nanos
Maternal effect genes, including oskar, define a shared pathway
leading to the provision of two determinants at the posterior pole of the embryo. One determinant
is the posterior body patterning morphogen Nanos, and the other directs germ cell formation.
Overexpression of oskar causes the shared pathway to be hyperactivated, with excess nanos
activity present throughout the embryo and a superabundance of posterior pole cells. In addition,
presumptive pole cells appear at a novel anterior position. Formation of these ectopic
pole cells is enhanced in nanos mutants, possibly reflecting competition between nanos and the germ cell determinant for a shared and limiting precursor (Smith, 1992).
Misexpression of NOS protein at the anterior of the embryo inhibits
translation of Bicoid mRNA, thereby suppressing head and thorax development (Gavis, 1994).
In Drosophila embryos, graded activity of the posterior determinant Nanos generates abdominal
segmentation by blocking protein expression from maternal transcripts of the Hunchback gene. When active
inappropriately at the anterior pole, NOS can also block expression of the anterior determinant Bicoid. Both regulatory interactions are mediated by similar sequences in the 3' untranslated region of each
transcript. These NOS response elements (NREs) are both necessary and sufficient to confer nos-dependent
regulation, the degree of regulation determined by the number and quality of the elements and the level of NOS in
vivo. Based on these and other results, it can be argued that NOS acts as a morphogen, controlling HB expression (and
hence abdominal pattern) as a function of its concentration-dependent interaction with the NREs. Thus, it would seem that the requirements for NOS mRNA localization, involving proteins of the posterior group are imposed by the presence of the NRE in BCD and HB mRNAs (Wharton, 1991).
giant is expressed in two broad gradients in precellular embryos, one in
anterior regions and the other in posterior regions.
GT patterns overlap with protein gradients specified by the gap genes hunchback and knirps.
Maternal factors are responsible for initiating gt expression, while
gap genes participate in the subsequent refinement of the pattern. The maternal morphogen Bicoid
initiates the anterior gt pattern, while nanos, by inhibiting hunchback, plays a role in the posterior pattern (Kraut, 1991).
High local concentrations of NOS protein in the posterior of the embryo are necessary to inhibit translation of the maternal transcript of hunchback in this region, and thus permit expression of genes required for abdomen formation. The restriction of HB protein expression permits the ordered expression of Krüppel, knirps and giant, three genes specifying the abdominal region (Gavis, 1994).
Primitive insects develop segmentation by heterchronic gene activation. Each posterior segment forms independently in a time dependent manner. nanos activity eliminates hunchback from the posterior precellular blastoderm region of primitive insects, and caudal, otherwise repressed by hunchback, can function there at a later stage. caudal proceeds to drive the addition of new segments over time. Thus, a gradient of Caudal would regulate the asynchronous segment development of primitive insects (Rivera-Pomar, 1995).
The closely linked POU domain genes pdm-1 and pdm-2 are first expressed early during
cellularization in the presumptive abdomen in a broad domain that soon resolves into two stripes. The broad abdominal domain
of pdm-1 protein is lacking in nanos- mutant embryos, and ectopic pdm-1 expression in nanos-
embryos leads to a partial restoration of abdominal segmentation (Cockerill, 1993).
In Drosophila, primordial germ cells (PGCs) are set aside from somatic cells and subsequently migrate through the embryo and associate with somatic gonadal cells to form the embryonic gonad. During larval stages, PGCs proliferate in the female gonad, and a subset of PGCs are selected at late larval stages to become germ line stem cells (GSCs), the source of continuous egg production throughout adulthood. However, the degree of similarity between PGCs and the self-renewing GSCs is unclear. Many of the genes that are required for GSC maintenance in adults are also required to prevent precocious differentiation of PGCs within the larval ovary. Following overexpression of the GSC-differentiation gene bag of marbles (bam), PGCs differentiate to form cysts without becoming GSCs. Furthermore, PGCs that are mutant for nanos (nos), pumilio (pum) or for signaling components of the decapentaplegic (dpp) pathway also differentiate. The similarity in the genes necessary for GSC maintenance and the repression of PGC differentiation suggest that PGCs and GSCs may be functionally equivalent and that the larval gonad functions as a 'PGC niche' (Gilboa, 2004).
GSC differentiation is repressed by extrinsic factors, such as Dpp, and also by intrinsic factors. To further test whether PGCs employ the same mechanisms as GSCs to repress differentiation, larval ovaries were examined that were mutant for the translational repressors Nanos (Nos) and Pumilio (Pum), which function within GSCs to repress their differentiation. Indeed, nos mutant LL3 gonads contained many developed cysts. pumilio (pum) mutant gonads also contained cysts, although less so than nos mutants. Gonads that were mutant for both nos and pum did not contain more cysts than gonads that were mutant for nos alone. Because the alleles that were used were very strong, this suggests that nos and pum function together in the repression of PGC differentiation (Gilboa, 2004).
In adult ovaries, the differentiation of cysts requires Bam, and increasing amounts of Bam are present during each subsequent mitotic division. A reporter construct of GFP under control of the bam promoter was used to follow bam expression in the larval cysts. Cysts found in nos ML3 larval gonads also expressed higher amounts of GFP as compared to single PGCs. As in adults, the intensity of GFP labeling corresponds to the developmental state of the cyst. In addition to precocious differentiation, nos mutant germ cells displayed aberrations in the shape of the branched fusome and increased amount of small fusomal material as compared with wild-type. It is concluded that both Nos and Pum, which are required for GSC maintenance, are also required to repress PGC differentiation (Gilboa, 2004).
To further test for a possible partnership between nos and pum in GSC maintenance, the time at which nos or pum mutant germ line clones, generated by the FLP-FRT method, were eliminated from the adult ovary was examined. In wild-type, clones of unmarked GSCs were induced in about 25% of the ovarioles and that percent decreased only slightly during the course of the experiment, probably due to the natural rate of GSC loss. nos and pum mutant GSCs, in contrast, were lost rapidly. GSC loss was observed as early as 4 days after clone induction, and by the 6th or 7th day, most ovarioles did not contain a mutant GSC. The striking similarity in the profiles of nos and pum GSC loss therefore suggests that these genes also function together within GSCs (Gilboa, 2004).
As of the fifth and sixth day after clone induction, it was found that many nos mutant cysts were eliminated from the ovary. These results agree with the death of cysts observed in nos and pum mutants and with the death of nos cysts in pupal ovaries, which may be the cause of the empty ovarioles observed in adult nos females. These results and the previously reported phenotypes of nos and pum suggest that these genes are continually required throughout germ cell life. In the embryo, nos and pum are required for correct migration, transcription, and viability. During larval stages, they are required for the repression of PGC differentiation and, in the adult, for the maintenance and viability of GSCs as well as for the viability of differentiating cysts (Gilboa, 2004).
The targets of Nos and Pum within GSCs remain elusive, and the relationship of these 'intrinsic' GSC maintenance factors to the 'extrinsic' Dpp signal is unclear. To test if Dpp could function partly through Nos, the Nos expression pattern was examined in wild-type and in tkv-mutant GSCs. In wild-type germaria Nos is expressed at intermediate levels in GSCs and their immediate daughters, at very low levels during mitotic divisions of the cyst, and at very high levels in a fraction of the 16-cell cysts. This expression pattern was unchanged in tkv-mutant germ cells. Similar results were obtained for larval PGCs; Nos was expressed at intermediate levels in wild-type and tkv mutant PGCs, at lower levels in cysts undergoing mitosis, and at very high levels in 16-cell cysts. This suggests that Nos expression is independent of Dpp signaling (Gilboa, 2004).
Next, whether nos is required for Dpp function was tested, by analyzing nos mutant PGCs that were overexpressing either Dpp or TkvQD, a constitutively activated form of Tkv. In nos mutant control gonads, fragmented fusomal material as well as branched cysts could be observed. The spherical fusome within nos mutant germ cells remained small or fragmented in nos gonads overexpressing Dpp. Most strikingly, single PGC/GSC like germ cells accumulated in these gonads, and no cysts could be found. Thus, although increased Dpp signaling cannot fully counteract the nos phenotype, it does prevent precocious differentiation of nos mutant PGCs. Similar results were obtained with PGCs expressing TkvQD. In most gonads no cysts could be observed, although occasionally a small branched fusome could be detected, suggesting that Dpp signaling acts directly on PGCs, rather than via a secondary signal. The genetic data show that PGCs that are mutant for nos, can still respond to a Dpp signal, which keeps them in an undifferentiated state (Gilboa, 2004).
During larval stages, PGCs proliferate rather than differentiate. The translational repressors Nos and Pum are required to repress PGCs differentiation during larval stages. It has also been show that the Dpp pathway functions in a similar manner. Both pathways are also required for GSC maintenance. The fact that the spherical fusome remains abnormal in nos mutant gonads even when Dpp is overexpressed may suggest that some of Nos function is downstream of Dpp. However, the Nos expression data and the fact that Dpp signaling can prevent nos mutant PGCs from differentiation are more compatible with the Nos pathway playing a role upstream or in parallel to the Dpp pathway. It remains unclear how these pathways converge within germ cells (Gilboa, 2004).
Germ cells may perceive a Dpp signal from the moment they form at the posterior pole of the embryo until they differentiate to form cysts. Indeed, pMad is present in embryonic pole cells, larval PGCs and adult GSCs. Dpp signaling is not only necessary for GSC maintenance but also required continually through larval stages to actively repress PGC differentiation. Thus, the larval ovary functions in a similar manner to the adult niche with regard to Dpp-mediated repression of differentiation. During the third-larval instar, the adult somatic niche forms, and repression of PGC differentiation may then become limited to the small area of the adult ovary, allowing PGCs outside the confinement of the niche to differentiate (Gilboa, 2004).
Repression of PGC differentiation is required for about 4 days, from the end of embryogenesis to the beginning of pupa formation, whereas GSCs are maintained in the adult for many days. Differences between the 'short-term' and the 'long-term' repression of differentiation may yet be found. However, all the genes tested, dpp, bam, nos, and pum, function similarly in GSCs and PGCs. This similarity suggests that there may not be a clear transition from a 'dividing' PGC to a 'self-renewing' GSC (Gilboa, 2004).
