Materials and methods

Collection of samples and isolations of Botrytis sp. In the months of November and December of 2016, and based on a randomized collection of samples (^{Steel et al., 1997}), blackberry fruits with symptoms of gray mold were gathered from commercial plantations in 17 municipalities distributed throughout five states in Mexico (Table 1). Each fruit was placed in an individual polyethylene bag, which were then sealed and transported in a cold ice cooler to the lab, where they were stored at 4 °C (^{Li et al., 2012a}). Later, the berries were placed in polyethylene bags containing moist sterilized paper towels at a temperature of 20 °C ± 1 and a relative humidity of nearly 100% (humid chamber technique) to stimulate its sporulation. Once sporulated, the conidia were scraped without touching the fruit and using a sterilized dissecting needle, they were suspended in 1 mL of sterilized distilled water and adjusted to a conidial concentration of 1 × 10⁶ spores mL^-1 (^{Li et al., 2012a}). Next, 200 µL of the suspensions were planted in a PDA (Potato Dextrose Agar) culture medium (Bioxon, PDA, 39 g L^-1 water) modified with lactic acid (0.1% v / v) in Petri dishes, which were sealed with Parafilm (Sigma-Aldrich) and incubated at 20 ± 1 °C in the dark for 36 h. The germinated conidia were purified using the monosporic conidia method in Petri dishes with a PDA culture medium (^{Li et al., 2012a}).

Pathogenicity tests. The 211 Botrytis sp. monosporic cultures were inoculated in Tupi variety blackberry fruits in a stage of physiological maturity, which were washed with tap water and disinfested in sodium hypochlorite at 2% for 3 min, then rinsed twice with sterilized distilled water and the excess moisture was eliminated. The inoculation procedure consisted in creating a lesion (using a sterilized dissection needle at a depth of 3 mm) and without a needle placing a 20 µL aliquot of a conidial suspension at a concentration of 1 × 10⁵ spores mL^-1 on the fruits; for each isolation, two control fruits were used, on which 20 µL of sterilized distilled water were placed (^{Saito et al., 2016}; ^{Zhou et al., 2014}). The inoculated fruits and the controls were placed in 15 × 15 cm plastic containers containing dampened sterilized paper towels, sealed to increase relative humidity and incubated at 20 ± 1 °C in the dark under a totally randomized experimental design, inoculating one fruit per isolation with four repetitions with their respective controls, where each monosporic isolation was considered as a treatment. The diameter of the lesion was measured every 24 h, ending at 72 h after inoculation (hai). Afterwards, reisolations were carried out in PDA culture to verify the pathogenicity of the inoculated isolations and complete Koch’s postulates; the experiment was carried out twice. An analysis of variance was carried out, as well as a comparison of averages (Least Significant Difference) using the program ^{SAS V.9.1 for Windows}.

Morphometric, morphological and cultural characterization. The variables determined to the isolations were growth rate - calculated with growth at 48 h minus the growth of 24 h (^{Zhou et al., 2014}), growth shape and color of mycelium, days to sporulation, days to formation of sclerotia, number of sclerotia, shape and color of sclerotia (Martínez et al., 2003, ^{Tanovic et al., 2009}, Tanovic et al., 2014). The sclerotia, conidia and conidiophores were measured (n=50) (^{Li et al., 2012}) using an OLYMPUS BX 41 compound microscope with an OLIMPUS U-CMAD3 T2, U-TV1X-2 T2 camera (Tokyo, Japan) and the morphometry was carried out using the program ImageJ (^{Schindelin et al., 2015}). The information obtained from the pathogenic, morphological and morphometric characterization was used to carry out a cluster by similarity or by morphotype for the identification of groups and selection of the representative isolations to carry out the molecular characterization.

