Full-Length Transcriptome Survey and Expression Analysis of Cassia obtusifolia to Discover Putative Genes Related to Aurantio-Obtusin Biosynthesis, Seed Formation and Development, and Stress Response

Yin Deng, Hui Zheng, Zicheng Yan, Dongying Liao, Chaolin Li, Jiayu Zhou, Hai LiaoInternational Journal of Molecular Sciences 2018 · DOI

TL;DR

This study presents the first full-length transcriptome of Cassia obtusifolia, a medicinal herb, using a hybrid sequencing approach. This comprehensive dataset provides crucial genomic insights into the biosynthesis of aurantio-obtusin, seed development, and stress responses, identifying key candidate genes for future research and breeding efforts.

What Problem Does This Paper Address?

Cassia obtusifolia (Jue-ming-zi) is a valuable medicinal herb, food source, and landscape plant, with its seeds being particularly important for pharmaceutical applications due to the presence of aurantio-obtusin, a key anthraquinone compound. Understanding the molecular mechanisms behind aurantio-obtusin biosynthesis, seed formation and development, and the plant's response to environmental stresses is crucial for its medicinal and agricultural utilization. However, previous genomic information for C. obtusifolia was limited, primarily relying on short-read sequencing technologies. These short reads often resulted in incomplete gene sequences (unigenes) and lacked detailed expression profiles across different organs, hindering a thorough understanding of the plant's biology. This paper aims to bridge this knowledge gap by generating a high-quality, full-length transcriptome and analyzing gene expression patterns across various organs.

Methods

The researchers employed a hybrid sequencing approach, combining two powerful technologies:

  • Next-Generation Sequencing (NGS): Specifically, the Illumina HiSeq™ 4000 platform was used to generate a massive amount of short reads (361,427,021 reads, 108.13 Gb). This provides broad coverage and high accuracy for gene expression quantification.
  • Single-Molecule Real-Time (SMRT) Sequencing: The PacBio RS II platform was utilized to generate long reads (4,315,774 reads, 9.66 Gb). This technology is crucial for obtaining full-length cDNA sequences, which are essential for accurate gene identification and functional annotation, especially for complex transcripts and those with repetitive elements.

The raw reads from both platforms were then processed through a sophisticated pipeline:

  1. Isoform Clustering: Reads were clustered to identify consensus isoforms.
  2. Quality Correction: Low-quality isoforms were corrected using Illumina short reads.
  3. Non-redundancy Filtering: Redundant transcripts were removed to obtain a clean set of nonredundant transcripts.
  4. Functional Annotation: Transcripts were annotated against public databases (Nr, Swiss-Prot, GO, COG, KEGG, Pfam) to assign functions and pathways.
  5. Gene Expression Analysis: Fragments per kilobase of transcript per million mapped reads (FPKM) values were used to quantify gene expression levels across different organs.
  6. Candidate Gene Validation: Key genes (CoDXS, CoDXR) were functionally validated using a bacterial complementation assay. Expression patterns of selected genes were further confirmed using quantitative Real-Time PCR (qRT-PCR).

Datasets

The study generated and analyzed two primary datasets:

  • Illumina HiSeq™ 4000 data: 361,427,021 short reads (108.13 Gb) from fifteen cDNA libraries (five organs: seed, root, stem, leaf, flower; three biological replicates per organ).
  • PacBio RS II SMRT data: 4,315,774 long reads (9.66 Gb) from five pooled organ samples.

These datasets were used to generate a comprehensive transcriptome assembly of 58,106 nonredundant transcripts.

Key Insights and Ideas

  • Hybrid Sequencing is Key: The combination of SMRT and NGS proved highly effective, yielding a significantly higher proportion of full-length transcripts and improved annotation accuracy compared to previous NGS-only studies. This highlights the power of integrating long and short-read technologies for comprehensive transcriptome analysis.
  • Seed is the Aurantio-Obtusin Hub: Aurantio-obtusin, a key medicinal compound, is predominantly found and likely synthesized in the seeds of C. obtusifolia. This finding directs future research efforts towards seed-specific gene discovery.
  • Unraveling Biosynthetic Pathways: The study identified candidate genes involved in both the early and late stages of aurantio-obtusin biosynthesis, pinpointing potential rate-limiting steps (ICS, DXS, IPPS) and key modification enzymes (CYPs, SAM-dependent methyltransferases, UDPGs).
  • Seed Development Orchestrated by TFs: Several seed-specific transcription factors (TFs), particularly MYB and MADS-box types, were identified as potentially crucial regulators of seed formation and development.
  • Hsp20s as Stress Guardians: Members of the Hsp20 gene family showed high expression in reproductive organs (seed and flower) and exhibited varied responses to abiotic stresses, suggesting their role in maintaining cellular homeostasis and protecting these vital tissues.