In C. elegans, mRNA production is initially repressed in the embryonic germline by a protein unique to C. elegans germ cells, PIE-1. PIE-1 is degraded upon the birth of the germ cell precursors, Z2 and Z3. A chromatin-based mechanism has been identified that succeeds PIE-1 repression in these cells. A subset of nucleosomal histone modifications, methylated lysine 4 on histone H3 (H3meK4) and acetylated lysine 8 on histone H4 (H4acetylK8), are globally lost and the DNA appears more condensed. This coincides with PIE-1 degradation and requires that germline identity is not disrupted. Drosophila pole cell chromatin also lacks H3meK4, indicating that a unique chromatin architecture is a conserved feature of embryonic germ cells. Regulation of the germline-specific chromatin architecture requires functional nanos activity in both organisms. These results indicate that genome-wide repression via a nanos-regulated, germ cell-specific chromatin organization is a conserved feature of germline maintenance during embryogenesis (Schaner, 2003).
The C. elegans nanos homologs nos-1 and nos-2 play important overlapping roles in worm primordial germ cell development. Knockdown of both NOS-1 and NOS-2 activities results in sterile animals, a population of which exhibit germ cells that begin proliferation prematurely. Loss of NOS-2 alone can also result in ectopic localization of the primordial germ cells. H3meK4 levels were examined in Z2/Z3 chromatin of embryos from nos-1(gv5);nos-2(RNAi) hermaphrodites. Animals homozygous for the nos-1(gv5) are viable and fertile; RNAi of nos-2 in these animals results in germ cell defects. nos-1 mutant animals were therefore injected with nos-2 dsRNA, and parallel broods were assessed for sterility as adults or stained for H3meK4 in Z2/Z3. In these experiments, 59% of the offspring of the injected animals grew up to be completely sterile adults. An almost identical percentage of parallel brood embryos exhibited inappropriate staining for H3meK4 in one or both of the primordial germ cells, Z2/Z3. The C. elegans NANOS homologs, NOS-1 and NOS-2, therefore play important roles in either establishing and/or maintaining the chromatin-based mode of transcriptional repression in Z2/Z3 (Schaner, 2003).
While the progenitors of the soma and germline are also separated from each other during early embryogenesis in Drosophila, the processes that give rise to the physical segregation of these two distinct cell types in flies are quite different from those in worms. In light of these differences, it was of interest to determine if distinct chromatin states are also established in the somatic and germline nuclei of fly embryos. Staining of early Drosophila embryos with the H3meK4 antibody has revealed that there is little detectable H3meK4 in any nuclei during the rapid synchronous nuclear divisions in the center of the embryo. Similarly, no H3meK4 was detected in the nuclei of newly formed pole cells, or in somatic nuclei when they first migrate to the periphery of the embryo. H3meK4 was first detected in somatic nuclei between nuclear division cycles 12 and 13, after the nuclei around the periphery of the embryo have already undergone several rounds of nuclear division. This change in the methylation status of histone H3 is coincident with, or slightly precedes, the time when transcription is broadly upregulated in the somatic nuclei of syncytial blastoderm embryos. High levels of H3meK4 are maintained through the cellular blastoderm stage, with all somatic nuclei staining with the H3meK4 antibody (Schaner, 2003).
In contrast to the somatic nuclei, little if any H3meK4 was detected in the pole cells (marked with anti-Vasa-specific antibody) of either syncytial blastoderm or cellular blastoderm embryos. Staining of embryos at later stages revealed that H3meK4 could still not be detected in germ cells in early gastrulation stage through midgut invagination. However, once the germ cells traversed the midgut wall and began migrating away from the hindgut toward the somatic gonadal precursor cells, H3meK4-specific signal became readily evident. Therefore, from the time of their formation until stage 9 of embryogenesis, there is little if any H3meK4 in the germ cells' chromatin. H3meK4 becomes readily detectable in the germ cells, however, coincident with the onset of transcription at stage 9/10. The accumulation of H3meK4 in the germ cells, as in the soma, correlates temporally with transcriptional activation in these cells (Schaner, 2003).
Embryos were stained for epitopes corresponding to H4acetylK5, H4acetylK8, H4acetylK12, H3diacetyl, and H3dimeK36. Only the antibodies specific for H3diacetyl and H4acetylK5 exhibited nuclear staining above background levels in early embryonic stages, and the appearance of these epitopes was temporally identical to that observed for H3meK4. In contrast to the H3meK4 modification, however, no differences were observed in the levels of these modifications in pole cell nuclei compared to somatic nuclei. This indicates that a global lack of H3meK4, in combination with the presence of other modifications that correlate with transcriptional competence, is a conserved feature of chromatin structure in transcriptionally inert germ cell nuclei. (Schaner, 2003).
nos function is required for either the establishment or maintenance of transcriptional quiescence in newly formed Drosophila pole cells. In light of results showing a premature appearance of H3meK4 in C. elegans germ cells in the absence of nos activity, an obvious question is whether nos is also required to block this methylation in the early fly germline. To answer this question, fly embryos produced by nos mutant mothers were probed with H3meK4- and Vasa-specific antibodies. H3meK4 can be detected in the pole cells of stage 4 nos− embryos at nuclear division cycle 14. In contrast, H3meK4 is completely absent from pole cell chromatin in wild-type embryos at this stage. Moreover, H3meK4-specific signal was also detected in pole cells in stage 5 embryos that have just initiated gastrulation. It is interesting to note that not all nos− pole cells (27/42) display H3meK4-specific staining. This is consistent with findings that transcription is not activated in every pole cell of nos− blastoderm stage embryos and supports the suggestion that additional factors contribute to the establishment/maintenance of transcriptional quiescence in fly pole cells (Schaner, 2003).
While methylation of lysine 4 of histone H3 is a conserved mark of transcriptionally competent or active chromatin, methylation of lysine 9 of histone H3 (H3meK9) is a highly conserved modification that is enriched in silenced genomic regions, such as either facultative or constitutive heterochromatin. Consequently, it was of interest to determine if the transcriptionally quiescent pole cell nuclei are enriched in H3meK9 as compared to the transcriptionally active somatic nuclei. Drosophila embryos were probed with antibodies against H3meK9 and Vasa to mark the pole cells. In contrast to H3meK4, cleavage stage Drosophila embryonic nuclei exhibited readily detectable H3meK9 even prior to their migration. The modification was also detected in somatic nuclei after nuclear migration, where it appeared enriched in nuclear regions adjacent to the periphery of the embryo that correspond to telomeric and centromeric heterochromatin in the Rabl configuration. These regions, conversely, were observed to lack H3meK4. H3meK9 is also detected in pole cells as soon as they are formed, with the level of H3meK9 antibody staining in the pole cell nuclei being considerably higher than in the somatic nuclei. The observed enrichment of H3meK9 in the pole cell nuclei coincided with loss of the mitosis-specific modification, H3 phospho-Ser10, suggesting that a specific remodeling of the pole cell chromatin accompanies their exit from the cell cycle (Schaner, 2003).
Since depletion of maternal nos activity results in a premature increased accumulation of H3meK4 in early pole cells, whether nos− pole cells exhibit a parallel reduction in H3meK9 was investigated. Indeed, nos− pole cell nuclei do not exhibit the characteristic enrichment for H3meK9-specific signal seen in wild-type embryos. Furthermore, more than 50% of the pole cells examined show considerably reduced staining specific for nuclear H3meK9 compared to the neighboring somatic nuclei. The known dichotomous roles for H3meK4 and H3meK9 in the regulation of transcriptional competence are therefore conserved in pole cells, and their relative abundances in pole cell chromatin are responsive to nos activity (Schaner, 2003).
Identifying the mechanisms that guide the separation of somatic and germ lineages is one of the oldest pursuits in developmental biology. How the totipotent germline is maintained during development has become increasingly relevant to modern science in the search for conserved mechanisms guarding stem cell identity. These data provide new evidence for at least two modes of germline-specific repression that guard the germline during C. elegans embryogenesis, one of which is conserved in Drosophila. In the earliest phase, maternal PIE-1 activity in the germline P blastomeres prevents mRNA production through a mechanism that does not involve substantial, germline-specific alterations in chromatin architecture. After the degradation of PIE-1, however, a second mode involving a dramatic and specific remodeling of chromatin arises in the germ cell precursors, Z2/Z3. Whereas the former mode appears to be unique to C. elegans, the latter mode, which bears the hallmarks of direct transcriptional repression provided by a specific mode of chromatin organization, is a conserved feature of germ cell maintenance (Schaner, 2003).
In both worms and flies, lineage restriction to germ cell fate is marked by a global absence of H3meK4 and a more condensed chromatin structure. In the case of C. elegans, the absence of H3meK4 is the result of a specific depletion; in Drosophila the absence arises from preventing H3meK4 accumulation. This absence is maintained in both organisms until zygotic activation of the genome in germ cells. Premature activation, marked by premature accumulation of H3meK4, results in the loss of the germ cell lineage in both worms and flies. In both species, the lack of H3meK4 occurs in the presence of high levels of other histone modifications that often correlate with an open chromatin configuration, indicating a 'dominance' of H3meK4 as an indicator of transcriptional activity (Schaner, 2003).
The conserved correlation between H3meK4 presence and absence and global activation and silencing, respectively, is striking. The inactive X chromosome in mammals is globally depleted of H3meK4, as are silenced regions of the genome in fission yeast. In Tetrahymena, the germinal micronucleus is transcriptionally silent and H3meK4 is missing. Sexual conjugation causes transformation of the micronucleus into a transcriptionally active somatic macronucleus, which becomes enriched in H3meK4. Thus, even in protozoans, genome-wide diminishment of H3meK4 is a property of the 'germline' and its accumulation accompanies 'somatic' activation (Schaner, 2003).
The lack of H3meK4 in fly pole cells is mirrored by enrichment for H3meK9. This is not observed in Z2/Z3 in C. elegans. The reason for this difference is not understood but may reflect a more substantial role for this modification in genome regulation in Drosophila, which has a more highly repetitive genome than that of C. elegans, and consequently more abundant classically defined heterochromatin. Indeed, in contrast to Drosophila, H3meK9 is only cytologically enriched at telomeres in C. elegans embryos and on the highly condensed, unpaired X chromosome during male meiosis. In both examples, the enrichment for H3meK9 is always mirrored by the absence of H3meK4. (Schaner, 2003).
