene 1 targets testis- specific poly(A) polymerase to the endoplasmic reticulum through protein-protein interactions. FEBS letters 582(8), 1203-1209.

Clement, F. C., Dip, R. and Naegeli, H. (2007). Expression profile of human cells in culture exposed to glycidamide, a reactive metabolite of the heat-induced food carcinogen acrylamide. Toxicology 240, 111–124.

Dam, A. H., Koscinski, I., Kremer, J. A., Moutou, C., Jaeger, A. S., Oudakker, A. R., Tournaye, H., Charlet, N., Lagier-Tourenne, C., van Bokhoven, H. and Viville, S. (2007). Homozygous mutation in SPATA16 is associated with male infertility in human globozoospermia. American journal of human genetics 81(4), 813-820.

Ehlers, A., Lenze, D., Broll, H., Zagon, J., Hummel, M. and Lampen, A. (2012). Dose dependent molecular effects of acrylamide and glycidamide in human cancer cell lines and human primary hepatocytes. Toxicology letters 217(2), 111-120.

Exon, J. H. (2006). A review of the toxicology of acrylamide. J. Toxicol. Environ. Health B. Crit. Rev. 9(5), 397-412.

Generoso, E. E., Sega, G. A., Lockhardt, A. M., Hughes, L. A., Cain, K. T., Cacheiro, N. L. A. and Shelby, M. D. (1996). Dominant lethal mutations, heritable translocations, and unscheduled DNA synthesis induced in male mouse germ cells by glycidamide, a metabolite of acrylamide. Mutation Research 371, 521-530.

Ghanayem, B. I., McDaniel, L. P., Churchwell, M. I., Twaddle, N. C., Snyder, R., Fennell, T. R. and Doerge, D. R. (2005a). Role of CYP2E1 in the epoxidation of acrylamide to glycidamide and formation of DNA and hemoglobin adducts. Toxicol Sci 88(2), 311-318.

Ghanayem, B. I., Witt, K. L., El-Hadri, L., Hoffler, U., Kissling, G. E., Shelby, M. D. and Bishop, J. B. (2005b). Comparison of germ cell mutagenicity in male CYP2E1-null and wild-type mice treated with acrylamide: evidence supporting a glycidamide-mediated effect. Biol Reprod 72(1), 157-163.

105 Ghanayem, B. I., Witt, K. L., Kissling, G. E., Tice, R. R. and Recio, L. (2005c). Absence of acrylamide-induced genotoxicity in CYP2E1-null mice: evidence consistent with a glycidamide-mediated effect. Mutat Res 578(1-2), 284-297.

Gu, A., Ji, G., Zhou, Y., Long, Y., Shi, X., Fu, G., Wang, S., Song, L. and Wang, X. (2010). Polymorphisms of nucleotide-excision repair genes may contribute to sperm DNA fragmentation and male infertility. Reproductive biomedicine online 21(5), 602-609.

Hagmar, L., Tornqvist, M., Nordander, C., Rosen, I., Bruze, M., Kautiainen, A., Magnusson, A. L., Malmberg, B., Aprea, P., Granath, F. and Axmon, A. (2001). Health effects of occupational exposure to acrylamide using hemoglobin adducts as biomarkers of internal dose. Scand. J. Work Environ. Health 27, 219-226.

Holt, J. E., Ly-Huynh, J. D., Efthymiadis, A., Hime, G. R., Loveland, K. L. and Jans, D. A. (2007). Regulation of Nuclear Import During Differentiation; The IMP alpha Gene Family and Spermatogenesis. Current genomics 8(5), 323-334.

Hombria, J. C. and Brown, S. (2002). The fertile field of Drosophila Jak/STAT signalling. Current biology : CB 12(16), R569-575.

Jansen, J., Olsen A.K., Wiger R, Naegeli H, de Boer P, et al. (2001) Nucleotide excision repair in rat male germ cells: low level of repair in intact cells contrasts with high dual incision activity in vitro. Nucleic acids Res 29: 1791-1800.

Jin, M., Ande, A., Kumar, A. and Kumar, S. (2013). Regulation of cytochrome P450 2e1 expression by ethanol: role of oxidative stress-mediated pkc/jnk/sp1 pathway. Cell death & disease 4, e554.

Kleene, K. C. (1996). Patterns of translational regulation in the mammalian testis. Mol Reprod Dev 43(2), 268-281.

Lee, T., Manjanatha, M. G., Aidoo, A., Moland, C. L., Branham, W. S., Fuscoe, J. C., Ali, A. A. and Desai, V. G. (2012). Expression analysis of hepatic mitochondria-related genes in mice exposed to acrylamide and glycidamide. Journal of toxicology and environmental health. Part A 75(6), 324-339.

Lopachin, R. M. and Decaprio, A. P. (2005). Protein adduct formation as a molecular mechanism in neurotoxicity. Toxicological sciences : an official journal of the Society of Toxicology 86(2), 214-225.

Lu, B., Poirier, C., Gaspar, T., Gratzke, C., Harrison, W., Busija, D., Matzuk, M. M., Andersson, K. E., Overbeek, P. A. and Bishop, C. E. (2008). A mutation in the inner mitochondrial membrane peptidase 2-like gene (Immp2l) affects mitochondrial function and impairs fertility in mice. Biology of reproduction 78(4), 601-610.

Mathur, P. P. and D'Cruz, S. C. (2011). The effect of environmental contaminants on testicular function. Asian journal of andrology 13(4), 585-591.

Nixon, B. J., Stanger, S. J., Nixon, B. and Roman, S. D. (2012). Chronic exposure to acrylamide induces DNA damage in male germ cells of mice. Toxicological sciences : an official journal of the Society of Toxicology 129(1), 135-145.

106 Olsen, A. K., Lindeman, B., Wiger, R., Duale, N. and Brunborg, G. (2005). How do male germ cells handle DNA damage? Toxicol Appl Pharmacol 207(2 Suppl), 521-531.

Rajender, S., Avery, K. and Agarwal, A. (2011). Epigenetics, spermatogenesis and male infertility. Mutation research 727(3), 62-71.

Rice, J. M. (2005). The carcinogenicity of acrylamide. Mutat Res 580(1-2), 3-20.

Sakamoto, J. and Hashimoto, K. (1986). Reproductive toxicity of acrylamide and related compounds in mice--effects on fertility and sperm morphology. Arch Toxicol 59(4), 201-205.

Sickles, D. W., Sperry, A. O., Testino, A. and Friedman, M. (2007). Acrylamide effects on kinesin-related proteins of the mitotic/meiotic spindle. Toxicol. Appl. Pharmacol. 222, 111- 121.

Sublet, V. H., Zenick, H. and Smith, M. K. (1989). Factors associated with reduced fertility and implantation rates in females mated to acrylamide-treated rats. Toxicology 55(1-2), 53- 67.

Sutherland, J. M., Keightley, R. A., Nixon, B., Roman, S. D., Robker, R. L., Russell, D. L. and McLaughlin, E. A. (2012). Suppressor of cytokine signaling 4 (SOCS4): moderator of ovarian primordial follicle activation. Journal of cellular physiology 227(3), 1188-1198.

Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S. and Tornqvist, M. (2000). Acrylamide: a cooking carcinogen? Chem Res Toxicol 13(6), 517-522.

Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S. and Tornqvist, M. (2002). Analysis of acrylamide, a carcinogen formed in heated foodstuffs. J Agric Food Chem 50(17), 4998- 5006.

Tyl, R. W. and Friedman, M. A. (2003). Effects of acrylamide on rodent reproductive performance. Reprod Toxicol 17(1), 1-13.

Tyl, R. W., Friedman, M. A., Losco, P. E., Fisher, L. C., Johnson, K. A., Strother, D. E. and Wolf, C. H. (2000). Rat two-generation reproduction and dominant lethal study of acrylamide in drinking water. Reprod. Toxicol. 14, 385-401.

Wang, Z., Lu, S., Liu, C., Yao, N., Zhang, C. and Ma, X. (2010). Acrylamide disturbs genomic imprinting during spermatogenesis. Front Biosci (Elite Ed) 2, 165-170.

WHO (2002). FAO/WHO Consultation on the Health Implications of Acrylamide in Food. Summary Report of a meeting held in Geneva, 25-27 June 2002. (http://www.who.int/foodsafety/publications/chem/en/acrylamide_full.pdf).

Yang, H. J., Lee, S. H., Jin, Y., Choi, J. H., Han, D. U., Chae, C., Lee, M. H. and Han, C. H. (2005). Toxicological effects of acrylamide on rat testicular gene expression profile. Reprod Toxicol 19(4), 527-534.

107 Tables & Figures

Note: Supplementary figures are located in Appendix C, Pg. 145. Supplementary tables are located on disc found on the inside back cover of this thesis.

Table 1

Dose-Response Modelling of Testis RNA Microarray Data from Mice Exposed to Acrylamide via the Drinking Water for 1, 6 or 12 Months.

1 Month 6 Months 12 Months b ToxResponse Modeler a10,237 probes 10,237 probes 10,237 probes Input c Sigmoidal Dose-Response 1,551 2,620 1,857 d Reasonable EC50 837 2,404 1,681 e EC50 Range: 0.001 - 0.01 µg/ml 120 47 33

0.01 - 0.1 µg/ml 422 906 507

0.1 - 1 µg/ml 136 449 329

1 - 10 µg/ml 159 1,002 812 a Microarray datasets were screened for probes with background expression > 8 (All Expressed Genes), totalling 10,237 probes (7,943 unique genes). b ToxResponse Modeler (Burgoon and Zacharewski, 2008) was used to perform automated dose-response modelling by identifying the best fit model to calculate EC50 values (see Materials and Methods section). c Probes with Sigmoidal Dose-Response expression were selected (EC50 values were not calculated for probes with Gaussian dose-response profiles). d Probes with EC50 values within the experimental dose range (0.001 - 10 µg acrylamide/ml), totalling 4,912 probes (3,704 unique genes) defined as dose-dependent genes. e The numbers of probes with Reasonable EC50 values, sorted according to EC50 dose range (complete gene lists included in supplementary data, sTable 4).

108 Figure 1

Fig. 1: Flowchart of microarray analyses conducted on testis RNA from acrylamide exposed mice. (A) Microarrays were conducted using an Illumina Sentrix Mouse ref8v2 Beadchip on samples from acrylamide doses (Control, 0.001, 0.01, 0.1, 1, and 10 µg/ml) at 1, 6 and 12 month time points. There were 2 to 3 replicates for each treatment group, totalling 46 testis RNA samples (see Materials and Methods section). One replicate was omitted as an outlier in the Principal Component Analysis (supplementary data, sFig. 2); hence 45 samples remained for analysis. (B) Significantly regulated genes were identified by Pair-wise ANOVA (±2 fold change expression, p < 0.05) and defined as differentially expressed genes. The differentially expressed genes common to each time point were depicted in a Venn diagram (Fig. 2A) and hierarchical clustering was conducted (Fig. 2B & 3). (C) In order to examine dose response effects on gene expression, microarray data was screened for probes with background expression greater than 8 (All Expressed Genes). Dose response modelling was carried out using ToxResponse Modeler (Burgoon & Zacharewski, 2008) to select for genes that exhibited dose-dependent changes and had ‘Reasonable EC50’ values (see Table 1). The genes regulated in a dose-dependent manner that are common to each time point were depicted in a Venn diagram and hierarchical clustering was conducted (Fig. 4). Ingenuity Pathways Analysis (IPA) was used to examine the top canonical pathways based on the dose-dependent genes identified (Fig.s 6 - 9).

109 Figure 2

Fig. 2: Microarray analysis was conducted on testis RNA from mice exposed to acrylamide via the drinking water (Control, 0.001, 0.01, 0.1, 1 and 10 µg/ml) over a 12 month period. (A) Venn diagram depicting numbers of differentially expressed genes (±2 fold change expression, p < 0.05) at time points 1, 6 and 12 months and the numbers of overlapping genes. The 18 genes common to all 3 time points are listed in Fig. 3A. (B) Heatmaps of differentially expressed genes at each time point showing dose-dependent patterns of gene expression (hierarchical clustering tree and gene names cropped out for simplicity). Red and green represent induction and repression of genes respectively. For interpretation of the references to colour in the figure caption, the reader is referred to the web version of the article.

110 Figure 3

Fig. 3: (A) The 18 differentially expressed genes common to all three time points as depicted in Venn diagram, Fig. 2A. Hierarchical clustering tree based on similarities in gene expression is included on the left (B) Heatmaps for the 18 genes common to all three time points with time and dose axes indicated (left to right: 1, 6 and 12 months; top to bottom: Control, 0.001, 0.01, 0.1, 1 and 10 µg/ml). Red and green represent induction and repression of genes respectively. Genes involved in spermatogenesis (Gsg1, Spata16) or imprinting/development (Igf2) were highlighted in black or grey respectively. For interpretation of the references to colour in the figure caption, the reader is referred to the web version of the article.

111 Figure 4

Fig. 4: Microarray analysis was conducted on testis RNA from mice exposed to acrylamide (Control, 0.001, 0.01, 0.1, 1 and 10 µg/ml) over a 12 month period. (A) Venn diagram depicting numbers of unique genes regulated in a dose-dependent manner with Reasonable

EC50 values at time points 1, 6 and 12 months. The 67 genes common to all 3 time points were included in supplementary data, sTable 5. (B) Heatmaps at each time point revealed a far greater number of genes with dose-dependent changes in gene expression in response to acrylamide compared to the differentially expressed genes identified in Fig. 2. Red and green represent induction and repression of genes respectively (hierarchical clustering tree and gene names cropped out for simplicity). For interpretation of the references to colour in the figure caption, the reader is referred to the web version of the article.

112 Figure 5

Fig. 5: Venn diagram of genes that were both differentially expressed (±2 fold change expression, p < 0.05) and met the criteria of dose-response modelling. No genes were found to be dose-dependently regulated at every time point. However, 10 of the 18 differentially expressed genes that were common to all time points (listed in Fig. 3) were dose- dependently regulated at more than one time point. The positions of these genes in the Venn diagram were annotated and genes involved in spermatogenesis (Gsg1, Spata16) or imprinting/development (Igf2) were highlighted in black, or grey respectively.

113 Figure 6

Fig. 6: The top 10 canonical pathways that were dose-dependently regulated in mouse testis following in vivo acrylamide exposure for 1, 6 and 12 months, as identified by IPA. Coloured bars represent the significance (p-value) of the association between the regulated genes and the canonical pathway, calculated using a right-tailed Fisher’s exact test (upper-axis, - Log P). Ratios of the proportion of regulated genes compared to the total number of molecules that make up the particular pathway are represented as a line graph (lower axis, Ratio).

114 Figure 7 1 Month Top Canonical Pathway – RAN Signalling

Fig. 7: RAN Signalling was the top canonical pathway regulated in mouse testis following 1 month of in vivo acrylamide exposure. Signalling pathway was generated by IPA based on data from the highest acrylamide dose, 10 µg/ml. Regulated molecules were colour coded according to their expression (red: up-regulated, green: down-regulated). For interpretation of the references to colour in the figure caption, the reader is referred to the web version of the article.

115 Figure 8 6 Months Top Canonical Pathway – JAK/STAT Signalling

Fig. 8: JAK/STAT Signalling was the top canonical pathway regulated in mouse testis following 6 months of in vivo acrylamide exposure. Signalling pathway was generated by IPA based on data from the highest acrylamide dose, 10 µg/ml. Regulated molecules were colour coded according to their expression (red: up-regulated, green: down-regulated). For interpretation of the references to colour in the figure caption, the reader is referred to the web version of the article.

116 Figure 9

12 Months Top Canonical Pathway – Nucleotide Excision Repair Pathway

Fig. 9: The Nucleotide Excision Repair Pathway was the top canonical pathway regulated in mouse testis following 6 months of in vivo acrylamide exposure. Signalling pathway was generated by IPA based on data from the highest acrylamide dose, 10 µg/ml. Regulated molecules were colour coded according to their expression (red: up-regulated, green: downregulated). For interpretation of the references to colour in the figure caption, the reader is referred to the web version of the article.

117 Chapter 5: Discussion

Human exposure to acrylamide is unavoidable due to the prevalence of acrylamide in foods and the numerous sources of industrial exposure. It is also difficult to evaluate the risk of acrylamide in the human diet as there is limited data regarding the consequences of chronic acrylamide exposure. Indeed, humans are exposed to a myriad of chemicals in the environment, though little data exists regarding the long term effects of the majority of these chemicals (Mattsson 2007). Thus, the outcomes of the current research not only contribute to the current understanding of acrylamide reproductive toxicity, but also provide insight into the effect of prolonged, low-level xenobiotic exposure on male reproductive health. In this final chapter, the major findings of this thesis will be reviewed in reference to the original aims of the project and future directions of this research will be discussed.

Aim 1: To examine the nature of genetic damage induced by acrylamide in isolated male germ cells.

The design of this research project could be divided into in vitro and in vivo components. The in vitro component formed the basis for the manuscript presented in Chapter 2, in which Cyp2e1 expression was examined at different stages of spermatogenesis and the effect of acrylamide was observed in isolated populations of mouse spermatocytes. The advantage of these in vitro experiments was the ability to examine direct effects of toxic exposure within specific tissues or cells; thus enabling better understanding of the molecular mechanisms that underpin toxicity in the whole animal.

Expression of Cyp2e1 was previously thought to be restricted to the interstitial cells of the testis (Healy et al. 1999). However, the results of this first manuscript indicated that male germ cells express Cyp2e1 and that this expression fluctuates throughout different stages of spermatogenesis. Indeed, immunostaining of testis sections and isolated germ cells indicated that spermatocytes express the highest levels of Cyp2e1 protein expression (Pg. 51). Furthermore, Cyp2e1 mRNA was observed to increase in spermatocytes following acrylamide treatment (Pg. 53). Induction of Cyp2e1 has been observed in response to other

118 xenobiotics in tissues such as kidney, intestine and lung (Amacher 2010; Pavek and Dvorak 2008); however to our knowledge this was the first study to examine cyp gene induction in male germ cells following acrylamide exposure. Spermatocytes were therefore found to not only express Cyp2e1, but respond to the presence of acrylamide by upregulating Cyp2e1 gene expression.

DNA damage assessment using the FPG modified Comet assay revealed significant, dose- dependent increases in DNA adducts in spermatocytes treated with acrylamide or glycidamide. The specific acrylamide/glycidamide induced lesions detected by FPG were most likely N7-GA-Gua, as these are the major DNA adducts formed by glycidamide (Fig. 4 Pg. 23). The DNA glycosylase, FPG, recognises a range of DNA lesions including oxidised purines as well as alkylated or ring-opened purines (Olsen et al. 2005). N7-alkylated guanine can spontaneously depurinate, forming alkyl-formamidopyrimidine (fapy)-G lesions that may be converted to methylated (fapy)-G lesions under the alkaline conditions of the Comet assay {Hansen, 2010 #212}. Indeed, these lesions can be recognised by the FPG enzyme. Additional characterisation of this damage using the hOGG modified Comet assay also provided some evidence of oxidative damage, particularly following glycidamide exposure, as hOGG has a more narrow substrate specificity for oxidised purines, such as 8-oxoguanine (Pg. 56). Previous mouse studies have demonstrated that both acrylamide and glycidamide induce damage in a number of tissues such as lung, liver and kidney, as well as in testis and testicular cells (Gamboa da Costa et al. 2003; Hansen et al. 2010). However, our results were the first to examine the effect of these compounds within a specific stage of male germ cell development.

These in vitro studies provided relevant new data regarding acrylamide-mediated damage in early male germ cells, and shed light on the mechanisms of xenobiotic metabolism present in the male germ line. It was also established that male germ cells from Swiss mice respond to xenobiotic exposure; and that the Comet assay, modified with enzymes, was a useful tool in detecting acrylamide mediated DNA damage in these cells; and indeed could be applied to genotoxic chemicals in general. Hence, this study provided the foundation for a large- scale, chronic exposure experiment in vivo.

119 Aim 2: To investigate whether chronic acrylamide exposure induces DNA damage in male germ cells in vivo.

The majority of acute acrylamide exposure studies examine overt reproductive endpoints, such as reduced fertility rates and dominant lethality, which are unlikely to be induced at exposures equivalent to human dietary intake (Tyl and Friedman 2003). However, these studies do not accurately reflect the human situation and thus the consequences of dietary exposure to acrylamide for male reproductive health remain unclear. The chronic exposure model described in Chapter 3 was the first study to systematically examine the effect of acrylamide exposure in the male, at doses in range of human acrylamide exposure. Whilst overt reproductive effects were not observed in this study, significant increases in germ line DNA damage were detected in mice exposed to acrylamide for 6 - 12 months. Damage to the germ line by acrylamide may not just impair fertility, but may also lead to heritable damage that impact on the health of future offspring (WHO 2002).

The chronic exposure model utilised an administered dose range of 0 - 2 mg/kg bodyweight/day, which encompassed the human exposure estimate of 0.5 µg/kg bodyweight/day (Exon 2006). Throughout the 12 month exposure period, no significant effects were observed in mouse bodyweight or water consumption. It was only after 6 months of exposure that significant increases in DNA damage could be detected using the Comet assay in male germ cells of mice exposed at the highest doses (Pg. 83). By the 12 month time point, significant levels of DNA damage were also found at the 0.01 µg/ml dose, which was equivalent to human exposure estimates. Expression of γH2A.X was also used to examine DNA damage, as γH2A.X is a useful marker of double strand breaks in meiotic germ cells (Mahadevaiah et al. 2001). Whilst control samples exhibited basal γH2A.X staining, due to normal homologous recombination events, a general increase in γH2A.X expression was found in both testis sections and germ cells from acrylamide exposed mice at the 12 month time point (Pg. 84).

The present study was the first to demonstrate reproductive toxic effects of acrylamide at clinically relevant doses, and suggests that human dietary exposure to acrylamide may present a risk to male reproductive health. Additionally, considering that no gross morphological effects were observed in testis histology and spermatogenesis, the damage

120 induced by acrylamide in developing germ cells may persist in mature spermatozoa and propagate to the next generation. Further research is required to ascertain the potential of chronic acrylamide exposure to induce transgenerational effects, and this is discussed in further detail in the future directions section (Pg. 124).

Aim 3: To elucidate the molecular mechanisms by which DNA damage is generated by acrylamide and the response of the male germ line to chronic toxic exposure.

To date, only a few studies have used microarray analyses to examine transcriptional changes induced by acrylamide exposure, the majority of which were based on doses up to 50 fold greater than those described in our chronic exposure model. In a study by Oshida et al. (2011) acrylamide was administered to mice at 50 mg/kg bodyweight for 5 days and regulation of genes associated with spermatogenesis, cytoskeletal remodelling, signalling, and apoptosis were observed in mouse testis. In rats, genes associated with testicular- functions, apoptosis, cellular redox, cell cycle, and nucleic acid-binding were regulated in testis in response to acrylamide at 60 mg/kg bodyweight/day (Yang et al. 2005). Other microarray studies in rodents have focussed on the pituitary-thyroid axis (Bowyer et al. 2008; Camacho et al. 2012), or liver gene expression following acrylamide exposure (Mei et al. 2008; Lee et al. 2012). Hence, the microarray analysis presented in Chapter 4 was unique, in that testicular gene expression was examined at a range of low acrylamide doses and across multiple time points.

Analysis of microarray data at the 1, 6 and 12 months revealed that despite the lack of gross morphological effects in testis histology and spermatogenesis (observed in Chapter 3), widespread transcriptional changes were induced in numerous pathways important for proper male germ cell development. These pathways were associated with spermatogenesis, signal transduction, epoxide metabolism, mitochondrial dysfunction, oxidative stress, cancer, junction signalling, and DNA repair. Interestingly, some of these pathways were similar to those identified in the acute exposure studies described above, such as responses associated with oxidative stress. Indeed, evidence from all three manuscripts presented in this thesis at least partially supported an oxidative mechanism of

121 acrylamide toxicity. DNA damage assessment using the FPG modified Comet assay in the in vitro study and chronic exposure model could be indicative of both oxidative and glycidamide adducts in male germ cells. This together with the identification of pathways involved in glutathione depletion and mitochondrial dysfunction in the testis microarray analysis of exposed mice was suggestive of a role of oxidative stress in acrylamide induced damage in the male.

However, the most striking finding of this study was the dose-dependent regulation of testicular genes that was identified via automated dose-response modelling by ToxResponse Modeler (Burgoon and Zacharewski 2008). This was especially evident at the 1 and 6 month time points, though dose-dependent regulation appeared to be diminished following 12 months of acrylamide exposure. IPA analysis indicated that the top pathways regulated at the earlier time points were primarily associated with signalling; however at 12 months, nucleotide excision repair was the top pathway identified, with remaining pathways relevant to junction signalling and blood/testis barrier dynamics (Pg. 114). Thus, it was speculated that acrylamide exposure in the short-term initiates signal transduction responses in the testis, which become saturated and overwhelmed following prolonged exposure to the compound. This mechanism may explain both the reduced sensitivity of genes to acrylamide exposure at the 12 month time point (Chapter 4), and the accumulation of DNA damage observed in male germ cells of exposed mice (Chapter 3). Particularly as downregulation of nucleotide excision repair and blood-testis barrier signalling pathways was observed at the 12 month time point (Pg. 117).

The other major finding of the microarray analysis was the identification of potential transcriptional markers of acrylamide exposure in the testis, such as Gsg1, Spata16, Ephx1, and Igf2 (Pg. 111). These genes were of interest due to their involvement in spermatogenesis, xenobiotic metabolism or imprinting/development. However, they were also selected because they were identified as dose-dependently regulated at early time points (1 and 6 months) and calculated by ToxResponse Modeler to have low EC50 values, indicative of their sensitive regulation at these time points. Thus the use of dose-response modelling in conjunction with microarray analyses not only provided insight into the mechanisms of chronic acrylamide exposure, but also offered the opportunity to identify testicular biomarkers of chronic acrylamide exposure.

122 Conclusions & Contribution to Current Literature

Based on the recommendations outlined by the FAO and WHO, research regarding acrylamide toxicity has been focussed in a number of key areas. This included strategies to improve human exposure estimates of acrylamide in the diet and reduce the concentrations of acrylamide generated in commercial food products to minimise human exposure (WHO 2011). Progress has been made in this area, and methods such as reduced cooking temperatures, changes in storage conditions and using low sugar/asparagine varieties of potatoes for chips and French fries (which contain the highest concentrations of acrylamide) have been found to significantly reduce the formation of acrylamide (Halford et al. 2012; Parker et al. 2012). However, as the compound forms naturally during the cooking process, elimination of all acrylamide from food is impossible. Furthermore, efforts to reduce concentrations of acrylamide in foods appear to have waned in the last decade (Sanderson 2012). Thus, due to the ubiquity of acrylamide in the food supply, the risk of human dietary exposure to the compound remains.

Acrylamide neurotoxicity is perhaps of lesser concern in regards to human dietary exposure, as these levels are well below those known to induce neurotoxicity in humans (WHO 2002). Understandably, significant emphasis has been placed on the carcinogenic activity of acrylamide, and is the focus of much research regarding acrylamide toxicity. Indeed, several recent exposure studies in rodents have examined cancer incidences in a variety of tissues for up to 2 years (National Toxicology Program (NTP 2012)); and epidemiological studies continue to evaluate the risk of cancer due to the consumption of acrylamide-rich foods (Wilson et al. 2009; Larsson et al. 2009; Hogervorst et al. 2007). However, research into acrylamide neurotoxicity and carcinogenicity also contributes to the understanding of acrylamide reproductive toxicity, as the mechanisms that underpin these effects likely overlap. Indeed, the genotoxic potential of acrylamide and its metabolites is relevant to mechanisms involved in both carcinogenicity and reproductive toxicity (Exon 2006).

The data presented in this thesis expands upon current research by examining acrylamide toxicity in mice within specific stages of spermatogenesis, and following exposures equivalent to human dietary intake. Our results support a genotoxic mode of action of acrylamide and suggest that oxidative damage may play a role in the damage induced by

123 this compound in male germ cells. Furthermore, a chronic exposure model of acrylamide was developed in Swiss mice, in which both dose and time-dependent increases in DNA damage in germ cells were observed. Microarray analysis of this study indicated potential biomarkers of chronic acrylamide exposure, and offered insight into mechanisms that result in deleterious effects in the testis. In conclusion, the outcomes of this research advance the understanding of acrylamide reproductive toxicity in the male, particularly at chronic levels, which certainly will aid in future risk assessment of acrylamide on human reproductive health.

Future Directions

The advantage of the chronic exposure model described in this thesis is that it enables the design of future animal studies using doses established to induce DNA damage in mice. Thus, the future directions of this project include further exposure studies, in which the efficacy of protective agents will be tested against acrylamide induced DNA damage in the male germ line. Indeed, a pilot study in male mice has already been conducted in our laboratory using food supplements, resveratrol and selenium. The use of resveratrol as a food supplement has been reported to have positive effects on health, such as anti-cancer, anti-inflammatory and cardio protection effects (Detampel et al. 2012). Selenium is an essential trace element that functions as a component of antioxidant enzymes, and is also thought to have beneficial antioxidant and anti-carcinogenic properties (Bodnar et al. 2012). Preliminary data suggests that mice exposed to acrylamide at 1 µg/ml for 6 months (the dose at which DNA damage was first detected) and injected with resveratrol have reduced levels of DNA damage in spermatocytes, compared to mice exposed to acrylamide alone (data not shown). However, further experiments are required to confirm the protective activity of resveratrol in the context of acrylamide toxicity.

The microarray analysis in Chapter 4 also opened interesting avenues of research worthy of further investigation. For example, the pathways regulated following chronic acrylamide exposure for 12 months were mostly involved with junction signalling, suggestive of interference with the blood/testis barrier in the exposed mice. As the blood/testis barrier is a major line of defence against xenobiotic exposure in the testis, further study of its

124 integrity in the context of acrylamide toxicity could be examined in future exposure studies. Additionally, a number of genes identified by microarray analysis and dose-response modelling may serve as possible biomarkers of acrylamide exposure in the testis. Testis RNA samples from all 5 time points of the chronic exposure study were retained for Q-PCR confirmation of gene expression, which would enable the verification of candidate genes as acrylamide biomarkers. Indeed, the use of xenobiotic biomarkers should be pursued in humans, as this would aid in exposure assessments and potentially shed light on the causes of infertility in males.

