The male short-beaked echidna has 63 chromosomes and the female 64 5. We characterized and classified the male karyotype by flow cytometric analysis, flow sorting, and chromosome painting. The chromosomes produced 36 peaks in the flow karyotype (Figure 1); 17 peaks represent single chromosome pairs and 4 peaks represent 2 chromosome pairs each. The homologues of chromosomes 1, 6, 16, and 27 (which are frequently heteromorphic) each sort in two different peaks. The nine remaining peaks represent single chromosomes, which we show to be the nine unpaired sex-chromosomes that constitute the meiotic chain in the male echidna. Thus, the male echidna has 27 autosome pairs and 9 sex chromosomes.

The chromosome paints were used to identify autosome pairs and sex chromosomes in male and female echidnas. Figure 2 shows a G-banded karyotype of male (upper) and female (lower) echidna chromosomes arranged in size and identified by the chromosome painting results (see below). The upper part of both the male and female karyotypes show 27 pairs of autosomes that should form bivalents at meiosis, and the lower parts show the nine unpaired sex-chromosomes in male and five paired sex-chromosomes in female. The paired autosomes can be divided into a group that contains submetacentric chromosomes (chromosomes 1 to 8) and a group of smaller metacentric and submetacentric chromosomes. Chromosomes 3, 6 and X₅ contain nucleolus organizer regions (Figure 2) determined by Ag-NOR (nucleolar organizing region) staining and by FISH using a probe specific for 28S rDNA (Figure 3).

Special attention was given to chromosome pair 27, as it shows complete homology with platypus X₄ and partial homology with platypus Y₃ and Y₄ (see below), and might, therefore, be an unrecognized sex chromosome. However, both the paints, produced from the two peaks representing the heteromorphic chromosome pair 27, painted both chromosomes 27 totally in both male and female metaphases (Figure 4a). Based on this analysis, the pair was defined as an autosome pair and this was confirmed by analysis of meiotic preparations, which revealed that 27 is not part of the chain.

Chromosome painting in echidna reveals non-specific signals that are present on more than one chromosome pair, indicated by coloured bars next to chromosomes in Figure 2. For instance, when the paint specific for chromosome 1 was hybridized to metaphases, stronger signals were visible on the region indicated on chromosome 1 and a region on chromosome 2. A hybridized chromosome 4 paint gave strong signals on chromosomes 4, 7, and 16 (and others as indicated), weaker signals on X₁ and Y₁, and even weaker signals on chromosomes 1 and 2.

The chromosome specificity of these regions was apparent using paints produced by microdissection of these regions. Hybridization to these regions (which hamper identification of chromosomes, and especially pairing regions of the sex chromosomes) was not blocked by pre-hybridization with echidna Cot-1 DNA. To facilitate chromosome identification, an image enhancement procedure was developed to remove these non-specific signals from the image 19. Molecular characterization of these regions is not considered in this paper.

Cross-species chromosome painting was used to define chromosome regions conserved between Oan and Tac and to identify rearrangements that differentiate the karyotypes of the two species. Figures 2a and 6 show the Oan and Tac homology maps, with the homologous regions indicated on the right of each chromosome.

In order to test the hypothesis that the bird Z chromosome is represented in the monotreme sex chromosome chain, platypus homologues of nineteen chicken-Z genes (together with chicken 2, 3, and 13 genes) were mapped to platypus chromosomes by two independent methods, PCR amplification of chromosome specific DNA and FISH localization of platypus bacterial artificial chromosome (BACs). Gene mapping results are summarized in Figures 11 and 12, and see Additional data file 1. This file also shows platypus contigs that contain the mapped and predicted genes extending the regions of homology.

Gene loci in chicken are designated in the following section according to their established conserved synteny with human chromosomes. For instance, chicken Z is homologous with regions of human chromosomes 5, 9, 8 and 18 and chicken chromosome 2, 3 and 13 also share homology with human chromosome 5 9. Figure 12 presents a standard idiogram, based on G-banding and chromosome painting, on which the present and previous 101122232425 gene assignments are shown. Consideration of the contigs in Ensembl to which these genes belong greatly expands the size of the syntenic regions (Additional data file 1). Most of the conserved syntenies that we have observed between platypus chromosomes and chicken Z are in five groups.

The first group includes platypus homologues of chicken Z and human 9 (Gallus gallus (GGA)-Z/Homo sapiens (HAS)-9) genes. Three genes mapped to platypus X₅, one mapped to X₂ and two mapped to X₃ (Figure 12). Consideration of the contigs to which these genes belong greatly expands the size of the homologous regions (Additional data file 1).

