You are on page 1of 16

Biogerontology DOI 10.1007/s10522-011-9325-8



The developmental aging and origins of health and disease hypotheses explained by different protein networks

Bruno Ce´ sar Feltes Joice de Faria Poloni Diego Bonatto

Received: 14 December 2010 / Accepted: 23 February 2011 Springer Science+Business Media B.V. 2011

Abstract One theory that attempts to explain how and why an organism ages is the developmental hypothesis of aging (DevAge), which describes how developmental programming leads to aging in adults. Interestingly, the developmental origins of health and disease hypothesis (DOHaD) asserts that some aging-associated diseases that occur in adults are closely related to development and to conditions in the intrauterine environment. Thus, both aging and aging- associated diseases can be viewed, at least in part, as the result of a developmental program that is activated early in embryogenesis and persists throughout the lifespan of the organism. We would expect this devel- opmental program to be regulated by a set of interacting protein networks that connect environmental and

Electronic supplementary material The online version of this article (doi:10.1007/s10522-011-9325-8) contains supplementary material, which is available to authorized users.

  • B. C. Feltes J. de Faria Poloni

Instituto de Biotecnologia, Universidade de Caxias do Sul,

Caxias do Sul, RS, Brazil

  • D. Bonatto (&)

Centro de Biotecnologia da Universidade Federal do Rio Grande do Sul—Sala 219, Departamento de Biologia Molecular e Biotecnologia, Universidade Federal do Rio Grande do Sul—UFRGS, Avenida Bento Gonc¸alves 9500—Pre´ dio 43421, Caixa Postal 15005, Porto Alegre, RS 91509-900, Brazil e-mail:

Biogerontology DOI 10.1007/s10522-011-9325-8 RESEARCH ARTICLE The developmental aging and origins of health and disease hypotheses explained10.1007/s10522-011-9325-8 ) contains supplementary material, which is available to authorized users. B. C. Feltes J. de Faria Poloni Instituto de Biotecnologia, Universidade de Caxias do Sul, Caxias do Sul, RS, Brazil D. Bonatto ( & ) Centro de Biotecnologia da Universidade Federal do Rio Grande do Sul—Sala 219, Departamento de Biologia Molecular e Biotecnologia, Universidade Federal do Rio Grande do Sul—UFRGS, Avenida Bento Gonc¸alves 9500—Pre´ dio 43421, Caixa Postal 15005, Porto Alegre, RS 91509-900, Brazil e-mail: molecular signals. However, the connection between aging and development is not clear. Thus, a systems biology approach that incorporates different ‘‘omic’’ databases for two mammalian models, Homo sapiens and Mus musculus , was used to evaluate how devel- opment and aging are interconnected. Interestingly, three major, evolutionarily conserved processes, namely the immune system, epigenetics, and aerobic metabolism, appear to regulate aging and development in both H. sapiens and M. musculus . Considering that these three processes are essential to embryogenesis, the protein networks within these processes are sub- jected to strong selective pressure to eliminate gross developmental abnormalities in early embryogenesis. This selective pressure becomes more relaxed in the adult organism, permitting the onset of aging-associ- ated diseases and inflammation-related aging; this concept echoes the antagonistic pleiotropy hypothesis of aging. Keywords Developmental aging (DevAge) Developmental origins of health and disease (DOHaD) Antagonistic pleiotropy hypothesis of aging Systems biology Inflammaging Introduction The mechanism of aging, which can be defined as the loss of physiological aptitude after an organ- ism reaches its maximum reproductive capability 123 " id="pdf-obj-0-45" src="pdf-obj-0-45.jpg">

molecular signals. However, the connection between aging and development is not clear. Thus, a systems biology approach that incorporates different ‘‘omic’’ databases for two mammalian models, Homo sapiens and Mus musculus, was used to evaluate how devel- opment and aging are interconnected. Interestingly, three major, evolutionarily conserved processes, namely the immune system, epigenetics, and aerobic metabolism, appear to regulate aging and development in both H. sapiens and M. musculus. Considering that these three processes are essential to embryogenesis, the protein networks within these processes are sub- jected to strong selective pressure to eliminate gross developmental abnormalities in early embryogenesis. This selective pressure becomes more relaxed in the adult organism, permitting the onset of aging-associ- ated diseases and inflammation-related aging; this

concept echoes the antagonistic pleiotropy hypothesis of aging.


Developmental aging (DevAge)

Developmental origins of health and disease (DOHaD) Antagonistic pleiotropy hypothesis of aging Systems biology Inflammaging


The mechanism of aging, which can be defined as the loss of physiological aptitude after an organ- ism reaches its maximum reproductive capability



(Vijg and Suh 2005), occurs in almost every living organism from prokaryotes to eukaryotes. Many hypotheses have been developed and tested to explain the aging process, and the resulting models attempt to explain aging at the organismal, organ, cellular, and molecular levels (Yin and Chen 2005). From those studies, other important hypotheses regarding the possible triggers of the aging process were defined, such as those concerning the accumu- lation of DNA mutations, free radical overproduction and accumulation, continuous cellular oxidative stress, and telomere shortening (Yin and Chen 2005). Despite intensive research, the mechanisms for how, when and why aging is initiated in multicellular organisms are virtually unknown. The antagonistic pleiotropy hypothesis (APH) of aging attempts to connect the evolutionary and molecular aspects of aging (Williams 1957; Leroi et al. 2005; Ungewitter and Scrable 2009). The APH defines aging as a consequence of the declining force of natural selection; traits that benefit young organ- isms can have unspecified deleterious effects that lead to aging phenotypes later in the adult organism (Williams 1957; Leroi et al. 2005; Ungewitter and Scrable 2009). If these traits are activated early in life, e.g., during embryogenesis, and favor correct tissue development and maturation, then these mech- anisms can escape natural selection but lead to aging later. Another unifying theory, the developmental theory of aging (DevAge), is garnering attention in the fields of both development and aging. DevAge, in its essence, was firstly proposed by the pioneering work of Dilman (1971, 1994) and further extended by the proteomic analysis performed by de Magalha˜ es and Church (2005). Briefly, DevAge suggests that the aging process is the result of a developmental program that remains active for the entire lifespan of an organism (de Magalha˜ es and Church 2005), and this theory agrees with the basic requirements of APH. Such developmental programming will impact not only aging, but also other physiological processes. This idea is corroborated by the obser- vation that tissue differentiation and body pattern formation in a fetus are mainly regulated by maternal nutrition and placental functions that maintain high rates of proliferation, growth and cell differentiation (Gluckman et al. 2010). Deficient

supplies of macro- and micro-nutrients, oxygen and endocrine signals will impact not only the growth of the fetus, but also its future health (Gluckman et al. 2010). In fact, it was demonstrated that mineral deficiencies during perinatal development can result in behavioral, immunological and biochemical abnor- malities that persist into adulthood (Keen et al. 2003). Although these persistent defects can be partially attributed to subtle morphological abnormalities, they may be secondary to development-associated changes in DNA methylation patterns (Keen et al. 2003). Epigenetic defects combined with subtle morpholog- ical abnormalities can influence an individual’s risk for certain chronic diseases and thus influence the risk for morbidity and mortality later in life (Keen et al.


