Mitochondrial DNA Heteroplasmy in Wheat, Aegilops and Their Nucleus-Cytoplasm Hybrids

Corresponding author: Chiharu Nakamura, Department of Biological and Environmental Science, Faculty of Agriculture, Kobe University, Rokkodai-cho, Nada-ku, Kobe 657-8501, Japan., nakamura{at}kobe-u.ac.jp (E-mail)

Communicating editor: B. S. GILL

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A mitochondrial (mt) transcriptional unit, nad3-orf156, was studied in the nucleus-cytoplasm hybrids of wheat with D/D2 plasmons from Aegilops species and their parental lines. A comparative RFLP analysis and sequencing of the random PCR clones revealed the presence of seven sequence types and their polymorphic sites were mapped. All the hybrids possessed the paternal copies besides the maternal copies. More paternal copies were present in the D2 plasmon hybrids, whereas more maternal copies were present in the D plasmon hybrids. Two major copies were present with different stoichiometries in the maternal Aegilops parents. However, only a major D plasmon copy was detected in the hybrids, irrespective of their plasmon types. The hexaploid wheat parent (AABBDD genome) possessed the major D plasmon copy in 5% stoichiometry, while no D plasmon-homologous copies were detected in the tetraploid wheat parent (AABB genome). The results suggest that the observed mtDNA heteroplasmy is due to paternal contribution of mtDNA. The different copy stoichiometry suggests differential amplification of the heteroplasmic copies among the hybrids and the parental lines. All editing sites and their editing frequencies were conserved among the lines, and only the maternal pattern of editing occurred in the hybrids.

THE genetic system of the cytoplasmic genome (plasmon) differs significantly from that of the nuclear genome in various aspects. A most prominent difference is that plasmon is inherited uniparentally, mostly through maternal lineage. A large extent of polymorphism can frequently be found in a given plasmon among different populations and individuals within a given species. Despite this apparent diversity and the high level of polyploidy, individual plasmons are genetically haploid because they often become homoplasmic by accumulating identical DNA molecules in individuals. It is generally considered that plasmon heteroplasmy, which is defined by the presence of different DNA molecules in organella and/or cells within individuals, can be rapidly purified during ontogeny, thus resulting in the homoplasmic state. Even in yeast, Paramecium, Oenothera, and Pelargonium, in which biparental inheritance occurs, heteroplasmic states rapidly change to homoplasmic states within a few generations (GRUN 1976 ).

The mitochondrial DNA (mtDNA) heteroplasmy is now widely recognized in a number of animal taxa, including Drosophila (SOLIGNAC et al. 1983 ; DE STORDEUR et al. 1989 ; KONDO et al. 1990 ; KANN et al. 1998 ; DOI et al. 1999 ), mouse (GYLLENSTEIN et al. 1991 ; JENUTH et al. 1997 ), marine mussel (HOEH et al. 1991 ; ZOUROS et al. 1992 ), human (JAZIN et al. 1996 ), and others. Recently, mtDNA heteroplasmy has attracted much attention because this phenomenon is expected to provide a potentially valuable means for clarifying the molecular mechanisms of mtDNA transmission and proliferation and for establishing animal models of human mitochondrial diseases (JAZIN et al. 1996 ; JENUTH et al. 1996 , JENUTH et al. 1997 ; CHINNERY et al. 2000 ; STEINBORN et al. 2000 ; COLLER et al. 2001 ). In plants, the most well-documented case of mtDNA heteroplasmy is related to cytoplasmic male sterility. Cytoplasmic male sterility is often associated with the presence and expression of novel chimeric genes produced by rearrangements of mtDNA, which are under the control of nuclear fertility restorer genes (MACKENZIE et al. 1985 ; HE et al. 1995 ; JANSKA et al. 1998 ). On the other hand, mtDNA heteroplasmy that is potentially associated with the paternal transmission of mtDNA molecules has not yet been proven. The first evidence for the presence of a paternal-identical mtDNA sequence in plants was obtained in the orf25 region in the human-made wheat/rye hybrid crop, triticale (LASER et al. 1997 ). The observation suggests that such a sequence might have been derived through leaky transmission of the paternal rye mtDNA. However, the rye sequence was detected in a small substoichiometric amount (0.1%) in the maternal wheat mitochondria. Although the origin and significance of this rye-homologous sequence in the wheat mtDNA remain unknown, it was considered that the sequence already present in the maternal wheat mtDNA was differentially amplified in this intergeneric hybrid crop.

Nucleus-cytoplasm (NC) hybrids or alloplasmic hybrids have been produced by the long-term recurrent backcrossing strategy in plants. This process enables the combination of a given nuclear genome with a particular plasmon from a different source. Such NC hybrids have been used extensively for studying phylogeny, co-evolution, and interaction of nuclear genomes and plasmons, particularly in the subtribe Triticinae (KIHARA 1951 ; TSUNEWAKI 1993 , TSUNEWAKI 1996 ). Because of the combination of different plasmons and nuclear genomes, these NC hybrids are expected to provide suitable materials for investigating the heteroplasmy that may persist in their plasmons throughout the repeated backcrosses and self-fertilization. In a previous study we detected mtDNA heteroplasmy in the polycistronic transcriptional unit nad3-orf156, which showed the simultaneous presence of the maternal and paternal sequences in tetraploid wheat NC hybrids with the D plasmon from Aegilops squarrosa (TSUKAMOTO et al. 2000 ). A similar result was obtained in a series of hexaploid wheat NC hybrids with various Triticum and Aegilops plasmons, including the D and D2 plasmons (TSUKAMOTO et al. 2000 ; Y. NAKAGAMI and C. NAKAMURA, unpublished results). We now report conclusive evidence that demonstrates the extensive presence of mtDNA heteroplasmy in the nad3-orf156 region in tetraploid and hexaploid NC hybrids with the D and D2 plasmons as well as in their parental pure lines. The presence of paternal-identical and paternal-derivative sequences in the NC hybrids and the apparent absence of such sequences in the maternal parents suggested the paternal contribution to the observed mtDNA heteroplasmy.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Plant materials:
Seven lines of NC hybrids of hexaploid (or common) wheat (Triticum aestivum L. Thell cv. "Chinese Spring"; AABBDD genome with B plasmon) were used together with their maternal and paternal parents (Table 1). These NC hybrid lines possess D or D2 plasmon derived from the following six Aegilops species: two accessions of Ae. squarrosa (KU20-2 2x and KU29 4x, both with D plasmon) and one accession each of Ae. ventricosa (4x, D), Ae. crassa (4x, D2), Ae. crassa (6x, D2), Ae. juvenalis (6x, D2), and Ae. vavilovii (6x, D2). Because of the repeated backcrosses, their nuclear genome is considered to be nearly equal to that of the paternal recurrent parent. One NC hybrid of T. durum Desf. cv. "Langdon" (AABB genome with B plasmon) with D plasmon was also used as reference. The maternal parent of this tetraploid NC hybrid is a colchicine-doubled tetraploid derivative, Ae. squarrosa (KU29), which was derived from diploid Ae. squarrosa (KU20-2) that was used for the production of the hexaploid NC hybrids with the D plasmon (OHTSUKA 1991 ).

