Skip Navigation


Human Molecular Genetics Advance Access originally published online on September 27, 2007
Human Molecular Genetics 2008 17(1):15-26; doi:10.1093/hmg/ddm281
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
17/1/15    most recent
ddm281v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Labialle, S.
Right arrow Articles by Naumova, A. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Labialle, S.
Right arrow Articles by Naumova, A. K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Coordinated diurnal regulation of genes from the Dlk1–Dio3 imprinted domain: implications for regulation of clusters of non-paralogous genes

Stéphane Labialle1, Lanjian Yang2, Xuan Ruan3, Aude Villemain3, Jennifer V. Schmidt6, Arturo Hernandez7, Tim Wiltshire8, Nicolas Cermakian3,4 and Anna K. Naumova1,2,5,*

1 Department of Obstetrics and Gynecology 2 Department of Human Genetics 3 Laboratory of Molecular Chronobiology, Douglas Mental Health University Institute 4 Department of Psychiatry and 5 The Research Institute of the McGill University Health Centre, McGill University, Montreal, QC, Canada 6 Department of Biological Sciences, The University of Illinois at Chicago, IL 60607, USA 7 Department of Medicine, Dartmouth Medical School, Lebanon, NH 03756, USA 8 Genomics Institute of the Novartis Research Foundation, San Diego, CA 92121, USA

* To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, Royal Victoria Hospital, Women's Pavilion, F3.32, 687 Pine Avenue West, Montreal, QC H3A 1A1, Canada. Tel: +1 5149341934 ext. 35906; Fax: +1 5148431662; Email: anna.naumova{at}muhc.mcgill.ca

Received May 28, 2007; Accepted September 21, 2007


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
The functioning of the genome is tightly related to its architecture. Therefore, understanding the relationship between different regulatory mechanisms and the organization of chromosomal domains is essential for understanding genome regulation. The majority of imprinted genes are assembled into clusters, share common regulatory elements, and, hence, represent an attractive model for studies of regulation of clusters of non-paralogous genes. Here, we investigated the relationship between genomic imprinting and diurnal regulation of genes from the imprinted domain of mouse chromosome 12. We compared gene expression patterns in C57BL/6 mice and congenic mice that carry the imprinted region from a Mus musculus molossinus strain MOLF/Ei. In the C57BL/6 mice, a putative enhancer/oscillator regulated the expression of only Mico1/Mico1os, whereas in the congenic mice its influence was spread onto Rtl1as, Dio3 and Dio3os, i.e. the distal part of the imprinted domain, resulting in coordinated diurnal variation in expression of five genes. Using additional congenic strains we determined that in C57BL/6 the effect of the putative enhancer/oscillator was attenuated by a linked dominant trans-acting factor located in the distal portion of chromosome 12. Our data demonstrate that (i) in adult organs, mRNA levels of several imprinted genes vary during the day, (ii) genetic variation may remove constraints on the influence of an enhancer and lead to spreading of its effect onto neighboring genes, thereby generating genotype-dependent expression patterns and (iii) different regulatory mechanisms within the same domain act independently and do not seem to interfere with each other.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Normal operation of the genome is associated with its architecture and maintenance of regulatory hierarchies. With few exceptions, the interactions between different regulatory programs within the same chromosomal domain are largely unknown (reviewed in 1). Genomic imprinting is differential epigenetic modification of the genome during male and female gametogenesis and results in a parent-of-origin dependent monoallelic gene expression in embryos. Imprinting anomalies lead to abnormal embryonic development and even death, establishing the notion that genomic imprinting is indispensable for normal development (reviewed in 2). However, it is not clear if the role of genomic imprinting as a gene dosage controlling mechanism is conserved during post-natal and adult life.

An important element of normal adult physiology is the integrity of biological rhythms. In mammals, circadian rhythms are established soon after birth (3) and ensure the coordinated expression of networks of genes in accordance with environmental changes in lighting (4,5). It was of particular interest for us to determine if diurnal oscillations in RNA levels were compatible with genomic imprinting.

The vast majority of imprinted genes are assembled into clusters and often share common regulatory elements (reviewed in 6,7). Therefore, studying the effect of biological rhythms on genes residing within the same imprinted chromosomal domain provides an opportunity to shed light into the mechanisms that control transcription of physical clusters of non-paralogous genes and determine if changes in transcriptional regulation may affect gene silencing imposed by genomic imprinting.

Multiple lines of evidence indicate that both trans- and cis-acting strain-specific genetic modifiers cause loss of imprinting of certain imprinted genes (e.g. Dlk1, Peg1, etc.) in brains and other organs of interspecific or intersubspecific mouse hybrids (8–11). In principle, one of such mechanisms could be a conflict between transcriptional variation and silencing imposed by genomic imprinting (9). However, it was not clear, if, in general, transcriptional variation could cause loss of imprinting and how different transcriptional mechanisms interact with each other. To clarify the interactions between different regulatory mechanisms within a single imprinted domain, we simplified the system and reduced the effect of trans-acting genetic modifiers of imprinting by generating several congenic strains that contain portions of the distal region of chromosome 12 including the imprinted domain and investigated genotype-dependent regulation of genes located in the imprinted domain of mouse chromosome 12 (Fig. 1). This domain spans about 1 Mb, harbors nine imprinted genes and multiple small non-coding RNAs. Most of these genes are actively expressed in the brain, pituitary and adrenal glands, and, therefore, are likely to share common neuroendocrine tissue-specific enhancers (12). Moreover, co-regulation of genes from the Dlk1–Dio3 imprinted domain has also been documented in sheep skeletal muscle (13).


Figure 1
View larger version (6K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. The Dlk1–Dio3 imprinted cluster on mouse chromosome 12 and adjacent non-imprinted genes. (A) Schematic representation of the genes whose expression was studied. Boxes represent genes and arrowheads represent the direction of transcription. Genes expressed from the maternal allele (MAT) in embryos are positioned above the line, genes expressed from the paternal allele (PAT) in embryos are positioned below the line. (B) Maps of congenic regions in the mouse strains used in the study. MOLF/Ei, NOD/Lt and CAST/Ei-derived genomic regions are represented as open bars. The name of the strain is given on the left. Genetic distances are not in scale.

 
The delta-like 1 homolog (Drosophila) (Dlk1) gene is expressed from the paternal chromosome only. DLK1 is an inhibitor of Notch-signaling (14) that promotes growth, is involved in neuronal differentiation (15), and prevents differentiation of preadipocytes into adipocytes (16). In the brain, Dlk1 mRNA is most abundant in the dopaminergic neurons of substantia nigra and the ventral tegmental area (17). The type 3 iodothyronine deiodinase gene (Dio3) codes for an enzyme that inactivates thyroid hormones and is highly expressed in the placenta and fetal tissues, as well as in developing and adult neuroendocrine tissues (reviewed in 18). Intriguingly, in the embryo, this gene is only partially imprinted, or imprinted in a specific type of cells, as it shows preferential paternal expression but not complete silencing of the maternal allele (19–21). In the adult brain Dio3 is not imprinted (A. Hernandez, unpublished data). The third paternally expressed gene, retrotransposon-like 1 gene (Rtl1) has no known function. All maternally expressed genes, maternal intergenic circadian oscillating 1 (Mico1), Mico1 opposite strand (Mico1os) (Labialle et al., submitted for publication), gene-trap locus 2 (Gtl2) (22), RNA imprinted and accumulated in nucleus (Rian) (23), Rtl1 antisense (Rtl1as) (24) and microRNA containing gene (Mirg) (25) are transcribed as untranslated RNAs that are particularly abundant in the brain and the pituitary gland. It has been suggested that all the maternally expressed genes in this domain might be alternative splice forms of a single precursor transcript (12). Imprinting of this gene cluster depends to a great extent upon two closely located regions, the intergenic differentially methylated region (IG DMR) and the Gtl2 DMR, that are methylated on the paternal and unmethylated on the maternal allele (26–28).

