Molecular Evolution and Recombination in Gender-Associated Mitochondrial DNAs of the Manila Clam Tapes philippinarum

Corresponding author: Marco Passamonti, Università degli Studi di Bologna, via Selmi 3, 40126 Bologna, Italy., mpassa{at}alma.unibo.it (E-mail)

Communicating editor: M. A. ASMUSSEN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Doubly uniparental inheritance (DUI) provides an intriguing system for addressing aspects of molecular evolution and intermolecular recombination of mitochondrial DNA. For this reason, a large sequence analysis has been performed on Tapes philippinarum (Bivalvia, Veneridae), which has mitochondrial DNA heteroplasmy that is consistent with a DUI. The sequences of a 9.2-kb region (containing 29 genes) from 9 individuals and the sequences of a single gene from another 44 individuals are analyzed. Comparisons suggest that the two sex-related mitochondrial genomes do not experience a neutral pattern of divergence and that selection may act with varying strength on different genes. This pattern of evolution may be related to the long, separate history of M and F genomes within their tissue-specific "arenas." Moreover, our data suggest that recombinants, although occurring in soma, may seldom be transmitted to progeny in T. philippinarum.

MITOCHONDRIAL DNA (mtDNA) has been studied in several animal groups, mainly for phylogenetic purposes. It is physically separate from the nuclear genome and, with few exceptions, it is circular and contains 37 genes (22 for tRNAs, 2 for rRNAs, 13 for proteins). mtDNA appears to have a strictly maternal inheritance in most organisms and is thought to lack recombination (WOLSTENHOLME 1992 ), although this is not without controversy (EYRE-WALKER and AWADALLA 2001 ).

The most serious exception to the matrilineal inheritance of animal mtDNA comes from a few bivalve mollusks, in which two different sex-related mtDNAs have been detected (the so-called M and F mtDNAs), showing unexpected levels of divergence (10–30%). Research on the Mytilus edulis-M. galloprovincialis species complex demonstrated that the F mtDNA is transmitted from the mother to both daughters and sons, whereas the M mtDNA is transmitted from father to sons only (SKIBINSKI et al. 1994A , SKIBINSKI et al. 1994B ; ZOUROS et al. 1994A , ZOUROS et al. 1994B ). In adult males the F mitochondrial genome prevails in somatic tissues, while the M mitochondrial genome is strongly predominant in gonads, with sperm carrying only M-type mtDNAs. Thus, a double mechanism of transmission, in which both M and F mtDNAs are inherited uniparentally, is realized. This peculiar pattern of mitochondrial inheritance was called "doubly uniparental inheritance" (DUI; ZOUROS et al. 1994A , ZOUROS et al. 1994B ) or "gender-associated inheritance" (SKIBINSKI et al. 1994A , SKIBINSKI et al. 1994B ). This pattern of transmission, detectable through the presence of two mtDNAs showing a tissue-specific distribution in males (SKIBINSKI et al. 1994A ; DALZIEL and STEWART 2002 ), was found in some species of the genus Mytilus (M. edulis, M. galloprovincialis, M. trossulus, and M. californianus), in Geukensia demissa (Mytilidae), and in some unionids (fresh water mussels: Pyganodon fragilis, P. grandis, Fusconaia flava, and Anodonta grandis; see HOEH et al. 1997 and references therein). Furthermore, the recent finding of mtDNA heteroplasmy in the venerid Manila clam Tapes philippinarum (PASSAMONTI and SCALI 2001 ) showed that the pattern also occurs in phylogenetically distant families and suggests that it might be widely distributed among bivalves.

Several hypotheses have been suggested for the evolutionary forces involved in sequence divergence of M and F genomes. Analyses of the Mytilus species complex showed that mussel mtDNA appears to evolve faster than that in other metazoans (HOFFMANN et al. 1992 ); this has been related to relaxed selective constraints acting in DUI systems of inheritance (HOEH et al. 1996 ). Moreover, several studies on Mytilus and Pyganodon demonstrated that the M lineage evolves faster than the F lineage (LIU et al. 1996 ; QUESADA et al. 1998 and references therein). Researchers suggested several explanations for this: a higher rate of M mtDNA duplication during spermatogenesis and early male embryo development, free-radical damage to sperm, positive selection, or effects of the smaller population size of the M genome (see STEWART et al. 1996 for details). Analyses were also performed to evaluate the best model of molecular evolution to be applied to the Mytilus system. The neutral model has been rejected (STEWART et al. 1996 ), but the nearly neutral theory (OHTA 1992 ) has been considered to best fit the observed data (QUESADA et al. 1998 ; SKIBINSKI et al. 1999 ). However, SKIBINSKI et al. 1999 stated that more information on effective population size for Mytilus M and F genomes is considered very important to discriminate a nearly neutral situation (OHTA 1996 ) from a model of positive selection.