Posttranscriptional regulation plays a major role in maintaining the embryonic germline in both worms and flies. The conserved classes of proteins involved include the pumilio (Pufs) and nanos families. nanos activities are similarly required for proper migration of the primordial germ cells, maintenance of their mitotic quiescence, proper proliferation of the germline after hatching, and, as shown in this report, the regulation of chromatin organization in Z2/Z3 in both species. In addition, nanos has also been shown to be required to maintain transcriptional quiescence in pole cells. Similar nanos functions may also be required in mammals, where two of three nanos homologs are required for fertility. The phenotype of nanos-3 null mice is consistent with specification of primordial germ cells occurring normally, followed by an inability to maintain germ cell identity during migration. This phenotype is strikingly similar to those observed in nanos(-) worms and flies (Schaner, 2003).
nanos function is essential to maintain embryonic germline repression prior to normal activation of proliferation, presumably through its characterized roles in posttranscriptional regulation. Only a few direct targets of nanos regulation in germ cells have been identified, including maternal cyclin mRNAs. Targets that are the effectors of H3meK4 addition have yet to be identified, but conceivably include an H3 lysine 4-specific methyltransferase. The phenotypes observed upon removal of nanos are also only partially penetrant in both organisms. It is thus likely that other conserved, partially redundant systems exist. A clear candidate is germ cell-less (gcl), which is required in Drosophila to maintain transcriptional quiescence. Preliminary experiments suggest that H3meK4 regulation is also disrupted in gcl mutants. A predicted C. elegans gene with substantial homology to gcl has been identified, but its role in these processes has not as yet been assessed (Schaner, 2003).
It appears clear in flies that nos is required to maintain the repressive state of the germline and that part of this involves preventing accumulation of H3meK4. The pole cells begin life lacking this modification, and Nanos function is required to prevent its addition to pole cell chromatin. It is not as clear-cut in worms, since H3meK4 is initially present and then removed at germline restriction. The detection of H3meK4 in Z2/Z3 of nos-depleted embryos could therefore conceivably reflect a role for nos in either (presumably indirectly) promoting the removal of H3meK4 after PIE-1 and/or preventing the re-addition of H3meK4 before hatching. Given the conservation of most other phenotypes caused by the loss of nos activity in multiple organisms, the latter role for C. elegans NOS proteins is favored (Schaner, 2003).
At approximately the 100-cell stage in C. elegans, PIE-1 levels decrease and this is concurrent with a global loss of a distinct subset of histone modifications, H3meK4 and H4acetylK8, in Z2/Z3. The presence of other histone tail modifications in Z2/Z3 indicates that the loss of H3meK4 and H4acetylK8 epitopes are not due to a general cleavage of histone tails, but rather a specific chromatin-remodeling event. While removal of H4acetylK8 is likely to be performed by a histone deacetylase (HDAC), an activity that demethylates lysines has yet to be identified in any organism. The loss of H3meK4 could therefore represent either a replication-independent or a replication-coupled replacement of histone H3, perhaps by a germline-specific H3 variant. A diminishment of H3meK4 in P4 after gastrulation and prior to its entry into mitosis, which could indicate an S phase-related event, is frequently observed (Schaner, 2003).
nanos (nos) specifies posterior development in the
Drosophila embryo by repressing the translation of maternal hb
mRNA. In addition to this somatic function, nos is required in the
germline progenitors, the pole cells, to establish transcriptional quiescence.
nos has been shown to be required to keep turned off in the germline of
both sexes the Sex-lethal establishment promoter, Sxl-Pe. nos also functions to repress Sxl-Pe activity in the surrounding soma. Sxl-Pe is inappropriately activated in the soma of male embryos from nos mothers, while Sxl-Pe can be repressed in female embryos by ectopic Nos protein. nos appears to play a global role in repressing transcription in the
soma since the effects of nos on promoter activity are correlated with
changes in the phosphorylation status of the carboxy terminal domain (CTD)
repeats of the large RNA polymerase II subunit. Finally, evidence is presented
indicating that the suppression of transcription in the soma by Nos protein is
important for normal embryonic development (Deshpande, 2005).
During the rapid nuclear division cycles in cleavage stage Drosophila
embryos, RNA polymerase II transcription is largely shut down and only a few
genes are actively transcribed. RNA polymerase II transcription in somatic
nuclei is upregulated soon after they migrate to the periphery of the embryo at
stage 9 and by nuclear cycle 10 and 11 many of the key segmentation genes are
already actively transcribed.
While RNA polymerase II activity is substantially augmented when the nuclei
reach the periphery of the embryo in the soma, the opposite occurs in the
germline pole cell nuclei. When these nuclei migrate into the posterior pole
plasm and pole cells are formed, transcription is shut down rather than
activated. Previous studies have implicated the posterior determinants
nos and pum in establishing/maintaining transcriptional
quiescence in pole cells. In embryos derived
from mothers mutant for either nos or pum, RNA polymerase II
transcription is not properly downregulated in the pole cells and several genes
that are normally active only in somatic nuclei are ectopically expressed (Deshpande, 2005).
Since only nos mRNA localized at the posterior pole is translated, pole cells have
the highest levels of Nos. However, translation of the localized message
generates a Nos gradient that extends to the center of the embryo. An obvious
question is whether this Nos gradient also affects RNA polymerase II activity
in somatic nuclei. Indeed, transcription of Sxl-Pe is upregulated in
somatic nuclei when Nos protein is removed, and is repressed when Nos protein
is ectopically expressed. The role of Nos protein in repressing transcription
is not restricted to the sex determination pathway since the activity of other
promoters also appears to be increased in the absence of nos function (Deshpande, 2005).
Several mechanisms
could potentially explain the ectopic activation of Sxl-Pe in the soma
and germline of nos mutant embryos. The most obvious is that this
promoter is turned-on by maternal Hb expressed in the absence of nos.
However, Sxl-Pe is upregulated in nos− embryos
even when maternal Hb is eliminated. In addition, ectopic expression of Hb from
a transgene lacking NREs seemed to repress rather than activate Sxl-Pe.
Another possibility is that the zygotic expression of one or more
of the X-linked numerators is elevated in nos embryos, upsetting X
chromosome to autosome counting. However, since none of the known numerators has a
recognizable NRE in the 3′UTR of its message, it seems unlikely that
these genes are direct targets for translational repression by Nos protein. In
addition, it is not at all clear why numerator genes (which are transcribed in
the zygote) would be subject to translational repression by Nos, while
autosomal denominator genes such as deadpan (which turns off
Sxl-Pe) would not (Deshpande, 2005).
For this reason, the idea is favored that
Sxl-Pe is activated in nosm− embryos at
least in part because RNA polymerase II activity is upregulated. Support for
this idea comes from analysis of CTD phosphorylation. When RNA polymerase is
transcriptionally engaged the CTD domain is phosphorylated on serine 2 and 5.
In wild-type pole cells, phospho-ser2 cannot be detected, while there is only
little phospho-ser5. In contrast, phospho-ser2 is found in
nosm− pole cells, while the level of
phospho-ser5 is increased. nos-dependent alterations in CTD
phosphorylation are also evident in the soma. When Nos is absent, the level of
ser2 and ser5 CTD phosphorylation is elevated, while both types of CTD
phosphorylation are reduced by ectopic Nos protein (Deshpande, 2005).
Additional evidence that nos has a global effect on transcription comes from the finding that nos regulates the methylation of histone H3 in the germline of worms and flies. In both
organisms, the methylation of histone H3 on lysine 4 (H3meK4) is upregulated in
the soma when zygotic transcription commences in early embryogenesis. In
contrast, little or no methylation H3 K4 is observed in the transcriptionally
quiescent germline. Inhibition of H3 K4 methylation in germ cells requires
nos and H3meK4 is markedly upregulated in nos−
germ cells. In light of these findings, K4 methylation was examined in the soma of
nosm− embryos. As might be expected from the
effects of nos on CTD phosphorylation, somatic H3meK4 is elevated
compared to wild type (Deshpande, 2005).
Since phosphorylation of
serines 2 and 5 are correlated with transcription, the nos-dependent
alterations in CTD phosphorylation are consistent with the idea that nos
has a global impact on RNA polymerase II activity. If this is the case, an
important question is whether CTD phosphorylation is the cause or the
consequence of nos induced changes in the activity of the
transcriptional apparatus. Because actively transcribing RNA polymerase has a
hyperphosphorylated CTD domain, any mechanism, which leads to a general
increase (or decrease) in transcription, would likely alter the level of CTD
phosphorylation. This makes it difficult to distinguish between cause and
effect. In contrast, besides being a characteristic feature of elongating
polymerase, CTD phosphorylation has been linked to the last steps in the
initiation process, promoter clearance and the formation of an elongation
competent RNA polymerase complex. Moreover, there is growing evidence that
these steps in the transcription cycle are subject to regulation (Deshpande, 2005).
The fact that CTD
phosphorylation may be a key control point in the transcriptional cycle raises
the possibility that nos exerts its effects on polymerase activity by
inhibiting the translation of some factor which promotes CTD phosphorylation.
In nos mutants, the level of this factor would increase, leading to a
general derepression of transcription. Conversely, the level of this factor
would decrease by ectopic Nos, reducing overall transcription (Deshpande, 2005).
While the results clearly
show that Sxl-Pe is inappropriately turned on in the soma of male
embryos and upregulated in the soma of female embryos in the absence of
nos activity, it was initially surprising to find that there is usually
not a very pronounced posterior-anterior activation gradient. In fact,
the smaller Sxl-Pe0.4 kb promoter is clearly
activated not only in the posterior but also in the anterior of nos
embryos, while the larger Sxl-Pe3.0, usually shows at
most only a very shallow posterior–anterior gradient of
β-galactosidase expression. Since the Nos gradient does not extend beyond
the midpoint of the embryo, and the repressive effects of Nos on hb mRNA
translation are restricted to this posterior domain, one might have expected
that the activation of Sxl-Pe in would be tightly restricted to the
posterior half of nos mutant embryos. However, proteins of average size
would be expected to diffuse (in water or even in cytoplasm) through the volume
of a fly embryo over a time scale of minutes, and the establishment of
gradients like those seen for Nos or Bcd are likely to require special
mechanisms including a localize source of product, as well as the sequestration
(e.g. nuclear localization) and degradation of the product. If the Nos target
for inhibiting general RNA Pol II activity is translated from a uniformly
distributed maternal mRNA and is able to equilibrate through the embryo during
the time between the onset of the very rapid nuclear divisions and the
formation of the cellular blastoderm, only a shallow gradient of this factor in
the soma might be expected at any one time in the presence or absence of Nos.
In contrast, in pole cells, where repression of this factor by Nos would
presumably be required to impose transcriptional quiescence, the formation of
the cell membrane would prevent factor synthesized in the soma from influencing
polymerase activity. This would enable Nos in the pole cells to reduce the
level of this factor below the threshold required for transcriptional
activation (Deshpande, 2005).
