DNA METHYLATION REGULATION in ACUTE KIDNEY INJURY By

DNA METHYLATION REGULATION IN ACUTE KIDNEY INJURY

Chunyuan Guo

Submitted to the Faculty of the Graduate School

of Augusta University in partial fulfillment

of the Requirements of the Degree of

Doctor of Philosophy

June

2017

ACKNOWLEDGEMENTS

This dissertation would not have become a reality without the support and the contributions of many people who gave me valuable assistance, guidance and encouragement in the preparation and completion of this project. Therefore, I would like to acknowledge all of them, including those not mentioned here by name.

First and foremost, I would like to express my utmost gratitude to my mentor Dr.

Zheng Dong, for his excellent guidance, infinite support and enthusiastic encouragement.

During the five-year training as a graduate student under Dr. Dong’s mentorship, I deeply appreciate his motivation, wisdom, and enthusiasm on research. I am heartily thankful to him for giving me the opportunity to join his amazing lab with an excellent atmosphere for doing research. It is my pleasure to thank all the current and former Dr. Dong’s lab members, who are always willing to help me and give their best suggestions, especially

Qingqing Wei, Man Jiang Livingston, Guie Dong, Shixuan Wang, Zhengwei Ma,

Navjotsingh Pabla, and Jianping Peng. Special thanks to Xiao Xiao for the initial work of this study. The experience gained here will benefit me for my entire scientific career.

Besides my mentor, I would also like to express my deepest gratitude to the rest of my advisory committee, Dr. Jiankang Chen, Dr. Adviye Ergul, Dr. Graydon Gonsalvez and Dr. Huidong Shi for all their efforts in guiding me through this process and giving me

precious advice on my research project. Particularly, I would thank Dr. Shi for providing me with the insights and knowledge in the field of DNA methylation. My sincere thanks also go to Dr. Sylvia B. Smith, Chair of the Department of Cellular Biology and Anatomy,

Dr. Mark Hamrick (current), and Dr. Ellen K. LeMosy (former), Program Director of the

Department of Cellular Biology and Anatomy for their support, assistance, and encouragement during my graduate study.

The support of the Graduate School is greatly appreciated and special thanks to Dr.

Mitchell A. Watsky, Dean of School of Graduate Studies, and Patricia L. Cameron, Vice

Dean of School of Graduate Studies.

I would like to thank the members of the Electron Microscopy and Histology Core, especially Donna Kumiski who helped me with all kidney tissue processing and sectioning.

My extreme appreciation goes to the administrative staff of the Department of

Cellular Biology and Anatomy: Sharon Lever, Cotton Mitchell, Heide Andrews, Donna

Tuten, Diane Riffe and Nan Eaton. Their constant support makes the graduate life much easier. Not forgetting, thank you all my friends in Augusta for all the emotional support, caring and fun. I will never forget the time I spent in Augusta with all of you.

Last but not the least, I owe my sincere and earnest thankfulness to my husband and my family for their devoted support, endless love, and encouragement during my entire

Ph.D. study.

ABSTRACT

CHUNYUAN GUO DNA Methylation Regulation in Acute Kidney Injury (Under the direction of ZHENG DONG)

DNA methylation is a critical epigenetic mechanism, which is heritable during cell division, but does not involve the change of DNA sequence. It plays an essential role in regulating gene transcription in physiological and disease conditions. However, little is known about DNA methylation in renal diseases, especially in acute kidney injury (AKI).

In this study, the role of DNA methylation in AKI was determined in both cell culture and mouse models. In cell culture, 5-aza-2’-deoxycytidine (5-aza), a pharmacological DNA methylation inhibitor, was used to inhibit DNA methylation. Interestingly, 5-aza increased both cisplatin- and hypoxia-induced apoptosis. These results suggest pharmacologic blockade of DNA methylation by 5-aza sensitizes renal tubular cells to apoptosis, supporting a cytoprotective role of DNA methylation in AKI. To determine the role of

DNA methylation in vivo, we first successfully established conditional knockout mice that were deficient in DNMT1, DNMT3a or both exclusively in proximal tubules. In cisplatin- induced AKI, consistent with the effects of 5-aza in the cell culture, ablation of DNMT1 from proximal tubules exacerbated cisplatin-induced AKI in mice, and primary proximal tubular cells from PT-DNMT1-KO mice were more sensitive to cisplatin-induced apoptosis than wild-type cells. In sharp contrast, PT-DNMT1/3a-DK mice attenuated cisplatin-induced AKI, and primary proximal tubular cells from PT-DNMT1/3a-DK mice were more resistant to cisplatin-induced apoptosis. However, PT-DNMT3a-KO mice and

PT-DNMT3a-WT mice showed similar AKI following cisplatin treatment. These results suggest different DNMTs play different roles in cisplatin-induced AKI. In ischemic AKI, none of the conditional knockout models showed differences in response to ischemia- reperfusion injury. Nevertheless, although ablation of both DNMT1 and DNMT3a in proximal tubular cells did not affect ischemia-reperfusion injury, it, indeed, suppressed renal fibroblast activation and ameliorated renal fibrosis. Furthermore, we found that Irf8 was regulated by DNA methylation during cisplatin treatment and knockdown of Irf8 in

RPTC cells inhibited cisplatin-induced apoptosis, supporting a pro-death role of Irf8 in renal tubular cells. In ischemic AKI, although Bcl6 is hypermethylated and repressed in mice, overexpression of Bcl6 in RPTC cells had no impact on hypoxia-induced apoptosis.

Collectively, these results suggest an important role of DNA methylation in AKI by regulating specific genes expression.

KEY WORDS: DNA Methylation, DNA methyltransferases, Acute kidney injury,

Cisplatin, Ischemic AKI, Hypoxia, Irf8, Bcl6.

Table of Contents

Chapter Page

I. INTRODUCTION ...... 1

A. Statement of the Problem ...... 1

B. Review of Related Literature ...... 5

II. MATERIALS AND METHODS ...... 25

III. RESULTS ...... 39

Specific Aim 1 ...... 39

Specific Aim 2 ...... 48

Specific Aim 3 ...... 64

Specific Aim 4 ...... 77

IV. DISCUSSION ...... 88

V. SUMMARY ...... 100

VI. REFERENCES ...... 102

List of Tables

Table 1: Staging of AKI by the KDIGO classification system ...... 9

Table 2: Primers used for genotyping ...... 29

Table 3: Primers used for real-time PCR ...... 34

Table 4: Genes with DMR at 5’ end or 5’UTR in cisplatin-induced AKI ...... 50

Table 5: Functions of genes with DMR at 5’ end or 5’UTR in cisplatin-induced AKI. ...51

Table 6: Genes with DMR at 5’ end or 5’UTR in ischemic AKI ...... 66

Table 7: Functions of genes with DMR at 5’ end or 5’UTR in ischemic AKI ...... 66

List of Figures

Figure 1: The structure of nephron ...... 6

Figure 2: Illustration of human kidney and nephron heterogeneity ...... 7

Figure 3: Classification of major causes of AKI ...... 10

Figure 4: Epigenetic modifications and their roles in gene expression regulation ...... 13

Figure 5: Mechanisms of DNA methylation and demethylation...... 19

Figure 6: Three major strategies for distinguishing methylated cytosine to unmethylated cytosine ...... 20

Figure 7: Breeding protocol to generate PT-DNMT1-KO mice ...... 26

Figure 8: Breeding protocol to generate PT-DNMT3a-KO mice ...... 27

Figure 9: Breeding protocol to generate PT-DNMT1/3a-DK mice ...... 28

Figure 10: Characterization of PT-DNMT1-KO mouse model ...... 41

Figure 11: Renal function and histology in PT-DNMT1-KO mice ...... 43

Figure 12: Characterization of PT-DNMT3a-KO mice ...... 44

Figure 13: Renal function and histology in PT-DNMT3a-KO mice ...... 46

Figure 14: Characterization of PT-DNMT1/3a-DK mouse model ...... 47

Figure 15: Genome-wide analysis of DNA methylation changes in cisplatin-induced

AKI ...... 49

Figure 16: 5-aza increases cisplatin-induced apoptosis in RPTC cells...... 52

Figure 17: Cisplatin-induced AKI is aggravated in PT-DNMT1-KO mice ...... 54

Figure 18: PT-DNMT1-KO mice induces more renal apoptosis after cisplatin treatment 55

Figure 19: DNMT1-KO primary proximal tubular cells are more sensitive to cisplatin- induced apoptosis ...... 57

Figure 20: PT-DNMT3a-KO mice show similar AKI following cisplatin treatment as PT-

DNMT3a-WT mice ...... 59

Figure 21: Cisplatin-induced AKI is attenuated in PT-DNMT1/3a-DK mice ...... 61

Figure 22: Cisplatin-induced renal apoptosis is suppressed in PT-DNMT1/3a-DK mice .62

Figure 23: DNMT1/3a-DK primary proximal tubular cells are more resistant to cisplatin- induced apoptosis ...... 63

Figure 24: Genome-wide analysis of DNA methylation changes in ischemic AKI ...... 65

Figure 25: 5-aza enhances hypoxia-induced apoptosis in RPTC cells ...... 68

Figure 26: DNMT1 deficiency in proximal tubules does not affect ischemic AKI ...... 70

Figure 27: Ablation both DNMT1 and DNMT3a in proximal tubular cells has no significant effects on renal ischemia-reperfusion-induced injury ...... 72

Figure 28: The expressions of fibrotic markers induced by unilateral renal ischemia- reperfusion are suppressed in PT-DNMT1/3a-DK mice ...... 74

Figure 29: PT-DNMT1/3a-DK mice exhibit much lower levels of collagen deposition compared to PT-DNMT1/3a-WT mice after UI30R2w...... 76

Figure 30: Real-time PCR analysis of selected hypermethylated and hypomethylated genes in cisplatin-induced AKI ...... 78

Figure 31: Real-time PCR analysis of selected hypermethylated and hypomethylated genes in ischemic AKI ...... 79

Figure 32: Irf8 is hypomethylated and induced in cisplatin-induced AKI ...... 81

Figure 33: Inhibition of DNA methylation by 5-aza increases Irf8 expression ...... 82

Figure 34: Silencing Irf8 suppresses cisplatin-induced apoptosis in RPTC cells ...... 83

Figure 35: Depletion of DNMT1 in mice does not affect Irf8 expression...... 85

Figure 36: Bcl6 is hypermethylated and suppressed in ischemic AKI ...... 86

Figure 37: Overexpression of Bcl6 does not affect hypoxia-induced apoptosis in RPTC cells ...... 87

I. INTRODUCTION

A. Statement of the Problem

Acute kidney injury (AKI) is a devastating kidney disease associated with a high mortality rate (>50%). Renal ischemia-reperfusion and cisplatin-induced nephrotoxicity are two major causes of AKI clinically.1, 2 Although the pathogenesis of AKI has been studied for decades, effective therapies for this disease are still lacking.

DNA methylation is a major form of epigenetic modification, which involves a direct chemical modification of cytosine by DNA methyltransferases (DNMTs, isforms1,

3a, and 3b). The fundamental role of DNA methylation in cell biology is transcriptional regulation. Generally, hypermethylation in the promoter region of a gene leads to heterochromatin with highly packed DNA, decreased accessibility of transcription factors, and loss of gene expression.3, 4 In contrast, DNA hypomethylation may correlate with the activation of gene transcription or lead to genomic instability.5 As a result, DNA methylation plays crucial roles in physiological conditions. Also, aberrant DNA methylation has been implicated in various disease conditions, such as cardiovascular diseases, neurological diseases, and cancers.6-13 However, very limited is known about the role of DNA methylation in renal diseases, especially AKI.

Recently, a few studies have implicated that DNA methylation was involved in

ischemic AKI. In 2006, Pratt et al.14 first reported DNA methylation changes in AKI. They found demethylation of a specific cytosine in renal C3 promoter in rat kidney treated with ischemia and subsequent reperfusion. Moreover, using quantitative methylation-specific

PCR, Mehta et al.15 found hypermethylation of calcitonin gene (CALCA) promoter in urine from kidney transplant patients compared with that from healthy controls, which could be used as a potential biomarker of ischemic AKI during transplantation. More recently,

Huang et al.16 using a mouse model of ischemic AKI showed that global 5- hydroxymethylcytosine (5hmC) was decreased, while global 5-cytosine (5mC) did not change. Obviously, these few studies are mainly descriptive. The role of DNA methylation in the pathogenesis of AKI remains elusive and the regulation of specific genes by DNA methylation in AKI is largely unknown.

Our preliminary data show that DNMT1 and DNMT3a are significantly increased during AKI induced by cisplatin nephrotoxicity and renal ischemia-reperfusion.

Importantly, our genome-wide DNA methylation analysis of kidney tissues has further unveiled aberrant DNA methylation changes in AKI. We have also identified a set of genes with differentially DNA methylated regions (DMRs) at their promoter regions in AKI, which may be regulated by DNA methylation. Based on these observations, the goal of my graduate research is to investigate the role of DNA methylation in the pathogenesis of AKI and the regulation of specific gene expression by DNA methylation.

To achieve this goal, the following specific aims were proposed:

Specific Aim 1: To establish proximal tubule-specific DNMTs knockout (KO) mouse models (PT-DNMT1 and PT-DNMT3a single KO mouse models, and PT-

DNMT1/3a double knockout (DK) mouse model).

Sub-aim 1a: To establish and characterize PT-DNMT1-KO mouse model.

Sub-aim 1b: To establish and characterize PT-DNMT3a-KO mouse model.

Sub-aim 1c: To establish and characterize PT-DNMT1/3a-DK mouse model.

Specific Aim 2: To determine the role of DNA methylation in cisplatin-induced

AKI using DNA methylation inhibitor and PT-DNMTs KO mouse models.

Sub-aim 2a: To determine the role of DNA methylation in cisplatin-induced AKI in vitro using DNA methylation inhibitor 5-aza-2’-deoxycytidine (5-aza).

Sub-aim 2b: To determine the role of DNA methylation in cisplatin-induced AKI using PT-DNMT1-KO mice.

Sub-aim 2c: To determine the role of DNA methylation in cisplatin-induced AKI using PT-DNMT3a-KO mice.

Sub-aim 2d: To determine the role of DNA methylation in cisplatin-induced AKI using PT-DNMT1/3a-DK mice.

Specific Aim 3: To determine the role of DNA methylation in ischemic AKI using DNA methylation inhibitor and PT-DNMTs KO mouse models.

Sub-aim 3a: To determine the role of DNA methylation in ischemic AKI in vitro using DNA methylation inhibitor 5-aza.

Sub-aim 3b: To determine the role of DNA methylation in ischemic AKI using PT-

DNMT1-KO mice.

Sub-aim 3c: To determine the role of DNA methylation in ischemic AKI using PT-

DNMT1/3a-DK mice.

Specific Aim 4: To determine the role of specific genes (Irf8 and Bcl6) with

DNA methylation changes in AKI.

Sub-aim 4a: To determine the role of Irf8 with DNA methylation changes in cisplatin-induced AKI.

Sub-aim 4b: To determine the role of Bcl6 with DNA methylation changes in ischemic AKI.

B. Review of Related Literature

1. The Kidney

The kidney is a bean-shaped organ that is highly differentiated and complexed. In humans, each kidney consists of about 1 million filtering units called nephrons. As shown in Figure 1, each nephron contains two main parts: a renal corpuscle and a system of renal tubules. The glomerulus is a bundled network of capillaries, which is encased in

Bowman’s capsule. Together with Bowman’s capsule, they form renal corpuscle, which filter the blood, letting the fluid and small molecules (e.g. amino acids, ions, glucose, and waste products et al.) pass through but not cells and large molecules, mostly proteins. The filtered fluid then flows into the renal tubules. The renal tubules are divided into 4 segments. 1) Proximal tubule. The proximal tubule, draining Bowman’s capsule, is the first part of the renal tubules. It makes up the majority of the kidney. The unique characteristic of proximal tubule is brush border, which provides surface area for reabsorption. The primary role of the proximal tubule is reabsorption. It reabsorbs 60-80% of the glomerular filtrate, including about two-third of the filtered salt and water, most filtered bicarbonate as well as most essential nutrients such as amino acids and glucose. 2)

The loop of Henle. The loop of Henle has a long, U-shape like structure, including descending thin limb, ascending thin limb, and ascending thick limb. The principal function of the loop of Henle is to further reabsorb water and sodium chloride, making urine more concentrated. 3) Distal convoluted tubule. The distal convoluted tubule is a short nephron segment, lying between the loop of Henle and collecting duct. It is responsible for potassium secretion, sodium chloride reabsorption, and magnesium and calcium handling.17 Also, it regulates pH by secreting hydrogen ions and ammonium ions. 4)

Collecting duct. Collecting duct is the final part of the renal tubules. It reabsorbs solutes and water, modulating the final composition of urine. Moreover, it facilitates the transportation of the urine to the renal pelvis and ureters.18, 19

Figure 1: The structure of nephron.

The functional unit of the kidney is nephron. Each nephron contains a corpuscle (glomerulus +

Bowman’s capsule) and renal tubules (① Proximal tubule, ② Loop of Henle, ③Distal tubule and

④Collecting duct). The figure is adapted from https://s-media-cache-

ak0.pinimg.com/736x/b0/71/0a/b0710aaaec469f7f0cc5e95fcfd1c16b--high-school-biology-your-

teacher.jpg.

Anatomically, the renal parenchyma can be divided into two major structures: renal cortex (the outer region) and renal medulla (the inner region) (Figure 2). All of the renal corpuscles are located in the renal cortex, which also contains the S1 and S2 segments of

the proximal tubules and distal tubules, with the S3 segments of the proximal tubule localizing in the outer stripe of outer medulla. There are two types of nephrons: one type with the short loop of Henle almost entirely within renal cortex called cortical nephron, and another with the long loop of Henle called juxtamedullary nephron, which extends deeply into the inner medulla. The majority of nephrons are cortical nephrons.19

Figure 2: Illustration of human kidney and nephron heterogeneity.

The left part shows the renal parenchyma is divided into cortex and medulla. The right part shows two

different types of nephrons: cortical nephron and juxtamedullary nephron. The figure is adapted from:

http://droualb.faculty.mjc.edu/Course%20Materials/Elementary%20Anatomy%20and%20Physiology%

2050/Lecture%20outlines/urinary_system.htm.

The kidney is a multifunctional organ. First, it is the major organ to maintain homeostasis, regulating water and electrolyte balance, body fluid osmolality, acid-base balance, and blood pressure. It also excretes wastes (such as toxins, urea, ammonium, and creatinine) by the formation of urine. Furthermore, it secretes, metabolizes and excretes

hormones. For example, it produces erythropoietin that stimulates the production of red blood cells. It also regulates vitamin D activation by producing calcitriol from inactive vitamin D molecules. During fast, the kidney also participates in gluconeogenesis.19

2. Acute Kidney Injury

Acute kidney injury (AKI), formerly referred as “acute renal failure”, is a devastating kidney disease with an abrupt reduction of kidney function within a few hours to days that leads to decreased glomerular filtration, accumulation of nitrogenous wastes, and imbalance of water, electrolytes, and acid-base. According to the KDIGO (Kidney

Disease Improving Global Outcome) Clinical Practice Guideline, AKI is defined as any of the following: increase in serum creatinine by ≥0.3 mg/dl (≥26.5 mmol/l) within 48 hours, or increase in serum creatinine to ≥1.5 times baseline, which is known or presumed to have occurred within the prior 7 days, or urine volume <0.5 ml/kg/h for 6 h.20 Based on the changes in serum creatinine and urine output, AKI is classified in three stages by KDIGO criteria (Table 1).20 The incidence of AKI is rapidly increasing, especially in hospitalized patients. AKI is potentially reversible if identified and appropriately managed in the early stages. Unfortunately, AKI lacks specific signs and symptoms. And the current diagnosis methods are largely based on the traditional biomarkers of blood urine nitrogen (BUN) and serum creatinine, which may not rise until half of the kidney function is lost, causing AKI is often recognized late. Patients who survive AKI also have an extremely high risk for developing chronic kidney disease (CKD) and end-stage renal disease (ESRD). AKI has emerged as a major public health concern associated with a high mortality rate (about 1.7 million deaths per year), increased length of hospital stay and cost.21

Table 1: Staging of AKI by the KDIGO classification system. Stage Serum Creatinine Urine output Others 1/Risk ≥1.5 but <2.0 time <0.5 ml/kg/h for 6-12 baseline hours OR ≥0.3 mg/dl (≥ 26.5 µmol/l) 2/Injury ≥2.0 but <3.0 times <0.5 ml/kg/h for ≥12 baseline hours 3/Failure ≥3.0 times baseline <0.3 ml/kg/h for ≥24 Initiation of renal hours replacement therapy OR ≥ 4.0 mg/dl OR Anuria for ≥12 OR in patients < 18 years, (≥353.6 µmol/l) hours decrease in eGFR to <35ml/min per 1.73 m2

There are many possible causes of AKI, which are commonly divided into three categories: prerenal, intrinsic renal, and postrenal (Figure 3).22-24 1) Prerenal AKI, the most common form of AKI, is caused by a sudden reduction of blood flow to the kidneys and a decrease in the glomerular filtration rate (GFR), resulting from severe injury, disease or medication. In prerenal AKI, the kidney itself does not have any problem. 2) Intrinsic renal AKI is a direct damage to the kidney itself, including one or more kidney structures

(i.e., kidney tubules, glomeruli, interstitium or vasculature). The most common causes of intrinsic renal AKI are renal ischemia-reperfusion, nephrotoxins, and sepsis. 3) Postrenal

AKI typically results from the disease states downstream of the kidney, which leads to obstruction of urine flow. This may be associated with obstructed urinary catheter, kidney stones, enlarged prostate, and bladder tumor, or injury.

