spindle E/homeless: Biological Overview | Evolutionary Homologs | Regulation | Developmental Biology | Effects of Mutation | References

Gene name - spindle E

Synonyms - homeless (hls)

Cytological map position - 89A5--89A6

Function - RNA helicase

Keywords - oogenesis

Symbol - spn-E

FlyBase ID: FBgn0004883

Genetic map position - 3-[58]

Classification - DE-H family of RNA-dependent ATPases

Cellular location - cytoplasmic

NCBI links: Precomputed BLAST | Entrez Gene
Recent literature
Ryazansky, S. S., Kotov, A. A., Kibanov, M. V., Akulenko, N. V., Korbut, A. P., Lavrov, S. A., Gvozdev, V. A. and Olenina, L. V. (2016). RNA helicase Spn-E is required to maintain Aub and AGO3 protein levels for piRNA silencing in the germline of Drosophila. Eur J Cell Biol [Epub ahead of print]. PubMed ID: 27320195
Germline-specific RNA helicase Spindle-E (Spn-E) is known to be essential for piRNA silencing in Drosophila that takes place mainly in the perinuclear nuage granules. Loss-of-function spn-E mutations lead to tandem Stellate genes derepression in the testes and retrotransposon mobilization in the ovaries. However, Spn-E functions in the piRNA pathway are still obscure. Analysis of total library of short RNAs from the testes of spn-E heterozygous flies revealed the presence of abundant piRNA ping-pong pairs originating from Su(Ste) transcripts. The abundance of these ping-pong pairs were sharply reduced in the library from the testes of spn-E mutants. Thus the ping-pong mechanism contributes to Su(Ste) piRNA generation in the testes. The lack of Spn-E caused a significant drop of protein levels of key ping-pong participants, Aubergine (Aub) and AGO3 proteins of PIWI subfamily, in the germline of both males and females, but did not disrupt of their assembly in nuage granules. Observed decline of the protein expression was not caused by suppression of aub and ago3 transcription as well as total transcription, indicating possible contribution of Spn-E to post-transcriptional regulation.

homeless, now termed spindle E, was initially detected in a P element insertion screen: a female sterile line was obtained in which the insertion mapped at 80A5-6. spindle E mutants contain mislocalized oocytes in a small percentage of their vitellogenic egg chambers. Ovaries dissected from mutants contain a range of late-stage phenotypes. A wild-type egg chamber at stage 14 of oogenesis possesses two dorsal eggshell respiratory appendages, just lateral to the dorsal midline. Ninety to ninety-five percent of the mutant egg chambers show aberrant appendage formation: the majority possess only one appendage or fused appendages emerging from one base on the dorsal midline. The dorsal appendage phenotype suggests that spn-E plays a role in dorsalization of the oocyte, while the mislocalization phenotypes suggest that spn-E is involved in an even earlier role in egg polarity (Gillespie, 1995).

spindle E encodes a DEAD box protein that is likely to function as an RNA helicase (an RNA unwinding function). The similarity of Spn-E to proteins that function through binding RNAs suggests a possible role for Spn-E in RNA processing, transport, or stabilization. The localization patterns of seven mRNAs known to be localized during oogenesis were examined. These transcripts fall into three classes: (1) those that fail to be transported or localized correctly in some fraction of spn-E egg chambers; (2) those that are localized correctly but are reduced in amount, and (3) those that remain unaffected in spn-E mutants. Gurken mRNA is localized appropriately in the majority of stage 9 and stage 10 spn-E mutant egg chambers. However, about 30% of hls mutant chambers reveal a defective pattern. Some show no GRK mRNA localization, some show an anterior ring, and some show a dorsal patch that is broader than normal. Although Oskar mRNA is transported to the oocyte normally in these early stages, later localization to the posterior pole is defective. In the majority of S10 egg chambers, OSK mRNA is diffuse throughout the oocyte. In spn-E mutant egg chambers clearly aberrant Bicoid mRNA distribution is observed in stage 8. In most egg chambers, some BCD mRNA is transported to the oocyte, but the majority remains in the nurse cells, concentrated at the apical cortex. In addition, much of the BCD message within the oocyte is not transported laterally toward the periphery but instead remains centrally located. The failure to transport BCD and OSK mRNAs is the earliest defect in RNA transport observed in spn-E mutants (Gillespie, 1995).

