Endophytic Trichoderma Species Isolated from Persea americana and Cinnamomum verum Roots Reduce Symptoms Caused by Phytophthora cinnamomi in Avocado

Avocado root rot caused by the oomycete Phytophthora cinnamomi is a severe disease that affects avocado production in Mexico and worldwide. The use of biological control agents such as Trichoderma species isolated from places where the disease is always present, represents an efficient alternative to reduce losses. Thus, the objective of this research was to evaluate the biocontrol ability of 10 endophytic Trichoderma spp. strains against P. cinnamomi tested both in vitro and in the greenhouse. The endophytic Trichoderma spp. were recovered from Persea americana and Cinnamomum verum roots, isolated and purified on potato–dextrose–agar medium. Ten strains were identified by phylogenetic reconstruction of the internal transcribed spacer region of rDNA sequences as T. asperellum (T-AS1, T-AS2, T-AS6, and T-AS7), T. harzianum (T-H3, T-H4, and T-H5), T. hamatum (T-A12), T. koningiopsis (T-K8 and T-K11), and P. cinnamomi (CPO-PCU). In vitro dual-culture assay, the percentage of inhibition of radial growth (PIRG) between Trichoderma spp. and P. cinnamomi strains was measured according to the Bell’s scale. PIRG results indicated that T-AS2 reached the highest value of 78.32%, and T-H5 reached the lowest value of 38.66%. In the greenhouse, the infection was evaluated according to the percentage of disease incidence. Plants with the lowest incidence of dead by avocado root rot were those whose seedlings were inoculated with T-AS2 and T-AS7, resulting in only 5% death by root rot caused by P. cinnamomi. The disease incidence of seedlings with wilt symptoms and death decreased more than 50% in the presence of Trichoderma spp. Relying on the results, we conclude that T. asperellum and T. harzianum contribute to the biocontrol of soil-borne pathogenic oomycete P. cinnamomi.


Introduction
The avocado (Persea americana Mill) is one of the most important fruit worldwide; its demand is increasing in the international market due to its nutraceutical properties and its use as a food supplement [1]. The global production of avocado was approximately 6048 million tons in 2017, of which Mexico produced approximately 583,426 tons on a surface of 188,723 ha −1 [2]. Currently, the yields of avocado plantations in the intended areas for this crop are decreasing, mainly due to phytosanitary problems that are difficult to prevent and control [3].
Avocado root rot is the major disease caused by Phytophthora cinnamomi Rands; this same oomycete infects a wide host range of wild and cultivated plants worldwide. This disease severely affects avocado plantations representing a threat to natural ecosystems [4,5].
This disease is also known in Mexico as avocado sadness in which the root and feeder roots shown typical root rot symptoms [1,6]. In Mexico, avocado root rot has been reported in the states of Queretaro, Guanajuato, Michoacan, and Puebla, with a mortality that reaches up to 100% attributed to P. cinnamomi infection [7].
In conventional avocado production, to mitigate losses caused by this pathogen, the most common practice is the indiscriminate application of high amounts of chemical fungicides and synthetic fertilizers [8], which cause a negative environmental impact on agroecosystems. In contrast, the use of cultural practices, biological control, and resistant rootstocks as a strategy to reduce the incidence of the disease offer sustainable alternatives to P. cinnamomi management [9].
The use of filamentous fungi as biocontrol agents represents an effective alternative for agricultural production systems because it reduces the numerous applications of various types of fungicides. For more than eight decades, research has been conducted on the use of different species of Trichoderma. Multiple mechanisms of action have been described, but the most relevant are (a) the production of secondary metabolites that strengthen the plant's immune system [10,11], (b) the action of natural mycoparasites and antagonistic agents [12] and (c) the promotion of seed germination and plant growth, as well as an increased root and foliage biomass and mineral assimilation [13][14][15].
Thus, different indigenous species of Trichoderma play an important role in agroforestry and natural ecosystems; they are considered biological control agents because they act against pathogens of agricultural importance, such as P. cinnamomi [16]. Investigations of dual in vitro confrontations with different plant pathogens and Trichoderma spp. have been frequently conducted [17,18]. However, there are no experiments performed in vitro between Trichoderma spp. and P. cinnamomi along with inoculated avocado seedlings under greenhouse.
Recently, different species of Trichoderma have been evaluated as antagonists of soil-borne, such as Phytophthora capsici with T. harzianum in pepper [15], Phytophthora nicotianae with T. harzianum and T. asperellum in pepper [19], Phytophthora melonis with T. harzianum in cucumber [20], P. cinnamomi with T. atroviride in tomato [18] and the oomycete Pythium ultimum with T. cerinum. This last antagonist was isolated from avocado roots [12,21]. In this sense, there are several studies related to indigenous species of Trichoderma recovered from the plant rhizosphere and their inference in the health of plants. For this study, some indigenous endophytic Trichoderma species isolated from avocado and cinnamon roots were tested in vitro against P. cinnamomi, and their influence on avocado seedling health under greenhouse conditions.

