Computational discovery of sense-antisense transcription in the human and mouse genomes
© Shendure and Church, licensee BioMed Central Ltd 2002
Received: 10 May 2002
Accepted: 15 July 2002
Published: 22 August 2002
Overlapping but oppositely oriented transcripts have the potential to form sense-antisense perfect double-stranded (ds) RNA duplexes. Over recent years, the number and variety of examples of mammalian gene-regulatory phenomena in which endogenous dsRNA duplexes have been proposed or demonstrated to participate has greatly increased. These include genomic imprinting, RNA interference, translational regulation, alternative splicing, X-inactivation and RNA editing. We computationally mined public mouse and human expressed sequence tag (EST) databases to search for additional examples of bidirectionally transcribed genomic regions.
Our bioinformatics approach identified over 217 candidate overlapping transcriptional units, almost all of which are novel. From experimental validation of a subset of our predictions by orientation-specific RT-PCR, we estimate that our methodology has a specificity of 84% or greater. In many cases, regions of sense-antisense overlap within the 5'- or 3'-untranslated regions of a given transcript correlate with genomic patterns of mouse-human conservation.
Our results, in conjunction with the literature, bring the total number of predicted and validated examples of overlapping but oppositely oriented transcripts to over 300. Several of these cases support the hypothesis that a subset of the instances of substantial mouse-human conservation in the 5' and 3' UTRs of transcripts might be explained in part by functionality of an overlapping transcriptional unit.
Characterized examples of endogenous antisense RNA in metazoans can be broadly divided into two categories (see [1,2] for extensive review). Antisense RNAs transcribed from loci distinct from their putative targets, such as lin-4 of Caenorhabditis elegans, are generally short and have the potential to form imperfect duplexes with complementary regions of their sense counterparts . In contrast, antisense transcripts that originate from the same genomic region (but with opposing orientation) have, by virtue of their common but complementary origin, the potential to form long perfect duplexes.
Experimental evidence suggests a functional role for sense-antisense pairings at a surprising variety of levels in mammalian gene regulation, including genomic imprinting [4,5], RNA interference , translational regulation , alternative splicing , X-inactivation , and RNA editing . Where the mode of regulation has been explored in detail each case has proved unique, so that it is difficult to make generalizations about mechanism or function. In transfection assays, full-length constructs of three splice variants of an endogenous coding transcript containing regions antisense to the FGF2 (fibroblast growth factor-2) mRNA can each suppress protein levels (but not mRNA levels) of FGF2 . A non-coding transcript antisense to a homeobox gene, MSX1, has a conserved transcription initiation site and is expressed in inverse correlation to the MSX1 protein . SCA8, an untranslated transcript implicated in spinocerebellar ataxia type 8, overlaps the 5' translation and transcription sites of KLHL1, a gene primarily expressed in the cerebellum . XIST and TSIX are conserved, overlapping, but oppositely oriented non-coding transcripts, which serve crucial functions in X-inactivation . Several imprinted loci generate multiple sense and antisense transcripts that are subject to reciprocal genomic imprinting, and a recent study demonstrated in vivo that premature termination of AIR, a non-coding imprinted antisense transcript, results in a failure to imprint several sense counterparts . Lipman  suggested that this phenomenon might be much more widespread than previously believed, and hypothesized that the existence of functionally relevant overlapping antisense transcripts might explain a subset of the many cases in which strong evolutionary conservation is observed in 5' and 3' untranslated regions of mammalian genes.
Overlapping transcriptional units in mammalian genomes previously described in the literature
Thousands of EST libraries, consisting in sum of millions of single-pass sequence reads, have been generated by investigators from around the world, using a variety of methods, and deposited into public sequence databases. UniGene  is an experimental algorithm developed at the NCBI, in which full-length mRNA and EST sequences are partitioned into a non-redundant set of gene-oriented clusters on the basis of nucleotide-level identity (using annotated mRNAs as initializing seeds), but these clusters are not further curated. The avoidance of spurious alignments by masking of transcribed repetitive elements, vector contaminants or low-complexity sequence is an important part of the UniGene build procedure. Each EST thus belongs to both an individual library from which it was sequenced (for example, a specific tissue from a specific individual) and a single UniGene cluster (along with other ESTs that are presumably derived from the same gene). We hypothesized that as a consequence of the automation of the UniGene build procedure, unannotated antisense transcripts might be found co-clustered with their sense counterparts. As nearly all annotated mRNAs in GenBank serve as UniGene cluster seeds (and therefore cannot be co-clustered with one another), such a strategy is strongly biased towards finding pairs of overlapping transcripts where one or both of the transcripts are unannotated.
