This experiment was conducted at least twice. Calculation of the EC50 values for potassium phosphite of each isolate was determined as described in Adaskaveg et al. . Briefly, the EC50 values were determined by fitting regression lines of the natural log-transformed phosphite concentration against the logit-transformed percent inhibition as compared with the control for each isolate. Regression equations were then solved for concentration at 50% inhibition using Microsoft Excel . The EC50 values for mefenoxam , fluopicolide , and oxathiapiprolin , currently used to control diseases caused by oomycete pathogens, were determined using the spiral gradient dilution method described by Förster et al. . Mefenoxam, oxathiapiprolin, and fluopicolide were dissolved in sterile water to make the corresponding stock solutions at 50, 5, and 100 µg/ml, respectively. Each fungicide was radially applied to a 15-cm 10% cV8A plate using a spiral plater . Isolates were grown for 7 days in the dark at 22°C on 10% cV8A media. Cultures were then cut into strips and placed between sterile hydrophilic cellophane strips on new 10% cV8A plates. Plates were incubated for another 7 days in the dark at 22°C or until the mycelium had evenly covered the strips. Mycelium-covered cellophane strips were then radially applied across the fungicide concentration gradient on the spiral gradient dilution fungicide plate.
The ranges of concentrations used were between 0.0025 to 0.6837 µg/ml , 0.0038 to 1.4054 µg/ml , procona London container and 0.0002 to 0.0572 µg/ml for mefenoxam, fluopicolide, and oxathiapiprolin, respectively. For each isolate, duplicated strips were placed at the opposite location on each spiral gradient dilution fungicide plate . Controls consisted of cV8A plates spirally plated with water, to which cellophane strips were placed as described above. Plates were incubated in the dark at 22°C. Two independent experiments were conducted. Two days after incubation, the mycelial growth of each isolate in each of the control plates was measured. The location where mycelial growth was inhibited by 50%, compared with growth on the control plates, was determined. This distance was used to determine the local fungicide concentration on the agar at the location of 50% inhibition using the SGE software . The EC50 values for each fungicide were calculated using the spiral gradient endpoint software as described in Förster et al. .Nine-month-old avocado rootstocks liners were inoculated with P. cinnamomi colonized millet seed. Millet inoculum was prepared according to the method of Drenth and Sendall . Briefly, 175 ml distilled water was added to 144 g of millet seed. Excess water was poured off and moistened millet was autoclaved in a 250 ml volumetric flask.
Autoclaving was repeated the following day. For each isolate, ten plugs of 6-mm-diameter were cut from the margin of a 4-day-old colony of P. cinnamomi plate and were added to the flask containing the sterilized millet seeds. Inoculated millet was incubated in the dark at 25°C for 21 days. The millet was then harvested, weighed, and homogenized in a plastic bag. A total of 4.4 g of inoculum per plant was used for inoculation by placing the inoculated millet around the root collar. Combination isolate inoculum was produced by thoroughly mixing equal amounts of each corresponding isolate inoculated millet seeds together. Dusaâ and PS.54 ungrafted clonal rootstocks were inoculated with individual isolates, a combination of Northern California isolates , a combination of Southern California isolates , and a combination of all isolates . Control plants were treated with uninoculated millet seeds that were sterilized and prepared in the same manner as the inoculated millet seeds. The experiment was conducted in a greenhouse using a randomized complete block design with five replicates per treatment. The average maximum temperature in the greenhouse was 28°C with an average minimum temperature of 14°C. To confirm that plants were successfully infected, roots samples were excised from harvested plants and 10 to 20 segments per plant were surface sterilized in 70% ethanol and plated onto the PARPH-V8 Phytophthora-selective media identified by Paglaccia et al. .
The A2 clade I contained P. cinnamomi isolates collected from 1989 to 2010, whereas the A2 clade II only contained a unique set of isolates collected in 2009 and 2010 from Southern California avocado growing regions. This study reported significant differences in vegetative growth, fungicide sensitivity, and virulence among all the isolates tested. Moreover, Southern isolates were less sensitive to potassium phosphite and have slower growth rate at 22℃ whereas Northern isolates were more sensitive to potassium phosphite and have higher growth rates. This trend observed suggests a correlation between mycelial growth rate at 22℃and potassium phosphite sensitivity with specific group of isolates. The group of Northern isolates corresponding to the A2 clade I mating type group exhibited higher mycelial growth rates than the Southern group of isolates at 22°C independent of the nutrient media used . Consistent with the results of Zentmyer et al. , there was a significant effect of nutrient media and mycelial growth rate of the P. cinnamomi study isolates, however this effect did not alter the split of the Northern and Southern isolates by their growth rate. The higher sensitivity of the A2 clade I isolates, collected from Northern California, to higher temperatures when compared with the Southern isolates might be explained by the fact that the median temperature is higher in Southern than in Northern California throughout the year.Previous studies reported phenotypic differences in colony morphology, growth rate, and optimum growth temperature among P. cinnamomi isolates, however the majority of these studies compared isolates from different mating types, origin , and host plants . Only a few studies have included P. cinnamomi associated with avocado when assessing phenotypic variability . In the majority of these studies, the authors could not assign a phenotypic trait to a specific group of isolates. Others have reported varying results on the linkage between genotype groups and colony morphology. Dobrowolski et al. reported that colony morphology of Australian P. cinnamomi isolates grouped with a particularly genotype. In contrast to these results, colony morphology among the twelve isolates tested in this study did not vary significantly. Lopez-Herrera and Perez-Jimenez reported significant differences on colony morphology among P. cinnamomi A2 isolates collected from avocado trees in Spain, but these isolates did not exhibit significant differences on mycelial growth rates. California produces 95% of the avocado crop for the U.S.A and PRR caused by P. cinnamomi is responsible for commercial losses totaling $40 million annually statewide. Mefenoxam and phosphite applications are widely used to prevent and combat this disease , however, there is a notably preference for phosphite products over mefenoxam among the growers and this is particularly true for California growers. This preference could explain why mefenoxam-resistant isolates were not found in this study. Moreover, the range of the EC50 values for mefenoxam was consistent with previous values reported for P. cinnamomi in the U.S.A . Duan et al. reported minor variation in sensitivity to mefenoxam among P. cinnamomi isolates collected from diseased ornamental plants in South Carolina. The EC50 values of the majority of these isolates were less than 0.1 µg/ml. Hu et al. found that there were more variations in mefenoxam sensitivity among P. cinnamomi isolates collected from different host species than from the same host.