Several attempts to assign a functional role to DUI have been made, but DUI systems are so scarcely known in their molecular aspects that the given explanations must remain largely speculative. In fact, even in non-DUI systems, the mechanisms of transmission of mitochondria to progeny are scarcely known in detail, but in recent years several studies, mainly on humans, have contributed somewhat to the issue. Despite a high mtDNA copy number (100,000) in oocytes, new mtDNA sequence variants segregate rapidly between generations. This paradox can be explained by the occurrence of narrow bottlenecks during oogenesis or early embryo development. Researchers speculate that only a few mitochondria are segregated to be passed on to the germline, and this mechanism can be related to the necessity of maintaining the genomic integrity of the organellar genome (JANSEN et al. 2000 ; SHOUBRIDGE 2000 ). Actually, highly rearranged and/or deleted mtDNAs are commonly observed in several somatic tissues (the so-called "sublimons"; see KAJANDER et al. 2000 ), which might be related to the high oxidative stress to which mitochondria are exposed.

Some evidence of mitochondrial recombination has been proposed by LAUDOKAKIS and ZOUROS 2001A , LAUDOKAKIS and ZOUROS 2001B in metazoans. As they point out, DUI systems represent a model for studying basic aspects of the evolution and transmission of the mitochondrial genome, mainly due to the presence of two different mtDNAs in the somatic tissues of males. The authors present evidence of recombination between male and female mitotypes in the DUI system of M. edulis. They also speculate that mtDNA recombination may occur at a high rate. On the basis of that, the extent to which the recombination generates new haplotypes may depend only on the frequency of biparental inheritance of mitochondrial genomes.

To test the generality of these hypotheses, we conducted a sequence analysis in the Manila clam T. philippinarum. We analyzed sequences obtained from somatic tissues of males as well as from mature gonad tissues of both males and females.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Sampling and DNA extraction:
Clams were collected from either the Adriatic Sea (Italy) or the Pacific Ocean (Washington state). Both stocks have been introduced for commercial purposes, so these populations do not correspond to the original species ranges (FISCHER-PIETTE and METIVIER 1971 ).

Fully mature, living clams were dissected and gonad liquid was collected with a glass capillary. Separately, a sample of somatic tissue (the terminal tip of the foot or adductor muscle) was also taken from each specimen. Gonad samples were inspected under a light microscope to detect mature eggs or sperm. Tissues, both fresh and frozen at -80°, were utilized for DNA isolation and purification. We analyzed 44 specimens (9 females and 35 males) from the Italian sample and 9 specimens (5 females and 4 males) from the U.S. sample.

Total genomic DNA was separately obtained from gonads and somatic tissues either according to the method described in PREISS et al. 1988 or by utilizing the DNAeasy tissue kit (QIAGEN, Chatsworth, CA). Enriched mtDNA fractions were obtained from tissues using the mtDNA Extractor CT kit (Wako Chemicals). All those extraction methods were suitable for obtaining good DNA for PCR amplification from clam tissues, but the best results were obtained with the Wako kit, especially for amplifying long fragments.

rRNA partial gene amplification and sequencing:
Partial sequence of the mitochondrial large ribosomal subunit RNA gene (rrnL) was amplified from various tissues of the 44 specimens from Italy as described in PASSAMONTI and SCALI 2001 . Fragments obtained from gonads were directly sequenced without cloning. In males we were able to obtain good sequences using only gonad liquid from fully mature specimens. At variance with what we did for gonads, we amplified and cloned the rrnL partial sequence obtained from somatic tissue of three specimens. Sequencing reactions were performed on both strands with the BigDye terminator cycle sequencing kit (Applied Biosystems, Foster City, CA) in a 310 Genetic Analyzer (ABI) automatic sequencer.

Long-PCR and sequencing:
Long-PCR amplifications were obtained from the nine U.S. samples with Herculase enhanced DNA polymerase (Stratagene, La Jolla, CA) according to the kit protocol. To increase amplification fidelity, we used no DMSO in the amplification mixture. The mtDNA region between the mitochondrial large ribosomal subunit RNA gene (rrnL) and the cytochrome oxidase subunit I gene (cox1; 9.2 kb) was amplified using 16sbl (PALUMBI et al. 1991 ) and HCO2198 (FOLMER et al. 1994 ) universal mtDNA primers.

The long-PCR products were randomly sheared into fragments of 1 kb by repeatedly passing the DNA through a narrow aperture under high pressure with a Hydroshear device (GeneMachines), blunt end repaired, and subcloned into a pUC18 vector to make a random insert library. For each sample, 384 recombinant clones were collected and analyzed. Recombinant plasmids were directly amplified from the crude bacterial colony using rolling circle amplification (DEAN et al. 2001 ) with the TempliPhi kit (Amersham Pharmacia). Sequencing reactions were performed on both strands using Amersham's ET dye terminator kit, reactions were purified using solid phase reversible immobilization (Perkin-Elmer), and electrophoretic separations and detection were performed with a MegaBACE 1000 automated DNA sequencer. The resulting sequences were processed using Phred (CodonCode) and assembled using Sequencher (GeneCodes). All assemblies and the overall sequence quality were verified by eye. Detailed protocols are available at http://www.jgi.doe.gov/Internal/prots_index.html.

