Large-scale transcriptome data reveals transcriptional activity of fission yeast LTR retrotransposons
© Mourier and Willerslev; licensee BioMed Central Ltd. 2010
Received: 20 October 2009
Accepted: 12 March 2010
Published: 12 March 2010
Retrotransposons are transposable elements that proliferate within eukaryotic genomes through a process involving reverse transcription. The numbers of retrotransposons within genomes and differences between closely related species may yield insight into the evolutionary history of the elements. Less is known about the ongoing dynamics of retrotransposons, as analysis of genome sequences will only reveal insertions of retrotransposons that are fixed - or near fixation - in the population or strain from which genetic material has been extracted for sequencing. One pre-requisite for retrotransposition is transcription of the elements. Given their intrinsic sequence redundancy, transcriptome-level analyses of transposable elements are scarce. We have used recently published transcriptome data from the fission yeast Schizosaccharomyces pombe to assess the ability to detect and describe transcriptional activity from Long Terminal Repeat (LTR) retrotransposons. LTR retrotransposons are normally flanked by two LTR sequences. However, the majority of LTR sequences in S. pombe exist as solitary LTRs, i.e. as single terminal repeat sequences not flanking a retrotransposon. Transcriptional activity was analysed for both full-length LTR retrotransposons and solitary LTRs.
Two independent sets of transcriptome data reveal the presence of full-length, polyadenylated transcripts from LTR retrotransposons in S. pombe during growth phase in rich medium. The redundancy of retrotransposon sequences makes it difficult to assess which elements are transcriptionally active, but data strongly indicates that only a subset of the LTR retrotransposons contribute significantly to the detected transcription. A considerable level of reverse strand transcription is also detected. Equal levels of transcriptional activity are observed from both strands of solitary LTR sequences. Transcriptome data collected during meiosis suggests that transcription of solitary LTRs is correlated with the transcription of nearby protein-coding genes.
Presumably, the host organism negatively regulates proliferation of LTR retrotransposons. The finding of considerable transcriptional activity of retrotransposons suggests that part of this regulation is likely to take place at a post-transcriptional level. Alternatively, the transcriptional activity may signify a hitherto unrecognized activity level of retrotransposon proliferation. Our findings underline the usefulness of transcriptome data in elucidating dynamics in retrotransposon transcription.
With only a few exceptions [1, 2], retrotransposons have been found in all analysed eukaryotic genomes. Although transcription of retrotransposons is an integral part of their life cycle, elements may be transcriptionally active without this resulting in proliferation of the elements within the host genome [3, 4]. Transcriptional activity of retrotransposons has been detected in a range of organisms and conditions, and may involve a multitude of elements that are simultaneously transcribed [5–7], or alternatively, single element loci driving transcription of nearby genes [8, 9]. The presence and transcriptional activity of retrotransposons may interfere with nearby genes , and hence presumably are subject to negative selection [11–13].
Unfortunately, the intrinsic sequence redundancy of retrotransposons has limited the resolution by which activity can be assigned to specific elements (or classes thereof) using genome-scale approaches [14, 15]. The recent advances in novel sequencing and hybridization technologies [16, 17] have permitted an unforeseen depth in detection of transcriptional activity. Recently, Faulkner and colleagues reported that 6-30% of cap-selected mammalian transcripts were initiated in repetitive elements . We set out to test if transcriptome data could provide information on the transcriptional activity of presumably functional (i.e. retrotransposition-competent) retrotransposons, and turned our attention to the single-celled fission yeast Schizosaccharomyces pombe. The genome of S. pombe is highly compact and well annotated , and harbours only a few families of Long Terminal Repeat (LTR) retrotransposons [19, 20]. LTR retrotransposons are transposable elements that typically contain gag and pol genes required for transposition, are related to retroviruses, and have their name from the two repeated LTR sequences flanking them. Two LTR sequences may recombine resulting in a solitary LTR sequence. All full-length LTR retrotransposons in the S. pombe reference genome (strain 972) belong to the Tf2 family, while all members of the other dominant LTR family, Tf1, are found as solitary LTR sequences [18, 21]. S. pombe LTR elements are predominantly inserted upstream of protein-coding genes [22, 23], where transcription activators are responsible for targeting the site of insertion . Intriguingly, the Tf1 elements were shown to harbour promoter regions restoring the regulatory functions that are disrupted by LTR integration .