Sampling of Symptomatic Avocado and Cinnamon Plants
To determine the presence of indigenous species of Trichoderma during the spring and summer seasons of 2018, avocado rootstock P. americana var. drymifolia roots were recovered from the states of Uruapan, Michoacan (19 • 34 31.80 W). At each site, roots with rhizosphere soil were collected from cinnamon and avocado plants that showed chlorosis and wilting foliage, five samples were collected at a depth of 0 to 30 cm and kept in plastic bags until further analysis.

Fungal Isolates
The roots from avocado and cinnamon were washed with tap water, followed by immersion in 1% sodium hypochlorite for 30 s, rinsed twice with sterile distilled water, wrapped with sterilized Plants 2020, 9, 1220 3 of 17 paper towels, and placed in a laminar flow chamber at 20 • C for 15 min to dry. Later, the roots were fragmented into 1 cm pieces and placed vertically by embedding a half centimeter in a Petri plate with potato dextrose agar (PDA, Dioxon) medium amended with chloramphenicol (Sigma-Aldrich, St. Louis, MO, USA) at 20 mg mL −1 . The plates were maintained at 25 • C for 40 h and then under white light at 20 • C for 3 days to stimulate sporulation above the root not embedded in the PDA medium ( Figure 1). After three days, the plates were examined, and they showed colonies with morphological characteristics similar to those of Trichoderma spp. They were transferred to fresh PDA to obtain pure cultures by the single conidium technique. A Phytophthora cinnamomi (CPO-PCU) was isolated on V8 juice agar medium amended with chloramphenicol at 20 µg mL −1 and then incubated in the dark at 28 • C for 72 h. The plates were examined, and they showed morphological characteristics (hyphal swellings, chlamydospore and sporangium) similar to P. cinnamomi [5,22]. The roots from avocado and cinnamon were washed with tap water, followed by immersion in 1% sodium hypochlorite for 30 s, rinsed twice with sterile distilled water, wrapped with sterilized paper towels, and placed in a laminar flow chamber at 20 °C for 15 min to dry. Later, the roots were fragmented into 1 cm pieces and placed vertically by embedding a half centimeter in a Petri plate with potato dextrose agar (PDA, Dioxon) medium amended with chloramphenicol (Sigma-Aldrich, St. Louis, MO, USA) at 20 mg mL −1 . The plates were maintained at 25 °C for 40 h and then under white light at 20 °C for 3 days to stimulate sporulation above the root not embedded in the PDA medium ( Figure 1). After three days, the plates were examined, and they showed colonies with morphological characteristics similar to those of Trichoderma spp. They were transferred to fresh PDA to obtain pure cultures by the single conidium technique. A Phytophthora cinnamomi (CPO-PCU) was isolated on V8 juice agar medium amended with chloramphenicol at 20 µg mL −1 and then incubated in the dark at 28 °C for 72 h. The plates were examined, and they showed morphological characteristics (hyphal swellings, chlamydospore and sporangium) similar to P. cinnamomi [5,22].

DNA Extraction, PCR Amplification and Sequencing
This procedure was performed with the 2% cetyl trimethylammonium bromide (CTAB) method according to Doyle and Doyle [23] with some modifications [24]. Genomic DNA was suspended in 100 µL of sterile HPLC water and quantified by spectrophotometry in a NanoDrop 2000 C (Thermo Scientific, Waltham, MA, USA). To determine the DNA quality, absorbance values between 1.8 and 2.2 at A280/260 and A230/260 nm were considered acceptable. Finally, the DNA was diluted to 20 ng µL −1 and then stored at −20 °C for PCR amplification.
All PCR products were cleaned with ExoSAP-IT (Affymetrix, Santa Clara, CA, USA), and both strands were individually sequenced using the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Carlsbad, CA, USA) in a 3130 Genetic Analyzer Sequencer (Applied

DNA Extraction, PCR Amplification and Sequencing
This procedure was performed with the 2% cetyl trimethylammonium bromide (CTAB) method according to Doyle and Doyle [23] with some modifications [24]. Genomic DNA was suspended in 100 µL of sterile HPLC water and quantified by spectrophotometry in a NanoDrop 2000 C (Thermo Scientific, Waltham, MA, USA). To determine the DNA quality, absorbance values between 1.8 and 2.2 at A280/260 and A230/260 nm were considered acceptable. Finally, the DNA was diluted to 20 ng µL −1 and then stored at −20 • C for PCR amplification.
All PCR products were cleaned with ExoSAP-IT (Affymetrix, Santa Clara, CA, USA), and both strands were individually sequenced using the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Carlsbad, CA, USA) in a 3130 Genetic Analyzer Sequencer (Applied Biosystems, Carlsbad, CA, USA) at Postgraduate College Facilities, Mexico, according to Juárez-Vázquez et al. [26].

