Abstract

Recent comprehensive studies of DNA sequences support the monophyly of Afrotheria, comprising elephants, sirenians (dugongs and manatees), hyraxes, tenrecs, golden moles, aardvarks, and elephant shrews, as well as that of Paenungulata, comprising elephants, sirenians, and hyraxes. However, phylogenetic relationships among paenungulates, as well as among nonpaenungulates, have remained ambiguous. Here we applied an extensive retroposon analysis to these problems to support the monophyly of aardvarks, tenrecs, and golden moles, with elephant shrews as their sister group. Regarding phylogenetic relationships in Paenungulata, we could characterize only one informative locus, although we could isolate many insertions specific to each of three lineages, namely, Proboscidea, Sirenia, and Hyracoidea. These data prompted us to reexamine phylogenetic relationships among Paenungulata using 19 nuclear gene sequences resulting in three different analyses, namely, short interspersed element (SINE) insertions, nuclear sequence analyses, and morphological cladistics, supporting different respective phylogenies. We concluded that these three lineages diverged very rapidly in a very short evolutionary period, with the consequence that ancestral polymorphism present in the last common ancestor of Paenungulata results in such incongruence. Our results suggest the rapid fixation of many large-scale morphological synapomorphies for Tethytheria; implications of this in relation to the morphological evolution in Paenungulata are discussed.

Introduction

Recent extensive analyses of large amounts of DNA sequence data have revealed many interordinal mammalian relationships (e.g., Madsen et al. 2001; Murphy et al. 2001a, 2001b). One of the most important outcomes from these studies was the identification of the superordinal clade Afrotheria (Springer et al. 1997; Stanhope et al. 1998a). This group includes modern Proboscidea (elephants), Sirenia (dugongs and manatees), Hyracoidea (hyraxes), Afrosoricida (tenrecs and golden moles), Tubulidentata (aardvarks), and Macroscelidea (elephant shrews). Comprehensive sequence data of DNA and amino acids, as well as chromosome painting analyses, suggest that they are monophyletic (Stanhope et al. 1998a, 1998b; van Dijk et al. 2001; Malia, Adkins, and Allard 2002; Douady and Douzery 2003; Fronicke et al. 2003; Yang et al. 2003; Robinson and Seiffert 2004; Robinson et al. 2004; Roca et al. 2004), although it remains unacceptable on the basis of anatomical and paleontological studies (Novacek 1992a, 2001; MacPhee and Novacek 1993; Asher 1999; Asher, Novacek, and Geisler 2003). In the clade Afrotheria, the Mirorder Tethytheria includes two extant orders of Proboscidea and Sirenia (McKenna 1975), and the Grandorder Paenungulata comprises Tethytheria and the Order Hyracoidea (Simpson 1945; Benton 2004). Concerning Paenungulata, since Gregory (1910), many morphologists have suggested their monophyly (Novacek, Wyss, and McKenna 1988; Novacek 1992a; Shoshani 1992, 1993). Many phylogenetic analyses based on various gene sequences also supported this conclusion (e.g., Lavergne et al. 1996; Stanhope et al. 1996; Springer et al. 1999; Murata et al. 2003). Therefore, the monophyly of Paenungulata is established from molecular and morphological viewpoints, although the monophyly of Afrotheria is not accepted from morphology.

In contrast to these two clades, monophyly of Tethytheria remains unsupported by molecular analyses, although this clade is strongly supported by morphological data. Since de Blainville first classified elephants and sirenians as a sister group in 1834 (summarized in Gregory 1910), many synapomorphies for Tethytheria have been recognized and, accordingly, most morphologists support the monophyly of elephants and sirenians among living paenungulates (e.g., Domning, Ray, and McKenna 1986; Tassy and Shoshani 1988; Novacek 1992b). From the molecular viewpoint, however, the phylogenetic relationship among elephants, sirenians, and hyraxes remains confused. Although various nuclear genes have been analyzed, concomitant with large data sets of nuclear combined with mitochondrial (mt) sequences, the results often challenge monophyly of Tethytheria and support variously one of the three topologies. On the other hand, analysis of complete mtDNA of all afrotherian orders strongly supported the monophyly of Tethytheria (Murata et al. 2003). Although mtDNA data sometimes mislead incorrect phylogenetic tree by several problems such as long-branch attraction (Phillips and Penny 2003; Gibson et al. 2005), long-branch attraction is not the case for monophyly of Tethytheria (Murata et al. 2003). Thus, the monophyly of Tethytheria is currently one of the most important questions in mammalian phylogenetics (Murphy, Pevzner, and O'Brien 2004).

On the other hand, the phylogenetic relationships among afrotherians other than paenungulates are also very interesting because some members of them exhibit brilliant examples of convergence during mammalian evolution. For example, hedgehog tenrecs and golden moles are very similar to hedgehogs and moles, respectively, in appearance (Nikaido et al. 2003a; Springer et al. 2004). Even comprehensive DNA sequence analyses, however, have not clarified phylogenetic relationships among aardvarks, tenrecs, golden moles, and elephant shrews. For instance, concatenated nuclear and mtDNA data suggested that aardvarks first diverged among them (Murphy et al. 2001b), whereas complete mtDNA analysis supports that aardvarks, elephant shrews, and golden moles are monophyletic (Murata et al. 2003; Gibson et al. 2005). Therefore, for further comprehension about their morphological evolutionary dynamics, it is important to shed light on the evolutionary history of African endemic mammals (Murphy, Pevzner, and O'Brien 2004). In the present study, we addressed these several issues by SINE insertion analysis.

SINEs are one of the major groups of retroposons, which propagate via reverse transcription of their RNA transcripts and are integrated into random sites in host genomes (Rogers 1985; Weiner, Deininger, and Efstratiadis 1986; Okada 1991a, 1991b). Insertions of retroposons are believed to be an irreversible event, and it is highly unlikely that a retroposon is inserted into the same genomic locus independently in different lineages during evolution. Such characteristics of retroposons provide us a nearly homoplasy-free character for phylogenetics (Rokas and Holland 2000; Shedlock and Okada 2000), and the insertion analysis of retroposons has already been used as a powerful marker to resolve phylogenetic relationships of various animals (Murata et al. 1993; Shimamura et al. 1997; Nikaido, Rooney, and Okada 1999; Nikaido et al. 2001; Schmitz et al. 2002; Salem et al. 2003; Roos, Schmitz, and Zischler 2004).

