Materials and methods

Antagonism of T. harzianum and T. koningiopsis against F. solani in dual cultures. The evaluation of the antagonism was carried out with strains T-H4 of T. harzianum and T-K11 of T. koningiopsis, both isolated from the root of Persea americana, the sequences of which are found in the National Center for Biological Information (NCBI) data base, with accession numbers MK779064.1 and MK791648.1, respectively. Likewise, F. solani strain MX-MIC 798, with accession number MK532475 was used, which was donated by the Plant Pathology Laboratory of the Agricultural and Forestry Research Institute of the Universidad Michoacana. All the biological material is stored in laboratory 204 of the Eco-Campus Valsequillo of the Institute of Science of the Benemérita Universidad Autónoma de Puebla (BUAP).

For the evaluation of mycelial growth, fragments of F. solani (MX-MIC 798), T. harzianum (T-H4) and T. koningiopsis (T-K11), 5 mm in diameter, were planted in Petri dishes with PDA (Potato Dextrose Agar), each in triplicate and incubated in the dark at 28 °C for 10 days. The mycelial diameter was measured every 12 h to estimate the growth speed (cm), which was calculated with the function of linear growth y=mx + b (where ‘y’ is distance, ‘x’ is time and ‘b’ is the constant factor) and it was expressed in centimeters per day (cm d^-1) (^{Zeravakis et al., 2001}). The diameter of the mycelial growth was determined with a digital caliper (CD-6 Mitutoyo) taking the same direction in triplicate.

The dual culture technique was used following ^{Andrade-Hoyos et al. (2019}) in triplicate to determine the percentage of inhibition of radial growth (PICR) for each trial evaluated, for 10 days using the formula PICR= [(R1-R2/R1) x 100], where R1 is the diameter of the control (average of the radial growth of the triplicates in each strain) and R2 is the diameter of the tested organism. To complement the evidence of the antagonism, each trial was compared and classified using the scale established by ^{Bell et al. (1982}) (Table 1).

Nursery tests to analyze the percentage of germination and rotting of seedlings. Miahuateco chili seeds were used, which were provided by farmers from Santa María Tecomavaca in the state of Oaxaca. The trial was held under controlled conditions in the nursery of the Eco-Campus Valsequillo, ICUAP-BUAP. Before the seeds germinated, they were disinfested with sodium hypochlorite at 0.3% (v/v) for 10 min, rinsed three times with sterile water and dried with sterile paper (^{Illa et al., 2019}). The pathogen F. solani was inoculated in a salt suspension at a concentration of 1x10⁸ conidia mL^-1 with a Neubauer chamber (PAUL MARIENFELD), in 400 seeds for each treatment under study; once the pathogen was soaked and the seeds were dry, T. harzianum and T. koningiopsis were inoculated. This was carried out with a suspension of teliospores at a concentration of 1x10⁸ conidia mL^-1 for two separate treatments, each one with 400 seeds (^{Reyes-Tena et al., 2019}).

Once the seeds were inoculated with the fungi, the seeds were placed in germinating trays with Peatmoss and Agrolita (1:1 v/v) after sterilizing it at 121 °C and 15 lb of pressure (^{Mannai et al., 2018}). The seeds were planted using a standard mechanical procedure; the humid substrate was placed to take up three quarters of the total cavity, and one seed was placed in each cavity at a depth of 1 cm (^{Castillo-Aguilar et al., 2015}). Mancozeb 80^® was used as a chemical treatment, following the manufacturer’s recommendations for the production of the seedling (1 kg of seed for every 5 g of product), along with the use of a control treatment, which consisted of seeds that were disinfested in a solution of sterile distilled water and sucrose at 3%, without fungal activity. Finally, the trays were wrapped with black plastic for 10 days and the temperature was kept between 10 and 27 °C during germination (^{Castillo-Aguilar et al., 2015}). The percentages of seed germination were calculated using the following formula: seed germination (%) = (number of germinated seeds/total number of seeds) × 100 (^{Al-Fadhal et al., 2019}). The data were taken when the control treatment presented a germination rate of ≥ 90%.

