TET-mediated hydroxymethylcytosine at the Pparγ locus is required for initiation of adipogenic differentiation.

BACKGROUND/OBJECTIVES: Adipose tissue is one of the main organs regulating energy homeostasis via energy storage as well as endocrine function. The adipocyte cell number is largely determined by adipogenesis. While the molecular mechanism of adipogenesis has been extensively studied, its role in dynamic DNA methylation plasticity remains unclear. Recently, it has been shown that Tet methylcytosine dioxygenase (TET) is catalytically capable of oxidizing DNA 5-methylcytosine (5-mC) to 5-hydroxymethylcytosine (5-hmC) toward a complete removal of the methylated cytosine. We investigate whether expression of the Tet genes and production of hydroxymethylcytosine are required for preadipocyte differentiation. SUBJECTS/METHODS: Murine 3T3-L1 preadipocytes were used to evaluate the role of Tet1 and Tet2 genes during adipogenesis. Changes in adipogenic ability and in epigenetic status were analyzed, with and without interfering Tet1 and Tet2 expression using small interfering RNA (siRNA). The adipogenesis was evaluated by Oil-Red-O staining and induced expression of adipogenic genes using quantitative polymerase chain reaction (qPCR). Levels of 5-hmC and 5-mC were measured by MassARRAY, immunoprecipitation and GC mass spectrometry at specific loci as well as globally. RESULTS: Both Tet1 and Tet2 genes were upregulated in a time-dependent manner, accompanied by increased expression of hallmark adipogenic genes such as Pparγ and Fabp4 (P<0.05). The TET upregulation led to reduced DNA methylation and elevated hydroxymethylcytosine, both globally and specifically at the Pparγ locus (P<0.05 and P<0.01, respectively). Knockdown of Tet1 and Tet2 blocked adipogenesis (P<0.01) by repression of Pparγ expression (P<0.05). In particular, Tet2 knockdown repressed conversion of 5-mC to 5-hmC at the Pparγ locus (P<0.01). Moreover, vitamin C treatment enhanced adipogenesis (P<0.05), while fumarate treatment inhibited it (P<0.01) by modulating TET activities. CONCLUSIONS: TET proteins, particularly TET2, were required for adipogenesis by modulating DNA methylation at the Pparγ locus, subsequently by inducing Pparγ gene expression.

Early epigenetic downregulation of WNK2 kinase during pancreatic ductal adenocarcinoma development.

Pancreatic ductal adenocarcinoma (PDAC) is usually incurable. Contrary to genetic mechanisms involved in PDAC pathogenesis, epigenetic alterations are ill defined. Here, we determine the contribution of epigenetically silenced genes to the development of PDAC. We analyzed enriched, highly methylated DNAs from PDACs, chronic pancreatitis (CP) and normal tissues using CpG island microarrays and identified WNK2 as a prominent candidate tumor suppressor gene being downregulated early in PDAC development. WNK2 was further investigated in tissue microarrays, methylation analysis of early pancreatic intraepithelial neoplasia (PanIN), mouse models for PDAC and pancreatitis, re-expression studies after demethylation, and cell growth assays using WNK2 overexpression. Demethylation assays confirmed the link between methylation and expression. WNK2 hypermethylation was higher in tumor than in surrounding inflamed tissues and was observed in PanIN lesions as well as in a PDAC mouse model. WNK2 mRNA and protein expressions were lower in PDAC and CP compared with normal tissues both in patients and mouse models. Overexpression of WNK2 led to reduced cell growth, and WNK2 expression in tissues correlated negatively with pERK1/2 expression, a downstream target of WNK2 responsible for cell proliferation. Downregulation of WNK2 by promoter hypermethylation occurs early in PDAC pathogenesis and may support tumor cell growth via the ERK-MAPK pathway.

Source and component genes of a 6-200 Mb gene cluster in the house mouse.

