Detailed Mapping of the Species Cytoplasm-Specific (scs) Gene in Durum Wheat

Corresponding author: Shahryar F. Kianian, 470G Loftsgard Hall, North Dakota State University, Fargo, ND 58105., s.kianian{at}ndsu.nodak.edu (E-mail)

Communicating editor: J. A. BIRCHLER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The compatibility-inducing action of the scs^ti (species cytoplasm-specific gene derived from Triticum timopheevii) and Vi (vitality) genes can be observed when a durum (T. turgidum) nucleus is placed in T. longissimum cytoplasm. These two genes restore compatibility between an otherwise incompatible nucleus and cytoplasm. The objective of this study was to localize the scs^ti gene on a linkage map of chromosome 1A, which could eventually be used to clone the gene. The mapping population consisted of 110 F₂ individuals derived from crossing a Langdon-T. dicoccoides chromosome 1A substitution line with a euplasmic (normal cytoplasm) line homozygous for the scs^ti gene. Through a series of testcrosses the genotypes of the 110 individuals were determined: 22 had two copies, 59 had one copy, and 29 had no copy of the scs^ti gene. Data from RFLP, AFLP, and microsatellite analysis were used to create a linkage map. The flanking marker loci found for the scs^ti gene were Xbcd12 and Xbcd1449-1A.2 with distances of 2.3 and 0.6 cM, respectively. Nearly 10% of individuals in this population were double recombinant for a genetic interval of <3 cM. A blistering phenotype reminiscent of the phenotype observed in maize brittle-1 mutable was also evident in these individuals. The higher frequency of double recombination within this region and seed-blistering phenotype could be an indication of a transposable element(s) in this locus.

MOST cultivated wheats are polyploid in nature. Common bread wheat, Triticum aestivum L. (2n = 6x = 42), is hexaploid, consisting of three different genomes designated as AABBDD. Durum wheat, T. turgidum L. var. durum (2n = 4x = 28), is a tetraploid consisting of two genomes, AABB. Wild Triticum species have various ploidy levels. Examples include the tetraploids, T. dicoccoides (Körn. Ex Asch. & Graebner) (2n = 4x = 28, AABB) and T. timopheevii Zhuk. (2n = 4x = 28, AAGG), and the diploid T. longissimum (S. & M.) Bowden (2n = 2x = 14, S^lS^l). Wild wheat species are an invaluable source of novel genes for cultivated wheat, including genes for high grain protein content and pest resistance (CANTRELL and JOPPA 1991 ). However, not all wild species are compatible with cultivated wheat (MAAN 1975 ). Typically, chromosome asynapsis or hybrid sterility results from incompatible crosses among related species (KIHARA 1951 ; MAAN 1975 ). Genes affecting nuclear-cytoplasmic interactions are directly or indirectly involved (MAAN 1975 ).

In the past, restoration of fertility (Rf) genes was used to overcome hybrid sterility induced by incompatibility, but Rf genes do not work in all situations (MAAN 1992B ). In particular, Rf genes do not restore compatibility between T. longissimum cytoplasm and the nuclei of T. turgidum L. var. durum. The species cytoplasmic-specific (scs) gene and the vitality (Vi) gene make up a two-gene compatibility restoration system described by MAAN 1992C . In alloplasmic (alien cytoplasm) wheat, the scs gene restores partial compatibility between the nucleus and cytoplasm, producing plump F₁ seeds and male-sterile plants with normal vigor. The Vi gene also restores partial compatibility between the nucleus and cytoplasm, producing plump F₁ seeds and plants with below-normal vigor. However, if these plants survive to maturity, they are male fertile (MAAN 1992A ). Together, the two genes restore complete compatibility between a T. longissimum cytoplasm and a T. turgidum nucleus.

