 Reference Base  The complete mitochondrial DNA sequence of the Greater In... |
|
Xu, Xiufeng; Janke, A.; Arnason, U., 1996. The complete mitochondrial DNA sequence of the Greater Indian rhinoceros, Rhinoceros unicornis, and the phylogenetic relationship among Carnivora, Perissodactyla, and Artiodactyla (+ Cetacea).. Molecular Biology and Evolution 13 (9): 1167-1173, fig. 1, tables 1-8
|
|
|
|
| |
| Location: |
World |
| Subject: |
Genetics |
| Species: |
Indian Rhino |
 |
|
Mitochondrial DNA sequence of Rhinoceros unicornis . [For tables, see file Xu & Arnason]
Introduction
Sequence data of mitochondrial DNA (mtDNA), notably the peptide-coding genes, have become a widely used tool for addressing phylogenetic relationships at various levels, especially among mammals. The importance of using comprehensive amounts of mitochondrial sequence data for inferring phylogenetic relationships was demonstrated by Cao et al. (1994), who, on the basis of analyses of complete mtDNAs, showed that different mtDNA genes provided different topologies for the ordinal relationship among Primates, Rodentia, Carnivora, Artiodactyla, and Cetacea. The topic that individual mtDNA genes (cytochrome b, COII) may, provide different tree topologies has also been addressed by Ho- neycutt and Adkins (1993) and Honeyeutt et al. (1995).
The mtDNA findings of Arnason and Johnsson (1992) and Janke et al. (1994) grouped Carnivora, Artiodactyla, and Cetacea into a superordinal clade, consistent with the nuclear data of Li et al. (1990) and Bulmer, Wolfe, and Sharp (1991). These analyses lacked, however, a suitable ungulate representation for conclusively assessing this relationship. The availability of the complete mtDNA of the horse (Xu and Arnason 1994) amended this shortcoming and a recent phylogenetic study including the horse and the hedgehog (Krettek, Gullberg, and Arnason 1995) identified the relationship among Perissodactyla (horse), Carnivora, and Artiodactyla (+ Cetacea) as essentially that of an unresolved triochotomy. The inclusion of the hedgehog was essential for these conclusions because of the proposal of phylogenetic affinities between Lipotyphla (hedgehog) and Carnivora (Miyamoto and Goodman 1986; MacPhee and Novacek 1993; Wyss and Flynn 1993). The analysis of Krettek, Gullberg, and Arnason (1 995) yielded no support to this understanding, however, showing that the Lipotyphla (as represented by the hedgehog) had a basal position among the eutherians included.
We report now the complete mtDNA molecule of the greater Indian rhinoceros (hereafter referred to as Indian rhinoceros), Rhinoceros unicornis, which represents a family, Rhinocerotidae, that is distantly related to the Equidae. On the basis of this improved perissodactyl representation and by including also the mtDNA of the domestic cat (Lopez et al. 1996). we reexamine the mtDNA relationships among Carnivora. Perissodactyla, and Artiodactyla + Cetacca and propose a molecular dating of the divergence between the families Rhinocerotidae and Equidae.
Materials and Methods
Mitochondrial DNA was isolated from a frozen kidney sample of Indian rhinoceros ('Miris') that died in Zoologischer Garten Berlin AG, Germany. The sample was kindly provided by Dr. Reinhard G?ltenboth. The isolation of mtDNA followed the same procedure as described in Arnason, Gullberg, and Widegren (1991). The enriched mtDNA was digested separately with Bgl II, Bln I, HindIII and Spe I. The products were ligated directly into M13 and cloned in E. coli JM 109. Positive clones were identified by hybridization using mtDNA fragments of the horse and donkey as labeled probes. Sequencing of cloned fragments was performed on single-stranded DNA applying the dideoxy termination technique with [35S] dATP The work was performed manually using both universal and numerous specific oligonucleotide primers. The entire mtDNA molecule was covered by natural (not PCR) clones. Complementary to the natural clones the repetitive portion of the control region was sequenced after M13 cloning of PCR-amplified fragments in order to determine the numbers of repetitive motifs.