Another important question that needs to be addressed in this project is whether the observed acrylamide-mediated DNA damage in male germ cells is of consequence to the next generation. To examine the transgenerational effects of paternal acrylamide exposure, future experiments will be conducted in mouse breeding studies in conjunction with a classic two-step skin tumour induction protocol (Yuspa 1994; Henderson et al. 2011). This assay involves measuring the development of skin papillomas in the presence of a tumour promoter, and can be used to test for an inherited predisposition to cancer in the F1 and F2 generation. The premise of this experiment will be to establish a link between chronic acrylamide exposure in the paternal male, and an increased mutational load in the unexposed offspring. Thus, these studies will provide confirmation of the extent to which paternal exposure to acrylamide impacts upon future generations.

At present, transmission of toxic effects to offspring are not well understood, thus this research may aid in elucidating how paternal toxic exposure affects the next generation. Further research into the molecular mechanisms of this compound may also shed light on the activity of other genotoxic agents in the male germ line, and contribute to the understanding of xenobiotic exposure and associated male reproductive toxicity.

125 References

Adler, I. D., Baumgartner, A., Gonda, H. and Friedman, M. A. (2000). 1-Aminobenzotriazole inhibits acrylamide-induced dominant lethal effects in spermatids of male mice. Mutagenesis 15, 133-136. Adler, I. D., Gonda, H., Hrabe de Angelis, M., Jentsch, I., Otten, I. S. and Speicher, M. R. (2004). Heritable translocations induced by dermal exposure of male mice to acrylamide. Cytogenet Genome Res 104(1-4), 271-276. Aitken, R. J. and Baker, M. A. (2004). Oxidative stress and male reproductive biology. Reprod Fertil Dev 16(5), 581-588. Amacher, D. E. (2010). The effects of cytochrome P450 induction by xenobiotics on endobiotic metabolism in pre-clinical safety studies. Toxicology mechanisms and methods 20(4), 159-166. Andreassen, A., Steffensen, I. L., Olsen, A. K., Tanaka, K. and Wiger, R. (2011). Spermatogenesis is not impaired in a nucleotide excision repair-deficient min mouse model with or without neonatal mutagen treatment. Journal of andrology 32(5), 541-549. Baarends, W. M., van der Laan, R. and Grootegoed, J. A. (2000). Specific aspects of the ubiquitin system in spermatogenesis. Journal of endocrinological investigation 23(9), 597- 604. Bergmark, E. (1997). Hemoglobin adducts of acrylamide and acrylonitrile in laboratory workers, smokers and nonsmokers. Chem Res Toxicol 10, 78-84. Bishop, J., Chapin, R., Fail, P., George, J., Grizzle, T., Sadler, B., Heindel, J. (1991). Acrylamide induced dominant lethality in mice following low dose chronic administration in drinking water. . Environ. Mol. Mutagen. (abstract) 17(suppl. 9):A25. Bodnar, M., Konieczka, P. and Namiesnik, J. (2012). The properties, functions, and use of selenium compounds in living organisms. Journal of environmental science and health. Part C, Environmental carcinogenesis & ecotoxicology reviews 30(3), 225-252. Bowyer, J. F., Latendresse, J. R., Delongchamp, R. R., Muskhelishvili, L., Warbritton, A. R., Thomas, M., Tareke, E., McDaniel, L. P. and Doerge, D. R. (2008). The effects of subchronic acrylamide exposure on gene expression, neurochemistry, hormones, and histopathology in the hypothalamus-pituitary-thyroid axis of male Fischer 344 rats. Toxicology and applied pharmacology 230(2), 208-215. Bull, R. J., Robinson, M., Laurie, R. D., Stoner, G. D., Greisiger, E., Meier, J. R. and Stober, J. (1984). Carcinogenic effects of acrylamide in sencar and A/J mice. Cancer Res 44(1), 107- 111. Burgoon, L. D. and Zacharewski, T. R. (2008). Automated quantitative dose-response modeling and point of departure determination for large toxicogenomic and high- throughput screening data sets. Toxicological sciences : an official journal of the Society of Toxicology 104(2), 412-418. Camacho, L., Latendresse, J. R., Muskhelishvili, L., Patton, R., Bowyer, J. F., Thomas, M. and Doerge, D. R. (2012). Effects of acrylamide exposure on serum hormones, gene expression, cell proliferation, and histopathology in male reproductive tissues of Fischer 344 rats. Toxicology letters 211(2), 135-143. Cederbaum, A. I., Lu, Y. and Wu, D. (2009). Role of oxidative stress in alcohol-induced liver injury. Arch Toxicol 83(6), 519-548.

126 Chapin, R. E., Fail, P. A., George, E. L., Grizzle, T. B., Heindel, J. J., Harry, G. J., Collins, B. J. and Teague, J. (1995). The reproductive and neural toxicities of acrylamide and three analogues in swiss mice, evaluated using the continuous breeding protocol. Fundam. Appl. toxicol. 27, 9-24. Clement, F. C., Dip, R. and Naegeli, H. (2007). Expression profile of human cells in culture exposed to glycidamide, a reactive metabolite of the heat-induced food carcinogen acrylamide. Toxicology 240, 111–124. Dearfield, K. L., Douglas, G. R., Ehling, U. H., Moore, M. M., Sega, G. A. and Brusick, D. J. (1995). Acrylamide: A review of its genotoxicity and an assessment of heritable genetic risk. Mutat. Res. 330, 71-99. Detampel, P., Beck, M., Krahenbuhl, S. and Huwyler, J. (2012). Drug interaction potential of resveratrol. Drug metabolism reviews 44(3), 253-265. Doroshyenko, O., Fuhr, U., Kunz, D., Frank, D., Kinzig, M., Jetter, A., Reith, Y., Lazar, A., Taubert, D., Kirchheiner, J., Baum, M., Eisenbrand, G., Berger, F. I., Bertow, D., Berkessel, A., Sorgel, F., Schomig, E. and Tomalik-Scharte, D. (2009). In vivo role of cytochrome P450 2E1 and glutathione-S-transferase activity for acrylamide toxicokinetics in humans. Cancer Epidemiol Biomarkers Prev 18(2), 433-443. Ehling, U. H. and Neuhauser-Klaus, A. (1992). Reevaluation of the induction of specific-locus mutations in spermatogonia of the mouse by acrylamide. Mutat Res 283(3), 185-191. Emery, B. R. and Carrell, D. T. (2006). The effect of epigenetic sperm abnormalities on early embryogenesis. Asian J Androl 8(2), 131-142. Exon, J. H. (2006). A review of the toxicology of acrylamide. J. Toxicol. Environ. Health B. Crit. Rev. 9(5), 397-412. FAO/WHO (2003). Food and Agriculture Organization and World Health Organization. Joint FAO/WHO Food Standards Programme Codex Commitee on Food Additives and Contaminants. Discussion paper on acrylamide. Rotterdam, The Netherlands, March. FAO/WHO (2005). Joint FAO/WHO Expert Commitee on Food Additives. Sixty-fourth meeting Rome, 8-17 February 2005 ftp://ftp.fao.org/es/esn/jecfa/jecfa64_summary.pdf. Favor, J. and Shelby, M. D. (2005). Transmitted mutational events induced in mouse germ cells following acrylamide or glycidamide exposure. Mutat Res 580(1-2), 21-30. Fennell, T. R., Sumner, S. C., Snyder, R., Burgess, J., Spicer, R., Bridson, W. E. and Freeman, M. A. (2005). Metabolism and hemoglobin adduct formation of acrylamide in humans. Toxicol. Sci. 85, 447-459. Feychting, M., Plato, N., Nise, G. and Ahlbom, A. (2001). Paternal occupational exposures and childhood cancer. Environ Health Perspect 109(2), 193-196. Friedman, M. and Levin, C. E. (2008). Review of methods for the reduction of dietary content and toxicity of acrylamide. J Agric Food Chem 56(15), 6113-6140. Friedman, M. A., Dulak, L. H. and Stedham, M. A. (1995). A lifetime oncogenicity study in rats with acrylamide. Fundam. Appl. Toxicol. 27, 95-105. Friedman, M. A., Zeiger, E., Marroni, D. E. and Sickles, D. W. (2008). Inhibition of rat testicular nuclear kinesins (krp2; KIFC5A) by acrylamide as a basis for establishing a genotoxicity threshold. J Agric Food Chem 56(15), 6024-6030. Fuhr, U., Boettcher, M. I., Kinzig-Schippers, M., Weyer, A., Jetter, A., Lazar, A., Taubert, D., Tomalik-Scharte, D., Pournara, P., Jakob, V., Harlfinger, S., Klaassen, T., Berkessel, A., Angerer, J., Sorgel, F. and Schomig, E. (2006). Toxicokinetics of acrylamide in humans after ingestion of a defined dose in a test meal to improve risk assessment for acrylamide carcinogenicity. Cancer Epidemiol Biomarkers Prev 15(2), 266-271.

127 Gamboa da Costa, G., Churchwell, M. I., Hamilton, L. P., Von Tungeln, L. S., Beland, F. A., Marques, M. M. and Doerge, D. R. (2003). DNA adduct formation from acrylamide via conversion to glycidamide in adult and neonatal mice. Chem Res Toxicol 16(10), 1328-1337. Generoso, E. E., Sega, G. A., Lockhardt, A. M., Hughes, L. A., Cain, K. T., Cacheiro, N. L. A. and Shelby, M. D. (1996). Dominant lethal mutations, heritable translocations, and unscheduled DNA synthesis induced in male mouse germ cells by glycidamide, a metabolite of acrylamide. Mutation Research 371, 521-530. Ghanayem, B. I., McDaniel, L. P., Churchwell, M. I., Twaddle, N. C., Snyder, R., Fennell, T. R. and Doerge, D. R. (2005a). Role of CYP2E1 in the epoxidation of acrylamide to glycidamide and formation of DNA and hemoglobin adducts. Toxicol Sci 88(2), 311-318. Ghanayem, B. I., Witt, K. L., El-Hadri, L., Hoffler, U., Kissling, G. E., Shelby, M. D. and Bishop, J. B. (2005b). Comparison of germ cell mutagenicity in male CYP2E1-null and wild-type mice treated with acrylamide: evidence supporting a glycidamide-mediated effect. Biol Reprod 72(1), 157-163. Gilbert, S. F. (2000). Developmental Biology 6th Edition. Ch. 19 'The saga of the germ line'. Sinauer Assoc. Inc. USA, 599-601. Guengerich, F. P. (2008). Cytochrome p450 and chemical toxicology. Chem Res Toxicol 21(1), 70-83. Hagmar, L., Tornqvist, M., Nordander, C., Rosen, I., Bruze, M., Kautiainen, A., Magnusson, A. L., Malmberg, B., Aprea, P., Granath, F. and Axmon, A. (2001). Health effects of occupational exposure to acrylamide using hemoglobin adducts as biomarkers of internal dose. Scand. J. Work Environ. Health 27, 219-226. Halford, N. G., Muttucumaru, N., Powers, S. J., Gillatt, P. N., Hartley, L., Elmore, J. S. and Mottram, D. S. (2012). Concentrations of free amino acids and sugars in nine potato varieties: effects of storage and relationship with acrylamide formation. Journal of agricultural and food chemistry 60(48), 12044-12055. Hansen, S. H., Olsen, A. K., Soderlund, E. J. and Brunborg, G. (2010). In vitro investigations of glycidamide-induced DNA lesions in mouse male germ cells and in mouse and human lymphocytes. Mutat Res 696(1), 55-61. Hashimoto, K., Sakamoto, J. and Tanii, h. (1981). Neurotoxicity of acrylamide and related compounds and their effects on male gonads in mice. Arch Toxicol 47, 179-189. Healy, L. N., Pluta, L. J. and Recio, L. (1999). Expression and distribution of cytochrome P450 2E1 in B6C3F1 mouse liver and testes. Chemico-biological interactions 121(2), 199-207. Henderson, C. J., Ritchie, K. J., McLaren, A., Chakravarty, P. and Wolf, C. R. (2011). Increased skin papilloma formation in mice lacking glutathione transferase GSTP. Cancer Res 71(22), 7048-7060. Hogervorst, J. G., Schouten, L. J., Konings, E. J., Goldbohm, R. A. and van den Brandt, P. A. (2007). A prospective study of dietary acrylamide intake and the risk of endometrial, ovarian, and breast cancer. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 16(11), 2304-2313. Holstein, A. F., Schulze, W. and Davidoff, M. (2003). Understanding spermatogenesis is a prerequisite for treatment. Reprod Biol Endocrinol 1, 107. IARC (1994). International Agency for Research on Cancer. Some industrial chemicals. Lyon: IARC; IARC Monographs on the evaluation of carcinogenic risks to human. 60. 389-433. Jansen, J., Olsen, A. K., Wiger, R., Naegeli, H., de Boer, P., van Der Hoeven, F., Holme, J. A., Brunborg, G. and Mullenders, L. (2001). Nucleotide excision repair in rat male germ cells:

128 low level of repair in intact cells contrasts with high dual incision activity in vitro. Nucleic Acids Res 29(8), 1791-1800. Johnson, K. A., Gorzinski, S. J., Bodner, K. M., Campbell, R. A., Wolf, C. H., Friedman, M. A. and Mast, R. W. (1986). Chronic toxicity and oncogenicity study on acrylamide incorporated in the drinking water of Fischer 344 rats. Toxicol Appl Pharmacol 85(2), 154-168. Klug, W. S. and Cummings, M. R. (2003). Genetics, A Molecular Perspective. Pearson Education Inc. USA, Ch 2, Pg 30. Larsson, S. C., Akesson, A., Bergkvist, L. and Wolk, A. (2009). Dietary acrylamide intake and risk of colorectal cancer in a prospective cohort of men. European journal of cancer 45(4), 513-516. Leduc, F., Nkoma, G. B. and Boissonneault, G. (2008). Spermiogenesis and DNA repair: a possible etiology of human infertility and genetic disorders. Systems biology in reproductive medicine 54(1), 3-10. Lee, I. P. and Dixon, R. L. (1978). Factors influencing reproduction and genetic toxic effects on male gonads. Environ Health Perspect 24, 117-127. Lee, T., Manjanatha, M. G., Aidoo, A., Moland, C. L., Branham, W. S., Fuscoe, J. C., Ali, A. A. and Desai, V. G. (2012). Expression analysis of hepatic mitochondria-related genes in mice exposed to acrylamide and glycidamide. Journal of toxicology and environmental health. Part A 75(6), 324-339. Lopachin, R. M. (2004). The changing view of acrylamide neurotoxicity. Neurotoxicology 25, 617-630. Lopachin, R. M. and Gavin, T. (2008). Acrylamide-induced nerve terminal damage: relevance to neurotoxic and neurodegenerative mechanisms. J Agric Food Chem 56(15), 5994-6003. Mahadevaiah, S. K., Turner, J. M., Baudat, F., Rogakou, E. P., de Boer, P., Blanco-Rodriguez, J., Jasin, M., Keeney, S., Bonner, W. M. and Burgoyne, P. S. (2001). Recombinational DNA double-strand breaks in mice precede synapsis. Nature genetics 27(3), 271-276. Marchetti, F. and Wyrobek, A. J. (2005). Mechanisms and consequences of paternally- transmitted chromosomal abnormalities. Birth Defects Res C Embryo Today 75(2), 112-129. Marlowe, C., Clark, M. J., Mast, R. W., Friedman, M. A. and Waddell, W. J. (1986). The distribution of [14C]acrylamide in male and pregnant Swiss-Webster mice studied by whole- body autoradiography. Toxicol. Appl. Pharmacol. 86, 457-465. Mattsson, J. L. (2007). Mixtures in the real world: the importance of plant self-defense toxicants, mycotoxins, and the human diet. Toxicology and applied pharmacology 223(2), 125-132. Mei, N., Guo, L., Tseng, J., Dial, S. L., Liao, W. and Manjanatha, M. G. (2008). Gene expression changes associated with xenobiotic metabolism pathways in mice exposed to acrylamide. Environ Mol Mutagen. Miller, M. S. and Spencer, P. S. (1985). The mechanisms of acrylamide axonopathy. Annu. Rev. Pharmacol. Toxicol. 25, 643-666. Mucci, L. A. and Adami, H. O. (2009). The plight of the potato: is dietary acrylamide a risk factor for human cancer? J Natl Cancer Inst 101(9), 618-621. Nebert, D. W. and Russell, D. W. (2002). Clinical importance of the cytochromes P450. Lancet 360(9340), 1155-1162. NTP (2005). NTP-CERHR monograph on the potential human reproductive and developmental effects of acrylamide. NTP CERHR MON(14), v-I-2, II-xi-166, III-161-174.

129 NTP (2012). Toxicology and carcinogenesis studies of acrylamide (CASRN 79-06-1) in F344/N rats and B6C3F1 mice (feed and drinking water studies). National Toxicology Program technical report series(575), 1-234. O'Flynn O'Brien, K. L., Varghese, A. C. and Agarwal, A. (2010). The genetic causes of male factor infertility: a review. Fertil Steril 93(1), 1-12. Olsen, A. K., Andreassen, A., Singh, R., Wiger, R., Duale, N., Farmer, P. B. and Brunborg, G. (2010). Environmental exposure of the mouse germ line: DNA adducts in spermatozoa and formation of de novo mutations during spermatogenesis. PloS one 5(6), e11349. Olsen, A. K., Lindeman, B., Wiger, R., Duale, N. and Brunborg, G. (2005). How do male germ cells handle DNA damage? Toxicol Appl Pharmacol 207(2 Suppl), 521-531. Oshida, K., Hirakata, M., Maeda, A., Miyoshi, T. and Miyamoto, Y. (2011). Toxicological effect of emodin in mouse testicular gene expression profile. Journal of applied toxicology : JAT 31(8), 790-800. Parker, J. K., Balagiannis, D. P., Higley, J., Smith, G., Wedzicha, B. L. and Mottram, D. S. (2012). Kinetic model for the formation of acrylamide during the finish-frying of commercial french fries. Journal of agricultural and food chemistry 60(36), 9321-9331. Parzefall, W. (2008). Minireview on the toxicity of dietary acrylamide. Food Chem Toxicol 46(4), 1360-1364. Pavek, P. and Dvorak, Z. (2008). Xenobiotic-induced transcriptional regulation of xenobiotic metabolizing enzymes of the cytochrome P450 superfamily in human extrahepatic tissues. Curr Drug Metab 9(2), 129-143. Raven, P. H., Johnson, G. B., Losos, J. B. and Singer, S. R. (2005). Biology. McGraw-Hill Companies, Inc. 7th Edition. Pg. 1069. Regidor, E., Ronda, E., Garcia, A. M. and Dominguez, V. (2004). Paternal exposure to agricultural pesticides and cause specific fetal death. Occup Environ Med 61(4), 334-339. Rice, J. M. (2005). The carcinogenicity of acrylamide. Mutat Res 580(1-2), 3-20. Russell, L. B., Hunsicker, P. R., Cacheiro, N. L. and Generoso, W. M. (1991). Induction of specific-locus mutations in male germ cells of the mouse by acrylamide monomer. Mutat Res 262(2), 101-107. Sakamoto, J. and Hashimoto, K. (1986). Reproductive toxicity of acrylamide and related compounds in mice--effects on fertility and sperm morphology. Arch Toxicol 59(4), 201-205. Sanderson, K. (2012). Bid to curb fried-food chemical goes cold. Nature 491(7422), 22-23. Savitz, D. A. and Chen, J. H. (1990). Parental occupation and childhood cancer: review of epidemiologic studies. Environ Health Perspect 88, 325-337. Schumacher, J. N., Green, C. R., Best, F. W. and Newell, M. P. (1977). Smoke composition. An extensive investigation of the water-soluble portion of cigarette smoke. J. Agric. Food Chem. 25, 310-320. Sega, G. A., Alcota, R. P., Tancongco, C. P. and Brimer, P. A. (1989). Acrylamide binding to the DNA and protamine of spermiogenic stages in the mouse and its relationship to genetic damage. Mutat Res 216(4), 221-230. Sega, G. A., Generoso, E. E. and Brimer, P. A. (1990). Acrylamide exposure induces a delayed unscheduled DNA synthesis in germ cells of male mice that is correlated with the temporal pattern of adduct formation in testis DNA. Environmental and Molecular Mutagenesis 16 (3), 137-142. Settels, E., Bernauer, U., Palavinskas, R., Klaffke, H. S., Gundert-Remy, U. and Appel, K. E. (2008). Human CYP2E1 mediates the formation of glycidamide from acrylamide. Arch Toxicol 82(10), 717-727.

130 Shalet, S. M. (2009). Normal testicular function and spermatogenesis. Pediatr Blood Cancer 53(2), 285-288. Shelby, M. D. (1996). Selecting chemicals and assays for assessing mammalian germ cell mutagenicity. Mutat Res 352(1-2), 159-167. Shelby, M. D., Cain, K. T., Cornett, C. V. and Generoso, W. M. (1987). Acrylamide: induction of heritable translocation in male mice. Environ Mutagen 9(4), 363-368. Shelby, M. D., Cain, K. T., Hughes, L. A., Braden, P. W. and Generoso, W. M. (1986). Dominant lethal effects of acrylamide in male mice. Mutat Res 173(1), 35-40. Shipp, A., Lawrence, G., Gentry, R., McDonald, T., Bartow, H., Bounds, J., Macdonald, N., Clewell, H., Allen, B. and Van Landingham, C. (2006). Acrylamide: review of toxicity data and dose-response analyses for cancer and noncancer effects. Crit Rev Toxicol 36(6-7), 481-608. Sickles, D. W., Sperry, A. O., Testino, A. and Friedman, M. (2007). Acrylamide effects on kinesin-related proteins of the mitotic/meiotic spindle. Toxicol. Appl. Pharmacol. 222, 111- 121. Sickles, D. W., Stone, J. D. and Friedman, M. A. (2002). Fast axonal transport: a site of acrylamide neurotoxicity? Neurotoxicology 23(2), 223-251. Smith, M. K., Zenick, H., Preston, R. J., George, E. L. and Long, R. E. (1986). Dominant lethal effects of subchronic acrylamide administration in the male Long-Evans rat. Mutat Res 173(4), 273-277. Sorgel, F., Weissenbacher, R., Kinzig-Schippers, M., Hofmann, A., Illauer, M., Skott, A. and Landersdorfer, C. (2002). Acrylamide: increased concentrations in homemade food and first evidence of its variable absorption from food, variable metabolism and placental and breast milk transfer in humans. Chemotherapy 48(6), 267-274. Sublet, V. H., Zenick, H. and Smith, M. K. (1989). Factors associated with reduced fertility and implantation rates in females mated to acrylamide-treated rats. Toxicology 55(1-2), 53- 67. Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S. and Tornqvist, M. (2000). Acrylamide: a cooking carcinogen? Chem Res Toxicol 13(6), 517-522. Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S. and Tornqvist, M. (2002). Analysis of acrylamide, a carcinogen formed in heated foodstuffs. J Agric Food Chem 50(17), 4998-5006. Tichy, E. D. and Stambrook, P. J. (2008). DNA repair in murine embryonic stem cells and differentiated cells. Exp Cell Res 314(9), 1929-1936. Trasler, J. M. (2009). Epigenetics in spermatogenesis. Mol Cell Endocrinol 306(1-2), 33-36. Tyl, R. W. and Friedman, M. A. (2003). Effects of acrylamide on rodent reproductive performance. Reprod Toxicol 17(1), 1-13. Tyl, R. W., Friedman, M. A., Losco, P. E., Fisher, L. C., Johnson, K. A., Strother, D. E. and Wolf, C. H. (2000a). Rat two-generation reproduction and dominant lethal study of acrylamide in drinking water. Reprod. Toxicol. 14, 385-401. Tyl, R. W., Marr, M. C., Myers, C. B., Ross, W. P. and Friedman, M. A. (2000b). Relationship between acrylamide reproductive and neurotoxicity in male rats. Reprod Toxicol 14(2), 147- 157. Wang, Z., Lu, S., Liu, C., Yao, N., Zhang, C. and Ma, X. (2010). Acrylamide disturbs genomic imprinting during spermatogenesis. Front Biosci (Elite Ed) 2, 165-170. WHO (2002). FAO/WHO Consultation on the Health Implications of Acrylamide in Food. Summary Report of a meeting held in Geneva, 25-27 June 2002. (http://www.who.int/foodsafety/publications/chem/en/acrylamide_full.pdf).

131 WHO (2011). Evaluation of certain contaminants in food. World Health Organization technical report series(959), 1-105, back cover. WHO/JECFA (2005). Joint FAO/WHO Expert Commitee on Food Additives. Sixty-fourth meeting Rome, 8-17 February 2005 ftp://ftp.fao.org/es/esn/jecfa/jecfa64_summary.pdf. Wilson, K. M., Mucci, L. A., Cho, E., Hunter, D. J., Chen, W. Y. and Willett, W. C. (2009). Dietary acrylamide intake and risk of premenopausal breast cancer. American journal of epidemiology 169(8), 954-961. Wise, L. D., Gordon, L. R., Soper, K. A., Duchai, D. M. and Morrissey, R. E. (1995). Developmental neurotoxicity evaluation of acrylamide in Sprague-Dawley rats. Neurotoxicol Teratol 17(2), 189-198. Yang, H. J., Lee, S. H., Jin, Y., Choi, J. H., Han, D. U., Chae, C., Lee, M. H. and Han, C. H. (2005). Toxicological effects of acrylamide on rat testicular gene expression profile. Reprod Toxicol 19(4), 527-534. Yuspa, S. H. (1994). The pathogenesis of squamous cell cancer: lessons learned from studies of skin carcinogenesis--thirty-third G. H. A. Clowes Memorial Award Lecture. Cancer Res 54(5), 1178-1189. Zenick, H., Hope, E. and Smith, M. K. (1986). Reproductive toxicity associated with acrylamide treatment in male and female rats. J Toxicol Environ Health 17(4), 457-472.

132

Appendices

133 Appendix A: Chapter 2 Supplementary Data

Manuscript: Mouse Spermatocytes Express Cyp2e1 and Respond to Acrylamide Exposure.

sTable 1. Primers used for QPCR including their product length and annealing temperatures for cyclophilin, hypoxanthine guanine phosphoribosyl transferase (Hprt), cytochrome P450 Cyp2e1 and cytochrome P450 Cyp1b1.

Product Annealing Gene Forward Primer Reverse Primer Length Accession No. Temp. (°C) (bp)

Cyclophilin CGTCTCCTTCGAGCTGTTT ACCCTGGCACATGAATCCT 100 50-60 NM_008907.1

Hprt GTCATGAAGGAGATGGGAG ATCTACAGTCATAGGAATGG 137 50-60 NM_0.13556.2

Cyp2e1 CGCATGGAACTGTTTCTGC CAATTGTAACAGGGCTGAGGTC 100 59 NM_021282.2

Cyp1b1 GTCTGTGAATCATGACCCAGC ACAGTTCCTCACCGATGCAC 152 59.4 NM_009994.1

134 Appendix A: Chapter 2 Supplementary Data Manuscript: Mouse Spermatocytes Express Cyp2e1 and Respond to Acrylamide Exposure.

sFig.1. A) Diagram outlining where male germ cells reside in the seminiferous tubule at different stages of spermatogenesis in the mouse. Also depicted are the supporting Sertoli cells and the tight junctions between Sertoli cells, which form the blood/testis barrier. Section of untreated mouse testis (shown on right) was stained with FITC-PNA, which labels the developing acrosome (green) and counter-stained with PI (red) to distinguish between different male germ cell types in the testis. B) Tubule staging in a mouse testis section stained with hematoxylin and eosin stain as identified according to the stages for the production of spermatozoa in the mouse seminiferous epithelium (Russel et al., 1990). Based on visual inspection, the tubule presented in (A) is in stage V – VIII, indicating that from the basement membrane to the lumen, germ cells can be identified as type B spermatogonia, pachytene spermatocytes, step 5 to 6 round spermatids and step 15 elongating spermatids.

135 Appendix A: Chapter 2 Supplementary Data Manuscript: Mouse Spermatocytes Express Cyp2e1 and Respond to Acrylamide Exposure.

sFig. 2: Q-PCR analysis of Cyp2e1 mRNA expression normalised to the geometric mean of reference genes cyclophiln and hprt in mouse testis at different developmental stages 2, 6, 11, 14, 18, 22, and 36 d after birth, and adult (older than 56 d). Normalisation to two reference genes produced a similar Cyp2e1 expression profile in developing testis, validating the data presented in Fig. 1. Data are representative of n = 2 experiments and depicted as transformed values, 2e-∆C(t) (Mean +SEM).

Capital letters indicate statistically significant differences between groups (F7,34 = 4.4, p < 0.05).

136 Appendix A: Chapter 2 Supplementary Data Manuscript: Mouse Spermatocytes Express Cyp2e1 and Respond to Acrylamide Exposure.

sFig. 3: Optimisation of FPG and hOGG1 in the Comet assay, conducted on spermatocytes treated with glycidamide at 50 nM, 0.5 µM, 5 µM for 18 h, or H2O2 at 50nM, 500nM, 5µM for 5 min. (A) Enzyme activity of FPG was confirmed as the addition of FPG increased detection of DNA damage in glycidamide treated cells. Enzyme activity of hOGG1 could be confirmed as increases in DNA damage were detected following hOGG1 treatment at 1:100 and 1:500, but not at 1:1000 dilution in glycidamide treated cells. (B) Both FPG and hOGG1 detected similar levels of damage in H2O2 treated cells at the concentrations of used in our study (1:1000 and 1:500 respectively). However increased hOGG1 concentration (1:100) detected more damage than FPG, demonstrating not only the activity of the hOGG1 enzyme but also the specificity of hOGG1 for oxidative damage.

137 Appendix A: Chapter 2 Supplementary Data Manuscript: Mouse Spermatocytes Express Cyp2e1 and Respond to Acrylamide Exposure.

sFig. 4. (A) Frequency of ‘hedgehog’ comets observed in the Comet assay ± FPG enzyme (Fig. 5 B). (B) Frequency of ‘hedgehog’ comets observed in the Comet assay + FPG enzyme, with increasing concentrations of acrylamide or (C) glycidamide (Fig. 5 C, D). The hydrogen peroxide positive control treatment for DNA damage had the highest frequency of ‘hedgehog’ comets; however, no statistically significant differences were found across all samples.