Plants other than wheat demonstrate evidence of nuclear-cytoplasmic incompatibility and compatibility. The most commonly observed phenotype is cytoplasmic male sterility (cms; HANSON 1991 ). It is believed that pollen-forming tissues have a lower threshold value for respiratory deficiencies than other plant tissues (LEON et al. 1998 ). Cytoplasmic male sterility has been reported in >150 different plant species, and in those examined, it was associated with the expression of a novel mitochondrial peptide (LEON et al. 1998 ). In Brassica, maize, and common bean, a cms gene encoding an extra peptide is cotranscribed with another mitochondrial gene such as atp6, orf25 (now known to be atp4), or atpa (DEWEY et al. 1986 ; CHASE 1994 ; SINGH et al. 1996 ; HEAZLEWOOD et al. 2003 ). In maize, there is a detoxification protein encoded by Rf2 that corrects cms (CUI et al. 1996 ). Nonchromosomal stripe (ncs) mutations may also arise due to certain nuclear-cytoplasmic genetic combinations and include infertility, leaf striping, loss of mitochondrial gene function, or severe growth impairment (NEWTON and COE 1986 ). In common bean, a novel polypeptide (pvs-orf239) accumulates only in reproductive tissue (ABAD et al. 1995 ). If the dominant nuclear factor, Fr, is added, the mitochondrial genome shifts the pvs-orf239 to substoichiometric levels within the genome, resulting in pollen fertility (JANSKA and MACKENZIE 1993 ). In Arabidopsis a nuclear gene, CHM, causes mitochondrial genomic rearrangements (LEON et al. 1998 ). These examples demonstrate the similarities, differences, and complexities of nuclear-cytoplasmic interactions in plants.

Molecular genetic research of nuclear-cytoplasmic interaction genes in wheat has been limited. ANDERSON and MAAN 1995 used molecular markers to determine the location of the scs^ti gene (species cytoplasm-specific gene derived from T. timopheevii). The mapping population consisted of the progeny from crossing two alloplasmic lines. Four markers were found linked to scs^ti and analysis of markers placed the scs^ti locus on the long arm of chromosome 1A near the centromere. The markers were found only on the centromeric side of the scs^ti gene. ASAKURA et al. 2000 used molecular markers to identify a segment introgressed from T. timopheevii into durum wheat carrying a gene (Ncc-tmp) for compatibility with Aegilops squarrosa (DD) cytoplasm. This gene, believed to be the same or similar to scs^ti, also located on chromosome 1A, was localized to a 9-cM segment near the centromere. Since alloplasmic populations were used in both studies for segregation analysis, selection against one of the genotypes (alloplasmic lines without the compatibility gene) could have affected the mapping results and scarcity of linked markers.

Studies of the scs genes may lead to a better understanding of wheat evolution and easier introgression of agronomically important genes from wild to cultivated wheat. Determination of the sequence would provide more immediate and far-reaching information about the gene itself, the sequence surrounding it, its function, and its role in the evolution of wheat. This information may, in turn, broaden our knowledge of nuclear-cytoplasmic interactions in plants. The first step toward cloning this gene is to produce a segregating population for the gene and identify flanking markers. Thus, the objective of this study was to determine a more precise location of scs^ti with respect to various DNA-based markers in a euplasmic mapping population. This mapping population could also provide the means for saturation mapping of the region surrounding this locus. The long-term goal of this project is to employ these markers to enable cloning of the scs^ti gene.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Population development:
A segregating population of 129 euplasmic individuals was developed to find markers associated with the scs^ti gene (Fig 1). This F₂ population was generated by crossing an intervarietal chromosomal substitution line, Langdon-T. diccocoides 1A (abbreviated LDN[Dic1A]), developed by JOPPA 1993 with a euplasmic line homozygous for the scs^ti gene developed by S. S. Maan (Fig 1).

View larger version (13K): In this window In a new window Download PPT slide   Figure 1. The mapping population was developed by crossing a euplasmic line homozygous for the scsti gene with a Langdon-T. dicoccoides intervarietal chromosome substitution line 1A [LDN (Dic 1A)] developed by JOPPA 1993 . The genotype of each F2 individual was determined by testcrossing to a hemizygous alloplasmic line and identifying the ratio of plump to shriveled seed. Seeds with (lo) - - genotype would be shriveled and nonviable in phenotype.