We have analysed the data from two recent studies: First, a high throughput sequencing of complementary DNAs generating short reads (30-51 nucleotides) from S. pombe growth phase and five time points during meiosis from the Bähler lab . This study is henceforth referred to as RNA-Seq. Second, a study from the Cairns lab  using a novel approach (called HybMap) in which RNA from growth phase was directly hybridized to a whole-genome microarray with 60 base pair DNA probes, followed by antibody procedures ensuring perfect matches as well as subsequent quantification by light emission . The direct hybridization approach hence allows the assignment of transcriptional activity to a specific strand. We refer to this study simply as HybMap. The available data samples are summarized in Additional file 1; Table S1. From RNA-Seq sequence reads and signal intensities of HybMap array probes we recorded the transcriptional activity from LTR sequences, analysed the extent and orientation of LTR transcription, and how the expression profiles of LTR sequences correlate with nearby genes.
Results and discussion
We retrieved 239 LTR sequences with lengths from 75 to 412 base pairs (bp) (median/average size: 346/321 bp) from the S. pombe genome annotation. Of these, 25 LTR sequences were residing in full-length LTR retrotransposons. This uneven number results from a single chimerical LTR retrotransposon with an [LTR-internal sequence-LTR-internal sequence-LTR] organization that potentially is a result of an ectopic recombination event. Two full-length LTR retrotransposons (SPBC1E8.04 and SPCC1494.11c) are frame shifted and annotated as pseudogenes http://www.genedb.org/. Genomic coordinates of the LTR sequences are provided in Additional file 1; Table S2. From a biological point of view, we are interested in distinguishing between transcriptional activity stemming from full-length LTR retrotransposons and from solitary LTRs. Analysis of transcriptional activity was therefore performed on these two sets of LTRs separately: 13 full-length LTR retrotransposons (each consisting of the internal sequence flanked by LTR sequences) and the remaining 214 solitary LTR sequences.
Data sets of LTR sequence sets and other genomic features
Total size (bp)
Full-length LTR elements
Highly expressed protein-coding genes
Lowly expressed protein-coding genes
Repair genes (rec and rad)
Full-length LTR retrotransposons
The fact that two almost identical flanking LTRs are present in each full-length retrotransposon, and the likelihood that an LTR mapping probe will also map to LTR sequences outside the set of full-length elements (and hence be excluded by our procedure), makes it difficult to compare directly the flanking LTRs and the internal sequence. We would expect transcription to initiate at the transcription start site in the 5' LTR (roughly halfway into the LTR sequence ) and terminate in the 3' LTR. In reality, as LTR-matching probes are often mapping to both flanking LTR sequences, we cannot establish from which end the activity stems, and for consistency we have simply mapped such probes to the 3' LTR in Figure 3. Interestingly, no clear distinction between forward and reverse strand transcriptional activity is observed for the LTR sequences.
Solitary LTR sequences
Transcriptional activity is also detectable from solitary LTR sequences, although only a subset of the 1298 probes mapping uniquely to solitary LTRs shows high levels of signal intensities (Figure 2B, blue curve). Further, transcriptional activity is similar from the forward and the reverse strand. To establish if the divergent intensities of signals from solitary LTR probes are a result of differences between LTR loci, or between probes within LTR sequences, we calculated the average signal intensity for each LTR locus. If the observed broad range of signal intensities in Figure 2B was a result of LTR loci with equal levels of transcriptional activity, but with certain parts of the LTR sequences being transcribed and some not, we would expect the average signal intensity for individual LTR sequences to be relatively similar. This is not the case (Figure 2B, grey curve), suggesting that individual solitary LTR loci differ in levels of transcriptional activity. Then why are some LTR loci apparently transcribed whereas others are not? We do not find any noticeable patterns between highly and lowly transcribed LTR sequences in terms of orientation and distance with respect to nearest neighbouring protein-coding gene (not shown). In fact, no correlation is found between the average signal intensity of LTR sequences and the average of the nearest gene (Additional file 2; Figure S2). Finally, to test if the observed transcriptional differences between specific LTR sequences are due to simple stochastic variation due to the relatively small sample size, we performed a permutation analysis in which probe intensities were randomly assigned to an LTR sequence (while keeping the distribution of uniquely mapping probes to LTRs constant, see Methods). When repeating this analysis 10.000 times, we find that the observed variance between the average signal intensities from each LTR sequence (forward variance: 2.08, reverse variance 2.03) fall well without the range of variances of the simulated sets (forward range 1.08-1.67, reverse range 1.11-1.65) (Additional file 2; Figure S3). This strongly suggests that distinct differences in transcriptional activity levels exist between S. pombe solitary LTR sequence loci.