○2 Intrinsic renal: ○1 Prerenal: X A direct damage to the kidney A sudden reduction of blood flow to the X kidneys X ○3 Postrenal: Resulted from the disease states downstream of the kidney

Figure 3: Classification of major causes of AKI.

The causes of AKI can be divided into three categories: ○1 prerenal, ○2 intrinsic renal, ○3 postrenal.

The picture is modified from:

http://www.renalnetwork.on.ca/info_for_patients/kidney_disease/#.WYDM2ITyuM8.

The key pathological characteristic of AKI is cell injury and death in kidney tubules, particularly the proximal tubules. After injury, adaptive repair response is activated. Apoptotic cells are cleared by macrophages. Some of the survived tubular cells begin dedifferentiation, migration, and proliferation to replace the lost cells, followed by redifferentiation to restore normal epithelial structure and function. However, if the injury is severe or sustained, it often results in maladaptive and incomplete repair, leading to renal fibrosis, ultimately CKD.25 Despite the progress made in the basic research and clinical care over the past decades, the underlying pathophysiology of AKI is not fully understood and other than dialysis, no effective pharmacological treatments for AKI are available.

Cisplatin-induced nephrotoxicity

Cisplatin-induced nephrotoxicity is one of the most common causes of AKI.

Cisplatin (cis-diamminedichloroplatinum II - CDDP), the most widely used and highly effective chemotherapy drug, is used for the treatment of a wide variety of cancer types, such as testicular, ovarian, head and neck, bladder, small and non-small cell lung, cervical cancers, and many other types of cancers like sarcomas and lymphomas.26, 27 Once inside the cell, cisplatin exerts its anticancer effects by creating intra-strand and inter-strand DNA cross-links, which blocks DNA replication and induces DNA damage, ultimately triggering cell growth arrest and apoptosis in tumors.26 However, cisplatin is also notorious for its side-effects in normal tissues and organs, especially in the kidney, which limits its use and therapeutic efficacy.28-32 The kidney is the predominant organ to excrete cisplatin. During the excretion process, cisplatin penetrates and accumulates in kidney tubular cells, particularly, in the renal proximal tubular cells of S3 segment (the straight segment of proximal tubule). The concentration can reach about five times greater than in the blood, causing proximal tubular cell injury and death, which is a key determinant of AKI.33

Following a single dose of cisplatin treatment, 25%-35% of the patients experience renal function deterioration.29

Recently, the pathogenesis of cisplatin-induced AKI has been intensively studied, and multiple signaling pathways have been implicated, such as apoptosis and necrosis of renal tubular cells, oxidative stress, ER stress, autophagy, mitochondrial dysfunction and inflammation.28-36 Despite these studies, the molecular basis of cisplatin nephrotoxicity remains unclear and no effective therapies are available for renoprotection during chemotherapy.

Ischemic AKI

Renal ischemia-reperfusion is another main cause of AKI, which occurs in a number of clinical conditions, such as dehydration, renal artery occlusion during major surgeries and kidney transplantation.37-39 It is characterized by a sudden reduction of blood flow to kidneys and followed by a restoration of blood supply and re-oxygenation. Renal ischemia (lack of blood supply) leads to deprivation of nutrients and oxygen, exerting energetic stress on kidney cells, particularly tubular epithelial cells that require high energy under normal conditions. Deprivation of oxygen in ischemic phase induces a metabolism switch from aerobic to anaerobic glycolysis, resulting in intracellular ATP depletion, acidosis and mitochondria dysfunction, ultimately causing kidney cell injury and death.

Subsequent reperfusion is not beneficial for cell survival but deleterious. The sudden restoration of blood supply rapidly returns oxygen level and pH to normal, causing oxidative stress, further mitochondria dysfunction and evoking inflammatory responses and cell death.39-41 The pathogenesis of ischemic AKI is very complicated, involving a complex interplay among tubular cells, vasculature, and inflammation.25, 37, 38, 42 In spite of the progress made in understanding the cellular and molecular basis of AKI, effective pharmacological strategies are still lacking.

3. Epigenetics

In 1942, the concept of epigenetics was first introduced by Waddington.43 What is epigenetics? Epi means above or over, and epigenetics means on top of genetics literally.

In biology, epigenetics refers to heritable changes in gene expression that does not involve the change of its primary nucleotide sequence.44 It can cause a change in phenotype but without changing its genotype. Epigenetics has been considered to be stable and heritable

during cell divisions; however, it is potentially reversible and can be affected by environment/lifestyle, age and disease state. Well-known epigenetic mechanisms include histone modifications, RNA interference, and DNA methylation (Figure 4).

Figure 4: Epigenetic modifications and their roles in gene expression regulation.

(1) DNA methylation. (2) Histone modifications. (3) MicroRNA interference. The figure is adapted

from Mas VR et al., 2016.

Histone modifications

Histones are highly conserved, basic, positively charged proteins that associate with

DNA (negatively charged) through electrostatic interaction and package it into highly condensed and ordered chromatin structure units called nucleosomes.45 There are two types of histones: core histones (H2A, H2B, H3 and H4) and linker histones (H1 and H5). Each

nucleosome consists of ~ 147 bp of DNA wrapped about 1.7 turns around a core histone octamer, which includes two copies of each core histone protein (two H2A-H2B dimers and one H3-H4 tetramer).46, 47 Histone modifications are covalent post-translational modifications of core histones, which occur predominantly, but not exclusively, on the N- terminal tails of histones. Histone modifications include acetylation, methylation, phosphorylation, ubiquitination, SUMOylation, citrullination, biotinylation, or ADP- ribosylation, which are considered to have a prominent role in regulating chromatin structure, status, and gene expression.46, 47 Histone modifications may change the structure of chromatin or serve as docking sites for transcriptional regulators that positively or negatively regulate gene expression. Among these modifications, histone acetylation, methylation, and phosphorylation make up the majority of histone modifications.

RNA interference (RNAi)

In addition to histone modifications, non-coding RNAs, such as long noncoding

RNAs and microRNAs, are also considered important epigenetic modulators. MicroRNAs are endogenous, small non-coding RNAs, usually containing 21-25 nucleotides, which function as essential post-transcriptional regulators of gene expression. The mature microRNAs predominantly bind to the 3’-untranslated region (UTR) of the target gene mRNAs, resulting in gene repression by either inducing mRNA degradation or, more commonly by blocking protein translation in mammals.48, 49 Approximately 1900 microRNAs have been identified in human listed in miRBase version 21

(http://www.mirbase.org) that regulate almost half of the human genome. By regulating gene expression, microRNAs have been shown to play important roles in diverse cellular and physiological processes, including cell cycle, cell proliferation, cell differentiation, cell

death, phenotype, and organ development.50-53 In addition, microRNAs have been implicated in the pathogenesis of various human diseases.53-57 Notably, accumulating evidence suggests that microRNAs have emerged as critical regulators in renal development and pathophysiology.58-64 The role of microRNAs in kidney diseases has been extensively reviewed.60, 61, 65-71

DNA Methylation

The epigenetic marks on DNA are DNA methylation. In eukaryotes, the methylation of DNA only occurs at cytosine residues.72 DNA methylation results from the covalent addition of a methyl group (CH3) commonly donated by S-adenosyl methionine

(SAM) to the 5-carbon position of the cytosine, creating 5mC by DNMTs.73 It occurs predominantly in CpG dinucleotides (C followed by G); however, there is now evidence showing DNA methylation in non-CpG contexts (CpH (H=A/C/T)), particularly in embryonic stem cells,74, 75 oocytes76 and brain tissues,77, 78 but with very low frequency. In mammalian genome, the majority of CpGs (70-80%) are methylated. CpG islands (CGI), which are CpG-enrich regions, are at least 200 bp long with a GC percentage over 50% and can be found in promoter regions, gene bodies as well as intragenic regions. CGIs within gene promoter regions are usually unmethylated (<10%). In mammals, DNA methylation is critical for many biological processes, such as embryonic development, genomic integrity, X chromosome inactivation (females), and genomic imprinting, and it is also linked to complexed human diseases including cancer.79-81

The mammalian DNA methylation machinery is composed of three parts: (I) DNA methylation writers – DNMTs to establish and maintain the methylation marks in the genome; (II) DNA methylation readers – methyl-CpG binding domain proteins (MBDs),

the zinc-finger proteins, and the ubiquitin-like with PHD and ring finger domain- containing (UHRF) proteins to read or interpret the methylation marks; (III) DNA methylation erasers – activation-induced cytidine deaminase/apolipoprotein B mRNA- editing enzyme complex (AID/APOBEC) and ten-eleven translocation (TET) methylcytosine dioxygenase for active demethylation.

DNA Methylation Writers: The DNMT family includes DNMT1, DNMT2, and

DNMT3 (3a, 3b, and 3L). DNMT1, DNMT3a, and DNMT3b are three major DNMTs required for DNA methylation and are conserved in mammals and plants. DNMT1 is the maintenance DNA methyltransferase which has a preference for hemimethylated DNA.

During cell division, the newly synthesized daughter DNA strands are recognized by

DNMT1 and methylated faithfully according to the parental DNA strands. DNMT1 is the most abundant DNMTs in most cell types, especially in proliferating cells. DNMT3a and

DNMT3b are de novo DNA methyltransferases, which establish the initial DNA methylation patterns during embryonic development.82, 83 DNMT3a and DNMT3b are highly expressed in embryonic tissues and undifferentiated embryonic stem cells, but are downregulated in differentiated cells and expressed at low levels in somatic tissues.84

DNMT3L (DNMT3-like) is not an active enzyme but stimulates the catalytic activity of

DNMT3a and DNMT3b by enhanced SAM binding.85 DNMT2 is not a DNA methylase but an RNA methylase for tRNAAsp, which methylates cytosine-38 of aspartic acid in the anticodon loop of tRNA.86 The major DNMTs are essential for development, as mice deficient in DNMT1 or DNMT3b are embryonic lethal, and DNMT3a deficient mice are born alive but die within a few weeks.87, 88

DNA Methylation Readers: The most important function of DNA methylation is transcriptional regulation. Generally, low levels of DNA methylation (hypomethylation) at gene promoters may correlate with activation of the gene through increased binding of transcription factors or lead to genomic instability.5 On the other hand, elevated DNA methylation (hypermethylation) in the promoter region of a gene leads to the loss of the associated gene expression.3, 4 DNA methylation can directly reduce gene expression by inhibiting the binding of transcription factors. Also, DNA methylation can repress gene transcription indirectly via specific proteins that bind to methylated CpGs referred as ‘DNA methylation readers’. DNA methylation readers are characterized into three families: the

MBD family, the zinc-finger proteins, and the UHRF family. There are five members in

MBD family, including methyl-CpG-binding protein 2 (MeCP2) and MBD1–4.89, 90 Except

MBD 4, all other MBDs bind to methylated cytosine and repress transcription by recruiting various repressor complexes through their transcriptional repression domain (TRD).90 For example, MeCP2 binds a single symmetrically methylated CpG site and recruits transcriptional co-repressors SIN3A and histone deacetylases (HDACs);91 MBD2 binds to methylated DNA and interacts with MeCP1, a large chromatin remodeling complex.92 The zinc-finger proteins include ZBTB4, ZBTB38 and Kaiso.89 The UHRF family consists of

UHRF1 and UHRF2. Unlike most of the MBD family and the zinc-finger proteins, the

UHRF family does not function to repress transcription but recruits DNMT1 to the hemimethylated DNA for maintenance of DNA methylation.90

DNA Methylation Erasers: DNA methylation has been traditionally considered to be very stable and can be inherited to daughter cells; however, recent studies showed that they are not permanent and can be erased, which is called demethylation.

Demethylation can take place through either passive or active mechanisms (Figure 5).

Passive methylation occurs in dividing cells. During cell division, if DNMT1 is inhibited or absent, the newly synthesized DNA strands cannot be methylated, leading to passive demethylation in the daughter cells and dilution of DNA methylation in the cell population.

On the other hand, active demethylation can be achieved in both dividing and non-dividing cells through the enzymatic replacement of 5mC to cytosine by a set of enzymes, such as

AID/APOBEC and TET family.93 The first active DNA methylation mechanism involves

AID/APOBEC, which removes the amino group from 5mC, generating a T/G mismatch. It is then recognized and corrected by thymine DNA glycosylase (TDG) and different DNA repair mechanisms.94 Another mechanism is mediated by the TET family of iron-dependent deoxygenases. TET family has three members (TET1, TET2 and TET3), which mediates the oxidation of 5mC to 5hmC. Two mechanisms have been reported to convert 5hmC to cytosine in mammals. First, 5hmC can be further oxidized to 5-formylcytosine (5fC) and

5-carboxylcytosine (5caC). Then 5fC and 5caC can be excised by TDG and replaced by cytosine via base excision repair (BER).95, 96 Second, 5hmC can be deaminated by

AID/APOBEC to form 5-hydroxymethyl-uracil (5hmU).97 Unlike the DNMT deficient mice, mice deficient in AID or TET1 or TET2, or both TET1 and TET2 are viable,98-100 but deletion of TET3 is neonatal lethal.101 Therefore, it seems TET3 is more important for early development.

TET 1,2,3 AID/APOBEC

Figure 5: Mechanisms of DNA methylation and demethylation.

The initial DNA methylation patterns are established by de novo DNA methyltransferases- DNMT3a

and DNMT3b during early development. DNMT1 maintains these marks in daughter cells. Passive

demethylation occurs when DNMT1 is inhibited or absent, and active demethylation can be achieved

by the enzymatic replacement of methyl-cytosine to cytosine by a set of enzymes, such as activation-

induced deaminase (AID) and ten-eleven translocation methylcytosine dioxygenase (TET). The figure

is modified from Wu SC et al., 2010.

4. Methods for DNA methylation analysis

There are three major strategies to distinguish methylated cytosine to unmethylated cytosine, including bisulfite conversion, restriction enzyme digestion and affinity purification (See Figure 6).102

B. Restriction enzyme digestion (1) (2) A. Bisulfite conversion

(Methylation- (Methylation- sensitive) insensitive)

C. Affinity purification

Figure 6: Three major strategies for distinguishing methylated cytosine to unmethylated cytosine.

(A) Bisulfite conversion. (B) Restriction enzyme digestion. (1) Genomic DNA is digested with either

HpaII (methylation-sensitive enzyme) or MspI (methylation –insensitive enzyme). (2) Genomic DNA is

digested with McrBC, which cuts methylated cytosine. (C) Affinity purification. The figure is modified

from Zilberman D et al., 2007.

Bisulfite conversion, the first method discovered to distinguish methylated cytosine and unmethylated cytosine, can only cause the conversion of unmethylated cytosine to uracil, but methylated cytosine remains intact. After PCR amplification, uracil becomes thymine. Therefore, after bisulfite conversion and PCR amplification, methylated cytosine is still cytosine, but unmethylated cytosine is converted to thymine.102-104 The advantage of bisulfite conversion based method is that it creates sequence-dependent

differences and can be utilized at single-base resolution. It is important to note, however, bisulfite conversion based methods depend on the complete conversion of unmethylated cytosines. Over the past few years, a number of different techniques have been developed that utilize bisulfite conversion. For instance, bisulfite cloning sequencing, pyrosequencing, and methylation specific PCR are methods for local specific DNA methylation analysis, and whole-genome shotgun bisulfite sequencing (WGSBS), reduced representation bisulfite sequencing (RRBS) and Illumina infinium Methylation BeadChip arrays are genome-wide DNA methylation methods.105

Restriction enzyme digestion-based methods employ methylation-sensitive enzymes or methylation-dependent enzymes. Methylation-sensitive enzymes (HpaII,

BstuI, TaqI, HhaI, HpyrictiCH4IV, Hpy99I) can only digest unmethylated DNA, which can be used to enrich for methylated DNA, while methylation-dependent enzymes (DpnI,

FspEI, LpnPI, McrBC, MspJI) can only cut methylated DNA, which can be used to enrich for unmethylaed DNA. For the enrichment of methylated DNA, the most commonly used pair of restriction enzymes are HpaII and MspI. HpaII, a methylation-sensitive enzyme, recognizes the sequence CCGG, but does not cut if it is methylated. In contrast, MspI, a methylation-insensitive enzyme, cleaves the same sequence regardless of the methylation status. In the experiment, genomic DNA is digested with either HpaII or MspI. Therefore, comparing these two libraries generated by HpaII or MspI, respectively, DNA fragments in HpaII but not in MspI libraries are derived from methylated regions.102 Examples of restriction enzyme digestion-based methods relying on HpaII and MspI are HpaII tiny fragment Enrichment by Ligation-mediated PCR (HELP), Methl-seq, luminometric methylation assay (LUMA).105 McrBC is a more recently discovered restriction enzyme

used to enrich for unmethylated DNA. McrBC is a DNA methylation-dependent enzyme, which only specifically cleaves between two methylated cytosines (mC: 5mC, 5hmC or

N4-methylcytosine), either AmC or GmC. McrBC can cut these sites within a distance of

40 - 3000 bp, but the optimal distance is 55 - 100 bp.102, 103 Hence, it is capable of digesting densely methylated DNA, enriching for unmethylated DNA. The techniques that utilize

McrBC include comprehensive high-throughput arrays for relative methylation (CHARM) and microarray-based methylation assessment of single samples (MMASS).105 Although there are many different restriction enzymes to choose from and it is easy to perform, the information acquired by restriction enzyme-based methods is limited on CpGs within or near the restriction site.

Another popular method is affinity purification of methylated DNA. There are two techniques used to capture methylated DNA, one is methylated DNA immunoprecipitation (MeDIP) using anti-5mC antibodies, and another is methyl-CpG binding domain-based capture (MBDCap) using MBD proteins. The enriched methylated

DNA fragments are then amplified and subjected to next generation sequencing or DNA microarray.102, 104 The advantage of using affinity purification of methylated DNA is that it is easy to implement and able to capture all CpG sites with no bias. However, the major limitation is that it is difficult to quantify and does not reveal detailed methylation patterns.

5. DNA Methylation in kidney injury and repair

Although DNA methylation is one of the first epigenetic modifications identified in 1948 and the most broadly studied and well-characterized epigenetic modification, the study of DNA methylation in kidney diseases is still in its infancy. Recently, several studies performed DNA methylation profiling in kidney diseases, including CKD,106-108 diabetic

nephropathy,109-112 and kidney fibrosis,113 which indicate the involvement of DNA methylation alterations in these diseases. Importantly, DNMT1, but not DNMT3a and

DNMT3b was induced in folic acid-induced renal fibrosis, and renal fibrosis was reduced in DNMT1+/- heterozygous mice. Moreover, 5′-azacytidine, a DNA methylation inhibitor, attenuated folic acid-induced renal fibrosis in mice.113 In addition, Klotho, a gene associated with renal fibrosis, was found promoter hypermethylated in the kidney tissues and peripheral blood mononuclear cells from patients with CKD. And the degree of its hypermethyaltion was associated with the severity of CKD.114

In AKI, very limited is known about the role and regulation of DNA methylation.

Several lines of evidence have shed light on DNA methylation involved in AKI, which suggests DNA methylation participates in AKI pathogenesis. DNA methylation alteration in AKI was first reported in 2006. Pratt et al.115 discovered DNA methylation changes was involved in ischemic AKI. In a rat model of ischemia-reperfusion, they found aberrant demethylation of a specific cytosine in the interferon-γ (IFN-γ) responsive element within the complement factor C3 promoter in response to cold ischemia and it was further demethylated during subsequent warm reperfusion. In another study,116 the same group further found demethylation of the C3 promoter not only occurred in acute injury phase (4 hours ischemic injury and transplantation for 48 hours) but also persisted in chronic injury phase (4 hours ischemic injury and transplantation for 6 months). They speculated that this aberrant demethylation in C3 promoter could account for the upregulation of C3 in ischemia-reperfusion injury. Another study examined the changes of global levels of 5hmC and 5mC in a mouse model of ischemic AKI.117 They showed that global level of 5hmC was decreased, while the global level of 5mC did not change after ischemic AKI.

Interestingly, these changes were associated with the decrease in TET1 and TET2 expression which are involved in active DNA demethylation by catalyzing the conversion of 5mC to 5hmC. In addition, their DNA hydroxymethylome profiling data showed that lower levels of 5hmC in the promoter regions of Cxcl10 and Ifngr2 with increased expression of these two genes after ischemia-reperfusion. Their following study118 first revealed the role of hyper-hydroxymethylation in gene body which is positively related to gene expression in mouse kidney. In contact-freezing induced AKI model,119 the authors developed a novel TaqMan-based assay to distinguish methylated and unmethylated DNA fragments. Using this method, they found the promoter region of Slc22a12, a urate transporter specifically expressed in proximal tubular cells, was hypomethylated in plasma from mice with AKI induced by contact-freezing. Moreover, in a human kidney transplantation study,120 using quantitative methylation-specific PCR, Mehta et al. found hypermethylation of CALCA promoter in urine from kidney transplant patients at day 2 after transplantation compared with that from healthy controls, and also in urine from deceased compared to living donor patients. Furthermore, in human AKI, another gene renal kallikrein (KLK1) was found promoter hypermethylation in blood DNA in established AKI compared to healthy controls, although KLK1 promoter methylation in urine tended to be elevated in AKI but without significant difference.121 These suggest

DNA methylation alteration could be used as a potential biomarker of AKI. These studies have shown that in AKI, DNA methylation changes occur at promoters of many genes, such as C3, CALCA, KLK1, and Slc22a12, indicating a role of DNA methylation in AKI.

