ESTABLISHMENT AND CHARACTERIZATION OF A PRENATAL METHYL-DEFICIENT NUTRITIONAL RAT MODEL

Get Complete Project Material File(s) Now! »

CHAPTER 3. ANALYSIS OF METHYLATION AND EXPRESSION OF TARGET GENES IN THE LIVER OF CON AND MD OFFSPRING DURING POSTNATAL DEVELOPMENT

Introduction

As was discussed in Chapter 2, the maternal methyl donor deficient diet caused intrauterine growth retardation in the offspring. However, unlike other prenatal programming animal models, the MD offspring did not develop overt signs of the Metabolic syndrome such as obesity, disturbed glucose metabolism or high blood pressure. The absence of phenotypic changes in the MD offspring does not exclude the possibility that, at the molecular level, regulation of expression of genes that contribute to development of adult diseases was affected. Hypothetically, these molecular changes could make the MD offspring more susceptible to the adult onset diseases if they encountered unfavorable conditions during postnatal development. This chapter describes experiments that were done to investigate DNA methylation and gene expression of selected genes since current hypotheses consider DNA methylation as a strong candidate for molecular mechanisms underlying developmental programming (Gluckman & Hanson, 2008; Godfrey, Lillycrop, Burdge, Gluckman, & Hanson, 2007).
The present study investigates molecular mechanisms in liver tissue, since it is known that the liver plays one of the central roles in regulation of glucose and lipid metabolism (van den Berghe, 1991). Disregulation of metabolic pathways in the liver is known to contribute to pathological conditions such as obesity (Ji & Friedman, 2007), fatty liver, which can progress to hepatic cirrhosis and hepatocellular carcinoma (Marchesini, Moscatiello, Di Domizio, & Forlani, 2008; Powell et al., 1990), atherosclerosis (Loria, Lonardo, & Targher, 2008) and insulin resistance (Cornier et al., 2008).
A small set of 8 genes was selected to study the effect of prenatal methyl donor deficiency on methylation levels and mRNA expression in liver tissue of the newborn offspring and 110 day old offspring. The first group included genes coding for glucocorticoid receptor (Gcr) and the type 2 11ß-hydroxysteroid dehydrogenase (Hsd11β2), two proteins involved in glucocorticoid signaling, which plays a role in the regulation of carbohydrate and lipid metabolism in the liver. The second group included two isotypes of a family of peroxisome proliferator-activated receptor alpha (Ppar) – α and γ, also known to regulate lipid metabolism. The third group included genes encoding proteins involved in the regulation of cell cycle, and known or suggested to have tumor suppressing abilities. The selection of the third group of genes was based on the hypothesis that the methylation and expression patterns of these genes could be influenced in utero, thus predisposing the offspring to neoplastic developments in adulthood. Additionally, some of the selected genes were shown to be involved in the regulation of pathways associated with the metabolic syndrome (J. G. Eriksson, 2008; Kendall, 2005; Robitaille et al., 2004).
Glucorticoid receptor (Gcr) expression in the liver plays an important role in the regulation of glucose and protein metabolism, and its disregulation is known to be one of the underlying causes of the metabolic syndrome (Vegiopoulos & Herzig, 2007). Tissue-specific expression of Gcr in a rat is controlled by multiple promoters located within exon 1, which has at least 11 alternative transcription initiation sites, with the exon 110 expressed predominantly in the liver (McCormick et al., 2000). A protein deficient diet during pregnancy was shown to decrease methylation of the promoter exon 110 and increase expression of the Gcr gene in liver tissue of the post-weaned offspring of deficient mothers (Lillycrop et al., 2005; Lillycrop et al., 2007).
Another gene investigated in this study is Hsd11β2, whose product deactivates corticosterone through it’s conversion to the less active cortisol in rodents. Hydroxysteroid dehydrogenases are involved in regulation of the access of local corticosterone to glucocorticoid receptors (Funder, Pearce, Smith, & Smith, 1988). The promoter region and exon 1 of the Hsd11β2 contains a CpG island; decreased expression of this gene was shown to be associated with increased methylation of the promoter region in both human and rat (Alikhani-Koopaei, Fouladkou, Frey, & Frey, 2004).
