Mytilus Mitochondrial DNA Contains a Functional Gene for a tRNA^Ser(UCN) With a Dihydrouridine Arm-Replacement Loop and a pseudo-tRNA^Ser(UCN) Gene

Corresponding author: David R. Wolstenholme, Department of Biology, University of Utah, 257 South 1400 East, Salt Lake City, UT 84112-0840., wolstenholme{at}biology.utah.edu (E-mail)

Communicating editor: G. B. GOLDING

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUDING REMARKS LITERATURE CITED

A 2500-nucleotide pair (ntp) sequence of F-type mitochondrial (mt) DNA of the Pacific Rim mussel Mytilus californianus (class Bivalvia, phylum Mollusca) that contains two complete (ND2 and ND3) and two partial (COI and COIII) protein genes and nine tRNA genes is presented. Seven of the encoded tRNAs (Ala, Arg, His, Met(AUA), Pro, Ser(UCN), and Trp) have the potential to fold into the orthodox four-armed tRNA secondary structure, while two [tRNA^Ser(AGN) and a second tRNA^Ser(UCN)] will fold only into tRNAs with a dihydrouridine (DHU) arm-replacement loop. Comparison of these mt-tRNA gene sequences with previously published, corresponding M. edulis F-type mtDNA indicates that similarity between the four-armed tRNA^Ser(UCN) genes is only 63.8% compared with an average of 92.1% (range 86.2–98.5%) for the remaining eight tRNA genes. Northern blot analysis indicated that mature tRNAs encoded by the DHU arm-replacement loop-containing tRNA^Ser(UCN), tRNA^Ser(AGN), tRNA^Met(AUA), tRNA^Trp, and tRNA^Pro genes occur in M. californianus mitochondria, strengthening the view that all of these genes are functional. However, Northern blot and 5' RACE (rapid amplification of cDNA ends) analyses indicated that the four-armed tRNA^Ser(UCN) gene is transcribed into a stable RNA that includes the downstream COI sequence and is not processed into a mature tRNA. On the basis of these observations the M. californianus and M. edulis four-armed tRNA^Ser(UCN) sequences are interpreted as pseudo-tRNA^Ser(UCN) genes.

THE mitochondrial (mt) genomes of almost all multicellular animals (metazoa) comprise a single circular molecule of 14–42 kb (WOLSTENHOLME 1992A ; WOLSTENHOLME and FAURON 1995 ). The mtDNA molecules of ~75 metazoans have been completely sequenced. Approximately two-thirds of these are from chordates (KRETTEK et al. 1995 ; ARNASON et al. 1996 ; ZARDOYA and MEYER 1997 ) but the remainder represent a variety of invertebrate phyla, including Cnidaria, Nematoda, Arthropoda, Mollusca, Annelida, and Echinodermata (see, for example, CLARY and WOLSTENHOLME 1985 ; JACOBS et al. 1988 ; CANTATORE et al. 1989 ; GAREY and WOLSTENHOLME 1989 ; OKIMOTO et al. 1991 , OKIMOTO et al. 1992 ; CROZIER and CROZIER 1993 ; BOORE and BROWN 1994 , BOORE and BROWN 1995 ; ASAKAWA et al. 1995 ; BEAGLEY et al. 1995 , BEAGLEY et al. 1998 ; HATZOGLOU et al. 1995 ; TERRETT et al. 1996 ).

Most metazoan mtDNAs contain the same set of genes for 13 energy-pathway proteins: cytochrome b (Cyt b); subunits I–III of cytochrome c oxidase (COI–COIII); subunits 6 and 8 of F_O ATP synthetase (ATPase6 and ATPase8); subunits 1–6 and 4L of the respiratory chain NADH dehydrogenase (ND1-ND6 and ND4L); and two RNAs that are homologous to the 16S and 23S ribosomal RNAs (rRNAs) of Escherichia coli (but smaller and usually referred to as s-rRNA and l-rRNA) and 22 transfer RNAs (tRNAs). The arrangement of genes within metazoan mtDNAs is very compact. Differences in gene arrangement occur among metazoan mtDNAs, and the extent of these differences appears to be a function of evolutionary distances (WOLSTENHOLME 1992A ; BOORE and BROWN 1994 , BOORE and BROWN 1995 ).

Metazoan mt-genomes are characterized by a number of unusual features. These include genetic code modifications, the use of unorthodox translation initiation codons, post-transcriptional generation of translation termination codons, translation by only 22 tRNAs that are mtDNA encoded and of diverse size and structural form, and replication by a unique asymmetrical mode (CLAYTON 1992 ; WOLSTENHOLME 1992A , WOLSTENHOLME 1992B ; WOLSTENHOLME and FAURON 1995 ).

HOFFMANN et al. 1992 determined nucleotide sequences that collectively represent ~80% of the ~17.1-kb mtDNA molecule of the blue mussel Mytilus edulis (phylum Mollusca, class Bivalvia). Because these sequences contained the terminal 5' and 3' ends of all protein and rRNA genes and all sequences between these genes including tRNA genes, they provided sufficient information to construct a complete gene map. The M. edulis mtDNA molecule was found to contain the gene complement common to fully mapped metazoan mtDNAs with two exceptions. The first is that there are two genes for methionine-specifying tRNAs: one with a CAU anticodon (tRNA^f-Met), as has been reported for all other metazoan mtDNAs so far examined, and a second with a UAU anticodon that might be expected to specifically recognize AUA codons, and is so far unique among metazoan mtDNAs. The second exception to gene content is that a gene for ATPase8 was not identified, a situation previously known among Metazoa only for nematode mtDNAs (WOLSTENHOLME et al. 1987 ; OKIMOTO et al. 1991 , OKIMOTO et al. 1992 ). Relative to the corresponding genes of Drosophila yakuba and mouse, considerable extensions were postulated at the 5' ends of the COI and ND1 genes, and at the 3' ends of the COI and COIII genes of M. edulis. The COIII gene 3' extension was of particular interest as the overall size of this gene and the C-terminal proximal amino acid sequence of the protein it encodes have been highly conserved in all organisms previously examined. Gene arrangement in the mtDNA molecule of M. edulis was found to be greatly different from gene arrangement in other sequenced metazoan mtDNAs, including those of the more recently sequenced polyplacophoran mollusk Katherina tunicata and the gastropod mollusks Albinaria coerulea and Cepaea nemoralis (BOORE and BROWN 1994 ; HATZOGLOU et al. 1995 ; TERRETT et al. 1996 ).

Subsequent to the report of HOFFMANN et al. 1992 it was demonstrated that in M. edulis, the closely related M. trossulus and M. galoprovincialis, and the more distantly related M. californianus, there are two highly distinct mtDNA sequence types that are gender limited in occurrence (FISHER and SKIBINSKI 1990 ; ZOUROS et al. 1992 , ZOUROS et al. 1994A , ZOUROS et al. 1994B ; SKIBINSKI et al. 1994A , SKIBINSKI et al. 1994B ; BEAGLEY et al. 1997 ). One mtDNA type (F) is found in both females and males, but the second type (M) is found only in males. The mtDNA molecule sequenced by HOFFMANN et al. 1992 has been identified as the F-type.

