Collection area

Avocado fruits were collected in orchards of the ejidos the Yerba, Camichin de Jauja and San Luis of Lozada belonging to the municipality of Tepic; Likewise, it was also sampled in the ejido of Xalisco that belongs to the municipality of Xalisco. In each of the municipalities indicated above, avocado fruits with anthracnose signs were taken directly from the tree, which were placed in low density polyethylene bags and transferred to the Food Technology Laboratory of the Autonomous University of Nayarit.

Isolation and morphological characterization of phytopathogenic fungus.

The fruits, leaves and flowers were washed with distilled water and 1% sodium hypochlorite to remove dust and external contaminants, then dried with paper towels. Subsequently, thin sections of approximately 2 to 3 mm of the diseased tissue were made, which were disinfected with 1% sodium hypochlorite solution for 3 min, rinsing with sterile distilled water; finally dried with sterile gels and placed in petri dishes containing potato dextrose agar (PDA Dibico), incubated at 28 ±3 °C, with a relative humidity of 80%, checking daily to observe the development of the pathogen. Successiveresharpening with PDA was carried out to purify phytopathogenic fungus (^{Arias et al., 2006}; ^{Rondón et al., 2006}; ^{Santander, 2012}).

The characterization of the fungus strains was placed through a PDA disk with growth of the pathogen in a Petri dish with this same medium, incubated for 7 days at 28 ±3 °C. For the characterization of the strains, the color was evaluated at the front, back and center of the colony, mycelial texture, colony shape, conidial size and shape and colony diameter. The colony diameter was determined by measuring the growth rate with a millimeter rule, obtaining the daily growth rate in cm, as well as the difference of the final diameter minus the initial diameter (mycelial disc placed). Long and wide conidia were evaluated, which were observed by a composite microscope at a magnification of 40x and measured with a micrometric ruler. Fifty randomly selected conidia from each of the isolates were measured. The shape of the conidia was determined based on the characteristics indicated by ^{Sutton (1992)}.

Phytopathogenicity tests

Once the avocado fungi were isolated, tests were carried out to determine which strains were able to generate symptoms of disease in the fruit and thus to verify the pathogenic capacity of them (^{Benbow and Sugar, 1999}).

The anthracnose free avocados were washed and disinfected with 1% sodium hypochlorite solution for 3 min, then washed with sterile distilled water and dried under asepsis conditions in a laminar flow hood. Two wounds per fruit were realized with a punch that allowed to create cavities of 4 mm of diameter by 3 mm of depth in different zones. A 25 µl aliquot of the Colletotrichum spp spore suspension was added to the wounds. (1x105 esp mL-1) taken from a culture with 7 days of growth. Fruits that were treated only with sterile distilled water were used as controls (^{Benbow and Sugar, 1999}; ^{Bautista, 2013}). Five fruits were used for each of the twenty strains and each was repeated three times.

The treated avocados were placed in plastic chambers with relative humidity higher than 90%, which favored the development of these phytopathogenic fungi. The chambers were maintained at room temperature (28 ± 3 ° C approximately) for 10 days. Daily check-ups were performed to detect, in a timely manner, the signs of the disease (necrotic lesions) (^{Benbow and Sugar 1999}).

Molecular identification of phytopathogenic fungi.

PCR (Polymerase Chain Reaction) was used to confirm the identification of Colletotrichum isolates. DNA extraction from pathogenic fungi was performed using the technique described by ^{Sambrook and Russel (2001)}.

The ITS1-5.8S-ITS2 region of the rDNA was amplified by primers ITS1 (5’- TCCGTAGGTGAACCCTGCGG-3’) and ITS4 (5’-TCCTCCGCTTATTGATATGC-3’) (^{White et al., 1990}).