The nuclear genome of hexaploid wheat is compatible with the D plasmon of Ae. squarrosa and thus euploid NC hybrids can be produced, because of the presence of a pair of NC compatibility genes, Ncc-sqr1D, on chromosome 1D from the paternal Ae. squarrosa (a D genome donor to hexaploid wheat species). On the other hand, the nuclear genome of a majority of tetraploid wheat species including T. durum is incompatible with the D plasmon (OHTSUKA 1991 ). The nuclear compatibility with the D plasmon in the tetraploid NC hybrids can be restored by the presence of Ncc-sqr1D on an added monosomic or disomic chromosome 1D (OHTSUKA and KIHARA 1976 ; TSUJI and MURATA 1976 ). A functionally homeologous NC compatibility gene, Ncc-tmp1A, is present in the timopheevi group of wheat (cultivated T. timopheevi and wild T. araraticum; AAGG genome with G plasmon). A viable, fertile, and euploid NC hybrid of T. durum cv. Langdon with the D plasmon of Ae. squarrosa was produced by introgressing a pair of the Ncc-tmp1A genes derived from T. timopheevi (ASAKURA et al. 1997 , ASAKURA et al. 2000 ).

PCR amplification and restriction fragment length polymorphism (RFLP) analysis of the mitochondrial nad3-orf156 region:
Total DNA was extracted according to TSUKAMOTO et al. 2000 from eight lines of NC hybrids and five parental pure lines (Table 1). An equal amount of DNA from 10 individual plants was bulked to equalize each line. The mitochondrial nad3-orf156 region, which encompasses four open reading frames (ORFs; the 3' end of nad3, rps12, and orf299, containing a duplicated part of the first exon of coxII and orf156), was amplified by PCR using the oligonucleotide primer set A: AF, 5'-ATGAATGGAAAAGGGGTGCTT-3' and AR, 5'-GGAGAAGACATAACCAGAAGA-3' (region I in Fig 1). Twenty-five cycles of PCR were performed using rTaq DNA polymerase (TOYOBO, Osaka, Japan) and GeneAmp PCR System 9600 (Perkin-Elmer/Cetus, Norwalk, CT). The PCR program was conducted as follows: a predenaturation step for 5 min at 94°, a denaturation step for 1 min at 94°, an annealing step for 2 min at 55°, and an extension step for 3 min at 72°, followed by a postextension for 2 min at 72°. The PCR products were digested with DdeI and MspI, both of which are known to generate polymorphic fragments from region I in a previous study (TSUKAMOTO et al. 2000 ). The restricted PCR products were fractionated by electrophoresis through 8% polyacrylamide gel and visualized by staining with ethidium bromide to reveal polymorphic fragments. The fragments were blotted onto nylon membranes (Boehringer Mannheim, Indianapolis) in 0.4 N NaOH. Southern blot hybridization was performed according to TSUKAMOTO et al. 2000 using the cloned PCR product of region I as a probe, which was obtained as a major copy from the paternal hexaploid wheat parent. Signals were analyzed by an enhanced chemiluminescence direct nucleic acid labeling and detection system (Amersham, Buckinghamshire, UK).

View larger version (11K): In this window In a new window Download PPT slide   Figure 1. A schematic illustration of the nad3-orf156 region. ORFs are indicated by boxes and the primer sets A (for region I) and B (region II) for PCR amplification are indicated by arrows. A primer C was used to sequence a region covering one insertion of a GAC triplet in the b3 and b4 copies (see Table 2). The numbers indicate nucleotide positions of the consensus b1 copy of B plasmon.

Identification of the sequence types in the nad3-orf156 region and estimation of their copy stoichiometry:
The PCR-amplified region I was cloned into pGEM-T vector (Promega, Madison, WI) and used to transform Escherichia coli JM109. Insert sequences of the random clones were determined by the automated fluorescence dye deoxy terminator cycle sequencing system using ABI Prism 310 Genetic Analyzer (PE Applied Biosystems). The inserts were reamplified by the described PCR program using the primer set A and their RFLP patterns were analyzed after digestion with the two restriction enzymes. The insert sequences and their PCR-RFLP patterns were compared to identify individual clones from region I. Two sequence types that could not be distinguished by PCR-RFLP patterns were identified by examining one polymorphic site between them (b1 and b2 copies; see RESULTS) after sequencing with a primer C, 5'-AGACCTAAATCGAAATGAAT-3' (Fig 1). On the basis of the identification of individual sequence types, their relative frequencies (stoichiometries) were estimated by counting the numbers of the corresponding random clones from each line.

