SEARCH

SEARCH BY CITATION

Keywords:

  • Botryotinia fuckeliana;
  • cytochrome b;
  • fluorescent probe;
  • grey mould;
  • QoI fungicides;
  • real-time PCR

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Botrytis cinerea field isolates collected in Japan were screened for resistance to Qo inhibitor fungicides (QoIs). Of the 198 isolates screened, six grew well on a medium containing azoxystrobin, a QoI, when salicylhydroxamic acid, an alternative oxidase inhibitor, was present. The resistance mutation in the cytochrome b gene (cytb) was characterized. All QoI-resistant isolates had the same mutation (GGT to GCT) in cytb that led to the substitution of glycine by alanine at position 143 of cytochrome b, which is known to confer QoI resistance in plant pathogens. To detect this mutation, a hybridization probe assay based on real-time PCR amplification and melting curve analysis was developed. Using DNA samples prepared from aubergines coinfected with QoI-resistant and QoI-sensitive B. cinerea isolates, two similar peak profiles with their corresponding melting temperatures were obtained. This result suggests that QoI-resistant and QoI-sensitive isolates may compete equally in terms of pathogenicity, and the assay may be used to assess the population ratio of mutant and wild-type isolates. However, the hybridization probe did not anneal to PCR products derived from the DNA samples of some QoI-sensitive isolates. Structural analysis of cytb revealed that B. cinerea field isolates could be classified into two groups: one with three introns and the other with an additional intron (Bcbi-143/144 intron) inserted between the 143rd and 144th codons. All 88 isolates possessing the Bcbi-143/144 intron were azoxystrobin-sensitive, suggesting that the QoI-resistant mutation at codon 143 in cytb prevents self-splicing of the Bcbi-143/144 intron, as proposed in some other plant pathogens.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Qo inhibitor fungicides (QoIs) such as azoxystrobin and kresoxim-methyl have proven to be very effective in controlling many plant diseases (Bartlett et al., 2002). QoIs inhibit respiration by binding to the Qo site of the cytochrome bc1 enzyme complex (complex III) (Von Jagow et al., 1986). Cytochrome b is a membrane protein that forms the core of the mitochondrial bc1 complex in the respiratory chain, and the gene of this protein is localized in the mitochondrial genome. Single amino acid substitutions in cytochrome b that confer resistance to QoIs have been found in different plant pathogens, including Alternaria alternata, Colletotrichum graminicola, Blumeria graminis, Pyricularia grisea, Pythium aphanidermatum, Podosphaera fusca, Pyrenophora teres and Pseudoperonospora cubensis (Sierotzki et al., 2000, 2007; Ishii et al., 2001; Gisi et al., 2002; Avila-Adame et al., 2003; Kim et al., 2003; Ma et al., 2003). In Saccharomyces cerevisiae, several target-site mutations of cytochrome b that confer resistance to QoIs have been described (di Rago et al., 1989; Fisher et al., 2004). However, few cases of QoI-resistant mutations have been found in field isolates of plant pathogens. In most cases, resistance was conferred by a single point mutation in the 143rd codon of the cytochrome b gene (cytb) (GGT to GCT, G143A mutation) that led to the substitution of glycine by alanine (G143[RIGHTWARDS ARROW]A). In some species, such as Pyricularia grisea and Pythium aphanidermatum, another amino acid substitution (F129[RIGHTWARDS ARROW]L) also conferred QoI resistance, even though the resistance level was lower than that conferred by the G143[RIGHTWARDS ARROW]A substitution (Gisi et al., 2002; Kim et al., 2003). Molecular techniques for the detection of QoI resistance have been reported in several plant pathogens and include PCR-restriction fragment length polymorphism (RFLP) in P. grisea and P. cubensis and allele-specific (AS)-PCR using real-time PCR in Alternaria spp., B. graminis and Pyrenophora teres (Fraaije et al., 2002; Kim et al., 2003; Ma & Michailides, 2004; Ishii, 2006; Kianianmomeni et al., 2007). Some plant pathogens have been shown to contain an intron between the 143rd and 144th codons (bi-143/144). QoI-resistant field isolates with the G143A mutation have been found only in species without the bi-143/144 intron, and QoI-resistant mutants with G143A have never been found in species possessing the intron, such as A. solani. This was recently explained as the prevention of proper splicing of the intron and maturation of cytb mRNA as a result of the presence of G143A (Grasso et al., 2006; Sierotzki et al., 2007).

Grey mould caused by Botrytis cinerea (teleomorph Botryotinia fuckeliana) is an economically important disease in many kinds of vegetables, fruits and ornamental plants. Control of grey mould has been adversely affected by the development of fungicide resistance. Botrytis cinerea strains resistant to benzimidazoles and dicarboximides are known to be distributed worldwide (Leroux et al., 2002; Ma et al., 2007; Banno et al., 2008) and laboratory mutants resistant to QoIs were recently isolated and characterized (Markoglou et al., 2006). A rapid and convenient method to detect resistance mutations is vital for managing fungicide resistance, for example, in the evaluation of the inherent risk of QoI-resistant B. cinerea present in the field.

