The Big Picture: Development, Epigenetics, and the "Diabesity" Epidemic

Web Sites

Centers for Disease Control and Prevention
Information on childhood overweight, with links to more resources.

The Future of Children
Free text version of a spring 2006 special issue on childhood obesity, with special attention to how a child's environment can exacerbate or intervene against obesity.

KidsNutrition.org
This site contains information about the Children's Nutrition Research Center at Baylor College of Medicine, including an archive of facts and answers on many topics in childhood nutrition.

Mayo Clinic
Clinical information for parents concerning childhood obesity.

MedLine Plus
Collection of links to more information about childhood obesity.

Shape Up America!
An initiative launched by former Surgeon General C. Everett Koop to educate the public on the importance of maintaining healthy weight. The Web site includes annual reports, information on childhood obesity, and proceedings of a 2003 national conference on diabesity.

Women, Infants, and Children (WIC)
This program of the USDA Food & Nutrition Service provides federal grants to states for supplemental foods, health care referrals, and nutrition education for low-income pregnant, breastfeeding, and non-breastfeeding postpartum women, and to infants and children up to age five who are found to be at nutritional risk. See their information on breastfeeding promotion and support.

Articles

Epigenetics and Diabesity

Shen L, Waterland RA. 2007. Methods of DNA methylation analysis. Curr. Opin. Clin. Nutr. Metab. Care 10: 576-581.

Waterland RA. 2006. Epigenetic mechanisms and gastrointestinal development. J. Pediatr. 149(5 Suppl): S137-142.

Waterland RA, Michels KB. 2007. Epigenetic epidemiology of the developmental origins hypothesis. Annu. Rev. Nutr. 27: 363-388.

Waterland RA, Travisano M, Tahiliani KG. 2007. Diet-induced hypermethylation at agouti viable yellow is not inherited transgenerationally through the female. FASEB J. 21: 3380-3385.

Robert Whitaker

Baughcum AE, Burklow KA, Deeks CM, et al. 1998. Maternal feeding practices and childhood obesity: a focus group study of low-income mothers. Arch. Pediatr. Adolesc. Med. 152: 1010-1014. Full Text

Baughcum AE, Chamberlin LA, Deeks CM, et al. 2000. Maternal perceptions of overweight preschool children. Pediatrics 106: 1380-1386. Full Text

Burdette HL, Whitaker RC. 2005. Resurrecting free play in young children: looking beyond fitness and fatness to attention, affiliation, and affect. Arch. Pediatr. Adolesc. Med. 159: 46-50. Full Text

Chamberlin LA, Sherman SN, Jain A, et al. 2002. The challenge of preventing and treating obesity in low-income preschool children: perceptions of WIC health professionals. Arch. Pediatr. Adolesc. Med. 156: 662-668. Full Text

Jain A, Sherman SN, Chamberlin LA, et al. 2001. Why don't low-income mothers worry about their preschoolers being overweight? Pediatrics 107: 1138-1146. Full Text

Whitaker RC. 2004. Predicting preschooler obesity at birth: the role of maternal obesity in early pregnancy. Pediatrics 114: E29-E36. Full Text

Whitaker RC, Dietz WH. 1998. Role of the prenatal environment in the development of obesity. J. Pediatr. 132: 768-776.

Whitaker RC, Wright JA, Pepe MS, et al. 1997. Predicting obesity in young adulthood from childhood and parental obesity. N. Engl. J. Med. 337: 869-873. Full Text

Rudolph Leibel

Leibel RL. 2006. The molecular genetics of the melanocortin pathway and energy homeostasis. Cell Metab. 3: 79-81.

Rosenbaum M, Goldsmith R, Bloomfield D, et al. 2005. Low-dose leptin reverses skeletal muscle, autonomic, and neuroendocrine adaptations to maintenance of reduced weight. J. Clin. Invest. 115: 3579-3586. Full Text

Rosenbaum M, Sy M, Pavlovich K, et al. 2008. Leptin reverses weight loss-induced changes in regional neural activity responses to visual food stimuli. J. Clin. Invest. 118: 2583-2591. Full Text

Stratigopoulos G, Padilla SL, LeDuc CA, et al. 2008. Regulation of Fto/Ftm gene expression in mice and humans. Am. J. Physiol. Regul. Integr. Comp. Physiol. 294: R1185-R1196.

Weisberg SP, Hunter D, Huber R, et al. 2005. CCR2 modulates inflammatory and metabolic effects of high-fat feeding. J. Clin. Invest. 116: 115-124. Full Text

Patrick Catalano

Catalano PM. 2007. Increasing maternal obesity and weight gain during pregnancy: the obstetric problems of plentitude. Obstet Gynecol. 110: 743-744.

HAPO Study Cooperative Research Group, Metzger BE, Lowe LP, et al. 2008. Hyperglycemia and adverse pregnancy outcomes. N. Engl. J. Med. 358: 1991-2002.

Kitzmiller JL, Block JM, Brown FM, et al. 2008. Managing preexisting diabetes for pregnancy: summary of evidence and consensus recommendations for care. Diabetes Care 31: 1060-1079.

Lain KY, Catalano PM. 2007. Metabolic changes in pregnancy. Clin. Obstet. Gynecol. 50: 938-948.

John Kral

Kral JG. 2007. A stitch in time versus a life in misery. Surg. Obes. Relat. Dis. 3: 2-5.

Kral JG. 2004. Preventing and treating obesity in girls and young women to curb the epidemic. Obes. Res. 12: 1539-1546.

Kral JG, Biron S, Simard S, et al. 2006. Large maternal weight loss from obesity surgery prevents transmission of obesity to children who were followed for 2 to 18 years. Pediatrics 118: e1644-1649. Full Text

Kral JG, Näslund E. 2007. Surgical treatment of obesity. Nat. Clin. Pract. Endocrinol. Metab. 3: 574-583.

Andreas Plagemann

Fahrenkrog S, Harder T, Stolaczyk E, et al. 2004. Cross-fostering to diabetic rat dams affects early development of mediobasal hypothalamic nuclei regulating food-intake, body weight, and metabolism. J. Nutr. 134: 648-654. Full Text

Harder T, Bergmann RL, Kallischnigg G, Plagemann A. 2005. Duration of breastfeeding and risk of overweight: a meta-analysis. Am. J. Epidemiol. 162: 397-403. Full Text

Plagemann A. 2008. A matter of insulin: developmental programming of body weight regulation. J. Matern. Fetal Neonatal Med. 21: 143-148.

Plagemann A, Harder T. 2005. Breastfeeding and the risk of obesity and related metabolic diseases in the child. Metab. Syndr. Relat. Disord. 3: 222-232.

Plagemann A, Harder T, Dudenhausen JW. 2008. The diabetic pregnancy, macrosomia, and perinatal nutritional programming. Nestle Nutr. Workshop Ser. Pediatr. Program. 61: 91-102.