The mtDNA sequence of the greater Indian rhinoceros has been deposited at the EMBL data bank with accession number X97336. Users of the sequence are kindly requested to refer to the present paper and not only to the accession number of the sequence.
Results and Discussion
General Features of the Mitochondrial Genome of Indian Rhinoceros
The length of the complete mtDNA sequence of the Indian rhinoceros, Rhinoceros unicornis, presented here is 16,829 base pairs. Like the mtDNAs of other perissodactyls, the control region of the Indian rhinoceros is characterized by tandemly arranged repetitive motifs. Therefore, the length of the molecule is not absolute. The nt composition of the L-strand and different features of the molecule is given in table 1. The nucleotide composition of different regions of the mitochondrial genome is consistent with that of other mammals (Janke et al. 1994). The underrepresentation of guanine in the L-strand is particularly pronounced at third codon position where all transitions are silent. Underrepresentation of this kind is not observable, however, in the control region.
The control region presently reported is 1,376 nt long with a continuous run of 36 identical repetitive motifs. 5'-CACATGTA. The repetitive motifs are located in the 3' part of the control region in positions 16194-16481 of the complete sequence. The motif has the purine/ pyrimidine alternation that characterizes most mammalian mtDNA motifs so far described (Ghivizzani et al. 1993; Hoelzel, Hancock, and Dover 1993; Xu and Arnason 1994). The number of repeats was determined in a total of 48 clones (23 natural and 25 PCR clones). The lowest number of repeats was 3 and the highest was 36. The characteristics of the repeat motifs in the three perissodactyl families, Rhinocerotidae, Tapiridae, and Equidae will he dealt with in a separate paper.
Two peptide-coding genes, NADH3 and NADH6, do not have a methionine start codon. Comparison with other mammalian mtDNAs suggests that the Indian rhinoceros has ATT (isoleucine) as the start codon of the NADH3 gene and GTG (valine) as the start codon of NADH6. Both these codons appear to be potential start codons in different mammalian mtDNAs (e.g., Xu and Arnason 1994). Three of the 13 peptide-coding genes, COIII, NADH3. and NADH4, are not terminated by a complete stop codon. Among mammals the occurrence of a complete stop codon in COIII has only been reported in the fin (Arnason, Gullberg, and Widegren 1991) and the blue (Arnason and Gullberg 1993) whales, and NADH3 is terminated by a complete stop codon only in the mouse (Bibb et al. 1981) and the rat (Gadaleta et al. 1989). A complete stop codon in NADH4 has not been described so far in any mammalian mtDNA. The COIII, NADH3, and NADH4 genes of the Indian rhinoceros have incomplete stop codons (T in COIII and NADH4, and TA in NADH3), consistent with the findings that the transcripts of such genes contain a stop codon created by posttranscriptional polyadenylation (Ojala, Montoya, and Attardi 1981).
The boundaries of the 22 TRNA genes of the mtDNA of the Indian rhinoceros were determined by analogy with the tRNA genes of different eutherians. Twenty of the tRNA genes of the rhinoceros mtDNA have the standard secondary structure discussed by Ku- mazawa and Nishida (1993). The features of these tRNAs, as well as those of the structurally atypical tRNA-Ser(AGY) and tRNA-Ser(UCN), conform with those described for other eutherians.
Comparison with the mtDNA of the Horse
In the present study we give an account of the molecular difference between two distantly related species of the order Perissodactyla, the Indian rhinoceros and the horse. The percentage nt composition of the mtDNA (L-strand) outside the control region of rhinoceros/ horse is similar: A = 33.8/32.5; C = 27.5/28.4; G = 12.6/13.3; T = 26.1/25.8. The corresponding percentages for the control region are: A = 32.5/27.3; C = 27.2130.9; G = 13.6/15.5; T = 26.7/26.3. Alignment of the mtDNA molecules of the rhinoceros and the horse, less the control region, shows a sequence difference of 15.6 %. In addition to indel (insertion/deletion) differences in genic regions this alignment shows three indels in non- genic regions, one at the junction between tRNASer(UCN) and tRNA-Asp, one between tRNA-Arg and NADH4L, and one in the loop of origin of L-strand replication.
