Fine Tuning of Real Time PCR as a First Tool for the Detection of G143A Substitution in Venturia inaequalis Samples ()
1. Introduction
Apple scab, caused by Venturia inaequalis (Cooke) Winter, is the main disease in almost all apple growing areas in the world, and in Italy requires numerous fungicide applications (from 10 to 20 and more) especially in Northern areas because of frequent rain and fairly high temperatures in the early growth stages.
Old protectant fungicides (dithiocarbamates, dithianon, captan and dodine) have an important role in apple scab control. In the early 1970s, the most important fungicides for scab control were the benzimidazoles, but in a few years they were dramatically affected by resistance development. This led to a return to the use of dodine until the late 1970s, when fenarimol’s registration started the Sterol Biosynthesis Inhibiting (SBI) fungicide period. These fungicides have been used for many years with triazoles but they were progressively affected by resistance from the middle of the 1980s to nowadays [1].
An improvement of control was made possible by anilinopyrimidines introduced in the middle of the 1990s and by strobilurins in the late 1990s. Strobilurin fungicides, belonging to the group of Quinone outside Inhibitors (QoIs), are highly effective against a wide range of fungal pathogens. The mode of action of QoIs consists in inhibiting mitochondrial respiration by binding at the so-called Qo site of cytochrome b. Cytochrome b is part of the cytochrome bc1 complex (or complex III), located in the inner mitochondrial membrane of fungi and other eukaryotes [2] [3]. Strobilurins, first as kresoxim-methyl and later as trifloxystrobin, permitted improvements in apple scab control because they made it possible to extend spray intervals between treatments even up to ten days [4].
The first occurrences of resistance to QoIs, also with a reduction in performance, were reported for several pathogens, including V. inaequalis [5] [6] [7]. The first mechanism of resistance observed in many phytopathogenic fungi was a single-nucleotide mutation in the cytochrome b (cytb) gene, leading to amino acid exchange [8]. Three of the most common amino acid substitutions were detected in several phytopathogenic fungi and Oomycetes that are resistant to QoIs: from glycine to alanine at position 143 (G143A), from phenylalanine to leucine at position 129 (F129L) and from glycine to arginine at position 137 (G137R) [9].
Sierotzki an collaborators detected for the first time the G143A substitution in cytochrome b gene in QoI-resistant field isolates of plant pathogens [10]. This target site mutation has been identified in more than twenty species, including phytopathogenic Ascomycetes such as V. inaequalis [11], several powdery mildews [10] [12] and Alternaria species [6] [13] [14] [15]. This mutation was able to cause severe decreases of sensitivity (Resistance factor [RF] often higher than 100) and field control failure, while the other detected mutation, F129L, determined moderate effects on sensitivity and field performances. The V. inaequalis situation was more complex because the high reduction of V. inaequalis population sensitivity was not always related to the detected mutations or the field control reductions [16] [17] [18]. In these situation, the alternative respiration pathway was assumed to have an important function, however its in planta activity has not been clarified [19]. Consequently, the results on V. inaequalis were conflicting, showing different levels of inhibition from very low to high [20]. Another resistance mechanism; an external esterase able to metabolise the fungicide conferring a partial loss of control of kresoxim-methyl was reported [21]. However, studies proved that it was much less effective in the case of other strobilurins [18].
In Italian apple areas, a reduced activity of QoIs on apple scab has been suspected since 2000 and 2001 [22]. The field control complaints made by farmers in the last few years in Italy led to reconsider the application of these fungicides. Strobilurins were abandoned in some areas, especially where control failures were rather frequent, or utilized only in low risk periods, while in other areas they continue to be applied with the usual anti resistance recommendations. Some authors reported the presence of resistant populations of V. inaequalis [23] [24] and they defined in vitro values for resistant populations similar to the one defined in our previous work [22]. Moreover, Fontaine and collabrators [24] were in agreement with our previous study supporting the concept that the G143A substitution was almost always present in resistant populations and sometimes also in sensitive ones [25]. Even when it is present at very low frequencies, the mutation can be identified utilizing quantitative Real-Time PCR [26]. Once the resistance mechanism is known, it is often faster to use molecular techniques compared to biological assays. The in vitro methods are widely used for sensitivity screening among fungal populations. However, they are labor-intensive and time-consuming; moreover, only a part of the samples from orchards (generally no more than 50% - 60%) have conidia with a sufficient viability to be multiplied in order to carry out the in vitro assays. The main advantage of qPCR methods is their high sensitivity and cultivation-independency, in fact, qPCR allows the detection of small amounts of the desired allele in bulk samples, thus it eliminates the need to maintain pure cultures of the pathogens.
