- Research article
- Open Access
Structure and expression of two nuclear receptor genes in marsupials: insights into the evolution of the antisense overlap between the α-thyroid hormone receptor and Rev-erbα
BMC Molecular Biologyvolume 11, Article number: 97 (2010)
Alternative processing of α-thyroid hormone receptor (TRα, NR1A1) mRNAs gives rise to two functionally antagonistic nuclear receptors: TRα1, the α-type receptor, and TRα2, a non-hormone binding variant that is found only in mammals. TRα2 shares an unusual antisense coding overlap with mRNA for Rev-erbα (NR1D1), another nuclear receptor protein. In this study we examine the structure and expression of these genes in the gray short-tailed opossum, Monodelphis domestica, in comparison with that of eutherian mammals and three other marsupial species, Didelphis virginiana, Potorous tridactylus and Macropus eugenii, in order to understand the evolution and regulatory role of this antisense overlap.
The sequence, expression and genomic organization of mRNAs encoding TRα1 and Rev-erbα are very similar in the opossum and eutherian mammals. However, the sequence corresponding to the TRα2 coding region appears truncated by almost 100 amino acids. While expression of TRα1 and Rev-erbα was readily detected in all tissues of M. domestica ages 0 days to 18 weeks, TRα2 mRNA was not detected in any tissue or stage examined. These results contrast with the widespread and abundant expression of TRα2 in rodents and other eutherian mammals. To examine requirements for alternative splicing of TRα mRNAs, a series of chimeric minigenes was constructed. Results show that the opossum TRα2-specific 5' splice site sequence is fully competent for splicing but the sequence homologous to the TRα2 3' splice site is not, even though the marsupial sequences are remarkably similar to core splice site elements in rat.
Our results strongly suggest that the variant nuclear receptor isoform, TRα2, is not expressed in marsupials and that the antisense overlap between TRα and Rev-erbα thus is unique to eutherian mammals. Further investigation of the TRα and Rev-erbα genes in marsupial and eutherian species promises to yield additional insight into the physiological function of TRα2 and the role of the associated antisense overlap with Rev-erbα in regulating expression of these genes.
Recent genome-wide studies have revealed far more extensive transcription of mammalian genomes than previously recognized [1–6]. One consequence of this pervasive transcription is the occurrence of widespread bidirectional transcription that results in overlapping transcripts antisense to most known mRNAs [5, 7–9]. Such antisense transcription takes several forms. Both long and short non-coding antisense transcripts have been described overlapping either the 5' or 3' ends of mRNAs. In other instances two mRNAs encoded on opposite strands share such an overlap. Although regulatory implications of antisense transcription are unclear, many studies suggest that the mechanisms associated with expression of antisense RNA in eukaryotes are remarkably diverse and mechanistically complex [9–14].
One of the first examples of an antisense overlap between two mRNAs involves two nuclear receptor genes, the α-type thyroid hormone receptor gene (TRα; also NR1A1 or THRA) and the Rev-erbα gene (NR1D1) [15–17]. These genes share an antisense overlap that reflects the presence of a novel alternatively spliced mRNA that is highly conserved in most mammals but is absent in non-mammalian vertebrates . Alternative processing of TRα pre-mRNAs yields two functionally distinct proteins. One is the canonical thyroid hormone (T3) receptor, TRα1, that activates target genes in a thyroid hormone (T3)-dependent manner. The other, TRα2, has a novel C-terminal sequence encoded by an exon antisense to Rev-erbα. TRα2 mRNA lacks both the hormone-binding domain and a critical activation domain present in TRα1 and may function as a weak dominant negative repressor, repressing expression of genes activated by TRα1 in the presence of T3 . Rev-erbα has a DNA-binding specificity distinct from TRα1 and TRα2. Like TRα1, Rev-erbα possesses a canonical ligand binding domain but, like TRα2, it lacks the AF-2 activation domain. Thus, both Rev-erbα and TRα2 are constitutive repressors, although they bind distinct sets of target genes.
TRα1 and Rev-erbα play important roles in metabolic and developmental regulation [19–21]. TRα1 mediates T3 activity critical for vertebrate development and homeostatic metabolism. Rev-erbα plays critical roles in adipogenesis and lipid metabolism, while occupying a specific node in the mammalian circadian clock [22–26]. While previously considered a ligandless orphan receptor, Rev-erbα has recently been shown to bind heme specifically within its canonical ligand binding domain (LBD) [24, 26]. Interactions of Rev-erbα with its co-repressors are enhanced in the presence of heme and, with heme bound, may respond to physiological regulators such as nitric oxide [23, 27, 28]. Thus, Rev-erbα has emerged as an important integrator of inputs from multiple physiologic pathways [29–31].
The overlap architecture of the region encoding TRα and Rev-erbα (here referred to as the TRα/Rev-erbα locus) and the physiological importance of these nuclear receptors raise important questions relating to the function, evolution and regulation of this locus. The evolutionary constraints imposed on such a bidirectional coding sequence argue that it is highly unlikely that such an overlap would occur without some functional benefit. Several lines of evidence support the hypothesis that expression of Rev-erbα mRNA may affect the balance between TRα1 and TRα2 through inhibition of TRα2 splicing [32–35]; however, such an interaction has not been demonstrated under physiological conditions.
In this study we first analyze the structure of the TRα/Rev-erbα locus in Monodelphis domestica, the gray short-tailed opossum, the first published marsupial genome , and compare it with that of other marsupials. Individually, TRα and Rev-erbα genes and transcripts are highly conserved in the opossum in comparison with those of eutherian mammals. However, the region corresponding to the Rev-erbα/TRα2 overlap in eutherian mammals displays several divergent features. Most of the C-terminal coding sequence unique to TRα2 is missing in M. domestica due to the introduction of an in-frame stop codon that eliminates the bidirectional coding overlap with Rev-erbα. Furthermore, expression of TRα2 mRNA was not detected in the opossum in any of a variety of tissues or developmental stages. These results are surprising given the remarkable conservation of TRα2 and its abundant and developmentally regulated expression in eutherian mammals. Analysis of chimeric minigenes comprised of opossum and rat sequences demonstrated that opossum sequences homologous to the TRα2-specific splice site (3'ss) are very poorly utilized in comparison with similar rat minigenes. These studies thus document the evolution of a novel alternatively spliced exon in mammals and the creation of antisense overlap between two important regulatory proteins.
Structures of eutherian TRα1 and Rev-erbα mRNAs are conserved in marsupials
Inspection of the genome of Monodelphis domestica reveals a single locus on chromosome 2 in which the TRα and Rev-erbα genes are arrayed in a convergent orientation similar to that in rat and other mammalian genomes (Figure 1). The coding regions for TRα1 (additional file 1 Panel A) and Rev-erbα are separated by only 3.85 kb, similar to the close spacing of the coding regions in the rat genome (3.71 kb). Furthermore, the non-coding sequence extending across and between the 3' ends of these genes is also very well conserved (Figure 1B). The most striking conservation is located at the extreme 3' end of TRα1 mRNA, in which a sequence of 140 bp is almost identical to that in the rat and other eutherian mammals. The region surrounding the 3' end of Rev-erbα is also highly conserved and is flanked upstream and downstream by homopolymeric runs of 3 or more C (C-rich or Cn) and G (G-rich or Gn) residues, respectively, as is the case in other mammals (Figure 1B and additional file 2).
Closer inspection of the sequence for M. domestica TRα, however, reveals that the first coding exon of TRα1 mRNA is entirely missing in the genome assembly due to a 5 kb gap . Also, apparent sequencing errors at positions 54 and 124 of exon 6 disrupt the reading frame for TRα1 by introducing, respectively, a one-base insertion and a stop codon. The complete coding sequence for M. domestica TRα1 was obtained by sequencing cDNA using primers complementary to conserved sequences in the 5' UTR and at intervals throughout the coding sequence. The TRα1 coding sequence for M. domestica from this analysis was identical in all other respects to the current genome assembly (additional file 1 Panel A). As shown in Figure 1 the amino acid sequence of opossum TRα1 is closely similar to that of rat, with 93% identity at the amino acid level. Rev-erbα is also well conserved with 84% identity. The 3' ends of TRα1 and Rev-erbα mRNAs from opossum were mapped via RNase protection assays and 3' RACE, respectively (results not shown), and were found to coincide precisely with the sites previously reported for rat [15, 32] as indicated in additional file 2. To confirm these observations, mRNA sequences for TRα1 and Rev-erbα were determined for a second marsupial, Potorous tridactylus (the long-nosed potoroo). The opossum and potoroo amino acid sequences, representing the two largest marsupial orders, Didelphia and Diprotodontia, about 80 My diverged , are very similar: 99% and 96% identity for TRα1 and Rev-erbα, respectively (additional file 1).