In this article we report the sequence of a 2500-nucleotide pair (ntp) segment of the F-type mtDNA molecule of M. californianus. From comparison of this latter sequence with the corresponding M. edulis sequence, and data obtained from Northern blot analysis, we show that Mytilus mtDNA molecules include a gene that encodes a functional dihydrouridine (DHU) arm-replacement loop-containing tRNA^Ser (UCN), not identified by HOFFMANN et al. 1992 . Further, we demonstrate that the sequence interpreted by HOFFMANN et al. 1992 as the Mytilus mt-tRNA^Ser(UCN) gene is likely a pseudogene, the transcript of which remains linked to the transcript of the COI gene. Our data also confirm that the Mytilus mt-tRNA^Met(AUA) gene encodes a functional tRNA, and provide confirmatory support for the conclusion that the Mytilus cytochrome c oxidase subunit III protein (COIII) has a unique, functional C-terminal extension.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUDING REMARKS LITERATURE CITED

Animals and mitochondrial nucleic acid isolations:
Specimens of M. californianus were collected from Monterey Bay by Michael O. Morris, Sea Life Supply, Sand City, CA. Mitochondria were isolated from mechanically disrupted adductor muscles of 10 M. californianus and mtDNA was prepared by methods previously described (WOLSTENHOLME and FAURON 1976 ; PONT-KINGDON et al. 1994 ). For the preparation of RNA, total soft tissue was removed from 20 live M. californianus, chilled on ice, immersed in 5 volumes of 50 mM Tris-HCl (pH 8.0), 300 mM mannitol, 3 mM EDTA, 0.1% bovine serum albumin, 1 mM ß-mercaptoethanol, 4 mM L-cysteine, and 0.1% polyvinylpyrrolidone, and mechanically disrupted in a Waring blender. The product was centrifuged for 15 min at 4,000 x g. From the resulting supernatant a crude mitochondrial pellet was obtained by centifugation for 20 min at 10,000 x g, and RNA was isolated from this pellet with a REX Total RNA extraction kit (Amersham Life Science, Piscataway, NJ), using a mitochondrial weight equivalent to that recommended by the supplier for total animal tissue. Finally, the RNA was ethanol (3 volumes 95%) precipitated, resuspended in diethylpyrocarbonate-treated water, and either used directly or stored at -20°.

Oligonucleotide probes:
Oligonucleotides were synthesized in an Applied Biosystem (Foster City, CA) Synthesizer 380B and used without further purification. Oligonucleotides complementary to the sense strand of various presumptive tRNA genes and the tRNA^Ser(UCN) gene (oligonucleotide F5) that were used in Northern blot experiments are shown in Figure 2. The sequence of an oligonucleotide complementary to E. coli tRNA^Tyr and used as a size marker in one Northern experiment was 5'GGAAGGATTCGAACCTTCGAA. These oligonucleotides were 5'-end-labeled using T4 polynucleotide kinase (BRL) and [-³²P]ATP. The gene locations of all other oligonucleotides (F1–F4) used in Northern blot experiments and/or as primers in PCR amplifications are shown in Figure 1B, and following are their nucleotide sequences together with their coordinates in the determined sequence (Figure 1A). Reverse-order numbers correspond to antisense sequences. Nonencoded restriction site-containing sequences are bracketed. F1: 1–30, 5' [TGGAATTC]TGGTACTGACTTTTGTAGATG; F2: 3195–3178 (not included in Figure 1A), 5' [AGGAATT]CCAGAACAATGCATAATTACC; F3: 2012–1992, 5' TAAATAGAAACCCTCTTGAGG; and F4: 2402–2383, 5' CTACTTCACCGTAACCGCCTC. Following are the nucleotide sequences of degenerate primers COI-11D and COI-12D (Figure 1B), which are based on highly conserved sequences in various metazoan COI genes. The coordinates shown are those of the Metridium senile intron-containing COI gene complex (BEAGLEY et al. 1996 ). Y = C or T, R = A or G, N = C, T, A, or G. Nonencoded, restriction site-containing sequences are bracketed. COI-11D: 712–731, 5' [CCGGATCC]TTYTGRTTYTTYGGNCAYCC; COI-12D: 1982–1964 (equivalent to ntp 1051–1070 of the uninterrupted COI gene), 5' [CCGGATCC]ACNACRTARTANGTRTCRTG.

View larger version (115K): In this window In a new window Download PPT slide   Figure 1. (A) Comparison of a continuous 2500-ntp sequence of the F-type mtDNA molecule of M. californianus (M.c) and two corresponding, partial F-type mtDNA sequences of M. edulis (M.e) taken from HOFFMANN et al. 1992 (GenBank accession nos. M83760 and M83761). The numbering to the right is shown only for the M. californianus sequence. All of the partial and complete protein genes shown (COI and COIII, cytochrome c oxidase subunits I and III; ND2 and ND3, NADH dehydrogenase subunits 2 and 3), the eight tRNA genes [Ser(UCN), Met (AUA), Arg, Trp, Ala, Ser(AGN), His, and Pro], and the sequence interpreted as a pseudo-tRNASer(UCN) gene [Ser(UCN)] are transcribed in the same direction, from left to right. Nucleotides 1–30 of the M. californianus sequence are one of a pair of PCR primers used to obtain the sequence: an EcoRI site followed by 24 nt corresponding to the M. edulis sequence. The sequence of the M. edulis ND2 gene corresponding to ntp 762–1053 of the M. californianus sequence, and the sequence of the M. edulis COI gene beyond that shown have not been determined. The anticodon of each tRNA gene is boxed. The inverted large arrowhead above the M. californianus COIII sequence indicates the position at which the COIII sequences of other mtDNAs terminate. A period indicates the absence in one nucleotide sequence of a nucleotide that occurs in the other nucleotide sequence. Vertical lines indicate similarities between the M. californianus and M. edulis nucleotide sequences. The predicted amino acid sequences of the protein genes are shown above the M. californianus sequence and below the M. edulis sequence. Termination codons are indicated by an asterisk. Below the last line of the M. edulis sequence are indicated amino acids corresponding in position to amino acids interpreted as the initial amino acids of predicted mouse and D. yakuba COI proteins. The boxed M (methionine) near the 5' proximal end of the COI gene is the amino acid tentatively interpreted as the first in the polypeptide product of this gene in M. edulis by HOFFMANN et al. 1992 . The small arrow indicates a nucleotide in the M. californianus sequence that (as shown) was T in each of two clones of one strand, but a C in each of two clones of the complementary strand. Modifications of the standard genetic code used in translation of the Mytilus mtDNA sequences are AGA and AGG = serine; TGA = tryptophan; ATA = methionine. An unspecified nucleotide and amino acid in one of the M. edulis sequences is indicated by X. (B) Gene map of the sequenced segment of M. californianus mtDNA. The arrowheads above and below the map show the locations and directions of the oligonucleotide primers (F1–F4) and degenerate primers in the COI gene (11D and 12D) used for PCR amplifications. The sequences of all of these oligonucleotides are given in MATERIALS AND METHODS. The dotted line near the 5' end of the COI gene corresponds to the 3' end of the sequenced segment shown in A. Crosshatched regions are tRNA genes identified above the map using the one-letter amino acid code. Serine and methionine tRNA genes are also identified by the codon family that the tRNA recognizes: S(A), Ser(AGN); S(U), Serine (UCN); M(A), Met(AUA); S(U), Ser(UCN). The numbers of nucleotides in sequences between identified genes (the largest three in black) are shown beneath the map.