Molecular identification. DNA extraction was carried out using the Alkaline phosphatase (AP) method (^{Ruiz et al., 2014}, ^{Sambrook and Russel, 2001}). The regions of the internal transcribed spacer (ITS) were amplified using primers ITS1 / ITS4 (^{Gardes and Bruns 1993}; ^{Staats et al., 2005}; ^{White et al., 1990}) in combination with the thermocycling program proposed by White et al. (1990). TE amplified products were sequenced in both directions using the Sanger method in the company Macrogen (http://dna.macrogen, Korea). The resulting sequences were analyzed using ^{DNASTAR (2001)} and ^{Sequencher (2014)}, and the alignment was carried out with Clustal W in MEGA 6.0 (^{Tamura et al., 2013}). The sequences were compared using the BLAST algorithm of the NCBI (http://blast.ncbi.nlm. nih.gov/Blast.cgi) and deposited in the GenBank.

Results

In blackberry fruits, symptoms of soft rotting were observed, colored maroon to black, over which abundant mycelia, conidiophores and conidia grew, altogether forming gray mold (Figure 1A and 1B). Two hundred eleven isolations of Botrytis sp. were obtained, isolated from blackberry fruits from different regions in Mexico. The pathogenicity tests confirmed that 100% of the isolations inoculated produced symptoms of gray mold 48 hours after inoculation (hai) on inoculated fruits, with or without lesions, and control fruits displayed no symptoms or signs of the fungus. When performing the morphometric, morphological and cultural characterization, the 211 isolations were identified as Botrytis cinerea (^{Ellis, 1971}; ^{Jarvis, 1977}; ^{Martínez et al., 2003}). The production areas studied were: Colima, State of Mexico, Jalisco, Michoacán and Morelos, from where 34, 25, 32, 80 y 40 isolations were taken, respectively (Table 1). Clustering by similarity helped identify 21 groups, from which a representative isolation of each was taken; the pathogenic, morphological and molecular description is described below.

Pathogenicity tests. Out of the 21 isolations inoculated, 100% was pathogenic in both inoculation methods. When no lesions were created, the inoculated fruits presented typical symptoms of the disease 48 hai (Figure 1C); significant differences were found (F = 2762.38, p = ˂ 0.0001); the diameter of the lesion varied between 4.2 and 15.9 mm., and at 72 hai the diameter of the lesion varied between 8.6 and 23.1 mm (F = 8289.98, p = ˂ 0.0001). In inoculated fruits with lesions, the typical symptoms manifested 48 hai, with significant differences (F = 3551.21, p = ˂ 0.0001) in the diameter of the lesion, which ranged between 7.3 and 19.6 mm. At 72 hai, the behavior of the isolations stabilized; the diameter of the lesion varied between 11.2 and 24.8 mm (F = 4287.86, p = ˂ 0.0001). In all cases, isolation MF12 was the most aggressive, since it produced the largest lesion diameter in comparison with isolation OS10, which presented the smallest lesion diameters (Table 2). Control fruits presented no symptoms or signs of the fungus (Figure 1C).

Morphometric, morphological and cultural characterization. The values of the mycelial growth rate varied between 0.7 cm d^-1 and 1.5 cm d^-1 where isolations 1.6 and 4.4 grew at a slower rate, while isolation MF21 was the quickest to grow (Table 3). The way of growth varied - 80.95% of isolations formed superficial mycelia and 19.05% formed aerial mycelia - as did the color, where 42.86% displayed a light gray color, 38.09% presented a gray color and 19.05%, dark gray, (Figure 1 D-G).

All isolations (100%) formed sclerotia, although isolations MF21 and 4.8 were more precocious, since they formed them 4 days after planting (dap), whereas isolation 4.1 took 11 dah. The shape of the sclerotia varied: 66.67% were irregular and 33.33% were round (Figure 1 H-J). The number of sclerotia formed in the Petri dish was 26 (isolation ZF10) to 333 (isolation MF21) (Table 3) and the largest sclerotia were formed by isolation 1.6, sized 2.5-(3.1)-4 × 2.0-(2.5)- 3.5 mm and isolation HC19 formed the smallest, sized 0.9-(1.2)-1.3 × 0.6-(0.9)-1.1 mm (Table 3).

