Comparative Phylogenetic Analysis of Blcap/Nnat Reveals Eutherian-Specific Imprinted Gene
http://www.100md.com
分子生物学进展 2005年第8期
* Department of Radiation Oncology, Duke University Medical Center; University Program in Genetics and Genomics, Duke University Medical Center, Durham; and Department of Genetics and Lineberger Comprehensive Cancer Center, University of North Carolina
E-mail: jirtle@radonc.duke.edu.
Abstract
Imprinted genes are parent-of-origin dependent, monoallelically expressed genes present in marsupials and eutherian mammals. Altered expression of imprinted genes plays a significant role in the etiology of a variety of human disorders and diseases. Nevertheless, the regulatory mechanisms of imprinting remain poorly defined. The imprinted gene Neuronatin (Nnat) is an excellent candidate for studying imprinting because it resides within the 8.5-kb intron of the nonimprinted gene Bladder Cancer-Associated Protein (Blcap) and is the only imprinted gene within the region. A phylogenetic comparison of this micro-imprinted domain in human, mouse, and rat revealed several candidates for imprint control, including tandem repeats and putative binding sites for trans- acting factors known to be involved in chromatin remodeling. Genome-wide phylogenetic comparisons of species from the three major extant mammalian clades failed, however, to show any evidence of Nnat outside the eutherian lineage. Thus, Nnat is the first identified eutherian-specific imprinted gene, demonstrating that imprinted genes did not arise at a single point during evolution. This finding also suggests that the complexity of imprinting regulation observed at other loci may, in part, be directly related to the amount of time they have been imprinted.
Key Words: Blcap ? eutherian ? evolution ? imprinting ? Nnat ? phylogenetics
Introduction
Genomic imprinting refers to parent-of-origin specific, monoallelic expression of a gene. The expression from a singular parental allele occurs despite identical sequence information and cellular environment. Imprinting regulation is complex with layers of control involving multiple cis- and trans-acting elements. The presence of tandem repeats, differential methylation of CpG islands, boundary elements, nontranslated RNA, and temporal differences in DNA replication have all been associated with imprinting (Reik and Walter 2001b; Delaval and Feil 2004).
Three general transcriptional regulatory motifs for imprint control have been proposed (Ferguson-Smith and Surani 2001; Reik, Dean, and Walter 2001). The most complex method uses long-range cis-acting elements to control multiple, clustered imprinted genes. Examples in humans of imprinted gene clusters regulated in this manner include the Beckwith-Wiedemann locus at 11p15.5 and the Prader-Willi/Angelman syndrome locus at 15q11-q13 (Paulsen et al. 1998; Nicholls and Knepper 2001). Secondly, some genes, such as Igf2r, are regulated in cis by the production of an imprinted antisense RNA that represses expression (Smilinich et al. 1999; Chamberlain and Brannan 2001; Sleutels, Zwart, and Barlow 2002). Thirdly, allele-specific promoter hypermethylation represses expression and leads to imprinting at some loci, including Neuronatin (Nnat) (Evans et al. 2001; John et al. 2001) and U2af1-rs1 (Nabetani et al. 1997). This last regulatory method is postulated to be the primordial imprint because it is the simplest in form (Reik and Walter 2001a).
Despite differences in the methods of imprinting regulation, all imprinted genes should share some imprinting characteristics. Inherited, differentially marked motifs that identify an imprinted allele must be present in the genome. Furthermore, sequences required for imprint regulation should be subject to the forces of natural selection, and thus be preferentially conserved both within and between species. Previous reports have shown phylogenetic comparisons between the three major extant mammalian clades Eutheria (placental mammals), Metatheria (marsupials), and Prototheria (monotremes) to be useful in identifying such conserved sequences (Killian et al. 2000; Weidman et al. 2004). These comparative phylogenetic studies were conducted with Igf2 and Igf2r, models for two of the distinct classes of imprint control—imprinted clusters and antisense RNA. A similar phylogenetic comparison has not been conducted, however, with a representative example for the third imprint motif—promoter hypermethylation. Consequently, we conducted a phylogenetic analysis of the imprinted gene Nnat, a candidate for this class of imprint regulation (Reik and Walter 2001a), in order to identify sequences potentially involved in its imprint control.
Nnat expresses highly in the central nervous system from midgestation through early postnatal development, correlating with the onset and termination of brain development in mice and humans (Joseph, Dou, and Tsang 1994; Wijnholds et al. 1995). Its expression appears to be limited, however, to the anterior pituitary in the adult brain. Nnat is also expressed in the developing pancreas, becoming restricted in adulthood to beta cells (Chu and Tsai 2005). Nnat is imprinted in both mouse and human, and all three exons of Nnat lie within the singular intron of the nonimprinted gene, Bladder Cancer-Associated Protein (Blcap) (Kagitani et al. 1997; Evans et al. 2001; John et al. 2001). This finding is especially surprising given that the intron of Blcap is only 8.5 kb, and imprinted genes can be coordinately regulated in clusters spanning distances up to 2 Mb (Buiting et al. 1995; Nicholls, Saitoh, and Horsthemke 1998; Thorvaldsen, Duran, and Bartolomei 1998). Therefore, we postulated that cis-acting sequences important for maintaining the transcriptional status of this microimprinted domain should be present within a relatively short distance of Nnat. Such sequences would allow for the differential allelic expression observed for Nnat while preventing the spread of imprinting to Blcap.
Our phylogenetic analysis of the orthologous imprinted domain of Nnat in human, mouse, and rat identified several highly conserved, noncoding sequences predicted to be important in regulating the imprinting of this gene. We also demonstrated that Nnat does not reside within the opossum Blcap gene (oBlcap) as it does in humans and mice. Moreover, we provide evidence that Nnat does not exist outside the eutherian lineage, thereby defining Nnat as the first known eutherian-specific imprinted gene.
Materials and Methods
Tissue and DNA Samples, Bacterial Artificial Chromosome Library
Tissues and DNA were obtained from the following resources—chicken: Department of Poultry Science, North Carolina State University, Raleigh, N.C.; didelphis opossum: Michael Stoskopf, North Carolina State University and the North Carolina Wildlife Commission, Raleigh, N.C.; elephant: North Carolina Zoo, Asheboro, N.C.; genomic yeast DNA: Invitrogen, Carlsbad, Calif.; human: Birth Defects Research Laboratory, University of Washington, Seattle, Wash.; lemur: Duke University Primate Center, Durham, N.C.; monodelphis opossum: Kathleen Smith, Duke University, Durham, N.C.; mouse kidney DNA: BD Biosciences, Palo Alto, Calif.; sheep: U.S. Department of Agriculture Meat Animal Research Center, Clay Center, Nebr.; and wallaby, echidna, and platypus: Barry Munday, University of Tasmania, Launceston, Tasmania. All tissues were stored at –80°C. Amplicon Express (Pullman, Wash.) constructed the 5x coverage opossum (Didelphis virginiana) genomic bacterial artificial chromosome (BAC) library. This study was performed in accordance with current regulations and standards of the U.S. Department of Agriculture, Department of Health and Human Services, and National Institutes of Health.
Probing BAC Membranes
Probes were generated using the RadPrime DNA Labeling System (Invitrogen) according to manufacturer's directions. Primers for Blcap polymerase chain reaction (PCR) template were 5'-CAG CTC CTG GAG AGA GAG TCG-3' and 5'-TGG GGA ATC GGA GCA GTG-3', and the product was amplified using Platinum Taq DNA Polymerase (Invitrogen). For Nnat, the product was amplified using the GC-RICH PCR System (Roche, Indianapolis, Ind.); primers were 5'-GGA CTC CGA GAC CAG TAG ACC-3' and 5'-TGG TGC CTA CGC CCA TAT C-3'. Cycling conditions for both reactions were 63°C annealing, 30 s extension, and 35 cycles. Membranes were prehybed in 10 ml of ExpressHyb (BD Biosciences) for 2 h at 65°C. Approximately 10 μl of denatured probe was added to 10 ml of fresh solution, and each membrane was hybridized overnight at 65°C. Membranes were washed as follows: twice 2 x standard saline citrate (SSC) for 5 min at room temperature, twice 0.5 x SSC/1% sodium dodecyl sulfate (SDS) for 30 min at 65°C, and twice 0.5 x SSC for 5 min at room temperature. Films were exposed for 5 days and then developed. Positive clones were ordered from Amplicon Express, grown, and prepped using a modified Qiagen midi protocol (Valencia, Calif.) in which the P1, P2, and P3 buffer volumes were increased to 10 ml. Elution was performed by five applications of 1 ml of elution buffer prewarmed to 65°C.
DNA Isolation
Tissue was homogenized and incubated with 30 μl of 20 mg/ml proteinase K (Qiagen) and 600 μl HCl proK buffer (50 mM Tris-HCl pH 8.0, 100 mM EDTA, 100 mM NaCl, 1% SDS) at 55°C. After overnight incubation, 10 μl of 10 mg/ml RNase A was added and incubated for 1 h at 37°C. DNA was subsequently extracted with phenol-chloroform followed by chloroform, then precipitated with an equal volume of isopropanol. The resulting pellet was suspended in 10 mM Tris-HCl, pH 8.0, and heated at 65°C to ensure complete resuspension.
Genomic and BAC Southern Blots
Approximately 10 μg of genomic DNA or 2 μg of BAC DNA were digested with 100 units of EcoRI and 100 units of BglII (New England Biolabs, Beverly, Mass.). Digested DNA was run on a 1% agarose gel. Transfer was accomplished by soaking the gel for 5 min in 250 mM HCl, twice for 15 min in 0.5 M NaOH/1.5 M NaCl, twice for 15 min in 0.5 M Tris-HCl pH 7.5/1.5 M NaCl, and for 10 min in 20 x SSC. DNA was transferred to a nylon membrane (Roche) with Turboblotter (Schleicher & Schuell, Keene, N.H.) according to manufacturer's directions. DNA was bound to the membrane by UV cross-linking.
For hybridization, probes were generated using the PCR DIG Probe Synthesis Kit (Roche). Blcap primers were 5'-CAG CTC CTG GAG AGA GAG TCG-3' and 5'-AAT GTG GTA CCC GCA GGA C-3'; PCR amplification conditions were 60°C annealing, 30 s extension, and 35 cycles. For Nnat, 1 μl of the PCR product used to generate the BAC probes was used as template with primers 5'-GGA CTC CGA GAC CAG TAG ACC-3' and 5'-TCT TCA TCA TTG TCG GTG C-3'; PCR amplification conditions were 63°C, 45 s extension, and 35 cycles. Approximately 20 μl of labeled probe was added to a prehybed membrane and incubated overnight at appropriate calculated temperatures. The membranes were washed twice with 2 x SSC/0.1% SDS at room temperature for 5 min, twice with 0.5 x SSC/0.1% SDS at 65°C for 15 min, blocked and developed using DIG Wash and Block Buffer Set (Roche) according to manufacturer's directions.
oBlcap Sequencing
Plasmids containing the subcloned 3-kb oBlcap fragment were sequenced at Duke University's Automated Sequencing Facility using 0.5 μg DNA and 1.67 mM primer. Direct sequencing of BACs was performed similarly using 2 μg DNA. Primer sequences are available upon request.
