The imprinted retrotransposon-like gene PEG11 (RTL1) is expressed as a full-length protein in skeletal muscle from Callipyge sheep

Abstract

Members of the Ty3-Gypsy retrotransposon family are rare in mammalian genomes despite their abundance in invertebrates and some vertebrates. These elements contain a gag-pol-like structure characteristic of retroviruses but have lost their ability to retrotranspose into the mammalian genome and are thought to be inactive relics of ancient retrotransposition events. One of these retrotransposon-like elements, PEG11 (also called RTL1) is located at the distal end of ovine chromosome 18 within an imprinted gene cluster that is highly conserved in placental mammals. The region contains several conserved imprinted genes including BEGAIN, DLK1, DAT, GTL2 (MEG3), PEG11 (RTL1), PEG11as, MEG8, MIRG and DIO3. An intergenic point mutation between DLK1 and GTL2 causes muscle hypertrophy in callipyge sheep and is associated with large changes in expression of the genes linked in cis between DLK1 and MEG8. It has been suggested that over-expression of DLK1 is the effector of the callipyge phenotype; however, PEG11 gene expression is also strongly correlated with the emergence of the muscling phenotype as a function of genotype, muscle type and developmental stage. To date, there has been no direct evidence that PEG11 encodes a protein, especially as its anti-sense transcript (PEG11as) contains six miRNA that cause cleavage of the PEG11 transcript. Using immunological and mass spectrometry approaches we have directly identified the full-length PEG11 protein from postnatal nuclear preparations of callipyge skeletal muscle and conclude that its over-expression may be involved in inducing muscle hypertrophy. The developmental expression pattern of the PEG11 gene is consistent with the callipyge mutation causing recapitulation of the normal fetal-like gene expression program during postnatal development. Analysis of the PEG11 sequence indicates strong conservation of the regions encoding the antisense microRNA and in at least two cases these correspond with structural or functional domains of the protein suggesting co-evolution of the sense and antisense genes.

Figure 1. Diagrammatic representation of the organization of imprinted genes located at the telomeric end of ovine chromosome 18.

(A) Representation of the approximate 1 Mbp region from BEGAIN to DIO3. The core imprinted genes affected by the callipyge mutation are colored while imprinted genes unaffected by the mutation are shown in grey. Paternally expressed genes are shaded blue and maternally expressed genes are shaded pink. The direction of transcription of each gene is indicated by the arrow in the gene symbol. Introns are not shown. The red asterisk denotes the position of the callipyge point mutation (CLPG). The precise lengths of the maternally expressed genes, which all produce non-coding RNAs, are unclear. The diagram is based on that deduced by supplemented with annotation for a miRNA cluster (MIRG) deduced by comparative sequence analyses with the orthologous murine and human sequence regions. (B) Representations of the PEG11 and PEG11as genes. A large black arrow denotes the direction of transcription of each gene. Small arrows show the relative positions of PCR primers. The region of PEG11 expressed as a recombinant protein (rPEG11) is also shown. The precise length of the PEG11as gene is unclear but it extends beyond the PEG11 gene in both directions (represented by broken lines).

Figure 2. Relative expression levels of PEG11/PEG11as mRNA in wild type (NN) and callipyge (NC^pat) genotypes during skeletal muscle developmental.

Expression levels were measured by qRT-PCR and normalised to RPLPO. (A) PEG11/PEG11as mRNA expression in SM muscle taken at 12 weeks (230 days post-fertilisation) of age from four wild-type (NN) (grey) and four callipyge paternal heterozygote (NC^pat) individuals (black). The mean expression values of the four individuals representing each genotype are illustrated in the inset. The qRT-PCR assay measured expression of both PEG11 and PEG11as as the former is wholly contained within the latter. However, the contribution of PEG11as is relatively minor in these genotypes and thus the assay primarily measures expression of PEG11 (see Materials and Methods). (B) PEG11/PEG11as expression in LD muscle during muscle development as a function of genotype. The samples were taken at 80, 100, 120, 150 and 230 days of development. Birth is at 147 days. The error bars denote the standard error of mean (n = 4). The asterisks denote significant (P<0.001) differences in expression levels between the NC^pat (black) and NN (grey) genotypes at each developmental stage.