View larger version (37K): In this window In a new window Download PPT slide   Figure 2. The eight mt-tRNA genes and the tRNASer(UCN) gene of M. californianus contained in the sequence in Figure 1, and the tRNASer(UCN) gene of M. edulis (HOFFMANN et al. 1992 ), shown in the predicted secondary structural forms of the corresponding tRNAs. Each solid line on the outside of some M. californianus tRNA genes, and the solid line (F5) on the outside of the M. californianus tRNASer (UCN) gene, identifies a segment to which a complementary oligonucleotide was made and used in hybridization experiments to test for the presence of the various corresponding tRNAs. The identity (derived from the anticodon) and size (in nucleotides, nt) of each tRNA gene are shown. In the DHU arm-replacement loop-containing tRNASer (AGN) gene, possible secondary interactions between nucleotides in the DHU arm-replacement loop and the variable loop are shown by dashed lines. Reconstruction of the TC arm of the M. edulis tRNAArg gene relative to that given by HOFFMANN et al. 1992 , based on comparisons of the M. californianus and M. edulis tRNAArg gene sequences, is shown. In each functional (or presumed functional) M. californianus tRNA gene, nucleotide substitutions that occur in the corresponding M. edulis tRNA gene (HOFFMANN et al. 1992 ) are shown circled outside the sequence. Nucleotides in the M. californianus sequence that are circled with an arrow pointing out are not found in the M. edulis sequence. Arrows attached to circled nucleotides outside the M. californianus sequences indicate the locations of nucleotides found in the M. edulis tRNA sequences but not in the M. californianus sequences. Circled nucleotides in the M. californianus tRNASer(UCN) gene and the M. edulis tRNASer(UCN) gene are those that are substituted in the corresponding sequence of the other species.

PCR amplification of M. californianus mtDNA:
A 435-ntp segment of the M. californianus COI gene was PCR amplified using the degenerate primer pair COI-11D and COI-12D, and then sequenced. An oligonucleotide (F2) based on this COI sequence, which included an added 5' end EcoRI site, together with an oligonucleotide (F1) corresponding to a segment of the M. edulis COIII gene, and also including a 5' end EcoRI site, were used to PCR amplify a 3.2-kb segment of M. californianus mtDNA. This 3.2-kb sequence was cloned in the two reverse orientations into bacteriophage M13mp19 DNA, and amplified in E. coli JM101. The two reversely oriented, cloned sequences were designated MC412 and MC413. Except as described below, details regarding PCR amplification, electrophoresis, cloning, and purification of single-strand M13 DNA are given or referred to in WAHLEITHNER and WOLSTENHOLME 1987 and BEAGLEY et al. 1996 , BEAGLEY et al. 1998 .

DNA sequencing and sequence analysis:
DNA sequences were obtained by the extension-dideoxy-termination procedure (SANGER et al. 1977 ) from nested sets of deletion clones (DALE et al. 1985 ) of each of the two M13mp19 M. californianus mtDNA clones MC412 and MC413 that represent both strands of the 2500-ntp segment shown in Figure 1A, all of the remaining 684 nt of one strand (MC412), and 393 nt of the complementary strand (MC413). Sequences were analyzed using Lasergene software from DNASTAR (Madison, WI). The 2500-ntp M. californianus mtDNA sequence has been submitted to GenBank under the accession no. AF090831.

Northern blotting:
For determination of approximate relative sizes of RNAs containing transcripts of some tRNA genes and the tRNA^Ser(UCN) gene, M. californianus whole mtRNA (3 µg/lane) was electrophoresed in a 2.2 M formaldehyde-containing 3% agarose gel, transferred to a nitrocellulose filter, and hydridized with ³²P-labeled gene sequence-specific oligonucleotides. A similar procedure was used to determine and compare the sizes of RNAs containing transcripts of the ND3, COI, and tRNA^Ser(UCN) genes, except that electrophoresis was carried out in a 2.2 M formaldehyde-containing 1% agarose gel.

For precise determination of tRNA sizes, M. californianus mtRNA (10 µg/lane) was electrophoresed in polyacrylamide gels, electroblotted, and probed with gene sequence-specific oligonucleotides as described in OKIMOTO and WOLSTENHOLME 1990 , with the following exceptions: the 15% polyacrylamide gel (containing 8 M urea) was 17 cm wide by 15 cm long; electrophoresis was continued until the cyanol dye ran completely off the gel; the blots were washed twice at 25° in 2x SSC, 0.1% SDS for 10 min, followed by two washes at hybridization temperature (37° rather than 35°) in 0.1x SSC, 0.1% SDS for 30 min each.

Location of the 5' end of the COI transcript:
5' RACE (rapid amplification of cDNA ends) analysis (FROHMAN 1990 ) was carried out to locate the 5' end of COI transcript-containing RNA molecules as follows: M. californianus whole mtRNA (10 µg) was reverse transcribed in a 100-µl reaction using 1.5 µM random hexamers (BRL) and 1000 units of Moloney-MLV reverse transcriptase under conditions recommended by the supplier (Gibco BRL, Gaithersburg, MD). Incubation was for 10 min at 25° followed by 30 min at 42°. Following extraction of the incubation product with phenol, then chloroform, cDNAs were purified using a Centricon 100 microconcentrator (Amicon, Beverly, MA). Deoxyadenosine monophosphates (dAs) were added to the 3' ends of cDNAs by incubation in a 50-µl reaction containing cDNAs, 2.0 µl of 10 mM dATP, and 34 units of terminal deoxynucleotidyl transferase (Amersham) under the supplier's recommended reaction conditions for 5 min at 37°. Following a further phenol/chloroform extraction and Centricon purification, a second-strand synthesis reaction was carried out under the reverse transcription conditions but using as primer 20 pM of a BamHI, BglII, and PstI site-containing adaptor oligonucleotide (TB17-1: 5' CCAGATCGGATCCTGACGTTTTTTTTTTTTTTTT; PONT-KINGDON et al. 1994 ). The product was phenol/chloroform extracted, Centricon purified, and taken up in 50 µl of Tris-EDTA (TE) to provide a 5' RACE pool. Amplification of the double-stranded cDNA that included the 5' end region of the COI transcript was accomplished in a 100-µl PCR (in a DNA thermal cycler (Perkin Elmer-Cetus, Norwalk, CT) using 5 µl of the RACE pool, 25 pM of a BamHI, BglII, PstI site-containing primer TB17-2 (5' CCAGATCTGGATCCTGCAG), 25 pM of oligonucleotide F4 [which is complementary to a nucleotide sequence within the 5' end region of the M. californianus COI gene (Figure 2)], and an annealing temperature of 58°. The PCR product was digested with BamHI, inserted into BamHI-cleaved M13mp19, and sequenced.

	RESULTS AND DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUDING REMARKS LITERATURE CITED

Complete 2500-ntp sequences were obtained from two reversely oriented inserts of a PCR-generated 3.2-kb M. californianus mtDNA segment, cloned in M13mp19. As the two 2500-ntp sequences obtained were identical, we concluded that they both originated from the same mtDNA type. That this was the F-type was determined by comparison of a portion of the COI gene sequence with the equivalent COI F-type and COI M-type sequences identified by BEAGLEY et al. 1997 .

