Metabolism disrupting chemicals and metabolic disorders

Metabolic syndrome (MetS) is a complex condition characterized by insulin resistance, abdominal obesity, dyslipidemia, hypertension, and hyperglycemia; it is a risk factor for cardiovascular disease, T2D, stroke, chronic kidney disease and cancers [ 1 , 2 ]. Its prevalence is increasing along with the increase in obesity, and it is reaching epidemic proportions affecting between 24% and 34% of the adult US population [ 3 ]. In the medical community, epidemics of metabolic diseases have largely been attributed to genetic background and changes in diet, exercise and aging. However, there is now considerable evidence that other environmental factors may contribute to the rapid increase in the incidence of obesity, T2D and other aspects of MetS observed over the past three decades [ 4 ]. One environmental factor that has begun to receive attention is a class of chemicals that can interfere with the action of hormones including metabolic hormones. These compounds, termed EDCs, are found in a wide variety of consumer products, and exposures are often widespread [ 5 ] Of particular concern is evidence that exposure to EDCs during critical periods when adipocytes are differentiating and the pancreas, liver, brain, etc. are developing can induce effects that manifest later in life, often as overt disease. This review will examine changes to the incidence of obesity, T2D and NAFLD and its associated hyperlipidemia, the contribution of genetics and describe the role of the endocrine system in these metabolic disorders. It will then specifically focus on the role of EDCs in metabolic diseases, focusing on their role in the etiology of obesity, T2D and NAFLD while finally integrating the information on EDCs on multiple metabolic disorders that could lead to MetS. We will specifically examine evidence linking EDC exposures during critical periods of development with metabolic diseases that manifest later in life and across generations.

The American Diabetes Association (ADA) defines Diabetes Mellitus (DM) as: “a group of metabolic diseases characterized by hyperglycemia resulting from defects in insulin secretion, insulin action, or both” [ 15 ]. DM can result from a deterioration in function and/or a loss of mass of pancreatic tissue [ 16 ] . T2D (formerly known as adult-onset or non-insulin-dependent diabetes or DM) accounts for 90-95% of diabetes cases and is characterized by increased insulin resistance and pancreatic beta cell dysfunction. More than 11% of individuals in the US older than 20 have diagnosed or undiagnosed T2D [ 7 ] and another 35% are estimated to be pre-diabetic. The World Health Organization (WHO) estimates that 347 million people globally suffer from diabetes (90% of which is T2D) [ 17 ]. Adolescents and even children have experienced significant increases in the prevalence of this disease over short periods [ 14 , 18 ]. Obesity is the main environmental factor driving the increased incidence of T2D; 70% of the risk associated with T2D is related to weight gain. Obesity is associated with insulin resistance that promotes beta cell proliferation, leading to hyperinsulinemia typical of early stages of T2D and MetS. However, obesity is neither necessary nor sufficient to cause T2D; these conditions can occur independently. Indeed, 20% of adults with T2D were not overweight and 57% of obese individuals do not have T2D [ 19 ].

The International Diabetes Federation estimates that 20-25% of the world’s adult population have MetS, which it defines as: “a cluster of the most dangerous heart attack risk factors: diabetes and prediabetes, abdominal obesity, high cholesterol and high blood pressure” ( https://www.idf.org/metabolic-syndrome ). The etiology of MetS is still a matter of research but insulin resistance and central obesity are significant contributors. Although there is still substantial debate, it is likely that components of MetS arise from insulin resistance. When insulin resistance occurs, there is an increase of fasting glucose and impaired glucose tolerance, often due to the abnormal expression of gluconeogenic enzymes. This metabolic state induces further insulin release, ultimately resulting in hyperinsulinemia. Hyperinsulinemia then simulates transcription factors such as Srebp-1c in the liver, which drive hypertriglyceridemia and hepatic steatosis [ 28 ]. In addition, the overproduction and secretion of insulin by pancreatic β-cells can result in their exhaustion and death, initiating the onset of T2D. The most prevalent form of insulin resistance is associated with abdominal obesity and dysfunction of adipose tissue, indicating an important central role for obesity in MetS.

Genetic factors are involved in the development of both type 1 and T2D. The most prominent genetic factor known to be associated with type 1 diabetes is located in the region of chromosome 6 that contains the highly polymorphic HLA class II genes and controls immune responsiveness [ 38 ]. Recently, whole-genome investigations have detected more than 20 other genetic variants associated with type 1 diabetes [ 39 ]. Twin and family studies have provided strong evidence that T2D also has a solid genetic predisposition [ 40 - 42 ]. Genome wide association studies (GWAS) identified genetic variants associated with T2D; most of the loci identified are related to lipid metabolism, obesity and β-cell pathways [ 43 , 44 ]. TCF7L2 demonstrated the strongest effect of >70 loci associated with the disease [ 43 ]. Similar to the genetic basis for obesity, it is assumed that predisposition to T2D involves multiple genes and SNPs.

Genetics have also been shown to contribute to variability in the occurrence of NAFLD [ 51 ]. In recent years, GWAS have identified two major genetic determinants associated with NAFLD. The most significant genetic linkage identified to date is a SNP in the gene patatin-like phospholipase domain-containing 3 (PNPLA3), a gene involved in the remodeling of hepatocellular lipid droplets [ 52 ]. This variant alters the mobilization of fatty acid, inhibits the activity of lipases and can cause hepatocellular accumulation of triglycerides [ 53 ]. Carriers of this genetic variant have increased susceptibility to liver damage when exposed to environmental stressors [ 53 ]. Another SNP in transmembrane six superfamily member 2 (TM6SF2) has been linked with NAFLD. This gene variant is located on chromosome 19 and associated with increased hepatic triglyceride content and lower serum lipoproteins [ 54 , 55 ].