However, whether DNA methylation is renoprotective or pathogenic in AKI is unknown.

II. MATERIALS AND METHODS

1. Antibodies and special reagents:

Antibodies used in this study were purchased from the following companies: rabbit monoclonal anti-DNMT1 (5032), rabbit monoclonal anti-Caspase-3 (9665), rabbit polyclonal anti-Cleaved caspase 3 (9661), rabbit monoclonal anti-PARP (9532), rabbit polyclonal anti-Vimentin (3932) and rabbit monoclonal anti-GAPDH (5174s) from Cell

Signaling Technology (Danvers, MA); rabbit polyclonal anti-Cyclophilin B (ab16045), rabbit polyclonal anti-Fibronectin (ab2413), rabbit polyclonal anti-α-SMA from Abcam

Inc. (Cambridge, MA); rabbit polyclonal anti-Bcl6 (sc-858), rabbit polyclonal anti-

DNMT3a (sc-20703) mouse monoclonal anti-DNMT3b (sc-81252), goat polyclonal anti-

ICSBP (Irf8, sc-6058) from Santa Cruz Biotechnology Inc. (Dallas, TX), and mouse monoclonal anti-β-actin (A5441) from Sigma-Aldrich (St. Louis, MO). Secondary antibodies for immunoblotting were from Thermo Scientific (Rockford, IL) and for immunohistochemistry staining from Dako (Carpinteria, CA), respectively. Fluorescein- labeled lotus tetragonolobus lectin (LTL) was bought from Vector Labs (Burlingame, CA).

Cis-Diammineplatinum(II) dichloride crystalline (Cisplatin, P4394-1G) and 5-aza-2’- deoxycytidine (A3656-50MG) were purchased from Sigma-Aldrich (St. Louis, MO).

Unless specifically indicated, all other reagents were purchased from Sigma-Aldrich.

2. Animals:

C57BL/6 mice were purchased from the Jackson Laboratory (Bar Harbor, ME).

DNMT1flox/flox and DNMT3aflox/flox mice were kindly provided by Dr. Guoping Fan’s laboratory at the University of California (Los Angeles, CA) and the generation of these two strains was described previously.122-124 DNMT1flox/flox mice have a mixed genetic background of Balb/C and 129Sv/J, and DNMT3aflox/flox mice have the C57BL/6 background. PEPCK-Cre mice were originally obtained from Dr. Volker Haase

(Vanderbilt University, Nashville, TN) with a mixed genetic background of Balb/C,

129Sv/J, and C57BL/6.125 To establish kidney proximal tubule-specific DNMT1 or

DNMT3a knockout mice, DNMT1flox/flox or DNMT3aflox/flox mice were crossed with

PEPCK-Cre mice according to the breeding protocol shown in Figure 7 and 8.

flox/flox +/+ Cre DNMT1 XY DNMT1 X X

flox/+ flox/+ Cre flox/flox DNMT1 XY DNMT1 X Y DNMT1 XX flox/+ DNMT1 XX flox/+ Cre DNMT1 X X

flox/+ flox/+ Cre DNMT1 XY DNMT1 X X

flox/flox flox/flox Cre DNMT1 XY DNMT1 X X

flox/flox flox/flox DNMT1 XX DNMT1 XY PT-DNMT1-WT flox/flox Cre flox/flox Cre DNMT1 X X DNMT1 X Y PT-DNMT1-KO

Figure 7: Breeding protocol to generate PT-DNMT1-KO mice.

flox/flox +/+ Cre DNMT3a XY DNMT3a X X

flox/+ flox/+ Cre flox/flox DNMT3a XY DNMT3a X Y DNMT3a XX flox/+ DNMT3a XX flox/+ Cre DNMT3a X X

flox/+ flox/+ Cre DNMT3a XY DNMT3a X X

flox/flox flox/flox Cre DNMT3a XY DNMT3a X X

flox/flox flox/flox DNMT3a XX DNMT3a XY PT-DNMT3a-WT

flox/flox Cre flox/flox Cre DNMT3a X X DNMT3a X Y PT-DNMT3a-KO

Figure 8: Breeding protocol to generate PT-DNMT3a-KO mice.

Proximal tubule-specific DNMT1 and DNMT3a double knockout mice (PT-

DNMT1/3a-DK) were generated by crossing DNMT1flox/flox DNMT3a+/+ XY mice with

DNMT1+/+ DNMT3aflox/flox XCreX mice as depicted in Figure 9. In these experiments, littermate wild-type and knockout mice were compared. All mice were maintained in the animal facility of Charlie Norwood Veterans Affairs Medical Center at Augusta under a

12 hour-light/12 hour-dark pattern with free access to food and water. All animal experiments were conducted following a protocol approved by the Institutional Animal

Care and Use Committee.

flox/flox +/+ +/+ flox/flox Cre DNMT1 DNMT3a XY DNMT1 DNMT3a X X

flox/+ flox/+ Cre flox/+ flox/+ DNMT1 DNMT3a X Y DNMT1 DNMT3a XX flox/+ flox/+ flox/+ flox/+ Cre DNMT1 DNMT3a XY DNMT1 DNMT3a X X

…… …… flox/flox flox/flox flox/flox flox/flox Cre DNMT1 DNMT3a XX (1/64) DNMT1 DNMT3a X Y (1/64)

flox/flox flox/flox Cre flox/flox flox/flox DNMT1 DNMT3a X X (1/64) DNMT1 DNMT3a XY (1/64)

flox/flox flox/flox flox/flox flox/flox DNMT1 DNMT3a XX DNMT1 DNMT3a XY PT-DNMT1/3a-WT flox/flox flox/flox Cre flox/flox flox/flox Cre DNMT1 DNMT3a X X DNMT1 DNMT3a X Y PT-DNMT1/3a-DK

Figure 9: Breeding protocol to generate PT-DNMT1/3a-DK mice.

3. Genotyping:

Genomic DNA extracted from the ear or kidney cortex was used for PCR genotyping. All the primers synthesized by Integrated DNA Technologies are listed in

Table 2.

For PEPCK-Cre, primers Cre-forward and Cre-reverse were used, and a 370 bp fragment was amplified in the presence of the PEPCK-Cre transgene. The PCR conditions are 94°C for 4 min; 30 cycles of 94°C for 1 min, 55°C for 30 s, and 72°C for 40 s; and

72°C for 10 min.

For DNMT1, primers DNMT1-P1 and DNMT1-P2 were used to determine the wild-type allele and the DNMT1-floxed allele, which amplify a 368 bp fragment from the

DNMT1-floxed allele and a 334 bp fragment from the wild-type allele. Primers DNMT1-

P1 and DNMT1-P3 were used to determine the recombinant DNMT1 allele, which amplify a fragment from DNMT1 recombinant allele. The PCR conditions are 94°C for 3 min; 35 cycles of 94°C for 30 s, 58°C for 30 s, and 72°C for 1 min; and 72°C for 10 min.

For DNMT3a, primers DNMT3a-P1 and DNMT3a-P2 were used. An 860 bp fragment was amplified from wild-type DNMT3a allele, a 1200 bp fragment was amplified from DNMT3a-floxed allele, and a 350 bp fragment from recombinant DNMT3a allele.

The PCR conditions are 94°C for 3 min; 35 cycles of 94°C for 1 min, 58°C for 1 min, and

72°C for 2 min; and 72°C for 10 min.

Table 2: Primers used for genotyping

Primer Name Sequence

Cre-forward 5'-ACCTGAAGATGTTCGCGATTATCT-3'

Cre-reverse 5'-ACCGTCAGTACGTGAGATATCTT-3'

DNMT1-P1 5'-GGGCCAGTTGTGTGACTTGG-3'

DNMT1-P2 5'-CTTGGGCCTGGATCTTGGGGA-3'

DNMT1-P3 5'-ATGCATAGGAACAGATGTGTGC-3'

DNMT3a-P1 5'-CTGTGGCATCTCAGGGTGATGAGC-3'

DNMT3a-P2 5'-GGTAGAACTCAAAGAAGAGGCGGC-3'

4. Animal model of cisplatin-induced AKI:

Male mice of 8 to 12 weeks were used for cisplatin treatment. Briefly, the mice were injected with a single dose of 30 mg/kg cisplatin intraperitoneally, and control animals received the saline injection. Our pilot study revealed that PT-DNMT1/3a mice were more sensitive to cisplatin-induced nephrotoxicity (~50% PT-DNMT1/3a-WT mice died at day 4 after cisplatin injection), therefore, we used 20 mg/kg cisplatin for PT-

DNMT1/3a mice. After indicated time, kidneys and blood samples were collected for analysis.

5. Animal models of renal ischemia-reperfusion:

Male mice of 8 to 13 weeks were used for renal ischemia-reperfusion surgery as described previously.126 Briefly, the mice were anesthetized by 50 mg/kg pentobarbital sodium (intraperitoneal injection) and kept on a homeothermic blanket. Renal ischemia 30 minutes reperfusion 48 hours (I30R48h) model: Flank incisions were made to expose both sides of renal pedicles for bilateral clamping for 30 minutes to induce renal ischemia. After

30 minutes reperfusion, clamps were released for reperfusion for 48 hours. Unilateral renal ischemia-reperfusion (UI30R2w) model: The left kidney was exposed for unilateral clamping for 30 minutes followed by 2 weeks reperfusion. For sham control, mice were followed by an identical procedure but without clamping the renal pedicle. After indicated time, blood and kidneys were harvested for further examinations.

6. Reduced representation bisulfite sequencing (RRBS):

RRBS was performed by Integrated Genomics Core and data was analyzed by

Bioinformatics Core in Augusta University as previously described.127 In brief, first, genomic DNA was extracted from kidney cortex and outer medulla using DNeasy Blood

& Tissue Kit from Qiagen (Hilden, Germany). Then 1 μg DNA for each sample was digested with a methylation-insensitive enzyme MspI (New England Biolabs, Ipswich,

MA). Subsequently, the fragment ends were filled in and adapters were ligated. Next, the ligation products were size selected by agarose gel electrophoresis with a size of 150 bp-

450 bp. Size-selected DNA fragments were then subjected to bisulfite conversion using

EpiTect Fast DNA Bisulfite Kit from Qiagen. The libraries were generated by PCR using the PfuTurbo Cx Hotstart polymerase (Agilent Technologies, Santa Clara, CA) and sequenced by an Illumina Genome Analyzer IIx (Ilumina, San Diego, CA).

7. Renal function analysis:

Renal function was analyzed by measuring blood urea nitrogen (BUN) and serum creatinine as described previously.128 In brief, at indicated time, blood samples were collected and centrifuged to collect serum. Then the BUN and serum creatinine levels were determined by analytical kits from Stanbio Laboratory (Boerne, TX). The detailed protocol for BUN measurement is as follows: BUN color reagent was mixed with BUN acid reagent

(color reagent : acid reagent = 1:2) to obtain the reaction mixture. Then serum or standard was added to reaction mixture, mixed and incubated at 100°C for 12 min in a heating block.

After that, the tubes were immediately placed on ice for 3 min. Absorbance was read and recorded at 520 nm. The detailed protocol for serum creatinine measurement is as follows:

First, the reaction mixture (Acid Reagent : Base Reagent=1:1) were pre-warmed at 37°C for 3 min. Then serum or standard was added to reaction mixture, mixed and immediately placed in cuvette well. Absorbance was read and recorded at 20s and 80s at 510 nm.

8. Histological examination:

Kidneys were harvested, fixed with 4% paraformaldehyde, embedded in paraffin and sectioned at 4 μm for histological examination. Renal histology was examined by hematoxylin and eosin (H&E) staining to analyze kidney tissue damage following a standard protocol. Loss of brush border, tubular dilation and disruption, protein/cell cast formation and cell lysis (nuclear loss) were considered as renal tubular damage. Kidney tissue damage was examined by an experimental pathologist in a blind manner and scored according to the percentage of renal tubular damage as follows: 0, no damage; 1, <25%; 2,

26–50%; 3, 51–75%; 4, >75%.

9. TUNEL assay:

Renal apoptosis was examined by terminal deoxynucleotidyl transferase (TdT) dUTP nick end labeling (TUNEL) assay using an in situ cell death detection kit from Roche

Applied Science (Indianapolis, IN) as described recently.128 In brief, kidney tissue sections were deparaffinized in xylene and rehydrated in graded ethanol series to water.

Subsequently, the tissue sections were permeabilized in 0.1 M sodium citrate (pH 6.0) at

65°C for 1 hour and then incubated with TUNEL reaction mixture (Enzyme solution :

Label solution = 1:9) in humidified chamber at 37°C for 1 hour. After PBS wash, the slides were mounted with Prolong Anti-fade Kit from Molecular Probes (Eugene, OR). The positive staining with nuclear DNA fragmentation was detected by fluorescence microscopy. For quantification, 10-15 fields were randomly selected from the kidney cortex and outer medulla in each sample and the number of TUNEL-positive cells per mm2 was evaluated.

10. Immunohistochemical staining:

Paraffin-embedded section preparation was the same as the procedures used in

H&E and TUNEL staining. After rehydration, antigen retrieval was performed in 10 mM

EDTA (pH 8.0) for DNMT1 or 10 mM sodium citrate (pH 6.0) for Cleaved caspase 3 and

Bcl6 at 100°C for 1 hour, followed by 1 hour blocking. Then tissue sections were exposed to primary antibody overnight at 4°C. After washing, tissue sections were incubated with secondary antibody with horseradish peroxidase (HRP) labelled polymer from Dako

(Carpinteria, CA) for 1 hour and then the slides were developed with ImmPACT DAB

Peroxidase Substrate (Vector Laboratories, Burlingame, CA). For DNMT1 staining, the tissue sections were further stained with fluorescein isothiocyanate (FITC)-labeled LTL, which is the proximal tubule-specific marker; and nuclei were counterstained with

Hoechst33342 (Molecular Probes, Eugene, OR). Finally, the slides were mounted with

Prolong Anti-fade Kit.

11. Masson’s trichrome staining:

Masson’s trichrome staining was used to detect collagen fibers in tissues for fibrosis analysis. Trichrome Stain (Masson) Kit from Sigma was used following a standard protocol as described previously.129 Briefly, after deparaffinized and rehydrated, the tissue sections were re-fixed in Bouin's solution at room temperature overnight. After washing, the sections were stained with Biebrich scarlet-acid fuchsin solution for 10 minutes, and then differentiated in phosphomolybdic-phosphotungstic acid solution for 13 minutes, followed by aniline blue staining for 15 minutes. Then, the tissues were differentiated in 1% acetic acid solution for 4 minutes. Finally, after washing and dehydration, the slides were mounted with resinous mounting medium. For quantification, 10-15 collagen-positive

stained fields were randomly selected for each sample and the percentage of fibrotic area was quantified by Image J.

12. Total RNA isolation and Real-time PCR:

Total RNA was isolated from cells or kidney cortex and outer medulla using mirVana™ miRNA Isolation Kit (Life Technologies, Carlsbad, CA) according to the manufacturer’s instructions. Reverse transcription into cDNA was then performed with 1

µg aliquots of total RNA from each sample using iScript™ cDNA Synthesis Kit (Bio-Rad,

Hercules, CA). Real-Time PCR was performed using iTaq™ Universal SYBR® Green

Supermix (Bio-Rad, Hercules, CA) with StepOne Real-Time System (Applied Biosystems,

Foster City, USA) and all reactions were performed in triplicate. For normalization, β-actin was used as internal control, followed by normalization to the control group. Real-time

PCR was conducted using the primers listed in Table 3.

Table 3: Primers used for real-time PCR

Primer Name Sequence

Irf8-mouse-forward 5'-TGTCTCCCTCTTTAAACTTCCC-3'

Irf8-mouse-reverse 5'-GAAGACCATGTTCCGTATCCC-3'

Irf8-rat-forward 5'-TGTCTCCCTCTTTAAACTTCCC-3'

Irf8-rat-reverse 5'-GAAGACCCTGTTCCGCATCCC-3'

Bcl6-mouse-forward 5'-TCGACACATGCAGGAAGTTC-3'

Bcl6-mouse-reverse 5'-ATATTGTTCTCCACGAACTCAC-3'

β-actin-mouse-forward 5’-GACTCATCGTACTCCTGCTTG-3’

β-actin-mouse-reverse 5’-GATTACTGCTCTGGCTCCTAG-3’

β-actin-rat-forward 5’-GGCATAGAGGTCTTTACGGATG-3’

β-actin-rat-reverse 5’-TCACTATCGGCAATGAGCG-3’

Hoxc10-mouse-forward 5’-CACCTCGGATAACGAAGCTAA-3’

Hoxc10-mouse-reverse 5’-AGCGTCTGGTGTTTAGTATAGG-3’

Sycp2-mouse-forward 5’-AGCGATGCTTGTGACTCTG-3’

Sycp2-mouse-reverse 5’-TGCACCATGACTGAGCTG-3’

Cuedc1-mouse-forward 5’-ACGACTTCAAGACTATGTTTCCT-3’

Cuedc1-mouse-reverse 5’-GCCTCCAGGTTCATCTGTAAC-3’

Spry4-mouse-forward 5’-TGCACCATGACTGAGCTG-3’

Spry4-mouse-reverse 5’-AGCGATGCTTGTGACTCTG-3’

Hdac5-mouse-forward 5’-GCACGTTTTGCTCCTGGA-3’

Hdac5-mouse-reverse 5’-AGCTTACCCACCGTCCT-3’

Nrip1-mouse-forward 5’-GATGTCATCGGAGTCTTCAGAT-3’

Nrip1-mouse-reverse 5’-TCGAAGGCGTGGACTGT-3’

13. Cells:

The immortalized rat kidney proximal tubular epithelial cell (RPTC) line was originally obtained from Dr. Ulrich Hopfer (Case Western Reserve University, Cleveland,

OH) and was maintained as described before.130 RPTC Irf8 knockdown or scramble cells were generated by transfection of Irf8 shRNA plasmids or scramble sequence shRNA plasmids with GFP tag from Origene (Rockville, MD) with Lipofectamine 2000 and selected by 1 g/ml puromycin (Clontech Laboratories, Mountain View, CA) for two weeks to obtain cells with stable shRNA expression. Then puromycin-resistant and GFP- positive cell colonies were selected and Irf8 knockdown efficiency was determined by immunoblot and real-time PCR. 35

14. Transient Transfection:

Transient transfection was performed using Lipofectamine 2000 (Thermo Fisher

Scientific, Waltham, MA) following the manufacturer’s instruction. Bcl6 overexpression plasmids and pCMV6-Entry plasmids (empty vector) were purchased from Origene. In brief, RPTC cells were plated at the density of 0.4 x 106 cells per 35 mm dish. After overnight growth, the cells reached 50-60% confluence and then were transfected with 1

µg Bcl6 plasmid or pCMV6-Entry plasmid. After 4 hours incubation, the serum-free medium was changed to full culture medium. And the transfected cells were used for treatment in the next day.

15. Primary proximal tubular cells isolation and maintenance:

Primary proximal tubular cells isolation and maintenance were performed as described previously.131 Briefly, kidney cortex and outer medulla were finely minced and then digested with collagenase IV with trypsin inhibitor in Hanks’ solution. After stopped by 10% horse serum, the renal tubules were collected by centrifugation at 50g for 2 minutes. Then the proximal tubules were purified using Percoll density gradient centrifugation (32% Percoll in DMEM/ F12 medium, 2000g for 10 minutes). After washes, the proximal tubules were cultured in DMEM/F12 medium supplemented with 10% FBS and growth factors, including 0.05 µM hydrocortisone, 5 mg/ml transferrin, 50 µM vitamin

C and 5 mg/ml insulin. All centrifugations were performed at 4°C. After 6-7 days of growth, the isolated primary proximal tubular cells were plated at 0.5 x106 cells per 35 mm dish.

16. Cisplatin treatment in cells:

For cisplatin treatment, the cells were seeded in 35 mm dishes and reached full confluence after overnight growth. RPTC cells and primary proximal tubular cells were incubated with 20 μM or 50 μM cisplatin in culture medium, respectively. To determine the effects of DNA methylation inhibitor, the cells were pretreated with 1 μM 5-aza-2’- deoxycytidine for 3 hours, followed by cisplatin treatment for 16 hours. After incubation for indicated time, apoptosis was determined by morphologic criteria and cells were harvested to collect cell lysates for various biochemical analysis.

17. Hypoxia treatment in cells:

RPTC cells were plated into 35 mm dish and reached full confluence on the day of hypoxia treatment. For hypoxia treatment, the cells were transferred into hypoxia (1% O2) chamber and changed with pre-balanced medium, then incubated for 48 hours to induce apoptosis. For the control group, the cells were kept in the normal oxygen condition (21%) chamber.

18. Morphological analysis of apoptosis in cultured cells:

Cell apoptosis was examined by morphology as described previously.130 In brief, after cisplatin treatment, cell nuclei were stained with 10 μg/ml Hoechst 33342. Then cellular and nuclear morphology were observed by phase contrast and fluorescence microscopy, respectively. Cellular shrinkage, the formation of apoptotic bodies and nuclear condensation and fragmentation are the typical apoptotic morphology. For each condition,

3 to 5 fields with approximately 200 cells per field were randomly selected, and the cells showing typical apoptotic morphology were counted to determine the percentage of apoptosis. Each experiment was repeated at least three times independently.

