Resistance to the Novel Fungicide Pyrimorph in Phytophthora capsici: Risk Assessment and Detection of Point Mutations in CesA3 That Confer Resistance

Pyrimorph is a novel fungicide with high activity against the plant pathogen Phytophthora capsici. We investigated the risk that P. capsici can develop resistance to pyrimorph. The baseline sensitivities of 226 P. capsici isolates, tested by mycelial growth inhibition, showed a unimodal distribution with a mean EC50 value of 1.4261 (±0.4002) µg/ml. Twelve pyrimorph-resistant mutants were obtained by repeated exposure to pyrimorph in vitro with a frequency of approximately 1×10−4. The resistance factors of the mutants ranged from 10.67 to 56.02. Pyrimorph resistance of the mutants was stable after 10 transfers on pyrimorph-free medium. Fitness in sporulation, cystospore germination, and pathogenicity in the pyrimorph-resistant mutants was similar to or less than that in the parental wild-type isolates. On detached pepper leaves and pepper plants treated with the recommended maximum dose of pyrimorph, however, virulence was greater for mutants with a high level of pyrimorph resistance than for the wild type. The results suggest that the risk of P. capsici developing resistance to pyrimorph is low to moderate. Among mutants with a high level of pyrimorph resistance, EC50 values for pyrimorph and CAA fungicides flumorph, dimethomorph, and mandipropamid were positively correlated. This indicated that point mutations in cellulose synthase 3 (CesA3) may confer resistance to pyrimorph. Comparison of CesA3 in isolates with a high level of pyrimorph resistance and parental isolates showed that an amino acid change from glutamine to lysine at position 1077 resulted in stable, high resistance in the mutants. Based on the point mutations, an allele-specific PCR method was developed to detect pyrimorph resistance in P. capsici populations.

Pyrimorph's mechanism of action is currently the subject of intensive investigation. Inhibition of P. capsici by pyrimorph might result from multiple modes of action including the impairment of the energy generation system and direct or indirect impairment of cell wall biosynthesis [23]. The mechanism of action for pyrimoph might be similar to that for carboxylic acid amide (CAA) fungicides because CAA-resistant laboratory mutants of P. capsici were also resistant to pyrimorph [26]. Recent research has identified amino acid changes in cellulose synthase 3 (CesA3) that confer resistance to CAA fungicides in the following oomycetes: Plasmopara viticola, Phytophthora infestans, Pseudoperonospora cubensis, and Phytophthora melonis [27][28][29][30][31]. However, the resistance mechanism of P. capsici to pyrimorph has not been elucidated at the molecular level.
Although pyrimorph has been marketed in China for about 3 years, it is still in the early stages of use on pepper, and resistance of P. capsici to pyrimorph has not yet been reported. Determining the response profiles of P. capsici field populations to pyrimorph and studying the characteristics of laboratory-generated mutants with pyrimorph resistance will be useful for estimating the risk that P. capsici can develop resistance to this new fungicide [32]. In the current study, we established the baseline sensitivity of P. capsici from 40 locations in 24 provinces of China to pyrimorph; we generated pyrimorph-resistant mutants with stable resistance by exposing isolates to pyrimorph in vitro; and we then assessed the risk that P. capsici can develop resistance to pyrimorph. By assaying cross-resistance between pyrimorph and other oomycete fungicides, we speculated that pyrimorph has a similar resistance mechanism to CAA fungicides, and we identified a novel point mutation at codon position 1077 in the PcCesA3 gene that is linked to pyrimorph resistance. Based on this mutation, a rapid and reliable allele-specific PCR method was developed to detect pyrimorph-resistant isolates in populations of P. capsici. . These fungicides were dissolved in dimethyl sulfoxide (DMSO) to make stock solutions (10 mg/ml), and the stock solutions were stored at 4uC in the dark.

Fungicides and Media
Potato dextrose agar (PDA) was prepared by boiling 200 g of potato tubers for 0.5 h and filtering the material through four layers of cheese cloth The filtrate was combined with 18 g of glucose, 15 g of agar, and sufficient distilled water to increase the volume to 1 L. For carrot agar (CA), juice from 200 g of ground carrots was filtered with cheese cloth; the filtrate was combined with 14 g of agar and sufficient distilled water to increase the volume to 1 L [33].