The entire, continuous 2500-ntp sequence of M. californianus F-type mtDNA is given in Figure 1A, and is compared with two sequences of M. edulis F-type mtDNA reported by HOFFMANN et al. 1992 . We interpret the M. californianus sequence to contain the following genes in the order given: COIII (3' end region), tRNA^Ser (UCN), tRNA^Met(AUA), ND2, tRNA^Arg, tRNA^Trp, tRNA^Ala, tRNA^Ser(AGN), tRNA^His, tRNA^Pro, ND3, tRNA^Ser(UCN); (see below), and COI (5' end region). The two M. edulis sequences, collectively, have the same gene content, but the sequence of a central region of the ND2 gene was not determined (Figure 1B).

Protein genes:
HOFFMANN et al. 1992 found that the M. edulis COI gene-containing open reading frame extended upstream beyond the codon corresponding to the translation initiation codons of the mouse and D. yakuba COI genes, for 21 and 22 codons, respectively. They suggested that, based on available information, a number of different triplets might act as the M. edulis COI translation initiation codon, and tentatively assigned this function to the ATA codon located 8 and 9 codons, respectively, upstream from the mouse and D. yakuba COI initiation codons, and at the same relative location as the Caenorhabditis elegans and Ascaris suum COI initiation codons. However, neither this ATA codon nor the ATT and ATA codons at the beginning of the M. edulis COI-gene containing open reading frame are conserved in the M. californianus sequence (Figure 1A). Therefore, from considerations of nucleotide sequence and predicted amino acid sequence conservation, it seems most likely that translation initiation of the COI gene occurs at the ATG codon in the open reading frame of both Mytilus species, located 17 and 18 codons upstream from the mouse and D. yakuba translation initiation codons (Figure 1A).

Data from nucleotide and amino acid sequence comparisons of the complete M. californianus and M. edulis ND3 genes, corresponding segments of the complete M. californianus ND2 gene and the two sequences of the M. edulis ND2 gene, and the portions of the M. californianus and M. edulis COIII genes that extend 3' relative to other metazoan COIII genes are given in Table 1. For the ND3 and ND2 genes, nucleotide and amino acid sequence similarities are 80.2 and 73.2%, and 94 and 80.7%, respectively. The corresponding values for comparisons of the M. californianus and M. edulis COIII 3' extensions are intermediate to those obtained for the ND3 and ND2 genes: 79.9 and 91.7%, respectively. Between the ND3 and ND2 genes of the two species, silent substitutions are in excess of replacement substitutions, and in silent positions, transitions exceed transversions, as has been reported for mt-protein genes of closely related species of other invertebrates and vertebrates (BROWN et al. 1982 ; BROWN and SIMPSON 1982 ; DESALLE et al. 1987 ). Between the M. californianus and M. edulis COIII extension sequences the proportion of substitutions that are silent, and the proportion of silent substitutions that are transitions are both greater than found for the ND3 and ND2 gene comparisons (Table 1). Taken together, similarities between patterns of nucleotide substitution among the COIII gene extensions and the ND3 and ND2 genes suggest that the COIII extension sequences have been subjected to selective pressures similar to those that have operated on the ND3 and ND2 gene sequences. Therefore, these data are consistent with the view (HOFFMANN et al. 1992 ) that the extra 46 codons of the M. californianus and M. edulis COIII genes encode a unique functional segment of the Mytilus COIII protein.

Transfer RNA genes:
Comparisons of the sequences of the M. californianus and M. edulis tRNA genes are given in Figure 2 and Table 2. The sequences of eight of the tRNA genes [Ala, Arg, His, Met, Pro, Trp, Ser(AGN), and Ser(UCN)] are highly conserved in regard to primary structure (mean, 92.1%; range, 86.2–98.5%; Table 2) and their potential to fold into secondary structures with features characteristic of mt-tRNAs (Figure 2). Compared with the M. edulis sequences, the M. californianus sequences have an average of only 4.5 nucleotide substitutions (range 1–9), 0.4 nucleotide deletions (range 0–1), and 0.3 nucleotide insertions (range 0–2; Table 2). These data clearly support the interpretation of each of these sequences as an mt-tRNA gene. The tRNA^Ser(UCN) gene, located between the COIII and tRNA^Met(AUA) genes, contains an 8 nucleotide (nt)-loop that replaces the DHU arm. A similar secondary structure is predicted for the tRNA^Ser(AGN) gene (Figure 2). Although a similarly located sequence [between the COIII and tRNA^Met(AUA) genes] in M. edulis mtDNA was noted to have stable secondary structure potential, it was not identified as a tRNA^Ser(UCN) gene by HOFFMANN et al. 1992 . Rather, they interpreted a sequence located between the ND3 and COI genes as the M. edulis mt-tRNA^Ser(UCN) gene. This sequence could be folded into a secondary structure that included most of the characteristic features expected for an mt-tRNA (Figure 2). A sequence in the corresponding location in M. californianus mtDNA can also be folded into a four-armed mt-tRNA-like structure (Figure 2). However, the sequences that form the TC arm and the 3' strand of the aminoacyl stem are not colinear in the mt-tRNA-like structures of the two species. Also, relative to the sequence contained in the M. californianus tRNA-like structure, there are 23 nucleotide substitutions in the corresponding M. edulis sequence. This degree of divergence contrasts with the highly conserved sequences of the other eight mt-tRNA genes of the two species (mean divergence = 5.1 nucleotides/tRNA; Figure 2, Table 2) and strongly supports the interpretation of this region of the two mt-tRNAs as a pseudogene that we designate tRNA^Ser(UCN).

View this table: In this window In a new window   Table 2. Comparisons of nucleotide sequences in eight tRNA genes and a tRNA gene in M. californianus F-type mtDNA with the nucleotide sequences of the corresponding tRNA genes of M. edulis F-type mtDNA

Experiments were carried out to test for the presence in M. californianus mitochondria of RNAs corresponding to the tRNA^Ser(UCN), tRNA^Ser(AGN), tRNA^Trp, and tRNA^Pro genes and the tRNA^Ser(UCN) sequence. Oligonucleotides complementary to each of these genes and the tRNA^Ser(UCN) sequence (Figure 2) were synthesized, ³²P-labeled, and hybridized to a nitrocellulose blot of whole mtRNA that had been electrophoresed in a 3% denaturing agarose gel. Examination of an autoradiograph of the product (Figure 3) revealed that the probes for each of the tRNA genes, but not that for the tRNA^Ser(UCN) sequence, had hybridized to an RNA of an approximate size (70 nt) expected for a tRNA. In contrast, the tRNA^Ser(UCN) probe hybridized to RNA molecules averaging ~2500 nt, indicating that although the tRNA^Ser(UCN) sequence is transcribed, it is rarely or never processed into a tRNA-sized molecule.

View larger version (70K): In this window In a new window Download PPT slide   Figure 3. Northern blot hybridization experiment to detect and determine the relative sizes of RNAs containing transcripts of genes for the DHU arm-replacement loop-containing tRNASer(UCN), the DHU arm-replacement loop-containing tRNASer(AGN), the four-armed tRNASer(UCN), tRNATrp, and tRNAPro predicted from gene sequences of M. californianus mtDNA. In the autoradiograph shown, each lane contains 3 µg of M. californianus mtRNA separated by electrophoresis through a 3% denaturing (2.2 M formaldehyde) agarose gel. The locations of the tRNASer(UCN), tRNASer (AGN), tRNATrp, tRNAPro, and tRNASer(UCN) gene-specific probes used are shown in Figure 2. Approximate band sizes (in kilobases to the right) were determined by comparison to a methylene blue-stained 0.24- to 7.46-kb ladder (Gibco-BRL; 5 µg/lane) produced by coelectrophoresis.

