*Corresponding author Introduction
Fire blight, caused by Erwinia amylovora, is a devastating disease affecting economically important species of family Rosaceae. Since its first description in northeastern United States at the end of the 18th century, fire blight rapidly spread throughout North America, Asia, Europe, and the Middle East (Bonn and Zwet, 2000; Doolotkeldieva et al., 2021; Gaganidze et al., 2021). In 2015, fire bight was first reported in Korea, and subsequently, the disease has extended rapidly in Korea (Myung et al., 2016; Park et al., 2016). Fire blight spreads through the transfer of bacterial ooze by insects, rain, strong wind, and other dispersal agents (Vanneste, 2000). E. amylovora infects apple and pear trees mainly via their flowers, where it rapidly develops and moves systemically through the parenchyma (Slack et al., 2017; Van Der Zwet and Keil, 1979). Although current control methods, such as antibiotics can protect apples and pears from infection, the only way to stop the spread of infection is to kill the infected trees or, in severe cases, to eradicate all trees in the affected orchard.
Black shoot blight caused by E. pyrifoliae is another bacterial disease affecting apple and pear trees, with symptoms are indistinguishable from those of fire blight. These two pathogens also share several morphological and biochemical characteristics, such that black shoot blight has been commonly misinterpreted as fire blight for extended periods (Rhim et al., 1999). Kim et al. (1999) proposed that E. pyrifoliae should be considered a single species, supported by DNA-DNA hybridization data and sequence analysis of 16S-23S intergenic transcribed spacer regions. Since then, numerous comprehensive studies of E. pyrifoliae have been conducted, including investigations of its host diversity, regional distribution, and genome (Jock et al., 2003; Jock and Geider, 2004; Shrestha et al., 2003, 2005).
Distinguishing fire blight from black shoot blight poses an enormous challenge due to the similarities of the causal species. Polymerase chain reaction (PCR)-based methods are valuable tools for this type of research (Bereswill et al., 1992; Jin et al., 2022; Kim et al., 2001; McManus and Jones, 1995; Salm and Geider, 2004). Given its widespread distribution worldwide, numerous specific primer sets for E. amylovora have been designed and applied in PCR detection (Bereswill et al., 1992; Powney et al., 2011; Taylor et al., 2001). In contrast, E. pyrifoliae is domesticated and only found on apple and pear trees in Korea, which has limited the availability of specific primers and detection methods for this pathogen (Ham et al., 2022). Since the introduction of E. amylovora in Korea in 2015, instances of both fire blight and black shoot blight have consistently been reported within the same city and year; this trend has increased in frequency from 2020 to 2022 (Choi et al., 2022; Lee et al., 2023), raising the possibility that these two Erwinia species could infect apple or pear trees within a single orchard or even coinfect a single tree in Korea (Choi et al., 2022). This risk poses a significant challenge for the detection and control of diseases in apple and pear orchards. Consequently, two PCR-based methods have already been developed to simultaneously detect fire blight and black shoot blight, one using a TaqMan duplex real-time PCR for simultaneous detection and quantification, and the other distinguishing E. amylovora from E. pyrifoliae based on their different amplicon sizes using a single primer set in one PCR reaction (Ham et al., 2022; Lehman et al., 2008).
In a previous study, we developed the droplet digital PCR (ddPCR) method for detecting and quantifying E. pyrifoliae (He et al., 2023). Digital PCR (dPCR) is a third-generation PCR technology that has shifted the detection standard for absolute quantification from fluorescence intensity to the presence or absence of fluorescence in target sequences. Specifically, the sample is partitioned into micro-reaction chambers prior to single nucleic acid amplification. The concentration of the targets is determined through Poisson distribution, calculated by assessing the ratio of micro-reaction chambers containing fluorescent signals to the total number of micro-reaction chambers (Vogelstein and Kinzler, 1999). The advantages of dPCR include precision and sensitivity, enabling the detection of lower target concentrations (Li et al., 2018). Moreover, dPCR offers superior reproducibility and repeatability and is unaffected by PCR inhibitors (Nilsen et al., 2014; Te et al., 2015). Therefore, dPCR has great potential for rapid pathogen screening at low levels, and has been successfully applied in medical practice, food safety detection, and agriculture (Maheshwari et al., 2017; Ramírez et al., 2019; Wang et al., 2018).
