Platypus TERT and its position in the evolution of vertebrate TERT genes
Analysis of the platypus TERT gene creates an important framework for the evolutionary comparison of TERT genes of eutherian mammals and sauropsid reptiles (including birds). Platypus, one of a few surviving species of monotremes, has a unique evolutionary position. Monotremes, followed by marsupials, first split from all other contemporary mammals after the divergence of the synapsid lineage leading to mammals and the sauropsid lineage leading to contemporary reptiles and birds . Monotremes and marsupials, therefore, might retain some features of the pre-mammalian synapsid TERT gene that were lost in eutherian mammals. The higher levels of sequence similarity of four principal TERT domain of OanTERT to the marsupial, and avian TERT than to the eutherian TERT support this hypothesis. Moreover, platypus and marsupials also retain three extended linker regions of a non-conserved sequence in the N-terminal region of the TERT protein which are about twice longer than the linkers in eutherians and their size matches more closely the linkers in sauropsids.
Interestingly, the long and intermediate linker regions present in most of the tetrapods except the eutherian mammals were not found in the well-studied group of ray-finned fish but are present in elephant shark TERT. Elephant shark belongs to the oldest evolutionary group of contemporary jawed vertebrates . It is, therefore, possible that the long TERT of the shark may represent the original jawed-vertebrate TERT structure and that the linkers were shortened independently in ray-finned fish and eutherian mammals. Divergent evolution of fish TERT is supported by the evolutionary tree built mostly on the conserved TERT sequences outside the linker regions that situates the elephant shark TERT closer to the tetrapod TERT and farther from ray-finned fish TERT (Figure 1d). These changes could be, at least partly, a consequence of genome duplication and increase in mutation rate in the teleost branch of ray-finned fish [72–75]. The long TERT structure described in sea urchin suggests an even deeper evolutionary root of the extended linker .
The long-term evolutionary stability of the extended linker regions suggests a possible function that constrains evolutionary change. The invertebrate sea urchin has two TERTs with extremely long linker regions and a specific function in embryogenesis were suggested for these structures . In vertebrate the presence of the longer linker region in TERT proteins of different species correlates with evolutionary distribution of interstitial telomeric sequences (ITS). ITS were detected in the platypus genome and while these repeats were described in most vertebrate species they occur at a higher frequency in frogs and birds but not eutherian mammals [64, 65, 67]. Several mechanisms contribute to the formation of ITS depending on their length and sequence organization [76, 77]. For the formation of most of the short ITS the direct involvement of telomerase during the repair of double-stranded breaks has been implicated. Therefore, one of possible functions of the extended linker region of the TERT protein might be to increase the recruitment of telomerase to sites of double-strand breaks. The precise function of the extended linker of TERT however remains to be established.
The shortening of the once extended TERT molecules could be also restricted by the necessity to adapt to interacting molecules. One of such molecule could be the telomerase RNA (TR) because the three TERT linkers are either inside or close to the TR-binding domain. The minimalist size of the TR that developed in ray-finned fish correlates with an extensive reduction of the linkers in the fish TERT protein . Further, the largest vertebrate TR has been described in sharks and the predicted size of the elephant shark TERT is also the largest of all vertebrates . However, the TERT-TR size correlation does not hold in birds. Birds encode a large TERT protein but the TR is of a similar size as that of mammals .
The evolution of AS variants
Alternative splicing plays an important role in creating proteome diversity in multicellular animals . More than 95% of human intron-containing genes express one or several alternatively spliced variants. While the alternative splicing of TERT transcripts has not been described in protists, it occurs extensively in multicellular organisms such as plants and Metazoa (for recent review see ). All vertebrate TERT transcripts are currently believed to be alternatively spliced, however, there are great differences in frequency and structure of AS variants . The largest number of AS TERT variants have been identified in chicken and human cells, 37 and 21, respectively ([58, 81–85] and unpublished data). In fish, amphibians and rodents only a few variants are known . In contrast to plants, almost all the AS TERT variants differ in structure between mammals and birds. There is only one known example of AS variants which is structurally identical, an in-frame deletion of exon 10 in the chicken and rat [58, 86]. We identified seven AS variants of OanTERT which likely represent all the major platypus TERT AS variants. The variant B is identical to chicken variant V3 and surprisingly, the structure of three other variants (A, A3, C), also resembles the chicken AS isoforms. This unprecedented conservation of AS variants between a basal mammal lineage and a bird suggests an important function for alternative splicing in TERT regulation. It also suggests that the repertoire of AS TERT variants identified in eutherian mammals evolved recently.
