Distribution and molecular characterization of Corynespora cassiicola isolates resistant to boscalid

Authors


E-mail: ta_miyamoto@agri.pref.ibaraki.jp

Abstract

A total of 618 isolates of corynespora leaf spot fungus (Corynespora cassiicola) collected from 24 commercial cucumber greenhouses in 12 cities in Ibaraki Prefecture, Japan, were tested for their sensitivity to boscalid. Boscalid-resistant isolates were detected in 17 out of 19 greenhouses with a history of use of this fungicide and detection frequencies of the resistant isolates exceeded 47% in nine greenhouses. Frequencies of very highly resistant (VHR) isolates with 50% effective concentration (EC50) values of boscalid exceeding 30 μg mL−1 were higher than those of moderately resistant (MR) isolates with EC50 ranging from 2·0 to 5·9 μg mL−1 in 11 greenhouses. Additionally, highly resistant (HR) isolates with EC50 from 8·9 to 10·7 μg mL−1 were first detected. Furthermore, molecular characterization of genes encoding succinate dehydrogenase (SDH) subunits (SdhA, SdhB, SdhC and SdhD) was carried out to elucidate the amino acid substitution responsible for the resistance to boscalid. All 23 VHR isolates had the same mutation from CAC to TAC in the SdhB gene leading to the substitution of histidine with tyrosine at amino acid position 278 (B-H278Y). At the same position, the substitution to arginine conferred by a mutation to CGC (B-H278R) was detected in all four HR isolates. Some MR isolates showed a substitution from serine to proline at position 73 in SdhC (C-S73P), from serine to proline or from glycine to valine at position 89 (D-S89P) and 109 (D-G109V), respectively, in SdhD. There was no common mutation in SDH genes of all MR isolates.

Introduction

Cucumber (Cucumis sativus) is commonly grown commercially in Japan. Recently, corynespora leaf spot caused by Corynespora cassiicola has become one of the most important diseases of cucumber. Growers generally apply fungicides at 7- to 10-day intervals during the growing season, which lasts for 4–5 months after transplanting of grafted seedlings, or at the first sign of disease development. However, control of the disease is laborious and some growers tend to shorten the period of cucumber cultivation because of heavy outbreaks of this disease. One of the reasons for an increase in this disease is the development of C. cassiicola isolates resistant to various fungicides, such as benzimidazoles, dicarboximides, N-phenylcarbamates and QoI fungicides (Hasama, 1991; Hasama & Sato, 1996; Date et al., 2004; Takeuchi et al., 2006; Ishii et al., 2007).

Boscalid is a new fungicide in the succinate dehydrogenase inhibitor (SDHI) group, previously known as carboxamides, with a broad spectrum of activity (Stammler, 2008). Boscalid resistance was first reported in field isolates of Alternaria alternata on pistachio in the USA (Avenot & Michailides, 2007). Subsequently, isolates resistant to this fungicide were found in Botrytis cinerea on grapevine and strawberry, Podosphaera xanthii on cucurbits and other cases (McGrath, 2008; McGrath & Miazzi, 2008; Miazzi & McGrath, 2008; Stammler, 2008; Stevenson et al., 2008). In Japan, this fungicide was registered commercially for the control of grey mould and sclerotinia rot of cucumber in January 2005, and for corynespora leaf spot of cucumber in July 2006. However, after only a few years of boscalid use, studies carried out to test the boscalid sensitivity of C. cassiicola isolates collected from greenhouses with a history of boscalid use in two cities in Ibaraki Prefecture revealed the existence of isolates expressing resistance to boscalid (Miyamoto et al., 2009). Thereafter, it was shown that boscalid-resistant isolates of C. cassiicola were widely distributed within Japan (Ushio & Takeuchi, 2009).

SDHI fungicides inhibit fungal respiration through the inhibition of the enzyme succinate dehydrogenase (SDH, also known as complex II) in the mitochondrial electron transport chain (Fungicide Resistance Action Committee, 2009). SDH consists of four subunits, the hydrophilic flavoprotein (SdhA), the iron-sulphur protein (SdhB), and two lipophilic transmembrane C- and D-subunits (SdhC and SdhD). Studies on the molecular mechanisms responsible for the acquisition of resistance to SDHI fungicides, such as carboxin, flutolanil or fenfuram in some resistant isolates of bacteria and Basidiomycete and Ascomycete fungi have shown that mutations which lead to amino acid substitutions in the SdhB, SdhC or SdhD subunits of SDH confer laboratory resistance (Keon et al., 1991; Broomfield & Hargreaves, 1992; Matsson et al., 1998; Skinner et al., 1998; Matsson & Hederstedt, 2001; Ito et al., 2004; Li et al., 2006; Shima et al., 2008). Furthermore, field isolates resistant to boscalid also have the mutations which lead to amino acid substitutions in the SdhB, SdhC or SdhD subunit of A. alternata (Avenot et al., 2008a,b, 2009), and SdhB of B. cinerea (Stammler, 2008; Stammler et al., 2008). Based on the nucleotide sequences obtained from studies or from databases, several molecular diagnostic methods have been developed for identifying resistance (Sakurai, 2007; Avenot et al., 2008a, 2009).