The purpose of this study was to screen for QoI-resistant isolates in Japanese fields. A mutation causing G143[RIGHTWARDS ARROW]A substitution in cytb was identified as a QoI-resistant mutation. To analyse the resistant isolates, a hybridization probe assay was developed to detect the G143A mutation using real-time PCR. Two adjacent oligonucleotide probes whose fluorescent labels communicate through fluorescence resonance energy transfer were used to detect any change in the nucleotide sequence of the template DNA. This assay was useful not only for the detection of the QoI-resistant mutation, but also for the pathogenicity competition assay between the QoI-sensitive and QoI-resistant isolates. The cytb gene structure in B. cinerea was compared and two types of isolate, with or without the Bcbi-143/144 intron in cytb, were identified.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Isolates and culture conditions

A total of 198 field isolates of B. cinerea were used in this study: 64 isolates were previously collected from 10 fields in Osaka Prefecture in 2004 (40 isolates) and 2005 (24 isolates) (Banno et al., 2008), while 134 isolates were newly collected from 41 fields in Kanagawa Prefecture in 2005 (103 isolates) and 2006 (31 isolates). Four strains were previously isolated from Japan during 2000. Of these, Bc-o-7, Bc-o-12 and Bc-o-26 were isolated in Osaka Prefecture, and Bc-56 was isolated in Gunma Prefecture (Oshima et al., 2006). These four were used as the QoI-sensitive isolates and were obtained from aubergine, tomato, cucumber, strawberry or eustoma plants. Botrytis cinerea isolates were maintained on potato dextrose agar (PDA) after single colony isolation.

Fungicide sensitivity assays

To assess the sensitivity of hyphal growth to fungicides, mycelial disks (5 mm in diameter) precultured on PDA medium were placed on PDA amended with azoxystrobin (Amistar 20% FL, Syngenta) at 0·008, 0·04, 0·2, 1, 5 or 25 µg mL−1, kresoxim-methyl (Stroby 41·5% FL, BASF Agro) at 0·008, 0·04, 0·2, 1, 5 or 25 µg mL−1, carbendazim (Wako) at 0·1 or 25 µg mL−1 or iprodione (Wako) at 0·4 µg mL−1. These discs were then cultured at 18°C for 3 days. Sensitivity to QoIs, azoxystrobin or kresoxim-methyl was tested in the presence of salicylhydroxamic acid (SHAM) at 50 µg mL−1. The mean colony diameter minus the diameter of the inoculation disk was then measured and expressed as a percentage of the mean colony diameter of the untreated control.

DNA extraction and sequencing

The genomic DNAs of the B. cinerea isolates were extracted from the mycelia cultured on PDA, as reported previously (Banno et al., 2008). Mycelia were collected and homogenized with a bead beater (FastPrep instruction, Qbiogene). After phenol extraction and isopropanol precipitation, genomic DNA was obtained, which was used as the template for PCR or real-time PCR. Alternatively, genomic DNA was extracted from spores (approximately 106) using the FastDNA kit (Qbiogene), according to the manufacturer's protocol. The sequence primers were designed based on the sequence of the cytochrome b gene (cytb) (accession no. AB262969) of B. cinerea (Table 1). The full-length cytb gene was amplified by PCR using the primers Bccytb-F4 and Bccytb-R2. The Bccytb-F1 and Bccytb-R1 primer set was used to confirm the mutation at the 143rd codon in cytb. For the sequencing reactions, amplified PCR fragments were purified using the QIAquick PCR Purification Kit (Qiagen) and used as templates. The DNA sequence was determined using an ABI Model PRISM310 Auto Sequence System (Applied Biosystems) and fluorescent dye terminator dideoxynucleotides, according to the manufacturer's PCR cycle sequencing protocol.

Table 1.  Sequence of primers and fluorescent probes for cytochrome b gene of Botrytis cinerea used in this work
Probe and primerSequence (5′[RIGHTWARDS ARROW]3′)aLocation in the cytb gene
  • Changes in nucleotide from the genome sequences are denoted in lower-case letters.

  • a

    FL, fluorescein-labelled; RED, LC-Red-640-labelled.

  • b

    Nucleotide position of the cytochrome b gene (cytb) of B. cinerea without the Bcbi-143/144 intron. Base A of the initiation codon of cytb was numbered 1, corresponding to base 455 of the sequence designated AB262969.

  • c

    Nucleotide position of cytb of B. cinerea with the Bcbi-143/144 intron. Base A of the initiation codon of cytb was numbered 1, corresponding to base 304 of the sequence designated AB428335. The Bcbi-143/144 intron was localized at positions 3389–4593. See Fig. 4 for more detail.