HybMap analysis from alternative procedures and growth conditions
Data from the Cairns group HybMap study include hybridizations using poly(A)-enriched RNA samples, as well as samples from alternative growth conditions (minimal medium, heat shock and DNA damage) (Additional file 1; Table S1) . In general, alternative growth conditions yield similar patterns of LTR transcriptional activity (Additional file 2; Figure S5). Notably, a relative higher forward strand activity from full-length LTR retrotransposons is found in poly(A)-enriched samples, consistent with polyadenylation of the full-length LTR retrotransposons (Additional file 2; Figure S5). In contrast, signal intensities from solitary LTR sequences are shifted markedly towards lower values in poly(A)-enriched samples, suggesting that no polyadenylation takes place (Additional file 2; Figure S5).
RNA-Seq analysis of LTR transcriptional activity
Sequence reads from the RNA-Seq study were mapped to the LTR sequences and other genomic features in a manner similar to the HybMap probes. Contrary to the HybMap approach, RNA-Seq involves conversion of sampled RNA into cDNA (in this case using a poly(dT) primer ). Presumably this step is responsible for the observed bias of higher RNA-Seq read densities towards the 3' end of the protein-coding genes in our reference set - a bias that is not observed for the HybMap intensities (Additional file 2; Figure S6). Consistent with this, when we plot the RNA-Seq reads mapping exclusively to full-length LTR retrotransposons onto the alignment, a pronounced 3' peak is observed (Figure 3C). Again, this strongly suggests that S. pombe LTR retrotransposons are actively transcribed in full length and polyadenylated.
For the full-length LTR elements, we assigned the RNA-Seq reads using two complementary procedures. First, reads mapping to multiple loci were assigned evenly between LTR loci (e.g. two LTRs sharing a read are assigned 0.5 read each). Second, multiple mapping reads were assigned to a single locus and the highest possible density of reads for a single locus was recorded. Similarly to the HybMap probe assignment procedure, these two alternatives reflect the two extremes of possible LTR transcription; all LTRs being transcribed at a similar level, or a minimum of LTR loci being responsible for all the detected transcription.
In summary, we find that the RNA-Seq data supports the finding of transcriptional activity from full-length LTR retrotransposons, and the notion that these transcripts are polyadenylated.
LTR transcription during meiosis
Our analysis of two independent transcriptome data sets clearly indicates that LTR retrotransposons are actively transcribed during growth phase, and that the retrotransposon transcripts are polyadenylated. When considering the evolutionary dynamics of S. pombe LTR elements, Kelly and Levin  speculated if only a few of the complete elements were responsible for majority of transposition events (by necessity preceded by transcriptional activity) as has been observed in Saccharomyces cerevisiae. Although we cannot conclude anything on the transcriptional levels of individual loci, our data strongly suggest that a minority of loci contribute the majority of the transcriptional output from LTR retrotransposons.
Assuming that high levels of retrotransposition are detrimental to the host, selection will favour regulation of retrotransposition. Such regulation may either take place pre-transcriptionally (e.g. methylation or chromatin condensation) or post-transcriptionally (e.g. degradation of transcripts). The observed transcription of S. pombe LTR retrotransposons suggests that post-transcriptional mechanisms are the predominant type of regulation during regular growth phase. The relatively high levels of transcriptional activity antisense to full-length LTR retrotransposons (Figures 2 and 3) may potentially be part of such post-transcriptional regulation pathways. The targeting of LTR insertion upstream of genes [23, 24] (which can be viewed as a means to minimize the deleterious effects of retrotransposition in a compact genome ) could limit the potential of regulating retrotransposition pre-transcriptionally in S. pombe, as any interference with transcriptional regulation presumably would also affect neighbouring protein-coding genes.