Key Results

  • Transcriptome Assembly: Generated 67,222 consensus isoforms, with 81.73% longer than 1000 bp, and 58,106 nonredundant transcripts.
  • Annotation Success: Achieved a high annotation ratio of 98.25% for nonredundant transcripts.
  • Organ-Specific Genes: Identified 658 genes specifically expressed in the seed, suggesting specialized roles in seed physiology.
  • Biosynthesis Genes: Identified 1214 transcripts involved in secondary metabolite biosynthesis, with a significant number related to pathways leading to anthraquinones.
  • Candidate Genes: Pinpointed specific CYPs, SAM-dependent methyltransferases, UDPGs, and transcription factors as promising candidates for further functional studies related to aurantio-obtusin biosynthesis, seed development, and stress response.
  • Expression Validation: qRT-PCR results largely corroborated the transcriptome expression patterns, validating the reliability of the data.

Limitations and Future Work

  • Functional Validation: While candidate genes have been identified, their precise functions require further experimental validation through methods like gene knockout or overexpression studies.
  • Alternative Splicing: Although alternative splicing events were detected, their functional significance in C. obtusifolia requires deeper investigation, particularly for TFs involved in seed development.
  • IncRNA Annotation: A minority of long non-coding RNAs (lncRNAs) could be annotated, suggesting a need for further research into their roles.
  • Stress Response Complexity: The varied responses of Hsp20 genes to different abiotic stresses indicate complex regulatory mechanisms that warrant further exploration.

The study provides a robust genomic foundation for future research, enabling targeted genetic engineering for improved aurantio-obtusin production, enhanced seed quality, and better stress tolerance in C. obtusifolia.

Summary updated

Abstract

The seed is the pharmaceutical and breeding organ of <i>Cassia obtusifolia</i>, a well-known medical herb containing aurantio-obtusin (a kind of anthraquinone), food, and landscape. In order to understand the molecular mechanism of the biosynthesis of aurantio-obtusin, seed formation and development, and stress response of <i>C. obtusifolia</i>, it is necessary to understand the genomics information. Although previous seed transcriptome of <i>C. obtusifolia</i> has been carried out by short-read next-generation sequencing (NGS) technology, the vast majority of the resulting unigenes did not represent full-length cDNA sequences and supply enough gene expression profile information of the various organs or tissues. In this study, fifteen cDNA libraries, which were constructed from the seed, root, stem, leaf, and flower (three repetitions with each organ) of <i>C. obtusifolia</i>, were sequenced using hybrid approach combining single-molecule real-time (SMRT) and NGS platform. More than 4,315,774 long reads with 9.66 Gb sequencing data and 361,427,021 short reads with 108.13 Gb sequencing data were generated by SMRT and NGS platform, respectively. 67,222 consensus isoforms were clustered from the reads and 81.73% (61,016) of which were longer than 1000 bp. Furthermore, the 67,222 consensus isoforms represented 58,106 nonredundant transcripts, 98.25% (57,092) of which were annotated and 25,573 of which were assigned to specific metabolic pathways by KEGG. <i>CoDXS</i> and <i>CoDXR</i> genes were directly used for functional characterization to validate the accuracy of sequences obtained from transcriptome. A total of 658 seed-specific transcripts indicated their special roles in physiological processes in seed. Analysis of transcripts which were involved in the early stage of anthraquinone biosynthesis suggested that the aurantio-obtusin in <i>C. obtusifolia</i> was mainly generated from isochorismate and Mevalonate/methylerythritol phosphate (MVA/MEP) pathway, and three reactions catalyzed by Menaquinone-specific isochorismate synthase (ICS), 1-deoxy-d-xylulose-5-phosphate synthase (DXS) and isopentenyl diphosphate (IPPS) might be the limited steps. Several seed-specific CYPs, SAM-dependent methyltransferase, and UDP-glycosyltransferase (UDPG) supplied promising candidate genes in the late stage of anthraquinone biosynthesis. In addition, four seed-specific transcriptional factors including three MYB Transcription Factor (MYB) and one MADS-box Transcription Factor (MADS) transcriptional factors) and alternative splicing might be involved with seed formation and development. Meanwhile, most members of Hsp20 genes showed high expression level in seed and flower; seven of which might have chaperon activities under various abiotic stresses. Finally, the expressional patterns of genes with particular interests showed similar trends in both transcriptome assay and qRT-PCR. In conclusion, this is the first full-length transcriptome sequencing reported in Caesalpiniaceae family, and thus providing a more complete insight into aurantio-obtusin biosynthesis, seed formation and development, and stress response as well in <i>C. obtusifolia</i>.
BibTeX
@article{deng2018full,
  author = {Yin Deng and Hui Zheng and Zicheng Yan and Dongying Liao and Chaolin Li and Jiayu Zhou and Hai Liao},
  title = {{Full-Length Transcriptome Survey and Expression Analysis of Cassia obtusifolia to Discover Putative Genes Related to Aurantio-Obtusin Biosynthesis, Seed Formation and Development, and Stress Response}},
  journal = {International Journal of Molecular Sciences},
  year = {2018},
  doi = {10.3390/ijms19092476},
}
Figure