Adipose tissue is the key regulator of energy balance and nutritional homeostasis and consists of white, brown and beige fat. It is an endocrine organ with more than 20 endocrine, paracrine and autocrine secretions. The adipose tissue consists of several depots located in subcutaneous, intra-abdominal (visceral) and intra-thoracic areas. Visceral adipose tissue depots are metabolically different from subcutaneous compartments [ 98 ]. Intra-abdominal (visceral) depots are associated with T2D and cardiovascular disease while subcutaneous depots seem protective against these diseases [ 99 ]. White adipose tissue stores energy as triglyceride and also signals to other organs on the status of energy reserves via hormones, growth factors and cytokines [ 100 ]. Two major secretions of white adipose tissue are leptin and adiponectin. Leptin is secreted in proportion to fat mass; it acts on the brain to reduce food intake and increase energy expenditure. However, obese individuals typically have increased serum leptin levels due to leptin resistance [ 101 ]. Adiponectin is also secreted from adipose tissue and induces fatty acid oxidation in liver, improves pancreatic beta cell function, enhances peripheral insulin sensitivity, suppresses hepatic glucose production and reduces inflammation [ 102 ]. Adipose tissue also contains immune cells, from both the adaptive (B and T lymphocytes ) or innate ( macrophages) immune system thus it is an immune orgain. Obesity is considered a proinflammatory condition in which both hypertrophied adipoctyes and resident immune cells produce and release proinflammatory cytokines, including IL-6 and TNFα, which are associated with chronic low-level systemic inflammation, insulin resistance and T2D. In contrast, in non-obese individuals anti-inflammatory cytokines including adiponectin, interleukin −10, and transforming growth factor beta (TGFb) are preferentially secreted which, among other functions, improve insulin sensitivity [ 103 ]. Brown adipose tissue is present throughout life, but with highest volume in newborns. It generates body heat via non shivering thermogenesis while beige fat appears to be bifunctional, changing to either white or brown fat depending on the stimuli [ 104 ]. The sympathetic nervous system controls lipolysis in white adipose cells and stimulates the cold response in brown adipocytes. In response to excess energy, adipocytes enlarge and/or increase in number [ 100 ]. Excess fat in these cells results in tissue dysfunction which leads to the development of other diseases and conditions including inflammation, T2D, heart disease, fatty liver, reproductive problems and some forms of cancer depending somewhat on the site of the added adipose tissue (e.g. visceral or subcutaneous) [ 100 ]. In humans, adipose tissue develops by the 14 th week of gestation [ 105 ] followed by a second period of increased cellularity that continues after birth and lasts through adolescence [ 106 ]. The number of white adipocytes is usually fixed after that time [ 100 ] however adipocytes are replaced at a rate of about 10% per year in adulthood [ 106 ], thus the tissue is not static. In mice, most subcutaneous adipogenesis occurs late in gestation and after birth; differentiation of gonadal fat only appears postnatally between birth and puberty [ 107 ]. Overall, fat mass can continue to grow due to high fat feeding which induces both hyperplasia and hypertrophy in rodents [ 107 ] with the hyperplasia occurring in the visceral tissue. Adult mice that are challenged with a high fat diet accumulate fat by hypertrophy in most adipose depots, with the exception of gonadal (visceral) fat which possesses higher capacity to expand by hyperplasia [ 107 ]. Specific genes play critical roles in fat cell development and control, including PPARγ and Runx2, often called the master regulators of fat cell differentiation [ 108 ] and sirtuins which play important roles in secretion of adipokines including leptin and adiponectin, hepatic glucose metabolism, insulin sensitivity and inflammation [ 109 ]. Adipocytes are derived from mesenchymal stem cells (MSCs), which can be neuroectodermal or mesodermal depending on where the fat body originates; differentiation of adipocytes requires a committed pre-adipocyte progenitor [ 110 , 111 ]. Visceral white adipose tissue (WAT) is primarily derived from the lateral plate mesoderm [ 112 ], brown fat is largely produced by the paraxial mesoderm [ 113 ] , and cranial WAT from the neural crest [ 114 ] . Beige (a.k.a brite) fat arises from WAT (precursors or mature cells). Despite this common origin, beige fat is thermogenic, like brown fat, so it plays a different metabolic role than WAT and has a correspondingly different transcriptional program than WAT [ 115 ] [ 116 ]. Mesenchymal stem cells harvested from adipose tissue or bone marrow can be made to differentiate into fat, bone, cartilage, and other lineages in culture [ 117 ]; commitment to each of these

The liver is the principal location of glucose storage as glycogen, and the main source of glucose for all tissues. Because the pancreatic veins drain into the portal venous system, every hormone secreted by the pancreas must traverse the liver before entering systemic circulation. The liver is a major target for pancreatic insulin and glucagon action as well as their site of degradation. In fact, 70% of hepatic glucose output occurs via liver glycogenolysis and 30% via gluconeogenesis. Insulin promotes glycogen synthesis and decreases its breakdown after enhancing the transcription of glucokinase and the activation of glycogen synthase through changes in its phosphorylation state. Insulin increases transcription of the glucokinase gene and other enzymes involved in glycolysis such as phosphofructokinase and pyruvate kinase, promoting glycolysis [ 140 ]. Insulin also inhibits gluconeogenesis by decreasing phosphoenolpyruvate carboxykinase (PEPCK) and fructose-1, 6-biphosphatase (FBPase) gene expression. As a result, insulin inhibits glucose production during the fed state, keeping glucose levels within the normal range. At the same time, high glucose levels inhibit glucose-6-phosphatase and decrease the activity of glycogen phosphorylase; all together, these processes considerably reduce the conversion of glycogen to glucose. Insulin also promotes the storage of fat by stimulating lipogenesis. It inhibits the oxidation of fatty acids by decreasing fatty acid transport into the mitochondria. Additionally, insulin stimulates fatty acid synthase (FAS). All together, these pathways promote the formation of triglycerides that can either be stored in the liver or exported as very low density-lipoproteins (VLDL). On the other hand, glucagon signaling in the liver plays a key role during fasting, as well as in the adaptive response to hypoglycemia. After binding at its receptors, glucagon activates the cAMP/PKA pathway, which decreases glycolysis via a modulatory action on pyruvate kinase [ 127 ]. Glucagon increases gluconeogenesis after up-regulation of glucose-6-phosphatase and PEPCK through the activation of coactivators such as CREB-binding protein (CBP), P300, PGC-1 and TORC2 [ 134 , 141 - 144 ]. In addition, glucagon also activates ketogenesis. All these effects favor hepatic release of glucose to maintain normal blood glucose levels during fasting. Glucagon also promotes the oxidation of fat in the liver, increasing the activity of the citric acid cycle and the generation of ketone bodies. Moreover, there is a glucagon-induced decrease of triglyceride, VLDL, cholesterol and fatty acid synthesis mediated by PPARα [ 145 ].