From the HybMap data transcriptional activity is detectable from both strands of solitary LTR sequences, and the transcription appears to be confined to the LTR sequences themselves. Analysis of the RNA-Seq data suggests that transcription of solitary LTRs upstream of genes is correlated with the transcriptional activity of the neighbouring genes. One might speculate that the presence of transcripts from both strands of the solitary LTRs could generate double-stranded RNAs, similar to the double-stranded RNAs involved in heterochromatin formation in S. pombe[33, 34]. However, this would suggest a negative correlation between the transcriptional activity of genes and LTR sequences. The observed positive correlation rather suggests that the LTR transcripts represent transcription facilitated by a physical association with actively transcribed genes, in parallel to the observed co-expression of linked genes .
Our analysis indicates that a clear distinction exists between solitary LTR sequences and full-length LTR elements in terms of transcriptional activity. Transcription of full-length LTR retrotransposons is for the most part derived from the forward strand and transcripts are polyadenylated. In contrast, solitary LTR transcription is found from both strands, with transcripts showing no signs of polyadenylation. Additionally, divergent expression levels during meiosis are observed between full-length LTR retrotransposons and solitary LTR sequences. In HIV retroviruses, a number of transcription factors bind the internal retrotransposon downstream of the 5' LTR . It is possible that the absence of such binding sites around solitary LTRs contributes to their changed transcription patterns. Similarly, in a range of retrovirus LTRs, polyadenylation signals are present downstream of the transcription start sites, but a minimum distance is required for polyadenylation to take place , so that transcription is initiated in the 5' LTR and polyadenylation only in the 3' LTR of full-length retroviruses. Certainly, this would preclude polyadenylation of transcripts initiated within solitary LTR sequences. Similar to the solitary LTR sequences, levels of transcriptional activity appear to be relatively similar from both strands of the LTRs from full-length retrotransposons. One possible scenario is therefore that - compared to full-length retrotransposons - the solitary LTR sequences have changed their transcriptional patterns due to loss of regulatory motifs, and lost the ability to generate polyadenylated transcripts, but have retained the ability to generate transcription from both strands (at equal, albeit relatively low levels).
Based on our analysis, we conclude that the application of large-scale transcriptome data allows the elucidation of retrotransposon transcriptional activity, but that the resolution by which transcription can be assigned to specific retrotransposon loci is still limited. A recent RNA-Seq approach revealed an up-regulation of transposons in methylation-defective Arabidopsis mutants . Additionally, mapping of capped sequence reads demonstrate wide-spread, regulated transcription initiated with mammalian retrotransposons . Hence, novel transcriptome analysis techniques will inevitably shed light on tissue-specific (if applicable) and temporal expression patterns of retrotransposons facilitating an assessment of the dynamics and immediate impact of these long-term residents of eukaryotic genomes.
Solitary LTR and full-length LTR retrotransposon alignments
LTR sequence coordinates were extracted from the S. pombe genome annotation files (version 16-08-2008) downloaded from the Sanger http://www.sanger.ac.uk ftp site. Full-length LTR retrotransposons were retrieved and aligned using MUSCLE . To construct the set of relatively similar solitary LTRs, all LTR sequences not being part of full-length LTR retrotransposons were aligned, and all pair-wise identity scores were recorded. LTRs were then clustered if their level of identity exceeded a certain threshold, and collapsed with other clusters if any member of one cluster had high enough similarity to any member of another cluster. By observing the changes in cluster sizes for different similarity thresholds, 70% identity was chosen as cut-off value. The members of the largest cluster were then re-aligned separately and subsequently trimmed manually removing low-similarity flanking sequences. The alignment of solitary LTR sequences is provided as Additional file 2; Figure S10 and the LTR sequences are marked as 'Context solitary LTRs' in Additional file 1; Table S2.
