Trichinella spiralis mtDNA: A Nematode Mitochondrial Genome That Encodes a Putative ATP8 and Normally Structured tRNAs and Has a Gene Arrangement Relatable to Those of Coelomate Metazoans

Corresponding author: Dennis V. Lavrov, Département de Biochimie, Université de Montréal, C.P. 6128, Montréal, QC H3C3J7, Canada., dlavrov{at}bch.umontreal.ca (E-mail)

Communicating editor: N. TAKAHATA

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS LITERATURE CITED

The complete mitochondrial DNA (mtDNA) of the nematode Trichinella spiralis has been amplified in four overlapping fragments and 16,656 bp of its sequence has been determined. This sequence contains the 37 genes typical of metazoan mtDNAs, including a putative atp8, which is absent from all other nematode mtDNAs examined. The genes are transcribed from both mtDNA strands and have an arrangement relatable to those of coelomate metazoans, but not to those of secernentean nematodes. All protein genes appear to initiate with ATN codons, typical for metazoans. Neither TTG nor GTT start codons, inferred for several genes of other nematodes, were found. The 22 T. spiralis tRNA genes fall into three categories: (i) those with the potential to form conventional "cloverleaf" secondary structures, (ii) those with TC arm + variable arm replacement loops, and (iii) those with DHU-arm replacement loops. Mt-tRNA(R) has a 5'-UCG-3' anticodon, as in most other metazoans, instead of the very unusual 5'-ACG-3' present in the secernentean nematodes. The sequence also contains a large repeat region that is polymorphic in size at the population and/or individual level.

MITOCHONDRIAL DNAs (mtDNAs) vary extensively in size and gene content across diverse eukaryotic groups; those of animals (Metazoa), however, are surprisingly uniform (LANG et al. 1999 ). A typical metazoan mtDNA is a circular molecule of 14–18 kb and encodes 37 genes: 13 for proteins [subunits 6 and 8 of the F₀ ATPase (atp6 and atp8), cytochrome c oxidase subunits 1–3 (cox1–cox3), apocytochrome b (cob), and NADH dehydrogenase subunits 1–6 and 4L (nad1–6 and nad4L)]; 2 for ribosomal RNAs [small and large subunit rRNAs (rrnS and rrnL)]; and 22 for tRNAs [designated by the one-letter code, with the two leucine and two serine tRNAs differentiated by their anticodon sequences (uag/uaa and ucu/uga, respectively)] (WOLSTENHOLME 1992 ). The arrangement of these genes in metazoan mtDNA is relatively well conserved, with some blocks of genes shared even among different phyla (BOORE 1999 ).

One metazoan group with mtDNA that deviates from the pattern just described is the phylum Nematoda. Complete mitochondrial gene arrangements are available for four nematode species: Ascaris suum and Caenorhabditis elegans (OKIMOTO et al. 1992 ), Meloidogyne javonica (OKIMOTO et al. 1991 ), and Onchocerca volvulus (KEDDIE et al. 1998 ); complete sequences are available for all except M. javonica. With the exception of the latter species, which has an unusually large noncoding region, the nematode mtDNAs are remarkably compact, ranging in size from 13,747 bp for O. volvulus to 14,284 bp for A. suum. These nematode mitochondrial genomes share several unusual features: most of their protein, rRNA, and tRNA genes are smaller than in other metazoans and one (atp8) is missing altogether; several of their protein genes initiate at the unusual start codons GTT and TTG, and none at an orthodox ATG codon (OKIMOTO et al. 1990 ); and all tRNAs encoded have secondary structures that lack either a TC or a DHU arm (OKIMOTO and WOLSTENHOLME 1990 ). The gene arrangements of these nematode mtDNAs are also very unusual: although the four share some arrangements among themselves, they differ at nearly every gene boundary from all other metazoans (BOORE 1999 ). An extreme case of nematode mitochondrial genome organization has been reported recently for the potato cyst nematode Globodera pallida (ARMSTRONG et al. 2000 ). The mitochondrial genome of this animal is multipartite and exists as a population of small circular DNAs of different sizes and gene contents, a unique organization among studied metazoans. The above features have helped to reinforce a widely held view that nematodes are a bizarre group, with unclear phylogenetic affinities to the major metazoan lineages.

The four species of nematodes for which complete mtDNA sequences and/or complete gene arrangements have been published are all in the class Secernentea, one of two traditionally recognized nematode classes (BRUSCA and BRUSCA 1990 ). There is far less information about mtDNA from representatives of another class, Adenophorea, which is often considered more primitive. A limited amount of sequence and gene arrangement data is available for the adenophorean species Romanomermis culicivorax (AZEVEDO and HYMAN 1993 ), and partial cox1 sequences are available from several species in the genus Trichinella (NAGANO et al. 1999 ). To complicate the matter, the monophyly of the class Adenophorea is questionable: no synapomorphies support its monophyly, and both morphological and DNA sequence data indicate that it may be paraphyletic (ADAMSON 1987 ; MALAKHOV 1994 ; BLAXTER et al. 1998 ; VORONOV et al. 1998 ). We describe here the mitochondrial genome of Trichinella spiralis, the first comprehensively studied mtDNA from a nematode outside the class Secernentea.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS LITERATURE CITED

mtDNA amplification and sequencing:
Total DNA from 10,000 larvae of the nematode T. spiralis was a gift from D. Despommier. Conserved primers designed in our laboratory were used to amplify portions of cox3, cob, nad5, and nad1. We designed two primers going in opposite directions for each of these gene fragments, designated:

• Trichi-cox3-F1, 5'-TACGTAGAATACCACACATCCAC-3';

• Trichi-cox3-R1, 5'-ATTCTTCCGTTTACTCCTCTCGA-3';

• Trichi-cob-F1, 5'-CAATCCATTAGGTACACACTCAC-3';

• Trichi-cob-R1, 5'-CCTGTAATTCTGTATCCTCCTCA-3';

• Trichi-nad5-F1, 5'-TTGGTAGTTGTGGTGGGTAAGTC-3';

• Trichi-nad5-R1, 5'-AACAACACCACCAACCTGAGCAC-3';

• Trichi-nad1-F1, 5'-CACTAGCACTTACCATTCCAGCC-3';

• Trichi-nad1-R1, 5'-GGTTGTTGCTAGGTTGTATGAGTC-3'.

Using a Perkin Elmer (Norwalk, CT) XL PCR kit and primer pairs cox3-F1-nad1-R1, cox3-R1-cob-F1, and cob-R1-nad5-R1, we amplified regions between nad1 and cox3 (4.4 kb), cox3 and cob (3.6 kb), and cob and nad5 (3.0 kb), respectively. Each PCR reaction yielded a single band when visualized with ethidium bromide staining after electrophoresis in a 1% or 0.7% agarose gel. Amplification of the remaining portion of mtDNA, downstream from nad1 and nad5, was very problematic. The flanking sequences of this region were amplified using Step-Out PCR (WESLEY and WESLEY 1997 ), and the entire region was later amplified using a TaKaRa LA Taq kit (Takara Shuzo Co.), but several products of different sizes were produced in all of the latter amplifications.