Although retroposon insertion analyses are quite powerful to estimate species phylogeny, there are cases where a pattern of retroposon insertion does not reflect a true species tree. If two speciation events occurred within a very short period and ancestral polymorphisms of retroposon presence/absence alleles were followed by random fixation of the alleles after speciation, a pattern of retroposon insertion may not be consistent with the species tree (Shedlock, Takahashi, and Okada 2004). Therefore, especially when a node is supported by only one locus of retroposon insertion, a possibility of such incomplete lineage sorting should be taken into consideration.

Recently, we characterized a novel SINE family designated AfroSINE, whose distribution is restricted to the genomes of afrotherian mammals (Nikaido et al. 2003b). This SINE family was divided into three subfamilies, namely, Anc, Ad, and HSP. Among these, Anc and Ad are detected in all afrotherians, whereas HSP resides only in the genomes of paenungulates. Therefore, we used HSP predominantly to elucidate phylogenetic relationships among paenungulates by SINE insertion analysis.

Materials and Methods

Retroposon Insertion Analysis

Genomic DNAs of the following afrotherian species were isolated according to standard protocols (Sambrook, Fritsch, and Maniatis 1989): the African elephant (Loxodonta africana), the Asian elephant (Elephas maximus), the West Indian manatee (Trichechus manatus), the dugong (Dugong dugon), the Cape hyrax (Procavia capensis), the aardvark (Orycteropus afer), the lesser hedgehog tenrec (Echinops telfairi), the Cape golden mole (Chrysochloris asiatica), and the black and rufous elephant shrew (Rhynchocyon petersi). To design primers of polymerase chain reaction (PCR) for the retroposon analysis we adopted two approaches. The first one is to refer to the human genome data (UCSC Genome Bioinformatics; http://genome.ucsc.edu/) and to design primers in two exons, the length between which is less than 2 kb, to detect retroposon insertions in introns. We performed interexon PCR for 40 such loci (introns) to examine phylogenetic relationships within Afrotheria. The second approach is to construct genomic libraries from DNAs of the African elephant, manatee, and hyrax and to design primers in the flanking region of a SINE-containing locus (Supplementary Table 1, Supplementary Material online) to analyze phylogenetic relationships in Paenungulata. To isolate orthologous loci, three sets of primers were designed for each side of the flanking regions at each locus because of the large genetic distances among the three orders. DNA clones containing SINE units were screened from the libraries using [α-32P]dCTP-labeled PCR products of AfroSINE-HSP as probes (Nikaido et al. 2003b) and sequenced. All PCR products were sequenced, and the nucleotide sequence data were deposited in GenBank (accession numbers AB208111AB208250).

Results

Relationships Within Afrotheria

To examine interordinal relationships in Afrotheria by retroposon insertion analysis, the orthologous locus where a retroposon was inserted must be isolated and be compared among various mammalian species from each order. Considering that the last common ancestor of Afrotheria can be traced back to about 80 MYA (Springer et al. 2003; Roca et al. 2004), isolation of such a locus is extremely difficult because of the large genetic distance among these species. To overcome such difficulty, we performed interexon PCR using primers designed on exons to amplify intron sequences (see Materials and Methods). We discovered five loci (INT245, INT292, INT298, INT1193, and INT1095) in which retroposon insertions indicate monophyly of Afrotheria and interordinal phylogenetic relationships within the clade (see Supplementary Table 2, Supplementary Material online). Intron information referred to in the human genome are as follows: INT245, 5th intron of F-box protein 32 (FBXO32) gene; INT292, 13th intron of IL-2–inducible T-cell kinase (ITK) gene; INT298, 9th intron of glutamate receptor subunit 3 (GRIA3) gene; INT1193, 49th intron of neurofibromin 1 (NF1) gene; and INT1095, 5th intron of FLJ44290 gene. Among them, four loci (INT245, INT292, INT298, and INT1193) supported monophyly of Afrotheria. An informative SINE inserted in the locus INT245 belongs to a new SINE family, designated “AfroSINE2,” while those in the other three belong to the previously identified AfroSINE family (Nikaido et al. 2003b). AfroSINE2 is restricted to only afrotherian species (data not shown), suggesting monophyly of Afrotheria, similar to that of the AfroSINE family.

As an example, figure 1A shows a schematic representation of the intron sequences for INT1193 locus in which the AfroSINE insertion indicates monophyly of Afrotheria. Insertions of other retroposons or unknown sequences indicated by white boxes in figure 1 are species specific, and they are represented on the data matrix (Supplementary Table 2A, Supplementary Material online) and the phylogeny in figure 4 only when they were specifically inserted into one of the afrotherian species. In the locus of INT1193, the distribution of a 150-bp insertion, as shown by “ins1” in figure 1A, supports monophyly of Paenungulata. Also, an 80-bp orthologous sequence is present in only aardvark and tenrec (“ins2” in figs. 1A and 2A). Although it is unknown for golden mole, its distribution is consistent with the data that elephant shrew diverged first among nonpaenungulate species of afrotherians (see below).

FIG. 1.—

A schematic representation of two introns. (A) INT1193: distribution of AfroSINE, ins1, and ins2 indicates monophyly of Afrotheria, monophyly of Paenungulata, and sister group of aardvark and Afrosoricida, respectively. (B) INT1095: distribution of MLT1B and AfroSINE indicates monophyly of Afroinsectiphillia and sister group of aardvark and Afrosoricida, respectively. The number corresponds to species as shown at the bottom. Black and gray boxes represent insertions indicating phylogenetic relationships in Afrotheria, whereas white boxes represent insertions of other retroposons or unknown sequences specific to species. Hatched boxes flanking introns denote exons, and regions flanked by “//” are unsequenced.

FIG. 1.—

A schematic representation of two introns. (A) INT1193: distribution of AfroSINE, ins1, and ins2 indicates monophyly of Afrotheria, monophyly of Paenungulata, and sister group of aardvark and Afrosoricida, respectively. (B) INT1095: distribution of MLT1B and AfroSINE indicates monophyly of Afroinsectiphillia and sister group of aardvark and Afrosoricida, respectively. The number corresponds to species as shown at the bottom. Black and gray boxes represent insertions indicating phylogenetic relationships in Afrotheria, whereas white boxes represent insertions of other retroposons or unknown sequences specific to species. Hatched boxes flanking introns denote exons, and regions flanked by “//” are unsequenced.