138 Appendix A: Chapter 2 Supplementary Data Manuscript: Mouse Spermatocytes Express Cyp2e1 and Respond to Acrylamide Exposure.

sFig. 5. Frequency of ‘hedgehog’ comets observed in the Comet assay ± hOGG1 enzyme (Fig. 6A). The hydrogen peroxide positive control treatment for DNA damage had the highest frequency of ‘hedgehog’ comets; however, no statistically significant differences were found across all samples.

139 Appendix B: Chapter 3 Supplementary Data

Manuscript: Chronic Exposure to Acrylamide Induces DNA Damage in Male Germ Cells of Mice.

sTABLE 1: Mean daily water consumption of male mice during 12 months of acrylamide exposure a. No statistically significant differences were found in water consumption as a result of acrylamide exposure.

1 month* 3 months* 6 months 9 months 12 months Exposure (µg/ml) Mean (ml) SD Mean (ml) SD Mean (ml) SD Mean (ml) SD Mean (ml) SD

Control 7.6 2.2 9.7 2.3 6.4 0.2 5.7 0.8 5.9 0.2 0.001 8.5 1.0 8.4 0.8 6.5 0.3 7.1 0.5 6.1 1.0 0.01 9.1 0.5 8.8 3.5 7.3 0.3 6.6 0.6 5.4 0.8 0.1 6.6 3.4 8.7 0.7 6.4 0.2 6.2 0.3 7.7 0.8 1 7.1 0.3 9.5 2.1 6.8 0.3 6.5 0.2 8.5 0.3 10 7.5 0.9 8.9 2.3 6.1 0.2 7.6 0.2 7.0 0.2

a There were 6 mice in each treatment group. b Daily water consumption was recorded for each cage and divided by the number of mice in one cage (maximum of six individuals per cage).

*Water consumption was slightly higher in the 1 and 3 month time points (by 0.6 - 3.6 ml) compared to all other time points (p < 0.05) however acrylamide treatment did not significantly affect mouse water consumption. Detailed statistical analyses of mouse water consumption shown in supplementary data (sTable 4).

140 Appendix B: Chapter 3 Supplementary Data

Manuscript: Chronic Exposure to Acrylamide Induces DNA Damage in Male Germ Cells of Mice.

sTABLE 2: Mean testis weights and testis to bodyweight ratios following 12 months of acrylamide exposure in male mice a. No statistically significant differences were found in testis weights as a result of acrylamide exposure.

1 month 3 months 6 months 9 months 12 months Exposure Meanb Te:BWc Mean Te:BW Mean Te:BW Mean Te:BW Mean Te:BW (µg/ml) SDd SD SD SD SD (mg) Ratio (mg) Ratio (mg) Ratio (mg) Ratio (mg) Ratio

0 (Control) 124 0.67 0.08 123 0.61 0.10 142 0.66 0.14 134 0.57 0.06 114 0.54 0.07 0.001 134 0.67 0.15 132 0.64 0.17 133 0.57 0.08 142 0.68 0.15 139 0.51 0.18 0.01 141 0.70 0.16 124 0.54 0.12 135 0.56 0.12 121 0.50 0.06 122 0.51 0.04 0.1 124 0.64 0.07 128 0.61 0.06 143 0.66 0.12 157 0.73 0.07 136 0.61 0.04 1 108 0.54 0.09 136 0.56 0.09 164 0.73 0.13 128 0.56 0.10 134 0.53 0.20 10 136 0.65 0.09 150 0.68 0.11 127 0.59 0.12 148 0.61 0.11 155 0.67 0.08 a There were six mice in each treatment group. b Average weight of both left and right testis. c The average testis to bodyweight ratio (Te:BW). Expressed as the [ratio of absolute testicular weight to body weight] x 100. d Standard deviation of testis to bodyweight ratio. No significant differences in mouse testis to bodyweight ratios were found.

141 Appendix B: Chapter 3 Supplementary Data

Manuscript: Chronic Exposure to Acrylamide Induces DNA Damage in Male Germ Cells of Mice. sTABLE 3: Statistical analysis of mouse bodyweight (Table 1). A significant difference across groups was found using the Kruskal-Wallis test with respect to exposure time (months) and acrylamide dose (μg/ml) (p < 0.05). A Steel-Dwass multiple comparisons test was used to determine which pairs of groups were statistically significant. In a two-way mixed model ANOVA however, significant differences were found only for time, and the interaction of time and dose; not for dose alone. In other words, the factor of dose did not have a significant effect on mouse bodyweight, except when it was associated with the factor of time. Power analysis showed the experiment had almost maximum power (power = 1) to detect small differences for time and the interaction of time and dose.

Mouse bodyweight (g) by Time (months) Mouse bodyweight (g) by Dose (µg/ml)

Kruskal Wallis Test Kruskal Wallis Test Test-statistic 49.83 Test-statistic 16.12 Degrees of Freedom 4 Degrees of Freedom 5 p - value <.0001* p - value 0.0065*

Steel-Dwass Multiple Comparisons Test Steel-Dwass Multiple Comparisons Test Months comparisons p - value Dose (µg/ml) comparisons p - value 6 - 1 <.0001* 0.01 - 0 0.0032* 9 - 1 <.0001* 1 - 0 0.0286* 12 - 1 <.0001* 0.01 - 0 0.0448* 3 - 1 0.0002* 10 - 0.001 0.1487 9 - 3 0.1473 1 - 0.1 0.2052 12 - 3 0.1321 0.01 - 0.001 0.26 6 - 3 0.4438 10 - 0 0.5543 12 - 6 0.5668 0.001 - 0.1 0.6023 9 - 6 0.8358 1 - 0.001 0.6549 12 - 9 0.9813 0.01 - 0 0.8098 10 - 0.001 0.9524 0.001 - 0.1 0.9592 0.1 - 0 0.9824 1 - 10 0.9885 0.01 - 1 0.9901 Two-way Mixed Model ANOVA for Mouse bodyweight (g) F Ratio 5.12 Degrees of Freedom 29 R Square value 0.52 p - value <.0001*

Effect Tests Degrees of Freedom F Ratio Prob > F Time (months) 4 6.35 0.0001* Dose (ug/ml) 5 1.22 0.30 Time (months)*Dose (ug/ml) 20 1.89 0.0176*

Power Analysis Power Time (months) 0.99 Dose (ug/ml) 0.42 142 Time (months)*Dose (ug/ml) 0.97 Appendix B: Chapter 3 Supplementary Data

Manuscript: Chronic Exposure to Acrylamide Induces DNA Damage in Male Germ Cells of Mice. sTABLE 4: Statistical analysis of mouse daily water consumption (sTable 1). A significant difference across groups was found using the Kruskal-Wallis test with respect to time (months) (p < 0.05) but not for acrylamide dose (μg/ml). Steel-Dwass multiple comparisons test was used to determine which pairs of groups were statistically significant, all of which were between the 1 and 3 month time points compared to other time points. This was due to the change in monitoring schedule after three months (detailed in the Materials and Methods section) rather than an effect of time on water consumption. A two-way mixed model ANOVA indicated significant differences among groups for the factors of time, dose, and the interaction between time and dose. However, these p-values also related to the differences between the 1 and 3 month time points compared with other time points (determined by Tukey Kramer multiple comparisons test showing all 436 comparisons, data not shown). Power analysis showed the experiment had almost maximum power (power = 1) to detect small differences for time, dose, and the interaction of time and dose.

Daily water consumption (ml) by Time (months) Daily water consumption (ml) by Dose (µg/ml) Kruskal Wallis Test Kruskal Wallis Test Test-statistic 56.87 Test-statistic 8.45 Degrees of Freedom 4 Degrees of Freedom 5 p - value <.0001* p - value 0.13

Steel-Dwass Multiple Comparisons Test Steel-Dwass Multiple Comparisons Test Months comparisons p - value Dose (µg/ml) comparisons p - value 3 - 1 0.0576 1 - 0 0.112 9 - 6 0.9824 10 - 0 0.1792 12 - 6 0.9999 0.001 - 0 0.2194 12 - 9 1 0.01 - 0 0.6178 12 - 1 0.0427* 1 - 0.1 0.6797 9 - 1 0.0040* 0.1 - 0 0.7987 6 - 1 0.0004* 10 - 0.1 0.9965 12 - 3 <.0001* 1 - 0.001 1 9 - 3 <.0001* 10 - 0.01 1 6 - 3 <.0001* 1 - 0.01 1 10 - 0.001 1 10 - 1 1 0.01 - 0.001 1 0.1 - 0.01 0.9782 0.1 - 0.001 0.8527 Two-way Mixed Model ANOVA for Daily water consumption (ml) F Ratio 4.78 Degrees of Freedom 29 R Square value 0.50 p - value <.0001*

Effect Tests Degrees of Freedom F Ratio Prob > F Time (months) 4 9.14 <.0001* Dose (ug/ml) 5 2.80 0.0193* Time (months)*Dose (ug/ml) 20 2.13 0.0057*

Power Analysis Power Time (months) 1.00 Dose (ug/ml) 0.82 Time (months)*Dose (ug/ml) 0.99 143 Appendix B: Chapter 3 Supplementary Data

Manuscript: Chronic Exposure to Acrylamide Induces DNA Damage in Male Germ Cells of Mice. sTABLE 5: Statistical analysis of DNA damage (Comet Tail DNA %, Figure 3A). A significant difference across groups was found using the Kruskal-Wallis test with respect to time (months) and acrylamide dose (μg/ml) (p < 0.05). In the Steel-Dwass multiple comparisons test almost all pairs of groups were found to be statistically significant.

A two-way mixed model ANOVA, indicated significant differences among groups for time and the interaction between time and acrylamide dose. R square value indicates that the model describes 46% of the variation within the data. Power analysis showed the experiment had almost maximum power (power = 1) to detect small differences for time and the interaction of time and dose.

DNA Damage (Tail DNA %) by Time (months) DNA Damage (Tail DNA %) by Dose (µg/ml) Kruskal Wallis Test Kruskal Wallis Test Test-statistic 887.65 Test-statistic 513.36 Degrees of Freedom 4 Degrees of Freedom 5 p - value <.0001* p - value <.0001*

Steel-Dwass Multiple Comparisons Test Steel-Dwass Multiple Comparisons Test Months comparisons p - value Dose (µg/ml) comparisons p - value 12 - 3 <.0001* 10 - 0 <.0001* 12 - 1 <.0001* 10 - 0.001 <.0001* 9 - 3 <.0001* 1 - 0 <.0001* 6 - 3 <.0001* 0.1 - 0 <.0001* 12 - 9 <.0001* 1 - 0.001 <.0001* 12 - 6 <.0001* 0.01 - 0 <.0001* 6 - 1 <.0001* 10 - 0.01 <.0001* 9 - 1 <.0001* 10 - 0.1 <.0001* 9 - 6 0.0178* 0.1 - 0.001 <.0001* 3 - 1 <.0001* 10 - 1 <.0001* 0.01 - 0.001 <.0001* 0.001 - 0 <.0001* 1 - 0.01 <.0001* 1 - 0.1 0.006 0.1 - 0.01 0.5388 Two-way Mixed Model ANOVA for DNA Damage (Tail DNA %) F Ratio 183.97 Degrees of Freedom 29 R Square value 0.46 p - value <.0001*

Effect Tests Degrees of Freedom F Ratio Prob > F Time (months) 5 6.95 <.0001* Dose (ug/ml) 4 0 1 Time (months)*Dose (ug/ml) 20 99.72 <.0001*

Power Analysis Power Time (months) 0.05 Dose (ug/ml) 0.9987 Time (months)*Dose (ug/ml) 1 144 Appendix C: Chapter 4 Supplementary Data

Manuscript: The Impact of 12-Month Chronic Acrylamide Exposure on Mouse Testicular Gene Expression.

Acrylamide administered via the drinking water Doses: 0.001, 0.01, 0.1, 1 and 10 µg/ml Time points: 1, 3, 6, 9 and 12 months

Testes extracted Testes extracted

Isolated early Fixed and RNA germ cells sectioned extracted

DNA damage analysis Microarray analysis (Nixon et al. 2012) Time points: 1, 6, 12 months

sFig. 1: Experimental Design. Acrylamide was administered to male mice via the drinking water at concentrations of 0, 0.001, 0.01, 0.1, 1 or 10 µg/ml and animals were exposed for 1, 3, 6, 9 or 12 months. Six individuals were allocated to each treatment group and testes were used for DNA damage analysis, as previously described (Nixon et al. 2012, depicted in grey). In the current study, testicular gene expression following acrylamide exposure was examined and the testes from one set of triplicate mice were used for RNA extraction and microarray analysis (depicted in white). Microarrays were performed on 1, 6 and 12 month time points only.

145 Appendix C: Chapter 4 Supplementary Data Manuscript: The Impact of 12-Month Chronic Acrylamide Exposure on Mouse Testicular Gene Expression.

1 month, 0.001 µg/ml

sFig. 2: Principle Component Analysis uses analysis of principal sources of variance in data and displays this graphically in a 3-dimensional plot. The array of 1 replicate sample of the 0.001 µg/ml dose, at the 1 month time point, was identified as an outlier and was omitted from subsequent analyses. Thus, 45 testis samples remained for microarray analyses in the present study.

146 Appendix C: Chapter 4 Supplementary Data Manuscript: The Impact of 12-Month Chronic Acrylamide Exposure on Mouse Testicular Gene Expression.

A. Tight Junction Signalling

(Figure continues on next page)

147 Appendix C: Chapter 4 Supplementary Data Manuscript: The Impact of 12-Month Chronic Acrylamide Exposure on Mouse Testicular Gene Expression.

B. Gap Junction Signalling

(Figure continues on next page)

148 Appendix C: Chapter 4 Supplementary Data Manuscript: The Impact of 12-Month Chronic Acrylamide Exposure on Mouse Testicular Gene Expression.

C. Germ Cell-Sertoli Cell Signalling

sFig. 4: Three of the top canonical pathways identified following 12 month chronic acrylamide exposure, which were relevant to junction signalling and the blood/testis barrier; (A) tight junction signalling, (B) gap junction signalling and (C) germ cell-Sertoli cell signalling. The signalling pathways were generated by IPA based on data from the highest dose, 10 µg/ml, as this was most representative of all doses measured. Regulated molecules were colour coded according to their expression (red: up-regulated, green: downregulated). Downregulation of molecules was predominantly observed in all three pathways.

149 Appendix C: Chapter 4 Supplementary Data Manuscript: The Impact of 12-Month Chronic Acrylamide Exposure on Mouse Testicular Gene Expression.

All supplementary tables (sTable 1 – 8) for this manuscript are compiled in an Excel spreadsheet (Appendix C_Supplementary Tables.xls) located on disc found on the inside back cover of this thesis.

150 sTable 1: The total numbers of testicular genes identified as differentially expressed by ANOVA (1.5 fold difference, p < 0.05) in mice exposed to acrylamide for 1, 6 or 12 months. Acrylamide Dose (µg/ml) 1 Month 6 Months 12 Months 0.001 10758 0.01 17 80 81 0.1 100 439 0 1 161 279 107 10 86 815 1 Total Regulated 248 865 180 sTable 2: All differentially expressed genes identified in testes of mice chronically exposed to acrylamide for 1, 6 or 12 months 1 Month 6 Months 12 Months 37316 40422 1110008P14Rik 0610037L13Rik 0610007C21Rik 1190002H23Rik 1110002B05Rik 0610009B22Rik 1700010D01Rik 1190002H23Rik 0610012G03Rik 1700011H14Rik 1700001P01Rik 0610037L13Rik 1700016D06Rik 1700003F12Rik 0910001L09Rik 1700021K19Rik 1700008F21Rik 1110001J03Rik 1700029F12Rik 1700018B24Rik 1110017D15Rik 1700063I17Rik 1700023B02Rik 1190007F08Rik 2310007A19Rik 1700023E05Rik 1500032L24Rik 2310039H08Rik 1700024G13Rik 1700001E04Rik 2900092C05Rik 1700025K23Rik 1700001J03Rik 3110070M22Rik 1700029I08Rik 1700001P01Rik 4921511K06Rik 1700047I17Rik1 1700003F12Rik 4930428D18Rik 1700063I17Rik 1700008F21Rik 4930432K21Rik 1700067K01Rik 1700008I05Rik 4930570C03Rik 1700095K08Rik 1700010D01Rik 4930583H14Rik 2010100O12Rik 1700010M22Rik 5730593F17Rik 2210418O10Rik 1700011F03Rik A330021E22Rik 2310007A19Rik 1700011H14Rik AA536749 2310039H08Rik 1700012A03Rik Abcb6 2400003C14Rik 1700012B09Rik Abhd4 2410001C21Rik 1700013N18Rik Acaa2 2610027L16Rik 1700015E13Rik Acsbg1 2610207I05Rik 1700016C15Rik Acss1 2810408A11Rik 1700016D06Rik Actb 2900092C05Rik 1700016G05Rik Agt 4121402D02Rik 1700017N19Rik Aldh2 4732415M23Rik 1700018F24Rik Aldoa 4733401H18Rik 1700019M22Rik Angel1 4930430E16Rik 1700019N19Rik Ap2b1 4930519N13Rik 1700021C14Rik Arl16 4930522H14Rik 1700023B02Rik Asah1 4930596D02Rik 1700023E05Rik Atp6ap1 4933407P14Rik 1700024G13Rik Azi1 4933428G09Rik 1700025E21Rik Bat2 6430706D22Rik 1700025F22Rik BC018465 Aamp 1700025K23Rik Bmi1 Acsbg1 1700026L06Rik C1qtnf2 Actb 1700029I08Rik C2 AI837181 1700029J07Rik Catsper1 Akap1 1700029P11Rik Ccdc113 Akap4 1700030J22Rik Ccdc117 Aldh1a1 1700034I23Rik Ccdc57 Aldoa 1700042B14Rik Cd59a Aqp11 1700047I17Rik1 Cd81 Arsa 1700049K14Rik Cdkn2c Aspscr1 1700057G04Rik Cep78 Atp5a1 1700057K13Rik Ciapin1 Atp6v1e2 1700063I17Rik Clk4 Atp6v1g1 1700081D17Rik Clmn Atxn10 1700095K08Rik Cnot4 Bap1 1700109H08Rik Cops7a Bat5 1700129C05Rik Cps1 BC005537 1810030N24Rik Crat BC061039 1810035L17Rik Crybg3 BC061237 2010100O12Rik Csl Bcas1 2210418O10Rik Ctsb Bhmt 2310005N03Rik Ctsh Brf2 2310006J04Rik Dak Bud31 2310007A19Rik Ddx6 C8b 2310036O22Rik Dhrsx Calm2 2410001C21Rik Dmwd Capns1 2410015M20Rik Dnajc5g Ccdc101 2410018C17Rik Dym Ccdc88b 2410025L10Rik E130016E03Rik Ccl28 2510003E04Rik Edem2 Ccrk 2610020H08Rik EG625054 Cd109 2610103J23Rik EG667977 Cd52 2700055A20Rik EG668668 Cdc42ep3 2810408A11Rik Egln2 Cdrt4 2810453I06Rik Ehd1 Cggbp1 2900010M23Rik Eif3i Ciz1 2900064A13Rik Eif4ebp1 Ckmt2 3000004C01Rik Ela1 Coil 3010026O09Rik Epb4.1l3 Commd1 3110070M22Rik Ephx1 Crisp2 3830406C13Rik Exosc5 Csnk2a2 4121402D02Rik Fbxo44 Cst9 4430402I18Rik Fcgr3 Cstb 4732415M23Rik Fdxr Cux1 4733401H18Rik Fntb Cuzd1 4921504I05Rik Fus Daf2 4921511H03Rik Fxyd6 Dbil5 4921511K06Rik Galntl5 Ddx4 4921528G01Rik Ggn Deadc1 4922502D21Rik Glrx Derl2 4930408O21Rik Glt8d1 Dkkl1 4930412F15Rik Gpha2 Dnajb6 4930430E16Rik Gpt2 Dnajc7 4930511I11Rik Gpx3 EG545047 4930511J11Rik Gsg1 Eif3k 4930522H14Rik Gstm1 Eif4el3 4930540L03Rik Gstm2 Fabp3 4930544G11Rik Hap1 Faim3 4930547C10Rik Hapln2 Fam134a 4930555G01Rik Hcr Fntb 4930578I06Rik Hdlbp Fscn3 4931417E11Rik Hist1h1c Galc 4931433A01Rik Hmgcs2 Gas2l1 4931440F15Rik Hnrpll Glrx2 4931440L10Rik Hsd17b11 Glt8d1 4932413O14Rik Hsd17b3 Gm128 4932702K14Rik Hsd3b2 Gm1821 4933400A11Rik Ibsp Got1l1 4933406M09Rik Igf2 Gpd2 4933407N01Rik Ino80b Gpr128 4933407P14Rik Insl3 Grin3b 4933413G11Rik Ipo4 Grtp1 4933428G09Rik Kif3a Gsg1 4933434E20Rik Klk1b26 Gstp1 4933434I06Rik Lcn2 Gtf2ird2 4933436I01Rik Lmo4 Hcfc2 5730437N04Rik LOC100045963 Hcr 6330548G22Rik Ly6e Herpud2 6430537H07Rik Map2k7 Hip2 6530404N21Rik Mipep Hist2h2ab 6720467C03Rik Mllt11 Hmgb2 A330021E22Rik Morn2 Hmgcs2 Aamp Mrps9 Hnrnpf Acaa2 Mupcdh Hsp90aa1 Acsbg1 Mxra8 Hspa8 Acta2 Nt5dc3 Hspb1 Actl6a Ogn Ibsp Actn2 Osr2 Id4 Actr10 Pafah1b1 Igf2 Adam24 Paox Inpp5k Adam39 Pcolce Itgae Adipor1 Pga5 Klhdc3 Adprh Phkg2 Lims2 Afap1 Pld3 LOC100040899 Agfg2 Plekha4 LOC100041703 Agpat3 Poli LOC100041729 AI314180 Polr1a LOC100044779 AI837181 Ppp1r10 LOC100046163 Aif1 Prcp LOC100046841 Akap1 Prdx6‐rs1 LOC100047052 Akap14 Prlr LOC100047749 Akap4 Prnp LOC100047935 Akap8l Psd3 LOC217341 Akr1a4 Ptpdc1 LOC218963 Akr1b3 Retsat LOC434960 Akr1e1 Rffl LOC544988 Akr7a5 Rgl1 LOC641240 Aldh3b1 Rnaset2b LOC668837 Aldoa rp9 Lrrc23 Alg5 Rpp38 Lrrc34 Allc Rps21 Lrrc8b Als2 Scrib Magmas Ankrd5 Sec23a Map1lc3a Ankrd7 Sepw1 Map2k7 Ap2b1 Serpinf1 Mea1 Apeh Sgpp1 MGC118210 Aplp2 Sh3gl3 Mrpl1 Apoc1 Slc25a10 Mrpl20 Apoh Slc40a1 Mrpl39 Aqp11 Slc7a8 Ms4a5 Aqp7 Smpdl3b Mylc2pl Arfgap2 Spata16 Ndufb10 Arhgap15 Spata7 Ndufb5 Arhgap29 Stambp Nosip Arid2 Stard7 Npm3 Arl3 Thap4 Oaz3 Arl6ip1 Tmem176b Odf2 Arpc1b Tnfrsf10b OTTMUSG00000007855 Arpc5 Tnks OTTMUSG00000015859 Arrdc3 Tomm7 OTTMUSG00000019001 Arsa Trip12 Pacs1 Art3 Tsc22d1 Pcbp2 Asnsd1 Ttll5 Pccb Ass1 Txnl4b Pebp1 Aste1 Uck2 Pfn2 Atf2 Ugt1a6a Pgam2 Atf4 Usp1 Pkib Atg2a Vcam1 Pmpcb Atox1 Wdr78 Polb Atp5a1 Zc3h15 Ppp3ca Atp5e Zfp36l1 Ppp4r1 Atp5h Zfp39 Prcp Atp6v1a Prdx6‐rs1 Atp6v1e2 Prm3 Atxn10 Psenen Aurka Psmd11 Aven Psmd8 B2m Ptpre B930041F14Rik Pvrl3 Bap1 Rabl5 Bat5 Ralgps2 Bbs9 Rffl BC017647 Rgl1 BC018465 Rnf220 BC051628 Rp23‐297j14.5 BC061039 rp9 BC061237 Rpl39l BC062650 Rps25 BC085271 Rpusd4 Bcas2 Rshl2a Bcl2l14 S100a11 Birc2 Sepp1 Bop1 Serpina5 Brp16 Serpinf1 Btrc Sfrs16 Bud31 Sh3gl3 Bzw1 Slc25a3 Calm2 Slc25a39 Calr Slc2a5 Catsper1 Slc35b1 Cbwd1 Sly Ccdc101 Smpd4 Ccdc134 Spata16 Ccdc28a Spata21 Ccdc53 Spata4 Ccdc70 Speer6‐ps1 Ccl27 Spert Ccrk Spink8 Cct6b Sqrdl Cct8 St13 Cd151 Stard7 Cdc2l6 Syce2 Cdc37l1 Tcam1 Cdca8 Tcp1 Cdk2 Tex101 Cdkal1 Tfb1m Cdkl2 Tmem176b Cdkn1c Tmem53 Cdrt4 Tmsb10 Cenpq Tnp2 Cep55 Trap1 Chchd6 Trp53bp1 Chchd7 Trpc4ap Chka Tsg101 Chl1 Ttll5 Chmp2a Tuba3a Chmp4b Tuba8 Churc1 Ube2u Ciz1 Ubxn6 Ckap2l Wbp4 Cklf Wdr31 Cldn11 Wfdc15b Clec4b1 Ybx2 Clk4 Ywhaq Clmn Zbtb3 Clptm1l Zkscan17 Clpx Znrd1 Cmtm2a Zpbp2 Cmtm2b Zscan21 Coil Zswim2 Col4a3bp Col5a1 Commd1 Cops4 Cops5 Copz2 Cox11 Cox17 Cox4nb Cox6c Cox7a2 Cox7b2 Cplx1 Cpxm1 Creb3l4 Crem Crls1 Crtap Crybg3 Csda Csl Csnk2a2 Csrp1 Cst8 Cstf1 Ctnna3 Ctnnb1 Cul1 Cuta Cyb5 Cyb5d1 Cyb5r2 Cyp17a1 Cypt2 Cypt6 Cypt8 Cypt9 D1Pas1 D2Ertd750e D3Ertd300e Daf2 Dap3 Dazap1 Dbf4 Dbi Dbil5 Dcakd Dcun1d5 Ddit3 Ddt Ddx19b Ddx4 Deadc1 Dedd Defb19 Defb36 Derl2 Dhh Dhx15 Dkkl1 Dmrtc2 Dnajb6 Dnajb9 Dnm1l Dusp13 Dym Ecsit Edf1 Eed Eef1b2 Eef2 Efhc2 EG432825 EG546282 EG625054 EG668668 Egfbp2 Egln2 Eif2b1 Eif3k Eif4h Elf2 Elovl2 Elovl5 Emg1 Emp3 Enpp5 Epb4.1l2 Ergic3 Erp29 Evl Fads1 Fads2 Fam108b Fam114a2 Fam134a Fam162a Fam173a Fank1 Fbp1 Fbxo21 Fbxo44 Fcf1 Fcgr3 Fem1c Fhl4 Fkbp6 Flot1 Fnip1 Fntb Fscn3 Ftmt Fzr1 Galm Galnt3 Galntl5 Gart Gemin6 Gk2 Gkap1 Glrx2 Glrx3 Glt8d1 Gm128 Gm136 Gm1527 Gm884 Gnl2 Gnptg Golga2 Gosr2 Gpd2 Gpr128 Gpt2 Gpx1 Gpx4 Grk4 Grtp1 Gsg1 Gstk1 Gstm2 Gstm5 Gtf2e2 Gtf2ird2 Gtl3 Gtpbp8 Gzmd H1fnt Hadhb Hap1 Hcfc2 Hcr Hdgf Heatr1 Hemt1 Herc4 Herpud2 Hip2 Hist1h2ao Hist2h3b Hist2h3c1 Hmgb2 Hmgb4 Hmgcs2 Hnrnpk Hnrpdl Hnrpll Hook2 Hoxc9 Hsd3b2 Hsf1 Hsf2 Hsp90aa1 Hsp90ab1 Hspa4 Hspa4l Hspa8 Hspb6 Hyal5 Ibsp Idh3g Ifngr2 Ift140 Ift20 Igf2 Il17d Inha Ints4 Irf3 Irgm1 Isg20 Itgae Itgb5 Itprip Kcnab2 Kcnk2 Kif27 Klf4 Klhdc3 Klhl10 Klk1 Klk1b9 Kpna2 Kpnb1 Krba1 Krcc1 Krtcap2 Lamp1 Lamp2 Lancl1 Lasp1 Lats2 Lcn2 Lgmn Lhcgr LOC100039532 LOC100041703 LOC100044087 LOC100044170 LOC100044779 LOC100046320 LOC100046841 LOC100047052 LOC100047368 LOC100047651 LOC100047749 LOC100047810 LOC100047834 LOC100047934 LOC100047935 LOC100047998 LOC100048020 LOC100048071 LOC100048105 LOC100048480 LOC100048710 LOC217341 LOC381629 LOC434960 LOC544988 LOC545732 LOC546015 LOC666676 LOC668837 LOC677317 Lpcat3 Lpgat1 Lrp10 Lrrc1 Lrrc18 Lrrc23 Lrrc34 Lrrc48 Lrrc49 Lrrc52 Lrrc67 Lrrc8b Lrriq3 Lrsam1 Lsm12 Lsm2 Ly6k Lyar Lypla2 Lyzs Mad2l2 Maged2 Magmas Magoh Man2c1 Map1lc3a Map1lc3b Map2k7 Mdga2 Mea1 Med1 Med28 Memo1 Metap2 Mff MGC107098 MGC118210 Mgst1 Mif4gd Mknk1 Mkrn1 Mkrn2 Mlf1 Mllt11 Mobkl2b Mod1 Morc2b Morn2 Mrpl11 Mrpl13 Mrpl16 Mrpl20 Mrpl47 Mrpl52 Mrps24 Ms4a10 Ms4a5 Mtch1 Mupcdh Mxi1 Mylc2pl Mylk Naca Naglu Nasp Nat5 Nbr1 Ncoa4 Ndufa5 Ndufb10 Ndufb3 Ndufb5 Ndufb9 Ndufs4 Nme2 Nme7 Nol14 Nola3 Nosip Noxo1 Nphp1 Npm3 Nsfl1c Nsun4 Nucb1 Nudc Nudt19 Nudt21 Nup133 Nup160 Oaz3 Obfc2a Odf2 Osr2 Ott OTTMUSG00000007855 OTTMUSG00000019001 Ovol2 P4hb Pa2g4 Pabpc1 Pacs1 Pafah1b1 Paox Pbrm1 Pcbp2 Pcolce Pdcd5 Pde6d Pdhb Pdpk1 Pebp1 Pex5l Pfdn2 Pfn2 Pgam2 Pgrmc1 Pgs1 Phc1 Pkib Pkig Pkn2 Pknox2 Plac8 Plcd4 Plcz1 Pld3 Plxdc1 Pmpcb Pnma1 Pno1 Pnpla2 Polb Poli Polr1c Polr2g Pom121 Ppap2b Ppp1ca Ppp1r10 Ppp2r5c Ppp3ca Ppp4r1 Ppp4r4 Prdx1 Prdx5 Prkaca Prox1 Prpf38a Prpf40a Prpsap2 Psap Psd3 Psenen Psip1 Psmd10 Psmd11 Psmd4 Psmd8 Ptpdc1 Ptpre Pttg1ip Pvrl3 Pxt1 Pygo2 Rab11fip5 Rab3il1 Rabl5 Rad52 Rai12 Ralgps1 Ralgps2 Rasl12 Rassf1 Rbbp7 Rchy1 Retsat Rfc4 Rffl Rfwd2 Riok1 Rnf10 Rnf138 Rnf139 Rnf166 Rnf19a Rnf32 Rnf6 Rnpepl1 Rnu6 Rogdi Ropn1 Rp23‐297j14.5 Rp23‐438h3.2 rp9 Rpl29 Rpl36a Rpl39 Rpl39l Rpp38 Rps10 Rps2 Rps21 Rps25 Rps3a Rps5 Rshl2a Rxra S100a11 Scgn Sdhd Sec11c Sec23ip Sel1l2 Sepp1 Serbp1 Serf1 Serpinf1 Setd3 Setx Sf3b2 Sgcb Sh3gl2 Sh3gl3 Sin3a Sirt3 Skiv2l Skp1a Slc25a17 Slc25a3 Slc25a39 Slc25a45 Slc30a6 Slc34a2 Slc35b1 Slc45a4 Sly Smc1b Smc5l1 Smek2 Smn1 Smpd4 Smpdl3a Snx1 Snx17 Snx3 Sod2 Spag16 Spam1 Sparc Spast Spata16 Spata21 Spata3 Spata6 Spatc1 Speer2 Speer6‐ps1 Spert Spink2 Spz1 Sqrdl Srp14 Srrm2 Srxn1 Ss18 Ssty2 St13 St5 Stard10 Stk22b Stmn1 Stx8 Sugt1 Sumo2 Sunc1 Syce2 Synj2bp Taf12 Taf9 Taldo1 Tbc1d19 Tbc1d20 Tcam1 Tcp1 Tctex1d1 Terf2 Tesk1 Tex101 Tex2 Tfam Tgif2lx Thoc4 Thoc5 Thop1 Timm8b Timp2 Tmem184a Tmem218 Tmem66 Tmem85 Tnp1 Tnp2 Tomm22 Tomm7 Tpi1 Tram1 Trim11 Trim33 Trim69 Trip12 Trpc4ap Tsc22d1 Tsg101 Tspan17 Tspan6 Ttc16 Ttc23 Ttc27 Ttll5 Tuba3a Tuba6 Tuba8 Tubb5 Tubb6 Tuft1 Uba1 Uba2 Ube2t Ube2u Ube3a Ubl3 Ubtd2 Ubxn10 Ubxn6 Ugt1a10 Uhrf1bp1l Usp1 V1rd20 Vapa Vapb Vasn Vps36 Vti1b Wbp2nl Wbp4 Wdr13 Wdr31 Wfdc10 Wnt6 Wtip Yif1a Ylpm1 Ythdf2 Ywhag Ywhaq Zbtb22 Zbtb3 Zbtb9 Zc3h10 Zc3h15 Zfp238 Zfp339 Zfp41 Zfp654 Zfyve21 Zkscan17 Znrd1 Zp3r Zpbp2 Zscan21 Zswim2 Zyx sTable 3: Summary of functions of the 18 differentially expressed genes that were regulated at all time points (1, 6 and 12 months) in testes of mice chronically exposed to acrylamide. Genes involved in spermatogenesis (Gsg1, Spata16) or imprinting/development (Igf2) were highlighted in black or grey respectively. Gene Name Summary of Function 2310007A19Rik regulatory subunit of type II PKA R-subunit (RIIa) domain con kinase regulator activity; signal transduction Gsg1 germ cell-specific gene 1 RNA polymerase binding 1700063I17Rik family with sequence similarity 24, member A unknown Sh3gl3 SH3-domain GRB2-like 3 Central nervous system development; signal transduction Aldoa aldolase A, fructose-bisphosphate embryonic glycolytic enzyme Rffl ring finger and FYVE like domain containing protein ubiquitin ligase Spata16 spermatogenesis associated 16 interacts with testis-specific poly(A) polymerase Glt8d1 glycosyltransferase 8 domain containing 1 glycosyl transferase Ttll5 tubulin tyrosine ligase-like family, member 5 protein modification; transcription Hcr coiled-coil alpha-helical rod protein 1 steroid metabolism Ibsp integrin binding sialoprotein structural protein of bone matrix Map2k7 mitogen-activated protein kinase kinase 7 protein kinase; signal transduction; response to environmental stress rp9 retinitis pigmentosa 9 (human) role in pre-RNA splicing Fntb farnesyltransferase, CAAX box, beta regulation of cell proliferation and cell cycle Igf2 insulin-like growth factor 2 Involved in development and growth. Paternally imprinted gene Hmgcs2 3-hydroxy-3-methylglutaryl-Coenzyme A synthase 2 ketone body production in the liver Serpinf1 serine (or cysteine) peptidase inhibitor, clade F, member 1 serine-type endopeptidase inhibitor activity Acsbg1 acyl-CoA synthetase bubblegum family member 1 fatty acid metabolic process sTable 4: All dose-dependent genes identified by ToxResponse Modeler that exhibited sigmoidal response curve and reasonable

