Introduction
The identification of cytoplasmic male sterility (CMS) and maintaining genotypes was a major step in the success of hybrid breeding programmes of sugarbeet. The first report of CMS in this crop was made by F. V. Owen, who found male sterile plants in an old cultivar, ‘US1’ (Owen, Reference Owen1945). He described that male sterility resulted from the interaction of two recessive nuclear genes (x and z) with the sterilizing cytoplasm and identified the maintainer plants, which had an xxzz genotype and normal fertile cytoplasm. Hybrid seed production in sugarbeet has relied entirely on this single CMS source, the so-called Owen CMS (Bosemark, Reference Bosemark and Draycott2006). Up to now, additional sources of CMS have been discovered in wild beet populations (e.g. Mikami et al., Reference Mikami, Kishima, Sugiura and Kinoshita1985; Halldén et al., Reference Halldén, Bryngelsson and Bosemark1988; Bosemark, Reference Bosemark, Frese, Panella, Srivastava and Lange1998; Touzet et al., Reference Touzet, Hueber, Bürkholz, Barnes and Cuguen2004) and might offer the opportunity to broaden the genetic base of the cytoplasm. During the past decade, male sterility-associated mitochondrial genes have been reported in three CMS sources in beets. We present in this study a brief overview of recent advances in our understanding of the mechanism of CMS in beets.
Owen CMS-associated gene
Because of its importance for breeding, the molecular basis of Owen CMS has been extensively investigated. Kubo et al. (Reference Kubo, Nishizawa, Sugawara, Itchoda, Estiati and Mikami2000) and Satoh et al. (Reference Satoh, Kubo, Nishizawa, Estiati, Itchoda and Mikami2004) determined the entire nucleotide sequences of the mitochondrial genomes from normal fertile and Owen CMS sugarbeet plants. In-depth sequence comparison of the two mitochondrial genomes, together with a mitochondrial protein assay (Yamamoto et al., Reference Yamamoto, Kubo and Mikami2005), indicated that the 5′ leader sequence of atp6 (designated preSatp6) encodes a variant 39 kDa protein that is possibly related to Owen CMS (Table 1). The 39 kDa preSATP6 is closely associated with the mitochondrial membrane and assembles into an Owen CMS-specific protein complex. Interestingly, this characteristic is shared by the sterility-related proteins identified in CMS-T maize and Ogura-CMS radish: URF13 and ORF138 (Grelon et al., Reference Grelon, Budar, Bonhomme and Pelletier1994; Krishnasamy and Makaroff, Reference Krishnasamy and Makaroff1994; Wise et al., Reference Wise, Gobelman-Werner, Pei, Dill and Schnable1999).
CMS sources in beets
a Chromosomal location is given in parentheses.
Other CMS sources
A second source of CMS, called I-12CMS(3), was derived from wild beets collected in Pakistan (Mikami et al., Reference Mikami, Kishima, Sugiura and Kinoshita1985). The I-12CMS(3) and Owen cytoplasms show different patterns of male sterility maintenance and male fertility restoration when crossed with the same pollen parents (Mikami et al., Reference Mikami, Kishima, Sugiura and Kinoshita1985). What is worthy of special mention is that preSATP6 is missing in the sterile anthers carrying the I-12CMS(3) cytoplasm, which instead express a CMS-correlated protein of 12 kDa (Yamamoto et al., Reference Yamamoto, Shinada, Onodera, Komaki, Mikami and Kubo2008). This 12 kDa protein proved to be encoded by an unusual mitochondrial gene, designated orf129 (Table 1). The translation product of orf129 was found to be primarily present in the matrix and loosely associated with the inner mitochondrial membrane, a feature shared by the PCF protein involved in petunia CMS (Nivison et al., Reference Nivison, Sutton, Wilson and Hanson1994). Further implicating orf129 with CMS is the observation that ORF129 causes pollen disruption in transgenic tobacco plants when targeted to the mitochondria (Yamamoto et al., Reference Yamamoto, Shinada, Onodera, Komaki, Mikami and Kubo2008).
Another interesting case is GCMS, which has a cytoplasmic origin in wild sea beets from France. Ducos et al. (Reference Ducos, Touzet and Boutry2001) reported that GCMS is associated with a mitochondrial cox2 gene lacking eight highly conserved, C-terminal amino acids, and that cytochrome oxidase activity is decreased by 50% in vegetative tissues (Table 1). The GCMS can be classified as a loss-of-function mutant, in contrast to Owen CMS and I-12CMS(3), which are considered to be gain-of-function mutants. Beets thus appear to maintain distinct CMS cytoplasm, each capable of conferring male sterility by an apparently different mechanism.