Our bioinformatics and experimental strategy involved five steps. We first identified EST libraries that were directionally cloned and sequenced (that is, ESTs were cloned and sequenced in a defined orientation with respect to the mRNA transcript); then, focusing exclusively on ESTs from such libraries, we searched for UniGene clusters containing a statistically significant number of misoriented ESTs. We then mapped the mRNA and EST sequences from candidate UniGene clusters to their genomic coordinates and evaluated whether putative sense and antisense ESTs in a given UniGene cluster represented distinct RNA species, on the basis of differential exon-intron splicing structures, the locations of poly(A) signals and tails, and patterns of mouse-human sequence conservation. Finally, we experimentally validated a subset of the predictions by orientation-specific reverse transcription PCR (RT-PCR).
As expected, a small fraction of ESTs (around 1.5% on average) from these libraries were incorrectly oriented. Our null hypothesis was that these represented random artifacts that would be distributed across the full set of UniGene clusters in a manner proportional to the size of the individual UniGene clusters. In other words, we expected that 98.5% of the directionally cloned ESTs in each UniGene cluster would be correctly oriented. Binomial statistics were applied to identify UniGene clusters that contained a statistically significant overrepresentation of incorrectly oriented ESTs. For the substantial subset of UniGene clusters for which the dominant transcriptional orientation was unannotated, we required that the misoriented ESTs be significantly overrepresented regardless of the correct orientation of the cluster. In total, we were able to identify 549 mouse and human clusters that significantly deviated from the null hypothesis.
We next sought to address the possibility that a significant number of our candidates could represent systematic errors (for example, systematic bias for directional-cloning artifacts to occur in association with specific transcripts) or errors of the UniGene clustering algorithm. We postulated that if two distinct, overlapping but oppositely oriented RNA species were present in a single UniGene cluster, they should map to the same genomic region, but should possess significantly distinguishable exon-intron splicing structures.
Experimental evaluation of candidates by directional RT-PCR
Sense (-)antisense (-)
Sense (+)antisense (+)
Sense (-)antisense (-)
Sense (+)antisense (+)
Properties of overlapping pairs of transcripts
Number of pairs
Tail to tail (3' to 3')
Head to head (5' to 5')
Transcript starts in intron of the other transcript
Transcript contained entirely within the other transcript
Difficult to classify
The misinterpretation of genomic contaminants as putative antisense candidates is a clear concern. The observation of transcript splicing, protein homology and/or derivation from multiple independent libraries for any given set of putative antisense ESTs is evidence against genomic contamination. We have flagged (as requiring particular caution) 18 candidate cases where the set of antisense ESTs derive from one or a few libraries, and are not observed to be spliced or have protein homologies (see Additional data files).
Experimental validation of a subset of our predictions by orientation-specific RT-PCR supports our bioinformatics methodology. We estimate that our approach has a specificity of 84%, as we were able to detect antisense transcription over 33 of 39 regions queried. This may be a low-end estimate as we only queried one tissue per candidate, and cell type and/or temporal specificity of antisense transcript expression might explain our inability to confirm antisense transcription for six of the candidates experimentally. These same factors (differential temporal or cell-type distribution) may explain why the sense transcript (all of which are annotated mRNAs in the 39 cases that we attempted to experimentally verify) was not detected for seven candidates that were positive for antisense transcription.
The observation of highly conserved regions in the 5' and 3' untranslated regions (UTRs) of a large fraction of mammalian genes  has been an intriguing mystery. Lipman  hypothesized that the existence of functionally relevant overlapping antisense transcripts might explain a subset of these cases. Indeed, with a number of candidates we do observe interesting correlations between mouse-human nucleotide-level conservation patterns in UTRs and their region(s) of overlap with the putative antisense species. This includes both cases where the antisense species has homology to a known protein (Figures 2, 7) and cases where it does not (Figure 4; UniGene cluster Mm.41304; UniGene cluster Mm.183060).