EC50 (between 0 - 10 µg/ml) in testes of mice chronically exposed to acrylamide for 1, 6 or 12 months. Genes are highlighted based on the range of their EC50 value (legend shown on right).

1 month EC50 (µg/ml) 6 months EC50 (µg/ml) 12 months EC50 (µg/ml) Legend

1700012G19RIK 8.907 ASNS 9.659 PPP4C 9.232 1 - 10 µg/ml EC50 Range

CPSF6 8.255 TEX11 8.798 5730410E15RIK 9.081 0.1 - 1 µg/ml EC50 Range

LARP5 7.512 CEP164 8.557 LOC100046883 9.077 0.01 - 0.1 µg/ml EC50 Range

SPATA7 7.065 TMEM176B 8.034 AHCYL1 8.411 0.001 - 0.01 µg/ml EC50 Range CAP1 6.981 DNAJC8 7.900 DGAT2 8.391 E2F6 6.481 LONP2 7.865 ZFP579 8.314 GABARAPL1 6.470 LOC238829 7.815 ALG5 8.245 LOC100045617 6.382 TUBA3A 7.717 GALNT1 8.244 ZFP423 6.355 RPS15 7.653 ZER1 8.155 MYSM1 6.330 LOC100046320 7.630 6720456B07RIK 8.118 TMSB10 6.279 2310033E01RIK 7.561 NDUFS8 7.731 BC017647 6.227 BDH1 7.507 MRPS24 7.343 PTGIS 6.192 GNA11 7.425 SPECC1L 7.321 GRM7 6.039 PEX6 7.372 HDLBP 7.228 CSTF3 5.990 KLHL15 7.278 ARFGAP2 7.197 FMNL3 5.956 TTC4 7.258 SLC39A7 7.188 TMED4 5.944 COPG 7.237 ATMIN 7.147 NDUFA10 5.815 CYPT10 7.135 PSMG2 7.101 CNGA1 5.737 PRLR 7.117 RNF135 7.074 CRTAP 5.730 CCDC96 6.958 RBBP7 6.973 HARS 5.543 WDR5 6.945 SLC25A45 6.962 EPN2 5.518 NUDT18 6.943 MFSD10 6.945 PFDN2 5.443 KLHL25 6.913 TBC1D10A 6.928 LSG1 5.396 CDKAL1 6.912 HIST1H2AO 6.850 NQO2 5.352 4833418A01RIK 6.894 UBA2 6.841 ZFP281 5.323 HIST1H2BF 6.857 EBPL 6.811 GM136 5.318 RPL30 6.836 RPL30 6.745 2310016E02RIK 5.304 SRP9 6.833 LOC676420 6.731 LANCL1 5.295 EHMT2 6.794 SUZ12 6.718 SNX13 5.266 S100PBP 6.785 AKAP10 6.703 OMA1 5.256 1200003C05RIK 6.783 PPP2CA 6.702 DDX24 5.233 4930535E21RIK 6.776 ADAMTS2 6.698 G3BP2 5.184 CYB5D1 6.770 AB112350 6.696 SURF2 5.058 D19ERTD721E 6.754 TIMM8A1 6.682 MGAT5 5.009 LOC631784 6.746 TGM2 6.663 EFCAB2 4.999 SGCB 6.740 PTGES2 6.657 RBM18 4.970 CCL25 6.718 CLN3 6.634 SOX5 4.949 RGL2 6.713 FIBCD1 6.615 ADRA2A 4.870 1500001M20RIK 6.705 CENPM 6.532 CEP72 4.839 TBC1D15 6.704 ODF3L1 6.529 ZC3H12A 4.815 SYT11 6.704 RAB11FIP5 6.513 ACAD10 4.770 OBFC2A 6.682 PPM1A 6.505 MBD3L2 4.756 DBNDD1 6.678 MGP 6.498 OTUB2 4.753 H2AFZ 6.580 YIPF3 6.495 FXYD5 4.747 TAL2 6.545 DARS 6.471 RGS7BP 4.741 ZFP326 6.533 YEATS2 6.445 ARL6IP2 4.668 CCDC91 6.533 PEX11A 6.429 SNTB2 4.627 ADRBK2 6.520 1700021K02RIK 6.420 NPM3 4.535 PCMT1 6.517 8430410K20RIK 6.418 AMMECR1L 4.518 ZFP457 6.512 IGFBP4 6.414 VPS33B 4.512 CD59B 6.505 ALG1 6.413 UBE2G1 4.512 CDKL2 6.494 XPC 6.396 NCAPD3 4.472 ARL4C 6.494 POMP 6.354 OLFR611 4.448 USP3 6.491 MS4A10 6.352 ACTA2 4.413 HERPUD1 6.483 DNAJA3 6.335 DDX19B 4.355 FKBPL 6.476 NME2 6.272 ALG8 4.352 HDDC2 6.463 ZFP282 6.210 FARSB 4.338 BC027231 6.458 RBBP5 6.209 TMEM9B 4.318 YWHAG 6.458 HSF1 6.193 SLC39A4 4.300 ELMO3 6.441 NME2 6.165 DHX38 4.273 HILS1 6.438 MVP 6.145 TRCG1 4.255 LOC100044101 6.423 COX17 6.124 APRT 4.183 LOC100047794 6.419 LEMD2 6.124 KHK 4.176 BCAS3 6.414 PPM1D 6.099 ACD 4.133 NDUFS4 6.367 LY6G6C 6.088 1200015N20RIK 4.124 RELL2 6.362 EIF3EIP 6.079 BRF2 4.094 ARMC6 6.361 SEPP1 6.076 CCBL1 4.078 USP39 6.356 TRFP 6.059 CCT6A 4.069 UBTD1 6.331 TAF12 6.049 VPS45 4.020 5830457O10RIK 6.326 LOC100046650 6.048 GANAB 3.994 1110002B05RIK 6.322 HES1 6.043 ALDH1A7 3.974 RRP1 6.314 NUP43 6.012 ZFP423 3.962 LSM14B 6.297 MAPK8IP2 6.009 9430015G10RIK 3.958 SLC22A21 6.281 ZMYM3 6.007 BTF3 3.927 B230317C12RIK 6.272 APOC1 5.995 NTAN1 3.909 YIPF1 6.261 SLC4A8 5.949 NECAP1 3.906 DEXI 6.252 B3GALT6 5.918 SCYL3 3.900 1500032D16RIK 6.251 SDF4 5.914 NME3 3.899 SYNGR2 6.231 TMEM70 5.897 NIT2 3.777 EID3 6.229 CCS 5.893 CSNK1D 3.768 TSG101 6.222 EG546166 5.885 ITM2B 3.741 RAB3IL1 6.220 FOXJ3 5.882 TMEM33 3.673 MRPS26 6.217 LOC100047911 5.855 BC016495 3.643 STX8 6.212 FIBP 5.847 NGFR 3.638 FGFR1OP 6.206 ZFP292 5.819 TIMM8A2 3.619 RNF6 6.205 METAP1 5.819 RANBP1 3.605 BFAR 6.203 SMURF1 5.819 KNTC1 3.579 CPEB2 6.197 NUCB1 5.818 ARL2BP 3.442 SMPD4 6.195 BC024997 5.803 CHFR 3.382 BC013529 6.194 BRD9 5.801 LOC236749 3.366 PURB 6.185 SEC16A 5.794 GNAS 3.323 NT5C2 6.179 4921517D22RIK 5.782 YIF1A 3.320 HPGD 6.176 ZFP35 5.782 RAI12 3.308 EG384813 6.172 CTNS 5.782 TDRD3 3.287 NFKBIL2 6.171 ACAD10 5.779 QPCT 3.276 CLCN2 6.166 ZDHHC14 5.770 MAP2K7 3.172 HSDL1 6.157 9030624J02RIK 5.770 HEBP1 3.163 6330503K22RIK 6.148 TCAM1 5.766 LOC546015 3.087 IGFALS 6.123 AACS 5.765 ALG12 3.072 4921523A10RIK 6.120 C85492 5.740 SNTB2 2.926 POLG 6.112 MBTPS1 5.685 LETMD1 2.879 SPATA16 6.108 CENTB5 5.681 MYLK 2.852 TTLL13 6.104 BB128963 5.681 DNAJC5 2.814 RHOX3A 6.097 BFAR 5.678 4932414N04RIK 2.579 BBS2 6.088 ACADM 5.666 FBXO9 2.576 SLC35B2 6.071 WBSCR18 5.664 NFATC3 2.571 KRBA1 6.062 ZSCAN10 5.650 DCTN5 2.538 KDELR2 6.059 OVCA2 5.632 MTHFD1 2.458 TRIM36 6.054 SLC4A2 5.625 2610024G14RIK 2.456 D0H8S2298E 6.052 ZFP711 5.614 EIF2B1 2.447 MTHFD2 6.043 CRELD2 5.609 ZFP709 2.406 1700016D06RIK 6.031 DAAM1 5.596 ALG5 2.403 ERGIC2 6.023 FAHD2A 5.595 NCDN 2.387 BC022687 6.002 SPINLW1 5.595 PABPC2 2.385 2310065K24RIK 5.997 P2RX4 5.592 POFUT1 2.257 RTEL1 5.997 RIMS2 5.583 NARS 2.227 RPAP1 5.996 DUSP16 5.566 DBT 2.171 MON1A 5.995 CCDC92 5.558 ARFRP1 2.140 TUBD1 5.979 9530068E07RIK 5.556 FXN 2.140 CHCHD4 5.969 DEXI 5.554 TRIM45 2.129 SPCS1 5.967 MYO7A 5.542 MRPL27 2.035 4921507P07RIK 5.956 ZFP330 5.530 LAPTM4A 2.007 1700071K01RIK 5.954 PTPN21 5.514 ZFP42 1.992 BC061237 5.951 ESPN 5.512 POLE3 1.974 CTSL 5.948 AU040320 5.487 EVI1 1.922 LOC668492 5.939 DAG1 5.436 1810073G14RIK 1.904 4921528G01RIK 5.939 SPCS3 5.416 5730596K20RIK 1.898 6230416J20RIK 5.929 TMEM16K 5.412 RAB5A 1.838 NUPL1 5.922 DBP 5.385 TRPC4AP 1.818 BTG4 5.920 ZFP207 5.378 ZMYND12 1.807 PKD2L2 5.905 GSPT1 5.377 SDF2 1.800 4732471D19RIK 5.905 SYNJ2 5.375 NSMAF 1.800 1810015A11RIK 5.896 SYT5 5.369 POLDIP3 1.759 LSM14B 5.886 GMPPB 5.359 ELAC2 1.752 REC8 5.873 ARIH2 5.346 GIPC1 1.720 C920005C14RIK 5.848 NFYC 5.343 HNRPK 1.680 NOL11 5.822 PIAS2 5.342 EYA3 1.648 LOC100046775 5.814 H47 5.337 MAGEB3 1.637 RNF41 5.810 ULK1 5.329 RGL2 1.599 POLDIP3 5.805 SNRP70 5.328 RPL36AL 1.593 PRPF40B 5.804 SERTAD3 5.322 CRYZL1 1.577 MANBAL 5.799 FNTA 5.321 GRIP1 1.569 SPATA2L 5.798 TNFAIP2 5.316 OLA1 1.550 MTERFD1 5.781 PXT1 5.309 HNRPL 1.497 NACA 5.763 ACTA2 5.287 TUBA3A 1.489 AMN 5.757 RAVER1 5.269 9130014G24RIK 1.446 TSGA8 5.751 AMFR 5.254 POP4 1.428 AGGF1 5.741 ALG5 5.253 1700113I22RIK 1.372 ODF2 5.718 ACAT2 5.244 NOL12 1.282 HSPB7 5.711 LOC100041103 5.227 OBFC1 1.269 ARFGAP1 5.703 YIF1A 5.220 A930009M04RIK 1.267 ADAM1B 5.691 RCL1 5.214 1110018J18RIK 1.256 DOT1L 5.688 DEFB19 5.209 CASC4 1.167 SOCS7 5.678 HSPB7 5.190 CALR 1.102 NUDC 5.676 ZC3H13 5.172 DLST 1.041 2310037I24RIK 5.671 SDC4 5.170 TTC16 1.019 SPINK4 5.665 JMJD2B 5.165 CLTA 1.012 1700034I23RIK 5.660 STX12 5.160 4931417E11RIK 1.001 SMC1B 5.634 SMARCC1 5.153 DEFB36 0.990 1700094C09RIK 5.615 SENP2 5.137 ATP6V0A2 0.975 DSCR3 5.614 PRMT2 5.123 SH3BP2 0.973 PIK3C2A 5.602 MYO9A 5.122 COX6B1 0.971 ADAM6 5.600 POLR2D 5.110 OSGEPL1 0.965 PIK3CB 5.546 DPYSL4 5.091 DUSP2 0.953 PGRMC1 5.539 FBXL10 5.088 GANAB 0.928 ZFP819 5.534 CCNB1 5.087 ZDHHC9 0.883 USP1 5.530 GBL 5.060 SPATA7 0.883 MRRF 5.522 DNMT3B 5.049 ARFGAP1 0.881 HIST1H2BK 5.511 BZRAP1 5.045 MYEF2 0.830 GOLGA1 5.503 GPS1 5.041 TMEM98 0.810 EDEM2 5.500 TEGT 5.027 HYI 0.796 CLNS1A 5.484 BNIP2 5.026 ZXDA 0.780 RPL19 5.479 PIK3IP1 5.023 6530406M24RIK 0.770 CHRNE 5.473 PATL1 5.007 TUBA3A 0.761 RIOK3 5.460 UBAP2L 5.002 D10ERTD322E 0.748 VPS53 5.449 KLHL23 4.997 NSMCE2 0.729 DEFB30 5.437 KANK1 4.989 ATPBD1C 0.725 PTGES2 5.432 ATP6AP2 4.982 ADAM1B 0.724 DDX20 5.426 BANF1 4.977 WDR37 0.712 TNFRSF10B 5.423 MYO18A 4.966 ANKRD50 0.699 HIST1H2BM 5.418 BRD4 4.956 JMJD5 0.660 PTPRS 5.405 HOXA13 4.952 LOC100048046 0.623 TBC1D19 5.395 POMGNT1 4.946 1700055M20RIK 0.544 UBE2U 5.394 GBA 4.944 8430432M10RIK 0.536 CCNB1 5.372 LDOC1L 4.938 JAZF1 0.530 EG237009 5.356 second 38961 4.929 1810007P19RIK 0.529 GEMIN4 5.354 CAPZB 4.925 PRIM2 0.500 CTNNA3 5.347 SMARCA2 4.908 LMAN1 0.461 NT5C2 5.343 1500019G21RIK 4.901 PROK2 0.455 SRRD 5.336 KLK1B24 4.897 4930427A07RIK 0.447 ECSIT 5.332 CLSTN3 4.891 MAST2 0.433 ULK2 5.331 GABARAPL1 4.891 CENPM 0.433 AKAP4 5.330 MAPKAPK3 4.888 SH3GL2 0.427 CDK2 5.325 HRMT1L2 4.879 SCNM1 0.421 MAPBPIP 5.322 BC002230 4.877 ASB2 0.409 PCSK4 5.284 EIF4G1 4.870 PTPRB 0.407 CDC5L 5.282 NANS 4.868 SNORA65 0.389 RSN 5.276 NISCH 4.863 ECM1 0.388 IFNGR2 5.266 TMEM41A 4.860 MYH9 0.387 ZFP318 5.261 0610038F07RIK 4.850 1700020N01RIK 0.377 MRPS7 5.258 SH3TC1 4.845 ZBTB8 0.372 HMX1 5.245 ZAN 4.832 HSPG2 0.372 UCHL3 5.242 NCAPD3 4.829 LRG1 0.367 IGFBP7 5.240 4930430E16RIK 4.812 GRIT 0.359 ZFP654 5.238 NMNAT3 4.809 1700012L04RIK 0.356 GEMIN5 5.227 UBAC2 4.807 4922505G16RIK 0.354 PPA2 5.221 KIF1B 4.807 ILVBL 0.351 GRCC10 5.221 STAG3 4.798 CTNNA3 0.336 COQ4 5.207 MTAP1B 4.793 PDE4A 0.336 ZFP780B 5.207 ABCF1 4.790 HSPA5 0.335 POLL 5.184 TMCO4 4.790 BC018242 0.327 PDK3 5.182 SNX33 4.783 TMEM176B 0.308 GIPC2 5.173 MAN2C1 4.758 NPHP1 0.302 3110037I16RIK 5.171 MCM2 4.757 PHC1 0.302 8430432M10RIK 5.170 ELAC2 4.745 BCL2L2 0.294 ZFP654 5.165 SERPINB6A 4.730 RBBP9 0.274 CUX1 5.162 GALNT10 4.730 0610009O20RIK 0.257 ZNHIT2 5.145 ETAA1 4.725 DIDO1 0.255 ZKSCAN17 5.128 TFDP1 4.715 RBM12 0.253 DMRTC2 5.126 CDC42BPB 4.710 COPS6 0.252 OXR1 5.115 SLC35E3 4.707 MUPCDH 0.250 PRPF3 5.110 LRRC59 4.705 LRIG3 0.227 CCNL1 5.109 CNOT2 4.697 SLC25A35 0.227 RARS2 5.107 HNRPH1 4.689 BHMT 0.225 RAGE 5.102 9130221H12RIK 4.686 RPL13A 0.222 LIMK2 5.101 GALNT10 4.684 CYP27A1 0.221 BC038286 5.094 NSF 4.684 EXTL2 0.204 SELL 5.079 GNA12 4.682 MIPEP 0.192 ACTR3B 5.079 RNF139 4.669 IDS 0.188 ACTL7B 5.058 RAD51C 4.664 P2RY6 0.185 EPN2 5.055 BDH1 4.655 FOXRED1 0.182 ADAM29 5.054 RNPEP 4.652 PRKD3 0.181 UBQLN1 5.052 TMEM201 4.650 ITFG2 0.180 SFMBT1 5.044 5330431N19RIK 4.643 HEXDC 0.178 TXNDC13 5.042 NLN 4.640 SSR4 0.173 TNFAIP8 5.028 SLC25A28 4.634 IL17RB 0.171 BTG1 4.984 ALAD 4.632 4933421E11RIK 0.169 FHIT 4.976 MED10 4.624 WHRN 0.167 EHBP1 4.974 MOBK1B 4.610 GJA6 0.161 4933400A11RIK 4.962 ZFP688 4.606 1300017J02RIK 0.157 BLM 4.951 HDGF 4.600 SULT1E1 0.155 PSTPIP1 4.947 FADS2 4.593 LOC100048554 0.152 MARK2 4.947 SPAG7 4.587 GM773 0.150 OTUD5 4.940 2310005E10RIK 4.582 DAB2 0.148 TMEM177 4.933 HEATR1 4.580 AHNAK 0.143 CCDC116 4.932 RGS9 4.580 TIMP1 0.143 PRUNE 4.929 PVRL3 4.577 APLP2 0.138 1110057K04RIK 4.925 ZFHX2 4.564 2310031L18RIK 0.137 RBL1 4.918 SDSL 4.553 CARHSP1 0.135 RIC8B 4.915 DUSP11 4.553 TTC17 0.135 THSD7B 4.914 PYCR2 4.547 GSTM1 0.134 RAB5A 4.906 SUPT3H 4.543 UBLCP1 0.131 D14ERTD231E 4.898 HSPA9 4.541 BBS9 0.130 PSMD7 4.890 BRP17 4.533 ZRANB1 0.130 D4BWG1540E 4.887 RNF10 4.528 2900024O10RIK 0.130 OLFR611 4.882 PPP3CC 4.512 PI4K2B 0.129 TFB2M 4.877 GBL 4.508 LOC677317 0.127 TARBP2 4.873 QPCT 4.499 CBFB 0.125 ELF1 4.865 LOC675813 4.491 NMNAT3 0.125 HDDC2 4.863 TFG 4.489 FOXJ3 0.124 DHX15 4.863 PPP2R5E 4.488 CORO1B 0.123 NDUFC1 4.858 LSM4 4.484 HSD17B4 0.119 RBL2 4.835 CCDC32 4.461 OGFR 0.119 5730410I19RIK 4.831 TRAM1 4.459 OSR2 0.118 1700010M22RIK 4.829 TJAP1 4.453 DIAP1 0.118 TYK2 4.821 DHRS1 4.451 IGF2 0.117 GORASP1 4.820 SLC9A1 4.443 SERPINA3H 0.117 COQ2 4.811 TSSC1 4.439 PRCP 0.117 TMEM150 4.807 ACAA1A 4.422 ZFPM2 0.116 SPEER4D 4.788 6720463M24RIK 4.419 LOC100046883 0.113 EG238829 4.780 PRKACA 4.417 B230339M05RIK 0.112 TMEM98 4.778 D6WSU176E 4.416 9430023L20RIK 0.112 SOX8 4.777 SUMO1 4.414 KIF3C 0.109 DTX2 4.771 SAE1 4.413 RILP 0.109 STX7 4.754 ZC3H12A 4.408 FDXR 0.109 BC049730 4.745 CDKN2AIPNL 4.401 ZBTB39 0.109 COPE 4.717 TMEM63C 4.400 TIMP1 0.108 KLKB1 4.703 BUB3 4.386 ACADM 0.107 ZPBP 4.703 CCDC59 4.383 HS3ST1 0.107 KPTN 4.699 CHTF18 4.381 UBE2J1 0.107 2610110G12RIK 4.696 AXL 4.374 FIGF 0.106 GYKL1 4.694 DERL3 4.368 GORASP2 0.104 CENPB 4.686 RPS8 4.366 BTBD14A 0.104 COMMD3 4.685 E430018J23RIK 4.365 SIRT5 0.104 PLCB2 4.678 TRIM26 4.353 MAN1A2 0.104 SLC8A1 4.672 LOC100047800 4.350 CDK5RAP1 0.104 RNF138 4.671 SAR1A 4.345 B930041F14RIK 0.103 1700027N10RIK 4.667 ASCC1 4.340 ELOVL2 0.102 CCT6B 4.651 TLE1 4.332 2310005E10RIK 0.102 RRM1 4.649 LOC216443 4.330 HNRPH1 0.102 EDF1 4.647 RANBP1 4.324 GLB1 0.102 SLC7A6 4.646 1700012G19RIK 4.321 SNX10 0.101 NUP50 4.636 ABCA7 4.321 EPS15L1 0.101 NUDT9 4.635 LOC677528 4.311 CCDC42 0.100 HERPUD1 4.616 GOT2 4.308 RPAP1 0.100 NDUFS4 4.615 PTP4A2 4.282 NRBP2 0.099 1700029I15RIK 4.609 FASTKD1 4.271 SERPINF1 0.099 2010001J22RIK 4.608 FNDC5 4.270 PTP4A2 0.099 SGTA 4.606 1190007F08RIK 4.268 PDE1B 0.099 PFKM 4.601 PTPLAD1 4.263 RASSF4 0.099 CLEC2D 4.600 TRPC4AP 4.256 PNPLA6 0.098 EEF1E1 4.588 1700036D21RIK 4.249 MOBKL2B 0.098 TPD52L2 4.580 GLT8D1 4.234 BC026439 0.098 DNAJC4 4.576 HIBADH 4.221 2610207I05RIK 0.098 PDCD10 4.574 PIP5K1B 4.212 SLC16A6 0.098 4930572J05RIK 4.570 JMJD5 4.210 VAMP8 0.097 ELOF1 4.546 APLP2 4.192 A230051G13RIK 0.097 SEC14L1 4.534 PRKCZ 4.188 SLCO3A1 0.097 SLC39A7 4.532 MYH11 4.183 9130404D14RIK 0.096 MLLT1 4.532 SUPT6H 4.181 BGN 0.096 DMRTC1A 4.525 MAP2K5 4.177 3830408P04RIK 0.096 IARS 4.522 TUBB2B 4.174 CMBL 0.095 2310005N03RIK 4.519 NR4A3 4.173 HSD3B7 0.095 DCTD 4.511 MANBA 4.172 EDC4 0.095 BRE 4.509 MNS1 4.164 CDIPT 0.095 TTC15 4.505 NDE1 4.163 SPRED1 0.095 3110070M22RIK 4.498 4930511J11RIK 4.161 EPHX1 0.095 TRIM17 4.494 SERPINA3N 4.157 CTF1 0.094 SYPL 4.494 ALDH5A1 4.149 SYAP1 0.094 RORA 4.492 DCP1A 4.142 SLC12A8 0.094 NHP2L1 4.492 FXYD5 4.127 ITGBL1 0.094 CST9 4.488 CMTM7 4.126 LOH11CR2A 0.093 TERF2IP 4.485 ZDHHC9 4.109 ADAM23 0.093 UBE3B 4.475 POLR2H 4.099 NDG2 0.092 ZFP770 4.472 ATP1A1 4.095 LOC100043671 0.091 TRAPPC4 4.466 ODC1 4.088 0610006I08RIK 0.091 PLCD1 4.464 PSMA5 4.088 USP54 0.091 DEXI 4.463 FEM1B 4.062 TCF12 0.091 MAN1A2 4.458 1110032A13RIK 4.043 3000004C01RIK 0.091 GPSM1 4.455 REPIN1 4.043 TBC1D2B 0.091 WDR31 4.442 PSMB6 4.038 WNT6 0.090 SFRS1 4.438 IMPAD1 4.037 PIK3R1 0.090 CENPJ 4.437 PRICKLE4 4.035 STAT3 0.089 LRSAM1 4.429 HIST1H2AH 4.030 EFCAB5 0.087 SLC25A41 4.425 TIAM1 4.029 TCEB3 0.087 TCTN3 4.420 INTS3 4.026 PTTG1IP 0.084 MRPL1 4.409 FBXO31 4.019 PIP4K2A 0.084 CLIP1 4.405 CIRBP 4.017 MLF2 0.084 MAD2L1 4.397 SF3B2 4.008 ADAM22 0.083 TRIM69 4.396 PPP3CB 3.956 BC026585 0.083 CCDC11 4.396 RFX1 3.955 FUNDC1 0.081 ACP6 4.391 ZFP771 3.948 TMEM160 0.081 H1FNT 4.385 AGXT2L2 3.913 RHOBTB3 0.081 EG212225 4.379 RNF167 3.908 FAIM3 0.080 ACADSB 4.364 YIF1A 3.906 GAB1 0.080 RGL2 4.355 AKAP9 3.906 TMC1 0.079 WDR78 4.353 GSG1 3.902 BAT5 0.079 5230400G24RIK 4.351 AP1S1 3.870 PDLIM7 0.079 8430415E04RIK 4.349 SUV420H2 3.870 JMJD1A 0.078 UPP2 4.349 PUF60 3.868 HPS3 0.077 HYAL5 4.347 PRKCSH 3.868 SAMD9L 0.077 TEX101 4.343 LEPROTL1 3.867 ZFP637 0.077 1700039E15RIK 4.341 TMED1 3.861 MTAP7D1 0.075 IL17D 4.333 MED12 3.831 A930005I04RIK 0.074 1810063B05RIK 4.324 FKBP8 3.822 KIF1B 0.074 4922503N01RIK 4.321 2310046K01RIK 3.821 CHRD 0.074 LSM8 4.317 A730017C20RIK 3.820 ABCA14 0.073 PSMD14 4.306 TIMM44 3.811 GPR128 0.073 NUP98 4.289 2310037I24RIK 3.807 NUDT19 0.072 BOP1 4.288 TPRKB 3.802 TGFBRAP1 0.072 RBAK 4.286 PRLR 3.796 CERCAM 0.072 KBTBD4 4.284 SPAG8 3.793 CMAS 0.071 SCRN1 4.282 COL6A1 3.793 H6PD 0.071 RHOX3F 4.280 CARM1 3.790 KLK1B26 0.070 COQ4 4.278 LENG9 3.788 ZW10 0.070 WDR53 4.270 GART 3.785 LOC329575 0.069 GAS8 4.263 GFRA4 3.783 PYGB 0.068 2610528E23RIK 4.261 VPS72 3.781 GLB1L2 0.068 1500005A01RIK 4.250 0610025P10RIK 3.777 PHF23 0.067 ITGA1 4.239 FHL1 3.777 JAM2 0.066 GATAD1 4.234 PSMD11 3.776 SYTL2 0.066 TAF6L 4.234 AP1M1 3.766 RDH14 0.066 IRGQ 4.234 ALK 3.762 DBR1 0.066 GTPBP8 4.222 RNF167 3.750 GPR19 0.066 4732415M23RIK 4.219 MAGI1 3.746 UBA2 0.066 COL16A1 4.218 TARBP2 3.739 C3 0.066 STARD4 4.217 RASL11B 3.728 MTCH1 0.065 RPL12 4.206 CHST1 3.725 LENG9 0.064 AHCTF1 4.202 TXNDC2 3.724 2700087H15RIK 0.064 MRPL52 4.202 CHIC2 3.717 5830434P21RIK 0.063 PLCB3 4.198 SVS5 3.717 TMSB4X 0.063 XRCC6 4.192 COX19 3.716 1700022C21RIK 0.062 TCP11 4.188 RBPMS 3.712 FHL1 0.062 CCRN4L 4.166 SIDT2 3.707 MGP 0.062 5730596K20RIK 4.164 NIPSNAP1 3.702 LRRC46 0.061 RAD23A 4.155 NDUFA13 3.680 NADK 0.061 CORIN 4.153 NFRKB 3.678 EIF2S2 0.060 AKAP4 4.149 HORMAD2 3.675 MTSS1 0.060 TPM2 4.148 PAK3 3.674 TTLL1 0.059 8430415E04RIK 4.142 ABCA3 3.661 PKP2 0.059 SLC7A8 4.131 LOC100044324 3.660 PIK3R1 0.058 TRAK1 4.129 FKBP4 3.660 COL6A1 0.058 DAZAP2 4.128 HIST1H2AD 3.655 1110007L15RIK 0.058 TEX19 4.116 PLK4 3.650 6330503K22RIK 0.057 SLC25A11 4.108 CASC3 3.635 LOC100041725 0.057 TGIF2LX 4.108 HDGF 3.635 DBP 0.057 ARHGAP15 4.108 CNOT2 3.632 DAZL 0.057 ARHGAP1 4.106 LASP1 3.628 ACN9 0.057 LOC100048589 4.106 TXNDC5 3.624 SPHK1 0.057 CHMP2A 4.095 G3BP1 3.618 ZBED4 0.057 CCDC46 4.088 LYZ2 3.608 BC017158 0.056 LOC100048105 4.087 VAMP4 3.602 COBLL1 0.056 NBR1 4.086 YKT6 3.599 RREB1 0.056 RHPN1 4.085 FSD2 3.594 1110007A13RIK 0.056 SECISBP2 4.085 2310003C23RIK 3.594 DAP3 0.055 PHLPP 4.084 PROX1 3.588 WTIP 0.055 UBFD1 4.084 CCND3 3.582 MTMR14 0.055 RABEP1 4.084 SNX17 3.573 LOC668837 0.055 WDR32 4.084 SLC25A14 3.569 PPIC 0.055 2610304G08RIK 4.083 PPT1 3.564 RBMS2 0.054 COMMD1 4.075 IRF2BP1 3.562 NFE2L1 0.053 PSMC1 4.074 MOSPD1 3.559 TPTE 0.053 PDE4DIP 4.067 HIRIP3 3.554 MRPL49 0.053 4930481F22RIK 4.063 ACTN3 3.553 LOC100048483 0.053 COPZ2 4.057 4931406P16RIK 3.547 SRD5A3 0.052 LOC216963 4.056 MTMR3 3.541 GTSE1 0.052 ABHD1 4.055 GGN 3.532 PIWIL2 0.052 FHL5 4.042 ZDHHC13 3.526 4930455C21RIK 0.052 LOC217341 4.038 TMEM66 3.522 EIF2B5 0.052 SNAPC3 4.034 CML1 3.499 TRAFD1 0.052 DUSP2 4.032 NPM3 3.498 LOC100040592 0.052 CGRRF1 4.031 MAPK1 3.493 FHL2 0.051 KLHL7 4.027 LRP11 3.492 DDX17 0.051 ZFP295 4.026 HEMK1 3.491 SLC25A45 0.051 MFN2 4.016 CYP17A1 3.491 GPBP1 0.051 4933425O20RIK 4.009 ADAM15 3.464 SMC1A 0.051 TCEAL1 3.975 OCIAD2 3.463 EIF2AK2 0.051 LRRC49 3.972 RAF1 3.459 ATP1A1 0.051 D4WSU132E 3.969 WIZ 3.443 TIAM1 0.050 MLYCD 3.969 CTNNA1 3.432 RAPGEF5 0.050 PHC1 3.961 2610204L23RIK 3.425 TRP53BP1 0.049 2610103J23RIK 3.960 1700021C14RIK 3.417 NHSL1 0.049 STRAP 3.960 UNG 3.416 RBM9 0.048 TFDP2 3.957 TBC1D25 