Analysis of RNA editing in the orf156:
Total RNA was extracted from 10 plants (2- to 3-week-old seedlings), each from the tetraploid (no. 1 in Table 1) and hexaploid (no. 2) wheat NC hybrids with the D plasmon of Ae. squarrosa and the three parental lines (nos. 9, 12, and 13). RNA extraction was performed according to the guanidinium thiocyanate method using ISOGEN (Nippongene, Tokyo, Japan). The first-strand cDNAs were synthesized using Rever Tra Ace- (TOYOBO) from region II containing the entire orf156 sequence using the primer set B: BF, 5'-CTGAGTTCTTCCCTCTTGAT-3' and BR, 5'-GAAAAACTTGCTTGCTTCTT-3' (Fig 1). The resulting cDNAs were amplified by the described PCR program using the primer set B and rTaq DNA polymerase. The amplified fragments were cloned into pGEM-T vector and sequenced. The sequences of randomly obtained RT-PCR products were compared with the corresponding genomic sequences, i.e., with the consensus sequences from the three parental lines.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Identification of the heteroplasmic sequence types in the mitochondrial nad3-orf156 region:
The mitochondrial nad3-orf156 region of common wheat contains the nad3 and rps12 genes, which respectively encode subunit 3 of NADH dehydrogenase and ribosomal protein S12 and two other open reading frames, orf299 and orf156 (Fig 1). The orf299 contains a duplicated first exon of coxII (cytochrome oxidase subunit II) and is presumably a pseudogene with no corresponding protein product (GUALBERTO et al. 1991 ). The orf156 has recently been shown to have homology with the mitochondrial atp8 (ATP synthase subunit 8) in sugar beet (KUBO et al. 2000 ). We have previously shown that this region (designated as region I with 1837 bp in Fig 1) assumes heteroplasmic states in both tetraploid and hexaploid NC hybrids of wheat with the D and D2 plasmons (TSUKAMOTO et al. 2000 ). Several other mtDNA regions were also found to be heteroplasmic, but by the PCR-RFLP analysis heteroplasmic states could be confirmed only in this region. To further characterize the heteroplasmic states, region I was amplified using DNA extracted from the two hexaploid hybrids, one with the D plasmon of Ae. squarrosa 2x (no. 2 of Table 1) and one with the D2 plasmon of Ae. crassa 4x (no. 5). The same region was amplified from the tetraploid hybrid with the D plasmon of Ae. squarrosa 4x (no. 1) and all five parental lines (nos. 9–13). PCR products were single digested with DdeI and MspI, and the restriction patterns after polyacrylamide gel electrophoresis were visualized by staining with ethidium bromide. The gels were blotted onto nylon membranes and Southern hybridization was performed using the consensus sequence (b1 copy) obtained from the paternal hexaploid wheat parent, cv. Chinese Spring, as a probe. Digestion with DdeI and MspI revealed the presence of the maternal, paternal, and novel fragments in the hybrids as compared with the parental lines. The D and D2 plasmons of the maternal Aegilops parents also showed different restriction fragments. No differences were observed between the D plasmon of the diploid and tetraploid Ae. squarrosa parents (nos. 9 and 10), the latter of which was produced by colchicine-mediated chromosome doubling of the former and used as a maternal parent to produce the tetraploid hybrid.

To identify the sequence copies that generated polymorphic restriction fragments, the PCR products were cloned into pGEM-T vector and transformed into E. coli, and their inserts were reamplified. Polyacrylamide gel electrophoresis of the reamplified inserts of the random clones after digestion with DdeI revealed three types of clones in the hybrids and their parents (Fig 2A). The same analysis with MspI revealed five types of clones. Nucleotide sequences of the random clones showing the representative PCR-RFLP patterns were then determined. This comparative analysis revealed that clones showing the three DdeI patterns consisted of a group of three maternal copies (designated as d1, d2, and d3), another group of three (b1, b2, and b3), and one (b4) of paternal copies. The three maternal copies, which could not be distinguished by the analysis with DdeI, could clearly be distinguished by the analysis with MspI. The sequence types and their PCR-RFLP fragments are schematically illustrated in Fig 2B.

View larger version (27K): In this window In a new window Download PPT slide   Figure 2. Five sequence types of the nad3-orf156 region identified by PCR-RFLP patterns of the cloned region. Region I was amplified by PCR with the primer set A, single digested with DdeI and MspI, resolved by polyacrylamide gel electrophoresis, and blotted onto nylon membrane. (A) Three sequence types identified by DdeI and five sequence types identified by MspI. Relative mobility of small DdeI fragments (fragments 4, 5, and 6) did not correspond to their molecular sizes in the polyacrylamide gel. (B) Restriction maps of DdeI and MspI sites. Fragment numbers correspond to those shown in A. The smaller fragments (one for DdeI and three for MspI) are not seen on the gels shown in A.

Combining the sequence data and the PCR-RFLP patterns of the clones, all of the detected polymorphic sites among the seven sequence types were mapped in region I (Table 2, Table 3, and Table 4). The polymorphisms were either single nucleotide substitutions (SNPs) or insertions/deletions. The SNPs and insertions/deletions occurred mainly in the intergenic regions and inside of the orf299. Among the four paternal sequence copies, a major b1 copy was identical to the previously reported sequence from the B plasmon of tetraploid and hexaploid wheat (GUALBERTO et al. 1991 ; TSUKAMOTO et al. 2000 ). The b2 copy had a single 28-bp deletion in the 3' part of the orf299 at the nucleotide position 1653–1680 as compared with the consensus b1 copy (Table 2). Due to this deletion, a new TAG stop codon occurred 35 bp downstream of the 3' end of the orf299 at position 1699 (Table 2 and Table 3). The b2 copy thus lacked four amino acids but possessed six additional amino acids in the C-terminal end of the ORF299. The b2 copy was previously detected as a paternal-derivative sequence copy in the tetraploid hybrid with the D plasmon (TSUKAMOTO et al. 2000 ), but it was not detected in any of the lines in this study, indicating that this is a minor copy. The b3 copy differed from the b1 copy only by the presence of a single GAC insertion (coding for aspartic acid) at the nucleotide position 776–777 (Table 2). The b4 copy was remarkable by possessing 50 SNPs, of which two resulted in two TAG stop codons at the positions 164 (inside of the rps12) and 1149 (inside of the orf299; Table 4). In the b4 copy, however, a single GAC insertion found in the b3 copy was conserved at the same position. Because of the new stop codon within rps12, this b1-derivative b4 copy apparently represents a defective version of the b1 copy.