PCR reaction products were purified by three serial passages through Ultrafree [30,000 nominal molecular weight limit (NMWL)] columns (Millipore, Bedford, MA) and used as templates in dye-terminator cycle-sequencing reactions according to supplier's (Perkin Elmer) instructions. Both strands of each amplification product were sequenced by primer walking, using an ABI Prizm 377 automated DNA sequencer (Perkin Elmer). The sequence has been submitted to GenBank under accession no. AF293969.

Sequence analysis:
Sequences were assembled using Sequencing Analysis and Sequence Navigator software (Perkin Elmer) and analyzed with MacVector 6.5 and GCG (Oxford Molecular Group) programs. Protein and ribosomal RNA gene sequences were identified by their similarity to published metazoan mtDNA sequences; tRNA genes were recognized initially by their potential to be folded into tRNA-like secondary structures, after which they were identified specifically by their anticodon sequences. The secondary structures of rRNA genes were derived by analogy to other published rRNA gene structures and drawn using the RnaViz program (DE RIJK and DE WACHTER 1997 ).

The amino acid sequences were inferred from mitochondrial protein genes of T. spiralis, A. suum, C. elegans (OKIMOTO et al. 1992 ), O. volvulus (KEDDIE et al. 1998 ), and Limulus polyphemus (LAVROV et al. 2000 ) and aligned using the ClustalW program (THOMPSON et al. 1994 ) in MacVector 6.5 (gap penalty = 5; extension penalty = 1; no gap separation distance; all other options at default settings), and percentage of their similarity was determined. Amino acid and codon usage on the different mtDNA strands were compared using ² analyses of contingency tables; when a 2 x 2 contingency table was used, the Yates correction for continuity was applied (YATES 1934 ). To illustrate the quantitative difference, the odds ratios (ORs) were calculated as the ratio of a particular amino acid (group of amino acids, codon, group of codons) to all other amino acids (codons) for one strand, divided by the same ratio for the second strand.

	RESULTS AND DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS LITERATURE CITED

Genome size and organization:
The estimated size of T. spiralis mtDNA varies between ca. 21 and 24 kb. This variation is due to an apparent size polymorphism of a region downstream from nad1 and nad2, as indicated by the results of PCR amplification and by Southern hybridization analysis (data not shown). Partial sequencing from the two ends of this region revealed the presence of two repeat units of 1323 bp, the first overlapping nad1 by 3 nucleotides and the second ending 153 nucleotides downstream from nad2 (Fig 1). The repeat unit closest to nad2 contains 50 bp of the inferred trnK; the remaining 12 bp of that gene is located in the adjacent sequence. The partially sequenced region between these repeat units includes smaller repeats and homopolymer runs, which interfere with further sequencing. The results of PCR amplifications using one primer complementary to a sequence inside the large repeat unit and a second primer complementary to a sequence in either nad1 or nad2 suggest the presence of additional large repeat units in this region (data not shown). The whole region downstream from nad1 and nad2 will, hereafter, be referred to as the repeat region. The sequence of the T. spiralis mtDNA, excluding the repeat region, is 13,902 bp in size and encodes 36 of the 37 genes (all but trnK).

View larger version (36K): In this window In a new window Download PPT slide   Figure 1. Gene map of T. spiralis mtDNA. Protein and rRNA genes are abbreviated as in the text; tRNA genes are abbreviated using the one-letter amino acid code; the two leucine and two serine tRNA genes are additionally identified by their anticodon sequences with trnL(uag) marked as L1, trnL(uaa)- as L2, trnS(ucu)- as S1, and trnS(uga)- as S2. Arrows indicate the direction of transcription of each gene. Positive numbers at gene boundaries indicate the number of intergenic nucleotides; negative numbers indicate the number of overlapping nucleotides. Asterisks mark incomplete stop codons (T or TA). The size of the repeat-containing region is not to scale; the unsequenced portion of this region is demarcated by curved lines.

In contrast with other nematodes studied, the gene arrangement of T. spiralis mtDNA can be easily related to those of several other metazoans by invoking a moderate number of rearrangements. The greatest similarity is to the primitive arthropod gene arrangement [exemplified by L. polyphemus (STATON et al. 1997 )] with which it shares three different blocks of three or more genes and four additional two-gene boundaries (Fig 2). In addition, the location of two tRNA(S) genes is similar in the two genomes: one is situated immediately downstream from cob, and the second is in the tRNA cluster downstream from nad3. However, the specificity of these genes is reversed in the two species [cob-trnS(ucu)/nad3-trnS(uga) in T. spiralis and cob-trnS(uga)/nad3-trnS(gcu) in L. polyphemus; Fig 2]. Mechanistically, this reversal could have arisen by either multiple rearrangements or anticodon switching. The latter hypothesis is supported by the phylogenetic analysis of mitochondrial trnS sequences, which tends to group T. spiralis trnS(ucu) and trnS(uga) with the trnS(kcu) and trnS(uga), respectively, of other animals (data not shown). A plausible mechanism for anticodon switching, involving tRNA gene duplication with consecutive changes in the anticodon sequence, has been proposed (CANTATORE et al. 1987 ). However, since the anticodons of the two serine tRNAs (UCU and UGA) differ at two positions and since a change at either would create an anticodon for a different amino acid, two simultaneous substitutions would be needed for conversion of one serine tRNA gene to the other. Evidence for a relatively high frequency of such mutational events has been recently provided (AVEROF et al. 2000 ).

View larger version (33K): In this window In a new window Download PPT slide   Figure 2. Comparison of gene arrangements in the mtDNAs of A. suum (OKIMOTO et al. 1992 ), L. polyphemus (STATON et al. 1997 ), and T. spiralis. Only coding sequences are shown. Protein and rRNA genes are indicated by open boxes, tRNA genes by hatched boxes. No pairwise gene arrangement is identical between A. suum and T. spiralis or A. suum and D. yakuba. Blocks of three or more genes shared between T. spiralis and D. yakuba are underlined and interconnected with arrows, and shared boundaries between two genes outside these blocks are marked with asterisks. All abbreviations and other symbols are as in Fig 1.

AGUINALDO et al. 1997 proposed that arthropods, nematodes, and several other "minor" phyla form a monophyletic group, the Ecdysozoa. While the T. spiralis mitochondrial gene arrangement is most similar to that primitive for arthropods, we found no synapomorphies in this or other metazoan gene arrangements that either support or refute the Ecdysozoa hypothesis.

