] Mol Evol (1991) 33:92-102
Mitochondrial DNA Evolution in Lagomorphs: Origin of Systematic Heteroplasmy and Organization of Diversity in European Rabbits Christophe Biju-Duval,' Hajer Ennafaa, 2 Nicole Dcnnebouy,' Monique Monnerot,' Frangoise Mignotte,' Ramon C. Soriguer, 3 Amel El Gaaied, 2 Ali El Hili, z and Jean-Claude Mounolou' Laborataire do Biologie Generale, Universile Paris-Sud, 91405 Orsay Cedex, France ' Faculte des sciences. Gcnctique, 1060 Tunis, Tunisie Estación Biológica de Doñana, Pabellon del Peru. Avenida Maria-Luisa, 41043 Sevilla. España
Summary. A characterization was conducted on mitochondria] DNA (mtDNA) molecules extracted separately from 107 European rabbits ( Oryctolagus cuniculus) both wild and domestic, 13 European hares (Le pus capensis), and I eastern cottontail (Sylvilagus floridanus). Experimentally this study took into account restriction site polymorphism, overall length variation of the noncoding region, and numbers of repeated sequences. Nucleotide divergences indicate that the mtDNAs from the three species derived from a common ancestor some 6-8 million years (Myr) ago. Every animal appeared heteroplasmic for a set of molecules with various lengths of the noncoding region and variable numbers of repeated sequences that contribute to them. This systematic heteroplasmy, most probably generated by a rate of localized mtDNA rearrangements high enough to counterbalance the cellular segregation of rearranged molecules, is a shared derived character of leporids. The geographic distribution of mtDNA polymorphism among wild rabbit populations over the western European basin shows that two molecular lineages are represented, one in southern Spain, the second over northern Spain, France, and Tunisia. These two lineages derived from a common ancestor some 2 Myr ago. Their present geographical distribution may be correlated to the separation of rabbits into two stocks at the time of Mindel glaciation. Finally the distribution of mtDNA diversity ex-
Offprint requests ro.-
C. Biju-Duval
hibits a mosaic pattern both at inter- and intrapopulation levels. Key words: Lagomorphs - Rabbit - Mitochondrial DNA - Heteroplasmy - Restriction site polymorphism - Evolution
Introduction The study of evolutionary relationships between individuals, populations, and species has been considerably renewed by the use of mitochondrial DNA ( mtDNA) as a marker (Wilson et al. 1985; Avisc et al. 1987; Harrison 1989). For animals, the advantage of using this informational molecule in this way stems from the constancy in number and function of the genes it carries, the lack ofduplication of these genes, the predominant maternal inheritance, and the apparent absence of recombination (Brown 1985; Harrison 1989). In addition the 37 genes of the vertebrate mitochondria] genome are closely clustered with little or no intergenic sequences (Attardi 1985; Tzagoloff and Myers 1986; Gadaleta et al. 1989). When compared in different species, the largest noncoding sequence is shown to comprise both variable and conserved domains in sequence and secondary structures (Brown et al. 1986; Mignotte et al. 1987; Saccone et al. 1987). Consequently, sequence comparisons between mitochondrial genes or the conserved regions of the large noncoding sequence provide information about the evolutionary relationships between species or distantly related
93
populations. Comparisons between the most variable positions whatever their location enable the study of relations and links between populations or even individuals (Avise et al. 1987, 1989). Moreover, in mammals the resolution of these DNA studies is increased by a faster accumulation of substitutions in the mitochondrial genome than in nuclear genes (Brown et al. 1979; Wilson et al. 1985). These valuable properties of mtDNA prompted us to investigate its evolution in lagomorphs and especially in European rabbits where several biological questions are pending. First, according to paleontological data, the genus Oryctolagus has not given rise to an abundant radiative speciation and it is now represented by a unique species, Oryctolagus cuniculus. We wondered consequently if in this evolution mitochondrial genetic diversity has been of limited extent or has evolved as in other genera. Moreover, the question of the origin of this genus and its relationship to Lepus and Sylvilagus is a matter of debate (Dawson 1981). We have consequently used a classical examination of nucleotide divergence between the mtDNAs of various European rabbits, hares, and a cottontail to get a new insight into their relatedness. The second goal of our study was to examine geographical organization of mtDNA diversity among wild rabbit populations distributed over western Europe. It has even been suggested that the animals of southern Spain could be the progeny of the rabbits that were pushed back in that region at the time of glaciations and that wild populations found over France would descend from a population that took refuge in southern France (Provence) and was separated by the ice cover of the Pyrenees from the first stock (Beaucournu L980; for review see Arthur 1989). Immunological data (Van der Loo et al. 1990) are consistent with this view. We took advantage of mtDNA to get more precise information about the history of expansion in western Europe after the ice age (including human interference). Finally, previous studies had demonstrated that rabbits apparently escape the segregation rule of ani mal mitochondria] genetics. Usually when mutation creates heteroplasmy in a cell, stochastic sampling at cell division progressively resolves it back to homoplasmy (Birky 1983; Birky et al. 1983; Solignac et al. 1984). Ennafaa et al. (1987) showed that every rabbit is heteroplasmic, carrying an array of mtDNA molecules differing by the length of the large noncoding region. This situation can only be the result of some events or processes that generate diversity to counteract segregation and thus perpetuate the heteroplasmic state [see Monnerot et al. (1986) for a discussion of a similar situation in Rana esculenta]. Some molecular understanding emerged from the sequencing of the Large noncoding region
Table 1.