The F₂ individuals were planted in the greenhouse in the fall of 1998. The greenhouse conditions were optimized for the growth and reduction of environmental stress to the durum wheat plants. The plants were first exposed to a 24-hr light cycle until heading, at which point the light regime was switched to a 16-hr light cycle. Parental and check durum lines, such as prevalent cultivars "Ben" and "Maier," were also planted as controls to observe environmental effects.

One head of each individual F₂ plant was self-pollinated and at least two alternate heads were used for testcrossing to determine genotype (Fig 1). Each male head for testcrossing was used to pollinate at least one female head. The seed produced from each male head was kept separate. The testcross line was alloplasmic in that it had a T. longissimum cytoplasm and a T. turgidum nucleus [(lo) durum]. The testcross line also was hemizygous for the scs^ti gene [(lo) scs^ti -]. Resulting ratios of plump to shriveled seed were used to determine the genotype of the male parents (Fig 1). All plump seed indicated two copies of the scs^ti gene in the F₂ individual. Three plump seeds to one shriveled seed indicated one copy of the scs^ti gene. One plump to one shriveled seed indicated no copy of the scs^ti gene. Various formulas from LIU 1998 were used to calculate the number of seeds required to determine a 3:1, 1:1, or 1:0 ratio, which were 38, 38, and 8, respectively, for a confidence level of 90%. For individuals whose genotype could not be determined at a 90% confidence level, testcross progeny were screened. By planting eight of the testcross progeny and fertilizing them with (d)- - pollen, the F₂ genotype was determined (Table 1).

Genotypic analysis:
Tissue was collected from the 129 F₂ individuals before jointing and flash frozen in liquid nitrogen. DNA was extracted according to the protocol of ANDERSON and MAAN 1995 and quantified with a fluorometer. Membranes for restriction fragment length polymorphism (RFLP) analysis were generated and hybridized according to the protocol of OTTO et al. 2002 . Clones were selected on the basis of their location on the long arm of chromosome 1A. Clones selected included those from barley cDNA libraries, bcd (M. E. Sorrells) and abc (A. Kleinhofs); barley genomic libraries, abg (A. Kleinhofs) and mwg (A. Graner); an oat cDNA library, cdo (M. E. Sorrells); a rice cDNA library, rz (M. E. Sorrells); and wheat genomic libraries, ksu (B. S. Gill), fbb (P. Leory), fba (P. Leory), psr (M. D. Gale), and wg (M. E. Sorrells).

Microsatellite primers were chosen for their location near the centromere on chromosome 1A (RODER et al. 1998 ). Amplification was done using the PCR protocol specified by RODER et al. 1998 . The products were electrophoresed on a denaturing polyacrylamide sequencing gel [6% Long Ranger (BioWhittaker Molecular Applications, Rockland, ME), 1x TBE buffer, and 7 M urea] for 2 hr at 70 W, allowing the 100-bp standard to run near the bottom of the gel. The DNA was visualized by silver staining with the Promega (Madison, WI) Silver Sequence DNA sequencing system kit.

The GIBCO BRL (Gaithersburg, MD) kit for amplified fragment length polymorphisms (AFLPs) was used with several modifications. First, 350 ng instead of 250 ng of DNA was digested for 2 hr at 37° followed by inactivation of the enzyme for 15 min at 70°. The ligation step was done as described, but the product was diluted only 1:5 with TE buffer. The preamplification step was done as directed with 2.5 µl of diluted ligated DNA in a 25.5-µl reaction. The preamplification product was diluted 1:25 with TE buffer. The selective amplification used 2.5 µl of the diluted preamplified product, 1 µl 10x PCR buffer, 0.3 µl 50 mM MgCl₂, 0.09 µl selected primer containing EcoRI adapter sequence, 2.25 µl selected primer containing MseI adapter sequence, 3.76 µl water, and 0.5 units platinum Taq polymerase (GIBCO BRL). The results were visualized by silver staining with the Promega Silver Sequence DNA sequencing system kit.