Here, we examined the expression levels of imprinted genes in the brain cortex of adult male mice from several mouse strains (Fig. 1B). We found that genetic regulatory variation and diurnal rhythms modify mRNA abundance of several genes without affecting their imprinted state. Furthermore, we propose a model that accounts for our data, while providing a framework to understand the multiple regulation levels of this genomic domain.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Expression of genes from the Dlk1–Dio3 imprinted domain during the day–night cycle
To investigate the interaction between two different mechanisms of gene regulation, circadian rhythms and genomic imprinting, we determined the mRNA levels of several imprinted and non-imprinted neighboring genes located in the distal part of mouse chromosome 12 in two strains of mice: C57BL/6 and B6.MOLF12A. The B6.MOLF12A congenic strain carries an 8 cM region of chromosome 12 from the Mus musculus molossinus strain MOLF/Ei, on a C57BL/6 genetic background (Fig. 1). The congenic region includes the Dlk1–Dio3 imprinted domain. Use of congenic mice allowed to study parent-of-origin dependent gene expression and to introduce limited genetic variation. Due to the high abundance of these genes in neuroendocrine tissues, we focussed our efforts on the brain. Moreover, as the phase and amplitude of diurnal oscillations of RNA levels of the same gene may be different in different cell types (reviewed in 5), we selected the cerebral cortex as a large, but a relatively homogeneous part of the brain with respect to types of cells. Cerebral cortex samples were collected at 4 h intervals over a 30 h period, from adult male mice entrained to a 12 h light–12 h dark cycle.

Imprinted genes Dlk1, Gtl2 and Rian, and non-imprinted genes YY1 transcription factor (Yy1), and dynein cytoplasmic 1 heavy chain 1 (Dync1h1) are transcribed in the same orientation and their mRNA isoforms are well characterized. A more complex situation arises with overlapping transcripts, such as Rtl1 and Rtl1as, Dio3 and Dio3os, and two novel overlapping maternally expressed transcripts Mico1 and Mico1os that were identified in our laboratory (Labialle et al., submitted for publication). Hence, a detailed characterization of these overlapping transcripts was necessary.

Dio3 and Dio3os
The three Dio3os transcripts that are most abundant in the mouse fetus are expressed biallelically (12), though at least six different Dio3os transcripts have been described in different tissues and developmental stages in the mouse (29). To identify the Dio3os isoform(s) that are expressed in the adult cerebral cortex, we designed several pairs of isoform-specific primers (Supplementary Material, Table S1). The products of isoform-specific PCRs were sequenced. Only one of the previously known Dio3os isoforms (Genbank accession number AY283181) was detected in the adult cortex. We also identified one exonic polymorphism within Dio3os that helped to determine the parental origins of expressed alleles.

Next, we measured Dio3 and Dio3os expression levels at different time points (Zeitgeber time (ZT) 2–22, which corresponds to time after dawn, in hours). Expression of the period homolog 2 (Drosophila) (Per2) gene, a component of the molecular circadian clock, was used as a rhythmic control. We did not find significant variation in Dio3 or Dio3os expression in the C57BL/6 mice (Fig. 2A). In contrast, in B6.MOLF12A mice, the mRNA levels of both genes varied during the day with peaks around ZT2 (Dio3os) and between ZT18 and ZT2 (Dio3) and troughs at ZT14 (Dio3os) and between ZT6 and ZT14 (Dio3). We compared the levels of Dio3 expression in individual mice at ZT2 (peak) and at ZT14 (trough). At ZT2, the level of the Dio3 mRNA in congenic mice was significantly higher than in C57BL/6 mice (t-test, P = 0.002), whereas at trough, the level of the Dio3 mRNA was similar to the one observed in C57BL/6 mice (t-test: P = 0.48).


Figure 2
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Diurnal variation in gene expression levels in the Yy1–Dync1h1 region of mouse chromosome 12. (A) Gene expression levels in the adult brain cortex. For all tested genes, expression levels were normalized to Hprt expression and the level of expression in C57BL/6 mice at ZT2 was set to 1. The X-axis corresponds to the ZT, in hours, which gives the time relative to the dark–light transition. At each time point a mix of RNA from three to six individual mice was tested. The gene Per2, a well-established molecular oscillator, is used as a control. (B) Enzymatic activity of the Dio3 gene product in adult mouse adrenals.

 
To determine if diurnal variation in Dio3 transcription was present in other organs and if it led to changes in protein levels, we determined the enzymatic activity of DIO3 in the adrenal glands of C57BL/6 and B6.MOLF12A mice. In the B6.MOLF12A mice, DIO3 activity was higher during the day than during the early night, while no rhythm was observed in C57BL/6 mice (Fig. 2B). This is consistent with the strain-specific patterns observed for Dio3 mRNA in the brain cortex (Fig. 2A). These results suggest that diurnal oscillations of Dio3 transcripts in the congenic mice are likely to occur in other organs and cause diurnal oscillations in enzymatic activity.

Rtl1as and Rtl1
Rtl1 and Rtl1as fully overlap and are transcribed in opposite directions. Rtl1 is expressed from the paternal, whereas Rtl1as is transcribed from the maternal allele during embryonic development (Fig. 1). To determine the orientation of gene transcription and the predominance of one or the other transcript in the adult cerebral cortex, we used primer-specific reverse transcription (RT) that allows us to distinguish between transcripts expressed from opposite DNA strands. In these experiments we tested Rtl1/Rtl1as expression in the adult cerebral cortex and as a positive control, we used midbrains from 8.5 and 14.5 days post-coitum (dpc) embryos (Supplementary Material, Fig. S1). Rtl1 was barely detectable in the adult brain and midbrains of 14.5 dpc embryos, but abundant in 8.5 dpc embryos. In contrast, expression of Rtl1as was readily detectable at all stages. Hence, in the adult cerebral cortex, we measure only Rtl1as transcripts.

Expression level of Rtl1as in the cortex of adult C57BL/6 mice did not vary with time (Fig. 2A). However, in B6.MOLF12A mice, Rtl1as exhibited a dramatic increase in its expression level with a morning peak at ZT2 and a trough at ZT10 and ZT14. The level of Rtl1as expression at ZT2 was about 6-fold higher than at ZT14 (P = 0.014). The trough level of B6.MOLF12A expression reached at ZT14 was also higher and statistically different from the levels observed at ZT2 (P = 0.034) or ZT14 (P = 0.011) in C57BL/6 mice.