Similar to skeletal muscle, insulin promotes recruitment of GLUT4 to the membrane and accelerates glucose transport into adipocytes. It then induces the breakdown of glucose to generate triglycerides. These triglycerides are stored in fat together with those delivered via the circulation as chylomicrons and VLDL. In addition, insulin inhibition of triglyceride lipase decreases triglyceride breakdown. Insulin decreases lipolysis through inhibition of hormone sensitive lipase in a cAMP-dependent manner [ 146 , 147 ]. Insulin promotes the synthesis of lipoprotein lipase (LPL), which is exported to the endothelial cell plasma membrane. Once anchored there, LPL cleaves triglycerides from VLDL and chylomicrons into glycerol and fatty acids that are taken up by nearby adipocytes to form triglycerides. Although the role of glucagon in WAT is controversial, recent results point to a role in lipolysis [ 127 ]. This lipolytic action has been attributed to glucagon-induced release of fibroblast growth factor 21 (FGF21) [ 148 ] as well as to signals from the sympathetic nervous system [ 149 ].

The liver is the largest and most metabolically complex organ in the human body. Hepatocytes make up over 80% of total liver mass and play a critical role in intermediary energy (lipid, carbohydrate, amino acid) and xenobiotic metabolism (Phase I-III metabolizing enzymes). Other liver-specific cell types include Kupffer cells, biliary epithelial cells, sinusoidal endothelial cells, and stellate cells. These cells have specialized functions ranging from protection against infection, bile duct flow, endocytosis and fibrosis. The liver arises from the hepatic diverticulum of the foregut during the fourth week of gestation. Hepatoblasts are bipotential progenitor cells arising from foregut endodermal cells that differentiate into hepatocytes and cholangiocytes. The liver is the principal organ for xenobiotic detoxification. Ligand-activated xenobiotic receptors induce foreign compound metabolism by cytochrome P450s. For example, the aryl hydrocarbon receptor (AhR) induces expression of CYP1A1, the constitutive androstane receptor (CAR) induces CYP2B10, and the pregnane X receptor (PXR) induces CYP3A4. In general, chemical ligands are metabolized by the P450s that they induce. In addition to foreign compound metabolism, xenobiotic receptors play an important role in the control of hepatic lipid and carbohydrate metabolism. It was recently proposed that activation and cross-talk of xenobiotic receptors by foreign compounds is a molecular initiating event in hepatic steatosis [ 159 ]. Likewise, interactions between environmental compounds and xenobiotic receptors regulate, in part, hepatic carbohydrate metabolism including gluconeogenesis and insulin resistance [ 160 , 161 ]. These mechanisms appear to account for the wasting syndrome associated with some dioxin-like chemicals that activate the AhR [ 161 ]. Owing to its critical role in xenobiotic and intermediary metabolism, the liver is a principle target organ for chemicals resulting in the development of steatosis. Steatosis may progress to steatohepatitis (steatosis with superimposed hepatic inflammation), cirrhosis and hepatocellular carcinoma, and ultimately liver-related death if liver transplantation does not occur. In the clinic, steatohepatitis is named according to its etiology: alcohol (alcoholic steatohepatitis, ASH), cancer medications (chemotherapy associated steatohepatitis, CASH), excess dietary lipids or carbohydrates (NASH), and industrial chemicals (toxicant associated steatohepatitis, TASH) [ 162 , 163 ]. While disease mechanisms vary by etiology [ 164 ], steatosis is invariable associated with an imbalance of hepatocyte lipid synthesis, oxidation, uptake, and efflux via VLDL [ 159 ].

In humans, there are important sex differences in the incidence and health consequences of obesity; men and women differ in the patterns of fat deposition, fat mobilization, utilization of fat, and the consequences of both excess and insufficient fat stores. Gonadal hormones appear to play a crucial role in shaping such differences. Women suffer fewer obesity-related disorders than men do. In fact women are resistant to free fatty acid-induced insulin release and are therefore less prone to T2D before manopause but the prevalence of these disorders increases dramatically after menopause[ 175 ]. The prevalence of T2D is higher in men before puberty compared to reproductive age females. It is noteworthy that T1D has a male predominance as well[ 176 ]. Androgens, adiposity and disease are clearly interrelated in humans. These asymmetries in energy balance traits probably reflect evolved adaptive differences due to differential investment and costs of reproduction in male and female mammals and are mainly shaped by gonadal hormones either during development (organizational effects) or at adulthood (activational effects) (for reviews see [ 177 - 179 ]). Development and maturation of brain circuits involved in the regulation of food intake and metabolism occur during the perinatal period. The current literature argues that there are multiple critical periods in which hormones organize energy balancing traits; besides the fetal and neonatal stage, the peripubertal period is also a time window when sexually dimorphic eating behaviors are established [ 180 ]. Sex differences in body fat composition and distribution, energy expenditure, orosensory physiology, taste and smell preference, food intake, binge eating, susceptibility to diet induced obesity, responses to leptin-, ghrelin,- or insulin-induced hyperphagia, POMC gene expression in the ARC nucleus, and many other traits are well documented (reviewed in: [ 181 - 183 ]). The POMC, melanocortin system, is sexually dimorphic [ 184 ]. In adults, females have increased responsiveness to leptin and decreased responsiveness to insulin in comparison to males. These differences are estrogen dependent [ 185 ], and they are perinatally organized by testosterone [ 186 ]. The NPY/AgRP circuit is also sexually dimorphic. In particular, in situ hybridization studies demonstrated sex differences in the distribution of NPY mRNA-containing cells in the rat ARC, and its modulation by testosterone in males [ 187 ]. Also, NPY immunoreactivity is sexually dimorphic in the ARC, the dorsomedial hypothalamus, and the PVN [ 188 ]; NPY-Y1 receptor expression is higher in females compared to males [ 189 ]. Male mice have higher levels of daily food intake, post-fast hyperphagia and leptin-induced hypophagia compared to female mice, and these behavioral differences are related to sexual dimorphisms in the ARC as far as the number of ARC cells containing NPY, AgRP, and POMC. Females perinatally treated with testosterone or DHT show male-like levels of food intake, post-fast hyperphagia and POMC gene expression and projection [ 186 ]. Estrogens play a pivotal role in regulating energy homeostasis, especially in female mammals, either by acting directly on the brain or through activation of estrogen receptors (ER) on adipocytes. Estrogens protect against increased adiposity/obesity through their effects of suppressing appetite and increasing energy expenditure; estradiol suppresses feeding by enhancing the potency of other anorectic signals (leptin, apolipoprotein, BDNF, cholecystokinin) and by decreasing the potency of orexigenic signals such as ghrelin and melanin concentrating hormone [ 87 , 185 , 190 , 191 ]. The liver is a major target for estrogen action in female mammals and the activity of the liver ERα is strictly associated with ovarian activity [ 192 ]. In the liver, ERα regulates fertility in response to protein consumption and controls lipid and cholesterol synthesis in relation to the reproductive cycle [ 193 ] . Since the liver is the major organ for the control of energy homeostasis, the activity of hepatic ERα also influences the synthesis and secretion of the signaling molecules necessary for coordinated responses among liver, fat, muscles and brain [ 192 ]. In mammals, including humans, the liver is a sexually dimorphic organ and exhibits major differences in the profile of steroid, lipid, foreign compound metabolism [ 194 ], and gene expression. These differentially expressed genes regulate a wide range of biological processes; accordingly, many enzymes, such as steroid hydroxylases belonging to the cytochrome P450 (CYP) superfamily, are expressed in the liver in unique, sexually biased patterns [ 195 ]. Such differences have implications for sex-related steroid metabolism, xenobiotic metabolism and pharmacokinetics, and differential susceptibility to some liver diseases [ 23 , 196 ]. The sexual dimorphism of liver gene expression is established and maintained, in part, by

