False phylogenies on wood mice due to cryptic cytochrome-b pseudogene

https://doi.org/10.1016/j.ympev.2008.12.008Get rights and content

Abstract

The phylogeny and phylogeography of the Old World wood mice (subgenus Sylvaemus, genus Apodemus, Muridae) are well-documented. Nevertheless, the distributions of species, such as A. fulvipectus and A. ponticus remain dubious, as well as their phylogenetic relationships with A. sylvaticus. We analysed samples of Apodemus spp. across Europe using the mitochondrial cytochrome-b gene (cyt-b) and compared the DNA and amino-acid compositions of previously published sequences. The main result stemming from this study is the presence of a well-differentiated lineage of Sylvaemus including samples of various species (A. sylvaticus, A. fulvipectus, A. ponticus) from distant locations, which were revealed to be nuclear copies of the mitochondrial cyt-b. The presence of this cryptic pseudogene in published sequences is supported by different pathways. This has led to important errors in previous molecular trees and hence to partial misinterpretations in the phylogeny of Apodemus.

Introduction

The phylogeny and phylogeography of the Old World wood mice (subgenus Sylvaemus, genus Apodemus, Muridae) are well-documented in numerous publications involving the mitochondrial cytochrome-b gene (cyt-b; e.g. Martin et al., 2000, Hille et al., 2002, Michaux et al., 2002, Michaux et al., 2003, Michaux et al., 2004, Michaux et al., 2005, Reutter et al., 2003, Balakirev et al., 2007, Hoofer et al., 2007, Suzuki et al., 2008) or both mitochondrial and nuclear genes (e.g. Michaux et al., 2002, Michaux et al., 2005, Suzuki et al., 2008). The western Eurasian species of the subgenus Sylvaemus, i.e. the alpine mouse A. alpicola, the wood mouse A. sylvaticus, the yellow-necked mouse A. flavicollis, the pygmy woodmouse (A. uralensis), the yellow-breasted mouse A. witherbyi, the Mt Hermon mouse A. hermonensis (a synonym of A. witherbyi; see Musser and Carleton, 2005) and the Caucasus mouse A. ponticus have been revealed to be genetically closely related (Michaux et al., 2002, Suzuki et al., 2008).

Within the species, the phylogeography of A. flavicollis, based on the cyt-b (Michaux et al., 2004) showed two well differentiated clades, a first in Turkey, Syria, Israel, and Iran (Southern of the Caucasus), and a second including western, central, and eastern Europe, and Russia (Fig. 1). For A. alpicola, a species restricted to the alpine region, Michaux et al. (2002) and Reutter et al. (2003) documented its monophyly. A thorough phylogeography of A. sylvaticus, based on the cyt-b (Michaux et al., 2003), highlighted a clear pattern with slightly divergent sub-clades involving different geographic areas such as (i) western, northern and central Europe, (ii) Italy and Balkans, or (iii) Sicily. Interestingly, Reutter et al. (2003), based on restricted sampling, found two additional well differentiated sequences of A. sylvaticus (G3.4 and G3.1), both from Karlsruhe in the southern part of Germany, from animals clearly identified as A. sylvaticus in the field. The first one (G3.4) was considered to belong to a lineage that separated very early, before the split separating the Italian animals (Michaux et al., 2003) and animals from the Pyreneans to the Ukraine. The authors were not able to explain the syntropic occurrence of these two lineages. The second one (G3.1) revealed to be closely related to some sequences ascribed to A. fulvipectus (but the identity is still uncertain) and A. ponticus from Georgia (Hille et al., 2002) and to a unique sequence of A. sylvaticus from Konstanz, Germany (Martin et al., 2000), a locality 150 km away from Karlsruhe. Both tissue samples were from biopsies of released animals without preserved voucher specimens. Reutter et al. (2003) suggested that an unrecognized population of A. fulvipectus might occur in southern Germany. As these sequences were fully coding the authors excluded the presence of a nuclear pseudogene and refuted the possibility of a DNA contamination in their laboratory, as they had never handled samples of A. fulvipectus from the Caucasus before. Moreover, the situation in Georgia, as studied by Hille et al. (2002), was all but clear. The assignment of some samples by morphology was not concordant with the genetic assignment (cyt-b), especially the assignment of A. ponticus and A. flavicollis from the same locality appeared often problematic. Finally, a recent phylogenetic study (Suzuki et al., 2008) including some of these sequences yielded similar results, which led the authors to the conclusion that the systematics of these species needs to be revised.

