Gene name - Stellate
Cytological map position - 12E1--2
Function - regulatory subunit of CK II
Symbol - Ste
FlyBase ID: FBgn0010086
Genetic map position - 1-45.7
Classification - ß subunit of CK2
Cellular location - cytoplasmic
The Y chromosome is known to be essential for male fertility in Drosophila melanogaster. Many aspects of the phenotype of flies lacking a Y chromosome (X0) reflect an unusual negative regulatory interaction that normally occurs between the X chromosome-linked Stellate (Ste) locus and the Y chromosome-linked Suppressor of Stellate [Su(Ste)] locus. That is, the Ste and Su(Ste) are normally silent. Deficiencies of Su(Ste) led to the derepression of the Ste elements in the male germ line and led to the mutant phenotype. Males lacking the Y linked Su(Ste) locus exhibit needle or star-shaped crystalline aggregates in the nuclei and the cytoplasm of primary spermatocytes and several meiotic defects, such as an undercondensation of meiotic chromosomes and an altered distribution of mitochrondria, leading, in many cases, to complete sterility. Both the formation and shape of crystals and the strength of the other meiotic abnormalities depend on the allelic state of the X-linked Ste locus (Bozzetti, 1995 and references therein).
How do the Ste loci maintain their coding capacity despite the fact that they are normally inactive and thus probably dispensable? The adventitious nature of these sequences is indicated by the finding that, with the exception of a single copy located in the Y chromosome of Drosophila simulans, such sequences are not present in any other member of the melanogaster group of Drosophila (Livak, 1984). Data showing the absence of Ste protein in normal males, further support the suggestion that the Ste-Su(Ste) system is dispensable and, under this hypothesis, identify an evolutionary strategy evolved by a parasitic genetic system to actively maintain itself. It consists in being nested into a meiotic physiological process so intimately that it mimics essential functions. Thus, any perturbation of the Ste-Su(Ste) system results in a perturbation of the entire meiotic process with a negative impact on fitness (Bozzetti, 1995).
The multiple testis-expressed Ste and Su(Ste) tandem repeats are localized in the D. melanogaster genome on the X and Y chromosomes, respectively (Livak, 1984; Balakireva, 1992; Tulin, 1997). ORF of the Stellate repeats are maintained by selective pressure (Tulin, 1997) and encodes protein with a striking homology to the ß subunit of Casein kinase II (CKII) (Livak, 1990; Bozzetti, 1995). Moreover, in vitro experiments have shown that Stellate-encoded protein can interact with the catalytic alpha subunit of the CKII enzyme, altering its activity (Bozzetti, 1995). The hyperexpression of Stellate genes is thought to be suppressed by the homologous Suppressor of Stellate [Su(Ste)] tandem repeats (Balakireva, 1992; Tulin, 1997: Palumbo, 1994). Deletion of the bulk of Su(Ste) repeats localized in the crystal locus of the Y chromosome (cry1Y chromosome) results in hyperexpression of Stellate in testes and causes meiotic abnormalities and accumulation of crystalline aggregates containing Stellate-encoded protein in primary spermatocytes (Bozzetti, 1995; Palumbo, 1994). In fly strains containing a high copy number of Ste repeats, this hyperexpression, due to the deletion of Su(Ste) repeats, caused male sterility (Livak, 1994). Su(Ste) repeats have an Ste-like region with about 90% nucleotide identity to the Stellate genes in a promoter and coding region with randomly positioned nucleotide substitutions. Each Su(Ste) repeat unit also contains a Y-specific region with no sequence similarity to Stellate genes and a 1360 (hoppel) transposon insertion (Balakireva, 1992; Kogan, 2000). In contrast to Stellate genes, all sequenced Su(Ste) repeats have damaged open reading frames and are considerably more diverged, suggesting the absence of selective pressure to sustain coding capacity (Balakireva, 1992; Kogan, 2000). Sense Su(Ste) transcripts with the site of polyadenylation located in a Y-specific region upstream of hoppel transposon insertion have been revealed (Kalmykova, 1998) as a result of testes cDNA library screening (Aravin, 2001 and references therein).
