Materials and methods

Isolates of E. perseae. Five isolates of E. perseae were selected from a larger pool of 25, all collected from various agroecological zones within Michoacán. These isolates are stored in the Collection of Phytopathogenic Fungi at the Facultad de Agronomía “Presidente Juárez” of the Universidad Michoacana de San Nicolás de Hidalgo, located in Uruapan, Michoacán.

To determine the isolates for examination, the sampling areas were categorized based on their elevation in meters above sea level (m.a.s.l.). The altitude range of 1200-2300 m.a.s.l. was identified as optimal for avocado cultivation, leading to the identification of five distinct areas. Within each of these areas, a single isolate was randomly chosen as follows: 1) ECC isolate from Cerro Colorado (1200-1420 m.a.s.l.), 2) EZU isolate from Zumpimito (1421-1620 m.a.s.l.), 3) EZA isolate from Zacandaro (1621-1820 m.a.s.l.), 4) EAR isolate from Ario de Rosales (1821-2080 m.a.s.l.), and 5) ECA isolate from Canacuas (2081-2300 m.a.s.l.). Given their slow growth, with a rate of 0.29 to 0.35 mm per day, only these five isolates were considered for evaluation in this investigation.

To conduct in vitro and pathogenicity assessments, the strains were reactivated and introduced into Petri dishes containing Potato Dextrose Agar (PDA) culture medium, with each dish holding 10 mL of the medium. These cultures were kept under controlled conditions of 28 °C temperature and 80% humidity. The experimentation resulted in the creation of 60 replicates, with six replicates established for each of the specified areas.

E. perseae pathogenicity and virulence assays. One-year-old nursery avocado plants of the Flor de María and Méndez varieties were used. These plants bore 6-9 cm fruits, corresponding to the fruit filling stage. These varieties were chosen due to known susceptibility to this fungus from earlier research. Notably, the Méndez variety is owned by the Californian Brokaw nursery, while the Flor de María variety is undergoing registration for plant breeders’ rights (^{Compean, 2020}). The plants were housed at the nursery of the Facultad de Agrobiología “Presidente Juárez.”

One month before starting the experiment, the plants were treated with Switch^® fungicide (1 g per liter of water) and Oleotech higuerilla (0.35 mL per liter of water) to eliminate potential pathogens. Additionally, the plants received 2 g monopotassium phosphate fertilizer each and were irrigated every third day with approximately 2 liters of water per plant to maintain health.

A double inoculation was carried out on avocado fruits. Firstly, 8-10 holes were created in an approximately 1.5 cm diameter area of the avocado’s outer skin using a fine entomological pin (number zero). This was done to make it easier for the fungus to enter. Following this, a piece of PDA culture medium containing E. perseae growth, with a diameter of 5 mm, was placed on the punctured area using a punch tool.

In addition to this method, another type of inoculation was performed. This involved putting a piece of cotton soaked in a spore suspension with a concentration of 1 x 10⁶ conidia mL^-1 onto the avocado fruit. The fungus used for inoculation was five months old. After inoculation, each avocado fruit was covered with parafilm paper and then enclosed in a polyethylene bag. This created a humid environment and was maintained for a period of three days.

To assess the virulence of the fungus, several parameters were monitored over a span of 150 days. These parameters included the number and size of spots that appeared on the fruit, the time it took for the first symptoms to show, and the rate of fruit drop. Observations were conducted every three days throughout this 150-day period.

The experimental design followed a complete randomization approach. This involved using five different isolates, with each isolate having three replicates. Additionally, there was an absolute control group with no inoculation. Each individual avocado fruit served as a replicate in this setup. To analyze the data, statistical methods were employed, including Tukey’s test with a significance level set at P<0.05.

Sensitivity of E. perseae isolates to different fungicides in vitro. Five chemical fungicides and four products approved for use in orchards following organic practices were employed in the study. The chemical fungicides consisted of Azoxystrobin 25SC (0.4 mL formulated product per liter of water), Thiabendazole 60WP (0.6 grams formulated product per liter), Pyraclostrobin 25CE (1.0 mL formulated product per liter), Cyprodinil+Fludioxonil 62.5WG (1.0 gram formulated product per liter), and Azoxystrobin+Propiconazole 20SE (0.4 mL formulated product per liter). On the other hand, the organic management products were Copper sulfate (1.0 gram formulated product per liter), Copper gluconate (2.0 mL formulated product per liter), Copper oxychloride (1.0 gram formulated product per liter), and Larrea tridentata extract (2.0 mL formulated product per liter). These fungicides and products were tested against the five different E. perseae isolates. Each compound was applied at its recommended dose as a formulated product. A control group without any fungicide was also included for comparison. The research was organized using a completely randomized design. Each fungicide-fungus combination, along with the control, was replicated five times. In this experimental setup, one Petri dish was used as one replication.

