Nucleoporin 107, 62 and 153 mediate Kcnq1ot1 imprinted domain regulation in extraembryonic endoderm stem cells

Abstract

Genomic imprinting is a phenomenon that restricts transcription to predominantly one parental allele. How this transcriptional duality is regulated is poorly understood. Here we perform an RNA interference screen for epigenetic factors involved in paternal allelic silencing at the Kcnq1ot1 imprinted domain in mouse extraembryonic endoderm stem cells. Multiple factors are identified, including nucleoporin 107 (NUP107). To determine NUP107's role and specificity in Kcnq1ot1 imprinted domain regulation, we deplete Nup107, as well as Nup62, Nup98/96 and Nup153. Nup107, Nup62 and Nup153, but not Nup98/96 depletion, reduce Kcnq1ot1 noncoding RNA volume, displace the Kcnq1ot1 domain from the nuclear periphery, reactivate a subset of normally silent paternal alleles in the domain, alter histone modifications with concomitant changes in KMT2A, EZH2 and EHMT2 occupancy, as well as reduce cohesin interactions at the Kcnq1ot1 imprinting control region. Our results establish an important role for specific nucleoporins in mediating Kcnq1ot1 imprinted domain regulation.

Fig. 1

Nucleoporin depletion disrupted Kcnq1ot1 ncRNA expression. a Real-time Kcnq1ot1 ncRNA expression levels normalized to Gapdh (n = 3 biological samples with four technical replicates per sample). b Allelic Kcnq1ot1 ncRNA expression in control and Nup-depleted XEN cells (n = 3 biological samples; n = 4 technical replicates per sample). c Absolute allelic Kcnq1ot1 transcript abundance determined by droplet digital PCR in control and Nup-depleted XEN cells, as measured by RNA copies µg⁻¹ (n = 3 biological samples). Center lines, medians; box limits, 25th and 75th percentiles as determined by R software; whiskers, 1.5 times the interquartile range from 25th and 75th percentiles; B6/maternal, red; CAST/paternal, blue; error bars, s.e.m.; *, significance p < 0.05 compared to siNT control (t test); WT wild type, Veh vehicle, siNT nontargeting siRNA, si107 Nup107 siRNA, si62 Nup62 siRNA, si98/96 Nup98/96 siRNA, si153 Nup153 siRNA

Fig. 2

Nucleoporin depletion disrupted Kcnq1ot1 ncRNA volume and Kcnq1ot1 domain positioning at the nuclear periphery. a Representative confocal nuclear images displaying Kcnq1ot1 DNA (magenta) Kcnq1ot1 ncRNA (cyan) and DAPI staining (blue) for G1-synchronized control and Nup-depleted XEN cells (n = 4; cell number = 109–123); upper panel, DNA FISH; middle panel, RNA FISH; lower panel, merge; M maternal domain, P paternal domain; white dashed line denotes nuclear rim. In these images, red and green fluorescence was converted to magenta and cyan. b, c Percent of cells with paternal or maternal Kcnq1ot1 ncRNA signals. d Percent of cells with Kcnq1ot1 ncRNA signal volume; low, 0–0.7 μm³; medium, 0.7–1.4 μm³; high, 1.4–2.1 μm³; very high, >2.1 μm³. e Distance of the paternal and maternal Kcnq1ot1 domain from the nuclear membrane in control and Nup-depleted XEN cells. The maternal Kcnq1ot1 domain was randomly positioned within the nucleus (expected nuclear periphery (NP) 15%; subnuclear periphery (SP) 30%; nuclear interior (NI) 60%), except for Nup153-depleted cells with a Kcnq1ot1 ncRNA signal (si153 M+). For these analyses, cells with no RNA but detectable DNA FISH signals were included, while those lacking DNA signals were excluded. NP, 0–0.5 μm; SP, 0.5–1.5 μm; NI, 1.5–4 μm; error bars, s.e.m.; *, significance p < 0.05 compared to vehicle control (t test); WT wild type, Veh vehicle, siNT nontargeting siRNA, si107 Nup107 siRNA, si62 Nup62 siRNA, si98/96 Nup98/96 siRNA, si153 Nup153 siRNA, si153 M- Nup153-depleted cells without a Kcnq1ot1 ncRNA signal; scale bar, 1 μm; n = 4; cell count number = 109–123

