Logo of narLink to Publisher's site
Nucleic Acids Res. 2009 Jan; 37(1): 268–278.
Published online 2008 Nov 26. doi:  10.1093/nar/gkn960
PMCID: PMC2615633

rasiRNA pathway controls antisense expression of Drosophila telomeric retrotransposons in the nucleus


Telomeres in Drosophila are maintained by the specialized telomeric retrotransposons HeT-A, TART and TAHRE. Sense transcripts of telomeric retroelements were shown to be the targets of a specialized RNA-interference mechanism, a repeat-associated short interfering (rasi)RNA-mediated system. Antisense rasiRNAs play a key role in this mechanism, highlighting the importance of antisense expression in retrotransposon silencing. Previously, bidirectional transcription was reported for the telomeric element TART. Here, we show that HeT-A is also bidirectionally transcribed, and HeT-A antisense transcription in ovaries is regulated by a promoter localized within its 3′ untranslated region. A remarkable feature of noncoding HeT-A antisense transcripts is the presence of multiple introns. We demonstrate that sense and antisense HeT-A-specific rasiRNAs are present in the same tissue, indicating that transcripts of both directions may be considered as natural targets of the rasiRNA pathway. We found that the expression of antisense transcripts of telomeric elements is regulated by the RNA silencing machinery, suggesting rasiRNA-mediated interplay between sense and antisense transcripts in the cell. Finally, this regulation occurs in the nucleus since disruption of the rasiRNA pathway leads to an accumulation of TART and HeT-A transcripts in germ cell nuclei.


RNA interference (RNAi) evolved as a defense mechanism against viruses that can generate double-stranded RNA (dsRNA). Long dsRNA is processed by a ribonuclease enzyme, Dicer, into small interfering RNAs (siRNAs), guiding degradation of homologous mRNA (1). This mechanism was adapted to control the expression and transposition of endogenous transposable elements in different organisms (2–12). A distinct class of short RNAs, repeat associated short interfering RNAs (rasiRNAs), 24–27 nt long, is involved in the silencing of retrotransposons in the Drosophila germ line (13). Importantly, rasiRNA production does not require processing of a dsRNA precursor by Dicer, a key enzyme for microRNA and siRNA production (13). Sense and antisense rasiRNAs specific to different classes of retrotransposons, long terminal repeat (LTR) and non-LTR elements are revealed in libraries of short RNAs (12,14) and by northern analysis (15,16). Although the mechanism of rasiRNA processing still remains unclear, it is obvious that antisense transcripts are important participants in this process.

First of all, antisense transcripts of transposable elements serve as a source for rasiRNA generation. A certain amount of rasiRNA is produced from heterochromatic loci enriched in damaged inactive copies of transposable elements (11,12). These small RNAs are believed to originate from a putative long, single-stranded precursor. The flamenco/COM locus has been shown to regulate the activity of the retroviral gypsy element (17) and two other retroelements, Idefix and ZAM (18). rasiRNAs originating from the flamenco locus are proposed to control the activities of Idefix, ZAM and gypsy retrotransposons in the Drosophila germline (12). Similar orientation of the copies of these elements in the flamenco locus results in the production of mainly antisense rasiRNAs. Some other described Drosophila and mouse loci generate small RNAs from both strands as a result of mixed orientations of transposable elements within the loci (11,12). Primary short RNAs guide the generation of additional short RNAs of both polarities. According to the ‘ping-pong’ amplification model, sense rasiRNAs result from the processing of long sense transcripts with the assistance of PIWI- or Aubergine (Aub)-associated antisense rasiRNAs and sense rasiRNAs in the complex with Ago3 guide the cleavage of antisense transcripts to produce additional antisense rasiRNAs (12,19). Being either the primary source of rasiRNA production or the template for subsequent amplification, antisense transcripts are apparently necessary for the RNA silencing process.

Antisense transcripts of transposable elements can be generated in different ways. They may be derived from read-through transcription from an adjacent external promoter, which has been documented, for example, in the Caenorhabditis elegans transposon Tc1 (4). Several examples of retrotransposon antisense internal promoters were described, although their functional significance is unclear. A testis-specific antisense promoter of the LTR retrotransposon micropia in Drosophila hydei drives expression of the transcript complementary to the reverse transcriptase open reading frame (ORF) (20). Antisense promoters located in the 5′ untranslated region (UTR) downstream of the sense transcription start site are found in the human non-LTR elements L1 (21–23) and the Drosophila F-element (24). Antisense L1-specific short RNAs complementary to the region of overlapping transcripts were shown to contribute significantly to the RNAi-mediated silencing of L1 (25). Additionally, the Chlamydomonas reinhardii element TOC1 (26), the Tripanosoma brucei retroelement TRS (27) and the Dictyostelium discoideum retroelement DRE (28) are transcribed in both directions. These data suggest multiple sources for transposon-specific antisense RNA origination.

The HeT-A, TART and TAHRE families of Drosophila non-LTR retrotransposons play important roles in the cell, providing telomere repair (29–32). Telomeres in Drosophila are maintained by transpositions of specialized telomeric retroelements, rather than by the telomerase activity that adds short DNA repeats to chromosome ends in other eukaryotes. These elements are found at Drosophila telomeres in mixed tandem head-to-tail arrays; their 3′-ends are orientated toward the centromere. Telomeric elements are characterized by the presence of long 3′- and 5′-UTRs and the unusual structure of their promoters. Most of the non-LTR retrotransposons utilize an internal promoter in order to transcribe a full-length RNA that serves as a template for transposition according to the target-primed reverse transcription mechanism (33–35). Promoters of HeT-A and TAHRE are localized in the 3′-UTR and drive transcription of a downstream element (32,36). The TART element was shown to be transcribed bidirectionally from internal sense and antisense promoters that were localized within nonterminal direct repeats in the TART 5′- and 3′-regions (37,38). Interestingly, TART antisense transcripts are spliced (38). The abundance of telomeric retroelement transcripts and the frequency of their transpositions onto chromosome ends are controlled by an rasiRNA-mediated mechanism (10,32). Nothing is yet known about the origin of antisense transcripts of the main structural telomeric element, HeT-A. Only sense HeT-A transcription was observed by northern (37) or in situ RNA hybridization analyses (10,39), while HeT-A antisense rasiRNAs were revealed to be among the cloned short RNA species in Drosophila (12,14) and by northern analysis (10,13).