The length of the NADH2 gene of the rhinoceros is 1,044 nt, 3 nt (one codon) more than that of the horse. Other peptide-coding genes have the same length in the two species. The nt differences (both total and conservative) were detailed according to codon position and type of substitution (transition or transversion) (table 2). The 13 peptide-coding genes differ by a total of 1.923 substitutions, 16.9%. The ratio for these differences with respect to codon positions 1, 2, and 3 is 2.3:1:8.9, as compared with 3.5:1:25.7 in the intrageneric comparison between the horse and the donkey (Xu, Guilberg, and Arnason 1996). The total number of conservative nt substitutions is 960 (8.4%). The codon position ratio for these differences is 1.6:1:3.6, as compared with 1.5:1: 2.1 in the horse/donkey comparison. The two sets of ratios for conservative nt substitutions show that the accumulation of substitutions in second codon position and the accumulation of nonsynonymous substitutions in first codon position have the same rate in the close (horse/donkey) and the distant (horse/rhinoceros) relationships and that this rate is considerably slower than that for accumulation of transversions in third codon position. The difference between the two sets of ratios for total nt substitution shows that there is a high degree of transition saturation in third codon position in rhinoceros/horse relative to horse/donkey. The findings suggest that there is a high degree of saturation in first codon position synonymous transitions (leucine) in the rhinoceros/horse comparison.
Table 3 shows the results of a pairwise comparison between the 13 mitochondrial peptide-coding genes of the Indian rhinoceros and the horse. The table shows, for each gene, the percent total amino acid (aa) difference, the percent conservative nt difference (Irwin, Ko- cher, and Wilson 1991), and the percent nt difference. The genes have been arranged according to increasing aa difference. Percent total aa difference was in the range 1.8 (COT) to 25.0 (ATPaseS). For conservative nt substitutions the range was 5.1% (COT) to 12.7% (NADH6). For total nt difference the corresponding range was 13.6% (NADH3) to 22. 1 % (ATPase8). The order of the genes, with respect to increasing difference. is highly similar between the two conservative modes of comparison, aa difference, and conservative nt sub- stitutions. The amplitude for total nt difference is relatively limited among the 13 genes, consistent with (primarily) third codon position mutational saturation, and the order based on percent total nt substitution deviates to some extent from the order based on the other two approaches, which are more conservative.
The results of a pairwise comparison between each of the 22 tRNA genes of the Indian rhinoceros and the horse are shown in table 4. The differences are detailed according to position in secondary structures. There are a total of 169 differences (11.1 %), 129 transitions (8.5%), 28 transversions (1.8%), and 12 indels (0.8%).
Very few differences (all transitions) occur in the D stem and in the AC loop. The alignment between the tRNA genes for cysteine, lysine, and leucine (CUN) shows a single transition at the junction between the D and AC stems. The difference between the two species is particularly marked in tRNA-Lys (17 differences) and tRNA-Thr (16 differences). The most conservative tRNAs are tRNA-Leu (CUN) (two substitutions), tRNA-Pro (three substitutions), and tRNA-Met (three differences). The transition/ transversion (Ti/Tv) ratio for the combined length of the tRNA genes is 4.6.
The tRNA-Pro and tRNA-Phe genes of the black rhinoceros were reported by Jama et al. (1993). In the tRNA-Pro gene the two rhinoceroses differ by five transitions, two more than horse and Indian rhinoceros. In the tRNA-Phe gene the two rhinoceroses differ by seven
transitions and one transversion, as compared with three transitions, four transversions and one indel (insertion/ deletion) between the Indian rhinoceros and the horse.
The 12S and 16S rRNA genes of the Indian rhinoceros and the horse differ at 128 (13.1%) and 180 (11.3%) positions, respectively. The differences are detailed in table 5. For the combined length of the rRNA genes the Ti/Tv ratio is 2.3.
|
|
|
File available
details
Original text on this topic: