Reagents, microbial strains, and fungicides

All used reagents (analytical grade) were purchased from Sigma Aldrich Corp. (St. Louis, MO, USA). Trichoderma isolates were obtained from soil near the root zone of healthy soursop trees in Compostela and San Blas, Nayarit, Mexico. Trichoderma sp., T. longibrachiatum, T. asperellum, and T. harzianum, provided by the Centro de Investigación en Alimentación y Desarrollo, A.C. (CIAD, A.C.), Unidad Cuauhtémoc, Chihuahua, were also included. The chemical fungicides evaluated were obtained from local commercial stores.

Strains of Lasiodiplodia theobromae and Fusarium falciforme were isolated and morphologically and molecularly characterized, from root tissues of diseased soursop trees from San Blas and Compostela, Nayarit, Mexico.

Trichoderma isolation and identification

Antagonistic microorganisms were isolated from soil near the rhizosphere of healthy-appearing soursop trees. In July 2020, in six commercial soursop orchards (Table 1), soil samples (500 g per tree) were collected from five trees located at the cardinal points and the center, subsequently, homogenized, and finally 200 g per orchard were recovered, and placed in transparent plastic bags (25 × 35 cm) for processing at the Laboratorio de Parasitología Agrícola of the Centro Multidisciplinario de Investigación Científica 03 (CEMIC-03) of the Universidad Autónoma de Nayarit (UAN).

Trichoderma isolation was done by serial dilutions (1:10) of the soil. Aliquots of 50 μL of the dilutions 10^-4 to 10^-6 were sown in Petri dishes (90 × 15 mm) with Potato-Dextrose-Agar (PDA) culture medium. The fungal microorganisms that showed antagonistic activity were purified using monohyphal cultures (^{Moreno & Albarracín, 2012}; ^{Pérez et al., 2015}; ^{Méndez et al., 2017}). All fungi with purified antagonistic qualities were preliminarily tested in vitro for their antifungal activity against F. falciforme and L. theobromae. Isolates that did not show antifungal properties were not considered in subsequent in vitro tests (^{Mendoza et al., 2020}).

In vitro biocontrol tests on Trichoderma strains

The ability of Trichoderma strains to suppress in vitro the fungal pathogens F. falciforme and L. theobromae was evaluated through an in vitro Dual Confrontation Assay. Both microorganisms (antagonists and pathogens) were confronted in Petri dishes (90 × 15 mm) containing solidified PDA medium. In the Trichoderma confrontation assays, explants (PDA with mycelium and conidia with 7 d of growth) of 6 mm in diameter of both the antagonist and the pathogen were used. Explants of both groups of microorganisms were simultaneously inoculated into the ends of Petri dishes containing PDA culture medium (^{Bell et al., 1982}). In these confrontations, the level of antagonism was determined according to the scale proposed by Bell et al. (1982) and the type of antagonism (antibiosis, mycoparasitism, and/or competition) according to ^{Infante et al. (2009}).

Experimental Petri dishes were incubated at 24 ± 2 °C in darkness until the control pathogen filled the Petri dish. Mycelial growth of pathogenic fungi (alone and confronted) was measured daily until the control pathogen filled the Petri dish (^{Rios et al., 2016}).

In vitro susceptibility testing of F. falciforme and L. theobromae to conventional fungicides

The poisoned medium technique was used according to ^{Albuqueque & Gusqui (2018)}, on solidified PDA + fungicide at the dose recommended by the manufacturer (Table 2) in Petri dishes (90 × 15 mm), 6 mm diameter explants (with surface growth of mycelium and conidia of 7 d of growth) of the pathogenic fungus were placed. Petri dishes with PDA without fungicide were used as controls. The experimental Petri dishes were incubated at 24 ± 2 °C in the dark until the control filled the Petri dish.

Molecular identification of antifungal Trichoderma isolates

Only the strains (T62, T1, and T10) of Trichoderma that inhibited mycelial growth of fungal pathogens by more than 60 % were molecularly characterized. Trichoderma genomic DNA (gDNA) was extracted according to the methodologies described by ^{Allers & Litchen (2000}) and ^{Bobadilla et al. (2020}), from fungal mycelium with 3 d of growth grown in liquid medium (broth) PDB (Potato Dextrose Broth). DNA quality was visualized by electrophoresis on a 1% agarose gel.

DNA was used to amplify the internal transcribed spacer (ITS) of rDNA, with the universal primers ITS5 (5'-GGAAGTAAAAGTCGTAACAAGG-3') and ITS4 (5'-TCTCCTCCGCTTATTGATATATGC-3'), an initial step at 94 °C/5 min, 30 cycles at 94 °C/30 s, 60 °C/30 s, 72 °C/10 min and final elongation at 72 °C/4 min were employed (^{White et al., 1990}; ^{Ruíz et al., 2017}). Detection of PCR products was performed by 1.5 % agarose gel electrophoresis for 110 min at 80V. The 100 bp molecular weight marker (Invitrogen by Thermo Fisher Scientific^®) was used. PCR products were sequenced by Macrogen company in Beotkkot-ro, Geumcheon-gu, Seoul (Gasan-dong, World Meridian I), South Korea. The sequences obtained were compared with those available in the National Center for Biotechnology Information (^{NCBI, 2022}) database by BLAST algorithm, and finally deposited in the NCBI database.

Statistical analysis

In the in vitro confrontations (Trichoderma and fungicides vs. F. falciforme and L. theobromae), nine Petri dishes were used in each confrontation, and a control group (culture of the phytopathogens in the absence of the antagonists-fungicides). In both fungal pathogens, Mycelial Growth Inhibition (MCI) by Trichoderma was evaluated, with the formula MCI= (R1-R2) /R1 ×100, where R1 is the radius of the control pathogen (F. falciforme-L. theobromae) and R2 is the radius of the pathogen in the confrontation (^{Rios et al., 2016}). All experiments were conducted in triplicate independently. MSI data were subjected to analysis of variance (ANOVA) with Statistical Analysis System version 9.0, 2002 (SAS, 2002), and means were separated with Tukey's range test (p= 0.05).

Isolation and in vitro confrontation of antagonistic microorganisms

A total of 12 fungi of the genus Trichoderma were isolated, 10 isolated from Compostela and two from San Blas, of which, in preliminary in vitro tests, only eight showed antagonistic activity against Fusarium falciforme and Lasiodiplodia theobromae. Different species of antagonists may exist, since the rhizosphere is considered a complex ecosystem, with a great diversity of microorganisms (^{Sokolova, 2015}), where climatic or soil conditions are determinants for the survival of microbial communities (^{Ordóñez et al., 2020}).

Inhibition of mycelial growth in vitro of Fusarium falciforme and Lasiodiplodia theobromae by Trichoderma

The strains evaluated showed antifungal potential against F. falciforme and L. theobromae, after 13 and two days of incubation, respectively (Figure 1). The mycelial growth inhibition values by the Trichoderma strains ranged from 0 to 21.72 % for L. theobromae and 32.13 to 75.31 % for F. falciforme. Trichoderma isolates T1, T10, and T62 native to soils near the root zone of soursop trees inhibited the growth of F. falciforme by more than 64.05 %, but against L. theobromae the growth inhibition of these three Trichoderma strains was less than 11.84 %. Isolate T62 was the most effective in inhibiting up to 75.31 % of the mycelial growth of F. falciforme (Figure 1b). The inhibition values of Trichoderma on F. falciforme in our study are among the ranges reported on other Fusarium species,^{Rios et al. (2016}) reported mycelial growth inhibitions on F. oxysporum from 43.3 to 52.3 %, by T. asperellum strains. On the other hand, ^{Cambero et al. (2020}) recorded mycelial growth inhibitions of 49.5 % to 57.9 % against L. pseudotheobromae with Trichoderma strains (T. longibrachiatum, T. harzianum, and T. asperellum) provided by CIAD, A.C., Unidad Cuauhtémoc, Chihuahua.

Equal literals within the same column show statistical equality according to Tukey's test (p = 0.05). 2*, 3*, and 4*= Level on the ^{Bell et al. (1982}) scale.

Figure 1 Trichoderma isolates inhibit the mycelial growth of pathogenic fungi Lasiodiplodia theobromae and Fusarium falciforme. 

Similarly, ^{Bhadra et al. (2014}) reported mycelial growth inhibitions of L. theobromae of up to 80 % with T. koningii and T. viridae attributing the antifungal capacity of these strains to mycoparasitism and its rapid growth, allowing them to compete for space and available nutrients. The mycelial growth inhibition values of L. theobromae by Trichoderma were lower (Figure 1a) than those documented against F. falciforme, this is possible since in this study L. theobromae could grow rapidly and fill the Petri dish in only two days after sowing, having a faster growth than Trichoderma spp. In the same way, the isolates of native Trichoderma presented higher percentages of mycelial growth inhibition than those evidenced by the strains provided by CIAD A.C., suggesting that they are more likely to adapt to the climatological conditions of the producing zones of A. muricata in Nayarit.

Trichoderma suppresses fungal pathogens through different mechanisms, such as mycoparasitism, antibiosis, as well as, space, light, and nutrient competition, as well as plant growth stimulation (^{Martínez et al., 2013}). Competition for space and available nutrients were the biocontrol mechanism evidenced in our Trichoderma strains on the pathogens L. theobromae and F. falciforme. Fungi of the genus Trichoderma have been reported as successful microbial biocontrol agents of phytopathogenic Fusarium species (^{Sánchez et al., 2017}), such as Fusarium oxysporum f. sp. cubense, F. roseum (^{Infante et al., 2009}; ^{Schuster & Schmoll, 2010}) and Lasiodiplodia such as L. pseudotheobromae (^{Cambero et al., 2020}).

In vitro sensitivity of F. falciforme and L. theobromae to conventional fungicides

In vitro, the sensitivity of pathogens to the fungicides tested is shown in Figure 2. Both pathogens showed sensitivity to the eight fungicides tested, with mycelial growth inhibitions of over 74.55 %. Fusarium falciforme and L. theobromae were completely inhibited (100 %) only with the fungicide Mancozeb, possibly as it belongs to the dithiocarbamate group, capable of inhibiting motor proteins and multisite activity (^{Medina et al., 2022}). The fungicides Cyprodinil+Fludioxonil and Mancozeb+Azoxystrobin inhibited the mycelial growth of L. theobromae by 100 %. Fusarium falciforme was less sensitive to the fungicides Boscalid+Pyraclostrobin and Chlorothalonil (Figure 2).

^{Piñeros et al. (2019}) evidenced high sensitivity (100 % MGI) in F. subglutinans and F. graminearum to the fungicides Prochloraz+Difenoconazole, Carboxin+Captan and Tebuconazole+Trifloxystrobin. Michael et al. (2018), obtained MGI of 76.67 % on Fusarium oxysporum f. sp. gladioli with the fungicide Metalaxyl+Chlorothalonil. ^{Tovar et al. (2013}) proposed the use of Cyprodinil+Fludioxonil, Pyraclostrobin+Boscalid, Prochloraz, Tebuconazole, and Iprodine as the most effective fungicides for in vitro control of L. theobromae on mamey. Fungicides belong to different chemical groups, therefore, they have different mechanisms of action, they can inhibit mycelial growth, conidial germination, and respiration, and alter the cell membrane, among others (^{FRAC, 2020}), however, in response to the pressure exerted by fungicides, plant pathogenic fungi can acquire resistance or lose sensitivity to them (^{Li et al., 2020}). To prevent pathogens from developing resistance to fungicides, it is suggested to use systemic and contact fungicides in rotation or combination (^{Denman et al., 2004}).

Both fungal pathogens (F. falciforme and L. theobromae) showed mycelial growth 10 d after being treated with the fungicides, evidencing only fungistatic action by Mancozeb+Azoxystrobin, Boscalid+Pyraclostrobin, Pidiflumethofen+ Difenoconazole and Chlorothalonil, possibly these fungicides do not completely inhibit conidia germination of these causal agents, which is essential for successful chemical control of pathogenic fungi in plants (^{Shin et al., 2014}), the development of resistance to fungicides may be one of the main reasons (^{Moreira et al., 2021}).

Means (± standard error) with equal literals indicate statistical equality according to Tukey's range test (p = 0.05). Man+Azo = Mancozeb+Azoxystrobin; Cip+Flu = Cyprodinil+Fludioxonil; Bos+Pir = Boscalid+Piraclostrobin; Pid+Dif = Pidiflumetofen+Diphenoconazole; Car = Carbendazim; *The microorganism evaluated against this fungicide continued with very slow growth (no measurement).

Figure 2 In vitro sensitivity of Lasiodiplodia theobromae and Fusarium falciforme to conventional fungicides. 

These control alternatives can be considered in integrated disease management and a timely diagnosis will allow us to have better results. On the other hand, the use of microorganisms can be applied preventively, and fungicides can be used only in cases where the problem begins to exceed the economic threshold.

Molecularly identified native antifungal Trichoderma isolates

Eight Trichoderma isolates were obtained from the municipalities of Compostela and San Blas, Nayarit, Mexico, only the isolates (T62, T1, T10) were identified as inhibiting >60 % mycelial growth of Fusarium falciforme. According to its molecular features and when compared with NCBI database available sequences, 100% similarity of isolates T62 with T. hamatum, T1 with T. asperellum, and isolate T10 with T. asperellum was observed. Sequences of these isolates were registered in GenBank under the following accession numbers (OQ696060, OQ696061, and OQ696062, respectively). It is worth mentioning that this is the first record of these antagonists in commercial soursop orchards in Nayarit, Mexico. Strains of T. hamatum, have been identified in Catamarca, Argentina, with antagonistic capacity against Verticillium dahliae KLEB (^{Gonzáles et al., 2021}), on the other hand, T. asperellum has been identified in Peru in avocado (Persea americana Mill.) orchards with biocontrol potential against Phytophthora cinnamomi (^{Morales et al., 2020}), although there is a wide diversity of Trichoderma species such as T. asperellum, T. atroviridae, T. koningiopsis, T. piluliferum, T. viridae among others (^{Sánchez et al., 2018}).