As noted above, approximately 1000 chemicals have been identified that meet the criteria of an EDC [ 213 , 222 ]. These compounds are used in a wide range of consumer products including food packaging, building materials, pesticides, clothing and upholstery, personal care products, detergents and other cleaning agents, thermal paper, plastics and medical equipment [ 210 , 215 , 227 , 228 , 256 , 257 ]. Some chemicals used in industrial processes lead to unintended contamination of food, water and air. Thus, routes of exposure can include oral, dermal, and inhalation, as well as subcutaneous and intravenous (via medical equipment). The US CDC’s National Health and Nutrition Examination Survey (NHANES) is a nationally representative biomonitoring program which assesses, among other things, exposures to environmental chemicals in the general population [ 258 ].The CDC has documented widespread exposures to a number of EDCs (e.g. [ 259 - 264 ]). Importantly, a large number of chemicals are not examined, and thus the number of exposed individuals, as well as the typical levels of exposure, remains unknown. Although the sampling of infants and young children is limited in the context of NHANES [ 265 ], other studies have revealed the presence of environmental chemicals in placenta, amniotic fluid and umbilical cord blood, documenting exposure throughout the most critical stages of development as well as across the lifespan (e.g. [ 266 - 272 ]).

In 2002, Baille-Hamilton wrote the first article relating environmental chemicals to obesity. Her article, “Chemical toxins: a hypothesis to explain the global obesity epidemic”, suggested that the current obesity epidemic was associated with the increase in production of environmental chemicals [ 300 ]. She reviewed published studies showing associations between exposure to a variety of environmental chemicals, including some pesticides, solvents, plastics, flame retardants and heavy metals, and increased weight gain; because these studies originally focused on weight loss and toxicity, their effects on weight gain had gone unnoticed. In 2006, Grun and Blumberg, published their now classic paper, in which they coined the term “obesogen” followed by a 2009 review “Endocrine disruptors as obesogens” [ 301 ]. They noted the existence of chemicals that alter regulation of energy balance to favor weight gain and obesity and proposed that obesogens derail the homeostatic and reward mechanisms important for weight control, such that exposed individuals have increased susceptibility to weight gain despite normal diet and exercise. The obesogen hypothesis makes two important points. First, susceptibility to obesity starts during development ( in utero and the first few years of life). Second, susceptibility to obesity is due in part to the influence of a specific subclass of EDCs that alter developmental programming, and thus disrupt the set point for weight gain later in life. "Obesogens" are defined functionally as chemicals that promote obesity by increasing the number of fat cells and/or the storage of fat in existing adipocytes. Obesogens can also act indirectly to promote obesity by shifting energy balance to favor calorie storage, by altering basal metabolic rate, by altering gut microbiota to promote food storage [ 302 ], and by altering hormonal control of appetite and satiety [ 303 - 307 ]. New obesogenic chemicals are being identified at an increasing rate. The obesogen field has recently expanded to include chemicals that cause or lead to diabetes [ 308 ] as well as altered lipid metabolism and fatty liver [ 309 ].

5.1 Adipogenesis, Subsequent Weight Gain and Obesity A number of MDCs have been shown to significantly alter the function (gene expression, hormone secretion) of white adipose tissue, adipose tissue mass (adipocyte number and/or volume), or body weight in animal models after developmental exposures ( Figure 2 ). Epidemiological studies also support the identification of obesogenic MDCs [ 311 , 312 ] and these studies focus mainly on weight gain and (body mass index) BMI as endpoints. This is typical for a new field, where the focus is on descriptive studies that show that a chemical can have an effect on an endpoint or disease of interest (e.g. weight gain). In many cases, effects of MDCs on adult adiposity and/or body weight are reported to be significant for only one sex, consistent with the sexually dimorphic responses that are a common feature of EDCs and thus MDCs [ 227 ]. Nicotine is an MDC where there are compelling data for its obesogenic properties from both animals and human studies [ 7 , 313 , 314 ]. Maternal smoking in pregnancy is a risk factor for subsequent obesity in offspring [ 315 ] even when exposure is limited only to early pregnancy [ 316 , 317 ]. Although multiple mechanisms have been proposed, associations may be partly attributable to impaired fetal growth, which as Barker and colleagues noted is a risk factor for subsequent rapid growth and long-term obesity [ 318 ]. Developmental exposure in mice to DES [ 319 ] has also been shown to increase weight gain which is specific to females and does not appear until puberty [ 320 ]. In another study, prenatal exposure to DES resulted in an increase in number of adipocytes in the gonadal fat pad of male mice[ 321 ]. Two epidemiological studies also link DES exposure to obesity; prenatal exposure to DES is associated with an increased likelihood of childhood obesity at age 7y [ 322 ] and increased risk of adult obesity in women that was most evident at lower doses [ 323 ]. Bisphenol A (BPA) [ 324 - 326 ], a chemical used to make polycarbonate plastic, epoxy resins that line food and beverage cans, and as a developer in cash register receipts has been shown to increase weight gain and body fat after developmental exposure in rats [ 327 - 329 ] and mice [ 321 , 330 - 332 ]. Some studies have not shown effects of BPA on weight gain [ 333 , 334 ] including government standardized study designs e.g. guideline studies, [ 335 , 336 ]. The two guideline studies with CD-SD rats,which appear to be relatively insensitive to xenoestrogens [ 337 ], showed no significant effects for any outcome measure. Also in one mouse study reporting no significant effect of BPA on body fat[ 333 ], the control animals were obese due to the use of casein-based feed, which increases body fat in CD-1 mice [ 338 , 339 ] . These studies differ with regard to several aspects including animal strain, doses, developmental windows, and diet which are likely responsible for the discordant results (reviewed in [ 340 ]). Also a distinction needs to be made between studies that only measured body weight (for example [ 331 ] ) which is recognized to be a poor indicator of significant changes in body fat in rodents [ 7 ] and studies that actually measured body fat. Human studies have shown that prenatal exposure to BPA was associated with increased body fat at age 7 [ 341 ] or BMI by age 9 [ 342 ] or accelerated postnatal growth without a change in BMI between age 2-5 [ 343 ] consistent with the DOHaD prediction that light at birth babies would experience increased rate of growth in childhood[ 14 , 344 ]. Not all epidemiology studies report a positive relationship between an exposure and a health outcome [ 342 , 343 , 345 - 347 ] which is not uncommon for studies linking environmental chemicals with adverse outcomes in humans, possibly due to potential for multiple environmental “stressors” to interact with chemical exposures [ 345 , 348 , 349 ]. Because of considerable within-person variability [ 350 - 353 ] and because most studies typically have only one sample to characterize exposure, BPA exposures may have been substantially misclassified and peak exposures, which generally occur in the evening, may have been underestimated [ 351 ]; these misclassifications may have increased the likelihood of a false negative outcome. Stronger attention is also needed for potential confounding by diet, which is only one source of BPA exposure [ 257 , 354 ]. Moreover, since the developing organism is more susceptible to BPA effects, epidemiology studies must consider the lifestage when exposure is measured. The inconsistencies in the data notwithstanding, there are data from both animal and human studies that support the hypothesis that developmental exposure to BPA can lead to an increase in weight gain later in life in exposed offspring. Phthalates are a class of chemicals that promote flexibility in plastic products such as tubing and vinyl flooring. Fragrances and a variety of household and