The injection of double-stranded RNA (dsRNA) has been shown to induce a potent sequence-specific inhibition of gene function in diverse invertebrate and vertebrate species. The homology-dependent posttranscriptional gene silencing (PTGS) caused by the introduction of transgenes in plants may be mediated by dsRNA. The analysis of Caenorhabditis elegans mutants impaired with dsRNA-mediated silencing and studies in plants implicate a biological role of dsRNA-mediated silencing as a transposon-repression and antiviral mechanism. The silencing of testis-expressed Stellate genes by paralogous Su(Ste) tandem repeats, which are involved in the maintenance of male fertility in Drosophila melanogaster, has been examined. Both strands of repressor Su(Ste) repeats are transcribed, producing sense and antisense RNA. The Stellate silencing is associated with the presence of short Su(Ste) RNAs. Cotransfection experiments revealed that Su(Ste) dsRNA can target and eliminate Stellate transcripts in Drosophila cell culture. The short fragment of Stellate gene that is homologous to Su(Ste) was shown to be sufficient to confer Su(Ste)-dependent silencing of a reporter construct in testes. Su(Ste) dsRNA-mediated silencing affects not only Stellate expression but also the level of sense Su(Ste) RNA providing a negative autogenous regulation of Su(Ste) expression. Mutation in the spindle-E gene relieving Stellate silencing also leads to a derepression of the other genomic tandem repeats and retrotransposons in the germline. It is concluded that homology-dependent gene silencing inhibits Stellate gene expression in the D. melanogaster germline, ensuring male fertility. dsRNA-mediated silencing may provide a basis for negative autogenous control of gene expression. The related surveillance system is implicated to control expression of retrotransposons in the germline (Aravin, 2001).
Strand-specific RT-PCR and Northern analysis show convincingly that both strands of Su(Ste) repeats are transcribed. Specific primers were designed to provide strand-specific reverse transcription of the Su(Ste) RNA, followed by PCR using primer pairs flanking intron 2 to distinguish the spliced and nonprocessed transcripts. Both unprocessed antisense Su(Ste) transcripts as well as spliced, and vestiges of nonspliced, sense Su(Ste) transcripts were detected in the testes of males carrying a normal Y chromosome. Two types of genomic Su(Ste) repeats differing by a 23 bp deletion (Balakireva, 1992) in an amplified region produce antisense as well as sense transcripts. A reverse transcription reaction using oligo-dT primer resulted in the same PCR products, with a great predominance of the spliced form, as was revealed with the sense-specific primer. This result suggests that antisense Su(Ste) transcripts failed to enter into an RT reaction with the oligo-dT primer and belong to a nonpolyadenylated RNA fraction (Aravin, 2001).
Northern blot analysis with the Su(Ste) strand-specific probe revealed antisense transcripts with an average size of 3 kb that correspond to about one Su(Ste)-repeating unit. The observed size heterogeneity of antisense Su(Ste) transcripts may be due to imprecise starts and/or terminations of transcription. Using a primer corresponding to the hoppel transposon sequence, it has been revealed that, in contrast to sense transcripts, antisense Su(Ste) transcripts include hoppel transposon sequences. The presence of hoppel sequences in antisense Su(Ste) transcripts indicates that the hoppel transposon might be responsible for the initiation of antisense Su(Ste) transcription. Mapping of the 5' end of the antisense Su(Ste) transcripts using primer extension and 5'-RACE experiments indicates that antisense transcription starts in different sites of the hoppel transposon body. The RT-PCR, Northern blot analysis, primer extension, and 5'-RACE experiments provide compelling evidence for the existence of antisense Su(Ste) transcripts starting in the hoppel transposon and extending through the Y-specific sequence into the region with a high sequence identity (90%) with the Ste genes (Aravin, 2001).
As in the cases of artificial RNAi, Stellate silencing is associated with the presence of small homologous RNAs species, presumably produced by Su(Ste) dsRNA cleavage. The presence of Su(Ste) dsRNA in a total testes RNA preparation was tested using treatment by RNase One, which degrades single-stranded RNA, followed by denaturation, reverse transcription using random primers, and PCR amplification. A denaturation-dependent amplification signal in PCR with Su(Ste)-specific primers was obtained using RNA isolated from normal males, but it was barely detectable in cry1Y males with a deletion of Su(Ste) repeats. This result indicates the presence of Su(Ste) dsRNA in the sample. However, dsRNA might be formed as a result of the annealing of sense and antisense strands during the isolation procedure. The presence of endogenous dsRNA in testes suggests that small RNAs that are derived from the dsRNA by endonucleolytic cleavage are being produced. The presence of small RNA species homologous to Stellate and Su(Ste) sequences was tested in the total testes RNA. Northern hybridization with sense or antisense Su(Ste) RNA probes revealed the presence of heterogeneous 25-27 nt RNA species in the total RNA isolated from normal males. No small RNAs were detected in cry1Y males, suggesting that these RNAs are produced from the Su(Ste) locus (Aravin, 2001).