Two further Northern hybridization experiments were carried out to determine the absolute sizes of the various M. californianus mt-tRNAs. Both experiments involved hybridizing ³²P-labeled tRNA gene-specific oligonucleotides (Figure 2) to blots of mtRNA electrophoretically separated on 15% acrylamide gels (OKIMOTO and WOLSTENHOLME 1990 ). In the first experiment, to provide a size marker, E. coli tRNA^Tyr (85 nt including the DNA-encoded 3' end-located CCA) was coelectrophoresed in a separate lane in the gel. Oligonucleotides specific for tRNA^Trp, tRNA^Ser(UCN), and tRNA^Met(AUA) sequences and the E. coli tRNA^Tyr sequence were hybridized separately to blots of the gel. In the second experiment the above-mentioned three oligonucleotides, together with oligonucleotides complementary to the mt-tRNA^Ser(AGN) and mt-tRNA^Pro genes (Figure 2), were hybridized in various pairs. Examination of autoradiographs of the hybridization products (Figure 4A and Figure B) and of a plot (Figure 4C) of relative migration distances of the various bands observed in Figure 4A confirmed that for each mt-tRNA gene tested there occurs in M. californianus mitochondria a tRNA of the size predicted from the respective gene, but to which three nucleotides, presumably CCA, have been added. In the lanes probed for tRNA^Met(AUA) and for tRNA^Ser(UCN) there is also visible a second, lighter band indicating the presence of an RNA either 2 or 3 nt smaller than the main band, a size expected for tRNAs lacking either CA or CCA at their 3' ends.

View larger version (28K): In this window In a new window Download PPT slide   Figure 4. (A and B) Northern blot hybridization experiments to precisely determine the sizes of tRNAs predicted from tRNATrp, tRNASer(UCN), tRNASer(AGN), tRNAMet(AUA), and tRNAPro gene sequences of M. californianus mtDNA. In the autoradiographs shown, each lane contains 10 µg of M. californianus mtRNA (M.c) or (A only), as a standard, 200 ng of E. coli (E.c) tRNATyr (85 nt including the 3' CCA) that have been electrophoresed through a 15% polyacrylamide gel containing 8 M urea. In A, the arrowheads that bracket the top of the gels identify the common 17-cm migration position. At the top of each lane containing M. californianus (M.c) mtRNA is shown the 32P-labeled tRNA probe or mixture of probes used (Figure 2). The expected number of nucleotides in the tRNA (or tRNAs) predicted from each gene sequence, plus CCA, are shown above each lane in A, and below each lane in B. The location of E. coli tRNATyr was also determined by hybridization of a 32P-labeled oligonucleotide complementary to that tRNA. (C) A plot of the expected size (log nucleotides) of each of the three mt-tRNAs (based on the size of the respective gene and the assumption that CCA is added post-transcriptionally at the 3' end of each tRNA) and E. coli tRNATyr (including the 3' terminal CCA) vs. the measured migration distance of the tRNAs in A.

tRNA^Ser(UCN)-containing transcripts:
The observation that an oligonucleotide complementary to the tRNA^Ser(UCN) gene sequence hybridized only to RNAs of ~2500 nt indicates that the tRNA^Ser(UCN) transcript remains linked to a transcript that includes the downstream COI gene and/or the upstream ND3 gene. To resolve the relationship of the tRNA^Ser(UCN) transcript and adjacent gene transcripts, Northern blot hybridizations were carried out using M. californianus whole mtRNA and probes consisting of an oligonucleotide complementary to the 3' end of the ND3 gene (F3, Figure 2 and Figure 5A), an oligonucleotide complementary to the tRNA^Ser(UCN) gene (F5, Figure 2 and Figure 5A), and an oligonucleotide complementary to the 5' end region of the COI gene (F4, Figure 2 and Figure 5A). The results are shown in Figure 5B. The ND3 probe hybridized to a single band of RNA of ~350 nt, the size (351 nt) expected for an RNA containing only a transcript of this gene. However, the tRNA^Ser(UCN) probe and the COI probe both hybridized to RNA of ~2400 nt. Therefore, these data indicate that transcripts of the tRNA^Ser (UCN) sequence and the COI gene, and probably the downstream ATPase6 gene (assuming identical gene order in this region of M. californianus and M. edulis mtDNAs), are contained in the same RNA molecule.

View larger version (52K): In this window In a new window Download PPT slide   Figure 5. Northern blot hybridization experiment to detect and determine the relative sizes of the transcripts containing the COI gene, the tRNASer(UCN) gene, and the ND3 gene of M. californianus mtDNA. (A) Gene map to show the locations and directions of oligonucleotide probes used (arrows labeled F3 and F4, Figure 1B; and F5, Figure 2). The number of nucleotide pairs in the ND3 and tRNASer(UCN) genes, and in the intergenic spaces, and the estimated number of nucleotide pairs (1635) in the COI gene are shown. (B) Autoradiographs of Northern blots. Each lane contains 3 µg M. californianus mtRNA separated by electrophoresis through a 1% denaturing (2.2 M formaldehyde) agarose gel. The COI, ND3, and tRNASer(UCN) gene-specific oligonucleotides used to probe each lane are shown. Sizes of bands (arrow a, 0.35 kb; arrows b, 2.5 kb) were determined by comparisons to a methylene blue-stained, 0.24- to 7.46-kb ladder (5 µg/lane; Gibco-BRL; represented in kilobases to the right).

5' RACE analysis was used to locate the 5' terminus of the M. californianus COI gene-containing transcript. cDNAs were generated from whole mtRNA using random hexamers and reverse transcriptase. A poly(A) sequence was added to the 3' end of the cDNAs using terminal deoxynucleotidyl transferase, and cDNA complementary strands were synthesized from a 35-nt primer (TB 17-1) containing a 19-nt sequence that includes BglII, BamHI, and PstI sites followed by 16 T's (PONT-KINGDON et al. 1994 ). Double-stranded cDNAs of the 5' end region of the COI-gene-containing transcript were selectively amplified in a polymerase chain reaction (PCR) using as primer oligonucleotide F4 (Figure 2 and Figure 5B) that is complementary to a sequence within the 5' end region of the COI transcript and a 19-nt sequence (TB 17-2) comprising the BamHI-BglII-PstI sequence of primer TB 17-1. The PCR product was inserted into M13mp19 DNA, and five clones of the COI transcript 5' end region were sequenced. All five sequences obtained were identical and comparison of this sequence with the mtDNA sequence indicated that the 5' end nucleotide of the COI gene-containing transcript is seven nucleotides downstream from the TAA termination codon of the ND3 gene (Figure 6).

View larger version (54K): In this window In a new window Download PPT slide   Figure 6. Autoradiograph showing the nucleotide sequence of a clone of the amplification product of a 3' polyadenylated (using terminal deoxynucleotidyl transferase) cDNA of the 5' end-proximal region of the M. californianus mitochondrial transcript that contains the tRNASer(UCN) and COI genes. As the sequence shown is the transcript equivalent, the terminal reiterated nucleotide is T rather than A. The upper small arrowhead indicates the 5' terminal nucleotide of the transcript. To the right is shown the sequence downstream from the poly-T sequence (equivalent to the sequence between the two small arrowheads). The written sequence above the large arrowhead is that between the termination codon of the ND3 gene (ter ND3) and the 5' end of the tRNASer(UCN)-COI gene transcript.