Orthologous mouse and human candidates
Antisense gene or homolog
No protein match
No protein match
No protein match
As the proposed mechanisms by which the formation of long duplex dsRNA can potentially affect gene regulation are so varied [4,5,6,7,8,9,10,11,12], it is difficult to draw inferences regarding function without further experiments. One of the few areas where functionally relevant information on sequences is available relates to the neoplastic versus non-neoplastic nature of the tissue of origin of EST libraries. An interesting example is candidate UniGene cluster Hs.288835 (Figure 3), which contains CIDEB (cell-death inducing DFFA-like effector B). Noting that the sense transcript encoded a potential tumor suppressor, we checked the annotated tissue origin of these ESTs, and found that a significantly greater fraction of the antisense ESTs (34/46 = ~0.74) than the sense ESTs (3/15 = ~0.2) were derived from neoplastic tissues (p = ~0.0002 by chi-squared statistic). As the sense transcript codes for a pro-apoptotic gene, the result immediately suggests the interesting hypothesis that upregulation of the antisense RNA species in cancer tissues has functional relevance with respect to suppression of the potentially tumor-suppressing sense gene.
It is worth noting that stages of our methodology may also be useful for determining the correct transcriptional orientation of UniGene clusters that contain no annotated mRNA sequences. Many probes on orientation-sensitive oligonucleotide arrays for unknown genes are often based on such ESTs, and knowledge of the correct transcriptional orientation of each cluster may help circumvent problems such as those recently encountered in the design of an Affymetrix mouse chip .
Subset of 217 candidates for overlapping transcription previously described in the literature
Antisense gene or homolog
Our results, in conjunction with the literature, bring the total number of predicted and validated examples of overlapping but oppositely oriented transcripts to over 300. Given the variety of gene-regulatory phenomena that long-duplex dsRNA has been suggested or shown to influence [4,5,6,7,8,9,10,11,12], experimental approaches are required to query whether and how each of these overlaps is functionally relevant.
Materials and methods
Identification of high-quality directionally cloned ESTlibraries
Human UniGene (Build 146) and mouse UniGene (Build 100) datasets were downloaded from NCBI on 16 January, 2002. A useful feature of the UniGene resource is the identification of a single sequence in each cluster as its longest high-quality member. We refer to this set of representatives as the best-of-UniGene (BOU) sequences. To assess the quality of directional cloning in EST libraries, we applied the MEGABLAST tool  to align ESTs to the BOU of the UniGene cluster to which they belonged. For each EST library, we then calculated the fraction of member ESTs that were deposited in the same orientation as the BOU sequence of the UniGene cluster to which they belonged. This fraction, a metric of the quality of directional cloning of each EST library, is referred to as the library quality score (LQS).
Our original analysis was revised in two ways to improve its accuracy. Our goal of calculating library quality by estimating the 'correctness' of EST orientation is complicated by the fact that not every UniGene cluster contains an mRNA with a defined open reading frame (ORF), and not every BOU sequence is deposited in the correct orientation (in other words, the correct orientation of many UniGene clusters is not known definitively). We therefore revised our analysis to calculate LQS scores exclusively from UniGene clusters whose BOU representative was an mRNA with an annotated ORF window (indicating that the BOU is deposited in the correct orientation). We subsequently refer to these as 'oriented BOUs'. Another caveat arises in that 3' sequencing reads of directionally cloned ESTs are generally not reoriented before deposit of sequences in GenBank. We resolved this issue by 'flipping' in silico sequences annotated as 3' reads. The results of this analysis on the human EST dataset are shown in Figure 1. In our final analysis (green bars in Figure 1), the distribution of LQS scores across the full set of UniGene EST libraries is roughly bimodal, with a peak near LQS = 0.5 (random orientation of ESTs) and a peak near LQS = 1.0 (correct orientation of nearly all ESTs). These peaks correspond broadly to libraries generated by non-directional and directional cloning methods, respectively.
Of the 6,525 human and 566 mouse EST libraries considered, 899 and 176, respectively, had an LQS of greater than 0.95, indicating that these libraries were generated by an efficient method of directional cloning. The remainder of our analysis focused exclusively on the 1,151,724 human ESTs and 550,567 mouse ESTs derived from the libraries with LQS scores of greater than 0.95. A full list of the mouse and human EST libraries and their LQS scores is available at our website .
Statistically significant overrepresentation of misoriented ESTs in a subset of UniGene clusters
Our null hypothesis was that the relatively small fraction of misoriented ESTs from high-quality directionally cloned EST libraries (approximately 1.5%) represented random artifacts, leading to the expectation that they would be distributed across the full set of UniGene clusters in a manner proportional to the sizes of individual UniGene clusters. We applied binomial distribution probability analysis to identify clusters that significantly deviated from this expectation with a p-value of less than 0.00001 (roughly equivalent to the number of hypotheses being tested). This analysis was sufficient for UniGene clusters with an oriented BOU (see above). To avoid excluding from consideration UniGene clusters without oriented BOU sequences, we again applied the binomial distribution probability test, with an additional requirement; the result had to be significant (p < 0.00001) regardless of whether the BOU was correctly or incorrectly oriented. For example, the observation of a UniGene cluster with 100 ESTs deposited in the same orientation as the BOU and 100 ESTs oriented opposite to the BOU would deviate significantly from the null hypothesis expectation of 98.5% regardless of whether the BOU was correctly oriented or not. This approach identified 297 human and 252 mouse UniGene clusters that contained a statistically significant over-representation of incorrectly oriented ESTs.