Nucleotide composition:
The A + T content of T. spiralis mtDNA, excluding the repeat region, is 65.2%, lower than those reported for other nematodes. Each of the large repeat units is 77.7% A + T. The two strands of T. spiralis mtDNA have significantly different nucleotide composition. The strand that contains the sense sequence of nine mRNAs, both ribosomal RNAs, and 12 tRNAs (hereafter referred to as the -strand) is AC rich (i.e., its A/T and C/G ratios are >1) and the other strand (hereafter the ß-strand) is GT rich. The difference is especially pronounced in the region containing coding sequences on the ß-strand (clockwise, from nad2 to trnP in Fig 1; nucleotides 1–4307 in the GenBank sequence) and is less extreme in the repeat region. The corresponding GC and AT skews [GC skew = (G - C)/(G + C) and AT skew = (A - T)/(A + T); PERNA and KOCHER 1995] for these two regions are -0.59, 0.48 and -0.25, 0.03, respectively; for the rest of the genome, GC skew = -0.33 and AT skew = 0.14. If the AT and GC skews are a consequence of asymmetrical mtDNA replication, as has been suggested (BROWN and SIMPSON 1982 ; ASAKAWA et al. 1991 ; REYES et al. 1998 ), the difference in nucleotide composition between the two strands of T. spiralis mtDNA implies that ß-strand replication precedes -strand replication in this mtDNA. This would be similar to the situation in arthropod and vertebrate mtDNAs, in which the sense sequence of most genes is located on the AC-rich strand that is also the lagging strand in mtDNA replication. By contrast, the sense sequence of all genes in the secernentean nematode mtDNAs that have been studied is located on the GT-rich strand, which, by the above criterion, is also the leading strand in mtDNA replication.

Protein genes
Size and sequence similarity: Thirteen protein genes are commonly present in metazoan mtDNAs; however, one of them (atp8) is absent from all nematode mtDNAs previously examined. Eleven T. spiralis protein genes (all but atp6 and atp8) were easily identified by sequence comparisons with other species' mtDNAs. In addition, two open reading frames (ORFs) were tentatively identified as atp6 and atp8. The first ORF, located between rrnL and cox3, has some sequence similarity to other metazoan atp6's, but is significantly larger [276 sense codons vs. 199 in A. suum and C. elegans (OKIMOTO et al. 1992 ), 224 in L. polyphemus (LAVROV et al. 2000 ), and 226 in human (ANDERSON et al. 1981 )]. The second ORF, located between trnD and nad3, also has some sequence similarity to other metazoan atp8's, but is smaller than those (41 sense codons vs. 51 in L. polyphemus and 68 in human). In addition to sequence similarities, the hydropathy profiles of the ORF-encoded proteins are similar to those of ATP6 and ATP8 in Limulus and human (Fig 3). The similarities are further enhanced by ending the presumptive atp6 with an abbreviated stop codon 45 codons upstream from the end of the ORF and by starting atp8 at the ATC codon 18 nucleotides (nt) downstream from the beginning of the ORF. This would result in a putative ATP6 of 232 amino acids, with a well-conserved C-terminal motif (EX₂-VX₃QX₂FX₂LX₃YX₂EX_n), and a putative ATP8 of 35 amino acids, with a partially conserved sequence at the N terminus. Both with and without the first 6 amino acids, the putative ATP8 would be shorter than its counterparts in other species; however, most of the size reduction is in the positively charged, hydrophilic domain, which is known to vary greatly in length in this protein (GRAY et al. 1998 ). Additional evidence that both ORFs encode functional proteins comes from the similarity in their codon nucleotide composition with those of other genes for -strand-encoded proteins (Fig 4A, Fig C, and Fig F), which have T-rich second and AC-rich third codon positions. Similar patterns of nucleotide usage prevail when only the first or, to a lesser extent, the last 50 codons are analyzed for the presumptive atp6 (Fig 4D and Fig E), which argues against the presence of an internal initiation codon and/or truncated stop codon in this ORF. Taking the analyses of hydropathy and codon nucleotide composition together, it is unclear if the presumptive atp6 ends with an incomplete termination codon or has a greatly expanded 3' end.

View larger version (36K): In this window In a new window Download PPT slide   Figure 3. Comparisons of T. spiralis, L. polyphemus, and human ATP6 and ATP8 hydropathy profiles. Each was calculated by the method of KYTE and DOOLITTLE 1982 . Window size = 7. Numbers below profiles designate amino acid positions in each protein. Arrows indicates a possible alternative end of ATP6 and a possible alternative beginning of ATP8 in T. spiralis, both of which would increase the similarity in hydropathy of the T. spiralis proteins to those of L. polyphemus and human.

View larger version (31K): In this window In a new window Download PPT slide   Figure 4. Comparisons of nucleotide composition at first, second, and third codon positions of T. spiralis genes for -strand-encoded proteins (except ATP6 and ATP8) (A), ß-strand-encoded proteins (B), putative ATP6 (C–E), and putative ATP8 (F). For atp6, nucleotide composition is shown for all codons (C), for the first 50 codons (D), and for the last 50 codons (E). Nucleotide percentages: black bars, T; dark gray bars, C; light gray bars, A; white bars, G. 1, 2, and 3 indicate first, second, and third codon positions, respectively; 3* indicates third codon positions in fourfold degenerate codon families.

Most mitochondrial protein genes in T. spiralis are slightly larger than their counterparts in other nematodes and slightly smaller than those in L. polyphemus (Table 1). The differences are within 5% of the T. spiralis gene length for all genes except atp6 and atp8 (discussed above); cox1 and nad3 (6.4% longer and 7.4% shorter, respectively, in O. volvulus); nad2, nad4, nad4L, and nad5 (>5% longer in L. polyphemus); and nad6 (8.3 and 7.6% shorter in A. suum and C. elegans, respectively). The comparison of amino acid sequences inferred from the protein genes of T. spiralis with those of three other nematode species and L. polyphemus revealed cox1 as the most conserved and atp6, nad2, and nad6 as the least-conserved genes, with amino acid identities of the encoded proteins ranging from 8.3 to 59.9% (Table 1). The size differences and low amino acid similarity of the putative ATP6 and ATP8 proteins made their alignments difficult, and the reported sequence identities for them should be regarded as preliminary estimates.

Translation initiation and termination signals: An ATG, ATT, or ATA codon occurs at the beginning of all inferred protein genes in T. spiralis mtDNA. Neither TTG nor GTT, both of which were reported as initiation codons of several protein genes in other nematodes (OKIMOTO et al. 1990 ), are used as such in T. spiralis. The use of ATG as an initiation codon in five mitochondrial protein genes of T. spiralis is also a departure from O. volvulus, A. suum, and C. elegans, none of which use it in this function (OKIMOTO et al. 1992 ; KEDDIE et al. 1998 ). Among the five T. spiralis genes initiated by ATG, three (cox1, cox2, and cob) share a sequence motif [5'-ATGATAAAATSA-3' (S = G or C)] at their 5' ends, and a fourth (cox3) has a slightly modified version of this motif (5'-ATGAATAAATCC-3'). The fifth gene with an ATG initiation codon (nad4) does not share this pattern.

All genes except cob and nad4 appear to end with complete termination codons (seven with TAA, four with TAG). The truncated stop codons inferred for cob (T) and nad4 (TA) are parts of TAG triplets that also contain the 5' ends of adjacent tRNA genes and are assumed to be completed by polyadenylation to TAA codons after tRNA excision (YOKOBORI and PAABO 1997 ; REICHERT et al. 1998 ). The observation that the next one or two nucleotides after a truncated stop codon may form a complete stop codon is quite frequent for metazoan mtDNAs and suggests that this may be a conserved feature to prevent readthrough of unprocessed transcripts. As presently inferred, atp6 overlaps cox3 by 8 bp and terminates with TAA. It is also possible that atp6 terminates after the T preceding the 5' end of cox3, or even earlier (see above). If this is the case, however, it would be unclear how the 3' end of atp6 transcript is formed, since there are no obvious sequence cues that could guide RNA processing at these positions (e.g., potential stem-loop structures; see BIBB et al. 1981 ; OKIMOTO et al. 1992 ).