Taking into account a crucial role of RNA silencing in the regulation of the telomeric element expression in the Drosophila germline (10), we studied HeT-A antisense expression as a putative source of antisense rasiRNAs in the ovaries. We mapped the antisense transcription start site to the 3′-UTR of this element. Multiple introns are present in the noncoding, antisense HeT-A transcripts, and their hypothetical significance is discussed.

Increasing numbers of natural sense–antisense transcript pairs are being identified in a variety of eukaryotic organisms (40–44). Experimental evidence suggests that some antisense transcripts may be involved in repression of the gene expression via natural RNAi (45–48). However, biogenesis of antisense RNAs is poorly understood. Endogenous antisense transcripts have never been considered as a natural target of the RNAi-based mechanism. Investigations are usually focused on the coding, sense transcripts, which, in the case of transposable elements, serve as a template for transpositions. However, their abundance is regulated by antisense short RNAs derived from long antisense transcripts. Here, we have shown that natural antisense transcripts, like the sense ones, are targets of the RNAi machinery. The abundance of antisense transcripts of Drosophila telomeric transposable elements is affected by mutations of the rasiRNA pathway genes. Thus, overall expression level of HeT-A and TART retrotransposons could be a result of a complex rasiRNA-mediated interplay of their sense and antisense transcripts. Our data demonstrate that this regulation occurs in the nucleus since the disruption of the rasiRNA pathway leads to the accumulation of TART and HeT-A transcripts in the germ cell nuclei in Drosophila.


Drosophila strains

Strains bearing spindle-E (spn-E) mutations were ru1 st1 spn-E1 e1 ca1/TM3, Sb1 es and ru1 st1 spn-Ehls3987 e1 ca1/TM3, Sb1 es. aub mutants were aubQC42/CyO and aubHN/CyO. vasa locus mutations were vigEP812 and vasPH165. These mutations affect two genes, vig and vasa, located at the same locus (49,50). We used piwi2 and piwi3 mutations and Drosophila melanogaster stock y1; cn1 bw1 sp1 (Bloomington Drosophila stock center #2057).

Northern analysis of short RNAs

Northern analysis of short RNAs was done as previously described (8). Ovarian RNAs enriched with short RNAs were isolated using the miRACLE miRNA isolation kit (Stratagene). P32-labeled riboprobes corresponding to sense or antisense strands of TART and HeT-A were synthesized. PCR fragments of retrotransposons cloned into pBS SK were used as transcription templates. The TART probe contained a fragment of the ORF2 corresponding to nucleotides 2377–2888 of the GenBank sequence DMU02279. The HeT-A probe contained a fragment of the 3′-UTR (nucleotides 6769–7092 of the GenBank sequence DMU06920). Hybridization with P32 end-labeled oligonucleotide 5′-ACTCGTCAAAATGGCTGTGATA-3′ complementary to mir13b1 or 5′-TACAACCCTCAACCATATGTAGTCCAAGCA-3′ complementary to 2S RNA was used as a loading control.

In situ RNA hybridization and immunohistochemistry

In situ RNA analysis was carried out according to the previously described procedure (51) using digoxigenin (DIG)-labeled strand-specific TART and HeT-A riboprobes (described above) and rhodamine- (diluted at 1/500) or alkaline phosphatase (AP)-conjugated (diluted at 1/2000) anti-DIG antibodies (Roche). For in situ RNA hybridization using AP-conjugated anti-DIG antibodies, the time of incubation with AP substrate was more than 2 h. An RNaseH control was done to exclude the possibility of RNA–DNA hybridization in these experiments. After hybridization and washing, ovaries were treated with RNaseH (0.3 U/μl) for 1 h at 37°C followed by immunodetection. The lacZ fragment was PCR amplified using 5′-TTCCAGTTCAACATCAGCCGCTAC-3′ and 5′-GTTGATGTCATGTAGCCAAATCGG-3′ primers, was cloned in pBS SK, and used to obtain a lacZ riboprobe. The combination of protein and RNA localization is described in Supplementary Materials and Methods. To stain DNA, ovaries were incubated in PBS containing 0.5 μg/ml DAPI. The ovaries were visualized by confocal microscopy Zeiss LSM510 Meta.

5′ RACE analysis

RNA samples were prepared from the dissected ovaries of D. melanogaster stock y1; cn1 bw1 sp1 and from Schneider2 (S2) cells transfected by the HeTA-s or HeTA-as constructs (∼7 × 106 cells) using TRIzol reagent (Invitrogen). RNA samples were treated with TURBO DNase (Ambion). 5′ RACE (rapid amplification of cDNA ends) analysis was performed using the 5′ RACE system (Invitrogen) according to the manufacturer's general guidelines. HeT-A-specific primers for the first strand cDNA synthesis and PCR (schematically represented in Figure 2) were: H-RT 5′-TACAATTTCCATGACGACTC-3′ and H4 5′-GGAACCCATCTTCAGAATTCCCTC-3′ or M1 5′-CTGTCTCCGTACCTCCACCAGC-3′. Primers for the first strand cDNA synthesis and PCR, corresponding to the CaSpeR-AUG-β-gal sequence, were 5′-GAAAATCACGTTCTTGTTG-3′ and 5′-TGGTCAAAGTAAACGACATGGTGAC-3′, respectively.

Figure 2.
HeT-A antisense transcripts contain multiple introns. Structures of HeT-A antisense transcripts deduced from the sequences of 5′ RACE products. The fragment of the telomeric clone ...

RT-PCR analysis

Semi-quantitative RT-PCR was done according to the previously described procedure (8). RNA samples were divided for RT and two RT+ reactions. cDNA was synthesized using a random primer or gene-specific primers. Using rp49 primers, amplification of the reverse transcription reaction obtained with a random primer was done as a loading control. It was observed that reverse transcriptase yields cDNA in the absence of primers, possibly as a result of unspecific priming. To avoid artifacts associated with this effect, we attached a universal adapter (5′-gcctgccccaacctcc-3′) to each gene-specific primer. A gene-specific part of such oligonucleotide serves as a primer for RT, and the adapter serves as a PCR primer. The results of the RT-PCR analysis were evaluated using the program ImageQuant5.0. Histograms display the quantification of at least three experiments.