Origin and number of rabbits
Origin
Locality or breed
Number of individuals
Wild rabbits France Ile de France Pays de Loire Camargue Spain Navarra
Versailles Cerizay La Tour du Valat
3 3 33
Andalusia Tunisia Tunis gulf Domestic rabbits [nstitut Pasteur 1NRA, Jouy en Josas
Sesma Tudela Arroniz Las Lomas
] 2 30
Zambia Island
17
Fauve de Bourgogne New Zealand
14 3
of rabbit mtDNA. Mignotte et al. (1990a,b) found that the variable domains of this sequence comprised two kinds of repeated sequences (SR and LR). Variations in numbers of both repeats from one molecule to another contribute to length differences and thus to the observed heteroplasmy. We were first interested in knowing whether this property was indeed shared by all wild rabbits of different origins and lineages. The observation of generalized heteroplasmy in Oryctolagus subsequently prompted us to ask if this property is specific to this genus or if it constitutes a derived character of the leporids (European rabbit, hare, and cottontail).
Materials and Methods A nimals Oryctolagus cuniculus. Wild animals were trapped in localities
where rabbits have not been recently imported for game interest. Domestic individuals were provided by ]nstitut Pasteur and by )nstitut National de la Recherche Agronomique (INRA). Table I indicates the numbers and origins of all the animals under study.
Lepus capensis (europaeus). The 13 hares came from a stock of INRA, which was built up with animals from Czechoslovakia and France. Sylvilagus fioridanus. The cottontail rabbit was provided by a private farm in France.
Techniques Mitochondria] DNA Extraction. Mitochondnal DNA was extracted usually from kidney or liver of freshly killed animals following Ennafaa et al. (1987). In brief, a mitochondria] fraction was first isolated by differential centrifugations in sucrose gra-
94 Table 2.
Numerical maps of mitochondrial DNAs
Restriction enzyme Aval
BainHI
Lo I 0.9 5.5 6.2 11.1 l1.3 12.4 14.8 15.8 3.1 4.2
Fbl
Bell
Clal
+
+
-e
+
+
EcoRl
+
+
1
+ +
+ +
+
+-
+
+
+
+ + + +
10.3 14.1
+
PsI
1.9
5.9 10.1 12.2
PvuII
+ +
5.8 6.4 9.3 2.6
+ +
+
+ + +
+
+ +
+
+
7.0 9.2
+ + + +
+ + + +
+
+
+
+
+ + +
3.2
2.3 3.7 6.7 7.6 8.5 9.6
+
+
6.3 Xbal
+
+
+
+
+
+ + + + +
+ +
+ +
+ +
+ +
13.0 .SsI
+
Sy
+
+
+
+
+ + +
6.6 8.6
+
+ + +
+ +
5.1
Hpal
+
+
Lc
+
+
+
Fb I
+
l 1.0
+
+ + +
Lo 1
+ + + + + +
+ +
0.0 0.6 1.8 2.0 3.5 5.1 5.5 6.9
7.2 8.6
+
+ +
12.1 14.9
+
+ +
+
10.1 1 0.6 1 L2 11.9
8.5
+
+ +
+ +
9.9 1.7 3.4
HinduT!
+
5.3 5.7 7.7
0.0 0.6 9.3
Sy
+
5.0 9.3 10.5 14.6 15.7 2.3 3.0 3.4 4.2
Le
Restriclion enzyme
+ +
+
+ +
+ +
+ +
+
+
+
+ +
+ + +
+ + +
Hare (Le), cottontail {Sy), one wild rabbit from southern Spain (Lot), and one domestic rabbit (Fbl). The numbers indicate the positions (in kb) of the sites along the molecule; the numbering begins at the conserved Avail site, located in the 12S rRNA gene
clients; after lysis of the mitochondria, DNA was purified by centrifugation in cesium chloride.
Detection of Repeated Sequences. Individual mtDNA extracts were digested by IIinf, electrophoresis was performed on 3% Nusieve agarose gels; DNA was transferred on Hybond-N (Amersham) membranes, then hybridized with the two probes pi 1 L and p l I S (Mignotte et al. 1990b), covering, respectively, the large and short repeats (LR and SR),
Restriction Site A nalysis. Usually 14 restriction endonucleases (Avai, Avail, BamHl, BclI, Bglll, Clal, EcoRl, ltindlll, Hpal, Psil, Fvull, Ssll, SIDI, Xbal) were used, however Avail, BgtII, and Steil results are not presented for hare and cottontail. After digestion, performed according to supplier's instructions. mtDNA fragments were end-labeled with a mixture ofall four a-"P dNTPs (Wright et al. 1983), and separated by electrophoresis on l % agarose gels (eventually on 5% polyacrylamide gels). Estimates of Nucleotide Divergence. The maps were aligned by the best adjustment of various-sized fragments (cf. Results) and shared fragments. Each site was attributed a coordinate,
which is its distance (in kb) to the conserved Avail site, located in the 12S rRNA gene (Attardi 1985). The alignment procedure in the mapping is based on parallel migrations and double digestions. It allows one to decide whether two sites the coordinates of which differ by less than 0.3 kb are located at identical positions (sites shared by two molecules) or not. Nucleotide divergence and nucleotide diversity were estimated following Nei and Li (1979), and Nei (1987).