The origin of diseases and onset of aging processes was firstly explained by Thrifty Phenotype (THP) hypothesis (Hales and Barker 1992, 2001) and further expanded to the so-called ‘‘developmental origins of health and disease’’ (DOHaD) or ‘‘developmen- tal programming’’ hypothesis (Langley-Evans 2006; Gluckman et al. 2010). In this scenario, an ensemble of molecular mechanisms senses the environment where the embryo is developing and translates these cues into biochemical information that will program both development and aging. An example of a molecular mechanism that could link DevAge and DOHaD is epigenetics. In embryonic stem cells (ESCs), epigenetic regulation is an essential factor for the differentiation of tissues and the formation of body patterns (Ringrose and Paro 2004; Bibikova et al. 2008). Additionally, many diseases that develop in adult organisms (e.g., type 2 diabetes or neurode- generative diseases) are affected by epigenetic pro- gramming that occurs during embryonic development (Aagaard-Tillery et al. 2008). However, other molec- ular mechanisms that require further explanation are operating during embryonic development to program the fetus for adult life. Here, we examine the DevAge theory compared to the DOHaD hypotheses at the molecular level using a systems biology approach for two mammalian mod- els: Homo sapiens and Mus musculus. Different physical protein–protein interaction (PPPI) networks derived from mammalian interactome projects are described. Moreover, clustering analyses of the major biological processes derived from the PPPI networks



and major centrality identification were conducted to evaluate the molecular and cellular mechanisms underlying the DevAge and DOHaD processes in H. sapiens and M. musculus.

Materials and methods

Interactome network design and global topological analysis

To design binary physical protein–protein interaction (PPPI) networks related to aging and development mechanism, the metasearch engine STRING 8.2 [] was initially applied by using proteins classically described in literature as being related to development (e.g., HOX, PAX, and BMPs proteins); aging and epigenetics (e.g., histone deacet- ylases and sirtuins). The STRING 8.2 software allowed us to visualize the physical connection among proteins related to development, aging, and epigenetic processes. In this sense, the following parameters were used within STRING 8.2: active prediction methods all enabled except text mining; no more than 50 interactions; high confidence score (0.700); and network depth equal to 2. In addition to STRING 8.2, the GeneCards [http://www.genecards. org/], PubMed [], Human Aging Genomic Resources (GenAge) [http://], NetAge database [], and iHop [http://] search engines were also employed using the default parameters. The results gathered from these search engines were subsequently analyzed using Cytoscape 2.6.3 (Shannon et al. 2003). To select the most relevant protein subnetworks from the interactome, the initial PPPI network was analyzed using Molecular Complex Detection (MCODE) (Bader and Hogue 2003), which is a Cytoscape plug-in freely available at The parameters used in MCODE to generate the subnet- works were: loops included; degree cutoff 2; deletion of single connected nodes from cluster (haircut option enabled); expansion of cluster by one neighbor shell allowed (fluff option enabled); node density cutoff 0.1; node score cutoff 0.2; kcore 2; and maximum depth of network 100. A MCODE score was calculated for each protein/small compound present in the interactome networks.

Network centralities and local topological analyses

Major network centralities (node degree and between- ness) were computed from the networks and local topologies. Bottleneck nodes were defined consider- ing all nodes with the highest threshold generated by each centrality calculated by Centiscape 1.0 (Scardoni et al. 2009a, b). A PPPI subnetwork for H. sapiens and M. musculus containing the major bottleneck nodes was drawn from Centiscape 1.0 data. The number of shared bottleneck nodes between H. sapiens and M. musculus PPPI subnetworks was obtained by Venn analysis using the Area-Propor- tional Venn Diagram Plotter and Editor software [].

Gene ontology analysis

The protein motifs generated from MCODE were further studied by focusing on the major biological- associated processes using Biological Network Gene Ontology (BiNGO) software (Maere et al. 2005), which is a Cytoscape plug-in available at http://www. The degree of functional enrichment for a given cluster and category was quantitatively computed (P value) using hypergeometric distribu- tion, and multiple test correction was also assessed by applying the false discovery rate (FDR) algorithm (Benjamini and Hochberg 1995), which was fully implemented through BiNGO software with a sig- nificance level of P \ 0.05.


Analyses of interactome networks and subnetworks associated to embryonic development, aging, and epigenetics in H. sapiens and M. musculus

The increasing number of available ‘‘omic’’ databases has enabled the development of new systematic tools for understanding the interplay between regulatory processes found within a cell. Considering a potential connection between DevAge and DOHaD, we devel- oped a systems biology study by investigating two binary PPPI networks and their associated pathways



using the data available for H. sapiens and M. musculus (Fig. 1). In this sense, a PPPI network that could link well described proteins in the literature (whose functions have been observed in adult tissues) to the aging and development processes (Fig. 1) was searched for both organisms. These networks were expected to contain

Biogerontology using the data available for H. sapiens and M. musculus (Fig. 1 ). In this

Fig. 1 Experimental approach employed to retrieve aging- and development-associated interactome data from human and mouse protein–protein interaction databases. The interactome data were obtained from an initial query using aging- and development-associated proteins, resulting in two physical protein–protein interaction (PPPI) networks. These two net- works were analyzed by Cytoscape software, and the union function of the Merge networks plugin was used to fuse both networks into a unique PPPI network. This final network was then analyzed by MCODE to identify the major associated subgraphs and by Centiscape to retrieve important nodes by centralities analysis. Biological processes found within sub- graphs were retrieved by employing the BiNGO plugin of the Cytoscape software. The information retrieved from literature data mining and network analyses was employed to design a model of how development interplays with aging in human and murine embryos and adult organisms (for more details, please refer to Supplementary Material 1)


proteins that participate in aging and development and have already been described in the literature, such as PAX, HOX, DNA methyltransferases (DNMTs), sirtuins, and histone deacetylases, among others. The data gathered demonstrated that the topologies of PPPI networks are very similar for both H. sapiens (Fig. 2) and M. musculus (Fig. 3). The PPPI network of H. sapiens (Fig. 2) contained 292 nodes and 2,559 connectors, while the PPPI network of M. musculus (Fig. 3) contained 331 nodes and 2,301 connectors. Both PPPI networks were composed of three subnet- works (Figs. 2, 3), each representing different bio- logical processes (Supplementary Material 1). For H. sapiens-associated PPPI network (Fig. 2), the three subnetworks can be classified into: (i) inflammatory response/immune system process; (ii) regulation of transcription/chromatin remodeling/glucose metabo- lism; and (iii) multicellular organismal development. For the PPPI network of M. musculus (Fig. 3), the three subnetworks are: (i) immune system/inflamma- tory response; (ii) regulation of transcription/devel- opment; and (iii) aerobic respiration/glycolysis. Additionally, the networks gathered for H. sapiens (Fig. 2) and M. musculus (Fig. 3) and their corre- sponding subnetworks (Supplementary Material 2) contained proteins that participate in biological processes such as the inflammatory response, tran- scription regulation, epigenetics, glucose metabolism, and developmental processes (Supplementary Mate- rial 1). Interestingly, a comparison of PPPI network of H. sapiens (Fig. 1) with the common signature of longevity and age-related disease proteins (ARDs) network (containing 643 nodes), as described in NetAge database, indicated that 57 nodes are com- monly present in both networks and distributed within the subnetworks (Supplementary Material 1). The topological similarities between H. sapiens (Fig. 2) and M. musculus (Fig. 3) networks prompted us to ask if common pathways or mechanisms of aging and development should be found in these two biological models. By applying centrality analysis to PPPI networks of H. sapiens and M. musculus, evolutionarily conserved pathways and nodes that represent a potential mechanism for the molecular control of DevAge/DOHaD processes were identi- fied. Thus, for both the H. sapiens (Fig. 2) and M. musculus (Fig. 3) networks, we obtained two small subnetworks containing only bottleneck nodes (Fig. 4a and b, respectively). A third subnetwork,


Biogerontology Fig. 2 A physical protein–protein interacting (PPPI) network of major proteins described in the literature

Fig. 2

A physical protein–protein interacting (PPPI) network

of major proteins described in the literature as associated with

aging and development in H. sapiens. Both clustered proteins

containing common (or shared) bottlenecks that define evolutionarily conserved points of control was further analyzed (Fig. 5).