Isolates of P. capsici
During 2006-2010, stems of pepper with typical brown lesions produced by P. capsici were sampled from 40 locations in 24 provinces of China; as noted earlier, these locations had never been treated with pyrimorph. Small pieces of tissue (563 mm) cut from margins of lesions on the stem were surface disinfested in 1% NaClO (vol:vol) for 3 minutes, rinsed three times in sterile water, and placed on PDA plates amended with 50 mg/ml of rifampicin (98% a.i., Tuoyingfang Biotech Co., Ltd., Beijing, China), 50 mg/ ml of penicillin (98% a.i., Tuoyingfang Biotech Co., Ltd., Beijing, China), and 50 mg/ml of pentachloronitrobenzene (40% a.i., Sanli Chemical Industry Co., Ltd., Shanxi, China). After 3 days at 25uC in the dark, mycelial plugs were cut from the colony margins and transferred to new PDA plates. A total of 226 isolates were obtained.
All the isolates were preserved on PDA slants covered with sterile mineral oil and were stored at room temperature. Single-tip transfers were made every 6 months.
Baseline Sensitivity of P. capsici to Pyrimorph The 226 P. capsici isolates were grown on PDA plates for 4 days at 25uC in the dark. A mycelial plug was taken from the colony margin with a cork borer (5 mm) and placed mycelia-side down in the centre of a PDA plate amended with 0, 0.1875, 0.375, 0.75, 1.5, 3, or 4 mg/ml of pyrimorph. The final concentration of DMSO was 0.1% in the medium. Each combination of isolate and concentration was represented by three replicate plates, and the experiments were performed three times. After 5 days at 25uC, the radial growth (the plug diameter was subtracted) on each plate was measured perpendicularly and averaged to calculate the percentage of growth inhibition (relative to the growth on plates without pyrimorph). EC 50 (the effective concentration for 50% inhibition of mycelial growth) was calculated by the regression of the probit of the percentage of inhibition of radial growth against the logarithmic value of pyrimorph concentration. In addition, the variance was analyzed to determine whether the baseline sensitivity to pyrimorph differed among the locations.

Generation of P. capsici Mutants Resistant to Pyrimorph
Mass selection from mycelial agar plugs. To obtain pyrimorph-resistant mutants from mycelial agar plugs, 10 wildtype isolates were used: Dz21, Hd3, Hd11, Hx18, Hb1-17, Mq12, N10, Pc112, Tj3-11, and Xs2. First, mycelial agar plugs (5 mm in diameter) were excised from 4-day-old colonies on PDA plates and placed mycelia-side down on PDA plates containing 20 mg/ml of pyrimorph, which was tentatively considered to be a discriminatory concentration for identifying mutants of P. capsici insensitive to pyrimorph. After 15-20 days at 25uC in the dark, the survivors were subsequently placed on PDA plates amended with the same pyrimorph concentration. This step was repeated until there was no significant difference in the linear growth of fast-growing sectors on the PDA plates with or without 20 mg/ml of pyrimorph. To obtain mutants with high resistance to pyrimorph, the lowresistant survivors were subjected to PDA plates amended with 100 mg/ml of pyrimorph. After 30 days at 25uC in the dark, fastgrowing sectors were transferred to new PDA plates amended with the same pyrimorph concentration. This step was repeated until there was no significant difference in the linear growth of fastgrowing sectors on the PDA plates with or without 100 mg/ml of pyrimorph.
For putative mutants that were resistant to 100 mg/ml of pyrimorph (as described in the previous paragraph), singlezoospore isolates were obtained using the method reported previously [34,35]. Isolates of P. capsici were cultured on CA plates (9 cm diameter) for 4 days in the dark at 25uC. Then the plates were placed under florescent light with 12-h light/12-h dark at 25uC. After 5 to 7 days, when sufficient numbers of sporangia had been produced, plates were flooded with 10 ml of steriledistilled water and placed at 4uC for 30 min and then at 25uC for 30 min. Zoospore suspensions of each isolate were spread onto 1.5% (wt/vol) water agar. After 16 h at 25uC in the dark, single germinated zoospores were transferred to PDA plates. These mutants were tested for their level and stability of resistance to pyrimorph as described later in the Methods.
Mass selection from zoospores. Wild-type isolates of P. capsici (Dz21, Hx18, Tj3-11, and Xs2) were used to select pyrimorph-resistant mutants from zoospores. For each isolate, zoospore suspensions were obtained following the method described in the previous section. The concentration of zoospore suspensions was adjusted to 10 7 /ml with the aid of a hemocytometer. The zoospore suspensions were inoculated onto PDA plates amended with pyrimorph at 1 mg/ml, at which concentration cystospore germination was completely inhibited for sensitive P. capsici isolates. After 7 days at 25uC in the dark, the survivors were transferred to new PDA plates with the same concentration of pyrimorph. This was repeated, and single-zoospore mutants were then generated and tested for their level and stability of resistance to pyrimorph as described in the next section.