19. Immunoblot analysis:

Proteins were extracted from cells or renal cortex and outer medulla using 2% sodium dodecyl sulfate (SDS) lysis buffer as previously described.130 For each sample, equal amounts (20 µg for cell lysate and 100 µg for tissue lysate) of protein were resolved by 10% SDS or Bis-tris protein gels, and were then transferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were blocked with 5% fat-free milk for 1 hour at room temperature and were subsequently incubated with primary antibody overnight at 4 °C. After washing with PBST [1 X PBS with 0.1% Tween 20 (MP

Biomedicals, Santa Ana, CA)), the membranes were incubated with HRP conjugated secondary antibody for 1 hour, and after washing, antigen-specific signals were visualized with ClarityTM Western ECL Blotting Substrates from Bio-Rad.

20. Statistical Analysis:

Quantitative data was expressed as means ± standard deviation (SD). Statistical differences between two groups were determined by 2-tailed paired t-test and statistical differences between multiple groups were determined by two-way ANOVA. p<0.05 was considered statistically significant.

III. RESULTS

Specific Aim 1: To establish proximal tubule-specific DNMTs KO mouse models (PT-DNMT1 and PT-DNMT3a single KO mouse models, and

PT-DNMT1/3a double knockout (DK) mouse model).

To investigate the role of DNA methylation in the pathogenesis of AKI, first, we decided to generate conditional knockout mouse models of DNMTs which are especially deleted in proximal tubular cells. For our study, we established PT-DNM1-KO, PT-

DNMT3a-KO and PT-DNMT1/3a-DK mouse models.

Sub-aim 1a: To establish and characterize PT-DNMT1-KO mouse model.

DNMT1 is responsible for the maintenance of DNA methylation. Since global knockout of DNMT1 is lethal,87 we decided to generate conditional (kidney proximal tubule-specific) knockout mice via the Cre-loxp system. DNMT1flox/flox mice were generated by inserting loxP sites flanking exons 4 and 5 in DNMT1 allele, which contain the localization domain and the entire catalytic domain.122 To generate PT-DNMT1-KO mice, DNMT1flox/flox mice were crossed with PEPCK (phosphoenolpyruvate carboxykinase)-Cre mice. PEPCK-Cre mice express the X-linked Cre-recombinase under control of a modified rat cytosolic PEPCK promoter, which is expressed predominantly in renal proximal tubules and marginally in periportal hepatocytes.125 In our experiments, only male mice were used to make sure the correct genotype because PEPCK-Cre is linked 39

to X chromosome. The breeding protocol is shown in Figure 7. Briefly, male mice with the floxed DNMT1 alleles (DNMT1flox/flox XY) were crossed with female PEPCK-Cre mice

(DNMT1+/+XCreX). After the first generation, heterozygous male progenies (DNMT1flox/+

XCreY) were selected to be further bred with DNMT1flox/flox XX female mice to generate

DNMT1flox/flox XY and DNMT1flox/flox XCreX mice, which are ideal breeders to produce

DNMT1flox/flox XY (PT-DNMT1-WT) and DNMT1flox/flox XCreY (PT-DNMT1-KO) littermate mice.

To confirm the genotypes, PCR was performed using 2 sets of primers (DNMT1-

P1, -P2 and -P1, -P3) as shown in Figure 10A. Primers P1 and P2 amplify wild-type allele

(334 bp) and floxed allele (368 bp), and primers P1 and P3 amplify the recombinant allele after deletion. Genomic DNA from kidney cortex was used to verify the deletion in PT-

DNMT1-KO mice (~70-80% of kidney cortex is proximal tubules), and DNA from the eye was used as negative control. As shown in Figure 10B, only the renal cortical tissues of

PT-DNMT1-KO mice showed the recombinant DNMT1 allele, but not in the renal cortical tissues of PT-DNMT1-WT mice or the eyes of PT-DNMT1-WT and -KO mice. This confirmed Cre-mediated deletion of DNMT1 in proximal tubular cells. The deletion of

DNMT1 in the proximal tubules was also confirmed by immunoblot (Figure 10C) and immunohistochemical staining (Figure 10D). By immunoblot analysis, the DNMT1 protein levels were suppressed in kidney cortex of PT-DNMT1-KO mice compared with those of littermate PT-DNMT1-WT mice. By immunohistochemical staining, lotus tetragonolobus lectin (LTL) staining was used to label proximal tubules, specifically their brush border. As shown in Figure 10D, positive DNMT1 staining was detected in all nuclei of proximal tubular cells in PT-DNMT1-WT mice, but it was absent in the majority of

proximal tubular cells in PT-DNMT1-KO mice. DNMT1 expression was observed in a small population of the kidney cortical tissues of PT-DNMT1-KO mice, likely reflecting the 70-80% efficacy of PEPCK-Cre-mediated gene ablation in proximal tubular cells125 and the presence of non-proximal tubular cells in the renal cortex.

Figure 10: Characterization of PT-DNMT1-KO mouse model.

(A) Schematic representation of the engineered locus of DNMT1 before and after PEPCK-Cre-

mediated deletion and the positions of the PCR primers. (B) PCR-based genotyping. Genomic

DNA was extracted from kidney cortex or eye (negative control) for PCR amplification to detect

WT-DNMT1 allele, floxed-DNMT1 allele, the recombinant allele of DNMT1 and PEPCK-Cre

allele. (C) Immunoblot analysis of renal cortical lysate confirming DNMT1 deletion in PT-

DNMT1-KO tissues. GAPDH was used as protein loading control. (D) Immunohistochemical

staining to verify DNMT1 ablation from proximal tubular cells in PT-DNMT1-KO mice. Kidney

tissues from PT-DNMT1-WT and -KO mice were fixed and processed for immunohistochemical

staining of DNMT1, followed by staining with LTL to show proximal tubules. Please note that

DNMT1 staining is in cell nuclei, whereas LTL staining is on the brush border of proximal tubules.

The merged images are the magnified images of the boxed areas and further show the absence of

DNMT1 in many proximal tubular cells in PT-DNMT1-KO tissues. Scale bar: 50 µm.

Under normal conditions, PT-DNMT1-KO mice showed no apparent abnormalities in kidney size, renal function or histology. They had similarly low BUN level as wild-type mice at the age of 8-14 weeks and 6 months (Figure 11A). And no phenotypic alterations were observed in kidney histology of these mice (Figure 11B).

Figure 11: Renal function and histology in PT-DNMT1-KO mice.

(A) Blood urea nitrogen (BUN) of PT-DNMT1-WT and -KO mice to indicate normal kidney function. (B) Representative hematoxylin and eosin (H&E) staining images of kidney cortex from

PT-DNMT1-WT and -KO mice. Scale bar: 50 µm. Quantitative data in (A) are expressed as mean

± SD.

Sub-aim 1b: To establish and characterize PT-DNMT3a-KO mouse model.

In addition, we also created PT-DNMT3a-KO mice by crossing DNMT3aflox/flox mice with PEPCK-Cre mice (See Figure 8 for the breeding protocol). DNMT3a mainly accounts for de novo DNA methylation and mice with deficiency in DNMT3a are not viable.88 Cre-mediated rearrangement deletes exon 19, which encodes the catalytic domain containing highly conserved PC (Pro-Cys) motif (Figure 12A). To verify the deletion of

DNMT3a in proximal tubular cells of PT-DNMT3a-KO mice, we first performed PCR to confirm the Cre-mediated rearrangement of DNMT3a alleles in proximal tubules of PT-

Figure 12: Characterization of PT-DNMT3a-KO mice.

(A) Schematic representation of the engineered locus of DNMT3a before and after PEPCK-Cre-

mediated deletion and the positions of the PCR primers. (B) PCR-based genotyping. Genomic

DNA was extracted from kidney cortex for PCR amplification to detect the recombinant allele of

DNMT3a and PEPCK-Cre allele. (C) Immunoblot analysis of renal cortical lysate confirming

DNMT3a suppression in PT-DNMT3a-KO tissues. Cyclophilin B was used as protein loading

control.

DNMT3a-KO mice (Figure 12B). Recombinant DNMT3a alleles (350 bp) were detected only in kidney cortex of PT-DNMT3a-KO mice. Immunoblot analysis showed the

DNMT3a protein levels were decreased in kidney cortical tissues of PT-DNMT3a-KO mice compared with those of PT-DNMT3a-WT mice (Figure 12C). And no major phenotypes were observed in PT-DNMT3a-KO mice under normal conditions. PT-

DNMT3a-KO mice showed normal kidney function (Figure 13A) and renal histology

(Figure 13B).

Figure 13: Renal function and histology in PT-DNMT3a-KO mice.

(A) BUN of PT-DNMT3a-WT and -KO mice to indicate normal kidney function. (B)

Representative H&E staining images of kidney cortex from PT-DNMT3a-WT and -KO mice.

Scale bar: 50 µm. Quantitative data in (A) are expressed as mean ± SD.

Sub-aim 1c: To establish and characterize PT-DNMT1/3a-DK mouse model.

Furthermore, to better understand the role of DNA methylation in AKI, double knockout or triple knockout DNMTs mouse model is needed. In this study, we generated

PT-DNMT1/3a-DK mice. DNMT1flox/flox DNMT3a+/+ XY mice were bred with DNMT1+/+

DNMT3aflox/flox XCreX mice. As shown in Figure 9, after 3 generations, we obtained PT-

DNMT1/3a-DK mice. DNMT1 and DNMT3a deletion in proximal tubular cells in PT-

DNMT1/3a-DK mice was confirmed by PCR (Figure 14A) and immunoblot (Figure 14B).

Similar to PT-DNMT1-KO and PT-DNMT3a-KO mice, at the age of 8-14 weeks, PT-

DNMT1/3a-DK mice did not show noticeable defects in kidney function or histology

(Figure 14C and D).

Figure 14: Characterization of PT-DNMT1/3a-DK mouse model.

(A) PCR-based genotyping. Genomic DNA was extracted from kidney cortex for PCR

amplification to detect the recombinant alleles of DNMT1 and DNMT3a, and PEPCK-Cre allele.

(B) Immunoblot analysis of renal cortical lysate confirming DNMT1 and DNMT3a decrease in

PT-DNMT1/3a-DK kidney tissues. GAPDH was used as protein loading control. (C) BUN of PT-

DNMT1/3a-WT and -DK mice to indicate normal kidney function. (B) Representative H&E

staining images of kidney cortex from PT-DNMT1/3a-WT and -DK mice. Scale bar: 50 µm.

Quantitative data in (C) are expressed as mean ± SD.

Specific Aim 2: To determine the role of DNA methylation in cisplatin-induced AKI using DNA methylation inhibitor and PT-DNMTs

KO mouse models.

Before examining the role of DNA methylation in cisplatin-induced AKI, we raised the question: what are the global DNA methylation changes in cisplatin-induced AKI? To determine the genome-wide DNA methylation changes in cisplatin-induced AKI, RRBS was utilized to identify DNA methylation changes at single-base resolution. Genomic DNA was isolated from kidney cortex and outer medulla of control and cisplatin-treated mice and then subjected to RRBS analysis. Totally about 1.5 and 1.9 million of CpG sites were analyzed in the control and cisplatin-treated kidney samples, respectively. Cisplatin treatment induced evident changes in DNA methylation as shown by the heat map (Figure

15A). We identified 215 DMRs between control and cisplatin-treated kidneys that showed significant (>±0.25) differences in methylation (Figure 15B), including 102 hypermethylated DMRs and 113 hypomethylated DMRs. In DMR genome-wide distribution analysis, 83% of the DMRs were in the intergenic region, intron, and coding

DNA sequence (CD), and only 7% of the DMRs were at the 5’ end and 5’ untranslated region (5’UTR) promoter or regulatory region of protein-coding genes (Figure 15C).

Figure 15: Genome-wide analysis of DNA methylation changes in cisplatin-induced AKI.

C57BL/6 mice were injected with cisplatin (cis3d) or saline (control) to collect kidney cortex and outer medulla at day 3 for the isolation of genomic DNA, which was subjected to genome-wide

DNA methylation analysis via reduced representation bisulfite sequencing (RRBS). 200 bp windows were used to identify differentially methylated regions (DMRs). DMRs were identified with a methylation difference over 0.25. (A) Representative heat map. Red color indicates high levels of DNA methylation and blue color indicates low levels of methylation. (B) The number of

DMRs identified in cisplatin-induced AKI. (C) Genome-wide distribution of the DMRs. The results were from 2 pairs of littermate mice. CD, coding DNA sequence.

This result is consistent with the recent study of genome-wide DNA methylation in CKD106 that also showed the distribution of the majority of the DMRs in intronic and transcription termination regions and 3’UTRs, but not in gene promoter regions. DNA methylation in gene promoter regions is considered to be critical for transcriptional regulation. Our analysis identified 15 genes with DMRs at their 5’end and 5’UTR regulatory regions (Table

4), and the functional analysis indicated these genes are involved in gene transcription, cell cycle, and apoptosis (Table 5). Taken together, these results indicate that cisplatin treatment induces aberrant changes in DNA methylation in kidney tissues and DNA methylation may play a role in cisplatin-induced AKI by regulating specific genes.

Table 4: Genes with DMR at 5’ end or 5’UTR in cisplatin-induced AKI

- Irf8 - Hdac5 - Nrip1 - Cuedc1 - Foxi1 - Foxk1 - Dhdh - Tdrd1 - Gm16617 - 1700012B15Rik - 2700086A05Rik + Spry4 + Sycp2 + Fbxo46 + Mir219-2 * “-” indicates hypomethylation, “+” indicates hypermethylation

Table 5: Functions of genes with DMR at 5’ end or 5’UTR in cisplatin-induced AKI

Transcription Cell Cycle Apoptosis MicroRNA Hyper Sycp2 Sycp2 Mir219-2 Foxk1 Foxk1 Irf8 Foxi1 Tdrd1 Hypo Nrip1 Hdac5 Irf8

In Specific Aim 2, we determined the role of DNA methylation in cisplatin-induced

AKI using DNA methylation inhibitor and PT-DNMTs KO mouse models.

Sub-aim 2a: To determine the role of DNA methylation in cisplatin-induced AKI in vitro using DNA methylation inhibitor 5-aza-2’-deoxycytidine (5-aza).

To determine whether DNA methylation plays a pathogenic role in cisplatin- induced AKI, we first tested the effects of 5-aza-2’-deoxycytidine (5-aza, a pharmacological DNA methylation inhibitor) on cisplatin-induced apoptosis in RPTC cells, a rat renal proximal tubular cell line. As a cytidine analog, 5-aza may incorporate into DNA and irreversibly bind to DNMTs resulting in inactivation of DNMTs and inhibition of DNA methylation. In addition, 5-aza can induce the proteasomal degradation of free (non-chromatin bound) DNMTs, causing the depletion of DNMTs nuclear pool, which also contributes to the hypomethylation induced by 5-aza.132-134 Pretreatment with

5-aza could diminish DNMT1 and DNMT3a expression (Figure 16A) without noticeable

Figure 16: 5-aza increases cisplatin-induced apoptosis in RPTC cells.

RPTC cells were treated with 20 μM cisplatin in the absence or presence of 1 μM 5-aza

pretreatment. (A) Immunoblot. After 16 hours cisplatin treatment, whole cell lysate was collected

for immunoblot analysis of indicated proteins. Cyclophilin B was used as protein loading control.

(B) Cell morphology. After 16 hours cisplatin treatment, cells were stained with Hoechst 33342 to

record cellular and nuclear morphology by phase contrast and fluorescence microscopy,

respectively. Scale bar: 200 µm. (C) Percentage of cell apoptosis. The cells with typical apoptotic

morphology were counted to determine the % of apoptosis. Data in (C) are presented as mean ±

SD; n = 4. *P < 0.05, significantly different from control; #P < 0.05, significantly different from

the cisplatin-only group. toxicity in RPTC cells (Figures 16B and C). Importantly, 5-aza significantly increased apoptosis during cisplatin treatment (Figure 16B and C). Typical apoptotic cells showed

cellular shrinkage and formation of apoptotic bodies viewed by phase contrast microscopy

(Figure 16B, upper panel) and nuclear condensation and fragmentation viewed by Hoechst

33342 nucleus staining (Figure 16B, bottom panel). Quantification of apoptotic cells showed that cisplatin induced 24% apoptosis, which was increased to 55% by 5-aza pretreatment (Figure 16C). The morphological results were verified by immunoblot analysis of the cleavage of caspase 3 and PARP, biochemical hallmarks of apoptosis, showing the enhancing effect of 5-aza (Figure 16A). Collectively, these results demonstrate a sensitizing effect of 5-aza on cisplatin-induced apoptosis, suggesting a cytoprotective role of DNA methylation in renal tubular cells.

Sub-aim 2b: To determine the role of DNA methylation in cisplatin-induced AKI using PT-DNMT1-KO mice.

To further delineate the role of DNA methylation in cisplatin-induced AKI, we compared cisplatin-induced AKI in PT-DNMT1-WT and -KO littermates in response to a single dosage of 30 mg/kg cisplatin. At day 4 of cisplatin injection, PT-DNMT1-WT mice showed a serum creatinine level of 1.9 ± 0.4 mg/dl, while PT-DNMT1-KO mice had 2.5 ±

0.5 mg/dl serum creatinine that was higher than that of wild-type mice with statistical significance (Figure 17A). Cisplatin-induced tubular damage in both PT-DNMT1-WT and

-KO mice were characterized by loss of brush border, tubular dilation, cast formation, and cell lysis in proximal tubules. Notably, consistent with serum creatinine results, significantly more tubular damage was shown in PT-DNMT1-KO mice (Figure 17B).

Histo-pathological grading showed a tubular damage score of 3.7 for PT-DNMT1-KO mice, which was higher than that of PT-DNMT1-WT mice (Figure 17C).

Figure 17: Cisplatin-induced AKI is aggravated in PT-DNMT1-KO mice. Figure 17: Cisplatin-induced AKI is aggravated in PT-DNMT1-KO mice. PT-DNMT1-WT (n=10) and -KO (n=10) littermate mice were injected with 30 mg/kg cisplatin or PT-DNMT1-WT (n=10) and -KO (n=10) littermate mice were injected with 30 mg/kg cisplatin or saline as control. Kidney tissues and blood samples were collected at day 4 after cisplatin injection saline as control. Kidney tissues and blood samples were collected at day 4 after cisplatin injection for analysis. (A) Serum creatinine. (B) Representative H&E images of kidney cortex from PT- for analysis. (A) Serum creatinine. (B) Representative H&E images of kidney cortex from PT- DNMT1-WT and -KO mice. Scale bar: 50 µm. (C) Kidney tissue damage score in cisplatin-treated DNMT1-WT and -KO mice. Scale bar: 50 µm. (C) Kidney tissue damage score in cisplatin-treated PT-DNMT1-WT and -KO mice. Quantitative data in (A) and (C) are expressed as mean ± SD. *P PT-DNMT1-WT and -KO mice. Quantitative data in (A) and (C) are expressed as mean ± SD. *P < 0.05, significantly different from PT-DNMT1-WT mice.

We also analyzed apoptosis in renal tissues by immunohistochemical staining of cleaved caspase 3. As shown in Figure 18A and B, PT-DNMT1-KO kidney tissues had a significantly higher level of cleaved caspase 3 staining than PT-DNMT1-WT kidney

tissues. Altogether, these results suggest that ablation of DNMT1 from kidney proximal tubules leads to more severe AKI following cisplatin exposure.

Figure 18: PT-DNMT1-KO mice induces more renal apoptosis after cisplatin treatment.

PT-DNMT1-WT and -KO littermate mice were injected with 30 mg/kg cisplatin or saline as

control. (A) Representative images of cleaved caspase 3 immunohistochemical staining. Scale bar:

50 µm. (B) Quantitative analysis of cleaved caspase 3 immunohistochemical staining. Quantitative

data in (B) are expressed as mean ± SD. *P < 0.05, significantly different from PT-DNMT1-WT

mice.

To further confirm the in vivo results, we examined cisplatin-induced apoptosis using primary proximal tubular cells isolated from PT-DNMT1-WT or -KO mice. As shown in Figure 19A, both DNMT1-WT and -KO proximal tubular cells showed normal morphology with minimal apoptosis under control condition. Cisplatin treatment induced 55

apoptosis in both groups of cells, but obviously more apoptosis was induced in DNMT1-

KO proximal tubular cells (Figure 19A). In quantification, cisplatin induced ~18% apoptosis in DNMT1-WT proximal tubular cells, whereas ~40% apoptosis in DNMT1-KO proximal tubular cells (Figure 19B). Moreover, immunoblot analysis detected much higher caspase 3 cleavage in DNMT1-KO proximal tubular cells after cisplatin treatment (Figure

19C). And repression of DNMT1 in DNMT1-KO proximal tubular cells was confirmed by immunoblot (Figure 19C). Thus, cisplatin induces higher toxicity in PT-DNMT1-KO primary proximal tubular cells than wild-type cells, an observation that is consistent with the above in vivo study.

Figure 19: DNMT1-KO primary proximal tubular cells are more sensitive to cisplatin- induced apoptosis.

Primary proximal tubular cells were isolated from PT-DNMT1-WT or -KO mice. Then primary cells were treated with 50 μM cisplatin for 24 hours. (A) Cell morphology. After 24h cisplatin treatment, cells were stained with Hoechst 33342 to record cellular and nuclear morphology by phase contrast and fluorescence microscopy, respectively. (B) Percentage of apoptosis. The cells with typical apoptotic morphology were counted to determine the % of apoptosis. Scale bar: 200

µm. (C) Immunoblot analysis of caspase 3 activation. After 24h cisplatin treatment, whole cell lysate was collected for immunoblot analysis of DNMT1, caspase 3, and Cyclophilin B as protein loading control. Caspase 3 activation is indicated by the appearance of cleaved (active) fragment of caspase 3. Quantitative data in (B) are presented as mean ± SD; n = 3. *P < 0.05, significantly different from control DNMT1-WT cells; #P < 0.05, significantly different from the cisplatin treated DNMT1-WT cells.

