Differential in vivo binding dynamics of somatic and oocyte-specific linker histones in oocytes and during ES cell nuclear transfer

Abstract

The embryonic genome is formed by fusion of a maternal and a paternal genome. To accommodate the resulting diploid genome in the fertilized oocyte dramatic global genome reorganizations must occur. The higher order structure of chromatin in vivo is critically dependent on architectural chromatin proteins, with the family of linker histone proteins among the most critical structural determinants. Although somatic cells contain numerous linker histone variants, only one, H1FOO, is present in mouse oocytes. Upon fertilization H1FOO rapidly populates the introduced paternal genome and replaces sperm-specific histone-like proteins. The same dynamic replacement occurs upon introduction of a nucleus during somatic cell nuclear transfer. To understand the molecular basis of this dynamic histone replacement process, we compared the localization and binding dynamics of somatic H1 and oocyte-specific H1FOO and identified the molecular determinants of binding to either oocyte or somatic chromatin in living cells. We find that although both histones associate readily with chromatin in nuclei of somatic cells, only H1FOO is capable of correct chromatin association in the germinal vesicle stage oocyte nuclei. This specificity is generated by the N-terminal and globular domains of H1FOO. Measurement of in vivo binding properties of the H1 variants suggest that H1FOO binds chromatin more tightly than somatic linker histones. We provide evidence that both the binding properties of linker histones as well as additional, active processes contribute to the replacement of somatic histones with H1FOO during nuclear transfer. These results provide the first mechanistic insights into the crucial step of linker histone replacement as it occurs during fertilization and somatic cell nuclear transfer.

Figure 1.

Exchange of somatic linker histones with oocyte-specific linker histones following transfer of R1 ES cell nuclei. (A-C) Cloned constructs were prepared as previously described (Gao et al., 2003, 2004). R1 ES cell nuclei were injected into enucleated MII-stage oocytes and then cultured as previously described (Gao et al., 2003, 2004) for the indicated time points. (A and B) Exchange of somatic linker histones with H1FOO in chromatin of injected R1 ES cell nuclei. Constructs were fixed and imaged for DNA content, and either oocyte-specific linker histone (A) or somatic linker histone (B) content as described (Gao et al., 2004). (C) H1F0-GFP expressed from a chromosomal locus in a R1 ES cell nucleus is removed from R1 chromatin with a kinetic similar to that of the endogenous somatic H1. To monitor fluorescence loss by bleaching a second nucleus was placed in the perivitelline space next to the ooplasm for comparison. Arrow indicates the injected nucleus; arrow head indicates the nucleus placed in the perivitelline space. Scale bars, 20 μm.

Figure 2.

Representative examples of the subnuclear distribution of H1F0-GFP and H1FOOα-GFP in somatic cells and oocytes. (A and C) Distribution of H1F0-GFP and H1FOOα-GFP in somatic 3134 nuclei. Both H1 isotypes show a distribution characteristic for linker histones. Cells were transiently transfected as described in Materials and Methods. (B and C) Distribution of H1F0-GFP and H1FOOα-GFP in GV stage oocytes. GV-stage oocytes were injected with in vitro-transcribed polyadenylated RNA as previously described (Tanaka et al., 2004). Only H1FOO-GFP shows the stage-specific surrounding nucleolus distribution (B), whereas H1F0-GFP shows aberrant predominantly nucleolar distribution and illuminates only a few nucleoplasmic foci (D). Live cells were monitored using confocal laser scanning microscopy as previously described (Becker et al., 2002). Arrowheads indicate nucleoli; the GV is indicated by a dashed line. Scale bars, 3 mm.

Figure 3.

The globular/N-terminal domain of H1FOO is responsible for the correct subnuclear distribution. (A) Schematic representation of the analyzed H1FOO deletion and chimeric constructs. N, N-terminal domain; G, globular domain; C, C-terminal domain. (B-I) Subnuclear distribution of H1FOO mutants in oocytes (B, D, F, and H) and 3134 cells (C, E, G, and I). (D and E) The N-terminal and globular domain of H1FOO are sufficient to mediated correct localization in oocyte and somatic nuclei. (F and G) The C-terminal domain of H1F0 fused to the N-terminal and globular domain of H1FOO shows a subnuclear localization similar to wild-type H1FOO in oocyte and somatic nuclei. (B, C, H, and I) Neither the C-terminal domain of H1FOO alone (B and C), nor the globular domain of H1F0 fused to the N-terminal and the C-terminal domain of H1FOO (H and I) shows correct localization in oocyte and somatic cell nuclei. Arrowheads indicate nucleoli. Scale bars, 3 mm.

Figure 4.

Dynamic exchange of linker histone isoforms and mutants with oocyte chromatin. (A and B) The nuclei of oocytes expressing either H1F0-GFP (A) or H1FOO-GFP (B) were imaged before and after photobleaching of chromatin foci at the indicated time points. The recovery of the fluorescent signal was monitored by time-lapse microscopy. A segment magnification of the bleached area indicated by the red rectangle is shown in false color below the corresponding panels. (C and D) Quantitation of recovery kinetics. For quantitation, at least four oocytes from two independent experiments were used. Scale bars, 3 mm.

Figure 5.

Dynamic exchange of linker histone isoforms and mutants with somatic chromatin. (A and B) 3134 cells expressing either H1F0-GFP (A) or H1FOO-GFP (B) were imaged before and after photobleaching of chromatin. The recovery of the fluorescent signal was monitored by time-lapse microscopy. (C and D) Quantitation of recovery kinetics. For quantitation, at least 10 cells from two independent experiments were used. Scale bars, 3 mm.

Figure 6.

Reduced dynamic exchange of H1F0CC-GFP is not sufficient for H1FOO-like localization in oocytes and does not block removal from R1 nuclei during SCNT. (A) Schematic representation of the H1F0CC-GFP chimeric protein. N, N-terminal domain; G, globular domain; C, C-terminal domain. (B and C) Quantitation of recovery kinetics of H1F0CC-GFP and H1F0-GFP in somatic cells (B) and oocytes (C). For quantitation of recovery kinetics in oocytes, four oocytes from two independent experiments were used. For quantitation of recovery kinetics in somatic cells, 10 cells from at least two independent experiments were used. Scale bar, 3 mm. (D and E) Representative example of the distribution of H1F0CC-GFP in 3134 nuclei (D) and oocyte nuclei (E). (F) R1 nuclear transfer experiment as described in Figure 1 (also see Materials and Methods). H1F0CC-GFP expressed from a chromosomal locus in a R1 ES cell nucleus is removed from R1 chromatin with a kinetic similar to that of the endogenous somatic H1 (compare Figure 1, B and C). To monitor fluorescence loss by bleaching a second nucleus was placed in the perivitelline space next to the ooplasm for comparison. Arrow indicates the injected nucleus; arrowhead indicates the nucleus placed in the perivitelline space. Scale bar, 20 mm.