The objectives of the current study were to (i) monitor the distribution of boscalid resistance of C. cassiicola isolates collected from cucumber greenhouses in 12 cities in Ibaraki Prefecture, Japan and (ii) sequence putative SdhA, SdhB, SdhC and SdhD genes and determine if mutations in these genes are responsible for boscalid resistance in C. cassiicola.

Materials and methods

Fungal isolates

For monitoring sensitivity to boscalid, 618 isolates were collected from 24 commercial cucumber greenhouses, with or without a history of boscalid applications, in Ibaraki Prefecture from October to November 2007. These isolates were maintained on 10-fold-diluted potato dextrose agar (PDA) slants at 5°C in the dark after single-spore isolation from lesions found on diseased leaves.

To analyse SDH genes, 56 isolates were collected from cucumber plants from different locations in Japan. Nine of the 56 isolates were collected and used in the above-mentioned test to monitor sensitivity to boscalid in Ibaraki. Twelve isolates collected in a previous study (Miyamoto et al., 2009) were also included. Thirty-three single-spore isolates were newly obtained from cucumber plants from Ibaraki, Chiba and Saga Prefectures between 2005 and 2008 and were maintained on PDA slants at 5°C in the dark. The remaining two isolates were gifts from Kagawa Prefectural Agricultural Experiment Station, and one isolate, a so-called ‘house isolate’ was used solely to design primers for the amplification of SDH genes of C. cassiicola.

Determination of C. cassiicola sensitivity to boscalid

The sensitivity test was performed by transferring a 4-mm-diameter disc of mycelium taken from the margin of a colony grown on a PDA plate at 25°C for 7 days onto YBA agar medium (yeast extract 1%, Bacto peptone 1%, sodium acetate 2% and agar 1·5%) (Stammler & Speakman, 2006; Ishii & Nishimura, 2007) amended with 0, 0·1, 1, 5, 10 or 30 μg mL−1 boscalid. Technical grade boscalid (94·4% a.i.) was dissolved in dimethylsulfoxide (DMSO) and added to cooled YBA agar medium after autoclaving. The final concentration of DMSO in the fungicide-amended and unamended media was 0·25% by volume. After incubation at 25°C in the dark for 4 days, the diameter of the mycelial colony was measured. Two replicate discs of mycelia were used per test, and each isolate was tested twice. For the rapid assessment of levels of resistance to boscalid, isolates were categorized depending on their sensitivity to the fungicide, as described in a previous report (Miyamoto et al., 2009), as sensitive (S; complete inhibition of mycelial growth on YBA agar amended with 10 μg mL−1, moderately resistant (MR; more than 75% inhibition at 30 μg mL−1) or very highly resistant (VHR; <55% inhibition). In addition, three isolates, IbCor0008, IbCor1481 and IbCor1689, which were isolated in the previous study (Miyamoto et al., 2009) were included in sensitivity tests as reference isolates of S, MR and VHR, respectively. In the present study, four isolates (IbCor3006, IbCor3009, IbCor3011 and IbCor3013) with inhibition rates at 30 μg mL−1 boscalid different from those of MR and VHR isolates were detected. These isolates were termed highly resistant (HR) isolates. For isolates shown in Table 4, the 50% effective concentration (EC50) value of boscalid for colony growth inhibition was calculated using log-linear model software kindly supplied by ZEN-NOH.

Table 4.   Deduced amino acid substitution in succinate dehydrogenase (SDH) from Corynespora cassiicola isolates
IsolateLocationEC50 to boscalid (μg mL−1)Sensitivity to boscaildDeduced substitution in SDH subunits
SdhBSdhCSdhD
  1. Asterisk indicates an isolate for which full-length sequeces of SdhA, SdhB, SdhC and SdhD genes were analysed. The sequences of SdhA gene were identical in 11 isolates.

  2. S: sensitive; MR: moderately resistant; HR: highly resistant; VHR: very highly resistant; WT: wild-type; NA: no analysis.