The second group includes platypus homologues of chicken Z and human 5 (GGA-Z/HSA-5) genes. Ten genes are distributed over platypus chromosomes 1, 2, 3 and the sex chromosomes. Only one GGA-Z/HSA-5 gene (PDE6A) mapped to platypus X₁ by both PCR and BAC-clone mapping. This gene mapped on the pairing region of X₁p and Y₁q, distal to the centromere on each (Figure 10b,c), and is contained in platypus contig 269 with nine other genes, four of which are homologous to GGA-13/HSA-5q (Additional data file 1). Two platypus genes (LMNB1 and DMXL1) homologous to chicken Z and human 5 (GGA-Z/HSA-5) mapped to platypus X₅ by both PCR and BAC-clone mapping. The homologue of the chicken gene LMNB1 PCR-mapped also to X₁ but not to Y₁, suggesting that the PCR product of X₁ may represent a section of the X₅ gene that has a copy (paralogue) on X₁. Likewise, DMXL1 was PCR-mapped also to Oan 10 as well as X₁ (but not Y₁), again indicating that a section of the DMXL1-X₅ gene has a copy (paralogue) on both Oan 10 and X₁. The seven other GGA-Z/HSA 5 genes map to Oan 1, 2 and 3 by PCR or BAC-FISH.

The third group includes platypus homologues of chicken 13 and human 5 (GGA-13/HSA-5), and chicken 3 and human 2 (GGA-3/HSA-2) genes. Five GGA-13/HSA-5 genes and one GGA-3/HSA-2 gene mapped to the pairing regions of X₁ and Y₁ (Figures 11 and 12). PANK3 also PCR-mapped to Oan 8, which might be a paralogue. Thus, the three GGA-Z/HSA-5 homologues on X₁ are accompanied by GGA-13/HSA-5 genes and one GGA-3/HSA-2 gene. The majority of genes in the respective contigs (10, 269, 847, 127) are GGA-13/HSA-5 genes, one is a GGA-3/HSA-8 gene and three are GGA-3/HSA-2 genes.

The fourth group includes platypus homologues of chicken 2 and human 5 (GGA-2/HSA-5), and chicken 2 and human 18 (GGA-2/HSA-18) genes. Three GGA-2/HSA-5 genes mapped to Oan Y₂ and X₃. These three genes are accompanied by the GGA-2/HSA-18 gene P15RS. This gene is in contig 29, which contains 7 GGA-2/HSA-5 genes, 3 GGA-2/HSA-9 genes and 3 GGA-2/HSA-18 genes.

The fifth group includes platypus homologues of chicken Z and human 8 (GGA-Z/HSA-8), and chicken Z and human 18 (GGA-Z/HSA-18) genes. Two GGA-Z/HSA-8 genes mapped to platypus chromosome 5, and one GGA-Z/HSA-18 gene mapped to platypus chromosome 3.

Thus, mapping platypus homologues of chicken Z genes revealed homology not only with platypus X₅ but also with platypus X₁, Y₁, X₂ and X₃. Other genes from human chromosomes 5 and 9, which are not homologous to the chicken Z, also mapped to platypus sex chromosomes. A set of GGA-Z genes homologous to HSA-5, 8 or 18 (see above), but so far not HSA-9, mapped to platypus autosomes. So far, no chicken Z genes have been mapped to platypus X₄, but this may not be surprising as X₄ is homologous to an echidna autosome.

A selection of these genes was PCR-mapped to echidna chromosome-specific DNA using the same primers. GPR98 (Oan 1) mapped to Tac 1, GRHL1 (Oan X1, Y1) mapped to Tac X₁ but not Y₁, RAPGEF6 (Oan X₁, Y₁) mapped to Tac X₁ and Y₁. MLLT3 (Oan X₂) mapped to Tac X₂, TRIO (Oan Y₂, X₃) mapped to Tac Y₂ and X₃, TSCOT (Oan X₃) mapped to Tac X₃, and LMNB1 (Oan X₁, X₅) mapped to Tac X₄. DMRT1 (Oan X₅) mapped to Tac X₄ by BAC-FISH (Figure 10a). These results confirm homologies between the platypus and echidna chromosomes determined by chromosome painting.

Ten chromosomes are completely conserved between echidna and platypus. The non-conserved chromosomes differ between the two species by centric rearrangements only, which are charted in Figures 2 and 6. Without comparative data from an outgroup it is not possible to distinguish fusions in one lineage from fissions in the other.