The allele-specific qPCR assay was successfully applied for QoI-resistant allele detection in response to QoI applications. In particular, the qPCR method has been applied for quantification of G143A substitution that confers QoI resistance in Alternaria spp. [27] Mycosphaerellagraminicola [28], Plasmoparaviticola [29] [30] Erysiphenecator and Pyrenophora teres [31] [32].
Qualitative or semi-quantitative techniques such as PCR-RFLP [33] allele-specific PCR (AS-PCR) and cleaved amplified polymorphic sequence (CAPS) assay were also used for detecting V. inaequalis resistance [24] [34]. Quantification of QoI resistant alleles in V. inaequalis populations has been developed [35] [36] [37] [38]. Our previous studies [22] [25] highlighted the occurrence of QoIs resistance in Italian V. inaequalis populations sampled from 2002 to 2005 with the presence of G143A substitution. The aim of the present work was to definitely substitute the labor intensive biological assays with a molecular one. In order to suggest the qPCR method as unique, robust and consistent, a large monitoring sampling V. inaequalis populations from 2002 to 2010 was carried out. A qPCR protocol was designed to quantitatively determine the frequency of G143A mutation and related to in in vitro sensitivity test performed to identify the QoIs resistance.
2. Materials and Methods
2.1. Origin and Maintenance of Venturiainaequalis Populations
Forty-nine bulk V. inaequalis populations were sampled in Northern Italy orchards during 2002-2010. V. inaequalis populations were collected in apple trees with different scab management: wild-type trees (WT), which have never been treated with fungicides and located in uncultivated areas of Northern Italy; untreated trees (UNT), which have never been exposed to QoIs; commercial orchards where several groups of fungicides, often including strobilurins, have been applied with good disease control (GC) or poor disease control (PC) (i.e. with practical resistance to QoI fungicides). Samples consisted of 40 - 50 scabbed leaves randomly collected in each orchard from May to July.
In order to obtain populations with high viability, a drop of sterile water was put on 30 - 40 scab lesions from each sample and conidia were harvested by rinsing with sterile water. The conidial concentration was adjusted to (1 − 3) × 105 spores/mL and inoculated on potted apple seedlings. Inoculated seedlings were incubated for 15 - 20 days at 21˚C in a greenhouse. All scabbed seedling leaves were then conserved in silica gel at 4˚C for fungicide sensitivity assays and molecular analysis. In addition, scabbed-infected leaf discs obtained by a sterile cork borer ( 9 mm Æ) from 15 - 20 leaves for each population were stored at −20˚C until DNA extraction.
2.2.In Vitro Fungicide Sensitivity Assays
Sensitivity of V. inaequalis to trifloxystrobin (Sigma-Aldrich, St Louis, MO, USA) was determined by spore germination assay. Trifloxystrobin was solubilized in acetone and added to water-agar (2% agar grade A; Becton, Dickinson and Company) to achieve the following concentrations: 0.001, 0.01, 0.1, 2 mg/L. The antibiotic streptomycin sulfate (Sigma-Aldrich) was added to water agar at a final concentration of 200 mg/L. In each in vitro test, pieces of scabbed leaves (0.012 - 0.020 g ) were randomly collected from each sample and introduced in micro tubes with 1 mL of sterile water. After shaking, the conidial concentration was adjusted to (1 − 3) × 105 spores/mL. Two drops of 20 µL of spore suspension were placed on agar plates for every fungicide concentration and incubated for 24 h at 20˚C. 150 conidia per replicate were visually assessed using a microscope, counting those that had germinated. Two in vitro tests were performed for each population. The resistance response for each population was expressed as mean percent relative germination “RG%” (percentage of germination at 2 mg/L/percentage of germination in untreated × 100), as previously described [39].
2.3. DNA Extraction
The fungal genomic DNA was extracted directly from lyophilised scabbed leaves which were ground in liquid nitrogen by a CTAB-based method and then purified through a Sepharose 6B (Sigma-Aldrich). Amount, purity and integrity of DNA samples were assessed on the basis of an absorbance ratio of 1.80 - 1.90 at 260/280nm and of 1.90 - 2.30 at 260/230nm using NanoQuant (Tecan, Grödig, Austria).