Sub-aim 2c: To determine the role of DNA methylation in cisplatin-induced AKI using PT-DNMT3a-KO mice.

Next, we determined the role of DNA methylation in cisplatin-induced AKI using

PT-DNMT3a-KO mice. After injection with cisplatin, we monitored BUN every day for 4 days. Cisplatin induced severe loss of renal function with significantly increased BUN level in both PT-DNMT3a-WT and -KO mice. However, we did not find any appreciable difference between PT-DNMT3a-WT and -KO mice at any time point (Figure 20A). Serum creatinine levels at day 4 of cisplatin treatment appeared to be lower in PT-DNMT3a-KO mice, but the difference was not statistically significant (Figure 20B). Consistently, histology analysis revealed similar tissue damage in PT-DNMT3a-WT and -KO mice

(Figure 20C), with a tubular damage score of 3.3 in both groups (Figure 20D). Thus, deletion of DNMT3a from proximal tubules does not have any significant effect on cisplatin-induced AKI.

Figure 20: PT-DNMT3a-KO mice show similar AKI following cisplatin treatment as PT-

DNMT3a-WT mice.

PT-DNMT3a-WT (n=4) and -KO (n=5) littermate mice were injected with 30 mg/kg cisplatin or

saline as control. (A) BUN of PT-DNMT3a-WT and -KO mice at indicated time point after

cisplatin injection. (B) Serum creatinine of mice treated with cisplatin at day 4. (C) Representative

H&E images of kidney cortex from PT-DNMT3a-WT and -KO mice. Scale bar: 50 µm. (D) Tissue

damage score in PT-DNMT3a-WT and -KO kidneys at day 4 after cisplatin injection. Quantitative

data in (A), (B) and (D) are expressed as mean ± SD.

Sub-aim 2d: To determine the role of DNA methylation in cisplatin-induced AKI using PT-DNMT1/3a-DK mice.

We then took advantage of PT-DNMT1/3a-DK mice. Initially, we tried 30 mg/kg cisplatin in PT-DNMT1/3a mice. However, 3 out of 6 PT-DNMT1/3a-WT mice died at day 4 after cisplatin injection, indicating PT-DNMT1/3a mice are more sensitive to cisplatin-induced nephrotoxicity, which may be caused by the different backgrounds of the mice. DNMT1flox/flox mice have a mixed genetic background of Balb/C and 129Sv/J, and

DNMT3aflox/flox mice have the C57BL/6 background. Therefore, we did pilot studies to determine the optimal conditions for this strain. We found that at day 3 after 20 mg/kg cisplatin injection, cisplatin induced moderate renal failure in PT-DNMT1/3a-WT mice with BUN and serum creatinine levels increased to 203.1 ± 59.1 mg/dl and 0.93 ± 0.23 mg/dl, respectively (Figure 21A and B). In contrast, littermate PT-DNMT1/3a-DK mice had significantly lower levels of BUN (132.8 ± 45.3 mg/dl) (Figure 21A) and serum creatinine (0.56 ± 0.12 mg/dl) (Figure 21B). Consistently, the renal histological analysis revealed that, compared with PT-DNMT1/3a-WT mice, cisplatin induced significantly less

Figure 21: Cisplatin-induced AKI is attenuated in PT-DNMT1/3a-DK mice.

PT-DNMT1/3a-WT (n=16) and -DK (n=12) littermate mice were injected with 20 mg/kg cisplatin

or saline as control. Kidney tissues and blood samples were collected at day 3 after cisplatin

injection for analysis. (A) BUN of mice treated with cisplatin at day 3. (B) Serum creatinine of

mice treated with cisplatin at day 3. (C) Representative H&E images of kidney cortex from PT-

DNMT1/3a-WT and -DK mice. Scale bar: 50 µm. (D) Percentage of tissue damage in PT-

DNMT1/3a-WT and -DK kidneys at day 3 after cisplatin injection. Quantitative data in (A), (B)

and (D) are expressed as mean ± SD. *P < 0.05, significantly different from PT-DNMT1/3a-WT

mice. tissue damage in PT-DNMT1/3a-DK mice (Figure 21C). Quantification of injured tubules suggested that cisplatin induced ~28% tissue damage in PT-DNMT1/3a-WT mice, whereas

~13% tissue damage in PT-DNMT1/3a-DK mice (Figure 21D). Further, cell apoptosis was 61

determined by TUNEL. In control condition, apoptotic cells were rarely detected in both

PT-DNMT1/3a-WT and -DK mice (Figure 22A). However, after cisplatin treatment, there were approximately 260 ± 33 apoptotic cells/mm2 in cortical tissues of PT-DNMT1/3a-

WT mice, which was markedly reduced to 159 ± 35 cells/mm2 in cortical tissues of PT-

DNMT1/3a-DK mice (Figure 22B). Therefore, these results indicate cisplatin-induced AKI is ameliorated in PT-DNMT1/3a-DK mice.

Figure 22: Cisplatin-induced renal apoptosis is suppressed in PT-DNMT1/3a-DK mice.

PT-DNMT1/3a-WT and -DK littermate mice were injected with 30 mg/kg cisplatin or saline as

control. (A) Representative images of TUNEL staining. Scale bar: 50 µm. (B) Quantification of

2 TUNEL-positive cells per mm . Quantitative data in (B) are expressed as mean ± SD. *P < 0.05,

significantly different from PT-DNMT1/3a-WT mice.

To complement the in vivo study, we isolated primary proximal tubular cells from

PT-DNM1/3a-WT or -DK mice and examined cisplatin-induced apoptosis. Under control condition, DNMT1/3a-DK proximal tubular cells showed similarly normal morphology with minimal apoptosis as DNMT1/3a-WT proximal tubular cells (Figure 23A).

Immunoblot analysis verified the suppression of both DNMT1 and DNMT3a in

DNMT1/3a-DK proximal tubular cells (Figure 23B). 24 hours after cisplatin treatment,

both groups of cells showed obvious apoptosis indicated by cellular and nuclear

Figure 23: DNMT1/3a-DK primary proximal tubular cells are more resistant to cisplatin-

induced apoptosis.

Primary proximal tubular cells were isolated from PT-DNMT1/3a-WT or -DK mice. Then primary

cells were treated with 50 μM cisplatin for 24 hours. (A) Cell morphology. (B) Immunoblot

analysis of DNMT1 and DNMT3a to verify the suppression of DNMT1 and DNMT3a in

DNMT1/3a-DK proximal tubular cells. β-actin was used as protein loading control. (C) Percentage

of apoptosis. The cells with typical apoptotic morphology were counted to determine the % of

apoptosis. Scale bar: 400 µm. (D) Immunoblot analysis of caspase 3 activation. β-actin was used

as protein loading control. Caspase 3 activation is indicated by the appearance of cleaved (active)

fragment of caspase 3. Data in (C) are presented as mean ± SD; n = 3. *P < 0.05, significantly

different from DNMT1/3a-WT cells.

morphology (Figure 23A). Importantly, less apoptosis was induced in DNMT1/3a-DK proximal tubular cells compared with DNMT1/3a-WT proximal tubular cells (Figure 23A and C). Furthermore, the morphology result was confirmed by immunoblot analysis of cleaved caspase 3. As shown in Figure 23D, compared with DNMT1/3a-WT proximal tubular cells, less cleaved caspase 3 was detected in DNMT1/3a-DK proximal tubular cells treated with cisplatin. Thus, primary proximal tubular cells isolated from PT-DNMT1/3a-

DK mice were more resistant to cisplatin-induced apoptosis, supporting our in vivo study findings. Together with in vivo study using PT-DNMT1/3a-DK mice, these results provide the evidence that depletion of both DNMT1 and DNMT3a in mouse proximal tubular cells could protect against cisplatin-induced nephrotoxicity.

Specific Aim 3: To determine the role of DNA methylation in ischemic AKI using DNA methylation inhibitor and PT-DNMTs KO mouse models.

Renal ischemia-reperfusion is another common cause of AKI. Using RRBS, we also determined the genome-wide DNA methylation changes in ischemic AKI. Totally about 1.5 and 1.4 million of CpG sites were analyzed in the sham control and ischemia- reperfusion kidney samples, respectively. Compared with sham control, kidney tissues after 30 minutes of renal ischemia and 48 hours of reperfusion showed aberrant DNA methylation changes (Figure 24A), resulting in the identification of 235 DMRs in 425 genes with 129 hypermethylated DMRs and 106 hypomethylated DMRs (Figure 24B).

Similar to cisplatin-induced AKI model, most of the DMRs were in the intergenic region, intron and CD, and only a small part (10%) of the DMRs were at the 5’ end and 5’UTR

promoter or regulatory region of protein-coding genes (Figure 24C). Totally, 24 genes were identified containing DMRs at the 5’ end and 5’UTR promoter or regulatory regions, which were listed in Table 6. The functions of the genes were grouped as shown in Table 7. Taken together, these indicate DNA methylation alterations occur in ischemic AKI and DNA methylation may play an important role in ischemic AKI by regulating specific gene expression.

Figure 24: Genome-wide analysis of DNA methylation changes in ischemic AKI.

C57BL/6 mice were subjected to ischemia 30 minutes reperfusion 48 hours (I30R48h) or sham.

Kidney cortex and outer medulla were collected for the isolation of genomic DNA, which was

analyzed by RRBS. 200 bp windows were used to identify DMRs. DMRs were identified with a

methylation difference over 0.25. (A) Representative heat map. Red color indicates high levels of

DNA methylation and blue color indicates low levels of methylation. (B) The number of DMRs

identified in ischemic AKI. (C) Genome-wide distribution of the DMRs. The results were from 2

separate experiments with 2 pairs of littermate mice.

Table 6: Genes with DMR at 5’ end or 5’UTR in ischemic AKI

- Irf8 - Hdac5 - Nrip1 - Gnas - Nespas - Nos3 - Tmem88 - Mir200c - Gm7120 - 1700012B15Rik - 1700123I01Rik - 2700086A05Rik - 3110070M22Rik + Bcl6 + Boll + Ehmt1 + Fnbp1 + Fkrp + Hoxc10 + ORF63 + Mir141 + BC016579 + Gm19557 + 4930524L23Rik * “-” indicates hypomethylation, “+” indicates hypermethylation

Table 7: Functions of genes with DMR at 5’ end or 5’UTR in ischemic AKI

Transcription Cell Cycle Apoptosis MicroRNA Bcl6 Boll Bcl6 Mir141 Boll Hoxc10 Hyper Ehmt1 Hoxc10 Hdac5 Irf8 Mir200c Hypo Irf8 Nos Nrip1

In Specific Aim 3, we determined the role of DNA methylation in ischemic AKI using DNA methylation inhibitor and PT-DNMTs KO mouse models.

Sub-aim 3a: To determine the role of DNA methylation in ischemic AKI in vitro using DNA methylation inhibitor 5-aza.

To gain insights into DNA methylation regulation in ischemic AKI, we first investigated the effects of 5-aza on hypoxic injury of RPTC cells. Hypoxia is a condition of decreased availability of oxygen, which is widely used to mimic the ischemia condition in vitro. For this purpose, RPTC cells were pretreated with 5-aza, then incubated in hypoxia chamber (1% oxygen) for 48 hours to induce apoptosis. As shown in Figure 25A, 5-aza only group showed similarly healthy morphology as the control group. After hypoxia incubation, it induced a significant increase of apoptosis in RPTC cells. Notably, the addition of 5-aza significantly increased the apoptosis during hypoxia incubation (Figure

25A). Quantification by cell counting showed that 5-aza increased hypoxia-induced apoptosis from ~13% to ~30% (Figure 25B). Consistently, under hypoxia treatment, cleaved caspase 3 was dramatically increased by 5-aza (Figure 25C). Altogether, these results suggest a deteriorative effect of 5-aza on hypoxic injury, indicating DNA methylation may play a cytoprotective role in ischemic AKI.

Figure 25: 5-aza enhances hypoxia-induced apoptosis in RPTC cells.

RPTC cells were treated with hypoxia for 48 hours in the absence or presence of 5-aza pretreatment. (A) Cell morphology. After 48 hours, cells were stained with Hoechst 33342 to record cellular and nuclear morphology by phase contrast and fluorescence microscopy, respectively. Scale bar: 200 µm. (B) Percentage of cell apoptosis. The cells with typical apoptotic morphology were counted to determine the % of apoptosis. (C) Immunoblot. After 48 hours, whole cell lysate was collected for immunoblot analysis of indicated proteins. Cyclophilin B was used as protein loading control. Quantitative data in (B) are presented as mean ± SD; n = 3. *P < 0.05, significantly different from control; #P < 0.05, significantly different from the hypoxia-only group.

Sub-aim 3b: To determine the role of DNA methylation in ischemic AKI using

PT-DNMT1-KO mice.

We further examined the role of DNA methylation mediated by DNMT1 in ischemic AKI using PT-DNMT1-KO mice. PT-DNMT1-WT and -KO littermates were subjected to sham or bilateral renal ischemia for 30 minutes followed by 48 hours reperfusion (I30R48h). I30R48h induced a significant induction of BUN with 227.6 ± 59.9 mg/dl in PT-DNMT1-WT mice and 183.2 ± 79.3 mg/dl in PT-DNMT1-KO mice, but with no statistical significant (Figure 26A). Additionally, PT-DNMT1-KO mice did not show noticeable difference in serum creatinine (1.29 ± 0.51 mg/dl) compared to PT-DNMT1-

WT mice (1.35 ± 0.57 mg/dl) after I30R48h (Figure 26B). Consistent with renal function analysis, renal histology showed similar cast formation, tubular necrosis, dilation and disruption in PT-DNMT1-WT and -KO mice (Figure 26C). This was further verified by quantification of tissue damage (Figure 26D). Collectively, DNMT1 deficiency in proximal tubules does not affect ischemic AKI.

Figure 26: DNMT1 deficiency in proximal tubules does not affect ischemic AKI.

PT-DNMT1-WT (n=10) and -KO (n=10) littermate mice were subjected to bilateral renal ischemia for 30 minutes followed by 48 hours reperfusion (I30R48h). Kidney tissues and blood samples were collected after 48 hours reperfusion for analysis. (A) BUN of mice treated with I30R48h. (B)

Serum creatinine of mice treated with I30R48h. (C) Representative H&E images of kidney cortex from PT-DNMT1-WT and -KO mice after I30R48h. Scale bar: 50 µm. (D) Percentage of tissue damage in PT-DNMT1-WT and -KO kidneys after I30R48h. Quantitative data in (A), (B) and (D) are expressed as mean ± SD.

Sub-aim 3c: To determine the role of DNA methylation in ischemic AKI using

PT-DNMT1/3a-DK mice.

Next, we determined the role of DNA methylation in ischemic AKI using PT-

DNMT1/3a-DK mice. PT-DNMT1/3a-WT and -DK littermates were challenged with

I30R48h. After 48 hours reperfusion, ischemic AKI induced in PT-DNMT1/3a-WT and -

DK mice showed no obvious difference. PT-DNMT1/3a-WT and -DK mice undergone

I30R48h had similar renal function loss indicated by the levels of BUN (Figure 27A) and serum creatinine (Figure 27B). Also, as demonstrated in Figure 27C and D, tissue damage in PT-DNMT1/3a-WT and -DK mice was similar. Thus, it is concluded that ablation both

DNMT1 and DNMT3a in proximal tubular cells has no significant effects on renal ischemia-reperfusion-induced injury.

Figure 27: Ablation both DNMT1 and DNMT3a in proximal tubular cells has no significant effects on renal ischemia-reperfusion-induced injury.

PT-DNMT1/3a-WT (n=7) and -DK (n=7) littermate mice were subjected to I30R48h. Kidney tissues and blood samples were collected after 48 hours reperfusion for analysis. (A) BUN of mice treated with I30R48h. (B) Serum creatinine of mice treated with I30R48h. (C) Representative

H&E images of kidney cortex from PT-DNMT1/3a-WT and -DK mice after I30R48h. Scale bar:

50 µm. (D) Percentage of tissue damage in PT-DNMT1/3a-WT and -DK kidneys after I30R48h.

Quantitative data in (A), (B) and (D) are expressed as mean ± SD.

After injury, tubular adaptive repair occurs in an attempt to restore tubular epithelial integrity and function, but incomplete or maladaptive repair leads to renal fibrosis, contributing to the progression of CKD. Recent studies have shed important light on the role of DNA methylation in renal fibrosis.106, 113, 135-137 We wondered if PT-DNMT1/3a-

DK mice could affect maladaptive repair after ischemic injury. To answer this question, we used a mouse model of unilateral ischemia-reperfusion without contralateral nephrectomy, which is a robust model to study renal fibrosis.138 PT-DNMT1/3a-WT and -

DK littermates were undergone 30 minutes unilateral ischemia followed by 2 weeks reperfusion (UI30R2w). Interestingly, as shown in Figure 28A, UI30R2w resulted in a markedly increase in DNMT1 and DNMT3a. We then detected the expression of several fibrotic hallmark proteins by immunoblot (Figure 28A). After UI30R2w, there was a dramatic upregulation of α-smooth muscle actin (α-SMA), a fibroblast activation marker, in PT-DNMT1/3a-WT mice. However, this upregulation was reduced in PT-DNMT1/3a-

DK mice. Similarly, compared to PT-DNMT1/3a-WT mice, UI30R2w-induced vimentin, another fibroblast marker, was also suppressed in PT-DNMT1/3a-DK mice. Importantly, fibronectin, an extracellular matrix protein, was markedly accumulated in kidney tissues of PT-DNMT1/3a-WT mice after UI30R2w, which was repressed in PT-DNMT1/3a-DK mice (Figure 28A). These results were confirmed by densitometry analysis of the immunoblot signals (Figure 28B, C, and D).

Figure 28: The expressions of fibrotic markers induced by unilateral renal ischemia-

reperfusion are suppressed in PT-DNMT1/3a-DK mice.

PT-DNMT1/3a-WT (n=7) and -KO (n=6) littermate mice were subjected to unilateral renal

ischemia 30 minutes reperfusion 2 weeks (UI30R2w) or sham control. Kidney tissues were

collected after 2 weeks reperfusion for analysis. (A) Immunoblot analysis of indicated proteins.

GAPDH was used as protein loading control. (B), (C) and (D) Densitometry analysis of α-SMA,

vimentin and fibronectin, respectively. After normalization with GAPDH, the protein signals of

PT-DNMT1/3a-WT sham control group were arbitrarily set as 1, and the signals of other groups

were normalized with it to calculate fold changes. Quantitative data in (B), (C) and (D) are

expressed as mean ± SD. *P < 0.05, significantly different from PT-DNMT1/3a-WT sham control

group; #P < 0.05, significantly different from PT-DNMT1/3a-WT UI30R2w group.

We further evaluated collagen deposition by Masson’s trichrome staining (Figure

29A). In sham control animals, Masson’s trichrome staining was rarely detected. In

contrast, a large increase in collagen deposition was shown after UI30R2w indicated by

Masson’s trichrome staining. Notably, kidney tissues from PT-DNMT1/3a-DK mice exhibited much lower levels of collagen deposition compared to PT-DNMT1/3a-WT mice.

Consistently, quantification of Masson’s trichrome-positive areas showed UI30R2w induced 21.3% renal fibrosis in PT-DNMT1/3a-WT mice, which was reduced to 15.5% in

PT-DNMT1/3a-DK mice (Figure 29B). Taken together, these results indicate that depletion of both DNMT1 and DNMT3a in proximal tubular cells attenuates renal fibroblast activation and interstitial renal fibrosis.

Figure 29: PT-DNMT1/3a-DK mice exhibit much lower levels of collagen deposition compared to PT-DNMT1/3a-WT mice after UI30R2w.

PT-DNMT1/3a-WT and -DK littermate mice were subjected to UI30R2w or sham control. Kidney tissues were collected after 2 weeks reperfusion for analysis. (A) Representative Masson’s trichrome staining images. Scale bar: 50 mm. (B) Quantification of Masson’s trichrome staining.

Quantitative data in (A) and (B) are expressed as mean ± SD. *P < 0.05, significantly different from PT-DNMT1/3a-WT UI30R2w group.

Specific Aim 4: To determine the role of specific genes (Irf8 and Bcl6) with

DNA methylation changes in cell culture model of kidney tubular cell injury.

Despite the above findings, it is important to identify the role of specific genes with

DMRs at the 5’ end and 5’UTR promoter or regulatory regions in AKI. From the genome- wide DNA methylation analysis, we identified a list of genes with DMRs at the 5’ end and

5’UTR promoter or regulatory regions in cisplatin-induced AKI (Table 4) and ischemic

AKI (Table 6). Real-time PCR was performed to determine if the expressions of these genes were correlated with their methylation changes. We selected several genes in each model, which are functionally relevant and with high DNA methylation difference. In cisplatin-induced AKI, using mRNAs from the same animals for RRBS, we analyzed

Sycp2, Cuedc1 and Spry4 which were hypermethylated genes, and Hdac5, Irf8 and Nrip1 that were hypomethylated genes. As shown in Figure 30, only the expression of Irf8 showed the correlation with its DNA methylation change. Hypomethylation in Irf8

Figure 30: Real-time PCR analysis of selected hypermethylated and hypomethylated genes

in cisplatin-induced AKI.

