Sampling and identification of the pathogen

Samples of stems from tomato plants were collected with symptoms typical of vascular withering disease caused by Fol in tomato-producing plots in six locations in the municipality of Metepec, Hidalgo, in commercial varieties: Reserva, Donnatello, El Cid, Palermo, Andino and Mezcal. The collected stems were deposited in polyethylene bags and subsequently transferred to the Agricultural Pathosystems Laboratory of the Department of Plant breeding at the Antonio Narro Autonomous Agrarian University (UAAAN).

Longitudinal cuts were made to the stem for the purpose of observing symptoms of internal brown necrosis in the conductive vessels and then cutting sections of the stem of approximately 3 mm. These stems were disinfected with 1% sodium hypochlorite solution for 3 min, then sown in a culture medium potato dextrose agar (PDA) in Petri plates and incubated at 25 °C for eight days, after which hypha tip transfer was performed to obtain monospiric crops (^{Amini, 2009}). The identification of the pathogen was made considering the symptomatology in diseased plants and by the morphological characteristics of mycelium and conidia in the growing medium at the microscopic level (^{Gerlanch and Nirenberg, 1982}; ^{Leslie and Summerell, 2006}).

Fol breed identification

10-day-old PDA cultures were washed with sterile distilled water to obtain inoculum suspension from the pathogen. The crops were then filtered, washed with sterile water and adjusted to a concentration of 1×10⁶ conidia per ml. The viability of conidia was verified by dilutions in plates in PDA media. For each insulation the concentration of spores was measured by counting macro conidia using a Neubauer hematocimeter under a 40X magnification microscope.

The pathogenicity of each isolate was tested on cv tomato seedlings. Bony Best, susceptible to breeds 1, 2 and 3 of Fol, following the methodology of immersion of root tips for 10 min in a fungal solution at a concentration of 1×10⁶ spores ml^-1 (^{Williams, 1981}; ^{Inami et al., 2012}). After inoculation the seedlings were transplanted into 10 cm diameter pots, filled with commercial peat moss substrate and kept in a greenhouse.

The identification of races of the isolates collected from Fol was carried out using the pathogenicity test in 4 differential tomato crops: Bonny Best, without resistance genes and susceptible to the three races, Manapal resistant only to race 1 by the presence of locus i, but susceptible to breeds 2 and 3, Walter resistant to breeds 1 and 2 due to the presence of loci i and i-2,but susceptible to race3 and I3R3 resistant to races 1, 2 and 3 due to the presence of the i-3 gene (^{Reis et al., 2004}; ^{Scott et al., 2004}). Differential materials were provided by Dr. John Paul Jones of the Institute of Sciences for Agriculture and Food at the University of Florida.

When the seedlings had three true leaves they were extracted from the substrate; its roots were washed with running water, rinsed with sterile distilled water and submerged for 10 min in a fungal solution at a concentration of 1×10⁶ spores ml^-1 (^{Inami et al., 2012}). The control plants were submerged in distilled water. All seedlings were transplanted into 10 cm diameter pots containing a sterile mixture of sand and soil, 15 g of NPK (15:15:15) were added to each pot. The pots were kept in a greenhouse at 23-28 °C, 60-70% relative humidity and 16 h of light, 8 h of darkness.

Evaluation of tomato genotype resistance for Fol breed 3

The incidence and severity of Fusarium oxysporum f. sp. lycopercisi breed 3was evaluated in seven tomato genotypes (K3, R1, F3, Y53, D4, D3, IR13), which are lines selected for their potential yield in previous work carried out in the Department of Phytomejoration of the Antonio Narro Autonomous Agrarian University (UAAAN). For inoculation, tomato seedlings were used 30 days after germination, using the root immersion technique in a suspension of 1x10⁶ conidia per ml. After inoculation, seedlings were transplanted into 3-litre polyethylene bags containing a mixture of soil and peat and they kept in a greenhouse for 48 days at a temperature of 25 ±2 °C.

The response of tomato plants to Fol inoculation was made using a severity scale of 1 to 5 according to ^{Marlatt et al. (1996)} modified to estimate the severity of the disease, where 1 corresponds to a symptom-free plant, 2 plant with mild chlorosis in the lower leaves, 3 moderate chlorosis, 4 severe chlorosis and 5 dead plant. With the severity scale values assigned to plants, a percentage disease index was estimated for each of the genotypes using the following formula: IE=∑i=1nXin0.2100 . Where: Xi= severity of the disease in the i-th seedling, n= number of seedlings evaluated. 0.2= correction factor for disease percentages.