Plagemann A, Harder T, Franke K, Kohlhoff R. 2002. Long-term impact of neonatal breast feeding on body weight and glucose tolerance in children of diabetic mothers. Diabetes Care 25: 16-22. Full Text

Rodekamp E, Harder T, Kohlhoff R, et al. 2006. Impact of breast-feeding on psychomotor and neuropsychological development in children of diabetic mothers: role of the late neonatal period. J. Perinat. Med. 34: 490-496.

Tanja Kral

Faith MS, Storey M, Kral TV, Pietrobelli A. 2008. The feeding demands questionnaire: assessment of parental demand cognitions concerning parent-child feeding relations. J. Am. Diet. Assoc. 108: 624-630.

Fisher JO, Kral TV. 2008. Super-size me: portion size effects on young children's eating. Physiol. Behav. 94: 39-47. Epub 2007 Nov 22.

Kral TV, Berkowitz RI, Stunkard AJ, et al. 2007. Dietary energy density increases during early childhood irrespective of familial predisposition to obesity: results from a prospective cohort study. Int. J. Obes. 31: 1061-1067.

Kral TV, Faith MS. 2008. Influences on child eating and weight development from a behavioral genetics perspective. J. Pediatr. Psychol. [Epub ahead of print]

Lowe MR, Kral TV. 2005. Stress-induced eating in restrained eaters may not be caused by stress or restraint. Appetite 46: 16-21.

Barry Levin

Bouret SG, Gorski JN, Patterson CM, et al. 2008. Hypothalamic neural projections are permanently disrupted in diet-induced obese rats. Cell Metab. 7: 179-185.

Levin BE. 2007. Neuronal glucose sensing: still a physiological orphan? Cell Metab. 6: 252-254.

Levin BE. 2007. Why some of us get fat and what we can do about it. J. Physiol. 583 (Pt 2): 425-430.

Patterson CM, Dunn-Meynell AA, Levin BE. Three weeks of early-onset exercise prolongs obesity resistance in DIO rats after exercise cessation. Am. J. Physiol. Regul. Integr. Comp. Physiol. 294: R290-301.

Patterson CM, Levin BE. 2008. Role of exercise in the central regulation of energy homeostasis and in the prevention of obesity. Neuroendocrinology 87: 65-70.

Sebastien Bouret

Bouret SG, Draper SJ, Simerly RB. 2004. Trophic action of leptin on hypothalamic neurons that regulate feeding. Science 304: 63-64.

Bouret SG, Gorski JN, Patterson CM, et al. 2008. Hypothalamic neural projections are permanently disrupted in diet-induced obese rats. Cell Metab. 7: 179-185.

Bouret SG, Simerly RB. 2007. Development of leptin-sensitive circuits. J. Neuroendocrinol. 19: 575-582. Review.

Bouret SG, Simerly RB. 2006. Developmental programming of hypothalamic feeding circuits. Clin. Genet. 70: 295-301.

Peter Nathanielsz

Hsieh AT, Anthony JC, Diersen-Schade DA, et al. 2007. The influence of moderate and high dietary long chain polyunsaturated fatty acids (LCPUFA) on baboon neonate tissue fatty acids. Pediatr. Res. 61(5 Pt 1): 537-545.

Kaufman D, Banerji MA, Shorman I, et al. 2007. Early-life stress and the development of obesity and insulin resistance in juvenile bonnet macaques. Diabetes 56: 1382-1386. Full Text

Nathanielsz PW, Poston L, Taylor PD. 2007. In utero exposure to maternal obesity and diabetes: animal models that identify and characterize implications for future health. Clin. Perinatol. 34: 515-526.

Jeremy Coplan

Gohil BC, Rosenblum LA, Coplan JD, Kral JG. 2001. Hypothalamic-pituitary-adrenal axis function and the metabolic syndrome X of obesity. CNS Spectr. 6: 581-586, 589.

Kaufman D, Smith EL, Gohil BC, et al. Early appearance of the metabolic syndrome in socially reared bonnet macaques. J. Clin. Endocrinol. Metab. 90: 404-408. Full Text

Speaker:
Robert Waterland, Baylor College of Medicine and USDA Children's Nutrition Research Center

Highlights

• Epigenetic processes are emerging as major factors in obesity, diabetes, and heart disease; these effects are not caused by genetic mutations, but are still maintained as our cells divide.
• Even genetically identical animals and humans can show differences in character, appearance, and physiology.
• Early nutrition might play a critical role in disease susceptibility; nutrients influence DNA methylation, thereby influencing gene activation and silencing.
• The Agouti viable yellow (A^vy) mouse, which naturally overeats, is a good model for investigating epigenetic effects on obesity.
• New technologies, such as a methylation-specific amplification microarray that amplifies hypermethylated regions of the genome, will be useful in identifying epigenetic modifications associated with obesity and other disorders.

Genes aren't everything

Obesity has been increasing steadily in most parts of the world over the past 30 years, and this rise is associated with a concomitant increase in prevalence of type 2 diabetes, dyslipidemia, and hypertension. What's behind this surge in adiposity? Why are we now in the midst of an obesity "epidemic?" It's logical to blame increasingly sedentary lifestyles and the wide availability of palatable foods. But why do some people succumb to these environmental factors by becoming fat whereas others remain lean? Does the answer lie in our genes?

In his keynote address, Robert Waterland of Baylor College of Medicine suggested that epigenetics, specifically the range of environmental influences on genetic expression during early development, may account for these differences.

"The best example of epigenetics at work is tissue-specific gene expression," Waterland explained. "Most cells in our body contain the same complement of DNA—the entire human genome. But our hepatocytes express a very different subset of genes from the cells in, for example, our colonic mucosa. And although those two cell types turn over throughout life, hepatocytes remember to express liver-specific genes; the pattern is established during early development and maintained throughout life through epigenetic mechanisms."

Waterland and others hypothesize that during critical periods of development, nutrition and other environmental stimuli can perturb developmental pathways, thereby leading to permanent changes in gene expression, metabolism, and chronic disease susceptibility. He and a colleague coined the term "metabolic imprinting" to denote a subset of adaptive responses to early nutrition characterized by a persistent effect lasting into adulthood, and with susceptibility limited to a critical period of development.

Rainbow and his clone, CC, share the same genes but look very different.

The 2002 cloning of a cat named Rainbow shows the importance of epigenetics, Waterland continued. One might reasonably expect a clone to be an exact duplicate of the original, but that's not the case. Although Rainbow and his clone, CC (nickname for Copy Cat) are genetically identical, they have very different phenotypes due to developmental epigenetics. Likewise, human monozygotic (genetically identical) twins are often phenotypically different; different mutations, stochastic development, various imprinting effects, differing environmental and nutritional inputs while in the uterus, and different societal inputs will all interact to cause dissimilarities between the two.

