Materials and methods

Characteristics of the orchards. In November of 2019, five avocado orchards, located in the municipality of Zitácuaro, Michoacán, were sampled. All orchards are located in a A(C)(w₁)(w) semi-warm subhumid climate with abundant rainsfall in the summer, winter rains below 5%, intermediate humidity, according to the Köppen climate system, modified by ^{García (1981}), and the established variety is ‘Hass’; in addition, they have heavy, clay-rich soils. Other characteristics of the orchards are indicated in Table 1.

Collection of samples and isolation. Root and soil samples were collected from the rhizosphere, one sample per tree and five trees per orchard, with symptoms of yellowing, defoliation and dieback. The samples were placed in labelled plastic bags in a cooler and transported to the Plant Pathology Laboratory of the Institute of Farming and Forestry Research (Instituto de Investigaciones Agropecuarias y Forestales - IIAF), of the Universidad Michoacana de San Nicolás de Hidalgo (UMSNH). In order to obtain a greater percentage of isolates from roots, the protocols by ^{Erwin and Ribeiro (1996}) and ^{Rodríguez-Padrón et al. (2018}) were modified. Lateral, partially necrotized roots were selected, washed with tap water and dried using sterilized paper towels. Approximately 2 cm cuts were made, with one healthy part and one necrotized part, placed in inclusion cassettes, which were submerged in commercial chlorine bleach at 2% (v/v) for 45 s and rinsed with sterile distilled water three times. The excess water was removed from the tissue cuts using sterilized paper towels and placed in Petri dishes with a NARPH-V8 (Natamycin 0.02 g L^-1, PCNB (Pentachloronitrobenzene) 0.10 g L^-1, Ampicillin 0.27 g L^-1, Rifampicin 0.01 g L^-1, Hymexazol 0.075 g L^-1) selective medium. The dishes were incubated at 25 °C in the dark for 24 to 48 h, until the growth of cenocytic mycelium and with the characteristic P. cinnamomi swellings were observed. The cultures were purified by the successive cultivation of hyphal tips on corn meal agar (HMA, Fluka^®) culture medium.

Compatibility type. The compatibility of the isolates obtained was determined using the reference P. cinnamomi isolate A1 (camellia isolate) and A2 (avocado isolate) which were donated by the University of California, Riverside to the Plant Pathology Laboratory (IIAF). In a Petri dish containing scV8 agar medium (900 mL distilled water, 100 mL Campbell´s V8 juice, 1 g CaCO₃), a mycelium disc of the isolate with compatibility type A1 was placed, and in another dish, the type A2. Likewise, in each dish, a mycelium disc of the isolate of interest was placed, approximately one centimeter away. The dishes were incubated at 25 ºC in the dark for 10 days. The isolates that formed oospores were assigned the complementary compatibility type in the cross.

Morphological characterization. The pure cultures were cultivated on HMA medium and cV8 agar medium (10 g of CaCO₃, 1 L of V8 juice, centrifuged at 4,000 rpm for 20 min) (^{Fernández-Pavía et al., 2020}). The shape of the culture was characterized, along with the appearance of the mycelium. For the production of sporangia, the protocol by ^{Hwang et al. (1975}) was modified. Three 5 mm mycelium discs from each isolate were plated in triplicate in Petri dishes measuring 60 x15 mm with 10 mL of V8 agar medium over a sheet of transparent cellophane and exposed to light for 24 h. The cellophane paper with the mycelium was transferred to new Petri dishes, 10 mL of liquid scV8 medium (900 mL of distilled water, 100 mL of V8 juice, 1 g CaCO₃), were added and incubated for 24 h at 18 ℃ in the dark. The liquid medium was then poured out, they were rinsed three times and were flooded with a salt solution, following the protocol by ^{Chen and Zentmyer, (1970}) (1.64 g of Ca (NO₃)₂, 0.05 g of KNO₃ and 0.48 g of MgSO4 were dissolved in one liter of distilled water and the solution was sterilized for 20 min at a pressure of 15 lb). They were then exposed to white light at 25 ℃ every 24 h, rinsed three times with the salt solution for two consecutive days and the reproductive structures observed were registered (^{Erwin and Ribeiro, 1996}; ^{Abad et al., 2019}).

