BIOLOGICAL OVERVIEW

nanos is required maternally for two processes during oogenesis. nanos is expressed in the early germarium where it is needed for continued egg chamber production. In mature eggs and the developing zygote it is also required to specify posterior identity. Nanos mRNA is localized to the pole plasm, a specialized cytoplasm later incorporated into pole cells, the precursors of the fly's reproductive system (Wang, 1991, 1994).

For its role in specifying posterior identity, nos is required briefly but immediately after fertilization. It has two known targets: hunchback and bicoid. Translation of both targets is inhibited by the binding of NOS protein to the 3' untranslated region of their mRNAs. While Nanos itself does not bind to the "Nanos response elements" of HB mRNA, Pumilio does bind and appears to bring Nanos protein into the complex (Murata, 1995). Inhibition of Hunchback mRNA, its maternal transcript in particular, guarantees an anterior to posterior gradient of HB protein. The restriction of HB mRNA permits the ordered expression of genes specifying the abdominal region: Krüppel, knirps and giant. In the absence of nos, all abdominal segments are lost. It would seem that one of the requirements for NOS mRNA localization, involving proteins of the posterior group are imposed by the presence of the Nanos response elements in BCD and HB mRNAs (Wharton, 1991).

Eight genes are required for NOS mRNA localization. Among these are oskar, a gene necessary for organization of the germ plasm, pumilio, whose coded protein binds the 3'UTR of NOS mRNA, and vasa, coding for an ATP dependent RNA helicase. Translation of NOS mRNA requires the posterior localization. Consequently when the NOS mRNA is freed from its association with the posterior pole NOS translation does not proceed and embryos develop abdominal defects notwithstanding the stability of the unlocalized NOS mRNA within their systems. The translational silencing is mediated by NOS's 3'UTR sequence. Repression can be alleviated by either replacement of the 3'UTR with heterologous 3'UTR sequence or by posterior localization (Gavis, 1994).

Although it was previously thought that Nos functions primarily to allow abdomen formation, Nos is now also known to be required for embryonic germ cell migration. In many animal groups, factors required for germline formation are localized in germ plasm, a region of the egg cytoplasm. In Drosophila embryos, germ plasm is located in the posterior pole region and is inherited in pole cells, the germline progenitors. Transplantation experiments have demonstrated that germ plasm contains factors that can form germline, and germ plasm also directs abdomen formation. Genetic analysis has shown that a common mechanism directs the localization of the abdomen and germline-forming factors to the posterior pole. The critical factor for abdomen formation is Nanos. Nos is also essential for germline formation in Drosophila; pole cells lacking Nanos activity fail to migrate into the gonads, and therefore do not become functional germ cells. A function for Nos protein in Drosophila germline formation is compatible with observations of its association with germ plasm in other animals (Kobayashi, 1996). The product of nos is required at the posterior pole of the embryo for the differentiation of abdominal structures, but not for pole cell formation. Kobayashi reported that nos also controls the timing of the initiation of transcription of germline-specific genes. The experiments that led to this hypothesis have been repeated and further experiments are reported. Contrary to what had been earlier reported, germ cells of embryos deriving from nos females do not show premature gene expression. Germ cells of such embryos, however, often show artifactual lacZ staining even in the absence of a lacZ gene (Heller, 1998).