5' Rapid Amplification of cDNA Ends and Reverse Transcriptase–Polymerase Chain Reaction
The 5' rapid amplification of cDNA ends (RACE) was conducted using 5'/3' RACE Kit (Roche) according to manufacturer's directions. The cDNA synthesis primer was 5'-GAG GAT CCT AAG TCC CCA CTA TC-3'. RACE products were obtained using seminested PCR with Platinum Taq DNA Polymerase (Invitrogen). Round 1: oligo dT anchor primer (provided) and 5'-TTT CCC CAA CAG CTG TAA CAG-3'; PCR amplification conditions were 63°C annealing, 1 min extension, and 35 cycles. Round 2: 5'-AAG GCT GCC AGG AAG ACC-3' and oligo dT anchor primer; PCR amplification conditions were 60°C annealing, 1 min extension, and 35 cycles. PCR products were purified using the High Pure PCR Product Purification Kit (Roche), ligated, and transformed with pGem T-Easy Vector System II (Promega, Madison, Wis.). Clones were prepped using Qiagen mini-prep, and 4 μl was sequenced with Sp6 and T7 primers (USB, Cleveland, Ohio.). The sequences obtained were confirmed using reverse transcriptase (RT)–PCR and the following primer sets: (RE2) 5'-CGC TCC AGG AGG AAG CTC-3' with (FE2) 5'-TTC AGA ATA CGG GAC AAC TTG-3' (positive control), (FI) 5'-TGT GGT ACC AAT ATG TGA CAC C-3' (intron), (F1E1) 5'-TGT CCA ACA CCC TCC TCA TC-3' (exon 1), and (F2E1) 5'-CCC TTC AGC CTT CCA GAC-3' (exon 1); PCR amplification conditions were 60°C annealing, 1 min extension, and 35 cycles.
RNA Isolation
Opossum liver tissue was homogenized and incubated in 1 ml RNA Stat-60 (TEL-TEST, Friendswood, Tex.) for 15 min at room temperature. RNA was extracted using 200 μl of chloroform and precipitated with 500 μl of isopropanol. The pellets were washed with 1 ml of 75% ethanol and dissolved in appropriate volumes of water.
Bioinformatics
The following Web sites were used for data analyses: Blast (www.ncbi.nlm.nih.gov/BLAST), VISTA (genomes.lbl.gov/vista/index.shtml), RepeatMasker (www.repeatmasker.org), WebGene (www.itb.cnr.it/sun/webgene), Grail (compbio.ornl.gov/grailexp), and Consite (mordor.cgb.ki.se/cgi-bin/CONSITE/consite).
Results
Opossum Blcap Isolation
The opossum Nnat/Blcap domain was obtained by probing our genomic opossum BAC library using the highly conserved coding sequence of mouse Blcap (table 1). We isolated two BAC clones containing oBlcap. To guard against sequence resulting from a clone containing deletions or rearrangements, both BAC clones were sequenced simultaneously to obtain genomic oBlcap sequence.
Table 1 Relative Species Homologies to Human Blcap and Nnat
Interestingly, after obtaining 14.5 kb of sequence that included oBlcap, no sequence homologous to Nnat was identified (GenBank accession number AY821914). To confirm this finding, we conducted 5' RACE on opossum liver RNA to obtain full-length oBlcap mRNA (figs. 1 and 2). The resulting RACE product revealed that only 2 kb of intronic sequence resides between exon 1 and exon 2 of oBlcap rather than the 8.5 kb of sequence present in human and mouse (fig. 2). Furthermore, these results definitively showed that Nnat does not reside within the intron of oBlcap, as in mice and humans (Evans et al. 2001; John et al. 2001).
FIG. 1.— Opossum Blcap RACE RT-PCR. (A) Genomic structure of opossum Blcap. Open boxes denote exons, long arrow indicates direction of transcription, and small arrows indicate positions of primers for RT-PCR. (B) oBlcap RACE RT-PCRs. RACE products were confirmed by RT-PCR. All PCRs were conducted with one primer residing within the known exon 2 of oBlcap (RE2) and with another 5' primer. Lanes 1–4 contain RT, lanes 5–8 do not contain reverse transcriptase, and lane 9 is a negative PCR control. Lanes 1 and 5 show a positive control using a second primer within exon 2 (FE2—250 bp); lanes 2 and 6 show a negative control using a primer within the intronic region of oBlcap (FI); lanes 3 and 7 show a primer within exon 1 of oBlcap (F1E1—300 bp), and lanes 4 and 8 show a second, distinct primer within exon 1 (F2E1—450 bp). All band sizes are those expected for cDNA products, and no contaminating DNA is present as shown by the absence of product in all negative controls.
FIG. 2.— Evolutionary comparison of Nnat and Blcap. Genomic structures for human (NT_011362 [GenBank] ), mouse (NT_039210), didelphis opossum (AY821914 [GenBank] ), zebrafish (BX001037 [GenBank] ), drosophila (NT_037436 [GenBank] ), and Caenorhabditis elegans (NC_003279 [GenBank] ) are drawn to scale based on sequence available from the National Center for Biotechnology Information. Gray boxes are Nnat exons, white boxes are Blcap exons, black boxes are Blcap coding sequences, arrows indicate direction of transcription, and double horizontal black lines are CpG islands. Several genomic reorganization events appear to have taken place during the course of this region's evolution including a restructuring of the Blcap exon/coding sequence and the insertion of Nnat into the intron of Blcap in a common ancestor of eutherian mammals.
Nnat—A Eutherian-Specific Gene
Because Nnat was not present within the oBlcap intron, we determined if Nnat was positioned elsewhere in the opossum genome or if this gene arose in a lineage specific to eutherian mammals. To resolve this issue, we first performed a Southern blot analysis of the opossum BACs containing oBlcap to confirm that Nnat was not located at a distant site on the BACs, either endogenously or from a BAC-specific rearrangement. Southern blot analyses indicated that Nnat did not exist in the two opossum BAC clones containing oBlcap (data not shown). We then hybridized our opossum genomic BAC library with a mouse Nnat probe. No positive clones were detected, indicating that Nnat does not exist within the opossum genome (data not shown).
Finally, we performed a Southern blot analysis using DNA from a number of species in the three major mammalian lineages—eutherians (human, lemur, mouse, sheep, and elephant), marsupials (monodelphis opossum, didelphis opossum, and wallaby), and monotremes (echidna and platypus); outlier groups used were chicken and yeast. The eutherian group included species from three of the four major mammalian clades: Euarchontoglires, Laurasiatheria, and Afrotheria. Nnat-specific bands were identified in human, lemur, mouse, sheep, and elephant, but bands were not detected in any of the marsupial or monotreme species (fig. 3). To check the integrity of the DNA, we also hybridized our blots with a probe for opossum Igf2r. Using this probe we detected bands in all three of the marsupial species examined (data not shown), demonstrating that our DNA was intact. For the Nnat Southern blot analysis, available genome sequence for mouse and human correlates with the observed band sizes for these two species. The lack of Nnat in any species outside placental mammals, and the presence of Nnat in the elephant, a member of the most ancestral eutherian superordinal clade (Afrotheria), indicates that Nnat originated in eutherians after their divergence from marsupials.
FIG. 3.— Southern blot analysis. DNA from eutherians (human, n = 2; lemur, n = 2; mouse, n = 2; sheep, n = 3; and elephant, n = 2), marsupials (monodelphis opossum, n = 2; didelphis opossum, n = 1; and wallaby, n = 2), monotremes (echidna, n = 1; and platypus, n = 2), chicken (n = 2), and yeast (n = 3) were used for Southern blot analysis. Mouse coding sequence for Nnat was used as a probe. Bands (arrowheads) were only observed for eutherian mammals; band sizes are as expected for species with known sequence (human = 5.8 kb and 700 bp, mouse = 3.5 kb). The top band in the mouse lane represents a Nnat pseudogene on mouse chromosome 7 (6 kb).
Blast Searches
To further our search for Nnat and its origins, we performed several Blast searches. We performed a Blast search of Nnat full-length mouse mRNA and protein against the nonredundant, EST, month, chromosome, and all species-specific databases presently available. Nnat ESTs were identified in human, mouse, rat, hamster, chimpanzee, cow, pig, and dog. No Nnat ESTs were identified from mRNA or protein searches against fungi, eukaryotes, chicken, frog, drosophila, malaria, nematode, plant, viral, sea urchin, zebrafish, or monodelphis opossum databases. These results support our finding that Nnat is indeed a eutherian-specific gene. Moreover, these Blast searches revealed no other gene homologous to Nnat, and it has not been classified into any known family.
VISTA Analysis
In order to identify cis-acting sequences that may be important for the imprinting of Nnat, we performed a VISTA analysis of the Nnat/Blcap domain using available sequence for all species known to be imprinted at this locus: human, mouse, and rat, all of which belong to the Euarchontoglires mammalian clade. Didelphis opossum was used as an outlier for comparison because this species represents the closest relation to the Euarchontoglires for which Nnat is definitively not imprinted. This examination revealed a highly conserved syntenic region between exon 1 and exon 2 of Blcap in human, mouse, and rat (fig. 4A). Conversely, the only sequence to show homology between human and opossum is the coding sequence and the very 3'untranslated region of Blcap. Because Nnat is absent in the opossum and imprinted in every eutherian mammal investigated, these highly conserved regions may be involved in transcriptional rather than imprinting regulation of Nnat. Nevertheless, some interesting characteristics worthy of note do appear from our phylogenetic comparative analysis.
FIG. 4.— Multispecies comparison of Blcap/Nnat domain. (A) VISTA analysis of human (GenBank, AL109614), mouse (GenBank, AL672259), rat (GenBank, NW_047659), and didelphis opossum (GenBank, AY821914). Genomic sequences were compared using VISTA to determine evolutionary conserved regions among all four species using human as the reference sequence. Exons are shown in blue and are conserved, conserved nonexonic regions are shown in pink. We used 75% homology and a 50-bp sliding window as our parameters for determining conservation. This region shows high homology in human, mouse, and rat with virtually no homology between human and opossum. The position of the small L2 repeat potentially responsible for transposition of Nnat is displayed as a small red bar at the 3' end of Nnat exon 3. Yellow bars indicate locations of tandem repeats. Dashed box represents the region enlarged in figure 4B. (B) Conserved region of interest in human, mouse, and rat that lies immediately 5' to Blcap exon 2 (blue box). It contains a CAGA simple repeat followed immediately by an A residue simple repeat (yellow box), a putative SUH-binding site (teal box), and a putative CTCF-binding site (purple box). The human sequence of the putative SUH- and CTCF-binding sites are shown (top lines) in comparison with their relative consensus binding sites (bottom line); mismatches are shown in red.
