Search

Additional file 1: Figure S1. DNA methylation and NuRD toolkit genes in mammals. (A) Proteins involved in DNA methylation and demethylation include Dnmt3A/B responsible for de novo 5-methyl-cytosine (5mC) formation, Dnmt1 required for 5mC maintenance during DNA replication, Uhrf1 which binds 5mC and recruits Dnmt1, and Tet and Tdg which are involved in active demethylation. Passive demethylation through cell divisions is also indicated. (B) The Nucleosome Remodeling and Deacetylase complex (NuRD complex) is recruited on methylated DNA and represses gene transcription. The NuRD complex is composed of two subcomplexes: one made of Mbd2/3 (which binds methylated cytosines), Gatad2a/b, and Chd3/4, and which acts on chromatin remodeling; and the other composed of Rbbp4/7, Mta1/2/3 and Hdac1/2, which stimulates histone deacetylation. For the sake of simplicity, the NuRD complex is depicted in a schematic manner that does not reflect its real stoichiometry.

Additional file 9: Figure S7. Additional expression data of 5mC and NuRD genes during regeneration. Whole-mount in situ hybridizations (WMISH) for the genes whose name is indicated are shown. In all panels except E, anterior is up and the regeneration stage for each picture is indicated. In all panels except E, red dotted lines indicate the amputation plane. Dorsal (D) and posterior (P) views are shown. Light blue arrowheads = ectodermal growth zone, light blue arrows = ectoderm of developing segment, red arrows = groups of internal cells in the segment adjacent to the amputation plane.

Additional file 5: Sequences used for phylogenetic analyses, multiple alignments, and phylogenetic trees. This file contains all protein sequences and multiple alignments used for phylogenetic analyses in fasta format, as well as the obtained phylogenetic trees in Newick format.

Additional file 3: Figure S2. Additional CpG o/e ratio calculations. Histograms of CpG o/e ratio for several species (whose name is indicated on top) for which this ratio has not been previously calculated. In each histogram, the red line indicates the estimated density, the vertical blue bar shows the estimated mean value, and the shaded blue bar represents bootstrap confidence intervals of 95%. PM = probability mass. Clusters are those defined in Aliaga et al. [48]. The color code for metazoan groups is indicated and is as in Fig. 2.

Additional file 8: Figure S6. A schematic representation of P. dumerilii posterior regeneration. On the top row is depicted a 3-4-month-old juvenile worm (with 30 to 40 segments). Its posterior part (5 segments, growth zone, and pygidum) is eliminated by amputation (the red dotted lines indicate the amputation plane). Regeneration occurs at the posterior extremity of the anterior body region. The region delineated by the dotted black lines corresponds to the part of the regenerating worms that is shown in WMISH pictures (Figs. 5 and S7). It comprises the posterior part of the posteriormost differentiated segment (S), in which parapodia (black arrows) associated with glands (asterisks) can be seen, plus the regenerated region (region indicated by the two-headed arrow; red dotted lines show the position of the amputation plane). Anterior is up and posterior down. The regenerated region increases in size as regeneration proceeds, from stage 1 to 5 of the process [37]. At stage 1 (reached 1 day post-amputation, 1dpa), wound healing is achieved. A small blastema is formed at stage 2 (2dpa) and molecular analyses indicate that the growth zone is already regenerated at this stage. A regenerated anus can be observed at this stage. At stage 3 (3dpa), the regenerated region has increased in size and small anal cirri are present. The use of molecular markers showed that the growth zone has already produced at least one segment and tissue differentiation has started in the pygidium. The large blastema found at stage 4 (4dpa) contains a differentiated pygidium with long anal cirri. Tissue differentiation also starts in the two or more segments that have been produced by the growth zone. Segments are however still not morphologically visible. Stage 5 (5dpa) corresponds to the end of regeneration. At this stage, a fully differentiated pygidium is found, and segmentation of the regenerated region becomes visible at the morphological level (presence of visible parapodial primordia for example). From this stage onwards, post-regenerative posterior growth starts and a growing number of morphologically well visible segments with parapodia and segmental grooves between adjacent segments is observed.