Two isotypes of a family of peroxisome proliferator-activated receptor alpha (Ppar) – α and γ were chosen to investigate in this study as well. Pparα is expressed in the liver at high levels and Pparγ is expressed in adipose tissue and in the liver (Memon et al., 2000; Vidal-Puig et al., 1996). Studies have demonstrated that Pparα is involved in glucose and lipid metabolism (Djouadi et al., 1998; Patsouris, Muller, & Kersten, 2004), and liver expression of both of these genes is up-regulated in a mouse model of obesity (Memon et al., 2000; Vidal-Puig et al., 1996). The promoter regions of Pparα and Pparγ contain CpG islands, methylation of which is thought to be involved in regulation of their expression (Lillycrop et al., 2005). In a rat, protein restriction during pregnancy was associated with decreased methylation of the Pparα gene, and its’ increased expression in the postweaned offspring (Burdge et al., 2007; Lillycrop et al., 2005).
The tumor suppressor p53 gene is involved in the induction of cell cycle arrest and apoptosis in response to cell damaging events (Bunz et al., 1998; Caelles, Helmberg, & Karin, 1994). Recently, it has been suggested that p53 might function beyond tumor suppression and be involved in the regulation of processes such as aging, glucose metabolism and stress response (reviewed by (Vousden & Lane, 2007). The p53 gene contains a CpG island within exons 6 and 7. A methyl-donor deficient diet in adult rats was shown to decrease p53 methylation in this region in preneoplastic hepatic nodules, which was associated with increased mRNA expression (Pogribny, Miller, & James, 1997).
One of the downstream targets of p53 is a cyclin-dependent kinase inhibitor 1a (Cdkn1a) encoding p21 protein, which inhibits cell cycle progression via binding to cyclin kinase complexes (Bendjennat et al., 2003; Harper, Adami, Wei, Keyomarsi, & Elledge, 1993). Besides involvement of these two genes in hepatocellular carcinoma (Hui, Makuuchi, & Li, 1998), one study demonstrated upregulation of the p53 and p21 genes in the liver of obese mice, and suggested their role in hepatocellular injury associated with fatty liver disease (Yahagi et al., 2004). In rat cell culture, methylation of the CpG island in the promoter region of the p21 gene was shown to be associated with down-regulation of expression of this gene (Allan, Duhig, Read, & Fried, 2000).
Dusp5 is another gene that is regulated by p53, and involved in control of cell cycle (Ueda, Arakawa, & Nakamura, 2003). It is involved in regulation of the mitogen activated protein kinase pathway (MAPK), mostly in the liver and placenta (Ishibashi, Bottaro, Michieli, Kelley, & Aaronson, 1994; Kovanen et al., 2003). Dusp5 activity leads to decreased phosphorylation of the insulin receptor substrate-1, and disturbance of this pathway could be involved in development of insulin resistance (Q. Fu, McKnight, Yu, Callaway, & Lane, 2006). Exon 2 of the Dusp5 gene contains a differentially methylated region, which was shown to be hypomethylated in the adult growth- retarded offspring in the bilateral uterine artery ligation rat model (Q. Fu et al., 2006). Altered methylation of Dusp5 was associated with long-term changes in it’s expression (Q. Fu et al., 2006).
A cyclin-dependent kinase inhibitor 2a (Cdkn2a, p16) was also demonstrated to be involved in negative regulation of the cell cycle (Koh, Enders, Dynlacht, & Harlow, 1995). It was shown that the promoter Cdkn2a often had abnormal DNA methylation in hepatocellular carcinomas (Harder et al., 2008). Similar to the p53 gene, the promoter of the Cdkn2a gene undergoes progressive changes in methylation in hepatic tumors, induced by methyl donor deficient diets (Pogribny & James, 2002).
The aim of this experiment was to investigate the short-term (newborn offspring) and the longterm (young adult offspring, 110 day old) effects of the maternal diet on methylation and expression of target genes in order to determine whether maternal methyl donor deficiency could affect the epigenetic regulation of genes involved in the pathogenesis of adult onset diseases. Furthermore, the age-specific (from d1 to d110) methylation and expression of target genes in the liver were investigated in order to gain a better understanding of possible postnatal molecular changes of these genes.

Materials and Methods

Tissue collection

Liver tissue was collected from the newborn (d1) and young adult offspring of the CON and MD mothers as described in Chapter 2 Section 2.2.3.1.