Codon usage: In contrast to the other nematode species examined, the proteins are encoded by both strands of T. spiralis mtDNA. Nine (ATP6, ATP8, COX1, COX2, COX3, COB, NAD1, NAD3, and NAD6) are encoded by the -strand, and four (NAD2, NAD4, NAD4L, and NAD5) are encoded by the ß-strand. Since the two strands have very different nucleotide compositions, the pattern of codon usage in protein genes with coding sequences on different strands was analyzed separately.

Nonsynonymous codon usage (amino acid composition): The amino acid frequencies differ significantly (² = 398, d.f. = 19, P < 0.001) in proteins encoded by the - and ß-strands of T. spiralis mtDNA: all amino acids with A- and/or C-rich (AC-rich) codons are more frequent in -strand-encoded proteins; those with GT-rich codons are more frequent in ß-strand-encoded proteins (Table 2). When the amino acids represented by GT- or AC-rich codon families were pooled in two groups and their frequencies in proteins encoded by the - and ß-strands were compared, we found them to be significantly different (P << 0.001), with the ratios of amino acids specified by GT-rich codon families to those specified by AC-rich equal to 0.74 and 3.6 for - and ß-strand encoded proteins, respectively. Individual differences were statistically significant for seven amino acids; six of those are specified by either AC-rich or GT-rich codon families and one (isoleucine) is specified by ATY codon family (Table 2). Thus, there exists a strong correlation between the biased nucleotide composition of the - and ß-strands and the amino acid composition of the proteins encoded by them. It is likely that asymmetrical mutational pressure, rather than specific amino acid requirements of the proteins, determines the observed codon-usage differences between the strands, since both the protein and ribosomal genes on each strand demonstrate similar nucleotide biases and since different proteins encoded by the same strand have similar biases in amino acid compositions (data not shown). We have made a similar observation for the mt-proteins of L. polyphemus (LAVROV et al. 2000 ).

Synonymous codon usage: Each amino acid in nematode mtDNAs is specified by either a two- or four-codon family, or by a combination of two such families. In all cases, when an amino acid is specified by a two-codon family, the two members of such a family [ending with either a purine (A or G) or a pyrimidine (T or C)] occur with significantly different frequencies in protein genes transcribed from different strands, in accordance with the nucleotide compositional biases of the two strands (Table 3). Likewise, the usage of codons within four-codon families is also significantly different in protein genes transcribed from the different strands. However, when the frequencies of individual codons from each four-codon family were compared in these genes, we found several cases in which they were not significantly different. Those cases, underlined in Table 3, may be due either to other constraints on codon usage, such as selection or dinucleotide bias (KARLIN and BURGE 1995 ), or to an artifact of insufficient sampling. The two amino acids that are each specified by two different codon families (serine and leucine) occur with similar frequencies on the two strands. However, the representation of the two leucine families (CTN and TTR) is highly uneven in protein genes encoded by the different strands (Table 2). In contrast, the frequencies of the two serine codon families (AGN and TCN) are not statistically different between these genes. This observation also accords with the strand biases: the TTR family of leucine is T rich, whereas both serine families lack a GT/AC bias.

The strong influence of mutational pressure on both synonymous and nonsynonymous codon usage can affect phylogenetic reconstruction, as suggested by FOSTER and HICKEY 1999 . It can also, in principle, explain the observation that highly rearranged mt-genomes often produce long branches in sequence comparisons (J. BOORE, personal communication). If rearrangements result in strand exchange (inversions) or in a change in the polarity of mtDNA replication, the new mutational pattern might "overwrite" the nucleotide and amino acid compositions of the genes transferred, thus creating long branches on phylogenetic trees inferred using those sequences.

rRNA genes
The T. spiralis mt-small and -large subunit ribosomal RNA genes (rrnS and rrnL, respectively) were identified by their sequence similarities to rrnS and rrnL in other metazoan mtDNAs. Both genes are encoded by the -strand and are separated from each other by trnV (Fig 1), an arrangement typical for many metazoan mtDNAs, but unlike that in the other nematode species examined. The 5' and 3' ends of rrnS are tentatively defined to be immediately adjacent to the 3' end of trnS(ucu) and the 5' end of trnV; those of rrnL are assumed to be immediately adjacent to the 3' end of trnV and the 5' end of atp6. Secondary structure models for both srRNA and lrRNA (Fig 5 and Fig 6) were derived based on the structures of the corresponding rRNAs of Escherichia coli (NOLLER and WOESE 1981 ; NOLLER et al. 1981 ), two other nematode species (OKIMOTO et al. 1994 ), and on generalized patterns of phylogenetic conservation observed in ribosomal genes across many different taxa (GUTELL et al. 1993 ; GUTELL 1994 ).

View larger version (33K): In this window In a new window Download PPT slide   Figure 5. Secondary structure model for T. spiralis mt-small subunit rRNA. The sequence is numbered every 25 nt from the 5' end. Helices that appear to be conserved relative to the E. coli 16S rRNA model (NOLLER and WOESE 1981 ; GUTELL 1994 ) are numbered in boldface; numbering is according to VAN DE PEER et al. 1994 . An alternative secondary structure (Alt) is shown for the boxed 5' end region.

View larger version (31K): In this window In a new window Download PPT slide   Figure 6. Secondary structure model for T. spiralis mt-large subunit RNA. The sequence is numbered every 25 nt from the 5' end. Helices that appear to be conserved relative to the E. coli 23S rRNA model (NOLLER et al. 1981 ; GUTELL et al. 1993 ) are numbered in boldface; numbering is according to DE RIJK et al. 1999 . Two helices potentially present in T. spiralis (H5N and D10N) are not present in E. coli. The boxed nucleotides at the 5' and 3' ends of the rRNA can form a helix similar to one observed in Eubacteria and most Archea. Nucleotides in domains G and E that are identical to those in similar locations of the E. coli 23S rRNA model are shown in boldface.