The sequence similarity between Stellate and Su(Ste) transcripts is sufficient to RNA interference in cell culture. To test the ability of Su(Ste) dsRNA to suppress Stellate expression, RNAi experiments were performed in Drosophila cell culture. Schneider 2 cells were cotransfected with the hsp70-Ste-lacZ reporter plasmid and the dsRNA corresponding to Stellate or Su(Ste) repeats. The reporter construct contains the bulk of Ste ORFs fused to the lacZ gene. The expression of the fused Ste-lacZ construct was measured by the evaluation of ß-galactosidase activity after a heat shock. While Ste dsRNA completely abolishes ß-galactosidase expression, cotransfection with Su(Ste) dsRNA leads to about an 8-fold decrease of expression. Thus, the extent of sequence identity between Ste and Su(Ste) repeats is sufficient for strong dsRNA-mediated silencing, although the strength of silencing by Su(Ste) dsRNA is significantly lower compared to Ste dsRNA (Aravin, 2001).
The short fragment of Stellate gene confers Su(Ste)-dependent silencing of the Ste-lacZ reporter in testes. To reveal the size of a Stellate gene fragment that is sufficient to induce Su(Ste)-dependent silencing, a set of transgene lacZ reporters was used, driven by 5' Ste fragments sharing homology to the Su(Ste) repeats. Transgenic flies carrying the Ste225-lacZ and Ste134-lacZ constructs demonstrate a drastic increase of ß-galactosidase expression in the testes of cry1Y males as compared to XY males, thus demonstrating a strong response of reporters to the elimination of Su(Ste) repeats. The Su(Ste)-dependent silencing was observed in all tested transgenic stocks (six and four stocks carrying the Ste225-lacZ and Ste134-lacZ constructs, respectively), independent of chromosomal localization of insertion. The 134 bp fragment of the Stellate gene in the Ste134-lacZ construct that is sufficient for establishing the repressed state contains only a 102 bp sequence with 89%-94% nucleotide identity to Su(Ste) repeats including 33 bp of 5'-transcribed sequence with Stellate ATG start codon fused to the lacZ ORF. Thus, a short region of homology to Su(Ste) repeats is shown to be sufficient to confer the Su(Ste)-dependent silencing. The Su(Ste)-dependent silencing of an intron-less construct suggested that the Ste/Su(Ste) interaction does not occur on the level of Ste transcript splicing (Aravin, 2001).
The RT-PCR analysis has revealed the presence of Su(Ste) antisense transcripts in cry1Y males, encoded by the remnant Su(Ste) repeats untouched by the cry1Y deletion. Both Northern blot and RT-PCR analysis show the drastic reduction of antisense Su(Ste) RNA in cry1Y testes as compared to the wild-type ones. However, in contrast to antisense transcripts, a steady-state level of sense Su(Ste) transcripts is significantly increased in cry1Y males, despite the deletion of the bulk of Su(Ste) repeats. This observation may be explained by the suggestion that the Ste/Su(Ste) interaction is mediated by the Su(Ste) dsRNA that targets both Stellate and sense Su(Ste) transcripts. Therefore, the cry1Y deletion, causing a drastic decrease of the Su(Ste) dsRNA level, results in an increase of sense Su(Ste) expression. Thus, dsRNA may be considered as a negative autogenous regulator of sense Su(Ste) expression. It should be noted that a decrease in the dsRNA level failed to cause an increase of antisense Su(Ste) transcripts (Aravin, 2001).
A weak expression of Su(Ste) repeats is apparent in heads of adult flies. Both sense and antisense Su(Ste) RNA were detected in heads. As in testes, the level of antisense transcripts in heads is drastically decreased in cry1Y males, while the level of sense transcripts is increased. This observation suggests that the Ste/Su(Ste)-silencing mechanism also operates in somatic tissues (Aravin, 2001).
aubergine (aub) and spindle-E mutations cause a relief of Stellate and sense Su(Ste) silencing. Stellate derepression in the presence of the intact Su(Ste) locus has been observed as a result of aubergine and spindle-E (spn-E) mutations, also known as sting and homeless, respectively. The Aubergine protein has homologs involved in PTGS and RNAi in plants, fungi, and animals. The spn-E gene encodes a putative RNA helicase that is also proposed as a participant in dsRNA-mediated silencing (Aravin, 2001).