Identification of important nodes within DevAge/ DOHaD-associated interactome networks and subnetworks by centrality analyses

Two major centrality analyses, node degree and betweenness, were performed in this work to identify the most important nodes within the PPPI networks of H. sapiens and M. musculus. ‘‘Node degree’’ is the simplest centrality measure in a given network, and it corresponds to the number of nodes adjacent (directly connected) to a given node (Scardoni et al. 2009a, b). The node degree represents the ‘‘popularity’’ of a given node, and highly connected nodes in a network are termed ‘‘hubs’’ (Yu et al. 2007). ‘‘Betweenness’’ indicates the extent to which a specific node is

(composing different subnetworks) and unclustered proteins are represented by nodes of different shapes and colors, as indicated in the inset

between all other nodes within the network (Newman 2005). In general, betweenness shows the influence of a node over the spread of information throughout the network. The measure of node degree and between- ness gives rise to a third piece of information called ‘‘node bottleneck,’’ which is defined as all nodes with high betweenness values. These nodes are central points that control the communication between other nodes within the network. ‘‘Bottleneck’’ also indi- cates all nodes that are ‘‘between’’ highly intercon- nected subgraph clusters, and removing a bottleneck could divide a network (Girvan and Newman 2002; Yu et al. 2007). Thus, for the human subnetwork, there were 69 bottleneck nodes (Fig. 4a; Supplementary Material 3), while the murine bottleneck-associated subnet- work was formed by 60 nodes (Fig. 4b; Supplemen- tary Material 3). Moreover, a Venn diagram analysis of the human and murine bottleneck-associated



Biogerontology Fig. 3 A PPPI network of major proteins associated with aging and development in M.

Fig. 3

A PPPI network of major proteins associated with aging and development in M. musculus. Clustered proteins (composing

different subnetworks) and unclustered proteins are represented by nodes of different shapes and colors, as indicated in the inset

Fig. 4 Smaller PPPI subnetworks derived from calculated centrality analyses for H. sapiens (a) and M. musculus (b). The color and shape of each node indicates the associated subnetwork (subnetworks 1–3 and unclustered proteins; please refer to the main text of the manuscript for more details)


Biogerontology Fig. 3 A PPPI network of major proteins associated with aging and development in M.


subnetworks identified 40 shared proteins, which represent approximately 58 and 66% of the total nodes observed in the human and murine bottleneck- associated subnetworks, respectively (Fig. 5; Supple- mentary Material 3).

Biogerontology subnetworks identified 40 shared proteins, which represent approximately 58 and 66% of the total nodes

Fig. 5 Area-proportional Venn diagram of the H. sapiens (a) and M. musculus (b) centrality-associated subnetworks. The number of total proteins, the degree of interpolation between subnetworks and the number of unique proteins for each subnetwork are indicated in the Venn diagram. A small PPPI subnetwork derived from proteins that are commonly found in both H. sapiens and M. musculus centrality-associated subnet- works is shown in the box


Interactome networks to link proteins associated with embryonic development, aging, and epigenetics in H. sapiens and M. musculus

The inflammation-epigenetic subnetwork and fetal intrauterine growth restriction linked to aging and epigenetic mechanisms

From the different subnetworks observed within the major interactome networks of H. sapiens (Fig. 2) and M. musculus (Fig. 3), the inflammation-epige- netic cluster contains very interesting proteins that could potentially connect DevAge and DOHaD. It has been reported that inflammation was impli- cated in embryo development by the participation of macrophages during the initial phase of mammalian development (Ovchinnikov 2008). In this sense, the inflammatory subnetworks for both H. sapiens and M. musculus contain cytokines and proteins that stimulate macrophage activation, such as interleukin 18 (IL18) and toll-like receptor 4 (TLR4; Supple- mentary Material 2). Macrophages present in embryonic tissues express a large number of cytokines and mitogenic factors, including IGF1, TGFa, PDGFA, TGFb, ILs, VEGFs, RANKL and LIF (Rappolee and Werb 1988; Hume et al. 2002; Ovchinnikov 2008), which probably stimulate cell proliferation and differentiation. Inter- estingly, a pro-inflammatory cascade observed at the feto-maternal interface (Hess et al. 2007) culminates in NF-jB nuclear translocation and production of secondary mediators, such as chemokines, COX-2, prostaglandins, MMPs, and pentraxin 3 (PTX3), and to the expression of adhesion molecules (Dinarello 1996; Medzhitov 2001; O’Neill 2002; Garlanda et al. 2008). Once the feto-maternal interface is established by inflammatory cytokines, the coordinated expres- sion of angiogenic growth factors, including vascular endothelial growth factor (VEGF) and placenta growth factor, and their respective receptors on invasive trophoblasts occurs (Leonard et al. 2006). Impaired endovascular trophoblast invasion causes inadequate conversion of the uterine arteries and reduced uteroplacental blood flow, leading to fetal intrauterine growth restriction (IUGR) (Redman and Sargent 2005; Garlanda et al. 2008). Fetal IUGR has been associated with premature aging and the



development of aging-associated diseases of adult individuals in various mammalian models (Jennings et al. 1999; Chen et al. 2009; Warner and Ozanne


The connection between the inflammation/immune systems (cluster 1; Figs. 2, 3) and transcriptional regulation/chromatin remodeling (cluster 2; Figs. 2, 3) is a poorly understood aspect of the DevAge/ DOHaD mechanisms. Some major proteins found in both the human and murine cluster 2 (Figs. 1, 2) were as follows: (i) DNA methyltransferase 1 (DNMT1), DNMT3a and DNMT3b; (ii) nuclear or class I histone deacetylase 1 (HDAC1), HDAC2, and HDAC3; (iii) class III HDACs termed sirtuin 1 (SIRT1), and SIRT2; and (iv) p300 (histone acety- lase; HAT) (Supplementary Material 2). These pro- teins belong to major complexes that alter the state of chromatin in response to environmental and/or phys- iological conditions (Ba¨ ckdahl et al. 2009), leading to gene silencing or gene expression induction (Sadoul et al. 2008). In addition to the connection observed between the immune system/inflammatory process and the epige- netic control of gene expression during development, other important proteins were also observed in cluster 2 in both the H. sapiens and M. musculus networks (Figs. 2, 3). These proteins are the tumor suppressor protein p53 and transcription factors, such as the HOX and PAX proteins, that control body pattern formation and organogenesis.