Mortality and rate of severity in plants 40 dat. We used 1,500 Miahuateco chili seedlings 15 days after emergence, provided by local farmers. The trial was conducted under open field conditions in the town of Santa María Tecomavaca, between parallels 17° 51’ and 18° 03’ latitude N, and meridians 96° 57’ and 97° 15’ longitude W, at an altitude between 400 and 2600 masl. The predominant climates in the area are dry warm (40.5%) and semidry warm (17.5%) (^{INEGI, 2017}). The ground was prepared by local farmers three months before planting. The field was plowed to a depth of 50 cm in order to reduce compaction and favor draining, along with two harrows-turns to a depth of 30 cm to ventilate the soil. The transplanting date was March 27, 2018, at a depth of de 3 cm. Fertilization was carried out with 90-80-80 of N, P and K after 15 days (dat). The experimental design was random block with three repetitions and the population density was six plants per m². During the development of the crop, the field was irrigated every five to six days, before the soil humidity reached 50% of the capacity of the field (^{Vázquez-Casarrubias et al., 2011}).

F. solani (MX-MIC 798) was inoculated at the neck of each chili plant (100 seedlings per treatment distributed into five random blocks) with 1 mL of solution at a concentration of 1x10⁸ conidia. After 36 h, the plants were inoculated with strains of Trichoderma (T-H4 and T-K11) at the same concentration as the pathogen (1x10⁸ conidia mL^-1), for each strain, applied to the root for all treatments. For the chemical treatment, Mancozeb 80^® was applied on the third day, following the recommendations provided by the manufacturer (480 g 200 L^-1 of water per ha). The control treatment was only given sterile water with no fungal activity. The plants were kept on the field for 40 days. The severity of the disease was calculated in the infected section in relation with the total root length (^{Perveen et al., 2010}) and classified on the scale of 0 to 3, where 0 = 0 to 25% of severity, 1 = 26 to 50%, 2 = 51 to 75% and 3 = 76 to 100% (^{Bokhari and Perveen, 2013}).

Finally, after week 6 in the field, reisolation was carried out. Small symptomatic root sections were taken (approx. 1 × 1 cm), sterilized with sodium hypochlorite at 1% for 30 s and rinsed three times with sterile distilled water. The segments were then placed in a PDA medium and incubated in the dark at 28 °C for 10 days. The cultures that developed were purified with monosporic or hyphae tips and the structures were identified using taxonomic keys by ^{Barnett and Hunter (1998}) in a microculture system using an optical microscope (Carl Zeiss, Jena, Germany) at a magnification of 1000x (^{Samson et al., 2014}), and were stored in a 20% glycerol solution at -84 °C (^{Morales et al., 2020}).

Statistical analysis. The data were analyzed with an ANOVA (two-ways) in the IBM SPSS Statistics statistical package, version 25. Growth speed and development rate were the response variables with three repetitions and they were validated in duplicate in a totally random statistical design. Means were compared with the method by Tukey-Kramer with a level of probability of p≤0.05. PICR, the percentage of germination, mortality (%) and the severity of the disease were expressed as percentages and transformed with an angular arccosine √x+1. They then underwent an analysis of variance (two-way), along with the variables of total height, root height and rootless plant; the Tukey-Kramer test was used to determine the significant differences between treatments with a significance of p≤0.05. In turn, the PICR was compared based on Student’s t test for paired samples (p≤0.05) using the same statistical package.