Sequence analysis:
Sequences were aligned using the Clustal algorithm of the MT Navigator PPC software (Applied Biosystems). Alignments were then edited manually, taking into account gene ends and structure. Proteins and ribosomal RNAs were analyzed by sequence comparison to Lumbricus terrestris (BOORE and BROWN 1995 ), Katharina tunicata (BOORE and BROWN 1994 ), and M. edulis (HOFFMANN et al. 1992 ). tRNA genes were identified by using tRNAscan-SE (version 1.1, http://www.genetics.wustl.edu/eddy/tRNAscan-SE; LOWE and EDDY 1997 ) or, whenever the program failed to predict tRNAs, by analyzing the sequence by eye, recognizing potential secondary structures. The 5' end of protein-coding genes was inferred to be at the first in-frame legitimate start codon (ATN, GTG, TTG, GTT; see WOLSTENHOLME 1992 ) and the 3' ends at the first observed in-frame stop codon (i.e., no "abbreviated" stop codons were inferred). The 5' and 3' ends of rRNA genes were assumed to be adjacent to the flanking genes.

The levels of divergence for each analyzed gene and for the whole 9.2-kb sequence were calculated using nucleotide pairwise P-distances, and then the mean value and standard errors (by the bootstrap procedure) of P-distances within each group (i.e., M vs. M sequences in males, F vs. F in females) were determined. The two groups were compared with a basic one-way analysis of variance (ANOVA) to test whether variability within M-type and F-type sequences is significantly different.

To analyze sequences for neutrality, we performed tests based on protein polymorphism only. We did not use neutrality tests based on DNA polymorphism (i.e., Tajima's test, etc.) because, as stated by several authors (MCDONALD and KREITMAN 1991 ; BROOKFIELD and SHARP 1994 ; QUESADA et al. 1998 ), these are strongly influenced by changes in population size. Actually, this is just what happened recently for these samples, since the analyzed populations have been introduced in the Italian and the U.S. seas from a small stock of clams obtained from hatcheries.

The number of polymorphic sites within M and F protein-coding sequences (showing either synonymous or nonsynonymous mutations) was measured, and a test of positive selection was performed by a one-tailed Fisher's exact test (ZHANG et al. 1997 ), as implemented by MEGA2 (KUMAR et al. 2001 ). If the resulting P-value is <0.05, then the null hypothesis of neutral evolution (i.e., strictly neutral and purifying selection) is rejected. Moreover, when the number of synonymous differences per synonymous site exceeds the number of nonsynonymous differences per nonsynonymous site, the test sets P = 1 to indicate the existence of purifying selection. This test is considered more appropriate than the Z-test when the number of nucleotide substitutions per sequence is small (NEI and KUMAR 2001 ), which is the case for our comparisons.

Comparisons between M- and F-types were performed by the test of MCDONALD and KREITMAN 1991 , as implemented in DnaSP 3.53 (ROZAS and ROZAS 1999 ). The test is based on the observation that, under neutrality, the ratio of nonsynonymous to synonymous fixed substitutions between M- and F-types should be the same as the ratio of nonsynonymous to synonymous polymorphism within types.

Male and female sequences were also analyzed for gene conversion and recombination using DnaSP (v. 3.53; ROZAS and ROZAS 1999 ), which incorporates the algorithm developed by BETRAN et al. 1997 for detecting gene conversion events.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Sequence features:
The sex of each sample was determined using light microscopy on gonads. Both populations showed no apparent sex ratio deviation. We sequenced 44 Italian samples (9 females and 35 males, all analyzed for the rrnL partial sequence only) and 9 U.S. samples (5 females and 4 males, analyzed for the 9.2-kb mtDNA fragment). As expected, sequences belong to two distinct groups: F-like sequences were invariably obtained from ovaries, whereas M-like sequences were obtained from testes. No clam carried haplotypes of the opposite sex, supporting the rarity of "masculinization" events in this species, when compared to Mytilus (see HOEH et al. 1997 and PASSAMONTI and SCALI 2001 for details).

In the 9.2-kb mtDNA region, we identified 29 genes (7 protein-encoding genes, 2 rRNA genes, and 20 tRNA genes). (Animal mitochondrial genomes typically contain 37 genes in total; BOORE 1999 .) The region contains no large noncoding regions. All sequenced genes are transcribed from the same strand, as is the case for M. edulis (HOFFMANN et al. 1992 ).