The sequence analysis and PCR-RFLP analysis after digestion of the random clones with MspI revealed the presence of three maternal sequence types. The d1 and d2 copies were the two major types and the d3 copy was the rare type in both D and D2 plasmons of the maternal Aegilops parents. The d1 copy was identical to the sequence previously found in the D plasmon of Ae. squarrosa and in the tetraploid hybrids with the D plasmon (TSUKAMOTO et al. 2000 ). The d1 copy differed from the b1 copy by 18 SNPs and one 5-bp insertion, resulting in 5 amino acid substitutions (Table 2). The d2 copy differed from the b1 copy by 36 SNPs, two 1-bp deletions, and one 6-bp insertion, resulting in 12 amino acid substitutions (Table 2). The 1-bp deletion at the position 1586 caused a new TAG stop codon at the position 1699. Due to this deletion, a frame shift occurred to cause substitutions of the entire amino acids downstream of position 1591 (Table 3). The d3 copy differed from the b1 copy by 24 SNPs, two 1-bp and one 6-bp insertion, and one 1-bp deletion, resulting in 10 amino acid substitutions (Table 2 and Table 3). One of the two 1-bp insertions at position 1282–1283 caused a new TAG stop codon at position 1298. Although the orf299 per se might be a pseudogene (GUALBERTO et al. 1991 ), the d3 copy apparently represents a defective version of the consensus d1 copy.

On the basis of the detection of the seven different sequence types, the heteroplasmic states in region I were further studied in the five other hexaploid wheat hybrids with the D and D2 plasmons from different Aegilops species (Table 1). The paternal sequence copies were detected simultaneously with the maternal sequence copies in all the hybrids by the PCR-RFLP analysis. Agreeing with the previous result (TSUKAMOTO et al. 2000 ), the PCR-RFLP analysis and Southern blot analysis using the b1 copy as a probe showed the presence of stoichiometrically different sequence copies in the hybrids; more maternal copies were detected in the D plasmon hybrids and more paternal copies were detected in the D2 plasmon hybrids (data not shown).

Estimation of the relative stoichiometry of the maternal and paternal copies in the nad3-orf156 region:
The relative stoichiometry of the maternal and paternal copies in the NC hybrids (nos. 1, 2, and 5 in Table 1) was studied on the basis of the number of each representative sequence class among the random PCR clones of region I. The relative stoichiometry of the different sequence copies in the parental lines (nos. 10–13) was also studied. Although we did not apply competitive PCR for this study, a method of random cloning of PCR products was shown to give an equivalent result in the quantification of different mtDNA copies in triticale (LASER et al. 1997 ). The estimated stoichiometry of the component sequence copies agreed well with that previously estimated by Southern blot analysis (TSUKAMOTO et al. 2000 ). Each sequence class was identified by the PCR-RFLP profiles based on the classification shown in Fig 2. Two paternal b1 and b3 copies, which could not be distinguished by the restriction analysis, were made distinguishable by sequencing the region covering the single GAC insertion in the b3 copy. This was performed using primer C (Fig 1). In the D plasmon hexaploid hybrid, all of the 11 detected b1/b3 clones were determined by this method (Fig 3). In the D2 plasmon hexaploid hybrid, the relative stoichiometry of the b1 and b3 copies was estimated based on the number of each sequence type among 22 clones randomly selected out of 78 b1/b3 clones.

View larger version (49K): In this window In a new window Download PPT slide   Figure 3. A schematic illustration of the relative stoichiometries of the seven sequence copies detected in region I. Figures in parentheses after the copy types indicate percentages of the corresponding clones. Figures in parentheses after the lines indicate total numbers of clones analyzed. Ldn, T. durum cv. Langdon with B plasmon; sqr, Ae. squarrosa with D plasmon; crs, Ae. crassa 4x with D2 plasmon; CS, T. aestivum cv. Chinese Spring with B plasmon; (sqr)-Ldn, NC hybrid of Ldn with Ae. squarrosa cytoplasm; (sqr)-CS, NC hybrid of CS with Ae. squarrosa cytoplasm; and (crs)-CS, NC hybrid of CS with Ae. crassa 4x cytoplasm. The solid arrows indicate the maternal parents and the dashed arrows indicate the paternal parents. The b2 copy, which was previously detected in the tetraploid NC hybrids with D plasmon, is not shown because it was not detected in any of the lines in this study.

The D plasmon hexaploid hybrid possessed more maternal d1 copy (73.3%) than the paternal b1, b3, and b4 copies, whereas the D2 plasmon hexaploid hybrid possessed more paternal copies (74.2%). In the D plasmon tetraploid hybrid, the paternal b1 copy occupied 8.7% of the random clones and the remainder were all maternal d1 copy. In the D plasmon of the maternal Ae. squarrosa parent, the d1 copy occupied 76.2% of the clones and the remainder were either d2 (19.0%) or d3 (4.8%) copies. In contrast, slightly more d2 copies than d1 copies (50.0 vs. 42.3%) were present in the D2 plasmon of the Ae. crassa 4x parent. It was noted that the maternal copy detected in the tetraploid and hexaploid NC hybrids was all d1 copy and no d2 and d3 copies were detected, irrespective of their plasmon types. Furthermore, the d1 copy was detected in 5 (4.8%) out of 105 clones obtained from the B plasmon of the paternal hexaploid wheat parent, which has the D genome in the nucleus. In the B plasmon of the paternal tetraploid wheat parent, there were 98.5% of the b1 copy and 1.5% of the b3 copy, but d1, d2, and d3 copies were not obtained. No b1 and b1-derivative copies were obtained in the D and D2 plasmons of the maternal Aegilops parents.