FIG. 2.—

Alignments of sequences for loci INT1193-ins2 (A), INT1095 (B), and Hy05 (C). Dots indicate nucleotides identical to those of the African elephant, whereas dashes indicate gaps inserted to improve the alignment. Bold lines above the alignments represent inserted sequences. Central region of each inserted sequence is omitted. Lines that flank the AfroSINEs in (B) and (C) are direct repeat sequences that were generated during SINE integration.

FIG. 2.—

Alignments of sequences for loci INT1193-ins2 (A), INT1095 (B), and Hy05 (C). Dots indicate nucleotides identical to those of the African elephant, whereas dashes indicate gaps inserted to improve the alignment. Bold lines above the alignments represent inserted sequences. Central region of each inserted sequence is omitted. Lines that flank the AfroSINEs in (B) and (C) are direct repeat sequences that were generated during SINE integration.

In the locus INT1095, an MLT1B element, a type of long terminal repeat (LTR)-retrotransposon, is inserted in aardvark, tenrec, golden mole, and elephant shrew and not in the other placental mammals (figs. 1B and 2B). This locus suggests monophyly of these four species of Afroinsectiphillia (Waddell, Kishino, and Ota 2001), namely, aardvark, tenrec, golden mole, and elephant shrew (fig. 4). Additionally, we detected an AfroSINE insertion in the MLT1B sequence in aardvark, tenrec, and golden mole but not in elephant shrew, again suggesting that elephant shrews diverged first among Afroinsectiphillia species of afrotherians.

Relationships Within Paenungulata

To resolve the Paenungulata phylogeny, we extensively screened SINEs from genomic libraries of African elephant, manatee, and hyrax and isolated a total of 213 SINE loci. Among them, a total of 93 SINE loci (33, 44, and 16 loci isolated from African elephant, manatee, and hyrax, respectively) were then chosen, and PCR was performed to characterize their presence or absence in related species. Among these, 26 loci could be successfully characterized among elephants, sirenians, and hyrax (four examples are shown in fig. 3). A data matrix of AfroSINE insertions for the 26 loci is constructed (Supplementary Table 2B, Supplementary Material online). A total of 14 loci were consistent with a history of insertion in the common ancestor of paenungulates. Regarding four of these loci, namely, E39, Ma036, Ma101, and Ma107, the orthologous loci from Afroinsectiphillia species were isolated and the absence of these SINEs was confirmed (see Supplementary Table 2B, Supplementary Material online), providing convincing evidence of Paenungulata monophyly. The remaining 10 loci were considered to be Paenungulata-specific insertions because they belong to the AfroSINE-HSP subfamily, which is distributed only in the genomes of paenungulates (Nikaido et al. 2003b). Besides these loci, five and four loci, respectively, supported the monophyly of Proboscidea and Sirenia (fig. 4), and two loci (Hy06 and E78Hy) indicated SINE insertions specific to the hyrax lineage.

FIG. 3.—

Electrophoretic profiles of PCR products of four loci. (A) Ma036: monophyly of paenungulates. (B) Ma158E: monophyly of proboscideans. (C) Ma030: monophyly of sirenians. (D) Hy05: possible sister group between sirenians and hyrax. The lane numbers correspond to the species that were used as PCR templates, as indicated at the lower right. The PCR primer sequences are shown in Supplementary Table 1 (Supplementary Material online). Sequence alignments for Hy05 are shown in figure 2C.

FIG. 3.—

Electrophoretic profiles of PCR products of four loci. (A) Ma036: monophyly of paenungulates. (B) Ma158E: monophyly of proboscideans. (C) Ma030: monophyly of sirenians. (D) Hy05: possible sister group between sirenians and hyrax. The lane numbers correspond to the species that were used as PCR templates, as indicated at the lower right. The PCR primer sequences are shown in Supplementary Table 1 (Supplementary Material online). Sequence alignments for Hy05 are shown in figure 2C.

FIG. 4.—

Phylogenetic relationships of Afrotheria reconstructed by retroposon insertions are shown in Supplementary Table 2 (Supplementary Material online). Closed vertical arrowheads denote insertions of retroposons into each lineage. The 10 SINE loci in a dashed line box gives the possible paenungulate-specific insertions because they belong to AfroSINE-HSP subfamily (see text). All the loci include insertions of SINEs except for that of MLT1B at INT1095-1 and INT1095Aa-2. It is unclear that the inserted sequences at INT1193-ins1 and -ins2 are retroposons.

FIG. 4.—

Phylogenetic relationships of Afrotheria reconstructed by retroposon insertions are shown in Supplementary Table 2 (Supplementary Material online). Closed vertical arrowheads denote insertions of retroposons into each lineage. The 10 SINE loci in a dashed line box gives the possible paenungulate-specific insertions because they belong to AfroSINE-HSP subfamily (see text). All the loci include insertions of SINEs except for that of MLT1B at INT1095-1 and INT1095Aa-2. It is unclear that the inserted sequences at INT1193-ins1 and -ins2 are retroposons.

Interestingly, one SINE, in the locus designated Hy05, was present in sirenians and hyrax but not in the orthologous loci of elephants, aardvark, or tenrec (see figs. 2C and 3D). This would appear to suggest that this SINE insertion is an indication of a sister-group relationship between sirenians and hyrax (fig. 4). However, considering that we characterized many SINE loci, inserted in the common ancestor of paenungulates and within the respective lineages of three orders, it is likely that these three lineages diverged very rapidly. In such a case, it is quite possible, perhaps very likely, that this result does not reflect the true species tree but instead may be due to the presence of ancestral polymorphism followed by incomplete lineage sorting of SINE presence/absence alleles (see Discussion and a recent review by Shedlock, Takahashi, and Okada 2004).

Discussion

Afrotherian Phylogeny and Evolution

Our SINE data supported monophyly of each of Afrotheria and Paenungulata, which were supported by 4 and 14 insertions of retroposons, respectively (fig. 4). Therefore, there is little doubt about the monophyly of Afrotheria as suggested by other molecular analyses, although there is no clear morphological support for this clade. From a biogeographic viewpoint, our data support the previous notion that afrotherians have radiated in Africa while they were physically isolated after the breakup of the ancient supercontinent Gondwana (Murphy et al. 2001b).

