Early embryogenesis, from the zygote to the blastocyst stage, is a time of complicated and delicate gene regulation. The most important event during early embryonic development is zygotic gene activation (ZGA), which must occur according to an exact timing for normal embryonic development. The earliest stages of embryogenesis are regulated by maternally inherited components, at least until the activation of the zygotic genome. Transcription from the newly formed zygotic genome occurs in different timing and can further be divided into two phases: a minor activation (minor ZGA) before cleavage and a major activation (major ZGA) at the two-cell stage (Hamatani et al.,2004). When ZGA occurs, the transcripts derived from the zygotic genome replace the maternal transcripts and promote a dramatic reprogramming of gene expression pattern (Nothias et al.,1995; Schultz et al.,1999; Schultz,2002; Kanka,2003). Although much experimental evidence indicates that ZGA occurs by the two-cell stage, results of more recent experiments indicate that the one-cell embryo is transcriptionally active (Ram and Schultz,1993; Matsumoto et al.,1994; Bouniol et al.,1995; Aoki et al.,1997). Many genes are known to be expressed during the preimplantation period, but only a few have been found to be essential for development (Stanton et al.,2003; Zeng and Schultz,2003). Further study of the function of regulatory genes expressed in this early embryonic stage can provide a better understanding of the molecular mechanism of early embryonic development.
Homeobox genes play important roles in many developmental processes (Rajkovic et al.,2004; Del Bene and Wittbrodt,2005; Garcia-Fernandez,2005). To identify novel homeobox genes involved in early embryogenesis, we performed silicon cloning with the NCBI dbEST databases of mouse early embryos (http://www.ncbi.nlm.nih.gov) to search for homeobox-containing genes. Obox6 was one of several genes identified in the mouse two-cell embryonic cDNA library (Yeh et al.,2002) and was classified as a member of the Obox family (Rajkovic et al.,2002). Obox6 spans 1.5 kb in the genome, which is located on mouse chromosome 7. Obox6 contains two exons, which encode 347 amino acids, and the 60 amino acid homeodomain is split across these two exons. Rajkovic et al. first identified Obox1 and Obox2 genes through in silico (electronic database) subtraction to identify expression sequence tags (ESTs) that are preferentially expressed in the adult mouse germ cells. The other four members, Obox genes 3 to 6, were further identified through a GenBank search for homologs of Obox genes 1 and 2. These six Obox members were all mapped to proximal chromosome 7, and transcripts were preferentially expressed in the gonads (Rajkovic et al.,2002).
By reverse transcriptase-polymerase chain reaction (RT-PCR) analysis, we found that Obox6 transcripts were specifically detected in the early embryonic stages of mice and exhibited elevated expression from the two-cell to the morula stage. The expression pattern of Obox6 in early embryos suggested that Obox6 might play a crucial role in early embryogenesis. To understand the function of Obox6 in vivo, we inactivated the mouse Obox6 using a gene targeting approach. The Obox6-mutant mice were born and grew normally like their wild-type littermates. The absence of detectable phenotypes in the developmental process of Obox6 mutants suggests a functional compensation from other Obox genes.
Generation of Obox6 Mutant Mice
To uncover the role of Obox6 in mouse development, we inactivated mouse Obox6 by gene targeting. The targeting vector was constructed to delete the entire coding region of Obox6 (Fig. 1A). After electroporation and drug selection, 216 embryonic stem (ES) cell clones were analyzed for homologous recombination. Three targeted ES clones were identified, and one of them was used in blastocyst microinjection. Chimeric mice were produced, and germline transmission was achieved. PCR analysis was used to identify the wild-type and mutant alleles of Obox6 in the mouse genotyping (Fig. 1B). As shown in Figure 1A, PCR primer pairs used to detect the mutant allele could specifically amplify the mutant allele resulted from homologous recombination. Southern analysis with both neo and 5′ probes further confirmed this result (Fig. 1C). The expression of Obox6 in early embryonic stages of these mice was identified by RT-PCR and quantitative RT-PCR analyses. In wild-type mice, the Obox6 expression profile was followed from egg to blastocyst, with maximum expression in two-cell and morula stages (Fig. 2A,C). Expression of Obox6 in adult tissues or late embryogenesis of wild-type animals was not identified (data not shown). Obox6 transcripts were totally absent in the Obox6 null embryos obtained from homozygous mutant parent crossing (Obox6−/− female × Obox6−/− male, Fig. 2B,C). To further understand whether the expression of Obox6 is of maternal origin, embryos from Obox6−/− female crossed with wild-type male were characterized. As shown in Figure 2D, Obox6 transcripts can be identified after two-cell stage but not in the eggs and zygotes. This result indicates that the zygotic expression is not of maternal origin. Furthermore, immunocytochemical analysis with an Obox6-specific antibody detected Obox6 protein in the nucleus of wild-type embryos but not in mutant embryos (Fig. 3A–H). All these results above reveal that the gene targeting strategy successfully generated Obox6 mutant mice.