These data indicate that the entire tRNA^Ser(UCN) transcript [which is separated from the 5' terminus of the COI gene by 126 nt (Figure 1 and Figure 5A)] remains linked to the COI transcript. The finding that the 5' end of the COI-gene-containing transcript is 81 nt upstream from the 5' end of the tRNA^Ser(UCN) sequence makes it unlikely that there has been selection for retention of the tRNA^Ser(UCN) sequence so that its secondary structure can act as a transcript processing site, a role suggested for interprotein gene tRNA sequences in mammalian multicistronic mtDNA transcripts (OJALA et al. 1981 ). Interestingly, however, in this regard the 5' end region of the M. californianus COI gene transcript can be folded into a stable stem and loop structure that includes the transcript cleavage site. This stable secondary structure potential is conserved in the corresponding sequence of M. edulis (F-type) mtDNA (Figure 7). Similar (but smaller) potential single stem and loop structures have been identified at the 3' ends of protein genes that are immediately followed by another protein gene in mtDNAs of a number of metazoan species including mouse, D. yakuba, C. elegans, and A. suum, and it has been suggested that these sequences may be important for precise transcript cleavage (BIBB et al. 1981 ; CLARY and WOLSTENHOLME 1985 ; OKIMOTO et al. 1992 ). HOFFMANN et al. 1992 noted secondary structure potential of both the sequence between the ND3 and tRNA^Ser(UCN) genes and the sequence between the tRNA^Ser(UCN) and COI genes in M. edulis mtDNA, but they concluded that the secondary structure of the first of these sequences was nonsignificant.

View larger version (15K): In this window In a new window Download PPT slide   Figure 7. Potential stem and loop configuration of the nucleotide sequence of the 5' end region of the COI transcript and presumed 3' end region of the ND3 transcript of F-type M. californianus mtDNA. The 5' end of the COI transcript determined by 5' RACE analysis (Figure 6) is indicated by an arrow. Transcription is from left to right. The ND3 translation termination codon is boxed. The location of the tRNASer (UCN) sequence within the COI transcript is shown. Circled nucleotides are substitutions in the corresponding transcripts of F-type M. edulis, predicted from the mtDNA. A nucleotide present in the M. edulis sequence but not in the M. californianus sequence is indicated by an arrow attached to a circled U. The predicted free energy changes shown were calculated using the free energy increments given by FREIER et al. 1986 .

	CONCLUDING REMARKS

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUDING REMARKS LITERATURE CITED

In metazoan mtDNAs, although most adjacent genes are separated by few or no nucleotides, there is a single region of apparently noncoding nucleotides that varies in size between 125 ntp and ~8000 ntp. As this region in both mammalian and Drosophila mtDNAs has been shown to contain the molecule's replication origin, and also in mammalian mtDNAs to contain promoters for transcription of multicistronic RNAs, it has been designated the control region (MONTOYA et al. 1981 ; CLAYTON 1982 , CLAYTON 1992 ; SHADEL and CLAYTON 1997 ). The larger size of some control regions of a variety of Metazoa result from the presence of repeated sequences of different sizes and numbers (for references see WOLSTENHOLME 1992A ). Cases have been reported of a repeated segment of mtDNA that contains whole or partial tRNA, rRNA, and protein genes. These include mtDNAs of different Cnemidophorus lizard species, the gecko Heteronotia binoei (MORITZ and BROWN 1986 , MORITZ and BROWN 1987 ; MORITZ 1991 ), the newt Triturus cristatus (WALLIS 1987 ), and the nematode Romanonermis culicivrax (HYMAN et al. 1988 ). However, whether or not any of these repeated genes might have a function has not been examined. The presently provided evidence that a sequence that appears to have acted as a tRNA^Ser(UCN) gene at some time in the past is now nonfunctional (at least in its original capacity) is the first direct demonstration of a pseudogene in metazoan mtDNA.

The single functional tRNA^Ser(UCN) gene of the two Mytilus species appears to be a 63-nt sequence located between the COIII and tRNA^Met(AUA) genes, which encodes a highly conserved DHU-replacement loop-containing tRNA. The interrelationship of the functional tRNA^Ser(UCN) gene and the tRNA^Ser(UCN) gene remains unclear. Sequence comparisons do not suggest that one is a duplicate of the other. As there is considerable divergence between the M. californianus and M. edulis mt-tRNA^Ser(UCN) gene sequences it remains possible that these sequences are the remnants of a gene for a tRNA other than one that recognizes UCN codons. However, this is not supported by sequence comparisons to other M. edulis mt-tRNA genes. It has been suggested that the tRNA^Ser(UCN) genes of the gastropods Euhadra herklotsi and Cepaea nemoralis (but not Albineria coerulea), the polyplacophoran mollusk Katherina tunicata, and the annelid Lumbricus terrestris encode DHU arm-replacement loop-containing structures (BOORE and BROWN 1994 , BOORE and BROWN 1995 ; HATZOGLOU et al. 1995 ; TERRETT et al. 1996 ; YAMAZAKI et al. 1997 ). However, the K. tunicata and L. terrestris predicted tRNAs can be folded to give the usual U₈, A₉, or G₉ between the aminoacyl and DHU arms, and a DHU arm with a stem of 3 ntp and a loop of 3 nt, but with three (rather than the usual one) nucleotides between the DHU and anticodon arms. Comparisons of the DHU arm-replacement loop-containing tRNA^Ser (UCN) genes of the two Mytilus species and the tRNA^Ser (UCN) gene of K. tunicata reveal conservation of 4/5 ntp in the TC stems, 3/5 ntp in the anticodon stems, and 7/7 nt in the anticodon loops.

The mt-tRNA^Ser(UCN) genes of arthropods, echinoderms, and chordates so far examined encode a four-armed tRNA and the primary sequences of these tRNAs of the latter two phyla are close to those of standard tRNAs. However, as in mtDNAs of mollusks and an annelid, genes for tRNAs expected to recognize UCN codons that have a DHU arm-replacement loop have also been found in nematode mtDNAs (OKIMOTO and WOLSTENHOLME 1990 ; OKIMOTO et al. 1992 ). Considerations of the various proposed relationships of invertebrate phyla (see, for example, WILSON et al. 1978 ; NIELSEN 1995 ) do not lend support to the view that the observed distribution of the genes for the two structural kinds of mt-tRNA^Ser(UCN) could have resulted from a single loss of a DHU arm-forming sequence from a mt-tRNA^Ser(UCN) gene during invertebrate evolution. Because it seems improbable that, once lost, a DHU arm-forming sequence of standard tRNAs would be regenerated (discussed in WOLSTENHOLME et al. 1987 ), then the most likely explanation for the presence of a DHU arm-replacement loop-containing mt-tRNA^Ser(UCN) in nematodes, some (but not all) mollusks, and an annelid, is that the gene for this kind of tRNA has been generated by multiple independent sequence losses. This would represent an unusual case of convergent structural evolution in metazoan mt-genomes.

The mtDNA sequence interpreted by HOFFMANN et al. 1992 as an M. edulis tRNA^Met(AUA) gene is highly conserved in M. californianus mtDNA. Our finding that an RNA molecule of the sequence and size expected for a mature tRNA encoded by the M. californianus tRNA^Met(AUA) gene occurs in M. californianus mitochondria provides strong evidence that this gene is functional.