Additional file 6: Figure S4. Phylogenetic trees of 5mC and NuRD toolkit proteins. Maximum likelihood (ML) trees constructed with PhyML are shown. Statistical supports (aLRT values) for all nodes are indicated with a color code provided in the inset. Terminal branches are colored using the shown color code also used in Fig. 2. P. dumerilii sequences are in bold and indicated by arrows. (A) DNMT proteins. The three subfamilies DNMT1, 2 and 3 are indicated. In the DNMT3 subfamily, the vertebrate-specific groups DNMT3A, B and -like are also shown. We used a distantly related sequence (Dnmt5) from Acanthamoeba castellanii (Acas, an amoeba) as outgroup to root the phylogenetic tree. We also retrieved Dnmt sequences from two choanoflagellates species, Monosiga brevicollis (Mbre) and Salpingoeca rosetta (Sros), which all belong to the DNMT2 subfamily. (B) TET proteins. The three vertebrate-specific groups TET1, 2 and 3 are indicated. We used midpoint rooting for this tree as we were unable to find Tet genes in choanoflagellates or another suitable outgroup. (C) TDG. The phylogenetic tree is rooted using choanoflagellate sequences as outgroup. (D) UHRF. The two vertebrate-specific groups UHRF1 and 2 are indicated. We used midpoint rooting for this tree as we were unable to find Uhrf genes in choanoflagellates or another suitable outgroup. (E) MBD proteins. The two subfamilies MBD1/2/3 and MBD4 subfamilies are indicated. Vertebrate-specific groups are also shown. While not found in S. rosetta and M. brevicollis, a single Mbd gene was found in three other choanoflagellates (Helgoeca nana (Hnan), Salpingoeca urceolata (Surc), and Acanthoeca spectabilis (Aspe)) for which extensive transcriptomic data have been produced. Two of these choanoflagellate Mbd sequences (Hnan and Aspe) form a monophyletic group that was used as outgroup to root the tree, while the third one (Surc) clusters with MBD 4 proteins. (F) CHD proteins. The three subfamilies CHD1/2, CHD3/4/5 and CHD6/7/8/9 are indicated. Vertebrate-specific groups are also shown. We used the distantly related SNF2L1 sequence from Human as outgroup to root the phylogenetic tree. (G) Class I HDAC proteins. The three subfamilies HDAC1/2, HDAC3 and HDAC8 are indicated. Vertebrate-specific HDAC1 and 2 groups are also shown. Class II and IV HDACs from Human were used as outgroup to root the phylogenetic tree. We also retrieved HDACs from choanoflagellates, three of which clustering with the outgroup, two with HDAC3 proteins, and two with HDAC1/2 proteins. (H) RBBP4/7 proteins. Vertebrate-specific RBBP4 and 7 groups are shown. The phylogenetic tree is rooted using choanoflagellate sequences as outgroup. (I) MTA1/2/3 proteins. Vertebrate-specific MTA1, 2 and 3 groups are shown. (J) GATAD2 proteins. Vertebrate-specific GATAD2-alpha and GATAD2-beta groups are shown. For (I) and (J), midpoint rooting has been used in the absence of choanoflagellate members of these two gene families and the absence of other suitable outgroups.

International audience; Research question Non-genetic heritability of susceptibility to chronic diseases is often different between male and female. The environmental factors to which an individual is exposed can leave an epigenetic footprint that dictates the coordinate expression of genes. A crucial period is the early development, when the epigenome is reprogrammed and therefore particularly sensitive to the environment. The sexual dimorphism results from the chromosomal sex (XX/XY) before gonad differentiation and a later-on complex mix of hormones and X/Y-linked genes regulating autosomal genes. The early exposure interacts with this subtle superposition, leading to specific reactions, adaptation and outcomes for men and women. Methods With several mouse model of maternal high-fat diet exposure, we demonstrate a striking sexual dimorphism of programming trajectories in different tissues (particularly placenta and liver) in response to the same environmental challenge. Results and Conclusion Our findings provide proof-of-concept that the epigenetic marks and modifiers may be part of the sex-specific causal variables. Their study represents a novel approach to identify sex-specific mechanisms in the origins of health and disease. The discovery of such factors should help physicians and patients anticipate disease susceptibility and may help the development of different preventative and treatment strategies precisely adapted to males and females.

National audience; The recent and rapid worldwide increase in non communicable diseases challenges the assumption that genetic factors are the primary contributors to such diseases. A new dimension, that of the "developmental origins of health and disease" (DOHaD), is at stake and therefore requires a paradigm shift. Maternal obesity and malnutrition predispose the offspring to develop metabolic syndrome, a vicious cycle leading to transmission to subsequent generation(s), with differences in response and susceptibility according to the sex of the individual. Placenta is a programming agent of adult health and diseases. Adaptation in placental phenotype in response to maternal diet and body composition alter fetal nutrient provision. This implies important epigenetic changes that are however still poorly documented in DOHaD studies, particularly concerning overnutrition. Our objective was to investigate the effects of a high fat diet (HFD) on mouse placental development. We used transcriptomic and epigenetic techniques and showed for the first lime that not only the gene sets but aIso the biological functions affected by the HFD differed markedly between the two sexes. Moreover, the expression of genes coding for epigenetic machinery enzymes, as well as global DNA methylation, were highly dynamic during the fetal period (3 stages), clearly differed between the 2 layers of placenta (labyrinth and functional zone) and showed conspicuous ontogenetic sexual dimorphism for epigenetic modifiers. These findings demonstrate a striking sexual dimorphism of programming trajectories in response to the same environmental challenge. Explaining the sex-specific causal variables and how males versus females respond and adapt to environmental perturbations should help physicians and patients anticipate disease susceptibility. Thus, explorations of changes in sexual dimorphism are potentially useful to identify sex-specific disease mechanisms for use in the development of different, more appropriate prevention and treatment strategies for males and females.