Bccytb-F1CGTCGGCCATATAAAAGGTC3160–3179b
Bccytb-F2ATTAAAGTTGTTTGTACGGATAAGTT3318–3343b
Bccytb-F3ATTCTAcCTTATTCTACAG3288–3306b
Bccytb-F4GACCGAATGGTGGGATCAAT−317 to −298b
Bccytb-F5GCCTAACGTATTAGGAGATAG4971–4991b
Bccytb-S-F01TGATATAGTCCATCCTCTAC349–368b
Bccytb-S-F02TGTAGGGTTTATAGAAGGTG979–998b
Bccytb-S-F03GTCTGATAGTTCTAAGGTTG1543–1562b
Bccytb-S-F04GGGATATTGTACAAGTTGTG2091–2110b
Bccytb-S-F05TCACCTGAAACAATAGAAAG2638–2657b
Bccytb-S-F06TACCGTGATAACAGAAATCC3755–3774b
Bccytb-S-F07AGATAATTTGTGAAGGCTCC4394–4413b
Bccytb-R1CTCCATCCACCATACCTACA3604–3623 (3′[RIGHTWARDS ARROW]5′)b
Bccytb-R2AATCCGAGATACCAGTAGCG5505–5524 (3′[RIGHTWARDS ARROW]5′)b
Bccytb-HPF1GTACAGGCGGAACTTTTAGTTA3224–3245b
Bccytb-HPR1GGTACAGCACTCATAAGATTTG3402–3423 (3′[RIGHTWARDS ARROW]5′)b
Bccytb-R-An1TGTACGGATAAGTTTGTATTTTGTATGTTCTGCCCTACGG-FL3330–3369b
Bccytb-R-Sn1RED-CAAATGTCACTGTGAGcTGCTA3371–3392b
Bccytb-in-F1GTCACTGTGAGcTTTTTTTGAGC3376–3398c
Bccytb-in-R1TGGAATACGTTTAGCCATTTGAT3507–3529 (3′[RIGHTWARDS ARROW]5′)c
Bccytb-in-R2GCTCAAAAAAAgCTCACAGTGAC3376–3398 (3′[RIGHTWARDS ARROW]5′)c
Bccytb-in-S-F01CCAGCTTTAATGATATCCAC3801–3820c
Bccytb-in-S-F02AGTGCTTAATGTACAGTCGG4437–4456c

Fluorescent probe assay for cytb genotyping using real-time PCR

The QoI-resistant mutation (G143A) was detected by using a LightCycler (LC) instrument (Roche Diagnostics) with hybridization probes (Table 1). Real-time PCR with the fluorescent probes was performed in glass capillaries in a total volume of 20 µL in an LC system (Roche Diagnostics). The hybridization probes consisted of two different oligonucleotides (Bccytb-R-Sn1 and Bccytb-R-An1), one labelled at the 5′-end with the acceptor fluorophore LC-Red 640 and the other labelled at the 3′-end with the donor fluorophore fluorescein. When these probes hybridize to two adjacent internal sequences of PCR-amplified products and the light source of the LC excites fluorescein, fluorescence resonance energy transfer occurs between fluorescein and LC-Red 640. This results in the emission of fluorescence, which is measured by the instrument. Mutations can thus be determined by analysis of melting curves. The reaction mixture contained an appropriate amount of genomic DNA, 1 × LC DNA Master Hybridization Probes mix (Roche Diagnostics), 3 mm of MgCl2, 0·5 µm primers, 0·2 µm fluorescein-labelled probe and 0·25 µm LC-Red 640-labelled probe. The PCR amplification comprised an initial denaturation step (95°C for 30 s) followed by 50 cycles of amplification (95°C for 10 s, 60°C for 10 s and 72°C for 8 s) and a melting curve step (denaturation at 95°C for 2·5 min; annealing at 40°C for 1 min; and melting, i.e. increasing the temperature to 95°C in increments of 0·1°C s−1).

PCR-RFLP assay for cytb genotyping

The Bccytb-F2, Bccytb-R1 and Bccytb-in-R1 primers were designed to amplify two types of B. cinerea cytb, i.e. with or without the Bcbi-143/144 intron inserted between the 143rd and 144th codons. Multiplex PCR with these three primers was performed using a 2720 thermal cycler (Applied Biosystems). The PCR mixtures contained 0·4 µm forward primer and 0·2 µm reverse primers, 200 ng genomic DNA, 250 µm dNTPs mix (dATP, dCTP, dGTP and dTTP), 2·5 mm MgCl2 and 2·5 U Bio Taq polymerase (Bioline) in a total volume of 50 µL. After being preheated at 94°C for 8 min, the reaction mixtures were cycled 35 times at 94°C (1 min) and 60°C (25 s), and then followed by an extension at 72°C (20 s). After amplification, the PCR product was digested with 10 U AluI restriction enzyme and electrophoresed on a 2% agarose gel to determine whether the G143A mutation was present.

The following procedure was used to construct cytb with the Bcbi-143/144 intron and the G143A mutation. DNA fragments containing cytb with the intron were amplified from the genomic DNA of the QoI-sensitive Bc-o-26 isolate by PCR using two primer sets, viz. Bccytb-F3 and Bccytb-in-R2, and Bccytb-in-F1 and Bccytb-in-R1. Bccytb-in-R2 and Bccytb-in-F1 both have an AluI recognition site. Both PCR fragments (111 and 154 bp) were ligated after AluI digestion, and the resultant DNA fragments, which had a partial QoI-resistant cytb gene with the Bcbi-143/144 intron, were then used as templates for the PCR-RFLP assay.