Linkage:
A linkage map containing the scs^ti gene was generated using Mapmaker 3.0b (LANDER et al. 1987 ). The "sequence" command was used to store the working sequence with {} to indicate an unknown order. Markers were grouped using the "group" command with a LOD = 4.0. Of the markers grouped with scs^ti, six were chosen to form a skeletal map on the basis of their previously identified locations. The "compare" command was used to order the six chosen markers at a LOD >3.0. The remaining markers grouped with scs^ti were placed into regions on the skeletal map using the "try" command if LOD was >2. The markers in the same region as scs^ti were then ordered and map distances determined using the Kosambi mapping function. Graphic analysis of all individuals in this population was performed using the method of OTTO et al. 2002 to allow visual examination of double recombinants within a given segment. Robustness of the linkage map was tested with the "ripple" command using a window of 5 and a LOD of 3.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Population development and genotypic determination:
The T. turgidum nucleus is incompatible with the T. longissimum cytoplasm, producing nonviable progeny. This incompatibility is improved by scs^ti. Alloplasmic lines of T. turgidum with T. longissimum cytoplasm [(lo) durum] that have the scs^ti gene have improved embryo-endosperm compatibility but are male sterile (Fig 2). The Vi gene, located on the short arm of chromosome 1B (1BS), in conjunction with the scs^ti gene restores male fertility and seed viability (Fig 2). In the absence of the Vi gene, alloplasmic plants that are hemizygous for the scs^ti gene have normal vigor, are male sterile, and produce a 1:1 ratio of viable to aborted seed when crossed to normal durum wheat (Fig 2). In the absence of the scs^ti gene, alloplasmic plants with the Vi gene are weak, but fertile (Fig 2). In the absence of both scs^ti and Vi genes, the T. turgidum nucleus is incompatible with the T. longissimum cytoplasm. Therefore, addition of these two nuclear genes overcomes a number of deleterious phenotypes.

View larger version (58K): In this window In a new window Download PPT slide   Figure 2. T. turgidum (2n = 4x = 28; AABB) nucleus is incompatible with T. longissimum (2n = 2x = 14; S1S1) cytoplasm, producing nonviable progeny. Two nuclear genes, scsti and Vi, ameliorate the deleterious effects of this interaction as shown. Alloplasmic plants with Vi alone have greatly reduced vigor (A, two plants on the right; and B) as compared to those plants with scsti and/or Vi having near-normal vigor (A, two plants on the left). Alloplasmic plants with both scsti and Vi grow normally and are fertile (C, spike with extruded anthers) as opposed to male-sterile plants with only scsti (D, spike without anthers). Alloplasmic plants hemizygous for scsti produce shriveled nonviable seeds without scsti (E, bottom row), and plump seeds with scsti (E, top row) that result in male-sterile plants. Blistered seeds having a variegated phenotype were observed in a few individuals (E, middle row).

The initial F₂ euplasmic population segregating for the scs^ti gene consisted of 129 individuals. Testcrosses allowed genotypes to be determined for 50 of the 129 individuals at a 90% level of confidence (Table 1). After planting eight plump seeds from the testcross progeny and crossing the plants with euplasmic durum ([d]- -), the genotypes for 60 additional individuals were determined at the 90% level of confidence (Table 1). The remaining 19 individuals could not be assigned a genotype due to a lack of seed produced from the testcrosses. Twenty-nine individuals were obtained with (d)- -, 59 with (d)scs^ti -, and 22 with (d)scs^ti scs^ti. The F₂ population fit the expected segregation ratio for scs^ti of 1:2:1 with a chi-squared value of 1.47. This evidence supports the idea of a single gene that follows Mendelian inheritance.

Genotypic analysis:
Microsatellites, RFLPs, and AFLPs were the three types of markers used to screen the parents and score the initial population. Of the five microsatellites screened, two were polymorphic, producing five polymorphic bands. Of the 47 RFLP clones screened, 19 were polymorphic, producing 48 polymorphic bands. Of the 64 AFLP combinations screened, the 8 combinations that produced at least three polymorphic bands each were used to screen a subset (107 individuals) of the initial population. In total the 8 AFLP combinations produced 40 polymorphic bands.