IbCor0008*Ibaraki0·4SWTWTWT
IbCor0101Ibaraki0·4SWTWTWT
IbCor1522*Ibaraki0·4SWTWTWT
IbCor1658*Ibaraki0·2SWTWTWT
IbCor1683*Ibaraki0·4SWTWTWT
IbCor1907*Ibaraki0·3SWTWTWT
20060410Chiba0·3SWTWTWT
050600-1Chiba0·2SWTWTWT
Mur9-1-1Kagawa0·5SWTWTWT
IbCor1218*Ibaraki5·3MRWTWTWT
IbCor1361*Ibaraki5·0MRWTWTS89P
IbCor1481*Ibaraki3·2MRWTS73PWT
IbCor1482*Ibaraki4·5MRWTS73PWT
IbCor1679*Ibaraki2·0MRWTWTWT
IbCor2429Ibaraki3·2MRWTS73PWT
Chikusei1-2Ibaraki2·7MRWTWTWT
Chikusei1-3Ibaraki3·5MRWTS73PWT
Chikusei1-11Ibaraki2·0MRWTWTWT
Chikusei1-14Ibaraki2·9MRWTNANA
Chikusei1-22Ibaraki2·3MRWTWTWT
Chikusei2-1Ibaraki3·5MRWTWTWT
Chikusei2-3Ibaraki2·6MRWTWTWT
Chikusei2-4Ibaraki5·4MRWTWTG109V
Chikusei2-5Ibaraki5·8MRWTWTWT
Chikusei2-7Ibaraki3·8MRWTWTWT
Chikusei2-8Ibaraki3·1MRWTWTWT
Chikusei2-9Ibaraki5·9MRWTNANA
Chikusei2-15Ibaraki4·8MRWTNANA
Chikusei2-26Ibaraki4·2MRWTNANA
IbCor3006Ibaraki10·7HRH278RWTWT
IbCor3009Ibaraki8·9HRH278RWTWT
IbCor3011Ibaraki9·8HRH278RWTWT
IbCor3013Ibaraki8·9HRH278RWTWT
IbCor1689*Ibaraki30<VHRH278YWTWT
IbCor2036BIbaraki30<VHRH278YWTWT
IbCor2529AIbaraki30<VHRH278YWTWT
IbCor2616AIbaraki30<VHRH278YWTWT
IbCor2617AIbaraki30<VHRH278YWTWT
Chikusei1-5Ibaraki30<VHRH278YWTWT
070327A1Chiba30<VHRH278YWTWT
070327A2Chiba30<VHRH278YWTWT
070327A3Chiba30<VHRH278YWTWT
070428B2Chiba30<VHRH278YWTWT
070508C3Chiba30<VHRH278YWTWT
070508D1Chiba30<VHRH278YWTWT
070508E1Chiba30<VHRH278YWTWT
070508F1Chiba30<VHRH278YWTWT
1-1-1Saga30<VHRH278YWTWT
1-2-2Saga30<VHRH278YWTWT
2-1-1Saga30<VHRH278YWTWT
2-2-1Saga30<VHRH278YWTWT
3-1-1Saga30<VHRH278YWTWT
3-2-2Saga30<VHRH278YWTWT
4-1-1Saga30<VHRH278YWTWT
4-3-1Saga30<VHRH278YWTWT
Toy2-1-1Kagawa30<VHRH278YWTWT

Inoculation tests

To evaluate the efficacy of boscalid, inoculation tests using potted cucumber plants were done using two VHR (IbCor3002 and IbCor3022), two HR (IbCor3006 and IbCor3013), two MR (IbCor3003 and IbCor3004) and two S (IbCor0008 and IbCor3001) isolates. The methods followed were as described by Miyamoto et al. (2009). After each fungal isolate was incubated on PDA plates at 25°C for 10 days in the dark, the aerial hyphae were removed with a sterile spatula and each plate was then incubated at 25°C for a further 3 days under black light/blue lamp to promote sporulation. Spores were then suspended in distilled water and the concentration adjusted to c. 104 spores per mL. The second true leaves of 3-week-old potted cucumber plants (cv. High-green 21) (four seedlings per treatment) grown at 25°C in a phytotron were then each sprayed with a suspension of boscalid (as Cantus 50 DF®) at a concentration of 334 μg a.i. mL−1, at the field dosage recommended by the manufacturer. Control plants were sprayed with tap water. After being dried in air, the plants were inoculated with the spore suspensions, stored at 28°C for 24 h in a chamber with high humidity (almost 100%), then maintained at 25°C for 4 days in a phytotron. The number of lesions of corynespora leaf spot was assessed 5 days after inoculation.

