Mitochondrial DNA-RFLP Analysis Distinguishes New CMS Sources in Pearl Millet (Pennisetum glaucum (L.) R. Br.)

Restriction fragment length polymorphism (RFLP) of mitochondrial (mt) DNA provides a rapid and effective method to assess heterogeneity among male sterile cytoplasms. Six isonuclear A-lines (81 A₁, with Tift 23A₁, cytoplasm, ICMA 88001 (= 81A_v) with Violaceum cytoplasm, 81A (=81A₄) with monodli = violaceum cytoplasm, Pb 310A₂ and Pb 311A₂ with A₂ cytoplasm from L 66A, and Pb 406A₃ with A₃ cytoplasm from L 67A), nine cytoplasmic male-sterility sources from Large-Seeded Genepool (LSGP 6, LSGP 14, LSGP 17, LSGP 22, LSGP 28, LSGP 36, LSGP 43, LSGP 55 and LSGP 66) and two each from Early Genepool (EGP 33 and EGP 15) and Population Varieties (PV 1 and PV 2) were characterized for variation in their mitochondrial genomes following Southern blot hybridizations using homologous (pearl millet 13.6 kb, 10.9 kb, 9.7 kb and 4.7 kb clones) and heterologous (maize atp6 and coxl clones) mitochondrial DNA (mtDNA) probes. Following cluster analysis based on similarity indices for the RFLP banding patterns observed, we identified seven cytoplasmic groups within LSGP. Two (LSGP 43 and LSGP 66) of these were quite distinct from each other as well as from other cytoplasms. This clearly indicates that besides serving as a source of diversity for agronomic and adaptation traits, broad-based gene pools can also provide diverse sources of cytoplasmic male sterility. These new CMS sources were also compared with standard CMS systems and cytoplasm-specific restriction fragments were identified.     

Mitochondrial DNA Polymorphisms Revealed by RAPD Assays Distinguish the Male-sterile and Male-fertile Cytoplasms in Pearl Millet

Diversification of cytoplamic male sterile (CMS) sources is of considerable significance in pearl millet, considering that almost all the commercially cultivated hybrids, particularly in India, are based on a single CMS source, A₁. We analyzed the mitochondrial DNA (mtDNA) polymorphisms among five pearl millet CMS sources (A₁ to A₅) and a male-fertile maintainer (B) line, all in an isonuclear background. Analysis using 21 random primers led to identification of polymorphic bands specific to the A₁, A₂, A3 and A₅ CMS lines. Two RAPD primers, OP-G12 and OP-G19, in combination were able to distinguish all the male-sterile and male-fertile cytoplasms. Highly effective for this purpose also were four RAPD markers, OP-B7, OP-D8, OP-F10 and OP-G12. Cluster analysis, followed by bootstrap analysis, of the mtDNA dataset revealed two distinct clusters: cluster-I comprising the A₁, A₂, A₃ CMS lines and the male-fertile line, and cluster-II comprising the A₄ and A₅ CMS lines.     

Downy mildew incidence of pearl millet hybrids with different male-sterility inducing cytoplasms

The use of different sources of cytoplasmic male sterility (CMS) in hybrid seed production of pearl millet [Pennisetum glaucum (L.) R. Br.] is advocated to avoid possible disease epidemics occurring due to cytoplasmic uniformity. The effects of commercially unexploited, but potentially exploitable, sources of CMS, like A₂, A₃ and A₄, on downy mildew [Sclerospora graminicola (Sacc.) Schroet] incidence were studied by using the disease incidence of isonuclear hybrids with male-sterile and fertile cytoplasm. The mean downy mildew incidence of hybrids carrying different male-sterile cytoplasm was similar to that of hybrids retaining the fertile cytoplasm. The cytoplasm accounted for only 0.6% of the total variation and its effect was non-significant; pollinators could explain most of the variation in determining the disease incidence of hybrids. This suggested that these male-sterile cytoplasms are not linked to downy mildew susceptibility and thus can be exploited commercially to broaden the cytoplasmic base of the male-sterile lines and, ultimately, of hybrids.     