HOFFMANN et al. 1992 stated that AUA specifies methionine in the M. edulis mt-genetic code, but they did not explain how they reached this conclusion. In fact, comparisons of ATA codons in the partial sequences of M. edulis Cyt b, COI, and COII genes (more highly conserved genes in metazoan mtDNAs), with similarly located codons in C. elegans, D. yakuba, and mouse mtDNAs, do not provide compelling evidence that AUA is more likely to encode methionine than isoleucine or leucine. However, interchange of ATA and ATG codons in the partial sequences of M. californianus F-type and M-type COI and ND5 genes (BEAGLEY et al. 1997 ) does strongly support the coding specificity of AUA as methionine. Of 15 ATA codons that occur in these M-type sequences, 6 have ATA and 7 have ATG in the corresponding locations of the F-type sequences; of 8 ATG codons that occur in the M-type sequences, 3 have ATG and 4 have ATA in the corresponding locations of the F-type sequences. Therefore, assuming that the tRNA^Met (AUG) gene is functional, which seems likely, as suggested by HOFFMANN et al. 1992 , Mytilus mtDNAs are unique among known metazoan mtDNA sequences in that they contain two methionine-specifying tRNAs. Based on sequence similarities in the aminoacyl and DHU arms, HOFFMANN et al. 1992 suggested that the tRNA^Met(AUA) gene may have arisen as a duplication of the tRNA^Met(AUG) gene.

It is uncertain why Mytilus alone among Metzoa examined to date should utilize two tRNAs to recognize methionine codons. In M. edulis, the tRNA^Met(AUG) would be expected to recognize the AUG initiation codons of 10 of the 12 mt-protein genes and internal AUG codons. In regard to its initiation function it is noteworthy that the predicted tRNA would have three consecutive G-C base pairs at the base of the anticodon stem. This is a characteristic feature of the E. coli initiator tRNA^Met and it has been shown that these three G-C base pairs are essential to initiation function (RAJBHANDARY and CHOW 1995 ). Also, three, two, or one of these G-C nucleotide pairs occur in the single mt-tRNA^Met of various other Metazoa (ANDERSON et al. 1982 ; CLARY et al. 1982 ; WOLSTENHOLME et al. 1994 ; BEAGLEY et al. 1995 ). All but one of the ATA codons in the M. edulis mtDNA molecule are internal. Only in the ND5 gene does it appear likely that the initiation codon is ATA. We have been unable to detect any convincing similarity between the Mytilus tRNA^Met(AUA) gene and the E. coli elongator tRNA^Met that recognizes only internal AUG codons.

	FOOTNOTES

¹ Present address: Department of Poultry Science, University of Arkansas, Fayetteville, AK 72701.

	ACKNOWLEDGMENTS

We thank Jane L. Macfarlane for excellent technical assistance and for discussions, and Robert Schackmann for oligonucleotides [partially subsidized by National Institutes of Health (NIH) grant CA-42014]. This work was supported by NIH grant GM-18375.

Manuscript received September 21, 1998; Accepted for publication March 5, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUDING REMARKS LITERATURE CITED

ANDERSON, S., A. T. BANKIER, B. G. BARRELL, M. H. L. DEBRUIJN, A. R. COULSON et al., 1982 Comparison of the human and bovine mitochondrial genomes, pp. 5–43 in Mitochondrial Genes, edited by P. SLONIMSKI, P. BORST and G. ATTARDI. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

ARNASON, U., X. XU, and A. GULLBERG, 1996  Comparison between the complete mitochondrial DNA sequences of Homo and the common chimpanzee based on nonchimeric sequences. J. Mol. Evol. 42:145-152[Medline].

ASAKAWA, S., H. HIMENO, K.-I. MIURA, and K. WATANABE, 1995  Nucleotide sequence and gene organization of the starfish Asterina pectinifera mitochondrial genome. Genetics 140:1047-1060[Abstract].

BEAGLEY, C. T., J. L. MACFARLANE, G. A. PONT-KINGDON, R. OKIMOTO, N. A. OKADA et al., 1995 Mitochondrial genomes of Anthozoa (Cnidaria), pp. 149–153 in Progress in Cell Research: Symposium on "Thirty Years of Progress in Mitochondrial Bioenergetics and Molecular Biology," edited by F. PALMIERI, S. PAPA, C. SACCONE and N. GADALETA. Elsevier Science BV, Amsterdam, The Netherlands.

BEAGLEY, C. T., N. A. OKADA, and D. R. WOLSTENHOLME, 1996  Two mitochondrial group I introns in a metazoan, the sea anemone Metridium senile: one intron contains genes for subunits 1 and 3 of NADH dehydrogenase. Proc. Natl. Acad. Sci. USA 93:5619-5623[Abstract/Free Full Text].

BEAGLEY, C. T., K. A. TAYLOR, and D. R. WOLSTENHOLME, 1997  Gender-associated diverse mitochondrial DNA molecules of the mussel Mytilus californianus.. Curr. Genet. 31:318-324[Medline].

BEAGLEY, C. T., R. OKIMOTO, and D. R. WOLSTENHOLME, 1998  The mitochondrial genome of the sea anemone Metridium senile (Cnidaria): introns, a paucity of tRNA genes, and a near standard genetic code. Genetics 148:1091-1108[Abstract/Free Full Text].

BIBB, M. J., R. A. VAN ETTEN, C. T. WRIGHT, M. W. WALBERG, and D. A. CLAYTON, 1981  Sequence and gene organization of mouse mitochondrial DNA. Cell 26:167-180[Medline].

BOORE, J. L. and W. M. BROWN, 1994  Complete DNA sequence of the mitochondrial genome of the black chiton Katharina tunicata.. Genetics 138:423-443[Abstract].

BOORE, J. L. and W. M. BROWN, 1995  Complete sequence of the mitochondrial DNA of the annelid worm Lumbricus terrestris.. Genetics 141:305-319[Abstract].

BROWN, G. G. and M. V. SIMPSON, 1982  Novel features of animal mtDNA evolution as shown by sequences of two rat cytochrome oxidase subunit II genes. Proc. Natl. Acad. Sci. USA 79:3246-3250[Abstract/Free Full Text].

BROWN, W. M., E. M. PRAGER, and W. A. WILSON, 1982  Mitochondrial DNA sequences of primates: tempo and mode of evolution. J. Mol. Evol. 18:225-239[Medline].

CANTATORE, P., M. ROBERTI, G. RAINALDI, M. N. GADALETA, and C. SACCONE, 1989  The complete nucleotide sequence, gene organization, and genetic code of mitochondrial genome of Paracentrotus lividus.. J. Biol. Chem. 264:10965-10975[Abstract/Free Full Text].

CLARY, D. O. and D. R. WOLSTENHOLME, 1985  The mitochondrial DNA molecule of Drosophila yakuba: nucleotide sequence, gene organization, and genetic code. J. Mol. Evol. 22:252-271[Medline].

CLARY, D. O., J. M. GODDARD, S. C. MARTIN, C. M-R. FAURON, and D. R. WOLSTENHOLME, 1982  Drosophila mitochondrial DNA: a novel gene order. Nucleic Acids Res. 10:6619-6637[Abstract/Free Full Text].

CLAYTON, D. A., 1982  Replication of animal mitochondrial DNA. Cell 28:693-705[Medline].