Linkage:
All scores from the markers were entered into Mapmaker 3.0b. Thirty-seven marker loci were grouped with the scs^ti genotype. Thirty-four of the 37 markers fit the expected codominant 1:2:1 ratio or the dominant 3:1 ratio. Marker loci were successfully ordered, resulting in a skeletal map with six markers spaced over chromosome 1A for a total length of 93.5 cM (LOD > 4.0, Fig 3). The remaining 30 markers were placed in regions using the "try" command (Fig 3).

View larger version (20K): In this window In a new window Download PPT slide   Figure 3. Molecular marker map of chromosome 1A in the euplasmic population segregating for scsti. Loci were either used to construct the framework skeletal map or placed at intervals. Also presented is the detailed linkage map of the region delineated by Xbcd1072 and Xbcd22 containing the scsti gene (LOD > 2.5) generated both before (right) and after (left) removing the scsti genotypes of the 11 apparently double-recombinant individuals. Distances are in Kosambi centimorgans. Arrow indicates the location of the centromere. *, removal of these markers increases the LOD score to 3.0.

The region containing the scs^ti gene was mapped in more detail and loci flanking the scs^ti gene, Xbcd12 and Xbcd1449-1A.2, were identified (Fig 3). The distance between these flanking loci was 1.4 cM before placing scs^ti and 14.0 cM after placement of the scs^ti gene, indicating that this locus was showing more double recombinants than expected (Fig 4). Graphic genotyping identified 11 individuals that appeared to have crossovers flanking both sides of the scs^ti gene. Removal of the scs^ti genotypes of these 11 individuals resulted in a map with distances of 2.3 and 0.6 cM from the scs^ti gene for the loci, Xbcd12 and Xbcd1449-1A.2, respectively (Fig 3). A comparison of the detailed maps before and after removing these 11 individual genotypes demonstrates the degree of difference (Fig 3).

View larger version (32K): In this window In a new window Download PPT slide   Figure 4. Distribution of recombination on chromosome 1A segment surrounding scsti calculated using the map units from the skeletal linkage map. The number of recombination events between markers on the skeletal linkage map is represented by the open bars. The solid bars show the number of recombination events between markers when including the double-recombination individuals. The major significant difference is associated with the scsti locus.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The evidence presented points toward a phenomenon in which a large number of apparent double crossovers are observed. This phenomenon may be observed as the result of a "hotspot" of recombination, gene conversion, somatic crossing over, different frequencies of chromosome transmission, gamete selection, or change in the allelic state of scs^ti possibly induced by transposable elements. The chromosome pairing was cytologically checked on a sib of the F₁ individual and appeared normal.

To test for a hotspot of recombination near scs^ti, the number of recombination events per map unit between each marker was graphed (Fig 4). The number of recombination events between Xbcd1072 and Xbcd22 more than doubles when the scs^ti genotypic data are included in the analysis. This increase demonstrates that a change in the allelic state of scs^ti is responsible for the difference. The recombination across the chromosome based on the markers used in this study was graphed (Fig 5) and compared to the published recombination distribution of wheat chromosome 1A (GILL et al. 1996 ). The trend was similar and also suggested that a hotspot of recombination around scs^ti did not exist (Fig 5).

View larger version (19K): In this window In a new window Download PPT slide   Figure 5. Recombination is unevenly distributed across the chromosome. Solid bars represent recombination found by GILL et al. 1996 on wheat chromosome 1A. The open bars represent the number of recombination events found along chromosome 1A on the basis of the euplasmic F2 population segregating for the scsti locus. Markers at the end of this chromosome were not mapped in this study.

By looking at the recombination distribution, it was observed that the number of recombinations around scs^ti seemed high. We expected 1 in 2500 (0.0004) double recombinants between the two codominant markers flanking scs^ti, but observed 11 in 110 (0.1000). Gene conversion was also considered, but we would not expect this event to occur with this high a frequency and cannot explain all instances, since double recombination occurred in five individuals showing homozygosity in the flanking markers. Somatic crossing over is one mechanism by which these double recombinants could arise. A few plants that had chimeric tillers (two tillers each producing different ratios of plump to shriveled seeds) were observed in the F₂ population. The observed frequency of chimeric plants is fairly low: 2 of the 110 individuals in the mapping population. The observed frequency of double crossovers for scs^ti is too high to be explained by this phenomenon.