Analysis of RNA editing in the orf156:
Using the primer set of AF/AR (Fig 1) a single fragment of 2250 bp in length was amplified; this confirmed that the nad3-orf156 region was contiguous. A considerable level of heteroplasmy was observed in the amplified fragment after digestion with various restriction enzymes (data not shown). The orf156 within region II (663 bp in length) was amplified using the primer set B, and the consensus sequences were determined by comparing sequences of the parental lines and the NC hybrids. The consensus sequence of the B plasmon (b1 copy) of the paternal tetraploid and hexaploid wheat parents was identical to that reported in hexaploid wheat (GUALBERTO et al. 1991 ). The B plasmon sequence and the D plasmon sequence (d1 copy) of the maternal Ae. squarrosa were distinguished by two 1-bp substitutions (C to A at the nucleotide position 2019 and A to G at 2074). The C to A conversion was synonymous and the A to G conversion was nonsynonymous, causing an amino acid substitution from arginine (in the B plasmon) to glycine (in the D plasmon).

We next studied post-transcriptional RNA editing in the orf156 region. We did this to confirm that the orf156 region truly represents the mitochondrial sequence and also to examine if heterozygous transcripts exist among the lines. The consensus genomic sequences served as references in comparing the sequences of randomly selected RT-PCR products. Four major C to U conversions occurred in the conserved sites in both B and D plasmons of the parents (Table 5), which were identical to the editing sites previously reported (GUALBERTO et al. 1991 ). Similarly, all the editing sites were conserved in the NC hybrids. However, only the maternal D-plasmon-derived transcripts were detected in the hybrids, despite the fact that they possessed varying amounts of the paternal copies of this region. No significant differences were observed in the frequency of editing at all the editing sites among the lines, except for site 2 (Table 5), which was a synonymous editing site and showed an editing frequency significantly lower than the other nonsynonymous sites. Although the number of clones studied was limited, conservation of the editing sites and their frequency, at least under our experimental conditions used for RNA extraction, suggest that ORF156 protein has an essential role as ATP subunit 8 in the wheat mitochondria.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have presented conclusive evidence for the presence of heteroplasmic sequence copies of the mitochondrial nad3-orf156 region in the tetraploid and hexaploid NC hybrids of wheat possessing the D and D2 plasmons from the Aegilops species as well as in their parental lines. The mtDNA heteroplasmy was manifested by the simultaneous presence of the maternal and paternal copies and their derivatives in the hybrids and the corresponding heteroplasmic copies in the parental lines. The heteroplasmic copies were identified and classified by mapping SNPs and insertions/deletions in comparison with the consensus sequences, i.e., the b1 copy in the B plasmon and the d1 copy in the D plasmon (Table 2, Table 3, and Table 4). Seven sequence types were distinguished among the lines, although a possible presence of some other hidden sequence types could not be excluded. We have recently obtained results that showed similar heteroplasmic conditions in the nad3-orf156 region among a series of hexaploid wheat hybrids with various plasmons other than the D and D2 from Aegilops and Triticum species (Y. NAKAGAMI and C. NAKAMURA, unpublished results). Our previous results showed that the heteroplasmy is not due to a random event among the individuals but is common within a hybrid population across at least several generations (TSUKAMOTO et al. 2000 ). These results show that the heteroplasmy in the nad3-orf156 region is a general phenomenon in wheat, Aegilops, and their NC hybrids.

An important question about the observed mtDNA heteroplasmy concerns its origin. Paternal transmission of mtDNA occurs in some animal taxa, allowing mtDNA heteroplasmy to arise as a function of organelle variation between parents (JENUTH et al. 1996 ; CHESSER 1998 ; CHINNERY et al. 2000 ). In mouse intra- and interspecific hybrids, participation of paternal mitochondria in fertilization, followed by complete or incomplete elimination of paternal mtDNA, has been demonstrated (KANEDA et al. 1995 ; SHITARA et al. 1998 ). Mitochondrial DNA heteroplasmy may also arise when mutations occur after fertilization, even if mitochondria are uniparentally inherited. Among the three paternal sequence copies detected in the hexaploid wheat NC hybrids, one copy was identical to the consensus b1 copy present in the B plasmon of the tetraploid and hexaploid wheat (Fig 3). No b1 and b1-derivative copies were obtained among random clones from the D and D2 plasmons of the maternal Aegilops species. Within the nad3-orf156 region, nucleotide sequences of the d1 and d2 copies differ significantly from that of the b1 copy (Table 2 and Table 3). Random mutations of these maternal D plasmon copies in the NC hybrids resulting in generation of the paternal-identical b1 copy are considered to be highly unlikely. The novel, apparently defective b4 copy, which was present in the hexaploid hybrids, had a large number of SNPs compared with the b1 and other b-derivative copies, but retained one trinucleotide insertion at the identical position to that in the b3 copy. The hexaploid hybrids used in this study have been maintained not more than 40 years by backcrossing and self-fertilization after their first crosses. Since neither recombination events nor mutation events seem to explain this remarkably high level of polymorphism during such a short period, the b4 copy likely has originated in the hexaploid wheat during evolution and has been transmitted to the hexaploid NC hybrids through the paternal parent. In fact, the b4 copy was detected in a small substoichiometric amount (2.9%) in the paternal hexaploid wheat parent (Fig 3).

The hexaploid wheat parent possessed the d1-identical copy of Ae. squarrosa in 4.8% stoichiometry, whereas no such d-homologous copies were detected in the tetraploid wheat parent among the random clones (Fig 3) or in Southern blots after prolonged exposure (data not shown). Although it is difficult to prove the absence of such copies in the tetraploid wheat parent, the result suggests the transmission of the d1 copy from Ae. squarrosa to the hexaploid wheat. Ae. squarrosa is known to have served as a paternal ancestor that contributed the D nuclear genome to the present-day hexaploid wheat species (KIHARA 1944 ). It is tempting to speculate that the D plasmon was transmitted from Ae. squarrosa to the hexaploid wheat during the natural hybridization process and maintained throughout the subsequent amphidiploidization and propagation by self-fertilization. We have obtained results that confirm the common presence of the d1 copy in synthetic hexaploid wheat lines with varying genetic backgrounds (T. KISHIDA and C. NAKAMURA, unpublished results). In Brassica, pollen transfer of a mitochondrial plasmid was clearly demonstrated (ERICKSON et al. 1989 ). Possible transmission of the paternal mitochondria and mtDNA through pollen during fertilization must be monitored in original hybrids between tetraploid wheat and Ae. squarrosa.