The role of intestine development, immune system, and glucose metabolism on aging

The p53 protein is important not only for its role in cell survival, but also for its debated role as an aging or anti-aging protein (Zafon 2007). In fact, our data analysis indicated that p53 is connected with CDX2, which is linked to the inflammatory/immune subnet- work (Fig. 1). CDX2 is a homeobox protein associated with early differentiation and maintenance of the intestinal epithelium in both the small and large intestines (Suh and Traber 1996). Because the pro- moter region of CDX2 contains an NF-jB binding site, it is clear that immune and inflammatory processes at early embryogenesis can modulate the expression of this gene and affect intestinal development. The intestines of IUGR neonates have reduced weight (proportionate to body weight), length, wall thickness,


villous height, and crypt depth (Baserga et al. 2004). A consequence of these intestinal changes induced by IUGR is that the neonate immune system becomes compromised because the intestinal barrier is involved in the first steps of postnatal immune system matura- tion, protection against food allergens, and protection against environmental microorganisms, affecting aging at later life (Schumann et al. 2005; Willing and Van Kessel 2009). Notably, lymphocytes first appear in the intestinal subepithelium early in development, indicat- ing a strong connection between the digestive and immune systems during development (Veereman- Wauters 1996). Another interesting occurrence observed in both PPPI networks was the presence of a subnetwork associated with glucose metabolism and aerobic respiration (Figs. 2, 3). The presence of citric acid cycle enzymes such as succinate dehydrogenase complex members (SDHA and SDHB; Figs. 2, 3; Supplementary Material 1), aconitase (ACO1 and ACO2; Figs. 2, 3; Supplementary Material 1), and enzymes associated with glucose metabolism such as glycerol-3-phosphate dehydrogenase 1 and 2 (GPD1/ GPD2; Figs. 2, 3; Supplementary Material 1) indi- cated the importance of these metabolic pathways during embryonic development. Other peptides/pro- teins that are important for glucose homeostasis and metabolism are also present in the PPPI networks and include insulin/insulin growth factor 1 (IGF-1) and mTOR (Olovnikov et al. 2009). The data gathered from the systems biology analysis showed that IGF-1 is a node that is connected to SIRT1 and p53 (Figs. 2, 3; Supplemen- tary Material 2). Interestingly, it was reported that lower baseline IGF-1 levels during embryogenesis predict the subsequent development of impaired glucose tolerance, type 2 diabetes and cardiovascular disease, all chronic conditions associated with aging in adults (Chakravarthy et al. 2008).

The role of molecular oxygen homeostasis on development and aging

In addition to their roles in embryonic glucose and general metabolism, both IGF-1 and mTOR are linked to hypoxia-inducible factor-1, subunit a (HIF1A; Figs. 2, 3), which is a transcription factor essential for cellular and systemic homeostatic responses to hypoxia.


During early embryogenesis, HIF1A is activated in response to inflammation and low O 2 tension (Imtiyaz and Simon 2010). This early embryogenesis environment induces the differentiation of various stem cell progenitors, including the differentiation of early mesoderm into hemangioblasts (Fraisl et al. 2009), which results in the early steps of blood vessel development through paracrine release of angiogenic cytokines, such as VEGF, by hematopoi- etic cells. Our data support the role of HIF1A in embryogenesis because this protein is directly linked to HDACs, VEGF nodes, and other proteins related to development and aging (Figs. 2, 3; Supplementary Material 2). Molecular oxygen can also modulate the epige- netic process by regulating HDACs (Okazaki and Maltepe 2006). During early embryogenesis, hypoxia and HIF1A expression correlate with HDAC1 and VEGF activation and result in angiogenesis (Okazaki and Maltepe 2006). Nonetheless, the role of O 2 in DOHaD is not clear, despite the observation that mouse embryos cultured in 20% oxygen in vitro before reimplantation show greater perturbations in global epigenetic patterns than those cultured in 5% oxygen or those that developed in vivo (Rinaudo et al. 2006). In humans, such perturbations cause less- than-optimal embryonic development (Waldenstrom et al. 2009) and lower rates of implantation and live births (Waldenstrom et al. 2009) after culture in an atmospheric (19–21%) oxygen environment. Inter- estingly, a condition termed ‘‘placental hyperoxia’’ has been described, in which high oxygen levels in the intervillous space of the placenta downregulate VEGF, restrict angiogenesis and lead to IUGR (Khaliq et al. 1999).

A common regulatory pathway for development and aging in H. sapiens and M. musculus

The bottleneck nodes shared by humans and mice contained important aging- and development-associ- ated proteins (Fig. 5; Supplementary Material 3), such as p53, EP300, HOXs, HDAC1, SIRT1, and NF- jB. It is important to note that a similar work published by de Magalha˜ es and Toussaint (2009) reports a connection between development and aging by exploring an H. sapiens proteomic network, supporting our data.

Moreover, other interesting aging- and develop- ment-associated bottleneck nodes were also repre- sented in this shared bottleneck subnetwork, including PAXs, BMP4, RAD50, and SMAD4 (Fig. 5; Supple- mentary Material 3). The functions of the PAX proteins are well known in development and stem cell maintenance (Lang et al. 2007). The PAX proteins are transcription factors that, together with HOXs, coordinate organ- ogenesis (Plaza et al. 2008). In this work, three PAX proteins are described: PAX2, PAX6, and PAX9 (Fig. 5; Supplementary Material 3). PAX2 is neces- sary for the development of the urogenital tract, eyes, and CNS, while PAX6 acts in the formation of the eyes, nose, central nervous system and pancreas. PAX9 is required for the development of the thymus, parathyroid glands, ultimobranchial bodies, teeth, skeletal elements of the skull, larynx, and distal limbs ( Interestingly, it has been demonstrated that IUGR induced by maternal protein restriction in a murine model leads to hypertension and a reduced number of glomeruli in the kidneys, which affects lifespan, especially when these animals are exposed to a high-protein diet (Chen et al. 2010). In addition, IUGR induces atrophy of the thymus (Cox and Marton 2009) and affects the development of pancreatic islets by altering the epigenetic mech- anisms of these cells, leading to the development of type 2 diabetes (Thompson et al. 2010). These facts show that PAX proteins appear to be important tar- gets of epigenetic control, and our data support this hypothesis in the context of major bottleneck proteins (Fig. 5). Unfortunately, and despite the intensive research related to PAXs, it is not completely clear how these proteins acts in aging. BMP4 is another interesting bottleneck that appears in the subnetwork (Fig. 5; Supplementary Material 3). This protein is necessary for cartilage and bone formation through mesoderm induction, tooth development, limb formation and fracture repair ( While the role of BMP4 in aging is less understood, some reports have indicated that BMP4 is an important mediator of atrophic age-related macular degeneration because it mediates oxidative stress-induced retinal pigment epithelium senescence in vitro via the SMAD and p38 pathways (Zhu et al. 2009). In addition, overexpres- sion of BMP4 was able to induce premature senes- cence in lung cancer cells (Su et al. 2009). The role of



BMP4 in epigenetic regulation and in IUGR is also not well understood, but a recent article by Muroh- ashi et al. (2010) showed that trophoblast cells pro- duce BMP4 by means of CDX2 to promote the proper growth of early mouse embryos. Because early embryogenesis correlates with rapid mitosis of ESCs (Luo et al. 1999), it is not surprising that DNA damage, especially DNA single- and double-strand breaks (SSBs and DSBs, respectively), could accumulate during this period. In fact, research has shown that DNA repair proteins are implicated in early embryogenesis (Chuykin et al. 2008). Similarly, RAD50 (Fig. 5; Supplementary Material 3), which is a major component of the RAD50/MRE11/NBS1 (MRS) complex that acts in DNA repair and chro- matin remodeling (Iijima et al. 2008), was present in the bottleneck-associated subnetwork. RAD50 is important during embryogenesis, and RAD50 mutant ESCs in a murine model are not viable, especially during the phase of accelerated cell proliferation (Luo et al. 1999). The failure of cells to proliferate could be attributed to the accumulation of unrepaired or misrepaired spontaneous DSBs over several cell cycles (Luo et al. 1999). Normally, however, rapid cell cycling and the formation of DSBs could be the initial trigger that results in the secretion of IL-6 and/ or other inflammatory cytokines, which promote angiogenesis and cell differentiation, and that has been observed for some tumor types (Rodier et al.