2.4. Cloning of Cytochromeb Gene of Sensitive and Resistant Strains of Venturiainaequalis
A 413 bp fragment of cytochromeb gene was amplified. Resistant and sensitive references were obtained from sensitive and resistant monoconidial isolates of V. inaequalis. CleavedSequencing reactions were done by Sanger methods using gene-specific primers (BMR Service, Padova, Italy). A BLASTN search was carried out in the NCBI database in order to confirm the strains. After PCR, excess primers were removed using GenElute PCR Clean-up kit (Sigma-Aldrich) and the species-specific PCR product ligated directly into the pGEM-T easy vector (Promega, Madison, WI, USA). Plasmids were transformed into Escherichia coli JM109 cells (Promega) according to a standard protocol, plasmid DNA was extracted using Quick Plasmid Miniprep Kit (Invitrogen, Carlsbad, CA), according to a manufacturer’s protocol.
2.5. Primers Set-Up
The G143A substitution in V. inaequalis population was detected using the qPCR method. A specific primers set was designed. The forward primers FwS5418 and FwR5418 were used respectively for sensitive (S) and resistant (R) alleles, while the reverse common primer, R5548U, was used (Table 1). With the aim of verifying the qPCR efficiency (E) and specificity of each primer pair, different percentages of reference sensitive and resistant plasmidial DNA were analyzed: 100% S, 1% S + 99% R, 5% S + 95% R, 10% S + 90% R, 50% S + 50% R, 99% S + 1% R, 100% R (Figure 1). Even when it is present at low frequencies, the mutation can be identified utilizing quantitative Real-Time PCR. The ratio (%) of the two alleles present in samples was calculated using the ΔCq method according to the equation [40] 10^ (Cq S allele − Cq R allele)/-slope); R (R/1 + R) × 100 = R allele; where Cqis the quantification cycle and reflects the cycle number at which the fluorescence generated within the reaction of each sample crosses the threshold and the slope was calculated according to the standard curves, which were specific for each allele. The data are expressed as percentages of mutated allele.
![]()
Table 1. Primer sequences selected for qPCR analysis.
![]()
Figure 1. Specificity of Fw S5418/R5548U (for wild type allele) and Fw R5418/R5548U (for mutant allele) primers tested according to different mix concentrations of plasmidial DNA template obtained from monoconidial isolates of V. inaequalis. The ratio (%) of the two alleles’ presence in samples was calculated using the ΔCq method (see Material and Methods section).
2.6. Standard Curves Development
Standard curves were obtained mixing S and R alleles strains (50% S and 50% R) 10-fold serial dilutions ranging from 20 ng × 10−3 to 20 ng × 10−6 of plasmid DNA. The standard curve efficiencies were: S allele, E = 101.9%, R allele, E = 100.1%. The high efficiency for each gene allowed the assumption that the genes are amplified with the same efficiency, and an average slope of −3.329 was used in the equation.
2.7. Quantitative Real-Time PCR Conditions
The qPCR analysis of the forty-nine V.inaequalis populations was performed using an ICycler-IQ5 (Bio-Rad). The reaction mixture was prepared in a final volume of 20 μL, including 10 μL of 2 × SYBR Green I (Bio-Rad), 1 μL of forward and reverse primers (12.5 μM each), and 8 μL of diluted DNA (1/5). Amplifications were performed in 96-well Hard-shell PCR plates (Bio-Rad). The following thermal cycling conditions were used: one cycle at 95˚C for 3 min followed by 40 cycles at 95˚C for 18 s, 55˚C for 12 s, and 72˚C for 18 s.
To verify the specificity of the product obtained, a melting curve was performed at the end of the PCR reaction with an increase of the temperature specificity of 0.05˚C/s, from 55˚C to 95˚C.
For each population, the specific primers for both QoI R and S alleles were analysed in each plate. The assays were carried out in duplicate, each experiment was repeated three times and the means comparisons was conducted. Data were calculated using the supplied Real-Time Detection System software version 3.0. for Windows.
3. Results
3.1. Sensitivity Assays
Eight WT and one UNT populations were found to be sensitive to trifloxystrobin with RG values from 0 to 9.2% (Table 2, Column B and Figure 2). Our
![]()
![]()
Table 2. Results of biological and molecular assays.
aWT: wild-type; UNT: untreated; NOS: no strobilurins used; GC: good control and use of strobilurins; PC: poor control and use of strobilurins; -: management characteristics not available. bStrobilurin sensitivity: S= sensitive EC50 < 0.065 mg/L; R = resistant EC50 > 0.12 mg/L. cMean percent relative germination (%RG) on medium amended with technical grade trifloxystrobin at 2 mg/L relative to that on non-fungicide amended medium. dFrequency of the resistant allele (A143) of populations determined by qPCR.