The relative stoichiometry of the observed heteroplasmic copies varied markedly among the different NC hybrids of hexaploid wheat (Fig 3), all of which possess a nearly identical nuclear genome. The parental lines also showed markedly different copy stoichiometries. However, it was noted that, despite the presence of three heteroplasmic copies in the D and D2 plasmons of Ae. squarrosa and Ae. crassa 4x, only the d1 copy was detected in the hexaploid hybrids with these plasmons. The complete selection of one copy type in the NC hybrids is striking, indicating some selective advantage of the d1 copy over the d2 copy in combination with the hexaploid wheat nuclear genome. A similar but preferential amplification of the paternal copies occurs among the tetraploid hybrids with the D plasmon under different nuclear backgrounds (TSUKAMOTO et al. 2000 ). Many factors could influence the extent of mtDNA heteroplasmy and the relative stoichiometry of the heteroplasmic copies. Random genetic drift and balancing selection have been proposed as two major mechanisms controlling the mtDNA heteroplasmy/homoplasmy (JENUTH et al. 1996 ; CHESSER 1998 ; CHINNERY et al. 2000 ; COLLER et al. 2001 ). In the case of cytoplasmic male sterility, nuclear fertility-restorer genes can control the relative stoichiometry of the fertility/sterility-associated mtDNA molecules (HE et al. 1995 ; JANSKA et al. 1998 ). Although a mechanism controlling the different equilibrium states of the different molecular forms has to be further studied, our results suggest that the differential amplification of the heteroplasmic mtDNA copies might be under control of NC interaction. Such NC interaction can lead to the fixation of certain mtDNA copies. A selective advantage for a particular mitochondrial genome that is provided by the nuclear-mitochondrial interaction has already been reported in Drosophila and human cell cultures (FOS et al. 1990 ; DUNBAR et al. 1995 ).

Three different sequence types with different stoichiometries were detected in the two Aegilops parents (Fig 3). Ae. crassa 4x is considered to have originated from the natural hybridization of Ae. squarrosa as a maternal parent with either one of the two wild Aegilops species, i.e., Ae. comosa or Ae. heldreichii, as a paternal parent (TSUNEWAKI 1996 ). The D2 plasmon of Ae. crassa 4x is known to cause photoperiod-sensitive cytoplasmic male sterility when combined with a specific nuclear genotype of common wheat (MURAI and TSUNEWAKI 1993 ; OGIHARA et al. 1997 ), whereas the D plasmon of Ae. squarrosa does not have such a function. Apparently, the D and D2 plasmons are functionally different in interaction with the wheat nuclear genome. Nevertheless, the simultaneous presence of the d1 and d2 copies, together with the rare d3 copy in both D and D2 plasmons, suggests that the observed heteroplasmic copies might have evolved in the D plasmon of Ae. squarrosa and have been maternally transmitted to the D2 plasmon of Ae. crassa 4x. Such heteroplasmic copies might have been maintained in different equilibrium states in the plasmons of these two Aegilops species under the different nuclear backgrounds.

Plant mtDNA is unique in assuming the complex multipartite structure, which consists of subgenomic circular and linear molecules that are generated by recombination-based rearrangements mediated by repeat sequences and structurally diverged sublimons (PALMER and SHIELDS 1984 ; QUETIER et al. 1985 ; SMALL et al. 1987 ; LONSDALE et al. 1988 ; FRAGOSO et al. 1989 ; FAURON et al. 1992 ; BACKERT et al. 1997 ). The presence of subcircular molecules and sublimons in heteroplasmic conditions undoubtedly contributes to increased genetic diversity of plant mtDNA. We showed that heteroplasmic states occur in at least five other mtDNA regions in the tetraploid NC hybrids with the D plasmon of Ae. squarrosa (TSUKAMOTO et al. 2000 ). Considering this, the nad3-orf156 region might assume only one of several other heteroplasmic sublimons in the wheat mitochondria. In such sublimons, different copies are maintained in different stoichiometric balances in the NC hybrids as well as in the parental species. The nad3-orf156 region is cotranscribed as a polycistronic mRNA followed by generation of different transcripts by processing in wheat (GUALBERTO et al. 1991 ). Editing sites and their frequencies in orf156, however, were completely conserved among the paternal parents and the NC hybrids under our experimental conditions (Table 5). Moreover, only the D-plasmon-derived transcripts were detected in the hybrids, although they possessed both the B and D plasmon copies. The result suggests a mechanism that prevents transcription of the paternal orf156 copies. Precise structures of the mtDNA heteroplasmy and the transcriptional heterogeneity must be further studied to elucidate the physiological significance of this interesting phenomenon in wheat, Aegilops, and their NC hybrids.

NC hybrids have provided valuable experimental tools, particularly in wheat and its related species in the subtribe Triticinae. These species are unique in their evolution: they comprise a series of allopolyploid species evolved by combining different nuclear genomes and plasmons from their diploid ancestors. The phenomenon of mtDNA heteroplasmy would provide a new insight into the phylogeny, evolution, and functional NC interaction in these important plant taxa.

	ACKNOWLEDGMENTS

We thank K. Tsunewaki and I. Ohtsuka for providing us with the original NC hybrids of hexaploid and tetraploid wheat used in this study. This is contribution no. 135 from the Laboratory of Plant Genetics, Faculty of Agriculture, Kobe University. This research was supported in part by a grant-in-aid from the Ministry of Education, Science, Culture and Sports, Japan (no. 13876001 to C.N.).

Manuscript received July 30, 2001; Accepted for publication January 3, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ASAKURA, N., C. NAKAMURA, and I. OHTSUKA, 1997  RAPD markers linked to the nuclear gene from Tritium timopheevii that confers compatibility with Aegilops squarrosa cytoplasm on alloplasmic durum wheat. Genome 40:201-210.