Finally, SMAD4, a bottleneck node that is highly connected with PAX, BMP, p53 and other proteins (Fig. 5; Supplementary Material 3), was conserved between murine and human models. The role of SMAD4 in embryogenesis has been well documented (Costello et al. 2009), and SMAD4-null embryos arrest shortly after implantation due to defects in the extra-embryonic lineages (Yang et al. 1998; Chu et al. 2004). Additionally, SMAD4 has important roles in inflammation and hematopoiesis (Karlsson et al. 2007) because this protein is necessary for promotion of hematopoietic stem cell renewal, as well as maintenance of colon structure and homeo- stasis (Karlsson et al. 2007). Other nodes could be observed in the bottleneck- associated subnetwork, including the SWI/SNF- related, matrix-associated, actin-dependent regulator of chromatin, subfamily a, member 4 (SMARCA4; Fig. 5 and Supplementary Material 3); the nuclear


receptor coactivator 3 (NCOA3; Fig. 4 and Supple- mentary Material 3); the transcription factor YY1 (Fig. 5 and Supplementary Material 3); and the polycomb group RING finger protein 4 (BMI1; Fig. 5 and Supplementary Material 3). These proteins were described as necessary for some aspects of development and for the control of epigenetic mech- anisms (Kim et al. 2006; Magnani and Cabot 2009; Kashyap and Gudas 2010), but their impact on aging is less understood, making these proteins interesting targets for DevAge/DOHaD studies.

A model linking DevAge with DOHaD in an antagonistic pleiotropy view

The idea that the developmental changes that occur in the intrauterine environment can have impacts on adult organisms can be found initially in the Thrifty Phenotype (THP) hypothesis. The THP hypothesis posits that during evolution a set of key genes (thrifty genes) were selected to provide resistance to starva- tion by triggering a fast insulin response that would facilitate the restoration of adipose tissue (Hales and Barker 1992, 2001). The hypothesis also suggests that this adaptive response in pregnant individuals could cause the regulation and growth of critical organs in the fetus while being simultaneously capable of causing the malformation of other organs, leading to changes in fetal and adult metabolism (McMillen and Robinson 2005). Moreover, the idea that the embry- onic process can adaptively respond to specific intrauterine changes raises the hypothesis that these errors in the developmental process could ‘‘program’’ certain conditions later in life because these early settings could predetermine metabolic function and morphological conditions (McMillen and Robinson 2005). Thus, essential responses or a process that takes place during embryonic development could be detrimental later in life. The THP hypothesis was further expanded to the so-called ‘‘Predictive Adap- tive Response’’ (PAR) hypothesis (Gluckman and Hanson 2006), which proposed that the fetus dynam- ically interacts with and assesses the environment into which it will be born and adapts to gain a future survival advantage (Gluckman and Hanson 2006). Finally, both the THP and PAR hypotheses were joined in DOHaD, which considers extrinsic and intrinsic factors that affect the developmental


program by changing the expression of specific genes associated with development and epigenetic control (Jones and Ozanne 2009). The accepted definitions of DOHaD and, as described before, of DevAge can be better understood if we consider the mechanism of APH. Both DOHaD and DevAge presuppose that early embryogenesis and its molecular pathways of development, which are regulated by the intrauterine environment, define the lifespan of the organism, including aging and aging-associated diseases. Thus, the molecular path- ways involved in development and differentiation are strongly subjected to natural selection and other evolutionary pressures (Richardson 1999). Epigenetic alterations (as a consequence of the embryo-associ- ated environment) affect gene expression in devel- oping tissues in response to selection to act against individuals that show gross anatomical or physiolog- ical abnormalities (Richardson 1999). Our model includes DOHaD/DevAge in the same conditions of development (Fig. 6), in which initial embryogenesis occurs in a low O 2 environment, leading to vasculogenesis and early oxidative stress signaling that culminate with the production of pro- inflammatory cytokines by the mother and by the differentiating ESCs (Fig. 6). Once the vascular connection occurs between the placenta and fetus, there is a gradual increase in the fetal tissue O 2 concentration, which establishes aerobic respiration in the fetus and starts the developmental program. Under normal conditions, adequate levels of O 2 and aerobic metabolism activate the developmental/ epigenetic program by regulating a series of proteins that target the chromatin structure, controlling the progression of the cell cycle and cell fate. During this phase of embryogenesis, fetal stem cell niches are formed in various tissues (Fig. 6), including intesti- nal, skin, neural, germ line, and hematopoietic tissues (Li and Xie 2005). These niches have important roles in the maintenance of tissue homeostasis during the adult life because they are required for functions such as wound healing and immune system modulation (Li and Xie 2005). The three major processes observed in our systems biology data model (immune system/inflammation, development/epigenetics, and aerobic respiration/ glucose metabolism; Fig. 6) possibly trigger the aging mechanism at the cellular and molecular levels because aging itself is important for tissue

differentiation and maturation. These processes should be strongly subjected to selective pressure (Akam 1998; Richardson 1999), and the same is most likely true for embryo-associated aging. This evolu- tive surveillance mechanism ensures that the embryo will develop correctly. When the fetus is establishing the placental connection with the mother, the intrauterine environ- ment is strongly regulated by external factors, such as the mother’s nutritional status (Fig. 6). These envi- ronmental factors can induce epigenetic effects during early development to influence various stable gene expression patterns that persist even after the disappearance of the initial signal and may continue to determine a distinct phenotype (Vaiserman 2008). In the adult organism, the same three major processes that are directly associated with DevAge/ DOHaD (Fig. 6) also participate in homeostasis. For example, wound healing is an essential mechanism that maintains the viability of adult tissues. The response to wounding is composed of three well- described and interdependent phases: (i) inflamma- tion, (ii) new tissue formation, and (iii) tissue remodeling (Gurtner et al. 2008). The initial reaction following cutaneous wounding is the production of inflammatory mediators, including IL-1, TNFa, and IL-6, in the dermis and epidermis. This first reaction is important for the initial removal of cellular debris and for the regulation of sequential responses after wounding, such as tissue formation and remodeling (Adams 2009). Overall, the processes of wound healing and immune modulation require the intense participation of adult stem cells (Rando 2006), which are mobilized from their niches (Rando 2006; Fig. 6). Glucose metabolism and aerobic respiration also have important roles in the aging and development of an adult organism. Chronic diseases, typically asso- ciated with aging and mainly regulated by environ- mental factors, e.g., type 2 diabetes (Kawahito et al. 2009), can affect the metabolism of glucose by inducing so-called ‘‘glucose toxicity’’, a clinical condition in which hyperglycemia itself reduces the insulin secretion capacity of pancreatic b-cells, and the resultant increase in insulin resistance leads to further hyperglycemia. This vicious cycle finally leads to the inability of b-cells to secrete insulin (Kawahito et al. 2009). As described before, insulin resistance could be pre-determined during embryo- genesis under IUGR conditions.



