Chorioallantoic fusion defects and embryonic lethality resulting from disruption of Zfp36L1, a gene encoding a CCCH tandem zinc finger protein of the Tristetraprolin family - PubMed (original) (raw)

Chorioallantoic fusion defects and embryonic lethality resulting from disruption of Zfp36L1, a gene encoding a CCCH tandem zinc finger protein of the Tristetraprolin family

Deborah J Stumpo et al. Mol Cell Biol. 2004 Jul.

Abstract

The mouse gene Zfp36L1 encodes zinc finger protein 36-like 1 (Zfp36L1), a member of the tristetraprolin (TTP) family of tandem CCCH finger proteins. TTP can bind to AU-rich elements within the 3'-untranslated regions of the mRNAs encoding tumor necrosis factor (TNF) and granulocyte-macrophage colony-stimulating factor (GM-CSF), leading to accelerated mRNA degradation. TTP knockout mice exhibit an inflammatory phenotype that is largely due to increased TNF secretion. Zfp36L1 has activities similar to those of TTP in cellular RNA destabilization assays and in cell-free RNA binding and deadenylation assays, suggesting that it may play roles similar to those of TTP in mammalian physiology. To address this question we disrupted Zfp36L1 in mice. All knockout embryos died in utero, most by approximately embryonic day 11 (E11). Failure of chorioallantoic fusion occurred in about two-thirds of cases. Even when fusion occurred, by E10.5 the affected placentas exhibited decreased cell division and relative atrophy of the trophoblast layers. Although knockout embryos exhibited neural tube abnormalities and increased apoptosis within the neural tube and also generalized runting, these and other findings may have been due to deficient placental function. Embryonic expression of Zfp36L1 at E8.0 was greatest in the allantois, consistent with a potential role in chorioallantoic fusion. Fibroblasts derived from knockout embryos had apparently normal levels of fully polyadenylated compared to deadenylated GM-CSF mRNA and normal rates of turnover of this mRNA species, both sensitive markers of TTP deficiency in cells. We postulate that lack of Zfp36L1 expression during mid-gestation results in the abnormal stabilization of one or more mRNAs whose encoded proteins lead directly or indirectly to abnormal placentation and fetal death.

PubMed Disclaimer

Figures

FIG. 1.

Gene-targeting strategy for Zfp36L1 inactivation. (A) Schematic representation of the normal mouse genomic locus for Zfp36L1. The two exons are represented by light gray shading, and the translational start site in exon 1 is indicated by the large black arrow. Genomic 5′ (black boxes) and 3′ (horizontal hatching) probes outside the targeted region are shown. (B) The targeting vector contained an interruption of the 5′ portion of exon 2 with a neomycin marker cassette (diagonal cross hatch), with the creation of new restriction sites for HindIII (box labeled H) and BamHI (box labeled B). A diptheria toxin resistance element (speckled hatching labeled DT) was also inserted into the vector. (C) The final targeted locus. The locations of PCR primers used to detect homologous recombination are indicated by the small arrows underneath the targeted locus. Cleavage sites for the following restriction enzymes are indicated: B, BamHI; E, Eco47III; H, HindIII; K, KpnI; N, NotI.

FIG. 2.

Southern and Northern analysis confirming Zfp36L1 targeting in ES cells. (A and B) Southern blots of mouse tail DNA from WT and Het. mice. (A) HindIII/XbaI cleavage of DNA and hybridization with the Zfp36L1 5′-flanking probe generated a diagnostic 7.5-kb fragment from WT (+/+) DNA (lane 1) and a 6.5-kb targeted Het. (+/−) DNA fragment (lane 2) (arrows). (B) BamHI cleavage of DNA and hybridization with the Zfp36L1 3′-flanking probe generated a diagnostic 6.0-kb WT DNA fragment (lane 3) and a 4.5-kb targeted (Het.) DNA fragment (lane 4) (arrows). Northern blots of total cellular RNA isolated from cultured MEFs are shown in panels C to F; each lane contained 15 μg of total cellular RNA from cells that were WT, Het., or KO for the targeted Zfp36L1 allele. The blots were probed with the indicated 32P-labeled cDNA probes for Zfp36L1 (C), GAPDH (D), TTP and cyclophilin (Cyclo) (E), and Zfp36L2 (F). Note that the apparently increased TTP and ZFP36L2 mRNA expression seen in the RNA from the KO cells in panels E and F is largely due to gel overloading in those lanes, as evidenced by the larger amounts of GAPDH and cyclophilin mRNA present in the same lanes in panels D and E.

FIG. 3.

Tissue distribution and developmental expression of Zfp36L1 mRNA. (A) Total cellular RNA from adult mouse tissues as well as from whole embryos at E12.5 and from ES cells were subjected to electrophoresis and Northern blotting using probes for Zfp36L1 and TTP mRNA, as indicated by the arrows. Each lane was loaded with 10 μg of total RNA. (B) A different Northern blot that used total RNA isolated from whole embryos at the embryonic day indicated or from the specified adult mouse tissues (B, brain; L, liver; T, testis). The blot was probed with either Zfp36L1 or cyclophilin (cyclo) cDNA probes, as indicated. Each lane contained 15 μg of total cellular RNA. (C) Northern blot using total cellular RNA from embryo head and placenta at E14.5. The same blot was probed for both TTP and GAPDH mRNAs, as indicated by the arrows.

FIG. 4.