A relief of Stellate silencing occurs as a result of the spn-E1 mutation: this was confirmed by studying the expression of the Ste-lacZ reporter construct in the spn-E1/+ and spn-E1/spn-E1 males. The expression of ß-galactosidase in testes is greatly enhanced in spn-E1/spn-E1 males as compared to the heterozygous ones. The effects of the aubsting-1 and spn-E1 mutations on the level of sense and antisense Su(Ste) transcripts were assessed. Both mutations, when homozygous, have no effect on the level of antisense Su(Ste) transcripts, but they do increase the level of sense Su(Ste) RNA. Thus, a common mechanism, assisted by the Aubergine and Spindle-E proteins is operated in Su(Ste) dsRNA-mediated silencing of Stellate and sense Su(Ste) expression. The effect of the spn-E1 mutation is restricted to the germline, since no increase in the level of sense Su(Ste) transcripts in the heads of homozygous flies was observed (Aravin, 2001).
Several, but not all, C. elegans mutants that are resistant to dsRNA injection exhibit the mobilization of endogenous DNA transposons and derepression of repetitive transgenes in the germline. spn-E mutation leads to derepression of retrotransposons and genomic tandem repeats in the germline. The steady-state level was examined of polyadenylated transcripts of several transposable elements and nonmobile genomic tandem repeats in hetero- and homozygous flies carrying the spn-E and aub mutations. The LTR retrotransposons mdg1 and 1731 as well as the non-LTR F-element were chosen for analysis. The expression of genomic germline expressed mst40 tandem repeats and histone H3 transcripts, both encoded by repeating, was also examined (Aravin, 2001).
No significant differences were detected in the level of histone H3 expression between the hetero- and homozygous aubsting-1 flies in testes or ovaries. The expression of 1731, mdg1, and F-element retrotransposons as well as mst40 tandem repeats was not affected by aubsting-1 mutations in testes, but a moderate decrease of mdg1 and mst40 expression was detected in ovaries of aubsting-1/aubsting-1 flies, while the level of 1731 retrotransposon RNA was slightly increased. On the whole, these results suggest that the aubergine mutation fails to cause systemic activation of retrotransposons and genomic tandem repeats in the germline. In contrast to aubsting-1, the levels of transcripts of all tested retrotransposons and mst40 tandem repeats were significantly increased in both the male and female germline of the spn-E1 homozygous flies. The similar effect of retrotransposons and mst40 activation was observed in the male and female germline. The most pronounced effect was detected for the 1731 and mdg1 expression. Curiously, in ovaries of spn-E1 homozygous females, the level of mst40 transcripts was shown to be significantly higher than in testes, although mst40 transcripts have been originally described as male specific. In contrast to mst40 and Ste/Su(Ste) tandem repeats, the level of histone H3 transcripts was unaffected in both the male and female germline. Thus, Spindle-E protein participates in the Ste/Su(Ste) interaction as well as in the silencing of different retrotransposons and genomic tandem repeats in the germline (Aravin, 2001).
It is concluded that Stellate repeats are silenced by dsRNA generated by the transcription of Su(Ste) repeats. The diverged repetitive X-linked Stellate and Y-linked Su(Ste) gene clusters are involved in balanced interactions sustaining male fertility in D. melanogaster. Stellate hyperexpression, due to the absence of repressor Su(Ste) repeats, is known as a cause of male sterility. The presented results support the conclusion that both sense and antisense transcription of Su(Ste) repeats leads to dsRNA formation that is involved in homologous silencing of Stellate genes. These results provide the first direct demonstration that Drosophila genes might be naturally regulated by homology-dependent silencing mediated by dsRNA (Aravin, 2001).