Biogerontology Fig. 6 Aging and development as unique processes in H. sapiens and M. musculus as

Fig. 6 Aging and development as unique processes in H. sapiens and M. musculus as observed by systems biology analysis. Embryogenesis takes place in the intrauterine environment, where the placenta connects the developing embryo with the mother’s major physiological systems, such as nutrient acquisition and metabolism. In addition, the placenta translates signals from the mother’s environment to the embryo. The establishment of the embryo–placenta–mother interface occurs together with the activation of the immune system/inflammatory process, epigenetics/development pro- cess, and glucose metabolism/aerobic respiration in embryonic tissues during early embryogenesis. Initially, the secretion of pro-inflammatory cytokines by the mother’s immune system due to the placenta–embryo connection, some embryonic structures (e.g., the intestine), and the low O 2 tension observed in the uterus stimulate embryo tissue differentiation. As a consequence of this developmental process, a molecular- cellular aging mechanism is established in the embryo that facilitates differentiation. Additionally, during embryogenesis, stem cell niches are formed in different tissues. The connection of the mother to the embryo by the placenta determines how the embryo will develop and age (developmental aging; DevAge) by adapting the embryo to the intrauterine and


maternal environmental conditions. This environmental adap- tation, which is mainly regulated by the immune/epigenetics/ aerobic metabolism processes, can also predispose the organ- ism to chronic diseases or physiological dysfunctions (devel- opmental origins of health and disease; DOHaD). It should be noted that the fetal immune system, epigenetics and development mechanisms, and glucose/aerobic metabolism are strongly subjected to natural selection, eliminating poorly adapted embryos. However, all surviving embryos that adapt to the uterine environment have the potential to develop into an adult individual. The same mechanisms that act during embryogenesis are also observed in adult tissues, promoting the physiological homeostasis of the organism and mobilizing adult stem cells from their niches to repair damaged tissues. However, unlike during embryonic development, the selective pressure becomes relaxed as the organism ages, allowing the accumulation of damage (e.g., oxidative damage) to molecules, biological processes, and tissues. Relaxed selective pressure allows the development of aging-associated chronic diseases (e.g., type 2 diabetes and cardiovascular diseases) as well as a mildly systemic chronic inflammation that correlates with aging in adults


Hyperglycemia, in turn, increases the production of ROS, leading to the formation of advanced glycation end products (AGEs) in tissues (Stoppera et al. 2003; Baynes 2001). A large body of evidence suggests that AGEs are important pathogenic medi- ators for various diseases, such as diabetic compli- cations, Alzheimer’s disease, atherosclerosis, and dialysis-related amyloidosis (Hou et al. 1997; Yan et al. 2003; Sun et al. 1998). Additionally, AGEs can induce inflammation in different tissues by binding to the AGE receptor (AGER) (Bierhaus et al. 2005). Interestingly, the activation of AGER by AGEs released from cell debris can contribute to the wounding response in a non-aged organism by inducing acute inflammation, which is an essential condition for tissue healing (Adams 2009). However, it is also possible that chronic activation of the wound response pathway by AGE accumulation and AGER signaling due to impaired glucose metabolism could increase the rate of tissue senescence by activating inflammation. Furthermore, senescence has been demonstrated to facilitate wound healing under normal physiological conditions (Adams 2009), sup- porting our idea that aging is a fundamental process for tissue differentiation and maturation. However, the overstimulation of wound healing may ultimately become a source of tissue damage and may even cause age-associated tissue degeneration, cancer (Adams 2009), or inflammaging (Fig. 6). Considering APH, the strong selective pressure observed during embryogenesis becomes relaxed as the adult organism ages. The cumulative damages induced by glucose metabolism/aerobic respira- tion, epigenetic modulation/development process, and immune system/inflammation augment the pre- determined conditions that originated from the DevAge/DOHaD programming, leading to aging- associated chronic diseases (Fig. 6). It should be noted that the relationship between aging-associated chronic diseases and development has been recently described by means of network analyses (Budovsky et al. 2009). In this sense, a comparison of the common signature of longevity and age-related disease proteins (ARDs) network present in the NetAge database (Tacutu et al. 2010) and described in the manuscript of Budovsky et al. (2009), with the PPPI network of H. sapiens, indicated the presence of 57 common proteins, all participating in epigenetic mechanism, development, and aging.

Interestingly, it has been proposed that epigenetic regulation could have a so deep impact in aging and aging-associated chronic diseases that the term ‘‘epi- genetic antagonistic pleiotropy’’ (EPH) was pro- posed, similarly to APH (Budovsky et al. 2006). The concept of EPH could be easily adapted to development, where a strong natural selection force molds the global epigenetic patterns of a developing embryo, and consequently defines how an organism will age and will develop aging-associated chronic diseases (Attig et al. 2010; Mun˜ oz-Najar and Sedivy 2011). Noteworthy, experimental data obtained from C. elegans using RNAi for the screening of long- lived individuals showed that early gene inactivations of essential genes associated to development, metab- olism, and chromatin structure caused growth arrest at larval stages, resulting in long-lived animals (Curran and Ruvkun 2007). In conclusion, the systems biology data gathered in this review indicate that the mechanisms that mod- ulate early embryogenesis and aging in mammals are strongly regulated by a network of different proteins that connect both aging and development and are subjected to environmental regulation. In other words, intrauterine factors epigenetically regulate the adaptability of the embryo to its environment. Once viable, the organism will develop to adulthood. The same mechanisms that controlled early embryo- genesis are still present in the adult but are less subject to natural selection pressures, which results in the origin of environment-associated chronic diseases and aging.

Acknowledgments This work was supported by research grants from the Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´ gico (CNPq; Grant number 471769/2007-0 and 300301/2009-0) and the Programa Institutos Nacionais de Cieˆ ncia e Tecnologia (INCT de Processos Redox em Biomedicina-Redoxoma; Grant number 573530/2008-4). Contract/grant sponsor: CNPq.


Aagaard-Tillery KM, Grove K, Bishop J, Ke X, Fu Q, McKnight R, Lane RH (2008) Developmental origins of disease and determinants of chromatin structure: maternal diet modifies the primate fetal epigenome. J Mol Endo- crinol 41:91–102 Adams PD (2009) Healing and hurting: molecular mechanisms, functions, and pathologies of cellular senescence. Mol Cell 36:1–14



Akam M (1998) Hox genes, homeosis and the evolution of segment identity: no need for hopeless monsters. Int J Dev Biol 42:445–451 Attig L, Gabory A, Junien C (2010) Early nutrition and epi- genetic programming: chasing shadows. Curr Opin Clin Nutr Metab Care 13:284–293 Ba¨ ckdahl L, Bushell A, Beck S (2009) Inflammatory signalling as mediator of epigenetic modulation in tissue-specific chronic inflammation. Int J Biochem Cell Biol 41:176–184 Bader GD, Hogue CW (2003) An automated method for finding molecular complexes in large protein interaction networks. BMC Bioinform 4:2 Baserga M, Bertolotto C, Maclennan NK, Hsu JL, Pham T, Laksana GS, Lane RH (2004) Uteroplacental insufficiency decreases small intestine growth and alters apoptotic homeostasis in term intrauterine growth retarded rats. Early Hum Dev 79:93–105 Baynes JW (2001) The role of AGEs in aging: causation or correlation. Exp Gerontol 36:1527–1537 Benjamini Y, Hochberg Y (1995) Controlling the false dis- covery rate: a practical and powerful approach to multiple testing. J R Stat Soc B Stat Methodol 57:289–300 Bibikova M, Laurent LC, Ren B, Loring JF, Fan JB (2008) Unraveling epigenetic regulation in embryonic stem cells. Cell Stem Cell 2:123–134 Bierhaus A, Humpert PM, Morcos M, Wendt T, Chavakis T, Arnold B, Stern DM, Nawroth PP (2005) Understanding RAGE, the receptor for advanced glycation end products. J Mol Med 83:876–886 Budovsky A, Muradian KK, Fraifeld VE (2006) From disease- oriented to aging/longevity-oriented studies. Rejuvenation Res 9:207–210 Budovsky A, Tacutu R, Yanai H, Abramovich A, Wolfson M, Fraifeld V (2009) Common gene signature of cancer and longevity. Mech Ageing Dev 130:33–39 Chakravarthy MV, Zhu Y, Wice MB, Coleman T, Pappan KL, Marshall CA, McDaniel ML, Semenkovich CF (2008) Decreased fetal size is associated with beta-cell hyper- function in early life and failure with age. Diabetes


