Loss of Daxx, a promiscuously interacting protein, results in extensive apoptosis in early mouse development

Abstract

The mammalian Daxx gene has been identified in a diverse set of yeast interaction trap experiments. Although a facilitating role for Daxx in Fas-induced apoptosis has been suggested, Daxx's physiologic function remains unknown. To elucidate the in vivo role of Daxx, we have generated Daxx-deficient mice. Surprisingly, rather than a hyperproliferative disorder expected from the loss of a pro-apoptotic gene, mutation of Daxx results in extensive apoptosis and embryonic lethality. These findings argue against a role for Daxx in promoting Fas-induced cell death and suggest that Daxx either directly or indirectly suppresses apoptosis in the early embryo.

Figure 1

Targeted disruption of the mouse Daxx gene. (A) A map of the mouse genomic Daxx locus. The seven exons are indicated as open boxes. The targeting vector includes a thymidine kinase (TK) gene for negative selection and a neo gene for positive selection. Homologous recombination and insertion of neo results in a 2.5-kb deletion including exon 1 and most of exon 2. The positions of an external probe from 5′ of Daxx and of an internal cDNA probe used for genotyping HindIII-digested DNA by Southern blot analysis are shown. (B) BamHI; (E) EcoRI; (H) HindIII; (K) KpnI; (S) SacI; (Sal) SalI; (X) XhoI; (Xb) XbaI. Not all K and S sites are shown. (B) PCR analysis of embryo yolk sac genomic DNA. (C) Northern blot analysis of total RNA from blastocyst-derived cell lines. A cDNA probe, generated by PCR and encompassing bp 1596–2274, detects a wild-type band of 2.4 kb. In mutant cells, a stable transcript of 1.4 kb is detected.

Figure 1

Targeted disruption of the mouse Daxx gene. (A) A map of the mouse genomic Daxx locus. The seven exons are indicated as open boxes. The targeting vector includes a thymidine kinase (TK) gene for negative selection and a neo gene for positive selection. Homologous recombination and insertion of neo results in a 2.5-kb deletion including exon 1 and most of exon 2. The positions of an external probe from 5′ of Daxx and of an internal cDNA probe used for genotyping HindIII-digested DNA by Southern blot analysis are shown. (B) BamHI; (E) EcoRI; (H) HindIII; (K) KpnI; (S) SacI; (Sal) SalI; (X) XhoI; (Xb) XbaI. Not all K and S sites are shown. (B) PCR analysis of embryo yolk sac genomic DNA. (C) Northern blot analysis of total RNA from blastocyst-derived cell lines. A cDNA probe, generated by PCR and encompassing bp 1596–2274, detects a wild-type band of 2.4 kb. In mutant cells, a stable transcript of 1.4 kb is detected.

Figure 1

Targeted disruption of the mouse Daxx gene. (A) A map of the mouse genomic Daxx locus. The seven exons are indicated as open boxes. The targeting vector includes a thymidine kinase (TK) gene for negative selection and a neo gene for positive selection. Homologous recombination and insertion of neo results in a 2.5-kb deletion including exon 1 and most of exon 2. The positions of an external probe from 5′ of Daxx and of an internal cDNA probe used for genotyping HindIII-digested DNA by Southern blot analysis are shown. (B) BamHI; (E) EcoRI; (H) HindIII; (K) KpnI; (S) SacI; (Sal) SalI; (X) XhoI; (Xb) XbaI. Not all K and S sites are shown. (B) PCR analysis of embryo yolk sac genomic DNA. (C) Northern blot analysis of total RNA from blastocyst-derived cell lines. A cDNA probe, generated by PCR and encompassing bp 1596–2274, detects a wild-type band of 2.4 kb. In mutant cells, a stable transcript of 1.4 kb is detected.

Figure 2

Daxx mutant embryos are growth retarded, highly disorganized, and undergo extensive apoptosis. (A) A pair of E9.5 wild-type and Daxx mutant littermates are compared. (B) A pair of E9.5 wild-type and mutant deciduae are compared. (C–E) Hematoxylin and eosin-stained paraffin-embedded sections of embryos from E7.5 +/− and −/− littermates (C) and E8.5 +/+ and −/− littermates (D). (a) Amnion; (e) epiblast; (mc) myocardium; (ne) neuroepithelium; (n) notochord; (nt) neural tube; (s) somites; (vs) vascular space; (ys) yolk sac. (E) Evidence of pyknotic nuclei (pn) in sections from −/− embryos at E7.5 and E8.5.

Figure 3

Extensive apoptosis in Daxx mutant embryos. TUNEL assay by digoxigenin–UTP end-labeling of nucleosome fragments of paraffin-embedded sections from E7.5 +/− and −/− littermates (A); E8.5 +/+ and −/− littermates (B); the head region, allantois, and vascular space of E8.5 −/− embryos (C).

Figure 4

Increased levels of apoptosis in Daxx-deficient cell lines. (A) The percentage of apoptotic cells in wild-type and mutant cell cultures was measured as the sub-G₁ peak following propidium iodide staining and FACS analysis. Each bar represents an independent cell line. Error bars indicate s.d. (B) A DNA fragmentation assay was performed by isolating low-molecular-weight DNA from 1 × 10⁷ wild-type or mutant cells, followed by separation on a 2% agarose gel and staining with ethidium bromide. Each lane represents an independent cell line.

Figure 5

Daxx is localized to the nucleus. (A) Total cell extracts were prepared from wild-type and Daxx-deficient cell lines. Total cell lysate (30 μg) or cell lysate that had been immunoprecipitated with α-Daxx (1.5 mg), was loaded on an 8% SDS–polyacrylamide gel. α-Daxx was used for immunoblotting. (B) Wild-type ES cells were fractionated into the following subcellular components: nucleus (n); membrane (mb); and cytosol (c). Immunoprecipitation and immunoblotting were performed with α-Daxx. (C) HeLa nuclear extracts were used for immunoprecipitation with α-Daxx or α-CAP. α-Daxx was used for immunoblotting. (D) Immunoblotting of subcellular fractions was performed with an antibody recognizing laminin B.