Retrieval and mapping of sequence reads and probes
For RNA-Seq data, fastq files were downloaded from ArrayExpress http://www.ebi.ac.uk/microarray-as/ae/, accession number E-MTAB-5. Reads with ambiguous calls (Ns) were omitted. Reads were then mapped onto the LTRs sets (solitary and full-length) as well as the other selected genomic features using the Tagger software . Only perfect matches were considered. Reads mapping to any set of genomic features were then mapped against the remaining genome, and reads and probes not mapping exclusively (for solitary LTRs and full-length LTR retrotransposons) or uniquely (all other genomic features) within a sequence set were excluded from the analysis.
HybMap data were downloaded from the Gene Expression Omnibus (GEO) at NCBI http://www.ncbi.nlm.nih.gov/, accession number GSE11619. Probes were mapped and filtered similarly to RNA-Seq sequence reads (although only probes mapping uniquely to solitary LTRs were considered), and their signal intensities normalised by a 'baseline' of intergenic values  were extracted. The total number of sequence reads and probes mapping to LTRs are shown in Additional file 1; Table S1. Mapping probes to LTR alignments were done by collecting the probes mapping exclusively to LTR sequences included in the alignment. The first instance of a mapping to an LTR sequence was selected, and the midpoint of the mapping position on the sequence was transferred to the corresponding column position in the alignment.
Genomic coordinates for histone, ribosomal, repair and tRNA genes, as well as introns were retrieved from the genome annotation. A set of H/ACA box snoRNA sequences were collected from supplementary Table S2 in reference . Genomic coordinates for snoRNAs were then established from genomic BLAST searches using stand-alone WUBLAST (now AB-BLAST http://www.advbiocomp.com/). SnoRNAs, histone, ribosomal and repair genes are listed as Additional file 1; Tables S3-S6. Intergenic sequences are defined as genomic sequence residing between protein-coding genes, tRNAs, snoRNAs and LTR sequences. To avoid putative untranslated regions (UTRs), 240 base pairs flanking both sides of protein-coding genes were omitted from the intergenic sequences (the size of 240 being adopted directly from the Cairns lab HybMap study ). Only intronic and intergenic sequences of at least 100 bp in size were included in the reference sets.
Variance and correlation analyses
To assess if transcriptional activity from solitary LTR sequences were randomly distributed between the sequences, the 649 forward and 649 reverse probes mapping uniquely to 177 LTR solitary sequences were collected (forward and reverse probes analysed separately). For each LTR sequence, the average signal intensity was calculated, and the observed variance between LTR signals was recorded (the real variance). The probes were then shuffled between LTR sequences, so that each LTR sequence was assigned the same number of probes as in the real data. For the simulated set, the average signal intensity for each LTR, and the variance between LTRs was similarly calculated (the simulated variance). The simulation procedure was repeated 10.000 times for both forward and reverse probes.
Correlation analysis of the transcriptional activity between solitary LTRs and their neighbouring genes was performed as follows: LTR sequences with high levels of uniquely mapped RNA-Seq reads were collected by filtering out LTRs with a minimum of 30 uniquely mapped reads from all stages combined, and at least 10 uniquely mapped reads from growth phase. These rather arbitrarily set thresholds resulted in the eight pairs of LTRs and protein-coding genes depicted in Figure 6. The permutated sets were constructed by randomly assembling eight sets of LTR time series by shuffling the real data while keeping time points constant. For example, all M1 time points were shuffled independently, as were time points M2, M3, M4, M5. The eight sets were then compared to the real data from protein-coding genes, eight product moment correlation coefficients were calculated, and the median of coefficients was recorded. This procedure was then repeated 10.000 times.
This work was supported by a personal grant to TM from the Lundbeck Foundation. Work at the Ancient DNA and Evolution group is funded by the Villum Kann Rasmussen Foundation; the Danish Natural Science Research Council; and the Danish National Research Foundation. We thank Daniel Jeffares for critical reading of the manuscript, Olaf Nielsen for advice on S. pombe biology, and two anonymous referees for constructive comments on an earlier version of this study.