Comparing male and female gene arrangements, we observed the same gene order with one notable exception: in females the tRNA immediately upstream of nad6 gene is for valine (tRNA-Val, anticodon TAC, 4927–4989 region), whereas in males, it is for lysine (tRNA-Lys, anticodon TTT, 4862–4921 region). However, both sexes also have copies of both of these tRNA genes downstream of nad6, so that the difference seems unlikely to be significant for mitochondrial function. The gene for tRNA-Met also appears in two tandem copies in both sexes, each having the same anticodon (CAT; Fig 1). The gene arrangement observed here, although incomplete, is novel among sampled metazoans, once more supporting the observed high rearrangement rate in molluscan mtDNAs (BOORE 1999 and references therein). The complete sequencing of the M- and F-like mtDNA genomes is now in progress.

View larger version (9K): In this window In a new window Download PPT slide   Figure 1. Partial gene arrangement of T. philippinarum M- and F-like sequences for the 9.2-kb analyzed sample (U.S. sample set). The difference between sexes is marked by shading.

Four out of a possible seven start codons were found to be used in T. philippinarum, namely ATA (female atp6, nad6; male nad4, cox1), ATG (female nad4; male nad6), GTG (female nad3; male atp6, nad3), and GTT (female nad5, cox3, cox1; male nad5, cox3). The stop codon TAA is used in females for nad4, nad3, nad5, nad6, and cox3 and in males for nad4, nad3, and nad6. The stop codon TAG is used for all other genes (the cox1 stop codon was not detected since its sequence is incomplete at the 3' end). No abbreviated stop codons, sensu BOORE and BROWN 2000 , were inferred for the analyzed portion of the mtDNA. The M sequence is shorter [8986 total nucleotides (nt)] compared with the F sequence (9207 nt). This is due mainly to the different length of male genes, which are commonly truncated at both the 5' and 3' ends (see especially nad4, cox3, and rrnS genes). It is unclear, however, whether these deletions may in any way interfere with male mtDNA function.

Sequence variability:
In T. philippinarum the variability of M sequences is greater than that of F sequences for all but one analyzed gene. Values of sequence divergence, evaluated by P-distance, are between 0.0020 (rrnL, nad3) and 0.0081 (nad6) in F-type and between 0.0028 (rrnL) and 0.0094 (cox3) in M-type sequences. However, ANOVA, performed on each sequenced gene, showed a quite complex situation, since differences in divergence levels are statistically significant in only some genes [rrnL, nad3, nad5, cox1, cox3, and (collectively) tRNAs] but not in others (rrnS, atp6, nad4, and nad6). Nonetheless, when the complete 9.2-kb sequences were compared, a highly significant value is obtained (P = 0.0022), thus indicating that a higher M-type sequence variability is a general feature of the mtDNA of T. philippinarum, even if some gene-specific differences do exist (Table 1).

Values of divergence are considerably higher for pairwise M- vs. F-type comparisons, ranging from 0.1549 (rrnL) to 0.3284 (nad6). Moreover, different genes show different levels of divergence, but it is worth mentioning that rRNA and tRNA genes are less divergent than protein-coding genes (Table 1). The whole set of data on sequence variability appears to be in line with what is observed in other DUI species, but the level of divergence within M- and F-type sequences still remains much lower in T. philippinarum than in M. edulis (see RAWSON and HILBISH 1995 ).

As we plotted the percentage of transitions as a function of sequence divergence, we observed a strong transition bias for all analyzed genes (single-gene comparisons not shown) either within or between M- and F-types, with the only exception being the partial sequence of rrnL, in which M- and F-type sequences appear to mutate without any bias toward transitions (Fig 2). It is unclear whether this deviation is somehow related to structural constraints on the LrRNA product or whether it is influenced by the Italian sample history (i.e., by the variation present in the founding populations).

View larger version (12K): In this window In a new window Download PPT slide   Figure 2. Plotting of the transition rate as a function of the percentage of sequence divergence on (top) rrnL partial sequence (Italian sample set) and on (bottom) the 9.2-kb mtDNA fragment (U.S. sample set). The following comparisons were carried out in T. philippinarum: M-type vs. M-type (), F-type vs. F-type (), and M-type vs. F-type ().

Comparisons within groups (i.e., M- and F-type sequences) by the Fisher's exact test (Table 2) reveal that a selection against nonsynonymous mutations is working for all analyzed protein-coding genes (i.e., purifying selection). McDonald and Krietman's test of neutrality indicates that, for six out of seven analyzed genes, M- and F-type sequences evolve over a long time of divergence in a nonneutral way (Table 3).

