ABSTRACT
Unisexual flowers have evolved repeatedly in the angiosperms. In Poaceae, multiple transitions from bisexual to unisexual flowers are hypothesized. There appear to be at least three distinct developmental mechanisms for unisexual flower formation as found in members of three subfamilies (Ehrhartoideae, Panicoideae, Pharoideae). In this study, unisexual flower development is described for the first time in subfamily Chloridoideae, as exemplified by Bouteloua dimorpha. Scanning electron microscopy (SEM) and anatomy were used to characterize the development of male (staminate) and female (pistillate) flowers, spikelets, and inflorescences. We found the developmental pathway for staminate flowers in B. dimorpha to be distinct from that described in the other three subfamilies, showing gynoecial arrest occurs at a different stage with possible loss of some cellular contents. However, pistillate flowers of B. dimorpha had some similarity to those described in other unisexual-flowered grasses, with filament and anther differentiation in abortive stamens. Comparing our findings with previous reports, unisexual flowers seem to have evolved independently in the four examined grass subfamilies. This analysis suggests the action of different genetic mechanisms, which are consistent with previous observations that floral unisexuality is a homoplasious condition in angiosperms.
ABSTRACT
Asparagus officinalis L. is a dioecious plant. A region called the M-locus located on a pair of homomorphic sex chromosomes controls the sexual dimorphism in asparagus. The aim of this work was to clone the region determining sex in asparagus from its position in the genome. The structure of the region encompassing M should be investigated and compared to the sex-determining regions in other dioecious model species. To establish an improved basis for physical mapping, a high-resolution genetic map was enriched with AFLP markers closely linked to the target locus by carrying out a bulked segregant analysis. By screening a BAC library with AFLP- and STS-markers followed by chromosome walking, a physical map with eight contigs could be established. However, the gaps between the contigs could not be closed due to a plethora of repetitive elements. Surprisingly, two of the contigs on one side of the M-locus did not overlap although they have been established with two markers, which mapped in a distance as low as 0.25 cM flanking the sex locus. Thus, the clustering of the markers indicates a reduced recombination frequency within the M-region. On the opposite side of the M-locus, a contig was mapped in a distance of 0.38 cM. Four closely linked BAC clones were partially sequenced and 64 putative ORFs were identified. Interestingly, only 25% of the ORFs showed sequence similarity to known proteins and ESTs. In addition, an accumulation of repetitive sequences and a low gene density was revealed in the sex-determining region of asparagus. Molecular cytogenetic and sequence analysis of BACs flanking the M-locus indicate that the BACs contain highly repetitive sequences that localize to centromeric and pericentromeric locations on all asparagus chromosomes, which hindered the localization of the M-locus to the single pair of sex chromosomes. We speculate that dioecious Silene, papaya and Asparagus species may represent three stages in the evolution of XX, XY sex determination systems. Given that asparagus still rarely produces hermaphroditic flowers and has homomorphic sex chromosomes, this species may be an ideal system to further investigates early sex chromosome evolution and the origins of dioecy.