Whole-mount in situ hybridization histochemistry of Zfp36L1 mRNA expression at approximately E8.0. WT embryos were processed for in situ hybridization histochemistry as described in Materials and Methods. Specific probe hybridization is indicated by the purple color. Note the higher-level expression in the neural folds, neurectoderm, and anterior neural ridge (ANR) as well as the patchy but readily detectable expression in the allantois (A to C). The specificity of staining with the Zfp36L1 probe was demonstrated by comparing staining of a WT embryo with that of its KO littermate embryo processed at the same time; no staining was seen (C). A control using a sense version of the probe was also negative (not shown). (D) Color development in the whole conceptus was stopped at an earlier time, demonstrating high-level expression in the allantois.

FIG. 5.

Phenotype of Zfp36L1-deficient embryos at E9.5. Right lateral views of an E9.5 WT embryo (A) and a KO littermate (B). Both embryos were photographed in the unfixed state. Note the abnormalities of the cranial neural tube in the KO embryo; the general underdevelopment of the abdominal organs and uncoiling of the heart tube; and the generally decreased size, abnormal body axis, and the presence of a large allantoic remnant or cyst (arrow).The head region (C and D) and thoracic region (E and F) of lateral confocal images of Lysotracker staining, indicating apoptosis. Note the widespread and striking staining in the neural tube of the KO embryo (D and F) compared to that of the WT embryo (C and E).

FIG. 6.

Phenotype of a _Zfp36L1_-deficient embryo at E13.5. (A and B) External appearance of the heads from WT (+/+) and KO (−/−) littermate mice at E13.5. The KO embryo was the longest surviving embryo in this series. Note the relatively normal facial structures in the KO mouse but the extreme abnormalities at the surface of the head. (C and D) Hematoxylin- and eosin-stained coronal sections through the heads of the same embryos at approximately the same level; one eye is present in each section (arrows). The extreme disorganization of the KO brain (D) should be contrasted with the brain from the WT littermate (C). Magnification, ×1.

FIG. 7.

Placental histology and cell proliferation at E9.5. Shown are sections through placentas from two WT and two KO embryos at E9.5 in which chorioallantoic fusion had occurred. The layers labeled include the maternal decidua (D), giant cell layer (G), spongiotrophoblast layer (S), labryrinthine layer (L), and allantois (A). The red-brown nuclear staining is for the Ki67 cell proliferation antigen; the counterstain was hematoxylin.

FIG. 8.

Placental histology and cell proliferation at E10.5. Shown are sections through three littermate placentas at E10.5, two WT (A and C) and one KO (B). The KO placenta was one in which chorioallantoic fusion had occurred. The sections were stained with an antibody to the cell proliferation antigen Ki67 (red-brown color) and counterstained with hematoxylin. The abbreviations are the same as those described in the legend for Fig. 7. Note the near absence of Ki67 staining in the decidual layer from all placentas; the relative atrophy of the embryonic layers in the KO placenta (B); and the markedly diminished Ki67 staining in the embryonic layers from the KO placenta (B).

FIG. 9.

GM-CSF mRNA stability in MEFs derived from Zfp36L1 WT and KO embryos. MEFs of the indicated genotype were stimulated with EGF (20 ng/ml for 30 min) (A to D) or mouse TNF (10 ng/ml for 60 min) (E and F) followed by actinomycin D (Act. D) (10 μg/ml in A to D, 5 μg/ml in E and F). The cells were then processed at different times for Northern blotting using a GM-CSF-specific probe. The indicated times represent the total elapsed time (in minutes) after the initial stimulation. (A and B) Comparison of results from WT and KO cells after EGF stimulation in one pair of MEFs derived from littermates of the WT and KO genotypes; after PhosphorImager normalization for the levels of GAPDH mRNA, the decay curves for the GM-CSF mRNAs after addition of actinomycin D (arrows) from this and a second identical experiment are shown in panels C and D. The solid lines are from the WT cells, and the dashed lines are from the KO cells. (E and F) Northern blots from WT and KO fibroblasts after stimulation with mouse TNF for 60 min followed by actinomycin D treatment. Note the similar relative expression levels of the two bands of the GM-CSF transcript in the two genotypes as well as the similar rapid decay rates of the mRNA in both genotypes. GAPDH mRNA levels are also shown in the same blots as loading controls.

References

1. Blackshear, P. J. 2002. Xenopus laevis genomic biomarkers for environmental toxicology studies, p. 339-353. In S. Wilson and W. A. Suk (ed.), Biomarkers of environmentally associated disease. CRC Press, Inc., Boca Raton, Fla.
1. Blackshear, P. J. 2002. Tristetraprolin and other CCCH tandem zinc-finger proteins in the regulation of mRNA turnover. Biochem. Soc. Trans. 30:945-952. - PubMed
1. Bortner, C. D., and J. A. Cidlowski. 1996. Absence of volume regulatory mechanisms contributes to the rapid activation of apoptosis in thymocytes. Cell Physiol. 40:C950-C961. - PubMed
1. Bulmer, J. N., L. Morrison, and P. M. Johnson. 1988. Expression of the proliferation markers Ki67 and transferrin receptor by human trophoblast populations. J. Reprod. Immunol. 14:291-302. - PubMed
1. Bustin, S. A., X. F. Nie, R. C. Barnard, V. Kumar, J. C. Pascall, K. D. Brown, L. M. Leigh, N. S. Williams, and I. A. McKay. 1994. Cloning and characterization of ERF-1, a human member of the Tis11 family of early-response genes. DNA Cell Biol. 13:449-459. - PubMed

Chorioallantoic fusion defects and embryonic lethality resulting from disruption of Zfp36L1, a gene encoding a CCCH tandem zinc finger protein of the Tristetraprolin family - PubMed (original) (raw)