ASAKURA, N., C. NAKAMURA, and I. OHTSUKA, 2000  Homoeoallelic gene Ncc-tmp of Triticum timopheevii conferring compatibility with the cytoplasm of Aegilops squarrosa in the tetraploid wheat nuclear background. Genome 43:503-511[Medline].

BACKERT, S., R. LURZ, O. A. OYARZABAL, and T. BORNER, 1997  High content, size and distribution of single-stranded DNA in the mitochondria of Chenopidium album (L.). Plant Mol. Biol. 33:1037-1050[Medline].

CHESSER, R. K., 1998  Heteroplasmy and organelle gene dynamics. Genetics 150:1309-1327[Abstract/Free Full Text].

CHINNERY, P. F., D. R. THORBURN, D. C. SAMUELS, S. L. WHITE, and H.-H. M. DAHL et al., 2000  The inheritance of mitochondrial DNA heteroplasmy: random drift, selection or both? Trends Genet. 16:500-505[Medline].

COLLER, H. A., K. KHRAPKO, N. D. BODYAK, E. NEKHAEVA, and P. HERRERO-JIMENEZ et al., 2001  High frequency of homoplasmic mitochondrial DNA mutations in human tumors can be explained without selection. Nat. Genet. 28:147-150[Medline].

DE STORDEUR, E., M. SOLIGNAC, M. MONNEROT, and J. C. MOUNOLOU, 1989  The generation of transplasmic Drosophila simulans by cytoplasmic injection: effects of segregation and selection on the perpetuation of mitochondrial DNA heteroplasmy. Mol. Gen. Genet. 220:127-132[Medline].

DOI, A., H. SUZUKI, and E. T. MAYSUURA, 1999  Genetic analysis of temperature-dependent transmission of mitochondrial DNA in Drosophila. Heredity 82:555-560.

DUNBAR, D. R., P. A. MOONIE, H. T. JACOBS, and I. J. HOLT, 1995  Different cellular backgrounds confer a marked advantage to either mutant or wild-type mitochondrial genomes. Proc. Natl. Acad. Sci. USA 92:6565-6566.

ERICKSON, C. M., R. KEMBLE, and E. SWANSON, 1989  The Brassica mitochondrial plasmid can be sexually transmitted: pollen transfer of a cytoplasmic genetic element. Mol. Gen. Genet. 218:419-422.

FAURON, C. M.-R., M. CASPER, R. GESTELAND, and M. ALBERTSEN, 1992  A muti-recombination model for the mtDNA rearrangements seen in maize cmsT regenerated plants. Plant J. 2:949-958.

FOS, M., M. A. DOMINGUEZ, A. LATORRE, and A. MOYA, 1990  Mitochondrial DNA evolution in experimental populations of Drosophila subobscura. Proc. Natl. Acad. Sci. USA 87:4198-4201[Abstract/Free Full Text].

FRAGOSO, L. L., S. E. NICHOLAS, and C. S. LEVINGS, III, 1989  Rearrangements in maize mitochondrial genes. Genome 31:160-168.

GRUN, P., 1976 Cytoplasmic Genetics and Evolution. Columbia University Press, New York.

GUALBERTO, J. M., G. BONNARD, L. LAMATTINA, and J. M. GRIENENBERGER, 1991  Expression of the wheat mitochondrial nad3-rps12 transcription unit: correlation between editing and mRNA maturation. Plant Cell 3:1109-1120[Abstract/Free Full Text].

GYLLENSTEIN, U., D. WHARTON, A. JOSEFSSON, and A. C. WILSON, 1991  Paternal inheritance of mitochondrial DNA in mice. Nature 352:255-257[Medline].

HE, S., A. LYZNIK, and S. MACKENZIE, 1995  Pollen fertility by nuclear gene Fr in CMS bean: nuclear-directed alteration of a mitochondrial population. Genetics 139:955-962[Abstract].

HOEH, W. R., K. H. BLAKLEY, and W. M. BROWN, 1991  Heteroplasmy suggests limited biparental inheritance of Mytillus mitochondrial DNA. Science 22:1488-1490.

JANSKA, H., R. SARRIA, M. WOLOSZYNSKA, M. ARRIETA-MONTIEL, and S. A. MACKENZIE, 1998  Stoichiometric shifts in the common bean mitochondrial genome leading to male sterility and spontaneous reversion to fertility. Plant Cell 10:1163-1180[Abstract/Free Full Text].

JAZIN, E. E., L. CAVELIER, I. ERIKSSON, L. ORELAND, and U. GYLLENSTEN, 1996  Human brain contains high levels of heteroplasmy in the noncoding regions of mitochondrial DNA. Proc. Natl. Acad. Sci. USA 93:12382-12387[Abstract/Free Full Text].

JENUTH, J. P., A. C. PETERSON, K. FU, and E. A. SHOUBRIDGE, 1996  Random genetic drift in the female germline explains the rapid segregation of mammalian mitochondrial DNA. Nat. Genet. 14:146-151[Medline].

JENUTH, J. P., A. C. PETERSON, and E. A. SHOUBRIDGE, 1997  Tissue-specific selection for different mtDNA genotypes in heteroplasmic mice. Nat. Genet. 16:93-95[Medline].

KANEDA, H., J. HAYASHI, S. TAKAHATA, C. TAYA, and K. F. LINDAHL et al., 1995  Elimination of paternal mitochondrial DNA in intraspecific crosses during early mouse embryogenesis. Proc. Natl. Acad. Sci. USA 92:4542-4546[Abstract/Free Full Text].

KANN, L. M., E. B. ROSENBLUM, and D. M. RAND, 1998  Aging, mating, and evolution of mtDNA heteroplasmy in Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 95:2372-2377[Abstract/Free Full Text].

KIHARA, H., 1944  Discovery of the DD-analyzer, one of the ancestors of Triticum vulgare. Agric. Hort. Tokyo 19:13-14. (in Japanese)..

KIHARA, H., 1951  Substitution of nucleus and its effects on genome manifestations. Cytologia 16:177-193.