Mapping of exon-intron organization of ESTs and mRNAs from candidate UniGene clusters to the mouse and human genomes
We downloaded the NCBI human genome draft assembly (Build 24)  and the Celera mouse genome draft assembly  in August 2001. Although the Celera mouse genome is not generally accessible, draft assemblies of the mouse genome based on the public sequencing effort have recently been released, and we anticipate that use of these assemblies would yield essentially equivalent results . The MEGABLAST tool  was used to identify the approximate genomic coordinates for each UniGene cluster (for example, the contig on which a given gene appeared to be located). The SIM4 tool  was then applied to map the exon-intron splicing coordinates of individual BOU and EST sequences more precisely. We have exploited mouse-human synteny and the availability of draft assemblies of the mouse and human genomes to generate a set of around 1.15 million mouse-human sequence alignments. These have been used to create overlay versions of each genome, in which the most conserved sequences (around 10% of each genome) is overlaid with homologous sequence of the other species. More detailed descriptions of the methodology followed and general statistics on this resource (HUMMUS) is available over the web . The graphical representations integrating information on transcript orientation, exon-intron structure, and mouse-human genomic conservation were generated using GNUPLOT . Graphical representations for the curated set of 144 human and 73 mouse candidates is available from our website , and an Excel-format summary is available as additional data with this paper.
Assessment of transcriptional directionality via RT-PCR assay
Primers were designed with the PRIMER3  algorithm and custom synthesized by Operon. For candidates, primers were selected to amplify a 100-200 base-pair (bp) sequence that was internal to a predicted region of transcriptional overlap. Control primers were designed to amplify 100-200 bp as well, either from non-overlapping regions of candidate transcripts or randomly selected regions of non-candidate transcripts. Templates included total human RNA from placenta, kidney, brain, thymus or uterus (Clontech). For each candidate or control, four RT-PCR reactions were carried out using total human RNA from a single tissue as template. We used the Qiagen One Step RT-PCR kit according to the manufacturer's protocol, except that reaction volume was reduced to 25 μl. Orientation of transcripts was assessed by restricting which primers were present during RT single-strand synthesis. The cycling parameters were as follows: (1) 50°C × 30 min, reverse transcription single-strand synthesis (with one, both or neither primer); (2) 95°C × 15 min, activate AmpliTaq polymerase, inactivate RT enzymes; (3) 4°C, add missing primers; (4) 94°C × 30 sec, commence PCR cycling; (5) 56°C × 30 sec; (6) 72°C × 30 sec; (7) go to step 4 (30 cycles in total); (8) 72°C × 10 min.
The inclusion of step (2) ensures that the RT will be inactivated before addition of missing primers. For each candidate or control primer pair, four RT-PCR reactions were carried out using total human RNA from a single tissue as template. In the first reaction, both primers were present during RT single-strand synthesis (positive control). In the second reaction, only the primer complementary to the antisense-orientation of the PCR product was present during RT single-strand synthesis (to assay for antisense transcription). In the third reaction, only the primer complementary to the sense-orientation of the PCR product was present during RT single-strand synthesis (to assay for sense transcription). In the fourth reaction, neither primer was present during RT single-strand synthesis (control for genomic contamination). Primers were designed to amplify regions of predicted bidirectional transcription for 39 of the human candidates and 18 negative controls. One of the 18 negative controls was discarded because no lane gave rise to a sharp band of the proper size. In all other cases, a sharp band of expected size was observed in one or more of the reactions.
Determination of protein homologies of sense and antisense oriented ESTs from candidate clusters
We applied the BLASTX tool [29,30] to blast each of the 45,588 relevant mRNA and EST sequences that belonged to both high-quality directionally cloned libraries and candidate UniGene clusters against the NCBI nr database (non-redundant database of protein sequences deposited in GenBank), with a threshold expectation value of 1e-10. Summary information for each candidate on the best protein alignment for ESTs oriented in each sense is available as additional data (see Additional data files).
Additional data files
We would like to thank Rob Mitra, Vasudeo Badarinarayana, and Yonatan Grad for helpful comments on the manuscript, and Fritz Roth for generously allowing us use of the LLAMA compute cluster.