PDA culture medium was utilized for the bioassay. After sterilizing the PDA, it was maintained at a temperature of 50 °C in Erlenmeyer flasks. The designated amount of each fungicide was then introduced into each flask. The mixture was subsequently poured into Petri dishes and allowed to solidify over a 24-hour period. Following this, a round agar disc with a diameter of 8 mm, containing mycelium from each of the five pathogen isolates, was individually placed at the center of the dishes. The control group consisted of each isolate grown on PDA without any fungicide. In total, five dishes were prepared for each fungicide.

The experiment was conducted under conditions of 24 °C temperature and 80% relative humidity. Darkness was maintained throughout the duration of the assay. As the experiment commenced, the diameter of growth for each E. perseae isolate was measured weekly using a vernier caliper. For the fungicides approved for use in organic orchards (copper sulfate, copper gluconate, copper oxychloride, and Larrea tridentata extract), measurements were taken for a span of one month. In contrast, measurements for the chemical fungicides (azoxystrobin, thiabendazole, pyraclostrobin, cyprodinil+fludioxonil, and azoxystrobin + propiconazole) extended over a period of three months. This difference was due to the relatively shorter half-life of organic products (15-20 days) compared to chemicals (90-120 days). Moreover, the sluggish growth of E. perseae, which could take up to a year to fill a Petri dish, necessitated an extended observation period. The calculation of the Mycelial Growth Inhibition Percentage (MIP) was performed using the formula established by ^{Pandey et al. (1982}).

Where dc = diameter of the control colony, dt = diameter of the treatments. MIP data were subjected to analysis of variance (ANOVA) and means were compared with Tukey’s test (P < 0.05). The JMP program of SAS Institute was used.

Results and discussion

All the tested isolates of E. perseae exhibited the development of small purple spots. As the disease advanced, these spots grew in size and eventually merged together, resulting in the formation of corkosis within 150 days. These visual indications were linked to the occurrence of avocado purple blotch disease, as depicted in Figure 1. The observed symptoms aligned closely with the descriptions provided by ^{Everett et al. (2011}), ^{Fan et al. (2017}), and ^{Morales (2017}; 2022).

Figure 1 Symptoms of purple blotch on avocado fruit produced in pathogenicity tests by isolates of E. perseae, on Flor de María (top) and Méndez (down) varieties. A) ECC (1,200-1,420 m.a.s.l.), B) EZU (1,421-1,640 m.a.s.l.), C) EZA (1,641-1,860 m.a.s.l.), D) EAR (1,861-2,080 m.a.s.l.), E) ECA (2,081- 2,300 m.a.s.l.). 

Symptoms appeared on the Flor de María variety first, at an average of around 41 days. The Méndez variety exhibited symptoms after 91 days. Furthermore, the Flor de María variety displayed a larger number and size of spots, averaging 7-8 spots with an average size of 3 mm. In contrast, the Méndez variety exhibited an average of 2-3 spots with an average size of 2 mm. These observations were made one month after the initial symptom manifestation (Figure 2).

Figure 2 Comparison of Flor de María and Méndez varieties. A) Degree of damage, B) Presence of symptoms (days), C) number of spots and D) size of spots of cultivars Flor de María and Méndez. Mean values ± standard deviation. Different letters indicate statistically significant differences according to Tukey’s test (α = 0.05). 

Based on the assessed parameters, a notable difference in susceptibility between the Flor de María and Méndez varieties is evident. This observation aligns with the findings of ^{Morales (2022}), who noted that among different avocado varieties inoculated with E. perseae, the Flor de María variety exhibited higher susceptibility and more pronounced symptoms. This contrasts with Morales’ (2017) report, which suggested a clear inclination of E. perseae to target the Méndez cultivar with greater frequency and severity in field conditions.

However, it’s important to consider that the Méndez variety holds a larger share of cultivation in the field compared to the Flor de María variety. This greater prevalence might explain why the susceptibility of the Flor de María variety to this disease hasn’t been as noticeable.

In terms of the virulence level of isolates from different agroecological regions, notable findings were observed. For the Flor de María variety, the fruits inoculated with isolates from Cerro Colorado (ECC) ranging from 1200 to 1420 meters above sea level and Canacuas (ECA) ranging from 2081 to 2300 meters above sea level, displayed symptoms earliest at 33 days. Among these isolates, the ECC strain stood out with a statistically significant count of 15 spots. However, there were no noteworthy variations in the size of the spots (Figure 3).

