2013 Annual Report
1a.Objectives (from AD-416):
Obj 1: Identify genes that show epigenetic dysregulation in obesity using a candidate-gene approach. Subobj 1A. Characterize developmental establishment of DNA methylation at hypothalamic genes known to affect food intake regulation. Subobj 1B. Compare DNA methylation and expression of these genes between lean and obese mice. Obj 2: Determine if methylation and expression of specific genes in hypothalamus and/or adipose tissue differ between lean and obese mice. Subobj 2A. Identify genomic loci in hypothalamus and adipose tissue at which epigenetic dysregulation is associated with obesity. Subobj 2B. Determine if interindividual epigenetic variation at these loci is found before the onset of obesity. Obj 3: Determine if maternal obesity and/or nutrition before and during pregnancy persistently alters epigenetic regulation in offspring hypothalamus or adipose tissue. Subobj 3A. Identify genomic loci at which epigenetic dysregulation is induced by maternal obesity. Subobj 3B. Determine if this induced epigenetic dysregulation can be prevented by altering maternal diet. Obj 4: Identify placental epigenetic mechanisms that affect fetal nutrition, growth and development. Subobj 4A. Use methylation-specific amplification in conjunction with MSAM of genomic DNA extracted from trophoblast of normal term placentas and androgenetic complete hydatidiform moles to identify novel imprinted genes and other epigenetically regulated genes in trophoblast that play a role in regulation of fetal nutrition. Subobj 4B. Analyze how NLRP7, which is mutated in women with biparental hydatidiform moles, contributes to imprinting in early embryo and placenta. NLRP7 protein, known to have a role in innate immunity, may be a link between environmental changes (such as suboptimal maternal nutrition) and imprinting alterations during development. Obj 5: Determine how programming of glucose intolerance, obesity, and epigenetic dysregulation of skeletal muscle-growth in mice is affected by maternal diet during development. Preliminary data show that offspring of mice exposed to maternal low-protein (MLP) diet have reduction in skeletal muscle mass at <1 yr of age and possibly glucose intolerance. Subobj 5A. Extend phenotypic studies on muscle development, glucose tolerance, growth and obesity to include maternal methyl-donor depleted (MDD) diets and to follow offspring exposed to MLP or MDD diets for up to 18 months. Subobj 5B. To find the molecular basis for the altered muscle phenotype after MLP diet, we will perform array-based gene expression and methylation profiling, as well as gene expression analysis and DNA methylation analysis of candidate genes in hindleg muscle-tissues from 21-d-old and 1-yr-old mice exposed to MLP diet. Obj. 6 Utilize genome-wide DNA methylation profiling to determine if epigenetic programming/reprogramming contribute to lineage-specific patterns of gene expression. Obj. 7 Develop a targeted knock-in mouse model.
1b.Approach (from AD-416):
Children's Nutrition Research Center scientists will use the agouti viable yellow (Avy) mouse model to study the effects of maternal obesity on the risk of obesity in the offspring. DNA methylation in two tissues is known to play important roles in body weight regulation, hypothalamus and adipose tissue. Scientists will study this by using both a candidate gene approach and a genome-wide DNA methylation profiling technique (MSAM). In human studies, our research team will use MSAM and commercially available methylation screening tools such as the Illumina Infinium methylation array to identify genomic regions of differential methylation in complete hydatidiform moles compared with normal term placentas. Also, to characterize the role of NLRP7 in genomic imprinting, we will use chromatin immunoprecipitation to assess interactions of NLRP7 with chromatin, identify DNA binding sites of NLRP7 by electromobility shift assays, and screen for NLRP7 protein binding partners using yeast two-hybrid assays. Lastly, previous studies in a mouse model of the effects of maternal low protein diet on skeletal muscle development in the offspring will be extended to include longer-term studies and additional types of early exposures. This research will investigate fundamental mechanisms regulating DNA methylation during development, and characterize their involvement in nutritional programming during critical ontogenic periods.
Note: DNA methylation is a developmentally regulated modification of DNA that helps control cell-type specific gene expression. For subobjective 2A, we wrote a manuscript describing our studies of how early postnatal overnutrition (in the suckling-period small litter mouse model) affects the lifelong propensity for obesity. Our manuscript, for the first time, identified persistent epigenetic changes in the hypothalamus that are induced by early postnatal overnutrition. (The hypothalamus plays a central role in the regulation of food intake and energy expenditure.) In response to reviewers' comments, we performed detailed analysis of the expression of genes that showed altered DNA methylation. Remarkably, expression of a few of these genes explained a large part of the individual variation in energy expenditure in adult mice, providing strong support for the hypothesis that early postnatal establishment of epigenetic regulation in the hypothalamus is an important determinant of adult risk of obesity. Our findings (Li-Kohorst, et al.) were published in the August 2013 issue of Diabetes.