Specific primer design for SDH genes

To amplify complete sequences of the SdhA, SdhB, SdhC and SdhD genes in C. cassiicola, four PCR primer sets were designed with the ‘house isolate’. The isolate was cultivated on PDA and mycelia were harvested from Petri dishes. RNA was extracted from the mycelia using a NucleoSpin RNA Plant Mini Kit (Macherey-Nagel) including DNase digestion according to the manufacturer’s instructions. RNA quality was assessed by gel electrophoresis. About 700–1000 ng RNA were used to synthesize cDNA using the Verso cDNA Kit (ABgene) and KES 283 a reverse transcription (RT) primer based on KES 187 described below, according to the manufacturer’s instructions. Primer sets KES 503/KES 504 for SdhA, KES 719/KES 729 for SdhB, KES 544/KES 519 for SdhC, and KES 750/KES 187 for SdhD were designed by aligning conserved sequences of the SdhA, SdhB, SdhC and SdhD genes of Aspergillus fumigatus, Aspergillus niger, B. cinerea, Septoria tritici, Magnaporthe grisea and Fusarium graminearum (Table 1). PCR amplification was performed using 2× Thermo-Start PCR Mastermix (ABgene) with the following parameters on an Eppendorf Mastercycler gradient: an initial preheat 95°C for 15 min was followed by 40 cycles of 15 s at 95°C, 30 s at 50°C for SdhA or 55°C for SdhB and 60 s at 72°C, and a final step for 5 min at 72°C. For SdhC and SdhD following parameters were used: an initial preheat 95°C for 15 min was followed by 50 cycles of 15 s at 95°C, 30 s at 55°C and 30 s at 72°C, plus a final step for 5 min at 72°C. The PCR products were purified with NucleoSpin Extract2 (Macherey-Nagel). The purified fragments were cloned into the pJET1.2 vector using the CloneJET PCR Cloning Kit (Fermentas) and employed to transform competent Escherichia coli cells (XL-1, Stratagene). Two colonies per sample were taken to isolate the plasmid from the E. coli cells using the NucleoSpin Plasmid Kit (Macherey-Nagel) and sequenced using a ABI Prism 377 DNA Sequencer (Applied Biosystems) with BigDye Terminator Version 3.1 (Applied Biosystems). Based on the obtained partial sequences of each gene, new primer sets specific for C. cassiicola were designed with the primerexpress software (Applied Biosystems). To obtain the whole sequences of the SDH genes, RACE reactions were performed using a CapFishing Full-length cDNA Premix Kit (Seegene) following the manufacturer’s instructions, and cDNAs of SDH genes were directly sequenced with PCR primers as mentioned above. Sequences were analysed with the lasergene software package (DNASTAR). Specific primer sets to amplify the whole sequences, i.e. KES 897/KES 903 for SdhA, KES 746/KES 747 for SdhB, KES 764/KES 751 for SdhC, and KES 862/KES 762 for SdhD (Table 1), were designed with the above-mentioned software.

Table 1.   Primers used to amplify the succinate dehydrogenase subunit (A, B, C and D) genes of Corynespora cassiicola
Target subunitPrimer combinationPrimer nucleotide sequence (5′ to 3′)
  1. F: forward primer; R: reverse primer.

  2. aKES 283 was used for only RT-PCR.

SdhAKES 503 (F)CTCGTGGTGAGGGTGGTTACCT
KES 504 (R)CGCTTGAAAGGTGGAACAGC
SdhAKES 897 (F)ATGAGTTGTCTCAGGATGCGTC
KES 903 (R)AGAATACAGGACAGCAGCAAACAA
SdhBKES 719 (F)CTBCCNCACACCTACGTCGTCAAGGAC
KES 729 (R)CTTCTTRATCTCVGCRATVGCC
SdhBKES 746 (F)ATGGCTTGCACACGCGC
KES 747 (R)CTACCCAACAGCTCACTTGCC
SdhBSDHMF-1 (F)TGYCCVTCBTACTGGTGGAA
SDHMB-1a (R)GGRCAKGYYCKNGWRCARTT
SdhCKES 544 (F)CGVCCCGTVTCVCCVCAYCT
KES 519 (R)AVVGGIGCRRCGAGGTAKGC
SdhCKES 764 (F)ATGGCTTCCCAGCGCGTC
KES 751 (R)CTAAACAAACAGAGAATAGTAGAGGGTGG
SdhDKES 750 (F)GTAYGAGTTYGARACVAAYGA
KES 187 (R)GGCCACGCGTCGACTAGTAC
KES 283 (R)aGGCCACGCGTCGACTAGTACTTTTTTTTTTTTTTTTTT
SdhDKES 862 (F)ATGAAGCGCACCTCGCCAATC
KES 762 (R)AAAGATCCTAATAATCACATGTATCCAAAC