In contrast to the strong conservation of TRα1 and Rev-erbα, sequences corresponding to the alternatively spliced exon 10 of TRα2 mRNA display a puzzling mix of conserved and non-conserved features. Sequences corresponding to the core 3'ss elements for this exon are nearly identical in rat and M. domestica, including the intronic polypyrimidine tract and the non-consensus AAG immediately preceding the 3'ss (Figure 2B and additional file 2). Sequence at the 5' end of exon 10, corresponding to 11 codons in rat TRα2, is also well conserved but at that point the similarity between marsupial and eutherian sequence ends abruptly until reaching the coding sequence for Rev-erbα. The divergent region includes a 12 nucleotide insertion in opossum and an in-frame stop codon immediately preceding the stop codon for Rev-erbα on the strand antisense to TRα. These adjacent stop codons, present in both species of opossum but absent in wallaby and potoroo, as described below, precisely eliminate the bidirectional coding overlap between TRα2 and Rev-erbα mRNAs seen in rat and other eutherian mammals [17, 38]. Finally, the Rev-erbα amino acid sequence encoded by its terminal exon is identical in opossum and rat. However, there are 26 synonymous substitutions in Rev-erbα sequence, almost all of which alter the coding potential of the TRα strand in opossum.
To further explore the significance of this sequence comparison, we examined 1.5 kb of genomic sequence that span the 3' end of TRα1 and extend across exon 8 and part of exon 7 of Rev-erbα in three other marsupial species: Didelphis virginiana (the Virginia opossum), P. tridactylus, and Macropus eugenii (the tammar wallaby). Alignment of all four marsupial sequences with those of rat shows a high level of conservation of marsupial features similar to that shown in Figure 1B and additional file 2. As expected, since both are of the order Didelphimorphia, the sequence of the opossum D. virginiana is very similar to that of M. domestica (89% identity across mostly non-coding sequence). Both include a 12-nucleotide insertion and in-frame stop codon in the region adjacent to the stop codon for Rev-erbα (Figure 2B). Sequences for wallaby and potoroo (both of the order Diprotodontia) are similar to those from the opossums in the region homologous to exon 10 of TRα2. Furthermore, the amino acid sequences encoded by the 3' exon of Rev-erbα in all four marsupials are identical to that of rat except for a single residue change in the potoroo, although each sequence contains many substitutions which are silent with respect to Rev-erbα (Figure 2C).
Despite their high degree of similarity overall, the potoroo and wallaby sequences differ from those of the opossums in one notable respect: there is a two-basepair deletion immediately downstream of sequences homologous to the 3'ss of exon 10 in TRα2 mRNA. This deletion disrupts the reading frame corresponding to the C-terminus of TRα2 and creates a novel open reading frame that extends past the Rev-erbα stop codon, overlapping the Rev-erbα coding region by 16 codons (Figure 2B and 2C).
Expression of TRα and Rev-erbα mRNAs in marsupials
Given the conservation of so many features within the TRα/Rev-erbα locus in marsupials, it seemed likely that marsupials might express an alternatively spliced mRNA yielding a truncated version of TRα2. To characterize expression of mRNAs from Rev-erbα and TRα genes we examined RNA from multiple tissues from M. domestica at six different ages ranging from newborn animals (0 days) to weanlings (9 weeks) and young adults (18 weeks).
Expression of both TRα1 and Rev-erbα mRNAs was readily detected at all stages by RT-PCR. Since newborn and one-week animals are only 10 mm and 16 mm in length, respectively, these animals were simply divided into two parts, heads and bodies, for extraction of RNA. Beginning with two-week animals, six different tissues were collected: cerebral cortex, cerebellum, liver, kidney, skeletal muscle and heart. For each tissue at each stage, both TRα1 and Rev-erbα expression were readily detected by realtime RT-PCR and expression of both mRNAs was confirmed by sequencing PCR amplification products. Results of these assays are summarized in Figure 3A for heads and bodies of newborn (0 or 1 week), and cerebral cortex and livers of older animals (2-18 weeks). Similar results are shown for four other tissues in additional file 3. Increasing expression of Rev-erbα correlates strongly with increasing age (with a linear correlation coefficient, r > 0.95 in skeletal muscle, kidney, cerebellum and liver, and r> 0.74 in brain and heart). The highest levels of Rev-erbα expression were observed in liver and skeletal muscle; the lowest levels were in heart. In all tissues except liver Rev-erbα showed a large, 4- to 6-fold increase in expression between 4 and 9 weeks (p < 0.001). In fact, TRα1 expression also showed a strong, if less pronounced, correlation with increasing age in all tissues except heart (r ≥ 0.95 for TRα1 in kidney and cerebellum and r ≥ 0.74 in brain, muscle and liver) (additional file 3).
In contrast to TRα1 and Rev-erbα mRNA expression, both realtime and conventional RT-PCR assays for TRα2 were consistently negative. To confirm this result, multiple pairs of primers were tested, including forward primers in exons 7, 8 and 9 and reverse primers annealing to different sites within the region homologous to TRα2 exon 10 and Rev-erbα exon 8. While the threshold values obtained for TRα1 were highly reproducible, typically varying by <0.2 cycles for a given set of replicas, replica samples run with primers for TRα2 were highly variable, either displaying large threshold values or altogether failing to cross the threshold. Melt-curves (not shown) and gel electrophoresis of these products revealed a heterogeneous array of non-specific products, as illustrated in the replica lanes shown in Figure 3D. Additional file 4 summarizes threshold values from realtime RT-PCR for TRα2 across all tissues and ages (including the testes from 2 and 4 month animals) and shows parallel values for TRα1 for comparison. Two different downstream primers for TRα2 were used for most of these assays. In none of the assays was a product obtained consistent with splicing of TRα2 mRNA in opossum.
To test the efficiency and sensitivity of the TRα2 primers, realtime RT-PCR was carried out with a short synthetic DNA template matching the predicted amplicon sequence of opossum TRα2 mRNA. Serial dilutions of this template were assayed in the presence of a fixed amount of a negative cDNA preparation of M. domestica total RNA from cerebral cortex. The plot in Figure 3E indicates that these TRα2-specific primers have a high efficiency, >99% , and a sensitivity at CT = 33 of less than 50 copies of TRα2 cDNA.
Expression of TRα and Rev-erbα mRNAs was also measured by RNase protection assays on total RNA from the head (day 0 and week 1 animals) or cerebral cortex (2-18 week animals) as shown in Figure 3B. From the specific activity of the probe, the level of TRα1 mRNA was determined to vary between 0.3 and 0.7 amol/μg RNA (1 amol = 10-18 moles) with no clear stage-specific trend. On the other hand, Rev-erbα increased steadily from less than 0.1 amol/μg RNA in day 0 animals to about 1.0 amol/ug RNA in 18 week animals (Figure 3C), a trend paralleling the increase observed by realtime RT-PCR in cerebral cortex and several other tissues (Figure 3A and additional file 3). TRα1 expression was assayed with a probe that extended across the 3' and 5' splice sites for exon 9A. The length of the protected fragment for TRα1 reveals efficient splicing of the former but not the latter site. In older opossums (9 and 18 weeks) a faint band corresponding in size to that expected for use of the 5'ss of TRα2 was observed. The presence of this band raises the possibility that an alternatively spliced product with a similar 5'ss to TRα2 is expressed in older animals. However, further efforts to identify such a product were unsuccessful.