Early nutritional influences

Epigenetic processes work synergistically to regulate both patterns of gene expression and the ability of different cell types to respond appropriately to endogenous signals such as nutrients and hormones, Waterland said. Among these processes, DNA methylation is the best characterized, and it is also directly influenced by diet. Therefore, Waterland and his group are focusing on this area in their quest to understand some of the mechanisms underlying obesity and diabetes.

Methyl (CH₃) groups influence whether genes are switched on or off. In humans, most cytosines within CpG dinucleotides (regions of DNA where a cytosine nucleotide occurs next to a guanine nucleotide) are methylated. Tissue-specific patterns of CpG methylation are established during early development.

The methylation patterns in sperm and egg genomes are largely erased shortly after fertilization in the early embryo, Waterland explained. This loss of epigenetic information coincides with the totipotency of the early embryo. During differentiation these patterns are restored in a cell type-specific way. This loss and subsequent restoration of epigenetic information includes a number of potential "critical windows" when environment could affect these processes.

"We might think, intuitively, that excesses or deficiencies of nutrients could affect the establishment of DNA methylation."

What does nutrition have to do with it? DNA methylation requires dietary donors and cofactors such as folate, vitamin B6, vitamin B12, methionine, and choline. These nutrients play critical roles in ensuring the flow of methyl donors into the DNA methylation pathway. "So we might think, intuitively, that excesses or deficiencies of some of these nutrients could affect the establishment of DNA methylation," Waterland observed.

To better understand the influences of early nutrition on obesity and other disorders, Waterland and his colleagues are working with an Agouti viable yellow (A^vy) mouse model. The agouti gene encodes a signaling molecule that regulates the formation of a yellow pigment. The molecule is normally expressed only in mouse hair follicles during a specific stage of hair growth, causing a yellow band on otherwise black hair and ultimately resulting in the brown (agouti) coat color in normal mice.

Insertion of a promoter from a transposable element into the agouti gene causes both genetic and epigenetic dysregulation in A^vy mice. Instead of being expressed in hair follicles, the promoter causes the agouti gene to be expressed all over the body all the time, leading to a yellow coat color and other effects, including hyperphagia and obesity, Waterland explained. From an epigenetic standpoint, methylation of CpG sites in the transposable element insertion site becomes entirely random. An A^vy mother might produce a litter in which some animals have a very low level of methylation at specific CpG sites, and thus the yellow coat color; other mice with high levels of methylation that silences the promoter and leads to a brown coat color; and still others expressing phenotypes in between.

In a landmark study in 2003, it was shown that giving a nutritional supplement to female A^vy mice induced hypermethylation at the A^vy locus, resulting in a permanent coat color shift to the brown phenotype in heterozygous (A^vy/a) offspring. The study was "the first clear demonstration that a transient nutritional exposure during a critical period of development could lead to a permanent change in phenotype due to an epigenetic mechanism," Waterland said. The findings are relevant to humans because about half the human genome is made of transposable elements, which could create metastable "epi-alleles" like the A^vy locus if the elements, which are usually silenced by methylation or other factors, regulate expression of neighboring genes in some instances. In turn, these loci might respond in a similarly labile way to early nutritional inputs.

The obesity connection

Dysregulation in epigenetic pathways can lead to obesity, Waterland continued. This has been documented in cloned mice, for example. When mice are successfully cloned from somatic cells of adult animals, the newborns often are normal weight, but go on to develop adult-onset obesity and diabetes. Although the mechanism by which this occurs is not yet understood, he explained, "it is clearly an epigenetic phenomenon" because the act of cloning doesn't change the mouse genotype. Therefore, something in the cloning process must induce epigenetic alterations in the cloned animals.

The agouti A^vy mice described earlier, which ubiquitously express the agouti protein, are another example. These mice are hyperphagic and obese because systemic expression of agouti protein impairs the satiety mechanism. A^vy/a mice who are hypermethylated at A^vy are protected from obesity—a clear example of epigenetics affecting obesity.

These A^vy/a sisters are genetically identical. The one on the right is hypermethylated at A^vy, whereas the one on the left is hypomethylated at A^vy.

In humans, some cases of Prader-Willi syndrome—a genetic syndrome that causes hyperphagia, hypogonadism, short stature, small hands and feet, and obesity—are caused by epigenetic variation. Although deletion of genes in a region of chromosome 15 most often causes the disease, in many instances it occurs because of maternal uniparental disomy 15 (inheriting both copies of chromosome 15 from the mother), resulting in inappropriate epigenetic silencing of the Prader-Willi syndrome region on both copies of chromosome 15.

Waterland wondered whether maternal obesity before and during pregnancy might cause some kind of "feed-forward chain" in a child's susceptibility, such that maternal obesity might lead to the enhanced possibility of children becoming obese via epigenetic mechanisms.

To test this hypothesis, the team turned to its hyperphagic agouti A^vy mice. They compared one group of pregnant mice given a supplement that enhanced methylation to a group that did not receive supplements. A series of investigations revealed a positive correlation between maternal adult body weight and offspring body weight in unsupplemented mice but not in the mice who received the supplement.

However, only newborn brown mice were totally protected from obesity; all the other coat colors experienced some degree of adiposity. But in every coat color class, mice whose mothers had received supplements tended to be a bit lighter than the unsupplemented mice. These and other findings suggested an interaction among genotype, epigenotype, and diet that influenced weight and adiposity, Waterland said.

Probing further

Waterland's team is now working to identify specific epigenetic modifications associated with obesity. To do this, they developed a methylation-specific amplification microarray that amplifies hypermethylated regions of the genome. They validated the microarray in a mouse model, showing that it is indeed capable of identifying relatively subtle changes in locus-specific CpG methylation in normal tissues, and that therefore it is a useful tool to study epigenetic alterations associated with obesity as well as environmental influences on these processes.

Investigations detected "an epigenetic storm" in the hypothalamus during the newborn and weaning period.

Using the microarray, they examined the hypothalamus of newborn and weanling non-agouti mice. Genetic mutations in leptin, melanocortin 4 receptor, neuropeptide Y, and other genes involved in hypothalamic regulation of energy balance all lead to obesity; however, "We know almost nothing about epigenetic regulation at these loci," Waterland stated.

Their investigations yielded a surprise, he said: unlike in the liver, where the team found very few epigenetic changes during the newborn and weaning period, they found "huge changes—an epigenetic storm" in the hypothalamus. Four-fold increases or decreases in methylation were seen during the 21-day period.

"It's not clear what it all means at this point, but this is where we're going," he concluded.