- Kumar M, Carmichael GG: Antisense RNA: function and fate of duplex RNA in cells of higher eukaryotes. Microbiol Mol Biol Rev. 1998, 62: 1415-1434.PubMedPubMed CentralGoogle Scholar
- Vanhee-Brossollet C, Vaquero C: Do natural antisense transcripts make sense in eukaryotes?. Gene. 1998, 211: 1-9. 10.1016/S0378-1119(98)00093-6.PubMedView ArticleGoogle Scholar
- Lee RC, Feinbaum RL, Ambros V: The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell. 1993, 75: 843-854. 10.1016/0092-8674(93)90529-Y.PubMedView ArticleGoogle Scholar
- Moore T, Constancia M, Zubair M, Bailleul B, Feil R, Sasaki H, Reik W: Multiple imprinted sense and antisense transcripts, differential methylation and tandem repeats in a putative imprinting control region upstream of mouse Igf2. Proc Natl Acad Sci USA. 1997, 94: 12509-12514. 10.1073/pnas.94.23.12509.PubMedPubMed CentralView ArticleGoogle Scholar
- Sleutels F, Zwart R, Barlow DP: The non-coding Air RNA is required for silencing autosomal imprinted genes. Nature. 2002, 415: 810-813.PubMedView ArticleGoogle Scholar
- 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. 10.1073/pnas.261562698.PubMedPubMed CentralView ArticleGoogle Scholar
- Li AW, Murphy PR: Expression of alternatively spliced FGF-2 antisense RNA transcripts in the central nervous system: regulation of FGF-2 mRNA translation. Mol Cell Endocrinol. 2000, 170: 233-242. 10.1016/S0303-7207(00)00440-8.PubMedView ArticleGoogle Scholar
- Munroe SH, Lazar MA: Inhibition of c-erbA mRNA splicing by a naturally occurring antisense RNA. J Biol Chem. 1991, 266: 22083-22086.PubMedGoogle Scholar
- Lee JT, Davidow LS, Warshawsky D: Tsix, a gene antisense to Xist at the X-inactivation centre. Nat Genet. 1999, 21: 400-404. 10.1038/7734.PubMedView ArticleGoogle Scholar
- Kumar M, Carmichael GG: Nuclear antisense RNA induces extensive adenosine modifications and nuclear retention of target transcripts. Proc Natl Acad Sci USA. 1997, 94: 3542-3547. 10.1073/pnas.94.8.3542.PubMedPubMed CentralView ArticleGoogle Scholar
- Blin-Wakkach C, Lezot F, Ghoul-Mazgar S, Hotton D, Monteiro S, Teillaud C, Pibouin L, Orestes-Cardoso S, Papagerakis P, Macdougall M, et al: Endogenous Msx1 antisense transcript: in vivo and in vitro evidences, structure, and potential involvement in skeleton development in mammals. Proc Natl Acad Sci USA. 2001, 98: 7336-7341. 10.1073/pnas.131497098.PubMedPubMed CentralView ArticleGoogle Scholar
- Nemes JP, Benzow KA, Moseley ML, Ranum LP, Koob MD: The SCA8 transcript is an antisense RNA to a brain-specific transcript encoding a novel actin-binding protein (KLHL1). Hum Mol Genet. 2000, 9: 1543-1551. 10.1093/hmg/9.10.1543.PubMedView ArticleGoogle Scholar
- Lipman DJ: Making (anti)sense of non-coding sequence conservation. Nucleic Acids Res. 1997, 25: 3580-3583. 10.1093/nar/25.18.3580.PubMedPubMed CentralView ArticleGoogle Scholar
- Lehner B, Williams G, Campbell RD, Sanderson CM: Antisense transcripts in the human genome. Trends Genet. 2002, 18: 63-65. 10.1016/S0168-9525(02)02598-2.PubMedView ArticleGoogle Scholar
- NCBI UniGene. [http://www.ncbi.nlm.nih.gov/UniGene]
- Camargo AA, Samaia HP, Dias-Neto E, Simao DF, Migotto IA, Briones MR, Costa FF, Nagai MA, Verjovski-Almeida S, Zago MA, et al: The contribution of 700,000 ORF sequence tags to the definition of the human transcriptome. Proc Natl Acad Sci USA. 2001, 98: 12103-12108. 10.1073/pnas.201182798.PubMedPubMed CentralView ArticleGoogle Scholar
- Zhang Z, Schwartz S, Wagner L, Miller WA: A greedy algorithm for aligning DNA sequences. J Comput Biol. 2000, 7: 203-214. 10.1089/10665270050081478.PubMedView ArticleGoogle Scholar