Different frequencies of chromosomal transmission or gamete selection were also contemplated. For these events to explain the amount of apparent double recombinants, one should observe an overall increase in one genotype. When observing the 11 aberrant individuals, the trend is toward having a scs^ti locus vs. not having one. Looking at the overall number of the genotypes in each class, particularly scs^tiscs^ti vs. - -(null), the trend is toward having more nulls. These two trends do not match, which leads to the conclusion that the scs^ti locus is not causing gametic selection or different chromosomal transmission frequencies.

The selfed seed from these 11 F₂ individuals, their testcross seed, and seed from the genotypic checks were also examined for abnormalities. Some seed seemed to be partially plump and partially shriveled or "blistered" in phenotype, but this shriveling was primarily limited to the nongerminating end of the seed (Fig 2). The environment does not appear to be an instigator of the blistering phenotype, because not all the seeds from a single cross were blistered nor were the seeds from the plants on either side of the crosses that exhibited the blistering. A similar blistering phenotype has been observed in maize with regard to a maize brittle-1 mutable (bt-1m) gene (SULLIVAN et al. 1991 ). The normal gene, bt-1, produces normal-appearing seed; the mutant locus produces shrunken kernels. The mutable gene, bt-1m, produces the blistering phenotype, which results from the insertion of a defective suppressor-mutator (dSpm) transposable element. The similarity of a blistering phenotype in the seed suggests similar circumstances, the presence of active transposable elements causing sectors of different genotypes in developing seeds.

Evidence has been found to support the presence of active retrotransposons and transposons in wheat. VICIENT et al. 2001 compared the sequences of known retrotransposons with various expressed sequence tag (EST) databases. The wheat EST databases had 1.26% of their ESTs matching known retrotransposon sequences. LISCH et al. 2001 identified a wheat cDNA sequence that was 96% similar to the mudrA sequence in maize. MudrA encodes a protein similar to bacterial transposases and regulates the maize mutator. Last, both KIHARA 1951 and MAAN 1991 presented data on alloplasmic lines of wheat suggesting the presence of active, mobile elements.

If a transposition event involving scs^ti had occurred, one may expect to observe a change in the allelic state of scs^ti. GEHLHAR et al. 2001 reported the allelic relationship of scs^d, derived from T. durum, to scs^ti. The scs^d allele did confer partial compatibility between the nucleus and cytoplasm, resulting in plump seed. The plants from the plump seed lacked vigor and grew to a height of 6 inches as compared to nearly 5 feet for the wild-type genotype. The scs^d allele originated from (lo)scs^ti- when it was transferred to durum cultivars for evaluation of yield (S. S. MAAN, personal communication). In alloplasmic lines presumed to have the scs^ti allele plants grown from plump seed grew half the height of normal durum and typically did not tiller (our unpublished results). These differences in phenotype may possibly be explained by the changes created from a footprint left after a transposition event.

Previous attempts to flank scs^ti with molecular markers by using alloplasmic populations have had limited success. In an alloplasmic situation, seeds with scs^ti produce plump seeds that grow to mature plants, while those without scs^ti are shriveled and inviable. Thus, selection against one of the genotypes could have affected the mapping results and scarcity of linked markers. In this study, flanking marker loci Xbcd12 and Xbcd1449-1A.2 were found for the scs^ti gene with distances of 2.3 and 0.6 cM, respectively. These marker loci will form the basic framework for more detailed studies of this important gene.