Mico1/Mico1os
We recently characterized novel maternally expressed overlapping transcripts Mico1/Mico1os located distal to the IG DMR and observed circadian oscillations in their mRNA levels, with a peak during the day (ZT2-6) and a trough during the night (ZT18-22), in the brain cortex and in the heart of C57BL/6 mice (Labialle et al., submitted for publication). The Mico1 and Mico1os genes fully overlap and are transcribed in opposite directions. Here, we determined their expression patterns in the B6.MOLF12A mice and did not detect a significant difference compared to C57BL/6 mice (Supplementary Material, Fig. S2). For both strains, expression levels at ZT2 were about 4-fold higher than at ZT18.

Dlk1
Expression levels of Dlk1 in the C57BL/6 and B6.MOLF12A mice varied at different time points (Fig. 2A). However, the observed variation results from inter-individual variance rather than rhythmic oscillations in RNA levels (Supplementary Material, Fig. S3, A). In parallel, in situ hybridization was used to examine the expression of Dlk1 in brain sections of C57BL/6 mice housed in constant darkness. Interestingly, the signal for Dlk1 transcripts was low in the parts of the brain cortex seen in the sections, but strong in the suprachiasmatic nucleus (SCN). Abundance of Dlk1 transcripts in the SCN did not vary over time (Supplementary Material, Fig. S3, B).

Gtl2 and Rian
The Gtl2 and Rian transcripts did not show rhythmic oscillations (Fig. 2A). For both genes, mRNA levels were decreased 2-fold in the cortices of B6.MOLF12A mice compared to C57BL/6 mice (Fig 2A). Thus, expression of Gtl2 and Rian is also influenced by strain-specific factors, however, Gtl2 and Rian are the only genes in this cluster that have higher expression levels in C57BL/6 mice.

Genes adjacent to the Dlk1–Dio3 imprinted domain
We tested expression levels of two genes, Yy1 and Dync1h1, located proximal and distal to the imprinted domain, respectively. Yy1 has a low expression level, whereas Dync1h1 is actively transcribed in the adult brain cortex. None of them exhibited rhythmic or genotype-specific variation in expression levels over the day–night cycle (Fig. 2A). Like in the case of Dlk1, variable levels of Yy1 reflect inter-individual variability (data not shown).

Imprinted genes from other chromosomal regions
To determine if oscillations occurred in other imprinted genes, we tested RNA levels of three imprinted genes located in other imprinted regions and highly expressed in the cerebral cortex: potassium voltage-gated channel, subfamily Q, member 1 (Kcnq1), neuronatin (Nnat), and ubiquitin protein ligase E3A (Ube3a). We found rhythmic oscillations in Kcnq1 transcripts (Fig. 3), suggesting that diurnal variations are not uncommon in imprinted genes.


Figure 3
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Gene expression levels of imprinted genes Kcnq1, Ube3a and Nnat as function of time. For all genes, expression levels were normalized to the Hprt expression level, and the level of expression in C57BL/6 mice at ZT2 was set to 1.

 
Lack of strain-specific differences in mouse locomotor activity
Since our data show changes in diurnal expression profiles of several genes, we wondered whether these changes led to changes in locomotor activity. The locomotor activity of B6.MOLF12A and C57BL/6 mice was monitored in running wheels. Entrainment to LD cycle and total activity counts in LD and DD were not different between strains (data not shown). The free-running period calculated under DD conditions was not different between B6.MOLF12A and C57BL/6 mice (23.60±0.15 h and 23.61±0.09 h, respectively, Student t-test, P = 0.87).

Imprinting status of chromosome 12 genes in the adult brain cortex
To determine if daily variation in expression levels interfered with genomic imprinting we established the imprinting status of all the tested genes located within or outside the imprinted domain. Parent-of-origin dependent expression was tested in the cerebral cortex of F1 mice at two time points, ZT2 and ZT10 (Fig. 4). At both time points, Dlk1 was expressed from the paternal chromosome; Gtl2 and Rtl1as were expressed from the maternal chromosome, whereas Dio3, Dio3os, Yy1 and Dync1h1 were transcribed from both alleles. Mico1/Mico1os are maternally expressed (Labialle et al., submitted for publication). Maintenance of parent-of-origin specific expression of Dlk1, Mico1/Mico1os, Gtl2 and Rtl1as suggests that variation in expression levels did not cause loss of imprinting, and, conversely, that genomic imprinting did not prevent rhythmic variation in gene expression.


Figure 4
View larger version (37K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. Parent-of-origin-specific expression analysis of genes within the Yy1–Dync1h1 region in the cortex of adult F1 male mice derived from crosses between C57BL/6 females and B6.MOLF12A males. SNPs are indicated by arrows.

 
Strain-specific differences in expression levels in midgestation embryos
The current understanding of the function of genomic imprinting as a mechanism of gene dosage control implies that variation in expression levels of imprinted genes in early embryos should lead to developmental anomalies and embryonic death (30,31). If this hypothesis were correct, up-regulation of Dio3 and Rtl1as in the B6.MOLF12A mice would be expected to cause developmental anomalies and even loss of embryos. However, comparison of litter-sizes and postnatal growth dynamics in C57BL/6 and congenic mice did not detect statistically significant differences between the strains (data not shown), suggesting either that changes in dosage of these genes do not affect embryonic development, prenatal and postnatal growth, or that their MOLF-specific up-regulation occurs only later in life. Therefore, we compared the expression levels of Dio3 and Rtl1as in 8.5 and 14.5 dpc embryonic midbrains from both strains. Our results indicate that these two genes show striking up-regulation in congenic embryos at 14.5 dpc (Fig. 5). However, 8.5 dpc embryos have similar levels of expression in both strains.


Figure 5
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. Rtl1as and Dio3 gene expression levels in embryonic midbrains at 8.5 and 14.5 dpc. For the 8.5 dpc, samples were pooled from 10 embryos. For 14.5 dpc, the averages of real-time PCR assays from eight individual midbrains are presented. Expression levels of Dio3 and Rtl1as were normalized to Hprt expression and the level of expression of C57BL/6 embryos at 8.5 dpc was set to 1. Statistical analysis was done using a t-test. Asterisk: P<0.00005 between C57BL/6 and B6.MOL12A values.