Pathogenicity tests. To perform the pathogenicity tests, four isolates were chosen to be inoculated on avocado fruits with physiological maturity; they were washed with soap and tap water and sprayed with alcohol at 75% to disinfest them. Two mycelium discs, 6 mm in diameter, from isolates ZITR-1-3, ZITR-2-5, ZITR-3-4 and ZITR-5-3, four days old and grown on HMA, were placed on the sides of the fruits; the discs were then covered in Janel™ brand masking tape. Three fruits were used for every isolate, and after inoculation, they were placed in a moist chamber and incubated at 25 °C. As controls, three fruits were inoculated with HMA discs.

Sensitivity to fungicides. The sensitivity of 15 isolates to potassium phosphite (Fp) (Nutriphite plus magnum 40% Gowan^®) were evaluated, at final concentrations of 0, 5, 10, 25, 50, 100, 300 and 600 µg mL^-1 of i.a. and metalaxyl-M (Mtx) (Ridomil Gold^® 480SL) 0, 0.5, 1, 3 and 5 µg mL^-1 of i.a. A completely randomized experimental design was used with a factorial arrangement with three replications. The effective concentration at which 50% of the mycelial growth was inhibited (value of EC₅₀) was calculated using the traditional method of dilution in agar (^{Gray et al., 2018}). The fungicide was added to the cV8-A medium at 10% to obtain the final concentrations. Discs measuring 6 mm with the mycelium from the four-day-old isolates cultivated on HMA were placed in the center of the Petri dishes containing the different concentrations. After three days of incubation in the dark at 22 °C, the growth of the culture was measured in two perpendicular directions, the measurement of the inoculum disc was subtracted and the average was obtained. The growth was calculated by dividing the diameter of the culture in the Petri dishes with fungicide by the average growth of the control dishes and it was expressed as a percentage (^{Hu et al., 2010}).

Effect of the temperature and culture medium on the growth of the mycelium. Six isolates were selected, based on the percentage of inhibition to Fp, along with three isolates with low percentage of inhibition and three with high percentage of inhibition (an A [high] and a B [low] were added at the end of the identification code of each isolate used in this test), to determine whether there was or was not a relationship between the velocity of growth and sensitivity. Growth rate (mm per day) was determined at 22 and 25 °C in the cV8 agar and HMA media, following the methodology described by ^{Belisle et al. (2019}b). Agar discs, 7 mm in diameter, were cut from the edge of the 4-day-old cultures plated on HMA medium, transferred to the dishes with the corresponding media and incubated in the dark for four days at both temperatures. The diameter of the culture was measured every day.

To calculate the values of the EC₅₀, the average percentage of inhibition of both repetitions was estimated for each concentration of the fungicides based on the percentage observed in the control group, using the formula ICR (%) = (CRT-CRF)/CRT)x100, where ICR is the percentage of inhibition of the radial growth, CRT is the radial growth of the control, and CRF is the radial growth in each of the concentrations. To obtain the EC₅₀, the natural logarithm of each of the studied concentrations was included and a probit analysis was carried out for each isolate related to 50% of its inhibition (^{Adaskaveg et al., 2015}). The experiments were carried out in duplicate.

To determine the effect of the temperature and of the medium on the isolate, the final growth was taken into consideration; the average was obtained and with these values, an analysis of variance and a multiple comparison of averages were carried out (Tukey, 0.05) for a 2x2 factorial experiment. Using the information on the final growth, interaction graphs were created to observe the effect of the factor over the variable. The analyses were carried out and graphs created using R version 4.0.1 (2020-06-06).