Requirements for splicing of TRα2
Although we found no evidence of TRα2 expression in M. domestica, the possibility remained that its expression is tightly restricted to some tissue, developmental stage or particular set of physiological conditions. To examine the intrinsic competency for splicing of conserved sequences corresponding to the 3'ss of exon 10 of TRα2, an opossum minigene was constructed based on the structure of a rat minigene that efficiently expressed TRα2 mRNA in transfection assays [32, 40]. The construct consisted of exons 7-10 of TRα, extending downstream to exon 7 of Rev-erbα, with all the intron sequences except for a large deletion within intron 9 that eliminated the competing TRα1 polyadenylation site and much of the preceding 3' UTR (Figure 4A). This minigene was efficiently expressed in transient transfection assays in HEK 293 cells. RNA processing was assayed for both spliced (TRα2) and partially spliced ("unspliced" TRα1-like) RNA by conventional PCR (Figure 4B) and realtime RT-PCR (Figure 4C), as well as RNase protection assays (Figure 4D). Bands corresponding to spliced TRα2 were evident for the rat minigene (R-R) in all three assays. Real-time PCR and RNase protection assays yielded quantitative measures of TRα2 splicing relative to unspliced TRα1 of 64-77%, consistent with earlier results for this minigene construct. Conventional PCR employing larger amplicons reveals a minor amount of a second spliced product corresponding to the minor spliced isoform, TRα3 (small arrow, Figure 4B). In contrast, no spliced product could be detected from the opossum minigene (O-O). The larger band evident in Figure 4B (lane O-O) corresponds to readthrough sequence, while the heterogeneous bands in Figure 4C represent non-specific amplification products observed after 44 cycles of amplification. Close examination of the RNase protection assays (Figure 4D) revealed no bands corresponding to splicing of exon 9A. Figure 4E summarizes the results of each of the assays. A low level of residual background activity was routinely detected with RNase protection assays and real-time PCR yielding heterogenous amplification products, as indicated by asterisks in Figure 4E.
Given the efficiency of TRα2 splicing in the rat construct (R-R), the complete lack of expression of TRα2 splicing with the opossum minigene (O-O) was surprising. Both the 5' and 3' splice sites corresponding to TRα2 are highly conserved in the rat and the opossum (Figure 2B and additional file 2), as are intronic sequences adjacent to 5' and 3' splice sites. Despite the high conservation of these sequences in marsupials and eutherians, the opossum TRα gene may lack cis-acting elements required for TRα2 expression [32, 38, 40]. To extend this analysis chimeric minigenes were constructed and assayed for their ability to express TRα2. One construct, O-R, consisting of opossum exons 7-9 upstream of the rat TRα2 3' splice site and exon10, expressed TRα2 transcripts at levels similar to the rat minigene (Figure 4B, C, and 4D). Sequencing of the RT-PCR product obtained from transfected cells confirmed that this product was accurately spliced from the homologous 5' splice site in exon 9 of opossum to exon 10 of rat. In contrast, the reciprocal construct (R-O) with rat sequences upstream (exons 7-9) and opossum sequences downstream (exon 10) was apparently unable to express spliced TRα2 mRNA. In all of the constructs tested, the upstream exons 7, 8 and 9 are efficiently and accurately spliced, but only in O-R, not O-O or R-O, is the downstream splice site corresponding to exon 10 efficiently used. These results demonstrate that sequences homologous to the TRα2 5' splice site in exon 9 of opossum are fully competent for splicing, while those homologous to the TRα2 3'ss in exon 10 are not.
To more precisely define cis-acting elements required for TRα2 splicing, two additional minigenes were tested. The first of these, designated O/Rx10, was constructed from the O-R chimera by replacing the rat sequence upstream of the TRα2 3' splice site but downstream of intron 9 deletion with the opossum sequence present in the O-O minigene. The other chimeric minigene, O/RpA, was constructed by incorporating opossum sequences into the length of exon 8, thus eliminating the rat 3'ss for TRα2 while retaining the rat TRα2 polyadenylation site (Figure 4A). When these chimeric minigenes were transfected into HEK 293 cells the efficiency of TRα2 splicing was greatly reduced or eliminated. Careful analysis revealed that the O/Rx10 minigene expressed a small amount of correctly spliced chimeric TRα2 mRNA, as indicated by both RPA and RT-PCR assays, but with greatly reduced efficiency, strongly suggesting that opossum sequences in the intron upstream of exon 10 are incompatible with TRα2 splicing.
Post-run analysis of both R-O and O/RpA real-time products yielded a weak band in the region expected for correctly spliced product as indicated by the asterisks in Figure 4C. However, the difference between TRα1 and TRα2 CT values (ΔCT in Figure 4C) was comparable or less than that observed for the O-O minigene. Correct splicing of these transcripts could not be confirmed by other assays. Cryptic splicing within the opossum sequence in O/RpA, 60 nt from the chimeric junction of opossum and rat sequence, was detected by sequencing a band obtained by conventional PCR. Conventional PCR also demonstrated that R-R, O-R and O/Rx10 all express a small amount of spliced TRα3, a minor variant of TRα2 that uses an alternative 3'ss 117 bp downstream of the TRα2 3'ss in exon 10 . For R-R and O-R this represents a minor product, slightly smaller than that for TRα2. For O/Rx10, however, this band is proportionately more intense, indicating a loss of specificity as well as a greatly reduced efficiency of TRα2 splicing for the O-Rx10 minigene.
The analysis described above strongly suggests that opossum sequences upstream of the 3'ss are incompatible with TRα2 splicing. The discrepancy between TRα2 splicing of O-R and O/Rx10 indicates that the TRα gene of M. domestica lacks upstream cis-acting elements required for efficient splicing of TRα2. Furthermore, inclusion of the rat TRα2 polyadenylation site, which is not conserved in marsupial species (additional file 2), is insufficient to restore accurate and efficient splicing in chimeric constructs O/Rx10 and O/RpA.
Expression of endogenous and transfected mRNAs in PtK1 cells
Analysis of the splicing of chimeric minigenes demonstrates that the opossum sequences homologous to the 3'ss of TRα2 splicing are not used efficiently. This most likely reflects the lack of cis-acting elements in opossum that are necessary for TRα2 expression [38, 40]. It is possible, however, that the failure of opossum tissues to express TRα2 reflects the presence or absence of a factor that specifically affects TRα2 processing. To test this possibility we examined the processing of the rat TRα minigene in potoroo PtK1 cells. These cells, derived from kidney epithelium, endogenously expressed both TRα1 and Rev-erbα, but not TRα2 (Figure 5A). PtK1 cells were then transfected with three previously characterized full-length rat minigenes : a wildtype minigene extending uninterrupted from exon 7 through exon 10 of TRα2 (and further downstream to exon 2 of Rev-erbα); a mutant with a single base replacement at the 5'ss of exon 9A (+5C/G), which overexpresses TRα2 mRNA; and a second mutant at that 5'ss with substitution at the adjacent site (+6T/G), which results in a decrease in TRα2 mRNA. When transfected into PtK1 cells, each of these constructs expressed TRα1 and TRα2 mRNAs at levels expected from previous expression studies in rodent and human cells (Figure 5B) . These results demonstrate that marsupial cells expressing only endogenous TRα1 are fully capable of splicing rat TRα2.
Results of these studies show that TRα1 and Rev-erbα mRNAs are widely expressed in opossum in a manner similar to that in rodents, where both receptors are readily detectable in most tissues throughout development [20, 42, 43], with expression peaking in post-natal, pre-weanling mice and rat. Since marsupials are born at a relatively early stage of development and mature more slowly than rats and mice, the ages used here correspond roughly to late pre-natal through post-weanling stages in rats. The striking increase in the expression of Rev-erbα in most tissues with increasing age is also consistent with that reported for rat . However, the failure to detect expression of TRα2 in a broad range of tissues at ages ranging from day 0 newborns to 4 month-old young adults contrasts strongly with the co-expression of all three of these mRNAs from this locus in rats, mice and other mammals.