- Arkhipova I, Meselson M: Transposable elements in sexual and ancient asexual taxa. Proc Natl Acad Sci USA. 2000, 97 (26): 14473-14477. 10.1073/pnas.97.26.14473.PubMed CentralPubMedView ArticleGoogle Scholar
- Gardner MJ, Hall N, Fung E, White O, Berriman M, Hyman RW, Carlton JM, Pain A, Nelson KE, Bowman S: Genome sequence of the human malaria parasite Plasmodium falciparum. Nature. 2002, 419 (6906): 498-511. 10.1038/nature01097.PubMedView ArticleGoogle Scholar
- Speek M: Antisense promoter of human L1 retrotransposon drives transcription of adjacent cellular genes. Molecular and cellular biology. 2001, 21 (6): 1973-1985. 10.1128/MCB.21.6.1973-1985.2001.PubMed CentralPubMedView ArticleGoogle Scholar
- Faulkner GJ, Kimura Y, Daub CO, Wani S, Plessy C, Irvine KM, Schroder K, Cloonan N, Steptoe AL, Lassmann T: The regulated retrotransposon transcriptome of mammalian cells. Nat Genet. 2009, 41 (5): 563-571. 10.1038/ng.368.PubMedView ArticleGoogle Scholar
- Evsikov AV, de Vries WN, Peaston AE, Radford EE, Fancher KS, Chen FH, Blake JA, Bult CJ, Latham KE, Solter D: Systems biology of the 2-cell mouse embryo. Cytogenet Genome Res. 2004, 105 (2-4): 240-250. 10.1159/000078195.PubMedView ArticleGoogle Scholar
- Peaston AE, Evsikov AV, Graber JH, de Vries WN, Holbrook AE, Solter D, Knowles BB: Retrotransposons regulate host genes in mouse oocytes and preimplantation embryos. Dev Cell. 2004, 7 (4): 597-606. 10.1016/j.devcel.2004.09.004.PubMedView ArticleGoogle Scholar
- Grandbastien MA, Audeon C, Bonnivard E, Casacuberta JM, Chalhoub B, Costa AP, Le QH, Melayah D, Petit M, Poncet C: Stress activation and genomic impact of Tnt1 retrotransposons in Solanaceae. Cytogenet Genome Res. 2005, 110 (1-4): 229-241. 10.1159/000084957.PubMedView ArticleGoogle Scholar
- Whitelaw E, Martin DI: Retrotransposons as epigenetic mediators of phenotypic variation in mammals. Nat Genet. 2001, 27 (4): 361-365. 10.1038/86850.PubMedView ArticleGoogle Scholar
- Dunn CA, Medstrand P, Mager DL: An endogenous retroviral long terminal repeat is the dominant promoter for human beta1,3-galactosyltransferase 5 in the colon. Proc Natl Acad Sci USA. 2003, 100 (22): 12841-12846. 10.1073/pnas.2134464100.PubMed CentralPubMedView ArticleGoogle Scholar
- Goodier JL, Kazazian HH: Retrotransposons revisited: the restraint and rehabilitation of parasites. Cell. 2008, 135 (1): 23-35. 10.1016/j.cell.2008.09.022.PubMedView ArticleGoogle Scholar
- Medstrand P, Lagemaat van de LN, Mager DL: Retroelement distributions in the human genome: variations associated with age and proximity to genes. Genome Res. 2002, 12 (10): 1483-1495. 10.1101/gr.388902.PubMed CentralPubMedView ArticleGoogle Scholar
- Lagemaat van de LN, Medstrand P, Mager DL: Multiple effects govern endogenous retrovirus survival patterns in human gene introns. Genome Biol. 2006, 7 (9): R86-10.1186/gb-2006-7-9-r86.PubMed CentralPubMedView ArticleGoogle Scholar
- Mourier T, Willerslev E: Does Selection against Transcriptional Interference Shape Retroelement-Free Regions in Mammalian Genomes?. PLoS ONE. 2008, 3 (11): e3760-10.1371/journal.pone.0003760.PubMed CentralPubMedView ArticleGoogle Scholar
- Mockler TC, Chan S, Sundaresan A, Chen H, Jacobsen SE, Ecker JR: Applications of DNA tiling arrays for whole-genome analysis. Genomics. 2005, 85 (1): 1-15. 10.1016/j.ygeno.2004.10.005.PubMedView ArticleGoogle Scholar
