Partitioning of [methyl-3H]methionine to methylated products and protein is altered during high methyl demand conditions in young Yucatan miniature pigs.

Methionine is the main source of methyl groups that are partitioned to synthesize various methylated products including creatine, phosphatidylcholine (PC), and methylated DNA. Whether increased methylation of 1 product can divert methionine from protein synthesis or other methylation products was the aim of this experiment. We used an excess of guanidinoacetate (GAA) to synthesize creatine to create a higher demand for available methyl groups in normal-weight (NW) (n = 10) and intrauterine growth-restricted (IUGR) (n = 10) piglets. Anesthetized piglets (15-18 d old) were intraportally infused with either GAA or saline for 2 h. A bolus of l-[methyl-(3)H]methionine was intraportally infused at 1 h, and hepatic metabolites were analyzed for methyl-(3)H incorporation 1 h later. Overall, 50-75% of label was recovered in creatine and PC with negligible amounts in DNA. In both NW and IUGR piglets, excess GAA led to an ≈ 80-120% increase in methyl incorporation into creatine (P < 0.05) with a concomitant decrease by ≈ 75-85% in methyl incorporation into PC (P < 0.05) as well as a 40% decrease in methyl incorporation into protein (P < 0.05), suggesting methyl groups were limited for PC synthesis and that methionine was diverted from protein synthesis. Compared with NW piglets, IUGR piglets had lower methyl incorporation into PC (P < 0.05), but not DNA or protein, suggesting IUGR affects methyl metabolism and could potentially impact lipid metabolism. The partitioning of methionine is sensitive to methyl supply in neonates, which has implications in infant diet composition and growth.

Intrauterine growth restriction leads to changes in sulfur amino acid metabolism, but not global DNA methylation, in Yucatan miniature piglets.

Intrauterine growth restriction (IUGR), in both animals and humans, has been linked to metabolic syndrome later in life. There has been recent evidence that perturbations in sulfur amino acid metabolism may be involved in this early programming phenomenon. Methionine is the precursor for cellular methylation reactions and for the synthesis of cysteine. It has been suggested that the mechanism behind the "fetal origins" of adult diseases may be epigenetic, involving DNA methylation. Because we have recently demonstrated the fetal origins phenomenon in Yucatan miniature swine, we hypothesized that sulfur amino acid metabolism is altered in IUGR piglets. In this study, metabolites and the activities of sulfur amino acid cycle enzymes were analyzed in liver samples of 3- to 5-day-old runt (IUGR: 0.85±0.13 kg) and large (1.36±0.21 kg) Yucatan miniature pig littermates (n=6 pairs). The IUGR piglets had significantly lower specific and total activities of betaine-homocysteine methyltransferase (BHMT) and cystathionine γ-lyase (CGL) than larger littermates (P<.05). Expression of CGL (but not BHMT) mRNA was also lower in IUGR piglets (P<.05). This low CGL reduced cysteine and taurine concentrations in IUGR pigs and led to an accumulation of hepatic cystathionine, with lower homocysteine concentrations. Methylation index and liver global DNA methylation were unaltered. Reduced prenatal growth in Yucatan miniature piglets impairs their remethylation capacity as well as their ability to remove cystathionine and synthesize cysteine and taurine, which could have important implications on long-term health outcomes of IUGR neonates.

Expression of the dnmt3 genes in zebrafish development: similarity to Dnmt3a and Dnmt3b.

The zebrafish differs from mammals in that they have six dnmt3 genes as opposed to the two that can produce a catalytically active protein in mammals. Zebrafish also do not show evidence of genomic imprinting and lack the Dnmt3l gene necessary to that process in mammals. As such, they offer a unique opportunity to compare the two genetic situations in order to define the roles of the multiple genes in developmental gene methylation. To this end, we have analyzed the developmental expression of the six dnmt3 genes in zebrafish and find that they fall into two distinct patterns. The expression patterns of the dnmt6 and dnmt8 genes, which more closely resemble the mammalian Dnmt3a gene in sequence, also show an expression pattern that is more similar to the expression of Dnmt3a rather than Dnmt3b. Conversely, the other four dnmt3 genes in zebrafish (dnmt3, dnmt4, dnmt5, and dnmt7) show an expression pattern that is more similar to Dnmt3b. The dnmt6 and dnmt8 genes are also expressed in the adult zebrafish and in the brain in particular. In situ expression analyses show that the dnmt6 and/or dnmt8 genes also show tissue-specific differences in expression with those two genes being more ubiquitously expressed in the developing zebrafish than the other dnmt3 genes. Although differences in dnmt3 function may exist between mammals and fish, our results showing similar expression patterns between the genes in fish and mammals suggest that the six dnmt3 genes in the zebrafish may be analogous to the two Dnmt3 genes in mammals.