Figure 1 Morphological characteristics of B. cinerea A. Blackberry fruit with symptoms in the field. B. Blackberry fruit with signs in the field. C. Top section, blackberry fruits inoculated with sterilized distilled water, bottom section diseased fruits inoculated with B. cinerea. D, E, F and G. Different growth forms and color of mycelium. H, I and J. Shape, color, number and disposition in the PDA growth medium of sclerotia. K. Conidiophore with conidia. L. Conidia. 

All isolations sporulated, and isolations MF21 and HC19 were the most aggressive, sporulating 9 dap, in comparison with isolation 1.5, which took place 20 dap. Conidia observed were erect, septated, branched maroon to olive in color, and the largest ones were formed by isolation HC19 with a size of 1175-1694 × 8-17 µm, whereas the shortest ones were formed by isolation 2.5 and measured 846-1492 × 6-15 µm (Figure 1K). Out of all the isolations, 61.90% formed elliptical conidia, 19.05 % lemon-shaped and 19.05 % round ones; 100% were hyaline to light maroon (Figure 1L), the largest conidia were formed by isolation 5.6 and measured 8.8-(10.4)-13.8 × 6.5-(7.2)- 7.9 µm, while the smaller conidia were formed by isolation OS10 and measured 5.6-(8.7)-10.30 × 5.2-(6.5)- 8.0 µm (Table 3).

Molecular identification. The access numbers of the ITS sequences obtained are shown in Table 1. Access numbers from isolations 2.5, 3.1, ZF10, 4.1, 4.8, HC19, 4.4, 5.6, 1.5, MF21 and MF12, presented a similarity of 100% with access number MG907605.1 of B. cinerea; isolations HAR3, HAR4, BP3, BP5 were similar (100%) to the deposit KX783612.1 of B. cinerea and isolations 1.6, T1 and HC25 presented a similarity of 100% with access number KX463512.1 of B. cinerea. Isolations OS5 and OS10 were similar to the same deposit, but by 99 % and isolation T6 was similar (97%) to access number KU992695.1 of B. cinerea.

Discussion

The symptoms observed in this research were typical of gray mold, and similar to those reported by ^{Li et al. (2012a}) and ^{Li et al., (2012b)}. In the pathogenicity tests of this study, 100% of the inoculated isolations were ineffective, showing symptoms at 48 hai. Isolation MF21 developed the largest lesion diameter at 48 and 72 hai in both forms of inoculation, reaching the largest lesion diameter when the cut was made. The cuts in the blackberry fruit facilitate the entry of B. cinerea, since the fungus penetrates and uses mechanisms such as the formation of appressoria, the production of phytotoxins, and the secretion of enzymes that degrade the cell wall (^{Choquer et al., 2007}; ^{Zhang et. al., 2016}). According to the results, the presence of a lesion induces a greater diameter of the lesion, which highlights the importance of avoiding physical damage during harvest and postharvest to increase the fruit’s shelf life.

The isolations MF21 and ZF10 presented a higher growth rate in PDA culture medium (Table 3). This was related with aggressiveness when they were inoculated in the host, since they caused large lesions in little time. These isolations come from fields in the state of Michoacán, specifically from the area of Peribán and Ziracuaretiro, respectively, which have a background of an intensive use of products, in comparison to isolation 1.6 from Texcoco, State of Mexico, which displayed a lower growth rate and aggressiveness, and which has received no fungicides. The mycelial growth data found in this investigation coincide with those reported by ^{Li et al. (2012a}).