Pathogenicity assays

To examine the pathogenicity of the B. cinerea isolates, mycelial disks (5 mm in diameter) were placed on detached second leaves of cucumber (cv. Sagamihanjiro), as described previously (Oshima et al., 2006) and incubated at 18°C in a humid, dark chamber. After 3 days, the spreading lesions of B. cinerea infection were photographed.

For the competitive infection assay between the sensitive and resistant isolates, an aubergine fruit was nicked in four places using a knife. Inoculation was carried out by placing two mycelial disks of each isolate in each nick and incubating the fruit at 18°C for 8–10 days in a humid, dark chamber. Spores produced on the infected fruit were collected in distilled water and their genomic DNA extracted. The ratio of resistant spores to sensitive ones was estimated by the hybridization probe assay for the G143A QoI-resistant mutation.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Isolation and characterization of the QoI-resistant field isolates of B. cinerea

The 64 isolates of B. cinerea from Osaka Prefecture (in 2004 and 2005) and the 134 isolates from Kanagawa Prefecture (in 2005 and 2006) were examined for QoI resistance. Azoxystrobin worked fairly well against ‘sensitive’ isolates such as Bc-56 and Bc-o-7 in terms of the inhibition of mycelial growth of B. cinerea on PDA medium at low concentrations (~1 µg mL−1); however, the QoI fungicide alone did not lead to complete inhibition of mycelial growth, even at 25 µg mL−1, and addition of 50 µg mL−1 of SHAM, an alternative oxidase inhibitor, was required to achieve this (Fig. 1). Among 198 B. cinerea field isolates, Bc-o-04-58 from Osaka (represented in Fig. 1), and Bc-05-k9, Bc-05-k38, Bc-05-k60, Bc-06-k10 and Bc-06-k18 from Kanagawa grew fairly well on the medium containing azoxystrobin (25 µg mL−1) and SHAM (50 µg mL−1). The resistant isolates showed similar levels of azoxystrobin resistance: their ED50 values (concentration required to inhibit radial growth by 50%) for azoxystrobin ranged from 19·3 to 21·0 µg mL−1 (Table 2). In contrast, ED50 values of sensitive isolates ranged from 0·07 to 0·64 µg mL−1. All azoxystrobin-resistant isolates grew on medium containing kresoxim-methyl (25 µg mL−1, also a QoI) and SHAM (50 µg mL−1), while the azoxystrobin-sensitive isolates did not (data not shown). All QoI-resistant isolates, with the exception of Bc-06-k18, were sensitive to iprodione and diethofencarb, but were resistant to carbendazim (Table 2). The BenA mutation of these isolates was identified by the genotyping assay described by Banno et al. (2008) and was shown to result in a single amino acid substitution (E198[RIGHTWARDS ARROW]A) in β-tubulin. Although these results suggest that the same strain spread in the fields, these isolates were obtained from different plants: Bc-05-k38, Bc-05-k60 and Bc-06-k10 were isolated from strawberry, while Bc-o-04-58, Bc-05-k9 and Bc-06-k18 were isolated from tomato, eustoma and cucumber, respectively, grown in different fields (Table 2). Moreover, Bc-06-k18 was resistant to both iprodione and carbendazim, and fungicide-resistant mutations caused more distinct amino acid substitutions in β-tubulin (E198[RIGHTWARDS ARROW]V) and BcOS1 histidine kinase (I365[RIGHTWARDS ARROW]S) than in the case of the other isolates (Table 2). When selected QoI-resistant isolates, Bc-o-04-58, Bc-05-k9 and Bc-05-k38, infected cucumber leaves, their lesions expanded equally, and their pathogenicity considered similar to that of the QoI-sensitive isolate Bc-o-7 (Fig. 2).

image

Figure 1. Mycelial growth of azoxystrobin-sensitive (Bc-56 and Bc-o-7) and -resistant (Bc-o-04-58) Botrytis cinerea isolates incubated on PDA medium containing 25 µg azoxystrobin mL−1 in the absence or presence of SHAM (50 µg mL−1) at 18°C for 60 h.

Download figure to PowerPoint

Table 2.  Fungicide resistance of Japanese Botrytis cinerea isolates
IsolateField (district)aHost plantYear Azoxystrobin ED50bcytb mutationcbi-143/144 intronSensitivity to carbendazimdBenA mutationeSensitivity to iprodionefBcOS1 mutationg
  • a

    Fields in the same district but affixed with different letters indicate closely located but different ones.

  • b

    Concentration that inhibited radial growth by 50% (ED50). Standard deviations in parentheses.

  • c

    A point mutation in the cytochrome b gene: G143A, GGT to GCT at codon 143.

  • d

    S, sensitive to 0.1 µg carbendazim mL−1; MR, resistant to 0.1 µg carbendazim mL−1 but sensitive to 25 µg carbendazim mL−1; R, resistant to 25 µg carbendazim mL−1.

  • e

    Point mutations in the β-tubulin gene: E198A, GAG to GCG at codon 198; E198V, GAG to GTG at codon 198; F200Y, TTC to TAC at codon 200.