Su(Ste) dsRNA and small 25-27 nt RNA species homologous to Stellate and Su(Ste) sequences were detected in total RNA preparation isolated from testes of normal, but not cry1Y, males carrying a deletion of the bulk of Su(Ste) repeats. Thus, the presence of these RNAs is associated with silencing. The 21-25 nt small RNAs are generally assumed to be a hallmark of dsRNA-mediated silencing. These RNAs have been proposed to guide the endonucleolytic cleavage of a target mRNA bearing a sequence complementary to that of the small RNAs in RNAi in Drosophila and C. elegans. The small RNA species are also detected in the cases of silencing caused by the introduction of artificial transgenes in plants. It is believed that small RNAs are produced from dsRNA by endonucleolytic cleavage. Recently, Dicer, a protein that cleaves the dsRNA to 21-23 nt fragments, was identified in Drosophila. The absence of small RNAs in testes of cry1Y males carrying a deletion of the bulk of Su(Ste) repeats suggests that these small RNAs are produced by the cleavage of Su(Ste) dsRNA. The size of small RNA species in the case of Stellate silencing (25-27 nt) is slightly longer then the size of 21-23 nt RNA produced in vitro by dsRNA cleavage in Drosophila embryo extracts, suggesting that small RNA-producing machinery may differ in some respect. It remains to be elucidated whether Dicer or other protein(s) perform Su(Ste) dsRNA processing in testes (Aravin, 2001).
Su(Ste) dsRNA produces silencing of the reporter hsp70-Ste-lacZ construct in Drosophila cell culture, but Su(Ste) dsRNA with about 90% nucleotide identity to Stellate sequence causes a less profound suppressor effect than the Ste dsRNA. This observation is in agreement with the previous studies that have demonstrated that the potential of dsRNA to induce silencing drops with a decrease in its sequence identity to a target, with a minimal threshold level of about 85%. Accordingly, the expression of the Stellate-related gene ßCK2tes (Kalmykova, 1997b), sharing only 72% sequence identity to Su(Ste) repeats, is unaffected (Kalmykova, 1997a) by the absence of Su(Ste) repeats in XO males (Aravin, 2001).
The observation that the silencing of Stellate genes is mediated by homologous Su(Ste) dsRNA suggests that the same mechanism might be directed to the Su(Ste) transcripts. Actually, the transcription of Su(Ste) repeats leads to a repression of their own sense expression. The level of sense Su(Ste) transcripts is increased in spite of the deletion of the bulk of Su(Ste) copies, demonstrating a negative mode of autogenous regulation of gene expression. Usually, negative autogenous regulation occurs when the protein gene product regulates transcription of a gene encoding this protein. In this case, negative autogenous regulation is operated by the production of the dsRNA that is supposed to be involved in the elimination of sense transcripts. The loss of a number of Su(Ste) copies prevents the accumulation of dsRNA and provides the basis for an increase of sense expression. Possibly, this mechanism of negative autogenous regulation occurs more widely and might operate in the regulation of a unique gene expression (Aravin, 2001).
The Ste/Su(Ste) interaction seems to be similar to the cosuppression phenomenon in plants and C. elegans in which silencing of both inserted transgenes and homologous endogeneous genes has been observed. Cosuppression of homologous transgenes was also demonstrated in Drosophila (Pal-Bhadra, 1997 and 1999). However, in all of the described cases, cosuppression has been caused by artificial gene manipulations. In contrast, the Ste/Su(Ste) interaction represents the first case of a naturally occurring cosuppression mechanism (Aravin, 2001).
Previous studies suggest that RNAi can target sense as well as antisense RNA strands for degradation. However, only sense Su(Ste) expression is upregulated in cry1Y males. At the same time, no increase of antisense Su(Ste) expression was revealed due to cry1Y deletion. Possibly, nonpolyadenylated antisense Su(Ste) transcripts are not exported from the nucleus, being less accessible for dsRNA-directed degradation machinery (Aravin, 2001).
Using transgenic flies carrying Ste-lacZ reporter constructs, it has been demonstrated that the 134 bp fragment of the Stellate gene, encompassing a 102 bp sequence with 89%-94% nucleotide identity to Su(Ste) repeats, is sufficient to confer Su(Ste)-dependent silencing. However, only 33 bp of this fragment are transcribed and will be a target for degradation if a posttranscriptional mechanism is operated. The recent observation that a dsRNA as short as 26 bp is still capable of inducing RNA interference in C. elegans suggests that it is not impossible. However, the presented evidence of the involvement of Su(Ste) dsRNA in silencing does not exclude the possibility of transcriptional silencing caused by dsRNA-driven modification of the Stellate chromatin structure. Recent study indicates that dsRNA corresponding to promoter sequences may trigger transcriptional gene silencing in plants, accompanied by the appearance of small 21-25 nt RNA (Mette, 2000). The involvement of the same chromatin-remodeling and methylation proteins in posttranscriptional and transcriptional gene silencing indicates the interconnection of these mechanisms in plants. Components of RNAi machinery were also shown to be involved in the cosuppression caused by repetitive transgene arrays associated with a change in the chromatin structure of the array in the C. elegans germline (Dernburg, 2000). Further investigation must address the relationship between posttranscriptional and transcriptional mechanisms in Stellate silencing (Aravin, 2001).
