Clovamide, a Hydroxycinnamic Acid Amide, Is a Resistance Factor Against Phytophthora spp. in Theobroma cacao

Abstract

The hydroxycinnamic acid amides (HCAAs) are a diverse group of plant-specialized phenylpropanoid metabolites distributed widely in the plant kingdom and are known to be involved in tolerance to abiotic and biotic stress. The HCAA clovamide is reported in a small number of distantly related species. To explore the contribution of specialized metabolites to disease resistance in cacao (Theobroma cacao L., chocolate tree), we performed untargeted metabolomics using liquid chromatography - tandem mass spectrometry (LC-MS/MS) and compared the basal metabolite profiles in leaves of two cacao genotypes with contrasting levels of susceptibility to Phytophthora spp. Leaves of the tolerant genotype 'Scavina 6' ('Sca6') were found to accumulate dramatically higher levels of clovamide and several other HCAAs compared to the susceptible 'Imperial College Selection 1' ('ICS1'). Clovamide was the most abundant metabolite in 'Sca6' leaf extracts based on MS signal, and was up to 58-fold higher in 'Sca6' than in 'ICS1'. In vitro assays demonstrated that clovamide inhibits growth of three pathogens of cacao in the genus Phytophthora, is a substrate for cacao polyphenol oxidase, and is a contributor to enzymatic browning. Furthermore, clovamide inhibited proteinase and pectinase in vitro, activities associated with defense in plant-pathogen interactions. Fruit epidermal peels from both genotypes contained substantial amounts of clovamide, but two sulfated HCAAs were present at high abundance exclusively in 'Sca6' suggesting a potential functional role of these compounds. The potential to breed cacao with increased HCAAs for improved agricultural performance is discussed.

Keywords: Phytophthora; Theobroma cacao; black pod rot; clovamide; hydroxycinnamic acid amide; metabolomics; oomycete; polyphenol oxidase.

Figure 1

Relative abundance of the 30 most abundant features in ‘Sca6’ of those >5-fold higher than in ‘ICS1’ (p < 0.05). Feature IDs (S-1 through S-30) match those in Table 1. MS counts represent mass spectrometer signal intensity of peaks integrated in XCMS Online (Tautenhahn et al., 2012). Error bars represent standard deviation. n = 3.

Figure 2

Semi-targeted LC-MS/MS analysis of HCAAs in leaf tissue (‘ICS1’ and ‘Sca6’). (A) Generic HCAA structure and diagnostic MS/MS fragments based on clovamide fragmentation (Arlorio et al., 2008) used to predict HCAAs in LC-MS/MS data. Two examples, Caffeoyl-DOPA (clovamide) and Coumaroyl-Tryptophan, are shown. (B) Relative abundance of putative HCAAs detected in LC–MS data. “N.D.” = not detected. ^* p < 0.05, ^** p < 0.01 (Welch’s t-test), n = 3. Error bars represent standard deviation.

Figure 3

Clovamide detection and quantification in cacao leaf and fruit peel. (A) High performance liquid chromatography – diode array detector (HPLC-DAD, 320 nm) chromatogram of ‘Sca6’ and ‘ICS1’ stage C leaf extracts, including UV absorbance spectra for peaks of interest. I.S. = internal standard. (B) Clovamide content (mg/gram tissue) from HPLC-DAD. ^** p < 0.01, ^*** p < 0.001, t-test (n = 5 for leaves, n = 3 for fruit peel). Error bars represent standard deviation. (C) Thin-layer Chromatography (TLC) plate with clovamide standard and stage C leaf extracts from ‘ICS1’ and ‘Sca6’. White line drawn near bottom of plate is origin of sample loading. Photograph taken under 365 nm UV excitation.