Phylogenetic Analysis
The sequences of both DNA strands obtained in this study were assembled and trimmed with BioEdit v7.0.5 software [27] to create consensus sequences for each strain. These sequences were compared with the sequences deposited in the GenBank database of the National Center for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/) with the BLASTN v2.2.19 algorithm [28]. Reference and related sequences of Trichoderma spp. were downloaded from the GenBank database and included along with the sequences obtained in this study (MK778890, MK779008, MK791646, MK791647, MK791650, MK780094, MK779064, MK784067, MK791649, and MK791648). All consensus sequences were compiled into a single file, and then multiple sequences were aligned using Muscle [29] included in MEGA 7 software [30].
Phylogenetic reconstruction was performed with Bayesian inference using MrBayes v3.2.2 [31] with the evolutionary model GTR + gamma + invariant positions [32]. According to the instructions, the first 25% of the generated trees was discarded as the burn-in phase in each analysis, and posterior probabilities were determined for the remaining trees. The phylogenetic tree was rooted with Hypomyces corticiicola K. Põldmaa strain CBS 137.71 sequence. The standard deviation of split frequencies was stopped when the value was <0.01. Additionally, the Phytophthora isolated in this study was named here as P. cinnamomi strain CPO-PCU (GenBank accession number JQ266267). It was identified by the BLASTn algorithm and the distance tree of the result as FJ746646) deposited by the American Type Culture Collection with 100% maximum identity.

Morphological Characterization of Trichoderma spp.
Isolates of Trichoderma spp. were cultured on PDA medium and incubated at 25 • C for 5 days to determine their characteristics according to the taxonomic keys of Barnett and Hunter [33]. Later, discs of PDA (5 mm in diameter) of each strain were transferred according to Harris' [34] modified method, in which microcultures were placed on a sterile rod triangle inside a Petri plate and incubated at 25 • C for 12 h and 24 h in the dark.
The macro and microscopic characteristics of Trichoderma spp. assessed were the presence of colony color, texture, presence of aerial mycelium and concentric rings, mycelium appearance on the upper and reverse sides, conidia shape and size (length and width), and morphology of 50 phialides were scored [15,[35][36][37][38]. All Trichoderma spp. structures were observed under an optical microscope (Zeiss Axioskop 2 plus, Göttingen, Germany) at 40× and 100× magnification.

Antagonistic In Vitro Assessment of Trichoderma spp.
The variables considered for this analysis were the percentage of inhibition of radial growth (PIRG) and colonization with Bell's scale [39]. For the in vitro test, T. asperellum (T-AS1, T-AS2, T-AS6, and T-AS7), T. harzianum (T-H3, T-H4, and T-H5), T. hamatum (T-A12), and T. koningiopsis (T-K8 and T-K11) were considered for dual confrontations versus P. cinnamomi in a completely randomized experimental design with 10 treatments and four repetitions. PDA discs (5 mm in diameter) with mycelia of Trichoderma spp. and P. cinnamomi were placed at the extremes of Petri plates containing PDA and incubated at 28 • C for 72 h. Then, mycelial growth was scored every 12 h until the first contact between the mycelia of each antagonist T. harzianum, T. asperellum, T. koningiopsis, and T. hamatum with the oomycete P. cinnamomi occurred [15,40,41].
Antagonism and percentage of inhibition were evaluated considering the mycelial growth radius of Trichoderma spp. and P. cinnamomi (with its respective control). Invasion of the antagonist or colonization on the surface of the P. cinnamomi mycelium was taken as the index of antagonism with the scale proposed by Bell [39] (Table 2). Table 2. Trichoderma strain antagonism evaluated in vitro using Bell's scale [39], considering the invasion of the surface.

Scale
Antagonistic Ability

1
Trichoderma completely overgrew the P. cinnamomi and covered the entire medium surface. 2 Trichoderma overgrew at least two-thirds of the medium surface. 3 Trichoderma and the P. cinnamomi each colonized approximately one-half of the medium surface and neither organism appeared to dominate the other. 4 P. cinnamomi colonized at least two-thirds of the medium surface and appeared to withstand encroachment by Trichoderma. 5 P. cinnamomi completely overgrew the Trichoderma and occupied the entire medium surface.