Experiments by another laboratory, however, show that Nos regulates certain germ cell markers. In order to probe the gene expression in pole cells, ten enhancer-trap lines that showed beta-gal expression in pole cells were screened. All of these enhancer-trap markers are fully activated in pole cells after their migration to the embryonic gonads. In the pole cells lacking Nos, the expression of nine out of ten enhancer-trap markers is affected. Among nine markers, five (Type-A) were prematurely expressed in the pole cells during the course of their migration. The expression of the other four markers (Type-B) initiates correctly after pole-cell migration, but their expression is significantly reduced. Thus, it is concluded that the maternal Nos plays a dual role in zygotic gene regulation in pole cells: to define the stages of expression for Type-A markers, and to enhance expression for Type-B markers. Contrary to the results presented here, Heller (1998) has recently reported that no premature expression of Type-A markers occurs in the pole cells of embryos derived from nos mutant females. This discrepancy is due to the difference in the nos mutant alleles used for these analyses. The current study used the much stronger allele, nos^BN. pumilio mutations, like nanos, affect the expression of the enhancer-trap markers in pole cells (Asaoka, et al. in preparation). This suggests that Nos cooperates with Pum in pole cells to regulate the gene expression by means of a translational repression mechanism. This is likely to occur with a similar mechanism by which Pum, along with Nos, represses the translation of maternal Hunchback mRNA. The premature expression of Type-A markers in nos pole cells can be explained by a failure of Nos to repress translationally the production of regulatory factor(s) responsible for Type-A enhancer activation. These activator(s) may be stored in pole cells as mRNA(s) whose translation is repressed by Nos. Once maternal Nos protein disappears from pole cells at around zygotic stage 15, the activator mRNA(s) is translated to produce activator(s) for Type-A enhancers. In contrast, Nos could act as a permissive factor that would allow the expression of Type-B enhancers (Asaoka, 1998).

To study the mechanism of nanos-mediated translational repression, the mechanism by which maternal Hunchback mRNA is translationally activated was examined. In the oocyte from wild-type females, where no HB translation is detected, the mRNA has a poly(A) tail length of approximately 30 nucleotides. However, concomitant with translation of the mRNA at between 0.5 and 1.5 hours after egg deposition, the poly(A) tail is elongated to approximately 70 nt. In the absence of nanos activity, the poly(A) tail of Hunchback mRNA is elongated to approximately 100 nt concomitant with its translation, suggesting that cytoplasmic polyadenylation directs activation. However, in the presence of nanos the length of the Hunchback mRNA poly(A) tail is reduced via the nanos response element present in the HB 3'UTR. To determine if nanos activity represses translation by altering the polyadenylation state of Hunchback mRNA, various in vitro transcribed RNAs were injected into Drosophila embryos and changes in polyadenylation were determined. nanos activity reduces the polyadenylation status of injected Hunchback RNAs by accelerating their deadenylation. Pumilio activity, which is necessary to repress the translation of Hunchback mRNA, is also needed to alter polyadenylation. An examination of translation indicates a strong correlation between poly(A) shortening and suppression of translation. These data indicate that nanos and pumilio determine posterior morphology by promoting the de-adenylation of maternal Hunchback mRNA, thereby repressing its translation (Wreden, 1997).

nanos plays an important role during zygotic development in downregulating mitosis and transcription during the development of the Drosophila germline. Germ cells in embryos derived from nos mutant mothers do not migrate to the primitive gonad and prematurely express several germline-specific markers. These defects have been traced back to the syncytial blastoderm stage. Pole cells in nos minus embryos fail to establish/maintain transcriptional quiescence; the sex determination gene Sex-lethal (Sxl) and the segmentation genes fushi tarazu and even-skipped are ectopically activated in nos minus germ cells. nos minus germ cells are unable to attenuate the cell cycle and instead continue dividing. Unexpectedly, removal of the Sxl gene in the zygote mitigates both the migration and mitotic defects of nos minus germ cells. Supporting the conclusion that Sxl is an important target for nos repression, ectopic, premature expression of Sxl protein in germ cells disrupts migration and stimulates mitotic activity (Deshpande, 1999).

Soon after formation, wild-type pole cells in Drosophila downregulate RNA polymerase II transcription until they have been incorporated into the primitive gonad. The premature activation of these germline-specific genes is likely to reflect a more general defect in transcriptional regulation that arises early in embryogenesis, soon after the pole cells are formed. Instead of shutting off RNA polymerase II transcription, nos^- pole cells inappropriately transcribe several somatic genes. Why do nos^- germ cells fail to regulate RNA polymerase II transcription? The only known regulatory target for nos in the embryo is the hb transcription factor. Nos together with the Pumilio protein is thought to bind to maternally derived hb mRNA and block its translation. Since Hb protein is produced throughout much of the posterior in the absence of Nos, one possibility is that this gap gene protein activates transcription in the pole cells. However, this explanation does not seem likely. Although hb regulates eve and ftz in the soma, it is not clear that the ectopic expression of only the Hb protein would be sufficient to activate either of these genes in the absence of other factors. In addition, hb has no known role in controlling the activity of Sxl-Pe. In fact, ectopic expression of Hb in the soma seems to repress rather than activate Sxl-Pe. Finally, germ cells derived from nos^-hb^- germline clones exhibit a similar set of developmental defects as nos^- germ cells. Conversely, these defects are not induced when hb is ectopically expressed in pole cells (Deshpande, 1999).