The present study found different growth forms, as well as the color of the B. cinerea mycelium, which coincides with different investigations (^{Li et al., 2012a}; ^{Lorenzini and Zapparoli, 2014}; ^{Ozer and Bayraktar, 2014}; ^{Tanovic et al. 2014}; ^{Zhou et al., 2014};). The results of the present investigation expose the high morphological variability of the pathogen to develop in a culture medium, but also share similarities between the members of a state in particular. For example, the 5 isolations from the state of Morelos displayed light gray mycelia and those from Jalisco, gray ones. In addition, the most aggressive isolation was observed to form darker colored mycelia. This correlation between pigmentation and aggressiveness may possibly be due to the higher presence of melanin in B. cinerea conidia (^{Doss et al., 2003}).

The size, shape, color and the arrangement of the sclerotia in the culture medium is important for the morphological identification of B. cinerea (^{Martínez et al., 2003}). The isolations obtained produce sclerotia that coincide in size with those reported by ^{Erper et al. (2015}) for B. cinerea in beans and in cranberries and grapevines (^{Saito et al., 2016}). The round and irregular shape, the black color and the arrangement in the culture medium of the present investigation coincides with B. cinerea isolated from blackberries (^{Li et al., 2012a}). The time taken for the fungus to form its sclerotia in vitro, as well as their number per Petri dish, gives us an idea of the survival of the isolations on the field. Regarding the number of sclerotia per Petri dish, isolation MF21 formed a large amount in a short time span, in comparison with the other isolations; it produced the greatest lesion diameter and growth rate in relation with the others.

The sporulation of B. cinerea in vitro is broadly related with its aggressiveness, because the faster it forms its reproductive structures, it can be disseminated and cause epidemics (^{Carisse et al., 2015}). The shape and color of the conidia found in the present investigation are similar to those reported by ^{Aktaruzzaman et al. (2017}); ^{Zhou et al. (2014}); ^{Lorenzini and Zapparoli (2014}) and ^{Xie et al. (2016}). The size of the conidiophores differs with that reported for blackberry by ^{Li et al. (2012a}); despite being the same species and host, they reported larger conidiophores. The size of the conidia is the most important morphometric variable to set these species apart, which helped us identify B. cinerea isolated from blackberries. Findings in this investigation coincide with reports for B. cinerea, although in other hosts by ^{Ozer and Bayraktar (2014}), ^{Rupp et al. (2017}). However, they differ with those reported by Zhou et al. (2014) and ^{Saito et al. (2016}), since these authors reported larger-sized conidia.There is a similarity in biological characteristics between the individuals obtained in a particular state, which is to say that the isolations with nearby geographic origins are similar. For example, the ability to induce diseases, aggressiveness, the color of the mycelium, the shape and morphometry of the sclerotia, conidia and conidiophores, amongst others. Each state shares a geographic origin and weather, which is similar in every sampling site by state from which the isolations were obtained (Table 1), which leads to the existence of biological similarities. However, there is a morphological, morphometric and pathogenic variability by states, but these are within the range reported for B. cinerea. For instance, the isolations from Michoacán differ in the biological characteristics with isolations from the rest of the states.

B. cinerea has been reported to cause gray mold in blackberries in Australia, China, New Zealand, South Africa, Norway and the United States. According to records from the USDA (^{Farr and Rossman, 2011}) and in the U.S., B. cinerea was reported only in species of Rubus in Alaska, California, North Carolina and Washington (^{Li et al., 2012a}), and in the present research, isolations were identified as B. cinerea causing gray mold in differents states in Mexico.

Conclusions

All isolations were pathogenic; variability was identified in the pathogenic, morphological and cultural characteristics, which are within the range of reports for this species. The isolations of the fungus obtained from the blackberry fruit were identified as Botrytis cinerea, taking the morphological, morphometric and cultural characteristics into consideration, confirmed with the analysis of sequences in the area of the internal transcribed spacers (ITS), and therefore the pathogen is reported as the causal agent of gray mold in the states of Colima, State of Mexico, Jalisco, Michoacán and Morelos.