Obox6 Mutant Mice Have Normal Growth and Reproductive Function
To determine the effects of Obox6 deficiency on mouse development, heterozygous mice of C57BL/6J background were crossed to obtain homozygous mutant mice. Genotyping analysis of the offspring from litters derived from heterozygous mating showed normal Mendelian ratio, averaging 9.6 pups per litter (Table 1), indicating that Obox6 inactivation does not cause embryonic lethality. Further observation and breeding indicated that Obox6 null mice were morphologically indistinguishable from their heterozygous and wild-type littermates and can reproduce normally, with a typical litter size of 7 to 12 pups. The same results were observed when heterozygous male (+/−) were crossed with homozygous female (−/−) or in Obox6-null mice intercrossing; the litter size (8.8 and 8.0 pups per litter, respectively) was not statistically different from the wild-type intercrossing (8.8 pups per litter; Table 1).
Table 1. Reproductive Performance of Obox6 Mutant Mice
No significant differences were found in number of animals born or in the average litter size (P > 0.05).
Normal Development in Homozygous Obox6 Mutant Embryos
According to the previous data shown in Figure 2A, we found that Obox6 is expressed exclusively in early embryonic stages and exhibits an elevated expression from the two-cell to the morula stage. For further characterization of the developmental process of the mutant embryos during preimplantation periods, mutant embryos were flushed out from the oviducts and cultured in vitro. All the embryos showed normal morphology in each developmental stage (Fig. 3I–P). This result suggests that Obox6 is not essential for embryonic development.
Expression of Obox Genes and Other Genes in Obox6 Null Embryos
Compensation among Obox genes is the most likely explanation for the lack of morphological and functional abnormalities in Obox6 mutant mice. The expression of Obox family genes in Obox6 null embryos was examined by RT-PCR and quantitative RT-PCR (Fig. 4A,B) analyses. As shown in Figure 4, in the wild-type embryos, transcripts of Obox1/Obox2, Obox3, and Obox5 are abundant in one-cell embryos and decrease significantly in further development. After the morula stage, no expression signals of these genes were observed. However, in Obox6 null embryos, expression of Obox1/Obox2, Obox3, and Obox5 maintained 25–35% of their zygotic expression level at the two-cell stage. Obox4, preferentially expressed in the testes, is expressed at a much lower level during early development (Rajkovic et al.,2002), and no expression signal was detected in the wild-type or Obox6 mutant embryos. The expression levels of genes associated with oocyte maturation and early embryonic developmental process were also analyzed. Both Oct4 and Nanog are required to maintain the pluripotency of cells (Ovitt and Scholer,1998; Chambers et al.,2003; Mitsui et al.,2003). According to the previous studies, Oct4 is expressed in embryos during the preimplantation period (Ovitt and Scholer,1998). No expression of Nanog was detected during the early cleavage stages until the morula stage (Chambers et al.,2003; Mitsui et al.,2003). Tcl1 mRNA is present in one-cell and two-cell embryos but declines rapidly during the late preimplantation stages (Narducci et al.,2002). In our studies, no significant change was identified among Oct4 and Nanog between the normal and Obox6 mutant embryos. However, we found that Tcl1 remained at a higher expression level in the two-cell mutant embryos than in the wild-type embryos at same stage (Fig. 4A,B). These findings suggest that functional redundancy among the Obox genes may compensate for the loss of Obox6, allowing the developmental process to progress normally. In addition, Obox6 knockout affected the expression levels of Tcl1 genes, which have been shown to be critical for early embryo development.
In this study, we used a gene targeting approach to investigate the in vivo function of the novel homeobox gene, Obox6, which is specifically expressed in the early embryos of mice. Obox6 homozygous mutant embryos from Obox6−/− parents show undetectable Obox6 expression in either mRNA or protein levels, confirming that the targeted disruption results in a loss-of-function allele. Obox6 mutant mice can grow and reproduce normally without any apparent defects, indicating that Obox6 is not essential during early embryonic development in mice.
