Читать книгу Biotechnology of Fruit and Nut Crops - Группа авторов - Страница 144

Molecular approaches help to resolve important questions with respect to the developmental biology of the oil palms. Jaligot et al. (2011) described studies of the mantled epigenetic flowering abnormality observed in oil palms produced in vitro. The large-scale development of somatic embryogenesis-based micropropagation protocols is hampered by the occurrence, in small but significant proportions, of the mantled somaclonal variant, which resembles the floral B class mutants of model plants (Jaligot et al., 2011; Jaligot and Rival, 2015). Abnormal palms may be sterile and therefore unproductive, thus cancelling out any potential genetic improvements made in the breeding programme from which they were obtained. Much interest therefore lies in identifying the molecular causes of the mantled phenotype, which has been demonstrated to be epigenetic in nature. Jaligot et al. (2011) reviewed current knowledge and priorities for future research on this complex but challenging phenomenon, which targeted molecular factors of particular interest, i.e. MADS-box genes and transposable elements, because of their involvement in related regulatory processes in model plants.

Beulé et al. (2011) used SSH and macroarray hybridization to compare transcriptome patterns between normal and mantled oil palm inflorescences. Two SSH libraries, enriched for complementary deoxyribonucleic acids (cDNAs) of either true-to-type or somaclonal variant material, were generated. Bioinformatics analysis of these two libraries allowed the identification of 1350 unique sequences and their annotation by a gene ontology-based approach. Macroarray hybridization was used to compare gene expression between normal and mantled inflorescences, and 32 genes were found to be differentially expressed. The temporal expression patterns of six genes were further investigated in more detail in relation to male and female inflorescence development. Full-length cDNAs were isolated and characterized for two of these genes, EgFB1 and EgRING1, both of which are downregulated in the mantled inflorescences and both of which encode proteins associated with proteolytic signalling complexes. The authors showed that EgFB1 and EgRING1 encode proteins which could potentially act in the same signalling pathway although this can only be a speculation given the lack of functional evidence. Nevertheless, it is interesting to note the similarity of the expression profiles observed for EgFB1 and EgRING1, both of which display reduced expression in the mantled inflorescence.

Shearman et al. (2013) performed RNA-Seq on developing flower and fruit samples of normal and mantled oil palm to characterize their transcriptomes. Expression data for all transcripts in normal and mantled flower and fruit samples were also studied, thus showing that many genes are differentially expressed, including several from pathways that may be the cause of the mantled phenotype if disrupted, i.e. genes involved in primary hormone responses, DNA replication and repair, chromatin remodelling and a gene involved in RNA-mediated DNA methylation. In addition, such gene expression data for developing flower and fruit can serve as a valuable resource for oil palm genetics and genomic studies. Among differentially expressed genes of normal and mantled samples there are several good candidates for explaining the phenotype including KTF1, the chromatin remodelling genes and AT1G21780. The most interesting find is differential expression in chromatin remodelling genes and histone methylation genes and is consistent with the hypothesis that somatic embryogenesis is disrupting the methylation pathway in a non-specific manner. Furthermore, non-specific disruption of the methylation pathway provides an explanation why different studies have found different patterns of methylation and different patterns of gene expression. This finding also suggests that the mantled phenotype is not a single phenotype, but rather a collection of phenotypes that include disruptions to other organs, i.e. leaf and root, but undergo selection at the plantlet stage to exclude any disruptions resulting in an apparent phenotype.

With the goal of exploring the relationship between epigenetic stability and the long-term in vitro proliferation of plant tissues, Rival et al. (2013) studied changes in genomic DNA methylation levels in oil palm embryogenic suspensions. Five embryogenic callus lines were obtained from selected hybrid seeds and then proliferated as suspension cultures. Cultures proliferated for 12 months and genomic DNA was sampled at 4-month intervals for the estimation of global DNA methylation rates through high performance liquid chromatography (HPLC) quantitation of deoxynucleosides. In vitro proliferation induces DNA hypermethylation in a time-dependent fashion. Moreover, this trend is statistically significant in several clonal lines and is shared among subclonal lines originating from the same genotype. Interestingly, the only clonal line undergoing loss of genomic methylation in the course of proliferation was unable to generate somatic embryos. Genome-wide DNA methylation changes in proliferating cells thus provide a powerful method for investigating mechanisms for maintaining genomic and epigenomic stability.

The mantled phenotype was associated with the DNA hypomethylation of a LINE retrotransposon (Karma) present in the intron of the B class gene DEFICIENS leading to alternative splicing and premature transcript termination (Ong-Abdullah et al., 2015). This epigenome-based approach confirmed previous results on the epigenetic origin of the mantled somaclonal variation which was previously deciphered by Jaligot et al. (2014) based on the epigenetic regulation of a MADS-Box candidate gene (Adam et al., 2006, 2007).

In order to elucidate the possible role of DNA methylation in the transcriptional regulation of EgDEF1, the APETALA3 orthologue of oil palm, Jaligot et al. (2014) studied this epigenetic marker within the gene in parallel with transcript accumulation in both normal and mantled developing inflorescences. The authors also examined the methylation and expression of two neighbouring retrotransposons that might interfere with EgDEF1 regulation. They showed that the EgDEF1 gene was essentially unmethylated and that its methylation pattern did not change with the floral phenotype; whereas expression is dramatically different, ruling out a direct implication of DNA methylation in the regulation of this gene. Jaligot et al. (2014) observed that both the gypsy element inserted within an intron of the EgDEF1 gene and the copia element located upstream from the promoter were heavily methylated and showed little or no expression. Interestingly a shorter, alternative transcript produced by EgDEF1 was identified and its accumulation was characterized with respect to its full-length counterpart. Depending on the floral phenotype, the respective proportions of these two transcripts changed differently during inflorescence development. This alternative splicing raises new questions in the search for the molecular mechanisms underlying the mantled phenotype in the oil palm.