As observed in many species, establishing transcriptional
quiescence in the newly formed pole cells during early embryogenesis is a
critical step in the development of the Drosophila germline. However, it
is not immediately obvious what role nos mediated down regulation of
polymerase activity would have in the development of the soma. Obviously,
hyperactivation of Sxl-Pe in nosm−
embryos could inappropriately switch on the Sxl autoregulatory feedback
loop in males. However, analysis of Sxl accumulation in post-blastoderm
stages suggests that only very few nosm− 1X/2A
embryos actually make the wrong choice in sexual identity. There is also little
evidence of a sex-bias in adult progeny of nos−
hb− germline clone mothers. The fact that the
Sxl autoregulatory loop is usually not activated in male
nosm− embryos, which are hemizygous for
Sxl, is not altogether surprising. In females that have only a single
wild type Sxl gene, activation of the autoregulatory loop is severely
compromised by conditions which diminish Sxl-Pe activity. Since the amount of Sxl
produced by Sxl-Pe in nosm− male embryos
is much less than that in wild-type females, the autoregulatory loop should be
activated infrequently (Deshpande, 2005).
While nosm− males
largely escape the effects of activating Sxl-Pe, the increased
polymerase activity appears to have other consequences. Previous studies have
shown that removal of maternal hb suppresses the posterior defects of
nosm− embryos. However, it has been shown that only about 40% of
the hb+/hb− embryos from
nos− hb− mothers survive to
adults. Likewise, it was found that only 60% of the embryos produced by
hb− nos− mothers hatch as
first instar larva, and that an even lower number survive to the adult stage.
Taken together, these experiments argue that nos has important functions
in the soma besides blocking translation of maternal hb mRNA. It seems
possible that the segmentation/developmental defects evident in progeny of
hb− nos− clone mothers could
arise from the upregulation of various patterning genes in the absence of
nos activity. It is presumed that in wild-type embryos the activity of the
transcriptional apparatus and of target zygote promoters is appropriately
adjusted to compensate for the repressive effects exerted by Nos. Because the
transcriptional apparatus is hyperactivated in the absence of Nos function,
this balance is perturbed and many genes are overexpressed (Deshpande, 2005).
In Drosophila, the maternal terminal system specifies cell fates at the
embryonic poles via the localised stimulation of the Torso receptor tyrosine
kinase (RTK). Signalling by the Torso pathway relieves repression mediated by
the Capicua and Groucho repressors, allowing the restricted expression of the
zygotic terminal gap genes tailless and huckebein. This study reports a
novel positive input into tailless and huckebein transcription
by maternal posterior group genes, previously implicated in abdomen and pole
cell formation. Absence of a subset of posterior group genes, or
their overactivation, leads to the spatial reduction or expansion of the
tailless and huckebein posterior expression domains, respectively.
The terminal and posterior systems converge, and
exclusion of Capicua from the termini of posterior group mutants is ineffective,
accounting for reduced terminal gap gene expression in these embryos. It is proposed
that the terminal and posterior systems function coordinately to alleviate
transcriptional silencing by Capicua, and that the posterior system fine-tunes
Torso RTK signalling output, ensuring precise spatial domains of tailless
and huckebein expression (Cinnamon, 2004).
Terminal gap gene expression must be tightly regulated for the correct
specification of terminal cell fates at the nonsegmented poles. Clearly, the Tor
pathway plays a key role in driving tll and hkb transcription,
given that terminal gap genes are not expressed at the posterior end of terminal
group mutants, and as a result terminal structures such as the terminal filzkorper (FK)
do not form. In this paper, a novel biological role is unraveled for the maternal posterior system, showing that members of this group, in particular Nos, positively
regulate transcription of the zygotic subordinate genes of the terminal system.
Torso response elements (TREs) in the tll upstream regulatory region, which are derepressed in cic mutants, also respond to alterations in maternal
osk dosage, and the Cic repressor is not excluded from the termini
of posterior group mutants. These results are consistent with the posterior system
feeding into the Tor signalling pathway, upstream of or at the level of the Cic
repressor. It is suggested that the concerted
activities of both the terminal and posterior systems, in their spatially
overlapping zones of action, generate accurate domains of terminal gap gene
expression at the posterior (Cinnamon, 2004).
It was originally proposed that the four
maternal systems that pattern the early Drosophila embryo act largely
independently of each other. Recent work, however, demonstrated interactions between the Tor
pathway and the anterior and D/V systems. For example, tll has been shown
to respond to the anterior determinant Bicoid (Bcd) even when Tor signalling is
genetically blocked. Indeed, cis-acting DNA elements responsive to these
three maternal systems have been found in the tll upstream regulatory
region. The current results now link the
terminal and posterior systems, previously thought to be independent of each
other, in terminal gap gene regulation, reinforcing the idea that maternal
systems that pattern the early embryo act in a coordinated manner (Cinnamon, 2004 and references therein).
Why has the positive input, by posterior group genes into terminal
patterning, been largely overlooked to date? Classical segmentation studies
mostly involved phenotypic analyses at the cuticular level. For this reason, and
when taking into account the primary contribution of the terminal system, the
delicate input by the posterior group has gone unnoticed. Thus, the unextended
FK that develops in posterior group mutant background, which may arise from
decreased terminal gap gene expression, had largely been attributed to
pleiotropic effects arising from abdominal defects. It has been possible to detect
the relatively subtle changes in tll and hkb gene expression
patterns only by investigating terminal gap gene regulation at the molecular
level. In fact, at least one other molecular study had previously reported
reduced terminal gap gene expression in osk mutant embryos (Cinnamon, 2004 and references therein).
One emerging concept is that, for the refinement of the expression
levels and spatial extents of RTK signalling targets, it is also imperative to
integrate accurately information originating from other, non-RTK sources.
In many cases this integration occurs at the level
of target gene enhancers, with various effectors of distinct signalling pathways
binding to specific DNA elements to regulate transcription. For example, D-Pax2
expression in the cone and pigment cells of the developing eye is regulated by
effectors of the EGFR RTK pathway, such as Pointed P2 and Yan, and also by the
Notch signalling component Suppressor of Hairless, as well as by the
transcription factor Lozenge.
The current study shows that terminal gap gene expression requires not only Tor RTK
pathway activity but also a contribution from the posterior system. In this
instance, inputs from these two maternal coordinate systems are interpreted and
linked not at the level of terminal gap gene promoters but at the level of the
Cic repressor. Thus, Cic functions as an integrator of multiple regulatory
inputs, with both the posterior and terminal systems acting to relieve
transcriptional silencing mediated by this repressor (Cinnamon, 2004).
Surprisingly, anterior tll and hkb
expression is also reduced in posterior group mutants. Similarly, others have
reported prolonged bcd expression and head defects in pum mutants.
It is speculated that low
levels of Osk and Nos, which escape translational repression, similarly regulate
terminal gap gene expression via Cic removal at the anterior. In accordance with
this, the dismissal of Cic from the anterior pole of posterior group mutants is
also ineffective (Cinnamon, 2004).
How does Nos, which has been assigned the role of a translational repressor, positively
regulate tll and hkb transcription? The results suggest that Nos
does so indirectly, by downregulating the accumulation of the Cic repressor at
the termini. The exact mechanism by which the Tor pathway mediates the exclusion
of Cic from terminal regions has not been established, but one model argues that
phosphorylation of Cic by MAPK causes degradation of the protein, as in the case
of Yan. Thus, Nos could be affecting this process in one of
several possible ways, at the level or downstream of MAPK. For example, Nos
could be facilitating the translocation of phosphorylated MAPK into the nucleus.
In posterior group mutants, then, activated MAPK would remain in the cytoplasm
rather than enter the nucleus, impeding Cic phosphorylation and degradation.
Alternatively, Nos may be modulating MAPK activity, or regulating adaptor
proteins that promote Cic phosphorylation by nuclear MAPK. Nos may also be
controlling the translation of factors that are involved in the nuclear
trafficking (import/export) or degradation of Cic, or perhaps may even be acting
on the cic message itself. Future studies will distinguish between these
possibilities, and may shed new light on the molecular mechanisms underlying role of
Nos in other developmental processes, for example, the
establishment/maintenance of transcriptional quiescence in pole cells. The positive
input by the posterior group genes is viewed as evolving to modulate terminal pathway
activity, merging with other varied modes of Tor regulation to ultimately ensure
accurate tll and hkb expression and, consequently, precise cell
fate determination (Cinnamon, 2004).
The Tor signal transduction pathway is
under multiple tiers of regulation, outside and inside the nucleus. For
instance, internalisation and trafficking of the activated Tor receptor to the
lysosome for degradation attenuates the signal, as evident by the spatial
broadening and temporal prolonging of Tor activation in mutants for hrs,
a component of the endosomal recycling machinery (Lloyd. 2002).
Yet another level of control is provided by the tyrosine
phosphatase corkscrew, which sharpens the gradient of Tor activity.
Additionally, multiple cytoplasmic
adaptor proteins take part in transducing the Tor signal, conceivably buffering against
surplus or deficiency in signalling (Cinnamon, 2004).
In the nucleus,
tll and hkb are subjected to silencing by several repressors.
Derepression of tll is observed in grainy-head and
tramtrack69 (ttk69) mutants, and the proteins encoded by these
genes bind tll promoter sequences. Cic and Gro appear to
play a leading role in terminal gap gene silencing, given that mutations in
cic and gro bring about a significant expansion of the tll
and hkb expression domains.
Intriguingly, however, tll expression never reaches
the middle of the embryo in these mutants. tll is uniformly
expressed, albeit weakly, throughout the embryo only when both the developmental
corepressors Gro and CtBP are removed concomitantly.
This broadened tll expression likely stems from the fact that
there is a redundancy in the activities that normally restrict terminal gap gene
transcription from inappropriately spreading into the central portion of the
embryo; by jointly removing the Gro and CtBP coregulators, activity of the above
repressors is compromised. Alternatively, CtBP might be acting in conjunction
with a novel, unidentified repressor that prevents tll transcription in
the middlemost region of the embryo (Cinnamon, 2004).
So what is the purpose
of the input by the posterior group genes into tll and hkb
transcription? Quantitative differences in Tor receptor activity have to be
eventually interpreted and translated into distinct cell fates at the termini.
Strong Tor activation induces both hkb and tll expression, whereas
weaker Tor activation only brings about tll expression. It is surmised that
the precision endowed by the Tor RTK cascade may not suffice for the complex
patterning of the termini, given that mere two-fold fluctuations in Tor
signalling result in defective embryonic development. For example, mutants with reduced
Tor RTK activity show partial tll expression and the complete loss of
hkb. These mutants consequently develop incomplete terminal structures
and die at the larval stage. Conversely, overactivation of the Tor pathway leads
to anterior expansion of the posterior tll expression domain, perturbing
segmentation in central body parts, likely as a result of downregulation of
abdominal gap genes by the Tll protein. Thus,
the precise spatial confinement of terminal gap gene expression domains requires
the coordinated integration of regulatory inputs, coming from two maternal
systems and converging on the same effector protein, Cic (Cinnamon, 2004).