Examination of two other anteriorly localized transcripts, the K10 and ORB mRNAs, identifies a second class of messages. In spn-E mutants, K10 transcripts are transported to the oocyte and later localized to the anterior edge, but the level of transcript is greatly decreased relative to wild type, specifically in vitellogenic stages. In spn-E mutants ORB mRNA accumulates normally in oocytes, but the anterior localization, while present, is significantly weaker in stage 8 to stage 10 oocytes. Bic-D and HTS mRNAs are unaffected in spn-E mutants (Gillespie, 1995).

Because the Spn-E protein has homology to RNA-binding proteins of the DE-H family, it seems likely that Spn-E is acting directly to localize RNAs in oogenesis. It is possible, however, that mutations in the gene are affecting RNA localization by disrupting the microtubule architecture that is a common component of the RNA's localization mechanism. Hence, the localization of a Kinesin heavy chain:beta-Galactosidase fusion protein was examined in a spn-E mutant background to analyze one aspect of microtubule structure and function. In wild-type stage 8 and stage 9 egg chambers, microtubule organizing centers (MTOCs) are located at the anterior of the oocyte and direct the formation of a gradient of microtubules whose plus ends extend toward the posterior pole. The fusion protein is thus propelled by the plus-end-directed kinesin motor function and is present in a band at the posterior of the oocyte, as detected by beta-Gal activity on an X-Gal substrate. In spn-E mutants, the fusion protein is detected in the center of S8 oocytes and in central and lateral portions of S9 oocytes, suggesting that microtubule structure or function is disrupted. Microtubule organization was examined directly in spn-E mutant egg chambers by comparing anti-alpha-Tubulin immunofluorescence in wild-type and mutant ovaries. In a majority of stage 8 spn-E chambers, a dense network of microtubules is present throughout the oocyte in place of the normal anterior MTOC localization. In all cases, the oocyte nucleus is correctly positioned at the dorsal anterior corner of the oocyte. Despite the presence of an aberrant network in stage 8 and 9 egg chambers, a normal rearrangement was observed in stage 10b spn-E oocytes. Thus the inappropriate formation of an extensive microtubule meshwork is confined to stage 8 and stage 9 egg chambers (Gillespie, 1995).

Spn-E could be functioning in one of at least two ways: (1) Spn-E could act downstream of the signaling pathways that induce correct microtubule reorganization during stages 8 and 9, actively interacting in microtubule reorganization, or, (2) Spn-E may be required for efficient transcription, pre-mRNA processing, localization, or translational regulation of products that control the kinetics of microtubule assembly, or to direct the reorganization of microtubule structures. The similarity of Spn-E protein to members of the DE-H family makes this second hypothesis attractive (Gillespie, 1995).

Subsequent studies reveal that spn-E phenotypes resemble a family of Drosophila mutants, all involved in a number of polarity determining steps during oogenesis. During wild-type oogenesis, the two cells in each germline cyst appear to be equivalent: these are the progeny of the first division of the cystoblast, derived from asymmetric division of a germ-line stem cell. Both cells enter meiosis to become pro-oocytes in region 2a of the germarium. In region 2b, one of these two cells is selected to develop as the oocyte and remains in meiosis, while the other exits meiosis and reverts to the nurse cell pathway of development. The event gives rise to the first asymmetry in egg development, the selection of one of two cells to become the oocyte. Later in oogenesis, anterior-posterior polarity originates when the oocyte comes to lie posterior to the nurse cells and signals through the Gurken/Egfr pathway to induce the adjacent follicle cells to adopt a posterior fate. This directs the movement of the germinal vesicle and associated Gurken mRNA from the posterior to an anterior corner of the oocyte, where Gurken protein signals for a second time to induce the dorsal follicle cells, thereby polarizing the dorsal-ventral axis. A group of five genes, the spindle loci, is described which is required for each of these polarizing events. The five spindle genes were originally identified in a screen for maternal-effect mutants on the third chromosome because homozygous mutant females lay ventralized eggs.