Characterization of Pyrimorph-resistant Mutants of P. capsici
Level and stability of pyrimorph resistance. For determination of the level of pyrimorph resistance of the mutants, mycelial plugs excised from the margins of 4-day-old colonies on PDA plates were placed upside-down on the centre of PDA plates (one plug per plate) amended with 0, 3.125, 6.25, 12.5, 25, 50, 100, or 150 mg/ml of pyrimorph. The EC 50 values were calculated. Resistance factors (RFs) were also calculated using the following formula: RF = EC 50 X/EC 50 P, where EC 50 X is the EC 50 value of the resistant mutant being examined, and EC 50 P is the EC 50 value of the parental wild-type isolate of that mutant.
For assessing the stability of resistant phenotypes, the resistant mutants were subjected to 10 successive transfers on fungicide-free PDA plates. RFs were determined after the first and the tenth subculture. A ''factor of sensitivity change'' (FSC) was calculated as the RF value in the tenth transfer divided by that in the first transfer. For further assessment of resistance stability, four resistant mutants were randomly selected, and 10 successive transfers with zoospores were conducted. Then, at least 50 single-zoospore isolates for each of the four resistant mutants were randomly sampled and their growth was assessed on PDA plates amended with 20 mg/ml of pyrimorph.
Mycelial growth as affected by temperature in vitro. The pyrimorph-resistant mutants and corresponding parental isolates were incubated on PDA plates at 15, 20, 25, 30, and 37uC. Colony diameters were measured perpendicularly after 5 days in the dark. Each combination of isolate and temperature was represented by four replicate plates.
Sporulation and cystospore germination in vitro. In vitro sporangial production and zoospore release by pyrimorphresistant mutants and corresponding parental isolates were measured as described earlier. Zoospore production was quantified with a hemocytometer and the number of zoospores per cm 2 of each CA plate (9 cm diameter) was calculated. After 12 h in the dark, more than 100 cystospores were examined, and cystospore germination was determined and expressed as a percentage; a cystospore was considered to have germinated if the length of its germ tube was greater than the diameter of the cystospore. These experiments were conducted at least two times.
Virulence and sporulation on detached bell pepper leaves. For determination of virulence of the pyrimorphresistant mutants and corresponding parental isolates, fully expanded bell pepper leaves of the same age and area were excised from pepper plants (cultivar Tedaqiemen) with four true leaves, rinsed three times with sterile-distilled water, and placed on 1% water agar in 20-cm-diameter Petri dishes. For each isolate, 12 replicate leaves were used. A mycelial plug (5 mm in diameter) from the margin of a 4-day-old colony on PDA plates was placed on the middle of the leaf. After 4 days in a growth chamber (Shanghai Yiheng Technical Co., Ltd., Shanghai, China) with 80% relative humidity, 25uC, and 12-h light/12-h dark, the percentage of leaf areas covered by lesions was determined.
For determination of sporangia production, the 12 diseased leaves per isolate were then placed in a 50-ml centrifuge tube containing 15 ml of distilled water and mixed with a vortex apparatus (Kylin-bell Lab Instruments Co. Ltd., Haimen, China) for 15 s. The sporangia were counted using a hemocytometer, and the number of sporangia per cm 2 of diseased leaf surface was calculated.
Virulence on pepper plants. Pepper seedlings (cultivar Tedaqiemen) were grown in plastic trays (540 mm6280 mm, 50 seedlings per tray) in the greenhouse. When the seedlings were 6 weeks old, they were inoculated with the pyrimorph-resistant mutants and corresponding parental strains following the method described by Sun [11], with slight modification. Zoospore suspensions of the isolates were prepared as described earlier.
The seedlings in the trays were watered and then inoculated by adding 3 ml of a zoospore suspension (1610 4 zoospores/ml) to the soil surface around each plant. Each isolate was represented by ten replicate plants. After inoculation, the trays were placed in a greenhouse (2762uC, 80% relative humidity, and 12-h light/12-h dark) and were watered twice daily [36]. After 7 days, virulence was assessed based on stem and leaf symptoms. Disease severity was rated on a scale of 0-5:0, no visible disease symptoms; 1, leaves slightly wilted with black lesions beginning to appear on stems or 10-29% of the entire plant diseased; 2, 30-49% of the entire plant diseased; 3, 50-69% of the entire plant diseased; 4, 70-90% of the entire plant diseased; 5, dead plant [37,38]. The experiment was conducted four times.
Control of pyrimorph-resistant mutants and wild-type isolates on bell pepper leaves treated with pyrimorph. Fully expanded bell pepper leaves of the same age were excised from glasshouse-grown pepper plants (cultivar Tedaqiemen). The leaves were rinsed three times in sterile-distilled water and placed on 1% water agar in 20-cm-diameter Petri dishes before they were sprayed with pyrimorph. For treatment of leaves with pyrimorph, pyrimorph was diluted with sterile-distilled water containing 0.1% DMSO and 0.005% Tween 20 to obtain 750 mg of pyrimorph/ml. A 1-ml volume of the pyrimorph solution (equivalent to the maximum dose of 450 g/ha in the field) or water containing 0.1% DMSO and 0.005% Tween 20 as a control was sprayed on the leaves in each dish with a hand sprayer (with 1.2610 5 pascals pressure). The leaves were then inoculated with wild-type isolates (Hd3, Hd11, and Dz21) or pyrimorph-resistant mutants (R3-1, R3-2, R3-3, R11-1, R11-2, and R-21) by placing one mycelial plug (obtained from the margin of a 4-day-old colony plates) on the middle of each leaf. Each isolate was represented by 12 replicate leaves. The Petri dishes with leaves were placed in a growth chamber (80% relative humidity, 25uC, 12-h light/12-h dark). After 4 days, the lesion area on each leaf was measured. The percentage of protection was calculated as % protection = 100 (1x/y), where x and y are lesion areas on pyrimorph-treated and control bell pepper leaves, respectively.