Firstly, VISTA plots show the presence of several repeats within the Blcap domain. A tandem repeat exists at both the 5' and 3' ends of the Blcap intron. These repeats abut the Blcap exons in human, mouse, and rat, but are absent in opossum. The VISTA plot also shows the presence of two L2 repeats within the Nnat gene. The largest L2 repeat resides within the first intron of Nnat, but other bioinformatics programs (RepeatMasker, WebGene, Grail) failed to recognize this sequence as a repeat. This region is highly A rich, a characteristic often observed for matrix attachment regions (MARs) (Cockerill and Garrard 1986). The second L2 is located at the 3' end of Nnat and is conserved in human, mouse, and rat.
Interestingly, a Blast search identified a mouse Nnat pseudogene on chromosome 7. The 6-kb band in the mouse lane of our Southern blot analysis correlates with the expected band size for this pseudogene. The Nnat pseudogene contains only exons 1 and 3 and lacks both introns and an open reading frame; however, the small L2 repeat was conserved. This finding indicated that the Nnat gene may derive from a retroelement and that the small L2 repeat may be responsible for the transposition of Nnat into the Blcap locus in an eutherian ancestor.
Finally, the conserved noncoding sequences delineated by VISTA were further screened for the presence of known DNA-binding motifs (Consite). Most of the binding sites identified were for transcription factors necessary for the regulation of gene expression. Nevertheless, we did detect a binding site for the human homolog of Suppressor of Hairless (SUH) in the region immediately 5' to Blcap exon 2 and adjacent to one of the tandem repeats previously mentioned (fig. 4B).
This region was also analyzed for the presence of putative CTCF-binding sites. Although CTCF binds to a wide variety of DNA sequences (Mukhopadhyay et al. 2004), a consensus imprinted gene-binding motif for CTCF has been identified from studies of the Igf2/H19 and DLK1/GTL2 JRIG domains (Wylie et al. 2000). We utilized this specific consensus sequence to screen for putative CTCF-binding sites in the Blcap/Nnat domain. Interestingly, we found the presence of one possible CTCF-binding site that is conserved in human, mouse, and rat, but is absent in the opossum. This putative binding site differs from the previous consensus site by a single residue and resides in the same location as the putative SUH-binding site and tandem repeat immediately upstream of Blcap exon 2 (fig. 4B).
Discussion
Comparative phylogenetics not only provides a powerful tool to identify crucial imprint regulatory sequences, but this technique can also offer insight into the process of imprinting evolution. The Nnat/Blcap micro-imprinted domain presents a unique region for study because the imprinted gene Nnat must maintain an expression pattern that is distinct from that of its close, nonimprinted neighbor, Blcap. We provide evidence herein by Southern blot and genomic sequence analysis that Nnat arose in a lineage specific to placental mammals, providing the first evidence of an eutherian-specific imprinted gene.
Comparative phylogenomic analyses of the Blcap/Nnat region allows for the visualization of the Blcap intronic expansion that accompanied the apparent transposition of Nnat into a common ancestor to eutherians. We have identified a small L2 repeat that may be responsible for the transposition of Nnat because this repeat is conserved in the human, mouse, and rat Nnat gene, as well as in the mouse Nnat pseudogene on chromosome 7. Similarly, U2af1-rs1 is imprinted in mouse and sits within intron 1 of the biallelically expressed Murr1 (Nabetani et al. 1997). This gene has also been proposed to have arisen through retrotransposition. Therefore, genes that transpose into the genome may trigger imprinting either through the transposition event itself or through the presence of sequence common to both transposition and imprinting regulation. Further evidence for the linkage between transposition and epigenetic controls comes from plants and mouse cells in which altered methylation levels enhance transposition (Miura et al. 2001; Ros and Kunze 2001; Singer, Yordan, and Martienssen 2001 Yusa, Takeda, and Horie 2004).
Where Nnat transposed from initially is still unknown. Genetic theory states that all genes arose from a rearrangement, deletion, mutation, or duplication of genes within a single ancestral genome (Cooper 1999). The intronless U2af1-rs1 most likely derived from transposition of its mouse paralogue U2af1-rs2. While Nnat has limited structural similarity to the proteolipid class of proteins (Dou and Joseph 1996), Blast searches with Nnat mRNA, exons, and protein did not reveal sequence homologies to any other gene. Thus, we found no evidence for an ancestral origin of this gene.
The eutherian-specific lineage and imprinting of Nnat is particularly interesting because expression profiles indicate that this gene plays an important role in brain development. In addition, mouse embryos carrying paternal duplications of the region containing Nnat (i.e., two functional copies of Nnat) have a notable increase in the depth of the cerebellar folds (Kikyo et al. 1997). Conversely, a reduction of cerebellar fold depth is present in embryos with a maternal duplication of this area (i.e., no functional copies of Nnat) (Kikyo et al. 1997). Therefore, it is tempting to speculate that the transposition of this small imprinted gene provided early eutherian mammals with a selective advantage in responding to cognitive pressures.
Moreover, our finding of Nnat as an eutherian-specific imprinted gene sheds new light on the evolution of imprinting. Previous studies demonstrated that Igf2r and Igf2 imprinting is present in therian mammals but not in monotremes (Killian et al. 2000; O'Neill et al. 2000; Killian et al. 2001). These findings suggest that genomic imprinting originated approximately 180 MYA in an ancestor common to both marsupials and eutherian mammals sometime after their divergence from monotremes (Penny et al. 1999; Hasegawa, Thorne, and Kishino 2003; Woodburne, Rich, and Springer 2003). Because Nnat does not exist in the marsupial lineage, the phenomenon of imprinting had already evolved by the time Nnat transposed into the eutherian lineage and became imprinted. Thus, our studies demonstrate for the first time that not all imprinted genes evolved at a single evolutionary time point because Nnat, and possibly other genes, acquired this unique form of regulation after the initial development of imprinting.
Our findings also lend weight to an evolutionary hypothesis of Kaneko-Ishino, Kohda, and Ishino (2003) that postulates that imprinted genes arose with the origination of placental tissues. Consequently, imprinting was vital to the speciation of therian mammals and therefore cannot be easily reversed. Once in place, other genes could usurp the imprinting machinery either by chance or by selective force. Interestingly, of the six genes examined within the placenta in their study, Nnat did not follow the extraembryonic expression patterns of the other five genes. Because we have shown that Nnat does not exist within the marsupial lineage, it was not present in the genome at the time of placentation. Consequently, Nnat would not have been involved in the initial establishment of imprinting, but would have acquired this mechanism of gene regulation after imprinting was established within the genome of therian mammals.
Our data, however, does not preclude other theories of imprinting evolution including the widely discussed kinship theory (Wilkins and Haig 2003). This theory states that imprinting of genes evolved due to a parental conflict between the genomes to control the extraction of resources from the mother by her offspring. Moreover, an imprinted gene's expression will confer a fitness benefit to a parental lineage. For example, genes expressed from the paternal allele often encourage fetal growth, while maternally expressed genes often suppress growth in an effort to conserve resources for the mother (Wilkins and Haig 2003). Accordingly, a role for Nnat in metabolism has recently been demonstrated because downregulation of this imprinted gene decreases insulin production in response to glucose stimulation (Chu and Tsai 2005). In addition, embryos carrying a maternal duplication of the region containing Nnat were 60%–80% smaller than embryos with normal Nnat genotypes (Kikyo et al. 1997). Alternatively, Nnat may provide a benefit to the patriarchal lineage by ensuring appropriate nurturing behavior in offspring as seen for other imprinted genes such as Peg1 and Peg3 (Lefebvre et al. 1998, Li et al. 1999).
The absence of Nnat in marsupials also supports the postulate that promoter hypermethylation represents the primordial imprint mark (Ferguson-Smith and Surani 2001; Reik, Dean, and Walter 2001). Because Nnat has clearly had less time to develop the layers of transcriptional control observed in genes known to be imprinted more ancestrally such as Igf2r, Igf2, and Peg1/Mest (Killian et al. 2000; O'Neill et al. 2000; Killian et al. 2001; Suzuki et al. 2005), imprinting at this locus may be mostly or even entirely regulated by the differential promoter methylation identified in eutherian mammals (Kikyo et al. 1997; Evans et al. 2001). Given enough evolutionary time, genes that utilize simple promoter hypermethylation may acquire additional levels of control and achieve the same degree of complexity as imprinted genes that arose earlier in the course of imprinting evolution, a hypothesis that will be further tested as more mammalian genomes are sequenced and compared.
Finally, our phylogenetic analysis of the Blcap/Nnat micro-imprinted domain reveals several intriguing, conserved structures including a possible MAR, putative binding sites for CTCF and SUH, and a tandem repeat both 5' and 3' of Nnat. With the exception of SUH, all of these factors have been implicated in establishing and/or maintaining imprinting (Neumann, Kubicka, and Barlow 1995; Greally et al. 1999; Bell and Felsenfeld 2000; Yoon et al. 2002). Interestingly, although we cannot define the boundaries of the ancestral Blcap intron versus that of eutherian mammals, neither tandem repeat is conserved in the opossum despite its virtual juxtaposition to the Blcap exons in eutherians. Furthermore, the conserved noncoding region containing one of these repeats also contains putative binding sites for the human homolog of SUH and CTCF. While SUH has not been implicated previously in imprinting, this finding is nonetheless intriguing because SUH acts as a transcriptional repressor when complexed with other factors such as histone deacetylase to form a chromatin remodeling complex (Hsieh et al. 1999). It is conceivable that such a complex would play a role in the imprinting process as histone deacetylases are known to be critical components of the imprinting machinery (Nan et al. 1998).
Our phylogenetic examination of the Nnat/Blcap domain has uncovered several genomic motifs whose role in the imprint regulation of Nnat would be interesting to determine experimentally. Moreover, even though Nnat is highly conserved in eutherian mammals and imprinted in mouse and human, it does not exist in species outside the eutherian lineage. These findings demonstrate that after imprinting evolved in an initial subset of genes in viviparous mammals during the Jurassic era (Killian et al. 2000; O'Neill et al. 2000; Killian et al. 2001), other genes continued to acquire imprinting, at least in eutherians. Indeed, approximately a 70 Myr span exists between the divergence of the Therian and Eutherian base species (Hasegawa, Thorne, and Kishino 2003; Woodburne, Rich, and Springer 2003), providing ample time for the formation of eutherian-specific imprinted genes such as Nnat. Continuing phylogenetic analyses of imprinted domains may reveal additional examples of the diversity of imprinting between species and thereby enhance our understanding of the mechanisms of imprinting, the history of imprint evolution, and the role these genes play in speciation and brain development.