Only a few studies have investigated the action of potential MDCs on neural circuits/cells and on the resulting feeding behavior and energy balance output. However, the neuroendocrine control of these features could be an important target of environmental chemicals. Prenatal exposure to low doses of BPA alters the development of the POMC system in a sexually differentiated way [ 440 ]. The differences are evident when adult are exposed to a high-fat diet; in particular, males show reduced POMC fiber innervation of the PVN and increased NPY and AGRP expression in the ARC. Females exposed to BPA show reduced POMC expression and ERα expression patterns in the ARC similar to those seen in males, suggesting a masculinizing effect of BPA. Also, fetal exposure to BPA in mice alters food intake during puberty and in adulthood as well as leptin and insulin levels, which in turn regulate the NPY system[ 321 ]. Organotin compounds such as TBT have not only peripheral effects, but also may activate elements in the brain, in particular in a crucial region for the regulation of food intake, the ARC [ 441 ] . Adult mice exposed to TBT for 4 weeks show profound alterations of the leptin-NPY-NPY-Y1 receptor system [ 96 , 188 ]. Prenatal exposure to TBT also induces hypothyroidism in the progeny, while the acute treatment of pregnant females induces a dose-dependent increase of T3-independent TRH transcription levels [ 442 ] in the hypothalamus. In summary, there are a number of important endpoints to study the effects of MDCs on the regulation of food intake and metabolism in the central nervous system. These include expression of the leptin receptor, ERα, thyroid hormone receptor (associated also with PPARy), the POMC-CART system, the NPY-AGRP system and their receptor systems, and the dopaminergic system.

Evidence that chemicals can disrupt the function of the endocrine pancreas dates to the early 1940s when alloxan, a glucose analogue which selectively destroys insulin producing cells in the pancreas, was shown to promote type 1 diabetes in rabbits [ 452 ]. This was followed by the discovery that streptozotocin similarly induced a diabetic state through selective β-cell destruction. Although humans are not exposed to alloxan or sreptozotocin, they are used in animal research on diabetes. Initial human evidence that synthetic chemicals promote the development of diabetes came from patients accidentally or intentionally exposed to pyrinuron (Vacor) [ 453 ]. Exposure to this rodenticide resulted in β-cell destruction and the development of type 1 diabetes [ 454 ]. More recently, a patient exposed to high levels of the fungicide chlorothalonil was reported to develop diabetic ketoacidosis, a condition arising from insulin deficiency [ 455 ]. These and other studies of environmental contaminants provide mechanistic insights into the pathways by which MDCs may promote diabetes pathogenesis through defects in β-cell physiology. 5.5.1 MDCs and Beta Cell Survival and Function 5.5.1.1 Cellular Studies A structurally diverse array of synthetic and inorganic toxicants disrupts β-cell survival and function in cell lines and isolated rodent islets. 2,3,7,8-tetrachlorodibenzodioxin (TCDD) reduces glucose-stimulated insulin secretion (GSIS) in isolated islets [ 415 , 456 , 457 ]. Similarly, DDT impairs both GSIS as well as insulin secretion in response to tolbutamide (a pharmacological insulin secretagogue) [ 458 ]. Among organotin compounds, triphenyltin was shown to disrupt cellular signaling in β-cells and impair insulin secretion in response to a variety of stimuli [ 459 ]. Similarly, inorganic contaminants including both inorganic and methylated arsenic [ 460 - 462 ], cadmium [ 463 - 465 ], and mercury [ 466 ] disrupt β-cell function ( Figure 2 ). Interestingly, several compounds disrupt β-cell signaling and function in a manner that promotes insulin release. In the RINm5F cell line, a PCB mixture (Aroclor 1254) increased insulin secretion into the culture media, an effect recapitulated by coplanar PCB congeners [ 467 ]; TCDD also promotes continuous insulin release [ 468 ]. Interestingly, the insulin secretory effect resulted in a depletion of cellular insulin content by PCBs [ 467 ] and TCDD was proposed to promote β-cell “exhaustion” [ 468 ]. This suggests that prolonged exposure to these compounds could result in insulin deficiency. BPA has been shown to augment GSIS; unlike TCDD and PCBs, low doses of BPA augmented β-cell insulin content in an ERα dependent manner [ 469 ]. A rapid action on insulin release was shown to be dependent on ERβ expression, and was confirmed in human as well as mouse islets [ 470 , 471 ]. Importantly, these effects of BPA may be representative of effects of other phenolic compounds because nonylphenol and octylphenol were also shown to augment GSIS in isolated rat islets [ 472 ]. In contrast to the extensive work examining MDC effects on β-cell physiology, less is known about the effects of environmental pollutants on α-cell biology. In early studies, cobalt was shown to be toxic to α-cells [ 473 ]. More recently, BPA has been shown to alter calcium signaling in α-cells [ 474 ]. Collectively, these data support the theory that the endocrine pancreas is an important target for the deleterious effects of MDCs on energy homeostasis.