CLAYTON, D. A., 1992 Transcription and replication of animal mitochondrial DNAs, pp. 217–232 in International Review of Cytology, Vol. 141, edited by D. R. WOLSTENHOLME and K. W. JEON. Academic Press, New York.

CROZIER, R. H. and Y. C. CROZIER, 1993  The mitochondrial genome of the honey bee Apis melifera: complete sequence and genome organization. Genetics 133:97-117[Abstract].

DALE, R. M. K., B. MCCLURE, and J. P. HOUGHINS, 1985  A rapid single-stranded cloning strategy for producing overlapping clones for use in DNA sequencing: application to sequencing the corn mitochondrial 18S rDNA. Plasmid 13:31-40[Medline].

DESALLE, R., T. FREEDMAN, E. M. PRAGER, and A. C. WILSON, 1987  Tempo and mode of sequence evolution in mitochondrial DNA of Hawaiian Drosophila.. J. Mol. Evol. 26:157-164[Medline].

FISHER, C., and D. O. F. SKIBINSKI, 1990 Sex-biased mitochondrial DNA heteroplasmy in the marine mussel Mytilus. Proc. R. Soc. Lond. Ser B 149–156.

FREIER, S. M., R. KIERZEK, J. A. JAEGER, N. SUGIMOTO, and M. H. CARUTHERS et al., 1986  Improved free-energy parameters for predictions of RNA duplex stability. Proc. Natl. Acad. Sci. USA 83:9373-9377[Abstract/Free Full Text].

FROHMAN, M. A., 1990 RACE: rapid amplification of cDNA ends, pp.28–38 in PCR Protocols, a Guide to Methods and Applications, edited by M. A. INNIS, D. H. GELFAND, J. J. SNINSKY and T. J. WHITE. Academic Press, San Diego.

GAREY, J. R. and D. R. WOLSTENHOLME, 1989  Platyhelminth mitochondrial DNA: evidence for early evolutionary origin of a tRNA-ser(AGN) that contains a dihydrouridine arm replacement-loop, and of serine-specifying AGA and AGG codons. J. Mol. Evol. 28:374-387[Medline].

HATZOGLOU, E., G. C. RODAKIS, and R. LECANIDOU, 1995  Complete sequence and gene organization of the mitochondrial genome of the land snail Albinaria coerulea.. Genetics 140:1353-1366[Abstract].

HOFFMANN, R. J., J. L. BOORE, and W. M. BROWN, 1992  A novel mitochondrial genome organization for the blue mussell, Mytilus edulis.. Genetics 131:397-412[Abstract].

HYMAN, B. C., J. L. BECK, and K. C. WEISS, 1988  Sequence amplification and gene rearrangement in parasitic nematode mitochondrial DNA. Genetics 120:707-712[Abstract/Free Full Text].

JACOBS, H. T., D. J. ELLIOT, V. B. MATH, and A. FARQUHARSON, 1988  Nucleotide sequence and gene organization of sea urchin mitochondrial DNA. J. Mol. Biol. 202:185-217[Medline].

KRETTEK, A., A. GULLBERG, and U. ARNASON, 1995  Sequence analysis of the complete mitochondrial DNA molecule of the hedgehog, Erinaceus europaeus, and the phylogenetic position of the Lipotyphla. J. Mol. Evol. 441:952-957.

MONTOYA, J., D. OJALA, and G. ATTARDI, 1981  Distinctive features of the 5' terminal sequences of the human mitochondrial mRNAs. Nature 290:465-470[Medline].

MORITZ, C., 1991  Evolutionary dynamics of mitochondrial DNA duplication in parthenogenetic geckos, Heteronotia binoei.. Genetics 129:221-230[Abstract].

MORITZ, C. and W. M. BROWN, 1986  Tandem duplication of D-loop and ribosomal RNA sequences in lizard mitochondrial DNA. Science 233:1425-1427[Abstract/Free Full Text].

MORITZ, C. and W. M. BROWN, 1987  Tandem duplications in animal mitochondrial DNAs: variation in incidence and gene content among lizards. Proc. Natl. Acad. Sci. USA 84:7183-7187[Abstract/Free Full Text].

NIELSEN, C., 1995 Animal Evolution: Interrelationships of Living Phyla. Oxford University Press, Oxford.

OJALA, D., J. MONTOYA, and G. ATTARDI, 1981  tRNA punctuation model of RNA processing in human mitochondria. Nature 290:470-474[Medline].

OKIMOTO, R. and D. R. WOLSTENHOLME, 1990  A set of tRNAs that lack either the TC arm or the dihydrouridine arm: towards a minimal tRNA adaptor. EMBO J. 9:3405-3411[Medline].

OKIMOTO, R., H. M. CHAMBERLIN, J. L. MACFARLANE, and D. R. WOLSTENHOLME, 1991  Repeated sequence sets in mitochondrial DNA molecules of root knot nematodes (Meloidogyne): nucleotide sequences, genome location and potential for host-race identification. Nucleic Acids Res. 19:1619-1626[Abstract/Free Full Text].

OKIMOTO, R., J. L. MACFARLANE, and D. R. WOLSTENHOLME, 1992  The mitochondrial genomes of two nematodes, Caenorhabditis elegans and Ascaris suum.. Genetics 130:471-498[Abstract].

PONT-KINGDON, G. A., C. T. BEAGLEY, R. OKIMOTO, and D. R. WOLSTENHOLME, 1994  Mitochondrial DNA of the sea anemone, Metridium senile (Cnidaria): prokaryote-like genes for tRNA^f-Met and small subunit ribosomal RNA, and standard genetic code specificities for AGR and ATA codons. J. Mol. Evol. 39:387-399[Medline].

RAJBHANDARY, U., and M. C. CHOW, 1995 Initiator tRNAs and initiation of protein synthesis, pp. 511–528 in tRNA Structure, Biosynthesis and Function, edited by D. SÖLL and U. RAJBHANDARY. American Society of Microbiology, Washington, DC.

SANGER, F., S. NICKLEN, and A. R. COULSON, 1977  DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467[Abstract/Free Full Text].

SHADEL, G. S. and D. A. CLAYTON, 1997  Mitochondrial DNA maintenance in vertebrates. Annu. Rev. Biochem. 66:409-435[Medline].

SKIBINSKI, D. O. F., C. GALLAGHER, and C. M. BEYNON, 1994a  Mitochondrial DNA inheritance. Nature 368:817-818[Medline].

SKIBINSKI, D. O. F., C. GALLAGHER, and C. M. BEYNON, 1994b  Sex-limited mitochondrial DNA transmission in the marine mussel Mytilus edulis.. Genetics 138:801-809[Abstract].

TERRETT, J. A., S. MILES, and R. H. THOMAS, 1996  Complete DNA sequence of the mitochondrial genome of Cepaea nemoralis (Gastropoda: Pulmonata). J. Mol. Evol. 242:160-168.

WAHLEITHNER, J. A. and D. R. WOLSTENHOLME, 1987  Mitochondrial plasmid DNAs of broad bean: nucleotide sequences, complex secondary structures, and transcription. Curr. Genet. 12:55-67[Medline].

WALLIS, M. J., 1987  Mitochondrial DNA insertion polymorphism and germline heteroplasmy in the Triturus cristatus complex. Heredity 58:229-238.

WILSON, E. O., T. EISNER, W. R. BRIGTS, R. E. DICKERSON, R. L. METZENBERG et al., 1978 Life on Earth, Ed. 2. Sinauer, Sunderland, Mass.