The expression of any one of these mRNAs, TRα1, TRα2 or Rev-erbα, is almost always accompanied by expression of the other two, although the relative ratios vary widely [20, 21, 33]. Our analysis focused on the expression of these receptors in the central nervous system, as TRα2 mRNA is expressed at much higher levels than TRα1 in both cerebral cortex and cerebellum [42, 43]. For example, in newborn rat brain TRα2 mRNA is more than 30-fold more abundant than TRα1 . Furthermore, TRα2 expression in rats is several fold greater than TRα1 expression in brain, liver and kidney at all ages examined (day 15 embryos to adult) , a striking contrast to the total absence of TRα2 mRNA observed here in opossum and PtK1 cells.
Conservation of sequences in TRα and Rev-erbα
The conservation of sequence spanning the 3' ends of TRα1 and Rev-erbα has several features of interest with respect to the evolution of TRα2 and the antisense overlap with Rev-erbα. The amino acid sequence encoded by the 3' exon of Rev-erbα is nearly identical in marsupial and eutherian mammals (100% identity in rat and opossum), suggesting that the coding sequence of Rev-erbα accommodated the evolution of its antisense overlap with TRα2 within its functional constraints as the coding sequence for TRα2 threaded through 67 (antisense) codons of Rev-erbα and into the final intron of the Rev-erbα gene. Interestingly, the one specific function ascribed to the unique C-terminal sequence of TRα2, that associated with phosphorylation of certain serine residues [45, 46], maps to positions in TRα2 that are encoded antisense to the final intron of Rev-erbα and are therefore not constrained by requirements for Rev-erbα coding as shown in Figure 2A.
The region in which the 3' ends of TRα1 and Rev-erbα mRNAs converge is framed by exceptionally conserved sequences within the 3'UTR of each mRNA (additional file 2 ), suggesting the presence of numerous functional elements important for the expression and regulation of both nuclear receptor proteins. The apparent conservation of core splice site elements important for TRα2 expression is puzzling, given the apparent absence of TRα2 mRNA in marsupials, and raises the question of whether these elements correspond to TRα2 or have other functions. In particular, several features are consistent with conservation of the alternatively spliced TRα2-specific exon in marsupials: (1) conservation of the non-canonical AAG trinucleotide identical to that found in all known eutherian genomes at the 3' end of exon 9 for TRα2; (2) the conservation of adjacent sequences corresponding to the 5' end of exon 10; and (3) the conservation of the polypyrimidine tract that is an essential feature of the core sequence of the 3'ss. In fact, the marsupial polypyrimidine tract appears a better match to the mammalian consensus sequence than that found in rat: a G at position -6 is replaced with T in all four marsupials and a somewhat higher proportion of T/C residues is present in the marsupial polypyrimidine tracts [47, 48].
However, other features of this region are consistent with the absence of TRα2 in opossum: (1) the open reading frame corresponding to the extended C-terminus of TRα2 is not conserved in four marsupials except for the first 11 codons of TRα2 exon 10 in the opossums; (2) the bidirectional coding overlap with Rev-erbα present in all eutherian species is eliminated in both opossums; (3) the TRα2 ORF is disrupted in wallaby and potoroo by a frame-shift deletion adjacent to the presumptive splice site; (4) a conserved polyadenylation signal corresponding to exon 10 of TRα2 is absent, and there is overall poor conservation of sequence within the final intron of Rev-erbα (additional file 2); and (5) there is poor conservation of intronic and exonic splicing enhancer sequences required for TRα2 splicing [38, 40]. These include the SEα2 enhancer downstream of the coding sequence for TRα1 , an intronic splicing enhancer near the 3' splice site of exon 10 and a exonic enhancer located within the bidirectional coding sequence of exon 10 . Taken together, these differences suggest that the conservation of the elements associated with TRα2 mRNA processing may reflect other functional requirements at this locus unrelated to TRα2 expression.
Splicing of TRα2 mRNA in chimeric minigenes
Analysis of splicing of opossum and rat minigenes corroborated our finding that alternatively spliced TRα2 mRNA is missing in marsupials. Most telling was the O/Rx10 construct, which includes the entire length of rat exon 10, identical to that in O-R, except for replacement of 180 bp of rat sequence upstream of the TRα2 3'ss with 220 bp of opossum sequence. The longer opossum sequence closely corresponds to a previously tested construct (ErbAΔXP) which was spliced as efficiently as the shorter sequence . The overall level of splicing of O/Rx10 was <10% that of O-R and the predominant spliced product corresponded to TRα3, a minor splice variant isoform, not TRα2, as shown by sequencing of RT-PCR products (Figure 4B). These experiments demonstrate that opossum sequences homologous to the TRα2-specific 3'ss are almost completely inactive in splicing transfected minigenes. In addition to the role of these intronic sequences in blocking TRα2 splicing, the opossum sequences also lacked a conserved polyadenylation signal hexanucleotide (CATAAA) that is conserved in all known TRα2 mRNAs. However, the inclusion of this element (together with much additional rat sequence) in the O/Rx10 and O/RpA minigenes was not sufficient to rescue efficient splicing.
Evolution of TRα2 mRNA in mammals
A number of studies suggest that the evolution of new alternative exons provides an important pathway for the creation of novel proteins and regulatory pathways in eukaryotic cells [49–52], reviewed in . Given the prominence of TRα2 as an abundant isoform in eutherian mammals , its proposed role as a dominant negative modulator of thyroid hormone activity  and the possible regulatory implications of its overlap with Rev-erbα , the evolution and function of TRα2 in mammals are of particular interest.
Three distinct evolutionary scenarios or models may account for similarities and differences of this locus in marsupial species as summarized in Figure 6. The first scenario, in which TRα2 expression is restricted to the eutherian lineage, suggests that some features of the marsupial TRα/Rev-erbα locus are intermediate between those of eutherian mammals and non-mammalian vertebrates. For example, the poly(A) sites of TRα1 and Rev-erbα are nearly identically spaced in rat and opossum and much closer than in the amphibian Xenopus tropicalis, where they are more than 10 kb apart . On the other hand, marsupials, like non-mammalian vertebrates, appear to lack an authentic TRα2 isoform. Also, the amino acid sequence of the marsupial TRα1 (additional file 1 Panel A) shows striking similarities to that of chicken. The opossum TRα1 amino acid sequence differs from chicken in only 22 positions as compared to 25 positions in rat (additional file 1 Panel A), despite a slightly greater similarity between the opossum and rat nucleotide coding sequences. In 20 of the 25 positions where opossum and rat TRα1 differ, the opossum sequence is identical to that of chicken. Most of these sites are also conserved in frog, a more distantly related vertebrate (result not shown). These observations suggest that the marsupial TRα locus has retained several characteristics reminiscent of non-mammalian vertebrates.
The second scenario, in which a truncated form of TRα2 is expressed in some limited (though as yet undetected) fashion, also suggests that the marsupial locus may represent a stage intermediate between the more complex eutherian locus, with its extended antisense overlap with Rev-erbα and abundant expression of TRα2, and the non-mammalian vertebrates, which entirely lack TRα2. Support for this model is found in the fact that many features characteristic of the marsupial TRα/Rev-erbα locus are associated with the creation of novel alternatively spliced exons, according to several recent studies [49–53]. These include inefficient splicing coupled with restricted or low levels of expression, as posited in the second model, and the absence of an extended open reading frame.
The third scenario, that TRα2 evolved early but was lost in the marsupial lineage, suggests that features characteristic of TRα2 that are shared by marsupial and eutherian mammals are vestiges of this loss. These vestigial features may have acquired other secondary functions within this locus. Although this scenario may appear overly complex and hence unlikely, careful analysis of nuclear receptor evolution throughout multiple metazoan phyla reveals that loss within a particular lineage is not at all uncommon . Among approximately 25 nuclear receptor proteins that appear to have been present in the last common ancestor of vertebrate and bilaterian invertebrate species, a number of nuclear receptors, including both thyroid hormone receptor genes, are missing entirely from major groups of animals such as nematodes or arthropods while specific orthologs for both thyroid hormone receptors and Rev-erb are found in more distantly related animals, such as mollusks, tunicates and flatworms [55–57].