In the present study, we therefore analysed 102 cyt-b sequences of different species and lineages of Apodemus, and compared their DNA and amino-acid compositions in order to explain the presence of three divergent cyt-b lineages of A. sylvaticus in Germany, i.e. (i) the lineage widely distributed in central and northern Europe (Michaux et al. 2003), (ii) the unexpected lineage (G3.4) found by Reutter et al. (2003), and (iii) the lineage including samples of A. fulvipectus from Germany (Martin et al., 2000, Reutter et al., 2003: G3.1), and A. fulvipectus and A. ponticus from Georgia (Hille et al., 2002).

Section snippets

Samples

We analyzed 102 samples of the Apodemus spp. collected in Europe (Table 1). We used Mus musculus (AB205312) as an outgroup. The set of samples included material from the collections of JR Michaux (JM) located at the University of Liège, Belgium, and from the Zoologisches Forschungsmuseum Alexander Koenig, Bonn, Germany (ZFMK). Specimens labelled with “HB” were biopsies without voucher specimens. Additional sequences (Table 1) were taken from GenBank (Martin et al., 2000, Hille et al., 2002,

Molecular phylogeny

We found 80 different haplotypes of 818 bp within the 102 analysed sequences, including 277 variable sites, of which 209 were parsimony informative. We did not find any insertions or deletions. In addition, no stop codons were found. Because the three phylogenetic methods yielded identical arrangements of the main branches, we show the relationship between haplotypes only for the ML analysis (Fig. 3).

Our phylogenetic analyses revealed seven different lineages, which included:

(I) Samples of A.

Discussion

Nuclear copies of mitochondrial genes have been described in a variety of animals and plants, and revealed to be common in mammals, particularly in Rodentia (e.g. Cooper et al., 2003, DeWoody et al., 1999, Mirol et al., 2000, Jaarola and Searle, 2004, Rat Genome Sequencing Project Consortium, 2004, Richly and Leister, 2004, Triant and DeWoody, 2007, Triant and DeWoody, 2008). These nuclear sequences can cause major problems in systematic analyses, including DNA barcoding and phylogeography,

Acknowledgments

For providing tissue samples, we thank our colleagues A. Bukhnikashvili (Georgia), W. Booth and I. Montgomery (Ireland), J. Searle and X. Lambin (Great Britain), G. Olsson (Sweden) and T.S. Hansen (Denmark). We also thank N. Di Marco for laboratory work. This study was supported by the Swiss National Science Foundation (SNSF) to S.D, and by a Belgian FNRS fellowship (mandat “chercheur qualifié”) and a financial grant of the Belgian FNRS (crédits aux chercheurs) to J.M.

References (36)

  • J.A. DeWoody et al.

    A translocated mitochondrial cytochrome b pseudogene in voles (Rodentia: Microtus)

    J. Mol. Evol.

    (1999)
  • M.G. Filippucci et al.

    Genetic variation and evolution in the genus Apodemus (Muridae, Rodentia)

    Biol. J. Linn. Soc.

    (2002)
  • S. Guindon et al.

    A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood

    Syst. Biol.

    (2003)
  • P.D. Hebert et al.

    Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulgerator

    Proc. Natl. Acad. Sci. USA

    (2004)
  • A. Hille et al.

    Morphometric, biochemical and molecular traits in Caucasian wood mice (Apodemus/Sylvaemus), with remarks on species divergence

    Acta Theriol.

    (2002)
  • S.R. Hoofer et al.

    New information for systematics, taxanomy, and phylogeography of the rodent genus Apodemus (Sylvaemus) in Ukraine

    J. Mammal.

    (2007)
  • J.P. Huelsenbeck et al.

    Evolution: Bayesian inference of phylogeny and its impact on evolutionary biology

    Science

    (2001)
  • G. Ibarguchi et al.

    Defeating numts: semi-pure mitochondrial DNA from eggs and simple purification methods for field-collected wildlife tissues

    Genome

    (2006)
  • Cited by (49)

    View all citing articles on Scopus
    View full text