3.406 LBH 0.048 SETD8 3.956 ZSWIM3 3.402 LYPD5 0.048 RDBP 3.944 LLGL1 3.400 ZFP612 0.048 TMEM77 3.942 MFN1 3.394 LOC100042427 0.048 PKN2 3.941 DAP3 3.380 PRRC1 0.048 EFHC2 3.939 HMGN2 3.380 LOC100047934 0.048 LRRC1 3.917 CYB561 3.377 TSPAN17 0.048 NAPEPLD 3.903 SLC1A2 3.377 RALY 0.048 ACAD9 3.902 FZR1 3.356 XPNPEP1 0.047 RNF216 3.901 KCTD5 3.354 1110057K04RIK 0.047 FLCN 3.891 C130074G19RIK 3.354 HTR5B 0.047 ZFP238 3.885 RXRB 3.345 BAG3 0.047 4930524E20RIK 3.877 WDR54 3.344 CUTA 0.046 GMCL1 3.867 ATP6V1B2 3.320 TEX2 0.046 LANCL1 3.856 CSNK1E 3.310 TMEM201 0.046 NDUFA7 3.853 BC046331 3.308 TRPD52L3 0.046 HRASLS5 3.850 ZFP60 3.298 ADI1 0.046 MEX3C 3.831 FUSIP1 3.298 UNC84A 0.046 RAB24 3.825 H2-DMA 3.296 MAGI1 0.046 UBE4B 3.824 IL17RE 3.293 LRBA 0.045 MCPH1 3.813 HEATR1 3.281 GPR120 0.045 NT5C2 3.813 ZFP212 3.274 PITPNA 0.045 ENOPH1 3.804 ALOXE3 3.273 RALBP1 0.045 UBL3 3.800 BC030476 3.272 SYVN1 0.045 CES3 3.797 PEX5 3.265 SAS 0.045 BC065085 3.784 KLK1B22 3.261 ZFP474 0.044 NAT5 3.777 ALDH1A2 3.259 VAPB 0.044 DEPDC5 3.771 UXT 3.253 1810055G02RIK 0.044 MTL5 3.763 LOC100047934 3.245 ATAD1 0.044 ANKAR 3.760 IGFBP7 3.241 TMEM147 0.043 KIF1B 3.756 ARHGEF4 3.231 ISY1 0.043 PRKRA 3.747 GRIT 3.225 ALDH1A1 0.043 SRP9 3.739 VKORC1 3.223 ARAF 0.043 ZSWIM2 3.733 ECD 3.191 GUSB 0.042 BC028528 3.719 VCP 3.186 LEPREL1 0.042 SLC12A3 3.710 COX18 3.183 DICER1 0.042 1700012B09RIK 3.710 FZR1 3.173 NAP1L4 0.042 FH1 3.700 KIF23 3.171 GSG1 0.042 MRPS33 3.699 PHKA2 3.145 4922503N01RIK 0.042 DTNA 3.696 MUS81 3.142 DPPA5 0.042 ANKRD40 3.687 NRD1 3.137 B2M 0.041 A530082C11RIK 3.686 CCDC41 3.133 CFL1 0.041 KCNAB1 3.683 SPNB2 3.133 TMEM118 0.041 TOP1MT 3.678 0910001L09RIK 3.126 D0HXS9928E 0.041 PTPDC1 3.662 MARCKS 3.112 DBNDD2 0.041 RNF24 3.661 D330028D13RIK 3.106 PORCN 0.040 STX5A 3.654 PRM1 3.093 SHPRH 0.040 RBM12 3.650 BC051227 3.060 5730472N09RIK 0.040 IGF2BP3 3.645 EPHX1 3.054 AKT3 0.040 GNL3 3.642 IFNGR1 3.050 NET1 0.040 SSXB2 3.638 VGLL2 3.045 MSH3 0.040 HIST1H2BJ 3.632 PPM1B 3.040 AKAP9 0.040 ASNSD1 3.621 PQLC3 3.037 UQCRC2 0.039 DPY30 3.606 MTAP 3.035 4931428L18RIK 0.039 SNX7 3.599 LOC100047651 3.025 5133400G04RIK 0.039 TIMELESS 3.593 FLCN 3.020 LOC218963 0.039 RAB3IP 3.592 GPR108 2.998 LRRC49 0.039 TPP2 3.589 MTCH1 2.997 ARMC8 0.039 MLF1 3.580 SNX14 2.997 TMTC2 0.039 LIG1 3.580 ARFGAP2 2.995 NPHP4 0.038 GMPR2 3.576 DEPDC1B 2.993 PLEKHM1 0.038 KHK 3.574 MRPL13 2.992 RAB31 0.038 ANKAR 3.568 FOXJ1 2.989 ZFP295 0.038 DCP1A 3.564 FTH1 2.984 DUSP28 0.037 4921525H12RIK 3.556 FKBP1A 2.979 NPC2 0.037 RNPS1 3.539 CCDC47 2.972 ZYX 0.037 1700081D17RIK 3.520 LRRC43 2.971 CKAP5 0.037 UBAC1 3.517 VPS25 2.957 KLK1B27 0.037 CHORDC1 3.507 EFTUD2 2.953 ZMYM3 0.037 FCF1 3.507 BCL2L12 2.947 ZFP532 0.037 IPO13 3.506 IL3RA 2.944 AGPS 0.037 MRPS16 3.505 TRPT1 2.930 RASL12 0.037 HBS1L 3.500 SFXN1 2.912 AP2B1 0.037 GABPA 3.497 GAMT 2.911 SPINLW1 0.036 MTAP1B 3.497 TMEM14C 2.908 ALDH3A1 0.036 DUSP6 3.489 NRF1 2.897 MOCS1 0.036 NAGLU 3.488 DCBLD1 2.894 TBC1D25 0.036 CSNK1D 3.474 HRBL 2.887 NDUFA5 0.036 MCPH1 3.465 LMNA 2.881 NIPA1 0.035 1200011O22RIK 3.458 GCDH 2.872 STARD8 0.035 3000004C01RIK 3.443 MANBA 2.864 ALDOC 0.035 SPAST 3.437 PEX5 2.853 NELF 0.035 DCP1B 3.435 MUPCDH 2.839 SCARA5 0.035 MCL1 3.428 UBB 2.836 CCDC3 0.035 AI314180 3.427 PLEKHB2 2.835 MLKL 0.035 STAU1 3.421 SCD2 2.830 RNASE4 0.035 RNF38 3.421 RABL3 2.830 1700023M03RIK 0.034 D030028O16RIK 3.418 THOC4 2.829 DDX6 0.034 1700042B14RIK 3.410 RAB3A 2.824 INPPL1 0.034 ATP6V1D 3.406 NUDT5 2.821 IRF3 0.034 CYB5R1 3.406 CYGB 2.813 ATP6V1H 0.033 ZFP286 3.396 NSUN4 2.809 MAFG 0.033 CEP57 3.392 GORASP2 2.805 ATG2B 0.033 RPO1-1 3.387 1810055G02RIK 2.799 5730453I16RIK 0.033 RNF166 3.385 PIGA 2.788 4930511J11RIK 0.033 GNAS 3.381 DOLPP1 2.785 TKTL1 0.033 TMEM48 3.370 TSPO 2.784 ZFP410 0.033 ZFP446 3.365 NUDC 2.779 S100A11 0.032 GABPB1 3.363 EG408196 2.771 5033414D02RIK 0.032 SIRT2 3.358 HSD17B7 2.754 HIBADH 0.032 ITIH2 3.357 PPM1B 2.728 ERGIC1 0.032 MGLL 3.350 PPP1CA 2.727 DHDH 0.032 SNX14 3.347 ZDHHC6 2.723 ITGA11 0.031 HIST1H4C 3.342 NUMA1 2.719 ATP6V0A1 0.031 ZBTB22 3.324 CHRNA9 2.715 MAP3K7IP1 0.031 KLK1 3.322 AK3 2.708 BIK 0.031 AOF1 3.316 LOC100045644 2.707 RNGTT 0.031 HSD17B12 3.315 MAPRE1 2.706 ADAMTS2 0.031 4933436I01RIK 3.314 LOC100047619 2.702 OSTM1 0.030 SETD8 3.312 2310061F22RIK 2.698 HEATR1 0.030 HIST1H3A 3.309 NAGS 2.690 2900092C05RIK 0.030 ZFP75 3.303 1200014J11RIK 2.684 MCTS1 0.030 ADAM26A 3.287 KCNK13 2.681 RPO2TC1 0.030 SKAP2 3.276 TNRC6A 2.678 FBXO27 0.030 PITPNM2 3.266 GAB1 2.671 SNX7 0.029 GALT 3.263 LRPPRC 2.664 D12ERTD647E 0.029 LOC433801 3.249 PSMB7 2.662 COL6A1 0.029 GLE1 3.245 CDCA2 2.644 CWF19L2 0.028 PIH1D1 3.244 TXNIP 2.641 BET1L 0.028 DGKH 3.233 SPINLW1 2.640 DDX6 0.028 4933424B01RIK 3.230 HAGHL 2.640 ZFP236 0.028 N4BP2 3.224 ZXDC 2.630 RGMA 0.028 CTNNB1 3.212 DPH4 2.607 USP39 0.028 CCDC117 3.201 VPS33A 2.604 GM648 0.027 LOC100048622 3.201 CKB 2.602 GALNT2 0.027 KIF9 3.201 TRIOBP 2.596 GALNT2 0.026 WDR18 3.196 CSNK1G2 2.578 VTI1B 0.026 FIS1 3.188 POLRMT 2.572 ABCA15 0.026 0610037L13RIK 3.183 NOL5A 2.568 ZBTB24 0.026 CUX1 3.178 BRF2 2.564 RREB1 0.026 ADAM39 3.160 PI4K2B 2.554 SCARA3 0.026 ST5 3.143 BCAT2 2.545 CDC37L1 0.025 DNMT1 3.137 2410025L10RIK 2.536 ZXDC 0.025 C1QTNF3 3.120 SUMF2 2.525 PSCD3 0.025 2410019G02RIK 3.116 ARAF 2.521 ANKRD10 0.025 NUCB1 3.115 DUSP16 2.518 MLLT11 0.025 ITGAD 3.114 6330534C20RIK 2.514 1110020P15RIK 0.025 LRTM1 3.097 UBE2B 2.500 THNSL2 0.025 BC089491 3.093 PLOD3 2.499 SATB1 0.025 1810048J11RIK 3.079 VPS45 2.490 RNF168 0.025 ZDHHC12 3.071 RAB11FIP3 2.483 LCA5 0.025 NDFIP2 3.066 AI747699 2.480 ENPP2 0.025 AES 3.064 ZC3H3 2.467 2010001J22RIK 0.025 FBXL4 3.062 STX18 2.467 PSIP1 0.024 TAF13 3.059 CLPTM1L 2.438 SPOCK2 0.024 ASNS 3.048 TXNDC15 2.424 SLC25A3 0.024 LOC100048613 3.047 RPL22L1 2.406 CD59A 0.024 SLTM 3.046 PRIM2 2.404 NRF1 0.024 FEM1C 3.043 NECAP1 2.396 GALT 0.024 FHL4 3.032 TRIT1 2.394 PTPN11 0.023 ALDH1A1 3.029 LRRK2 2.391 SLC24A3 0.023 LMF1 3.020 6430537H07RIK 2.383 NOL5 0.023 TEX264 3.019 BCKDK 2.381 APPBP1 0.023 FBXO7 3.018 TAZ 2.368 CX3CR1 0.023 WINS2 3.004 DOK3 2.362 LSM8 0.022 SURF1 2.995 ORF28 2.350 AK3 0.022 SEC61A2 2.994 RAB5C 2.349 TBC1D5 0.022 RANBP6 2.983 S100A1 2.340 CCND1 0.022 HIST1H2BC 2.974 0610006I08RIK 2.310 MBNL2 0.022 SIRT1 2.966 SHBG 2.305 RPS27A 0.022 IFT172 2.963 TGFBI 2.305 TIMM8B 0.022 GNL2 2.962 USP9Y 2.301 SCAMP5 0.022 UHRF1 2.957 DDX17 2.296 ABCC5 0.021 POLK 2.954 PPP1R14A 2.284 EXDL2 0.021 KRBA1 2.954 ACRV1 2.281 2600009E05RIK 0.021 CCDC66 2.951 LOC100047707 2.281 CDK2 0.021 CDK5RAP1 2.948 SLC9A8 2.279 NUP88 0.021 NOSIP 2.945 GPRASP1 2.275 ZHX1 0.021 ICA1L 2.944 ADRBK1 2.272 ENTPD5 0.021 CTDP1 2.941 SART1 2.264 SNX2 0.021 F5 2.933 GYLTL1B 2.264 1700072E05RIK 0.021 TMSB10 2.925 D4ERTD22E 2.261 DCLRE1B 0.021 PSMD5 2.920 DDAH1 2.249 CENTG2 0.021 F730014I05RIK 2.911 STARD5 2.240 ZFP579 0.020 EEF1D 2.908 CPSF4L 2.217 VAMP2 0.020 PKNOX1 2.906 LOC100044294 2.215 BMI1 0.020 2410018M08RIK 2.884 HP1BP3 2.204 ADAMTS2 0.020 COG2 2.881 RHOX8 2.201 TCFCP2 0.020 KLHL32 2.881 DUSP28 2.191 EHD4 0.020 PAQR9 2.880 CTPS 2.188 C530028I08RIK 0.019 METTL6 2.877 DHX29 2.187 ELL 0.019 GJC2 2.874 COX6B1 2.183 IQCB1 0.019 TNP2 2.862 BCCIP 2.182 FASTKD1 0.019 DEK 2.860 ALG3 2.167 DOM3Z 0.019 MORN3 2.848 RCHY1 2.166 AA407270 0.019 PIGA 2.847 2810453I06RIK 2.163 FAH 0.019 LOC100046891 2.843 2610101N10RIK 2.156 ASAH1 0.019 SERHL 2.831 FSIP1 2.141 FKBP9 0.019 COPS4 2.822 CDCA4 2.137 C730025P13RIK 0.019 F730014I05RIK 2.812 IFT140 2.134 ALG5 0.018 EXOSC8 2.810 LMAN2L 2.129 6430550H21RIK 0.018 HSP90AB1 2.807 2310005N01RIK 2.124 E4F1 0.018 GOLGA2 2.805 LRP12 2.118 LOC100040899 0.018 TTC23 2.803 RAC1 2.116 MRPL20 0.018 CDKL2 2.802 C1QDC2 2.092 6030443O07RIK 0.018 DCUN1D4 2.795 FLOT1 2.078 VPS24 0.018 HIST1H2AG 2.793 LRRC49 2.071 EIF3EIP 0.018 4833427G06RIK 2.791 RILPL2 2.052 PPP4R1 0.018 HOMER1 2.787 D6WSU176E 2.050 LDB1 0.018 CLPX 2.772 PRF1 2.043 CXXC1 0.018 WDR93 2.765 FAHD2A 2.036 FDPS 0.018 MORC2B 2.748 THOC4 2.029 UNC119B 0.017 LOC100046393 2.743 CDKN1C 2.015 IDH2 0.017 DCUN1D5 2.735 CCDC130 2.013 4930547N16RIK 0.017 TCEA2 2.734 VPS53 2.004 ETF1 0.017 COPE 2.729 GLYCTK 2.004 UBP1 0.017 CDK10 2.727 KPNA1 1.998 4930550C14RIK 0.017 SH3KBP1 2.721 MGLL 1.993 FBXW8 0.017 SDHC 2.720 SPOP 1.993 ING4 0.017 COX18 2.715 SLC2A1 1.967 3110050N22RIK 0.017 TULP4 2.712 LYZL4 1.962 EVI5 0.017 CYPT8 2.706 SMPD1 1.956 D6WSU176E 0.016 ILF3 2.700 TUT1 1.955 UVRAG 0.016 TAF5 2.694 SMN1 1.946 AMIGO2 0.016 2510012J08RIK 2.682 LOC100045551 1.946 SMG7 0.016 LOC100045882 2.682 1200016B10RIK 1.921 RANGAP1 0.016 LOC100048858 2.682 CST9 1.920 PSMA4 0.016 RNF167 2.676 TOMM34 1.912 KPNA3 0.015 ISG20 2.676 OSTM1 1.910 SERPINA5 0.015 SPSB3 2.676 MXRA8 1.904 GBA 0.015 CCDC124 2.671 ANKAR 1.903 SLC40A1 0.015 ZBTB9 2.666 B4GALT1 1.884 DBI 0.015 4921521F21RIK 2.657 ADCK5 1.869 MAT2A 0.015 TEX16 2.657 MCM6 1.866 PIK3R3 0.015 COX4I1 2.654 HTATIP 1.865 TGFBR1 0.015 GLRX5 2.644 TDRD7 1.862 CHMP7 0.015 CYP4F39 2.641 LRRC8D 1.861 LOC631784 0.015 RHOBTB2 2.639 2400003C14RIK 1.854 STX7 0.015 CUGBP1 2.639 PMVK 1.854 PGRMC1 0.015 DNAJC15 2.632 ATP6AP2 1.835 CSNRP2 0.015 MTF1 2.621 TSPAN3 1.832 CHD8 0.015 HTF9C 2.605 LRRC34 1.826 SLC39A11 0.015 2810470D21RIK 2.590 EHBP1L1 1.818 NPM1 0.015 PRR19 2.588 NARG1L 1.813 CHRNA5 0.015 LOC100048436 2.585 BC048599 1.807 BCL2L2 0.015 YOD1 2.582 MIB2 1.775 MRO 0.015 MRPS26 2.580 5830434P21RIK 1.770 LRRC48 0.014 CLCN3 2.569 LRPAP1 1.770 CASK 0.014 CCDC3 2.563 ACOT7 1.768 AA881470 0.014 LAGE3 2.563 LOC666676 1.760 ARL6IP1 0.014 2600011E07RIK 2.560 RABAC1 1.754 GJA1 0.014 ZMYM5 2.557 EPB4.1L3 1.751 4732496O08RIK 0.014 PLK2 2.548 LRP10 1.738 PSMA7 0.014 UQCC 2.546 RNF214 1.737 AGXT2L2 0.014 MED1 2.531 TMEM82 1.734 PPIL3 0.013 KLK1B5 2.515 BC048562 1.728 6030408B16RIK 0.013 RPS21 2.515 FXYD6 1.725 6720456B07RIK 0.013 SSBP2 2.515 2610207I05RIK 1.718 CSTB 0.013 IKZF5 2.513 0610006I08RIK 1.710 AW555464 0.013 LSM2 2.503 CUGBP1 1.709 SPO11 0.013 SKIV2L 2.498 UBE2D2 1.696 TBK1 0.013 ITGAE 2.497 CTPS 1.692 ABCF2 0.013 ZFML 2.492 UBE2D3 1.690 AKR1A4 0.013 1110020C03RIK 2.490 MRPS5 1.682 SOAT1 0.013 NFE2 2.489 VKORC1 1.676 2010107E04RIK 0.012 VPS26A 2.480 GCN5L2 1.675 RBBP7 0.012 SNAPIN 2.476 THADA 1.674 PLEKHA1 0.012 KARS 2.467 2310003H01RIK 1.669 TNRC15 0.012 RBM11 2.466 MAPK6 1.660 SMC1A 0.012 LOC545732 2.465 CLMN 1.656 KCNK13 0.012 RNF141 2.449 KLF1 1.643 RASL11B 0.012 NFS1 2.448 ESAM1 1.642 4732418C07RIK 0.012 1110008F13RIK 2.443 2700087H15RIK 1.637 CDK5RAP3 0.012 ZFP512 2.433 1200015F23RIK 1.635 UBL7 0.011 NDOR1 2.425 PAOX 1.626 FAF1 0.011 MNAT1 2.417 CENPM 1.607 ABI2 0.011 SORT1 2.413 SLC1A3 1.587 NIPSNAP1 0.011 SH3GL3 2.413 VPS28 1.586 SNN 0.011 COG3 2.392 PPP1CA 1.582 KCTD20 0.011 CD164 2.387 SFRS14 1.581 SCARF2 0.011 SCD1 2.374 SMYD2 1.578 HINT3 0.011 LMAN2 2.368 CCDC86 1.577 KPNB1 0.011 COPS5 2.366 WNK1 1.555 ECHS1 0.011 UBE3C 2.362 ZFP187 1.547 2410116G06RIK 0.011 SEC61G 2.360 2310079N02RIK 1.546 ZFP202 0.011 UROS 2.353 PEX3 1.536 FNIP1 0.011 CCDC101 2.353 HMGN3 1.535 E430028B21RIK 0.011 IBTK 2.351 HDAC3 1.528 SERINC3 0.010 CTTNBP2NL 2.346 DNAJC5 1.524 WIZ 0.010 LOC668837 2.332 4921520G13RIK 1.509 9130011E15RIK 0.010 EFTUD2 2.331 ABHD12 1.500 ZFP326 0.010 ARHGAP29 2.327 TRIM23 1.493 BAT1A 0.010 ATF7IP 2.326 TMEM9B 1.483 SLC36A3 0.010 UBE2K 2.325 VGLL2 1.482 3300001P08RIK 0.010 ZMYND11 2.323 MAGOHB 1.480 WBP11 0.010 AP1S1 2.318 2610101N10RIK 1.471 CCDC117 0.009 GM136 2.304 LOC666904 1.467 OCIAD1 0.009 CASC1 2.300 RNGTT 1.463 WASL 0.009 CORO1C 2.299 PDIA5 1.459 DCAKD 0.009 KBTBD7 2.296 PPP1CA 1.458 COX18 0.009 HUS1 2.294 ASCC2 1.454 FBXO28 0.009 ENPP5 2.292 PLA2G2C 1.442 PTS 0.009 ZFP787 2.290 TXNRD2 1.440 SLC12A2 0.009 NME7 2.277 ZRSR2 1.434 SH2B1 0.009 ATOH8 2.273 TIGD2 1.433 USP33 0.009 0610009B22RIK 2.271 PRRX1 1.430 D930001I22RIK 0.009 CSDA 2.262 PIK3R1 1.416 HUS1 0.009 AK1 2.245 DNAJC2 1.409 CTAGE5 0.009 0610031J06RIK 2.238 POLR3B 1.404 2410025L10RIK 0.009 BHMT 2.238 GBL 1.402 BRDT 0.009 AA673488 2.237 RRP12 1.401 STAMBP 0.008 UQCRC1 2.233 PRKACB 1.393 RAB2A 0.008 HSD3B4 2.231 CXXC1 1.391 SCOTIN 0.008 DEDD2 2.230 BEX4 1.389 4632411B12RIK 0.008 ABCF1 2.212 LUC7L 1.374 ECH1 0.008 NR1H3 2.209 ABCF1 1.365 XBP1 0.008 PPM2C 2.203 GPSN2 1.363 DNAJA3 0.008 TCOF1 2.199 PPM1G 1.357 ZFP105 0.008 RRAGB 2.188 CBX5 1.357 RNF103 0.008 DDX26B 2.182 SPOCK2 1.341 CHORDC1 0.008 1700021C14RIK 2.176 WFDC10 1.340 ILF2 0.008 NCKAP1 2.169 RNF14 1.335 TAX1BP1 0.008 BOLA3 2.168 NRIP3 1.318 CBFB 0.008 GPHN 2.147 AGT 1.316 ANKRD47 0.008 UFSP2 2.142 POLR2F 1.297 MAP2K3 0.007 THAP7 2.141 HBP1 1.293 HYAL2 0.007 1700112C13RIK 2.122 MTAP1B 1.293 BC031781 0.007 ACTL7A 2.122 BTBD12 1.283 RAB5C 0.007 SEC13 2.114 LOC100046320 1.279 RBM16 0.007 PMPCB 2.110 1110019N10RIK 1.268 MYST4 0.007 ARHGAP17 2.110 ZYX 1.246 PGS1 0.007 UBE2G1 2.108 PRMT5 1.240 UBXD3 0.007 PNMA1 2.096 TMEM118 1.228 DDX47 0.007 LOC100047052 2.071 CUTC 1.219 TBC1D9 0.007 PCF11 2.070 BTBD10 1.218 A130092J06RIK 0.007 1700008I05RIK 2.054 SLC39A12 1.211 PCCA 0.007 RRN3 2.053 AVPI1 1.206 ATP6V0B 0.006 1700049K14RIK 2.051 LDHC 1.192 SNRK 0.006 PIBF1 2.040 ANGPT2 1.191 INPPL1 0.006 ZSCAN20 2.038 HOOK2 1.185 WSB2 0.006 1700017N19RIK 2.036 LOC100048083 1.184 PJA2 0.006 GLRX2 2.021 ANKRD50 1.181 TOB1 0.006 BARD1 2.018 QRICH2 1.174 ZFP41 0.006 ATP5H 2.018 ACADVL 1.168 SYPL 0.006 PGAM2 2.016 ACTB 1.168 SMU1 0.006 GANAB 2.016 POLE4 1.166 MIR16 0.006 A430005L14RIK 2.015 ODZ3 1.154 ZFA 0.006 ADAM24 2.014 D3ERTD300E 1.152 SYBL1 0.006 CETN3 2.011 PRKRA 1.147 GLRX 0.006 SRA1 2.004 SKP2 1.139 LOC100046746 0.006 MFGE8 2.001 CCDC85B 1.124 GM1698 0.006 4932431H17RIK 1.993 DUSP4 1.122 HS2ST1 0.006 TOP1 1.979 ARL8A 1.119 DAPL1 0.006 DHODH 1.972 C230093N12RIK 1.116 PNRC2 0.006 SH3TC1 1.970 PLA1A 1.112 RFC1 0.005 NDST1 1.962 BC026585 1.112 RNPEPL1 0.005 RALGPS1 1.951 SDCCAG3 1.110 RIN1 0.005 SPARC 1.939 MTMR14 1.106 NUPL2 0.005 4732418C07RIK 1.938 NME7 1.100 SNX4 0.005 MMD2 1.932 NCOA4 1.098 RNF139 0.005 RBM6 1.923 MRPL54 1.094 VGLL2 0.005 2610510J17RIK 1.906 SPARC 1.094 FBXL14 0.005 OPRS1 1.900 LOC100046120 1.090 PCMTD2 0.005 CREM 1.900 WDR32 1.084 ACOT8 0.005 UNC84A 1.894 APOA1BP 1.083 ANK 0.005 FUS 1.874 HOMER1 1.080 RAB11A 0.005 RHOX5 1.865 SEC22B 1.076 2310079N02RIK 0.005 LOC631002 1.865 PIBF1 1.074 BC013529 0.004 LOC100044087 1.863 NFIC 1.073 TRIM36 0.004 EG245376 1.846 CLSPN 1.061 MAPRE1 0.004 LOC670044 1.845 LOC100046406 1.061 TRAPPC1 0.004 RNF139 1.834 MEF2C 1.057 RDH11 0.004 BRPF1 1.824 STK19 1.055 IMP3 0.004 PPS 1.821 ME2 1.047 PDLIM3 0.003 COPS6 1.813 2410016O06RIK 1.029 ADRBK1 0.003 TEKT3 1.804 LOC100046568 1.027 SFRS7 0.003 AIFM1 1.799 AQP7 1.026 0610010K06RIK 0.003 RAD51 1.780 SLC38A10 1.023 TIMM8A1 0.003 1700108M19RIK 1.774 NCKAP1L 1.019 TBC1D22A 0.003 1810030N24RIK 1.767 SEC61B 1.017 LGALS3 0.003 LOC382010 1.763 ZFML 1.012 CYP2D22 0.003 SPAG9 1.751 DLD 1.011 PIGK 0.003 UBE1C 1.742 ZFYVE16 1.010 ZCCHC6 0.003 ART5 1.742 INHA 1.008 1110020P15RIK 0.003 MRPS28 1.737 ZC3H10 1.006 EML4 0.003 ZWILCH 1.735 1110017D15RIK 1.005 AP2M1 0.003 BCAR1 1.733 CBLB 1.003 NUDT5 0.003 LRPPRC 1.731 SORT1 1.001 SART3 0.003 CCDC38 1.727 WBP2 1.000 SERTAD2 0.003 KPNA3 1.726 RANGAP1 0.997 DPM1 0.002 ADAM15 1.723 DAG1 0.988 CLDN23 0.002 RAB1 1.722 EG433182 0.982 DHRS7B 0.002 RBM26 1.715 DYNLL1 0.981 NDUFA5 0.002 KLHDC4 1.708 TBC1D14 0.980 SUMO1 0.002 COPG 1.701 GM136 0.972 RNF44 0.002 RFFL 1.699 TCTN3 0.968 MORC2A 0.002 ZFP704 1.697 FARSB 0.966 LOC667250 0.002 CMBL 1.695 RAB21 0.965 HISPPD1 0.002 BTBD9 1.694 HK1 0.962 1700027A23RIK 0.002 SNAI3 1.688 MED8 0.957 TCTN3 0.002 LOC100039532 1.682 RP23-297J14.5 0.947 2310046K01RIK 0.002 MON1B 1.676 LRRC44 0.940 DEDD 0.002 PPP2R2B 1.671 MCL1 0.933 2010003O18RIK 0.002 D2ERTD750E 1.659 PCNA 0.928 SPATA22 0.002 WARS 1.646 ZFP318 0.916 PTBP1 0.001 CD9 1.645 AV249152 0.915 HNRPH1 0.001 BAG5 1.644 SESN1 0.911 CLCN7 0.001 GTF3C1 1.640 AES 0.911 EN2 0.001 UBE2Q2 1.638 MAK10 0.909 1810064F22RIK 0.001 TTC3 1.637 ILVBL 0.905 TGM2 0.001 SRP19 1.631 SLC30A1 0.904 ADAM9 0.001 KCNK2 1.623 RBM4 0.901 PCGF5 0.001 LOC100047935 1.619 RAP2C 0.900 EID3 0.001 4930519N16RIK 1.609 FAHD2A 0.899 LYRM2 1.599 MRPS18B 0.893 NID1 1.596 6330505N24RIK 0.890 ZCRB1 1.591 PPFIBP2 0.890 TTC24 1.579 VGLL4 0.888 SUMO1 1.568 HSPA9 0.886 ADAM4 1.553 GP38 0.885 4921515J06RIK 1.543 PIGYL 0.883 ADSL 1.537 MDM1 0.879 SPRED1 1.535 COMMD3 0.875 POLB 1.522 LIG4 0.868 2610020H08RIK 1.521 SLC25A26 0.856 SSBP3 1.506 SEC14L1 0.855 KLC4 1.501 HDC 0.851 LOC434960 1.500 PRPF3 0.848 MRPL20 1.498 WFDC6A 0.847 ATP2A2 1.482 PRSS35 0.846 5330431N19RIK 1.476 PHF23 0.843 BZW1 1.475 STARD5 0.843 EG625054 1.468 SFMBT1 0.843 ALS2 1.463 SLC12A2 0.841 DLGAP4 1.454 TLK2 0.833 ECE1 1.453 MNAT1 0.830 ACAD9 1.450 RPL22 0.818 CBFB 1.448 PRKCB1 0.815 CHIC2 1.448 ARS2 0.810 1200014J11RIK 1.443 PAIP2B 0.805 STK39 1.435 SH3GLB2 0.773 UNC93B1 1.435 MED15 0.761 CYPT9 1.425 IBTK 0.741 1700023B02RIK 1.425 IGSF11 0.728 TMEM79 1.413 ATP6V0A2 0.721 NBL1 1.407 2700060E02RIK 0.721 4930579C15RIK 1.407 BRD9 0.704 COIL 1.400 PDLIM4 0.693 HMGB4 1.399 1810049H19RIK 0.684 BRDT 1.394 CTPS 0.680 ZFP57 1.392 MGEA6 0.678 EIF2AK1 1.389 MYLK 0.677 GRWD1 1.379 PAPSS1 0.675 WDR8 1.366 STRA13 0.668 JOSD3 1.363 IGF2 0.658 CCDC53 1.363 ODF4 0.638 CREBL1 1.359 COPA 0.637 RHOG 1.350 GRN 0.633 SEMA7A 1.345 SCHIP1 0.626 PREB 1.343 HSPD1 0.616 BC050811 1.342 SYF2 0.615 MOBK1B 1.340 IPO13 0.595 1600012H06RIK 1.337 ERGIC3 0.594 NRBP1 1.331 C1QC 0.591 SLC25A1 1.327 CCDC88A 0.585 ZMPSTE24 1.327 ATP9A 0.581 TCF20 1.324 PDCD10 0.578 PHF7 1.323 RABGAP1 0.577 NIBAN 1.323 TADA3L 0.560 1700123K08RIK 1.317 KLF13 0.547 1700018C11RIK 1.316 KLF12 0.542 DHDH 1.309 GM129 0.541 DHRSX 1.308 RBAK 0.535 MRPL48 1.308 DTD1 0.533 6720467C03RIK 1.301 SLC17A7 0.529 MMP14 1.300 TANK 0.527 SLC38A2 1.293 4930432K21RIK 0.515 SMPD1 1.286 SORT1 0.512 