The following primers were used (sequence of the adapter is given in bold):










Analysis of promoter activity

Constructs were made in the CaSpeR-AUG-β-gal vector. The HeT-A sequence was derived from clone z2 containing the fragment of the HeT-A element attached to a terminal deleted X chromosome (32). Clone z2 was used as a template to amplify 434 bp of the HeT-A promoter region. An HeT-A PCR fragment obtained with 5′-ATGAATTCATCCATCGCCCGCAACATG-3′ and 5′-ATGGATCCTTTGCTGGTGGAGGTACGGAGAC-3′ primers was inserted into the EcoRI and BamHI polylinker sites of CaSpeR-AUG-β-gal to obtain the HeTA-s construct. The HeT-A PCR fragment obtained with 5′-ATGAATTCTTTGCTGGTGGAGGTACGGAGAC-3′ and 5′-ATGGATCCATCCATCGCCCGCAACATG-3′ primers was inserted into the EcoRI and BamHI polylinker sites of CaSpeR-AUG-β-gal to obtain the HeTA-as construct. The constructs were verified by sequencing. Drosophila cell culture S2 was used for transfection, which was performed with FuGENE 6 Transfection Reagent (Roche) according to the manufacturer's recommendations. The β-galactosidase assay was done as previously described (52). Four experiments were done to evaluate the mean ± SD.


HeT-A antisense transcription start sites in ovaries

HeT-A antisense short RNAs were detected by northern analysis in Drosophila ovaries, using a probe corresponding to the ORF (10). In the ovarian short RNA libraries, antisense HeT-A rasiRNAs corresponding to 3′-UTR were also detected (12,14). Considering these data, we have proposed that antisense HeT-A RNA is transcribed from an unidentified promoter located at the 3′-UTR or just downstream. 5′ RACE analysis was applied to identify the HeT-A antisense transcription start site in Drosophila ovaries. Total RNA was extracted from ovaries of the D. melanogaster stock y1; cn1 bw1 sp1, which was used in genome sequencing. 5′ RACE analysis yielded a number of PCR products. Heterogeneity of the 5′ RACE products results from the alternative splicing of antisense transcripts (described below). We have found that antisense transcription is initiated at heterogeneous sites within the region ∼220 bp upstream of the polyadenylation site. Nucleotides 6870, 6872 and 6873 (according to HeT-A element 23Zn-1 from the genomic clone U06920, Figure 1) are frequently used as initiation sites of antisense transcription. The major antisense start site is located ∼190 nt upstream of the HeT-A sense transcription start site (36). However, one of the 5′ RACE products starts as far as 30 bp upstream of the sense transcription start site (position 7029, according to HeT-A element 23Zn-1 from the genomic clone U06920).

Figure 1.
Positions of the HeT-A antisense transcription initiation sites identified by 5′ RACE. A schematic representation of the HeT-A element is given. Positions of the sense transcription start site and polyadenylation site are indicated in accordance ...

In spite of the fact that primers used for 5′ RACE analysis match only a subset of HeT-A elements, the obtained products proved to be polymorphic, which allows us to conclude that antisense transcripts are derived from different elements. Most of them are highly homologous to HeT-A elements from the second and fourth chromosomes. For several other products we failed to determine the exact genomic origination. Analysis of 5′ RACE products that were obtained using a HeT-A-specific primer that anneals to the 3′-most end (Figure 2, primer M1) revealed that they had arisen from antisense transcription of the truncated HeT-A copy into an upstream HeT-A element. The sequence of such 5′ RACE products, including the boundary between HeT-A elements, aligns perfectly to the chromosome 4R telomeric clone (GenBank # AC010841; Figure 2), indicating that telomeric elements contribute to the overall level of HeT-A antisense expression. Thus, the antisense HeT-A promoter is localized at the 3′-UTR and is capable of driving read-through transcription from downstream into an upstream element in the telomere.

HeT-A antisense transcripts contain multiple introns

Sequencing of 25 5′ RACE clones revealed that most of the HeT-A antisense transcripts are spliced. Figure 2 shows structures of HeT-A antisense transcripts deduced from the sequences of 5′ RACE products. A fragment of the telomeric clone AC010841, containing the boundary between two HeT-As, was chosen to represent a genomic HeT-A sequence (Figure 2). The right panel of Figure 2 (transcripts 6–9) shows read-through HeT-A antisense transcripts (described above) that were derived from these particular HeT-A elements. In these transcripts, different patterns of intron splicing are detected, suggesting the existence of alternative ways in which transcripts derived from a single element are processed. It is noteworthy, that other antisense HeT-A transcripts (transcripts 1–5, aligned to the left HeT-A copy in Figure 2) are heterogeneous and have different genomic origins. The HeT-A copy represented in Figure 2 has been chosen for alignment as it is the most similar to transcripts 1–5 (Figure 2, left panel). Good matches to splice site consensus sequences are present at the exon/intron boundaries for all revealed introns in the HeT-A antisense transcripts. In some cases, intron sizes vary in different HeT-A copies (Figure 2). This was expected, taking into account the sequence polymorphism of HeT-A 3′-UTRs. Splicing in different polymorphic HeT-A copies is likely to occur at cryptic splice sites. Furthermore, introns were identified in the GenBank cDNA and EST clones, which correspond to HeT-A antisense transcripts (Supplementary Figure 1). HeT-A antisense transcripts are noncoding, as no long ORFs were found among these transcripts.

The 3′-end of HeT-A has antisense promoter activity

HeT-A sense promoter has been mapped to the 3′-end of the element (36). This promoter drives expression of a downstream element in a telomere. We have studied the antisense promoter activity of the HeT-A 3′-region, using transfection of cultured Drosophila cells by reporter constructs. A fragment of the HeT-A 3′-UTR containing 434 bp upstream of the polyadenylation site was cloned in both the sense and antisense orientations into the pCaSpeR-AUG-β-gal vector containing the lacZ reporter gene (HeTA-s and HeTA-as constructs, respectively, Figure 3). The enzyme activity was evaluated in extracts of transfected S2 cells. We detected low but reproducible reporter gene activity of the HeTA-as construct, which was evaluated at about 15% activity of the sense promoter, according to β-galactosidase activity assay (Figure 3). To determine the transcription start site in the HeTA-as construct, we performed the 5′ RACE analysis on the RNA isolated from S2 cells transfected by this construct. Primers corresponding to the lacZ reporter gene were used. We have found that transcription of the fused HeT-A-lacZ transcript initiates at heterogeneous start sites within region 217–235 nt upstream of the poly(A) site. Positions of antisense transcription start sites in the construct correspond closely to those determined for endogenous HeT-A copies (Figure 1). Also, like endogenous HeT-A antisense transcripts, transcripts from the HeTA-as construct are spliced (Figure 3). We performed 5′ RACE for the HeTA-s construct using the same conditions and found that, in this case, transcription is initiated at two major sites, 30–32 and 90–95 nt upstream of the HeT-A polyadenylation site (not shown). These data coincide well with the previously determined HeT-A transcription start sites (36,38) and confirm the reliability of the 5′ RACE data for HeT-A antisense transcription.