In the Drosophila embryo, Nanos and Pumilio collaborate to repress the translation of hunchback mRNA in the somatic cytoplasm. Both proteins are also required for repression of maternal Cyclin B mRNA in the germline; it has not been clear whether they act directly on Cyclin B mRNA, and if so, whether regulation in the presumptive somatic and germline cytoplasm proceeds by similar or fundamentally different mechanisms. This report shows that Pumilio and Nanos bind to an element in the 3' UTR to repress Cyclin B mRNA. Regulation of Cyclin B and hunchback differ in two significant respects. (1) Pumilio is dispensable for repression of Cyclin B (but not hunchback) if Nanos is tethered via an exogenous RNA-binding domain. Nanos probably acts, at least in part, by recruiting the CCR4-Pop2-NOT deadenylase complex (see twin), interacting directly with the NOT4 subunit. (2) Although Nanos is the sole spatially limiting factor for regulation of hunchback, regulation of Cyclin B requires another Oskar-dependent factor in addition to Nanos. Ectopic repression of Cyclin B in the presumptive somatic cytoplasm causes lethal nuclear division defects. It is suggestd that a requirement for two spatially restricted factors is a mechanism for ensuring that Cyclin B regulation is strictly limited to the germline (Kadyrova, 2007).
Thus Nos and Pum directly regulate maternal CycB mRNA, binding to an NRE in its 3' UTR. Differences in the spacing and arrangement of protein-binding sites within the hb and CycB NREs appear to account for the regulation of hb but not CycB by Brat. For regulation of CycB, the main function of Pum is to recruit Nos, a role that can be bypassed by tethering Nos via an exogenous RNA-binding domain. CycB-bound Nos is then likely to act, at least in part, by recruiting a deadenylase complex, interacting with its NOT4 subunit. Regulation of CycB is limited to the PGCs to avoid the deleterious consequences of repression in the presumptive somatic cytoplasm. The requirement for both Nos plus at least one additional germline-restricted factor may be part of a mechanism to ensure that CycB regulation is strictly limited to the PGCs (Kadyrova, 2007).
The co-crystal structure of human Pum bound to a fragment of the
hb NRE shows that a single Pum RBD directly contacts eight
nucleotides of the RNA. However, Puf proteins bind with essentially wild-type affinity to many mutant sites, suggesting that all eight nucleotides are not rigidly specified. How, then, do Puf proteins
recognize specific mRNA targets in vivo (Kadyrova, 2007)?
Part of the answer appears to be that, within functional NREs, more than eight nucleotides are recognized, at least by Drosophila Pum. Mutations that simultaneously disrupt Pum binding in vitro and regulation in vivo are spread over 20 nts of the hb NRE and 18 nts of the CycB NRE. These extended Pum mutational 'footprints' are too large to be accounted for by binding of a single RBD; it is suggested that two or more Pum RBDs bind each NRE, an idea supported by the detection of two RNA-protein complexes in gel mobility shift experiments using both the CycB and hb NREs. This model disagrees with earlier experiments that suggested only a single Pum RBD binds to the hb NRE. Further biochemical and structural studies will be required to resolve the issue (Kadyrova, 2007).
The distribution of Pum- and Nos-binding sites within the CycB and hb NREs is different. In the former, the Nos binding site lies 5' to the Pum-binding site(s), whereas in the latter, the Nos-binding site is flanked by nucleotides recognized by Pum. It is assumed that the different arrangement of Nos- and Pum-binding sites is responsible for the assembly of Pum-NRE-Nos complexes with different topographies, such that Brat is recruited to hb but not to CycB. Further definition of each RNP structure will ultimately be required to understand the combinatorial assembly of different repressor complexes on each NRE (Kadyrova, 2007).
In addition to the NRE, Pum also binds with high affinity to at least two other sites in the CycB 3' UTR; however, binding to
these sites does not mediate translational repression in the PGCs, perhaps because neither supports recruitment of Nos. These sites may simply bind Pum fortuitously, or they may mediate Nos-independent regulation at other stages of development. Pum has been suggested to destabilize bcd mRNA at the anterior of the embryo in a Nos-independent manner. Another Nos-independent function of Pum is the repression of CycB translation throughout the prospective somatic cytoplasm during the early syncitial nuclear cleavages. These processes might be mediated by elements in Fragments A and F of the 3' UTR, that bind Pum but not Nos (Kadyrova, 2007).
A general framework has been provided for understanding how Puf proteins act to control either the translation or stability of target mRNAs (Goldstrohm, 2006). The yeast Puf protein MPT5 interacts directly with Pop2, one of the catalytically active subunits of a large deadenylase complex. Subsequent deadenylation could either silence the mRNA or cause its degradation, depending on other signals in the transcript or the composition of the deadenylase complex (or both). The Puf-Pop2 interaction is conserved across species (including Drosophila), supporting the idea that the mechanism uncovered for MPT5 might generally be applicable to Puf proteins (Kadyrova, 2007).
In this context, it is surprising that Pum is dispensable if Nanos is tethered to CycB via MS2 CP. It is suggested that yeast Puf proteins both recognize target mRNAs and recruit the deadenylase, but that in the Drosophila germline these functions are partitioned, with Pum primarily responsible for target mRNA recognition and Nos primarily responsible for effector recruitment. This model has the attraction of attributing an important role to Nos, which is essential for Puf-mediated regulation in Drosophila, and probably other metazoans as well. What, then, might be the role of the conserved interaction between Pum and Pop2? One possibility is that it acts cooperatively with Nos to recruit the deadenylase; unlike CycB, other mRNA targets (e.g. hb) might require recruitment by both Nos and Pum to ensure efficient deadenylation. Another possibility is that it plays an essential role for mRNAs regulated by Pum but not Nos (Kadyrova, 2007).
Oscillations in CycB activity underlie normal cell cycle progression. During the early embryonic syncitial nuclear cleavages, degradation in the vicinity of the nuclei is thought to deplete CycB locally. Recent work has shown that Pum can inappropriately repress de novo translation of CycB mRNA during the initial nuclear cleavages if not antagonized by
the PNG kinase, resulting in mitotic failure. This early Pum-dependent repression is thought to be Nos-independent, as it occurs efficiently in the anterior, where Nos activity is undetectable (Kadyrova, 2007).
The results show that if CycB is inappropriately subjected to Pum+Nos-dependent repression via the hb NRE, CycB is locally depleted, resulting in mitotic failure during nuclear division cycles 10-13. Since it is thought to be the case during the early cycles (1-7), de novo synthesis of CycB apparently is required to counteract the local degradation that probably occurs during M phase of each cycle. The CycB NRE must therefore be precisely tuned to repress translation only in the PGCs and not in the presumptive somatic cytoplasm (Kadyrova, 2007).
Osk is the limiting factor for assembly of pole plasm in the embryo; the results suggest that it stimulates the accumulation or activity of at least one factor in addition to Nos that is required for repression of CycB in the PGCs. The existence of a co-factor is inferred from the finding that ectopic Nos can repress CycB in the somatic cytoplasm only in the presence of ectopic Osk. Regulation of CycB may depend on more than one germline-restricted factor to ensure that potentially deleterious repression does not occur in the somatic cytoplasm (Kadyrova, 2007).
A germline Nos co-factor might act in a variety of ways. It could bind to the CycB NRE adjacent to Pum and Nos, substituting functionally for Brat, which is recruited to the Pum-hb NRE-Nos complex. The 50 nt CycB NRE is inactivated by a truncation at both ends that leaves the Pum- and Nos-binding sites intact, consistent with the idea that another factor binds to the element. Another possibility is that the co-factor is a germline-specific component of the adenylation/deadenylation machinery, as is the case for the GLD-2 cytoplasmic poly(A)-polymerase in C. elegans.
Distinguishing among these ideas awaits identification of the cofactor (Kadyrova, 2007).
Nanos (Nos) is a translational regulator that governs
abdominal segmentation of the Drosophila embryo in
collaboration with Pumilio (Pum). In the embryo, the mode
of Nos and Pum action is clear: they form a ternary
complex with critical sequences in the 3'UTR of Hunchback
mRNA to regulate its translation. Nos also regulates germ
cell development and survival in the ovary. While this
aspect of its biological activity appears to be evolutionarily
conserved, the mode of Nos action in this process is not yet
well understood. In this report it is shown that Nos interacts
with Cup, which is required for normal development of the
ovarian germline cells. nos and cup also interact genetically:
reducing the level of cup activity specifically suppresses
the oogenesis defects associated with the nosRC allele. This
allele encodes a very low level of mRNA and protein that,
evidently, is just below the threshold for normal ovarian
Nos function. Taken together, these findings are consistent
with the idea that Nos and Cup interact to promote normal
development of the ovarian germline. They further suggest
that Nos and Pum are likely to collaborate during
oogenesis, as they do during embryogenesis (Verrotti, 2000).
To identify proteins that interact with Nos, a
yeast two-hybrid screen was performed using full-length Nos fused to the
GAL4 DNA-binding domain as the bait. Two of the interactors
proved to be fragments of Cup, a novel cytoplasmic protein of
unknown biochemical function that is required for normal
oogenesis (Keyes, 1997). In the ovaries of cup
mutant females, egg chamber maturation arrests between
stages 5 and 14, and the nurse cells have aberrant nuclear
morphology. However, all of the extant cup alleles encode
detectable protein, and thus the null phenotype may be
stronger. The interaction with Nos appears to be specific, since
Cup fails to interact with a variety of other baits in yeast. In
particular, Cup does not interact with the RNA-binding domain
of Pum (Verrotti, 2000).
To test the interaction between Nos and Cup in vitro, the minimum region of Cup required for interaction in yeast was defined by deletion analysis. Residues
593-963 constituted the smallest fragment of Cup tested that
interacted with Nos in yeast. A GST fusion protein bearing these Cup
residues was prepared in bacteria and was incubated
with embryonic extracts from either wild-type
or transgenic flies that produced a Myc epitope-tagged Nos that was fully functional and rescued the defects in otherwise nos-
embryos and ovaries. Approximately 10% of the Nos-Myc from the
extract is retained by GST-Cup under the reaction conditions, whereas a negligible
amount of Nos-Myc is retained in a control reaction with GST. In summary, Nos appears to interact specifically with Cup in yeast and in vitro (Verrotti, 2000).
Which portion of Nos mediates the interaction with Cup?
