The Power of Plasticity in Polyploid Persimmon.

Most plants are hermaphrodites, producing perfect flowers with both male and female functions. In roughly 6%of plants, however, male (usually XY) plants produce only male flowers and female (XX) plants produce only female flowers. Thesedioecious plants cannot self-pollinate, ensuring genetic diversity and facilitating the breeding process. In the dioecious diploid persimmon (Diospyros lotus), sex is determined by the autosomal gene MeGI (Japanese for female tree) and the Y-encoded pseudogeneOGI (male tree) (Akagi et al., 2014).MeGIencodesahomeodomain transcription factor that regulates flower development and anther fertility in a dosage-dependent manner, acting as a feminizing agent. In females, MeGI builds to high levels, repressing pollen formation. Inmales,OGIproducesa small RNA (smRNA) that targets and repressesMeGI, therebypreventing the accumulation of MeGI. The situation is more complex in the autohexaploid persimmon species D. kaki. These trees are either fully female ormonoecious (male and female flowers on separate branches). Monoecious trees are genetically male, prompting Akagi et al. (2016) to wonder how plants with Y chromosomes produce female flowers. Their analysis uncovered an intriguing epigenetic regulatory mechanism not found in D. lotus. The authors began by comparing three types of flowers: female flowers from female trees, female flowers frommonoecious trees, andmaleflowers frommonoecious trees.Like D. lotus flowers, male D. kaki flowers harbor MeGI targetedbysmRNA,withallsixcopiesof this gene equally targeted. As expected, OGI expression was completely undetectable in female flowers from both female andmonoecious D. kaki trees. Surprisingly, OGI expression was also undetectable in male flowers from monoecious trees. It turns out that the OGI promoter in D. kaki is interrupted by a 268-bp insertion resembling a SINE (short interspersed nuclear element)-like retrotransposonelement,whichappears tobeunique to monoecious D. kaki cultivars. This element exhibits heavy smRNA accumulation, along with strong cytosinemethylation, in bothmale and female buds and flowers fromgenetically male plants, suggesting it represses OGI expression in D. kaki. The position of a D. kaki bud on a branch affects the sex of the resulting flowers the following year (see figure). In June, each branch bears only male or female flowers and developing buds of unknown sex. The following year, the buds on these parental branches develop into new, secondary branches, each bearing only male or female flowers. Apical buds on female branches almost always give rise to female secondary branches, whereasmedial buds on the same brancheshaveanearlyequalchanceof forming male or female branches. By contrast, buds from male parental branches almost always develop into male branches, regardless of their position on the parental branch. This pattern suggests that an epigenetic mechanismisatplay. Indeed, themethylation pattern across MeGI in D. kaki strongly mirrors this trend, with the highest methylation levels found inmaturemale flowers andbuds and the lowest in their female counterparts. Therefore, gene silencing by methylation appears to regulate sexdetermination inD.kaki. This differential methylation pattern is much stronger in D. kaki than in D. lotus, which mainly uses OGI to silence this feminizing gene, suggesting that methylation of MeGI represents an alternative mechanism for sex determination in this hexaploid persimmon. Treatment with the nonspecific methylation inhibitor zebularine helped confirm this model: Zebularine treatment tended to promote pistil development and inhibit anther development inmale flowers ofmale-based Male and female flower development in monoecious D. kaki. For each branch, the relative percentages of male (blue) branches, female (pink) branches, and branches with no flowers (gray) are represented as stacked bars. (Adapted from Akagi et al. [2016], Figure 3B.)