Figure 3 Comparison of presence of symptoms and number of spots of isolates from each agroecological area on Flor de María variety. Mean values ± standard deviation. Different letters indicate statistically significant differences according to Tukey’s test (α = 0.05). 

Symptoms emerged on the Méndez variety at 87 days, with the Zacandaro (EZA) and Canacuas (ECA) isolates exhibiting the earliest signs. In terms of the number of spots, it was the EZA isolate that displayed the greatest count on the fruit. Meanwhile, variations in the size of the spots were not statistically significant (Figure 4). Based on the assessed criteria, the ECC isolate demonstrated the highest level of virulence in the Flor de María variety, whereas the EZA isolate exhibited the greatest virulence in the Méndez variety. Subsequently, the ECA isolate displayed intermediate virulence levels in both varieties.

Figure 4 Comparison of presence of symptoms and number of spots of isolates from each agroecological area in Méndez variety. Mean values ± standard deviation. Different letters indicate statistically significant differences according to Tukey’s test (α = 0.05). 

The origin of E. perseae isolates from distinct agroecological areas appears to have exerted an influence on their virulence. This inference is supported by a statistically significant disparity observed in the manifestation of symptoms and the count of spots. ^{López (2006}) documented variations in the virulence of Histoplasma capsulatum isolates across diverse geographic regions. This phenomenon can be attributed to the organism’s adaptation to distinct environmental circumstances, leading to variations in its behavior, including incidence and severity, based on the prevailing conditions.

Products authorized for use in organically managed orchards. All the products evaluated lost effect with the passage of time. However, L. tridentata and copper oxychloride showed the highest mycelium inhibition percentage of 58% from the first week. While copper gluconate showed very low mycelial inhibition from the first week with 22% (Table 1).

As outlined by ^{IRET (2022)}, the average lifespan of a fungicide denotes the duration (measured in days) required for the conversion of 50% of the substance into other forms within various environmental components (namely water, air, soil, and biota). Additionally, ^{TECAGR (2013)} points out that the effectiveness of organic products diminishes within a span of 1 to 7 days. Nevertheless, the extent of this decline hinges on the product’s formulation and prevailing environmental conditions.

The observed impact of L. tridentata on E. perseae isolates, leading to a 58% inhibition of fungal mycelium (as shown in Table 1), might be attributed to the presence of nordihydroguaiaretic acid within the extract. This notion finds support in the findings of ^{Arteaga et al. (2005}), ^{Gowan (2022)}, Vargas et al. (2006), and ^{Lira (2003}), who attest to the compound’s potential for curbing aflatoxin-producing fungi. In line with this, E. perseae is known for its synochrome production-a type of toxin akin to those originating from the Aspergillus genus-according to the insights of ^{Daub and Chung (2009)}.

Copper oxychloride exhibited a higher level of inhibition (57.5%) against the mycelium of the tested isolates compared to the other copper-containing compounds. Copper sulfate displayed an inhibition rate of 42%, while copper gluconate exhibited 21.8% inhibition (Table 1). This difference in inhibition could be attributed to the concentration of copper oxychloride, which contains 35-70% metallic copper (^{Adama, 2021}), and its higher ionization potential, resulting in a greater presence of Cu2+ ions. These ions are responsible for the fungicidal action and the inhibition of spore germination (^{TECAGR, 2013}), surpassing the effects of copper sulfate (25%) and copper gluconate (<9%) (^{Adama, 2021}). Furthermore, the lower solubility of copper oxychloride causes a slower release of copper ions, leading to a prolonged and sustained effect (^{TECAGR, 2013}). Copper demonstrates its impact on six distinct structures or processes within the fungal cell: the nucleus, ribosome, mitochondrion, smooth and rough endoplasmic reticulum, plasma membrane, and chromatin. This multifaceted approach contributes to its enhanced efficiency (^{Adama, 2021}).

The relatively low in vitro inhibition rate of copper gluconate at 21.8% against E. perseae isolates (Table 1) might be attributed to its distinct interaction with the plant, as it was specifically formulated to enhance copper assimilation within plants (^{ATEEC, 2022}). ^{Kirkby and Römheld (2007}) have pointed out that copper contributes to various plant processes, including photosynthesis, respiration, and the detoxification of superoxide radicals. Additionally, copper aids in the synthesis of phytoalexins that impede spore germination and fungal growth, as well as the formation of lignins that establish a physical barrier, bolstering plant resistance against diseases. Given these roles, evaluating its effectiveness in field conditions would be a worthwhile endeavor.