Microsporogenesis breakdown in Owen CMS
In Owen CMS plants, male meiosis proceeds normally until the tetrad stage, after which the microspores degenerate either during tetrad formation or immediately after microspore liberation from the tetrads. Concurrently, the anther tapetum shows marked symptoms of abnormality. The most common irregularity is extensive vacuolation and enlargement (hypertrophy) of the tapetal cells, accompanied by mitochondrial disorganization (Kaul, Reference Kaul1988; Majewska-Sawka et al., Reference Majewska-Sawka, Rodriguez-Garcia, Nakashima and Jassen1993). At this stage of development, demand for energy or particular mitochondrial biosynthetic products may be markedly high, so the impairment of mitochondrial function is devastating (Budar and Berthomé, Reference Budar, Berthomé and Logan2007). Consistent with this hypothesis is the observation that plants expressing an antisense mitochondrial pyruvate dehydrogenase subunit gene in tapetal cells are male sterile (Yui et al., Reference Yui, Iketani, Mikami and Kubo2003). Male sterile transgenic plants, in common with Owen CMS plants, exhibited poor development of tapetal mitochondria and aberrant formation of microspore walls. These results led us to speculate that the synthesis of preSATP6 protein in the tapetum of Owen CMS plants results in dysfunctional mitochondria, which in turn could cause pollen abortion.
Restorer-of-fertility (Rf) genes
Hagihara et al. (Reference Hagihara, Itchoda, Habu, Iida, Mikami and Kubo2005a) found that male fertility restoration in Owen CMS by a Japanese breeding line is controlled by a single dominant gene (designated Rf1). They constructed a regional map encompassing the Rf1 locus (Hagihara et al., Reference Hagihara, Matsuhira, Ueda, Mikami and Kubo2005b). The mapping data also provided a clear indication of the location of the Rf1 locus on chromosome III, and the locus most likely corresponds to the X gene described previously by Owen (Pillen et al., Reference Pillen, Steinrücken, Herrmann and Jung1993; Table 1). The Rf1 locus could be further narrowed to the 250 kb region, which was delimited by two DNA markers (Hagihara et al., Reference Hagihara, Matsuhira, Ueda, Mikami and Kubo2005b). A second Rf locus (Rf2, Z) of Owen CMS has been located on chromosome IV (Schondelmaier and Jung, Reference Schondelmaier and Jung1997; Hjerdin-Panagopoulos et al., Reference Hjerdin-Panagopoulos, Kraft, Rading, Tuvesson and Nilsson2002; Table 1). In GCMS, at least two genes were reported to restore fertility, and one (RfG1) was mapped on chromosome VIII and was not co-located with either Rf1 or Rf2 (Touzet et al., Reference Touzet, Hueber, Bürkholz, Barnes and Cuguen2004; Table 1). This is consistent with the conclusion that the causal mechanism of male sterility in Owen CMS is different from that in GCMS.
An additional source of CMS (designated HCMS) from wild beets has been described by Boudry et al. (Reference Boudry, Mörchen, Saumitou-Laprade, Vernet and Van Dijk1993), but its sterilizing gene remains unclear. Genetic analysis of the fertility restoration of HCMS led to the conclusion that a single dominant gene (R1H) is involved (Laporte et al., Reference Laporte, Merdinoglu, Saumitou-Laprade, Butterlin, Vernet and Cuguen1998). Intriguingly, R1H is located on chromosome IV and is linked to the gene for monogerm seed character (Table 1). Several authors have reported weak linkage between the monogermy locus and the Rf2 gene associated with Owen CMS (Hogaboam, Reference Hogaboam1957; Roundy and Theurer, Reference Roundy and Theurer1974). The question inevitably arises whether the same mitochondrial gene (preSatp6) is responsible for both Owen CMS and HCMS or whether these two types have a different CMS mechanism but share a common Rf gene. This is worthy of further investigation.
Future prospects
How does the accumulation of CMS gene products, such as preSATP6 and ORF129, in anther tissue lead to failed pollen development? This issue has hardly been addressed. As mentioned earlier, it has been shown that sexual reproduction, and pollen production in particular, is highly sensitive to mitochondrial dysfunction. It is therefore interesting to define a range of anther-specific genes that may act downstream of preSatp6 or orf129 and that are responsible for aberrant biochemical and physiological processes leading to defective microsporogenesis, in response to mitochondrial dysfunction (Matsuhira et al., Reference Matsuhira, Shinada, Yui-Kurino, Hamato, Umeda, Mikami and Kubo2007). Another important clue to the CMS mechanism will certainly come from identification of the Rf gene products. This research is underway in our laboratory.