Acknowledgements
We wish to thank individuals and institutions that aided our efforts in collecting tissues for this study, including Guy Liechty and the North Carolina Zoo. We are also grateful to Scott Langdon and the Duke Sequencing Facility for their assistance in obtaining sequence described in this work. This study was supported by National Institutes of Health grants CA25951, ES08823, and ES13053.
References
Bell, A. C., and G. Felsenfeld. 2000. Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene. Nature 405:486–489.
Buiting, K., S. Saitoh, S. Gross, B. Dittrich, S. Schwartz, R. D. Nicholls, and B. Horsthemke. 1995. Inherited microdeletions in the Angelman and Prader-Willi syndromes define an imprinting centre on human chromosome 15. Nat. Genet. 9:395–400.
Chamberlain, S. J., and C. I. Brannan. 2001. The Prader-Willi syndrome imprinting center activates the paternally expressed murine Ube3a antisense transcript but represses paternal Ube3a. Genomics 73:316–322.
Chu, K., and M. J. Tsai. 2005. Neuronatin, a downstream target of BETA2/NeuroD1 in the pancreas, is involved in glucose-mediated insulin secretion. Diabetes 54:1064–1073.
Cockerill, P. N., and W. T. Garrard. 1986. Chromosomal loop anchorage of the kappa immunoglobulin gene occurs next to the enhancer in a region containing topoisomerase II sites. Cell 44:273–282.
Cooper, D. N. 1999. Human gene evolution. BIOS Scientific Publishers, Oxford.
Delaval, K., and R. Feil. 2004. Epigenetic regulation of mammalian genomic imprinting. Curr. Opin. Genet. Dev. 14:188–195.
Dou, D., and R. Joseph. 1996. Cloning of human Neuronatin gene and its localization to chromosome 20q11.2-12: the deduced protein is a novel ‘proteolipid’. Brain Res. 723:8–22.
Evans, H. K., A. A. Wylie, S. K. Murphy, and R. L. Jirtle. 2001. The Neuronatin gene resides in a "micro-imprinted" domain on human chromosome 20q11.2. Genomics 77:99–104.
Ferguson-Smith, A., and M. Surani. 2001. Imprinting and the epigenetic asymmetry between parental genomes. Science 293:1086–1089.
Greally, J. M., T. A. Gray, J. M. Gabriel, L. Q. Song, S. Zemel, and R. D. Nicholls. 1999. Conserved characteristics of heterochromatin-forming DNA at the 15q11-q13 imprinting center. Proc. Natl. Acad. Sci. USA 96:14430–14435.
Hasegawa, M., J. L. Thorne, and H. Kishino. 2003. Time scale of eutherian evolution estimated without assuming a constant rate of molecular evolution. Genes Genet. Syst. 78:267–283.
Hsieh, J. J.-D., S. Zhou, L. Chen, D. B. Young, and S. D. Hayward. 1999. CIR, a corepressor linking the DNA binding factor CBF1 to the histone deacetylase complex. Proc. Natl. Acad. Sci. USA 96:23–28.
John, R. M., S. A. Aparicio, J. F. Ainscough, K. L. Arney, S. Khosla, K. Hawker, K. J. Hilton, S. C. Barton, and M. A. Surani. 2001. Imprinted expression of Neuronatin from modified BAC transgenes reveals regulation by distinct and distant enhancers. Dev. Biol. 236:387–399.
Joseph, R., D. Dou, and W. Tsang. 1994. Molecular cloning of a novel mRNA (Neuronatin) that is highly expressed in neonatal mammalian brain. Biochem. Biophys. Res. Commun. 201:1227–1234.
Kagitani, F., Y. Kuroiwa, S. Wakana, T. Shiroishi, N. Miyoshi, S. Kobayashi, M. Nishida, T. Kohda, T. Kaneko-Ishino, and F. Ishino. 1997. Peg5/Neuronatin is an imprinted gene located on sub-distal chromosome 2 in the mouse. Nucleic Acids Res. 25:3428–3432.
Kaneko-Ishino, T., T. Kohda, and F. Ishino. 2003. The regulation and biological significance of genomic imprinting in mammals. J. Biochem. 133:699–711.
Kikyo, N., C. M. Williamson, R. M. John, S. C. Barton, C. V. Beechey, S. T. Ball, B. M. Cattanach, M. A. Surani, and J. Peters. 1997. Genetic and functional analysis of Neuronatin in mice with maternal or paternal duplication of distal chr 2. Dev. Biol. 190:66–77.
Killian, J. K., J. C. Byrd, J. V. Jirtle, B. L. Munday, M. K. Stoskopf, R. G. MacDonald, and R. L. Jirtle. 2000. M6P/IGF2R imprinting evolution in mammals. Mol. Cell 5:707–716.
Killian, J. K., C. M. Nolan, N. Stewart, B. L. Munday, N. A. Andersen, S. Nicol, and R. L. Jirtle. 2001. Monotreme IGF2 expression and ancestral origin of genomic imprinting. J. Exp. Zool. 291:205–212.
Lefebvre, L., S. Viville, S. C. Barton, F. Ishino, E. B. Keverne, and M. A. Surani. 1998. Abnormal maternal behaviour and growth retardation associated with loss of the imprinted gene Mest. Nat. Genet. 20:163–169.
Li, L., E. B. Keverne, S. A. Aparicio, F. Ishino, S. C. Barton, M. A. Surani. 1999. Regulation of maternal behavior and offspring growth by paternally expressed Peg3. Science 284:330–333.
Miura, A., S. Yonebayashi, K. Watanabe, T. Toyama, H. Shimada, and T. Kakutani. 2001. Mobilization of transposons by a mutation abolishing full DNA methylation in Arabidopsis. Nature 411:212–214.
Mukhopadhyay, R., W. Yu, J. Whitehead et al. (15 co-authors). 2004. The binding sites for the chromatin insulator protein CTCF map to DNA methylation-free domains genome wide. Genome Res. 14:1594–1602.
Nabetani, A., I. Hatada, H. Morisaki, M. Oshimura, and T. Mukai. 1997. Mouse U2af1-rs1 is a neomorphic imprinted gene. Mol. Cell. Biol. 17:789–798.
Nan, X., H. H. Ng, C. A. Johnson, C. D. Laherty, B. M. Turner, R. N. Eisenman, and A. Bird. 1998. Transcriptional repression by the methyl-CpG binding protein MeCP2 involves a histone deacetylase complex. Nature 393:386–389.
Neumann, B., P. Kubicka, and D. P. Barlow. 1995. Characteristics of imprinted genes. Nat. Genet. 9:12–13.
Nicholls, R. D., and J. L. Knepper. 2001. Genome organization, function, and imprinting in Prader-Willi and Angelman syndromes. Annu. Rev. Genomics Hum. Genet. 2:153–175.
Nicholls, R. D., S. Saitoh, and B. Horsthemke. 1998. Imprinting in Prader-Willi and Angelman syndromes. Trends Genet. 14:194–200.
O'Neill, M. J., R. S. Ingram, P. B. Vrana, and S. M. Tilghman. 2000. Allelic expression of IGF2 in marsupials and birds. Dev. Genes Evol. 210:18–20.
Paulsen, M., K. R. Davies, L. M. Bowden et al. (15 co-authors). 1998. Syntenic organization of the mouse distal chromosome 7 imprinting cluster and the Beckwith-Wiedemann syndrome region in chromosome 11p15.5. Hum. Mol. Genet. 7:1149–1159.
Penny, D., M. Hasegawa, P. J. Waddell, and M. D. Hendy. 1999. Mammalian evolution: timing and implications from using the LogDeterminant transform for proteins of differing amino acid composition. Syst. Biol. 48:76–93.
Reik, W., W. Dean, and J. Walter. 2001. Epigenetic reprogramming in mammalian development. Science 293:1089–1093.
Reik, W., and J. Walter. 2001a. Evolution of imprinting mechanisms: the battle of the sexes begins in the zygote. Nat. Genet. 27:255–256.
Reik, W., and J. Walter. 2001b. Genomic imprinting: parental influence on the genome. Nat. Rev. Genet. 2:21–32.
Ros, F., and R. Kunze. 2001. Regulation of activator/dissociation transposition by replication and DNA methylation. Genetics 157:1723–1733.
Singer, T., C. Yordan, and R. A. Martienssen. 2001. Robertson's mutator transposons in A. thaliana are regulated by the chromatin-remodeling gene Decrease in DNA Methylation (DDM1). Genes Dev. 15:591–602.
Sleutels, F., R. Zwart, and D. P. Barlow. 2002. The non-coding Air RNA is required for silencing autosomal imprinted genes. Nature 415:810–813.
Smilinich, N. J., C. D. Day, G. V. Fitzpatrick et al. (16 co-authors). 1999. A maternally methylated CpG island in KvLQT1 is associated with an antisense paternal transcript and loss of imprinting in Beckwith-Wiedemann syndrome. Proc. Natl. Acad. Sci. USA 96:8064–8069.
Suzuki, S., M. B. Renfree, A. J. Pask, G. Shaw, S. Kobayashi, T. Kohda, T. Kaneko-Ishino, and F. Ishino. 2005. Genomc imprinting of IGF2, p57(KIP2) and PEG1/MEST in a marsupial, the tammar wallaby. Mech. Dev. 122:213–222.
Thorvaldsen, J. L., K. L. Duran, and M. S. Bartolomei. 1998. Deletion of the H19 differentially methylated domain results in loss of imprinted expression of H19 and Igf2. Genes Dev. 12:3693–3702.
Weidman, J. R., S. K. Murphy, C. M. Nolan, F. S. Dietrich, and R. L. Jirtle. 2004. Phylogenetic footprint analysis of IGF2 in extant mammals. Genome Res. 14:1726–1732.
Wijnholds, J., K. Chowdhury, R. Wehr, and P. Gruss. 1995. Segment-specific expression of the Neuronatin gene during early hindbrain development. Dev. Biol. 171:73–84.
Wilkins, J. F., and D. Haig. 2003. What good is genomic imprinting: the function of parent-specific gene expression. Nat. Rev. Genet. 4:359–368.
Woodburne, M. O., T. H. Rich, and M. S. Springer. 2003. The evolution of tribospheny and the antiquity of mammalian clades. Mol. Phylogenet. Evol. 28:360–385.
Wylie, A. A., S. K. Murphy, T. C. Orton, and R. L. Jirtle. 2000. Novel imprinted DLK1/GTL2 domain on human chromosome 14 contains motifs that mimic those implicated in IGF2/H19 regulation. Genome Res. 10:1711–1718.
Yoon, B. J., H. Herman, A. Sikora, L. T. Smith, C. Plass, and P. D. Soloway. 2002. Regulation of DNA methylation of Rasgrf1. Nat. Genet. 30:92–96.