The recent and rapid worldwide increase in noncommunicable diseases challenges the assumption that genetic factors are the primary contributors tc such diseases. A new dimension, thalof the "developmental origins of health and disease" (DOHaD), is at stake and therefore requires a paradigm shift. Maternai obesity and malnutrition p1*edispose the offspring to develop metabolic syndrome, a vicious cycle leading to transmission to subsequent generation(s), with differences in response and susceptibility according to the sex of the individual. Placenta is a programming agent of aduheah and diseases. Adaptation in placenta! phenotype in response tc maternai diet and body composition alter fetal nutrient provision. This implies important epigenetic changes that are however still poorly documented in DOHaD studies, particularly concerning overnutrition. 0ur objective was tc investigate the effects of a high fat diet (HFD) on meuse placenta! development. We used transcriptomic and epigenetic techniques and showed for the first time that not onlythe gene sets but aIso the biological functions affected by the HFD differed markedly between the two sexes. Moreove1*, the expression of the epigenetic machinery enzymes, as weil as global DNA methylabon, were highly dynamic during the fetal period (3 stages), clearly differed between the 2 layers of placenta (labyrinth and junctional zone) and showed conspicuous sexual dimorphism. Thus, these findings demonstrate a striking sexual dimorphism of programming trajectories in response to the same environmental challenge. Explaining the sex-specific causal variables and how males versus females respond and adapt to environmental perturbations should help physiciens and patients an1icipate disease susceptibility.

There is mounting evidence that placenta can be considered as a programming agent of adult health and diseases. Placental weight and shape at term are correlated with the development of metabolic diseases in adulthood in humans. Maternal obesity and malnutrition predispose the offspring to develop metabolic syndrome, a vicious cycle leading to transmission to subsequent generation(s), with differences in response and susceptibility according to the sex of the individual. Adaptations in placental phenotype in response to maternal diet and body composition alter fetal nutrient provision. This implies important epigenetic changes. However the epigenetics of placental development in DOHaD studies is still poorly documented, particularly concerning overnutrition. We used histology, microarrays analysis and epigenetic techniques to investigate the effects of a high fat diet (HFD) on mouse development. We showed for the first time that not only the gene sets but also their biological functions affected by the HFD differed markedly between the two sexes. Remarkably, genes of the epigenetic machinery as well as global DNA methylation levels also showed sexual dimorphism. Imprinted genes expression was altered, with locus-specific changes in DNA methylation. Thus, these findings demonstrate a striking sexual dimorphism of programming trajectories in response to the same environmental challenge. Explaining the sex-specific causal variables and how males versus females respond and adapt to environmental perturbations should help physicians and patients anticipate disease susceptibility.

Undernutrition during gestation is associated with an increased susceptibility to metabolic and cardiovascular diseases. Placenta, as a widely recognized programming agent, contributes to the underlying processes. Alterations in both placental development and activity are well known to constitute programming events for offspring’s physiology and metabolism in adulthood. Growing experimental evidences suggest that epigenetic marks may serve as a memory of exposure to inappropriate environments and thus could be implicated in foetal programming.1 Our aim was to explore whether maternal undernutrition could disturb epigenetic processes in the placenta of intrauterine growth-restricted (IUGR) foetuses. Two experimental IUGR models were used: pregnant Wistar rats were subjected to a 70% food restriction along the gestation (FR30 model);2 or to a 50% food restriction during the last week of gestation (FR50).3 We investigated the global level of four epigenetic marks in full-term placentas. DNA methylation was assessed using LUMA and performed western blot assays for H3K9me3, H3K4me3 and H4K16ac, three important histone marks.4 We did not observe any change in H3K9me3, H3K4me3 and DNA methylation, but a decrease in placental H4K16ac, in both models and in both sexes. High-performance liquid chromatography/high-performance capillary electrophoresis quantified the decrease of H4 monoacetylation: 212% in FR30 males, 218% in FR50 males and 222% in FR30 and FR50 females. As both models were similarly affected, our findings suggest that the last third of gestation may be a critical period for H4K16ac set-up in placenta. This epigenetic mark may constitute a nutritional sensitive target during foetal programming and may be an important link between nutrition and epigenetic programming during foeto-placental development.

National audience

absent

Ouvrage édité à l'occasion de l'exposition présentée par la Ville de Nancy au Musée lorrain du 7 mai au 5 septembre 2005 sous le titre "Sous les pavés, la place ; archéologie et histoire de la place Stanislas". Numéro hors série de La Gazette Lorraine, juin 2005