  • f

    S, sensitive to 0.4 µg iprodione mL−1; R, resistant to 0.4 µg iprodione mL−1.

  • g

    Point mutations in the BcOS1 gene: I365S, ATC to AGC at codon 365; V368F, GTC to TTC at codon 368; Q369H, CAG to CAC at codon 369.

Bc-o-04-58Habikino-a (Osaka)Tomato200420·0 (2·0)G143AAbsentRE198AS
Bc-05-k9Hadano-a (Kanagawa)Eustoma200520·5 (1·9)G143AAbsentRE198AS
Bc-05-k38Hadano-b (Kanagawa)Strawberry200520·3 (3·6)G143AAbsentRE198AS
Bc-05-k60Isehara-a (Kanagawa)Strawberry200520·1 (0·3)G143AAbsentRE198AS
Bc-06-k10Atsugi (Kanagawa)Strawberry200619·3 (2·5)G143AAbsentRE198AS
Bc-06-k18Isehara-b (Kanagawa)Cucumber200621·0 (0·4)G143AAbsentRE198VRI365S
Bc-o-7Habikino-b (Osaka)Aubergine2000 0·07 (0·01)AbsentRE198VRI365S
Bc-o-12Tondabayashi (Osaka)Aubergine2000 0·13 (0·06)AbsentSRV368F, Q369H
Bc-05-k12Hiratsuka (Kanagawa)Tomato2005 0·55 (0·14)AbsentRE198VS
Bc-06-k22Kawasaki (Kanagawa)Tomato2006 0·23 (0·08)AbsentSS
Bc-56Itakura (Gunma)Cucumber2000 0·64 (0·25)PresentSS
Bc-o-26Tondabayashi (Osaka)Aubergine2000 0·16 (0·04)PresentSS
Bc-05-k88Fujisawa-a (Kanagawa)Tomato2005 0·36 (0·11)PresentMRF200YS
Bc-06-k56Fujisawa-b (Kanagawa)Tomato2006 0·21 (0·02)PresentRE198VRI365S
image

Figure 2. Pathogenicity of azoxystrobin-sensitive (Bc-o-7) or resistant (Bc-o-04-58, Bc-05-k9, and Bc-05-k38) Botrytis cinerea isolates on a detached cucumber leaf. After incubation on PDA medium at 18°C for 3 days, mycelial disks (5 mm in diameter) were placed on a detached first leaf and incubated at 18°C for 3 days in a humid, dark chamber.

Download figure to PowerPoint

The G143A mutation in cytb was reported to confer high resistance to QoIs in most plant pathogens (Gisi et al., 2002). Six QoI-resistant isolates of B. cinerea showed similar resistance levels to azoxystrobin. An approximately 0·5-kb region of cytb, containing the 143rd codon, was sequenced for four selected isolates, i.e. Bc-o-04-58, Bc-05-k9, Bc-05-k38 and Bc-05-k60, all of which were found to have a point mutation (GGT to GCT, designated the G143A mutation), as in the case of other plant pathogens (Table 2). No mutations were found at the 129th codon in cytb (TTC for phenylalanine) in these QoI-resistant isolates. This is the first report of QoI resistant field-isolates of B. cinerea.

Development of the hybridization probe assay to detect the G143A mutation

The BenA and BcOS1 genotyping assays were previously developed using hybridization probes to detect mutations for benzimidazoles and dicarboximides in B. cinerea (Banno et al., 2008). This method is based on PCR amplification and thermal analysis and is a reliable detection procedure that does not involve post-amplification steps such as restriction enzyme digestion and electrophoresis of the DNA fragment. To detect a point in the G143A mutation within cytb, hybridization probes and PCR primers were designed (Table 1). The Bccytb-HPF1 and Bccytb-HPR1 primers specifically amplified a 200-bp region in cytb that contained the 143rd codon. A sensor probe (Bccytb-R-Sn1, labelled at the 5′ end with LC-Red 640) could be hybridized to a region containing the mutation site, and an anchor probe (Bccytb-R-An1, labelled at the 3′ end with fluorescein) could be hybridized 1 bp away from the sensor probe (Fig. 3a). Therefore, two fluorophores were localized in sufficient proximity to allow fluorescence resonance energy transfer. After PCR amplification of the target sequence of cytb with the Bccytb-HPF1 and Bccytb-HPR1 primers, thermal analysis was performed using the QoI-resistant and QoI-sensitive cytbs. The resistant cytb that had the G143A mutation had a melting curve peak at 66·2°C, while the sensitive cytb had this peak at 57·1°C (Fig. 3b). The PCR fragments containing the G143A mutation (a perfect match with the sensor probe) formed more stable hybrids with the sensor probe and showed a higher melting peak temperature than the hybrids formed with the PCR products containing the wild-type sequence. This assay successfully distinguished all six QoI-resistant isolates from sensitive isolates such as Bc-o-7 and Bc-o-12, and the results indicated that all resistant isolates had the same G143A mutation in cytb.