Genomic DNA extraction

Genomic DNA was extracted using modified standard phenol/chloroform/isoamyl alcohol method (Sambrook, Fritsch, & Maniatis, 1989). The following chemicals were used for this protocol: Ammonium acetate (Sigma), Chloroform (BDH, AnalaR quality), Ethanol (99% pure, BDH), Hydroquinone (Sigma), Isoamyl alcohol (BDH), Isopropanol (2-Propanol, BDH), Phenol, buffer saturated (Invitrogen), Sodium dodecyl sulfate (SDS) (Sigma-Aldrich), UltraPureTM DNase/RNase-free distilled water (nuclease free water) (Invitrogen).
The procedure was as follows: a piece of frozen tissue (10 – 50 mg) was powdered in liquid nitrogen (5-10ml) on dry ice. Powdered tissue was incubated overnight with 1ml of lysis buffer (100mM NaCl, 10mM Tris-Cl, pH8, 25mM EDTA, pH8 and 0.5% SDS) and 15μl of 10mg/ml proteinase K (30U/mg of enzyme, Sigma) solution at 550C with constant rotation. The next day, tissue lysate was mixed with phenol/chloroform/isoamyl alcohol using 25:24:1 ratio, and incubated with gentle shaking for 20min at room temperature. Then the lysate was centrifuged at maximum speed (13rpm standard tabletop centrifuge), at room temperature for 15min. The upper aqueous phase was collected into a fresh 2ml tube. DNA was precipitates with 0.5 volumes (of the initial volume of the lysate) of ammonium acetate (7.5M, Sigma) and 1 volume of isopropanol (1ml). The precipitate was centrifuged at maximum speed for 15 min at 40C, after which the supernatant was removed and discarded. The genomic DNA pellet was washed with 1ml of 70% ethanol, vortexed briefly, and again centrifuged for 15min at 40C. The wash step was repeated one more time. After that, DNA pellet was dried briefly and resuspended in 500μl of MilliQ water. RNase A (Sigma) was added to the final concentration of 100ng/μl and RNase T1 (Roche) was added to the final concentration of 1000U/ml and incubated at 370C for 1 hour. After incubation, DNA was re-extracted with phenol/chloroform/isoamyl alcohol mix as described above. Next, aqueous phase was mixed with chloroform/isoamyl alcohol (49:1 respectively), incubated at room temperature for 15min with gentle shaking, and centrifuged at maximum speed for 15min at room temperature. The aqueous phase was collected and genomic DNA was precipitated using 0.5 volume of ammonium acetate and 2 volumes of 99% ethanol. Precipitated DNA was centrifuged at maximum speed at 40C for 15min and washed with 70% ethanol twice, as described above. The DNA pellet was resuspended in nuclease free water. Concentration and DNA quality was measured using a NanoDrop (NanoDrop, Wilmington, DE, USA) spectrophotometer. DNA solutions with a 260/280 ratio above 1.78 were considered as sufficient purity genomic DNA.

. Global DNA methylation analysis

The level of global DNA methylation was determined by HPLC quantification of 5-methyl 2’deoxycytidine (5’mdCyt) and 2’-deoxycytidine (dCyt) using a modification of the method described by Friso and colleagues (Friso, Choi, Dolnikowski, & Selhub, 2002). Genomic DNA was prepared from newborn and adult liver tissue using the phenol/chloroform/isoamyl alcohol method as above (Sambrook et al., 1989). DNA aliquots (100ng/μl) were incubated consecutively with 1/10 volume of 0.1M ammonium acetate (pH5.3) and 0.2U of Nuclease P1 (MP Biomedicals) per 100ng of gDNA for 2 hours at 450C; ammonium bicarbonate (1M) and 0.0002U of venom phosphodiesterase (Roche) per 100ng of gDNA for 1.5hours at 370C, and 0.1U of shrimp alkaline phosphatase (Promega) per 100ng of gDNA for 2 hours at 370C. Iododeoxycytidine (Sigma) was added as an internal standard to each sample prior to HPLC (end concentration 20μM). Standard curve dilutions of 2’-deoxycytidine (Sigma) 200μM, 100μM, 50μM, 20μM and 4μM, and of 5-methyl-2’-deoxycytidine (MP Biomedicals, US) 5μM, 2.5μM, 1.25μM, 0.5μM and 0.25μM were prepared at the same time. Samples or standards were injected into an HPLC C18 Luna column (250×4.6mm, 3μm; Phenomenex), the separation of dCyt and 5’mdCyt was achieved by gradient elution at 0.6ml/min flow rate on an Alliance 2690 HPLC system (BAE, Systems-AlphaTech), column temperature 300C and sample temperature 100C, and UV detection was done using a Waters 2996 PDA detector at 254 nm wave length. Raw data were collected using Millenium32® software (Waters, US). The area under the peak was used to calculate methylation levels that were defined as the percentage of 5’mdCyt of the overall amount of dCyt and 5’mdCyt. The R2 for dCyt standard curve was 0.9992 and for 5’mdCyt R2 was 0.9978.