Nos contains two regions -- a well-conserved C-terminal Zn2+-binding domain that mediates the interaction with Pum and HB mRNA, and a poorly conserved N-terminal region. The N-terminal
region of Nos mediates interaction with Cup.
No biochemical function has previously been
ascribed to this portion of Nos, which is very
poorly conserved even among closely related Dipteran homologs. Further deletion analysis of the N-terminal region reveals that it contains at least two redundant sub-domains that can interact with Cup (Verrotti, 2000).
The function of the N-terminal region of Nos in vivo is not
clear. A recent analysis of 60 nos- alleles reveals that mis-sense mutations that eliminate both ovarian and embryonic
function alter residues in the conserved C-terminal domain,
consistent with the idea that this domain is required for function in both
tissues (Arrizabalaga, 1999). In contrast, no such mis-sense mutations are found in the coding region for the poorly conserved N-terminal domain. Microinjection of
mRNAs encoding deletion derivatives of Nos into nos-
embryos suggests that no single part of the N-terminal region
is essential for regulation of HB mRNA.
No comparable analysis of Nos residues required for ovarian
function has been reported (Verrotti, 2000).
To identify residues of Nos essential for its activity in the
ovary, nos transgenes were prepared that encode deletion
derivatives. No part of the N-terminal region
of Nos is essential for Nos to function in either the embryo or the
ovary. Maternal expression from a single transgene
encoding each deletion derivative rescues the ovarian
morphology and egg-laying defects associated with
nosRC. Each deletion derivative also rescues the
abdominal segmentation defects associated with
nosBN either completely (a full complement of 8
abdominal segments) or nearly completely (6-8
abdominal segments). One derivative, deltaG, appears to
have somewhat less activity than the others; however,
expression from two maternal copies of the deltaG
transgene completely rescues abdominal segmentation in
100% of embryos. Thus, it is concluded that the N-terminal
region of Nos contains no unique sequence that is essential for
its activity in either the embryo or the ovary. This latter
observation is consistent with the finding that interaction with
Cup is mediated by redundant elements in the N-terminal
region of Nos (Verrotti, 2000).
To determine whether the interaction between Nos and Cup is
functionally significant, it was asked whether lowering the level
of Cup modified any of the ovarian or embryonic phenotypes
associated with altered Nos function. In one case,
a strong genetic interaction was observed: introduction of a single cup allele substantially suppresses the oogenesis defects in hemizygous
nosRC mutant females. Cystoblasts that give rise to the germline
components of the egg chamber do not develop normally in
nosRC mutant ovaries, and germline stem cells that give rise to
cystoblasts are not maintained. As a result, nosRC mutant
ovaries contain only rare mature egg chambers. In
contrast, in cup-/+; nosRC/Df(nos) ovaries, many of the egg chambers appear normal and mature into oocytes that are
fertilized and oviposited. (The resulting embryos
develop no abdominal segments, presumably because they
lack sufficient Nos activity to repress HB translation.) The
cup-/+; nosRC /Df(nos) females lay eggs for at least 3 weeks, suggesting that germline stem cells are maintained and
function normally. Thus, reducing the level of Cup appears to
specifically suppress the oogenesis defects associated with the
nosRC allele, but not the embryonic defects.
Nine different cup alleles tested suppress the defects
associated with nosRC, suggesting that it is simply a reduction
of Cup activity that suppresses the oogenesis phenotype. In
contrast, the genetic interaction appears to be specific to the
RC allele; the nosRD mutant encodes an unstable protein
bearing a substitution at one of the conserved Cys residues in
the C-terminal domain. This allele exhibits oogenesis defects similar to nosRC, but these defects are not ameliorated by lowering the level of Cup, presumably because the level of active Nos protein is insufficient. In addition, reducing the level of Cup has no effect on the oogenesis defects associated with two different allelic combinations of pum. Thus,
reduction of Cup activity does not appear to globally suppress oogenesis defects resulting from alterations in Nos or Pum activity, but specifically suppresses the defects associated with nosRC (Verrotti, 2000).
The genetic interaction between nos and cup
suggests that expression of the protein encoded
by each gene coincides, and previous reports
support this idea (Keyes, 1997). However, it was of interest to
visualize the distributions of Nos and Cup
simultaneously to determine whether the
spatiotemporal distribution of the proteins is
consistent with the observed genetic interaction. Using anti-Nos antibodies, Nos protein could not be reliably detected in the germarium.
Therefore, the localization of Cup and Myc-tagged Nos was examined in the transgenic flies that carry a fully functional nos+ transgene altered to encode a Myc epitope tag at the C terminus of the protein. Cup is present
throughout the cytoplasm of all the germ cells in
the germarium, the terminal region of the ovary
that contains the stem cells, cystoblasts and most
immature egg chambers. In contrast, the
Nos-Myc distribution is not uniform. It is present in the
germline stem cells and cystoblasts in region 1 of the
germarium, falls beneath the level of detection during the early
cystoblast cleavages in regions 1 and 2, rises to relatively high
levels in the germline cysts in region 2, and falls to somewhat
lower levels in the maturing cysts of region 3. The significant
finding is that Nos and Cup co-localize to the cytoplasm of
the stem cells, the cystoblasts and the cysts, consistent with the
genetic interaction described above (Verrotti, 2000).
The data reported above support the idea that Cup interacts
with the N-terminal region of Nos and thereby lowers its
activity, perhaps titrating it away from regulatory targets. This
follows from the observed physical interaction and
the genetic interaction between nosRC and cup. However, nosRC, which bears a mutation in the splice donor
of intron 1 in the pre-mRNA, has been described as a null
allele, and no mature mRNA is detectable by either in situ
hybridization or Northern blot. How then can a physical interaction
between Cup and Nos account for the genetic interaction
between cup and nosRC?
To address this question, it was asked whether nosRC actually
encodes a very low level of functional protein. Semi-quantitative RT-PCR was used to
determine whether nosRC flies contain low levels of
mRNA. Using primers that flank intron 1, a low level of
NOS mRNA was detected in extracts prepared from whole flies. Since the major site of transcription in adult females is the ovary, it is assumed that most of this mRNA is derived from the rudimentary nosRC ovaries. Two major cDNA
species were detected by ethidium bromide staining of the
PCR product following electrophoresis. To further
characterize these cDNAs, the PCR products were subcloned
and six individual clones were sequenced. The six clones
appear to represent mRNAs generated by processing from
cryptic splice sites; the open reading frame is preserved in
three different clones, and one of these encodes a Nos
derivative that is four amino acids larger than wild type. This cDNA clone plausibly
represents the mRNA species that gives rise to the nosRC-encoded
protein (Verrotti, 2000).
In an attempt to detect protein encoded by nosRC directly,
transgenic flies were prepared bearing a nosRC-myc gene that encodes an epitope-tagged protein that is otherwise identical
to the nos+-myc gene described above. Using Western blots, a
very low level of nearly full-length protein was detectable in
nosRC-myc ovaries from five different transgenic lines. Comparison with dilutions of extracts prepared from nos+-myc
transgenic ovaries suggests that the level of protein
encoded by nosRC is in the order of 1%-2% of wild type. By
crossing the nos+-myc and nosRC-myc transgenes into cup-/+
backgrounds and comparing the level of Nos protein in ovarian
extracts, it was found that reducing the level of Cup does not
significantly affect the level of protein encoded by either
transgene. Thus, low levels of Cup do not appear to
suppress the nosRC ovarian phenotype by stabilizing Nos. It is
concluded that, in the presence of reduced levels of Cup, 1%-2%
of the wild-type level of Nos is sufficient to promote normal
maintenance of the germline stem cells and differentiation of
the cysts (Verrotti, 2000).
It is concluded that reducing the level of functional Cup suppresses the oogenesis defects in hemizygous nosRC ovaries. This finding
suggests that Cup acts to inhibit the residual protein encoded
by nosRC and prevent it from acting on potential regulatory
targets. The identities of such targets are not currently known.
In addition, the sequence of Cup sheds no light on its function.
A search of the current database reveals no significant
homologies to proteins of known function, neither does it
possess recognizable motifs using programs such as Prosite,
although Keyes (1997) has suggested that Cup
may be a microtubule-associated protein. A fragment of human
sequence bears high homology to a part of the fly protein, and
thus it seems likely that one or more of the Cup functions are
evolutionarily conserved. Further analysis of the significance
of the Nos-Cup interaction awaits definition of the biochemical
activities of Cup and the identification of Nos-regulated genes
in the ovary (Verrotti, 2000).
While physiological levels of Cup are capable of inhibiting
the low level of Nos activity in hemizygous nosRC flies, it is
not understood what role the Cup-Nos interaction
plays in the ovaries of wild-type flies. Over-expression of Nos
is deleterious in many different tissues -- the embryo, the eye
imaginal disc and the male germline -- suggesting that Nos is a potent
regulator of gene expression. Consistent with this idea, it has been found
that extremely low levels of Nos, in the order of a few percent
of the wild-type amount, suffice for biological activity in the
ovary. Thus, it seems possible that the interaction with Cup
helps restrict Nos activity, which otherwise might interfere
with normal ovarian development. Alternatively, Nos and Cup
may act together to govern some aspect of germ cell
development. Cup appears to play a role in early germline
development, since cup and ovarian tumor interact genetically in
the ovary, leading to over-proliferation of germline cells (Keyes, 1997; Verrotti, 2000).
The Drosophila Nanos protein is a localized repressor of Hunchback mRNA translation in the early embryo, and is required for the establishment of the anterior-posterior body axis. 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. Full-length and C-terminal Nanos proteins bind to RNA in vitro with high affinity, but with little sequence specificity. Mutations affecting the Hunchback mRNA target sites for Nanos-dependent translational repression are found to disrupt translational repression in vivo, but have little effect on Nanos RNA binding in vitro. Thus, the Nanos zinc domain does not specifically recognize target Hunchback RNA sequences, but might interact with RNA in the context of a larger ribonucleoprotein complex (Curtis, 1997).
Translational regulation of Hunchback mRNA is essential for posterior patterning of the Drosophila embryo. This regulation is
mediated by sequences in the 3'-untranslated region of HB mRNA (the Nanos response elements or NREs), as well as two trans-acting
factors -- Nanos and Pumilio. Pum
binds to a pair of 32-nucleotide sequences (named Nanos
response elements -- NREs) in the 3'-UTR of
maternal HB mRNA in order to repress HB
translation in the posterior of the embryo. This translational repression is essential
for normal abdominal segmentation. The RNA-binding domain of Pum is structurally similar to
that of another translational regulator, FBF (fem-3
mRNA-binding factor) found in C. elegans (Zhang, 1997).
