Successful pod infections by Moniliophthora roreri result in differential Theobroma cacao gene expression depending on the clone's level of tolerance

Abstract

An understanding of the tolerance mechanisms of Theobroma cacao used against Moniliophthora roreri, the causal agent of frosty pod rot, is important for the generation of stable disease-tolerant clones. A comparative view was obtained of transcript populations of infected pods from two susceptible and two tolerant clones using RNA sequence (RNA-Seq) analysis. A total of 3009 transcripts showed differential expression among clones. KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis of differentially expressed genes indicated shifts in 152 different metabolic pathways between the tolerant and susceptible clones. Real-time quantitative reverse transcription polymerase chain reaction (real-time qRT-PCR) analyses of 36 genes verified the differential expression. Regression analysis validated a uniform progression in gene expression in association with infection levels and fungal loads in the susceptible clones. Expression patterns observed in the susceptible clones diverged in tolerant clones, with many genes showing higher expression at a low level of infection and fungal load. Principal coordinate analyses of real-time qRT-PCR data separated the gene expression patterns between susceptible and tolerant clones for pods showing malformation. Although some genes were constitutively differentially expressed between clones, most results suggested that defence responses were induced at low fungal load in the tolerant clones. Several elicitor-responsive genes were highly expressed in tolerant clones, suggesting rapid recognition of the pathogen and induction of defence genes. Expression patterns suggested that the jasmonic acid-ethylene- and/or salicylic acid-mediated defence pathways were activated in the tolerant clones, being enhanced by reduced brassinosteroid (BR) biosynthesis and catabolic inactivation of both BR and abscisic acids. Finally, several genes associated with hypersensitive response-like cell death were also induced in tolerant clones.

Keywords: RNA-Seq; differential transcriptome; frosty pod rot; hemibiotroph; plant defence; tolerance.

Figure 1

Differential transcriptomes between tolerant Theobroma cacao clones CATIE‐R4 and CATIE‐R7 and susceptible cacao clones Pound‐7 and CATIE‐1000 in response to frosty pod rot caused by Moniliophthora roreri. Numbers of transcripts induced (A) and repressed (B) in the tolerant relative to the susceptible clones. Transcripts are RNA reads identified from RNA sequence (RNA‐Seq) libraries aligned to the coding sequences of the cacao genome (Argout et al., 2010). The differential expression (P ≤ 0.05) level was determined using the count data of each gene with three replicates for each library.

Figure 2

Shift in the number of differentially expressed genes among key biological processes involved in the disease response between tolerant and susceptible Theobroma cacao clones in response to Moniliophthora roreri infection in pods. Differentially expressed genes were identified using RNA sequence (RNA‐Seq) analysis between two tolerant clones CATIE‐R4 and CATIE‐R7 and two susceptible clones CATIE‐1000 and Pound‐7. Gene ontology analysis was carried out using the program Blast2GO (Conesa et al., 2005). ABA, abscisic acid; BR, brassinosteroid; ET, ethylene; GA, gibberellic acid; HR, hypersensitive response; JA, jasmonic acid; PCD, programmed cell death; SA, salicylic acid.

Figure 3

Relative expression of fungal reference genes (fungal load, Fr), symptoms of infection and expression of 36 Theobroma cacao genes analysed over 74 artificially infected and uninfected pods representing two susceptible clones (Pound‐7 and CATIE‐1000) and three tolerant clones (UF‐273, CATIE‐R7 and CATIE‐R4). The expression levels for each of the three Moniliophthora roreri reference genes (Refs) in each sample were calculated as % cacao Ref(1–3) = 100 × [(E)^{Δ CT}], where E is the primer efficiency. Fr was estimated as the geometric mean of the expression levels of the three M. roreri  Refs. Symptoms of pod infection: –3, uninfected; –1, malformed; +1, chlorotic; +3, necrotic. Gene expression is the log₁₀ value of the relative mRNA expression (% cacao Ref) in each pod. Real‐time quantitative reverse transcription polymerase chain reaction (real‐time qRT‐PCR) was conducted using RNA harvested from infected (+) and uninfected pods (C) at regular time intervals (0 and 30 days after inoculation).

Figure 4

Relative expression of fungal reference genes (fungal load, Fr), symptoms of infection and expression of 36 cacao genes analysed over 140 infected pods representing two susceptible clones (Pound‐7 and CATIE‐1000) and three tolerant clones (UF‐273, CATIE‐R7 and CATIE‐R4). The expression levels for each of the three Moniliophthora roreri reference genes (Refs) in each sample were calculated as % Theobroma cacao  Ref(1–3) = 100 × [(E)^{Δ CT}], where E is the primer efficiency. Fr was estimated as the geometric mean of the expression levels of the three M. roreri Refs. Symptoms of pod infection: –3, uninfected; –1, malformed; +1, chlorotic; +3, necrotic. Gene expression is the log₁₀ value of the relative mRNA expression (% cacao Ref) in each pod. Real‐time quantitative reverse transcription polymerase chain reaction (real‐time qRT‐PCR) was conducted using RNA from pods harvested from infected plots with different stages of infection.

Figure 5

Principal components plot of the relative mRNA expression of the 36 genes selected on the basis of RNA sequence (RNA‐Seq) analysis and previous reports. (A) Pods from all the clones compared among different stages of infection. (B) All the infected pods compared for different clones. (C) All the malformed and chlorotic pods compared among different clones. (D) All the necrotic pods compared among different clones. Real‐time quantitative reverse transcription polymerase chain reaction (real‐time qRT‐PCR) was conducted using RNA from samples harvested from infected and uninfected pods of cacao.