Dickeya species belong to the family of soft rot Pectobacteriaceae (Adeolu et al. 2016) from which many if not all can cause plant diseases. In 2005, D. chrysanthemi, D. dadantii, D. dianthicola, D. paradisiaca, D. solani and D. zeae were described as species (Samson et al. 2005). More recently D. aquatica D. fongzhangdai and D. lacustris were described (Parkinson et al. 2014; Tian et al. 2016; Hugouvieux-Cotte-Pattat et al. 2019). From the various Dickeya species, D. dianthicola, D. dadantii and D. solani have been described as causative agents of potato blackleg (Czajkowski et al. 2011, Toth et al. 2011). For detection of the pathogen, various Dickeya species-specific TaqMan assays have been developed, such as for D. solani and D. dianthicola (Pritchard et al. 2013, van der Wolf et al. 2014), D. dadantii, D. paradisiaca, D. chrysanthemi, D. zeae and D. dieffenbachiae (van der Wolf et al. 2014). However, detection of multiple species with individual assays in large-scale testing programs is laborious and expensive. There is obviously a need for assays that enable faster and more cost-effective testing of seed lots. In 1996, a conventional PCR assay for the generic detection of all Dickeya species was developed (Nassar et al. 1996). Combined with an enrichment step to increase densities of Dickeya species in plant extracts, this PCR test seemed to be suitable as a tool in a disease management program (Tsror et al. 2012). However, conventional PCR assays require the use of gel electrophoresis as a post amplification step to analyse the amplicons, which make the assays less suitable for high throughput screening. Because TaqMan assays involve fluorescence measurements that can be performed without opening the PCR tube, the risk of carry-over product is greatly reduced (Kalinia et al. 1997). Moreover, the use of a fluorescent probe in TaqMan assays allows real time measurements which reduces labour costs and increases specificity. The aim of this study was to develop a TaqMan assay that detects all Dickeya species that can affect potato, i.e. D. dadantii, D. dianthicola and D. solani, but not (related) non-target bacteria, and that can be successfully applied for testing of potato tissues for blackleg-causing Dickeya species.

In order to design primers and probes that detect all known species of Dickeya present on potato in Europe, they need to be conserved in Dickeya, but they should not be found in non-target organisms. To find these conserved regions, the genome of a D. solani strain (IPO2222) was selected as the reference as it consisted of a single 4.9 Mb contig (Khayi et al.,2016). This reference sequence was split up in 500 bp fragments, resulting in 9898 sequences (set A). After mapping the sequences from set A with BWA MEM (Li, 2013) to D. chrysanthemi, D. dadantii, D. dianthicola, D. dieffenbachiae, D. solani and D. zeae, sequences aligning to all genomes were selected (conserved set B). Sequences from set B mapping to Pectobacterium atroseptica, P. brasiliensis, P. carotovorum, P. odoriferum, P. betavasculorum and P. wasabiae were discarded. The remaining conserved 500 bp regions (set C) were sorted based on the number of mismatches within the group of blackleg-causing Dickeya species.

Two sets of primers and probes for the TaqMan PCRs for generic detection of all Dickeya species were designed on the basis of the selected conserved regions (set C) in Dickeya using Primer Express 3.0 (Applied Biosystems). The MGB TaqMan probes were labelled with a 5′-end reporter dye (FAM-6-carboxyfluorescein (FAM)) and a 3′-end quencher dye (BHQ-1). The first set was derived from a gene encoding a ferrodoxin subunit involved in aerobic respiration. It consisted of forward primer Fw109 (5’-GCGCGCAGCACTHGAT-3′), H = A or C or T; reverse primer Rv109 (5’-CGACGGCACGCTCAGAAT-3′) and probe P109 (5’-AAGCCGCGGAAAT-3′). The other set was derived from a potassium transporter Kup involved in potassium-ion uptake under hyper osmotic stress at a low pH. It consisted of forward primer Fw284 (5’-TGTGCGTTTTCGGGCTASTC-3′), S=C or G; reverse primer Rv284 (CCYTGTCTTCTGTTATCAATTCATTAAC-3′), Y=C or T and probe P284 (5’-AACCAGAATAAGGCCC-3′). DNA was isolated from bacterial strains listed in Table S1. DNA was extracted from pure cultures of bacteria grown for 48 h on TSA at 27 °C. The bacteria were scraped from the agar surface and stored at −20 °C until DNA extraction. The DNA was extracted using the Wizard Magnetic DNA purification System for Food (Promega, Leiden, The Netherlands). DNA yield was determined by fluorescence using the Pico® Green I dye (Invitrogen) and a TECAN Infinite® M200 pro microplate reader (Tecan Group Ltd., Switzerland). The selected primer sets and corresponding probes were tested in TaqMan assays performed in a 96-well format in an ABI Prism 7500 apparatus. For each TaqMan assay, 2 μl sample was mixed with 23 μl reaction mix containing 12.5 μl 2X Premix Ex Taq (Takara Bio Inc. Otsu, Shiga, Japan), 0.5 μl 50X ROX reference Dye II (Takara Bio Inc.), 100 nM probe and 300 nM of each forward and reverse primer. The reaction conditions were: an initial incubation step at 50 °C for 2 min; 95 °C for 10 min; 40 cycles of 95 °C for 15 s followed by 60 °C for 60 s. Analysis of the data was done by automatic threshold calculation within the 7500 system software version 1.2.3. Ct-values were interpreted qualitatively. Reactions with Ct-values below 35 were considered positive.