Maternal or zygotic Obox6 expression in the egg and early embryonic stage was characterized in Obox6 mutants generated from Obox6−/− females crossed with either wild-type or mutant males. This result reveals that the Obox6 expression in eggs and zygotes are maternal in origin. The Obox6 mRNAs were synthesized and stored in oocytes as maternal inherited components. Fertilization triggers the initial zygotic transcription (minor ZGA) before the first cleavage. The robust Obox6 expressions from two-cell to morula are the results of major ZGA. These mechanisms regulating Obox6 expression might be a good model system to study the minor and major ZGA processes. Preimplantation development represents a tightly controlled program of cell division and differentiation. More than 15,700 mouse genes expressed during preimplantation development have been identified from cDNA sequences deposited in the UniGene database of the National Institutes of Health (Stanton et al.,2003). Spatial and temporal differences in gene expression play an important role in establishment of axes at the two-cell stage and development of the trophectoderm and inner cell mass after eight-cell stage embryo compaction. This expression profile suggests that Obox6, a homeodomain-containing transcription factor, might play a role in early embryogenetic events such as ZGA. Genes whose expression is activated by Obox6 might then be involved in the regulation of cell cycle and differentiation modulation. However, Obox6 homozygous mutant mice did not exhibit any obvious phenotypes, and showed normal growth and reproductive activity. Careful characterization of preimplantation embryos during in vitro culture also revealed no morphological differences between wild-type and Obox6 null embryos. Genetic redundancy may be a good explanation for the phenomena. Previous sequence alignment has shown that the homeodomain of Obox6 is highly homologous to that of other Obox members (60% identity and 80% similarity; Rajkovic et al.,2002). During normal development, Obox1/Obox2, Obox3, and Obox5 transcripts are abundant in one-cell zygotes, but decrease significantly by the two-cell stage. Our RT-PCR and quantitative RT-PCR analysis of Obox6 mutant results showed that Obox1/Obox2, Obox3, and Obox5 elevated their expression levels at the two-cell stage, when Obox6 was expressed most abundantly in the wild-type embryos. The extensive expression of Obox1/Obox2, Obox3, and Obox5 at the two-cell stage in Obox6 null embryos may suggest functional compensation between Obox gene members.
Expression patterns of Oct4 and Nanog show no obvious difference between wild-type and Obox6 mutant embryos in our investigation. However, expression of Tcl1 was maintained at a similar level as the Obox members. It is well known that the overexpression of TCL1 in the B-cell lineage promotes B-cell chronic lymphocytic leukemia and B-cell lymphomas in mice (Bichi et al.,2002; Hoyer et al.,2002). The Tcl1 protein has been shown to interact with Akt to participate in transduction of antiapoptotic and proliferative signals (Laine et al.,2000; Pekarsky et al.,2000; Kang et al.,2005). The Tcl1 gene is normally expressed in ovaries, testes, preimplantation embryos, fetal thymus, and bone marrow (Narducci et al.,1997; Hallas et al.,1999). Functional studies of Tcl1 have demonstrated that lack of maternally derived Tcl1 impairs the embryo's ability to undergo normal cleavage (Narducci et al.,2002). A recent study showed that the numbers of thymocytes and bone marrow B-lineage cells were significantly reduced in Tcl1mutant mice (Hoyer et al.,2002). These observations indicate that Tcl1 plays a major role in the proliferation of early embryos and lymphocytes. The maternal effect genes were expressed abundantly in oocytes or one-cell embryos and decreased rapidly in two-cell embryo stage. Whether the elevated (or prolonged) expression of Tcl1 in the two-cell stage was the effect of the sustained expression of other Obox family members or through other pathways unrelated to these Obox proteins remain to be elucidated.
In summary, we have shown that Obox6 is specifically expressed in preimplantation embryos and is not essential for normal embryogenesis in mice. Our study also revealed that there might be functional redundancy among Obox members based upon their high homology and expression level alteration. Further studies of Obox genes will be necessary to understand the functional roles of these evolutionarily conserved transcription factors in the developmental process. The elucidation of the regulatory network among Obox genes and other development related genes could also significantly advance our understanding of early embryonic development.
C57BL/6J strain mice were used in this study. All animals were maintained in a specific pathogen-free environment under standard laboratory conditions and handled following the guidelines of the institutional animal committee.