Figure 3.
Antisense promoter activity of the HeT-A 3′-region. HeTA-s and HeTA-as constructs contain the HeT-A 3′ fragment cloned in sense and antisense orientation, respectively, in the pCaSpeR-AUG-β-gal vector. Enzymatic activity of the ...

HeT-A rasiRNAs in RNAi mutants

Northern blot analysis was used to reveal HeT-A-specific rasiRNAs in the ovaries. Riboprobes containing sense or antisense strand sequences corresponding to the 3′-UTR were used. We detected HeT-A rasiRNAs of both polarities in the ovaries of spn-E1/TM3, aubQC42/Cy and piwi2/Cy flies. No sense or antisense HeT-A rasiRNAs were detected in the spn-E1/spn-Ehls3987 and piwi2/piwi3 ovaries. A decreased amount of HeT-A rasiRNA was observed in aubQC42/aubHN ovaries (Figure 4).

Figure 4.
Effect of mutations of the rasiRNA pathway genes on the presence of sense and antisense HeT-A-specific rasiRNAs in ovaries. Northern analysis of the RNA isolated from ovaries of spn-E1/TM3, spn-E1/spn-Ehls3987, piwi2/CyO, piwi2/piwi3, aubQC42/CyO and ...

In the ovarian short RNA libraries, both sense and antisense rasiRNAs were found for HeT-A and TART elements (12). The presence of the sense HeT-A and TART rasiRNAs suggests that the antisense transcripts are the targets of the RNAi machinery.

HeT-A and TART antisense expression in RNAi mutants

According to previously obtained data, sense HeT-A and TART transcripts accumulated in ovaries of the rasiRNA pathway component mutants (10,50), however, the patterns of their expression are different. HeT-A transcripts were detected both in oocyte and nurse cells, whereas TART transcripts substantially accumulated in nurse cells at the late stages of oogenesis. In parallel experiments, we failed to detect changes in the abundance and localization of HeT-A and TART antisense transcripts in the rasiRNA pathway mutants using in situ RNA hybridization with AP-conjugated anti-DIG antibodies (10). No sense and antisense HeT-A and TART transcripts were detected in the ovarian nuclei. However, in situ RNA hybridization using AP-conjugated anti-DIG antibodies and prolonged time of incubation with the AP substrate allowed us to reveal the accumulation of antisense HeT-A and TART transcripts in the ovaries with a compromised rasiRNA pathway (Figure 5, Supplementary Figure 2). Next, RNA fluorescence in situ hybridization (RNA FISH) was applied to compare cellular localization of sense and antisense HeT-A and TART transcripts in the ovaries of RNAi mutants.

Figure 5.
HeT-A and TART sense and antisense transcripts accumulate in spn-E mutants. The results were obtained by in situ RNA hybridization with HeT-A or TART riboprobes using anti-DIG AP-conjugated antibodies (A and B, upper panels) or anti-DIG rhodamin-conjugated ...

Both HeT-A (Figure 5A) and TART (Figure 5B) antisense transcripts accumulated in the nuclei of nurse cells in the developing egg chambers of spn-E/spn-E mutants. Signals corresponding to the TART antisense transcripts are visible as bright dots in the nuclei of nurse cells (Figure 5B). HeT-A antisense transcripts are found in nuclei and cytoplasm of the nurse cells in spn-E mutants (Figure 5A). No hybridization signal was detected with a probe for the lacZ, which indicates the specificity of hybridization (not shown). Accumulation of TART antisense transcripts in the nuclei of nurse cells of aub and vasa locus mutants was detected (Supplementary Figure 2).

As demonstrated previously, sense HeT-A transcripts are localized in the growing oocyte in rasiRNA pathway component mutants (10,50). We showed that, in spn-E/spn-E mutants, HeT-A sense transcripts are localized in the cytoplasm of the oocyte where the Orb protein is detected (Figure 5E). It is noteworthy that HeT-A and TART sense transcripts are accumulated not only in cytoplasm of germ cells as was shown previously, but also in nurse cell nuclei in spn-E/spn-E mutants (Figure 5C, D). HeT-A sense transcripts from nurse cells are most likely to be effectively transported to the oocyte. TART sense transcripts are accumulated in the nurse cell cytoplasm and nuclei at later stages of oogenesis and are not detected in oocytes of spn-E/spn-E mutants (Figure 5D). It is significant that hybridization conditions allowed us to detect RNA rather than genomic DNA. The treatment of ovaries by RNaseH after RNA hybridization did not affect the intensity of the nuclear signals (data not shown). Different patterns of hybridization with complementary sense and antisense single-stranded TART riboprobes is also a strong argument against the possibility of RNA–DNA hybridization. Actually, TART antisense transcripts are detected in nurse cell nuclei at different stages of oogenesis (Figure 5B), while sense transcripts are revealed in nuclei of nurse cells only at the late stage (Figure 5D).

The RT-PCR analysis of the cellular fractions of ovarian RNA was done to compare HeT-A and TART sense and antisense transcript abundance in spn-E mutants. The purity of cellular fractionation was determined by western and northern blot analysis (Supplementary Figure 3). Using RT-PCR with strand-specific primers, we detected substantial accumulation of HeT-A and TART antisense transcripts in the nuclear fraction of the ovarian RNA from spn-E/spn-E females (Figure 6). Sense transcript abundance in nuclei is also controlled by the spn-E gene, according to the RT-PCR analysis with strand-specific primers (Figure 6). These data confirm the accumulation of sense and antisense HeT-A and TART transcripts in ovarian nuclei of transheterozygous spn-E mutants obtained by RNA FISH analysis (Figure 5). The main difference between sense and antisense transcript behavior in RNAi mutants is a less pronounced accumulation of antisense transcripts in the cytoplasm than in the nuclei of mutant ovaries (Figure 6). Northern blot analysis confirms accumulation of sense HeT-A and TART transcripts and antisense TART transcripts in ovaries of spn-E/spn-E flies (Supplementary Figure 4). Long antisense HeT-A transcripts are not detected in this experiment. TART antisense transcripts are more abundant than sense ones in the ovaries of spn-E/+ flies (Supplementary Figure 4). This observation corresponds well with the previously revealed ratio of sense to antisense TART transcripts in wild type flies (37). Thus, HeT-A and TART antisense transcripts are the targets of the RNAi-based machinery and accumulated predominantly in the nuclear compartment of rasiRNA pathway mutant ovaries.