Mutations in spn-E give rise to an oocyte displacement phenotype, but also affect the oocyte cytoskeleton and mRNA localization, even when the oocyte is at the posterior of the egg chamber. Double spindle mutants reveal a phenotype even earlier in oogenesis, one where both pro-oocytes develop as oocytes, by delaying the choice between these two cells. spindle mutants inhibit the induction of both the posterior and dorsal follicle cells by disrupting the localization and translation of Gurken mRNA. The transient mislocalization of Gurken mRNA to an anterior ring in spn mutant stage 9 egg chambers is very similar to the mislocalization of Gurken mRNA observed in fs(K10) mutants. However, K10 mutations produce a dorsalization of the egg chamber rather than a ventralisation, because the mislocalization of Gurken mRNA directs Gurken signaling to the follicle cells on all sides of the oocyte. In different spindle mutants, from 19% to 100% of egg chambers show a strong reduction or a complete absence of Gurken protein in the oocyte membrane. The oocyte often fails to reach the posterior of mutant egg chambers and it differentiates abnormally. This analysis of spindle phenotypes suggests that spindle genes are likely to be involved in the localization and/or translation of Gurken mRNA without having any discernible effect on the Gurken mRNA level, yet dramatically reduced amounts of Gurken protein are produced. K10 mutants cause a similar mislocalization of Gurken mRNA without significantly affecting protein expression. Thus, spindle mutants reveal a novel link between oocyte selection, oocyte positioning and axis formation in Drosophila, leading to a proposal that the spindle genes act in a process that is common to several of these events (Gonzalez-Reyes, 1997).

How the asymmetry between the two pro-oocytes arises is unknown, but it has been proposed that it could be generated during the first division of the cystoblast to give rise to a two-cell cyst. During this division, a vesicular structure called the spectrosome (see Drosophila Spectrin) associates with one pole of the mitotic spindle and is asymmetrically partitioned between the two daughter cells. Since each of these cells gives rise to one pro-oocyte and seven nurse cells, this asymmetry might determine which pro-oocyte is fated to become the oocyte. Whatever mechanism generates the initial asymmetry, it seems that the key step in the selection of the oocyte is the accumulation of Bicaudal-D and Egalitarian proteins in a single cell. Null mutants in either gene block the localization of the protein encoded by the other to the presumptive oocyte and prevent all other known steps in oocyte differentiation, such as the formation of an active MTOC in this single cell and the subsequent microtubule-dependent localization of oocyte-specific transcripts, such as Oskar. Although it is unclear at what point in the pathway of oocyte selection the spindle genes act, they must function upstream of the process that results in the localization of Bic-D (and presumably Egl) to a single cell. It is most likely that the spindle proteins are directly involved in this process: (1) the spn double mutant combinations delay but do not block the choice between the two pro-oocytes, suggesting that they do not remove the initial asymmetry, but slow down its expression. (2) A reduction in Bic-D or Egl activity later in oogenesis leads to the same ventralized phenotype that is produced by the single spn mutations. This raises the possibility that the spn gene products interact with Bic-D and Egl at two different stages of oogenesis, first to select the oocyte and then to regulate Gurken expression once the oocyte has formed (Gonzalez-Reyes, 1997).

Telomere elongation is under the control of the RNAi-based mechanism in the Drosophila germline; mutations in the spn-E and aub cause an increase in the frequency of telomeric element retrotransposition to a broken chromosome end