The effects of MDCs on β-cell physiology in adult human studies are limited. In a Northern Mexican cohort, inorganic arsenic exposure was associated with a reduction in β-cell function, with the effect amplified among those with T2D [ 479 ]. Along with larger epidemiologic literature linking arsenic exposure to diabetes [ 480 - 482 ], other studies have also found arsenic to be associated specifically with measures of β-cell dysfunction or reduced insulin secretion, more strongly than with measures of insulin resistance [ 483 , 484 ]. Epidemiological studies also suggest that BPA, phthalates, dioxins, and POPs including DDT metabolites and PCBs are associated with measures of altered glucose homeostasis including T2D [ 345 , 485 - 489 ]. In one recent study, urinary BPA concentration in US adults was associated with an increase in β-cell function, hyperinsulinemia and insulin resistance [ 490 ] preferentially in males. These results are similar to those obtained from studies performed in mice [ 469 ]. Consistent with animal studies [ 491 ], a number of human studies —including studies among children and numerous studies in adults— suggest that DEHP is more strongly associated than are other phthalates with diabetes and other markers of impaired glucose metabolism [ 365 , 492 - 495 ], perhaps because of greater activation of PPARs. However, several other studies [ 496 - 498 ] found stronger evidence of associations with butyl phthalates, which have a weaker PPARγ affinity than do DEHP metabolites [ 499 ]. Because data thus far are limited, it is uncertain to what extent the mixed results in humans may be due to factors such as differences in exposures [ 496 ], measurement errors in estimates of exposure [ 350 ], or sex-specific effects [ 500 ]. Moreover, to date very few epidemiological studies have examined the developmental or perinatal exposures thought to have the most potent diabetogenic effects [ 501 ], though recent animal studies support adverse effects of ongoing exposures including those in adulthood [ 502 , 503 ]. Nonetheless, overall, these studies support the idea that MDC-induced disruptions in β-cell function may mediate some of the observed associations between environmental chemicals and diabetes in human populations.

5.5.2.1 Cellular models In addition to data demonstrating that MDCs disrupt insulin production, an increasing body of evidence suggests that a variety of MDCs have the capacity to impair peripheral insulin action. In diverse cell line and organ culture models of insulin-responsive tissues, an array of compounds have been shown to impair insulin signal transduction or insulin-stimulated glucose disposal, including TCDD [ 508 , 509 ], tolylfluanid [ 510 ], inorganic and methylated arsenic species [ 511 , 512 ], DEHP [ 513 , 514 ], and POPs [ 416 ]. In one model, BPA was also shown to inhibit insulin-stimulated glucose utilization in 3T3-L1 adipocytes [ 417 ] and another study showed that BPA can increase basal and insulin-stimulated glucose uptake in 3T3-F442A cells [ 412 ]. BPA also stimulated secretion of pro-inflammatory cytokines IL-6 and TNF while inhibiting the anti-inflammatory cytokine adiponectin from human adipoctyes in culture {Ben-Jonathan, 2009 #2081}. Collectively, these data suggest impairments in insulin action may result from exposure to a variety of environmental MDCs; however, dose, duration, and model system may alter the phenotypic response of some tissues to these compounds.

Various models have suggested that imbalances in insulin action can arise from MDC exposures during development. BPA enhanced the insulin resistance of pregnancy, with female offspring demonstrating higher insulin levels and males exhibiting glucose intolerance with systemic insulin resistance [ 536 ]. Insulin resistance was also observed in BPA-exposed rats [ 537 ], while another mouse model similarly demonstrated glucose intolerance with insulin resistance; however, the effect was observed only at low doses[ 321 ]. Metabolic stressors like high fat diet may be additive to the BPA-induced insulin resistance [ 331 ]. Indeed, high-fat diet potentiated GSIS impairments elicited by low doses of BPA given subcutaneously [ 538 ]. In the CD-1 mouse, however, developmental exposure to BPA did not alter glucose homeostasis in adult mice fed a normal chow or high fat diet [ 333 ]. Overall, these findings suggest that developmental BPA exposure can alter metabolism, albeit the ultimate metabolic phenotype may be modulated by genotype, diet and exposure patterns. Additional studies have suggested that developmental exposure to other compounds can promote alterations in insulin action. For example, exposure to low doses of PFOA in midlife were shown to increase insulin levels [ 398 ], while PFOS exposure during gestation and early postnatal development resulted in glucose intolerance and insulin resistance [ 539 ]. Rats exposed to PFOS from gestation day 0 to postnatal day 21 also shown exhibited glucose intolerance with elevated insulin levels [ 540 ]. Developmental exposure to DEHP induces a complicated phenotype with the development of hyperglycemia with reduced insulin levels in female rats, while male offspring had elevated insulin levels but normal glucose tolerance [ 491 ]. In another model, DEHP exposure led to glucose intolerance with insulin resistance in the offspring, although this model also revealed central defects in β-cell function as well [ 505 ]. Sex-specific effects of DEHP on measures of insulin resistance have been reported in some epidemiological studies [ 495 ], but not others [ 541 ]. These data suggests that both adult and developmental exposures to various MDCs have the capacity to modulate insulin action globally as well as at the cellular level. This conclusion is further supported by a number of studies, not discussed, in which MDC exposure promoted glucose intolerance without examination of insulin levels or action. However, the precise mechanism(s) by which this common phenotype occurs remains somewhat obscure. Examining the totality of the data, several molecular pathways are implicated as potential mechanisms of altered insulin action. These include increased production of proinflammatory cytokines that induce insulin resistance such as TNFα and IL-6 [ 508 , 524 , 542 ], increased oxidative stress [ 466 , 477 , 515 , 525 , 543 ], and mitochondrial dysfunction [ 529 ], which may also increase oxidative damage. Further work is required to fully characterize the modes by which MDCs promote impaired insulin action to devise strategies to mitigate their adverse effects on global energy homeostasis.