Although none of the three models can be definitively ruled out, on balance, the first scenario might be judged most likely. The failure to detect TRα2 expression in tissues and stages where it is abundantly expressed in eutherians, the poor efficiency of TRα2 minigene splicing, the absence of a discernible TRα2 polyadenylation signal element and differences between the presumptive coding sequence for TRα2 in two marsupial orders weigh against the second model. The differences and similarities among the marsupial sequences within the 3' UTR of Rev-erbα (Figure 2 and additional file 2), including strict conservation of 3' processing sites for TRα1 and Rev-erbα, seem most consistent with the first model, in which sequence conservation reflects regulatory requirements for these convergent mRNAs rather than vestiges of a once functional, alternatively spliced mRNA, as suggested in the third model.
If the first scenario is correct, apparent similarities between marsupials and eutherians in the region corresponding to the TRα2 3' ss and exon 10 must reflect other conserved functions. The first model thus suggests core elements of the TRα2 splice site were coopted (or "exapted") [58, 59] during the evolution of TRα2. What, then, are the possible functions of these sequences? It is likely that they reflect the overlay of elements involved in regulating expression of TRα and Rev-erbα, as well as other possible roles. Accumulating evidence suggests that sequences at the 3' ends of genes play multiple roles in regulating gene expression. These include target sites for microRNA binding , sites for binding of proteins regulating transcription termination, 3' end processing, turnover and translation [61, 62], and promoters for expression of non-coding transcripts that are often found within the 3' UTRs of transcripts [1, 3, 7].
Physiological correlates of nuclear receptor function
The physiological role of TRα2 has remained elusive since its discovery more than two decades ago, despite numerous genetic and molecular investigations of its function [63–65]. Although many studies have focused on its role as a weak dominant negative competitor of TRα1 or TRβ isoforms [19, 66–68], other studies have pointed to its phosphorylation-dependent RNA-binding activity and cytoplasmic localization [46, 69, 70]. These properties, along with the developmental regulation of its expression, suggest a role for TRα2 in developmental regulation of T3 activity [71, 72]. The present study strongly indicates that its function is specific to the eutherian lineage, and thus may ultimately be resolved by comparative analysis of relevant aspects of marsupial and eutherian physiology.
The multiple roles ascribed to Rev-erbα and TRα1 in metabolic, circadian and developmental signaling [22, 27, 73–75] require a diversity of regulatory inputs that determine the expression and physiological activity of these receptors [20, 21]. The evolution of the antisense overlap between TRα2 and Rev-erbα mRNAs in eutherian mammals is likely to be associated with additional mechanisms that affect the expression or activity of both receptors.
Analysis of the TRα/Rev-erbα locus in four marsupial species strongly suggests that the alternatively spliced TRα2 isoform, which is widely and abundantly expressed in all known eutherian mammals, is not present in marsupials. Three observations support this conclusion: first, we can find no evidence for TRα2 expression in any of multiple tissues or developmental stages surveyed; second, the coding sequence specific to TRα2 is severely truncated in opossums and completely altered in the wallaby and potoroo; and third, opossum sequences homologous to the 3'ss specific for TRα2 are utilized very inefficiently for splicing of chimeric minigenes. The antisense overlap between TRα2 and Rev-erbα mRNAs, which is tightly conserved in eutherian mammals, appears to have evolved within the constraints of the conserved C-terminal coding sequence of Rev-erbα. These results suggest that comparative analysis of these genes in marsupials and eutherian mammals will provide further insight into the evolution, function and expression of TRα2 and the regulatory implications of its antisense overlap with Rev-erbα.
Tissues and cells
Monodelphis domestica tissues were harvested from staged animals obtained from stocks maintained at the Southwest Foundation for Biomedical Research. The opossums were maintained under standard conditions for this species . The Southwest Foundation for Biomedical Research is accredited by the Association for the Assessment and Accreditation of Animal Care, International. All procedures were approved by the Institutional Animal Care and Use Committee and conformed to the Public Health Service Policy on Humane Care and Use of Laboratory Animals. DNA was collected from animals at birth (0 days), 1, 2 and 4 weeks, 2 months (9 weeks) and 4 months (18 weeks). DNA from Didelphis virginiana was isolated from a purchased sample of brain tissue (Pel-Freez; Rogers, AR). Potorous tridactylus PtK1cells (CRL-6493) were obtained from ATCC (Manassas, VA) and passaged in Dulbeccos modified Eagles medium with 10% fetal bovine serum.
Nucleic acid sequencing and cloning
DNA was sequenced from the overlap regions spanning the TRα and Rev-erbα genes of Didelphis virginiana and Potorous tridactylus following amplification of total DNA using primers complementary to conserved sites within exon 9 of TRα1 and exons 7 and 8 of Rev-erbα. mRNA sequencing was similarly carried out with cDNA obtained by randomly primed reverse transcription of total RNA (Superscript II; Invitrogen) following homogenization of tissue in TRIzol according to manufacturer's instructions (Invitrogen). Amplified genomic products were also cloned into pGEM-T vector (Promega; Madison, WI) and subcloned into pErbA construct as previously described . PCR amplifications were carried out with Taq polymerase (Promega). All primers were from Integrated DNA Technologies (Coral, IA)
Minigene plasmid constructs are based on the rat construct, pErbAΔXE (R-R in Figure 4). Both the rat and possum constructs include all but the first 20 bp of exon 7, and all of exons 8, 9A and 10 [32, 38]. In this plasmid intron 9 sequences are deleted from an XbaI site 490 bp downstream of the 5'end of exon 9 in rat and a site 186 bp upstream of the exon 10 3'ss. Sequences of both rat and opossum minigenes extend to a unique BstEII site in Rev-erbα exon 7 that is conserved in M. domestica. The opossum minigene, pErbA(O-O) ("O-O" in Figure 4) includes 620 bp of intron 9 upstream of the deletion and 226 nts downstream following insertion of an XbaI site to facilitate swapping of rat and opossum sequences in construction of the chimeric plasmids pErbA(O-R) and pErbA(R-O) (Figure 4A). Chimeric plasmids pErbA(O/RpA) and pErbA(O/Rx10) were prepared by two-step recombinant PCR as previously described . Wild-type plasmid pCMV-ErbAm and the corresponding 5'ss point mutants, +5C/G and +6T/G, have been previously described .
The following DNA and mRNA sequences (with NCBI accession) were analyzed using Vector NTI software (Invitrogen): M. domestica NW_001581875.1, XM_001370259, XM_001362861; Ratus novegicus NW_047339, NM_001017960, NM_031134, NM_001113422; Gallus gallus NM_205313. The Macropus eugenii genomic sequence was assembled from the following files in the trace archives: ti:1070151041, ti:1437108730, ti:1378054157, ti:1582562684, ti:1623537866, ti:1484317168, ti:1467742023, ti:1070151372.
The following sequences have been deposited in the GenBank: M. domestica TRα1 mRNA (HM149330), P. tridactylus TRα1 (HM149329), P. tridactylus NR1D1 (HM149328), P. tridactylus TRα1/Rev-erbα genomic sequence (HM149331), D. virginiana TRα1/Rev-erbα genomic sequence (HM149332). The M. eugenii TRα1/Rev-erbα genomic sequence has been deposited in the NCBI Bankit database (BK007078).
Transfection assays and RNA analysis
Plasmids were transfected into HEK 293 cells using calcium phosphate as described  and into PtK1 cells with Lipofectamine 2000 (Invitrogen). Total RNA was prepared from transfected cells with TRIzol 48 h post-transfection. RNA was assayed by RNase protection assays as previously described [32, 33]. Riboprobes for rat TRα1/TRα2 exon 9 and Rev-erbα exon 5 have been previously described . Similar riboprobes were prepared for opossum Rev-erbα exon 5, TRα1 exons 9A, and TRα1 poly(A) site by in vitro transcription of PCR products cloned into pGEM-T vector (Promega) using primers described in (additional file 5). The riboprobe for opossum, unlike that for rat, extended upstream into intron 8 as well as downstream of the site homologous to exon 9A in TRα2 (Figure 4A). 3'RACE was carried out on RNA from liver and cerebral cortex from 4 month animals using gene specific primers for opossum Rev-erbα. Direct sequencing of PCR products obtained with two gene-specific primers confirmed the presence of a major poly(A) site coincident with that found in rat, human and mouse in the EST database.