Telomeres in Drosophila are maintained by transposition of specialized telomeric retroelements HeT-A, TAHRE, and TART instead of the short DNA repeats generated by telomerase in other eukaryotes. This study implicates the RNA interference machinery in the control of Drosophila telomere length in ovaries. The abundance of telomeric retroelement transcripts is up-regulated owing to mutations in the spn-E and aub genes, encoding a putative RNA helicase and protein of the Argonaute family, respectively, which are related to the RNA interference (RNAi) machinery. These mutations cause an increase in the frequency of telomeric element retrotransposition to a broken chromosome end. spn-E mutations eliminate HeT-A and TART short RNAs in ovaries, suggesting an RNAi-based mechanism in the control of telomere maintenance in the Drosophila germline. Enhanced frequency of TART, but not HeT-A, attachments in individuals carrying one dose of mutant spn-E or aub alleles suggests that TART is a primary target of the RNAi machinery. At the same time, enhanced HeT-A attachments to broken chromosome ends were detected in oocytes from homozygous spn-E mutants. Double-stranded RNA (dsRNA)-mediated control of telomeric retroelement transposition may occur at premeiotic stages, resulting in the maintenance of appropriate telomere length in gamete precursors (Savitsky, 2006).

The problems of end-under-replication and stability of linear chromosomes are resolved by telomeres. The lengthening of terminal regions of linear eukaryotic chromosomes is often provided by RNA-templated addition of repeated DNA by reverse transcriptase enzyme, telomerase. In most eukaryotes, telomeric DNA is maintained by the action of telomerase, which is responsible for the synthesis of short 6-8-nucleotide (nt) arrays using an RNA component as a template. In contrast, telomeres of Drosophila are maintained as a result of retrotranspositions of specialized telomeric non-long-terminal repeat (LTR) HeT-A, TAHRE, and TART retrotranspositions (Biessmann, 1992b; Levis, 1993; for review, see Pardue, 2003; Abad, 2004b). Retrotransposons are also found in telomeric regions of such diverse organisms as Bombyx mori, Chlorella and Giardia lamblia. HeT-A, TAHRE, and TART are found at Drosophila telomeres in tandem arrays. HeT-A, the most abundant Drosophila telomeric element, contains a single ORF encoding a Gag-like RNA-binding protein, but lacks reverse transcriptase (RT). It is proposed that the RT necessary for its transposition might be provided in trans, perhaps by TART (Rashkova, 2002). TART ORF2 encodes a reverse transcriptase related to the catalytic subunit of telomerase. The recently discovered TAHRE element shows extensive similarity to HeT-A, but contains a second ORF, which encodes a reverse transcriptase (Abad, 2004b). A HeT-A promoter located in the 3' region of the element directs synthesis of a downstream neighbor (Danilevskaya, 1997). The TART element was shown to be transcribed bidirectionally using a putative internal sense promoter and antisense one that was localized within the 1-kb region of the TART 3' end (Danilevskaya, 1999). Maintenance of Drosophila telomere length is mediated by HeT-A and TART transpositions to chromosome ends as well as by terminal recombination/gene conversion (Mikhailovsky, 1999; Kahn, 2000). Most of the observed spontaneous attachments to telomeres are HeT-A transpositions (Biessmann, 1992a; Kahn, 2000; Golubovsky, 2001), but TART attachments (Sheen, 1994) were also detected (Savitsky, 2006 and references therein).

The spn-E and aub genes, encoding an RNA helicase and a protein of Argonaute family, respectively, are involved in double-stranded RNA (dsRNA)-triggered RNA interference (RNAi) in embryos, in transcriptional silencing of transgenes, and in the control of Drosophila retrotransposon transcript abundance in the germline, especially in ovaries. No effects of RNAi gene mutations on HeT-A and TART expression and telomere structure were observed in somatic tissues (Perrini, 2004). This study shows that increased HeT-A and TART transcript abundance in ovaries, owing to RNAi mutations, is correlated with a high frequency of telomeric element attachments to broken chromosome ends. Addition of HeT-A or TART to a truncated X chromosome, with a break in the upstream regulatory region of yellow, activates yellow expression in aristae, which enables monitoring of the elongation events (Kahn, 2000; Savitsky, 2002). Using this genetic system, the effects of RNAi mutations were studied on the frequency and molecular nature of telomeric attachments. A high frequency of TART but not HeT-A attachments in heterozygous RNAi mutants suggests that TART may be the primary target of the RNAi-based silencing mechanism. These results highlight for the first time the importance of TART, but not the more abundant HeT-A element, in Drosophila telomere maintenance. The disappearance of short TART and HeT-A RNAs was found in spn-E mutant ovaries, strongly suggesting an RNAi-based pathway in the control of telomere maintenance in the Drosophila germline (Savitsky, 2006).