We identified and analyzed the genes Sp100, Csprs, and Ifi75 in two members of the genus Mus, M. musculus and M. caroli. Sp100 is a nuclear dot gene; Csprs and Ifi75 are novel genes encoding a putative G-protein coupled receptor (GPCR) and a putative transcriptional coactivator, respectively. A fourth gene, Sp100-rs, occurs in M. musculus, but not in M. caroli. Sp100-rs is a chimeric gene which arose by fusion of Sp100 and Csprs copies. Sp100-rs and Ifi75 are components of a repeat cluster that extends over 6-200 Mb of the M. musculus genome. The Sp100-rs fusion gene arose only 1-2 million years ago and has become fixed and amplified in M. musculus. Although the gene is transcribed, it appears to have no function. The repeat cluster may have become fixed in the species as a 'hitchhiker' in a 'selective sweep'.

Modular bacterial artificial chromosome vectors for transfer of large inserts into mammalian cells.

To facilitate the use of large-insert bacterial clones for functional analysis, we have constructed new bacterial artificial chromosome vectors, pPAC4 and pBACe4. These vectors contain two genetic elements that enable stable maintenance of the clones in mammalian cells: (1) The Epstein-Barr virus replicon, oriP, is included to ensure stable episomal propagation of the large insert clones upon transfection into mammalian cells. (2) The blasticidin deaminase gene is placed in a eukaryotic expression cassette to enable selection for the desired mammalian clones by using the nucleoside antibiotic blasticidin. Sequences important to select for loxP-specific genome targeting in mammalian chromosomes are also present. In addition, we demonstrate that the attTn7 sequence present on the vectors permits specific addition of selected features to the library clones. Unique sites have also been included in the vector to enable linearization of the large-insert clones, e. g., for optical mapping studies. The pPAC4 vector has been used to generate libraries from the human, mouse, and rat genomes. We believe that clones from these libraries would serve as an important reagent in functional experiments, including the identification or validation of candidate disease genes, by transferring a particular clone containing the relevant wildtype gene into mutant cells or transgenic or knock-out animals.

Transcription and proper splicing of a mammalian gene in yeast.

The house mouse strain C57BL/6 harbours 64 copies of the multicopy gene Sp100-rs. Three of these are contained in the yeast artificial chromosome (YAC) clone yMm75. Four Sp100-rs transcripts of 3.0, 2.6, 1.6 and 1.3kb were detected by Northern hybridization in the yMm75-harbouring line of Saccharomyces cerevisiae. Additional and less abundant transcripts were detected by RT-PCR. With one exception, the YAC-derived Sp100-rs transcripts were a subset of those found in the C57BL/6 mouse. This indicates transcription and proper splicing of murine pre-mRNAs in yeast. Analysis of the splice sites shows that the yeast splicing machinery accepts splice sites that deviate from the standard yeast consensus sequences. It may be feasible, therefore, at least in a fair proportion of cases, to exploit the mammalian mRNAs present in transgenic yeast for gene recognition of YAC-inserts.

A modular, positive selection bacterial artificial chromosome vector with multiple cloning sites.

To construct large-insert libraries for the sequencing, mapping, and functional studies of complex genomes, we have constructed a new modular bacterial artificial chromosome (BAC) vector, pBACe3.6 (GenBank Accession No. U80929). This vector contains multiple cloning sites located within the sacB gene, allowing positive selection for recombinant clones on sucrose-containing medium. A recognition site for the PI-SceI nuclease has also been included, which permits linearization of recombinant DNA irrespective of the characteristics of the insert sequences. An attTn7 sequence present in pBACe3.6 permits retrofitting of BAC clones by Tn7-mediated insertion of desirable sequence elements into the vector portion. The ability to retrofit BAC clones will be useful for functional analysis of genes carried on the cloned inserts. The pBACe3.6 vector has been used for the construction of many genomic libraries currently serving as resources for large-scale mapping and sequencing.

Pericentric satellite DNA and molecular phylogeny in Acomys (Rodentia).

Satellite DNAs (stDNAs) of four Acomys species (spiny-mice), A. cahirinus, A. cineraceus, A. dimidiatus and A. russatus, belong to closely related sequence families. Monomer sizes range from 338 to 364 bp. Between-species sequence identity was from 81.0% to 97.2%. The molecular phylogeny of the sequences helps to clarify the taxonomy of this 'difficult' group. The A. dimidiatus genome contains about 60000 repeats. According to the restriction patterns, repeats are arranged in tandem. The stDNA maps to the centromeric heterochromatin of most autosomes, both acrocentric and metacentric, but appears to be absent in the centromeric region of Y chromosomes. A well-conserved centromere protein B (CENP-B) box is present in the stDNA of A. russatus while it is degenerated in the other species.