Both platypus and echidna have prominent NORs on the short arm of chromosome 6, which was thought to be homologous, because of their similar size and morphology 21. However, chromosome painting shows that platypus chromosome 6 is homologous to echidna chromosome16. In addition, echidna has NORs on X₄ and chromosome 3. Non-homology of NOR-bearing chromosomes is not surprising in view of their rearrangement in many closely related species.

The constitution of the sex chromosome system in Tac differs from that in Oan in their number, order and in the identity of one XY pair. Firstly, chromosome painting shows that there are five Xs and five Ys in platypus but five Xs and only four Ys in the echidna. The missing platypus Y₅ is a very small chromosome, and it had been previously supposed to have been lost from the echidna lineage 2127. However, the presence of a strongly hybridizing region on the larger echidna Y₃ using the platypus Y₅ paint suggests that the content of this platypus Y is incorporated into echidnaY₃. The surprising finding is that these two Y-chromosomes are at different locations in the chain. Secondly, Oan X₅ and Tac X₄ are shown to be homologous chromosomes by chromosome painting, and by sharing the DMRT1 gene cluster (Figure 10a). However, they occupy different positions in the chain. Thirdly, the most telling result we obtained was the finding that the platypus and echidna sex chromosome chains contain elements (Oan chromosomes Y₃, X₄ and Y₄, and Tac X₅ and Y₄) that are not homologous; platypus X₄ paints an autosome (chromosome 27) in echidna and echidna X₅ paints an autosome (chromosome 12) in platypus.

Thus, the chain in these monotreme species differs in both order and constitution, indicating that the chain continued to evolve after the divergence of platypus and echidna approximately 25 MYA 28. One can speculate that the primary sex determining locus is more likely to be found on the non-pairing regions of the X or Y chromosomes that are shared between platypus and echidna rather than on those that are autosomal in either.

Although the comparative painting reported here is a genome-wide comparison between the two monotreme species, it cannot, of course, reveal the gene content of the chromosomes, which requires comparative gene mapping.

Comparative gene mapping was used to establish that homologous genes are together in the same contiguous region in both species indicating chromosome homology. This does not exclude the possibility of non-orthology for some of the contiguous genes, but the likelihood of multiple, independent exceptional events is reduced as more homologues are discovered in the region. As the chance of independent evolution of syntenic regions is reduced, the likelihood of shared descent from the same chromosomal region in a common ancestor is increased.

The comparative mapping results show that monotreme sex chromosomes contain genes homologous to the chicken Z chromosome and chicken autosomes, with the implication that the sex determining system might be related to an ancestral sauropsid system. The early finding of the DMRT1 gene on platypus X₅ prompted our search for other chicken Z genes on this chromosome by mapping homologues of GGA-Z/HSA-9 genes followed by homologues of GGA-Z/HSA-5 genes. This led to the observation of chicken Z and autosomal homologues on other platypus sex chromosomes.

With regard to platypus X₁, the preliminary results of the draft genome sequence of the female platypus (Ensembl release 44) and a recent comparative mapping study of therian X-linked genes (F Veyrunes, personal communication) show that platypus X₁ does not share homology with the therian X as previously reported 1415161718. X-linked genes on human Xq assigned by FISH localization of BACs so far all map to platypus chromosome 6 10. These on human Xp map to platypus chromosome 15 and 18 23 (F Veyrunes, personal communication). The results presented here reveal instead that platypus X₁ shares homology with chicken chromosomes 3, 13, and Z and the corresponding human chromosomes 2, 5, 8, and 9 (Additional data file 1). Thus, our mapping data show that X₁ shares homology with those chicken chromosomes that are homologous to human autosomes.

The implications of some of the platypus gene assignments require further discussion. For example, the motivation to map RAPGEF6 was that this gene is close to the chicken Z HINT1 homologue in human (HSA-5q), although on a different chromosome in chicken (GGA-13). HINT1 might be involved in the chicken sex determining pathway, so may be a candidate for the primary sex determining locus in platypus; mapping a HINT1 homologue was unsuccessful. However, the localization of RAPGEF6 to Oan X₁, Y₁ makes it unlikely that this region contains the primary sex determinant. Note that platypus X₁p and Y₁q contain homologues of genes on chicken 3, 13 and Z (mapped in this report). The chicken 13 and the chicken Z regions were presumably syntenic before the divergence of Prototheria and Theria as these regions are syntenic in both monotremes and human.