Sodium bisulfite DNA conversion

Genomic DNA was extracted using modified phenol/chloroform/isoamyl alcohol method as described above. The sodium bisulfite DNA conversion protocol was a modified version of the protocol described by Frommer and colleagues (Frommer et al., 1992). Chemicals were obtained from following suppliers: Hydroquinone (Sigma), Isoamyl alcohol (BDH), Isopropanol (2Propanol, BDH), Phenol (Invitrogen), Sodium acetate (BDH), Sodium hydroxide (BDH, AnalaR grade), Sodium metabisulfite (Sigma), UltraPureTM DNase/RNase-free distilled water (nuclease free water) (Invitrogen).
Five micrograms of gDNA were denatured in 0.3M NaOH (10µl total volume) for 30min at 370C in a PCR machine (Mastercycler® Gradient, Eppendorf, Germany). Sodium bisulfite solution (2.3M, pH5) was prepared fresh each time on the day of an experiment. It consisted of 4.4g sodium metabisulfite mixed with 0.5ml of 10mM Hydroquinone solution (0.006g of Hydroquinone mixed with 5 ml of MilliQ water), 0.8 ml of NaOH and MilliQ water up to 9 ml. The solution pH was adjusted to 5 with more 3M NaOH, and the volume was adjusted to 10ml with MilliQ water.
Denatured gDNA (10μl) was mixed with 90µl of 2.3M Sodium bisulfite solution and conversion was done in a PCR machine (Mastercycler® Gradient, Eppendorf, Germany) at 950C for 2min and 500C for 30min, repeated 28 times, with an indefinite hold at 40C.
After sodium bisulfite conversion, the DNA was desalted by two consecutive precipitations. During the first desalting step, 0.1 volumes (of initial sodium bisulfite converted (SBC) DNA solution volume) of Ammonium acetate 7.5M, 5 volumes (of initial SBC DNA volume) of water and 6µl of Linear polyacrylamide (recipe in Appendix II) were mixed with SBC DNA solution. An equal volume of isopropanol was added, and the solution was centrifuged at maximum speed for 10min at RT.
The supernatant was removed, and the pellet was re-dissolved in 300µl of MilliQ water and 30µl of 3M Sodium acetate pH7. DNA was precipitated with 2 volumes of 99% ethanol and centrifuged for 5min, at maximum speed at room temperature (RT). The supernatant was removed, and the pellet washed with 70% ethanol.
After washing, the pellet was redissolved in 200µl of 0.3M NaOH and incubated at 370C for 30min. DNA was precipitated with 0.5 volumes of 7.5M Ammonium acetate and 2 volumes of 99% ethanol. The solution was centrifuged in a table top centrifuge for 10min at maximum speed, at 40C. The DNA pellet was washed with 70% ethanol (the same as in the protocol for gDNA extraction). The DNA pellet was dissolved in 20-30µl of nuclease free water and stored at –800C.

Quantitative DNA methylation analysis using base-specific cleavage and Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS)