Chen JH, Martin-Gronert MS, Tarry-Adkins JL, Ozanne SE (2009) Maternal protein restriction affects postnatal growth and the expression of key proteins involved in lifespan regulation in mice. PLoS ONE 4:e4590 Chen J, Xu H, Shen Q, Guo W, Sun L (2010) Effect of postnatal high-protein diet on kidney function of rats exposed to intrauterine protein restriction. Pediatr Res


Chu GC, Dunn NR, Anderson DC, Oxburgh L, Robertson EJ (2004) Differential requirements for Smad4 in TGFbeta- dependent patterning of the early mouse embryo. Devel- opment 131:3501–3512 Chuykin IA, Lianguzova MS, Pospelova TV, Pospelov VA (2008) Activation of DNA damage response signaling in mouse embryonic stem cells. Cell Cycle 7:2922–2928 Costello I, Biondi CA, Taylor JM, Bikoff EK, Robertson EJ (2009) Smad4-dependent pathways control basement membrane deposition and endodermal cell migration at early stages of mouse development. BMC Dev Biol 9:54

Cox P, Marton T (2009) Pathological assessment of intrauter- ine growth restriction. Best Pract Res Clin Obstet Gynaecol 23:751–764 Curran SP, Ruvkun G (2007) Lifespan regulation by evolu- tionarily conserved genes essential for viability. PLoS Genet 3:e56 de Magalha˜ es JP, Church GM (2005) Genomes optimize reproduction: aging as a consequence of the develop- mental program. Physiology 20:252–259 de Magalha˜ es JP, Toussaint O (2009) GenAge: a genomic and proteomic network map of human ageing. FEBS Lett


Dilman VM (1971) Age-associated elevation of hypothalamic threshold to feedback control and its role in development, aging and disease. Lancet 1:1211–1219 Dilman VM (1994) Development, aging, and disease. A new rationale for an intervention. Harwood Academic Pub- lisher, Chur Dinarello CA (1996) Biologic basis for interleukin-1 in dis- ease. Blood 87:2095–2147 Fraisl P, Mazzone M, Schmidt T, Carmeliet P (2009) Regu- lation of angiogenesis by oxygen and metabolism. Dev Cell 16:167–179 Garlanda C, Maina V, Martinez de la Torre Y, Nebuloni M, Locati M (2008) Inflammatory reaction and implantation:

the new entries PTX3 and D6. Placenta 29:129–134 Girvan M, Newman ME (2002) Community structure in social and biological networks. Proc Natl Acad Sci USA


Gluckman PD, Hanson MA (2006) The consequences of being born small—an adaptive perspective. Horm Res 65:5–14 Gluckman PD, Hanson MA, Buklijas T (2010) A conceptual framework for the developmental origins of health and disease. J Dev Orig Health Dis 1:6–18 Gurtner GC, Werner S, Barrandon Y, Longaker MT (2008) Wound repair and regeneration. Nature 453:314–321 Hales CN, Barker DJP (1992) Type 2 (non-insulin-dependent) diabetes mellitus: the thrifty phenotype hypothesis. Dia- betologia 35:595–601 Hales CN, Barker DJP (2001) The thrifty phenotype hypoth- esis. Br Med Bull 60:5–20 Hess AP, Hamilton AE, Talbi S, Dosiou C, Nyegaard M, Nayak N, Genbecev-Krtolica O, Mavrogianis P, Ferrer K, Kruessel J, Fazleabas AT, Fisher SJ, Giudice LC (2007) Decidual stromal cell response to paracrine signals from the trophoblast: amplification of immune and angiogenic modulators. Biol Reprod 76:102–117 Hou FF, Chertow GM, Kay J, Boyce J, Lazarus JM, Braatz JA, Owen WF Jr (1997) Interaction between beta 2-micro- globulin and advanced glycation end products in the development of dialysis related-amyloidosis. Kidney Int


Hume DA, Ross IL, Himes SR, Sasmono RT, Wells CA, Ravasi T (2002) The mononuclear phagocyte system revisited. J Leukoc Biol 72:621–627 Iijima K, Ohara M, Seki R, Tauchi H (2008) Dancing on damaged chromatin: functions of ATM and the RAD50/ MRE11/NBS1 complex in cellular responses to DNA damage. J Radiat Res 49:451–464



Imtiyaz HZ, Simon MC (2010) Hypoxia-inducible factors as essential regulators of inflammation. Curr Top Microbiol Immunol 345:105–120 Jennings BJ, Ozanne SE, Dorling MW, Hales CN (1999) Early growth determines longevity in male rats and may be related to telomere shortening in the kidney. FEBS Lett


Jones RH, Ozanne SE (2009) Fetal programming of glucose- insulin metabolism. Mol Cell Endocrinol 297:4–9 Karlsson G, Blank U, Moody JL, Ehinger M, Singbrant S, Deng CX, Karlsson S (2007) Smad4 is critical for self- renewal of hematopoietic stem cells. J Exp Med 204:


Kashyap V, Gudas LJ (2010) Epigenetic regulatory mecha- nisms distinguish retinoic acid-mediated transcriptional responses in stem cells and fibroblasts. J Biol Chem


Kawahito S, Kitahata H, Oshita S (2009) Problems associated with glucose toxicity: role of hyperglycemia-induced oxidative stress. World J Gastroenterol 15:4137–4142 Keen CL, Clegg MS, Hanna LA, Lanoue L, Rogers JM, Daston GP, Oteiza P, Uriu-Adams JY (2003) The plausibility of micronutrient deficiencies being a significant contributing factor to the occurrence of pregnancy complications. J Nutr 133:1597S–1605S Khaliq A, Dunk C, Jiang J, Shams M, Li XF, Acevedo C, Weich H, Whittle M, Ahmed A (1999) Hypoxia down- regulates placenta growth factor, whereas fetal growth restriction up-regulates placenta growth factor expression:

molecular evidence for ‘‘placental hyperoxia’’ in intra- uterine growth restriction. Lab Investig 79:151–170 Kim SY, Paylor SW, Magnuson T, Schumacher A (2006) Juxtaposed Polycomb complexes co-regulate vertebral identity. Development 133:4957–4968 Lang D, Powell SK, Plummer RS, Young KP, Ruggeri BA (2007) PAX genes: roles in development, pathophysiol- ogy, and cancer. Biochem Pharmacol 73:1–14 Langley-Evans SC (2006) Developmental programming of health and disease. Proc Nutr Soc 65:97–105 Leonard S, Murrant C, Tayade C, van den Heuvel M, Watering R, Croy BA (2006) Mechanisms regulating immune cell contributions to spiral artery modification-facts and hypotheses—a review. Placenta 27:S40–S46 Leroi A, Bartke A, de Benedictis G, Franceschi C, Gartner A, Gonos ES, Fedei ME, Kivisild T, Lee S, Kartaf-Ozer N, Schumacher M, Sikora E, Slagboom E, Tatar M, Yashin AI, Vijg J, Zwaan B (2005) What evidence is there for the existence of individual genes with antagonistic pleiotropic effects? Mech Ageing Dev 126:421–429 Li L, Xie T (2005) Stem cell niche: structure and function. Annu Rev Cell Dev Biol 21:605–631 Luo G, Yao MS, Bender CF, Mills M, Bladl AR, Bradley A, Petrini JH (1999) Disruption of mRad50 causes embry- onic stem cell lethality, abnormal embryonic develop- ment, and sensitivity to ionizing radiation. Proc Natl Acad Sci USA 96:7376–7381 Maere S, Heymans K, Kuiper M (2005) BiNGO: a cytoscape plugin to assess overrepresentation of gene ontology cate- gories in biological networks. Bioinformatics 21:3448–