The PCR amplifications were performed through a thermocycler (Applied Biosystem® GeneAmp® PCR System 9700, California, USA) with the following conditions: denaturation at 95 °C for 2 min, followed by 30 cycles of denaturation at 95 °C per 1 min, alignment at 50 °C for 30 s and primer extension at 72 °C for 2 min. With a final extension at 72 °C for 10 min. The amplification products were separated by 1% agarose gel electrophoresis (SIGMA-Aldrich®, Germany), which were stained with methyl bromide (0.2 μg mL-1). the DNA amplicons were visualized in a transilluminator (BioDoc -IT system image, UVP®, USA) (^{Ochoa et al., 2007}) and sequenced (Genewize Inc. New Jersey, USA). The DNA sequences were aligned in the NCBI online database (NCBI BLAST) (http://blast.ncbi.nlm.nhi.gov/Blast.cgi) (^{Altschul et al., 1997}).

Growth of the germinal tube and spore size of Colletotrichum spp.

To determine the percentage of germination and growth of the spore germ of Colletotrichum spp. the technique of slide culture was used (^{Cai et al., 2009}), for which cropping was taken in PDA medium of 10 mm2 and placed on a slide. The 30 μL of a spore suspension (1x106 spores μL-1) were then deposited on the edges of the cut and covered with coverslip plates (^{Johnston and Jones, 1997}). The cultures were incubated for 8 hours at room temperature (28 °C) and under high humidity conditions in plastic boxes with lid (^{Chaky et al., 2001}).

The length and width of 30 spores and the germinative tube of each insulation were recorded. Microscopic observations and measurements were made under an optical microscope at a magnification of 40x. The structures on each plate were photographed with the program (Motic Images Plus 2.0), the images were used to measure the length and width of the spores and the germinal tube of Colletotrichum spp. (^{Rodríguez et al., 2009}).

Pathogenicity evaluation was performed with 10 replicates per strain. Data obtained from the characterization of fungus strains (conidial size = long x width) and germ tube as well as pathogenicity were evaluated using a completely randomized design. The data were subjected to a variance analysis and multiple means comparison using the Tukey test at a significance level of p≤ 0.05, using the statistical package Statistic Analysis System (SAS, 2009) version 9.

Isolation of the pathogen and morphological characterization

A total of 86 fungal strains were isolated, among which 20 strains of Colletotrichum spp. using molecular techniques, which were identified as C. gloeosporioides (13 strains), C. hymenocallidis (1 strain), C. siamense (5 strains) and C. tropicale (1 strain) (Table 1). From these strains, C. gloeosporioides has been reported in different species such as papaya, citrus varieties, avocado, coffee, mango, grape, soursop, tomato and strawberry (^{Beltrán and García, 2006}). As well as C. tropicale, several crops have been reported, such as guanabana, mango and avocado among others (^{Rojas et al., 2010}).

However, the rest of the species have not been reported for avocado, as is the case of C. hymenocallidis which has been reported mainly in spider lily and some cases in chile and tomato (^{Yang et al., 2009}). Like C. siamense, the information is not available if it causes damage to the avocado, but if in the various crops such as the tree of bread, coffee tree, fig, cocoa, good herb, black pepper, rosemary (^{James et al., 2014}).

The strains were grouped into 8 categories according to their macroscopic morphological characteristics such as growth type, mycelia, color and conidial mass, which are presented in Table 1. All these isolates have characteristics similar to those obtained by ^{Rodríguez et al. (2009)} that together with ^{Pérez et al. (2003)} mention that the pathogen that causes antracosis called Colletotrichum spp. has great morphological variability due to the parameters such as incubation conditions, temperature and humidity, however, in the present work the environmental conditions or culture medium were not varied, so we can highlight the variations that may exist between species and even among strains of the same species, which according to ^{Gutiérrez et al. (2001)} may be due to the great genetic plasticity of adaptation to different conditions

Phytopathogenicity in avocado fruits

All isolates identified as Colletotrichum caused the symptoms of anthracnose in avocados inoculated from day six. The pathogenicity test of the different isolates Colletotrichum spp. showed that the 20 strains were able to generate, at 10 days after inoculation (ddi), the same signs of the disease that occurred in the lesions from which they were isolated (Figure 1), whereby, re-isolates from artificially infected avocados share the characteristics of the strains that were inoculated, that is, they comply with Koch’s postulates.

Figure 1 Damage caused by Colletotrichum spp. in the avocado ‘Hass’ during 10 days of incubation. 