Analysis of nucleotide sequences of SDH genes

To compare the sequences of SDH genes among boscalid-sensitive and -resistant isolates in C. cassiicola, the whole sequences of the SdhA, SdhB, SdhC and SdhD genes of five S (IbCor0008, IbCor1522, IbCor1658, IbCor1683 and IbCor1907), five MR (IbCor1218, IbCor1361, IbCor1481, IbCor1482 and IbCor1679)and one VHR isolate (IbCor1689) of C. cassiicola were cloned and sequenced. Eleven isolates were cultivated on PDA and mycelia were harvested from Petri dishes. Genomic DNA was extracted from the mycelia using the NucleoSpin DNA Plant Mini Kit according to the manufacturer’s instructions. Primer sets KES 897/KES 903, KES 746/KES 747, KES 764/KES 751 and KES 862/KES 762 (Table 1) were used to amplify the SdhA, SdhB, SdhC and SdhD genes, respectively. PCR amplifications were performed using the following parameters: an initial preheat 98°C for 1 min was followed by 40 cycles of 10 s at 98°C, 30 s at 62°C for SdhA, 68°C for SdhB and SdhC, or 66°C for SdhD, 30 s at 72°C and a final step for 5 min at 72°C. PCR products were separated on a 2% agarose gel containing ethidium bromide, with 1× Tris-acetate EDTA (TAE) buffer. PCR products of 2340 bp (SdhA), 1071 bp (SdhB), 629 bp (SdhC) and 1083 bp (SdhD) were purified with NucleoSpin Extract2 and were ligated into the pJET1.2 vector and transformed into competent E. coli cells (XL-1). Transformants were selected on Luria-Bertani (LB) agar plates containing ampicillin at 100 μg mL−1. Individual clones were used to inoculate LB broth containing 100 μg mL−1 ampicillin and grown overnight at 37°C with shaking. Plasmids were purified using the NucleoSpin Plasmid Mini Kit. The plasmids thus purified were digested with BglI to confirm that inserts were present. The plasmids were sequenced as mentioned above.

Furthermore, partial sequences of the SdhB gene, and the whole sequences of the SdhC and SdhD genes of four S, 15 MR, four HR and 22 VHR isolates were directly sequenced with their PCR products. Genomic DNA was extracted from mycelia cultured on PDA for 5 days using the method of Saitoh et al. (2006). Primer sets SDHMF-1/SDHMB-1a (Sakurai, 2007), KES 764/KES 751 and KES 862/KES 762 (Table 1) were used to amplify the SdhB, SdhC and SdhD genes, respectively. Primer set SDHMF-1/SDHMB-1a was used to amplify the partial region encoding a proline residue at position 231 and a histidine at position 278, which were previously reported to be responsible for SDHI sensitivity in other plant pathogens (Keon et al., 1991; Broomfield & Hargreaves, 1992; Skinner et al., 1998; Avenot et al., 2008a; Stammler et al., 2008). PCR amplifications were carried out in a Thermal Cycler Dice Gradient (Takara Bio) with an initial pre-heat for 2·5 min at 94°C, followed by 40 cycles for SdhB and 30 cycles for SdhC and SdhD of denaturation at 94°C for 30 s, annealing at 48°C for 1 min and extension at 72°C for 1 min, terminating with a final extension at 72°C for 10 min. The PCR products from each isolate were purified using the MinElute PCR Purification Kit (Qiagen) according to the instructions supplied by the manufacturer. Sequence analyses of PCR products were performed with the automated DNA sequencer Prism Model (Applied Biosystems) by using fluorescent-dye-labelled dideoxy terminators or the commercial DNA sequence service by Greiner Bio-one.

Results

Distribution of C. cassiicola isolates resistant to boscalid in Ibaraki Prefecture, Japan

Boscalid was used in 19 of 24 greenhouses (Table 2). Except for greenhouses in Jousou City and Koga City, this fungicide was generally applied as a tank-mix with other fungicides. Corynespora cassiicola isolates resistant to boscalid were detected in 17 greenhouses which consisted of eight with a history of two or three applications of boscalid per crop, and nine with only one application. Moreover, resistant isolates were also obtained from two greenhouses (Tsukuba-B and -C) without a prior history of boscalid use. On the other hand, resistant isolates were not detected in two greenhouses, Shirosato-A and -B with totals of six or nine prior applications of boscalid, respectively.

Table 2.   Monitoring of boscalid sensitivity in Corynespora cassiicola isolates collected from cucumber greenhouses with or without a history of boscalid use in Ibaraki prefecture, Japan
Location of greenhousesaNo. of isolates testedbNo. of resistant isolatesFrequency of resistant isolates (%)cNo. of total boscalid sprays before isolation
MRHRVHRMR + HR + VHR
  1. Asterisk indicates that boscalid was always used as a solo product.

  2. MR: moderately resistant; HR: highly resistant; VHR: very highly resistant.

  3. aLetter after city name indicates different greenhouse in the same city.

  4. bCollection of isolates in October or November 2007 in cucumber greenhouses.

  5. cFrequency of resistant isolates (%) = no. of MR + HR + VHR isolates/no. of isolates tested × 100.

  6. dChikusei-A to -G were shown in previous study (Miyamoto et al., 2009).

