Comparative Transcriptome Analysis of Two Species of Curcuma Contrasting in a High-Value Compound Curcumin: Insights into Genetic Basis and Regulation of Biosynthesis

Turmeric (Curcuma longa L), one of the widely used spices and herbal medicines, is rich in biologically active compound curcumin. Efficacy of curcumin in treating various diseases has been established through over 1000 published in vivo and in vitro studies. Curcumin content is reported to vary between accessions and subject to agro-climatic conditions, and the reasons are unexplored. Gaining an understanding on the molecular mechanism underlying curcumin biosynthesis will help in improving its content and maintaining stability under all conditions of cultivation. To investigate the genes involved in curcuminoid biosynthesis, we used Illumina sequencing platform and generated a substantial amount of expressed sequence tag (EST) dataset from two species viz., C. longa and its wild relative Curcuma aromatica Salisb. contrasting in curcumin content. The data reads of Illumina sequencing have been deposited in the National Center for Biotechnology Information Sequence Read Archive under BioProject ID PRJNA270561 and PRJNA277549 for C. longa and C. aromatica, respectively. De novo assembly produced a total of 99,482 and 104,514 contigs with an average length of 367 and 359 bp for C. longa and C. aromatica, respectively. The transcripts were assembled by BLAST similarity searches and annotated with Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) orthology identifiers. Bioinformatics analysis using BLASTX revealed that both the transcriptomes contained all the ten genes putatively involved in curcuminoid biosynthesis. Two novel polyketide synthase genes showing similarity to that of Musa accuminata (clpks1 and clpks2) were found to be upregulated in C. longa in comparison with C. aromatica. Transcription factors with putative roles in phenylpropanoid biosynthetic pathway were also reported for the first time in Curcuma spp. We also identified 5488 and 5620 putative simple sequence repeats (SSRs) for C. longa and C. aromatica, respectively. By using the assemblies as reference, 190 single-nucleotide polymorphisms (SNPs) for C. longa and 108 SNPs for C. aromatica in the candidate genes of curcuminoid biosynthetic pathway were identified. We could also detect 68 and 64 transcripts as microRNA (miRNA) targets from C. longa and C. aromatica, respectively. This study generated an extensive transcriptome data of C. longa and C. aromatica using deep sequencing. The candidate genes for enzymes involved in curcuminoid biosynthesis were identified from both the species. Differentially expressed genes, novel polyketide synthases, transcription factors, SSRs, SNPs and miRNA targets were identified from the transcriptomes. These EST sequences are a valuable public information platform for functional studies in turmeric and form a rich resource for studies on marker development and turmeric breeding.