In this study, we developed a ddPCR platform to detect and quantify E. amylovora. In addition, we applied duplex-ddPCR with two independent fluorescence reading channels (FAM and HEX/VIC), which enables the simultaneous detection and absolute quantification of both E. amylovora and E. pyrifoliae.
Materials and Methods
Bacteria strains and DNA extraction.
Erwinia amylovora strain TS3128 and E. pyrifoliae strain YKB12327 were initially stored in 40% (v/v) glycerol at −80°C and later revitalized separately on Luria-Bertani (LB) agar plates at 28°C. After 12 hr, a single colony of each species was selected and inoculated at 28°C in liquid LB with shaking at 180 rpm for 12 hr for further study. Genomic DNA from both bacteria and plant tissues was extracted using the G-spin Total DNA Extraction kit (iNtRON Biotechnology, Seongnam, Korea), following the manufacturer's instructions. DNA quality and concentration were assessed using a NanoDrop spectrophotometer (NanoPhotometer NP80; Implen, Munich, Germany). Bacterial DNA concentrations were initially measured in ng/μl and converted into copies/μl before ddPCR was performed (Pan et al., 2020; Papić et al., 2017).
ddPCR assay.
Primer/TaqMan probe sets used in the ddPCR assays are listed in Table 1. These sets detect E. amylovora (Salm and Geider, 2004) and E. pyrifoliae (He et al., 2023) specifically.
Table 1.
Sequences of primers and TaqMan probes used for the quantification of Erwinia amylovora and E. pyrifoliae
| Species | Primer and probe | Sequence (5’→3’) | Reference |
|---|---|---|---|
| Erwinia amylovora | P29TF (forward) | CACTGATGGTGCCGTTG | Salm and Geider (2004) |
| P29TR (reverse) | CGCCAGGATAGTCGCATA | ||
| Probe | HEX-TACCTCCGCAGCCGTCATGG-BHQ | ||
| E. pyrifoliae | Pyr-F (forward) | CGGCGCGGGATTTATGTAT | He et al. (2023) |
| Pyr-R (reverse) | CCATGCTGCGTTAGTTGATATTG | ||
| Probe | FAM-AGAAGACTAT CAGCGGGAAGCCTACA-BHQ |
The ddPCR assays were performed on a QX200 Droplet Digital PCR System (Bio-Rad, Hercules, CA, USA). The detection conditions were optimized to eliminate intermediates and enhance the accuracy of the results. We previously established the optimal detection conditions for E. pyrifoliae (He et al., 2023). To optimize the detection of E. amylovora, we tested a thermal gradient with annealing temperatures ranging from 52.8°C to 64.8°C in singleplex ddPCR assays. The ddPCR reaction mixture included 10 μl of 2× ddPCR Supermix for Probes (Bio-Rad), 1 μl each primer (0.5 mM), 0.5 μl each probe (0.25 mM), 2 μl genomic DNA or cell culture template, and ultrapure water up to 22 μl. For droplet generation, 20 μl reaction mixture was transferred into an eight-channel DG8 cartridge (Bio-Rad), and 70 μl Droplet Generation Oil for Probes (Bio-Rad) was added to each oil well. The generated emulsion (40 μl) was transferred into a 96-well PCR plate (Bio-Rad) and sealed with a pierceable cover using a PX1 plate sealer (Bio-Rad). End-point PCR was performed using a C1000 thermal cycler (Bio-Rad) with the following conditions: an initial denaturation of 98°C for 10 min, followed by 40 cycles of a two-step thermal profile consisting of 94°C for 30 sec and 52.8°C to 64.8°C for 60 sec at a ramp rate of 2.0°C/sec. The final extension was performed at 98°C for 10 min, followed by cooling to 4°C. After the reaction, the plate was transferred to a QX200 droplet reader, and the data were analyzed using QuantaSoft software (Bio-Rad).
Fruit inoculation.
Healthy immature pear fruits (cv. ‘Shin-go’) were surface-sterilized by immersion in 2% sodium hypochlorite for 2 min, and then washed twice with sterilized water. For bacterial inoculation, we wounded the fruits to a depth of approximately 3 mm using sterilized toothpicks and then applied 10 μl of bacterial culture suspension containing varying ratios of E. pyrifoliae to E. amylovora (100:0, 90:10, 70:30, 50:50, 30:70, 10:90, and 0:100 v/v). The negative control was inoculated with sterile water. All treated fruits were placed in plastic boxes with high humidity and incubated in a dark chamber at 28°C.