The platypus homologues that map to X₂ (MLLT3), X₃ (TSCOT, PALM2), and both X₃ and Y₂ (CDH12, DNAH5, TRIO, P15RS) assign their respective contigs to these sex chromosomes (Figure 12 and Additional data file 1), indicating that parts of chicken 2 and chicken Z were fused before the divergence of Prototheria and Theria, and that the fission of chicken chromosome 2 into the regions on human chromosome 5, 9, and 18 occurred after the divergence of Prototheria and Theria.

Previous work showed that platypus chromosome X₅ contains the DMRT1-2-3 complex, whose homologue maps to the Z chromosome in chicken, and to human chromosome 9 68. We show here that several other genes with homologues on the chicken Z also map to the platypus X₅. Identification of the contigs that include these genes maps a large fraction of the chicken Z to platypus X₅, indicating that a large region of human 9 homologous to chicken Z is conserved on platypus chromosome X₅. Platypus major histocompability complex (MHC) class III genes have been mapped recently to the pairing segments of X₅ and Y₄ and MHC class I and II genes have been mapped to the pairing regions of X₃ and Y₃ 29. We note that platypus X₄ (homologous to the echidna autosome Tac 27) separates these two clusters, while in echidna the homologues of platypus X₅ and Y₄ (that is, Tac X₄ and Y₄) that carry the MHC genes are adjacent to X₃ and Y₃, suggesting that the insertion of X₄ in platypus was a later event in the evolution of the monotreme meiotic chain. So far, no homologues of chicken Z map to platypus X₄, possibly because it is homologous to an echidna autosome.

The gene mapping results described here assign several genes to the pairing regions of platypus Y chromosomes. The sequence reported in Ensembl does not provide this information as it is based on the female genome, so Y-linked genes are not included.

Not all chicken Z homologous genes are located on the monotreme sex chromosomes. Seven additional GGA-Z/HSA-5 genes map to platypus autosomes 1, 2, or 3 (Figures 11 and 12). The GGA-Z/HSA-8 genes LPL and CHRNB3 map to platypus chromosome 5 and the GGA-Z/HSA-18 gene ATP5A1 maps to platypus chromosome 3. Platypus chromosome 3 contains homologues of genes localized on chicken Z and human 5 and 18. This means that these two human regions were syntenic before monotreme-therian divergence and became separated only in the mammalian lineage after this divergence.

The monotreme regions homologous to chicken Z are considerably rearranged and distributed over autosomes and at least four X chromosomes and the corresponding Ys. The (single) Z homologues on X₁, Y₁ and X₂ are accompanied by other non-Z homologues in their respective contigs. X₃ seems to have a large region homologous to chicken Z as it contains contig 278 with a size around 0.8 Mb. This contig has chicken Z genes that are homologous only to human chromosome 9 and 5. In Ensembl release 45, contig 278 is part of ultracontig 84 with a size of around 8 Mb containing more chicken Z (human 5 and 9) homologous genes. Only chicken Z genes have so far been mapped to platypus X₅ and most of these are homologous to human 9 and a few to human 5. As both X₃ and X₅ contain human 5 and 9 genes (that are homologous to chicken Z), the separation into these human 5 and human 9 regions must have occurred later in the therian lineage. Finally, the platypus autosomes 1, 2 and 3 also contain chicken Z genes, suggesting that these chromosomes may represent the partners in original translocations involving a putative ancestral Z chromosome.

The results shown in Figure 12 all confirmed the homologies found by chromosome painting in this report and in our previous paper 7. In particular, a number of FISH assignments confirmed the pairing regions between X and Y chromosomes. These early mapping results should help in the identification of the primary sex-determining locus, in the investigation of mechanisms of dosage compensation and in understanding the evolution of vertebrate sex chromosomes.

An exchange mechanism for the evolution of the multiple sex chromosomes of the platypus was postulated previously 721. The chain development was suggested to start with an ancestral pair of differentiated sex chromosomes, one of which was repeatedly involved in exchanges with autosomes. An alternative view 2730 suggests that the chain arose as the result of hybridization between two ancestral monotreme populations, each with a different set of Robertsonian translocations resulting in a male heterozygous for unpaired sex chromosomes. Common to all models is that the rearranged autosomes in the chain evolved into Y chromosomes. Our finding that different rearrangements occurred in the two monotreme lineages after the platypus-echidna divergence (25 MYA 28) is easier to reconcile with a model of successive translocation, rather than the unlikely alternative of additional hybridizations between populations differing in other Robertsonian rearrangements.