Assessment of cytoplasmic differences of near-isonuclear male-sterile lines in pearl millet

Four near-isonuclear polycytoplasmic versions of 81A and two of Pb 402A male-sterile lines of pearl millet (Pennisetum typhoides) were used in factorial matings with five inbred male testers in different combinations in three sets. The cytoplasmic differences were studied for several agronomic traits using mean values and general combining effects (gca) of male-sterile lines, and specific combining ability effects of hybrids. The fertility/ sterility behaviour of different male-sterile lines in crosses with common male parents was also studied. Significant differences among near-isonuclear polycytoplasmic lines were observed in mean values for a few traits such as plant height, leaf length and peduncle length, but the differences for combining ability were more pronounced. The A₃ cytoplasm was a better general combiner than the A₂ cytoplasm for grain yield and both A₂ and A₃ cytoplasms were better general combiners for leaf length and peduncle length. In addition, superiority of A₃ cytoplasm for gca was observed for plant height and ear characters over the A₂ cytoplasm in set II. A differential behaviour of cytoplasms, both in combination with a common pollinator and across pollinators, was observed for several traits. The results provide evidence for the distinctiveness of different cytoplasmic sources in pearl millet and for the influence of cytoplasmic factors on the phenotypic expression of nuclear genes. A diversification of male sterility sources in the breeding of pearl millet hybrids is suggested.     

Identification of mitochondrial DNA sequence variation and development of single nucleotide polymorphic markers for CMS-D8 in cotton

Cytoplasmic male sterility (CMS), which is a maternally inherited trait and controlled by novel chimeric genes in the mitochondrial genome, plays a pivotal role in the production of hybrid seed. In cotton, no PCR-based marker has been developed to discriminate CMS-D8 (from Gossypium trilobum) from its normal Upland cotton (AD1, Gossypium hirsutum) cytoplasm. The objective of the current study was to develop PCR-based single nucleotide polymorphic (SNP) markers from mitochondrial genes for the CMS-D8 cytoplasm. DNA sequence variation in mitochondrial genes involved in the oxidative phosphorylation chain including ATP synthase subunit 1, 4, 6, 8 and 9, and cytochrome c oxidase 1, 2 and 3 subunits were identified by comparing CMS-D8, its isogenic maintainer and restorer lines on the same nuclear genetic background. An allelic specific PCR (AS-PCR) was utilized for SNP typing by incorporating artificial mismatched nucleotides into the third or fourth base from the 3′ terminus in both the specific and nonspecific primers. The result indicated that the method modifying allele-specific primers was successful in obtaining eight SNP markers out of eight SNPs using eight primer pairs to discriminate two alleles between AD1 and CMS-D8 cytoplasms. Two of the SNPs for atp1 and cox1 could also be used in combination to discriminate between CMS-D8 and CMS-D2 cytoplasms. Additionally, a PCR-based marker from a nine nucleotide insertion–deletion (InDel) sequence (AATTGTTTT) at the 59–67 bp positions from the start codon of atp6, which is present in the CMS and restorer lines with the D8 cytoplasm but absent in the maintainer line with the AD1 cytoplasm, was also developed. A SNP marker for two nucleotide substitutions (AA in AD1 cytoplasm to CT in CMS-D8 cytoplasm) in the intron (1,506 bp) of cox2 gene was also developed. These PCR-based SNP markers should be useful in discriminating CMS-D8 and AD1 cytoplasms, or those with CMS-D2 cytoplasm as a rapid, simple, inexpensive, and reliable genotyping tool to assist hybrid cotton breeding.     

Molecular diversity of male sterility inducing and male-fertile cytoplasms in the genus Helianthus

The organisation of mtDNA was investigated for 28 sources of cytoplasmic male sterility (CMS) and a fertile line (normal cytoplasm) of Helianthus annuus by Southern hybridisation. In addition to nine known mitochondrial genes (atp6, atp9, cob, coxI, coxII, coxIII, 18S, 5S and nd5) three probes for the open reading frames in the rearranged area of PET1, orfH522, orfH708 and orfH873, were used. Genetic similarities of the investigat-ed cytoplasms varied between 0.3 and 1. Cluster analyses using the UPGMA method allowed the distinction of ten mitochondrial (mt) types between the 29 investigated cytoplasms. Most mitochondrial types comprise two or more CMS sources, which could not be further separated, like the PET1-like CMS sources (with the exception of ANO1 and PRR1), or ANN1/ANN2/ANN3, ANN4/ ANN5, ARG3/RIG1, BOL1/EXI1/PEF1/PEP1 and GIG1/ PET2. ANL1, ANL2 and the fertile cytoplasms are also regarded as one mitochondrial type. Unique banding patterns were only observed for ANT1 (atp6), MAX1 (atp6, orfH522 and orfH708) and PRR1 (coxII). However, four of the mitochondrial types showed unique hybridisation signals: ANN4/ANN5 had characteristic bands for atp6 and orfH708, PEF1/PEP1/EXI1/BOL1 for atp6 and coxII, and PET2/GIG1 for atp9. The PET1-like cytoplasms all shared the same patterns for orfH522, orfH708 and cob (except ANO1). It could be demonstrated that CMS sources, like, e.g., PET2 and PEF1, are different from PET1 in mtDNA organisation and the CMS mechanism. Therefore, these CMS sources represent interesting candidates for the development of new hybrid breeding systems based on new CMS mechanisms. Received: 20 April 2001 / Accepted: 3 August 2001     