Origin of the chromosome 1 HSR of the house mouse detected by CGH.

The polymorphic Sp100-rs repeat cluster in chromosome band 1D of the house mouse, Mus musculus, makes up as much as 0.1-5% of the haploid genome. 'High-copy' versions of this long-range repeat cluster are cytogenetically apparent as DAPI-negative chromomycin-A3-positive homogeneously staining regions (HSRs). The cluster is a relatively recent acquisition in the genus Mus; the related species M. caroli possesses neither the Sp100-rs cluster nor even the Sp100-rs gene. Except for chromosomes with high-copy clusters, no major rearrangements are visible in chromosomes 1 from M. musculus and M. caroli: they have the same order of G-bands, DAPI-bands and chromomycin A3-bands. Comparative genomic hybridization (CGH) visualizes the cluster in M. musculus and detects a single region of sequence homology to the cluster in M. caroli chromosome band 1D. This indicates that the M. musculus cluster has evolved in situ from sequences originally present in the same chromosome band.

Restoration of the Mendelian transmission ratio by a deletion in the mouse chromosome 1 HSR.

The house mouse, Mus musculus, harbours a variable cluster of long-range repeats in chromosome 1. As shown in previous studies, some high-copy clusters such as the MUT cluster are cytogenetically apparent as a homogeneously staining region (HSR) and are associated with a distortion of the Mendelian recovery ratio when transmitted by heterozygous females. The effect is caused by a decreased viability of +/+ embryos. It is compensated by maternal or paternal MUT. In this study, a deletion derivative of MUT, MUTdel, shows normal transmission ratios and no compensating capability. In this respect, MUTdel behaves like a wild-type cluster. Hence, both properties--transmission ratio distortion and compensating capability--map to the deleted region. The deletion comprises three-quarters of the MUT HSR and does not extend to the nearest markers adjacent to the HSR.

Evolution by fusion and amplification: the murine Sp100-rs gene cluster.

Sp100 is a single-copy gene in the human and the mouse. A related gene, Sp100-rs, occurs in multiple copies and forms a conspicuous cluster in the mouse chromosome 1. Murine Sp100 and Sp100-rs are homologous from the promoter up to a position in intron 3, but they differ 3' of that position. In the genus Mus, Sp100-rs is present in one phylogenetic branch, represented by the house mouse, M. musculus, but probably does not exist in another branch, represented by M. caroli. Thus, Sp100-rs arose relatively late in the evolution of the genus Mus, whereas Sp100 existed in the common ancestor of the human and the mouse. The Sp100-rs gene cluster probably evolved by gene fusion followed by amplification and diversification.

Structure and expression of the murine Sp100 nuclear dot gene.

The human SP100 gene encodes an autoantigen that colocalizes with two other proteins, PML and NDP52, in distinct nuclear domains, called "nuclear dots" (NDs). NDs do not overlap with other known subnuclear structures, and their function is still unknown. Patients suffering from the autoimmune disease primary biliary cirrhosis often produce antibodies against the SP100 protein. The present study describes the structure and expression of the murine Sp100 gene. In the species Mus caroli, Sp100 consists of 17 exons that are distributed over a range of 52 kb. The human and murine Sp100 promoters are very similar, and both harbor an interferon-stimulated response element. Like its human counterpart, the murine Sp100 gene is responsive to interferon treatment. The house mouse, Mus musculus, harbors the Sp100 gene and a second gene with homology to Sp100, the multicopy Sp100-rs gene. However, in contrast to the genuine mouse homolog, Sp100-rs shares only segmental homology with the human Sp100 gene. Replacement of the murine Sp100 gene by a defective copy is now feasible and should shed light on its function in an animal model.

Distortion of Mendelian recovery ratio for a mouse HSR is caused by maternal and zygotic effects.