Speakers:
Rudolph Leibel, Columbia University
Patrick Catalano, Case Western Reserve University
Andreas Plagemann, University of Medicine Berlin
Robert Whitaker, Temple University
Tanja Kral, University of Pennsylvania
John Kral, SUNY Downstate Medical Center

Highlights

• Genetic and epigenetic factors render an individual more or less vulnerable to obesity and its many social and health consequences.
• Maternal obesity and gestational diabetes are dangerous for mother and child. Despite normal weight, newborns of obese mothers have increased body fat, and risk becoming overweight.
• Obesity surgery is safe and effective. Young women unable to reduce weight should consider surgical weight loss before becoming pregnant to have a healthier pregnancy.
• Breastfeeding protects against development of obesity and related metabolic diseases and improves brain development. Milk from diabetic mothers is healthier than commercial formulae.
• Eating in the absence of hunger is a heritable trait that may promote obesity in children. Parents' improper monitoring of children's eating can increase the risk of chronic over-eating.
• Behavioral strategies to reduce childhood obesity include regular family meals, establishing sleep routines, and increasing unstructured outdoor play. Clinicians should work with parents in light of their values to achieve these goals.

The "big" picture

Epigenetics suggests that nutrition and other environmental factors present during very early development can strongly influence an individual's susceptibility to diseases such as obesity and diabetes. Thus, predisposing factors passed through the generations—for example, diet and pollution— can influence an individual's health.

Genetic, epigenetic, environmental, and behavioral factors act in concert to determine vulnerability to obesity and its consequences.

But figuring out which influences exert the most powerful effects, and how they interact within an individual, is extremely complicated. At least 50 genes have been implicated in the development of obesity; understanding how each of them is influenced by environmental factors, and by each other, is a whole area of investigation unto itself. Another is identifying specific nutritional factors in the mother, for example, that influence the pre-conception environment, the gestational period, and, of course, postnatal environment when the child is exposed to the mother's feeding practices.

The speakers in this session tackled various aspects of this complex picture, showing how genetic and epigenetic factors, as well as environmental influences and behavior, all act in concert to render an individual more or less vulnerable to obesity and its consequences.

Where genes fit in

Rudolph Leibel of Columbia University is working to identify key molecular genetic aspects of obesity and type 2 diabetes. Both are complex disorders with strong genetic predispositions, he said. His group and others have concluded that "genes predisposing to obesity act through a permissive environment to produce an obese individual. Obesity interacts primarily with a largely different set of genes that predispose to type 2 diabetes, probably through β-cell biology. That is, obesity per se imposes a stress on β-cells [in the pancreas], leading to type 2 diabetes in genetically predisposed individuals." Thus, the two disorders are "locked together," with the result that 80% of type 2 diabetic individuals are obese and 50% of obese individuals are diabetic.

But that general understanding of how the two disorders interact is only the beginning. Researchers still don't know how the 50-odd genes believed to influence relative fatness interact with each other to mediate susceptibility; how various cell types and genes that respond to leptin—a key regulator of energy homeostasis—and other factors exert effects on food intake and energy expenditure; and how environmental and stochastic factors come into play. Compounding the challenge is the fact that different genes seem to mediate susceptibility to obesity in different racial/ethnic groups, and genes that influence adiposity may differ between children and adults. The most common variants in these genes are likely to be non-coding, subtly affecting levels of transcripts, rather than nonsense and missense mutations.

The bottom line is that an individual's weight or body fat content really represents "just the tip of a physiological iceberg below which are a complex series of cellular and molecular and developmental physiologies," Leibel said.

Fortunately, new technologies are helping to identify genes that seem to play pivotal roles in obesity. Recently, a technique that permits investigators to interrogate the entire genome to find associations between genetic markers and phenotypes of interest—in this case, obesity and diabetes—is yielding some fruit. Several large, whole-genome association studies have linked a region of chromosome 16 containing part of the FTO gene with high body mass index; FTO is also a demethylase, and so could affect genetic imprinting.

A neighboring gene, FTM, encodes a ciliary protein that may be co-regulated with FTO. Ciliary proteins are disrupted in Bardet-Biedl syndrome that includes obesity.

Further investigation revealed that CUTL1—a transcription factor that regulates expression of both FTO and FTM, thereby possibly affecting energy metabolism—binds to non-coding DNA in the FTO gene at a site with sequence variants that may affect risk of obesity. But it's not yet known whether these alleles affect obesity in humans due to effects on FTO or FTM, or both—or to other genes in the vicinity. Future work in this area will clarify the connection. Meanwhile, this new genomic strategy is also likely to reveal many genes that were previously unrecognized as playing a role in obesity or diabetes.

Weighty issues

Patrick Catalano of Case Western Reserve University addressed yet another aspect of the diabesity picture: the consequences of maternal obesity on the mother, the developing fetus, and the child. Until recently, most research on fetal origins of adult disease focused on the growth-restricted fetus; however, mounting recent evidence shows significant increased risks for macrosomic (greater than 90th percentile for gestational age and gender, or 9 1bs, 15 ounces) offspring, as well.

The Institute of Medicine recommendations for weight gain in pregnancy were published almost 20 years ago. The original criteria were developed to help decrease the risk of the small or growth-restricted fetus. Because of the increase in prevalence of overweight and obesity in the population and an increase in birth weights, the criteria for weight gain recommendations in pregnancy are being re-examined. The average weight of women at the time of delivery is significantly higher in many parts of the world than it was 15 or 20 years ago, corresponding to concomitant increases in weight among children and adolescents.

Catalano's work is mainly concerned with body composition in pregnant women and their offspring. He explained that at the time of delivery, the newborn's body is composed of about 12%–15% fat mass and 85%–88% fat-free or lean mass. Fat-free mass may represent genetic growth, whereas fat mass represents the influence of other intrauterine factors, he postulated. In a series of studies, his group found that the strongest correlate for fatness in a newborn at birth is the mother's pregravid metabolic status; for example, the degree of insulin resistance related to obesity and gestational diabetes. The more insulin resistant the mother is, the greater the adiposity in the offspring, even if the offspring is normal weight.

Other studies suggested that maternal pregravid BMI is the strongest correlate for macrosomia defined as fat mass or adiposity at birth. The newborns of overweight/obese mothers had larger babies because of increase in fat mass and not lean body mass.

Finally, follow-up studies in offspring showed that the risk of a child having metabolic syndrome at age 10 was 50% if the child was born large for gestational age and his or her mother had gestational diabetes, compared with 20% if the child's mother did not have gestational diabetes. If the child was born at an appropriate weight for gestational age, the risk of metabolic syndrome was 20% if the mother had gestational diabetes.

Taken together, these findings provide evidence of both short- and long-term effects of maternal obesity on the intrauterine environment as it relates to fetal growth, neonatal body composition, and metabolic dysfunction in late childhood.

The kindest cut?

What to do about obesity and excess weight gain during pregnancy? John Kral of SUNY Downstate Medical Center showed that most non-surgical interventions aimed at durably reducing or curbing weight—e.g., education, advice about diet and exercise, weight-loss drugs—are largely ineffective. Certainly, trying to intervene before pregnancy, which is the ideal time, won't work when the rate of unplanned pregnancies is 50% nationally and more than 90% in inner-city, poor populations. Moreover, there are very few studies of restricting weight gain during pregnancy, he noted, because until recently, it simply wasn't done—it was unthinkable in times of global hunger. Thus, there were no studies following offspring of voluntarily weight-restricted mothers over time.