- Florea L, Hartzell G, Zhang Z, Rubin GM, Miller W: A computer program for aligning a cDNA sequence with a genomic DNA sequence. Genome Res. 1998, 8: 967-974.PubMedPubMed CentralGoogle Scholar
- NCBI Human Genome. [http://www.ncbi.nlm.nih.gov/genome/guide/human]
- Celera. [http://www.celera.com]
- HUMMUS. [http://arep.med.harvard.edu/hummus.html]
- Sense and antisense (Church Lab). [http://arep.med.harvard.edu/antisense.html]
- Duret L, Dorkeld F, Gautier C: Strong conservation of non-coding sequences during vertebrate evolution: potential involvement in post-transcriptional regulation of gene expression. Nucleic Acids Res. 1993, 21: 2315-2322.PubMedPubMed CentralView ArticleGoogle Scholar
- Affymetrix Murine Genome U74 Array Set. [http://www.affymetrix.com/support/technical/product_updates/mgu74_product_bulletin.affx]
- Shoemaker DD, Schadt EE, Armour CD, He YD, Garrett-Engele P, McDonagh PD, Loerch PM, Leonardson A, Lum PY, Cavet G, et al: Experimental annotation of the human genome using microarray technology. Nature. 2001, 409: 922-927. 10.1038/35057141.PubMedView ArticleGoogle Scholar
- Ensembl Mouse Genome. [http://www.ensembl.org/Mus_musculus/]
- Gnuplot Central. [http://www.gnuplot.info]
- Primer 3 software distribution. [http://www-genome.wi.mit.edu/genome_software/other/primer3.html]
- Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. J Mol Biol. 1990, 215: 403-410. 10.1006/jmbi.1990.9999.PubMedView ArticleGoogle Scholar
- NCBI BLAST. [http://www.ncbi.nlm.nih.gov/BLAST]
- Wutz A, Smrzka OW, Schweifer N, Schellander K, Wagner EF, Barlow DP: Imprinted expression of the Igf2r gene depends on an intronic CpG island. Nature. 1997, 389: 745-749. 10.1038/39631.PubMedView ArticleGoogle Scholar
- Whitehead CM, Winkfein RJ, Fritzler MJ, Rattner JB: ASE-1: a novel protein of the fibrillar centres of the nucleolus and nucleolus organizer region of mitotic chromosomes. Chromosoma. 1997, 106: 493-502. 10.1007/s004120050271.PubMedView ArticleGoogle Scholar
- Lee YJ, Park CW, Hahn Y, Park J, Lee J, Yun JH, Hyun B, Chung JH: Mit1/Lb9 and Copg2, new members of mouse imprinted genes closely linked to Peg1/Mest(1). FEBS Lett. 2000, 472: 230-234. 10.1016/S0014-5793(00)01461-7.PubMedView ArticleGoogle Scholar
- Zavadil J, Svoboda P, Liang H, Kottickal LV, Nagarajan L: An antisense transcript to SMAD5 expressed in fetal and tumor tissues. Biochem Biophys Res Commun. 1999, 255: 668-672. 10.1006/bbrc.1999.0256.PubMedView ArticleGoogle Scholar
- McGuinness T, Porteus MH, Smiga S, Bulfone A, Kingsley C, Qiu M, Liu JK, Long JE, Xu D, Rubenstein JL: Sequence, organization, and transcription of the Dlx-1 and Dlx-2 locus. Genomics. 1996, 35: 473-485. 10.1006/geno.1996.0387.PubMedView ArticleGoogle Scholar
- Liu JK, Ghattas I, Liu S, Chen S, Rubenstein JL: Dlx genes encode DNA-binding proteins that are expressed in an overlapping and sequential pattern during basal ganglia differentiation. Dev Dyn. 1997, 210: 498-512. 10.1002/(SICI)1097-0177(199712)210:4<498::AID-AJA12>3.0.CO;2-3.PubMedView ArticleGoogle Scholar
- Zambrowicz BP, Imamoto A, Fiering S, Herzenberg LA, Kerr WG, Soriano P: Disruption of overlapping transcripts in the ROSA beta geo 26 gene trap strain leads to widespread expression of beta-galactosidase in mouse embryos and hematopoietic cells. Proc Natl Acad Sci USA. 1997, 94: 3789-3794. 10.1073/pnas.94.8.3789.PubMedPubMed CentralView ArticleGoogle Scholar
- Smilinich NJ, Day CD, Fitzpatrick GV, Caldwell GM, Lossie AC, Cooper PR, Smallwood AC, Joyce JA, Schofield PN, Reik W, et al: A maternally methylated CpG island in KvLQT1 is associated with an antisense paternal transcript and loss of imprinting in Beckwith-Wiedemann syndrome. Proc Natl Acad Sci USA. 1999, 96: 8064-8069. 10.1073/pnas.96.14.8064.PubMedPubMed CentralView ArticleGoogle Scholar