image

Figure 3. Detection of the QoI-resistant G143A mutation in cytb of Botrytis cinerea using the hybridization probe assay. (a) DNA sequence of sensor probe and G143A mutations conferring QoI resistance in cytb. Hyphens (-) indicate nucleotides identical to those in the sensor probe. The sensor probe Bccytb-R-Sn1 was a perfect match to the cytb G143A gene, so the wild-type and G143A mutant cytb genes had a single mismatch, G[RIGHTWARDS ARROW]C. Codon 143 of the cytb gene is underlined. (b) Derivative melting curves [–(dF/dT) vs. T] of wild-type and QoI-resistant cytbs; fluorescence signal (F) was continuously monitored during the temperature ramp and then plotted against temperature (T). Melting curves were subsequently converted to derivative melting curves [–(dF/dT) vs. T]. Melting peaks of the wild-type cytb gene (broken line) and the G143A mutant gene (thick line) were at 57·1°C and 66·2°C, respectively. (c) Aubergine fruits coinfected with QoI-sensitive and QoI-resistant isolates. Mycelial disks (5 mm in diameter) of the sensitive Bc-05-k12 isolate and the resistant Bc-05-k9 isolate were inoculated onto the same fruits and incubated at 18°C for 8 days in a humid, dark chamber. Fruit on left: resistant isolate inoculated top and bottom and sensitive isolate inoculated in centre. Fruit on right: inoculated in opposite manner. (d) Competition assay between azoxystrobin-sensitive and -resistant isolates using hybridization probe. Genomic DNA was extracted from spores produced on the coinfected aubergines shown in (c), and this was used as template DNA for the hybridization probe assay. Melting peaks for the QoI-sensitive and QoI-resistant cytbs were at 57·1°C and 66·2°C, respectively. Thick line indicates the melting profile derived from the DNA sample prepared from the aubergine on the left; broken line indicates that obtained from the aubergine on the right.

Download figure to PowerPoint

This method was used to assess competitive ability in terms of pathogenicity. Mycelial disks of Bc-05-k9 (resistant) and Bc-05-k12 (sensitive) were placed on an aubergine fruit in order to trigger infection; after 8 days in a humid chamber these fruits were rotten and covered with spores (Fig. 3c). Genomic DNA was extracted from the spores produced on the coinfected aubergines, and this was then used for the hybridization probe assay (Fig. 3d). Two peaks corresponding to the two cytbs from the sensitive and resistant isolates were detected, and their signal intensities were found to be similar. A similar result was obtained with another combination: Bc-o-04-58 (resistant) and Bc-o-7 (sensitive) (data not shown). These results suggest that QoI-resistant isolates may be able to compete on the plant.

Gene structure of cytb in B. cinerea

The hybridization probe assay exhibited a clear peak at 66·2°C for all QoI-resistant isolates as described above; however, further analysis of sensitive isolates revealed that some of these, represented by Bc-56, Bc-o-26, Bc-05-k88 and Bc-06-k56 (shown in Table 2), did not exhibit any peaks in the melting analysis. This observation suggested that the structure of the probe-binding region of cytb in these sensitive isolates might be different from that in other sensitive isolates such as Bc-o-7, Bc-o-12, Bc-05-k12 and Bc-06-k22. Sequence analysis of the entire cytb of these eight isolates revealed that B. cinerea had two distinct cytb structures. Botrytis cinerea isolates represented by Bc-o-7 possessed three introns in cytb. These introns were designated Bcbi-67/68 (insertion between the 67th and 68th codons), Bcbi-131/132 (insertion between the 131st and 132nd codons) and Bcbi-164 (insertion in codon 164). In contrast, other isolates represented by Bc-o-26 had an additional 1205-bp long intron between the 143rd and 144th codons (designated the Bcbi-143/144 intron) in cytb (Fig. 4). It was assumed that this fragment belongs to the group-I intron family, in which the exonic base immediately upstream from the 5′ splice site is always T (U in pre-mRNA) and the base following the 3′ splice site is G, as reported in some other organisms (Cech, 1988). This intron contained an ORF that encoded a putative mRNA maturase, and its amino acid sequence was 52% and 41% identical to the mRNA maturases in the cytb introns of Puccinia triticina (DQ209276) and S. cerevisiae (AAB24190), respectively. Introns inserted between the 143rd and 144th codons in cytb were found in several fungi, such as A. solani, Puccinia spp., and Pyrenophora teres (Grasso et al., 2006; Sierotzki et al., 2007). The cytb sequence of the Bc-o-26 isolate with the Bcbi-143/144 intron (AB428335) was identical to that of the isolate without the Bcbi-143/144 intron (AB262969), except for the insertion of the Bcbi-143/144 intron. There was no polymorphism between two types of cytb gene.

image

Figure 4. Gene structures of mitochondrial cytb. (a) B. cinerea cytb without the Bcbi-143/144 intron (AB262969). (b) B. cinerea cytb with the Bcbi-143/144 intron (AB428335). (c) Puccinia triticina cytb (DQ209276). (d) Saccharomyces cerevisiae cytb (AJ011856). Boxes indicate exons, lines indicate introns. Lengths of exons and introns not to scale.