Results and discussion

Evaluation of the antagonism of T. harzianum and T. koningiopsis against F. solani in dual cultures. Areas of interaction appeared between T. harzianum (TH-4) and T. koningiopsis (T-K11) against F. solani (MX-MIC 798), where parasitism was observed after 132 h. The rate of development and speed of growth presented significant differences (p≤0.05), where T. harzianum obtained the highest value (Table 2) with 0.972±0.008 mm h^-1 and 1.3238±0.0024 cm d^-1, respectively. ^{Morales et al. (2020}) reported a higher speed of growth than in the present study for strain TH-4 with 1.86± 0.22 cm d^-1 and 1.67 ± 0.01 mm h^-1 for the development rate. F. solani presented the lowest speed of growth (0.4718±0.00063 cm d^-1). A study carried out by ^{Andrade-Hoyos et al. (2019}) reaffirm the results obtained in the present investigation, in which different species of Trichoderma spp. have superiorities over the mycelial growth of F. oxysporum (0.22 mm). The reduction in the growth rate of dual cultures is an indicator of the antagonistic ability of Trichoderma (^{Guigón-López et al., 2010}). The percentage of inhibition of radial growth (PICR) of F. solani exerted by T. koningiopsis and T. harzianum presents significant differences (p<0.0001) from day eight (Table 2), when the strain T-H4 was 1.33% higher than T. koningiopsis T-K11 and a class II classification (Figure 1 A and B), according to the scale established by ^{Bell et al. (1982}).

The present investigation coincides with ^{Sundaramoorthy and Balabaskar (2013}), who found an inhibition of 53% in the mycelial growth of F. oxysporum f. sp. lycopersici. Likewise, ^{Jagraj et al. (2018}) reported a PICR of 55.6% for F. oxysporum against T. koningiopsis, for strains isolated from tomato plants. ^{Michel-Aceves et al. (2019}) obtained a class II antagonism on F. oxysporum with T. harzianum, which were similar results to those obtained in this investigation.

^{Suárez et al. (2008}) presented 12 isolations of T. harzianum confronted with F. solani, where they obtained a PICR between 60 and 70% of the antagonism. ^{Nagendran et al. (2016)} reported that 15 isolations of Trichoderma spp. reached an inhibition of over 50% on the mycelial growth of F. solani, in which strain Phyto-4 obtained the highest inhibition (66.1%). Likewise, ^{Sánchez-García et al. (2017}) reported a PICR of 59.1% for F. solani against T. harzianum (Tri-4), which are higher results than those obtained in the present investigation. This indicates that T. harzianum may have a higher rate of incorporation of nutrients and a higher metabolism than F. solani, as well as different mechanisms for the secretion of hydrolytic enzymes, including cellulases, chitinases, glucanases, xylases and proteases, which may be involved in biocontrol mechanisms, helping it make better use of nutrients from the medium and deprive the pathogen of the resources (^{Michel-Aceves et al., 2005}). T. harzianum is also capable of inhibiting the growth of phytopathogens, since it produces numerous antibiotics, such as trichodermine, suzukacillin, alameticin, dermadin, penicillin, trichothecenes, trichorzianines and others (^{Andrade-Hoyos et al., 2020}).

Figure 1 Antagonism of Trichoderma with F. solani on the scale by^{Bell et al. (1982}) after 132 h in dishes with PDA medium, incubated at 28 °C for 10 days, (A) class II antagonism of T. harzianum (T-H4) and F. solani (MX-MIC 798), (B) class III antagonism of T. koningiopsis (T-K11) and F. solani (C) F. solani without interaction. 

Mycoparasitism is another action mechanism displayed by the genus Trichoderma. In this investigation, we found granulation from the strain MX-MIC 798 induced by T. harzianum, as well as the breaking of septa and strangulation (Figure 2). This highlights the eminent mycoparasitic ability of strain TH-4. ^{Duarte-Leal et al. (2017}) found mycoparasitism between hyphae in T. asperellum (Ta.25), in which they observed lysis in hyphae of F. oxysporum f. sp. ciceri (F-50), as reported in the present investigation. Other studies, carried out by ^{Hyder et al. (2017}), refers to the enzyme degradation of the cell wall of phytopathogenic fungi during the mycoparasitic action, which causes lysis and thus the disorganization of the cytoplasmic content. For T. koningiopsis, this investigation was unable to observe the mycoparasitic mechanism or breaking of the septae.

The growth rate and antagonism tests are tools to predict and determine the biocontrolling ability. According to ^{Morales et al. (2020}), these studies help select efficient antagonistic strains for evaluation under field conditions and obtain the best results.