An HSR in chromosome 1 which is found in many feral populations of Mus musculus domesticus was shown in previous studies to consist of a high-copy long-range repeat cluster. One such cluster, MUT, showed distorted transmission ratios when introduced by female parents. MUT/+ offspring were preferentially recovered at the expense of +/+ embryos in the progeny of male MUT/+ x female +/+ but were found at the expected 1:1 ratio in reciprocal crosses. Preferential recovery of maternal MUT was due to lethality of postimplantation +/+ embryos. There was no distortion of the recovery ratio in MUT/+ x MUT/MUT progeny: maternal MUT and + clusters were present among live implants at a 1:1 ratio. Maternal and zygotic effects therefore contribute to the phenomenon. The mechanism of their interaction is unknown.

Absence of correlation between Sry polymorphisms and XY sex reversal caused by the M. m. domesticus Y chromosome.

Mus musculus domesticus Y chromosomes (YDOM Chrs) vary in their ability to induce testes in the strain C57BL/6J. In severe cases, XY females develop (XYDOM sex reversal). To identify the molecular basis for the sex reversal, a 2.7-kb region of Sry, the testis-determining gene, was sequenced from YDOM Chrs linked to normal testis determination, transient sex reversal, and severe sex reversal. Four mutations were identified. However, no correlation exists between these mutations and severity of XYDOM sex reversal. RT-PCR identified Sry transcripts in XYDOM sex-reversed fetal gonads at 11 d.p.c., the age when Sry is hypothesized to function. In addition, no correlation exists between XYDOM sex reversal and copy numbers of pSx1, a Y-repetitive sequence whose deletion is linked to XY sex reversal. We conclude that SRY protein variants, blockade of Sry transcription, and deletion of pSx1 sequences are not the underlying causes of XYDOM sex reversal.

Copy numbers of a clustered long-range repeat determine C-band staining.

A cluster of long-range repeats (LRRs) with a repeat size of roughly 100 kb is part of band D of chromosome 1 of the house mouse, Mus musculus. The cluster is cytogenetically polymorphic: it is either C-band negative or C-band positive. Our results show that the differential staining behavior depends on the LRR copy number and not on differences in DNA composition. There is a threshold between 105 and 175 LRR copies per haploid genome; clusters with lower copy numbers stain C-band negative, whereas those with higher copy numbers are C-band positive. Above this threshold, the size of the C-band is linearly correlated with the LRR copy number. The results imply that sequences capable of forming heterochromatin may be dispersed throughout the genome but are not recognized as such by cytogenetic techniques, unless they reach the threshold amount and concentration.

A transcript family from a long-range repeat cluster of the house mouse.

A family of closely related genes is a component of the polymorphic long-range repeat cluster D1Lub1 of the house mouse. Members of the gene family have diverged from one another by rearrangements and point mutations. D1Lub1 cluster have low (approximately 50) or high (> or = 500) copy numbers. In mice with high-copy clusters five or six poly(A)+ RNAs are found, while in mice with low-copy clusters only a single member of the RNA family is detected. The RNA family is synthesized in a tissue-independent manner. Each member of the RNA family is defined by a set of DNA probes. Cross hybridization with the probes reveals common 5' regions and variable remaining parts. The RNA variants are probably transcribed from different gene copies.

A member of the mouse LRR transcript family with homology to the human Sp100 gene.

A previously isolated cDNA sequence with homology to the long-range repeat (LRR) cluster in chromosome 1 of the house mouse, Mus musculus, was identified as derived from a 1.3 kb polyadenylated RNA. This transcript belongs to a family of polyadenylated RNAs which are synthesized from a multicopy gene included in the LRR copies. The representation of the 1.3 kb transcript in genomic DNA was studied in lambda and cosmid clones from the LRR cluster. Two different types of LRRs were detected with respect to the arrangement of coding regions. In the type-1 arrangement, the sequence is split into five exons, and in the type-2 arrangement, into six exons. The respective exons with their flanking regions were sequenced. The analysis of splice signals revealed that LRR copies with a type-1 arrangement are presumably the source of the 1.3 kb transcript. The 1.3 kb transcript has sequence homology to a human gene encoding Sp100, a nuclear antigen recognized by autoantibodies from patients suffering from some autoimmune diseases including primary biliary cirrhosis. Mouse exons II and III exhibit 71% homology at the nucleotide level and 56% homology at the amino acid level to the human Sp100 cDNA. We mapped the human Sp100 gene to chromosome 2. This location corroborates the assumption that the human Sp100 gene and the mouse LRR gene are homologous, as the human chromosome 2 contains the segment which is homologous to the mouse LRR region.