Influence of A1 cytoplasmic substitution on the downy-mildew incidence of pearl millet

Large-scale cultivation of pearl millet [Pennisetum glaucum (L.) R. Br. F₁ hybrids in India has led to increased incidence of downy-mildew (Sclerospora graminicola). There is concern that the A₁ male-sterile cytoplasm used in all the hybrids released so far is responsible for this increase. The influence of A₁ malesterile cytoplasm on downy-mildew incidence in pearl millet was studied by comparing the disease reaction of 40 pairs of F₁ hybrids, each pair carrying respectively a₁ male-sterile and normal B cytoplasm. Mean downy-mildew incidence was similar in the hybrids carrying either A₁ male-sterile or B cytoplasm. The general combining ability of lines with and without A₁ cytoplasm was found to be similar for downy-mildew incidence. These results indicated that in pearl millet A₁ cytoplasm is not associated with increased downymildew incidence. The possible danger of using only one source of cytoplasm has been briefly discussed.     

Nucleo-cytoplasmic interactions affect RNA editing of cox2, atp6 and atp9 in alloplasmic male-sterile rice (Oryza sativa L.) lines

RNA editing plays an important role in the regulation of mitochondrial gene expression in flowering plants. In this study, we examined RNA editing of the mitochondrial genes cox2, atp6 and atp9 in five isonuclear alloplasmic male-sterile lines (IAMSLs) of rice to investigate whether different cytoplasmic types affect RNA editing. Although many editing sites were conserved among the three genes, we found that the editing efficiency of certain sites was significantly different between different IAMSLs or between IAMSLs and their corresponding cytoplasmic donor CMS lines. Furthermore, several editing sites were found to be either present or absent in certain IAMSLs and their corresponding CMS lines. These results indicate that nuclear loci, as well as unknown editing factors within the mitochondria of different cytoplasmic types, may be involved in RNA editing, and they suggest that RNA editing in plant mitochondria is affected by nucleo-cytoplasmic interactions.     

Effect of A1 cytoplasm on the combining ability for smut severity in pearl millet

Among the various available sources of male-sterile cytoplasm in pearl millet [Pennisetum glaucum (L.) R.Br.], the A₁ source has been exploited the most for the breeding of commercial F₁ hybrids. The effect of this source on the combining ability (CA) for smut severity was studied since it is the CA that determines the performance of hybrids. The effect was estimated by comparing the CA estimates of 5 pairs of lines and 35 pairs of crosses with and without A₁ cytoplasm. The cytoplasm showed either a significantly desirable or at least no adverse effect on the CA of 4 out of the 5 line pairs and 56 out of 70 pairs of comparison of crosses in two environments. The differential effect of cytoplasm in some pairs might be due to its interaction with nuclear genes. These results further substantiated that the A₁ cytoplasm is not linked with increased smut severity in pearl millet hybrids.     

Diversity among male-sterility-inducing and male-fertile cytoplasms of onion

Hybrid-onion (Allium cepa) seed is produced using systems of cytoplasmic-genic male sterility (CMS). Two different sources of CMS (S and T cytoplasms) have been genetically characterized. Testcrosses of N-cytoplasmic maintaining and restoring genotypes to S and T cytoplasmic lines demonstrated that different alleles, or loci, restore male fertility for these two male-sterile cytoplasms. Other sources of CMS have been used or reported in Europe, Japan and India, and their relationships to S and T cytoplasms are not clear. Restriction fragment length polymorphisms were identified in the organellar genomes among commercially used male-sterile cytoplasms from Holland, Japan and India, and were compared to S and T cytoplasms. Mitochondrial DNA diversity among 58 non-S-cytoplasmic open-pollinated onion populations was also assessed. All five putative CMS lines selected from the Indian population Nasik White Globe were identical to S cytoplasm for all polymorphisms in the chloroplast genome, and always possessed the same-sized mitochondrial fragments as S cytoplasm. T cytoplasm, the male-sterile cytoplasm used to produce the Dutch hybrid Hygro F₁, and two sources of CMS from Japan, were similar and showed numbers of mitochondrial polymorphisms similar to those observed among the 58 non-S-cytoplasmic open-pollinated populations. This research demonstrates that the same, or very similar, male-sterile cytoplasms have been independently isolated and exploited for hybrid-seed production in onion. Received: 27 October 1999 / Accepted: 12 February 2000     

Development of a SCAR marker for early identification of S-cytoplasm based on mitochondrial SRAP analysis in pepper (Capsicum annuum L.)