In effect, interventions that have been shown to offer the greatest potential benefits are those with the highest associated risks and costs—drugs and various types of surgery. This creates a "therapeutic dilemma," Kral acknowledged. Nonetheless, he believes antiobesity surgery in young women can be an effective preventive strategy. Despite adverse effects that may include, among others, thrombo-embolism, bleeding, stenosis, wound infection, and long-term vitamin and mineral deficiencies, surgery effectively lowers a mother's body mass index and virtually cures type 2 diabetes and prevents gestational diabetes, hypertension of pregnancy, and pre-eclampsia. Importantly, it also more than halves the risk of having an obese child (from 27% to 10%).

Kral's work has shown that offspring of obese mothers who underwent bariatric surgery were more likely to be normal weight and less likely to become obese in their teens. Follow-up studies into adolescence suggest that pre-conception surgery overrides both the obese genotype and the toxic home by normalizing the hypercaloric intra-uterine environment of chronic maternal overnutrition.

A closer look at breast milk

If the intrauterine environment can promote obesity in offspring, what about breast milk? Breast milk arguably contains a host of nutrients from the mother, and breastfeeding has long been considered the best way to nurture newborns. That is still the case, according to Andreas Plagemann of the University of Medicine Berlin. However, recent evidence from his group and others suggests that breast milk from mothers who are obese and diabetic may not be quite as beneficial.

Plagemann began by reviewing the evidence for positive effects of breastfeeding. He observed that it has been shown to protect against atopic dermatitis, respiratory infections and asthma in childhood, type 1 and type 2 diabetes, and hypertension. It also promotes emotional, psychomotor, and cognitive development. Importantly for this talk, breastfeeding has also been shown to lower the risk of obesity in later life by as much as one third.

That said, most studies of the long-term effects of breastfeeding were performed in healthy women; none considered what might happen if the mother was affected by a metabolic disease during lactation, Plagemann noted. So his group, in the Kaulsdorf Cohort Study, did just that.They knew that breast milk composition in mothers with diabetes has increased glucose concentrations, increased energy content, and increased insulin concentrations. Might this affect neonates? Preliminary evidence from the study suggests that, indeed, the neonates' risk of becoming overweight and having impaired glucose tolerance in childhood may be increased, rather than decreased.

Subsequent studies in a rat model of diabesity showed a similar increased risk to offspring when dams were obese and diabetic. In addition, the young rats showed impairments in the hypothalamic circuits that regulate food intake, body weight, and metabolism, raising the possibility that the offspring had been "programmed" to become obese and/or diabetic as a result of breastfeeding.

Taken together, the findings point to the need for rigorous studies to clarify whether breastfeeding may have negative consequences for the offspring of mothers with diabetes, Plagemann urged. In addition, it's important to determine whether concentrations of insulin, leptin, and other hormones in breast milk play a "programming" role in the offspring's brain that forms the basis for deleterious effects later in life.

Parenting and childhood obesity

Robert Whitaker of Temple University proposed behavioral strategies to help break the cycle of obesity "transmission" from one generation to the next. He noted that 40% of newborns who are large for their gestational age and who have at least one obese parent will be obese as young adults, and that by age six, about 70% of overweight children with at least one obese parent will be obese as young adults. Although genetic and epigenetic factors may play important roles, Whitaker is convinced that influencing a child's environment through lifestyle changes can still make a difference.

Although no socioeconomic or racial group has been left untouched by the spiraling increase in obesity, low-income families seem to have been especially hard hit, Whitaker pointed out. He suggested that the critical period for childhood obesity prevention for this group is from before conception until kindergarten, adding that "families are the social unit with the greatest potential influence in this part of the life cycle." Focusing on that critical period, Whitaker and colleagues have tried to identify effective strategies for helping families.

Interventions during the pre-conception stage could include helping mothers to attain and maintain a healthy weight during pregnancy and to lengthen the time between pregnancies, he suggested. But his work has shown that, other than encouraging and supporting breastfeeding, post-conception strategies for childhood obesity prevention remain a challenge. "One of the struggles is that parents with young children often don't think their children are overweight or feel it's a cause for concern," Whitaker said. "Parents are more concerned with other areas, like their children's cognitive, social, and emotional development." So Whitaker is now exploring a different strategy by targeting household behaviors and routines that can lead to healthy weight for children but which also promote those other developmental outcomes that appear to be more meaningful to parents. These household routines include having regular family meals, establishing sleep routines, and increasing unstructured outdoor play.

If clinicians are attempting to reduce obesity, "is it devious to address a different target?" Whitaker asked. "My position is that we have to meet parents where they are, and we either don't address obesity at all or we address it as we are addressing other issues for children that many parents seem more concerned about. When you try to approach the issue of maternal-child feeding interactions, you're immediately talking about parenting. So, when you bring up the child's weight, parents know that you are talking about how they are parenting. Understandably, parents get defensive." So, instead of focusing on eating and weight, he advised, "form an alliance" with parents about their global goals for their children. Focusing first on the parents' goals—for example, having a happy child who gets along with others—it is still possible to address the long-term health outcomes related obesity.

Eating, but not hungry

What impact does a child's eating behavior have on his propensity to become overweight and stay that way? The answer is not straightforward, according to Tanja Kral of the University of Pennsylvania. The fact that obesity runs in families suggests that shared genes may operate in a particular family environment to promote overweight in children, she acknowledged. But it's also important to consider intermediary behavioral traits that promote increased energy intake and excessive weight gain in children.

Children seem to be born with an innate ability to regulate their energy intake based on their physiological needs, Kral observed. However, over time this ability seems to weaken, as children become more responsive to environmental factors and adhere less to their internal hunger and satiety cues. This could lead to overeating and excessive weight gain.

Eating in the absence of hunger in children may be a "behavioral precursor" for disinhibited eating during adulthood.

A series of competing models exist which can be used to test how genetic and non-genetic factors can shape an eating trait and weight gain. One model posits that that the correlation between an eating trait and weight gain is due entirely to genetic pathways that influence both traits. Another model posits that the correlation is due entirely to environmental pathways which impact both traits. But both genetic and environmental pathways are likely at play, said Kral. For example, a child's genotype has been shown to influence his caloric compensation ability, but so, too, have environmental factors, one of them being the use of overly controlling feeding practices.

Eating in the absence of hunger (EAH) is an example of a behavior that is likely to be influenced by genetic and environmental factors. It's a behavioral trait that refers to children's susceptibility to eating when satiated in response to the presence of palatable foods, Kral explained. The behavior shares behavioral characteristics with dietary "disinhibition" in adults, and is a risk factor for overeating and excessive weight gain in children. Some evidence suggests that disinhibition is heritable and that sequence variations in candidate genes such as GAD2 and NMB are contributing to its development.