Figure 3. Immunoblot analysis for PEG11 protein in nuclear fractions from ovine SM skeletal muscle samples obtained from different individuals at 12 weeks of age.

Nuclear extract (100 µg) from each muscle sample was separated by SDS-PAGE, transferred to nitrocellulose and probed with immunoaffinity-purified rabbit antibody raised to recombinant ovine PEG11 (rPEG11). Lanes 1–3, nuclear protein fractions from SM muscle obtained from three wild type (NN) individuals; lanes 4 and 5, two callipyge (NC^pat) samples; lane S, protein size standards. Proteins designated A (∼Mr = 146,000 Da) and B (∼Mr = 190,000 Da) were identified by mass spectrometry analysis as PEG11 and myosin heavy chain, respectively.

Figure 4. Mass spectrometry analyses of two peptides derived from the tryptic digest of protein contained within the PEG11 immuno-reactive region of the SDS-PAGE gel.

Seven unique peptides were identified by using two parallel mass spectrometry approaches. Example product ion ms/ms spectra for peptides identified by each technique are shown. (A) LC-MALDI-MS/MS, and; (B) LC-ES-MS/MS. For each peptide, the amino acid sequence is tabulated along with the theoretical m/z values corresponding to the y and b ions. The experimentally detected sequence ions are denoted by bold typeface and grey shading (insets). A selection of the sequence ions is labelled in the ms/ms spectra.

Figure 5. Sequence coverage of PEG11 peptides.

The peptides sequenced by LC-MALDI-MS/MS and ESI-LC-MS/MS are bolded and underlined. The combined peptide sequences represented 8.2% of the ovine PEG11 sequence. The putative N-terminus was deduced from the longest open reading frame of the mRNA sequence in conjunction with the marked DNA sequence conservation transition occurring 5′ to the putative initiating methionine.

Figure 6. Conserved regions of the ovine PEG11 gene correspond with antisense miRNA.

The PEG11 open reading frame is shown in blue while the locations of antisense miRNA (mir-431, mir-433, mir-127, mir-432 and mir-136) are shown in red. The black arrow denotes the direction of transcription of PEG11. The graph shows the extent of PEG11 gene conservation using the UCSC PhastCons Conserved Elements 17-way Vertebrate Multiz Alignment and Conservation tool .

Figure 7. Structural domains in the PEG11 protein.

(A) Probability of coiled coil formation. All sequences were extracted from the UCSC genome browser (http://genome.ucsc.edu/). Coiled coils were predicted using Coils and a 14 amino acid window . The relative positions in the PEG11 protein corresponding with regions encoding antisense miRNA are shown in red. (B) Structural domains common to mammalian PEG11 proteins. The diagram is based on the ovine sequence but is representative of mammalian sequences. Structural domains were identified by searching the Pfam protein families database . PF03732 (capsid-like or retrotransposon gag protein domain; Expect score = 3.4e−11); PF00077 (retroviral aspartyl protease domain; Expect score = 0.33); PF00098 (zinc finger knuckle; Expect score not significant); PF00078 (reverse transcriptase domain; Expect score  = 1.4e−5); PF00075 (RNaseH; Expect score not significant) domain; PF00665 (integrase core domain; Expect score not significant). Mutations to key residues in some of the PEG11 retrotransposon domains have compromised the significance of identification of some Pfam families, which are often biased in their weightings of catalytic residues. The positions corresponding to regions encoding antisense miRNA are shown in red. The green triangles denote positions where there are large insertions of repetitive sequence in the murine and rat protein sequences. The sequence insertions at each site for these two species are not conserved in sequence or length.