Results Restriction Site Mapping and Nucleotide Divergence of Leporid mtDNAs The restriction map of domestic rabbit mtDNA (Fbl) previously published by Ennafaa et al. (1987) was enlarged with sites for two additional enzymes (Avail and Steil). Because wild rabbits from Las Lomas (Andalusia)
95
Fig. I. Phylogenetic tree of Oryctolagus, Lepus, and Sylvilagus mtDNAs. This UPGMAphenogram illustrates Table 3. The time scale is based on the assumption that mtDNA nucleotide divergence occurs in leporids at a mean rate of 2% per million years.
exhibited mtDNA digestion profiles very different from that already obtained with Fbl, a map was independently established for one animal taken as reference (Lol). Restriction site maps were also established for hare (Le) and cottontail (Sy) m1DNAs. Table 2 presents the 4 maps for the 11 endonucleases after alignment of shared sites. Each map presents an average of 35 sites of 6 nucleotides each, representing a set of about 200 nucleotides (1.1% of the complete genome). Comparisons of these four maps allowed an estimation of nucleotide divergence between molecules taken pairwise (Table 3). The resulting distance matrix was employed to build a UPGMA (unweighted pairgroup method using arithmetic averages) phenogram (Fig. 1): the two Oryctolagus mtDNA types (Lot and Fb1), although noticeably divergent, are more closely related to each other than to Sylvilagus and Lepus mtDNAs.
Organization of Diversity in European Rabbits For O. runiculus, mtDNAs of 90 wild animals from eight localities, and of 17 domestic animals from two breeds (Fauve de Bourgogne and New Zealand) were analyzed. On the whole, 70 sites were recognized by the 14 endonucleases used, and 37 of them are polymorphic. All of these sites have been mapped; they are spread out all along the molecule. Their distributions allowed the definition of 17 mtDNA types. Table 4 presents all the sites mapped with their positions on the molecules and their occurrence in the different mtDNA types. Table 5 indicates how these mtDNA types are represented within the populations surveyed. From these data a parsimonious network was established connecting the 17 mtDNA types with a minimum number of mutations (Fig. 2). Two sets are clearly separated: one with the seven mtDNA types detected in a population of southern Spain (Las Lomas) and the other
Fig. 2. Ph ylogenetic relationships between the 17 mtDNA types. A parsimonious network interconnecting all The mtDNA types was built from Table 4 data. Each circle represents one mtDNA type, with a size proportional to the number of individuals. Black points represent potential intermediates. The two main groups of mtDNA types are separated by at least 17 site changes.
Table 3. Nucleotide divergence between Lepus, Sylvilagus, and two types of Oryctolagus mtDNA
Lo 1 Fb l Le Sy
Lot
Fb I
Le
Sy
37 4.2 13,2 12.2
28 35 15.1 14.2
16 14 33 11.5
17 15 17 34
Abbreviations are as in Table 2. Data were obtained by comparison of four restriction maps (Table 2) for 11 restriction endonucleases (10 with r = 6, 1 with r = 16/3). Total numbers of examined sites are given on the diagonal, numbers of shared sites are above, and nucleotide divergence (%) is below
one with all the mtDNA types observed in populations from France, Tunisia, northern Spain, as well as in domestic animals. In three localities, the sample size was large enough to allow the estimation of within-population diversity. In Las Lomas (Spain) among 30 rabbits seven different mtDNA types are found; the nucleotide diversity is 0.4%. In La Tour du Valat (France) 32 rabbits exhibited the same mtDNA type and 1 individual was heteroplasmic (Tvl-Tv2): an additional site (Bel 14.2) was found in a fraction of its molecules. The nucleotide diversity can be roughly estimated to be 0.01%, and this is probably overestimated. On Zembra Island (Tunisia) I rabbit among 17 carried a mtDNA type with an additional site (Sat 13.9). The nucleotide diversity is about 0.02%.
Length Heteroplasmy in the Family Leporidae For rabbit mtDNA the detection of length heteroplasmy is particularly easy when restriction fragments including the variable region (namely the noncoding region) are smaller than 5 kb. The variability appears, on gels, as a fuzzy area instead of a well-defined band (Ennafaa et al. 1987). The survey
96 Table 4.
Distribution of the 17 mtDNA types in the diverse localities and breeds mtDNA types
Origin Spain Las Lomas Sesma Tudela Arroniz
Fb
Tv
Lo 1
2
3
4
5
6
7
10
8
5
2
2
2
1
France La Tour du Valat Cerizay Versailles Fauve de Bourgogne (d) New Zealand (d) Tunisia Zembra
Se
Tu
Az
1
2
33
l`
Ce
1
2
Ze 1
2
16
1
1 2 1
1
2 3
13
1
3
Each mtDNA type is noted according to the locality or breed where it was described first accompanied by a number if there is more than one type in the locality or in the breed. The table indicates the number of individuals bearing the mtDNA type. In La Tour du Valat, one individual (") among the 33 exhibits the two different mtDNA types Tv I and Tv2. (d), domestic stocks.
Length heterogeneity in different digest patterns. Individual mtDNA extracts (each noted by its mtDNA type, see Table 4) were digested with different endonucleases (Avail for four rabbits. Aral for three hares, and F.coltl for the sole cottontail). In each lane one band (*) corresponding to the fragment including the noncoding region appears fuzzy and its spreading pattern varies from one individual to the other. Fig. 3.