However, the risk of developing EAH is also increased in children whose parents use overly restrictive feeding practices and increased monitoring. The finding that EAH is relatively stable throughout childhood suggests that EAH may be a "behavioral precursor" for disinhibited eating during adulthood.

Can EAH be abrogated, thereby possibly reducing both the risk of overeating and excessive weight gain during childhood? Possibly, Kral said. Additional research is needed to determine whether it is possible to teach children who exhibit EAH to only eat in response to internal hunger and satiety cues, thereby re-establishing normal eating regulation; the extent to which the association between eating traits and excessive weight gain in children is mediated by genetic factors, environmental factors, or both; whether dietary interventions may need to be tailored to a child's familial predisposition to obesity, since the child's response to these interventions may depend on his genotype; and whether, as Whitaker suggested, parents' involvement in feeding practices and their structuring of the home food environment will be critical components in such interventions.

Speakers:
Barry Levin, University of Medicine and Dentistry of New Jersey
Sebastien Bouret, University of Southern California
Peter Nathanielsz, University of Texas Health Science Center
Jeremy Coplan, SUNY Downstate Medical Center

Highlights

• Work in animals is shedding light on biological processes that underlie gene–environment interactions leading to diabesity.
• Studies in offspring of obesity-prone and obesity-resistant animals show that the tenacity of the obesity-prone phenotype from the early postnatal period through adulthood is likely due to polygenic inheritance of these traits.
• Perinatal undernutrition and overnutrition during critical developmental periods not only predisposes to diabesity, particularly in genetically susceptible individuals; it also impairs the development of neural circuits in the hypothalamus and other parts of the brain.
• Exercise at an early age is one environmental intervention that abrogates the obesity-prone phenotype in rats. However, it is uncertain whether such early onset exercise might have the same effect in humans.
• Maternal nutrient restriction results in fetal abnormalities of the pancreas, liver, and other organs.
• Stress in early life can lead to early onset of metabolic syndrome and changes in the brain that are associated with elevations in body mass index and plasma cholesterol.

Rat models of obesity

In animal models, researchers have opportunities to explore what happens on a biomolecular basis when genes and the environment interact—in this case, regulation of energy homeostasis, the balance between energy intake and energy expenditure and storage. Barry Levin of the East Orange VA Medical Center and New Jersey Medical School has created a rat model of diet-induced obesity comprising genetic lines of obesity-prone and obesity-resistant animals. His work is providing insights into the processes involved in weight gain and its consequences, particularly in the context of a genetic propensity to obesity.

In a series of experiments, his group made dams obese during pregnancy and looked at the effects in offspring, not only with respect to diabetes and obesity, but also in the brain. "We saw major outcomes," he said. For example, if a rat is obesity-prone and its mother is obese during pregnancy and lactation, that rat will become more obese as an adult than offspring of lean mothers, even if fed a very low-fat diet; by contrast, obesity-resistant rats stay lean even if their mother is obese. However, if a resistant pup is weaned by an obese dam, that resistant pup will become obese. What's behind this variability?

"Our first clue that something was going on with the obesity-prone animals came at about five weeks of age when, fed a 31% fat diet—less fat than the typical American diet—these rats ate more than the obesity-resistant animals right out of the box," Levin explained. After about three days of overeating "something magical happened" [they reached their 'set point'] and both groups of animals downregulated their food intake. But in contrast to their lean counterparts, the obesity-prone animals didn't compensate very well and never came back to their baseline weight, despite a "huge" increase in leptin production, which should have markedly reduced food intake. "Thus, once they're exposed to the 31% fat diet, they're on an upward trajectory and we don't know anything that will reverse it permanently, except illness or GI surgery," Levin said. The fact that the pups overate despite the surge in leptin also suggested that they had inherited leptin resistance from the dams. Further investigation revealed that, indeed, these animals had fewer leptin receptors in the brain's arcuate nucleus even before they became obese.

Having an obese mother during pregnancy and lactation is a double whammy for obesity-prone animals.

In other experiments, the team found that the offspring of the obese dams also had 50% more fat (assessed by measuring fat pads) than their lean counterparts even when they ate a low-fat diet from weaning. This difference was magnified when they ate a high-fat diet. "So, having the proper genetic background—a propensity to obesity—and making the mother obese during pregnancy and lactation is a double whammy for these [obesity-prone] animals," Levin observed. The obese dams' pups also had half the amount of norepinephrine transporters in the hypothalamus compared with obesity-resistant pups; fewer transporters increases norepinephrine levels, which increases food intake.

Levin and colleagues investigated other environmental factors that could influence obesity. They induced stress in pups by separating them from the dam and found that if the pups were left on a low-fat diet, their weight, fat percentage, and leptin levels were normal; but once they were exposed to a high-fat diet, they got fatter. "Postnatal stress has predisposed them to become obese as adults," Levin explained.

Similarly, additional studies of the offspring of resistant pups who were raised by obese dams found that the animals had increased levels of agouti-related peptide (which can increase food intaker and lower energy expenditure), more fat, leptin insensitivity, and a 50% reduction in insulin sensitivity. Thus, "by raising lean animals in obesogenic environment, we can make these animals permanently obese."

Levin doubts that adult obesity can be cured, so his efforts are now focused on prevention. In that regard, his team has found that when obesity-prone animals exercised for at least three weeks in the immediate post-weaning stage of development, they were protected from becoming obese for months after exercise cessation, even when fed a high-fat diet. He speculates that the early exercise may have increased leptin responsiveness, among other factors, thus thwarting the obesity phenotype. Identifying exactly when the animals are vulnerable to an exercise intervention is critical to "making the translational leap" to humans, he concluded.

Neural circuits and obesity

Working in animal models similar to Levin's, Sebastien Bouret of the University of Southern California investigated the biological underpinnings of metabolic imprinting; that is, how the environment affects the development of neural circuits related to energy homeostasis. His team first determined that in normal animals, projections from the arcuate nucleus develop during the first two to three weeks of life, suggesting that this may be the critical period that affects later metabolism. Coinciding with this critical period, on day 16, the team documented a "dramatic surge" in leptin. Bouret suggested that these elevated leptin levels might have a trophic effect on the projections from the arcuate nucleus, noting that the increased leptin did not affect the animals' weight.

Subsequent experiments in leptin-deficient obese mice showed that leptin deficiency led to "dramatic reductions" in projections from the arcuate nucleus, including reductions in length and density. Although the changes seemed permanent, Bouret wondered whether injecting leptin could reverse the abnormalities. Indeed, giving leptin during the early postnatal period did reverse the adverse effects of leptin deficiency; however, injecting adult animals with leptin had no effect. The findings suggested that leptin may act to promote formation of pathways from the arcuate nucleus "exclusively during a restricted critical period in the first two weeks of life," Bouret said.

Under- or overnutrition in the early postnatal period can lead to permanent changes in phenotype.