Yusa, K., J. Takeda, and K. Horie. 2004. Enhancement of sleeping beauty transpostion by CpG methylation: possible role of heterochromatin formation. Mol. Cell. Biol. 24:4004–4018.(Heather K. Evans*,, Jenni)
E-mail: jirtle@radonc.duke.edu.
Abstract
Imprinted genes are parent-of-origin dependent, monoallelically expressed genes present in marsupials and eutherian mammals. Altered expression of imprinted genes plays a significant role in the etiology of a variety of human disorders and diseases. Nevertheless, the regulatory mechanisms of imprinting remain poorly defined. The imprinted gene Neuronatin (Nnat) is an excellent candidate for studying imprinting because it resides within the 8.5-kb intron of the nonimprinted gene Bladder Cancer-Associated Protein (Blcap) and is the only imprinted gene within the region. A phylogenetic comparison of this micro-imprinted domain in human, mouse, and rat revealed several candidates for imprint control, including tandem repeats and putative binding sites for trans- acting factors known to be involved in chromatin remodeling. Genome-wide phylogenetic comparisons of species from the three major extant mammalian clades failed, however, to show any evidence of Nnat outside the eutherian lineage. Thus, Nnat is the first identified eutherian-specific imprinted gene, demonstrating that imprinted genes did not arise at a single point during evolution. This finding also suggests that the complexity of imprinting regulation observed at other loci may, in part, be directly related to the amount of time they have been imprinted.
Key Words: Blcap ? eutherian ? evolution ? imprinting ? Nnat ? phylogenetics
Introduction
Genomic imprinting refers to parent-of-origin specific, monoallelic expression of a gene. The expression from a singular parental allele occurs despite identical sequence information and cellular environment. Imprinting regulation is complex with layers of control involving multiple cis- and trans-acting elements. The presence of tandem repeats, differential methylation of CpG islands, boundary elements, nontranslated RNA, and temporal differences in DNA replication have all been associated with imprinting (Reik and Walter 2001b; Delaval and Feil 2004).
Three general transcriptional regulatory motifs for imprint control have been proposed (Ferguson-Smith and Surani 2001; Reik, Dean, and Walter 2001). The most complex method uses long-range cis-acting elements to control multiple, clustered imprinted genes. Examples in humans of imprinted gene clusters regulated in this manner include the Beckwith-Wiedemann locus at 11p15.5 and the Prader-Willi/Angelman syndrome locus at 15q11-q13 (Paulsen et al. 1998; Nicholls and Knepper 2001). Secondly, some genes, such as Igf2r, are regulated in cis by the production of an imprinted antisense RNA that represses expression (Smilinich et al. 1999; Chamberlain and Brannan 2001; Sleutels, Zwart, and Barlow 2002). Thirdly, allele-specific promoter hypermethylation represses expression and leads to imprinting at some loci, including Neuronatin (Nnat) (Evans et al. 2001; John et al. 2001) and U2af1-rs1 (Nabetani et al. 1997). This last regulatory method is postulated to be the primordial imprint because it is the simplest in form (Reik and Walter 2001a).
Despite differences in the methods of imprinting regulation, all imprinted genes should share some imprinting characteristics. Inherited, differentially marked motifs that identify an imprinted allele must be present in the genome. Furthermore, sequences required for imprint regulation should be subject to the forces of natural selection, and thus be preferentially conserved both within and between species. Previous reports have shown phylogenetic comparisons between the three major extant mammalian clades Eutheria (placental mammals), Metatheria (marsupials), and Prototheria (monotremes) to be useful in identifying such conserved sequences (Killian et al. 2000; Weidman et al. 2004). These comparative phylogenetic studies were conducted with Igf2 and Igf2r, models for two of the distinct classes of imprint control—imprinted clusters and antisense RNA. A similar phylogenetic comparison has not been conducted, however, with a representative example for the third imprint motif—promoter hypermethylation. Consequently, we conducted a phylogenetic analysis of the imprinted gene Nnat, a candidate for this class of imprint regulation (Reik and Walter 2001a), in order to identify sequences potentially involved in its imprint control.
Nnat expresses highly in the central nervous system from midgestation through early postnatal development, correlating with the onset and termination of brain development in mice and humans (Joseph, Dou, and Tsang 1994; Wijnholds et al. 1995). Its expression appears to be limited, however, to the anterior pituitary in the adult brain. Nnat is also expressed in the developing pancreas, becoming restricted in adulthood to beta cells (Chu and Tsai 2005). Nnat is imprinted in both mouse and human, and all three exons of Nnat lie within the singular intron of the nonimprinted gene, Bladder Cancer-Associated Protein (Blcap) (Kagitani et al. 1997; Evans et al. 2001; John et al. 2001). This finding is especially surprising given that the intron of Blcap is only 8.5 kb, and imprinted genes can be coordinately regulated in clusters spanning distances up to 2 Mb (Buiting et al. 1995; Nicholls, Saitoh, and Horsthemke 1998; Thorvaldsen, Duran, and Bartolomei 1998). Therefore, we postulated that cis-acting sequences important for maintaining the transcriptional status of this microimprinted domain should be present within a relatively short distance of Nnat. Such sequences would allow for the differential allelic expression observed for Nnat while preventing the spread of imprinting to Blcap.
Our phylogenetic analysis of the orthologous imprinted domain of Nnat in human, mouse, and rat identified several highly conserved, noncoding sequences predicted to be important in regulating the imprinting of this gene. We also demonstrated that Nnat does not reside within the opossum Blcap gene (oBlcap) as it does in humans and mice. Moreover, we provide evidence that Nnat does not exist outside the eutherian lineage, thereby defining Nnat as the first known eutherian-specific imprinted gene.
Materials and Methods
Tissue and DNA Samples, Bacterial Artificial Chromosome Library
Tissues and DNA were obtained from the following resources—chicken: Department of Poultry Science, North Carolina State University, Raleigh, N.C.; didelphis opossum: Michael Stoskopf, North Carolina State University and the North Carolina Wildlife Commission, Raleigh, N.C.; elephant: North Carolina Zoo, Asheboro, N.C.; genomic yeast DNA: Invitrogen, Carlsbad, Calif.; human: Birth Defects Research Laboratory, University of Washington, Seattle, Wash.; lemur: Duke University Primate Center, Durham, N.C.; monodelphis opossum: Kathleen Smith, Duke University, Durham, N.C.; mouse kidney DNA: BD Biosciences, Palo Alto, Calif.; sheep: U.S. Department of Agriculture Meat Animal Research Center, Clay Center, Nebr.; and wallaby, echidna, and platypus: Barry Munday, University of Tasmania, Launceston, Tasmania. All tissues were stored at –80°C. Amplicon Express (Pullman, Wash.) constructed the 5x coverage opossum (Didelphis virginiana) genomic bacterial artificial chromosome (BAC) library. This study was performed in accordance with current regulations and standards of the U.S. Department of Agriculture, Department of Health and Human Services, and National Institutes of Health.
Probing BAC Membranes
Probes were generated using the RadPrime DNA Labeling System (Invitrogen) according to manufacturer's directions. Primers for Blcap polymerase chain reaction (PCR) template were 5'-CAG CTC CTG GAG AGA GAG TCG-3' and 5'-TGG GGA ATC GGA GCA GTG-3', and the product was amplified using Platinum Taq DNA Polymerase (Invitrogen). For Nnat, the product was amplified using the GC-RICH PCR System (Roche, Indianapolis, Ind.); primers were 5'-GGA CTC CGA GAC CAG TAG ACC-3' and 5'-TGG TGC CTA CGC CCA TAT C-3'. Cycling conditions for both reactions were 63°C annealing, 30 s extension, and 35 cycles. Membranes were prehybed in 10 ml of ExpressHyb (BD Biosciences) for 2 h at 65°C. Approximately 10 μl of denatured probe was added to 10 ml of fresh solution, and each membrane was hybridized overnight at 65°C. Membranes were washed as follows: twice 2 x standard saline citrate (SSC) for 5 min at room temperature, twice 0.5 x SSC/1% sodium dodecyl sulfate (SDS) for 30 min at 65°C, and twice 0.5 x SSC for 5 min at room temperature. Films were exposed for 5 days and then developed. Positive clones were ordered from Amplicon Express, grown, and prepped using a modified Qiagen midi protocol (Valencia, Calif.) in which the P1, P2, and P3 buffer volumes were increased to 10 ml. Elution was performed by five applications of 1 ml of elution buffer prewarmed to 65°C.
DNA Isolation
Tissue was homogenized and incubated with 30 μl of 20 mg/ml proteinase K (Qiagen) and 600 μl HCl proK buffer (50 mM Tris-HCl pH 8.0, 100 mM EDTA, 100 mM NaCl, 1% SDS) at 55°C. After overnight incubation, 10 μl of 10 mg/ml RNase A was added and incubated for 1 h at 37°C. DNA was subsequently extracted with phenol-chloroform followed by chloroform, then precipitated with an equal volume of isopropanol. The resulting pellet was suspended in 10 mM Tris-HCl, pH 8.0, and heated at 65°C to ensure complete resuspension.
Genomic and BAC Southern Blots
Approximately 10 μg of genomic DNA or 2 μg of BAC DNA were digested with 100 units of EcoRI and 100 units of BglII (New England Biolabs, Beverly, Mass.). Digested DNA was run on a 1% agarose gel. Transfer was accomplished by soaking the gel for 5 min in 250 mM HCl, twice for 15 min in 0.5 M NaOH/1.5 M NaCl, twice for 15 min in 0.5 M Tris-HCl pH 7.5/1.5 M NaCl, and for 10 min in 20 x SSC. DNA was transferred to a nylon membrane (Roche) with Turboblotter (Schleicher & Schuell, Keene, N.H.) according to manufacturer's directions. DNA was bound to the membrane by UV cross-linking.
For hybridization, probes were generated using the PCR DIG Probe Synthesis Kit (Roche). Blcap primers were 5'-CAG CTC CTG GAG AGA GAG TCG-3' and 5'-AAT GTG GTA CCC GCA GGA C-3'; PCR amplification conditions were 60°C annealing, 30 s extension, and 35 cycles. For Nnat, 1 μl of the PCR product used to generate the BAC probes was used as template with primers 5'-GGA CTC CGA GAC CAG TAG ACC-3' and 5'-TCT TCA TCA TTG TCG GTG C-3'; PCR amplification conditions were 63°C, 45 s extension, and 35 cycles. Approximately 20 μl of labeled probe was added to a prehybed membrane and incubated overnight at appropriate calculated temperatures. The membranes were washed twice with 2 x SSC/0.1% SDS at room temperature for 5 min, twice with 0.5 x SSC/0.1% SDS at 65°C for 15 min, blocked and developed using DIG Wash and Block Buffer Set (Roche) according to manufacturer's directions.
oBlcap Sequencing
Plasmids containing the subcloned 3-kb oBlcap fragment were sequenced at Duke University's Automated Sequencing Facility using 0.5 μg DNA and 1.67 mM primer. Direct sequencing of BACs was performed similarly using 2 μg DNA. Primer sequences are available upon request.