Most cultivated wheats are polyploid in nature. These allopolyploids arose from stepwise interspecific hybridizations involving diploid ancestors. An important factor in the successful formation of a hybrid combination is the interaction between the nucleus and cytoplasmic genomes. Triticum species differ with regard to compatibility with alien cytoplasms and interspecific nuclear-cytoplasmic interactions produce a variety of phenotypes, including maternally inherited male sterility, delayed maturity, and reduced plant vigor in alloplasmic wheat lines (MAAN 1991 ). As is evident, nuclear-cytoplasmic interaction and genes involved were important in the formation of polyploid wheat and speciation of different species. Therefore, molecular characterization of genes involved in nucleo-cytoplasmic compatibility could lead to a better understanding of the interplay between the various genomes in a cell.

	ACKNOWLEDGMENTS

We thank Justin B. Hegstad, Kay M. Carlson, and the rest of the Wheat Germplasm Enhancement group for their assistance in making this work possible. We also thank P. McClean and J. Faris and the anonymous reviewers for their suggestions in improving this manuscript and R. L. Phillips for helpful discussion and suggestions on the bt-1m blistering phenotype of maize. This material is based upon the work supported by the United States Department of Agriculture-National Research Initiative grant no. 99-35311-8252, National Science Foundation-Plant Genome Research Program contract agreement no. DBI-9975989, and the North Dakota Wheat Commission to S.F.K.

Manuscript received July 2, 2003; Accepted for publication August 28, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ABAD, A. R., B. J. MEHRTENS, and S. A. MACKENZIE, 1995  Specific expression in reproductive tissues and the fate of a mitochondrial sterility-associated protein in cytoplasmic male-sterile bean. Plant Cell 7:271-285.[Abstract]

ANDERSON, J. A. and S. S. MAAN, 1995  Interspecific nuclear-cytoplasmic compatibility controlled by genes on group 1 chromosomes in durum wheat. Genome 38:803-808.[Medline]

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]

CANTRELL, R. G. and L. R. JOPPA, 1991  Genetic analysis of quantitative traits in wild emmer (Triticum turgidum L. var. dicoccoides). Crop Sci. 31:645-649.[Abstract/Free Full Text]

CHASE, C. D., 1994  Expression of CMS-unique and flanking mitochondrial DNA sequences in Phaseolus vulgaris L. Curr. Genet. 25:245-251.[Medline]

CUI, X., R. P. WISE, and P. S. SCHNABLE, 1996  The rf2 nuclear restorer gene of male-sterile t-cytoplasm maize. Science 272:1334-1336.[Abstract]

DEWEY, R. E., C. S. LEVINGS, III, and D. H. TIMOTHY, 1986  Novel recombinations in the maize mitochondrial genome produce a unique transcriptional unit in the Texas male-sterile cytoplasm. Cell 44:439-449.[Medline]

GEHLHAR, S. B., S. F. KIANIAN, S. S. MAAN and K. J. SIMONS, 2001 An allelism test for species cytoplasmic specific (scs) genes, p. 47 in North Dakota Academy of Science 93rd Annual Meeting, edited by J. JACKSON. North Dakota Academy of Science, Bismarck, ND.

GILL, K. S., B. S. GILL, T. R. ENDO, and T. TAYLOR, 1996  Identification and high-density mapping of gene-rich regions in chromosome group 1 of wheat. Genetics 144:1883-1891.[Abstract]

HANSON, M. R., 1991  Plant mitochondrial mutations and male sterility. Annu. Rev. Genet. 25:461-486.[Medline]

HEAZLEWOOD, J. L., J. WHELAN, and A. H. MILLER, 2003  The products of the mitochondrial orf25 and orfB genes are FO components in the plant F1FO ATP synthase. FEBS Lett. 537:201-205.

JANSKA, H. and S. A. MACKENZIE, 1993  Unusual mitochondrial genome organization in cytoplasmic male sterile common bean and the nature of cytoplasmic reversion to fertility. Genetics 135:869-879.[Abstract]

JOPPA, L. R., 1993  Chromosome engineering in tetraploid wheat. Crop Sci. 33:908-913.[Abstract/Free Full Text]

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

LANDER, E. S., P. GREEN, J. ABRAHAMSON, A. BARLOW, and M. J. DALY et al., 1987  MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1:174-181.[Medline]

LEON, P., A. ARROYO, and S. A. MACKENZIE, 1998  Nuclear control of plastid and mitochondrial development in higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 49:453-480.