 
Mapping of regulatory factors responsible for strain-specific differences in expression levels of genes from the imprinted domain
Our data show strain-specific coordinated expression of five non-paralogous genes, Mico1, Mico1os, Rtl1as, Dio3os and Dio3. In the congenic B6.MOLF12 mice, mRNAs of all five genes are oscillating, whereas in C57BL/6 mice only Mico1 and Mico1os mRNAs oscillate. As these two strains differ by a 8 cM (11.9 Mb) region in the distal part of chromosome 12, the observed strain differences are most likely due to variant cis-acting or linked trans-acting genetic factors. To map these factors, we tested the expression levels of the Rtl1as, Dio3 and Dio3os in the brain cortex of mice from additional strains and crosses at two time points that corresponded to high and low expression in the B6.MOLF12A mice (Fig. 6). For all three genes, these time points were ZT2 (high) and ZT10/ZT14 (low). Mice from the F1 crosses had expression patterns distinct from those of B6.MOLF12A mice (Fig. 6). Expression of Dio3 and Dio3os was upregulated only in homozygous B6.MOLF12A mice, but not in heterozygous animals. Rtl1as had 2-fold higher expression in heterozygous compared to C57BL/6 animals, but did not reach the 8-fold increase seen in homozygous B6.MOLF12A mice. This suggests that the C57BL/6 strain contains a repressor of the morning expression surge of Rtl1as, Dio3 and Dio3os that is closely linked to the imprinted region. Moreover, mice from the B6.MOLF12B, B6.NOD12 and B6.CAST12 congenic strains did not show upregulation of Dio3 and Dio3os. Rtl1as mRNA levels were increased about 2-fold in the B6.NOD12 mice compared to C57BL/6 mice at both time points. However, in B6.MOLF12B and B6.CAST12 mice, Rtl1as mRNA levels were similar to those observed in C57BL/6 mice. This suggests that at least two regulatory elements with strain-specific variants control Rtl1as expression. Based on the data from B6.MOLF12B mice, we excluded the region between D12Mit279 and the proximal part of the IG DMR as the candidate region for the C57BL/6-specific repressor of oscillations. The chromosome 12 distal region of the NOD/Lt strain is highly similar to that of the MOLF/Ei strain in a large interval that includes Dlk1 and the maternally expressed genes, but not Dio3 or Dio3os. The B6.CAST12 strain also contains a region derived from the Mus musculus castaneus genome that is highly similar, if not identical, to that of MOLF/Ei in the interval between positions 109.3 and 111 Mb in the Ensembl built v.44. Applying the most stringent hypothesis that the C57BL/6, NOD/LtJ and CAST/Ei strains carry the same, but the MOLF/Ei strain carries a different variant of the putative repressor of oscillations, and using the GNF2 dataset and the SNP TOOL from the Mouse Phenome Database (http://www.jax.org/phenome/snp.html), we searched for an inferred extended haplotype that would contain at least three SNPs that would fulfill the above criteria. We excluded most of the distal part of chromosome 12 and identified five regions that might contain the putative trans-acting factor. All of these regions are located outside of the imprinted region. These candidate regions partially overlap with or are located in the vicinity of the following genes: tumor necrosis factor, alpha-induced protein 2 (Tnfaip2), vasoactive intestinal peptide receptor 2 (Vipr2), protein tyrosine phosphatase, receptor type, N polypeptide 2 (Ptprn2), Rap guanine nucleotide exchange factor (GEF) 5 (Rapgef5), and ATP-binding cassette, subfamily B (MDR/TAP), member 5 (Abcb5).


Figure 6
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 6. Mapping of the genetic factors that control strain-specific expression variation of Rtl1as, Dio3, Dio3os and Rian using congenic strains of mice. For Rtl1as, Dio3 and Dio3os, gene expression levels at ZT2 and ZT14 or ZT10 in the adult brain cortex from different strains of mice and F1 crosses are shown. Each value corresponds to the mean of three to four individual mice. For Rian, the C57BL/6 and B6.MOLF12A values correspond to the average of all samples collected at different time points, all other Rian values represent the results obtained from pooled samples (n=3–6) collected at ZT2 and ZT14. Statistical analysis was done using a t-test. Asterisk: P < 0.002 compared to C57BL/6 value at ZT2; International currency symbol: P < 0.002 between value at ZT2 and value at ZT14 or ZT10, for the same genotype; N.D.: not done.

 
To map the strain-specific regulatory element of Rian, we compared Rian expression levels in the F1 mice (Fig. 6). The level of Rian mRNA in (B6.MOLF12A x C57BL/6) F1 mice was similar to the one observed in B6.MOLF12A mice; conversely, the gene expression level observed in (C57BL/6 x B6.MOLF12A) F1 hybrids was similar to the one observed in C57BL/6 mice. Therefore, expression of Rian depended upon the genotype of the maternal allele. Next, Rian RNA levels were examined in the brain cortex of the additional congenic strains, B6.MOLF12B and B6.NOD12. In B6.MOLF12B mice, Rian RNA levels were similar to those observed in C57BL/6 mice, whereas in B6.NOD12 they were similar to those observed in B6.MOLF12A mice (Fig. 6). Using the congenic mapping data, the GNF2 dataset and the SNP TOOL from the Mouse Phenome Database (http://www.jax.org/phenome/snp.html) we mapped the Rian regulatory elements to the interval between the distal boundary of the congenic region in the B6.MOLF12B strain (Fig. 1) and the distal boundary of homology between MOLF/Ei and NOD/Lt which is located near position 110.2 Mb of the NCBI mouse genome built reference assembly 36.1. This region harbors the whole cluster of maternally expressed genes including Rian. It is therefore likely that downregulation of the MOLF/Ei alleles of Rian is due to a variant cis-acting regulatory element.

Putative binding sites for CLOCK/BMAL1 are located in the vicinity of Mico1/Mico1os, Rtl1/Rtl1as and Dio3/Dio3os genes
RNA oscillations that we observed in this study have a 24 h period and hence, are likely the result of circadian rhythmicity. The CLOCK/BMAL1 complex is one of the main transcription factors regulating expression of genes controlled by circadian rhythms. This complex binds to E-boxes and, therefore, many of the clock genes and clock-controlled genes contain E-boxes in their promoter regions (32–35). Hence, we searched the sequences surrounding Mico1/Mico1os, Rtl1/Rtl1as and Dio3/Dio3os for E-boxes. We detected four CACGTG sites and three non-canonical CACGTT sites within the interval between Mico1/Mico1os and exon 1 of Gtl2 (Fig. 7). One E-box is located within the transcript and is the only one in this region that is conserved in Homo sapiens. We sequenced this region in the MOLF/Ei strain and found a mutation (CACATG) in one of the non-conserved E-boxes (Fig. 7). Another interesting feature of the region is the presence of a cAMP-responsive element (CRE) in the promoter of Gtl2, which is conserved in other species (36) and known to be present in the promoter of other rhythmic genes (37). Another group of E-boxes is located about 5 kb proximal to Mico1/Mico1os. Because of the relatively high density of E-boxes in the vicinity of Mico1/Mico1os it is a reasonable conjecture that this region contains a regulatory element that is responsible for the circadian oscillations of Mico1 and Mico1os. The Dio3/Dio3os and Rtl1/Rtl1as regions also contain several E-boxes each (Fig. 7).


Figure 7
View larger version (7K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 7. Map of the E-boxes within or near the genes that show oscillations. Black rectangles represent genes, open rectangles represent canonical E-boxes, and gray rectangles represent non-canonical E-boxes. Orientations of the non-canonical E-boxes are given as arrowheads. The CRE is shown as an open circle. Orientation of gene transcription is given as arrows. Distances between genes and E-boxes are given in bp. Star indicates the mutant E-box.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Chromosomal domains are believed to reflect the compartmentalization of genes with shared transcriptional control. Several lines of evidence suggest that physical proximity of genes is often associated with their co-regulation (38,39) (reviewed in 40). Here, we examined the interaction between two different mechanisms of gene regulation, genomic imprinting and diurnal/circadian rhythms, both of which control transcription of a cluster of non-paralogous genes in the distal region of mouse chromosome 12. We found that RNA levels of several imprinted and non-imprinted genes located within the boundaries of the imprinted domain oscillate with a 24 h period, suggesting circadian regulation. The caveat here is that due to the experimental design we would not detect the effect of biological rhythms with shorter or longer periods.