Greenhouses with a history of two or three applications of boscalid per crop
 Chikusei-Hd309451860·04
 Chikusei-I3022032583·311
 Daigo-A3060182480·010
 Daigo-B1810151688·915
 Jousou-A211091047·67*
 Kawauchi2800113·62<
 Sakuragawa30204620·012
 Shirosato-A600000·06
 Shirosato-B2800000·09
 Shirosato-C6200233·32
Greenhouses with a history of only one application of boscalid per crop
 Jousou-B2540101456·06*
 Jousou-C3000181860·04*
 Jousou-D302001030100·04*
 Jousou-E30203516·75*
 Koga2200229·15*
 Namegata3018042273·36
 Ryugasaki3000113·31
 Shirosato-D17101211·81
 Tsukuba-A32103412·53
Greenhouses without a history of boscalid use
 Hitachiomiya3000000·00
 Mito-A3000000·00
 Mito-B3000000·00
 Tsukuba-B3210013·10
 Tsukuba-C23008834·80

Two hundred of 473 isolates collected from 19 greenhouses with a history of boscalid use were resistant to boscalid. Detection frequencies of resistant isolates exceeded 47% in nine greenhouses. Eighty-nine MR and 107 VHR isolates were detected in 14 and 16 greenhouses, respectively. Unlike the previous study (Miyamoto et al., 2009), the frequency of VHR isolates was higher than that of MR isolates in 11 of 19 greenhouses. The sensitivity of the remaining four HR isolates (IbCor3006, IbCor3009, IbCor3011 and IbCor3013) was different from that of MR and VHR isolates. The inhibition rate of HR isolates ranged from 65% to 71% on YBA agar amended with 30 μg mL−1 boscalid (data not shown). Furthermore, EC50 values of boscalid for MR and VHR isolates were 2·0–5·9 μg mL−1 and >30 μg mL−1, respectively, whilst those of HR isolates ranged from 8·9 to 10·7 μg mL−1. HR isolates were detected in only one greenhouse (Chikusei-H).

Efficacy of boscalid in controlling disease caused by HR isolates

Two isolates selected from each of the S, MR, HR and VHR categories were used for the inoculation tests (Table 3). S isolates were completely controlled by boscalid at 334 μg a.i. mL−1. In contrast, decreased efficacy was recorded against MR, HR and VHR isolates. Boscalid still slightly inhibited lesion formation in terms of number of lesions when leaves were inoculated with either MR or HR isolates compared with inoculation with VHR isolates.

Table 3.   Efficacy of boscalid against corynespora leaf spot disease of cucumber inoculated with boscalid-sensitive, -moderately resistant, -highly resistant and -very highly resistant isolates of Corynespora cassiicola
IsolateSensitivity to boscalidTreatmentaAverage no. of lesionsbDisease control (%)c
  1. S: sensitive; MR: moderately resistant; HR: highly resistant; VHR: very highly resistant.

  2. aBoscalid was sprayed at the concentration of 334 μg a.i. mL−1.

  3. bAverage number of lesions assessed 5 days after inoculation with spore suspensions (c. 104spores per mL) on second leaves of four seedlings.

  4. cDisease control (%) of individual fungicides was calculated as (no. of lesions on tap-water treated leaves−no. of lesions on boscalid-treated leaves)/no. of lesions on tap-water treated leaves × 100.

IbCor0008SBoscalid0100
Tap water50·5
IbCor3001SBoscalid0100
Tap water33·0
IbCor3003MRBoscalid16·875
Tap water67·0
IbCor3004MRBoscalid13·563
Tap water36·8
IbCor3006HRBoscalid42·539
Tap water69·3
IbCor3013HRBoscalid26·048
Tap water49·8
IbCor3002VHRBoscalid52·05
Tap water49·5
IbCor3022VHRBoscalid36·05
Tap water38·0

Nucleotide sequence of SDH genes from C. cassiicola isolates

Primer sets (Table 1) KES 897/KES 903, KES 746/KES 747, KES764/KES 751 and KES 862/KES 762, were successfully used to amplify 2340-, 1071-, 629- and 1083-bp fragments, respectively, from the genomic DNA of 11 isolates of C. cassiicola. These fragments were cloned and sequenced. The assembled SdhA, SdhB, SdhC and SdhD gene sequences contained open reading frames (ORFs) of 647, 307, 177 and 165 amino acids, respectively. ORFs of SdhA, SdhB, SdhC and SdhD genes were interrupted by three (206, 56 and 47 bp), two (64 and 59 bp), one (95 bp) and two (250 and 120 bp) introns, respectively. These resulting sequences were subjected to a homology search using the blastx search program in the DNA Data Bank of Japan (DDBJ) database. Deduced amino acid sequences of SdhA, SdhB, SdhC and SdhD of C. cassiicola were similar to those of the well-characterized mitochondrial flavoprotein (71% identity, accession no. B3LQV3), iron-sulfur protein (71%, B3LTD3), cytochrome b (34%, B3LQV9) and membrane anchor (32%, B3LGA5) of succinate dehydrogenase of Saccharomyces cerevisiae, respectively. Additionally, these sequences of C. cassiicola were highly homologous to putative SdhA (84%, B8XCQ1), SdhB (89%, B2BZ64), SdhC (67%, B8XSR3) and SdhD (78%, B8XSR4) of A. alternata.