The disease rates obtained were used to determine the progression of the disease and the response of these materials to the inoculation of the pathogen by calculating the area under the disease progress curve (ABCPE) according to the following equation: ABCPE=∑i=1n-1yi+1+yi2ti+1-ti Where: ‘t’ is the time of each reading ‘y’ is the percentage of plants affected at each reading and ‘n’ is the number of readings (^{Shaner and Finney, 1977}).

Statistical analysis

All pathogenicity tests were performed on a randomized complete block design, with three repetitions. The experimental unit consisted of 5 pots per genotype. With the values obtained for disease index and ABCPE a variance analysis was performed using the GLM procedure and comparison of means using the Tukey test (p≤ 0.05). For disease index, the variance test was performed with values obtained at 16, 32 and 48 days after inoculation.

Isolation and identification of the pathogen

Of thirty isolations obtained from plants with symptoms typical of the disease, six produced colonies in the PDA culture medium with aerial, dispersed and abundant mycelium with variation in white to pale violet, with abundant macroconidia of short and medium length, curved, thin-walled and septate. These characteristics coincide with that described by ^{Leslie and Summerell (2006)} for the morphological identification of Fusarium oxysporum, in addition to the symptomatology observed in artificially inoculated tomato plants allowed identification as Fusarium oxysporum f. sp. lycopersici.

Fol breed identification

Results on the reaction of differential cultivars for Fol were obtained 16 days after inoculation. Susceptible seedlings had typical symptoms of the disease (yellowing, defoliation, withering and death), while resistant seedlings showed no symptomatology to the strains used. Bonny Best and Manapal had complete susceptibility in the six strains evaluated (Table 1), thus ruling out the presence of race 1. Walter showed complete susceptibility to four of the six insulations and resistance in two, indicating that these two isolations can be identified as race 2, since the Walter variety is resistant to breeds 1 and 2 and susceptible to race 3.

Some plants of the I3R3 variety considered resistant to race 3 showed some signs of susceptibility probably due to a genetic impurity in the crop, due to the loss of minor genes during the introgression process by backcrossing, which are able to modulate the expression of the reaction of resistance to this pathogen (^{Reis et al., 2004}).

The presence of two breeds was detected in plots sampled in a ratio of 33% corresponding to race 2 and 67% to race 3. Therefore, it is necessary to use crops adapted to the region with resistance to strains 2 and 3. ^{Ortega (2010)}; ^{Sánchez et al. (2010)}; ^{Hernández et al. (2014)} also reported the presence of Fol strains 2 and 3 in the state of Morelos, Sinaloa and San Luis Potosí, respectively. ^{Valenzuela-Ureta et al. (1996)}; ^{Carrillo-Facio et al. (2003)}; ^{Ascencio-Álvarez et al. (2008)} report the presence of breeds 1, 2 and 3 in commercial lots of tomato in Culiacán Sinaloa; ^{Holguín-Peña (2005)} in Baja California Sur and ^{Ortega (2010)} in Morelos, it is likely that the origin of these breeds in Mexico was by introduction of contaminated seed.

With the recent increase in the use of commercial tomato seeds produced outside Mexico, there is greater potential to introduce and disseminate the pathogen in areas where it has not been previously reported. The appearance of new breeds may also be due to the selection and mutation of pre-established or isolated avirulent races (^{Shahi et al., 2016}). Fusarium oxysporum lacks sexual reproduction, therefore, genetic exchange is limited to the parasexual cycle and genetic transformation, which requires heterokaryosis, leading to hyphal fusion followed by cell lysis (^{Strom and Bushley, 2016}).

The plants have developed a complex defense system against various pathogens (^{Agrios, 2004}). Once pathogens overcome mechanical barriers to infection, plant receptors initiate signaling pathways that drive the expression of defense response genes that depend on their ability to recognize harmful molecules, perform signal transduction, and respond defensively through ways involving many genes and their products (^{Collinge et al., 1994}). Similarly, pathogens actively try to evade and interfere with response ways, selecting a decentralized and multi-component immune system (^{Andersen et al., 2018}).