- Wang SM: Understanding SAGE data. Trends Genet. 2007, 23 (1): 42-50. 10.1016/j.tig.2006.11.001.PubMedView ArticleGoogle Scholar
- Hu Z, Zhang A, Storz G, Gottesman S, Leppla SH: An antibody-based microarray assay for small RNA detection. Nucl Acids Res. 2006, 34 (7): e52-10.1093/nar/gkl142.PubMed CentralPubMedView ArticleGoogle Scholar
- Schuster SC: Next-generation sequencing transforms today's biology. Nature methods. 2008, 5 (1): 16-18. 10.1038/nmeth1156.PubMedView ArticleGoogle Scholar
- Wood V, Gwilliam R, Rajandream MA, Lyne M, Lyne R, Stewart A, Sgouros J, Peat N, Hayles J, Baker S: The genome sequence of Schizosaccharomyces pombe. Nature. 2002, 415 (6874): 871-880. 10.1038/nature724.PubMedView ArticleGoogle Scholar
- Levin HL, Weaver DC, Boeke JD: Two related families of retrotransposons from Schizosaccharomyces pombe. Molecular and cellular biology. 1990, 10 (12): 6791-6798.PubMed CentralPubMedView ArticleGoogle Scholar
- Weaver DC, Shpakovski GV, Caputo E, Levin HL, Boeke JD: Sequence analysis of closely related retrotransposon families from fission yeast. Gene. 1993, 131 (1): 135-139. 10.1016/0378-1119(93)90682-S.PubMedView ArticleGoogle Scholar
- Kelly FD, Levin HL: The evolution of transposons in Schizosaccharomyces pombe. Cytogenet Genome Res. 2005, 110 (1-4): 566-574. 10.1159/000084990.PubMedView ArticleGoogle Scholar
- Behrens R, Hayles J, Nurse P: Fission yeast retrotransposon Tf1 integration is targeted to 5' ends of open reading frames. Nucleic Acids Res. 2000, 28 (23): 4709-4716. 10.1093/nar/28.23.4709.PubMed CentralPubMedView ArticleGoogle Scholar
- Bowen NJ, Jordan IK, Epstein JA, Wood V, Levin HL: Retrotransposons and their recognition of pol II promoters: a comprehensive survey of the transposable elements from the complete genome sequence of Schizosaccharomyces pombe. Genome Res. 2003, 13 (9): 1984-1997. 10.1101/gr.1191603.PubMed CentralPubMedView ArticleGoogle Scholar
- Leem YE, Ripmaster TL, Kelly FD, Ebina H, Heincelman ME, Zhang K, Grewal SI, Hoffman CS, Levin HL: Retrotransposon Tf1 is targeted to Pol II promoters by transcription activators. Molecular cell. 2008, 30 (1): 98-107. 10.1016/j.molcel.2008.02.016.PubMed CentralPubMedView ArticleGoogle Scholar
- Wilhelm BT, Marguerat S, Watt S, Schubert F, Wood V, Goodhead I, Penkett CJ, Rogers J, Bahler J: Dynamic repertoire of a eukaryotic transcriptome surveyed at single-nucleotide resolution. Nature. 2008, 453 (7199): 1239-1243. 10.1038/nature07002.PubMedView ArticleGoogle Scholar
- Dutrow N, Nix DA, Holt D, Milash B, Dalley B, Westbroek E, Parnell TJ, Cairns BR: Dynamic transcriptome of Schizosaccharomyces pombe shown by RNA-DNA hybrid mapping. Nat Genet. 2008, 40 (8): 977-986. 10.1038/ng.196.PubMed CentralPubMedView ArticleGoogle Scholar
- Cam HP, Noma K, Ebina H, Levin HL, Grewal SI: Host genome surveillance for retrotransposons by transposon-derived proteins. Nature. 2008, 451 (7177): 431-436. 10.1038/nature06499.PubMedView ArticleGoogle Scholar
- Greenall A, Williams ES, Martin KA, Palmer JM, Gray J, Liu C, Whitehall SK: Hip3 Interacts with the HIRA Proteins Hip1 and Slm9 and Is Required for Transcriptional Silencing and Accurate Chromosome Segregation. J Biol Chem. 2006, 281 (13): 8732-8739. 10.1074/jbc.M512170200.PubMedView ArticleGoogle Scholar
