Animal mitochondrial genomes

This Article

	Abstract
	Print PDF (369K)
	Alert me when this article is cited
	Alert me if a correction is posted

Services

	Email this article to a friend
	Similar articles in this journal
	Similar articles in ISI Web of Science
	Similar articles in PubMed
	Alert me to new issues of the journal
	Add to My Personal Archive
	Download to citation manager
	Search for citing articles in: ISI Web of Science (452)
	Request Permissions
	Commercial Re-use Guidelines for Open Access NAR Content

Google Scholar

	Articles by Boore, J. L.
	Search for Related Content

PubMed

	PubMed Citation
	Articles by Boore, J. L.

Social Bookmarking

	What's this?

Nucleic Acids Research Pages 1767-1780

Animal mitochondrial genomes
Introduction
   Chordata and Hemichordata
   Echinodermata
   Arthropoda
   Mollusca
   Annelida
   Nematoda
   Platyhelminthes
   Cnidaria
Conclusions
Acknowledgements
References

Animal mitochondrial genomes

Jeffrey L. Boore*

Department of Biology, University of Michigan, 830 North University Avenue, Ann Arbor, MI 48109-1048, USA

Received November 12, 1998; Revised and Accepted February 15, 1999

ABSTRACT

Animal mitochondrial DNA is a small, extrachromosomal genome, typically ~16 kb in size. With few exceptions, all animal mitochondrial genomes contain the same 37 genes: two for rRNAs, 13 for proteins and 22 for tRNAs. The products of these genes, along with RNAs and proteins imported from the cytoplasm, endow mitochondria with their own systems for DNA replication, transcription, mRNA processing and translation of proteins. The study of these genomes as they function in mitochondrial systems-‘mitochondrial genomics’-serves as a model for genome evolution. Furthermore, the comparison of animal mitochondrial gene arrangements has become a very powerful means for inferring ancient evolutionary relationships, since rearrangements appear to be unique, generally rare events that are unlikely to arise independently in separate evolutionary lineages. Complete mitochondrial gene arrangements have been published for 58 chordate species and 29 non-chordate species, and partial arrangements for hundreds of other taxa. This review compares and summarizes these gene arrangements and points out some of the questions that may be addressed by comparing mitochondrial systems.

INTRODUCTION

Mitochondria play a central role in metabolism (1), apoptosis (2), disease (3) and aging (4). They are the site of oxidative phosphorylation, essential for the production of ATP, as well as a variety of other biochemical functions. Within these subcellular organelles is a genome, separate from the nuclear chromatin, referred to as mitochondrial DNA (mtDNA), very commonly used in studies of molecular phylogenetics (5). Among multicellular animals this is nearly always a closed-circular molecule; only the cnidarian classes Cubozoa, Scyphozoa and Hydrozoa have been found to have linear mtDNA chromosomes (6).

In animals, mtDNA is generally a small (15-20 kb) genome containing 37 genes. Although much larger mitochondrial genomes have occasionally been found, these are the product of duplications of portions of the mtDNA rather than variation in gene content (7,8). The typical gene complement encodes 13 protein subunits of the enzymes of oxidative phosphorylation, the two rRNAs of the mitochondrial ribosome, and the 22 tRNAs necessary for the translation of the proteins encoded by mtDNA. For historical reasons, these genes have designations specific to animal mtDNA; Table 1 lists these along with their synonyms. This gene content has been shown to vary only in nematodes (9-11), which lack A8, a bivalve (12) which both lacks A8 and contains an extra tRNA, and cnidarians (13-18), which have lost nearly all tRNA genes and gained one or two additional genes not found so far in other mtDNAs (see below). All of the 37 genes typically found in animal mtDNA have homologs in the mtDNAs of plants, fungi and/or protists (19-22). There is also generally a single large non-coding region which, for a few animals (23), is known to contain controlling elements for replication and transcription. Whether these ‘control regions’ (assumed to be the largest non-coding segment) are homologous between distantly related animals or, alternatively, have arisen from various non-coding sequences independently in separate evolutionary lineages is often unclear since they share no sequence similarity except among closely related animals. Although the largest non-coding region is illustrated herein for many of these genomes, this is not necessarily to imply homology, and it is noteworthy that many mtDNAs also have other smaller non-coding regions that are not illustrated that may or may not contain controlling elements.