Download figure to PowerPoint

Botrytis cinerea isolates were classified into two groups with respect to the presence of the Bcbi-143/144 intron. One forward primer, Bccytb-F2, and two reverse primers, Bccytb-R1 (to detect isolates without the intron) and Bccytb-in-R1 (to detect those with the intron), were designed to amplify both types of cytbs. Mixtures of these three primers were used to amplify 306- and 212-bp fragments from cytb, without or with the Bcbi-143/144 intron, respectively (Fig. 5). AluI digestion of the 306-bp fragments, derived from cytb without the intron produced 27- (not shown), 69- and 210-bp fragments (Fig. 5). In contrast, the 212-bp fragments, derived from cytb with the intron and the artificially introduced G143A mutation (see Materials and methods), were digested into 69- and 143-bp fragments by the enzyme; however, the 212-bp fragment without the G143 mutation remained uncut. To analyse the B. cinerea population, the presence of the Bcbi-143/144 intron in the 198 field isolates was determined by multiplex PCR. The percentages of isolates with the intron-containing cytb were 68% (27/40) in Osaka, 2004; 46% (11/24) in Osaka, 2005; 39% (40/103) in Kanagawa, 2005; and 32% (10/31) in Kanagawa, 2006 (Table 3). Although isolates with the Bcbi-143/144 intron existed at high frequencies in Osaka and Kanagawa fields, none of them were QoI-resistant. Using this multiplex PCR-RFLP assay, 30 isolates with the intron were designated as non-G143 mutants. The sensitivity of the intron-carrying isolates (Bc-56, Bc-o-26, Bc-05-k88 and Bc-06-k56) to azoxystrobin was very similar to that of the sensitive isolates without the intron (Table 2).

image

Figure 5. Multiplex PCR-RFLP assay to detect the Bcbi-143/144 intron and the G143A mutation of cytb in Botrytis cinerea. (a) Multiplex PCR-RFLP strategy for the G143A mutation. Three primers, Bccytb-F2, Bccytb-R1 and Bccytb-in-R1, amplified the 306- and 212-bp fragments of cytb with and without the Bcbi-143/144 intron, respectively. (b) Agarose gel electrophoresis patterns after multiplex PCR-RFLP. The G143A mutation was detected by AluI digestion: two of three DNA fragments (69 and 210 bp) from cytb without the Bcbi-143/144 intron and two DNA fragments (69 and 143 bp) from artificially mutated cytb with the Bcbi-143/144 intron (See Materials and methods) were detected.

Download figure to PowerPoint

Table 3.  Frequency of Botrytis cinerea isolates without or with bi-143/144 intron in the cytochrome b gene
YearRegionAzoxystrobin-sensitiveAzoxystrobin-resistantTotal
− intron+ intron− intron+ intron
2004Osaka122710 40
2005Osaka131100 24
2005Kanagawa604030103
2006Kanagawa191020 31

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The discovery of QoI-resistant B. cinerea isolates in two distinct locations (Osaka and Kanagawa) would suggest that they are widely distributed in Japanese fields, although their population seemed to be low (ca. 3%, 6 out of 198 isolates). Three of the six resistant isolates were obtained from strawberry plants in different fields of Kanagawa Prefecture. There were no detailed records of the use of QoI fungicides in each field. However, in general, QoI fungicides had been tentatively used to control grey mould, anthracnose and powdery mildew in strawberry. High frequency of QoI application might promote the appearance of QoI-resistant isolates, although further studies are needed to elucidate the factors that influence the evolution of QoI resistance in B. cinerea. All resistant isolates had the G143A point mutation in cytb (Table 2), which was identified as the most common cause of QoI resistance in a wide variety of fungi (Gisi et al., 2002). Similar to other plant pathogens, the G143A mutation in cytb conferred B. cinerea with azoxystrobin resistance. The results indicated that two types of B. cinerea isolates, with or without the Bcbi-143/144 intron in cytb, existed in Japanese fields. It was shown that the pathogenicity of some isolates, which were classified into two types in this study, was indistinguishable (Banno et al., 2008). The two types of B. cinerea isolates, possessing different cytb, may be of similar environmental fitness, since approximately half of the isolates were shown to carry the intron (Table 3).