Molecular markers, coxII SCAR, atp6-2 SCAR and accD-U, have been used for marker-assisted selection of cytoplasmic male sterility (CMS) in pepper. However, the presence of these markers at the sub-stoichiometric level in maintainer lines affects the reliable selection of male sterile (S-) cytoplasm. This study aimed to develop a new CMS-specific molecular marker, SCAR₁₃₀, for reliable identification of S-cytoplasm in pepper, while the new and three previous molecular markers were used to determine the cytoplasm types of pepper lines. Based on mitochondrial genome sequence related amplified polymorphism (SRAP) analysis of the CMS lines and the maintainer lines, SCAR₁₃₀ was developed from a 10-bp deletion at the SRAP primer binding site in the CMS line (130 bp) compared with that in the maintainer line (140 bp). S-cytoplasm could be unambiguously selected from the pepper lines by the different length of the marker bands. Application of the four molecular markers to various pepper lines revealed that SCAR₁₃₀ is more reliable than the other three previous markers, orf507, ψatp6-2 and accD-U. Homology alignment with BLAST showed that the marker was located between trnE and trnS in the Nicotiana tabacum mitochondrial genome. Furthermore, expression of the marker-linked gene was significantly higher at the pollen abortive stage in the CMS line (HW203A) than in the maintainer line, which indicated that the marker was closely related to male sterility. Hence, factors other than orf507 and ψatp6-2 may exist for the regulation of male sterility in pepper.     

Mitochondrial DNA genetic polymorphism in thirteen rice cytoplasmic male sterile lines

Key message

Thirteen rice CMS lines derived from different cytoplasms were classified into eight groups by PCR amplification on mtDNA. The orf79 gene, which causes Boro II CMS, possibly results in Dian1-CMS.

Abstract

Thirteen rice cytoplasmic male sterile (CMS) lines derived from different cytoplasms are widely used for hybrid rice breeding. Based on 27 loci on mitochondrial DNA, including single nucleotide polymorphisms and segmental sequence variations between typical indica and japonica as well as high-polymorphism segmental sequence variations and single nucleotide polymorphisms among rice CMS lines, the 13 rice CMS lines were classified into eight groups: (I) wild-abortive CMS, Indonesian Shuitiangu CMS, K-CMS, Gang CMS, D-CMS and dwarf abortive CMS; (II) Maxie-CMS; (III) Honglian CMS; (IV) Boro II CMS; (V) Dian1-CMS; (VI) Liao-CMS; (VII) Lead CMS; and (VIII) Chinese wild rice CMS. According to their pollen abortion phenotypes, groups I and II (including 7 CMS lines) were classified as sporophytic CMS lines, the cytoplasmic genetic relationships among which were very close. They could have originated from similar, or even the same, cytoplasm donors. Groups III–VIII (including 6 CMS lines) were categorized as gametophytic CMS lines, the cytoplasms of which differed from one another, with some having relatively far genetic relationships. Dian1-CMS was found to harbor the orf79 gene, which causes Boro II CMS, whereas Liao-CMS had an orf79 structure that does not result in Lead CMS. Therefore, we speculated that orf79 is associated with Dian1-CMS but not with Liao-CMS. The atp6–orf79 structure related to sterility was also found to experience multiple evolutionary turnovers. All sporophytic CMS lines were indica-like. Except the Honglian CMS line, which was indica-like, all gametophytic CMS lines were japonica-like.     

Cytoplasmic male sterility in sorghum: Organization and expression of mitochondrial genes in Indian CMS cytoplasms

Cytoplasmic male sterility in sorghum has been reported in a number of varieties originating in different geographical regions (India, Africa and America). We have attempted to characterize three male sterile cytoplasms of Indian origin designated as Maldandi, Guntur and Vizianagaram by studying restriction fragment length polymorphisms (RFLPs) and expression patterns of 14 mitochondrial genes. Our results indicate that the cytoplasms, classified tentatively as Indian A₄ types, are distinct from the American A₄ and A₁ types. Although they are identical to each other with respect to the location of 10 of the mitochondrial genes selected, they can be distinguished from each other on the basis of RFLPs inatp6, atp9 andrrn18. Further the three cytoplasms differ from their maintainers in the location ofnad3, rpsl2 andatpA. Differences are also observed in the pattern of expression ofatpA between all the sterile lines and their respective maintainers.