An insertion of MLT1B at INT1095 indicates monophyly of Afroinsectiphillia, namely, aardvark, tenrec, golden mole, and elephant shrew (fig. 4). This conclusion is also supported by three different types of recent data, namely, complete mtDNA (Murata et al. 2003), concatenated nuclear sequences (Murphy et al. 2001b), and chromosome painting analyses (Robinson et al. 2004; Svartman et al. 2004). Therefore, although we could isolate only one locus, this consistency suggests that the data of MLT1B at INT1095 really reflect the species tree and are not a result of random sorting of ancestral presence/absence alleles of the locus. In addition, monophyly of aardvarks, tenrecs, and golden moles is newly demonstrated by two insertions in the present study. Many studies based on nuclear and mt sequences have so far provided only ambiguous results regarding relationships in Afroinsectiphillia, although monophyly of tenrecs and golden moles, proposed as Afrosoricida, is supported by relatively high bootstrap values. We proposed here monophyly of aardvarks and Afrosoricida, with elephant shrews as their sister group, which was a phylogenetic issue that morphological cladistics has failed to resolve, probably because of their extreme morphological diversification.

The pattern of morphological diversification and evolution involving these orders needs to be reexamined. Accordingly, future paleontological studies will recognize some fossil records as ancient afrotherians by reconsideration of Afotherian molecular phylogeny. For example, Ptolemaiidae includes genera such as Ptolemaia, Qarunavus, and Cleopatrodon and is known only from Africa in late Eocene to early Oligocene (McKenna and Bell 1997). Although the phylogenetic position of Ptolemaiidae among mammals is in controversy, Butler (1969) suggested close similarity between Qarunavus and elephant shrews from dental morphology, while Simons and Gingerich (1974) suggested a possible relationship between Ptolemaia and aardvarks. These observations provide a possibility that Ptolemaiidae might be another member of Afrotheria, particularly of Afroinsectiphillia.

Reexamination of Nuclear Gene Sequences for Paenungulate Phylogeny

A large number of nuclear gene phylogenetic analyses have been published which address relationships within Paenungulata. As described in Introduction, these results tend to be very mixed in their support for one branching arrangement or another. Amrine-Madsen et al. (2003) in a thoroughly analyzed 17.7-kb concatenation, comprising largely nuclear gene sequence data, for 44 interordinal mammalian taxa, found that the most likely branching arrangement was elephant + hyrax (E + H). We used the amino acid sequences of 19 nuclear genes (listed in Supplementary Table 3, Supplementary Material online) available for three species of paenungulates and three out-groups including human, mouse, and one of the four species of aardvarks, tenrecs, golden moles, and elephant shrews. For each alignment, we examined the number of phylogenetically informative sites supporting each of E + S, S + H, and E + H sister groups. It is clear that more than half of the informative sites supported the E + H sister-group relationship, irrespective of out-group sequences (see “Sum” in table 1). Thus, based on phylogenetic analysis and examination of informative sites, nuclear genes appear to have a tendency to support the E + H sister group, which contradicts both the morphological and mtDNA data.

Table 1

Nuclear Sequence Data Most Likely Support the Elephant + Hyrax Sister Group


 

 

Number of Informative Sites
 
                  
 
Out-group
 
Phylogenetic Hypothesis
 
ADORA3 (107)a
 
ADRA2B (397)
 
ADRB2 (275)
 
APOB (445)
 
AQP2 (111)
 
ATP7A (225)
 
BDNF (189)
 
BRCA1 (954)
 
CNR1 (331)
 
EDG1 (326)
 
GHR (305)
 
IRBP (412)
 
PNOC (106)
 
PRNP (310)
 
RAG1 (303)
 
SCA1 (49)
 
TYR (142)
 
VWF (427)
 
ZFX (68)
 
Sum (5414)
 
Aardvark (Elephant, Sirenians) 
 (Sirenians, Hyrax) 
 (Elephant, Hyrax) 12 
Tenrec (Elephant, Sirenians)     
 (Sirenians, Hyrax)     
 (Elephant, Hyrax)     
Golden mole (Elephant, Sirenians)    
 (Sirenians, Hyrax)    
 (Elephant, Hyrax)    12 
Elephant shrew (Elephant, Sirenians)  
 (Sirenians, Hyrax)  

 
(Elephant, Hyrax)
 
0
 
0
 
0
 
1
 
0
 
0
 
0
 
4
 
0
 
0
 
0
 
1
 
1
 
0
 

 
0
 
0
 
1
 
0
 
8
 

 

 

Number of Informative Sites
 
                  
 
Out-group
 
Phylogenetic Hypothesis
 
ADORA3 (107)a
 
ADRA2B (397)
 
ADRB2 (275)
 
APOB (445)
 
AQP2 (111)
 
ATP7A (225)
 
BDNF (189)
 
BRCA1 (954)
 
CNR1 (331)
 
EDG1 (326)
 
GHR (305)
 
IRBP (412)
 
PNOC (106)
 
PRNP (310)
 
RAG1 (303)
 
SCA1 (49)
 
TYR (142)
 
VWF (427)
 
ZFX (68)
 
Sum (5414)
 
Aardvark (Elephant, Sirenians) 
 (Sirenians, Hyrax) 
 (Elephant, Hyrax) 12 
Tenrec (Elephant, Sirenians)     
 (Sirenians, Hyrax)     
 (Elephant, Hyrax)     
Golden mole (Elephant, Sirenians)    
 (Sirenians, Hyrax)    
 (Elephant, Hyrax)    12 
Elephant shrew (Elephant, Sirenians)  
 (Sirenians, Hyrax)  

 
(Elephant, Hyrax)
 
0
 
0
 
0
 
1
 
0
 
0
 
0
 
4
 
0
 
0
 
0
 
1
 
1
 
0
 

 
0
 
0
 
1
 
0
 
8
 

NOTE.—Sum of informative sites of amino acid sequences among the three phylogenetic hypotheses for every four out-group species is shown at the right.

a

Gene names and length of amino acid sequences in parentheses.