Figure 6.
Cellular distribution of sense and antisense HeT-A and TART transcripts in the ovaries of spn-E mutants. RT-PCR analysis with strand-specific primers was done to compare HeT-A and TART sense and antisense transcript amounts in the total ovarian RNA (T), ...

Accumulation of telomeric element transcripts of both polarities in ovarian nuclei of RNAi mutants suggests that the RNA silencing machinery and short RNAs guiding its specificity should be present in the nuclei. The intracellular distribution of HeT-A-specific rasiRNAs of both polarities was investigated by northern analysis of RNA from ovarian cell fractions. Distributions of sense and antisense HeT-A rasiRNAs do not differ markedly; both of them are present in cytoplasmic and nuclear fractions (Supplementary Figure 5). Thus, our results indicate that RNAi-based regulation of retrotransposon expression may occur in the nuclear compartment through the assistance of the rasiRNAs.


In this study, we have investigated the peculiarities of the antisense expression of the telomeric retrotransposons in the Drosophila germline. We have demonstrated the antisense expression of non-LTR telomeric retrotransposon HeT-A from the internal promoter. Taking into account previously reported antisense expression of other retroelements, including telomeric Drosophila element TART (38), Drosophila F-element (24) and human L1 (21), it becomes clear that bidirectional transcription, at least of non-LTR retrotransposons, is an intrinsic feature of the retrotransposon expression. Sense and antisense promoters of the investigated retroelements are located in close proximity to each other. In this situation, it looks as if the enhancer-like element drives bidirectional expression of these elements. Enhancer-like elements, capable of functioning in either orientation, were identified in promoter regions of the Drosophila F-element and human L1 (53,54). For TART and HeT-A retrotransposons, a presence of such elements remains to be determined. Despite of the proximity of the sense and antisense TART promoters, they are activated at different stages of oogenesis (Figure 5B, D). Sense TART transcripts are detected only at late stages of oogenesis, while antisense transcripts begin to appear at the earlier stages. Antisense Su(Ste) transcripts involved in the RNAi-mediated silencing of testis-expressed Stellate genes were shown to accumulate in the nuclei of early spermatocytes before the appearance of sense transcripts (55). Thus, sense and antisense transcripts are coordinately expressed. Considering the earlier appearance of antisense transcripts, it is tempting to speculate that they might also play a role in short antisense RNA generation prior to sense transcript accumulation; this may occur with the assistance of the ‘ping-pong’ mechanism, guided by the maternally submitted sense rasiRNAs (12,56,57).

Another enigmatic feature of the antisense expression of retrotransposons is the presence of introns in antisense transcripts. Sense transcripts that serve as a template for transposition are usually not spliced, although such examples have been described (58–60). Antisense transcripts are spliced in alternative ways, as has been demonstrated for the Drosophila TART (38) and HeT-A (present study) elements and human L1 (21,23). The role of this splicing for retrotransposon silencing, if any, remains elusive.

5′ RACE analysis of endogenous HeT-A antisense transcripts revealed transcripts of a telomeric nature: specifically, read-through transcripts from the fourth chromosome. However, we were not able to identify some of the 5′ RACE sequences with the particular HeT-A copies despite the fact that the D. melanogaster stock with the sequenced genome was used in our study. HeT-A-related sequences were previously reported to be present in the pericentromeric regions of the Drosophila chromosomes (61). We do not exclude the possibility of a heterochromatic origin of the portion of antisense HeT-A transcripts.

We have shown for the first time that endogenous antisense transcripts of transposable elements are the targets of the RNAi machinery. Thus, sense expression and antisense expression, being the source of short RNAs of corresponding polarities, are mutually controlled via an RNAi-based mechanism. An interesting feature of the antisense transcript silencing is nuclear compartmentalization of the process. Vast amounts of natural antisense transcripts identified in the mouse transcriptome are present in the nucleus (62). Nuclear localization of antisense testes-expressed Su(Ste) repeat transcripts was previously reported (55). Northern analysis detected TART antisense transcripts in the nuclear fraction of the Drosophila cells (37). We have demonstrated the accumulation of the antisense retrotransposon transcripts in the nucleus as a result of rasiRNA pathway disruption. This suggests that an important pathway of the rasiRNA mechanism operates in the nucleus and participates in the biogenesis of antisense RNAs and possibly in the generation of a primary pool of rasiRNAs. Surprisingly, we did not observe significant accumulation of antisense RNAs in the cytoplasm, where they serve as a substrate for the ‘ping-pong’ amplification loop. The presence of multiple introns in antisense transcripts suggests a role of processing for the stabilization and nuclear-to-cytoplasm transport of these transcripts. Most likely, these transcripts are the targets of the nonsense-codon mediated decay (NMD) machinery in the cytoplasm (63) because the antisense HeT-A and TART transcripts are noncoding. We believe that the regulation of the antisense expression may be a result of cumulative effects of both NMD and rasiRNA pathways. The role of the NMD machinery in the antisense retrotransposon transcript abundance remains to be determined.

RNAi-based machinery operates in the nucleus to induce transcriptional silencing through DNA or histone modification in plants, mammals and yeast (64). Evidence of the nuclear RNAi-mediated degradation of the target in human cells was previously reported (65). In Drosophila, the rasiRNA pathway component spn-E causes changes in the chromatin state of retroelements in ovaries (66). We have demonstrated that HeT-A and TART transcripts of both polarities accumulate in the germ cell nuclei as a result of rasiRNA pathway mutations. This fact clearly indicates that this mechanism controls transcript abundance in the nucleus. HeT-A and especially TART transcript hybridization signals in nuclei look like distinct bright dots, which may reflect the accumulation of nascent transcripts at the sites of transcription. This observation argues in favor of a transcriptional or co-transcriptional mechanism of regulation. The revealed nuclear population of the rasiRNAs may be involved in the rasiRNA-mediated silencing in the nucleus. The ratio between the amount of sense and antisense HeT-A-specific RNAs found in the nucleus and cytoplasm, however, appears to be the same. Most components of the Drosophila rasiRNA pathway were shown to be localized in the perinuclear cytoplasmic granules known as nuage (67–70), except for the nuclear PIWI protein (71). Thus, a distinct nuclear rasiRNA pathway may not exist; rather, a crosstalk between the cytoplasmic and nuclear rasiRNA pathway components ensures retrotransposon silencing in cell.