Control of pyrimorph-resistant mutants and wild-type
isolates on pepper plants treated with pyrimorph. Pepper plantss (cultivar Tedaqiemen) with four true leaves growing in plastic trays (540 mm6280 mm, 50 plants per tray) were treatd with 450 g ha 21 pyrimorph or water containing 0.1% DMSO and 0.005% Tween 20 as a control 24 h before they were inoculated with zoospore of wild-type isolates (Hd3, Hd11, and Dz21) or pyrimorph-resistant mutants (R3-1, R3-2, R3-3, R11-1, R11-2, and R-21). Suspensions of the isolates containing at least 10 4 zoospores/ml were prepared as described earlier. A pipette was used to deliver 3 ml of each zoospore suspension to the soil surface near the roots of each plant. Each isolate was represented by ten replicate plants. The trays were placed in a greenhouse (2762uC, 80% relative humidity, and 12-h light/12-h dark) and were watered twice daily. The severity of disease was assessed 7 days after inoculation based on a scale of 0-5 and the percentage of protection was calculated as described earlier. The experiment was performed twice. Cross-resistance. The sensitivity of pyrimorph-resistant mutants and pyrimorph-sensitive isolates to fungicides with different modes of action was determined. The experiment included three CAA fungicides (flumorph, dimethomorph, and mandipropamid), azoxystrobin (a quinone outside inhibitor), chlorothalonil (chloronitrile), cyazofamid (a quinone inside inhibitor), cymoxanil (cyanoacetamideoxime), etridiazole (thiadiazole), fluazinam (an uncoupler of oxidative phosphorylation), metalaxyl (phenylamide), and zoxamide (benzamide). EC 50 values were calculated as described earlier but with the fungicide concentrations as indicated in Table 1. Each combination of isolate and concentration was represented by three replicate plates, and the experiments were performed twice.