Statistical analysis.
Linear regression analysis was performed using GraphPad Prism v9.5.1 software (GraphPad Software, Boston, MA, USA). QuantaSoft v1.7.4 software (Bio-Rad) was used to discriminate positive droplets from negative droplets according to a fluorescence amplitude threshold. For ddPCR data analysis, only reactions with more than 10,000 accepted droplets per well were included.
Results
Detection of E. amylovora by ddPCR methods.
In a previous study, we developed and optimized ddPCR platform for detecting E. pyrifoliae (He et al., 2023). To develop duplex detection of E. amylovora and E. pyrifoliae, we developed a singleplex species-specific detection method for E. amylovora. To optimize the ddPCR conditions, we applied an annealing temperature gradient for E. amylovora detection, which ranged from 52.8°C to 64.8°C. The annealing temperature plays a crucial role in ddPCR, as it facilitates the separation of positive and negative droplets. Thus, optimizing the annealing temperature enhances the sensitivity of the PCR, enabling the detection of low target concentrations. The application of different annealing temperatures resulted in variable separation of positive and negative signals. Signals dropped significantly at annealing temperatures above 62.5°C, with positive and negative droplets becoming almost indistinguishable at 64.8°C (Fig. 1A). Although signal separation improved at temperatures of 52.8°C-60.3°C, the nonspecific amplification and droplet impairment rates increased at these low annealing temperatures, which could lead to difficulty or errors in interpreting the results, potentially impacting the accuracy of experimental outcomes. Based on a previous study, the optimal annealing temperature for E. pyrifoliae ddPCR detection is 62.5°C (He et al., 2023). Therefore, in further experiments, we conducted simultaneous detection of both E. amylovora and E. pyrifoliae using duplex ddPCR based on an annealing temperature of 62.5°C, resulting in a clear separation between positive and negative droplet clusters with minimal intermediate “rain” (Fig. 1A).
Fig. 1.
Optimization of annealing temperature and linear regression analysis of ddPCR assays. (A) Erwinia amylovora cell culture (10⁴ cfu/ml) was used as the template for thermal gradient PCR amplification with annealing temperatures ranging from 52.8°C to 64.8°C. The pink line represents the threshold separating positive (green) and negative (gray) droplets. (B) Linear correlation between log-transformed genomic DNA or E. amylovora cell culture concentrations and their corresponding ddPCR-measured values. These experiments were repeated three times with similar results. ddPCR, droplet digital polymerase chain reaction.
The sensitivity of the singleplex ddPCR E. amylovora assay was determined by performing serial dilutions of genomic DNA and cell culture, ranging from 101 to 108 copies/ml or cfu/ml. The results showed that the lower detection limits for cell cultures and genomic DNA were 3.43±0.313 log cfu/ml and 3.91±0.173 log copies/ml, respectively. The upper detection limit for cell cultures was 7.34±0.0016 log cfu/ml, and that of genomic DNA was 7.92±0.013 log copies/ml (Table 2). Samples with templates of 108 copies/ml or cfu/ml from cell cultures and genomic DNA were unquantifiable due to the saturation of all droplets by positive signals. Despite the potential influence of sample preparation on the detection limit, these results suggest that the detection range for E. amylovora using singleplex ddPCR is from 103 to 107 copies/ml or cfu/ml in both genomic DNA and cell culture templates. (Fig. 1B, Table 2). To examine the linearity and dynamic range of the results in greater detail, regression curves were generated for the two templates. The genomic DNA and cell culture templates demonstrated strong correlations, with coefficient of determination (R²) values of 0.9957 and 0.9878, respectively (Fig. 1B).
Table 2.
Quantification of genomic DNA and cell cultures of E. amylovora using ddPCR
Performance of the duplex ddPCR assay.
To evaluate the quantification efficiency of the duplex ddPCR for the two Erwinia species and the linearity of the resulting data, templates were prepared by mixing cell cultures or genomic DNA from E. pyrifoliae and E. amylovora at a concentration of 10⁷ cfu/ml or copies/ml. Duplex ddPCR was performed to test templates including varying ratios of E. pyrifoliae to E. amylovora (100:0, 90:10, 70:30, 50:50, 30:70, 10:90, and 0:100 v/v). The results confirmed effective simultaneous detection and quantification of both Erwinia species in a single duplex ddPCR reaction (Fig. 2A,B), based on thresholds set automatically set by QuantaSoft. Positive droplet events detected in both cell culture and genomic DNA mixtures are shown in Fig. 2C,D.