Recombination:
Analysis failed to show any trace of recombination in the 9.2-kb partial mtDNA sequences obtained from the U.S. samples. Although this could reflect a sampling bias, the whole sequencing effort (a total of 132 kb analyzed) might reduce such a shortcoming. It should be noted that we analyzed samples obtained from gonads so that our sequencing procedures were targeted toward detecting the haplotype that is commonly passed to progeny. This approach was taken mainly because we wanted to know whether or not recombinant mtDNAs, as detected by LAUDOKAKIS and ZOUROS 2001A , are commonly inherited. However, if the PCR sample is cloned and different clones are analyzed, it is virtually impossible to determine whether or not the analyzed sequences are really from germline cells, due to the inevitable contamination by somatic cells.

Out of 30 analyzed clones of rrnL partial sequences obtained from three samples of somatic tissue, we obtained a single sequence that may be interpreted as recombinant in origin and appears fully similar to that observed by LAUDOKAKIS and ZOUROS 2001A . This sequence, although clearly male-like, presents a series of contiguous female-like diagnostic sites in the 429–459 region (Fig 3, GenBank accession no. AF492465).

View larger version (105K): In this window In a new window Download PPT slide   Figure 3. Putative recombinant clone (rrnL partial sequence) obtained from the somatic tissue of an analyzed male (Italian sample set). M-type sequence regions are marked by light shading; F-type sequence regions are marked by dark shading.

This may indicate that recombination of somatic mitochondria is more common in T. philippinarum than in the germline. However, it has to be mentioned that for this clone a PCR artifact cannot be excluded (see PAABO et al. 1990 ).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Molecular evolution of sex-related mtDNA genomes:
These data confirm the presence of a sex-related heteroplasmy in the Manila clam, T. philippinarum. The observed tissue-specific distribution of M and F mitochondrial sequences fully parallels that predicted by a doubly uniparental mechanism of mitochondrial inheritance (SKIBINSKI et al. 1994A , SKIBINSKI et al. 1994B ; ZOUROS et al. 1994A , ZOUROS et al. 1994B ). Moreover, the analysis over a large part of the mitochondrial genome shows an overall higher sequence variability in M than in F mtDNA. However, not all genes behave the same way, because some do not show a significantly higher level of variability in the M sequence (see Table 1).

As already mentioned, three main mechanisms have been proposed to explain a higher sequence variability of M mtDNA: enhanced duplication rate during spermatogenesis, free-radical damage to sperm, or the effects of the smaller population size of the M genome. However, all of these mechanisms would produce an evenly distributed higher level of mutation of the M genome, because they invariably act on the entire mtDNA molecule. Although the previously mentioned mechanisms may act on the M genome, the observed gene-specific variation in effect seems to indicate that different selective forces are acting on individual M genes.

If so, we might also speculate that the hypothesis of STEWART et al. 1996 fits our data. This hypothesis differentiates three selective "arenas" for mitochondria, namely the somatic cell line, the female germline, and the male germline, and assumes that there might be some tradeoffs in terms of optimal functioning. In DUI species, the M genome has to function in only one of the three arenas (i.e., the male germline), whereas the F genome has to function in two (i.e., somatic tissue and the female germline); in non-DUI species, the sole mtDNA type must function in all three. Therefore, perhaps those functioning in fewer arenas can evolve more quickly, adapting to that specific environment, whereas the need to function in more arenas would result in greater purifying selection, accounting for the varying rates of evolution. It is possible that the clam mtDNAs are adapted with greater specialization to the arenas in which they function, accounting for the M and F mtDNAs not having switched roles in T. philippinarum (i.e., low "masculinization" frequency), in contrast to the observation in Mytilus (HOEH et al. 1997 ).

This is also in line with the observation that between M and F genomes, divergence is very high for both synonymous and nonsynonymous positions. In fact, the splitting of the M and F lineages seems to have occurred near the origin of the species itself, as suggested by the relative-rate test performed on rrnL (PASSAMONTI and SCALI 2001 ). The McDonald and Kreitman test also showed that the two variants largely diverged in a nonneutral way, even if some exception occurs (nad3). The ratio of fixed synonymous to nonsynonymous substitutions is significantly different from the ratio of polymorphic synonymous to nonsynonymous substitutions, which does not meet the prediction of a neutral model of evolution. This should be taken as evidence that some sort of selective pressure might act during the divergence of M and F genomes, most likely to better adapt them to their specialized tasks. It may be that gene-specific selection has produced in T. philippinarum two stable (i.e., purifying selection detected within mtDNAs) and divergent (i.e., positive selection detected between mtDNAs) mitochondrial genomes.

This scenario largely parallels results by STEWART et al. 1996 in American M. edulis and M. trossulus, which rejects neutral evolution of mussels' mitochondrial genes, but seems to contrast with the results of QUESADA et al. 1998 , which found an excess of nonsynonymous substitutions in European mussels.

It has to be noted, however, that we cannot totally exclude a nearly neutral pattern of evolution for T. philippinarum (OHTA 1992 ), according to Quesada's rationale (QUESADA et al. 1998 ). Actually, in a nearly neutral scenario, repeated fluctuations of effective population size can result in a relaxed selection on slightly deleterious mutations (OHTA 1972 , OHTA 1992 ); while the major part of deleterious mutations would be eliminated over a long time, some would be permitted by the more relaxed selection on the M genome (still in a gene-specific pathway), thus resulting in replacement substitutions between the M and F genomes.