LISCH, D. R., M. FREELING, R. J. LANGHAM, and M. Y. CHOY, 2001  Mutator transposase is widespread in the grasses. Plant Physiol. 125:1293-1303.[Abstract/Free Full Text]

LIU, B. H., 1998 Statistical Genomics, Linkage, Mapping and QTL Analysis. CRC Press, Boca Raton, FL.

MAAN, S. S., 1975 Cytoplasmic variability and speciation in Triticinae, pp. 255–281 in Prairie: A Multiple View, edited by M. K. WALI. University of North Dakota Press, Grand Forks, ND.

MAAN, S. S., 1991 Nucleo-cytoplasmic genetics of wheat, pp. 175–194 in International Symposium on Nuclear and Organeller Genetics of Wheat, edited by T. SASAKUMA and T. KINOSHITA. Hokkaido University, Sapporo, Japan.

MAAN, S. S., 1992a  A gene for embryo-endosperm compatibility and seed viability in alloplasmic Triticum turgidum.. Genome 35:772-779.

MAAN, S. S., 1992b  Genetic analyses of male fertility restoration in wheat: IV. Fertile line without major Rf genes. Crop Sci. 32:24-28.

MAAN, S. S., 1992c  The scs and Vi genes correct a syndrome of cytoplasmic effects in alloplasmic durum wheat. Genome 35:780-787.

NEWTON, K. J. and E. H. COE, JR., 1986  Mitochondrial DNA changes in abnormal growth (nonchromosomal stripe) mutants of maize. Proc. Natl. Acad. Sci. USA 83:7363-7366.[Abstract/Free Full Text]

OTTO, C. D., S. F. KIANIAN, E. M. ELIAS, R. W. STACK, and L. R. JOPPA, 2002  Genetic dissection of a major Fusarium head blight QTL in tetraploid wheat. Plant Mol. Biol. 48:625-632.[Medline]

RÖDER, M. S., V. KORZUN, K. WENDEHAKE, J. PLASCHKE, and M. H. TIXIER et al., 1998  A microsatellite map of wheat. Genetics 149:2007-2023.[Abstract/Free Full Text]

SINGH, M., N. HAMEL, R. MENASSA, X.-Q. LI, and B. YOUNG et al., 1996  Nuclear genes associated with a single brassica CMS restorer locus influence transcripts of three different mitochondrial gene regions. Genetics 143:505-516.[Abstract]

SULLIVAN, T. D., L. I. STELOW, C. A. ILLINGWORTH, R. L. PHILLIPS, and O. E. NELSON, JR., 1991  Analysis of maize brittle-1 alleles and a defective suppressor-mutator-induced mutable allele. Plant Cell 3:1337-1348.[Abstract/Free Full Text]

VICIENT, C. M., M. J. JÄÄSKELÄINEN, R. KALENDAR, and A. H. SCHULMAN, 2001  Active retrotransposons are a common feature of grass genomes. Plant Physiol. 125:1283-1292.[Abstract/Free Full Text]

This article has been cited by other articles:

				S. B. Gehlhar, K. J. Simons, E. M. Elias, S. S. Maan, and S. F. Kianian A Species Cytoplasm Specific Gene in Euplasmic Durum Wheat Does Not Alter Field Performance Crop Sci., August 1, 2005; 45(5): 1704 - 1707. [Abstract] [Full Text] [PDF]

				S. B. Gehlhar, K. J. Simons, S. S. Maan, and S. F. Kianian Genetic Analysis of the Species Cytoplasm Specific Gene (scsd) Derived from Durum Wheat J. Hered., July 1, 2005; 96(4): 404 - 409. [Abstract] [Full Text] [PDF]

				K. G. Hossain, O. Riera-Lizarazu, V. Kalavacharla, M. I. Vales, S. S. Maan, and S. F. Kianian Radiation Hybrid Mapping of the Species Cytoplasm-Specific (scsae) Gene in Wheat Genetics, September 1, 2004; 168(1): 415 - 423. [Abstract] [Full Text] [PDF]