Total mRNA was extracted from the same mice used for RRBS. Then real-time PCR was

performed to analyze Sycp2, Cuedc1 and Spry4 which were hypermethylated genes (A), and

Hdac5, Irf8 and Nrip1 that were hypomethylated genes (B). β-actin was used as internal control. increased its transcription. In ischemic AKI, hypermethylated genes including Hoxc10 and

Bcl6, and hypomethylated genes such as Hdac5, Irf8 and Nrip1 were examined (Figure

31). Hoxc10, Bcl6 and Hdac5 showed significantly differential expression after renal ischemia- reperfusion. The expressions of both Hoxc10 and Bcl6 were inversely associated with DNA methylation status, but not Hdac5. Altogether, these results indicate Irf8 in

cisplatin-induced AKI, and Hoxc10 and Bcl6 in ischemic AKI may be regulated by DNA methylation and play a role in the pathogenesis of AKI.

Figure 31: Real-time PCR analysis of selected hypermethylated and hypomethylated genes

in ischemic AKI.

Male mice were subjected to I30R48h or sham. After 48 hours reperfusion, kidneys were collect.

Total mRNA was extracted from the kidney cortical tissues. Then real-time PCR was performed

to analyze Hoxc10 and Bcl6 which were hypermethylated genes (A), and Hdac5, Irf8 and Nrip1

that were hypomethylated genes (B). Quantitative data in (A) and (B) are expressed as mean ± SD;

n=4. *P < 0.05, significantly different from sham group.

In Specific Aim 4, we determined the role of Irf8 and Bcl6 with methylation changes in cell culture model of kidney tubular cell injury.

Sub-aim 4a: To determine the role of Irf8 with DNA methylation changes in cisplatin-induced AKI.

Irf8 (interferon regulatory factor 8) stood out as a good candidate gene for further study because Irf8 is a pro-apoptotic gene that is regulated by DNA methylation. As shown in Figure 32A, Irf8 contained a DMR at 5’ UTR region with hypomethylation when mice were challenged with cisplatin. We further confirmed our real-time PCR results using 7 different sets of samples, concluding that, along with Irf8 hypomethylation, cisplatin induced Irf8 expression (Figure 32B). Immunoblot further verified Irf8 induction at the protein level in cisplatin treated mouse kidneys (Figure 32C).

Figure 32: Irf8 is hypomethylated and induced in cisplatin-induced AKI.

Male mice were injected with 30 mg/kg cisplatin or saline as control. Kidney tissues were collected at day 3 after cisplatin injection for analysis. (A) The RRBS result of Irf8 gene methylation changes shown by UCSC genome browser screenshot. The hypomethylated CpG sites following cisplatin- treatment are highlighted in green. (B) Real-time PCR analysis to verify Irf8 induction in kidney cortical tissues following cisplatin treatment. (C) Immunoblot analysis of Irf8 expression in renal cortical lysate. Cyclophilin B was used as protein loading control. Irf8 signals were then analyzed by densitometry (bottom). After normalization with Cylcophilin B, the protein signal of lane 1 was arbitrarily set as 1, and the signals of other lanes were normalized with lane 1 to calculate fold changes. Quantitative data in (B) are expressed as median ± interquartile range; n=7. Quantitative data in (C) are expressed as mean ± SD; n=4. *P < 0.05, significantly different from control.

To further confirm the role of DNA methylation in regulating Irf8 expression, we analyzed the effects of 5-aza on Irf8 expression in RPTC cells. As shown in Figure 33, 5- aza induced Irf8 at both mRNA and protein levels. Together, these results suggest that Irf8 is induced at least partly through hypomethylation in cisplatin AKI.

Next, to verify the pro-apoptotic role of Irf8 in renal tubular cells, we determined

Figure 33: Inhibition of DNA methylation by 5-aza increases Irf8 expression.

RPTC cells were treated with 20 μM cisplatin for 16 hours in the absence or presence of 5-aza

pretreatment. Irf8 expression was shown by real-time PCR (top) and immunoblot (bottom)

analysis. Quantitative data are expressed as mean ± SD; n=4. *P < 0.05, significantly different

from control; #P < 0.05, significantly different from the cisplatin-only group. the effects of Irf8 knockdown on cisplatin-induced apoptosis in RPTC cells. Compared with scramble sequence shRNA, Irf8 shRNA significantly suppressed Irf8 expression

(Figure 34A and B). Under control condition, both scramble cells and Irf8 knockdown cells exhibited a minimal degree of apoptotic cells (Figure 34C). After 20 hours cisplatin treatment, ~43% apoptosis was observed in scramble shRNA cells, but only 17% in Irf8 knockdown cells (Figure 34D). Consistently, silencing Irf8 significantly reduced caspase

3 cleavage (Figure 35E). Thus, loss of Irf8 expression protected RPTC cells from cisplatin- induced apoptosis, supporting an injurious role of Irf8 in renal tubular cells.

Figure 34: Silencing Irf8 suppresses cisplatin-induced apoptosis in RPTC cells.

Irf8 shRNA and scramble sequence plasmids were separately transfected into RPTC cells to select

stably transfected cells. The cells were then treated with 20 μM cisplatin for 20 hours. (A)

Immunoblot and (B) real-time PCR analysis of Irf8 to confirm the silencing of Irf8. Cyclophilin B

was used as protein loading control. (C) Cell morphology. Cells were stained with Hoechst 33342

to record cellular and nuclear morphology by phase contrast and fluorescence microscopy,

respectively. Scale bar: 200 µm. (D) Percentage of cell apoptosis. The cells with typical apoptotic

morphology were counted to determine the % of apoptosis. (E) Immunoblot analysis of caspases

3 and Cyclophilin B as protein loading control. Quantitative data in (B) and (D) are presented as

mean ± SD; n = 3. *P < 0.05, significantly different from control scramble sequence-transfected

cells. #P < 0.05, significantly different from cisplatin-treated scramble sequence-transfected cells.

$P<0.05, significantly different from control Irf8 shRNA sequence-transfected cells.

Then we wondered if the aggravated AKI in PT-DNMT1-KO mice after cisplatin treatment is due to the induction of Irf8. We examined Irf8 mRNA level by real-time PCR in PT-DNMT1-WT and -KO mice kidney tissues to determine if Irf8 expression depends on DNMT1. Although there was an induction of Irf8 after cisplatin treatment, PT-DNMT1-

KO and -WT mice showed similar levels of Irf8 in both control and cisplatin-treated conditions (Figure 35). Taken together, these results suggest that Irf8 expression during cisplatin-induced AKI is regulated by DNA methylation, but may not depend on DNMT1 alone. Furthermore, these results indicate that the exacerbated renal injury in PT-DNMT1-

KO mice after cisplatin treatment is not due to Irf8 transcription regulation but other genes regulated by DNMT1.

Figure 35: Depletion of DNMT1 in mice does not affect Irf8 expression.

PT-DNMT1-WT or -KO littermate mice were injected with 30 mg/kg cisplatin or saline as control.

Irf8 mRNA levels in kidney cortex and outer medulla were detected by real-time PCR.

Quantitative data are presented as mean ± SD; n=5. *P<0.05, significantly different from control

PT-DNMT1-WT mice; #P<0.05, significantly different from control PT-DNMT1-KO mice.

Sub-aim 4b: To determine the role of Bcl6 with DNA methylation changes in ischemic AKI.

Bcl6 (B-cell lymphoma 6) is another gene identified from the genome-wide DNA methylation analysis. As shown in Figure 36A, Bcl6 was hypermethylated after ischemia- reperfusion. Also, its transcription was inhibited, correlated well with the hypermethylation level at its 5’UTR (Figure 31A). In addition, the protein levels of Bcl6 were also suppressed in ischemic AKI (Figure 36B). Immunohistochemistry further confirmed the decrease of

Bcl6 in ischemic AKI (Figure 36C). Bcl6 is a transcriptional repressor, which is suggested to play an essential role in cell proliferation and survival. Therefore, Bcl6 is a good candidate gene for further study.

Figure 36: Bcl6 is hypermethylated and suppressed in ischemic AKI.

Male mice were subjected to I30R48h or sham. Kidney tissues were collected after 48 hours

reperfusion for analysis. (A) The RRBS result of Bcl6 gene methylation changes shown by UCSC

genome browser screenshot. (B) Immunoblot analysis of Bcl6 expression in renal cortical lysate.

Cyclophilin B was used as a protein loading control. (C) Representative images of Bcl6

immunohistochemical staining. Scale bar: 50 µm.

Next, we explored the role of Bcl6 in ischemic AKI by Bcl6 overexpression in vitro.

RPTC cells were transiently transfected with the full-length Bcl6 plasmids or empty vector plasmids. 24 hours after transfection, the cells were then incubated with hypoxia for 48 hours to induce apoptosis. 48 hours after incubation in hypoxia, Bcl6 overexpression cells showed similar levels of apoptosis as empty vector transfected cells (Figure 37A). This was confirmed by quantification of the percentage of apoptotic cells (Figure 37B). As shown in Figure 37C, immunoblot analysis verified the Bcl6 overexpression in Bcl6

overexpression cells. Taken together, Bcl6 overexpression had no effect on hypoxia- induced apoptosis.

Figure 37: Overexpression of Bcl6 does not impact hypoxia-induced apoptosis in RPTC cells.

Bcl6 or empty vector plasmids were transiently transfected into RPTC cells. 24 hours after

transfection, the cells were then treated with hypoxia for 48 hours. (A) Cell morphology. Cells

were stained with Hoechst 33342 to record cellular and nuclear morphology by phase contrast and

fluorescence microscopy, respectively. Scale bar: 200 µm. (B) Percentage of cell apoptosis. The

cells with typical apoptotic morphology were counted to determine the % of apoptosis. (C)

Immunoblot analysis of Bcl6 and β-actin as protein loading control. Quantitative data in (B) are

presented as mean ± SD; n = 3. *P < 0.05, significantly different from control empty vector

transfected cells. E, empty vector.

IV. DISCUSSION

DNA methylation is a major form of epigenetic modification that regulates gene expression in mammalian development and disease pathogenesis. In kidneys, DNA methylation changes have been implicated in kidney fibrosis, diabetic nephropathy, and

CKD.106-110, 113, 135 However, the information of DNA methylation in AKI is very limited.

In this regard, Mehta and colleagues120 demonstrated the evidence of hypermethylation of

CALCA gene promoter in the urine samples from kidney transplant patients, suggesting its potential as a biomarker of ischemic AKI during transplantation. Pratt et al.115 reported that a specific cytosine in C3 promoter was demethylated in rat kidney during renal ischemia- reperfusion. More recently, Huang et al.117 showed a decrease in global 5hmC during ischemic AKI in mice, whereas global 5mC remained unchanged. It is noteworthy that these studies are mainly superficial and the role of DNA methylation in the pathogenesis of AKI remains elusive and the regulation of specific pathogenic genes by DNA methylation in AKI is largely unknown.

In the current study, we have provided the first analysis of genome-wide DNA methylation changes in both cisplatin-induced and ischemic AKI. We further determined the pathological role of DNA methylation in these two AKI models. 5-aza augmented cisplatin- and hypoxia-induced apoptosis in RPTC cells and genetic ablation of DNMT1,

DNMT3a or both from proximal tubules played different roles in these two AKI models.

Furthermore, deficiency in both DNMT1 and DNMT3a was found to be renoprotective in the model of unilateral renal ischemia-reperfusion. Finally, we identified Irf8 that was hypomethylated and induced during cisplatin treatment to contribute to tubular cell injury and death induced by cisplatin. However, Bcl6 did not play a major role in hypoxic injury in RPTC cells, although it was hypermethylated and downregulated in the mouse model of ischemic AKI.

Alterations of DNA methylation have been shown to be related to various human diseases. Genome-wide DNA methylation analysis is needed to provide better understanding of the disease states and may pinpoint the hyper- or hypo-methylation of specific genes in diseases, uncovering novel therapeutic targets. Recently, several advanced techniques have been developed on a genome-wide scale based on microarray or deep sequencing, making it possible to quantitatively interrogate DNA methylation sites across the genome at single-nucleotide resolution or even a single-cell resolution. Illumina infinium Methylation BeadChip arrays are microarray-based methods, which are cost- effective and require a little amount of input DNA. Infinium MethylationEPIC BeadChip, the most recent implementation, allows the researchers to analyze more than 850,000 methylation sites at a single-nucleotide resolution. It covers 95% of CpG islands and 99% of RefSeq genes, with higher coverage of regulatory element including enhancer regions and other content categories.139 However, the drawback is that it is only applicable to human samples. RRBS is a deep sequencing based technique, which is an efficient and high-throughput technology with single nucleotide resolution. It provides the sensitivity of whole-genome bisulfite sequencing, but is more cost-effective by only sequencing a reduced, representative sample of the whole genome, which covers 85% of CpG islands

and 60% of promoters.140 Using these techniques, several groups performed genome-wide

DNA methylation analysis in kidney diseases, including CKD106-108, 113 and diabetic nephropathy109-112. However, genome-wide DNA methylation analysis is lacking in AKI.

Previously, Huang117 and his colleagues determined the overall DNA methylation status in ischemic AKI. They found that the total amount of 5mC was not altered in ischemic AKI, but 5hmC was decreased. However, they used dot blotting and immunohistochemical staining of 5mC, which is semi-quantitative and not able to reveal detailed methylation patterns. In the present study, we conducted the first genome-wide DNA methylation profiling in both cisplatin-induced and ischemic AKI using RRBS. The reason why we chose RRBS is that it is an efficient, high-throughput and cost-effective technology with single nucleotide resolution, and is applicable to mouse samples. As shown in the heat map

(Figure 15A and 24A), both models induced wide-spread DNA methylation alterations throughout the genome in kidney tissues. The analysis identified a number of DMRs following cisplatin treatment or renal ischemia-reperfusion (Figure 15B and Figure 24B), indicating AKI induced profound changes in DNA methylation in kidneys. Indeed, we identified 15 or 24 genes with DMRs in their promoters or promoter regulatory regions, which are likely to be regulated by DNA methylation during cisplatin treatment or renal ischemia-reperfusion, respectively. Of note, these DMRs account for only a small portion, and the DMRs in non-promoter regions are also valuable because DNA methylation in the enhancers and gene bodies may also contribute to transcription modulation.106, 141-144 The genome-wide DNA methylation analysis not only reveals the novel epigenetic changes in

AKI, but also allows us to identify new therapeutic targets.

Moreover, our study investigated the role of DNA methylation in AKI using both in vitro cell culture and in vivo mouse models. In RPTC cells, we examined the effects of

5-aza on cisplatin- and hypoxia-induced apoptosis. 5-aza is one of the most commonly used

DNA methylation inhibitors in the experimental studies. 5-aza decreased DNMT1 and

DNMT3a expression in RPTC cells (Figure 16A), indicative of its inhibitory effect.

Importantly, 5-aza markedly increased RPTC apoptosis during cisplatin treatment and hypoxia incubation (Figures 16 and Figure 25). These suggest a renoprotective role of

DNA methylation in cell culture models of AKI.

To study the role of DNA methylation in mouse, we established three conditional knockout mouse models, in which different DNMTs (DNMT1, DNMT3a or both) were specifically deleted from kidney proximal tubules. DNMT1 is the maintenance DNA methyltransferase that is the most abundant DNMT in various cell types, whereas DNMT3a and DNMT3b are de novo methyltransferases, which establish the initial DNA methylation patterns.73, 145 Global deleting DNMT1 from the genome results in embryonic lethality, while DNMT3a knockout mice are born alive but die within a few weeks.87 In this case, we generated proximal tubule-specific DNMT1 or DNMT3a-knockout mouse models

(Figure 10 and 12). We first examined cisplatin-induced AKI using these conditional knockout mouse models. Not surprisingly, PT-DNMT1-KO and PT-DNMT3a-KO mice responded differently to cisplatin-induced nephrotoxicity. PT-DNMT1-KO mice were more sensitive to cisplatin-induced AKI (Figure 17 and 18), while PT-DNMT3a-KO mice developed similar kidney injury as their wild-type littermates after cisplatin treatment

(Figure 20). These results indicate that the maintenance DNA methyltransferase DNMT1 plays a more important role in cisplatin-induced AKI than de novo DNA methyltransferase

DNMT3a. Altogether, the effect of 5-aza in RPTC cells and the effect of PT-DNMT1-KO in mice support a renoprotective role of DNA methylation in cisplatin-induced AKI.

Although compared to PT-DNMT1-WT mice, PT-DNMT1-KO mice developed significantly more injury in renal tubules following cisplatin treatment, the effect is not dramatic. This may be caused by several factors. First, DNMT1 was not ablated from all proximal tubular cells in our conditional knockout model (Figure 10D), which would lead to a partial effect on phenotype. Dr. Haase’ group125 showed mosaic expression of PEPCK-

Cre in proximal tubules (in ~70% of S1 and S2 segment and all S3 segments of proximal tubules). Second, it is possible that DNMT3a and DNMT3b may function as maintenance

DNA methyltransferases to compensate for the loss of DNMT1. Although DNMTs have been widely speculated to be classified as maintenance (DNMT1) and de novo (DNMT3a and DNMT3b) DNA methyltransferases, increasing evidence has implicated the overlap function of these two types.146-151 It has been reported that DNMT3a and DNMT3b can methylate the cytosine which is missed by DNMT1 during DNA replication. Deletion of

DNMT1 alone may lead to partial demethylation. For example, in neural progenitor cells, deletion of DNMT1 caused 30-50% demethylation.152 In HCT116 cells, deletion of

DNMT1 alone only reduced 20% of overall DNA methylation, deletion of DNMT3b alone caused 3-5% reduction, but deletion of both DNMT1 and DNMT3b increased the reduction to 95%.153 Finally, it is possible that DNMT1 ablation may not significantly block DNA methylation in some of the proximal tubular cells in our model. This is because DNMT1 is the maintenance DNMT that mainly catalyzes DNA methylation in dividing or proliferating cells. In our model, DNMT1 ablation was driven by PEPCK-Cre at ~3 weeks after birth in mice,154 a time-point when kidney development has mostly completed. Under

these conditions, the role of DNMT1 in DNA methylation may be limited due to the low rate of cell proliferation. However, upon AKI and especially during kidney repair after

AKI, there is a remarkable increase of proliferation in tubule cells.155 Thus, DNMT1 ablation may specifically affect these proliferative cells. In line with this possibility, primary tubular cells from PT-DNMT1-KO mice (that were proliferative) had more than two times increase in apoptosis during cisplatin treatment than wild-type cells (Figure 19).

To rule out the compensatory role of other DNMTs, we generated PT-DNMT1/3a-

DK mouse model (Figure 14) to study the role of DNA methylation in cisplatin-induced

AKI. Strikingly, PT-DNMT1/3a-DK showed opposite effects but not augmented effects on cisplatin-induced AKI compared with PT-DNMT1-KO. Unlike PT-DNMT1-KO mice, PT-

DNMT1/3a-DK mice were more resistant to cisplatin than PT-DNMT1/3a-WT mice

(Figure 21 and 22). Why is that PT-DNMT1/3a-DK was protective, but PT-DNMT1-KO was injurious? First, one possibility is that the overall DNA methylation change may not be the causal factor, but the DNA methylation changes in specific genes or sequences may contribute to cisplatin-induced AKI. It is possible that in PT-DNMT1/3a-DK mice, predominantly, renoprotective genes are restored the expression. Moreover, methylation of some loci requires the cooperation of different DNMTs. For example, methylation of long interspersed nuclear element (LINE-1) promoter requires both DNMT1 and one of the de novo DNMTs in mouse embryonic stem cells.147 Therefore, ablation of DNMT1 or ablation both DNMT1 and DNMT3a could affect different sets of genes. It is possible that some of the renoprotective genes regulated by DNA methylation are mediated not by DNMT1 alone, but by the cooperation of DNMT1 and DNMT3a. When both DNMT1 and DNMT3a are absent, these renoprotective genes are reactivated, protecting proximal tubular cells

from cisplatin-induced cell death. Third, DNMT1 not only has DNA methylation- dependent function, but it also plays a role in DNA damage and repair pathway that is DNA methylation-independent. When DNA damage occurs, DNMT1 was rapidly and transiently recruited to the double strand breaks and interacted with proliferating cell nuclear antigen

(PCNA), as well as the Rad9–Hus1–Rad1 (9–1–1) clamp.156, 157 Additionally, knockdown of DNMT1 induced checkpoint kinases 1 and 2 (Chk1 and -2) phosphorylation and histone

H2AX phosphorylation, which was DNA methylation-independent.158 Thus, the injurious effects in PT-DNMT1-KO mice might be due to the methylation-independent function of

DNMT1, not methylation-dependent function. However, when both DNMT1 and

DNMT3a are deleted, DNMT1 may function primarily through the methylation-dependent pathway.

Surprisingly, the DNA methylation inhibitor 5-aza in cell culture has some different

(detrimental) phenotypes in comparison to the PT-DNMT1/3a-DK mice in cisplatin- induced AKI. The discrepancy between these two models can be explained by several factors. First, 5-aza could inhibit all DNMTs for DNA methylation, but in PT-DNMT1/3a-

DK mice, we only deleted DNMT1 and DNMT3a. Other DNMTs may also play a role in cisplatin-induced AKI. Second, the in vitro model of cisplatin-induced AKI could not perfectly mimic the in vivo model. The pathogenesis of AKI in mice involves a complex interplay among tubular, microvascular, and inflammatory factors. However, the in vitro model only contains proximal tubular cells, lacking the interplay with other cell types.

Third, we used rat proximal tubular cells in vitro, but in vivo we used proximal tubular- specific KO mouse models. DNA methylation may play different roles in cisplatin-induced

AKI in different species.