Molecular characterization. Isolates ZITR-1-3 and ZITR-5-3 were grown in a green bean medium (green bean and zucchini baby food: Gerber Nestlé^®, 339 g, agar 15 g and water 646 mL) for five days. The mycelium was collected, dehydrated at 39 °C for 24 h, stored at -20 °C for 24 h, ground, and the genomic DNA extracted following an extraction protocol based on CTAB (^{Fernández-Pavía et al., 2020}). The DNA was quantified in a spectrophotometer (Varioskan Flash Thermo Scientific NanoDrop 2000^®) and diluted to obtain a concentration of 12 ng µL^-1. The DNA was amplified by PCR using oligonucleotides for the region of the internal transcribed spacer (ITS). The reactions contained 6.25 μL of GoTaq^® Hot Start Master Mix (Promega, Madison, WI, U.S.A.), 0.0675 μL of each of the oligonucleotides ITS4 (TCCTCCgCTTATTgATATgC) and ITS6 (gAAgAAggTgAAgTCgTAACAAgg) (100 pmol μL^-1), 3 μL of genomic DNA (12 ng) and ultrapure distilled water, until a final volume of 13 μL was obtained. The amplification reactions were carried out in the following conditions: an initial denaturalization for 2 min at 94 °C, 35 denaturalization cycles for 1 min at 94 °C, alignment for 1 min at 53 °C, extension for 1 min at 72 °C, and a final extension for 10 min at 72 °C. The products of the PCR were sequenced (Macrogen, Seoul, South Korea) and the sequences were edited using the programs PreGap and Gap (http://staden.sourceforge.net). The consensus sequences obtained were analyzed with the BLAST program, from the NCBI [National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov)] to compare with other sequences of oomycetes from this database.

Results

Isolates obtained. Plating was carried out for every root sample out of a total of 25 samples. Fifteen isolates were obtained (Table 2) with morphological characteristics of P. cinnamomi, and in the rest of the dishes, bacterial growth was observed, therefore they were discarded.

Compatibility Type. The crosses of the isolates with a P. cinnamomi strain with compatibility type A1 produced plerotic oospores with amphigynous antheridia. The crosses of the isolates with a strain with compatibility type A2 only produced globose chlamydospores (Figure 1D). The results indicate that the isolates are compatibility type A2.

Morphological and molecular characterization. The isolates developed rosaceous shaped cultures with wavy edges, with abundant, cotton-like, coraloid mycelia, coenocytic hyphae with swellings, with globose chlamydospores on HMA culture medium, whereas in the sV8-A medium, the cultures were star-shaped (Figures 1F-G). The isolates produced sporangia without papillae, ovoid, ellipsoid, persistent and with internal proliferation in the cultures flooded with a salt solution (Figure 1A-E). The characteristics observed coincide with those reported for P. cinnamomi (^{Erwin and Ribeiro 1996}; ^{Hardham, 2005}; ^{Abad et al., 2019}). The analyzed sequences of the ITS region of the isolates selected for molecular identification ZITR-1-3 and ZITR-5-3 coincide with those reported for P. cinnamomi.

Pathogenicity tests. The tests on the avocado fruits turned out positive. Seventy-two hours after inoculation (hai) dark circular lesions with irregular edges were found on the epidermis. At 96 hai, the diameter of the lesions were between 3 and 6 cm. The control fruits displayed no symptoms and all the isolates were reisolated from the inoculated fruits, thus proving the pathogenicity completing Koch’s postulates (Figure 1H).

Figure 1 Morphological characteristics and pathogenicity of P. cinnamomi. A) Coraloid mycelium; B-C) Sporangia without papillae, ovoid-ellipsoid; D) Globose chlamydospores in groups; E) Internal proliferation of sporangia; F) Cottonlike, star-shaped culture, in a cV8A medium; G) Rosaceous-shaped culture in a HM medium; H) Pathogenicity in fruits of the isolates (downwards) ZITR-2-5, ZITR-3-4, ZITR-5-3, ZITR-1-3 and control. 

Sensitivity to fungicides. The effect of fungicides Fp and Mtx on the growth of the culture and the percentage of inhibition of the isolates between treatments was significant (P>0.001). No fungicide-resistant isolates were found. With Fp, no statistical differences were found between isolates for growth (P>0.63) or for inhibition (P>0.982) of the mycelium, unlike with the different concentrations, in which significance was found. Additionally, something in particular that was observed was the promotion of the growth of eight isolates (ZITR-1-3, ZITR-1-5, ZITR-2-1, ZITR-2-4, ZITR-3-2, ZITR-3-4, ZITR-5-1, ZITR-5-3) by Fp at 5 μg mL^-1 (1.15-18.61%) (Figure 2). In the comparison of averages, six groups were formed; no significant difference was found in growth between the control (16.76 mm) and the first concentration (16.02 mm). Growth was mainly inhibited at 100, 300 and 600 μg mL^-1, with these concentrations it resulted statistically equal, for both growth and for inhibition (Table 2). The values of the effect of Fp on the inhibition of growth ranged between 54.7 and 74.2 % regarding the control, and for the EC₅₀, between 11.8 and 37.7 μg mL^-1 with a mean of 24.62 μg mL^-1 for potassium phosphite. For Mtx, a lower variation was observed between treatments regarding Fp; at concentrations of 3 and 5 μg mL^-1, the mycelium was mostly inhibited; the percentage of inhibition ranged between 80.4 and 100% and the values of EC₅₀ ranged from 0.08 to 0. 34 μg mL^-1 with a mean of 0. 215 μg mL^-1 of metalaxyl-M. There was no significance in the growth (P> 0.996) or inhibition (P>1.0) of the mycelium among isolates (Table 3).