Fig. 2.
Analysis of Duplex ddPCR Assays for E. pyrifoliae and E. amylovora Mixtures. (A, B) One-dimensional plots of duplex ddPCR assays showing specific detection across varying ratios of E. pyrifoliae (Ep) and E. amylovora (Ea). Normalized (A) bacterial cultures and (B) genomic DNA mixtures at different ratios (E. pyrifoliae/ E. amylovora: 100:0, 90:10, 70:30, 50:50, 30:70, 10:90, 0:100, v/v) were used as PCR templates. Blue dots indicate E. pyrifoliae positive droplets (FAM fluorescence-positive droplets), green dots indicate E. amylovora positive droplets (HEX fluorescence-positive droplets) and gray dots represent negative detection signals. (C, D) Quantification sensitivity analysis of E. pyrifoliae and E. amylovora mixtures at the same ratios using (C) cell cultures and (D) genomic DNA as templates. The Y-axis represents the number of droplets, while the X-axis shows the expected sample ratios. These experiments were repeated three times with similar results.
To evaluate the accuracy of the duplex ddPCR, fractional abundances were analyzed. The fractional abundances of both mixture templates closely matched the expected proportions of E. pyrifoliae and E. amylovora. However, compared to cell cultures, gDNA templates are easier to precisely adjust (Fig. 3A,B). Linear regression analysis was performed using the expected fractional abundances of E. pyrifoliae in prepared samples alongside those obtained from duplex ddPCR measurements. The results demonstrated excellent linearity, with slope values of 1.002 and 1.005, and R² values of 0.994 and 0.999 for genomic DNA and cell culture templates, respectively (Fig. 3C).
Fig. 3.
Fractional abundance plots show the percentage frequency of E. pyrifoliae and E. amylovora. (A) Cell culture mixtures with varying ratios (E. pyrifoliae/ E. amylovora: 100:0, 90:10, 70:30, 50:50, 30:70, 10:90, 0:100, v/v) were used as templates. (B) Genomic DNA mixtures containing the same ratios as the cell culture mixtures were used as templates. (C) Linear regression analysis of the expected versus measured fractional abundance in the duplex ddPCR assays. These experiments were repeated three times with similar results. ddPCR, droplet digital polymerase chain reaction.
Duplex ddPCR detection of two Erwinia species in plant tissues.
Bacterial mixtures with an initial concentration of 107 cfu/ml, containing different ratios of E. pyrifoliae to E. amylovora (100:0, 90:10, 70:30, 50:50, 30:70, 10:90, and 0:100, v/v), were inoculated into immature pear fruits. After 4 days, symptoms were observed in cut side sections of the fruits (Fig. 4A). In all inoculated fruits, water-soaking symptoms had developed at the inoculation site, and subsequently spread to the surrounding areas (Fig. 4A). To detect and quantify the two Erwinia species in inoculated fruit tissues, 0.3-g samples were collected around the inoculation points and ground in 3 ml sterile distilled water. A total of 20 µl genomic DNA was extracted from each sample and diluted 10-fold. Then, 2 µl samples of the diluted DNA were used as templates for detection by duplex ddPCR. Both E. pyrifoliae and E. amylovora were detected by ddPCR assays in all fruit samples inoculated with each Erwinia species alone or in mixed ratios (Fig. 4B). When E. pyrifoliae or E. amylovora was inoculated alone (ratio of E. pyrifoliae to E. amylovora, 100:0 or 0:100), no positive reaction was observed for the non-target species, confirming the species specificity of the assays (Fig. 4B). These results indicate that duplex ddPCR can sensitively and simultaneously detect both Erwinia species in plant tissues. This method offers a rapid and reliable tool for the early detection of initial infections, including in asymptomatic and subsymptomatic sample.
Fig. 4.