The minimal RNA-binding domain of each protein consists of eight
imperfect repeats plus flanking residues. These structural similarities
define a conserved 'Puf' motif (Pum and FBF)
that is found in proteins from diverse organisms from yeast to humans. However, the RNA partner of no
other Puf domain protein has been identified, nor is it clear whether other
Puf proteins regulate translation or some other aspect of RNA metabolism.
Thus, Pumilio recognizes the NREs via a conserved binding motif. The mechanism of Nanos action has
not been clear. In this report protein-protein and protein-RNA interaction assays in yeast and in vitro were used to show that Nanos forms a
ternary complex with the RNA-binding domain of Pumilio and the NRE. Mutant forms of the NRE, Nos, and Pum that do not regulate HB mRNA normally in
embryos do not assemble normally into a ternary complex. In particular, recruitment of Nos is dependent on bases in the center of the NRE, on the carboxy-terminal
Cys/His domain of Nos, and on residues in the eighth repeat of the Pum RNA-binding domain. These residues differ in a closely related human protein that also binds
to the NRE but cannot recruit Drosophila Nos. Taken together, these findings suggest models for how Nos and Pum collaboratively target HB mRNA. More
generally, they suggest that Pum-like proteins from other species may also act by recruiting cofactors to regulate translation (Sonoda, 1999).
In one model, Nos simultaneously makes specific contacts with Pum and nucleotides 17-20 of the NRE. On their own, neither the Nos-Pum nor the Nos-NRE
contacts are strong enough to recruit Nos to hb mRNA (at least in the presence of competitor proteins and RNAs), because binary complexes with Nos are not
detectable. In another model, unbound Pum cannot interact with Nos, but binding to the NRE induces a conformational change in Pum, which subsequently recruits
Nos via protein-protein contacts. In this model, nucleotides 17-20 of the NRE interact with Pum to induce the conformational change without affecting its affinity for
the RNA, and nonspecific interactions between Nos and other portions of the RNA help stabilize the complex. Either model is consistent with the nonspecific
RNA-binding activity reported for the carboxy-terminal portion of Nos in vitro and the RNA-Nos cross-link found in this study.
Further structural and biochemical experiments will be required to distinguish between these (or alternative) models (Sonoda, 1999 and references therein).
The mechanism by which the ternary complex blocks translation is not yet clear. mRNAs subject to Nos- and Pum-dependent repression
are deadenylated in vivo. In addition, Nos and Pum have been shown to regulate internal ribosome
entry site (IRES)-dependent translation in imaginal disc cells, suggesting that their regulatory target lies downstream of cap recognition and scanning. It is assumed that some surface of the ternary complex, formed jointly by Nos and Pum, targets a component of the polyadenylation or translation
machinery. This surface appears to be altered in the Pum680 mutant protein, which binds the NRE normally but is defective in regulating HB translation in the embryo. The Pum680 mutant recruits Nos into a ternary complex normally and thus apparently is defective in a subsequent step of the
repression reaction. The RNA-binding domain of Pum therefore appears to have at least three different functions in regulating HB: recognizing the NRE,
recruiting Nos, and acting as a corepressor (with Nos) to block translation (Sonoda, 1999 and references therein).
In the experiments reported in this study, focus was placed on discrete regions of both Nos (the carboxy-terminal 97 amino acids) and Pum (the minimal RNA-binding domain),
which play an essential role in formation of the ternary complex. However, other regions of Nos are known to be required for its function in repressing translation in the embryo. In addition, residues elsewhere in Pum play an unknown role in augmenting the intrinsic translational repression activity of the
RNA-binding domain. Thus, the ternary complex formed by the 157-kD, full-length Pum protein may be stabilized by auxillary
protein-protein or protein-RNA interactions in addition to those that mediate recruitment of the carboxy-terminal domain of Nos by the RNA-binding
(or Puf domain) of Pum. The results suggest that Puf domain proteins generally may act by recruiting cofactors to specific RNA binding sites. Cofactor specificity may be
mediated, at least in part, by the eighth repeat of the Puf domain. Although Puf domain proteins have been described in organisms from yeast to humans, for
only one protein other than Drosophila Pum, C. elegans FBF, is the relevant RNA regulatory target known. FBF regulates the sperm/oocyte switch in the
hermaphrodite germ line by governing the translation of fem-3 mRNA (Zhang, 1997). The FBF
RNA-binding domain interacts with one of the C. elegans Nos homologs (Kraemer, 1999). Further experiments will be required to determine whether
the Pum/fly Nos complex and the FBF/worm Nos complex function in a similar manner (Sonoda, 1999 and references therein).
Translation regulation plays an essential role in the differentiation and development of animal cells. One
well-studied case is the control of Hunchback mRNA during early Drosophila embryogenesis by the trans-acting factors Pumilio, Nanos, and Brain Tumor. This study reports a crystal structure of the critical region of
Pumilio, the Puf domain, that organizes a multivalent repression complex on the 3' untranslated region of Hunchback mRNA. The similarity between Pum RBD and that of another translation regulator FBF, which binds to the 3'UTR of fem-3 mRNA in C. elegans, defines a Puf (Pum and FBF) domain, which is
conserved in organisms as diverse as plants, yeast, and humans. The Puf domain is characterized by eight imperfect repeats of ~36 amino acids (Puf repeats), followed by a C-terminal extension. All eight repeats appear to be required for proper folding of the Puf domain, since limited proteolysis fails to yield stable smaller fragments. The Puf domain is thus amongst the largest sequence-specific RNA binding motifs to be discovered; the RRM, the KH domain (70 residues), and the dsRBD (65 residues), are much smaller. The PUF domain structure reveals an extended, rainbow shaped molecule, with tandem helical repeats that bear unexpected resemblance to the armadillo repeats in beta-catenin and the HEAT repeats in protein phosphatase 2A. Based on the structure and genetic experiments, putative interaction surfaces for Hunchback mRNA and the cofactors Nanos and Brain Tumor are identified. This analysis suggests that similar features in helical repeat proteins are used to bind extended peptides and RNA (Edwards, 2001).
Two lines of evidence from mutagenesis studies support the idea that the Pum concave surface binds RNA. (1) A gene encoding the 322 residue minimal Pum RBD was randomly mutagenized and variants were isolated that
bind normally to the wild-type NRE in yeast. Collectively, these variants bear substitutions at 61 residues, 55 of which map to the structure; the remaining six are in the putative 9th repeat, not in this crystal structure. Of these, only 3 (presumably silent) substitutions fall on the solvent exposed concave
surface, with the remaining 52 lying elsewhere. The relative paucity of substitutions within the inner surface is consistent with this being the area that contacts the RNA. (2) Based on the structure, single substitutions were introduced in solvent-exposed residues along the inner surface in five of the eight Puf domains and RNA binding activity was tested in yeast. Each of these mutants is inactive. Thus, the concentration of positive charge and the distribution of both silent and inactivating substitutions together suggest that the RNA interacts with the inner concave surface (Edwards, 2001).
It is proposed that HB mRNA binds to this inner surface in an extended single-stranded conformation. Algorithms that predict RNA structure suggest the NRE does not adopt a stable secondary or tertiary structure. The minimal NRE for high affinity Pum binding consists of nucleotides 3-27, which bracket specific contacts with nucleotides 9, 11-13, and 21-24. The length of this minimal NRE, in an extended single-stranded conformation (112 Å), agrees approximately with the contour length (90 Å) of the concave surface of the Puf domain. It is noteworthy that beta-catenin also has the highest concentration of positive charge within its concave surface (or groove), which is the proposed binding site for segments of cadherins, APC, and members of the LEF-1/TCF family of transcription factors. A recent crystal structure of a beta-catenin/TCF complex shows the TCF segment tethered along the positively charged groove. In the case of karyopherin-alpha, the concave surface is the binding site for the NLS peptide. Taken together, the binding of ligands to concave surfaces is a recurring theme in helical repeat proteins. The Pum Puf domain shows that this type of extended surface can be used to bind RNA, as well as peptides (Edwards, 2001).
Repression of HB mRNA depends not only on Pum, but also on the recruitment of Nanos and Brat to form a quaternary complex. Previous work suggested that Nos is recruited via residues in Puf repeat 8. These residues map to the extra long loop between helices H1 and H2 in repeat 8, that is the main protrusion from an otherwise relatively smooth outer Pum surface. Two different insertions into this loop have no effect on Pum-RNA binding but eliminate recruitment of Nos. To further define the Nos interaction surface, the collection of Pum mutants that bind normally to RNA was tested for Nos recruitment in yeast. Of the 61 substitutions distributed throughout the domain, only two abrogate interaction with Nos. One is a substitution in the putative ninth Puf repeat that is not represented in this structure, while the other changes the solvent exposed phenylalanine on the H1/H2 loop to a serine (F1367S). Thus, the Pum surface that interacts with Nos appears to be limited to a small region that includes the eighth repeat and the C-terminal tail. If this tail indeed does fold into a ninth Puf repeat, then the Pum-Nos interface would span a length of ~15-20 Å on the outer convex surface. It is tempting to think that the C-terminal tail may only fold when Pum binds to the RNA, thereby explaining why Nos is only recruited to the Pum/NRE binary complex and not to Pum alone. The insertions into the long flexible loop in repeat 8 may modify its conformation such that F1367 is no longer exposed for interaction with Nos. The proposed Phe-Nos interaction is reminiscent of the way in which a solvent exposed phenylalanine on the receptor CD4 interacts with the HIV gp120 glycoprotein (Edwards, 2001).
Development of the Drosophila abdomen requires repression of maternal Hunchback (HB) mRNA translation in the posterior of the
embryo. This regulation involves at least four components: nanos response elements within the HB 3' untranslated region and the activities
of Pumilio (Pum), Nanos (Nos), and Brain tumor. To study this regulation, an RNA injection assay was developed that faithfully recapitulates the regulation of the endogenous HB message. Previous studies have suggested that Nos and Pum can regulate translation by directing poly(A) removal. RNAs that lack a poly(A) tail and cannot be polyadenylated and RNAs that contain translational activating sequences in place of the poly(A) tail are still repressed in the posterior. These data demonstrate that the poly(A) tail is not required for regulation and suggest that Nos and Pum can regulate HB translation by two mechanisms: removal of the poly(A) tail and a poly(A)-independent pathway that directly affects translation (Chagnovich, 2001).
To determine whether injected RNAs require the same factors for regulation as the endogenous maternal HB transcript, the effect of nos, pum, and NRE mutations on the translation of the HFH mRNAs was studied.
