Use of the growing environment as a source of variation to identify the quantitative trait transcripts and modules of co-expressed genes that determine chlorogenic acid accumulation

Abstract

Developing Coffea arabica seeds accumulate large amounts of chlorogenic acids (CGAs) as a storage form of phenylpropanoid derivatives, making coffee a valuable model to investigate the metabolism of these widespread plant phenolics. However, developmental and environmental regulations of CGA metabolism are poorly understood. In the present work, the expression of selected phenylpropanoid genes, together with CGA isomer profiles, was monitored throughout seed development across a wide set of contrasted natural environments. Although CGA metabolism was controlled by major developmental factors, the mean temperature during seed development had a direct impact on the time-window of CGA biosynthesis, as well as on final CGA isomer composition through subtle transcriptional regulations. We provide evidence that the variability induced by the environment is a useful tool to test whether CGA accumulation is quantitatively modulated at the transcriptional level, hence enabling detection of rate-limiting transcriptional steps [quantitative trait transcripts (QTTs)] for CGA biosynthesis. Variations induced by the environment also enabled a better description of the phenylpropanoid gene transcriptional network throughout seed development, as well as the detection of three temporally distinct modules of quantitatively co-expressed genes. Finally, analysis of metabolite-to-metabolite relationships revealed new biochemical characteristics of the isomerization steps that remain uncharacterized at the gene level.

Figure 1

Influence of developmental and environmental factors on chlorogenic acid (CGA) metabolism in the coffee seed. (a) Hierarchical cluster analysis (Euclidian distance, Ward grouping method) of the eight CGA isomers detected according to their profiles in the 16 locations sampled. (b) CGA isomer profiles: the X-axis represents seed developmental stages in chronological order, and the Y-axis represents the seed content in each CGA expressed as percentage dry mass (DM). For each isomer, the different lines show the accumulation patterns in the 16 experimental plots. Significant positive (+) and negative (−) correlations between air temperature and seed isomer contents are indicated (one symbol, P < 0.05; two symbols, P < 0.01; three symbols, P < 0.001; CQA, caffeoyl quinate; FQA, feruloyl quinate). (c) Correlations (P < 0.001) between air temperature and the seed content in the eight isomers. For each isomer, stages showing the highest level of significance (as estimated by P and R values) were retained to illustrate temperature–CGA correlations.

Figure 3

Simplified diagram showing the three ‘control boxes’ of chlorogenic acid (CGA) biosynthesis evidenced in the developing coffee seed: delay in core phenylpropanoid gene activation caused by low temperatures (control box 1), and upstream (2) and downstream (3) Quantitative trait transcripts (QTTs) for 5-CQA and 5-FQA synthesis. Correlations between temperature and transcript abundances, as well as between transcript abundance and metabolite contents, were significant at P < 0.05. The isomerization steps characterized by a constant conversion rate over locations and time are also presented. Conversion rates were estimated from the slope of regression lines (R > 0.8; P < 0.01).

Figure 2

Influence of the environment on the expression profile of genes involved in chlorogenic acid (CGA) and monolignol metabolic pathways. For each gene, the different patterns correspond to the 16 experimental plots, the X-axis represents seed developmental stages in chronological order and the Y-axis represents the normalized expression value. Significant positive (+) and negative (−) correlations between air temperature and transcript abundance are indicated: one symbol, P < 0.05; two symbols, P < 0.01. Dashed lines indicate the alternative routes towards 5-caffeoylquinic acid and 5-feruloylquinic acid. C3'H, p-coumaroyl CoA 3-hydroxylase; C4H, trans-cinnamate 4-hydroxylase; CAD, cinnamyl alcohol dehydrogenase; CCoAOMT, caffeoyl-CoA 3-O-methyltransferase; CHS, chalcone synthase; CHR, chalcone reductase; 4CL, 4-coumarate:CoA ligase; CCR, cinnamoyl-CoA reductase; COMT, caffeic acid O-methyltransferase; F5H, ferulate 5-hydroxylase; FQT, feruloyl-CoA quinate feruloyl transferase; HCT, hydroxycinnamoyl-CoA:shikimate/quinate hydroxycinnamoyl transferase; HQT, hydroxycinnamoyl-CoA quinate hydroxycinnamoyl transferase; PAL, phenylalanine ammonia lyase; POD, secretory peroxidase.

Figure 4

Isomerization to minor CQA: relationships between minor mono- and di-CQA isomers during late stages of development (R > 0.8 and P < 0.01).

Figure 5

Network of positively co-expressed phenylpropanoid genes during seed development. Modules of co-expression were designed using a P value cut-off of 0.001 (corresponding to a Pearson's correlation coefficient >0.80). At each developmental stage, genes actively transcribed (expression value >0.1) are boxed. Significant correlations between the transcript abundances are shown by a straight line.