Partial sequences of SdhB, and whole sequences of the SdhC and SdhD genes of the remaining 45 isolates were directly sequenced from PCR products. Three SDH genes from these isolates were identical in their sequences to those of 11 isolates cloned, except for the mutations mentioned below.

Nucleotide sequence comparison of SDH genes among boscalid-S, MR, HR and VHR C. cassiicola isolates

The sequences of the C. cassiicola SDH genes were compared among isolates which were grouped into four classes based on their differential sensitivity to boscalid (Table 4). Complete sequences of the SdhA, SdhB, SdhC and SdhD genes of five S, five MR and one VHR isolate were cloned and sequenced. The SdhA genes of these isolates were identical in their nucleotide sequences and consequently in deduced amino acid sequences as well (data not shown). When the full-length of sequence of the SdhB gene was analysed using a VHR isolate (IbCor1689), it was found that CAC was replaced with TAC, resulting in the substitution from histidine to tyrosine at amino acid position 278 (B-H278Y) in the S3 cluster (Fig. 1). MR isolates were found to carry no mutations in the SdhB gene. In two MR isolates, IbCor1481 and IbCor1482, TCG was replaced with CCG, resulting in the substitution of serine with proline at position 73 (C-S73P) of SdhC. Furthermore, the SdhD gene of the MR isolate IbCor1361 was mutated from TCC to CCC, resulting in a serine to proline substitution at amino acid position 89 (D-S89P). No mutation was found in the SDH genes of two other MR isolates, IbCor1218 and IbCor1679.

Figure 1.

 Comparison of the amino acid residues substituted in Corynespora cassiicola isolates resistant to boscalid and flanking regions of SdhB, SdhC and SdhD subunits from organisms in which SDHI-resistant mutants have been found or in which the structure of SDH has been determined. Sequences are from Alternaria alternata (Aal), Aspergillus oryzae (Aor), Coprinopsis cinerea (Cci), C. cassiicola (Cca), Escherichia coli (Eco), Mycosphaerella graminicola (Mgr) and Paracoccus denitrificans (Pde). Asterisks indicate residues that interact with ubiquinone and carboxin (Horsefield et al., 2006). Bold letters indicate residues that are substituted in C. cassiicola isolates resistant to boscalid. Underlined letters indicate residues that are substitutions responsible for SDHI fungicide resistance in other organisms.

For the remaining four S, 15 MR, four HR and 22 VHR isolates, partial sequences of the PCR products of the SdhB gene and the whole SdhC and SdhD genes were determined directly. A mutation corresponding to B-H278Y was found in the SdhB gene in all of the 22 VHR isolates. The SdhB gene of four HR isolates was mutated from CAC to CGC, causing the substitution of histidine with arginine at the same position (B-H278R). Two MR isolates, IbCor2429 and Chikusei1-3, carried a mutation responsible for the substitution of C-S73P. The SdhD gene of Chikusei2-4 was mutated from GGC to GTC, resulting in a glycine to valine substitution at amino acid position 109 (D-G109V). However, eight MR isolates possessed no mutations in three genes examined.

Accession numbers of these sequences were AB548737 (SdhA), AB548738AB548740 (SdhB), AB548741 and AB548742 (SdhC), and AB548743AB548745 (SdhD).

Discussion

In a previous study (Miyamoto et al., 2009), C. cassiicola isolates resistant to boscalid were found in cucumber greenhouses with a history of boscalid use in two cities in Ibaraki Prefecture. The present study suggests that boscalid resistance is widespread in many cities in Ibaraki. Most of the cucumber growers in those areas applied boscalid in a tank mixture with non-selective fungicides, such as mancozeb, TPN or captan, to concurrently control other diseases, such as downy or powdery mildew. Furthermore, to avoid the development of boscalid resistance, several growers limited the number of boscalid spray applications to only one per crop. However, these anti-resistance strategies were not always effective. To make matters worse, the frequency of VHR isolates is now higher than that reported in the previous study (Miyamoto et al., 2009). This finding indicates that it is getting more and more difficult to control corynespora leaf spot using boscalid. It has already been recommended that growers discontinue the use of boscalid in the control of this disease in Ibaraki. Recently, the development of boscalid resistance in cucumber corynespora leaf spot fungus was also reported in some other prefectures, such as Chiba (Ushio & Takeuchi, 2009) and Kagawa (M. Mori, unpublished data). This increasing distribution of resistance is problematical for cucumber cultivation in many regions of Japan.