Supplementary Data are available at NAR Online.


RAS program for Molecular and Cell Biology (to A.K.); Russian Foundation for Basic Researches (06-04-48493); Russian Foundation for Science School (3464.2008.4). Funding for open access charge: Russian Foundation for Basic Researches (06-04-48493).

Conflict of interest statement. None declared.

Supplementary Material

[Supplementary Data]


We thank V. Gvozdev for critical reading of this article, I. Olovnikov for helpful discussion, Y. Abramov for the assistance with fly maintenance and crosses and S. Lavrov, L. Olenina and A. Stolyarenko for methodological advices.


1. Hammond SM. Dicing and slicing: the core machinery of the RNA interference pathway. FEBS Lett. 2005;579:5822–5829. [PubMed]
2. Baulcombe D. RNA silencing in plants. Nature. 2004;431:356–363. [PubMed]
3. Tabara H, Sarkissian M, Kelly WG, Fleenor J, Grishok A, Timmons L, Fire A, Mello CC. The rde-1 gene, RNA interference, and transposon silencing in C. elegans. Cell. 1999;99:123–132. [PubMed]
4. Sijen T, Plasterk RH. Transposon silencing in the Caenorhabditis elegans germ line by natural RNAi. Nature. 2003;426:310–314. [PubMed]
5. Shi H, Djikeng A, Tschudi C, Ullu E. Argonaute protein in the early divergent eukaryote Trypanosoma brucei: control of small interfering RNA accumulation and retroposon transcript abundance. Mol. Cell Biol. 2004;24:420–427. [PMC free article] [PubMed]
6. Svoboda P, Stein P, Anger M, Bernstein E, Hannon GJ, Schultz RM. RNAi and expression of retrotransposons MuERV-L and IAP in preimplantation mouse embryos. Dev. Biol. 2004;269:276–285. [PubMed]
7. Wu-Scharf D, Jeong B, Zhang C, Cerutti H. Transgene and transposon silencing in Chlamydomonas reinhardtii by a DEAH-box RNA helicase. Science. 2000;290:1159–1162. [PubMed]
8. Aravin AA, Naumova NM, Tulin AV, Vagin VV, Rozovsky YM, Gvozdev VA. Double-stranded RNA-mediated silencing of genomic tandem repeats and transposable elements in the D. melanogaster germline. Curr. Biol. 2001;11:1017–1027. [PubMed]
9. Kalmykova AI, Klenov MS, Gvozdev VA. Argonaute protein PIWI controls mobilization of retrotransposons in the Drosophila male germline. Nucleic Acids Res. 2005;33:2052–2059. [PMC free article] [PubMed]
10. Savitsky M, Kwon D, Georgiev P, Kalmykova A, Gvozdev V. Telomere elongation is under the control of the RNAi-based mechanism in the Drosophila germline. Genes Dev. 2006;20:345–354. [PMC free article] [PubMed]
11. Aravin AA, Sachidanandam R, Girard A, Fejes-Toth K, Hannon GJ. Developmentally regulated piRNA clusters implicate MILI in transposon control. Science. 2007;316:744–747. [PubMed]
12. Brennecke J, Aravin AA, Stark A, Dus M, Kellis M, Sachidanandam R, Hannon GJ. Discrete small RNA-generating loci as master regulators of transposon activity in Drosophila. Cell. 2007;128:1089–1103. [PubMed]
13. Vagin VV, Sigova A, Li C, Seitz H, Gvozdev V, Zamore PD. A distinct small RNA pathway silences selfish genetic elements in the germline. Science. 2006;313:320–324. [PubMed]
14. Aravin AA, Lagos-Quintana M, Yalcin A, Zavolan M, Marks D, Snyder B, Gaasterland T, Meyer J, Tuschl T. The small RNA profile during Drosophila melanogaster development. Dev. Cell. 2003;5:337–350. [PubMed]
15. Blumenstiel JP, Hartl DL. Evidence for maternally transmitted small interfering RNA in the repression of transposition in Drosophila virilis. Proc. Natl Acad. Sci. USA. 2005;102:15965–15970. [PMC free article] [PubMed]
16. Pelisson A, Sarot E, Payen-Groschene G, Bucheton A. A novel repeat-associated small interfering RNA-mediated silencing pathway downregulates complementary sense gypsy transcripts in somatic cells of the Drosophila ovary. J. Virol. 2007;81:1951–1960. [PMC free article] [PubMed]
17. Pelisson A, Song SU, Prud'homme N, Smith PA, Bucheton A, Corces VG. Gypsy transposition correlates with the production of a retroviral envelope-like protein under the tissue-specific control of the Drosophila flamenco gene. EMBO J. 1994;13:4401–4411. [PMC free article] [PubMed]
18. Desset S, Meignin C, Dastugue B, Vaury C. COM, a heterochromatic locus governing the control of independent endogenous retroviruses from Drosophila melanogaster. Genetics. 2003;164:501–509. [PMC free article] [PubMed]
19. Gunawardane LS, Saito K, Nishida KM, Miyoshi K, Kawamura Y, Nagami T, Siomi H, Siomi MC. A slicer-mediated mechanism for repeat-associated siRNA 5′ end formation in Drosophila. Science. 2007;315:1587–1590. [PubMed]
20. Lankenau S, Corces VG, Lankenau DH. The Drosophila micropia retrotransposon encodes a testis-specific antisense RNA complementary to reverse transcriptase. Mol. Cell Biol. 1994;14:1764–1775. [PMC free article] [PubMed]
21. Speek M. Antisense promoter of human L1 retrotransposon drives transcription of adjacent cellular genes. Mol. Cell Biol. 2001;21:1973–1985. [PMC free article] [PubMed]
22. Matlik K, Redik K, Speek M. L1 antisense promoter drives tissue-specific transcription of human genes. J. Biomed. Biotechnol. 2006;2006:71753. [PMC free article] [PubMed]
23. Nigumann P, Redik K, Matlik K, Speek M. Many human genes are transcribed from the antisense promoter of L1 retrotransposon. Genomics. 2002;79:628–634. [PubMed]