Statistical Analysis
Data were analyzed by the general linear model (GLM) procedure with Statistical Analysis System software (version 9; SAS Inc., Cary, NC). Fisher's least significant difference (LSD) multiple range test was used to separate means. For assessment of cross-resistance, EC 50 values were transformed to logarithmic values (logEC 50 ) and subjected to Spearman's rank correlation analysis.
Cloning and Analysis of the PcCesA3 Gene of the Parental Wild-type Isolates and Pyrimorph-resistant Mutants in P. capsici Bioinformatics. For the identification of the PcCesA3 gene in P. capsici, CesA3-specific consensus primers were designed on the basis of the multiple aligned CesA3 sequences (the GENBANK/ EMBL/uniprot data libraries) in Phytophthora infestans (ABP96904), Plasmopara viticola (ADD84672), and the genome data of P. capsici v1.0 (http://genome.jgi-psf.org/PhycaF7/PhycaF7.home.html). The multiple alignment was generated using ClustalW [39].
Cloning and analysis of the PcCesA3 gene of the parental wild-type isolates and pyrimorph-resistant mutants in P.
capsici. DNA was extracted from frozen mycelium by a CTAB (hexadecyltrimethylammonium bromide) procedure [40]. The  MyCycler TM Thermal Cycler (Bio-Rad) with the following parameters: an initial preheating for 5 min at 95uC; followed by 35 cycles of denaturation at 94uC for 30 s, annealing at 62uC for 30 s, and extension at 72uC for 4 min; and a final extension at 72uC for 10 min.
All PCR products were analyzed on 1% agarose gels, and fragments of expected size were purified from the gels with the Easy Pure Quick Gel Extraction Kit (TransGen Biotech Co., Beijing, China). The purified fragments were cloned into a pEASY-T3 cloning vector (TransGen Biotech Co., Beijing, China) and further processed following the manufacturer's instructions. Sequencing reactions were done by Invitrogen life Technologies (Beijing, China). The PCR products were also directly sequenced for heterozygous alleles. A BLASTN search was carried out in the NCBI database (http://www.ncbi.nlm.nih.gov/) to identify correct clones. Prediction of complete PcCesA3 open reading frames (ORFs) was done using the NCBI ORF finder program (http:// www.ncbi.nlm.nih.gov/projects/gorf/). The programs in the DNAMAN software (5.2.9 Demo version) were used to predict the PcCesA3 amino acid sequences and to compare amino acid sequences of the wild-type isolates with those of the pyrimorphresistant mutants.
An Allele-specific PCR Method for Rapid Detection of the Novel Point Mutation in the PcCesA3 Gene that Confers High Resistance to Pyrimorph According to the single point mutation at codon position 1077 in the PcCesA3 gene from pyrimorph-resistant isolates of P. capsici, we designed an allele-specific pair of primers. Based on the single point mutation from C to A at the codon position 1077, a nucleotide A at the 39-end of the forward primer R1077A (59-TCTTCGGGTTATTCGTGATGAGCA-39) was designed to match the A at this codon position. To improve the specificity of traditional allele-specific PCR amplification [41][42][43][44], an artificial mismatch base T was introduced at the second nucleotide at the 39-end of the primer, and the new primer was named R1077B (59-TCTTCGGGTTATTCGTGATGAGTA-39). Reverse primer R1077 (59-CTTCGTAGTAGACCTGCCACA-39) and non-specific forward primer CF1077 (59-GTTCTTCGGGTTCTTCGTAATGAG-39) were designed on the basis of the PcCesA3 sequence.
The specificity of the allele-specific PCR primers was determined with the pyrimorph-resistant mutants (R3-2, R3-3, R11-1) and the corresponding parental isolates (Hd3, Hd11). The PCR was performed as described earlier. PCR amplification parameters were as follows: an initial preheating for 5 min at 95uC; followed by 35 cycles of denaturation at 94uC for 30 s, annealing at 58uC for 30 s, and extension at 72uC for 1 min; and a final extension at 72uC for 10 min. When the PCR was completed, a 5-mL aliquot of PCR product from each sample was analyzed by electrophoresis using a 1.5% agarose gel in TAE buffer.