Percentage of germination and rot caused by F. solani in Miahuateco chili seedlings in the nursery. The germination of the seeds from the control treatment was displayed after 12 days with 91% and significant differences (p=0.0023) in regard to the remaining treatments (Table 3), in which the effect of the inoculation from F. solani (MX-MIC 798), which affected the neck of the root, causing rot and death of the plant, as well as the lowest percentage of germination (30%). In this investigation, the Miahuateco chili seeds inoculated with strain MX-MIC 798 treated with T. harzianum T-H4 showed a germination rate of 82%, even higher than the value reported by ^{Ezziyyani et al. (2004}), who obtained a percentage of germination of 60% in bell pepper seeds treated with T. harzianum after 10 days. In untreated plants inoculated with F. solani, root rot reached level 4.0 and a mortality of 70% (Table 3) in the scale proposed by ^{Ezziyyani et al. (2004)}. ^{Madhavi et al. (2006}) reported a percentage of germination of 90.3% in chili seedlings (C. annuum) inoculated with T. harzianum-m1 and F. solani under nursery conditions, similar results to thos found in the present investigation.

Figure 2 Mycoparasitism of T. harzianum (T-H4) on F. solani (MX-MIC 798) in PDA medium, incubated at 28 °C for 10 days; A) interaction of T. harzianum conidia vs. F. solani septated macroconidia; B) granulation and breaking of septa of the strain MX-MIC 798 induced by T. harzianum; C) strangling of F. solani hyphae by strain TH-4. 

The microbial antagonists reduced rot caused by F. solani under nursery conditions in Miahuateco chili seeds significantly (22%), whereas T. harzianum presented the lowest percentage of plant mortality (18%), higher than strain T-K11, which obtained 26%. The group treated with Mancozeb 80^® reduced seed mortality by 28% and presented no significant differences with T. koningiopsis; the control treatment accounted for a feasibility of 91% in seed germination (Table 3).

The Mancozeb 80® chemical treatment presented high mortality results, which could indicate the creation of resistance by F. solani to this fungicide. Some fungicides used, such as triazoles, tend to develop resistance to the fungal populations due to genic factors (accumulative with time), since they only have one place of action (^{Leroux and Walker, 2011}; ^{Ziogas and Malandrakis, 2015}). ^{Chen et al. (2014}) reported that Fusarium fujikuroiuses a resistance mechanism generated by the mutation of β2tub by the fungicide carbendazim, making its management difficult, although it opens the possibility of generating new control measures with antagonistic agents.

Mortality and rate of severity in plants 40 dat. The Miahuateco variety chili seedlings that were inoculated with F. solani at the moment of transplanting displayed rot in the roots 40 days after inoculation. Long reddish lesions were found, which covered 10% of the total surface of the main root and reddish-brown spots were also observed in the section of the stem found under the soil surface, along with a considerable loss of lateral roots. This coincided with descriptions by ^{Agrios (2006}).

The field results showed that the treatments with antifungal activity were effective to reduce mortality by wilting and root rot in Miahuateco chili plants under induced infection by 48 and 51% respectively, and it reduced the severity of the disease 40 dat by 50% (Table 4). Strain T-H4 presented the lowest percentage of plant mortality (48%), yet it displayed no significant differences with the Mancozeb 80^® fungicide, which accounted for 47% of the mortality. Similarly, ^{Thangavelu et al. (2004}) showed that T. harzianum (Th-10) efficiently controlled the fusarial wilting of the banana tree, comparable to chemical control. ^{Nagendran (2016)}, found the highest percentage (36.9%) of the reduction of the disease with strain Phyto-6, and also displayed a lower severity in witling, as shown by the present investigation. T. koningiopsis T-K11 obtained a mortality of 51%, presenting significant differences with the control group (23%), which obtained the lowest mortality.