rrnS: The size of T. spiralis mt-rrnS, as defined above, is 688 bp, similar to those of other nematodes (697 bp in C. elegans; 700 bp in A. suum; 684 bp in O. volvulus), but shorter than those of most other metazoans [e.g., 789 bp in Drosophila yakuba (CLARY and WOLSTENHOLME 1985 ); 955 bp in mouse (BIBB et al. 1981 )]. In conformity with the general model (NOLLER and WOESE 1981 ), the structure we propose for T. spiralis mt-srRNA (Fig 5) can be partitioned into four domains bounded by the three sets of long-range interactions that form helices 3, 22, and 32. The structures at the domain boundaries are well conserved in T. spiralis, as are most other elements of the core structure (RAUE et al. 1988 ; GUTELL 1994 ), with the notable exception of helices 31 and 48, which, if real, are much shorter than those in other srRNAs. [We note, however, that the reductions in the lengths of both helices are unaccompanied by a decline in the total numbers of nucleotides in the corresponding stem-loop structures, which are about the same or even greater than in the related secondary elements in the C. elegans/A. suum model (OKIMOTO et al. 1994 ).] In addition, alternative folding is possible for several structures (e.g., helices 3, 22, 23, and 39), and some of these determine the way other structures are formed. Thus, two alternatives are possible for helix 3, which, in turn, lead to alternative foldings for helices 1, 4, and 16 (Fig 5). Since alternative foldings of the 5' end domain have also been proposed for the other nematode srRNAs (e.g., compare OKIMOTO et al. 1994 and GUTELL 1994 ), it is clear that further studies of srRNAs from closely related species are needed to test these structural alternatives.

rrnL: The estimated size of T. spiralis mt-rrnL, 947 bp, is similar to those of other nematodes (953 bp in C. elegans; 960 bp in A. suum; 987 bp in O. volvulus), but shorter than those of most other metazoans (e.g., 1325 bp in D. yakuba; 1581 bp in mouse). The two 3'-most nucleotides of helix H5N (Fig 6) plus the six nucleotides following them form an octomer (5'-GUACAAAA-3') that is complementary to the sequence 27 nucleotides downstream from the inferred 5' end of rrnL(5'-UUUUGUAU-3'). Although the potential for pairing of the two ends of lrRNA is known for Eubacteria and most Archea, it has not been observed previously in either cytoplasmic or mitochondrial lrRNAs of eukaryotes (DE RIJK et al. 1999 ), with the exception of Metridium senile mt-lrRNA (BEAGLEY et al. 1998 ).

Structurally, T. spiralis mt-lrRNA is typical of mt-lrRNAs from other triploblastic metazoans: the 5' half is drastically reduced in size, with a concomitant loss of structures, whereas the 3' half is conserved and structurally similar to even E. coli's lrRNA (RAUE et al. 1988 ; GUTELL et al. 1993 ). The structural loss in the 5' half of the molecule is especially extreme in T. spiralis, as in other nematodes (OKIMOTO et al. 1994 ; KEDDIE et al. 1998 ); most of the distinctive elements in domains A–C and F are gone, and only those bounded by helices D6 and D17 are identifiable in domain D.

By contrast, structures in domains E and G are relatively well conserved in T. spiralis and other triploblasts, with the exception of helices E19 and E20, which are missing, and of helices E23, E25, and several in the terminal region bounded by helix G2, which are reduced in size relative to those in E. coli lrRNA (Fig 6). The reduction in the region bounded by helix G2, which is believed to be associated with the ribosomal E site, is more extreme in nematodes than in other metazoans (OKIMOTO et al. 1994 ) and is especially pronounced in T. spiralis, which has lost all helices between G2 and G6 (Fig 6).

tRNA genes
T. spiralis has the 22 mt-tRNA genes typical of metazoans; the genes vary in size from 53 (trnH) to 65 (trnW) bp. Twelve can be folded into structures characteristic for other nematodes (WOLSTENHOLME et al. 1987 ); in those, the TC arm and variable loop are replaced by a loop (the TV loop). Similarly, in both serine tRNAs, the DHU arms are replaced by unpaired loops (Fig 7). The remaining 8 can be folded into conventional cloverleaf structures.

View larger version (23K): In this window In a new window Download PPT slide   Figure 7. Consensus secondary structures for three groups of tRNAs in T. spiralis. Numbering of nucleotides is based on the convention used for yeast tRNA F (ROBERTUS et al. 1974 ); numbering in TV loops follows WOLSTENHOLME et al. 1994 . Open circles with numbers, nucleotides present in all tRNAs in each group; solid gray circles, nucleotides present in some, but not all, tRNAs; solid black circles with letters, nucleotides conserved in anticodon loops, TC, DHU, and variable arms, or their replacement loops, of all or all but one of the tRNAs in each of these groups. K = G or T; R= A or G; Y = C or T; W = A or T. The pattern of nucleotide conservation is not shown for tRNAs with a DHU-replacement loop, due to the limited sample size. Broken lines indicate possible tertiary interactions.

Each tRNA has been inferred to have an aminoacyl acceptor stem of 7 bp, an anticodon stem of 5 bp, and an anticodon loop of 7 nt. Fifteen mismatches were found among the aminoacyl acceptor stems, and three were found at the base of anticodon stems. The most common mismatch position was between nucleotides 7 and 66, at the base of the aminoacyl acceptor stem. This position was mismatched in 8 of the 12 tRNAs with TV loops (R, A, N, E, Q, G, F, P, and V), but in no others. Interestingly, mismatches at this position are also common in the tRNAs of the other nematodes (WOLSTENHOLME et al. 1987 , WOLSTENHOLME et al. 1994 ; KEDDIE et al. 1998 ). Additional mismatches in the aminoacyl acceptor stems of T. spiralis mt-tRNAs are between nucleotides 1 and 72 in tRNAs R and V and between nucleotides 3 and 70 in tRNA Y. Mismatches at the base of the anticodon stem occur in tRNAs E, L(tag), and V.

In tRNAs with a DHU arm, the stem is usually 4 bp long [3 bp in tRNAs C, L(uag), L(uaa), and Y] and the loop is between 3 nt [tRNA(H)] and 12 nt [tRNA(E)]. The DHU-replacement loops are 5 and 4 nt in tRNA (S)(ucu) and tRNA(S)(uga), respectively. The TC arm, when present, has a stem of 2, 3, or 5 bp and a loop of 3 to 8 nt. The variable loop in these cases is either 4 or 5 nt. When the variable loop and TC arm are absent, they are replaced by a TV loop of 6 to 8 nt.

The anticodons in T. spiralis mt-tRNAs are generally the same as those in other nematode mt-tRNAs. However, that for T. spiralis tRNA(R) is 5'-TCG-3', as in most other metazoans, instead of the very unusual 5'-ACG-3' present in the secernentean nematodes (WATANABE et al. 1997 ).

Conserved nucleotides and possible tertiary interactions: The tRNAs encoded by prokaryotic, nonanimal organellar and nuclear genomes (referred to as standard tRNAs) have several invariable and semi-invariable nucleotide positions (DIRHEIMER et al. 1995 ). Nucleotides at some of these positions are also conserved in previously studied nematode mt-tRNAs (WOLSTENHOLME et al. 1994 ) and T. spiralis mt-tRNAs (Fig 7), but not in mt-tRNAs from most other metazoans (WOLSTENHOLME 1992 ). Since nucleotides at several conserved positions were shown to be involved in the tertiary interactions in standard tRNAs and nematode mt-tRNAs (KIM et al. 1974 ; ROBERTUS et al. 1974 ; WATANABE et al. 1994 ; OHTSUKI et al. 1998 ), we evaluated the potential for similar tertiary interactions in T. spiralis mt-tRNAs. We found such for previously described hydrogen bondings between nucleotides 8·4·21, L3(46)·22-13, L2(45)·10-25, 15·L4(48), 9·23-12, and 26·44(L1) (Fig 7). We also found three deviations from the previously described patterns of nucleotide conservation.