The aubergine (sting) and spindle-E (homeless) mutants have been shown to upregulate Stellate expression (Schmidt, 1999; Stapleton, 2001). This study confirms the role of spn-E in Stellate silencing and shows that both mutations lead to an increase in the level of sense Su(Ste) transcripts. The spn-E, but not aub, mutation increases the expression of different retrotransposons and genomic tandem repeats (Aravin, 2001).
Both aub- and spn-E-encoded proteins control translation and localization of specific mRNAs during oogenesis. The Aubergine protein is homologous to C. elegans RDE-1, Neurospora crassa QDE-2, and Arabidopsis AGO-1 proteins. All of these proteins have been shown to be involved in RNAi and PTGS phenomena. Recently described are the PAZ- and piwi-conserved protein domains, shared by numerous proteins including Aubergine, RDE-1, QDE-2, and AGO-1; these proteins are implicated in gene silencing and stem-cell maintenance in plants, fungi, and animals, but their precise biochemical function is unknown. The Dicer protein containing the PAZ domain was shown to operate in the cleavage of dsRNA to 21-23 nt fragments in the Drosophila cell culture. Spindle-E encodes putative RNA helicase with the DExH domain. RNA helicase has been postulated as a component of dsRNA-mediated silencing machinery that functions in dsRNA unwinding to provide sequence-specific target recognition. Recently, the smg-2 gene encoding the RNA helicase involved in the nonsense-mediated decay pathway has been shown to be required for the persistence of RNA interference in C. elegans (Aravin, 2001 and references therein).
While studying the role of aub and spn-E mutations in the relief of Stellate silencing, it was observed that these mutations increase the level of sense Su(Ste) transcripts, exerting no effect on antisense transcripts. This observation suggests that corresponding proteins are involved in the mechanism of silencing downstream of antisense RNA production. The expression of spn-E is restricted to the germline. Accordingly, no influence of spn-E mutation on the expression of Su(Ste) was detected in heads. However, an increased level of sense Su(Ste) transcripts in heads of cry1Y males compared to normal XY males indicates that other proteins might participate in Su(Ste) dsRNA-mediated repression in somatic tissues. This conclusion is in agreement with recent findings that the RNAi effect in somatic tissues of D. melanogaster may be produced by transgene-encoded dsRNA (Aravin, 2001 and references therein).
The expression of the aubergine-lacZ reporter construct has been detected only in a tip of testis in which the stem cells are situated, but expression of the Ste-lacZ reporter is observed in all germ cells of testes, except for a tip. If both reporter constructs reflect the natural expression patterns of aub and Stellate, then their expression areas are not spatially overlapping in testes. These results argue that Aubergine may participate in the earlier stage of the establishment of silencing and may be dispensable for the later steps, as has been shown for its C. elegans homolog, RDE-1, in RNAi (Aravin, 2001 and references therein).
There is ample evidence to implicate the operation of a host surveillance system, acting against mobile elements and viruses, as a natural function of dsRNA-mediated silencing. Homologous RNA-mediated silencing of the non-LTR retroposon I-element has been demonstrated in D. melanogaster, and an involvement of dsRNA has been proposed (See Malinsky, 2000). The spn-E, but not aubergine, mutation causes derepression of the mst40 genomic tandem repeats and a wide spectrum of non-LTR and LTR-containing retrotransposons in the D. melanogaster germline. The spn-E encoding putative RNA helicase inhibits the expression of repetitive elements, which may be considered as selfish, but exerts no effect on essential repetitive histone genes. A mutation in the mut6 gene encoding DEAH RNA helicase impairs PTGS in Chlamydomonas and leads to an increase in the steady-state level of transposable element transcripts by preventing their degradation. The Aubergine protein is involved in dsRNA-mediated Stellate repression, but seems to be unrelated to the silencing of retrotransposons and other genomic repeats in the germline. This observation is in accord with a report that the relief of transposon silencing is not observed in mutants of the rde-1 gene, the C. elegans homolog of aubergine involved in RNAi (Aravin, 2001 and references therein).