![]()
Figure 2. Effect of trifloxystrobin on spore germination of V. inaequalis. (a) Sensitive population (N.62) and (b) resistant population (N.87) on water agar amended with 2 mg/L of trifloxystrobin.
results concur with the reports of others [17]. Populations obtained from good control orchards had RG ranging from 0 to 14.6% (Table 2, Column B). They often showed slightly low sensitivity with respect to wild types but they are also considered sensitive.
Thirty-two populations sampled mainly in orchards that were reported to have poor control (PC) by strobilurins demonstrated resistance with a wide range of RG values from 15.4 to 97.3% (Table 2, Column B and Figure 2). As regards the further four populations, three from treated orchards without use of strobilurins in the sampling year (No 201, 600, BA3) and one from an orchard with unknown field management (No 612), they showed low sensitivity with mean RG values ranged from 14.6% to 71.8%.
3.2. Standard Curves and Primer Set up for qPCR
The qPCR analysis showed a specific amplicon using the primer-set for both R and S alleles. The melting curve analyses for all amplicons showed a single peak, and no non-specific products or primer-dimer formation were detected (data not shown). The standard curves obtained using mixed strains (50% S + 50% R) in 10-fold serial dilution showed good efficiency (E) with these primer sets (FwS5418/5548URev), (FwR5418/5548URev) for qPCR ranging from E = 100.1% to E = 101.9%, with a linear correlation coefficient of 0.999 (Figure 3). The high efficiency for each gene allowed the assumption that the genes are amplified with the same efficiency, and an average slope of −3.329 was used in the equation.
The specificity primer tests were carried out utilizing FW S5418/R5548U and FW R5418/5548U primer sets (Figure 1).
Successful allele discrimination was achieved and an excellent reliability related to the ΔCq calculation method was observed.
3.3. Relation between R-Allele Frequencies and Relative Germination (RG)
In vitro resistance response to trifloxystrobin was expressed as RG. The populations were classified as sensitive with RG < 14.6% and they were found to contain mutated alleles from 0.015% to 21.32%. Referring to WT and UNT populations, when the relative germination is 0%, the mutated alleles generally ranged from 0% to 2.28%. Sometimes there was no clear relationship between the frequency of A143 allele and RG. For instance, when populations (No 426, 158,
![]()
Figure 3. Standard curve with both S5418 sensitive primer (a) and R5418 resistant primer (b) qPCR Standard curve: Ten-fold dilutions obtained using mixed strains (50%S + 50%R) of plasmidial DNA ranging from 20 ng × 10−3 to 20 ng × 10−6 were used, as indicated on the x-axis, whereas the corresponding cycle threshold (Cq) values are presented on the y-axis. Each dot represents the result of triplicate amplifications of each dilution. The correlation coefficient and slope values of the regression curve were calculated.
228, 226) had RG ranging from 0% to 2.4%, the resistant allele frequency was 17.55% to 21.32%, the same as a population sampled from the orchard where strobilurin gave good disease control and had high R-allele frequency (84.5%) (No 427).
Starting from No 319, the populations obtained from orchards that were reported to have poor control (practical resistance) were defined as resistant by conidial germination test and showed progressively high levels of RG (from 15.4% to 97.3%) (Table 2, Column B). They exhibited the G143A substitution with a higher R-allele frequency, compared to the wild type populations, ranging from 23.01% to 99.54%. Only two cases (Nos. 201 and 161) were outside this range and, the R-allele frequency was much lower (0.015% and 5.97%) (Table 2, Column C).
4. Discussion
In this study, an allele-specific qPCR method with a newly designed primer set was successfully developed to quantitatively determine the frequency of QoI-resistant A143 allele in populations of V. inaequalis in order to substitute the biological assays. In order to be able to suggest only this methodology for the monitoring of V. inaequalis we determined how the frequency of mutated alleles relates to the strobilurin resistance, and fundamental was the relationship between in vitro sensitivity tests and molecular assay results.