First, while a strong correlation exists in the occurrence of nucleotides at positions 13-22 of the DHU-stem and L3(46) of the TV-replacement (variable size) loop in the inferred T. spiralis mt-tRNAs, its pattern differs from the usual R46(L2)·R22-Y13. We found that in all but one case the same nucleotide and not necessarily a purine, is present at position 22 of the Watson-Crick 13-22 pair and at L3(46) [G46(L2)·G22-U13 in six tRNAs, U46(L2)·U22-A13 in three tRNAs, and A46(L2)·A22-T13 in six tRNAs]. The only exception is tRNA(K), which has an A22-T13 pair in the DHU stem but G at position 46. There are mismatches between positions 13 and 22 in the DHU stems of four additional tRNAs. In all these cases there are different nucleotides at position 22 and L3(46).

Second, the nature of bond III (usually RL2(45)·R10-Y25 in standard tRNAs and in the mt-tRNAs of other nematodes) appears to vary among T. spiralis tRNAs with different secondary structures. Although the R10-Y25 pair is present in all tRNAs except tRNA(I), six of eight tRNAs with cloverleaf structures have a pyrimidine at position 45, whereas all those with TV loops have a purine at the corresponding position (L2). A purine is also present at L2 in all other nematode mt-tRNAs with TV loops.

Third, there are differences between T. spiralis and other nematode mt-tRNAs in the presence of specific nucleotides at positions 9, 12, and 23, which are involved in the formation of the hydrogen bond V. In secernentean nematodes nucleotide 9 is always A and the 12-23 pair is always W12-W23 (WOLSTENHOLME et al. 1994 ; KEDDIE et al. 1998 ). However, nucleotide 9 is G in four T. spiralis mt-tRNAs [E, G, L(uag), L(uaa)], and the 12-23 pair is S12-S23 in six [A, R, E, G, H, M]. The combinations of nucleotides at these positions other than A9·W23-W12 are also very common in standard tRNAs.

tRNA-like structure: In addition to the set of 22 tRNA genes commonly present in metazoan mtDNAs, a sequence between trnG and trnD, designated trnM₂ in Fig 1, has the potential to form a tRNA-like structure with an anticodon (5'-UAU-3') that would recognize methionine codons. Two genes for tRNA(M), one with anticodon 5'-CAU-3' and the second with 5'-UAU-3', were reported in Mytilus edulis mtDNA (HOFFMANN et al. 1992 ). Both trnM(uau) and its transcription product have also been found in the related species M. californianus (BEAGLEY et al. 1999 ). The location of T. spiralis trnM₂ within a set of six tRNA genes suggests that it is transcribed and most likely processed. However, it lacks some well-conserved nucleotides (T in position 33, R in position 37), has an unusual secondary structure with a very large (21 nt) TC loop, and overlaps the downstream trnG by 2 bp, all of which suggest that it may not be functional. Interestingly, a sequence identical to part of trnM₂ is found in the noncoding region bounded by trnT and trnP.

Noncoding regions
The region between nad1 and nad2 contains at least two copies of a large (1232 bp) repeat, which, though mostly noncoding, also includes part of trnK. The repeat units proximal to each end of the region were completely sequenced and found to differ at three positions. Two potential stem-loop structures were found in each repeat unit. Both have 14-bp stems; the one proximal to nad1 has a 7-nt loop, while that proximal to nad2 has a 15-nt loop; the latter also has a poly(T) tract, a feature common to the class of stem-loop structures implicated as possible origins of mtDNA replication in metazoans (WOLSTENHOLME 1992 ). The structures do not appear to be artifactual: the probability of their occurring by chance in a sequence of equal length and nucleotide composition to that of the repeat unit is <0.01, as estimated by computer simulation (LAVROV et al. 2000 ). Only a small amount of sequence between the two flanking repeat units was determined. However, as stated above, it is likely that additional repeat units are present between the two sequenced. It is also likely that the size variation in this region is caused by the differences in the number of repeat units among different mtDNA molecules in the same and/or different individuals, as previously observed (e.g., DENSMORE et al. 1985 ; MORITZ and BROWN 1987 ; LA ROCHE et al. 1990 ). Another relatively large noncoding region (168 bp), between trnT and trnP, contains a 56-bp sequence identical to part of trnM₂, a part of which has the potential to form a structure with an 11-bp stem and a 3-nt loop. Aside from the two noncoding regions just described, 117 additional noncoding base pairs are present in 15 small intergenic regions. These range in size from 1 to 40 bp, have no shared sequence motifs or potential to form structures, and can be characterized as intergenic "spacers" (Fig 1).

	CONCLUSIONS

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS LITERATURE CITED

The mtDNA of T. spiralis establishes a link between typical metazoan mtDNAs and the nematode mtDNAs described previously. In several respects, T. spiralis mtDNA is more similar to those of non-nematode metazoans: it has the 37 genes typical of most metazoan mtDNAs; its gene arrangement has clear affinities with those of coelomate metazoans; its protein genes initiate with standard ATN codons; and tRNA(R) encoded has a typical metazoan 5'-UCG-3' anticodon. Thus, the unusual gene arrangements, initiation codons, 5'-ACG-3' anticodon in tRNA(R), and the lack of atp8 observed in the mtDNAs of secernentean nematodes appear to be derived features that arose within that lineage after the divergence of secernentean nematodes from other metazoan groups. In other respects, T. spiralis mtDNA is more similar to those of other nematodes or intermediate between them and those of non-nematode metazoans: it encodes rRNAs that are similar to their counterparts in other nematodes both in size and structure; most of its protein genes are intermediate in size between those of other nematodes and coelomate metazoans; and some of its tRNAs have conventional cloverleaf structures, whereas others have the "bizarre" structures that are characteristic of secernentean nematode mt-tRNAs.

	ACKNOWLEDGMENTS

We thank D. Despommier and R. Polvere for T. spiralis DNA, J. Boore for help with data analysis, and K. Helfenbein and three anonymous reviewers for helpful comments and suggestions on an earlier version of this manuscript. This work was supported by National Science Foundation (NSF) dissertation improvement grant DEB 9972712 (to W.M.B. and D.V.L.) and NSF grant DEB 9807100 (to W.M.B).

Manuscript received March 28, 2000; Accepted for publication October 11, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS LITERATURE CITED

ADAMSON, M. L., 1987  Phylogenetic analysis of the higher classification of the Nematoda. Can. J. Zool. 65:1478-1482.

AGUINALDO, A. M. A., J. M. TURBEVILLE, L. S. LINFORD, M. C. RIVERA, and J. R. GAREY et al., 1997  Evidence for a clade of nematodes, arthropods and other moulting animals. Nature 387:489-493[Medline].

ANDERSON, S., A. T. BANKIER, B. G. BARRELL, M. H. L. DE BRUIJN, and A. R. COULSON et al., 1981  Sequence and organization of the human mitochondrial genome. Nature 290:457-465[Medline].