Magnani L, Cabot RA (2009) Manipulation of SMARCA2 and SMARCA4 transcript levels in porcine embryos differ- entially alters development and expression of SMAR- CA1, SOX2, NANOG, and EIF1. Reproduction 137:


McMillen IC, Robinson JS (2005) Developmental origins of the metabolic syndrome: prediction, plasticity, and pro- gramming. Physiol Rev 85:571–633 Medzhitov R (2001) Toll-like receptors and innate immunity. Nat Rev Immunol 1:135–145 Mun˜ oz-Najar U, Sedivy JM (2011) Epigenetic control of aging. Antioxid Redox Signal 14:241–259 Murohashi M, Nakamura T, Tanaka S, Ichise T, Yoshida N, Yamamoto T, Shibuya M, Schlessinger J, Gotoh N (2010) An FGF4-FRS2a-Cdx2 axis in trophoblast stem cells induces Bmp4 to regulate proper growth of early mouse embryos. Stem Cells 28:113–121 Newman MEJ (2005) A measure of betweenness centrality based on random walks. Soc Netw 27:39–54 O’Neill LA (2002) Toll-like receptor signal transduction and the tailoring of innate immunity: a role for Mal? Trends Immunol 23:296–300 Okazaki K, Maltepe E (2006) Oxygen, epigenetics and stem cell fate. Regen Med 1:71–83 Olovnikov IA, Kravchenko JE, Chumakov PM (2009) Homeo- static functions of the p53 tumor suppressor: regulation of energy metabolism and antioxidant defense. Semin Cancer Biol 19:32–41 Ovchinnikov DA (2008) Macrophages in the embryo and beyond: much more than just giant phagocytes. Genesis


Plaza S, Prince F, Adachi Y, Punzo C, Cribbs DL, Gehring WJ (2008) Cross-regulatory protein–protein interactions between Hox and Pax transcription factors. Proc Natl Acad Sci USA 105:13439–13444 Rando TA (2006) Stem cells, ageing and the quest for immortality. Nature 441:1080–1086 Rappolee DA, Werb Z (1988) Secretory products of phago- cytes. Curr Opin Immunol 1:47–55 Redman CW, Sargent IL (2005) Latest advances in under- standing preeclampsia. Science 308:1592–1594 Richardson MK (1999) Vertebrate evolution: the develop- mental origins of adult variation. Bioessays 21:604–613 Rinaudo PF, Giritharan G, Talbi S, Dobson AT, Schultz RM (2006) Effects of oxygen tension on gene expression in preimplantation mouse embryos. Fertil Steril 86:1252–1265 Ringrose L, Paro R (2004) Epigenetic regulation of cellular memory by the Polycomb and Trithorax group proteins. Annu Rev Genet 38:413–443 Rodier F, Coppe´ JP, Patil CK, Hoeijmakers WA, Mun˜ oz DP, Raza SR, Freund A, Campeau E, Davalos AR, Campisi J (2009) Persistent DNA damage signalling triggers senes- cence-associated inflammatory cytokine secretion. Nat Cell Biol 11:973–979 Sadoul K, Boyault C, Pabion M, Khochbin S (2008) Regulation of protein turnover by acetyltransferases and deacetylases. Biochimie 90:306–312 Scardoni G, Petterlini M, Laudanna C (2009) Analyzing bio- logical network parameters with CentiScaPe. Bioinfor-

matics 25:2857–2859



Schumann A, Nutten S, Donnicola D, Comelli EM, Mansou- rian R, Cherbut C, Corthesy-Theulaz I, Garcia-Rodenas C (2005) Neonatal antibiotic treatment alters gastrointestinal tract developmental gene expression and intestinal barrier transcriptome. Physiol Genomics 23:235–245 Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T (2003) Cytoscape: a software environment for integrated models of biomo- lecular interaction networks. Genome Res 13:2498–2504 Stoppera H, Schinzelb R, Sebekovac K, Heidlandd A (2003) Genotoxicity of advanced glycation end products in mammalian cells. Cancer Lett 190:151–156 Su D, Zhu S, Han X, Feng Y, Huang H, Ren G, Pan L, Zhang Y, Lu J, Huang B (2009) BMP4-Smad signaling pathway mediates adriamycin-induced premature senescence in lung cancer cells. J Biol Chem 284:12153–12164 Suh E, Traber PG (1996) An intestine-specific homeobox gene regulates proliferation and differentiation. Mol Cell Biol


Sun M, Yokoyama M, Ishiwata T, Asano G (1998) Deposition of advanced glycation end products (AGE) and expression of the receptor for AGE in cardiovascular tissue of the diabetic rat. Int J Exp Pathol 79:207–222 Tacutu R, Budovsky A, Fraifeld VE (2010) The NetAge database: a compendium of networks for longevity, age- related diseases and associated processes. Biogerontology


Thompson RF, Fazzari MJ, Niu H, Barzilai N, Simmons RA, Greally JM (2010) Experimental intrauterine growth restriction induces alterations in DNA methylation and gene expression in pancreatic islets of rats. J Biol Chem


Ungewitter E, Scrable H (2009) Antagonistic pleiotropy and p53. Mech Ageing Dev 130:10–17 Vaiserman AM (2008) Epigenetic engineering and its possible role in anti-aging intervention. Rejuvenation Res 11:39–42 Veereman-Wauters G (1996) Neonatal gut development and postnatal adaptation. Eur J Pediatr 155:627–632

Vijg J, Suh Y (2005) Genetics of longevity and aging. Annu Rev Med 56:193–212 Waldenstrom U, Engstrom AB, Hellberg D, Nilsson S (2009) Low-oxygen compared with high-oxygen atmosphere in blastocyst culture, a prospective randomized study. Fertil Steril 91:2461–2465 Warner MJ, Ozanne SE (2010) Mechanisms involved in the developmental programming of adulthood disease. Bio- chem J 427:333–347 Williams GC (1957) Pleiotropy, natural selection, and the evolution of senescence. Evolution 11:398–411 Willing BP, Van Kessel AG (2009) Intestinal microbiota dif- ferentially affect brush border enzyme activity and gene expression in the neonatal gnotobiotic pig. J Anim Physiol Anim Nutr 93:586–595 Yan SF, Ramasamy R, Naka Y, Schmidt AM (2003) Glycation, inflammation, and RAGE: a scaffold for the macrovas- cular complications of diabetes and beyond. Circ Res


Yang X, Li C, Xu X, Deng C (1998) The tumor suppressor SMAD4/DPC4 is essential for epiblast proliferation and mesoderm induction in mice. Proc Natl Acad Sci USA


Yin D, Chen K (2005) The essential mechanisms of aging:

irreparable damage accumulation of biochemical side- reactions. Exp Gerontol 40:455–465 Yu H, Kim PM, Sprecher E, Trifonov V, Gerstein M (2007) The importance of bottlenecks in protein networks: cor- relation with gene essentiality and expression dynamics. PLoS Comput Biol 3:e59 Zafon C (2007) Jekyll and Hyde, the p53 protein, pleiotropics antagonisms and the thrifty aged hypothesis of senes- cence. Med Hypotheses 68:1371–1377 Zhu D, Deng X, Xu J, Hinton DR (2009) What determines the switch between atrophic and neovascular forms of age related macular degeneration?—the role of BMP4 induced senescence. Aging 1:740–745