Furthermore, among the oscillating genes, three genes, Mico1, Mico1os and Rtl1as, which are transcribed from the maternal allele, conserve their imprinted state. Based on data for these genes, we conclude that diurnal oscillations do not interfere with genomic imprinting, and that the two different regulatory mechanisms are not in conflict. Conversely, our data confirm the observations by Charlier et al. (13) that genomic imprinting does not prevent changes in expression levels in imprinted genes, a fact that has to be kept in mind when conclusions regarding imprinting are made solely on the basis of gene expression levels.

The F1 band of mouse chromosome 12 contains a number of genes that are preferentially expressed in the central nervous system (CNS) (GNF SymAtlas v.1.2.4 http://symatlas.gnf.org/SymAtlas). The region of preferential CNS expression extends from the imprinted domain to non-imprinted Dync1h1 and Hsp90aa1, whose imprinting status is unknown (Fig. 8A). Interestingly, the proximal cerebral cortex-specific expression domain boundaries are different between the maternal and paternal chromosomes: on the maternal chromosome, the domain of high cerebral cortex-specific expression starts at the Mico1/Mico1os locus, whereas on the paternal chromosome, its homologous region contains paternally expressed genes that are either poorly transcribed (Dlk1) or not expressed at all (Rtl1) shifting the proximal boundary of cerebral cortex-specific expression closer to the Dio3 locus (Fig. 8A). Such a parent-of-origin dependent difference in the functional expression domain is consistent with the limited co-expression of Dlk1 and Gtl2 observed in other tissues (41).


Figure 8
View larger version (26K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 8. Strain-specific and parent-of-origin dependent effects on expression of genes in the imprinted domain of chromosome 12. (A) Extents of functional expression domains influenced by different regulatory mechanisms in the brain cortex. Boundaries of functional expression domains determined based on gene expression profiles are shown as gradient. Black box indicates the distal boundary of the oscillation domain in C57BL/6 mice that is lost in the B6.MOLF12A strain. Open circles represent the positions of Gtl2 and Rian that are excluded from the oscillation domain. (B) Model for strain-specific loss of insulator/insulator bypass of oscillatory expression in MOLF/Ei mice. Chromosomal regions from C57BL/6 and MOLF/Ei genotypes are represented as black and white rectangles, respectively. Genes are shown as gray bars. Arrows represent constant (straight arrows) or oscillatory (wavy arrows) gene expression. A putative enhancer/oscillator (Os) drives rhythmic expression. A C57BL/6-specific repressor/insulator (I), located in the distal part of chromosome 12 interacts with an insulator sequence (open box) and prevents the expansion of the oscillatory expression domain. Large shaded arrows represent the spreading of the oscillatory expression domain. As the repressor/insulator acts in trans, it prevents spreading of oscillations in F1 mice as well as in C57BL/6 homozygous mice.

 
We observed striking strain-specific differences in imprinted mRNA levels of Rtl1as and Dio3 in adult brains and in the brains of midgestation embryos. These changes in expression levels, however, were not accompanied by developmental anomalies. Our data suggest that either genomic imprinting controls expression levels of some, but not all imprinted genes, or that strict control of imprinted gene dosage is crucial during early embryonic stages, but becomes less important later in development. The observed strain-specific differences in imprinted gene expression levels in embryos raises again the question of whether genomic imprinting evolved as a mechanism of fine regulation of gene dosage and embryonic growth (30,31), or as an epigenetic security lock that prevents parthenogenesis and reinforces the necessity for sexual reproduction (42–44).

Diurnal variation in gene regulation in the cerebral cortex shows strain-specificity: in C57BL/6, only the overlapping genes Mico1 and Mico1os show oscillations in their expression, whereas homozygosity for MOLF/Ei alleles in this region causes rhythmic oscillations in Rtl1as, Dio3 and Dio3os mRNA levels. These oscillations do not occur in heterozygous mice suggesting the involvement of a linked trans-acting factor. We mapped this putative trans-acting factor to the distal part of mouse chromosome 12, outside of the imprinted region. We found five candidate regions for the trans-acting factor with the caveat that the regions that are different between MOLF/Ei and the three other strains may, in fact, be larger than defined by SNP mapping because of non-informative SNPs, for example.

In the present study, we have used different congenic strains, harboring overlapping sections of the distal region of mouse chromosome 12 on a C57BL/6 genetic background. Using mice whose genome is entirely of the C57BL/6 origin except for very specific and well-delineated regions allows simplifying the system and finding cis-acting or linked trans-acting regulatory regions, without the confounding influence of unlinked trans-acting modifiers (9). The caveat here is that our reductive approach precludes an overall comparison of circadian rhythms between C57BL/6 and molossinus strains. In contrast, this approach allowed the deduction of regulatory regions that include elements for circadian regulation of the imprinted domain of chromosome 12.

In principle, the strain-specific differences between diurnal expression profiles of Rtl1as, Dio3os and Dio3 may arise from several different mechanisms. The most parsimonious model would explain the strain-specific effects by loss, in the MOLF/Ei strain, of an insulator complex involved in the formation of the distal boundary of the domain where diurnal oscillations occur (Fig. 8B). This model implies that only one enhancer/oscillator controls expression of all five genes. Such a conjecture is supported by the similar profiles of oscillations with peaks in the morning and troughs in the evening hours. Moreover, genes transcribed in opposite orientations (Mico1/Mico1os and Dio3/Dio3os) have similar phases and even amplitudes of rhythmic expression. We propose that the gene encoding a protein involved in the formation of the distal Mico1/Mico1os oscillatory domain boundary is located in the distal part of chromosome 12. The MOLF/Ei variant of this protein is either non-functional or has a significantly reduced expression level in the cerebral cortex, which leads to insulator bypass and spreading of oscillatory expression to adjacent genes (Fig. 8B). This model may be easily modified to accommodate a scenario where the MOLF/Ei congenic region carries a factor that facilitates interaction between Dio3, Dio3os, Rtl1as and a distant enhancer/oscillator. It is not clear how Gtl2 and Rian are protected from the influence of such an oscillator. However, exclusion from circadian regulation of genes within the same chromosomal cluster of co-regulated genes has been observed in Drosophila (38) and might reflect chromatin architecture. Another possibility is that the Gtl2 and Rian RNAs are more stable, a feature that precludes oscillations, as oscillations require a short half-life of the RNA (reviewed in 45).

RNA levels of maternally expressed genes Gtl2, Rian and Rtl1as are also affected by strain-specific regulatory variants. These are likely to be cis-acting factors as they are located in the genomic interval comprising the genes involved. Rtl1as shows the most complex pattern of strain-specific effects in this region with a combination of MOLF/Ei-specific up-regulation of steady-state expression levels (in B6.MOLF12A, F1 and B6.NOD12 mice) and C57BL/6-specific repression of oscillations.