Inference of Species Tree in Paenungulata

Although monophyly of Paenungulata is confirmed here, the phylogenetic relationship among species within this clade has not been solved. Morphological data and total mtDNA analysis support the Tethytheria hypothesis (E + S sister group). As shown above, 19 amino acid sequences from nuclear genes tend to support the E + H sister group, and one SINE insertion (Hy05) supports the S + H sister group. These discordant results suggest that the species diversification event that defined the three orders of Paenungulata occurred over a relatively short evolutionary time period. To argue what the species tree is, we here assume that two successive speciation events occurred as described below. In our following discussion of the putative historical events, we refer to the more recent species divergence as the first divergence and the earlier or more ancient one as the second divergence. An ancestral population is also referred to as the first or second ancestral population according to the species divergence. The time period between the two species divergence events could be of the same order or shorter than the coalescence time in the first ancestral population, so that random sorting of alleles in the second ancestral population could cause discordance between species tree and gene tree. On the other hand, if the coalescence time in the first ancestral population is much shorter than the duration of the two successive speciation events, the resulting phylogeny could reflect the species tree. In other words, a phylogeny based on characters for which mutations have a relatively short coalescence time more likely reflects the species relationship. In this sense, it is likely that SINEs and nuclear gene sequences do not tend to support a single phylogenetic relationship because they are neutral and thus the coalescence time is expected to be 4Ne generations, where Ne is the effective population size. For mtDNA, however, the coalescence time is one-fourth of that of nuclear DNA. Thus, the mtDNA tree is more likely to reflect the species tree. Similarly, although the expectation of coalescence time of morphological data is not easy to determine, if we assume that morphological characters evolve under strong selective pressure, they also may have a shorter coalescence time than 4Ne generations. It is possible therefore that morphological data in this case also reflect the true species tree.

This discussion implies a possible scenario shown in figure 5. In the successive species divergences of the Paenungulata, Proboscidea and Sirenia may be sister groups as supported by morphological studies and mtDNA analysis. Coalescence time periods of SINE dimorphic alleles and nuclear genes are longer than the period of the two species divergences (see arrow “b” in fig. 5A), which could cause the inconsistent gene tree characteristics of the molecular phylogenetic examinations of Paenungulata.

FIG. 5.—

(A) A likely scenario of evolution of paenungulates in relation to the SINE presence/absence dimorphism at locus Hy05. “+” and “−” represent the presence and absence of a SINE allele, respectively. The AfroSINE-inserted allele remained dimorphic during continuous divergences from their last common ancestor followed by random fixation in each ancestral genome of Proboscidea (−), Sirenia (+), and Hyracoidea (+). Although morphological traits may become fixed in a population during a short evolutionary period (arrow “a”), a relatively long period is required for SINE-inserted alleles to become fixed because they may behave as neutral mutations (arrow “b”). (B) The most likely evolutionary scenario of Afrotheria suggested by the present study.

FIG. 5.—

(A) A likely scenario of evolution of paenungulates in relation to the SINE presence/absence dimorphism at locus Hy05. “+” and “−” represent the presence and absence of a SINE allele, respectively. The AfroSINE-inserted allele remained dimorphic during continuous divergences from their last common ancestor followed by random fixation in each ancestral genome of Proboscidea (−), Sirenia (+), and Hyracoidea (+). Although morphological traits may become fixed in a population during a short evolutionary period (arrow “a”), a relatively long period is required for SINE-inserted alleles to become fixed because they may behave as neutral mutations (arrow “b”). (B) The most likely evolutionary scenario of Afrotheria suggested by the present study.

Revisiting Morphological Data and the Evolution of Paenungulata

As described above, the time period between the two speciation events could be shorter than the coalescence time in the first ancestral population, suggesting that many large-scale morphological synapomorphies in Tethytheria might evolve over a very short evolutionary period. The oldest fossil Tethytheria, known as a proboscidean Phosphatherium escuilliei, was found in northern Africa (Gheerbrant, Sudre, and Cappetta 1996; Gheerbrant et al. 1998, 2003), which dates back to the early Eocene, as well as other early proboscideans, such as the middle Eocene proboscidean Numidotherium (Mahboubi et al. 1986) and late Eocene Moeritherium (Matsumoto 1923; Tassy 1981). On the other hand, the most basal sirenians are known from the early middle Eocene of Jamaica (Savage, Domning, and Thewissen 1994; Domning 2001), approximately 48–50 MYA. Sirenians of similar age are known from other parts of the world, such as Egypt (Gingerich 1992) and Indo-Pakistan (Zalmout, Ul-Haq, and Gingerich 2003). Therefore, if Tethytheria is valid, the ancestry of sirenians should be in Africa and origin of the order must be older than the earliest (Jamaican) representatives.

The oldest known fossil hyracoidean is found at Ouled Abdoun in Morocco, as well as Phosphatherium (Gheerbrant et al. 2003), indicating that early proboscideans and early hyracoids are in close geographic and temporal proximity, probably close to the origin of Tethytheria. This supports the notion that Paenungulata radiated very rapidly. Based on this overall picture, the origin of the three extant orders of Paenungulata was probably in Africa in the Paleocene or Cretaceous, and then they expanded outside of Africa. Recent molecular analysis also suggests that they diverged about 60–65 MYA (Springer et al. 2003).

All the early elephants listed above are associated with marine sediments, suggesting that they lived in near-shore marine waters. Consistently, embryological analysis of elephants suggests that they evolved from an aquatic ancestor (Gaeth, Short, and Renfree 1999). The Jamaican basal sirenians are also thought to have spent most of their time in the water (Domning 2001). Additionally, desmostylians, another extinct order of Tethytheria, also inhabited the semiaquatic environment of the North Pacific from the middle Oligocene to the late Miocene (Domning, Ray, and McKenna 1986; Inuzuka 2000). Therefore, it is reasonable to speculate that a common ancestor of Tethytheria may have lived in a semiaquatic environment, which is probably shallower water in the vicinity of the ancient Tethys Sea (Janis 1988; Shoshani and Tassy 1996). If Tethytheria is valid, as shown in figure 5A, and the synapomorphies in the ancestry of tethytheres evolved very rapidly, this may have resulted from their rapid adaptation to a semiaquatic existence.