A more likely possibility is that nos^- germ cells have a defect in the system responsible for attenuating RNA polymerase II activity. If this is true, there must be additional target(s) for nos regulation, in addition to maternal hb mRNA. In this context it is interesting to note that the failure to establish/maintain transcriptional quiescence in nos^- pole cells is reminiscent of the defects seen in the germ cell lineage of C. elegans pie-1 mutants. In pie-1 mutants, the germ cells transcribe many genes that are normally expressed only in the soma, assuming an inappropriate identity. The Pie-1 protein contains two copies of a C₃H zinc finger motif found in proteins implicated in pre-mRNA metabolism. Antibody staining indicates that Pie-1 is restricted to germ cells and is localized preferentially in the nucleus. A plausible explanation for why pie-1 mutants fail to repress transcription comes from studies on the phosphorylation of the RNA polymerase II large subunit carboxy-terminal domain (CTD). The CTD contains tandem repeats of a seven-amino acid sequence that contains two serine residues (2 and 5) that are targets for phosphorylation. Phosphorylation is thought to play an important role in polymerase elongation and in the recruitment of pre-mRNA modifying enzymes. In wild-type C. elegans, RNA polymerase II phosphorylated in the serine-2 residues of the CTD is detected in somatic cells but is not observed in the germline. In contrast, in pie-1 mutants, phosphorylated CTD serine-2 is detected in both cell types. The available evidence suggests that Pie-1 may directly inhibit CTD serine-2 phosphorylation, perhaps by interfering with a protein that recognizes the CTD (Deshpande, 1999 and references therein).

Like the C. elegans germline lineage, pole cells in Drosophila also have greatly reduced levels of phosphorylated CTD serine-2, which could be responsible for the inhibition of RNA polymerase II in the fly germline. However, it is not clear at this point whether the failure to establish/maintain transcriptional quiescence in nos^- pole cells is due to a defect in the system that regulates serine-2 phosphorylation. The level of CTD serine-2 phosphorylation in nos^- pole cells is not greatly different from that seen in wild type. It is possible that the changes in the level of CTD serine-2 phosphorylation in nos^- pole cells were too small to detect. Alternatively, RNA polymerase II activity in the germline might be controlled not only by CTD phosphorylation, but also by some other unknown mechanism which is the target for the Nos protein. Supporting this later possibility is the finding that only a subset of the genes expressed in the soma of early embryos are ectopically activated in nos^- pole cells. By contrast, changes in the status of CTD phosphorylation might be expected to have quite global effects on transcription. Further studies will be required to resolve this question (Deshpande, 1999 and references therein).

Another feature that distinguishes germline precursors from the soma is the cell cycle. Since hb has no known role in cell cycle regulation, the cause of the cell cycle defect was suspected to be a failure in the regulation of some other target gene. One possible candidate is cyclin B. However, no obvious abnormalities could be detected in cyclin B accumulation in nos^- pole cells at earlier stages of embryogenesis. Since cell cycle defects are evident in nos^- pole cells as early as the syncitial blastoderm stage, the reduction in cyclin B midway through embryogenesis might be simply a consequence of continued cycling rather than arresting the cycle in G2 at the blastoderm stage. Another possible cell cycle candidate is String. Germ cells are blocked in G2 because Cdc2 is inhibited by the phosphorylation of two amino acid residues, Thr-14 and Tyr-15. Since phosphorylated Cdc2 is normally reactivated by the string phosphatase, cdc25^stg, it is possible that Nos prevents the translation of maternally derived stg mRNA in germ cells (Deshpande, 1999 and references therein).

In wild-type embryos, transcription factors, such as Runt, Sisterless-a, and Scute, are responsible for activating the Sxl establishment promoter, Sxl-Pe. The genes encoding these positive regulators are on the X chromosome, and they are expressed in the early precellular zygote in direct proportion to the number of gene copies. Sufficient quantities of the X-linked activators are produced by 2X/2A nuclei to activate Sxl-Pe, while quantities produced by 1X/2A nuclei are insufficient to activate Sxl-Pe. Pole cells differ from the surrounding soma in that these activators are not expressed at detectable levels in the germline precursors, and Sxl-Pe remains off in both sexes. The failure to express these activators most likely reflects the global downregulation of RNA polymerase II transcription in wild-type pole cells. Thus, one mechanism that might account for the inappropriate activation of Sxl-Pe in nos^- pole cells would be a general derepression of somatically active genes. As a consequence, the genes encoding the X-linked activators would be expressed, and these in turn would activate Sxl-Pe. Although it seems reasonable to believe that the ectopic expression of the X-linked activators could contribute to the activation of Sxl-Pe in nos^- pole cells, it does not readily explain why Sxl-Pe is turned on not only in 2X but also in 1X pole cells. Moreover, when Scute protein expression was examined in nos^- embryos, the level of Scute protein in pole cells was less than that seen in 1X/2A somatic nuclei. For this reason, it is suspected that Sxl-Pe may be activated in nos^- pole cells by a mechanism that, at least in part, bypasses the normal regulation of this promoter by the X/A counting system (Deshpande, 1999).

Although Nos protein is likely to control Sxl-Pe activity by an indirect mechanism, a number of lines of evidence indicate that Sxl is an important nos regulatory target. In wild-type, Sxl proteins are normally not expressed in the germline until after the formation of the primitive gonad, and at this stage expression is restricted to the female germline. As a consequence of the ectopic activation of Sxl-Pe, Sxl proteins are present in nos^- pole cells at the blastoderm stage. It would appear that the premature appearance of Sxl proteins in the pole cells is an important contributing factor to the nos^- phenotype. The migration and cell cycle defects of nos^- germ cells can be alleviated by the elimination of the Sxl gene. Conversely, it is possible to induce both of these defects in wild-type germ cells by ectopically expressing Sxl protein. While the removal of Sxl mitigates some of the defects of nos^- germ cells, it should be noted that these cells are still abnormal. They fail to establish/maintain transcriptional quiescence, and they cannot form a functional adult germline. This finding indicates that Sxl is not the only target for nos regulation (Deshpande, 1999).

Why does ectopic expression of Sxl protein (either in the absence of nos or in the presence of the Sxl transgene) disrupt germ cell migration and induce cell cycle defects? Sxl encodes an RNA-binding protein that functions in the soma as both a splicing and translational regulator. Since the Sxl protein is predominantly localized in the cytoplasm of nos^- pole cells, it is imagined that Sxl also functions to regulate the translation of mRNAs encoding proteins critical to migration or cell cycle control. An important goal for future study will be the identification of these Sxl targets (Deshpande, 1999).

GENE STRUCTURE

Exons - three

Bases in 3' UTR - 874

PROTEIN STRUCTURE

Amino Acids - 400

Structural Domains

The C-terminal region of Nanos is homologous to Xcat-2, an RNA binding protein of Xenopus. Although NOS has a much higher molecular weight than X-cat2, the Nanos and X-cat2 shared domain forms a putative zinc finger, potentially able to bind to either RNA or protein (Mosquere, 1993).

Analysis of nanos mutants reveals that a small, evolutionarily conserved, C-terminal region is essential for Nanos function in vivo, while no other single portion of the Nanos protein is absolutely required. Within the C-terminal region are two unusual Cys-Cys-His-Cys (CCHC) motifs that are potential zinc-binding sites. One equivalent of zinc is bound with high affinity by each of the CCHC motifs. nanos mutations disrupting metal binding at either of these two sites in vitro abolish Nanos translational repression activity in vivo (Curtis, 1997).

date revised: 12 September 98

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.