24. Minchiotti G, Di Nocera PP. Convergent transcription initiates from oppositely oriented promoters within the 5′ end regions of Drosophila melanogaster F elements. Mol. Cell Biol. 1991;11:5171–5180. [PMC free article] [PubMed]
25. Yang N, Kazazian HH., Jr L1 retrotransposition is suppressed by endogenously encoded small interfering RNAs in human cultured cells. Nat. Struct. Mol. Biol. 2006;13:763–771. [PubMed]
26. Day A, Rochaix JD. Conservation in structure of TOC1 transposons from Chlamydomonas reinhardtii. Gene. 1991;104:235–239. [PubMed]
27. Murphy NB, Pays A, Tebabi P, Coquelet H, Guyaux M, Steinert M, Pays E. Trypanosoma brucei repeated element with unusual structural and transcriptional properties. J. Mol. Biol. 1987;195:855–871. [PubMed]
28. Schumann G, Zundorf I, Hofmann J, Marschalek R, Dingermann T. Internally located and oppositely oriented polymerase II promoters direct convergent transcription of a LINE-like retroelement, the Dictyostelium repetitive element, from Dictyostelium discoideum. Mol. Cell Biol. 1994;14:3074–3084. [PMC free article] [PubMed]
29. Biessmann H, Champion LE, O’Hair M, Ikenaga K, Kasravi B, Mason JM. Frequent transpositions of Drosophila melanogaster HeT-A transposable elements to receding chromosome ends. EMBO J. 1992;11:4459–4469. [PMC free article] [PubMed]
30. Levis RW, Ganesan R, Houtchens K, Tolar LA, Sheen FM. Transposons in place of telomeric repeats at a Drosophila telomere. Cell. 1993;75:1083–1093. [PubMed]
31. Abad JP, De Pablos B, Osoegawa K, De Jong PJ, Martin-Gallardo A, Villasante A. TAHRE, a novel telomeric retrotransposon from Drosophila melanogaster, reveals the origin of Drosophila telomeres. Mol. Biol. Evol. 2004;21:1620–1624. [PubMed]
32. Shpiz S, Kwon D, Uneva A, Kim M, Klenov M, Rozovsky Y, Georgiev P, Savitsky M, Kalmykova A. Characterization of Drosophila telomeric retroelement TAHRE: transcription, transpositions, and RNAi-based regulation of expression. Mol. Biol. Evol. 2007;24:2535–2545. [PubMed]
33. McLean C, Bucheton A, Finnegan DJ. The 5′ untranslated region of the I factor, a long interspersed nuclear element-like retrotransposon of Drosophila melanogaster, contains an internal promoter and sequences that regulate expression. Mol. Cell Biol. 1993;13:1042–1050. [PMC free article] [PubMed]
34. Minchiotti G, Contursi C, Di Nocera PP. Multiple downstream promoter modules regulate the transcription of the Drosophila melanogaster I, Doc and F elements. J. Mol. Biol. 1997;267:37–46. [PubMed]
35. Swergold GD. Identification, characterization, and cell specificity of a human LINE-1 promoter. Mol. Cell Biol. 1990;10:6718–6729. [PMC free article] [PubMed]
36. Danilevskaya ON, Arkhipova IR, Traverse KL, Pardue ML. Promoting in tandem: the promoter for telomere transposon HeT-A and implications for the evolution of retroviral LTRs. Cell. 1997;88:647–655. [PubMed]
37. Danilevskaya ON, Traverse KL, Hogan NC, DeBaryshe PG, Pardue ML. The two Drosophila telomeric transposable elements have very different patterns of transcription. Mol. Cell Biol. 1999;19:873–881. [PMC free article] [PubMed]
38. Maxwell PH, Belote JM, Levis RW. Identification of multiple transcription initiation, polyadenylation, and splice sites in the Drosophila melanogaster TART family of telomeric retrotransposons. Nucleic Acids Res. 2006;34:5498–5507. [PMC free article] [PubMed]
39. Walter MF, Biessmann H. Expression of the telomeric retrotransposon HeT-A in Drosophila melanogaster is correlated with cell proliferation. Dev. Genes Evol. 2004;214:211–219. [PubMed]
40. Misra S, Crosby MA, Mungall CJ, Matthews BB, Campbell KS, Hradecky P, Huang Y, Kaminker JS, Millburn GH, Prochnik SE, et al. Annotation of the Drosophila melanogaster euchromatic genome: a systematic review. Genome Biol. 2002;3 RESEARCH0083. [PMC free article] [PubMed]
41. Yelin R, Dahary D, Sorek R, Levanon EY, Goldstein O, Shoshan A, Diber A, Biton S, Tamir Y, Khosravi R, et al. Widespread occurrence of antisense transcription in the human genome. Nat. Biotechnol. 2003;21:379–386. [PubMed]
42. Kiyosawa H, Yamanaka I, Osato N, Kondo S, Hayashizaki Y. Antisense transcripts with FANTOM2 clone set and their implications for gene regulation. Genome Res. 2003;13:1324–1334. [PMC free article] [PubMed]
43. Osato N, Yamada H, Satoh K, Ooka H, Yamamoto M, Suzuki K, Kawai J, Carninci P, Ohtomo Y, Murakami K, et al. Antisense transcripts with rice full-length cDNAs. Genome Biol. 2003;5:R5. [PMC free article] [PubMed]
44. Yamada K, Lim J, Dale JM, Chen H, Shinn P, Palm CJ, Southwick AM, Wu HC, Kim C, Nguyen M, et al. Empirical analysis of transcriptional activity in the Arabidopsis genome. Science. 2003;302:842–846. [PubMed]
45. Billy E, Brondani V, Zhang H, Muller U, Filipowicz W. Specific interference with gene expression induced by long, double-stranded RNA in mouse embryonal teratocarcinoma cell lines. Proc. Natl Acad. Sci. USA. 2001;98:14428–14433. [PMC free article] [PubMed]
46. Shinagawa T, Ishii S. Generation of Ski-knockdown mice by expressing a long double-strand RNA from an RNA polymerase II promoter. Genes Dev. 2003;17:1340–1345. [PMC free article] [PubMed]