The specificity of the generic Dickeya TaqMan assays was evaluated using 0.4 ng DNA of pure cultures of target and non-target strains. Results are summarized in Table 1 and full results are shown in Table S1. All 36 strains of the 6 target species, i.e. D. chrysanthemi, D. dadantii, D. dianthicola, D. dieffenbachia, D. solani and D. zeae, were detected by both TaqMan assays. TaqMan reactions of the 77 strains of more than 15 non-target species tested were all negative. In general, Ct-values of both assays were similar. Ct-values of the tested strains of D. chrysanthemi, D. dadantii, D. dianthicola, D. dieffenbachia, D. solani and D. zeae varied between 15.4 and 23.2 when TaqMan set 109 was used and between 15.8 and 24.2 for TaqMan set 284. However, Ct-values of the D. paradisiaca strains were higher and varied between 29.3 and 36.0. Three out of four strains of D. paradisiaca were positive when TaqMan assay set 109 was used and two out of four were positive when TaqMan set 284 was used. None of the 77 non-target strains were positive, including 61 strains of various taxonomically related Pectobacterium species.

Table 1 Results of TaqMan assays (with set 109 and set 284), developed for generic detection of Dickeya species, with target and non-target bacteria
Full size table

TaqMan assays 109 and 284 were highly specific and detected all plant pathogenic Dickeya species involved in the studies, with the exception of D. paradisiaca for which relatively high Ct-values were found. Also in other studies, D. paradisiaca was found divergent from the other Dickeya species (Naushad et al. 2014). This pathogen has a narrow host range, being isolated only from Musa species and Zea mays that were grown in tropical regions. It has never been described in association with potato.

The TaqMan assays were further evaluated using ten lots of seed potatoes from different places in the Netherlands and belonging to different cultivars, i.e. Amorosa (Groningen), Binella (Zeeland), Diamant E (Groningen), Elmundo (Noord Holland), Fabula E (Noordoostpolder), Kondor (Zeeland), Monalisa A (Friesland), Monalisa (Friesland), Picasso A (Noord Holland) and Zafira A (Noordoostpolder). From the same ten seed lots regular (commodity) samples were collected as done by seed inspectors, as well as samples of refused tubers that do not meet the quality standards. The refused tubers showed bruises, were misshaped and/or showed decay. We hypothesized that Dickeya would be more readily detected in refused samples. Two times ten subsamples of 20 tubers per seed lot were analysed, ten subsamples for the commodity samples and ten subsamples for the regular samples per seed lot. These 20 tubers from each subsample were put in 100 μm thick bags (Topa, Lelystad) and 5 ml water was added. The bags were vacuumed and incubated for 7 days at room temperature. Subsequently 1 ml of the rotten extract was used for DNA extraction using the Wizard Magnetic DNA purification System for Food (Promega, Leiden, The Netherlands). One hundred regular potato subsamples and 97 refused potato subsamples were analysed with the TaqMan assays 109 and 284 for generic detection of Dickeya species with TaqMan assays 109 and 284.

Full results are shown in Table S2 and results are summarized in Table 2. Of 100 regular subsamples 90 subsamples were negative and four subsamples were positive for the presence of Dickeya species by both generic assays (Set 109 and Set 284). Four subsamples were positive with Set 109 and negative for set 284. Conversely, two subsamples were positive with set 284 and negative with set 109. Ct-values of the positive regular subsamples were relatively high (between 30 and 35) except for a highly positive sample of Elmundo with Ct-values of 17.0 and 17.7 for set 284 and set 109 respectively. Of the 97 refused subsamples, 82 (84.5%) were tested negative for the presence of Dickeya species by both generic TaqMan assays. Thirteen subsamples (13.4%) were tested positive by both generic assays. Two subsamples (2.1%) were positive with set 109 and negative with set 284.

Table 2 Evaluation of TaqMan assays for detection of plant pathogenic Dickeya species with potato tubers collected from seed lots
Full size table

The sensitivity, specificity and accuracy of both assays are summarized in Table 3. Results of both assays are similar and the calculated values of the selected characteristics are high. With set 284 there was a lower number of subsamples positive, which may be a disadvantage if the generic assay is used for initial screening. In the refused samples more Dickeya was found than in the regular samples (Table 2) which confirmed our hypothesis that Dickeya is more frequently present in rejected tubers that are misshaped, wounded or rotten.

Table 3 Characteristics of the TaqMan assays for generic detection of Dickeya species in potato seed lots
Full size table

In order to determine the detection ranges of the Dickeya TaqMan assays in a background of potato extract, 10-fold dilution series (10 fg – 10 ng) of genomic DNA of D. solani strain IPO 2187 were prepared in a background of ca. 34 ng of DNA extracted from potato extract from regular tubers tested negative for the presence of Dickeya. When set 109 was used, the response was linear with a slope of −3.29, an R2 of 0.996 and a Y-intercept of 39.0 (Fig.1A). The response was also linear when set 284 was used (Fig. 1B) with a slope of −3.58, an R2 of 0.975 and a Y-intercept of 41.5. For both TaqMan assays the detection level was 100 fg bacterial DNA per reaction in a background of potato DNA.

Fig. 1
figure 1

Standard curves showing the relationship between Ct (cycle threshold) and quantity of genomic DNA of Dickeya solani strain IPO 2187 in a background of DNA from potato extract using the generic Dickeya TaqMan assays. (A) Ct-values for set 109 plotted against the quantity (log base 10) of Dickeya DNA (fg). (B) Ct-values for set 284 plotted against the quantity (log base 10) of Dickeya DNA (fg)

Full size image

This study has shown that two TaqMan assays have been developed that can detect all Dickeya species found on potato. If necessary, samples positive in the generic assays can be further analysed with Dickeya species-specific assays.