Double-stranded RNA may be considered as a signal for recognition and silencing of repetitive elements in a genome. Applying the technique used to detect the Su(Ste) dsRNA, preliminary results were obtained suggesting that the transcription of several D. melanogaster retrotransposons can potentially lead to the formation of dsRNA in the germline. Together with the observation of increasing retrotransposon transcript levels in the spn-E homozygous flies, this result suggests that retrotransposon expression in the Drosophila germline is controlled by the mechanism that is related, but not identical, to dsRNA-mediated silencing of Ste/Su(Ste) repeats. Antisense transcripts of mobile elements may be produced by a read-through mechanism from promoters of adjacent genes or by an internal antisense promoter. Intriguingly, the dsRNA-mediated mechanism of Stellate suppression might have evolved as a result of hoppel DNA transposon insertion in the genomic Su(Ste) repeats, leading to antisense transcription. The transcription of both sense and antisense strands, which provides a potential source of dsRNA, has been shown for numerous Drosophila transposable elements, including mdg1 retrotransposon and the F-element (Aravin, 2001).
Recent reports of RNAi experiments in mammals suggest the possibility that related silencing mechanisms of repetitive genes might exist in vertebrates. One may speculate that dsRNAs are implicated in the silencing of repetitive genes, since it has been shown that heterogeneous nuclear ribonucleoprotein (hnRNP) particles isolated from mammalian cells contain dsRNA; a significant part of these duplexes is represented by repetitive sequences. The future studies of RNAi-related phenomenon may result in unexpected findings, uncovering the role of dsRNA-mediated silencing in genome surveillance, especially in germ cell development (Aravin, 2001).
The D. melanogaster DNA segment in the recombinant phage lambda Dm2L1 contains at least eight copies of a tandemly repeated 1250-base pair (bp) sequence (henceforth called the 2L1 sequence). Testes from XO D. melanogaster males contain an abundant 800-base RNA species that is homologous to a 520-bp region of the 2L1 sequence. Blotting experiments show that the 2L1 sequence is repeated in the D. melanogaster genome and is present on both the X and Y chromosomes. With the use of X-Y translocations, the 2L1 sequence has been mapped to a region between kl-1 and kl-2 on the long arm of the Y chromosome. In Oregon-R wild type there are an estimated 200 copies of the 2L1 sequence on the X chromosome and probably at least 80 copies of the Y chromosome. In some other strains the repetition frequency on the Y chromosome is about the same, but the copy number on the X chromosome is much reduced. On the basis of the five strains investigated, there is a correlation between copy number of the 2L1 sequence on the X chromosome and the presence of a particular allele of the Stellate locus (Ste; 1-45.7). It seems that low copy number corresponds to Ste+ and high copy number corresponds to Ste mutant alleles. The Ste locus determines whether single or star-shaped crystals are observed in the spermatocytes of XO males. Studies using D. simulans and D. mauritiana DNA show that the 2L1 sequence is homologous to restriction fragments in male DNA but not female DNA, indicating that this sequence is present only on the Y chromosome in these two species. In DNA derived from D. erecta, D. teissieri and D. yakuba, there is very little, if any, hybridization with the 2L1 sequence probe (Livak, 1984).
The X-linked Stellate locus contains two major size classes of a tandemly repeated gene. The steady-state level of Stellate RNA is much higher in XO testis than in XY testis. Sequencing of six cDNA clones derived from XO testis RNA shows that there are two major introns in the Stellate genes (see GenBank record X15899). Primer extension and RNase protection analyses show that these introns are spliced much more efficiently in XO than in XY testis. These results also indicate the major transcriptional start site for Stellate RNA. P element transformation results using a marked Stellate gene demonstrate that at least one of the genes sequenced contains a functional promoter, which generates low levels of RNA in XY testis and high levels of RNA in XO testis. This promoter does not contain a TATA element in the -30 region relative to the transcriptional start. Previous results had implicated a specific region of the Y chromosome, designated here as the Su(Ste) locus, in the control of the Stellate genes on the X. Analysis using segmental Y deficiencies shows that the Su(Ste) region suppresses both the high levels and efficient splicing of Stellate RNA (Livak, 1990).