During their development, the plants infected by F. solani presented symptoms of yellowing in the basal part, yellowing of the nervations in 30% of leaves, as described by ^{Black et al. (1993}) in chili plants (C. annuum) that survived the infection on the field. The analysis of variance for the total plant height displayed highly significant differences (p= 0.0002) between treatments. Mancozeb 80® obtained the greatest height (25.6±2.78 cm), followed by plants inoculated with T. harzianum (24±1.98 cm) and T. koningiopsis (21±1.81 cm). The lowest height appeared recurrently in plants inoculated with F. solani with 16±0.28 cm, as shown in Figure 3A. ^{Agrios (2006}) mentions that plants affected by Fusarium sp. reduce their growth, due to the ability of the pathogen to colonize roots, preventing the adequate nutrition of the plant.

The genus Trichoderma can parasite pathogenic fungi and produce antibiotics. It also presents positive effects in plant growth, yield, absorption of nutrients, fertilizer use efficiency and the systemic resistance of plants to diseases (^{Hyder et al., 2017}; ^{Andrade-Hoyos et al., 2020}). In a study carried out by ^{Cole and Zvenyika (1988}), they noticed that the infection caused by F. solani in tobacco plants could be controlled when applying T. harzianum in disease management procedures. Likewise in a study by ^{Yedidia et al. (1999}), when inoculating T. harzianum in early growth, where the absorption of nutrients in the plant improved. These investigations agree with studies by ^{Tsegaye-Redda et al. (2018}), who mention that Trichoderma sp. has the ability to antagonize the phytopathogens present in the soil, and also induces the promotion of plant growth and protects plants from biotic and abiotic stress. The studies by ^{Dubey et al. (2007}) have been confirmed in T. harzianum, where seed germination, root health and sprout length have improved and the frequency of healthy plants has increased, as confirmed by the present investigation.

Forty dat with the strain MX-MIC 798 that led to the development of necrotic lesions in the roots and stems of evaluated plants, we proceeded to re-isolate in order to satisfy Koch’s postulates. The fungal cultures developed abundant aerial mycelia, with a cotton-like texture and white in color with purple tones (Figure 3B), which appeared more intense in the middle of the culture medium (^{Rentería-Martínez et al., 2018}). The duration of the microculture until chlamidospores were visible was 14 days; for macroconidia, seven days; and for phialides and microconidia, four days. Oval-shaped septated mycelia and abundant microconidia were observed, with an average size of 2.4 x 6.3 μm in length by 2 x 4 μm in width (Figure 3C), along with the typically crescent-moon shaped macroconidia, one to four septa (Figure 3F), an average size of 2.4 x 4.1 μm in width and 11.8 x 36.4 μm long (^{Barnett and Hunter, 1998}). Likewise, chlamidospores were also observed (Figure 3E), 8.4-10.5 μm in diameter and with particular morphometric characteristics reported by ^{Montealegre et al. (2003}).

Figure 3 Reproduction structures of F. solani (MX-MIC 798) after 10 days in a PDA medium at 28 °C. A) dead Capsicum annuum var Miahuateco plant inoculated with F. solani MX-MIC 798, B) F. solani culture in PDA medium; C) microconidia of F. solani; D) phialides and microconidia in false F. solani heads; E) F. solani chlamydospores; F) F. solani micro and macroconidia. 

Conclusions

The strains T. harzianum (TH-4) and T. koningiopsis (T-K11) displayed classes II and III of antagonism, respectively, on F. solani on the Bell scale. In additionT. harzianum TH-4 displayed the highest rate of development, speed of growth and percentage of inhibition of PICR radial growth on F. solani MX-MIC 798 (53%) under in vitro conditions.

In nursery conditions, T. harzianum TH-4 (82%) and T. koningiopsis T-K11 (74%) presented the best response in the germination of Miahuateco variety chili plants, with the highest percentage of germination in comparison with the chemical treatment (72%), although T. harzianum TH-4 reduced rot caused by the F. solani strain by 75%.

In the town of Santa María Tecomavaca, in the state of Oaxaca, Mexico, T. harzianum (T-H4) and the Mancozeb 80® chemical treatment presented the best response in the reduction of the mortality and severity of the disease in Miahuateco chili plants 40 dat.