In the present study, we showed afrotherian interordinal phylogenetic relationships as well as monophyly of Afrotheria by retroposon insertion analysis. Although monophyly of Afroinsectiphillia is supported by only one locus (INT1095-1), this conclusion can be accepted and is probably not a result of random sorting of ancestral presence/absence alleles of the locus because other recent molecular studies also support this clade. In the case of the locus Hy05, however, it is likely that this insertion reflects random lineage sorting of ancestral polymorphism because of rapid radiation of paenungulates that we demonstrated here. Generally, isolation and characterization of more than one SINE informative locus will be safer for us to induce any conclusion for species phylogeny unless the phylogeny in question is backed up by other plural phylogenetic data. Here, we discussed about the evolutionary history of mammalian diversification in ancient isolated Africa from molecular and paleontological viewpoints. Such information should prove useful for future studies concerning evolutionary history of morphological convergence, mechanism for parallel adaptation between afrotherians and other mammals, and paleontological interpretations of fossil records.

Supplementary Material

Supplementary tables 1, 2, and 3 are available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/).

Dan Graur, Associate Editor

We thank Y. Fukumoto for providing the elephant and dugong tissue samples and B. van Vuuren and S. Maree for providing the aardvark and golden mole samples. We also thank two anonymous reviewers for helpful comments and suggestions on an earlier version of this paper. This work was supported by research grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to N.O.).

References

Amrine-Madsen, H., K. P. Koepfli, R. K. Wayne, and M. S. Springer.
2003
. A new phylogenetic marker, apolipoprotein B, provides compelling evidence for eutherian relationships.
Mol. Phylogenet. Evol.
 
28
:
225
–240.
Asher, R. J.
1999
. A morphological basis for assessing the phylogeny of the “Tenrecoidea” (Mammalia, Lipotyphla).
Cladistics
 
15
:
231
–252.
Asher, R. J., M. J. Novacek, and J. H. Geisler.
2003
. Relationships of endemic African mammals and their fossil relatives based on morphological and molecular evidence.
J. Mammal. Evol.
 
10
:
131
–193.
Benton, M. J.
2004
. Vertebrate palaeontology. Blackwell, Oxford.
Butler, P. M.
1969
. Insectivores and bats from the Miocene of East Africa: new material. Pp. 1–38 in L. S. B. Leakey, ed. Fossil vertebrates of Africa. Academic Press, New York.
Domning, D. P.
2001
. The earliest known fully quadrupedal sirenian.
Nature
 
413
:
625
–627.
Domning, D. P., C. E. Ray, and M. C. McKenna.
1986
. Two new Oligocene desmostylians and a discussion of tethytherian systematics.
Smithson. Contrib. Paleobiol.
 
59
:
1
–56.
Douady, C. J., and E. J. P. Douzery.
2003
. Molecular estimation of eulipotyphlan divergence times and the evolution of “Insectivora.”
Mol. Phylogenet. Evol.
 
28
:
285
–296.
Fronicke, L., J. Wienberg, G. Stone, L. Adams, and R. Stanyon.
2003
. Towards the delineation of the ancestral eutherian genome organization: comparative genome maps of human and the African elephant (Loxodonta africana) generated by chromosome painting.
Proc. R. Soc. Lond. B Biol. Sci.
 
270
:
1331
–1340.
Gaeth, A. P., R. V. Short, and M. B. Renfree.
1999
. The developing renal, reproductive, and respiratory systems of the African elephant suggest an aquatic ancestry.
Proc. Natl. Acad. Sci. USA
 
96
:
5555
–5558.
Gheerbrant, E., J. Sudre, and H. Cappetta.
1996
. A Paleocene proboscidean from Morocco.
Nature
 
383
:
68
–71.
Gheerbrant, E., J. Sudre, H. Cappetta, and G. Bignot.
1998
. Phosphatherium escuilliei du Thanetien du Bassin des Ouled Abdoun (Maroc), plus ancien proboscidien (Mammalia) d'Afrique.
Geobios
 
30
:
247
–269.
Gheerbrant, E., J. Sudre, H. Cappetta, C. Mourer-Chauvire, E. Bourdon, M. Iarochene, M. Amaghzaz, and B. Bouya.
2003
. Les localites a mammiferes des carrieres de Grand Daoui, bassin des Ouled Abdoun, Maroc, Ypresien: premier etat des lieux.
Bull. Soc. Geol. France
 
174
:
279
–293.
Gibson, A., V. Gowri-Shankar, P. G. Higgs, and M. Rattray.
2005
. A comprehensive analysis of mammalian mitochondrial genome base composition and improved phylogenetic methods.
Mol. Biol. Evol.
 
22
:
251
–264.
Gingerich, P. D.
1992
. Marine mammals (Cetacea and Sirenia) from the Eocene of Gebel Mokattam and Fayum, Egypt.
Univ. Mich. Pap. Paleontol.
 
30
:
1
–84.
Gregory, W. K.
1910
. The orders of mammals.
Bull. Am. Mus. Nat. Hist.
 
27
:
1
–524.
Inuzuka, N.
2000
. Primitive late Oligocene desmostylians from Japan and phylogeny of Desmostylia.
Bull. Ashoro Mus. Paleontol.
 
1
:
91
–124.
Janis, C. M.
1988
. New ideas in ungulate phylogeny and evolution.
Trends Ecol. Evol.
 
3
:
291
–297.
Lavergne, A., E. Douzery, T. Stichler, F. M. Catzeflis, and M. S. Springer.
1996
. Interordinal mammalian relationships: evidence for paenungulate monophyly is provided by complete mitochondrial 12S rRNA sequences.
Mol. Phylogenet. Evol.
 
6
:
245
–258.
MacPhee, R. D. E., and M. J. Novacek.
1993
. Definition and relationships of Lipotyphla. Pp. 13–31 in F. S. Szalay, M. J. Novacek, and M. C. McKenna, eds. Mammal Phylogeny. Springer-Verlag, New York.
Madsen, O., M. Scally, C. J. Douady, D. J. Kao, R. W. DeBry, R. Adkins, H. M. Amrine, M. J. Stanhope, W. W. de Jong, and M. S. Springer.
2001
. Parallel adaptive radiations in two major clades of placental mammals.
Nature
 
409
:
610
–614.
Mahboubi, M., R. Ameur, J. Y. Crochet, and J. J. Jaeger.
1986
. El Kohol (Saharan atlas, Algeria): a new Eocene mammal locality in northwestern Africa.
Palaeontogr. Abt. A
 
192
:
15
–49.
Malia, M. J. Jr., R. M. Adkins, and M. W. Allard.
2002
. Molecular support for Afrotheria and the polyphyly of Lipotyphla based on analyses of the growth hormone receptor gene.
Mol. Phylogenet. Evol.
 