Mycoparasitism and Colonization by Trichoderma spp. over P. cinnamomi
The colonization ability of Trichoderma spp. strains was evaluated by a dual-culture technique [42] that consisted of measuring the mycelial overgrowth of Trichoderma sp. every 12 h for 8 days from when the first contact occurred until the pathogen P. cinnamomi was completely covered (Figure 2). To calculate the percentage of Trichoderma colonization above P. cinnamomi, the formula proposed by Camporota [41] was used (Equation (2)).
where C% = The percentage of colonization; DTP = The distance of the route through which the colony of Trichoderma spp. grows over the pathogen Phytophthora cinnamomi colony, taking into account the axis that separates both colonies; DE = The distance between both colonies. As a described mechanism of action of Trichoderma spp., macroscopic observations were made on the dual-cultures, considering invasion of the antagonist on the surface of the pathogenic mycelium as the index of mycoparasitism. The mycoparasitism was determined by microscopic observation on a Zeiss Axioskop plus (Zeiss, Göttingen, Germany) at 40× and 100× magnifications; the interactions of the antagonistic hyphae with those of the pathogen, either by coiling, vacuolation or penetration were documented.
Individual growth radii of the Trichoderma spp. antagonists and oomycete P. cinnamomi were measured by using a Vernier every 12 h until the Petri plate containing PDA was completely overgrown.
The oomycete P. cinnamomi was incubated on PDA for 7 days. For the production of sporangia of P. cinnamomi, agar discs (5 mm in diameter) with mycelia were placed into 500-mL flasks with V8 juice liquid medium [5] and incubated at 28 °C for 8 days. Subsequently, the flasks were placed at 4 °C for 5 min; after that, the number of zoospores per milliliter was quantified with a hemocytometer. Finally, the zoospores were collected and adjusted to a final concentration of 10 6 zoospores mL −1 .

Avocado Seedling Inoculation
A greenhouse experiment was carried out to evaluate the antagonistic ability of 10 Trichoderma isolates recovered from avocado and cinnamon roots. For the experiment, the substrate was prepared with a mixture of peat moss, agricultural soil and vermiculite (2:2:1) sterilized with wet steam at 15 psi for 4 h and placed into 1 L plastic pot. Avocado seeds var. drymifolia were collected from trees in the municipality of Atlixco in Puebla, Mexico, which has been recognized as having moderate resistance to infection by P. cinnamomi [5]. They were disinfested with 3% sodium hypochlorite for 10 min, rinsed with sterile distilled water, and planted in plastic pots.
A total of 12 treatments, with 4 replicates per treatment (Table 1) were distributed in the greenhouse using a complete randomized design. Inoculations of 20-60 day old avocado seedlings (15 to 20 cm high) with Trichoderma spp. and P. cinnamomi were performed by adding 5 mL of 10 6 conidia mL −1 and 10 6 zoospores mL −1 to the substrate, while control seedlings were inoculated with As a described mechanism of action of Trichoderma spp., macroscopic observations were made on the dual-cultures, considering invasion of the antagonist on the surface of the pathogenic mycelium as the index of mycoparasitism. The mycoparasitism was determined by microscopic observation on a Zeiss Axioskop plus (Zeiss, Göttingen, Germany) at 40× and 100× magnifications; the interactions of the antagonistic hyphae with those of the pathogen, either by coiling, vacuolation or penetration were documented.
Individual growth radii of the Trichoderma spp. antagonists and oomycete P. cinnamomi were measured by using a Vernier every 12 h until the Petri plate containing PDA was completely overgrown. Later, they were exposed for 3 days to white light to induce the formation of conidia. Finally, conidia were harvested by filtering through a sterile cheesecloth [42]. Then, the conidial suspension was adjusted to a final concentration of 10 6 conidia mL −1 using a hemocytometer (Marienfeld, Lauda-Königshofen, Germany).
The oomycete P. cinnamomi was incubated on PDA for 7 days. For the production of sporangia of P. cinnamomi, agar discs (5 mm in diameter) with mycelia were placed into 500-mL flasks with V8 juice liquid medium [5] and incubated at 28 • C for 8 days. Subsequently, the flasks were placed at 4 • C for 5 min; after that, the number of zoospores per milliliter was quantified with a hemocytometer. Finally, the zoospores were collected and adjusted to a final concentration of 10 6 zoospores mL −1 .

Avocado Seedling Inoculation
A greenhouse experiment was carried out to evaluate the antagonistic ability of 10 Trichoderma isolates recovered from avocado and cinnamon roots. For the experiment, the substrate was prepared with a mixture of peat moss, agricultural soil and vermiculite (2:2:1) sterilized with wet steam at 15 psi for 4 h and placed into 1 L plastic pot. Avocado seeds var. drymifolia were collected from trees in the municipality of Atlixco in Puebla, Mexico, which has been recognized as having moderate resistance to infection by P. cinnamomi [5]. They were disinfested with 3% sodium hypochlorite for 10 min, rinsed with sterile distilled water, and planted in plastic pots.
A total of 12 treatments, with 4 replicates per treatment (Table 1) were distributed in the greenhouse using a complete randomized design. Inoculations of 20-60 day old avocado seedlings (15 to 20 cm high) with Trichoderma spp. and P. cinnamomi were performed by adding 5 mL of 10 6 conidia mL −1 and 10 6 zoospores mL −1 to the substrate, while control seedlings were inoculated with sterile distilled water; an additional control was P. cinnamomi alone. The avocado seedlings were maintained in the greenhouse for three months.
The disease incidence (Ii), based on visual symptoms such as the number of asymptomatic seedlings and the appearance of wilt symptoms and dead seedlings was evaluated at 20 days after inoculation (dai) with the following equation (Equation (3)): where Ii = Incidence of diseased seedlings at time i; ni = Number of diseased seedlings at time i; Ni = Total population of inoculated seedlings.

