Lily is a popular and economically ornamental crop around the world. However, its high production of pollen grains causes serious problems to consumers, including allergies and staining of clothes. During anther development, the tapetum is a crucial step for pollen formation and microspore release. Therefore, it is important to understand the mechanism of tapetum degradation and microspore development in lily where free pollen contamination occurs. Here, we used the cut lily cultivar ‘Siberia’ to characterize the process of tapetum degradation through the use of cytology and transcriptomic methods. The cytological observation indicated that, as the lily buds developed from 4 cm (Lo 4 cm) to 8 cm (Lo 8 cm), the tapetum completed the degradation process and the microspores matured. Furthermore, by comparing the transcriptome profiling among three developmental stages (Lo 4 cm, Lo 6 cm and Lo 8 cm), we identified 27 differentially expressed genes. These 27 genes were classed into 4 groups by function, namely, cell division and expansion, cell-wall morphogenesis, transcription factors, LRR-RLK (leucine-rich repeat receptor-like kinases), plant hormone biosynthesis and transduction. Quantitative real-time PCR was performed as validation of the transcriptome data. These selected genes are candidate genes for the tapetum degradation and microspore development of lily and our work provides a theoretical basis for breeding new lily cultivars without pollen.
Fusarium wilt (caused by Fusarium oxysporum f. sp. Lilii) is one of the most damaging diseases in lily (Lilium sargentiae Wilson). Although some F. oxysporum-resistant lily varieties have been identified and are being utilized in resistant breeding, the regulation network of the resistance-associated mechanisms is yet to be studied due to the lack of reliable reference genes for qRT-PCR (quantitative reverse transcription PCR) normalization. The reliability of results by qRT-PCR relies mainly on the stability of the reference genes. This study investigated the reliability of nine candidate reference genes (CYP, EF1-α, GAPDH, TUB, UBQ, AQP, HIS, PGK, and RPL13) for qRT-PCR analysis of F. oxysporum-resistant genes. Expression stability analysis via common programs GeNorm, BestKeeper, and NormFinder, at different time points post-inoculation of F. oxysporum, revealed that all nine genes met the basic requirements of reference genes. Amongst them, HIS and GAPDH displayed the highest and the lowest expression stability, respectively. The reliability of HIS was further validated by analyzing the expression levels of four resistance-related candidate genes. The expression patterns of the four target genes were consistent with their responses to pathogenetic fungi in other plants. Our results show that HIS is the most suitable reference gene for accurately normalizing F. oxysporum-resistant genes’ expressions in L. sargentiae.