Table 1. The 13 protein, two ribosomal RNA and 22 tRNA genes typically found in animal mitochondrial genomes For historical reasons these have designations commonly used only for animal systems. To aid comparison with other works, column three lists the synonymous gene labels used for non-animal systems.

In some mtDNAs all genes are transcribed from one strand, whereas in others the genes are distributed between the two strands. This is indicated in the text and figures by underlining right-to-left (as drawn) oriented genes. In the few cases where it has been studied, each strand is transcribed as a single large polycistron which is post-transcriptionally processed into gene-specific messages. Transfer RNA genes, whose secondary structures are thought to signal cleavage (24), punctuate the polycistron of some mtDNAs, but others have tRNA genes clustered so as to preclude such a mechanism. Although potential secondary structures have been proposed as substitutes at some of these gene junctions, no obvious secondary structures can be identified for most cases. It is possible that the ‘polycistron model’ may not be appropriate for many of the animals whose mitochondrial gene expression remains unstudied. Whatever the mechanisms of gene rearrangement (discussed below), it is clear that a new gene order must preserve, or enable substitutes for, whatever signals are necessary for transcription and processing of RNAs from the rearranged genome.

Although animal mitochondrial sequences are known to evolve rapidly, their gene arrangements often remain unchanged over long periods of evolutionary time. For example, the gene arrangements of human (25-27) and trout (28) mtDNAs are identical (Table 1). With few exceptions, gene arrangements are relatively stable within major groups, but variable between them, and comparisons of these gene arrangements have great potential for resolving some of the deepest branches of metazoan (multicellular animal) phylogeny. The great number of potential gene arrangements makes it very unlikely that different taxa would independently adopt an identical state, so complex rearrangements are unlikely to be convergent. Comparisons of mitochondrial gene arrangements have provided convincing phylogenies in several cases where all other data were equivocal, including the relationships among major groups of echinoderms (29) and arthropods (30,31).

Many aspects of genome evolution (e.g. evolution of gene rearrangements, gene regulation, message processing systems and replication mechanisms) may be tractable for mitochondrial systems at a level of detail and understanding not currently possible for nuclear systems. Questions that may be addressed by ‘mitochondrial genomics’ include: What are the patterns of evolution of gene expression mechanisms? Which types of genes rearrange most often? What aspects of genomic variation are correlated with aspects of physiology, molecular mechanisms or life history? How do interacting molecular factors coevolve? Is rate of sequence evolution correlated with rate of gene rearrangements within a lineage?

Comparisons of mitochondrial genomes may result in significant insights into the evolution both of organisms and of genomes. Other recent reviews summarize information about the mitochondrial genomes of plants (20), protists (21) and fungi (22), and address issues of the genetics and molecular biology of mitochondrial systems (18,20,23). This summary of animal mitochondrial gene arrangements is given largely in the hope of stimulating further expansion of this data set, of encouraging broad phylogenetic comparisons of the basic molecular biology of mitochondrial systems, and of advocating the development of better techniques for comparison and analysis.

CHORDATA AND HEMICHORDATA

Of the 87 complete metazoan mtDNA sequences that have been published, 58 are from chordates with all but two of these from vertebrates. Table 2 lists these 58 species. There is little variation in gene arrangement; the 44 species listed without asterisks share an identical arrangement of all 37 genes. Many hundreds of other vertebrate species that are too numerous to include here have been sampled for short regions identical to this common vertebrate arrangement (many of these can be found at: www.biology.lsa.umich.edu/~jboore ).

Table 2. Chordates for which the complete mitochondrial gene arrangement has been determined

All the species share an identical arrangement of the 37 mitochondrial genes except those 14 marked with an asterisk. Database accession numbers are in parentheses. See Figure 1 for the specific deviations.