Baseline Sensitivity of P. capsici to Pyrimorph
A total of 226 isolates of P. capsici from 40 locations in 24 provinces in China were investigated ( Table 2). The mean EC 50 values varied among locations and were highest (2.0095 mg/ml) for isolates from Wuhu, Anhui Province, and lowest (1.0232 mg/ ml) for isolates from Bayannuoer, Inner Mongolia Autonomous Region. However, the variation in EC 50 values did not follow a geographical pattern ( Table 2). The frequency distribution of EC 50 Table 4. Fitness of pyrimorph-resistant mutants (mutant designations begin with the letter R) and the corresponding parental isolates of Phytophthora capsici a .

In vitro b
In planta c Isolate Sporulation (10 5 /cm 2 ) Cystospore germination (%) Lesion area (cm 2 ) Sporulation (10 5 /cm 2 ) Disease score d values was unimodal (Figure 1), ranging from 0.7578 to 3.1610 mg/ml, with a mean of 1.4261 (60.4002) mg/ml. The low EC 50 values and their narrow range indicated the absence of pyrimorph-resistance among the field populations.

Generation of P. capsici Mutants Resistant to Pyrimorph
When exposed to pyrimorph, a total of 12 pyrimorph-resistant mutants with spontaneous mutation were obtained from parent isolates Hd3, Hd11, Hx18, and Dz21. The survival frequency, calculated as the number of resistant mutants divided by the total number of inoculations, was approximately 1610 24 . However, no resistant mutants were obtained from other parent isolates Hb1-17, Mq12, N10, Pc112, Tj3-11, and Xs2. Also, in spite of many attempts, no resistant mutants were obtained from zoospores of the wild-type isolates Dz21, Hx18, Tj3-11, or Xs2.

Characterization of P. capsici Mutants Resistant to Pyrimorph
Level and stability of pyrimorph resistance. The initial EC 50 values for the pyrimorph-resistant mutants ranged from 10.697 to 68.677 mg/ml, and the initial RF values ranged from 10.67 to 56.02 (Table 3). After 10 transfers on fungicide-free PDA, the resistance of some pyrimorph-resistant isolates had changed somewhat, as indicated by FSC values other than 1.0 (Table 3). For example, resistance increased for mutants R18-6 but decreased for R11-2, R18-3, R18-4, and R18-5 ( Table 3). The pyrimorph resistance of selected single-zoospore progeny of the mutants (R11-1, R11-2, R3-2, and R3-1) was stable in that .50 single-zoospore progeny of each mutant grew well on pyrimorphamended PDA after 10 asexual generations. These data indicate that the pyrimorph resistance of the mutants was relatively stable.
Mycelial growth as affected by temperature in vitro. For all the pyrimorph-resistant mutants and parental isolates of P. capsici tested, the optimal temperature for mycelial growth was 25uC ( Figure 2). For Hd3 and Dz21 and their corresponding resistant mutants, the mutants grew at the same rate or faster than the parental isolates at 15,20,25,30, and 37uC (Figure 2A, D); in these cases, the general pattern was that mycelial growth was greater for the mutants than for parents at all temperatures tested. For Hd11 and the corresponding resistant mutants (Figure 2B), the mutants grew at the same rate or slower than the parent isolate at 15 and 20uC but grew at the same rate or faster than their parental isolate at 25, 30, and 37uC. This indicated that the growth for this group of resistant mutants was decreased at temperatures ,25uC but increased at temperatures $25uC. For Hx18 and its resistant mutants ( Figure 2C), the mutants grew at the same rate or slower than Hx18 at 15uC but faster at 20, 25, 30, and 37uC (with the exceptions of R18-2 and R18-3 at 20 and 37uC). This indicated that mycelial growth of resistant mutants derived from Hx18 was reduced at 15uC but enhanced at temperatures .15uC. In summary, the response of the pyrimorph-resistant mutants to different temperatures varied, indicating that resistance is independent from growth characteristics at various temperatures.  isolates and mutants that were tested, sporulation and cystospore germination for the pyrimorph-resistant mutants were lower than or similar to sporulation and germination for the corresponding parental isolates both in vitro and on detached bell pepper leaves (Table 4). However, cystospore germination was higher for R3-1 and R3-3 (p,0.05) than for the parental isolate Hd3 (Table 4).
Virulence on detached bell pepper leaves and on pepper plants. Both pyrimorph-sensitive isolates and pyrimorph-resistant mutants were virulent, causing typical and severe symptoms on detached bell pepper leaves and on entire plants (Table 4). However, the size of the lesions caused by the resistant mutants was either significantly smaller than (p,0.05) or similar to that caused by their parent isolates on detached leaves. On entire plants, disease scores of all the resistant mutants were either less than or similar to those of the parental isolates ( Table 4). -type  isolates  on  bell  pepper  leaves  treated  with pyrimorph. Both the wild-type isolates and the pyrimorphresistant mutants caused typical and severe symptoms on DMSO (0.1%)-treated bell pepper leaves (data not shown). On leaves treated with 750 mg of pyrimorph/ml, disease control was highest for the wild-type isolates, intermediate for the moderately resistance mutants, and low for the highly resistant mutants (resistance ratings were based on the EC 50 values in Table 3). More specifically, disease control was 83, 81, and 84% for the wild-type isolates Hd3, Hd11, and Dz21; 77, 76, and 74% for the moderately resistant mutants R3-1, R11-2, and R-21; and 0, 28, and 0% for the highly resistant mutants R3-2, R3-3, and R11-1. Disease control was significantly lower (p,0.05) with the highly resistant mutants than with the moderately resistant mutants or the wild-type isolates.