of mtDNA polymorphism in Leporidae gave us many opportunities for such observations. Intraindividual length heterogeneity has also been shown to be correlated in domestic rabbits with the exis-
tence of a variable number of two types of repeatsLR, 153 bp and SR, 20 bp (Mignotte et al. 1990b). Although this approach in characterizing heteroplasmy provides greater resolution at the molecular level, it may not detect other possibilities of length variation and does not allow an appraisal of the range of overall size differences. Figure 3 illustrates the results obtained from the first approach: every individual mtDNA from Oryctolagus, Lepus, or Sylvilagus appears heterogeneous in size. The lengths of the longest and the shortest molecules in an individual extract constitute a characteristic of the animal. In Oryctolagus the size of mtDNA molecules varies between 16.9 and 17.9 kb; the maximal difference within an individual being around 0.6 kb. For Lepus (animals issued from the same breeding), the sizes range from 17.4 to L8.2 kb, and for the sole Sylvilagus studied 17.2 and 17.9 kb are the extremes. As illustrated in Fig. 4 every individual extract hybridizes to both p l IL and p l 1S. Furthermore the fragments that are homologous to these probes exhibit a pattern similar to that of domestic rabbits previously observed ( Mignotte et al. 1990b) and also presented in that figure. Fragments hybridizing with the p1 1S probe vary in size by a few nucleotides and this is indicative of the presence of short repeated sequences. Fragments hybridizing with p11 L probe vary in size by some 150 bp, and this signals the occurrence of larger repeated sequences. These two types of fragments are characterized by their respective numbers of repeats subsequently named SR and LR by analogy with the rabbit situation: when DNAs are taken
97 as a whole the number of repeats varies from 6 to 20 for the SR and from 4 to 8 for the LR, but each animal has its own qualitative and quantitative combinations of both SR and LR repeats. Thus, mtDNA length heteroplasmy in Lepus, Sylvilagus, and wild Oryctolagus involves, as it does in domestic rabbits, variation of repeat numbers.
Discussion Mitochondrial DNA Length Heteroplasmy within Lagomorphs Mutations generate genetic diversity among the several hundreds or thousands (Hauswirth and Laipis 1985; Veltri et al. 1990) of mtDNA molecules present in a cell. This molecular heterogeneity called heteroplasmy is usually transient as mitochondria and consequently mtDNA molecules are sorted out at cell divisions by repeated stochastic samplings. However, the experimental observation of heterogeneity among the molecules extracted from a tissue, an organ, or an organism reveals the occurrence of a heteroplasmic situation at some stages in the cellular lineage that gave rise to this tissue, organ, or organism. Thus, the study of mtDNA heterogeneity allows an analysis ofheteroplasmy and its resolution in the progeny of the heteroplasmic animal. How long this state is perpetuated through cell and animal generations depends on the sampling system as well as on selective process (Solignac et al. 1987; de Stordeur et at. 1989). In Drosophila, Gryllus, and Romanomermis it is known to be carried over for many generations (Solignac et al. 1984; Rand and Harrison 1986; Hyman and Slater 1990) and in mammals to be segregated in a few (Hauswirth and Laipis 1982). In the last case Hauswirth and Laipis (1985) suggested that some bottleneck in the transmission of mitochondria] genomes during oogenesis and/or early development rapidly drives the cells to homoplasmy. Contrary to this situation Ennafaa et at. (1987) showed that each rabbit sampled in a Tunisian wild population or a domestic stock harbors a heterogeneous population of mtDNA molecules. This diversity is due to length changes of the noncoding region of the mtDNA and each individual can be characterized by the sizes of its shortest and longest observable molecules and the distribution pattern of molecules in between. Later, Mignotte et al. (I 990a,b) demonstrated that variations in the numbers of two types of repeated sequences are basically responsible for the observed mtDNA length differences in the rabbits they studied. The present paper reports the extension of the previous observations to animals of very different geographic origins and confirms that individual heteroplasmy is
Length differences in some Hiatt fragments of mtDNA from different animals. Each lane corresponds to an individual mtDNA extract whose mtDNA type (see Table 4) is noted above. Hybridizations were performed with two different probes from cloned domestic rabbit mtDNA fragments, one (p l IS) containing the small repeats (SR), the other (p11 L) containing the large ones (LR) ( Mignotte et al. 199ob).1n every lane, each probe hybridizes with fragments of different sizes. Fragments hybridizing with pI IS vary by a pace of a few base pairs. The size differences between successive fragments hybridizing with pl1L are about 150 base pairs. Numbers of both repeats are deduced from the comparison with Fb mtDNA and are indicated on the right, Fig.4.
of systematic occurrence in Oryctolagus. In principle this situation implies either a high mutation rate (Clark 1988; Birky et al. 1989) in the noncoding region or a very slow mtDNA purification (or both). By now there are no data to support the second proposal. This is confirmed by the rarity of heteroplasmy judged through restriction site variability in the coding domain of molecules of the same mtDNA extracts. Thus it seems reasonable to consider systematic heteroplasmy as a consequence of a high rate of molecular rearrangements strictly localized in the noncoding region of the mitochondrial genome. A similar situation has already been documented and discussed in the case of Rana esculenta ( Monnerot et al. 1986). The examination of Lepus and Sylvilagus mtDNAs indicates that heteroplasmy due to length variations of the noncoding region and involving repeated sequences is also observed. Although only one cottontail was examined, 13 hares were analyzed, and we suggest that in this species heteroplasmy may also be of general occurrence. In consequence, systematic heteroplasmy is a shared derived character of leporids. Because it is not present in the nearest groups (rodents, primates) we suggest that both mtDNA rearrangements and the nuclear-encoded potential to promote them if any
98 'fable 5.