Quantitative realtime RT-PCR was carried out as follows. Total RNA was randomly primed and reverse transcribed with Superscript II. For expression assays with possum tissues, 4 ug of total RNA was used per reaction and the equivalent of 0.1 μg of RNA product was used in a 25 μl reaction containing Bio-Rad SybrGreen Supermix and 150 nM primers (additional file 5 ). RNA from transfected cells was similarly treated except less RNA was used. PCR values from opossum tissues were normalized to expression of opossum β-actin with primer to exons 5 and 6. Realtime PCR assays were carried out on a MyiQ Realtime Thermocycler using IQ SyberGreen Supermix and analyzed according to the manufacturer's instructions (Bio-Rad Laboratories; Hercules, CA).
The following primers were used for amplification of chimeric gene products as shown in Figure 4C, with expected amplicon lengths (in base pairs) given in parentheses: for TRα1 in O-O, O-R, O/Rx10 and O/RpA: 1F/2R (290); for TRα1 in R-R and R-O: 15F/16R (291); (2) for TRα2 in O-O and O/RpA 1F/6R (271); for TRα2 in O-R and O/Rx10 1F/17R (222); for TRα2 in R-O 15F/6R (251); and for TRα2 in R-R 15F/17R (202). For conventional PCR shown in Figure 4B (40 cycles) the same primers for TRα1 but for TRα2 longer amplicons were used: O-O and O/RpA 20F/21R (722); O-R and O/Rx10 20F/23R (728); R-R 22F/23R (471); and R-O 22F/21R (464).
The efficiency of TRα2 primers was determined from the slope of the semilogarithmic plot of template concentration vs. CT using the equation, efficiency = [10^(-1/slope)] -1 . As a further test of primers used to probe for TRα2 expression, the "forward" TRα2 primer was paired with a "reverse" primer specific for TRα1, while the "reverse" TRα2 primers were paired with a "forward" primer specific to the 3' UTR of Rev-erbα. Each of these primer combinations efficiently amplified the expected mRNAs, supporting the conclusion that the failure to detect TRα2 mRNA does not reflect shortcomings of primer design (data not shown). Relative expression of TRα1 and TRα2 and percent splicing of TRα2 were calculated from realtime RT-PCR assuming 100% amplification efficiency. The significance of changes in expression was evaluated by Student's two-sample unpooled t test.
Carninci P, Kasukawa T, Katayama S, Gough J, Frith MC, Maeda N, Oyama R, Ravasi T, Lenhard B, Wells C, Kodzius R, Shimokawa K, Bajic VB, Brenner SE, Batalov S, Forrest AR, Zavolan M, Davis MJ, Wilming LG, Aidinis V, Allen JE, Ambesi-Impiombato A, Apweiler R, Aturaliya RN, Bailey TL, Bansal M, Baxter L, Beisel KW, Bersano T, Bono H: The transcriptional landscape of the mammalian genome. Science. 2005, 309 (5740): 1559-1563. 10.1126/science.1112014
Jacquier A: The complex eukaryotic transcriptome: unexpected pervasive transcription and novel small RNAs. Nat Rev Genet. 2009, 10 (12): 833-844. 10.1038/nrg2683
Kapranov P, Cheng J, Dike S, Nix DA, Duttagupta R, Willingham AT, Stadler PF, Hertel J, Hackermuller J, Hofacker IL, Bell I, Cheung E, Drenkow J, Dumais E, Patel S, Helt G, Ganesh M, Ghosh S, Piccolboni A, Sementchenko V, Tammana H, Gingeras TR: RNA maps reveal new RNA classes and a possible function for pervasive transcription. Science. 2007, 316 (5830): 1484-1488. 10.1126/science.1138341
Kapranov P, Willingham AT, Gingeras TR: Genome-wide transcription and the implications for genomic organization. Nat Rev Genet. 2007, 8 (6): 413-423. 10.1038/nrg2083
Katayama S, Tomaru Y, Kasukawa T, Waki K, Nakanishi M, Nakamura M, Nishida H, Yap CC, Suzuki M, Kawai J, Suzuki H, Carninci P, Hayashizaki Y, Wells C, Frith M, Ravasi T, Pang KC, Hallinan J, Mattick J, Hume DA, Lipovich L, Batalov S, Engström PG, Mizuno Y, Faghihi MA, Sandelin A, Chalk AM, Mottagui-Tabar S, Liang Z, Lenhard B: Antisense transcription in the mammalian transcriptome. Science. 2005, 309 (5740): 1564-1566. 10.1126/science.1112009
Wu JQ, Du J, Rozowsky J, Zhang Z, Urban AE, Euskirchen G, Weissman S, Gerstein M, Snyder M: Systematic analysis of transcribed loci in ENCODE regions using RACE sequencing reveals extensive transcription in the human genome. Genome Biol. 2008, 9 (1): R3- 10.1186/gb-2008-9-1-r3
He Y, Vogelstein B, Velculescu VE, Papadopoulos N, Kinzler KW: The antisense transcriptomes of human cells. Science. 2008, 322 (5909): 1855-1857. 10.1126/science.1163853
Makalowska I, Lin CF, Makalowski W: Overlapping genes in vertebrate genomes. Comput Biol Chem. 2005, 29 (1): 1-12. 10.1016/j.compbiolchem.2004.12.006
Munroe SH, Zhu J: Overlapping transcripts, double-stranded RNA and antisense regulation: a genomic perspective. Cell Mol Life Sci. 2006, 63 (18): 2102-2118. 10.1007/s00018-006-6070-2
Beiter T, Reich E, Williams RW, Simon P: Antisense transcription: a critical look in both directions. Cell Mol Life Sci. 2009, 66 (1): 94-112. 10.1007/s00018-008-8381-y
Farnebo M: Wrap53, a novel regulator of p53. Cell Cycle. 2009, 8 (15): 2343-2346.
Houseley J, Rubbi L, Grunstein M, Tollervey D, Vogelauer M: A ncRNA modulates histone modification and mRNA induction in the yeast GAL gene cluster. Mol Cell. 2008, 32 (5): 685-695. 10.1016/j.molcel.2008.09.027
Mahmoudi S, Henriksson S, Corcoran M, Mendez-Vidal C, Wiman KG, Farnebo M: Wrap53, a natural p53 antisense transcript required for p53 induction upon DNA damage. Mol Cell. 2009, 33 (4): 462-471. 10.1016/j.molcel.2009.01.028
Lavorgna G, Dahary D, Lehner B, Sorek R, Sanderson CM, Casari G: In search of antisense. Trends Biochem Sci. 2004, 29 (2): 88-94. 10.1016/j.tibs.2003.12.002
Lazar MA, Hodin RA, Darling DS, Chin WW: A novel member of the thyroid/steroid hormone receptor family is encoded by the opposite strand of the rat c-erbA alpha transcriptional unit. Mol Cell Biol. 1989, 9 (3): 1128-1136.
Miyajima N, Horiuchi R, Shibuya Y, Fukushige S, Matsubara K, Toyoshima K, Yamamoto T: Two erbA homologs encoding proteins with different T3 binding capacities are transcribed from opposite DNA strands of the same genetic locus. Cell. 1989, 57 (1): 31-39. 10.1016/0092-8674(89)90169-4
Lazar MA: Thyroid hormone receptors: multiple forms, multiple possibilities. Endocr Rev. 1993, 14 (2): 184-193.
Forrest D, Sjoberg M, Vennstrom B: Contrasting developmental and tissue-specific expression of alpha and beta thyroid hormone receptor genes. EMBO J. 1990, 9 (5): 1519-1528.