The results of conidial germination assays (RG) demonstrated differences in sensitivity to strobilurins among populations collected from orchards with different fungicide use and different disease control by these fungicides. In sensitive populations, RG detected within wild-type and well-controlled populations were lower than < 14.6%. Additionally, in most cases, the A143 allele frequencies were lower than 2.28% while RG was 0%. By contrast, frequency distributions of the mutated allele of resistant populations were variable. In fact, in some cases, a high in vitro sensitivity with low RG corresponded to a moderately high detection of R-allele frequency in four wild type populations (Nos. 426, 158, 226, 228) and one well controlled (No 427) population. Detection of R-alleles in wild type populations may be explained by the natural flow of spores from strobilurin-treated orchards to wild-types, with the spores travelling up to 15 kilometres. It is important to notice that the resistant one can reproduce and spread over the sensitive populations, and this can lead to an increase of resistant isolates [35] [41] [42]. This indicates, as suggested by Fraaije et al., [28] that isolates with A143 alleles might play an important role in long-distance dispersal of QoI-resistant genotypes. Moreover, the Fungicide Resistance Action Committee (FRAC) has also shown that it is possible to detect G143A substitutions in fungal populations never exposed to QoI fungicides. An explanation for this phenomenon is, that mutations occur naturally in fungal populations, usually at low frequencies [43]. Mutations can also be found, at low, but variable, frequencies, in situations where the use of QoI based products is providing perfectly acceptable disease control. Furthermore, it should be clarified why mutated alleles with 17.55% to 21.32% frequencies, especially in wild types, is not reflected in in vitro sensitivity tests. This case can be explained by the detection of mutated allele in DNA from both non-viable and viable fungal material using qPCR.
The thirty-two populations collected from orchards where strobilurins were applied and poor control was noticed showed lower sensitivity to trifloxystrobin and they presented RG from 15.4% to 97.3%, therefore they were classified as resistant. It was observed that populations defined as resistant (PC, NOS,) in in vitro tests almost always had high mutated allele frequency (from 23.01% to 99.54%). However, two populations, No 161(PC) and No 201(NOS) associated with low sensitivity responses in vitro had a low relative abundance of the A143 allele (5.97% and 0.015%). This difference may be explained by the presence of other mechanisms causing reduced sensitivities. Indeed, the mutation could not always explain the QoI-resistant phenotype, this it t may be due to the alternative respiration pathway but this phenomenon has been considered of little importance under field conditions [44]. Moreover, the other effective mechanisms responsible for QoI resistance are still to be characterized and could be clarified with studies on biochemical and genetic aspects [20] [44]. Many basic aspects of QoI resistance (e.g. genetic stability, segregation) are still not well understood [45].
In this study, quantitative assessments using qPCR followed a similar pattern to that obtained using in vitro conidial germination tests in very sensitive and very resistant populations. In fact, in most cases, it was noted that when RG was <10%, mutated allele frequency was <10% and when RG was >70%, very high mutated allele (>80%) was detected. Similar results were observed in a large number of monoconidial isolates (data not shown). However, some variability between the two tests was observed in heterogeneous populations (presence in the samples of leaf lesions colonised by either mutant or wild type isolates and/or presence of heteroplasmic fungal strains containing mitochondria with or without the G143A substitutions in the leaf samples.
The results demonstrated that a high relative abundance of the A143 allele was associated with high resistance to QoIs. Furthermore, in the current study, the A143 allele was higher than 23% in the orchards, where QoIs failed to control apple scab. Similarly, Villiani et al., [36] reported that in the US, commercial orchards with practical resistance to QoI fungicides had a relative abundance of the A143 allele > 22%.
The qPCR method is able to determine the development of fungicide resistance in apple scab populations and it is a fundamental step in being able to predict possible trends in fungicide resistance on a large scale. More often, it is necessary to detect the presence of the pathogen early and quickly, and to determine the frequencies of resistant and sensitive isolates in one or more regions. The most commonly used test for determining QoI sensitivity of V. inaequalis is a biological test, often conducted at several concentrations and which is very complex, this conventional methods to detect resistance are time-consuming and labor-intensive. Additionally, culturing of V. inaequalis is a difficult step because the growth is very slow on culture medium and cross contamination of other fungi occurs very often. The method developed here was designed as a substitution to the traditional method in vitro sensitivity test.
5. Conclusion
In conclusion, with this work, after processing a large amount of data with different scab management, we can conclude that qPCR analysis could be performed alone without the biological test especially with a high number of samples in order to efficiently and quickly determine the presence of mutated alleles to identify the practical resistance risk in the orchards, with the certainty that the results are consistent and close to the field situation.
Acknowledgements
We thank Dr. Riccardo Fiaccadori (University of Bologna) for providing some important strains of V. inaequalis and sharing his experiences in bioassays.
This study was partially funded by the Emilia Romagna region within the Rural Development Plan 2014-2020 Op. 16.1.01—GO PEI-Agri—FA 4B, Pr. 5004934 and coordinated by CRPV, the authors are thankful for this financial support.