Statistical Analysis
The PIRG, in vitro colonization percentage (C%) and in vivo incidence of disease (Ii%), were expressed as percentages and transformed with angular arccosine √ x + 1. The mean values of the four repetitions (five plants per repetition) of the biological samples were subjected to an analysis of variance with Statistical Analysis System software SAS [43], using the Tukey test to determine the significant differences between the treatments and level of significance of p < 0.05.

Isolates Recovered
Ten Trichoderma spp. and one Phytophthora isolate were recovered from avocado and cinnamon roots from five states in Mexico. They were studied to establish their identity based on phylogenetic and morphological characteristics. Additionally, to determine the Trichoderma spp. biocontrol activity over the oomycete Phytophthora sp. in vitro and under greenhouse conditions.

Phylogenetic Analysis
This analysis was performed with sequences of a 621 bp fragment of the ITS region corresponding to Trichoderma spp. and 849 bp of P. cinnamomi. The isolates in study were clustered within four subclades belonging to T. asperellum (T-AS1, T-AS2, T-AS6, and T-AS7), T. hamatum (T-A12), T. harzianum (T-H3, T-H4, and T-H5), and T. koningiopsis (T-K8 and T-K11). A phylogenetic tree constructed with Bayesian inference was run until 2 million generations with a final standard deviation of 0.008126.
The evolutionary history of the indigenous Trichoderma species represented in the phylogenetic tree ( Figure 3) shows that they represented four species, which are part of the soil microorganism community in the avocado ecological niche. Furthermore, these species can coexist in the same root area of the same avocado tree of the drymifolia variety.

Morphological Characterization of Trichoderma spp.
The results indicated the presence of 10 Trichoderma strains that belong to four species. They were characterized by the presence of a light to dark green colony color, aerial mycelium, light yellow staining of the PDA, and one to three concentric rings (Figure 4A,D,G,J and Table S1).
Strains of T. asperellum were characterized by the absence of color on the reverse side of the Petri plate. T. harzianum presented light to dark green colonies with a dusty-cottony texture and aerial mycelia ( Figure 4A), T. hamatum showed yellow pigment on the reverse side of the Petri plate, and T. koningiopsis presented flocculated aerial mycelia that adhered to the culture medium ( Figure 4G).
The conidia of T. harzianum were subglobose, with elongated globus phyloids distributed asymmetrically ( Figure 4B,C). T. asperellum presented bottle-shaped phialides with a length of 10.2 to 11.8 µm and arranged in spiral shapes of 2, 3, and 4, and the conidia were globose to subglobose and occasionally ovoid (Table S1). The chlamydospores in T. asperellum appeared at the end of the mycelium with a globose to subglobose form ( Figure 4K,L); the morphological characteristics of the colonies of Trichoderma spp. isolated in this study are already mentioned above, described in Table  S1.

Morphological Characterization of Trichoderma spp.
The results indicated the presence of 10 Trichoderma strains that belong to four species. They were characterized by the presence of a light to dark green colony color, aerial mycelium, light yellow staining of the PDA, and one to three concentric rings (Figure 4A,D,G,J and Table S1).

Percentage of Growth Inhibition In Vitro
In the dual-assays, some Trichoderma spp. could compete for space and inhibit the growth of P. cinnamomi (Table 3). Double confrontation between T. asperellum (T-AS1 and T-AS2) and T. harzianum (T-H3 and T-H4 strains) showed an inhibition above 65.3% (p < 0.05). However, the highest percentage of inhibition, 78.32%, was obtained with T. asperellum (T-AS2 strain). Nevertheless, according to the results, these strains have the potential to inhibit the mycelial growth of the P. cinnamomi strain CPO-CPU. In the same way, T. harzianum strain T-H3 presented 73.33% inhibition ( Table 3). The data analysis showed that mycelial growth inhibition of P. cinnamomi by Trichoderma species had significant differences (p < 0.05).
The lowest percentage of inhibition was observed for T. harzianum strain T-H5 with 38.66%. In addition, T-H4 and T-AS1 strains belonging to T. harzianum and T. asperellum, respectively, presented moderate inhibition of 69.15%. The species with low to moderate performance were T. asperellum T-AS7 and T-AS6 strains, with 51.5% and 53.33%, respectively; T. hamatum strain T-A12 with 50.99%; T. koningiopsis T-K8 and T-K11 strains with 46.99% and 47.33%, respectively (Table 3). Strains of T. asperellum were characterized by the absence of color on the reverse side of the Petri plate. T. harzianum presented light to dark green colonies with a dusty-cottony texture and aerial mycelia ( Figure 4A), T. hamatum showed yellow pigment on the reverse side of the Petri plate, and T. koningiopsis presented flocculated aerial mycelia that adhered to the culture medium ( Figure 4G).
The conidia of T. harzianum were subglobose, with elongated globus phyloids distributed asymmetrically ( Figure 4B,C). T. asperellum presented bottle-shaped phialides with a length of 10.2 to 11.8 µm and arranged in spiral shapes of 2, 3, and 4, and the conidia were globose to subglobose and occasionally ovoid (Table S1). The chlamydospores in T. asperellum appeared at the end of the mycelium with a globose to subglobose form ( Figure 4K,L); the morphological characteristics of the colonies of Trichoderma spp. isolated in this study are already mentioned above, described in Table S1.