Developmental studies examining MDCs and liver health endpoints have been conducted in laboratory animals, but epidemiology studies examining the developmental basis of these diseases are lacking. This is likely due to the relatively long time to develop clinically apparent human liver disease owing to the slowly progressive nature of hepatic fibrosis. Compounding the problem, routine clinical biomarkers (e.g. alanine aminotransferase) may be insensitive for the detection of environmental liver disease[ 549 ] Novel biomarkers for fatty liver disease and fibrosis [ 550 ] are being developed for clinical use, and perhaps these could be applied to future environmental epidemiology studies. Due to the relative lack of epidemiological data on developmental MDC exposures in steatosis and hyperlipidemia, post-developmental studies (adolescent and adult) provide ‘proof of concept’ and thus are reviewed below. 5.6.1 Adult MDC Exposures, Steatosis and Hyperlipidemia Hepatic steatosis is likely to be the most common pathologic liver responses to chemical exposures [ 163 ]. Indeed, hepatic steatosis was noted in approximately 8-10% of rodent studies warehoused in the Environmental Protection Agency’s (EPA) Integrated Risk Information System (IRIS) database and the Toxicological Reference Database (pesticide) [ 162 , 551 ]. Furthermore, in the Chemical Effects in Biological Systems (CEBS) data repository of the US National Toxicology Program (NTP), 329 rodent studies of 81 unique chemicals reported hepatic steatosis [ 551 ]. Whatever its etiology, hepatic steatosis is invariably associated with insulin resistance [ 163 ]. However, this interaction is complex as fatty liver disease is both a cause and effect of insulin resistance. While some chemicals (e.g. vinyl chloride) appear to directly cause steatosis and steatohepatitis [ 552 ], other chemicals such as non-dioxin like PCBs merely modify the hepatic response to diet-induced obesity [ 535 ]. These chemicals may be a ‘second hit’ in the progression of diet-induced steatosis to frank steatohepatitis, which may further progress to cirrhosis and hepatocellular carcinoma. Many non-dioxin like PCBs interact with NR1 class nuclear receptors such as PXR and CAR [ 553 ]. However, the role of MDC-nuclear receptor interactions in fatty liver disease is likely to be complex, as PPAR? agonists (e.g. obesogens) have been proposed as treatments for non-alcoholic fatty liver disease and associated insulin resistance [ 554 ]. Nuclear receptor crosstalk especially at the liver X receptor response element is also likely to be important, but is currently understudied. Other proposed mechanisms include oxidative/carbonyl stress, endoplasmic reticulum stress, mitochondrial dysfunction, increased cytokine production, increased hepatic lipid synthesis/uptake and impaired VLDL synthesis and secretion [ 162 , 163 ]. Many of the environmental chemicals associated with steatosis are organochlorines. Of the compounds in the US Environmental Protection Agency’s Integrated Risk Information System (IRIS) that induced steatosis in rodents following oral (dietary) treatment, the most potent (mirex, chlordecone, chlordane) were structurally similar, highly chlorinated molecules (>8 chlorines) [ 162 ]. While these data suggest that highly halogenated environmental chemicals could be of particular interest in steatosis, more data are required. Key chemical classes identified in adult steatosis studies include solvents and volatile organic chemicals; POPs and pesticides; and metals [ 163 , 551 ]. Solvent exposures have historically been associated with steatosis and liver injury in the occupational health literature [ 552 , 555 ]. These data were recently reviewed by the Institute of Medicine and the National Research Council which concluded that there was “…evidence of an association between chronic exposure to solvents in general and hepatic steatosis that could persist after cessation of exposure” [ 556 , 557 ]. Unfortunately, it is difficult to assess biomarkers of prior solvent exposures, and epidemiological data on the impact of solvent exposures in human cohorts are lacking. Exposure to BPA is associated with liver enzyme abnormalities reflective of liver injury in population-based studies [ 344 , 558 ], although pathologic data were not provided. Regarding dyslipidemia, non-significant trends were observed for BPA in pediatric NHANES populations [ 559 ]. However, prolonged (8-month) exposure of male mice to low BPA doses induced hypercholesterolemia with upregulation of key genes involved in cholesterol biosynthesis including sterol regulatory element-binding protein 2. Whole body de novo cholesterol synthesis was also increased as seen by the plasma lathosterol-to-cholesterol ratio [ 503 ]. Interestingly, the food contaminant 1,3-dichloro-2-propanol induced hyperlipidemia with increased LDL/HDL ratio in mice via the AMP-activated protein kinase (AMPK) signaling pathway [ 560 ]. Exposures to