KONDO, R., Y. SATTA, E. T. MATUURA, H. ISHIKAWA, and N. TAKAHASHI et al., 1990  Incomplete maternal transmission of mitochondrial DNA in Drosophila. Genetics 126:657-663[Abstract].

KUBO, T., S. NISHIZAWA, A. SUGAWARA, N. ITCHODA, and A. ESTIATI et al., 2000  The complete nucleotide sequence of the mitochondrial genome of sugar beet (Beta vulgaris L.) reveals a novel gene for tRNA(Cys)(GCA). Nucleic Acids Res. 28:2571-2576[Abstract/Free Full Text].

LASER, B., S. MOHR, W. ODENBACH, G. OETTLER, and U. KUCK, 1997  Paternal and novel copies of the mitochondrial orf25 gene in the hybrid crop-plant triticale: predominant transcriptional expression of the maternal gene copy. Curr. Genet. 32:337-347[Medline].

LONSDALE, D. M., T. BREARS, T. P. HODGE, S. E. MELVILLE, and W. H. ROTTMANN, 1988  The plant mitochondrial genome: homologous recombination as a mechanism for generating heterogeneity. Philos. Trans. R. Soc. Lond. B Biol. Sci. 319:149-163.

MACKENZIE, S. A., D. R. PRING, M. J. BASSETT, and C. D. CHASE, 1985  Mitochondrial DNA rearrangement associated with fertility restoration and cytoplasmic reversion to fertility in cytoplasmic male-sterile Phaseolus vulgaris L. Proc. Natl. Acad. Sci. USA 85:2714-2717.

MURAI, K. and K. TSUNEWAKI, 1993  Photoperiod-sensitive cytoplasmic male sterility in wheat with Aegilops crassa cytoplasm. Euphytica 67:41-48.

OGIHARA, Y., K. FUTAMI, K. TSUJI, and K. MURAI, 1997  Alloplasmic wheats with Aegilops crassa cytoplasm, which express photoperiod-sensitive homeotic transformations of anthers, show alterations in mitochondrial DNA structure and transcription. Mol. Gen. Genet. 255:45-53[Medline].

OHTSUKA, I., 1991  Genetic differentiation in wheat nuclear genomes in relation to compatibility with Aegilops squarrosa cytoplasm and application to phylogeny of polyploid wheat. J. Fac. Agric. Hokkaido Univ. 65(part 2):127-198.

OHTSUKA, I. and H. KIHARA, 1976  The effect of 1D chromosome in Aegilops squarrosa cytoplasm, that manifest in gametes and zygotes. Jpn. J. Genet. 51:433-434.

PALMER, J. D. and C. R. SHIELDS, 1984  Tripartite structure of the Brassica campestris mitochondrial genome. Nature 307:437-440.

QUETIER, F., B. LEJEUNE, S. DELORME and D. FALCONET, 1985 Molecular organization and expression of the mitochondrial genome of higher plants, pp. 413–420 in Molecular Form and Function of the Plant Genome, edited by L. VAN VLOTEN-DOTING, G. S. P. GROOT and T. C. HALL. Plenum Press, New York.

SHITARA, H., J. I. HAYASHI, S. TAKAHAMA, H. KANEDA, and H. YONEYAMA, 1998  Maternal inheritance of mouse mtDNA in interspecific hybrids: segregation of the leaked paternal mtDNA followed by the prevention of subsequent paternal leakage. Genetics 148:851-857[Abstract/Free Full Text].

SMALL, I. D., P. G. ISSAC, and C. J. LEAVER, 1987  Stoichiometric differences in DNA molecules containing the atpA gene suggest mechanisms for the generation of mitochondrial genome diversity in maize. EMBO J. 6:865-869[Medline].

SOLIGNAC, M., M. MONNEROT, and J. C. MOUNOLOU, 1983  Mitochondrial DNA heteroplasmy in Drosophila mauritiana. Proc. Natl. Acad. Sci. USA 80:6942-6946[Abstract/Free Full Text].

STEINBORN, R., P. SCHINOGL, V. ZAKHARTCHENKO, R. ACHMANN, and W. SCHERNTHANER et al., 2000  Mitochondrial DNA heteroplasmy in cloned cattle produced by fetal and adult cell cloning. Nat. Genet. 25:255-257[Medline].

TSUJI, S. and M. MURATA, 1976  Specific interaction between the D genome and the three alien cytoplasms in wheat. II. Seed inviability induced by the alien cytoplasms. Jpn. J. Genet. 51:327-336.

TSUKAMOTO, N., N. ASAKURA, N. HATTORI, K. TAKUMI, and N. MORI et al., 2000  Identification of paternal mitochondrial DNA sequences in the nucleus-cytoplasm hybrids of tetraploid and hexaploid wheat with D and D2 plasmons from Aegilops species. Curr. Genet. 38:208-217[Medline].

TSUNEWAKI, K., 1993  Genome-plasmon interactions in wheat. Jpn. J. Genet. 68:1-34.

TSUNEWAKI, K., 1996 Plasmon analysis as the counterpart of genome analysis, pp. 271–299 in Methods of Genome Analysis in Plants, edited by P. P. JAHUHAR. CRC Press, New York.

ZOUROS, E., K. R. FREEMANN, A. O. BALL, and G. H. POGSON, 1992  Direct evidence for extensive paternal mitochondrial DNA inheritance in the marine mussel Mytilus. Nature 359:412-414[Medline].

This article has been cited by other articles:

				M. E. Welch, M. Z. Darnell, and D. E. McCauley Variable Populations Within Variable Populations: Quantifying Mitochondrial Heteroplasmy in Natural Populations of the Gynodioecious Plant Silene vulgaris Genetics, October 1, 2006; 174(2): 829 - 837. [Abstract] [Full Text] [PDF]

				T. Stadler and L. F. Delph Ancient mitochondrial haplotypes and evidence for intragenic recombination in a gynodioecious plant PNAS, September 3, 2002; 99(18): 11730 - 11735. [Abstract] [Full Text] [PDF]