5' Rapid Amplification of cDNA Ends and Reverse Transcriptase–Polymerase Chain Reaction
The 5' rapid amplification of cDNA ends (RACE) was conducted using 5'/3' RACE Kit (Roche) according to manufacturer's directions. The cDNA synthesis primer was 5'-GAG GAT CCT AAG TCC CCA CTA TC-3'. RACE products were obtained using seminested PCR with Platinum Taq DNA Polymerase (Invitrogen). Round 1: oligo dT anchor primer (provided) and 5'-TTT CCC CAA CAG CTG TAA CAG-3'; PCR amplification conditions were 63°C annealing, 1 min extension, and 35 cycles. Round 2: 5'-AAG GCT GCC AGG AAG ACC-3' and oligo dT anchor primer; PCR amplification conditions were 60°C annealing, 1 min extension, and 35 cycles. PCR products were purified using the High Pure PCR Product Purification Kit (Roche), ligated, and transformed with pGem T-Easy Vector System II (Promega, Madison, Wis.). Clones were prepped using Qiagen mini-prep, and 4 μl was sequenced with Sp6 and T7 primers (USB, Cleveland, Ohio.). The sequences obtained were confirmed using reverse transcriptase (RT)–PCR and the following primer sets: (RE2) 5'-CGC TCC AGG AGG AAG CTC-3' with (FE2) 5'-TTC AGA ATA CGG GAC AAC TTG-3' (positive control), (FI) 5'-TGT GGT ACC AAT ATG TGA CAC C-3' (intron), (F1E1) 5'-TGT CCA ACA CCC TCC TCA TC-3' (exon 1), and (F2E1) 5'-CCC TTC AGC CTT CCA GAC-3' (exon 1); PCR amplification conditions were 60°C annealing, 1 min extension, and 35 cycles.
RNA Isolation
Opossum liver tissue was homogenized and incubated in 1 ml RNA Stat-60 (TEL-TEST, Friendswood, Tex.) for 15 min at room temperature. RNA was extracted using 200 μl of chloroform and precipitated with 500 μl of isopropanol. The pellets were washed with 1 ml of 75% ethanol and dissolved in appropriate volumes of water.
Bioinformatics
The following Web sites were used for data analyses: Blast (www.ncbi.nlm.nih.gov/BLAST), VISTA (genomes.lbl.gov/vista/index.shtml), RepeatMasker (www.repeatmasker.org), WebGene (www.itb.cnr.it/sun/webgene), Grail (compbio.ornl.gov/grailexp), and Consite (mordor.cgb.ki.se/cgi-bin/CONSITE/consite).
Results
Opossum Blcap Isolation
The opossum Nnat/Blcap domain was obtained by probing our genomic opossum BAC library using the highly conserved coding sequence of mouse Blcap (table 1). We isolated two BAC clones containing oBlcap. To guard against sequence resulting from a clone containing deletions or rearrangements, both BAC clones were sequenced simultaneously to obtain genomic oBlcap sequence.
Table 1 Relative Species Homologies to Human Blcap and Nnat
Interestingly, after obtaining 14.5 kb of sequence that included oBlcap, no sequence homologous to Nnat was identified (GenBank accession number AY821914). To confirm this finding, we conducted 5' RACE on opossum liver RNA to obtain full-length oBlcap mRNA (figs. 1 and 2). The resulting RACE product revealed that only 2 kb of intronic sequence resides between exon 1 and exon 2 of oBlcap rather than the 8.5 kb of sequence present in human and mouse (fig. 2). Furthermore, these results definitively showed that Nnat does not reside within the intron of oBlcap, as in mice and humans (Evans et al. 2001; John et al. 2001).
FIG. 1.— Opossum Blcap RACE RT-PCR. (A) Genomic structure of opossum Blcap. Open boxes denote exons, long arrow indicates direction of transcription, and small arrows indicate positions of primers for RT-PCR. (B) oBlcap RACE RT-PCRs. RACE products were confirmed by RT-PCR. All PCRs were conducted with one primer residing within the known exon 2 of oBlcap (RE2) and with another 5' primer. Lanes 1–4 contain RT, lanes 5–8 do not contain reverse transcriptase, and lane 9 is a negative PCR control. Lanes 1 and 5 show a positive control using a second primer within exon 2 (FE2—250 bp); lanes 2 and 6 show a negative control using a primer within the intronic region of oBlcap (FI); lanes 3 and 7 show a primer within exon 1 of oBlcap (F1E1—300 bp), and lanes 4 and 8 show a second, distinct primer within exon 1 (F2E1—450 bp). All band sizes are those expected for cDNA products, and no contaminating DNA is present as shown by the absence of product in all negative controls.
FIG. 2.— Evolutionary comparison of Nnat and Blcap. Genomic structures for human (NT_011362 [GenBank] ), mouse (NT_039210), didelphis opossum (AY821914 [GenBank] ), zebrafish (BX001037 [GenBank] ), drosophila (NT_037436 [GenBank] ), and Caenorhabditis elegans (NC_003279 [GenBank] ) are drawn to scale based on sequence available from the National Center for Biotechnology Information. Gray boxes are Nnat exons, white boxes are Blcap exons, black boxes are Blcap coding sequences, arrows indicate direction of transcription, and double horizontal black lines are CpG islands. Several genomic reorganization events appear to have taken place during the course of this region's evolution including a restructuring of the Blcap exon/coding sequence and the insertion of Nnat into the intron of Blcap in a common ancestor of eutherian mammals.
Nnat—A Eutherian-Specific Gene
Because Nnat was not present within the oBlcap intron, we determined if Nnat was positioned elsewhere in the opossum genome or if this gene arose in a lineage specific to eutherian mammals. To resolve this issue, we first performed a Southern blot analysis of the opossum BACs containing oBlcap to confirm that Nnat was not located at a distant site on the BACs, either endogenously or from a BAC-specific rearrangement. Southern blot analyses indicated that Nnat did not exist in the two opossum BAC clones containing oBlcap (data not shown). We then hybridized our opossum genomic BAC library with a mouse Nnat probe. No positive clones were detected, indicating that Nnat does not exist within the opossum genome (data not shown).
Finally, we performed a Southern blot analysis using DNA from a number of species in the three major mammalian lineages—eutherians (human, lemur, mouse, sheep, and elephant), marsupials (monodelphis opossum, didelphis opossum, and wallaby), and monotremes (echidna and platypus); outlier groups used were chicken and yeast. The eutherian group included species from three of the four major mammalian clades: Euarchontoglires, Laurasiatheria, and Afrotheria. Nnat-specific bands were identified in human, lemur, mouse, sheep, and elephant, but bands were not detected in any of the marsupial or monotreme species (fig. 3). To check the integrity of the DNA, we also hybridized our blots with a probe for opossum Igf2r. Using this probe we detected bands in all three of the marsupial species examined (data not shown), demonstrating that our DNA was intact. For the Nnat Southern blot analysis, available genome sequence for mouse and human correlates with the observed band sizes for these two species. The lack of Nnat in any species outside placental mammals, and the presence of Nnat in the elephant, a member of the most ancestral eutherian superordinal clade (Afrotheria), indicates that Nnat originated in eutherians after their divergence from marsupials.
FIG. 3.— Southern blot analysis. DNA from eutherians (human, n = 2; lemur, n = 2; mouse, n = 2; sheep, n = 3; and elephant, n = 2), marsupials (monodelphis opossum, n = 2; didelphis opossum, n = 1; and wallaby, n = 2), monotremes (echidna, n = 1; and platypus, n = 2), chicken (n = 2), and yeast (n = 3) were used for Southern blot analysis. Mouse coding sequence for Nnat was used as a probe. Bands (arrowheads) were only observed for eutherian mammals; band sizes are as expected for species with known sequence (human = 5.8 kb and 700 bp, mouse = 3.5 kb). The top band in the mouse lane represents a Nnat pseudogene on mouse chromosome 7 (6 kb).
Blast Searches
To further our search for Nnat and its origins, we performed several Blast searches. We performed a Blast search of Nnat full-length mouse mRNA and protein against the nonredundant, EST, month, chromosome, and all species-specific databases presently available. Nnat ESTs were identified in human, mouse, rat, hamster, chimpanzee, cow, pig, and dog. No Nnat ESTs were identified from mRNA or protein searches against fungi, eukaryotes, chicken, frog, drosophila, malaria, nematode, plant, viral, sea urchin, zebrafish, or monodelphis opossum databases. These results support our finding that Nnat is indeed a eutherian-specific gene. Moreover, these Blast searches revealed no other gene homologous to Nnat, and it has not been classified into any known family.
VISTA Analysis
In order to identify cis-acting sequences that may be important for the imprinting of Nnat, we performed a VISTA analysis of the Nnat/Blcap domain using available sequence for all species known to be imprinted at this locus: human, mouse, and rat, all of which belong to the Euarchontoglires mammalian clade. Didelphis opossum was used as an outlier for comparison because this species represents the closest relation to the Euarchontoglires for which Nnat is definitively not imprinted. This examination revealed a highly conserved syntenic region between exon 1 and exon 2 of Blcap in human, mouse, and rat (fig. 4A). Conversely, the only sequence to show homology between human and opossum is the coding sequence and the very 3'untranslated region of Blcap. Because Nnat is absent in the opossum and imprinted in every eutherian mammal investigated, these highly conserved regions may be involved in transcriptional rather than imprinting regulation of Nnat. Nevertheless, some interesting characteristics worthy of note do appear from our phylogenetic comparative analysis.
FIG. 4.— Multispecies comparison of Blcap/Nnat domain. (A) VISTA analysis of human (GenBank, AL109614), mouse (GenBank, AL672259), rat (GenBank, NW_047659), and didelphis opossum (GenBank, AY821914). Genomic sequences were compared using VISTA to determine evolutionary conserved regions among all four species using human as the reference sequence. Exons are shown in blue and are conserved, conserved nonexonic regions are shown in pink. We used 75% homology and a 50-bp sliding window as our parameters for determining conservation. This region shows high homology in human, mouse, and rat with virtually no homology between human and opossum. The position of the small L2 repeat potentially responsible for transposition of Nnat is displayed as a small red bar at the 3' end of Nnat exon 3. Yellow bars indicate locations of tandem repeats. Dashed box represents the region enlarged in figure 4B. (B) Conserved region of interest in human, mouse, and rat that lies immediately 5' to Blcap exon 2 (blue box). It contains a CAGA simple repeat followed immediately by an A residue simple repeat (yellow box), a putative SUH-binding site (teal box), and a putative CTCF-binding site (purple box). The human sequence of the putative SUH- and CTCF-binding sites are shown (top lines) in comparison with their relative consensus binding sites (bottom line); mismatches are shown in red.