Detection of E. amylovora and E. pyrifoliae in artificially inoculated immature pear fruits using duplex ddPCR. (A) Symptoms observed in pear fruits 4 days post-inoculation with bacterial mixtures of varying E. pyrifoliae/ E. amylovora (Ep/Ea) ratios: 100:0, 90:10, 70:30, 50:50, 30:70, 10:90, and 0:100 (v/v). The negative control was treated with sterile water. (B) Duplex ddPCR detection of E. pyrifoliae and E. amylovora in total DNA extracted from inoculated pear fruits. Blue dots represent positive detection of E. pyrifoliae, green dots represent positive detection of E. amylovora, and gray dots indicate negative droplets without amplification. ddPCR, droplet digital polymerase chain reaction.
Discussion
Despite extensive government control efforts, fire blight and black shoot blight continue to spread across several regions in Korea, resulting in significant economic losses. In the absence of effective preventive measures, rigorous and precise monitoring programs are crucial for the early identification and management of these pathogens. Previous studies have developed several methods to detect the responsible pathogens (Bereswill et al., 1992; Dreo et al., 2012; Jin et al., 2022; Llop et al., 2000; Salm and Geider, 2004). However, because these diseases often occur in separate locations and E. pyrifoliae is a domestic pathogen in Korea, most existing methods are designed to detect only one of them. With the increasing risk of coinfection in Korean orchards and the difficulty of distinguishing E. pyrifoliae and E. amylovora, there is an urgent need for a rapid, sensitive technology capable of simultaneous detection (Ham et al., 2022).
Specific primers play a crucial role in pathogen detection, and their integration with TaqMan technology further boosts detection efficiency (Reynisson et al., 2006). The precise design of these primers ensures that they match the unique gene sequences of the target pathogen, enhancing detection accuracy and reducing false positive or false negative rates. TaqMan technology also allows quantification, enabling insights into infection levels and disease progression while maintaining high specificity by excluding interference from non-target pathogens or environmental DNA (Watson and Li, 2005). In this study, we employed primer and probe sets designed in previous research for ddPCR detection. For E. amylovora, a 112 bp product was amplified from the stable plasmid pEA29, which is present in all wild-type E. amylovora strains (Mohammadi, 2010; Salm and Geider, 2004). We slightly modified the probe by replacing the 5′-end reporter dye (FAM) with HEX and the 3′-end quencher (TAMRA) with BHQ. For E. pyrifoliae, primers and probes were designed from a putative protein-coding gene in the chromosome. The primer set produces a 106 bp product, with specificity for E. pyrifoliae verified through in silico and electrophoresis analyses (He et al., 2023).
The ddPCR method has become a widely used tool across scientific fields due to its unique advantages (Hou et al., 2022). In plant pathogen detection, ddPCR offers several benefits over conventional PCR-based methods, as it does not require external standards to quantify unknown samples and produces results quickly and visually (Dreo et al., 2014; Dupas et al., 2019; Whale et al., 2013). In addition, ddPCR is less affected by PCR inhibitors, allowing for reproducible data even under challenging sampling conditions (Hou et al., 2022). This robustness supports multiplex PCR applications, enabling the simultaneous detection of multiple target DNAs in a single reaction. Although ddPCR methods were previously developed for E. amylovora and E. pyrifoliae individually (Dreo et al., 2014; He et al., 2023), the risk of coinfection was not addressed. In this study, we standardized the detection conditions and utilized advanced equipment to develop a novel ddPCR method for E. amylovora, achieving a detection limit as low as 10³ copies or cfu/ml, with an upper limit of 10⁷ copies or cfu/ml in genomic DNA and cell culture templates. In addition, we developed a highly sensitive duplex ddPCR method for the simultaneous detection of E. amylovora and E. pyrifoliae using dual-labeled probes.
The duplex ddPCR assay demonstrated high accuracy in artificially inoculated fruits, effectively detecting both pathogens even when present at varying ratios. However, despite the high sensitivity of the assay, we did not detect a direct correlation between pathogen load and symptom severity, indicating that factors beyond pathogen concentration may influence disease manifestation. Future studies should further investigate interactions between E. amylovora and E. pyrifoliae, including competition and population dynamics within the host; such research could be facilitated by duplex ddPCR.
Present study, we validated a duplex-ddPCR assay for the quantification of E. amylovora and E. pyrifoliae. The system yielded highly sensitive and quantitative results, offering a fast, reliable, and visual method for the detection of fire blight and black shoot blight. This tool may be applied to investigate the behavior of these two Erwinia species, particularly at low levels of infection, and monitor their population dynamics, either in co-culture or during coinfected in the same plant.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation
Print