The test mRNA (HFH) contains the maternal HB 5' UTR, the F-Luc coding region, and the maternal HB 3' UTR.
HFH transcripts injected into nos or pum mutant embryos are translated equally well in the anterior versus the posterior. This finding is consistent with the requirement for Nos and Pum in regulating HB translation in vivo. Further, HFH transcripts containing mutant NREs in which the six guanosines have been changed to uracil (GU) show no significant difference in translation between the anterior and posterior of the embryo, regardless of the genetic background. This NRE mutation disrupts Pum binding to the NRE in vitro and eliminates translational repression of maternal HB mRNA in vivo. Thus factors that are required for the regulation of the endogenous maternal HB transcript also are required for the regulation of the injected transcripts (Chagnovich, 2001).
HFH reporter mRNAs were synthesized that contained the Drosophila histone H1 3' terminal stem loop (HSL) in place of the poly(A) tail (HFH HSL).
To determine whether the NRE represses translation of HSL-containing mRNAs by directing removal of the HSL, just as it directs removal of the poly(A) tail, radiolabeled, m7GpppG-capped HB 3' UTRs containing either a poly(A) tail or the HSL were injected into WT embryos. Consistent with previous findings, RNAs containing a poly(A) tail maintain the poly(A) tail in the anterior, but the poly(A) tail is rapidly removed in the posterior. However, when RNAs containing the HSL are injected, the HSL is maintained in the anterior and posterior of the embryo, demonstrating that the NRE complex is not directing removal of HSL. Next tested was whether the differences in translation of the HSL-containing reporters were caused by differential stability of the mRNAs. In these experiments, the levels of HSL or poly(A)+ reporter RNAs compared with a poly(A)+ control RNA containing a mutant NRE were examined. The relative levels of the reporter mRNAs are not significantly altered in the anterior versus the posterior for either the polyadenylated or the HSL-containing mRNAs. Together these data demonstrate that translational regulation of HB does not require removal of the poly(A) tail. It is concluded that NRE-directed repression can be independent of the poly(A) tail (Chagnovich, 2001).
Bicoid is a key determinant of anterior Drosophila development. The prototypical Puf protein Pumilio temporally regulates bicoid (bcd) mRNA translation via evolutionarily conserved Nanos response elements (NRE) in its 3'UTR. Disruption of Pumilio-bcd mRNA interaction by either Pumilio or bcd NRE mutations causes delayed bcd mRNA deadenylation and stabilization, resulting in protracted Bicoid protein expression during
embryogenesis. Phenotypically, embryos from transgenic mothers that harbor bcd NRE mutations exhibit dominant anterior patterning defects and similar head defects have been discovered in embryos from pum minus mothers. Hence, Pumilio is required for normal anterior development. Since bcd mRNA resides outside the posterior gradient of Nanos, the canonical partner of Pumilio, the data suggest that Pumilio can recruit different partners to specifically regulate distinct mRNAs (Gamberi, 2002).
To identify sequences regulating bcd mRNA expression, focus was placed on the perfect bipartite NRE sequence GUUGU-N5-AUUGUA (A box-N5-B >box) in the 3'UTR of bcd, starting 50 nucleotides downstream of the bcd translational stop codon. This bcd motif was noticed previously, but its role in normal development was unclear because it resides outside the Nanos embryonic domain. The hb 3'UTR contains two NRE motifs, while bcd has one NRE and an additional B box at position +79 (termed 1 1/2 NREs). By aligning the bcd and hb 3'UTRs from all available species, it was found that the bcd motifs are closer to the second hb NRE. Moreover, the 1 1/2 NREs was absolutely conserved in the bcd 3'UTR of eight fly species that diverged more than 60 million years ago, underscoring functional constraint. Thus, the role the NREs play in bcd expression and their contribution to normal embryonic development were analyzed (Gamberi, 2002).
Pumilio temporally regulates bcd mRNA expression: its mutation causes delayed deadenylation and stabilization of the bcd message, resulting in protracted Bicoid protein expression. Disruption of this molecular control perturbs normal Drosophila head development (Gamberi, 2002).
While affected Bicoid targets are presently only speculative, it was fortuitously noticed that late zygotic hb expression (hbzyg) was increased in Northern blots of bcd NRE mutants, hinting at one potential affected molecule. This is consistent with the pum- data. A second target candidate arises from the defective mouth hook (mh) base present in both bcd NRE mutant transgenics and pum- embryos. This alteration, which is suggestive of a maxillary segment defect, similarly occurs when orthodenticle (otd) is expressed ectopically. Interestingly, Bicoid activates otd transcription and resulting Orthodenticle has the same DNA-binding specificity as Bicoid. Hence, the prolonged Bicoid expression in mutant bcd NRE transgenics and pum- embryos may interfere with normal head development through a complex pattern of interactions (Gamberi, 2002).
Pum was originally characterized as a posterior group gene: Pumilio and Nanos cooperate to repress maternal hb (hbmat) in the posterior of the embryo, allowing abdominal patterning. However, ubiquitous expression of Pumilio in excess of hb implies it could possess additional function(s) elsewhere. This study demonstrates that Pumilio also participates in Drosophila anterior embryonic patterning. pum embryos exhibit head defects. The Pumilio anterior function is mediated via bcd post-transcriptional expression, since similar anterior abnormalities occur when Pumilio's presumptive bcd mRNA-binding site is mutated (Gamberi, 2002).
It was asked if bcd NRE regulation required the Pumilio canonical partner Nanos. When bcd mRNA is injected posteriorly or Nanos is expressed anteriorly by genetic means, Pumilio and Nanos can affect bcd expression because all factors co-exist. In each case, large Nanos amounts are present and head morphogenesis is inhibited (Gamberi, 2002).
A major Nanos role in normal head formation seems unusual because Nanos and bcd mRNA reside at opposite ends of the embryo. Surprisingly Nanos does influence bcd expression and subsequent anterior development to some degree. This suggests undetectable Nanos amounts may regulate bcd mRNA in the anterior. Analogously, a contribution of low Nanos levels in oogenesis has been reported (Gamberi, 2002).
bcd mRNA might encounter low Nanos levels via the NRE-dependent back-up mechanism postulated to repress it when it escapes localization, diffuses posteriorly and intercepts the Nanos gradient. Alternatively, sufficient Nanos moieties might diffuse anteriorly, analogous to when enough Bicoid molecules exist in the posterior of the embryo to elicit hairy stripe 7 expression or to cooperate with Caudal in knirps activation. In a different scenario, nos mRNA translational repression throughout the embryo may be leaky, yielding low basal Nanos levels everywhere, including the anterior. How a Pumilio-bcd mRNA complex can recruit enough Nanos for action and whether this involves additional (anterior?) factors to modulate Nanos activity are questions for future studies (Gamberi, 2002).
nos- severe head involution defects occur at a significantly lower frequency than in pum- cuticles (4% versus 81%; null versus presumptive null), raising the intriguing possibility that an additional partner(s) for Pumilio exists at the anterior that affects bcd NRE function independently of Nanos. Consistently, the sequence between the A and B boxes of the bcd and hb NREs diverges at two of the four nucleotide positions known for hb recruitment of Nanos. While Pumilio and Nanos are usually thought of as functioning in concert, they have only partially overlapping roles in the Drosophila germline and may function independently in oogenesis. The alternate Pumilio partner for bcd might be an anterior Nanos paralog (although only one nos gene was found in the fly genome) or a distinct moiety. Interestingly, S. cerevisiae has five Puf proteins involved in mRNA metabolism but no Nanos homologs, suggesting some Puf proteins can function with novel partners (Gamberi, 2002).
Molecular data indicate the bcd NREs act temporally, repressing translation in a deadenylation dependent way. Mutating either Pumilio or the bcd NREs results in protracted Bicoid expression. Presently, it cannot be distinguish if the bcd NREs primarily constitute a translational control element with mRNA deadenylation and instability accompanying specific repression or a regulated instability element whose downstream effects are seen at the protein level. Interestingly, in addition to detecting specific Pumilio-dependent bcd NRE regulation, a second effect of pum- mutation was noticed: stabilization of maternal mRNAs devoid of NREs. While it is unclear whether this effect is direct, it may reflect a novel Pumilio function in general NRE-independent mRNA turnover (Gamberi, 2002).
Complementary phenotypic analyses of bcd NRE mutant transgenes has revealed that prolonged Bicoid expression interferes with maxillary segment determination, which may affect head involution by altering the intersegmental contacts required for appropriate head morphogenetic movements. Incomplete overlap between the highly penetrant mouth hook defect and the partially penetrant head involution defect might reflect the complexity of fly head development, which is subjected to redundancy and fail-safe mechanisms (Gamberi, 2002).
The conservation of the bcd and hb NREs, their Pumilio association, and their ability to direct translational regulation imply functional similarity between these elements. However, the hb regulatory system operates on a uniformly distributed mRNA to repress its expression in the embryonic posterior where Nanos is most concentrated. By contrast, bcd mRNA is spatially restricted to the anterior via localization, which conceivably impacts NRE action and predicts underlying functional differences between bcd and hb NREs (Gamberi, 2002).
The novel Pumilio role in anterior development documented here raises the exciting possibility that the prototypical Puf protein Pumilio operates more generally than previously thought, regulating multiple physiological pathways in different Drosophila embryonic locales. Furthermore, since Pumilio is also expressed in the adult fly and pum- flies exhibit additional uncharacterized phenotypes, Pumilio may function in mRNA metabolism throughout the life of the fly (Gamberi, 2002).
To date, NREs have been identified in three mRNA species: hb, bcd and cyclin B. For each, NRE organization differs: hb and bcd contain two and 1 1/2 copies, respectively, of the basic (A box-N5-B box) NRE motif, while cyclin B contains one NRE motif with a larger spacer. Furthermore, hbmat and hbzyg mRNA have identical NREs, but hbzyg mRNA seems relatively insensitive to regulation by Pumilio/Nanos. Differences among NREs combined with distinct distributions of NRE-containing mRNAs and their known effectors underlie a potential combinatorial model of NRE recognition in which a common factor (Pumilio) associates with the mRNA target sequence and subsequently recruits different (sets of) factors (e.g. Nanos, Brat for hbmat mRNA) to regulate ultimately and specifically unique target expression. How Pumilio functions on different NRE-containing mRNAs, what factor combinations are employed in distinct situations and whether Nanos homologs are involved in every case are experimental questions begging to be answered (Gamberi, 2002).
Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
nanos:
Biological Overview
| Evolutionary Homologs
| mRNA localization and post-transcriptional regulation
| Developmental Biology
| Effects of Mutation
| References
Society for Developmental Biology's Web server.