Numerical maps of the 17 rabbit m1PNA types. Continued on page 99
Awl]
Avail
Tv
Lo
Restriction enzyme 5.5 11.1 1 1.3 12.4 15.8 0.0 0.8
1
2
3
4
5
6
7
+ + +
+ +
+ + +
+ +
+ + +
+ + +
+ + +
+ +
+ +
+ +
t
+
+
+
+
+
BamHl
Bell
BgllI
Clal EcoRl
2
Ce
1
Hpal
Pvutl
Ss'FI
5'rul
2 +
+
+
f
+
+
+ +
+ +
+ +
+
+
+
+
+
+
+
+
+
+
+
+
+ + + +
+
+
+
+
+
+
+
+
+
+ +
+ +
+ +
+ +
+ +
+
+
+ +
A+
+
+
+
+
+ +
+
+
+ +
+
+
4-
4
+
+
+
+ +
+ +
+ +
+ +
+ +
+ +
+
2.3 3.0 5.3 5.7 10.1
+ +
+ +
+ +
+ + +
+ + +
[ 0.6 11.7 14.2
+
+
+ + + +
1.7 7.3 12.8 13.1 15.0 0.6
+
+ +
+
+
+
4-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ + + + + + +
+ +
+
+
+ + + + + +
+ + + +
+ + +
+ + +
+ + + + +
+ + +
+ + +
+ + +
+ +
+
+
+ + + +
+ +
+ +
+
+
+
+
+ + +
4-
+
+
+
+
+
+
+
+
+
+
+
+
+
+ +
+ +
+ +
+ +
+
+
+
4-
+
+
+
-F
+
+
+
+
+
+
+ +
+ +
+ +
+ +
+
+
+
+
+ + + +
4-
+
+
-F
+ + + + + +
+ + + + +
+ + + + +
+ + + + +
+ + + + +
+ + + + +
+
+
+
+
+
+ + + +
+ + +
+ + +
+
+
+
+
+ +
+ +
+ +
+ +
+
-F
+
+
+
+
+
+
+
+
+ +
+
-F
+ +
+
+
+
+
+
+
+
+
+
+
+
4-
+
+
+
+
+
+
+
+
+
+ +
+ +
+ +
+ +
+
+
+
+
+ +
+ +
+ +
+ +
+ + +
-F
-1-
+
+ +
+ +
+ +
+ + +
+ +
0.6 1.8 7.0
+ +
+ +
+ 4-
5.1 6.9 8.6 11.0 1.9
+
+ + +
4-
+
+
+
+
4-
+
+
+
+
4-
+
+
+
-F
+ +
+ +
+ +
-F
+
+
+
+
+
+
+
+
+
+
+
+ +
+
+
+
+
+
+ + + +
+ +
+
+ + + +
+ +
+
+ + + +
+ +
4-
+ + + +
+ + +
+
+
+
+
+
+
+
+
+ + +
-F
+
+
+
+
+
+
+
+
+
+
+ + +
4-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ + + +
+
+
+
+
+
+
+
+
+
+
1 4.1
-F
+
+
4-
+
+
+
+
+
+
-F
+
+
+
+
+
+
5.9 12.2 5.8 9.3
-F
+
+
+
+
+
+
+
+
+
+
+
+
+
t
+ +
+ +
+ +
+ +
+ +
+ +
+
+
+
+
4-
+
+
+
+
+
+
+
+
+ + +
+ + + +
+ +
5.1 6.6 8.6 10.3 Psi].
1
+ + +
15.6 Hindlll
7-e
2
A-
+
15.0 3.1 9.3 1 0.5 15.7
9.9 3.4 8.5 1 2.1 14 .9
1
Fb
Tu
+
2.7 5.9 8.9 9.2
Az
-
Se
+
+
+
+
+
+
+
4-
+
+
+
+
+
+
+
+ +
+ +
+ +
+
+
+
+
+
+
+
+
+ +
+
13.0 2.6 3.2 6.3 13.9
+ + +
z,z
+
+
+
+
+
+
+
+
4.6 8.2
+
+ + +
+ + +
+
-
-
+
+
-
+
+
+
+
4-
+
-
-
-
+
+
-
+
+
+
+
+
+ + +
+
8.7 9.6
+ + +
+
+
-
-
-
-
+
+
+
+
4-
+
+
-
-
-
-
+
+
+
+
+ +
+ +
+ +
+ +
+ + +
4-
+
+
+
+
+
+
-F
+
+
+
+
+
+
+
4-
+ +
+ +
99 Table 5.
Continued
Xbal
Tv
Lo
Restriction enzyme 13.3 14.1 0.0 2.6 3.7 6.7 8.5
Z.c
Fb
1
2
3
4
5
6
7
Sc
Tu
Az
I
2
Ce
1
2
1
2
+ +
+ +
+
+
I +
+ +
+ +
-
-
-
+ + +
+ + +
+
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+
+ +
+
+
+
+
+
+
+
+
+
+
+
+
+
+
The numbers indicate the positions (in kb) of the sites along the molecule; the numbering begins at the conserved Avail site, located in the 12S rRNA gene. One Sstt polymorphic site has not been mapped yet (noted x.x), and some information is still lacking for SLUT
originated and/or developed in a common ancestor of leporids [before 6-8 millon year (Myr)]. A study of the other family of the order, namely the ochotonids, would be interesting in order to know whether this character is common to all lagomorphs and traces back to their early differentiation. In general we would like to know whether in some lagomorph species mtDNA does not exhibit any repeated sequence or only one kind of repeats. Finally, the question of whether the presence of repeats automatically leads to systematic heteroplasmy could be asked. Relationships between the Three Genera Lepus, Oryctolagus, and Sylvilagus In the phylogenctic tree presented in Fig. 1 Lepus and Sylvilagus mtDNAs appear to have diverged slightly more recently than both did from Oryctolagus mtDNA. Uncertainties on nucleotide divergence calculations leave, however, the relative position of the two branching points open to criticism. More restriction sites and the introduction of an outer genus (possibly taken in the ochotonid family) would be necessary to confirm the true mtDNA phylogeny. Without additional results we consider that Oryctolagus, Sylvilagus, and Lepus mtDNAs diverged almost at the same time. Evolutionary relationships within leporids have not been clearly established in detail (Dawson 1981). Classically, hares (21 species of the genus Lepus) are morphologically distinguished from all other leporids distributed among nine genera and collectively named rabbits. This separation is substantiated by studies on chromosome numbers (Robinson et al. 1983). However, the consideration of different specific characters can lead to other proposals. For instance, in contrast to the other leporids, Oryctolagus and Lepus share a small quantity of pericentromeric heterochromatin (Robinson et al. 1983). Comparing fossil and present dental characteristics, Hibbard (1963) proposed that the three genera (Lepus, Or-