- NCBI LocusLink Record 114044. [http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=114044]
- Wroe SF, Kelsey G, Skinner JA, Bodle D, Ball ST, Beechey CV, Peters J, Williamson CM: An imprinted transcript, antisense to Nesp, adds complexity to the cluster of imprinted genes at the mouse Gnas locus. Proc Natl Acad Sci USA. 2000, 97: 3342-3346. 10.1073/pnas.050015397.PubMedPubMed CentralView ArticleGoogle Scholar
- Murphy PR, Knee RS: Identification and characterization of an antisense RNA transcript (gfg) from the human basic fibroblast growth factor gene. Mol Endocrinol. 1994, 8: 852-859. 10.1210/me.8.7.852.PubMedGoogle Scholar
- Seroussi E, Kedra D, Pan HQ, Peyrard M, Schwartz C, Scambler P, Donnai D, Roe BA, Dumanski JP: Duplications on human chromosome 22 reveal a novel Ret Finger Protein-like gene family with sense and endogenous antisense transcripts. Genome Res. 1999, 9: 803-814. 10.1101/gr.9.9.803.PubMedView ArticleGoogle Scholar
- Cooper PR, Smilinich NJ, Day CD, Nowak NJ, Reid LH, Pearsall RS, Reece M, Prawitt D, Landers J, Housman DE, et al: Divergently transcribed overlapping genes expressed in liver and kidney and located in the 11p15.5 imprinted domain. Genomics. 1998, 49: 38-51. 10.1006/geno.1998.5221.PubMedView ArticleGoogle Scholar
- NCBI LocusLink Record 93653. [http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=93653]
- NCBI LocusLink Record 93654. [http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=93654]
- NCBI LocusLink Record 93655. [http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=93655]
- Jong MT, Gray TA, Ji Y, Glenn CC, Saitoh S, Driscoll DJ, Nicholls RD: A novel imprinted gene, encoding a RING zinc-finger protein, and overlapping antisense transcript in the Prader-Willi syndrome critical region. Hum Mol Genet. 1999, 8: 783-793. 10.1093/hmg/8.5.783.PubMedView ArticleGoogle Scholar
- Potter SS, Branford WW: Evolutionary conservation and tissue-specific processing of Hoxa 11 antisense transcripts. Mamm Genome. 1998, 9: 799-806. 10.1007/s003359900870.PubMedView ArticleGoogle Scholar
- Campbell CE, Huang A, Gurney AL, Kessler PM, Hewitt JA, Williams BR: Antisense transcripts and protein binding motifs within the Wilms tumour (WT1) locus. Oncogene. 1994, 9: 583-595.PubMedGoogle Scholar
- Silverman TA, Noguchi M, Safer B: Role of sequences within the first intron in the regulation of expression of eukaryotic initiation factor 2α. J Biol Chem. 1992, 267: 9738-9742.PubMedGoogle Scholar
- Faurholm B, Millar RP, Katz AA: The genes encoding the type II gonadotropin-releasing hormone receptor and the ribonucleoprotein RBM8A in humans overlap in two genomic loci. Genomics. 2001, 78: 15-18. 10.1006/geno.2001.6650.PubMedView ArticleGoogle Scholar
- Wagener R, Kobbe B, Aszodi A, Aeschlimann D, Paulsson M: Characterization of the mouse matrilin-4 gene: a 5' antiparallel overlap with the gene encoding the transcription factor RBP-l. Genomics. 2001, 76: 89-98. 10.1006/geno.2001.6589.PubMedView ArticleGoogle Scholar
- Borsu L, Presse F, Nahon JL: The AROM gene, spliced mRNAs encoding new DNA/RNA-binding proteins are transcribed from the opposite strand of the melanin-concentrating hormone gene in mammals. J Biol Chem. 2000, 275: 40576-40587. 10.1074/jbc.M006524200.PubMedView ArticleGoogle Scholar
- Sureau A, Soret J, Guyon C, Gaillard C, Dumon S, Keller M, Crisanti P, Perbal B: Characterization of multiple alternative RNAs resulting from antisense transcription of the PR264/SC35 splicing factor gene. Nucleic Acids Res. 1997, 25: 4513-4522. 10.1093/nar/25.22.4513.PubMedPubMed CentralView ArticleGoogle Scholar