Disease index and ABCPE in tomato genotypes inoculated with Fol

One way to measure Fol damage in tomato plantations is by quantifying the incidence, which measures the number of affected plants expressed in percentage, similarly used to determine resistance to this pathogen. The area under the disease progress curve is a quantitative summary of disease intensity over time, useful for comparing genotype response to pathogen inoculation. The results obtained in this study show significant differences (p≥ 0.05) for disease index and ABCPE in the genotypes evaluated (Table 2), the above highlights the difference in the genetic basis of the lines evaluated and the resistance or susceptibility to Fol.

The first symptoms of the disease (chlorosis) occurred 16 days after inoculation, with increased severity over time. Y53 was the genotype with the lowest percentage of incidence of the disease (60.6%) and ABCPE of 1 100, while IR13 and R1 had the highest incidence values for the disease

(76 and 75.3%, respectively), with ABCPE values of 1 700 and 1 450, respectively (Table 3). The lowest values of ABCPE correspond to materials with the lowest incidence of disease, that is. with a higher level of resistance (^{Bautista et al., 2009}).

In this regard, plants and pathogens have carried out complex attack and defense mechanisms. The plant defense system is the ability to perceive pathogens and activate effective defense responses (^{Grube et al., 2000}). Resistance in plants involves resistance proteins (R) that detect specific effecting proteins (Avr) produced by the pathogen. So far, some of the disease resistance genes to Fol have been used successfully in plant breeding. The use of tomato varieties with genetic resistance has represented a good sustainable production alternative replacing frequent pesticide applications. In general, most commercial varieties of tomato contain one or two of the i-1 and i-2 genes against Fol disease (^{El Mohtar et al., 2007}).

Plant disease management has always been one of the main goals in genetic improvement programs. However, host-pathogen interactions involving resistance are complex, plants develop resistance mechanisms, pathogens develop strategies to overcome plant resistance; plants, at the same time, develop new defensive measures that select additional changes in the pathogen (^{Stahl and Bishop, 2000}; ^{Gururani et al., 2012}). Plant pathogens have strategies to recognize the right host, penetrate and invade plant tissue, overcome plant defenses, and optimize their growth within plant.

To carry out these processes, generally, the fungus has to perceive the chemical and physical signals of the host and respond with the metabolic and morphogenetic changes required for pathogenic development (Pietro et al., 2001). Figure 1 shows the progress of the disease during plant development over time, it is seen that D3 and Y53 had the lowest incidence values of the disease in the first few days of evaluation, delaying the onset of the disease probably as a mechanism of defense of plants to attack the pathogen. This does not prevent the plants from being infected, but rather reduces the rate of disease increase in each of the infection points, thus delaying the spread of the disease and the development of plant disease epidemics in the field (^{Van der Plank, 1984}).

Figure 1 Response to Fusarium oxisporum f. sp. lycopersici inoculation in seven tomato genotypes. 

Pathogens actively try to evade and interfere with response ways, selecting a decentralized, multi-component immune system (^{Andersen et al., 2018}), which is finally expressed in a generalized infection.

In this study, disease incidence values of more than 50% were observed, in contrast to other studies, the incidence and severity was variable, because the environmental conditions, variety and virulence of the pathogen are different. In order to evaluate the resistance of tomato genotypes to Fol breed 3, ^{Báez-Valdez et al. (2010)}, carry out root immersion inoculations on four tomato rootstocks, finding severity values below 10%. ^{Mitov and Pérez (1973)} when determined resistance

to F. oxysporum in tomato varieties obtained values of 15% incidence in the evaluations carried out. For his part, ^{Mitidieri et al. (2005)} determined resistance to F. oxysporum in tomato rootstocks, reporting severity percentages of 13% after 30 days of evaluation.

The use of tomato varieties resistant to the breeds of Fol is an alternative for its control, therefore, it is essential to have germplasm with resistance to this disease. In this work, the genotypes evaluated showed differences in resistance to this pathogen, showing its broad genetic base. In results obtained by ^{Ascencio-Álvarez et al. (2008)}, based on the incidence of vascular wilt presented in 27 tomato accessions, it was found that four of them presented resistance to Fusarium oxysporum f. sp. lycopersici, being three of wild origin, two of S. pimpinellifolium, one of S. peruvianum and one of S. lycopersicum var. Motelle.

On the other hand, in a study conducted by ^{Dordevic et al. (2012)} about the reaction of different tomato crops to Fusarium oxysporum f. sp. lycopersici breed 1, it was found that of 24 tomato genotypes (between these pure and hybrid lines) 15 were not affected by this breed, four were tolerant and five susceptible.