A series of subsequent experiments in mutant mouse models also revealed that under- or overnutrition in the early postnatal period can also lead to permanent changes in phenotype. By manipulating litter size, the researchers found that undernutrition (large litter) decreases body weight and the size and length of the newborn, whereas overnutrition (small litter) had the opposite effects. Moreover, the changes were enduring; overnourished animals remained overnourished through adulthood, and vice versa. Interestingly, both groups of mice had some neural abnormalities, suggesting that "both low and high leptin levels may be negative," Bouret concluded.

Consequences of nutrient restriction

Peter Nathanielsz of the University of Texas Health Science Center is investigating the effects of nutrition on diabesity from another perspective; that is, how maternal nutrient restriction affects fetal development. In response to questions about the relevance of maternal nutrient restriction at a time when so many mothers are obese, Nathanielsz provided the following data and information:

• In 2002, 34.9 million Americans experienced food insecurity, compared to 31 million in 1999.
• In 2002, 11.1% of U.S. households experienced either food insecurity or hunger; Utah, Texas, Mississippi, Arkansas, New Mexico, and Oklahoma all had rates above 14%.
• The number of hungry people in the U.S. is greater now than when the 1996 World Food Summit set hunger-cutting goals.
• The U.S. lags behind in its pledge to cut the number of Americans living in hunger from 30.4 million to 15.2 million by 2010.
• UNICEF estimates 852 million people worldwide experienced food insecurity in 2004.
• Many teenagers restrict their diet for cosmetic reasons; when they become pregnant, their fetus will be nutrient-deprived.
• In the presence of suboptimal vascular function (e.g., uterine arterial disease and attendant placental insufficiency), the fetus is nutritionally deprived; many women put off pregnancy until well past the optimal age; age adversely affects maternal vascular perfusion.

Nathanielsz works in a baboon model, which may be more informative with respect to humans than rodent and sheep models, he said. Moreover, human reagents generally cross-react with nonhuman primate tissues, enabling the study of key organs related to growth and insulin signaling.

Rodent work revealed that insulin, IGF-1, and IGF-II are potential growth factors in the developing fetal pancreas, Nathanielsz observed. Following maternal nutrient restriction in pregnant rodents, fetal pancreatic growth is decreased, IGF-II and insulin concentrations are decreased, and apoptosis is increased. Similarly, his group's studies in the baboon model showed impaired levels of IGF-I, IGF-II, and insulin in response to a global 30% nutrient restriction in mothers. The team also showed that nutrient restriction decreases maternal weight compared with controls, and that placental weight and volume are significantly reduced at term.

Taken together with the results of other studies in this model, Nathanielsz concluded that the impairments in pancreatic development documented in rodents are also seen in primates, even at a fairly moderate level of nutrient restriction. Moreover, the brain is also affected; the subventricular zone in the fetus of the nutrient-restricted mother is half the thickness of that of controls.

Subsequent studies of the fetal liver showed decreased β1-adrenergic receptor density in response to maternal nutrient restriction, suggesting that such restriction might alter glucose metabolism, resulting in reduced lipolysis and thus predisposing to fatty liver. In addition, data from the kidney, liver, and brain midway through gestation—and the heart at term—showed effects on global methylation, indicating changes in gene expression in these organs. The team is now studying the longer-term effects of maternal nutrient restriction on offspring weight and behavior, Nathanielsz concluded.

Stress and metabolic syndrome

Using another primate model, Bonnet macaques, Jeremy Coplan of SUNY Downstate Medical Center is studying how early stress—and concomitant overaction of the corticotrophin releasing factor/HPA axis system and reduction of neurotrophic gene expression—affects the brains of young animals. Coplan's "variable foraging demand" model leads to unpredictable periods of separation between the mother and her offspring, as well as an unpredictable supply of food.

This "early life adversity" results in a rapid onset of primate metabolic syndrome (visceral adiposity and insulin resistance), as well as decreased metabolites in the left hippocampus (the side that mediates calmness and a positive emotional state), asymmetry in hippocampal volume (the right side, which is involved with threat and danger appraisal, is larger than the left; this also affects plasma lipids and body weight), and decreased neurogenesis in the dentate gyrus.

Taken together, the findings support the concept that early-life adversity can lead to long-term epigenetic effects, resulting in metabolic syndrome and impairments in brain development and function, Coplan concluded.

Moving forward

During the discussion period that followed the formal presentations, participants expressed concerns and offered some potential solutions to the challenges raised by the complex and seemingly tenacious problems of childhood and subsequent adult obesity.

Scientists "have unusual partners in obesity prevention."

An audience member who works at the NYC Department of Health, Dr. Mary Bassett, found the presentations "a little depressing because they suggest we have the equivalent of a contagious condition in the population." The health department intervenes by trying to change the food environment and the care of pregnant women and newborns, but these techniques naturally cannot address genetic propensities, she said. Moreover, she questioned the implication that breastfeeding might be harmful if the mother is obese.

Andreas Plagemann responded that breastfeeding should be promoted "irrespective of a woman's weight or metabolic status." He reiterated that breastfeeding protects children in a variety of ways and emphasized that although benefits may be reduced in diabetic mothers, breast milk has protective effects and is superior to formula feeding. He called for universal screening for gestational diabetes, arguing that with "tight metabolic control, there is a lot of potential for improvement of the perinatal environment." Barry Levin added that although high insulin levels during the third trimester is bad for rats, the same may not necessarily be true for humans.

One of the speakers urged greater education for young people, noting that in the United States "most people are biologically illiterate; they know more about what goes under hood of their car than in their body." Unless women learn more about their health pre-pregnancy, "we've got a huge problem," he said, adding that Sweden and Scandinavian countries provide education on how to become a parent, but that isn't the case here.

There was some discussion about changes underway in the WIC program and Head Start —e.g., improving communication with parents and foods provided to families—that could help curb childhood obesity.

Robert Whitaker concluded by calling for better communication among professionals in diverse disciplines. "We have unusual partners in obesity prevention, and we have to stop operating in a silo way," he said. People who work in WIC and Head Start need to communicate with pediatricians, he suggested, as well as with people working to prevent sexually transmitted diseases and unwanted pregnancies. "They're all important partners."

Epigenetics and Diabesity

How do the epigenome and the genome interact with each other and the environment to cause obesity and other diseases and disorders?

What specific factors may cause hypo- or hypermethylation during early development?

How does epigenetic regulation play out in the hypothalamus, the brain region that regulates energy intake and expenditure?

Diabesity in Humans

How can eating patterns be improved to reduce the risk perpetuating obesity through the generations?

Which genes have the greatest impact on obesity, and how do they interact in a given individual?

How can weight best be restricted in obese mothers during pregnancy to avoid adverse consequences for the mother and fetus?

To what extent is the association between eating traits and excessive weight gain in children mediated by genetic factors, environmental factors, or both?

Animal Models of Diabesity

What are the mechanisms by which lack of abundant food early in life influences the developing brain?

How can the obesity phenotype be abrogated to reduce a genetically susceptible individual's risk of becoming obese for life?

