Biocontrol of the Potato Blackleg and Soft Rot Diseases Caused by Dickeya dianthicola

Abstract

Development of protection tools targeting Dickeya species is an important issue in the potato production. Here, we present the identification and the characterization of novel biocontrol agents. Successive screenings of 10,000 bacterial isolates led us to retain 58 strains that exhibited growth inhibition properties against several Dickeya sp. and/or Pectobacterium sp. pathogens. Most of them belonged to the Pseudomonas and Bacillus genera. In vitro assays revealed a fitness decrease of the tested Dickeya sp. and Pectobacterium sp. pathogens in the presence of the biocontrol agents. In addition, four independent greenhouse assays performed to evaluate the biocontrol bacteria effect on potato plants artificially contaminated with Dickeya dianthicola revealed that a mix of three biocontrol agents, namely, Pseudomonas putida PA14H7 and Pseudomonas fluorescens PA3G8 and PA4C2, repeatedly decreased the severity of blackleg symptoms as well as the transmission of D. dianthicola to the tuber progeny. This work highlights the use of a combination of biocontrol strains as a potential strategy to limit the soft rot and blackleg diseases caused by D. dianthicola on potato plants and tubers.

FIG 1

Identification of Dickeya- and Pectobacterium-antagonistic strains. (A) Distribution (Venn diagram) of the 241 isolates according to their capacity for growth inhibition against D. dianthicola RNS04.9 (Dd04.9), D. solani 3337 (Ds3337), P. atrosepticum CFBP6276 (Pa6276), and P. carotovorum subsp. carotovorum RNS98.1 (Pcc98.1). (B) Distribution of the retained 96 bacterial isolates according to their genus identification. (C) Distribution (Venn diagram) of the retained 58 isolates according to their capacity for limiting symptoms induced by each pathogen in potato tuber assays. The successive screening and selection to keep 241, then 96, and finally 58 isolates are described in the text.

FIG 2

Pathogen fitness in the presence and absence of the growth-inhibiting strains The means of absolute fitness (W) of the pathogens (from the top, Dd04.9, Ds3337, Pa6276, and Pcc98.1) were calculated in the absence (−) or presence of the bacterial strains P. fluorescens PA4C2 and PA3G8 and P. putida PA14H7 (left panels) and P. brassicacearum PP1-210F and PA1G7 and B. simplex BA2H3 (right panels). Significantly different W values (by Kruskal-Wallis test then pairwise comparison test [α = 0.05]) are indicated by different letters.

FIG 3

MLSA-based relationship tree of the Pseudomonas strains. Concatenated gyrB, recA, rpoB, and 16S rRNA genes were used to construct a relationship tree of the Pseudomonas antagonistic strains (P. fluorescens PA3G8 and PA4C2, P. putida PA14H7, and P. brassicacearum PA1G7 and PP1-210F) and several well-identified strains of the Pseudomonas genus. An Azotobacter strain concatenated sequence was used to root the phylogenetic tree. The sequences were aligned with ClustalW, and then a neighbor-joining tree was created. The accession numbers of the NCBI sequences are indicated in parentheses.

FIG 4

Blackleg symptom dynamics in greenhouse assays. Two independent assays (B1 and B2) were conducted in Brittany and two others (N1 and N2) in Nord-Pas-de-Calais. In the four assays, a set of 20 plants was used under each of the 6 compared conditions: uninfected plants and plants infected with D. dianthicola RNS04.9 alone (Dd04.9) and in combination with each of the biocontrol P. fluorescens strains PA3G8 and PA4C2 and P. putida strain PA14H7 and a mixture of them (All 3). Left panels, disease progression curves (DPCs); right panels, Δt values (in days) and AUDPCr ratios.

FIG 5

Pathogen propagation in potato tubers. In experiments B2 in Brittany (upper graph) and N2 in Nord-Pas-de-Calais (lower graph), PCR detection of D. dianthicola RNS04.9 (Dd04.9) was performed on asymptomatic tubers harvested from asymptomatic plants which were cultivated in the presence and absence of the biocontrol strains (P. fluorescens PA3G8 and PA4C2, P. putida PA14H7, and all the three). Statistically significant decreases (Fisher's test; α = 0.05) of Dd04.9 presence are indicated by asterisks.

FIG 6

Dynamics of the pathogenic and biocontrol populations. In experiments B2 (left graphs) and N2 (right graphs), the population level in soil was quantified by qPCR when D. dianthicola RNS04.9 was inoculated alone (D.d04-9), and in combination with each of the biocontrol P. fluorescens strains PA3G8 and PA4C2 and P. putida strain PA14H7 and a mixture of all of them (All 3).