Transcriptome profiling of Camelina sativa to identify genes involved in triacylglycerol biosynthesis and accumulation in the developing seeds

Abstract

Background: Camelina sativa is an emerging dedicated oilseed crop designed for biofuel and biodiesel applications as well as a source for edible and general-purpose oils. Such valuable oilseed crop is subjected to plant breeding programs and is suggested for large-scale production of better seed and oil quality. To accomplish this objective and to further enhance its oil content, a better understanding of lipid metabolism at the molecular level in this plant is critical. Here, we applied tissue transcriptomics and lipid composition analysis to identify and profile the genes and gene networks associated with triacylglycerol (TAG) biosynthesis, and to investigate how those genes are interacting to determine the quantity and quality of Camelina oil during seed development.

Results: Our Camelina transcriptome data analysis revealed an approximate of 57,854 and 57,973 genes actively expressing in developing seeds (RPKM ≥ 0.1) at 10-15 (Cs-14) and 16-21 (Cs-21) days after flowering (DAF), respectively. Of these, 7932 genes showed temporal and differential gene expression during the seed development (log2 fold change ≥1.5 or ≤-1.5; P ≤ 0.05). The differentially expressed genes (DEGs) were annotated and were found to be involved in distinct functional categories and metabolic pathways. Furthermore, performing quantitative real-time PCR for selected candidate genes associated with TAG biosynthesis validated RNA-seq data. Our results showed strong positive correlations between the expression abundance measured using both qPCR and RNA-Seq technologies. Furthermore, the analysis of fatty-acid content and composition revealed major changes throughout seed development, with the amount of oil accumulate rapidly at early mid seed development stages (from 16-28 DAF onwards), while no important changes were observed in the fatty-acid profile between seeds at 28 DAF and mature seeds.

Conclusions: This study is highly useful for understanding the regulation of TAG biosynthesis and identifying the rate-limiting steps in TAG pathways at seed development stages, providing a precise selection of candidate genes for developing Camelina varieties with improved seed and oil yields.

Keywords: Camelina sativa; Fatty-acid profiling; Lipid metabolism; Transcriptome profiling; Triacylglycerol biosynthesis.

Fig. 1

Expression profiling of selected Camelina TAG-associated genes. DGAT1 (a), MGAT1 (b), PDCT (c), WRl1 (d), PDAT1 (e), WSD1 (f), and LPAT3 (g) examined by quantitative real-time PCR in different camelina tissues; developing seeds, leaves, and flower tissues. Gene expression levels were normalized with respect to the internal control Actin2. Data bars represent the mean ± SE level of relative transcript abundance of three replicates. DAF days after flowering. Abbreviated names for the genes are described in Additional file 1: Table S1

Fig. 2

Rate of oil deposition during Camelina seed development. Oil content values are expressed as total oil amount (mg or µg, gray line) and  % FW (black line). The significance of the effect of developmental seed stages was tested by ANOVA (F and P values). Data represent the mean of three independent measurements ±SE

Fig. 3

Distribution of RPKM values (in log2 scale) for the genes identified in Camelina developing seeds. The means for both the original and log2 transformed RPKMs, maximum RPKMs, 75 % percentile (Q75), the total number of genes analyzed, and the genes with RPKM >0.1 are shown

Fig. 4

Venn diagram showing the actively expressing genes in developing seeds and leaf tissues. Among the genes, 49,586 are expressed in seeds and leaf tissues, 5481 are co-expressed in Cs-14 and Cs-21, 1087 are co-expressed in Cs-14 and leaf, and 1564 are co-expressed in Cs-21 and leaf. The number of tissue—specifically expressed genes—is 1700 (Cs-14), 1342 (Cs-21), and 5756 (leaf), respectively. Cs-14: 10–15 DAF; Cs-21: 16–21 DAF; Cs-Leaf: leaf

Fig. 5

Changes in gene expression profiles among the different developmental stages of Camelina seeds and leaf tissues. The number of differentially expressed genes between Cs-14 and Cs-21, Cs-14 and Cs-Leaf, Cs-21 and Cs-Leaf, is summarized. Between Cs-14 (10–15 DAF) and Cs-21 (16–21 DAF), there are 4223 genes up-regulated and 3709 genes down-regulated, between Cs-14 and Cs-Leaf, there are 8676 genes up-regulated and 10,737 genes down-regulated, while there are 9356 genes up-regulated and 10,773 genes down-regulated between Cs-21 and Cs-Leaf. Cs-14 sample was used as a control in Cs-14 vs Cs-21 comparison, while Cs-Leaf sample was used as a control in both Cs-14 vs Cs-Leaf and Cs-21 vs Cs-Leaf comparisons