It has been suggested that all maternally expressed genes in this region are derived from a single precursor transcript (12). Our data indicate that although Gtl2 and Rian might be products of the same precursor transcript, Mico1 and Rtl1as are likely to originate from distinct transcripts as their expression patterns differ from each other and those of Gtl2 and Rian.

In conclusion, our data provide evidence for circadian co-regulation of several imprinted genes that are located in the distal part of mouse chromosome 12, suggesting that genomic imprinting does not interfere with other regulatory mechanisms and does not prevent variation in gene expression levels. Our data and proposed model provide a general framework to understand how genetic variation may lead to dramatic changes in diurnal expression profiles of multiple genes located within a given chromosomal domain.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Mice and crosses
Procedures involving animals were done in accordance with guidelines of the Canadian Council on Animal Care. The C57BL/6 strain was purchased from Charles River Laboratories. To generate the B6.MOLF12A congenic strain, mice from the congenic B6.MOLF12 strain (9) were crossed to C57BL/6 mice, the offspring was backcrossed to C57BL/6 and the next generation progeny was genotyped to find a recombinant with a shorter congenic region. The B6.MOLF12B mice were derived from the B6.MOLF12A mice using the same strategy.

The B6.NOD12 strain was derived in the following way: F1 (C57BL/6 x NOD/Lt) mice were intercrossed and their progeny was genotyped to find mice carrying a recombinant chromosome that would have a recombination breakpoint between Yy1 and Dlk1. The recombinant mouse was then backcrossed to C57BL/6. The B6.CAST12 strain has been described elsewhere (20).

Mice were housed with free access to food and water under conditions of 12 h of light and 12 h of darkness (LD), unless otherwise specified.

For the study of locomotor activity, three adult male B6MOLF12A mice and six age-matched male C57BL/6 mice (Charles River) were housed singly in running-wheel cages (Actimetrics) and wheel turns were monitored using the ClockLab program (Actimetrics). Mice were housed for 2 weeks under LD and then put in constant darkness (DD) for over 3 weeks. Actograms were generated and analyzed using the ClockLab analysis program and the free-running period was calculated from the onset of activity for cycles in DD.

Embryo collection
Embryos were collected at 8.5 and 14.5 dpc at ZT2, i.e. 2 h after lights were on. Embryonic age was estimated based on the time of detection of vaginal plug (0.5 dpc) and verified by morphological analysis.

RT and PCR
Total RNA was extracted from cerebral cortexes or embryos using Trizol (Invitrogen). All RT reactions were performed on 1 µg of total RNA treated by DNase I Amplification Grade (Invitrogen), using MMLV-Reverse Transcriptase and RNase Out (Invitrogen) following company instructions. RT reactions were primed using oligo-dT primers (Invitrogen) unless specified otherwise.

For analysis of gene expression level over the day–night cycle, RT reactions were performed on a mixture in equal proportions of RNAs from three to six animals for each time point. Diurnal differences in gene expression levels were confirmed by performing analysis on three to four individual samples per genotype. For analysis of Rtl1/Rtl1as expression in embryos, a mixture of equal proportions of RNAs from 10 (8.5 dpc) or eight (14.5 dpc) individual embryonic midbrains of each strain was used as a template for RT reactions.

Semi-quantitative PCR was performed using Taq DNA polymerase (Fermentas), ‘Taq Buffer +KCl –MgCl2’ (Fermentas) and 1 mM MgCl2, on a T3 thermocycler (Biometra). Primer sequences for RT–PCR and real-time RT–PCR experiments are provided in Supplementary Material, Table S1.

Orientation of transcription at the Rtl1/Rtl1as region was assayed by RT–PCR using specific RT primers qRtl1F or qRtl1R instead of the oligo-dT primer (Supplementary Material, Fig. S1).

Real-time PCR
Real-time PCR experiments were performed using the ABI Prism 7900HT (PE Applied Biosystems) in 384 micro-well plates. The reactions were done using the SYBR Green PCR master mix kit (Applied Biosystems) following manufacturer's instructions. Relative quantification was achieved following a standard curve method. We used 4 ng of cDNA generated from total RNA samples as templates for the PCR reactions. For the standard curves, we used 10-fold serial dilutions of cDNA. Reactions were performed over 40 cycles of 20 s at 95°C, 30 s at 60°C and 45 s at 72°C, and followed by a dissociation stage analysis. All the reactions, including the standards and non-template control (H2O) were run in duplicate or triplicate. The data presented in Figure 2 correspond to averages of at least two independent experiments. The data presented in Figures 3, 5 and 6 correspond to a single experiment done in triplicate. Primer sequences are provided in Supplementary Material, Table S1.

Parent-of-origin-specific expression
To test parent-of-origin specific expression, F1 male mice were generated from crosses between C57BL/6 females and B6.MOLF12A males. Total RNA was extracted from the cerebral cortexes of F1 mice at different time points and used as templates for RT–PCR. PCR products were sequenced. Parental origin of the transcribed alleles was established based on exonic polymorphisms. All polymorphisms that were used in these experiments were deposited into the NCBI SNP database. Sequencing analysis was performed by the Sequencing Platform of the McGill University and Genome Quebec Innovation Centre.

DIO3 enzymatic activity in adrenal glands
DIO3 enzymatic activity was determined as previously described (46). In brief, adrenal glands were homogenized in 400 µl of 10 mM Tris–HCl, 0.25 sucrose, 1 mM dithiothreitol (DTT) pH 7.5 buffer. Appropriate volumes of tissue homogenate were used in the enzymatic reaction to ensure that deiodination did not exceed 30% and was proportional to the amount of protein content. Activities were determined in duplicate in two different volumes of tissue homogenates. Homogenates were incubated at 37°C for an hour with 2 nM 125I-triiodothyronine (Perkin Elmer) in the presence of 20 mM DTT. Deiodination was determined based on the amount of 125I-3,3'-diiodothyronine produced. The latter was determined after separation of reaction products by paper chromatography (as described in 47).

In situ hybridization on brain coronal cryosections
For this experiment, C57BL/6 mice were entrained to a LD cycle for 2 weeks and then transferred to constant darkness (DD). Mice were killed by decapitation every 4 h over a 24 h period on the second day in DD. Brains were collected and in situ hybridization was performed on 10 µm-thick brain coronal cryosections. The Dlk1 antisense riboprobe, corresponding to nucleotides 441–1295 of the coding sequence (NM_010052), was produced using the Promega riboprobe in vitro transcription system in the presence of [{alpha}-35S] UTP, and then purified on an Ambion NucAway column. Brain sections were fixed in 4% paraformaldehyde and then processed for in situ hybridization (as described in 48). Slides were exposed to a Biomax MR film (Kodak).


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
FUNDING
 REFERENCES
 
Supplementary Material is available at HMG Online.


   FUNDING
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
REFERENCES
 
This work was supported by an operating grant from the Canadian Institutes of Health Research (CIHR) to A.N.; grants from Natural Science and Engineering Research Council and Human Frontier Science Program to N.C.; a NIDDK grant DK054716 to A.H. and grant HD042013 from the National Institutes of Health to J.V.S. A.N. is a recipient of the John R. and Clara M. Fraser Memorial Award of the Faculty of Medicine of McGill University. N.C. holds a salary award from the Fonds de la Recherche en Santé du Québec.