Two AS sites evolved in the second exon of OanTERT which yield four AS variants, two of which retain an ORF. In contrast to variants containing PTCs, which are targeted by nonsense mediated decay (NMD), these AS variants will likely produce TERT proteins . The A variant, which maintains an original ORF, is highly expressed in liver, lung, ovary, and testes, like the corresponding chicken variant A. Interestingly, the chicken and platypus A variants, as well as AS variant A4, have maintained the TEN domain which plays a role in DNA-binding, however, based on their structure, they would not possess telomerase activity [88, 89]. Several telomerase activity-independent functions have been attributed to the TERT protein including the enhancement of cell proliferation. The alternatively spliced protein-coding TERT isoforms might provide some of these functions [6, 7].
The tissue-specific expression of TERT
Telomerase activity is downregulated during development in most somatic cells simultaneously with a reduction in their rate of proliferation and subsequent differentiation . This process is accompanied by the transcriptional downregulation of TERT. However, the levels of repression differ between cold-blooded (fish and amphibian) and warm-blooded species (mammals and birds). In adult organs of many mammals and birds, the expression of TERT is repressed, with very low or undetectable levels expressed in heart, skeletal muscle and brain [11, 41, 62, 90]. In contrast, the expression of TERT in all adult organs of fish remains at high levels [40, 42, 48]. Likewise, telomerase activity is widespread in the tissues of the toad Xenopus tropicalis including heart, brain, and muscle . However, there are significant differences even among mammals [63, 92]. Most mammals with telomeres longer than 20 kb, including small rodents, bats, some felines, lagomorphs, elephant shrews, and the Virginia opossum, express significant levels of telomerase in adult somatic cells. The platypus, with telomeres longer than 20 kb and telomerase activity in all adult tissues, belongs to these species. Telomerase activity in the platypus was even detected in skeletal muscle, brain, and heart, though in lower levels than in other organs with higher cell proliferation rates. The pattern of telomerase expression in platypus, as well as the length of its telomeres, correlates well with its relatively small size (1–2 kg), supporting the hypothesis that telomerase activity and telomere length inversely correlate with body mass in mammals .
The location of OanTERT on the X3/Y2 sex chromosomes
The chromosomal position of platypus TERT is, to our knowledge, the first instance among amniote vertebrates to localize TERT on sex chromosomes. The homology of platypus and echidna Y2/X3 chromosomes suggests that this is common in monotremes . TERT genes of the other amniotes analyzed to date are located on autosomes [43, 93]. Platypus sex chromosomes have large regions orthologous to avian ZW chromosomes while they generally lack orthologs of genes located at therian XY sex chromosomes [53–55]. Interestingly, the region on human autosome 5 containing the TERT gene is located in proximity to a region homologous to a part of the avian sex chromosome Z . Further, the inspection of the Ensembl database shows that similar juxtaposition is also found in other mammals, including macaque, marmoset, mouse, cow, pig, horse, and opossum. These observations lead to an interesting possibility that the last common ancestor of all mammals might have the TERT gene chromosomal region associated with an avian-type Z chromosome. The TERT position on sex chromosomes may not be an evolutionary innovation of monotremes, but one of the features of mammalian ancestors preserved in the platypus genome and only lost in therian mammals as a result of the evolution of the new XY sex chromosomes .