Each evaluated isolate from the various agroecological regions displayed distinct levels of sensitivity, mirroring findings by ^{Espinoza et al. (2017}). When examining the sensitivity of 60 Colletotrichum acutatum isolates to fungicides thiophanate-methyl and azoxystrobin in two strawberry-producing regions, these researchers found notable variations. In the Maravatío Valley, situated at 2032 meters above sea level, the mean effective dose (DE₅₀) of thiophanate-methyl ranged from 0.3 to 9.7 mL L^-1, while in the Zamora Valley (at 1580 meters above sea level) it spanned from 1.4 to 3.0 mL L^-1. For azoxystrobin, the range was 0.04 to 0.36 mL L^-1 in Maravatío and 0.07 to 0.99 mL L^-1 in Zamora. This underscores the distinct sensitivity levels across different regions.

Conversely, as outlined by Fuentes et al. (2014), fungi-generated melanins can offer protection against certain antifungal agents. In the case of E. perseae, it produces elsinocromos starting from the second month.

Treatment with L. tridentata exhibited the most pronounced inhibitory effect, registering a 74% growth reduction in the Zumpimito (EZU) isolate and 75% in the Canacuas (ECA) isolate. Conversely, the Zacandaro (EZA) isolate demonstrated a relatively milder impact, resulting in a 34% growth reduction. In the case of the EZU isolate, copper oxychloride showcased robust growth control, achieving a 72% inhibition rate that significantly outperformed the other isolates. Among them, copper sulfate demonstrated the highest inhibition rate at 60%, specifically in the ECC isolate. However, its inhibitory effect was least pronounced in the EAR isolate, registering a 27% reduction in growth. Conversely, when applying copper gluconate treatment, no notable distinctions in mycelial growth emerged among the evaluated isolates from the agroecological regions (Table 2).

Chemical products. The bioassays involving chemical agents were conducted over a span of three months. During the initial month, all products managed to completely inhibit the growth of mycelium. Moving into the second month, azoxystrobin+propiconazole, thiabendazole, and azoxystrobin achieved a remarkable outcome: there was an absence of mycelial growth, thus resulting in a 100% inhibition rate. In contrast, the control group exhibited a growth of 3 cm in diameter. Both cyprodinil + fludioxonil and pyraclostrobin displayed a decline in effectiveness during the second month of isolate evaluation, registering an inhibition rate of 93%. This trend might stem from the efficacy duration of the product, linked to its degradation timeline. According to ^{Syngenta (2022)}, cyprodinil + fludioxonil remains effective for a period ranging from 10 to 30 days, pyraclostrobin from 2 to 36 days, and azoxystrobin from 17 to 30 days. In contrast, propiconazole sustains its efficacy for up to 85 days, while thiabendazole remains effective for as long as 120 days. However, it’s noteworthy, as stated by ^{NPIC (2022)}, that fungicides may exhibit diverse half-lives under varying environmental conditions.

During the third month, the effectiveness of azoxystrobin+propiconazole and thiabendazole products persisted at a 100% inhibition rate against mycelial growth. In contrast, cyprodinil+fludioxonil and pyraclostrobin achieved slightly lower inhibitions, at 87.5 and 86.3% respectively. These reductions were statistically significant when compared to the inhibitions achieved by azoxystrobin+propiconazole and thiabendazole. Notably, azoxystrobin’s efficacy declined, resulting in a 93.3% inhibition rate, which, although lower, was not statistically distinct from azoxystrobin+propiconazole and thiabendazole. It’s worth highlighting that all tested products exhibited a significant level of mycelium inhibition when compared to the control group (Table 3).

With cyprodinil+fludioxonil and pyraclostrobin, significant differences were observed at 60 and 90 days, with different percentages of mycelium inhibition in the isolates from the different agroecological areas, some being more sensitive than others (Table 4). A lower susceptibility to these products was observed for the 67% ECA isolate.

Conclusions

The five E. perseae isolates from each agroecological area exhibited varying virulence levels on avocado fruit. Nursery plants with the Flor de María variety were more susceptible to the pathogen based on spot count, spot size, and symptom appearance time. E. perseae isolates from different agroecological areas showed varying in vitro sensitivity to tested fungicides (Larrea tridentata, copper oxychloride, copper sulfate, cyprodinil+fludioxonil, and pyraclostrobin). Larrea tridentata and copper oxychloride achieved 74.1% and 72.4% inhibition, respectively, against the EZU isolate (1421-1640 m.a.s.l.), outperforming copper gluconate, which controlled a maximum of 37.7% inhibition. The fungus displayed higher in vitro sensitivity (100%) to chemical fungicides thiabendazole and azoxystrobin + propiconazole, as well as to orchard-approved organic options: Larrea tridentata and copper oxychloride (58%).