   ACKNOWLEDGEMENTS
 
We are grateful to: Sylvie Croteau and Valerie Bélanger for technical assistance and Jacquetta Trasler, Cindy Goodyer and Kennneth Morgan for comments on an earlier version of the manuscript. We thank Scott Gurd for advice concerning the real-time PCR assays.

Conflict of Interest statement. None declared.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 

  1. Kleinjan D.A., van Heyningen V. Long-range control of gene expression: emerging mechanisms and disruption in disease. Am. J. Hum. Genet. (2005) 76:8–32.[CrossRef][Web of Science][Medline]

  2. de la Casa-Esperon E., Sapienza C. Natural selection and the evolution of genome imprinting. Annu. Rev. Genet. (2003) 37:349–370.[CrossRef][Web of Science][Medline]

  3. Sladek M., Sumova A., Kovacikova Z., Bendova Z., Laurinova K., Illnerova H. Insight into molecular core clock mechanism of embryonic and early postnatal rat suprachiasmatic nucleus. Proc. Natl Acad. Sci. USA (2004) 101:6231–6236.[Abstract/Free Full Text]

  4. Bell-Pedersen D., Cassone V.M., Earnest D.J., Golden S.S., Hardin P.E., Thomas T.L., Zoran M.J. Circadian rhythms from multiple oscillators: lessons from diverse organisms. Nat. Rev. Genet. (2005) 6:544–556.[CrossRef][Web of Science][Medline]

  5. Ko C.H., Takahashi J.S. Molecular components of the mammalian circadian clock. Hum. Mol. Genet. (2006) 15:R271–R277.[Abstract/Free Full Text]

  6. Edwards C.A., Ferguson-Smith A.C. Mechanisms regulating imprinted genes in clusters. Curr. Opin. Cell. Biol. (2007) 19:281–289.[CrossRef][Web of Science][Medline]

  7. Verona R.I., Mann M.R., Bartolomei M.S. Genomic imprinting: intricacies of epigenetic regulation in clusters. Annu. Rev. Cell. Dev. Biol. (2003) 19:237–259.[CrossRef][Web of Science][Medline]

  8. Vrana P.B., Guan X.J., Ingram R.S., Tilghman S.M. Genomic Imprinting is disrupted in interspecific Peromyscus hybrids. Nat. Genet. (1998) 20:362–365.[CrossRef][Web of Science][Medline]

  9. Croteau S., Roquis D., Charron M.C., Frappier D., Yavin D., Loredo-Osti J.C., Hudson T.J., Naumova A.K. Increased plasticity of genomic imprinting of Dlk1 in brain is due to genetic and epigenetic factors. Mamm. Genome (2005) 16:127–135.[CrossRef][Web of Science][Medline]

  10. Ogawa H., Wu Q., Komiyama J., Obata Y., Kono T. Disruption of parental-specific expression of imprinted genes in uniparental fetuses. FEBS Lett. (2006) 580:5377–5384.[CrossRef][Web of Science][Medline]

  11. Shi W., Krella A., Orth A., Yu Y., Fundele R. Widespread disruption of genomic imprinting in adult interspecies mouse (Mus) hybrids. Genesis (2005) 43:100–108.[CrossRef][Web of Science][Medline]

  12. Tierling S., Dalbert S., Schoppenhorst S., Tsai C.E., Oliger S., Ferguson-Smith A.C., Paulsen M., Walter J. High-resolution map and imprinting analysis of the Gtl2-Dnchc1 domain on mouse chromosome 12. Genomics (2006) 87:225–235.[CrossRef][Web of Science][Medline]

  13. Charlier C., Segers K., Karim L., Shay T., Gyapay G., Cockett N., Georges M. The callipyge mutation enhances the expression of coregulated imprinted genes in cis without affecting their imprinting status. Nat. Genet. (2001) 27:367–369.[CrossRef][Web of Science][Medline]

  14. Baladron V., Ruiz-Hidalgo M.J., Nueda M.L., Diaz-Guerra M.J., Garcia-Ramirez J.J., Bonvini E., Gubina E., Laborda J. dlk acts as a negative regulator of Notch1 activation through interactions with specific EGF-like repeats. Exp. Cell Res. (2005) 303:343–359.[CrossRef][Web of Science][Medline]

  15. Christophersen N.S., Gronborg M., Petersen T.N., Fjord-Larsen L., Jorgensen J.R., Juliusson B., Blom N., Rosenblad C., Brundin P. Midbrain expression of Delta-like 1 homologue is regulated by GDNF and is associated with dopaminergic differentiation. Exp. Neurol. (2007) 204:791–801.[CrossRef][Web of Science][Medline]

  16. Smas C.M., Green D., Sul H.S. Structural characterization and alternate splicing of the gene encoding the preadipocyte EGF-like protein pref-1. Biochemistry (1994) 33:9257–9265.[CrossRef][Web of Science][Medline]

  17. Jensen C.H., Meyer M., Schroder H.D., Kliem A., Zimmer J., Teisner B. Neurons in the monoaminergic nuclei of the rat and human central nervous system express FA1/dlk. Neuroreport (2001) 12:3959–3963.[CrossRef][Web of Science][Medline]

  18. Hernandez A. Structure and function of the type 3 deiodinase gene. Thyroid (2005) 15:865–874.[CrossRef][Web of Science][Medline]

  19. Hernandez A., Fiering S., Martinez E., Galton V.A., St Germain D. The gene locus encoding iodothyronine deiodinase type 3 (Dio3) is imprinted in the fetus and expresses antisense transcripts. Endocrinology (2002) 143:4483–4486.[Abstract]

  20. Yevtodiyenko A., Carr M.S., Patel N., Schmidt J.t. Analysis of candidate imprinted genes linked to Dlk1-Gtl2 using a congenic mouse line. Mamm. Genome (2002) 13:633–638.[CrossRef][Web of Science][Medline]

  21. Tsai C.E., Lin S.P., Ito M., Takagi N., Takada S., Ferguson-Smith A.C. Genomic imprinting contributes to thyroid hormone metabolism in the mouse embryo. Curr. Biol. (2002) 12:1221–1226.[CrossRef][Web of Science][Medline]

  22. Schuster-Gossler K., Simon-Chazottes D., Guenet J.L., Zachgo J., Gossler A. Gtl2lacZ, an insertional mutation on mouse chromosome 12 with parental origin-dependent phenotype. Mamm. Genome (1996) 7:20–24.[CrossRef][Web of Science][Medline]

  23. Hatada I., Morita S., Obata Y., Sotomaru Y., Shimoda M., Kono T. Identification of a new imprinted gene, Rian, on mouse chromosome 12 by fluorescent differential display screening. J. Biochem. (2001) 130:187–190.[Abstract/Free Full Text]

  24. Seitz H., Youngson N., Lin S.P., Dalbert S., Paulsen M., Bachellerie J.P., Ferguson-Smith A.C., Cavaille J. Imprinted microRNA genes transcribed antisense to a reciprocally imprinted retrotransposon-like gene. Nat. Genet. (2003) 34:261–262.[CrossRef][Web of Science][Medline]