Mitochondrial cytb in fungi has group-I introns, which are self-splicing RNAs (Cech, 1988). The number of introns within cytb may vary among different fungi, but the number and locations are highly conserved within a single species (Grasso et al., 2006; Sierotzki et al., 2007). This is apparently the first report that describes isolates belonging to the same species that carry one of two types of cytb, i.e. with or without the Bcbi-143/144 intron. In B. cinerea, one group possessed three introns and the other had an additional intron in their cytb (Fig. 4). Four cytb introns each encode a putative maturase that is considered to stimulate the RNA-splicing reaction of each intron. These B. cinerea cytb maturases in the Bcbi-67/68, Bcbi-131/132, Bcbi-143/144 and Bcbi-164 introns are quite distinct from each other (data not shown), probably because they are involved in their own RNA splicing. Nevertheless, putative B. cinerea cytb maturases showed high degrees of homology to each of the corresponding maturases that intervene with 143rd and 144th codons of cytbs within fungi (data not shown). Individual introns might have evolved from distinct origins. More importantly, Grasso et al. (2006) proposed that the presence of Bcbi-143/144 may prevent the occurrence of the G143A QoI-resistant mutation in cytb. The G143A mutation GGT to GCT, which is adjacent upstream of the Bcbi-143/144 intron, would prevent self-splicing and lead to a deficient cytochrome b. Both of the G143A mutations in cytb with the Bcbi-143/144 intron and an insertion of the intron in G143A-mutated cytbs might occur. However, if the above-mentioned hypothesis is true, in both cases, functional cytochrome b would not be produced and the cell would be expelled from the environment. In fact, the G143A mutation was not found in plant pathogens that possess the cytb Bcbi-143/144-type intron, such as A. solani, Puccinia spp. and Plasmopara viticola (Grasso et al., 2006; Sierotzki et al., 2007; Toffolatti et al., 2007). The present results also support this hypothesis, since none of the B. cinerea isolates with the G143A mutation had the Bcbi-143/144 intron in cytb. The mechanism underlying this phenomenon remains unclear, but the results may indicate that the presence of the Bcbi-143/144 type intron in cytb, which may guarantee a low risk of QoI resistance, might not always be fixed in plant pathogens.

Two types of assay were developed here, the hybridization probe assay (Fig. 3) and the multiplex PCR-RFLP assay (Fig. 5), for rapid detection of the G143A mutation in cytb that is associated with QoI resistance in B. cinerea. The hybridization probe assay was based on the hybridization property of DNA and fluorescence resonance energy transfer between two fluorophores and could detect all the G143A-resistant cytbs with a melting peak at 66·2°C. Non-mutant cytbs without the Bcbi-143/144 intron also exhibited a melting peak at 57·1°C. This method is a reliable detection procedure that does not involve post-amplification steps. Furthermore, melting-curve analyses could detect individual cytb genes from a mixture of DNAs from both sensitive and resistant isolates, and the fluorescence intensity of the peaks was almost proportional to the DNA content. Therefore, this method was used to estimate the rates of QoI-sensitive and QoI-resistant isolates from coinfected aubergines. DNA isolated from spores produced on the aubergine fruits produced two peaks of similar strength at their corresponding melting temperatures (Fig. 3c,d). These results indicated that the G143A-resistant isolates had the ability to invade the aubergine fruit and produce conidia, as in the case of the sensitive isolates. The use of this hybridization probe assay is proposed, not only for detecting the G143A mutation, but also for elucidating the ability of fungicide-resistant isolates to compete against sensitive isolates.

Multiplex PCR-RFLP using three primers could detect the QoI-resistant G143 mutation in two types of cytbs, with or without the Bcbi-143/144 intron. In many cases, the G143A (GGT to GCT) mutation virtually created an ItaI or Fnu4HI (GCNGC) recognition site, which was utilized in the PCR-RFLP analysis of cytb (Sierotzki et al., 2000; Ma et al., 2003; Araki et al., 2005; Ishii et al., 2007). However, such enzymes may not cleave cytb with the G143 mutation and Bcbi-143/144-type intron unless the intron has the GC sequence at the 5′ end. The G143A mutation in most plant pathogens, including B. cinerea, creates an AluI (AGCT) recognition site in the exon region; therefore, it was not affected by the DNA sequence following it (Fig. 5). The results supported the explanation that occurrence of the G143A QoI-resistant mutation may be suppressed by the presence of the Bcbi-143/144 intron. However, careful monitoring should be carried out to confirm this hypothesis. Botrytis cinerea is a suitable plant pathogen for examining this hypothesis because it has two types of cytb genes (with or without the Bcbi-143/144 intron). The reported multiplex PCR-RFLP assay protocol can detect the G134A mutation in both cytbs and is useful for monitoring the occurrence of the mutation in B. cinerea.

It was demonstrated that QoI-resistant isolates of B. cinerea were distributed in Japanese fields. The QoI fungicides have been characterized as being at ‘high risk’ of resistance development. In addition, the fungus B. cinerea is a typical high-risk pathogen with regards to the possibility of fungicide resistance (Myresiotis et al., 2007). Although QoI resistance was reported in many pathogens in Japan (Ishii et al., 2007), the frequency of QoI-resistant isolates was still low. This may be attributed to the fact that the fungicide programmes employed for grey mould control have mainly used fungicides such as phenylpyrroles, anilinopyrimidines, hydroxyanilides and carboxamides, rather than QoIs, in Japan. Alternatively, two types of isolates with different cytb genes may contribute to reduce the risk of QoI resistance. Further studies are necessary to elucidate (i) why two types of cytochrome b gene exist in B. cinerea but are not found in other plant pathogens, and (ii) the effect of QoI fungicides on the population structure of the two types of isolates. The molecular diagnostic methods developed in this study will prove useful in clarifying these open questions and contribute to the monitoring of QoI resistance in B. cinerea for efficient grey mould control.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This work was partly supported by the University-Industry Joint Research Project for Private Universities, with a matching fund subsidy from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References