In this study, we also examined ischemic AKI using these conditional knockout mouse models. However, depletion of DNMT1, or both DNMT1 and DNMT3a had no effect on ischemic AKI (Figure 26 and 27). This may indicate that overall DNA methylation change caused by the ablation of DNMTs may not contribute to the pathogenesis of ischemic AKI. This may be true because genome-wide DNA methylation analysis revealed that the numbers of hypermethylated DMRs and hypomethylated DMRs are similar, suggesting it did not induce an overall shift to DNA hypermethylation or hypomethylation, which is consistent with Huang’s study117 in ischemic AKI. These observations suggest that the DMRs in specific genes or sequences, other than the overall shift of DNA methylation status, may contribute to ischemic AKI. Another possibility is that DNMTs ablation may not significantly block DNA methylation in the proximal tubular cells in our models as we discussed previously.

Renal ischemia-reperfusion initiates the tubular epithelial cell injury, if it is severe, it leads to cell death by either necrosis or apoptosis. Activation of repair response is followed by injury, including adaptive repair and maladaptive repair. Severe or sustained injury often causes failed or maladaptive repair, contributing to renal fibrosis, ultimately

CKD.25 Recently, accumulating evidence suggests that DNA methylation contributes to the development of renal fibrosis. In 2010, Zeisberg’s group113 first demonstrated that inhibition of DNA methylation suppressed fibroblast activation and fibrogenesis in the kidney. In their study, they identified the aberrant hypermethylation of RASAL1 promoter in fibroblasts in a mouse model of renal fibrosis induced by folic acid. Administration of

5’-azacytidine inhibited RASAL1 hypermethylation and attenuated folic acid-induced fibroblast activation and renal fibrosis in mice. Moreover, RASAL1 hypermethylation was

mediated by DNMT1, and renal fibrosis was ameliorated in DNMT1+/− heterozygous mice treated with folic acid or nephrotoxic serum. Furthermore, the same group expanded their analysis to different mouse models of renal fibrosis: UUO, 5/6 nephrectomy,

COL4A3-deficient Alport mice and streptozotocin treated CD1mice. All these models showed similar results as folic acid model: RASAL1 hypermethylation and decreased

RASAL1 expression. They further found BMP7 reduced renal fibrosis in these models by reversing RASAL1 hypermethylation, which was mediated by TET3.135 In another study,136 Krüppel-like factor 4 (KLF4) was found to be hypermethylated with decreased expression in UUO model and HK2 cells treated with TGF-β1. Overexpression of KLF4 suppressed the expression of α-SMA and fibroblast-specific protein 1 (FSP-1). These results suggest DNA methylation repressed KLF4, contributing to renal fibrosis. Klotho, another gene associated with renal fibrosis, has been reported to be regulated by DNA methylation by several studies.114, 137, 159 In our study, even though DNA methylation does not play an important role in the acute injury phase of ischemic AKI, it may be involved in the repair phase, especially in the maladaptive repair. To test this, PT-DNMT1/3a-WT and

-DK littermates were undergone unilateral renal ischemia-reperfusion to induce renal fibrosis. Consistent with other studies, deficiency in both DNMT1 and DNMT3a alleviated unilateral renal ischemia-reperfusion-induced renal fibrosis, which inhibited fibroblast activation and renal fibrogenesis (Figure 28, 29).

From the genome-wide DNA methylation analysis, we identified Irf8 as a good candidate that was regulated by DNA methylation and contributed to cisplatin-induced apoptosis (Figure 32, 33 and 34). Irf8 is also known as interferon consensus sequence- binding protein (ICSBP), belonging to the IRF transcription factor family. In 2007, Irf8

was reported to be regulated by DNA methylation.160 Hypermethylation in Irf8 promoter region repressed Irf8, which was attributed to the apoptotic resistance and metastatic phenotype in tumor cells.160 After that, several studies verified the regulation of Irf8 expression by DNA methylation.161-163 In the present study, we showed hypomethylation in Irf8 at its 5’UTR during cisplatin treatment (Figure 32A), which was associated with

Irf8 induction (Figure 32B and 32C). Previously, Irf8 was thought to be exclusively expressed in myeloid and lymphoid cells,164, 165 but recently it has been found that Irf8 is expressed in a variety of cell and tissue types, such as kidney, liver, lung, and colon.163 We also detected Irf8 expression in mouse kidney and rat proximal tubular cells. Under control condition, Irf8 was expressed at a low level, but after cisplatin treatment, Irf8 was induced markedly (Figure 32B and C). Also treatment of RPTC cells with 5-aza resulted in dramatic increase of Irf8 expression (Figure 33), further confirming the regulation of Irf8 by DNA methylation in renal tubular cells. Together, these data support a role of Irf8 hypomethylation and induction in cisplatin-induced kidney injury.

Irf8 is a central mediator of IFN-γ signaling pathway.161, 166 As a transcription factor, Irf8 can act as either transcription activator or repressor. The target genes of Irf8 include Fas, Bax, FLIP, STAT1 and iNOS, suggesting an essential role of Irf8 in apoptosis.167-171 Consistently, knockdown of Irf8 attenuated cisplatin-induced apoptosis in

RPTC cells (Figure 34). Thus, hypomethylation of Irf8 during cisplatin-induced AKI may lead to the up-regulation of this pro-apoptotic transcription factor for tubular cell injury and death.

In PT-DNMT1 mice, WT and KO kidney tissues showed similar levels of Irf8 induction during cisplatin treatment (Figure 35), indicating that DNMT1 deficiency did not

significantly affect Irf8 expression under this condition. This observation also suggested that Irf8 may not be a key to the cisplatin injury sensitivity of PT-DNMT1-KO mice. A previous study also showed that knockdown of DNMT1 alone in HCT116 cells did not lead to Irf8 induction, which required the knockdown of both DNMT1 and DNMT3b.161

Thus, in the absence of DNMT1, other DNMTs or mechanisms may regulate Irf8 methylation to control its expression. Moreover, the injury sensitivity of PT-DNMT1-KO mice may involve the methylation regulation of other genes.

Bcl6, a conserved zinc finger transcription factor, usually functions as a transcription repressor. Bcl6 blockage or loss of function suppresses cell proliferation and survival.172 Under certain stress conditions, Bcl6 overexpression has been found to suppress ROS production and apoptosis.173-175 p53, an important gene associated with AKI, is also regulated by Bcl6.174, 176 Additionally, Bcl6 contributes to the transcriptional regulation of human organic anion transporter 1 (OAT1) which is an organic anion transporter in renal proximal tubular cells.177 Together, these indicate Bcl6 may be beneficial for AKI. In our study, Bcl6 was found to be hypermethylated and repressed in the mouse model of ischemic AKI (Figure 31A and Figure 36), suggesting Bcl6 is regulated by DNA methylation. However, overexpression of Bcl6 in RPTC cells did not protect tubular cells from hypoxia-induced apoptosis (Figure 37A and B). These results suggest

Bcl6 is not associated with hypoxia-induced cell apoptosis.

It should be noted that besides Irf8 and Bcl6, other genes regulated by DNA methylation may contribute significantly to the pathogenesis of AKI as well. As presented in Table 5 and Table 7, there are many other genes with DMRs at their 5’end and 5’UTR that function in the regulation of gene transcription, cell cycle, and apoptosis. For example,

Hoxc10 is a transcription factor, which belongs to homeobox (Hox) protein family. As a

Hox protein, Hoxc10 plays a pivotal role in development and cellular differentiation.

Especially, Hoxc10 has been reported to be involved in the kidney development.178-180

Moreover, Hoxc10 has been implicated in cell cycle control181 and regeneration.182 In addition, genes with DMRs in other regions may be also important for AKI. Future investigation should extend to these genes to have a comprehensive understanding of DNA methylation in AKI.

Furthermore, using different renal ischemia models may identify different DNA methylation alterations. Renal ischemia 30 minutes model is a severe AKI model for

C57BL/6 mice and most of the mice subjected to surgery die within 72 hours. By contrast, short-time ischemia model, such as 25 minutes, is a moderate AKI model, which induces initial moderate injury at early time point (e.g. 1 day), and then the injury is recovered gradually (e.g. 5, 7, 14 days and 1 month). Different set of genes may be identified with

DNA methylation changes using different renal ischemia models and different time points.

The genes identified in our I30R48h model are probably involved in the injury phase.

V. SUMMARY

This study investigated the role of DNA methylation in the pathogenesis of both cisplatin-induced AKI and ischemic AKI. Further, we determined the role of specific genes regulated by DNA methylation in both AKI models.

Specific Aim 1:

We have successfully established conditional knockout mice that are deficient in

DNMT1, DNMT3a or both exclusively in proximal tubules.

Specific Aim 2:

 DNA methylation inhibitor 5-aza increases cisplatin-induced apoptosis in RPTC cells.

 Cisplatin-induced AKI is worse in PT-DNMT1-KO mice and primary proximal tubular cells from PT-DNMT1-KO mice are more sensitive to cisplatin-induced apoptosis than WT cells.

 PT-DNMT3a-KO mice show similar AKI following cisplatin treatment as PT-

DNMT3a-WT mice.

 Cisplatin-induced AKI is attenuated in PT-DNMT1/3a-DK mice, and primary proximal tubular cells from PT-DNMT1/3a-DK mice are more resistant to cisplatin- induced apoptosis than WT cells

Specific Aim 3:

 DNA methylation inhibitor 5-aza enhances hypoxia-induced apoptosis in RPTC

100

cells.

 All of the conditional knockout mouse models exhibit similar response to ischemic

AKI as WT mice.

 Deficiency in both DNMT1 and DNMT3a alleviates unilateral renal ischemia- reperfusion-induced renal fibrosis, which inhibits fibroblast activation and renal fibrogenesis.

Specific Aim 4:

 Irf8 expression during cisplatin-induced AKI is regulated by DNA methylation, and silencing Irf8 suppresses cisplatin-induced apoptosis in RPTC cells.

 Bcl6 is hypermethylated and suppressed in ischemic AKI in mouse; however, overexpression of Bcl6 in RPTC cells does not impact hypoxia-induced apoptosis.

In conclusion, this study suggests DNA methylation plays a critical role in AKI by regulating specific genes. The findings of this study will gain significant insights into the molecular mechanism of AKI, which may lead to the discovery of novel strategies for the prevention and treatment of this devastating disease.

101

VI. REFERENCES

1. Dirkes S. Acute Kidney Injury: Not Just Acute Renal Failure Anymore? Critical Care Nurse 2011; 31: 37-50.

2. Edelstein C. Biomarkers of acute kidney injury. Adv Chronic Kidney Dis 2008; 15: 222-234.

3. Song F, Smith JF, Kimura MT, et al. Association of tissue-specific differentially methylated regions (TDMs) with differential gene expression. Proceedings of the National Academy of Sciences of the United States of America 2005; 102: 3336- 3341.

4. Weber M, Davies JJ, Wittig D, et al. Chromosome-wide and promoter-specific analyses identify sites of differential DNA methylation in normal and transformed human cells. Nature genetics 2005; 37: 853-862.

5. Rodriguez J, Frigola J, Vendrell E, et al. Chromosomal instability correlates with genome-wide DNA demethylation in human primary colorectal cancers. Cancer Research 2006; 66: 8462-8468.

6. Robertson KD. DNA methylation and human disease. Nature reviews Genetics 2005; 6: 597-610.

7. Kanai Y, Hirohashi S. Alterations of DNA methylation associated with abnormalities of DNA methyltransferases in human cancers during transition from a precancerous to a malignant state. Carcinogenesis 2007; 28: 2434-2442.

8. Stenvinkel P, Karimi M, Johansson S, et al. Impact of inflammation on epigenetic DNA methylation - a novel risk factor for cardiovascular disease? J Intern Med 2007; 261: 488-499.

9. Murphy TM, Perry AS, Lawler M. The emergence of DNA methylation as a key modulator of aberrant cell death in prostate cancer. Endocr-Relat Cancer 2008; 15: 11-25.

102

10. Weng YL, An R, Shin J, et al. DNA modifications and neurological disorders. Neurotherapeutics : the journal of the American Society for Experimental NeuroTherapeutics 2013; 10: 556-567.

11. Bergman Y, Cedar H. DNA methylation dynamics in health and disease. Nat Struct Mol Biol 2013; 20: 274-281.

12. Kulis M, Esteller M. DNA Methylation and Cancer. Adv Genet 2010; 70: 27-56.

13. Robert MF, Morin S, Beaulieu N, et al. DNMT1 is required to maintain CpG methylation and aberrant gene silencing in human cancer cells. Nature genetics 2003; 33: 61-65.

14. Pratt JR, Parker MD, Affleck LJ, et al. Ischemic epigenetics and the transplanted kidney. Transplant P 2006; 38: 3344-3346.

15. Mehta TK, Hoque MO, Ugarte R, et al. Quantitative detection of promoter hypermethylation as a biomarker of acute kidney injury during transplantation. Transplant P 2006; 38: 3420-3426.

16. Huang N, Tan L, Xue ZG, et al. Reduction of DNA hydroxymethylation in the mouse kidney insulted by ischemia reperfusion. Biochem Bioph Res Co 2012; 422: 697-702.

17. Subramanya AR, Ellison DH. Distal convoluted tubule. Clin J Am Soc Nephrol 2014; 9: 2147-2163.

18. Kasper DL. Harrison's principles of internal medicine, 19th edition / editors, Dennis L. Kasper, MD, William Ellery Channing, Professor of Medicine, Professor of Microbiology, Department of Microbiology and Immunobiology, Harvard Medical School, Division of Infectious Diseases, Brigham and Women's Hospital, Boston, Massachusetts and five others . edn. McGraw Hill Education: New York, 2015.

19. Widmaier EP, Raff H, Strang KT, et al. Vander's human physiology : the mechanisms of body function, 11th edn. McGraw-Hill Higher Education: Boston, 2008.

20. Section 2: AKI Definition. Kidney Int Suppl (2011) 2012; 2: 19-36.

21. Mehta RL, Burdmann EA, Cerda J, et al. Recognition and management of acute kidney injury in the International Society of Nephrology 0by25 Global Snapshot: a multinational cross-sectional study. Lancet 2016; 387: 2017-2025.

22. Winterberg PD, Lu CY. Acute kidney injury: the beginning of the end of the dark ages. Am J Med Sci 2012; 344: 318-325.

103

23. Rahman M, Shad F, Smith MC. Acute kidney injury: a guide to diagnosis and management. Am Fam Physician 2012; 86: 631-639.

24. Harrison's principles of internal medicine. Choice: Current Reviews for Academic Libraries 2008; 46: 138-138.

25. Zuk A, Bonventre JV. Acute Kidney Injury. Annu Rev Med 2016; 67: 293-307.

26. Wang D, Lippard SJ. Cellular processing of platinum anticancer drugs. Nat Rev Drug Discov 2005; 4: 307-320.

27. Cepeda V, Fuertes MA, Castilla J, et al. Biochemical mechanisms of cisplatin cytotoxicity. Anticancer Agents Med Chem 2007; 7: 3-18.

28. Launay-Vacher V, Rey JB, Isnard-Bagnis C, et al. Prevention of cisplatin nephrotoxicity: state of the art and recommendations from the European Society of Clinical Pharmacy Special Interest Group on Cancer Care. Cancer chemotherapy and pharmacology 2008; 61: 903-909.

29. Pabla N, Dong Z. Cisplatin nephrotoxicity: mechanisms and renoprotective strategies. Kidney Int 2008; 73: 994-1007.

30. Finkel M, Goldstein A, Steinberg Y, et al. Cisplatinum nephrotoxicity in oncology therapeutics: retrospective review of patients treated between 2005 and 2012. Pediatric nephrology 2014; 29: 2421-2424.

31. dos Santos NA, Carvalho Rodrigues MA, Martins NM, et al. Cisplatin-induced nephrotoxicity and targets of nephroprotection: an update. Archives of toxicology 2012; 86: 1233-1250.

32. Miller RP, Tadagavadi RK, Ramesh G, et al. Mechanisms of Cisplatin nephrotoxicity. Toxins 2010; 2: 2490-2518.

33. Linkermann A, Chen G, Dong G, et al. Regulated cell death in AKI. J Am Soc Nephrol 2014; 25: 2689-2701.

34. Peres LA, da Cunha AD, Jr. Acute nephrotoxicity of cisplatin: molecular mechanisms. J Bras Nefrol 2013; 35: 332-340.

35. Ozkok A, Edelstein CL. Pathophysiology of cisplatin-induced acute kidney injury. Biomed Res Int 2014; 2014: 967826.

36. Wang S, Zhang C, Hu L, et al. Necroptosis in acute kidney injury: a shedding light. Cell Death Dis 2016; 7: e2125.

104

37. Bonventre JV, Yang L. Cellular pathophysiology of ischemic acute kidney injury. J Clin Invest 2011; 121: 4210-4221.

38. Sharfuddin AA, Molitoris BA. Pathophysiology of ischemic acute kidney injury. Nat Rev Nephrol 2011; 7: 189-200.

39. Chatauret N, Badet L, Barrou B, et al. Ischemia-reperfusion: From cell biology to acute kidney injury. Prog Urol 2014; 24 Suppl 1: S4-12.

40. Yellon DM, Hausenloy DJ. Myocardial reperfusion injury. N Engl J Med 2007; 357: 1121-1135.

41. Menting TP, Wever KE, Ozdemir-van Brunschot DM, et al. Ischaemic preconditioning for the reduction of renal ischaemia reperfusion injury. Cochrane Database Syst Rev 2017; 3: CD010777.

42. Kinsey GR, Okusa MD. Pathogenesis of acute kidney injury: foundation for clinical practice. Am J Kidney Dis 2011; 58: 291-301.

43. Waddington CH. The epigenotype. 1942. Int J Epidemiol 2012; 41: 10-13.

44. Egger G, Liang G, Aparicio A, et al. Epigenetics in human disease and prospects for epigenetic therapy. Nature 2004; 429: 457-463.

45. Kimura H. Histone modifications for human epigenome analysis. J Hum Genet 2013; 58: 439-445.

46. Chen R, Kang R, Fan XG, et al. Release and activity of histone in diseases. Cell Death Dis 2014; 5: e1370.

47. Bhaumik SR, Smith E, Shilatifard A. Covalent modifications of histones during development and disease pathogenesis. Nat Struct Mol Biol 2007; 14: 1008-1016.

48. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004; 116: 281-297.

49. Chua JH, Armugam A, Jeyaseelan K. MicroRNAs: biogenesis, function and applications. Curr Opin Mol Ther 2009; 11: 189-199.

50. Alvarez-Garcia I, Miska EA. MicroRNA functions in animal development and human disease. Development 2005; 132: 4653-4662.

51. Flynt AS, Lai EC. Biological principles of microRNA-mediated regulation: shared themes amid diversity. Nat Rev Genet 2008; 9: 831-842.

105

52. Bushati N, Cohen SM. microRNA functions. Annu Rev Cell Dev Biol 2007; 23: 175-205.

53. Chang TC, Mendell JT. microRNAs in vertebrate physiology and human disease. Annu Rev Genomics Hum Genet 2007; 8: 215-239.

54. Mendell JT, Olson EN. MicroRNAs in stress signaling and human disease. Cell 2012; 148: 1172-1187.

55. Erson AE, Petty EM. MicroRNAs in development and disease. Clin Genet 2008; 74: 296-306.

56. Tufekci KU, Oner MG, Meuwissen RL, et al. The role of microRNAs in human diseases. Methods Mol Biol 2014; 1107: 33-50.

57. Ullah S, John P, Bhatti A. MicroRNAs with a role in gene regulation and in human diseases. Mol Biol Rep 2014; 41: 225-232.

58. Karolina DS, Wintour EM, Bertram J, et al. Riboregulators in kidney development and function. Biochimie 2010; 92: 217-225.

59. Saal S, Harvey SJ. MicroRNAs and the kidney: coming of age. Curr Opin Nephrol Hypertens 2009; 18: 317-323.

60. Wei Q, Mi QS, Dong Z. The regulation and function of microRNAs in kidney diseases. IUBMB Life 2013; 65: 602-614.

61. Bhatt K, Mi QS, Dong Z. microRNAs in kidneys: biogenesis, regulation, and pathophysiological roles. Am J Physiol Renal Physiol 2011; 300: F602-610.

62. Ho J, Kreidberg JA. The long and short of microRNAs in the kidney. J Am Soc Nephrol 2012; 23: 400-404.

63. Chandrasekaran K, Karolina DS, Sepramaniam S, et al. Role of microRNAs in kidney homeostasis and disease. Kidney Int 2012; 81: 617-627.

64. Ho J, Kreidberg JA. MicroRNAs in renal development. Pediatr Nephrol 2013; 28: 219-225.

65. Khalid U, Bowen T, Fraser DJ, et al. Acute kidney injury: a paradigm for miRNA regulation of the cell cycle. Biochem Soc Trans 2014; 42: 1219-1223.

66. Aguado-Fraile E, Ramos E, Conde E, et al. MicroRNAs in the kidney: novel biomarkers of acute kidney injury. Nefrologia 2013; 33: 826-834.

106

67. Liu Z, Wang S, Mi QS, et al. MicroRNAs in Pathogenesis of Acute Kidney Injury. Nephron 2016.

68. Ma L, Qu L. The function of microRNAs in renal development and pathophysiology. J Genet Genomics 2013; 40: 143-152.

69. Trionfini P, Benigni A, Remuzzi G. MicroRNAs in kidney physiology and disease. Nat Rev Nephrol 2015; 11: 23-33.

70. Mukhadi S, Hull R, Mbita Z, et al. The Role of MicroRNAs in Kidney Disease. Non-Coding RNA 2015; 1: 192-221.

71. Bhatt K, Kato M, Natarajan R. Mini-review: emerging roles of microRNAs in the pathophysiology of renal diseases. Am J Physiol Renal Physiol 2016; 310: F109- 118.