24
:
91
–101.
Matsumoto, H.
1923
. A contribution to the knowledge of Moeritherium.
Bull. Am. Mus. Nat. Hist.
 
48
:
97
–139.
McKenna, M. C.
1975
. Toward a phylogenetic classification of the Mammalia. Pp. 21–46 in W. P. Luckett and F. S. Szalay, eds. Phylogeny of the primates: a multidisciplinary approach. Plenum Press, New York.
McKenna, M. C., and S. K. Bell.
1997
. Classification of mammals above species level. Columbia University Press, New York.
Murata, S., N. Takasaki, M. Saitoh, and N. Okada.
1993
. Determination of the phylogenetic relationships among Pacific salmonids by using short interspersed elements (SINEs) as temporal landmarks of evolution.
Proc. Natl. Acad. Sci. USA
 
90
:
6995
–6999.
Murata, Y., M. Nikaido, T. Sasaki, Y. Cao, Y. Fukumoto, M. Hasegawa, and N. Okada.
2003
. Afrotherian phylogeny as inferred from complete mitochondrial genomes.
Mol. Phylogenet. Evol.
 
28
:
253
–260.
Murphy, W. J., E. Eizirik, W. E. Johnson, Y. P. Zhang, O. A. Ryder, and S. J. O'Brien.
2001
a. Molecular phylogenetics and the origins of placental mammals.
Nature
 
409
:
614
–618.
Murphy, W. J., E. Eizirik, S. J. O'Brien et al. (11 co-authors).
2001
b. Resolution of the early placental mammal radiation using Bayesian phylogenetics.
Science
 
294
:
2348
–2351.
Murphy, W. J., P. A. Pevzner, and S. J. O'Brien.
2004
. Mammalian phylogenomics comes of age.
Trends Genet.
 
20
:
631
–639.
Nikaido, M., Y. Cao, N. Okada, and M. Hasegawa.
2003
a. The phylogenetic relationships of insectivores with special reference to the lesser hedgehog tenrec as inferred from the complete sequence of their mitochondrial genome.
Genes Genet. Syst.
 
78
:
107
–112.
Nikaido, M., F. Matsuno, H. Hamilton et al. (11 co-authors).
2001
. Retroposon analysis of major cetacean lineages: the monophyly of toothed whales and the paraphyly of river dolphins.
Proc. Natl. Acad. Sci. USA
 
98
:
7384
–7389.
Nikaido, M., H. Nishihara, Y. Fukumoto, and N. Okada.
2003
b. Ancient SINEs from African endemic mammals.
Mol. Biol. Evol.
 
20
:
522
–527.
Nikaido, M., A. P. Rooney, and N. Okada.
1999
. Phylogenetic relationships among cetartiodactyls based on insertions of short and long interpersed elements: hippopotamuses are the closest extant relatives of whales.
Proc. Natl. Acad. Sci. USA
 
96
:
10261
–10266.
Novacek, M. J.
1992
a. Mammalian phylogeny: shaking the tree.
Nature
 
356
:
121
–125.
———.
1992
b. Fossils, topologies, missing data, and the higher level phylogeny of eutherian mammals.
Syst. Biol.
 
41
:
58
–73.
———.
2001
. Mammalian phylogeny: genes and supertrees.
Curr. Biol.
 
11
:
R573
–R575.
Novacek, M. J., A. R. Wyss, and M. C. McKenna.
1988
. The major groups of eutherian mammals. Pp. 31–71 in M. J. Benton, ed. The phylogeny and classification of the tetrapods, Vol. 2. Mammals. Clarendon Press, Oxford.
Okada, N.
1991
a. SINEs.
Curr. Opin. Genet. Dev.
 
1
:
498
–504.
———.
1991
b. SINEs: short interspersed repeated elements of the eukaryotic genome.
Trends Ecol. Evol.
 
6
:
358
–361.
Phillips, M. J., and D. Penny.
2003
. The root of the mammalian tree inferred from whole mitochondrial genomes.
Mol. Phylogenet. Evol.
 
28
:
171
–185.
Robinson, T. J., B. Fu, M. A. Ferguson-Smith, and F. Yang.
2004
. Cross-species chromosome painting in the golden mole and elephant-shrew: support for the mammalian clades Afrotheria and Afroinsectiphillia but not Afroinsectivora.
Proc. R. Soc. Lond. B Biol. Sci.
 
271
:
1477
–1484.
Robinson, T. J., and E. R. Seiffert.
2004
. Afrotherian origins and interrelationships: new views and future prospects.
Curr. Top. Dev. Biol.
 
63
:
37
–60.
Roca, A. L., G. K. Bar-Gal, E. Eizirik, K. M. Helgen, R. Maria, M. S. Springer, S. J. O'Brien, and W. J. Murphy.
2004
. Mesozoic origin for West Indian insectivores.
Nature
 
429
:
649
–651.
Rogers, J. H.
1985
. The origin and evolution of retroposons.
Int. Rev. Cytol.
 
93
:
187
–279.
Rokas, A., and P. W. Holland.
2000
. Rare genomic changes as a tool for phylogenetics.
Trends Ecol. Evol.
 
15
:
454
–459.
Roos, C., J. Schmitz, and H. Zischler.
2004
. Primate jumping genes elucidate strepsirrhine phylogeny.
Proc. Natl. Acad. Sci. USA
 
101
:
10650
–10654.
Salem, A. H., D. A. Ray, J. Xing, P. A. Callinan, J. S. Myers, D. J. Hedges, R. K. Garber, D. J. Witherspoon, L. B. Jorde, and M. A. Batzer.
2003
. Alu elements and hominid phylogenetics.
Proc. Natl. Acad. Sci. USA
 
100
:
12787
–12791.
Sambrook, J., E. F. Fritsch, and T. Maniatis.
1989
. Molecular cloning: a laboratory manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Savage, R. J. G., D. P. Domning, and J. G. M. Thewissen.
1994
. Fossil Sirenia of the West Atlantic and Caribbean region. V. The most primitive known sirenian Prorastomus sirenoides Owen, 1855.
J. Vertebr. Paleontol.
 
14
:
427
–449.
Schmitz, J., M. Ohme, B. Suryobroto, and H. Zischler.
2002
. The colugo (Cynocephalus variegatus, Dermoptera): the primates' gliding sister? Mol.
Biol. Evol.
 
