An imprinted transcript, antisense to Nesp, adds complexity to the cluster of imprinted genes at the mouse Gnas locus

Abstract

The Gnas locus in distal mouse chromosome (Chr) 2 is emerging as a complex genomic region. It contains three imprinted genes in the order Nesp-Gnasxl-Gnas. Gnas encodes a G protein α-subunit, and Nesp and Gnasxl encode proteins of unknown function expressed in neuroendocrine tissue. Together, these genes form a single transcription unit because transcripts of Nesp and Gnasxl are alternatively spliced onto exon 2 of Gnas. Nesp and Gnasxl are expressed from opposite parental alleles, with Nesp encoding a maternal-specific transcript and Gnasxl encoding a paternal-specific transcript. We now identify a further imprinted transcript in this cluster. Reverse transcription–PCR analysis of Nesp expression in 15.5-days-postcoitum embryos carrying only maternal or paternal copies of distal Chr 2 revealed an isoform that is exclusively paternally, rather than maternally, expressed. Strand-specific reverse transcription–PCR showed that this form is an antisense transcript. The existence of a paternally expressed antisense transcript was confirmed by Northern blot analysis. The sequence is contiguous with genomic sequence downstream of Nesp and encompasses Nesp exons 1 and 2 and an intervening intron. We propose that Nespas is an additional control element in the imprinting region of mouse distal Chr 2; it adds further complexity to the Gnas-imprinted gene cluster.

Genomic imprinting is a phenomenon whereby genes are differentially expressed according to parental origin (1). Most imprinted genes in the mouse are located within nine imprinting regions distributed across six autosomes [C.V.B. and B. M. Cattanach (Medical Research Council Mammalian Genetics Unit, Harwell, Oxfordshire, U.K.), World Wide Web Site: Genetic and Physical Imprinting Map of the Mouse; http://www.mgu.har.mrc.ac.uk/anomaly/anomaly.html]. One of the first described imprinting regions was distal chromosome (Chr) 2 (2). Mice with two maternal copies of the region (MatDp.dist2) are hypoactive; they have long, flat-sided bodies and die within a few hours of birth. By contrast, mice with two paternal copies of the region (PatDp.dist2) have an opposite phenotype because they are hyperactive; they are also edematous, have short, square bodies, and survive for several days after birth (2, 3). It was shown from genetic studies that the phenotypes must be due to at least two imprinted genes, one of which is maternally imprinted and the other which is paternally imprinted (4).

Using representational difference analysis, based on parent-of-origin methylation differences, we recently have identified two oppositely imprinted transcripts, Nesp and Gnasxl, at the Gnas locus in distal Chr 2 that are candidates for the imprinting phenotypes (5, 6). Nesp is paternally imprinted/maternally expressed, and Gnasxl is maternally imprinted/paternally expressed. Both determine proteins found in neuroendocrine tissues although their functions are unknown (7, 8). Remarkably, Nesp, Gnasxl, and Gnas are all part of the same transcription unit, as Nesp and Gnasxl transcripts splice onto Gnas exon 2 (5). The human homologues, NESP55 and XLαs, have been shown to be imprinted and in a similar manner to the mouse (refs. 9 and 10, respectively). The Gnas/GNAS1 locus is the first example of which we are aware of a cluster of imprinted genes in which two oppositely imprinted transcripts share the same exons.

The order of genes in the Gnas cluster is Nesp-Gnasxl-Gnas with 15 kb separating Nesp and Gnasxl, whereas Gnasxl lies 30 kb upstream of Gnas (5, 6). Nesp is associated with a 2.8-kb region of paternal methylation, and Gnasxl is associated with a 5.5-kb region of maternal methylation that extends 3.3 kb upstream of the extra-large exon. There was no evidence of parental-specific methylation (5) associated with a Gnas promoter (11) despite the existence of good biochemical and clinical evidence that Gnas/GNAS1 shows maternal-specific expression in a subset of tissues (12, 13).

Parent-specific methylation is the simplest explanation for monoallelic expression of Nesp and Gnasxl because both genes carry methylation marks, with the expressed allele being unmethylated. An expression competition model (14, 15) in which methylation regulates the availability in cis of shared regulatory elements also could account for the opposite imprinting of Nesp and Gnasxl. Antisense transcripts can act as regulatory elements. We describe a maternally imprinted antisense transcript of Nesp that, we predict, regulates expression of genes within the Gnas cluster. This imprinted antisense transcript adds further complexity to the cluster of imprinted genes in the distal imprinting region of mouse Chr 2.

Methods

Distal Chr 2-Duplication/Deficient Mice.

Mice with maternal duplication of distal Chr 2 (MatDp.dist2) and the reciprocal paternal duplication (PatDp.dist2) were generated by intercrossing heterozygotes for the reciprocal translocation T(2;8)26H (16). The duplication offspring were identified by typing for the marker D2Mit226 (17). These mice and wild-type siblings were used for expression analysis.

Reverse Transcription–PCR (RT-PCR) Analysis.

For RT-PCR analysis, approximately 1 μg of poly(A)⁺ RNA, isolated by using the FastTrack 2.0 mRNA isolation kit (Invitrogen), was reverse-transcribed by mouse murine leukemia virus reverse transcriptase (Life Technologies, Gaithersburg, MD) using oligo(dT)₁₅ primer (Promega). Conditions for PCR were 25 cycles of 1 min each at 94°C, 55°C, and 72°C by using Thermoprime Plus DNA polymerase (Advanced Biotechnologies, Columbia, MD). The positions of the primers NL3 (5′-AGTGGAGGCACCTCTCGGA-3′, nucleotides 85–103, GenBank accession no. AF125315), NR3 (5′-CTCTGGCTCTGCAGAGAGT-3′, nucleotides 354–372, accession no. AF125315), and R7 (5′-TTAGTGACGCCGGATGGGGA-3′, nucleotides 997–978, accession no. AF175305) are shown in Fig. 2c. 5′ Rapid amplification of cDNA ends (RACE)-PCRs were performed on poly(A)⁺ RNA from 15.5-days-postcoitum (dpc) PatDp.dist2 embryos by using the SMART RACE kit (CLONTECH). PCR products were subcloned by using the Invitrogen TA cloning kit.

Figure 2.

Identification of an imprinted antisense transcript for Nesp. (A) RT-PCR demonstrating the differential expression of Nesp transcripts in 15.5-dpc embryos. Oligo(dT)-primed cDNA was derived from 15.5-dpc PatDp.dist2 (P), MatDp.dist2 (M), and wild-type sib (wt) embryos. Twenty-five cycles of amplification were performed by using the primers NL3 and NR3. The arrows indicate the Nesp 193-bp spliced and 288-bp unspliced PCR products. RT reactions were carried out in the presence (+) and absence (−) of reverse transcriptase. (B) Southern hybridization of strand-specific RT-PCR products demonstrating that the paternally expressed unspliced isoform lies antisense to Nesp. The first lane is a control RT, using oligo(dT) to show the presence of a product, and the third lane is a negative control for RT without exogenous primer to show the absence of priming by endogenous factors in the RNA preparations. Primer NL3 is specific for the antisense transcript, and R7 is specific for the sense strand. PCR was performed by using primers F1 and R7, which lie 3′ of the intron in Nesp exon 2; 35 cycles were required to detect the less abundant paternal transcript in PatDp.dist2, and 25 cycles were required for the more abundant maternal transcript in MatDp.dist2. The probe was a 229-bp PCR product derived from cDNA extending from primer F2 to exon 2 of Gnas. (C) Diagram showing the approximate position of the primers on genomic sequence (accession nos. AJ245401 and AJ251480). The reverse transcriptase primers (half-arrowheads) specific for the antisense and sense are shown above and below the map, respectively; PCR primers are designated by normal arrows; primer sequence is given in Methods. The approximate position of the IMAGE clone is shown together with PCR products from transcript walking by RT-PCR. The 296-bp product was isolated by RACE. The figure is not to scale.

Strand-Specific RT-PCR.

Poly(A)⁺ RNA, isolated as described above, was treated with RNase-free DNase I by using the Message Clean kit (GenHunter Corporation, Nashville, TN). Each sample was set up in duplicate for +reverse transcriptase and −reverse transcriptase reactions. Strand-specific primers and 1 μg of oligo(dA) [(dA)₈₀; Genosys, The Woodlands, TX] were added to 0.15 μg of poly(A)⁺ RNA, and the mixture was heated at 70°C for 10 min. The (dA)₈₀ oligonucleotide was added to all samples, except those with oligo(dT)₁₅ primer, to trap any oligo(dT) that might have copurified with the poly(A)⁺ RNA (18). First-strand cDNA was synthesized at 50°C for 50 min by using either sense or antisense primers with Superscript II (200 units; Life Technologies). The enzyme was inactivated at 80°C for 45 min. First-strand cDNA was amplified by PCR as described above. The relative position of the reverse transcriptase primers, R7 and NL3 (specific for the sense and antisense transcripts, respectively), and primers for subsequent PCR, F1 (5′-ACCAGTCACTCACTCAGCGT-3′, nucleotides 711–730, accession no. AF175305) and R7, are shown in Fig. 2c. All PCR products were probed with a 229-bp PCR product derived from cDNA extending from F2 (5′-CAAGGAGGAAAACAGGCAGC-3′, nucleotides 883–902, accession no. AF175305) to exon 2 of Gnas (5′-CTCCGTTAAACCCATTAACATGCA-3′, nucleotides 205–182; ref. 19); the primers are shown in Fig. 2c.

Southern Hybridization.

DNA was transferred onto charged nylon membranes (Hybond N⁺; Amersham Pharmacia) by alkaline transfer. PCR products were radiolabeled with 25 μCi of [α-³²P]dCTP (NEN) by using Megaprime (Amersham Pharmacia). The Southern filters were hybridized by using Church and Gilbert hybridization buffers (20).

Northern Blot Analysis.

Poly(A)⁺ RNA was treated with DNase I, and Northern blots were prepared by using the NorthernMax–Gly kit (Ambion, Austin, TX). Riboprobes were made by using the Strip-EZ RNA labeling kit (Ambion) and [α-³²P]UTP (Amersham Pharmacia). Sense and antisense riboprobes for Nesp were prepared as described previously (5) by transcription with T3 polymerase from clones 317 and 330 (opposite orientation), which contain a 423-bp genomic fragment extending from exon 1 to exon 2 of Nesp. The actin riboprobe was generated from pTRI-β-actin-mouse template DNA (Ambion). Hybridization was done overnight by using ULTRAhyb (Ambion), and blots were washed by using Ambion's Low and High Stringency Wash Buffers. The CLONTECH Northern was hybridized under standard Express Hyb conditions (CLONTECH).

Sequencing.

A genomic phage lambda clone for the Nesp locus was isolated from a library of 129/SvJ mouse DNA in Lambda FIX II (Stratagene; library no. 946309) by hybridization with the CpG island clone M1/1 (ref. 6; RZPD clone EDIUp123NO611Q4 at http://www.rzpd.de). A 14.5-kb XhoI fragment was subcloned into the SalI site of pDELTA 2 (Life Technologies), and a series of nested deletions was prepared according to the manufacturer's instructions. Sequencing was done with SP6 and T7 promoter primers as described below, and sequences were assembled by using the gap4 program of the staden package. The ABI Prism dichlororhodamine Terminator Cycle Sequencing Ready Reaction kit (Applied Biosystems) was used for sequencing. All sequencing products were electrophoresed on an ABI 377 (Perkin–Elmer) automated sequencer. Sequences were analyzed for similarity by using the blast program accessed at http://www.ncbi.nlm.nih.gov/blast.

Results

Identification of a Paternally Expressed Nesp Transcript.

Nesp had been identified as a candidate imprinted gene by the isolation of a differentially methylated HpaII fragment with a spliced maternally expressed transcript that lacks a 95-bp intron present in genomic DNA (5, 6). The intron is upstream of the Nesp ORF. Sequence alignments showed that bovine NESP55 cDNA (accession no. U77614; ref. 7) and human NESP cDNA (accession no. AJ009849; ref 9) were unspliced forms (Fig. 1). Therefore, to determine whether there were unspliced Nesp isoforms in the mouse, the oligonucleotide primers NL3 and NR3 were designed across the 95-bp intron (Fig. 2) and RT-PCR analysis was carried out on whole 15.5-dpc embryos. As expected, a 193-bp spliced PCR product was observed in the MatDp.dist2 (lane 5) and wild-type sib cDNA (lane 3) but not in the PatDp.dist2 cDNA (Fig. 2a, lane 1). In addition, a 288-bp unspliced RT-PCR product of lower intensity and exclusive paternal expression was observed. This band was seen in the PatDp.dist2 (Fig. 2a, lane 1) and wild-type sib cDNA (lane 3) but not in the MatDp.dist2 cDNA (lane 5). Genomic DNA contamination in the RNA samples, which could account for the 288-bp unspliced form, was ruled out by the absence of the product in the controls without reverse transcriptase. Sequencing of the MatDp.dist2 and PatDp.dist2 RT-PCR products confirmed that the 95-bp intron is absent in the 193-bp maternal RT-PCR product but present in the 288-bp paternal RT-PCR product. Similar results were obtained by using newborn MatDp.dist2 and PatDp.dist2 tissues obtained using another Chr 2 translocation, T(2;19)68H (refs. 5 and 21; data not shown).

Figure 1.

Multiple alignment of mouse 95-bp intron and flanking Nesp sequence with human and bovine sequence. The intron is upstream of the Nesp ORF. Mat, maternally expressed mouse transcript (accession no. AF175305); Mus, mouse genomic (accession no. AF125315); Pat, paternally expressed mouse transcript (accession no. AF173359); Hum, human sequence (accession no. AJ009849; ref 9); and Bov, bovine mRNA (accession no. U77614; ref. 7). Bases in common between the mouse paternal transcript and at least one other species are shaded. Putative splice donor and acceptor sites are underlined in the mouse genomic sequence (Mus) and are not present in the antisense orientation with respect to Nesp.

Sequence alignment of the 1,083-bp paternally expressed transcript (derived from primers NL3 to R7) with bovine NESP55 cDNA (accession no. U77614) and human NESP cDNA (accession no. AJ009849) showed that there is sequence conservation within the region of the 95-bp intron (Fig. 1). Apart from the 95-bp intron, the sequence of the paternally expressed transcript (accession no. AF173359) matched with exons 1 and 2 of the maternally expressed Nesp transcript (accession no. AF175305; alignment not shown). A mouse IMAGE clone 932324 (accession no. AI561892) contains the 95-bp intron and represents a sequence apparently transcribed in the opposite orientation with respect to Nesp (Fig. 2c). Analysis of the IMAGE clone sequence (accession nos. AI561892 and AA530580) revealed that the transcript extends at least 198 bp downstream of Nesp exon 2, and this is contiguous with the genomic sequence (AJ245401). In the reverse orientation there are no consensus donor and acceptor splice sites around the 95-bp intronic sequence, thus suggesting the paternally expressed isoform may be derived from an antisense transcript.

Orientation of the Paternally Expressed Transcript with Respect to the Maternally Expressed Nesp Transcript.

Strand-specific RT-PCR was performed on RNA derived from PatDp.dist2 15.5-dpc embryos to determine whether the paternally expressed isoform was derived from a sense or antisense transcript. MatDp.dist2 material was also tested to confirm the strand specificity of the primers. Poly(A)⁺ RNA was reverse-transcribed by using primers specific for each strand; primer NL3 designed from the 5′ end of Nesp (Fig. 2c) was specific for the antisense strand, and primer R7, designed from the 3′ end of Nesp, was specific for sense. The strand-specific cDNA was amplified with primers F1 and R7, designed from the 3′ end of Nesp as shown in Fig. 2c. In PatDp.dist2 embryos that express the unspliced form of Nesp, the expected PCR product of 287 bp was detected in the oligo(dT) and NL3-primed cDNA as shown in Fig. 2b (lanes 1 and 5, respectively). No amplification was detected in the sample with oligo(dA) alone and in the R7-primed PatDp.dist2 cDNA (Fig. 2b, lanes 3 and 7, respectively). These results show that the paternally expressed transcript is obtained when an antisense-specific primer is used, indicating that it is derived from an antisense transcript. The opposite was observed with MatDp.dist2 embryos that express the spliced form of Nesp; a PCR product was detected in the oligo(dT) and R7-primed cDNA as shown in Fig. 2b (lanes 1 and 7, respectively) but no amplification was detected in the sample with oligo(dA) alone (Fig. 2b, lane 3) and the NL3-primed MatDp.dist2 cDNA (lane 5). This shows that the maternally expressed Nesp transcript is found only when a sense-specific primer is used, indicating that this product is derived from a sense transcript. Similar PCR results were obtained by using cDNA reverse transcribed with other sense and antisense primers across the Nesp exons (data not shown). Because the antisense transcript overlaps exons 1 and 2 of Nesp, we have designated this transcript Nespas (Nesp antisense).