Firstly, VISTA plots show the presence of several repeats within the Blcap domain. A tandem repeat exists at both the 5' and 3' ends of the Blcap intron. These repeats abut the Blcap exons in human, mouse, and rat, but are absent in opossum. The VISTA plot also shows the presence of two L2 repeats within the Nnat gene. The largest L2 repeat resides within the first intron of Nnat, but other bioinformatics programs (RepeatMasker, WebGene, Grail) failed to recognize this sequence as a repeat. This region is highly A rich, a characteristic often observed for matrix attachment regions (MARs) (Cockerill and Garrard 1986). The second L2 is located at the 3' end of Nnat and is conserved in human, mouse, and rat.
Interestingly, a Blast search identified a mouse Nnat pseudogene on chromosome 7. The 6-kb band in the mouse lane of our Southern blot analysis correlates with the expected band size for this pseudogene. The Nnat pseudogene contains only exons 1 and 3 and lacks both introns and an open reading frame; however, the small L2 repeat was conserved. This finding indicated that the Nnat gene may derive from a retroelement and that the small L2 repeat may be responsible for the transposition of Nnat into the Blcap locus in an eutherian ancestor.
Finally, the conserved noncoding sequences delineated by VISTA were further screened for the presence of known DNA-binding motifs (Consite). Most of the binding sites identified were for transcription factors necessary for the regulation of gene expression. Nevertheless, we did detect a binding site for the human homolog of Suppressor of Hairless (SUH) in the region immediately 5' to Blcap exon 2 and adjacent to one of the tandem repeats previously mentioned (fig. 4B).
This region was also analyzed for the presence of putative CTCF-binding sites. Although CTCF binds to a wide variety of DNA sequences (Mukhopadhyay et al. 2004), a consensus imprinted gene-binding motif for CTCF has been identified from studies of the Igf2/H19 and DLK1/GTL2 JRIG domains (Wylie et al. 2000). We utilized this specific consensus sequence to screen for putative CTCF-binding sites in the Blcap/Nnat domain. Interestingly, we found the presence of one possible CTCF-binding site that is conserved in human, mouse, and rat, but is absent in the opossum. This putative binding site differs from the previous consensus site by a single residue and resides in the same location as the putative SUH-binding site and tandem repeat immediately upstream of Blcap exon 2 (fig. 4B).
Discussion
Comparative phylogenetics not only provides a powerful tool to identify crucial imprint regulatory sequences, but this technique can also offer insight into the process of imprinting evolution. The Nnat/Blcap micro-imprinted domain presents a unique region for study because the imprinted gene Nnat must maintain an expression pattern that is distinct from that of its close, nonimprinted neighbor, Blcap. We provide evidence herein by Southern blot and genomic sequence analysis that Nnat arose in a lineage specific to placental mammals, providing the first evidence of an eutherian-specific imprinted gene.
Comparative phylogenomic analyses of the Blcap/Nnat region allows for the visualization of the Blcap intronic expansion that accompanied the apparent transposition of Nnat into a common ancestor to eutherians. We have identified a small L2 repeat that may be responsible for the transposition of Nnat because this repeat is conserved in the human, mouse, and rat Nnat gene, as well as in the mouse Nnat pseudogene on chromosome 7. Similarly, U2af1-rs1 is imprinted in mouse and sits within intron 1 of the biallelically expressed Murr1 (Nabetani et al. 1997). This gene has also been proposed to have arisen through retrotransposition. Therefore, genes that transpose into the genome may trigger imprinting either through the transposition event itself or through the presence of sequence common to both transposition and imprinting regulation. Further evidence for the linkage between transposition and epigenetic controls comes from plants and mouse cells in which altered methylation levels enhance transposition (Miura et al. 2001; Ros and Kunze 2001; Singer, Yordan, and Martienssen 2001 Yusa, Takeda, and Horie 2004).
Where Nnat transposed from initially is still unknown. Genetic theory states that all genes arose from a rearrangement, deletion, mutation, or duplication of genes within a single ancestral genome (Cooper 1999). The intronless U2af1-rs1 most likely derived from transposition of its mouse paralogue U2af1-rs2. While Nnat has limited structural similarity to the proteolipid class of proteins (Dou and Joseph 1996), Blast searches with Nnat mRNA, exons, and protein did not reveal sequence homologies to any other gene. Thus, we found no evidence for an ancestral origin of this gene.
The eutherian-specific lineage and imprinting of Nnat is particularly interesting because expression profiles indicate that this gene plays an important role in brain development. In addition, mouse embryos carrying paternal duplications of the region containing Nnat (i.e., two functional copies of Nnat) have a notable increase in the depth of the cerebellar folds (Kikyo et al. 1997). Conversely, a reduction of cerebellar fold depth is present in embryos with a maternal duplication of this area (i.e., no functional copies of Nnat) (Kikyo et al. 1997). Therefore, it is tempting to speculate that the transposition of this small imprinted gene provided early eutherian mammals with a selective advantage in responding to cognitive pressures.
Moreover, our finding of Nnat as an eutherian-specific imprinted gene sheds new light on the evolution of imprinting. Previous studies demonstrated that Igf2r and Igf2 imprinting is present in therian mammals but not in monotremes (Killian et al. 2000; O'Neill et al. 2000; Killian et al. 2001). These findings suggest that genomic imprinting originated approximately 180 MYA in an ancestor common to both marsupials and eutherian mammals sometime after their divergence from monotremes (Penny et al. 1999; Hasegawa, Thorne, and Kishino 2003; Woodburne, Rich, and Springer 2003). Because Nnat does not exist in the marsupial lineage, the phenomenon of imprinting had already evolved by the time Nnat transposed into the eutherian lineage and became imprinted. Thus, our studies demonstrate for the first time that not all imprinted genes evolved at a single evolutionary time point because Nnat, and possibly other genes, acquired this unique form of regulation after the initial development of imprinting.
Our findings also lend weight to an evolutionary hypothesis of Kaneko-Ishino, Kohda, and Ishino (2003) that postulates that imprinted genes arose with the origination of placental tissues. Consequently, imprinting was vital to the speciation of therian mammals and therefore cannot be easily reversed. Once in place, other genes could usurp the imprinting machinery either by chance or by selective force. Interestingly, of the six genes examined within the placenta in their study, Nnat did not follow the extraembryonic expression patterns of the other five genes. Because we have shown that Nnat does not exist within the marsupial lineage, it was not present in the genome at the time of placentation. Consequently, Nnat would not have been involved in the initial establishment of imprinting, but would have acquired this mechanism of gene regulation after imprinting was established within the genome of therian mammals.
Our data, however, does not preclude other theories of imprinting evolution including the widely discussed kinship theory (Wilkins and Haig 2003). This theory states that imprinting of genes evolved due to a parental conflict between the genomes to control the extraction of resources from the mother by her offspring. Moreover, an imprinted gene's expression will confer a fitness benefit to a parental lineage. For example, genes expressed from the paternal allele often encourage fetal growth, while maternally expressed genes often suppress growth in an effort to conserve resources for the mother (Wilkins and Haig 2003). Accordingly, a role for Nnat in metabolism has recently been demonstrated because downregulation of this imprinted gene decreases insulin production in response to glucose stimulation (Chu and Tsai 2005). In addition, embryos carrying a maternal duplication of the region containing Nnat were 60%–80% smaller than embryos with normal Nnat genotypes (Kikyo et al. 1997). Alternatively, Nnat may provide a benefit to the patriarchal lineage by ensuring appropriate nurturing behavior in offspring as seen for other imprinted genes such as Peg1 and Peg3 (Lefebvre et al. 1998, Li et al. 1999).
The absence of Nnat in marsupials also supports the postulate that promoter hypermethylation represents the primordial imprint mark (Ferguson-Smith and Surani 2001; Reik, Dean, and Walter 2001). Because Nnat has clearly had less time to develop the layers of transcriptional control observed in genes known to be imprinted more ancestrally such as Igf2r, Igf2, and Peg1/Mest (Killian et al. 2000; O'Neill et al. 2000; Killian et al. 2001; Suzuki et al. 2005), imprinting at this locus may be mostly or even entirely regulated by the differential promoter methylation identified in eutherian mammals (Kikyo et al. 1997; Evans et al. 2001). Given enough evolutionary time, genes that utilize simple promoter hypermethylation may acquire additional levels of control and achieve the same degree of complexity as imprinted genes that arose earlier in the course of imprinting evolution, a hypothesis that will be further tested as more mammalian genomes are sequenced and compared.
Finally, our phylogenetic analysis of the Blcap/Nnat micro-imprinted domain reveals several intriguing, conserved structures including a possible MAR, putative binding sites for CTCF and SUH, and a tandem repeat both 5' and 3' of Nnat. With the exception of SUH, all of these factors have been implicated in establishing and/or maintaining imprinting (Neumann, Kubicka, and Barlow 1995; Greally et al. 1999; Bell and Felsenfeld 2000; Yoon et al. 2002). Interestingly, although we cannot define the boundaries of the ancestral Blcap intron versus that of eutherian mammals, neither tandem repeat is conserved in the opossum despite its virtual juxtaposition to the Blcap exons in eutherians. Furthermore, the conserved noncoding region containing one of these repeats also contains putative binding sites for the human homolog of SUH and CTCF. While SUH has not been implicated previously in imprinting, this finding is nonetheless intriguing because SUH acts as a transcriptional repressor when complexed with other factors such as histone deacetylase to form a chromatin remodeling complex (Hsieh et al. 1999). It is conceivable that such a complex would play a role in the imprinting process as histone deacetylases are known to be critical components of the imprinting machinery (Nan et al. 1998).
Our phylogenetic examination of the Nnat/Blcap domain has uncovered several genomic motifs whose role in the imprint regulation of Nnat would be interesting to determine experimentally. Moreover, even though Nnat is highly conserved in eutherian mammals and imprinted in mouse and human, it does not exist in species outside the eutherian lineage. These findings demonstrate that after imprinting evolved in an initial subset of genes in viviparous mammals during the Jurassic era (Killian et al. 2000; O'Neill et al. 2000; Killian et al. 2001), other genes continued to acquire imprinting, at least in eutherians. Indeed, approximately a 70 Myr span exists between the divergence of the Therian and Eutherian base species (Hasegawa, Thorne, and Kishino 2003; Woodburne, Rich, and Springer 2003), providing ample time for the formation of eutherian-specific imprinted genes such as Nnat. Continuing phylogenetic analyses of imprinted domains may reveal additional examples of the diversity of imprinting between species and thereby enhance our understanding of the mechanisms of imprinting, the history of imprint evolution, and the role these genes play in speciation and brain development.
Acknowledgements
We wish to thank individuals and institutions that aided our efforts in collecting tissues for this study, including Guy Liechty and the North Carolina Zoo. We are also grateful to Scott Langdon and the Duke Sequencing Facility for their assistance in obtaining sequence described in this work. This study was supported by National Institutes of Health grants CA25951, ES08823, and ES13053.