Figure 2 Inhibition of the mycelial growth of P. cinnamomi isolates collected in Zitácuaro, Michoacán against potassium phosphite. 

Effect of the temperature and culture medium on the growth of the mycelium. The analyses displayed a significant (P>0.001) effect on the mycelium growth rate at both temperatures, in both nutrient media and their interactions. In the test for the comparison of means, six groups were formed and the isolates with the greatest inhibition (%) to Fp presented a greater growth speed than those with the lowest inhibition. However, isolate ZITRA-3-3, with the highest inhibition (%) displayed the lowest growth rate (11.25 mm) and ZITRB-1-5, with the lowest inhibition (%), presented a higher one (21.75 mm), although the latter also displayed a greater variation in data (Figure 3). The growth of isolates ZITRA-1-4, ZITRA-5-5, ZITRB-3-4 and ZITRB-2-4 fluctuated between 18.66 and 15.91 mm. A higher growth rate was observed in cultures at 25 °C and in the cV8-A culture medium, with the exception of isolate ZITRB-1-5, which displayed a higher growth rate at 22 °C (Figure 4). The morphology of the culture varied according to the culture medium. In the isolates in cV8-A, they displayed star-shaped cultures, and in HMA, rosaceous-shaped cultures.

Discussion

This is the first study on P. cinnamomi isolates related to wilt in avocado in eastern Michoacán, in the municipality of Zitácuaro. Other studies on this disease have reported the presence of P. cinnamomi in other areas of the state (^{Ochoa-Fuentes et al., 2007}; ^{Ochoa-Fuentes et al., 2018}; ^{Hernández-Pérez et al., 2019}). The surface with avocado plantations in Zitácuaro has increased in recent years and could potentially be an important production area in the future. However, the presence of the pathogen P. cinnamomi represents a threat to the crop. This area displays certain agroclimatic characteristics, such as the presence of Acrisol soils, which promote the development of this disease (^{Zentmyer, 1980}; ^{Guillén-Andrade et al., 2007}). The sampled trees were adults, established in orchards with high contents of clay and poor drainage, which contributed to the presence of the disease (Zentmyer, 1980; ^{Erwin and Ribeiro 1996}; ^{Hardham, 2005}). Although in this study only isolates with morphological characteristics of P. cinnamomi in trees with root rot were recovered, the presence of other oomycetes is not discarded.

Figure 3 Effect of the temperature and culture medium on the mycelial growth of six P. cinnamomi isolates obtained from avocado roots with rotting 

Figure 4 Effect of 22 and 25 ºC temperatures on six P. cinnamomi isolates from avocado roots with rotting. 

All isolates corresponded to compatibility type A2, which indicates that the oomycete is reproducing asexually. Despite ^{Ochoa-Fuentes et al. (2005}) reporting both compatibility types with molecular markers in Michoacán, they did not prove it with crosses, as performed in this study, which is considered the most accurate way to determine them. The morphological characteristics observed in the isolates coincide with those described for P. cinnamomi according to ^{Abad et al. (2019}). This identity was verified with the amplification of the ITS region in selected isolates and the pathogenicity was proven on avocado fruits. P. cinnamomi is one of the 10 main phytopathogenic oomycetes, it infects over 5000 species (^{Hardham, 2005}) and its pathogenicity has been proven (^{Linde et al., 1999}; ^{Rodríguez-Padrón et al., 2018}; ^{Belisle et al., 2019b}).