CSE1L 1.284 AP3B1 0.495 4933424B01RIK 1.284 SLC11A1 0.485 ZFP36L1 1.282 BAT1A 0.484 LOC100045617 1.276 PDCD6IP 0.479 LYAR 1.264 MINPP1 0.479 SURF4 1.264 PCF11 0.471 YWHAZ 1.260 EMB 0.468 RTN1 1.256 1700030J22RIK 0.456 CSL 1.254 GSG1 0.451 APIP 1.254 GTF2H4 0.450 EXDL2 1.251 SORT1 0.449 MAST2 1.249 LSM3 0.445 SAPS2 1.246 PLCD4 0.445 GPS1 1.239 DGCR2 0.442 CDKL1 1.238 PHF2 0.432 TMEM131 1.237 AUP1 0.426 FASTKD5 1.234 RPP21 0.426 NDUFB5 1.222 ASB4 0.423 GORASP2 1.215 RPL39 0.408 3830408P04RIK 1.212 TPPP3 0.405 2510010F15RIK 1.207 RSRC2 0.397 INTS5 1.205 GTSE1 0.395 DNM1L 1.204 POLR2E 0.393 SSSCA1 1.203 FEN1 0.389 MYH10 1.202 KCTD10 0.385 DTNBP1 1.188 ORF9 0.376 HEXIM1 1.187 SPATA7 0.376 KNS2 1.184 SLC27A6 0.365 TMEM55A 1.180 2610034M16RIK 0.364 NUPR1 1.179 GNPDA1 0.357 CUTA 1.167 ZFP523 0.353 SHBG 1.166 SQLE 0.348 FOXQ1 1.163 ANKFY1 0.344 DYRK2 1.162 SEC61A2 0.340 GALNT1 1.158 4930473A06RIK 0.339 1700034I23RIK 1.153 SMARCD2 0.338 FANCL 1.151 RTN3 0.331 NDUFA3 1.149 PIWIL1 0.326 HIST1H4I 1.144 DDX49 0.324 NUDT21 1.143 TRP53INP2 0.323 NDUFS4 1.140 2400010D15RIK 0.318 VPS13A 1.139 CCDC27 0.305 RRAGD 1.138 LAMB3 0.303 1700057G04RIK 1.127 SPATA7 0.297 2810485I05RIK 1.125 MID1IP1 0.296 1700109H08RIK 1.120 ANXA11 0.296 LTBP4 1.119 SLC35A4 0.293 SNRPD1 1.118 PPFIBP1 0.289 PEX5L 1.118 PLA2G6 0.289 MTBP 1.116 40057 0.288 LOC100048299 1.113 GAPDHS 0.286 STBD1 1.109 AKAP3 0.286 SPAST 1.107 CCNB1 0.284 ACTR10 1.099 EIF4EBP2 0.276 PEA15 1.098 NOC4L 0.275 AMIGO2 1.091 2610003J06RIK 0.274 CCDC108 1.089 GGT7 0.272 LOC100046568 1.083 RBMS2 0.267 GPRASP1 1.081 NR1H2 0.264 TBL1X 1.080 RAPSN 0.257 NCDN 1.070 TMEM33 0.255 NOB1 1.065 PMM2 0.253 ITM2B 1.064 AP2A1 0.251 SUMO2 1.063 PLEKHG2 0.246 KLHL9 1.060 4921536K21RIK 0.245 SERPINA3H 1.057 KRBA1 0.245 PLEKHO2 1.056 SUMO1 0.241 NOS3 1.054 MMD2 0.240 HSPA2 1.049 PMPCB 0.240 SOX30 1.046 GSTM1 0.239 SLC39A4 1.045 ATG3 0.238 SNX14 1.042 AI662250 0.238 SPG21 1.041 ATMIN 0.231 FNDC5 1.041 ASPSCR1 0.229 AKR1A4 1.039 ASB3 0.228 ORF28 1.038 CTNNB1 0.226 VAPB 1.038 IFNGR2 0.223 VDR 1.037 ZBTB32 0.223 CSK 1.036 MAPK6 0.211 HECTD2 1.032 NIBAN 0.210 GM648 1.030 PCSK4 0.210 STX1A 1.029 NLRP14 0.209 PDCD2L 1.029 VPS33B 0.208 LOC100044324 1.025 SNX2 0.205 GPRASP1 1.024 4933426M11RIK 0.202 ZFP474 1.022 MED15 0.202 UBAC1 1.021 CD96 0.199 WFDC10 1.020 AKT3 0.197 CARHSP1 1.019 TMC7 0.192 PEX3 1.019 PKIG 0.191 CRKL 1.016 ADRM1 0.190 SEL1L2 1.015 M6PR 0.189 KCNG4 1.015 OBFC2B 0.188 IMP3 1.015 2410014A08RIK 0.188 SSXB5 1.010 ABCA14 0.188 STARD5 1.006 LOC100048046 0.187 LOC100044475 1.006 DDT 0.186 TYMS 1.005 ARHGAP1 0.184 SYNCRIP 1.004 SMARCE1 0.183 OGDH 1.002 ZFPL1 0.181 PSME3 1.000 BC022224 0.180 TMEM204 0.998 ADAL 0.179 AIFM1 0.998 GAS6 0.176 CNTD1 0.991 VPS33B 0.175 2610034E13RIK 0.991 FKBPL 0.174 NIP7 0.987 P2RX2 0.170 2510006D16RIK 0.984 NUDT19 0.169 CAPZB 0.983 HTATIP 0.168 SESN2 0.982 LMNA 0.167 ATP6V0D1 0.982 CS 0.167 TULP2 0.981 AA467197 0.160 SLC25A20 0.980 TNFRSF21 0.159 MYH11 0.980 AGPAT1 0.158 AKT3 0.979 KCNK2 0.156 EG668668 0.979 GEMIN7 0.155 SUMO1 0.978 PRPF38A 0.155 ORMDL3 0.976 NUDT16L1 0.151 TNIP1 0.975 ATP13A1 0.150 SLC25A38 0.972 TPM4 0.150 SAR1A 0.970 SLC9A10 0.149 KIF4 0.966 PPS 0.149 PRMT5 0.965 PBK 0.149 KLK1B9 0.964 SLTM 0.148 SAMD4B 0.962 TMEM8 0.148 UNC45A 0.960 D0H8S2298E 0.146 4933434I06RIK 0.958 ZFP189 0.144 SUSD3 0.957 TTC4 0.144 DDX4 0.956 PKM2 0.144 TXNL4A 0.955 RARS 0.142 ESAM1 0.952 ALDOA 0.142 2700055A20RIK 0.952 MAPK9 0.142 OGFOD2 0.952 ACADSB 0.141 SF3B4 0.951 AHCTF1 0.141 NEDD8 0.950 AA536749 0.140 ARFIP2 0.949 5430432M24RIK 0.139 ACTB 0.949 RAD17 0.138 ZFP217 0.949 TMEM19 0.138 LOC100046457 0.948 BC056474 0.137 SRA1 0.948 AQP11 0.136 SBDS 0.947 TAPBP 0.136 EGFL7 0.945 DHPS 0.136 9430015G10RIK 0.944 SRI 0.135 CCDC39 0.943 1700013G24RIK 0.133 OBFC2B 0.942 COPS7A 0.131 TSC22D1 0.941 ACTR10 0.130 NDUFAF1 0.941 TIMM10 0.129 TSSC1 0.940 TRIP12 0.128 RNF219 0.939 ITIH2 0.128 BCAT2 0.936 SLC9A8 0.128 ZFP771 0.935 3300001P08RIK 0.127 SPATA7 0.934 FBXO9 0.127 CEBPB 0.933 BRDT 0.126 WDR31 0.933 BC016423 0.126 E330018D03RIK 0.932 AMY1 0.125 ITGB5 0.927 SETDB1 0.124 TG 0.927 NUDT4 0.124 PROK2 0.926 ZFP238 0.124 SLC12A6 0.926 1110006G06RIK 0.123 STIM1 0.922 NDUFS6 0.123 2700062C07RIK 0.922 METTL9 0.121 DUS1L 0.920 MTMR2 0.120 SGCA 0.920 AMDHD2 0.120 4921523A10RIK 0.918 RHOB 0.120 CNTD1 0.917 PPP1R10 0.119 LARP2 0.914 1110039B18RIK 0.119 APBA1 0.913 TMEM141 0.119 GOSR2 0.912 KLC2 0.118 BECN1 0.910 LANCL1 0.118 RNF25 0.908 TRAFD1 0.118 CCDC131 0.908 VPS16 0.116 FSIP1 0.908 TEX16 0.116 GOSR2 0.903 GTF2E2 0.116 5330431N19RIK 0.903 TMEM186 0.116 NCOR1 0.895 ZFP512 0.116 BC067068 0.894 UCHL5 0.115 1700055M20RIK 0.892 TSNAXIP1 0.115 SOCS7 0.888 RPL28 0.114 4930555G01RIK 0.887 AAMP 0.113 1700080E11RIK 0.886 IHPK1 0.112 USP39 0.884 RHBDD1 0.112 ASCC2 0.881 HSD17B12 0.112 CDSN 0.880 ATP6V0D1 0.111 ZFP346 0.880 WIPI2 0.111 TEX9 0.879 DCPS 0.111 LOC100045439 0.877 SPCS1 0.111 BCDIN3D 0.876 CSNK1G3 0.111 DYNLL2 0.875 ASNSD1 0.111 FLOT2 0.868 KLC4 0.111 6430527G18RIK 0.861 SPAG9 0.110 CD55 0.852 SPINK8 0.110 ENSA 0.848 GTPBP8 0.110 HSPA9 0.848 TSN 0.110 LOC100048480 0.843 RHEB 0.109 CLEC4G 0.842 UBE2M 0.109 SPATA6 0.841 CD74 0.109 PLCZ1 0.841 BAT2 0.109 MRPL22 0.840 4930572J05RIK 0.108 GNB1 0.840 NDUFB5 0.108 ODF4 0.839 COX6A1 0.108 EFR3A 0.836 TMEM192 0.108 LOC100047615 0.832 RPS27L 0.107 1700018F24RIK 0.832 SLK 0.107 SGK3 0.813 COPS7B 0.107 LIPH 0.811 1700034H14RIK 0.107 TRPC4AP 0.807 SLC15A4 0.107 CRLS1 0.802 IK 0.107 ENAH 0.794 MRPS15 0.106 MGC118210 0.792 FLNA 0.106 ANXA6 0.788 TGFBR2 0.106 NMD3 0.786 SDHC 0.106 DHRSX 0.784 EIF4G2 0.106 RPP21 0.781 6530404N21RIK 0.105 SLC22A14 0.777 DGCR8 0.105 PRPF38A 0.765 MRPS35 0.105 OAZ2 0.758 UQCRH 0.105 LOC100047214 0.752 ZSCAN2 0.105 4932413O14RIK 0.737 SF1 0.105 MCM7 0.735 GGA1 0.105 TESSP3 0.726 LAMB2 0.104 IQGAP1 0.726 1700012L04RIK 0.104 POLE3 0.712 GABARAP 0.104 1700025E21RIK 0.709 BC050811 0.104 EHD4 0.707 AI449175 0.104 GRIPAP1 0.691 1700003M02RIK 0.104 ARFRP1 0.689 2610301B20RIK 0.103 H2AFY 0.682 UPP2 0.103 HNRPL 0.664 4930428D18RIK 0.102 1700029H14RIK 0.659 3110037I16RIK 0.102 GORASP2 0.657 MRPL43 0.102 EG545047 0.654 1700016K19RIK 0.102 LOC100044170 0.646 1700034I23RIK 0.102 CML1 0.641 GGN 0.102 BCL2L1 0.632 ZFP346 0.101 4933407N01RIK 0.629 HERC4 0.101 IK 0.629 THTPA 0.101 LOC625480 0.621 C80913 0.101 LOC100047368 0.620 RAI14 0.101 40057 0.615 SLC41A3 0.100 RBM15 0.611 POU6F1 0.100 EG433182 0.610 LOC100044087 0.100 AURKC 0.600 SAMD4B 0.100 PFN1 0.587 SUMO2 0.100 SLC30A7 0.579 MORN2 0.100 TRIM54 0.578 PHOSPHO1 0.100 ODF1 0.573 LIMD1 0.099 PPP2R5C 0.572 SMU1 0.099 SCMH1 0.571 MAT2A 0.099 ZFP313 0.570 CALR 0.099 DDX19B 0.569 1700022C21RIK 0.099 NID2 0.559 MRPL40 0.099 ZFP143 0.557 NDST1 0.099 MED15 0.556 PMFBP1 0.098 1300001I01RIK 0.556 SPEER2 0.098 VDAC1 0.554 TIMP1 0.098 LRRC46 0.553 MAD1L1 0.098 TMEM111 0.545 MT1 0.098 NLN 0.541 BHLHB9 0.098 KIF18A 0.531 LOC100048105 0.098 MRPL51 0.519 NRBP1 0.098 MFAP3 0.514 DAK 0.097 ABHD4 0.508 MTERFD1 0.097 LOC100048105 0.501 ARL4A 0.097 NUPL1 0.495 ERO1LB 0.097 PSME4 0.488 LOC677317 0.097 NCL 0.488 CLEC16A 0.097 CCDC71 0.479 SAV1 0.097 EG638695 0.477 2610209M04RIK 0.097 DERL2 0.472 PRPF40B 0.097 METAP1 0.471 HEXIM2 0.097 XRN2 0.468 FBXO8 0.097 CD320 0.468 FRS2 0.096 CC2D1B 0.458 BC021790 0.096 MAGOHB 0.447 H1FX 0.096 PSMD7 0.440 LRCH4 0.096 A630095E13RIK 0.437 TEX22 0.096 MAP2K7 0.420 MDH2 0.096 FOXJ3 0.419 1700012H17RIK 0.095 RWDD4A 0.413 SNX7 0.095 RPAIN 0.408 PREP 0.095 ANKRD54 0.406 GPRASP1 0.095 TOR1AIP1 0.401 MYG1 0.095 TPM1 0.400 FHAD1 0.095 CCDC23 0.395 RAD23A 0.095 ANKRD7 0.395 SCOC 0.094 TEAD2 0.393 CAPNS1 0.094 AGPAT3 0.392 HDAC6 0.094 GSG2 0.389 NUP85 0.094 USP3 0.388 ELF1 0.094 HSPBAP1 0.383 ANKRD54 0.093 NBR1 0.380 PER1 0.093 FBXO36 0.378 PDCD5 0.093 OTUB1 0.375 2900062L11RIK 0.093 PPP1R14A 0.362 SNX14 0.093 ROM1 0.357 JMJD1A 0.093 PDCD4 0.355 MEMO1 0.093 EIF4ENIF1 0.354 KLF4 0.093 PPAT 0.352 TLE3 0.092 GABARAP 0.344 PCBP1 0.092 FKBPL 0.342 SOCS7 0.092 ZSCAN2 0.340 ZFP474 0.092 ACTA2 0.336 MAFG 0.092 ZFP523 0.334 MAK10 0.092 COX5B 0.333 ERP29 0.092 RBM42 0.327 ADAM9 0.092 RAF1 0.327 TUBB2C 0.090 NOS3 0.322 SIRT1 0.090 NFX1 0.320 SETD1B 0.090 R3HDM1 0.320 1700013N18RIK 0.090 WDR51B 0.314 ZFP451 0.089 PAIP1 0.312 LRRC52 0.089 KLF4 0.310 FNIP1 0.089 CBLB 0.308 METT11D1 0.089 CTPS 0.303 HSP90B1 0.089 PON3 0.297 FNIP1 0.088 HARBI1 0.285 LYPLA1 0.088 RAD51 0.285 HNRPM 0.088 ARHGEF18 0.283 COPS5 0.088 CD82 0.283 CCDC115 0.087 AATF 0.279 ATPAF2 0.087 TMEM85 0.278 CPSF1 0.087 PTPRR 0.275 RBM18 0.087 STK4 0.272 NARG1 0.087 LTK 0.271 4932418E24RIK 0.087 AI894139 0.268 BECN1 0.086 TCEB2 0.268 ZFP219 0.086 POLR2D 0.262 LOC100045963 0.086 PGLS 0.262 FUS 0.086 PABPC1 0.258 HNRPUL1 0.086 BCL10 0.257 CCDC123 0.086 CLN6 0.256 MIPEP 0.086 ABCC3 0.253 SLC4A1AP 0.086 MFAP3L 0.253 GSTM5 0.086 ZFP282 0.251 SPG7 0.085 DPEP3 0.250 MTMR3 0.085 PDIA4 0.243 LOC100045617 0.085 JOSD1 0.240 CLN6 0.085 VIM 0.240 RPO1-3 0.084 HMGA1 0.238 DONSON 0.083 USP8 0.237 LOC100045887 0.083 1110039B18RIK 0.236 DDX51 0.083 ARL6IP2 0.233 RNF14 0.083 GM906 0.231 FAH 0.082 EFHA1 0.230 STK31 0.082 LOC100046741 0.230 TMC6 0.082 NUP133 0.229 GK2 0.082 SUCLG2 0.229 EFTUD1 0.081 AP3B1 0.228 BTBD1 0.081 ACR 0.228 TUBA3A 0.081 ABHD12 0.226 ACAD10 0.081 PRIM2 0.225 UNC119 0.081 PEX7 0.225 MTMR12 0.081 SMC1A 0.225 PHKG2 0.081 LOC667609 0.223 PRPF40A 0.080 1700052N19RIK 0.222 TTC29 0.080 CREB3L4 0.221 CDKN1A 0.080 HCFC1 0.217 MARK2 0.080 ATPAF2 0.215 PPA2 0.080 BTF3 0.215 MTAP1S 0.080 BAG3 0.213 CALCA 0.079 FBXL13 0.211 2010007H12RIK 0.079 POMT1 0.209 4933407N01RIK 0.079 EIF2S3X 0.209 SMG7 0.079 4933404M02RIK 0.208 THOC1 0.079 SNX13 0.206 PRDX1 0.078 FUBP1 0.200 WIPI2 0.078 TCP1 0.200 PKN2 0.078 DCBLD1 0.199 CDV3 0.077 LAPTM4A 0.199 2810452K22RIK 0.076 ZFP523 0.199 ERCC3 0.076 RHBDD1 0.194 COX5A 0.076 A830059I20RIK 0.194 SNTB2 0.076 BAZ1A 0.192 EHD1 0.075 5730596K20RIK 0.192 OTTMUSG00000000971 0.074 PTPRS 0.191 PSMC5 0.074 ISCA1 0.191 BAG5 0.074 HSF5 0.191 PDE1C 0.073 BC017643 0.190 CHCHD7 0.073 ACTB 0.189 CATSPER1 0.073 HSPA8 0.188 4930408O21RIK 0.073 SS18 0.187 SLC12A2 0.073 GPRC5A 0.182 MARCKSL1 0.073 MBOAT2 0.181 RABEP1 0.072 DMKN 0.179 PWWP2B 0.072 1810055G02RIK 0.178 BTG4 0.070 4632411B12RIK 0.177 CDC45L 0.070 TUBG1 0.177 BC085271 0.069 HNRNPA2B1 0.175 ARPC4 0.069 LOC100039571 0.173 HSF5 0.069 DUSP13 0.172 SMEK2 0.068 LOC100047717 0.170 NDUFA1 0.067 TIMM8A2 0.168 CSTF2 0.067 GNA12 0.167 PI4KB 0.067 TLE1 0.164 CHKB 0.067 ZFP653 0.164 TFRC 0.067 ATP8A2 0.163 IQGAP2 0.067 LOC100047214 0.162 TOPORS 0.066 NELL2 0.162 CTTNBP2NL 0.066 BC002230 0.159 RWDD3 0.066 DNTTIP1 0.157 APPBP1 0.066 ZDHHC1 0.157 PAPD1 0.066 RASSF7 0.157 TSC2 0.066 UGP2 0.157 1500012F01RIK 0.066 CLCN7 0.156 FETUB 0.066 ACTR5 0.155 ARFGAP1 0.066 BC048651 0.153 AP3B1 0.066 4930504E06RIK 0.152 EFCAB2 0.064 MFSD11 0.151 LOC100048295 0.064 UTP6 0.150 ACTR3 0.064 SOCS2 0.150 CAPRIN1 0.064 1700003M02RIK 0.150 LOC625480 0.064 ADD1 0.150 SERBP1 0.063 MLLT1 0.148 RRAS2 0.063 CHD4 0.148 ADORA1 0.063 ARPC4 0.147 2610029G23RIK 0.063 EXOSC2 0.147 BC038156 0.063 ARL2BP 0.147 MTCH2 0.063 CCDC115 0.147 TTLL6 0.062 1700067C01RIK 0.145 TERF2IP 0.062 DTNB 0.142 SEPW1 0.062 C430004E15RIK 0.141 UPF3A 0.062 IGSF3 0.141 SELL 0.061 SSU72 0.140 LOC100048071 0.061 TSEN54 0.139 RPS16 0.061 DGAT1 0.138 E4F1 0.061 RNU6 0.138 STK40 0.061 TNFRSF12A 0.137 KLHL7 0.061 TES 0.137 FH1 0.060 TAF9 0.136 SERINC1 0.060 BC050210 0.136 SLC39A6 0.060 LPGAT1 0.135 ZPBP 0.059 ZFP661 0.135 FBXO18 0.059 SQLE 0.135 BC003993 0.059 CITED4 0.135 SLC25A25 0.059 GRLF1 0.135 AP2B1 0.059 1500001M20RIK 0.134 ACTN1 0.059 DNAJB6 0.134 ARMC10 0.059 GALNT1 0.133 KCTD6 0.059 SDCCAG3 0.133 YWHAB 0.059 CCDC77 0.132 TRAPPC1 0.058 RNF167 0.132 HSPA1L 0.057 PEX2 0.131 TTYH1 0.057 2410016O06RIK 0.130 MFAP3 0.057 AA536749 0.130 GM1698 0.056 FABP9 0.130 CRIPT 0.056 ARL6 0.129 NXT1 0.056 POU6F1 0.129 ADAM6 0.056 GDE1 0.129 PLEKHJ1 0.056 SLC39A3 0.127 ZFP654 0.056 1200015F23RIK 0.127 PEX19 0.055 DBR1 0.127 USP2 0.055 EEF2 0.126 TMPO 0.055 LOC100046792 0.126 9330134C04RIK 0.054 LRP12 0.126 BC053749 0.054 PEX5 0.126 DUSP6 0.054 SPSB1 0.126 CCDC95 0.054 ACOX3 0.125 CDH23 0.054 SAMD8 0.125 RFX2 0.053 1700113I22RIK 0.125 ABCF1 0.053 PFKM 0.125 TIMP1 0.053 SOCS6 0.124 COPG 0.053 PARD6G 0.124 1700081D17RIK 0.053 9130404D14RIK 0.121 EVL 0.053 TMED4 0.121 4933430I17RIK 0.053 D6WSU163E 0.121 NT5E 0.053 CTGF 0.120 POLG 0.053 SCRIB 0.120 GTF2E2 0.052 CDR2 0.120 VAT1 0.052 MLKL 0.120 BCL7B 0.052 CCDC85B 0.119 PRM3 0.052 CASC1 0.118 SDCBP 0.052 SLC25A44 0.118 EMG1 0.052 DTX4 0.117 COQ5 0.052 NUT 0.117 NOL10 0.052 GALC 0.116 HOOK1 0.051 OS9 0.116 RNF185 0.051 SLC30A5 0.115 P42POP 0.051 KIFC2 0.115 INTS12 0.051 PCMT1 0.115 GCAP14 0.051 POMGNT1 0.115 RILPL1 0.051 ARHGAP20 0.114 NMD3 0.051 GPR108 0.114 MTPN 0.050 LMNB1 0.114 LOC100044776 0.050 TMEM39A 0.114 PRKAR1A 0.050 CSTF3 0.114 FAIM 0.049 DNAJC2 0.113 LOC100046996 0.049 TMEM50B 0.113 MAPK15 0.049 INSIG1 0.113 UBE2J2 0.049 PIAS2 0.113 FBXL6 0.049 RREB1 0.112 SPATA3 0.049 ZC3H3 0.112 HDDC2 0.048 ITPKA 0.112 ARF2 0.048 TPST1 0.111 ARGLU1 0.048 PSEN1 0.111 EIF4EL3 0.048 NPM1 0.111 MTF2 0.048 GLO1 0.111 ACSL1 0.048 AMHR2 0.111 SLC35C1 0.048 WIZ 0.110 MRPL3 0.048 LOC100047619 0.109 SEC22A 0.048 SOHLH1 0.109 THAP1 0.047 2700038C09RIK 0.109 AVEN 0.047 RAPSN 0.109 TBC1D8 0.047 ZMPSTE24 0.108 UBE2G1 0.047 DAG1 0.108 ZFP472 0.047 CLSTN3 0.108 SLC9A3R1 0.047 PROK2 0.108 BTG1 0.047 LOC677144 0.108 AI894139 0.047 ZFP410 0.108 ELF2 0.046 USP22 0.107 MATR3 0.046 STAP2 0.107 FBXO18 0.046 SLC24A3 0.106 PTP4A2 0.046 GOLGA7 0.106 NACA 0.046 DAPK3 0.105 EI24 0.046 TIMP3 0.105 IL11RA1 0.046 LEMD2 0.105 4933407N01RIK 0.046 PRKACB 0.105 BC087945 0.045 DMKN 0.105 ABAT 0.045 CHMP4B 0.105 KCNB1 0.045 SAPS1 0.105 ATP5A1 0.045 NRBP2 0.105 PSG23 0.045 FPGS 0.105 BCAP29 0.045 HDAC3 0.105 H2AFX 0.045 UTX 0.105 SOCS7 0.045 TMEM198 0.104 ARHGAP18 0.044 SDCBP 0.104 SSFA2 0.044 4930403O06RIK 0.104 DDX4 0.044 GM249 0.104 FKBP7 0.044 VPS33A 0.104 UBAC1 0.044 ERLIN1 0.104 DNAJB6 0.044 PLEKHJ1 0.104 RAD23B 0.044 ANXA11 0.103 PJA2 0.044 PPM1G 0.103 3830406C13RIK 0.044 NFKB1 0.103 ACTR5 0.044 USPL1 0.103 EIF4E3 0.043 SETD1A 0.103 4833439L19RIK 0.043 TGM2 0.102 CLTB 0.043 TTLL8 0.102 1700007I06RIK 0.043 RPO1-3 0.102 DCP1B 0.043 LZTR1 0.102 ZHX1 0.043 ELF2 0.102 RAB9 0.043 ADRBK1 0.102 ADORA1 0.043 STARD5 0.102 COX6C 0.043 UFSP2 0.101 PFKFB2 0.043 LMNA 0.101 UBE2Q2 0.043 MKNK2 0.101 RAB33B 0.042 ATP5J 0.101 CCDC23 0.042 NEO1 0.101 PLXDC1 0.041 2600001B17RIK 0.101 KLHDC3 0.041 ADIPOR1 0.101 EFTUD2 0.041 CIDEA 0.100 NDST1 0.041 FARSB 0.100 LRTM1 0.041 TSSC1 0.100 H2AFY 0.041 MED19 0.100 SLC35C2 0.041 STARD7 0.099 BEX2 0.041 LOC100046406 0.099 IFT122 0.040 TRAPPC3 0.099 EIF5 0.040 PIAS3 0.099 SGCB 0.040 RABGAP1 0.099 RAB28 0.040 OLFR748 0.098 LRRC23 0.040 CACNA2D1 0.098 PAPOLG 0.039 DES 0.098 GOSR2 0.039 MYL1 0.098 UCHL1 0.039 SCAMP4 0.098 PACSIN1 0.039 ATP9B 0.098 GM773 0.039 FOLR1 0.097 CCDC71 0.039 TLK2 0.097 AGGF1 0.039 D10BWG1379E 0.097 CEP70 0.038 GM962 0.097 IMMT 0.038 ANK2 0.097 D10WSU52E 0.038 5033414D02RIK 0.097 1700109H08RIK 0.038 WFDC6B 0.097 CMPK 0.038 ADRM1 0.097 MRPS30 0.038 GANC 0.097 AI449175 0.038 ETSRP71 0.096 TFDP2 0.038 UGCG 0.096 KRBA1 0.037 BC016495 0.096 PLD3 0.037 SPERT 0.096 LIAS 0.037 SLC6A9 0.096 PHOSPHO2 0.037 UNC13B 0.096 SDCCAG1 0.037 MBD3 0.096 ZFP715 0.037 PTPLA 0.096 M6PRBP1 0.037 40238 0.096 TIAL1 0.037 LRRC28 0.096 PPP1CB 0.037 2900006K08RIK 0.096 FNDC5 0.037 USP50 0.096 DDX1 0.036 MCM4 0.095 SHPRH 0.036 DMTF1 0.095 USP6NL 0.036 RAB3D 0.095 AOX3 0.036 MTAP7D1 0.095 OPTN 0.036 MYD116 0.095 SMARCE1 0.036 KIFAP3 0.095 1700010A17RIK 0.036 CTNS 0.095 CCDC38 0.036 WDR21 0.095 DCTN4 0.035 PCYOX1 0.095 KPNA3 0.035 LOC100045981 0.095 EIF3S10 0.035 DYDC2 0.095 DDX1 0.035 LOC100047911 0.094 RTKN 0.035 GALNT11 0.094 G430022H21RIK 0.035 DNAJC12 0.094 1700025G04RIK 0.035 ATG7 0.094 KPNA6 0.035 PTPRK 0.094 HOOK1 0.034 TLE6 0.094 5330431N19RIK 0.034 D930001I22RIK 0.094 LOC100048858 0.034 MTHFS 0.094 ZMPSTE24 0.034 NPTN 0.093 FXN 0.034 CALR3 0.093 DKKL1 0.034 ACTN4 0.093 TMSB10 0.034 TCF4 0.093 SLC35B1 0.034 2410014A08RIK 0.093 2500003M10RIK 0.034 LOC676420 0.093 BCKDK 0.034 DNAJB3 0.092 1700008I05RIK 0.033 UBE2G1 0.092 GLRX2 0.033 38961 0.092 SERF1 0.033 MAFG 0.092 CSF2RA 0.033 TRUB2 0.092 ATP9B 0.033 DCTN2 0.092 2410042D21RIK 0.033 AACS 0.091 IGF2BP3 0.033 ALG9 0.091 THNSL1 0.032 RTN2 0.091 TUBA3A 0.032 TUSC2 0.090 SYF2 0.032 SERINC1 0.090 RDH14 0.032 RGS3 0.090 NSFL1C 0.032 SENP7 0.090 RPS3 0.032 MPPED1 0.090 OSBPL8 0.032 PFKM 0.089 PFN2 0.032 FRMD8 0.089 CRNKL1 0.032 ZBP1 0.089 FHIT 0.032 EDNRA 0.089 ZFP423 0.032 ACVR2B 0.089 EXOC8 0.032 ENDOGL1 0.089 1700113O17RIK 0.032 SLC10A3 0.089 AA673488 0.031 IDE 0.089 ASB8 0.031 L2HGDH 0.088 POLL 0.031 CASKIN1 0.088 SDCBP 0.031 CCDC65 0.088 PHF7 0.031 RAP2C 0.088 METTL4 0.031 NT5DC3 0.088 DTNB 0.030 HSPA9 0.088 SFRS2 0.030 TM2D3 0.088 POLR3A 0.030 TRAF3 0.088 VBP1 0.030 1200015N20RIK 0.088 TCHP 0.030 PTPRD 0.088 NDUFA12 0.030 MYLC2B 0.087 SH3GLB1 0.029 USP16 0.087 PMPCA 0.029 MDK 0.087 6330503K22RIK 0.029 CPSF1 0.087 MRPS6 0.029 4933417E01RIK 0.087 CCT3 0.029 2510009E07RIK 0.087 CBFB 0.028 TEX264 0.086 PLEKHM1 0.028 ISCU 0.086 ZDHHC13 0.028 ROCK1 0.086 OAZ2 0.028 LRIG1 0.086 PHPT1 0.028 GFOD2 0.086 MED11 0.028 SH3GL2 0.086 ADAM29 0.027 CCDC41 0.086 RBM12 0.027 BTBD1 0.086 PLEKHG4 0.027 NECAP1 0.086 CLMN 0.027 IER3IP1 0.086 GMCL1 0.026 ANP32A 0.085 NDFIP1 0.026 PRRX1 0.085 SPAG6 0.025 D10BWG1364E 0.085 SLK 0.025 PKIA 0.085 GMCL1 0.025 TMOD4 0.085 2600011E07RIK 0.025 RUNDC1 0.085 EG625054 0.025 PRMT5 0.085 CDC37L1 0.025 LOC100047837 0.085 1700025F22RIK 0.025 TMEM38B 0.084 SLC7A8 0.024 BC017158 0.084 EFHD1 0.024 MRPL23 0.084 ANKRD40 0.024 BAX 0.084 SEPHS2 0.024 DOK3 0.084 CCNL1 0.024 IL18BP 0.084 