For subobjective 3A, as reported in last year's annual report, we found that the effect of maternal obesity on offspring obesity (in the Avy/a mouse model) occurs during fetal development. Moreover, we have shown that this effect is specific to female offspring, and is caused by a persistent down-regulation of spontaneous physical activity. These findings are now in press in The International Journal of Obesity. Additionally, we have isolated neuronal and glial DNA from hypothalamus of wild-type offspring born to lean (a/a) and obese (Avy/a) mothers, and performed DNA methylation profiling. This approach identified a number of interesting gene regions showing cell type-specific evidence of persistently altered DNA methylation. We are currently working on validating some of these by bisulfite pyrosequencing. If these can be validated, these will provide a potential mechanism to explain how in utero exposure to maternal obesity induces persistent changes in body weight regulation.
In subobjective 4A, we have continued work to confirm methylation analysis data of term placentas compared to typical molar pregnancies (known as androgenetic hydatidiform moles; HM) and to determine whether they are expressed in human placentas in an imprinted manner (meaning preferentially from the copy of the gene inherited from the mother or the father). For this we collected a set of placentas for which we also have DNA from the parents of the pregnancy. We are testing imprinted expression of several genes using polymorphic markers that can differentiate which parent contributed the expressed gene. These experiments are still in progress.
For subobjective 4B, we have found several interacting partners of the genes NLRP7, and C6orf221 (a.k.a KHDC3L), the second gene implicated in a special form of molar pregnancies, biparentally inherited hydatidiform moles (HM). We found that these proteins bind to each other and that NLRP7 binds to different factors that regulate imprinting. We combined this data with data from methylation arrays on human embryonic stem (ES) cells in which we inactivated NLRP7. We found many methylation changes in these cells and identified changes in how these cells can be forced to develop into trophoblast. Although part of this work was not supported by ARS, the combined data are critical for understanding regulation of imprinting in the placenta and have been submitted for publication. A revision of the paper is under re-review in the journal Human Molecular Genetics. Our FY2013 milestone related to subobjective 5A was met in a previous year; however, the data obtained for this goal and new published literature motivated us to explore other aspects of the effect of maternal low protein diet on the offspring. We found altered body composition (lean mass vs. fat distribution), and we investigated behavior and found evidence of increased anxiety in offspring exposed prenatally to a maternal low protein diet. We also found altered sleep patterns and are now studying if these could reflect depression or reflect circadian rhythm abnormalities (abnormalities in the daily wake-sleep pattern) in the offspring.
In subobjectives 6A and 6B we have completed an integrated analyses based on genome-wide profiling of both DNA methylation and gene expression changes in human embryonic stem cells before and after induced differentiation (a process by which a less specialized cell becomes a more specialized cell type), and published the results (Yu et al. 2013 Mol Cell Biol). Our data demonstrate that DNA methylation plays a direct role in regulating gene expression during development. Specifically, we discovered certain genomic regions called (CpG islands) that gain methylation after embryonic stem cells differentiate. Both global gene expression profiling and quantitative gene expression validation indicated opposing effects of CpG island methylation in transcriptional regulation of developmental genes, with promoter methylation repressing and 3' (located at the end of gene) methylation activating gene expression. By studying diverse human tissues and mouse models, we further confirmed that developmentally programmed 3' CpG island methylation confers tissue and cell-type specific gene activation. Importantly, our functional assays provided evidence that 3' CpG island methylation regulates gene activation via a mechanism that affects the interaction with gene enhancers. These findings expand the classic view of mammalian DNA methylation as a mechanism for transcriptional silencing, and indicate a functional role for 3' CpG island methylation in developmental gene regulation.
In subobjective 7A, we are developing a novel epigenetic mouse model to study the functional role of DNA methylation. In the epigenetic field, a major obstacle is the lack of experimental systems to directly test epigenetic causality and functional significance. Genetic engineering has become the "gold standard" tool for functional studies of gene mutations. Our goal is to develop an analogous "epigenetic engineering". Our approach builds upon our previous work which identified a specific cis-element (a region of DNA or RNA that regulates the expression of genes located on that same molecule of DNA) associated with programmed promoter DNA methylation during normal development. In mice, we have successfully introduced targeted delivery of DNA methylation at a tumor-suppressor gene called p16. Our preliminary data show that targeted promoter methylation is sufficient to induce gene silencing and cause tumors in mice. Currently we are confirming these results in a large cohort of animals. If successful, this will be the first epigenetic mouse model to demonstrate the functional role of aberrant DNA methylation in cancer. This contribution is significant because it is the first step in a continuum of research that is expected to lead to a broadly used strategy to advance our understanding of epigenetic causality of human diseases in general. In addition, the model systems developed will be enormously useful in determining whether nutrition and dietary supplement exacerbate the tendency for hypermethylation and consequent cancer susceptibility.
NLRP7 and genomic imprinting. NLRP7, a protein that is expressed in germ cells, which are key for reproduction, and the early embryo causes abnormal imprinting in the placenta. Using various types of cells, scientists at the Children's Nutrition Research Center in Houston, Texas, found that NLRP7 affects DNA methylation at multiple genes. Using a model system, we also found that NLRP7 plays a role in the earliest stages of development of the placenta and affects methylation at several genes. These findings are important and provide a new understanding about imprinting regulation in the placenta, which is secondarily relevant to understanding how placental genes regulate nutrient transfer to the fetus.