As noted above there are MDCs that can result in obesity, T2D or lipid disorders. What is striking is that when multiple endpoints were examined within individual studies, in many cases an MDC caused disruptions in multiple disease pathways leading to what could be called MetS. Table 1 shows that indeed many EDCs should be characterized as MDCs as they can cause multiple disease/dysfunctions, even when not all of the metabolic endpoints were measured in the same experiment. We highlight examples here to show that indeed there are MDCs that can affect multiple tissues leading to multiple metabolic disorders and in some cases to MetS due to their ability to induce weight gain, glucose intolerance and lipid disorders. These data indicate it is important to examine more than one endpoint and tissue to define an action of a suspected MDC. The first example comes from developmental exposure to BPA, which can induce glucose intolerance and insulin resistance [ 331 ], and impairments β-cell function and mass [ 507 ]. Developmental exposure to BPA has also been shown to cause weight gain in offspring in some animal models and human studies (reviewed in [ 340 ]). BPA exposure throughout gestation and perinatal development exacerbates a nonalcoholic steatohepatitis-like phenotype in male rats that were fed a high-fat diet post-weaning; in particular, BPA worsened the accumulation of lipids in hepatocytes as well as liver inflammation and oxidative stress fibrosis [ 600 ]. Similarly, rats exposed to DEHP throughout gestation and perinatal development exhibited hyperglycemia in the presence of reduced insulin levels [ 504 ] along with reductions in β-cell mass, reduced islet insulin content, and disruptions in β-cell ultrastructure [ 491 ]. Sex-specific effects of DEHP on measures of insulin resistance have been reported in some epidemiological studies[ 495 ]. Gestational exposures to DEHP resulted in hepatic steatosis in offspring as well as decreased glycogen storage in males, but with a delayed shift to glycogen-dependent metabolism of the mature hepatocyte [ 602 ]. Prenatal exposure of mice to DEHP results in increased body weight as well as numbers and size of adipocytes in male offspring in several studies from different labs using different models [ 356 - 359 ]. DDT and its metabolites have been associated with increased risk of higher body weight, insulin resistance, T2D and dyslipidemia in human studies [ 573 , 607 ]. In rodent studies, female offspring exposed prenatally to DDT followed by a high fat diet for 12 weeks in adulthood developed glucose intolerance, hyperinsulinemia, dyslipidemia and altered bile acid metabolism as well as reduced energy expenditure and impaired thermogenesis [ 389 ]. DDT effects were also transmitted across generations resulting in obesity in the F3 generation [ 608 ]. PBDEs have also been associated with development of fatty liver disease and/or abnormal hepatic lipid metabolism in rodent studies [ 368 , 378 , 551 , 605 ]. Exposure to BDE-47 and DE-71 resulted in transcriptomic enrichment of genes of lipid metabolism in rat livers [ 605 , 606 ], including long-term systematic activation of pathways of α, ω, and β-oxidation of fatty acids. Developmental exposure in a rodent model to a specific PBDE mixture, Firemaster 550, resulted in weight gain in the offspring [ 410 ]. Prenatal exposure to TBT in mice promotes adipocyte differentiation that results in increased lipid accumulation and adipose tissue while reducing muscle mass that persists into adulthood and across generations [ 309 , 367 ] . It also increases adiposity in zebrafish [ 373 ] . One epidemiology study noted that prenatal TBT exposures were associated with a non-significant trend towards higher weight gain in the first three months of life [ 374 ]. In a transgenerational study, TBT resulted in hepatic steatosis through the F3 generation. Finally TBT was shown to promote hyperglycemia with reduced circulating insulin levels accompanied by increased islet apoptosis and reduced cellular proliferation, suggesting a β-cell defect as a contributing lesion to TBT-induced metabolic dysfunction [ 476 ]. Among the other chemicals summarized in table 1 , developmental exposure to PAHs in a rodent model induce obesity, insulin resistance and inflammation in adults on a high fat diet [ 376 ] [ 377 ]. In a human cohort, children of mothers with the highest PAHs exposure during pregnancy had increased weight at 5 and 7 years of age [ 379 ]. Prenatal exposure to specific PCB congeners in birth cohort studies was shown to result in increased BMI in offspring [ 387 ] [ 390 , 393 ]. Arsenic also deserves mention because of its specific association with insulin dysregulation and T2D and liver toxicity [ 477 , 522 , 528 , 582 ]

Many of the studies discussed above highlight the importance of development as a sensitive time for programming all aspects of metabolism. Environmental chemicals with endocrine activity can alter programming of metabolism; this fact, along with the importance of diet during development and throughout life on metabolism, and the role for exercise in controlling weight and glucose metabolism, leads to the perfect storm for metabolic disease. In utero , and the first few years of life, are critical periods where the sensitivity or set point for obesity, diabetes and liver disease are established. We have herein shown that sensitivity or set points for the development of these diseases can be altered by MDCs that interfere with the normal developmental trajectories of adipose tissue, pancreas, muscle, liver, GI tract and the brain. These set points are also influenced by diet and nutrition in utero and early childhood years, thus nutrition and MDC exposures during development are the key along with genetic background for setting the stage for all metabolic diseases. These metabolic diseases may not manifest until later in life when the system is challenged by over-nutrition and/or lack of exercise. MDCs can change the expression of genes involved in the control of adipogenesis as well as glucose and lipid metabolism. Exposure to MDCs, together with excess calories and lack of exercise, would increase the susceptibility to these disease epidemics. Thus, we propose that some individuals are more prone to gain weight due to both their genetic background and the effects of developmental exposures to MDCs that are critical for setting the sensitivity or susceptibility of the tissues for metabolic disruption later in life. Some recent publications support increased susceptibility to obesity due to environmental exposures. Developmental exposure to BPA induced weight gain due to increased food intake, changes in brain satiety neurons [ 619 ] and decreased activity and energy expenditure in females [ 620 ]. Similarly, prenatal nicotine leads to increased body weight due to a marked hypertrophy of adipocytes and fat deposition along with decreased spontaneous physical activity later in life, cold intolerance, and also increased sensitivity to the effects of high fat diet [ 313 ]. In addition, there are many examples of the effects of developmental exposures to MDCs that are exacerbated by high fat diet later in life, again indicating that developmental exposures increase the susceptibility to obesity (and other metabolic diseases) but may need a “second hit” later in life for the effects to actually become apparent as disease. For example, developmental exposure to PAH results in obesity, insulin resistance and inflammation only after a high fat diet as adults [ 376 ] [ 377 ]; the effect of atrazine on insulin resistance was exacerbated by high fat diet [ 529 ]; and BPA-induced insulin resistance may be additive with high fat diet [ 331 ]. Thus there are emerging data supporting a role for developmental exposures to MDCs in altering the set point of susceptibility for metabolic diseases later in life. The second hit of high fat diet and lack of exercise then results in onset of the metabolic diseases. Of course, throughout life there are likely to be multiple exposures to MDCs that can also increase sensitivity to metabolic disruption leading to weight gain, altered glucose tolerance and lipid disorders. For example, young mice exposed to BPA for 30 days showed significantly increased body weight and fat mass on a chow diet [ 332 ] but not on HFD (45%) which could be due to the overwhelming effect of the HFD. If true, then it would be difficult to control or treat metabolic disorders with pharmaceuticals later in life, as they would need to be able to offset the increased sensitivity programmed during early development. Indeed, the current state of science and medicine focuses on losing weight, and restoring glucose homeostasis and liver lipids after they are disrupted; it is clear that this approach is not working as the incidence of these diseases continues to increase. Furthermore, it is well documented that while it is possible to lose weight and keep it off for an extended time, the vast majority of people will gain the weight back, perhaps indicating they are fighting against a set point or sensitivity to develop these metabolic problems that favors calorie storage over the long term [ 437 , 438 , 621 , 622]. For this reason, a consequence of the MDC hypothesis is that a focus should be on prevention instead of intervention. If indeed a set point for body weight, diabetes and/or METS is developed in early life, then a better approach would be to focus on limiting factors that can alter programming during these sensitive times; for example, addressing these metabolic epidemics will require reducing MDC exposures and improving early-life nutrition. In this way, the MDC hypothesis offers the ability to actually prev