Generation of Obox6 Mutant Mice
We screened the NCBI dbEST databases and performed in silico subtraction of homeobox genes of two-cell stage ESTs against the entire collection of ESTs. Obox6 was one of the EST clones identified from the mouse two-cell cDNA libraries (Rajkovic et al.,2002; Yeh et al.,2002). The Obox6 genomic DNA was isolated from a 129/SvJ mouse genomic library and subcloned into the pBluescript SK vector (Stratagene). A 1-kb upstream fragment and a 5.5-kb downstream fragment of the Obox6 gene were subcloned into XhoI and SalI sites of the pOSdupdel vector, respectively. The entire coding region of the Obox6 gene was replaced by an MCI-neo cassette. The targeting vector was linearized with NotI and electroporated into ES cells. Clones resistant to G418 and gancyclovir were screened by PCR analysis using the following primers: Obox6F2 (5′-CAAACCTACTTCCACCTTCCCG-3′) and NeoR (5′-GAAAACCACACTGCTCGACGAA-3′). Chimeric mice were generated by standard methods and bred with female C57BL/6J mice to test for germline transmission. To generate the Obox6−/− congenic strain on a C57BL/6J background, Obox6+/− mice were backcrossed to strain C57BL/6J for more than seven generations.
Genotyping of the Obox6 Mutant Mice
Genomic DNA was isolated from tail biopsy specimens of weaned mice. Southern analyses were performed using both 32P-labeled Obox6-specific and Neor fragment probes. After digestion with SphI, the sizes detected by the Obox6-specific probe in the wild-type and mutant alleles were 9 and 3.5 kb, respectively. A 6.2-kb genomic fragment from the BglII-digested mutant allele was detected by Neor probe. PCR analyses were performed on progeny after their parents' genotypes had been confirmed by Southern blot analyses. The primers used for detection of wild-type alleles were Obox6F1 (5′-ATGCTTCAATACAATCAGAGCCC-3′: exon 1 sequence) and Obox6R1 (5′-TCAATC CCCCAACTCATCAAAGT-3′: exon 2 sequence). Primers Obox6F2 and NeoR were used to detect the target alleles.
Embryos were obtained at the appropriate stage, and 50 embryos were used for total RNA extraction according to the guanidinium thiocyanate procedure (Shim et al.,1997). These total RNAs were reverse transcribed with Superscript II RNase H− reverse transcriptase (Invitrogen) and PCR was carried out with Advantage2 polymerase (Clontech) for 30 cycles. Sequences of the primers used are listed in Supplementary Table S1, which can be viewed at http://www.interscience.wiley.com/jpages/1058-8388/suppmat.
Total RNA was extracted from pools of 50 embryos and used for reverse transcription as previous described. Quantitative real-time PCR was conducted using a LightCycler 2.0 System (Roche) with the LightCycler FastStart DNA Master SYBR green kit (Roche) according to the manufacturer's instructions. Sequences of the primers used are listed in Supplementary Table S1. Amplification specificity was controlled by production of a melting curve. After the validation of primer pair efficiency by construction of a standard curve, expression level of each gene was measured three times independently in every sample. Gene expression data was normalized by using β-Actin as internal control.
Embryos at the appropriate stages were fixed in 2.4% paraformaldehyde in M2 medium for 1 hr at room temperature. Fixed embryos were washed and incubated for 3 hr in phosphate buffered saline (PBS) containing 0.1 M glycine and 0.3 mg/ml bovine serum albumin (BSA), and then permeabilized in PBS containing 0.1% Triton X-100 for 15 min at room temperature. After washing through three changes of PBS containing 1 mg/ml BSA (PBS-BSA), immunostaining procedure was processed. An overnight incubation at 4°C in the presence of the rabbit anti-Obox6 (1:2,000) antibody, raised against His6-Obox6 fusion protein, was followed by washing in PBS-BSA and 1 hr of incubation with the secondary antibody Cy3-conjugated anti-rabbit IgG (1:500, Jackson ImmunoResearch) at room temperature. Obox6 was detected in samples by laser microscopy (Olympus IX70).
Embryo Isolation and In Vitro Culture
Embryos of appropriate stages were collected from the oviducts or uterus with M2 medium and morphologies were observed by optical microscope. For in vitro culture, fertilized eggs were collected by M2 medium and cultured in M16 medium until the blastocyst stage.
We thank Dr. Harry Wilson and Ms. M. Loney for their critical reading of this manuscript.