Percentage of Growth Inhibition In Vitro
In the dual-assays, some Trichoderma spp. could compete for space and inhibit the growth of P. cinnamomi (Table 3). Double confrontation between T. asperellum (T-AS1 and T-AS2) and T. harzianum (T-H3 and T-H4 strains) showed an inhibition above 65.3% (p < 0.05). However, the highest percentage of inhibition, 78.32%, was obtained with T. asperellum (T-AS2 strain). Nevertheless, according to the results, these strains have the potential to inhibit the mycelial growth of the P. cinnamomi strain CPO-CPU. In the same way, T. harzianum strain T-H3 presented 73.33% inhibition ( Table 3). The data analysis showed that mycelial growth inhibition of P. cinnamomi by Trichoderma species had significant differences (p < 0.05). The lowest percentage of inhibition was observed for T. harzianum strain T-H5 with 38.66%. In addition, T-H4 and T-AS1 strains belonging to T. harzianum and T. asperellum, respectively, presented moderate inhibition of 69.15%. The species with low to moderate performance were T. asperellum T-AS7 and T-AS6 strains, with 51.5% and 53.33%, respectively; T. hamatum strain T-A12 with 50.99%; T. koningiopsis T-K8 and T-K11 strains with 46.99% and 47.33%, respectively (Table 3).

Mycoparasitism, Percentage of Colonization, and Growth Rate
In the dual confrontation test, the 10 Trichoderma strains had the ability to colonize the P. cinnamomi strain CPO-CPU, reducing mycelial growth ( Table 4). The best colonization percentage was observed with T. koningiopsis T-K11 (67.83%) and T. harzianum T-H3 (60%) strains. Generally, most of the Trichoderma showed significant differences (p < 0.05) and they reduced the mycelial growth of the P. cinnamomi strain CPO-CPU. However, the strains of T. asperellum (T-AS2, T-AS6, and T-AS7) and T. koningiopsis (T-K8) presented intermediate colonization. Table 4. Test of means of the growth rate of Trichoderma spp. and Phytophthora cinnamomi.

Strain
Trichoderma The results showed that the inhibition and colonization of all Trichoderma strains were efficient considering the competition for space and nutrients. In addition, the parasitic ability against the P. cinnamomi strain CPO-CPU was observed by light microscopy and corroborated in microcultures ( Figure 5A-D). Additionally, in the dual-assays, which were performed by facing up the indigenous strains of Trichoderma spp. and P. cinnamomi strain CPO-CPU, there were different types of mycoparasitism as vacuolation, surrounding and lysing the mycelium of the pathogen and abnormal mycelium, and massive hyphal curl of T. hamatum and T. koningiopsis covering the pathogen P. cinnamomi ( Figure 5A-D).
The results showed that the inhibition and colonization of all Trichoderma strains were efficient considering the competition for space and nutrients. In addition, the parasitic ability against the P. cinnamomi strain CPO-CPU was observed by light microscopy and corroborated in microcultures ( Figure 5A-D). Additionally, in the dual-assays, which were performed by facing up the indigenous strains of Trichoderma spp. and P. cinnamomi strain CPO-CPU, there were different types of mycoparasitism as vacuolation, surrounding and lysing the mycelium of the pathogen and abnormal mycelium, and massive hyphal curl of T. hamatum and T. koningiopsis covering the pathogen P. cinnamomi ( Figure 5A-D).
According to Bell's scale [40] (Tables 2 and 3), strains of the Trichoderma species were classified into two types: In scale 1, when biocontrol completely covers the surface, it should be occupied by the pathogen. In this case, T. koningiopsis (T-K8 and T-K11), T. harzianum (T-H3), and T. asperellum (T-AS6) were overgrown on the P. cinnamomi strain CPO-CPU. In scale 2, when Trichoderma overgrew at least two-thirds of the medium surface, it should be occupied by the pathogen. For this case, T. harzianum (T-H4 and T-H5), T. hamatum (T-A12), and T. asperellum (T-AS1) strains presented a colonization value between 23.83% and 30% on the P. cinnamomi strain CPO-CPU (Table 3).  According to Bell's scale [40] (Tables 2 and 3), strains of the Trichoderma species were classified into two types: In scale 1, when biocontrol completely covers the surface, it should be occupied by the pathogen. In this case, T. koningiopsis (T-K8 and T-K11), T. harzianum (T-H3), and T. asperellum (T-AS6) were overgrown on the P. cinnamomi strain CPO-CPU. In scale 2, when Trichoderma overgrew at least two-thirds of the medium surface, it should be occupied by the pathogen. For this case, T. harzianum (T-H4 and T-H5), T. hamatum (T-A12), and T. asperellum (T-AS1) strains presented a colonization value between 23.83% and 30% on the P. cinnamomi strain CPO-CPU (Table 3).
The P. cinnamomi strain CPO-CPU had uniform radial growth with no significant differences in the growth rate, while the Trichoderma strains showed significant differences (p < 0.05) related to the individual growth rate. The strains that showed the best development rates were T. koningiopsis (T-K8) and T. hamatum (T-A12); otherwise, the lowest levels were presented by T. harzianum (T-H5) and T. asperellum (T-AS2, T-AS6, and T-AS7), with similar behaviors (Table 4). These results indicated the antagonistic capacity of Trichoderma species in controlling the P. cinnamomi strain CPO-CPU growth under in vitro conditions.