References
Bell, A. C., and G. Felsenfeld. 2000. Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene. Nature 405:486–489.
Buiting, K., S. Saitoh, S. Gross, B. Dittrich, S. Schwartz, R. D. Nicholls, and B. Horsthemke. 1995. Inherited microdeletions in the Angelman and Prader-Willi syndromes define an imprinting centre on human chromosome 15. Nat. Genet. 9:395–400.
Chamberlain, S. J., and C. I. Brannan. 2001. The Prader-Willi syndrome imprinting center activates the paternally expressed murine Ube3a antisense transcript but represses paternal Ube3a. Genomics 73:316–322.
Chu, K., and M. J. Tsai. 2005. Neuronatin, a downstream target of BETA2/NeuroD1 in the pancreas, is involved in glucose-mediated insulin secretion. Diabetes 54:1064–1073.
Cockerill, P. N., and W. T. Garrard. 1986. Chromosomal loop anchorage of the kappa immunoglobulin gene occurs next to the enhancer in a region containing topoisomerase II sites. Cell 44:273–282.
Cooper, D. N. 1999. Human gene evolution. BIOS Scientific Publishers, Oxford.
Delaval, K., and R. Feil. 2004. Epigenetic regulation of mammalian genomic imprinting. Curr. Opin. Genet. Dev. 14:188–195.
Dou, D., and R. Joseph. 1996. Cloning of human Neuronatin gene and its localization to chromosome 20q11.2-12: the deduced protein is a novel ‘proteolipid’. Brain Res. 723:8–22.
Evans, H. K., A. A. Wylie, S. K. Murphy, and R. L. Jirtle. 2001. The Neuronatin gene resides in a "micro-imprinted" domain on human chromosome 20q11.2. Genomics 77:99–104.
Ferguson-Smith, A., and M. Surani. 2001. Imprinting and the epigenetic asymmetry between parental genomes. Science 293:1086–1089.
Greally, J. M., T. A. Gray, J. M. Gabriel, L. Q. Song, S. Zemel, and R. D. Nicholls. 1999. Conserved characteristics of heterochromatin-forming DNA at the 15q11-q13 imprinting center. Proc. Natl. Acad. Sci. USA 96:14430–14435.
Hasegawa, M., J. L. Thorne, and H. Kishino. 2003. Time scale of eutherian evolution estimated without assuming a constant rate of molecular evolution. Genes Genet. Syst. 78:267–283.
Hsieh, J. J.-D., S. Zhou, L. Chen, D. B. Young, and S. D. Hayward. 1999. CIR, a corepressor linking the DNA binding factor CBF1 to the histone deacetylase complex. Proc. Natl. Acad. Sci. USA 96:23–28.
John, R. M., S. A. Aparicio, J. F. Ainscough, K. L. Arney, S. Khosla, K. Hawker, K. J. Hilton, S. C. Barton, and M. A. Surani. 2001. Imprinted expression of Neuronatin from modified BAC transgenes reveals regulation by distinct and distant enhancers. Dev. Biol. 236:387–399.
Joseph, R., D. Dou, and W. Tsang. 1994. Molecular cloning of a novel mRNA (Neuronatin) that is highly expressed in neonatal mammalian brain. Biochem. Biophys. Res. Commun. 201:1227–1234.
Kagitani, F., Y. Kuroiwa, S. Wakana, T. Shiroishi, N. Miyoshi, S. Kobayashi, M. Nishida, T. Kohda, T. Kaneko-Ishino, and F. Ishino. 1997. Peg5/Neuronatin is an imprinted gene located on sub-distal chromosome 2 in the mouse. Nucleic Acids Res. 25:3428–3432.
Kaneko-Ishino, T., T. Kohda, and F. Ishino. 2003. The regulation and biological significance of genomic imprinting in mammals. J. Biochem. 133:699–711.
Kikyo, N., C. M. Williamson, R. M. John, S. C. Barton, C. V. Beechey, S. T. Ball, B. M. Cattanach, M. A. Surani, and J. Peters. 1997. Genetic and functional analysis of Neuronatin in mice with maternal or paternal duplication of distal chr 2. Dev. Biol. 190:66–77.
Killian, J. K., J. C. Byrd, J. V. Jirtle, B. L. Munday, M. K. Stoskopf, R. G. MacDonald, and R. L. Jirtle. 2000. M6P/IGF2R imprinting evolution in mammals. Mol. Cell 5:707–716.
Killian, J. K., C. M. Nolan, N. Stewart, B. L. Munday, N. A. Andersen, S. Nicol, and R. L. Jirtle. 2001. Monotreme IGF2 expression and ancestral origin of genomic imprinting. J. Exp. Zool. 291:205–212.
Lefebvre, L., S. Viville, S. C. Barton, F. Ishino, E. B. Keverne, and M. A. Surani. 1998. Abnormal maternal behaviour and growth retardation associated with loss of the imprinted gene Mest. Nat. Genet. 20:163–169.
Li, L., E. B. Keverne, S. A. Aparicio, F. Ishino, S. C. Barton, M. A. Surani. 1999. Regulation of maternal behavior and offspring growth by paternally expressed Peg3. Science 284:330–333.
Miura, A., S. Yonebayashi, K. Watanabe, T. Toyama, H. Shimada, and T. Kakutani. 2001. Mobilization of transposons by a mutation abolishing full DNA methylation in Arabidopsis. Nature 411:212–214.
Mukhopadhyay, R., W. Yu, J. Whitehead et al. (15 co-authors). 2004. The binding sites for the chromatin insulator protein CTCF map to DNA methylation-free domains genome wide. Genome Res. 14:1594–1602.
Nabetani, A., I. Hatada, H. Morisaki, M. Oshimura, and T. Mukai. 1997. Mouse U2af1-rs1 is a neomorphic imprinted gene. Mol. Cell. Biol. 17:789–798.
Nan, X., H. H. Ng, C. A. Johnson, C. D. Laherty, B. M. Turner, R. N. Eisenman, and A. Bird. 1998. Transcriptional repression by the methyl-CpG binding protein MeCP2 involves a histone deacetylase complex. Nature 393:386–389.
Neumann, B., P. Kubicka, and D. P. Barlow. 1995. Characteristics of imprinted genes. Nat. Genet. 9:12–13.
Nicholls, R. D., and J. L. Knepper. 2001. Genome organization, function, and imprinting in Prader-Willi and Angelman syndromes. Annu. Rev. Genomics Hum. Genet. 2:153–175.
Nicholls, R. D., S. Saitoh, and B. Horsthemke. 1998. Imprinting in Prader-Willi and Angelman syndromes. Trends Genet. 14:194–200.
O'Neill, M. J., R. S. Ingram, P. B. Vrana, and S. M. Tilghman. 2000. Allelic expression of IGF2 in marsupials and birds. Dev. Genes Evol. 210:18–20.
Paulsen, M., K. R. Davies, L. M. Bowden et al. (15 co-authors). 1998. Syntenic organization of the mouse distal chromosome 7 imprinting cluster and the Beckwith-Wiedemann syndrome region in chromosome 11p15.5. Hum. Mol. Genet. 7:1149–1159.
Penny, D., M. Hasegawa, P. J. Waddell, and M. D. Hendy. 1999. Mammalian evolution: timing and implications from using the LogDeterminant transform for proteins of differing amino acid composition. Syst. Biol. 48:76–93.
Reik, W., W. Dean, and J. Walter. 2001. Epigenetic reprogramming in mammalian development. Science 293:1089–1093.
Reik, W., and J. Walter. 2001a. Evolution of imprinting mechanisms: the battle of the sexes begins in the zygote. Nat. Genet. 27:255–256.
Reik, W., and J. Walter. 2001b. Genomic imprinting: parental influence on the genome. Nat. Rev. Genet. 2:21–32.
Ros, F., and R. Kunze. 2001. Regulation of activator/dissociation transposition by replication and DNA methylation. Genetics 157:1723–1733.
Singer, T., C. Yordan, and R. A. Martienssen. 2001. Robertson's mutator transposons in A. thaliana are regulated by the chromatin-remodeling gene Decrease in DNA Methylation (DDM1). Genes Dev. 15:591–602.
Sleutels, F., R. Zwart, and D. P. Barlow. 2002. The non-coding Air RNA is required for silencing autosomal imprinted genes. Nature 415:810–813.
Smilinich, N. J., C. D. Day, G. V. Fitzpatrick et al. (16 co-authors). 1999. A maternally methylated CpG island in KvLQT1 is associated with an antisense paternal transcript and loss of imprinting in Beckwith-Wiedemann syndrome. Proc. Natl. Acad. Sci. USA 96:8064–8069.
Suzuki, S., M. B. Renfree, A. J. Pask, G. Shaw, S. Kobayashi, T. Kohda, T. Kaneko-Ishino, and F. Ishino. 2005. Genomc imprinting of IGF2, p57(KIP2) and PEG1/MEST in a marsupial, the tammar wallaby. Mech. Dev. 122:213–222.
Thorvaldsen, J. L., K. L. Duran, and M. S. Bartolomei. 1998. Deletion of the H19 differentially methylated domain results in loss of imprinted expression of H19 and Igf2. Genes Dev. 12:3693–3702.
Weidman, J. R., S. K. Murphy, C. M. Nolan, F. S. Dietrich, and R. L. Jirtle. 2004. Phylogenetic footprint analysis of IGF2 in extant mammals. Genome Res. 14:1726–1732.
Wijnholds, J., K. Chowdhury, R. Wehr, and P. Gruss. 1995. Segment-specific expression of the Neuronatin gene during early hindbrain development. Dev. Biol. 171:73–84.
Wilkins, J. F., and D. Haig. 2003. What good is genomic imprinting: the function of parent-specific gene expression. Nat. Rev. Genet. 4:359–368.
Woodburne, M. O., T. H. Rich, and M. S. Springer. 2003. The evolution of tribospheny and the antiquity of mammalian clades. Mol. Phylogenet. Evol. 28:360–385.
Wylie, A. A., S. K. Murphy, T. C. Orton, and R. L. Jirtle. 2000. Novel imprinted DLK1/GTL2 domain on human chromosome 14 contains motifs that mimic those implicated in IGF2/H19 regulation. Genome Res. 10:1711–1718.
Yoon, B. J., H. Herman, A. Sikora, L. T. Smith, C. Plass, and P. D. Soloway. 2002. Regulation of DNA methylation of Rasgrf1. Nat. Genet. 30:92–96.
Yusa, K., J. Takeda, and K. Horie. 2004. Enhancement of sleeping beauty transpostion by CpG methylation: possible role of heterochromatin formation. Mol. Cell. Biol. 24:4004–4018.(Heather K. Evans*,, Jenni)