yctolagus, and Sylvilagus) have been simultaneously separated from a unique common ancestor. Our data are consistent with this hypothesis. If the estimate of mtDNA divergence upon time determined for other mammalian species (2% per Myr: Brown et al. 1979; Wilson et al. 1985) stands true for leporids, the molecules representative of the three genera, Lepus, Oryctolagus, and Sylvilagus, diverged 6-8 Myr ago. Different views on the date of emergence of the leporids have been presented in the literature. Hibbard (1963) has rooted the branches leading to these genera in the Pleistocene, less than 2 myr ago. However the differentiation of the genus Oryctolagus could have taken place much earlier as the first recognizable Oryctolagus partial fossils are estimated to be 6-6.5 Myr old (LopezMartinez 1977a). Our data on mtDNA are consistent with the latter possibility although they do not prove it. Diversity in European Wild Rabbits Two striking facts emerge from the present survey of European wild rabbit diversity: (1) the occurrence of two clearly divergent mtDNA lineages and their distribution in two separated geographic areas; and (2) the relatively high value of divergence between two mtDNAs (Lo I and FbI) representative of the two sets of molecules, 4%. Assuming as before a mtDNA divergence rate of 2% per Myr, the two molecular lineages diverged 2 Myr ago. At that time according to paleontological data (Lopez et al. 1976; Lopez-Martinez 1977a,b) two species distinguishable by skeletal characteristics coexisted. One, Oryctolagus lacosti, had settled over Spain and southern France. The second, Oryetolagus laynensis, was found only in Spain. Present European rabbits are believed to have speciated from O. laynensis in Spain 0.9 Myr ago. Because the two molecular lineages we find today diverged some 2 Myr ago O. cuniculus speciation must have occurred in populations with noticeable mtDNA di-
100 versity. Then one group of animals defined by the population of Las Lomas (Andalusia, southern Spain) inherited only mtDNA molecules of the first lineage. Animals found in all other populations (northern Spain, France, Tunisia) belong to the other group and share mtDNAs only derived from the second molecular lineage. The occurrence of two sets of O. cuniculus was first postulated by Beaucournu (1980) when he realized that rabbits from southern Spain and from southern France suffer specifically from ectoparasites ofdifferent species. More recently, Van der Loo et al. (1990) in their immunoglobin allele diversity survey demonstrated that although wild rabbits share some alleles across Europe, the population of Andalusia, Portugal, and the Azores can be distinguished from all the others by specific alleles. On the whole, and because the same Las Lomas population has been studied by Van der loo et al. and us, the three different approaches (ectoparasites, immunoglobin genes, mtDNA) are consistent. All the results indicate a genetic and geographic differentiation between the two groups of rabbits. It is tempting to relate this genetic distinction to that based on morphological traits of two subspecies: O. cuniculus cuniculus and O. cuniculus algirus (Saint-Girons 1973). At this point the only argument in favor of a correspondance between the two classifications is the geographic location. Indeed O. cuniculus algirus is reported in Spain and northern Africa, and O. cuniculus cuniculus in France. Whatever the relation of the two mtDNA lineages to any subspecies, the origin of their separated location can be explained through the following model proposed by Beaucournu (1980). At the time of glaciations rabbits took refuge in two areas: southern Spain and southeastern France. Southern Spain is still occupied by the progeny of the first stock when the second one spread and populated France (and northern Spain?). The estimated divergence time (2 Myr) between the mtDNAs of the present populations in these areas suggests that the separation of the two stocks could have occurred at the time of Mindel glaciation or even before. A different and interesting insight on rabbit social biology is given by the examination of the distribution of mtDNA diversity in northern Spain and France populations. Although all the mtDNAs belong to the same molecular lineage each mtDNA type (Az, Ce, Se, Tu, and Tv) is found in only one locality with the exception of a molecular form (Fbl ), which was detected in two domestic and two wild stocks. These results suggest that at least for mtDNA transmission wild rabbit populations are organized as a mosaic implying the absence of female exchanges, a situation classified as category III of the phy-
logeographic patterns proposed by Avise et al. (1987): phylogenetic continuity, spatial separation. The identity of mtDNA type detected in rabbits from Versailles with the molecular form found in domestic stock (Fbl) indicates that animals taken for creating this population in the X V I century have the same matriarchal origin as rabbits domesticated two to three centuries later. This and the existence of a close relationship between mtDNA types, respectively, found in animals from Zembra (introduced by man more than 2000 years ago, see below) and in domestic stocks suggest that man has sampled several times in the same pool during the last thousand years. Intrapopulation samplings have been large enough in Zembra (Tunisia), Las Lomas (Spain), and La Tour du Valat (France) to analyze the local mtDNA diversity. This study reveals two clearcut situations. In Las Lomas the diversity is noticeable (0.4%) and according to the assumptions discussed previously the last common ancestor to the seven mtDNA types observed today existed some 200,000 years ago. This indicates that neither the Wurm glaciation (30,00070,000 years ago), nor