Biocontrol of P. cinnamomi by Trichoderma Strains in the Greenhouse Assay
Avocado seedlings inoculated in the greenhouse with both Trichoderma and P. cinnamomi strain CPO-CPU showed the first visible symptoms at 20 days. The disease incidence (Ii) caused by P. cinnamomi leads to a high mortality rate of avocado seedlings; although, it was observed that the antagonistic effect of the different isolates of Trichoderma spp. reduced the absence of symptoms of wilt or root rot in relation to the seedling control group ( Figure S1A-E), showing significant differences (p < 0.05).
When the antagonists inoculated avocado seedlings previously infected with P. cinnamomi strain CPO-CPU, there were significant differences (p < 0.05) due to the absence of wilt or root rot symptoms in relation to the seedling control. The inoculation of avocado seedlings with T. asperellum (T-AS2 and T-AS7) resulted in only 5% death by root rot caused by P. cinnamomi strain CPO-CPU ( Figure 6).

Discussion
In this work, we have demonstrated that indigenous endophytic Trichoderma spp. recovered from avocado and cinnamon roots were efficient in vitro and under greenhouse conditions as biological control agents against the soil-borne pathogen P. cinnamomi, reducing the incidence of root rot in avocado seedlings.
Species of the genus Trichoderma, well known as green-spored fungi, have been reported widely as biological control agents against plant diseases caused mainly by soil-borne pathogens [44,45]. They are filamentous fungi that exhibit different modes of action such as growth regulators, competition by space and nutrients, and infection sites and reduce the damage by pathogens that affect the development of the roots [8,17,46]. Additionally, the Trichoderma genome has revealed the presence of genes involved in a mutualistic association between plants and biological control agents, in which some species of this genus can be established in the rhizosphere of the plant [47].
In this study, Trichoderma species, such as T. asperellum, T. hamatum, T. harzianum, and T. koningiopsis, were clearly identified based on phylogenetic and morphological approaches and have been frequently isolated from the rhizosphere worldwide [38,48]. Phylogenetic reconstruction did not show any discrepancy with the morphological features found in this work, and phylogenetic studies using the ITS region were able to classify each strain within the corresponding species.
The four species of Trichoderma identified in this study are inhabitants of different regions and tropical systems around the world [46,49]. Their ecological plasticity is related to the capacity to The disease incidence of seedlings with wilt symptoms and death caused by P. cinnamomi strain CPO-CPU decreased more than 50% in the presence of Trichoderma spp. The strain with the best behavior for controlling this pathogenic oomycete was T. asperellum (T-AS7), with 80% of seedlings without evident symptoms ( Figure 6).