Mitochondrial recombination in the Manila clam DUI system:
Absence of recombination in metazoan mtDNA has been a "tenet" for several years and is still considered one of the most important features of mitochondrial DNA as a phylogenetic tool.

However, the subject of mitochondrial recombination has again become controversial in recent years (ARCTANDER 1999 ; AWADALLA et al. 1999 ; EYRE-WALKER et al. 1999 ; HAGELBERG et al. 1999 ; MERRIWEATHER and KAESTLE 1999 ; INGMAN et al. 2000 ; KIVISILD and VILLEMS 2000 ; ELSON et al. 2001 ). Although it is quite clear that there is no molecular reason to consider recombination structurally absent from metazoan mitochondria (THYAGARAJAN et al. 1996 ; LAUDOKAKIS and ZOUROS 2001A , LAUDOKAKIS and ZOUROS 2001B ), it has to be noted that within an all-maternal, homoplasmic mtDNA population, recombinants are hardly detectable by direct observation of gene-converted tracts, since recombinants will generally be identical in sequence to the typical mtDNAs. For this reason, doubly uniparental inheritance represents a unique system to test recombination events, because these species show a high level of mtDNA heteroplasmy. Following this, LAUDOKAKIS and ZOUROS 2001A pointed out the possibility that the ratio of recombinants, which are commonly produced in metazoan cells, is directly proportional to the level of heteroplasmy. Further, even if we accept that mtDNA recombination is possible, then we still must ask whether recombinants are commonly inherited by progeny. This issue is limited by lack of understanding of mechanisms for germline mitochondrial segregation.

Data on T. philippinarum may give some insights on these issues. As reported above, recombinant sequences were not detected in mtDNAs obtained from gonads, but a recombinant sequence was obtained from the soma. This may be an indication that clams do not transmit recombinants to their progeny at a detectable level, even if recombination may occur in their cells with characteristics similar to that of M. edulis (LAUDOKAKIS and ZOUROS 2001A ). This may interface with studies that show a narrow "bottleneck" in the passing on of mitochondrial genomes each generation (not limited to DUI organisms), whereby only a few mitochondria are partitioned into the primordial germ plasm of the zygote. Each resulting egg of the subsequent generation, then, will be populated with mitochondria descending from a small number of progenitors, and those that were founded from deficient organelles can be eliminated by metabolic screening.

In conclusion, from present analysis a quite different picture is emerging for mitochondrial inheritance and selection, and further analyses of DUI systems will probably play a pivotal role in building up a new model for metazoans. A better understanding of the distribution of mtDNAs in both soma and germline may illuminate these processes, and this can be addressed by exploiting the different, co-occurring sex-related haplotypes of mtDNA in DUI species.

	FOOTNOTES

Sequence data from this article have been deposited with the EMBL/GenBank Data Libraries under accession nos. AF484288, AF484289, AF484290, AF484291, AF484292, AF484293, AF484294, AF484295, AF484296, AF484297, AF484298, AF484299, AF484300, AF484301, AF484302, AF484303, AF484304, AF484305, AF484306, AF484307, AF484308, AF484309, AF484310, AF484311, AF484312, AF484313, AF484314, AF484315, AF484316, AF484317, AF484318, AF484319, AF484320, AF484321, AF484322, AF484323, AF484324, AF484325, AF484326, AF484327, AF484328, AF484329, AF484330, AF484331, AF484332, AF484333, AF484334, AF484335, AF484336, AF484337, AF484338, AF484339, AF484340 and AF492465.
² Present address: 2800 Mitchell Dr., Walnut Creek, CA 94598.

	ACKNOWLEDGMENTS

We gratefully acknowledge H. Matthew Fourcade (DOE Joint Genome Institute) for his skillful lab assistance and Cristian Balanzoni (University of Bologna) for his precious help in sampling and characterizing the Italian samples. This work has been supported by Italian MIUR funds. Part of this work was performed under the auspices of the U.S. Department of Energy, Office of Biological and Environmental Research, and by the University of California, Lawrence Berkeley National Laboratory, under contract no. DE-AC03-76SF00098.

Manuscript received May 7, 2002; Accepted for publication February 17, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ARCTANDER, P., 1999  Mitochondrial recombination? Science 284:2090-2091.