  25. Seitz H., Royo H., Bortolin M.L., Lin S.P., Ferguson-Smith A.C., Cavaille J. A large imprinted microRNA gene cluster at the mouse Dlk1-Gtl2 domain. Genome Res. (2004) 14:1741–1748.[Abstract/Free Full Text]

  26. Takada S., Paulsen M., Tevendale M., Tsai C.E., Kelsey G., Cattanach B.M., Ferguson-Smith A.C. Epigenetic analysis of the Dlk1-Gtl2 imprinted domain on mouse chromosome 12: implications for imprinting control from comparison with Igf2-H19. Hum. Mol. Genet. (2002) 11:77–86.[Abstract/Free Full Text]

  27. Lin S.P., Youngson N., Takada S., Seitz H., Reik W., Paulsen M., Cavaille J., Ferguson-Smith A.C. Asymmetric regulation of imprinting on the maternal and paternal chromosomes at the Dlk1-Gtl2 imprinted cluster on mouse chromosome 12. Nat. Genet. (2003) 35:97–102.[CrossRef][Web of Science][Medline]

  28. Steshina E.Y., Carr M.S., Glick E.A., Yevtodiyenko A., Appelbe O.K., Schmidt J.V. Loss of imprinting at the Dlk1-Gtl2 locus caused by insertional mutagenesis in the Gtl2 5' region. BMC Genet. (2006) 7:44.[CrossRef][Medline]

  29. Hernandez A., Martinez M.E., Croteau W., St Germain D.L. Complex organization and structure of sense and antisense transcripts expressed from the DIO3 gene imprinted locus. Genomics (2004) 83:413–424.[CrossRef][Web of Science][Medline]

  30. Haig D. Genomic imprinting and the theory of parent-offspring conflict. Semin. Dev. Biol. (1992) 3:153.

  31. Kato Y., Rideout W.M. III, Hilton K., Barton S.C., Tsunoda Y., Surani M.A. Developmental potential of mouse primordial germ cells. Development (1999) 126:1823–1832.[Abstract]

  32. Gekakis N., Staknis D., Nguyen H.B., Davis F.C., Wilsbacher L.D., King D.P., Takahashi J.S., Weitz C.J. Role of the CLOCK protein in the mammalian circadian mechanism. Science (1998) 280:1564–1569.[Abstract/Free Full Text]

  33. Hogenesch J.B., Gu Y.Z., Jain S., Bradfield C.A. The basic-helix-loop-helix-PAS orphan MOP3 forms transcriptionally active complexes with circadian and hypoxia factors. Proc. Natl Acad. Sci. USA (1998) 95:5474–5479.[Abstract/Free Full Text]

  34. Jin X., Shearman L.P., Weaver D.R., Zylka M.J., de Vries G.J., Reppert S.M. A molecular mechanism regulating rhythmic output from the suprachiasmatic circadian clock. Cell (1999) 96:57–68.[CrossRef][Web of Science][Medline]

  35. Yamaguchi S., Mitsui S., Miyake S., Yan L., Onishi H., Yagita K., Suzuki M., Shibata S., Kobayashi M., Okamura H. The 5' upstream region of mPer1 gene contains two promoters and is responsible for circadian oscillation. Curr. Biol. (2000) 10:873–876.[CrossRef][Web of Science][Medline]

  36. Zhao J., Zhang X., Zhou Y., Ansell P.J., Klibanski A. Cyclic AMP stimulates MEG3 gene expression in cells through a cAMP-response element (CRE) in the MEG3 proximal promoter region. Int. J. Biochem. Cell Biol. (2006) 38:1808–1820.[CrossRef][Web of Science][Medline]

  37. Travnickova-Bendova Z., Cermakian N., Reppert S.M., Sassone-Corsi P. Bimodal regulation of mPeriod promoters by CREB-dependent signaling and CLOCK/BMAL1 activity. Proc. Natl Acad. Sci. USA (2002) 99:7728–7733.[Abstract/Free Full Text]

  38. McDonald M.J., Rosbash M. Microarray analysis and organization of circadian gene expression in Drosophila. Cell (2001) 107:567–578.[CrossRef][Web of Science][Medline]

  39. Kalmykova A.I., Nurminsky D.I., Ryzhov D.V., Shevelyov Y.Y. Regulated chromatin domain comprising cluster of co-expressed genes in Drosophila melanogaster. Nucleic Acids Res. (2005) 33:1435–1444.[Abstract/Free Full Text]

  40. Dillon N. Gene regulation and large-scale chromatin organization in the nucleus. Chromosome Res. (2006) 14:117–126.[CrossRef][Web of Science][Medline]

  41. da Rocha S.T., Tevendale M., Knowles E., Takada S., Watkins M., Ferguson-Smith A.C. Restricted co-expression of Dlk1 and the reciprocally imprinted non-coding RNA, Gtl2: Implications for cis-acting control. Dev. Biol. 306:810–823.

  42. Varmuza S., Mann M. Genomic imprinting - defusing the ovarian time bomb. Trends Genet. (1994) 10:118.[CrossRef][Web of Science][Medline]

  43. Pardo-Manuel de Villena F., de la Casa-Esperon E., Sapienza C. Natural selection and the function of genome imprinting: beyond the silenced minority. Trends Genet (2000) 16:573–579.[CrossRef][Web of Science][Medline]

  44. Kono T., Obata Y., Wu Q., Niwa K., Ono Y., Yamamoto Y., Park E.S., Seo J.S., Ogawa H. Birth of parthenogenetic mice that can develop to adulthood. Nature (2004) 428:860–864.[CrossRef][Medline]

  45. Harms E., Kivimae S., Young M.W., Saez L. Posttranscriptional and posttranslational regulation of clock genes. J. Biol. Rhythms (2004) 19:361–373.[Abstract/Free Full Text]

  46. Hernandez A., Martinez M.E., Fiering S., Galton V.A., St Germain D. Type 3 deiodinase is critical for the maturation and function of the thyroid axis. J. Clin. Invest. (2006) 116:476–484.[CrossRef][Web of Science][Medline]

  47. Galton V.A., Hiebert A. The ontogeny of the enzyme systems for the 5'- and 5-deiodination of thyroid hormones in chick embryo liver. Endocrinology (1987) 120:2604–2610.[Abstract/Free Full Text]

  48. Dardente H., Poirel V.J., Klosen P., Pevet P., Masson-Pevet M. Per and neuropeptide expression in the rat suprachiasmatic nuclei: compartmentalization and differential cellular induction by light. Brain Res. (2002) 958:261–271.[CrossRef][Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Endocr. Rev.Home page
B. Gereben, A. M. Zavacki, S. Ribich, B. W. Kim, S. A. Huang, W. S. Simonides, A. Zeold, and A. C. Bianco
Cellular and Molecular Basis of Deiodinase-Regulated Thyroid Hormone Signaling
Endocr. Rev., December 1, 2008; 29(7): 898 - 938.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
17/1/15    most recent
ddm281v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Labialle, S.
Right arrow Articles by Naumova, A. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Labialle, S.
Right arrow Articles by Naumova, A. K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?