Transcriptome analysis of leaves, flowers and fruits perisperm of Coffea arabica L. Reveals the differential expression of genes involved in raffinose biosynthesis

Abstract

Coffea arabica L. is an important crop in several developing countries. Despite its economic importance, minimal transcriptome data are available for fruit tissues, especially during fruit development where several compounds related to coffee quality are produced. To understand the molecular aspects related to coffee fruit and grain development, we report a large-scale transcriptome analysis of leaf, flower and perisperm fruit tissue development. Illumina sequencing yielded 41,881,572 high-quality filtered reads. De novo assembly generated 65,364 unigenes with an average length of 1,264 bp. A total of 24,548 unigenes were annotated as protein coding genes, including 12,560 full-length sequences. In the annotation process, we identified nine candidate genes related to the biosynthesis of raffinose family oligossacarides (RFOs). These sugars confer osmoprotection and are accumulated during initial fruit development. Four genes from this pathway had their transcriptional pattern validated by quantitative reverse transcription polymerase chain reaction (qRT-PCR). Furthermore, we identified ∼24,000 putative target sites for microRNAs (miRNAs) and 134 putative transcriptionally active transposable elements (TE) sequences in our dataset. This C. arabica transcriptomic atlas provides an important step for identifying candidate genes related to several coffee metabolic pathways, especially those related to fruit chemical composition and therefore beverage quality. Our results are the starting point for enhancing our knowledge about the coffee genes that are transcribed during the flowering and initial fruit development stages.