Comparison of the genes surrounding the TERT locus in platypus, human, and chicken (Figure 5c and Additional file 6: Table S2) suggests that the formation of a discontinuous sex chromosome chain in platypus lead to the relocation of some genes to different chromosomes. The genes found in one region in human chromosome 5 and chicken chromosome 2 are divided into three X chromosomes (X1, X2, and X3) in platypus. Similarly, chicken Z-homologous genes are located on four platypus X chromosomes (X1, X2, X3, and X5) . This suggests that the evolution of the sex chromosome chain might have created instability and possibly increased the rate of interchromosomal rearrangements of the chromosomes involved [95, 96].
The OanTERT locus is located in a pseudoautosomal region of the platypus sex chromosomes. This location, however, does not automatically guarantee identical regulation of the X and Y copies of TERT. Different spacing of two BAC clones (165 F5, 151 O20) hybridizing spots at Y2q and X3p might indicate different chromatin folding and epigenetic state of these two chromosome arms or alternatively incomplete homology . These differences may lead to differential expression of the two TERT copies.
Gender-dependent telomerase expression
Gender-specific differences between telomere length in several mammalian species are well established. Female mice and rats have significantly longer telomeres than age-matched males [71, 97]. In human, female telomeres are longer than male, with the exception of the Amish people [98–100]. These differences in telomere length positively correlate with life expectancy [101, 102]. Telomere length is the result of a complex regulation involving both the synthesis and attrition of telomeric repeats [103, 104]. Telomerase activity is responsible for telomere synthesis and generally correlates with telomere length. Therefore, telomerase activity likely contributes to gender-specific differences in telomere length and, in turn, life expectancy. However, very limited information is available concerning gender-specific telomerase activity in mammals. Cardiac myocytes obtained from female rats express significantly higher levels of telomerase than male cells . Our results demonstrating that female platypuses express higher telomerase activity than males in at least two tissues, liver and heart, extend this observation. Telomerase activity in these tissues may also correlate with lifespan. While this issue will require further study, there are indications that the lifespan of the female platypus in the wild is greater than that of the male. Studies performed over a 30 year period in the upper Shoalhaven River demonstrated that the recapture rate is significantly biased toward females . In addition, the oldest female captured in these studies was 21 years old, while the oldest male was 7.
Recently, it has been shown that there are gender differences in telomere length in the kakapo, Strigops habroptilus, a parrot endemic to New Zealand . The male kakapos have longer telomeres than females which is in distinct contrast to mammals. Previously, it has been suggested that estrogen in mammals is at least partly responsible for the increased telomere length in females since estrogen is a positive regulator of TERT transcription . Female birds, however, also express high levels of estrogen like their mammalian counterparts, therefore, this mechanism cannot apply to birds [108, 109]. Moreover, in contrast to human, where paternal inheritance of telomere length to daughters and sons have been described, the kakapo telomere length is inherited maternally to sons [110, 111]. To reconciliate these opposing patterns of regulation and inheritance of telomere length it has been proposed that heterogametic sex (either XY in mammals or ZW in birds) chromosomes X and Z have a central role in determination of gender-specific differences in telomere length.
The chromosomal localization of TERT on platypus sex chromosomes together with gender-specific expression supports this hypothesis and suggests that gender-specific differences in telomere length are related to the original localization of the TERT gene on sex chromosomes. One possible scenario is that the TERT activator was located on the sex chromosome of a common ancestor of mammals and birds. This localization might be retained in monotremes with the hypothetical TERT activator located on a X specific part of the X chromosome which in platypus may be only partially dosage-compensated . The large gender-specific differences in telomerase expression as well as the lifespan of platypuses may be the result of concomitant localization of TERT and its activator on sex chromosomes. A large part of the platypus sex chromosomes is orthologous to sex chromosome Z in birds, which also does not involved a chromosome-wide dosage compensation mechanism [54, 113, 114]. The localization of a TERT activator on the avian Z chromosome could result in higher levels of telomerase activity in male than female birds because the expression of genes localized on bird sex chromosomes is also weakly dosage-compensated. Finally, therian mammals developed a new sex chromosome system and evolved additional levels of regulation beyond the already existing regulatory circuits . In case of TERT gender-specific regulation this relationship could remain conserved in therian mammals while subjected to a level of additional control of the new evolved X chromosome.