What societal interventions might quell the prevalence of childhood and adult obesity and diabetes?

"Diabesity"—a convergence of type 2 diabetes and obesity resulting from chronic overnutrition—has reached catastrophic proportions worldwide. A new report by the Centers for Disease Control and Prevention shows that 24 million people in the United States now have diabetes—approximately 8% of the U.S. population. This is an increase of more than 3 million in two years, with most cases being linked with obesity, poor diet, and lack of exercise. According to the U.S. Surgeon General and the World Health Organization, "urgent action" is required to contain this growing public health crisis. Finding the best opportunity to intervene, however, is an ongoing challenge.

As Barry Levin of the University of Medicine and Dentistry of New Jersey explained at an April 15, 2008, meeting of the Academy's Diabetes & Obesity Discussion Group, "Many of us feel we're not going to cure adult obesity, but if we can learn the antecedents we may be able to prevent it," sparing future generations. Researchers believe that the period from gestation to puberty is most likely the optimal time to intervene. But doing so will hardly be straightforward. More than 50 genes are involved in obesity, and genes account for perhaps 60% to 70% of the problem.

Additional organizers of the meeting were John Kral of SUNY Downstate Medical Center and Gerard P. Smith of Weill Cornell Medical College.

"We used to think genes were everything—that in any given environment, genes alone would determine who remains healthy or succumbs to disease, and who becomes obese and who stays lean," said Robert Waterland of Baylor College of Medicine in his keynote address. But that no longer seems to be the case. Waterland is among the pioneers of an emerging area of research, epigenetics, which postulates that various influences during prenatal and early childhood development also affect individual susceptibility to disease and disorders such as obesity, as well as aspects of character and appearance.

Epigenetics—literally, "above" genetics—is the study of mitotically heritable alterations in gene expression potential that are not caused by changes in DNA sequence. Simply put, this means that epigenetic processes can alter gene activity in DNA without changing the genes themselves. Like a software program that tells a computer what to do, epigenetic processes tell DNA when, where, and how to express our genes. Just as genetic mutations can adversely affect health, so, too, can epigenetic dysregulation. And so interventions to thwart obesity and diabetes will have to target not only specific genes, but also specific epigenetic processes that influence them.

Figuring out which influences exert the most powerful effects, and how they interact within an individual, is extremely complicated. But scientists are making some inroads in their studies of obesity in humans, and utilizing animal models to gain a better understanding of the biological bases of diabesity.

Genetic and perinatal origins of diabesity

Rudolph Leibel of Columbia University observed that an individual's weight or body fat composition really represent "just the tip of a physiological iceberg below which are a whole series of cellular and molecular and developmental physiologies." Known for his work in cloning the gene for leptin, a hormone important for regulating body weight, he and his colleagues are working to identify other candidate genes that play pivotal roles in obesity. A better understanding of how those genes interact with epigenetic factors—particularly maternal nutrition—could lead to novel interventions in the future.

The average weight of women at the time of delivery is significantly higher in many parts of the world than it was 15 or 20 years ago, said Patrick Catalano of Case Western Reserve University. This corresponds to concomitant increases in weight among children and adolescents. His group found that the strongest correlate for fatness in a newborn at birth is the mother's pregravid insulin status; the more insulin resistant the mother is, the greater the adiposity in the offspring—even if the offspring is normal weight. This finding, along with others, underscores both short- and long-term effects of maternal obesity on the intrauterine environment as it relates to fetal growth, neonatal body composition, and metabolic dysfunction in late childhood.

"Many of us feel we're not going to cure adult obesity, but if we can learn the antecedents we may be able to prevent it."

Robert Whitaker of Temple University proposed clinical strategies to help break the cycle of obesity "transmission" from one generation to the next. Medical doctors can play an important role in helping parents to recognize the dangers of overweight in children, although he has also found that discussing these concerns with mothers can be difficult Incorporating strategies to address behavioral factors in obesity in the context of global goals for their children—e.g., having a happy child who gets along with others—is probably the best way convince mothers to help their children eat more healthfully and exercise regularly, he suggested.

According to John Kral, bariatric surgery in the mother before the child's birth could be another solution. His work has shown that offspring of obese mothers who underwent surgery were more likely to be normal weight and less likely to become obese. Andreas Plagemann of the University of Medicine Berlin emphasized the value of breastfeeding in reducing child health risks, but showed that the benefits are reduced somewhat if mothers are diabetic. Tanja Kral of the University of Pennsylvania emphasized the importance of creating a health-promoting family environment to mitigate the genetic susceptibility of some children to eat even when they're not hungry.

Body-brain connections

Animal studies are shedding light on the biological processes that underlie the gene/environment interactions leading to diabesity. Using a rat model of diet-induced obesity comprised of genetic lines of obesity-prone and obesity-resistant animals, Barry Levin is exploring the processes involved in weight gain and its consequences, particularly in the context of a genetic propensity to obesity. Since he doubts that adult obesity can be cured, his current efforts focus on prevention. His team found that when obesity-prone animals exercised for at least three weeks postnatally (roughly up to age 8 in humans), they were protected from becoming obese for life. Identifying exactly when the animals are vulnerable to an exercise intervention is critical to "making the translational leap" to humans, he said.

Working in animal models similar to Levin's, Sebastien Bouret of the University of Southern California is investigating the biological underpinnings of metabolic imprinting; that is, how the environment affects the development of neural circuits related to energy homeostasis. His research has revealed that under- or overnutrition in the early postnatal period can lead to permanent changes in phenotype, mainly through the action of leptin. However, both underfed and overfed animals showed neural abnormalities, suggesting that both low and high leptin levels can have a negative effect.

Peter Nathanielsz of the University of Texas Health Science Center is focusing on the effects of maternal nutrient restriction on fetal development. He noted that while it's true that many more pregnant women are overweight now than ever before, nutrient restriction is a problem in a number of population groups, including young girls who restrict their weight for cosmetic purposes, and households that struggle to have enough food. Working in a baboon model, he determined that maternal nutrient restriction results in fetal abnormalities of the pancreas, liver, and other organs.

Stress in early life can lead to early onset of metabolic syndrome and changes in the brain that are associated with elevations in body mass index and plasma cholesterol, according to Jeremy Coplan of SUNY Downstate Medical Center. In work in Bonnet macaques he has shown that adversity in early life can have lifelong consequences.

Call to action

Following a discussion period in which participants expressed concerns about the challenges of childhood and subsequent adult obesity, Robert Whitaker called for better communication among professionals in diverse disciplines. "We have unusual partners in obesity prevention, and we have to stop operating in a silo way," he said. People who work in family assistance programs need to communicate with pediatricians, as well as with professionals working to prevent sexually transmitted diseases and unwanted pregnancies. "They're all important partners for obesity prevention, and we need to bring them together over issues of mutual concern, even if, on the surface, the concerns don't seem to be similar," he concluded.