Primers were designed from genomic sequence between Nesp and Gnasxl (accession no. AJ245401) to allow walking along the antisense transcript by strand-specific RT-PCR. PCR products from NL3 strand-specific cDNA were generated by primers F2 (Fig. 2c, nucleotides 2374–2393 of accession no. AJ245401) and R3 (nucleotides 2994–2975) and by F3 (nucleotides 2975–2994) and R4 (nucleotides 3802–3783) to give products of 621 bp and 828 bp, respectively, which were sequenced. These products extended the sequence of Nespas beyond the IMAGE clone by 1.4 kb. Although attempts to define the transcriptional start site by RACE were unsuccessful, sequence analysis of 5′ RACE products generated by the gene-specific primer F4 (nucleotides 3783–3806) and the RACE primer gave a further 296 bases of sequence, thus providing a total of 2.2 kb of sequence extending from primer NL3 for Nespas, and this is contiguous with genomic sequence (accession no. AJ245401 and AJ251480) and appears to lack coding potential. Genomic sequence (AJ251480 and AJ245856) further 5′ to Nespas gave an excellent match on the complementary strand to Aa955517/Aa955518.em_est5 (rat cDNA, source undefined, 89% over 240 bp by bestfit) and AI047184.em_est6 (mouse ES cell cDNA, 99% over 494 bp by bestfit), and work is in progress to determine whether they are part of the paternally expressed Nespas transcript.

Expression Pattern of Nespas by Northern Blot Analysis.

Northern blot analyses were performed to confirm the existence of an antisense transcript to Nesp by a nonamplifying method. A riboprobe extending from exon 1 into exon 2 of Nesp, specific for antisense, revealed a smear and bands in most mouse tissues on a Northern blot (CLONTECH). These were strongly expressed in heart (Fig. 3a), consistent with the isolation of IMAGE clone 932324 from a heart library. The Nespas-specific riboprobe recognized both a smear and a 4.4-kb band in the heart of PatDp.dist2 and wild type but not in MatDp.dist2, confirming the paternal-specific expression of antisense (Fig. 3b). That the antisense appeared as a smear on a Northern blot suggests that it could be either an unusually large RNA that is degraded in preparation using standard methods or a collection of differently sized RNAs. It therefore is possible that the antisense bands detected in Fig. 3 a and b are nothing more than artifacts resulting from the presence of ribosomal RNA bands that act to concentrate an RNA smear above and below the 28S and 18S bands. However, the band in skeletal muscle (Fig. 3a) is probably a genuine transcript because there is no evidence of a smear in this tissue. The Northern blot in Fig. 3b showed an inverse correlation between the expression of the sense and antisense transcripts. A clear dosage effect was seen with MatDp.dist2, which expressed a double dose of the sense transcript (Nesp) and no antisense transcript (Nespas) whereas with PatDp.dist2 there was no sense transcript but enhanced expression of antisense. In wild type, expected to have one dose of sense and one of antisense, there was intermediate expression. The detection of both Nesp and Nespas transcripts from opposite parental alleles in heart supports a proposal that antisense controls expression of the sense transcript from the paternal allele.

Figure 3.

Northern blot analysis of Nespas expression. (A) CLONTECH multiple tissue Northern blot from adult mouse probed with Nespas-specific riboprobe. The position of the RNAs in the 0.24- to 9.5-kb RNA ladder (Life Technologies) is shown to the left of the blot. (B) Northern blot of approximately 2 μg of poly(A)⁺ RNA extracted from the heart of newborn PatDp.dist2 (P), MatDp.dist2 (M), and wild-type (wt) offspring from the T26H intercross. The Nespas-specific riboprobe recognized a smear and a 4.4-kb band in PatDp.dist2 and wt (overnight exposure) whereas the Nesp-specific riboprobe hybridized to a discrete, 3.9-kb transcript in MatDp.dist2 (4-h exposure). Actin was included as a loading control (30-min exposure).

Discussion

Previous results have shown that the imprinting at the Gnas/GNAS1 locus in mice and humans is complex (5, 9, 10). Three genes, Nesp, Gnasxl, and Gnas, were found to be part of the same transcription unit, and two of these, Nesp and Gnasxl, show exclusive monoallelic expression. For Nesp, only the maternally derived allele is expressed, and Gnasxl expression is just from the paternally derived allele (5, 9, 10). The results presented here now indicate that the situation is even more complex; there is expression of Nesp antisense from the paternally derived allele.