72. Schubeler D. Function and information content of DNA methylation. Nature 2015; 517: 321-326.

73. Jeltsch A. Beyond Watson and Crick: DNA methylation and molecular enzymology of DNA methyltransferases. Chembiochem 2002; 3: 274-293.

74. Ramsahoye BH, Biniszkiewicz D, Lyko F, et al. Non-CpG methylation is prevalent in embryonic stem cells and may be mediated by DNA methyltransferase 3a. Proc Natl Acad Sci U S A 2000; 97: 5237-5242.

75. Lister R, Pelizzola M, Dowen RH, et al. Human DNA methylomes at base resolution show widespread epigenomic differences. Nature 2009; 462: 315-322.

76. Shirane K, Toh H, Kobayashi H, et al. Mouse oocyte methylomes at base resolution reveal genome-wide accumulation of non-CpG methylation and role of DNA methyltransferases. PLoS Genet 2013; 9: e1003439.

77. Guo JU, Su Y, Shin JH, et al. Distribution, recognition and regulation of non-CpG methylation in the adult mammalian brain. Nat Neurosci 2014; 17: 215-222.

78. Zhu HY, Liu MY, Hong Q, et al. Role of microRNA-181a in the apoptosis of tubular epithelial cell induced by cisplatin. Chin Med J (Engl) 2012; 125: 523-526.

79. Messerschmidt DM, Knowles BB, Solter D. DNA methylation dynamics during epigenetic reprogramming in the germline and preimplantation embryos. Genes Dev 2014; 28: 812-828.

80. Portela A, Esteller M. Epigenetic modifications and human disease. Nat Biotechnol 2010; 28: 1057-1068.

107

81. Li E, Zhang Y. DNA methylation in mammals. Cold Spring Harb Perspect Biol 2014; 6: a019133.

82. Turek-Plewa J, Jagodzinski PP. The role of mammalian DNA methyltransferases in the regulation of gene expression. Cell Mol Biol Lett 2005; 10: 631-647.

83. Ting AH, Jair KW, Schuebel KE, et al. Differential requirement for DNA methyltransferase 1 in maintaining human cancer cell gene promoter hypermethylation. Cancer Research 2006; 66: 729-735.

84. Okano M, Xie S, Li E. Cloning and characterization of a family of novel mammalian DNA (cytosine-5) methyltransferases. Nat Genet 1998; 19: 219-220.

85. Gowher H, Liebert K, Hermann A, et al. Mechanism of stimulation of catalytic activity of Dnmt3A and Dnmt3B DNA-(cytosine-C5)-methyltransferases by Dnmt3L. J Biol Chem 2005; 280: 13341-13348.

86. Goll MG, Kirpekar F, Maggert KA, et al. Methylation of tRNAAsp by the DNA methyltransferase homolog Dnmt2. Science 2006; 311: 395-398.

87. Li E, Bestor TH, Jaenisch R. Targeted mutation of the DNA methyltransferase gene results in embryonic lethality. Cell 1992; 69: 915-926.

88. Okano M, Bell DW, Haber DA, et al. DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 1999; 99: 247-257.

89. Bogdanovic O, Veenstra GJ. DNA methylation and methyl-CpG binding proteins: developmental requirements and function. Chromosoma 2009; 118: 549-565.

90. Moore LD, Le T, Fan G. DNA methylation and its basic function. Neuropsychopharmacology 2013; 38: 23-38.

91. Nan X, Ng HH, Johnson CA, et al. Transcriptional repression by the methyl-CpG- binding protein MeCP2 involves a histone deacetylase complex. Nature 1998; 393: 386-389.

92. Ng HH, Zhang Y, Hendrich B, et al. MBD2 is a transcriptional repressor belonging to the MeCP1 histone deacetylase complex. Nature Genetics 1999; 23: 58-61.

93. Wu SC, Zhang Y. Active DNA demethylation: many roads lead to Rome. Nat Rev Mol Cell Biol 2010; 11: 607-620.

94. Ramiro AR, Barreto VM. Activation-induced cytidine deaminase and active cytidine demethylation. Trends Biochem Sci 2015; 40: 172-181.

108

95. Kohli RM, Zhang Y. TET enzymes, TDG and the dynamics of DNA demethylation. Nature 2013; 502: 472-479.

96. Pastor WA, Aravind L, Rao A. TETonic shift: biological roles of TET proteins in DNA demethylation and transcription. Nat Rev Mol Cell Bio 2013; 14: 341-356.

97. Guo JU, Su Y, Zhong C, et al. Hydroxylation of 5-methylcytosine by TET1 promotes active DNA demethylation in the adult brain. Cell 2011; 145: 423-434.

98. Dawlaty MM, Ganz K, Powell BE, et al. Tet1 is dispensable for maintaining pluripotency and its loss is compatible with embryonic and postnatal development. Cell Stem Cell 2011; 9: 166-175.

99. Ko M, Bandukwala HS, An J, et al. Ten-Eleven-Translocation 2 (TET2) negatively regulates homeostasis and differentiation of hematopoietic stem cells in mice. Proc Natl Acad Sci U S A 2011; 108: 14566-14571.

100. Dawlaty MM, Breiling A, Le T, et al. Combined deficiency of Tet1 and Tet2 causes epigenetic abnormalities but is compatible with postnatal development. Dev Cell 2013; 24: 310-323.

101. Gu TP, Guo F, Yang H, et al. The role of Tet3 DNA dioxygenase in epigenetic reprogramming by oocytes. Nature 2011; 477: 606-610.

102. Zilberman D, Henikoff S. Genome-wide analysis of DNA methylation patterns. Development 2007; 134: 3959-3965.

103. Gupta R, Nagarajan A, Wajapeyee N. Advances in genome-wide DNA methylation analysis. Biotechniques 2010; 49: iii-xi.

104. Yong WS, Hsu FM, Chen PY. Profiling genome-wide DNA methylation. Epigenetics Chromatin 2016; 9: 26.

105. Olkhov-Mitsel E, Bapat B. Strategies for discovery and validation of methylated and hydroxymethylated DNA biomarkers. Cancer Med 2012; 1: 237-260.

106. Ko YA, Mohtat D, Suzuki M, et al. Cytosine methylation changes in enhancer regions of core pro-fibrotic genes characterize kidney fibrosis development. Genome Biol 2013; 14: R108.

107. Wing MR, Devaney JM, Joffe MM, et al. DNA methylation profile associated with rapid decline in kidney function: findings from the CRIC study. Nephrol Dial Transplant 2014; 29: 864-872.

109

108. Smyth LJ, McKay GJ, Maxwell AP, et al. DNA hypermethylation and DNA hypomethylation is present at different loci in chronic kidney disease. Epigenetics 2014; 9: 366-376.

109. Marumo T, Yagi S, Kawarazaki W, et al. Diabetes Induces Aberrant DNA Methylation in the Proximal Tubules of the Kidney. J Am Soc Nephrol 2015; 26: 2388-2397.

110. Brennan EP, Ehrich M, O'Donovan H, et al. DNA methylation profiling in cell models of diabetic nephropathy. Epigenetics 2010; 5: 396-401.

111. Swan EJ, Maxwell AP, McKnight AJ. Distinct methylation patterns in genes that affect mitochondrial function are associated with kidney disease in blood-derived DNA from individuals with Type 1 diabetes. Diabet Med 2015; 32: 1110-1115.

112. Bell CG, Teschendorff AE, Rakyan VK, et al. Genome-wide DNA methylation analysis for diabetic nephropathy in type 1 diabetes mellitus. BMC Med Genomics 2010; 3: 33.

113. Bechtel W, McGoohan S, Zeisberg EM, et al. Methylation determines fibroblast activation and fibrogenesis in the kidney. Nat Med 2010; 16: 544-550.

114. Chen J, Zhang X, Zhang H, et al. Elevated Klotho promoter methylation is associated with severity of chronic kidney disease. PLoS One 2013; 8: e79856.

115. Pratt JR, Parker MD, Affleck LJ, et al. Ischemic epigenetics and the transplanted kidney. Transplant Proc 2006; 38: 3344-3346.

116. Parker MD, Chambers PA, Lodge JP, et al. Ischemia- reperfusion injury and its influence on the epigenetic modification of the donor kidney genome. Transplantation 2008; 86: 1818-1823.

117. Huang N, Tan L, Xue Z, et al. Reduction of DNA hydroxymethylation in the mouse kidney insulted by ischemia reperfusion. Biochem Biophys Res Commun 2012; 422: 697-702.

118. Wang H, Huang N, Liu Y, et al. Genomic distribution of 5-Hydroxymethylcytosine in mouse kidney and its relationship with gene expression. Ren Fail 2016; 38: 982- 988.

119. Endo K, Kito N, Fukushima Y, et al. A novel biomarker for acute kidney injury using TaqMan-based unmethylated DNA-specific polymerase chain reaction. Biomed Res 2014; 35: 207-213.

110

120. Mehta TK, Hoque MO, Ugarte R, et al. Quantitative detection of promoter hypermethylation as a biomarker of acute kidney injury during transplantation. Transplant Proc 2006; 38: 3420-3426.

121. Kang SW, Shih PA, Mathew RO, et al. Renal kallikrein excretion and epigenetics in human acute kidney injury: expression, mechanisms and consequences. BMC Nephrol 2011; 12: 27.

122. Jackson-Grusby L, Beard C, Possemato R, et al. Loss of genomic methylation causes p53-dependent apoptosis and epigenetic deregulation. Nat Genet 2001; 27: 31-39.

123. Fan G, Beard C, Chen RZ, et al. DNA hypomethylation perturbs the function and survival of CNS neurons in postnatal animals. J Neurosci 2001; 21: 788-797.

124. Kaneda M, Okano M, Hata K, et al. Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting. Nature 2004; 429: 900-903.

125. Rankin EB, Tomaszewski JE, Haase VH. Renal cyst development in mice with conditional inactivation of the von Hippel-Lindau tumor suppressor. Cancer Res 2006; 66: 2576-2583.

126. Wei Q, Dong G, Chen JK, et al. Bax and Bak have critical roles in ischemic acute kidney injury in global and proximal tubule-specific knockout mouse models. Kidney International 2013; 84: 138-148.

127. Pei L, Choi JH, Liu J, et al. Genome-wide DNA methylation analysis reveals novel epigenetic changes in chronic lymphocytic leukemia. Epigenetics 2012; 7: 567- 578.

128. Jiang M, Wei Q, Dong G, et al. Autophagy in proximal tubules protects against acute kidney injury. Kidney Int 2012; 82: 1271-1283.

129. Livingston MJ, Ding HF, Huang S, et al. Persistent activation of autophagy in kidney tubular cells promotes renal interstitial fibrosis during unilateral ureteral obstruction. Autophagy 2016; 12: 976-998.

130. Guo C, Wei Q, Su Y, et al. SUMOylation occurs in acute kidney injury and plays a cytoprotective role. Biochim Biophys Acta 2015; 1852: 482-489.

131. Wei Q, Dong G, Franklin J, et al. The pathological role of Bax in cisplatin nephrotoxicity. Kidney Int 2007; 72: 53-62.

111

132. Patel K, Dickson J, Din S, et al. Targeting of 5-aza-2'-deoxycytidine residues by chromatin-associated DNMT1 induces proteasomal degradation of the free enzyme. Nucleic Acids Res 2010; 38: 4313-4324.

133. Christman JK. 5-Azacytidine and 5-aza-2'-deoxycytidine as inhibitors of DNA methylation: mechanistic studies and their implications for cancer therapy. Oncogene 2002; 21: 5483-5495.

134. Michalowsky LA, Jones PA. Differential nuclear protein binding to 5-azacytosine- containing DNA as a potential mechanism for 5-aza-2'-deoxycytidine resistance. Mol Cell Biol 1987; 7: 3076-3083.

135. Tampe B, Tampe D, Muller CA, et al. Tet3-mediated hydroxymethylation of epigenetically silenced genes contributes to bone morphogenic protein 7-induced reversal of kidney fibrosis. J Am Soc Nephrol 2014; 25: 905-912.

136. Xiao X, Tang W, Yuan Q, et al. Epigenetic repression of Kruppel-like factor 4 through Dnmt1 contributes to EMT in renal fibrosis. Int J Mol Med 2015; 35: 1596- 1602.

137. Zhang Q, Yin S, Liu L, et al. Rhein reversal of DNA hypermethylation-associated Klotho suppression ameliorates renal fibrosis in mice. Sci Rep 2016; 6: 34597.

138. Le Clef N, Verhulst A, D'Haese PC, et al. Unilateral Renal Ischemia-Reperfusion as a Robust Model for Acute to Chronic Kidney Injury in Mice. Plos One 2016; 11.

139. Pidsley R, Zotenko E, Peters TJ, et al. Critical evaluation of the Illumina MethylationEPIC BeadChip microarray for whole-genome DNA methylation profiling. Genome Biol 2016; 17: 208.

140. Gu H, Bock C, Mikkelsen TS, et al. Genome-scale DNA methylation mapping of clinical samples at single-nucleotide resolution. Nat Methods 2010; 7: 133-136.

141. Hellman A, Chess A. Gene body-specific methylation on the active X chromosome. Science 2007; 315: 1141-1143.

142. Shukla S, Kavak E, Gregory M, et al. CTCF-promoted RNA polymerase II pausing links DNA methylation to splicing. Nature 2011; 479: 74-79.

143. Maunakea AK, Chepelev I, Cui K, et al. Intragenic DNA methylation modulates alternative splicing by recruiting MeCP2 to promote exon recognition. Cell Res 2013; 23: 1256-1269.

144. Magnusson M, Lu EX, Larsson P, et al. Dynamic Enhancer Methylation--A Previously Unrecognized Switch for Tissue-Type Plasminogen Activator Expression. PLoS One 2015; 10: e0141805.

112

145. Jin B, Robertson KD. DNA methyltransferases, DNA damage repair, and cancer. Adv Exp Med Biol 2013; 754: 3-29.

146. Jones PA, Liang G. Rethinking how DNA methylation patterns are maintained. Nat Rev Genet 2009; 10: 805-811.

147. Liang G, Chan MF, Tomigahara Y, et al. Cooperativity between DNA methyltransferases in the maintenance methylation of repetitive elements. Mol Cell Biol 2002; 22: 480-491.

148. DNMT3A and TET2 Cooperate and Compete to Maintain HSC DNA Methylation. Cancer Discov 2016; 6: 944.

149. Egger G, Jeong S, Escobar SG, et al. Identification of DNMT1 (DNA methyltransferase 1) hypomorphs in somatic knockouts suggests an essential role for DNMT1 in cell survival. Proc Natl Acad Sci U S A 2006; 103: 14080-14085.

150. Robertson KD. DNA methylation, methyltransferases, and cancer. Oncogene 2001; 20: 3139-3155.

151. Subramaniam D, Thombre R, Dhar A, et al. DNA methyltransferases: a novel target for prevention and therapy. Front Oncol 2014; 4: 80.

152. Feng J, Zhou Y, Campbell SL, et al. Dnmt1 and Dnmt3a maintain DNA methylation and regulate synaptic function in adult forebrain neurons. Nat Neurosci 2010; 13: 423-430.

153. Rhee I, Bachman KE, Park BH, et al. DNMT1 and DNMT3b cooperate to silence genes in human cancer cells. Nature 2002; 416: 552-556.

154. Wei Q, Bhatt K, He HZ, et al. Targeted deletion of Dicer from proximal tubules protects against renal ischemia-reperfusion injury. J Am Soc Nephrol 2010; 21: 756-761.

155. Basile DP, Bonventre JV, Mehta R, et al. Progression after AKI: Understanding Maladaptive Repair Processes to Predict and Identify Therapeutic Treatments. Journal of the American Society of Nephrology : JASN 2016; 27: 687-697.

156. Ha K, Lee GE, Palii SS, et al. Rapid and transient recruitment of DNMT1 to DNA double-strand breaks is mediated by its interaction with multiple components of the DNA damage response machinery. Hum Mol Genet 2011; 20: 126-140.

157. Mortusewicz O, Schermelleh L, Walter J, et al. Recruitment of DNA methyltransferase I to DNA repair sites. Proc Natl Acad Sci U S A 2005; 102: 8905- 8909.

113

158. Unterberger A, Andrews SD, Weaver IC, et al. DNA methyltransferase 1 knockdown activates a replication stress checkpoint. Mol Cell Biol 2006; 26: 7575- 7586.

159. Sun CY, Chang SC, Wu MS. Suppression of Klotho expression by protein-bound uremic toxins is associated with increased DNA methyltransferase expression and DNA hypermethylation. Kidney Int 2012; 81: 640-650.

160. Yang D, Thangaraju M, Greeneltch K, et al. Repression of IFN regulatory factor 8 by DNA methylation is a molecular determinant of apoptotic resistance and metastatic phenotype in metastatic tumor cells. Cancer Res 2007; 67: 3301-3309.

161. McGough JM, Yang D, Huang S, et al. DNA methylation represses IFN-gamma- induced and signal transducer and activator of transcription 1-mediated IFN regulatory factor 8 activation in colon carcinoma cells. Mol Cancer Res 2008; 6: 1841-1851.

162. Lee KY, Geng H, Ng KM, et al. Epigenetic disruption of interferon-gamma response through silencing the tumor suppressor interferon regulatory factor 8 in nasopharyngeal, esophageal and multiple other carcinomas. Oncogene 2008; 27: 5267-5276.

163. Zhang Q, Zhang L, Li L, et al. Interferon regulatory factor 8 functions as a tumor suppressor in renal cell carcinoma and its promoter methylation is associated with patient poor prognosis. Cancer Lett 2014; 354: 227-234.

164. Tamura T, Ozato K. ICSBP/IRF-8: its regulatory roles in the development of myeloid cells. J Interferon Cytokine Res 2002; 22: 145-152.

165. Nguyen H, Hiscott J, Pitha PM. The growing family of interferon regulatory factors. Cytokine Growth Factor Rev 1997; 8: 293-312.

166. Contursi C, Wang IM, Gabriele L, et al. IFN consensus sequence binding protein potentiates STAT1-dependent activation of IFNgamma-responsive promoters in macrophages. Proc Natl Acad Sci U S A 2000; 97: 91-96.

167. Yang D, Thangaraju M, Browning DD, et al. IFN regulatory factor 8 mediates apoptosis in nonhemopoietic tumor cells via regulation of Fas expression. J Immunol 2007; 179: 4775-4782.

168. Yang J, Hu X, Zimmerman M, et al. Cutting edge: IRF8 regulates Bax transcription in vivo in primary myeloid cells. J Immunol 2011; 187: 4426-4430.

114

169. Yang D, Wang S, Brooks C, et al. IFN regulatory factor 8 sensitizes soft tissue sarcoma cells to death receptor-initiated apoptosis via repression of FLICE-like protein expression. Cancer Res 2009; 69: 1080-1088.

170. Kubosaki A, Lindgren G, Tagami M, et al. The combination of gene perturbation assay and ChIP-chip reveals functional direct target genes for IRF8 in THP-1 cells. Mol Immunol 2010; 47: 2295-2302.

171. Simon PS, Sharman SK, Lu C, et al. The NF-kappaB p65 and p50 homodimer cooperate with IRF8 to activate iNOS transcription. BMC Cancer 2015; 15: 770.

172. Geng H, Brennan S, Milne TA, et al. Integrative epigenomic analysis identifies biomarkers and therapeutic targets in adult B-acute lymphoblastic leukemia. Cancer discovery 2012; 2: 1004-1023.

173. Kurosu T, Fukuda T, Miki T, et al. BCL6 overexpression prevents increase in reactive oxygen species and inhibits apoptosis induced by chemotherapeutic reagents in B-cell lymphoma cells. Oncogene 2003; 22: 4459-4468.

174. Phan RT, Dalla-Favera R. The BCL6 proto-oncogene suppresses p53 expression in germinal-centre B cells. Nature 2004; 432: 635-639.

175. Shao Y, Ling CC, Liu XQ. High concentrations of glucose suppress etoposide- induced cell death of B-cell lymphoma through BCL-6. Biochem Bioph Res Co 2014; 450: 227-233.

176. Ranuncolo SM, Wang L, Polo JM, et al. BCL6-mediated attenuation of DNA damage sensing triggers growth arrest and senescence through a p53-dependent pathway in a cell context-dependent manner. The Journal of biological chemistry 2008; 283: 22565-22572.

177. Wegner W, Burckhardt G, Henjakovic M. Transcriptional regulation of human organic anion transporter 1 by B-cell CLL/lymphoma 6. Am J Physiol-Renal 2014; 307: F1283-F1291.

178. Andersson O, Reissmann E, Ibanez CF. Growth differentiation factor 11 signals through the transforming growth factor-beta receptor ALK5 to regionalize the anterior-posterior axis. EMBO Rep 2006; 7: 831-837.

179. Yallowitz AR, Hrycaj SM, Short KM, et al. Hox10 genes function in kidney development in the differentiation and integration of the cortical stroma. PLoS One 2011; 6: e23410.

180. Patterson LT, Potter SS. Atlas of Hox gene expression in the developing kidney. Dev Dyn 2004; 229: 771-779.

115

181. Gabellini D, Colaluca IN, Vodermaier HC, et al. Early mitotic degradation of the homeoprotein HOXC10 is potentially linked to cell cycle progression. EMBO J 2003; 22: 3715-3724.

182. Christen B, Beck CW, Lombardo A, et al. Regeneration-specific expression pattern of three posterior Hox genes. Dev Dyn 2003; 226: 349-355.

116