Figure 4

Clovamide’s contribution to enzymatic browning as a polyphenol oxidase (PPO) substrate. (A) Diagram of PPO-mediated formation of quinones from generic o-diphenol substrate and subsequent melanization/browning. Adapted from Yamane et al. (2010). (B) PPO activity (quinone formation) of ‘ICS1’ and ‘Sca6’ stage C leaf protein extracts with clovamide as substrate. [Clovamide] = 0.5 mM in all treatments shown. p-values from t-test, n = 3. (C) Browning in supernatant (Abs_{418 nm}) of ground stage C leaf disks in water (^** p < 0.01, n = 4). (D) Enhanced browning in supernatant (ΔAbs_{418 nm}) by addition of clovamide (~27.7 μg clovamide added per leaf disk, see methods; ^* p < 0.05, n = 4). Clov = clovamide. Error bars in (B–D) represent standard deviation.

Figure 5

Enzyme inhibition by clovamide and effect of cacao stage C leaf protein pectinase activity. (A) Proteolysis inhibiton by clovamide. “PK” = proteinase K. PK included in all treatments shown, with the addition of ‘Sca6’ protein, clovamide (2 mM), or both. Data represents two experiments (n = 5 from each). (B) Pectolysis inhibition by clovamide. “Pase” = pectinase from A. niger. “EGCG” = epigallocatechin gallate. Pase included in all treatments, with the addition of clovamide (2 mM), EGCG (2 mM), ‘Sca6’ stage C leaf protein, or ‘Sca6’ protein in combination with either phenolic compound. Data represents two experiments (n = 3 from each). (C) Enhancement of pectinase (A. niger) activity by cacao stage C leaf protein (n = 3). No pectinase (“-Pase”) and pectinase (“+Pase”) with or without addition of ‘ICS1’ or ‘Sca6’ leaf protein. Shared letters mean no difference by Tukey-HSD at p < 0.0001 (A), p < 0.05 (B), or p < 0.0001 (C). Error bars represent standard deviation.

Figure 6

Phytophthora spp. growth inhibition by clovamide. (A) Growth inhibition (%) by clovamide of three Phytophthora species on V8 media. (B) V8 media plates with P. megakarya mycelia with 0 mM clovamide (top) and 2 mM clovamide (bottom). (C) Growth inhibition (%) by clovamide of three Phytophthora species on semi-synthetic Henniger/Casein/Pectin (“HenCasPec”) media. (D) HenCasPec media plates with P. megakarya mycelia with 0 mM clovamide (top) and 2 mM clovamide (bottom), stained with calcofluor and photographed under 365 nm excitation for contrast between white mycelia and white media. Growth inhibition data from two experiments (n = 4 from each), except P. palmivora on HenCasPec, which represents one experiment. p < 0.01 (^**), p < 0.001 (^***), p < 0.0001 (^****), t-test. Error bars in (A) and (C) represent standard deviation.

Figure 7

LC-MS metabolomics of pod (fruit) infection by P. palmivora overview. (A) Principle Components Analysis (PCA) of LC-MS data, P. palmivora-infected or mock inoculated. (B) Loading chart showing contribution of individual metabolite features to each principle component. Greater deviation from zero on either axis represents a larger contribution to the respective principle component. Major contributors to each principle component are indicated with median mass to charge ratios (m/z) and putative annotations. All m/z shown are suspected molecular ions ([M-H]⁻) except one indicated as an isotope.

Figure 8

Changes in four HCAAs during fruit/pod infection by P. palmivora. (A) Putative structures of two sulfated HCAAs, feruloyl-DOPA, and clovamide. Parent ion m/z ([M-H]⁻) and diagnostic MS/MS fragments shown. (B) Metabolite abundance of the same four HCAAs in ‘ICS1’ and ‘Sca6’ fruit/pod tissue, mock inoculated or infected by P. palmivora. MS Counts represent mass spectrometer signal intensity of peaks integrated in XCMS Online (Tautenhahn et al., 2012). Shared letters mean no difference by Tukey-HSD (p < 0.05, n = 3). Error bars represent standard deviation.