An RNAi-based mechanism has been proposed to evolve in order to immobilize transposable elements and was found to control expression of endogenous transposable elements and their mobility in different species. Drosophila telomeres are maintained by successive transpositions of specialized telomeric retroelements HeT-A and TART. This study shows that transposition of both telomeric elements is under the control of the spn-E and aub genes, known to be related to the RNAi machinery. Hence, an RNAi-based mechanism may be considered not only as a defense against retrotransposon expansion, but also as a regulatory system responsible for proper telomere length maintenance in Drosophila (Savitsky, 2006).

spn-E is required for appropriate localization of mRNA and proteins involved in the establishment of axis formation in the embryo and encodes a member of the DEAD/DE-H protein family possessing RNA-binding and RNA helicase activity. aub encodes a protein of the Argonaute family that was shown to be a component of the RNAi effector complex RISC. aub and spn-E mutations strongly diminished effects of the injected dsRNA into mature oocytes. Both genes are implicated in small interfering RNA (siRNA)-dependent silencing of testis-expressed Stellate genes. Thus, spn-E and aub are components of RNAi-based silencing pathways in Drosophila. Mutations in these genes result in the derepression of a wide spectrum of retrotransposons in the germline, including the HeT-A telomeric element (Aravin, 2001; Stapleton, 2001; Kogan, 2003). This study demonstrates that spn-E and aub mutations increase the frequency of telomeric element retrotranspositions to broken chromosome termini, suggesting that the RNAi machinery controls telomere length in Drosophila (Savitsky, 2006).

Both telomeric elements are shown to be the targets of RNAi. The present results emphasize the differences in the response of HeT-A and TART elements to RNAi mutations. Surprisingly, two different spn-E mutant alleles and an aub mutation in the heterozygous state increase considerably TART mobility, whereas attachments of HeT-A to broken chromosome ends were detected much more rarely in spn-E1/+ ovaries and are not observed in ovaries of spn-Ehls3987/+ and aubQC42/+ flies. One copy of a spn-E mutation is sufficient to increase TART transcript abundance. Strong accumulation of HeT-A transcripts is found only in homozygous mutants, correlating with a high frequency of HeT-A attachments to the broken chromosome ends in the developing oocytes. This observation argues that TART is a primary target of the RNAi machinery in ovaries (Savitsky, 2006).

TART and HeT-A, in spite of sharing the region of integration, are dissimilar in their structure and expression strategy. While both sense and antisense TART transcription has been demonstrated, antisense transcripts are more abundant. In situ RNA analysis detected sense and antisense TART transcripts in the cytoplasm of nurse cells in the late-stage egg chambers, suggesting a possibility of dsRNA formation. However, it was found that the level of antisense TART transcripts is not affected in RNAi mutants. Only sense HeT-A transcription was observed by Northern or by in situ RNA analyses. Nevertheless, HeT-A- and TART-specific siRNAs were revealed among the cloned short RNA species in Drosophila, and short RNAs corresponding to both HeT-A and TART elements are detected by Northern analysis. Antisense HeT-A RNA is probably transcribed at a low level from an unidentified promoter, possibly, from the HeT-A internal region. Actually, a low level of antisense activity of the HeT-A 3' end has been observed . While TART transcripts were observed only in the nurse cells, HeT-A transcripts were detected both in the growing oocyte and nurse cells. It is proposed that TART is a primary target of the RNAi controlling system, since one dose of an RNAi mutation causes preferential TART, but not HeT-A, attachments to broken chromosome ends in ovaries. In contrast, one dose of a mutant Su(var)205 gene (HP1) considerably increasess the frequency of HeT-A rather than TART attachments to the chromosome ends (Savitsky, 2002). Thus, a specific effect of RNAi components on telomeric element expression is observed . Although TART copies are much less abundant in the genome than HeT-A and no TART elements are detected in some telomeres, TART is a conserved component of telomeres in distant Drosophila species. TART was considered as a source of RT production, thus ensuring retrotranspositions of both TART and HeT-A elements. One may propose that TART supplies an RNAi-regulated template for RT production, thus providing telomere-specific transpositions of both elements (Savitsky, 2006).