Control of pyrimorph-resistant mutants and wild
Control of pyrimorph-resistant mutants and wild-type isolates on pepper plants treated with pyrimorph. For pepper plants treated with 450 g ha 21 pyrimorph, disease control was significantly higher (p,0.05) for the wild-type isolates and the moderately resistance mutants than that for the highly resistant mutants. More specifically, disease control was 72, 76, and 75% for the wild-type isolates Hd3, Hd11, and Dz21; 71, 64, and 71% for the moderately resistant mutants R3-1, R11-2, and R-21; and 7, 13, and 0% for the highly resistant mutants R3-2, R3-3, and R11-1. Pyrimorph was almost ineffective in controlling Phytophthora blight caused by the highly resistant mutants. Cross-resistance. EC 50 values for pyrimorph were highly and positively correlated with EC 50 values for the CAA fungicides dimethomorph, flumorph, and mandipropamid ( Figure 3A-C); the correlation was also positive and significant for azoxystrobin ( Figure 3D). Based on EC 50 values, mutants with high resistance to pyrimorph had high resistance to the CAA fungicides dimethomorph, flumorph, and mandipropamid. However, mutants with moderate resistance to pyrimorph were sensitive to these three CAA fungicides. No cross-resistance was detected between pyrimorph and the other fungicides tested (chlorothalonil,  . Specificity of allele-specific PCR primers for detection of Phytophthora capsici mutants with high pyrimorph resistance. R3-2, R3-3, and R11-1 are mutants with high resistance to pyrimorph and were obtained by exposure to the fungicide on agar; isolates Hd3 and Hd11 are the corresponding parental isolates. doi:10.1371/journal.pone.0056513.g005 resistant mutants (Hd11 and R11-1; Hd3 and R3-2) (unpublished data), mutants with high resistance can probably survive in the field although with some loss of fitness. It is therefore necessary to monitor pyrimorph resistance in P. capsici populations in the field. To facilitate this monitoring, we designed a molecular diagnostic method for detection of the mutation of Q1077K in pyrimorphresistant isolates of P. capsici.
Compared to conventional fungicide-resistance assays, the allele-specific PCR method described here is rapid; it can determine the pyrimorph-resistance phenotype in approximately 4 h by direct sampling of diseased tissue. The AS-PCR primers designed in our study effectively identified pyrimorph resistance in P. capsici. Unlike the traditional AS-PCR primers, the new primers contain an additional mismatch nucleotide 'T' at the second nucleotide of the 39-end; the introduction of this mismatch was previously reported to increase specificity of the allele-specific primer [31,41]. The AS-PCR primers described here will be useful for detecting pyrimorph-resistant isolates of P. capsici in field populations.
As noted earlier, the pyrimorph-resistant mutants of P. capsici with a mutation of Q1077K were also resistant to CAA fungicides. This indicates that this newly discovered nucleotide substitution at codon position 1077 can confer CAA fungicide resistance. Although resistance based on this novel mutation has not been detected in the field, the method developed here will serve as a supplement to monitor the development of point mutations in PcCesA3 conferring resistance to CAA fungicides in P. capsici.