AWADALLA, P., A. EYRE-WALKER, and J. MAYNARD-SMITH, 1999  Linkage disequilibrium and recombination in hominid mitochondrial DNA. Science 286:2524-2525.[Abstract/Free Full Text]

BETRÁN, E., J. ROZAS, A. NAVARRO, and A. BARBADILLA, 1997  The estimation of the number and the length distribution of gene conversion tracts from population DNA sequence data. Genetics 146:89-99.[Abstract]

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

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 DNA sequence of the mitochondrial genome of the annelid worm, Lumbricus terrestris.. Genetics 141:305-319.[Abstract]

BOORE, J. L. and W. M. BROWN, 2000  Mitochondrial genomes of Galatheolinum, Helobdella, and Platynereis: sequence and gene arrangement comparisons show that Pogonophora is not a Phylum and Annelida and Arthropoda are not sister taxa. Mol. Biol. Evol. 17:305-319.

BROOKFIELD, J. F. Y. and P. M. SHARP, 1994  Neutralism and selectionism face up to DNA data. Trends Genet. 10:485-488.

DALZIEL, A. C. and D. T. STEWART, 2002  Tissue-specific expression of male-transmitted mitochondrial DNA and its implications for rates of molecular evolution in Mytilus mussels (Bivalvia: Mytilidae). Genome 45:348-355.[Medline]

DEAN, F. B., J. R. NELSON, T. L. GIESLER, and R. S. LASKEN, 2001  Rapid amplification of plasmid and phage DNA using Phi29 DNA polymerase and multiply-primed rolling circle amplification. Genet. Res. 11:1095-1099.

ELSON, J. L., R. M. ANDREWS, P. F. CHINNERY, R. N. LIGHTOWLERS, and D. M. TURNBULL et al., 2001  Analysis of European mtDNAs for recombination. Am. J. Hum. Genet. 68:145-153.[Medline]

EYRE-WALKER, A. and P. AWADALLA, 2001  Does human mtDNA recombine? J. Mol. Evol. 53:430-435.[Medline]

EYRE-WALKER, A., N. H. SMITH, and J. MAYNARD-SMITH, 1999  How clonal are human mitochondria? Proc. R. Soc. Lond. B Biol. Sci. 266:477-483.[Medline]

FISCHER-PIETTE, E., and B. METIVIER, 1971 Révision des Tapetinae (Mollusques Bivalves). Memoires du Museum National d'Histoire Naturelle, Paris.

FOLMER, O., M. BLACK, W. HOEH, R. LUTZ, and R. VRIJENHOEK, 1994  DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol. Marine Biol. Biotechnol. 3:294-299.[Medline]

HAGELBERG, E., N. GOLDMAN, P. LIÓ, S. WHELAN, and W. SCHIEFENHÖVEL et al., 1999  Evidence for mitochondrial DNA recombination in a human population of island Melanesia. Proc. R. Soc. Lond. B Biol. Sci. 266:485-492.[Medline]

HOEH, W. R., D. T. STEWART, B. W. SUTHERLAND, and E. ZOUROS, 1996  Cytochrome c oxidase sequence comparisons suggest an unusually high rate of mitochondrial DNA evolution in Mytilus (Mollusca: Bivalvia). Mol. Biol. Evol. 13:418-421.[Medline]

HOEH, W. R., D. T. STEWART, C. SAAVEDRA, B. W. SUTHERLAND, and E. ZOUROS, 1997  Phylogenetic evidence of role-reversals of gender-associated mitochondrial DNA in Mytilus (Bivalvia: Mytilidae). Mol. Biol. Evol. 14:959-967.[Abstract]

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

INGMAN, M., H. KAESSMANN, S. PÄÄBO, and U. GYLLENSTEN, 2000  Mitochondrial genome variation and the origin of modern humans. Nature 408:708-713.[Medline]

JANSEN, R. P., J. A. BARRIT, C. A. BRENNER, S. WILLADSEN, and J. COHEN et al., 2000  Germline passage of mitochondria: quantitative considerations and possible embryological sequelae. Hum. Reprod. 15(Suppl. 2):112-128.

KAJANDER, O. A., A. T. ROVIO, K. MAJAMAA, J. POULTON, and J. N. SPELBRINK et al., 2000  Human mtDNA sublimons resemble rearranged mitochondrial genomes found in pathological states. Hum. Mol. Genet. 9:2821-2835.[Abstract/Free Full Text]

KIVISILD, T. R., and R. VILLEMS, 2000 Questioning evidence for recombination in human mitochondrial DNA. Science 288: 1931.

KUMAR, S., K. TAMURA, I. B. JAKOBSEN and M. NEI, 2001 MEGA2: Molecular Evolutionary Genetics Analysis. Arizona State University, Tempe, AZ.

LAUDOKAKIS, E. D. and E. ZOUROS, 2001a  Direct evidence for homologous recombination in mussel (Mytilus galloprovincialis) mitochondrial DNA. Mol. Biol. Evol. 18:1168-1175.[Abstract/Free Full Text]

LAUDOKAKIS, E. D. and E. ZOUROS, 2001b  Recombination in animal mitochondrial DNA: evidence from published sequences. Mol. Biol. Evol. 18:2127-2131.[Free Full Text]

LIU, H. P., J. B. MITTON, and S. K. WU, 1996  Paternal mitochondrial DNA differentiation far exceeds maternal mitochondrial DNA and allozyme differentiation in the freshwater mussel, Anodonta grandis grandis.. Evolution 50:952-957.