For many years, phosphites have been used to control root rot. Avocado producers use potassium phosphite more frequently than metalaxyl-M, due to its higher efficiency and lower cost than others (^{Adaskaveg et al., 2015}; ^{Belisle et al., 2019b}). Our results prove that there are no resistant isolates to the fungicides evaluated in Zitácuaro, Michoacán. However, regarding potassium phosphite, a lower sensitivity was observed in the isolates than towards metalaxyl-M. The values obtained for EC₅₀ fluctuated between 11.8 and 37.7 μg mL^-1 of potassium phosphite, and for growth inhibition, between 54.7 and 74.2% in comparison with the control, which are similar to those obtained by ^{Wilkinson et al. (2001}), who obtained inhibition values ranging from 59 to 100%. Eight of the isolates displayed an enhancement in growth when they were exposed to the concentration of 5 μg mL^-1 of i.a of potassium phosphite, and these results coincide with those by Wilkinson et al. (2001) and ^{Hunter et al. (2018}). This may be due to the possibility that, at low potassium phosphite concentrations, in the cV8 nutrient-rich medium, it may be easier for P. cinnamomi to alter its growth as a survival strategy, along with the fungistatic effect not affecting the mycelial growth (^{Guest and Grant, 1991}; ^{Hunter et al., 2018}). Another reason is the loss in sensitivity of the isolates due to the frequent exposure to the fungicide, since they undergo a pressure of selection, which may eventually lead to the development of resistance, even if there is no sexual recombination (^{Dobrowolski et al., 2008}; ^{Hunter et al., 2018}). Not only has P. cinnamomi, been reported as sensitive to potassium phosphite, also P. syringae and P. citrophthora, although the inhibition of growth of the species in vitro is not indicative of the disease control held in the field as being effective (^{Hao et al., 2021}).

On the other hand, although ^{Marin et al. (2021}) reported the development of resistance in Phytophthora species to metalaxyl-M (mefenoxam), our study observed an average growth inhibition of 88.24% and a EC₅₀ value of 0.08 and 0. 34 μg mL^-1 of metalaxyl-M mL^-1, suggesting that the P. cinnamomi populations obtained from avocado are sensitive to the fungicide. In a similar study, the isolates obtained from ornamental plants in nurseries were less sensitive, with a EC₅₀ value of 0.01-0.08 μg mL^-1 of metalaxyl (^{Hu et al., 2010}). This may be due to the applications being more controlled and less frequent in the nurseries than in the field. In North Carolina, sensitivity tests performed on P. cinnamomi isolated from Christmas trees gave percentages of inhibition that fluctuate between 89 and 100%, despite metalaxyl-M applications being carried out in the region two or three times a year (^{Benson and Grant, 2000}). This confirms that P. cinnamomi is sensitive to metalaxyl-M, along with the fact that high percentages of inhibition of mycelial growth by metalaxyl-M have also been observed in other species such as P. nicotianae (73. 5%) and P. × pelgrandis (97.5%) (^{Pánek and Tomšovský et al., 2017}).

On the other hand, ^{González et al. (2014}) indicated that the V8 culture medium promotes the growth of Phytophthora in regard to other media. These results are consistent with those obtained in this work, since the isolates with the most growth were those plated on V8. A difference in growth was found between both temperatures, with 25 ºC being more favorable, since it is within the optimum growth temperature range of P. cinnamomi that ranges between 24 and 28 ºC (^{Erwin and Ribeiro 1996}).

In Mexico, the sensitivity studies performed on P. cinnamomi isolates to fungicides are scarce. The results of this work provide an estimation of the range of sensitivity to potassium phosphite and metalaxyl-M of P. cinnamomi isolates that cause root rot in avocado produced in the municipality of Zitácuaro, Michoacán. However, it is necessary to carry out further studies which include a greater number of isolates from the main producing municipal areas, in order to help us determine if the populations of the pathogen are becoming tolerant to these fungicides.

Conclusions

The results of this work indicate that in Zitácuaro, Michoacán, P. cinnamomi is the main oomycete present in avocado roots with rot. Out of 25 root samples, 15 isolates were obtained with the morphological characteristics of P. cinnamomi. The isolates were inhibited by more than 98% at concentrations of 300 and 600 μg mL^-1 of potassium phosphite and over 96 % at 3 and 5 μg mL^-1 with metalaxyl- M.