Zhang J, Lazar MA: The mechanism of action of thyroid hormones. Annu Rev Physiol. 2000, 62: 439-466. 10.1146/annurev.physiol.62.1.439
Bookout AL, Jeong Y, Downes M, Yu RT, Evans RM, Mangelsdorf DJ: Anatomical profiling of nuclear receptor expression reveals a hierarchical transcriptional network. Cell. 2006, 126 (4): 789-799. 10.1016/j.cell.2006.06.049
Yang X, Downes M, Yu RT, Bookout AL, He W, Straume M, Mangelsdorf DJ, Evans RM: Nuclear receptor expression links the circadian clock to metabolism. Cell. 2006, 126 (4): 801-810. 10.1016/j.cell.2006.06.050
Preitner N, Damiola F, Lopez-Molina L, Zakany J, Duboule D, Albrecht U, Schibler U: The orphan nuclear receptor REV-ERBalpha controls circadian transcription within the positive limb of the mammalian circadian oscillator. Cell. 2002, 110 (2): 251-260. 10.1016/S0092-8674(02)00825-5
Pardee KI, Xu X, Reinking J, Schuetz A, Dong A, Liu S, Zhang R, Tiefenbach J, Lajoie G, Plotnikov AN, Botchkarev A, Krause HM, Edwards A: The structural basis of gas-responsive transcription by the human nuclear hormone receptor REV-ERBbeta. PLoS Biol. 2009, 7 (2): e43- 10.1371/journal.pbio.1000043
Raghuram S, Stayrook KR, Huang P, Rogers PM, Nosie AK, McClure DB, Burris LL, Khorasanizadeh S, Burris TP, Rastinejad F: Identification of heme as the ligand for the orphan nuclear receptors REV-ERBalpha and REV-ERBbeta. Nat Struct Mol Biol. 2007, 14 (12): 1207-1213. 10.1038/nsmb1344
Rogers PM, Ying L, Burris TP: Relationship between circadian oscillations of Rev-erbalpha expression and intracellular levels of its ligand, heme. Biochem Biophys Res Commun. 2008, 368 (4): 955-958. 10.1016/j.bbrc.2008.02.031
Yin L, Wu N, Curtin JC, Qatanani M, Szwergold NR, Reid RA, Waitt GM, Parks DJ, Pearce KH, Wisely GB, Lazar MA: Rev-erbalpha, a heme sensor that coordinates metabolic and circadian pathways. Science. 2007, 318 (5857): 1786-1789. 10.1126/science.1150179
Wu N, Yin L, Hanniman EA, Joshi S, Lazar MA: Negative feedback maintenance of heme homeostasis by its receptor, Rev-erbalpha. Genes Dev. 2009, 23 (18): 2201-2209. 10.1101/gad.1825809
Yin L, Wu N, Lazar MA: Nuclear receptor Rev-erbaα: a heme receptor that coordinates circadian rhythm and metabolism. Nucl Recept Signal. 8: e001 10.1101/gad.1825809
Alenghat T, Meyers K, Mullican SE, Leitner K, Adeniji-Adele A, Avila J, Bucan M, Ahima RS, Kaestner KH, Lazar MA: Nuclear receptor corepressor and histone deacetylase 3 govern circadian metabolic physiology. Nature. 2008, 456 (7224): 997-1000. 10.1038/nature07541
Estall JL, Ruas JL, Choi CS, Laznik D, Badman M, Maratos-Flier E, Shulman GI, Spiegelman BM: PGC-1α negatively regulates hepatic FGF21 expression by modulating the heme/Rev-Erbα axis. Proc Natl Acad Sci USA. 2009, 106 (52): 22510-22515. 10.1073/pnas.0912533106
Burris TP: Nuclear hormone receptors for heme: REV-ERBα and REV-ERBβ are ligand-regulated components of the mammalian clock. Mol Endocrinol. 2008, 22 (7): 1509-1520. 10.1210/me.2007-0519
Hastings ML, Ingle HA, Lazar MA, Munroe SH: Post-transcriptional regulation of thyroid hormone receptor expression by cis-acting sequences and a naturally occurring antisense RNA. J Biol Chem. 2000, 275 (15): 11507-11513. 10.1074/jbc.275.15.11507
Hastings ML, Milcarek C, Martincic K, Peterson ML, Munroe SH: Expression of the thyroid hormone receptor gene, erbAalpha, in B lymphocytes: alternative mRNA processing is independent of differentiation but correlates with antisense RNA levels. Nucleic Acids Res. 1997, 25 (21): 4296-4300. 10.1093/nar/25.21.4296
Lazar MA, Hodin RA, Cardona G, Chin WW: Gene expression from the c-erbA α/Rev-ErbAα genomic locus. Potential regulation of alternative splicing by opposite strand transcription. J Biol Chem. 1990, 265 (22): 12859-12863.
Munroe SH, Lazar MA: Inhibition of c-erbA mRNA splicing by a naturally occurring antisense RNA. J Biol Chem. 1991, 266 (33): 22083-22086.
Mikkelsen TS, Wakefield MJ, Aken B, Amemiya CT, Chang JL, Duke S, Garber M, Gentles AJ, Goodstadt L, Heger A, Jurka J, Kamal M, Mauceli E, Searle SM, Sharpe T, Baker ML, Batzer MA, Benos PV, Belov K, Clamp M, Cook A, Cuff J, Das R, Davidow L, Deakin JE, Fazzari MJ, Glass JL, Grabherr M, Greally JM, Gu W: Genome of the marsupial Monodelphis domestica reveals innovation in non-coding sequences. Nature. 2007, 447 (7141): 167-177. 10.1038/nature05805
Samollow PB: The opossum genome: insights and opportunities from an alternative mammal. Genome Res. 2008, 18 (8): 1199-1215. 10.1101/gr.065326.107
Salato VK, Rediske NW, Zhang C, Hastings ML, Munroe SH: An exonic splicing enhancer within a bidirectional coding sequence regulates splicing of an antisense mRNA. RNA Biology. 2010, 7 (2): 179-190. 10.4161/rna.7.2.11182
Wong ML, Medrano JF: Real-time PCR for mRNA quantitation. Biotechniques. 2005, 39 (1): 75-85. 10.2144/05391RV01
Hastings ML, Wilson CM, Munroe SH: A purine-rich intronic element enhances alternative splicing of thyroid hormone receptor mRNA. RNA. 2001, 7 (6): 859-874. 10.1017/S1355838201002084
Mitsuhashi T, Nikodem VM: Regulation of expression of the alternative mRNAs of the rat alpha-thyroid hormone receptor gene. J Biol Chem. 1989, 264 (15): 8900-8904.
Keijzer R, Blommaart PJ, Labruyere WT, Vermeulen JL, Doulabi BZ, Bakker O, Tibboel D, Lamers WH: Expression of thyroid hormone receptors A and B in developing rat tissues; evidence for extensive posttranscriptional regulation. J Mol Endocrinol. 2007, 38 (5): 523-535. 10.1677/jme.1.02125
Strait KA, Schwartz HL, Perez-Castillo A, Oppenheimer JH: Relationship of c-erbA mRNA content to tissue triiodothyronine nuclear binding capacity and function in developing and adult rats. J Biol Chem. 1990, 265 (18): 10514-10521.
Jannini EA, Mitsuhashi T, Nikodem VM: Developmental expression of mRNAs from a rat C-erbA genomic locus. Biochem Biophys Res Commun. 1992, 184 (2): 739-745. 10.1016/0006-291X(92)90652-2
Katz D, Reginato MJ, Lazar MA: Functional regulation of thyroid hormone receptor variant TRα2 by phosphorylation. Mol Cell Biol. 1995, 15 (5): 2341-2348.