LOWE, T. M. and S. R. EDDY, 1997  tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 25:955-964.[Abstract/Free Full Text]

MCDONALD, J. H. and M. KREITMAN, 1991  Adaptive protein evolution at the Adh locus in Drosophila.. Nature 351:652-654.[Medline]

MERRIWEATHER, A. and A. F. KAESTLE, 1999  Mitochondrial recombination? Science 285:835.

NEI, M., and S. KUMAR, 2001 Molecular Evolution and Phylogenetics. Oxford University Press, New York.

OHTA, T., 1972  Population size and rate of evolution. J. Mol. Evol. 1:305-314.[Medline]

OHTA, T., 1992  The nearly neutral theory of molecular evolution. Annu. Rev. Ecol. Syst. 23:263-286.

OHTA, T., 1996  The current significance and standing of neutral and nearly neutral theories. Bioessays 18:673-677.[Medline]

PÄÄBO, S., D. M. IRWING, and R. G. HARRISON, 1990  DNA damage promotes jumping between templates during enzymatic amplification. J. Biol. Chem. 265:4718-4721.[Abstract/Free Full Text]

PALUMBI, S., A. MARTIN, S. ROMANO, W. O. MCMILLAN, L. STICE et al., 1991 The Simple Fool's Guide to PCR. Department of Zoology and Kewalo Marine Laboratory, University of Hawaii, Honolulu.

PASSAMONTI, M. and V. SCALI, 2001  Gender-associated mitochondrial DNA heteroplasmy in the venerid clam Tapes philippinarum (Mollusca: Bivalvia). Curr. Genet. 39:117-124.[Medline]

PREISS, A., D. A. HARTLEY, and S. ARTAVANIS-TSAKONAS, 1988  Molecular genetics of enhancer of split, a gene required for embryotic neural development in Drosophila.. EMBO J. 12:3917-3927.

QUESADA, H., M. WARREN, and D. O. F. SKIBINSKI, 1998  Nonneutral evolution and differential mutation rate of gender-associated mitochondrial DNA lineages in the marine mussel Mytilus. Genetics 149:1511-1526.[Abstract/Free Full Text]

RAWSON, P. D. and T. J. HILBISH, 1995  Evolutionary relationships among the male and female mitochondrial DNA lineages in the Mytilus edulis species complex. Mol. Biol. Evol. 12:893-901.[Abstract]

ROZAS, J. and R. ROZAS, 1999  DnaSP version 3: an integrated program for molecular population genetics and molecular evolution analysis. Bioinformatics 15:174-175.[Abstract/Free Full Text]

SHOUBRIDGE, E. A., 2000  Mitochondrial DNA segregation in the developing embryo. Hum. Reprod. 15(Suppl. 2):229-234.[Free Full Text]

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]

SKIBINSKI, D. O. F., C. GALLAGHER, and U. QUESADA, 1999  On the roles of selection, mutation and drift in the evolution of mitochondrial DNA diversity in British Mytilus edulis (Mytilidae; Mollusca) populations. Biol. J. Linn. Soc. 68:195-213.

STEWART, D. T., E. R. KENCHINGTON, R. K. SIGH, and E. ZOUROS, 1996  Degree of selective constraint as an explanation of the different rates of evolution of gender-specific mitochondrial DNA lineages in the mussel Mytilus. Genetics 123:585-595.

THYAGARAJAN, B., R. A. PADUA, and C. CAMPBELL, 1996  Mammalian mitochondria possess homologous DNA recombination activity. J. Biol. Chem. 271:27536-27543.[Abstract/Free Full Text]

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

ZHANG, J., S. KUMAR, and M. NEI, 1997  Small-sample tests of episodic adaptive evolution: a case study of primate lysozymes. Mol. Biol. Evol. 14:1335-1338.[Medline]

ZOUROS, E., B. A. OBERHAUSER, C. SAAVEDRA, and K. R. FREEMAN, 1994a  Mitochondrial DNA inheritance. Nature 368:817-818.

ZOUROS, E., B. A. OBERHAUSER, C. SAAVEDRA, and K. R. FREEMAN, 1994b  An unusual type of mitochondrial DNA inheritance in the blue mussel Mytilus. Proc. Natl. Acad. Sci. USA 91:7463-7467.[Abstract/Free Full Text]

This article has been cited by other articles:

				B. S. Ort and G. H. Pogson Molecular Population Genetics of the Male and Female Mitochondrial DNA Molecules of the California Sea Mussel, Mytilus californianus Genetics, October 1, 2007; 177(2): 1087 - 1099. [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]