ARMSTRONG, M. R., V. C. BLOCK, and M. S. PHILLIPS, 2000  A multipartite mitochondrial genome in the potato cyst nematode Globodera pallida.. Genetics 154:181-192[Abstract/Free Full Text].

ASAKAWA, S., Y. KUMAZAWA, T. ARAKI, H. HIMENO, and K. MIURA et al., 1991  Strand-specific nucleotide composition bias in echinoderm and vertebrate mitochondrial genomes. J. Mol. Evol. 32:511-520[Medline].

AVEROF, M., A. ROKAS, K. H. WOLFE, and P. M. SHARP, 2000  Evidence for a high frequency of simultaneous double-nucleotide substitutions. Science 287:1283-1286[Abstract/Free Full Text].

AZEVEDO, J. L. and B. C. HYMAN, 1993  Molecular characterization of lengthy mitochondrial DNA duplications from the parasitic nematode Romanomermis culicivorax.. Genetics 133:933-942[Abstract].

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].

BEAGLEY, C. T., R. OKIMOTO, and D. R. WOLSTENHOLME, 1999  Mytilus mitochondrial DNA contains a functional gene for a tRNASer (UCN) with a dihydrouridine arm-replacement loop and a pseudo-tRNASer(UCN) gene. Genetics 152:641-652[Abstract/Free Full Text].

BIBB, J. M., 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].

BLAXTER, M. L., P. DE LEY, J. R. GAREY, L. X. LIU, and P. SCHELDEMAN et al., 1998  A molecular evolutionary framework for the phylum Nematoda. Nature 392:71-75[Medline].

BOORE, J. L., 1999  Animal mitochondrial genomes. Nucleic Acids Res. 27:1767-1780[Abstract/Free Full Text].

BROWN, F. F. 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].

BRUSCA, R. C., and G. J. BRUSCA, 1990 Invertebrates. Sinauer Associates, Sunderland, MA.

CANTATORE, P., M. N. GADALETA, M. ROBERTI, C. SACCONE, and A. C. WILSON, 1987  Duplication and remoulding of tRNA genes during the evolutionary rearrangement of mitochondrial genomes. Nature 329:853-855[Medline].

CLARY, D. O. and D. R. WOLSTENHOLME, 1985  The ribosomal RNA genes of Drosophila mitochondrial DNA. Nucleic Acids Res. 13:4029-4045[Abstract/Free Full Text].

DENSMORE, L. D., J. W. WRIGHT, and W. M. BROWN, 1985  Length variation and heteroplasmy are frequent in mitochondrial DNA from parthenogenetic and bisexual lizards (genus Cnemidophorus). Genetics 110:687-707.

DE RIJK, P. and R. DE WACHTER, 1997  RnaViz, a program for the visualization of RNA secondary structure. Nucleic Acids Res. 25:4679-4684[Abstract/Free Full Text].

DE RIJK, P., E. ROBBRECHT, S. DE HOOG, A. CAERS, and Y. VAN DE PEER et al., 1999  Database on the structure of large subunit ribosomal RNA. Nucleic Acids Res. 27:174-178[Abstract/Free Full Text].

DIRHEIMER, G., G. KEITH, P. DUMAS and E. WESTHOF, 1995 Primary, secondary and tertiary structures of tRNAs, pp. 93–126 in tRNA: Structure, Biosynthesis, and Function, edited by D. SÖLL and U. RAJBHANDARY. American Society for Microbiology, Washington, DC.

FOSTER, P. G. and D. A. HICKEY, 1999  Compositional bias may affect both DNA-based and protein-based phylogenetic reconstructions. J. Mol. Evol. 48:284-290[Medline].

GRAY, M. W., B. F. LANG, R. CEDERGREN, G. B. GOLDING, and C. LEMIEUX et al., 1998  Genome structure and gene content in protist mitochondrial DNAs. Nucleic Acids Res. 26:865-878[Abstract/Free Full Text].

GUTELL, R. R., 1994  Collection of small subunit (16S- and 16S-like) ribosomal RNA structures: 1994. Nucleic Acids Res. 22:3502-3507[Abstract/Free Full Text].

GUTELL, R. R., M. W. GRAY, and M. N. SCHNARE, 1993  A compilation of large subunit (23S- and 23S-like) ribosomal RNA structures. Nucleic Acids Res. 21:3055-3074[Free Full Text].

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

KARLIN, S. and C. BURGE, 1995  Dinucleotide relative abundance extremes: a genomic signature. Trends Genet. 11:283-290[Medline].

KEDDIE, E. M., T. HIGAZI, and T. R. UNNASCH, 1998  The mitochondrial genome of Onchocerca volvulus: sequence, structure and phylogenetic analysis. Mol. Biochem. Parasitol. 95:111-127[Medline].

KIM, S. H., F. L. SUDDATH, G. J. QUIGLEY, A. MCPHERSON, and J. L. SUSSMAN et al., 1974  Three-dimensional tertiary structure of yeast phenylalanine transfer RNA. Science 185:435-440[Abstract/Free Full Text].

KYTE, J. and R. F. DOOLITTLE, 1982  A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105-132[Medline].

LANG, B. F., M. W. GRAY, and G. BURGER, 1999  Mitochondrial genome evolution and the origin of eukaryotes. Annu. Rev. Genet. 33:351-397[Medline].

LA ROCHE, J., M. SNYDER, D. I. COOK, K. FULLER, and E. ZOUROS, 1990  Molecular characterization of a repeat element causing large-scale size variation in the mitochondrial DNA of the sea scallop Placopecten magellanicus.. Mol. Biol. Evol. 7:45-64[Abstract].

LAVROV, D. V., J. L. BOORE, and W. M. BROWN, 2000  The complete mitochondrial DNA sequence of the horseshoe crab Limulus polyphemus.. Mol. Biol. Evol. 17:813-824[Abstract/Free Full Text].

MALAKHOV, V. V., 1994 Nematodes: Structure, Development, Classification, and Phylogeny. Smithsonian Institution Press, Washington, DC.

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].

NAGANO, I., Z. WU, A. MATSUO, E. POZIO, and Y. TAKAHASHI, 1999  Identification of Trichinella isolates by polymerase chain reaction—restriction fragment length polymorphism of the mitochondrial cytochrome c-oxidase subunit I gene. Int. J. Parasitol. 29:1113-1120[Medline].

NOLLER, H. F. and C. R. WOESE, 1981  Secondary structure of 16S ribosomal RNA. Science 212:403-411[Abstract/Free Full Text].

NOLLER, H. F., J. KOP, V. WHEATON, J. BROSIUS, and R. R. GUTELL et al., 1981  Secondary structure model for 23S ribosomal RNA. Nucleic Acids Res. 9:6167-6189[Abstract/Free Full Text].

OHTSUKI, T., G. KAWAI, and K. WATANABE, 1998  Stable isotope-edited NMR analysis of Ascaris suum mitochondrial tRNAMet having a TV-replacement loop. J. Biochem. 124:28-34[Abstract/Free Full Text].