Drosophila telomeres contain a multisubunit protein complex forming a chromosome cap protecting chromosomes from DNA repair and end-to-end fusions. However, no HeT-A or TART sequences were detected at the stably maintained broken chromosome end, which is protected from telomere fusions. Thus, a sequence-independent system performs telomere capping functions. The capping complex contains HP1, HOAP (HP1/ORC associated protein), as well as ATM-kinase and DNA repair MRN complex and the Ku70/Ku80 heterodimer. HP1 and the Ku heterodimer act also as negative regulators of telomere elongation by retrotransposition of telomeric elements. Deficiencies that remove either the Ku70 or the Ku80 gene increase the transposition rate of HeT-A and TART elements but exert no effect on the HeT-A expression, suggesting that Ku proteins control the accessibility of the telomere to transposition events. At the same time, mutations in the Su(var)205 gene increase both transcript abundance of HeT-A and TART and the frequency of their attachments to chromosome ends. RNAi affects both telomeric retrotransposon expression and the rate of transposition to the telomere. Probably, this effect is mediated through HP1 recruitment and silencing of HeT-A and/or TART chromatin (Savitsky, 2006).

siRNAs produced from telomeric elements TART and HeT-A belong to the long size class (25-29 nt) in contrast to 21-22-nt RNAs guiding post-transcriptional RNAi. In plants, long siRNAs are associated with RNA-directed DNA methylation and play an essential role in the transcriptional retrotransposon silencing. dsRNA and proteins of the RNAi machinery can direct chromatin alteration to homologous DNA sequences and induce transcriptional silencing. RNAi mutations cause delocalization of HP1 in yeast and Drosophila. Actually, the increase in accessibility of HeT-A chromatin and its enrichment in K9-acetylated H3 histone were revealed in ovaries of spn-E mutants. It is also possible that TART and/or HeT-A short RNAs can be targeted to telomeric repeats in a transcriptional silencing complex (Savitsky, 2006).

RNAi disruption affects neither HeT-A and TART expression, nor telomere fusions in somatic cells. No effect was observed of spn-E mutations on HeT-A expression, even in actively dividing cells of imaginal discs, where HeT-A expression was found. The data indicate a crucial role of the RNAi machinery in the regulation of telomere elongation in germinal cells. The appearance of a cluster of individuals with identical retroelement attachments indicates that dsRNA-mediated control of terminal elongation may occur at premeiotic stages of oogenesis (Savitsky, 2006).

This study has demonstrated that expression and retrotransposition of specific telomeric repeats is under control of an RNAi-based system in the Drosophila germline. In this case, the telomerase-dependent mechanism of telomere stability is substituted by retrotranspositions. Interestingly, telomerase-dependent telomere functioning during meiosis in the yeasts Schizosaccharomyces pombe and Tetrahymena is also under the control of RNAi machinery. These observations and the current data indicate that dsRNA-mediated regulation of telomere dynamics in the germline may be a general phenomenon independent of a mode of telomere maintenance (Savitsky, 2006).


Transcript length - 4.6 kb and 11 kb (either a pre-mRNA or an alternatively spliced message)


Amino Acids - 4359

Structural Domains

Sequence analysis of the spn-E cDNA predicts a protein with amino-terminal homology to members of the DE-H family of RNA-dependent ATPases and putative helicases. Similarity of 51% in the amino-terminal third of the protein was found to two yeast splicing factors, PRP2 and PRP16, and to Drosophila Maleless (Mle) (Kuroda, 1991), which is required for dosage compensation. Within the seven broadly defined domains of these proteins, Spn-E shares 29% identity with PRP2, PRP16 and Mle. The C-terminal two-thirds of the predicted protein reveals no homology to other known protein sequences (Gillespie, 1995).

homeless/spindle E: Evolutionary Homologs | Regulation | Developmental Biology | Effects of Mutation | References

date revised: 6 February 98

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

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