Discussion
In this work, we have demonstrated that indigenous endophytic Trichoderma spp. recovered from avocado and cinnamon roots were efficient in vitro and under greenhouse conditions as biological control agents against the soil-borne pathogen P. cinnamomi, reducing the incidence of root rot in avocado seedlings.
Species of the genus Trichoderma, well known as green-spored fungi, have been reported widely as biological control agents against plant diseases caused mainly by soil-borne pathogens [44,45]. They are filamentous fungi that exhibit different modes of action such as growth regulators, competition by space and nutrients, and infection sites and reduce the damage by pathogens that affect the development of the roots [8,17,46]. Additionally, the Trichoderma genome has revealed the presence of genes involved in a mutualistic association between plants and biological control agents, in which some species of this genus can be established in the rhizosphere of the plant [47].
In this study, Trichoderma species, such as T. asperellum, T. hamatum, T. harzianum, and T. koningiopsis, were clearly identified based on phylogenetic and morphological approaches and have been frequently isolated from the rhizosphere worldwide [38,48]. Phylogenetic reconstruction did not show any discrepancy with the morphological features found in this work, and phylogenetic studies using the ITS region were able to classify each strain within the corresponding species.
The four species of Trichoderma identified in this study are inhabitants of different regions and tropical systems around the world [46,49]. Their ecological plasticity is related to the capacity to produce enzymes to degrade substrates with a high organic matter content found in soil, rhizosphere, leaf litter, wood, aquatic systems, and less frequently in biological tissues [36,50]. The strains belonging to T. harzianum have a greater capacity than the soil-borne plant pathogen to compete for root space [16,41]. Similar behavior was observed in T. asperellum [17,48], T. koningiopsis, and T. hamatum, which act as antagonistic agents during root colonization [49,51]. However, this last species has not yet been isolated from cinnamon and avocado roots to date. It can be indicative that they have evolved with the host to settle in the rhizosphere or live like an endophytic fungus inside the roots of these trees.
The endophytic capacity and ability of a certain group of individuals to create a strengthened relationship between the host and other microbial species, allowing them to coexist like an ecological unit, is known as a halobiont [52]. This feature explains the recasts in the complex association between Trichoderma spp. with avocado roots [15,53].
Antagonistic activity of these Trichoderma spp. evaluated with Bell's scale [39] showed that most strains were scored in class 1, with Trichoderma completely overgrown on P. cinnamomi strain CPO-CPU, covering the entire surface of the culture medium. Occasionally, some Trichoderma strains were in class 2, covering at least two-thirds of culture medium surface, suggesting the importance of these strains as potential biological control agents.
Additionally, when healthy avocado seedlings with the absence of symptoms were inoculated with T. harzianum (T-H5), T. asperellum (T-AS2), and T. koningiopsis (T-K11), they presented a lower Ii of root rot or wilt symptoms along with T. asperellum (T-AS7). In the case of T. harzianum (T-H4 and T-AS6) that exhibited wilting symptoms and dead seedlings by root rot, the results indicated highly significant differences concerning the growth of the P. cinnamomi strain CPO-CPU. The comparison performed with the control seedlings inoculated only with this oomycete revealed symptoms of root rot, growth retardation, wilting, defoliation, root necrosis, and seedling death. In contrast, in the control (without an inoculum) avocado seedlings, these symptoms did not occur, and seedling development was considered normal. Similar results were obtained in studies focused on the activity of Trichoderma in roots from other crops [54,55].
Furthermore, these results reflect that the strains of Trichoderma spp. act as antagonists and play significant roles in plant growth, and others display an efficient mycoparasitism mechanism against soil-borne plant pathogens, including competition for space within the ecological niche, appressorium formation, and antibiosis. In this response, plants activate induced systemic resistance, a process that has been well documented [44,46,56]. Observations made in vitro between the inter-action points of the plant pathogen P. cinnamomi and Trichoderma spp. were supported by microscopic observations of mycoparasitism, in which vacuolation in the mycelium of the oomycete, disintegration, and parallel mycelial growth were found [46]. The colonization capacity of the biological control agents is considered essential to establish themselves in the rhizosphere, especially in root endophytes or those inhabiting the ecological niche of the avocado and cinnamon roots.
The mechanisms of action of Trichoderma species prevent the host from being affected by other kinds of pathogens. In the case of T. asperellum, it contributes to the active mechanisms against biotic and abiotic stress (salinity, low availability of nutrients in the soil, and root infection), conferring resistance to the plant in the presence of soil-borne pathogens [57]. Another interesting aspect of T. koningiopsis is the release of koninginol, the compounds isolated from the endophytic fungus T. koningiopsis (10S-7-isopropyl-4,10-dimethylbicyclo), which has potent antimicrobial activity, preventing infection by bacteria and fungi spread through the soil [58].

Conclusions
This study reports to T. asperellum, T. hamatum, T. harzianum, and T. koningiopsis isolated from avocado and cinnamon roots, which have been very poorly studied in Mexico in those hosts. Most species of Trichoderma isolated in this work partially or totally covered oomycete growth, reducing the incidence rate of root rot, wilting, and death of avocado seedlings in the greenhouse. On the other hand, inoculation with Trichoderma spp. showed that the percentage of P. cinnamomi infection severity decreased, showing an increase in the health of avocado seedlings by the reduction in the percentage of wilting, root rot, and death of plants. The results of this study provide evidence that endophytic indigenous species of Trichoderma isolated from avocado and cinnamon roots efficiently contributed to in vitro and greenhouse biocontrol against the plant pathogen oomycete P. cinnamomi, the causal agent of root rot of avocado worldwide.