Primary fibroblast cultures from the short-beaked echidna (T. aculeatus, 2n = 63 male, 64 female) were established routinely in standard medium at 32°C (AEEC permit no. R.CG.07.03 and AEC permit no S-049-2006, NSW P&W permit S10443). Flow sorting, chromosome paint production and FISH were performed according to the protocol described previously 732. Platypus (2n = 52) chromosome paints and metaphases were generated as previously described for the characterization of the platypus sex chromosome complement 7.

Cot-1 DNA was prepared as described 33, except that the renatured (double-stranded) Cot-1 DNA was separated from the digested (single-stranded) DNA using a phenol-chloroform-extraction method. Purified echidna Cot-1 DNA (30 mg) was used for FISH.

Meiotic cells were obtained from animals captured at the upper Barnard river, New South Wales, Australia during breeding season (AEEC and AEC permits to FG as above). The captured animals were euthanased with an intraperitoneal injection of pentobarbitone sodium (Nembutal, Boehringer Ingelheim, NSW, Australia) at a dose of 0.1 mg/g body weight. Meiotic cells were obtained by disaggregating the testis. The material was either directly fixed in methanol/acetic acid (3:1) or incubated in 0.075 KCl M at 37°C as hypotonic treatment to improve spreading of metaphase cells and then fixed.

Microdissection-derived chromosome specific paints of specific blocks were generated as described previously 34. In short, short-beaked echidna metaphases were dropped onto wet cover slips and subsequently digested with 0.015% trypsin for 1 minute and stained with Giemsa. Microdissected material was collected in a drop containing 10 mM NaCl, 10 mM Tris-HCL, 1 mM EDTA, 0.1% SDS, 0.1% Triton ×100, 1.44 mg/ml Proteinase K (Sigma-Aldrich, Gillingham, Dorset, UK), 30% glycerol. After 1 hour incubation at 60°C this drop was transferred to 5 μl of 1× sequenase buffer (USB, Staufen, Germany), 0.2 mM dNTPs, 5 μM 6 mW primer. A PCR protocol for low temp cycles was next: 5 minutes at 92°C followed by 8 cycles of 2 minutes 20 s at 25°C, 2 minutes at 34°C, 1 minute at 90°C. At the start of each cycle 0.4 units of sequenase (USB) was added. For the high temp cycles 45 μl 1× Buffer D (Invitrogen, Paisley, UK), 0.2 mM dNTP, 5 μM 6 mW primer, 0.3 u SuperTaq polymerase (HT Biotechnology, Cambridge, UK), 0.05% W1 (Sigma) were added. This solution was subjected to 33 cycles of 1 minute at 92°C, 2 minutes at 56°C, 2 minutes at 72°C with a final extension for 5 minutes at 72°C. The DNA was labeled using a standard protocol 32.

Images were captured using the Leica QFISH software (Leica Microsystems, Milton Keynes, UK) and a cooled CCD camera (Photometrics Sensys, Photometrics, Tucson, AZ, USA) mounted on a Leica DMRXA microscope equipped with a 63×, 1.3 NA objective. Cy3, FITC 9 (fluorescein isothiocyanate) and DAPI (4',6-diamidino-2-phenylindole) signals were captured separately as 16 bit black and white images, and merged to a color image. The DAPI image was enhanced with a spatial filter to obtain enhanced chromosome bands. All image processing was performed with Leica CW4000 software.

The Tac paints produced were hybridized to male (three individuals) and female (one individual) Tac metaphase preparations. Multicolor chromosome painting was used to ensure that different peaks represent different chromosomes, to define the order of the unpaired chromosomes and to determine the homologous parts that link these chromosomes in the meiotic chain.

Twenty-nine genes were mapped to platypus chromosomes (Table 1), and eight of these to echidna chromosomes. Two methods, PCR and BAC-clone mapping (indicated by 'a' and 'b' in Table 1) were used to localize the platypus homologues.

Comparative gene mapping was used for the assessment of chromosome homology. It is important to consider whether the homologous genes are likely to be true orthologues. The definition of orthologues is two genes from two different species that derive from a single gene in the last common ancestor of the species 36. Absolute proof of orthology is difficult on this criterion and was not pursued in this report. Instead, support for chromosome homology was provided when homologous genes were together within the same contiguous region in both species. Instances of non-orthology in the syntenies may exist but the likelihood of multiple, independent exceptional events is reduced when several gene homologues are found in one region.