TSG101 0.023 SCARA5 0.084 MAPK3 0.023 TBC1D20 0.084 ZHX1 0.023 ZFAND6 0.083 GATAD1 0.023 CDCA7 0.083 1200003C05RIK 0.023 ATP1A1 0.083 D130059P03RIK 0.023 GM1527 0.083 DCUN1D4 0.023 4930562D19RIK 0.083 RNASEN 0.023 SLC9A8 0.082 LOC434172 0.022 STAT4 0.082 LOC100040573 0.022 PTPRV 0.082 GPR137 0.022 TST 0.082 PSMC3 0.022 LAMC3 0.082 LOC100048331 0.022 RPS12 0.082 SPARC 0.022 HDHD2 0.082 2810470D21RIK 0.022 PTPN11 0.081 TTC35 0.022 ADI1 0.081 TMEM184A 0.022 UPP2 0.081 CAP1 0.021 RNF44 0.081 GLRX5 0.021 UBAC1 0.081 CDC5L 0.021 MPHOSPH6 0.081 EPB4.1L5 0.021 ZFP238 0.081 EG545056 0.021 ELP3 0.080 CDK10 0.021 POLS 0.080 TNFRSF10B 0.021 METTL1 0.080 TMEM11 0.020 HS2ST1 0.080 CTDSPL2 0.020 CNBP2 0.080 GJA1 0.020 RWDD1 0.079 5830468K18RIK 0.020 A730017C20RIK 0.079 PSMB1 0.020 LMO4 0.079 ST6GALNAC2 0.019 TTC29 0.079 9530077C05RIK 0.019 LEPROTL1 0.079 RAB3IL1 0.019 ARD1 0.079 PSMD10 0.019 0610010K06RIK 0.079 SLC39A13 0.018 USP20 0.078 NSMAF 0.018 MYG1 0.078 SRP9 0.018 EPB4.1L3 0.078 CCDC101 0.018 MED7 0.077 ATP10A 0.018 PPP5C 0.077 TRIM33 0.018 HRBL 0.077 CLGN 0.018 ACAA2 0.077 1700034H14RIK 0.018 AMFR 0.077 PAFAH1B1 0.017 ST3GAL6 0.077 NUT 0.017 PABPN1 0.076 ZFP238 0.017 SLC17A8 0.076 4922502D21RIK 0.017 CCNA1 0.076 LOC100044170 0.017 LLGL1 0.076 PGAM2 0.017 TIMM8A2 0.076 ATP8B3 0.017 GYLTL1B 0.076 NECAB2 0.016 CYBA 0.076 RARS2 0.016 2810432L12RIK 0.075 D15WSU169E 0.016 LOC100047490 0.075 ACSL3 0.016 BIRC2 0.075 RTTN 0.016 LYPLA1 0.075 EIF2AK1 0.015 TWSG1 0.075 SUPT5H 0.015 UBE2N 0.075 TANK 0.015 NMNAT1 0.074 UBQLN1 0.015 ANKRD50 0.074 NIF3L1 0.015 MTCH1 0.074 PDK3 0.014 WBP2 0.074 LOC100047935 0.014 TMEM184A 0.074 RDH11 0.014 SLAMF9 0.073 NUBP2 0.014 ADD3 0.073 ADCY2 0.013 MTX1 0.073 RBM18 0.013 CCNB1 0.073 LOC100044177 0.013 UTP11L 0.072 DCI 0.013 ACAA2 0.072 RNF5 0.013 SMYD2 0.071 RBM12 0.012 PRPF3 0.071 ASZ1 0.012 TMCO4 0.071 CRY2 0.011 NVL 0.071 2810004N20RIK 0.011 RAD51L3 0.071 NRP1 0.011 WDR24 0.071 RPS24 0.011 ATP6V0D1 0.071 NCK1 0.011 TMSB4X 0.071 PTPRJ 0.011 CD47 0.070 CHERP 0.010 PWWP2B 0.070 IRGM 0.010 BHLHB9 0.070 COQ2 0.010 ECD 0.070 CYHR1 0.010 GADD45G 0.070 CDC27 0.009 BSCL2 0.070 DCP1A 0.009 TBX1 0.069 MCRS1 0.009 CAPN7 0.069 UBE2Q2 0.009 PAQR9 0.069 THSD7B 0.009 SLC26A8 0.069 LOC100048187 0.009 SSRP1 0.069 PARP2 0.009 COTL1 0.069 CD63 0.008 IFT140 0.069 1500032D16RIK 0.008 ZC3HC1 0.069 FBXO38 0.008 NELF 0.069 RALGPS1 0.007 EXOC4 0.069 EMD 0.007 LOC100047707 0.069 NDRG1 0.007 CUL3 0.068 TXNL4A 0.007 IMMT 0.068 1700019L03RIK 0.006 PHACTR1 0.067 ZC3HC1 0.006 DSTN 0.067 LOC100047214 0.006 AGPAT9 0.067 GLT25D1 0.006 PLA2G10 0.067 SOD1 0.006 D030070L09RIK 0.067 SOX5 0.005 PPM1A 0.067 1500001M20RIK 0.005 PFDN5 0.067 2210012G02RIK 0.005 CHMP1B 0.067 DAZAP1 0.004 KEAP1 0.067 COX8C 0.004 BC018465 0.067 MAPKAP1 0.003 LOC100044159 0.067 GATAD1 0.003 RSL1D1 0.066 SPSB3 0.003 2810046L04RIK 0.066 WIPI1 0.002 TWF2 0.066 RNF6 0.002 STARD7 0.066 ZDHHC12 0.001 MAN2C1 0.066 GM711 0.001 ADO 0.066 FIBCD1 0.066 PRMT5 0.065 CREM 0.065 PGS1 0.065 CAR2 0.065 SLC9A1 0.065 TMED4 0.065 TIAM1 0.065 IL11RA1 0.065 BCCIP 0.065 SH3BP2 0.065 ENTPD6 0.065 CDYL 0.065 LOC100045300 0.065 39142 0.065 NT5E 0.064 SDCBP2 0.064 4833426J09RIK 0.064 ATP6AP2 0.063 USH1C 0.063 VPS18 0.063 PALLD 0.063 4931440F15RIK 0.063 GRINA 0.063 SNX33 0.063 OXCT2A 0.063 PER1 0.062 RNF103 0.062 2900024O10RIK 0.062 ADAM4 0.062 NOB1 0.061 SCT 0.061 RAB40C 0.061 RARS2 0.061 PCBP3 0.061 LOC100046953 0.061 ABCF3 0.061 COBLL1 0.060 RABL4 0.060 LDOC1L 0.060 PTTG1IP 0.060 PPA1 0.060 METTL3 0.060 TCEAL1 0.060 DCPS 0.059 LRMP 0.059 DHPS 0.059 ACAA2 0.059 ACSL5 0.059 INPP5E 0.059 TMOD4 0.059 ANLN 0.059 YBX2 0.059 SLC6A6 0.058 EPHB1 0.058 OSBPL2 0.058 ANAPC5 0.058 TIPIN 0.058 TBK1 0.058 BC046331 0.058 NR2F6 0.058 MEOX1 0.058 KATNA1 0.058 PDLIM4 0.058 SULT1A1 0.058 COPS2 0.058 UBAC2 0.057 IAH1 0.057 NPEPL1 0.057 1500001M20RIK 0.057 HOXC6 0.057 ANKRD49 0.057 RAPSN 0.057 METT11D1 0.057 LOC100046081 0.057 APBB1 0.056 AI987944 0.056 OGFR 0.056 BZRAP1 0.056 ARL11 0.056 GLYCTK 0.056 CDAN1 0.056 MON1B 0.056 PPP1R12C 0.056 DDX54 0.056 AOC3 0.056 1110017D15RIK 0.056 MSRB3 0.056 UBP1 0.056 LZTS2 0.056 SCOC 0.056 XPNPEP1 0.055 QRICH2 0.055 STT3B 0.055 GJA1 0.055 ARL6IP2 0.055 PES1 0.055 PPM1B 0.055 LGALS8 0.055 RPL35 0.055 KCNIP2 0.055 RGS3 0.054 NDE1 0.054 EXTL3 0.054 TBC1D2B 0.054 RNF113A2 0.054 ARAF 0.054 PPP4R1 0.054 UCK2 0.054 HAP1 0.054 GSDMDC1 0.054 ATG2B 0.054 APPL2 0.054 LPIN1 0.054 CAPRIN1 0.054 PSMA2 0.054 NDUFA10 0.053 OVCA2 0.053 SMC1A 0.053 1700058C13RIK 0.053 TRAFD1 0.053 ACSS1 0.053 ZFP397 0.053 ZFP27 0.053 NCOA5 0.053 RPS26 0.053 IRF2 0.053 STXBP1 0.053 PRC1 0.053 ASAH3L 0.053 AGPAT9 0.052 2310008H09RIK 0.052 ABCA3 0.052 MYNN 0.052 4933427D14RIK 0.052 PHOSPHO1 0.052 SERBP1 0.052 NEK9 0.052 PLEKHF1 0.052 SLC30A7 0.052 OLFML2B 0.052 KRAS 0.052 ARAF 0.052 CASP7 0.052 SUDS3 0.051 ERCC5 0.051 MRPL2 0.051 ABCF2 0.051 LOC100047834 0.051 DBP 0.051 INTS7 0.051 INPPL1 0.051 PTPRE 0.051 ATP6V0D1 0.051 2010005J08RIK 0.051 GM347 0.050 3000004C01RIK 0.050 FBXO44 0.050 2310044H10RIK 0.050 TIMM8B 0.050 SFXN1 0.050 NXT2 0.050 COX6A1 0.050 KCNMB4 0.050 SERPINA1B 0.050 ARL4A 0.050 CCDC114 0.050 FBXW7 0.050 VKORC1 0.050 GDI1 0.050 ZFP532 0.050 1110007M04RIK 0.049 ZCCHC8 0.049 MGST2 0.049 BAG1 0.049 PCDH17 0.049 DENND3 0.049 ELMO1 0.049 HSD17B3 0.049 UACA 0.049 HERC2 0.049 LRPAP1 0.049 JMJD6 0.049 2610039C10RIK 0.049 TIMM8A1 0.049 LRIG3 0.049 SSBP2 0.049 NUP88 0.048 IFNA6 0.048 ITSN2 0.048 NUDCD3 0.048 DYNC2LI1 0.048 FEN1 0.048 CSTB 0.048 4933405O20RIK 0.048 ZFP574 0.048 LOC676420 0.048 2310008H09RIK 0.048 RAB35 0.047 SKP2 0.047 ACRV1 0.047 ARPP19 0.047 DACT2 0.047 ITFG3 0.047 ZBTB24 0.047 ARL6IP5 0.047 GATA2 0.047 GNS 0.047 COL4A1 0.047 RNF181 0.047 MCM10 0.047 ASB13 0.047 CYB5R2 0.047 PACS2 0.046 PPIE 0.046 HECA 0.046 PARP2 0.046 UTX 0.046 SF4 0.046 ZFP346 0.046 2700087H15RIK 0.046 BTBD14A 0.046 IL17RB 0.046 ELMO1 0.046 RASSF4 0.045 WDR63 0.045 COPS4 0.045 NR1H2 0.045 NRAS 0.045 PDLIM3 0.045 ILK 0.045 TMOD1 0.045 ITPR1 0.045 PI16 0.045 HMGN3 0.045 RPL37 0.045 MEF2C 0.045 NDUFA12 0.044 ARTN 0.044 TPI1 0.044 LITAF 0.044 CACNA1H 0.044 ELMO1 0.044 IRF2BP1 0.044 VANGL2 0.044 MVP 0.044 BHMT2 0.043 WARS 0.043 ASB8 0.043 ACSBG1 0.043 TLE4 0.043 YIPF3 0.043 MOCS2 0.043 LOC100047800 0.043 TMEM41B 0.043 EPN2 0.043 SCUBE3 0.043 TNIP1 0.043 LSM14A 0.043 PJA2 0.043 GSS 0.043 ZFP35 0.042 JAK3 0.042 PLEKHG3 0.042 RFX1 0.042 IL4I1 0.042 SLC39A11 0.042 AQP9 0.042 CEP63 0.042 STX12 0.042 KLK1B8 0.041 KANK1 0.041 CSNRP2 0.041 SRPRB 0.041 ZC3H12A 0.041 MCTS1 0.040 SIDT2 0.040 WIPI1 0.040 RIC3 0.040 LOC100047651 0.040 GNAO1 0.040 MIDN 0.040 MYST4 0.040 WDFY1 0.040 GGN 0.040 1700030K09RIK 0.040 3110043J09RIK 0.040 TAF7L 0.039 PGAM5 0.039 SLC35A4 0.039 MAT2B 0.039 TTLL6 0.039 SORT1 0.039 SHROOM3 0.039 ATP5C1 0.039 5730453I16RIK 0.039 PIP5KL1 0.039 ABCA15 0.039 RPL4 0.039 ST8SIA1 0.039 4933409N07RIK 0.039 MARCKS 0.038 C85627 0.038 CWF19L2 0.038 RBMS2 0.038 2610507B11RIK 0.038 BC027231 0.038 ENOX1 0.038 DUSP2 0.038 KIF1A 0.038 FNDC5 0.038 SOHLH1 0.038 2310035K24RIK 0.038 CDT1 0.037 ZFP36 0.037 CDC42EP5 0.037 EIF4A1 0.037 ZMYM3 0.037 UBE3B 0.037 ADARB1 0.037 OPRS1 0.037 SAP30L 0.037 ORAI1 0.037 SLC36A1 0.037 HEXDC 0.037 RIMS3 0.037 PDZD11 0.037 B230396O12RIK 0.037 A230050P20RIK 0.037 BRP17 0.037 PRM1 0.036 PRPF40B 0.036 LOC100048076 0.036 ETF1 0.036 NHSL1 0.036 AA407659 0.036 CHMP7 0.036 PTPRD 0.036 LOC100045403 0.036 REPIN1 0.036 MYO6 0.036 RAB11FIP5 0.036 CDC37L1 0.036 MFN1 0.036 HMGB2L1 0.036 ADK 0.036 IMPACT 0.035 CENPA 0.035 ZFP644 0.035 CRELD1 0.035 5730410E15RIK 0.035 HMGCS2 0.035 LDHC 0.035 ANGEL2 0.035 4933430I17RIK 0.035 LOC732482 0.035 RETSAT 0.035 LOC100045551 0.035 LOC218963 0.035 LOC100047238 0.034 ABI2 0.034 ACSS2 0.034 AFAP1 0.034 PPP2CB 0.034 MFSD10 0.034 MAN2B1 0.034 DCTN1 0.034 4933400C05RIK 0.034 INTS4 0.034 MYCBP2 0.034 GSPT1 0.034 2010011I20RIK 0.034 SEMA5B 0.033 HPS3 0.033 EIF2S2 0.033 ZDHHC14 0.033 NXF2 0.033 SENP3 0.033 SORT1 0.033 NRP1 0.033 HSPA1L 0.033 KCNK5 0.033 PSMG2 0.033 NT5C1B 0.033 HERPUD2 0.033 PQLC3 0.033 TXNRD3 0.033 BC037034 0.032 NKD2 0.032 TNFAIP1 0.032 AEBP2 0.032 TEX2 0.032 ZMYND12 0.032 6330505N24RIK 0.032 ZKSCAN14 0.032 PPP1R3C 0.032 COL18A1 0.032 AP1B1 0.032 ZBTB17 0.032 UBE2D2 0.032 MRPS5 0.032 SLFNL1 0.032 IARS2 0.032 ZBTB8 0.032 COMTD1 0.031 MBNL1 0.031 SDCBP 0.031 AMPD1 0.031 SIAH1B 0.031 SPATS1 0.031 NFE2L1 0.031 RNF185 0.031 KLHDC4 0.031 1700030J22RIK 0.031 GLT6D1 0.031 BLVRB 0.031 NAIF1 0.031 MEIS1 0.030 BCLAF1 0.030 STATIP1 0.030 D15WSU169E 0.030 RAB3GAP2 0.030 ATPIF1 0.030 AQP7 0.030 MBD4 0.029 KLHL26 0.029 PSCDBP 0.029 GPS1 0.029 RABIF 0.029 WBSCR22 0.029 SH3KBP1 0.029 2700060E02RIK 0.029 CXXC1 0.029 PFN2 0.029 AADACL1 0.029 RILPL1 0.029 TRAPPC1 0.029 SNX10 0.029 RBM5 0.029 EVI5L 0.029 MED22 0.029 TPI1 0.029 PPIF 0.029 RASGRF1 0.029 C78339 0.029 INPPL1 0.029 BAHD1 0.028 HSD17B10 0.028 LOC100047353 0.028 CASP9 0.028 UHRF2 0.028 1700120B06RIK 0.028 1110057K04RIK 0.028 ANAPC7 0.028 TSPAN3 0.028 POLDIP3 0.028 RHOT2 0.028 WDR31 0.028 CD8B1 0.028 MCM6 0.028 HPCAL4 0.028 DHH 0.027 HECTD2 0.027 EIF4H 0.027 6030443O07RIK 0.027 BC063749 0.027 IVD 0.027 SLC1A2 0.027 YPEL3 0.027 NT5E 0.027 DDAH1 0.027 ABHD2 0.027 ARMC9 0.027 RNF40 0.027 USP38 0.027 MTMR14 0.027 TMC4 0.027 PPM1F 0.027 PYROXD1 0.026 ETFA 0.026 MCEE 0.026 FERMT2 0.026 VAMP2 0.026 SYAP1 0.026 CCDC92 0.026 RNPEPL1 0.026 9130213B05RIK 0.026 BC025546 0.026 MYT1L 0.026 CCNF 0.026 ADAM17 0.026 RABGGTA 0.026 LAMP1 0.025 UBOX5 0.025 AUP1 0.025 ATP6V0A1 0.025 CYHR1 0.025 AP4B1 0.025 SDC3 0.025 RETSAT 0.025 TRIM45 0.025 NAPG 0.025 DNAJB13 0.025 AKR7A5 0.025 ATG5 0.025 TBC1D25 0.024 AHNAK 0.024 PIK3R6 0.024 CHRD 0.024 1600016N20RIK 0.024 FMNL3 0.024 1110007A13RIK 0.024 SLC44A2 0.024 APLP2 0.024 EHMT2 0.024 OSGEPL1 0.024 SLC9A10 0.024 GJA6 0.024 AURKA 0.024 CSNK1G2 0.023 UNC45A 0.023 SF3A3 0.023 MKLN1 0.023 4930563P21RIK 0.023 FDXR 0.023 ARTN 0.023 2610024G14RIK 0.023 MRPS22 0.023 KLHDC1 0.023 LOC100048301 0.023 FBXO2 0.023 NFYB 0.023 USP45 0.023 PLOD3 0.023 EYA4 0.023 ZFAND3 0.023 ALDH4A1 0.023 RPAP1 0.022 RAE1 0.022 SPOP 0.022 SRPK1 0.022 SBNO1 0.022 4930504O13RIK 0.022 MAN1B1 0.022 2410025L10RIK 0.022 E2F6 0.022 XPNPEP1 0.022 PIK3C3 0.022 PNRC2 0.022 PLRG1 0.021 TM2D3 0.021 GTF3C2 0.021 RBJ 0.021 ACYP2 0.021 ZFP251 0.021 USP39 0.021 TRIM32 0.021 PDE1B 0.021 AP2B1 0.021 PTGDS 0.021 TMEM57 0.021 D11WSU47E 0.021 DBNDD2 0.020 NCAPH2 0.020 TSSK6 0.020 PI4K2B 0.020 PAIP2B 0.020 EARS2 0.020 PRNPIP1 0.020 SNX2 0.020 ZFP12 0.020 MGEA6 0.020 FKBP4 0.019 RSL1D1 0.019 RAB11FIP3 0.019 MMEL1 0.019 4930583H14RIK 0.019 VASN 0.019 SMAD2 0.019 UBE2O 0.019 TDRD12 0.019 TMEM93 0.019 KLHL15 0.019 IFT140 0.019 HDGF 0.019 1700037H04RIK 0.019 BBS7 0.019 D5ERTD579E 0.019 ZYX 0.019 CDKN1A 0.019 9630028B13RIK 0.019 INPP4A 0.019 MNAT1 0.018 LOC100048372 0.018 LIAS 0.018 ARHGEF3 0.018 ACTN4 0.018 ELAVL2 0.018 2810453I06RIK 0.018 1700027J05RIK 0.018 FLVCR2 0.017 NPTX2 0.017 MREG 0.017 WWC2 0.017 B9D1 0.017 EIF3B 0.017 AFAP1 0.017 GEMIN4 0.017 AMPD1 0.016 HAGHL 0.016 2610003J06RIK 0.016 MORN3 0.016 IDH3A 0.016 PIGK 0.016 MYO9B 0.016 DPH3 0.016 FIZ1 0.016 MSH6 0.016 ILVBL 0.016 HYAL2 0.016 CDC23 0.016 COPA 0.015 SRM 0.015 CHTF18 0.015 DDX19B 0.015 MANBA 0.015 1110057K04RIK 0.015 NDUFA8 0.015 5930416I19RIK 0.015 DPP6 0.015 SUMF2 0.015 BEX4 0.015 VPS16 0.015 SLC4A2 0.014 NDUFV1 0.014 WDR33 0.014 PDHA1 0.014 LOC100047427 0.013 DECR2 0.013 FSCN1 0.013 RNASET2 0.013 EPHX1 0.013 4930572J05RIK 0.013 TMEM201 0.013 SDC2 0.012 FZR1 0.012 PPP2R5D 0.012 ATMIN 0.012 ABCF2 0.012 EG633640 0.012 AKT1S1 0.012 DNAHC2 0.012 P2RX2 0.012 PKD2L1 0.012 CNOT2 0.012 TDRD7 0.012 PRRG2 0.012 THNSL2 0.012 GGT7 0.012 PLEKHM1 0.011 GTF3A 0.011 HEATR1 0.011 SMOC1 0.011 PTP4A3 0.011 D9ERTD280E 0.011 HDGFRP2 0.011 FADD 0.011 4932411G14RIK 0.011 SLC29A3 0.011 CPSF1 0.011 EDEM2 0.011 GTPBP2 0.010 SF3A1 0.010 EHBP1L1 0.010 LOC100048384 0.010 PPME1 0.010 ANK 0.010 SPSB3 0.010 CDKL2 0.010 GTRGEO22 0.009 VPS45 0.009 POLRMT 0.009 WASF1 0.009 PORCN 0.009 GSTZ1 0.009 MICAL3 0.009 ZFP263 0.008 ICK 0.008 ZFP637 0.008 RNF185 0.008 COPS3 0.008 TLK1 0.008 CCIN 0.008 TBC1D1 0.008 SH3GL2 0.008 D030070L09RIK 0.007 IHPK1 0.007 RAB11B 0.007 IQCB1 0.006 DBN1 0.006 GM614 0.005 MAK 0.005 ADAM22 0.005 MTAP4 0.004 FXYD5 0.004 UCHL1 0.004 TUBA3A 0.004 MAG 0.004 AHCYL1 0.003 GM166 0.003 MRPS24 0.003 ANKRD53 0.003 C85492 0.003 PROX1 0.003 ERLIN2 0.003 HBP1 0.002 KNTC1 0.002 EIF3EIP 0.002 SERPINF1 0.002 TDRD6 0.002 BOLA1 0.002 ASPSCR1 0.001 sTable 5: All genes identified by ToxResponse Modeler that were dose-dependently regulated at all time points (1, 6 and 12 months) in testes of mice chronically exposed to acrylamide (67 genes). Hierarchical clustering analysis was also conducted on all genes at each time point to determine whether genes were up or downregulated with increasing acrylamide dose (clustering trend,↑ dose-dependently upregulated, ↓ dose-dependently downregulated, ~ no clear dose dependency observed). For some genes, more than one probe at each time point was detected by ToxResponse Modeler as dose-dependently regulated. The EC50 value, clustering trend, and gene function is included. 1 month 6 months 12 months Gene Function EC50 (µg/ml) Clustering Trend EC50 (µg/ml) Clustering Trend EC50 (µg/ml) Clustering Trend 1810055G02RIK 0.044 ↑ 0.178 ↑ 2.799 ↓ Unknown 2410025L10RIK 0.009 ↑ 0.022 ↑ 2.536 ↓ Fibrosin-like 1 2700087H15RIK 0.064 ↑ 0.046 ↑ 1.637 ↓ Mesoderm induction early response 6330503K22RIK 0.057 ↑ 6.148 ↑ 0.029 ~ Cell cycle ACTA2 4.413 ↓ 0.335 ↑ 5.287 ↑ smooth muscle actin protein ADRBK1 0.003 ↑ 0.102 ↑ 2.272 ↓ phosphorylates beta-adrenergic and G-protein-coupled receptors AKT3 0.040 ↑ 0.980 ↑ 0.197 ↑ ATP binding; kinase activity; ANKRD50 0.699 ↓ 0.074 ↑ 1.181 ↑ ankyrin repeat domain 50 AP2B1 0.037 ↑ 0.021 ↑ 0.059 ↑ clathrin binding, protein binding and protein transporter activity APLP2 0.138 ↑ 0.024 ↑ 4.192 ↓ amyloid beta (A4) precursor-like protein 2 ARAF 0.043 ↑ 0.054 ↑ 2.521 ↑ v-raf murine sarcoma 3611 viral oncogene homolog ARFGAP1 0.088 ↓ 5.700 ↓ 0.066 ↑ ADP-ribosylation factor GTPase activating protein 1 ATP1A1 0.051 ↑ 0.083 ↑ 4.095 ↓ ATPase, Na+/K+ transport BRDT 0.009 ↓ 1.394 ↓ 0.126 ↑ Regulation of transcription 0.125 ↓ 1.448 ↓ 0.028 ↑ CBFB PEBP2/CBF family transcription factor 0.008 CDC37L1 0.025 ↑ 0.036 ↑ 0.025 ↓ Cell cycle COX18 0.009 ↓ 2.715 ↑ 3.183 ↓ Respiratory chain complex IV assembly CXXC1 0.018 ~ 0.029 ↑ 1.391 ~ Histone H3-K4 methylation DBP 0.057 ↑ 0.051 ↑ 5.385 ↓ Transcription factor EIF3EIP 0.018 ↑ 0.002 ↑ 6.079 ↓ Translational initiation EPHX1 0.095 ↑ 0.013 ↑ 3.054 ↓ Toxin response, epoxide metabolism FARSB 4.338 ↓ 0.100 ↑ 0.966 ↓ tRNA synthesis FOXJ3 0.124 ↑ 0.419 ↑ 5.882 ↓ Embryo development FXYD5 4.747 ↑ 0.004 ↑ 4.127 ↓ Ion transport regulator GJA1 0.014 ↑ 0.055 ↑ 0.020 ↓ Gap junction, apoptosis GM136 5.318 ↑ 2.304 ↓ 0.972 ↑ Chromosome 6 open reading frame 163 0.104 ↓ 1.215 ↑ 2.805 ↓ GORASP2 Golgi reassembly stacking protein 2 0.657 0.030 ↓ 0.011 ~ 4.580 ↑ HEATR1 HEAT repeat containing 1 3.281 0.351 ↑ 0.016 ↑ 0.905 ↓ ILVBL ilvB (bacterial acetolactate synthase)-like 0.905 0.074 ↑ 3.756 ↑ 4.807 ↓ KIF1B Mitochondria and synaptic vesicle transport 4.807 0.015 ↑ 1.726 ↓ 0.035 ↑ KPNA3 Karyopherin (importin) alpha 3 0.035 LANCL1 5.295 ~ 3.856 ↑ 0.118 ↓ Biosythesis of antimicrobial peptides LOC100045617 6.382 ↓ 1.276 ↓ 0.085 ↑ Unknown LRRC49 0.039 ~ 3.972 ↓ 2.071 ↑ Leucine rich repeat containing 49 MAFG 0.033 ↑ 0.092 ↑ 0.092 ↑ Embryonic development MTCH1 0.065 ↑ 0.074 ↑ 2.997 ↓ Apoptosis MTMR14 0.055 ↓ 0.027 ↑ 1.106 ↓ Myotubularin related protein 14 NECAP1 3.906 ↓ 0.086 ↑ 2.396 ↓ NECAP endocytosis associated 1 PI4K2B 0.129 ↑ 0.020 ↑ 2.554 ↑ Phosphatidylinositol 4-kinase type 2 beta PJA2 0.006 ↓ 0.043 ↑ 0.044 ↑ Metal ion binding PLEKHM1 0.038 ↑ 0.011 ↑ 0.028 ↑ Pleckstrin homology domain containing family M PRIM2 0.500 ↑ 0.225 ↑ 2.404 ↑ DNA primase, p58 subunit 0.253 ↓ 3.650 ↓ 0.027 ↑ RBM12 Nucleic acid binding 0.012 RBMS2 0.054 ~ 0.038 ↑ 0.267 ~ Single stranded RNA interacting protein 2 RNF139 0.005 ↓ 1.834 ↓ 4.669 ↑ Ligase activity SNX2 0.021 ↓ 0.020 ↑ 0.205 ↓ sorting nexin 2 SNX7 0.029 ↑ 3.599 ↓ 0.095 ↑ Cell communication 7.065 ~ 0.934 ↑ 0.376 ↑ SPATA7 Spermatogenesis 0.883 0.297 0.002 ↑ 1.568 ↑ 4.414 ↓ SUMO1 Cytokine-mediated signalling 0.978 0.241 TBC1D25 0.036 ↑ 0.024 ↑ 3.406 ~ TBC1 domain family, member 25 TCTN3 0.002 ↓ 4.420 ↓ 0.968 ↓ Apoptosis TGM2 0.001 ↑ 0.102 ↑ 6.663 ↓ Apoptosis, PLC activation TIAM1 0.050 ↑ 0.065 ↑ 4.029 ↓ T cell lymphoma invasion and metastasis 1 TIMM8A1 0.003 ↑ 0.049 ↑ 6.682 ↑ Chaperone-mediated protein transport TMEM201 0.046 ↑ 0.013 ↑ 4.650 ↓ Transmembrane protein 201 TMSB10 6.279 ↓ 2.925 ↑ 0.034 ↑ Actin polymerisation TRAFD1 0.052 ↑ 0.053 ↑ 0.118 ↑ TRAF type zinc finger domain containing 1 TRAPPC1 0.004 ↑ 0.029 ↑ 0.058 ↓ Vesicle-mediated transport TRPC4AP 1.818 ↓ 0.807 ↑ 4.256 ↓ Transient receptor potential cation channel, subfamily C 1.489 ↓ 7.717 ↓ 0.081 ↑ TUBA3A Microtubule based functions 0.761 0.004 0.032 4.512 ↓ 2.108 ↓ 0.047 ↓ UBE2G1 Ubiquitination 0.092 VPS45 4.020 ↓ 0.009 ↑ 2.490 ↓ Protein trafficking WIZ 0.010 ↑ 0.110 ~ 3.443 ↓ Widely-interspaced zinc finger motifs ZC3H12A 4.815 ↑ 0.041 ↓ 4.408 ↓ Angiogenesis, apoptosis ZFP474 0.044 ↑ 1.022 ↓ 0.092 ↓ Zinc finger protein 474 ZMYM3 0.037 ↑ 0.037 ↑ 6.007 ↓ Zinc finger, MYM-type 3 ZYX 0.037 ↑ 0.019 ↑ 1.246 ↓ Zinc ion binding sTable 6: The genes that were identified as both differentially expressed and dose-dependently regulated at at least 1 time point. Of the 18 differentially expressed genes that were identified at all time points (see sTable 3), 10 were identified as dose-dependently regulated at one or more time points (highlighted in bold). EC50 values are included in the table.