Six imprinted genes with antisense transcripts are now known (22–29). For two of these genes, Igf2 and ZNF127/Zfp127, both the sense and antisense transcripts are expressed from the paternal allele (23, 26, 27), but for the remaining four, UBE3A, Igf2r, KvLQT1/Kvlqt1, and now Nesp, the antisense transcript is expressed from the opposite allele to the sense, protein-encoding transcript (5, 23–25, 28, 29). Furthermore, these four genes are paternally imprinted/maternally expressed and their antisense transcripts are maternally imprinted/paternally expressed. These four would conform to the “expression competition” model of genomic imprinting whereby expression of the antisense transcript from the paternal allele represses the expression of the sense transcript from the same allele (14, 15). For the other two genes, Igf2 (23) and ZNF127/Zfp127 (26, 27), in which the overlapping sense and antisense transcripts are expressed from the paternal allele, the regulation of imprinted gene expression is likely to require a different mechanism (29).

The finding of an antisense transcript of Nesp has implications for the regulation of the Gnas cluster. If Nespas is postulated to repress the expression of Nesp in cis from the paternal allele, Nespas can be regarded as an “imprintor” and Nesp as the imprinted target. It could lead to nonexpression of the sense transcript by any of the currently proposed methods: occlusion of the sense promoter, inactivation of the paternal allele by localized heterochromatinization, and competition for shared transcription factors or enhancers by the sense and antisense promoters (15, 30). One noted feature has been that the Nespas transcript appears to be less abundant than the Nesp transcript. The Nespas transcript therefore either may be less stable or only weakly expressed but, even so, still able to regulate the silencing of the sense Nesp transcript from the paternal allele.

Although the position of the 5′ end of Nespas is not yet defined, it must lie downstream of the Nesp promoter. The differentially methylated region that is downstream of Nesp and upstream of Gnasxl is a candidate for the 5′ end of Nespas and its promoter; it is also the candidate region for the Gnasxl promoter (Fig. 4). From this region, which is unmethylated on the paternal allele, there may be bidirectional transcription from this allele to lead to the production of both Gnasxl and Nespas transcripts. The situation is different for the maternal allele. The region is maternally methylated, and this can account for nonexpression of Gnasxl and Nespas from the maternally derived chromosome. An expression competition model in which methylation regulates the availability of shared regulatory elements (14, 15) also could account for the expression of Nesp and lack of expression of Gnasxl from the maternal allele.

Figure 4.

Summary of oppositely imprinted transcripts at the Gnas locus in distal mouse Chr 2. Gene order was determined previously (5, 6); the orientation with respect to centromere and telomere is not known. Shaded blocks show exons exclusive to Nesp, the hatched block represents the exon specific for Gnasxl, and the solid block represents Gnas exon 2. Mat and Pat refer to maternally and paternally derived alleles, respectively. The maternally and paternally methylated regions represent the approximate positions of differentially methylated HpaII sites identified by methylation-sensitive representational difference analysis (6). The maternally expressed Nesp transcript that extends into Gnas exon 2, the paternally expressed Gnasxl transcript that also extends into exon 2, and the paternally expressed transcript that lies antisense to Nesp are represented by arrows showing the direction of transcription; the Gnas transcripts are not shown. The figure is not to scale.

The function of the protein products of Nesp and Gnasxl are not yet known. NESP55, however, is a neuroendocrine secretory protein originally identified in bovine chromaffin cells and resembles the chromogranin-like polypeptides (7). A more extensive examination of NESP55 mRNA in rat brain revealed a significant overlap with noradrenergic, adrenergic, and serotonergic transmitter systems (31). The product of Gnasxl has conserved regions for guanine nucleotide binding and is tightly membrane-associated at the trans-Golgi network. It therefore may function in secretory vesicle formation (8). Both Nesp and Gnasxl are expressed in neuroendocrine tissue (7, 8) and may function in a common pathway in which it is important that their expression is mutually exclusive and monoallelic. Nespas is a candidate gene for the imprinting phenotypes associated with PatDp.dist2 and MatDp.dist2 (2, 3). PatDp.dist2 will have a double dose of the Nespas transcript whereas MatDp.dist2 will lack Nespas transcripts.

Abbreviations

Chr

chromosome

dpc

day postcoitum

RT-PCR

reverse transcription–PCR

RACE

rapid amplification of cDNA ends