Quantitative DNA methylation analysis of CpG sites was done using the Sequenom-developed method, which employs the MassARRAY Compact System (http://www.sequenom.com). This method is based on Homogeneous MassCLEAVETM (hMC) base-specific cleavage and matrixassisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) for the detection and quantitative analysis of DNA methylation (Ehrich et al., 2005). The analysis was done as described previously (Ehrich et al., 2005). Primers specific for bisulfite converted sequence of target genes were designed using MethPrimer software (http://www.urogene.org/methprimer/index1.html) (L. C. Li & Dahiya, 2002). Primer sequences can be found in Table 3.2, and gDNA sequences can be found in the Results section. To avoid bias in amplification of methylated or unmethylated DNA, primers did not contain any CpG sites in their sequence. Each reverse primer had a T7-promotor tag for in vitro RNA transcription (5′- cagtaatacgactcactatagggagaaggct-3′), and the forward primer was tagged with a 10mer (5′aggaagagag-3′) for annealing temperature balance.
3.2.5.1. Brief overview of the method
The quantitative methylation analysis technique developed by Sequenom® utilizes a C/T polymorphism created by sodium bisulfite DNA conversion. Sodium bisulfite treatment, described in the previous section, results in conversion of unmethylated cytosine to uracil, whereas methylated cytosine remain unaffected (Figure 3.1, step 1). During the following PCR amplification (step 2, with primers tagged with T7 promoter sequence), uracil is amplified as thymine. A partially methylated CpG site in studied tissue is represented by a pool of PCR products, whereby a proportion of them would have thymine at that particular CpG site and the rest of PCR products would have cytosine at the same CpG site. After PCR, unincorporated dNTPs are inactivated with Shrimp alkaline phosphatase and the PCR product is used as a template for in vitro RNA synthesis (Figure 3.1, step 3). RNA synthesis is done using a T7DNA&RNA polymerase that can incorporate rNTPs as well as dNTPs. As a substrate, a mixture of dCTP and rATP, rGTP and rUTP nucleotides is used. RNase A cleaves RNA 3’ to cytosine and uracil. However, in this case, cleavage of in vitro transcribed RNA by RNaseA is specific for uracil, because dCTP is non-cleavable by RNaseA. RNaseA digestion of the RNA fragment synthesised from the reverse strand of PCR product generates same length fragments that differ by A/G nucleotides, which results in a 16Da mass difference per one CpG site (Figure 3.1, step 4). MALDI-TOF MS is used to detect differences in mass between the same length fragments containing A and G nucleotides. The abundance of each fragment is detected as signal intensity, which represents the amount of methylation in the studied sequence (Figure 3.1, step 5). Each RNA fragment containing one or more CpG sites is defined as a CpG unit. Depending on the specific sequence, a CpG unit could contain one CpG site or two or more than two CpG sites. Some CpG units in the sequence cannot be analyzed due to several reasons: the RNA fragment was outside of the mass spectrometer detection range, close interfering peaks or overlapping peaks.

TABLE OF CONTENTS
ABSTRACT
DEDICATION
ACKNOWLEDGEMENTS
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
LIST OF ABBREVIATIONS
CHAPTER 1. INTRODUCTION
1.1. DEVELOPMENTAL ORIGINS OF HEALTH AND DISEASE
1.2. BIOLOGICAL MECHANISMS UNDERLYING DEVELOPMENTAL PROGRAMMING
1.3. DNA METHYLATION AS AN EPIGENETIC MECHANISMS OF GENE REGULATION
1.4. SCOPE OF THESIS
CHAPTER 2. ESTABLISHMENT AND CHARACTERIZATION OF A PRENATAL METHYL-DEFICIENT NUTRITIONAL RAT MODEL
2.1. INTRODUCTION
2.2. MATERIALS AND METHODS
2.3. RESULTS
2.4. DISCUSSION
CHAPTER 3. ANALYSIS OF METHYLATION AND EXPRESSION OF TARGET GENES IN THE LIVER OF CON AND MD OFFSPRING DURING POSTNATAL DEVELOPMENT
3.1. INTRODUCTION
3.2. MATERIALS AND METHODS
3.3. RESULTS
3.4. DISCUSSION
CHAPTER 4. EFFECTS OF MATERNAL METHYL DONOR DEFICIENCY ON EPIGENTIC REGULATION OF IMPRINTED GENES IGF2 AND H19
4.1. INTRODUCTION
4.2. MATERIAL AND METHODS
4.3. RESULTS
4.4. DISCUSSION
CHAPTER 5. STUDIES OF BEHAVIOR AND EPIGENETIC MECHANISMS IN THE HIPPOCAMPUS IN THE CON AND MD OFFSPRING
5.1. INTRODUCTION
5.2. MATERIALS AND METHODS
5.3. RESULTS
5.4. DISCUSSION
CHAPTER 6. FINAL CONCLUSIONS
6.1. FINAL SUMMARY
6.2. LIMITATIONS OF THE STUDY
6.3. FUTURE DIRECTIONS
APPENDIX I
APPENDIX II
APPENDIX III
APPENDIX IV
REFERENCES

GET THE COMPLETE PROJECT
EFFECTS OF PRENATAL DIETARY METHYL DONOR DEFICIENCY ON DEVELOPMENT AND EPIGENETIC MECHANISMS IN OFFSPRING – STUDIES IN THE RAT