47. Okada Y, Tashiro C, Numata K, Watanabe K, Nakaoka H, Yamamoto N, Okubo K, Ikeda R, Saito R, Kanai A, et al. Comparative expression analysis uncovers novel features of endogenous antisense transcription. Hum. Mol. Genet. 2008;17:1631–1640. [PubMed]
48. Borsani O, Zhu J, Verslues PE, Sunkar R, Zhu JK. Endogenous siRNAs derived from a pair of natural cis-antisense transcripts regulate salt tolerance in Arabidopsis. Cell. 2005;123:1279–1291. [PMC free article] [PubMed]
49. Styhler S, Nakamura A, Swan A, Suter B, Lasko P. vasa is required for GURKEN accumulation in the oocyte, and is involved in oocyte differentiation and germline cyst development. Development. 1998;125:1569–1578. [PubMed]
50. Vagin VV, Klenov MS, Kalmykova AI, Stolyarenko AD, Kotelnikov RN, Gvozdev VA. The RNA interference proteins and vasa locus are involved in the silencing of retrotransposons in the female germline of Drosophila melanogaster. RNA Biol. 2004;1:54–58. [PubMed]
51. Kogan GL, Tulin AV, Aravin AA, Abramov YA, Kalmykova AI, Maisonhaute C, Gvozdev VA. The GATE retrotransposon in Drosophila melanogaster: mobility in heterochromatin and aspects of its expression in germline tissues. Mol. Genet. Genomics. 2003;269:234–242. [PubMed]
52. Kalmykova AI, Kwon DA, Rozovsky YM, Hueber N, Capy P, Maisonhaute C, Gvozdev VA. Selective expansion of the newly evolved genomic variants of retrotransposon 1731 in the Drosophila genomes. Mol. Biol. Evol. 2004;21:2281–2289. [PubMed]
53. Contursi C, Minchiotti G, Di Nocera PP. Functional dissection of two promoters that control sense and antisense transcription of Drosophila melanogaster F elements. J. Mol. Biol. 1993;234:988–997. [PubMed]
54. Yang Z, Boffelli D, Boonmark N, Schwartz K, Lawn R. Apolipoprotein(a) gene enhancer resides within a LINE element. J. Biol. Chem. 1998;273:891–897. [PubMed]
55. Aravin AA, Klenov MS, Vagin VV, Bantignies F, Cavalli G, Gvozdev VA. Dissection of a natural RNA silencing process in the Drosophila melanogaster germ line. Mol. Cell Biol. 2004;24:6742–6750. [PMC free article] [PubMed]
56. Harris AN, Macdonald PM. Aubergine encodes a Drosophila polar granule component required for pole cell formation and related to eIF2C. Development. 2001;128:2823–2832. [PubMed]
57. Megosh HB, Cox DN, Campbell C, Lin H. The role of PIWI and the miRNA machinery in Drosophila germline determination. Curr. Biol. 2006;16:1884–1894. [PubMed]
58. Miller K, Rosenbaum J, Zbrzezna V, Pogo AO. The nucleotide sequence of Drosophila melanogaster copia-specific 2.1-kb mRNA. Nucleic Acids Res. 1989;17:2134. [PMC free article] [PubMed]
59. Avedisov SN, Ilyin YV. Identification of spliced RNA species of Drosophila melanogaster gypsy retrotransposon. New evidence for retroviral nature of the gypsy element. FEBS Lett. 1994;350:147–150. [PubMed]
60. Belancio VP, Hedges DJ, Deininger P. LINE-1 RNA splicing and influences on mammalian gene expression. Nucleic Acids Res. 2006;34:1512–1521. [PMC free article] [PubMed]
61. Losada A, Agudo M, Abad JP, Villasante A. HeT-A telomere-specific retrotransposons in the centric heterochromatin of Drosophila melanogaster chromosome 3. Mol. Gen. Genet. 1999;262:618–622. [PubMed]
62. Kiyosawa H, Mise N, Iwase S, Hayashizaki Y, Abe K. Disclosing hidden transcripts: mouse natural sense-antisense transcripts tend to be poly(A) negative and nuclear localized. Genome Res. 2005;15:463–474. [PMC free article] [PubMed]
63. Chang YF, Imam JS, Wilkinson MF. The nonsense-mediated decay RNA surveillance pathway. Annu. Rev. Biochem. 2007;76:51–74. [PubMed]
64. Matzke MA, Birchler JA. RNAi-mediated pathways in the nucleus. Nat. Rev. Genet. 2005;6:24–35. [PubMed]
65. Robb GB, Brown KM, Khurana J, Rana TM. Specific and potent RNAi in the nucleus of human cells. Nat. Struct. Mol. Biol. 2005;12:133–137. [PubMed]
66. Klenov MS, Lavrov SA, Stolyarenko AD, Ryazansky SS, Aravin AA, Tuschl T, Gvozdev VA. Repeat-associated siRNAs cause chromatin silencing of retrotransposons in the Drosophila melanogaster germline. Nucleic Acids Res. 2007;35:5430–5438. [PMC free article] [PubMed]
67. Findley SD, Tamanaha M, Clegg NJ, Ruohola-Baker H. Maelstrom, a Drosophila spindle-class gene, encodes a protein that colocalizes with Vasa and RDE1/AGO1 homolog, Aubergine, in nuage. Development. 2003;130:859–871. [PubMed]
68. Snee MJ, Macdonald PM. Live imaging of nuage and polar granules: evidence against a precursor-product relationship and a novel role for Oskar in stabilization of polar granule components. J. Cell Sci. 2004;117:2109–2120. [PubMed]
69. Kennerdell JR, Yamaguchi S, Carthew RW. RNAi is activated during Drosophila oocyte maturation in a manner dependent on aubergine and spindle-E. Genes Dev. 2002;16:1884–1889. [PMC free article] [PubMed]
70. Lim AK, Kai T. Unique germ-line organelle, nuage, functions to repress selfish genetic elements in Drosophila melanogaster. Proc. Natl Acad. Sci. USA. 2007;104:6714–6719. [PMC free article] [PubMed]
71. Cox DN, Chao A, Lin H. piwi encodes a nucleoplasmic factor whose activity modulates the number and division rate of germline stem cells. Development. 2000;127:503–514. [PubMed]

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...