- Durand-Dubief M, Sinha I, Fagerstrom-Billai F, Bonilla C, Wright A, Grunstein M, Ekwall K: Specific functions for the fission yeast Sirtuins Hst2 and Hst4 in gene regulation and retrotransposon silencing. Embo J. 2007, 26 (10): 2477-2488. 10.1038/sj.emboj.7601690.PubMed CentralPubMedView ArticleGoogle Scholar
- Hansen KR, Burns G, Mata J, Volpe TA, Martienssen RA, Bahler J, Thon G: Global Effects on Gene Expression in Fission Yeast by Silencing and RNA Interference Machineries. Mol Cell Biol. 2005, 25 (2): 590-601. 10.1128/MCB.25.2.590-601.2005.PubMed CentralPubMedView ArticleGoogle Scholar
- Morillon A, Benard L, Springer M, Lesage P: Differential Effects of Chromatin and Gcn4 on the 50-Fold Range of Expression among Individual Yeast Ty1 Retrotransposons. Mol Cell Biol. 2002, 22 (7): 2078-2088. 10.1128/MCB.22.7.2078-2088.2002.PubMed CentralPubMedView ArticleGoogle Scholar
- Kim JM, Vanguri S, Boeke JD, Gabriel A, Voytas DF: Transposable Elements and Genome Organization: A Comprehensive Survey of Retrotransposons Revealed by the Complete Saccharomyces cerevisiae Genome. Genome Research. 1998, 8 (5): 464-478.PubMedGoogle Scholar
- Reinhart BJ, Bartel DP: Small RNAs correspond to centromere heterochromatic repeats. Science. 2002, 297 (5588): 1831-10.1126/science.1077183.PubMedView ArticleGoogle Scholar
- Volpe TA, Kidner C, Hall IM, Teng G, Grewal SI, Martienssen RA: Regulation of heterochromatic silencing and histone H3 lysine-9 methylation by RNAi. Science. 2002, 297 (5588): 1833-1837. 10.1126/science.1074973.PubMedView ArticleGoogle Scholar
- Batada NN, Urrutia AO, Hurst LD: Chromatin remodelling is a major source of coexpression of linked genes in yeast. Trends in Genetics. 2007, 23 (10): 480-484. 10.1016/j.tig.2007.08.003.PubMedView ArticleGoogle Scholar
- Pereira LA, Bentley K, Peeters A, Churchill MJ, Deacon NJ: A compilation of cellular transcription factor interactions with the HIV-1 LTR promoter. Nucleic Acids Res. 2000, 28 (3): 663-668. 10.1093/nar/28.3.663.PubMed CentralPubMedView ArticleGoogle Scholar
- Guntaka RV: Transcription termination and polyadenylation in retroviruses. Microbiol Rev. 1993, 57 (3): 511-521.PubMed CentralPubMedGoogle Scholar
- Lister R, O'Malley RC, Tonti-Filippini J, Gregory BD, Berry CC, Millar AH, Ecker JR: Highly integrated single-base resolution maps of the epigenome in Arabidopsis. Cell. 2008, 133 (3): 523-536. 10.1016/j.cell.2008.03.029.PubMed CentralPubMedView ArticleGoogle Scholar
- Edgar RC: MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004, 32 (5): 1792-1797. 10.1093/nar/gkh340.PubMed CentralPubMedView ArticleGoogle Scholar
- Iseli C, Ambrosini G, Bucher P, Jongeneel CV: Indexing strategies for rapid searches of short words in genome sequences. PLoS ONE. 2007, 2 (6): e579-10.1371/journal.pone.0000579.PubMed CentralPubMedView ArticleGoogle Scholar
- Li S-G, Zhou H, Luo Y-P, Zhang P, Qu L-H: Identification and Functional Analysis of 20 Box H/ACA Small Nucleolar RNAs (snoRNAs) from Schizosaccharomyces pombe. J Biol Chem. 2005, 280 (16): 16446-16455. 10.1074/jbc.M500326200.PubMedView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.