Fig. 3

NUP107/62 and NUP153 interaction with the Kcnq1ot1 ICR, and the Osbpl5, Kcnq1, Cd81 promoters. a The Kcnq1ot1 domain with regions of analysis (arrowheads) at the Kcnq1ot1 ICR, IC3, IC4; enhancer element, E1, E2; imprinted gene promoters; Os1, Os2, Ph1, Ph2, Sl1, Sl2, Ck1, Kc1, Kc2, Ts1, Ts2, Cd1, Th1; and negative control sites, Ctrl1, Ctrl2. Quantitative ChIP analysis using b mAb414 antibodies and c NUP153 antibodies in wild-type XEN cells at regions across the domain, respectively (n = 3 biological samples with four technical replicates per sample). Quantitative allelic analysis for d mAb414 and e NUP153 in siNT- and nucleoporin-depleted XEN cells. Allelic proportions are represented as percent of the total enrichment levels (n = 3 biological samples with four technical replicates per sample). f Quantitative ChIP analysis using mAb414 antibodies was performed in control and Nup153-depleted XEN cells at sites of NUP153 enrichment (n = 3 biological samples with three technical replicates per sample). g Quantitative ChIP analysis using NUP153 antibodies was performed in control and Nup107- and Nup62-depleted cells at sites of mAb414 (NUP107/62) enrichment (n = 3 biological samples with three technical replicates per sample). Error bars, s.e.m.; *, significance p < 0.05 compared to the IgG or siNT control (t test)

Fig. 4

Nup107, Nup62, and Nup153 depletion reactivate a subset of paternal alleles at the Kcnq1ot1 domain. Absolute allelic transcript abundance of imprinted genes determined by droplet digital PCR in control and Nup-depleted XEN cells, as a measure of RNA copies µg⁻¹ (n = 3 biological samples). Center lines, medians; box limits, 25th and 75th percentiles as determined by R software; whiskers, 1.5 times the interquartile range from 25th and 75th percentiles; Mat, maternal; Pat, paternal; Pat R, reactivated paternally silent allele; *, significance p < 0.05 compared to the siNT control (t test). Error bars, s.e.m.

Fig. 5

Nup107, Nup62, and Nup153 depletion disrupts RNAPII enrichment and histone modifications at the Kcnq1ot1 ICR and imprinted gene promoters. a RNAPII and H3K4me3 ChIP at the maternal and paternal Kcnq1ot1 ICR and imprinted gene promoters in control and Nup-depleted XEN cells (n = 3; n = 3 technical replicates). b H3K9me2 and H3K27me3 ChIP at the maternal and paternal Kcnq1ot1 ICR and imprinted gene promoters in control and Nup-depleted XEN cells (n = 3 biological samples with three technical replicates per sample). The y-axis indicates total and allelic ChIP enrichment levels represented as percent of input. Allelic proportions are represented as a percent of the total ChIP enrichment level. Error bars, s.e.m.; *, significance p < 0.05 of paternal/maternal levels compared to the siNT paternal/maternal control (t test)

Fig. 6

Nup107, Nup62, and Nup153 depletion disrupts KMT2A, EHMT2, and EZH2 occupancy at the Kcnq1ot1 ICR and imprinted gene promoters. a KMT2A, b EHMT2 and c EZH2 ChIP at the maternal and paternal Kcnq1ot1 ICR and imprinted gene promoters in control and Nup-depleted XEN cells (n = 3 biological samples with three technical replicates per sample). The y-axis indicates total ChIP allelic enrichment levels represented as percent of input. Allelic proportions are represented as a percent of the total ChIP enrichment level. Error bars, s.e.m.; *, significance p < 0.05 compared to the siNT control (t test)

Fig. 7

SMC1A and SMC3 enrichment at the paternal Kcnq1ot1 ICR reduced upon nucleoporin depletion. Quantitative ChIP analysis and allelic analysis for a SMC1A and b SMC3 at positive mAb414 and NUP153 enrichment sites in control and Nup107/Nup62-double, Nup98/96- (IC3, IC4 only) and Nup153-depleted XEN cells (n = 3 biological samples with three technical replicates per sample). Allelic proportions are represented as a percent of the total enrichment level. Error bars, s.e.m.; *, significance p < 0.05 compared to the siNT control (t test)

Fig. 8

NUP107 is required for NUP153 and NUP62 interaction with CTCF, SMC1A, and SMC3. a NUP-immunoprecipitation with NUP153 and mAb414 antibodies was performed on control (siNT) and Nup107/62- and Nup153-depleted XEN nuclear lysates, following which western analysis was conducted using CTCF, SMC1A, and SMC3 antibodies (n = 3 biological replicates). b NUP-immunoprecipitation with NUP153 antibodies was performed on Nup107- and Nup62-depleted XEN lysates, followed by western analysis using CTCF, SMC1A, and SMC3 antibodies (n = 3 biological replicates)