recurrent myxomatosis, nor humans or any other predator inflicted severe reductions in the effective population size and created bottleneck effects in its evolution. The Las Lomas population in Spain is up to now the only one that has retained traces of molecular evolution and maintained mtDNA polymorphism. Living habits of rabbits (Garson 1981; Soriguer 1983) may have contributed to this situation and their effects on mtDNA polymorphism will be investigated. In contrast, the two other populations (Zembra, Tunisia, and La Tour du Valat, France) appear quite homogeneous (mtDNA diversity around 0.010.02%). In Zembra (Tunisia) the observation is consistent with a recent introduction of rabbits on the island probably by humans when they dispersed rabbits from Spain over the Mediterranean world as reported historically (Bodson 1978). Indeed present archaeological data (Vigne 1988) have documented the presence of rabbits on Zembra island in Roman times (2000 years ago). As discussed previously southern France was a refuge area at glaciation times and we expected to find in this area enough mtDNA diversity to account for the polymorphism of populations that emerged from it (Arthur 1989). Consequently the quasimonomorphism of the population of La Tour du Valat (France) has come as a surprise. Remains of the original polymorphism may not exist any longer there as man and diseases have profoundly modified all of this Mediterranean area. But an image of the original polymorphism of the Provencal refuge area may still be found in the various mtDNA types
101
probably spread out following the dispersal of rabbits after the last glaciation and presently located in different places in France and northern Spain. In conclusion our data first support the differentiation of two mtDNA lineages in O. cuniculus and suggest that humans have sampled only in one of them to create domestic stocks. Second, the mosaic pattern of mtDNA polymorphism distribution shows that rabbit populations are well separated at least for mitochondrial transmission and female exchanges. A more extensive study in intrapopulation diversity and a survey of new populations in search of new molecular lineages and mixed situations are now awaited for to test these speculations. Acknowledgments. We thank C. Arthur, J. Aubineau, M. Carpentier, O. Ceballos-Ruiz, B. Morel, and P. Van de Waite for their help in providing animals in France and northern Spain. We are indebted to Mr. Jose Ramon Mora Figueroa and his son Fernando for lodging and sampling facilities in Las Lomas. This work was supported by a NATO grant (no. 0336/87) and carried on with CNRS, MEN, and university facilities.
References Arthur CP (1989) Origine et histoire du lapin. Bull Mens Off Nation Chasse 135:13-21 Anardi G (1985) Animal mitochondrial DNA: an extreme example of genetic economy. lot Rev Cytol 93:93-145 Avise JC, Arnold J, Ball RM, Bermingham E, Lamb T, Neigel JE, Reeb CA, Saunders NC (1987) Intraspecific phylogcography: the mitochondrial bridge between population genetics and systematics. Annu Rev Ecol Syst 18:489-522 Avise JC, Bowen BW, Lamb T (1989) DNA fingerprints from hypervariablemitochondrial genotypes. Mol Biol Evol 6:258264 Beaucoumu JC (1980) Les ectoparasites du lapin de garenne, Oryctolagus cuniculus: apports 8 son histoire. Bull Mens Off Nation Chasse, N° Sp Scien Tech, Dec 1980, pp 23-36 Birky CW Jr (1983) The partitioning of cytoplasmic organelles at cell division. Int Rev Cytol [Suppl] 15:49-89 Birky CW Jr, Maruyama T, Fuerst P (1983) An approach to population and evolutionary genetics theory for genes in mitochondria and chloroplasts, and some results. Genetics 103: 513-527 Birky CW Jr, Fuerst P, Maruyama T (1989) Organelle gene diversity under migration, mutation, and drift: equilibrium expectation, aproach to equilibrium, effects an heteroplasmic cells, and comparison to nuclear genes. Genetics 121:613627 Hodson L (1978) Donnces antiques de zoogeographie. L'expansion des Lkporidcs clans la mediterranee classique. Les Naturalistes Helps 59:66-81 Brown GG, Gadaleta G, Pepe G, Saccone C, Sbisa E (1986) Structural conservation in the D-loop-containing region of vertebrate mitochondrial DNA. J Mol Biol 192:503-51 ] Brown WM (1985) The mitochondrial genomes of animals. In: Maclntyre RD (ed) Molecular evolutionary genetics. Plenum, New York, pp 85-130 Brown WM, George M, Wilson AC (1979) Rapid evolution of mitochondrial DNA. Proc Natl Acad Sci USA 76:1967-1971