Different modes of hypervariability in (GATA)n simple sequence repeat loci.

Only a few prominent simple sequence repeat (SSR) loci of the type (GATA)n are found in the genome of the mealmoth Ephestia kuehniella Zeller. Therefore this moth was chosen as a model organism for the genetic and molecular analysis of hypervariability of SSR loci. We characterized alleles of (GATA)n loci in different Ephestia strains by cloning and genomic restriction mapping. Some variants appeared to be mere variable number of tandem repeat (VNTR) alleles, others showed considerable changes in the sequence neighbourhood of the GATA repeats. These may be produced by major rearrangements or by transposition of the (GATA)n block together with flanking sequences into a different sequence environment.

The recA-recBCD dependent recombination pathways of Serratia marcescens and Proteus mirabilis in Escherichia coli: functions of hybrid enzymes and hybrid pathways.

The physical maps of cloned recBCD gene regions of Serratia marcescens and Proteus mirabilis were correlated to genes located in this region. The genes thyA, recC, recB, recD and argA were organized as in Escherichia coli. The 3 rec genes code for the 3 different subunits of the RecBCD enzyme and produced enzymes promoting recombination and repair of UV damage in E coli. The recBCD-dependent stimulation of recombination at specific nucleotide sequences called Chi (Chi-activation) was determined in lambda red-gam-crosses. Chi-activation by the different RecBCD enzymes decreased in the order E coli greater than S marcescens greater than P mirabilis. In E coli cloned subunits genes from S marcescens and P mirabilis led to the formation of functional hybrid enzymes consisting of subunits from 2 or even 3 species. The origin of the RecC subunit present in the hybrid enzymes affected the degree of Chi-activation. Further, changes in Chi-activation occurred when the RecD subunit in the enzyme from E coli was replaced by RecD proteins from S marcescens or P mirabilis. This suggested that the RecD subunit determines not only whether or not Chi-activation is possible but also to which extent it occurs. Finally we have reconstituted recombination pathways of S marcescens and P mirabilis by combining the cloned recA and recBCD genes from these species in E coli deleted for recA and recBCD. Both pathways can efficiently promote recombination and repair. Studies are summarized which showed that levels of repair and recombination promoted by the recA-recBCD genes are mostly higher when the recA and recBCD genes came from the same species than from 2 different species (hybrid RecBCD recombination pathway). The data are interpreted to provide evidence that in vivo the RecA protein co-operates with the RecBCD enzyme in recombination and repair of UV damage.

Functional analyses of Proteus mirabilis wild-type and mutant RecBCD enzymes in Escherichia coli reveal a new mutant phenotype.

Cloned into Escherichia coli the recB, recC and recD genes of Proteus mirabilis produce a recBCD enzyme (exoV) functional in recombination and DNA repair. The direction of transcription of recB, recC and recD, the sizes of the enzyme subunits, and their composition in the active enzyme are similar to that observed for the E. coli enzyme. In lambda crosses, the P. mirabilis enzyme has only about 40% of the Chi activity of the E. coli enzyme. The recBCD genes were also cloned from an exoV mutant of P. mirabilis which is u.v.-sensitive and partly deficient in exoV. The defect was attributed to the recB gene by complementation studies. In a recBCD deletion strain of E. coli, the enzyme from the mutant produced 40% of conjugational recombinants and had retained about 25% of Chi activity. However, it did not restore normal DNA repair, cell viability or recombination in lambda crosses and P1 transduction. The new mutant phenotype is discussed in the light of the assumption that prokaryotic recBCD enzymes can promote recombination in a Chi-dependent and a Chi-independent manner.