In the in vivo tests, the ANq (C. siamense) and ANc (C. gloeosporioides ) strains had the greatest developmental diameter of the disease symptoms, followed by a group of C. gloeosporioides, C. siamense and lower virulence in avocado were found C. hymenocallidis and C. tropicale, however, all strains caused considerable damage to the fruit (Figure 2).

Figure 2 Strains of Colletotrichum spp. evaluated in the development of the symptoms of the disease in vivo. 

The 20 strains present high morphological, pathogenic and genetic variability, which according to ^{Morales et al. (2009)}, is related to the place where the pathogen was isolated. In the present study, the most pathogenic strains were isolated from orchards where trees were not pruned and had the humidity and temperature conditions for the development of pathogens. ^{Montero et al. (2010)} mentions that the variability of C. gloeosporioides is due to its biochemical aspects; while ^{Marroquin et al. (2016)} says that there is great pathogenic variability, which makes the differential reactions expressed in the interactions of Colletotrichum gloeosporioides are different.

^{Domínguez et al. (2012)} found that the different species of Colletotrichum spp. have great genetic and molecular variability, causing the degree of damage to be variable between them. For this reason the first signs of anthracnose in the in vivo evaluation began to appear between the third and fourth day of incubation and the damage was increased when the fruit began to mature, which may be due to latent disease, i.e. infection occurs in early stages of fruit development and even from flowering and remains latent until the fruit reaches optimal conditions such as the maximum production of ethylene, which triggers enzymatic events that stimulate the development of the fungus, as well as the decrease in the concentration of some compounds that inhibit its development (^{Alarcon and Chavarriaga, 2007}), as is the flavonoid epicatechin and other polyphenols, which maintain Colletotrichum spp. in latent state, however, during maturation of the fruit the levels of this compound and others that are antifungal decrease, which causes the activation of the fungus (^{Beno and Prusky 2000}et al., 2005; ^{Rodríguez et al., 2013}), which is consistent with what was observed in the present study.

Growth of the germinal tube and spore size of Colletotrichum spp.

According to studies by ^{Adaskaveg and Hartin (1997)}, morphology is a characteristic that can vary according to the isolate. In the present study, Colletotrichum strains were classified into four major groups with reference to germ tube development for eight hours (Figure 3). The first group consists of the NAa strain, which presented a growth of 102.9 µm; the second group by the strain NAc (71.6 µm), the third group by the strains NAd, NAb, NAf, NAj, NAm, Nag, NAt, NAh, Nai, NAn, NAr, NAq, NAk, and NAs (46.1-22.5 µm), while the fourth group is composed of strains NAl, NAe, NAp and Nao (20.5-9.2 µm). Growth of the highest germination tube at 8 hours after sowing was reported by Colletotrichum gloeosporioides.

Figure 3 Development of the germinal tube of 20 strains of Colletotrichum spp. isolated from the avocado crop at 8 h after inoculation. 

In relation to the size of the spore, it was found that there is variability both in the length and the width of the spore, with a length between 14.8 and 33.9 µm, whereas in the width one has a range that goes from 5.2 µm to 8.6 µm.

In the Table 2 presents the Pearson correlation coefficient between the morphological characteristics of the spores and the development of the pathogen, where it is shown that there is no statistically significant correlation between the spore size and the development of the fungi, either in vivo or in vitro (p≤ 0.05); however, germ tube growth has a statistically significant correlation with in vitro and in vivo pathogen development, with a Pearson r of 0.548 and 0.552 respectively (p≤ 0.05).

^{Rodríguez et al. (2013)} mention that the development of the appressorium is not necessary to be able to do damage to the host, but depends on the germination and the development of the germinative tube to be able to form the synthesis of a mucilaginous layer that adheres to the host; as well as ^{Cascino et al. (1990)} and ^{Kim et al. (1999)} mention that not all species of Colletotrichum form appressorium, but can penetrate directly with the help of the germinative tube.

The results in this study clearly show that there are at least four species of avocado pathogenic Colletotrichum; these species present morphological and pathogenic diversity even among strains of the same species.