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

OKIMOTO, R., J. L. MACFARLANE, and D. R. WOLSTENHOLME, 1990  Evidence for the frequent use of TTG as the translation initiation codon of mitochondrial protein genes in the nematodes, Ascaris suum and Caenorhabditis elegans.. Nucleic Acids Res. 18:6113-6118[Abstract/Free Full Text].

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

OKIMOTO, R., J. L. MACFARLANE, and D. R. WOLSTENHOLME, 1994  The mitochondrial ribosomal RNA genes of the nematodes Caenorhabditis elegans and Ascaris suum: consensus secondary-structure models and conserved nucleotide sets for phylogenetic analysis. J. Mol. Evol. 39:598-613[Medline].

PERNA, N. T. and T. D. KOCHER, 1995  Patterns of nucleotide composition at fourfold degenerate sites of animal mitochondrial genomes. J. Mol. Evol. 41:353-358[Medline].

RAUE, H. A., J. KLOOTWIJK, and W. MUSTERS, 1988  Evolutionary conservation of structure and function of high molecular weight ribosomal RNA. Prog. Biophys. Mol. Biol. 51:77-129[Medline].

REICHERT, A., U. ROTHBAUER, and M. MÖRL, 1998  Processing and editing of overlapping tRNAs in human mitochondria. J. Biol. Chem. 273:31977-31984[Abstract/Free Full Text].

REYES, A., C. GISSI, G. PESOLE, and C. SACCONE, 1998  Asymmetrical directional mutation pressure in the mitochondrial genome of mammals. Mol. Biol. Evol. 15:957-966[Abstract].

ROBERTUS, J. D., J. E. LADNER, J. T. FINCH, D. RHODES, and R. S. BROWN et al., 1974  Structure of yeast phenylalanine tRNA at 3 A resolution. Nature 250:546-551[Medline].

STATON, J. L., L. L. DAEHLER, and W. M. BROWN, 1997  Mitochondrial gene arrangement of the horseshoe crab Limulus polyphemus L.: conservation of major features among arthropod classes. Mol. Biol. Evol. 14:867-874[Abstract].

THOMPSON, J. D., D. G. HIGGINS, and T. J. GIBSON, 1994  CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680[Abstract/Free Full Text].

VAN DE PEER, Y., I. VAN DEN BROECK, P. DE RIJK, and R. DE WACHTER, 1994  Database on the structure of small ribosomal subunit RNA. Nucleic Acids Res 22:3488-3494[Abstract/Free Full Text].

VORONOV, D. A., Y. V. PANCHIN, and S. E. SPIRIDONOV, 1998  Nematode phylogeny and embryology. Nature 395:28[Medline].

WATANABE, Y., H. TSURUI, T. UEDA, R. FURUSHIMA, and S. TAKAMIYA et al., 1994  Primary and higher order structures of nematode (Ascaris suum) mitochondrial tRNAs lacking either the T or D stem. J. Biol. Chem. 269:22902-22906[Abstract/Free Full Text].

WATANABE, Y., H. TSURUI, T. UEDA, R. FURUSIHIMA-SHIMOGAWARA, and S. TAKAMIYA et al., 1997  Primary sequence of mitochondrial tRNA(Arg) of a nematode Ascaris suum: occurrence of unmodified adenosine at the first position of the anticodon. Biochim. Biophys. Acta 1350:119-122[Medline].

WESLEY, U. V. and C. S. WESLEY, 1997  Rapid directional walk within DNA clones by step-out PCR. Methods Mol. Biol. 67:279-285[Medline].

WOLSTENHOLME, D. R., 1992  Animal mitochondrial DNA: structure and evolution. Int. Rev. Cytol. 141:173-216[Medline].

WOLSTENHOLME, D. R., J. L. MACFARLANE, R. OKIMOTO, D. O. CLARY, and J. A. WAHLEITHNER, 1987  Bizarre tRNAs inferred from DNA sequences of mitochondrial genomes of nematode worms. Proc. Natl. Acad. Sci. USA 84:1324-1328[Abstract/Free Full Text].

WOLSTENHOLME, D. R., R. OKIMOTO, and J. L. MACFARLANE, 1994  Nucleotide correlations that suggest tertiary interactions in the TV-replacement loop-containing mitochondrial tRNAs of the nematodes, Caenorhabditis elegans and Ascaris suum.. Nucleic Acids Res. 22:4300-4306[Abstract/Free Full Text].

YATES, F., 1934  Contingency tables involving small numbers and the ² test. J. R. Stat. Soc. 1(Suppl.):217-235.

YOKOBORI, S. and S. PÄÄBO, 1997  Polyadenylation creates the discriminator nucleotide of chicken mitochondrial tRNA(Tyr). J. Mol. Biol. 265:95-99[Medline].

This article has been cited by other articles:

				S. E. Masta and J. L. Boore Parallel Evolution of Truncated Transfer RNA Genes in Arachnid Mitochondrial Genomes Mol. Biol. Evol., May 1, 2008; 25(5): 949 - 959. [Abstract] [Full Text] [PDF]

				L. Podsiadlowski, A. Braband, and G. Mayer The Complete Mitochondrial Genome of the Onychophoran Epiperipatus biolleyi Reveals a Unique Transfer RNA Set and Provides Further Support for the Ecdysozoa Hypothesis Mol. Biol. Evol., January 1, 2008; 25(1): 42 - 51. [Abstract] [Full Text] [PDF]

				D. V. Lavrov Key transitions in animal evolution: a mitochondrial DNA perspective Integr. Comp. Biol., November 1, 2007; 47(5): 734 - 743. [Abstract] [Full Text] [PDF]

				S. Tang and B. C. Hyman Mitochondrial Genome Haplotype Hypervariation Within the Isopod Parasitic Nematode Thaumamermis cosgrovei Genetics, June 1, 2007; 176(2): 1139 - 1150. [Abstract] [Full Text] [PDF]

				M. Arita, T. Suematsu, A. Osanai, T. Inaba, H. Kamiya, K. Kita, M. Sisido, Y.-i. Watanabe, and T. Ohtsuki An evolutionary 'intermediate state' of mitochondrial translation systems found in Trichinella species of parasitic nematodes: co-evolution of tRNA and EF-Tu Nucleic Acids Res., October 6, 2006; 34(18): 5291 - 5299. [Abstract] [Full Text] [PDF]

				D. S. Zarlenga, B. M. Rosenthal, G. La Rosa, E. Pozio, and E. P. Hoberg Post-Miocene expansion, colonization, and host switching drove speciation among extant nematodes of the archaic genus Trichinella PNAS, May 9, 2006; 103(19): 7354 - 7359. [Abstract] [Full Text] [PDF]

				D. Papillon, Y. Perez, X. Caubit, and Y. Le Parco Identification of Chaetognaths as Protostomes Is Supported by the Analysis of Their Mitochondrial Genome Mol. Biol. Evol., November 1, 2004; 21(11): 2122 - 2129. [Abstract] [Full Text] [PDF]

				T. A. Rawlings, T. M. Collins, and R. Bieler Changing identities: tRNA duplication and remolding within animal mitochondrial genomes PNAS, December 23, 2003; 100(26): 15700 - 15705. [Abstract] [Full Text] [PDF]