Xu B, Koenig RJ: An RNA-binding domain in the thyroid hormone receptor enhances transcriptional activation. J Biol Chem. 2004, 279 (32): 33051-33056. 10.1074/jbc.M404930200
Burge CB, Tuschl T, Sharp PA: Splicing of precursors to mRNAs by the spliceosomes. The RNA World. Edited by: Gesteland RF, Cech TR, Atkins JF. 1999, 525-560. Cold Spring Harbor, NY: Cold Spring Harbor Press, second
Sheth N, Roca X, Hastings ML, Roeder T, Krainer AR, Sachidanandam R: Comprehensive splice-site analysis using comparative genomics. Nucleic Acids Res. 2006, 34 (14): 3955-3967. 10.1093/nar/gkl556
Xing Y, Lee C: Evidence of functional selection pressure for alternative splicing events that accelerate evolution of protein subsequences. Proc Natl Acad Sci USA. 2005, 102 (38): 13526-13531. 10.1073/pnas.0501213102
Modrek B, Lee CJ: Alternative splicing in the human, mouse and rat genomes is associated with an increased frequency of exon creation and/or loss. Nat Genet. 2003, 34 (2): 177-180. 10.1038/ng1159
Nurtdinov RN, Mironov AA, Gelfand MS: Rodent-specific alternative exons are more frequent in rapidly evolving genes and in paralogs. BMC Evol Biol. 2009, 9: 142- 10.1186/1471-2148-9-142
Zhang XH, Chasin LA: Comparison of multiple vertebrate genomes reveals the birth and evolution of human exons. Proc Natl Acad Sci USA. 2006, 103 (36): 13427-13432. 10.1073/pnas.0603042103
Xing Y, Lee C: Alternative splicing and RNA selection pressure--evolutionary consequences for eukaryotic genomes. Nat Rev Genet. 2006, 7 (7): 499-509. 10.1038/nrg1896
Bertrand S, Brunet FG, Escriva H, Parmentier G, Laudet V, Robinson-Rechavi M: Evolutionary genomics of nuclear receptors: from twenty-five ancestral genes to derived endocrine systems. Mol Biol Evol. 2004, 21 (10): 1923-1937. 10.1093/molbev/msh200
Paris M, Brunet F, Markov GV, Schubert M, Laudet V: The amphioxus genome enlightens the evolution of the thyroid hormone signaling pathway. Dev Genes Evol. 2008, 218 (11-12): 667-680. 10.1007/s00427-008-0255-7
Paris M, Escriva H, Schubert M, Brunet F, Brtko J, Ciesielski F, Roecklin D, Vivat-Hannah V, Jamin EL, Cravedi JP, Scanlan TS, Renaud JP, Holland ND, Laudet V: Amphioxus postembryonic development reveals the homology of chordate metamorphosis. Curr Biol. 2008, 18 (11): 825-830. 10.1016/j.cub.2008.04.078
Wu W, Niles EG, LoVerde PT: Thyroid hormone receptor orthologues from invertebrate species with emphasis on Schistosoma mansoni. BMC Evol Biol. 2007, 7: 150- 10.1186/1471-2148-7-150
Brosius J, Gould SJ: On "genomenclature": a comprehensive (and respectful) taxonomy for pseudogenes and other "junk DNA". Proc Natl Acad Sci USA. 1992, 89 (22): 10706-10710. 10.1073/pnas.89.22.10706
Makalowska I, Lin CF, Hernandez K: Birth and death of gene overlaps in vertebrates. BMC Evol Biol. 2007, 7: 193- 10.1186/1471-2148-7-193
Bartel DP: MicroRNAs: target recognition and regulatory functions. Cell. 2009, 136 (2): 215-233. 10.1016/j.cell.2009.01.002
Moore MJ, Proudfoot NJ: Pre-mRNA processing reaches back to transcription and ahead to translation. Cell. 2009, 136 (4): 688-700. 10.1016/j.cell.2009.02.001
Licatalosi DD, Darnell RB: RNA processing and its regulation: global insights into biological networks. Nat Rev Genet. 11 (1): 75-87. 10.1016/j.cell.2009.02.001
Izumo S, Mahdavi V: Thyroid hormone receptor alpha isoforms generated by alternative splicing differentially activate myosin HC gene transcription. Nature. 1988, 334 (6182): 539-542. 10.1038/334539a0
Mitsuhashi T, Tennyson GE, Nikodem VM: Alternative splicing generates messages encoding rat c-erbA proteins that do not bind thyroid hormone. Proc Natl Acad Sci USA. 1988, 85 (16): 5804-5808. 10.1073/pnas.85.16.5804
Flamant F, Samarut J: Thyroid hormone receptors: lessons from knockout and knock-in mutant mice. Trends Endocrinol Metab. 2003, 14 (2): 85-90. 10.1016/S1043-2760(02)00043-7
Tagami T, Kopp P, Johnson W, Arseven OK, Jameson JL: The thyroid hormone receptor variant alpha2 is a weak antagonist because it is deficient in interactions with nuclear receptor corepressors. Endocrinology. 1998, 139 (5): 2535-2544. 10.1210/en.139.5.2535
Reginato MJ, Zhang J, Lazar MA: DNA-independent and DNA-dependent mechanisms regulate the differential heterodimerization of the isoforms of the thyroid hormone receptor with retinoid × receptor. J Biol Chem. 1996, 271 (45): 28199-28205. 10.1074/jbc.271.45.28199
Yang YZ, Burgos-Trinidad M, Wu Y, Koenig RJ: Thyroid hormone receptor variant alpha2. Role of the ninth heptad in dna binding, heterodimerization with retinoid × receptors, and dominant negative activity. J Biol Chem. 1996, 271 (45): 28235-28242. 10.1074/jbc.271.45.28235
Xu B, Koenig RJ: Regulation of thyroid hormone receptor alpha2 RNA binding and subcellular localization by phosphorylation. Mol Cell Endocrinol. 2005, 245 (1-2): 147-157. 10.1016/j.mce.2005.11.010
Colley SM, Leedman PJ: SRA and its binding partners: an expanding role for RNA-binding coregulators in nuclear receptor-mediated gene regulation. Crit Rev Biochem Mol Biol. 2009, 44 (1): 25-33. 10.1080/10409230802661719
Hodin RA, Meng S, Chamberlain SM: Thyroid hormone responsiveness is developmentally regulated in the rat small intestine: a possible role for the alpha-2 receptor variant. Endocrinology. 1994, 135 (2): 564-568. 10.1210/en.135.2.564
Wills KN, Zhang XK, Pfahl M: Coordinate expression of functionally distinct thyroid hormone receptor alpha isoforms during neonatal brain development. Mol Endocrinol. 1991, 5 (8): 1109-1119. 10.1210/mend-5-8-1109
Gatfield D, Le Martelot G, Vejnar CE, Gerlach D, Schaad O, Fleury-Olela F, Ruskeepaa AL, Oresic M, Esau CC, Zdobnov EM, Schibler U: Integration of microRNA miR-122 in hepatic circadian gene expression. Genes Dev. 2009, 23 (11): 1313-1326. 10.1101/gad.1781009
Le Martelot G, Claudel T, Gatfield D, Schaad O, Kornmann B, Sasso GL, Moschetta A, Schibler U: REV-ERBalpha participates in circadian SREBP signaling and bile acid homeostasis. PLoS Biol. 2009, 7 (9): e1000181- 10.1371/journal.pbio.1000181
Wang J, Lazar MA: Bifunctional role of Rev-erbalpha in adipocyte differentiation. Mol Cell Biol. 2008, 28 (7): 2213-2220. 10.1128/MCB.01608-07
VandeBerg JL, Williams-Blangero S: The laboratory opossum. The UFAW Handbook on the Care and Use of Laboratory and Other Research Animals. Edited by: Hulbrecht R, Kirkwood J. 2010, Chapter 19: 246-261. full_text. West Sussex: Wiley-Blackwell, 8
We would like to thank Janice MacRossin for assistance with collection of opossum tissues, Tessa Hoff for assistance with RNA sequencing and analysis, and David Durica (University of Oklahoma), Carol Munroe (Boston University) and Mitch Lazar (University of Pennsylvania) for their comments and advice on the manuscript. We are grateful for the support of National Institutes of Health (Grant R15 DK075418 to SHM) and the Robert J. Kleberg, Jr. and Helen C. Kleberg Foundation for support of the Monodelphis domestica breeding colony at the Southwest Foundation for Biomedical Research.
SHM and BCR initiated this study in consultation with JLV; BCR and MSB performed the experiments and SHM, MSB and BCR wrote the manuscript. All authors approved the final manuscript.
Brandon C Rindfleisch, M Scott Brown contributed equally to this work.