Immunological detection of changes in genomic DNA methylation during early zebrafish development.

DNA methylation reprogramming, the erasure of DNA methylation patterns shortly after fertilization and their reestablishment during subsequent early development, is essential for proper mammalian embryogenesis. In contrast, the importance of this process in the development of non-mammalian vertebrates such as fish is less clear. Indeed, whether or not any widespread changes in DNA methylation occur at all during cleavage and blastula stages of fish in a fashion similar to that shown in mammals has remained controversial. Here we have addressed this issue by applying the techniques of Southwestern immunoblotting and immunohistochemistry with an anti-5-methylcytosine antibody to the examination of DNA methylation in early zebrafish embryos. These techniques have recently been utilized to demonstrate that development-specific changes in genomic DNA methylation also occur in Drosophila melanogaster and Dictyostelium discoideum, both organisms for which DNA methylation was previously not thought to occur. Our data demonstrate that genome-wide changes in DNA methylation occur during early zebrafish development. Although zebrafish sperm DNA is strongly methylated, the zebrafish genome is not detectably methylated through cleavage and early blastula stages but is heavily remethylated in blastula and early gastrula stages.

Novel splice variants associated with one of the zebrafish dnmt3 genes.

BACKGROUND: DNA methylation and the methyltransferases are known to be important in vertebrate development and this may be particularly true for the Dnmt3 family of enzymes because they are thought to be the de novo methyltransferases. Mammals have three Dnmt3 genes; Dnmt3a, Dnmt3b, and Dnmt3L, two of which encode active enzymes and one of which produces an inactive but necessary cofactor. However, due to multiple promoter use and alternative splicing there are actually a number of dnmt3 isoforms present. Six different dnmt3 genes have recently been identified in zebrafish. RESULTS: We have examined two of the dnmt3 genes in zebrafish that are located in close proximity in the same linkage group and we find that the two genes are more similar to each other than they are to the other zebrafish dnmt3 genes. We have found evidence for the existence of several different splice variants and alternative splice sites associated with one of the two genes and have examined the relative expression of these genes/variants in a number of zebrafish developmental stages and tissues. CONCLUSION: The similarity of the dnmt3-1 and dnmt3-2 genes suggests that they arose due to a relatively recent gene duplication event. The presence of alternative splice and start sites, reminiscent of what is seen with the human DNMT3s, demonstrates strong parallels between the control/function of these genes across vertebrate species. The dynamic expression levels of these genes/variants suggest that they may well play a role in early development and this is particularly true for dnmt3-2-1 and dnmt3-1. dnmt3-2-1 is the predominantly expressed form prior to zygotic gene activation whereas dnmt3-1 predominates post zygotic gene activation suggesting a distinct developmental role for each.

Variations in DNA (cytosine-5)-methyltransferase-1 expression during oogenesis and early development of the zebrafish.

We report the determination of zebrafish DNA (cytosine-5-) methyltransferase ( dnmt-1) temporal and spatial patterns of expression in gonadal tissues and during early development. Only one dnmt-1 message of around 5 kb was observed in all tissues examined and its levels were highest in gonadal tissues. During the course of oogenesis, early oocytes contain significant amounts of dnmt-1 transcript while message abundance declines as oocytes mature. During early embryogenesis message levels remain low until the blastula stage. Methyltransferase enzyme assays reveal that the maternal dnmt-1 message accumulated during oogenesis is translated into protein presumably providing necessary dnmt-1 stockpiles to support early embryonic development prior to zygotic gene activation. Such spatial and temporal regulation of dnmt-1 expression suggests specific functions for the enzyme during oogenesis and early development of zebrafish.