Clark AG (1988) Deterministic theory of heteroplasmy. Evolution 42:621-626 Dawson MR (1981) Evolution of the modern lagomorphs. In: Myers K, Maclnnes CD (eds) Proceedings of the world lagomorph conference. University Guelph Press, Guelph, Ontario, pp 1-8 de Stordeur E, Solignac M, Monnerot M. Mounolou JC (1989) The generation of tamsplasmic Drosophila simulans by cytoplasmic injection: effects of segregation and selection on the perpetuation of mitochondrial DNA heteroplasmy. Mol Gen Genet 220:127-132 Ennafaa H, Monnerot M, El Gaaied A, Mounolou JC (1987) Rabbit mitochondrial DNA: preliminary comparisons between some domestic and wild animals. Genet Sel Evol l9: 279-288 Gadaleta G, Pepe G, De Candia G, Quagliariello C, Sbisa E, Saccone C (1989) The complete nucleotide sequence of the Rattus norvegicus mitochondrial genome: cryptic signals revealed by comparative analysis between vertebrates. J Mol Evol 28:497-516 Garson PJ (1981) Social organization and reproduction in the rabbit: a reviw. In: Myers K, Maclnnes CD (eds) Proceedings of the world lagomorph conference. University Guelph Press, Guelph, Ontario, pp 256-270 Harrison RG (1989) Mitochondrial DNA as a genetic marker in population and evolutionary biology. Trends Ecol Evol 4: 6-11 Hauswirth WW, Laipis PJ (1982) Mitochondria! DNA polymorphism in a maternal lineage of holstein cow. Proc Nat] Acad Sci USA 79:4686-4690 Hauswirth WW, Laipis PJ (1985) Transmission genetics of mammalian mitochondria: a molecular mode] and experimental evidence. In: Quagliariello E, Slater EC, Palmieri F, Saccone C, Kroon AM (eds) Achievements and perspectives of mitochondrial research, vol ll. Elsevier, Amsterdam. pp 49-59 Hibbard CW (1963) The origin of the P3 pattern of Syhvilagus, Caprolagus, Oryctolagus and Lepus- J Mammal 44:1-15 Hyman BC, Slater TM (1990) Recent appearance and molecular characterisation of mitochondrial DNA deletions within a defined nematode pedigree. Genetics 124:845-853 Lopez N, Michaux J, Villata de JF (1976) Rongeurs et lagomorphes de Bagur-2 (province de Gerone, Espagne), nouveau remplissage de fissure du debut du picistocene moyen. Acta Gen] Hisp 11,2:46-54 Lopez-Martinez Ni (I977a) Revision sistematica y biostratigralica de los lagomorphos (Mammalia) de] ncogcno y cuaternario de Espana. Tesis doctoral, Universidad de Madrid Lopez-Martinez N (1977b) Nuevos iagomorphos (Mammalia) del neogeno y cuaternario Espana!. Trabajos M/Il 8:7-45 Mignotte B, Dunon-Bluteau D, Reiss C, Mounolou JC (1987) Sequence deduced physical properties in the D-loop region common to five vertebrate mitochondrial DNAs. J Theor Biol 124:57-69 Mignotte F, Biju-Duval C. Champagne AM, Gueride M, Monncrot M, Mounolou JC (1990a) Repeats and variations in the control region of rabbit mtDNA. In: Quagliariello E, Papa S, Palmieri F, Saccone C (eds) Structure, function and biogenesis of energy transfer system. Elsevier, Amsterdam, pp 47-101 Mignotte F, Gueride M, Champagne AM, Mounolou JC (I990b) Direct repeats in the noncoding region of rabbit mitochondria] DNA: involvement in the generation of intra and inter-individual heterogeneity. EurJ Biochem 194:561-571 Monnerot M, Mounolou JC, Solignac M (1986) Intra-individual length heterogeneity of Rana esculenta mitochondria] DNA. Biol Cell 52:213-218
102
PDF
Nei M (1987) DNA polymorphism within and between populations. In: Nei M (ed) Molecular evolutionary genetics. Columbia University Press, New York, pp 255-286 Nei M, Li WH (1979) Mathematical model for studying variation in terms of restriction endonucleases. Proc Natl Acad Sci USA 76:5269-5273 Rand DM, Harrison RG 41986) Mitochondria) DNA transmission genetics in crickets. Genetics 114:955-970 Robinson TJ, Elder FBB, Chapman JA (1983) Karyotypic conservatism in the genus Lepus (order Lagomorpha). Can J Gcnct Cytol 25:540-544 Saccone C, Attimonelli M, Sbistl E (1987) Structural elements highly preserved during the evolution of the D-loop-containing region in vertebrate mitochondria! DNA. 1 Mol Evol 26: 205-211 Saint-Girons MC (1973) Lagomorphes. In: Saint-Girons MC (ed) Les mammiferes de France et du Benelux. Doin Editions, Paris, p 402 Solignac M, Genermont J. Monnerot M, Mounoiou JC (1984) Mitochondria] genetics of Drosophila: mt DNA segregation in heteroplasmic strains of D. rnaaritiana. Mol Gen Genet 197:183-188 Solignac M, G6nermonl J, Monnerot M, Mounolou JC (1987) Drosophila mitochondrial genetics: evolution ofheteroplasmy through germ line cell divisions. Genetics 117:687-696 Soriguer RC (1983) El concjo: papel ccologico y estrategia de
vida en los ecosystemas mediterraneos. In: Caslroviejo 1, Blanc JM (eds) Actas del XV Congreso International de Fauna Cinegelica Y Silvestre. Trujillo (Caceres), Espana, 198 I, Federation Espanola de Gaza, Sevilla, pp 517-543 Tzagolof A, Myers AM (1986) Genetics of mitochondria] biogenesis. Annu Rev Biochem 55:249-288 Van der Loo W, Ferrand N, Soriguer RC (1990) Estimation of gene diversity at the b-locus of the constant region of the immunoglobulin light chain in natural populations of European rabbits (Oryctolagus cuniculus) in Portugal, Andalusia and on Azorean islands. Genetics (in press) Vcltri KL, Espiritu M, Singh G (1990) Distinct gnomic copy number in mitochondria of different mammalian organs. J Cell Physiol 143:160-164 Vigne JD (1988) Donnees preliminaires sur l'histoire du pcu' plement mammalien de I ilot de Zembra {Tunisie). Mammalia 52:567-574 Wilson AC, Cann RL, Carr S, George M, Gyllenstein UB, HelmBychowsky KM, Higuchi RG. Palumbi SR. Prager EM, Sage RD, Stoneking M (1985) Mitochondria] DNA and two perspectives on evolutionary genetics. Biol J Linn Soc 26:375400 Wright JW, Spolsky C, Brown WM (1983) The origin of the parthenogenetic lizard C'nemidophorus laredoensis inferred from mitochondrial DNA analysis. Herpetologica 39:410416
PDF