1 month EC50 (µg/ml) 6 months EC50 (µg/ml) 12 months EC50 (µg/ml) 2610207I05RIK 0.098 0610009B22RIK 2.271 4930428D18RIK 0.140 2900092C05RIK 0.030 0610037L13RIK 3.183 4930432K21RIK 1.168 ALDH1A1 0.043 1110017D15RIK 0.056 AA536749 1.316 BAT5 0.079 1700008I05RIK 2.054 ACTB 0.142 BHMT 0.225 1700010M22RIK 4.829 AGT 0.059 BRF2 4.094 1700012B09RIK 3.710 ALDOA 0.109 CSTB 0.013 1700016D06RIK 6.031 AP2B1 0.073 FAIM3 0.080 1700017N19RIK 2.036 BAT2 1.656 GPR128 0.073 1700018F24RIK 0.832 CATSPER1 0.131 GSG1 0.042 1700021C14RIK 2.176 CLMN 0.097 IGF2 0.117 1700023B02RIK 1.425 COPS7A 0.025 LOC100040899 0.018 1700025E21RIK 0.709 DAK 0.075 LOC218963 0.039 1700030J22RIK 0.031 EG625054 1.751 LOC668837 0.055 1700034I23RIK 1.153 EHD1 3.054 MAP2K7 3.172 1700042B14RIK 3.410 EPB4.1L3 0.086 MRPL20 0.018 1700049K14RIK 2.051 EPHX1 1.725 NPM3 4.535 1700057G04RIK 1.127 FUS 3.532 PPP4R1 0.018 1700081D17RIK 3.520 FXYD6 4.234 PRCP 0.117 1700109H08RIK 1.120 GGN 3.902 S100A11 0.032 1810030N24RIK 1.767 GLT8D1 0.239 SERPINA5 0.015 2310005N03RIK 4.519 GSG1 7.228 SERPINF1 0.099 2410025L10RIK 0.022 GSTM1 0.658 SLC25A3 0.024 2610020H08RIK 1.521 HDLBP 0.086 TMEM176B 0.308 2610103J23RIK 3.960 IGF2 0.086 TMSB10 6.279 2700055A20RIK 0.952 LOC100045963 0.100 TRP53BP1 0.049 2810453I06RIK 0.018 MIPEP 2.839 TRPC4AP 1.818 3000004C01RIK 0.050 MORN2 1.904 TUBA3A 0.761 3110070M22RIK 4.498 MUPCDH 0.017 4732415M23RIK 4.219 MXRA8 1.626 4921528G01RIK 5.939 PAFAH1B1 0.081 4930555G01RIK 0.887 PAOX 0.037 4931440F15RIK 0.063 PHKG2 0.119 4932413O14RIK 0.737 PLD3 3.796 4933400A11RIK 4.962 PPP1R10 0.062 4933407N01RIK 0.629 PRLR 0.024 4933434I06RIK 0.958 SEPW1 0.376 4933436I01RIK 3.314 SLC7A8 0.021 6720467C03RIK 1.301 SPATA7 0.128 ACAA2 0.059 TNFRSF10B ACSBG1 0.043 TRIP12 ACTA2 0.336 ACTR10 1.099 ADAM24 2.014 ADAM39 3.160 ADIPOR1 0.101 AFAP1 0.017 AGPAT3 0.392 AI314180 3.427 AKAP4 4.149 AKR1A4 1.039 AKR7A5 0.025 ALS2 1.463 ANKRD7 0.395 AP2B1 0.021 APLP2 0.024 AQP7 0.030 ARHGAP15 4.108 ARHGAP29 2.327 ASNSD1 3.621 ATP5H 2.018 AURKA 0.024 BC018465 0.067 BC061237 5.951 BIRC2 0.075 BOP1 4.288 BZW1 1.475 CCDC101 2.353 CCDC53 1.363 CCT6B 4.651 CDC37L1 0.036 CDK2 5.325 CDKAL1 6.912 CDKL2 0.010 CHMP2A 4.095 CHMP4B 0.105 CLPX 2.772 COIL 1.400 COMMD1 4.075 COPS4 0.045 COPS5 2.366 COPZ2 4.057 CREB3L4 0.221 CREM 0.065 CRLS1 0.802 CSDA 2.262 CSL 1.254 CTNNA3 5.347 CTNNB1 3.212 CUTA 1.167 CYB5D1 6.770 CYB5R2 0.047 CYPT8 2.706 CYPT9 1.425 D2ERTD750E 1.659 DCUN1D5 2.735 DDX19B 0.015 DDX4 0.956 DERL2 0.472 DHH 0.027 DHX15 4.863 DMRTC2 5.126 DNAJB6 0.134 DNM1L 1.204 DUSP13 0.172 ECSIT 5.332 EDF1 4.647 EEF2 0.126 EFHC2 3.939 EG625054 1.468 EG668668 0.979 EIF4H 0.027 ELF2 0.102 ENPP5 2.292 FBXO44 0.050 FCF1 3.507 FEM1C 3.043 FHL4 3.032 FZR1 0.012 GLRX2 2.021 GM136 2.304 GM1527 0.083 GNL2 2.962 GOLGA2 2.805 GOSR2 0.903 GTPBP8 4.222 H1FNT 4.385 HAP1 0.054 HDGF 0.019 HEATR1 0.011 HERPUD2 0.033 HMGB4 1.399 HMGCS2 0.035 HSP90AB1 2.807 HSPA8 0.188 HYAL5 4.347 IFNGR2 5.266 IFT140 0.019 IL17D 4.333 INTS4 0.034 ISG20 2.676 ITGAE 2.497 ITGB5 0.927 KCNK2 1.623 KLF4 0.310 KLK1 3.322 KLK1B9 0.964 KRBA1 2.954 LAMP1 0.025 LANCL1 3.856 LOC100039532 1.682 LOC100044087 1.863 LOC100044170 0.646 LOC100046320 7.630 LOC100047052 2.071 LOC100047368 0.620 LOC100047651 0.040 LOC100047834 0.051 LOC100047935 1.619 LOC100048105 0.501 LOC100048480 0.843 LOC217341 4.038 LOC434960 1.500 LOC545732 2.465 LOC668837 2.332 LPGAT1 0.135 LRRC1 3.917 LRRC49 3.972 LRSAM1 4.429 LSM2 2.503 LYAR 1.264 MAN2C1 0.066 MAP2K7 0.420 MED1 2.531 MGC118210 0.792 MLF1 3.580 MORC2B 2.748 MRPL20 1.498 MRPL52 4.202 MRPS24 0.003 MTCH1 0.074 NACA 5.763 NAGLU 3.488 NAT5 3.777 NBR1 0.380 NDUFB5 1.222 NDUFS4 1.140 NME7 2.277 NOSIP 2.945 NUCB1 3.115 NUDC 5.676 NUDT21 1.143 NUP133 0.229 OBFC2A 6.682 ODF2 5.718 PABPC1 0.258 PEX5L 1.118 PFN2 0.029 PGAM2 2.016 PGRMC1 5.539 PGS1 0.065 PHC1 3.961 PKN2 3.941 PLCZ1 0.841 PMPCB 2.110 PNMA1 2.096 POLB 1.522 PPP2R5C 0.572 PPP4R1 0.054 PROX1 0.003 PRPF38A 0.765 PTPDC1 3.662 PTPRE 0.051 PTTG1IP 0.060 RAB11FIP5 0.036 RAB3IL1 6.220 RALGPS1 1.951 RETSAT 0.025 RFFL 1.699 RNF138 4.671 RNF139 1.834 RNF166 3.385 RNF6 6.205 RNPEPL1 0.026 RNU6 0.138 RPS21 2.515 SEL1L2 1.015 SERBP1 0.052 SERPINF1 0.002 SGCB 6.740 SH3GL2 0.008 SH3GL3 2.413 SKIV2L 2.498 SMC1B 5.634 SMPD4 6.195 SPARC 1.939 SPAST 1.107 SPATA16 6.108 SPATA6 0.841 SPERT 0.096 SS18 0.187 ST5 3.143 STX8 6.212 SUMO2 1.063 TAF9 0.136 TBC1D19 5.395 TBC1D20 0.084 TCP1 0.200 TEX101 4.343 TEX2 0.032 TGIF2LX 4.108 TIMM8B 0.050 TMEM184A 0.074 TMEM85 0.278 TNP2 2.862 TPI1 0.029 TRIM69 4.396 TRPC4AP 0.807 TSC22D1 0.941 TSG101 6.222 TTC23 2.803 TUBA3A 0.004 UBE2U 5.394 UBL3 3.800 USP1 5.530 VAPB 1.038 VASN 0.019 WDR31 0.028 WFDC10 1.020 YWHAG 6.458 ZBTB22 3.324 ZBTB9 2.666 ZFP238 0.081 ZFP654 5.165 ZKSCAN17 5.128 ZSWIM2 3.733 ZYX 0.019 sTable 7: The most over-represented functions of the dose-dependent genes (sTable 4) as identified by IPA analysis in testes of mice chronically exposed to acrylamide. Functions were identified on the basis of their z-score at one or more doses, indicating a significant over representation of genes involved in that function. Positive and negative z-scores indicate activation (red) or inhibition (green) respectively. Functions relevant to male reproduction and spermatogenesis were highlighted in bold Function Z-score 1 month 0.001µg/ml 0.01µg/ml 0.1µg/ml 1µg/ml 10 µg/ml proliferation of mesangial cells 2.213 migration of breast cancer cell lines 2.147 cell movement of lymphocytes 2.053 autophagy of cells 2.042 2.259 metabolism of protein 2.181 quantity of gonadal cells 2.17 internalization of protein 2.207 fusion of endosomes 2.137 neovascularization of cornea -2 cell movement of epithelial cells -2.151 hydrolysis of GTP -2.392 6 months contraction of cells 2.576 2.099 2.576 transport of molecule 2.448 3.258 3.762 3.474 replication of Influenza A virus 2.19 2.773 2.104 2.773 senescence of fibroblast cell lines 2.168 replication of virus 2.084 3.452 2.403 3.552 senescence of cells 2.045 transport of alpha-amino acid 2.012 2.012 2.012 2.012 Viral Infection 2.932 2.77 3.044 infection of cervical cancer cell lines 2.401 2.367 2.266 outgrowth of neurites 2.392 growth of neurites 2.35 infection of tumor cell lines 2.308 2.054 cell cycle progression of fibroblasts 2.213 2.213 2.213 organization of cytoplasm 2.19 2.193 2.016 growth of plasma membrane projections 2.178 cytolysis of red blood cells 2.176 2.176 blebbing 2.138 cell death of embryonic cell lines 2.131 2.326 sealing of plasma membrane 222 infection of cells 2.178 cytokinesis 2.198 2.198 proliferation of cells 2.115 remodeling of actin filaments 2 repair of DNA -2.865 development of reproductive system -2.55 gonadogenesis -2.55 development of genital organ -2.55 development of gonadal cells -2.236 gametogenesis -2.236 size of embryo -2.11 spermatogenesis -2 quantity of LDH -2 -2 -2 Movement Disorders -2.19 -2.19 -2.19 congenital anomaly of skeletal bone -2.2 -2.2 -2.2 DNA damage response of cells -2 malformation of brain -2 cell death of tumor cell lines -2.063 malformation -2.433 12 months cell death of B-lymphocyte derived cell lines 2.223 dilation of left ventricle 2.449 organismal death 2.564 2.537 ubiquitination 2.261 G1/S phase transition of tumor cell lines 2.078 2.078 proliferation of hepatocytes 2.153 gametogenesis -2.592 -2.592 spermatogenesis -2.401 -2.401 transport of molecule -2.167 -2.097 -2.238 -2.96 development of genital organ -3.094 -2.021 -2.557 biosynthesis of nucleoside triphosphate -2.098 -2.098 quantity of benign tumor -2.63 -2.63 -2.63 metabolism of nucleoside triphosphate -2.098 -2.098 biosynthesis of purine ribonucleotide -2.098 -2.098 gonadogenesis -3.094 -2.021 -2.557 protein kinase cascade -2.076 development of reproductive system -3.094 -2.879 -2.557 synthesis of ATP -2.098 -2.098 quantity of papilloma -2 -2.021 -2 hydrolysis of nucleotide -2.531 contraction of heart -2.021 I-kappaB kinase/NF-kappaB cascade -2.138 -2.759 -2.138 formation of muscle cells -2.148 -2.673 -2.148 infection of embryonic cell lines -2.013 protein kinase cascade -2.021 -2.478 -2.076 hydrolysis of nucleotide -2 -2.252 -2.252 proliferation of bone cancer cell lines -2.486 cell transformation -2.304 apoptosis of muscle cells -2.011 infection by lymphocytic choriomeningitis virus -2.2 phosphorylation of L-threonine -2 transport of protein -2.63 sTable 8: The most over-represented toxicological responses of the dose-dependent genes (sTable 4) as identified by IPA analysis in testes of mice chronically exposed to acrylamide. Toxicological responses were listed by time point as p-values and ratios were the same at each acrylamide dose. Toxicological Response 1 month p -value Ratio Recovery from Ischemic Acute Renal Failure (Rat) 6.78E-05 5 / 13 Liver Proliferation 6.46E-03 15 / 197 Hypoxia-Inducible Factor Signaling 1.16E-02 7 / 67 Renal Necrosis/Cell Death 4.46E-02 23 / 426 Xenobiotic Metabolism Signaling 4.71E-02 18 / 317 6 month Mitochondrial Dysfunction 1.87E-04 28 / 134 Anti-Apoptosis 2.37E-03 9 / 30 Swelling of Mitochondria 1.40E-02 5 / 15 p53 Signaling 1.98E-02 16 / 90 Glutathione Depletion - CYP Induction and Reactive Metabolites 2.78E-02 4 / 12 12 month Increases Renal Proliferation 1.54E-03 17 / 105 RAR Activation 3.07E-03 23 / 170 Mitochondrial Dysfunction 8.82E-03 18 / 134 Anti-Apoptosis 1.97E-02 6 / 30 NRF2-mediated Oxidative Stress Response 2.78E-02 24 / 217