The heterochromatic array of Stellate repeats is divided into three regions by a 4.5-kb DNA segment of unknown origin and a retrotransposon insertion: the A region (approximately 14 Stellate genes), the adjacent B region (approximately three Stellate genes), and the C region (about four Stellate genes). The sequencing of Stellate copies located along the discontinuous cluster reveals a complex pattern of diversification. The lowest level of divergence is detected in nearby Stellate repeats. The marginal copies of the A region, truncated or interrupted by an insertion, escaped homogenization and demonstrate high levels of divergence. Comparison of copies in the B and C regions, which are separated by a retrotransposon insertion, reveals a high level of diversification. These observations suggest that homogenization takes place in the Stellate cluster, but that inserted sequences may impede this process (Tulin, 1997).
Sequence analysis reveals that the Ste protein shares extensive homolgy with the ß subunit of Casein kinase 2, an enzyme that is able to aggregate in filamentous structures and, among its multiple functions, is involved in regulating topoisomerase II activity, which is essential for chromosome condensation and segregation. Moreover, in vitro experiments have shown that this protein can interact with the catalytic ß subunit of casein kinase 2 enzyme, altering its activity (Bozzetti, 1995 and references therein).
The 30-kb cluster comprising close to 20 copies of tandemly repeated Stellate genes is localized in the distal heterochromatin of the X chromosome. Of 10 sequenced genes, nine contain undamaged open reading frames with extensive similarity to protein kinase CK2 beta-subunit; one gene is interrupted by an insertion (Tulin, 1997).
The peculiarities of molecular evolution and divergence of paralogous heterochromatic clusters of the testis-expressed X-linked Stellate and Y-linked Su(Ste) tandem repeats have been examined. It has been suggested that Stellate and Su(Ste) clusters affecting male fertility are the amplified derivatives of the unique euchromatic gene betaCK2tes (Suppressor of Stellate-like) encoding the putative testis-specific beta-subunit of protein kinase CK2. The putative Su(Ste)-like evolutionary intermediate, located at 60B--C, has been detected on the Y chromosome as an orphon outside of the Su(Ste) cluster (Note: An orphon is a gene identified in a chromosomal location outside the main loci). The orphon shows extensive homology to the Su(Ste) repeat, but contains several Stellate-like diagnostic nucleotide substitutions, as well as a 10-bp insertion and a 3' splice site of the first intron typical of the Stellate unit. The orphon looks like a pseudogene carrying a drastically damaged Su(Ste) open reading frame (ORF). The putative Su(Ste) ORF, as compared with the Stellate one, carries numerous synonymous substitutions leading to the major codon preference. It is concluded that Su(Ste) ORFs evolved on the Y chromosome under the pressure of translational selection. Direct sequencing shows that the efficiency of concerted evolution between adjacent repeats is 5-10 times as high in the Stellate heterochromatic cluster on the X chromosome as that in the Y-linked Su(Ste) cluster, judging by the frequencies of nucleotide substitutions and single-nucleotide deletions (Kogan, 2000).
An earlier described CK2(beta)tes gene of Drosophila melanogaster has been shown to encode a male germline specific isoform of regulatory beta subunit of casein kinase 2. Western analysis using anti-CK2(beta)tes Ig revealed CK2(beta)tes protein in Drosophila testes extract. Expression of a CK2(beta)tes-beta-galactosidase fusion protein driven by the CK2(beta)tes promoter was found in transgenic flies at postmitotic stages of spermatogenesis. Examination of biochemical characteristics of a recombinant CK2(beta)tes protein expressed in Escherichia coli revealed properties similar to those of CK2beta: (1) CK2(beta)tes protein stimulates CK2alpha catalytic activity toward synthetic peptide; (2) it inhibits phosphorylation of calmodulin and mediates stimulation of CK2alpha by polylysine; (3) it is able to form (CK2(beta)tes)2 dimers, as well as (CK2alpha)2(CK2(beta)tes)2 tetramers. Using the yeast two-hybrid system and coimmunoprecipitation analysis of protein extract from Drosophila testes, an association between CK2(beta)tes and CK2alpha has been demonstrated. Northern analysis has shown that another regulatory (beta') subunit found recently in D. melanogaster genome is also testis specific. Thus, this is the first example of two tissue specific regulatory subunits of CK2 that might serve to provide CK2 substrate recognition during spermatogenesis (Kalmykova, 2002).
date revised: 15 July 2002
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.