Interestingly, resistant isolates were not detected from two greenhouses, Shirosato-A and -B, despite boscalid being sprayed a total of six and nine times, respectively. Unlike most other greenhouses, in which cucumbers were cultivated twice per year and corynespora leaf spot disease occurred heavily in the later stages of cultivation, cucumber was cultivated only once per year in the two Shirosato greenhouses and the disease pressure was low according to the data so far examined (data not shown). With other fungicides, resistant fungal isolates emerge more slowly at lower disease pressure (Staub, 1991; Brent & Hollomon, 2007). By contrast, resistant isolates were also detected in two greenhouses (Tsukuba-B and -C) in which no boscalid applications were performed. Similar results were obtained in A. alternata resistant to boscalid in California, USA (Avenot & Michailides, 2007). The reason for the results obtained from these four greenhouses is as yet unclear, but the method of spore dispersal may be a critical factor for the dispersal of boscalid-resistant isolates of C. cassiicola; in several monitoring studies, long-distance dispersal of resistant fungal conidia by wind has already been proposed (Foster & Staub, 1996; Ishii et al., 2001) and was demonstrated in QoI resistance of Mycosphaerella graminicola (Fraaije et al., 2005).

Sequence comparison among boscalid-sensitive and -resistant isolates of C. cassiicola showed that HR and VHR isolates had the amino acid substitutions B-H278R and B-H278Y, respectively. This histidine residue, located in a region associated with the S3 cluster, was highly conserved in various organisms (Broomfield & Hargreaves, 1992). Structural analysis of the quinone-binding site (Q-site) in SDH from E. coli demonstrated that one potential Q-site was very close to the conserved histidine residue, at the equivalent position to C. cassiicola SDH, suggesting key roles in ubiquinone binding and reduction (Horsefield et al., 2006). The same study revealed carboxin docked in the same way as ubiquinone in close proximity to the conserved histidine. Broomfield & Hargreaves (1992) showed that substitution of this histidine to a leucine in Ustilago maydis SDH resulted in resistance to carboxin. The same results were also observed for M. graminicola (Skinner et al., 1998) and Paracoccus denitrificans (Matsson & Hederstedt, 2001). Recently, it was reported that boscalid-resistant isolates of several fungi, such as B. cinerea (Stammler, 2008; Stammler et al., 2008) and A. alternata (Avenot et al., 2008a) had the substitution of this histidine residue. These results strongly suggest that substitution of this histidine residue is responsible for boscalid resistance in C. cassiicola.

HR isolates with B-H278R and VHR isolates with B-H278Y also showed different phenotypes in their growth on boscalid-containing medium and in their sensitivity to boscalid on cucumber leaves. In laboratory studies on Aspergillus oryzae, three types of mutants with different substitutions of this histidine revealed different phenotypes in their growth on carboxin-containing medium and in their SDH activity (Shima et al., 2008). The substitution from this histidine to arginine or tyrosine was also found in boscalid-resistant isolates of A. alternata (Avenot et al., 2008a) and B. cinerea (Stammler, 2008; Stammler et al., 2008), but it was not reported that the different substitutions resulted in differential sensitivities to the fungicide.

Unlike the substitutions in HR and VHR isolates, there was a clear relationship between boscalid sensitivity and the residues corresponding to three substituted residues found in the SdhC and SdhD genes of MR isolates, namely C-S73P, D-S89P and D-G109V. These substitutions in these two genes were not found in other fungal isolates resistant to SDHI (Matsson & Hederstedt, 2001; Ito et al., 2004; Avenot et al., 2008b, 2009; Shima et al., 2008). D-G109 in C. cassiicola was a highly conserved residue in a wide range of organisms, although C-S73 and D-S89 were not particularly conserved. These mutations were found in only some MR isolates, whilst other MR isolates (IbCor1218 and IbCor1679) had no mutations in any SDH genes. If these three mutations (C-S73P, D-S89P and D-G109V) are responsible for the reduced sensitivity to boscalid, it has still to be proven.

Although MR isolates were the most sensitive of those showing resistance to boscalid, a decrease in control efficacy of boscalid against corynespora leaf spot was observed in greenhouses even when only MR isolates were detected (Miyamoto et al., 2009). Although the molecular diagnosis of VHR and HR isolates may be developed by using mutations responsible for B-H278Y and B-H278R (e.g. Sakurai, 2007; Avenot et al., 2008a), this diagnosis method will be not practical for monitoring sensitivity to boscalid because MR isolates cannot be thus detected. Consequently, a molecular marker to detect MR isolates is necessary for the development of a practical molecular diagnosis method for boscalid resistance in this fungus.

Acknowledgements

We express our thanks to M. Mori (Kagawa Prefectural Agricultural Experiment Station) and M. Inada (Saga Prefectural Agriculture Research Center) for providing Corynespora cassiicola isolates or assisting in the isolation of this fungus for this work. We also thank S. Uzuhashi (National Institute for Agro-Environmental Sciences) and M. Mizuno (Ibaraki Agricultural Center) for technical assistance.

Ancillary