Fig. 6

GO classification: Gene ontology distribution of the differentially expressed transcripts in Camelina developing seeds. Cs-14 (a) and Cs-21 (b). The results are summarized in three main categories of GO classification; biological process, molecular function, and cellular component

Fig. 7

Working model for the genes/gene networks involved in fatty acid and TAG biosynthesis in Camelina sativa. The expression abundance (RPKMs in log2 scale) for the selected candidate genes are highlighted in different color scales in Camelina developing seeds at 10–15 days after flowering (Seed 14 DAF) and at 16–21 DAF (seed 21 DAF) as well as in leaf tissues. The full name for the metabolites shown in the pathways is pyruvate, acetyl-CoA, malonyl-CoA, malonyl-ACP acyl-carrier protein, Acetoacetyl-ACP, Acyl-ACP, sn-Glycerol 3-phosphate G3P, LPA lysophosphatidic acid, PA Phosphatidic acid, MAG monoacylglycerol, FAA free fatty acid, DAG 1,2-Diacylglycerol, TAG Triacylglycerol, PC phosphatidylcholine, LPC lysophosphatidylcholine, and glycerol. The enzymes shown here are pyruvate dehydrogenase E1-α (PDH-E1-α), PDH-E1-β pyrxuvate dehydrogenase E1-β, PDC pyruvate dehydrogenase complex, ACC1 Acetyl-CoA carboxylase, ACCase-α acetyl-CoA carboxylase, a-carboxyltransferase, EMP3147 acyl-carrier-protein S-malonyltransferase, KASI ketoacyl-ACP synthase I; KASII ketoacyl-ACP Synthase II; KASIII ketoacyl-ACP synthase III, WRl1 wrinkled 1, ENR1 enoyl-ACP reductase, ATFATA fatA acyl-ACP thioesterase, FATA acyl-ACP thioesterase A, FATB fatty-acyl-ACP thioesterase B, SAD stearoyl-ACP desaturase, LACS long chain Acyl-CoA synthase, NHO1 protein-similar to glycerol kinase, MGAT monoacylglycerol acyltransferase, AAPT1 Choline/ethanolaminephosphotransferase, GPAT9 glycerol-3-phosphate acyltransferase 9, LPAT2 lysophosphatidyl acyltransferase 2, SDP1 sugar-dependent protein, SDP1-L sugar-dependent 1-like protein, PAH1 phosphatidic acid phosphohydrolase 1, PAH2 phosphatidic acid phosphohydrolase 1, DGAT1 diacylglycerol O-acyltransferase 1, DGAT2 diacylglycerol O-acyltransferase 2, PDAT1 phospholipid:diacylglycerol acyltransferase 1, PDAT2 phospholipid:diacylglycerol acyltransferase 2, DGK diacylglycerol kinase, AGK acylglycerol kinase, MGLL acylglycerol lipase, PDCT phosphatidylcholine: diacylglycerol cholinephosphotransferase, AAPT aminoalcoholphosphotransferase, LPCAT1 lysophosphatidylcholine acyltransferase 1, LPCAT2 lysophosphatidylcholine acyltransferase 2, PLA2 phospholipase A2, Ole1 Oleosin1, Ole2 Oleosin 2, and Ole4 Oleosin 4. This model is modified from the model published by Wang and colleagues in 2012 [44], Dussert and colleagues in 2013 [45], and the glycerolipids metabolism pathway in KEGG database http://www.genome.jp/kegg/. The Gene/enzyme names are modified from the names available in the TAIR database (http://www.Arabidopsis.org) and the Camelina genome database (http://www.camelinadb.ca)

Fig. 8

Expression of TAG biosynthesis-associated genes in Camelina developing seeds and leaf measured by qRT-PCR. Relative combined expression of all three copies of CsWRl1, CsGPAT9, CsLPP1, CsLPP2, CsLPAT2, CsDGAT1, CsDGAT2, CsPDAT, CsWSD1, CsPDCT, CsOle1, CsOle4, and CsMGAT1. The descriptive gene names are available in supplemental tables. The bars represent the fold change in log2 scale as measured by real-time qPCR from cDNA derived from Camelina seeds at 10–15 (Cs-14) and 16–21 (Cs-21) days after flowering (DAF) and from leaf tissue (Cs-Leaf). The leaf sample was used as the calibrator for the remaining samples. Error bars represent the standard error ± SE of three biological replicates. The quantification of the genes is measured relative to the expression of the indigenous housekeeping gene β-actin