19
:
2308
–2312.
Shedlock, A. M., and N. Okada.
2000
. SINE insertions: powerful tools for molecular systematics.
Bioessays
 
22
:
148
–160.
Shedlock, A. M., K. Takahashi, and N. Okada.
2004
. SINEs of speciation: tracking lineages with retroposons.
Trends Ecol. Evol.
 
19
:
545
–553.
Shimamura, M., H. Yasue, K. Ohshima, H. Abe, H. Kato, T. Kishiro, M. Goto, I. Munechika, and N. Okada.
1997
. Molecular evidence from retroposons that whales form a clade within even-toed ungulates.
Nature
 
388
:
666
–670.
Shoshani, J.
1992
. The controversy continues: an overview of evidence for Hyracoidea-Tethytheria affinity.
Isr. J. Zool.
 
38
:
233
–244.
———.
1993
. Hyracoidea-Tethytheria affinity based on myological data. Pp. 235–256 in F. S. Szalay, M. J. Novacek, and M. C. McKenna, eds. Mammal phylogeny. Springer-Verlag, New York.
Shoshani, J., and P. Tassy.
1996
. The Proboscidea. Oxford University Press, New York.
Simons, E. L., and P. D. Gingerich.
1974
. New carnivorous mammals from the Oligocene of Egypt.
Ann. Geol. Surv. Egypt
 
4
:
157
–166.
Simpson, G. G.
1945
. The principles of classification and a classification of mammals.
Bull. Am. Mus. Nat. Hist.
 
85
:
1
–350.
Springer, M. S., H. M. Amrine, A. Burk, and M. J. Stanhope.
1999
. Additional support for Afrotheria and Paenungulata, the performance of mitochondrial versus nuclear genes, and the impact of data partitions with heterogeneous base composition.
Syst. Biol.
 
48
:
65
–75.
Springer, M. S., G. C. Cleven, O. Madsen, W. W. de Jong, V. G. Waddell, H. M. Amrine, and M. J. Stanhope.
1997
. Endemic African mammals shake the phylogenetic tree.
Nature
 
388
:
61
–64.
Springer, M. S., W. J. Murphy, E. Eizirik, and S. J. O'Brien.
2003
. Placental mammal diversification and the Cretaceous-Tertiary boundary.
Proc. Natl. Acad. Sci. USA
 
100
:
1056
–1061.
Springer, M. S., M. J. Stanhope, O. Madsen, and W. W. de Jong.
2004
. Molecules consolidate the placental mammal tree.
Trends Ecol. Evol.
 
19
:
430
–438.
Stanhope, M. J., O. Madsen, V. G. Waddell, G. C. Cleven, W. W. de Jong, and M. S. Springer.
1998
b. Highly congruent molecular support for a diverse superordinal clade of endemic African mammals.
Mol. Phylogenet. Evol.
 
9
:
501
–508.
Stanhope, M. J., M. R. Smith, V. G. Waddell, C. A. Porter, M. S. Shivji, and M. Goodman.
1996
. Mammalian evolution and the interphotoreceptor retinoid binding protein (IRBP) gene: convincing evidence for several superordinal clades.
J. Mol. Evol.
 
43
:
83
–92.
Stanhope, M. J., V. G. Waddell, O. Madsen, W. W. de Jong, S. B. Hedges, G. C. Cleven, D. Kao, and M. S. Springer.
1998
a. Molecular evidence for multiple origins of Insectivora and for a new order of endemic African insectivore mammals.
Proc. Natl. Acad. Sci. USA
 
95
:
9967
–9972.
Svartman, M., G. Stone, J. E. Page, and R. Stanyon.
2004
. A chromosome painting test of the basal eutherian karyotype.
Chromosome Res.
 
12
:
45
–53.
Tassy, P.
1981
. Le crane de Moeritherium (Proboscidea, Mammalia) de l'Eocene de Dor El Talha (Libye) et le probleme de la classification phylogenetique du genre dans les Tethytheria McKenna, 1975. Bull. Mus. Nat. Hist. Nat. Paris, 4th series, section C3:87–147.
Tassy, P., and J. Shoshani.
1988
. The Tethytheria: elephants and their relatives. Pp. 283–315 in M. J. Benton, ed. The phylogeny and classification of the tetrapods, Vol. 2. Mammals. Clarendon Press, Oxford.
van Dijk, M. A., O. Madsen, F. Catzeflis, M. J. Stanhope, W. W. de Jong, and M. Pagel.
2001
. Protein sequence signatures support the African clade of mammals.
Proc. Natl. Acad. Sci. USA
 
98
:
188
–193.
Waddell, P. J., H. Kishino, and R. Ota.
2001
. A phylogenetic foundation for comparative mammalian genomics.
Genome Inform. Ser. Workshop Genome Inform.
 
12
:
141
–154.
Weiner, A. M., P. L. Deininger, and A. Efstratiadis.
1986
. Nonviral retroposons: genes, pseudogenes, and transposable elements generated by the reverse flow of genetic information.
Annu. Rev. Biochem.
 
55
:
631
–661.
Yang, F., E. Z. Alkalaeva, P. L. Perelman, A. T. Pardini, W. R. Harrison, P. C. O'Brien, B. Fu, A. S. Graphodatsky, M. A. Ferguson-Smith, and T. J. Robinson.
2003
. Reciprocal chromosome painting among human, aardvark, and elephant (superorder Afrotheria) reveals the likely eutherian ancestral karyotype.
Proc. Natl. Acad. Sci. USA
 
100
:
1062
–1066.
Zalmout, I. S., M. Ul-Haq, and P. D. Gingerich.
2003
. New species of protosiren (Mammalia, Sirenia) from the early middle Eocene of Balochistan (Pakistan).
Contrib. Mus. Paleontol. Univ. Mich.
 
31
:
79
–87.

Author notes

*Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, Yokohama, Japan; †Department of Biosystems Science, Graduate University for Advanced Studies (Sokendai), Kanagawa, Japan; ‡Department of Anatomy, Northeastern Ohio Universities College of Medicine; §School of Biology and Biochemistry, Queen's University of Belfast, Belfast, United Kingdom; ∥Department of Population Medicine and Diagnostic Sciences, College of Veterinary Medicine, Cornell University; and ¶Division of Speciation Mechanism, National Institute of Basic Biology, Okazaki, Japan