In this study, cut flower transport bucket the authors classified isolates as sensitive to mefenoxam when their EC50 values ranged from 0.01 to 0.02 µg/ml and as intermediate when EC50 values ranged from 0.03 to 0.08 µg/ml. The EC50 values for mefenoxam in this study were also less than 0.1 µg/ml suggesting that the repeated use of this fungicide to control PRR in nurseries and avocado orchards does not appear to have reduced the mefenoxam sensitivity of P. cinnamomi isolates. In contrast to mefenoxam, we detected a significant variability among the isolates in potassium phosphite sensitivity but we did not find potassium phosphite-resistant isolates. The Northern isolates corresponding to the A2 clade I group had EC50 values of < 34.1 µg/ml, whereas the Southern isolates had EC50 values of > 98.9 µg/ml. It is worrisome that the Southern group of isolates including the A2 clade II isolates that exhibited more virulence in the moderate resistant Dusaâ, which is the current industry standard rootstock among California avocado growers, were less sensitive to potassium phosphite. Their higher EC50 values could represent a selection from higher doses of potassium phosphite being necessary to suppress and control avocado root rot where these isolates are present. A more detailed study with a larger number of isolates and history of phosphonate applications in the field are needed it to test this hypothesis. Dobrowolski et al. demonstrated that P. cinnamomi isolates exposed to long periods of phosphite treatment in avocado orchards in Australia exhibited reduced sensitivity to phosphite when evaluated on avocado, lupin, and eucalyptus suggesting the onset of resistance to this fungicide. Similar results have been reported for P. cinnamomi isolates from avocado orchards in South Africa . This study reports for the first time, the presence of P. cinnamomi isolates, collected from PRR diseased avocado roots in California, that are less sensitive to potassium phosphate. To help delay the development of phosphite-resistant P cinnamomi isolates, care should be taken to alternate or mix phosphite products with other effective fungicides with different mode of action to control avocado PRR. Phosphite and mefenoxam rotation with alternative fungicides is commonly used to prevent or reduce the emergence of Phytophthora resistant isolates , however there are no other fungicides tested or registered to control P. cinnamomi in avocado. Fluopicolide and oxathiapiprolin are two new oomycete-targeted fungicides that have been tested for several Phytophthora spp., but not for P. cinnamomi . In this study, we report for the first time that the EC50 values for mycelial growth inhibition of P. cinnamomi avocado isolates are within the range of the EC50 values reported for other Phytophthora spp. using these two fungicides . Gray et al. recently reported the Fluopicolide and oxathiapiprolin EC50 values for several Phytophthora species associated with citrus on California. Fluopicolide EC50 values for P. citrophthora, P. syringae, P. nicotianae, and P. hibernalis ranged from 0.031 to 0.087, 0.02 to 0.0461, 0.039 to 0.095, and 0.017 to 0.018 µg/ml, respectively. Of the four fungicides evaluated in this study, oxathiapiprolin had the lowest EC50 values . This range is similar to EC50 values reported for other Phytophthora spp. including P. sojae, another member of the phylogenetic clade 7, which includes P. cinnamomi . In agreement to our study, Gray et al. , also found that this fungicide had the lowest EC50 values among five different fungicides tested. The authors reported oxathiapiprolin EC50 values for P. citrophthora, P. syringae, P. nicotianae, and P. hibernalis ranged from 0.0002 to 0.0015, 0.0002 to 0.0003, 0.0003 to 0.001, and < 0.0003 µg/ml, respectively. Interestingly, in contrast to potassium phosphite sensitivity, the Southern isolates in this study exhibited lower EC50 values than the Northern A2 clade I isolates. These results suggest the potential to rotate oxathiapiprolin with phosphonates and mefenoxam for controlling avocado PRR in California reducing the risk of the emergence of phosphonate-resistant P. cinnamomi isolates. Significant variation in pathogenicity and virulence among P. cinnamomi isolates collected from different host plants have been previously reported . Few cases have reported no differences in virulence among P. cinnamomi isolates collected from different hosts and unique host including one study in avocado . In this study we found significant differences in virulence among three representative P. cinnamomi isolates from the two A2 clonal groups identified by Paglaccia et al. when we inoculated a moderate resistant avocado rootstock but not the most susceptible one. The use of highly susceptible plants to distinguish between isolates with different levels of virulence is challenging and could explain why we did not detect significant differences in virulence among our isolates when we used a susceptible avocado rootstock . Although the mixture of the A2 clade II isolates were more virulent than the mixture of the A2 clade I isolates , we observed that the mixture of all the isolates together exhibited the same disease severity as the mixture of the less aggressive ones . This observation might be explained by the level of competition between these two groups of might reflect the induction of specific plant defense responses triggered by these two distinct groups of isolates, a more comprehensive study is required to test these two possibilities.