- Hastings ML, Milcarek C, Martincic K, Peterson ML, Munroe SH: Expression of the thyroid hormone receptor gene, erbAα, in B lymphocytes: alternative mRNA processing is independent of differentiation but correlates with antisense RNA levels. Nucleic Acids Res. 1997, 25: 4296-4300. 10.1093/nar/25.21.4296.PubMedPubMed CentralView ArticleGoogle Scholar
- Colombo P, Yon J, Garson K, Fried M: Conservation of the organization of five tightly clustered genes over 600 million years of divergent evolution. Proc Natl Acad Sci USA. 1992, 89: 6358-6362.PubMedPubMed CentralView ArticleGoogle Scholar
- Khochbin S, Lawrence JJ: An antisense RNA involved in p53 mRNA maturation in murine erythroleukemia cells induced to differentiate. EMBO J. 1989, 8: 4107-4114.PubMedPubMed CentralGoogle Scholar
- Celano P, Berchtold CM, Kizer DL, Weeraratna A, Nelkin BD, Baylin SB, Casero RA: Characterization of an endogenous RNA transcript with homology to the antisense strand of the human c-myc gene. J Biol Chem. 1992, 267: 15092-15096.PubMedGoogle Scholar
- Krystal GW, Armstrong BC, Battey JF: N-myc mRNA forms an RNA-RNA duplex with endogenous antisense transcripts. Mol Cell Biol. 1990, 10: 4180-4191.PubMedPubMed CentralView ArticleGoogle Scholar
- Fremeau RT, Popko B: In situ analysis of myelin basic protein gene expression in myelin-deficient oligodendrocytes: antisense hnRNA and readthrough transcription. EMBO J. 1990, 9: 3533-3538.PubMedPubMed CentralGoogle Scholar
- Tosic M, Roach A, de Rivaz JC, Dolivo M, Matthieu JM: Post-transcriptional events are responsible for low expression of myelin basic protein in myelin deficient mice: role of natural antisense RNA. EMBO J. 1990, 9: 401-406.PubMedPubMed CentralGoogle Scholar
- Laabi Y, Gras MP, Brouet JC, Berger R, Larsen CJ, Tsapis A: The BCMA gene, preferentially expressed during B lymphoid maturation, is bidirectionally transcribed. Nucleic Acids Res. 1994, 22: 1147-1154.PubMedPubMed CentralView ArticleGoogle Scholar
- Adelman JP, Bond CT, Douglass J, Herbert E: Two mammalian genes transcribed from opposite strands of the same DNA locus. Science. 1987, 235: 1514-1517.PubMedView ArticleGoogle Scholar
- Murashov AK, Wolgemuth DJ: Sense and antisense transcripts of the developmentally regulated murine hsp70.2 gene are expressed in distinct and only partially overlapping areas in the adult brain. Brain Res Mol Brain Res. 1996, 37: 85-95. 10.1016/0169-328X(95)00288-4.PubMedView ArticleGoogle Scholar
- Batshake B, Sundelin J: The mouse genes for the EP1 prostanoid receptor and the PKN protein kinase overlap. Biochem Biophys Res Commun. 1996, 227: 70-76. 10.1006/bbrc.1996.1469.PubMedView ArticleGoogle Scholar
- Bender TP, Thompson CB, Kuehl WM: Differential expression of c-myb mRNA in murine B lymphomas by a block to transcription elongation. Science. 1987, 237: 1473-1476.PubMedView ArticleGoogle Scholar
- Farrell CM, Lukens LN: Naturally occurring antisense transcripts are present in chick embryo chondrocytes simultaneously with the down-regulation of the alpha 1 (I) collagen gene. J Biol Chem. 1995, 270: 3400-3408. 10.1074/jbc.270.7.3400.PubMedView ArticleGoogle Scholar
- Belhumeur P, Lussier M, Skup D: Expression of naturally occurring RNA molecules complementary to the murine L27' ribosomal protein mRNA. Gene. 1988, 72: 277-285. 10.1016/0378-1119(88)90153-9.PubMedView ArticleGoogle Scholar
- PubMed. [http://www.ncbi.nlm.nih.gov/pubmed]
- LocusLink. [http://www.ncbi.nlm.nih.gov/LocusLink]
- Antisense transcripts in the human genome. [http://www.hgmp.mrc.ac.uk/Research/Antisense/]