Fig. 1.

Allele-specific expression in preimplantation embryos. RTPCR was used to analyse allele-specific expression of genes within the Kcnq1 domain in C57BL/6J (B6)×Mus spretus-distal chromosome 7 (SD7) two-cell and blastocyst stage embryos. Maternal (M) and paternal (P) alleles were distinguished using RFLP or deletion/insertion polymorphisms described in the Materials and methods. For each gene, a representative sample is shown with its corresponding reverse-transcriptase negative control. (A) A schematic of the Kcnq1 domain, with the non-coding RNA Kcnq1ot1 labelled in black, ubiquitously imprinted genes labelled in green and placental-specific imprinted genes labelled in purple. DMRs are marked by black circles. (B) Kcnq1ot1 exhibits monoallelic paternal expression at the two-cell stage. (C) Kcnq1ot1 and Kcnq1 are paternally and maternally expressed, respectively; however, placental specific imprinted genes Tssc4 and Cd81 are biallelically expressed at the blastocyst stage. (D) Table showing the tissue specific imprinting of genes in the locus.

Fig. 2.

Allele-specific expression in ES and TS cells. RT-PCR was used to analyse the allele-specific expression of genes within the Kcnq1 domain in undifferentiated C57BL/6J (B6)×M.m. castaneus (Ca) ES and TS cells. Maternal (M) and paternal (P) alleles were distinguished using RFLP or deletion/insertion polymorphisms described in the Materials and methods. The non-coding RNA Kcnq1ot1 is labelled in black, ubiquitously imprinted genes are labelled in green and placental specific imprinted genes are labelled in purple. For each gene, a representative sample is shown with its corresponding reverse transcriptase-negative control. Phlda2, Cdkn1c and Kcnq1ot1 exhibit monoallelic expression, and Tssc4 and Osbpl5 show biallelic expression in both cell types. Cd81 shows expression from both alleles although there is a bias towards the maternal allele. We corrected for primer bias by densitometry and normalisation to a mix of 50% B6 cDNA and 50% Cast cDNA. The parental ratio of Cd81 in ES cells is 1:1, whereas TS cells show a ratio of 3.5:1.

Fig. 3.

Histone modifications in ES cells, TS cells and E7.5 EPC. (A) ChIP analysis of the previously mentioned ES and TS cell lines was carried out to analyse allele-specific histone modifications with antibodies against H3Ac, H3K4me2, H3K9me2 and H3K27me3. The modifications associated with active chromatin regions are marked in green (light green for ES, dark green for TS cells), while those associated with repressive chromatin are marked in red (light red for ES, dark red for TS cells). The parental alleles are distinguished by SNPs which are separated on SSCP gels. The non-coding RNA Kcnq1ot1 is labelled in black, ubiquitously imprinted genes are labelled in green and placental specific imprinted genes are labelled in purple. Phlda2, Cdkn1c and Kcnq1ot1 show an allelic bias in histone modifications (marked by arrows) and Osbpl5, Tssc4, Cd81 and Ascl2 show no bias in both cell types. Each panel is a representative example of the ChIP, beside it is a graphical representation of the ratio of the bound maternal allele to the bound paternal allele (normalised according to the input) shown for the active modifications and the ratio of paternal over maternal for repressive modification, except for Kcnq1ot1 where the ratios are reversed. (B) RT-PCR was used to analyse allele specific expression of Tssc4 in C57BL/6J (B6)×M. spretus-distal chromosome 7 (SD7) E7.5 embryos and EPCs. Maternal (M) and paternal (P) alleles were distinguished using an RFLP polymorphism described in the Materials and methods. Embryos show biallelic expression while EPCs show maternal expression. Carrier ChIP analysis was used to analyse the histone modification H3K4me2 at Tssc4 in C57BL/6J (B6) × SD7 (SD7) E7.5 EPC. The parental alleles are distinguished by a SNP which is separated on an SSCP gel. There is a clear bias in the distribution of K4me2 with the majority associated with the active maternal allele.