Figure 1A depicts this common vertebrate gene arrangement along with that of the primitive cephalochordate Branchiostoma (‘amphioxus’) (32,33). The gene arrangement of Branchiostoma varies from the common vertebrate arrangement only in the location of four tRNA genes: F, G, M and N (marked with an asterisk in Fig. 1A). In order to determine whether the primitive gene arrangement for Chordata (Vertebrata plus Cephalochordata) is more like the common vertebrate arrangement or more like that of Branchiostoma, one can compare with the arrangements of lesser related taxa. The position of F is similar in the common vertebrate arrangement and in the mtDNAs of echinoderms (Fig. 2) and the positions of G and M are identical in the common arrangement and in the mtDNA of many arthropods (Fig. 3), so the most parsimonious interpretation is that the primitive arrangement of these tRNA genes for Chordata is the same as in the common vertebrate arrangement, with later, independent translocations resulting in their positions in Branchiostoma mtDNA. Since the position of N in Branchiostoma mtDNA is identical to that of the hemichordate Balanoglossus mtDNA (34), this must be the chordate primitive state, with its translocation from N-W-A-C-Y to W-A-N-C-Y occurring at the base of Vertebrata (32).

As noted with asterisks in Table 2 and depicted in Figure 1, small-scale rearrangements have been found in several chordate lineages. In each of these cases where a gene (i.e. other than the non-coding region, see below) rearrangement is shared it is a valid marker of evolutionary relationships, although these particular relationships are not generally controversial.

The sea lamprey (35) has a translocation of E-Cytb and changes in the locations of the non-coding regions. It was initially unclear if this arrangement could be the primitive state for vertebrates, but further outgroup comparisons convincingly demonstrate that it is, rather, derived for the sea lamprey (see above). Several marsupials (36-38) also have a small scale change, a shared, derived rearrangement of W-A-N-C-Y to A-C-W-N-Y. This is not found in a monotreme (39) or in eutherian mammals.

Several reptiles have small scale, unique rearrangements. The Texas blind snake (40), has a translocation of Q not shared by other studied reptiles. An amphisbaenian reptile, Bipes, has an independent duplication of T-P followed by the degeneration of one of these tRNAs into a pseudogene (see more below) (41). Several acrodont and agamid reptiles share an exchange of position between I and Q (42,43). An alligator (44) and a saltwater crocodile (45) share an exchange of the non-coding region and F and this alligator shares with a crocodile, caiman (40) and several squamate reptiles (40) an exchange in position of S(AGY) and H.

Many birds share an exchange of position between the blocks Cytb-T-P and ND6-E (46-53) (Table 2; Fig. 1A). In addition, a recent study (54) identified two arrangements of bird mtDNAs (termed ‘gene order B’ and ‘gene order C’) differing in the location of the large non-coding region. This is based on five complete mtDNA sequences (Rhea, Aythya and Vidua having gene order B and Falco and Smithornis having gene order C) and a sampling of short fragments of many species representing 16 orders. When these two gene arrangements are plotted on a phylogenetic tree derived from other types of data, it is apparent that there must have been several convergent, independent origins of gene order C. This suggests the model proposed in Figure 1C, where the original translocation between Cytb-T-P and ND6-E was accompanied by a duplication of the large non-coding region. Random loss of one or the other of these non-coding regions could then have produced either of the states found among birds. Supporting this model is the commonality of duplicate non-coding regions [e.g. in mtDNAs of Tuatara (45) and of several snakes (40,55,56)] and the presence of a secondary non-coding region between E and F in birds having gene order C (the location of the single non-coding region of gene order B). Although the authors designate this as ‘nc’ for ‘non-coding’ to differentiate it from the ‘control region’, it may be more correct to view this as the degenerating vestige of a duplicate control region; Figure 1 therefore designates it as a pseudo-control region. This view is reinforced by noting that the two species from which sequence of this region has been determined show no interspecies similarity (as though they result from independent paths of degeneration) and that, in one of the cases (Smithornis), there is substantial sequence similarity between the designated ‘nc’ and ‘control region’.

Although their nearest outgroup relative, the crocodile, does not share this rearrangement (45), the tuatara has two large non-coding regions, one in the same relative position as in birds and the other in the same relative position as in the common vertebrate arrangement (45,57). This may be a retained primitive feature of the original rearrangement that occurred prior to the divergence of birds from reptiles, or it may have occurred convergently in these independent lineages.

Nucleic Acids Research